Skip to main content

Full text of "Proceedings of the Society for Experimental Biology and Medicine"

See other formats

This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project 
to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 
to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 
are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 
publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 

We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liability can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 

at |http : //books . google . com/ 

+ -I- f -I- 


+ ■♦- + + 

1 M 





+ + + + H 

1 M I 



+ + + + 4 






+ + + 4- H 

-I- ^ + -f ^ 





+ + + + 



agry Qfr 




+ + -♦■ 















,ITY O/. 






-I- + + 




+ + + ^ 




this animal 
comes with 

1 aHmtitmr eutb 2 Orbftal Enucfaoiion 

^3 Plnoalectomy 

_4 HypoptiyHictOfTiy 

10 Adrenaloclomy 
11 I 

12 Ovariectomy 
13 OophorttysterectCNiiy 
14 HysMfttdoniy 

ISUraterUgition 16¥Mec to my 17CMtratton 

(17 routine surgical procedures) 

'he number of routine surgical rrxxji- 
'cations we perform has grown C 1 7 
procedures) Many other non-routine 
procedures can also be done to your 
speafications All surgery is performed 
or our premises to retain control of 
3^ environmentai factors 

^0 be sure of accurate and consistent 
^ta always start with a dependable. 
'^thy animal that has consistertt 

'^ invite you to compare the quality of 
ouf arvimals and our services with your 
resent supplier Whether you want 
animals that come with, or without", 
you can be sure you'll be getting the 
tiest ammal for your needs when you 
aarr with laconic 


Olfactory Bulb 

Orbital Enucleation 

Salivary Duct 


Ureter Ligation 

D Yes. I am interested in your animals 
U Yes. I am interested in your procedures 
D Rease have someone call me. 












Gemiantown, New York 12526 
518537 6206 

All Surgically Modified animals are packed 
in Disposable Filtered Shipping Containers 
equipped with TRANSIT KITS ' 

* a disposable A-ater sysiem composed ol a wvater 
tilled plastic pouch reservcxr and automatic dnnK- 
■ng valves (Oirncai additives added m percentage 
required deperxjing on dictates of rrxxlitication 
performed ) 

Council 1978-79 

President, DeWitt Stetten, Jr. 
National Institutes of Health 

President' Elect, Robert W. Berliner 
Yale University 

Past President, Dennis W. Watson 
University of Minnesota 

Treasurer, Gregory W. Siskind 
Cornell Medical Center 

Secretary and Ass't Treasurer, Mero R. Nocenti 
Columbia University 


Univ. of Kansas 

A. H. Briggs '81 
University of Texas 

H. F. DeLuca '81 
University of Wisconsin 

P. P. FoA '79 
Sinai Hosp. of Detroit 

I. J. Fox '81 
Univ. of Minnesota 

J. P. Gilmore'81 
Univ. of Nebraska 

M. Orsini '81 
Univ. of Wisconsin 

R. J. Peanasky '79 
Univ. of South Dakota 

E. E. Selkurt '81 
Univ. of Indiana 


Univ. of California 


Cornell Univ. 


^'Membership in the Society for Experimental Biology and Medicine is open to all individuals who 
have independently published original meritorious investigations in experimental biology or experimen- 
tal medicine and who are actively engaged in experimental research. In general, applicants should be 
beyond a supervised post-doctoral experience in order to be able to demonstrate the ability to conduct 
independent investigations. 

Application forms may be obtained from the Office of the Secretary, Society for Experimental 
Biology and Medicine, 630 W. 168lh St., N.Y., N.Y. 10032." 

80 2 I 

21; 013 C 



Board of Editors 

M. R. 


Managing Editor 

630 W. 

168th Street 

New York, N. Y. 10032 



R. R. Gala 


R. B. Roberts 

SON, Jr. 

R. C. Gallo 

C. W. Lloyd 

R. Ross 


Z. N. Gaut 




G. L. Gebber 




J. Genest 

G. J. MacDonald 

W. Sawyer 

f. Bach 

D. G. Gilmour 

D. F. Malamud 

B. B. Saxena 



L Mandl 

A. J. Sbarra 


M. Greenwood 

A. J. Marcus 




A. Mazur 



N. S. Halmi 

S. M. McCann 



C. G. Harford 

L. C. McLaren 

N. J. Schmidt 


P. C. Harpel 

J. Meites 

H. A. Schneider 



T. C. Merigan, Jr. 

E. E. Sfikurt 



F. H. Meyers 

J. H. Shaw 


J. J. Holland 

F. N. Miller 

E. M. Shevach 




N. Shock 



C. R. Morgan 

M. M. SiGEL 



S. L Morse 



E. D. Jacobson 


N. E. Sladek 


H. D. Janowitz 


A. A. Spector 


D. C. Johnson 

D. Nathan 

R. S. Spiers 


R. C. Johnson 


J. G. Stevens 


T. J. Kindt 

S. Oparil 

A. Stracher 


S. Klahr 

P. Y. Paterson 

E. D. Thomas 



P. N. Patil 

G. J. Thorbecke 


C. A. Krakower 

W. E. Paul 

M. L. Tyan 


L. C. Krey 

M. J. Peach 

J. L. Vaitukaitis 



V. A. Pedrini 

C. M. Veneziale 

P. L. LaCelle 

G. L. Plaa 

C. S. Vestling 


B. N. LaDu 

S. A. Plotkin 

S. R. Wagle 


M. E. Lamm 

D. D. Porter 

M. E. Weksler 


C. A. Lang 

A. S. Rabson 

J. M. Weller 


J. H. Laragh 

J. A. Ramaley 

R. M. Welsh 


C. Lenfant 

M. M. Rapport 



C. E. Leroy 

W. D. Reid 

E. E. Windhager 


R. Levere 

J. A. Resko 



J. V. Levy 



Editorial and Publications Committee 

M. ZucKER, '82, Chairperson; L Clark, '80; M. Hilleman, '82; 

S. L Morse, '78; S. Seifter, '82. 

The President, President- Elect and Secretary 


Volume 159, Number 1, October 1978 


Copyright © 1978 by the Society for ExperimenUl Biology and Medidne 
All Rights Reserved 

No part of this publication may be reproduced or transmitted in any form or by any means, etoctniaic 
or mechanical, including photocopy, recording, or any information storage and retrieval system, withoat 
permission in writing from the copyright owner. 

