Mechanisms of Hard Tissue Destruction Marine Biological Laboratory Library Woods Hole, Mass. Presented by Stephen A, Wainwright June 4, 1964 S C7 Mechanisms of Hard Tissue Destruction A Symposium Presented at the Philadelphia Meeting of The American Association for the Advancement of Science December 29 and 30, 1962 Edited by REIDAR F. SOGNNAES School of Dentistry Center for the Health Sciences University of California at Los Angeles f^^f 1 Publication No. 75 of the AMERICAN ASSOCIATION FOR THE ADN'ANCEMENT OF SCIENCE \Vashington, D. C, 1963 Copyright © 1963 by the American Association for the Advancement of Science Library of Congress Catalog Number 63-23050 Printed in the United States of America Preface Mechanisms of Hard Tissue Destruction is based on a four-session symposium organized by the Section on Dentistry of the American Association for the Advancement of Science, held during the 129th annual meeting of the AAAS in Philadelphia, Pennsylvania, on December 29 and 30, 1962. The symposium was cosponsored by the AAAS Sections on Den- tistry (Nd), Medicine (N), and Zoology (F) and by the Interna- tional Association for Dental Research, North American Division, the American College of Dentists, and the American Dental Associa- tion. A multidisciplinary approach was chosen with a view to covering a theme which could serve as a logical sequence — the other side of the coin, as it were — to a previous AAAS symposium which dealt with the formative aspects of hard tissue biology (Calcification in Biological Systems, AAAS Publication No. 64, Washington, D. C, 1960). In the present volume the oral presentations have been supple- mented by three additional manuscripts (chapters 5, 10, and 23), making a total of twentv-six chapters by forty-eight authors and coauthors, including fourteen from institutions outside the United States. The international participation in this symposium was made pos- sible in part by a conference grant from the National Institute of Dental Research, negotiated together with Dean Lester W. Burket and Dr. Ned B. Williams, University of Pennsylvania School of Dentistry, the latter serving as 1962 AAAS Vice President and Chairman of AAAS Section on Dentistry (Nd). Serving with me as co-chairmen of the individual symposium ses- sions were Drs. Seymour J. Kreshover, National Institute of Dental iii IV PREFACE Research (now Secretary, AAAS Section Nd); Franklin C. McLean, University of Chicago; and George Nichols, Jr., Harxard University. The cost of the color plate (facing page 218) was covered by a grant from tlie Miami Vallev Laboratories of the Procter & Gamble Companv, Cincinnati, Ohio. From the initial planning until the final preparation of subject and author indexes, I have enjoved assistance bevond the hours of duty from Mrs. Dorothy Good, Administrative Assistant, and Mrs. Phyllis Lessin, Secretary, School of Dentistry of the University of California, Los Angeles. I am grateful for the excellent cooperation of all the authors here and abroad, and for other expert help in editing this volume. Each of the individual chapters that follow is concluded with a summary, and the brief tabulation below mav serve to orient the SUMMARY OF CONTENTS Chapter Destructive Structures Biological NUMBERS PROCESSES INVOLVED influences 1 to 4 IJoring canals < 'Calcareous rocks Coral reefs Shells ^Tusks and teeth MoUusks Sponges Gastropods Postmortem fungi (?) 4 and 5 'Attrition Abrasion Erosion ."Erosion" Teethl Teeth] Enamel, dentin Enamel Mastication, bruxism Saliva, bacteria {?) Postmortem algae (?) 0 to 10 Caries Teeth Bacteria ■= Resorption Antler 1 10 to ^24 Bone Cementum Dentin > Multinucleated giant cells ^"Osteolysis" Enamel Bone Osteocytes 25 Chelation Shells, bones, teeth (?) Sequestering agents 26 Proteolysis Collagen Collagenase PREFACE V reader regarding the extent to which different reports have shed hght on related mechanisms. The primary purpose of the symposium was to examine the con- ditions in which mineraUzed structures — inchiding rocks, corals, shells, antlers, bone, ivorv, cementum, dentin, and enamel — are subject to destruction by various marine and subterranean organisms such as boring sponges, mollusks, snails, octopuses, worms, algae, and fungi, as well as by the action of the giant cells tvpical of lacunar resorption and the oral bacteria responsible for tooth decay. Beyond the morphological and cellular levels of observation, the symposium also served to delineate present knowledge and various areas of ignorance regarding specific chemical agents which lead to the disruption and dissolution of the inorganic salts and organic matrices of mineralized structures; e.g., glandular secretions and various extracellular and intracellular metabolites, acids, chelators, enzymes, and combinations of chemical and physical factors. When comparing the various mineralized structures that succumb to decalcification in biological systems, one is impressed by the broad spectrum of their chemical and physical properties, as well as bv the varietv of the biological organisms and biochemical agents involved in their dissolution. Rock-boring organisms can disintegrate not onlv relatively soft sedimentary rock, but also densely mineral- ized calcareous products. Boring sponges burrow not only into corals, but also into limestone and shells whether composed of cal- cite or of aragonite. Gastropods "drill" into the shells of bivalves as well as those of their own fellow snails. Excised gastropod boring organs can act on hard tissues other than shells and will produce etchings when the calcium phosphate crystals of human enamel and dentin are exposed to them in vitro. Indigenous oral microorganisms evidently are endowed with a dual capacity to produce agents which can dissolve and digest hard tissues of as contrasting composition as enamel and dentin, so as to cause tooth decay, now definitely established as being of bacterial origin. Subterranean fungi under postmortem conditions produce boring canals by a dissolution of the collagen and calcium phosphate in buried bone, ivory, cemen- tum, and dentin, but leave dental enamel alone. Marine fungi at- tack the shells of bivalves (calcite) and snails (aragonite) and VI - PREFACE have the biochemical capacity to digest the organic conchioHn shell matrix. Vertebrate hard tissues prone to biological destruction are not entirely uniform with regard to the nature of their organic scaf- folding and inorganic building blocks. Presumably there may also be different molecular bonds which bridge the two, i.e., the organic- inorganic linkage which renders the biological whole — be it shell, pearl, ivory, or bone — something far more complex ( as well as more beautiful) than a simple summation of the chemical parts. The com- plexity of analyzing this problem has been illustrated h\ a number of biological systems described in this volume. All together at least half a dozen destructive influences appear to be at work: acid demineralization, chelation, enzymatic diges- tion, proteolysis, molecular bond disruption of the organic-inorganic linkage, cellular ingestion, possibly phagocytosis, physical motion, and mechanical friction. A combination of two or more if not all of these mechanisms may well exert their influence at some stage of destruction within one and the same biological system. The more readily understood physical forces are at work in the case of large multicellular organisms, e.g., the twisting motion of rock-boring bivalves and the drilling action of the snail rasping holes in a shell region partially softened by decalcification. Yet anyone who has observed the cinematographic recordings of the lively process of experimental bone resorption in tissue culture will have a vivid impression that the osteoclast — aside from its complex biochemical apparatus — does in fact move around in a slow-motion "twist," rub- bing its pseudopodia along the presumably softened walls of an eroding Howship's lacuna. Though the discussion is primarily focused on the destructive as- pects of hard tissue biology, it is noteworthy that a variety of sys- tems in fact exhibit closely related constructive (biopositive) and reparative phenomena; in other words, that we are dealing with interrelated three-way processes: formation, destruction, and re- formation at the cytological level; mineralization, demineralization, and remineralization at the molecular level; in short, cellular and chemical remodeling. In coral reef remodeling the extensive "erosion" of the dying PREFACE Vll coral skeletons by boring sponges is countered by extensive rebuild- ing through the calcifying powers of the living coral polyps. As rock-boring mussels channel their way into mineralized structures, the dissolved calcareous material is deposited on the walls of the burrows. In vertebrate hard tissues, redeposition of new large in- organic crystals takes place, at least at the ultrastructural level, within superficially altered tooth substance both in erosion and in caries. Such intermittent recrystallization may in part have a "re- parative" significance. For example, the large crystals filling the den- tinal tubules in dental erosion could possibly explain the failure of oral microorganisms to invade the dentin substance; and, similarly, in dental caries the large crystals noted in partially demineralized areas have been found to contain an exceptionally high amount of fluoride, which presumably would make such tooth substance less soluble. Moreover, when the dental enamel is exposed to a de- mineralizing solution, the "first order" diffusion-controlled reaction can be inhibited bv deposition of protective reaction products in equilibrium with acid solutions (dicalcium phosphate and calcium fluoride on the surfaces of hydroxyapatite and fluorapatite respec- tively). In brief, dental erosion and caries, destructive as they ar^«^ can no longer be looked upon as entirely one-way processes, at^T^rRflk not from the point of view of molecular biology. In terms of protective mechanisms there is limited knowledge regarding certain organic coatings which appear to modify the de- structive processes in teeth as well as in shells, and possibly in bone. Thin salivary films which cover the tooth surfaces appear to have a significant bearing on the relative protection of the thin external layer of enamel in early caries. Furthermore, it has been suggested that dental erosion may in part be due to the absence of the pro- tective action of such a salivary film. The mussels, whose rock- boring capacity is assisted by a calcium-dissolving secretion, have an organic protection against decalcification of the mussels' own shells, these being covered by a thick periostracal horny covering. When snails and octopuses erode the shells of oysters and abalone, it must be presumed also that a preferential shell destruction of their prey can occur only if the gastropod's own radula is protected from both mechanical and chemical action through the presence of Viii PREFACE either some organic coating or a different crystal structure of tlie denticles, or both. In addition to such potential protection of the inorganic phase against demineralization, other factors appear to control the en- zymatic breakdown of the underlying organic framework. Observa- tions on the action of collagenase suggest that the amorphous ground substance may serve as a protective coating wliich could modifv the coUagenolytic activity. The pertinence of this concept in dissolution of bones and teeth has not been fully established. The very same substance, presumably an acid mucopolysaccharide, has been thought to be involved in the process of calcification ( see Cal- cification in Biological Systems). The two concepts could conceiv- ably be reconciled, however, were one for the moment simply to suggest that the ground substance serves a stabilizing function. When all is said and done, it will appear that the weakest link in our present fundamental understanding of the mechanisms involved in dissolution of mineralized structures relates to the specific chem- ical agents located in immediate juxtaposition to the dissolving sur- faces. Whereas the living culprits of destruction generally can be identified at the "scene of the crime" — be they gastropods, mollusks, sponges, algae, fungi, osteoclasts, or bacteria — the precise micro- environments in which these biological systems operate present great difficulties in research and consequently certain differences in . interpretation. It is hoped that this volume may serve as a springboard for fur- ther research on hard tissue biology throughout the animal king- dom, especially at the relatively unexplored level of molecular biology. REmAR F. SOGNNAES, Editof Program Chairman and Retiring Secretary, AAAS Section on Dentistry September, 1963 Contributors C. R. Barnicoat, Cawthion Institute, Nelson, New Zealand Leonard F. Belanger, Department of Histology and Embryology, Faculty of Medicine, Uniyersity of Ottawa, Ottawa, Canada Sol Bernick, Department of Anatomy, Uniyersity of Southern Cali- fornia, Los Angeles, California Surindar N. Bhaskar, Department of Dental and Oral Pathology, United States Army Institute of Dental Research, Walter Reed Army Medical Center, Washington, D. C. Brian Boothroyd, Department of Histology, Uniyersity of Liyer- pool, Liyerpool, England Melbourne R. Carriker, Systematics-Ecology Program, Marine Biological Laboratory, ^^'oods Hole, Massachusetts David V. Cohn, Kansas City Veterans Administration Hospital, Kansas City, Missouri D. Harold Copp, Department of Physiology, Faculty of Medicine, Uniyersity of British Columbia, Vancouyer, Canada Arthur I. Darling, Uniyersity of Bristol Dental School, Bristol, England C. Dawes, Department of Chemistry, Harvard School of Dental Medicine, Boston, Massachusetts C. M. Dowse, Department of Radiation Biology, School of Medicine and Dentistry, University of Rochester, Rochester, New York Bernard K. Forscher, School of Dentistry, University of Kansas City, Kansas City, Missouri, and (new aflRliation) Mayo Clinic, Rochester, Minnesota Marion D. Francis, Miami Valley Laboratories, The Procter & Gamble Company, Cincinnati, Ohio Paul Goldhaber, Department of Oral Histopathology and Perio- dontologv, Harvard School of Dental Medicine, Boston, Massa- chusetts ix X CONTRIBUTORS Thomas F. Goreau, Department of Physiology, University of the West Indies, Kingston, Jamaica Richard J. Goss, Department of Biology, Brown University, Provi- dence, Rhode Island John A. Gray, Miami Valley Laboratories, The Procter & Gamble Gompany, Gincinnati, Ohio Jerome Gross, Department of Medicine, Harvard Medical School at the Massachusetts General Hospital, Boston, Massachusetts Norman M. Hancox, Department of Histology, University of Liver- pool, Liverpool, England Ghester S. Handelman, Harvard School of Dental Medicine and Forsyth Dental Genter, Boston, Massachusetts Willard D. Hartman, Peabody Museum of Natural History and Department of Biologv, Yale Uni^'ersitv, New Haven, Gonnecti- cut James T. Irving, Harvard School of Dental Medicine and Forsyth Dental Genter, Boston, Massachusetts G. Neil Jenkins, Department of Physiology, Medical School, King's Gollege, Newcastle upon Tyne, England Erling Johansen, Department of Dentistry and Dental Research, School of Medicine and Dentistry, University of Rochester, Rochester, New York Harold V. Jordan, Laboratory of Microbiology, National Institute of Dental Research, Bethesda, Maryland Jenifer Jo\vsey, Department of Microradiography, Albert Einstein Medical Genter, Northern Division, Philadelphia, Pennsylvania, and (new affiliation) Section of Surgical Research, Mayo Glinic, Rochester, Minnesota Paul H. Keyes, Laboratory of Histology and Pathology, National Institute of Dental Research, Bethesda, Maryland K. Lane, Department of Radiation Biology, School of Medicine and Dentistry, University of Rochester, Rochester, New York Gharles M. Lapiere, Institut de Medecine, Hopital de Baviere, Liege, Belgium, and (visiting research fellow) Department of Medicine, Harvard Medical School at the Massachusetts Gen- eral Hospital, Boston, Massachusetts CONTRIBUTORS XI Norman S. MacDonald, Departments of Nuclear Medicine, Bio- physics, and Radiology, School of Medicine, Center for the Health Sciences, University of California, Los Angeles, Cali- fornia Garland N. Martin, Jr., Laboratory of Histology and Pathology, National Institute of Dental Research, Bethesda, Maryland William V. Mayer, Department of Biology, Wayne State Univer- sity, Detroit, Michigan Franklin C. McLean, Department of Physiology, University of Chicago, Chicago, Illinois B. B. MiGicovsKY, Animal Research Institute, Research Branch, Canada Department of Agriculture, Ottawa, Canada Milton J. Moss, Division of Orthopedics, Department of Surgery, School of Medicine, Center for the Health Sciences, University of California, Los Angeles, California M. W. Neuman, Department of Radiation Biology, School of Medi- cine and Dentistry, University of Rochester, Rochester, New York W. F. Neuman, Department of Radiation Biology, School of Medi- cine and Dentistry, University of Rochester, Rochester, New York George Nichols, Jr., Departments of Medicine and Biochemistry, Harvard Medical School, Boston, Massachusetts Ingjald Reichborn-Kjennerud, Faculty of Odontology, University of Oslo, Oslo, Norway Jacques Robichon, Department of Histology and Embryology, Fac- ulty of Medicine, University of Ottawa, Ottawa, Canada Robert E. Rowland, Radiological Physics Division, Argonne Na- tional Laboratory, Argonne, Illinois, and (new affiliation) Department of Radiation Biology, School of Medicine and Dentistry, University of Rochester, Rochester, New York David B. Scott, Laboratory of Histology and Pathology, National Institute of Dental Research, Bethesda, Maryland William A. Skoog, Department of Medicine, School of Medicine, Center for the Health Sciences, University of California, Los Angeles, California Xll CONTRIBUTORS Reidar F. Sognnaes, School of Dentistry and School of Medicine, Center for the Health Sciences, University of California, Los Angeles, California Marshall R. Urist, Division of Orthopedics, Department of Sur- gerv. School of Medicine, Center for the Health Sciences, Uni- versity of California, Los Angeles, California Jacques Vincent, Department of Anatomy, Universite Lovaninm, Leopoldville, Congo CM, YoNGE, Department of Zoology, University of Glasgow, Glas- gow, Scotland Richard W. Young, Department of Anatomy, School of Medicine, Center for the Health Sciences, University of California, Los Angeles, California J^ Contents 1 Rock-Boring Organisms, bv C. M. Yonge 1 2 Boring Sponges as Controlling Factors in the Formation and Maintenance of Coral Reefs, bv Thomas F. Goreau and WiLLARD D. Hartman 25 3 Demineralization Mechanism of Boring Gastropods, bv VIelbourne R. Carriker, David B. Scott, and Garland N. Martin, Jr. 55 4 Dental Hard Tissue Destruction with Special Reference to Idiopathic Erosions, by Reidar F. Sognnaes 91 5 Attrition of the Hypsodont (Sheep's) Tooth, by C. R. Barnicoat 155 6 Microstructural Changes in Early Dental Caries, by Arthur I. Darling 171 7 Ultrastructural and Chemical Observations on Dental Caries, bv Erling Johansen 187 8 Physical Chemistrv of Enamel Dissolution, by John A. Gray and Marion D. Francis 213 9 Factors Influencing the Initiation, Transmission, and Inhi- bition of Dental Caries, by Paul H. Keyes and Harold V. Jordan 261 10 Eftect of Hibernation on Tooth Structure and Dental Caries, by William V. Mayer and Sol Bernick 285 11 Dento- Alveolar Resorption in Periodontal Disorders, by Ingjald Reichborn-Kjennerud 297 12 Bone Remodeling during Dental Eruption and Shedding, bv Surindar N. Bhaskar 321 13 The Deciduous Nature of Deer Antlers, bv Richard J. Goss ' " 339 14 Internal Remodeling of Compact Bone, by Franklin C. McLean and Robert E. Rowland 371 xiii xiv CONTENTS 15 Rarefying Disease of the Skeleton: Observations Dealing with Aged and Dead Bone in Patients with Osteoporo- sis, by Marshall R. Urist, Norman S. MacDonald, Milton J. Moss, and William A. Skoog 385 16 Microradiography of Bone Resorption, by Jenifer Jowsey 447 17 Histophysical Studies on Bone Cells and Bone Resorption, by Richard W. Young 471 18 Structure-Function Relationships in the Osteoclast, by Norman M. Hancox and Brian Boothroyd 497 19 Bone Destruction by Vlultinucleated Giant Cells, by James T. Irving and Chester S. Handelman 515 20 Resorption without Osteoclasts ( Osteolvsis ) , by Leonard F. Belanger, Jacques Robichon, B. B. Migicovsky, D. Harold Copp, and Jacques Vincent 531 21 In Vitro Studies of Bone Resorptive Mechanisms, by George Nichols, Jr. 557 22 In Vitro Carbohydrate Metabolism of Bone: Effect of Treatment of Intact Animal with Parathyroid Extract, by Bernard K. Forscher and David V. Cohn 577 23 Metabolic Action of Parathvroid Hormone on Rat Cal- varia, by C. M. Dowse, M. W. Neuman, K. Lane, and W. F. Neuman 589 24 Some Chemical Factors Influencing Bone Resorption in Tissue Culture, by Paul Goldhaber 609 25 The Possible Role of Clielation in Decalcification of Bio- logical Systems, by G. Neil Jenkins and C. Dawes 637 26 Animal Collagenase and Collagen Metabolism, by Charles M. Lapiere and Jerome Gross 663 Indexes 695 Rock-Boring Organisms C. M. YONGE, Department of Zoology, University of Glasgow, Glasgow, Scotland BEARING in mind the problems presented by this mode of hfe, the habit of rock boring is surprisingly widespread among marine or- ganisms. Among plants, it is found in a variety of green, blue-green, and red algae and also in some fungi. Boring animals include certain sponges, a flatworm ( Turbellaria ) , various sipunculid and polychaete worms, certain echinoid echinoderms, a genus of barnacles, and a diversity of gastropod and, above all, bivalve molluscs. In general these organisms are inhabitants of shallow, most often intertidal, waters. Although of wide occurrence where suitable substrates exist in temperate and tropical seas, they are undoubtedly most abundant on tropical coral reefs and within mudstones in temperate waters. Personal experience of boring organisms has been gained during the course of the Great Barrier Reef Expedition of 1928-1929 and, more recently, on the central California coast while working during 1949 and subsequently at the University of California, Berkeley, and at the Hopkins Marine Station of Stanford University at Pacific Grove. The habit of boring is obviously not primitive. The substrate bored is either relatively soft sedimentary rock or else the calcareous product of animal secretion, notably coral skeletons and the shells of molluscs, especially the larger Bivalvia. Certain organisms, it may be noted, such as serpulid polychaetes which secrete a calcareous tube and the neogastropod Coralliophilidae, e.g. Magilus, settle 1 C. M. YONGE upon corals and extend their shells to keep pace with the growth of these. The final appearance gives a misleading impression of bor- ing. Organisms bore either by mechanical or, if the rock be largely or in part calcareous, by chemical means. The firmer the substrate (e.g. many bivalve shells), the greater the need for at least some chemical assistance in boring. Significance and Mode of Boring The ability to bore invariably confers a high degree of protection, and this certainly represents the biological reason for the prevalence of the habit. Unlike the wood-boring bivalve Teredinidae (ship- worms) or the crustacean Limnoria (gribble), which obtain much of the energy for boring from the material into which they pene- trate, it is the exception for rock borers to obtain energy in this manner. Nevertheless we may conveniently begin by considering cases where boring is either certainly or possibly associated with feeding. Some Association with Nutrition Plants. The one certain case where energy is obtained by the borer is that of the fungi which ramify through dead or living bivalve shells, utilizing the energy present in the organic conchiolin matrix of the shell. The best-known instance is provided by the causal agent of Dutch shell disease, which formerly did great damage to Euro- pean stocks of oysters, often spreading as spores to the living oyster shells from dead shells used as a settling surface or "cultch." Korringa (1952) has described the life history; although oysters die if the shells become heavily infected, this is due to reaction by the mollus- can tissues. The fungus itself never penetrates the tissues. Presum- ably boring is along the areas of conchiolin, although some actual penetration of the calcified regions bv either mechanical or, more probably, chemical means may well be necessarv. A variety of filamentous green, blue-green, and also some red algae penetrate into calcareous rock or shells, either deeply or superficially. Blue-green algae are abundant between tidal levels on coral reefs, causing a softening which may be due to the action of CO2 or other ROCK-BORING ORGANISMS 3 acids produced in metabolism (Newliouse, 1954; Newell, 1955, 1956). Less eas)' to understand is the boring activity of the green algae which penetrate into living coral rock, where thev form a green zone up to 3 mm within the skeleton. They may also occur in dead coral fragments. The commonest genus appears to be Acht/Ia, but others are listed by Utseumy (1942). The little that is known about the manner of boring indicates that this is due to the combined action of acid production and growth pressure. There is also the suggestion that entrance may be associated with the presence of nutritive matter, presumably derived from the small organic content within the skeleton (Duerden, 1902, 1905). Odum and Odum (1955) have recently suggested that there may be some symbiotic relation be- tween the corals and these algae, but, in the absence of any experi- mental evidence, this is difficult to accept. A full understanding of the mode of boring and of the source of material needed for protein svnthesis in these algae is badly needed. But the problems are complex and investigation is difficult.* Animals. Possible direct association. Certain animals which bore into coral rock may possiblv gain all or a significant part of their food from the boring algae and the almost equally common and widely ramifying sponges (described later). This could apply to the sipunculid worms, such as the numerous species of Aspidosiphon, which commonly bore into coral rock. The small cuticular plates present in the skin of sand-burrowing sipunculids are here fused to form massive shields with which the animal bores. Since the primi- tive habit is burrowing into sand, which is swallowed in great quanti- ties with digestion of the small amount of contained organic matter, change to a rock-boring habit presents no major problem. It is cer- tainly possible that the rock into which the worms bore may contain as much organic matter, in the form of boring plants and sponges, as the sand. What applies to these sipunculids may also, although less prob- ably, apply to some of the coral-boring polychaetes. Gardiner ( 1903) has listed the more important families. According to Crossland * Since the above was written, knowledge has been significantly advanced by Wain Wright (1963). C. M. YONGE (quoted by Gardiner, 1931), the largest Indo-Pacific species "are Leocrates (Eunice) siciliensis, Lijsidice coUaris and some of the CirratuHdae. The former two have enlarged gouge-shaped lower jaws, with hard, though calcareous, cutting edges, which are ob- viously their boring tools, but the latter have soft bodies and no conspicuous hard parts at all." Unlike that of the sipunculids, the mode of feeding of related nonboring species does not indicate that thev may get food bv boring. Like the temperate-water Polijdora (described later), they may bore solely for protection. Study of these forms in life would solve this problem. The turbellarian flatworm Pseudostylachus ostreophagus, which is a native of Japanese waters but was introduced into Puget Sound with consignments of Japanese oysters (Crassostrea gigas), attacks the oysters when young, making a keyhole-shaped opening in the shell and then eating the contained animal. The process of boring is unknown, but in the absence of jaws or any other hard structure must involve chemical means. Certain carnivorous Gastropoda, nota- bly the prosobranch Muricidae and Naticidae, bore through the firm calcareous shells of their prey, usually bivalves, by the mechanical action of the radula assisted by acid secretion from an accessory boring organ ( Carriker, 1961 ) . Later the flesh of the prey is rasped out by means of the radula. Full description of the processes in- volved is provided by Carriker and his co-workers in chapter 3 of this symposium. Indirect association. We are here concerned with animals which scrape the surface of calcareous rock to obtain the algae which grow upon it or bore superficially into it. This is true of species of largely intertidal prosobranch Gastropoda, of such genera as Nerita and Turbo, which cut into the softened rock surface with the broad radula and assist in the formation of the characteristic intertidal "nick" along the landward margin of coral reefs ( Doty and Morrison, 1954; Newell, 1955, 1956; Newell and Imbrie, 1955)'. But in no real sense can these gastropods be said to bore. Much the same habit exists in a variety of echinoid echinoderms (sea urchins) which excavate new borings or enlarge old ones by means of the teeth and spines. This is purely for protection, primarily against wave action but to some extent on exposed shores against ROCK-BORING ORGANISMS O desiccation, and occurs only when such dangers exist. It is a purely facultative habit and is found in various intertidal echinoids. Umb- grove (1947) considered Echinometra mathaei to be a major agent of erosion in East Indian coral reefs. Only when the borings become deep and finally flask-shaped are the animals imprisoned within them. Usually they are able to leave the "boring" and forage on the surrounding rock surface. Food con- sists largely of encrusting algae, and probably sufficient fragments are carried in by water movements to support imprisoned animals. But any algae growing on or within the rock in the borings will be a source of food. There is no question here of chemical action, in- deed the rock may not be calcareous, and boring is to a large extent fortuitous. The whole matter has been most thoroughly reviewed and discussed by Otter ( 1932 ) . No Association with Nutrition Sponges. A variety of sponges, most notably and ubiquitously the numerous species of the genus Cliona, bore into calcareous rocks, into coral skeletons, and into mollusc shells. Cliona celata, which bores into calcareous structures, especiallv ovster shells, in European waters is probably the best-known species. The larvae settle from the plankton and the developing sponge quickly penetrates the shell, forming extensive galleries so that the whole structure becomes fria- ble and crumbles away. There can be no question here of the sponge's gaining food in this way; like all sponges, it feeds on nano- plankton drawn into the body in currents created by the flagella. It does not necessarily bore, indeed this would appear to depend on the nature of the substrate on which the larvae settle. The method of boring remains obscure; the siliceous spicules could assist mechan- ically, but, in view of the calcareous substrate, acid could obviously be used. Growth processes could also assist, as suggested in the case of algae. Indeed, as noted by Gardiner (1903), the algal Achijla and Cliona resemble each other in their mode of growth; Gardiner adds that "their ramifications are most delicate, imperceptible to the un- aided eye, and wander all over the coral skeleton. In Pocillopora their terminal filaments extend so close to the ends of the branches that only the very thinnest layer of corallum separates the polyp r. M. yongp: tissues from them. When a branch of this genus is decalcified and the polyp layer carefully stripped off, either of these boring organ- isms will be seen to have formed a close-meshed network, showing accurately the shape of the original branch with all its twigs." Gar- diner considered that neither of these organisms bores into anything but newly secreted coral skeletons. As noted above, thev could pro- vide food for boring sipunculids and, less probablv, polvchaetes. The different, much more finely ramifying, type of growth within coral as compared with bivalve skeletons mav well be related to differences in the skeleton, possibly in the organic content. Goreau and Hart- man add greatly to our knowledge of the boring of sponges into corals in their contribution to this symposium (chapter 2). PoLYCHAETES. The bcst-kuown polvchaete rock borers, always apparently boring into calcareous substrates, are species of Polijdora, which may be serious pests of oysters in many parts of the world. These small worms initially form mucous tubes to which mud ad- heres. They tend to back into crevices and can then bore if the sub- strate is suitable. They do so with the bodv bent double into a U, the arms of which are separated by a partition of mud and debris consolidated in mucus. Boring must be in part, if not wholly, me- chanical, possibly bv the agencv of enlarged dorsal setae on the fifth setigerous segment. Since the substrate is apparently always calcareous, boring could be assisted by some initial chemical action, although there is no evidence of anv glands on the surface of the body. Other species inhabit crevices in shale and sandstone. There seems to be some doubt as to whether these ever bore; if not, it could be because the rock is harder, not because it is noncalcareous. CiRRiPEDES. The attached habit of barnacles has led, following fixation to the body of animals, to various degrees of commensalism culminating in parasitism. It has also, most notablv in the genus Lithotrya, led to penetration of the rock on which the c\'pris larvae settle. Species of this genus are among the commonest borers on both Indo-Pacific and Atlantic coral reefs (Cannon, 1935; Seymour Sewell, 1926). The chitinous covering of the contractile peduncle contains numerous "nail-like bodies, the head of each nail being ROCK-BORING ORGANISMS originalh^ enclosed in a small calcareous head" (Seymour Sewell, 1926). B)' this means the animal excavates a cavity into which the contracted peduncle fits, with the opercular plates just flush with the surface, so that even when common these barnacles are initially most difficult to detect. The peduncle is attached to the boring by way of a calcareous basal plate a little distance from the tip on the carinal side (Fig. 1). This supplies the necessary purchase for bor- FiG. 1. LitJiotrya nicoharica, probable appearance when withdrawn within boring. (Modified after Seymour Sewell, 1926.) ing, although, of course, the position of attachment must continually alter during growth (the same problem arises in the case of the boring bivalve Tridacna crocea, as described below). The process of boring must be purely mechanical. Like other barnacles, LitJio- trya feeds on zooplankton organisms by means of its setous thoracic appendages. Bivalve molluscs. The basic structure of these animals ideally fits them for boring. Enclosure of the entire body by the mantle and shell has led to loss of the head and enlargement of the gills, which have assumed the added function of collecting food consisting of phytoplankton. Except in a few primitive and some specialized C, M. YONGE groups, both intake and extrusion of the essential water current take place at the posterior end. Hence when these animals have succeeded in penetrating into rock they have achieved maximum protection while continuing to feed by ciliary currents in the normal manner. There is no question of any of them obtaining anything of food value from the rock. The shell valves are always the prime agents of boring, although the manner in which they operate to this end differs greatly. Only in certain members of one superfamily (vii below) is the mechanical action of the shell valves undoubtedly assisted by chemical means. Origin and Nature of Boring in the Bivalvia So well are these animals basically fitted for boring that members of no less than seven superfamilies of the Bivalvia have, independ- ently of one another, taken to this mode of life. It is interesting also that they have arrived at this final habit by one of two routes. Primitively the Bivalvia are members of the infaima, i.e. they live within soft substrates through or down into which they burrow by means of the initially hatchet-shaped foot terminally dilated by blood pressure. By further development of the habit of deep burrowing they have come to penetrate stiffer and stiffer substrates, finally becoming true borers. This is true of some or all of the mem- bers of the first two superfamilies ( i and ii ) described below. Certain Bivalvia, e.g. marine mussels, have secondarily become permanently attached in adult life to a hard substrate, i.e. they have become members of the epifauna. For reasons given elsewhere (Yonge, 1962), this is probably due to retention into adult life of the mechanism of attachment by byssal threads. This probably was originally a larval structure concerned solely with temporary at- tachment of the newly settled larva, and this certainly constitutes its sole function in the more primitive infaunal bivalves. The epi- faunal habit brings the bivalves into intimate association with rock, and this is the certain, or very probable, route whereby the boring habit has been reached in superfamilies iii to vii. The degree of speciafization exhibited within the different groups KOCK-BORING ORGANISMS 9 of rock-boring bivalves varies very greatlv, as brieflv outlined below. Infatinal Origin i. SuPERFAMiLY Myacea, Plalijodon cancellatiis, the sole boring species (and genus), somewhat resembles the soft-shelled clam, Mija arenaria, to which it is closely related. It is confined to the California coast between 33° and 38° north latitude, where it oc- curs between tidemarks, boring in relatively soft mudstones (Yonge, 1951), It bores mechanically by means of the shell valves, which are ridged and much eroded in the umbonal regions. Compared with Mija, the valves are more convex (see Fig. 2), and the con- A B Fig. 2. Platyodon cancellatus. A, animal in boring, showing erosion of shell around umbones, also periostracal "scales" near tip of siphons; B, section through boring in umbonal region. (After Yonge, 1951.) densed ligament or chondrophore is displaced somewhat posteriorly. The horny periostracum which covers the long siphons is thickened near the tip to form four "scales," which represent the only obvious adaptation for boring. Platyodon bores directly into the rock, i.e. without twisting. As a result the boring is not rounded in cross section, but, in the region occupied by the shell, has "dorsal" and "ventral" ridges, the former, which lies between the rounded umbones, being very pronounced ( Fig. 2B ) . Boring is the result of lateral pressure by the valves due to taking of water into the mantle cavity, followed by closure and withdrawal of the siphons. Such pressure can be only slightly in- creased by the opening thrust of the somewhat reduced ligament. 10 C. M. YONGE but there may be some rocking on the median fulcrum of the re- duced hinge and hgament by reason of uneven contractions of the two adductor muscles. The valves gape both anteriorly and pos- teriorly. Judging by the greater erosion, most work may be done by the umbonal (but also oldest) regions of the shell. The siphonal extension of the boring is certainly enlarged by means of the four periostracal scales. The reduced foot has no apparent function in boring. ii. SuPERFAMiLY Adesmacea. All mcmbcrs of this large super- family are borers, the more specialized Xylophaginidae and Tere- dinidae living in wood, the majority of the Pholadidae boring into stiff clay or rock. Species of Zirphaea which burrow into veiy heavy clay probably indicate the manner in which the boring habit has evolved. But these are now the only members of the superfamily which do not bore. There is in all, including Zirphaea, great modifi- cation of structure (see Lloyd, 1897; Purchon, 1955fl) in relation to the purely mechanical process of boring. The valves tend to be elongate, to bear rows of teeth or ridges, and to be very convex. By a reduction of the mantle isthmus which runs between the two lobes of the mantle, the ligament is either much reduced or com- pletely lost, i.e. there is separation of the mantle lobes and shell valves. Hinge teeth disappear and are replaced by dorsal articulat- ing processes on which the shell valves, which gape anteriorly and posteriorly, rock. In certain genera (e.g. Pholadidea, as shown in Fig. 3, VA) there are secondary ventral articulatory surfaces. Both adductors migrate dorsally so that they come to lie in the same line as the dorsal articulating surface. In the anterior adductor this in- volves a rolling upward of the anterodorsal region of the valves. By cross fusion of the inner muscular lobe of the mantle margin, a third, mid-ventral, adductor muscle is formed. Following the al- ternate contraction of the anterior and posterior adductors, the shell valves rock on the fulcrum of the dorsal articulation (and ventral articulation where this is present ) . Attachment to the head of the boring, essential for effective bor- ing action, is by way of the foot modified to form a sucker. This projects forward through a wide anteroventral gape between the ROCK-BORING ORGANISMS DA 11 Fig. 3. Pholadidea peniUi, left valve, showing dorsal {DA) and ventral (VA) articulations, anterior {AA) and posterior {PA) adductors, and apophy- sis {AP) to which retractor muscles of foot are attached. Foot protrudes through pedal gape anteroventrallv. Shell valves rock on axis of motion indi- cated by broken line. (After Lloyd, 1897.) valves. The changed position, and function, of the foot involves change in disposition of the pedal muscles. These are now attached to a pair of bladelike shell processes which extend ventrallv from within the hinge region. These apophvses (Fig. 3, AP) are found only in the Adesmacea. During boring the animal rotates, the attaching foot moving first in one direction, then in the other. Thus a smooth boring is cut by the rocking of the valves on the median fulcrum. The effective abrasive action is performed bv the anterior halves of the valves, which are pulled apart bv the action of the posterior adductors. In section the boring is round with no trace of dorsal or ventral ridges. In the most specialized of the adesmacean rock borers, Pholadidea and Parapholas, the animal excavates a boring of a definite size and then the foot with its musculature atrophies and the anteroventral gape is filled in by inward growth of the mantle margins, which secrete a thin callum over the entire surface. This is not true of Zirphaea, Pholas, or Barnea. Apart from the stiff clay in which species of Zirphaea occur, to depths of up to 50 cm, the pholads bore into a wide variety of sub- 12 c. M. yongf: strates including slate, shales, sandstone, chalk, marl, and even peat and submerged wood. Pholadidea penita has been found in concrete casings. In all, the fused siphons are long, the basal half, which is encased in a protective periostracum, representing pos- terior extension of the mantle cavity and housing the posterior region of the gills. The pholads do not appear to bore significantly into limestone and, although usually the most important rock borers in other regions, are certainly largely absent from coral reefs. Epifaunal Origin iii, SuPERFAMiLY Veneracea. Family Petricolidae. The typi- cal venerid burrows very superficially, but in certain species byssal attachment persists, and these forms inhabit rock crevices. In the Petricolidae this habit is taken further and the animals bore into the rock. There is an interesting range of specialization within the genus Petricola. Thus in P. carditoides (Fig. 4A), common in holes in rocks on the California coast, the animal does no more than en- large these as it grows; the process is entirely mechanical by means of the shell valves, water pressure being possibly materially assisted by the opening thrust of the long and unmodified ligament. The byssus is not retained in boring. Great secretion of mucus, both in the mantle cavity and in the siphonal embayment, is associated with this mode of life (Yonge, 1958). The shell may be almost globular or elongate; it is often very irregular in form. At the other extreme comes P. pholadiformis (Fig. 4B) — so named because of the super- ficial resemblance of its elongate, ridged shell to that of a pholad — which bores with great efficiency (Purchon, 1955/?). Species of Petricola bore into a variety of substrates, from stiff mud, mudstone, and shales to limestone. Petricola lapicida, with boring capacities intermediate between those of P. carditoides and P. pholadiformis, is not uncommon in beach rock, coral boulders, etc., on the Great Barrier Reef ( Otter, 1937 ) . iv. SuPERFAMiLY Saxicavacea. Genus HiatclUi ( Saxicava ) . Spe- cies of this widely distributed genus may either live in rock crevices attached by byssus threads or bore. It is not clear whether certain species always bore and others always "nestle"; indeed, there is ROCK-BORING ORGANISMS 13 Fig. 4. A, Petricola carditoides; B, P. pholadiformis; showing greater specialization of the shell for boring in the latter. (From Yonge, 1958.) evidence that the boring habit is facultative and not obHgatory. In his account of Hiatella gallicana and H. arctica, Hunter (1949) states that the habit of the adult is determined by the nature of the substrate on which the larvae settle: if it be the surface of a soft homogeneous rock they will bore, if it be a hard but creviced rock surface they will become byssally attached. The byssus threads (but not the gland) are lost in animals which bore, the process being entirely mechanical (the valves becoming much eroded) by means of water pressure forcing the shell valves apart (Fig. 5). Hunter points out that "the boring movements evolved with little 14 r. M. YONGE B Fig. 5. Hiatella (Saxicava) sp.: A, showing full extension of siphons for feeding and respiration; B, contraction of siphons (openings closed) ready for boring, valves forced apart by water pressure, and thickened siphons grip- ping sides of boring. (After Hunter, 1949.) modification from the protective reactions of non-boring animals." Hiatella is confined to temperate seas in both the Atlantic and the Pacific. It bores into comparatively soft rocks, i.e. muddy lime- stones, mudstones, sandstones, calcite, and chalk (Hmiter, 1949). Although all these rocks are broken down by acid, i.e. the binding cement is acid soluble, there is no evidence that boring is aided chemicallv. The form of the shell \aries so greatly, especialh' in boring individuals, that specific identification is unusually difficult. However, species of Hiatella are most efficient borers, cutting deep borings which are round in cross section. V. SuPERFAMiLY Gastrochaenacea. Genus RoceUaria {Gastro- chaena). Members of this genus are highlv specialized borers of warm-temperate and tropical seas, always boring into calcareous substrates. They are common in Indo-Pacific coral reefs (Otter, 1937). As shown in Fig. 6, there is some superficial resemblance to the Pholadidae, notably in the similar possession of a suckerlike foot projecting through a wide anteroventral gape and also of small internal shell ridges to which the pedal muscles are attached (Purchon, 1954). In other respects, however, there are major dif- ferences, most obviouslv in the ligament, which is here very long, so that the valves cannot possibly operate in boring as do those of ROCK-BORING ORGANISMS R 15 Fig. 6. Rocellaria (Gastrochaena) cuneiformis. Above, animal in situ in boring; below, transverse sections in regions a to d. CL, calcareous lining secreted by walls of siphons (not shown) and forming projecting exhalant (£S) and inhalant (IS) siphonal tubes; F, suckerlike foot; H, hinge region; M,' mantle exposed in anterior pedal gape; P, pedal opening; R, rock; V, shell valve. Broken line denotes inner limit of secreted calcareous lining of boring. (After Otter, 1937.) the Pholadidae, but must abrade as a result of water pressure aided by the opening thrust of the long ligament. A striking feature of Rocellaria (indeed of the entire superfamily) is the capacity of the siphonal tissues to secrete calcium carbonate (Shiiter, 1890), so lining the boring and extending this beyond the rock surface into two tubes formed by the separated ends of the siphons (Fig. 6, ES, IS ) . The boring is oval in section ( Fig. 6 c, d), indicating, together with the fixed position of the siphonal tubes, that the animal bores directly into the rock. The borings appear to be much longer than even the long siphons, and possibly, unlike the Pholadidae, these animals can move backward and forward within the boring. A byssus gland persists, and occasionally threads are secreted. This fact, together with the habits of related genera, indicates that here also boring has followed an initial epifaunal habit. 16 C, M. YONGE vi. SuPERFAMiLY Cardiacea. Family Tridacnidae. THdacna cro- cea. This species, confined to the mid-tropics in the Indo-Pacific, where it is often extremely abundant, boring into coral boulders, is perhaps the most unexpected of bivalve borers. It belongs to a family which includes the largest bivalve ever evolved, THdacna deresa, the giant clam. All species in the family begin life attached by byssus threads. In the larger ones the bvssus and the shell gape through which it projects disappear and the animals finally main- tain themselves by their unaided weight. But in T. crocea the byssus, and the gape through which it emerges, become larger with age, and at the same time the animal proceeds to bore doivnward into the calcareous rock (Fig. 7). This is a consequence of the Fig. 7. Tridacna crocea, viewed, from left side, in situ within boring in coral rock, shown in longitudinal section. Animal lies with hinge (H) under- most, massive byssus (obscured by shell) attached to pillar (R) anterior to this. A, P, anterior, posterior. (After Yonge, 1936.) unique form of these animals. The shell valves have changed their position in relation to the enclosed body so that the hinge comes to lie beside the byssal gape on the underside of the body. The result is that the siphonal tissues (normally at the posterior end) ex- tend along the upper, i.e. dorsal, surface. They also extend laterally, forming a broad band of brilliantly pigmented tissues facing up- ward to the light. Within these tissues are contained countless mil- ROCK-BORING ORGANISMS 17 lions of unicellular algae or zooxanthellae. The Tridacnidae literally "farm" plants in these illuminated tissues, and this is the biological explanation for the remarkable change in form (Yonge, 1936, 1953). The presence of the hinge, instead of the free margin of the valves, on the underside alongside the byssus has made boring possible. Running between the bvssus and the inner surface of the valves are extremely powerful pedal (or byssal) retractor muscles. Alternate contraction of these muscles causes the animal to rock in the longitudinal axis on the fulcrum of the byssus. These rocking movements are totallv distinct from those of the Pholadidae, which are in the median transverse plane and caused by alternate con- tractions of the two adductors. Entry in T. crocea is vertical but to some extent diagonal to allow for undercutting of the byssal attachment on the one side while new threads are formed on the other. The byssus is attached to a pillar ( Fig. 7, R ) at the base of the boring which fits into the byssal gape on the underside of the shell. The thick-ridged shell valves are always greatly eroded where thev bear against the surface of the rock. After they begin to bore (when about 1.4 cm long), the animals must quickly penetrate to the depth of the shell valves, which remain flush with the surface during growth to a final length of some 10 cm, when the boring will be about 7 cm deep. The animal is imprisoned within the boring, the length of the shell be- ing greater than that of the opening (see Fig. 7). Except when withdrawn during exposure at low tide, the hypertrophied siphonal tissues spread out over the surface of the rock, completely covering both the margins of the valves and the sides of the boring. The high temperature at which these animals spawn, about 30 °C, con- fines them to shallow water in the mid-tropics. vii. SuPERFAMiLY Mytilacea. Family Mytilidae. This large family, which includes the common mussels, contains two important genera of rock borers, Botula and Lithophaga. Both are byssally at- tached throughout life, but whereas the former bores only into soft, usually noncalcareous, rocks such as mudstones and shales, the latter is confined to calcareous rocks — limestones, calcareous shales, 18 c. M. yon(;e carbonate-cemented sandstones, shells of other molluscs, and above all coral rock — and boring is certainly assisted by initial chemical softening of the rock ( Yonge, 1955 ) . In both genera the shell is elongated and almost tubular, a modi- fication of the somewhat triangular shell form of the typical mussel. This form, brought about by modification of the components of shell growth (see Yonge, 1955), enables them to penetrate rock with great efficiency. In Botula (Fig. 8) a massive byssus is secreted. B Fig. 8. Botula falcata. A, boring exposed to show dorsal side of animal, siphons extended and animal withdrawn from head of boring; erosion of shell on either side of hinge region shown. B, boring opened to reveal attachment to "floor" of boring of byssal threads in large anterior and smaller posterior groups, former pulled against when animal bores. C, boring opened at head end. (From Yonge, 1955.) So attached to the floor of the boring, contraction of the posterior byssal retractor muscles drives the anterior end of the shell forward against the head of the boring. There can be no question here of water pressure within the mantle cavity because there are no true siphons, but the ligament is long and its opening thrust powerful enough to cause widening of the boring. Although the shell has a thick covering of brown periostracum, this and the underlving ROCK-BORING ORGANISMS 19 calcareous laver are invariably deeply eroded around the umbones ( Fig. 8A ) . Boring is straight into the rock, with a pronounced dorsal and a smaller ventral ridge (Fig. 8C). B. californiensis and B. falcata are both common borers, largely between tidemarks, in soft rocks along the coasts of central California where thev were ex- amined ( Yonge, 1955 ) . Lithophaga (Lithodomus) lithophaga, the date mussel of the Mediterranean, has been known since classic times. There are many other species in warm-temperate and tropical seas, and thev are particularly common on coral reefs, five species occurring on Low Isles, Great Barrier Reef (Otter, 1937). Compared with Botula, the shell is more completely rounded in section and is everywhere covered by a particularly thick and dark-colored periostracum. This is never eroded. In most species of Lithophaga anterior and pos- terior areas of the shell are covered with a granular calcareous deposit (Fig. 9). Although bvssal attachment persists throughout 0 0 a b c d Fig. 9. Lithophaga cumingiana. Above, animal //) situ, in boring; below, transverse sections in regions a to d. Small extent of byssal attachment shown, also adherent calcareous debris over anterior and posterior regions of shell (but no erosion). Calcareous lining of boring (as in Rocellaria) secreted by walls of siphons, indicated in section by thick line, marginall)- (ventrally and anteriorly) by broken line. (After Otter, 1937.) 20 C. M. YONGE life, the number of threads and strength of attachment are far less than in Botula (Yonge, 1955). This condition is correlated with an initial chemical softening of the rock. The mantle tissues protrude from between the valves anteriorlv, spreading out laterally over their surfaces. When the animal is pushed against the head of the boring, these tissues are applied to the surface of the rock ( Fig. 10 ) . Fig. 10. Lifhophaga phimula, boring opened from above to show anterior end of shell with mantle tissues protruded against head of boring. Arrows show direction of ciliarv currents. (From Yonge, 1955.) Although the presence of acid, as such, cannot be demonstrated, softening and dissolution of calcareous rock against which the tis- sues are applied has been demonstrated by Kiihnelt ( 1930 ) , Yonge (1955), and Hodgkin (1962). The last named has also demon- strated experimentally that, when placed in a suitable-sized bore hole in mudstone, L. plumula is unable to enlarge it, although it can certainly rotate within it. It is quite certain that Lithophaga cannot bore without some initial chemical softening of the rock. Its own shell is protected against chemical action by the thick, and in- variably unbroken, layer of periostracum. The glands responsible for the softening, probably acid-contain- ing, secretion are situated within the middle mantle fold immedi- ately anterior and posterior to the ligament. Similar glands also oc- cur in Botula and in all Mytilidae. In these mussels they secrete mucus, which entangles particles that fall on these regions (es- HOCK-BORING ORGANISMS "21 pecially posteriorly) when the valves are open. The anterior glands are large only in Lithophaga, where alone the tissues in which they are situated can be protruded. The granular layer on both anterior and posterior shell surfaces in Lithophaga and over posterior regions only in Botula consists of calcareous fragments and other debris (only the latter in Botula) which are consolidated in mucus secreted by these glands ( Yonge, 1955 ) . The presence of acid in the anterior glands — and probably also in the posterior glands, which may assist in the widening of the boring — is confined to Lithophaga. Some species of Lithophaga rotate in the boring, e.g. L. phimula (Yonge, 1955); others do not, e.g. L. cumingiana (Otter, 1937). In the majority of species — there is some degree of variation within the genus — the posterior region of the boring has a secreted cal- careous lining, resembling that described in RoceUaria although not so well developed. This is produced by the walls of the siphons, here consisting of the inner lobe only of the mantle margin, the in- halant siphon being produced by apposition, not fusion, of the ventral margins. In certain species, e.g. L. phimula, the siphons may also be responsible for adventitious posterior extension of the shell valves. Although Botula and Lithophaga doubtless descend from com- mon ancestors, conditions in the former, exclusively mechanical, borers foreshadow those in the latter. With the development of the means of chemical softening of the rock, Lithophaga has lost full efficiency for mechanical boring but has exploited to the full the added powers of boring into calcareous rocks of various kinds. Discussion and Summary Although, as noted in the introduction, the habit of rock boring is unexpectedly widespread, this brief review indicates how many of the borers are not specifically adapted for this mode of life. In many cases boring is facultative, not obligatory. The condition in the plant borers is somewhat obscure, but among the animals boring is not obhgatory in the sponge Cliona, in the echinoids, in Polydora, or, among the bivalves, in Hiatella and the less specialized species of Petricola. All these animals are members of the epifauna, with 22 C. M. YONGE a basic sessile or nestling habit which docs, where conditions are suitable, lead to boring. Really specialized obligatory borers include certain sipunculids and polvchaetes, the carnivorous shell borers including the turbel- larian Psemlosti/Iachus and the prosobranch gastropods (reiving wholly or in part on chemical agencies), the cirriped Lithotrija, and most of the bivalve borers. In the last-named group, where we have fuller information than about an\' other group apart from the carnivorous gastropods, boring is primarily mechanical by means of the valves. The majority grind by lateral and forward movements caused by pressure of water in the mantle cavity aided to greater or lesser extent by the opening thrust of the ligament. In the highly specialized Adesmacea, boring is due to the rocking action of the valves on the median fulcrum of the hinge region caused by the alternate contraction of the anterior and posterior adductor muscles. The ligament is greatlv reduced or lost. In the boring Tridacnidae, where the ligament is not reduced, the animal grinds its way downward by the alternate contraction of the pedal (or byssal) retractor muscles. This causes rocking in the longitudinal plane on the fulcrum of the massive byssal attachment. In the bor- ing Mytilidae, contraction of the posterior byssal muscles causes the anterior end of the shell to bear against the head of the boring, while lateral widening must be due to the opening thrust of the powerful, secondarily extended ligament. But in Lithopliago, con- fined to calcareous rocks, the means of mechanical boring are re- duced, with accompanying modification of preexisting mantle glands to supply the means for chemical softening of the rock. But, as with other boring organisms, apart from the carnivorous gastro- pods, we lack knowledge about the precise manner in which the rock is softened although this presumably involves acid production. Protection has been revealed as the major biological advantage of rock boring. Where borers are numerous, notably on coral reefs, they may be major agents of erosion. Reefs are maintained only by the exceptional powers of calcification possessed by hermatypic (reef-building) corals. Fungi, sponges, flatworms, polychaetes, and drills are all major pests of ovsters. On the other hand, unlike the ROCK-BORING ORGANISMS 23 wood borers, rock borers do little damage to structures erected in sea water, which are usually too hard for them to penetrate. References Cannon, H. G. 1935. On the rock-boring barnacle, Lithotnja valentiana. Sci. Rept. Great Barrier Reef Exped. 1928-29, Brit. Museum (Nat. Hwf.),Vol. 5,pp. 1-17. Carriker, M. R. 1961. Comparative functional morphology of boring mech- anisms in gastropods. Am. Zoologist, 1, 263-266. Doty, M. S., and Morrison, J. P. E. 1954. Interrelationships of the organ- isms on Raroia aside from man. Atoll Research Bull, No. 35. Pacific Science Board. Duerden, J. E. 1902. Boring algae as agents in the disintegration of corals. Bull. Am. Museum Nat. Hist., 16, 323-332. Duerden, J. E. 1905. Recent results on morphology and development of coral polyps. Smithsonian Inst. Misc. Collections, 47, 93-111. Gardiner, J. S. 1903. The Maldive and Laccadive Groups, with notes on other coral formations in the Indian Ocean. Fauna and Geography, Maldive and Laccadive Archipelagoes, 1, 333-341. Gardiner, J. S. 1931. Coral Reefs and Atolls. Macmillan and Co., London. Hodgkin, N. M. 1962. Limestone boring by the mytilid Lithophaga. The . Vcliger, 4, 123-129. Hunter, W. R. 1949. The structure and behaviour of "Hiatella gallicana" (Lamarck) and "H. arctica" (L.) with special reference to the boring habit. Proc. Roy. Soc. Edinburgh, B, 63, 271-289. Korringa, P. 1952. Recent advances in oyster biologv. Quart. Rev. Biol, 27,266-308,339-365. Kiihnelt, W. 1930. Bohrmuschelstudien. I. Palaeohiologica, 3, 53-91. Lloyd, F. E. 1897. On the mechanisms in certain lamellibranch boring molluscs. Trans. N. Y. Acad. Sci., 16, 307-316. Newell, N. D. 1955. Bahamian platforms. Bull. Geol Soc. Am., Spec. Paper, 62, 303-316. Newell, N. D. 1956. Geological reconnaissance of Raroia (Kon Tiki) Atoll, Tuamotu Archipelago. Bull. Am. Museum Nat. Hist., 109, 317- 372. Newell, N. D., and Imbrie, J. 1955. Biogeological reconnaissance in the Bimini area. Great Bahamas Bank. Trans. N. Y. Acad. Sci., ser. 2, 18, 3-14. Newhouse, J. 1954. Floristics and plant ecology of Raroia Atoll, Tuamotu. Part 2. Ecological and floristic notes on the Myxophyta of Raroia. Atoll Research Bull, No. 33. Pacific Science Board. 24 ■ C. M, YONGE Odum, H. T., and Odum, E. P. 1955. Trophic structure and productivity of a windward coral reef community on Eniwetok Atoll. Ecol. Mono- graphs, 25, 291-320. Otter, G. W. 1932. Rock-burrowing echinoids. Biol. Rev., 7, 89-107. Otter, G. W. 1937. Rock-destroying organisms in relation to coral reefs. Sci. Repf. Great Barrier Reef Exped. 1928-29, Brit. Museum {Nat. ///5^), Vol. 1, pp. 323-352. Purchon, R. D. 1954. A note on the biology of the lamellibranch Rocellaria (Gastrochaena) cuneijormis Spengler. Proc. Zool. Soc. London, 124, 17-33. Purchon, R. D. 1955a. The structure and function of the British Pholadidae (rock-boring Lamellibranchia ) . Proc. Zool. Soc. London, 124, 859- 911. Purchon, R. D. 1955b. The functional moiphology of the rock-boring lamellibranch Petricola pholadifonnis Lamarck. /. Marine Biol. Assoc. U. K., 34, 257-278. Seymour Sewell, R. B. 1926. A studv of LitJwtn/a nicoharica Reinhardt. Records Indian Museum, 28, 269-330. Sluiter, G. Ph. 1890. Uber die Bildung der Kalkrohren von Gastrochaena. Natuurkund. Tiidschr. Ned. Indie, 50, 45-60. Umbgrove, J. H. F. 1947. Coral reefs of the East Indies. Bull. Geol. Soc. Am., 58, 129-778. Utseumy, F. 1942. Lime boring algae. South Sea Sci. {Kagaku Nanifo), 5, 123-128. Wainwright, S. A. 1963. Skeletal organization in the coral, Pocillopora damicornis. Quart. J. Microscop. Sci., 104, 169-183. Yonge, G. M. 1936. Mode of life, feeding, digestion and symbiosis with zooxanthellae in the Tridacnidae. Sci. Rept. Great Barrier Reef Exped. 1928-29, Brit. Museum (Nat. Hist.), Vol. 1, pp. 283-321. Yonge, G. M. 1951. Stvidies on Pacific coast mollusks. II. Structure and adaptations for rock boring in Plattjodon cancellatus (Gonrad). Univ. Calif. Puhl. Zool, 55, 401-407. Yonge, G. M. 1953. Mantle chambers and water circulation in the Tridac- nidae. Proc. Zool. Soc. London, 123, 551-561. Yonge, G. M. 1955. Adaptation to rock boring in Botula and Lithophaga (Lamellibranchia, Mytilidae) with a discussion on the evolution of this habit. Quan. J. Microscop. Sci., 96, 383-410. Yonge, G. M. 1958. Observations on Petricola carditoides (Gonrad). Proc. Malacol. Soc. London, 33, 25-31. Yonge, G. M. 1962. On the primitive significance of the byssus in the Bivalvia and its effects in evolution. /. Marine Biol. Assoc. U. K., 42, 113-125. Boring Sponges as Controlling Factors in the Formation and Maintenance of Coral Reefs THOMAS F. GOREAU, Department of Physiology, University of the West Indies, Kingston, Jamaica WILLARD D. HARTMAN, Peabody Museum of Natural History and Department of Biology, Yale University, New Haven, Connecticut TROPICAL reef communities harbor a diverse biota of indwelling plants and animals which destroy corals, shells, and limestone (Otter, 1937). Verrill and Smith (1873) suggested that the break- down of insoluble skeletal carbonates by marine borers is a signifi- cant factor in the calcium balance of the seas. Gardiner (1903, 1931 ) stated that boring algae, mollusks, and worms play important roles in the breakdown and erosion of atoll reefs, whereas Yonge (1930) believed the most obvious agents of reef destruction to be the bivalve mollusks, Ginsbin-g (1957) affirmed that the weakening of reefs by organic action makes them more susceptible to wave erosion, and cited boring sponges as examples. Until the present, the destruction of reefs by burrowing organisms has been studied only in extreme shallow-water environments where the effects of boring are masked by rapid biological calcification and energetic wave attrition. However, a series of recent investiga- tions carried out in Jamaica on the coral communities of the steep 25 26 T. F. OORKAIT AND W. D. HARTMAN Figures 1 to 7 CONTROL or CORAL REEFS BY BORING SPONGES 27 fore-reef slope habitat at depths between 30 and 70 meters drew our attention to tlie existence of major imdescribed erosional fea- tures which are determined by the activity of boring sponges under conditions where CaCOa deposition in corals is not so fast as at the surface and the effect of wave turbulence is normally negligible. In such localities the consequences of sponge boring go far beyond the mere hollowing out of a few cavities: our observations show that the ecologv of the reef coral population, the movement of reef sediments, and the modeling of the reef edge are all strongly in- fluenced. Nassonow's important work (1883) showing that clionid sponges produce fine calcareous debris in the course of their boring ac- tivities has been generallv overlooked, and more recent authors (Revelle and Fairbridge, 1957; Ginsburg, 1957; Cloud, 1959) assert that sponges bore by chemical action and imply that all the ex- cavated carbonate is removed by solution. The observations of Nas- sonow and others suggest, however, that the burrowing sponges may be of major significance in the reduction of reef limestone to verv fine detritus. Fig. 1. Incurrent and excurrent papillae of boring stage of Cliona cclata Grant. Oscule in vipper right. (X 3.6.) (From Hartman, 1958.) Fig. 2. Massive, free-living stage of Cliona celata Grant. (X 0.36.) (From Hartman, 1958.) Fig. 3. Broken edge of a colonv of the coral Agaricia showing extensive excavation bv Cliona tampa de Lau'benfels. (X 0.9.) 38 meters, Maria Buena Bay, Jamaica. (Photo by T. F. Goreau.) Fig. 4. Surface of colonv of Agaricia showing incurrent and excurrent papillae of Cliona lampa de Laubenfels. An expanded incurrent papilla may be seen in the upper right; papillae with central holes are excurrent. (X 1.2.) 38 meters, Maria Buena Bay, Jamaica. (Photo by T. F. Goreau.) Fig. 5. Broken edge of a colony of Montastrea annularis (E. and S.) ex- cavated by a clionid. Beneath the coral polyp layer at the top is a layer of boring algae, seen as a dark line. The boring sponge has died and has left extensive galleries in the coral. (X 0.9.) 38 meters, Maria Buena Bay, Jamaica." (Photo by T. F. Goreau.) Fig. 6.' Larva of Cliona spreading out over a fragment of oyster shell. Lines etched into the calcareous matter are clearly seen. (From Nassonow, 1883.) Fig. 7. Pattern of depressions left in an oyster shell after the removal of calcareous chips by a clionid. (From Nassonow, 1883.) T. F. GOREAU AND W. D. HARTMAN -Ostium ncurrent Channel Calcareous Substrate '^J \iii^>wi^)), a tissue completelv devoid of even proto- plasmatic extensions of any antecedent cellular elements. The den- tin, which undergoes typical lacunar resorption both in normal shedding and under pathological conditions of internal resorption ("pink tooth"), has no connective tissue cells within its matrix, albeit permeated by long cytoplasmic processes from the pulpal odonto- blasts. In the case of cementum there are to be sure cells embedded in the widest periapical zone of secondary cementum (cemento- cytes ) , but the primary cementum is completelv acellular and yet in no way spared from the same resorbabilitv in the presence of typi- cal multinucleated giant cells which obviouslv must originate from the adjacent environment (e.g. enamel resorption, see Fig. 30e). DENTAL HARD TISSUE DESTRUCTION 97 As a second aspect of general significance, there appears to be no common link between the conditions under which these gener- ally protected dental hard tissues succumb to resorption. In no case — even in physiological shedding — can one definitely point to a generalized systemic influence, such as is found in the case of deer antler resorption, where a definitive sexual cycle is involved (see chapter 13 by Goss). Unlike antlers, the teeth of man resorb one after another, summer and winter, over a long period of time be- tween the ages of 6 and 13. But from a local environmental point of view, the shedding teeth share with the shedding antlers a juxtaposition to a highly vascularized pad of granulation tissue studded with giant cells which clearlv must originate from the ad- jacent connective tissue environment. Even in pathological tooth resorption — be it internal (pulpal) or external (periodontal), of local origin or caused by metabolic bone disease (hyperparathy- roidism), traumatic occlusion, or radiation injury — one topically finds proliferation of highly vascularized granulation tissue loaded with giant cells in immediate juxtaposition to the resorbing ce- mentum, dentin, or enamel, as the case may be. Finally, there are certain important differences in the structural and chemical nature of the three dental hard tissues, cementum, dentin, and enamel, which may help to clarify the question as to whether or not the basic phenomenon of resorbability is dependent to any significant degree upon the quality of the structures being resorbed. In general, it may be said that the prevailing concepts of hard tissue biology hitherto have been largely evolved from studies of bone tissue characterized by minute inorganic crystals and a col- lagenous organic framework — whether the concepts deal with the process of resorbability or of calcifiability (Sognnaes, 1960fl). But in the dental apparatus we are faced with a less than satisfactory ex- planation for the deposition and removal of the large inorganic crystals contained within the inadequately identified organic matrix of enamel, a product of ectodermal cells, recently classified as a cross-;8-linkage protein (Glimcher et ah, 1961). At best this is a sparsely distributed organic framework. Thus, irrespective of its chemical nature, the quantity is so low as compared with the or- 98 R. F. SOGNNAES ganic matter in other resorbable tissues that one is almost forced to disregard the basic role of inherent organic components within the tissues being resorbed — as we have already disregarded the role of interstitial cellular components — in the basic process of resorbability. In conclusion, then, it is believed that the general significance of dental resorption phenomena rests on the observation that resorb- ability must be largely controlled by the consistent presence of a highly vascularized adjacent tissue environment rather than by some peculiarly consistent feature characterizing the structures that are being resorbed, whether these be enamel, dentin, ivory, cemen- tum, or bone, in various states of calcification or with various chemi- cal and physical characteristics of the organic and inorganic constit- uents (Sognnaes, 1955, 1960fl, 1960fo). Dental Caries The problem of dental caries, the most universal of all human ills, is being examined in several other chapters of this volume, from the point of view of microstructure (Darling, chapter 6), ultra- structure (johansen, chapter 7), chemistry (Gray and Francis, chap- ter 8), and bacteriology (Keves and Jordan, chapter 9). As a preamble, however, to the principal theme of the present chapter — in vivo, in vitro, and postmortem erosions — it seems im- portant to recognize several characteristic features of dental caries which are in sharp contrast to other types of hard tissue destruction in general and to other types of dental pathology, notably dental erosion, in particular (Sognnaes, 1959, 1962). In the first place, dental caries is characterized by a deep sub- surface demineralization ( up to 1000 microns deep ) prior to collapse of the tooth structure (i.e. prior to extensive proteolysis and "cavity" formation). Secondly, the pathway of this process is intimately re- lated to certain preformed structural patterns within the dental hard tissues themselves. These are, in the enamel (a) the incre- mental lines of Retzius, (b) the interprismatic substance, and (c) the intraprismatic cross striations of the individual enamel prisms (these zones are probably related; Sognnaes, 1949); and in the den- tin (a) the normal growth rings of von Ebner, (b) the Owen's contour lines (probably representing less than adequate calcification DENTAL HARD TISSUE DESTRUCTION 99 to begin with), and (c) the pathways of the dentinal tubules, which contain protoplasmatic extensions from the pulpal odontoblasts and may be subject to secondary calcifications, decalcifications, and bacterial invasion as a late feature of caries. As will be noted below, these manifestations of tooth destruction in caries are significantly different from those occurring in dental erosion in vivo and in vitro, on the one hand, and those occurring during postmortem destruction of the teeth, on the other. Dental Erosion There are numerous reports in the literature regarding erosion of teeth caused bv various chemical agents with and without concur- rent abrasion caused by mechanical friction. Stafne and Lovestedt ( 1947 ) have observed a number of cases in which dissolution of tooth substance was caused bv various kinds of acids, juices, soft drinks, etc. Similar eftects have been observed as a result of certain medicaments (James and Parfitt, 1953), thera- peutic use of dilute hydrochloric acid (Stafne, 1933), general hy- drogen ion concentration (Elsburv, 1952), long-continued vomiting accompanying obstipation (Bargen and Austin, 1936), diabetes insipidus (Finch, 1957), excessive consumption of acidified candies (West and Judy, 1938), miscellaneous natural juices (Gortner and Kenigsberg, 1952), acidified beverages (McClure, 1943; McCay and Will, 1949; Restarski et ah, 1945; Gortner et al, 1945), various types of fruit drinks (Miller, 1950; Holloway et al., 1958), dietary oxalate (Gortner et al., 1946), citrate and lactate (McClure and Ruzicka, 1946). These reports also contain observations on partial control of such erosions by various buffering agents, fluoride, etc. (Restarski et al., 1945; Holloway et al., 1958) as well as many cross references to other literature not cited here. The degree to which abrasive action is involved in the progress of erosion-like lesions of the teeth is not fully understood. As a rule the lesions referred to above have the appearance of rather diffuse dissolution, whereas tvpical abrasion and atypical erosion (idio- pathic, see below) mav ha\'e a more definitive pattern depending upon the location of the frictional and or chemical influences. 100 R. F. SOGNNAES A most interesting pattern of erosion-like defects has recently been reported by Rost and Brodie (1961), whose observations sug- gest that mechanical friction without any chemical action had caused localized destruction, otherwise more typical of erosion. For example, in several cases where amalgam or gold fillings had been placed for the express purpose of repairing such lesions, the eroded pattern was subsequently found on the fillings themselves. In one case the erosion lines were found within 8 months on the surface of a plastic (methyl methacrylate ) filling which had been placed in the center of the labial surface of an upper central incisor for the repair of a small congenital defect. In other cases delicate lines could be seen in the plastic base of dentures. The authors suggested that a possible cause of these lines might be some "hyperactivity" of the soft tissue environment of the structures involved, whether natural or artificial, and that such environmental abrasion might be one of the causes of the lesions observed. Since the plastic material would be resistant to any chemical action encountered in the mouth, one must conclude — in the opinion of this writer — that here are cases in which physical forces have been at work, possibly a twitch- ing of soft tissue elements rubbing against the plastic material or perhaps even streams of saliva continuously running like miniature rivers in confined areas. There can be no question that the hard tissues of the mouth are susceptible to destructive actions of mechanical or chemical nature or both. Hence one must expect to encounter mechanical abrasion and chemical erosion as well as combined erosion-abrasion of dual etiology. Calcium phosphate is obviously soluble in calcium-binding chemical agents, whether these be of exogenous or endogenous origin, as long as the agents are present sufficiently long, in adequate quantities, and in susceptible locations. Thus it is not surprising, as noted above, that chemical erosion of rats' teeth occurs after pro- longed ingestion of what in human equivalents would be enormous quantities of acidic beverages. Similarly, it is to be expected that frictional forces will wear away the dental hard tissues. Perhaps it is more surprising that the dental organs can in fact last the pro- longed lifetime of modern man without being either dissolved or DENTAL HARD TISSUE DESTRUCTION 101 worn away in the face of the many known injurious influences within the oral environment. There is, however, a more obscure type of dental hard tissue destruction that needs to be discussed in greater detail, namely the pecuhar wasting of tooth substance which, for the moment, may be best classified as "idiopathic erosion," The following subsections will consider this problem from the point of view of ( 1 ) gross dis- tribution, (2) microradiography, (3) histopathology, (4) tooth structure, (5) oral environment, (6) general constitution, and (7) clinical management. Gross Distribution The commonly held concept that dental erosion may be largely attributed to mechanical friction from tooth brushing became widelv accepted following a series of studies by W. D. Miller, published in 1907. He based his conclusion on the observation that dental erosion is much more common in modern than in ancient man. Out of a total of 36,000 ancient skulls surveyed, none were said to have shown typical lesions of dental erosion, whereas among twice as many contemporarv individuals dental erosion was found, and was claimed to be confined to those who had become introduced to the use of a toothbrush. In retrospect, it is somewhat surprising how widely this circum- stantial evidence was to be accepted, conflicting clinical experience notwithstanding. Indeed, even earlier writings on the subject had emphasized that the topographic configuration and distribution of dental erosion frequenth' failed to conform to the destructive char- acteristics that would be expected from mechanical friction of a toothbrush. Figure 1 illustrates this point. Some of these photo- graphs are reproductions of plaster casts from patients seen in the private practice of Dr. Maurice Peters, who, in addition, has very kindly provided some of the specimens that have been further ex- plored in my microscopic follow-up studies reported below. It will be noted from the nine cases shown in Fig. 1 that a con- siderable variation can exist in location and configuration of the lesions. In some instances the lesions are located near the incisal 102 R. F. SOGNNAES Fig. 1. Variability in the gross morphological distribntion ot dciita] erosion: A, B, narrow horizontal grooves; C, D, E, shallow, flat, and disc-shaped; F, irregular horizontal and vertical erosion separated by intact tooth; G, figured; H, I, circumscribed, wedge-shaped, labial (H) and lingual (I). For details see text. edge, in others near the gingival margin, in others at both sites. In some mouths there are combinations of lesions running horizontally and vertically, and yet separated by completely unaffected teeth (Fig. IF). This would be extremely difficult to account for on the basis of tooth brushing. Occasionallv extreme lesions can be seen lingually in the gingival area of the lower anterior teeth, a location completely incompatible with a tooth-brushing etiology. In one such case, shown in Fig. 1 1, the erosion ultimately progressed so far that the destruction circumscribed the whole cervical area, both lingually, interproximally, and labiallv, so as to produce an hour- glass appearance interestingly similar to a natural rock "erosion" ( see Fig. 2 with insert ) . The several examples of dental erosion shown in Fig. lA to I con- form (from the point of view of gross topography) to the classical description of G. V. Black, as retained in later textbooks (Black, 1936 ) , including narrow horizontal grooves (A, B ) , disc-shaped and DENTAL HARD TISSUE DESTRUCTION 108 Fig. 2. Rock "erosion," curiously similar to certain circular erosions of human teeth (insert). (Reproduced from "The Toby-jug Stone — Kannesteinen near Maal0y, Norway," with the permission of Mittet Foto A/S, Oslo, Norway.) In the insert is shown the ultimate "hourglass" appearance of one of the mandibular incisors seen in Fig. 1 I at an earlier stage of erosion. At the time of extraction, the erosion process had extended all around the lingual, labial, mesial, and distal surfaces. 104 R. F. SOGNNAES flattened (C, D, E), irregular (F), figured (G), circumscribed, and wedge-shaped (H, I). Though tooth brushing had been considered an etiological factor in erosion prior to the studies of Miller, it is noteworthy that several years before the turn of the century, Kirk (1887) emphasized that erosion could be found even in people who only brushed occasion- ally, sporadically, or not at all. Further, he noted, as we have amply illustrated above (Fig. 1), that there could be a great variation in the intraoral orientation of the lesions on difi^erent teeth. He found, furthermore, that people who brushed horizontally could neverthe- less have lesions of erosion in the form of vertical grooves, that eroded teeth could be separated by non-eroded ones, and that ero- sion could occur, as we have noted, in teeth which were not acces- sible to the toothbrush at all. Though it is true that typical dental erosion may ultimately as- sume a very wedge-shaped appearance with sharp angles, this is actually an advanced stage of pathology. In an earlier stage, we Fig. 3. Segment of labial surface of incisor with beginning erosion of enamel (upper part) passing through dentin-enamel junction at one point and, near cementum-enamel junction (below), producing deep wedge into the dentin. (X 12.) Fig. 4. View of outer enamel surface removed from top of segment seen in Fig. 3. To the left of the erosion grooves, prior to gross loss, is seen a slight chalklike change in the opacity of the enamel. ( X 12.) DENTAL HARD TISSUE DESTRUCTION 105 have noted that the lesions, when observed grossh' in tooth seg- ments hke those shown in Fig. 3, may have a multiple, rather rounded, "lacunar" profile, afl^ecting similarly both enamel and den- tin. The peripheral surface of the beginning lesion in enamel is ex- tremely hard to detect with the naked eve. With some magnifica- tion, however, as shown in Fig. 4, one may occasionallv detect in the enamel surface, just outside the grossly visible defect, a very slight opacity or chalkiness. Clinically one would be unlikely to diagnose this important incipient stage, because to detect this, the tooth surface would have to be extremely clean, thoroughly dehy- drated, and inspected with a lens under ideal lighting. We shall later discuss the structural basis for this appraisal. Suffice it to say, from the point of \'iew of gross appearance, that if a beginning, posteruptive, chalky lesion in the enamel surface is seen readily with the naked eve prior to gross destruction, one can be reasonably sure that it is probably not erosion, but the beginning of caries (Sognnaes, 1940). X-Raij Microscopij It would seem a foregone conclusion that if dental erosion were caused purely by mechanical friction, then the surface of the lesion should exhibit a simultaneous and complete removal of all ingre- dients of the tooth substance, inorganic and organic. Unfortunately, there is limited knowledge regarding the precise microscopic sequence of events in erosion. Textbooks on the pathol- ogy and histopathologv of the teeth do not include any documentary evidence to indicate that the conclusions have been based on micro- scopic examinations. Indeed, it would appear that the charac- teristics of dental erosion mentioned, for example, in Kronfeld's Histopathologi/ of the Teeth (re-edited by Boyle, 1955) are, in fact, based on the classical gross description of the lesions, stating that the lesions have "sharply outlined borders" and that the "floor is clean, hard and smooth in appearance." This is not to say that information is readily obtained by routine histological examination. Neither ground sections nor decalcified sections examined in the optical microscope would in themselves Figures 5 to 12 106 DENTAL HARD TISSUE DESTRUCTION 107 necessarily reveal gradients in demineralization near the surface. Fortunateh', howe\er, through the recently refined method of inicroradiograph\-, utilizing soft x-ra\s, between 5 and 30 kv, and a yery fine-grained photographic emulsion (Kodak spectroscopic plates, 649-GH), it is now possible to produce yery accurate x-rays of thin sections of bones and teeth which can be enlarged photo- graphically under the microscope up to 1000 times. The accuracy of this method can be as good as 1 micron, making it possible to detect differences in microdensity and potential demineralization gradients near the surface of dental erosions. Our own microradiographic obseryations haye so far primarily been made on the superficial areas in lesions extending into the dentin, as exemplified in Fig. 5. In the bottom of these typical wedge-shaped lesions, we haye been able to demonstrate that there can occur a change in the surface of the tooth substance, charac- terized by a gradient in the microdensity and indicating that de- mineralization may occur to a maximal depth of 100 microns (Fig. 6). This, it should be noted (as pointed out elsewhere: Sognnaes, 1959), is less than one-tenth of the demineralization gradient which can be observed in dental caries. In other words, in erosion the alterations are strictly confined to a surface effect. Occasionally the microstructure of the surface may exhibit a some- what uneven configuration ( Figs. 7 to 9 ) , but even then there is no evidence of the deep destruction along the dentinal tubules so characteristic of caries. At other times, in certain parts of the lesion, the surface mav be very smooth (Fig. 10), with little or no evidence Figs. 5 to 12. Microradiographs prepared by placing ground sections on fine-grained spectroscopic plates (Kodak 649-GH), exposed to soft x-rays (25 kv). In Fig. 5 the deep wedge-shaped buccal erosion cavity appears to exhibit no remarkable surface changes at low magnification (x 4). Figures 6 to 10 show at 100 times magnification that the erosion surface may exhibit a periph- eral demineralization gradient. Figure 6, enlarged from bottom of wedge- shaped defect in Fig. 5, shows that this demineralization can reach a maximal depth of 100 microns. At certain stages the erosion lesions can be slightly irregular (Figs. 7 to 9), and at others quite smooth and dense as though there had been a redeposition of minerals in the periphery (Figs. 6 and 10). By comparison, mechanical abrasion, whether produced in vivo (Fig. 11) or in vitro (Fig. 12), shows verv smooth lesions with no subsurface loss in microdensity. 108 R. F, SOGNNAES of a detectable demineralizatioii gradient. Indeed, the very periph- ery of certain parts of the lesion can at certain stages show a somewhat elevated microdensity (Fig. 6). It is qnite possible that the lesion mav assume a transitionallv chronic stage in which there may be some slight redeposition of minerals in the organic matrix of the previously demineralized tooth substance. For comparative purposes, we have examined lesions believed to represent uncomplicated abrasion in vivo (Fig. 11) as compared with destruction produced in vitro by mechanical friction (Fig. 12), For the latter purpose, extracted teeth were placed in a so-called tooth-brushing machine (courtesy of the Miami Vallev Laboratories of the Procter & Gamble Company, Cincinnati, Ohio), and exposed to brushing with a tooth paste slurry for a period of time equivalent to years of tooth brushing under practical conditions. The surfaces of these lesions were similar in vivo and in vitro (Figs. 11 and 12), exhibiting a smooth contour without anv evidence of subsurface demineralization. Admittedly a surface which has become demin- eralized, even though to a slight depth, could well be susceptible to further loss by mechanical friction, if within the accessible range of rigorous tooth brushing. In that case, this would be a secondary factor, assuming that our observation of a superficial demineraliza- tion gradient is indicative of some other chemical agent which, as a primary factor, would tend to demineralize the tooth surface. Histopathologij Following the obsei^vation that dental erosion can exhibit a surface demineralization, at least at certain stages, it seemed of interest to explore more precisely, in stained decalcified sections, whether or not the lesion mav be exhibiting anv other charac- teristics which might explain this peculiar peripheral pattern. Paraffin sections were prepared following decalcification accord- ing to the method of Morse ( 1945 ) . From a microscopic point of view, the active lesions did not always look as "clean" as would appear from gross appearance. Thus, we noted a thin zone of organic material attached to portions of the eroded surface, as illustrated in Figs. 13 to 18. In some instances (Figs. 13 to 15) the tooth surface was very straight and smooth, even though over- DENTAL HARD TISSUE DESTRUCTION 109 lain by a thin mucous plaque containing microorganisms. In such cases we have noted a \'erv thin cuticular meml^rane separating the microorganisms from the tooth surface proper. At higher magnification (Fig. 17) this thin membrane may best be seen when stained with toluidine blue, with which it stains green, that is orthochromatically, in contrast to the bacteria-laden mucous film, which is intensely metachromatic with this dye. In other instances (Fig. 16) a metachromatic mucous plaque laden with bacteria appears to be in direct contact with the tooth surface without being separated by the orthochromatic organic film. Under this circumstance a superficial destruction of the tooth substance may be visualized. At higher magnification (Fig. 18) it is note- worthy that the surface destruction is primarily occurring in the matrix proper between the dentinal tubules. This is in sharp contrast to the bacterial invasion of the teeth in caries, in which the dentinal tubules become invaded at an early stage and to considerable depth prior to intertubular matrix destruction. Figure 18 illustrates this characteristic feature of erosion. The dark material represents bac- teria situated in a mucous film, and it will be noted that the inter- tubular matrix is becoming destroyed without any widening of the dentinal tubules. The possible bacterial implication in the etiology of dental ero- sion, suggested by these findings, does not appear to have received serious attention in the literature. This is perhaps not surprising in view of the fact that dental erosion usually looks so extremely clean to the naked eye. It must be remembered, however, that what is seen in the microscopic sections represents an extremely thin film, and since we have noted its relative absence in some areas of the lesions as compared with others, this film may be transitory and not always present. The basic issue raised by these observations is whether or not the pathogenesis of erosion, albeit clearly different from that of caries (Sognnaes, 1959), is of inherently dental or oral environ- mental origin. Some peculiar characteristic of the tooth substance itself could theoretically invite destruction of the intertubular ma- trix along a surface area rather than invasion through the dentinal tubules in depth. On the other hand, the superficial environmental R. F. SOGNNAES IM(,1 i- I » £b a. o (T ? si Q. V) : oi O < O X if) Q. U o 2 O O ^' o O UJ ^. < (/) o 1 _l UJ _J < a -Z. o k ' i= o ^ \1 *^ yr 5 X: Ul CC CD 5 \- < -I _i"jr. ^2 ** < t; X ' o o .9 >tr S2 CHEMI PHYSI (Microb 4 < if) 1 ^ _i -. 1 lu H^^J • < < < c M^ o CD 3 1 1 I o T3 O • I 1 Dep Min With -o- o o CD •- ^ r- :t X ^ '^ D o 4—* +-* cu , ^ -^ o ' c: n U1 o C/5 -M m 0) to c3 c •"• rt on M-H CD )0 'A a> >, iT^ S 1— I o CU f- lU q=l rt C3 o bO hr bJCTS n c3 ^ (■^ Q o f2 *-' QJ O 1— 1 o (_i U^ CO ") n3 o f^ 11 \M C/2 ■-7-1 ,15 > DENTAL HARD TISSUE DESTRUCTION 147 Dental Caries The pattern of the initial carious lesion is intimately related to certain structural pathways within the dental hard tissues that are exposed to the external environment of the mouth. The destructive process is pathologicallv specific. One of the most distinctive micro- structural features of dental caries is a deep subsurface demineraliza- tion of about 1000 microns, whether it be enamel (Fig. 30A) or dentin (Fig. SOB), prior to invasion of the tissues by oral micro- organisms (Fig. 30a) and proteolvtic disintegration of the organic matrix (Fig. 30b). In this respect, dental caries is markedly different from other tvpes of hard tissue destruction (see Fig. 31), in which subsurface demineralization and preformed structural pathwavs do not appear to be b\' far so significantlv involved in the pathogenesis of the lesions. The relativelv higher resistance to demineralization observed within a narrow surface zone of incipient enamel caries ( Fig. 30A ) ma\^ be explained in part by the higher fluoride content now known to be present in this zone and in part by what has been suggested to be a protective interaction of organic surface films de- posited on the inorganic phase of the tooth substance from the adja- cent liquid phase of saliva. Dental Erosion The gross distribution and morphological variety of idiopathic dental erosion defects preclude a simple mechanism of mechanical wear as a primary cause of this disease. Microradiographic observations of ground sections have shown at certain stages a minute subsurface demineralization gradient (Fig. 30C) never seen to exceed 100 microns in depth (i.e., less than one- tenth of what can be found in incipient caries ) . Consequently, one cannot, in beginning erosion, detect any change in opacity by naked- e\e inspection so as to distinguish the lesion from simple abrasion, in which there is no subsurface demineralization at all (Fig. 30D). Histological examination of stained decalcified sections suggests that dental erosion is not necessarily de\'oid of surface deposits of bacteria-laden mucus (Fig. 30C), though the lesion may look clean to the naked e\'e. Occasionalb, these superficial deposits may be 148 R. F. SOGNNAES separated from the tooth surface by a thin cuticular membrane, which stains orthochromatically with tokiidine bhie, presumably originates from sahva, and possibly may be of protective significance. Tentatively, one may suggest that a three-stepped sequence of events may be involved in the pathogenesis of "idiopathic" dental erosion: (1) primary absence or secondary loss of some protective salivary organic coating on the tooth surface; (2) drainage of min- erals from the unprotected and, hence, more soluble peripheral tooth substance by some decalcifying agent present in or brought into contact with the oral environment; and ( 3 ) destruction of this super- ficially decalcified tooth surface through biochemical and biophysical influences or even simple mechanical friction by lip, cheek, tongue, food, or toothbrush. Resorption On the basis of what has been learned from dento-alveolar resorp- tion processes (Fig. 30E and F), one cannot point to the content and distribution of cellular elements (Fig. 30d and e) or to any consistent proportion of inorganic and organic ingredients to ac- count for the origin of the multinucleated osteoclasts, on the one hand, and the susceptibility to resorption on the other. It is sus- pected, therefore, that changes in the environment adjacent to the disintegrating surfaces must be explored for factors that induce the virtually simultaneous removal of organic and inorganic elements of these tissues in the process of resorption (Fig. 31). Postmortem Destruction Postmortem changes have been observed in human teeth of pre- historic, ancient, and recent times originating from Palestine, Egypt, Greece, Iceland, Norway, and Central America. The histological ob- servations indicate that the commonest postmortem change consists in large irregular boring canals which penetrate the dentin and cementum, leaving sharply defined margins with no evidence of any gradients in demineralization (Fig. 30G). The microscopic pattern of the postmortem destruction is suggestive of an invasion by fungi (Fig. 30f). Postmortem changes were rarely found in the enamel and were limited to localized areas of erosion of the enamel surface. DENTAL, HARD TISSUE DESTRUCTION 149 Remineralization The theoretical concept of redeposition of minerals in partially demineralized enamel has been reappraised on the basis of recent observations on radioisotope exchange reactions between enamel and saliva in primate teeth. Experimental application of acid to the enamel surface of teeth in vivo resulted in demineralization in depth beyond what can be detected by gross inspection or by surface rep- lica techniques. In the living animal the demineralization has a definite preferential pattern along the incremental lines of Retzius before grossly visible loss of surface contour. In the normal animal these areas of demineralization appear to be prone to redeposition of minerals, demonstrable within a week or two. However, despite the regained high microdensity, these remineralized areas appear to contain an organic substance which has a greater dye-binding ca- pacity than intact mature enamel. It is believed that enamel in which minerals are redeposited following experimental demineraliza- tion is not completely reconstituted to its original form, and that it would be inappropriate to label this mechanism as a "genuine healing process." Acknowledgments. This research was supported by a general Institu- tional Research Grant from the National Institutes of Health, and by a Project Research Grant (D-1413) from the National Institute of Dental Research for "Further Studies on Hard Tissue Destruction." References Andresen, V. 1926. TJie Physiological and Artificial Mineralization of the Enamel. Einar Dancke, Oslo. Badanes, B. B. 1930. Sources and clinical manifestations of the oxalates as related to dental caries and erosion. Dental Cosmos, 72, 477-490. Bargen, J. A., and Austin, L. T. 1936. Obstipation with long continued vomiting: Effect on the teeth. Am. J. Digest. Diseases, 3, 198. Barnicoat, C. R. 1957. Wear in sheep's teeth. New Zealand J. Sci. and Technol, 38A, 583-632. Barnicoat, C. R. 1960. Attrition of incisors of grazing sheep. Nature, 185, 179. Bird, C. K. 1931. Erosion and abrasion of natural teeth: The remedy or correlation of these conditions. Dental Cosmos, 73, 1204-1208. 150 R. F. SOGNNAES Black, A. D. 1936. Opcnithc Dentistry, \o\. 1. Medico-Dental Publishing Co., Chicago, 111. Bodecker, C. F. 1933. Denial erosion: Its possible causes and treatment. Dental Cosmos, 75, 1056-1062. Bodecker, C. F. 1945. Local acidit}': A cause of local erosion-abrasion. Ann. Dentistry, 4, 50-55. Bodecker, C. F. 1953. Again, erosion-al:)rasion (editorial). New York State Dental J., 19, 154-156. Boyle, P. E. (editor). 1955. Kronf eld's Histopathology of tlie Teeth, 4th edition, p. 83. Lea and Febiger, Philadelphia, Pa. Chievitz, O., and Hevesy, G. 1936. Radioactive indicators in the study of phosphorus metabolism in rats. Nature, 136, 754-755. Darby, E. T. 1892. Dental erosion and the gouty diathesis: Are they usually associated? Dental Cosmos, 34, 629-640. Elsbury, W. B. 1952. Hydrogen-ion concentration and acid erosion of the teeth. Brit. Dental j., 93, 177-179. Ericsson, Y. 1953. Investigations on the occurrence and significance of citric acid in the saliva. /. Dental Research, 32, 850-858. Finch, L. D. 1957. Erosion associated with diabetes insipidus. Brit. Dental ]., 103, 280-282. Frank, R. M., and Sognnaes, R. F. 1960. Electron microscopy of matrix formation and calcification in rat enamel. Arch. Oral Biol, 1, 339-348. Frank, R. M., Sognnaes, R. F., and Kern, R. 1960. Calcification of dental tissues with special reference to enamel ultrastructure. In Calcifica- tion in Biological Sy.ste7ns (R. F. Sognnaes, editor), pp. 163-202. American Association for the Advancement of Science, Washington, D. C. Glimcher, M. J., Bonar, L. C, and Danial, E. J. 1961. The molecular structure of the protein matrix of bovine dental enamel. /. Mol. Biol., 3, 541-546. Gortner, R. A., and Kenigsberg, R. K. 1952. Factors concerned with differ- ent erosive effects of grapefruit and grapefruit juice on rats' molar teeth. /. Niitr., 46, 133-140. Gortner, R. A., Jr., McCay, C. M., Restarski, J. S., and Schlack, C. M. 1946. Some effects of dietary oxalate on teeth of white rats. /. Nutr., 32, 121-131. Gortner, R. A., Jr., Restarski, J. S., Bieri, J. G., and McCay, C. M. 1945. Factors influencing destructive effects of acidic beverages on teeth of white rats and hamsters. Arch. Biochem., 8, 405-414. Hill, J. T. 1949. Oral Pathology, 4th edition. Lea and Febiger, Phila- delphia, Pa. Holloway, P. J., Mellanby, M., and Stewart, R. J. C. 1958. Fruit drinks and tooth erosion. Brit. Dental J., 104, 305-309. DENTAL HARD TISSUE DESTRUCTION 151 Hopewell-Smith, A. 1918. Normal and Pathological Histology of the Mouth, Vol. 2, p. 34. P. Blakiston's Son and Co., Philadelphia, Pa. James, P. M. C, and Parfitt, G. J. 1953. Local effects of certain medica- ments on teeth. Brit. Med. J., 2, 1252-1253. Kirk, E. C. 1887. A contribution to the etiology of erosion. Dental Cosmos, 29, 50-58. Kronfeld, B. 1932. Preliminary report of clinical observations of cervical erosions, a suggested analysis of the cause and the treatment for its relief. Dental Items Interest, 54, 90.5-909. McCav, C. M., and Will, L. 1949. Erosion of molar teeth by acid beverages. J.Nittr., 39, 313-324. McClure, F. J. 1943. The destructive action, in vivo, of dilute acids and acid drinks and beverages on the rats' molar teeth. /. Niitr., 26, 251. McClure, F. J., and Ruzicka, S. J. 1946. Destructive effect of citrate vs. lactate ions on rats' molar teeth surfaces in vivo. J. Dental Research, 25, 1-12. Miller, C. D. 1950. Enamel erosive properties of fruits and fruit juices. /. Ntitr., 41, 63-71. Miller, W. D. 1907. Experiments and observations on the wasting of tooth tissue variously designated as erosions, abrasion, chemical abrasion, denudation, etc. Dental Cosmos, 49, 1-23, 109-124, 225-247. Morse, A. 1945. Formic acid-sodium citrate decalcification and butyl al- cohol dehydration of teeth and bones for sectioning in paraffin. /. • Dental Research, 24, 143-153. Piez, K. A., and Likins, R. C. 1960. The nature of collagen. II. Vertebrate collagens. In Calcification in Biological Systems ( R. F. Sognnaes, editor), pp. 411-420. American Association for the Advancement of Science, Washington, D. C. Restarski, J. S., Gortner, R. A., Jr., and McCay, C. M. 1945. Effect of acid beverages containing fluorides upon the teeth of rats and puppies. /. Am. Dental Assoc.: 32, 668. Rheinwald, V., and Staehle, G. 1949. Untersuchungen zur Frage der Remineralisation des Zahnschmelzes. Stoma, 2, 259-314. Rost, T., and Brodie, A. G. 1961. Possible etiologic factors in dental ero- sion. /. Dental Research, 40, 385. Roux, W. 1887. tjber eine im Knochen lebenden Gruppen von Faden- pilzen. Z. wiss. Zool., 45, 227. Rushton, B. 1957. Cosmetic reshaping of the natural teeth. Intern. Dental /., 7 (3), 403-404. Schaffer, J. 1895. Bemerkungen zur Geschichte der Bohrkanale im Knochen und Ziihnen. Anat. Anz., 10, 459-464. Schatz, A., Karlson, K. E., Martin, J. J., and Schatz, V. 1957. The pro- teolysis-chelation theory: Applications tc caries and erosions. Ann. Dentistry, 16, 37-49. 152 R. F. SOGNNAES Shulman, E. H., and Robinson, H. B. G. 1948. Salivary citrate content and erosion of teeth. /. Dental Research, 27, 541-544. Sognnaes, R. F. 1940. The importance of a detailed clinical examination of carious lesions. /. Dental Research, 19, 11-15. Sognnaes, R. F. 1941. Eflfect of topical fluorine application on experi- mental rat caries. Brit. Dental J., 70, 433-437. Sognnaes, R. F. 1947. Preparation of thin "serial" ground sections of whole teeth and jaws and other highly calcified and brittle structures. Anat. Record, 99, 133-144. Sognnaes, R. F. 1949. The organic elements of the enamel. III. The pattern of the organic framework in the region of the neonatal and other incremental lines of the enamel. /. Dental Research, 28, 558-664. Sognnaes, R. F. 1955. Microstructure and histochemical characteristics of the mineralized tissues. In Recent Advances in the Study of the Struc- ture, Composition, and Growth of Mineralized Tissues ( R. W. Miner, editor; R. O. Creep and A. E. Sobel, chairmen), pp. 545-572. Ann. N. Y. Acad. Sci., 60, 541-806. Sognnaes, R. F. 1956. Histologic evidence of developmental lesions in teeth originating from paleolithic, prehistoric, and ancient man. Am. J. Pathol, 32, 547-577. Sognnaes, R. F. 1957fl. Relative significance of cellular and chemical re- modeling of bones and teeth as revealed by radioactive isotopes. Oral Surg., Oral Med. and Oral Pathol, 10, 167-174. Sognnaes, R. F. 1957Z;. Some observations and speculations on the physio- logical and pathological significance of the fluid-solid equilibrium of the mouth. Proc. Univ. Rochester Dental Research Fellowship Pro- gram, 25th Year Celebration, pp. 199-208. University of Rochester, Rochester, N. Y. Sognnaes, R. F. 1959. Microradiographic observations on demineralization gradients in the pathogenesis of hard tissue destruction. Arch. Oral Biol, 1, 106-121. Sognnaes, R. F. (editor). 1960fl. Calcification in Biological Systems. Amer- ican Association for the Advancement of Science, Washington, D. C. Sognnaes, R. F. 1960Z?. Dental aspects of the structure and metabolism of mineralized tissues. In Mineral Metabolism — An Advanced Trea- tise (C. L. Comar and F. Bronner, editors), pp. 678-733. Academic Press, Inc., New York. Sognnaes, R. F. 1960c. The ivory core of tusks and teeth. In "Festschrift" for Franklin McLean (M. R. Urist, editor). Clin. Orthopaed., Suppl, 17, 43-62. Sognnaes, R. F. (editor). 1962. Chem,istri/ and Prevention of Dental Caries. Charles C. Thomas, Springfield, 111. Sognnaes, R. F., Shaw, J. H., and Bogoroch, R. 1955. Radiotracer studies DENTAL HARD TISSUE DESTRUCTION 153 on bone, cementum, dentin and enamel of rhesus monkeys. Am. J. Physiol, 180, 408-420. Stafne, E. C. 1933. The effect of therapeutic doses of dikite hydrochloric acid on the teeth. Proc. Staff Meetings Mayo Clinic, 8, 157. Stafne, E. C, and Lovestedt, S. A. 1947. Dissolution of tooth substance by lemon juice, acid beverages and acids from some other sources. /. Am. Dental Assoc., 34, 586-592. Steel, J., and Browne, R. C. 1953. Effect of abrasion upon the acid de- calcification of the teeth. Brit. Dental J., 94, 285-288. Swartz, M. L., and Phillips, R. W. 1952. Solubility of enamel on areas of known hardness. /. Dental Research, 31, 293-300. Wedl, C. 1865. Uber einen im Zahnbein und im Knochen keimenden Pilz. Sitzber. kgl. Akad. Wiss. Wien, 50, 171-193. Werner, H. 1937. Scheinbare und wirkliche Karies an prahistorischen Zahnen. Z. Rassenk., 5, 70. West, E. S., and Judy, F. R. 1938. Destruction of tooth enamel bv acidified candies. /. Dental Research, 17, 499. Zipkin, I., and McClure, F. J. 1949. Citrate and dental erosion: Procedure for determining citric acid in saliva — dental erosion and citric acid in saliva. /. Dental Research, 28, 61^-626. Attrition of the Hypsodont (Sheep's) Tooth C. R. BARNICOAT, Cawthron Institute, Nelson, New Zealand EXCESSIVE wear of incisor teeth of sheep is an economic problem in New Zealand which has become aggravated during the past thirty to forty years. Teeth wear much more rapidly on improved pasture, chiefly rye grass (Lolium perenne) and white clover (Trifolium re pens), than on the fine native grasses of low carrving capacitv. Excessive wear of sheep's teeth is now becoming a problem in other countries where extensive methods of farming are being replaced by intensive practices through the introduction of improved pastures. Most of the results reported in this paper were obtained with Rom- ney sheep, the predominant breed in New Zealand. Since the problem is of regional occurrence — certain districts be- ing noted for sheep "not holding their mouths" — first impressions were that it was of nutritional origin, related to soils and pasture composition and possibh^ to defective mineral metabolism. It was eventually concluded that the cause is not nutritional in the gen- erally accepted sense, however, since the defect is quite difi^erent from caries; and there were no skeletal abnormalities, which are usually associated with deranged calcification. The wear is also dis- tinct from abrasion caused by the presence of foreign gritty particles in the food. The various factors concerned compare and contrast interestingly with those responsible for dental caries in man ( Barni- coat, 1957). 155 156 c. r. barnicoat General The eiglit incisors ("teeth") of sheep (actually six incisors, the two corner teeth being canines) are confined to the mandible and bite against a pad in the upper jaw. The molars ("grinders"), twelve in each jaw, are mainly used in rumination and are not of particular interest to this discussion. Deciduous teeth are shed at 12 to 15 months, after which four pairs of permanent incisors erupt in due order when the sheep is about 15, 24, 33, and 48 months of age. It is then "fresh full- mouthed." The first pair (centrals) often contain as much tooth substance as the rest, and their rate of wear controls that of the others. The hypsodont tooth consists of a crown covered with enamel, and when it is first cut only about half of the crown appears above the gum ( Fig. 1 ) . The remainder of the crown erupts as the animal grows and matures, mostly during the summer and autumn. If the tooth is not subject to wear, it retains its "chisel" edge, and may attain a length of 20 mm or even more when the animal is 5 years old.* Slight wear gives teeth with flat biting surfaces, which are preferred by most fanners. When wear is excessive, however, the unerupted crown responds by emerging from the gum more rapidlv than normal. When all the crown has appeared the tooth may wear eventually to a "gummy" state. On the other hand, teeth of older sheep not fully erupted ma\^ actually lengthen when they are transferred from sparse to generous conditions of feeding. Histological studies show that in the earliest stages of wear, when the thin layer of enamel on the incisal surface has just worn down, dead tracts in the dentine appear between it and the apex of the pulp chamber, and the deposition of secondary dentine is initiated. Sheep's teeth have a surprising facility for forming secondary den- tine, which plugs the pulp canal as it is paid out of the gum with the tooth as it wears (Figs. 2 and 3). Incisal surfaces of older sheep may consist entirely of secondary dentine. The secondary dentine is relatively soft, according to the Tukon Hardness Tester, as shown * Hill country sheep are graded and sorted when 5 years old, the condition of the mouth being an important factor. ATTRITION OF THE HYPSODONT TOOTH 157 in the accompanying table. Yet when mouths are nearing the "gummy" stage the phigs of secondary dentine appear to be the most resistant part of the tooth toward wear. Knoop hardness numbers (air-dried teeth) Average Range Enamel 256 (220-288) Dentine 49 (41- 56) Secondary dentine 25 (13- 32) During the course of this investigation, which involved the ex- amination of thousands of sheep's mouths, dental caries was never detected, though the mouths usually contained much food debris and fibrous portions lodged between the teeth ( Fig. 3 ) . The sheep's incisor tapers from the incisal surface and therefore narrows as it wears, which causes gaps to appear between teeth (Figs. 1 and 3). This gives an appearance of spreading, which was not, however, confirmed by measurements of casts of mouths of in- dividual "full-mouthed " ewes taken at different ages. Factors Responsible for Wear in Sheep's Teeth These may be classified as ( 1 ) managemental ( farming ) prac- tices, (2) anatomical (hereditary) differences in individual mouths, and (3) nutritional factors. Managemental Practices Changes in farming practices may be of paramount importance in overcoming the problem for individual farmers. Suffice it to say that the type of pasture, as well as its length as controlled by grazing stock, has great influence on rate of wear of sheep's teeth. Anatomical (Hereditary) Defects Short and long lower jaws* are common faults in sheep. Only after studying photographs of the mouths of over 100 individual * The terms undershot and overshot are avoided since they have opposite mean- ings in different countries. 158 C. R. BARNICOAT Fig. 1. Central incisors from 18-month-oId (two-tooth) ewes (x 3). The gum level is shown by grass stain (arrows) in A and C. The junction of enamel- covered crown and root is marked with pencil, and the unerupted part of the crown lies between arrow and pencil mark. A contained approximately twice as much "useful tooth substance" as C; B, a wide tooth, though rather less in weight than A, is a better type of tooth. sheep taken at intervals until they attained the age of 5 to 6 years did other well defined faults become evident. They are: narrow teeth, "splayed" or "fanned" central incisors, overlapping central incisors, deep "V" in central incisors, and cleft palate. All these abnormalities are also detrimental to the lasting quality of the teeth, and it is unusual to find only one type of fault present in a defective mouth. Moreover, certain defects in 5-year-old mouths are totally different in appearance at the early stages; for example. ATTRITION OF THE HYPSODONT TOOTH 159 Fig. 2. Longitudinal transparent section of sheep's tooth, about half worn, showing plug of secondary dentine. Note thicker layer of enamel on labial (top) surface. (X 5.) overlapping central incisors usually develop into a deep "V." Except in grossly defective mouths, the permanent teeth are not necessarily similar to the deciduous; and the mouths of twins are usually un- related in type. Short lower jaw is the commonest mouth defect in New Zealand sheep. This type of mouth wears down quickly, and even the presence of this defect to only a mild extent, not usually recognized as a fault by farmers, increases the rate of wear. These defects are probably all of hereditary origin and some are congenital. The laws governing the inheritance of sliape and align- ment of sheep's teeth are, as yet, little understood. The faults are widespread, but by adopting a long-term policy of selecting desir- 160 C. R. BARNICOAT Fig. 3. Excessively worn mouth of 4-year-old ewe, showing caries-like ap- pearance due to staining of plugs of secondary dentine. able mouths of breeding ewes and, more important, of rams, it is possible to "grade up" gradually the quality of the mouths of the whole flock. An interesting feature is that in flocks of sheep noted for sound teeth, these faults may merely be latent in the animals while grazing "good teeth" country. When the sheep are transferred at an early age to "improyed" high-yielding pastures, a surprising range of de- fective mouths usually develops as they mature. Nutritional Factors Results of nutritional trials with sheep emphasize the importance of adequate intakes of calcium and phosphorus. Franklin (1950) found that diets (grains) which were deficient, particularly in calcium, caused a general stunting of the whole skele- ATTRITION OF THE IIYPSODONT TOOTH 161 tal structure, as well as irregularities and gaps in the molars. Mc- Roberts and Hill (1962) concluded that "diet deficient in calcium had a greater effect on the teeth than on the skeleton, while the diet deficient in phosphorus and in Vitamin D had a greater effect on the skeleton than on the teeth." Recommended minimum adequate requirements of calcium and phosphorus for sheep (Natl. Research Council Natl. Acad. Sci. (U. S.), 1949) as percentage of dry matter are: Ca P Ca/P Growing lambs 0.18 0.15 1.2 Ewes 0.20-0.23 0.18-0.24 1.2-1.6 The average composition of herbage eaten by New Zealand sheep in relation to extent of wear of teeth is recorded in Table I. The in- take of calcium and phosphorus is satisfactory by accepted stand- ards; and, in fact, the better teeth are found in animals ingesting herbage with the lower mineral content. No relationships between quality of sheep's teeth and deficiencies or imbalances of microelements (i.e. the biologically important cop- per, molybdenum, cobalt, iron, fluorine, iodine* ) were established; nor is there any apparent relationship with geological formation or soil type. There is no evidence to support the theory that vitamin A (as the provitamin carotene) or D is implicated, and the water- soluble vitamins are believed to be more than amply provided by the rumen microflora (Barnicoat, 1959). Analyses of enamels and dentines of teeth of New Zealand sheep are recorded in Table II. Their stoichiometric relationships are given in Table III. Figures vary little, and are not influenced by the quality of the mouth. These analytical results are similar to those reported by numerous workers for human teeth (Leicester, 1949). Spectrographic analyses of enamel and dentine for 19 microele- ments of biological importance, including aluminium, cobalt, copper, iron, molybdenum, silicon, vanadium, and zinc, showed no differ- * Paradontal disease, chiefly affecting molars, is associated with selenium de- ficiency (Hartley and Grant, 1961 ). 162 C. R. BARNICOAT O O H O < > O ;? o H P^ H ^ H w < Q g w fa * o o fa o X OQ 5 § > p-l OS O o o q p. ^ t^ o o ft ^ ^ bC oj C o -1^ -s o3 (^ M fc- o M . o OJ > 'x is <: •Oj — £= -S -^ w 03 O o GO ilO CO o o d d V O G _- --^, "r^ ^^. — ^ fl j3 ^ O -(^ C"* o O 'bb P '-(^ & ci X 02 f^ 3 rt o 3 o ?"— ' ^ "S S d oj oj ►^ ~ s cc tc rp w « cc .;:; 3 cc Oj o3 Oj •-^ !_ ^ CU hC ^^ «2 g o3 9 SS TJ it: a O •« ATTRITION OF THE HYPSODONT TOOTH 163 rr 13 .— c H cr h w r;2 CU r-" a 71 ~f^ r^ r^ ' Cl ~ ;:^ a r^ X W 'ji pi , o . f-H o o g O CO -^ ?= i§ qC 3 fcj ^ ? -G 1 -, OJ n H - o (>l -f <^ — f" P ^ (M_ 1^5 I — I oa ' — ^ o ' — ' O , — , -t< , — , O , — , I , , 00 CD t^ lO Cq lO £2 00 c^ "-^ d ■— ' CO ■— ' 3'—' d ^ a; p p oc' lO o p CD Ci -TtH ^ 00 I — , ^ , I ^ , , <^. 00 "O 00 >o t^ o ' ' d ' ' (m' ' — ' "^ I — 1^1 — , P lO P lO ,_i ^,^1 1£3, , 00, rt^i — .CM, ^co, pCOpCDp-* (Nc)0^00»OcDOTo CO 2S ' ' "^ ' — ' °' ' — ' ^' — ^ '^' ' — ' ^' ' — t^ , — , C3 , — ^ CO , , , '^ 00 CO I CO -H _H 00 t>6 O ' — ' o d o d ~ o • r3 -— ; oC T3 CJ C3 ^,,^^ ^ O cc^ CD o -^J _o tB c3 t/3 p 03 C^l a bc a. ^-^ o tc 03 5 *-■ <: fe 'c^ o h*i )-H ff^ ^ 164 C. R. BARNICOAT a H X < m o ^ w o3 bX3 H J- tn Cm ^ 1 q3 ^ fl g ■J c5 o cc 03 Sh V -tJ OJ Dh jii 3 1 o s X 12; o < b L^ -i^ -tj c3 Q o o T 00 O o O o o o o w -1-3 o g 1 o S d o d Q l"~{ o CO « 3 , p cS p H 'bb *; *^ o !=; *p o IS (M o^ o ' — ' ' — ' O ^ r-^ ^ r^ 4.i -fJ K o o O (—1 O r^' ^ 1? !=^ o ^ -1-3 w o s fe ^ sS >-. rt -o -: ^ ^ H f S ^ ^ ^ J; -tf ^ ^3 ■ZS a i=l OJ C C; « aj oi o o p:;OH HrH ^ (M CO 't lO lO ^-( (M ,—1 H o q CO '*. o lO iq 05 id 00 CO I-'^ X 1>^ i-H CO S CO 00 lO rH CO to CO O T-H • i-« q q 00 "t q i>. iq CO o id 00 CO t^ 00 x' r-^ CO CO 00 lO lO lO Oi Co Oi o q cc -t q CO iq lO id GC CO t-^ X x' ^ CO a CO 00 lO -iU ^ -t< ; o -t q Ol lO lO CO Sh t^ oi (>1 -^ ^ id r-^ id ._o CO Ci CD ^ tJ t— -+' lO ^ ^ q 00 iq 00' q 00 iq 0) ui b/j o_ Q G o TO A.D.J. i o ?^ ^■^ 2 y p = 180 MICROSTRrCTURAL CHANGES IN EARLY CARIES 181 cules bigger than those of water, whereas the largest will accept octanol and quinoline with ease. This zone is graduallv transformed into the body of the lesion by the steady growth of the spaces in the dark zone. The bodv of the lesion contains at least 5 per cent of spaces and may contain up to 25 per cent or possiblv 50 per cent or more. All these spaces are large enough to admit octanol and quinoline and possibly even larger molecules. Thus the zone of translucence described by Nishi- mura (1926) and Gustafson (1961) just between the dark zone and the body of the lesion would be explained in the same way and would be distinguished because the more highly demineralized parts of the bodv of the lesion probablv have an altered refractive index which mav produce some form birefringence in media of R.I. 1.62 (Crabb, 1962). Using this knowledge of the spaces and their form birefringence, it is possible to demonstrate something of the pattern of spread (Fig. 8) (Darling, 1961). This begins as fine streamers spreading into the normal enamel along the so-called interprismatic markings. From these the cross striations are involved, and soon afterward the whole of the prism seems to be affected. By combining the evidence from microradiography and observa- tions of form birefringence, it is now possible to give a description of the various stages as follows. The translucent zone shows no evidence of demineralization, though occasionallv it may be hypermineralized. When the latter condition is not found, i.e. in the great majority of cases, this zone is characterized by the sudden development of large spaces in the sites where normal enamel has minute spaces (Fig. 7). These spaces of the translucent zone persist in the dark zone but are overlain bv the development of minute spaces penetrable only bv the molecules of water. These grow throughout the dark zone until they become large spaces, big enough to admit octanol, quino- line, and balsam. When all the spaces reach this size the dark zone is transformed into the body of the lesion. The body of the lesion and to some extent the dark zone show demineralization. There can be little doubt that the minute spaces of the dark zone growing into the body of the lesion represent the results of demineralization, but 182 A. I. DARLING 1 / »i Fig 8. Ground section of tarly caries of human dental enamel mounted in quinoline and seen by polarized light in the 45° position, (x 750.) The illustra- tion shows normal enamel at the lower border and passes through translucent zone and dark zone to the body of the lesion at the upper border. it is not yet possible to say definitely what material is lost to produce the spaces in the translucent zone. The explanations offered on the quality of the attack are to some extent modified by opinions on the precise structures involved and their nature. Most workers now agree that there is a differential decalcification of certain structures. Most seem to accept that the pattern of attack, after the surface is penetrated, begins in the so-called interprismatic region, then involves the cross striations or something very like them, after which the prism becomes more completely involved, leaving rings of unaffected enamel which correspond in size to prisms. The precise location of these structures, however, is difficult to determine. All seem to agree that the surface zone, rings of enamel correspond- MICROSTRUCTURAL CHANGES IN EARLY CARIES 183 ing in size to the prism cortices, zones beneath striae of Retziiis, and Hnes parallel to the cross striations are not affected until late in the process of demineralization, but there is considerable difference of opinion as to why this should be. Gustafson (1961) speaks chiefly of variations in the degree of mineralization as the cause of this differential demineralization. Brudevold (1961) has suggested that it is the chemistry of the struc- tures and in particular the distribution of fluoride which is responsi- ble. Poole ( 1962 ) believes that it may be largely a matter of the accessibility of the structures to the pathways of attack, and Little (1959) and Darling (1958) have suggested that it is related to dif- ferences in the organic matrix of the structures. There seems no reason why all these possible causes should not play some part, but the marked difference in susceptibility between the surface zone and the deeper enamel makes it unlikely that this difference will depend to any great extent on the degree of mineral- ization or the accessibility of the structure to the pathways of attack. It seems that here at least the resistance to attack is more likely to depend on fundamental chemical or structural differences. To recognize such differences as exist between these two groups of structures it is necessary to delineate them anatomically. Here again there is some difficulty. Most workers seem to agree that the early attack is at the margins of the structures known as prisms, but whether it is truly interprismatic or in the prism cortex or just out- side or just within this structure, or even along spaces at the junction of the prism and the so-called interprismatic structure, seems to be open to argument. Similar arguments seem to apply to other stages in the anatomv of the attack. Probably the best and most acceptable description which can be given at present is that the attack appears to take place first along the interprismatic or border region of the prisms, from which it spreads to involve the cross striations and later what seems to be the prism core, leaving stripes parallel to the cross striations, and, more obviouslv, what look like the cortices of the prisms, relatively unaffected. In the surface attack there seem to be two possibilities which appear to conflict. Guzman et al. ( 1957 ) believe that entry is along 184 A. I. DARLINCx the prism cores, which can be seen on the outer surface of the en- amel. Mortimer ( 1962 ) tends to agree so far as occlusal lesions are concerned, but it is very difficult to understand how such an attack can leave the surface zone apparently intact. It might be argued that what they see is the late spread of the attack back into the sur- face zone from its undersurface along the prism cores. The evidence in both cases is microradiographic. Darling (1956, 1958), on the other hand, both by birefringence and bv microradiograph\', shows evidence of attack along the striae of Retzius passing through the surface zone into the lesion beneath it. Admittedly, the whole of the surface zone is involved in a process similar in some wa\'s to that in the rest of the lesion, but it rarelv passes the stages of translucent and dark zones before the whole of the surface breaks down. Even dark zones are rarely seen in the surface zone except along the striae of Retzius. It seems tremendously important that tlie pathwavs of attack should be clearlv identified on an anatomical basis. As for the qualitv of the changes in the various zones, we know that the evidence obtained from the dark zone and the body of the lesion by various methods of examination suggests that the changes in these zones are produced bv the progressive effects of selective or differential demineralization. The translucent zone, on the other hand, shows no exidence of mineral loss bv microradiography or in birefringence. On the basis of work by Rowles (1955), Darling and Mortimer (1959), and Stack ( 1954 ), Darling (1961) has suggested that this zone mav be caused bv loss of organic material or carbon- ate. Darling and Mortimer ( 1959 ) have claimed that similar zones can be produced by removal of soluble organic material, though this work is being reexamined bv them because of some doubts. The theory on which they base this work suggests that the selective de- calcification of the enamel structures is related to the presence of predominantly "soluble" matrix in the "susceptible" structures and of "insoluble" matrix in the "resistant" structures. Both types of ma- trix are known to exist, but it is difficult to prove their precipe distribution. There can be no doubt that we are approaching a real understand- ing of the carious process in enamel, but the end is not yet and will I MICRO STRUCTURAL CHANGES IN EARLY CARIES 185 not be reached amel structure. not be reached without much greater understanding of normal en Summary The evidence on the structural changes in demineralization of enamel by dental caries is reviewed. The findings by microradiog- raphy and observations of intrinsic birefringence are in agreement and are accepted by most workers. By the use of form birefringence, spaces caused by the attack can be demonstrated. There is some difference of opinion on the paths of penetration through the enamel surface and on the precise detail and causes of the spread within the enamel, but it is generally agreed that there is a differential de- mineralization of the structures of the enamel. The zones of the enamel are identified and an explanation is offered for their appear- ances. Progressive demineralization is found in the dark zone and the bodv of the lesion and the intermediate zone, well in advance of histologically recognizable organic change. The translucent zone, though occasionally hypermineralized, usually shows a stage of at- tack without evidence of demineralization and because of this may represent a different type of attack. The nature of the attack in this zone is not clear, though suggestions are offered. References Applebaum, E. 1932. Incipient dental caries. /. Dental Rcseorcli, 12, 619- 627. Bergman, G., and Engfeldt, B. 1954. Studies on mineralised dental tissues. II. Microradiography as a method for studying dental tissues and its application to the study of caries. Acta Odontol. Scaiul., 12, 99-132. Brudevold, F. 1961. Personal communication. Crabb, H. S. M. 1962. Areas simulating carious lesion in the enamel of teeth from dentigerous cysts. Intern. Assoc. Dental Research, British Sect. /. Dental Research 41, 1253. Darling, A. I. 1956. Studies of the early lesion of enamel caries with trans- mitted light, polarised light, and radiography. Brit. Dental J., 101, 289-297. Darling, A. I. 1958. Studies of the early lesion of enamel caries, its natunt. mode of spread and points of entry. Brit. Dental J., 105, 119-135. 186 A. I. DARLING Darling, A. I. 1961. The selective attack of caries on the dental enamel. Ann. Roy. Coll. Surg. EngZ., 29, 354-369. Darling, A. I., and Mortimer, K. V. 1959. Further observations on the early lesion of enamel caries. /. Dental Research, 38, 1226. Darling, A. I., Mortimer, K. V., Poole, D. F. G., and OlHs, W. D. 1961. Molecular sieve behaviour of normal and carious human dental enamel. Arch. Oral Biol, 5, 251-273. Gustafson, G. 1945. The structure of human dental enamel. Odontol. Tidsskr., 53, SuppL, 1-150. Gustafson, G, 1957. The histopathology of caries of human dental enamel. Acta Odontol. Scand., 15, 13-55. Gustafson, G. 1961. Human dental enamel in polarised light and contact microradiography. Acta Odontol. Scand., 19, 259-287. Guzman, C., Brudevold, F., and Mermagen, H. 1957. A soft roentgen-ray study of early carious lesions. /. Am. Dental Assoc, 55, 509-515. Keil, A. 1935. Uber den Wandel der Doppelbrechung des Zahnschmelzes bei Entlcalkung, Wiirmeeinwirkung und Karies. Z. Zellforsch., 22, 633-649. Keil, A. 1936. Beitriige zur Kenntnis der Doppelbrechung des mensch- lichen Zahnschmelzes. Z. Zellforsch., 25, 204-220. Kostlan, J. 1962. Translucent zones in the central part of the carious lesion of enamel. Brit. Dental /., 113, 244-248. Little, K. 1959. Electron microscope studies on human dental enamel. /. Roy. Microscop. Soc, 78, 58-66. Miller, J. 1958. Note on the earlv carious lesion in enamel. Brit. Dental J., 105, 135-136. Mortimer, K. V. 1962. Personal communication. Nishimura, T. 1926. Histologische Untersuchungen iiber die Anfange der Zahnkaries, speziell der Karies des Schmelzes. Schweiz. Monatsschr. Zahnheilk., 36, 491-545. Poole, D. F. G. 1962. Personal communication. Poole, D. F. G., Mortimer, K. V., Darling, A. I., and Ollis, W. D. 1961. Molecular sieve behaviour of enamel. Nature, 189, 998-1000. Rowles, S. L. 1955. Some observations on histological techniques. /. Dental Research, 34, 778. Schmidt, W. J., and Keil, A. 1958. Die gesunden und die erkrankten Zahn- gewehe des Menschen unci der Wirbeltiere im Polarisationsmikro- skop, pp. 297-314. Garl Hanser Verlag, Munich. Stack, M. V. 1954. Organic constituents of enamel. /. Aw. Dental Assoc, 48, 297-306. 7 Ultrastructural and Chemical Observations on Dental Caries ERLING JOHANSEN, Department of Dentistry and Dental Research, School of Medicine and Dentistry, University of Rochester, Rochester, New York CARIOUS lesions are manifestations of a complex disease phenome- non affecting the calcified tissues of teeth. It is known that numerous factors including the saliva will influence the course of the disease, and that the basic process by which lesions develop is generated by microbial metabolism in localized areas on external tooth surfaces. The acid environment thus created by metabolism of carbohydrates favors dissolution of dental tissues. Enamel and dentin do not re- spond to this injury by cellular regeneration or repair, but by a de- fensive obliteration of dentinal canals and development of secondary dentin between the lesion and the pulp. These formations at the same time constitute a barrier to the penetration of pulpal fluids into the lesion. On these grounds it can be assumed that the ultrastruc- tural and chemical changes which have occurred in dental caries are mainly the results of the carious process with modifying influences primarily limited to factors associated with the fluid environment of the oral cavitv. The original ultrastructural studies on carious lesions of enamel and dentin to be summarized in the present report were carried out on zones of advanced destruction. This part of the tissue was chosen in order to observe maximal changes brought about bv the carious 187 188 E. JOIIANSKX process. Similar zones were used for the chemical analyses of dentin, whereas technical difficulties introduced some variation in the sam- pling of enamel. Observations on the ultrastructure and composition of carious lesions in various stages of development have been reported by other investigators. Hohling (1961) and Lenz (1961) noted morphological changes in cr\'stallites from advanced carious lesions in enamel, and Scott and Albright (1954) compared the matrix of different zones within enamel lesions. In dentin, the distribution of mineral within zones of advanced lesions was studied bv Takuma and Kurahashi ( 1962), and Bernick et ah ( 1954) noted the presence of cross-striated collagenous fibrils in carious matrix. These and other electron micro- scopic studies of carious tissues have recentlv been comprehensively considered bv Helmcke ( 1962 ) . Chemical studies by Coolidge and Jacobs ( 1957 ) of carious lesions of enamel with intact surfaces (white spot lesions) showed a decrease in the amount of calcium, phosphorus, and carbonate on a volume basis. Stack (1954) and Bhussry (1958) reported an increase in organic material in such lesions. In carious dentin, Manly and Deakins (1940) described three physically and chemically distinct zones and noted that volume loss of organic and inorganic constituents (from these zones) was associated with gain in moisture. The amino acid composition of carious dentin and its mucopobsaccharide content have been in- vestigated by Armstrong (1961rt, 1961Z?). Comparable studies on sound tissues have recentlv been reviewed in detail bv Brudevold (1962), Frank et ah (1960), and Johansen (1963, 1964). Even though areas of controversy and incomplete knowledge remain in this field of investigation, the information available provides a basis for comparison with conditions in altered tissues. Materials and Methods Fully formed permanent human teeth with active carious lesions and extracted in the citv of Rochester, New York,* were used for both the chemical and the ultrastructural studies. For observations ^ The watt'i" supply of Hoch(\ster contains 1 ppni of fluoride. ULTKASTRVCTURAL AND CHEMICAL STUDIES ON CARIES 189 Oil carious enamel, lesions with and without break in surface con- tinuity were selected. The ultrastructural studies were performed on areas of advanced destruction in both types of lesions. For the chem- ical analyses, lesions with intact surfaces were used for determina- tion of major elements, carbonate, and fluoride; lesions showing cavitation were studied for trace elements. In the study of dentinal lesions all samples were obtained from soft carious dentin. For the electron microscopic work whole teeth or specimens of sound and carious tissue were fixed immediately after operative re- moval in cold (4°C) 1 per cent osmic acid solution containing sucrose (0.22 m) and buffered to pH 7.2-7.4 with 0.025 m veronal acetate buffer; or in neutral 10 per cent formalin or in 70 per cent alcohol. After fixation some of the material was dehvdrated in al- cohol and embedded in butyl methacrylate catalvzed with 1 per cent benzoyl peroxide. Some material was stored in 70 per cent al- cohol and sectioned without further treatment. A few observations were also made on fresh untreated tissues. Obser\'ations on the inorganic phase of enamel and dentin were made on ultrathin sections cut with a Porter-Blum microtome equipped with a glass knife, and an LKB Ultratome equipped with a diamond knife. Samples of enamel gentlv triturated in distilled water were studied without embedding. The collagen-crvstallite re- lationships in dentin were examined in tissue samples homogenized in distilled water emploving a Dounce ground-glass homogenizer. For study of the organic phase the tissues were usuallv demineral- ized prior to embedding. To protect the enamel matrix, the tissue samples were maintained within a dialysis bag dining demineraliza- tion with EDTA (0.46 m, pH 7.2-7.4) (Burrows, 1961), and also during the subsequent washing, dehydration, and infiltration pro- cedures. After embedding in butyl methacrylate and sectioning, the tissue was stained with a lead preparation (Watson, 1958). Dentin samples were demineralized in EDTA or in formic acid-sodium citrate decalcification solution (Morse, 1945) prior to embedding. Sections of dentin were stained with 10 per cent phosphotungstic acid solutions for periods from 10 minutes to 2 hours. Micrographs were taken with a Siemens Elmiskop la operated at 60 or 80 kv. (For details of procedures, see Johansen and Parks, 1961, 1962.) 190 E. JOHANSEN The chemical studies were performed on samples of enamel and dentin prepared by mechanical separation of the tissues using dia- mond wheels, burs, or hand instruments. This procedure has been found to give enamel and dentin samples of high puritv. In studies of the inorganic phase the organic matrix was eliminated by ashing at temperatures of 550 to 600° C. Dry weights of the samples were obtained by drying in partial vacuum for 24 hours at 105° C. For specific elements and for carbonate determinations the follow- ing procedures were used. Calcium and magnesium combined were determined by complexometric titration with EDTA, using erio- chrome black T as indicator (Biedermann and Schwarzenbach, 1948). Magnesium was then determined separately according to a colorimetric method (McCann, 1959), For phosphorus the colori- metric method of Chen et al. (1956) was employed. Carbonate de- terminations were performed on the standard VanSlvke manometric apparatus (VanSlyke and Folch, 1940), modified for solid samples (Sobel etal., 1943) with solutions prepared according to MacFadyen ( 1942 ) . In the determination of fluoride the traditional steam dis- tillation was employed ( Willard and Winter, 1933 ) , and fluoride in the distillate was determined by spectrophotometric methods (Me- gregian, 1954). (Specifics of procedures will be published in de- tailed reports on these studies.) Results Ultrastructure A conspicuous feature of many sections of carious enamel in an advanced stage of destruction was the occurrence of distinct struc- tureless areas of varying width delimiting enamel rods ( Fig. 1 ) , and corresponding to rod sheaths in location (Figs. 4 and 11). Another characteristic was the sparse distribution of crystallites within the carious tissue (Figs. 1 and 2). These findings indicated extensive demineralization within rods and in rod sheath and interrod areas, resulting in spaces which appeared to form a system of microchannels in the tissue. The occurrence of such a system was readily demon- strated by immersing a carious lesion of enamel in silver nitrate solution. Within a few minutes the electron-dense silver was found ULTRASTRUCTURAL AND CHEMICAL STUDIES ON CARIES 191 in spaces of enamel rods ( Fig. 3 ) . The sound tissue showed no evi- dence of penetration of silver in the same period of time. Observations on the morphology, size, and electron density of crystallites within individual enamel rods at times suggested re- gional differences in these characteristics. The crystallites located at the periphery of rods appeared larger, more electron dense, and generally better preserved than those within rods (Fig. 2). It may be speculated that such crystallites have been selectively spared from dissolution or selectively recrystallized in the carious process. Individual crystallites studied in sections and in triturated samples of carious enamel showed evidence of morphological changes (Figs. 5 and 6) referable to the carious process. The changes included localized and generalized surface erosion with establishment of spaces between adjoining crystallites (Figs. 6a and 6b). Also, per- forations in transverse direction (Fig. 6c) were common findings. Occasionally, crystallites appeared to present perforations in longi- tudinal directions (Fig. 5). The most severely affected crystallites were seen as mere vestiges with little resemblance to the crystallites of sound enamel. Others (Fig. 6d) appeared remarkably similar to segments of crystallites observed in sound enamel (Figs. 7 and 8). The many apparent spaces seen in carious tissues suggested that many crystallites had been entirely dissolved. Organic material was found in all samples of carious enamel stud- ied after artificial demineralization, embedding, sectioning, and stain- ing. Frequently, rod sheaths appeared distinctly outlined (Fig. 9), displaying a fine fibrillar component. In some preparations similar fibrils were also seen within rods and in interrod areas, while in other preparations these areas were devoid of matrix. (Inteipreta- tion of these findings in relation to the process of dental caries must, however, await further clarification of the modifying influence of certain technical procedures.) In other sections there appeared to be more organic material in the carious (Fig. 10) than in the sound tissue (Figs. 11 and 12). This finding may represent the presence of extraneous organic material derived from oral fluids. With the stain- ing procedures available at present, however, it was not possible to distinguish between extraneous and intrinsic organic material. Carious dentin studied in sections displayed a most interesting 19^2 E. JOIIANSEN Figures 1 to 4 ULTRASTRUCTURAL AND CHEMICAL STUDIES ON CARIES 193 pattern of distribution of mineral. The greatest concentration was invariably found in areas lining the canals, often forming a hyper- mineralized pericanalicular laver interposed between the lumen of the canal and the extensively demineralized intercanalicular matrix (Figs. 13 and 14). It is not known whether this layer represents a persistence of the pericanalicular zone found in sound dentin ( Fig. 15), or whether its occurrence is related to the carious process. The most surprising finding in the studv of carious dentin was the presence of distinct, relativel)' large crystallites in both pericanalicu- lar and intercanalicular areas (Figs. 16 and 17). Stereoscopic ob- servations (Johansen and Parks, 1960) showed that these structures resembled the crystallites of sound dentin in size and in their plate- like shape (Figs. 18 and 19). This unexpected finding led us to pro- pose that these crystallites persisted in the lesion because they were chemically different from the crystallites which had been dissolved. It was considered that fluoride preferentially deposited in the lining of dentinal canals and in persisting crystallites of intercanalicular areas could be responsible for this phenomenon (Johansen and Parks, 1959, 1961). Crystallites of carious dentin were also found to be closely as- sociated with collagenous fibrils isolated from homogenate prepara- tions. Usually the crystallites were oriented with their long axes parallel to the long axis of the fibril (Fig. 20) in an arrangement similar to that seen in the heavily mineralized fibrils of sound dentin (Fig. 21) isolated by similar techniques. Fig. 1. Section of undecalcified carious enamel showing enamel rods (R) outlined by wide rod sheath areas (RS). The sparse distribution of crystallites indicates extensive demineralization. (X approx. 14,250.) Fig. 2. Section of undecalcified carious enamel. Crystallites (AC) at the periphery of rods are generalK' more conspicuous than those within rods. ( X approx. 11,400.) Fig. 3. Thin fragment of carious enamel exhibiting silver particles (Ag) within spaces in the tissue. The carious lesion was exposed to silver nitrate solution to test the permeability of the tissue. Dry preparation of specimen. ( X approx. 66,500.) Fig. 4. Section of developing deciduous himian enamel, illustrating the gen- eral organizational pattern of the tissue. Rods (R) delimited by rod sheaths (RS) and separated by so-called interrod areas (IR) are seen. (X approx. 15,200.) 194 E. JOHANSEN Figures 5 to 8 ULTRASTRXJCTURAL AND CHEMICAL STUDIES ON CARIES 195 Artificial demineralization of soft carious dentin and staining with phosphotungstic acid or lead preparations revealed an abundant matrix of collagenous fibrils (Fig. 22). The presence of typical cross-banding in longitudinal sections (Fig. 23) suggested that most fibrils were intact, resembling those of sound dentin (Fig. 26). However, areas with marked reduction in numbers of fibrils were seen, both in close proximity to bacteria (Fig. 24) and, in some regions, at a distance from bacteria. Observations on homogenate preparations of carious dentin also revealed collagenous fibrils with typical banding but devoid of crystallites ( Fig. 25 ) . The persistence of these fibrils in the carious tissue strongly supports the point of view that demineralization precedes proteolysis in dentinal caries. (For comparison with bone resorption see Hancox and Boothroyd, chapter 18 in this volume.) Composition Interpretation of the electron microscopic observations presented above raised a number of questions pertaining to the chemical com- position of the carious tissues. In an attempt to clarify some of these points, a number of analyses were carried out. Several pertinent find- ings will be summarized by representative data. Determination of major elements and carbonate of carious enamel revealed a reduction of calcium, phosphorus, magnesium, and car- FiG. 5. Section of carious enamel showing crystallites of varying size and shape. (X approx. 40,500.) Fig. 6. Crystallites from carious enamel. 6a. Surface erosion with resulting space (S) between adjacent crystallites. Triturated preparation. (X approx. 77,000.) 6b. Apparent localized erosion of adjacent crystallites with estabhshment of channel-like space (S). (X approx. 124,750.) 6c. Transverse perforation in crystallite (S). (X approx. 124,750.) 6d. Segments of well preserved crvstallites within the lesion. ( X approx. 124,750.) Fig. 7. Section of sound permanent enamel showing distinct outlines of in- dividual crystallites. Several apparent junctions (/) between ends of crystallites have been marked. ( X approx. 103,700.) Fig. 8. Section of sound permanent enamel illustrating obliquely cut crys- tallites. (X approx. 124,750.) 19G E. JOHANSEN "-IR 9' 10 It, RS ^ '<♦ ■'*^«. IR' \ f' , ^i. " \. «' .12 Figures 9 to 12 ULTEASTRUCTURAL AND ('HF:^[ICAL STUDIES ON CARIES 197 TABLE I. Major Elements, Carbonate, and Water Content of Sound and Carious Human Enamel" (Pit cent dry weight; means of 20 samples. From Singh and Johansen, to be pubhshed) Ca P Mg CO2 Ca/P HoO Sound 3G.75±().17 17.41 ±0.04 0.54 ± 0.01 2.42 ± 0.02 2.09 ± 0.02 2.02 ± 0.04 Cai'ious 35.95 zt 0.21 17.01 ±0.06 0.40 ± 0.01 1.56 ±0.03 2.08 ± 0.03 3.07 ± 0.05 " Lesions with intact surfaces (white spots). bonate content (Table I). This was particularly noticeable on a volume basis, but the change was also statisticalh' significant on a dry weight basis. The data show that the calcium phosphorus ratio remained relatively unchanged in spite of the demineralization proc- ess. On a percentage basis the loss of carbonate and magnesium was much higher than that of calcium and phosphorus. This finding mav indicate a preferential loss of magnesium and carbonate in the dis- solution of crystallites, or it could possibly reflect a low concentra- tio.n of these ions during the assiuned recrvstallization process. Observations on the fluoride content of carious enamel are found in Table IL In study I are presented fluoride values for total lesions compared with a corresponding sample of sound enamel. It can be seen that carious lesions on an ash basis contain 5 times more fluoride than the sound tissues. Since it is known that the fluoride concentra- FiG. 9. Section of organic matrix from demineralized carious enamel stained with lead preparation. Rod sheaths (RS) are prominent in outline. Fine fibrils are frecjuentlv missing in interrod areas (/R) but are present within rods (R). ( X approx. 10,500. ) ^ Fig. 10. Section of organic matrix from demineraHzed carious enamel stained with lead preparation. An apparent increase in the amount of organic material is observed. ( X approx. 35,650.) Fig. 11. Section of organic matrix from demineralized sound enamel stained with lead preparation. The morphologic pattern of the tissue is readily recog- nized bv rods (R), rod sheaths (RS), and so-called interrod areas (/R). (X approx. 8550.) Fig. 12. Section of organic matrix from demineralized sound enamel stained with lead preparation. Longitudinallv cut rods (R) delimited by rod sheaths (RS) exhibit a feltwork of minute fibrils. (X approx. 29,200.) 198 E. JOHANSEN Figures 13 to 15 ULTRASTRUCTURAL AND CHEMICAL STUDIES ON CARIES 199 TABLE II. The Fluoride Content of Sound and Carious Human Enamel" (Ash weight. From Johanson and Nordback, 1962) Fluoride (ppm) Study I* Study 11'^ Carious Sound: outer and inner laj-ers Sound: outer layer Sound: inner layer 580 120 490 310 90 ° Lesions with intact surfaces (white spots). ^ Pooled sample of 14 teeth. The sound sample was obtained by preparing cavities similar in size and shape to those resulting from removal of white spots. "= Pooled sample of 31 teeth. The sample labeled "outer layer" represents the somid enamel from the surface inward to one-half the depth of cavities resulting from re- moval of white spots. The "inner layer" represents the remaining enamel to the full depth of the cavity. tion of surface enamel is higher than that of subsurface enamel (Brudevold et al., 1956), it was also of interest to compare the fluoride content of the carious lesion with that of layers of sound enamel. The results of this comparison are presented in study II. Carious lesions again showed a higher concentration of fluoride than corresponding sound tissues. Some trace elements other than fluoride were also studied, and the results are presented in Table III. It was found that carious enamel showed a slight increase in lead, iron, and aluminum. Studies on the composition of soft carious dentin on a weight basis Fig. 13. Section of undecalcified soft carious dentin showing dentinal canals (C) filled with microorganisms (B). The intercanalicular matrix (7C) is largely but incompletely decalcified by the carious process and contains some scattered microorganisms (B) within lacunae. The polystyrene latex spherules (courtesy of Dow Chemical Company) are approximately 0.26 micron in diameter (x approx. 8000.) Fig. 14. Undecalcified section of soft carious dentin showing a thick, densely calcified layer interposed between the lumen of the canal (C) and the ex- tensively demineralized intercanahcular matrix (IC). (X approx. 26,000.) Fig. 15. Section of embedded, undecalcified sound dentin. A zone of high mineralization (PC), approximately 0.5 micron wide, surrounds the lumen of the canals (C). The intercanahcular matrix (IC) shows an irregular but gen- erally less dense distribution of mineral matter, occasionally displaying a banded appearance. (X approx. 15,000.) •200 E. JOHANSEN ,» «««^ - - -HI t^' 44 ^ 4 ^ '.-** AC 20 Figures 16 to 21 ULTBASTRUCTUEAL AND CIIFIMICAL STUDIES ON CARIF:S 201 TABLE III. Trace Elements of Sound and Carious Human Enamel" (Ash weight.'' From Johansen, Steadman, and Steadman, to be published) I: lement (ppm) Pb Sn Mn Fe Al Sr Pound Carious 60 93 <3 24 <0.4 <1 25 52 32 93 94 99 " Soft carious enamel without surface layer; samples removed with pj con excavator. '' Pooled sample of 25 teeth. revealed an increase in water and organic material and a decrease in ash (Table IV). Analysis of the ash fraction for major elements showed a marked decrease in magnesinm with relatively minor changes in calcium and phosphorus contents. The calcium phos- phorus ratio was somewhat lower in the carious tissue than in sound dentin. On a dry weight basis, carious dentin also contained about 50 per cent less carbonate than sound dentin (Fig. 27), and the difference in rate of release at temperatures below 600 °C might in- FiG. 16. Section of undecalcified soft carious dentin embedded in methac- ryhite. Crystallites (AC) seen in edge view present thin, dark profiles; those seen in broad surface view are wider and less dense. At left is the lumen of a dentinal canal (C). (X approx. 74,650.) Fig. 17. Undecalcified soft carious dentin, showing the somewhat crumbled edge of a section taken from unembedded material. Most crystallites appear in broad surface view. ( X approx. 96,800. ) Fig. 18. Section of undecalcified sound dentin embedded in methacrylate. Crystallites (AC) seen in edge view present thin, dark profiles; those seen in broad surface view are wider and less dense. ( X approx. 74,650.) Fig. 19. Undecalcified sound dentin, showing the somewhat crumbled edge of a section taken from unembedded material. In the thinnest parts, the crys- tallites appear in broad surface view; in the thicker parts they are seen in edge view. (X approx. 96,800.) Fig. 20. Isolated collagenous fibril from homogenized preparation of carious dentin. A few crystallites (AC) in both broad surface view and narrow profile view are seen associated with the fibril. Very faint indication of banding is seen. ( X approx. 69,600. ) Fig. 21. Isolated collagenous fibrils from homogenized preparation of sound dentin. The crystallites are oriented with their long axes parallel to the long axis of the fibrils. The alternating light and dark transverse bands are seen to be in register where the two fibrils join to form a single unit. (X approx. 70,000.) 202 E. JOHANSEN Figures 22 to 26 ULTRASTRUCTURAL AND CHEMICAL STUDIES ON CARIES 203 TABLE IV. Composition of Sound and Carious Human Dentin (From Johanseii and Nordbat-k, 1962) Or- Inor- Tissue Water ganic ganic — > Ca P JNIg CO2 Ca/P Wet weight Ash iveight Sound 11.2 20.7 68.1 .37.8 ± 0.83 17.7.5 ± 0..34 1.10 ±0.18 2.20 ±0.37 2.13 ±0.04 Carious 58.0 27.4 14. G 38.4 ±0.74 18.38 ± 0.42 0.27 ± 0.20 1.07 ±0.36 2.08 ± 0.04 Dry weight'' Sound - 23.0 77.0 27.5 13.0 0.83 3.50 Carious 65.6 34.4 12.8 6.2 0.10 0.74 ° Organic and bound water. * Inorganic composition corrected for CO2 loss (see Fig. 27). dicate differences in the location of carbonate in apatite crystallites from these two sources. Of particular interest for the interpretation of the electron micro- scopic studies were the observations on the fluoride content of carious dentin. The data presented in Table V show that the inorganic phase of soft carious dentin contains about 10 times more fluoride than the corresponding sound tissue. The high fluoride content might thus be the most important factor in maintaining the crystallites within the carious lesion. Studies on trace elements other than fluoride in carious dentin showed a fivefold increase in the zinc content (Table VI). Smaller Fig. 22. Section of soft carious dentin decalcified and stained with phos- photungstic acid. The intercanahcular matrix (M) displays collagenous fibrils (CF) in transverse section. Dentinal canals are filled with microorganisms (JB). (X approx. 10,450.) Fig. 23. Section of soft carious dentin decalcified and stained with phos- photungstic acid. Collagenous fibrils (CF) with typical banding are found in close proximity to bacteria (B). (X approx. 53,700.) Fig. 24. Section of soft carious dentin decalcified and stained with phospho- tungstic acid. Numerous microorganisms (B) are seen within the intercanahcu- lar matrix. A few collagenous fibrils (CF) persist between the microorganisms. ( X approx. 37,300. ) Fig. 25. Isolated collagenous fibril from homogenate preparation of carious dentin, exhibiting banding and devoid of crystallites. (X approx. 68,500.) Fig. 26. Section of intercanahcular matrix of sound dentin stained with phosphotungstic acid, showing a group of collagenous fibrils with distinct bands and interbands. (X approx. 31,100.) ^204 K. JOIIANSEN 5^ 2.0 - V— -v SOUND O— O CARIOUS 100 200 300 400 500 600 700 800 TEMPERATURE IN "C Fig. 27. The effect of heat treatment on the carbonate content of sound and cariou.s human dentin. increases in lead and iron were also noted. The relative significance of these elements in the carious process is not known. Discussion A matter of particular interest in relation to the carious process was the pattern of distribution of mineral in carious enamel and dentin. It has been shown that dissolution of enamel may extend to a depth TABLE V. Fluoride Content of Sound and Soft Carious Human Dentin (Ash weight; means of 38 samples. From Joliansen and Nordbaok, 1962) Fluoride (ppm) Sound Carious Carious/sound ratio: 10.4 ± 7.0 180 ± 25 1730 ± 1400 L'LTRASTRrCTURAL AND CHEMICAL STUDIES ON CARIES 205 TABLE VI. Trace Elements of Sound and Soft Carious Human Dentin (Ash weight; means of 12 samples. From Johansen, Steadman, and Steadman, to be pubHshed) Element (ppm) Zn Pb Sn Mn Fe Al Ag >Sr Sound Carious 318 1556 136 298 2 14 1.5 4 48 182 67 57 <0.4 1.3 120 106 1000 microns be\'ond the location of the l^acteiial plaque on the tooth surface (Sognnaes, 1962). Such lesions frequently show an intact and densely mineralized surface laver, while deeper regions may be extensively but not uniformly demineralized (Gustafson, 1957 ) . The electron microscopic findings show that apparent regional differences in degree of demineralization extend to the ultrastruc- tural level. In enamel it was occasionally seen that crystallites at the periphery of rods were better preserved than the majority of crystallites within rods. Similar advanced demineralization of prism cores has also been noted in microradiographs (Darling, chapter 6 in this volume). The electron microscopic observations on dentin further showed more extensive demineralization in intercanalicular areas than in pericanalicular zones. On the basis of the latter ob- servations it was hypothesized that the persisting crystallites in these regions were altered chemically and had become more resistant to dissolution (Johansen and Parks, 1959, 1961). This hypothesis is supported by the chemical data previously discussed (Johansen and Nordback, 1962) and summarized in this report. It is known that the surface layer of enamel shows an increase in fluoride and a decrease in carbonate with age (Brudevold et al., 1956; Little and Brudevold, 1958). This phenomenon may be part of a slow "maturation process" of dental tissues, not necessarily limited to the enamel surface. Deeper within the tooth, correspond- ing changes may occur on a limited scale in areas of relatively greater permeability. In enamel it is likely that rod sheath areas, because of their higher organic content, represent pathways of dif- fusion. In dentin, pericanalicular zones would appear to be more directly exposed to tissue fluid than would intercanalicular matrix. 206 E. JOHANSEX By a slow exchange phenomenon, cr\'stanites in regions of higher permeabihtv might pick up fluoride from the surrounding fluid and possibly lose carbonate. Such a change in composition could account for the greater resistance to dissolution observed in crystallites in specific regions. Lack of such a "maturation process " in the area might be associated with a more uniform pattern of demineraliza- tion. The marked increase in fluoride and the pronounced decrease in carbonate and magnesium observed in carious enamel and dentin, however, suggests more profound chemical changes than can be attributed to a maturation process, and an associated selective de- mineralization. The fact that manv crystallites in all regions ap- peared well preserved lends support to the idea that these crystallites at least in part are products of a recrvstallization process (Johansen and Parks, 1961 ) . By such a mechanism the fluoride liberated from dissolved crystallites could be incorporated into those persisting; also, additional fluoride (Dowse and Jenkins, 1957) and other ions might be available from oral fluids for incorporation into the apatite lattice. The low magnesium and carbonate values, on the other hand, might reflect a low concentration of these ions in the fluid environ- ment during recrystallization as well as preferential loss during de- mineralization. The recrystallization process would be dependent on changes in pH within the lesion and might involve an inter- mediate product of CaHPOi (Neuman and Neuman, 1958) as proposed by Gray and Francis (chapter 8 in this volume). The persistence of cr)'stallites within advanced lesions might thus be explained on the basis of a reduction in solubility resulting from changes in composition. The fate of the organic matrix in the carious process has long been a matter of controversv. The electron microscopic findings show that lesions in both enamel and dentin contain abundant organic ma- terial, and the chemical data confirm this observation. It has not, however, been possible to distinguish between extraneous and in- trinsic organic material in lesions of enamel. In dentin, the matrix of collagenous fibrils appears relativeh' intact in areas of advanced demineralization, but its composition and susceptibility to break- ULTRASTRUCTURAL AND CHEMICAL STUDIES ON CARIES 207 down by collagenase appear different from that of the sound tissue ( Armstrong, 196 !« ) . In conchision, it might be stated that the findings lend support to the general idea that demineralization precedes proteolysis in the development of carious lesions. It is evident, however, that the carious process in all its phases is a most complex phenomenon, at present not readily characterized in a single concept. Summary A series of electron microscopic and chemical studies were carried out to characterize carious lesions in enamel and dentin. The ultra- structural observations on the inorganic phase of the tissues were carried out on ultrathin sections, homogenate preparations, and triturated samples. After demineralization, the organic phase was studied in ultrathin sections stained with lead preparations and phosphotungstic acid, in unstained sections, and in homogenate preparations. The chemical composition was studied in samples that had been dried to constant weight or ashed. In carious enamel, ultrastructural observations on the inorganic phase often revealed structureless areas of varying width delimiting enamel rods. Crystallites were found to be sparsely distributed, with some evidence for the existence of microchannels between them. Those at the periphery of enamel rods occasionally appeared to be in a better state of preservation. Individual crystallites were found to vary considerably in morphology; some showed generalized and localized surface erosion, some displayed perforations, and others appeared as mere vestiges with little resemblance to crystallites found in the sound tissue. Segments of seemingly unaltered crystal- lites were also encountered. Demineralized carious enamel studied in stained sections showed evidence of organic material. Persistence of the organic matrix was indicated by the occurrence of rod sheaths, even though rods and interred areas frequently lacked the fine fibrils observed in the sound tissue. Sometimes an apparent increase in organic material was encountered. This finding was thought to be related to an influx of organic material from oral fluids. The electron microscopic observations on soft carious dentin re- 208 E. JOHANSEN vealed an interesting pattern of distril)otion of mineral. The highest concentration of crystalHtes was usually found adjacent to dentinal canals, whereas intercanalicular areas were extensively demineral- ized. Individual crystallites from both regions often appeared con- spicuously large, indicating resistance to dissolution, possibly due to a change in their composition. Studies of the organic phase showed a remarkably intact collagenous matrix with individual fibrils displaying banding and cross striations. It was concluded that proteolysis follows demineralization in dentinal caries. In the chemical studies of carious enamel and dentin it was found that the two tissues were in several respects similarly affected by the carious process. Both tissues showed an increase in water (and organic material) but a decrease in ash. Analyses of the ash fraction revealed relatively minor changes in calcium and phosphorus con- tent, but a marked decrease in magnesium and carbonate. The fluoride content of carious enamel and dentin was much higlier than that of the sound tissue. It was theorized that these chemical changes were, in part, brought about by a recrystallization process which re- sulted in the formation of crystallites of reduced solubility. Acknowledgments. Financial support for the original studies dis- cussed in this report was obtained from the following sources: the National Institute of Dental Research, United States Public Health Serv- ice, Research Grant D-689; the American Chicle Company; the Colgate- Palmolive Company; and the University of Rochester. The technical assistance of Diane Anderson in the preparation of the illustrative material is gratefully acknowledged. References Armstrong, W. G. 1961rt. A quantitative comparison of the amino acid composition of sound dentin, carious dentin and the collagenase re- sistant fraction of carious dentin. Arcli. Oral Biol., 5, 115-124. Armstrong, W. G. 1961/j. Reaction //} vitro of sound dentin with glucose, glucosamine and carbohydrate fermentation and degradation prod- ucts. Arch. Oral Biol, 5, 179-189. Bernick, S., Warren, O., and Baker, R. F. 1954. Electron microscopy of carious dentin. /. Dental Research, 33, 20-26. Bhussry, B. R. 1958. Chemical and physical studies of enamel from human ULTRASTRUCTURAL AND CHEMICAL STUDIES ON CARIES 209 teeth. Ill, Specific gravity, nitrogen content, and histologic charac- teristics of opaque white enamel. /. Dental Research, 37, 1054-1059. Biedermann, W., and Schwarzenbach, G. 1948. Die complexometrische Titration der Erdalkalien und einigen anderer Metalle mit Erio- chromschwarz T. Chimia, 2, 56-60. Brudevold, F. 1962. Chemical composition of the teeth in relation to caries. In Chemistry and Prevention of Dental Caries (R. F. Sogn- naes, editor), pp. 32-88. Charles C. Thomas, Springfield, 111. Brudevold, F., Gardner, D. E., and Smith, F. A. 1956. The distribution of fluoride in human enamel. /. Dental Research, 35, 420-429. Burrows, L. R. 1961. Hydroxyproline in enamel. /. Dental Research, 40, 749-750. Chen, P. S., Jr., Toribara, T. Y., and Warner, H. 1956. Microdetermination of phosphorus. Anal. Chem., 28, 1756-1758. Coolidge, T. B., and Jacobs, M. H. 1957. Enamel carbonate in caries. /. Dental Research, 36, 765-768. Dowse, C. M., and Jenkins, G. N. 1957. Fluoride uptake in vivo in enamel defects and its significance. /. Dental Research, 20, 816 (abstract). Frank, R. M., Sognnaes, R. F., and Kern, R. 1960. Calcification of dental tissues with special reference to enamel ultrastructure. In Calcifica- tion in Biological Systems (R. F. Sognnaes, editor), pp. 163-202. American Association for the Advancement of Science, Washington, D.C. Gustafson, G. 1957. The histopathology of caries of human dental enamel, with special reference to the division of the carious lesion into zones. Acta Odontol. Scand., 15, 13-55. Helmcke, J. G. 1962. Dental caries in the light of electron microscopy. Intern. Dental J., 12, 322-334. Hohling, H. J. 1961. Elektronenmikroskopische Untersuchungen am gesunden und kariosen Zahnschmelz unter besonderer Berijcksichti- gung der Ultramicrotomschnitt-Technik an nicht entmineralisierter Substanz. Deut. zahndrztl Z., 16, 694-705. Johansen, E. 1963. Microstructure of enamel and dentin. Proc. Conf. Phosphate and Dental Caries, Massachusetts Institute of Technology, Cambridge, Mass. In press. Johansen, E. 1964. On the ultrastructure and composition of enamel and dentin. /. Dental Research (in press). Johansen, E., and Nordback, L. G. 1962. The chemistry of carious lesions. III. The fluoride content of carious dentin. Intern. Assoc. Dental Re- search, Ahstr. 40th General Meeting, No. 143, p. 39. Johansen, E., and Parks, H. F. 1959. Preliminary electron-microscopic observations on carious dentine. /. Dental Research, 38, 693 (ab- stract). 210 E. JOHANSEN [ohansen, E., and Parks, II. F. 1960. Election microscopic observations on the three-dimensional morphology of apatite cr\'stallites of human dentine and bone. /. BiopJii/s. and Biochcin. (li/foL, 7, 743-746. |()hansen, E., and Parks, II. F. 1961. Electron-microscopic observations on soft carious human dentine. /. Dental Research, 40, 235-248. Johansen, E., and Parks, H. F. 1962. Electron-microscopic observations on sound human dentine. Arch. Oral Biol., 7, 185-193. Johansen, E., Steadman, L. T., and Steadman, S. G. To be published. Lenz, H. 1961. Elektronenmikroskopische Untersuchungen der Mine- ralisation des Zahnschmelzes und der beginnenden Schmelzkaries. Arch. Oral Biol, 4, 34-39. Little, M. F., and Brudevold, F. 1958. A study of inorganic dioxide in in- tact enamel. /. Dental Research, 37, 991-1000. MacFadyen, D. A. 1942. Determination of amino acids in plasma by nin- hvdrin-carbon dioxide reactions without removal of proteins. /. Biol. Chem., 145, 387-403. Manly, R. S., and Deakins, M. L. 1940. Changes in the volume per cent of moisture, organic and inorganic material in dental caries. /. Dental Research, 19, 165-170. McCann, H. G. 1959. Determination of microgram quantities of magne- sium in mineralized tissues. Anal. Chem., 31, 2091-2092. Megregian, S. 1954. Rapid spectrophotometric determination of fluoride with zirconium-eriochrome cvanine R lake. Anal. Chem., 26, 1161- 1166. Morse, A. 1945. Formic acid-sodium citrate decalcification and butyl alcohol dehydration of teeth and bones for sectioning in paraffin. /. Dental Research, 24, 143-153. Neuman, W. F., and Neuman, M. W. 1958. The Chemical Dynamics of Bone Mineral. University of Chicago Press, Chicago, 111. Scott, D. B., and Albright, J. T. 1954, Electron microscopy of carious enamel and dentin. Oral Surg., Oral Med. and Oral Pathol, 7, 64-78. Singh, B., and Johansen, E. To be published. Sobel, A. E., Rochenmacher, M., and Kramer, B. 1943. Microestimation of the inorganic constituents of bone. /. Biol. Chem., 152, 255-266. Sognnaes, R. F. 1962. Microstructure and histochemistry of caries. In Chemistry and Prevention of Dental Caries (R. F. Sognnaes, editor), chap. 1. Charles C. Thomas, Springfield, 111. Stack, M. V. 1954. The organic content of "chalky" enamel. Brit. Dental J., 96, 73-76. Takuma, S., and Kurahashi, Y. 1962. Electron microscopy of various zones in a carious lesion in human dentine. Arch. Oral Biol, 7, 439-444. VanSlyke, D. D., and Folch, J. 1940. Manometric carbon determination. /. Biol Chem., 136, 509-541. ULTRASTRUCTURAL AXD CHEMICAL STUDIES ON CARIES 211 ^^'atson, M. L. 1958. Staining of tissue sections for electron microscopy with heavy metals. II. Application of solutions containing lead and barium. /. Biophys. and Biochem. Ci/fol, 4, 727-729. W'illard, H. H., and Winter, O. B. 1933. \'olumetric method for determi- nation of fluorine. Iml. and Eng. Chem., AnaJ. Ed., 5, 7-10. 8 Physical Chemistry of Enamel Dissolution JOHN A. GRAY and MARION D. FRANCIS, Miami Valley Labora- tories, The Procter & Gamble Company, Cincinnati. Ohio INCIPIENT dental caries in enamel begins with the appearance of an opaque lesion, commonly known as a "white spot."' The formation of this stage of caries is the result of chemical dissolution of enamel substance and, therefore, should be amenable to description in terms of physical-chemical principles. Such a description requires an iden- tification of the factors affecting the process and a measurement of the magnitude of their effect on the process. The objective of this exposition is to present the physical and chemical factors affecting or controlling formation of an incipient carious lesion and to de- scribe the process on a physical-chemical basis. Recognition of the fundamentally chemical character of caries formation, beginning shortly before 1900 (Magitot, 1878; Miller, 1905), provided the inspiration for many investigations into the mechanism of this process. Concurrently with these investigations, a variety of systems, some of which included bacteria, were devised for producing the lesions in vitro, and, indeed, these methods were eminently successful. The essential process consists of the diffusion of the reactant to the reaction site, which in this case is the solid, enamel, followed by reaction and diffusion of the products away from the reaction site. Such an interaction between a solution and a solid is known as a heterogeneous reaction. In most cases, the rate of heterogeneous reactions is controlled by the rate of diffusion of 213 214 J. A. GRAY AND M. D. FRANCIS the reactaiit or solute to the sohd surface. The diffusion rate is de- termined h\ the concentration of the reactant, and, therefore, tlie rate of reaction with the sohd is directly proportional to the concen- tration of the reactant. Equilibrium solubilitv must also be con- sidered because it mav be attained frequentb, and studies at equilibrium provide information which is difficult or impossible to obtain from rate data. These phvsical-chemical principles provide the basis for examining the formation of the carious lesion. The following presentation can be regarded as a study of a model system for producing caries-like lesions irrespective of the chemical agents which cause caries. The primarv reactant examined in this discussion is acid or acidic buffer. The model based on such an acid reactant is considered most promising because ( 1 ) adequate quanti- ties of acid have been shown to exist in the immediate en\ ironment of the teeth, (2) an adequate substrate, hvdrox\'apatite, exists on which the acid can operate, and (3) caries-like lesions have been produced in vitro with acid. The only other agents ever considered extensively as the primarv reactants for enamel are proteolvtic enzvmes and complexing com- pounds. However, several attempts to produce a proteolvtic enzyme effect in enamel resulting in caries-like lesions have not been success- ful (Burnett and Scherp, 1952; Jenkins, 1961). Furthermore, the rather small supply of substrate (i.e., organic content of enamel) upon which the proteolytic agents can act ( Orban, 1957; Schatz et al., 1956) makes this approach less logical. It does appear conceiv- able that compounds which can form strong soluble complexes with calcium, such as ethylenediaminetetraacetate, could be used to pro- duce caries-like lesions in enamel (Ravnik and Loe, 1961), although no major attempts have been made to complete such a study. Many complexing agents have the necessary capability for dissolving the substrate, the inorganic portion of enamel, just as does acid. How- ever, this concept becomes untenable in vivo because complexing agents of sufficient strength and in adequate concentration in the vicinity of the enamel surface (other than hydrogen ion) seem scarcely possible on the basis of a quantitative consideration of po- tential sources such as the organic moietv of enamel, bacteria, or I PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 215 other organic constituents of plaque. A critique of axailable experi- mental evidence supports this view (Jenkins, 1S61). On the other hand, hydrogen ion, in reality an extremely strong complexing agent for phosphate and hydroxyl groups, is adequately available. Should other agents be found important, however, they would also have to complv with the physical-chemical principles set forth earlier. Morphology^ and Formation of Incipient Carious Lesions: A Brief Review Carious Lesion Morphology The morphology of an incipient carious lesion has been well de- scribed by Nishimura (1926), Darhng (1956), Gustafson (1957), and Schmidt and Keil ( 1958 ) , and is reviewed by Darling in chapter 6 of this volume. An example of an incipient carious lesion as de- scribed in these works is illustrated with photomicrographs ( Figs. 1, 2, 3a, 3b, and 3c) taken of a section through a natural white spot. The typical early lesion consists first of an apparently sound layer of enamel at the surface. Beneath this surface layer is a region of decalcification where enamel substance has been removed in quan- titv. These regions can be observed by light microscopy (Fig. 1), but the proof of relative degrees of demineralization is provided by microradiography (Fig. 2). It is clearly apparent that there is a reduced mineral content in the decalcified subsurface region and a high mineral content in the relatively sound outer layer. Some loss of enamel substance from this relatively sound outer layer has been demonstrated bv comparison with the outer layer of normal enamel, using precise microradiographic measurements ( Soni and Brudevold, 1959). The incipient carious lesion can be separated further into zones by observation with polarized light using a full-wave plate, first order red (Gustafson, 1957). A lesion showing these zones of birefringence ( Fig. 3a, the same specimen as in Fig. 1 ) can be compared with the diagrammatic representation of Gustafson (Fig. 3b). In darkfield illumination, the lesion has a characteristic lighter appearance ( Fig. 3c). 216 J. A. GRAY AND M. D. FRANCIS 2 Figs. 1 and 2. A section through a typical natural incipient carious lesion with the original enamel surface retained, cut transversely from the mesial sur- face of an upper right first molar. Fig. 1. Viewed with ordinary transmitted light, the lesion is apparent by the increase in contrast of the structural features as compared with normal enamel. (X 200.) Fig. 2. A microradiograph of the lesion demonstrates the loss of material in the lesion that accentuates the interrod spaces. The relatively sound layer near the surface is readily apparent. ( X 200. ) Therefore, an incipient carious lesion will be defined as a sub- surface region of decalcification overlain by a relatively sound layer, with the surface completely intact, at least according to visual or light microscopic observations. It thus becomes apparent that the reactant must, for the most part, be by-passing the surface to diffuse into the enamel and react with subsurface enamel components. PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 217 Carious Lesion Formation Incipient carious lesions of enamel formed in the oral cavity com- monly occur under a bacterial plaque. The metabolic products of the bacteria that compose the plaque are almost certainly the de- structive agents responsible for the attack on enamel. Unfortunately, definition and characterization of plaque is too incomplete to permit a chemical simulation of this system. In addition to establishing the composition and morphology of the entire plaque, which consists mostly of bacteria, it is necessary to define the composition of the free aqueous phase. Furthermore, knowledge of the interfaces exist- ing at the exposed plaque surface and the enamel-plaque junction is needed for understanding the transport of the various chemical species from one site to the other. At the moment, one of the most deficient areas in caries research is this exact knowledge of the plaque system, which can at best be described as a complicated mix- ture of bacteria, enzymes, acids, other organic compounds, and in- organic ions such as calcium and phosphate. This enamel-plaque complex is the system of ultimate interest that needs to be described by rational, scientific principles as well as simulated in the laboratory. The earliest attempts at producing carious lesions in the laboratory (Magitot, 1878; xMiller, 1905; Pickerill, 1912) made use of mixtures of saliva with foodstuffs such as sugar or bread, and, in some cases, acidic solutions alone. These mixtures incubated with teeth for vary- ing periods of time, but usually quite long, produced defects re- sembling both the incipient stage and the fully developed lesion with cavity formation. Such lesions covered the entire enamel sur- face unless the area of exposure had been limited by protective coatings. The importance of these experiments lay in the fact that caries-like lesions could be produced outside the oral cavity. Com- plicated as some of these systems were, they were more suitable for experimental study than the oral situation. The next advances in the in vitro production of caries-like lesions followed two directions. One consisted in studying known bacterial cultures and nutrient systems, while the other involved elimination of bacteria to use only chemical systems. The classical example of the former is the "artificial mouth" devised by Pigman et al. ( 1952 ) . Figure 3 Fig. 3a. The section of the natural lesion of Fig. 1 as viewed with polarized light using a full-wave plate, first order red, has the typical appearance de- scribed and illustrated (Fig. 3b) by Gustafson (1957). All zones except the last highly decalcified fifth zone of an advanced lesion are visible. ( X 200. ) Fig. 3b. A schematic representation of an incipient carious lesion mounted in Canada balsam, as viewed with polarized light using a full-wave plate, first order red, according to Gustafson, shows the five characteristic zones, (l^epro- duced from Acta Odoutol. Scand., 15, Fig. X (1957), with the kind permission of the author and editor. ) Fig. 3c. The natural lesion of Fig. 1 as viewed in darkfield illumination with ordinary light. A lesion viewed this way appears lighter than the usual bluish gray of normal enamel and sometimes will almost glow as the surface does in this sample. However, the surface glow in this case was an edge reflection effect unrelated to the presence of the lesion, (x 200.) Fig. 3d. A section of a lesion produced by a 96-hour exposure at 37 °C to a medium consisting of 0.002 m (^aCL plus 6 per cent hydroxyethyl cellulose adjusted to pH 4.5 with hvdrochloric acid, as viewed with polarized light using a full-wave plate, first order red. The similarity to the natural lesion of Fig. 3a and the diagram of Fig. 3b is apparent. (X 200.) Figs. 3e and 3f. The same sections as in Figs. 8 and 9 and Figs. 12 and 13, showing the effect of increasing exposure time on lesions produced at 4°G with a medium of 0.05 m lactic acid and 0.03 m CaCL plus 6 per cent hv- droxvethyl cellulose adjusted to pH 3.5 when viewed with polarized light, full- wave plate, first order red. Fig. 3e. After a 30-hour exposure only zone 1 of dark blue or purple de- scribed bv Gustafson was visible. The sound outer layer was just appearing. (X200.)' Fig. 3f. At 120 hours, zone 2 (gold band) had appeared. (X 200.) Fig. 3g. A section of a lesion produced by a 96-hour exposure at 37 °C to a medium consisting of 0.01 m lactic acid and 0.03 m CaClo plus 6 per cent h\'droxyethvl cellulose (with high buffer content) adjusted to pH 3.5. At this high rate of decalcification a whitish zone was formed in place of the usual gold appearance of zone 2. ( X 200. ) Fig. 3h. A section of a lesion produced by interrupted exposures at 37 °C to a medium consisting of 0.05 M lactic acid and 0.03 m GaClo plus 6 per cent hydroxyethyl cellulose adjusted to pH 4.0. The sample was removed every 48 hours, washed 5 minutes with distilled water, and returned to tlie decalcification medium. The total exposure was 240 hours. Zone 2 (gold band) became very broad by reason of the long cycling exposure. (X 200.) Fig. 3i. A section of enamel (same as in Fig. 37) after a 15-day exposure at room temperature to a solution of hydrochloric acid at pH 4.5, viewed with polarized light using a full-wave plate, first order red. A decalcification near the svnface is visible and there are signs of zones similar to those seen in Figs. 3a and 3b. (X 200.) Fig. 3j. A section of enamel (same as in Fig. 38) after a 15-day exposure at room temperature to an aqueous solution of 0.0005 m CaCL and 0.0005 M NaHoP04 adjusted to pH 4.5, viewed with polarized light using a full-wave plate, first order red. The appearance of the lesion bears no resemblance to Figs. 3a and 3b. The uniformitv throughout the lesion is striking. (X 200.) 218 Figure 3 PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 219 This system permitted the production of lesions in all the various stages and demonstrated the wide variety of defects obtainable with different l)acterial strains. Several other methods have been devel- oped for producing incipient caries-like lesions using a nutrient solution mixed with saliva or bacteria (Gore, 1940; Darling, 1956; Newbrun et ciL, 1959; Soni and Bibb\-, 1961). Francis and Meckel ( 1963 ) devised such a method using an agar liase containing the nutrient and salix a that produced lesions in a matter of days and was sufficiently well controlled for use as a routine test method. A bacterial medium inoculated with Lactobacillus casei and sterilized after a suitable incubation time was used to show that the products of bacterial metabolism, such as acids, were responsible for produc- ing the caries-like lesions rather than the l:)acterial enz\mes or the bacteria per se (Briner, 1963). Teetli that were immersed in such a medium developed the same defects as those found in a medium containing viable organisms. The production of such lesions in en- amel has been achiexed using onlv acid mixed with a viscous or gel base, omitting entireh both the organisms and the nutrients (von Bartheld, 1958; Miihlemann, 1960; Opdyke, 1962). One of the more extensive studies of a chemical svstem for pro- ducing incipient caries-like lesions in enamel was done using acidic solutions containing calcium and phosphate (Coolidge et al., 1955). A definite relationship among the hydrogen, calcium, and phosphate ion concentrations was found to be necessarv in order to obtain char- acteristic incipient lesions. Similar observations were made using a novel technique whereby the progress of the lesion could be ob- served on a section of a tooth (Wachtel and Brown, 1963). A de- tailed study was also made of lesions produced bv lactic acid alone (Hals et al., 1955), and the various combinations of subsurface dis- solution and surface dissolution were descril:)ed. Another investiga- tion was made using acetic acid alone, where the lesions were examined with polarized light and microradiography (Soni and Brudevold, I960). In a stud\' of the dissolution of enamel in acidic bufi^ers, the im- portant factors afi^ecting the reaction between acid and hvdroxy- apatite were isolated (Gray, 1962). The rate of dissolution was found to be primarily dependent on and proportional to the undis- 220 J. A. GRAY AND M. D. FRANCIS sociated acid concentration. The rate was also inhibited by the presence of cations and selected anions. Although this study dealt only with surface decalcification, and the white spots typical of incipient carious lesions were not produced, the relationships de- termined also apply to some stages of incipient carious lesion forma- tion. On this basis, an elementary mechanism was proposed for incipient carious lesion formation (Gray et ah, 1962) founded on the diffusion of undissociated acid into enamel to cause subsurface decalcification. In addition, it was suggested that this system was influenced by the presence of an organic coating to protect the en- amel surface, as will be discussed subsequently. More recently, the rate of growth of incipient carious lesions was measured (Kapur et ah, 1962) using a technique similar to that of Wachtel and Brown ( 1963 ) . The lesions were produced with aque- ous solutions of either lactic acid or acetic acid, and the formation rate was determined by measuring the depth of the lesion as a func- tion of exposure time. The rates of formation were then measured as a function of pH and buffer concentration. The significant con- clusion reached in this study was that the rate of formation was proportional to the concentration of undissociated acid following, in essence, the mechanism proposed by Gray et ah ( 1962 ) . Study of Incipient Carious Lesion Formation Using Chemical Systems Cardinal Factors Affecting Decalcification Systems The study of the physical chemistry of incipient carious lesion formation required the development of a system that would imitate the in vivo circumstances and would be suitable for a quantitative evaluation of the variables of the system. Acid or acidic buffer con- stituted the fundamental element of the dissolution or decalcification medium. Lactic acid was chosen on the basis of its being one of the important acid products of bacterial metabolism. The organic coating protecting the enamel surface, mentioned previously, was considered essential, and a material to provide such a coating con- stituted another fundamental element of the decalcification medium. PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 221 The major reaction products — calcium and phosphate — also were incorporated into the decalcification medium at appropriate stages in the study. Carbonate, another reaction product, was not con- sidered in this preliminary investigation, but may have to be ac- counted for in later, more sophisticated systems. It is obvious that if a homogeneous solid sample of hydroxyapatite is exposed to acid, attack will begin at the surface. It is also obvious that if diffusion pathways within such a sample are available, the acid will tend to diffuse into those regions of the sample having lower acid concentrations. The presence of diffusion pathways in normal sound enamel has been amply demonstrated (Pickerill, 1912; Sullivan, 1954 ) . Therefore, the logical expectation for a system con- sisting of enamel exposed to acid is that acid will diffuse both to the enamel surface and into the enamel. If the acid is consumed very rapidly at the surface, then little or no acid will remain to diffuse into the enamel. If, on the other hand, the surface reaction is re- tarded, more acid will diffuse farther into the enamel until subsur- face reaction occurs. Thus, conceivably, there could be competition between ( 1 ) reaction at the surface and ( 2 ) diffusion into enamel with subsequent subsurface reaction. In the event that the rate of the first step could be greatly reduced relative to the second step, then the result would be an incipient caries-like lesion. The studies by Hals et al. (1955) support this idea of a competition between surface and subsurface reaction. The wide variety of conditions encountered in the formation of incipient carious lesions in vivo and in vitro poses the problem of how the surface reaction can be uniformly retarded in all these systems. A review of both the natural and the artificial systems which have been reported indicates, as far as can be determined, that in every case organic material was present. This conclusion is based on the following considerations. In the oral cavity there are present the enamel cuticles, bacterial plaque, and saliva. In vitro systems using bacteria contain proteins, polysaccharides, mucoproteins, or mucopolysaccharides. The enamel surface as it occurs in the oral cavity is less soluble than subsurface enamel, and this property has been ascribed, in part at least, to the presence of certain inorganic 222 J. A. GRAY AND M. D. FRANCIS ions, such as fluoride (Brudevold, 1948). It lias been shown, how- ever, that a reduction in solubihtx approaching ahnost complete protection results from the naturally occurring organic coating on enamel, the acquired enamel cuticle ( Meckel, 1961 ) , and this de- posit has since been found to penetrate enamel to a depth of several microns, which makes its removal difficult. In previous studies of chemical svstems which produced caries-like lesions in enamel, the original enamel surface was left essentially intact, retaining either all or remnants of the insoluble organic coatings and the inorganic ions which reduce enamel solubilitv. On the basis of the preceding argument, an organic material was considered to be a necessarv component of anv decalcification me- dium in addition to acid and acidic buffer. Furthermore, in order to define the role of the organic material to be used in making in- cipient carious lesions in vitro, the enamel should retain none of the original surface and should consist entirelv of subsurface enamel. For this purpose, all samples of enamel used in the present studies were acid etched after mounting and ground with abrasive to re- move the surface laver as ascertained by solubility rate measure- ments (Gray, 1962). Enamel samples consisted of squares (ap- proximately 4X4 mm- ) cemented to the ends of a plastic rod with a dental plastic. All sides of the enamel were coated with the plastic cement, leaving onlv the surface exposed. A suital:)le organic material was selected experimentalh'. An ex- amination of past effective in vitro systems indicated that such a material should be polymeric. Meckel (personal communication) in studying the components of the agar-saliva system (Francis and Meckel, 1968) found agar, saliva, proteins, and selected amino acids to be effective to varying degrees. A variety of water-soluble poly- mers was then tested using synthetic chemicals, natural products, and derivatives of natural products in an attempt to find an effective, well characterized material. The test consisted of mixing the polymer with a solution of 0.05 m lactic acid, immersing a sample of enamel in a few grams of the medium, and examining the sample visuallv at intervals for defects resembling incipient carious lesions. Con- trol areas were provided by coating a corner of the enamel with a dental cavitv lining varnish. The qualitative results of such tests PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 223 TABLE I. Effectiveness of Water-Soluble Polymers INIixed with Acidic Buffer for the Formation of Incipient Caries-like Lesions AND Inhibition of Dissolution of Enamel Surface Name Effectiveness Hydroxyethyl cellulose Excellent SeaKem (carrageenan, an acidic polysaccharide) Good Tomato juice agar Good Carboxymethyl cellulose Fair Carboxymethylhydroxyethyl cellulose Fair Polyvinylpyrrolidone Fair Methyl cellulose Fair to poor Polyox (ethylene oxide condensate) Poor Cab-o-sil (fineh' divided silica) No effect Glucose No effect Polyvinyl alcohol No effect with several materials are listed in Table I. Synthetic polymers of simple structure, such as ethylene oxide condensates or polyvinyl alcohol, as well as an inorganic thickening agent, finely divided silica, were completely ineffective. The solutions dissolved the enamel surface just as would have occurred if the polymer had been absent. The most effective compounds were polysaccharides, such as derivatives of cellulose. The testing of eligible materials was by no means exhaustive, but complexity, lack of purity, and unde- sirable ionic properties precluded interest in many undoubtedly ef- fective compounds. Hydroxyethyl cellulose, a nonionic synthetic derivative of cellulose, was unusually effective and was selected as the polymeric ingredient of the decalcification medium. The choice of hydroxyethyl cellulose was based on (1) effectiveness, (2) non- ionic character, (3) inertness to salt effects, (4) the fact that hy- droxyethyl cellulose thickened rather than solidified the solutions and permitted a good contact ]:)etween enamel and the decalcifica- tion medium, and (5) minimal changes in viscosity over the tem- perature range of interest. The particular samples used were found to contain an acidic buffer salt, sodium acetate, which had to be taken into account when total undissociated acid was calculated. The amount of this buffer capacity was determined from pH titra- tion curves using both acid and base titrants. 224 J. A. GRAY AND M. D. FRANCIS Comparison of in Vitro and Natural Lesions The equivalence of natural incipient carious lesions to those pro- duced by the in vitro system will be established, and, at the same time, the effects of some of the variables will be qualitatively de- scribed. The microscopic appearance, using ordinary transmitted light (Fig. 4), of a lesion produced in 24 hours with a solution of 5 Figs. 4 and 5. A section through an incipient caries-hke lesion prodviced by a 24-hour exposure at 4°C to a medium consisting of 0.05 m lactic acid plus 6 per cent hydroxyethyl cellulose adjusted to pH 3.5. Fig. 4. In ordinary transmitted light, the lesion can be seen to have pene- trated about 30 microns. The right side of the surface, protected from the medium by a varnish coating, provides a control for comparison. ( X 200. ) Fig. 5. A microradiograph shows that a considerable loss of material from the subsurface lesion had occurred. There is slight evidence of a relatively sound outer layer at the surface over the decalcified region of the lesion. The surface of this sample, when viewed under a metallurgical microscope with top illumination, was slightly dissolved by the medium. (X 200.) PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 225 hydroxyethyl cellulose and lactic acid adjusted to pH 3.5, along with a microradiograph (Fig. 5) of the same section of enamel, illustrates the type of result obtainable. Lesions produced by this system definitely had a more sound layer, albeit not very deep, overlying the subsurface decalcified region. The surface of the enamel sample was only slightly changed, as determined by micro- scopic observation using a metallurgical microscope with top illumi- nation. The depth of the relatively sound outer layer at the surface can be enhanced by several devices: (1) increasing the exposure to the de- calcification medium, (2) removing and returning the enamel sample to the decalcification medium at intervals, (3) adding cal- cium ions initially in the decalcification medium, (4) adding both calcium and phosphate ions initially in the decalcification medium, (5) increasing the buffer concentration, and (6) including one of many select cations, such as zinc, in the decalcification medium. Examples of the lesion that resulted from adding calcium as CaCL to the decalcification medium are presented in Figs. 6 to 13 for exposures of 16, 30, 90, and 120 hours. Figures 14 and 15 illus- trate the effect of including both calcium ( as CaCl2 ) and phosphate (as NaH2F04) to the medium (all at pH 3.5). There was a pro- gressive increase in the depth of the relatively sound outer layer and in the loss of material in the decalcified region with each successive addition to the decalcification medium. Also, the enamel surface was now unchanged as determined microscopically. Adding phosphate (as NaH2P04) alone caused severe surface damage. An even more dramatic improvement was found by doubling the lactate buffer concentration (Figs. 16 and 17) or by interrupting the exposure periodically (Figs. 18 and 19). The similarity of these lesions to natural lesions can be seen by comparison with Figs. 1 and 2. Add- ing zinc as ZnCl2 had the same qualitative effect as CaCL, although higher concentrations of ZnCL were required for quantitative equivalency. The various zones observed in natural lesions using polarized light with a full-wave plate, first order red, have been well described pre- viously ( Gustafson, 1957 ) , and a diagrammatic representation is re- produced (Fig. 3b). A chemically produced lesion with all zones S226 .1. A. CRAY AND M. I). FRANCIS PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 227 except number 5 is shown in Fig. 3d. Zone 5, vvhicli represents the ahnost completely destro)'ed enamel, is beyond the scope ot this study, but can be obtained by carrying out sufficiently long ex- posures. The progression of the lesion began with zone 1 or zone 3 (dark blue or purple band near the surface, Fig. 3e), these zones being indistinguishable from each other. The relatively sound outer layer was formed next. This was followed by formation of zone 2 (gold band), which then separated zones 1 and 3 (Fig. 3f). Zone 4 (light blue band above zone 2) developed as the lesion progressed, fol- lowed by zone 5. Although a detailed stud\ of the formation of these zones as a function of the man\^ \ ariables of the decalcification was not made, a few relationships were discovered. The most significant was that at very high decalcification rates the yellow or gold birefringence colors of zone 2 became almost white with overtones of yellow, green, or blue (Fig. 3g). As the rate of decalcification decreased, hints of yellow and finally the familiar gold color appeared. At high decalcification rates, zone 2 was usually very narrow (Fig. 3f). Periodic interruption of tlie decalcification tended to broaden zone Figs. 6 to 13. The effect of increasing exposure time on the incipient carious lesion formed by exposure of enamel to a medium at 4°C consisting of 0.05 m lactic acid and 0.03 m CaCL plus 6 per cent hydroxyethyl cellulose adjusted to pH 3.5 is demonstrated with light micrographs (Figs. 6, 8, 10, and 12) and microradiographs (Figs. 7, 9, 11, and 13) of sections from enamel samples ex- posed to the medium for different time periods. Figs. 6 and 7. After a 16-hour exposure, decalcification at or near the sur- face, extending about 10 microns deep, was visible. (X 200.) Figs. 8 and 9. After a 3()-hour exposure, the lesion had progressed to about 25 microns and the relatively sound outer layer had appeared. ( X 200.) Figs. 10 and 11. After 90 hours, the lesion had penetrated about 50 mi- crons and the relativelv sound outer laver was about 5 microns wide. ( X 200.) Figs. 12 and 13. After 120 hours, there was a further increase in loss of subsurface enamel material and an enhancement of the relativelv sound outer layer. About 50 per cent of the enamel had been dissolved from the lesion, according to calculations from the measured dissolved phosphate shown in Fig. 26. The surfaces of all the samples were almost untouched by the de- calcification medium as determined microscopicallv. The depth of the lesion across each sample was also quite uniform. (X 200.) 228 J. A. GRAY AND M. D. FRANCIS ^6 17 HHHHHI 19 Figs. 14 and 15. A section of an incipient carious lesion produced by a 96-hour exposure at 4°C to a medium consisting of 0.04 m lactic acid, 0.03 m CaCl2, and 0.01 m NaHoP04 plus 6 per cent hydroxyethyl cellulose adjusted to pH 3.5. The right side of the enamel surface was protected from the medium by a varnish coating. Fig. 14. The light micrograph, taken with ordinary light, shows a lesion about 75 microns in depth and containing three bands. ( X 200. ) Fig. 15. The microradiograph shows both the very great loss of subsurface enamel and the very considerable relatively sound outer layer. The surface of the enamel was unchanged as determined microscopically with top illu- mination. The increase in decalcification as compared with Figs. 10 and 11 [Continued at bottom of following page] PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 229 2 (Fig. 3h). Thus the decalcification mediums not only reproduce the zone differences of a natural incipient carious lesion, but also offer a means of studying these peculiar and little-understood changes. An additional comparison of the natural and in vitro lesions was made by examination with electron microscopy (Figs. 20, 21, and 22 ) . Diamond knife sections cut from undemineralized samples were used (Neal and Murphy, in preparation). Loss of substance from the interrod region was clearly apparent in both the natural (Fig. 20) and the chemically produced lesions (Figs. 21 and 22). The higher content of matter in the relatively sound outer layer was also demonstrable, as were the open pathways between the rods in this region. The advancing front of the lesion deeper in the enamel could be seen as an enlargement of the space between the rods that gradually disappeared in the normal enamel, while the body of the lesion consisted of large spaces with only scattered remains of the original enamel structure. The attack could be seen as beginning at resulted from the increase in acidic buffer capacity of phosphate, although the presence of phosphate further reduced attack on the enamel surface. ( X 200.) Figs. 16 and 17. A section of an incipient carious lesion produced by a 24-hour exposure at 4°C to a medium consisting of 0.10 m lactic acid and 0.03 M CaCl2 plus 6 per cent hydroxyethyl cellulose can be compared with Figs. 8 and 9 to see the efiFect of increasing lactate buffer concentration. Fig. 16. In the light micrograph a very sharply defined relatively sound outer layer can be seen above the body of the lesion. ( X 200. ) Fig. 17. The microradiograph confirms the presence of the relatively sound outer layer and the loss of subsurface mineral. ( X 200.) Figs. 18 and 19. A section of an incipient carious lesion produced by five 24-hour exposures at 4°C to fresh samples of a medium consisting of 0.05 m lactic acid and 0.03 m CaClo plus 6 per cent hydroxyethyl cellulose adjusted to pH 3.5. Fig. 18. A light micrograph shows the depth of the lesion and some bands within the lesion. ( X 200.) Fig. 19. A microradiograph proves the presence of a very wide relatively sound outer layer, amounting to 20 microns, and the loss of material from deep within the enamel. Interrupting or cycling the attack led to an increase in the sound outer layer, deeper penetration of the lesion, and slower de- calcification of the body of the lesion. (X 200.) ^30 J. A. CRAY AND M. D. FRANCIS FlGLTTlES 20 AND 21 PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION '^Sl the outside of the rods and slowly permeating to the center of the rod. Diamond knife sections of these samples with the embedding ma- terial ( methyl methacrylate ) remo\ ed were then demineralized with hydrochloric acid (0.005 n) to remove the inorganic substance and expose the organic content (Figs. 23 and 24). Normal mature enamel (Fig. 23) had a sparse organic residue which generally collapsed during the demineralizing procedure, so that little or no correlation with the original enamel structure was possible. In the incipient carious lesion region, produced either naturally or in vitro ( in vitro, Fig. 24), the content of organic residue appeared to be much more appreciable and was easily correlated with the enamel rod structure, similar to enamel matrix of unerupted teeth, as well as the original undemineralized specimen (Fig. 21). The higher than normal or- ganic content of natural lesions has been demonstrated by other investigators using chemical analyses and light microscopy ( Bhussry, 1958; Johansen, 1962). The organic residue found in the present study seems to be concentrated in the enamel rod proper, where the greatest mineral content was observed. Furthermore, the highest organic content was always found in the relatively sound outer laver of the lesion at the surface and appeared to be located in the same vicinity as the mineral. Additional studies are in progress to identify the source of these residues and their relationship to the organic moietv of normal enamel. The reality of the spaces observed in the incipient carious lesion by electron microscopy (Figs. 20, 21, and 22) can be demonstrated bv infiltrating the enamel with a material which is visible in electron Fig. 20. An electron micrograpli of a section through a natural incipient carious lesion cut from an undemineralized sample of adult human enamel us- ing a diamond knife. The interrod spaces have been enlarged by the decay process and material has been leached out of the rod proper. (X 5000.) The appearance of normal enamel can be seen in parts of Fig. 22 or 36. Fig. 21. An electron micrograph of a section through an incipient cariovis lesion produced by a 72-hour exposure at 37°C to a medium consisting of 0.05 M lactic acid and 0.03 m CaCb plus 6 per cent hydroxyethyl cellulose at pH 3.5. The appearance is the same as that found in the natural lesion of Fig. 20, with loss of mineral from the interrod region and some leaching out of the rod interior. Undemineralized. ( X 5000. ) 232 J. A. GRAY AND M. D. FRANCIS Fig. 22. An electron micrograph at low magnification showing an over-all view of an undemineralized section of a lesion produced by a 24-hour exposure at 4°C to a medium consisting of 0.10 m lactic acid and 0.03 m CaClo plus 6 per cent hydroxyethyl cellulose adjusted to pH 3.5, from the same sample as Figs. 16 and 17. The enamel surface is in the upper right corner. The large areas containing no enamel coincide with the subsurface decalcified region seen in the microradiograph of Fig. 17. The relatively sound outer layer is still heavily mineralized, but some channels through this region following the inter- rod spaces are visible. The decalcification front approaching normal enamel is on the left side and was very obviously moving along between the rods. Sections of normal enamel have little or no space between the rods (upper left-hand corner of this figure or lower right-hand corner of Fig. 36). (X 2000.) micrographs. The embedding material, methyl methacrylate, proved suitable for this purpose. It was found that the methacrylate pene- trated to the very depths of the lesion and filled extremely small voids. The location of the methacrylate can be better visualized by demineralizing the sections with hydrochloric acid. Not only were extensive voids created in enamel during incipient carious lesion fonnation, but these voids were readily available to diffusing ma- terials. The lesions made in vitro advanced between the rods, along the striae of Retzius, and across prism cross striations. By all present PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 233 methods of observation, including histochemistry (Opdyke, 1962), the incipient carious lesion produced by the in vitro system is iden- tical in appearance with natural lesions. Variables Affecting Rate of Carious Lesion Formation The rate of formation of the lesion produced in vitro was studied by quantitative measurements of calcium and phosphate released from enamel as a function of pH, buffer concentration, calcium con- centration, and temperature. The dissolution of enamel during incipient carious lesion forma- tion was measured as a function of exposure time. Both an integral method and a differential method for determination of reaction rates were used (Laidler, 1950). The amount of enamel dissolved during exposure to the various mediums was determined by an- alyzing chemically for calcium (Welcher, 1958) and phosphate ( Lucena-Conde and Prat, 1957). The calcium and phosphate ana- lytical results were converted to weight of enamel, using its reported composition (Orban, 1957), and rechecked by analyzing known weights of dissolved enamel. By way of an example, the first medium to be discussed contained 0.05 m lactic acid, and hydroxyethyl cellu- lose was added in an amount of 6 gm per 100 ml of the lactic acid solution (hereafter referred to as 0.05 m lactic acid plus 6 per cent hydroxyethyl cellulose). The initial pH was adjusted with concen- trated sodium hydroxide or hydrochloric acid as required. The amounts of enamel dissolved, determined by both phosphate and calcium analyses of the same solution (Fig. 25), were in good agree- ment, indicating no appreciable preferential dissolution of one com- ponent over the other during incipient carious lesion formation. This correspondence held during all subsequent measurements, and only the results of the phosphate analyses are presented in the figures to follow. The coeflBcient of variation in the determination of dissolved enamel was ±11 per cent based on phosphate analysis and ±13 per cent based on calcium analysis across different enamel samples and experimental conditions. The enamel samples used in these studies were examined for surface damage with a metallurgical microscope using top illumination, and then were sectioned with a special dental saw (Gray and Opdyke, 1962) for observation by 234 J. A. GRAY AND M. D. FRANCIS Figures 23 and 24 PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 235 niicroradiograph\' and light microscopv and as a prcliminaiv step in sample preparation for electron microscop\'. The thickness of the sections xaried between 50 and 100 microns. The sections were mounted on glass slides with Permount after microradiographs were taken. The rate of enamel dissolution during incipient carious lesion formation followed a typical solubility curve (Figs. 25, 26, and 27). In every case, the rate of dissolution decreased with increasing length of exposure time, and the curves appeared to be approaching equilibrium. The decrease was not due to saturation of the bulk solution in all cases, as identical rates were found when the volume of the decalcification medium was increased tenfold. It is apparent that the dissolution rate is limited bv changes in the composition of the medium in the vicinitv of the enamel because of lack of time for equilibration with the bulk solution. Adding calcium chloride to a medium containing 0.05 m lactic acid plus 6 per cent hvdroxyethvl cellulose at pH 3.5 (Fig. 26) retarded the rate ( Fig. 27 ) . As pointed out previously, however, this addition caused a better-defined "relatively sound outer layer." In- creasing pH of this same medium at a constant buffer strength caused a marked reduction in the rate of enamel dissolution (Fig. 27). Increasing temperature from 4° to 37 °C had almost no effect (Figs. 25 and 26). Further quantitative evaluation of the different variables was done bv a differential method measuring the rate of dissolution at only Fig. 23. An election micrograpli of a section of normal enamel cnt with a diamond knife and then demineralized with 0.005 n hydrochloric acid. The residue, which appears to be organic, is believed to be the remains of the or- ganic matrix of enamel, but little of the original enamel rod pattern is recog- nizable. (X 5000.) Fig. 24. An electron micrograph of a section cut from the sample of Fig. 21 and demineralized with 0.005 n hydrochloric acid. The incipient carious lesion was produced bv a 72-hour exposure at 37°C to a medium consisting of 0.05 iM lactic acid and 0.03 m CaCL plus 6 per cent hydroxvethyl cellulose at pH 3.5. The enamel rod pattern is very apparent and resembles the picture in Fig. 21. The organic residue is located primarily in the rods and is more highly concentrated than in normal enamel. The highest concentration is always found at the surface (upper left-hand corner) in the relatively sound outer laver. (X 5000.) ^36 J. A. GRAY AND M. D. FRANCIS 16 E o E Q IxJ §8. CO V) o _J UJ < 4 z UJ AT 4°C • CALCULATED FROM PO4 ANALYSES A CALCULATED FROM Ca ANALYSES AT 37°C o CALCULATED FROM PO4 ANALYSES / • / 8 40 80 120 EXPOSURE TIME (HOURS) 160 Fig. 25. Enamel dissolution as a function of time during exposure to a me- dium consisting of 0.05 M lactic acid plus 6 per cent hydroxyethyl cellulose (with low buffer content) adjusted to pH 3.5. The amounts of enamel dissolved as determined by either calcium or phosphate analysis agreed well. Tempera- ture had little effect at 90 hours exposure. The enamel dissolution rate during incipient carious lesion formation decreased with increasing exposure time. a single exposure period, thus simplifying the experimental measure- ments. Ideally, the initial rate at time zero is used, but an atypical condition exists during the initial time period (approximately 1 hour) while the various diffusion gradients are established. How- ever, the dissolution or decalcification rate was close to being linear during the first 4 to 6 days, so that the relative effect of the different variables could be determined using a long time interval. A period of 0 to 96 hours was chosen in order to give the necessary latitude ex- perimentally and yet approach the conditions of a true rate measure- ment as a function of initial conditions. This requires that neither the initial reaction or solution conditions have been altered enough nor the reaction products accumulated sufficiently to alter signifi- cantly the reaction rate. PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 237 CALCULATED FROM PO4 ANALYSES • 4 "C ^ 37 "C 80 120 160 EXPOSURE TIME (HOURS) Fig. 26. Enamel dissolution as a function of time during exposure at 4°C and 37°C to a medium consisting of 0.05 m lactic acid and 0.03 m CaCl2 plus 6 p6r cent hydroxyethjl cellulose (with low buffer content) adjusted to pH 3.5. Temperature had little effect on the enamel dissolution rate. The rate was decreased by addition of CaClo (compare with Fig. 25). The next series of observations deal with enamel dissolution rate (expressed as mg of enamel dissolved per cm- of enamel surface exposed per 96 hours exposure) as a function of hydrogen ion con- centration (i.e., pH), lactate concentration, and calcium concentra- tion (Figs. 28 to 33). These quantitative results are higher than the preceding measurements because a different sample of hydroxyethyl cellulose was used that contained a greater quantity of the buffer salt impurity. The results become comparable when the decalcifica- tion rate is plotted as a function of total acid. The rate of enamel dissolution during incipient carious lesion formation increased with increasing hydrogen ion concentration. The rate is a nonlinear function of the hydrogen ion concentration, and the slope of the function decreases as the hydrogen ion concen- ^238 J. A. CxRAY AND M. D. FRANCIS 16 lI2 E o ^' E Q CO LlI <4 ijj • CALCULATED FROM PO4 ANALYSES ^v /' / ^ /' / ^ '' / m^^,^"^""^^ / / ^ '' / ^'■■'''^^ * /'' / •^.--^^''^ / / ^ /' / ^^ * / / ••^'^ / ^--^ ir 40 80 '20 160 EXPOSURE TIME (HOURS) 200 240 Fig. 27. Comparison of enamel dissolution rate as a function of time during exposure to mediums containing 0.05 m lactic acid plus 6 per cent hvdroxvethvl cellulose (with low buffer content) at different CaCl. concentrations and pH values. Increasing the CaCL concentration or the pH value decreased the rate of enamel dissolution during incipient carious lesion formation. The line of short dashes is from Fig. 25 and the line of long dashes is from Fig. 26. tiat!on is increased. The enamel dissolution rate was measured as a function of pH for a medium at 0.05 m lactate concentration and two different calcium chloride concentrations. The data are plotted as a function of hydrogen ion concentration (Fig. 28). Increasing calcium chloride concentration caused a decrease in the rate and hence in the slope of the rate curve. The rate of enamel dissolution increased linearly with increasing lactate concentration at an\ pH level or calcium chloride concentra- tion. The effect of lactate concentration on rate was determined for two different pH values and three different calcium chloride con- centrations (Fig. 29). The slope of these linear plots varied with the different hydrogen ion and calcium chloride concentrations. PHYSICAL, CHEMISTRY OF ENAMEL DISSOLUTION 40 . ^.rl. ^0D£0_ 239 0 pH5 pH4 [H+] X 10^ (moles/liter) Fig. 28. Enamel dissolution rate as a function of hvdrogen ion concentra- tion during exposure at 37°C to a medium containing 0.05 m lactic acid plus 6 per cent hydroxyethvl cellulose with and without 0.03 m CaCL. In either case, the enamel solubility rate increased with increasing hydrogen ion concen- tration, but with a continuously decreasing slope. This latter effect is related to the amount of undissociated acid formed from the buffers. CaCL caused a de- crease in the rate of enamel dissolution. Also, the rate did not fall to zero at zero lactate concentration, re- flecting the buffer capacity of the hydroxyethvl cellulose sample. The rate of enamel dissolution decreased with increasing calcium concentration, but not linearly (Fig. 30). However, replotting the data in logarithmic form resulted in a straight-line relationship ( Fig. 31). The effect of calcium chloride on the dissolution rate was meas- ured at two different pH levels and lactate concentrations. A logarith- mic plot is a typical procedure for determining the order of reaction with respect to the variable under study. In this case, because of mathematical requirements, the logarithm of the difference in rate with and without calcium present is plotted versus the logarithm of the calcium chloride concentration. The slope of this plot gives the J. A. GRAY AND M. D. FRANCIS 0 0.1 0.2 0.3 0.4 0.5 TOTAL LACTATE CONCENTRATION (moles/liter) Fig. 29. Enamel dissolution rate as a function of lactate concentration dur- ing exposure at 37° C to a medium containing lactic acid and CaCL plus 6 per cent hydroxyethyl cellulose adjusted to either pH 3.5 or 4.5. The rate was linear and increased with increasing lactate concentration. Decreasing pH or CaCU concentration increased the slope of these linear functions. The intercept at zero lactate concentration resulted from the effect of the buffer content of the hy- droxyethyl cellulose sample. order, which appears to be equal to about %. The fractional aspect of the order is quite certain, but further data are required to establish the quantitative exactness. It should also be pointed out that under conditions giving very high enamel dissolution rates, involvement of the surface became more and more extensive, leading to the condition, described by Hals et al. ( 1955 ) , of an "inner" and an "outer" spot. However, there appeared to be no detectable discontinuity in the rate when this effect was encountered and included in the measurement of enamel dissolution ( Fig. 29 ) . The very highest rates resulted in severe sur- PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 241 10 Average of 15 values (Range 4.6-6.5 -etLl^L^:^!^ LACTATE 0.01 0.02 0.03 0.04 ADDED CaCl2 CONCENTRATION (moles/liter) Fig. 30. Enamel dissolution rate as a function of CaClo concentration dur- ing exposure at 37°C to a medium containing lactic acid and CaCL plus 6 per cent hydroxyethyl cellulose. Addition of CaCL decreased the rate at two dif- ferent pH levels and lactate concentrations. face dissolution (e.g., at pH 3.5, 0.61 m lactate, 0.03 m CaClo). The maximum rate of enamel dissolution before surface damage became detectable by visual or microscopic observation was about 20 mg of enamel per cm- of surface per 96 hours exposure. Although not all the levels of composition variables investigated gave ideal in- cipient carious lesions, some measurements outside of the conditions for producing a typical lesion were necessary in order to assess the effects within the preferred region. The importance of undissociated acid in addition to the hydrogen ion for controlling the rate of enamel dissolution during incipient carious lesion formation has been suggested ( Gray et ah, 1962 ) . The recent results mentioned earlier (Kapur et al., 1962) support this hypothesis. The data from the acid-organic polymer systems studied in this presentation were plotted versus total acid concentration (Figs. 32 and 33) to test this concept further. The undissociated lactic acid was calculated from the dissociation constant (Gray, 24'2 .1. A. CRAY AND M. D. FRAXriS I0.0-- < — (D CD \ 3 CM ^ i en \ rl E 1.0- 0.001 1 1 \ — I — I — I I I I 1 1 i 1 — I — TTT SLOPE = I I I I I I I J I I I I I I I 0.01 0.1 ADDED CaClg CONCENTRATION (moles / li ter ) Fig. 31. A logarithmic plot of the data from Fig. 30 for the enamel dis- solution rate during incipient carious lesion formation as a function of CaCL concentration was a straight line. The rate was, therefore, proportional to the CaCL) concentration taken to a power equal to the slope of this plot. 1962) and the known lactate concentration vvhicli had been added. The undissociated acid content arising from hydroxyethyl celkdose was determined from pH titration curves. The hydrogen ion con- centration was calculated from the pH measurement. The contribu- tion of the various acid sources is given in Table II. It is immediately apparent that the rate of enamel dissolution over all pH values and lactate concentrations can be summarized by a single straight-line function of total available acid concentration for a given calcium chloride concentration. A separate linear function with a different slope is required for each calcium level. In the absence of organic material, lactate buffer of any apprecia- ble concentration destroyed the surface and only a slight advancing front of decalcification occurred (extending only 5 to 10 microns into the enamel for 0.05 m lactic acid at its "as is" pH of 2.6 ) . The loss of enamel and, to some extent, the advancing decalcification front are apparent in a photomicrograph (Fig. 34) or in a micro- PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION ^243 - 40- o 0 / u> / i. / f /a to 0> / M B / A e / 5 30- / \ / o» / E / *— ' / UJ / H- < °= 20- / A > A PH H 3.0 0 _j /■ 3.5 u 3 4.0 V o 10- /A 4.2 ■ cn 4.5 A _i 5.0 • UJ A 6.0 A S / • < A LlI 0- I 0 0.05 0.10 0.15 TOTAL ACID CONCENTRATION (moles/liter) [H+] + [UNDISSOCIATED BUFFER] Fig. 32. The enamel dissolution rate during incipient carious lesion forma- tion was a straight-line function of total acid concentration for a medium con- taining lactic acid plus 6 per cent hydroxvethvl cellulose over a pH range of 3 to 6 at 37°C. A very good lesion was obtained even at pH 6. Therefore, the rate can be expressed by a simple equation in terms of acid concentration (see equation 2, page 252). radiograph (Fig. 35); the advancing front in this same sample can be seen in an electron micrograph ( Fig. 36 ) . The appearance of this defect in the electron micrograph was quite different from that of an incipient carious lesion (Fig. 22). As the concentration of total lactate was decreased at a constant pH value of 4.5 until only h> - drogen ions were present (i.e., as HCl), the ratio of subsurface to surface dissolution increased ( microradiograph, Fig. 37; photomicro- graph, Fig. 3i), but surface dissolution was always present. The addition of calcium and phosphate ions to the acid solutions, still in the absence of organic pohiner, further increased the ratio of sub- surface to surface dissolution and permitted a higher acidic buffer concentration to be used (microradiograph. Fig. 38). The lesions 244 J. A. GRAY AND M. D. FRANCIS 40 NO CaCl2 ADDED / / / / / 0^^ 30 / / / / / / ■ X 0.03 M 20 / / / • / / • / '' ■ / ■ 1^ CaCl2 ADDED PH 3.0 • 3.5 ■ 10 / / 4.0 ' 4.5 ' ) 0.05 0.10 0.15 020 TOTAL ACID CONCENTRATION (moles/liter) [H+] + [UNDISSOCIATED BUFFER] Fig. 33. The enamel dissolution rate was a straight-line function of total acid concentration for a medium at 37°C containing lactic acid plus 6 per cent hvdroxyethyl cellulose with or without 0.03 m CaCL added. The CaCL in- hibited the rate proportionally across all acid concentrations and pH values. The dashed line is taken from Fig. 32. appeared to be excellent examples of incipient carious lesions, but again surface attack was nev^er completely arrested. However, the character of the subsurface dissolution was completely different from that in the lesions produced in vivo or in vitro with the organic polymer-acid systems (Fig. 3j). The lesion was continuous to the surface with no relatively sound outer layer, and had no birefrin- gence, conditions indicative of very advanced decalcification as de- fined by Gustaf son ( 1957 ) . Nevertheless, further studies of such systems should provide additional valuable information concerning the dynamics and equilibria of incipient carious lesion formation once correlation with the organic polymer-acid svstem has been established. The decalcification rate of enamel during incipient carious lesion PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 245 formation also depended on the size and shape of the enamel snrface area that was exposed to the decalcification medium. Experimentally, this means that for direct quantitative comparisons, all samples should be of the same size; 15 mm- samples were used throughout the present study. The rate per unit area increased linearly with TABLE II. Acid Concentration of Decalcification Medium IN Moles per Liter Undissociated acid in Total acid" Total Undissociated hydroxj-ethyl (dissociated + lactate pH [H+] lactic acid" cellulose'' undissociated) 0.05 3.0 10 X 10-* 0.042 0.050 0.093 0.00 3.5 3.2 X 10-* — 0.045 0.045 0.01 ti " 0.006 '■'■ 0.051 0.05 u <( 0.031 ti 0.076 0.12 n 11 0.074 (( 0.119 0.25 tl (t 0.154 tl 0.199 0.61 u It 0.376 It 0.421 0.05 4.0 1.0 X lO-* 0.017 0.042 0.059 0.05 4.2 0.64 X 10-" 0.012 0.039 0.051 0.00 4.5 0.32 X 10-" — 0.031 0.031 0.01 If 0.001 It 0.032 0.05 (( 0.007 It 0.038 0.06 11 0.008 ti 0.039 0.12 li 0.017 " 0.048 0.25 ii 0.034 tt 0.065 0.61 it 0.084 ti 0.115 0.05 5.0 0.1 X 10-" 0.002 0.017 0.019 0.05 6.0 0.01 X 10-" 0.000 0.005 0.005 0.61 " ** 0.003 a 0.008 " Calculated and checked by titration. '' Determined by titration. The other sample of low buffer content contained un- dissociated acid of 0.026 m at pH 3.5 and 0.022 m at pH 4.5. decreasing sample area, approximately doubling for a 50 per cent reduction in surface area (Fig. 39). The effect could be the residt of the greater volume available for diffusion of reactant and reaction products to and from the enamel at the sample edges as compared with the middle of the sample, and, therefore, would be a function of the perimeter of the sample. '240 J. A. GRAY AND M. D. FRANCIS Figs. 34 and 35. A section of enamel after exposure for 7 hours to an rui- agitated aqueous solution of 0.05 m lactic acid at its natural pH of 2.6. The right side was protected w ith varnish to serve as a control. Fig. 34. A light micrograph shows the complete loss of enamel to a depth of 50 microns and the damage extending slightly into the remaining sample. (X2000.) Fig. 35. A microradiograph also shows the complete loss of the enamel surface and confirms the presence of a decalcification front. (X 2000.) Fig. 36. An electron micrograph of a section cut from the sample of Figs. 34 and 35. This enamel sample was exposed 7 hours to an aqueous solution of 0.04 M lactic acid at the "as is" pH of 2.6. Although most of the enamel was disso]\'ed awav as shown in Figs. 34 and 35, a slight advancing front of de- calcification extended into the enamel about 5 microns. (X 5000.) PHYSICAL CHEMISTRY (W ENAMEL DISSOLUTKIX "247 Fig. 37. A microradiograph of a section of enamel after a 15-day exposure at room temperature to an acjueous solution of hydrochloric acid at pH 4.5 shows that decalcification had occurred near the surface but had not penetrated very far. Examination of the surface with a metallurgical microscope showed that light pitting had occurred. ( X 200. ) Fig. 38. A microradiograph of a section of enamel after a 15-day exposure at room temperature to an aqueous solution of 0.0005 m CaClo and 0.0005 m NaH^POj, adjusted to pH 4.5, shows that extensive decalcification had occurred. The subsurface dissolution penetrated quite far but was fairly uniform through- out the damaged region. The surface was much pitted by the exposure, but this could be seen only with a metallurgical microscope. ( X 200. ) 12- E ^ 8-1 < OQ 4. o en '^ 2 (mm ) 20 24 z 0 4 8 12 ENAMEL SURFACE AREA Fig. 39. The enamel dissolution rate was measured in a medium at 37°C consisting of 0.05 m lactic acid and 0.002 m CaCU plus 6 per cent hydroxyethyl cellulose adjusted to pH 4.5, using enamel samples with different surface areas. The smaller the surface area, the faster was the absolute rate of dissolution. This effect is a result of the geometry of the experimental set-up and is probably related to diffusion conditions. 248 -T. A. GRAY AND M. D. FRANCIS Important Factors in Incipient Carious liEsiON Formation Preservation of Enamel Surface In a previous section, it was indicated that the enamel surface is protected principally by the presence of polymeric organic material. The exact nature of this protective mechanism is not understood, but a protective coating resulting from adsorption of the organic compound on the apatite crystals appears most logical. The higher organic content of incipient carious lesions as compared with normal enamel demonstrates that organic substances diffuse into these en- amel defects. Therefore, protective coatings of organic origin can be expected within the enamel near the surface, and thus could account for the relativelv sound outer layer overlying the main de- calcified region. Indeed, electron micrographs do show that this outer layer has the highest residue after demineralization. Obviously, some dissolution occurring in this relatively sound outer layer would promote the diffusion of acid and organic material within the enamel. The reduced density of the outer layer of a lesion produced in vitro can be seen in microradiographs. In previous studies (Soni and Brudevold, 1959), it has been shown quite clearly that this outer layer in natural lesions has a lower density than normal enamel. The hardness is also slightly reduced within this layer (Gustafson, 1957), consistently with the reduced density. In electron micrographs, the channels between the rods going through this layer are clearly apparent. The organic material apparently diffuses into the enamel as the acid creates voids and enlarges the interrod spaces. In order for the organic material to be protective (Table I), it must be water soluble or in a colloidal suspension so as to gain access to the hydroxyapatite surface, and it must be adsorbed. Polymeric saccharides and proteins are known to adsorb on many substrates, and these materials are highly effective in protecting the enamel surface during subsurface decalcification. The monomers have little or no such effect. The organic material, constantly present and dif- PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 249 fusing into the enamel through the enlarged pathways, will begin, at some point, to have a larger and larger effect in reducing the attack within the enamel lesion as more and more exposed apatite crystals become covered with the polymeric substances. For this reason, it is also very doubtful whether the organic component can ever be completely omitted in an in vitro system for reproducing the typical natural lesion. The protection of the surface layer is increased by the presence of acid-insoluble inorganic compounds, particularly those resulting from reaction with enamel, such as calcium fluoride or stannous phosphate. Diffusion within the Enamel In a system which restricts interaction with the enamel surface, hydrogen ions will tend to diffuse into any region lower in acid concentration. That such regions must exist and are accessible to a solution phase has been demonstrated by the diffusion of radioactive isotopes through enamel under in vivo conditions (Sognnaes et ah, 1955). It is obvious from experimental results that although the sur- face apatite crystals are protected by the organic polymers, the dif- fusion of acid into enamel is not blocked by this coating. Hence, any acid penetrating beyond the limits of the organic protective coating will react to dissolve subsurface enamel. At the point of reaction be- tween acid and hydroxyapatite, the principles involved in the sur- face dissolution of enamel in acid, as determined in a previous study (Gray, 1962), can be invoked. As the acid reacts at specific sites in the enamel, calcium and phosphate accumulate and retard or arrest the dissolution reaction. Then, acid will tend to diffuse farther into enamel for reaction at advanced unprotected sites, but it is limited by the available diffusion pathways and the depleted acid concen- tration. Diffusion of soluble calcium and phosphate out of enamel is favored by the wide diffusion pathways already established and by the favorable concentration gradient to the exterior solution. At the same time, any increase in pH, either ahead of or behind the reaction site, will cause reprecipitation of calcium phosphate, the particular phase depending on the existing conditions. As calcium 250 J. A. GRAY AND M. D. FRANCIS and phosphate concentration is reduced by precipitation and/or diffusion, the reaction with hydrogen ion is renewed and the lesion below the surface slowly progresses. Concent rat io)i Gradients in the Solution Phase In the course of incipient carious lesion formation, different solu- tion compositions must exist inside and outside the enamel with a gradient extending from one to the other. For the in vitro system the external or bulk solution is defined bv the initially prepared composition. At or near the enamel surface, however, there will be a region of different composition as solution components diffuse into the enamel and products diffuse out. The hydrogen ion and undissociated acid concentrations will certainly be lower and the calcium and phosphate concentrations ( from the reaction products ) will be higher at the surface than in the external bulk solution. In- side the enamel, the hvdrogen ion concentration will be lower than in the bulk solution, owing to reaction of hvdrogen ions with enamel. Defining the concentrations of reaction products relative to the ex- terior bulk solution is not alwavs so straightforward. When such materials are not incorporated initially in the decalcification medium, the concentration within the enamel will obviously be higher. When, however, calcium and phosphate are added initially as part of the decalcification medium, sometimes in large amounts, the concentra- tions in the exterior phase may approach or even exceed those in the solution phase inside the enamel. As the concentration of cal- cium, for instance, is raised in the external decalcification medium, the rate of enamel dissolution decreases owing to the decrease in rate of removal of the reaction products; it will be recalled that diffusion rate is controlled principally bv the difference in concen- tration from one site to the other. At sufficiently high calcium con- centrations, the enamel dissolution reaction stops altogether. The amount of calcium required in the decalcification medium to inhibit incipient carious lesion formation is greater than would be expected on the basis of enamel solubility rate measurements (Gray, 1962). The existence within the enamel of other species of calcium and phosphate in the form of soluble complexes, for example acid phosphate and calcium lactate, would permit a greater rate of out- PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION '251 ward diffusion, in spite of the high concentrations of calcium and phosphate in the exterior solution. The dissociation, or association, of these complexes under the different conditions existing in the external solution would allow this process to continue. It should also be pointed out that the conditions of the solution phase will determine the solid phase, which will either reprecipitate and/or exist at anv equilibrium that is attained. Chemical Kinetics and Equilibrium Kinetics of Dissolution In the presence of an acid maintained at a constant concentration, the rate of enamel dissolution, as well as the various concentration gradients just described, will shift onh' gradually ])ut continuously as the mineral substrate breaks down. It is this pseudo steady state that can be measured quantitatively and for which data have been presented. If the acid supply increases and decreases, as would be expected under oral conditions, there will occur a cycling of events that is difficult to describe mathematicalK . Consequently, only the condition of a single, continuous exposure will now be considered and treated mathematically. It has been shown that the rate of incipient carious lesion forma- tion is determined largely by the rate of reaction of acid with the inorganic portion of enamel. The small effect of temperature on rate of enamel dissolution during incipient carious lesion formation supports the concept of diffusion control. As shown above, the rate depends on total available acid concentration, but hydrogen ion contributes little directly to the rate because of its low concentra- tion. The undissociated acids, alwa\ s relatively high in concentration as compared with hydrogen ions, provide the hydrogen ions for reaction through diffusion to and dissociation at the apatite crystal surface. The linear relationship of the enamel dissolution rate as a function of total acid concentration shows that the rate of dissolution is "first order" with respect to acid concentration (Figs. 32 and 38). The dissociated buffer anion appears to have no direct controlling effect on the rate of dissolution during incipient carious lesion forma- tion. The linear relation of the plot of Fig. 29 supports this conclu- 252 J. A. GRAY AND M. D. FRANCIS sion, as deviations from linearity would he expected if either inhibi- tion or enhancement were involved. The anion could have been expected to change the rate because of its effect on dissociation of lactic acid or because of its weak calcium complex ( Bjerrum, 1957 ) . Mathematicallv, the rate of incipient carious lesion formation (i.e., the enamel dissolution during incipient carious lesion formation) can be expressed as follows, using the usual first order rate expression ( Laidler, 1950 ) , assuming that the buffer anion is inert and all other conditions (e.g., calcium and phosphate concentrations) are con- stant: ^ = A-o[//+] + UHB] (1) where dEn/dt = rate of enamel dissolution during incipient carious lesion formation (mg enamel/cm-/ 96 hrs.) [i/+] = hydrogen ion concentration (moles/liter) \HB^ = undissociated acid concentration (moles/liter) A'o, fci = constants For combinations of several different l)uffers, the expression would take the following general summation form : ^ = k,[m] + Z lUHB]^ (2) Then, for any buffer system and in terms of total buffer concentra- tion (for ease of calculation), the following expressions (3 and 4) for the total buffer concentration [TB] and the dissociation constant K of the buffer are substituted in equation 2: [TB] = [HB] + [B-] (3) ^ [m]\B-] [HB] ^^ where [■B~] = dissociated acid buffer anion concentration (moles/ liter) The result of the substitution is the following equation: dt ~^"^^ J + ^^'"K„, + [//+] ^^^ PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 253 If the strength of the buffer decreases ( i.e., the dissociation constant decreases), the rate should increase, reach a maximum, and then decrease. For most practical cases, the first term as well as the [H+] in the denominator of the second term can be ignored since they are small, and then : f^ = A,Ifl[ra] (6) For these expressions, all other variables are assumed to be con- stant, such as the cation and organic polymer concentration of the decalcification medium. To incoi"porate these other variables into the function requires additional terms which must be added to or multiplied by the existing terms. The preceding equation, however, is the fundamental expression of the rate of incipient carious lesion formation. Calcium suppresses the rate, and this inhibition can be expressed by including a term, k[Ca]'\ in the immediately preceding equation. The value of the exponent n is determined from the slope of the logarithmic plot (Fig. 31, n = ¥4) of the rate versus calcium concen- tration. It has not yet been determined how best to incorporate this additional term to augment the fundamental expression for rate of formation of incipient carious lesion. Phase Identification by Equilibrium Solubility Studies The mechanism of rate inhibition by calcium is complicated but is almost certainly not a reversal of the reaction to re-form hydroxy- apatite (i.e., common ion effect) (Gray, 1962). One reason for this conclusion is that phase diagram data (Van Wazer, 1958) indicate that dicalcium phosphate, not hydroxyapatite, will be the existing phase under the conditions present during enamel dissolution. An- other supporting fact is that many, if not all, cations forming in- soluble phosphate salts have a similar repressing effect (although to widely different degrees ) , leading to the conclusion that inhibition is a generalized effect of cations. To a much more limited extent, similar effects can be expected and are found for anions, particularly those which form insoluble calcium salts. ^254: J. A. GRAY AND M. D. FRANCIS At present, the onl\' mechanism bv which all the different effects observed experimentallv can be satislactorih explained (Gray et al., 1962 ) is the precipitation of some of the reaction products as phases that are different from the original hydrox\ apatite. These phases form on the h\ droxxapatite surface and limit further reaction or, at equilibrium, completely prevent access of hydrogen ions to the en- amel or egress of calcium and phosphate ions to the bulk solution phase. The existence of these phases has been proxed by equilibrium solubility measurements, as will now be described. Detection of the surface phases as described in the previous para- graph by currently available physical methods is very difficult be- cause, in the first place, the la\'ers formed are onh' a few angstrom units thick (approximateh^ 25 A), on the basis of calculations from chemical and phvsical data. Secondlv, isolation of the solid samples is at present impracticable because any change of the equilibrium solution in contact with the surface will tend to change the phase. If, however, the original postulation is correct, then the solution in equilibrium with the surface coating should have a composition satisfving the solubilitv product for the phase existing on the hy- droxyapatite surface. These measurements of solubilitv product were made in simple acidic solutions (i.e., not decalcification me- diums). Brieflv, the experimental work was as follows (Francis, in preparation). Samples of dental enamel (human and bovine, both intact and powdered) and s\'nthetic hvdroxvapatite (Victor Chem- ical Works) as well as dicalcium phosphate were agitated in solu- tions of lactic or acetic acid buffers in a pH range from 4.6 to 6.5. The slurries were filtered or centrifuged to remove the solids, and the solution pH, calcium concentration, and phosphate concentra- tion were measured. Determinations as a function of exposure time were made to confirm that equilibrium conditions had been attained. Corrections for complexing of calcium bv lactate and acetate anion species were required, but ionic strength corrections could be ig- nored. Several ratios of solution to solid, and addition of calcium (as CaCL.) and phosphate (as NaHL>P04) initiallv to the acid solu- tion, were also studied. Regardless of conditions, the Ksi- consistentlv observed was that of dicalcium phosphate (Van Wazer, 1958), as summarized in Table III. This proved that tlie solution was in equi- Q PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION ''^OD librium with a phase different from hydroxyapatite, which must, therefore, be covered l)y this other phase. Previous work ( Gray et aJ., 1962 ) has shown that the initial sohi- bihty rates of normal enamel and fluoridated enamel were the same. TABLE III. Equilibrium Solubility Constants Determined from Synthetic and Natural Phosphates in Acid Buffer Solutions Buffer Solid solution Solid concentration ratio pll r inge Ksp" [Ca++][HP04 ']) Dicalcium phosphate 0.20 M acetate 0.1 4.7- f).7 .5.4 X 10-1 Dicalcium phosphate 0.18 M acetate 0.1 ■).! 4.7 X 10-' 5.3 X 10-' Dicalcium phosphate 1.0 M acetate 0.1 4.8- 4.9 5.5 X 10-' J Synthetic apatite 0.20 M acetate 0.1 .-).0- 5.3 2.7 X 10-" ' Synthetic apatite 0.20 M lactate 0.1 4.(; 8.1 X 10-' Synthetic apatite 1.0 M acetate 0.1 4.8- 4.9 4.5 X 10"' Synthetic apatite 0.20 M acetate 0.01 4.9- 5.2 3.8 X 10-' [ 4.5 X 10' Synthetic apatite 1.0 M acetate 0.01 4.9 4.4 X 10-' Synthetic apatite + CaCi? 0.18 M acetate 0.001 4.8. 5.1 1.5 X 10-' Synthetic apatite + NaH2P04 0.18 M acetate 0.001 .-..0. 5.1 (i.5 X 10' Bovine incisor enamel 0.20 M acetate 0.1 r,.0- (1.2 5.0 X 10' Bovine incisor enamel + CaCh + NaHoPOi 0.20 M acetate 0.1 4.9 4.5 X 10' Bovine incisor enamel 0.20 M acetate 0.01 4.9 7.8 X 10-' Bovine incisor enamel 0.20 M lactate 0.01 .'■).9 4.2 X 10-' 5.5 X 10-' Bovine incisor enamel + CaCh 0.20 M acetate 0.01 .-^.o, 5.2 (i.l X 10-' Bovine incisor enamel + CaCh + NaH2P04 0.20 M acetate 0.01 4.9, 5.0 5.1 X 10' Bovine incisor enamel + NaH2P04 0.18 M acetate 0.001 5.0 ll.l X 10-' Intact human incisor enamel 0.20 M acetate - 4.4 8.3 X 10' Human molar enamel and dentin 1.0 M acetate 0.1 5.1 12.0 X 10' . 7.1 X 10-' Human molar enamel and dentin 1.0 M acetate 0.01 4.8 8.0 X 10-' Human molar enamel and dentin 1.0 M acetate 0.004 4.9 3.7 X 10-' Human molar enamel and dentin 0.18 M acetate 0.002 5.1 3.0 X 10' Rat bone 1.0 M acetate 0.01 5.0 12.0 X 10' . 8.5 X 10' Rat bone 0.18 M acetate 0.002 5.2 4.8 X 10-" Hvunan enamel Synthetic apatite Synthetic apatite Saliva + lactic acid (Fosdick and Starke, 1939) 4.4 X 10' HCl + NaCl (Ericsson, 1949) 2.8 X 10"' Saliva + lactic acid (Ericsson, 1949) 4.5 X 10-' 4.1 X 10- " The value of the Ksp determiiu'd in the present work across all systems (85) was 5.4 (± 2.7) X lO"''. An average value of 4.1 (± 2.0) X 10"' was ol)taine(l from data which included 61 systems from the literature. Fluoridated enamel became less soluble only after fluoride accumu- lated in the acid solution, suggesting that inhibition resulted from the deposition of calcium fluoride from the reaction products onto the enamel surface. Furthermore, there is no fundamental reason 256 -T- A. GRAY AND M. D. FRANCIS why fluorapatite should have a lower initial rate of solubility in acid than hydroxyapatite when the reaction of hydrogen ion with the phosphate group must be responsible for the breakdown of the apatite lattice. Measurements of the solution phase in equilibrium with fluorapatite mineral in acidic buffers gave the solubility product for calcium fluoride ( Table IV ) . The measured solubility product of TABLE IV. Equilibrium Solubility Constants of Calcium Fluoride AND Fluorapatite Determined in Lactate Buffer Buffer Solid/solution KsP»([Ca+^][F-]^) Solid concentration ratio pH CaF2 0.20 M 0.1 3.4, 4.3 2.7 X 10-i« CaF2 0.47 M 0.01 3.5 3.9 X 10-1" Synthetic fluorapatite 0.20 M 0.1 4.0 4.7 X 10-i« Synthetic fluorapatite 0.20 M 0.01 4.6 2.5 X 10-1° Asparagus stone'' 0.20 M 0.002 3.5 2.0 X 10-1° Intact asparagus stone'' 0.20 M — 3.5 3.4 X 10-1" "The value of the Ksp determined acros.s uU systems (10) was 2.8 (±1.2) X IQ-i". '' A naturally occurring mineral form of fluorapatite [Caio(P04)6F2]. calcium fluoride in lactate buffer is given for comparison because it was found that the value was lower, corresponding to literature values, when determined in water or unbuffered acidic solutions. Although pH of the solubilizing solution and composition of the enamel may have only an indirect effect in determining how fast enamel dissolves (or an incipient carious lesion forms), they will have a major role in determining the phase at any equilibrium and, therefore, the rate at which the equilibrium is reached. Furthermore, it is necessary to determine the fate of the phosphate phase laid down during a low pH condition when the external pH conditions return to values once more establishing apatite as the stable phase. Unfortunately, solubility product measurements cannot answer this question, but new physical techniques, such as back-scattering elec- tron diffraction and infrared absorption spectroscopy by attenuated total reflection, give promise of confirming the presence of these thin coatings which have been discovered by chemical means. Among the problems which remain to be solved in the description PHYSICAL CHEMISTRY OF P:NAMEL DISSOLUTION 'io t of the formation of incipient carious lesions are: the positive iden- tification of the organic fractions in the lesion; the events leading to the relatively sound outer layer of enamel, especially with a view to separating the effects of by-passing and reprecipitation; diffusion measurements through enamel, particularlv relating to transport of acid and reaction products; and the role of acidic forms of phosphate as well as the various weak complexes of calcium with phosphate, bicarbonate, and lactate in the transport of the reaction products. Summary Incipient carious lesions of enamel, tvpical of those formed in vivo, have been produced bv purely chemical svstems. These svstems were used to investigate the variables affecting incipient carious lesion formation and elucidate the mechanisms involved. The essential factors necessary for the formation of incipient cari- ous lesions are (1) a reactant (acid) to dissolve the hydroxyapatite and (2) an organic polymer to protect the external surface of the enamel. The inclusion of the reaction products — calcium and phos- phate— has mediating effects ranging from complete inhibition of dissolution to subtle alterations in subsurface decalcification, per- mitting closer approach to duplication of the natural lesions. Incipient carious lesion formation can be explained for the general case by the dynamics of the heterogeneous system consisting of an acidic solution and a calcium phosphate solid. This involves chemical phase transformation, equilibria, and kinetics. The rate of enamel dissolution during incipient carious lesion formation is directly proportional to total acid concentration (dissociated plus undis- sociated), a "first order" diffusion-controlled reaction. This reaction can be inhibited or brought to equilibrium by the presence of ions which form protective deposits on apatite surfaces as salts of the reaction products, such as dicalcium phosphate or calcium fluoride. The existence of dicalcium phosphate on the surface of enamel, bone, and synthetic hydroxyapatite, and the existence of calcium fluoride on the surface of fluorapatite, in equilibrium with acid solu- tions, has been established. 258 J- A. GRAY AND M. D. FRANCIS References Bhussiy, B. R. 1958. Chemical and physical studies of enamel from hu- man teeth. II. Specific gravity, nitrogen content, and hardness rating of discolored enamel. III. Specific gravity, nitrogen content, and histologic characteristics of opaque white enamel. /. Dental Research, 37, 104.5-1053, 1054-1059. Bjerrum, J. 1957. Stability Constants of Metal-Ion Complexes, with Solit- bilitij Products of Inorganic Substances, Part 1, p. 12. The Chemical Society, London. Briner, W. W. 1963. Role of metabolic products of Lactobacillus casei in the decalcification of enamel. /. Dental Research, 42, Sept.-Oct. Brudevold, F. 1948. A study of the phosphate solubility of the human enamel surface. /. Dental Research, 27, 320-329. Burnett, G. W., and Scherp, H. \^^ 1952. The accessibility of the organic dentinal matrix. /. Dental Research, 31, 776-790. Coolidge, T. B., Besic, F. C, and Jacobs, M. H. 1955. A microscopic comparison of clinically and artificially produced changes in enamel. Oral Sura., Oral Med. and Oral Pathol, 8, 1204-1210. Darling, A. I. 1956. Studies of the early lesion of enamel caries with transmitted light, polarised light, and radiography. Rrit. Dental }., 101, 289-297. Ericsson, Y. 1949. Enamel apatite solubility. Investigations into the cal- cium phosphate equilibrium between enamel and saliva and its re- lation to dental caries. Acta Odontol. Scand., 8, Suppl. 3, 44, 49, 50. Fosdick, L. S., and Starke, A. C, Jr. 1939. Solubility of tooth enamel in saliva at various pH levels. /. Dental Research, 18, 417-430. Francis, M. D., and Meckel, A. H. 1963. The in vitro formation and quantitative evaluation of carious lesions. Arch. Oral Biol., 8, 1-12. Gore, J. T. 1940. Saliva and enamel decalcification. IV. Factors involved in the prevention of caries. /. Dental Research, 19, 455-471. Gray, J. A. 1962. Kinetics of the dissolution of human dental enamel in acid. /. Dental Research, 41, 633-645. Gray, J. A., Francis, M. D., and Griebstein, W. J. 1962. Chemistry of enamel dissolution. In Chemistry and Prevention of Dental Caries (R. F. Sognnaes, editor), pp. 164-179. Charles C. Thomas, Spring- field, 111. Gray, J. A., and Opdyke, D. L. J. 1962. A device for thin sectioning of hard tissues. /. Dental Research, 41, 172-181. Gustafson, G. 1957. The histopathology of caries of human dental enamel. With special reference to the division of the carious lesion into zones. Acta Odontol. Scand., 15, 13-55. Hals, E., Morch, T., and Sand, H. F. 1955. EflFect of lactate buffers on PHYSICAL CHEMISTRY OF ENAMEL DISSOLUTION 259 dental enamel in vitro as observed in polarizing microscopy. Acta Odontol Scaiul, 73, 85-122. Jenkins, G. X. 1961. A criti(|iie of the proteolysis-chelation theory of caries. Brit. Dental J., Ill, 311-330. Johansen, E. 1962. The nature of the carious lesion. In Dental CAinics of North America. I. Dental Caries (E. Johansen, editor), pp. 305-320. W. B. Saunders Co., Philadelphia, Pa. Kapur, K. K., Fischer, E. E., and Manly, R. S. 1962. Influence of type of acid, pH, molarity, and saturation on rate of enamel penetration. /. Dental Research, 41, 760-770. Laidler, K. J. 1950. Chemical Kinetics, pp. 1-24. McGraw-Hill Book Co., New York. Lucena-Conde, F., and Prat, L. 1957. A new reagent for the colorimetric and spectrophotometric determination of phosphorus, arsenic, and germanium. A)ial. Chim. Acta, 16, 473-479. Magitot, E. 1878. Treatise on Dental Caries. Experimental and Thera- peutic Investigations (translated by T. H. Chandler), pp. 157-166. Houghton, Osgood and Co., Boston, Mass. Meckel, A. H. 1961. Formation and properties of organic films on teeth. /. Dental Research, 40, 754. Miller, W. D. 1905. A study of certain questions relating to tlie pathology of teeth. Dental Cosmos, 47, 18-39. Miihlemann, H. R. 1960. Experimental modifications of the enamel sur- face. Helv. Odontol. Acta, 7, 5-24. Newbrun, E., Brudeyold, F., and Mermagen, H. 1959. A microroentgeno- graphic investigation of demineralized enamel, comparing natural and artificial lesions. Oral Surg., Oral Med. and Oral Pathol., 12, 576-584. Nishimura, T. 1926. Histologische Untersuchungen iiber die Anfiinge der Zahnkaries, speciell der Karies des Schmelzes. Schiceiz. Monatsschr. Zahnhcilk., 36, 491-545. Opdyke, D. L. J. 1962. The histochemistry of dental decay. Arch. Oral Biol, 7, 207-219. Orban, B. J. 1957. Oral Histologi/ and Emhryologi/, 4th edition, pp. 55-58. C. V. Mosby Co., St. Louis, Mo. Pickerill, H. P. 'l912. The Prevention of Dental Caries and Oral Sepsis, pp. 18, 104. Bailliere, Tindall and Cox, London. Pigman, W., Elliott, H. C, Jr., and Laffre, R. O. 1952. An artificial mouth for caries research. /. Dental Research, 31, 627-633. Raynik, C, and Loe, H. 1961. The effect of various demineralizing agents on dog tooth enamel. Acta Odontol. Scand., 19, 495-506. Schatz, A., Karlson, K. E., and Martin, J. J. 1956. Trace element stimula- tion of keratin (hair) degradation bv oral keratinolvtic microflora. Experientia, 12, 308-309. 260 J- A. GRAY AND M. D. FRANCIS Schmidt, W. J., and Keil, A. 1958. Die gesunden tind die erkrankten Zahngewebe des Menschen und der Wirbeltiere im Polarisations- mikroskop, pp. 297-314. Carl Hanser Verlag, Munich. Sognnaes, R. F., Shaw, J. H., and Bogoroch, R. 1955. Radiotracer studies on bone, cementum, dentin and enamel of rhesus monkeys. Am. J. Physiol, 180, 408-420. Soni, N. N., and Bibby, B. G. 1961. Enamel decalcification in food-saliva mixtures. /. Dental Research, 40, 185-189. Soni, N. N., and Brudevold, F. 1959. Microradiographic and polarized light studies of initial carious lesions. /. Dental Research, 38, 1187- 1194. Soni, N. N., and Brudevold, F. 1960. Microradiographic and polarized- light studies of artificiallv produced lesions. /. Dental Research, 39, 233-240. Sullivan, H. R. 1954. The formation of early carious lesions in dental enamel. Part II. /. Dental Research, 33, 231-244. Van Wazer, J. R. 1958. Phosphorus and Its Compounds. Vol. I. Chemistry, pp. 510-522. Interscience Publishers, New York, von Bartheld, F. 1958. Decalcification in initial dental caries. Tijdschr. Tandheelk., 65, 76-89. Wachtel, L. W., and Brown, L. R. 1963. In vitro caries. Factors influenc- ing the shape of the developing lesion. Arch. Oral Biol., 8, 99-107. Welcher, F. J. 1958. The Analytical Uses of Ethylenediaminetetraacetic Acid, p. 114. D. van Nostrand Co., Princeton, N. J. Factors Influencing the Initiation, Transmission, and Inhibition of Dental Caries PAUL H. KEYES, Laboratory of Histology and Pathology, National In- stitute of Dental Research, Bethesda, Maryland HAROLD V. JORDAN, Laboratory of Microbiology, National Institute of Dental Research, Bethesda, Maryland THE discussion which follows represents observations, speculations, questions, and a few pertinent criticisms. In an attempt to present a broader perspective on dental caries we have tried to scan facets of the problem and not to focus narrowly on any special one. Our ulti- mate objective is to gain greater insight, to open doors, to reexamine postulations, to challenge concepts, and not to forge any idea into a fact. We feel that the emphasis belongs on a better understanding of the biology of caries and not on arguments over hypotheses. Dental caries is a complicated infection which, for the most part, can be considered to originate outside the body proper. Lesions be- gin as a demineralization of enamel, a tissue of ectodermal origin, and later penetrate into the dentin and pulp, both of mesodermal origin. For many years most investigators of the disease have recog- nized that the substrate upon which the pathogenic bacteria sub- sist is composed largely of ingested food substances which come into contact with the teeth when conditions are favorable. Much atten- tion has been directed to the role of acidogenesis, lactobacilli, and 261 >-2()'-2 p. II. KEYES AND II. V. JORDAN the mechanisms of carbohvdrate degradation and associated phos- phorylation. It seems reasonable to expect that tlie exploration of some of the non-acidogenic aspects of the disease might lead to in- formation of \'alue in strengthening a program of comprehensive therapeutic control. Students of dental caries recognize that in addition to salivary components, the highly variable nature of such di\'erse substrates as food residues and mineralized tissue makes the disease more complex to study than many other infections in which the intra- and extra- cellular components of the tissues furnish the principal substi'ate for the invading organisms. Some of the experimental findings to be discussed lend support to the idea that colonization and invasion of the cariogenic flora may be related to properties of the enamel and dentin which the pathogens find beneficial to their life cvcle. Etiological Factors There is no single cause of caries, it is a multifactorial disease as are other infections. Etiolog\' can be considered in terms of con- tributing factors in the host, the oral microbiota, and the dietarv substrate. It is possible to gain a broader perspective on the disease processes in caries bv considering somewhat analogous interactions between host, microbe, and over-all environment in other experi- mental infections, e.g. tuberculosis (Ratcliffe et ah, 1953, 1957). Additional insight into the complexities of biological interactions and into their analysis can be gained from the discourse "Cause and Effect in Biology" ( MaM", 1961 ) . In Fig. 1, three overlapping circles have been used to depict the possible relationships in the factorial triad contributing to caries activity, i.e. the host, the microbiota, and the diet. When the disease is active, these three sets of factors are in concentricitv. Investigators are becoming increasingly aware that each one of these components behaves in a dynamic and variable wa\' and that under laborator\' conditions only the most delicate balance of the three will induce the disease and favor its progression. However, the magnitude of each contributing factor is still difficult to identify positively and is rarely amenable to control bv conventional experimental methods. FACTORS INFLUENCING DKNTAL CARIES 263 Fig. 1. Three overlapping circles depict the possible relationships of the factorial groups responsible for caries activity. There are numerous variables in each parameter. When the infection is active, conditions in each are conducive; but when it is inactixe, to know which of the groups is in a noncontributory status is exceedingl\- difficult and often impossible. Despite the prevalence of caries in human populations, we might also expect to find that rather specific conditions are necessary for activity. It follows, therefore, that in order to establish the caries- inactive host as immune or resistant, one must be sure about which factors are in a noncontributory status; such definition has rarely, if ever, been attained with certainty. Many experiments in the field of dental caries have not been con- trolled in terms of the single-factor variable described in their de- sign, and as a result such studies are open to various interpretations in terms of mechanisms responsible for the results. As suggested above, the limiting factor is obviouslv in the biological complexities of the disease itself rather than in the experimental design as it has been conceived in the mind of the investigator. "In view of the high number of multiple pathways possible for most biological processes, and in view of the randomness of manv of the biologi- cal processes, particularly on the molecular level, causality in biological systems is at best onlv statistically predictive" (Mayr, 1961). ^(u p. h. keyes and h. v. jordan Host and Teeth Although factors in the host and teeth are known to contribute importantly to caries activity, these are undoubtedly the most dif- ficult to determine in a way that can be considered definitive. There are many factors in the oral cavity and in the teeth themselves which are thought to be either conducive or nonconducive to caries activity. Some of these are obviously related to the inherent nature of the host and to his over-all constitutional make-up. Of great importance and particular interest are those nonconducive factors which grad- ually develop within the host himself despite an inherent tendency toward susceptibility. For example, the older host, both animal and human, seems to be more refractory to this disease. In part this has been attributed to maturation factors in the teeth and to the presence of fluoride in the enamel. More information is becoming available as to what these properties may be in terms of chemical composition and physical properties (Brudevold, 1962rt; Johansen, 1962; Zipkine^flL, 1962). Dietary Substrate The substrate formed by the ingesta and saliva provides a source of nutrient both to the macroorganism (host) and to the many microorganisms of the mouth and alimentary canal. If some of the bacteria ( parasites ) have a cariogenic potential, and if the food sub- strates are of the proper quality and quantity, colonization and plaque formation occur on certain coronal surfaces of the teeth. See Fig. 2. The microflora associated with caries does not induce lesions unless carbohydrates are present, but the cariogenic potential of carbohydrates in both animals and humans is related to such factors as quality, quantity, and physical consistency ( Stephan, 1948; Haldi et al, 1953; Gustafsson et al, 1954; Shaw and Griffiths, 1960; Volker, 1962), and the carbohydrates must also be present in the mouth a certain length of time (Kite et al., 1950; Larson et al., 1962). These same conditions probably appl\' to other essential nutrients and not to carbohydrates alone. FACTORS INFLUENCING DENTAL CARIES 265 Fig. 2. When young hamsters harbor cariogenic microorganisms, coloniza- tion (plaque formation) rapidly develops on the coronal sm-faces of the teeth. The photograph above was taken of a living animal to show the plaque which had covered the lingual sulcus and parts of the crown of the maxillary second molar. Bacterial activity in this accumulation had largely destroyed the disto- buccal cusp of this tooth (arrow) about 3 weeks after commencement of an appropriate diet. It is evident that there are various levels at which nutrients will be (1) adequate for the host and inadequate for the microflora, (2) adequate for the microflora and inadequate for the host, (3) un- favorable for both, or (4) satisfactory for both. See Fig. 3. The studies of Shaw (1949) and Shaw and Griffiths (1960) show that an experimental diet can be adequate for growth, reproduction, and general health of the host and at the same time favor bacterial ac- tivity associated with caries. Although some workers have suggested a difi^erent interpretation, the available evidence indicates little association between nutritional status and dental caries. This holds for animal studies and also for human populations (Russell et ah, 1961 ) . In the interest of clarity and scientific accuracy, the question is whether one should accept the nutrition-caries interrelationship on exceedingly tenuous grounds or emphasize the importance of eating habits and dietary factors as they really are. The property of food substrates which has been investigated for 2(50 P. TI. KEYES AND H. V. JORDAN INGESTA-SALIVA SUBSTRATE COMPOSITION and/or QUANTITY (Chemical Physical) HOST 1. Adequate 2. Inadequate 3. Inadequate 4. Adequate (Proportion Duration' PARASITE(S) Inadequate Adequate Inadequate Adequate Fig. 3. Correlations between nutrition and caries activity are exceedingly difficult because it is necessary to recognize that the requirements of macro- organism (host) and microorganisms (parasites) may be either alike or different. "cariogenicity" most frequently is acidogenic potential (Bibby, 1961; Morch, 1961). Although the statement that "carboh\'diate foodstuffs are the principal causative agents of dental caries" tends to oversimplify matters, one cannot discount the essential importance of carbohydrates. However, not too much consideration has been given to the role of other foods whose residues can be conducive to the maintenance and propagation of the pathogenic organisms in- volved. Is it unreasonable to assume that individual foods or various combinations which favor and support the pathogenic microbiota might have a contributory potential? For example, dairy products are not generally considered conducive to caries, but rather to be more or less protective (Shaw, 1950; Shaw et al, 1959; Bibby, 1961). In view of the frequency and numerous wa\'s these products are used in modern diets, it seems hazardous to assume that such substrates are always innocent bystanders. To some extent this idea can be inferred from the findings of iMcClure and Folk (1955) in regard to increases in caries activity associated with heat treatment of skim milk powders. No one would question the nutritional excellence of milk products, but the designation of these products as "protective" in regard to FACTORS INFLUENCING DENTAL CARIES "267 caries is open to question because of the recognized capacity of such substances to favor the maintenance and growth of bacteria. In many experimental diets, milk products furnish the main source of protein for microorganisms as well as host, and two recent reports of human studies merit thoughtful consideration. Magnusson ( 1962 ) has reported that the saliva of newborn infants remains neutral to weakly alkaline during the first 12 hours of life and becomes increas- ingly acid during the remainder of the first and second day; on the third day the pH levels off to a range between 5.0 and 5.5. "The gradual increase of aciditv of the saliva during the first days of life seems to proceed at the same rate as the bacterial colonization of the oral cavity." In the light of these findings the question raised by Fass (1962), "Is bottle feeding of milk a factor in dental caries?" deserves attention not only for the cases he discusses, but also from the viewpoint of the dietary and eating patterns of older individuals. Because the implications of these observations are of considerable magnitude, it might be desirable to recognize this possibility and to consider whether anv measures should be recommended to re- duce the potential of a deleterious effect after the consumption of certain other foods not usuallv considered to be cariogenic and high in carboh\'drate. The foregoing remarks are intended to point out that evaluations of the cariogenic potential of foodstuffs are exceedingly difficult because of the conditions illustrated in Fig. 3, and there would seem to be a number of reservations and limitations to information ob- tained from both in vivo and in vitro systems which attempt to de- fine the role of monosaccharides, disaccharides, adhesiveness, and the like. Microflora Generally speaking, caries has been regarded more as a nonspecific disease than as a specific one ( MacDonald, 1962 ) . Numerous acido- genic organisms have been implicated, with most attention directed toward lactobacilli and streptococci. In gnotobiotic studies with rats, certain strains of enterococci and streptococci have shown an unquestionable cariogenic potential ( Orland et al., 1955; Fitzgerald 268 p. 11. KEYES AND II. V. JORDAN et ah, 1960). In his studies to date, Fitzgerald ( 1962) has not found caries in rats monoinfected with rat strains of Lactobacillus fermenti, Lactobacillus acidophilus. Streptococcus lactis, and Streptococcus fecalis var. zijmogenes. It was previously demonstrated that after the cariogenic flora of animals had been depressed with an antibiotic, their caries experi- ence was markedly less than in nondepressed controls or in reinfected littermates (Keyes, 1960^; Keyes and Fitzgerald, 1963). See Fig. 4. INFECTED FEMALES ANTIBIOTIC (PENICILLIN) DEPRESSES FLORA NON-INFECTED SAME CAGE BOTH CARIOUS CARIOGENIC STREPTOCOCCI LABELLED" STRAINS STREPTOCOCCI PLAQUE c "LABELLED" STREPTOCOCCI SAME CAGE •LABELLED" CARIES TRANSMISSION OF CARIOGENIC FLORA TO OFFSPRING ?9 CARIOUS CARIES c/? "LABELLED" (/9 CARIES o- ? "LABELLED" CARIES o- ? CARIOUS o" 9 'LABELLED" CARIES CARIOUS Fig. 4. In the schematic diagram above, on the left are represented several generations of hamsters which were caries-inactive as a result of depression of the cariogenic microbiota. The infection has been reintroduced by contact with infected animals, by an inoculation of both original sti-ains of streptococci and "labeled" streptomycin-resistant strains, and also by the transfer of cariogenic plaque. Following reintroduction of the cariogenic microbiota, caries is again transmitted by natural passage as indicated by solid lines on the right. FACTORS INFLUENCING DENTAL CARIES 269 Other findings in coincidence with these include those of Zipkin et al (1960) and of Larson and Zipkin (1961). By the use of bacterially depressed or similar noninfected animals it is possible to show that caries is related to an infectious agent which can be passed from cagemate to cagemate and from mother to young. The disease can also be induced by inoculating either feces or carious dentin and plaque into the mouth of a noninfected young animal. The point has been emphasized that because the ali- mentary canal microbiota with its cariogenic components is not stable but varies with different experimental diets and conditions, it is difficult to design and interpret many caries studies in animals. For example, many attempts have been made to correlate caries susceptibility with prenatal diets and those consumed during the early stages of tooth formation. The mechanisms whereby protein- carbohydrate ratios, ash and salt mixtures, dietary deficiencies, etc., influence susceptibility of the host cannot be detennined if the diet induces unrecognized changes in the alimentary canal microflora of mother animals, who then transmit their altered flora to the young. It has also been pointed out that in hamsters the dietary history of the mother and her own caries experience may be independent of the susceptibihty to caries of her offspring (Keyes, 1960Z?). A susceptible and infected female will not develop caries unless main- tained on a cariogenic diet; but since she harbors a pathogenic flora, her offspring will develop caries if a suitable diet is provided while the animals are young. It has also been reported (Keyes and Fitz- gerald, 1962 ) that a female can harbor cariogenic bacteria and con- sume a caries-conducive ration, and still be caries-inactive for various reasons, e.g. maturation factors, etc. Yet such an animal can transmit the flora to her offspring, who will develop the disease. Since the entire alimentary canal microbiota is involved and not the mouth alone, one would not expect to find offspring from females whose teeth had been extracted to be appreciably different from animals whose dams were allowed to develop the disease in retained teeth. This possibility, as postulated by Shaw et al. (1962), would seem unlikely. Indeed, animals in which caries activity has been sup- pressed by fluoride continue to harbor pathogenic organisms and to produce infected offspring (Keyes et al., 1962), and even though the 270 !'• H. KEYES AND II. V. JORDAN combination of penicillin and a concomitant change in diet will abrupth' arrest caries in the teeth, it does not always depress the cariogenic microbiota so completely that offspring are caries-inactive when tested ( Keyes and Fitzgerald, 1963 ) . The implications of experimental caries transmissibilitv have not been readilv understood. Probably the most important implication of caries transmission is that conventional laboratory animals, nota- bly hamsters, can be obtained which harbor a negligible number of cariogenic microorganisms. Such animals can be used in the design of a "normar" experimental model for studying biological interac- tions associated with caries. See Fig. 4. The procedure involves less complicated technics than the gnotobiotic ones, and the methods are within reach of almost anv laboratory Investigators have a method for studving caries in conventional animals which is not basically different from those used in the studies of other experimental infec- tions, e.g. tuberculosis (Ratcliffe and Palladino, 1953). Manv attempts have been made to induce experimental caries by an inoculation of organisms thought to have a cariogenic potential (Etchells and Devereux, 1932-1933; Rosebury et al, 1934; Belding and Belding, 1943; and others). However, unequivocal results were not obtained before the use of the previouslv discussed gnotobiotic technics (Orland et al, 1955; Fitzgerald et al, 1960). It has been interesting to find that dental caries can be induced by an inoculation of pure cultures of certain organisms isolated from the mouth of caries-active hamsters (Fitzgerald and Keyes, 1960, 1963). Rampant coronal caries has onlv followed the inoculation of specific strains of hamster streptococci in association with the proper diet. See Fig. 5. It has not followed the inoculation of various other strains of acidogenic streptococci isolated from the mouths of either caries-free hamsters or caries-active rats, nor the inoculation of lacto- bacilli and numerous other acidogenic and non-acidogenic organ- isms; i.e., coronal caries has not been induced bv inoculations of 15 strains of fecal and oral streptococci, lactobacilli, and diphtheroids, and 5 strains of acido2;enic filaments, sarcinas, fusiforms, and Gram- O JIT negatix'e rods, all strains indigenous to the hamster (Fitzgerald and Keyes, 1960, 1963). Although cariogenic streptococci seem to have a proclivity for forming plaque, it should be emphasized that not all FACTORS INFLUENCING DENTAL CARIES ^271 Fig. 5. A, no cavitation has occurred in teeth in control animal because it did not harbor cariogenic bacteria during the earlv stages of its life. B, extensive caries followed the inoculation of cariogenic streptococci isolated from a carious lesion in another hamster. •272 p. H. KEYES AND H. V. JORDAN plaques found on hamster teeth have been associated with typical caries. Apparently plaques do not form unless specific types of organ- isms are present. In hamsters, cariogenic plaque formation does not appear to be simply the result of bacterial growth in food residues coincidently retained on the teeth, although in certain instances this may be the case. The plaque which so rapidly forms in either naturally infected or inoculated animals appears to be the product of an interaction between bacteria and the diet. There seem to be specific conditions which determine whether colonization occurs or not. This is to say that, experimentally, the formation of the plaque itself is not the result of one set of factors but of the simultaneous interaction of two sets, or three if one adds those contributed by the enamel. Question of Bacterial Specificity It seems important to clarify the etiologic role of the acidogenic microbiota in caries initiation in vivo. On the basis of what is gen- erally recognized about the specific nature of the biochemical, nutri- tional, and environmental factors which determine implantation and propagation of microorganisms, it does not seem unreasonable to postulate a degree of specificity to caries initiation. This does not imply that ever)^ lesion would have exactly the same microbiological component; indeed, some observations in rats and hamsters suggest that this may not be so, but probably the point will require gnoto- biotic technics to determine conclusively. Nevertheless, the position of streptococci in an environment with high concentrations of carbo- hydrate would seem to be more delicate than that of lactobacilli. The fact that streptococci are not so aciduria suggests that these organisms would thrive better in a locus in which their ovm acid by-products could be readily neutralized, as well as those from other neighboring organisms. A question not too frequently asked is, "What does the carious lesion, i.e. the invasion of the enamel and dentin, mean to the organ- isms involved?" Certainly from numerous histopathological studies caries appears to be a highly invasive infection. Figure 6 shows sev- eral examples of bacterial invasion in hamster molars. Is this phe- FACTORS INFLUENCING DENTAL CARIES 273 A ^^^■■r ^^^ MM 2 1 -Jw ■ ' ^K'^^ """"■^ Jm\ •sv ?^^;^""•k^ Fig. 6. Decalcified sections of carious lesions in the hamster. A, placjue (1) formation in the gingival sulcus associated with massive invasion of enamel (2); and characteristic spread along the junction of enamel and dentin (3). B, Gram- positive spherical organisms invading enamel. C, cross sections of dentinal tubules filled with Gram-positive cocci. D, tubular invasion often can be fol- lowed for considerable distances in the dentin. Only coccoid forms of organisms have been found in the depths of lesions, approaching the pulp, and far in advance of generalized necrosis. Photographs B, C, and D were taken under oil immersion. 274 P- II- KKYES AND II. Y. JORDAN iiomenon merely accidental or coincidental? Orland (1946) noted that streptococci were found in great nuni])ers in hamsters with highly active caries. Although he recovered many lactobacilli, among other bacterial t\pes, he mentioned that the streptococci might be less aciduric than lactobacilli, and went on to speculate that the source of these bacteria might be in the less acid environment of deeper cavities. Harrison (1948) stated: "It is presumed that bac- teria mav cause dissolution of one or more components of the en- amel structure, either to secure their own metabolic requirements or as an incidental or accidental consequence of the release of harmful metabolic products on the enamel surface," In the hamster, plaques develop on coronal surfaces of the teeth and in the gingival sulcus when certain dietarv substrates are avail- able and specific types of bacteria are harbored in the alimentary canal. As previouslv stated, neither one of these factors by itself appears adequate to induce the phenomenon. It cannot be too strongly emphasized that none of those regions of the body, to which bacteria can gain entrance by the normal portals, provides a virgin soil in which any newcomer may flourish. It is just as impossible to ensure the proliferation of a particular bacterial species by introducing it into the mouth, as it is to ensure a crop of a particular plant by scattering seeds in a field already occupied by a pre-existing plant-association. The newcomer will have little chance of survival, un- less it is adapted to occupy some definite place in its new environment. It is fairly certain that those pathogenic bacteria which spread readily from host to host owe their capacities in this direction to their ability to colonize on skin or mucous surfaces, or to escape from the superficial environment to the underlying tissues, (\^^ilson and Miles, 1955.) The findings of Fitzgerald and Keyes ( 1960 ) have indicated that many types of acidogenic bacteria normally found in the mouths of hamsters do not cause plaque formation even when a high-carbo- hydrate low-fat diet passes through the mouth and alimentary canal; i.e., the teeth generally remain remarkablv free of these bacterial accumulations. When colonization of cariogenic microorganisms does occur and when the tooth is in a susceptible state (newly erupted and low in fluoride content), demineralization and invasion rapidly occur (Keyes, 1959). The findings of Fitzgerald and Keyes (1960) FACTORS INFLUENCING DENTAL CARIES XiO have been misconstrued by Scliatz and Martin ( 1962 ) : "If decay were due to acid, all the aforementioned organisms should have been cariogenic since all are acidogenic." The reasoning here is open to question because in the experimental hamster model not all acido- genic organisms will colonize and produce plaques on enamel sur- faces. The results of Fitzgerald and Keyes (1960) should not be used to support the proteolvsis-chelation theorv. The discussion of streptococci b\' Burnett and Scherp ( 1962 ) is quoted below because many obserxations on caries activity seem to be in coincidence with what is known about this group of bac- teria: The environmental and nutritional requirements of streptococci diflFer widely among various strains and species, ranging from the most exacting conditions to the abilitv to grow in a very adverse environment. The hu- man pathogenic strains are generally more exacting in their environmental and nutritional requirements than are the nonpathogenic strains. The following essentials have been recognized: Carbohydrates — a readily available source of carbohydrate is necessary for the growth of aerobic streptococci, although it may not be essential for the growth of some species of anaerobic streptococci. The amount of carbohydrate available in media may be critical; if too much is available, sufficient acid is pro- duced to inhibit or kill streptococci; if not enough is present, growth does not occur. Carbon dioxide — a proper amount of carbon dioxide (usually 5 per cent) favors growth and utilization of nutritive material and may act, for instance, as a substitute for purines. Vitamins — some water-soluble vitamins such as biotin, pantothenic and nicotinic acid, thiamine, riboflavin, pyridoxal, and folic acid are essential for growth, varying with the species. Purines, pyrimidines, and as yet undefined growth-promoting substances, e.g., yeast or liver extracts, may enhance or even be essential for the growth of the more fastidious streptococci, bi- organic salts and buffers — although a balanced inorganic ion system is needed for the growth of streptococci, their specific requirements have not been adequately defined. Good buffers are necessary for the neutrali- zation of acids produced during metabolism. Questions about the requirements for COi> and for inorganic salts and buffers have been of interest to us. Is there a special need when certain organisms have to deal with high loads of carbohydrates? It seems reasonable to postulate that those organisms which have invaded the tooth are surviving and probably finding a fa\orable 276 P. H. KEYES AND H. V. JORDAN environment. If inorganic ions form part of this milieu, which ones and why? There are many interesting questions related to the specific re- quirements of microorganisms and how they invade. Along these lines, the mechanisms whereby fluoride exerts its protective effect might again be examined in terms of what the inability to invade might mean to the organisms involved. The two effects most often advanced to explain the action of fluoride are a decreased solubility of the apatite mineral and an antienzymatic effect which inhibits glycolysis. Recent observations in hamsters have shown that caries can be effectively inhibited by adding fluoride to the drinking water at the time of third molar eruption, i.e., during the very early erup- tive and posteruptive period. Although lesions which started before the addition of the fluoride ion continued to progress slowly and known cariogenic streptococci were recovered from the mouth in high numbers (Keyes et al., 1962), the third molars were exceed- ingly well protected. Also, this tooth acquired a long-lasting re- sistance, as indicated by its ability to remain intact long after with- drawal of the drug. Apparently the colonization of pathogenic organisms was not pre- vented sufiiciently to explain the lowered level of caries activity. It would be of interest to know whether the organisms were behaving as usual. Is it possible that the reduced solubility of the apatite (Gray et al., 1962; Zipkin et al., 1962) and its altered chemical com- position (Brudevold, 1962«, 1962^?; Nikiforuk et al, 1962) prevent the pathogenic bacteria from obtaining essential requirements at a critical time? For example, the amount of salivary CO2 available at the interface between cell membrane and crystal and deep in the lesion is unknown, but it may not be enough for organisms faced with the problem of rapid metabolic activity, especially since salivary CO2 probably does not diffuse readily through the acid plaque and lesion. The reduced solubility of enamel mineral and replacement of hydroxyl ions could conceivably reduce the availa- bility of buffers necessary for neutralization of the acids produced during metabolism. One is able to think of many possibilities in regard to what the effect of a change in the availability of inorganic ions can mean to FACTORS INFLUENCING DENTAL CARIES 277 bacteria which are confronted with the metabohsm of copious amounts of carbohydrates, often in the form of hypertonic sugar sokitions. Perhaps the organisms have httle choice other than rapid metabohsm, accelerated reproduction, or polysaccharide storage (Gibbons and Socransky, 1962). (The latter phenomenon may also represent a protective mechanism whereby less aciduric organisms can delay the conversion of large quantities of carbohydrate and thus control the release of unfavorable amounts of free acid. Poly- saccharide storage apparently immobilizes part of the carbohydrate for future metabolization at a gradual rate. ) We have found these questions intriguing and have found it in- teresting to speculate on possible mechanisms which might affect bacterial activity from what is known about the inorganic ion re- quirements of streptococci, changes in the carbonate content of enamel containing fluoride (Brudevold, 1962fl; Nikiforuk et at., 1962; Middleton, 1962), and the loss of carbonate and magnesium from carious enamel and dentin ( Johansen, 1962 ) . Along these lines the report of Brown et al. (1962) merits com- ment. These workers postulated that essential nutrients diffusing from within the tooth might support growth of invading micro- organisms and thus play a role in caries. In an in vitro system of special design it was possible to demonstrate that organisms pene- trated dentinal tubules when the crowns of teeth were covered with a niacin-deficient medium, and the growth of Lactobacillus arabi- nosiis depended on the diffusion of niacin from the pulp side of the teeth. The principles of this system and the implications of the find- ings are of considerable interest, as are the original demonstrations. Invasion seems to be related to dependence of organisms upon es- sential factors. Clinical and Therapeutic Implications The matters discussed have a bearing on the clinical disease. We do not know whether human caries is largely a specific type of microbial infection, particularly in the initial stages. The etiologic role lactobacilli and other acidogenic organisms play in human caries needs further clarification. On the basis of such findings im- 278 p. H. KEYES AND II. V. JORDAN proved diagnostic tests might be developed. If specific organisms occiipv a ke\' position in caries initiation, chemotherapeiitic agents would have a greater potential in anticaries therapv. Contagiousness and transmissibilit\' of the cariogenic agent (streptococci or other- wise) are of more concern to laboratory workers, who need to balance animals microbiologically in order to control their experi- ments and who need to determine the bacterial conditions in the mouth and intestinal tract in order to interpret their results. The next few vears should see the development of more effective programs for caries control which would employ a combination of therapeutic measures based on the considerable amount of informa- tion available. It is doubtful, human nature being what it is, that any single measure will be entirelv effective in preventing and ar- resting this disease. Changes in diets and the avoidance of unfavora- ble eating habits are to be stronglv advocated (Bibb\% 1962; Nizel, 1962; and others). The consumption of fluoridated water is of un- questionable benefit, and the supplemental use of fluorides appears to be indicated on theoretical grounds. Further work on other modes of fluoride application is certainlv needed. Antibacterial drugs have occasionally been used in anticaries therapy for almost four decades (Bunting et ah, 1927; Stralfors, 1962); but few, if any, have been thoroughly tested in a comprehensive program of treatment. In the treatment of this infection, with its highly invasive characteristics, it will be necessary to emplov the proper tvpe of drug and to devise methods of application which not only will ensure its presence in susceptible areas but also will keep it in sufficient concentration and for adequate periods of time. Summary Dental caries activity is determined by interactions in a factorial triad comprising the host, the diet, and the microflora. As yet only specific strains of streptococci have induced typical coronal caries in noninfected hamsters and in more rigidly controlled germ-free ( gnotobiotic ) rats. In hamsters kept under conventional laboratory conditions, lactobacilli and various other acidogenic organisms, in- cluding other strains of streptococci, have not induced highlv active FACTORS INFLUENCING DENTAL CARIES '279 disease. The demonstration of caries infectivitv in hamsters shows investigators how to produce quasi gnotobiotes which can be used for more refined experimental models to study some of the bio- mechanisms involved. These observations raise the question whether caries is a more specific t\'pe of bacterial infection than is usually supposed. It certainly appears that all acidogenic organisms are not equally cariogenic. Some lesions might follow the activity of organ- isms which are adapted to use the tooth as a source of beneficial substrate, especially when faced with high levels of carbohydrates in their immediate environment. Further leads on diagnosis, causa- tion, and control could conceivably follow research directed toward the disclosure of other bacterial characteristics besides acid produc- tion, e.g. the nature of the organic components of the cells, their cell walls, and extracellular products. Fully effective caries control can be achieved under laboratory conditions. Probably equally effec- tive control can be attained in humans by comprehensive therapeutic measures analogous to those used in the treatment of tuberculosis, i.e. improvement in eating habits, increasing resistance of tissues (enamel), chemotherapeutic depression of pathogenic organisms, and necessary operative (surgical) procedures. Acknowledgment. Part of the text and several of the figures used in this discussion formerly appeared in "Recent Advances in Dental Caries Research. Bacteriology" (Keyes, 1962). These have been reprinted with the kind permission of the International Dental Journal. References Belding, P. H., and Belding, L. J. 1943. Is caries acuta transmissible? /. Am.. Dental Assoc, 30, 713-717. Bibby, B. G. 1961. Cariogenicity of foods. /. Am. Med. Assoc, 177, 316- 321. Bibby, B. G. 1962. Caries prevention without fluorides. In Dental Clinics of North America, pp. 411-424. W. B. Saunders Co., Philadelphia, Pa. Brown, L. R., Wachtel, L. W., and Wheatcroft, M. G. 1962. Diffusion of niacin through extracted human teeth and its effect on bacterial penetration into dentin. /. Dental Research, 41, 684-894. Brudevold, F. 1962fl. Chemical composition of the teeth in relation to caries. In Chemistri/ and Prevention of Dental Caries (R. F. Sogn- naes, editor), pp. 32-88. Charles C. Thomas, Springfield, 111. 280 P. H. KEYES AND H. V. JORDAN Brudevold, F. 1962/;. Fluorides in prevention of dental caries. In Dental Clinics of North America, pp. 397-409. W. B. Saunders Co., Phila- delphia, Pa. Bunting, R. W., Nickerson, G., and Hard, D. G. 1927. Further studies of Bacillus acidophilus in its relation to dental caries. /. Am. Dental Assoc, 14, 416-417. Burnett, G. W., and Scherp, H. W. 1962. Streptococci and streptoccal in- fections. In Oral Microbiologi/ and Infectious Disease, pp. 578-615. Williams and Wilkins Co., Baltimore, Md. Etchells, J. L., and Devereux, E. D. 1932-1933. Relationship of acidogenic bacteria to diet in production of dental caries in the white rat. Proc. Soc. Exptl. Biol, and Med., 30, 1042-1043. Fass, E. N. 1962. Is bottle feeding of milk a factor in dental caries? /. Dentistry for Children, 29, 245-251. Fitzgerald, R. J. 1962. Gnotobiotic contribution to oral microbiology. /• Dental Research, 42, 549-552. Fitzgerald, R. J., Jordan, H. V., and Stanley, H. R. 1960. Experimental caries and gingival pathologic changes in the gnotobiotic rat. /. Dental Research, 39, 923-935. Fitzgerald, R. J., and Keyes, P. H. 1960. Demonstration of the etiologic role of streptococci in experimental caries in the hamster. /. Am. Dental Assoc, 61,9-19. Fitzgerald, R. J., and Keyes, P. H. 1963. Ecologic factors in dental caries. The fate of antibiotic-resistant cariogenic streptococci in hamsters. Arji. J. Pathol, 42, 759-772. Gibbons, R. J., and Socransky, S. S. 1962. Intracellular polysaccharide storage by organisms in dental plaques. Its relation to dental caries and microbial ecologv of the oral cavity. Arch. Oral Biol, 7, 73-80. Gray, J. A., Francis, M. D., and Griebstein, W. J. 1962. Chemistry of enamel dissolution. In Chemistrii and Prevention of Dental Caries (R. F. Sognnaes, editor), pp. 164-179. Charles C. Thomas, Spring- field, 111. Gustafsson, B. E., Quensel, C. E., Swenander Lanke, L., Lundquist, C, Grahnen, H., Bonow, B. E., and Krasse, B. 1954. The eflFect of dif- ferent levels of carbohydrate intake on caries activity in 436 indi- viduals observed for five years. Acta Odontol. Scand., 16, 232-364. Haldi, J., Wynn, W., Shaw, J. H., and Sognnaes, R. F. 1953. The relative cariogenicity of sucrose when ingested in the solid form and in solu- tion by the albino rat. /. Niitr., 49, 295-306. Harrison, R. W. 1948. Lactobacilli versus streptococci in the etiology of dental caries. /. Am. Dental Assoc, 37, 391-403. Johansen, E. 1962. The nature of the carious lesion. In Dental Clinics of North America, pp. 305-320. W. B. Saunders Co., Philadelphia, Pa. Keyes, P. H. 1959. Dental caries in the Syrian hamster. VIII. The indue- FACTORS INFLL^ENCING DENTAL CARIES 281 tion of rampant caries activity in albino and golden animals. /. Dental Research, 38, 525-533. Keyes, P. H. 1960a. The infectious and transmissible nature of experi- mental dental caries. Findings and implications. Arc]}. Oral Biol., 1, 304-320. Keyes, P. H. 1960Z;. Questions raised by the infectious and transmissible nature of experimental caries. /. Dental Research, 39, 1086. Keyes, P. H. 1962. Recent advances in dental caries research. Bacteriology. Intern. Dental J., 12, 443-464. Keyes, P. H., and Fitzgerald, R. J. 1962. Dental caries in the Syrian hamster. IX. Arch. Oral Biol, 7, 267-278. Keyes, P. H., and Fitzgerald, R. J. 1963. Dental caries in the Syrian hamster. X. The natural history of the disease in a single familv. Intern. Dental /., 13, 86-109. Keyes, P. H., Fitzgerald, R. J., Jordan, H. V., and White, C. L. 1962. The effect of various drugs on caries and periodontal disease in albino hamsters. Arch. Oral Biol., Siippl., 7, 159-177. Kite, O. W., Shaw, J. H., and Sognnaes, R. F. 1950. The prevention of experimental tooth decay bv tube-feeding. /. Niitr., 42, 89-103. Larson, R. H., Rubin, M., and Zipkin, I. 1962. Frequency of eating as a factor in experimental dental caries. Arch. Oral Biol., 7, 463-468. Larson, R. H., and Zipkin, I. 1961. Effect of tetracycline on the transmis- sion of dental caries in rats. /. Dental Research, 40, 264-267. MacDonald, J. B. 1962. Microbiology of caries. In Chemistry and Pre- vention of Dental Caries (R. F. Sognnaes, editor), pp. 89-125. Charles C. Thomas, Springfield, 111. Magnusson, B. 1962. The pH of the saliva in newborn infants. Odontol. Tidsskr., 70, 503-514. Mayr, E. 1961. Cause and effect in biology. Science, 134, 1501-1506. McClure, F. J., and Folk, J. E. 1955. Observations on the production of smooth-surface rat caries by diets containing skim milk and whey powders. /. Niitr., 55, 589-600. Middleton, J. D. 1962. Effect of fluoride on enamel carbonate. /. Dental Research, 41, 1257. Morch, T. 1961. The acid potentialitv of carbohydrates. Acta Odontol. Scand., 19, 355-385. Nikiforuk, C, McLeod, I. M., Burgess, R. C, Grainger, R. M., and Brown, H. K. 1962. Fluoride-carbonate relationship in dental enamel. /. Dental Research, 41, 1477. Nizel, A. E. 1962. Food, nutrition, and dental caries. In Dental Clinics of North America, pp. 335-346. W. B. Saunders Co., Philadelphia, Pa. Orland, F. J. 1946. The oral bacterial flora as related to dental caries in the Syrian hamster. /. Dental Research, 25, 455-467. Orland, F. J., Blayney, J. R., Harrison, R. W., Reyniers, J. A., Trexler, OM P. n. KEYES AND II. Y. JORDAN P. C, Ervin, R. F., Gordon, H. A., and Wagner, M. 1955. Experi- mental caries in germ-free rats inoculated with enterococci. /. Am. Dental Assoc, 50, 259-272. RatcliflFe, H. L., and Merrick, J. V. 1957. Tuberculosis induced by droplet nuclei infection. Its developmental pattern in guinea pigs and rats in relation to dietary protein. Am. J. Pathol., 33, 1121-1135. Ratcliffe, H. L., and Palladino, \\ S. 1953. Tuberculosis induced by droplet nuclei infection. Initial homogeneous response of small mam- mals ( rats, mice, guinea pigs, and hamsters ) to human and to bovine bacilli, and the rate and pattern of tubercle development. /. Exptl. Med, 97, 61-68. Rosebury, T., Karshan, M., and Foley, G. 1934. Studies of lactobacilli in relation to caries in rats. I. Effect of feeding lactobacilli with caries- producing diets. /. Dental Research, 14, 231. Russell, A. L., Gonsolazio, C. F., and White, C. L. 1961. Dental caries and nutrition in Eskimo scouts of the Alaskan National Guard. /. Dental Research, 40, 594-603. Schatz, A., and Martin, J. J. 1962. The proteolysis-chelation theory of dental caries. /. Am. Dental Assoc, 65, 368-375. Shaw, J. H. 1949. Carious lesions in cotton rat molars. Ineffectiveness of certain essential nutrients in prevention of tooth decay in cotton rat molars. Proc Soc Exptl. Biol, and Med, 70, 479-483. Shaw, J. H. 1950. Effects of dietary composition on tooth decay in the albino rat. /. Nutr., 41, 13-24. Shaw, J. H., Ensfield, B. J., and Wollman, D. H. 1959. Studies on the re- lation of dairv products to dental caries in caries-susceptible rats. /. Nutr., 67, 253-273. Shaw, J. H., and Griffiths, D. 1960. Partial substitution of hexitols for sucrose and dextrin in caries-producing diets. /. Dental Research, 39, 377-384. Shaw, J. H., Griffiths, D., and Terborgh, A. 1962. Attempts to alter the genetically expected caries activitv in rats by manipulation of the oral flora. ^Arch . Oral Biol, 7, 693-706. Stephan, R. M. 1948. Relative importance of polysaccharides, disaccha- rides, and monosaccharides in the production of caries. /. Am. Dental Assoc, 37, 530-536. Stralfors, A. 1962. Disinfection of dental plaques in man. Odontol. Tidsskr., 70, 183-203. Volker, J. F. 1962. Prevention of dental caries: General principles. In Dental Clinics of North America, pp. 385-396. W. B. Saunders Co., Philadelphia, Pa. Wilson, G. S., and Miles, A. A. 1955. The normal flora of the human body. In Topley and Wilson's Principles of Bacteriology and Immunity, Vol. 2, pp. 2249-2269. Williams and Wilkins Co., Baltimore, Md. FACTORS INFLUENCING DENTAL CARIES ^283 Zipkin, I., Larson, R. H., and Rail, D. P. 1960. Reduced caries in offspring of rats receiving tetracycline during various prenatal and post-partum periods. Proc. Soc. Exptl. Biol, and Med., 104, 158-160. Zipkin, I., Posner, A. S., and Eanes, E. D. 1962. The effect of fluoride on the X-ray diffraction pattern of the apatite of human bone. Biochim. et Biophijs. Acta, 59, 255-25(S. 10 Effect of Hibernation on Tooth Structure and Dental Caries WILLIAM V. MAYER, Department of Biology, Wayne State University, Detroit, Michigan SOL BERNICK, Department of Anatomy, University of Southern Cah- fornia, Los Angeles, California AT A TIME when the pubhc press, popular science writers, and even some professional scientists have speculated on hibernation as a mechanism whereby humans could travel great distances through space in the state of torpor accompanying the phenomenon of hiber- nation, it is strange that we have so little evidence of the effect of hibernation on the morphology and physiology of mammals. The studies of Mayer ( 1960 ) and Mayer and Bernick have dealt with aspects of hibernation in relation to the structure and function of endocrine glands (1959), the digestive system (1958), and protein and carbohydrate metabolism (1957). Almost nothing, however, is known of the effect of hibernation on bones and teeth. Sarnat and Hook, in 1942, concluded that all stages of tooth de- velopment, including growth, calcification, and eruption, were se- verely retarded in proportion to the time the animal was in hiberna- tion. However, they made no detailed histological studies of the effects of hibernation on tooth and bone development. The study of Richardson et al. ( 1961 ) did not provide unequivocal evidence regarding the effect of hibernation on the dental tissues of the 13- 285 086 w. V. :mayer and s. bernick lined ground squirrel. In t!ie present inxestigation the autliors have examined histologically the effects of hibernation on the teeth and on the surrounding bones and periodontium of a hibernating animal. Hibernation is a relatively rare phenomenon. The term is correctlv used to identify that metabolic state involving cessation of externally observable activity and a marked lowering of the body temperature of the homoiotherm. Though it is commonh' thought of as a phe- nomenon of the high Arctic, it is in realit\' a temperate-zone phe- nomenon, to judge from the geographic distribution of the animals which actually hibernate. The phenomenon is limited to relatively few genera of birds and mammals and is considered by the senior author to be a highly stressful and relatively unsatisfactory method of meeting environmental extremes, at least for the individual ani- mal, although it is sufficiently effective for species survival. The selection of the Arctic ground squirrel as the hibernator for use in these experiments is based on the facts that it is a large animal readily kept in captivity and that it can be induced to hibernate relatively easily. Mayer (1953a, 1953/?) has reported on the ecolog- ical relationships of the Arctic ground squirrel and its patterns of hibernation. Although in this paper such terms as "3 months in hibernation" are used, it is to be understood that hibernation is not a continuous process, operating from September until Mav in the case of the Arctic ground squirrel, but rather a series of hibernations from which the animal periodically awakens at roughly 3-week in- tervals. Thus, a 3-month period of hibernation would undoubtedly have included at least four such awakenings. The Arctic ground squirrels for this study were obtained from Alaska and maintained in colonies in our laboratories. Materials and Methods The animals used for this study were Arctic ground squirrels (CiteUtis undulatus) , ranging in age from 3 to 4:1 years. In captivity they had been subsisting on a natural stock ration ( Purina ) supple- mented with vegetables, such as lettuce, cabbage, carrots, or celery. Animals were sacrificed and the skulls removed, decalcified, em- EFFECT OF HIBERNATION ON TEETH %H7 bedded in celloidin, sectioned, and prepared for examination his- tologically in the routine manner. Observations The dentin of a nonhibernating Arctic ground squirrel presents a wide, homogeneously calcified outer layer which extends from its outer surface through the uncalcified predentin toward the odonto- blastic laver ( Fig. 1 ) . After about 3 weeks in hibernation, however, there is evidence of a disturbed calcification process (Fig. 2). One may still see a thinner outer zone of dentin which is homogeneously calcified. There is, however, an increase in the number of inter- globular spaces in the deficient dentin extending toward the odonto- blastic layer. In nonhibernating Arctic ground squirrels the interradicular and the interseptal bone show a high degree of calcification and both are compact and normal in histological configuration (Fig. 3). In contrast, after an animal has been in hibernation for 3 months there is a progressive loss or resorption of both interradicular and inter- septal bone. Figure 4 shows a marked osteoporosis of the spongiosa arid the presence of calculus in the interproximal gingivae. Even though the alveolar bone proper and the attachment fibers are rela- tively unaffected in this animal, alveolar bone frequently undergoes degenerative changes during hibernation. Caries appears rather consistently in both the molar and premolar teeth of squirrels kept in captivity for extended periods of time. As field-caught wild animals do not show such a predisposition to caries, it is believed that decay is not related so much to hibernation as to conditions of captivity, including diet. However, as the effects of hibernation on noncarious normal teeth are drastic, in carious teeth these effects even further aggravate tooth destruction. All stages of caries, from early lesions to fracturing of the coronal part of the tooth, accompany hibernation. Figure 5 shows early carious lesions. Two plaques occur in the occlusal pits, and from these plaques dark-staining lamellae project through the decalcified enamel toward the dentin. Figure 6 shows a more advanced carious lesion which involves an invasion of the dentinal tubules. The in- 288 W. V. MAYER AND S. BERNICK Figures 1 to 6 EFFECT OF HIBERNATION ON TEETH 289 fected tubules extend from the surface lesion deep into the substance of the dentin. In Fig. 7, a further aggravated lesion, in which the dentin has been locally decalcified, has produced a typical carious cavity containing a heavily stained plaque. Pulpal involvement is of common occurrence as demonstrated in Fig. 8, which represents a buccal-lingual section of the upper first molar. The carious lesion has invaded the pulpal area, and the proc- ess of degenerative change has resulted in an open communication with the oral cavity. The disintegration of the pulp extends into the root to produce an apical abscess. The inflammatory lesion has infiltrated the interradicular region and has extended to the adjacent bone. In this figure the periapical abscess is pronounced. In an animal in hibernation for 3 months, a buccal-lingual section of an upper molar (Fig. 9) demonstrates open communication of the pulp with the oral cavity. In this case, the destruction of the Abbreviations Used in Figures A, abscess Ep, epithelium B, bone G, gingiva C,_ cementum IT, infected tubules Ca, calculus Od, odontoblastic layer CE, cemento-enamel junction P, plaque Cr, carious lesion . PA, pulpal abscess Cu, cuticle PP, periodontal pocket D, dentin Pu, pulp DD, deficient dentin R, root E, enamel RF, root fragment EA, epithelial attachment Fig. 1. Incisal dentin of a warm and active Arctic ground squirrel. The dentin is homogeneously calcified. Fig. 2. Incisal dentin of squirrel in hibernation for 3 weeks. The deficient dentin shows an increased number of interglobular spaces. Fig. 3. Upper molar of a warm and active squirrel. The compactness of the interseptal and interradicular bone is normal. Fig. 4. Upper molar of an animal in hibernation for 3 months. Osteoporosis of the spongiosa of both the interseptal and the interradicular bones is obvious. Calculus is present in the interproximal gingivae. Fig. 5. Upper molar of an animal in hibernation for 3 weeks. From the plaques in the occlusal pits heavily stained lamellae project toward the dentin. Fig. 6. Upper molar of an animal in hibernation for 3 weeks. The dentinal tubules have become infected by carious invasion. •^oo W. V. MAYER AND S. BERNICK ^Ef> n EFFECT OF HIBERNATION ON TEETH '291 pulp has extended down into the root of the tooth. This pulpal in- flammation has hollowed out the root canal, resulting in a thinning of the root while the necrotic pulp fills with debris. In addition, the epithelial attachment has proliferated deeply along the sides of the root toward the apex, with extensive destruction of the gingiva on the buccal face of the molar. In another animal in hibernation for 3 months, the pulpal involve- ment has become so extensive as to lead to a fracturing of the coronal part of the tooth (Fig. 10). The epithelium has invaded the base of the fracture, and both the surface epithelium and the inflammatory cells have proliferated deeply to the apex of the tooth fragments. One fractured fragment has been encircled by epithelium. The encircling of fragments by epithelium takes place at various times during hibernation, as shown in Fig. 11, which is a section of a premolar from an animal in hibernation for 1 month. In this case, a persistent fragment of the tooth is encapsulated by epithelium. Pus cells are seen adjacent to the open end of the tooth, and the epithelium has proliferated to the apical end of the fragment, form- ing a deep periodontal pocket on the distal surface of the tooth frag- ment. The mesial fragment from this tooth has been exfoliated, and a periodontal pocket is observed on its distal surface. Fig. 7. Upper molar of an animal in hibernation for 1 month. The carious cavity in the locally decalcified dentin is extensive. Fig. 8. A buccal-lingual section of an upper molar of an animal in hiberna- tion for 3 weeks. The carious lesion has invaded the pulp, resulting in its open communication with the oral cavity. The periapical abscess has invaded the adjacent bone. Fig. 9. A buccal-lingual section of an upper molar of an animal in hiberna- tion for 3 months. The destruction of the pulp has extended deep into the roots, resulting in a thinning of the root. Downgrowth of epithelium toward the apex and destruction of the gingiva on the buccal surface have also occurred. Fig. 10. A premolar of an animal in hibernation for 3 months. The coronal part of the tooth has fractured, and the epithelium has invaded the base of the fracture. Fig. 11. A premolar of an animal in hibernation for 1 month. A persistent fragment of the tooth is encapsulated bv epithelium. The deep periodontal pocket on the distal surface of the tooth fragment is indicative of extensive degeneration. Fig. 12. The interproximal region between two premolars of an animal in hibernation for 3 weeks. Pocket formation and the presence of calculus on all surfaces is obvious. 292 W. V. MAYER AND S. BERNICK In addition to the effects of hibernation on caries and its effects on bone and dentinogenesis, animals in hibernation for periods of from 3 weeks to 3 months exhibit various degrees of periodontal involve- ment. Osteoporosis of bone was shown in Fig. 4 in an animal which had been hibernating for 3 months. Figure 12 represents the inter- proximal region of two premolars of an animal hibernating for 3 weeks. There is a loss of trabeculation, although less than that ob- served in Fig. 4. There is, however, a pocket formation on both the mesial and distal surfaces of the two adjacent teeth, and the distal surface of the second premolar exhibits a deep pocket that extends to the lower half of the root. In addition, the presence of calculus is obvious. In still another hibernating squirrel the distal region of the first upper molar exhibits a degeneration of the surface epithe- lium and the presence of calculus on the cementum ( Fig. 13 ) . There is also a noticeable loss of interseptal bone and a further downgrowth of the epithelial attachment. After 3 months in hibernation a section of the interproximal region between two molars of a squirrel shows an epithelial attachment which has proliferated to the apex of the root ( Fig. 14 ) . The crest of the bone has been lowered to the apical region. The cemental Fig. 13. Interproximal region between two molars of a hibernating animal. Further downgrowth of the epithelial attachment and loss of interseptal bone are common during hibernation. Fig. 14. Interproximal region between two molars of an animal in hiberna- tion. The epithelial attachment has proliferated to the apex of the root. The crest of the bone has been lowered to the apical region. Cemental tears on the distal surface of the molariform tooth, a periodontal pocket in the gingiva of the same surface, and the pulpal abscess indicate the deleterious effects of hibernation on teeth. Fig. 15. Interproximal region bet\^een two molars of an animal in hiberna- tion. The presence of epithelium in the interradicular region, the lowering of the crest of the bone, and the carious lesion in the dentin further attest to the stresses of hibernation. Fig. 16. Upper first molar of an animal in hibernation for 3 months. The bifurcation involvement with loss of attachment fibers and bone, the periapical abscess, and the cemental tears indicate further deterioration in the oral cavity during hibernation. Fig. 17. Upper first molar of a hibernating animal. The complete encapsula- tion of a root fragment by epithelium is a response to the degeneration of tooth structure during hibernation. EFFECT OF HIBERNATION ON TEETH 293 294 W. V. MAYER AND S. BERNICK tears on tlie distal surface of the molariform tooth are obvious. A periodontal pocket with a proliferating epithelial attachment is pres- ent in the gingivae of the same surface. A pulpal abscess is present and its communication with the oral cavitv is noticeable. The interproximal region between two molars from another hil^er- nating squirrel shows a downgrowth of the epithelial attachment, which proliferates into the interradicular area (Fig. 15). The in- vasion of the epithelium into the bifurcation region is accompanied by a disorganization of the attachment fibers and a lowering of the alveolar bone. This degeneration is accompanied bv carious lesions of the dentin. In Fig. 16, showing a first upper molar of a squirrel in hibernation for 3 months, continued deterioration is demonstrated. Periodontal pockets on the mesial and distal surface exhibit such extensive bi- furcation invohement that no bone remains in the interradicular region. In addition, an apical abscess is present and there is an in- flammatory invasion of surrounding connective tissue. Epithelial encapsulation of tooth fragments is frequently noted in squirrels in hibernation. Figure 17 shows a root fragment fullv encapsulated bv epithelium. There can be no doubt that hibernation has a dramatic eftect on bone, dentinogenesis, and dental caries. The reasons why this should be so, however, are not elucidated in the literature. In the work of Willet et al. (1957) it was indicated that protease may accelerate caries in several ways. The work of Hunt et al. (1957) showed that submaxillary gland extracts of caries-susceptible rats had 5j times the protease activity of gland extracts from more resistant animals. The work of Mayer and Bernick, however, demonstrated an involu- tion of the salivary glands during hibernation and thus an inhibi- tion of protease activity. Thus, it is not to the salivaries that we can turn for our explanation of aggravated carious lesions during hiber- nation. Rosen et al. (1957) indicate that it is highly probable that micro- organisms are the immediate causes of dental caries. During hiberna- tion, however, there is not only a lowered oral cavit\' temperature, but also a drier oral environment, due to cessation of salivary ac- tivity. Thus, one would be led to anticipate less and not more mi- EFFECT OF HIBERNATION ON TEETH ^95 crobial activit\' dnriiicr hibernation l:)ecause the factors of moisture o and temperature during hibernation are inimical to growth of micro- organisms. Hunt et ill. (1957) cited work which conckided that endocrine secretions are a factor in dental caries, but here again Mayer and Bernick ( 1959 ) have indicated a general involution of the endocrine system during hibernation. Two points in Hunt's paper may point the way to a possible explanation of the extensive dental caries in captive Arctic ground squirrels. One involves dietarv factors, which have an evident effect on caries. The other is that a desalivated animal, in which conditions similar to those of an animal in hiberna- tion are simulated, is not able to wash food awav from its teeth as well as the animal with active salivary secretions. The resulting prolonged contact of food with teeth, regardless of lowered tempera- ture, may contribute greatly to the accelerated development of caries in hibernating animals. Summary and Conclusions The observations reported in this paper underscore the lability of bone minerals during hibernation, when both teeth and bone are called upon to contribute calcium to the metabolism of the hiber- nator. Hibernation results in deficient dentinogenesis, a reduction and osteoporosis of both interseptal and interradicular bone, and de- generation of alveolar bone, as well as a downgrowth of surface epithelium, and pocket formation on mesial and distal surfaces of adjacent teeth. There is also a disorganization of attachment fibers and the presence of calculus on the cementum, together with ce- mental tears. Carious lesions and pulpal and apical abscesses are also aggravated in the teeth of hibernating animals. The process of hibernation can thus be seen to have severe effects on the metabolism of the hibernating animal. During hibernation there is a severe drain on mineral resources within the body. This drain increases with time and gives one pause in postulating hiberna- tion as an effective mechanism for meeting severe environmental stress, for the process itself is an exceptionally stressful one. 296 W. V. MAYER AND S. BERNICK Acknowledgments. This research was supported, in part, by the United States Air Force under Contract No. AF18(600)-843, monitored by the Arctic Aeromedical Laboratory, Alaskan Air Command, Ladd Air Force Base, Alaska. Contribution No. 94 from the Biology Department of Wayne State University, Detroit, Michigan. References Hunt, H. R., Hoppert, C. A., and Rosen, S. 1957. The causes of dental caries. Centennial Rev., 1, 264-285. Mayer, W. V. 1953fl. A preliminary study of the Barrow ground squirrel, Ciiellus parriji barrowensis. J. Mammal., 34, 334-345. Mayer, W. V. 1953Z?. Some aspects of the ecology of the Arctic ground squirrel, Citellus parriji barrowensis. Stanford Univ. Publ., Univ. Ser., Biol. Sal, 11, 48-55. Mayer, W. V. 1960. Histological changes during the hibernating cycle in the Arctic ground squirrel. VII. Bull. Museum Comp. Zool. Harvard Coll., 124, 131-157. Mayer, W. V., and Bernick, S. 1957. Comparative histochemistry of selected tissues from active and hibernating Arctic ground squirrels, Spermophilus undtdatus. J. Cellular and Comp. Physiol., 50, 277-292. Mayer, W. V., and Bernick, S. 1958. Comparative histological studies of the stomach, small intestine, and colon of warm and active and hibernating Arctic ground squirrels, Spermophilus undulatus. Anat. Record, 130, 747-758. Mayer, W. V., and Bernick, S. 1959. Comparative studies of the thyroid, adrenal, hypophysis, and the islands of Langerhans in warm and ac- tive and hibernating Arctic ground squirrels, Spermophilus undulatus. Trans. Am. Microscop. Soc., 78, 89-96. Richardson, R. L., Fisher, A. K., and Folk, G. E., Jr. 1961. The dental tissues of wild and laboratory-raised hibernating and non-hibernating 13-lined ground squirrels. /. Dental Research, 40, 1029-1035. Rosen, S., Benarde, M. A., Fabian, F. W., Hunt, H. R., and Hoppert, C. A. 1957. Several properties of saliva from Hunt-Hoppert caries-resistant and caries-susceptible rats. /. Dental Research, 36, 87-91. Sarnat, B. C, and Hook, W. E. 1942. Effects of hibernation on tooth de- velopment. Anat. Record, 83, 471-493. Willett, N. P., Rosen, S., Stafseth, H. J., Hunt, H. R., and Hoppert, C. A. 1957. A comparison of salivary protease activity in the Hunt-Hoppert caries-resistant and caries-susceptible rats. /. Dental Research, 36, 223-229. Dento-Alveolar Resorption in Periodontal Disorders INGJALD REICHBORN-KJENNERUD, Faculty of Odontology, Uni- versity of Oslo, Oslo, Norway PERIODONTAL disorders are diseases clinically characterized by (a) inflammatory color changes and swelling of the gingiva, (b) deepening of the gingival pockets, and sometimes (c) increased mobility and migration of the teeth. To some degree, one or more of these symptoms are nearly always found somewhere in the mouths of man in all age groups. Alveolar bone resorption can be detected very frequently in x-rays of periodontal disorders, and tooth resorption may also occur. In periodontal disease, pathological changes in the gingiva always are seen at the microscopic level. This is the case even if the typical clinical signs of a gingivitis are missing. Microscopic sections of a human periodontium without inflammatory reactions are rare indeed. Many theories have been proposed regarding the etiology of bone and tooth resorption in periodontal disorders. These theories will be discussed in this chapter. Theories Regarding Alveolar Bone Destruction It has been suggested that alveolar bone destruction is a result of a gingival inflammation (James and Counsell, 1928; Lang, 1923; 297 Q98 I- REICIIBORN-KJENNERUD Hiiupl, 1925; Iliiupl and Lang, 1927), and that it is caused by in- creased masticaton' functional stress (Karolyi, 1905) or by a dystro- phy not necessarily related to gingivitis (Weski, 1921, 1928). Ac- cording to Gottlieb ( 1925 ) , alveolar bone resorption can be due to an atrophv, or to a gingival inflammation caused by dental calculus or tartar and retention of food ( "Schmutzpyorrhoe" ) . Some authors, among them Thoma and Goldman (1937), have attributed the breakdown of alveolar bone to local (paradontitis) and systemic (paradontosis) factors and to combinations of the two factors. Glick- man (1949, 1952) is of the opinion that both local and systemic factors mav be responsible for the alveolar bone destruction, the outcome being governed bv a so-called "bone factor," defined as the systemic regulatorv influence upon the alveolar bone. The last theory to be mentioned, as advanced by Lang ( 1923) and by Haupl and Lang ( 1927 ) , differs in many respects from the others, and necessitates an explanation in greater detail of its main points, as follows. Hiiupl and Lang, in accordance with the general concept of Pommer ( 1885, 1925), take the position that circulatory disturbances which increase the tissue tonus in the bone cause bone resorption in all parts of the skeleton. They maintain more specifically that circulatory disturbances of this type develop in the alveolar bone through an interaction of factors which can be divided into three groups: (1) external factors such as irritants initiating a gingival inflammation, and circulatory disturbances in the periodontium caused by mechanical functional stress above the physiological limit; (2) inherited and acquired anatomical and physiological periodontal conditions which directly promote the development of or aggravate disturbances in the blood circulation in the alveolar bone; and (3) inherited and acquired systemic factors which in- directly influence the periodontal blood circulation, e.g., the kinds of influences that are involved in the over-all parasite-host relation of the organism. Hiiupl and Lang assert that the loss of alveolar bone in periodontal disorders is determined by the relation between resorption and for- mation of alveolar bone. Furthermore, they are of the opinion that DENTO-ALVEOLAR RESORPTION IN PERIODONTAL DISORDERS 290 formation of alveolar bone ma^ be clue to three factors: inherited characteristics, substances produced in the inflamed periodontal tissues, and masticator\ functional impulses. The latter concept is based on the fact that weakening of the alveolar bone as a result of resorption exposes the rest of it to in- creased functional stress, or what is called hvperfunction. In all parts of the skeleton, hvperfunction to a certain limit results in compensa- tory bone apposition, provided that the bone tissue has a sufficient supplv of hormones, xitamins, calcium salts, and other nutritional elements necessary for bone production. In a growing organism compensatory bone apposition and hereditary bone growth take place at the same time. In this period heredity governs the gross pattern of bone formation, while functional impulses modifv the form and structure of the bones. The stimulating effect of mechanical functional impulses on bone production was established bv Roux in 1895, and his findings have been confirmed bv clinical observations and a series of experiments, of which onlv those performed b\' Hiiupl and Psanskv ( 1938 ) and by Watt and \Mlliams (1951) will be mentioned here. The former authors observed that alveolar bone and connective tissue in the pericementum had been produced in humans as a reaction to an increase in masticatory functional stress for three nights by means of activators (Fig. 6). Watt and Williams found that the size and densitv of the mandible in growing and adult rats increased with a rise in the masticatory functional stress. Both experiments support the assumption that alveolar bone resorption is a trigger mechanism which, subject to certain conditions, incites alveolar bone produc- tion. Instead of any further comment on the different theories, the problems pertinent to the loss of alveolar bone in periodontal dis- orders will be discussed from the standpoint of (1) the relation between circulatory disturbances in the periodontium and loss of alveolar bone, (2) the effect of increased functional stress on the alveolar bone, (3) the influence of periodontal and systemic condi- tions on the loss of alveolar bone, and (4) the effect of periodontal pathologv on resorption of the teeth themselves. 300 i. reichborn-kjennerud The Relation between Circulatory Disturbances in THE Periodontium and Loss of Alveolar Bone Gingival inflammations are the most common cause of circulatory disturbances in the periodontium. An increased amount of blood must be transported to and from the inflamed gingiva through the alveolar bone. Clinical and x-ray observations indicate that a chronic gingivitis very often is accompanied by a breakdown of alveolar bone. Elimi- nation of a gingivitis through successful treatment usually stops the resorption, and sometimes it is followed by an apposition of alveolar bone visible on x-rays. In histological sections it can also be seen that alveolar bone re- sorption is related to periodontal circulatory disturbances. If they are confined to the supra-alveolar connective tissue, a horizontal septum resorption occurs. When the vessels passing through the marginal part of the pericementum are involved, vertical septum resorption takes place. In cases of gingival inflammation, a breakdown of the central part of the septa frequently is observed on x-rays. This resorption may extend from the margin to the base of the septa, and not seldom the lamina dura of the adjacent teeth is broken down in some places (Figs. 1 and 9). According to Haupl and Lang (1927), these resorp- tions are due to a remote effect of a gingivitis. When an increased amount of blood is transported to an inflamed gingiva through the interalveolar arteries, they will be dilated and compress the veins in the alveolar bone marrow more or less (Fig. 2). The effect of this is an increase of the permeability of the vessels in the septa, and that again enlarges the amount and changes the quality of the transudate to the bone marrow. In histological sections it can be observed that the alterations in the transudation lead not only to alveolar bone resorption, but also to fibrosis of the alveolar bone marrow. The latter process, called bone phlegmasia according to Recklinghausen (1891), very often can be seen in the whole length of the septa (Fig. 1). Circulatory disturbances in the interalveolar vessels may also extend through DENTO-ALVEOLAR RESORPTION IN PERIODONTAL DISORDERS 301 \J.. Fig. 1. A, marginal, B, basal part of the same septum. On the left side of the septum a vertical resorption has occurred. Through the whole length of the septum (A and B) there are circulatory disturbances and fibrosis of the bone marrow. Only some pieces of the horizontal bone trabeculae are left. In some places circulatory disturbances and loose connective tissue extend from the bone marrow into the pericementum (arrows). These changes are remote effects of the gingivitis. (X 15.) their branches into the pericementum of the adjacent teeth, result- ing in a breakdown of pericemental fibers (Fig. 1). All these ob- servations support the assumption that alveolar bone resorption is caused by circulatory disturbances. In two types of periodontal disorder, however, there seems to be no relation between gingival inflammation and loss of alveolar bone. Tn some patients a chronic gingivitis is not accompanied by break- 30''2 I. REICIIBORN-KJENNERUD down of alveolar bone as shown by x-rays; in others, alveolar bone destruction appears when few or no clinical symptoms of gingivitis are present. The former tvpe of periodontal disorder is usually seen in growing indi\'iduals, seldom in \oung adults. In histological sections from the periodontia of growing indixiduals with gingivitis and no loss of alveolar bone, it can be observed that thev have alveolar bone resorption, but that a pronounced bone production simultaneously occurs, and often in the same bone marrow space (Figs. 3 and 4). 4.. '■ Fig. 2. An extended artery in a septnm has compressed a dilated vein. There is a fil:)rosis between the vessels. ( X 200.) DENTO-ALVEOLAR RESORPTION IX PERIODONTAL DISORDERS 3()i Fic o. The iiiaigiiial part of a periodontium from a 19-vear-o!d individual. Inflammatory changes are seen in the gingiva. The epithehal attachment is located at the enamel-cementum junction. On the top of the septum osteoid has been laid down. In the three marginal round bone marrow spaces bone resorption and apposition are taking place, as seen in Fig. 4. In the other bone marrow spaces there is apposition in spite of the circulatory disturbances and the fibrosis. ( X 30. ) Resorption on one side of a bone trabecula and apposition on the other side is also frequenth' seen. In these patients the destruction of aheohir bone is compensated or exceeded by alveolar bone formation owing to hereditary factors and to functional impulses. The same bone production is also re- sponsible for the very quick repair of ah'eolar bone in children after orthodontic treatment. In young adults with gingivitis and no loss of alveolar bone, the bone formation incited b\' functional impulses is able to replace the resorbed alveolar bone. This response is seldom seen in adults in our populations. Among east Greenland Eskimos subsisting as hunt- ers far from the white population, it is, however, common to find 304 I. REICIIBORN-KJENNERUD ^SZ^ -' ^4^^ Fig. 4. Osteoclasts (vertical arrow) and osteoblasts (horizontal arrow) in the same bone marrow space as that shown in the marginal part of the septum in Fig. 3. On the top of the septum, osteoid lined by osteoblasts can be seen. (X 150.) very little or no loss of alveolar bone in adults who have had a chronic gingivitis for a long time. This has been reported by Peder- sen ( 1937, 1939 ) and Hilming and Pedersen ( 1940 ) , who also found that gingivitis in the Eskimos on the west coast of Greenland is followed by the same breakdown of alveolar bone that Europeans have in cases of gingivitis. These Eskimos live together with the white population and eat their type of food. The lack of alveolar DENTO-ALVEOLAR RESORPTION IN PERIODONTAL DISORDERS 305 bone destruction in the east Greenland Eskimos may be due to the composition and consistency of their food. The latter ensures a very good functional development of the periodontal tissues. An explanation should also be given as to the loss of alveolar bone when very little or no color change and swelling is present in the gingiva. This type of periodontal disorder can be found in all age groups, more often in females than in males. A clinical ex- amination does not in these cases disclose local factors which can account for the loss of alveolar bone (Seidler et ah, 1950). In these patients very often the fixed and the movable gingiva have the same light pink color, with a normal mucous membrane. In gingival biopsies a regressive inflammation is found, usually with little or no hyperemia and exudative reactions. In autopsy, material bone formation is very rare even in sections from growing individuals. Patients with anemic gingiva frequently have cold hands and feet and diminished salivation. In some of them an increase in the blood cholesterol up to double the normal amount has been found. It has been suggested by Wang-Norderud ( 1951 ) that biopsies should be taken from the palate in the region of the second premolar in cases of anemic gingiva accompanied by advanced destruction of alveolar bone. The biopsies reveal that the lumen of the vessels in the palate usually is reduced owing to thickening of their walls. Loss of elastic fibers and degeneration in the vessels also occur (Fig. 5). Similar changes have been found by Quintarelli ( 1957 ) in the vessels of the jaws in patients with periodontal disorders. In biopsies from the palate of diabetic patients the pathological changes in the vessels can be very pronounced. These histological findings indicate that insufiBcient bone produc- tion is the main cause of the loss of alveolar bone when there is a decrease in the blood circulation in the periodontium. From the observations mentioned above, the conclusion can be drawn that circulatory disturbances increasing the tissue tonus in the periodontium initiate alveolar bone resorption and that a reduced flow of blood in the periodontium may impede the production of alveolar bone. 306 I. REICIIBORN-KJENNERl I) /^O^-- J;^:;'^^ ■vs. *;^-^i»^ ■ if Fig. 5. Reduction in the lumen ot an artery, from a biopsy of a patient with anemic gingiva. In the wall of the vessel, which is verv thick, atrophic and degenerative changes can be seen. ( X 200.) The Effect of Increased Functional Stress ON THE Alveolar Bone It has been explained that the masticatorv functional stress in- creases when alveolar bone is resorbed and that this may incite a compensatory bone formation. A corresponding production of ce- mentum and pericemental fibers also takes place ( Reichborn-Kjen- nerud, 1956). Mechanical functional stress does not have a formative effect, but a destructive one if it is raised above a certain limit. This has been proved in the previously mentioned experiments bv Haupl and Psansky ( 1938 ) , who by means of activators increased the magni- tude and frequency of the intermittent masticatory impulses. When the appliance had been in the mouth for 3 nights, compensatory tissue production was dominant; onlv in some areas were circulatory disturbances and bone resorption observed (Fig. 6). In sections from patients who had used the activators for 7 nights, extensive DENTO-ALVEOLAR RESORPTION IN PERIODONTAL DISORDERS 307 Fig. 6. Drawings of two premolars to demonstrate the reactions after 3 (left) and 7 (right) nights' use of an activator. The masticatory stress has mainly been increased in the direction of the arrows. Where the lamina dura is drawn as a double broken line, apposition has occurred. In the areas where the lamina dura is represented bv a solid line, aplasia was found. The zigzag lines indicate resorption areas. resorption of the lamina dura could be seen in the pressure zones of the pericementum ( Hiiupl and Psansky, 1938; Andresen et ah, 1957 ) . From these and other experiments it is known that circulatory disturbances are the first reaction to an increase in mechanical func- tional stress above the physiological limit. If the stress is not de- creased below this limit in one way or another, the circulatory disturbances are succeeded by atrophy, degeneration, mechanical inflammation, and necrosis (Fig. 7). These alterations which occur in the periodontium and in other organs are due to disturbances in the liying conditions of the cells ( Hiiupl and Lang, 1927; Kronfeld, 1931; Coolidge, 1938; Tillotson and Coventry, 1950; Reichborn- Kjennerud, 1956). That an exposure of the periodontal tissues to constant mechanical stress above the physiological limit will also break down the alveolar bone is seen in orthodontic treatment. In animal experiments where an uncontrolled, often traumatic effect of the masticatory forces is obtained bv cementing crowns which are too high on single teeth, alveolar bone resorption also occurs (Gottlieb and Orban, 1931; Wentz et al., 1958). In the same type of experiments in monkeys, Glickman and Smulow ( 1962 ) could observe alveolar bone resorp- 308 Compensatory hypertrophy (production of specific tissue) Regeneration -- Atrophy I. REICHBORN-KJENNERUD Ph. limit Pathological changes Circ. disturbances, atrophy, degeneration, mechanical inflam., necrosis Hypofunction -N- Hyperfunction Fig. 7. Abscissa, mechanical functional stress, increasing from left to right. Ordinate, tissue reactions to changes in the functional stress. N, normal func- tional stress activating tissue regeneration. Hvperf unction from N to the physi- ological limit is followed by compensatory tissue formation. Above this limit it results in pathological alterations and destruction of tissue. Production of normal tissues of different kinds decreases rapidlv. tion and development of inflammation in the snpra-alveolar gingiva. It is obvious that an increase of the functional stress above the physiological limit will aggravate pathological changes already pres- ent in the periodontium, and it has been found that it influences the spread of a gingival inflammation (Orban, 1954; Macapanpan and Weinmann, 1954; Glickman and Smulow, 1962). Alterations in the periodontal tissues caused by increased func- tional stress during mastication can be studied clinicallv in cases of hyperfunction in the front region due to a bite sinking in the lateral segments (Ramfjord, 1959). A reduction in the height of the bite in the premolar and molar region may have no effect on the front teeth, v^hich indicates that compensatory periodontal tissue formation has taken place. But very often, and particularly in older persons, the result is increased mobil- ity and migration of the front teeth. These are clinical symptoms of alveolar bone resorption. In the upper jaw, more or less or all of the bone facial to migrated front teeth may be lost (Fig. 8). This destruction is frequently followed by a deepening of the gingival DENTO-ALVEOLAR RESORPTION IN PERIODONTAL DISORDERS 309 A 11 Fig. 8. A, destruction of the maxillai\- bone in the labial alveolar region due to migration of the teeth following a bite sinking; osteophytes are seen on the facial side of the right lateral incisor. B, x-ray of the teeth in A. pockets on the lingual side of the migrated teeth. Resorption of the lamina dura of the teeth and of their root tips is also seen. The progression of these alterations can best be observed in x-ravs of the lower front region ( Fig. 9 ) . Here the first reaction usually is resorption in the central part of the septa, followed by an extension of their nutritional canals, widening of the apical part of the perice- mentum, and root tip resorption. When the changes are not accom- pa,nied by a breakdown in the marginal part of the alveolar bone. Fig. 9. Reactions to h)perfunction in mandibular incisor region. A, pulp stones in the region of the tooth necks, dentin formation in the crowns. B, pulp stones, dentin formation, breakdown of the central parts of the septa, widening of the apical part of the pericementum, root tip resorption. C, there is very little left of the pulp, the nutritional canals in the septa are visible, the apical part of the pericementum is extended, the root tips are blunt. In A and B there is no marginal destruction of the septa, indicating that the reactions in their pro- found parts are not due to a gingivitis, but to hyperf unction. 310 I. REICHBORN-KJENNERUD it nia\^ be assumed tliat tbev are due to increased functional stress. Another finding sustaining this concept is the nearly constant ap- pearance of pulp stones and the retraction of the crown pulps in the front teeth in these cases. These reactions must result from circula- tory disturbances in the apical part of the pericementum due to in- creased moxements of the root tips during mastication, A clinical s\ mptom which verv often precedes and always follows an alveolar bone resorption manifesting itself in an increased tooth mobility is dullness by transversal percussion of the teeth. Histolog- ical observations and experiments indicate that this symptom is due to circulatory disturl:)ances in the pericementum ( Reichborn-Kjen- nerud, 1956). Percussion "dullness" is a means of diagnosing an increase in the functional stress above the physiological limit. It can be concluded that an increase in the masticatory functional stress has the following effects. When it is raised to a certain limit it stimulates a compensatory production of alveolar bone. If it in- creases above this limit, this reaction does not occur, but pathological changes appear, followed by destruction of alveolar bone. Functional impulses of this type will thus impede the formation of alveolar bone and incite destruction of it. The Influence of Periodontal and Systemic Conditions ON THE Loss OF Alveolar Bone Because circulatory disturbances increasing the tissue tonus are followed by bone resorption, all external periodontal conditions con- tributing to the development of these changes in the periodontium will promote destruction of alveolar bone. The internal anatomy and the physiology of the periodontium and of all other organs will always influence the initiation and fur- ther spread of disorders, and are very often responsible for the re- lapse of diseases. The structure of the periodontium ma\' be more or less favorable for the loss of alveolar bone. It is probable that the resistance of east Greenland Eskimos and other peoples to loss of alveolar bone is partly due to the anatomy of their periodontia. Anatomical con- ditions having the opposite efl^ect are increased permeability and DENTO-ALVEOLAR RESORPTION IN PERIODONTAL DISORDERS 311 thickening of the periodontal vessels (Figs. 5 and 10). These and Other alterations facilitating the initiation of periodontal circulatory disturbances may be due to systemic disorders. A very quick change in the physiological periodontal conditions followed by loss of alveolar bone occurs frequentlv after irradiation treatment of tumors in the jaws (Stafne and Bowing, 1947; Bruce and Stafne, 1950; Shapiro et ah, 1960). Gingival inflammations and Fig. 10. Circulatory disturbances and fibrosis in a bone marrow space in an alveolar septum of the jaw. A large and a small artery seen in the picture both have a narrow lumen due to thickening of their walls. ( X 250. ) advanced destruction of the periodontal bone happen so often in these cases that extraction of the teeth in the region which is to be exposed to irradiation is recommended by many radiologists. Systemic disorders aftect the resorption and formation of alveolar bone in many ways. They can reduce the periodontium's resistance against irritants and thus contribute to the development of gingival inflammations. It is well known that hormonal disturbances, nutri- tional deficiencies, leukemia, and manv other diseases and old age can have this effect (Sheridan, 1959; Brinch, 1937; Boyle et al., 1937; Boyle, 1938; Glickman, 1946, 1948fl, 1948??; Dreizen, 1956; 312 I. REICHBORN-KJENNERrD Duffy and Driscoll, 1958; Biirket, 1961; Frandsen and Becks, 1962). Furthermore, some systemic disorders manifest themselves in periodontal circulatory disturbances, as is the case in scurvy and in hvperparathvroidism. They can also in different ways impede regeneration and repair of alveolar bone and even lead to excess formation of it. Clinical examinations sometimes do not reveal local or systemic conditions responsible for an abnormal alveolar bone picture. It may, however, be due to metabolic disturbances as in- dicated by observations made by Person ( 1959 ) . Metabolic studies should therefore be performed in these cases (Barr and Bulger, 1930; Albright and Reifenstein, 1948; Person, 1959; Silverman et al., 1962). Inherited factors should be mentioned because observations and experiments in animals suggest that they may be of importance for alveolar bone destruction (Baer and Lieberman, 1960; Baer et al., 1961; Shklar et al, 1962). In summarizing it can be said that periodontal and systemic con- ditions may influence the loss of alveolar bone by promoting the development and aggravation of periodontal circulatory disturb- ances, and by preventing formation of alveolar bone. In individual cases these effects may or may not appear, owing to the duration and the severity of the systemic disorder, and to the influence of many other factors. This explains the controversial reports as to the effect of systemic disorders on the periodontium. Dental Resorption in Periodontal Disorders It has been noted that the root tips and the surrounding bone can be resorbed because of circulatorv disturbances incited by hyper- function, which mav be the cause of tooth resorptions in other parts of the pericementum. Tooth resorptions are also seen in the marginal region starting from the gingiva. They appear in various types of periodontal dis- orders, most frequently in cases of hyperplastic gingivitis. Extensive gingival tooth resorptions are sometimes found in hyperplastic gin- gival inflammations following a trauma, and in cases of epulis tumors. Gingival resorptions from the connective tissue in gingival DENTO-ALVEOLAR RESORPTION IN PERIODONTAL DISORDERS 313 pockets may, however, occur when there is very httle or no swelHng of the gingiva. This happens frequently on the hngual side in the upper jaw. In histological sections it can be observed that progressive gingi- val resorptions usually are caused by fingerlike proliferations of inflamed gingival connective tissue. They may leave trabeculae of dentin in the resorption cavities, giving them a bonelike structure. This happens so frequently that a differential diagnosis between subgingival caries and resorption cavities can be based on this find- ing. In some cases bone or a hard tissue similar to bone is produced in resorption cavities (Fig. II). Fig. 11. Resorption extending far into the crown of a molar from proliferat- ing gingival granulation tissue. (X 10.) Gingival resorptions of the teeth by means of perforating vessels also occur. They may start from the gingival papilla and extend deep into the crown (Figs. II and 12), This type of resorption is very often observed in the periodontal bone during the eruption of the permanent teeth. Resorptions of the teeth in the marginal region relatively seldom appear as compared with the nearly constant alveolar crest resorp- tions in periodontal disorders. Advanced resorptions of the alveolar 314 I. REICIIBORN-KJENNERUD Fig. 12. Detail of Fig. 11. A vessel from the gingiva is entering the resorp- tion cavitv, which is filled with connective tissue and bonelike structures. (X35.) bone from the pericemental side of the lamina dura mav be found while no resorptions or very small ones are seen on the opposing tooth surface. It has not been explained why the alveolar bone often is resorbed before the teeth, as is also the case in orthodontic treatment. Ac- cording to histological observations, resorptions do not appear on a root surface where pericemental fibers are intact. This indicates that the fillers must be broken down before the resorption can take place. On the pericemental side of the lamina dura there usually are some places not covered by pericemental fibers, particularlv near the canals where branches of the interalveolar vessels enter the pericementum. This may be the reason why alveolar bone is re- sorbed before the teeth. In most of the resorption cavities it can be seen that resorption has been followed by apposition of a bonelike hard tissue, which again may be broken down. This is probably due to changes in the tissue tonus in the resorbing connective tissue. DENTO-ALYEOLAR RESORPTION IN PERIODONTAL DISORDERS 315 The clinical obseivatioiis and histological findings mentioned above indicate that tooth resorptions are caused by circulatory dis- turbances. An explanation has been suggested as to the reason why tooth resorptions occur more seldom than resorptions of the al\ eolar bone. Summary Alveolar bone resorption in periodontal disorders has been dis- cussed in relation to local circulatory disturbances, increased func- tional stress, and general systemic influences. 1. Histological observations indicate that circulatory disturbances, which increase the tissue tonus in the periodontium, initiate alveolar bone resorption, and that a reduced flow of blood in the periodon- tium may impede the production of alveolar bone. 2. When the masticatory functional stress is raised to a certain limit it stimulates a compensatory production of alveolar bone, whereas if it is increased above this limit pathological changes appear, including destruction of alveolar bone. 3. Systemic conditions may influence the loss of alv^eolar bone by' inciting and aggravating periodontal circulatory disturbances, and by preventing new formation of alveolar bone. 4. On the pericemental side of the lamina dura there usually are some areas not covered by pericemental fibers, particularly near the canals where branches of the interalveolar vessels enter the pericementum. This may be the reason why alveolar bone is resorbed before the teeth, which, according to histological observations, do not undergo root surface resorption where pericemental fibers are intact. References Albright, F., and Reifenstein, E. C, Jr. 1948. The Parathyroid Glands and Metabolic Bone Disease. Williams and Wilkins Co., Baltimore, Md. Andresen, ^^, Hiiupl, K.. and Petrik, L. 1957. Funktions-Kieferorthopadie. Johann Ambrosias Barth, Munich. Baer, P. N., and Lieberman, J. E. 1960. Periodontal disease in six strains of inbred mice. /. Dental Research, 39, 215-255. 31 G I. REICHBORN-KJENNERUD Baer, P. N., Crittenden, L. B., Jay, G. E., and Lieberman, J. E. 1961. Studies on periodontal disease in the mouse. II. Genetic and material effects. /. Dental Research, 40, 23-33. Barr, D. P., and Bulger, H. A. 1930. Clinical syndrome of hyperpara- thyroidism. Am. J. Med. ScL, 179, 449-476. Boyle, P. E. 1938. Dietary deficiencies as a factor in the etiology of diffuse alveolar atrophy. /. Am. Dental Assoc, 25, 1436-1446. Boyle, P. E., Bessey, O., and Wolback, S. B. 1937. Experimental produc- tion of the diffuse alveolar bone atrophy type of periodontal disease by diets deficient in ascorbic acid (vitamin C). /. Am. Dental Assoc., 24, 1768-1777. Brinch, O. 1937. Dentale und paradentale Gewebsveranderungen bei Skorbut. Paradentium, 9, 120-135. Bruce, K. W., and Stafne, E. C. 1950. The effect of irradiation on the dental system as demonstrated by the roentgenogram. /. Am. Dental Assoc, 41, 684-689. Burket, L. W. 1961. Oral Medicine. Diagnosis and Treatment, 4th edition. Pitman Medical Publishing Co. Ltd., London. Coolidge, E. D. 1938. Traumatic and functional injuries occurring in the supporting tissues of human teeth. /. Am. Dental Assoc, 25, 343-357. Dreizen, S. 1956. Current knowledge of nutrition relative to the support- ing dental structures. /. Periodonfol., 27, 262-270. Duffy, J. H., and Driscoll, E. J. 1958. Oral manifestations of leukemia. Oral Surg., Oral Med. and Oral Pathol, 11, 484-490. Frandsen, A. M., and Becks, H. 1962. The effect of hypovitaminosis A on bone healing and endochondral ossification in rats. Oral Surg., Oral Med. and Oral Pathol, 15, 474-487. Glickman, I. 1946. The periodontal structures in experimental diabetes. N. Y. /. Dentistry, 16, 226-251. Glickman, I. 1948a. Acute vitamin C deficiency and periodontal disease. I. The periodontal tissues of the guinea pig in acute vitamin A de- ficiency. /. Dental Research, 27, 9-23. Glickman, I. 1948Z?. Acute vitamin C deficiency and periodontal disease. II. The effect of acute vitamin C deficiency upon the response of the periodontal tissues of the guinea pig to artificial induced inflam- mation. /. Dental Research, 27, 201-210. Glickman, I. 1949. The experimental basis for the "bone factor" concept in periodontal disease. /. Periodonfol, 20, 7-22. Glickman, I. 1952. Interrelation of local and systemic factors in periodon- tal disease: Bone factor concept. /. Am. Dental Assoc, 45, 422-429. Glickman, I., and Smulow, J. B. 1962. Alterations in the pathway of gingival inflammation into the underlying tissues induced by exces- sive occlusal forces. /. Periodontol, 33, 7-13. DENTO-ALYEOLAR RESORPTION IN PERIODONTAL DISORDERS 317 Gottlieb, B. 1925. Schmiitzpijorrhoe, Parodentalptjorrlwc and Alveolar- atrophie. Urban und Schwarzenberg, Vienna. Gottlieb, B., and Orban, B. 1931. Die Vercinderunpen der Geivebe bei iihermtissiger Beanspruchung der Zlihne. Georg Thieme, Leipzig. Haupl, K. 1925. Mikroslcopische Befunde bei Gingivitis und den durch sie bedingten Veranderungen der Alveolarkuppe nebst einigen Bemerkungen zur Frage der sogenannte Alveolarpyorrhoe. Viertel- jahresschr. Zahnheilk., 41, 1-48. Haupl, K., and Lang, F. J. 1927. Die marginale Paradentitis. Hermann Meusser, Berlin. Haupl, K., and Psansky, R. 1938. Histologische Untersuchungen iiber die \\^irkungsweise der in der Funktionskieferorthopadie gebrauchten Apparate. Deiif. Zahn-, Minid- u. Kieferheilk., 5, 214-224, 485-505, 691-648. Hilming, F., and Pedersen, P. O. 1940. Uber die Paradentalverhaltnisse und die Abrasion bei rezenten ostgronliindischen Eskimos. Para- dentium, 12, 69-86. James, W. W., and Counsell, A. 1928. The primary lesion in so-called pyorrhaea alveolaris. Brit. Dental J., 49, 1129-1135. Karolyi, M. 1905. Studium liber Alveolar-Pvorrhea. Osterr. Z. Stomatol., 3, 316-318. Kronfeld, R. 1931. Histologic study of the influence of function on the human periodontal membrane. /. Am. Dental Assoc, 18, 1242-1274. Lang, F. J. 1923. Zur Kenntnis des Knochenschwundes und -anbaues bei der bisher als Alveolarpyorrhoe bezeichneten Paradentitis. Vier- teljahresschr. Zahnheilk., 39, 489-501. Macapanpan, L. C., and Weinmann, J. P. 1954. The influence of injury to the periodontal membrane on the spread of gingival inflammation. /. Dental Research, 33, 263-272. Orban, B. 1954. Biologic basis for correcting occlusal disharmonies. /. Periodontal, 25, 257-263. Pedersen, P. O. 1937. Dental conditions among the west Greenlanders in former times as well as nowadavs. Tandlaegehladet (Trans. Congr. intern, assoc. recherches paradontopathies) , pp. 58-61. Pedersen, P. O. 1939. Ernahrung und Zahnkaries primitiver und ur- banisierter Gronliinder. Dent. Zahn-, Mund- n. Kieferheilk., 6, 728- 734. Person, P. 1959. Metabolic studies of human alveolar bone disease. Oral Surg., Oral Med. and Oral Pathol, 12, 610-635. Pommer, G. 1885. Untersuchungen iiber Osteomalacie und Rachitis nebst Beitrdgen zur Kenntnis der Knochenresorption und Apposition in verschiedenen Altersperioden unci der durchbohretiden Gefdsse. F. C. Vogel, Leipzig. 318 I. REICIIBORN-KJENNERUD Pommer, G. 1925. Uber Osteoporose, iliren Urspning und ilire differential- diaenostische Bedeutune. Arch. klin. Cliir., 136', 1-35. Quintarelli, G. 1957. Histopathology of the human mandibular artery and arterioles in periodontal disease. Oral Siirf^., Oral Med. and Oral Pathol, 10, 1047-1059. Ramfjord, S. P. 1959. Periodontal reaction to functional occlusal stress. /. Periodonfol, 30, 95-112. Recklinghausen, F. von. 1891. Die fibrose oder deformierende Ostisis, die Osteomalacic und die osteoplastische Carcinose in ihren gegenseiti- gen Beziehungen. In Festschrift der Assistentcn R. Virchoivs. G. Reimer, Berlin. Reichborn-Kjennerud, I. 1956. Funktionell-protetische Behandlung und Prophylaxe der Zahnlockerung und der Dysgnathien. In Die Zahn-, Miind- und Kieferlicilkunde; Ein Uandhuch fiir die zahmirztliche Praxis, Vol. 4, pp. 481-640. Urban und Schwarzenberg, Munich/Ber- lin. Roux, W. 1895. GcsammeJte Ahhandhingcn idjcr EntwickJungsmechanik der Organismen. W. Engelmann, Leipzig. Seidler, B., Miller, S. C., and \\ olf, W. 1950. Systemic aspects of pre- cocious advanced alveolar bone destruction: Preliminary report. /. Am. Dental Assoc., 40, 49-58. Shapiro, M., Brat, V., and Ershoff, B. H. 1960. Periodontal changes follow- ing multiple sublethal doses of X-irradiation in the mouse. /. Dental Research, 38, 668. Sheridan, R. C., Jr. 1959. Epidemiology of diabetes mellitus. II. A study of 100 dental patients. /. Periodontol, 30, 298-323. Shklar, G., Cohen, M. M., and Yerganian, G. 1962. A histopathologic study of periodontal disease in the Chinese hamster with hereditary dia- betes. /. Periodontol, 33, 14-22. Silverman, S., Gordan, G. S., Grant, T., Steinbach, H., Eisenberg, E., and Manson, R. 1962. The dental structures in primary hvperparathy- roidism. Oral Surg., Oral Med. and Oral Pathol, 15, 426-436. Stafne, E. C, and Bowing, H. H. 1947. The teeth and their supporting structures in patients treated bv irradiation. Am. J. Orthodontics, 33, 567-581. Thoma, K. H., and Goldman, H. M. 1937. The classification and his- topathology of parodontal disease. /. Ain. Dental Assoc, 24, 1915- 1928. Tillotson, J. F., and Coventry, M. B. 1950. Spontaneous ischemic necrosis of the anterior tibial muscle. Proc. Staff Meetings Mayo Clinic, 25, 223-227. Wang-Norderud, R. 1951. Sirkulasjonsforstyrrelsenes betydning for para- dentopathiene. — En orientering. Norske Tannlaegeforen. Tidsskr., 61, 385-393. DENTO-ALVEOLAR RESORPTION IN PERIODONTAL DISORDERS 319 Watt, G., and Williams, C. H. M. 1951. The effects of the physical con- sistency of food on the growth and development of the mandible and the maxilla of the rat. Am. J. Orthodontics, 37, 895-928. Wentz, F. M., Jarabak, J., and Orban, B. 1958. Experimental occlusal trauma imitating cuspal interferences. /. Periodontol., 29, 117-127. Weski, O. 1921. Rontgenologisch-anatomische Studien aus dem Gebiete der Kieferpathologie. Vierteljahrcsschr. ZahnJieilk., 37, 3-56. Weski, O. 1928. Allgemeine Gesichtspunkte zur Aetiologie der Paradentose und ihrer kausalen Therapie. Zahnarztl. Rundschau, 37, 1337-1343, 1408-1413. 12 Bone Remodeling during Dental Eruption and Shedding SURINDAR N. BHASKAR, Department of Dental and Oral rathology. United States Army Institute of Dental Research, Walter Reed Army Medical Center, Washington, D. C. THE process of physiologic resorption and apposition of bone tissue in a hone is termed "modeling" or "remodeling" and it is an essential part of osteogenesis. This process begins almost as soon as the pri- mary centers of ossification of a bone are formed and continues throughout life. During the active phase of bone growth, as for example in the tibia of a child, modeling resorption occurs ver\' rapidly, whereas in the adult or later periods of life it slows down. Modeling of bones is species and organ specific, that is, it varies in intensity and location from bone to bone as well as in the same bone of different species. Whereas the process of endochondral as well as intramembranous ossification is identical in different species, the pattern of bone remodeling is markedly different. Thus, at the cartilaginous model stage, the tibia of a dog and that of a monkey may be very similar, but soon after the appearance of the ossifica- tion centers and the beginning of remodeling, they quickly begin to assume different shapes. This is also true for different bones in the same skeleton. In addition to contributing to their morphogenesis, modeling is also important in the functional adaptation of bones. This means that 321 322 S. N. BHASKAR through modehng and internal reconstruction bones are adapted to the functional stresses placed upon them. Like the rest of the skeleton, the mammalian jaws are subjected to external and internal remodeling. The process in the jaws, how- ever, is much more dramatic and precise and has more far-reaching effects than in the other bones. This is because the development, eruption, functioning, and shedding of teeth are intimately con- nected with, and are in part dependent upon, the internal remodel- ing in the maxilla and the mandible. This interrelationship is the subject of the present discussion. Bone Remodeling during Dental Eruption The presence of the epithelial organs, the tooth germs, within the maxilla and the mandible is a unique anatomic feature of these bones ( Fig. 1 ) . There are 52 such structures within the human jaws. Fig. 1. Frontal section through the maxiha of a fetus 8 months in itfero. Note eight tooth germs in various stages of development. ( X 6. ) The growth and development of each of these organs is somewhat similar to that of the others, but the period of development varies greatly from tooth to tooth. Odontogenesis, or tooth development, begins in liter o and is not completed for almost two decades. During this long period, tooth germs grow in size, erupt, function, and BONE REMODELING IN DENTAL ERUPTION AND SHEDDING S'23 adjust their positions within the jaws. In this comphcated proeess, niodehng resorption plavs a dominant role. During their hfe span, the tooth germs and the teeth go through the following major phases ( Bhaskar, 1962 ) : growth, movement out of the jaws, and functional adaptation. The remodeling of bone as related to each of these phases will be discussed separately. Growth Period The tooth germs originate from the oral epithelium and from there extend into the underlying bone. Thev are separated from the latter bv mesenchyme, and the cayity in which they lie is called the bony crypt (Fig. 2). From this point on, the growth, location, and moyement of the tooth germs is dependent upon and closely linked to the modeling changes in the crypt. During the growth Fig. 2. A maxillary tooth germ with its bony crypt ( X 10 ^ 324 S. N. BIIASKAR phase, the tooth germ enlarges and the bone tissue of the crypt undergoes rapid bone resorption (Fig. 3). Since the growth of the tooth germs is eccentric and varies from tooth to tooth, the intensity of resorption on the wall of the bony crypt varies markedly in differ- ent crypts as well as in different areas of the same crypt. During this phase bone resorption on the crypt wall is striking, but the epithelial tooth germ does not come into contact with bone tissue. It is separated from it by a zone of highly cellular mesenchvme -i»£S***-^****" Fig. 3. Osteoclastic resorption of the wall of a bony crypt opposite the epithelial (top) tooth germ. (X 80.) which in some areas ( usuallv apical ) contains a great deal of inter- cellular fluid (Fig. 4). Movement Out of the Jaws After the tooth germ has reached a certain size (completion of the anatomic crown ) , it begins a pronounced movement toward the oral cavity. The mechanism of this outward migration is not fully known, but the movement is accompanied by rapid bone formation on the wall of the crypt (Fig. 5). As in the earlier resorptive phase, the rate of bone apposition varies in different crypts, in different areas of the same crypt, and at different periods of this phase. During periods of rapid growth the apposition occurs as widely BONE REMODELING IN DENTAL ERUPTION AND SHEDDING 3^25 N '^-- Fig. 4. A portion of the tooth germ (top) is separated from the bony crypt (bottom) by a wide zone of loose, almost edematous connective tissue. (X 67.) Spaced trabeculae of embryonic bone (Fig. 5), while in areas or periods of slower growth, surface apposition of bone can be seen. The extent as well as the direction of migration of teeth into the oral cavity is intimately related to the apposition pattern of bone in the crypt. Functional Adaptation After the tooth erupts and comes into function, the remodeling of the bone of the crypt, now called the socket, continues. It is, however, much less marked. In man, fully erupted teeth undergo wear on their occlusal, anterior (mesial), and posterior (distal) surfaces. This wear is accompanied by an anterior (mesial) migra- tion of all human teeth. The migration is associated with a charac- teristic remodeling pattern in the socket. It consists of bone apposi- 326 S. X. BIIASKAR Fig. 5. A molar tooth germ in the mandible. Note rapid bone formation on the wall of the crypt. (X 10.) (From Bhaskar, 1962.) tion on the posterior (distal) walls and bone resorption on the anterior ( mesial ) walls of all tooth sockets ( Fig. 6A and B ) . It has been suggested that the anterior migration of human teeth, which is of great importance in maintaining the integrity of the dental arch, is the result of the remodeling of bone in their sockets. From the brief description above it is apparent that, first, the development and eruption of teeth are intimately associated with the internal remodeling of the maxilla and mandible; and, second, the remodeling process is very intense in the early phases of odonto- genesis, slows down progressivelv, but continues throughout life. The intimate relationship between bone remodeling and tooth BONE REMODELING IN DENTAL ERUPTION AND SHEDDING 3'27 Fic. (i. I ppci figure (A) shows the posterior (distal) wall of the socket of an erupted tooth. In this area the bone shows apposition. Lower figure (B) shows the anterior (mesial) wall of the socket of an erupted tooth. Note osteo- clasts and bone resorption. (Both figures X 67.) 328 S. N. BHASKAR eruption is illustrated by numerous reports. Dramatic illustration of this interdependence is seen in recessive mutations in the rat (Greep, 1941; Schour et al, 1949; Bhaskar et al, 1950, 1952), mouse (Grune- berg, 1935, 1936, 1937, 1938), and rabbit (Pearce and Brown, 1948; Pearce, 1948, 1950rt, 19505). Although the changes in these mutations are similar in nature, only those of the ia rat will be described. Dentition and the Bone Tissue of the ia Rat The ia mutation in the albino rat is characterized by a retardation of resoi'ption of bone tissue in the entire skeleton ( hereditary hypo- osteoclasis) {Bhaskar et al, 1950; Bhaskar, 1953; Mohammed, 1957). As a consequence, the developing and growing bones show general- ized osteosclerosis. In the jaws, the lack of physiologic modeling resorption of the alveolar process leads to the distortion and anky- losis of developing teeth (Figs. 7 and 8) (Schour et al., 1949). The proliferation and differentiation of the odontogenic epithelium is normal. However, because of the failure of remodeling of the bony crypt, the growing epithelial dental organ approaches the wall of the crypt and then extends into the surrounding marrow spaces. Although the odontogenic epithelium continues to proliferate at a normal rate, the normal growth pattern of the tooth is disturbed owing to the persistence of the adjacent bone trabeculae. The grow- ing odontogenic epithelium follows the course of least resistance and invades the bone marrow spaces in a haphazard and disorderly fashion. The odontogenic epithelium maintains its physiologic poten- tialities, that is, it differentiates into pulp tissue and odontoblasts, and this leads consequently to the formation and calcification of dentin and enamel. The irregular invasion of the odontogenic epi- thelium into the bone marrow spaces leads to the presence of strands and islands of dental tissues in different stages of development, on and between the bone trabeculae. This has two consequences. First, the teeth are locked to the surrounding bone, causing ankylosis, which prevents first any movement of the tooth germ and later the eruption of the tooth. Secondly, the continuous growth and the differentiation of the odontogenic epithelium leads to the accumula- BONE REMODELING IN DENTAL ERUPTION AND SHEDDING 329 Figs. 7 and 8. Incisor and premaxilla of a normal (Fig. 7) and an ia (Fig. 8) rat at 10 days of age. The ia rat incisor has not erupted. (Both figures X 21.) tion of the dental tissues and the formation of tumoilike masses. In summary, it may be stated that the cause of the development of the odontome-like formation in the ia rat is the resistance of the bone tissue to resorption in the early days of life. The proliferation 330 S. N. BlIASKAR and histodiffeieiitiation of the individual cells of the odontogenic epithelium are not disturbed. It is apparent that a lack of modeling in the maxilla and mandible not onh aborts the eruption of teeth, but leads to formation of odon- tome-like structures. The reason for the failure of ia bone to undergo resorption is as yet unclear. It has been shown that ( 1 ) the ia bone does undergo resorption, but this process is slow and delayed (Bhaskar et ah, 1950); (2) the ia skeleton does have osteoclasts (Bhaskar et ah, 1956) (Fig. 9); (3) high dosage of parathyroid hormone promotes Fig. 9. Bone of ia rat with ostt'otlasts 750) bone resorption and eruption of ia teeth (Bhaskar et ah, 1952); (4) the ia bones are hypercalcified (Kenny et al., 1958); (5) the ia de- fect is not due to hypoparathvroidism (Kenny et al., 1958). In view of the above findings and in the light of our present knowledge concerning bone resorption, it would appear that hypo- BONE REMODELING IN DENTAL ERUPTION AND SHEDDING 331 Figs. 10 and 11. Bone tissue of the normal (Fig. 10) and ia (Fio-. 11) rat at 20 days of age. (Both figures X 300.) " ^ 332 S. N. BIIASKAR osteoclasis in the ia rat is prol)ably due to some defect in the bone tissue itself rather than in the osteoclasts or the parathyroid glands. This belief is strengthened by the fact that grossly the ia bone looks more basophilic, is composed of thinner trabeculae, and contains more resting lines than normal (Figs. 10 and 11). The ia jaws demonstrate the complete dependence of normal odontogenesis upon the modeling resorption of the bony crypt. Bone Remodeling in Shedding of Teeth Man has two sets of teeth, the primary or deciduous and the permanent or secondary. The deciduous or primary teeth are lost or exfoliated at, or before, the emergence of the permanent teeth. This process is called shedding of teeth. It consists of a progressive re- FiG. 12. A portion of the human mandible showing the bicuspid teeth (un- erupted) and deciduous molars. The bicuspid on the right has already replaced part of the socket of the deciduous tooth. ( X 6.) BONE REMODELING IN DENTAL ERUPTION AND SHEDDING 333 sorption of first the socket and then the roots of the primary tooth (Figs. 12 and 13), and it begins with the eruption of the permanent successor. As the permanent tooth moves toward the oral cavity, pressure is brought to bear, first upon the alveolar bone tissue which surrounds the deciduous roots and then on the roots of the decidu- ous teeth (Fig. 14). In the connective tissue intervening between these structures, osteoclasts (or odontoclasts) differentiate and the resorption of the bony socket as well as the root can be seen. One can see such areas as irregular, deep excavations in the root surface Fig. 13. Not only has the permanent cuspid (left) produced resorption of a part of the socket of the deciduous tooth, but the root of the latter is also undergoing resorption. (X 6.) 334 S. N. BHASKAR Fig. 14. Enamel of the crown of a permanent tooth (lower right) is sepa- rated from the root of a deciduous tooth bv connective tissue. The deciduous root shows Howship's lacunae. (X 6.) which contain multinucleated giant cells. It is apparent that the part of the deciduous roots which first undergoes resorption depends upon the location of the permanent tooth and the direction in which it is moving. The process is much more complicated in the area of the bicuspids than it is in the incisor region of the maxilla and the mandible. This is because the anterior teeth are single-rooted, whereas in the bicuspid area the multi-rooted deciduous teeth are replaced by multi-rooted successors. It is apparent from the description that during shedding of teeth, as in eruption, the remodeling of bone plays an essential role. Not only is the resorption of the bonv crypt of the deciduous tooth es- sential for the eruption of the permanent successor, but after the latter has erupted, the formation of a new socket and a new suspen- sory ligament must occur. Unless this new socket is formed through a process of remodeling, the new tooth is virtually useless. BONE remodf:ling in dental eruption and shedding 33o Discussion During eruption of a tooth a series of very complicated tooth movements occur within the jaws. These are essential so that the tooth may (a) keep pace with the growing jaws, (b) maintain a certain relationship to other teeth, and (c) maintain a certain rela- tionship to the oral epithelium. Although there are only two sets of teeth in man, chronologically, the 26 teeth in each jaw are distinct organs. It is apparent, then, that for a normal dentition, the move- ment of teeth during and before eruption must be very precise. We have seen that the remodeling of bone is a constant companion to this movement. Whether the remodeling is the result or the cause of tooth movements is a much debated question. There are two main beliefs. One is that the tooth germ grows and produces pres- sure on the surrounding bone, which then vmdergoes osteoclastic resorption, and as a tooth germ moves away from a crypt wall, bone apposition occurs on this surface. According to this concept, bone remodeling within the jaws is a passive phenomenon. The second belief is that the internal remodeling of the jaws is a predetermined genetic pattern, in much the same way as the modeling pattern of a tibia or a femur is a genetically determined pattern. According to this concept, the teeth move as a result of this pattern of bone apposition and resorption. Regardless of which theory is correct, it is undeniable that the internal modeling of the maxilla and the mandible has a great deal to do with the normal development and eruption of teeth. In the case of shedding of teeth, the eruptive force of a perma- nent tooth plays an important role in the loss of the deciduous tooth. This eruptive pressure first produces resorption of the bony socket, and then the resorption of the root of the deciduous tooth. These resorptive changes are accompanied by osteoclasts and odontoclasts. The changes which induce the resorptive phenomenon are seen in a multitude of circumstances in the skeleton. Pressure resorption of the bone tissue adjacent to tumors, cysts, aneurysms, and orthodontic appliances are examples of this phenomenon. Re- gardless of the mechanism, it is certain that bone remodeling is im- portant in shedding of teeth as it is in their eruption. 336 s. n. bhaskar Summary It is apparent that during both the eruption and the shedding of human teeth there is an intimate correlation between the changes in the tooth and those in the surrounding bone. During tooth devel- opment and growth, the growth of the epithelial dental organ pro- duces changes in the surrounding mesenchyme which lead to resorp- tion of bone. Whether this bone resorption is a result of growth pressure of the tooth germ and surroimding mesenchyme on the bony crypt or is a function of a genetically predetermined pattern of bone remodeling (as in the metaphyseal area of long bones) is difficult to determine. During the movement of teeth out of the jaws, the internal reconstruction of the sockets consists essentially of bone apposition. Whether this process is the result or the cause of eruption has not been established, but its intimacy with the active eruptive phase is unquestioned. The period of tooth shedding is again marked by extensive re- modeling of the alveolar process. The sockets of the deciduous teeth are destroyed by the erupting permanent teeth and rebuilt to adapt to their much larger roots. In short, the remodeling of alveolar processes of the jaws is a continuous process which begins with the beginning of odonto- genesis and does not cease until all the teeth have been lost. References Bhaskar, S. N. 1953. Growth pattern of the rat mandible from 13 days in- semination age to 30 days after birth. Am. J. Anat., 92, 1-54. Bhaskar, S. N. 1962. Synopsis of Oral Histology. C. V. Mosby Co., St. Louis, Mo. Bhaskar, S. N., Mohammed, C. I., and Weinmann, J. P. 1956. A morpho- logical and histo-chemical study of osteoclasts. /. Bone and Joint Surg., 38A, 1335-1345. Bhaskar, S. N., Schour, I., Creep, R. O., and Weinmann, J. P. 1952. The corrective eflFect of parathyroid hormone on genetic anomalies in the dentition and the tibia of the ia rat. /. Derital Research, 31, 257-270. Bhaskar, S. N., Weinmann, J. P., Schour, I., and Creep, R. O. 1950. The growth pattern of the tibia in normal and ia rats. Am. J. Anat., 86, 439-478. BONE REMODELING IN DENTAL ERUPTION AND SHEDDING 337 Greep, R. O. 1941. An hereditary absence of the incisor teeth. /. Heredity, 32, 397-398. Gruneberg, H. 1935. A new sub-lethal color mutation in the house mouse. Proc. Roy. Soc. (London), Ser. B, 118, 321-342. Gruneberg, H. 1936. Grey-lethal, a new mutation in the house mouse. /. Heredity, 27, 105-109. Gruneberg, H. 1937. The relation of exogenous and endogenous factors in bone and tooth development. The teeth of the grey-lethal mouse. /. Anat., 71, 236-244. Gruneberg, H. 1938. Some new data on the grey-lethal mouse. /. Genet., 36, 153-170. Kenny, A. D., Toepel, W., and Schour, I. 1958. Calcium and phosphorus metabolism in the ia rat. /. Dental Research, 37, 432-443. Mohammed, C. I. 1957. Growth pattern of the rat maxilla from 16 days insemination age to 30 days after birth. Am. J. Anat., 100, 115-165. Pearce, L. 1948. Hereditary osteopetrosis of the rabbit. II. X-ray hema- tologic and chemical observations. /. Exptl. Med., 88, 597-620. Pearce, L. 1950«. Hereditary osteopetrosis of the rabbit. III. Pathologic observations; skeletal abnormalities. /. Exptl. Med., 92, 591-600. Pearce, L. 1950/:>. Hereditary osteopetrosis of the rabbit. IV. Pathologic observations; general features. /. Exptl. Med., 92, 601-624. Pearce, L., and Brown, W. H. 1948. Hereditary osteopetrosis of the rabbit. I. General features and course of disease; genetic aspect. /. Exptl. , Med., 88, 579-596. Schour, I., Bhaskar, S. N., Greep, R. O., and Weinmann, J. P. 1949. Odon- tome-like formations in a mutant strain of rats. Am. J. Atiat., 85, 73- 112. \av 13 The Deciduous Nature of Deer Antlers RICHARD J. GOSS, Department of Biology, Brown University, Provi- dence, Rhode Island FEW events in nature rival the unique mechanisms by which the annual renewal of antlers is achieved. The rate at which antlers must elongate in order to attain lengths of up to 4 feet or more in growing seasons of only a few months may exceed 1 centimeter per day, a growth rate probably unequaled elsewhere in the animal kingdom. Once the ultimate dimensions are reached, the entire antler dies, whereupon the \'elvet\^ skin peels off revealing the compact bone beneath. As nonliving structures penetrating the epidermis (Fig. 9), antlers represent one of the rare instances, together with teeth and the placoid scales of elasmobranchs, in which the continuity of the skin is naturallv, though temporarilv, interrupted. Serving as weap- ons and status symbols during the rutting season and for varying periods thereafter, the burnished antlers are shed as the succeeding year's set begin to grow. This is achieved by osteoclastic activity at the junction between the dead bone of the basal part of the antler and the living bone of the frontal pedicle from which the antler originates. The shedding of antlers represents an extreme instance of a rather common biological phenomenon, usually represented by the loss of certain epidermal structines in various animals following their respective breeding seasons. The pronghorn antelope, for example, sheds the outer keratinized sheaths of its horns annually 339 3-40 R. J. GOSS (Caton, 1877; Skinner, 1922; Noback, 1932). Among birds a variety of epidermal structures may be shed (Rawles, 1960). The colorful bills of puffins of both sexes are detached from the basal portions after the breeding season, as described by Bent (1919), Forbush ( 1936 ) , and Lockley ( 1953 ) . In the male white pelican there is present during the breeding season a cornilied crest on the upper beak which, according to Coues (1872), "appears to be shed and renewed in a manner analogous to the casting of deer's horns." These and other more familiar cases of molting involve structures exclusively of epidermal origin. Teeth and antlers, however, rep- resent the sole instances of the natural loss of mineralized vertebrate tissues bv shedding. All these cases are examples of how the body gets rid of excess dead tissue from its external surface. The phenom- enon of antler shedding, however, has inspired some rather unique interpretations, not the least of which is the notion of Ries ( 1948 ) that surplus sexual hormones are stored in antlers during the grow- ing period. Upon the completion of antler growth these hormones are supposedly free to induce rutting behavior. The legendary belief that antlers possess medicinal (including aphrodisiac) prop- erties (cf. Scherbatoff, 1933) is cited as evidence in favor of such an incredible hypothesis. The distinction between truth and myth is not, unhappily, always so obvious. Mechanism of Antler Shedding To appreciate fully what is known of the mechanism of antler shedding, one does well to adopt the view of Waldo and Wislocki ( 1951 ) that "shedding of the old antler is in reality incident to growth of the new one." In a sense, shedding of the antler repre- sents the crucial point of transition between what Gruber ( 1952a ) refers to as the bionegative, or destructive, processes and the bio- positive, or constructive, processes in the antler cycle. Indeed, shedding is necessary only because, in the nature of things, death of the antler is the culmination of its growth and development. The normal course of antler development in the sika deer is illustrated in Figs. 1 to 7. Death of the mature antler is characterized by three principal DECIDUOUS NATURE OF DE?:R ANTLERS .'Ul events, namely, hardening of the bone, vascular constriction, and shedding of the velvet. These events are set in train by increased testosterone secretion, which also brings about the period of rut following antler death. From this time on there exists, perforce, a union between the compact dead bone of the antler base and the live bony trabeculae of the pedicle. The line of demarcation is at first usually convex, with respect to the pedicle, as is illustrated by the concave base of an antler caused, by castration, to be shed in the early autumn (Fig. 10). During the intervening months the antler apparently dies back gradually, for when shedding occurs in the spring, either naturally or because of castration, its base tends to be convex (Figs. 11 and 12). In either case, the breakage plane is characterized by numerous small bony spicules which are re- sponsible for the rough texture of the antler base ( Fig. 8 ) . This base is also remarkable for the normal absence of any traces of blood, a fact indicating that it is the dead, not the living, bone that is eroded prior to shedding. In 1859-1861, Wyman described the absorption of bone from around the haversian canals just below the "burr," until each cavity united with adjacent ones to effect a separation of the antler from the pedicle. Amplified by the subsequent investigations of Lieber- Kiihn (1861), Kolliker (1873), Caton (1875), Macewen (1920), Gruber (1937, 1952^, 1952Z>), Wislocki (1942), and Waldo and Wislocki (1951), Wyman's original account still represents an ac- curate description of the basic process of antler shedding. Kolliker (1873) emphasized that the events preceding shedding involve hyperemia in the frontal pedicle and resorption of haversian la- mellae in the central and peripheral regions of the antler-pedicle junction. Thus, in its final days the antler remains attached only by osseous trabeculae in a circular, intermediate zone, a fact that can be verified by close examination of an antler base after shedding (Fig. 8). In his investigations of roe deer and fallow deer, Gruber ( 1937, 1952a, 1952&) described and illustrated the numerous osteoclasts so conspicuously arranged along the trabeculae of the distal end of the pedicle. Under their influence, the bone is eroded and the inter- trabecular spaces are correspondingly enlarged to accommodate I'^KURES 1 TO 9 342 DECIDUOUS NATURE OF DEER ANTLERS 343 Fig. 10. Sagittal section showing concave base of antler caused, by castra- tion, to be shed on October 25. Note the absence of compact bone at plane of shedding. Fig. 11. Antler shed on April 2 by a castrated deer, showing convex base. Fig. 12. Tvpical convex base of an antler shed in the spring bv a normal deer. the congested blood vessels of the region. Emphasizing the possible importance of neurovascular influences in antler shedding, Gruber ( op. cit. ) says that upon loss of the antler the vessels closest to the shedding zone contract as soon as their contents have flowed out to produce a scab. Beneath this scab, a vascularized syncytial mesen- chymal tissue grows out of the marrow cavities of the pedicle. Wound healing and antler regeneration take place subsequently. Confirming and extending many of the earlier observations, Wis- locki ( 1942 ) mentioned the existence of a layer of fibrocellular tissue on a Virginia deer pedicle from which the antler had recently been shed. This tissue, probably equivalent to Gruber's syncytium, is located immediately beneath the blood clot and above the can- cellous trabeculae of the pedicle. In later studies, Waldo and Wis- locki (1951) and Wislocki and Waldo (1953) claimed that this Figs. 1 to 7. Representative stages in the development of the antler of the sika deer. Figure 1 shows animal's right pedicle shortly after antler was shed; no scab has formed vet, in contrast to opposite pedicle, from which antler was detached a day or two earlier. Fig. 8. Magnified view of base of shed antler illustrating (top to bottom) the burr, a ring of adherent epidermis, a peripheral zone of relatively smooth bone erosion, and an inner zone of delicate bony spicules. ( X 5.) Fig. 9. Longitudinal section through distal pedicle and part of antler, show- ing abrupt termination of pedicle epidermis (arrow) against the antler-pedicle bone (at left). (X 6.) 344 R. J. Goss fibrocelliilar connective tissue, with the aid of dermal connective tissue at the periphery, holds the antler to the pedicle for an unde- termined length of time between the final dissolution of the osseous antler-pedicle connections and the eventual shedding of the antler. How fibrous connective tissues can hold a heavv antler so rigidlv attached to the skull is left to the intuition of the reader. It was further contended that the blood clot forms before the antler is shed, a statement which is obviously incompatible with the usuallv im- maculate condition of the antler base after shedding. Having had the good fortune to witness the shedding of several antlers, I can testifv to the fact that bleeding does not normallv precede shedding. The fresh wound on the pedicle exposes a pink layer of tissue which within a few moments becomes obscured by blood ( Fig. 1 ) . A final misconception in the otherwise laudable investigations of Waldo and Wislocki ( 1951 ) is the assumption that shedding even- tually occurs, not as a combined result of the weight of the antler and its weakened bony attachments to the pedicle, but because of the upward pressure exerted by the growing pedicle skin on the flangelike burr around the base of the antler. It is true that the distal pedicle skin becomes somewhat tumescent prior to shedding, an indication that incipient regeneration has been initiated, but there is no reason to consider that this plays a mechanical role in detach- ment of the antler. Examinations by the author of the swollen pedi- cle skin in sika deer just prior to shedding revealed no indications of pressure being exerted on the burr above. Indeed, circumcision of the distal centimeter of skin from the right pedicles of two sika bucks on March 28 was followed in one case by loss of the operated antler in about 4 days and the contralateral one a few days later. In the other deer, both antlers were shed 3 weeks later, on April 18. It must be concluded, therefore, that the pedicle skin, which is so inti- mately associated with the histogenesis of the new antler, cannot be directly involved in the shedding mechanism. Inasmuch as the burr is sometimes absent from the very small first antlers of yearling bucks, which are nevertheless shed normally, the role of this struc- ture cannot be to act as a rim against which pressure is exerted. Rather, it has probably evolved as a protective ridge by which the pedicle skin is shielded from injury. DECIDUOUS NATURE OF DEER ANTLERS 345 Clearly there is a significant relation between shedding and re- growth of antlers, such that one never occurs without the other. To learn if renewed growth might be affected by preventing shedding, in each of three sika deer the old antler on one side was sawed off within a centimeter of its base, a hole was drilled longitudinally down into the pedicle, and a tantalum screw was tightly inserted (Fig. 13). In two of these deer both antlers were shed within a few days of each other, apparently because the screw was not driven deep enough into the pedicle. In the third deer, however, the con- trol antler was shed in less than a week, but the operated one re- mained attached for another 7 weeks. Despite this obstruction, the pedicle skin on the operated side developed into antler tissue which Fig. 13. Appearance of antler soon after it was secured to pedicle with tantalum screw. Fig. 14. Same antler approximately 1 month later. Incipient antler tissue is bulging around the sides and is beginning to put up a branch from posterior margin. Fig. 15. Photograph taken on day the old antler was finally lost (7 weeks after opposite control antler was shed) . Fig. 16. At end of summer the posterior branch had elongated considerably, while a curved outgrowth had been produced from the central area. 840 K. J. GOSS of necessity bulged laterally on all sides ( Fig. 14 ) . The attachments of the old antler to the pedicle were conconiitantK disrupted. Nevertheless, this loosened antler remained attached bv its screw, while the pedicle skin healed the wound beneath it. Prevented from growing as a direct extension of the pedicle, the new antler re- generated a straight, unbranched protuberance from its posterior margin. When the old antler was finally shed, there was revealed a flattened disc of abortive antler tissue overlving the pedicle (Fig. 15). Apparently in this case growth pressure was responsible for the eventual detachment of the old antler, which in fact had been physiologicallv shed some 7 weeks before. During the rest of the summer, the posterior branch grew almost as long as the opposite control antler, while another twisted outgrowth was produced be- latedly from the center of the pedicle (Fig. 16). Thus, shedding of the old antler is not prerequisite to regenera- tion of the new one. Rather, its detachment coincides with the in- itiation of renewed growth. Indeed, one never occurs without the other, except in cases where replacement of the antler may be secondarilv delayed after healing of the pedicle, as in older males of certain species that shed their antlers in the autumn but do not resume growth until spring. Whatever may be the nature of the regulating mechanism(s), it is not improbable that shedding and regrowth are simultaneous reactions to a common stimulus. Control of Antler Shedding Influence of Environmental Factors The seasonal nature of antler shedding strongly implicates the operation of one or more environmental factors as the stimulating agent(s). Mediated through secondarv phvsiological mechanisms, such influences are probably also responsible for many other annual occurrences in deer, e.g., molting, color changes, migrations, and the various events associated with the reproductive cycle. Though each species of deer sheds its antlers at a characteristic season, the variations among different kinds of deer are considerable even in the same geographical area. For example, in the north temperate regions, shedding may occur in the fall, winter, or spring. In adult DECIDUOUS NATURE OF DEER ANTLERS 347 sika deer (Cervus nippon) and fallow deer {Dama dama) shedding most often occurs in May, whereas elk and red deer {Cervus cana- densis and C. elaphus) usually cast their antlers in March. Winter is the time when the moose {Alces aloes), Virginia deer (Odocoi- leus virginianus), and black-tailed deer (O. coUnnbianus) shed their antlers. Still other species lose their antlers in the autumn soon after the rutting season. The European roe deer (Capreolus ca- preolus) sheds in December, as do mature male reindeer and caribou (Rangifer tarandus), whereas Pere Dayid's deer (Elaphurus davi- dianus) drops its antlers in Noyember.* The pronghorn antelope (Antilocapra americana) likewise sheds its horn sheaths in Noyem- ber. Some species of deer natiye to neotropical regions also shed their antlers at regular seasons. In India, the Cashmere stag {Cervus cashmirianus) generally sheds in March (Blanford, 1888-1891), whereas the hog deer {Hijehphus porcinus) (Blanford, 1888-1891) and barasingha (C. duvauceJi) (Brander, 1923) drop their antlers in April, though yariations from this norm are not uncommon (Mohr, 1932). The Indian muntjak {Cervulus muntpk) casts its antlers in May (Lydekker, 1898). Other species may vary according to habitat. Eld's deer {Cervus eJdi) is said to lose its antlers in June in Manipur and in September in Lower Burma (Blanford, 1888- 1891). The sambar (Cervus unicolor) generally sheds in March on the Indian peninsula and a month later in the Himalayan region, but individuals ma\ depart markedly from the average (Blanford, 1888-1891). In all deer there is some diyergence between different individuals of the same age and species, and even in the same individual in different years. This variation may become exaggerated among some species of deer native to tropical regions. Thus, in a given locality the male population may represent all stages of antler growth at a given time. The chital {Cervus axis) of India, for example, sheds its antlers at any season of the year, as has been noted by Blanford * According to certain records (Lydekker, 1898; Pocock, 1923; Bedford, 1949, 1952; Wood Jones, 1951) stags have occasionally grown two sets of antlers a year. The summer antlers were shed in the fall, whereupon a smaller set was grown in December, to be sh?d later in the winter. This phenomenon has not been observed in recent years, 3-18 R. J. GOSS (1888-1891), Lvdekker (1898), and Brander (1923). In Ceylon, Phillips (1927-1928) has recorded that these deer also exhibit ir- regular antler cycles, but that in Southern and Uva provinces 75 per cent of them shed their antlers in April and May. Where the shedding dates are not seasonally synchronized, the other aspects of the antler cycle, as well as the period of rut and birth, are similarly irregular (cf. Zuckerman, 1953). Yet despite the lack of uniformity^ within a tropical population, there is a strong tendency for individuals to grow successive sets of antlers at approximately 12-month intervals. According to the studies of Valera (1955) on the Philippine deer ( Riisa sp. ) , antler shedding by different individ- uals has been observed at a wide variety of times of year. It is note- worthy that these males nevertheless adhered to annual antler cycles, while the does underwent repeated estrus, sometimes as frequently as every 15 days. In agreement with this is the informa- tion generously supplied to the author by Sr. Tomas Blohm ( 1962- 1963) of Caracas, Venezuela, whose careful records of the antler growth cycle in a captive Venezuelan deer indicate that antlers are normally replaced at yearly intervals. In this particular specimen {Odocoileus gijmnotis), successive sets of antlers were shed on De- cember 8, 1958, December 1, 1959, December 19, 1960, and January 24, 1962. (Velvet was rubbed off in March or April of each year.) The most recent shedding, however, occurred precociously on July 29, 1962, apparently induced by disease (gastroenteritis). Despite this irregularity, the deer did not start to grow new antlers until its usual time of vear in December. It then shed the velvet in April 1963, indicating that the basic annual rhythm had not been dis- turbed. In temperate species of deer the breeding period is seasonally controlled and coincides in the male with the maturation of the antlers, an occurrence involving the shedding of the velvet, which is induced by the increased secretion of testosterone characteristic of the onset of rut. Should a similar situation prevail among tropical deer, the polyestrous female could mate with only that fraction of the available male population bearing mature antlers. That this is not the case, however, is indicated by Mohr's ( 1932 ) account of the hog deer (Hyelaphus porcinus) in the Hamburg Zoo: "At first sight DECIDUOUS NATURE OF^DEER ANTLERS 349 it may seem astonishing that births occur every month, and hence copulation as well, although the bucks have rubbed-off antlers only for a comparatively brief period of the year, and the last of them shed theirs only when the first begin to rub the velvet ofi^ in June. As hog deer live in pairs, and the distribution of the sexes is approxi- mately equal, the buck dees not have to do much battling to possess its female. Its antlers are not as necessary to it as a weapon as in the larger species, and it willingly and fertilely copulates even when in velvet. The same is true of Axis ( and Melanaxis ) , where the females are likewise regularlv in rut." The validity of this interesting aspect of the problem has been verified by histological examination of testes from 4 Venezuelan deer {Odocoileus gymnotis) generously made available to the author by Dr. Pedro Trebbau, Director of the Jardin Zoologico in Caracas. Mature sperm were present in abundance regardless of whether or not the velvet had been shed from the antlers, a condition which contrasts markedly with the seasonal variations in spermatogenesis that were observed by Wislocki (1949) in Odocoileus virginianus inhabiting temperate North America. The data presented by Cabrera and Yepes (1940) indicate that in South America the seasons of shedding are increasingly irregular in proportion to the proximity of habitat to the equator. The hue- mul, Hippocamehis bisulcus, which inhabits the Andes Mountains of Argentina and Chile, sheds its antlers in the winter and grows new ones in the spring. ( Gigoux ( 1929 ) , however, claims that antlers are shed in December (i.e. late spring) in this species.) The pampas deer (Ozotocerus bezoarticus) , which lives between 5° and 41° south latitude, sheds its antlers at variable times but most frequently in May (i.e. late autumn). The swamp deer {Blastocerus dichotomns) inhabiting southern Brazil and adjacent countries does not shed at any fixed season. Mazama rufa, the red brocket, inhabits the near-equatorial regions of Brazil, Venezuela, Colombia, and the Guianas. Accordingly, it sheds its antlers at any time of year. In Venezuela, data gathered bv the author on Odocoileus gymnotis indicate that antler cycles are probably asynchronous among the members of local populations. Comparable infonnation regarding other South American species (e.g., Hippocamelus antisensis and 350 R. J. GOSS Pudella mephistophelis of the mountains of Ecuador and Peru, and Fiidii piuhi of the Chilean Andes) is unfortunately not axailable. At the equator the lengths of da\ and night are essentially un- altered and equal throughout the year, and the annual fluctuations for some distance on either side of the equator are too insignificant to be of biological importance. Therefore, it is not possible for deer to exhibit annual cycles governed by seasonal variations in dav length. Presumably this explains why the members of some tropical species of deer sometimes retain their antlers for periods longer than one year. The earliest account of such an interesting phenom- enon was by Forsyth (1889), who wrote (pp. 234-235), "I have taken much pains to assure myself of a fact, of which I am now per- fectly convinced, namely, that neither in the case of the sambar nor the spotted deer . . . are the antlers regularly shed every vear in the Central Indian forests." Indeed, individual stags were observed not to shed their antlers for successive years in the same localitv. With regard to the sambar and the chital, or axis deer, of Ceylon, Phillips ( 1927-1928 ) reported that the antlers are renewed annu- ally (though not in unison) for the first few vears of life, but there- after they are shed at variable seasons and may occasionally' per- sist for several years without being replaced. Similarly, Cabrera and Yepes (1940) reported that in the red brocket of tropical South America the antlers may sometimes last more than a year. Finally, Pocock ( 1912 ) mentions a South American deer identified as Mazama bricenii, which was observed in captivity in England to shed its antlers in April of 1908 and 1909 but not again until May 1911, a period of 25 months having elapsed. Thus, there appears to be sufficient evidence to support the contention that the members of some species of tropical deer mav retain their antlers for unusuallv prolonged periods of time, though the\' also possess a strong in- clination to shed their antlers annuallv. In the latter cases, however, one wonders how an annual rhythm might be established and maintained in the absence of seasonal diurnal variations. Although it seems doubtful whether animals could recognize and respond to seasonal modifications in the declination of the sun, such meteoro- logical conditions as rainy versus drv seasons suggest themselves as an obvious answer, but experimental proof of this is lacking. Alter- DPX'IDUOUS NATURE OF DEER ANTLERS 351 natively, annual cycles of antler growth and reproduction in the ab- sence of environmental stimuli could represent an atavistic reflec- tion of temperate progenitors. Whatever the explanation, there is reason to believe that when births are distributed throughout the vear, the date at which a buck sheds his antlers may bear a regular relation to the time when he happened to be born. Lvdekker (1898), noting the persistence of irregular antler and reproductive cycles among chitals (Cervus axis) living in England, reported that when birth occurred in December, subsequent antler shedding usu- ally took place each October. Similarly, a deer born in July sheds its antlers in May or June. In the Philippine Rusa, Valera (1955) has recorded that fawns begin to grow their first antlers at about one vear of age, implying that subsequent shedding may occur a few months before each succeeding birth anniversary. Unhappily, avail- able data on these and other related phenomena among tropical deer are fragmentarv and too often dependent on hearsay. Accurate information on the life histories of deer native to equatorial regions is, therefore, urgently needed if the relation of antler growth and shedding to the reproductive cycle is to be more clearly understood. Sufficient reliable evidence along these lines may provide clues as to which endocrinological factors might best be subjected to experi- mental investigations in attempts to determine how antler loss and regeneration are induced. In view of the foregoing comparisons between deer inhabiting different latitudes, it would be expected that antler growth cycles (as well as other periodicities, e.g. molting, reproduction) would be responsive to annual variations in day length and that the degree of reaction might be proportional to the amplitude of these seasonal fluctuations. In tropical regions, where diurnal alterations are mini- mal, deer species tend to be verv irregular in the dates at which their antlers are shed. That this is due to an inbred inability to re- spond to day-length changes, and not to a lack of environmental stimulation, has been demonstrated by the persistence of irregular breeding and antler cvcles in such Indian deer as the chital and the sambar after prolonged residence in England (Lydekker, 1898; Py- craft, 1914; Bedford and Marshall, 1942). Unfortunately, I know of no records of the reverse situation, namely, the transfer of temperate 352 R- J- Goss species to tropical zones. It would be interesting to learn if such deer and their offspring would continue on an annual schedule coincident with that of their original habitats in the near absence of seasonal variations in the environment. Attempts have been made, however, to transfer various species of deer from northern temperate zones to New Ziealand (Donne, 1924; Marshall, 1936, 1937). Following such transequatorial shifts of red deer, fallow deer, moose, wapiti, and Virginia deer, the antler and reproductive cycles readjusted to the seasonal reversal. Clearly, these animals shed their antlers in response to the seasonal changes to which thev are directly exposed. Indeed, it is possible to change the antler growth cycles according to artificially altered day lengths, as has been proved by Jaczewski ( 1954 ) . In these classic experi- ments, red deer which had already begun to grow their antlers were confined in a darkened building from 4 p.m. until 8 a.m. daily. Sub- jected to suddenly decreased day length from late March or early April until June or July, these animals completed the development of their antlers, shed their velvet, and manifested precocious rutting behavior. Several weeks after they were returned to the normal summertime day lengths, the antlers were shed and new outgrowths were produced during what remained of the summer. The velvet was next shed in September, but antlers were retained until the usual shedding dates the following spring. Thus, it has been possible to induce the formation of two sets of antlers in a single year by the appropriate manipulation of day lengths. Decreases in day length accelerate maturation of the antlers and hasten the onset of rut, whereas increased day lengths stimulate shedding of the antlers in the red deer. Experiments by French et al. (1960), however, in which Virginia deer were exposed to artificial 16-hour days and 8- hour nights starting in October, failed to have marked effects on antler growth cycles. Under these conditions, antlers were shed only 2 to 3 weeks earlier than controls despite sustained exposure to prolonged day lengths for li years. The fact that Virginia deer shed their antlers in the winter when the days are short (whereas red deer shed in the spring) may account for the relatively poor response in the former species to increased lengths of daylight. Nevertheless, investigations such as these tend to indicate that DECIDUOUS NATURE OF DEER ANTLERS 353 variations in the daily amount of light play a greater role than does the absolute duration of daylight in regulating antler and reproduc- tive periodicities. Influences of Internal Factors The control of antler shedding is also affected by certain second- ary factors, not the least of which is age. Young bucks bearing their first set of antlers, which are usually unbranched spikes, invariably shed later than do older members of their species. The difference may be rather small, as in sika and fallow deer, in which the year- lings shed their antlers in late May or early June whereas mature animals do so usually in late April or early May. In other species, there is a somewhat greater age difference. Young roe deer usually lose their antlers in January or February, whereas mature ones shed in late November or December. Very old bucks are reported to have shed in late October or early November. In the moose, year-old males drop their antlers in May or June, young adults shed in February or March, fully grown bulls usually lose their antlers in late December or January, and old ones may shed them as early as November. A similar pattern is exhibited among reindeer and caribou, the young males losing their antlers about April and the oldest ones shedding them in late October or early November, with intermediate ages in between. Initiation of new antler growth is deferred until spring in most species of deer regardless of whether the previous antlers were shed in the autumn, winter, or spring. Nevertheless, older animals begin to grow new antlers earlier than younger ones. Thus, the larger the antlers, the longer is the growing season in which they can develop. In reindeer and caribou, antler shedding is influenced by sex, for females as well as males carry antlers. As mentioned above, adult males shed their antlers in fall or winter. Calves of both sexes, how- ever, do not shed them until spring. In the adult female the antlers are likewise retained until spring, eventually to be shed either just before, or more commonly soon after, parturition. The temporal correlation between these two events is so marked as to suggest a causal relationship. Indeed, it has been noted by Kelsall (1962), Lent (1962), and McEwan (1962) that barren female caribou 8o4 R. J. GOSS usuallv shed their antlers before fertile ones do, i.e., in the winter or early spring. Therefore, the condition of pregnancy appears to delay the loss of antlers, perhaps because of the heightened produc- tion of sex hormones. As will be described below, shedding of ant- lers in other species of deer can be postponed bv injections of either testosterone or estrogen. There is good reason to believe that the sequence cf events lead- ing from day-length changes to antler shedding is mediated by hormones. This is suggested bv such correlations as the relation between shedding and rut. In older and presumably more virile males, antlers are shed sooner after the breeding season than in vounger animals. In the female reindeer and caribou, one would suspect that antler shedding is stimulated by endocrine factors as- sociated with parturition or lactation. Wislocki et al. (1947) sug- gested that the male counterpart of the lactogenic hormone might logically be involved in antler growth regulation. The hormonal basis of antler shedding has been most clearly established, however, by investigations of the effects of castration. The profound consequences of castration have been studied bv nu- merous investigators, notablv Gaskoin (1856), Caton (1877), Fowler (1894), Rorig (1907), Tand'ler and Grosz (1913), and Zawadowsky ( 1926), as well as more recent investigators to be mentioned below. This operation, when performed on fawns prior to the initiation of antler development, precludes antler formation altogether. Its effect on mature antler-bearing animals, however, depends upon the con- dition of the antlers at the time of the operation. Castration of bucks with growing antlers inhibits shedding of the velvet and results in the permanent retention of viable antlers. But when a deer is castrated after the velvet has been rubbed off and the antlers are burnished and dead, shedding occurs approximatelv a month later and new antlers are grown which remain in velvet permanently thereafter. Such castrate antlers continue to grow each succeeding spring and summer, adding to what has previously been formed or replacing what mav have been lost by necrosis after winter freezing. In this way, rather bizarre headpieces may develop. In some species, such as the sika and fallow deer, the antlers may become crooked outgrowths lacking the normal number of branches. A Virginia deer DECIDUOUS NATURE OF DEER ANTLERS 355 which was protected from freezing for a number of \ears de\'eloped numerous accessory branches leading to the production of massi\'e bouquets of antlers (Wislocki et ah, 1947). An axis deer which was observed for 5 years following castration (Bullier, 1948) exhibited increasingly abnormal antler morphology as frozen tines were lost and regrown. These and other species of deer tend also to develop unusually thickened bases, from which varying numbers of tuberosi- ties may grow. Excessive thickening of antler branches may lead to the production of "cactus bucks," as described b\- Mearns ( 1907 ) in the mule deer {Odocoileus hemionus) . Most remarkable of all, how- ever, are the tumorlike masses into which the antlers of castrated roe bucks develop (Olt, 1927; Blauel, 1935, 1936; Bickel, 1936; Kleesiek, 1953). These amorphous outgrowths, known as peruke antlers, may grow down over the animal's head like a wig, obstruct- ing the vision and ultimately resulting in death from necrosis and infections. Such abnormalities may be attributed in part to incom- plete ossification in the absence of adequate amounts of testosterone. The pronghorn antelope, incidentally, responds to castration in a comparable manner by failing to shed the horn sheaths. Thus the successive annual increments of horn remain attached seriatim in an' abnormally curved accumulation of sheaths (Pocock, 1905). In contrast to the dramatic effects of castration on deer of the usual kinds, reindeer and caribou antlers appear to be considerably less dependent on sex hormones, a condition probablv correlated with the bisexual occurrence of antlers in these animals. As in other kinds of deer, castration of males after the velvet has peeled elf results in shedding of antlers in 2 to 3 weeks (Hadwen and Palmer, 1922). Males castrated while their antlers are in velvet, however, retain their antlers in this condition until spring, when the skin is lost and the antlers are shed (Tandler, 1910; Tandler and Grosz, 1913; Hadwen and Palmer, 1922; Fisher, 1939). Antlers are grown annually thereafter, although they are heavier, less calcified, and permanently in velvet, and are shed later than normal. Tandler and Grosz (1913) reported that a spayed female reindeer reacted in this same manner. These deer are thus capable of annual renewal of antlers even in the absence of gonads, but how the velvet can be lost in the presumed absence of testosterone remains to be ex- 356 R. J. Goss plained. In other deer, the velvet cannot normallv be shed unless testosterone is present, nor are the antlers lost and replaced unless the velvet has been shed. The premature loss of burnished antlers brought about by castra- tion is unquestionablv the result of reduced testosterone. Not only are unilateral castration and cryptorchidism ineffectual (Zawadow- sky, 1926; Jaczewski, 1952), but, according to experiments con- ducted by the author, replacement therapy prevents the shedding of antlers following castration. In these investigations,* 4 deer were castrated in the autumn and injected at operation with 500 mg of testosterone phenylacetate (Perandren)^ in microcrvstalline aque- ous suspension in a concentration of 50 mg per ml. Thereafter for 3 months each deer was injected intramuscularly with 100 mg testos- terone twice a week, followed bv three final injections at weekly intervals, the last administered 112 days after the beginning of the experiment. Excluding one animal that died accidentally after 84 days of treatment, the remaining 3 deer shed their antlers an average of 151 days after castration, or 39 days after the last injec- tion (Table I). Another 4 deer were similarly castrated, but given 25 mg (250,000 lU) of estrogen (Theelin)^ in aqueous suspension (5 mg/ml) at operation, and 5 mg of estrogen thereafter according to the same regimen as above. Antler shedding was delayed an aver- age of 141.5 days after castration, or 29.5 days after the terminal injection. In comparison, 14 control animals castrated at various times between September 24 and April 20 shed their antlers an average of 7 weeks later (Table I). Therefore, both testosterone and estrogen can delay loss of antlers in castrated deer, but 3 to 7 weeks after injections of these hormones are stopped, shedding occurs as in castrates not benefiting from replacement therapy. In all cases, new antler rcHeneration occurred after the old ones were lost, regardless of the time of year. These results extend the earlier "Experiments were performed on sika bucks (Cerviis nippon) anesthetized with succinylchohne (Anectine; Burroughs-Wellcome and Co.) in doses of 0.6 to 1.2 mg per deer. Anesthetics and hormones were administered subcutaneously or intra- muscularly with automatically injecting projectile syringes shot from a Cap-chur rifle (Palmer Chemical and Equipment Co., Douglasville, Georgia). f Samples of Perandren generously supplied by Ciba Pharmaceutical, Inc. I Theelin was generously supplied by Parke, Davis and Co. TABLE I. Effects of Drugs on Antler Shedding in Castrated Deer Interval from Interval from last operation until injection until Deer Date antler shedding" antler shedding" Drug no. castrated (days) (days) Testosterone 52 11-21 163 51 53 11-21 149 37 54 11-21 141.5 Average 151 29.5 Average 39 Estrogen 57 11-21 132.5 20.5 58 11-21 148 36 59 11-21 144.5 32.5 60 11-21 141 Average 141.5 29 Average 29.5 Enovid 43 11-21 139 27 44 11-21 140 28 45 11-21 139 27 46 11-21 134 Average 138 22 Average 26 Norethynodrel 92 10-25 142 30 93 10-25 157.5 45.5 94 10-25 154 Average 151 42 Average 39 Cortisone . 99 11-29 61 Controls 26 9-24 35 19 10-7 79 28 10-7 49 30 10-7 65 24 10-29 28 51 11-21 37 56 11-21 98 61'' 11-21 112-f 32 11-26 32 29 12-31 60 41 1-23 21 35 2-18 44.5 31 3-18 14 23 4-20 8 Average 48.7 " Figures represent averages of left and right antlers. '' Animal died of unknown causes without having shed antlers. 357 358 i{. .1. Goss observations of Wislocki et al. ( 1947 ) and Waldo and W'islocki (1951), who showed that testosterone administered to normal Vir- ginia deer in winter and spring delayed the normal shedding of antlers for several months. They show further that estrogen exerts effects very similar to those of testosterone on the antlers of adult deer, and corroborate the earlier observations of Blauel (1935) that the antlers of castrated roe bucks are not shed until almost 2 months after injections of estrogen (Progvnon) cease. Castration-induced shedding mav be simplv a withdrawal s\ mp- tom based upon testosterone depletion. Alternatively, the drop in testosterone levels may be attended by a rise in the amounts of other hormones secreted, one of which might directlv stimulate osteoclastic activity at the pedicle-antler junction. Since it is well known that the secretion of gonadotropins from the pituitary increases when the target organs are removed, an experimental attempt to inhibit the secretion of these hormones was undertaken bv administering the contraceptive Enovid * to 4 castrated bucks. Injection schedules were the same as in the previously described experiments, starting with 100 mg of Enovid (50 mg/ml sesame oil) at the time of castra- tion and 50 mg per injection thereafter. As shown in Table I, this effectively delayed antler shedding for an average of 138 days, or 26 days after the final injection. However, caution is indicated in concluding that these results represent solelv the effects of reduced gonadotropin secretion, for in addition to the active ingredient, 17- ethynyl-17-hydroxy-5( 10)-estren-3-one ( norethynodrel ) , these prep- arations also contained 1.5 per cent of the 3-methyl ether of ethynyl estradiol, which would be expected to mimic the effects of estrogen as described above. To obviate this objection, 3 other castrated deer were similarly injected with purified preparations of norethynodrel * over a period of 112 days. Their antlers were retained an average of 39 days after the last injection (Table I), indicating the possibilitv that antler shedding may be stimulated by a pituitary hormone, per- haps a gonadotropin. The hypothesis advanced by Tachez}' (1956) that gonadotropin might be responsible for the shedding if not the regrowth of antlers * The author is especially grateful to G. D. Searle and Co. for their cooperation and generosity in supplying Eno\'id and purified samples of norethynodrel. DECIDUOUS NATURE OF DEER ANTLERS 359 was tested on a final group of 4 deer which were not castrated. These were given injections of 5000 lU of chorionic gonadotropin (Antni- trin-"S"; Parke, Davis and Company) according to the same sched- ule as before. One animal died accidentallv after a month, but the 3 others exhibited no signs of shedding or growth during the course of 4 months and the accumulated administration of 150,000 lU of gonadotropin per deer. Unhappily, such negative results are not necessarily meaningful, for the preparations were highly soluble proteins which might possibly have been effective had they been injected in higher doses at more frequent intervals. Nevertheless, as Wislocki ( 1943 ) has pointed out, gonadotropin secretion in nor- mal deer is minimal when shedding and initial antler growth occur. Therefore, its role would appear to be that of stimulating testoster- one secretion and spermatogenesis, as the breeding season ap- proaches, rather than influencing antler shedding several months earlier. The deciduous character of deer antlers cannot be considered in its natural perspective except as a component phase in the annual reproductive periodicity typical of most deer. Hence, the shedding of antlers nearly always occurs sometime after rut and before the season of birth. Thus, newborn animals are not jeopardized by the presence of antler-bearing adults except in the case of reindeer and caribou cows that may not drop their antlers until soon after calving. An- other interesting exception to this generalization is represented by the roe deer. Not only does this deer possess mature antlers from early spring (some 2 months before the fawns are to be born) through late fall, but it is unique in that its antlers normally develop during the winter instead of the spring and summer as in other species. Moreover, it mates in July and August but implantation of the ova is delayed until December, approximately 5 months before birth. Considering these various departures from the more common sequence of events, it would be surprising if the roe deer did con- form to the reproductive patterns typical for most other deer. More specifically, the shedding of antlers may be correlated with reduced spermatogenesis and diminished testosterone secretion such as occur after the rutting season (Wislocki, 1949; Stieve, 1950; Frankenberger, 1954). Accordingly, castration brings about pre- 360 R. J. GOSS mature loss of antlers in male deer, and parturition is accompanied by shedding in female reindeer and caribou. Common to these two phenomena is an abrupt decrease in the secretion of sex hormones, an effect which might be causally related to antler shedding. It remains to be determined how the effects of decreased testoster- one are related to antler shedding and replacement. Conceivably, simple release from the inhibiting effects of testosterone might bring about shedding. Yet experiments by the author in which 4 normal male sika deer were injected with heroic doses of testosterone pro- pionate (Oreton; Schering Corporation), amounting to 5.5 gm per deer in 6 doses over a 3- week period from September 21 to October 11, failed to induce the shedding of antlers upon abrupt cessation of injections. Thus, a relative decrease in testosterone levels from super- normal to normal is not sufficient to mimic the effects of castration on antler shedding. It is possible that a stimulating effect might be involved in antler shedding. If so, such an influence could be mediated either by nerves or by hormones. With reference to the former, the experiments of Wislocki and Singer (1946) show that denervated antlers grow abnormally ( probablv owing to injuries resulting from loss of sensory innervation) but that the velvet is lost and the antlers are shed at normal times. Waldo and Wislocki ( 1951 ) , however, mention cer- tain abnormalities in the shedding mechanism of the denervated antler. The first step in analyzing the operation of endocrine factors in regulating antler loss and regeneration is to determine what glands might be involved. With the exception of gonadectomies, no en- docrine deletion experiments have been performed on adult deer and only a few have been attempted on immature ones. Therefore, one cannot rule out the possibility that almost any gland might secrete an antler-influencing hormone. Literature on the relation of the thyroid gland to antler growth cycles is somewhat contradictory. Grafflin ( 1942 ) noted no signifi- cant seasonal changes in thyroid histology in the Virginia deer. Freundova ( 1955 ) , however, described decreased thyroid activit) characterized by low epithelium and large follicles during periods of sexual repose in the red deer. In correlation with the onset of DECIDUOUS NATURE OF DEER ANTLERS 361 spermatogenesis the follicles decreased in size, stored less colloid, and exhibited more columnar epithelium. The effects of hvpothyroid- ism have been studied by Wislocki et al. ( 1947 ) , who reported nor- mal antler development subsequent to thyroidectomy of a 2-month- old fawn. Lebedinsky (1939) noted enhanced antler growth in roe bucks which had received thyroxin injections as fawns. In adults of this species, however, Bruhin (1953) was unable to affect antler growth by oral administration of 1 mg thyroxin per day for 1 month during the antler growing period. But in vearling reindeer and fallow deer thyroxin injections induced strikingly greater antler growth than in controls. Inasmuch as antler shedding is effected by localized erosion of bone, a phenomenon well known to be under the control of the parathyroid gland, it is surprising that the parathyroids of deer have not been more thoroughly investigated. What we do know, how- ever, does not indicate a direct relationship of this gland to antler shedding. Grafflin (1942) reported the lack of histologically de- tectable annual changes in the parathyroid glands of Virginia deer. Moreover, experiments by the author involving semiweekly injec- tions of 1 gm of powdered ox parathvroid glands into each of 3 adult sika bucks for 8 weeks in the autumn failed to bring about any indications of antler shedding. There is possible evidence that the adrenal cortical hormones play roles in antler growth. Wislocki ( 1943) stated that the adrenal glands of male Virginia deer exhibit no histological alterations correlated with seasonal changes. More recently, however, Hucin (1957) has described increased secretorv activity in the zona glomerulosa dur- ing the period of antler growth in the red deer ( a correlation between mineral metabolism and mineralocorticoids ) , but only slight changes in the activities of the inner zones associated with the rutting season. It is interesting to note that Doutt and Donaldson (1959) have re- ported the unusual occurrence of a female Virginia deer bearing antlers which might have been attributable to androgens secreted by an adrenal cortical tumor. Cortisone probably does not influence antler development, for Taft et al. (1956) reported that ACTH injections exerted little or no effect on the antlers of Virginia deer. The present author has shown that 8 weekly injections of 500 mg of 36''2 * , R. J. Goss cortisone acetate into 4 castrated sika deer with antlers in velvet failed to induce shedding of the velvet (in contrast to the effective- ness of testosterone, estradiol, and norethynodrel ) . Similarly, semi- weeklv injections of 250 mg of cortisone acetate into a castrated sika buck with mature antlers failed to prevent shedding of the antlers, which were lost 61 days after operation (Table I). In view of the known effects of testosterone and estrogen on antlers, it would seem especiallv important to undertake studies of other sex steroids. Waldo and Wislocki (1951) administered pro- gesterone to a normal deer without obtaining noteworthy efiFects on subsequent antler development. In another animal lacking antlers because of prior castration as a fawn, this hormone failed to bring about antler growth ( in contrast to the efficacv of testosterone under similar circumstances). Neither did progesterone stimulate antler growth in a female deer, an effect which has been achieved with testosterone, as reported by Wislocki et al. ( 1947), Aub et al. ( 1949- 1950), and Waldo and Wislocki (1951). Progesterone has also been investigated by the present author and found incapable of inducing shedding of the velvet in 3 castrated sika deer given 8 weeklv doses of 500 mg. Partlv by default and partly by analogy with other secondary sexual characters, the possible existence of an antler-stimulating hormone of pituitary origin has often been assumed (Wislocki, 1943; Tachezy, 1956). This is supported by the reports of Taft et al. ( 1956 ) and Hall et al. ( 1960 ) that hypophysectomy of an 8-month- old Virginia deer resulted in the failure of antlers to grow and in the absence of molting during the next 13 months that the animal survived. Had this deer not yet developed pedicles, it would have been comparable to a castrated fawn, inasmuch as the lack of gonad- al development following hypophvsectomv would have precluded the development of pedicles owing to insufficient testosterone secre- tion. Growth of antlers in the adult, however, is not dependent on testosterone, as castration studies have shown. Since this animal possessed distinct, albeit small, pedicles at the time of operation in January (Taft, 1962), its subsequent inability to grow antlers was probably the direct result of hypophysectomy and not of secondary testicular atrophy. Taft et al. (1956) also indicated that massive DECIDUOUS NATURE OF DEER ANTLERS 363 doses of growth hormone were practicalh without effect on antler growth, and Blauel ( 1935) observed tliat the growing antlers of the roe deer were not influenced by administration of whole pituitary powder. In recent experiments by the author, no effects were ob- served on the autumn antlers of 2 sika deer injected with 1 gm of lyophilized, defatted l^eef pituitary twice a week for 2 months. In view of these incomplete data, the course of research in the future is inescapable. It is to approach the problem of antler shed- ding in particular, and that of the recurrent annual cycle of antler replacement in general, from a variety of directions. From the com- parative point of view, valuable insight into the relationships be- tween antler cycles and reproductive c\'cles may be gained by seeking fuller knowledge of these phenomena in such aberrant forms as reindeer and caribou, which have antlers in both sexes, and in tropical species of deer, which often exhibit verv irregidar cvcles of reproduction and antler succession. A sur\'ev of endocrine influences, by extirpation experiments as well as hormone injections, may re- veal how the various phases of the annual antler cycle are regulated. Finally, it will be important to examine more closely the actual mechanism of antler shedding. Since the earliest stages of a process are so often the most important, by learning more about the de- ciduous nature of antlers our understanding of their genesis and growth may be enriched. Summary Antler shedding is accomplished l^v osteoclastic erosion of the haversian lamellae along the line of demarcation between the dead bone of the antler and the living bone of the pedicle. It is an integral part of the sequence of histological events associated with the initia- tion of renewed antler growth. In temperate regions, male deer lose their antlers in a fixed season depending upon the species. This is determined by reactions to sea- sonal changes in dav length, as has been proved by noting responses to transequatorial shifts or exposure to experimentally altered day- to-night ratios. Tropical deer show a tendency toward asynchronous, and occasionally prolonged, antler replacement periodicities, to- 304 R. J. GOSS gether with a loss in correlation between antler and reproduction cycles. Antlers are shed earlier in older deer than in younger ones. De- layed by pregnancy, shedding coincides with the calving season in female caribou and reindeer. Castration of deer with mature, dead antlers induces precocious shedding and subsequent regrowth of antlers that remain permanently in velvet. Castration-induced antler loss is precluded by injections of testosterone, estrogen, Enovid, or norethvnodrel, but not by cortisone. In normal deer, antler shedding could not be induced by repeated injections of chorionic gonado- tropin or of pituitary or parathyroid preparations, nor by abrupt cessation of testosterone injections. Hypophysectomv, however, ap- pears to prevent antler growth. Acknowledgments. Original investigations by the author were sup- ported by a research grant (B-923) from the National histitutes of Health, United States Public Health Service. I am indebted to Mrs. Sally Hoedemaker for valuable technical assist- ance, and to Messrs. Daniel and Justin Southwick of the Southwick Animal Farm, Blackstone, Massachusetts, for their generous cooperation in the acquisition and maintenance of experimental deer and for the many cour- tesies without which my investigations could not have been so successfully pursued. References Aub, J. G., Wislocki, G. B., and Waldo, C. M. 1949-1950. The growth of the deer antler. N. Y. State Conservationist, 4, 4-5. Bedford, Duke of. 1949. The Years of Transition. Andrew Dakers Ltd., London. Bedford, Duke of. 1952. Pere David's deer. Zoo Life, 7, 47-49. Bedford, Duke of, and Marshall, F. H. A. 1942. On the incidence of the breeding season in mammals after transference to a new latitude. Proc. Roy. Soc. (London), Ser. B, ISO, 396-399. Bent, A. C. 1919. Life histories of North American diving birds. V. S. NatL Museum, Bull. No. 107, pp. 92-93. Bickel, A. 1936. Drei Periickenbocke. Wild und Hund, 42, 674-675. Blanford, W. T. 1888-1891. The Fauna of British India, including Ceylon and Burma. Mammalia. Taylor and Francis, London. Blauel, G. 1935. Beobachtungen liber die Entstehung der Periicke beim Rehbock. Endokrinologie, 15, 321-329. DECIDUOUS NATURE OF DEER ANTLERS 365 Blauel, G. 1936. Beobachtungen iiber die Entstehung der Periicke beim Rehbock. 2. Mitt. Endokrinologie, 17, 369-372. Blohm, T. 1962-1963. Personal communications. Brander, A. A. D. 1923. Wild Animals in Central India. Edward Arnold and Co., London. Bruhin, H. 1953. Zur Biologic der Stirnaufsatze bei Huftieren. Physiol. comp. Oecol, 3, 63-127. Bullier, P. 1948. Curieuse anomalie des bois chez un cerf castre {Axis axis). Mammalia, 22, 271-274. Cabrera, A., and Yepes, J. 1940. Historia natural Ediar; mamiferos siid- americanos (vida, costumbres y descripcion) . Cia argentina de edi- tores, soc. de resp. Itda, Buenos Aires. Caton, J. D. 1875. On the structure and casting of the antlers of the deer. Am. Naturalist, 8, 348-353. Caton, J. D. 1877. The Antelope and Deer of America. Forest and Stream Publishing Co., New York. Coues, E. 1872. Kei/ to North American Birds. Naturalists' Agency, Salem, Mass. Donne, T. E. 1924. The Game Animals of New Zealand; an Account of Their Introduction, Acclimatization, and Development. John Murray, London. Doutt, J. K., and Donaldson, J. C. 1959. An andered doe with possible masculinizing tumor. /. ManDuaL, 40, 230-236. Fisher, C. 1939. The nomads of arctic Lapland. Natl. Geogr. Mag., 76, 641-676. Forbush, E. H. 1936. Tufted Puffin. In Birds of America (T. G. Pearson, editor), pp. 17-18. Garden City Publishing Co., Garden City, N. Y. Forsyth, J. 1889. The Highlands ^of Central India. Chapman and Hall Ltd., London. Fowler, G. H. 1894. Notes on some specimens of antlers of the fallow deer, showing continuous variation, and the effects of total or partial castra- tion. Proc. Zool. Soc. London, 1894, 485-494. Frankenberger, Z. 1954. Interstitialni bunky jelena (Cervus elaphus L.). (Interstitial cells in the deer {Cervus elaphus L.).) Cesk. Morfol, 2, 36-41. French, C. E., McEwen, L. C, Magruder, N. D., Rader, T., Long, T. A., and Swift, R. W. 1960. Responses of white-tailed bucks to added artificial light. /. Mammal, 41, 23-29. Freundova, D. 1955. Stitna zlaza jelena {Cervus elaphus L.) a jeji souvis- lost s pohlavnim cycklem. (Thyroid gland of the deer {Cervus ela- phus L.) and its relationship to the sex cycle.) Cesk. Moiiol, 3, 205-211. Gaskoin, J. S. 1856. On some defects in the growth of the antlers and S60 R. J. GOSS some results of castration in the Cervidae. Proc. Zool. Soc. London, 24, 151-159. Gigoux, E. E. 1929. El huemiil. Reo. chilcna hist, nut., 33, 573-582. Grafflin, A. L. 1942. A study of the thyroid gland in specimens of Virginia deer taken at intervals throughout the vear. /. MoiplioL, 70, 21-40. Gruber, G. B. 1937. Morphobiologische Untersuchungen am Cerviden- Geweih. Werden, Wechsel und W'esen des Rehgehorns. Nachr. Ges. Wiss. Gottingen, N. F., Matli.-phijsik. KL, Fachgruppc VL Biol., 3, 9-63. Gruber, G. B. 1952(/. Uber das Wesen den Cerviden-Geweihe. Dent. tiercirzfL Wochschr., 5.9, 225-228, 241-243. Gruber, G. B. 1952Z;. Studienergebnisse am Geweih des Cervus capieolus. Zentr. all gem. Pathol, ii. pathol. Anat., 88, 336-345. Hadwen, S., and Palmer, L. J. 1922. Reindeer in Alaska. U. S. Depf. Agr. Bull. No. 1089. Hall, T. C., Ganong, W. F., Taft, E. B., and Aub, J. C. 1960. Endocrine control of deer antler growth. First Intern. Congr. Endocrinol.., Copen- hagen, p. 525. Hucin, B. 1957. Histologicke zmeny v kuie nadledvin v prubehu rocniho cyklu u jelena. (Histological changes in the adrenal cortex during annual cycles in the deer. ) Cesk. Morfol., 5, 376-387. Jaczewski, Z. 1952. Wplyw jednostronnej kastracji na wzrost rogow u sarny (Capreoliis caprcohish.) i jelenia (Cervus clapluis h.). (Effect of unilateral castration on antler growth in roe deer (Capreolus capreolus L. ) and common stag (Cervus elaphus L.).) Fragmenta Faun. Mus. Zool. Polon., 6, 199-205. Jaczewski, Z. 1954. \\^plyw oswietlenia dziennego na poroze jelenia, Cervus elaphus L. (The effect of changes in length of daylight on the growth of antlers in the deer, Cervus elaphus L.) Folia Biol. (Warsaw), 2, 133-143. Kelsall, J. P. 1962. Personal communication. Kleesiek, C. 1953. Hormonbehandlung eines Periickenbocks. Wild und Hund, 56, 117. Kolliker, A. 1873. Die normale Resorption des Knochengewebes und ihre Bedeutung fiir die Entstehung der tijpischen Knochenformen. Leip- zig. Lebedinsky, N. G. 1939. Beschleunigung der Geweihmetamorphose beim Reh (Capreolus capreolus L.) durch das Schilddriisenhormon. Acta Biol. Latvica, 9, 125-132. Lent, P. C. 1962. Personal communication. Lieberkiihn, N. 1861. Ueber den Abfall der Geweihe und seine Ahnlichkeit mit dem cariosen Process. Arch. Anat., PhijsioL u. wiss. Med., 4, 748- 759. DECIDUOUS NATURE OF DEER ANTLERS 367 Lockley, R. M. 1953. Puffins. J. M. Dent and Sons Ltd., London. Lydekker, R. 1898. The Deer of All Lands. A History of the Fcnnihj Cervidae Living and Exiinct. Rowland Ward Ltd., London. Macewen, W. 1920. The Growth and ?)heddino of fjjc Antler of tJie Deer. The Histohigieal Phenomena and Their Rehiiion to the Growtli of Bone. Maclehose, Jackson and Co., Glasgow. Marshall, F. H. A. 1936. Sexual periodicity and the causes which deter- mine it. The Croonian Leeture; Phil. Trans. Roij. Soc. (London), Ser. B, 226, 42.3-456. Marshall, F. H. A. 1937. On the change over in the oestrons cycle in ani- mals after transference across the e((uator, with further observations on the incidence of the breeding seasons and the factors controlling sexual periodicity. Proe. Roij. Soc. (London), Ser. B, 122, 413-428. McEwan, E. H. 1962. Personal communication. Mearns, E. A. 1907. Mammals of the Mexican boundary of the United States. [/. S. Natl. Museum, Bull. No. 56. Mohr, E. 1932. Materialien iiber die Hirschzuchten des ehemaligen Ham- burger Zoo. Der Zool. Garten, N. F., 5, 3-15. Noback, C. V. 1932. The deciduous horns of the pronghorn antelope, Antilocapra americana. Btdl. N. Y. Zool. Soe., 35, 195-207. Olt, A. 1927. Die Periicke der Cerviden und das Karzinom. Deut. Tierdrztl. Woehschr., 35, 131-133. Phillips, W. W. A. 1927-1928. Guide to the mammals of Ceylon. Part VI. , Ungulata. Ceylon J. Sci., B14, 1-50. Pocock, R. I. 1905. The effects of castration on the horns of a prongbuck (Antiloeapra americana) . Proe. Zool. Soc. London, 1905, 191-197. Pocock, R. I. 1912. On antler-growth in the Cervidae, with special refer- ences to Elaphurus and Odocoileus (Dorcelaphus) . Proe. Zool. Soc. London, 1912, 773-783. Pocock, R. I. 1923. On the external characters of Elaphurus, Lli/droptes, Pudu and other Cervidae. Proe. Zool. Soc. London, 1923, 181-207. Pvcraft, W. P. 1914. The Courtship of Animals. Henry Holt and Co., New York. Rawles, M. E. 1960. The integumentary system. In Biology and Compara- tive Physiology of Birds (A. J. Marshall, editor), pp. 189-240. Aca- demic Press, Inc., New York. Ries, J. 1948. Das Hirschgeweih als Aufspeicherungsort fiir Sexualhor- mone. Schweiz. Rundschau Med., Praxis, 37, 368-370. Rorig, A. 1907. Gestaltende Correlationen zwischen abnormer Korperkon- stitution der Cerviden und Geweihbildung derselben. Arch. Entwick- lungsmech. Organ., 23, 1-150. ScherbatoflF, N. 1933. Der Altai-Maral, ein eigenartiges, aber nicht un- wichtiges Haustier Asiens. Der Zool. Garten, N. F., 6, 44-47. 368 R- J- Goss Skinner, M. P. 1922. The prong-horn. /. Mammal, 3, 82-105. Stieve, H. 1950. Anatomische Untersuchungen iiber die Fortpflanzungs- tatigkeit des europaischen Rehes (Capreolus capreolus capreolus L.). Z. mikroskop.-anat. Forsch., 55, 427-530. Tachezy, R. 1956. Uber den Einfluss der Sexualhormone auf das Geweih- wachstum der Cerviden. Saugetierkimdl. Mitt., 4, 103-112. Taft, E. B. 1962. Personal communication. Taft, E. B., Hall, T. C, and Aub, J. C. 1956. The growth of the deer antler (2). N. Y. State Conservationist, 10, 4-5. Tandler, J. 1910. Bericht iiber den Einfluss der Geshlechtsdriisen auf die Geweihbildung bei Rentieren. Anz. kaiserl. Akad. \i iss. Wien, Math.- naturw. Kl, 47, 252-257. Tandler, J., and Grosz, S. 1913. Die biologischen Griindlagen der sekun- ddren GescJdechtscharaktere. Springer, Berlin. Valera, R. B. 1955. Observations on the breeding habits, the shedding and the development of antlers of the Philippine deer (Riisa sp.). Philip- pine }. Forestry, 11, 249-257. Waldo, C. M., and Wislocki, G. B. 1951. Observations on the shedding of the antlers of Virginia deer {Odocoileiis virginianus boreaJis). Am. J. Anat., 88, 351-396. Wislocki, G. B. 1942. Studies on the growth of deer antlers. I. On the structure and histogenesis of the antlers of the Virginia deer ( Odocoi- leiis virginianus borealis). Am. J. Anat., 71, 371-416. Wislocki, G. B. 1943. Studies on the growth of deer antlers. II. Seasonal changes in the male reproductive tract of the Virginia deer ( Odocoi- leus virginianus borealis) with a discussion of the factors controlling the antler-gonad periodicity. In Essays in Biology, in Honor of Herbert M. Evans, pp. 631-653. University of CaHfornia Press, Berkeley, Galif. Wislocki, G. B. 1949. Seasonal changes in the testes, epididymes and seminal vesicles of deer investigated by histochemical methods. En- docrinology, 44, 167-189. Wislocki, G. B.; Aub, J. C., and Waldo, G. M. 1947. The effects of gonadec- tomy and the administration of testosterone propionate on the growth of antlers in male and female deer. Endocrinology, 40, 202-224. Wislocki, G. B., and Singer, M. 1946. The occurrence and function of nerves in the growing antlers of deer. /. Comp. Neurol, 85, 1-19. Wislocki, G. B., and Waldo, G. M. 1953. Further observations on the his- tological changes associated with the shedding of the antlers of the white-tailed deer (Odocoileiis virginianus borealis). A^uif. Record, 117, 353-376. Wood Jones, F. 1951. A contribution to the history and anatomy of Pere David's deer (Elaphurus davidianus). Proc. Zool. Soc. London, 121, 319. DECIDUOUS NATURE OF DEER ANTLERS 369 Wyman, J. 1859-1861. Account of some observations on the shedding of the antlers of the American red deer. Proc. Boston Soc. Nat. Hist., 7, 167-168. Zawadowsky, M. M. 1926. Bilateral and unilateral castration in Cervus clama and Cervus elaphiis. Trans. Lab. Exptl. Biol. Zoo-Park Moscow, 1, 18-48. (In Russian; English summary.) Zuckerman, S. 1953. The breeding seasons of mammals in captivity. Proc. Zool. Soc. London, 122, 827-950. 14 Internal Remodeling of Compact Bone FRANKLIN C. McLEAN. Department of I'hysiology. I'niversity of Chicago, Chicago, Illinois ROBERT E. ROWLAND,* Radiological Physics Division, Aigonne Na- tional Laboratory, Argonne, Illinois BONES increase in length by the functioning of the growth appara- tus, inchiding the epiphyseal cartilage plate and the metaphysis of the long bones. They increase in diameter bv apposition of new periosteal bone, while the marrow cavity is being enlarged by re- sorption at the endosteal surface. In addition to the changes in size and shape of the bones, as a result of the remodeling incident to growth, there is continuous internal remodeling throughout the life of the individual; this serves an important physiologic function, essential to homeostatic control of the calcium level in the blood plasma, and therefore to life itself. In compact bone there is first the formation of absorption cavities, described by Tomes and De Morgan in 1853. The appearance of these cavities is associated with the presence of osteoclasts, and the cavities are extended until they assume the form of tunnels. Because of certain similarities to rock boring by biologic organisms, as de- scribed elsewhere in this volume (chapters 1, 2, and 3), and be- cause of the possible significance of absorption or resorption cavities in the evolution of the skeleton, it seems desirable to treat the sub- ° Present address: Department of Radiation Biology, Sehool of Medicine and Dentistry, University of Rochester, Rochester, New York. 371 372 F. C. MCLEAN AND R. E. ROWLAND ject of internal remodeling of compact bone as a topic in the com- parative biology of calcified tissues. Neither rock boring nor the excavation of tunnels in compact bone should be regarded solely as a destructive process; both serve to per- form physiologic functions, and can only be understood by considera- tion of the cycles in which they participate. The rock borers and the boring gastropods engage in the removal of calcified tissues in search of either shelter or food, or both. The tunneling in compact bone, by osteoclastic resorption, is unique in that it is followed by reconstruction of new osteons, or haversian systems, which provide a continuing supply of new and reactive bone, meeting the metabolic needs of the organism while leading to maturation and formation of structural bone. The internal structure of compact bone has been well described for more than a century; its haversian systems and canals, providing for blood vessels, and its lacunae, housing the osteocytes, with inter- connecting canalicules, are easily demonstrable in ordinary histologic sections. Such sections, however, do not reveal any contrast in the density of the numerous osteons and interstitial lamellae that make up the mass of bone. It has also been known that there is an ex- change of ions between the bones and the circulating fluids of the body, but not until Chievitz and Hevesy, in 1935, exploited radio- active phosphorus, P^-, in the study of bone was the rapiditv of this exchange recognized, and its physiologic significance is still under study, largely by means of tracer elements. With the advent of microradiography, and especially when it was employed together with autoradiography, it was found that the bone mineral, instead of being distributed homogeneously in compact bone, exhibits wide variation in the density of the osteons. Figure 1 shows a low-power microradiograph of a transverse section through the shaft of the tibia of adult man, in which every gradation from beginning calcification of new osteons to maximum density of ma- ture and fully mineralized osteons appears. Figure 2 illustrates two osteons from the radius of a dog, the same section being viewed at high magnification by four different methods: A, unstained his- tologic section; B, microradiograph; C, autoradiograph, recording alpha tracks from radium-226, administered 2 days before sacrifice; INTERNAL REMODELING OF COMPACT BONE 378 o *2 O fc o 374 F. C. MCLEAN AND R. E. ROWLAND C o o „ fcjD.- -^ 5 :s m .2 > m . V.'- i^ Ol o oj i O # o '^' " >- o O Td y ?^ a.' 5 2 ^ ^ t/D ■u .~ s ^ ^ qj c -G - •= .„ ^ .B ■-t; ii "^ ~ L^ 'T^ ^ - c: 2 "^ pc; ^ ?2 ■£ ii .^ 5 <: . 5 r ^ ^ g ^ J ■i; o r 5 -g _^ cc _ pC ^ -G V fcjD ^ .ti W S '^ ": y) ?^ > ° -^ I 0; o Y^ rp *^ t-'> W oi J H I ^ J ^ ci c _>^rl^ ^ '^ "c rT ? j£ o ^ tJ-.U; Mh r- 3 (U -I-' ^ 00 G ST > C^ O INTERNAL REMODELING OF COMPACT BONE 375 and D, fluorescent image with ultraviolet liglit, revealing deposition of tetracvcline administered 4 days before sacrifice. Treatment of sections of undecalcified bone by these different methods adds greatly to the information to l^e obtained from their study. The autoradiographs and the fluorescent image reveal that both radium and tetracycline are preferentialh' deposited in preosseous tissue, as l:one is laid down in concentric layers in the rebuilding of osteons. When visualized in three dimensions, the tunnel, or resorption cavity, is seen to contain blood vessels and connective tissue cells. It is excavated in the long axis of the bone, cutting through old csteons and lamellar bone, without relation to the previous structure. While resorption is in progress, the cavity is lined with osteoclasts; when rebuilding begins, these are replaced by osteoblasts. The tunnel is then filled in, from the walls toward the center, by apposi- tion of bone in successive layers or lamellae, incorporating the csteocvtes, their lacunae, and the canalicules. The formation of new lavers continues until the canal reaches its final diameter, usually approximately 20 microns; at this time the osteoblasts assume a resting form, and are no longer recognizable. Using lead as a marker, Vincent ( 1957 ) has determined the rate of formation and of maturation of new osteons. An average resorp- tion cavity in a dog takes roughly 3 weeks to form, this being the period required for tunneling or excavation. The building of the new osteon, including partial minerafization of the organic matrix, requires some 6 to 12 weeks. Primary mineralization, to about 70 per cent of the final density, occurs rapidly during and immediately after the deposition of new layers of organic matrix; completion of secondary mineralization, to maximum density, takes much longer and has been found to be incomplete for as long as 18 weeks. Frost et al. (1960), using tetracycline as a marker, found the mean osteon formation time in a 57-year-old man to be 5 weeks. The biologic half life of the osteons was calculated as 2.7 years for the femur and 8.6 years for the tilMa. In another study, also with tetracy- cline. Frost and Villanueva (1960) report that an average of 0.9 micron of new osseous tissue per da\' is formed in actively forming haversian systems in man. As to the formation of resorption cavities, little is known of the 37G F. C. MCLEAN AND R. E. ROWLAND mechanism, except that it appears to be identical with osteoclastic resorption in the removal of bone throughout the skeleton. Here the situation seems to be analogous to that in rock boring; in both instances there is speculation concerning the chemical factors in dissolution of calcified tissues; in both cases acid formation and chelation are given consideration as possible mechanisms; in neither have the factors participating in destruction of the hard tissue been elucidated. Rock boring by biologic organisms and tunnel formation bv osteoclastic resorption are sufficiently similar in their nature and their manifestations to suggest a common pathway in the evolution of hard tissues. In connection with the comparative biology of calcified tissues, Amprino and Godina ( 1953 ) reported that remodeling of compact bone in the teleost Thijnniis thynnus L. is initiated by the formation of tunnels by osteoclastic resorption. On the other hand, Norris et al. (1963), studying the sea bass Epinephelus striatiis (Bloch), observed resorption apparently in the absence of osteoclasts; primary bone was penetrated by blood vessels, forming tunnels. Acellular bone, arranged circumferentially around thin-walled blood vessels, was then laid down by osteoblastic activity, a thin sheet of osteoid appearing beneath a single layer of osteoblasts. The diameter of the lumen of the acellular osteons varied from 28 to 41 microns. There remains the important problem of physiologic regulation. At least two regulatory processes are involved: (1) regulation of initiation and formation of resorption cavities or tunnels; and (2) regulation of the transfer of calcium in such a manner as to maintain the constancy of the Ca++ concentration in the blood, in spite of the rapid turnover of the plasma calcium. Little is known concerning the regulatory control of the formation of resorption cavities. Jowsey et al. (1958) have studied the re- modeling of compact bone in parathyroidectomized dogs, and have found that tunnel formation continues after removal of the para- thyroid glands, although at a somewhat reduced rate; in any event a continuing supply of reactive bone is maintained, adequate to support life, and osteoclastic resorption continues. Since osteoclastic resorption, incident to the growth of bone, also continues in the al^- sence of the parathyroids, it appears that osteoclastic activity is not INTERNAL REMODELING OF COMPACT BONE 377 wholly dependent upon parathyroid function; a rate-determining relation between the parathyroids and osteoclastic activity seems likely. It is not known whether in the absence of the parathyroids vitamin D assists in promoting the resorption incident to remodeling; such an influence is a possibility. Once a tunnel is completed, there is a reversal in activity, and osteoblastic deposition of bone begins; there is established a con- tinuing supply of preosseous tissue, and deposition of bone mineral in the organic matrix is initiated. It is in these locations that the reactivity of the newly formed tissue is demonstrable by observation of the uptake of radioactive calcium or radium. When new tissue is laid down, gradually filling in the cavity until only enough space is left for the haversian canal, there is always a fresh surface lining the forming haversian systems, available for deposition of mineral. It is in such locations that accretion continues throughout life. Deposition of new mineral, as seen bv autoradiographs following administration of Ca^"' or Ra^-'', is concentrated in these newly formed layers of preosseous tissue, lining the haversian systems; these areas have acquired the designation hotspots, as illustrated in Fig. 3. The mineral so deposited is often referred to as exchange- able, labile, or reactive; there is as yet no direct evidence that it is from these locations that mineral leaves the bone to replace that lost in the rapid turnover of the blood calcium. In addition to the concentration in hotspots, there is diffuse deposition of the isotopes, in much lower concentration and more or less uniformly, throughout the compact bone; this has been designated as the diffuse component (Marshall et al., 1959). Of the total amount deposited in the skele- ton, following administration of a single dose of Ca**^ or Ra^^'', ap- proximately one-half is distributed in the diffuse component, while the other half is concentrated in the hotspots, occupying only a small proportion of the total space in the bones. Bauer et al. ( 1955 ) accounted for the transfer of calcium between blood and bone, in both directions, by a combination of accretion, resorption, and exchange. The mineral concentrated in the newly forming osteons, as demonstrable by the use of tracer Ca^^ or Ra"^*', is attributed to accretion. Resorption, associated with the presence of osteoclasts, is generally believed to result from an osteolytic action 378 F. c. mcl?:an and r. e. Rowland •Jjk,-. ;mm.. c INTERNAL REMODELING OF COMPACT BONE 379 of these cells upon both the mineral and the matrix of bone; it occurs mainly on the surfaces of bone and in the formation of re- sorption cavities; there is also some indication that osteocvtes mav contribute to the mobilization cf bone mineral. The diffuse com- ponent, visualized bv administration of tracer elements, is regarded by Marshall et al. (1959) as representing the portion of the mineral undergoing long-term exchange. It has been stated above that there is a rapid turnover of the calcium in the lilood; this i> so rapid that in adult man one out of four calcium ions leaves the blood everv minute ( Bauer et al., 1961 ) . The calcium can go cnly into the skeleton, and is replaced by other ions from the bones. For the minute-to-minute equilibrium betv^een blood and bone, it is believed that this transfer occurs bv ion ex- change; neither osteoclastic resorption nor long-term exchange is sufficiently rapid to account for it. The source of the calcium trans- ferred from bone to blood, by ion exchange, has not been definitelv localized, although the opinion is held that it comes from the re- centlv deposited mineral in the osteons undergoing mineralization. These osteons constitute the reactive or metabolic bone; in contrast, the fullv mineralized bone, in which deposition of new mineral oc- curs only as the diffuse component, is designated as structural bone (Vincent and Haumont, 1960). There is a pronounced difference between the amounts of stable and of labile, reactive, or exchange- able bone mineral; less than 1 per cent of the mineral is described as exchangeable (Neuman and Neuman, 1958). McLean and Urist ( 1961 ) have proposed a dual mechanism to account for homeostatic control of the calcium ion concentration in the body fluids, including the blood plasma. The function of the parathyroid glands is to monitor this concentration. This function is illustrated, in cybernetic terms, in Fig. 4. The parathyroid glands respond to changes in the Ca++ concentration in the plasma by altering the output of parathyroid hormone; the net result is that this concentration is held, in normal man, at approximately 10 mg per 100 cc. It seems certain that the effect of the parathyroid hor- mone on release of calcium from the bones is mediated through the cells of bone, although the mechanisms by which this is accomplished are not fully understood. The three cell tvpes — osteoblasts, osteo- 380 F. C. MCLEAN AND R. E. ROWLAND Error , Proportional Derivative Integral i BONE sensing ! -A Co"'"' -dCa-'Vdt -/(A Ca'-'ldt ! E=^ - " ' "^S H|^H HAKAIHYKOIU ^H GLANDS ^^^^H ARTERIAL BLOOD^^^H BONE CELLS Osteoblasts Osteocytes Osteoclasts Feedback Control signal 1 transducers transducers PARATHYROID HORMONE CALCITONIN (?) Fig. 4. Diagram of control system regulating Ca++ concentration in the blood. For explanation, see text. cytes, and osteoclasts — are potentialK' capal^le of performing this function; available evidence suggests that this is done by a com- bination of osteoclastic resorption with release of H+ from the gly- colytic cvcle of the osteocytes; there is no specific suggestion that the osteoblasts contribute to the release of calcium from bone. In the absence of the parathyroid glands there is a decrease in the concentration of calcium in the blood plasma, with 7 mg per 100 cc as a representative figure; the transfer of calcium between bone and blood continues under these conditions. The dual mecha- nism of McLean and Urist attributes this to ion transfer, not under the control of the parathyroids. Osteoclastic resorption continues, as demonstrated bv Jowsey et al. ( 1958 ) ; new haversian systems are formed; and the glycolytic cycle of the cells of bone doubtless con- tinues to be active. There must remain in bone all of the mecha- nisms concerned in ion transfer or ion exchange; what is lost is the parathyroid mechanism, which detects and corrects any deviation from the normal concentration of Ca++ in the blood plasma. One further proposal remains for consideration, i.e., the possibility that there is a change in the chemical nature of the mineral of bone, INTERNAL REMODELING OF COMPACT BONE 381 as it matures during the mineralization of osteons. The terms labile and stable bone mineral have been in use for some time, but there has not been a chemical definition of these terms. There is, however, particularly in the work of Dallemagne et al. (1961), a suggestion that as bone mineral matures it may change in chemical composi- tion, and hence in reactivity. This possibilitv may be explored, al- though any tentative conclusions must be regarded as speculative. The basic structure of the bone mineral, as now generally ac- cepted, is that of hijdroxyapatite, Caio(P04)6(OH)2. Dallemagne, however, has for some years supported the view that the mineral, as it exists in living bone, is a hydrated tricalciuni phosphate, Ca9(P04)6H2(OH)2 (alpha tricalciuni phosphate), and has believed that when this mineral is isolated from its organic matrix, it assumes the more stable form of hydroxy apatite. Posner and Perloff ( 1957 ) , however, have proposed the concept of a calcium-deficient apatite, and regard Dallemagne's hydrated calcium phosphate as a special case of such a mineral. Posner et al. (1960) propose that such a cal- cium-deficient apatite can be accounted for by hydrogen bonding. One example may be Ca-, ( PO4 ) 4 ( OH ) ( 0.,P0— H— OPOs ) . The hypothesis is here suggested that the highly reactive form of the bone mineral, subject to rapid ion exchange at the surfaces of the crystals, may be regarded as a calcium-deficient apatite, with hydrogen bonding accounting for its crystal structure, and that as this mineral matures with aging it is transformed into the more stable hydroxyapatite. Summary Internal remodeling of compact bone continues throughout the life of the individual. By providing a continuing supply of reactive bone mineral, from which calcium lost from the blood in a rapid turnover of this element is replaced, this performs a function essen- tial to life. Internal remodeling is accomplished by formation of resorption cavities, or tunnels, in the bone, as a result of osteoclastic resorption; an analogy with rock boring by biologic organisms, as described elsewhere in this volume, is suggested. 38^2 F. C. MCLEAN AND R. E. ROWLAND The resorption caxities are filled in hv formation of new bone in eoncentric layers or lamellae, insuring a constant snpplv of new or metabolic bone; this new bone takes up radioisotopes, such as Ca^' and Ra"", preferentially, leading to concentration in hotspots, de- monstrable bv autoradiography. As metabolic bone matures it loses its reactivity, and adds to the mass of structural bone. It is suggested that maturation of the bone, with a decrease in its reactivity, repre- sents a change in the chemical nature of the mineral. The reactive bone in the newly forming osteons participates in the exchange of Ca++ between bone and blood; this is in part under control of the parathyroid glands, and in part independent of them; the parathyroids, however, serve to control the level of Ca++ in the blood plasma. It is suggested also that vitamin D mav plav a part in the regulation of exchange of calcium between blood and bone. Acknowledgment. This study was aided by Research Grant A-4225 from the National Institute of Arthritis and Metabolic Diseases, National Institutes of Health, United States Public Health Service. Kefkhences Amprino, R., and Godina, G. 1953. Le renouvellement structurel du tissu osseux de poissons teleosteens. Compt. rend, assoc. onat., 40, 573- 580. Bauer, G. C. H., Carlsson, A., and Lindquist, B. 1955. Evaluation of accre- tion, resorption, and exchange reactions in the skeleton. Kgl. Fysio- graf. SciUskop. Lund, Fcirh., 25, 1-16. Bauer, G. G. H., Garlsson, A., and Lindquist, B. 1961. Metabolism and homeostatic function of bone. In Mineral McfaboJisJii (G. L. Gomar and F. Bronner, editors), Vol. I, Part B, pp. 609-676. Academic Press, Inc., New York. Ghievitz, O., and Hevesv, G. 1935. Radioactive indicators in the study of phosphorus metabolism in rats. Nature, 136, 754-755. Dallemagne, M. J., ^^'inand, L., Richelle, L., Herman, H., Francois, P., Fabry, G., and Gloosen, J. 1961. Les sels osseux. Etat actuel de la question. Bull. acad. ray. mcd. Belg., ser. 7, 1, 749-808. Frost, H. M., and Villanueva, A. R. 1960. Observations on osteoid seams. Henry Ford Hosp. Med. Bid!., 8, 212-219. Frost, H. M., Villanueva, A. R., and Roth, H. 1960. Measurement of bone INTERNAL REMODELING OF COMPACT BONE 383 fonnation in a 57 vear old man by means of tetracyclines. Henry Ford Hosp. Med. Bull., 8, 239-254. Jowsey, J., Rowland, R. E., Marshall, J. H., and McLean, F. C. 1958. The eflFect of parathyroidectomv on haversian remodeling of bone. En- door inologij, 63, 903-908. Marshall, J. H., Rowland, R. E., and Jowsey, J. 1959. Microscopic metabo- lism of calcium in bone. V. The paradox of diffuse activity and long- term exchange. Radiation Research, 10, 258-270. McLean, F. C, and Urist, M. R. 1961. Bone; An Introduction to the Phijsiologi/ of Skeletal Tissue. University of Chicago Press, Chicago, 111. Neuman, W. F., and Neuman, M. W. 1958. The Chemical Dynamics of Bone Mineral. University of Chicago Press, Chicago, 111. Norris, W. P., Chavin, \^^, and Lombard, L. S. 1963. Studies of cal- cification in a marine teleost. Ann. N. Y. Acad. Sci., 109, 312-336. Posner, A. S., and Perloff, A. 1957. Apatites deficient in divalent cations. /. Research Natl. Bur. Std., 58, 279-286. Posner, A. S., Stutman, J. M., and Lippincott, E. R. 1960. Hydrogen-bond- ing in calcium-deficient hvdroxyapatites. Nature, 188, 486-487. Tomes, J., and De Morgan, C. 1853. Observations on the structure and development of bone. PJiil. Trans. Roy. Soc. (London), Scr. B, 143, 109-139. Vincent, J. 1957. Les remaniements de I'os compact marque a I'aide de plomb. Rev. beige pathol. et mcd. exptl., 26, 161-168. Vincent, J., and Haumont, S. 1960. Identification autoradiographique des osteones metaboliques apres administration de Ca 45. Rev. franc, etudes din. et biol, 5, 348-353. 15 Rarefying Disease of the Skeleton: Observations Dealing with Aged and Dead Bone in Patients with Osteoporosis MARSHALL R. URIST, Division of Orthopedics, Department of Surgery, School of Medicine, Center for the Health Sciences, University of Cali- fornia, Los Angeles, California NORMAN S. MacDONALD, Departments of Nuclear Medicine, Bio- physics, and Radiology, School of Medicine, Center for the Health Sci- ences, LTniversity of California, Los Angeles, California MILTON J. MOSS, Division of Orthopedics, Department of Surgery, School of Medicine, Center for the Health Sciences, University of Cali- fornia, Los Angeles, California WILLIAM A. SKOOG, Department of Medicine, School of Medicine, Center for the Health Sciences, University of California, Los Angeles, California THE subject of rarefying disease of bone in clinical medicine in- cludes at least ten relatively uncommon conditions and one very common disorder. Some of the uncommon conditions are: osteo- malacia, osteitis fibrosa generalisata, renal osteodystrophy, hypo- phosphatasia, metastatic bone disease, osteitis deformans, fibrous dysplasia of bone, multiple myeloma, scurvy, and osteogenesis im- perfecta. The one common disorder is osteoporosis, a condition very difficult to diagnose at an early stage and often superimposed upon physiological atrophy of bone of old age. In the advanced stage, 385 38() URIST, MACDONALD, MOSS, AND SKOOG osteoporosis is recognized by hone failure or spontaneous collapse of vertebral bodies, ballooned discs, and thin cortex in radiographs of the dorsal and lumbar regions of the spine. After a thorough clinical and laboratory study of the patient, it is easy to identif\' osteoporosis as a disease by these three radiographic changes. Pa- tients with osteomalacia and hyperparathyroidism may present the same radiographic picture, l^ut the clinical and laboratory findings are different and the number of cases is relatively small. Patients with diffuse decrease in radiographic bone density (a highly sub- jective observation), without anv deformity of the vertebrae, are classified as examples of physiologic atrophy of bone of old age. Patients in an early stage of the disease process of osteoporosis can- not be diagnosed by clinical or laboratory methods that are now available to physicians.* Controversy exists about whether the osteoporosis is an endocrine (metabolic) disorder, a degenerative disease, or an integral part of the time-dependent process of aging. Some investigators claim that osteoporosis is simply mild chronic osteitis fibrosa generalisata caused by prolonged calcium deficiency (Fraser, 1962); others con- tend that osteoporosis consists of generalized bone atrophy that is complicated by either osteitis fibrosa or osteomalacia ( Clerkin et ah, 1962) or both (Meroney ef ah, 1959) and, therefore, is not a dis- crete condition; others (Little et al., 1962; Casuccio, 1962) postulate that osteoporosis is primarily a degenerative disease of bone tissue. A larger number of cases were reported in 1961-1962 than in the preceding 5 years, and more new information was obtained about pathogenesis of the disorder than in all of previous time. The etiol- ogy, however, is not known and treatment is unsatisfactory. " Osteoporosis may be present or absent in individuals with a number of diverse conditions, as, for example, acromegaly, diabetes, hyperthyroidism, gonadal agenesis, Cushing's syndrome (endogenous and exogenous), pernicious anemia, starvation, tumors, cirrhosis, von Gierke's disease of the liver, posttraumatic metabolic syndrome, and postpregnancy state. The proportion of these diagnoses in the total number of patients with osteoporosis is 19 per cent. The proportion of osteoporosis in otherwise healthy individuals, between 50 and 75 years of age, generally classified in hospital records as postmenopausal or senile types is 81 per cent (Moon and Urist, 1962). Osteoporosis is frequently more severe in patients with than in patients without the above diseases, but the causative relation is indirect and involves poorly understood complex reactions of bone tissue. RAREFYING DISEASE OF THE SKELETON 387 The object of this communication is to ( 1 ) siilimit observations on 5-year follow-up examinations of 100 women with an a^'erage age of 85; (2) add two cases of osteoporosis to six cases previously re- ported and intensively studied by correlated methods; (3) analyze the literature of the period between 1961 and 1963 and some pre- vious articles that lend support to current research work. Articles appearing between 1957 and 1961 (Urist et al, 1962; Urist, 1962) and reviews of previous years have already been discussed in detail (Vincent and Urist, 1961; Gordan, 1961; Fraser, 1962; Moon and Urist, 1962). Case Material and Results Case No. 7 Bl. S., a childless housewife, 67 years of age, had severe backache for a period of 3 vears. The condition began when she lifted a chair, experienced sharp pain in the low back, and was found to have spontaneous collapse of two vertebrae. The pain subsided after treat- ment with a corset and sex hormones for 9 months. One year later, she had another attack of backache associated with spontaneous fractures of the lower dorsal spine. The response to the treatment was unsatisfactory. The past history was interesting in that the pa- tient was treated for anemia for 8 years. She lost 15 pounds and 1 inch in height in a period of 1 year. Physical examination revealed a flattening of the dorsal and lumbar spinal curves, and deep cir- cumferential skin folds between the costal margin and iliac crests. Radiographs showed old compression fractures of the 7th, 9th, and 11th dorsal, and 1st, 2nd, and 4th lumbar vertebrae; intervertebral discs between the unfractured segments were ballooned and the cortex was everywhere thin (Figs. lA to IC). The results of labora- tory studies were as follows: sternal marrow cytologv was negative for mveloma cells; hemoglobin, 11.0 gm %; sedimentation rate, 40 mm/hr. (Wintrobe); white blood cells and differential counts were normal; serum protein 6.9, albumin 3.7, and globulin 3.2 gm % with AG ratio of 1.1; protein-bound iodine, 5.7 mg %; serologic tests for syphilis were negative; cholesterol, 262 mg 7c; serum calcium, 11.0 mg 7c; inorganic phosphorus, 3.5 mg %; alkaline phosphatase, 8.2 388 URIST, MACDONALD, MOSS, AND SKOOG Fig. 1A. Lateral- view radiograph of dorsal spine (Case No. 7, Bl. S.) shows bone failure or collapse of 7th, 9th, and 11th dorsal to 1st lumbar verte- brae. RAREFYING DISEASE OF THE SKELETON 389 Fig. IB. Lateral-view radiograph of lumbar spine in case shown in Fig. lA, showing thin cortex, ballooned discs, and bone failure in the 1st, 2nd, and 4th lumbar vertebrae. 390 ITRIST, MACDONALD, MOSS, AND SKOOG Fig. IC. Snapshot showing the patient (Case No. 7) performing bicycling exercises. Note the foreshortened trnnk, cine to collapsed vertebral bodies in the spine. King Armstrong units; serum citric acid, 2.9 mg %; quantitative de- terminations for fat in the stool were within normal limits; routine urinalysis presented no abnormalities. Proline tolerance test. The rate of bone formation in Case No. 7, as compared with normal growing children and an osteoporot- ic patient with a fracture of the hip, was in\ estigated by means of RAREFYING DISEASE OF THE SKELETON 391 the proline tolerance test of Summer ( 1961) and Ibsen et al. ( 1963). In this test the rate of disappearance of proline from the blood was increased during the period of rapid formation of callus and new bone. In Case No. 7, the rate was the same as in normal nonosteo- porotic subjects, but lower than in \'oung individuals and osteoporot- ics during a period of callus formation ( Fig. 2 ) . 160^ TIME HOURS Fig. 2. Graph showing results of prohne tolerance test in Case No. 7, show- ing .normal bone formation rate as determined by the rate of disappearance of an intravenously injected dose of 65 mg proline per kg body weight. Metabolic balance. During a control period of 20 days, Case No. 7 was in negative nitrogen, phosphorus, and calcium balance. This was restored to equilibrium with respect to nitrogen and phos- phorus, but positive calcium balance developed during a period of 15 days of treatment with oral administration of 65 mg per kg of proline per day. In order to determine whether the effect was upon intestinal absorption, 500 mg of proline was administered daily by intramuscular injection for an additional 15 days; the patient re- mained in nitrogen and phosphorus equilibrium but returned to negative calcium balance. Proline was empirically selected for this study because it is a precursor to collagen, known from animal ex- periments with proline-H" to be deposited in bone matrix ( Fig. 3 ) . Radioisotope kinetics. Figure 4 illustrates the excretion of an injection of 5 /jlc of Sr^^ as determined by the changes in the amount 392 URIST, MACDONALD, MOSS, AND SKOOG CASE Bl. S. PHOSPHORUS BALANCE PROLINE OXYTETRACYCLINE i CASE Bl. S. OXYTETRACYCLINE | GM 12 10 8 6 4 2 0 n^-^ NITROGEN BALANCE URINARY EXCRETION ' '-' ""T-i ^j«^ vj J FECAL EXCRETION iii[im|ii]][iill|iiiniiii|iiiniiii|iiii|ii]i I '2 3 '4 '5 6*7 ' 8 9 10 FIVE-DAY PERIODS GM 1.6 H 1.4 1.2 1.0- 08- 0.6- 0.4- 02 00 MG% 5 -4 ■3 URINARY EXCRETION iiiiiiiiiiiiiirn FECAL EXCRETION GM 08 06 04 02 00 CALCIUM BALANCE URINARY EXCRETION FECAL EXCRETION 234'56'7'8'9'I0 FIVE-DAY PERIODS MG% 12 II 10 Fig. 3. Metabolic balance studies for nitrogen, calcium, and phosphorus in Case No. 7, showing loss of nitrogen, phosphorus, and calcium in the pretreat- ment period, restoration of equilibrium during oral administration of proline, and return to negative calcium balance during treatment with intramuscular injections of proline. detected by means of whole body counting in Case No. 7. The close correspondence between the results of excretion as determined bv retention in the skeleton and assay of absolute amounts of Sx^'' in the urine and feces demonstrated that the use of a whole body counting facility can eliminate the costly and more time-consuming balance studies for calcium in the future. The osteogram and Bauer- RAREFYING DISEASE OF THE SKELETON 80 393 20 O EXCRETA ASSAYS A TOTAL BODY COUNTING 10 15 DAYS POST INJECTION Fig. 4. Graph showing the rate of excretion of Sr*-"' in Case No. 7, as meas- ured both by the percentage of the close detected by the total body counting technic, and by direct radio assay of total vuinary and fecal output. The results of the two methods are entirely comparable and equally accurate. Carlssen-Lindquist calculations were not completed in this case be- cause of mechanical failure of the coimting equipment during the period of the balance study. Tetracycline labeling and biopsy. Oxytetracycline was infused intravenously in a dose of 1.0 gm per liter of 0.9 per cent solution on the 5th and 20th days of the control period and again on the 50th day. A biopsy was performed on the upper end of the tibia at the conclusion of this studv on the 50th dav. The thickness of the cortex, measured with the aid of a micrometer, was 0.6 mm or 50 per cent of that of young adult women. On this basis, assuming that the thickness of the cortex of the upper tibia reflects the density of the whole skeletal system ( Lindahl and Lindgren, 1962 ) , the skele- ton in this patient weighed approximately 5.6 kg. The spaces be- tween the tetracycline lines, considering the 30- versus the 15-day interval, indicate no increase in bone formation rate (Fig. 5A). On the histological level, in the low-power magnification with the light microscope, the bone that persists (in Case No. 7) was laid down in the normal configuration and had the appearance of normal 894- URIST, MACDONALD, MOSS, AND SKOOG Fig. 5A. Fluorophotomicrograph showing triple labeling of cortical bone in Case No. 7. The space between the two outer rings reflects the amount of bone deposited during a 15-day control period, and the space between the middle and inner ring indicates the amount deposited in a 30-day period on treatment with proline. With due allowance for the well known fact that the bone de- position rate is high at the beginning and low at the end of the period of forma- tion of a new osteon, the longer interval and wider space indicate that the treat- ment did not increase the mean osteon formation time for the bone in the upper end of the tibia. ( X approx. 450. ) bone. In high-power magnification, however, some parts of the skeleton presented an abnormal change in that they had large areas in which the lacunae had lost their osteocytes. The cell-free areas were frequently in the outer layers of the osteons and in the inter- stitial lamellae of cortical bone. The matrix also stained irregularly basophilic and metachromatic rather than as a homogeneous eosino- philic structure. Occasionally, it was possible to see strips of tissue in which the collagen fiber bundles accepted the Wilder stain ir- regularly. In some areas of the surface compacta, as, for example, in the neck of the femur and the anterior cortex of the vertebral bodies RAREFYING DISEASE OF THE SKELETON 305 (in autopsy subjects), there were patches of basophihc metachro- matic, acellular necrotic bone that predisposed the patient to a spontaneous fracture. The aboxe clianges also ojcurred in nonosteoporotic bcue from both men and women of comparable age, but quantitatively the degree was significantly less than in osteoporotic subjects (Table VI). The osteoc\'tes in inner lamellae around blood vessels averaged 6.5 cells per high-power field in nonosteoporotics and 2.7 in osteo- porotics. In the outer lamellae, the outstanding difi^erence was in the number of retracting osteoc\tes, 6.0 and 3.7, and calcified osteo- cytes, 3.0 and 1.5, in osteoporotics and nonosteoporotics respectively. In the interstitial lamellae, the average number of enlarged empty lacunae was 28.2, compared with 22.0. To judge from the appearance of increasing amounts of bone with emptv lacunae with incieasing age and with chronic peripheral vascular disease, it was assumed that this condition represents devitalized bone from prolonged cir- culatory insufficienc)' (Sherman and Selako\itch, 1957). We have reviewed the biopsy sections of our previously reported cases ( Urist et ah, 1962) and observed that osteocytes move to one side and then enlarge the lacuna bv absorption of matrix between the canalic- uli, possibly through the action of a proteolvtic enzvme. Later, the cell undergoes autolysis, and an emptv enlarged spherical (rather than stellate and ovoid) lacuna remains. These morphologic changes were unspecific l^ut approximately 30 per cent more extreme during middle-aged life in patients with severe osteoporosis (Figs. 5B to 51). Quantitative analyses of oxytetracycline in the biopsy sample by the method of Ibsen et al. ( 1963) revealed 4 y per gm of bone. This represented the amount deposited in the skeleton in Case No. 7 after infusion of 3.0 gm in a period of 45 days. This was 3.2 y less than the amount to be expected, inasmuch as the average nonosteo- porotic deposited 2.4 y after one infusion of 1.0 gm. Fluorescent microscopy demonstrated that very few osteons, approximately 5 per cent of the total number in the section, were growing and re- moving oxytetracychne from the extracellular fluid; in control sub- jects of comparable age, approximatelv 10 per cent were reactive. Examination of the active osteons by white light microscopy re- 396 URIST, MACDONALD, MOSS, AND SKOOG Fig. 5B. Photomicrograph showing a typical osteon in cortical bone in osteo- porosis. The osteocvtes in the inner lamellae near \ essels appear normal, whereas osteocvtes in the outer lamellae (large arrow) are retracting. The lacunae in the interstitial lamellae between osteons are enlarged and transformed from a thin stellate to a spherical cavity either containing a pyknotic nucleus or en- tirely empty (small arrow). Hematoxylin, eosin, and azure II stain, (x ap- prox. 450.) vealed normal stellate osteocytes in both the inner and outer lamellae of the growing osteons. Nearly all the nongrowing osteons had re- tracting, enlarged, or devitalized osteocytes in the outer lamellae ( Fig. 5C ) and empty lacunae in the interstitial lamellae. There was no way of measuring the age of the inactive osteons (those with a hypercalcified inner margin), but the oxytetracycline label at three intervals suggested that approximately 12 weeks was the time that RAREFYING DISEASE OF THE SKELETON 397 Fig. 5C. Photomicrogiaph showing a typical area of osteoporotic interstitial bone. The lacunae are enlarged, spherical, smooth-walled, and empty (arrow). The broad lines between the lamellae represent thick layers of metachromatic- staining cement substance, presumably deposited on exposed sin-faces of matrix under conditions in which the rate of bone formation is abnormally low. (x approx. 400.) an osteoporotic required to build a new osteon. However, many osteons stopped growing and developed a sclerotic inner margin before they were three-fourths closed in aged and osteoporotic in- dividuals. The life expectancy of an osteocyte in an outer lamella ( the area of the osteon farthest from the central blood vessel ) in aged persons with osteoporosis and chronic circulatory insufficiency is probably not more than 12 weeks. Further investigations with the aid of con- tinuous administration of oxytetracycline are in progress to meas- ure this interval in both nonosteoporotic and osteoporotic subjects. 398 rmST, MACDOXALD, MOSS, AND SKOOG Fig. 5D. Photomicrograph showing dii area of devitahzed bone in the center of the cortex of the tibia in severe osteoporosis. The lacunae are en- kirged everywhere and the osteocytes are either pvknotic, fragmented, or absent. The thin hne inchcated by the arrow represents a rapidly deposited metachromatic-staining cement substance on the outer wall of a resorption cavity; this is the reversal line that marks the beginning of construction of a new osteon. ( X approx. 400.) Case No. 8 M. R., a retired school teacher, 71 )'ears of age, had severe back- aches, pam in both lower extremities, and was unable to walk with- out the aid of two canes. At the age of 68, while lifting an emptv bookcase weighing less than 25 pounds, she incurred a compression fracture of the 12th dorsal vertebra. One month later, three addi- tional vertebrae, the 9th and 10th dorsal and the 1st lumbar vertebra, collapsed following a trixial injurw She lest 6 cm in height and 25 pounds in weight in a period of 6 months (Figs. 6A and 6B). The treatment of the fractures consisted simply of bed rest and 5 mg RAREFYING DISEASE OF THE SKELETON 399 Fig. 5E. Photomicrograph illustrating a complex of six areas of interstitial bone enclosed in thin reversal cement lines between two half-closed osteons. The matrix of old bone surrounds enlarged emptv lacunae and is strongly basophilic and metachromatic rather than eosinophilic (open arrow). Only the new bone surrounding large vascular channels contains small li\'ing oste- ocytes (indicated by black arrows) . ( X approx. 400.) of methandrostenolone (Dianabol) per day for 6 weeks. She re- gained most of the weight loss, but the backache persisted. The past history was remarkable in that the patient was treated for hyper- thyroidism, gastroenteritis, and herpes zoster when she was 66 years of age. The diagnosis of osteoporosis was not considered because, during the 7 months preceding the first fracture, she received small doses of ACTH and triamcinolone empirically for treatment of vague bone and joint pains. This was discontinued as soon as radiographs of the spine revealed collapsed vertebrae (Figs. 6C to 6E). The calcium intake was estimated at 700 mg per day, and her food habits in general were evaluated as normal. Routine hospital labora- 400 URIST, MACDONALD, MOSS, AND SKOOG Fig. 5F. Photomicrograph showing trabecular bone in severe osteoporosis. The total number of trabeculae is low, and some are attenuated, but a few are decreased in length and increased in thickness. The enlarging empty lacunae and shrinking osteocytes are in the core (open arrow), while the living osteocytes (black arrow) are on the surface. (X approx. 400.) tory studies showed no abnormality. The hiboratorv tests given Case No. 7 were also given Case No. 8, and the results were all within the limits of normal. Metabolic balance study. During a control period of 15 days, the patient complained of backache and spent 20 hours a day in bed. She lost weight and was in negative calcium, phosphorus, and nitrogen balance. Polysaccharides (Dingwall et al., 1962; Westcott et at., 1962 ) administered by intramuscular injection, 5 mg per day, between the 15th and 30th days, did not prevent further decline in weight, or change the calcium, phosphorus, or nitrogen balance. Hu- man growth hormone (supplied by the Endocrine Study Section, RAREFYING DISEASE OF THE SKELETON 401 *■. # * f ' N lite-' '- *■ • /V FlG. 5G. Microradiograph of the endosteal surface of the cortex showing old bone in a low-density osteon with a sclerotic inner wall and calcified osteocytes (arrow on the right). The arrow near the center indicates an en- larged lacuna with a rarefied zone of matrix around it. The arrow on the left points to hypermineralized interstitial bone in which many of the lacunae are filled with calcium salts. The osteon with the large vascular channel in the lower right corner of the picture has a sclerotic inner ring; this occurs when bone formation ceases before the system is half closed; quantitatively, this appears more frequently in osteoporotics than in nonosteoporotics and causes increased porosity of the cortex. (X approx. 400.) National Institutes of Health, United States Public Health Service, 1962 ) was administered by intramuscular injection of 2 mg per day from the 50th to the 65th day; this produced 2.0 gm of positive nitrogen balance but little or no improvement in the negative cal- cium and phosphorus balance; the urinary calcium excretion, nor- mally onlv 50 to 100 mg per dav in aged osteoporotics, increased from 100 to 200 mg per day. These effects of growth hormone were associated with further decline in weight, and probably did not 402 URIST, MACDONALD, MOSS, AND SKOOG Flu. 511. Photomicrograph ilhistrating an enJargctl lacuna in hypermincr- ahzed interstitial bone (arrow). This cell is also shown in Fig. 51. Osteocytes with normal nuclei and lacunae of normal shape and size are found only in perivascular lamellar bone. (X approx. 400.) reflect specific action upon either the skeletal svstem or the osteo- porosis (Fig. 7). Radioisotope kinetic study. An intravenous injection contain- ing 5 fjiC of Ca^', 5 fic of Sr^^, and 1 fxc of P^^ labeled serum albumin was administered during the period of treatment with daily injec- tions of polysaccharides extracted from bone, and again during the period of treatment with human growth hormone. Osteograms of the midshaft of the tibia during polysaccharide treatment showed a slow change in rate of uptake, rising over a period of more than RAREFYING DISEASP: OF THE SKELETON 403 Fig. 5 I. Microradiograph showing the identical area and osteocyte lacuna (see arrow) illustrated in Fig. .5H. Note the hvpermineralized interstitial bone and the low-density mineralized lamellar bone in the lower right osteon. The sclerotic inner margin suggests that this osteon represents a deposit of old bone. (X approx. 400.) 1 hour; on treatment with human growth hormone, the rate of ac- cumulation was the same as in normal subjects, where the accumula- tion of the radioisotopes reached a plateau in 15 minutes. Other areas of the skeleton responded in the same way as the tibia (Fig. 8). The daily loss of Sr^^ in the urine during treatment with poly- saccharide, especially during the first 3 days, was less than with human growth hormone. One-half of the Sr^° was excreted in 5 days when the patient received polysaccharides; on human growth hormone, 50 per cent was excreted in only 2 days (Fig. 9). Ca^' was much better retained in the skeleton than Sr" '. The 50 per cent excretion point was reached after 18 days on polysaccharide treat- ment but only 5 days on human growth hormone. This clearly 404 ^*\ URIST, MACDONALD, MOSS, AND SKOOG B <^^ ^tJ Fig. 6A. Photograph of 71-year-olcl retired school teacher (Case No. 8, M. R.) showing fixed flexion of the cervical spine and flattening of the dorsal and lumbar spine. This patient lost 6 cm in height. Fig. 6B. View of the back (Case No. 8) illustrating skin folds (arrow) produced by telescoping of the torso. EAREFYIXG DISEASE OF THE SKELETON ■lOo Fig. 6C. Lateral-view radiograph of the dorsal spine, showing bone failure and collapse of the 9th to the 11th dorsal vertebrae (Case No. 8). 406 URIST, MACDOXALD, MOSS, AND SKOOG Fig. 6D. Lateral-view radiograph of the himbar spine of the case shown in Fig. 6C, illustrating ballooned discs, thin cortex, and collapse or bone failure of the 3rd lumbar vertebra. RAREFYING DISEASE OF THE SKELETON m 40' Fig. 6E. Anteroposterior view of the pelvis showing accentuation of the trabeculae of the internal architecture, and decrease in the density of the lateral cortex of the neck of the femur in the case shown in Figs. 6A to 6D. demonstrated the established fact that strontium is not reabsorbed as efficiently as calcium by the kidney tubules (Fig. 10). The rela- tive values of the exchangeable pool and calcium excretion rate, as shown in Table I, are calculated on the basis of urinary clearance of strontium, and therefore are not absolute values but can be used to compare one case with another. Our impression was that the size of the exchangeable calcium pool (on the basis of the osteogram) was the same during the entire period of treatment, but the accre- tion rate was greater with polysaccharide than with human growth hormone. Calcium balance was unchanged, however, and it is neces- sary to assume that the higher rate of accretion was associated with a higher rate of resorption. This is reflected in the urinary excretion curves of Fig. 10. The dailv loss of Sr^^ in the urine accompanying 408 CASE MR ? LTRIST, MACDONALD, MOSS, AND SKOOG CASE MR 9 LBS 147 146 145 144 143 142 GM 16 BODY WEIGHT NITROGEN BALANCE URINARY EXCRETION ""^j^i/t^LT"^ FECAL EXCRETION TTTT|TTTTpTn|TTTTjnil|MII|llll|llll|llll|llll|llll|llli 5 6 7 8 9 10 II 12 13 14 FIVE - DAY PERIODS I8H 16 PHOSPHORUS BALANCE URINARY EXCRETION :!llLjJV|rA^-wi[y^ Fig. 7. Charts showing metabolic balance studies in Case No. 8. This patient lost weight and was in continuous negative calcium, phosphorus, and nitrogen balance. Polysaccharide treatment (Osseofac, Squibb) did not re- store equilibrium. Human growth hormone produced nitrogen retention but did not correct negative calcium and phosphorus balances. polysaccharide treatment was less during the first 10 days than with the growth hormone. Thereafter, more Sy^'' was lost per day during the polysaccharide regime, suggesting a higher rate of resorption of bone which had been labeled with the isotope. Thus, the Sr""' ex- creted during the early days comes mainly from soft tissue and from Sr^^ deposited on preexisting bone salt crystals by exchange rather than by incorporation into newly forming crystals (accretion). At RAREFYING DISEASE OF THE SKELETON 409 47 ^GIVEN TOGETHER I.V. 1ST STUDY (POLYSACCHARIDE) J I I I I I I I I I ' I -I 2ND STUDY (HUMAN GROWTH HORMONE) J I I I I I I I I I I \ I I L_J 20 40 60 MINUTES POST INJECTION 80 Fig. 8. Radioisotope osteogram of MacDonald (1960) showing the changes in rate of uptake of Sr^-"' and Ca^' in Case No. 8. The upper curve (first study) represents the reaction of the tibia during a period of treatment with intra- muscular injections of polysaccharides; this suggests a high rate of deposition of bone mineral. The lower curve (second study) represents the response to treatment with growth hormone; this is the same as obtained in untreated nonosteoporotic subjects bv MacDonald. later times, however, the daily loss probably is derived primarilv from continual resorption of mineral crystals which contained Sr'*^ laid down earlier by accretion. Bone biopsy and tetracycline uptake. Oxvtetracvcline was in- fused intravenously, in doses of 1.0 gm per liter of 0.9 per cent salt solutions, on the 1st, 16th, and 39th days. A biopsy was performed on the 42nd day on the proximal end of the shaft of the tibia. Fluores- cent microscopy revealed a triple line of fluorescence. The amount rRIST, MACDONALD, MOSS, AND SKOOG 25 35 5 DAYS POST INJECTION Fig. 9. Charts illustrating the retained amounts of Sr^^ and Ca^" as meas- ured by whole body counting in Case No. 8. See Fig. 8. At all times, the amounts of both isotopes retained by the skeleton were greater during treat- ment with polysaccharide (first study) than with human growth hormone (second study). The time required to lose 50 per cent of the Sr^^ ^^g 5 days for the first study, as compared with 2 days for the second. of bone between the two inner lines represented the amount of bone deposited during the period of 15 days of treatment with poly- saccharide; the amount between the middle and outer lines, the control period of 15 days of observation preceding treatment. The distance between the two lines was approximately equal. In non- osteoporotic control cases, the rate of bone formation decreased and the space between the outer and inner lines diminished in the proc- ess of closure of a labeled osteon. There was, however, considerable variation between individual osteons. Some showed the diminishing space between lines, and, therefore, no conclusive evidence of stimu- lation of bone formation (Fig. 11). The histophysiologic changes observed in Case No. 7 were even more extreme in Case No. 8. All sections showed many empty lacu- RAREFYING DISEASE OF THE SKELETON 20- 10- 411 UJ O Q A Sr^^ I ST STUDY A Sr^^ 2ND STUDY o Ca"*^ 2ND STUDY J L. J L_^ L J I I L 5 10 15 DAYS POST INJECTION 20 Fig. 10. Chart showing urinai) excretion of Sv^'' and Ca^"^ in Case No. 8. See Fig. 8. The total loss of Sr"*"' isotope during the first 3 days was only 23 per cent on polysaccharide therapy (first study), as compared with 37 per cent on human growth hormone (second study). These results suggest more ac- cretion of bone salts after treatment with polysaccharide, as compared with osteoporotics treated with human growth hormone, or osteoporotics observed without treatment of any kind. However, an enhanced resorption rate is tend- ing to counteract the increased accretion, so the net effect over a period of several weeks is small. nae and irregular staining of matrix of both haversian and interstitial lamellae (Figs. 5B to 5G). After infusion of 3.0 gm of oxytetracy- cline, onlv a rare osteon fluoresced, and the total uptake was only 1.0 y per gm of bone, or 84 per cent less than in nonosteoporotic controls. The over-all impression gained from these observations was that in Case No. 8 as much as 75 per cent of the bone in the tibia was dead! 412 URIST, MACDONALD, MOSS, AND SKOOG Fig. 11. Fluorophotomicrograph of unckcalcified section of the proximal end of the tibia in Case No. 8, illustrating triple labeling of an obliquelv cut new osteon on the endosteal surface of the cortex. The space between the two outer lines (i and 2) represents the amount of bone deposited during 15 days on polysaccharide therapy, and the two inner lines (2 and 3) the control period off therapy. The equal spacing, with due consideration for the diminish- ing rate of deposition of bone, and the mean osteon formation time in this area of the skeleton, suggests a slight beneficial effect of treatment on osteogenesis in this case. RAREFYING DISEASE OF THE SKELETON 413 TABLE I. Case No. 8, M. R.: Intravenous Injections OF A Mixture of Sr^^ and Ca^' Treatment Polysaccharide (Osseofac) Human growth hormone Tibia osteogram Knee osteogram Urinary excretion (3 days) Total excretion (total body counting) : time to lose 1st 50% of dose Exchangeable calcium pool" From Sr**'^ data From Ca^^ data Accretion rate for calcium" From Sr^^ data From Ca'*'' data Both Ca« and Sr«-^ still rising after 70 minutes Both Ca^^ and Sr«^ still rising after 70 minutes Sr, 23%; Ca, lost Both Ca^' and Sr«^ leveled off after 15 minutes Both rising after 70 min- utes but less steeply Sr, 37%;Ca, 14% Sr, 5 days; Ca, 18 days Sr, 2 days; Ca, 5 days 1.9 gm 0.06 gm/day 1.5 gm 5.9 gm 0.04 gm/day 0.22 gm/day " Calculated by method of Eauer et al. (1961). Summary of Correlated Observations in Cases 7 and 8, and Previous Six Cases of Severe Osteoporosis Table II includes previously tabulated data (Urist et al., 1962) and data obtained from Cases Nos. 7 and 8. This table demonstrates that correlation of the results of physiologic and morphologic meth- ods of investigation was extremely valuable. The information gained from one method tested the validitv of the other. Cases Nos. 7 and 8 (while receiving approximately the same amount of calcium as there was in the diet to which they were accustomed) were in negative calcium, phosphorus, and nitrogen balance. Case No. 7 was restored to equilibrium or positive calcium balance by proline, but tetracycline labeling showed no improvement in bone formation rate or alteration of the course of osteoporosis. In Case No. 8, bal- ance studies showed no improvement in calcium retention, while the radioisotope osteograms revealed high or normal rates of uptake of Sr^^ and Ca^' on polysaccharide therapy. The accretion rate cal- culated by the Bauer et al. (1961) formulations was also higher. The tetracycline uptake in new bone was approximately the same 414- URIST, MACDONALD, MOSS, AND SKOOG as in nonosteoporotic subjects of comparable age. Judging from the extraordinary decrease in the thickness of the cortical bone, and assuming an equal loss of cancellous bone, it was apparent that both patients had lost approximately half of the mass of the skeleton. But in the face of this great deficiency, balance studies demonstrated that osteoporotics were able to turn over almost as much calcium TABLE II. Summary oi Observations Ca Ca Case no. intake balance Sr^^O Ca^'A Cx SM TC c* 0.8 ±25 25 0.80 100 11.6 2.4 1 0.7 + 50 50 0.72 50 5.6 2.0 2 0.8 -ISO 21 0.19 ()(i 7.3 4.1 3 0.5 -100 () 0..50 ()f) 7.3 3.5 4 1.0 + 5 23 0.75 50 5.6 2.4 5 0.5 -5 10 0.51 50 5.6 0.4 6 0.4 -50 — — 50 5.6 0.3 7 0.7 -100 — — 50 5.6 4.0 8 0.8 -400 15 0.22 50 5.6 1.0 Abbreviations: C*, average of subjects with no osteoporosis. Ca intake, average daily intake of calcium in the diet in grams. Ca balance, daily gain or loss of calcium in diet in milligrams per daj'. Sr*''0, time in minutes during which cortical bone of the tibia produced an increase in the counting rate of the .scintillation probe-photomulti- plier tube-ratemeter system. A, accretion rate in grams of calciiun per day, estimated by method of Bauer et al. (1961); Sr*^ was used in all but Case No. 8, where Ca''^ was used. Cx, percentage of the thickness of the cortex of the tibia of nonosteoporotic sub- jects of comparable age. SM, bone weight in kilograms estimated on the basis of normal bone mass = 16 per cent of body weight. TC, total uptake of tetracycline in y per gram of weight of fresh cortical bone obtained at liiopsy. and phosphorus as patients with 100 per cent of their bone tissue. Tetracycline labeling of osteoporotic bone correlated with chemical balance and radioisotope studies revealed that this activity was ac- complished chiefly by the exchangeable, labile, reactive, or metabolic bone tissue — which constitutes only approximately 1 per cent of the total bone mass in normal subjects and perhaps not more than 2 to 3 per cent in osteoporotics. Thus, kinetic studies were unable to provide a measurement of the activity of either the total bone mass or the relatively nonreactive structural bone, which is where the deficiency exists in patients with osteoporosis. rarefying disease of the skeleton 415 The Progress of Osteoporosis in Five- Year Radiographic Survey of Aged Females The survey of 100 women with average age of 85, in residence at an Eastern Star Home in Los Angeles, was made in 1957 and re- ported at the Lankenau Hospital Conference in 1958 ( Urist, lQ60b ) . TABLE III. Five-Year Follow-up Radiographic Examinations OF 100 White Females, Average Age 85, Previously Reported BY Urist (19606) Radiologic observations Per cent Osteoporosis (diagnosis based on collapsed vertebrae and ballooned discs in 1957) 26 Deaths in group of both nonosteoporotics and osteoporotics in period 1957-1962, ages 70 to 103 75 Survivors with osteoporosis, ages 74 to 89 3 Survivors without osteoporosis, ages 71 to 91 22 Further fractures in osteoporotics, 1957-1962 10 Further fractures in nonosteoporotics, 1957-1962 0 Additional cases of osteoiDorosis appearing in 22 survivors previously free of compression fractures of the spine 0 Table III presents the results of a second radiographic survey de- signed to determine whether additional cases of the disorder (as diagnosed by collapsed vertebrae ) appeared during a 5-year period in previously unafflicted individuals. In addition, there was also the question whether the disorder (also determined by the number of collapsed vertebrae ) progressed in proportion to a period of time in the course of aging. Only 25 per cent of the women survived the period between 1957 and 1962; 22 per cent were nonosteoporotics and 3 per cent were osteoporotics. Ten per cent of the osteoporotics, but none of the nonosteoporotics, were treated for further fractures or spontaneous collapses of vertebral bodies during the 5 years. Thus, it appeared that the disorder was progressive in patients who had the disease, but was not incurred bv others who had reached the 8th and 9th decades of life. This was clearly evident from the radio- graphs of 22 women, with an average age of almost 88, who, as yet, had no collapsed vertebrae. It was also evident that osteoporosis generally occurred at a much earlier age, possibly between 50 and 416 URIST, MACDONALD, MOSS, AND SKOOG TABLE IV. Indicpzs of Physiologic Aging of Bone and Osteoporosis IN White Females between 5th and 7th Decades of Life Estimated per cent change Characteristic or component Aging, <65 Osteoporotic Authority Density (w/v, 10 bones) Bone tissue water Variation in density of osteons No. osteons less than ^4 closed Enlarged vascular channels, porosity Vascular channels with plugs of amorphous calcium de- posits Osteocytolysis Osteocyte lacunae filled with calcium salts Surface area occupied with resorption Rate of retention of bone- seeking isotopes Hexosamine/collagen ratio EDTA-soluble fraction of mucopolysaccharides Degraded acid-insoluble collagen (solid residue) Loss of fibrillar form of archi- tecture of collagen Bone highly reactive with tetracycline -12-30 -33-50 Trotteref aZ. (1960); Urist et al. (1962) -20 + 10 +20 ? +20 +30 Robinson (1960) Amprino and Engstrom (1952) ; author Jowsey (1960) + 10 +20 Jowsey (1960) +2 + 15 +5 +30 Author Sherman and Selakovitch (1957); Casuccio (1962); author + 1 +5 Casuccio (1962) + 10 -10 -10 + 15-40 -15 -20 Jowsey (1960); Frost (1961) Bauer et al. (1961) Sobele^a/. (1954); author -20 -50 Casuccio (1962) dbl + 10 Little et al. (1962) ±1 + 10 Little et al. (1962) + 10 + 10 Author 65, and did not progress at an increasing rate with advancing time. The indices hsted in Tables IV and V indicate that all morphological and chemical changes attributed to aging occurred in osteoporotics, probably at an earlier age and to a more extreme degree. These ob- servations were interpreted to suggest that osteoporosis is caused by an aberration or acute development of the aging changes in bone, perhaps in persons who are genetically predisposed to rapid earlier deterioration of the skeletal system. RAREFYING DISEASE OF THE SKELETON 417 TABLE V. Urinary Indices of Physiologic Aging OF Bone and Osteoporosis Estimated per cent change Aging non- Characteristic or component osteoporotic Osteoporotic Authority Response to ACTH ±10 ±15 Urist (19606) ; Gallagher and Spencer (1962) 1 1-deoxy-l 7-ketosteroi(ls — 5 -10 Urist and Vincent (1960) Calciuria 0 + 20 Urist (1962) Ability to reduce total urinai'v calcium on low calcium diet + 10 -25 Nordin (1962c) Calcium excretion index, Ca^yCa^", accumulative +5 + 10 Bronncr et al. (1962) Free hydroxj'proline + 10 +20 Raveni et al. (1962) Mucopolysaccharide + 10 +20 Bertolin and Greco (1962) Accelerated Changes of Aging in Osteoporotics The distinguishing feature of patients with severe osteoporosis, described in Table II, as compared with patients of comparable age without osteoporosis, was bone failure or collapse of the vertebral bodies of dorsal and lumbar spine. Figure 12 illustrates the relation between the reduction in bone mass and the age. The number of osteoporotic patients was too small for statistical evaluation, but suf- ficient to suggest a relationship to aging. The shape of the curve that describes the rapid rate of reduction of bone mass with time in osteoporotics, and the more gradual decline that describes the rate of normal or physiologic aging, indicated that the disease is most active during the period from 50 to 65, and mav progress, but rarely arises, after age 75. The various tissue (Table IV) and urinarv (Table V) indices of change in bone structure and metabolism that have been estimated and attributed to physiologic aging were also found in patients with osteoporosis. The large amount of tissue with empty lacunae and the patches of necrotic bone suggested bone circulatory insujQficiency in osteoporotics to a greater degree than in nonosteo- porotic aged persons (Table VI). Spontaneous fracture or bone failure occurs in the spine, the ribs, and the neck of the femur, where the cortex is thin and the necrotic osteoporotic bone is revas- 418 URIST, MACDONALD, MOSS, AND SKOOG 100 ^0.3 90_ 80. 70. 5 60. 50. 40 30. 20. lOj 0, _0.2 _0.l CASE _._^ NO. 7 CASE No. 8 • 29 RANDOM AUTOPSY SUBJECTS (LINDAHL AND LINDGREN, 1962 ) X 7 OSTEOPOROTIC PATIENTS (URlST et Ql.,1962) 10 20 30 40 50 AGE ~~i 1 1 1— 60 70 80 90 100 Fig. 12. Graph illustrating relation between change in bone density and age in random autopsy subjects and in patients with severe osteoporosis. The amount of retention and the rate of change in bone mass after age 70 is almost the same in the two groups of white women. The important characteristic of the group with severe osteoporosis is the sudden change in the quantity of bone retained in the skeleton between 50 and 65 years of age. The chief feature distinguishing the osteoporotic group from the group with physiologic rate of decrease in bone mass or aging was bone failure, or spontaneous collapse of vertebral bodies. It was assumed that borderline cases of osteoporosis, with or without slight bone failure, could occur in older age groups, but the separa- tion of the two groups was generally quite clear-cut. cularized and resorbed too fast to be replaced by new bone (Figs. 13A to 13C). Recent and Related Literature Race, Sex, and Genetic Factors Moon and Urist (1962) reported 87 cases, 77 women and 10 men, with severe osteoporosis admitted to a new hospital that received 90,000 patients over a period of 4 years. All were Caucasian. Trotter et al. (1960) measured the density of whole bones of anatomical specimens by the ratio of weight to displacement volume and ob- RAREFYING DISEASE OF THE SKELETON 419 TABLE VI. OsTEOCYTE Counts: Number of Stellate (S), Retracting (R), Calcified (C), and Empty (E) Lacunae per High Field in Cortical Bone in Osteoporotic and Nonosteoporotic Autopsy Subjects Age In Os teoc3'tes ner lamel ae Outer lamellae Interstitial lamellae Case no. S R C E S R 6 C E S R C E 1 42 3 3 1 0 1 3 1 0 5 0 24 2 44 3 2 1 1 1 5 4 3 0 4 1 26 3 50 3 4 0 0 1 6 3 3 0 5 0 30 4 54 3 3 1 0 2 6 3 4 0 8 0 29 5 V6 3 2 1 0 1 6 5 3 1 4 0 32 6 84 3 2 1 1 1 6 3 3 0 7 1 28 / 67 2 3 2 1 0 5 4 4 0 5 2 30 8 71 2 2 3 1 1 7 5 4 0 7 0 27 Average osteopo- rotic 63 2.7 2.6 1.2 0.4 1.0 6.0 3.7 3.0 0.1 5.7 0.5 28.2 CI 48 7 1 0 0 4 5 1 1 s 6 0 15 C2 50 / 1 0 0 4 4 0 1 s (1 1 18 C3 71 i 1 1 0 4 2 2 2 4 3 2 27 C4 80 5 2 1 1 3 4 3 2 1 2 1 28 Average nonosteo - porotic control 62 6.5 1.2 0.5 0.2 3.7 3.7 1.5 1.5 5.2 4.2 1.0 22.0 served that the bones of white males were normally denser than those of females; Negro females had denser bones than white; the rate of decline with age was parallel in all sex and age groups. Smith et al. (1960) made an x-ray survey of 218 ambulatory outpatient cases with osteoporosis; almost all were Caucasians, a very few were of Latin groups, and none were Negro, although the hospital was geographically located in the area of a large Negro population. This observation was reinforced further by the fact that the incidence of spontaneous fracture of the hip, a common symptom of osteo- porosis in patients over 70 years of age, was very much lower in Negro than in white racial groups (Moldawer et al, 1961). Vincent and Urist (1961) reported that osteoporosis occurred in a high 4^20 FRIST, MACDOXALD, MOSS, AND SKOOG Fig. 13A. Radiograph showing the head and neck of the femur (natural size) excised following a bilateral intertrochanteric fracture in a 78-year-old woman with severe osteoporosis. The split in the superior margin of the head was produced by insertion of a hip skid. Note the accentuation of the vertical trabecular markings. The thin cortex and hollow neck are indications for ex- cisional surgery or bone pegs, rather than internal fixation with metallic ap- pliances. Fig. 13B. Microradiograph of the superior cortex of the neck of the femur in the case shown in Fig. 13A. Note the hypermineralized avascular bone on the right and the attenuated discontinuous cortex in the center. ( X approx. 75.) RAREFYING DISEASE OF THE SKELETON 421 Fig. 13C. Photomicrograph of the center of the cortex of the neck of the femur shown in Figs. 13A and 13B. The visceral capsule or vestigial periosteum of the hip is shown at the top. The dense-staining bone matrix lies beneath it. The lacunae are either empty or partly filled with pyknotic nuclei and ab- normal osteocytes (arrow), (x approx. 400.) percentage of males with chronic debihtating diseases, but the condition was invariably less severe than in females. The possibility of a hereditary or genetic factor associated with osteoporosis of the 422 URIST, MACDONALD, MOSS, AND SKOOG spine has long been suspected bnt has yet to be subjected to a sys- tematic investigation. Densitometric Methods of Diagnosis Renewed effort has been made to perfect densitometry to enable physicians to diagnose osteoporosis at an early stage of the disease and evaluate the efficacy of dijfferent methods of treatment. Nordin et al. ( 1962a ) devised an interesting technic in which lateral tomo- grams were prepared of the patient and a dissected specimen of the lumbar spine was placed next to the subject. The film images were fed through a Laurence-Locarte recording densitometer. The ratio of the densities of the subject and the standard was called the relative vertebral density. It was concluded that the codfish spine occurred when the vertebral body was less dense than the inter- vertebral discs. Mavo ( 1961 ) made a similar survey and concluded that the individual could lose a relatively large amount of bone mineral and still fall within the range of normal individuals, and that the early diagnosis of osteoporosis is limited by these considera- tions. Such limitations did not applv to the use of quantitative methods for measuring changes in the bones with time in the same individual. Doyle (1961) surveved the proximal end of the ulna, and Wagner and Schaaf (1961) examined the thumb, and corrob- orated these results. Vose ( 1962 ) described microradiographic changes in cortical bone. The mineral content varied from 0.94 to 1.19 gm of hydroxyapatite per cc of bone in normals and from 1.12 to 1.50 in osteoporotics. Thus, while the bone mass was reduced in volume in osteoporosis, the bone tissue that persisted was hypermineralized. This incre- ment was apparently in amorphous plugs in blood vessels and in the interstitial lamellae of old deposits. In view of the technical diffi- culties, physical variables, and other complexities of con\'entional x-rays for measurement of bone densit\', it is now desirable to search for entirely different principles for obtaining quantitati\ e informa- tion about osteoporosis. Jacobson (1962) is testing dichromatic ab- sorption radiography for calcium in mineralized tissues; this method gives greater contrast than an ordinary radiograph (Jacobson, 1953). RARKFYIN(; DISEASE OF THE SKELETON 4''23 Plujsiologic Aging of Bone Structural physical and chemical alterations occiu- in the human skeleton in the course of chronological or physiologic aging. Sher- man and Selakovitch (1957) noted that osteocytes begin to disap- pear from the outer lamellae of haversian bone during adolescence, and the number of empty lacunae increases after the 3rd decade. These changes were attributed to circulatory insufficiency and physi- ologic necrosis of bone in old age. Atkinson et al. (1962) measured the dry weight and displacement volume of blocks of bone, w/v = D, and described decrease in thickness and density of cortical bone with increasing age in both sexes after 50; the metaphyseal de- creased faster than the diaphyseal cortex. Sissons (1962) noted that in the aged many osteons ceased to grow when they were half closed, and confirmed Amprino and Engstrom's ( 1952 ) observation that the time between osteon formation and completion of minerali- zation increased with aging. In aged individuals the osteon became rapidly mineralized to 90 per cent of the capacity, but a larger pro- portion of the total bone was always less than 100 per cent mineral- ized. It was easy to distinguish the incompletely mineralized old boile from new bone because the osteon in which old bone is found contains a sclerotic inner edge before it is half closed. Lindahl and Lindgren ( 1962 ) observed that in 60 subjects be- tween 14 and 91 years of age, the decrease in bone density was parallel in the upper end of the tibia and the vertebral column. Samples of bone were treated with hot water to remove marrow and with xylene to remove fat, and cut into cubes to measure volume of bone tissue; the density of the sample was the weight per unit volume of the cube. Irrespective of age, osteoporosis was present when the density was 0.20 mg per mm^ of spongy bone. These observations confirmed the conclusions of Trotter et al. ( 1960) that bone density diminishes with age. This is a generalized process, but the rate differs in different bones, in males and females, and in the white and Negro races. According to Morgan et al. (1962), the decline was more pronounced between ages 50 and 65 in women, and usually did not occur in men until age 70. In women, the 424 URIST, MAfDONALD, MOSS, AND SKOOG severity of the radiographic changes had no relation to the length of time following the menopause (Moon and Urist, 1962). The relation between the time-dependent process of aging and osteoporosis was obscure. Groen et al. ( 1960 ) described alveolar atrophy in osteoporotic subjects and concluded that senilitv was not the cause. Berglund and Lindquist ( 1960 ) described osteo- porosis or osteopenia in adolescents. The appearance of the dis- order in rare instances in young men, the so-called idiopathic oste- oporosis (Hall and Kennedy, 1962), also suggested that a specific factor, in addition to physiologic aging, endocrinopathy, and nutri- tion, had to be taken into consideration. Degradation of the Collagen of Bone An interesting theory appeared in 1962 to propose that osteoporo- sis is primarily a degenerative disease, rather than an endocrine or metabolic disorder. Little et al. ( 1962 ) examined normal and osteoporotic bone tissue bv electron microscopv and x-rav diffrac- tion of the fibrous proteins. Bone was defatted with alkali and decalcified with citric acid in order to enhance the solul:)ilitv of collagen in acid buffer, and, thereby, separate soluble eucollagen. This was reprecipitated in pure form by dialysis. The nonsoluble collagen and ground substance were analyzed together as solid residue. The x-rav diffraction pattern of the solid residue suggested that the crystal structure of the collagen was altered or degraded. Since these alterations were not seen in bone from nonosteoporotic aged individuals, or in dead bone, and resembled degradation pro- duced bv irradiation damage, it was suggested that an active degrad- ing agent may be involved. The patterns also included the lines of a fatty acid, possibly myristic acid. Electron micrographs of bone showed loss of the normal architecture and fibrillar form of the collagen in various areas of osteoporotic bone. Little et al. (1962) suggested that osteoporosis was the result of a local agent emanat- ing from osteocytes rather than a general chemical effect. The com- position of the mineral phase of bone was the same in osteoporotic and nonosteoporotic individuals of comparable age, and the apatite crystallites probably played only a passive role (Holdeway et ah, 1962). The degradation of collagen by irradiation and starvation. RAREFYING DISEASE OF THE SKELETON -i^O and the inhibition of conversion of sokible to insoluble collagen in irradiated experimental animals (Gerber et ah, 1962), illustrated some of the possible derangements that could occur in aging and osteoporotic bone. The level of the urinary hydroxyproline has also been interpreted to reflect either the process of degradation or turnover of bone collagen. Bastomsky and Dull (1962) described hydroxyprolinuria in rats treated with parathyroid hormone and triiodothyronine. Klein et al. ( 1962 ) found a 60 per cent increase in a patient with hvper- parathyroidism with bone disease, as compared with no increase in two patients without bone disease. Raveni et al. (1962) measured total and free urinary hydroxyproline in normal and osteoporotic patients and concluded that an increase in the free fraction reflects increased destruction or catabolism of bone in this disorder. Ibsen et al. ( 1963 ) evaluated a proline tolerance in rabbits with healing fractures and concluded that bone formation increased the rate of proline utilization. Further investigations may establish proline metabolism as a useful indicator of bone matrix metabolism in health and disease. Alterations in the Mticoproteins of Bone Casuccio et al. ( 1962 ) described abnormalities of the osteocytes in bone tissue and a decrease in several fractions of the mucopoly- saccharides of bone matrix in the skeleton of aged and osteoporotic subjects. The perilacunar sheath of Rouget-Neumann was accen- tuated, metachromasia was absent, nuclei were pyknotic, the canalic- uli were reduced in number and showed terminal club-shaped thickenings. The osteocyte sheaths were shrunken, shriveled, or absent. The mucoprotein cylinders that normally fill out the canalic- uli were absorbed and the osteocytes retracted their cytoplasmic processes. This suggests intralacunar resorption or osteocytolysis as described by Heller et al. ( 1950 ) . Casuccio ( 1962 ) observed a 50 per cent decrease in protein nitrogen but only a 6 to 9 per cent de- crease in inorganic mineral in bone tissue in osteoporotic patients; the bones of aged individuals without osteoporosis were similarly altered, although to a lesser degree. Four fractions of the mucopoly- saccharides of bone were influenced by both aging and osteoporosis 426 URIST, MACDONALD, MOSS, AND SKOOG as follows : A versene-soluble component, rich in galactosamine, was reduced 50 per cent, and the galactosamine/glucosamine ratio was changed from 3:1 in young to 1:8.1 in aged or osteoporotic subjects; a versene-insoluble component, consisting of collagen and interfibril- lary substance, was altered so that the hydroxyproline was relatively high and the amino sugar content was relatively low; an alkali- soluble component, consisting of collagen, was unremarkable; an alkali-insoluble fraction, consisting of a small amount of uncharacter- ized residue rich in carbohydrate, did not change with age. These observations suggested that the interfibrillary mucopolysaccharide gradually decreased progressively with aging and more abiaiptly with osteoporosis of bone. Bertolin and Greco ( 1962 ) measured an in- creased daily rate of excretion of a mucopolysaccharide with a high galactosamine/glucosamine ratio in the urine of patients with osteo- genesis imperfecta and osteoporosis. These observations substan- tiated the findings of Sobel et al. (1954) and Sobel (1960) that the rate of deposition of hexosamine-containing substances decreases in the skin and bones of rats with the progress of growth and aging. These changes were probably unspecific, but it would be of interest to know whether they occur in osteoporosis associated with endog- enous and exogenous hyperadrenal corticoidism, hyperthyroidism, acromegaly, and other endocrinopathies. Nutrition and Metabolism of Calcium Though a calcium-deficiency disease has never been clearly estab- lished as an entitv in adult human beings (Nicolaysen, 1960), it is generally assumed that low calcium intake is one of the factors in the development of osteoporosis (Malm, 1961). The hypothesis rests on the observation that some patients who are in negative calcium balance on 0.8 gm per dav will retain substantial amounts on an intake of 2.0 gm. Nordin (1962fl) reasoned that if osteoporo- sis is due to increased bone resorption, one would expect to find reduced calcium intake, malabsorption, or high excretion in osteo- porosis. A dietarv survev revealed that 128 patients with osteopo- rosis had a significantly lower intake of calcium and protein than 128 patients with normal bones (Nordin, 1962/;). As many as 15 per cent had steatorrhea, but this disorder did not impair the ab- RAREFYING DISEASE OF THE SKELETON 4''27 sorption of calcium when the intake was raised to 2.0 gm per day. Some patients with osteoporosis had p\ elonephritis and a "calcium leak" that caused negative calcium balance (Nordin, 1962/;). Of 29 balance studies on 17 patients, 5/17 were in negative calcium bal- ance, and 12 17 were in slightly positive calcium balance. Inas- much as the calcium lost in sweat was not measured in balance studies, it was assumed that all the patients were in negative calcium balance. The contention was, first, that osteoporotics were not able to reduce their urinary calcium on low calcium diets, but had the ability to retain calcium on high calcium diet (Bhandarkar and Nordin, 1962 ) ; second, that the cause of the disorder was prolonged negative calcium balance in subjects with higher than normal cal- cium requirement or inability to adapt to inadequate intake. Some severe osteoporotics, however, were able to adapt to low calcium intake. In any case, the condition was relieved by dietary supple- ments of calcium, but x-ray was not sensitive enough to detect the new bone that was presumed to have been added (Nordin, 1962c). Bhandarkar et al. ( 1961 ) fed 7.5 ^tc of Ca^ ' or 5 /xc of Ca^'' to meas- ure calcium absorption, and noted that the majority of patients with osteoporosis absorbed calcium normallv. Kinetic studies were inter- preted to show that the rate of bone accretion in osteoporotics was also the same as normal (Nordin et ah, 1962/?). Fraser (1962) divided patients with osteoporosis into three groups, Cushing's svndrome, at\'pical or idiopathic, and tvpical or postmenopausal, and corroborated Nordin's contentions. However, sex hormones were prescribed in addition to calcium supplements for senile osteoporosis. Haas et al. ( 1962 ) measured retention of an infusion of calcium in 36 osteoporotics and 12 normal subjects, and concluded that the lowest levels occurred in Cushing's syn- drome, castration, and idiopathic osteoporosis. In the past, some investigators have assumed that typical osteo- porosis represents chronic low-grade hyperparathyroidism of bone. The calcium deficiency hypothesis ultimately leads to such an as- sumption for an explanation of the bone tissue deficit. Gordan et al. (1962) rejected hyperparathyroidism secondary to malabsorption of calcium as a cause of osteoporosis; patients with hyperparathy- roidism had only 35 to 75 per cent tubular reabsorption of phos- 4'28 I'RIST, MACDONALD, MOSS, AND SKOOG phate, whereas patients with osteoporosis had 80 to 100 per cent. Goldsmith et al. ( 1962 ) employed an infusion of 180 mg of calcium in the form of calcium glucoheptonate and observed that phosphate excretion was unchanged in hyperparathyroidism, but lowered in other bone conditions. Frost ( 1961 ) measured the number of How- ship's lacunae per unit yolume of bone in undecalcified thin ground sections of cortical biopsies of 12 cases of osteoporosis, and con- cluded that osteoclasis was four times greater than in normal sub- jects. Though bone biopsies and tests of intestinal tract and kidnev excretion of calcium and phosphate were extremely yaluable, it was not possible to diagnose low-grade hyperparathyroidism with- out an exploration of the neck and a positive identification of en- larged parathyroid glands (Henneman et ah, 1958). Adrenal Cortical Hormones and Intermediari/ Metabolism of Protein Postmenopausal or senile osteoporosis is radiographically and histologically indistinguishable from osteoporosis associated with Cushing's syndrome, either the endogenous or the exogenous form (Murray, 1961). The mechanism of depletion of the bone tissue by the action of glucocorticoid hormones has been attributed to inhi- l:)ition of formation of osteoblasts with relatively little effect on the normal bone-resorbing activity of osteoclasts. Suppression of bone formation alone explains the decrease in absorption of calcium from the intestinal tract that is generally found in Cushing's syndrome, but direct effects have also been reported by Bhandarkar et al. (1961). Children and women after the menopause were more sus- ceptible than men to bone destruction by cortisone (Schlesinger et al., 1961). Anabolic steroids protected, but vitamin D increased, the vulnerability of the skeleton (Bassett et al., 1960). One of the intriguing aspects of the mineral metabolism was the effect of adrenalectomy. Patients with endogenous Cushing's syn- drome were relieved of a backache and further fractures of the spine, vet the radiographic picture was relatively unchanged follow- ing this operation (Martinez and Greenblatt, 1962). Myers (1962) reviewed the literature on calcium metabolism in adrenalectomized animals, including patients with Addison's disease, and concluded RAREFYING DISEASE OF THE SKELETON 429 that the adrenal cortex has a dual effect upon calcium metabolism. Excess glucocorticoids induced negative calcium balance, and in- sufficiency caused change in calcium equilibrium characterized by hypercalcemia. The latter was independent of the bone, and the mechanism was different from that produced bv parathyroid hor- mone. Furthermore, this hypercalcemia persisted in parathvroidec- tomized animals, and cortisone in physiologic amounts restored the serum calcium to normal (Kolb and Cohen, 1962). Walser et al. (1962) described 58 samples of the serum of adrenalectomized dogs and attributed the hypercalcemia to a combination of factors, such as hyperproteinemia, due to hemoconcentration; increased affinity of calcium for protein, due to hyponatremia; increased citrate levels, as much as three times normal; and increased amounts of uniden- tified calcium complexes. The effective or free calcium ion concen- tration was not altered bv adrenalectomv. Urist ( 1960/? ) reported the case of a 78-year-old woman with severe osteoporosis treated by adrenalectomy. The glands were low in weight, as is usual in persons of advanced age, and there was atrophy of the zona fasciculata of the right, but not of the left, adrenal cortex. Urist and Vincent ( 1960 ) observed that the de- crease in urinary ll-deoxy-17-ketosteroids that occurs in aged women w^as more extreme in women with severe osteoporosis. Cald- well (lQ62b) examined the adrenal glands in 31 autopsy subjects, including 8 cases of senile osteoporosis, and found no evidence of involution or atrophv, and possiblv slightlv greater than normal adrenal cortical cell activitv. These observations suggested that further observations upon the adrenal cortical hormone metabolism in castrates and aged patients with osteoporosis may be of consid- erable value. For some unknown reason, adrenalectomy increased the sensitivity of animals to the side effects of glucocorticoid hor- mones. Both the adrenals and the gonads are markedly diminished in size in birds with osteoporosis (Urist, 1958). Whereas the influ- ence of adrenal hypercorticoidism upon osteoporosis was clearly demonstrable, the effect of endocrine deficiency disorders was quite obscure, and an antiosteoporosis factor produced by the adrenal cortex, the liver, or the skeleton itself, but dependent upon normal endocrine gland function, had to be considered (Urist, 1960b). 430 URIST, MACDONALD, MOSS, AND SKOOG Teng et al. (1961<7, 1961/;) described osteoporosis in 7 children with a systemic disorder of protein metaboHsm caused by tumors, cirrhosis, and von Gierke's disease. As in patients with senile osteo- porosis, the urinary excretion of calcium was decreased, owing to the low rate of turnover of bone tissue. Apparently skin, muscle, and bone had a low priority rating and were deprived of their supply of proteins to fulfill the needs of organs of vital bodily functions. A 60-year-old osteoporotic woman, who had only 50 per cent of the normal bone mass, was calculated to have had a 10 per cent deficit in bodv protein. Adrenal 11-oxy corticosteroids decreased the rate of protein svnthesis, or increased the rate of protein breakdown. Hence, metabolic reactions, nutritional factors, and hormonal influ- ences were interdependent, and the rate of differentiation of oste- oblasts and deposition of bone must have been regulated by each and all. Skeletal Kinetics The recent literature on measurement of turnover of bone tissue by tracer techniques with radioisotopes, stable strontium, and tetra- cycline has alreadv been reviewed in detail (Bauer et al., 1961; Eisenberg and Gordan, 1961; Urist, 1962; Urist et al, 1962). When the results of two or three tracer technics were correlated with the results of conventional chemical balance studies and bone biopsy, we concluded that the use of tracer dilution formulas measured the activity of less than 1 per cent of the total bone tissue (Urist et al., 1962) and gave no information about the volume of the total bone mass. Less than 1 per cent of the skeleton was capable of turning over 0.5 to 1.0 gm of calcium, the daily rec|uirement of an adult human being. Therefore, the defect in osteoporosis was in the proc- ess of storage of calcium in structural bone, not in the activity of reactive or metabolic bone. Nevertheless, Eisenberg and Gordan ( 1961 ) identified two divergent groups in 26 patients with osteo- porosis. One group had small miscible pools and low bone formation rates, as, for example, osteoporotics of advanced age or with Cush- ing's SMidrome. Another group had large miscible pools and rapid bone formation rates, as, for example, in thyrotoxicosis and acro- megaly, in which there is excessive bone resorption. Thus, in cases of RAREFYING DISEASE OF THE SKELETON 431 extreme alteration in the structure of the bone tissue, the percentage of the skeleton that was reactixe or exchangeable with tracer sub- stances was proportional to the \ olume of the percentage of skeleton that was relatively nonreactive or nonexchangeable. Our impression was that in individuals in earlv stages of the disease, in which quan- titative data would have been most valual^le for diagnosis and treatment, it was not possible to distinguish normal from osteo- porotic individuals. Further development of the mathematical treat- ment of radioisotope kinetic data and the osteogram of MacDonald (1960) may increase the value of measurements of skeletal turn- over with Sr^^, Ca^', and other isotopes in patients with local, as well as systemic, bone disease (Dow and Stanbury, 1960; Green- berg et al., 1961; Bauer et al, 1961). Fractures We assumed in previous reports that physiologic aging or atrophv of the spine did not produce gross deformities or spontaneous col- lapse of vertebrae, and that radiographic evidence of multiple com- pression fractures, ballooned discs, and thin cortex distinguished osteoporosis as a disorder (Urist et al., 1959; Urist, 1960Z?; Urist and Moon, 1961 ) . The degenerative changes in the bone tissue were the same in many respects in physiologic aging and in osteoporosis (Table V), and the only difference could have been in the rate of decline of the bone mass. In both conditions, the ratio of weight to volume or densitv declined from 0.3 to 0.2, but in osteoporosis the decline occurred at an earlier age and a faster rate (Fig. 12). Bone failure or spontaneous fractures of the spine occurred when the density fell below 0.15, or 50 per cent of that of the normal adult human skeleton at age 20. It was interesting in this connection that many patients with osteoporosis reported sudden and recent weight loss. The high incidence of fractures in the aged has been attributed to lower bone density, but the correlation with osteoporosis is in- complete. The incidence of severe osteoporosis is 26 per cent in healthy aged women (with average age 85), but it is 76 per cent in women, average age 73, with all types of fractures of the hip (Urist et al, 1959; Urist, 1960Z7; Urist and Moon, 1961). Stevens 4'>*-2 URIST, MACDONALD, MOSS, AND SKOOG et al. ( 1962 ) noted that fractures of the hip are not more common in osteoporotics until after 70 years of age, but that intertrochanteric fractures are more typical of osteoporosis than other tvpes. Although the aged are more susceptible to fractures because of the physiologic decline in bone density, it is clear that incidence of fracture increases further in individuals in older age groups with severe osteoporosis. Previously, Buhr and Cooke (1959) attributed fractures of the hip to "senile matrix of old age." Little et al. (1962), however, con- tended that the bone matrix was degraded or defective in severe osteoporotics, as compared with nonosteoporotic individuals of com- parable age. Osteoporosis in Experimental Animals The discovery of osteoporosis in the White Leghorn hen ( Urist and Deutsch, 1960a) was followed by the realization that the loss of bone mass is a manifestation of cage layer fatigue (CLF) (Urist, 1960« ) . Urist ( 1959 ) had previously observed that hypertrophy of the adrenal cortex developed when normal ^^'hite Leghorns were fed a calcium-deficient diet. But the same variety of laving hens with CLF had atrophied adrenal glands, and the relationship of adrenal hypercorticoidism to avian osteoporosis has not been clari- fied. The problem is the same as with osteoporosis in man. Severe osteoporosis was produced in hens by administration of ACTH (Urist and Deutsch, 1960/?) or cortisone (Urist and Deutsch, 1960c). Roosters were highly resistant; caponized birds were susceptible to bone-destructive properties of these agents. The skeleton of the rooster initially possessed a larger mass of bone than that of the capon or hen, and required a longer period and a larger amount of bone resorption to develop the disorder, but other factors may have been involved. It was not possible to deplete the skeleton of the rooster to the extent that occurs in the hen with CLF. Taylor et al. (1962), Taylor (1962), and Bell and Siller (1962) corroborated our observation that osteoporosis is the basic lesion in CLF, and proposed that the cause was genetic and the mechanism was hyperactivity of the pituitary. Bell and Siller (1962) described 40 cases in Brown Leghorn hens, and noted hyperosteoclasis of cortical bone and atrophy of muscle as well as bone. Birds with RAREFYING DISEASE OF THE SKELETON 433 CLF fed a low-calcium diet continued to lay heavily and produced well calcified eggshells. Normal birds produced one or two thin- shelled eggs and discontinued ovulation when the intake of calcium was restricted. Taylor et al. ( 1962) suggested that the pituitary cut- off mechanism was generally verv efficient, but lines bred for inten- sive laying continued to ovulate when transferred to a low calcium intake, and it was these individuals that were susceptible to CLF. Restriction in cages and insufficient physical activity was an impor- tant factor in CLF; some birds recovered simply by being released to exercise ad libitum. Thus, in birds as in humans, calcium-deficient diet and sedentary life were precipitating factors in the pathogenesis of osteoporosis, but the main cause was unknown, although probably genetic in origin. Silberberg and Silberberg (1962) reported that senile osteoporosis in mice is associated with genetic factors. The DBA strain was more susceptible than other inbred strains to a disorder characterized by thin, rarefied, deformed bones appearing between 24 and 36 months, an advanced age for mice. Young rats fed diets with less than 0.1 per cent calcium and some vitamin D develop osteoporosis that is cured by treatment on high calcium intake (McClendon et al., 1962). During adult life, rats retain their bone mass more tenaciously than other mammals, and are generally resistant to osteoporosis (Nicolaysen, 1960). Despite continuous administration of cortisone, in doses that produce osteoporosis in birds, mice, rabbits, and man, rats will retain the bulk of the calcium in the skeleton. Caldwell ( 1962a ) noted that cortisone inhibits bone growth, but the effect in rats is temporary, and these animals develop resistance to the hormone. Furthermore, by measuring the amount of calcium per gram of wet weight of whole bones, Caldwell (1962a) observed that adrenalectomy and orchiectomy destroyed the resistance to cortisone, and high calcium intake did not prevent or cure the osteoporosis. Urist, MacDonald, Johnson, Deutsch, and Vincent (unpublished) performed similar experiments and measured the capacity of the adreno-orchiectomized rat to retain an injection of Sr''"', but found no significant abnormality of bone. Nevertheless, adrenalectomized animals are more generally sensitive than normal animals to cortisone, and further experiments on osteoporosis in 434 URIST, MACDONALD, MOSS, AND SKOOG cortisone-treated adrenalectomized rats would be of great interest. The possibility of an antiosteoporosis factor present in normal adult males, and essential for bone retention during aging, requires inten- sive investigation (Urist, 1960/^, 1962). The action of adrenal cortical hormone upon bone in rats was attributed to inhibition of absorption of calcium from the intes- tinal tract or physiologic calcium deficiency. Malabsorption was assumed to stimulate the parathyroids and lead to osteoporosis. Clark and Smith (1962), however, noted hypercalciuria in parathy- roidectomized rats, the same as in normal rats, and Clark and Roth (1961) concluded that hvdrocortisone decreased the reabsorption of calcium and phosphate by the kidney tubules. Under these condi- tions, weanling rats increased the density of their bones, but grew at a rate so much less than normal that presumablv the over-all re- tention of calcium was less. The reaction of the skeleton has long been known to be entirely different before and after a rat reaches maturity and the plateau of growth, and the endocrine relations were difi^erent in many respects in rats, other rodents, and higher mammals. Calcium deficiencv was a contributing factor in the de- velopment of osteoporosis; Storey (1961) fed low-calcium diets to rabbits treated with cortisone and described osteoporosis associated with extensive osteoclastic activity and secondary hyperparathyroid- ism. How endocrine interrelationships influence osteoporosis in man is not known. Schlesinger et al. (1961) observed that adrenal steroids produce osteoporosis in children and postmenopausal women more frequently than in men, and generally from treatment on lower doses and in less time. It could have been argued that the reason is simply that young adults have more bone to begin with than children and senile women. However, the subject was considerably more complex; 85 per cent of the patients of comparable ages and skeletal conditions did not develop osteoporosis on comparable steroid therapy. Other, as vet unknown, factors were evidently of primary importance. Kao et al. (1962) reported that the glycoprotein decreased in bone and increased in costal cartilage in rats in the process of aging between 1 and 36 months. Previouslv, Sobel et al. ( 1954), Sobel and RAREFYING DISEASE OF THE SKELETON 435 Marmorston ( 1958 ) , and Sobel ( 1960 ) observed tliat the hexos- amine/collagen ratio decreased in the femora of rats with aging. In view of recent reports of decrease in galactosamine-rich muco- proteins of bone matrix of patients with osteoporosis (Casuccio, 1962), it is now necessary to assume that quahtative, as well as quantitative, changes occur in the skeleton in aged and osteoporotic bones. The nature of the relation between the time-dependent proc- ess of aging and osteoporosis has not been defined in meaningful terms. Treatment A record of the progress of a control series of cases of osteoporosis over a long period of time is essential for evaluation of any one, or any combination, of the man\' forms of treatment that have been prescribed for the disorder. It was possible that in manv patients the skeleton was reduced within a few years to 70 or 50 per cent of normal bone mass, and there was no further progress of the disorder over a period of 5 to 20 years. This possibilitv created uncertainty about whether therapv had had even a preventive effect either on further loss of bone or on the occurrence of spontaneous fractures (Urist, 1962). Though sex hormones (Gordan, 1961; Vincent and Urist, 1961), anabolic steroids (Kuzell et ah, 1962), fluoride (Rich and Ensinck, 1961), calcium supplements (Nordin, 1962a, 1962i>), or combinations of these agents (Lichtowitz et al., 1962; Riccitelli, 1962) produced svmptomatic improvement, it was reasonable to assume that placebo effects were responsible in many cases. By means of Ca^'' kinetic studies, Bhandarkar et al. ( 1961 ) claimed to produce an increase in rate of bone formation from calcium supplements, and Bronner et al. ( 1962 ) measured acceleration of rate of bone uptake following estrogen therapy. Improvement in radiographic density or increase in the mass of bone tissue in biopsies in series of cases of osteoporosis is essential for proof of a curative effect, but this has not been accomplished (Fig. 14). Discussion Present knowledge of osteoporosis consists of a collection of a large quantitv of negative and a small quantity of positive informa- 436 URIST, MACDOXALD, MOSS, AND SKOOG 14% 23% 22% 32% 9% NON-FAT MILK HIGH PROTEIN DIET CORSET, EXERCISE CALCIUM GLYCERO- PHOSPHATE VITAMINE D SEX HORMONES ANABOLIC AGENTS & NEW AGENTS • + - PLACEBO PSYCHO- THERAPY RESEARCH TO UNCOVER AN ANTIOSTEOPOROSIS OR BONE RETENTION FACTOR Fig. 14. Diagrammatic illustration of the treatment recorded in the hospital charts of 100 patients with severe osteoporosis over a period of 10 years. Nine per cent changed their diet to include a higher intake of dairy food and protein; 32 per cent made some effort, in addition to the alterations in diet, to increase the amount of physical exercise, and also to wear a corset for comfort; 22 per cent, especiallv those intolerant of dairy food, received calcium supple- ments, chiefly calcium glycerophosphate, and maintenance doses (1000 lU) of vitamin D; 23 per cent received sex hormones or anabolic steroids in ad- dition to the above measures; 14 per cent received tranquilizers, muscle re- laxants, vitamin B, and other drugs that had placebo effects in that there was no detectable influence upon the skeletal system. No improvement in bone density was observed by serial radiographs of the spine following any one, or any combination, of these treatments. tion. The etiology of osteoporosis is not known. The disorder ap- pears in its severest form much more frequently in females than in males. Sedentary life after the menopause seems to be a factor. Men and women who do heavy labor usually have spinal osteo- phytes (spondylosis) and high bone density, whereas sedentary individuals with osteoporosis generally are free of spondylosis and have low bone density. Many patients with this disorder have the normal capacity to produce new bone, but by some insidious mech- anism all lose a large percentage of their bone mass in a relatively RAREFYING DISEASE OF THE SKELETON 437 short time, perhaps one or two years, generally at some interval between ages 50 and 65. Recent writers contend that polysaccharide is lost, that collagen is degraded, and that the disorder is basically a degenerative process. The morphology and biochemistry of the changes in bone associated with osteoporosis resemble in most respects abnormalities attributable to the physiologic process of aging. The disappearance of osteccytes and the occurrence of patches of necrotic bone are striking. It appears that in the aged, and more so in the osteoporotic, the movement of calcium from the gut to the blood plasma is slower than normal, movement of calcium from bone to extracellular fluid is accelerated, and movement of calcium from plasma to feces and urine is normal or high. Emphasis is often placed upon the low calcium intake and insufficient peptides (from protein) in the diet in osteoporotic women. Our view is that negative calcium and nitrogen balance mav reflect the process of resorption of dead bone and mav be the result rather than the cause of osteoporosis. Bone mineral is among the products of breakdown of structural bone, and this is the source of the excess calcium excreted in both the urine and the feces. A low level of osteogenetic activity is apparent during the later stages of the process of osteon formation, when reactive new bone is normallv converted to structural bone. The movement of calcium from gut to plasma to bone therefore declines in rate, and significantlv greater than normal amounts of calcium move from the plasma to the urine and feces. In young adults, the daily net absorption of calcium, regardless of the total amount (800 mg) in the diet of an adult, is approximately 150 mg (Dent, 1956). Though radioisotope kinetic studies reveal the internal movement of Ca++ and a high rate of turnover of bone in the acute stage of osteoporosis, the low dailv net absorption of calcium remains to be accuratelv measured in the chronic stages. Indeed, in the aged osteoporotic, to judge from the total amount of calcium in 24-hour samples of urine, this may be as little as 50 mg. Twenty-five years ago, the emphasis in research on osteoporosis was on the endocrine control of calcium metabolism. The occur- rence of the disorder in patients with acromegaly, hyperthyroidism, diabetes, and hypogonadism was attributed to endocrine imbalance. 438 VRIST, :MArDOXALD, MOSS, AND SKOOG It is also possible that endociinopathy initiates nonspecific degen- erative changes and premature aging of bone tissue. Hyperparathy- roidism has specific effects on bone and causes osteoporosis b\ osteoclasis and osteoc\ tohsis, and Cu.shing's svndrome inhibits the proliferation of osteoblasts and produces bone resorption without osteoclasts, but these conditions may also accelerate the aging process and deterioration of osteocytes. The retention of bone and control of the rate of reduction of bone mass, however, could be determined by an antiosteopowsis factor. This seems to be associated with such phenomena as maintenance of blood flow, completion of osteons and development of structural bone, androgen and protein biochemistry, effects of mechanical stimuli of phvsical exercise upon bone tissue, liver metabolism, and the intracellular phvsiologv of the bone cells. The point of view set forth above suggests the possibilit\' that treatment of osteoporosis by exercise, with improved diet and injec- tions of sex hormones, may stimulate the rate of capillary circulation of bone and thereby slow the process of aging of bone tissue. Meas- urements of urinarv products of bone breakdown such as hydroxy- proline and mucopolvsaccharides are now being made to estimate the rate of aging of bone and to evaluate current methods of treatment of osteoporosis. Summary 1. Two cases have been added to six previously reported cases of osteoporosis, investigated by metabolic balance, radioisotope, and tetracvcline technics. The data suggest that the metabolic functions of the skeleton are performed by the 1 per cent of the bone that is reactive or exchangeable, and that this is not significantly abnormal in patients with osteoporosis. The deficiency is in the malfunction of the 99 per cent that is the nonreactive, nonexchangeable, stable, or structural bone. 2. Twenty-five survivors from a group of 100 previouslv reported aged women (average age 85) were reexamined after 5 years. Three cases had osteoporosis (as previouslv determined by collapsed ver- tebrae), further fractures, and significant progress of the disease. RAREFYING DISEASE OF THE SKELETON 439 Twenty-two cases, previously free of fractures, did not acquire collapsed vertebrae, and had onlv a physiologic rate of decline in bone density. These observations suggested that spontaneous col- lapses of vertebrae are indicative of structural hone failure, and this is the cardinal sign of osteoporosis. Bone failure distinguishes the disorder from physiologic bone atrophy of the aging. Bone fail- ure occurs when bone mass declines more rapidly than normal to 50 per cent of the quantity that is found in voung adult women. 3. The morphologic and biochemical indices of aging are qualita- tively the same, but quantitatively more extreme in osteoporotics than in nonosteoporotics. Many osteons stop growing and calcifying and develop a sclerotic inner margin before they are one-half closed. Bone cells retract, enlarge the lacunae, and disappear in the outer and interstitial lamellae that are farthest from the blood vessel. The matrix becomes more basophilic, metachromatic, and irregular stain- ing. These morphologic and histochemical changes are unspecific, but are approximately 30 to 50 per cent more extreme during middle- aged life in patients with severe osteoporosis. As much as 50 per cent of the bone tissue in the tibia mav be dead. 4. The recent literature may be interpreted to propose a unifying hypothesis that extreme dietary deficiencies of calcium and protein, various endocrinopathies, and many degenerative vascular system disorders produce premature or accelerated aging of bone, and that osteoporosis appears as a result. The conditions responsible for reten- tion of bone or the maintenance of an antiosteoporosis factor that regulates the physiologic rate of aging of the skeleton are possibly related to the genetic constitution of the individual, but otherwise are entirely unknown. 5. A regimen of vigorous constitutional exercises, including deep breathing, bicycling, swimming, and hiking, should be added to medical or orthopedic treatment to improve the circulation and retention of bone in the aging and osteoporotic skeleton. Acknowledgments. These investigations were supported by Contract No. AT(04-1)-GEN-12 between the Atomic Energy Commission and the University of California Medical Center, Los Angeles, and by grants-in- aid from the National Institutes of Health, United States Public Health 440 URIST, MACDONALD, MOSS, AND SKOOCi Service (No. A-3793); the Easter Seal Foundation; the Society for Crip- pled Children; and the Ayerst Laboratories Inc. Dr. Moss is an orthopedic Research Fellow, The writers announce their indebtedness and present their thanks to James A. Dingwall, M.D., and the Squibb Institute for Medical Research, who supplied polysaccharides extracted from calf bone; and to the mem- bers of the Endocrinology Study Section of the National Institutes of Health, United States Public Health Service, who provided human growth hormone for these investigations. References Amprino, R., and Engstrom, A. 1952. Studies on x-ray absorption and diffraction of bone tissue. Acta Anat., 15, 1-22. Atkinson, P. J., Weatherill, J. A., and Weidman, S. M. 1962. Changes in density of human femoral cortex with age. /. Bone and Joint Si/rg., 44B, 496-502. Bassett, S. H., Figueroa, W. C, Tuttle, S. C, and Jordan, T. 1960. Meta- bolic studies in Cushing's syndrome. The effect of steroid withdrawal, androgen, and vitamin D on calcium, phosphorus, and nitrogen bal- ance. Clin. OrtJiopaed., 17, 304-332. Bastomsky, C, and Dull, T. 1962. Urinary hydroxyproline response in in- tact and parathyroidectomized rats treated with triiodothyronine. Clin. Researclu 10, 399. Bauer, G. C. H., Carlsson, A., and Lindquist, B. 1961. Metabolism and homeostatic function of bone. In Mineral Metabolism (C. L. Comar and F. Bronner, editors), Vol. 1, Part B, pp. 609-677. Academic Press, Inc., New York. Bell, D. J., and Siller, W. G. 1962. Cage layer fatigue in Brown Leghorns. Research Vet. Sci., 3, 219-229. Berglund, C, and Lindquist, B. 1960. Osteopenia in adolescence. Clin. Orthopaed., 17, 259-264. Bertohn, A., and Greco, B. 1962. Prime indagini sull' escrezione urinaria dei mucopolisaccaridi in soggetti osteopsatirotici: Osservazioni ezio- patogenetiche. Boll. soc. ital. hiol. sper., 38 (in press). Bhandarkar, S. D., Bluhm, M. M., MacGregor, J., and Nordin, B. E. C. 1961. An isotope test of calcium absorption. Brit. Med. J., U, 1539- 1541. Bhandarkar, S. D., and Nordin, B. E. C. 1962. Effect of low calcium diet on urinary calcium in osteoporosis. Brit. Med. J., I, 145-147. Bronner, F., Saville, P. D., Nicholas, J. A., Cobb, J. R., Wilson, P. D., Jr., and Wilson, P. D. 1962. Calcium metabolism in man. Quantitation of calcium absorption and excretion index. In Radioisotopes and Bone RAREFYING DISEASE OF THE SKELETON 44- 1 (F. C. McLean, P. Lacroix, and A. M. Budy, editors), pp. 17-34. Blackwell Scientific Publications, Oxford, England. Buhr, A. J., and Cooke, A. M. 1959. Fracture patterns. Lancet, I, 531-536. Caldwell, R. A. 1962<7. The effect of sex hormones in experimental osteo- porosis in rats. Brit. J. Pathol, 73, 103-115. Caldwell, R. A. 1962b. Assessment of adrenal cortical function after death in cases of senile osteoporosis. /. Clin. Pathol., 15, 361-366. Casuccio, C. 1962. Concerning osteoporosis. /. Bone and Joint Surg., 44B, 453-463. Casuccio, C, Bertolin, A., and Falzi, M. 1962. Dell' osteoporosi senile, ricerche anatomo-radiologiche-istochemiche, chemico-analitiche e biofisiche sul tessuto osseo vertebrale umano in rapporto all' eta. Clin, ortoped., 14, 1-15. Clark, I., and Roth, M. L. 1961. The effects of adrenal cortical steroids on bone calcium and phosphorus. In Inflammation and Diseases of Con- nective Tissue (L. C. Mills and J. H. Moyer, editors), pp. 404-409. W. B. Saunders Co., Philadelphia, Pa. Clark, I., and Smith, M. R. 1962. Effect of hydrocortisone on Ca^'' and P'^- metabolism in intact and parathvroidectomized rats. Federation Proc., 21, 2. Clerkin, E. P., Haas, H., and Canery, J. J. 1962. Osteomalacia in thyrotoxi- cosis. Clin. Research, 10, 399. Dent, C. E. 1956. Discussion on surgical aspects of disorders of calcium metabolism. Proc. Roy. Soc. Med., 49, 715-722. Dingwall, J. A., Millonig, R. C, and Westcott, W. L., assignors to Olin Mathieson Chemical Corp. 1962. Inorganic Salts of Mucoprotein. Can. Pat. 645,973, July 31, 1962. Dow, E. C, and Stanbury, J. B. 1960. Strontium and calcium metabolism in metabolic bone diseases. /. Clii}. Invest., 39, 885-903. Doyle, F. H. 1961. Ulnar bone mineral concentration in metabolic bone diseases. Brit. J. Radiol, 34, 698-712. Eisenberg, E., and Gordan, G. S. 1961. Skeletal dynamics in man measured by radioactive strontium. /. Clin. Invest., 40, 1809-1825. Fraser, R. 1962. The problem of osteoporosis. /. Bone and Joint Surg., 44B, 485-495. Frost, H. M. 1961. Postmenopausal osteoporosis: A disturbance in osteo- clasia. /. Am. Geriat. Soc., 9, 1078-1084. Gallagher, T. F., and Spencer, H. 1962. Steroid hormone metabolism in osteoporosis and response to adrenocorticotropin. /. Clin. Endocrinol, and Metab., 22, 1122-1126. Gerber, G. B., Gerber, G., Altman, K. I., and Hempelmann, L. H. 1962. Studies on the metabolism of tissue proteins. Intern. J. Radiation Biol, 4, 609-614. 44'2 URIST, MACDONALD, MOSS, AND SKOOG Goldsmith, R. S., Mason, A. D., Hall, W. H., and Powers, J. F. 1962. A rapid calcium infusion test for hyperparathyroidism. /. Clin. En- docrinol and Metah., 22, 741-747. Gordan, G. S. 1961. Osteoporosis diagnosis and treatment. Texas State J. Med., 37, 740-744. Gordan, G. S., Eisenberg, E., Loken, H. F., Gardner, B., and Hayashida, T. 1962. Clinical endocrinology of parathyroid hormone excess. Re- cent Pro^r. Hormone Research, 18, 297-336. Greenberg, E., Pazianos, A., Corey, K. R., Kenney, P., Laughlin, J. S., and Pearson, O. H. 1961. Radioactive calcium (Ca^") tracer studies in patients with bone lesions. Proc. Conf. Researcli Radiotherapy Can- cer, pp. 158-172. American Cancer Society, New York. Groen, J. J., Duyvensz, F., and Halsted, J. A. 1960. Diffuse alveolar atrophy of the jaw (non-inflammatory form of paradental disuse and pre- senile osteoporosis). Gerontol. Clin., 2, 68-86. Haas, H. G., Canery, J. J., Meyer, R. J., and Schaaf, M. 1962. Further evidence for calcium deficiency in several types of osteoporosis. Clin. Research, 9, 180. Hall, James W., Ill, and Kennedy, B. J. 1962. Idiopathic osteoporosis. Arch. Internal Med., 108, 448-455. Heller, M., McLean, F. C, and Bloom, W. 1950. Cellular transformations in mammalian bones induced by parathyroid extract. Am. /. Anat., 87, 315-348. Henneman, P. H., Benedict, P. H., Forbes, A. P., and Dudley, H. R. 1958. Idiopathic hypercalciuria. New England J. Med., 259, 802-807. Holdeway, E. M., Ruhr, A. J., and Little, K. 1962. The mineral phase in osteoporotic bone. Nature, 196, 181-182. Ibsen, K., Uyeno, S., and Urist, M. R. 1963. Proline utilization and healing of fractures. Nature, 197, 803-804. Jacobson, B. 1953. Dichromatic absorption radiography. Dichromatog- raphy. Acta Radiography, 39, 437-452. Jacobson, B. 1962. Personal communication. jowsey, J. 1960. Age changes in human bone. Clin. Orthopaed., 17, 210- 218. Kao, K.-Y. T., Hitt, W. E., Dawson, R. L., and McGavack, T. H. 1962. Connective tissue. VII. Changes in protein and hexosamine content of bone and cartilage of rats at different ages. Proc. Soc. Exptl. Biol, and Med., 110, 538-543. Klein, L., Albertsen, K., and Curtiss, P. H. 1962. Urinary hydroxyproline in hyperparathyroidism: A study of three cases with and without bone lesions. Metah., Clin, and Exptl., 11, 1023-1027. Kolb, F. O., and Cohen, P. 1962. The influence of the adrenals on calcium homeostasis in hypoparathyroidism. Clin. Research, 10, 87. RAREFYING DISEASE OF THE SKELETON 443 Kiizell, W. C, Glover, R. P., Brims, D. L., and Gibbs, J. O. 1962. Methan- drostenolone in rheumatic diseases and osteoporosis. Geriotrics, 17, 428-441. Lichtowitz, A., de Seze, S., Parlier, R., Hioco, D., Miravet, L., and Lanham, C. 1962. Le traitement actuel des osteoporoses. Semaine hop. Paris, 38, 2055-2065. Lindahl, O., and Lindgren, A. G. H. 1962. Grading of osteoporosis in autopsy specimens. Acta Orthopaed. Scand., 32, 85-99. Little, K., Kelly, M., and Courts, A. 1962. Studies on bone matrix in nor- mal and osteoporotic bone. /. Bone and Joint Surg., 44B, 503-519. MacDonald, N. S. 1960. Radioisotope osteogram — Kinetic studies of skele- tal disorders in humans. Clin. Orthopaed., 17, 154-164. Malm, O. y. 1961. Lack of adaptation to low calcium intakes and its possi- ble relation to calcium deficiencv osteoporosis in man. Voeding, 22, 567-590. Martinez, O. M., and Greenblatt, R. B. 1962. The role of corticoids in osteoporosis. Geriatrics, 15, 555-562. Mayo, K. M. 1961. Radiological assessment of bone density. In Symposium on Radiological A.ssessmcnt of Bone Densitij. II. Quantitative Meas- urement of Bone Mineral in Norfnal Adult Bone. Brit. }. Radiol, 34(407), 693-695. McClendon, J. F., Jowsey, J., Gershon-Cohen, J., and Foster, W. C. 1962. The curative effect of a high-calcium diet on senile osteoporosis. /. Nutr., 77, 299-302. Meroney, W. H., Rubini, M. E., Rosch, P. J., Austen, F. K., Herdon, E. G., and Blythe, W. B. 1959. Decreased density of bone: An etiologic ap- proach to diagnosis. Metah., Clin, and Exptl., 8, 293-375. Moldawer, M., Collins, L. C, \Miite, B. L., and Mayfield, J. D. 1961. Com- parison of the incidence of osteoporosis in white and Negro individ- uals. Clin. Research, 9, 52. Moon, N. F., and Urist, M. R. 1962. Gerontal and gerontoid osteoporosis. Clin. Orthopaed., 23, 269-282. Morgan, A. F., Gilium, H. L., Gifford, E. D., and Wilcox, E. B. 1962. Bone density in aging population. Am. }. Clin. Nutr., 10, 337-346. Murray, R^ O. 1961. Steroids and the skeleton. Radiology, 77, 729-743. Myers^ W. P. 1962. Studies of serum calcium regulation. Advan. Internal Med., 9, 163-214. Nicolaysen, R. 1960. The calcium requirement of man as related to dis- eases of the skeleton. Clin. Orthopaed., 17, 226-234. Nordin, B. E. C. 1962«. L'osteoporose. Schweiz. med. Wochschr., 92, 883- 891. Nordin, B. E. C. 1962/;. The problem of osteoporosis. Gerontol. Clin., 4, 19-32. 444 IRIST, MACDONALD, MOSS, AND SKOOG Nordin, B. E. C. 1962f. Calcium balance and calcium requirement in spinal osteoporosis. Am. J. Clin. Niitr., 10, 384-390. Nordin, B. E. C, Barnett, E., McGregor, J., and Nisbet, J. 1962fl. Lumbar spine densitometry. Brit. Med. J., I, 1793-1796. Nordin, B. E. C, Bluhm, M., and MacGregor, J. 1962^. In vitro and in vivo studies with bone seeking isotopes. In Radioisotopes and Bone (F. C. McLean, P. Lacroix, and A. M. Budy, editors), pp. 105-126. Blackwell Scientific Publications, Oxford, England. Raveni, G., Rubegni, M., and Del Giovane, L. 1962. L'idrossiprolinuria neir osteoporosi. Boll. soc. ital. biol. sper., 38, 266-267. Riccitelli, M. L. 1962. The management of osteoporosis in the aged and infirm. /. Am. Geriat. Soc., 10, 498-504. Rich, C., and Ensinck, J. 1961. EflFect of sodium fluoride on calcium me- tabolism of human beings. Nature, 191, 184-185. Robinson, R. A. 1960. Chemical analysis and electron microscopy of bone. In Bone as a Tissue (K. Rodahl, J. T. Nicholson, and E. M. Brown, Jr., editors), pp. 186-250. McGraw-Hill Book Co., New York. Schlesinger, B. E., Forsyth, C. C, White, R. H. R., Smelhe, J. M., and Stroud, C. E. 1961. Observations on the clinical course and treatment of one hundred cases of Still's disease. Arch. Disease Childhood, 36, 65-76. Sherman, M. S., and Selakovitch, W. G. 1957. Bone changes in chronic circulatory insufficiency. /. Bone and Joint Surg., 39A, 892-901. Silberberg, M., and Silberberg, R. 1962. Osteoarthrosis and osteoporosis in senile mice. Gerontolo^'ia, 6, 91-101. Sissons, H. A. 1962. Age changes in the structure and mineralization of bone tissue in man. In Radioisotopes and Bone (F. C. McLean, P. Lacroix, and A. M. Budy, editors), pp. 443-466. Blackwell Scientific Publications, Oxford, England. Smith, R. W., Eyler, W. R., and Mellinger, R. C. 1960. On the incidence of senile osteoporosis. Ann. Internal Med., 52, 773-781. Sobel, H. 1960. Studies on the measurement of aging. Puhl. Am. Inst. Biol. Sci., 6, 274-278. Sobel, H., and Marmorston, J. 1958. Hormonal influences upon connective tissue changes of aging. Recent Progr. Hormone Research, 14, 457- 482. Sobel, H., Marmorston, J., and Moore, F. J. 1954. Collagen and hexosamine content of femurs of rats. Proc. Soc. Exptl. Biol, and Med., 87, 346- 349. Stevens, J., Freeman, P. A., Nordin, B. E. C, and Barnett, E. 1962. The incidence of osteoporosis in patients with femoral neck fractures. /. Bone and Joint Surg., 44B, 520-527. Storey, E. 1961. Cortisone-induced bone resorption in the rabbit. En- docrinologij, 68, 533-542, RAREFYING DISEASE OF THE SKELETON 445 Summer, G. K. 1961. Oral proline tolerance in osteogenesis imperfecta. Science, 134, 1527-1528. Taylor, T. G. 1962. Calcium absorption and metabolism in the laying hen. In Nutrition of Pigs and Poultry (J. T. Morgan and D. Lewis, edi- tors), pp. 148-157. Butterworth Co., London. Taylor, T. G., Morris, T. R., and Hertelendy, F. 1962. The effect of pitui- tary hormones on ovulation in calcium-deficient pullets. Vet. Record, 74, 123-125. Teng, C. T., Daeschner, C. W., Singleton, E. B., and Rosenberg, H. S. 1961fl. Liver diseases and osteoporosis in children. II. Etiological consideration. /. Pediat., 59, 703-709. Teng, C. T., Daeschner, C. W., Singleton, E. B., Rosenberg, H. S., Cole, W., Hill, L. L., and Brennan, J. C. 19615. Liver diseases and osteo- porosis in children. I. Clinical observations. /. Pediot., 59, 68.5-702. Trotter, M., Broman, G. E., and Peterson, R. R. 1960. Density of bones of white and Negro skeletons. /. Bone and Joint Surg., 42A, 50-58. Urist, M. R. 1958. The problem of osteoporosis. Clin. Research, 6, 377- 385. Urist, M. R. 1959. The effects of calcium deprivation upon the blood, adrenal cortex, ovary, and skeleton in domestic fowl. Recent Progr. Hormone Research, 15, 455-481. Urist, M. R. 1960rt. Cage layer osteoporosis. Endocrinologij, 67, 879-880. Urist, M. R. 1960/?. Observations bearing on the problem of osteoporosis. In Bone as a Tissue (K. Rodahl, J. T. Nicholson, and E. M. Brown, ■ Jr., editors), pp. 18-45. McGraw-Hill Book Co., New York. Urist, M. R. 1962. Osteoporosis. Ann. Rev. Med., 13, 273-286. Urist, M. R., and Deutsch, N. M. 1960fl. Osteoporosis in the laying hen. Endocrinology, 66, 377-391. Urist, M. R., and Deutsch, N. M. 19605. Influence of ACTH upon avian species and osteoporosis. Proc. Soc. Exptl. Biol, and Med., 104, 35- 39. Urist, M. R., and Deutsch, N. M. 1960c. Effects of cortisone upon blood, adrenal cortex, gonads, and the development of osteoporosis in birds. Endocrinology, 66, 805-818. Urist, M. R., Mazet, R., and Bechtol, C. O. 1959. Senile osteoporosis as a disorder influencing treatment and end results of fractures of tlie hip: With a preliminary report on the use of collapatite. Atin. Surg., 25, 883-890. Urist, M. R., and Moon, N. F. 1961. Fractures of the hip in patients with severe osteoporosis. /. Trauma, 1, 373-391. Urist, M. R., and Vincent, P. J. 1960. The excretion of various fractions of 17 ketosteroids in the urine in women with postmenopausal or senile osteoporosis. CAin. Orthopaed., 18, 199-208. Urist, M. R., Zaccahni, P. S., MacDonald, N. S., and Skoog, W. A. 1962. 446 URIST, MACDONALD, MOSS, AND SKOOG New approaches to tlie problem of osteoporosis. /. Bone and Joint S'f/rg., 44B, 464-484. \ incent, P. J., and Urist, M. R. 1961. The appearance of osteoporosis in ambulatory institutionalized males. Clin. OrthopaecL, 19, 245-252. Vose, G. P. 1962. Quantitative microradiography of osteoporotic compact bone. Clin. 0)ihopoed., 24, 206-212. \V'agner, A., and Schaaf, J. 1961. Vergleichende Untersuchungen mit und ohne photometrischen Messverfahren iiber den Grad osteoporotischer Veranderungen und Rontgenbild. Dent. Arch. klin. Med., 207, 364- 385. Walser, M., Robinson, B. H. B., and Duckett, J. W. 1962. The hyper- calcemia of adrenal insufficiency. Clin. Research, 10, 404. Westcott, VV. L., Dingwall, J. A., and Millonig, R. C, assignors to Olin Mathieson Chemical Corp. 1962. Miicopoli/saccharide Preparation. Can. Pat. 645,975, July 31, 1962. 16 icroradiography of Bone Resorption JENIFER JOWSEY,* Department of Microradiography, Albert Einstein Medical Center, Northern Division, Philadelphia, Pennsylvania FOR a normal skeleton to be maintained there must be a balance between the addition and the removal of bone tissue. This turnover of bone may be taking place rapidly as in a growing individual, or slowly as in young adults, but whatever the rate, the balance must be maintained to result in the compact, well calcified tissue of a normal skeleton. A number of abnormalities of the skeleton are the result of an upsetting of this balance in turnover of bone; there is an increase or decrease in removal of tissue without a corresponding increase or decrease in addition of bone, and the result is a change in the gross structure of the bone, generally a loss of tissue, which seriously affects its supporting function. If the abnormality existing in various bone diseases is to be understood, the immediate cause of the upset in the balance must first be found, and this may be done by measuring the amount of bone formation and bone resorp- tion and also measuring the extent of decrease in mass of the tissue, or its porosity, a figure which will indicate for how long the balance has been upset. The Measurement of Bone Turnover A number of attempts have been made to measure bone foimation by equating the amount of new bone laid down in the skeleton at ** Present address: Section of Surgical Research, Mayo Clinic, Rochester, Minnesota. 447 448 J. JOWSEY anv time with the amount of mineral being added at that time (Bauer et al., 1955; Heaney and Whedon, 1958). It is most impor- tant to distinguish between these two processes. Mineral metabolism involves the movement of calcium and phosphorus between the bod\^ fluids and the bone and can be going on actively in a skeleton that is completely inert in terms of the formation and resorption of tissue. Bone turnover refers to actual addition and removal of tissue to and from the skeleton. Radioactive isotopes of calcium, radium, and strontium, when injected or ingested in vivo in man, will be taken up in new bone as it is mineralized, and for this reason their retention by the skeleton has been measured bv external counting techniques as an indication of bone formation. However, these elements are also firmly retained by fully mineralized nongrowing bone by a mechanism not associated with the laying down of new bone or even necessarilv with net addition of mineral (Arnold et al., 1960; Marshall et al., 1959). Quantitative measurements of this fraction in radium-burdened bone have indicated that it is a signifi- cant part of the body burden; in 16 individuals measured, an aver- age of 47 per cent of the total amount was in this diffuse fraction and not associated with new bone formation (Rowland, 1960). Tetracyclines, which are incorporated into mineralizing bone ( Milch et al., 1957 ) , have also been used in an attempt to measure bone formation in small bone samples (Urist et al., 1962), but since they have a distribution in bone similar to that of calcium isotopes (Harris et al., 1962), these results must suffer from the same diffi- culty of interpretation as the data from calcium and strontium re- tention studies. A preliminary experiment (Jowsey, 1962) involving microdissection of human biopsy samples labeled with tetracycline has shown that 24 hours after injection or ingestion, only 31 to 62 per cent of the total amount of the dose is associated with formation of bone, 38 to 62 per cent being deposited in nongrowing areas. Investigations into the retention of radioactive isotopes of sulfur and carbon that label the organic fraction of bone would suffer from the same difficulty of intei^pretation, as they too have a diffuse com- ponent not associated with bone formation (Kent et al., 1956; Bloom etal,lMl). Since it has so far been impossible to separate these two fractions MICRORADIOGRAPHY OF BONE RESORPTION 449 in in vivo procedures or in chemical analyses of whole biopsy sam- ples, there must always be uncertainty in these methods as to the proportion of the retained isotope measured that reflects bone for- mation. The most successful attempts to measure bone formation have consisted of sampling small areas of growth in sections of bone from specific sites in the skeleton (Jowsey, 1960; Frost and Villanueva, 1961 ) . Microradiographs of undecalcified sections are perhaps more informative than the more conventional decalcified sections, since they demonstrate the distribution of mineral in the tissue besides showing a most characteristic appearance in areas of formation and resorption. It is possible with tetracyclines (Vanderhoft et al., 1962; Harris et ah, 1962) or a bone-seeking radioactive isotope (Jowsey et al, 1953; Plonlot, 1960) to see with the aid of a microscope where growth has taken place, as the new tissue will be marked by a band of the fluorescent material or the radioisotope, or will lie between two bands if the label is given twice at different times. By compar- ing such areas of growth with a microradiograph of the same section, the characteristic appearance of new bone formation has been estab- hshed as an area of low density with new lamellae lying parallel to a smooth surface (Fig. 1); most characteristic is the continuous decrease in density right up to the edge of the bone. A bone surface that has recently been growing but has stopped will show a sclerotic ring adjacent to the haversian canal (Fig. 3). Measurements of bone formation may be made using microradiographs of unlabeled material. It becomes apparent on looking at such measurements in bone from normal elderly people and in specimens of some bone disorders that, generally, there is no great variation in bone formation. It would seem that it is increases or decreases in resorption that are important in producing the gross changes seen in abnormal bone, so that attempts to evaluate bone formation only would not measure the cause of the upset in the balance of bone turnover. Resorption must be measured. Bone resorption is unfortunately more difficult to characterize than bone formation, since it consists of the removal of tissue, and it is possible onlv to deduce from circumstantial evidence what a 450 J. JOWSEY Fig. 1. Microradiograph of a cross section through the mid-shaft of the femur of a 2-year-old male. Bone formation is occurring at a and b while resorption is occurring at c and d. There are a number of areas of interstitial bone containing enlarged lacvniae (e and /) which are characteristic of woven or rapidly growing bone. ( X 20. ) surface from which bone is being removed looks like in, for instance, a microradiograph of a bone section. This evidence comes from a number of sources. The periosteal surface of the metaphvsis of a long bone from a growing animal is an area where removal of tissue must be occurring as the diameter of the shaft becomes smaller. In addition, histological preparations show osteoclasts on such surfaces, and osteoclasts have been observed to resorb bone in tissue culture (Goldhaber, 1960). Such a periosteal surface must therefore be undergoing active resorption. This surface appears in a micro- radiograph as an irregular, highlv calcified surface. The resorption of bone has also been associated with the deposi- tion of radioactive isotopes of the rare earths and, in particular, of yttrium. Y^^ has been shown to be localized in vivo in areas where MICRORADIOGRAPHY OF BONE RESORPTION 451 bone is being remo\ed in the process of remodeling in the metaph- vsis, and in large irregular ca\'ities in cortical bone (Jowsev et cil., 1956). Further studies (Jowsev et cil, 1958; Neuman et al., 1960) have shown that the deposition of \'ttrium on the surface of hydroxy- apatite is associated with citric acid production or a change in pH which in turn is produced by injections of parathyroid hormone, which causes resorption of bone in conjunction with the production of osteoclasts (Martin et ah, 1958). Y'*^ may well be a better indica- tion of bone resorption than the presence of osteoclasts, since it depends on the mechanism of tissue resorption for its deposition in bone, the isotope being taken up on bone mineral in areas where the yttrium in the bodv fluid is rendered diffusible by a local increase in acidity. Though deposition of yttrium probably always indicates resorption and its absence in high concentrations indicates lack of resorption, the absence of an osteoclast does not always mean that no resorption is occurring in that area ( Hancox, 1956; Fell and Mel- lanby, 1952; Urist and Deutsch, 1960 ) . Therefore the deposition of yttrium in bone can be used to confirm the characteristic micro- radiographic appearance of a resorbing surface which has been established by comparing the microradiograph with the histological preparation. There should be a high correlation between the sites of deposition of yttrium and the microradiographic appearance of resorption; experimental data confirm this. Autoradiographs of bone sections from a young monkey injected with Y'" 24 hours before death were compared with microradiographs of the same sections. The sites of concentrated Y"^ deposition were correlated with the sites in the microradiograph where the highly calcified surface indi- cated active resorption to be occurring. Accurate measurements on enlarged photographs of the lengths of surface involved showed that the yttrium deposition and microradiographic appearance of re- sorption corresponded in 89 per cent of the areas, the remaining 11 per cent being sites either of yttrium deposition which were judged as not showing resorption in the microradiograph or of microradiographic indications of resorption with no concentrated uptake of yttrium. The correlation coefficient between the two sets of results was 0.99. If, on the other hand, the results from the auto- 45^2 J. .towsp:y radiographic measurements of resorption and the microradiographic measurements of resorption are compared directly, paying no atten- tion to whether the measurements correspond, there is a 99.99 per cent similarity between the results. Therefore, assuming the results from the ) ttrium autoradiographs to be correct, the results from the microradiographs will be 99.99 per cent similar numerically though they will include 10 per cent mistakes. This correlation between the deposition of yttrium and the microradiographic appearance of high density confirms the criteria established for bone resorption by com- parison with the presence of osteoclasts in a histological preparation, and also the quantitative data suggest that very nearlv 100 per cent of all resorption surfaces are being measured by this method. These correlative data show that a bone surface that is under- going resorption appears in a microradiograph as an irregular, crenated, highlv calcified surface with the lamellae Iving at all angles to the surface rather than around the cavity ( Fig. 1 ) . It is most important to distinguish between a bone surface that is actively being resorbed and appears microscopically crenated and irregular and one where resorption has been taking place but has ceased. The latter surfaces are highly calcified and irregular and have the lamellae running at angles to the surface, like areas of active resorption, but are microscopically smooth and have a narrow rim of high mineralization immediately adjacent to the vascular cavity. Apart from the areas of active resorption, yttrium is also taken up diffusely throughout the bone tissue, probably in association with low levels of citric acid, and it is also retained to some extent on the highly mineralized surfaces of completed osteons ( Jowsey et ah, 1956, 1958). Therefore yttrium and the rare earths probably cannot be used to measure bone resorption with any certainty b\' methods using in vivo external counting or chemical measurements on large pieces of bone, because, as with the alkaline earths, it is not possible to differentiate between the fraction of isotope retained as a result of the localized intense activity of the bone, that is, the resorption of tissue, and the slow metabolic activity that occurs throughout the bone. MICRORADIOGRAPHY OF BONE RESORPTION 453 Summary Any investigation of bone may profitably include a measurement of turnover, that is, the addition and removal of tissue from the skeleton. Quantitative measurements of growth and resorption in normal tissue may be compared with those in bone from metabolic diseases and other diseases of the skeleton to discover the imme- diate cause of the abnormality. A number of methods have been devised for measuring bone formation, the most direct and accurate one being the quantitative estimation of growth in a specific area of a bone sample bv the characteristic appearance of the surface of the bone, preferablv in a microradiograph of an undecalcified sec- tion. In vivo administration of radioactive isotopes of calcium and other alkaline earths and of sulfur in animals have made it possible to establish the characteristic appearance of a growing surface as one where the lamellae are of low density and lie parallel to a smooth surface. Bone resorption, which is probably the most im- portant cause of imbalance in tissue turnover, can be identified by comparison with histological preparations showing osteoclasts and with the localization of yttrium and the rare earths, which depend on the mechanism of resorption for their deposition in high con- centrations. Resorbing surfaces can be identified in a microradio- graph by the presence of highly calcified bone with lamellae lying at angles to an uneven, crenated surface. The sampling method of studying bone turnover has some dis- advantages, the most obvious being that it is not easy to be sure exactly how relevant the results from any one sample are to the turnover in the rest of the skeleton. However, the results from one sample area in different individuals may perhaps be treated as com- parative data without reference to the rest of the skeleton. Also samples from different parts of the skeleton will produce data that will indicate the magnitude of the variations to be expected through- out the skeleton. The disadvantages are surpassed by the advantages of being able to measure bone formation and, most important, bone resorption without doubt, and also of being able to investigate local lesions in the skeleton. 45-i J. JOWSEY Material and Techniques The method adopted for the present stud\ of bone turnover was the quantitative evaluation of bone formation and resorption in microradiographs of bone sections. Both normal and pathological material was used in this studv. Normal material was collected from accident cases and sudden deaths from heart attacks. Gross x-rav pictures were made of 1.0-cm slices through the 5th lumbar vertebra to determine whether any osteoporosis was present in the normal individuals; if such x-rays showed abnormally low density, these individuals were excluded from the studv. Pathological material came from individuals with bone diseases, in particular osteoporosis. This group consisted of 10 individuals with x-rav evidence of osteo- porosis such as decreased density of the skeleton and crush fractures of the spine; about half of the group had also fractured the neck of one femur. The specimens were taken from the mid-shaft of the femur, the neck of the femur, the iliac crest, and the 5th lumbar vertebra, though the results reported here are mainly those from the mid-femoral shaft. The specimens are fixed in 70 per cent alcohol, deh)'drated in absolute alcohol for 2 or 3 days, and then put in unpolymerized methyl methacrylate monomer for 24 hours. The specimens are then placed in semipolvmerized monomer and the polymerization is com- pleted by heating at 32 °C for 2 to 3 davs. The blocks are trimmed on a handsaw and sections 100 microns thick are cut on a commercial milling machine using a circular steel saw. The sections are further ground, if necessary, until they are 100 ± 2 microns thick, and con- tact microradiographs are made using Eastman 649-0 spectroscopic plates and an x-ray beam from a copper target. The microradio- graphs are inspected with a microscope and the areas of bone forma- tion, resorption, and inactivity identified. Quantitative measurements are made by making a print at a magnification of 30 times, and since formation and resorption of bone are activities that take place ex- clusively on the surface, the lengths of all bone surfaces that are undergoing formation or resorption or are inacti\'e are measured with a map measure. The results are expressed as the length of surface involved in either formation or resorption in a certain area MICRORADIOGRAPHY OF BONE RESORPTION 455 of a section, as a percentage of the inactive surface in that same area. A standard sample area in each site is analyzed. In the mid- shaft of the femur one-third of an entire cross section constitutes the sample area, one-quarter of the area being taken from each of four quadrants. Analysis of smaller samples produces data with considerably less significance. Measurements of the porosity of the bone are also made. In cor- tical bone from the femur this porosity measurement consists of the percentage of osteons which are less than two-thirds closed; that is, the diameter of the central canal is more than one-third the diameter of the whole osteon up to its cement line. The thickness of the cortex in the specimens is also measured. Summary Undecalcified bone sections are cut from normal and pathological material and microradiographs made. Bone formation and resorption are identified and quantitative measurements made on photographic enlargements. Bone porosity is also measured. Discussion of Results Resorption Resorbing surfaces are obviously places of great metabolic and chemical activity and are in intimate contact with the body fluids as indicated by electron microscope work (Scott and Pease, 1956; Cameron and Robinson, 1958) and the uptake of the rare earths (Jowsey et al., 1958). Calcium^"' and tetracycline are also deposited on these surfaces for a short time ( Fig. 2 ) , possibly by an exchange mechanism, and may account to some extent for some of the skeletal retention found at short intervals after an injection of radioisotopes such as Ca^^ in areas of fracture or in osteoporotic bone, where a great deal of resorption is known to be occurring. Areas of par- ticularly active resorption appear microradiographically as very ir- regular surfaces with numerous uneven projections into the wall of the cavity. This bizarre resorption was first described by Rowland et al. ( 1959 ) in microradiographs of radium-burdened bone, but has since been found in almost every bone disorder involving rapid re- 456 J. JOWSEY .<* '^ «>• « B. Fig. 2. Microradiograph (A) and Ca^^ autoradiograph (B) of a cross section through the mid-shaft of the femur of an adult rabbit with cortisone- induced osteoporosis. The animal was killed \\ hours after Ca^^ injection, and the isotope is heavily concentrated in areas of bone resorption. ( X 22. ) sorption, such as osteoporosis and renal failure, and also occasionallv in normal bone (Fig. 3). Despite these indications that there may be different rates of resorption in different areas, in the quantitative measurements it has been supposed that an axerage \'alne is xalid. » w Fig. 3. Microradiograph of a cross section through the mid-shaft of the femur of a 15-year-old female with renal failure, showing the irregularity of surface characteristic of rapid resorption at a, h, and c, and more normal resorption at d and e. At g and / are two recently formed osteons with sclerotic rings characteristic of cessation of new bone formation. (X 22.) MICRORADIOGRAPHY OF BONE RESORPTION 457 Quantitative Measurements Normal. In order that abnormality may be appreciated, the normal state of affairs must first be described. It is always a little difficult to choose criteria for what shall be considered normal and what abnormal. Since one of the main purposes of this study was to compare normal with osteoporotic bone, certain individuals were rejected from the normal group because a gross slab x-ray through a 1-cm-thick longitudinal slice of the 5th lumbar vertebra showed definite porosis; only those showing unremarkable density in this x-ray were retained as normals. This normal group shows the char- acteristic pattern of bone turnover that has already been demon- strated (Jowsey, 1960), with the exception that, as a result of ex- cluding those individuals that showed loss of bone mass in the spine, there was less increase in the amount of resorption in older individ- uals. From Fig. 4 it seems that from the age of 49 onward the level of resorption is above that of formation, but that the difference does not continue to increase after the age of 70. The result of this imbalance is visible in the values for porosity; that is, the number of osteons less than two-thirds closed is increased, while the thick- ness of the cortex decreases. These graphs do not paint the whole picture: from the microradiographs themselves (Fig. 5) it can be seen that there are generally a greater number of incompletely mineralized osteons in the older individuals and that the increase in number of osteons less than two-thirds closed appears to be the result of a preponderance of resorption cavities which have become only partly filled in with new bone. Resorption and formation rates. The actual addition or sub- traction of tissue mass to or from the bone can be calculated by mul- tiplying the values for surface activities by the rate of formation or resorption which is seen in the thickness of new tissue in a cross section. The results reported here assume that the rates of bone formation and resorption are similar and that they do not vary a great deal without leaving an indication in the microradiograph. Measurements have been made of the rate of new bone formation in cross sections of cortical bone using tetracycline markers (Frost, 458 J. JOWSEY % inactive 16 Bone formation and resorption 14 12 10 Normal n Formation o Resorption Osteoporotic ■ Formation • Resorption Cortical thickness % 60 oNorma •Osteop. _i 1 i 1 — I 0 Porosity:<-3 closed osteons Relation between inactive surface and area /xxlO" /mm' 40- o Normal • Osteop _i I I I I I I 0 o Normal • Osteop. _j I I I I I I 20 40 60 80 20 40 60 80 20 40 60 80 Age Age Age Fig. 4. Turnover data from the mid-shaft of the femur of normal and osteoporotic individuals. See text for details. 1960), and the average value is 1 micron per day. Resorption rates cannot be directly measured, but since a mass of bone that is not losing or gaining tissue, such as cortical bone in a normal 20- to 30- year-old human, shows approximately equal lengths of surface in- volved in resorption and in formation, it is probably not incorrect to assume that the rate of resorption is approximately equal and also averages 1 micron per dav. However, rates of resorption and formation must vary from place to place and time to time. Fomia- tion is somewhat more rapid when an osteon is just beginning to be laid down, and the rate becomes slower as the osteon is com- MICRORADIOGRAPHY OF BONE RESORPTION 450 '''^. -'<'*• '-.^4. /.'.-.^ '( ) # ♦ < . • v • v^** •-^ • y^ ^ .. •/. # ■* ' ^ ■^ t()() J. .lOWSEY pleted. As large resorption cavities are more frequent than small ones, and as the average length of surface per cavity is higher for resorption than for formation, it is probable that resorption also takes place more rapidly in a new, small resorption cavitv than in an older, large one. However, despite the local variations in the rates of formation and resorption, an average figure should be valid, and measurements of the length of surface occupied by resorption and formation should also be valid and refer to actual bone turnover. The measurements can be converted to mass of tissue added to or removed from bone if one assumes the figure of 1 micron per day for the rate of formation and resorption to be correct and knows the relation between the amount of inactive surface and the area of the sample (Fig. 4) (20 ju, X 10 "-/mm^ in 20- to 40-year-old individuals and SO [x X lO""/!!!!""^" hi 60- to 80-year-old individuals). This in- crease in the amount of surface per unit area of bone means that if resorption and formation are to be considered in terms of area or mass rather than surface, the actual values will show a greater in- crease in old as compared with young adults. In other words, from the graph (Fig. 4), resorption in young adults is 2 /x/100 fx of in- active surface, or 1.5 mm^/100 mm^/year, which is balanced by similar values for bone formation. In older individuals the value in terms of length of surface is 4.8 per cent and in terms of mass it is 5.2 mni'VTOO mm'Vyear, while bone formation is 3.3 per cent in terms of length of surface and 3.6 per cent per year in terms of mass. Therefore, at this rate a piece of cortical bone should lose 16 per cent of its mass over a 10-vear period, a figure that is consistent, as far as can be judged, with the values for porosity and cortical thickness. Osteoporosis In comparison with the normal series, the amount of bone forma- tion in osteoporotic bone seems not to be different from the levels of formation found in normals of similar age ( Fig. 4 ) . However, 2 out of the 10 cases showed remarkably low values, and these had been bedridden for 2 to 12 months before autopsy; there seems to be an effect of partial immobilization on bone formation. The levels MICRORADIOGRAPHY OF BONE RESORPTION 461 Fig. 6. Microradiograph of a cross section through the mid-shaft of the femur of a 67-year-old female with osteoporosis. There are a large number of resorption cavities and a high proportion of incompletely closed osteons. (Xll.) of bone resorption are very much higher in this group than in the normal series ( Fig. 6 ) . In older individuals there appears to be some equilibrium, and the amount of resorption decreases toward normal values while the levels of bone formation remain the same. The porosity figures are significantlv higher in the osteoporotic group than in the normal series, which is to be expected since the porosity represents the cumulative effect of the difference between the levels of resorption and of formation. The values for cortical thickness are generally lower than normal. It would seem, therefore, that osteo- porosis is the result of increased resorption of bone, which in the mid-shaft of the femur appears to be a factor of 2 or 3 times above normal values. These differences are exaggerated in the iliac crest, both in the osteoporotics and in the normal series, and for this rea- son the iliac crest is probably a better biopsy site. 462 J. JOWSEY Paget' s Disease Grossly this a])normality varies in its appearance from one of de- creased density to one of sclerosis, and the microradiographic ap- pearance of three stages — porosis, sclerosis, and presarcomatous — has been described bv Kelly et ah ( 1961 ) . Figure 7A demonstrates the porotic stage, where bone resorption is going on most actively at levels 15 times above the normal level. A higher magnification (Fig. 7B) shows the characteristic mosaic appearance and great variation in degree of mineralization and also the very irregular surfaces of the bizarre resorption cavities associated with rapid disappearance of tissue. At short intervals after injection of tetra- cvcline such irregular surfaces are strongly labeled, indicating par- ticularly high metabolic activity. Quantitative values for formation and resorption are in the order of 20 times the normal levels. Hi/perparathijwidism Earlv stages of hyperparathyroidism appear microradiographically like osteoporosis. However, advanced hyperparathyroidism accom- panied bv a much elevated blood calcium level results in calcification of the fibrous connective tissue in the haversian canals and resorp- tion cavities (Fig. 8). Both bone resorption, which is a factor of 10 or more times above the normal level, and bone formation, which is also increased, appear to be going on rapidly. In both Paget's disease and hyperparathyroidism the increased resorption levels may result in generalized or local porosis and even- tually fracture of the bone. Osteogenesis Imperfecta Measurements of bone formation and resorption show that in osteogenesis imperfecta there is both a reduction in formation to less than a third of its normal value and a threefold increase in the amount of resorption. It would seem that this situation has existed for some time, since there is a considerable amount of unremodeled primary periosteal and endochondral bone with large resorption cavities, and the cortex is very thin (Fig. 9). The bone itself appears MICRORADIOGRAPHY OF BONE RESORPTION 463 Fig. 7. A, low-power (X 9), and B, high-power (X 30) microradiographs of a cross section of the femur of a 78-year-old female showing the porotic phase of Paget's disease with the large number of spaces, mosaic pattern of mineralization and bone formation (a and b), and resorption (c, d, and e) taking place on nearly all surfaces (formation 50 per cent, resorption 140 percent). 464 J. JOWSEY Fig. 8. MiLioiacIiograph of a cross section through the teimir of an indi- vidual with hyperparathyroidism, showing recent new bone formation at a, b, and c, and minerahzation of connective tissue in haversian spaces at d, e, and /. Most bone surfaces are in the process of resorption. (X 22.) normal. In osteogenesis imperfecta, abnormal levels of both forma- tion and resorption contribute toward the decreased mass of tissue. Rickets The microradiographic appearance of rickets is most unusual. The gross appearance (Fig. lOA) is one of porosity with many areas of decreased density; the high-power appearance (Fig. lOB) shows the decreased mineral density of the cement lines surrounding the osteons and the lack of mineral around the osteocytes and their canaliculi. It is the enlarged canaliculi that are mainly responsible for the unusual appearance, and whether this represents incomplete mineralization as the bone is laid down or demineralization and MICRORADIOGRAPHY OF BONE RESORPTION 465 Fig. 9. Microradiograph of a cross section through the femur shaft of a 12-year-oId male with osteogenesis imperfecta. The cortex is thin and there is a high proportion of unremodeled primary bone. Compare with 15-year-old male in Fig. 5. (Formation 9.0 per cent, resorption 54 per cent.) ( X 6.) leaching out of mineral is unsure. The surfaces of most osteons are lined with uncalcified osteoid tissue of great thickness. The presence of osteoid tissue does not always indicate bone formation; in rickets and hypoparathyroidism the low blood calcium levels may cause failure of mineralization of osteoid tissue that has been laid down some time ago, and this may be true in other abnormalities or even in normal bone. It is worth while to point out the differences be- tween the appearance of the osteocyte lacunae in rickets and that of the somewhat irregular and enlarged lacunae of the osteocytes in rapidly growing or woven bone (Fig. 1), where the walls of the lacunae are smooth and sharply defined and the canaliculi normal in diameter in contrast to tlie fuzzy appearance of the osteocyte lacunae and involvement of the canaliculi seen in rickets. This ap- pearance is also seen in osteomalacias and in the early development of Paget's disease. 4()() J. JOWSEY Fig. 10. A, low-power (X 9), and B, high-power (X 30) microradiographs of a cro.ss section through the fibula of a 9-year-old male with vitamin D-re- sistant rickets, showing the porosity, unmineralized cement lines (a), and lack of mineral around lacunae and canaliculi {h and c). microradiography of boxe rp:sorptiox 467 Conclusion The results have demonstrated some of the facts that can be learned from microradiographic studies of bone and have shown the importance of resorption in both the normal and the abnormal skeleton. The quantitation provides some sort of base line for com- parison and makes possible the constructive evaluation of therapv in bone disorders. SrMMARY Bone turnover was measured in a defined area bv techniques in- volving semiquantitati\'e analvsis of microradiographs. Formation and resorption of bone as well as its porosity were measured and related to the age of the individual, and comparisons were made be- tween normal and abnormal tissue. It was apparent that in normal individuals an increase in the amount of bone resorption is what produces the porosity character- istic of aging bone. Osteoporotic bone appeared to be an exaggera- tion of the normal aging process in that the increased resorption caused the porosity. Bone from individuals with Paget's disease, hyperparathyroidism, osteogenesis imperfecta, and rickets had a distinctive appearance. Acknowledgments. I should like to thank Dr. Gershon-Cohen for his support and William Lafferty for his excellent technical assistance. References Arnold, J. S., Jee, W. S. S., and Johnson, K. 1960. Observations and quan- titative radioautographic studies in calcium-45 deposited in vivo in forming haversian systems and old bone of rabbit. Am. }. Anat., 99, 291-314. Bauer, G. C. H., Carlsson, A., and Lindquist, B. 1955. Evaluation of ac- cretion, resorption, and exchange reactions in the skeleton. Kgl. Fysiogrof. SaUskap. Lund, Forh., 25, 1-16. Bloom, W., Curtis, H. J., and McLean, F. C. 1947. The deposition of C'^ in bone. Science, 105, 45. Cameron, D. A., and Robinson, R. A. 1958. The presence of crystals in the 468 J. jowsEY cytoplasm of large cells adjacent to sites of bone absorption. /. Bone and Joint Surg., 40A, 414-418. Fell, H. B., and Mellanby, E. 1952. The effect of hypervitaminosis A on embryonic limb-bones cultivated in vitro. J. Physiol. (London), 116, 320-349. Frost, H. M. 1960. Lamellar osteoid mineralized per day in man. Henry Ford Hosp. Med. Bull, 8, 267-272. Frost, H. M., and Villanueva, A. R. 1961. Human osteoblastic activity. Henry Ford Hosp. Med. Bull, 9, 87-96. Goldhaber, P. 1960. Behavior of bone in tissue culture. In Calcification in Biological Systems (R. F. Sognnaes, editor), pp. 349-372. Ameri- can Association for the Advancement of Science, Washington, D. C. Hancox, N. M. 1956. The osteoclast. In The Biochemistry and PJu/siology of Bone (G. H. Bourne, editor), pp. 213-250. Academic Press, Inc., New York. Harris, W. H., Jackson, R. H., and Jowsey, J. 1962. The in vivo distribu- tion of tetracyclines in canine bone. /. Bone and Joint Surg., 44A, 1308-1320. Heaney, R. P., and \^'hedon, G. D. 1958. Radiocalcium studies of bone formation rate in human metabolic bone disease. /. Clin. Endocrinol, 18, 1246-1267. Jowsey, J. 1960. Age changes in human bone. Clin. Orthopaed., 17, 210- 218. Jowsey, J. 1962. The structure of normal and osteoporotic bone. /. Bone and Joint Surg., 44A, 1255-1256. Jowsey, J., Owen, M., and \^aughan, J. 1953. Microradiographs and auto- radiographs of cortical bone from monkeys injected with ^"Sr. Brit. J. Exptl Pathol, 34, 661-667. Jowsey, J., Rowland, R. E., and Marshall, J. H. 1958. The deposition of the rare earths in bone. Badiation Bcsearch, 8, 490-501. Jowsey, J., Sissons, H. A., and \^aughan, J. 1956. The site of deposition of Y^^ in the bones of rabbits and dogs. /. Nucl Energy, 2, 168-176. Kelly, P. J., Peterson, L. A. F., Dahlin, F. C., and Plum, G. E. 1961. Osteitis deformans (Paget's disease of bone). Radiology, 77, 368-375. Kent, P. W., Jowsey, J., Steddon, L. M., Oliver, R., and Vaughan, J. 1956. The deposition of ^^S in cortical bone. Biochem. J., 62, 470-476. Marshall, J. H., Rowland, R. E., and Jowsey, J. 1959. The paradox of the diffuse activity and long-term exchange. Badiation Besearch, 10, 258-270. Martin, G. R., Firschein, H. E., Mulryan, B. J., and Neuman, W. F. 1958. Concerning the mechanism of parathvroid hormone. II. Metabolic effects. /. Am. Chcm. Soc, 80, 6201-6204. Milch, R. A., Rail, D. P., and Tobie, J. E. 1957. Bone localization of tetracychnes. /. Natl. Cancer Inst., 19, 87-93. MICRORADIOGRAPHY OF BONE RESORPTION 469 Neuman, W. F., Mulryan, B. J., and Martin, G. R. 1960. A chemical view of osteoclasis based on studies with yttrium. Clin. Orthopaed., 17, 124-134. Plonlot, R. 1960. Le radiocalcium dans Tetude des os. Arscia, Brussels. Rowland, R. E. 1960. Radioisotopes in the skeleton: Late observations of the distribution of radium in the human skeleton. In Radioisotopes in the Biosphere (R. S. Caldecott and L. A. Snyder, editors), pp. 339-353. University of Minnesota Press, Minneapolis, Minn. Rowland, R. E., Marshall, J. H., and Jowsey, J. 1959. Radium in human bone. The microradiographic appearance. Radiation Research, 10, 323-334. Scott, B. L., and Pease, D. C. 1956. Electron microscopy of the epiphyseal apparatus. Anat. Record, 126, 46.5-495. Urist, M. R., and Deutsch, N. M. 1960. Effects of cortisone upon blood, adrenal cortex, gonads, and the development of osteoporosis in birds. Endocrinologi/, 66, 805-818. Urist, M. R., Zaccalini, P. S., MacDonald, N. S., and Skoog, W. A. 1962. New approaches to the problem of osteoporosis. /. Bone and Joint Surg., 44B, 464-484. Vanderhoft, P. J., Peterson, L. F. A., and Kelly, P. J. 1962. A method for correlative analysis of microradiogram and tetracycline fluorophore of puppies' bone. Proc. Staff Meetings Mayo Clinic, 37, 229-235. 17 Histophysical Studies on Bone Cells and Bone Resorption RICHARD W. YOUNG, Department of Anatomy. School of Medicine, Center for the Health Sciences, University of California, Los Angeles, California ONE approach to the study of the mechanisms of hard tissue de- struction directs attention to possible variations in the tissues them- selves in regions preferentiallv disposed to such destructive processes. In regard to bone, this relationship has frequently been investigated in systemically accelerated resorptive states induced by parathyroid stimulation, or by injection of parathvroid extract. The results of these studies have led, however, to contradictorv conclusions. Under such circinnstances, "stable" rather than "labile" bone is said to be preferentiallv resorbed. "Stable" bone apparently cor- responds to older, more heavilv mineralized bone, while "labile" bone refers to newer, less completely calcified bone" (McLean and Urist, 1961). A number of studies indicate, in fact, that under parathyroid stimulation there is selective resorption of older, fully mineralized bone (Woods and Armstrong, 1956; Clark and Geoffrov, 1958; Elliott and Talmage, 1958; Talmage et al, 1959, 1960, 1961) . Other data, however, suggest that it is actuallv newly formed bone, in " Stable bone has also been referred to as "nonexchangeablc," "unavailable," "deeper," "well established," and "structural" bone. Labile bone has also been designated "reactive," "exchangeable," and "metabolic" bone. 471 472 R. W. YOUNG areas of normally active osteogenesis, which is preferentially re- moved ( Jaffe et at, 1931a, 1931&; Heller et ah, 1950; Talmage et ah, 1953; Lotz et al., 1954; Bronner, 1961). The rapid destruction of metaphyseal trabcculae is often noted, and this is a region of rela- tively low mineralization (Weidman and Rogers, 1950; Engstrom and Bergendahl, 1958; Arnold, 1960). Indeed, it has been reported that matrix need not be calcified at all to yield to resorption (Carnes, 1950; Follis, 1952), although some studies indicate that uncalcified matrix is refractory to resorption (Weinmann and Schour, 1945; Reitan, 1951), and other experiments indicate that matrix need not be calcified at the time of resorption, so long as it was calcified at some time in the past (Irving and Dale, 1962). Finally, there are some reports (Engfeldt and Zetterstrom, 1954) indicating that both old and new bone are removed, following sys- temic acceleration of resorption; and published microradiograms indicate that resorption may extend through areas with varying degrees of calcification (e.g. Engfeldt and Zetterstrom, 1954; Sogn- naes, 1959; Jowsey, 1960). In the light of these inconsistent findings, it appeared worth while to summarize a number of previously unpublished observations which bear on the relation of the state of the intercellular matrix to preferential sites of resorption. Furthermore, since it appears to be generally accepted that bone resorption is an active, cellular process (McLean, 1954; Goldhaber, 1961; Talmage, 1962), attention has also been directed to the cellular element in resorption, and to controlling mechanisms which operate at the cellular level. Materials and Methods Young rats of the Long-Evans strain have been used throughout. The bones of normal animals have been analyzed by a number of techniques, and have been compared with those of animals in which bene resorpticn had been accelerated by the prior intraperitoneal injection of parathyroid extract ( PTE ) . * Generally, a total of 75 to ** Injection parathyroid U.S. p., Lilly; in part donated by Eli Lilly Company, Indian- apolis, Indiana. BONE CELLS AND BONE RESORPTION 473 100 units of the extract was administered in 3 or 4 doses over a period of 7 to 8 hours in suckUng rats between 5 days and 2 weeks of age at the time of injection. When weanHng rats were used (3 to 4 weeks of age), the dose was increased to a maximum total of 200 units administered over an 8-hour period. The costo-chondral junction of ribs, the tibiae (both regions sectioned longitudinally), and the parietal bones of the cranial vault (sectioned coronally) have been examined. Unless specifically stated otherwise, specimens were fixed in Bouin-Hollande solution for 3 days, decalcified in 18.5 per cent versene at pH 7, double-embedded in celloidin-paraffin, sectioned at 5 microns, and mounted in Permount. The following analyses were carried out on this material. Histological Stains Sections were stained with hematoxylin and eosin, chrome hema- toxylin-phloxine, Masson's trichrome, Bodian's silver (Gridley, 1960), Gomori's silver (Romeis, 1948; fixation 10 per cent formalin in 95 per cent ethanol for 24 hours), and toluidine blue (fixation in Rossmann's fluid at 4°C overnight; 0.1 per cent aqueous stain, 20 minutes at 45°C, with and without alcoholic differentiation). Methyl Green-Pyronin Bones were fixed in 10 per cent neutral formalin for 16 hours, and stained with methyl green-pyronin Y, according to Brachet ( Pearse, 1960). Some sections were incubated in a ribonuclease (Worthing- ton Biochemical Corporation ) solution ( 1 mg/ml distilled water ) at 37°C for 3 to 4 hours, or in water alone under the same condi- tions, prior to staining. Periodic Acid-Schiff Bones were fixed in Rossmann's fluid at 4°C overnight, and de- calcified in 2 per cent nitric acid in 80 per cent ethanol. Sections were spread on 70 per cent ethanol, dried at 37°C, coated with 1 per cent celloidin, and stained with periodic acid-Schiff (PAS). Some sections were incubated in saliva or in water at 37 °C for 1.5 hours, prior to staining. All sections were counterstained with hema- toxylin, and mounted in Canada balsam. 474 K. W. YOUNG Supraoilal Staiiiitig Calvaria only were used. Immediately upon sacrifice, the calvaria were placed in a solution of 0.01 per cent neutral red in 0.9 per cent saline at 37 °C for 50 minutes, then were fixed in 1 per cent mercuric chloride for at least 1 hour. The preparations were examined with the dissecting microscope, intact and unmounted. Microradiography with Uhrasoft X-Ratjs Bones were fixed in neutral formalin and embedded in paraffin. Sections were floated onto Kodak spectroscopic plates (649-0) ac- cording to the technique of Greulich ( 1960 ) . Exposure to the ultra- soft (wavelength > 10 A) polychromatic x-rays varied from 45 to 50 minutes at 1.5 kv and 1 ma (copper target). The plates were developed in Dektol (Eastman Kodak) for 5 minutes at 20°C. Sec- tions were floated off the plates before development, remounted on glass slides, and stained with hematoxylin and eosin, or with PAS- hematoxylin. Microradiography with Soft X-Rays Bones were fixed in absolute ethanol and embedded without de- calcification in methyl methacrylate. Sections were cut on a rotary saw at 250 microns, and ground by hand on abrasive paper to about 15 microns. The sections were then placed on Kodak spectroscopic plates (649-0) and exposed to soft (wavelength < 10 A) poly- chromatic x-rays for 10 to 12 minutes at 12 kv, 19 ma ( copper target and beryllium filter) at a tube-to-film distance of 15 cm. The ex- posed plates were developed in Dektol for 5 minutes at room tem- perature. Sections were examined separately in the phase contrast microscope without removal of the embedding medium. Autoradiography Animals were injected intraperitoneally at 6 days of age with 5 fic/gm body weight of glycine-H'^ ( New England Nuclear Corpora- tion, sp. act. 194 mc/mmole) in 0.1 ml of isotonic saline, and sac- rificed at various intervals after injection. Autoradiograms of sections BONE CELLS AND BONE RESORPTION 475 were prepared using NTB2 liquid emulsion (Eastman Kodak). Following 1 week's exposure, the preparations were developed in Dektol for 2 minutes at 17°C. Some sections were stained with PAS prior to autoradiography. All sections were stained with hematoxylin following development of the autoradiograms. FiNDINOS Differences in Degree of Calcification and in Organic Mass In soft x-ra\' microradiograms of undecalcified sections, less heav- ily calcified regions were distinguished from those of greater miner- alization bv the greater exposure (blackening) of the emulsion in the former, due to the increased permeabilit\' of the matrix to the x-irradiation (Figs. 10, 13, and 20). In the immature bones and calcified cartilage of these young rats, such differences are not promi- nent. Uncalcified matrices of cartilage or of bone (osteoid) do not perceptibly absorb the soft x-rays, and thus were not distinguishable in the microradiograms. Thev could be detected by comparison of the microradiograms with the corresponding ground sections (Figs. 13 and 14). In ultrasoft microradiograms of dehydrated, decalcified sections, differences in dr)- (organic) mass content (Greulich, 1961) were reflected in the greater blackening of the underlying emulsion in regions of decreased mass (Figs. 5 and 7). Distinction beticeen Old and New Bone Differences in the relative ages of different regions of bone may often be deduced from the established sequence of events in normal bone growth. Such differences were objectively demonstrated in autoradiograms of bones from glycine-HMnjected animals ( Figs. 8, 9, and 22 ) . Bone matrix present prior to the time of injection ( "old" bone) remained unlabeled, although the walls of osteocyte lacunae within these regions became labeled if they were being remodeled at the time of injection. Bone matrix formed within 4 hours after injection was heavilv labeled, and appeared as a prominent reaction band. Matrix formed for several days thereafter was diffusely and progressively weakly labeled (cf. Young, 1962b). — 'jt^i III I 111 imiiiii !«Eaa^- ' - ■«.. iiiim.., iIihimiiiiiiiii i;' ® "''^^^ 'I Figures 1 to 5 476 BONE CELLS AND BONE RESORPTION 477 Distinction between the Matrices of Calcified Cartilage and Bone Calcified cartilage was readily distinguished from bone in ribs and tibiae by the absence of cells (Fig. 1), and by staining differences with all histological procedures used. In ultrasoft x-ray microradio- grams, the matrix of calcified cartilage was sharply demarcated from that of bone by its considerably reduced organic mass content (Figs. 6 and 7 ) . However, no appreciable differences were detected in the degree of calcification of these two matrices in the soft x-ray micro- radiograms prepared from undecalcified sections ( Fig. 19 ) . Distinction between Woven and Lamellar Bone Woven bone could be distinguished from lamellar bone (most easily in the skull) owing to differences in the matrix which are re- flected in increased basophilia, in greater irregularity of the collagen bundles ( Figs. 2 and 12 ) , and in affinity for the PAS reaction ( Figs. 4 and 22). Furthermore, in woven bone osteocytes were generally larger, closer together, and more irregularly arranged ( Figs. 2, 4, 5, 11 to 14). Woven bone, in contrast to lamellar bone, was also char- FiG. 1. Metaphyseal trabecula from the tibia of a 10-day-old rat. Note the unstained core of calcified cartilage matrix. The canaliculi of osteocytes situated within the surface coating of bone do not penetrate the cartilage. Bodian's silver. (X 800.) Fig. 2. Parietal bone of a 26-day-old rat, showing regions of woven (w) and lamellar (/) bone. Note the differences in the arrangement of collagen fibers and in the distribution of osteocytes in these two types of bone matrix. Gomori's silver. (X 800.) Fig. 3. Tibial metaphysis of a 26-day-old rat, illustrating normal, sub- periosteal resorption accompanying remodeling of the "funnel" region. Note that osteoclasts (right) are engaged in the simultaneous resorption of calcified cartilage (arrow) and bone. Gomori's silver, (x 200.) Fig. 4. Part of the parietal bone, near the parietal suture, in a 17-day-old rat. A segment of woven bone, stained heavily with periodic acid-Schiff, is indicated by an arrow. The boundaries of the diploic spaces are unrelated to the distribution of either woven or lamellar bone. PAS-hematoxylin. (X 200.) Fig. 5. Ultrasoft x-ray microradiogram of the region shown in Fig. 4. Note the greater organic mass of lamellar bone as compared with woven bone. (X200.) '"» -n^ "--'*' ' • -sr^, -^; >*'... (T) ■ ^y -'.iX' (T) '■■ ■v. f . ;i ""^.'? A ^;- r r ' ->••• Figures 6 to 10 478 BONE CELLS AND BONE RESORPTION 479 acterized by a slightly reduced organic mass content (Fig. 5), and by a slightly increased degree of calcification (Figs. 13 and 20). Sites of Preferential Resorption in Normal Rats Sites of resorption associated with growth remodeling vary among different bones, and in the same bone at different stages during maturation. However, at the costo-chondral junction of ribs, and in the metaphyseal region of tibiae, resorption not only is particularly rapid, but occurs in a characteristic and similar manner, as judged bv the presence of osteoclasts, by the uptake of glycine-H'' (which is minimal in resorptive areas: Young, 1962Z?, 196-3fl), and by the progressive removal of labeled matrix (formed elsewhere) in these resorptive zones (Figs. 8 and 9). In both ribs and tibiae of these Fig. 6. Tangential section through the shaft of a rib from an 11-day-old rat injected with 100 units of parathyroid extract 1 day prior to sacrifice. Scattered segments of calcified cartilage matrix are present. Hematoxylin and eosin. (X 320.) Fig. 7. Ultrasoft x-rav microradiogram of the region shown in Fig. 6. The cartilaginous remnants are readily apparent, owing to their lower organic mass content as compared with that of the surrounding bone matrix. The prominent nucleoli within the osteoclast nuclei (top right) are characterized by high organic mass content. Arrows indicate two osteoclasts ( or perhaps two segments of a single osteoclast) engaged in the simultaneous resorption of matrices of differing dry mass. ( X 320. ) Fig. 8. Tibial metaphysis of a 6-day-old rat sacrificed 2 hours after in- jection of glycine-H'^. The radioactive material, incorporated in large amounts l3y osteoblasts engaged in matrix formation, is concentrated in the zone of bone apposition (a), and is sparse in the zones of cartilage invasion (i) and trabecular resorption (r). Calcified cartilage matrix (arrow) remains un- labeled. Autoradiogram, PAS-hematoxylin. (X 200.) Fig. 9. Zone of trabecular resorption in the tibial metaphysis of a 7-day- old rat sacrificed 32 hours after injection of glycine-H''. Heavily reactive bone matrix, formed in the zone of bone apposition in the first 4 hours after injection, is now restricted to the distal extremities of the longest trabeculae (lower left). Resorption has removed the major part of the metaphyseal spongiosa which was present 32 hours previously. Autoradiogram, PAS-hematoxylin. ( X 320.) Fig. 10. Soft x-ray microradiogram of a segment of parietal bone from a 23-day-old rat injected with 150 units of PTE 1 day prior to sacrifice. The diploic spaces have been markedly enlarged by resorption, which has pro- ceeded without regard to regional variations in the degree of calcification of the matrix. (X 80.) ^^^fi^^'^^mt.m^. ♦13 «••■"*»•, " -«Bsr '"^^"^ A «r -'*!;, "^ >'^«^1Le.iC':-^j^^jii ^^ 1 .■^" ^ K\ -«>- _^ Figures 11 to 17 480 BONE CELLS AND BONE RESORPTION 481 young rats, there is continued removal of some ( but not all ) of the calcified cartilaginous intercellular partitions in the zone of cartilage invasion at the distal edge of the growth cartilage. (In the central regions of the metaphyseal trabeculae, in the zone of bone apposi- tion, a thin coating of bone is laid down on those cartilage partitions which escape resorption. ) Similarly, in both bones resorption occurs at the distal extremities of the mixed ( bone and calcified cartilage ) metaphyseal trabeculae in the zone of trabecular resorption, keeping pace with lengthening of the trabeculae, which occurs through vas- cular invasion of the growth cartilage (Figs. 8 and 9). On the periosteal surface of the metaphyseal regions of these bones, there is a characteristic region of resorption associated with maintaining the "funnel" shape of the tibia, and the cuwature of the ribs (Fig. 3). In the crania of these rats (e.g., the parietal bones) resorption is observed endocranially, just within the peripheral bone margin, and to a limited extent ectocranially, in the central region of the in- dividual bones. These findings, largely derived from glycine-H^- injected rats ( cf . Young, 1962Z? ) , are consistent with the distribution •Fig. 11. Parietal bone of a 20-day-olcl rat, illustrating formation of diploic spaces by resorption in a region where a segment of woven bone lies midway between the inner and outer tables. The osteoclast on the right is engaged in simultaneous resorption of woven and lamellar bone. Hematoxylin and eosin. (X 320.) Fig. 12. The same region shown in Fig. 11, photographed in polarized light. The more irregular arrangement of collagen fibers in woven bone, as compared with lamellar bone, is illustrated by the reduced refractility of the former. (X 320.) Fig. 13. Soft x-ray microradiogram of a segment of parietal bone from a 23- day-old rat. A thin remnant of woven bone (arrow), characterized by a more irregular disposition of osteocyte lacunae, is slightly more heavily miner- alized than the surrounding lamellar bone. Boundaries of the diploic space are unrelated to these minor diflFerences in the degree of calcification. ( X 200. ) Fig. 14. Phase contrast photomicrograph of the central portion of the section from which the preceding microradiogram was made. A layer of osteoid is present on the upper surface of the bone. The osteoid layer is coated by osteoblasts, and is thick enough to contain an osteocyte (arrow). Note that osteocyte canaliculi are visible within the bone, but not in the osteoid. ( X 320. ) Figs. 15 to 17. Four cells, apparently phagocytized, are shown within cytoplasmic vacuoles (arrows) in osteoclasts. Chrome hematoxylin-phloxine. (X800.) 482 H. W. YOUX(; «urv > • iiJi "Hf^: ^ ^^-*^ - ►ii^'^^^i" ■ 4B,atsLlJ Figures 18 to 22 BONE CELLS AND BONE RESORPTION 483 of osteoclasts observed on the surface of crania stained supravitally with neutral red (cf. Young, 1959). At about 10 days of age and thereafter, resorption becomes prominent within the vascular spaces of the parietal bones in association with the development of the diploe. Fig. 18. Rib shaft of a 10-clay-old rat, stained according to Bodian's silver technique. Osteogenic surfaces (above) are characterized by a granular stain- ing of the matrix. Osteocyte canaliculi are not stained within this new matrix, but are clearly visible in the underlying, older matrix. Resorptive surfaces (below) are sharply outlined, and are further distinguished by the staining of associated canaliculi. ( X 800.) Fig. 19. Soft x-ray microradiogram of the tibial metaphysis of an 11-day- old rat, injected with 100 units of parathyroid extract 1 day prior to sacrifice. Growth cartilage is on the right, marrow cavity on the left. Note the apparent fracture of the shaft (upper left). An increase in resorption in the zone of trabecular resorption (left) has resulted in a shortening of the metaphyseal trabeculae, and a similarK' accelerated resorption in the zone of cartilage in- vasion (x) has led to the apparent disengagement of some trabeculae from their normal attachment to the calcified parts of the growth cartilage. (X 80.) Fig. 20. Soft x-ray microradiogram of a part of the parietal bone of a 15- day-old rat, injected with 100 units of PTE 1 day prior to sacrifice. Endocranial resorption predominates in this region, in which development of the diploe has not yet been initiated. A thin remnant of woven bone (arrow) is slightly more heavily mineralized than the surrounding lamellar bone. Sites of prefer- ential resorption (indicated bv the irregular surface, lower left) are unrelated to these differences in calcification. The vertical line (center) results from a crack in the section. ( X 200.) Fig. 21. Phase contrast photomicrograph of the left part of the section from which the preceding microradiogram was made. Note the absence of osteoid, and the presence of an osteoclast (arrow) apparently engaged in resorbing bone of varying degrees of mineralization, (x 320.) Fig. 22. Parietal bone from an 18-day-old rat, injected with 150 units of PTE 1 day prior to sacrifice. The animal had been given a single injection of glvcine-H-^ at 6 days of age. A thin layer of woven bone, stained heavily with periodic acid-Schiff, is present centrally. The matrix of the woven bone is not labeled, indicating that it had been formed prior to the injection of glycine-H^. Within this region, however, the walls of osteocyte lacunae are reactive. Reaction bands (arrows) are present at the junction of the woven bone and the thick layers of lamellar bone above and below. These heavily labeled lamellae were formed within the first 4 hours after injection of gly- cine-H'^. The remainder of the lamellar bone (which is weakly and diffusely labeled) was formed subsequent to the injection. Accelerated resorption within the diploe is unrelated to the age or histologic organization of the matrix. Note that the ectocranial, but not the endocranial, surface (below) shows an increased affinitv for the PAS stain. Autoradiogram, PAS-hematoxylin. (X320.) ' _^^i^- A' &■ 484 R. W. YOUNG State of Matrix in Sites of Preferential Resorption in Normal Rats No relationship was apparent between sites of preferential resorp- tion and the age, degree of calcification, amount of organic mass, histological organization, or chemical variations (differential stain- ing) of the calcified matrices of bone and cartilage. For example, although no difterences could be detected among the calcified car- tilage intercellular partitions in regard to these variables, some of these partitions are resorbed in the zone of cartilage invasion, while others are spared. At the distal extremities of the metaphvseal tra- becule, resorption simultaneously removes the calcified cartilage cores and the thin osseous coating, despite difterences in the age ( the cartilage is slightly older than the bone ) , histological organiza- tion, organic mass content, and relative amounts of matrical con- stituents (Figs. 3, 8, and 9).* Similarly, in the skull, resorption of the walls of the diploic spaces involves removal of both woven and lamellar bone, without regard to their age ( the woven bone usually is older: Young, 1962i>) or any of the other variables studied, in- cluding degree of calcification (Figs. 4, 5, 11, and 13). Sites of Preferential Resorption in Rats Treated with Parathyroid Extract Bone resorption was markedly accelerated in all animals which received parathyroid extract, as indicated by the regional disappear- ance of calcified matrices in PTE-injected animals, compared with untreated controls. The greatest resorption occurred in regions which in uninjected animals could be identified as sites of normal resorp- tion. In addition, resorption tended to spread from normally resorp- tive zones into adjacent areas. For example, an increase of resorption in the zone of cartilage invasion resulted in the occasional partial or apparently complete detachment of metaphyseal trabeculae from the calcified intercel- lular partitions of the growth cartilage. At the distal end of the metaphyseal spongiosa, resorption was also accelerated, noticeably * In some species significant differences in calcification occur as well ( Owen et ah, 1955). BONE CELLS AND BONE RESORPTION 485 shortening the trabeculae (Fig. 19). Furthermore, resorption ex- tended into the mid-trabecular region, normally characterized by osseous apposition. In extreme cases, the metaphyseal trabeculae were entirely resorbed. Similarly, an acceleration of subperiosteal resorption in the metaphyseal region of both ribs and tibiae resulted in the thinning of bone at this site. Not infrequently, the shaft in this region was perforated or fractured (Fig. 19). In the skull, resorptive sites were related to the age of the animal, as in controls. In young animals (less than 10 days old), an increase in resorption (and in osteoclasts, stained supravitally with neutral red) was most marked endocranially, particularly near the bone margins. Ectocranial resorption near the bone centers also appeared to be increased. In older rats, in which development of the diploe had been initiated, the diploic spaces were noticeably widened in hyperparathyroid animals (Figs. 10 and 22). Surface resorption con- tinued to predominate endocranially (Figs. 20 and 21), but was less prominent than in younger animals. State of Matrix in Sites of Preferential Resorption in Rats Treated with Parathyroid Extract In hyperparathyroid rats, as in normal animals, there was no re- lation between sites of selective resorption and anv of the several variable features of the calcified matrices which were studied. This finding was typified by the prominent resorption within the diploe of hyperparathyroid rats more than 10 days old. The accelerated removal of bone was characterized by a progressive enlargement of the diploic spaces, without regard to whether the involved matrix was a "younger" region of lamellar bone, or an older region of woven bone with slightly less organic mass, slightly more baso- philic material, appreciably more irregularly arranged collagen, slightly greater impregnation with mineral salts, and larger, more closely packed and irregularly arranged osteocytes (Figs. 10 and 22). In some of the older rats, woven bone which was resorbed in this process had been formed several weeks earlier (Fig. 22), whereas that simultaneously removed from the metaphyseal spon- giosa was only a few hours old. 480 R. W. YOUNG The Cellular Component in Resorption The presence of osteoclasts was a characteristic, but not inevita- ble, accompaniment to demonstrable resorption. In some cases, osteoprogenitor cells (but never osteoblasts) were present in such regions. There was no evidence of decalcification (in agreement with Sognnaes, 1959), changes in organic mass (in contrast to Greulich, 1961; Takuma, 1962), or modifications in staining proper- ties of the organic matrix beneath osteoclasts. Occasionally, bone surfaces with increased affinitv for the PAS reagent were observed in animals treated with parathvroid extract (Heller-Steinberg, 1951; Gaillard, 1955^, 19555; Laskin and Engel, 1956; cf. Kroon, 1958), but this could not be correlated with demonstrable resorption ( Fig. 22). Frequentlv, a single osteoclast could be observed apparently engaged in resorbing a segment of matrix containing components of varying age, degree of calcification, histological organization, etc. (Figs. 3, 7, 9, 11, and 21). No cvtoplasmic ingestion of calcium salts by osteoclasts was detected, nor was glvcine-H^ concentrated by these cells, even when thev were present on hea\'ih' labeled bone surfaces (cf. Arnold and Jee, 1957). Osteoclasts were generallv characterized by large, pale-staining nuclei, prominent, RNA-containing (pyroninophilic, RNase-digesti- ble) nucleoli of high organic mass (Fig. 7), and variable but appreciable cytoplasmic RNA, In normal animals ( seldom in hvper- parathyroid rats ) some osteoclasts were seen with scanty cytoplasm and small, dark-staining nuclei, with insignificant nucleoli. Meta- chromatic (toluidine blue) granules and numerous small, PAS-posi- tive granules (retained after incubation in saliva) were observed within the cytoplasm of osteoclasts. Nucleated inclusions, contained within cytoplasmic vacuoles, were a prominent feature of osteoclasts in hyperparathyroid animals, but were rare in controls. The included nuclei generally had little associated cytoplasm, and were small, dark-staining or pyknotic, of high organic mass, and characteristically lobulated (Figs. 15 to 17). There is considerable variability in the size, shape, and arrange- ment of osteocytes in normal bone (Figs. 1, 2, 11, and 18). In regions of resorption, no unusual modifications in osteocytes or their sur- BONE CELLS AND BONE RESORPTION 487 roundings were observed. Occasionally, small groups of apparently enlarged osteocytes, well removed from bone sm-faces, were seen in animals treated with massive doses of PTE. This was an infrequent rather than a common finding. In these hvperparathxroid rats, how- ever, osteocytes were nniformlv depleted of their normal stores of cytoplasmic glycogen (saliva-digestible, PAS-positive material). Chondrocytic glycogen was not noticeably decreased in these ani- mals. In regions of osteogenesis in normal animals, newly formed osteo- cytes near the bone surface were characterized by large size, abun- dant cytoplasmic and nucleolar RNA, large, pale-staining nuclei, absence of glycogen, and active synthesis of bone matrix (cf. Young, 1962/;, 1963a). A surface layer of osteoid was generally discernible in such regions (Fig. 14). Osteogenic surfaces were inevitably coated with osteoblasts, and were characterized by a granular appearance of the matrix in preparations stained bv Bodian's silver technique (Fig. 18). Within this new matrix, the walls of osteocyte lacunae and canaliculi were not stained. Slightlv farther from the surface, and within all other regions of the bone, they were clearly delineated (Figs, land 18). In animals given large doses of parathyroid extract beginning 24 hours before sacrifice, no osteogenic surfaces, osteoid, or newly formed osteocytes were detected, owing to the rapid cessation of bone formation. During this process, osteoblasts became depleted of RNA, lost their ability to synthesize bone matrix (cf. Vacs and Nichols, 1962), and reverted to the osteoprogenitor state (Young, 1963fl). Osteoprogenitor cells were characterized by a propensity to fuse in resorptive areas, forming increased numbers of osteoclasts, and by regionally accentuated cell proliferation, which led to a sig- nificant thickening of the endosteum and of the cellular lining of the diploic spaces. This proliferative response was associated with the production of a network of argyrophilic fibers (cf. Kroon, 1958). Discussion The techniques used in this study have made it possible to dis- tinguish a large number of variable features within the mineralized 488 R. W. YOUNG matrices of bone and calcified cartilage. In no case was there evi- dence that any of these features was effective either in determining regions in which resorption occurred, or in modifying rates of re- sorption within these regions. The possible resistance of osteoid to resorption could not be studied, since in normal animals uncalcified matrix was uniformly present in sites of osteogenesis, and in PTE-treated animals no osteoid was detected at the intervals ex- amined. The abnormally pronounced resorption occasioned by administra- tion of parathyroid extract appeared to represent an exaggeration of normal resorption already in progress (cf. Engfeldt and Zetter- strom, 1954) and a spreading of this process into adjacent regions. This conclusion is consistent with experiments which have demon- strated that there is a lag between the time of uptake of bone-seek- ing isotopes and the time at which they are released from bone following parathyroid stimulation. At the time of administration, these osseous precursors are incorporated mainly in sites of osteo- genesis (e.g., in the metaphyseal zone of bone apposition). Later, some of the labeled sites are reached by resorptive processes (e.g., in the metaphyseal zone of trabecular resorption ) . The lag between initial uptake and subsequent normal resorption is influenced by the age (rate of growth) of the animal. In 6-day-old rats, for ex- ample, regions of apposition in the metaphysis (labeled by matrix precursors such as glycine-H'^ ) mav be found undergoing resorption as early as 4 to 6 hours later (Young, unpublished material). In older animals, this interval is progressively and appreciably length- ened. The metaphyseal zone of long bones represents a major site of resorption following parathyroid stimulation. Consequently, raised parathvroid hormone levels will accelerate the release of bone-seek- ing isotopes within hours after their administration in young animals, but will not stimulate their removal until later in older animals. Thus, findings derived from experiments of this nature using young animals (interpreted as showing an effect on new bone) and those from old animals (interpreted as showing an effect on old bone) may be brought into harmony under the explanation oflFered above, that the hormone accelerates resorption in areas already character- ized by resorptive processes. An acceleration of normal resorption BONE CELLS AND BONE RESORPTION 489 is also indicated by the finding that fluctuations in calcium levels in tissue fluids may occur within minutes after changes in para- thyroid secretion in nephrectomized animals ( Talmage, 1956, 1962 ) . This suggests an immediate effect on the rate of resorption already in progress. Since it is becoming increasingly apparent that the several types of bone cells represent different functional states of the same cell (Bloom etal, 1958; Young, 1962^, l96Sa, 1963&), and, further, that these cellular specializations are reversible, it is evident that all bone cells contain the genetic potential for bone resorption. In which of these functional specializations is this potential realized? There appears no longer to be any doubt that osteoclasts are capa- ble of resorbing bone and calcified cartilage (as well as dentin, cementum, and enamel: Sognnaes, 1959). Some reports claim, how- ever, that osteoclasts may appear after resorption has taken place (Burrows, 1938; Gaillard, 1955<7, 1955t>), or may not appear at all, despite recognizable resorption (Engel, 1952; Storey, 1957; Urist and Deutsch, 1960). Apparently, groups of bone cells in which the resorptive response has been activated tend to coalesce, creating a multinucleated cell, yet need not necessarily do so in order to carry out this function. The ability of osteoclasts to phagocytize whole cells (indicated by the presence of nucleated inclusions within cytoplasmic vacuoles: cf. Arey, 1917; Jaffe, 1933) has been noted. This suggests the presence of a variety of lytic enzymes in these cells. Many workers have brought forward evidence which suggests that osteocytes may be capable of limited resorption of their lacunar walls (Jaffe, 1933; Kind, 1951; Heller-Steinberg, 1951; Lipp, 1956; Kroon, 1958; Frost et ah, 1960). In the present material, however, no consistent alteration in osteocytes or their surroundings suggesting a significant activation of the resorptive potential was observed, even following massive doses of parathyroid extract. The loss of glycogen by osteocytes under such conditions (Heller-Steinberg, 1951; Laskin and Engel, 1956) probably represents a direct response to the hormone, since no glycogen depletion occurred in chondro- cytes. It has occasionally been stated (e.g., Follis, 1952) that osteoblasts -4i)0 R. W. YOUNG are capable of resorbiiig bone. In the present material, however, cells characterized morphologically as osteoblasts uniformly were engaged in formation (rather than destruction) of bone matrix (cf. Young, 1962/7, 1963r/). The same may be said of immature osteo- cytes, which retain osteoblast characteristics for some time after being trapped in bone. Furthermore, osteoblasts were sparse or ab- sent in PTE-injected rats in which resorption was increased. Emlothelial cells have been implicated in resorption (Jaffe, 1933; Cameron, 1961). Others have suggested that histocytes are capable of resorption ( Muhlethaler, 1953; Goldhaber, 1961) or can serve as precursors of osteoclasts (Hancox, 1956; Jee and Nolan, 1962). Conceivably, these cells (and others) represent mesenchymal spe- cializations which have retained the capacity for resorption, and are capable of activating it under suitable circumstances. A number of stimuli are known which appear capable of evoking the resorptive potential in bone cells. For example, resorption often occurs when bone is relieved of its normal, functional stresses ( Geiser and Trueta, 1958), or when its component cells have been injured or killed. Pressure on bone generally leads to its resorption ( Reitan, 1951; Storey, 1955; Young, 1959). Parathyroid hormone, cortisone (Storey, 1957; Urist and Deutsch, 1960), changing levels of sex hormones (in birds: Bloom et al., 1958), and possiblv vitamin A (Barnicot and Datta, 1956) all accelerate resorption. Diploic re- sorption may occur in response to proliferation of associated hemo- poietic tissue (Ascenzi and Marinozzi, 1958). In tissue culture, resorption is stimulated by high levels of oxygen (Goldhaber, 1961). Evidently, a wide variety of stimuli is capable of provoking a similar response from the cells of bone. This is a strong indication that the complexity lies in the reacting system, the cells themselves. The cellular response involves an integrated mobilization of meta- bolic machinery, which apparentlv precludes the simultaneous utili- zation of these resources for other specialized functions (such as bone formation ) . Elicitation of the resorptive response probablv involves the co- ordinated activation of a group of interrelated genetic loci, which code information required for the synthesis of the specific proteins (including enzymes) involved in resorption. Conceivablv, the vari- BONE CELLS AND BONE RESORPTION 491 ous resorptive stimuli might act through "regulator" genes, which, by controlling the acti\ ity of sexeral functionalh' related "structural" genes, could facilitate the evocation of integrated, complex cellular responses by relatively simple microenvironmenfal modifications (Young, 1963Z?). By microen\ironment is implied the entire complex of physical and chemical factors which impinge upon an individual cell. The functional specialization of individual bone cells is largely deter- mined by the microen\'ironment in which they are situated ( Bassett, 1962; Young, 1963a, 1963/?). The central region of the metaphyseal trabeculae, for example, represents a microenvironment which elicits activation of the osteogenic complex, whereas conditions at the distal extremity of these trabeculae predispose the affected cells to resorp- tion. At the overlapping edges of these microenvironmental "fields," intermediate cell t\'pes are found. These are apparently osteoblasts which are "de-specializing," as they come under the influence of the advancing resorptive field (cf. Young, 1962a). Bone as a tissue provides a complex system of multitudinous, graded microenvironments. A systemic stimulus to resorption, such as parathvroid hormone, probablv accentuates the tendency of these varied microen\'ironments to elicit the resorptive capacities of the involved cells. The initial response to raised hormone levels would presumably be an increase in resorptive rate by cells already en- gaged in resorption. Prolongation or accentuation of this stimulus would lead to an activation of the resorptive complex in cells at the periphery of existing resorptive fields. Such systemic resorptive stim- uli would also modifv normally osteogenic microenvironmental fields, slowing bone formation, and in extreme cases ultimately leading to its cessation. Summary Analvsis of bone and calcified cartilage by a variety of techniques has made possible the identification of calcified matrices varying in age, degree of calcification, histological organization, and chemical composition. None of these variable features proved to be critical in determining sites of preferential resorption in normal young rats. MH K. W. YOUNG or in rats in which resorption had been stimulated by the admin- istration of parathyroid extract. In the latter group, resorption ap- peared to represent an acceleration of resorption already in progress, and an extension of this process into adjacent areas. The role of the cellular microenvironment in eliciting the resorptive capacities of bone cells has been emphasized. The cellular response to appropriate microenvironmental modifications apparently involves the complex, coordinated activation of integrated portions of the cell's total ge- netic information, resulting in the onset of resorption, which then proceeds without regard to structural variations in the resorbable tissue. Acknowledgments. Parts of this work were conducted in the labora- tories of Professors R. Amprino (Bari, Italy) and A. Engstrom (Stock- holm, Sweden) during the tenure of a National Science Foundation postdoctoral fellowship (1959-1960). The remainder was supported by United States Public Health Service Grants A-4243 and D-1413. The technical assistance of Mrs. Mirdza Berzins is gratefully acknowl- edged. References Arey, L. B. 1917. Phagocytosis by osteoclasts. Anat. Record, 13, 269-272. Arnold, J. S. 1960. Quantitation of mineralization of bone as an organ and tissue in osteoporosis. Clin. Orthopaed., 17, 167-175. Arnold, J. S., and Jee, \W. S. S. 1957. Bone growth and osteoclastic activity as indicated by radioautographic distribution of plutonium. Am. J. Anat., 101, 367-417. Ascenzi, A., and Marinozzi, V. 1958. Sur le "crane en brosse" au cours des polyglobulies secondaires a I'hypoxemie chronique. Acta Haematol., 19, 253-262. Barnicot, N. A., and Datta, S. P. 1956. Vitamin A and bone. In The Bio- chemistry and Physiology of Bone (G. H. Bourne, editor), pp. 507- 538. Academic Press, Inc., New York. Bassett, G. A. L. 1962. Current concepts of bone formation. /. Bone and Joint Surg., 44A, 1217-1244. Bloom, M. A., Domm, L. V., Nalbandov, A. V., and Bloom, W. 1958. Medullary bone of laying chickens. Am. J. Anat., 102, 411-453. Bronner, F. 1961. Parathyroid effects on sulfate metabolism: Interrelation- ships with calcium. In The Parathyroids (R. O. Greep and R. V. Talmage, editors), pp. 123-138. Gharles G. Thomas, Springfield, 111. BONE CELLS AND BONE RESORPTION 493 Burrows, R. B. 1938. Variations produced in bones of growing rats by parathyroid extracts. Am. J. Anat., 62, 237-290. Cameron, D. A. 1961. Erosion of the epiphysis of the rat tibia by capil- laries. /. Bone and Joint Surg., 43B, 590-594. Carnes, W. H. 1950. The demonstration of an eflFect of parathyroid ex- tract on bone matrix. Am. J. Pathol., 26, 736 (abstract). Clark, I., and Geoffrov, R. 1958. Studies in calcium metabolism. /. Biol. Chem., 233, 203-205. Elliott, J. R., and Talmage, R. V. 1958. Removal of Ca^o and Ca*= from bone by citrate as influenced by the parathyroids. Endocrinology, 62, 709-716. Engel, M. B. 1952. Mobilization of mucoprotein by parathyroid extract. A. M. A. Arch. Pathol, 53, 339-351. Engfeldt, B., and Zetterstrom, R. 1954. Biophysical and chemical investi- gation on bone tissue in experimental hyperparathyroidism. En- docrinology, 54, 506-515. Engstrom, A., and Bergendahl, G. 1958. Note on the distribution of min- eral salts and "bone seeking" radioisotopes in spongious bone tissue. Expfl. Cell Research, 15, 265-268. Follis, R. H., Jr. 1952. Cartilage and bone matrix: Chemical structure, formation, and destruction. Trans. 4th Macy Conf. Metabolic Inter- relations, pp. 11-31. Frost, H. M., Villanueva, A. R., and Roth, H. 1960. Halo volume. Henry Ford Hosp. Med. Bull, 8, 228-238. Gaillard, P. J. 1955fl. Parathyroid gland tissue and bone in vitro. I. Exptl Cell Research, Suppl 3, 154-169. Gaillard, P. J. 1955/?. Parathyroid gland tissue and bone in vitro. III. Koninkl Ned. Akad. Wetenschap., Proc, Ser. C, 58, 286-293. Geiser, M., and Trueta, J. 1958. Muscle action, bone rarefaction and bone formation. /. Bone and Joint Surg., 40B, 282-311. Goldhaber, P. 1961. Oxygen-dependent bone resorption in tissue culture. In The Parathyroids (R. O. Creep and R. V. Talmage, editors), pp. 243-254. Charles C. Thomas, Springfield, 111. Greulich, R. C. 1960. Application of high resolution microradiography to qualitative experimental morphology. Proc. 2nd Intern. Symp. X-Ray Microscopy and X-Ray Microanalysis, pp. 273-287. Elsevier Publish- ing Co., Amsterdam. Greulich, R. C. 1961. Organic mass distribution in bone matrix under- going osteoclastic resorption. Arch. Oral Biol, 3, 137-142. Gridley, M. F. 1960. Manual of Histologic and Special Staining Technics, 2nd edition, pp. 162-163. McGraw-Hill Book Co., New York. Hancox, N. 1956. The osteoclast. In The Biochemistry and Physiology of Bone (G. H. Bourne, editor), pp. 213-250. Academic Press, Inc., New York. 494 R. W. YOUNG Heller, M., McLean, F. C, and Bloom, W. 1950. Cellular transformations in mammalian bones induced bv parathyroid extract. Am. J. Anat., 87, 315-348. Heller-Steinberg, M. 1951. Ground substance, bone salts and cellular ac- tivity in bone formation and destruction. Am. J. Anat., 89, 347-379. Irving, J. T., and Dale, J. G. 1962. The resorption of decalcified bone and of osteoid. Intern. Assoc. Dental Research, Ahstr. 40th General Meeting, No. 251, p. 67. JafiFe, H. L. 1933. Hyperparathyroidism (Recklinghausen's disease of bone). Arch. Pathol, 16, 63-112, 236-258. Jaffe, H. L., Bodansky, A., and Blair, J. E. 193k/. Fibrous osteodystrophy (osteitis fibrosa) in experimental hvperparathvroidism of guinea pigs. Arch. Pathol, 11, 207-228. Jaffe, H. L., Bodansky, A., and Blair, J. E. 19315. The sites of decalcifica- tion and of bone lesions in experimental hvperparathvroidism. Arch. Pathol, 12, 715-728. Jee, W. S. S., and Xolan, P. 1962. Origin of osteoclasts from coalescence of histiocytes. Anat. Record, 142, 310 (abstract). Jowsev, J. 1960. Age changes in human bone. Clin. Orthopaedy 17, 210- 2i8.' Kind, H. 1951. Studien zur Frage der Osteolvse. Beitr. pathol. Anat. ii allgem. Pathol, 111, 283-312. Kroon, D. B. 1958. Effect of parathyroid extract on osteogenic tissue. Acta Morphol Neerl-Scancl, 2, 38-58. Laskin, D. M., and Engel, M. B. 1956. Bone metalK)lism and bone resorp- tion after parathyroid extract. A. M. A. Arch. Pathol, 62, 296-302. Lipp, W. 1956. Neuuntersuchungen des Knochengewebes. HI. Histo- logisch erfassbare Lebensiiusserungen der Osteozyten in embryonalen Knochen des Menschen. Anat. Anz., 102, 361-372. Lotz, W. E., Talmage, R. V., and Comar, C. L. 1954. Effect of parathyroid extract administration in sheep. Proc. Soc. Exptl Biol, and Med., 85, 292-295. McLean, F. C. 1954. Biochemical and biomechanical aspects of the resorption of bone. J. Periodontol, 25, 176-182. McLean, F. C., and Urist, M. R. 1961. Bone; An Introduction to the Physiology of Skeletal Tissue, 2nd edition. University of Ghicago Press, Chicago, 111. Muhlethaler, J. P. 1953. La resorption de I'os mort etudiee par la methode de culture des tissus. Z. Zellforsch., 38, 69-77. Owen, M., Jowsey, J., and Vaughan, J. 1955. Investigation of the growth and structure of the til)ia of the rabbit by microradiographic and autoradiographic techniques. /. Bone and Joint Surg., 37B, 324-342. Pearse, A. G. E. 1960. Histochemistry, 2nd edition. J. and A. Churchill Ltd., London. BONE CELLS AND BONE RESORPTION 495 Reitan, E. 1951. Tlie initial tissue reaction incident to orthodontic tootli movement. Acta Odontol. ScancL, Siippl. 6, 5-240. Romeis, B. 1948. Mikwskopische Technik. Leibniz Verlag, Municli. Sognnaes, R. F. 1959. Microradiographic observations on demineraliza- tion gradients in the pathogenesis of liard tissue destruction. Arch. Oral Biol, 1,106-121. Storey, E. 1955. Bone changes associated witli tooth movement. Australian J. Dentistry, 59, 147-161, 209-224. Storey, E. 1957. The effect of continuous administration of cortisone and its withdrawal on bone. Australian and New Zealand J. Surg., 27, 19-30. Takuma, S. 1962. Electron microscopy of cartilage resorption bv chondro- clasts. /. Dental Research, 41, 883-889. Talmage, R. V. 1956. Studies on the maintenance of serum calcium levels by parathvroid action on bone and kidnev. Ann. N. Y. Acad. Sci., 64, 326-335. ' Talmage, R. V. 1962. Parathvroid function: A calcium replacement mechanism. Am. Zoologist, 2, 353-360. Talmage, R. V., and Kraintz, F. W. 1961. The effect of sodium chloride acidosis on parathyroid function in the rat as studied by peritoneal lavage. Gen. and Cotnp. Endocrinol., 1, 341-347. Talmage, R. V., Lotz, W. E., and Comar, C. L. 1953. Action of para- tliyroid extract on bone phosphorus and calcium in the rat. Proc. Soc. Exptl. Biol, and Med., 84, 578-582. Talmage, R. V., Toft, R. J., and Buchanan, G. D. 1959. Relationship of the parathyroids to bone phosphorus. Endocrinology, 65, 1-6. Talmage, R. V., Wimer, L. T., and Toft, R. J. 1960. Additional evidence in support of McLean's feedback mechanism of parathyroid action on bone. Clin. Orthopaed., 17, 195-205. Urist, M. R., and Deutsch, N. M. 1960. Effects of cortisone upon blood, adrenal cortex, gonads, and the development of osteoporosis in birds. Endocrinology, 66, 805-818. Vaes, G. M., and Nichols, G., Jr. 1962. Effects of a massive dose of parathy- roid extract on bone metabolic pathways. Endocrinology, 70, 546-555. Weidman, S. M., and Rogers, H. J. 1950. Studies on the skeletal tissues. 1. The degree of calcification of selected mammalian bones. Bio- chem. ]., 47, 493-497. \^^einmann, J. P., and Schour, I. 1945. Experimental studies in calcifica- tion. Am. J. Pathol, 21, 833-856. Woods, K. R., and Armstrong, W. D. 1956. Action of parathyroid extracts on stable bone mineral using radiocalcium as a tracer. Proc. Soc. Exptl. Biol and Med., 91, 255-258. Young, R. W. 1959. The influence of cranial contents on postnatal growth of the skull in the rat. Am. }. Anat., 105, 383-416. 496 R. W. YOUNG Young, R. W. 1962fl. Cell proliferation and specialization during en- dochondral osteogenesis in young rats. /. Cell Biol., 14, 357-370. Young, R. W. 1962Z;. Autoradiographic studies on postnatal growth of the skull in young rats injected with tritiated glycine. Anat. Record, 143, 1-13. Young, R. W. 1963fl. Nucleic acids, protein synthesis and bone. Clin. Orthopaed., 26, 147-160. Young, R. W. 1963/;. Specialization of bone cells. In Bone Biodynamics (H. Frost, editor). Henry Ford Hospital International Symposia (in press ) . IS Structure-Function Relationships in the Osteoclast NORMAN M. HANCOX and BRIAN, BOOTHROYD, Department of Histology, University of Liverpool, Liverpool, England IT WOULD nowadays be generally accepted that the osteoclast is actively involved in some way in the resorption of bone (Hancox, 1963), but its exact role and mode of action remain very obscure. The results to be described below bear on this problem. They come partly from morphological studies with the electron microscope, and partly from cytochemical work at the light microscope level. Materials and Methods The electron micrographs were obtained from embryonic fowl bone fixed in Palade's buffered osmium tetroxide. Araldite was used for embedding, and the results were much superior to those we had obtained in earlier work (Hancox and Boothroyd, 1961) with methacrylate. The tissue cultures to be described were simple 24- to 48-hour hanging-drop preparations (fowl plasma and embryo extract) of embryonic fowl bone. Osteoclasts wander out from the bone into the surrounding plasma (Hancox, 1946). The cultures were pre- pared by Miss S. Warner, who also performed the cytochemical tests. 497 4J)S N. M. IIANCOX AND B. IJOOTIIROYD Some of the main topographical features of a typical embryonic avian osteoclast as they would appear in a low-power electron micro- graph are shown in diagrammatic form in Fig. 1. Several zones along the line where the cytoplasm adjoins the bone can be differentiated Fig. 1. Diagrammatic representation of the main features seen in osteoclasts with low-power electron microscopy. Bone matrix represented as black, above; osteoclast below. Seven nuclei (N) are shown. The rectangles lettered A, B, and C refer to zones discussed in detail in the text; zone A is the site of the traditional "brush" or "striated" border. P, pinosome; M, mitochondrion; V, cytoplasmic vesicles. STRUCTURE-FUNCTION RELATIONSHIPS IN OSTEOCLAST 499 on the basis of striictiiral differences. Three of these will be discussed below; their main characteristics are as follows. Xone C The cytoplasm overlying the matrix is usually poor in organelles such as ribosomes, rough or smooth-walled sacs and vesicles, mito- chondria, etc.; it has a finely granular amorphous appearance (Figs. 2, 3, 4, and 6). The cell membrane follows the contour of the bone edge closely except that here and there it seems to form short, blunt channels leading inward ( Fig. 3 ) . Loose or detached bone salt crys- tals can sometimes be distinguished within these (Fig. 6). Appearances suggest that cvtoplasm in this area was neither elaborating any secretion nor pouring it forth onto the bone surface at the moment the cell was fixed. It is interesting that Dudley and Spiro ( 1961, Fig. 13 ) obtained a similar cytoplasmic picture in what they describe as "endosteal lining cells," which they regard as form- ing a more or less inert covering membrane. Zone B This constitutes a transition zone whose length is very variable; it may be practically absent. The chief characteristics are the pres- ence, first, of cytoplasmic vesicles gathered near the cell membrane, and, second, of more numerous and longer channels leading in from the surface. Bone salt crystals can be recognized in these, and they have obviously traveled some distance from the matrix ( Fig. 4 ) . Zone A This, of course, is the "brush" or "striated" border area of light microscopy (Kolliker, 1873; Kroon, 1954). Scott and Pease (1956) seem to have been the first to describe its ultrastructure. They thought it similar to the folded membrane of macrophages described by Palade ( 1955 ) and called it the "ruffled" border; this name is generallv employed now in electron microscopy. In our material the border consists of a complex system of cyto- plasmic folds and projections separated b\ cleftlike spaces of varia- ble appearance, and an associated system of cvtoplasmic vacuoles and vesicles. At one end of the scale are relativelv fine canals, often 500 N. M. IIANCOX AND B. BOOTIIROYD 3):*. FlGLULS 2 AXD 3 STRUCTURE-FUNCTION RELATIONSHIPS IN OSTEOCLAST 501 branching, which can sometimes be seen to terminate in sacUke vesicles bounded by a single membrane (Figs. 7 and 9). Generally there are multitudes of other vesicles, structurally identical, in the near-by cytoplasm. At the other end of the scale ( Figs. 2, 5, 7, 8, and 10) are broader channels leading into large vacuoles, probably pino- somes ( Hancox and Boothroyd, 1961 ) . Bone salt crystals can be recognized in both the finer and the broader channels and also in the more distal pinosomes. Their ap- pearance in such places has been described in most of the published work on the electron microscopy of osteoclasts (Scott and Pease, 1956; Gonzales and Kamovsky, 1961; Hancox and Boothroyd, 1961; Dudley and Spiro, 1961; Cameron and Robinson, 1958). The occurrence of base collagen fibrils between folds of the ruffled border does not seem to have been encountered by other workers. Collagen occurs constantly, however, in our material (Figs. 5, 7, 10, 11, and 12). The fibrils seem to protrude from the surface of the calcified matrix and are closely invested by ruffled border folds. The interfold spaces continue outward into the cytoplasm in fine chan- nels which may branch before ending in vesicular sacs. More rarely, collagen cross-banding can be identified in the larger and more distal pinosomes ( Fig. 10 ) . Partly demineralized fibrils are also seen. Discussion Turning next to the possible functional implications of these struc- tural findings, two points particularly seem worth discussing, namely, the presence of free or detached bone salt crystals and the occur- rence of collagen fibrils apparently denuded of their mineral. Figures 2 to 1.2 are electron micrographs made from unstained sections of osmium-fixed bone embedded in Araldite. Fig. 2. Low-power electron micrograph of part of an osteoclast, at right, in contact with a trabeculum of bone labeled BM. The position of zone C is indicated by that letter, and the arrow points to the amorphous cytoplasm typi- cal of this zone, which is seen at higher magnification in Figs. 3 and 6. The position of the ruffled border (RB) of zone A is arrowed. (X 8000.) Fig. 3. Bone matrix (BM) at left; apposed to it is the amorphous, inert- looking cytoplasm typical of zone C. It contains a few ribosomes and vesicles. A short, blunt cytoplasmic channel is arrowed. (X 45,000.) o02 N. M. HANCOX AND B. BOOTHROYD ;*•*•*•' 'mm #^' -4^^ Figure 4 STRUCTURE-FUNCTION RELATIONSHIPS IN OSTEOCLAST 503 Loose, detached crystals have been described within cytoplasmic channels in the ruffled border and its vacuoles in all previous pub- lished reports on the electron microscopy of the osteoclast ( Scott and Pease, 1956; Cameron and Robinson, 1958; Gonzales and Karnovsky, 1961; Hancox and Boothroyd, 1961; Dudley and Spiro, 1961). Their strict confinement to these sites and their constant occurrence seem to rule out any question of mechanical shift by the knife edge dur- ing sectioning, and one can conclude that loose crystals genuinely occur in such areas as zones A and B. Concerning zone C, however, the question of artifact has to be considered carefully. Here again, however, artifact seems unlikely; the tissue samples, of course, were supported bv and permeated with hard plastic, and if the crystals had been translocated by the knife edge one would expect to find telltale scars or tears in the Araldite, but these are not seen. In zone C, free crystals are encountered be- low what seems to be inert cvtoplasm. It is very interesting that similar loose crystals can be distinguished in Dudley and Spiro's Fig. 13 ( 1961 ) , where they are present below the inert-looking cyto- plasm of an "endosteal lining cell." This observation suggests that crystals can be detached from the surface of normal bone, or, rather, that the osteoclast is not required for the detachment of ciystals, and of course it raises the whole question of the nature of the forces which normally bind crvstals to collagen, and what is normal bone. Passing from zone C through B to A, more crystals can be seen to have been shifted and over longer distances. How the crystals are detached is uncertain. Other workers, of course, have proposed that the primarv change is solution of the collagen; if it is removed first, it would, of course, leave the crystals free. This view of the sequence of events is based chieflv on failure to detect demineralized collagen fibrils in the subjacent matrix ( Scott and Pease, 1956; Gonzales and Karnovsky, 1961; Dudlev and Spiro, 1961). However, both with Araldite in the present work and with methacrylate in our previous Fig. 4. Bone matrix runs vertically down the left margin, and osteoclast cytoplasm occupies rest of field. At top, zone C is indicated. There is a tran- sition to zone B features below; arrows indicate its short cytoplasmic channels, containing bone salt crystals. Note innumerable cytoplasmic vesicles and mito- chondria. (X 35,000.) 504 !^ ^ -4* .7- N. M. HANCOX AND B. BOOTIIROYD i m •^.^^ ■S^l' :. mr t^ Figures 5 and 6 STRUCTURE-FUNCTION RELATIONSHIPS IN OSTEOCLAST 505 study (Hancox and Boothroyd, 1961), collagen fibrils completely or partially ( Fig. 5 ) denuded of their apatite crystals occur constantly. This strongly suggests that the primary change is the removal of crystals, collagen disappearing subsequently. The reasons for this interesting discrepancy are not clear. Naturally the question of arti- fact has to be reviewed carefully v^ith respect to the possibility of loss of crystals during specimen preparation or in the electron beam. It is unlikely, however, that they would be removed from collagen in this way while persisting in near-by channels and vacuoles. All the same, this point merits further study, and Boothroyd is currently working on it. Another possibility is that a species difference is in- volved. Scott and Pease (1956) worked with kitten tissue; Gonzales and Karnovsky (1961) used rat tissue; Dudley and Spiro (1961) studied material from man and 10-week-old chickens, but give no indication in their paper of the species of the osteoclasts illustrated. If crystals indeed leave the fibril first, then presumably the forces which anchor them must be weakened or abolished. What the forces are, and how they can be affected, remains obscure. It is difficult to see how acid decalcification ( Kolliker, 1873 ) or chelation ( McLean and Urist, 1955) can be involved; these should lead to disappear- ance of the crystals rather than their shift from matrix to cell. At all events, it would follow that as demineralization proceeds, the underlying fibrils will be exposed. Those which protrude from the resorbing edge seem to become gathered up by folds in the ruffled border, and their fate, presumably, is to be digested away by enzyme action as erosion of the matrix deepens. No solid formed bodies are seen in the osteoclast cytoplasm which might be the carriers of enzymes. Dense bodies, such as have been Fig. 5. Edge of bone matrix runs vertically downward just beyond left- hand margin. The field is occupied by ruffled border of zone A. Channels, some wide and pinocytotic, and others finer, lead (horizontally) from matrix into cytoplasm. A demineralized collagen fibril (horizontal arrow) can be seen toward top; another (vertical arrow) still appears to have a few crystals at- tached. (X 30,000.) Fig. 6. Zone C. Bone matrix runs vertically down left side of figure. The typical "inert-looking" cytoplasm fills rest of field. There are several short, blunt cytoplasmic channels (arrows) apparently containing free bone salt crystals. (X 70,000.) Fig. 7. Zone A, ruffled border. Bone matri.x to left. Note pinosoiiu i /') at top, and others in varying section plane (horizontal arrows). Finer channels (vertical arrows) also lead off from matrix edge. Demineralized collagen (F) can be seen between cytoplasmic folds. ( X 70,000.) 506 Fig. 8. Pinosomes from Fig. 5 enlarged to show their contained bone salt crystals. Note ribosomes and vesicles scattered in the cytoplasm. (X 45,000.) 50'; ■mm.' ^-"liimteKjm «"' '"J Fig. y. lUillled border of zone A. Fine channels lead down from bone above. Bone salt cr)stals can be seen within. Channels probably terminate in saclike vesicles (arrows); these also contain crystals. (X 50,000.) 508 Fic; 10 BoiK' niatiix luns vertically down left margin. Ruffled border adjoins Collagen denuded of bone salt seen cut in longitudinal section (hori- zontal arrows) and, possibly, transversely (vertical arrow). A pinosome (P) is indicated. In original electron micrograph collagen cross-banding is discern- ible at arrows, but it is difficult to make out in positive print. (X 40,000.) 509 .10 X. M. HANCOX AXD B. BOOTIIROYD Fig. 11. Ruffled border folds enclosing collagen fibrils. (X 30,000.) Fig. 12. Collagen fibril surrounded by ruffled border folds. Channels lead off from the fold and end, apparently, in a saclike vesicle (arrow). (X 35,000.) STRUCTURE-FUNCTION RELATIONSHIPS IN OSTEOCLAST 511 identified with Ivsosomes, the carriers of hvdrolvtic enzvmes in Other cells ( de Man et al, 1960; Holt and Hicks, 1961 ) , are not to be seen in the osteoclast. On the other hand, as the electron micro- graphs show, the cytoplasm of this cell is characteristically packed with small vesicles bounded by a single membrane. It is natm-ally tempting to think of these as the carriers of the enzyme cocktail needed for solubilization of the collagen and the ground substance. They might be a form of, or related to, "lysosomes," or "phago- somes" (de Duve, 1959; Novikofl:", 1960), or "cytolysomes" (Novi- koff and Essner, 1962). Perhaps they empty into the cytoplasmic folds, much as "dense bodies" believed to contain hydrolytic en- zymes have been thought to coalesce with liver cell vacuoles (de Man et al., 1960). Rose (1957), of course, has watched small dark cytoplasmic bodies moving toward and fusing with pinosomes in living cells and has proposed that this may represent enzyme trans- fer. It is very unfortunate that little or nothing is known about the cytochemistry of the osteoclast at the electron microscope level. However, there is quite a lot of information from the light micro- scope, and osteoclasts are known to contain a wide range of enzymes. Leaving aside their oxidases, dehydrogenases (Walker, 1961; Bur- stone, 1960/^), and carbonic anhydrase (Simasaki and Yagi, 1960), they possess acid phosphatase (Burstone, 1960rt, 1960c,- Changus, 1957), leucine aminopeptidase (Burstone, 1960«,- Lipp, 1959), and Q:,^-glycosidase and galactosidase (Schlager, 1959, 1960). Such enzymes are often considered to be involved in the breakdown of soft tissue constituents elsewhere in the body ( de Duve, 1959; Novi- koff, 1960; Cabrini, 1961), and there seems no reason why they should not have a similar efi^ect on bone matrix once decalcified. Cytochemical observations made in our laboratory (Hancox and Warner, unpublished) on whole osteoclasts in tissue culture have provided some suggestive results. The advantage of these over sec- tioned cells is that cytological detail is very much better preserved. Thus far, acid phosphatase, leucine aminopeptidase, and yS-glucuron- idase have been studied. The reaction product is located in the cyto- plasm in the form of droplets and granules whose size and number seem to coincide quite well with those of the larger vesicles seen 51^2 N. M. IIANCOX AND B. BOOTIIROYD in electron micrographs. Obviously, the next step is to locate the reaction product in the electron microscope. A further point of interest is that the various osteoclasts present in a given culture differ in their content of a particular enzyme. Some are strongly positive, some weakly so, and others more or less negative. This perhaps might be another example of the existence of different degrees of "readiness" for phagocytosis as postulated by Holt and Hicks (1961). Summary Electron microscopy seems to indicate that bone salt crystals are somehow loosened from embryonic avian bone matrix and are swept up into channels and vacuoles in the osteoclast; the exposed collagen is enclosed in folds of the ruffled border in a way which strongly suggests a process of continuous digestion. Cytochemical work with the light microscope shows that lysosome-like hydrolytic enzymes occur in the cells. But much more information is needed, particu- larly about the shift of crystals and about the enzyme cytochemistry of the cell at the electron microscope level. References Burstone, M. S. 1960rt. Histochemical observations on enzymatic processes in bones and teeth. Ann. N. Y. Acad. Sci., 85, 431-444. Burstone, M. S. 1960L>. Histochemical demonstration of succinic dehydro- genase activity in osteoclasts. Nature, 185, 860. Burstone, M. S. 1960c. Hydrolytic enzymes in dentinogenesis and osteo- genesis. In Calcification in Biological Systems (R. F. Sognnaes, edi- tor), pp. 217-243. American Association for the Advancement of Science, Washington, D. C. Cabrini, R. L. 1961. Histochemistry of ossification. Intern. Rev. Cytol, 11, 283-306. Cameron, D. A., and Robinson, R. A. 1958. The presence of crystals in the cytoplasm of large cells adjacent to sites of bone absorption. /. Bone and Joint Surg., 40A, 414-418. Changus, G. W. 1957. Convenient histochemical phosphatase technique. Cancer, 10, 1157-1161. de Duve, C. 1959. Lysosomes, a new group of cytoplasmic particles. In STRUCTURE-FUNCTION RELATIONSHIPS IN OSTEOCLAST 513 Siibcellulor Particles (T. Havashi, editor), pp. 128-159. Ronald Press Co., New York. de iMan, J. C. H., Daems, W. Th., Willighagen, R. J. C, and Van Rijssel, Th. G. 1960. Electron-dense bodies in liver tissue of the mouse in relation to the activity of acid phosphatase. /. Ultrastriict. Research, 4, 43-57. Dudley, H. R., and Spiro, D. 1961. The fine structure of bone cells. /. Biophys. and Biochem. Cijtol, 11, 627-649. Gonzales, F., and Karnovsky, M. J. 1961. Electron microscopy of osteo- clasts in healing fractures of rat bone. /. Biophys. and Biochem. Cijtol, 9, 299-316. Hancox, N. M. 1946. On the occurrence in vitro of cells resembling osteo- clasts. /. Physiol, 105, 66-71. Hancox, N. M. 1963. In The Biology of Cells and Tissues in Culture ( E. N. Willmer, editor ) . Academic Press, London. In press. Hancox, N. M., and Boothroyd, B. 1961. Motion picture and electron microscope studies on the embryonic avian osteoclast. /. Biophys. and Biochem. Cytol., 11, 651-661. Holt, S. J., and Hicks, R. M. 1961. The localization of acid phosphatase in rat liver cells as revealed by combined cytochemical staining and electron microscopy. /. Biophys. and Biochem,. Cytol., 11, 47-66. Kolliker, A. 1873. Die normale Resorption des Knochengewebes. Vogel, Leipzig. Kroon, D. B. 1954. The bone-destroying function of osteoclasts. (Koel- liker's 'brush border.') Acta Anat., 21, 1-18. Lipp, W. 1959. Aminopeptidase in bone cells. /. Histochem. and Cyto- chem., 7, 205. McLean, F. C., and Urist, M. R. 1955. Bone; An Introduction to the Physiology of Skeletal Tissue, 1st edition. University of Chicago Press, Chicago, 111. Novikoff, A. B. 1960. Biochemical and staining reactions of cytoplasmic constituents. In Developing Cell Systeins and Their Control, pp. 167- 203. Ronald Press Co., New York. NovikoflF, A. B., and Essner, E. 1962. Cytolysomes and mitochondrial de- generation. /. Cell Biol, 15, 140-145. Palade, G. 1955. Relations between the endoplasmic reticulum and the plasma membrane in macrophages. Anat. Record, 121, 445 (abstract). Rose, G. G. 1957. Microkinetospheres and VP satellites of pinocytic cells observed in tissue cultures of Gey's strain HeLa with phase contrast cinematographic techniques. /. Biophys. and Biochem. Cytol, 3, 697. Schlager, F. 1959. Vorkommen und Lokalisation der /3-S-Galactosidase in Knochen, Knorpel und benachbarten Geweben. Acta Histochem., 8, 176-184. 514 N. M. HANCOX AND B. BOOTHROYD Schla^er, F. 1960. /3-8-Glucosicleaseaktivitat in Knochen, Knorpel iind Skeletalmuskulatur. Acta Hisiochem., 9, 320-328. Scott, B. L., and Pease, D. C. 1956. Electron microscopy of the epiphyseal apparatus. Anat. Record, 126, 465-495. Simasaki, M., and Yagi, T. 1960. Histochemistry of carbonic anhydrase with special reference to the osteoclast. Dental Bull. Osaka Univ., 1, 89-98. Walker, D. G. 1961. Citric acid cvcle in osteoblasts and osteoclasts. Bull. Johns Hopkins Hosp., 108, 80-99. 19 Bone Destruction by Multinucleated Giant Cells JAMES T. IRVINCrand CHESTER S. HANDELMAN, Harvard School of Dental Medicine and For.s\th Dental Center, Boston. IVIassachusetts THE osteoclastic ability of foreign body giant cells was noted by Bujard ( 1946 ) , who injected particles of ground-up bone under the skin or into muscles and found them surrounded by giant cells which behaved like osteoclasts. A little later Ham and Gordon (1952) implanted dead autogenous bone into muscles and observed the formation of typical osteoclasts around the implants. They con- cluded that these cells did not need to arise from osteogenic cells or from osteoblasts. Bujard considered that these osteoclasts could arise from mesenchymatous cells that were in the "etat histocytaire." In the present experiments a similar technique was used in the investigation of several problems: (1) to determine whether these giant cells had the same cytochemical properties as osteoclasts found around bone in its normal position; (2) to study the effect of chemical or biological alterations of implanted bone upon the ability of these giant cells to resorb it; in the present paper, the reaction with rachitic osteoid is described; ( 3 ) to determine whether the osteoclasis of the devitalized bone implants was dependent on the parathyroid glands. 515 olg j. t. irving and c. s. iiandelman Methods Rats. Young male animals of the Holtzman strain weighing 80 to 100 gm were used. Operative technique. Under ether anesthesia, a piece of the scapula was removed aseptically. Part was preserved for histological examination (called the reference section) and the other part was treated in various ways and ultimately implanted into the same ani- mal under the skin of the back, the implants thus being autogenous. At stated intervals the animals were sacrificed and the bone and surrounding tissue removed for examination. Some animals in each group were injected with trypan blue before the implant was re- moved, to detect the presence of phagocytic cells around the implant. Treatment of bone implants. Decalcification was carried out with 0.5 M ethylenediamine tetraacetic acid at pH 7.0 for 7 to 10 days. The bone was well washed before implantation. Some of the reference bones that had been decalcified were ashed at 900° C to determine if decalcification had been complete. Devitalization was carried out by repeated freezing to — 40°C and thawing. The presence of empty osteocyte lacunae was taken as evidence of de- vitalization. In no case was any fixative employed. Histological methods. The following were used: hematoxylin and eosin; the von Kossa method for phosphate in calcified tissues, the sections being treated with thiosulfate before mounting; the Gomori (1933) method for calcified tissues. Histoenzymologic methods. For the demonstration of en- zymes, decalcified scapulae which had been implanted for 2 weeks were frozen in isopentane, cooled by a dry ice-acetone mixture. Sections 10 microns thick were cut on a cryostat, mounted on cover- slips, and incubated at room temperature in the various substrates. The following procedures were used: for cytochrome oxidase, that of Burstone (1961a), using 8-hydroxy-l:4-naphthoquinone as a coupling agent; Burstone's method (1961^) for acid phosphatase, using naphthol AS-BI phosphate as substrate, and fast red violet LB BONE DESTRUCTION BY MULTINUCLEATED GIANT CELLS 517 salt as the coupling agent; for succinic dehydrogenase, the method of Sehgman and Rutenburg (1951); the method of Nachlas et al. (1958a) for diphosphopyridine nucleotide-linked dehydrogenases, malic, lactic, and glutamic; for DPN diaphorase, DPNH (0.5 mg/ ml) was substituted as substrate in the procedure of Nachlas et al. (1958a); for the triphosphopyridine nucleotide-linked isocitric de- hydrogenase, that of Nachlas et al. ( 1958Z? ) ; and for TPN diaphorase demonstration, TPNH (0.5 mg/ml) was substituted as substrate in the last-mentioned procedure. In the demonstration of the above diaphorases and dehydrogenases, tetranitro blue tetrazolium was substituted for nitro-BT to minimize artifactitious staining on lipid- aqueous interfaces (Rosa and Tsou, 1961). Control sections were incubated in the absence of substrate, and in the demonstration of cytochrome oxidase were treated with 10 ~^ mole of NaCN prior to incubation. All incubations were for ^ o hour except for acid phos- phatase and isocitric deh\xlrogenase, which were for 1 hour. Maxil- lae and tibiae of 3-day-old rats were processed to compare the en- zyme activities of skeletal osteoclasts with those of the giant cells about the implants. Details of the Experimental Groups Group 1. The scapula sample was removed, decalcified, and implanted 14 days later. Groups of 3 to 7 animals were killed at weekly intervals after implantation up to 8 weeks. In some groups, prior to implantation, the decalcified scapulae were blotted on filter paper and weighed to the nearest 0.1 mg, and after removal were similarly reweighed. Group 2. Rats were put on the rachitogenic diet of Jephcott and Bacharach ( 1926) for 28 days. Portions of the scapula were removed and devitalized by repeated freezing and thawing but were not decalcified. The animals were in the meantime returned to a normal diet and 2 weeks later the bone was implanted. Groups of 4 to 17 animals were killed at weekly intervals up to 4 weeks later, when the implant was removed and sectioned. 518 J. T. IRVING AND C. S. IIANDELMAN Group 3. Animals were treated similarly to those of group 2, but the bone was decalcified and implanted 2 weeks after removal. Groups of 4 or 5 animals were killed 1 to 3 weeks later. Group 4. Pieces of muscle or skin were removed from normal animals and treated with EDTA for the same length of time as the bones in group 1. After implantation, 4 animals were killed weekly up to 4 weeks later. Group 5. Animals had bacterial agar hydrocolloid implanted under the skin. Four animals were killed weeklv up to 6 weeks later. Group 6. This part of the experiment is in a preliminary stage. Decalcified and devitalized bone implants were used. Parathyroid extract (100 units, TOO gm bodv weight for 4 days) was adminis- tered to some of the animals before the implants were removed, and other animals were parathvroidectomized with a cautery just prior to bone implantation. Groups of 4 rats were killed at intervals up to 8 weeks. Results Group 1 Decalcified bone had been deliberately chosen for the study of the giant cells, since it was possible to cut the specimens without further decalcification and thus better preserve enzyme and morpho- logical details. The morphology and number of giant cells were the same as was found around devitalized calcified bone. All implants were completely devitalized as judged by empty osteocyte lacunae. They were also completely decalcified before implantation, as shown by analyses of the reference sections. None of the implant sites be- came septic, and this applied to all the implants in all groups. There was a strong small-celled reaction around each implant early in the experiment. This became less intense with time, and later a fibrous capsule surrounded the implant. After 1 week multi- nucleated cells could occasionallv be seen around the bone, occur- ring specially at the sectioned ends. By 2 weeks man\' more were seen, and to judge from slide counts, the number was maximal at BONE DESTRUCTION BY MULTINUCLEATED GIANT CELLS 519 / r^ •• * ♦* '^'>^.^*«sC". /-.•• / a1^ 7 ' . * Fig. 1. Avitogenous devitalized bone after 2 weeks of implantation. A num- ber of giant cells (arrows) are seen especially toward the ends of the bones. Note associated cellular reaction. Hematoxylin and eosin. (X 350.) Fig. 2. Autogenous bone decalcified before implantation which subse- quently recalcified (black areas). Implant was removed after 3 weeks. Note giant cell in Howship's lacuna, and the nuclei on the periphery away from the bone. Von Kossa, hematoxylin and eosin. ( X 350.) about this time (Fig. 1). Some of the bones slowly recalcified after implantation, as could be seen in the von Kossa and Gomori sections. Before this happened, however, giant cells were seen in many places in contact with uncalcilied surfaces. Later, as calcifica- tion became more extensive, these cells attacked uncalcified and calcified bone indiscriminately (Fig. 2). A number of macrophages which had engulfed trypan blue were seen around all the implants, o20 J. T. IRVING AND C. S. HANDELMAN but at some distance from the bone. In no case had the giant cells engulfed trypan blue. Almost all the giant cells of diameters up to 100 microns had up to 12 nuclei arranged circumferentially around the cell, and very few had less than 6 nuclei. They were usually much larger than osteoclasts found around bone, in which the nuclei are usually more central. The cytoplasm was generally acidophilic, and all the cells were seen tightly applied to bone. The only other for- eign bodies around which they were found were hairs which had accidentallv been introduced. The giant cells were sometimes pres- ent in a long, thin shape, around projecting parts of bone, or else were in typical Howship's lacunae, where they were circular or oval in shape. No intermediate types with 2 or 4 nuclei were convincingly seen. This might be due to the timing procedure adopted, which failed to catch cells at this stage; but it seemed that the appearance of the multinucleated cells was a very sudden process. Loss of weight of the implants was evident by 2 weeks, the time at which the giant cells began to appear in significant numbers. Thereafter it proceeded at a constant rate ( Fig. 3 ) . The giant cells about the decalcified implants were exceedingly active for all the oxidative enzymes studied, the activity of these cells being equivalent in intensity to that of the osteoclasts of the skeleton of the newborn rat (Figs. 4 to 9). Acid phosphatase ac- tivity was less intense in the giant cells about the implant than that seen in the skeletal osteoclasts (Figs. 10 and 11). In some cases, the giant cells were flattened and difficult to differentiate from the surrounding cells (Fig. 7). When an enzyme process was applied, it became immediately obvious which they were ( Fig. 8 ) . Group 2 The sections of devitalized bone from rachitic animals showed a difi^erent picture from those of group 1. The von Kossa and Gomori techniques showed that calcification of the rachitic osteoid began soon after implantation. By 1 week a few giant cells were seen on calcified surfaces, but they were never seen on osteoid till it calcified (Figs. 12 and 13). As the experiment proceeded, the bone became more heavily calcified and more giant cells were seen attacking it. BONE DESTRUCTION BY MtLTINUCLEATED GIANT CELLS 5^21 Group 3 This group of rachitic decalcified scapulae was a control on group 2 and was included to ensure that omitting decalcification was not responsible for the inability of the giant cells to attack osteoid. Giant cells appeared around the implants 2 weeks after implantation and were observed in increasing numbers with the passage of time. It was of course impossible to distinguish with anv certainty rachitic osteoid from decalcified bone, but osteoclasis proceeded with the usual vigor. i ■ — . Co from the bicarbonate medium) from a glucose substrate in considerable amounts even under aerobic conditions. Moreover, acid production was increased by previous treatment of the animals with parathyroid extract. These experiments are illus- trated in the left-hand part of Fig. 1. However, chemical analysis of the medium after incubation revealed that the acid anions which accumulated were almost entirely lactate, citrate accumulation being only about 1 70 as great. Two other observations of importance were made : First, both lactate and citrate production were increased by parathyroid treatment ( Borle, Nichols, and Nichols, 1960Z? ) ; and, second, acid production measured as lactate and citrate accumula- tion was always considerably larger than that which could be meas- ured manometrically as free H+ ion (Borle, Nichols, and Nichols, 1960a). These experiments confirmed with direct evidence Neuman's find- ing of acid production bv bone and its enhancement by parathyroid hormone, but indicated that lactic rather than citric acid was the chief end product. Since May 1959, when these experiments were performed, these observations have been confirmed many times 560 G. NICHOLS, JR. □ " or m a I P T E 1- + 1 ON LACTATE 4 0 4.0 3 0 3.0 1 2 0 2.0 1 0 1 1 0 n n CIT R ATE = 0.4 0 3 0 2 Fig. 1. The mean accumulation per hour of H+, lactate", and citrate— ions in the incubation medium surrounding fragments of cartilage-free metaphyseal bone from 50- to 60-day-old mice. All data have been calculated in terms of /xmoles accumulated per mg of alkali-soluble nitrogen ("cell nitrogen") present in the system. Incubations were carried out for 2 hours in Kiebs-Ringer media without Ca++ containing glucose 2 mg/ml buffered to pH 7.4 with bicarbonate under a gas phase of 5 per cent COo and air at 37.5°C. H+ ion accumulation was measured manometrically as CO2 displaced from the incubation medium. Lactate and citrate were measured chemically. Note that the units on the ordinate for citrate— are one-tenth those on the ordinates for the other two ions. (Data in this figure have been replotted from Borle, Nichols, and Nichols, 1960a, and Borle, Nichols, and Nichols, 1960Z?.) (Dowse and Neuman, 1961; Kenny, 1961; Cohn and Forscher, 1962; Deiss et ciL, 1962). All that remained, it seemed, was to establish the relation between acid production and Ca mobilization, and Schartum undertook to obtain the necessary evidence. Although the details of his experiments have been published (Schartum and Nichols, 1961a, 1961fo), I will review them briefly. The in vitro system seemed an excellent model, since Ca concen- trations in the medium could be considered to bear a relation to the bone fragments similar to that which the Ca concentration in extracellular fluid bears to the skeleton in vivo. In preliminary ex- IN VITRO STUDIES OF BONE RESORPTIVE MECHANISMS 561 0.80- 0.60 0.40 0.20 0.00 • SURVl VING BONES O HEAT- I NACTI VATED BONES 3 6 9 12 HOURS OF INCUBATION Fig. 2. Calcium concentrations found in incubation media surrounding fresh and heat-inactivated calvaria from 7- to 9-week-old mice after various periods of incubation. Krebs-Ringer media (pH 7.4) buffered with bicarbonate, a gas phase of 5 per cent COo and air, and glucose 2 mg/ml as substrate were used. No calcium was present in the media at the start of incubation, but phosphorus 0.4 mmole/ liter was included. Calvaria were inactivated by im- mersion for 1.5 minutes in medium previously brought to boiling. A "steady state" distribution of Ca and P between bone sample and incubation medium was assumed to hold after 4 hours of incubation on the basis of these findings. (This diagram has been reproduced from Schartum and Nichols, 1961^, with the permission of the Journal of Clinical Investigation. ) periments using calvaria from mice, Schartum was able to show, as indicated in Fig. 2, that the Ca concentration in incubation medium (Ca-free at the start of the experiment) rapidly reached a plateau, and the level maintained by fresh bone, which was producing acid, was 18 per cent higher than that maintained by inactivated bone which produced no acid. On the basis of this observation the total 5()2 G. NICHOLS, JR. medium Ca concentration at steach state in vitro could he divided into two fractions: one dependent on mineral solubility alone — the concentration maintained by inactive bone — and an additional in- crement dependent on the presence of cell metabolism. Moreover, 100 CALCIUM LACTATE COHTROUS HEATED , , „" °„ 3 , '" "", ^'uf^ " ' "">"0B. Fig. 3. Ca++ and lactate" concentrations in incubation media after 6 hours' incubation of calvaria from 7- to 9-week-old mice. Krebs-Ringer bicar- bonate-buffered media, containing no Ca++ but 0.4 mmole/liter of phosphorus, and a temperature of 37.5°C were used. The gas phase was 5 per cent COo in air or No as appropriate. Glucose when present was at a concentration of 2 mg/ml. lodoacetate when used was added to give a final concentration of 0.9 X 10-^ M. The data in this diagram have been plotted as percentages of the values obtained in simultaneous aerobic incubations of normal samples in media containing glucose and no inhibitors. The scale for calcium is on the left, for lactate on the right. (The data shown have been recalculated from Schar- tum and Nichols, 19615.) metabolic acid production and this increment in Ca concentration in the media seemed at least qualitatively related. His next series of experiments is summarized in Fig. 3. Previous experience (Borle, Nichols, and Karnovsky, 1960; Kenny, 1961) in- dicated that lactate and H+ ion accumulation in the medium could be increased by incubation under N2 and sharply reduced by the omission of glucose from the medium or by metabolic inhibitors. IN VITRO STUDIES OF BONE RESORPTIVE MECHANISMS 563 Unfortuiiatelv, manipulation of lactate accumulation by these means failed to reveal any clear relationship of lactate to Ca concentration once steady state was achieved. Although, as can be seen, iodoace- tate and the absence of glucose reduced lactate accumulation 90 per cent or more, Ca concentration declined only 10 per cent. Under a N2 atmosphere lactate was increased 20 per cent, but Ca levels were if anything slightly lower than normal for fresh bone. Vaes in experiments carried on at the same time was equally unsuccessful in demonstrating anv relation between citrate concentration in the medium and Ca levels ( Vaes and Nichols, 1961 ) . Meanwhile, Raisz and his collaborators (Raisz et ah, 1960), us- ing fresh bone and an analogous in vitro svstem, had shown that parathyroid treatment increased the level of Ca maintained in the medium. Schartum therefore determined to see how this parathyroid- induced increment in medium Ca concentration might be distributed between the two fractions of the total Ca concentration and whether it could be related to the changes in cell metabolism known to be produced by parathyroid extract. The results of these experiments are shown in Fig. 4. In essence, parathyroid treatment of the animals produced two effects in Schartum's system. First, the total Ca con- celitration maintained by fresh bone was increased (confirming Raisz's observation), but, most importantly, this effect was entirely the result of an increase in the fraction of the total which was de- pendent on the soliihilitij of the hone mineral, the fraction related to cellular metabolism remaining essentially the same. Second, al- though the total fraction dependent on cell metabolism remained constant, more of it was related to carbohydrate metabolism and the availability of glucose as substrate than was true of bone from normal animals. Despite these changes in calcium, no change in phosphorus distribution was observed in any of these experiments. These observations, we l^elieve, have important implications re- garding both the in vitro model system and the mechanisms involved in Ca and P distribution between the skeleton and extracellular fluids in vivo. In the first place, the lack of change in total medium F despite changes in Ca suggests that mechanisms are present in the skeleton which allow these two ions to be handled separately to some degree. Subsequent experiments (Nichols et ah, 1963; 564 Co % I 20 G. NICHOLS, JR. 100 90 80 7 0 -L PASSIVE SOLUBILITY NON-GLUCOSE METABOLISM ^H H GLUCOSE METABOLISM 1 1 CONTROLS PT E Fig. 4. Components of the "steady state" Ca++ concentration maintained by bone samples (calvaria) from normal and parathyroid-treated mice after 6 hours' incubation. Parathyroid-treated animals received 0.15 ml Lilly para- thyroid extract subcutaneously daily for 3 days prior to sacrifice. Conditions of incubation, etc., were the same as for Figs. 2 and 3. The fraction denoted "solubility" was taken as the concentration maintained by heat-inactivated samples (cf. Fig. 2). All data are plotted as percentages of the values obtained using fresh bone from normal animals in a complete medium. (This diagram has been reproduced from Schartum and Nichols, 1961/^, with the permission of the Journal of Clinical Investigation.) Schartum and Nichols, 1961a) using Uving bone from animals treated with other agents known to affect Ca metabolism have provided further support for this view. The results of these experiments are contrasted in Fig. 5 with those of the parathyroid experiments. These data indicate that vitamin D treatment causes an increase in both Ca and P, parathyroid extract an increase in Ca only, and cortisone no change in Ca but a notable depression of P. Although not shown in the diagram, the vitamin D effects — as in the parathyroid experi- IN VITRO STUDIES OF BONE RESORPTIVE MECHANISMS 565 VITAMIND PTE CORTISONE + 20 20 1 □ co I I + 20 -2 C Fig. 5. Differences in final medium calcium and phosphorus concentrations from simultaneously incubated controls following treatment with vitamin D, parathyroid extract, or cortisone. Calvaria from mice 7 to 9 weeks old were used in all experiments. Animals were prepared as follows: In the vitamin D experiments, 4-week-old mice were placed on a vitamin D-free synthetic diet containing approximately 1 per cent Ca and 1 per cent P for 3 to 4 weeks. Those treated with vitamin D received a single dose of 8000 units 3 days prior to sacrifice. Animals that had received no vitamin D served as controls. The data for parathyroid extract treatment have been recalculated from Fig. 4. Cortisone acetate 2.5 mg was injected subcutaneously in normal mice daily for 8 days preceding sacrifice to give the changes seen in the last section of the diagram. Conditions of incubation, etc., were as in Figs. 2, 3, and 4. (These data have been recalculated from Schartum and Nichols, 1961a; Schartum and Nichols, 1961/?; and Nichols et al, 1963.) ments — were almost entirely a reflection of change in mineral solu- bility. It is of great interest that alteration in mineral solubility rather than the fraction dependent on cell metabolism seems to determine these changes in Ca and P distribution, at least as seen in this in vitro model. So far we have found only one case where this is not true : in a mutant strain of rats which lack the ability to resorb bone yet have normal serum Ca concentrations, low serum P, and ap- parently normal parathyroids (Vaes and Nichols, 1963). Figure 6 shows that the Ca level maintained by inactive bone from normal and ia rats, as this strain has been named (Creep, 1941), is the 566 G. NICHOLS, JR. I 50- 130- H < SO- TO PASSIVE SOLUBILITY ^^M ^CELLULAR METABOLISM ^^M """""■""""■ ■ IHHI NORMAL BONE ia BONE Fig. 6. Components of the total calcium concentration maintained in incu- bation media at "steady state" by metaphyseal bone from normal and ia rats 22 days old. Conditions of incubation, etc., were as in previous figures. The large increment in the fraction of the total calciimi concentration dependent on cellular activity in ia bone can be seen to account entirely for the 60 per cent increment above controls in the total Ca concentration found in media surrounding fresh samples. (These data have been replotted from \"aes and Nichols, 1963.) same, but the level maintained by living bone is much higher in ia animals. Thus, in this case, the increment in fresh bone is entirely in the fraction dependent on cellular acti\ itv. The reason for this difference between normal and ia rats remains unknown, but met- abolic studies have shown no excessive lactate or citrate accumula- tion in the medium surrounding ia bone, although a partial defect in citrate oxidation appears to be present. Although these experiments illustrated se\'eral important features of the distribution of Ca and P between bone and its surrounding fluids, the information needed to relate cellular metabolic activity directly to Ca and P mobilization into the medium and to changes in bone mineral solubility was still lacking. In all these experiments, however, the in vitro svstem had been viewed, na'ivelv, as a two- phase system, bone and medium. It became increasingly apparent IN VITRO STUDIES OF BONE RESORPTIVE MECHANISMS 567 that it is, in fact, at least a three-phase system consisting of mineral- ized fibers covered by a very thin layer of fluid which is separated from the incubation medium by a layer of metabolicalh' active liv- ing cells embedded in a stroma of collagen and polysaccharide ground substance, as illustrated in Fig. 7. This view, which is consistent with the histology of the tissue, points out possible ways to explain such observations as the dis- crepanc)' between the accumulation of H+ ion and the accumulation of organic anions in the medium. As previously noted (Fig. 1), only 1 equivalent of H+ ion (measured by the manometric technique) was released into the medium for every 3 equivalents of lactate plus citrate. In the three-phase model the retained protons could be either retained in the layer of fluid bathing the mineralized fibers, thereby maintaining the pH of the layer at a lower level than the inculcation medium, or taken up by the hydroxyapatite crystals. Moreover, the possibility was suggested that the ratio of protons and other ions retained in the sample to those released into the medium might l)e modified by changes in cell metabolism or the per- meability of the cell layer, which might change under the influence of some of the factors studied ( Borle, Nichols, and Karnovsky, 1960 ) , Thvis the three-phase model suggested possible mechanisms for the observed modification of l)one mineral solubility, on the one hand, and for the influence of cell metabolism on the steady-state medium Ca concentration on the other, while pointing to possible reasons for the failure of organic anion concentration in the medium to cor- relate with medium Ca and P concentrations. With these views in mind, Schartum determined to seek evidence that the pH at the mineral surface was kept at a lower level than the inculcation medium bv living cells and that the presence of this H+ ion gradient was responsible for the cell-dependent fraction of the total Ca concentration in the incubation medium (Schartum and Nichols, 1962). His argument was simple: If inactiyating the cells of a bone sample abolished such a H+ ion gradient and thus lowered the steady-state Ca concentration in the medium, a similar effect should be observed if the gradient were abolished by another means in incubations using samples with actively metabolizing cells. Previous experiments ( Borle et al., unpublished data ) had indicated 568 G. NICHOLS, JR. - E C F p H 7.4 -CELu LAYER INNER FLUID LAYER pH6.8? CALCIFIED COLLAGEN Fig. 7. A schematic view of the "interface" between the calcified col- lagenous matrix of bone and the circulating extracellular fluids (ECF) (see text). In the in vitro svstem described here, the incubation medium substitutes for the extracellular fluid. As conceived at present, movement of ions between calcified matrix and extracellular fluids is driven by diffusion along concentra- tion gradients only. Such ion movement (and exchange) occurs, not directly between calcified matrix and extracellular fluid, but rather via the "inner fluid layer"; hence, a three-compartment system. The "inner fluid layer" is thought to be of very small relative volume and indeed may be partly or totally filled with an extension of the fibrillar and amorphous matrix in which the cells are embedded. Despite the ability of ions to diffuse rapidly between this inner layer and the extracellular fluids, some considerable diff^erences in its ionic composition could be maintained under steady-state conditions thanks to its very small volume in relation to cellular metabolic activitv and the high change density present in close proximity to the mineral crystals. One possible dif- ference— in pH — is indicated in the diagram with a question mark. Some evidence in favor of this idea is shown in Fig. 8. IN VITRO STUDIES OF BONE RESORPTIVE MECHANISMS 569 normal lactate production by bone cells incubated at pH's as low as 6.6. The simplest approach, therefore, was to incubate samples, both fresh and inactivated, at lower pH's and see whether the differ- ence between the two was abolished. Figure 8 shows the results ob- 25 20 I 5 - 10 - 7.6 7 4 7.2 7.0 6 8 6.6 6.4 6.2 PH MEDIUM INC U BAT ION Fig. 8. The percentage increments in total calcium concentration main- tained by fresh bone over heat-inactivated bone during aerobic incubation for 6 hours at 37.5°C in acetate buffers varying in pH from 7.1 to 6.2 have been plotted with data from incubations in bicarbonate-buffered media at pH 7.6 and 7.4. Each point represents the mean of 4 to 6 observations. (This diagram has been reproduced from Schartum and Nichols, 1962, where the details of these experiments are discussed. It is reproduced with the permission of the Journal of Clinical Investigation.) tained. It is apparent that the difference in Ca concentration be- tween fresh and inactivated samples declined very sharply below pH 7.2 and disappeared by pH 6.6, even though lactate production was unchanged — a result which fits nicely with published data on bone mineral solubility (Nordin, 1957). Thus on its first test the 570 G. NICHOLS, JR. three-phase model appeared to be vahd, and further evidence in favor of an acid mechanism for the mobihzation of Ca from the skeleton was provided. The foregoing review of our experiments with regard to bone mineral resorption has perhaps indicated the lines which our think- ing has followed and some of the difficulties we have encountered. Obviously more questions have been raised than answered, and there is patently much more to be done. It would seem, however, that such in vitro model systems not only are feasible but are prob- ably sufficiently good prototypes to offer by analogy possible ex- planations of in vivo phenomena. Especially encouraging is the evidence suggesting that the bone fragment in vitro functions as a three-phase system, indicating that it may be an even better model of the situation in vivo than we had hoped. Though the majority of our work so far has been concerned with the resorption of inorganic material, experiments have been begun to study resorption of the organic phase. The first of these grew from evidence recently obtained by Flana- gan and others that C^^-labeled proline from the incubation medium is incorporated into the stable collagen of fresh bone fragments dur- ing incubation (Flanagan and Nichols, 1962; Deiss et ah, 1962). This observation suggested a way of examining at a chemical level the notion advanced from histologic and physiologic observations (Lacroix, 1956; Jowsev, I960; Talmage et al., 1959) that bone re- sorption occurs at sites microscopically remote from areas of new bone formation, and at the same time possibly developing a method for studying collagen turnover time in bone. Since proline label in- troduced during in vitro incubation in labeled medium would be incorporated into the stable collagen only at sites of new bone for- mation, further incubation of such samples in medium containing only unlabeled proline, with examination of the collagen labeling of samples removed at various times thereafter, should indicate the biological life of the collagen deposited during the initial labeling. Woods, who is carrying on these studies, has obtained the results shown in Fig. 9, in two t\'pical experiments using metaphyseal bone from 50-day-old rats. The closed circles show the counts found in IN VITRO STUDIES OF BONE RESORPTIVE MECHANISMS 571 TIME DAYS Fig. 9. Radioactivity present in the collagen of metaphyseal bone from rats at varying intervals following a 4-hour "pulse" labeling with proline-1-C^*. All incubations were carried out in Krebs-Ringer bicarbonate-buffered media, pH 7.4, at 37.5 °C, with L-proline 0.5 mmole liter and glucose 2 mg. ml present as substrates. Penicillin 5 u/ml and streptoniNcin 0.01 mg/ml were added to prevent bacterial growth. stable collagen of ])oiie samples at times ranging from immediately after transfer of the washed samples of bone into unlabeled media to 48 hours. It can be seen that immediately after transfer, collagen radioactivity increased, confirming Flanagan's observation of trans- fer of proline label from the cells of the sample into collagen. This was followed bv a small abrupt decline in counts whose significance is not vet clear. From then on, however, collagen radioactivity re- mained quite stable sa\'e for a very slow rise. The open circles are for a longer experiment with less frequent sampling. At 4 days all the labeled collagen was still present although the cells were still metabolically active as evidenced by continued lactate production 57^2 G. NICHOLS, JR. and ability to incorporate further label if transferred back to radio- acti\e niedimn. There are, howe\'er, pitfalls in this approach. In another experiment antibiotic was omitted by error from the me- diinn, which soon developed a healthy growth of bacteria — ap- parently an organism capable of breaking down collagen, since radioactivity was rapidly lost and was almost totally gone by 48 hours. Although no true bone resorption has been demonstrated in these preliminar)- experiments, the results offer strong corroboration to the view that bone resorption occurs at sites remote from bone for- mation, and that most bone collagen once laid down has a biological life of considerable duration. Moreo\ er, it would appear likelv that the cells at sites of new collagen formation are not displaced and continue synthetic acti\it\ during prolonged experiments in vitro; certainh' the\' do not begin to resorb the newlv formed bone during this time. The contaminated experiment further supports this ^•iew of the integrit) of the synthetic sites, since when collagenoK tic bac- teria are present, the newly incorporated label is rapidly removed. Finally, Woods and I have recentlv begun to look for exidence of the collagenoK tic acti\'ity which must be present somewhere in bone. For these experiments we ha\'e used a system of macerated fresh bone incubated at 37°C under N- in phosphate buffers ranging in pll from 5.5 to 8. Tn this SNstem the cells are intact but free swim- ming rather than bound in their normal anatomical sites. Since known collagenases break up collagen into small peptides rather than individual amino acids (Gallop ct al., 1957), we have used an increase in total ultrafiltrable hvdrox\proline following incubation as an index of "collagenase-like" acti\it\'. So far only a few qualita- tive results ha\e been obtained. Howe\er, ultrafiltrable peptide hydroxyproline can be shown to increase after 4 hours of incuba- tion in this SN'stem, a phenomenon which is completelv abolished by previous heating of the s\stem to 70 °C for 5 minutes. The pH optimum of this acti\'itv is still in doubt, although it seems at present to lie between pH 6 and 7. Though onlv of the most preliminar\- sort, these few obserxations and Gross's elegant experiments ( Gross and Lapiere, 1962), which he will describe later in this s\niposium ( chapter 26 ) , encourage us to believe that soon much more evidence IN VITRO STUDIES OF BONK RESORPTIVE MECHANISMS 578 of the presence of a collagenase in bone and perhaps something of its physiological regulation will be forthcoming. Summary and Conclusions It is apparent from this review that no definitive answers can yet be derived from our experience to the perplexing question of the relationship of l^one cellular activity to bone resorption. Some evi- dence has been obtained, however, which would appear important. In the realm of mineral resorption it has been possible to show : ( 1 ) that steady-state Ca concentrations are in part dependent upon the presence of activelv metabolizing cells; (2) that part Init not all of the cellular effect is related to carbohydrate metabolism; (3) that increases in steady-state Ca concentrations ( such as those produced by parathyroid extract or vitamin D treatment ) usually are the result of increases in mineral solubility rather than in the size of the cell- dependent fraction; (4) that, nevertheless, the cell-dependent frac- tion of Ca concentration can be increased above normal under special circumstances; and finally (5) that Ca and P concentrations can potentially be regulated separately by mechanisms inherent in the skeleton itself. In addition, these experiments have suggested a model for interpreting results obtained in in vitro systems of the tvpe used here — a model which has already proved useful in supply- ing corroboration of the importance of metabolic acid production to mineral mobilization. For studies of organic matrix resorption the in vitro approach has also proved encouraging. Not only has it been possible to indicate more directlv the separation of sites of resorption from those of new matrix synthesis, but, in addition, some evidences of collagenase ac- tivity have also liegun to emerge. Moreover, Bollett ( 1962, personal communication ) and Fell et al. ( 1962 ) have indicated that mecha- nisms for removal of polysaccharide ground substance and the bone cells themselves may be investigated in such systems. Thus, it would appear, the tools for making the next steps toward understanding bone resorption at the chemical level are at hand. What must be done now is to exploit them with imagination and industry. 574 G. NICHOLS, JR. Rkfkhk.nces Bollett, A. S. 1962. Personal coninuinication. Borle, A. B., Nichols, G., Jr., and Karnovsky, M. ]. 1960. Some effects of adrenalectomy and prednisolone administration on extracellular fluid and bone composition in the rat. Endocrinology, 66, 508-516. Borle, A. B., Nichols, N., and Nichols, G., Jr. 1960fl. Metabolic studies of bone in vitro. I. Normal bone. /. Biol. Chem., 235, 1206-1210. Borle, A. B., Nichols, N., and Nichols, G., Jr. 1960/;. Metabolic studies of bone in vitro. II. The metabolic patterns of accretion and resorp- tion. /. Biol. Chem., 235, 1211-1214. Cohn, D. v., and Forscher, B. K. 1962. Aerobic metabolism of glucose by bone. /. Biol. Chem., 237, 615-618. Deiss, W. P., Jr., Holmes, L. B., and Johnston, C. C., Jr. 1962. Bone matrix biosynthesis in vitro. I. Labeling of hexosamine and collagen of normal bone. /. Biol. Chem., 237, 3555-3559. Dowse, G. M., and Neuman, W. F. 1961. Possible fundamental action of parathyroid hormone in bone. In The Faraihijroids (R. O. Greep and R. V. Talmage, editors), pp. 310-332. Charles C. Thomas, Springfield, 111. Fell, H. B., Dingle, J. T., and Webb, M. 1962. Studies on the mode of action of excess of vitamin A. 4. The specificity of the effect on embryonic chick-limb cartilage in culture and on isolated rat-liver lysosomes. Biochem. J., 83, 63-69. Firschein, H., Martin, G., Mulryan, B. J., Strates, B., and Neuman, W. F. 1958. Concerning the mechanism of action of parathyroid hormone. I. Ion gradients. /. Am. Chem. Soc., 80, 1619-1623. Flanagan, B., and Nichols, G., Jr. 1962. Collagen biosynthesis by surviving bone fragments in vitro. Federation Proc., 21, 79. Frost, H. M., Villanueva, A. R., and Roth, H. 1960. Halo volume. Henry Ford Hosp. Med. Bull, 8, 228-238. Gaillard, P. J. 1961. Parathyroid and bone in tissue culture. In The Para- thyroids (R. O. Greep and R. V. Talmage, editors), pp. 20-48. Charles G. Thomas, Springfield, 111. Gallop, P. M., Seifter, S., and Meilman, E. 1957. Studies on collagen. I. The partial purification, assay, and mode of actixation of bacterial collagenase. /. Biol. Chem., 227, 891-906. Greep, R. O. 1941. An hereditary absence of the incisor teeth. /. Heredity, 32, 397-398. Gross, J., and Lapiere, C. M. 1962. Collagenolytic activity in amphibian tissues: A tissue culture assay. Proc. Natl. Acad. Sci., 48, 1014-1022. Jowsey, J. 1960. Age changes in human bone. Clin. Orthopaed., 17, 210- 215. IN VITRO STI'DIES OF BONE RESORPTIVE MECHANISMS 575 Kenny, A. D. 1961. Citric acid production by bone. In The Pomflu/roids (R. O. Creep and R. V. Talmage, editors), pp. 275-298. Charles C. Thomas, Springfield, 111. Lacroix, P. 1956. The histological remodelling of adult bone — an auto- radiographic study. In Bone Structure and Metabolism (G. E. W. Wolstenholme and C. M. O'Connor, editors), pp. 36-44. Little, Brown and Co., Boston, Mass. McLean, F. C, and Urist, M. R. 1961. Bone; An Introduction to the Physiology of Skeletal Tissue, 2nd edition. University of Chicago Press, Chicago, 111. Neuman, W. F., Firschein, H., Chen, P. S., Jr., Mulryan, B. J., and Di Stefano, V. 1956. On the mechanism of action of parathormone. /. A??!. Chem. Soc, 78, 3863. Neuman, W. F., and Neuman, M. W. 1953. The nature of the mineral phase of bone. Chem. Rev., 53, 1-45. Nichols, C, Jr., Schartum, S., and Vaes, C. M. 1963. Some effects of vitamin D and parathyroid hormone on the calcium and phosphorus metabolism of bone in vitro. Acta Physiol. Scand., 57, 51-60. Nordin, B. E. C. 1957. The solubility of powdered bone. /. Biol. Chem., 227, 551-564. Raisz, L. C, Au, W. Y. W., and Tepperman, J. 1960. Effects of para- thyroids on bone metabolism in vitro. Clin. Research, 8, 27. Schartum, S., and Nichols, C, Jr. 1961<7. Influence of adrenal gluco- corticoids on distribution of calcium and phosphorus between bone . and its surrounding fluids. Proc. Soc. Exptl. Biol, ami Med., 108, 228-231. Schartum, S., and Nichols, C, Jr. 19615. Calcium metabolism of bone in vitro: Influence of bone cellular metabolism and parathyroid hor- mone. 7. Clin. Invest., 40, 208.3-2091. Schartum, S., and Nichols, C, Jr. 1962. Concerning pH gradients between the extracellular compartment and fluids bathing the bone mineral surface and their relation to calcium ion distribution. /. Clin. Invest., 41, 1163-1168. Talmage, R. V., Toft, R. J., and Buchanan, G. D. 1959. Relationship of the parathyroids to bone phosphorus. Endocrinology, 65, 1-6. Vaes, C. M., and Nichols, C, Jr. 1961. Metabolic studies of bone in vitro. III. Citric acid metabolism and bone solubility: Effects of para- thyroid hormone and estradiol. /. Biol Chem., 236, 3323-3329. Vaes, C. M., and Nichols, C, Jr. 1963. A comparison of the in vitro metabolism of bone from normal and a mutant strain of rats which lack bone resorption. Am. J. Physiol. ( in press) . Vincent, J., and Haumont, S. 1960. Identification autoradiographique des osteones metaboliques apres administration de Ca 45. Rev. frang. etudes din. et biol, 5, 348-353. 22 In Vitro Carbohydrate Metabolism of Bone: Effect of Treatment of Intact Animal with Parathyroid Extract BERNARD K. FORSCHER,* School of Dentistry, University of Kansas City, Kansas City, Missouri DAVID V. COHN, Kansas City Veterans Administration Hospital, Kansas City, Missouri BONE resorption induced by parathyroid extract provides a con- venient experimental svstem and has been used by a number of investigators in attempts to determine the chemical alterations re- sponsible for dissolving the mineral of bone in vivo. Although resorp- tion of bone has been observed to result from a variety of biologic stimuli, the effect of parathyroid extract is the simplest to produce experimentally, and it is hoped that elucidation of the mechanism of this phenomenon ma\' provide clues for explaining resorption initiated by some of the other stimuli. Consideration of the chemical aspects of the problem of dissolving the minerals of bone suggests three possible mechanisms. One of these would require an alteration of metabolism within the bone cells, such that an increased concentration of molecules or ions capa- ble of chelating calcium would be produced. The second would re- quire metabolic alteration leading to an increased concentration of hydrogen ions in the immediate environment of the bone crystals. " Present address: Mayo Clinic, Rochester, Minnesota. 577 578 B. K. FORSCHER AND D. V. COHN The third would involve the binding of phosphate ions by conver- sion of inorganic phosphate to organic phosphoric esters. Only the first two of these mechanisms will be considered here, as it is thought that the third is unlikely because of the high phosphatase activity of bone. Unless specifically stated otherwise, all the data mentioned in this review came from similar experiments. In general, parathyroid ex- tract was administered to animals at one or more times, the animals were sacrificed, and samples of bone were taken for in vitro metabolic studies. Control values for comparison were obtained from similar studies of samples of bone from normal, untreated animals. Let us consider first the possibility of a chelation mechanism. The binding of calcium by citrate is a well known phenomenon and is the basis for a theory to explain the demineralization of bone. The theor\' holds that an excessive production of citrate occurs locally. Although the major end product of the aerobic metabolism of glu- cose by slices of bone is lactate, there is some Krebs cycle activity ( Cohn and Forscher, 1962^ ) . The direct demonstration, in bone, of some of the Krebs cvcle enzymes also has been reported (Balogh et al., 1961; Kuhlman, 1960; Van Keen, 1959). An increased net production of citrate bv lione in tissue culture was reported ( Kenny et al., 1959) when incubations were carried out in an atmosphere of 5 per cent CO2 in oxygen, as compared with incubations in 5 per cent CO12 in air; even in this study, however, production of lactate was 50 to 100 times greater than production of citrate. Thus, there is no question that bone cells can produce citrate, but there is some question as to whether significant concentrations of citrate accumu- late and also whether parathyroid hormone influences this accumula- tion. Production and oxidation of citrate liy bone from normal mice and from mice treated with parathyroid hormone was studied by Vaes and Nichols ( 1961 ) . They found the rate of citrate oxidation in 5 per cent CO2 in air to be about 10 times greater than the rate of citrate production. Although treatment with parathyroid hormone increased the rate of production of citrate ( when further metabolism was blocked), the rate of oxidation remained sufiiciently rapid to more than handle the increased amounts of citrate. MECHANISM OF PARATHYROID HORMONE ACTION 579 The strongest support for the hypothesis that parathyroid hormone induces an increased accumulation of citrate in bone came from the in vivo work of Neuman and co-workers (Neuman et al., 1956; Firschein et al, 1958; Martin et al, 1958) and from the in vitro study bv Lekan et ah ( 1960 ) . However, subsequent work has pro- duced considerable data incompatible with this hypothesis. No evidence of an increase in the accumulation of citrate related to mineral dissolution was found either by chemical analysis (Borle et al., 1960; Raisz et al, 1961) or by radioisotope tracer methods ( Cohn and Forscher, 1961 ) . Although the citrate hypothesis should not be abandoned merely because of contradictory evidence, alter- nate possibilities should certainlv be seriously considered. The second hypothesis mentioned related to alterations in the concentration of hydrogen ions. Although the major metabolic end product in bone is an acid, lactic acid, no evidence has been found of an effect of parathvroid honnone on lactate that relates to the dissolving of minerals (Cohn and Forscher, 1962/?; Schartum and Nichols, 1961; Raisz et al, 1961). We proposed that the parathyroid hormone induces a local increase in the amount of available hydro- gen ion bv increasing the production of CO2 (Cohn and Forscher, 1962c). The nature of this effect is shown in Table I. The data from the control slices qualitatively indicate hexose monophosphate-shunt activity, which has been verified by calculations based on lactate yields ( Cohn and Forscher, 1962fl ) . TABLE I. Efffxt of Parathykoid Extract on ^Mkta holism OF Glucose-C" in Bone C"02 (cpm) Donor animal" Glucose-1-C" Glucose-6-C" Glucose-U-C" Control PTE treated Per cent increase Control PTE treated Per cent increase 17,200 23,300 35 27,300 32,100 18 3,900 6,700 72 6,100 11,300 85 11,100 16,000 45 17,600 29,300 66 " Rabbit. 580 B. K. FORSCHER AND D. V. COHN There is an important relationship between the relative magnitude of the increase in C^^02 and the position of the label in the glucose molecule. It shows that we are dealing with a specific metabolic alteration rather than a generalized increase in cellular metabolic activity or an increase in the number of metabolizing cells. Increased production of CO2 can also be demonstrated with pvruvate-2-C^^ as the substrate (Cohn and Forscher, 1961). The differential effect shown in Table I could be the result of one of two possible types of alteration. The data would be consistent with ( 1 ) an increased flow of metabolites through the Embden- Meyerhof and the Krebs pathways with no change in the rate of operation of the hexose monophosphate shunt, or with (2) an in- creased rate of operation of some pathway involving preferential oxidation of the 6-carbon of glucose. When incubations are carried out in the presence of fluoroacetate, as shown in Table II, the in- TABLE II. Effect of Fluoroacetate on Metabolism in Bone End products C»02 (cpm) Citrate-C" (cpm) Lactate (/xmoles) Donor animal Substrate Substrate" +FA'" Substrate Substrate +FA Substrate Substrate +FA Control PTE treated 4,300 650 5,400 610 630 14,000 630 10,300 4.0 3.8 4.1 3.8 " Fach flask contained 10 Mmoles glucose, labeled with 2 X 10'' cpm ghicose-6-C'S in 9.5 ml of Ivrebs bicarbonate buffer. ' Fluoroacetate, 50 ^moles added to solution of substrate. creased CO2 production due to pretreatment with parathyroid hor- mone disappears, indicating that the extra CO2 from C-6 of glucose originates in the Krebs cycle. This observation correlates with that of Raisz and co-workers ( 1961 ) that in the presence of fluoroacetate there was decreased mobilization of calcium into the medium despite an increased accumulation of citrate. They also found that iodoace- tate decreased the mobilization of calcium, in agreement with Schartum and Nichols (1961). Thus the metabolic activity respon- MECHANISM OF PARATHYROID HORMONE ACTION 581 sible for dissolving the minerals is in the Embden-Meyerhof-Krebs system, and the honnone effect also is exerted in this system. There are several apparent contradictions to our carbonic acid theory that must be resolved. Vaes and Nichols ( 1962) used glucose- U-C^^ as a substrate and could find no change in the CO2. However, they used a substrate concentration of 11.1 mM, and we have found ( Cohn and Forscher, 1962b ) that in bone the CO2 effect is sensitive to the concentration of the substrate. We have demonstrated this for glucose-U-C^^ as well as for glucose-6-C^^. With the uniform label, the effect could not be detected when the concentration of the substrate was 4 mM or more. Using manometric methods, Laskin and Engel (1956) studied the effects of pretreatment with para- thyroid extract on the metabolic activity of slices of bone and did not detect the changes in the production of CO2 which we describe as evidence of the basic action of the hormone. We also did not find these changes when we used a Warburg respirometer; as Laskin and Engel reported, we found no evidence of utilization of exogenous glucose with such incubations. In fact, were it not for the greater sensitivity of radioisotope tracer methods, one might conclude that there was no utilization of exogenous glucose. From the appearance of C" in the CO2, in lactate, and in other intermediates, we know that the added substrate is metabolized. Our hypothesis is that excess parathyroid hormone causes an in- creased production of CO2 from glucose or from metabolites derived from glucose. Calculations based on the specific activity of the added glucose and on the radioactivity of the CO2 produced show that the difference between the production of CO2 in the control and in the experimental flasks would be about 0.2 /A per hour, an increment which is not detectable in the standard Warburg respirometer. The apparent failure of in vitro slice preparations to use an exogenous substrate is known (Robinson, 1949), and we make the assumption that, in the body, exogenous glucose is a major substrate for cellular metabolism. Two alternate explanations of our data must be settled. One possi- bility is that the metabolic effect we observed was due to a general or random influence of the hormone and was not necessarily related 582 B. K. rORSCHER AND D. V. COIIN to its specific physiologic action. A second possibility is that the metabolic effect was secondary to the abnormally high concentra- tion of serum calcium associated with treatment with the parathyroid hormone. To rule out the first possibility, we performed experiments similar to those performed with slices of bone, ])ut used instead slices of kidney, liver, and muscle. Kidney is also a target organ of para- thyroid hormone, and results were similar to those obtained with bone (Table III). Other tissues from the same animals, such as liver and muscle, did not produce the CO2 effect (Table III). TABLE II L Effect of Parathyroid Extract on Production of €^■^02 BY Kidney' and Other Tissues Tissue GIuco se C»02 Experiment Position of label /imoles Control (cpm) PTE (cpm) Per cent change A Bone Kidney Liver C-1 C-1 c-1 10 10 10 4,200 18,900 4,300 5,700 23,400 3,700 36 24 -14 B Liver Heart Muscle C-6 C-6 C-6 10 10 10 1,850 750 90 1,400 410 70 -24 -45 -22 C Kidney C-1 C-6 10 10 10,500 6,200 13,600 11,200 30 81 D Kidney C-6 C-6 3 24 49,700 13,700 90,000 26,200 82 91 The second possibilitv was examined in two ways : bv experiment- ing with rabbits whose concentration of serum calcium had been artificially maintained at a high level, and bv experimenting with parathyroidectomized rats. In the first of these studies an increased concentration of serum calcium was maintained in rabbits with subcutaneous injections of a neutral suspension of calcium chloride in olive oil. After 6 hours, these animals and their controls were sacrificed. In vitro metabolic studies were then carried out on samples of bone and kidney. The data (Table IV) show two important findings. One is that the high calcium levels in the serum do not affect the metabolic pattern of MECHANISM OF PARATHYROID HORMONE ACTION TABLE IV. Effect of Serum Calcium Concentration ON Production of C"02 583 C»02 (cpm) Bone Kidney Experiment Substrate Control Increased calcium" Control Increased calcium" I Glucose-U-C" II Glucose-6-C" Glucose-1-C" 15,060 6,180 14,870 (-1.3%) 5,080 (-18%) 52,010 37,470 55,710 77,510 (+49%) 67,310 (+80%) 97,030 (+74%) " Mean serum calcium 8 mEq/liter more than in control. the bone. The other finding of significance is that of the effect of the calcium treatment on the kidney. In this tissue there was an in- crease in CO2 production, but of a nonspecific nature, since there was no differential between carbons of position 1 and position 6 of glucose. This further indicates that the differential after treatment with the hormone resulted from a metabolic alteration. The second study to rule out the effect of high concentrations of serum calcium was carried out on parathyroidectomized rats. C^O^ production from glucose-6-C^^ by slices of rat bone in vitro was not influenced by pretreatment of normal animals with parathyroid ex- tract. However, C^^02 production was reduced with bone from parathyroidectomized donors. It was normal with bone from such rats treated with parathyroid extract ( Table V ) . It is important to consider whether or not the increased CO12 TABLE V. Effect of Parathyroidectomy on Production OF C"02 BY Bone Donor animal" Pretreatment Ci^O. (cpm)^ Normal Parathyroidectomized None 200 units PTE None 200 units PTE 17,000 16,000 12,000 16,000 "Rat. <> Substrate, gluco.se-6-C". 584 B. K. FORSCHER AND D. V. COIIN production that we have described could be responsible for dissolv- ing the minerals in bone. We have no data concerning this aspect of the problem and must rely on reasoning and on evidence from other studies. From the increased radioactivity of the CO2, when the substrate was uniformly labeled glucose, we estimated the in vivo effect of the parathyroid hormone to result in a 50 per cent increase in production of CO2. Using the Henderson-Hasselbach equation, and assuming the initial ratio of salt anion to undissociated acid to be about the same as that in blood, that is, 20 to 1 for the bicarbonate-carbonic acid system, this increase in CO2 would be equivalent to decreasing the pH about 0.2 unit. The significance of such a change in pH would lie not in its magnitude so much as in its persistence. It has been suggested that the solubility of minerals of bone is very sensitive to change in pH in the region of pH 7 (Neuman and Neuman, 1958). Even if the absolute effect on solu- bility were small, living bone constitutes an open-end system in equilibrium with the blood, and a shift in the equilibrium position in the direction of the solution phase would mean a continuous drain of mineral components into the blood. This shift would be main- tained as long as the altered metabolic pattern in the bone cells re- mained in effect. Such a shift to the solution phase has been demon- strated in in vitro incubations of slices of bone from animals pre- viously treated with parathyroid hormone (Vaes and Nichols, 1962; Raisz et al., 1961; Schartum and Nichols, 1961; and others). If the effect of parathyroid hormone on bone is due to carbonic acid, could one demonstrate this by detecting a change in local pH? Probably not. In fact, two recent reports ( Raisz et al., 1961; Schartum and Nichols, 1961 ) indicate specifically that in the in vitro incuba- tions studied, the increased concentrations of calcium that devel- oped in the medium around slices of bone from animals treated with parathyroid hormone were not accompanied by related changes in pH. The reason for this is that the bone mineral acts as a buffer by dissolving in response to an increased local production of hydrogen ions. It has been suggested (Schartum and Nichols, 1962) that pH changes are involved by virtue of a pH gradient between a fluid phase in bone and the circulating fluids. MECHANISM OF PARATHYROID HORMONE ACTION 585 If the mechanism of action of a hormone on one tissue is due to a specific metaboUc alteration, and if that same hormone has an- other physiologic action on another tissue, we should make an at- tempt to explain the second action in terms of the same biochemical mechanism that was suggested for the first action. The effect of the parathyroid hormone on the kidney is related to the excretion of phosphate. Without going into the details of the renal handling of phosphate, we might suggest that possibly the ionic form of the phosphate is important. An increased local production of carbonic acid would alter the ratios of the different forms of phosphate to one another; the major change would be a shift from the monohy- drogen form to the dihydrogen form. The bulk of present evidence is in agreement concerning the necessity of cellular metabolism for dissolving minerals of bone. Much of this evidence suggests that parathyroid hormone increases the rate of some part or parts of cellular metabolism in bone. It seems more reasonable that the effect of the hormone on the struc- ture of bone is mediated by an end product of this metabolism than by an intermediate product in the metabolic process. However, the need for active metabolism does not necessarily mean that the prpcess of dissolution requires energy. It appears that the activity of the hexose monophosphate shunt is not related to the effect of the hormone, so we shall concentrate on the glycolytic and Krebs pathways. The marked accumulation of lactate, even under aerobic conditions, suggests that the limiting step in the sequence occurs after pyruvate. Since we found that the parathyroid hormone increases CO2 production with pyruvate-2-C^^ as the substrate but not with citrate-l,5-C^^ (Cohn and Forscher, 1961), and with lactate-1-C^^ as substrate but not with succinate- 1,4-C^^ ( Cohn, 1962 ) , we might narrow our attention to the sequence between pyruvate and citrate. Raisz and co-workers (1961) also did not find an effect on CO2 labeling with acetate- 1-C^^ as substrate; but, in their experiments, glucose was present in high concentrations. The CO2 production relevant to the effect of parathyroid hormone requires the operation of the Krebs cycle. We suggest that the rate of operation of the Krebs cycle is controlled by the entry of me- 586 B. K. FORSCIIF^R AND D. V. COIIN tabolites through the pyruvic acid-oxidase complex to acetyl coen- zyme A and through the condensing system or "citrogenase." Excess parathyroid hormone increases the flow through this bottleneck ( Table VI ) . This opening of a bottleneck is suggested not only by TABLE VI. Summary" of Effect of Parathyroid Extract ON Production of C"02 by Bone C'^O, (cpm ) Substrate Control PTE Per cent change Glucose-U-Ci^ 11,100 16,000 44 Glucosc-1-C" 17.200 23,300 35 Glucose-6-C'^ 6,390 10,320 62 Pyruvate-2-C" 52,000 65,800 27 Lactate-l-Ci" 27,870 40.240 44 Citrate-1,5-C" 3,900 4,300 N.S. Succinate-1,4-C^^ 16,900 10,300 -39 ° Taken from different experiments with slightly different experimental conditions. However, for each substrate, conditions were identical for tissues derived from control animals and from animals treated with paratnyroid extract. increased amounts of C'^02 from C^ ^-labeled precursors before the bottleneck, but also by the decreased amounts of C^^02 from pre- cursors after the bottleneck, which decrease presumably results from dilution secondary to increased flow. There is no real accumulation of metabolic citrate, since the reactions of the Krebs cycle follow in sequence. If more citrate is produced in the cell, more is oxodized. The real consequence of increasing the flow through the bottleneck is a greater production of CO2. We do not yet haye sufficient data to establish clearly which of the two enzyme systems between pyruvate and citrate is the actual controlling factor. At this time we might tentatively postulate that the crucial point in the sequence, and the site of hormone action, is in the pyruvic acid-oxidase complex. This suggestion derives from the fact that lactate, which accumulates in large amounts, is the major end product. If the conversion of pyruvate to acetyl coenzyme A proceeded rapidly, and the condensation to citrate were the limit- ing reaction, acetate or some other derivative of acetyl coenzyme A should be the accumulating end product. MECHANISM OF PARATHYROID HORMONE ACTION 587 SlMMAUY We propose that an excess of parathyroid hormone inckices an increased production of CO2 from gkicose or metaboHtes of glucose by cells in bone. This theory is in accord with data from many in- dependent studies. Of the various h}potheses posed to explain the mechanism of action of parathyroid hormone, this concept explains the present evidence in a more satisfactory manner than theories based on the accumulation of citrate or lactic acid. Acknowledgments. These studies were supported in part by Grants D-836, D-1197, D-1523, and A-1132 from the United States Public Health Service. We acknowledge the technical assistance of Mr. Raymond Newman, Miss Anna Smaich, and Mrs. Betty Young. References Balogh, K., Jr., Dudley, H. R., and Cohen, R. B. 1961. Oxidative enzyme activity in skeletal cartilage and bone: A histochemical study. Lab. Invest., 10, 839-845. Borle, A. B., Nichols, N., and Nichols, C, Jr. 1960. Metabolic studies of • bone in vitro. II. The metabolic patterns of accretion and resorption. /. Biol. Chem., 235, 1211-1214. Cohn, D. V. 1962. Unpublished data. Cohn, D. v., and Forscher, B. K. 1961. The effect of parathyroid hor- mone on the oxidation of carbohydrate by bone. Biochim. et Biophys. Acta, 52, 596-599. Cohn, D. v., and Forscher, B. K. 1962fl. Aerobic metabolism of glucose by bone. /. Biol. Chem., 237, 615-618. Cohn, D. v., and Forscher, B. K. 1962Z?. Effect of parathyroid extract on the oxidation in vitro of glucose and the production of C^^02 by bone and kidney. Biochim. et Biophys. Acta, 65, 20-26. Cohn, D. v., and Forscher, B. K. 1962c. The influence of parathyroid ex- tract on the production of CO2 from glucose by bone and kidney. Federation Proc, 21, 207 (abstract). Firschein, H., Martin, C, Mulryan, B. J., Strates, B., and Neuman, W. F. 1958. Concerning the mechanism of action of parathyroid hormone. I. Ion-gradients. /. Am. Chem. Soc, 80, 1619-1623. Kenny, A. D., Draskoczy, P. R., and Coldhaber, P. 1959. Citric acid pro- duction bv resorbing bone in tissue culture. Am. J. Physiol., 197, 502-504. 588 15. K. FORSCHER AND D. V. TORN Kuhlman, R. E. 1960. A niicrochemical study of the developing epiphyseal plate. /. Bone and Joint Surg., 42A, 457-466. Laskin, D. M., and Engel, M. B. 1956. Bone metabolism and bone re- sorption after parathyroid extract. A. M. A. Arch. Pathol, 62, 296-302. Lekan, E. C, Laskin, D. M, and Engel, M. B. 1960. EflFect of para- thyroid extract on citrate metabolism in bone. Am. J. PhysioJ., 199, 856-858. Martin, G. R., Firschein, H. E., Mulryan, B. J., and Neuman, W. F. 1958. Concerning the mechanism of action of parathyroid hormone. II. Metabolic effects. /. Am. Chem. Soc, 80, 6201-6204. Neuman, W. F., Firschein, H., Chen, P. S., Jr., Mulryan, B. J., and Di Stefano, V. 1956. On the mechanism of action of parathormone. /. Am. Chem. Soc., 78, 3863-3864. Neuman, W. F., and Neuman, M. W. 1958. The Chemical Dynamics of Bone Mineral, pp. 32-33. University of Chicago Press, Chicago, 111. Raisz, L. C, Au, W. Y. W., and Tepperman, J. 1961. Effect of changes in parathyroid activity on bone metabolism in vitro. Endocrinology, 68, 783-794. Robinson, J. R. 1949. Some effects of glucose and calcium upon the metabolism of kidney slices from adult and newborn rats. Biochem. J., 45, 68-74. Schartum, S., and Nichols, C, Jr. 1961. Calcium metabolism of bone in vitro: Influence of bone cellular metabolism and parathyroid hor- mone. 7. Clin. Invest., 40, 2083-2091. Schartum, S., and Nichols, C, Jr. 1962. Concerning pH gradients between the extracellular compartment and fluids bathing the bone mineral surface and their relation to calcium ion distribution. /. Clin. Invest., 41, 1163-1168. Vaes, G. M., and Nichols, C, Jr. 1961. Metabolic studies of bone in vitro. III. Citric acid metabolism and bone mineral solubility: Effects of parathyroid hormone and estradiol. /. Biol. Chem., 236, 3323-3329. Vaes, G. M., and Nichols, G., Jr. 1962. Effects of a massive dose of para- thyroid extract on bone metabolic pathways. Endocrinology, 70, 546- 555. Van Reen, R. 1959. Metabolic activity in calcified tissues: Aconitase and isocitric dehydrogenase activities in rabbit and dog femurs. /. Biol. Chem., 234, 1951-1954. 23 Metabolic Action of Parathyroid Hormone on Rat Calvaria C. M. DOWSE, M. W. NEI MAN, K. LANE, and W. F. NEUMAN, Department of Radiation Biology, School of Medicine and Dentistry, University of Rochester, Rochester, New York DESPITE a long history of polemical argument, it is now generally agreed that parathyroid hormone has a direct action on bone, caus- ing an increase in the population of resorbing cells, the osteoclasts, a mobilization of bone mineral, and perhaps a general increase in the availability of the bone mineral, at least for exchange with parenterally administered isotopes (Creep and Talmage, 1961; Neu- man and Neuman, 1958). More recently, attention has focused on possible hormonally induced changes in cellular metabolism in bone. Presumably an understanding of these metabolic changes might clarify the biochemical mechanisms by which the hormone exerts its osteolytic eflFects. Bone, however, is so obviously unsuited for conventional bio- chemical studies of metabolism that progress has been rather limited and published reports have been few (Laskin and Engel, 1956; Borle et al, 1960a, 1960Z?; Lekan et al, 1960; Neuman and Dowse, 1961; Cohn and Forscher, 1961, 1962fl, 1962Z?; Vaes and Nichols, 1961, 1962; Schartum and Nichols, 1961, 1962). In most cases crushed or sliced metaphyses have been employed despite the obviously serious contamination by marrow cells. The investigations reported below were conducted on the calvarium of the newborn rat. 589 590 DOWSE, NEUMAN, LANE, AND NEUMAN Calvaria have been extensively employed in a variety of studies designed to show histological effects of parathyroid hormone ( Gail- lard, 1959 ) , resorption in surviving culture ( Goldhaber, 1960 ) , and the cellular role in maintaining calcium levels above those due to passive solubility only (Raisz et ah, 1961). However, no comprehen- sive biochemical studies of calvaria have to our knowledge yet ap- peared. This is a ready-made tissue slice with a minimum of cut or damaged cells. It has some mineralized matrix, but the cell popula- tion, for bone, is high. There is but little contamination with marrow cells. Morphologically, however, at least five general cell types pre- dominate: osteoblasts, osteocvtes, osteoclasts, cartilage cells, and preosseous mesenchymal cells. The principal advantages of this tissue are convenience, reproducibilitv, and responsiveness to parathyroid hormone in vitro (Gaillard, 1959). Nonetheless, the calvarium still represents something of a compromise, and it is therefore emphasized that the studies to be reported are on the metabolism of calvarium. Though the data matj apply to bone cells in general, no such claim is made. Experimental Procedure Randomized newborn rat pups were decapitated within 15 hours of birth. The calvarium, a round section from the top of the skull, including most of the parietal and parts of the frontal and occipital bones, was quickly removed, freed of bits of connective tissue, and halved at the midline. Left and right halves were placed alternately in each of two weighed collecting bottles containing ice cold in- cubation medium. Wet weight was used as the basis of reference, since it was found at least as reliable as dry weight, noncollagen nitrogen ( NGN ) , or deoxyribonucleic acid ( DN A ) . The medium emplo\ed was either Krebs-Ringer phosphate or bicarbonate with the calcium reduced to a level of 1.2 mM or ab- sent. Incubations were performed in a Warburg apparatus at 37 °G after equilibration with the appropriate gas. Gontrol and experi- mental measurements were made on paired flasks containing 6 to 10 alternating halves of the same calvaria. PARATHYROID HORMONE AND RAT CALVARIA METABOLISM 591 Methods Staiidaid methods were employed for calcium (Chen and Tori- bara, 1953), PO4 (Chen et ah, 1956), lactate (Barker and Summer- son, 1941), citrate (Ettinger et a]., 1952), carbonate (Umbreit et al., 1959), NCN (Lilienthal et al, 1950), DNA (Leslie and Davidson, 1951), glucose (Saifer and Gerstenfeld, 1958), and glycogen after alcohol precipitation (Umbreit et al., 1959). The calvarium was selected for study because it was presumed that its cell population would consist primarily of osteocytes and osteoblasts. Histological examination of the tissue, however, showed a significant number of preosseous mesenchymal cells. How much of the characteristics of osteoblasts the latter have assumed is un- determinable. TABLE I. General Analytical Findings on Calvaria (All concentrations given in terms of wet weight) Average wet weight 15.0 ± 0.2 mg Wet wt/dry wt ratio 4.63 d= 0.05 Calcium 33.8 dz 0.6 mg/gm Phosphorus 17.9 zb 0.2 mg/gm Ca/P mole ratio 1.47 ± 0.05 Citrate 1.08 ± 0.08 mg gm Carbonate (as CO2) 3.7 ± 0.1 mg gm DNA 1.78 ± 0.03 mg/gm NCN 9.2 ± 0.1 mg/gm Table I summarizes some of the tissue analyses performed. The values are given plus or minus the standard error of the mean, and represent at least 25 analyses. They indicate that, chemically, the tissue is intermediate between soft tissue and mature bone. DNA and NCN were at first considered as possible bases for met- abolic reference, but the constancy of the ratio they bore to wet weight rendered these measurements unnecessary. Further, in the case of NCN, the finding that no less than 50 per cent of the NCN originally present in the tissue passed into the medium during in- cubation made this parameter, in our view, of questionable value. It will be seen that the oxidative and glycolytic properties of this 592 DOWSE, NEUMAN, LANE, AND NEUMAN tissue are distinctly different from those of most soft tissue and must be representative, in a qualitative way, of new bone or bone-forming tissue. Results and Discussion Because of the many metabolic aspects to be covered, a discussion of each is presented along with the results. Oxidation The rate of oxygen consumption of calvaria in phosphate medium containing no added substrate is presented in Fig. lA. Despite the HOURS HOURS Fig. 1. Oxygen consumption of calvaria incubated in phosphate medium in vitro. Qq.-, expressed as /imoles per gm wet weight. A, data showing the continued hnear oxygen consumption over an 8-hour period. B, data showing the enhanced oxygen consumption from the addition of 10 mM succinate (S) and subsequent inhibition by malonate (M); numbers indicate concentration in mM. C, data showing the effects of additions of the inhibitor fluoroacetate; numbers indicate its concentration in mM. PARATHYROID HORMONE AND RAT CALVARIA METABOLISM 593 constancy of the rate over an incubation period of 8 hours, all sub- sequent studies were conducted for 3 hours or less. The average Qo.2 of 10 different experiments, 18.4 ± 2.5 /u,moles/gm wet wt/hr., is intermediate between earlier values reported for metaphysis and for a truly active soft tissue such as liver: Qot Ainioles/mg Qot Mmoles/mg NCN DNA Metaphysis (Borle et al, 1960a, 19606) 1.1 Metaphysis (Laskin and Engel, 1956) 57 Calvarium 2.0 230 Liver (Spector, 1956) 850 No effect on the Qoo could be found upon addition of a variety of substrates, including glucose, pyruvate, acetate, fumarate, malate, oxaloacetate, citrate, and alpha ketoglutarate either singly or in various combinations best calculated to stimulate oxygen uptake. Similar findings were reported bv Laskin and Engel using meta- physeal preparations (Laskin and Engel, 1956). Succinate was the only substrate found to have any effect (Fig. IB) : a 50 per cent increase at a concentration of 10 mM or higher. This stimulation was abolished by the subsequent addition of mal- onate at an equal or higher concentration. Note that, whereas the succinate effect is immediate, the malonate effect takes some 15 to 20 minutes to develop. Because of its ability to participate in sub- strate-linked oxidative phosphorylation independent of the pyr- idine nucleotides, however, succinate occupies a special place in the Krebs ( TCA ) cycle, and stimulation of oxygen uptake in a tissue upon its addition is not evidence of increased TCA cycle activity, but only of the availability of an excess capacity in the succinic de- hydrogenase-flavoprotein system. Malonate does more than abolish the increase due to succinate. It produces, in turn, a further reduc- tion in the Q02 which, relatively to the endogenous rate, is equal to that produced by malonate additions alone ( not shown ) . In experi- ments both with and without succinate, a concentration effect of malonate was noted: 1 mM not affecting the Q0.2, 30 niM reducing it by some 25 to 30 per cent, 10 niM being intermediate. The con- 594 DOWSE, NEUMAN, LANE, AND NEUMAN centratioii dependency of the effect is to be expected for a competi- tive inhibitor of this type. Addition of fmnarate, after malonate, at levels up to twice that of the malonate concentration used, did not relieve the l^lock, nor did addition of oxaloacetate. The latter was of interest because malonic acid has been shown to be an inhibitor of fumarase as well as of succinic dehydrogenase ( Massev, 1953). It would appear, therefore, that malonic acid was eliminating, bv a competitive inhibition of succinic deh\ drogenase, O- uptake due to succinate oxidation, l)ut was not reducing the availabilitv of dicar- boxylic acids for the functioning of the rest of the cycle. A single addition of fluoroacetate at the 5-mM level had no effect on endogenous Qn.^ (Fig. IC), whereas a second addition at the same concentration, or single additions at the 10- or 20-mM level, reduced the Qi,.^ by some 25 per cent. There was no greater effect at the highest concentration. This, as well as the delav period before the effect became manifest, is consistent with the noncompetitive nature of the inhibition, and with the necessity for prior incorpora- tion of the fluoroacetate into the TCA cycle to form fluorocitric acid. The continued oxygen uptake, with fluoroacetate at the 20-mM level, could be due to an incomplete block at the aconitase stage, perhaps because of an adequate supply of endogenous pyruvate. If, instead of being added during the incubation, 10-mM fluoroacetate was pres- ent from the beginning, the reduction in Qi)., could be prevented b\' the coincident addition of p\'ruvate at 10 mM, but not of fumarate at 10 niM or glucose at 17 niM (Table II). This protective property TABLE IT. Thk Effects of Fluoroacetate and Substrates ON Oxygen Consumption (Phosphate medium; 100 per cent Oo, 2-hour incubation; averages of 4 to 8 flasks) Qo, (Mmoles/gm wet wt/hr ) Fluoroacetate 10 mn No fluoroacetate Additions No glucose Glucose 17 mM No glucose Glucose 17 mM None Fumarate 10 niM Pyruvate 10 niM Fumarate 10 niM : pyru\ •ate 10 niM 11.6 9.8 18.2 18.2 10.7 11.2 17.4 17.8 17.4 17.4 17.8 17.8 17.4 17.0 17.4 17.0 PARATHYROID HORMONE AND RAT CALVARIA METABOLISM 595 of pyruvate is most satisfactoriK' explained in terms of a competition iDetween pyruvate and fiuoroacetate for coenzvme A, for it is well known that if p\'ru\'ate and fiuoroacetate are present together, the greater amount of carbon incorporated into the cvcle is derived from the pyruvate. As a consequence of its inhibitor action upon aconitase, fiuoroace- tate poisoning of a tissue with a functioning TCA cycle usually results in the accumulation of citric acid. Because of the large quan- tity of citric acid which is present in bone mineral, however, in- creased citrate concentration in response to fiuoroacetate injection in vivo has not been shown to occur (Beaulieu, 1953). In a closed system in vitro, changes in citrate content are easily demonstrable. Analyses of the medium and tissue showed that glucose addition did not increase citrate accumulation, whereas, as shown in Table III, fumarate on the one hand and fiuoroacetate on the other, alone TABLE III. Thp: Effect of Fluoroacetate and Substrates ON Citrate" and Lactate Production (Bicarbonate medium; glucose 17 mu; fumarate 5 mM; fluoroacetate 5 mM; 2-hour incubation; averages of 4 to 8 flasks) Lactic acid Citric acid (jumoles/gm wet wt) production (juinoles/gm Medium Tissue Total wet wt) Glucose 0.9 3.8 4.7 30.4 Glucose : fumarate L7 4.6 6.3 31.3 Glucose : fluoroacetate L2 5.2 6.4 29.8 Glucose : fumarate : fluoroac •etate 2.6 5.1 7.7 30.2 Original tissue content 5.6 0.2 " Note that with ghicose only, citrate is actually utilized despite its appearance in the medium. Because of ease of exchange of citrate between tissue and medium it is necessary to analyze the system /'/( toto. or in combination, did bring about an increase in total citrate. Even with fumarate and fluoroacetate present at the same time, the rate of appearance of citric acid in the medium was very considerably less ( 15 times ) than that of lactate. Lactic acid production was not significantly affected by the various additions, except, of course, by the addition of glucose. 590 DOWSE, NEUMAN, LANE, AND NEUMAN There would seem, considering all the fragmentary data collected here and elsewhere (Cohn and Forscher, 1962a, 1962i>; Lekan et al., 1960), to be no reason to doubt the presence of the TCA cycle in bone. In the calvarium it is present from succinate to citrate, and citrate can be further metabolized. The failure to show effects on the Q02 by addition of intermediates is not really surprising, for it is necessary to deplete the endogenous supply of substrate in a tissue before this can be achiev^ed. Subsequent addition of the inter- mediate, then, merely results in a return of the Qo._, to the initial yalue. There is also, in the data presented, no reason to conclude that the oxygen uptake results from s) stems other than the TCA cycle, for in a tissue exhibiting a continuing aerobic glycolysis (see below) the ayailability, of pyruyate at least, for the TCA cycle will not be limiting. Glycolysis Like other "bone" preparations (Cohn and Forscher, 1962/?; Laskin and Engel, 1956; Borle et a]., 1960fl, 1960/?), calyaria show high rates of glycolysis eyen under aerobic conditions. With no added substrate, glycolysis (defined as lactate production) de- creases with time; with added glucose, the rate is linear for at least 3 hours (Fig, 2A). Anaerobicalh', the initial glycolytic rates are 2 to 2.5 times higher. Though added glucose maintains the initial rate for 3 hours, in the absence of added substrate the rate falls off even more rapidly than it does aerobically. Thus, after 2 or 3 hours' in- cubation total lactate accumulation is about the same, anaerobically or aerobicall)'. This may account for the failure of some investigators to observe the Pasteur effect (Lekan et ah, 1960). When shorter periods of incubation are employed, the effect of oxygen on endogenous glycolysis is readily apparent, as seen in Table IV. Here periods of 30 and 60 minutes were used, and a Pasteur ef- fect of some 70 per cent was found. Lactate production increases with increasing concentrations of glucose up to about 4 niM (Fig. 2B). Thereafter the Ql increases less rapidly, especially in the bicarbonate medium. A marked effect of the buffer present, seen in Fig. 2B, is further demonstrated in Table IV. In the bicarbonate-buffered medium, the PARATHYROID HORMONE AND RAT CALVARIA METABOLISM GLUCOSE 597 4 8 GLUCOSE mM Fig. 2. Lactate production by calvaria incubated in vitro. A, the Qj^ in fimoles per gm wet weight with time in the presence and absence of added glucose. B, the variation in Qj^ as /j.moIes/gm/hr. with varying glucose con- centration in both phosphate and bicarbonate media. C, the variation in Q^^ with variation in pH; phosphate medium, glucose 17 mxi. bicarbonate ion concentration was 25 niM, phosphate 1.2 niM; in the phosphate medium, phosphate was 16 niM, bicarbonate being absent. Interestingly enough, the medium effect was missing under anaerobic conditions (Table IV), lactate production in both cases TABLE IV. The Effect of Medium on Lactate Production (Mean values of six flasks) Lactic acid production (jumoles/gm wet wt) 30-minute incubation 60-minute incubation Medium Aerobic Anaerobic Aerobic Anaerobic Bicarbonate Phosphate 3.2 5.0 10.8 10.9 5.8 7.5 17.1 16.5 598 DOWSE, NEUMAN, LANE, AND NEUMAN l:)eiDg approximatcK twice that in the phosphate liiiffer aerobiealh'. It is difficult to find an adequate explanation for effects of this kind in\()l\ing phosphate and bicarbonate; effects which are not confined to bone cells. Sexeral possibilities, however, come to mind. One invohes the intracellular level of inorganic phosphate, on the assumption that inorganic phosphate is one of the regulatorv factors of the ghx'olytic rate. Thus, in ascites tumor cells, glycolvsis can be markedh' increased b\ increasing the concentration of inorganic phosphate in the mediimi and therebx', indirecth , in the cell ( Racker and Wu, 1959 ) . However, though increase in phosphate ion concen- tration to levels of 40 niM led to some further stimulation of glv- colysis, it was not very great, and not of the order obtained with ascites preparations. Also, addition of bicarbonate suppressed the phosphate effect. Thus, in a medium containing both phosphate at 16-mM and l^icarbonate at 25-mM concentrations the Ql was almost identical with tlie quantitv found in tlie standard bicarbonate- buffered medium. We have been unable to find a difference between Qo.^ in bicar- bonate-buffered medium on the one hand and phosphate-buffered medium on the other using the COl- buffer technique suggested bv Pardee and developed in detail by Krebs ( Umbreit et ah, 1959 ) . If Warburg's "indirect" method is used, however, the Q0.2 appears to be increased some threefold in the presence of bicarbonate. We can suggest no explanation for the lack of agreement between the results obtained with the two methods, though a \'ariable CO2 retention bv the tissue, as postulated bv Borle (Borle et ah, 1960fl, 1960/?), is one possibility. Until this problem can be resolved, the possible effect of different oxidation rates upon glycolysis in the two media cannot be determined. Another explanation involves control of the intracellular pH. One would expect extracellular bicarbonate to exert a more efficient con- trol over the intracellular pH than phosphate, because of the differ- ence in permeability characteristics of the two ions. However, the crucial experiment has not been devised to test this hypothesis. Lactate production is only slightly affected b\' changes in the pH in phosphate medium between 7.3 and 7.8; below this pH, Qi. is decreased ( Fig. 2C ) . PARATHYROID HORMONE AND RAT ( ALVARIA METABOLISM 599 Substrate Utilization The Qo.2 remains stable for manv hours with no added substrate. From this, it is obvious that an adequate supplv of the intermediates and substrates exists for the system or systems responsible for oxygen uptake. At no time during this period will addition of TCA cycle intermediates (with the exception of succinate as discussed above) alter the Qo.^. If oxygen uptake is taking place in the TCA cycle, it follows that a continuous supplv of dicarboxvlic acids is available, either through CO2 fixation or by transamination reactions involving alpha ketoglutarate. The intracellular store of carbohvdrate which is being oxidized must be of considerable size. Glycogen is histolog- ically demonstrable in the tissue and, chemically, is present in ade- quate amount to maintain the observed Qo.^ for over 8 hours. The content is some 4.5 mg per gm wet weight, equivalent to 25 moles of glucose and requiring 150 moles of oxygen for complete oxidation. Based on a ()(.o of 18.4 moles/gm wet wt/hr., it would therefore be adequate for just over 8 hours if the lactate initially formed were later metabolized. The glycogen content was determined before and after a 2-hour incubation and the results are given in Table V, expressed as the quantitv of glvcogen hvdrolyzed, i.e. no longer precipitable with alcohol, with its calculated glucose equivalent. By expressing the lactic acid production and the Qo.j in the same manner, a striking TA15LK y. IvxDOGKNOus and I'>xogenou.s Substrate I'tilization (Phospliatc medium; 100 i)('r cent Oj; mean values of 6 to 12 flasks) /imoles/gm/2 hr. Glucose equivalent ()LimoIes/gm/2 hr.) Endogenous Glja-ogen liydrulysis Lactic acid production Qo. Exogenous'' Glucose uptake Lactic acid production 2.30" 13.S 3.3 13.0 26.1 12.8 ^:-ry 12.4 13.0 13.1 " Mg/gm wet weight. '' Glucose, 30 mg %. 600 DOWSE, NEUMAN, LANE, AND NEUMAN equivalence between the two is obtained. Though this is not con- ckisive proof, it is suggestive evidence that glycogen is indeed the principal, and possibly the only, endogenous substrate present. Attempts were made to deplete the endogenous glycogen to per- mit a studv of the utilization patterns of exogenous substrates. How- ever, aeration for a period of up to 4 hours, with medium changes to eliminate the lactate formed, did not accomplish this. Longer periods of aeration may well be adequate, but deterioration of the tissue under such circumstances raises questions concerning the validity of the data so obtained. Exogenous glucose uptake is also related to glycolysis ( Table V ) . Uptake was determined by difference after a 2-hour incubation, and the fact that an equivalent amount of lactic acid is produced in- dicates that exogenous glucose is metabolized entirely to lactic acid. lodoacetate, at a concentration of 2 X 10~^ m (sufficiently high to inhibit completely lactic acid production), reduces the Qon to zero. This would suggest, as a first approximation, that oxygen up- take is completely supported by the utilization of a carbohydrate substrate. Conclusions from data obtained by the use of iodoacetate must be interpreted with caution, however. In any event, no con- clusion may be drawn as to the sequences involved. lodoacetate may well restrict carbohydrate oxidation in the hexose monophosphate shunt shown to be present in bone (Cohn and Forscher, 1962fl, 1962b) if the cell is dependent for reoxidation of the reduced TPN formed in the first two stages upon coupling the reoxidation with either the DPN formed in the pyruvate-to-lactate step or the TPN formed in the malic enzvme reaction. Ability of the cell to utilize the TPN formed in synthetic mechanisms or transamination reac- tions would, of course, relieve it of dependence on later glycolytic intermediates as electron acceptors. Uncoupling Uncoupling of oxidative phosphorylation in a tissue can be accom- plished with 2,4-dinitrophenol (DNP) at a concentration of 2 X 10~^ M. With DNP, there was an increase in the Q0.2 of some 50 per cent and an increased accumulation of lactic acid under aerobic conditions ( Table VI ) . Both endogenous and exogenous lactate pro- PARATHYROID HORMONE AND RAT CALVARIA METABOLISM 601 TABLE VI. The Effects of an Uncoupler on Glycolysis (Glucose 17 niM; 2,4-DNP 2 X 10^'* m; bicarbonate medium) Lactic acid production (/tmoles/gm/hr.) Additions Aerobic Anaerobic None Dinitrophenol 14.2 39.1 37.1 36.1 duction were increased to that found under anaerobic conditions. There was a shght increase in the quantity of exogenous glucose taken up by the tissue, but not enough to account for the increased lactic acid production. It would seem that, in the presence of dini- trophenol, the endogenous supply of carbohydrate is more rapidly glycolyzed. These data also support the conclusion that the TCA cycle is functioning in this tissue and that it is, as in all tissues studied with this uncoupler, intimately linked with glycolysis as far as control is concerned. This point is also deducible from the fact that the Pasteur effect is present both endogenously and when ex- ternal substrate is added. Parathyroid Action The parathyroid hormone, a purified preparation kindly donated by Dr. Rasmussen, was added in vitro at a level of 5 units per ml, and the various parameters so far described were studied. Because of the increase in citrate production observed in vivo (Firschein et al., 1958 ) , it was sui-prising that no significant alteration in citric acid levels was observed, either in the medium or in the tissue. No effect was noted on glucose uptake or on the Qoo in phosphate me- dium. No significant changes in calcium or phosphate levels in the medium could be demonstrated. The only significant finding was that shown in Table VII, an in- crease in the accumulation of lactate in the system. This was found to occur in phosphate and in bicarbonate medium, at low and high levels of glucose, but only under aerobic conditions. The differences between the means, in phosphate medium, have p values of about 0.05. If comparisons are made of paired flasks containing the al- ternating halves of the same calvaria, the p values are less than 0.02. ()0'2 DOWSE, NEUMAN, LANE, AND NEUMAN TABLE VII. The Effects of Parathyroid Hormone ON Lactate Production Ql" Conditions (fimoh Bs/gm/hr.) Ghicose Iiicrca.'^c Medium Gas'' (mM) Control PTH (5 u/ml) {%) I'hosphate Air L6 12.7 ± 0.6 (4) 15.0 ± 0.4 (4) 18 Phosphate Air 17.0 18.9 zbO.o (4) 21.1 ± 0.3 (4) 12 J bicarbonate N2 17.0 35.0 ± 0.4 (8) 34.5 ± O.S (S) 0 1 bicarbonate 0, 0 5.2 ± 0.2 (4) 6.3 ± 0.3 (4) 21 P)icarbonate 0, 1.6 13.5 ±0.5 (4) 17.0 ± 0.5 (4) 25 liicarbonate ()•' 13.0 Ki.l ± 0.4 (4) 19.4 ±0.5 (4) 21 I bicarbonate 0, 17.0 15.2 ±0.3 (12) 18.9 ±0.4 (12) 24. " Means ± standard error; numbers of flasks in parentheses. * X2 and O2 contained 5 per cent CO2 for buffering. -■Significant at /) < 0.001. In bicarbonate medium, the findings are obviously significant. The absence or presence of ghicose and the level at which it is present seem to be of no detectable importance. On the hvpothesis that the hormonal effect could well be time dependent, longer incubation periods have been tried (Fig, 3A). A continuing effect was observed, not increasing but rather decreasing with time. Similarlv, with variation in concentration of the hormone, a dose response curve is obtained (Fig. 3B). The response can no longer be produced if the hormone is allowed to undergo inactiva- tion bv being removed from the deep freeze for a few weeks. Conclusion Calvarium in these studies behaved qualitatively in a fashion simi- lar to metaphyseal chips: low oxvgen consumption, excess reserve succinate dehydrogenation capacitv, high glycolytic capacity, and. PARATHYROID HORMONE AND RAT CALVARIA METABOLISM 603 150- 100- 4 6 HOURS Q, •/CONTROL 4 B I LOG DOSE Fig. 3. The effects of paiathxroid hormone on lactate production by calvaria. A, Qr„ /xmoles per gm wet weight over 8 hours in the presence (5 units per ml) and absence of parathyroid hormone. B, the relation between parathyroid hormone concentration and lactate production; Aq^ represents the difference, in /(.moles gm/hr., between paired control incubations and the indicated dosage of parathyroid hormone expressed as log units per ml + 1. Thus, the highest dose corresponded to 45 units per ml of Rasmussen polypep- tide. despite a Pasteur effect, a high rate of aerobic glycolysis (Laskin and Engel, 1956; Borle et al, 1960a; Cohn and Forscher, 1961). Though these studies represent the first reported (Neuman and Dowse, 1961) metabolic effects in bone resulting from purified para- thyroid hormone added in vitro, the effects were similar to those observed in metaphvses from animals given massive doses of hor- mone parenterally (Laskin and Engel, 1956; Borle et al, 1960a; Cohn and Forscher, 1961). In view of the rather profound physiological effects of parathy- roidectom\' in most species, the observed metabolic changes from excess hormone seem rather minor. In fact, the number of parameters which are unaffected is remarkable and suggestive in itself: no change in Qo.^, no change in glucose uptake, no change in citrate production, no change in lactate production anaerobically. One gains the impression that any hormonal effects on carbohy- 604 DOWSE, NEUMAN, LANE, AND NEUMAN drate metabolism are indirect and, in large measure, dependent upon experimental conditions. Thus, in intact dogs, citrate produc- tion by bone is increased (Firschein et al., 1958); in calvarium and metaphyseal chips no effects are seen in vitro sometimes (Borle et al, 1960a; Neunian and Dowse, 1961; Cohn and Forscher, 1961), though significant increases in labeled citrate (from labeled py- ruvate, Lekan et al, 1960) or labeled CO- (from C-6-labeled glu- cose via the TCA cycle, Cohn and Forscher, 1962Z?) are also reported. The most consistent finding is increased aerobic production of lactate, and this, too, is not a dramatic change, usually 20 to 30 per cent (Laskin and Engel, 1956; Borle et al, 1960a; Neuman and Dowse, 1961 ) . Such a small shift in glycolysis has a number of plausi- ble metabolic explanations, among which increased intracellular levels of inorganic phosphate ( Pi ) is perhaps most appealing ( Egawa and Neuman, 1963). Surely P, can be a controlling factor between competing glycolytic and oxidative mechanisms. Increased P, could increase aerobic lactate production and could also explain variable condition-dependent changes noted above in the operation of the TCA cycle. If we assume, for the moment, that parathyroid hormone does increase P,, this change is in itself likely to be the result of some more primary event such as a change in transport mechanisms of ions in and out of the cell. This kind of speculation led us to examine the movement of labeled inorganic phosphate in a variety of tissues, particularly bone, kidney, and intestine. Hope for some modest success rested in the classic phosphaturic activity ascribed to the hormone ( Albright and Reifen- stein, 1948 ) . Indeed, it has been possible to show significant altera- tions in the movement of labeled phosphate in all three tissues (Egawa and Neuman, 1963; Borle et al, 1963). Of particular inter- est, here, are studies on calvarium. Because calvarium contains a fair quantity of bone mineral which represents a large pool of ex- changeable phosphate, it was not possible to sample the cellular in- organic phosphate directly. Instead, the acid-soluble ester fraction was isolated as an indicator of intracellular label. As shown in Fig. 4, there was a significant increase in labeling of the ester-P fraction in response to the hormone. PARATHYROID HORMONE AND RAT CALVARIA METABOLISM 60.5 0.5 I 1.5 HOURS AFTER P' Fig. 4. The increased incorporation of labeled phosphate into the acid- soluble esters of calvarium after parathyroid hormone. Newborn rat pups were given 10 units Lilly Para-Thor-Mone each, or control injections, 30 minutes prior to labeled phosphate. The size of the circles indicates the standard error for each point. (Data taken from Egawa and Neuman, 1963.) Thus, it is our present opinion that explorations of carbohydrate metabohsm have not yet yielded important clues as to the primary sites of action of parathyroid hormone. Whether further exploration of phosphate transport will prove more profitable, only time will tell. Summary Calvaria from newborn rats were taken for biochemical study of the action of parathyroid hormone for several reasons: responsive- ness of the tissue, its low mineral-high cell content, its low marrow contamination, and its ease of preparation with little cell damage. 006 DOWSE, NEUMAN, LANE, AND NEUMAN Though control studies indicated the operation of the Krebs cvcle as the principal oxidatixe pathwa\-, the most striking metabolic properties of calvaria were: almost exclusive utilization of glucose (or gh'cogen), a high and medium-dependent ghcolytic rate, a high aerobic glycolysis (incomplete Pasteur mechanism), and a low but medium-independent oxidative rate. Purified parathyroid polypeptide added to the medium showed relatively little influence on metabolic activit\ . There were no sig- nificant changes in citrate production or utilization, glucose uptake, Qo.,, or levels of Ca++ or inorganic phosphate in the medium. The only significant finding was an increased production of lactate, and tliis only under aerobic conditions. The impression is gained that the action of the hormone on this tissue under conditions in vitro may be on some system other than glucose metabolism, and that the metabolic changes, though definite, are small and indirect in character. Arguments are presented, along with preliminary evidence, that the hormone may influence the transport of phosphate in and/or out of the cell. References Albright, F., and Reifenstein, E. C, Jr. 1948. The Parathyroid Glands and Metabolic Bone Disease. Williams and Wilkins Co., Baltimore, Md. Barker, S. B., and Summerson, W. H. 1941. The colorimetric determina- tion of lactic acid in biological material. /. Biol. Chem., 1-38, .535- 554. Beaulieu, M. M. 1953. X-rays and the citric acid of the bones. Bull. soc. chim. biol, 35, 491-500. Borle, A. B., Keutmann, H. T., and Neuman, W. F. 1963. The role of parathyroid hormone in phosphate transport across the rat duodenum. A7n. J. Physiol, 204, 705-709. Borle, A. B., Nichols, N., and Nichols, G., Jr. 1960fl. Metabolic studies of bone in vitro. I. Normal bone. /. Biol. Chem., 235, 1206-1210. Borle, A. B., Nichols, N., and Nichols, G., Jr. 1960/;. Metabolic studies of bone in vitro. II. The metabolic patterns of accretion and resorption. /. Biol. Chem., 2.35, 1211-1214. Chen, P. S., Jr., and Toribara, T. Y. 1953. Determination of calcium in biological material by flame photometry. Anal. Chem., 25, 1642-1644. Chen, P. S., Jr., Toribara, T. Y., and Warner, H. 1956. Microdetermina- tion of phosphorus. Anal. Chem., 28, 1756-1758. PARATHYROID HORMONE AND RAT CALVARIA METABOLISM 607 Cohn, D. v., and Forscher, B. K. 1961. The effect of parathyroid hormone on the oxidation of carbohydrate by bone. Biochim. et Biophys. Acta, 52, 596-598. Cohn, D. v., and Forscher, B. K. 1962a. Effect of parathyroid extract on the oxidation in vitro of gkicose and the production of ^^COo by bone and kidney. Biochim. et Biophys. Acta, 65, 20-26. Cohn, D. v., and Forscher, B. K. 1962Z?. Aerobic metabohsm of gkicose by bone. /. Biol. Chem., 237, 615-618. Egawa, J., and Neuman, W. F. 1963. The effects of parathyroid hormone on phosphate turnover in bone and kidney. Endocrinology, 72, 370- 376. Ettinger, R. H., Goldbaum, L. R., and Smith, L. H. 1952. A simphfied photometric method for the determination of citric acid in biological fluids. /. Biol. Chem., 199, 531-536, as modified by K. Lane and P. S. Chen, Jr., University of Rochester Atomic Energy Project Report 579. Firschein, H., Martin, G., Mulryan, B. J., Strates, B., and Neuman, W. F. 1958. Concerning the mechanism of action of parathyroid hormone. I. Ion gradients. /. Am. Chem. Soc, 80, 1619-1623. Gaillard, P. J. 1959. Parathyroid gland and bone in vitro. XI. Develop. Biol.,1, 152-181. Goldhaber, P. 1960. Behavior of bone in tissue culture. In Calcification in Biological Systems (R. F. Sognnaes, editor), pp. 349-372. Amer- ican Association for the Advancement of Science, Washington, D. C. Creep, R. O., and Talmage, R. V. (editors). 1961. The Parathyroids. Charles C. Thomas, Springfield, 111. Laskin, D. M., and Engel, M. B. 1956. Bone metabolism and bone re- sorption after parathvroid extract. A. M. A. Arch. Pathol, 62, 296- 302. Lekan, E. C, Laskin, D. M., and Engel, M. B. 1960. Effect of parathyroid extract on citrate metabolism in bone. Am. J. Physiol, 199, 856-858. Leslie, I., and Davidson, J. N. 1951. The chemical composition of the chick embrvonic cell. Biochim. et Biophys. Acta, 7, 413-428. Lilienthal, J. L.,' Jr., Zierler, K. L., Folk, B. P., Buka, R., and Riley, M. J. 1950. A reference base and svstem for analysis of muscle constituents. /. Biol. Chem., 182, 501-508.' Massey, V. 1953. Studies on fumarase. IV. Biochem. /., 55, 172-177. Neuman, W. F., and Dowse, C. M. 1961. The possible fundamental action of parathvroid hormone in bone. In The Parathyroids (R. O. Creep and R. V.Talmage, editors), pp. 310-326. Charles C. Thomas, Spring- field, 111. Neuman, W. F., and Neuman, M. \X. 1958. The Chemical Dynamics of Bone Mineral Universitv of Chicago Press, Chicago, 111. Racker, E., and Wu, R. 1959.'Clycolysis and the Pasteur effect. In Regula- 608 DOWSE, NEUMAN, LANE, AND NEUMAN Hon of Cell Metabolism (G. E. W. Wolstenholme and C. M. O'Con- nor, editors), pp. 205-218. Little, Brown and Co., Boston, Mass. Raisz, L, C, Au, W. Y. W., and Tepperman, J. 1961. Effect of changes in parathyroid activity on bone metabolism in vitro. Endocrinology, 68, 783-794. Saifer, A., and Gerstenfeld, S. 1958. The photometric microdetermination of blood glucose with glucose oxidase. /. Lab. and Clin. Med., 51, 448-460. Schartum, S., and Nichols, C, Jr. 1961. Calcium metabolism of bone in vitro: Influence of bone cellular metabolism and parathyroid hor- mone. /. Clin. Invest., 40, 2083-2091. Schartum, S., and Nichols, C, Jr. 1962. Concerning pH gradients be- tween the extracellular compartment and fluids bathing the bone mineral surface and their relation to calcium ion distribution. /. Clin. Invest., 41, 1163-1168. Spector, W. S. 1956. Handbook of Biological Data. W. B. Saunders Co., Philadelphia, Pa. Umbreit, W. W., Burris, R. H., and Stauffer, J. F. 1959. Manometric Tech- niques. Burgess Publishing Co., Minneapolis, Minn. Vaes, G. M., and Nichols, C, Jr. 1961. Metabolic studies of bone in vitro. III. /. Biol. Chem., 236, 3323-3329. Vaes, G. M., and Nichols, G., Jr. 1962. EflFects of a massive dose of para- thyroid extract on bone metabolic pathways. Endocrinology, 70, 546- 555. 24 Some Chemical Factors Influencing Bone Resorption in Tissue Culture PAUL GOLDHABER, Department of Oral Histopathology and Perio- dontology, Harvard School of Dental Medicine, Boston, Massachusetts DESPITE the pioneering work of Gaillard (1955fl, 1955/?, 1957, 1961«, 1961&), who demonstrated a direct effect of parathyroid tissue or extract on bone in tissue culture, others, inchiding ourselves, were not able to obtain similar evidence until most recently. In the course of our work, however, we did find a method of producing bone resorption routinely in tissue culture, without adding para- thyroid extract to our system, thereby permitting investigation of some of the morphological and biochemical changes related to bone resorption per se. Of particular interest to us was the development of a tissue culture system which would enable us to study the mecha- nism of action of potential stimulators or inhibitors of resorption. The purpose of the present paper is to describe the development of such a system and some of the significant chemical factors which have been found to exert a pronounced effect on the phenomenon of bone resorption. Observations Regulation of Bone Resorption by Oxygen Tension In our earlier experiments ( Goldhaber, 1958, 1960 ) it was found that bone resorption could be induced within young mouse calvaria 609 010 p. GOLDHABER grown in stationaiv Leighton tubes by introducing high concentra- tions of oxygen into the gas phase every 2 to 3 days at the time of refeeding the cultures. By using various oxygen concentrations in the gas mixture (from 0 to 100 per cent) it has been possible to demonstrate a direct correlation between oxygen concentration and extent of bone resorption (Goldhaber, 1961). Furthermore, on Figures 1 to 6 BONE RESORPTION IN TISSUE CULTURE 611 changing to roller tubes it was found that good bone resorption took place with hitherto ineffective concentrations of ox\gen, whereas high concentrations of ox\gen were toxic. Of interest was the finding that \ariation in extent of bone resorption (from slight to marked) could be obtained in roller tubes bv utilizing a relatively small spectrum of oxygen concentrations. From Figs. 1 to 6 it may be seen that after 7 davs of culture, whereas gassing with 10 per cent oxvgen resulted in onlv slight resorption in the median suture area of the frontal bone, 20 per cent oxygen resulted in almost com- plete resorption of this region. In both instances, however, the parie- tal bone was relatively unaffected. On the other hand, cultures gassed with 30 per cent O- rcNcaled widespread resorption in the median suture area, occasionalh' affecting the parietal bones as well. It seems clear, therefore, that in this culture svstem bone resorption not onlv is oxygen dependent but can be regulated bv alterations in the oxvgen concentration. Observations of Bone Resorption with Time-Lapse Microcinematographij Attempts to visualize the resorptive process with time-lapse micro- ciiiematographv at magnifications up to 660 times have revealed Fig. 1. Seven-day roller-tube culture gassed with 10 per cent oxygen (balance nitrogen). Note slight resorption (R) in sagittal suture of frontal bone area, (x 14.) Fig. 2. Seven-dav roller-tube culture gassed with 20 per cent oxygen (balance nitrogen). Note moderate amount of resorption (R) in sagittal suture area of frontal bone. (X 14.) Fig. 3. Seven-day roller-tube culture gassed with 80 per cent oxygen (balance nitrogen) . Note marked resorption (R) in sagittal suture area of frontal bone. (X 14.) Fig. 4. Parietal bone of same calvarium as Fig. 1. Culture gassed with 10 per cent oxvgen (balance nitrogen). Note absence of resorption. (X 14.) Fig. 5. Parietal bone of same calvarium as Fig. 2. Culture gassed with 20 per cent oxygen (balance nitrogen). Slight resorption (R) evident. (X 14.) Fig. 6. Parietal bone of same calvarium as Fig. 3. Culture gassed with 30 per cent oxygen (balance nitrogen). Note marked resorption (R). (X 14.) (Figures 1 to 6 were reproduced by permission from the author's paper in Ciucmicrogiaphy in Cell Biology, edited by George G. Rose and published by the Academic Press, Inc., New York, 1963.) 012 p. GOLDHABER FiGUHES 7 AND 8 BONE RESORPTION IN TISSUE CULTURE 613 that both macrophages and osteoclasts are capable of resorbing bone (Figs. 7 and 8), The formation of Howship's lacunae, however, is a task restricted to the osteoclasts. It is of interest that large numbers of macrophages appear relatively late in the culture period and that, to date, we have not observed anv fusion of these mononucleated cells to form large, multinucleated osteoclasts. Although we have observed osteoclastic activity as early as 1 day in culture, it is prob- able that these cells were present at the time of explantation and were stimulated by the experimental conditions rather than being formed de novo. With regard to the fate of the osteoclast, the longest period that we have been able to follow one cell has been approxi- mately 48 hours, during which time the cell appeared to enlarge and become more granular. In several instances we have observed the death of such granular osteoclasts, which was manifested by a sudden gelation of the cytoplasm and cessation of cell movement. Significantly, further bone resorption did not continue at these sites. The aggressive action of these cells against the resorbing bone sur- face supports the idea that osteoclasts actively participate in the process of bone resorption. The hypothesis that osteoclasts actively secrete some product in- volved in bone resorption is further supported by some of our time- lapse studies, wherein it is possible to visualize the production of large vacuoles at the border of the resorbing bone. Occasionally, in cultures exhibiting extremely rapid and marked bone resorption, the presence, growth, and fusion of huge bubblelike vacuoles has been noted in resorbing areas at the height of resorptive activity. Time- lapse films of such giant "bubbles" suggest that these unusual struc- tures are intracytoplasmic. Their sudden, rapid contraction several Fig. 7. Four-day living culture, showing osteoclast (O), outlined by arrows, in Howship's lacunae. BM, bone margin. (Enlargement of 16-mm frame, X approx. 540. ) Fig. 8. Same field, 6* hours later, showing resorption of some of the bone margin (BM) as the osteoclast (O) burrows deeper into the bone. (Enlarge- ment of 16-mm frame, X approx. 540.) (Figures 7 and 8 were reproduced by permission from the author's paper in Cinemicrographij in Cell Biology, edited by George G. Rose and published by the Academic Press, Inc., New York, 1963.) ()U p. GOLDIIABER hours after reaching maximum size is reminiscent of the contractile vacuole of unicellular organisms. The presence of large amounts of succinic dehydrogenase demon- strated histochemicalh' in osteoclasts has been reported independ- ently bv several laboratories simultaneousK' ( Scliajowicz and Ca- brini, 1960; Burstone, 1960; Goldhaber and Barrnett, 1960). The !•>.- J #«. /: 11 ,14 Figures 9 to 14 BONE RESORPTION IN TISSUE CULTURE 615 study liy Goldhaber and Barrnett (1960), however, was carried out on resorl)ing bone tissue cultures under time-lapse microcinematog- raphy. This procedure permitted visualization of active osteoclastic bone resorption prior to the histochemical staining of the very same osteoclasts (Figs. 9 to 14). In view of the intimate association of this enzyme with the mitochondria, the likelihood appears that osteoclasts have a high metabolic activit}', supporting the interpreta- tion of previous cinematographic evidence concerning the active role of osteoclasts in the process of bone resorption. Some Chemical Inhibitors of Resorption The high energy requirement of bone-resorbing cells is suggested by the finding that addition to the culture medium of 2,4-dinitro- phenol (5 X 10 "' m) inhibits resorption during the 2-week culture period, presumably by inhibiting svnthesis of adenosine triphosphate (Figs. 15 and 16). Under these conditions, bone formation was not totally inhibited (Figs. 17 and 18). Of further interest was the find- ing that addition to the medium of sodium malonate (5 X 10~'' m), a powerful inhibitor of succinic dehvdrogenase, inhibits bone resorp- FiG. 9. Two-dav resorbing culture showing indistinct osteoclast (O) against hollowed-out bone margin (BM). (Portion from enlargement of 16-mm frame, X approx. 400.) Fig. 10. Continuation from Fig. 9. Active resorption proceeding, as may be seen from new position of osteoclast (O) and the resorbing bone margin (BM). (X approx. 400.) Fig. 11. Continuation from Fig. 10. Resorption still progressing. Note new position of osteoclast ( O ) and adjacent bone margin ( BM ) . Total time interval between Fig. 9 and Fig. 11 was IGJ hours. ( X approx. 400.) Fig. 12. Continuation from Fig. 1 1 after reaction mixture for histochemical disclosure of succinic dehydrogenase was added to tissue culture medium. Note that outline of osteoclast (O) is becoming more distinct as a result of the initiation of the color reaction. BM, bone margin. ( X approx. 400.) Fig. 13. Continuation from Fig. 12. Development of color reaction pro- ceeding in osteoclast (O). No further bone resorption occurring owing to toxicity of the histochemical procedure. BM, bone margin. (X approx. 400.) Fig. 14. Continuation of histochemical reaction for succinic dehydrogenase. Note intensity of color reaction in osteoclast (O). Time interval between Fig. 12 and Fig. 14 was 8 hours. BM, bone margin. (X approx. 400.) (Figures 9 to 14 were reproduced by permission from the author's paper in Cinemicrography in Cell Biology, edited by George G. Rose and published by the Academic Press, Inc., New York, 1963.) 616 p. GOLDHABER Fig. 15. Six-day control calvaiium (roller tube at 20 per cent Oo). Note re- sorption (R) in median suture area. (X 14.) Fig. 16. Six-day culture containing approximately 5 X 10 '^•'' m 2,4-dinitro- phenol (roller tube at 20 per cent Oo). Note inhibition of resorption, although outgrowth is healthy. ( X 14.) Fig. 17. Histological section through control calvarium maintained in cul- ture for 2 weeks. Note thick layer of new osteoid (O) on surface of darker, original bone (OB). ( X 325. ) Fig. 18. Histological section through 2-week culture exposed continuously to 5 X 10~^ M 2,4-dinitrophenol. Note patch of defective osteoid (O) on surface of original bone (OB) . ( X 325.) tion (Figs. 19 and 20). Again, under these conditions, bone forma- tion was not completely inhibited (Figs. 21 and 22), suggesting that bone formation and resorption are independent processes hav- ing diflFerent metabolic pathways. The above findings support the concept that the Krebs cycle is intimately involved in the process of bone resorption. Along these lines, previous concomitant chemical analyses of the supernatant fluid of resorbing bone cultures have revealed a cumulative increase in citric acid over a 10-day culture BONE RESORPTION IN TISSUE CULTURE 617 21 OB-^ \ Fig. 19. Se\cii-da) control calvarium (statioiiai) tubes gassed witli 95 per cent O2 + 5 per cent COo). Note resorption (R) in median suture area. (X 14.) • Fig. 20. Seven-day culture containing 5 X 10-^ m sodium malonate (sta- tionary tubes gassed with 95 per cent Oo + 5 per cent CO2 ) . Note inhibition of resorption. (X 14.) Fig. 21. , Histological section through control calvarium maintained in cul- ture for 9 days. Note multinucleated osteoclast (OCL) at edge of original bone spicule (OB). (X 325.) Fig. 22. Histological section through 9-day culture exposed continuously to 5 X 10~2 M sodium malonate. Note layer of new osteoid (O) on surface of original bone (OB). (X 325.) period, amounting to 8 to 19 times the total citric acid content of the original calvaria ( Kenny et ah, 1959 ) . Effect of Parathyroid Extract Recently, we have found that it is possible to partially inhibit the resorption of bone in response to oxygen by eliminating the embryo extract component of the supernatant. Surprisingly, with this new culture medium (80 per cent horse serum and 20 per cent Gey's 018 p. GOLDHABER balanced salt solution containing penicillin and streptomycin) we have been able to demonstrate a dramatic, clear-cut effect of Lilly parathyroid extract in roller-tube cultures gassed with 50 per cent O2. This finding differed from our previous results in that resorption did not cease after approximately 5 davs, but continued until prac- tically all the bone was destroyed. The gross, massive bone resorp- tion resulting from the addition of 0.5 unit of parathyroid extract per ml of medium every 2 days for 15 davs mav be seen in Fig. 23. Fig. 23. Gross appearance of 15-day cultures. Both control and experimental roller tubes were gassed with 50 per cent Oo. Note almost complete resorption of calvaria in tubes containing 0.5 unit of Lilly parathyroid extract per ml of medium (80 per cent horse serum + 20 per cent Gey's balanced salt solution). It should be noted that both the control and experimental groups were gassed with 50 per cent Ol'. Evidence that this effect was due to the hormone per se rather than to some contaminant in the com- mercial parathyroid extract was obtained by the use of more highly purified parathxroid preparations" which gave similar results. " Kindly pro\ ided by Dr. Paul L. Muiison. BONE RESORPTION IN TISSUE CULTURE 619 In view of our earlier results with chick embr\ o extract-coutaiuiui^ uiedium, wherein it was possible to dciuoustrate regulation of extent of bone resorption by alteration of oxvgen tension, it was significant that the response to parathyroid extract could be altered bv variation in oxygen tension. From Figs. 24 and 25 it mav be seen that whereas increasing the oxygen tension from 10 per cent to 50 per cent had little effect on the gross appearance of 12-da\ control cultures of mouse cal\ aria, there was a distinct effect on the experimental cul- tures containing 0.5 unit of parathyroid extract per ml of medium. Although the hormone-containing group gassed with 10 per cent oxygen showed little diff^erence from its corresponding control group without hormone, the experimental groups gassed with 20 per cent, 30 per cent, and 50 per cent ()j showed graded enhancement of bone resorption in response to increasing oxxgen tensions. The interdependence of oxvgen tension and parath\roid extract in stimulating bone resorption in this svstem was demonstrated by the observation that despite optimum oxvgen tensions (50 per cent O2) in the gas phase, decreased bone resorption resulted from decreasing the concentration of parathvroid extract in the medium (Fig. 26). Although cultures gassed with 50 per cent O2 and exposed to 0.01 or 0.05 unit PTE per ml showed no gross difference from control tubes not receiving hormone, those treated with 0.1 or 0.5 unit PTE per ml (and gassed with 50 per cent Oi) showed enhanced bone resorption after 2 weeks in culture. It should be noted that distinct differences in extent of resorption for the different dose levels of parathvroid extract were determined readilv during the 1st week of culture bv microscopic examination and scoring of the living cul- tures. Another means of comparing the eff^ects of different doses of parathyroid hormone on bone resorption in the previous experiment was to perform calcium determinations (Munson et ah, 1955) on all supernatants throughout the experiment. From Fig. 27 it may be seen that the largest differences in calciimi level occurred in the media collected at the 4th, 6th, and 8th days of culture. This interval corresponded to the period of most rapid bone resorption. Although no statistical analysis was performed, it appears that during this rapid resorption period the calcium levels in the media of tubes containing 0.5, 0.1, and 0.05 unit of PTE per ml were significantly mo p. GOLDHABER Figures 24 and 25 BONE RESORPTION IN TISSUE CULTURE 621 different from one another and from the control tubes without hor- mone. In view of its sensitivity to small quantities of parathyroid hormone, the present tissue culture system may provide a simple assay method for the hormone, based upon scoring severity of re- sorption microscopically. This approach has been of practical value already in studies concerning the extraction of parath\'roid hormone from rat and human parathyroids." Citrate and Lactate Production in Resorhing and Nonresorbing Bone Cultures In a previous study (Kenny et al., 1959) it was shown that the media from our resorbing bone cultures showed consistent increases in citric acid, calcium, and phosphorus, which paralleled the bone resorption seen morphologically. Lactic acid, on the other hand, accumulated in the nonresorbing, control cultures. Whether the accumulation of citric acid in the media of the resorbing cultures and lactic acid in the media of the nonresorbing cultures was sig- nificant with respect to the mechanism of bone resorption was not known. This pattern of acid accumulation might have been coinci- dental, for in those studies bone resorption was initiated by gassing the experimental stationary cultures with 95 per cent O2 and 5 per cent CO2, whereas the nonresorbing stationary cultures were main- tained by gassing with 95 per cent air and 5 per cent CO2. It was to be expected, therefore, that the experimental cultures would favor aerobic metabolism and citrate production, whereas the control cul- tures would favor anaerobic metabolism and lactic acid formation. The striking ability of parathyroid extract to stimulate bone re- sorption in our tissue culture system ( both control and experimental * Studies in progress in collaboration with Drs. Paul L. Munson, Philip F. Hirsch, and Amien J. Tashjian. Fig. 24. Gross appearance of control roller tubes ( 12 days in culture) gassed with 10, 20, 30, or 50 per cent oxygen. It should be noted that some resorption was evident microscopicalhj in tlie tubes gassed with the higher concentrations of oxygen. Fig. 25. Gross appearance of experimental roller tubes (12 days in culture) containing 0.5 unit of parathyroid extract per ml of medium. Note the increasing effectiveness of the hormone on resorption as the oxygen tension is increased. ()'-2^2 p. (JOLDIIABER O.l u/ml ^mHb 0,5 u/ml 0.0 5 u/ml Fig. 26. Appearance of 2-\veek roller-tube cultures treated with various concentrations of parathyroid extract and gassed with 50 per cent Oo. Gross effects may be noted at the level of 0.1 u/ml and 0.5 u/ml. Microscopically, 0.05 u/ml of hormone caused some enhancement of resorption. BONE RESORPTION IN TISSUE CULTURE 623 EFFECT OF DIFFERENT CONCENTRATIONS OF PARATHYROID EXTRACT (LILLY) ON CALCIUM LEVEL IN TISSUE CULTURE MEDIA I6r 15 14- 13- 3 O < o 12- 10- 9- U/ML 0.5— ^^ ./\ '••^. (C 0.1 — .•'*' 0.0 5 -^y^ o.oi-\ X 0 00 - ^ ^ - ^>v^- X < o UJ Hydroxyproline- H found in media hydrolyzates of parathyroid extract-treated cultures vs. controls (DPM/ml whole media hydrolyzate) 500 1200 - 900 - PTE r0.5U/AfLJ+50% Or, ^ 600 - O X d 300 Q. O DAYS Fig. 28B. Graph of hydroxyproline-H'' (DPM/ml whole media hydroly- zate) found in media of parathyroid extract-treated cultures versus controls. Note that peak release of labeled collagen occurs in media collected on the 4th day, which corresponds roughly to the period of most rapid bone resorption, as visualized morphologically or as measured by determining the calcium level in the media. (Data reproduced from Stern et ah, 1963rt, 1963/?.) form, but was recoverable both as small peptides and as the free amino acid. Effect of Vitamin A It is well known that hypervitaminosis A can intensify bone re- sorption in vivo, leading to bone deformities and fractures in the rat (Barnicot and Datta, 1956). Although Barnicot's experiment con- cerning the intracerebral implantation of combined fragments of parietal bone and parathyroid gland tissue has been widely ac- BONE RESORPTION IN TISSUE CULTURE Hydroxyproline-H^ found in medio hydroiyzotes of parathyroid extrad-treoted cultures vs. controls (Cumulative DPM/ml wtiole media hydrolyzate) 3500 ]^ 3000 < N ^ 2500 O a: Q ^ 2000 (HI _J UJ > ^ Q UJ 500 LH 1000 o X 500 6 8 DAYS 10 14 Fig. 28C. Same data as shown in Fig. 28B, bnt graphed as cumulative counts. Note that at the end of 2 weeks the cumulative count of hvdroxvprohne- H'^ per ml of media hydrolyzate of parath\ roid extract-treated cultures totals more than 3 times the value derived from the control cultures. (Data repro- duced from Stern ct al, 1963fl, 1963/;.) X ^ claimed (McLean, 1956), his similar findings of intense bone resorp- tion with crystalline vitamin A and calciferol ( Barnicot, 1948 ) have not received much attention in this countr>'. Although his experi- ments stronglv suggested a direct effect of these agents on iDone, clear-cut proof of such a potential can be obtained only in tissue culture. Such evidence has been obtained for parathyroid tissue and extract, as discussed earlier in this paper. Similar evidence of a direct effect of n itamin A on bone in tissue culture has been ac- cumulated during the past decade bv Dr. Honor B. Fell and her co-w^orkers. Fell and Mellanbv ( 1952 ) found that addition of xitamin A acetate or alcohol to the medium resulted in a profound effect on 628 p. GOLDHABER fetal mouse bones grown in tissue culture. Both the cartilage matrix and the bone were destroyed, the effect being more pronounced the younger the limb-bone rudiments and the higher the concentration of vitamin A. Biochemical evidence was obtained in support of the hypothesis that vitamin A acts on cartilage matrix by enhancing the enzymic activity of chondrocytes and by increasing proteolytic ac- tivity through alteration of lysosome permeability (Dingle et ah, 1961; Lucy et ah, 1961; Dingle, 1961). Of interest was the finding that vitamin A in low concentrations released a protease from iso- lated lysosomes of rat liver cells (Fell et ah, 1961). This in vitro effect of vitamin A upon lysosomes has been demonstrated recently in an in vivo system by Janoff and McCluskey (1962). These in- vestigators have reported that peritoneal macrophages and poly- morphonuclear leucocytes of vitamin A-treated guinea pigs showed a marked loss of acid phosphatase, presumabh' due to disruption of the intracellular lysosomes which contain the enzvme. Although Fell and Mellanby ( 1952 ) reported the presence of only a few osteoclasts in most of their vitamin A experiments, it would be of interest to learn whether the acid phosphatase content of these cells is depleted. Attempts to demonstrate a direct effect of vitamin A on mouse calvaria in our own tissue culture system have been successful. In cultures gassed with 50 per cent 0-, the addition of 30 units of crystalline vitamin A alcohol per ml of medium produced bone re- sorption. The addition of 3 units of vitamin A per ml of medium had less effect, whereas 300 units of vitamin A per ml had the least effect (Fig. 29). It should be noted that our effective dose of approxi- mately 30 units of vitamin A per ml was similar to that found by Fell and Mellanby ( 1952 ) . Of further interest was the finding that there was an interdepend- ence between vitamin A and oxygen tension similar to that found for parathyroid extract. From Figs. 30 and 31 it may be seen that vitamin A-treated cultures (24 units per ml) showed no gross ef- fect as compared with untreated controls when both groups were gassed with 10 per cent O2. Definite enhancement of resorption was obtained with 20 per cent O2 and vitamin A. The most intensive resorption, comparable to the effect observed with 0.5 unit of para- BONE RESORPTION IN TISSUE CULTURE 629 Fig. 29. Gross appearance of 2-week roller-tube cultures exposed to 50 per cent Oo and various concentrations of crystalline vitamin A alcohol. Although bone resorption was obtained with 3 or 300 units of vitamin A per ml, the gi'eatest effect \\'as found with 30 units per ml. thyroid extract per ml of medium and 50 per cent O2, was obtained in those tubes containing vitamin A and 30 per cent Oi-. Surprisingly, the response to vitamin A and 50 per cent O2 was not so marked, suggesting that the optimal oxygen tension for bone resorption with vitamin A-containing cultures is at a lower level than that for para- thyroid extract-containing cultures. Attempts to demonstrate enhancement of bone resorption in our tissue culture system with /3-carotene were unsuccessful, confirming Barnicot's negative results with the vitamin A precursor (Barnicot, 1948). Effect of Vitamin D Although one of the major functions of vitamin D is control of calcium absorption from the gut, it is believed that the vitamin can also influence bone resorption (McLean and Urist, 1961). Not only 630 P. GOLDIIABER does this latter function resemble the action of parathyroid hormone, but some investigators have even postulated that parathvroid hor- mone acts by modifying the action of xitamin D (Neuman and Dowse, 1961). Figures 30 and 31 BONE RESORPTION IN TISSUE CULTURE 631 In view of the controversy which still exists concerning a direct action of vitamin D on the skeleton, it was deemed worth while to test the effect of this vitamin in our tissue culture system (50 per cent O2). Dihydrotachysterol, a member of the vitamin D group which is closely related to calciferol and has calcemic activity in vivo, caused a clear-cut enhancement of resorption when added to our system. The extent of bone resorption was directlv related to the concentration of dihydrotachysterol in the medium. The micro- scopic appearance of living cultures exposed to 0.001 mg, 0.01 mg, and 0.1 mg of dihydrotachysterol per ml of medium for 1 week may be seen in Figs. 32 to 35. Preliminarv experiments with crystalline vitamin Di- or Da in our system ( 50 per cent O- ) have demonstrated enhancement of bone resorption when the vitamins were added at the level of 0.1 mg (4000 units) per ml to the medium. The extent of resorption, which was similar to that found with 0.1 mg per ml of dihydrotachysterol, was not so marked as that obtained with 0.5 unit of parathyroid extract per ml of medium. Further experiments are continuing with the D vitamins to determine optimum dosages and the influence of oxygen tension on their ability to induce bone resorption. Discussion Several points about these findings seem worth emphasizing and discussing briefly. It should be stressed that variation in culturing techniques or media can make a vast difference in the response obtained to a given Fig. 30. Gross appearance of 12-day roller-tube cultures containing 24 units vitamin A per ml and exposed to 10 per cent and 20 per cent Oo. Note the enhanced resorption induced by the presence of vitamin A as compared with controls, as well as the influence of 20 per cent O2 as compared with 10 per cent Oo. Fig. 31. Gross appearance of 12-day roller-tube cultures containing 24 units vitamin A per ml and exposed to 30 per cent and 50 per cent Oo. Note that the best resorption in the experiment occurs with the combination of 24 units of vitamin A per ml and 30 per cent Oo. Although the vitamin A cultures gassed with 50 per cent Oo show less of an effect at this stage, it should be noted that this group demonstrated the best resorption microscopically during the first 6 days. 632 p. GOLDHABER Figs. 32 to 35. Microscopic appearance of 7-day roller-tube cultures gassed with 50 per cent Oo and treated with various concentrations of dihydrotachy- sterol. Note the increased frontal bone resorption (R) as compared with the control (Fig. 32) as the concentration of dihydrotachvsterol is raised from 0.001 mg/ml (Fig. 33), to 0.01 mg/ml (Fig. 34), to 0.1 mg/ml (Fig. 35). (All figures X 14.) test substance. The fact that elimination of embryo extract from our supernatant fluid results in decreased bone resorption in response to oxygen indicates that the "oxygen effect" with our embryo extract- containing medium depends on the simultaneous presence of some unknown bone-resorbing cofactor in the extract. Paradoxically, the elimination of embryo extract permits parathyroid extract, in the presence of oxygen, to stimulate bone resorption, thereby suggesting the presence of a parathyroid inhibitor in embryo extract. This latter possibility is analogous to the findings of Chen ( 1954 ) concerning the inhibitory effect of embryo extract on insulin in tissue culture. Of interest is the finding that the effect of parathyroid extract is directly dependent on the amount of oxygen in the system. This effect of oxygen may be due to a generalized metabolic priming of the target cells, making them more susceptible to the action of bone- BONE RESORPTION IN TISSUE CULTURE 633 resorbing factors — including parathyroid extract. Perhaps the en- hancing effect of oxygen is directly related to increased formation of ATP, since the cells involved in bone resorption are oliviously most active and must require large amounts of energ)' to carry out the complicated task of bone resorption. The inhibition of bone resorption in our system by the addition to the medium of 2,4-dini- trophenol, an uncoupler of oxidative phosphorylation, provides ex- perimental evidence favoring this hvpothesis. The influence of oxygen tension on vitamin A-induced resorption and probablv on vitamin D-induced resorption may be explained on a similar basis. Since the oxygen tensions delivered to the bone-resorbing cells in the calvaria were probably within physiological limits, it is possible that regulation of bone resorption in vivo might be accom- plished by variation of local oxygen tension through alterations in the local vasculature. Such a scheme would allow local control or modification ( enhancement or inhibition ) of the effects of a variety of systemically circulating bone resorption stimulators. The finding that several D vitamins ( dihydrotachysterol, crystal- line vitamin D2 or Da ) exhibit marked enhancement of bone resorp- tion in our tissue culture system provides additional evidence favoring the concept that vitamin D may have a similar direct effect on bone in vivo. Although enhanced bone resorption was demonstrated in our sys- tem by the addition of either parathyroid extract, crystalline vitamin A, dihydrotachysterol, or crystalline vitamin D2 or D3, the difference in results obtained with such closely related compounds as /3-caro- tene and vitamin A emphasizes the specificity of the stimulation. How these compounds alter the metabolism of bone cells must be the subject of continued experimentation. Summary 1. The extent of bone resorption can be regulated in roller-tube cultures containing chick embryo extract by variation of the oxygen tension in the gas phase. 2. Parathyroid extract, when introduced into the supernatant me- dium from which embryo extract had been eliminated, induced marked and progressive bone resorption which could be detected ().'U p. GOLDIIABER grossly. Control tubes showed minimal resorption despite the pres- ence of adequate oxvgen tensions. 3. An interrelation between oxygen tension and parathyroid ex- tract could be demonstrated. Alteration of either factor affected the extent of bone resorption. 4. Biochemical studies performed during the first 4 davs of culture on parathyroid-induced resorbing calvaria revealed increased citrate and lactate levels in the media as compared with nonresorbing con- trols. By the 6th to 8th day of culture, lactate levels (which corre- lated with glucose utilization) were higher in the nonresorbing control media. The citrate levels remained elevated in the resorbing culture media and paralleled the calcium and phosphate levels. 5. Biochemical analyses of the media of resorbing calvaria con- taining isotopically labeled collagen revealed significantly more labeled hydroxyproline in the media from parathxroid-treated sam- ples than from the controls. 6. Crystalline vitamin A alcohol ( approximatelv 30 units ml ) produced marked bone resorption in tissue culture. As is the case with parathyroid extract, the effect of vitamin A is dependent on the oxygen tension. 7. Both dihydrotachysterol (0.01 to 0.1 mg/ml) and crvstalline vitamin D- or Dh (0.1 mg/ml) significantly enhanced bone resorp- tion in roller-tube cultures gassed with 50 per cent 0-, a finding which supports the concept of a direct action on the skeleton of large doses of vitamin D. Acknowledgments. This work was supported by United States Public Health Service Research Grant D-1298. The author is a United States Public Health Service Research Career Development Awardee. The author is most grateful to Miss Gunta Cirulis and Miss Lidija Trencis for their technical assistance. References Barnicot, N. A. 1948. Local action of calciferol and xitamin A on bone. Nature, 162, 848-849. Barnicot, N. A., and Datta, S. P. 1956. \^itamin A and bone. In The Bio- chemistry and Physiologij of Bone (G. H. Bourne, editor), pp. 507- 538. Academic Press, Inc., New York. BONE RESORPTION IN TISSUE CULTURE 635 Burstone, M. S. 1960. Histochemical observations on enzymatic processes in bones and teeth. Ann. N. Y. Acad. Sci., 85, 431-444. Chen, M. J. 1954. The effect of insuhn on embryonic hmb-bones cultivated in vitro. }. Physiol, 125, 148-162. Dingle, J. T. 1961. Studies on the mode of action of excess of vitamin A. 3. Release of a bound protease by the action of vitamin A. Biochem. J., 79, 509-512. Dingle, J. T., Lucy, J. A., and Fell, H. B. 1961. Studies on the mode of action of excess of vitamin A. 1. Effect of excess of vitamin A on the metabolism and composition of embryonic chick-limb cartilage grown in organ culture. Bioclwm. J., 79, 497-500. Fell, H. B., Dingle, J. T., and Webb, M. 1961. Studies on the mode of action of excess of vitamin A. 4. The specificity of the effect on em- bryonic chick-limb cartilage in culture and on isolated rat-liver lyso- somes. Biochem. ]., 83, 63-69. Fell, H. B., and Mellanby, E. 1952. The effect of hypervitaminosis A on embrvonic limb-bones cultivated in vitro. J. Phijsiol (London), 116, 320-349. Gaillard, P. J. 1955(7. Parathyroid gland tissue and bone //) vitro. I. Exptl Cell Research, Siippl. .3,' 154-169. Gaillard, P. J. 19555. Parathyroid gland tissue and bone in vitro. II. Koninkl.'Ned. Akad. Wetenschap., Proc, Ser. C, 58, 279-293. Gaillard, P. J. 1957. Parathyroid gland and bone in vitro. Schiceiz. mcd. Wochschr., 87, Suppl. 14, 447-450. Gaillard, P. J. 196k/. The influence of parathyroid extract on the explanted radius of albino mouse embryos. II. Koninkl. Ned. Akad. Wetenschap., Proc, Ser. 0,64,119-128. Gaillard, P. J. 1961Z?. Parathyroid and bone in tissue culture. In The Parathi/roids (R. O. Greep and R. V. Talmage, editors), pp. 20-48. Charles C. Thomas, Springfield, 111. Goldhaber, P. 1958. The eflfect of hyperoxia on bone resorption in tissue culture. A. M. A. Arch. Pathol, 66, 635-641. Goldhaber, P. 1960. Behavior of bone in tissue culture. In Calcification in Biological Systems (R. F. Sognnaes, editor), pp. 349-372. American Association for the Advancement of Science, Washington, D. C. Goldhaber, P. 1961. Oxygen-dependent bone resorption in tissue culture. In The Parathyroids (R. O. Greep and R. V. Talmage, editors), pp. 243-254. Charles C. Thomas, Springfield, 111. Goldhaber, P., and Barrnett, R. 1960. Succinic dehydrogenase in osteo- clasts in resorbing-bone tissue cultures. /. Dental Research, 39, 728. JanoflF, A., and McCluskey, R. T. 1962. Effect of excess vitamin A on acid phosphatase content of guinea pig peritoneal leucocytes. Proc. Soc. Exptl Biol and Med., 110, 586-589. C36 P. GOLDHABER Kenny, A. D., Draskoczy, P. R., and Goldhaber, P. 1959. Citric acid pro- duction by resorbing bone in tissue culture. Am. J. Physiol., 197, 502-504. Lucy, J. A., Dingle, J. T., and Fell, H. B. 1961. Studies on the mode of action of excess of vitamin A. 2. A possible role of intracellular pro- teases in the degradation of cartilage matrix. Biochem. J., 79, 500- 508. McLean, F. C. 1956. The parathyroid glands and bone. In The Bio- chemistnj and Physiology of Bone (G. H. Bourne, editor), pp. 705- 727. Academic Press, Inc., New York. McLean, F. C, and Urist, M. R. 1961. Bone; An Introduction to the Physiology of Skeletal Tissue. University of Chicago Press, Chicago, 111. Mecca, C. E., Martin, G. R., and Goldhaber, P. 1963. Alteration of bone metabolism in tissue culture in response to parathyroid extract. Proa. Soc. Exptl Biol, and Med., Ill, 538-540. Munson, P. L., Iseri, O. A., Kenny, A. D., Cohn, V. H., and Sheps, M. C. 1955. A rapid and precise semimicro method for the analysis of cal- cium. /. Dental Research, 34, 714-715. Neuman, W. F., and Dowse, C. M. 1961. Possible fundamental action of parathyroid hormone in bone. In The Parathyroids (R. O. Creep and R. V. Talmage, editors), pp. 310-326. Charles C. Thomas, Spring- field, 111. Schajowicz, F., and Cabrini, R. L. 1960. Histochemical distribution of succinic dehydrogenase in bone and cartilage. Science, 131, 1043. Stern, B., Mechanic, G., Glimcher, M. J., and Goldhaber, P. 1963fl. Bone collagen degradation products from mouse calvaria resorbing in tissue culture. Intern. Assoc. Dental Research, Abstracts of 41st General Meeting, p. 79. Stern, B., Mechanic, G., GHmcher, M. J., and Goldhaber, P. 1963??. Re- sorption of bone collagen in tissue culture. Biochein. and Biophys. Research Commtm., 13, 137-143. 25 The Possible Role of Chelation in Decalcification of Biological Systems G. NEIL JENKINS, Dej)artinent of Physiology, Medical School, King's College, Newcastle upon Tyne, England C. DAWES, Department of Chemistry, Harvard School of Dental Medi- cine. Boston, jMassachusetts NEUMAN and Neuman ( 1958 ) have pointed out in their stimulating monograph the three most reasonable hypotheses to solve the prob- lem of how the calcified tissues may dissolve in the surrounding fluids which are in general believed to be already supersaturated with respect to bone mineral. The most discussed possibility is that the environment of the bone cells is maintained at a lower pH and is therefore locally unsaturated as a result of metabolic acids' leaving the cells. A second possibility is that either the cells or the bone matrix releases a chelating substance which can dissolve the bone salts even in an environment which, on an ionic basis, is saturated or supersaturated. Thirdly, if the cell were surrounded by a high concentration of certain ions which take part in surface exchange (e.g. magnesium, carl^onate, or citrate), these ions could increase the apparent solubility of the bone. The Nature of Chelation and the Chelating Agents Present in Tissues In addition to forming the familiar salts with metallic ions, certain acids are capable of forming complex ions such as ferricyanides or 637 (i88 G. N. JENKINS AND C. DAWES platiiiicliloiides. In some cases, owing to the existence of two or more groups in the acid molecule which from their nature and steric position can combine with a metal, the latter becomes part of a ring compound (Fig. 1). Such compounds are called "chelates," and the I Co '3 H Fig. 1. The structural formula of one of the possible forms of calcium lac- tate chelate. acids possessing the necessary groupings are called chelating or sequestering agents. For example, acetic acid combines with cal- cium to form the salt calcium acetate, but when an amino group is introduced into the molecule ( thus converting it into glycine ) , calcium can be attached both to the acid group (b\- ionic bonds) and to the amino group (bv coordinate covalent l)onds), thus form- ing the ring. Although the term "chelate" should be used onlv for this special tvpe of complex ion in which the metal is part of a ring, there has been some tendency to use it looselv for complex ions in general. Chelates are characterized by great stability, especially if the metal is attached to the molecule at several sites. Some acids can combine with metals in two wavs: as an ordinary salt, and as a chelate. The chelating properties of a substance are greath' in- fluenced by pH and tend to increase as pH rises, because the metal ions compete directly with hvdrogen ions. This is of great practical importance because manv chelating agents are capable of dissolving at neutral or alkaline pH values salts such as calcium phosphate previously regarded as virtually insoluble except in acids. Chelators may dissolve these salts even in solutions such as blood and saliva which, from the ionic point of view, are already satmated. Many naturally occurring substances have a structure which en- dhles them to form chelates, and this fact has suggested the possibil- ROLE OF CHELATION IN DECALCIFICATION SYSTEMS 639 ity that chelation plays an important part in biological processes. Mandl et al. (1952, 1953) and Mandl and Neuberg (1956) have listed the groups of such substances, which include the salts of nucleotides (e.g. adenosine triphosphate), polyphosphates (such as pyrophosphates), phosphorvlated sugars, amino acids, and the hy- droxy, keto, dicarboxylic, and tricarboxylic acids. Among the biochemical processes which it has been suggested mav occur bv chelation are the absorption of insoluble salts from the soil bv the root hairs of plants, the transport and deposition of metallic ions (e.g. calcium), and the control of enzvme action by chelation of inhiliitors or activators. In the present paper we are concerned with the possible role of chelation in dissolving the hard tissues in tissue fluids already saturated with calcium phosphate. The Phenomena to Be Explained Calcified tissues are removed during the physiological processes of bone growth and remodeling, and of resorption of the roots of deciduous teeth, and in meeting a calcium stress such as pregnancy and lactation, especially on a low calcium diet. Pathological proc- es.ses involving decalcification are dental caries, hyperparathyroid- ism, and infection of bones, including periapical abscesses. There are no reasons for believing that all these diverse phenomena can be accounted for bv one single process. Although several of the above processes are accompanied by the presence of osteoclasts — and it can no longer be doubted that osteo- clasts do take an active part in resorption — it is far from certain that these are the onlv cells concerned. Apart from the osteolysis described bv Belanger et al. in this symposium (chapter 20), several workers (Woods and Armstrong, 1956; Talmage and Elliott, 1958) have reported that in experimental hyperparathyroidism resorption occurs from bone deposited several months previously which is in- accessible to osteoclasts. Heller-Steinberg ( 1951 ) reported histolog- ical changes, presumably related to bone dissolution, in the matrix surrounding the lacunae containing osteocytes. A photomicrograph by Raisz et al. ( 1961 ) also showed that parathyroid hormone treat- ment produced resorption in the neighborhood of osteocytes and 640 G. N. JENKINS AND C. DAWES that osteoclasts were not present. The strong imphcation of all this work is that osteocytes can take part in resorption. The Unique Position of Citrate As Neuman and Neuman ( 1958 ) pointed out, citrate is a unique ion in that it can act in all the three ways in which they suggested bone could be dissolved in tissue fluid. It is an acid with three car- boxylic groups, it forms a stable chelate with calcium, and it can exchange with the phosphate of apatite to form a more soluble crystal. The idea that bone dissolves by the local formation of citrate is therefore very attractive. The evidence which has accumulated during the past decade or so favoring the view that citrate is im- portant in bone resorption has been so well reviewed elsewhere (Neuman and Neuman, 1958; Freeman, 1960) that it need not be repeated here. Neuman and Neuman also proposed a hypothesis relating the actions of vitamin D and parathyroid hormone to citrate metabolism. They suggested that vitamin D promotes citrate syn- thesis, perhaps by activating an enzyme cofactor, e.g. DPN (NAD), whereas parathyroid hormone might prevent oxidation of citrate by inhibiting TPN (NADP) -linked reactions. In the past two or three years, however, it has become clear from in vitro studies of bone metabolism that lactate and not citrate is the major end product of bone, a finding which has cast doubt on the citrate theory (Borle et ah, 1960a, lQ60b). Although lactate is an a-hydroxy acid and might be expected to have chelating proper- ties, as some workers have, in fact, reported (Johnston, 1956), re- cent opinion is that its chelating action on apatite crystals is negligi- ble (Leach and Dodge, 1961; Gray et ah, 1961). If lactate dissolves bone in vivo, then it does so as an acid rather than as a chelator. Citrate versus Lactate: Recent Evidence Although it is tempting to reject the citrate theory and with it reject the main support for the idea that chelation is important in hard tissue destruction, such a rejection is probably premature. It might be thought that one way of testing whether citrate or lactate is the bone solvent would be to find whether bone resorption ROLE OF CHELATION IN DECALCIFICATION SYSTEMS 641 is invariably accompanied by an increase in citrate or lactate pro- duction. This has been attempted by various groups of workers, but their results are contradictory. One practical reason for this may be the analytical difficulties of detecting changes in the very small absolute amounts of citrate which are produced. This approach may be unsound, however, because if both osteoblasts and osteoclasts are concerned in bone removal, they may act by different mecha- nisms, and there may not be one uniform metabolic product in- variably associated with bone removal. Even if citrate production is verv low in relation to lactate pro- duction, it is still possible to envisage mechanisms by which citrate might be of importance. We might consider the following possibili- ties : ( 1 ) that citrate production of bone may occur mainly in areas of resorption; (2) that small changes in citrate concentration may, because of its special properties, have quite large effects on bone solubility; (3) that changes in citrate production by bone cells may alter the citrate concentration and solubility properties of the bone crystal. Hypothesis 1. In most bone samples, osteoblasts greatly outnum- ber osteoclasts, so that if, as might be speculated, the former pro- duced lactate and the latter citrate, there might be very high local concentrations of citrate in the vicinity of, or within, osteoclasts in areas of resorption, but very little citrate in the surrounding medium. Few of the workers who have studied the metabolism of bone in vitro have reported on the relative numbers of osteoblasts and osteoclasts in the tissues they incubated, so that it is impossible to test this hypothesis rigorously with existing data. Many workers have, however, found an increased citrate produc- tion by bone after parathyroid treatment, which would be expected to increase the number of osteoclasts. It is also of interest that Raisz et al. (1961) showed that pretreatment with parathyroid extract (PTE) increased the amount of calcium dissolving in vitro from incubated bone without an increase in osteoclasts. This presumably osteoblastic resorption was not accompanied by any change in citrate formation, but lactate production did increase under anaerobic con- ditions. One piece of evidence against the citrate theory was pre- 04'2 (l. N. JENKINS AND C. DAW?:S sented by Vaes and Nichols ( 1961 ) . They found that the amount of calcium dissolving in vitro from PTE-treated bone was higher than from controls even in a medium containing 1.5 mM citrate, i.e. so high a level that the small addition from the cells would not be ex- pected to have any influence. The cells might, however, be produc- ing such high concentrations in their immediate vicinity that the citrate was still active. The probability that this occurred was in- creased somewhat bv the finding that the absolute amounts of bone dissolving from the control samples were no greater in the citrate buffer than in the other media, suggesting that, for some reason, the citrate buffer was not dissolving more than other buffers. Recent work on the fundamental action of parathyroid hormone ( PTH ) has emphasized the possibilitv that one of its effects may be on an individual enzyme svstem. Could not its action be a stage farther back than this, and is there not evidence that one of its fun- damental actions is to stimulate osteoclast formation? If these cells produce citrate, as is speculated here, then their formation would account for the increased citrate production by bone without in- volving reactions at the enzvme level (presumablv the formation of osteoclasts is more complicated than an interaction with one enzvme system). It is realized that this cannot be the only effect in bone, since, as already mentioned, PTH seems also to bring about resorp- tion by osteocvtes. Hypothesis 2. In view of the special powers of citrate in making bone soluble, it seems quite reasonable to suggest that although the citrate is only 1 to 2 per cent of the total acid produced by bone, it may have a much higher proportional effect on dissolving bone. This is a difficult point to test in vivo, but unpublished in vitro tests have shown that adding citrate to molar lactic acid buffer so that the citrate/lactate ratio is 1 100 may increase the amount of calcium phosphate dissolving by as much as 5 to 10 per cent. Anomalous results are obtained with some other buffers, however. If this hypothesis is substantiated, this could be a means bv which citrate plays an important part in controlling bone solubility even if citric and lactic acids are both formed bv the same cells. ROLE OF CHELATION IN DECALCIFICATION SYSTEMS 643 Hypothesis 3. There is good evidence that citrate can exchange with tlie phosphate in h\ droxyapatite and that this is associated with an increased sohibihtv of the crystals. Schartnni and Nichols ( 1961 ) concluded that much of the effect of pretreatment with PTE on the dissolving of bone in vitro could be accounted for by changes in passive solubility rather than by effects on cell metabolism, since it could be detected in bone inac- tivated by heat. The\' suggest that the increased citrate concentra- tion in bone found bv others after parathyroid treatment might account for this change in solubility. In a later paper Schartum and Nichols (1962) showed that if the solubilities of living bone and bone killed bv heating were compared in media over a range of pH values, the living bone dissolved more readily above pH 6.5, but below this pH both samples dissolved to the same extent. This could mean that living bone maintains the fluid bathing the crystal surface at a pH of about 6.5, so that if the medium was at that pH already the living bone had no means of increasing solubility; this would favor the idea that acid is the solubilizing agent. These results were in good agreement with the conclusion of Nordin ( 1957 ) that plasma is in equilibrium with bone if the pH on the surface of the crystal is about 6.6 to 6.8. Schartum and Nichols were reluctant to conclude that their results supported the acid theory, since, as they pointed out, at a low pH value some mechanisms normally at work, such as citrate production, might be impaired. Hartles (1961) has summarized the results from his own labora- tory on the factors influencing the concentration of citrate in bone. Briefly, he reports that diets deficient in vitamin D and calcium led to the formation of bone low in calcium and citrate, and that the animals developed tetany. With the same low calcium diet to which normal or high levels of vitamin D were added, tetany did not occur despite the fact that the concentration of calcium in the bone was not significantly altered, although the bone citrate was much higher ( Table I ) . Hartles and Leaver ( 1961 ) point out that the bone with the higher citrate appeared to be able to supply calcium to the blood (presumably in response to the increased parathyroid stimulus as a result of the low calcium diet ) and thus avoid tetany, whereas the 64-1 G. N.^JENKINS AND C. J)AWES TABLE I. Composition of Rat Femurs on Various Diets (From Hartles and Leaver, 1961) Ca P Citric acid (%) (%) (%) Control 24.3 12.1 0.51 Low calcium 19.7 9.9 0.76 No tetany Low vitamin D 23.5 11.6 0.46 Low calcium and low vitamin D 19.9 9.7 0.24 Tetany Conclusion: Calcium readily mobilized from bone with high citrate but not from bone with low citrate. bone with the low citrate was unable to supply sufficient calcium to avoid tetany. Studies on the extraction of citrate from bone show that some fractions of it are more readily dissolved out than others, and Hartles and Leaver tentatively suggest that some of it may be regarded as a "constant fraction" and some as a "variable fraction." This work suggests that one of the functions of vitamin D is to in- crease the citrate in the variable fraction and make the calcium available to parathyroid influence. How this happens \^ not known, but it might be speculated that the simultaneous presence in bone of the citrate and calcium facilitates the release of a calcium citrate complex. The suggestion is, then, that the citrate in the bone, rather than citrate newly formed by bone cells, is an important factor in calcium mobilization. Hartles and Leaver seem to have overlooked another possible ex- planation of their results, namely, that vitamin D prevented tetany by increasing the absorption of calcium from the low concentrations in the diet and digestive juices. Incidentally, Hartles and Leaver point out that their results with diets containing normal levels of vitamin D but low in calcium sup- ported Neuman and Neuman's hypothesis on vitamin D action in that the bone citrate was raised under conditions which would be expected to stimulate the parathyroid. In the groups without vitamin D, however, the hypothesis would suggest that with both calcium and vitamin D low in the diet (i.e. with parathyroid stimulated), the citrate should be higher than in a diet with normal calcium and low in vitamin D, The reverse was found. ROLE OF CHELATION IN DECALCIFICATION SYSTEMS 645 The Dissolving of Apatite within Osteoclasts Four groups of workers (Cameron and Robinson, 1958; Hancox and Boothroyd, 1961; Gonzales and Karnovsky, 1961; Scott and Pease, 1956 ) have now demonstrated, in electron micrographs, bone salt crystals within the osteoclast. The crystals enter by pinocytosis in the "ruffled border" and can be seen in the cytoplasmic vacuoles. There is at present no means of knowing whether all the crystals removed by osteoclasts enter the cell. The crystals presumably dis- solve in the cytoplasmic vacuoles and the dissolved products are shed into the tissue fluid. The question now becomes: What is the nature of the intracellular mechanism for dissolving the crystals and discharging the salts in a form in which they will remain soluble in the tissue fluids? The contents of cells are stated to be acid ( Cald- well, 1954); much of the intracellular calcium is likely to be bound to protein, and the phosphate is also largely organically bound. These conditions of acidity and low concentrations of calcium and phosphate ions would favor the dissolving of the crystals perhaps without the intervention of chelating sul^stances. However, since the osteoclasts are presumably taking in many apatite crystals, the capacity for binding the calcium and phosphate would become rapidly used up — unless the organization of the cell can prevent the contents of the vacuoles from having access to the protein and en- zymes of the cytoplasm. The probability is, then, that the cell needs chelating agents to dissolve the crystals as well as to discharge the calcium into the circulation in a form in which it will remain soluble. Possible Importance of Chelating Agents Released from the Matrix There remains the problem of how the crystals are dislodged from the bone matrix. The electron micrographs of Hancox and Boothroyd showed naked collagen libers, from which they concluded that the removal of the apatite crystals occurs first, and is followed by the disintegration of the matrix. An extracellular solvent, perhaps citrate or lactate, may dissolve some of the crystals and thus expose the matrix. Scott and Pease (1956) and Gonzales and Karnovsky (1961), on the other hand, saw no matrix in their micrographs and 046 G. X. JENKINS AND C. DAWES concluded that the coUagen was broken down first, leadmg to the release of the crystals. Hancox and Boothroyd, in this symposium (chapter 18), have shown that detaclied collagen enters the osteoclasts, where it is presumably broken down into products which might be important in chelating the apatite crystals. Greulich ( 1961 ) showed by ultrasoft microradiography that the actiye osteoclast is surrounded bv a band of organic matter 1.0 to 1.5 microns wide. The nature of this material is uncertain, but it could contain chelating agents which might be concerned in the dissolution of the crystals. Johnston et al. ( 1961 ) reported a large reduction in the hexosamine content of bone following PTE injections in rats, implying that breakdown of some constituents of the matrix was taking place. The products might play a part in chelating the calcium. Cretin (1951), in a paper giving few details and no photomicro- graphs, stated that in freehand sections of unfixed bone, indicators showed that the vicinity of osteoclasts was acid up to a distance of 20 to 30 microns from the cell. This work would be worth repeating, as it might help to decide which cells, if any, were producing acid even though it would not distinguish between citric, lactic, and other acids. The results of Dulce and Siegmund ( 1960 ) and Siegmund and Dulce (1960) are also of interest. They found that inhibition of carbonic anhydrase by Diamox caused a fall in the calcium of blood plasma in laying hens. The explanation they suggested was that inhibition of carbonic anhydrase by Diamox prevented acid secre- tion and thus bone resorption by osteoclasts. This suggests that carbonic acid might be important, which is support for the views of Forscher and Cohn in this symposium ( chapter 22 ) . The Dissolution of Infected Bone Very little is known about the means b\' which infected bone is remo\'ed. The presence of pus and inflammation will increase pres- sure locally, which is probably a stimulus to osteoclast formation (Hancox, 1956), and if so, the problem becomes the same as that of osteoclastic removal of bone elsewhere. Dubos ( 1955) summarizes ROLE OF CHELATION IN DEOALOIFICATION SYSTEMS ()47 the evidence that areas of inflammation are usually acid, although he states that when infection is present, acidity may not develop owing to damage to the acid-producing cells by the bacteria. The intense autolvtic activit\' in pus might be expected to release the acid contents of cells as well as manv substances with chelating properties. It seems possible that pus might have a direct solvent action on bone and not be dependent on osteoclasis. As few data could be found on the chemistry of pus, a few pre- liminary observations have been made. Five samples have been collected with pH values ranging between 6.1 and 6.7, and lactate estimations suggested that this was the main acid. It has not yet been possible to obtain pus from infected bone. These very limited data suggest tentatively that acidity brought about largely by lactate ( presumably released from autolyzing cells or formed from their contents ) is probably one factor in the removal of infected bone. It is realized that study of pH alone cannot decide the importance of acid. It is essential to know the concentrations of calcium and phosphate ions in order to decide whether, even under acid conditions, the environment of the bone is or is not saturated with bone salt. Chelation as a Mechanism for Demineralization hij Miiricid Boring Gastropods In this symposium (chapter 3) Carriker et al. showed that an alkaline buffer extract of the secretory gland associated with the boring mechanism of a muricid boring gastropod was capable of etching ovster shells. As this extract was alkaline, the most reasonable explanation of the etching is that a chelating agent is involved. Chelation and Proteolysis-Chelation Theories of Dental Caries Martin et al. (1954) pointed out that many substances known to occur in the mouth have chelating properties, and suggested that chelation under neutral or alkaline conditions might therefore be the cause of dental caries. After further consideration of possible sources of chelators, Schatz and Martin ( 1955, 1962 ) proposed that 048 G. N. JENKINS AND C. DAWES proteolytic bacteria might attack the proteins of enamel, releasing from them various breakdown products which would then form chelation compounds with the mineral matter of the enamel (the proteolysis-chelation theory). Then the organic and mineral phases of enamel would be destroyed more or less simultaneously. This approach is a direct challenge to the acid theory, which in its modern form is based on results which are interpreted as showing that tooth substance does not dissolve in saliva or plaque until acids have lowered the pH of the environment to a critical value, usuallv be- tween 5 and 6. As long ago as 1949, Eggers Lura suggested a nonacid decalcifica- tion theory of caries, and although he has modified his views he still believes that chelators arising in the plaque are more important than acids (Eggers Lura, 1949, 1957, 1961). Detailed critiques of these theories have recently been published (Bibby et al., 1958; Jenkins, 1961), and only the main points will be discussed here. Evidence Needed to Prove the Proteolysis-Chelation Theorij Before the proteolysis-chelation theory is accepted, experimental evidence must provide positive answers to the following questions: Do bacteria exist in the mouth which can attack the organic matter of enamel without previous decalcification? Are the products of this breakdown chelating agents? If so, are the amounts of chelators formed adequate to dissolve the inorganic matter to form a carious cavity? The existence of bacteria which can attack undecalcified ENAMEL. No study appears to have been made of the effect of bacteria on isolated enamel protein, presumably owing to the dif- ficulty of obtaining this material. On the assumption ( not supported by the analyses of Stack, 1955) that enamel protein is a keratin, Schatz et al. ( 1957, 1958 ) cultured oral bacteria in media containing keratin from carefullv washed hair, wool, and feathers as the only source of nitrogen. They reported that both human and animal saliva contained bacteria which grew on these media, and considered that the existence of oral keratinolvtic bacteria was thereby proved. This ROLE OF CHELATION IN DECALCIFICATION SYSTEMS 649 conclusion was criticized by Mandel and Ellison (1957) on the ground that the bacteria may not have been metabolizing the kera- tin, but various nitrogenous contaminants. The thorough washing to which these sources of keratin were subjected makes this unlikely, although Schatz et al. did not prove conclusively that all non-keratin nitrogenous substances had been removed. Whether we accept or reject this evidence for the presence of oral keratinolytic bacteria is probably unimportant because the majority of the enamel protein is not keratin, but mucoprotein and polypeptides (Stack, 1955). There is little doubt that some oral bacteria could attack these substances. The question whether enamel organic matter can be attacked without previous removal of the calcium salts is more controversial. Most of the evidence suggests that the calcium salts, both in enamel and in dentine, protect the protein from enzvme attack (Miller, 1890; Bibby, 1932; Evans and Prophet, 1950; Roth, 1957), but some previous workers have disagreed (Pincus, 1937; Hurst et al., 1953). Schatz et al. ( 1955 ) measured the oxygen uptake by cultures of oral bacteria in the presence of whole teeth, hair, rat and human dentine, and human enamel ( Fig. 2 ) . Respiration increased with every addi- tion, especially with the human teeth and rat dentine. With human enamel, the rise in oxygen uptake was extremely small and of doubt- ful significance, although in later publications larger eflFects from human enamel were reported (Fig. 3). These results were inter- preted as showing that the organic matter in intact enamel could be metabolized by oral bacteria. Evidence for the chelating action of the substances released FROM incubated HARD TISSUES. In another series of experiments by Schatz et al. (1955), bone containing Ca'*'^ was incubated with and without oral proteolytic bacteria and the rises in radioactivity of the media were compared. In the presence of the bacteria 77 per cent more Ca"*^ entered the medium (pH 8.4) than in the con- trol medium which remained at pH 7.0, from which it was concluded not only that the bacteria could attack the undecalcified bone, but also that the products of proteolysis dissolved the calcium by chela- tion under these alkaline conditions. 050 G. N. JENKINS AND C. DAWES 800 - Gelatin-grown 600 - o Q. O 400 - 200 - Hours Fig. 2. Respiration ot a strain of proteolytic bacteria isolated from a human mouth and later grown in the presence of autoclaved human tooth (1), un- treated human tooth (2), hair (3), rat molar dentine (4), rat incisor dentine (5), human dentine (6), human enamel (7), no addition (autorespiration) (8). (Courtesy of Dr. Albert Schatz and the New York State Dental Journal.) Experiments on somewhat similar lines were carried out b\ Hashimoto (1958, I960), who incubated cow and human teeth with bacteria and detected a rise in concentration of nonprotein nitrogen, calcium, magnesium, and phosphate in the medium. The parallel- ism between the rise in nonprotein nitrogen and in mineral matter was interpreted as evidence for proteohsis followed b\- chelation of the metallic ions by the products of proteolysis. Although these conclusions are quite reasonable, they are by no means the only possible explanation of the results, and in anv case their relevance to caries is speculative. In particular, the media in which these incubations were carried out were not saturated with calcium phosphate, unlike saliva at neutrality and probabh plaque at neutralit\' (Dawes and Jenkins, 1962). It is therefore very prob- able that at least some of the mineral matter was dissolved in the media without involving chelation, and exposed the organic matrix, which was then attacked by bacteria. These experiments did not, ROLE OF CHELATION IN DECALCIFKWTION SYSTEMS 180 O Q. CM O MINUTES 60 120 Fig. 3. Respiration of pioteol) tic bacteria isolated from a human mouth and grown in the presence of different amounts of human enamel. (Courtesv of Dr. Albert Schatz and the New York State Dental Journal.) therefore, prove tliat the organic matter of the hard tissues was attacked l^efore decalcification, nor that the mineral matter had been dissolved bv chelation. Also, in some of the experiments (es- pecially with ground enamel and dentine), some of the organic matter is quite freelv soluble (Stack, 1955; Evans and Prophet, 1950), and its release does not necessarilv prove proteolvtic action. There is no reason to doubt that many of these products of proteol- ysis are chelating agents. It is, however, impossible to decide how effective these are or what proportion of the mineral matter which dissolved did so bv chelation. Other workers whose results have been quoted in support of the proteolysis-chelation theory of caries are Saito (1957), Stiiben (1959), Wandelt (1959), and Eggers Lura (1949, 1957, 1961). In our view all their results are inconclusive as far as caries is concerned, either because their experiments were inadequately controlled (Wandelt and Eggers Lura) or because they did not reproduce conditions of saturation with calcium phos- phate which resembled those in saliva. 652 G. N. JENKINS AND C. DAWES A Criticism of Experiments on tlie ''Critieal })ir and Its Investigation Since saliva is saturated at neutialit) with calcium and phosphate, tooth sulDstance would be expected to be insoluble. If the pH falls steadily, it will eventually reach a value at which the calcium and phosphate concentrations are no longer able to saturate the saliva, and below this value tooth substance would be expected to dissolve. After the addition of chelating agents, however, apatite would dissolve even in saliva which, on an ionic basis, was previously saturated. In other words, the concept of a critical pH would be invalid. Experiments designed to test the h\"pothesis of a critical pH have consisted of measuring the solubilitv of tooth substance either in saliva at different pH values (Fosdick and Starke, 1939; Ericsson, 1949; Hills and Sullivan, 1958) or in buffers saturated with calcium phosphate oxer a pH range (Enright et ciL, 1932; Hills and Sullivan, 1958). In the saliva experiments, toluene or thymol was added to prevent bac^^erial activity from making the pH unstable, but it was not realized that this would also prevent the formation of chelators either from sali\'a constituents or from enamel. It seemed to us desirable, therefore, to find out whether mineral matter from enamel could dissolve in saliva incubated without car- bohydrates or toluene. Such a saliva would become alkaline, and the pH would be so much higher than the critical value that no apatite could dissolve unless chelating agents were formed. Over fortv ex- periments were carried out, and the results (Table II) showed that TAl^LE II. Calcium and Phosphate Concentrations (jug, ml) before and AFTER Incubating Salivary Sediment with Teeth for 48 Hours After incubation I )ifl'erence Before incubation (teeth - control) Control W ith teeth pH — 7.8 7.9 +0.1 Calcium — 90 88 -2 Inorganic phosphorus 170 187 206 + 19 Total phosphorus 280 280 280 0 ROLE OF CHELATION IN DECALCIFICATION SYSTEMS 653 although the concentration of dissolved calcium did not change during the incubation of teeth with saliva (or saliva sediment), there was a fairly consistent rise in the inorganic phosphate as com- pared with control media containing saliva but no teeth (Jenkins, 1959; Jenkins and Dawes, 1963). To determine whether this was a loss of phosphate from the tooth or a breakdown of organic phos- phates in saliva, total phosphate (i.e. organic plus inorganic) was estimated in many of the experiments, and in most cases it showed no change. It was concluded that the presence of the tooth surface favored the breakdown of organic phosphate in the same way as White and Hess ( 1956 ) had found that ashed bone appeared to act as though it contained phosphatase. The explanation of the few experiments in which total phosphate in solution did rise (again with no evidence of a rise in calcium) is not clear; but possibly adsorbed phosphate was released. How it entered a medium al- ready saturated with calcium phosphate and whv it should be re- leased in some experiments and not others is unexplained. These experiments provide no evidence for the formation of chelating agents either from salivary constituents or from enamel. It is possi- ble, of course, that chelators might have formed but in too low a concentration to dissolve detectable quantities of calcium. In any case, this negative evidence in no way disproves the proteolysis- chelation theory, but merelv makes it somewhat less probable. Quantitative Considerations An obvious difficultv of proteolysis-chelation as a mechanism of enamel caries arises from the low concentration of organic matter in enamel. Could 0.6 per cent of organic matter (excluding lipid), even if it were broken down as Schatz suggests and even if the whole of it had chelating properties, dissolve the 99.4 per cent of mineral matter? Bibbv et ah (1958) considered this to be one in- superable difficultv in the theor\'. It has been answered by Schatz et al. in several ways. First, they speculate that small areas of de- struction of enamel may open up pathways through which other chelators from plaque or food can enter and extend the ca\ity. Secondly, they have calculated from the amount of bone dissolved during 2 hours in their in vitro experiments that 11 mg of enamel 654 G. N. JENKINS AND C. DAWES could be dissolved by chelation in a year. They assumed that the rate of decalcification in a cavity in vivo was the same as in their experiments in vitro, allowing for the much lower protein content of enamel as compared with bone. Though fullv realizing the limi- tations of this calculation, Schatz et al. at least think that it shows that chelation could have effects of the right order of magnitude. However, the rate in vivo is verv unlikelv to be as rapid or as con- stant as that in vitro, and the whole basis of the calculation is there- fore unsound. In conclusion, the proteolvsis-chelation theor\' is original and stim- ulating, but most of the experiments so far carried out to answer the questions raised above are either negative or susceptible to alternative explanations, or do not resemble sufficientlv the condi- tions in carious cavities to be relevant. When this evidence is con- trasted with the mass of evidence on the acid theor\', which — though mcsth circumstantial and inconclusixe — so largeh' faxors the theor\', the probabilities seem against proteolvsis-chelation in the form so far discussed. Possible Chelators in Placjue The original chelation theorv of Schatz et al. ( 1955 ) , which did not confine the source of chelators to the organic matter of enamel, is more plausible than the later proteolvsis-chelation theorv. It can- not be denied that man\' substances in foods, saliva, and plaque possess chelating powers. The question that arises is whether the concentrations are sufficientlv high to decalcifv enamel, and in par- ticular whether their action is quantitativelv comparable with that of acids. Plaque is known to have a high calcium concentration (Allen and Moore, 1957; Dawes and Jenkins, 1962), and it is possi- ble that much of it is present in a chelate or complexed form. If so, the chelators mav be alreadv bound to so much calcium that they are unlikelv to combine with the much less soluble surface of the enamel. Eggers Lura (1960) has tested the efl^ects of certain chelating agents dissolved in ammonia buffer at pH 7.8 on the enamel surface. He stated that alkaline solutions of nucleic acid, PTE, salivary mucin, and amino acids produced signs of decalcification, but the experi- KCLE OF CHELATION IN DECALCIFICATION SYSTEMS 655 ments were inadequately controlled, and, since the media were not saturated with calcium phosphate, some solution of tooth substance would be expected in these mixtures bv ordinary ionic equilibria witliout in\'olving complex formation. Sucrose, ATP and other or- ganic phosphates, and sialic acid were also suggested by Eggers Lura as possible chelating agents. Chelation by sucrose is of particular interest because it provides a possible explanation, on the chelation theory, of the undoubted fact that the ingestion of much sucrose increases caries. Blackwell et al. (1961) have shown, however, that high unphysiological pH values are necessary for sucrose to attack the enamel surface. ATP and most other organic phosphates of plaque are present inside the bacterial cells and out of contact with the enamel surface. If these substances were released from auto- lyzing dead cells, much of the organic phosphate would be broken down by the enzymes of the plaque. These substances, although powerful chelators, would not therefore seem to be present under conditions in which they could have much eflFect on the enamel. The position of sialic acid will be discussed below. Tests on the Chelating Towers of Tlaqtie Extracts In an attempt to test the effecti\'eness of the chelators in plaque, we have made salivarv extracts of plaque pooled from over a dozen subjects and shaken them with a piece of bone containing Ca^'^ (Jenkins and Dawes, 1963). Control media contained a similar piece of bone shaken with saliva supernatant only. The radioactivity en- tering the two media — salixarv supernatant alone and saliva super- natant containing any soluble chelating agents of plaque— was identical and could be accounted for by ionic exchange, due allow- ance being made for variations in surface area of the bone samples bv prior incubation of both bone samples with salivary supernatant alone. These experiments were carried out four times in all, and lent no support to the suggestion that alkaline plaque contained effective concentrations of chelating agents. Admittedly, the plaque constituents were diluted sixfold in making the extracts, but this is probably compensated for by the great sensitivity of the radio- active method of detecting whether the bone dissolved. We should like to emphasize the importance of studying salivary extracts of ()56 G. N. JENKINS AND C. DAWES plaque. A very difFerent, and we l^elieve a misleading, result might have been obtained if aqueous extracts of plaque had been used. The crucial issue which we believe needs testing is whether chela- tors are active in a salivary medium saturated with calcium phos- phate. The Concentration of Certain Chelating Agents in Phique The concentrations in plaque of several substances which have been suggested as chelating agents have also been investigated. Citrate concentrations were estimated in plaque collected before and after a glucose rinse, and the rate of breakdown of citrate added to plaque was determined. The results (Table III) show TABLE III. Citrate and Lactate Concentrations in Dental Plaque No. of Mg/100 mg estimates wet plaque Citrate before glucose rinse 10 7.3 Citrate after glucose rinse 18 3.5 Rate of tireakdown of 10 /xg of added citrate At pH 6.S 30 8.5/min. At pH .5.0 12 4.9/min. Lactate (pH above 7.0) 60-1- 60-270 that the concentrations were extremely low; they are, in fact, near the limit of detection by the method. The concentration after the glucose rinse was significantly ( p < 0.05 ) lower than before, which is surprising, as it was expected that carbohydrate metabolism would have increased the citrate concentration. The rapid rate of breakdown of added citrate presumably explains why plaque citrate is at too low a concentration to act as a chelator. Lactate concentrations in over 60 plaques whose pH exceeded 7.0 ranged from 0.06 to 0.27 mg/100 mg wet plaque. These figures are in fair agreement with those of Moore et al. (1956). Leach and Dodge (1961) found that as high a concentration as 0.9 per cent lactate exerted no detectable chelating action, from which it may be concluded that the much lower concentration of lactate in neu- tral plaque is of no importance as a chelating agent. ROLE OF CHELATION IN DECALCIFICATION SYSTEMS 657 The sialic acid concentration of plaque was also found to be very low (averaging 3.7 /xg/lOO mg wet plaque), and, as with citrate, plaque rapidly metabolized added sialate. None of this evidence from the experiments on plaque extracts and from the concentration of individual substances reported to have chelating properties supports the hypothesis that chelation is important in caries. Conclusion The position may be summarized as follows: There is no evidence which clearly supports the hypothesis that decalcification in caries results from chelators in the plaque. The evidence for acids is very considerable, in spite of uncertainties about the pH on the inner side of the plaque or about the critical pH necessary for decalcification in that area. On the other hand, very few experiments have been carried out on the action of plaque chelators. Since it cannot be denied that some chelating agents are present in the plaque, it is possible that they may be active during the in- tervals between meals and during sleep, when the plaque is in the neutral or alkaline conditions which favor chelation. The experi- meiits reported here suggest that such activity must be weak and probably unimportant as compared with the action of acid. More experiments are needed on the quantitative effects and distribution within the plaque both of chelators and of acids before a final de- cision can be reached. There are no grounds for believing that chelators are alone re- sponsible for caries. There is no experimental evidence for the view that enamel protein is a source of chelators in caries. The low con- centration of organic matter in enamel is a weighty objection to the theory, and the answers given to this objection are purely specula- tive. Summary 1. The nature of chelation and its possible biological importance are described. 2. The unique solubilizing powers of citrate are pointed out, al- ().jS G. N. JENKINS AND C. DAWES though tlie low concentration of citrate as compared with lactate produced by bone in vitro tends to minimize its importance. 3. It is speculated that even small amounts of citrate might be important in bone if they were localized in sites of resorption such as those within or in the immediate vicinity of the osteoclasts. Evidence is reviewed that citrate may enter the bone crystal and increase its solubility. 4. The suggestions on the possible role of chelation in dental caries made by Schatz et at. are described and the evidence for them is critically reviewed. 5. It is concluded that, although there are gaps in the chain of evidence for both the acid and the chelation theories, most of the evidence supports acid as the means of decalcification in caries. References Allen, W., and Moore, B. W. 1957. Calcium content of plaque and saliva. Intern. Assoc. Dental Research, Ahstr. 35th General Meetino No. 108, p. 41. Bibby, B. G. 1932. The organic structure of dental enamel as a passive defense against caries. /. Dental Research, 12, 99-116. Bibby, B. G., Gustafson, G., and Davies, G. N. 1958. A critique of three theories of caries attack. Intern. Dental J., 8, 685-695. Blackwell, R. Q., McMillan, L., and Fosdick, L. S. 1961. Sucrose retarda- tion of acid etching in dental enamel: An electron microscopic study. /. Dental Research, 40, 16-22. Borle, A. B., Nichols, N., and Nichols, G., Jr. 1960fl. Metabolic studies of bone in vitro. I. Normal bone. /. Biol. Chem., 235, 1206-1210. Borle, A. B., Nichols, N., and Nichols, G., Jr. 19605. Metabolic studies of bone in vitro. II. The metabolic patterns of accretion and resorption. /. Biol. Chem., 235, 1211-1214. Caldwell, J. P. 1954. A glass microelectrode suitable for the measurement of the intracellular pH of large cells. /. Physiol., 124, 18. Cameron, D. A., and Robinson, R. A. 1958. The presence of crystals in the cytoplasm of large cells adjacent to sites of bone absorption. /. Bone and Joint Surg., 40A, 414-418. Cretin, A. 1951. Contribution histochemique a I'etude de la construction et de la destruction osseuse. Presse med., 59, 1240-1242. Dawes, C, and Jenkins, G. N. 1962. Some inorganic constituents of dental ROLE OP^ CHELATION IN DECALCIFICATION SYSTEMS 659 plaque and their relationship to earlv calcukis formation and caries. Arch. Oral Biol., 7, 161-172. Dubos, R. 1955. The microenvironment of inflammation or Metchnikoff revisited. Lancet, July, 1-5. Diilce, H.-J., and Siegmund, P. 1960. Zur Biochemie der Knochenaiiflos- iing. II, III. Hoppe-Seilcr's Z. physiol Chem., 320, 160-162, 163-167. Eggers Lura, H. 1949. Die Enzi/me des Speichels. Munich. Eggers Lura, H. 1957. Non-acid agents in the caries pathogenesis and their relation to the prophylaxis. IV. ORCA report. Odontol. Revij, 8, 323-328. Eggers Lura, H. 1960. Beziehungen den Speichelsialins zur Karies. Dent. zaJmdrztl. Z., 15, 963-966. Eggers Lura, H. 1961. The biochemical analogy between tooth resorption and caries. N. Y. State Dental }., 27, 75-79. Enright, J. J., Friesell, H. E., and Trescher, M. O. 1932. Studies of the cause and nature of dental caries. /. Dental Research, 12, 759-851. Ericsson, Y. 1949. Enamel-apatite solubility. Acta Odontol. Scand., 8, Siippl. 3. Evans, D. G., and Prophet, A. S. 1950. Disintegration of human dentine by bacterial enzymes. Lancet, 1, 290-293. Fosdick, L. S., and Starke, A. C. 1939. Solubility of tooth enamel in saliva at various pH levels. /. Dental Research, 18, 417-430. Freeman, D. J. 1960. Citrate metabolism with special reference to citrate and bone. In Bone as a Tissue (K. Rodahl, J. T. Nicholson, and E. M. . Brown, editors), pp. 314-329. McGraw-Hill Book Co., New York. Gonzales, F., and Karnovsky, M. J. 1961. Electron microscopy of osteo- clasts in healing fractures of rat bone. /. Biophijs. and Biochem. Cijtol, 9, 299-316. Gray, J. A., Francis, M. D., and Griebstein, W. J. 1961. Chemistry of enamel dissolution. In Chemistnj and Prevention of Dental Caries (R. F. Sognnaes, editor), pp. 164-169. Charles C. Thomas, Spring- field, 111. Greulich, R. 1961. Organic mass distribution in bone matrix undergoing osteoclastic resorption. Arch. Oral Biol., 3, 137-142. Hancox, N. M. 1956. The osteoclast. In The Biochemistry and Physiology of Bone (G. H. Bourne, editor), pp. 220-224. Academic Press, Inc., New York. Hancox, N. M., and Boothroyd, B. 1961. Motion picture and electron microscope studies on the embryonic avian osteoclast. J. Biophys. and Biochem. Cytol, 11, 651-661. ' Hartles, R. L. 1961. The role of citric acid in mineralized tissues. Brit. Dental J., 111,52-56. Hartles, R. L., and Leaver, A. G. 1961. Citrate in mineralized tissues. III. 660 G. N. JENKINS AND C. DAWES The effect of purified diets low in calcium and vitamin D on the citrate content of the rat femur. Arch. Oral Biol., 5, 38-44. Hashimoto, K. 1958. New theory of mechanism of dental caries. Japan. }. Storruitol, 7, 322. Hashimoto, K. 1960. Biochemical problems of dental caries. Seikatsueisei (Osaka), 4,188-197. Heller-Steinberg, M. 1951. Ground substance, bone salts and cellular ac- tivity in bone formation and destruction. Am. J. Anat., 89, 347-379. Hills, J., and Sullivan, H. R. 1958. Studies on acid decalcification of hu- man enamel. Australian Dental J., 3, 6-18, 101-106. Hurst, v., Nuckolls, J., Frisbie, H. E., and Marshall, M. S. 1953. In vitro studies on the initiation of enamel caries. Oral Surg., Oral Med. and Oral Pathol, 6, 976-994, 1236-1238, 1450-1454. Jenkins, G. N. 1959. Some observations relevant to the proteolysis-chela- tion theory of caries. /. Dental Research, 38, 1225 (abstract). Jenkins, G. N. 1961. A critique of the proteolvsis-chelation theorv of dental caries. Brit. Dental J., Ill, 311-330. Jenkins, G. N., and Dawes, G. 1963. Experiments on the chelating proper- ties of dental plaque and saliva. Brit. Dental J. ( in press ) . Johnston, G. C, Jr., Deiss, W. P., Jr., and Holmes, L. B. 1961. Effect of parathyroid extract on bone matrix hexosamine. Endocrinology, 68, 684-691. Johnston, H. W. 1956. Ghelation between calcium and organic ions. Neio Zealand J. Sci. and TechnoL, B, 37, 522-537. Leach, S. A., and Dodge, A. M. 1961. The effect of various anions on the solubility of enamel and dentine in acid. /. Dental Research, 40, 1284-1285 (abstract). Mandel, I. D., and Ellison, S. A. 1957. Some comments on the proteolysis- chelation concept of caries initiation. N. Y. State Dental /., 23, 171- 172. Mandl, I., Grauer, A., and Neuberg, G. 1952. Solubilization of insoluble matter in nature. I. The part played by the salts of adenosine triphos- phate. Biochim. et Biophys. Acta, 8, 654-663. Mandl, I., Grauer, A., and Neuberg, G. 1953. Solubilization of insoluble matter in nature. II. The part played by organic and inorganic acids occurring in nature. Biochim. et Biophys. Acta, 10, 540-569. Mandl, I., and Neuberg, G. 1956. Solubilization, migration, and utilization of insoluble matter in nature. Advan. Enzymol., 17, 135-158. Martin, J. J., Isenberg, H. G., Schatz, V., Trelawny, G. S., and Schatz, A. 1954. Ghelation or metal binding as a new approach to the problem of dental caries. Euclides (Madrid), 14, 311-317. Miller, W. D. 1890. Microorganisms of the Human Mouth. S. S. White D. M. Go., Philadelphia, Pa. ROLE OF CHELATION IN DECALCIFICATION SYSTEMS (561 Moore, B. W., Carter, W. J., Dunn, J. K., and Fosdick, L. S. 1956. The formation of lactic acid in dental plaques. /. Dental Research, 35, 778-785. Neuman, W. F., and Neuman, M. W. 1958. The Chemical Dynamics of Bone Mineral. University of Chicago Press, Chicago, 111. Nordin, B. E. C. 1957. The solubility of powdered bone. /. Biol. Chem., 227, 551-564. Pincus, P. 1937. Caries attack on enamel protein in an alkaline medium. Brit. Dental]., 63, ^ll-5U. Raisz, L. C, Au, W. Y. W., and Tepperman, J. 1961. Effect of changes in parathyroid activity on bone metabolism in vitro. Endocrinology, 68,783-794. Roth, G. 1957. Proteolytic organisms of the carious lesion. Oral Stirg., Oral Med. and Oral Pathol, 10, 1105-1117. Saito, T. 1957. Fundamental study on the tooth decalcification mechanism in dental caries. Part 1. On the chelating agents. /. Dental Soc. Nihon Univ. ( Tokyo), 31, 71-76. Schartum, S., and Nichols, C, Jr. 1961. Calcium metabolism of bone in vitro: Influence of bone cellular metabolism and parathyroid hor- mone. /. Clin. Invest., 40, 2083-2091. Schartum, S., and Nichols, C, Jr. 1962. Concerning pH gradients between the extracellular compartment and fluids bathing the bone mineral surface and their relation to calcium ion distribution. /. Clin. Invest., 41, 1163-1168. Schatz, A., Karlson, K. E., Adelson, L. M., and Schatz, V. 1957. The pro- teolysis-chelation theory. Applications to caries and erosion. Ann. Dentistry (N.Y.), 16, 37-49. Schatz, A., Karlson, K. E., and Martin, J. J. 1955. Destruction of tooth organic matter by oral keratinolytic microorganisms. N. Y. State De7italJ., 21, 438-446. Schatz, A., Karlson, K. E., Martin, J. J., Schatz, V., and Adelson, L. M. 1958. Some philosophical considerations on the proteolysis-chelation theory of dental caries. Proc. Penn. Acad. Sci., 32, 20-48. Schatz, A., and Martin, J. J. 1955. Speculations on lactobacilli and acid as possible anti-caries factors. N. Y. State Dental J., 21, 367-379. Schatz, A., and Martin, J. J. 1962. The proteolysis-chelation theory of dental caries. /. Am. Dental Assoc, 65, 368-374. Scott, B. L., and Pease, D. C. 1956. Electron microscopy of the epiphyseal apparatus. A7iat. Record, 126, 465-495. Siegmund, P., and Dulce, H.-J. 1960. Zur Biochemie der Knochenauflos- ung. I. Hoppe-Seilers Z. physiol. Chem., 320, 149-159. Stack, M. V. 1955. The chemical nature of the organic matrix of bone, dentin and enamel. Ann. N. Y. Acad. Sci., 60, 585-595. 66'2 G. X. JENKINS AND C. DAWES Stiiben, J. 1959. Uber die Ca-bindende Affinitiit verschiedener Metall- komplexbildner, eimittelt an Schmelzproben xor und nach Behand- lung niit Fhiornatriun-i. Dcuf. ZahnarztcJ)!., 13, 251-254. Talmage, R. \., and Elliott, |. R. 1958. Effect of time of injection on re- moval of Ca'"' and Sr^"' bv peritoneal la\age. Proc. Soc. Exptl. Biol. and Med., 99, 306-309. Vaes, G. M., and Nichols, G., Jr. 1961. Metabolic studies of bone in vitro. III. Citric acid metabolism and bone mineral solubility: Effects of parathyroid hormone and estradiol. /. Biol. Chem., 236, 3323-3329. Wandelt, S. 1959. Uber die Wirkung von Chelation auf die Zahnhart- substanz. Deiif. zahndrztl. Z., 14, 1255-1270. White, A. A., and Hess, W. C. 1956. Phosphatase, peroxidase and oxidase activity of dentin and bone. /. Dental Research, 35, 276-285. Woods, K. R., and Armstrong, W. D. 1956. Action of parathyroid extracts on stable bone mineral using radiocalcium as tracer. Proc. Soc. Exptl. Biol, and Med., 91, 255-288. 26 Animal Gollagenase and Collagen Metabolism CHARLES M. LAPIERE* and JEROME GROSS, Department of Medicine, Harvard Medical School at the Massachusetts General Hos- pital, Boston, Massachusetts TISSUE remodeling during growth and development requires that synthesis and removal of structural elements be precisely synchro- nized in time and space. Cell migration, aggregation, differential multiplication, and selectively timed and placed cell death are major features of the process. The resultant complexities of the com- plete tissues have made it difficult to dissect and analyze the relevant mechanisms involved in these processes. Because of the ubiquitous distribution of the extracellular struc- tural element collagen, because of the singular and specific char- acteristics of this protein, and because of the large body of knowl- edge concerning its structure, chemistry, and metabolism, we have chosen to follow its fate in a controllable remodeling system, the metamorphosing tadpole. Metamorphosis can be initiated by thy- roid compoimds, producing rapid resorption of large collagen-con- taining structures sucli as tail fin and gill (Gudernatsch, 1912; Allen, 1929; Frieden, 1961). Simultaneously, there is hypertrophy of other collagenous tissues such as the skin of the back. Local intricate changes are induced in the shape of specialized structures such as the mouth parts, opercular areas, limbs, eye positions, etc. * Permanent address: Institut de Medecine, Hopital de Baviere, Liege, Belgimn. 663 GG4 r. M, LAPIERE AND J. GROSS The rapid removal of collagen which occurs during bone growth and during retrogression of the postpartum uterus has stimulated an intensive search for an animal collagenase operating under phvs- iologic conditions (Morrione and Seifter, 1962; Woessner, 1962; Rouiller, 1961). The many examinations of tissue extracts for col- lagenolytic activity operating at physiologic pH and temperature have uniformly been unsuccessful ( Mandl, 1961 ) . By culturing tis- sues on reconstituted collagen gels we have recently demonstrated the appearance of a collagenolytic enzyme in amphibian connective tissues (Gross and Lapiere, 1962). Similarly, we have also detected collagenolytic activity in rat uterine tissue and in bone. These studies have been extended and we wish to discuss the results of simultaneous experiments on the metabolic turnover of collagen and collagenolytic activity in resting and metamorphosing bullfrog tadpoles. We have compared the collagen of the tail fin with that of back skin since in the former complete removal occurs during metamor- phosis, whereas in the latter there is increased production. Gross Changes in Tail Fin and Back Skin Connective Tissue in Metamorphosis The dramatic remodeling of the frog tadpole in metamorphosis has been amply described (Taylor and Kollros, 1946; Lynn and Wachowski, 1951; Etkin, 1955; Kollros, 1961). There have been several recent examinations of the detailed structural changes in the body skin during metamorphosis (Kemp, 1959, 1961). Thus far, however, there has been little quantitative chemical information available on structural element modifications. We have examined the alterations in the various collagen fractions, free proline, and water and cellular content of both tail and back skin in order to obtain information needed for studies on collagen turnover and collagenolytic activity. Rana catesbiana tadpoles used in these stud- ies were all at the same stage of development, either legless or with hind legs less than 1 cm in length. Metamorphosis was induced by adding thyroxin in concentrations of 10"^ M to the aquarium water. Control animals were starved, ANIMAL COLLAGENASE AND COLLAGEN METABOLISM 665 since it was observed that thyroxin-treated tadpoles failed to feed. Figure 1 illustrates the marked alterations induced in Rana cates- biana exposed to thyroxin for 11 da vs. The rate at which the tail was resorbed is charted in Fig. 2. Fig. 1. Resting tadpole Rana caicshiana; abo\e, tadpole after 9 davs' treat- ment with thyroxin. The most dramatic change in composition in the tissues of meta- morphosing animals is a consideral^le loss of water, about 50 per cent in the fin and 25 per cent in the back skin (Table I) after 6 days' exposure to thyroxin. Collagen concentration, as well as free proline and DNA (representing cell content), was proportionately increased. Noncollagenous protein, on the other hand, showed relatively little increase in concentration. Composition changes of the whole tail fin, however, conformed closelv to the visible observa- tion of tissue resorption. Total tail fin weight diminished by three- fourths, with collagen and DNA falling approximately 50 per cent. Free proline dropped to two-thirds of the control value. The non- collagenous protein fraction* fell even lower, decreasing to one- quarter of the original level. The loss in tissue water during metamorphosis is well known; the literature was recentlv reviewed by Frieden ( 1961 ) . * Represented by the protein of the extract which does not precipitate with col- lagen on dialysis against water. 6G() C. M. LAPIERE AND J. GROSS to --0 o--. Starved Controls Thyroxin ^ Treated 50 DAYS Thyroxin Fig. 2. Decrease in tail length of tadpoles treated with thyroxin as a func- tion of time of exposure to the hormone. Controls were starved, since thyroxin- treated animals ceased feeding. Comparing tail fin and back skin in the resting state, collagen concentration is fourfold higher in the latter, with free proline and noncollagen protein concentration about twice as high. There is some resorption of collagen from the back skin, associated with the change in shape and organization which occurs in this area during metamorphosis (Kemp, 1961). ANIMAL COLLAGENASE AND COLLAGEN METABOLISM 667 TABLE I. Composition of Tail Fin and Back Skin in Thyroxin-Treated AND NONTREATED TaDPOLES" Tail fin Back skin Control Treated Control Treated No. fins and back skins 40 40 40 40 Wet weight (mg) per organ 473 107 354 256 Per cent dry weight 5.7 10.0 7.7 10.5 Mg/gin wet tissue Mg/gm wet tissue Collagen 10.5 21.7 45.8 56.5 Free prohne 0.07 0.16 0.14 0.16 Proline (noncollagen protein) ^ 0.20 0.21 0.37 0.44 DNA 1.0 2.7 0.63 0.77 Mg per fin Mg per back skin Collagen 4.9 2.3 16.2 14.5 Free proline 0.03 0.02 0.05 0.04 Proline (noncollagen protein ) - 0.09 0.02 0.13 0.11 DNA 0.7 0.3 0.22 0.20 " Six days with 6 X 10~" m thyroxin in the water. * Fraction of extracted noncollagenous protein not precipitated by dialysis against water. ■ Metabolism of Collagen in Resting and Metamorphosing Tadpoles The numerous studies conducted on turnover of the various col- lagen fractions in mammalian tissues have provided a working theory of collagen metabolism (Harkness et ah, 1954; Jackson and Bentley, 1959; Piez and Likins, 1957; Hausmann and Neuman, 1961; Prockop et al., 1962; Gould and Manner, 1962; Green and Lowther, 1959; Gross, 1959). The tropocollagen molecule (Gross et al, 1954; Schmitt et al., 1955), synthesized within the fibroblast, is secreted into the intercellular space and organized into fibrils. That fraction which can be extracted in cold physiologic saline is considered to be most recentlv synthesized, probabh' existing in the tissue as loose molecular aggregates in fibrillar form. Probabh' the newlv secreted collagen remains in the dispersed state for a very brief period in most situations, since little collagen is extr actable at body tempera- 668 C. M. LAPIERE AND J. GROSS ture. Successive extractions with solvents of increasing ionic strength remove older collagen, which is more highly organized into fibrils. Acid-extractable collagen is tho fibrils. The remaining collagen ca ght to be derived from still older be removed onlv bv denaturation Fig. 3. Schematic representation of a hypothesis formulated to help explain occurrence of the different extractable collagen fractions. The rodlike units repre- sent tropocollagen particles. Cold physiological saline (0.14 m NaCl) extracts the most recently formed collagen molecules (and perhaps also those resulting from physiological degradation ) , which are completely dissociated or in the loosest association; hypertonic salt solutions extract the same material plus older collagen in a more ordered state of aggregation; acid citrate buffer extracts all the above plus some of the older collagen in the typical fibrillar form. The in- soluble fibrils were of a sufficient age so that the degree of cross linking has prevented solubilization. (From Gross, 1959.) procedures and is most likely in a very stable, cross-linked form. This hvpothesized scheme is represented diagrammaticallv in Fig. 3. It should be noted that these collagen fractions are operational, and not necessarily phvsiologically or metabolicallv homogeneous. For example, it has been suggested bv Jackson ( 1957 ) that a portion of ANIMAL COLLAGENASE AND COLLAGEN METABOLISM 661) the cold saline-extractable collagen might originate from degrada- tion of nonextractable, previously deposited fibrils. Examination of the collagen fractions reveals marked differences in distribution between back skin and tail fin, shov^n in Table II. TABLE II. Collagen (Hydroxyproline) Fractions in Tissues OF Thyroxin-Treated and Nontreated Tadpoles" Av. total hydroxyproline (/imoles per fin or back skin) Acid Wet weight Neutral extract per organ extract (0.1 M acetic (gm) (1 M NaCl) acid) Insoluble Total Control tail fin 0.47 0.73 1.00 0.59 2.32 (31.5%) (42.8%) (25.7%) (100%) Thyroxin tail fin 0.11 0.46 0.43 0.19 1.08 (42.8%) (39.8%) (17.4%) (100%) Control back skin 0.35 0.29 2.52 5.53 8.34 (3.5%) (30.2%) (06.3%) (100%) Thyroxin ])ack skin 0.26 0.36 3.02 3.38 6.77 (5.4%) (44.6%) (50%) (100%) • " Average values for 140 whole fins and 140 back skins pooled and analyzed in 7 groups. In the former, cold neutral-extractable collagen accounts for only 3 per cent of the total, with the insoluble fibers accounting for close to 70 per cent. In the tail fin there is a roughly equal distribu- tion between the cold neutral-extractable, the acid-extractable, and the insoluble fractions, the last accounting for onlv 26 per cent of the total. There is a significant reduction in the proportion of in- soluble collagen in the tail fin of thyroxin-treated animals, 17 per cent as compared with 26 per cent in the controls. There is a coin- cident rise in neutral-extractable collagen from 32 per cent to 43 per cent. In the back skin, there is a proportionally smaller loss of in- soluble collagen associated with a significant increase in the acid- extractable fraction. A single dose of tritiated proline, 0.14 /xc per gm bodv weight, was injected intraperitoneally into each tadpole. The thyroxin- treated animals received the label at the same time as the controls. 670 C. M. LAPIERE AND J. GROSS after 6 da) s of continuous exposure to the liormone. Groups of ani- mals were sacrificed at chosen inter\als up to 30 hours; the tail fins and back skins were removed, minced, and extracted three times with 5 volumes of cold 1 m NaCl, followed bv three extractions with cold 0.1 M acetic acid. The extracts were dialvzed against large volumes of water, quantitatively precipitating the collagen with a small amount of globulin, leaving other noncoUagenous pro- tein in tlie supernatant fluid. The dialyzate contained most of the free proline and was dried by lyophilization. The material inside the bag was separated by centrifugation and the precipitates and supernatants were hydrolyzed. Hydroxyproline and proline were isolated from paper chromatograms developed with a phenol-water- ethanol solvent (Tanzer and Gross, in press). Hydroxyproline and proline are widely separated on such chromatograms and may be obtained from the paper in isotopically pure form. An aliquot part of the eluted imino acids was chemically measured and another portion counted in a liquid scintillation counter.* It is important to note that although the labeled proline was always given in a single injection, these are not truly "pulse" ex- periments, since, as seen from Fig. 4, the specific activity of free proline levels off rapidly at a fairly high level, approximately 10,000 cpm/i"^niole even at 30 hours post injection. Metabolism of Total Collagen Let us first examine the incorporation of proline-H'^ into the whole collagen pool (measured as hydroxyproline) of back skin and tail fin. The data represent the sum of all three fractions. Because of the diminishing tail size and change in water content, net incorporation of label could not be satisfactorily related to a per gram wet weight basis. As the fins and back skins were removed in a standard manner, reproducibly obtaining most of the tissue, the above measurements could best be expressed on the basis of content per "organ." Over the 30-hour period of measurement there was only an 11 per cent loss of collagen from the tail fin and 9 per cent from the back skin. The net loss of collagen from back skin may only be ap- * All these procedures will be described in detail in the definitive publications. ANIMAL COLLAGENASE AND COLLAGEN METABOLISM 671 I05 ^ 10^- ^ Q. ^ 10- . \ - \ CONTROL - ' 0 0 Tail • ■* Body . 1 \ - A\ THYROXIN - 0 o Tail \ \ ^ — ^ \ ^ \ \ • — •- (V. ^ _^ "" ~" -o 1 1 1 1 1 1 10 20 HOURS AFTER PROLINE-H 25 30 Fig. 4. Specific activit}- of the free proline in 1 m NaCl extracts of tail fin and back skin. parent and probably reflects technical difficulties in making a com- plete dissection because of the radical changes in body and head shape. There was greater net incorporation of labeled proline in the back skin collagen of metamorphosing animals than in the controls ( Fig. 5). In the tail fin, on the other hand, the reverse occurred. The spe- cific activity of the back skin collagen was also higher, whereas there was little difference in tail fin specific activities up to 20 hours (Fig. 6). In the resting tadpole the amount of tail fin collagen was constant, and there was a progressive increase in both total and specific radio- activity over a period of 20 hours. The rates of synthesis and deg- radation must be the same, but whether there is some selective removal of old "cold" collagen or a nonselective random removal ()72 ('. M. LAPIERE AND J. CROSS Fig. 5. skin. 400- TAIL FIN 300 200 700 600 500 400 300 200 100 Ax BACK SKIN I ^ I / <» 5 10 15 20 25 HOURS AFTER PROLINE- H^ 30 Total radioactivity of hydroxyproline isolated from tail fin and back of newly deposited and old fibrils in this equilibrium state cannot be established from the data. In the metamorphosing animal there was marked reduction in total tail fin collagen, and diminished net incorporation, but a spe- cific activity increment identical with the normal (up to 20 hours) (Figs. 5 and 6). If the data (Fig. 5) for thyroxin-treated tail fins were "normalized" to account for the loss of about 50 per cent dry ANIMAL COLLAGENASE AND COLLAGEN METABOLISM 673 300 200- TAIL FIN • CONTROL --0 THYROXIN Rx 0 5 10 15 20 25 30 HOURS AFTER PROLINE-H^ Fig. 6. Specific activity of hydroxypioline isolated from the total collagen weight (including collagen and cells), the two curves would be superimposable. These observations suggest that the rate of syn- thesis of collagen was unchanged but the rate of resorption in- creased. In the back skin there was an increase in both total and specific activities in the metamorphosing animals as compared with the con- trols ( Figs. 5 and 6 ) , indicating increased synthesis. Metabolism of the Extractable and Insoluble Collagen Fractions Separation of collagen into soluble and insoluble fractions gives us a more detailed picture of events. In Table II is found the amount 674 C. M. LAPIERE AND J. GROSS of collagen and percentage of the total obtained in sequential neutral and acid extracts and in the insoluble residues of control and thyroxin-treated tail fin and back skin. The amounts of collagen in the different fractions of the tail fin were roughly similar, the insoluble representing the smallest amount. In the back skin, on the other hand, there was very little neutral-extractable collagen and the insoluble fraction represented the largest portion. This latter distribution is more characteristic of that found in terrestrial mam- mals. Thyroxin treatment for 6 days ( in this experiment ) resulted in 50 per cent diminution in the amount of neutral-extractable and in- soluble collagen per tail fin and a loss of 57 per cent in the acid- extracted. By far the largest fractional loss was in the insoluble fraction. In contrast, in the back skin there was an increase in neu- tral- and acid-extractable collagen accompanied by a diminution in the insoluble fraction. The net result was 53 per cent loss of total collagen in the tail fin and 19 per cent loss in the back skin. Thus, partition of the different collagen fractions differed markedly be- tween tail fin and back skin, and the influence of thyroxin on this pattern in the two tissues was also dissimilar. In all the isotope data we were struck by the fact that incorpora- tion of labeled proline into the neutral-extractable and insoluble collagen fractions occurred at nearly the same rapid rates. Examination of the radioactivity of the individual collagen frac- tions in the tail fin discloses some interesting changes during meta- morphosis. Net incorporation in the saline-extracted fraction of the control animals was about three times as great as in the insoluble residue (Fig, 7). In the resorbing tail a striking alteration is noted in the relationships of the net incorporation in the different fractions. The level of radioactivity of the neutral fraction was about equal to that of the insoluble collagen and considerably lower than in the corresponding control. The insoluble collagen net incorporation was about the same in both groups. The acid-extractable collagen was least labeled and its curye of incorporation resembled that of the insoluble fraction. Thus, the 50 per cent lower net incorporation seen in tlie total collagen of the metamorphosing tadpole tail fin ANIMAL COLLAGENASE AND COLLAGEN METABOLISM 675 10' - CONTROL \ M NaCI Extract ■ — — ^ Insoluble Acid Extract THYROXIN Rx 30 0 10 HOURS AFTER H^Pro Fig. 7. Total activity of hydroxyproline isolated from the three collagen fractions of tail fin. (Fig. 5) is due primarily to diminution in net incorporation in the saline-extracted fraction. Looking at the specific activity relationships (Fig. 8), in the tail fin we observe that the specific activity of the insoluble component is greatly increased over that of the neutral-extractable fraction in the resorbing tail, and is also much higher than that of the control insoluble collagen. Similarly, while there is a fairly constant specific activity in tlie acid-extractable component in the control, there is a progressive increase in that of the other group. It would appear likely that the rise in specific activity, coupled with a diminished amount of insoluble collagen, results from prefer- ential removal of an unlabeled, hence older, portion of this pool, indicating that the pool is metabolically inhomogeneous. Depression of total and specific activity of the neutral-extractable fraction in 676 C. M. LAPIERE AND J. GROSS 10^ CONTROL THYROXIN Rx 30 Fig. 8. 20 25 30 0 5 10 15 HOURS AFTER PROLINE-h' Specific activity of the three coHagen fractions isolated from tail fin thyroxin-treated animals along with 43 per cent diminution of the total amount of this fraction (Table II) suggests either diminished synthesis, accelerated breakdown, or increased conversion to the insoluble pool. Diminished net synthesis but not rate would be ex- pected because of the loss in total tissue dry weight, as discussed earlier. Since the total counts in the insoluble fraction are un- changed, there may also be an increased conversion rate. On examination of the back skin data ( Figs. 9 and 10 ) , it appears that in the neutral-extractable fractions, total and specific activities we:e practically identical for control and treated groups. Specific activity of the insoluble fraction, though somewhat elevated above that of the control, did not approach the differences seen in the tail fin. The specific activity of the neutral-extracted fraction always remained considerablv greater than that of the insoluble component although the total radioactivity was about the same for both. There ANIMAL COLLAGENASE AND COLLAGEN METABOLISM 677 CONTROL THYROXIN Rx |.0*-X L HOURS AFTER PROLINE-H^ Fig. 9. Total activity of the three collagen fractions isolated from back skin does appear to be more rapid increase in the radioactivity of the acid-extractable pool in metamorphosing tadpole back skin. In the back skin some resorption seems to be occurring, but at a slower rate than in the tail fin. The higher rate of collagen forma- tion in the back skin as compared with the tail fin ( Fig. 5 ) of resting tadpoles is not altered by metamorphosis. This picture of the dif- ferences between back skin and tail fin is consistent with the ob- servations of thickening of the dermal layer of the body during metamorphosis, associated with partial removal of the well ordered basement lamella and its replacement by a new, less well ordered terrestrial type of dermis ( Kemp, 1961 ) . Compared with the collagen distribution and turnover data from other animal species such as rat (Harkness et al, 1954) and guinea pig (Jackson, 1957; Jackson and Bentley, 1959), there are some significant differences in the tadpole. In the rodent skin the insolu- 678 C. M. LAPIERE AND J. GROSS CONTROL THYROXIN Rx J I I L_ 20 30 0 10 20 HOURS AFTER PROLINE - H^ 30 Fig. 10. Specific activity of the three collagen fractions isolated from back skin ble fraction is by far the largest, the acid-extractable next, and the neutral-extractable the least. The situation in the back skin of the tadpole is very similar, but that in the tail fin very different, as noted. There is a striking difference in radioactivity distribution. In mam- mal skin the acid-extractable fraction is labeled to a greater degree and more rapidly than the insoluble, although much less than the cold neutral-extractable moiety. In contrast, in the tadpole the acid-extractable collagen of both tail and back skin is tlie least and slowest labeled. The shape of the curve closely parallels that of the insoluble fraction rather than that of the neutral-extractable com- ponent. Another striking feature of the data is the nearly equal rate of ANIMAL COLLAGENASE AND COLLAGEN METABOLISM G79 incorporation of isotope in the neutral and insoluble fractions. This suggests that either there is an extremely rapid insolubilization of the neutral-extractable collagen, or both are derived independently from a common precursor. The fairly constant level of radioactivity obtained by all three collagen fractions over a 20- to 30-hour period is consistent with the observations of Jackson and Bentley ( 1959 ) in guinea pig skin. In their studies only the lower ionic strength cold neutral extracts showed a more rapid "turnover." The meaning is obscure, although the fairly constant high levels of labeled free proline may play a role in our experiments. It should be kept in mind that, although induction of meta- morphosis with thyroxin closely resembles the natural process, in- cluding increased thyroid activit\', it is possible that some of the isotope results are a direct effect of thvroxin per se. Though this is unlikely, it has been shown, for example, that thyroid compounds in the range of concentration employed here do stimulate protein synthesis. However, at most this would onlv influence the results in back skin and would not affect the findings associated with tail resorption. Collagen Degradation: Detection and Measurement OF A Collagenolytic Enzyme in Tadpole Tissues The long-recognized removal of collagen during remodeling has encouraged an intensive quest for an animal collagenase ( see Mandl, 1961, for review). To qualify as a true collagenase, the enzyme should digest native collagen at physiologic pH and temperature. The usual approach has been to apply whole tissue homogenates, or fractions therefrom, to a dissolved or solid collagen substrate. Measurements of viscosity decrement or the release of hydroxypro- line or free amino groups have been generally used as assay pro- cedures, analogous to the measurement of bacterial collagenase (Gallop et ah, 1957). The well known proteolytic enzymes such as trypsin, chymotrypsin, the cathepsins, and papain will hardly attack native collagen under physiologic conditions (Mandl, 1961; Gross, 1958; Oken and Boucek, 1957; Hodge et al, 1960; Kuhn et al, 1961). 680 C. M. LAPIERE AND J. GROSS Enzymes have been isolated from mammalian white blood cells (Sherry et ah, 1954) and from liver lysosomal fractions (Franklin and Wynn, 1961) and uterus (Morrione and Seifter, 1962; Woessner, 1962 ) which attack collagen to a slight extent in the acid pH range. A type of collagenolvtic activity in certain pancreatic extracts has been claimed ( Houck and Patol, 1962 ) . The only well characterized, true collagenase has been isolated from species of Clostridium (Mandl, 1961). In considering the possible reasons for failure to detect an animal collagenase, we suspected that such an enzyme must be present in extremely low concentrations in highly localized regions, otherwise uncontrolled degradation incompatible with functional stability would occur. It seemed unlikely that any of the current assay pro- cedures would be sensitive enough to detect low concentrations of this enzyme. In addition, it has been established (Gallop et al., 1957) that bacterial collagenase binds to collagen fibers and is re- leased only after digesting the protein; this being the case, the usual extraction procedures might fail to release animal collagenase from the collagen still present in the tissues. It also seemed possible that tissue collagenase might have a very short active life and either be rapidly inhibited or be degraded by other enzymes. For these reasons we set out to detect collagenase activity by culturing small bits of tissue on a reconstituted fibrous collagen substrate. It was thought that any newly produced or activated collagenase might diffuse away from the tissue, bind to the fibrils of the substrate, digest them, and thereby reveal activity as an expanding area of lysis. This prediction was borne out and we have found collagenolytic activity in cultured tadpole tissues and in mam- malian uterus and bone (Gross and Lapiere, 1962; Gross, Lapiere, and Tanzer, 1963). Figure 11 presents photographs of a typical culture of a collagen gel, showing the expanding areas of lysis around the explants of tail fin tissue. The collagen substrates used were either acid-extracted calf skin collagen or saline-extracted radioactive guinea pig skin collagen which were purified, lyophilized, and dissolved in cold physiologic salt solutions in concentrations of approximately 0.1 per cent. The collagen polymerized to typical cross-striated fibrils (characteristic ANIiMAL COLLAGENASE AND COLLAGEN METABOLISM 681 Fig. 11. Four tadpole fin explants on reconstituted calf skin collagen gel before (A) and after (B) incubation for 24 hours at 37°C. 682 C. M. LAPIERE AND J. GROSS of native structure) when warmed for short periods at 37°C (Gross and Kirk, 1958). These opalescent fibrous gels were resistant to attack by the common proteolytic enzymes (Gross and Lapiere, 1962). The tissues of tadpoles (previously sterilized by adding penicillin-streptomycin and chloramphenicol to the aquarium water ) were cultured on sterile collagen gels in small ring chambers or petri dishes using a medium composed of amino acid- and vitamin- supplemented amphibian Tvrode solution. The following collageno- l)tic properties of tadpole tissues were recently described by the authors (Gross and Lapiere, 1962). 1. The area of lysis expanded in exponential fashion and was linearly related to the size of the tissue explant. 2. Living tissues were required for collagenolytic action. Freezing and thawing the explant prevented collagenolytic activity. 3. The pH of the medium remained in the physiologic range, and up to 80 per cent of the collagen substrate could be digested to dialyzable peptides. No free hydroxvproline was released. 4. Twelve different tadpole organs were examined for collageno- lytic activity. Only four were active, tail fin and back skin, gill, and gut. These four tissues undergo the most dramatic remodeling and resorption of connective tissue during metamorphosis. The back skin showed considerably less activity than did the tail fin. 5. Tissue cultures could be used for quantitative microassay; the area of lysis was directly proportional to the release of radioactive collagen fragments when radioactive collagen substrates were used. The lytic process could be readily compared with that of a standard amount of purified bacterial collagenase placed on the gel. Further studies have revealed changes in collagenolytic activity in the cultured tail fin tissues of animals in metamorphosis. In the first experiment, groups of tadpoles were exposed to thyroxin for 2 and 6 days. Sections of tail fin tissue approximateh' 1 mm- in area were cultured on radioactive collagen gels at 27°C. At appropriate intervals the cultures were assayed for collagenolytic activity. The results are shown in Fig. 12. Two davs of treatment produced little increase in activity. After 6 days, however, when the tail had dimin- ished by approximately 30 per cent in length, collagenolytic activity AiSriMAL COLLAGENASE AND COLLAGEN METABOLISM 683 20 30 40 50 HOURS OF INCUBA TION Fig. 12. Collagenolytic activity of tadpole tail fin in culture on radioactive collagen gels at 27°C. was fourfold greater than in the control group. In several other re- peat experiments the increase in activity of the treated animal tissues ranged from two to four times that of the controls. In a second experiment, groups of tadpoles were maintained in a thyroxin-containing bath for inter\als ranging from 1 to 7 days. Figure 13 illustrates collagenohtic activity in tail fin and back skin along with the decrement in tail length. The direct relationship of 684 C. M. LAPIERE AND J. GROSS I i 50- 40 30- 20 10- 0---0 BACK SKIN X / _ r/ ^ _ •-..^,jj^_xr' Ayt^^^'"^ V - 1 1 1 1 1 1 1 . — o 1 - 15 - 10 - 0 ^ J 5 ^ $ ^ S 0 1 2 3 4 5 6 7 DAYS AFTER THYROXIN Fig. 13. Collagenolytic activity of tadpole tail fin and back skin, and de- crease in tail length, as functions of time of exposure to thyroxin. increase in enzyme activity with loss in tail length is clear; the ac- tivity of the back skin shows no relation to the change in tail size. Isolation and purification of the tadpole collagenase from mass cultures was made possible by the development of two sensitive assay techniques which detect as little as 0.01 />tg of bacterial col- lagenase. The first assay depends upon the release of radioactive peptides from radioactive collagen, the released fragments being measured aftei incubation ( Lapiere et al., submitted for publica- tion). The second method is based upon decreased collagen fibril formation after incubation of a reaction mixture containing the en- zyme preparation and dissolved collagen at 25 °C. The temperature is then elevated to 37 °C, polymerization of the collagen to an opales- cent gel occurring in a few minutes. The reduction in opacity is a direct measure of collagenolytic activity (Nagai et al., unpublished observations ) . By growing large amounts of tadpole tail fin in petri ANIMAL COLLAGENASE AND COLLAGEN METABOLISM 685 dishes on filter paper in Tyrode solution, collagenase could be har- vested from the culture medium (Nagai et al., 1963). After several days of incubation, tissue fragments were removed, the breakdown products were dialyzed away, and the fluid in the bag was lyophil- ized to a dry powder. This material has been partially purified on Sephadex columns and compared with bacterial collagenase. Par- tially purified tadpole collagenase has a pH optimum between 6.5 and 7.8, is destroyed by heating between 50 and 60 °C for 10 min- utes, and is reversibly inhibited by EDTA. It probably has a lower molecular weight than the Clostridium collagenase as indicated by elution time from Sephadex columns. Using the radioactive collagen microassav, we have reexplored the possibility of extracting substantial amounts of collagenase from tadpole tail tissues, by making extracts of control and resorbing tail fin with and without EDTA. (It has been found that the firm en- zyme-substrate complex between bacterial collagenase and solid collagen can be broken with this chelating agent (Gallop et al., 1957). The activity of the enzyme can be restored with calcium.) We also carefully reexamined the effect of known proteases on our reconstituted collagen substrate (Table III). We have found a small but definite amount of digestion of reconstituted collagen fibrils by very high concentrations of trypsin, chymotrypsin, papain, elastase, and pronase. This activity had to be differentiated from that of collagenase in order to evaluate the true effect of tissue extracts. It was noted that, unlike that of collagenase, the action of the other proteolytic enzvmes was not progressive with time but was complete within 1 to 2 hours. Moreover, considerably larger amounts of en- zyme were required for detection of this minimal activity, except for pronase, which was effective at lower concentrations. A second significant difference between true collagenolytic action and that of the other enzymes was found in response to EDTA. Both bacterial and tadpole collagenase showed complete reversible inhibition by low concentrations of EDTA, whereas the other proteases were only slightly affected (Table III). By tabulating the fractional inhibition caused by EDTA (last column. Table III) we have a sensitive index whereby collagenase activity may be differentiated from nonspecific 686 C, M. LAPIERE AND J. GROSS TABLE III. COLLAGENOLYTIC ACTIVITY OF PROTEASES AND TiSSUE EXTRACTS Cpm released from collagen subs trate - after 3 hours at 37 °C" EDTA C No EDTA Co A Co - C C Bacterial collagenase'' 0.1 Mg 1 252 251 251 Tadpole collagenase* 0.2 nig 6 497 491 82 Trypsin 1 mg 56 96 40 0.7 Chymotiypsin 1 mg 164 171 i 0.04 Pronase 10 Mg 153 198 45 0.3 Cathepsin C + SH 3 units 6 4 0 0 Papain + SH 1 mg 160 170 10 0.06 Elastase 0.5 mg 100 117 17 0.2 Cpm released after 24 hours Tadpolr fin extract- 1 day th>Toxin Rx 6S 113'^ 45 0.7 2 davs " " 66 98 32 0.5 3 " 57 83 16 0.3 ^ i( (i (( 59 82 23 0.4 5 " 74 102 28 0.4 6 " 47 93 46 1.0 Tadpole body skin 1 day thyroxin Rx 70 90 20 0.3 2 days 61 92 31 0.5 ^ w it '< 50 67 17 0.3 5 " 83 102 19 0.2 6 " 68 81 13 0.2 " Collagen substrate in each tube contains 810 cpm. Total volume is 0.6 ml. '' Purified CI. histolyticum collagenase (courtesj- of Dr. P. Gallop) and crude tadpole collagenase (lyophilized tail fin culture medium). ■^ 100 jul of extract representing 20 mg of fresh tissue homogenate in 0.12 M NaCl + 0.008 M EDTA. •^ Excess of Ca^"*" added prior to assay. protease activity. Values greater than 1 which increase with time of incubation are indicative of the presence of collagenase activity. Extracts of fresh tail fin and back skin from nonmetamorphosing and th)a-oxin-treated animals have, thus far, rexealed onh' proteohtic type activity on collagen. animal collagenase and collagen metabolism 687 Discussion The accretion or diminution of structural elements such as colla- gen is probal)ly accomplished by altering the balance between rates of synthesis and rates of degradation. The manner in which this might occur is not yet known. The present studies suggest that where rapid collagen removal occurs, as in the resorbing tadpole tail, there is an increase in degradative activity. Synthetic activity seems to remain constant. It is of interest that there is substantial collagenolytic activity in cultures of nonmetamorphosing tadpole tissues but such activity is essentially nondetectable in fresh tissue extracts. It may be that the culture technique allows the tissue to reveal its full potential, which it cannot do under in vivo restraints. Such hypothetical inhibitors were not found when large amounts of tadpole tissue extracts were added to the culture medium to prevent the appearance of colla- genase. Earlier experiments (Gross and Lapiere, 1962) indicate that there is no storage of the active enzyme in the tissues, but that its accumulation in culture is a result of new synthesis or activation of a zymogen molecule. The possibility of induction or inhibition by substrate or its breakdown products requires further inquiry. The increase in collagenolytic activity in cultures of metamorphos- ing tail fin ranged from two- to fourfold in repeated experiments. This increment might be largely ascribed to an increased con- centration of enzyme-producing cells rather than to increased syn- thesis per cell. The concentration effect might result from the loss in tissue water. We propose that collagen resorption increases in metamoi-phosis because the cells making collagenase carry the en- zyme to the substrate when they invade the collagenous basement lamella. We hypothesize from these metabolic studies that the old insoluble collagen framework is preferentially removed and replaced continu- ously by a new insoluble scaffolding. It is possible that either specific cells responsible for removal can select out the older fibrils or the newly deposited fibrils are protected in some manner from attack by these cells. The ground substance, which presumably is produced 688 C. M. LAPIERE AND J. GROSS in concert with new collagen, may provide a protective coat. There is some evidence to indicate that newly formed collagen (reticulin) is surrounded by ground substance not found in older mature collagen bundles (Robb-Smith, 1957). At this stage of the game there are other equally plausible hypotheses. Hay and Revel (1963), using radioautography, have noted that newly deposited collagen first be- comes evident in the regenerating basement lamella adjacent to the epithelial cells. During thyroid-induced tail resorption we (Usuku and Gross, unpublished observations) noted the earlv invasion of the deeper layers of the lamella by mesenchymal cells with local collagen disruption; only later were the more peripheral, presumably newly deposited, collagen layers attacked. There is also the possi- bility that the spreading apart of the deeper collagen layers early in resorption is a result of a shift of water into this layer. Loosening of the structure might then facilitate the entrance of the mesen- chvmal cells. The differences in collagen turnover between back skin and tail fin suggest that the same stimulus, thyroxin, affects skin in two adjacent regions in different ways. There was no detectable increase in collagenolytic activitv in the back skin, nor was there nearly the same degree of dehydration. The net incorporation of label in the back skin of thyroxin-treated animals was greater than that in the normal, in sharp contrast to the picture in the tail tissues. We know from older studies such as those of Clausen (1930) that body skin transplanted to the tail does not resorb during metamorphosis, whereas tail fin transplanted to the back undergoes resorption at the same rate as in its parent site. There may be some difference in the properties or composition of the collagen of the tail fin and the back skin to account for this observation. Certainly there is a marked contrast in collagen concentration and distribution be- tween these two sites. Thus far, examination of denaturation tem- perature indicates that the stability of the acid-extracted collagen in the two regions is identical, and even young and old bullfrog collagens are no different in this respect. Perhaps other analyses may demonstrate a difference which could alter susceptibility to collagenase. However, it seems more likely that the cells responsible for the reorganization of the tissues in different regions respond ANIMAL COLLAGENASE AND COLLAGEN METABOLISM 689 differently to thyroid hormone. Perhaps in the loci where very in- tricate remodeling takes place, as in the mouth parts, the differential response to thyroid hormone in highly localized regions plays a key role. By analyzing small in vitro systems we hope to be able to map out regions of differential response and ultimately determine the nature of this selectivity. It might also be asked whether the cells responsible for collagen synthesis can produce collagenase or whether there is segregation of function in different cell tvpes. In either case one wonders what the stimuli are which tell the cells to synthesize or digest collagen. Would both potentialities exist in the same cell or would there be further specialization? At the present time we have no idea as to which cells produce collagenase. We hope to answer the question with cell suspensions and single-cell cultures. With regard to the beha\dor of mesenchymal tissues in morpho- genesis, it is conceivable that the extracellular framework plays a role in limiting growth and form of an organ, either as competitor for space, or as a contact inhibitor. If this turns out to be the case, then the remodeling of mesenchvme under hormonal control may have significance beyond that of a model system. Summary Resorption and new formation of large collagen-containing struc- tures have been explored at the chemical level in a controllable remodeling svstem, the metamorphosing and resting tadpole. Ex- periments on collagenolvtic activity and metabolic turnover of col- lagen have been performed in thyroxin-induced metamorphosis, comparing tadpole tail fin collagen (which is completely removed during metamorphosis) with back skin collagen (which undergoes increased production ) within the same animals. The most dramatic change in composition in the tissues of meta- morphosing animals is a considerable loss of water, with collagen and cell concentrations proportionately increased. Tritiated proline was injected intraperitoneally into thyroxin- treated and control animals. There was greater net incorporation of labeled proline in the back skin collagen of metamorphosing animals 690 C. M. LAPIERE AND J. GROSS than in the controls, whereas in the tail fin the reverse occurred. In the back skin there was an increase in both total and specific activities in the metamorphosing animals as compared with the controls, suggesting increased synthesis. The elevated specific ac- tivit}' in the face of diminished total collagen suggests a preferential attack on the older "cold" collagen. Partition of the dift'erent collagen fractions differed markedly be- tween tail fin and back skin, and the influence of thvroxin on this pattern in the two tissues was also dissimilar. In the resorbing tail a striking alteration was noted in the relation- ships of the net incorporation in the different fractions. The specific radioactivitv of tlie neutral-extractable fraction was well below that of the insoluble collagen and was also considerably lower than in the corresponding control. A rise in specific activity, coupled with a diminished amount of insoluble collagen, appeared to result from preferential removal of an unlabeled, hence older, portion of a meta- bolically inhomogeneous pool. We speculate from these metabolic studies that the old insoluble collagen framework is preferentiallv removed and replaced continu- ously by a new insoluble scaffolding. The newly deposited fibrils may be protected in some manner from attack bv the enzyme. Per- haps the ground substance, produced in concert with new collagen, mav provide a protective coat. Alternatively, the older fibers are attacked first because of geographic proximity to the cells with lytic power. Collagenase activity was tested by culturing small bits of tissue on a reconstituted fibrous collagen substrate. Living tissues were required for collagenolytic activity. Of twelve different tadpole or- gans examined for collagenolytic activitv, four were active, namely tail fin and back skin, gill, and gut, i.e., organs which undergo the most dramatic remodeling and resoi^ption of connective tissue during metamorphosis. Where rapid collagen removal occurred, as in the resorbing tad- pole tail, there was an increase in degradative activitv. This effect might simply be due to a concentration of enzyme-producing cells as a result of tissue water loss. The increased lysis of collagen may be ascribed to the invasion of the collagenous basement lamella by ANIMAL COLLAGENASE AND COLLAGEN METABOLISM 691 the lytic cells, i.e., they carry the enzyme to its substrate. It seemed unlikely that storage of the active enzyme in the tissue occurred. Its accumulation in culture appeared to be a result of new synthesis or activation of a zvmogen molecule. By growing large amounts of tadpole tail fin in petri dishes on filter paper in Tyrode solution, coUagenase could be harvested from the culture medium. Partially purified tadpole collagenase had a pH optimum between 6.5 and 7.8, was destroyed bv heating between 50 and 60 °C for 10 minutes, and was reversibly inhibited by EDTA. A small but definite amount of reconstituted collagen fibrils was digested by very high concentrations of trypsin, chymotrypsin, pa- pain, elastase, and pronase. Unlike that of collagenase, the action of other proteolytic enzymes was not progressive with time but was complete within 1 to 2 hours; and considerably larger amounts of enzyme were required for detection of this minimal activity. Both bacterial and tadpole collagenase showed complete reversible in- hibition by low concentrations of EDTA, whereas the other proteases were only slightly affected. It seems likely that the cells responsible for the reorganization of the tissues in different regions respond differently to thyroid hor- mone. Perhaps in the loci where very intricate remodeling takes place, as in the mouth parts, the differential response to thyroid hormone in highly localized regions plays a key role. Acknowledgments. This work was supported by United States Public Health Service Grant AM-05142-02. Publication No. 342 of the Robert W. Lovett Memorial Group for the Study of Diseases Causing Deformities, Harvard Medical School at the Massachusetts General Hospital, Boston, Massachusetts. Refer ENCE.S Allen, B. M. 1929. The influence of the thyroid gland and hypophysis upon growth and development of amphibian larvae. Quart. Rev. Biol, 4, 325-352. Clausen, H. J. 1930. Rate of histolysis of anuran tail fin and muscle during metamorphosis. Biol. Bull, 59, 199-210. Etkin, W. 1955. Metamorphosis. In Analysis of Development (B. H. Wil- 692 C. M. LAPIERE AND J. GROSS ler, P. Weiss, and V. Hamburger, editors), pp. 631-663. W. B. Saun- ders Co., Philadelphia, Pa. Franklin, D. M., and Wynn, C. H. 1961. The degradation of acid-soluble collagen by rat-liver preparations. Biochem. J., 85, 276-282. Frieden, E. 1961. Biochemical adaptation and anuran metamorphosis. Am. Zoologist, 1, 115-149. Gallop, P. M., Seifter, S., and Meilman, E. 1957. Studies on collagen. I. The partial purification, assay and mode of activation of bacterial collagenase. /. Biol. Chcm., 227, 891-906. Gould, B. S., and Manner, G. 1962. Some aspects of collagen biosvnthesis and resorption. In Collagen (N. Ramanathan, editor), pp. 311-326. Interscience Publishers, Inc., New York. Green, N. M., and Lowther, D. A. 1959. Formation of collagen hvdroxv- proline in vitro. Biochem. J., 71, 55-66. Gross, J. 1958. Studies on the formation of collagen. I. Properties and fractionation of neutral salt extracts of normal guinea pig connective tissue. /. Exptl. Med., 107, 247-263. Gross, J. 1959. On significance of the soluble collagens. In Connective Tissue, Thrombosis and Atherosclerosis (I. H. Page, editor), pp. 77- 95. Academic Press, Inc., New York. Gross, J., Highberger, J. H., and Schmitt, F. O. 1954. Collagen structure considered as states of aggregation of a kinetic unit. The tropocolla- gen particle. Proc. Natl Acad. Sci., 40, 679-688. Gross, J., and Kirk, D. 1958. The heat precipitation of collagen from neutral salt solutions: Some rate-regulating factors. /. Biol. Chem., 233, 355-360. Gross, J., and Lapiere, C. M. 1962. Collagenolytic activity in amphibian tissues: A tissue culture assay. Froc. Natl. Acad. Sci., 48, 1014-1022. Gross, J., Lapiere, C. M., and Tanzer, M. L. 1963. Organization and dis- organization of extracellular substances: The collagen system. In Cytodifferentiation and Macromolecular Synthesis (M. Locke, edi- tor), pp. 175-202. Academic Press, Inc., New York. Gudernatsch, J. F. 1912. Feeding experiments on tadpoles. I. The in- fluence of specific organs given as food on growth and differentiation. A contribution to the knowledge of organs with internal secretion. Arch. Entwicklungsmcch . Organ., 35, 457-483. Harkness, R. D., Marko, A. M., Muir, H. N., and Neuberger, A. 1954. The metabolism of collagen and other proteins of the skins of rabbits. B/oc/?cw./., 56', 558-569. Hausmann, H., and Neuman, W. F. 1961. Conversion of proline to hy- droxyproline and its incorporation into collagen. /. Biol. Chem., 236, 149-152. Hay, E. D., and Revel, J. P. 1963. Autoradiographic studies of the origin of the basement lamella in Amhlystoma. Develop. Biol., 7, 152-168. ANIMAL COLLAGENASE AND COLLAGEN METABOLISM 693 Hodge, A. J., Highberger, J. H., Deffner, O. J., and Schmitt, F. O. 1960. The effect of proteases on the tropocollagen macromolecule and on its aggregation properties. Proc. Natl. Acad. Sci., 46, 197-206. Houck, J. G., and Patol, Y. M. 1962. The collagenolvtic activity of pan- creas. Ann. N. Y. Acad. Sci., 93, 331-349. Jackson, D. S. 1957. Connective tissue growth stimulated by carrageenin. I. The formation and removal of collagen. Biochem. J., 65, 277-284. Jackson, D. S., and Bentley, J. P. 1959. On the significance of the ex- tractable collagens. /. Biophys. and Biochem. CytoL, 7, 37-48. Kemp, N. E. 1959. Development of the basement lamella of larval anuran skin during metamorphosis. Develop. Biol., 1, 459-475. Kemp, N. E. 1961. Replacement of the larval basement lamella by adult- type basement membrane in anuran skin during metamoi-phosis. Develop. Biol, 3, 391-410. Kollros, J. J. 1961. Mechanisms of amphibian metamorphosis: Hormones. Am. Zoologist, 1,107-114. Kuhn, K., Kuhn, J., and Mannig, K. 1961. Einwirkung von Trypsin auf gelostes Kollagen. Z. physiol. Chem., 326, 50-60. Lapiere, C. M., Nagai, Y., and Gross, J. Submitted for publication. Lynn, W. G., and Wachowski, H. E. 1951. The thyroid gland and its func- tions in cold blooded vertebrates. Quart. Rev. Biol., 26, 123-168. Mandl, I. 1961. Collagenase and elastases. Advan. Enzym.ol., 23, 174-264. Morrione, T. G., and Seifter, S. 1962. Alteration in the collagen content of the human uterus during pregnancy and post partum involution. /. Exptl. Med., 115, 357-365. Nagai, Y., Lapiere, C. M., and Gross, J. 1963. Isolation and partial char- acterization of an animal collagenolytic enzyme. Federation Proc, 22, 648. Nagai, Y., Lapiere, C. M., and Gross, J. Unpublished observations. Oken, D. E., and Boucek, R. J. 1957. Proc. Soc. Exptl. Biol, and Med., 96, 367-369. Piez, K. A., and Likins, R. C. 1957. The conversion of lysine to hydroxyly- sine and its relationship to biosynthesis of collagen in several tissues of the rat. /. Biol. Chem., 229, 101-109. Prockop, D. J., Peterkofsky, B., and Udenfriend, S. 1962. Studies on the intracellular localization of collagen synthesis in the intact chick embryo. /. Biol. Chem., 237, 1581-1588. Robb-Smith, A. H. T. 1958. The relationship of reticulin to other "colla- gens." In Recent Advances in Gelatin and Glue Research (G. Stainsby, editor), pp. 38-44. Pergamon Press, Oxford. Rouiller, G. 1961. Collagen fibres of connective tissue. In The Biochemistry and Physiology of Bone (G. H. Bourne, editor), pp. 138-147. Aca- demic Press, Inc., New York. Schmitt, F. O., Gross, J., and Highberger, J. H. 1955. Tropocollagen and (i94 c. M. lapie;re and j. gross the properties of fibrous collagen. Exptl. Cell Research, Suppl. 3, 326-334. Sherry, S., Troll, W., and Rosenbkim, E. D. 1954. "Collagenase" activity of cathepsins. Proc. Soc. Exptl. Biol, and Med., 87, 125-128. Tanzer, M. L., and Gross, J. Collagen metabolism in normal and lathyritic chicks. /. Exptl. Med. ( in press ) . Taylor, A. C, and Kollros, J. J. 1946. Stages in the normal development of Rana pipiens larvae. Anat. Record, 94, 7-23. Usuku, G., and Gross, J. Unpublished observations. Woessner, J. F. 1962. Catabolism of collagen and non collagen protein in the rat uterus during post partum involution. Biochem. J., 83, 304- 314. Author Index Numbers in boldface type indicate pages with complete bibliographical data, to permit cross reference to authorship abbreviations (et al., etc.) appearing in the text. Numbers in italics indicate the first and last pages of authors' chapters in this book. Abbott, R. T., 61, 87 Adelson, L. M., 648, 65.3, 654, 661 Albertsen, K., 425, 442 Albright, F., 312, 315, 604, 606 Albright, J. T., 188, 210 Allen, B. M., 663, 691 Allen, W., 654, 658 Altman, K. I., 425, 441 Amprino, R., 376, 382, 416, 423, 440 Andresen, ^^, 140, 143, 149, 307, 315 Ankel, W. E., .56, 59, 60, 65, 87 Applebaum, E., 173, 185 Arev, L. B., 489, 492 Armstrong, W. D., 471, 495, 639, 662 Armstrong, W. G., 188, 207, 208 Arnold, J.^S., 448, 467, 472, 486, 492 Ascenzi, A. 490, 492 Association of Official Agricultural Chemists, 532, 554 Atkinson, P. J., 423, 440 Au, W. Y. W., 563, 575, .579, 580, 584, 585, 588, 590, 608, 639, 641, 661 Aub, J. C..3.54, 355, 3.58, 361, 362, 364, 366, 368 Austen, F. K., 386, 443 Austin, L. T., 99, 149 Bacharach, A. L., 517, 530 Badanes, B. B., 114, 149 Baer, P. N., 312, 315, 316 Bailie, J. M., .528, 529 Baker, B. B., 61, 87 Baker, B. L., 523, 529 Baker, G., 166, 169 Baker, R. F., 188, 208 Balogh, K., Jr., 578, 587 Bargen, J. A., 99, 149 Barker, S. B., 591, 606 Barnett,E., 422, 431,444 Barnicoat, C. R., 94, 149, 155-170, 155, 161, 167, 169 Barnicot, N. A., 490, 492, 626, 627, 629, 634 Barr, D. P., 312, 316 Barrnett, R., 614, 635 Bassett, C. A. L., 491, 492 Bassett, S. H., 428, 440 Bastomsky, C., 425, 440 Bauer, G. C. H., 377, 379, 382, 392, 413, 416, 4.30, 431, 440, 448, 467 Beaulieu, M. M., .595, 606 Bechtol, C. O., 431, 445 Becks, H., 312, 316 Bedford, Duke of, 347, .351, 364 Belanger, C., 531, .535, 554 Belanger, L. F., 531-556, 531, 535, 539, .543, 554, 556, 639 Belding, L. J., 270, 279 Belding, P. H., 270, 279 Bell, D. J., 432, 440 Benarde, M. A., 294, 296 Benedict, P. H., 428, 442 Bent, A. C., 340, 364 695 C96 AUTHOR INDEX Bentlev, J. P., 667, 677, 679, 693 Bergeiidahl, G., 472, 493 Berglund, G., 424, 440 Bergman, G., 173, 185 Bernick, S., 94, 188, 208, 285-296, 285, 294, 296 Bertolin, A., 417, 425, 426, 440, 441 Besic, F. C, 219, 258 Bessev, 0.,311,316 Bhandarkar, S. D., 427, 428, 435, 440 Bhaskar, S. N., 94, 96, 321-SS7, 323, 328, 330, 336, 337 Bhussrv, B. R., 188, 208. 231, 258 Bibbv,'B. G., 219, 260, 266, 278, 279, 648, 649, 653, 658 Bickel, A., 355, 364 Biedeimann, W.. 190, 209 Bieri, J. G., 99, 150 Bird, C. K., 112, 149 Bjerrum, J., 252, 258 Black, A. D., 102, 150 Black, G. v., 102 Blackwell, R. O., 655, 658 Blair, J. E., 472, 494 Blake,T. W., 57, 87 Blanford, W. T., 347, 364 Blauel, G., 355, 358, 363, 364, 365 Blavney, J. R., 267, 270, 281 Bloiim, T., 348, 349. 365 Bloom, M. A., 489, 490, 492 Bloom, W., 425, 442, 448, 467, 472, 489, 490, 492, 494 Bluhm, M., 427, 428, 435, 440, 444 Blythe, W. B., 386, 443 Bodansky, A., 472, 494 Bodecker, C. F., 112, 113, 116, 150 Boettger, C. R., 56, 87 Bogoroch, R., 141, 152, 249, 260 Bollett, A. S., 558. 573, 574 Bonar, L. C., 97, 150 Bonow, B. E., 264, 280 Boone, S. M., 59, 88 Boothrovd, B., 195, 497-514, 497, 501, 503, 505, 513, 645, 646, 659 Borle, A., 557, 558, 559, 560, 562, 567, 574, 579, 587, 589, 593, 596, 598, 603, 604, 606, 640, 658 Boucek, R. J., 679, 693 Bowing, H.H., 311, 318 Bovle, P. E., 105, 150, 311, 316 Brander, A. A. D., 347, 348, 365 Brat, v., 311,318 Brennan, J. C., 430, 445 Brinch, 0".,311, 316 Briner, W. W., 219, 258 Brodie, A. G., 100, 151 Broman, G. E., 416, 418, 423, 445 Bronner. F., 417, 435, 440, 472, 492 Brown, H. K., 276, 277, 281 Brown, L. R., 219, 260, 277, 279 Brown, W. H., 328, 337 Browne, R. C., 116, 153 Brnce, K. W., 311, 316 Brudevold, F., 173, 175, 183, 185, 186, 188, 199, 205, 209, 210, 219, 222, 248, 258, 259, 260, 264, 276, 277, 279, 280 Bruhin, H., 361, 365 Brnns, D. L., 435, 443 Buchanan, G. D., 471, 495, 570, 575 Bucher, W. H., 59, 87 Buhr, A. ]., 424, 432, 441, 442 Bujard, E., 515, 529 Buka, R., 591, 607 Bulger, H. A., 312, 316 Bullier, P., 349, 355, 365 Bunting, R. W., 117, 278, 280 Burgess, R. C., 276, 277, 281 Burket, L. W., 312, 316 Burnett, G. W., 214, 258, 275, 280 Burris, R. H., 591, 598, 608 Burrows, L. R., 189, 209 Burrows, R. B., 489, 493 Burstone, M. S., 511, 512, 516, 529, 530, 614, 635 Cabrera, A., 349, 350, 365 Cabrini, R. L., 511, 512, 527, 528, 530, 614, 636 Caldwell, J. P., 429, 433, 441, 645, 658 Cameron, D. A., 455, 467, 490, 493, 501, 503, 505, 512, 645, 658 Canerv, J. J., 386, 427, 441, 442 Cannon, H. G., 6, 23 AUTHOR INDEX 697 Carlsson, A., 377, 382, 393, 413, 416, 430,431,440,448,467 Carnes, W. H., 472, 493 Carriker, M. R., 4, 23, 55-89, 56, 57, 59, 81, 87, 88, 647 Carter, W. J., 656, 661 Casuccio, C, 386, 416, 425, 435, 441 Caton, J. D., 340, 341, 354, 365 Changus, G. W.,511,512 Chavin, W., 376, 383 Chen, M. J., 632, 635 Chen, P. S., Jr., 190, 209, 559, 575, 579,588,591,606,607 Chievitz, O., 140, 150, 372, 382 Clark, I., 434, 441, 471,493 Clausen, H. J., 688, 691 Clerkin, E. P., 386, 441 Cloosen, J., 381, 382 Cloud, P." E., 27, 52 Cobb, J. R., 417, 435, 440 Cohen," M. M., 312, 318 Cohen, P., 429, 442 Cohen, R. B., 578, 587 Cohn, D. v., 560, 574, 577-588, 578, 579, 581, 585, 587, 589, 596, 600. 603, 604, 607, 646 Cohn, V. H., 619, 636 Cole, W., 430, 445 Collins, L. C, 419, 443 Comar, C. L., 472, 494, 495 Consolazio, C. F., 265, 282 Cooke, A. M., 432, 441 Coolidge, E. D., 307, 316 Coolidge, T. B., 188, 209, 219, 258 Copp, D. H., 531-556, 533, 545, 554, 639 Corey, K. R., 431, 442 Cotte, J., 33, 34, 52 Coues, E., 340, 365 Counsell, A., 297, 317 Courts, A., 386, 416, 424, 432, 443 Coventry, M. B., 307, 318 Crabb, H. S. M, 181, 185 Cretin, A., 646, 658 Crittenden, L. B., 312, 316 Crossland, C, 3 Curtis, H. J., 448, 467 Curtiss, P. H., 425,442 Daems, W. Th., 511, 513 Daeschner, C. W., 430, 445 Dahlin, F. C, 462, 468 Dale, J. C, 472, 494 Dallemagne, M. H., 381, 382 Danial, E. J., 97, 150 Darby, E. T., 117, 150 Darling, A. 1., 98, 171-186, 173, 179, 181, 183, 184, 185, 186, 205, 215, 219, 258 Datta, S. P., 490, 492, 626, 634 Davidson, A. G. F., 545, 554 Davidson, J. N., 591, 607 Davies, G. N., 648, 653, 658 Dawes, C, 114, 116, 637-662, 650, 653, 654, 655, 658, 660 Dawson, R. L., 434, 442 Deakins, M. L., 188, 210 Deffner, O. J., 679, 693 Deiss, W. P., Jr., 560, 570, 574, 646, 660 Del Giovane, L., 417, 425, 444 deMan,J.C.H., 511,513 De Morgan, C, 371, 383 Dent, C. E., 437, 441 Deutsch, N. M., 432, 433, 445, 451, 469, 489, 490, 495 Devereux, E. D., 270, 280 Dingle, J. T., 558, 573, 574, 628, 635, 636 Dingwall, J. A., 400, 441, 446 Di Stefano, V., 559, 575, 579, 588 Dodge, A. M., 640, 656, 660 Domm, L. V., 489, 490, 492 Donaldson, J. C, 361, 365 Donne, T. E., 352, 365 Doty, M. S., 4, 23 Doutt, J. K., 361, 365 Dow, E. C, 431, 441 Dowse, C. M., 94, 206, 209, 560, 574, 589-608, 589, 603, 604, 607, 630, 636 Doyle, F. H., 422, 441 Draskoczy, P. R., 578, 587, 617, 621, 624, 635 Dreizen,S., 311,316 Driscoll, E. J., 312, 316 Dubos, R., 646, 659 698 AUTHOR INDEX Duckett, J. W., 429, 446 Dudley, H. R., 428, 442, 499, 501, 503, 505, 513, 578, 587 Duerden, J. E., 3, 23 Duffv, J. H., 312, 316 Dugal, L. P., 68, 88 Dulce, H.-J., 646, 659, 661 Dull, T., 425, 440 Dunn, J. K., 656, 661 deDuve, C, 511, 512 Du^-vensz, F., 424, 442 Eanes, E. D., 264, 276, 283 Egawa, J., 604, 605, 607 Eggers Lui-a, H., 648, 651, 654, 659 Eidinger, D., 535, 555 Eisenberg, E., 312, 318, 427, 430, 441, 442 Elliott, H. C, Jr., 217, 259 Elliott, J. R., 471, 493, 531, 556, 639, 662 Ellison, S. A., 649, 660 Elsburv, W. B., 99, 150 Emery, K. O., 50, 52 Engel, M. B., 486, 489, 493, 494, 579, 581, 588, 589, 593, 596, 603, 604, 607 Engfeldt, B., 173, 185, 472, 488, 493, 551, 555 Engstrom, A., 416, 423, 440, 472, 493 Enright, J. J., 652, 659 Ensfield,"B.J.,266, 282 Ensinck, J., 435, 444 Ericsson,Y., 114, 115, 150, 255, 258, 652, 659 Ershoff, B. H.,311,318 Ervin, R. F., 267, 270, 282 . Essner, E., 511, 513 Etchells, J. L., 270, 280 Etkin, Wi, 664, 691 Ettinger, R. H., 591, 607 Evans, D. G., 649, 651, 659 Eyler, W. R., 419, 444 Fabian, F. W., 294, 296 Fabry, C, 381, 382 Fairbridge, R., 27, 53 Fales, F. W., 533, 555 Falzi, M., 425, 441 Fass, E. N., 267, 280 FeU, H. B., 451, 468, 558, 573, 574, 627, 628, 635, 636 Figueroa, W. G., 428, 440 Finch, L. D., 99, 150 Firschein, H., 559, 574, 575, 579, 587, 588, 601, 604, 607 Firschein, H. E., 451, 468, 579, 588 Fischer, E. E., 220, 241, 259 Fischer, P., 32, 52 Fischer, P. H., 56, 59, 60, 65, 88 Fischman, D. A., 527, 530 Fisher, A. K., 285, 296 Fisher, C., 355, 365 Fitzgerald, R. J., 267, 268, 269, 270, 274, 275, 276, 280, 281 Flanagan, B., 557, 570, 571, 574 Folch, J., 190, 210 Foley, G., 270, 282 Folk, B. P., 591, 607 Folk, G. E., Jr., 285, 296 Folk, J. E., 266, 281 Follis," R. H., Jr., 472, 489, 493 Fontaine, R., 547, 555 Forbes, A. P., 428, 442 Forbush, E. H., 340, 365 Forscher, B. K., 560, 574, 577-588, 578, 579, 581, 585, 587, 589, 596, 600, 603, 604, 607, 646 Forsyth, C. C., 350, 365, 428, 434, 444 Fosdick, L. S., 255, 258, 652, 655, 656, 658, 659, 661 Foster, W. G., 433, 443 Fowler, G. H., 354, 365 Francis, M. D., 98, 133, 206, 213- 260, 219, 220, 222, 241, 254, 255, 258, 276, 280, 640, 659 Francois, P., 381,382 Frandsen, A. M., 312, 316 Frank, P., 547, 555 Frank, R., 547, 555 Frank, R. M., 113, 116, 150, 188, 209 Frankenberger, Z., 359, 365 Franklin, D. M., 680, 692 Franklin, M. G., 160, 169 Eraser, R., 386, 387, 427, 441 AUTHOR INDEX 699 Freeman, D. J., 640, 659 Freeman, P. A., 431, 444 French, C. E., 352, 365 Fretter, V., 56, 88 Freundova, D., 360, 365 Frieden, E., 663, 665, 692 Friesell, H. E., 652, 659 Frisbie, H. E., 649, 660 Frost, H. M., 375, 382, 416, 428, 441, 449, 457, 468, 489, 493, 552, 553, 555, 557, 574 Gaillard, P. J., 486, 489, 493, 557, 574, 590, 607, 609, 635 Gallagher, T. F., 417, 441 Gallop, P. M., 572, 574, 679, 680, 685, 692 Galtsoff, P. S., 60, 88 Ganong, W. F., 362, 366 Gardiner, J. S., 3, 4, 5, 23, 25, 52 Gardner, B., 427, 442 Gardner, D. E., 199, 205, 209 Gaskoin, J. S., 354, 365 Geiser, M., 490, 493 Gemmell, G. G., 165, 169 Geoffroy, R., 471, 493 Gerber,'G., 425, 441 Gerber, G. B., 425, 441 Gershon-Cohen, J., 433, 443 Gerstenfeld, S., 591, 608 Gibbons, R. J., 277, 280 Gibbs, J. O., 435, 443 Gifford", E. D., 423, 443 Giglioli, M. E. C., 56, 88 Gigoux, E. E., 349, 366 Gilium, H. L., 423, 443 Ginsburg, R. N., 25, 27, 52 Glegg, R. E., 535, 555 Glickman, I., 298, 307, 308, 311, 316 Glimcher, M. J., 97, 150, 624, 626, 627, 636 Glover, R. P., 435, 443 Godina, G., 376, 382 Goldbanm, L. R., 591, 607 Goldhaber, P., 94, 450, 468, 472, 490, 493, 578, 587, 590, 607, 609-636, 609, 610, 614, 617, 621, 623, 624, 625, 626, 627, 635, 636 Goldman, H. M., 298, 318 Goldsmith, R. S., 428, 442 Gomori, G., 516, 530, 535, 555 Gonzales, F., 501, 503, 505, 513, 528, 530, 645, 659 Gordan, G. S., 312, 318, 387, 427, 430, 435, 441, 442 Gordon, H. A., 267, 270, 282 Gordon, S., 515, 530 Gore, J. T., 219, 258 Goreaii, T. F., 6, 25-54, 27, 35, 50, 52 Gortner, R. A., Jr., 99, 150, 151 Goss, R. J., 95, 97, 339-369 Gottlieb, B., 298, 307, 317 Gould, B. S., 667, 692 Grafflin, A. L., 361, 366 Graham, A., 56, 68, 88 Grahnen, H., 264, 280 Grainger, R. M., 276, 277, 281 Grant, A. B., 161, 169 Grant, R. E., 31, 52 Grant, T., 312, 318 Grauer, A., 639, 660 Gray, J. A., 98, 133, 206, 213-260, 219, 220, 222, 233, 241, 249, 250, 253, 254, 255, 258, 276, 280, 640, 659 Greco, B., 417, 426, 440 Green, N. M., 667, 692 Greenberg, E., 431, 442 Greenblatt, R. B., 428, 443 Greep, R. O., 328, 330, 336, 337, 565, 574, 589, 607 Gregoire, C., 60, 81, 88 Greulich, R. C., 486, 493, 646, 659 Gridley, M. F., 473, 493 Griebstein, W. J., 220, 241, 254, 255, 258, 276, 280, 640, 659 Griffiths, D., 264, 265, 269, 282 Groen, J. J., 424, 442 Gross, J., 572, 574, 663-694, 664, 667, 668, 670, 679, 680, 682, 685, 687, 692, 693, 694 Grosz, S., 354, 355, 368 Gruber, G. B., 340, 341, 366 Gruneberg, H., 328, 337 Gudernatsch. J. F., 663, 692 Gunter, G., 56, 88 700 AUTHOR INDEX Gustafson, G., 172, 173, 175, 176, 177, 178, 181, 183, 186, 205, 209, 215, 218, 225, 244, 248, 258, 648, 653, 658 Gustafsson, B. E., 264, 280 Guzman, C, 173, 183, 186 Haas, H. G., 386, 427, 441, 442 Hadwen, S., 355, 366 Haldi, J., 264, 280 Hale, C. W., 535, 555 Hall, D. M., 167, 168, 169 Hall, J. W., Ill, 424, 442 Hall, T. C, 361, 362, 366, 368 Hall, W. H., 428, 442 Hals, E., 219, 221, 240, 258 Halsted, J. A., 424, 442 Ham, A. W., 515, 530 Hancock, A., 32, 53 Hancox, N. M., 195, 451, 468, 490, 493, 497-514, 497, 501, 503, 505, 511,513,645,646,659 Handelman, C. S., 515-530 Hard, D. G., 278, 280 Hai-kness, R. D., 667, 677, 692 Harris, W. H., 448, 449, 468 Harrison, R. W., 267, 270, 274, 280, 281 Hartles, R. L., 643, 644, 659 Hartley, W. J., 161, 169 Hartman, W". D., 6, 25-54, 27, 30, 53 Hartnett, A., 535, 554 Hashimoto, K., 650, 660 Haumont, S., 379, 383, 557, 575 Haupl, K., 298, 299, 300, 306, 307, 315,316,317 Hausmann, H., 667, 692 Hay, E. D., 527, 530, 688, 692 Ha'vashida, T., 427, 442 Heaney, R. P., 448, 468 Heller, M., 425, 442, 472, 494 Heller-Steinberg, M., 486, 489, 494, 531,551,555,639,660 Helmcke, J. G., 188, 209 Hempelmann, L. H., 425, 441 Henneman, P. H., 428, 442 Herdon, E. G., 386, 443 Herman, H., 381,382 Hertelendv, F., 432, 445 Hess, W. C., 653, 662 Hevesv, G., 140, 150, 372, 382 Hicks,'R. M., 511, 512, 513 Highberger, J. H., 667, 679, 692, 693 Hill, J. T., 117,150 Hill, L. L., 430, 445 Hill, R., 161, 170 Hills, J., 652, 660 Hilming, F., 304, 317 Hioco,b.,435, 443 Hirsch, G. G., 56, 60, 65, 81, 88 Hirsch, P. F., 621 Hitt, W. E., 434, 442 Hodge, A. J., 679, 693 Hodge, H. C., 163, 169 Hodgkin, N. M., 20, 23 Hohling, H. J., 188, 209 Holdewav, E. M., 424, 442 Hollowav, P. J., 99, 150 Holmes,' L. B., 560, 570, 574, 646, 660 Holt, S. J., 511, 512, 513 Hooghwinkel, G. J. M., 535, 555 Hook, W. E., 285, 296 Hopewell-Smith, A., Ill, 127, 151 Hoppert, G. A., 294, 295, 296 Houck, J. G., 680, 693 Huber, L., 168, 170, 547, 555 Hucin, B., 361, 366 Hunt, H. R., 294, 295, 296 Hunter, W. R., 13, 14, 23 Hurst, v., 649, 660 Ibsen, K., 391, 395, 425, 442 Imbrie, J., 4, 23 Irving, J. T., 472, 494, 515-530, 528, 529 Isenberg, H. G., 647, 660 Iseri, O. A., 619, 636 Jackson, R. H., 448, 449, 468, 667. 668,677,679,693 Jacobs, M. H., 188, 209, 219, 258 Jacobson, B., 422, 442 Jaczevvski, Z., 352, 356, 366 Jafte, H. L., 472, 489, 490, 494 James, P. M. G., 99, 151 James, W. W., 297, 317 AUTHOR INDEX 701 Janes, J. M., 552, 555 janofF,A., 628, 635 Jarabak, J., 307, 319 Jay, G. E., 312, 316 Jee, W. S. S., 448, 467, 486, 490, 494 Jenkins, G. N., 114, 116, 206, 214, 215, 259, 637-662, 648, 653, 654, 655, 658, 660 Jephcott, H., 81, 88, 517, 530 Johansen, E., 98, 187-211, 188, 193, 197, 199, 201, 204, 205, 209, 210, 231, 259, 264, 277, Johnson, 433 Johnson, K., 448, 467 Johnston, C. C, Jr., 560, 570, 646, 660 Johnston, H. W., 640, 660 Jones, L. H. P., 166, 169 Jordan, H. V., 98, 261-283, 268, 270,276,280,281 Jordan, T., 428, 440 Jowsev, J., 376, 377, 380, 383, 433, 442, 443, 447-469, 448, 451, 452, 455, 457, 468, 469, 484, 494, 557, 570, 574 Judv, F. R., 99, 153 492, 209, 650, 189, 206, 280 574, 269, 416, 449, 472, Kao. K.-Y. T., 434, 442 Kapur, K. K., 220, 241, 259 Karlson, K. E., 115, 116, 151, 214, 259, 648, 649, 653, 654, 661 Karnovskv, M. J., 501, 503, 505, 513, 528, 530, 558, 562, 567, 574, 645, 659 Karolyi, M., 298, 317 Karshan, M., 270, 282 Keil, A., 177, 178, 186, 215, 260 Kellenberger, E., 547, 555 Kelly, M.,^386, 416, 424, 432, 443 Kellv, P. J., 449, 462, 468, 469, 552, 555 Kelsall, J. P., 353, 366 Kemp, N. E., 664, 666, 677, 693 Kenigsberg, R. K., 99, 150 Kennedy, B. J., 424, 442 Kennev^P.,431, 442 Kenny, A. D., 330, 337, 560, 562, 575, 578, 587, 617, 619, 621, 624, 635, 636 Kent, P. W., 448, 468 Kern, R., 116, 150,188,209 Keutmann, H. T., 604, 606 Keyes, P. H., 98, 261-283, 268, 269, 270, 274, 275, 276, 279, 280, 281 Kind, H., 489, 494 Kirk, D., 682, 692 Kirk, E. C, 104, 117, 151 Kite, O. W., 264, 281 Klapper, Z. F., 523, 529 Kleesiek, C, 355, 366 Klein, L., 425, 442 Klein, M., 547, 555 Kolb, F. O., 429, 442 Kolliker, A., 341, 366, 499, 505, 513 Kollros, J. J., 664, 693, 694 Korringa, P., 2, 23 Kostlan, J., 176, 186 Kraintz,F.W., 471,495 Kramer, B., 190, 210 Krasse, B., 264, 280 Kronfeld, R., 105, 112, 151, 307, 317 Kroon, D. B., 486, 487, 489, 494, 499, 513 Kuhlman, R. E., 578, 588 Kuhn, J., 679, 693 Kuhn, K., 679, 693 Kiihnelt, W., 20, 23 Knrahashi, Y., 188, 210 Kuzell, W. C, 435, 443 Lacroix, P., 557, 570, 575 Ladd, H. S., 50, 52 Laffre, R. O., 217, 259 Laidler, K. J., 233, 252, 259 Lane, K., 94, 589-608, 591, 607 Lang, F. J., 297, 298, 300, 307, 317 Lanham, C., 435, 443 Lapiere, C. M., 572, 574, 663-694, 664, 680, 682, 685, 687, 692, 693 Larson, R. H., 264, 269, 281, 283 Laskin, D. M., 486, 489, 494, 579, 581, 588, 589, 593, 596, 603, 604, 607 Langhlin, J. S., 431, 442 Leach, S. A., 640, 656, 660 702 Leaver, A. G., 643, 644, 659 Lebedinsky, N. G., 361, 366 Leblond, C. P., 535, 555 Leicester, H. M., 161, 169 Lekan, E. C., 579, 588, 589, 596, 604, 607 Lent, P. C., 353, 366 Lenz, H., 188, 210 Leslie, I., 591, 607 Letellier, A., 34, 53 Lichtovitz, A., 435, 443 Lieberkiihn, N., 341, 366 Lieberman, J. E., 312, 315, 316 Likins, R. C.", 116, 151, 667, 693 Lilienthal, J. L., Jr., 591, 607 Lindahl, O., 393; 418, 423, 443 Lindgren, A. G. H., 393, 418, 423, 443 Lindquist, B., 377, 382, 393, 413, 416, 424, 430, 431, 440, 448, 467 Lipp, W., 489, 494, 511, 513, 551, 553, 555 Lippincott, E. R., 381, 383 Little, K., 183, 186, 205, 210, 386, 416, 424, 432, 442, 443 Llovd, F. E., 10, 11,23 Locklev, R. M., 340, 367 Loe, H., 214, 259 Loken, H. F., 427, 442 Lombard, L. S., 376, 383 Long, T. A., 352, 365 Lotz, W. E., 472, 494, 495 Lovestedt, S. T., 99, 153 Lowenstam, H. A., 35, 53 Lowther, D. A., 667, 692 Lucena-Conde, F., 233, 259 Lucy, J. A., 628, 635, 636 Lundquist, C., 264, 280 Lydekker, R., 347, 348, 351, 367 Lynn, W. G., 664, 693 Macapanpan, L. C., 308, 317 Macdonald, J. B., 267, 281 MacDonald,"N. S., 95, 385-446, 395, 409, 413, 416, 418, 430, 431, 433, 443, 445, 448, 469 Macewen,W., 341,367 MacFavden, D. A., 190, 210 AUTHOR INDEX MacGregor, J., 427, 428, 435, 440, 444 Magitot, E., 213, 217, 259 Magnusson, B., 267, 281 Magruder, N. D., 352, 365 Maisto, O. A., 526, 530 Majno, G., 551, 552, 553, 556 Malm, O. J., 426, 443 Mandel, I. D., 649, 660 Mandiangu, L., 547, 556 Mandl, 1., 639, 660, 664, 679, 680, 693 Manfredi, E. E., 528, 530 Manlv, R. S., 163, 169, 188, 210, 220, 241, 259 Manner, G., 667, 692 Mannig, K., 679, 693 Manson, R., 312, 318 Marinozzi, V., 490, 492 Marko, A. M., 667, 692 Marmorston, J., 416, 426, 434, 435, 444 Marshall, F. H. A., 351, 352, 364, 367 Marshall, J. H., 376, 377, 380, 383, 448, 451, 452, 455, 468, 469 Marshall, M. S., 649, 660 Martin, G., 559, 574, 579, 587, 601, 604, 607 Martin, G. N., Jr., 4, 55-89, 647 Martin, G. R., 451, 468, 469, 579, 588, 623, 625, 636 Martin, J. J., 115, 116, 151, 214, 259, 275, 282, 647, 648, 649, 653, 654, 660, 661 Martinez, O. M., 428, 443 Mason, A. D., 428, 442 Massey, V., 594, 607 Masson, P., 535, 555 Mayer, W. V., 94, 285-296, 285, 286, 294, 296 Mayfield, J. D., 419, 443 Mayo, K. M., 422, 443 Mavr, E., 262, 263, 281 Mazet, R., 431, 445 McCann, H. G., 190, 210 McGav, G. M., 99, 150, 151 McClendon, J. F., 433, 443 McGlure, F. J., 99, 114, 115, 151, 153, 266, 281 AUTHOR INDEX 703 McCluskey, R. T., 628. 635 McEwan, E. H., 353, 367 McEwen, L. C, 352, 365 McGavack, T. H., 434, 442 McGregor, J., 422, 444 McLean, F. C, 371-383, 376, 379, 380, 383, 425, 442, 448, 467, 471, 472, 494, 505, 513, 557, 575, 627, 629, 636 McLeod, I. M., 276, 277, 281 McMillan, L., 655, 658 McPherson, G. D., 533, 554 McRoberts, M. R., 161, 170 Mearns, E. A., 355, 367 Mecca, C. E., 623, 625, 636 Mechanic, J., 624, 626, 627, 636 Meckel, A. H., 219, 222, 258, 259 Megregian, S., 190, 210 Meilman, E., 572, 574, 679, 680, 685, 692 Mellanby, E., 451, 468, 627, 628, 635 Mellanby, M., 99, 150 Mellinger, R. C., 419, 444 Mensen, E. D., 533, 554 Merea, C., 527, 530 Mermagen, H., 173, 183, 186, 219, 259 Meroney, W. H., 386, 443 Merrick, J. V., 262, 282 Meyer, R. J., 427, 442 Middleton, J. D., 277, 281 Migicovskv," B. B., 531-556, 531, 539, 543, 554, 556, 639 Milch, R. A., 448, 468 Miles, A. A., 274, 282 Miller, C. D., 99, 151 Miller, J., 173, 186 Miller, S. C., 305, 318 Miller, W. D., 101, 116, 151, 213, 217, 259, 649, 660 Millonig, R. G., 400, 441, 446 Miravet, L., 435, 443 Moghadam, H., 533, 554 Mohammed, G. I., 328, 330, 336, 337 Mohr, E., 347, 348, 367 Moldawer, M., 419, 443 Moon, N. F., 385, 387, 418, 424, 431, 443, 445 Moore, B. W., 654, 656, 658, 661 Moore, F. J., 416, 426, 434, 444 Morch, T., 219, 221, 240, 258, 266, 281 Morgan, A. F., 423, 443 Morrione, T. G., 664, 680, 693 Morris, T. R., 432, 445 Morrison, J. P. E., 4, 23 Morse, A., 108, 151, 189, 210 Mortimer, K. V., 175, 179, 184, 186 Moss, M. J., 95, 385-446 Movvry, R. W., 535, 555 Miihlemann, H. R., 219, 259 Muhlethaler, J. P., 490, 494 Muir, H. N., 667, 677, 692 Mulryan, B. J., 451, 468, 469, 559, 574, 575, 579, 587, 588, 601, 604, 607 Munson, P. L., 618, 619, 621, 636 Murphy, A. P., 229 Murray, R. O., 428, 443 M) ers, W. P., 428, 443 Nachlas, M. M., 517, 530 Nagai, Y., 685, 693 Naibandov, A. V., 489, 490, 492 Nassonov, N., 32, 33, 53 Nassonow, N., 27, 31, 32, 33, 34, 35, 53 National Research Gouncil (U. S.), 161, 170 Neal, R. J., 229 NesterofF, W. D., 50, 53 Neuberg, G., 639, 660 Neuberger, A., 667, 677, 692 Neuman, M. W., 94, 206, 210, 379, 383, 559, 575, 584, 588, 589-608, 589, 607, 637, 640, 644, 661 Neuman, W. F., 94, 206, 210, 379, 383, 451, 468, 469, 558, 559, 560, 574, 575, 579, 584, 587, 588, 589- 608, 589, 601, 603, 604, 605, 606, 607, 630, 636, 637, 640, 644, 661, 667, 692 Nevvbrun, E., 219, 259 Newell, N. D., 3, 4, 23 Newhouse, J., 3, 23 Nicholas, J. A., 417, 435, 440 704 AUTHOR INDEX Nichols, G. Jr., 487, 495, .557-575, 558, 559, 560, 561, 562, 563, 564, 565, 566, 567, 569, 570, 574, 575, 578, 579, 580, 581, 584, 587, 588, 589, 593, 596, 598, 603, 604, 606, 608, 640, 641, 643, 658, 661, 662 Nichols, N., 557, 558, 559, 560, 574, 579, 587, 589, 593, 596, 598, 603, 604, 606, 640, 658 Nickerson, G., 278, 280 Nicolavsen, R., 426, 433, 443 Nikiforuk, G., 276, 277, 281 Nisbet, J., 422, 444 Nishimura, T., 178, 181, 186, 215, 259 Nizel, A. E., 278, 281 Noback, C. V., 340, 367 Nolan, P., 490, 494 Nordback, L. G., 199, 204, 205, 209 Nordin, B. E. C., 417, 422, 426, 427, 428, 431, 435, 440, 443, 444, 569, 575, 643, 661 Norris, W. P., 376, 383 Novikoff, A. B.,511,513 Nuckolls, J., 649, 660 Nvlen, M.U.,65, 88 Odum, E. P., 3, 24 Odum, H. T., 3, 24 Oken, D. E., 679, 693 Old, M. C., 32, 34, 53 Oliver, R., 448, 468 Ollis, W. D., 179, 186 Olt, A., 355, 367 Opdvke, D. L. J., 219, 233, 258, 259 Orban, B., 307, 308, 317, 319 Orban, B. J., 214, 233, 259 Orland, F. J., 267, 270, 274, 281 Otter, G. W., 5, 12, 14, 15, 19, 21, 24, 25, 53 Owen, M., 449, 468, 484, 494 Palade, G., 499, 513 Palladino, V. S., 262, 270, 282 Palmer, L. J., 355, 366 Parfitt, G. J., 99, 151 Parks, H. F., 189, 193, 205, 206, 209, 210 Parlier, R., 435, 443 Patol, Y. M., 680, 693 Pazianos, A., 431, 442 Pearce, L., 328, 337 Pearse, A. G. E., 473, 494 Pearson, O. H., 431, 442 Pease, D. C., 455, 469, 499, 501, 503, 505, 514, 528, 530, 645, 661 Pedersen, P. O., 304, 317 Peirce, A. W., 165, 170 Perloff, A., 381, 383 Person, P., 312, 317 Peterkofskv, B., 667, 693 Peters, M.,' 101 Peterson, L. F. A., 449, 462, 468, 469, 552, 555 Peterson, R. R., 416, 418, 423, 445 Petrik, L., 307, 315 Phillips, R. W., 138, 153 Phillips, W. W. A., 348, 350, 367 Pickerill, H. P., 217, 221, 259 Piez, K. A., 116, 151, 667, 693 Pigman, W., 217, 259 Pincus, P., 649, 661 Plonlot, R., 449, 469 Plum, G. E., 462, 468 Pocock, R. L, 347, 350, 355, 367 Pommer, G., 298, 317, 318 Poole, D. F. G., 175, 179, 183, 186 Posner, A. S., 264, 276, 283, 381, 383 Powers, J. F., 428, 442 Prat, L., 233, 259 Prockop, D. J., 667, 693 Prophet,A. S., 649, 651, 659 Psansky, R., 299, 306, 307, 317 Pu^h, M. H., 65, 88 Purchon, R. D., 10, 14, 24 Pycraft, W. P., 351, 367 Quensel, C. E., 264, 280 Quintarelli, G., 305, 318 Racker, E., 598, 607 Rader, T., 352, 365 Raisz, L. G., 563, 575, 579, 580, 584, 585, 588, 590, 608, 639, 641, 661 Rail, D. P., 269, 283, 448, 468 Ramfjord, S. P., 308, 318 Ratcliffe, H. L., 262, 270, 282 Raveni, G., 417, 425, 444 AUTHOR INDEX 705 Ravnik, C, 214, 259 Rawles, M. E., 340, 367 Recklinghausen, F. von, 300, 318 Reichborn-Kjennerud, I., 94, 96, 279- 319, 306, 307, 310, 318 Reifenstein, E. C, Jr., 312, 315, 604, 606 Reitan, E., 472, 490, 495 Restarski, J. S., 99, 150, 151 Revel, J. P., 688, 692 Revelle, R., 27, 53 Reyniers, J. A., 267, 270, 281 Rheinwald, V., 140, 151 Riccitelli, M. L., 435, 444 Rich, C, 435, 444 Richardson, R. L., 285, 296 Richelle, L., 381, 382 Ries, J., 340, 367 Riley, M. J., 591, 607 Robb-Smith, A. H. T., 688, 693 Robichon, J., 531-556, 535, 554, 639 Robinson, B. H. B., 429, 446 Robinson, H. B.C., 114, 152 Robinson, J. R., 581,588 Robinson, R. A., 416, 444, 455, 467, 501, 503, 512, 645, 658 Rochenmacher, M., 190, 210 Rogers, H. J., 472, 495 Romeis, B., 473, 495 Rorig, A., 354, 367 Rosa, C. G., 517, 530 Rosch, P. J., 386, 443 Rose, G. G., 511, 513 Rosebury, T., 270, 282 Rosen, S., 294, 295, 296 Rosenberg, H. S., 430, 445 Rosenblum, E. D., 680, 694 Rost, T., 100, 151 Roth, G., 649, 661 Roth, H., 375, 382, 489, 493, 557, 574 Roth, M. L., 434, 441 Rouiller, C., 168, 170, 547, 555, 664, 693 Roux, W., 128, 151, 299, 318 Rowland, R. E., 371-383, 376, 377, 380, 383, 448, 451, 452, 455, 468, 469 Rowles, S. L., 184, 186 Rubegni, M., 417, 425, 444 Rubin, M., 264, 281 Rubini, M. E., 386, 443 Rushton, B., 145, 151 Russell, A. L., 265, 282 Rutenburg, A. M., 517, 530 Ruth, E. B., 531, 552, 555, 556 Rutishauser, E., 168, 170, 533, 547, 551, 552, 555, 556 Ruzicka, S. J., 99, 115, 151 Sabin, F. R., 528, 530 Saifer, A., 591, 608 Saito, T., 651, 661 Sand, H. F., 219, 221, 240, 258 Sarnat, B. G., 285, 296 Saville, P. D., 417, 435, 440 Schaaf, J., 422, 446 Schaaf, M., 427, 442 Schaffer, J., 128, 151 Schajowicz, F., 527, 530, 614, 636 Schartum, S., 557, 560, 561, 562, 563, 564, 565, 567, 569, 575, 579, 580, 584, 588, 589, 608, 643, 661 Schatz, A., 115, 116, 151, 214, 259, 275, 282, 647, 648, 649, 650, 651, 654, 658, 660, 661 Schatz, v., 115, 116, 151, 647, 648, 653, 654, 660, 661 Scherbatoff, N., 340, 367 Scherp, H. W., 214, 258, 275, 280 Schiemenz, P., 56, 65, 88 Schlack, C. M., 99, 150 Schlager,F., 511,513,514 Schlesinger, B. E., 428, 434, 444 Schmidt, W. J., 177, 186, 215, 260 Schmitt, F. O., 667, 679, 692, 693 Schour, I., 328, 330, 336, 337, 472, 495, 528, 530 Schwarzenbach, G., 190, 209 Scott, B. L., 455, 469, 499, 501, 503, 505, 513, 528, 530, 645, 661 Scott, D. B., 4, 55-S,9, 65, 88, 188, 210, 647 Seidler, B., 305, 318 Seifter, S., 572, 574, 664, 679, 680, 685, 692, 693 706 AUTHOR INDEX Selakovitcli, W. G., 395, 416, 423, 444, 552, 556 Seligman, A. M., 517, 530 Selye, H., 552, 556 Sevmour Sewell, R. B., 6, 7, 24 de Seze, S., 435, 443 Shapiro, M., 311, 318 Sharp, D. C, 68, 73, 89 Shaw, J. H., 141, 152, 249, 260, 264, 265, 266, 269, 281, 282 Sheps, M. C, 619, 636 Sheridan, R.C., Jr., 311, 318 Sherman, M. S.", 395, 416, 423, 444, 552, 556 Sherry, S., 680, 694 Shkla'r, G., 312, 318 Shuhnan, E. H., 114, 152 Siegmnnd, P., 646, 659, 661 Silberberg, R., 433, 444 Siller, W. G., 432, 440 Silverman, S., 312, 318 Simasaki, M., 511, 514 Singer, M., 360, 368 Singh, B., 197, 210 Singleton, E. B., 430, 444, 445 Sissons, H. A., 423, 444, 451, 468 Skinner, M. P., 340, 368 Skoog, W. A., 95, 385-446, 395, 416, 418, 430, 445, 448, 469 Sluiter, C. Ph., 15, 24 Smellie, J. M., 428, 434, 444 Smith, F. A., 199, 205, 209 Smith, L.H., 591,607 Smith, M. R., 434, 441 Smith, R. W., 419, 444 Smith, S. I., 25, 53 Smits, G., 535, 555 Smulow, J. B., 307, 308, 316 Sobel, A.E., 190,210 Sobel, H., 416, 426, 434, 435, 444 Socranskv, S. S., 277, 280 Sognnaes, R. F., 91-153, 92, 95, 96, 97, 98, 107, 109, 111, 113, 116, 118, 120, 126, 131, 133, 141, 142, 143, 145, 146, 150, 152, 188, 205, 209, 210, 249, 260, 264, 281, 472, 486, 489, 495 452, 413. Soni, N. N., 219, 248, 260 Specter, W. S., 593, 608 Spencer, H., 417, 441 Spiro, D., 499, 501, 503, 505, 513 Stack, M. v., 184, 186, 188, 210, 648, 649, 651, 661 Staehle, G., 140, 151 Stafne, E. G., 99, 153, 311, 316, 318 Stafseth, H. J., 294, 296 Stanburv, J. "B.,431,441 Stanley,' H". R., 268, 270, 280 Starke, A. G., 255, 258, 652, 659 Stauffer, J. P., 591, 598, 608 Steadman, L. T., 201, 205, 210 Steadman, S. G., 201, 205, 210 Steddon, L. M., 448, 468 Steel, J., 116, 153 Steinbach, H., 312, 318 Stephan, R. M., 264, 282 Stern, B., 624, 626, 627, 636 Stevens, J., 431, 444 Stewart, R. J. G., 99, 150 Stieve, H., 359, 368 Storey, E., 434, 444, 489, 490, 495 Stralfors, A., 278, 282 Strates, B., 559, 579, 587, 601, 604, 607 Stroud, G. E., 428, 434, 444 Stiiben, J., 651, 662 Stutman", J. M., 381, 383 Sullivan, H. R., 221, 260, 652, 660 Smnmer, G. K., 391, 445 Summerson, W. H., 591, 606 Swartz, M. L., 138, 153 Swenander Lanke, L., 264, 280 Swift, R. W., 352, 365 Tachezv, R., 358, 362, 368 Taft, E; B., 361, 362, 366, 368 Takuma, S., 188, 210, 486, 495, 528, 530 Talmage, R. V., 471, 472, 489, 493, 494,^495, 531, 553, 556, 570, 575, 589, 607, 639, 662 Tandler, J., 354, 355, 368 Tanzer, M. L., 670, 680, 692, 694 Tashjian, A. J., 621 Taylor, A. G., 432, 445, 664, 694 AUTHOR INDEX 707 Teng, C. T., 430, 445 Tepperman, J., 563, 575, 579, 580, 584, 585, 588, 590, 608, 639, 641, 661 Terboigh, A., 269, 282 Thoma, K. H., 298, 318 Tillotson, J. F., 307. 318 Tobie, J. E., 448, 468 Toepel, W., 330, 337 Toft, R. J., 471, 495, 570, 575 Tomes, J., 371, 383 Topley, W. W. C, 274, 282 Topsent, E., 32, 33, 53 Toribara, T. Y., 190, 209, 591, 606 Tracev, J. I., 50, 52 Trelawnv, G. S., 647, 660 Trescher, M. O., 652, 659 Trexler, P. C, 267, 270, 281 Troll, W., 680, 694 Trotter, M., 416, 418, 423, 445 Trueta, J., 490, 493 Tsou, K. C, 517, 530 Tsujii, T., 73, 89 Turner, H. J., 56, 89 Tuttle, S. G., 428, 440 Udenfriend, S., 667, 693 Umbgrove, J. H. F., 5, 24 Umbreit, W. W., 591, 598, 608 Urist, M. R., 95, 379, 383, 385-446, 387, 391, 395, 413, 415, 416, 417, 418, 419, 424, 425, 429, 430, 431, 432, 433, 434, 435, 442, 443, 445, 446, 448, 451, 469, 471, 489, 490, 494, 495, 505, 513, 557, 575, 629, 636 Usuku, G., 688, 694 Utseumv, F., 3, 24 Uyeno, S., 391, 395, 425, 442 Vaes, G. M., 489, 495, 557, 563, 565, 566, 575, 578, 581, 584, 588, 589, 608, 641, 662 \'alera, R. B., 348, 351, 368 \'anderhoft, P. J., 449, 469 Van Reen, R., 578, 588 Van Rijssel, Th. G., 511, 513 VanSlvke, D. D., 190, 210 Van Wazer, J. R., 253, 254, 260 Vaughan, J., 448, 449, 451, 452, 468, 484, 494 \^errill, A. E., 25, 53 Mllanueva, A. R., 375, 382, 449, 468, 489, 493, 557, 574 Vincent, J., 375, 379, 383, 5.3i-556, 531, 543, 547, 556, 557, 575, 639 Vincent, P. J., 387, 417, 419, 429, 433, 435,445,446 \'olker, J. F., 264, 282 von Bartheld, F., 219, 260 Vose, G. P., 442, 446 Vosmaer, G. C. J., 29, 33, 35, 53 Wachowski, H. E., 664, 693 Wachtel, L. W., 219, 260, 277, 279 Wagner, A., 422, 446 Wagner, M., 267, 270, 282 Wainvvright, S. A., 3, 24 Waldo, C. M., 340, 341, 343, 344, 354, 355, 358, 360, 362, 364, 368 Walker, D. G., 511, 514, 517, 530 Walser, M., 429, 446 Wandelt, S., 651, 662 Wang-Norderud, R., 305, 318 Warburton, F. E., 31, 32, 34, 54 Wardrop, I. D., 166, 169 Warner, H., 190, 209, 591, 606 Warner, S., 497, 511 Warren, O., 188, 208 \A^atabe, N., 68, 73, 89 Watson, M.L., 189,211 Watt, G., 299, 319 Weatherill, J. A., 423, 440 Webb, M., 558, 573, 574, 628, 635 Wedl, C., 128, 129, 130, 153 Weidman, S. M., 423, 440, 472, 495 Weinmann, J. P., 308, 317, 328, 330, 336, 337, 472, 495, 528, 530 W^elcher, F. J., 233, 260 Wentz, F. M., 307, 319 Werner, H., 129, 153 Weski, O., 298, 319 West, E. S., 99, 153 Westcott, W. L., 400, 441, 446 Wheatcroft, M. G., 277, 279 Whedon, G. D., 448, 468 708 White, A. A., 653, 662 White, B. L., 419, 443 White, C. L., 265, 269, 276, 281, 282 White, R. H. R., 428, 434, 444 Wilbur, K. M., 60, 68, 73, 89 Wilcox, E. B., 423, 443 Will, L., 99, 151 Willard, H. H., 190,211 Willett, N. P., 294, 296 Williams, C. H. M., 299, 319 Willighas^en, R. J. C, 511, 513 Wilson, G. S., 274, 282 Wilson, P. D., 417, 435, 440 Wilson, P. D., Jr., 417, 435, 440 Wimer, L. T., 471, 495 Winand,L., 381,382 Winberg, J., 551, 555 Winter, O.B., 190,211 Wislocki, G. B., 341, 343, 344, 354, 355, 358, 359, 360, 361, 362, 364, 368 Woessner, J. F., 664, 680, 694 Wolbach,S.B. ,31 1.316 Wolf, W., 305, 318 Wollman, D. H., 266, 282 Wood Jones, F., 347, 368 Woods, J., 557, 570, 572 Woods, k. R., 471, 495, 639, 662 AUTHOR INDEX Wu, R., 598, 607 Wyckoff, R. W. G., 65, 88 Wyman, J.,341,369 Wynn, C. H., 680, 692 Wynn, W., 264, 280 Yagi, T., 511, 514 Yepes, J., 349, 350, 365 Yerganian, G., 312, 318 Yonge, G. M., 1-24, 8, 9, 12, 13, 16, 17, 18, 19, 20, 21, 24, 25, 54 Young, R. W., 471-496, 475, 479, 481, 483, 484, 487, 488, 489, 490, 49L 495, 496 Zaccalini, P. S., 395, 413, 416, 418, 430, 445, 448, 469 Zans, V. A., 35, 54 Zavvadowsky, M. M., 354, 356, 369 Zawisch-Ossenitz, G., 531, 556 Zetterstrom, R., 472, 488, 493, 551, 555 Ziegelmeier, E., 56, 59, 89 Zierler, K. L., 591, 607 Zipkin, I., 114, 153, 264, 269, 276, 281, 283 Zuckerman, S., 348, 369 Subject Index Terms have been indexed according to American usage only, e.g., dentin for dentine, mollusk for mollusc, and the like. ABO (accessory boring organ), 56, 79,86 see also under Boring; Deminerali- zation; Shell etching acid secretion, 4 contact with shells, 63 demineralizing activity, 67 diameter of gland, 56 etching, 62, 83, 86 hemostatic pressure, 57 secretory cap, 67 secretory disk, 68, 83 secretory globules, 68 ABO activity aragonite, 77 calcite, 77 shells, 69, 75 ABO etching see also under Shell etching calcareous substrates, 62 organic substrates, 62 ABO homogenate, 60, 86 demineralizing action, 68 pH, 65 shell dissolution, 64 ABO secretion, 60, 69, 75, 77, 85 absence of acid, 81 calcase, 60 calcium phosphate dissolution, 79 conchiolinase, 60 pH, 60, 68, 86 shell boring, 56, 65 Abrasion, 99, 155 in vitro, 108 in vivo, 108 sheep's teeth, 166 Abrasive siliceous materials, 166 Abscesses jawbone, 95, 291 periapical, 639 pulpal, 294 Accelerated aging, osteoporosis, 417 Accelerated resorption, 488 Accessory boring organ, see ABO Accretion bone mineral, 377, 411, 538 haversian systems, 377 Acellular bone, 376 Acellular osteons, 376 Acetate, erosion effect, 167 Acetic acid, caries in vitro, 219 Acetyl coenzyme A, parathyroid ac- tion, 586 Achijla, 3, 5 Acid, absence in ABO secretion, 81 Acid action, inhibition bv organic films, 117 Acid-containing secretion, 20 Acid decalcification, bone resorption, 505 Acid-etched shells, 75 Acid-hydroxyapatite reaction, 249 Acid mechanism, calcium, 570 Acid mucopolysaccharides, 535 aging, 426 avian bone, 536 Acid-organic polymer systems, enamel dissolution, 117, 241 Acid pH range, collagenase, 680 709 710 Acid pliosphiitase giant cells, 520, 527 osteoclasts, 511 \'itamin A effect, 628 Acid production algae, 3 calcium mobili/.atioii. 560 Acid secretion ABO, 4 boring mechanism, 20, 22 Lithophaga, 21 Acid soil, 130 Acidic beverages, 91, 99, 100 Acidic forms of phosphate, 257 Acidity, dental erosion, 113 Acidogenesis, 261 Acidogenic bacteria, hamsters, 274 Acidogenic filaments, 270 Acidogenic microbiota, 272 Acidogenic potential, food substrates, 266 Acids dissolution of shell, 55 Krebs cycle, 166 metabolism, 3 plants, 167 Acromegaly, 426, 430 Acropora, 42 ACTH, see Adrenal cortical hormones Activators bone formation, 299 intermittent masticatory impulses, 306 Addison's disease, 428 Adenosine triphosphate, see ATP Adesmacea, 10 Adesmacean rock borers, 11 Adhesiveness, foodstuffs, 267 Adrenal cortex, calcium metabolism, 429 Adrenal cortical hormones antler growth, 361 joint pains, 399 osteoporosis, 428, 432 Adrenal glands autopsy subjects, 429 laying hens, 432 SUBJECT INDEX Adrenalectomy citrate level, 429 mineral metabolism, 428 osteoporosis, 429 Adreno-orchiectomized rats, 433 Aerobic glycolysis, bone, 603 Agar-saliva svstem, 222 Agaricia, 26,27, 42, 43 communities, 50 cucuUata, 49 fragilis, 49 tindata, 40, 41, 49, 52 Age influences, antler shedding, 353 Aging, 91 accelerated, 417 acute bone changes, 416 antiosteoporosis factor, 434 bone density, 418, 459 bone mass, 417 collapsed vertebrae, 45 endocrinopathy, 438 lacunar mineralization, 552 mucopoh'saccharides in bone, 426 osteoporosis, 386, 467 Alaska ground squirrels, 286 Albino rat, ia mutation, see ia rat Aloes olces, 347 Alcian blne-PAS stain, 535 Algae, 1, 4, 25, 42 acid production, 3 boring, 2, 3, 26, 27, 33, 55 buried teeth, 129 cementation, 47 growth pressin-e, 3 growth processes, 3, 5 shell-boring, 55 Alimentary canal microflora cariogenic components, 269 mother animals, 269 Alkaline earths, 452 Alkaline phosphatase, large osteocvtes, 553 Alpha track autoradiography, 372, 374 Alpharadiography, 531, 535, 538, 543 norethandrolone-treated chicks, 539 rickets, 541 Aluminum, carious enamel, 199 SUBJECT INDEX 711 Alveolar atrophy, osteoporosis, 424 Alveolar bone, 93 aplasia, 307 atrophy, 298 blood circulation, 298 circulatory disturbances, 298 compensatory production, 310 degeneration, 295 functional stress, 299, 306, 307 hibernation effect on, 287 mechanical stress, 306 nutritional canals, 309 orthodontic treatment, 307 production, 305 repair, 303 trigger mechanism, 299 Alveolar bone destruction sec also Alveolar bone resorption circulatorv disturbances, 310 inherited factors, 312 periodontal conditions, 310 systemic condition, 310 theories, 297 Alveolar bone factor, 298 Alveolar bone marrow, fibrosis, 301 Alveolar bone production, 302, 303 flow of blood, 315 Alveolar bone remodeling, 94, 95 orthodontic activators, 306 Alveolar bone resorption, 94, 145 see also Alveolar bone destruction anemic gingiva, 305 circulatorv disturbances, 300, 310 orthodontic treatment, 314 percussion dullness, 310 periodontal disorders, 297, 313 summary, 315 Alveolar bone septum, resorption, 300 American Chicle Company, 208 American Indians, ancient teeth, 121, 127 Amino acids, 166 Amorphous calcium deposits, 416 Amorphous plugs, blood vessels, 422 Anabolic steroids, bone metabolism, 428 Anaerobic glycolysis, 68 Ancient teeth see also Exhumed teeth; Postmor- tem Central America, 120, 127 Egvpt, 120, 126 Greece, 120, 123, 126, 127 Iceland, 120,127 Norwav, 120, 121, 127 Palestine, 120, 121, 127 Androgens, antler growth, 361 Anectine, 356 Anemic gingiva, 306 alveolar bone resorption, 305 Animal caries conditions of captivity, 287 experimental model, 270 feces inoculation, 269 gnotobiotic technics, 270 infected offspring, 269 therapeutic implications, 277 Animal Research Institute, Canada Department of Agriculture, 554 Ankylosis, tooth germ, 328 Annual rhvthm, antler shedding, 348 Antecedent cellular elements, osteo- clasts, 96, 144, 146 Antelope castration, 355 horns, keratinized sheath, 339, 347 Antibacterial drugs, anticaries therapy, 278 Antibiotics, cariogenic flora, 26S Anticaries therapv antibacterial drugs, 278 chemotherapeutic agents, 278 Antienzymatic effect, fluoride, 276 Antilocapra antciicana, 347 Antiosteoporosis factor, 429, 436, 438 aging, 434 genetic constitution, 439 Antipatharia, 39, 40, 42, 48, 49 Antler abnormal forms, 354, 355 aphrodisiac properties, 340 bisexual occurrence, 355 deciduous nature, 339 histogenesis, 344 712 Antler — Continued hormonal influences, 360; see also under Antler growth; Antler shedding maturation, 348 pedicle, see Antler pedicle resorption, 97 status symbol, 339 successive sets, 348 Antler cycle, 97, 348 bionegative phase, 340 biopositive phase, 340 endocrinological factors, 351 environmental factors, 346 reproductive periods, 363 Antler death, 341 Antler growth, 361 adrenal cortical hormones, 361 androgens, 361 cortisone, 361 endocrine factors, 360 gonadectomies, 360 gonads, 355 growth hormone, 363 hypophysectomy, 362 hypothyroidism, 361 lactogenic hormone, 354 mineralocorticoids, 361 progesterone, 362 rate, 339 thyroid gland, 360 thyroidectomy, 361 thyroxin, 361 zona glomerulosa, 361 Antler pedicle, 339, 343 fibrocellular connective tissue, 344 skin, circumcision, 343, 344 Antler shedding, 95, 97, 339, 341 age influences, 353 annual rhythm, 348 bleeding, 344 castration, 341, 354, 357; see also Castrated deer control, 346, 353 cortisone, 357, 362 day-length changes, 353, 354 denervated antlers, 360 endocrine factors, 354 SUBJECT INDEX Enovid, 357 environmental factors, 346 estrogen, 354, 357 gonadotropins, 358 growth pressure, 346 hormones, 353, 354, 357 hyperemia, 341 internal influences, 353 lactation, 354 mechanism, 340 neurovascular influences, 343 norethynodrel, 357 parathyroid gland, 361 parturition, 354 pituitary hormone, 358 pregnancy, 354 sex hormones, 354 sex influence, 353 spermatogenesis, 359 summary, 363 testosterone, 354, 357 Antler-stimulating hormone, pituitary gland, 362 Antler velvet shedding, 341, 348 castration, 349, 354 testosterone, 356 Antuitrin, 359 Apatite crystals, 166 carbonate, 203 denuded collagen, 505 lactate, 640 lattice, 256 organic coating, 248 stable phase, 256 surface, 251 Apatite dissolution intracellular mechanism, 645 protective surface deposits, 257 uptake within osteoclasts, 645 Aplasia, alveolar bone, 307 Appositional pattern, jaw bone, 325 Aragonite, 62, 75, 83, 86 ABO gland activity, 77 electron micrography, 83 needles in corals, 46 shell surface, 75 Aragonitic mollusk shells, 71 Aragonitic nacreous shell, 73 SUBJECT INDEX 713 Arctic Aeromedical Laboratory, 296 Arctic ground squirrel, 286 Argyrophilic fibers, cell proliferation, '487 Arteries, interalveolar, 300 Articulating processes, shell valves, 10 Artifact, bone crystals, 503 Artificial etchings, shells, 64 Artificial mouth, caries, 217 Ascites tumor, glycolysis, 598 Asparagus stone, 256 Aspartic acid, 167 Aspido siphon, 3 Assay techniques, collagen, 684 Atoll reefs, erosion, 25; see also Coral reefs Atomic Energy Commission, 439 ATP, 639 bone resorption, 633 dental plaque, 655 Atrophy, alveolar bone, 298 Attrition, dental, 92, 93 Attrition of sheep's teeth, 155 herbage, 162 nutritional factors, 160, 165 pasture, 157 summary, 168 Australia, fluorosed sheep's teeth, 165 Autogenous implants, 516 Autopsy subjects adrenal glands, 429 bone mass, 418 Autoradiographv, 372 alpha track, 372, 374 bone resorption, 451, 474 Avalanches, coral reefs, 47 Avian bone, acid mucopolysaccha- rides, 536 Avian osteoclast, micrography, 498 Avian osteoporosis, 429, 432 Axis deer antlers, 349, 350 castration, 355 Ayerst Laboratories Inc., 440 Azurophilia, 536 Back-scattering electron diffraction, 256 Bacteria see also Microorganisms caries, 91 dental erosion, 109 inorganic ions, 277 interaction with diet, 272 shell-boring, 55 Bacterial collagenase, 679, 680, 685 lytic process, 682 Bacterial colonization, oral cavity, 267 Bacterial enzymes, 219, 679 Bacterial invasion caries, 99, 109, 171 enamel caries, 171 Bacterial metabolism caries-like lesions, 219 lactic acid, 220 Bacterial plaque, 111 carious lesions, 217 Bacterial specificity, caries, 272 Balance of bone turnover, 447, 449 Ballooned discs, 386, 431 lumbar spine, 389 Barasingha, antlers, 347 Barnacles, 1, 6, 55 bristles, 56 nail-like bodies, 6 setous thoracic appendages, 7 Barnea, 11 Basement lamella collagen, 687, 688 lytic cells, 691 Basophilia, 536 organic matrix, 531 woven bone, 477 Bauer-Carlsson-Lindquist calculations, 392 Beak pelican, 340 puffin, 340 Bedridden patients, bone turnover, 460 Beryllium intoxication, osteocytes, 522 Bicarbonate-carbonic acid system, Henderson-Hasselbach equation, 584 714 SUBJECT INDEX Biochemical studies bone resorption, 557 calvaria, 590 osteol\tic effects, 589 Biologic half life, osteons, 375 Biological calcification, coral reefs, 25 Biological decalcification, h\potheses, 637 Biological life span, collagen, 580 Biology of caries, 261 Bionegative process, antler cycle, 340 Biopositive process, antler cycle, 340 Biopsy bone, 393 iliac crest, 461 palate, 305 Biotic zones, coral reefs, 35 Birds, osteoporosis, 429, 432 Birefringence carious zones, 218, 227 enamel caries, 177, 180 incipient carious lesion, 215 striae of Retzius, 177 Bisexual occurrence of antlers, 355 Biyalye mollusks, sec Biyalyes Bivalve shells, 2, 71 Bivalves, 1, 7, 25, 39, 55, 73, 86 boring, nature and origin, 8 byssal threads, 8 California, 9, 12, 19 ciliary currents, 8 epifaunal habit, 8 foot sucker, 10 glands, 20 Great Barrier Reef, 12 hatchet-shaped foot, 8 infaunal origin, 9 mechanism of attachment, 8 secretion of mucus, 12 suckerlike foot, 14 Bivalvia, sec Bivalves Bizarre resorption Paget's disease, 462 radium-burdened bone, 455 Black Sea ovsters, 31 Black-tailed deer, antlers, 347 Blastocerus dichotomus, 349 Blood and bone calcium transfer, 377 ion transfer, 380 niinute-to-minute ecjuilibrium, 379 Blood calcium, 371, 380 Blood cells, multinucleated, 527 Blood cholesterol, 305 Blood circulation alveolar bone, 298 bone production, 305 Blood flow alveolar bone production, 315 periodontium, 305 Blood plasma, calcium level, 371 lilood pressure, bivalve boring, 8 Blood vessels amorphous phigs, 422 permeability, 300 Bodian's silver stain, 483 Body burden, radium, 378, 448 Bone accretion sites, 538 aerobic glycolysis, 603 alpharadiography, 535 autoradiography, 451 carbohydrate metabolism, 577 cellular specializations, 489 circulatory disturbances, 298 collagenolytic activity, 664, 680 diffuse fraction, 448 electron micrography, 497 functional adaptation, 321 glycogen, 599 glycolysis, 596 glycolytic pathway, 585 hexosamine, 426, 646 in hibernation, 285, 294 inner fluid layer, 568 internal reconstruction, 322 in vitro metabolism, 578 ionic exchange, 372 Krebs cycle, 578, 585, 593, 596 lactate production, 596 micrography, 447, 498 mucoproteins, 425 myristic acid, 424 normal, sec Normal bone old and new, 475 SUBJECT INDEX 715 Bone — Continued organic mass content, 475 Pasteur effect, 596, 601 pH gradient, 584 phase contrast microscopy, 535 physiologic aging, 416, 423 porosity, 455 radium burdening, 448 rarefying diseases, 385 solubility, 637 substrate utilization, 599 TCA cycle, 596 tetracycline, 375, 393 tissue tonus, 298 turnover, 447, 460 x-rav microradiography, 535 Bone aging, capillar\' circulation, 438 Bone apposition, 95, 303 lamina dura, 307 Bone as tissue, graded microenviron- ments, 491 Bone biopsies, 393, 409, 533 parathyroidectomy, 534 Bone and blood, see Blood and bone Bone boring, rock-boring analogy, 372 Bone breakdown, urinary products, 417, 438 Bone catabolism, 425 Bone cells citric acid, 559 coalescence, 489 enyironment, 637 functional states, 489 genetic information, 492 histophysical studies, 471 leucine aminopeptidase, 553 metabolic acids, 637 microenyironment, 491 resorptive potential, 490, 491 summary, 491 Bone citrate calcium mobilization, 644 parathyroid hormone, 559 tetany, 643 Bone collagen degradation, 424 turnover, 425, 580 Bone collagenase, 573 Bone crystals artifact, 503 mechanical shift, 503 Bone cultures citrate, 621 lactate, 631 Bone demineralization, 578 Bone density, 418 aging, 418 Bone destruction see also Bone resorption; Osteolysis cortisone, 428 multinucleated giant cells, 94, 333, 515 osteoclasts, see Osteoclasts theories, 297 yitamin D, 428 Bone diseases, 447 Bone dystrophy, 298 Bone and extracellular fluids, interface, 568 Bone failure, 417, 431 lumbar spine, 389 osteoporosis, 386 Bone formation alveolar process, 302 balance with resorption, 447 blood circulation, 305 immobilization, 460 lead lines, 375 orthodontic activators, 299, 306 retained isotope, 449 tetracycline lines, 375, 448 tetracycline markers, 457 Bone growth, collagen removal, 664 Bone implants see also Implant macrophages, 519 parathyroid glands, 515 summary, 529 Bone junction, osteoclasts, 498 Bone lamellae, crenated surfaces, 453 Bone marrow circulatory disturbances, 300 fibrosis, 301 osteoblasts, 304 . ^ osteoclasts, 304 transudate, 300 716 SUBJECT INDEX Bone mass aging, 417 autopsy subjects, 418 tracer dilution formulas, 430 Bone matrix amino sugar content, 426 chelating agents, 645 electron micrography, 505 galactosamine/'glucosamine ratio, 426 glycine-H^-injected animals, 475 hydroxyproline, 426 parathyroid action, 485 proline metabolism, 425 Bone metabolism anabolic steroids, 428 ia rats, 566 in vitro model, 563 in vitro studies, 557 lactate, 559 parathyroid action, 589 parathyroid extract, 577 radioisotope kinetics, 391 summary, 605 vitamin D, 564 Bone mineral accretion, 377 Caio(P04)o(OH)o, 381 Ca9(P04)6Ho(OH)o, 381 Cag ( PO4 ) 4 ( OH ) ( O3PO— H— OP03),381 chemical nature, 380 diffuse component, 377, 379 homeostatic control, 371 long-tenn exchange, 379 reactive form, 381 solubility, 563, 569 Bone of old age, 386 Bone phlegmasia, 301 Bone phosphorus, cortisone, 564 Bone remineralization, 519 Bone remodeling, 321, 371 eruption of teeth, 321, 322 mutations, 326 parathyroid function, 377 shedding of teeth, 321, 332 teleost, 376 vitamin D, 377 yttrium, 450 Bone repair, orthodontic treatment, 303 Bone resorption see also Bone destruction; Osteolysis acid decalcification, 505 ATP, 633 autoradiography, 474 biochemical level, 557 calciferol, 627, 631 carbonic acid theory, 581 /3-carotene, 629 chelation, 505, 578 chemistry, 557, 558, 573, 609 chronic gingivitis, 300 citrate, 579, 640 citrate versus lactate, 640 citric acid increase, 617 degree of calcification, 475 dihvdrotachysterol, 631 endogenous mechanism, 558 hibernation, 285, 287 histophvsical studies, 471 hvpervitaminosis A, 626 Krebs cycle, 616 macrophages, 613 methvl green-pyronin, 473 microcinematography, 611 microradiography, 447, 474 model for, 527 multinucleated giant cells, 94, 333, 515 osteoclasts, see Osteoclasts oxygen cofactor, 632 oxygen tension, 609 parathyroid extract, 472, 577, 617 periodic acid-Schiff reaction, 473 polarized light, 481 rates of process, 457 renal failure, 456 sea bass, 376 supravital staining, 474 tissue culture, 609 vitamin A, 628 vitamin D, 629 without osteoclasts, 438 SUBJECT INDEX 717 Bone resorption — Continued yttrium, 451 Bone retention factor, 436 Bone salt crystals cytoplasmic channels, 503 cytoplasmic vesicles, 499 mobilization, 558 osteoclasts, 499, 508 pinosomes, 501, 507 ruffled border area, 508 Bone-seeking elements, 447 Bone-seeking isotopes, 449, 488 parathyroid stimulation, 488 rate of retention, 416 Bone solubility citrate/lactate ratio, 642 hypotheses, 641 Bone solvent, 640 Bone therapy, evaluation, 467 Bone tissue, ia rat, 328 Bone tissue culture, cytochemical tests, 497 Bone tissue water, 416 Bone turnover balance, 447, 449 bedridden patients, 460 measurement, 447 method of studv, 453, 454 normal rate, 457 osteoporosis, 458, 460 summary, 467 Bonelike hard tissue, resorption cavi- ties, 314 Bony crypt, 323 Bore holes, snail, 69, 71 ABO gland effect, 56 demineralized zone, 69 electron microscopy, 60 microradiography, 60 subsurface demineralization, 69 surface replicas, 69 uniformity, 83 Boring acid, 3, 4, 20, 22 chemical softening, 22 concrete casings, 12 mechanical process, 7, 10, 22 mechanical rasping, 60 pressure, 9 rocking action, 9, 22 twisting, 9 Boring algae, 2, 3, 26, 27, 33, 55 Boring associated with feeding, 2 Boring canals, exhumed teeth, 93, 124, 145, 148 cementum, 125 dentin, 125 Boring fungi, 33 Boring gastropods, see Snail Boring mechanism, 4 gastropods, 86; see also ABO octopods, 55 Boring organ, see ABO Boring organisms, erosional effects, 35, 39 Boring plants, 2 Boring snails, see Snail Boring sponges, see Sponge Bottle feeding, dental caries, 267 Botida, 17 calif orniensis, 19 falcata, 18 Bristles barnacles, 56 polychaetes, 56 sipunculid worms, 56 Brush border, osteoclasts, 498 ultrastructure, 499 Bruxism, 93 Bryozoans, 39, 42, 55 Buffer requirement, caries bacteria, 275 Buffering agents, 99 Buried teeth see Ancient teeth; Exhumed teeth Burr, shed antler, 343 Burrowing organisms, 25 Burrowing sponges, 27 Byssal attachment, 12, 19 Byssal muscles, rocking movements, 17 Byssal threads, 8, 16 Byssus gland, 15 718 C"-labeled proline, 580 CaCOg, 62, 77, 79 see also Carbonate aragonite, 86 calcite, 86 coral, 27, 35 dissolution, 68 ovster shells, 32 secretion, 15, 77 Cactus bucks, 355 Cage laver fatigue, osteoporosis, 432 Calcareous algae, 29 sand-producing, 37 Calcareous basal plate, 7 Calcareous deposit, mussel, 19 Calcareous exoskeletons, 55 Calcareous lining, 19 Calcareous matter, sponges, 35 Calcareous mud, 49 Calcareous product of secretion, 1 Calcareous residue, coral reefs, 47 Calcareous rock, 4, 17 dissolution, 20 Calcareous substrate, ABO etching, 55, 62 dissolution, 34 Calcareous worm tubes, 29 Calcase, 81 snail secretion, 60 Calciferol, bone resorption, 627, 631 Calcifiability, 97 Calcification hibernation, 287 rachitic osteoid, 520 regional variations, 479 Calcification powers, corals, 22 Calcified cartilage matrix, 477 organic mass content, 477 resorption, 484 Calcified osteocytes, 395, 401 Calcified surface, resorption, 452 Calcified tissues, comparative biologv, 376 Calcite, 62, 86 ABO activity, 77 etched crvstals, 31 SUBJECT INDEX shell surface, 75 Calcitcostracum, 62 Calcitic moUusk shells, 71, 75 Calcitic nacreous shell, 73 Calcium acid mechanism, 570 daily net absorption, 437 exchangeable pool, 407 excretion rate, 407 sheep requirement, 161 steady state, 564 Calcium balance, 414 osteoporosis, 392, 426 sea water, 25 Calcium binding, citrate, 578 Calcium carbonate, sec CaCO,-.; Car- bonate Calcium citrate, 640 Calcium concentration, blood, 380 Calcium deficiency, 386 Calcium-deficiency disease, 426 Calcium-deficient apatite, 381 Calcium-EDTA chelate, 533 Calcium excretion index, 417 Calcium fluoride enamel surface, 255 ec^uilibrium solubilitv, 256 reaction products. 257 solubility product, 256 Calcium glucoheptonate, 428 Calcium intake, 414 Calcium ion concentratioTi, homeo- static control, 379 Calcium isotopes, 448 Calcium lactate chelate, 638 Calcium leak, 427 Calcium level, blood plasma, 371 Calcium metabolism, 426 adrenal cortex, 429 Calcium mobilization, 533 acid production, 560 bone citrate, 644 endocrine control, 437 Calcium phosphate, 86, 92 effect of ABO secretion, 77, 79 Calcium phosphate substrate, 62 Calcium reabsorption, kidney tubules, 407 SUBJECT INDEX 719 Calcium-sequestering agent, 114 see also Chelating agents Calciuria, 417 Calculus, dental, 292, 298 hibernation, 289, 291 Calf bone, polysaccharides, 440 California, bivalves, 12, 19 Callus formation, proline tolerance test, 391 Calvaria biochemical studies, 590 chemical findings, 591 glycolytic properties, 591 lactate concentrations, 562 mice, 561 oxygen consumption, 502 preosseous mesenchvmal cells, 590, 591 Calvarium destruction, 94 Calvarium metabolism, parathvroid action, 601 Canada Department of Agriculture, 554 Canaliculi, 375, 425 leaching of mineral, 464 mucoprotein cylinders, 425 Ca/P ratio enamel, 138 herbage, 162 Caio(P04),.(OH)o, bone mineral, 381 Ca9(P04),;H.,(OH).,, bone mineral, 381 Ca,, ( PO^ ) 4 ( OH ) ( 0;,P0— H— OPO..^), bone mineral, 381 Cap-chur rifle, 356 Capillary circulation, bone aging, 438 Capreolus caprcolus, 347 Captivity conditions, animal caries, 289 Carbohydrate cariogenic potential, 264 intercellular store, 599 Carbohydrate degradation, 262 Carbohydrate metabolism, bone, 577 Carbohydrate oxidation, hexose mono- phosphate shunt, 600 Carbon dioxide, see Cd Carbonate, 79 apatite crystallites, 203 carious dentin, 201 cemented sandstones, 18 enamel, 133, 139, 277 enamel caries, 184, 197 sedimentation, 35 Carbonic acid local pH, 584 produced by sponges, 35 Carbonic acid theory, bone resorption, 581 Carbonic anhydrase, 646 osteoclasts, 511 Carboxvlic groups, 640 Cardiacea, 16 Caribou, antlers, 347, 353 Caries, 93, 98, 145, 147 animal, see Animal caries bacterial invasion, 99, 109 bacterial specificity, 272 bottle feeding, 267 chelation theory, 647 chemical character, 213 chemistry, 187 contrast to erosion, 93 contrast to other hard tissue de- struction, 98 critical pH, 652 demineralization gradients, 146 dentin, see Carious dentin; Dentin caries desalivated animals, 295 dietary substrate, 262, 264, 266 enamel, see Carious enamel; Enamel caries endocrine secretions, 295 etiological factors, 262 experimental hamster model, 275 factorial triad, 262, 263 gnotobiotic studies, 267 hibernation, 285, 287 in vitro, 213, 217, 219 infectious agent, 269 inhibition, 261 initiation, 261 invasive infection, 272 microflora, 55, 264, 267 720 Caries — Continued microstructural changes, 171 milk powders, 266 nonacid decalcification theory, 648 nutritional status, 265 pattern of attack, 182 polarized light, 178 protease, 294 proteolysis-chelation theory, 647 recrystallization process, 206 resting period, 176 salivary substrate, 266 subsurface demineralization, 98 transmission, 261 ultrastructure, 187, 190 Caries bacteria buffer requirement, 275 COo requirement, 275 inorganic salt requirement, 275 nutrients from tooth, 277 polysaccharide storage, 277 Caries control, 278 Caries initiation, specificity, 272 Caries in vitro, 213, 217 acetic acid, 219 lactic acid, 219 Caries-like cavities, exhumed teeth, 130 Caries-like lesions, 217 bacterial metabolism, 219 model system, 214 Caries resistance, fluoride ion eff^ect, 276 Caries transmissibility, 270 summary, 278 Cariogenic components, alimentary canal microbiota, 269 Cariogenic flora, 262 antibiotics, 268 contagiousness, 268 Cariogenic microbiota, hamsters, 268 Cariogenic microorganisms, coloniza- tion of, 274 Cariogenic plaque, 268 hamsters, 272 Cariogenic potential carbohydrates, 264 SUBJECT INDEX enterococci, 267 foodstuffs, 267 lactobacilli, 267 streptococci, 267 Cariogenic streptococci, plaque, 270 Cariogenicity, 266 Carious dentin carbonate, 201 collagenous fibrils, 200, 201 fluoride content, 203 intercanalicular matrix, 193, 198, 199 trace elements, 205 ultrastructure, 191 zinc content, 203 Carious enamel aluminum, 199 crystallites, 194, 195 fluoride content, 197 iron, 199 lead, 199 organic material, 191 organic matrix, 196, 197 surface layer, 204 trace elements, 201 Carious invasion, hibernation, 289 Carious lesion bacterial plaque, 217 hamster, 273 incipient, see Incipient carious le- sion Carious lesion formation, 217 chemical systems, 220 dynamics, 244 elementary mechanism, 220 enamel surface, 248 mathematical treatment, 252 polarized light, 225 Carious process, organic matrix, 206 Carious teeth, chemical composition, 195 Carious zones, birefringence, 218, 227 Carnivorous Gastropoda, 4; see also Snail /^-Carotene, 629 Carriers of enzymes, 505 Cartilage invasion, 479, 481 resorption, 483 SUBJECT INDEX 721 Cartilage matrix, 477 Ivsosome permeability, 628 vitamin A, 628 Cartilage partitions, 481 Cartilaginous model stage, 321 Case studies, osteoporosis, 387, 413 Cashmere stag, antlers, 347 Castrated deer, 343, 358 abnormal antler forms, 354, 355 antelope, 355 antler shedding, 341, 354, 357, 358 axis deer, 355 contraceptive, 358 induced shedding, 358 replacement therapv, 356 roe buck, 355, 358 ' velvet shedding, 349, 354 Castrates ACTH metabohsm, 429 calcium retention, 427 Catabolism of bone, 425 Cavities, coral, 45 Cavity formation, 98 Cell membrane, osteoclasts, 499 Cell proliferation, argyrophilic fibers, 487 Cell types, collagenase production, 689 Cellular component in resorption, 486 Cellular elements, resorption, 472 Cellular metabolism, exogenous glu- cose, 581 Cellular microenvironment, resorptive capacities, 492 Cellular specializations, bone, 489 Cemental tears, hibernation, 292 Cementation, algae, 47 Cementocytes, 96 Cementum, 91, 95, 96 boring canals, 125 production, 306 resorption, 97 Centers of ossification, 321 Central America, ancient teeth, 120, 121, 127 Cervical erosion, restoration, 119 Cervuhis miintjak, 347 Cervus axis, 347 canadensis, 347 cashmirianus, 347 dtitauceli, 347 cJaphiis, 347 cldi, 347 uippon, 347, 356 unicolor, 347 Chalkiness, enamel surface, 134 Chelating agents, 81, 655 bone matrix, 645 boring snail secretion, 647 citrate, 656 and collagenase activity, 685 and demineralization, 60 dental plaque, 654, 655, 656, 658 EDTA, 533, 685 enamel proteins, 115, 648 herbage, 166 hydroxy group, 167 incubated hard tissues, 649 sucrose, 655 in tissues, 637 Chelating influence in dental erosion, 114, 115 Chelating properties influenced by pH, 638 Chelation bone resorption, 505 enzyme action, 639 nature, 637 role in decalcification, 637 Chelation mechanism, 637 bone resorption, 578 Chelation theories dental caries, 647 summary, 657 Chemical action rock boring, 5 snail drilling, 57 sponges, 27 Chemical boring, 1 see also under Boring Chemical composition carious teeth, 195 sheep's teeth, 163 Chemical dissolution, sponges, 35 Chemical erosion, 100 rats' teeth, 100 1^2'2 SUBJECT INDEX Chemical factors bone resorption, 609 dissolution of calcified tissues, 376 Chemical inhibitors, resorption, 615 Chemical kinetics, 251 Chemical level, bone resorption, 573 Chemical mechanism bone resorption, 558 shell destruction, 55 Chemical nature of bone mineral, 380 Chemical phase transformation, 257 Chemical repair, enamel, 145 Chemical softening Lithophaga, 21 rock boring, 18, 20, 22 Chemical systems carious lesion formation, 220 incipient carious lesions, 257 Chemistry of bone resorption, 557, 558' Chemistry of caries, 213 sinnmary, 207 Chemistry of pus, 647 Chemomechanical means, perforation of shells, 56 Chemotherapeutic agents, anticaries therapy, 278 Chewing habits, 93 Chicks parathyroid extract, 582 vitamin D, 536 Chital, antlers, 347, 350 Chitinous covering, barnacles, 6 Chondrocytes, glycogen depletion, 489 Chondrocytic glycogen, 487 Chondrophore, 9 Chorionic gonadotropin, 359 Chronic gingivitis, bone resorption, 300 " Chronic staiie, dental erosion, 108 Chymotrypsin, collagen digestion, 685 Ciliary currents, bivalves, 8, 20 Circulatory disturbances alveolar bone, 298, 300, 310 bone marrow, 300 fibrosis, 301 periodontium, 298 tooth resorptions, 315 Circulatory insufficiency, devitalized bone, 395 Cirratulidae, 4 Cirripedes, 6 Citelhis undiilatus, 286 Citrate, 99 see also Citric acid bone cultures, 621 bone resorption, 640 calcium binding, 578 chelator, 656 dental erosion, 114 dental plaque, 656 dissolution of sheep's teeth, 167 parathyroid hormone, 578 parathyroid treatment, 641 unique position, 640 variable fraction, 644 Citrate concentration fluoroacetate, 594 fumarate, 595 Citrate content, saliva, 114 Citrate hypothesis, bone resorption, 579 Citrate versus lactate bone resorption, 640 bone solubility, 642 resorbing calvaria, 623 Citrate level, adrenalectomy, 429 Citrate metabolism, parathyroid hor- mone, 640 Citrate oxidation, ia rat bone, 566 Citrate synthesis, vitamin D, 640 Citrate theory, hard tissue destruction, 460 Citric acid see also Citrate bone cells, 559 effect on sheep's teeth, 167 yttrium, 451 Citric acid increase, bone resorption, 617 Citrogenase, parathyroid action, 586 Clay, boring in, 10 Cleftlike spaces, osteoclasts, 499 Cliona, 5, 29, 30 celata, 26, 27,28,29, 31 lampa, 26, 27 larva, 26, 27 SUBJECT INDEX CUona — Continued sfationis. 31 vdfitifica. 30, 33 Clionid sponges, 27 mechanism of boring, 31 siliceous spicules, 30 Clionidae, 29 Clostridium, collagenase, 680, 685 CO. dissolution of shell, 55 production by parathyroid hormone, 581 requirement of caries bacteria, 275 streptococcal growth, 275 Coalescence, bone cells, 489 Code information, resorption, 490 Codfish spine, 422 Colgate-Palmolive Company , 208 Collagen assay techniques, 684 basement lamella, 687, 688 biological life span, 570 cold, removal, 671 conversion rate, 676 cross linking, 668 cross-striated fibrils, 680 crystal binding, 503 dispersed state, 667 lysis, 680 metabolic turnover, 664 net synthesis, 676 osteolysis, 553 protective coat, 688 rate of resorption, 673 rate of synthesis, 673 ruffled border area, 501, 510 solution, 503 specific activitv, 675 terrestrial mammals, 674 Collagen breakdown, proteolytic en- zymes, 679 Collagen degradation bone, 424 collagenolytic enzyme, 679 irradiation, 424 resorbing calvaria, 624 Collagen digestion, proteolytic en- zvmes, 685 Collagen fractionation, hypothesis, 668 Collagen fractions, 668 extractable, 673 guinea pig skin, 679 insoluble, 674 isotope data, 674 rodent skin, 677 specific activity, 678 Collagen gels, reconstituted, 664 Collagen metabolism, 663 metamorphosing tadpoles, 667 theory, 667 Collagen microassay radioactivity, 685 tissue cultures, 682 Collagen molecules, 668 Collagen pool h\'droxyproline, 673 proline incorporation, 670 Collagen protection, ground substance, 687, 690 Collagen remodeling collagenase, 679 summary, 689 Collagen removal, bone growth, 664 Collagen turnover time, bone, 580 Collagenase see also CollagenoKtic enzyme acid pH range, 680 bone, 573 Clostridium, 680 collagen remodeling, 679 cysteine inhibition, 685 dentin caries, 207 EDTA inhibition, 685 pH optimum, 685 tissvie storage, 687 Collagenase activity, index, 685 Collagenase-like activitv, ultrafiltrable hydroxyproline, 572 Collagenase production, cell tvpes, 689 Collagenolytic activity bone, 664, 680 proteases, 686 tissue extracts, 664 Collagenolytic enzyme see also Collagenase 724 SUBJECT INDEX Collagenolvtic enzyme — Continued collagen degradation, 679 connective tissues, 664 Collagenolytic properties, tadpole tis- sues, 682 Collagenous fibers, exhumed teeth, 123 Collagenous fibrils carious dentin, 200, 201 sound dentin, 200, 201 Collapse, dorsal spine, 388 Collapsed vertebrae, aging, 415 Collodion-carbon replicas, 69, 70, 71 Colonization of cariogenic microorgan- isms, 274 Colony form, reef corals, 39 Commensalism, 6 Compact bone remodeling, 371 summary, 381 Comparative biology, calcified tissues, 376 Compensatory bone apposition, 299, 306 alveolar bone, 310 Complex ions, 638 Complexing compounds, 214; see also Chelating agents Compression fractures, 398, 431 Concentration gradients, enamel dis- solution, 250 Conchiolin, 81, 85 oyster shell, 32 Conchiolin matrix, 2 Conchiolinase, ABO secretion, 60 Concrete casings, boring, 12 Confluent lacunae, 547 osteogenesis imperfecta, 549 Connective tissue metamorphosis, 664 collagenolvtic enzvmes, 664 Contagiousness, cariogenic agent, 278 Contraceptive, castrated bucks, 358 Contractile peduncle, 6 Contradictory conclusions, resorptive states, 471 Controlling mechanisms, resorption, 472 Conversion rate, collagen, 676 Coral attachment, 45 Coral block fracture, earthquake shocks, 47 (.'oral-boring bivalve, 16 Coral-boring polychaetes, 3 Coral-boring sponges, 56 Coral CaCOs, 27, 35 Coral-encrusted detritus, 50 Coral fallout, 46 Coral fauna, 37 Coral polyps, 26, 27, 42 Coral reef biotope, 50 Coral reef remodefing, summary, 50 Coral reefs, 1 Atlantic, 6 avalanches, 46, 47, 48, 49 biotic zones, 35 calcareous residue, 47 ecology, 35 fore-reef slope, 27, 37 formation, 25 frame-cementing biota, 50 Indo-Pacific, 6 intertidal "nick," 4 Jamaica, 35; see also Jamaica coral reefs Pacific Ocean, 50 I^ed Sea, 50 sediments, 47 skeletal carbonates, 49 skeletal debris, 47 structure, 35 subsidence and slump, 47 summary, 50 talus fallout, 46 talus formation, 46 Coral skeletons, 1 Coral shdes, 46 Coralline algae, 37 Coralliophilidae, 1 Corals, 29 see also Reef corals aragonite needles, 46 CaCOs deposition, 27 cavities, 45 deep-water, 41 effect of boring sponges, 41 encrusting biota, 42 SUBJECT INDEX Corals — Continued erosion, 45, 50 growth, 42 polyps, 26, 27, 42 rate of calcification, 50 skeletal debris, 46 zooxanthellae, 39 Corals and algae, symbiotic relation, 3 Cortical biopsies, 428 Cortical bone fluorophotomicrograpliy, 394 osteons, 396 resorption rates, 458 Cortisone antler growth, 361 antler shedding, 357 bone destruction, 428 bone phosphorus, 564 Cortisone acetate, antler shedding, 362 Cortisone-induced osteoporosis, rabbit, 456 Cosmetic remodeling, enamel, 145 Costo-chondral junction, 479 Covalent bonds, 638 Crania, rats, 481 Cranial vault, 473 Crassostrea gigas, 4 virginica, 59, 61, 62, 69, 71, 73, 77, 79, 81, 86 Crenated surface, bone lamellae, 453 Critical pH decalcification, 657 dental caries, 652 Cross-/3-linkage protein, enamel ma- trix, 97 Cross linking, collagen, 668 Cross striations enamel prisms, 173 collagen fibrils, 680 Crush fractures, spine, 454 Cryptorchidism, 356 Crystal binding, collagen, 503 Crystal outlines, shells, 73 Crystallites, carious enamel, 194, 195 Crystals, serrated edges, 75 Cushing's syndrome, 386, 427 Cuticle, enamel, 222 725 Cuticular membrane, protection, 109 Cuticular plates, 3 Cypris larvae, 6 Cysteine inhibition, collagenase, 685 Cytochemical tests, bone tissue cul- ture, 497 Cytochemistry, osteoclasts, 497, 511 Cytolysomes, 511 Cytoplasmic channels bone salt crystals, 503 free bone salt, 505 Cytoplasmic folds, osteoclasts, 499 Cytoplasmic ingestion, osteoclasts, 486 Cytoplasmic RNA, osteoclasts, 486 Cytoplasmic tentacles, sponge, 29 Cytoplasmic vacuoles, 645 nucleated inclusions, 489 osteoclasts, 481, 486 Cytoplasmic vesicles, bone salt crys- tals, 499 Dama damn, 347 Dark zone, enamel caries, 172 Date mussel, Mediterranean, 19 Dayhght, antler shedding, 353, 354 Dead bone, osteoclasts, 515 Dead tracts, dentin, 156 Decalcification see also Demineralization chelation potential, 637 critical pH, 657 polymeric ingredient, 223 resorption process, 486 systems of, 220 Deciduous nature of deer antlers, 339 Deciduous teeth, 95 Howship's lacunae, 144, 334 Deep-water corals, sponge burrows, 41 Deer antlers, see Antler Degenerative disease, osteoporosis, 424 Degree of calcification, resorption, 472, 475, 484 Dehydrogenase giant cells, 523 osteoclasts, 511 Demineralization see also Decalcification 7^26 SUBJECT INDEX Demineralization — Continued bone, 578 chelating agents, 60 dentin caries, 195 Embden-Meverhof-Krebs s\stem, 581 enamel caries, 171 in vitro, 131 invertebrate phyla, 55 plant groups, 55 preceding proteolysis, 207, 505 shell, 79 Deinineralization-boring mechanism, 56, 60 mollusks, 56 snails, 56, 60; see also ABO Demineralization gland, see ABO Demineralization gradients, 131 dental caries, 98, 147 dental erosion, 107, 146 postmortem destruction, 146 resorption, 146 Demineralization mechanism, boring gastropods, 55 Demineralization organ, snail, see ABO Demineralization patterns, enamel caries, 175 Demineralization solution, organic polymer, 133 Demineralization tests, rat molars, 138 Demineralized collagen, 505, 506 subjacent matrix, 503 Demineralized subsurface zone, incipi- ent enamel caries, 133 Demineralizing action, ABO homoge- nate, 67, 68 Demineralizing agents, hard tissues, 131 Demospongiae, 29 Denervated antlers, shedding mecha- nism, 360 Densitometry, osteoporosis, 422 Density bone, 418 osteons, 372, 416 Dental attrition, 92, 93, 155 Dental calculus, 292, 298 Dental caries, see Caries Dental cement, orthophosphoric acid, 137 Dental erosion, 99 see also under Erosion acidity, 113 bacterial involvement, 109 buried teeth, 120 cavity effect, 111 cervical, 119 chelating mechanism, 115 chelation influence, 114 chronic stage, 108 circumscribed, 104 clinical management, 117 cuticular membrane, protection, 109 demineralization gradients, 107, 146 diagnosis, 118 disc-shaped, 102 distribution, 101, 102 enamel, 104, 105 figured, 104 fluoride therapv, 118 frictional forces, 100 histopathologv, 108 horizontal grooves, 102 hourglass appearance, 102 idiopathic, 91 irregular, 104 lacunar profile, enamel, 105 mechanical friction, 100, 101, 116 metachromatic deposit. 111 microradiography, 105, 107 mucous plaque, 109 mucous secretion, 114 occlusal wear, 112 opacity, enamel, 104 oral environment, 113 orthochromatic membrane. 111 oxalate, 114 pathogenesis, 112 postmortem, see Postmortem erosion prevention, 118 prognosis, 118 salivary citrate, 114 subsurface changes, 131 summary, 145 surface demineralization, 108 SUBJECT INDEX 727 Dental erosion — Continued susceptibility, 111 therapy, 118 tin fluoride, 118 tooth-brushing etiology, 102 tooth structure, 111 topographic configuration, 101 wedge-shaped, 104 Dental eruption, see Eruption of teeth Dental hard tissues, preformed struc- tural patterns, 98 Dental plac[ue ATP, 655 chelating agents, 656 chelating powers, 655 citrate, 656 lactate, 656 salivary COo, 276 salivary extracts, 656 sialic acid, 657 Dental resorption, 98 periodontal disorders, 312 Denticles, radular, 57, 60, 71, 81, 86 Dentin, 79, 85, 86, 96, 98, 131 boring canals, 125 dead tracts, 156 internal resorption, 528 intertubular matrix destruction, 109 organic bonding material, 169 stnicture, hibernation, 287 Dentin caries see also Caries; Carious dentin collagenase, 207 demineralization, 195 large crystallites, 193 organic matter, 188 pericanalicular layer, 193 proteolysis, 195 proteolysis following demineraliza- tion, 207 Dentin erosion, proteolysis, 168 Dentin resorption, 96, 97 Dentinal canals, microorganisms, 198, 199, 202, 203 Dentinal tubules, 99, 124 obliteration, 112 Dentinogenesis, 292 Dento-alveolar destruction, 92, 145, 148 histopathological characteristics, 145 spectrum of pathology, 92 Dento-alveolar hard tissues, 91 Dento-alveolar resorption, 148 periodontal disorders, 297 summary, 145, 315 Dento-alveolar structures, hard tissue destruction, 91, 145, 297 Denuded collagen, 509 apatite crystals, 505 Depolymerization, mucopolysaccha- rides, 531 Desalivated animal, caries, 295 De-specializing osteoblasts, microen- vironmental fields, 491 Detached bone crystals, 501 in osteoclasts, 503 Devitalized bone, circulatory insuffi- ciency, 395 Devitalized osteocytes, 396 Diabetes, 99 osteoporosis, 437 periodontal disorders, 305 Diamox, 646 Dianabol, 399 Diaphorase, giant cells, 523 Diaphyseal bone osteogenesis imperfecta, 551 parathyroid adenoma, 551 Dicalcium phosphate, 253, 254 reaction products, 257 Dichocoenia, 39 Dichromatic absorption, radiography 422 Dietary substrate, caries, 262, 264, 266 Differential decalcification, enamel caries, 182 Differential rate of solution, crystals, 79 Diffuse component, bone mineral, 377, 379, 448 Diffuse disintegration, exhumed teeth, 125 728 Diffusion control, enamel dissolution, 251 Diffusion gradients, enamel dissolu- tion, 236 Diffusion pathways, enamel, 221, 249 Diffusion of radioactive isotopes, enamel, 249 Diffusion rate, heterogeneous reaction, 214 Digestive system, hibernation, 285 Dihydrogen form, phosphate, 585 Dihydrotachysterol, bone resorption, 631 Dinitrophenol, inhibition of resorption, 615 Diphtheroids, oral flora, 270 Diploe, resorption, 483 Diploic spaces hyperparathyroid animals, 485 parietal bone, 481 resorption, 484 Diphria, 39 Direct bone action parathyroid hormone, 589, 609 vitamin A, 627 vitamin D, 631 Disc-shaped erosion of teeth, 102 Disodium ethylenediamine tetraace- tate, see EDTA Dispersed state, collagen, 667 Dissolution calcareous substrate, 34 calcified tissues, chemical factors, 376 calcium carbonate, 68 infected bone, 646 tooth substance, 99 Dissolution of calcareous rock, 20 pH, 34 Dissolution of shell acids, 55 CO2, 55 enzymes, 55 sequestering agents, 55 Division of Biology and Medicine, United States Atomic Energv Commission, 554 SUBJECT INDEX Dogs bone biopsies, 533 parathyroid extract, 532, 543 Dorsal spine, collapse, 388 Dosinia discus, 62, 67, 71, 73, 86 Drilling gastropods, demineralization- boring mechanism, 56 see also ABO Drilling site, snails, 57 Dutch shell disease, 2 Dye-binding capacity enamel, 149 rehardened enamel, 139 Dye penetration, Retzius lines, 140 Dynamics carious lesion formation, 244 heterogeneous system, 257 Early caries, microradiography, 175 Early diagnosis, osteoporosis, 422 Earthquake shocks, coral block frac- ture, 47 East Indian coral reefs, erosion, 5 Easter Seal Foundation, 440 Eastern oysters, 61 Eating habits, caries control, 278 Eating patterns, older individuals, 267 Echinoderms, 1 Echinoid echinoderms, 4 Echinometra mathaei, 5 Ecology, reef coral, 27, 35 Ectocranial resorption, 485 EDTA (ethylenediamine tetraace- tate), 214 infusion of, 533 EDTA-bound calcium, plasma, 533 EDTA inhibition, collagenase, 685 EDTA venoclysis, 545 Egg capsule, 81 Egypt, ancient teeth, 120, 126 Elaphurus davidianus, 347 Elasmobranchs, placoid scales, 339 Elastase, collagen digestion, 685 Eld's deer, antlers, 347 Electron micrographs aragonite surface, 83 bone matrix, 505 calcite surface, 83 SUBJECT INDEX 729 Electron micrographs — Contin ucd erosion, 113, 168 etched shell region, 75 incipient carious lesion, 231 osteoclasts, 528 shell etching, 75 Electron microscope level, enzyme cytochemistry, 512 Electron microscopy, 65 bore holes, 60 etched shell, 71 in vitro caries, 229 osteoclasts, 497, 498, 501, 503 osteoporosis, 424 Elk, antlers, 347 Embden-Meyerhof-Krebs system demineralization, 581 parathyroid action, 581 Embden-Meyerhof pathway, 580 Embryonic bone, 498 Embryonic fowl, bone electron micro- graphs, 497 Enamel, 79, 85, 91, 96, 98, 131 absence of keratin, 649 Ca/P ratio, 138 carbonate, 139, 277 caries, see Enamel caries chemical repair, 145 cosmetic remodeling, 145 diffusion pathways, 221, 249 diffusion of radioactive isotopes, 249 dye-binding capacity, 149 enzyme attack, 649 equilibrium solubility, 253 exchange reactions, 149 hardness, 138 in vivo demineralization, 134 in vivo etching, 137 intrinsic birefringence, 179 ionic exchange mechanism, 140 mechanical remodeling, 145 mineralization period, 142 natural surface coatings, 133 opacity, 131 organic coating, 222 organic-inorganic linkage, 140 organic matter, 97, 116 pathological demineralization, 143 permeability, 141 phosphate deposition, 139, 141 pohshed surfaces, 62 radiodensity, 136 radioisotope gradients, 142 redeposition of minerals, 149 reprecipitation, 257 resorption, 96, 144, 148, 489 solubility, 138 solubility rates, 255 stainability, 137 white spots, 215 Enamel caries see also Caries; Carious enamel bacterial invasion, 171 birefringence, 177, 180 carbonate, 197 carbonate loss, 184 dark zone, 172 differential decalcification, 182 form birefringence, 178, 181 incremental striae, 173 magnesium, 197 microradiography, 173, 215 organic content, 188, 231 organic material loss, 184 pathways of attack, 184 patterns of demineralization, 175 polarized light, 176 prism cores, 173 prism cortex, 183 proteolysis, 171 radiolucence, 173 recrystallization, 191 remineralization, 176 selective demineralization, 175 soluble matrix, 184 submicroscopic spaces, 179 translucent zone, 172 summary, 185 ultrastructure, 190 Enamel-cementum junction, 303 Enamel crystallites perforation, 194, 195 surface erosion, 191, 194, 195 Enamel cuticles, 221 780 SUBJECT INDEX Enamel demineralization, 171 lines of Retzius, 134 Enamel diffusion, intenod spaces, 248 Enamel dissolution, 236 acid-organic polymer systems, 241 concentration gradients, 250 diffusion control, 251 diffusion gradients, 236 equilibrium, 251 first order diffusion, 251 hydrogen ion concentration, 237 kinetics, 251 lactate concentration, 240 mathematical treatment, 251 phase identification, 253 phosphate salts, 253 physical chemistry, 213 reaction rate, 235', 236, 238 reaction products, 250 summary, 257 undissociated acid, 241 Enamel matrix, 97 Enamel plaque, 217 Enamel-pla(|ue complex, 217 Enamel-placjue junction, 217 Enamel prisms, cross striations, 173 Enamel protein, 116 absence of keratin, 649 mucoprotein, 649 source of chelators, 657 Enamel remineralization, 130 "Heilungsyorgang," 140 hypothesis, 143 theoretical basis, 140 Enamel resorption, 96, 148 osteoclasts, 144, 489 Enamel-saliva contact, liquid-solid equilibrium, 143 Enamel shells, exhumed teeth, 126 Enamel surface, 79 calcium fluoride, 255 carbonate content, 133 carious lesions, 248 chalkiness, 134 fluoride content, 133 microradiographs, 134 protectiye mechanism, 248 reconstitution, 145 replicas, 136 Enamel uptake, radioactivated saliva, 142 Encrusting algae, 5, 39 Encrusting biota, corals, 42 Encrusting sponges, 45 Endocranial resorption, cartilage in- vasion, 483 Endocranial surface, PAS stain, 483 Endocrine control, calcium metabo- lism, 437 Endocrine factors antler cycle, 35 1 antler growth, 360 antler shedding, 354 Endocrine glands, hibernation, 285 Endocrine influences, osteoporosis, 434 Endocrine secretions, dental caries, 295 Endocrinopathies, 426 Endocrinopathy aging of bone tissue, 438 osteoporosis, 424 Endogenous mechanism, bone resorp- t'ion, 558 Endogenous oral debris, 117 Endosteal lining cell, 499, 503 Endothelial cells, resorption, 490 Enovid, 358 antler shedding, 357 Enterococci, cariogenic potential, 267 Environment of bone cells, 637 Environmental factors, antler shed- ding, 346 Environmental stress, hibernation, 295 Enzyme sec aha separate entries for individ- ual enztpnes dissolution of shell, 55 juices of herbage, 168 osteoclasts, 511 Enzyme action chelation, 639 resorbing edge, 505 Enzyme attack, enamel, 649 Enzyme cofactor, 640 Enz\me content, giant cells, 523 SUBJECT INDEX 731 Enzyme cvtochemistn , electron micro- scope level, 512 Enzyme-producing cells, tissue water loss, 690 Enzyme transfer, pinosomes, 511 Epifauna, 8 Epifaunal habit, bi\alves, 8 Epifaunal origin, bivalves, 12 Epincphehis striafiis, 376 Epiphyseal cartilage plate, 371 Epithelial attachment downgrowth, 292, 294 enamel, 303 hibernation, 291 Epulis tumors, tooth resorptions, 312 Equilibrium, enamel dissolution, 251 Equilibrium solubility, 214, 255 calcium fluoride, 256 enamel, 253 fluorapatite, 256 measurements, 254 Eroded tooth substance. 111 Eroded tooth surface, organic ma- terial, 108 Erosion see also Dental erosion; Erosion of sheep's teeth boring organisms, 39 corals, 45, 50 East Indian coral reefs, 5 periostracum, 18 rock, 22, 102 shell, 9, 67 shell valves, 9, 17 sponges, 31, 45 wave effect, 25 Erosion-abrasion, 93, 100 Erosion agents, rock boring, 22 Erosion of atoll reefs, 25 Erosion and caries, reverse lelation- ship, lis Erosion control, vestibular shield, 120 Erosion of corals boring sponges, 37, 46 sediment distribution, 46 Erosion-like defects, 99, 100 Erosion patient, general constitution, 117 Erosion of sheep's teeth see also Attrition of sheep's teeth acetate, 167 chelating agents, 166 citrate, 167 electron micrography, 168 in vitro, organic acids, 166 Erosion of teeth, see Dental erosion Erosional eftects, boring organisms, 37 Erosional process, boring sponges, 50 Erosional remodeling, coral reefs, 46 Erosive effects, worn teeth, 166 Eruption of teeth, 91, 95, 321 bone remodeling, 321, 322 summar\', 336 Eskimos, periodontal tissues, 305 Estradiol, 358 Estrogen, 356 antler shedding, 354, 357 Etat histocvtaire, osteoclasts, 515 Etched shell, see Shell etching Etching see also Shell etching calcite crystals, 31 hydroxyapatite crystals, 62 //) vivo enamel, 137 oyster shell, 31 rat molars, 139 sheep's teeth, 167 teeth, 138 tooth mineral, by ABO secretion, 62, 79 variability in, 68 Ethylenediamine, 64, 75 effect on shells, 79 Ethylenediamine tetraacetate, see ' EDTA Etiology dental caries, 262 periodontal disorders, 297 FAipleiira, 63, 69 caiidata etterae, 59, 61, 83, 86, 87 European oyster shells, 5 Evolution of skeleton, resorption cavi- ties, 371 Ewes, 161 incisors, 158 wear of teeth, 162 73^2 SUBJECT INDEX Ewing's sarcoma, 549 Exchangeable bone mineral, 379 Exchangeable pool, calcium, 407 Excised ABO, 77, 83, 85 see also ABO etchings, 62, 81 Exhumed teeth see also Ancient teeth; Postmortem algae, 129 boring canals, 124, 126, 145, 148 caries-like cavities, 130 collagenous fibers, 123 diffuse disintegration, 125 erosion, 120, 125 fungoid excavations, 128, 129 histopathology, 120 petrification, 125 shells of enamel, 126 Wedl's canals, 128 Excessive wear, sheep's teeth, 165 Exchange reactions, enamel, 149 Excretion rate, calcium, 407 Excurrent oscules, sponge, 30 Excurrent papilla, sponge, 31 Exercise, osteoporotic patients, 390, 439 Exogenous glucose, cellular metabo- lism, 581 Exposed collagen, osteoclasts, 512 Experimental animals, osteoporosis, 432 Experimental bone remodeling, ortho- dontic acti\'ators, 306 Experimental demineralization, 130 rhesus monkey, 135 Experimental hamster model, caries, 270, 275 External canals, postmortem destruc- tion, 126 External counting, radioactive iso- topes, 448 vttrium, 452 Extractable collagen fractions, 673 Facsimile, saliva ions, 143 Factorial triad, caries activity, 262, 263 Fallow deer, antlers, 341, 347, 352, 354 Fathometer profile, 47 Faulty tooth structure. 111 Fecal and oral streptococci, 270 Feces inoculation, animal caries, 269 Female antlers abnormal, 361 caribou, 353 reindeer, 354 Fetus, maxilla, 322 Fibrocellular connective tissue, antler pedicle, 344 Fibrosis, 302, 303 alveolar bone marrow, 301 circulatory disturbances, 301 Fibrous capsule, implant, 518 Fibrous dysplasia of bone, 385 Figured erosion of teeth, 104 Filamentous algae, 2 First order difl-usion-controlled reac- tion, enamel caries, 251, 257 Flagella, 29 sponges, 5 Flagellated chambers, 28, 29 Flatworms, 1, 4, 55 Fluid-solid equilibrium, mouth, 145 Fluorapatite, equilibrium solubility, 256 Fluorescent microscopy, 395 Fluoride, 99, 161 antienzvmatic effect, 276 carious dentin, 203 carious enamel, 197 enamel surface, 133 erosion therapv, 118 glycolysis, 276 intercanalicular dentin, 193 long-lasting anticaries effect, 276 manure, 165 New Zealand sheep's teeth, 165 New Zealand soils, 165 osteolysis, 552 Fluoroacetate citrate concentration, 594 lactate production, 595 SUBJECT INDEX 733 Flvioroacetate — Continued mobilization of calcium, 580 oxygen uptake, 594 Fluorophotomicrography, cortical bone, 394 Fluorosed sheep's teeth, Australia, 165 Fluorosis, osteocytes, 522 Folded membrane, macrophages, 499 Food substrates, acidogenic potential, 266 FoodstuflFs adhesiveness, 267 cariogenic potential, 267 Foot sucker, bivalves, 10 Foraminifera, 39, 42 Ford Foundation, 87 Fore-reef slope habitat, 27 Foreign body giant cells, osteoclastic ability, 515 Form birefringence, enamel caries, 178, 181 Fowl bone, tissue cultures, 497 Fractures, skeletal, 431 Frame-cementing biota, coral reefs, 50 Free bone salt, cvtoplasmic channels, 505 Frictional forces, dental erosion, 100 Frog tadpole, metamorphosis, 664 Fruit drinks, 99 Fruit juice, salivarv citrate, 114 Full-mouthed ewes, 157 Fumarate, citrate concentration, 595 Functional adaptation bones, 321 tooth eruption, 325 Functional impulses, structure of bones, 299 Functional states, bone cells, 489 Functional stress alveolar bone, 299, 306, 307 pericementum, 299 periodontium, 298 resorption, 490 tissue reactions, 308 Fungi, 1, 2 boring, 33 exhumed teeth, 129 shell-boring, 55 Fungoid excavations, exhumed teeth, 127 Funnel shape, tibia, 481 Fusiforms, 270 Galactosamine/glycosamine ratio, bone matrix, 426 Galactosidase, osteoclasts, 511 Gastric acid, 93 Gastrochaenacea, 14 Gastropod, 1, 73 see also Snail Gastropod bore holes see Bore holes, snail Gastropod boring mechanism, sum- marv, 86 Gastropod families, 56 Gastropod mantle, 68 Generalized bone atrophy, 386 Generalized bone disease, 95 Genetic cell potential, resorption, 489 Genetic constitution, antiosteoporosis factor, 439 Genetic factors, osteoporosis, 418, 433 Genetic information, bone cells, 492 Genetic loci, resorptive response, 490 Giant cell formation, phthioic acid, 528 Giant cells, 96 acid phosphatase, 520, 527 blood elements, 527 dehvdrogenase, 523 diaphorase, 523 enzvme content, 523 glucuronidase, 527 Howship's lacunae, 520 oxidase, 523 oxidative enzymes, 520 parathyroid extract, 523 phosphatase, 527 phosphoamidase, 527 polyvinyl sponge implants, 523 resorption, 515 summarv, 529 trypan blue, 520 Giant cells versus osteoclasts, 527 Giant clam, 16 734 Gingiva anemic, see Anemic gingiva color change, 305 hibernating animals, 289 swelling, 297 Gingival biopsies, regressive inflamma- "tion, 305 Gingival granulation tissue, tooth re- sorption, 313 Gingival inflammation, 113, 297, 298, 300, 305 radiation treatment, 311 Gingival pockets migrated teeth, 309 periodontal disorders, 297 Gingival sulci, 113 Gingivitis, 297, 301, 303 chronic, bone resorption, 300 hvperplastic, 312 remote effect, 300 Glands of Leiblein, 68 Glossiness enamel surface, 138 shell, 60 Glucocorticoid hormones, 428 Glucose metabolism, parath\roid ef- fect, 564 Glucose rinse, 656 Glucuronidase giant cells, 527 osteoclasts, 511 Glutaric acid, 167 Glycine, 638 Glycine-H"*' bone matrix, 475 osteoclasts, 486 Glycogen bone, 599 osteocytes, 487, 489 Glycogen depletion, chondrocvtes, 489 Glycolysis ascites tumor, 598 bone, 596 fluoride, 276 Glycohtic cycle, osteoc) tes, 380 Glycolytic pathway, bone, 585 Glycolytic properties, calvaria, 591 Glycosidase, osteoclasts, 511 SUBJECT INDEX Gnotobiotic studies, caries, 267 Gnotobiotic technics, animal caries, 270 Gold fillings, erosion, 100 Gonadectomies, deer, 360 Gonadotropins, antler shedding, 358 Gonads antler growth, 355 osteoporosis, 429 Gorgonacea, 35, 41, 42 Graded microenvironments, bone as a tissue, 491 Gradients in demineralization erosion of teeth, 106, 107 hard tissue pathologv, 144, 145, 146 Gram-negative rods, 270 Granulation tissue, 97 Grass, see Herbage Grass and clover juices, solvent action, 167 Grass sap, malic acid, 166 Gravitv, displacement of coral detritus, 46 Great Barrier Reef, 1, 19 bivalves, 12 Greece, ancient teeth, 120, 123, 126, 127 Greenland Eskimos, 303, 310 Ground squirrel Alaska, 286 hibernation, 286 Ground substance collagen protection, 687, 690 reticulin, 688 Growing individuals, periodontia, 302 Growth cartilage, 483 corals, 42 microorganisms in hibernation, 295 Growth hormone antler growth, 363 osteoporosis, 401 Growth period, tooth germs, 323 Growth pressure algae, 3 antler shedding, 346 Growth processes, algae, 5 SUBJECT INDEX 785 Growth rate antlers, 339 haversian systems, 375 Growth remodeling, 479 Growth rings of von Ebner, 98 Gruber's s\'ncvtium, 343 Guatemala, ancient teeth, 121, 127 Guinea pig skin, collagen fractions, 679 Hamster acidogenic bacteria, 274 cariogenic microbiota, 268 cariogenic plaque formation, 272 carious lesions, 273 streptococci, 270 Hamster caries, proteolvsis-chelation theory, 275 Hard exoskeletons, perforation of, 56 Hard tissue biology, concepts of, 97 Hard tissue destruction citrate theory, 640 demineralizing agents, 131 dental, 91 den to-alveolar aspects, 92, 297 dento-alveolar structures, 145 pathogenesis, 147 resorption cavities, 313 Hardness enamel, 138 sheep's teeth, 167 Hatchet-shaped foot, bivalves, 8 Haversian canal, sclerotic ring, 449 Haversian systems, 372 accretion, 377 growth rate, 375 hotspots, 377 "Heilungsvorgang," enamel reminerali- zation, 140 Hemopoietic tissue, resorption, 490 Hemostatic pressure, 83 snail boring organ, 57 Henderson-Hasselbach equation, bi- carbonate-carbonic acid system. 584 Herbage Ca/P ratio, 162 chelating agents, 166 enzymes, 168 mineral content, 161 organic acids, 94 phytoliths, 169 press juice pH, 166 species, 162 wear of sheep's teeth, 157, 161 Hereditary hypoosteoclasis, 328 Hereditarv jaw defects, sheep, 159 Hermatypic (reef-building) corals, powers of calcification, 22 Heterogeneous reaction, 213 Heterogeneous s\'stem, dvnamics of, 257 ■ ' Hexosamine, bone, 426, 646 Hexosamine collagen ratio, 416 Hexose monophosphate shunt, 579 carbohydrate oxidation, 600 Hiofclla, 12 saxicava, 14 Hibernating animals, 286 Hibernation, 94 alveolar bone, 287 bone, 285, 294 bone resorption, 287 calcification process, 287 calculus, 289, 291 caries, 285, 287 carious invasion, 289 cemental tears, 292 dentin structure, 287 digestive system, 285 endocrine glands, 285, 295 environmental stress, 295 epithehal attachment, 291 gingivae, 289 ground squirrel, 286 growth of microorganisms, 295 involution of endocrines, 295 lability of bone minerals, 295 microbial activity, 295 osteoporosis, 289 periodontal pocket, 291 periodontium, 286 physiology of mammals, 285 pulpal involvement, 289 salivary glands, 294 736 Hibernation — C out in ued salivary protease, 294 space travel, 285 stress, 286 summary, 295 teeth, 285 tooth structure, 285 torpor, 285 High resolution microradiographv, 131 Hippocamchis antisensis, 349 hisiilcus, 349 Histochemistry, in vitro caries, 233 Histocytes osteoclast precursors, 490 resorption, 490 Histoenzymologic methods, 516 Histogenesis, antlers, 344 Histopathological characteristics, den- to-alveolar destruction, 145 Histopathology dental erosion, 108 exhumed teeth, 120 Histophysical studies bone cells, 471 bone resorption, 471 Histophysics of bone resorption, sum- mary, 491 Hog deer, antlers, 347, 348 Homeostatic control bone mineral, 371 calcium ion concentration, 379 Homoiotherm, 286 Horizontal grooves, erosion of teeth, 102 Hormones see also separate entries for individ- ual hormones antler shedding, 353, 354, 360 Horns, antelope, keratinized sheaths, 339, 347 Horny periostracum, 9 Host versus teeth, 264 Hotspots, haversian systems, 377 Hourglass appearance, erosion of teeth, 102 Houses of ivory, 92 SUBJECT INDEX Howship's lacunae, 94, 96 deciduous teeth, 144, 334 giant cells, 520 Kuemul, antlers, 349 Human growth hormone, 413 osteoporosis treatment, 400 Hydrated tricalcium phosphate, 381 Hvdrogen bonding, 381 Hydrogen ion concentration see also pH enamel dissolution, 237 snail boring organ, 65 Hydroids, 42 Hvdrolytic enzymes, osteoclast, 511, 512 Hydroxy group, chelating activity, 167 Hydroxyapatite, 79, 86, 219, 221, 253, 381 phosphate group, 256 reaction with acid, 249 Hydroxyapatite crystals, etching, 62 Hydroxyapatite surface, reaction prod- ucts, 254 Hydroxyethyl cellulose, 131, 133, 218, 223, 224, 233 Hydroxyproline bone matrix, 426 collagen pool, 673 metamorphosis, 669 resorbing calvaria, 626 ultrafiltrable, 572 urinary excretion, 425 Hydroxyproline removal, peritoneal lavage, 553 Hyelaphus porcinus, 347, 348 Hvperactivity, oral soft tissue, 100 Hyperadrenal corticoidism, 426 Hypercorticoidism, osteoporosis, 429 Hyperemia, antler shedding, 341 Hyperfunction alveolar bone, 299 mandibular incisors, 309 Hypermineralized bone, 402 Hyperosteoclasis, 432 Hyperparathyroid animals diploic spaces, 485 resorption, 485 SUBJECT INDEX 737 Hvperparathvroidism, 95, 97, 425, 426, 427 inicroradiographv, 462 periodontal disorders, 312 Hyperplastic gingivitis, tooth resorp- tions, 312 Hyperproteinemia, 429 Hypertonic sugar solutions, 277 Hypertrophic lacunae, 522 Hypertrophic osteocyte, 538, 539 salt removal, 552 Hypervitaminosis A, bone resorption, 626 Hypocalcemia, 533 Hyponatremia, 429 Hypoosteoclasis, 332 Hypoparathyroidism, 330 Hypophosphatasia, 385 Hypophysectomy, antler growth, 362 Hypothesis biological decalcification, 637 bone solubility, 641 collagen fractionation, 668 enamel remineralization, 143 vitamin D action, 644 Hypothyroidism, antler growth, 361 Hypsodont (sheep's) tooth, 155 la bone, rat, citrate oxidation, 566 la mutation, albino rat, 328 la rat bone metabolism, 566 bone tissue, 328 incisors, 329 odontoma, 329 Iceland, ancient teeth, 120, 127 Iceland spar, 35 Idiopathic dental erosion, 91, 147 see also Dental erosion Idiopathic osteoporosis, 424, 427 Iliac crest, biopsy, 461 Immature bone, 536 Immobilization, bone formation, 460 Implant autogenous bone, 515, 516 destruction of, 527 fibrous capsule, 518 phagocytic cells, 516 rachitic bone, 526 osteoclasts, 515 tropism, 528 Improved pastures. New Zealand, 165 7;i vitro caries, 213, 217 electron microscopy, 229 histochemistry, 233 In vitro demineralization, 131 In vitro erosion, organic acids, 166 In vitro metabolic studies, bone, 557, 578 In vitro model, bone metabolism, 563 three-phase system, 567, 570 In vitro studies, resorptive mecha- nisms, 557 In vivo demineralization, enamel, 134 In vivo etching, enamel, 137 Inactive osteons, 396 Incipient carious lesion birefringence, 215 chemical system, 257 electron micrography, 231 formation, 215 microradiography, 226, 227 morphology, 215 polarized light, 218 rate of formation, 253 Incipient enamel caries demineralized subsurface zone, 133 white spot, 213 Incisors rhesus monkey, 134 sheep, 155, 158 Incremental striae, 98 enamel caries, 173 Incubated hard tissues, chelating ac- tion, 649 Incurrent papilla, sponge, 31 Incurrent pores, sponge, 30 Index collagenase activity, 685 protease activity, 685 India, deer, 347 Individual bone cells, microenviron- ment, 491 Indo-Pacific coral boulders, 16 Indo-Pacific coral reefs, 6, 14 738 SUBJECT INDEX Inert-look iiig c-Ntoplasm, osteoclasts, 505 Iiitauna, 8 Iiifaunal bivalves, 8, 9 Infected bone, dissolution, 92, 646 Infected offspring, animal caries, 269 Infectious agent, caries, 269 Inflammatorv reactions, human perio- dontium, 297 Infrared absorption spectroscopy, 256 Inherited factors, alveolar bone de- struction, 298, 312 Inhibition of caries, 261 Inhibition of resorption, 609 dinitrophenol, 65 malonate, 615 Initiation of caries, 261 Inner fluid layer, bone, 568 Inorganic constituents, mollusk shells, 77 Inorganic ion requirement, strepto- cocci, 277 Inorganic ions, bacteria, 276 Inorganic material, resorption of, 570 Inorganic phase, osteoclasis, 528 Inorganic salt requirement, caries bac- teria, 275 Insoluble collagen fractions, 674 preferential removal, 687 Intact surface layer, 133 Interalveolar arteries, 300 Interalveolar vessels, pericementum, 314 Intercanalicular matrix, carious dentin, 193, 198, 199 fluoride, 193 Intercellular matrix, resorption sites, 472 Intercellular store, carbohydrate, 599 Intercrystalline spaces, 73, 79 Interface, bone and extracellular fluids, 568 Interglobular spaces, dentin. 111, 287 Intermittent masticatory impulses, ac- tivators, 306 Internal canals, postmortem destruc- tion, 127 Internal influences, antler shedding, 353 Internal reconstruction, bone, 322 Internal remodeling, compact bone, 371 Internal resorption, 96 dentin, 528 Interprismatic cross striations, 98 Interprismatic regions, subsurface de- mineralization, 139 Interprismatic substance, 98 Interrod spaces, enamel difi^usion, 248 Interstitial basophilic resorption, 545 Interstitial lamellae, 372, 395 Intertidal echinoids, 5 Intertidal nick, coral reefs, 4 Intertrochanteric fractures, 420, 432 Intertubular matrix destruction, den- tin, 109 Intracellular mechanism, apatite dis- solution, 645 Intracellular pH, 598 Intracellular phosphate, 604 Intralacunar resorption, 425 Intrinsic birefringence, enamel, 179 Invasion of dentinal tubules, 287 Invasive infection, caries, 272 Invertebrate phyla, demineralization, 55 Involution of endocrines, hibernation, 295 Iodine, 161 Ion exchange, 379, 380 bone, 372 enamel, 140 Ion transfer, bone and blood, 380 Ionic bonds, 638 Iron, carious enamel, 199 Irradiation collagen degradation, 424 soluble collagen, 425 Irregular erosion of teeth, 104 Irregular surfaces, resorption, 455 Isotope data, collagen fractions, 674 Isotropic enamel, 177 Ivory, 91, 131 Ivory houses, 92 SUBJECT INDEX 739 Jamaica coral reefs communities, 25 deep fore-reef slope, 37 fore-reef, 37 reef crest, 35 reef systems, 36, 37 Japanese oysters, 4 Jaw bone apposition pattern, 325 metabolic disease, 95 Jaw defects of sheep, hereditary, 157 Jaws modeling resorption, 323 osteosclerosis, 328 remodeling, 322, 323 tumors, 92 Jaws and teeth, hard tissue destruc- tion, 91, 297 Joint pains, ACTH effect, 399 Juices of herbage enzymes, 168 pH,' 166 Kannesteinen (Tobv-Jug Stone), Nor- way, 103 Keratin absence in enamel matrix, 649 chelating effect of, 115 Keratinized horn sheaths, antelope, 339 Keratinohtic bacteria, 116, 648 Keratinous media, 115 «-Ketoglutarate, transamination reac- tions, 599 Kidney, parathyroid hormone, 582 Kidney tubules calcium reabsorption, 407 strontium reabsorption, 407 Kinetics bone metabolism, 430 enamel dissolution, 251 Knoop hardness number, 157, 167 Krebs cycle, 166, 167 bone, 580, 585, 593 bone resorption, 616 enzymes of bone. 578 Labeled bone matrix, 479 Labile bone, 471 Labile bone mineral, 377 hibernation, 295 Lactate, 99 see also Lactic acid apatite crystals, 640 bone cultures, 621 bone metabolism, 559 dental plaque, 656 parathyroid hormone, 559 Lactate concentration calvaria, 562 enamel dissolution rate, 240 Lactate production bone, 596 fluoroacetate, 595 parathyroid hormone, 602 pH effect, 598 Lactation, antler shedding, 354 Lactic acid, 64, 75 see also Lactate bacterial metabolism, 220 caries in vitro, 219 effect on enamel, 218 Lactobacilli, cariogenic potential, 267 Lactobacillus acidophilus, 261, 267, 268 arahinosus. Til casei, 219 fennenti, 268 Lactogenic hormone, antler growth, 354 Lacunar resorption, trabecular bone, 551 Lacunar profile, erosion of enamel, 105 Lamina dura, 95 bone apposition, 307 bone resorption, 300, 307, 309 Lamellae of bone, low-densit\- areas, 449 Lamellar bone, 477 Lankenau Hospital Conference, 415 Large crystallites, dentin caries, 193 Large osteoc)tes alkaline phosphatase, 553 metachromasia, 551 Lathyric seed, osteolysis, 552 Laurence-Locarte densitometer, 422 740 Laving hens adrenal glands, 432 cage layer fatigue, 432 Diamox, 646 Leaching of minerals, canaliculi, 465 Lead, carious enamel, 199 Lead lines, bone formation, 375 Leocrates siciliensis, 4 Leucine aminopeptidase bone cells, 553 osteoclasts, 511 resorption, 553 Life cycle, osteoclast, 613 Life expectancy, osteocytes, 397 Light intensity, coral skeletogenesis, 51 Light microscopy, striated cell border, 499 Limestone, 35 Liinnoria, 2 Lines of Retzius, see Retzius hnes Liquid-solid equilibrium, saliva and enamel, 143 Lithophaga, 17 acid secretion, 21 chemical softening, 21 curningiana, 19 phinmJa, 20 Lithothamnioid algae, 39 Lithofrijo, 6, 7 Living prey, snail boring, 71 Local agent, osteoporosis, 424 Local pH, 584 Localized resorption, 95 LoJiiim perenne, 155, 166 Long-term exchange, bone mineral, 379 Loose teeth, scurvy, 95 Lovett Memorial Group for Study of Diseases Causing Deformities, 691 Low-density bone lamellae, 449 Lumbar spine, ballooned discs, 389 Lumbar vertebrae bone failure, 389 osteoporosis, 454 Lubricating action, salivary secretion, 93, 112 SUBJECT INDEX Lijsidice collaris, 4 Lysis, collagen, 680 Lvsosome, 511 osteoclasts, 511 Lvsosome permeabilitv, cartilage ma- trix, 628 Lytic cells, basement lamella, 691 Lytic enzymes, osteoclasts, 489 Lytic process, bacterial collagenase, 682 Macrophages bone implants, 519 bone resorption, 613 folded membrane, 499 Madracis, 42 Magilus, 1 Magnesium, enamel caries, 197 Malic acid, 167 grass sap, 166 Malonate inhibition of resorption, 615 oxygen uptake, 593 Mammalian jaws, remodeling, 322 Mandible, 322 tooth germs, 322 Mandibular incisors, hyperfunction, 309 Mantle ca\it\', water pressure, 18 Manure, fluoride content, 165 NLarine Biological Laboratory, 87 Marine borers, 25 Marine mussels, 8 Marine rock-boring organisms, 1 Masson trichrome stain, 535 Mastication, periodontal tissues, 308 Masticatory function, 93 Masticatory functional stress, 298, 299, 306 Masticatory impulses, orthodontic acti- vators, 299, 306 Masticatorv wear, 91 Mathematical treatment carious lesion formation, 252 enamel dissolution, 251 radioisotope kinetic data, 431 ALatrix of bone, 477 see also under Organic matrix SUBJECT INDEX Matrix of bone — Continued precursors, 488 preferential bone resorption, 484 Matrix of calcified cartilage, 477 Matrix of enamel, see under Organic matrix Maturation of antlers, 348 _ Maxilla, 322 fetus, 322 tooth germs, 322, 323 Mazama bricenii, 350 rufa, 349 Mechanical abrasion, 100 postlar\al sponges, 34 Mechanical action shell valves, 8 sponge cells, 33 Mechanical boring, 1, 7, 10, 22 Mechanical friction dental erosion, 101, 116 toothbrush, 101 Mechanical functional stress, alveolar bone, 306 Mechanical inflammation, periodon- tium, 307 Mechanical rasping, boring, 56, 60 Mechanical remodeling, enamel, 145 Mechanical shift, bone crystals, 503 Medical Research Council of Canada, 554 Melanaxis, 349 Mercenaria mercenoria, 59, 62, 68 Meristematic tissues, 166 Mesenchymal cells, pseudopodial ex- pansions, 33 Mesenchymal specializations, resorp- tion, 490 Mesenchvmatous cells, 515 Metabolic acids, bone cells, 637 Metabolic action, rat calvaria, 589 Metabolic balance studies, osteoporo- sis, 400 Metabolic balance tests, osteoporosis, 391 Metabolic bone, 379, 471 tissue, 414 741 Metabolic disease jaw bone, 92, 95, 97 skeleton, 453 Metabolic disturbances, periodontal disorders, 312 Metabolic turnover, collagen, 664 Metabolism calcium, 426 protein, 428 total collagen, 670 Metabolite signal, prey of snails, 57 Metabolizing leaves, 166 proteinases, 94, 168 Metachromasia, 536 large osteocytes, 551 osteoclasts, 486 Metachromatic deposit, dental erosion, 111 Metachromatic osteocytes, 539 Metamorphosed sponges, 31 Metamorphosing tadpole collagen metabolism, 667 remodeling s)'stem, 663 Metamorphosis connective tissue, 664 frog tadpole, 664 hvdroxvproline, 669 proline, 669 thyroid compounds, 663 tissue water, 665 Metaph\seal spongiosa, 479 Metaphvseal trabeculae, 472, 477 Metaphvseal zone, parathyroid stim- ulation, 488 Metaphysis, 371 periosteal surface, 450 Metastatic bone disease, 385 Metastatic tumors, 95 Methandrostenolone, osteoporosis treatment, 399 Meth\'l green-pyronin, bone resorp- tion, 473 Methyl methacrylate filling, 100 Mice calvaria, 561, 609 osteoporosis, 433 Microbial activity, hibernation, 295 74^2 SUBJECT INDEX MicTOcinematogiapln , lioiie resorp- tion, 611 Microdensitv, peripheral enamel, 137 Microelements, sheep's teeth, 161 Microenvironment, indixidual bone cells, 491 Microenvironmental fields, de-special- izing osteoblasts, 491 Microenvironmental modifications, structural genes, 491 Microflora alimentary canal, 269 caries, 264, 267 Microorganisms see also Bacteria caries, 55 dentinal canals, 198, 199, 202, 203 Microradiographic appearance, yt- trium deposition, 451 Microradiographic changes, osteoporo- sis, 422 Microradiography bone resorption, 447, 474 bore holes, 60 dental erovion, 105, 107 dento-alveolar destruction, 145 early caries, 175 enamel caries, 173, 215 enamel surface, 134 etched shells, 65 hy^oerparathyroidism, 462 monkey incisors, 137 osteogenesis imperfecta, 462 osteoporosis, 460 Paget's disease, 462 rickets, 462, 464 semiquantitatiye anahsis, 467 tibia, 372 tooth demineralization, 133 Microradiography of bone, summarN', 467 ^ ' Microstructural changes, dental caries, 171 Migrated teeth, gingival pockets, 309 Migration teeth, 297, 308, 309, 326 tooth germ, 325 Milk powders, caries activity, 266 Mineral content, herbage, 161 Mineral densitw atrinii bone, 459 Mineral metabolism, 448 adrenalectom\', 428 mineralocorticoids, 361 Mineral solubility, vitamin D effect, 564 Mineralization failure, osteoid tissue, 465 Mineralization of lacunae, aging, 552 Mineralization period, enamel, 142 Mineralocorticoids antler growth, 361 mineral metabolism, 361 Minute-to-minute ecjuilibrium, blood and bone, 379 Mitochondria, osteoclasts, 615 Mobility, teeth, 297 Mobilization of bone mineral, 558 osteocvtes, 379 Mobilization of calcium, fluoroacetate, 580 Model for bone resorption, 527 Model system caries-like lesions, 214 remodeling of mesenchyme, 689 Modeling resorption, 321, 323 Modiolus demisstis, 59 Mollusk shells, 59, 71, 83, 86 aragonitic, 71 calcitic, 71 inorganic constituents, 77 nacreous surfaces, 72, 73 Mollusks, 1, 25, 29, 42 demineralization-boring mechanism, 56 Molting, 340 Monkey, see also Rhesus monkey Monkey incisors, microradiographs, 137 Monkey teeth, radiotracer studies, 141 Monohydrogen form, phosphate, 585 Monfastrea aunidaris, 26, 27, 44 Moose, antlers, 347, 352, 353 Morphogenesis, 321 Morphology of incipient carious le- sions, 215 SUBJECT INDEX 743 Mother animals, alimentarv canal mi- croflora, 269 Mouse calvaria, resorption, 609 Mouth, fluid-solid equilibrium, 145 Mouth defects, sheep, 157 Mouth of tadpole, intricate remod- eling, 691 Mucoid of saliva, 133 Mucopolysaccharides aging, 426 depolymerization, 531 osteolysis, 553 osteoporosis, 426 Mucoprotein cylinders, canaliculae, 425 Mucoproteins bone, 425 enamel, 649 osteoporosis, 435 Mucous glands, erosion, 114 Mucous plaque, erosion, 109 Mucous substance, saliva, 140 Mucous tubes, worms, 6 Mucus mussels, 20 snails, 57 Mudstones, 1 Mule deer, antlers, 355 Multinucleated giant cells, 94, 333 see also Giant cells; Osteoclasts bone destruction, 515 Multiple myeloma, 385 Murex hrevifrons, 59, 86 fulvescens, 61, 62, 67, 71, 73, 77, 79, 81, 86, 87 Muricid snails, 56, 68 ABO gland, 85 bore holes, 59, 85 pH of secretion, 65 sole of the foot, 56 Muricidae, 4, 56, 57, 59, 61 Mussa, 39 Mussels boring, 17 calcareous deposit, 19 mucus, 20 Mutant strain, rats, 565 Mutations bone remodeling, 328 ia rat, 328 Mya, 9 arenaria, 9 Myacea, 9 Mijcetophyllia, 39 Myristic acid, bone, 424 Mvtilidae, 17 Nacreous shell, 68 Nacreous surfaces, 66, 67 shells, 72, 73, 77, 81 Nail-like bodies, barnacles, 6 Naked collagen fibers, 645 Nanoplankton, 5 Naticid snails, 56, 59 ABO gland, 85 borings, 57 proboscis, 56, 85 Naticidae, 4, 56, 57, 59, 61 National Institute of Arthritis and Met- abolic Disease, 382 National Institute of Dental Research, 149, 208, 492, 529, 587, 634 National Institutes of Health, 149, 208, 364, 382, 439, 492, 529, 587, 634, 691 National Science Foundation, 52, 492 Natural juices, 99 Natural surface coatings, enamel, 133 Necrotic bone, 417 Nephrectomized animals, parathyroid secretion, 489 Nerita, 4 Net synthesis, collagen, 676 Neurovascular influences, antler shed- ding, 343 New Zealand sheep, see Sheep; Sheep's teeth New Zealand soils, fluorine, 165 Newborn infants, saliva, 267 Nitrogen balance, osteoporosis, 392 Nonacid decalcification theory, dental caries, 648 Norethandrolone-treated chicks, 531 alpharadiographv, 539 744 Norethynodrel, 358 antler shedding, 357 Normal bone criteria, 457 osteocytes, 486 preferential resorption, 479 turnover rate, 457 Normal resorption, acceleration of, 488 parathyroid extract, 488 Norway, ancient teeth, 120, 121, 127 Nucleated inclusions, cytoplasmic vac- uoles, 489 Nucleolar RNA, osteogenesis, 487 Nucleotides, 639 Nutrients from tooth, caries bacteria, 277 Nutritional canals, alveolar bone, 309 Nutritional factors, attrition of sheep's teeth, 160, 165 Nutritional requirements, streptococci, 275 Nutritional status, dental caries, 265 Obliteration of dentinal tubules, 112 Obstipation, dental erosion, 99 Occlusal wear, dental erosion, 112 Octopods, boring mechanism, 55 Odocoilcus columhianiis, 347 gymnofis, 348, 349 hem ion us, 355 virginianiis, 347 Odontoblastic layer, 287 Odontoblasts, 99 Odontoclasts, 96, 333 Odontogenesis, 322 Odontogenic epithelium, 328 Odontogenic tumors, 95 Odontome, ia rat, 329 Official osteoclasts and giant cells, 527 Old and new bone, 475 Older individuals, eating patterns, 267 Opacity enamel, 131 enamel erosion, 104 Opal phytoliths, abrasion of sheep's teeth, 166 Opercular plates, 7 SUBJECT INDEX Optical microscopy, shell etching, 65 Oral cavity, bacterial colonization, 267 Oral citrate, 115 Oral debris, endogenous, 117 Oral environment, erosion, 113 Oral epithelium, 323 Oral microbiota, 262 Oral soft tissue hvperactivity, 100 twitching, 100 Oral streptococci, 270 Orchiectomy, 433 Oreton, 360 Organic acids herbage, 94 in vitro erosion, 166 Organic bonding material, dentin, 169 Organic coating apatite crystals, 248 protection of enamel surface, 220, 222 Organic content, enamel caries, 231 Organic films, inhibition of acid action, 117 Organic-inorganic linkage, enamel, 140 Organic mass bone, 475 calcified cartilage, 477 degree of calcification, 475 osteochist nuclei, 479 resorption, 486 woven bone, 479 Organic material dentin caries, 188 enamel, 116 enamel caries, 184, 188, 191 eroded tooth surface, 108 resorbable tissues, 97 Organic matrix see also Matrix basophilia, 531 carious enamel, 196, 197 carious process, 206 enamel, 97 osteoclasis, 528 Organic phase, resorption of, 580 SUBJECT INDEX 745 Crganic polymer-acid systems, 244 Organic polymers, 131, 133 demineralization solution, 133 Organic substrates, shell etching, 62 Organic surface films, protective inter- action, 147 Orthochromatic membrane, erosion of teeth, 111 Orthodontic activators alveolar bone remodeling, 306 bone formation, 306 experimental bone remodeling, 306 masticatory impulses, 306 Orthodontic appliances, pressure re- sorption, 335 Orthodontic bands, 137 Orthodontic tooth movement, 95 Orthodontic treatment, 92 alveolar bone, 307 alveolar bone resorption, 314 bone repair, 303 Orthophosphoric acid demineralization of enamel, 137 dental cement, 137 Osseofac, 408, 413 Osseous precursors, 488 Ossification centers of, 321 testosterone, 355 Osteitis deformans, 385 Osteitis fibrosa generalisata, 385 Osteoblast characteristics, osteocyte relationship, 490 Osteoblasts, 375 bone marrow, 304 osteoprogenitor state, 487 resorption effect, 489, 641 Osteoclasis, 428 inorganic phase, 528 organic matrix, 528 Osteoclast nuclei, organic mass, 479 Osteoclast precursors, 96, 144, 489, 527 histocytes, 490 Osteoclastic ability, foreign body giant cells, 515 Osteoclastic action, time sequence, 528 Osteoclastic activity, pedicle-antler junction, 358 Osteoclastic resorption, 376 parath\roid glands, 376 Osteoclasts, 91, 375 acid phosphatase, 511 aggressive action, 613 antecedent cellular elements, 96 bone junction, 498 bone marrow, 304 bone salt, 499 brush border, 498 carbonic anhydrase, 511 carriers of enzymes, 505 cell membrane, 499 cleftlike spaces, 499 cytochemistry, 497, 511 cytoplasmic folds, 499 cytoplasmic ingestion, 486 cytoplasmic RNA, 486 cytoplasmic vacuoles, 481, 486 dead bone, 515 dehydrogenases, 511 detachment of crystals, 503 electron microscopy, 497, 498, 501, 503, 528 enamel resorption, 96, 144, 489 enzymes, 511 etat histocytaire, 515 exposed collagen, 512 galactosidase, 511 glycine-H3, 486 glycosidase, 511 glycuronidase, 511 hydrolytic enzymes, 511, 512 implants, 515 inert-looking cytoplasm, 505 leucine aminopeptidase, 511 life cycle, 613 lysosomes, 511 lytic enzymes, 489 metachromasia, 486 mitochondria, 615 mode of action, 497 origin, 96, 146 osteolytic action, 377 oxidases, 511 pH, 646 740 Osteoclasts — Continued phagocytosis, 489, 512, 528 pinosomes, 501, 506 precursors, see Osteoclast precursors readiness for phagocytosis, 512 resorbing surface, 451 resorption, 639 ribosomes, 499 ruffled border, 499, 501, 506 saclike vesicles, 508 striated border, 498 structure-function relationships, 497 succinic dehydrogenase, 614 summarv, 512 tissue culture, 511 trypan blue uptake, 520 ultrastructure, 497 vacuoles, 613 Osteoclasts versus giant cells, 527 Osteocyte canaliculi, 481 Osteocyte counts, osteoporosis, 419 Osteocvte lacunae, 483 osteomalacia, 465 Osteocvte sheaths, 425 Osteocytes, 372, 375 beryllium intoxication, 522 dual role, 552 fluorosis, 552 glycogen, 487, 489 glycolytic cvcle, 380 life expectancy, 397 mobilization of bone mineral, 379 normal bone, 486 osteoblast characteristics, 490 osteolathyrism, 522 parathyroid extract, 489 proteolytic enzvme, 395 resorption, 639, 640 resorptive potential, 489, 642 rickets, 465 woven bone, 465, 477 Osteocytolysis, 416, 425, 438 Osteodystrophy, 385 Osteogenesis active, 472 nucleolar RNA, 487 remodeling, 321 Osteogenesis imperfecta, 385, 426 SUBJECT INDEX confluent lacunae, 549 diaphyseal bone, 551 microradiography, 462 unremodeled bone, 462 Osteogenic complex, 491 Osteogenic microenvironment, 491 Osteogenic sarcoma, 549 Osteogenic surfaces parathyroid action, 487 rib shaft, 483 Osteogram, 392, 413 Osteoid, 536 phase contrast photomicrography, 481 PTE-treated animals, 488 rachitic, 515, 520 Osteoid tissue, failure of mineraliza- tion, 465 Osteolathyrism, osteocytes, 522 Osteolysis, 531, 639 see also Bone destruction; Bone re- sorption collagen, 553 fluoride, 552 lathyric seed, 552 mechanism of, 553 mucopolysaccharides, 553 pregnancy, 552 Osteolytic action, osteoclasts, 377 Osteolytic effects, biochemical mecha- nisms, 589 Osteolytic phenomenon, 552 Osteomalacia, 385 osteocyte lacunae, 465 Osteons, 372 active, 395 biologic half life, 375 cortical bone, 396 density, 372 porosity, 455 resorption cavities, 457 reversal cement lines, 399 sclerotic rings, 456 Osteopenia, 424 Osteoporosis, 95, 385 accelerated aging, 417 adrenal cortical hormones, 428, 432 adrenalectomy, 429 SUBJECT INDEX Osteoporosis — Continued aged females, 415 aging, 386, 467 alveolar bone atrophv, 424 birds, 429, 432 bone failure, 386; see also Bone failure bone turnover rate, 458, 460 cage laver fatigue, 432 calcium balance, 392, 426 case material, 387 castrates, 429 degenerative disease, 424 densitometry of bone, 422 diabetes, 437 earlv diagnosis, 422 electron microscopy, 424 endocrine influences, 434 endocrinopathy, 424 experimental animals, 432 genetic factors, 418, 433 gonadal influence, 429 growth hormone, 401 hibernation, 289 hypercorticoidism, 429 idiopathic, 424 increased resorption, 461 local agent, 424 lumbar vertebrae, 454 metabolic balance studies, 400 metabohc tolerance tests, 391 mice, 433 microradiographv, 422, 460 mucopolysaccharides, 426 mucoproteins, 435 nitrogen balance, 392 osteocyte counts, 419 peptides, 437 phosphorus balance, 392 phvsiologic aging, 416 placebo ps\chotherapy, 436 postmenopausal, 386 proline tolerance test, 390 racial influence, 418 radioisotope kinetics, 402 rats, 433 sedentarv life, 433 sex, 418 74-7 sex hormones, 387 structural bone, 414 summary, 438 unifying hypothesis, 439 urinary indices, 417 vitamin D, 433 weight loss, 431 White Leghorn hen, 432 x-ray diffraction, 424 Osteoporosis treatment, 435 exercise, 439 human growth hormone, 400 methandrostenolone, 399 polvsaccharides, 408 proline injections, 392 tranquilizers, 436 Osteoprogenitor cells, resorptive areas, 486, 487 Osteoprogenitor state, osteoblasts, 487 Osteosclerosis, jaws, 328 Owen's contour lines, dentin, 98 Oxalate, 99 dental erosion, 114 saliva, 114 Oxeas, 30 Oxidase giant cells, 523 osteoclasts, 511 Oxidative enzymes, giant cells, 520 Oxidative phosphor\'lation, uncou- pling, 600 Oxycorticosteroids, protein break- down, 430 Oxygen, resorption, 490 Oxygen consumption, calvaria, 502 Oxygen effect, bone-resorbing cofac- tor, 632 Oxygen tension bone resorption, 609 parathyroid action, 619 and parathyroid extract, interde- pendence, 619 and vitamin A, interdependence, 628 Oxygen uptake fluoroacetate, 594 nialonate, 593 748 Oxvgen uptake — Continued succinate, 593 TCA cycle, 596 Oxvtetracvcline, 393 Ovstershells, 26, 27, 31 CaCOo, 32 conchiolin, 32 etched lines, 31 Oysters, 2, 71 disease, 2 pests, 6, 22 Ozotocerus bezoorticiis, 349 Pacific Ocean, coral reefs, 50 Paget's disease, 549 bizarre resorption, 462 microradiography, 462 Palate biopsies, 305 Palestine, ancient teeth, 120, 121 Pampas deer, antlers, 349 Papain, 168 collagen breakdown, 679 collagen digestion, 685 Paradontitis, 298 Paradontosis, 298 Parasite-host relation, periodontal dis- orders, 298 Parasitism, 6 Parathormone, sec Parathyroid extract; Parathyroid hormone Para-Thor-Mone, see Parathyroid ex- tract Parathyroid action acetyl coenzyme A, 586 bone matrix, 485 bone metabohsm, 589 bone remodeling, 377 calyaria metabolism, 601 citrogenase, 586 Embden-Meyerhof-Krebs system, 581 glucose metabolism, 564 osteogenic surfaces, 487 oxygen tension, 619 passiye solubility, 564 preferential resorption, 484 pyruvic acid-oxidase complex, 586 summary, 587, 605 SUBJECT INDEX Parathyroid adenoma, 549 spontaneous fractures, 549 Parathyroid assay, tissue culture sys- tem, 621 Parathyroid extract bone metabolism, 577 bone resorption, 472, 577, 617 chicks, 532 dogs, 532 giant cells, 523 osteocytes, 489 osteoid, 488 and oxygen tension, interdepend- ence, 619 rats, 532 resorption, 488 tissue culture calcium, 623 Parathyroid glands, 332, 379, 428 antler shedding, 361 bone implants, 515 osteoclastic resorption, 376 Parathyroid hormone, 425, 429 bone citrate, 559, 578 bone-seeking isotopes, 488 citrate metabolism, 578, 640 COo production, 581 direct action, 589, 609 fundamental action, 642 kidney, 582 lactate production, 559, 602 phosphaturic activity, 604 primary action, 605 target cells, 633 tooth eruption, 330 vitamin D, 630 yttrium, 451 Parathyroid secretion, nephrectomized animals, 489 Parathyroid stimulation bone-seeking isotopes, 488 metaphyseal zone, 488 resorption, 471 Parathyroid treatment, citrate, 641 Parathyroidectomy, bone biopsies, 534 Parietal bones, 473, 477 diploic spaces, 481 vascular spaces, 483 Parotid glands, 114 SUBJECT INDEX PAS, see Periodic acid-Schiff Passive solubility, 643 parathyroid effect, 564 Pasteur effect, 596, 601 Pasture see also Herbage improved, 165 wear of sheep's teeth, 157, 165 Pathogenesis hard tissue destruction, 147 idiopathic dental erosion, 109, 148 Pathways of attack, enamel caries, 182, 184 Parturition, antler shedding, 354 Pedal epithelium, snail, 68 Pedal sac, snail, 63 Pedicle, antler, 339, 343 Pedicle-antler junction, osteoclastic ac- tivity, 358 Pedicle epidermis, antler, 343 Pelecypods, valve edges, 56 Pelican beak, 340 Peptides, osteoporosis, 437 Perandren, 356 Percussion dullness, alveolar bone resorption, 310 Pere David's deer, antlers, 347 Perennial rye grass, 155, 166 Perforation chemomechanical means, 56 enamel crystallites, 194, 195 hard exoskeletons, 56 Periapical abscesses, 291, 639 Pericanalicular layer, dentin caries, 193 Pericemental fibers, 301, 306 root surface resorption, 315 Pericementum, 301, 309 functional stress, 299 interalveolar vessels, 314 pressure zones, 307 Perilacunar sheath, 425 Periodic acid-Schiff (PAS) reaction, bone resorption, 473, 483, 486, 535 Peripheral enamel, microdensity, 137 Periodontal blood circulation, 298 749 Periodontal conditions inherited, 298 radiation treatment, 311 Periodontal disorders, 92, 94 alveolar bone resorption, 297, 310 alveolar crest resorption, 313 dental resorption, 312 diabetic patients, 305 etiology, 297 hyperparathyroidism, 312 gingival pockets, 297 inflammatory reactions, 297 metabolic disturbances, 312 parasite-host relation, 298 percussion dullness, 310 scurvy, 312 summary, 315 Periodontal pockets, 291 hibernation, 291 Periodontal tissues Eskimos, 305 mastication, 308 Periodontal vessels, 311 Periodontium, 303 blood flow, 305 circulatory disturbances, 298 functional stress, 298 growing individuals, 302 hibernation, 286 mechanical inflammation, 307 physiology, 310 structure, 310 tissue tonus, 305 Periosteal bone, 371 Periosteal surface active resorption, 450 metaphysis, 450 Periostracal layer, 69 Periostracal scales, 9 Periostracum, 19, 57, 69, 71, 81 erosion, 18 shell protection, 20 Permeability blood vessels, 301 enamel, 141 Peritoneal lavage, hydroxyproline re- moval, 553 Peruke antlers, 355 750 SUBJECT INDEX Pests of oysters, 6, 22 Petricola carditoides, 13 Petricolidae, 12 Petrification, exhumed teeth, 125 pH ABO homogeiuite, 65 ABO secretion, 68 critical level, 652, 657 herbage press juice, 166 niuricid snails, 65 osteoclasts, 646 pH conditions, phosphate phase, 256 pH determinations ABO secretion, 86 snail boring organ, 64 pH effect chelating properties, 638 dissolution of calcareous material, 34 lactate production, 598 solubility of minerals, 584 yttrium deposition, 451 pH gradient, bone, 584 pH indicator, boring organ, 60 pH levels, saliva, 267 pH optimum, collagenase, 685 Phagocytic cells, implant, 516 Phagocytosis, osteoclasts, 489, 512, 528 Phagosomes, 511 Phase contrast microscopy bone, 535 osteoid, 481 Phase identification, enamel dissolu- tion, 253 Philippine deer, antlers, 348, 351 PJwIadidea, 10 penita, 11, 12 Pholas, 11 Phoronids, 55 Phosphatase, giant cells, 527 Phosphate acidic forms, 257 dihydrogen form, 585 enamel, 139, 141 intracellular, 604 monohydrogen form, 585 renal handling, 385 transport mechanisms, 604 tubular reabsorption, 427 Phosphate excretion, 428 Phosphate group, hydrox) apatite, 256 Phosphate phase, pH conditions, 256 Phosphate salts, enamel dissolution, 253 Phosphatic manures, 165 Phosphaturic activity, parathyroid hor- mone, 604 Phosphoamidase, giant cells, 527 Phosphorus, plasma, 533 Phosphorus balance, osteoporosis, 392 Phosphorus requirements, sheep, 161 Phosphorylated compounds, 166 Phosphorylated sugars, 639 Phosphorylation , 262 Phlegmasia, bone, 301 Phthioic acid, giant cell formation, 528 Physical chemistry of enamel dissolu- tion, 213 Physiologic aging bone, 416, 423 osteoporosis, 416 urinary indices, 417 Physiologic bone atrophy, 386 Physiologic resorption, 321 Physiologic shedding, 96 Physiology of mammals, hibernation, 285 ' Physiology of periodontium, 310 Phytoliths, herbage, 169 Phytoplankton, 7 Pink tooth, 92, 96 Pinocytosis, 645 Pinosomes bone salt crystals, 501, 507 enzyme transfer, 511 osteoclasts, 501, 506 Pituitary cutofli^ mechanism, 433 Pituitary gland, antler-stimulating function, 362 Pituitary hormone, antler shedding, 358 Placebo psychotherapy, osteoporosis, 436 Placoid scales, elasmobranchs, 339 Plankton, 5 SUBJECT INDEX 751 Plants acids in, 167 demineialization action, 55 proteinases, 168 Plaque, 264 calcium turnover, 376 cariogenic streptococci, 270 chelators, 648, 654 enamel, 217 Plaque system, 217 Plasma calcium turnover, 376 EDTA-bound calcium, 533 phosphorus, 533 Platyodon canccllatus, 9 PociUopora, 5 Polarized light bone resorption, 481 caries, 178, 225 enamel caries, 176 incipient lesions, 218 Poliniccs duplicatiis, 59, 61, 67, 77, 86 Polychaetes, 1, 6, 55 bristles, 56 Polijdora, 4, 6 Polyestrous female deer, 348 Polymeric ingredient, decalcification medium, 223 Polymers, organic, see Organic poly- mers Polvphosphates, 639 Polyps, coral, 26, 27, 42 Polysaccharide storage caries flora, 277 protective mechanism, 277 Polvsaccharides calf bone, 440 osteoporosis treatment, 408 Polyvinyl sponge implants, giant cells, 523 Porifera, 29 Porites, 42 osfreoides, 44 Porosity aging bone, 455, 467 osteons, 455 Postlarval sponges, mechanical abra- sion, 34 Postmenopausal osteoporosis, 386, 428 Postmortem canals in teeth, 123 dentin, 125 external, 126 internal, 127 Wedl's, 128 Postmortem destruction of teeth, 148 demineralization gradients, 146 Postmortem erosion, 120 distribution, 126 gross examination, 121 microscopic examination, 124 origin, 127 Postmortem saprophytes, 145 Prebone, 538 Precursors of osteoclasts, 527 Predentin resorption, 528 Predentin zone, 124 Preferential bone resorption critical conditions, 491 normal bone, 479 organic matrix, 484 parathyroid action, 484 Preferential collagen removal, insolu- ble fraction, 687 Preformed structural patterns, dental hard tissues, 98 Pregnancy antler shedding, 354 osteolysis, 552 Prehistoric teeth, see Exhumed teeth Premaxilla, 329 Preosseous mesench\'mal cells, cal- varia, 590, 591 Preosseous tissue, 377 Presarcomatous bone, 462 Pressure zones, pericementum, 307 Pressure resorption, 490 orthodontic appliances, 335 Primary action, paratlnroid hormone, 605 Primary cementiun, 96 Primary change in resorption, demin- eralization, 505 Prism cores, enamel caries, 173 Prism cortex, enamel caries, 183 Proboscis, snail buccal muscles, 59 iO'^l SUBJECT INDEX Proboscis, snail — Continued naticids, 56 tip, 57, 83 Progesterone, antler growth, 362 Prognosis, dental erosion, 118 Projection microradiography, 65 Proline incorporation, collagen pool, 670 Proline injection, osteoporosis, 392 Proline metabolism bone matrix, 425 metamorphosis of tadpole, 669 Proline tolerance test callus formation, 390 osteoporosis, 390 Pronase, collagen digestion, 685 Pronghorn antelope, see Antelope Propodial folds, 85 Propodial proboscis, 83 Propodium, 57 Prosobranchia, 56 Protease caries, 294 reconstituted collagen, 685 vitamin A effect, 628 Piotease activity collagenolytic, 686 index, 687 salivary glands, 294 Protective coat, collagen, 688, 690 Protective cuticular membrane, 109 Protective deposits, apatite surfaces, 257 Protective interaction, organic surface films, 147, 220 Protective mechanism enamel surface, 220, 248 polysaccharide storage, 277 Protective periostracum, 12 Protective property, pyruvate, 594 Protein breakdown, oxycorticosteroids, 430 Protein metabolism, 428 Protein synthesis boring, 3 thyroid compounds, 679 Proteinases see also Proteolytic enzymes metabolizing leaves, 94, 168 Proteins of enamel chelation compounds, 648 proteolytic bacteria, 648 Proteolysis, 98 dentin caries, 195 dentin erosion, 168 enamel caries, 171 secondary, in caries, 207, 657 secondary, in resorption, 505 sheep's teeth, 94 summary, 689 Proteolysis-chelation theory dental caries, 647 hamster caries, 275 insuperable inconsistency, 653 Proteolytic bacteria, 115 proteins of enamel, 648 Proteolytic enzymes, 214 see also Proteinases; Collagenase collagen digestion, 679, 685 osteocvtes, 395 sponges, 33 Protoplasmic processes, sponge, 33 Pseudo replicas, etched shells, 73, 77, 81 Pseudopodia, sponge fragments, 32 Pseudopodial expansions, mesenchy- mal cells, 33 Pseudostylachiis ostreophagiis, 4 PTE, see Parathyroid extract Pudella mephisioplielis, 350 Pudii piidu, 350 Puffins, beaks, 340 Pulp canal, sheep's teeth, 156 Pulp stones, 309, 310 Pulpal abscesses, 294 Pulpal involvement, hibernation, 289 Pulpal odontoblasts, 98 Pus chemistry, 647 solvent action, 647 Pyelonephritis, 427 Pyrophosphates, 639 Pyruvate, protective propert\-, 595 Pyruvate-to-lactate step, 600 Pyruvic acid-oxidase complex, para- thyroid action, 586 SUBJECT INDEX 753 Rabbit, cortisone-induced osteoporosis, 456 Rachitic birds, parathyroid extract, 535 Rachitic bone, implantation, 526 Rachitic osteoid, 515 calcification, 520 Racial factor in osteoporosis, 418 Radiation injury, 92, 97 Radiation treatment gingival inflammations, 311 periodontal conditions, 311 Radioactivated saliva, enamel uptake, 142 Radioactive calcium, 377, 402 Radioactive collagen microassay, 685 Radioactive elements, salivary circula- tion, 141 Radioactive isotopes diffuse component, 448 external counting, 448 rare earths, 450 Radioactive phosphorus, 372 uptake by monkey teeth, 141 Radioactive strontium, 391, 402 Radioassay, total body counting, 393 Radiodensity, enamel, 136 Radioisotope gradients, enamel, 142 Radioisotope kinetics bone metabolism, 391 mathematical treatment, 431 osteoporosis, 402 Radioisotope uptake, saliva, 141 Radiolucence, enamel caries, 173 Radiotracer studies, monkey teeth, 141 Radium, 372, 377 body burden, 378 Radium-burdened bone, 448 bizarre resorption, 455 Radula, snail, 4, 56, 57, 81, 83 rasping action, 56, 57 Radular denticles, snail, 57, 60, 71, 81,86 Radular teeth, see Radular denticles Rana cateshiana, 664, 665 Rangifer tarandus, 347 Rare earths, radioactive isotopes, 450 Rarefying disease of skeleton, 385 Rarefying jaw lesions, 95 Rasping period, snail, 57 Rasping radula, see Radula Rat calvaria, metabolic action, 589 Rat molars demineralization tests, 138 etching, 139 Rat teeth, chemical erosion, 100 Rate of calcification, corals, 50 Rate of formation, incipient carious le- sions, 253 Rate of penetration, shell, 59 Rate of reaction, enamel dissolution, 236 Rate of retention, bone-seeking iso- topes, 416 Rate of resorption bone, 457 collagen, 673 Rate of synthesis, collagen, 673 Ratios, solution to sohd, 254 Rats crania, 481 mutant strain, 565 osteoporosis, 433 parathyroid extract, 532 Reaction products calcium fluoride, 257 dicalcium phosphate, 257 enamel dissolution, 250 hydroxvapatite surface, 254 transport, 257 Reaction rate, enamel dissolution, 236 Reactive form, bone mineral, 381 Reactivity, bone mineral, 381 Reconstituted collagen gels, 664 proteases, 685 Reconstitution, enamel surface, 145 Recrvstallization CaHPO^, 206 caries process, 206 enamel caries, 191 Red deer, antlers, 347, 352 Red Sea, coral reefs, 50 Redeposition of minerals, enamel, 149 Reef, see Coral reefs 754 SUBJECT INDEX Reef coral communities, 25, 39 Reef corals see also Corals attachment, 39, 45 colony form, 39 light intensity, 51 Reef erosion, 47 Reef limestone, 27 Reef sediments, 27 Reef silt, 37 Reef subsidence, 47 Regressive inflammation, gingival bi- opsies, 305 Regulator genes, resorptive stimuli, 491 Rehardened enamel, dye-binding ca- pacity, 139 Reindeer antlers, 347, 353 thyroxin injections, 361 Remineralization, 137, 149 bone, 519 enamel caries, 176 Remineralization of enamel, 130 theoretical basis for, 140 Remineralizing salt mixture, 143 Remodeling of bone, 371 compact bone, 371 jaws, 322, 323 mammalian jaws, 322 osteogenesis, 321 summarv, 381 Remodeling of mesenchyme, 689 Remodeling svstem, metamorphosing tadpole, 663 Remote eff^ect, gingivitis, 300, 301 Renal failure, 456 bone resorption, 456 Renal handling of phosphate, 385 Repair of alveolar bone, 303 Replacement therapy, castrated deer, 356 Replicas, 81 enamel surface, 136 shell etching, 70, 72 Reprecipitation, enamel, 257 Reproductive cvcles, antler shedding, 363 ' ^ Resorbability, 97, 98, 148, 491 Resorbable tissues organic matter, 98 structural variations, 96, 144, 492 Resorbing bone surfaces, 94 Resorbing calvaria citrate versus lactate production, 623 collagen degradation, 624 hvdroxyproline, 626 vitamin A effect, 626 Resorbing cementum, 97 Resorbing dentin, 97 Resorbing edge, enzyme action, 505 Resorbing enamel, 96, 97 Resorbing surface osteoclast, 451 vttrium deposition, 451 Resorption accelerated, 488 alkahne earths, 452 alveolar bone, 95 alveolar septum, 300, 301 bone microradiography, 447 calcified cartilage, 484 calcified surface, 452 cartilage invasion, 483 cellular elements, 144, 472 cellular influence, 486 chemical inhibitors, 615 code information, 490 controlling mechanisms, 472 decalcification, 486 degree of calcification, 472, 483, 484 demineralization gradients, 146 diploic spaces, 483, 484 enamel, 144, 145, 489 endothelial cells, 490 functional stresses, 490 genetic cell potential, 489 giant cells, 515 hemopoietic tissue, 490 histocytes, 490 hvperparathyroid rats, 485 inhibitors, 609, 615 inorganic material, 570 interstitial basophilia, 545 SUBJECT INDEX iOO Resorption — Continued intralaciinar, 425 irregular surfaces, 455 jaw bones, 95 lamina dura, 300, 301, 307, 309 leucine aminopeptidase, 553 mesenchymal specializations, 490 mouse calvaria, 609 organic mass, 486 organic phase, 570 osteoblasts, 489 osteoclasts, 639; .sec also under Osteoclasts osteocytes, 489, 640 osteoprogenitor cells, 486 oxygen tension, 490 parathyroid extract, 488 parathyroid stimulation, 471 PAS staining, 486 periosteal surface, 450 predentin, 528 pressure, 490 root surface, 314 stimulators, 609 svstemic acceleration, 472 uncalcified matrix, 472 vitamin A, 490 Resorption cavities, 375 bonelike hard tissue, 314 evolution of skeleton, 371 hard tissues, 313 osteons, 457 subgingival caries, 313 Resorption processes, 92 Resorption rate, 491 cortical bone, 458 Resorption sites intercellular matrix, 472 skull, 485 Resorption surfaces Bodian's silver stain, 483 yttrium autoradiographs, 452 zones of action, 499 Resorption in tissue culture, summary, 633 Resorption without osteoclasts, 376, 531 Resorptive areas, osteoprogenitor cells, 487 Resorptive complex, 491 Resorptive fields, 491 Resorptive mechanisms, in vitro studies, 557 Resorptive potential bone cells, 490, 491 cellular microenvironment, 492 osteocytes, 489 Resorptive response, genetic loci, 490 Resorptive states, contradictory con- clusions, 471 Resorptive stimuli, 491 regulator genes, 491 Resting period of caries, 176 Restoration, cervical erosion, 119 Retained isotope, bone formation, 449 Reticulin, ground substance, 689 Retzius lines, 137, 173 birefringence, 177 dye penetration, 140 enamel demineralization, 134 subsurface changes, 137 Reversal cement lines, osteons, 399 Reverse relationship, erosion and ca- ries, 118 Rhesus monkey in vivo demineralization of enamel, 135 incisors, 134 radioactive phosphorus, 141 Rib shaft, osteogenic surfaces, 483 Ribosomes, osteoclasts, 499 Rickets alpharadiography, 541 microradiography, 464 osteocyte, 465 Ring compound, 638 Rocellaria, 14 cuneiformis, 15 Rock boring, 1 agents of erosion, 22 bone boring analogy, 372 chemical action, 5 chemical softening, 18 Rock-boring bivalves, specialization, 9 756 Rock-boring organisms, 1, 376 summary, 21 Rock erosion, 102, 103 Rocking action, boring organisms, 11, 17,22 Rodent skin, coUagen fractions, 677 Roe deer antlers, 341, 353 reproductive pattern, 359 Roentgenological examinations, 95 Romney sheep, 155 Root hairs of plants, soil salt solution, 639 Root surface resorption, 314 pericemental fibers, 315 Root tip resorption, 309 Rouget-Neumann sheaths, 425 Ruffled border, osteoclast, 499, 501, 506, 645 bone salt crystals, 508 collagen fibrils, 501 Rumen microflora, 161 Rumination, 156 Rut, efl^ect on antlers, 341, 348 Rye grass, 155, 166 Saclike vesicles, osteoclasts, 508 Saliva citrate content, 114 facsimile of ions, 143 lubricating action, 94, 112 miniature rivers, 100 mucous substance, 133, 140 newborn infants, 267 pH levels, 267 oxalate, 114 radioactivated, 141 Saliva and enamel, liquid-solid equilib- rium, 143 Saliva substrate, caries, 266 Salivary circulation, radioactive ele- ments, 141 Salivary citrate dental erosion, 114 fruit juices, 1 14 Salivary COo, dental plaque, 276 Salivary environment, 143 Salivary extracts, dental plaque, 655 SUBJECT INDEX Salivary flora, 115 Salivary glands accessory, snafl, 68 hibernation, 294 protease activity, 294 Salivarv mucin, 131 Salivary protease, hibernation, 294 Salivarv salt mixture, enamel reminer- alization, 143 Salivary secretion, systemic effect, 117 Salivary sediment, 652 Salivation, 305 Salt mixture, remineralizing, 143 Salt removal, hypertrophic osteocyte, 552 Sambar, antlers, 347, 350 Sampling method, bone turno\er, 453 Sand-burrowing sipunculids, 3 Sand-producing calcareous algae, 37 Saprophytic erosions, 91 Sarcinas, 270 Saxicavacea, 12 Schmutzpyorrhoe, 298 Sclerotic rings haversian canal, 449 osteons, 456 Scurvy, 95, 385 periodontal disorder, 312 Sea bass, bone resorption, 376 Sea urchins, 4 Secondary calcifications, 99 Secondarv cementum, 96 Secondary dentin, 156, 159, 187 Secondarv hvperparath\roidism, 434 Secondarv proteolvsis in caries, 207, 657 in resorption, 505 Secreted calcareous lining, 21 Secretion calcium carbonate, 15, 77 mucus, bivalves, 12 siphonal tissues, 15 Secretion of ABO gland, 60 absence of acid, 81 Secretory cap, ABO gland, 67, 83 Secretory disk, ABO gland, 68 Secretory epithelium, 57, 83 Secretory globules, ABO gland, 68 SUBJECT INDEX Sedentary life, osteoporosis, 433 Sediment coral reef, 47 erosion of corals, 46 Sedimentary rock, 1 Selective demineralization, enamel caries, 175 Selenium deficiency, paradontal dis- ease, 161 Semiquantitative analysis, microradio- graphs, 467 Senile bone matrix, 432 Senile osteoporosis, sex hormones, 427 Sequestering agents, 55, 638 sec also Chelation Sequestrene, 81 Serrated edges, crystals, 75 Serum citric acid, 390 Setous thoracic appendages, barnacles, 7 Severe osteoporosis case studies, 413 trabecular bone, 400 Sex factor, osteoporosis, 418 Sex hormones antler shedding, 353, 354 osteoporosis, 387 senile osteoporosis, 427 Sex influence, antler shedding, 353 Shadowed replicas, 65 Shedding of antlers, see Antler shed- ding Shedding mechanism, 92, 340 denervated antlers, 360 Shedding of teeth, 91, 92, 95, 97, 321 bone remodeling, 321, 331, 332 summary, 336 Sheep calcium requirements, 161 incisor teeth, 155 jaw defects, 157 phosphorus requirements, 161 Sheep's teeth attrition, see Attrition of sheep's teeth chemical composition, 163 excessive wear, 165 fluorine content, 165 757 hardness, 157, 167 proteolysis, 94 pulp canal, 156 spectrographic analyses, 161 Shell boring ABO gland, 65, 75 algae, 55 bacteria, 55 fungi, 55 rate of penetration, 59 snails, 55, 58, 59 Shell dissolution, 69 ABO homogenates, 64 chemical means, 55 Shell drilling, see Shell boring Shell etching, 31, 65, 66, 67, 86 ABO homogenate, 65 ABO secretion, 62, 83, 75 ABO-substrate preparations, 63 crystal outlines, 73 electron microscopy, 71, 75 excised ABO gland, 62 pattern, 75 pseudo replicas, 73, 77, 81 replicas, 70, 72 variability, 68, 85 Shell of gastropod prey, 86 artificial etching, see Shell etching bore holes, see Bore holes chemical agents, effect, 75 demineralization, 79 gloss, 60 Shell morphology, ABO activity, 69 Shell protection, periostracum, 20 Shell substrates, 62, 85 Shell valves abrasive action, 11 agents of boring, 8 articulating processes, 10 erosion, 9, l7 mechanical action, 8 rocking, 11 rows of teeth, 10 Shortened lower jaw, sheep, 159 Shrinking osteocytes, 400 Shunt activity, hexose monophosphate, 579 Sialic acid, dental plaque, 657 758 Siderastrea, 39 Sika deer, antlers, 340, 343, 347, 354, 356 Siliceous bodies, sponge, 32 Siliceous spicules, clionids, 30 Silicon, 161 worn sheep teeth, 165 Siniim per.spcctivum, 61, 67, 73, 86 Siphonal tissues, bivalves calcium carbonate secretion, 15 unicellular algae, 16 Sipunculid worms, 3 bristles, 56 Skeletal carbonates, 25 coral reefs, 49 Skeletal debris, coral reefs, 46, 47 Skeletal diseases, 385, 453 summarv, 438 Skeletal fractures, 431 Skeletal kinetics, 430 Skeletogenesis, corals, light intensit\ , 51 Skeleton metabolic diseases, 453 tvn-no\er, 453 Skull resorptive sites, 485 trabecula, 547 Small osteocytes, 549 Snail, 57 accessorv boring organ, sec ABO boring mechanism, 55, 86 chelation, 647 demineralization mechanism, 55, 56 demineralization organ, see ABO drilling site, 57 metabolite signal, 57 muricid, 56 naticid, 56 radula, 4, 56, 57, 81, 83 radular denticles, 57, 60, 71, 81,. 86 shell boring, 55, 58, 59 species, 61 tentacles, 63 Snail boring chemical action, 57 living prey, 71 Snail boring organ, see ABO SUBJECT INDEX Snail drilling, see Snail boring Squirrel, Arctic ground, 286 Societv for Crippled Children, 440 Sodium versenate, 75 Soft drinks, 99 Soft x-rays, 474 Soft-shelled clam, 9 Soil salt solution, root hairs of plants, 639 Sole of foot, muricids, 56 Solubility apatite, 276 bone, 637 bone mineral, 563, 569 enamel, 138, 255 Solubility of minerals, changes in pH, 584 Solubility product calcium fluoride, 256 measurements, 256 Solubility rates, enamel, 255 Soluble collagen, irradiation, 425 Soluble matrix, enamel caries, 184 Solution of collagen, 503 Solution to solid, ratios, 254 Solvent action, pus, 647 Sound dentin, collagenous fibrils, 200, 201 South America, seasons of antler shed- ding, 349 South American deer, 350 Space travel, hibernation, 285 Specialization, rock-boring bivalves, 8 Specific activity collagen, 675 collagen fractions, 678 Specificity, caries initiation, 272 Spectrographic analyses, sheep's teeth, 161 Spectrum of dento-alveolar destruc- tion, 92 Spectrum of \ertebrate hard tissues, 91 Spermatogenesis, antler shedding, 359 Spinal osteoph\tes, 436 Spine, crush fractures, 454 Spirasters, 30 Spisula solidissima, 62, 75, 86 ♦H SUBJECT INDEX Spondylosis, 436 Sponge, 1, 55 boring activities, 27, 33, 45, 47 calcareous matter, 35 carbonic acid production, 35 chemical action, 27 chemical dissolution, 35 erosion of shell, 31 erosion of coral, 37, 39, 45, 50 erosional remodeling, 46 flagella, 5 proteolytic enzyme, 33 protoplasmic processes, 33 jorotoplasmic strands, 33 siliceous bodies, 32 Sponge burrows, deep-water corals, 45 Sponge cells, mechanical action, 33 Sponge-encrusted corals, 42 Sponge fragments, pseudopodia, 31 Sponge larvae, 31 Spontaneous fracture, 395, 417, 419 parathyroid adenoma, 549 Spotted deer, antlers, 350 Stable bone, 471 Stable collagen, C^ ^-labeled proline, 580 Stable phase, apatite, 256 Stainability, enamel, 137 Stained plaque, 289 Steady state, calcium, 564 Steatorrhea, 426 Stephanocoenia, 39 Stereoscopic projection, microradiog- raphy, 69 Storage of calcium, structural bone, 430 Streptococcal growth, carbon dioxide, 275 Streptococci cariogenic potential, 267 hamster caries, 270 inorganic ion requirements, 277 nutritional requirements, 275 Streptococcus fccalis, 268 lactis, 268 Stress, hibernation, 286 Striae of Retzius, see Retzius lines 759 Striated border, osteoclast, 498, 499 Strontium reabsorption, kidney tu- bules, 407 Strontium retention studies, 448 Structural bone, 372, 379, 471 osteoporosis, 414 storage of calcium, 430 Structural genes, microenvironmental modifications, 491 Structural variations, resorbable tis- sues, 144, 145, 492 Structure-function relationships, osteo- clast, 497 Subgingival caries, resorption cavities, 313 Subjacent matrix, demineraHzed col- lagen, 503 Submicroscopic mineralization, 143 Submicroscopic spaces, enamel caries, 179 Subsidence and slump, coral reefs, 47 Substrate utilization, bone, 599 Subsurface changes dental erosion, 131 Retzius lines, 137 Subsurface demineralization, 98, 131, 133, 216, 220 interprismatic regions, 139 shell bore holes, 69 Subsurface enamel components, 216 Subterranean saprophytes, 92 Successive sets of antlers, 348 Succinate, oxygen uptake, 593 Succinic acid, 167 Succinic dehydrogenase Havoprotein system, 593 osteoclasts, 614 Suckerhke foot, bivalves, 14 Sucrose, chelation, 655 Summary alveolar bone resorption, 315 antler shedding, 363 attrition of sheep's teeth, 168 bone cells and bone resorption, 491 bone implants, 529 bone metabolism, 605 bone metabolism in vitro, 573, 587 bone remodeling, 381, 467 760 Summary — Continued bone turnovei' measurement, 453 caries pathology, 147, 185 chelation theories, 657 chemistry of caries, 207, 257 collagen remodeling, 689 compact bone remodeling, 381 coral reef remodeling, 50 dental erosion, 145 dento-alveolar resorption, 315 enamel caries, 185 enamel dissolution, 257 eruption of teeth, 336 gastropod boring mechanism, 86 giant cells, 529 hibernation and teeth, 295 histophysics of bone resorption, 491 microradiography of bone, 467 osteoclasts, 512, 529 osteolysis, 553 osteoporosis, 438 parathyroid action, 587, 605 periodontal disorders, 315 proteolysis, 689 resorption, 315, 467, 491, 529 rock-boring organisms, 21 shedding of antlers, 363 shedding of teeth, 336 skeletal diseases, 438 tissue culture resorption, 633 transmission of caries, 278 ultrastructure of caries, 207 Supporting tissues of teeth, 94 Supra-alveolar gingiva, 308 Supravital staining, bone resorption, 474 Surface demineralization, dental ero- sion, 108 Surface erosion enamel crystallites, 191, 194, 195 exhumed teeth, 125 Surface exchange, 637 Surface layer, carious dentin, 205 Surface replicas, shell boring, 69 Swamp deer, antlers, 349 Symbiotic relation, corals and algae, 3 Synthesis, zymogen molecule, 687 Synthetic fluorapatite, 256 SUBJECT INDEX Synthetic hydroxyapatite, 254 Systematics-Ecology Program, Marine Biological Laboratory, 87 Systemic acceleration, resorption, 472 Systemic conditions, alveolar bone loss, 310 Systemic effect, salivary secretion, 117 Tadpole, metamorphosing, 663 Tadpole tissues, collagenolytic proper- ties, 682 Tadpoles, thyroxin-treated, 665 Talus fallout, coral reefs, 46 Talus formation, coral reefs, 46 Target cells, parathyroid, 632 Tartar, 298; see also Calculus TCA cycle bone, 596 oxygen uptake, 596 Teeth boring canals, see Exhumed teeth effect of hibernation, see under Hi- bernation migration, 297, 308, 309, 326 mobility, 297 postmortem destruction, 148; see also Exhumed teeth radioactive phosphorus, 140 radular, snail, see Radular denticles Teeth of sheep, see Sheep's teeth Teeth and spines, sea urchins, 4 Teeth versus host, 264 Teleost, bone remodeling, 376 Telescoping of torso, 404 Teredinidae, 1, 10 Terrestrial mammals, collagen, 674 Testosterone antler shedding, 354, 357 ossification, 355 secretion, 341 shedding of antler velvet, 348, 356 Testosterone phenvlacetate, 356 Testosterone propionate, 360 Tetany, bone citrate, 643 Tetracycline, bone formation, 375, 393, 409, 448, 457 Thaididae, 56 SUBJECT INDEX 761 Theelin, 356 Therapeutic hydrochloric acid, enamel erosion, 99 Therapeutic implications, animal ca- ries, 277 Three-phase system, in vitro model, bone metabolism, 567, 570 Thijnnus tlnjnniis, 376 Thyroid compounds protein synthesis, 679 tadpole metamorphosis, 663 Thyroid gland, antler growth, 360 Thyroid histology, Virginia deer, 360 Thyroidectomy, antler growth, 361 Thyrotoxicosis, 430 Thyroxin injections antler growth, 361 reindeer, 361 Thyroxin-treated tadpoles, 665 Tibia, funnel shape, 481 Tin fluoride, erosion therapy, 118 Tissue culture, 94 bone resorption, 609 collagen microassay, 682 fowl bone, 497 osteoclasts, 511 vitamin A, 628 «, Tissue culture calcium, parathyroid ex- tract, 623 Tissue cidture resorption, summary, 633 Tissue culture system, parath)roid as- say, 621 ^ Tissue extracts, collagenolytic activity, 664 *^ Tissue reactions, functional stress, 308 Tissue storage, collagenase, 687 Tissue tonus bone, 298 periodontium, 305 Tissue water loss enzyme-producing cells, 690 tadpole metamorphosis, 665 Tissues chelating agents in, 637 resorbability, 98, 144, 492 Toby- Jug Stone — Kannesteinen, Nor- way, 103 Tomograms, 422 Tooth brushing, 101 etiology of erosion, 102 Tooth-brushing machine, 108, 116 Tooth development, 322 hibernation, 285 Tooth eruption bone remodeling, 321 functional adaptation, 325 parathyroid hormone, 330 tooth resorption, 313 Tooth germ, 322 ankylosis, 328 growth period, 323 migration, 325 Tooth mineral, etching, 62, 79 Tooth mobility, 310 percussion dullness, 310 Tooth resorption circulatory disturbances, 315 eruption process, 313 epulis tumors, 312 gingival granulation tissue, 313 hyperplastic gingivitis, 312 periodontal disorders, 297 Tooth structure erosion. 111 hibernation, 285 Tooth versus bone resorption, 315 Topographic configuration, dental ero- sion, 101 Torpor, hibernation, 285 Total body counting, radioassay, 393 Total bone mass, 414, 430 Total collagen, metabolism, 670 TPN (NADP)-linked reactions, 640 Trabecular bone lacunar resorption, 551 severe osteoporosis, 400 Trabecular markings, 420 Trabecular resorption, 479 Trabeculation, 292 Trace elements carious dentin, 205 carious enamel, 201 Tracer dilution formulas, bone mass", 430 762 Tracer techniques, turnover of bone, 430 Tranquilizers, osteoporosis treatment, 436 Transamination reactions, a-ketoglu- tarate, 599 Translocated crystals, 503 Translucent zone, enamel caries, 172 Transmission of caries, 261, 270 summary, 278 Transmission of cariogenic flora, 268 Transport mechanisms, phosphate, 604 Transport of reaction products, enamel caries, 257 Transudate, bone marrow, 300 Traumatic occlusion, 92, 97, 112 Triamcinolone, 399 a-Tricalcium phosphate, 381 Tricarboxylic acids, 639 Tridacna crocca, 7, 16 dcresa, 16 Tridacnidae. 16 Trifolium repens, 155 Trigger mechanism, alveolar bone, 299 Triiodothyronine, 425 Tropical deer, 347 Tropical reef communities, 25 Tropism, implants, 528 Tro^oocollagen, 668 Tropocollagen molecule, 667 Trypan blue giant cells, 520 osteoclasts, 520 Trypsin, collagen digestion, 685 Tube sponges, 44 Tube worms, 39 Tubular reabsorption, phosphate, 427 Tukon Hardness Tester, 156 Turbellaria, 1 Turbo, 4, 81 Turnover bone, 447, 453, 460 collagen, 425 plasma calcium, 376 SUBJECT INDEX tracer technique, 430 Twitching, oral soft tissues, 100 Tylostyles, 30 Ultrafiltrable hydroxyproline, collagen- ase-like activity, 572 Ultrasoft x-rays, 474 Ultrastructural topography, shell etch- ings, 85, 86 Ultrastructure brush border, 499 caries, 187, 190, 207 carious dentin, 191 osteoclasts, 497 Unavailable bone, 471 Uncalcified matrix, resorption, 472 Uncoupling, oxidative phosphoryla- tion, 600 Underwater landslides, 46 Undissociated acid, enamel dissolu- tion, 241 Unicellular algae, 29 in siphonal tissues, 17 Unifying hypothesis, osteoporosis, 439 Unit area of bone, 460 United States Air Force, 296 United States Atomic Energy Com- mission, 439, 554 United States Fish and W'ildlife Ser\- ice, 87 United States Navy Office of Naval Research, 52 United States Public Health Service, 149, 208, 364, 382, 439, 492, 529, 587, 634 Univalve, 62 Unremodeled bone, osteogenesis im- perfecta, 462 Urinary excretion curves, 407 Urinary hydroxyproline, 425 Urinary indices osteoporosis, 417 physiologic aging, 417 Urinary products, bone breakdown, 4i7, 438 Urosalpinx cine tea foUycusis, 59, 61, 63, 69, 73, 75, 77, 83, 85, 86, 87 SUBJECT INDEX 763 Vacuoles, osteoclasts, 613 Valve edges, pelecypods, 56 \'^alves, shell, see Shell valves Variable fraction, citrate, 644 Vascular invasion, cartilage, 481 Vascular spaces, parietal bones, 483 Vascularized granulation tissue, 97 Velvet shedding, see Antler velvet shedding Veneracea, 12 Venezuelan deer, 348 Venoclysis, EDTA, 545 Versene, 65, 81 effect on shells, 75, 79 Vertebrae, collapsed, 415 Vertebral density, relative, 422 Vertebrate hard tissues, spectrum of, 91 Vesicular sacs, collagen, 501 * Vestibular shield, dental erosion con- trol, 120 Virginia deer antlers, 343, 347, 352, 354 thvroid histologv, 360 \'itamin A acid phosphatase loss, 628 bone resorption, 490, 628 cartilage matrix, 628 direct bone effect, 627 induced resorption, 633 and oxvgen tension, interdepend- ence, 628 protease release, 628 resorbing calvaria, 626 tissue culture, 628 \'itamin D bone destruction, 428 bone metabolism, 564 bone remodeling, 377 bone resorption, 629 citrate svnthesis, 640 direct action, 631 hypothesis on action, 644 induced resorption, 633 mineral solubility, 564 osteoporosis, 433 parathvroid hormone, 630 \^itamin D-resistaat rickets, 466 \'omiting, dental erosion, 99 \ on Gierke's disease, 386 Wapiti, antlers, 352 Wasting of tooth substance, 101 Water pressure, mantle cavitv, 18 Wave attrition, 25 Wave erosion, 25 Wave surge, 37 Wave turbulence, 27 Wear of sheep's teeth, see Attrition of sheep's teeth Wedge-shaped erosion of teeth, 102, 104 restoration, 119 ^^'ed^s canals, exhumed teeth, 128 Weight loss, osteoporosis, 431 \\'hite clover, 155 White Leghorn hen, osteoporosis, 432 White spot, incipient enamel caries, 213, 215 Whole-body counting facility, 392 Wood-boring organisms, 1 Woods Hole Marine Biological Labo- ratory, 87 Worms, 1, 25, 42 mucous tubes, 6 Worn teeth, sheep erosive agents, 166 silicon content, 165 Woven bone, 450, 477 basophilia, 477 organic mass content, 479 osteocytes, 465, 477 Xerostomia, 94 X-ray diffraction, 62 osteoporosis, 424 X-rav microradiography, bone, 535 Xylophaginidae, 10 Yttrium autoradiographs of resorption sur- faces, 452 bone remodeling, 450 bone resorption, 451 764 Yttrium — Coitiniicd titrif acid, 451 oxtnnal coiiiitiny;, 152 niicroradiograpliN of bono, 451 paralliMoid lionnoiie, 451 pi 1 rclatioiisliip, 451 rcsoihiiiti; snriat'cs, 451, 452 Zinc, carious dculiu, 203 SUBJECT INDEX ZirpliaccL 10, 11 Zoua fasciculata, 429 Zoua glomerulosa, antler growth, 361 Zone of partial dcmincrali/ation, shell drilling, 58, 59 Zones, resorption surfaces, 499 Zooplankton organisms, 7 Zooxanthellae, 1?, 39 Zymogen molecule, s\ulhcsis, 687 X'/''