The appearance of the code at the bottom of the first page of an article in this journal mdicaiet At ' 
copyright ownefs consent that copies of the article may be made for personal or internal uae, or te Ar^ - 
personal or internal use of specific clients. This consent is given on the condition, however, that the copiir:^ 
pay the stated per copy fee through the Copyright Clearance Center, Inc. (Operations Staff, P.O. BOK 7iS«/ 
Schenectady, New York 12301) for copying beyond that permitted by Sections 107 or 108 of the UJL '; 
Copyright Law. This consent does not extend to other kinds of copying, such as copying tot gOMnlC 
distribution, for advertising or promotional purposes, for creating new collective works, or for fcsalc Ccpf ^^ 
fees for pre- 1978 articles are the same as those shown for current articles. 

Proceedings of The Society for Experimental Biology and Medicine. Vol. 159. No. 1, October 1978. Published monlhly 
excepl August by Academic Press, Inc., Ill Fifth Avenue, New York, N. Y. 10003. Second class postage paid at New 
York. N. Y. and at additional mailing offices. 1978: Subscription per year $48.00 U.S.A.; $62.00 outside U.S.A. All prices 
include postage. Send notices of change of address to the Office of the Publisher at least 6-8 weeks in advance. Please 
include both the old and new addresses. Copyright © 1978 by the Society for Experimental Biology and Medicine. 

Members can help considerably by mentioning our Proceedings 
when communicating with or ordering supplies from our advertisers. 



Osborne / Mendel 



Certified Pathogen Free 

Caesarean derived • Barrier reared 

jperlative is the word that best describes the 
?w CAMM Certified Pathogen Free Rats 

id you can now get them in Sprague-Dawley. 
sborne Mendel or Wistar strains. 

esearchers and scientists from facilities all 
ound the country have been writing to us 
lling us how pleased they are with these new 
AMM rats. They are available now. Ready for 
lipment today by company trucks or Direct 
oute Air Express. Please contact us with your 
quirements. Ask for our price list. 


By Company Truck and 

Direct Route Air Express 



414 Black Oak Ridge Road, Wayne. New Jersey 07470 



If you are having diffi- 
culty growing cells in Tissue 
Culture, our Growth Factors 
will help you. They in- 
crease survival times, 
accelerate growth 
rates and improve 
plating efficiency. In 
many instances, they 
also affect DNA. RNA 
and protein synthesis rates 
and a host of transport 
mechanisms. Collaborative 
Research Growth Factors 
now available from stock 

Epidermal Growth Factor (CR-EGF) 

Fibroblast Growth Factor (CR-FGF) 

Multiplication Stimulating Activity (CR-MSA) 

Nerve Growth Factor (CR-NGF) 

Endothelial Cell Growth Supplement (CR-ECGS) 

Human Thrombin (CR-HT) 

In addition, we have [^"1] labelled derivatives and antisera for EGF. 
FGF, MSA and NGF. 

To find out more about our Growth Factors and what they can mean 
to your research, call or write for our free technical bulletins. Or for im- 
mediate delivery of these products call our Order Department 61 7-899-1 133. 

For Research use Only 

Research Products Division 

Collaborative Research, mc 

1365 Mam Street. Waltham. Mass 02154 
Tel (617) 899-1133 TWX 710-324-7609 

Members can help considerably by mentioning our Proceedings 
when communicating with or ordering supplies from our advertisers. 


for the most demanding 

■ Superior t)uCib(e cleaf<ng abitiiy a rtjmoiv ■ . ; 
he\d protit H<lh pushbutton switch aid spet.i 
Tfon. afe advanEages no( Shaded by other moiuri^i;; 
dJikjiers » Accuracy V'= reproduceabiUty 2^. - 
Sup&rfor chpmico" resistance Reagents c,or\^sc' 
T^it„« kf-i p and sapphire ■ Vofyme dI ddyeni anj 
MTTippt' <irt jrtdependenTJy wT by means of digital 
micrcimeters readfng directly m volume -Range ^^ I to 
& nnl - DilyTron ratio <5 cor^linuoUSty adjustable trom 
?500 |,^^^ a,t,i.i,..r,Hii .nrL.fH,al„.,f, .:a|i ..f ikriic 



Members and subscribers are requested to cooperate with our advertisers 
Camm Research Inst v 

Charles River Cover 4 

Collaborative Research vi 

Gilson Medical Electronics vii 

Taconic Farms i 


Annual Report b 

Guest Editors Report bit 


Competition Binding Assay Using o-Methyl (^Hl-Demethyl-y 

Amanitin for Study of RNA Polymerase B G. M. Garrity, A. Brown 


Kinetics of Testosterone Induced-Cholesterol Synthesis in Rat A. K. Singhal, D. P. Bonner, C. P. 

Ventral Prostate Schaffner 

The Role of Cyclic AMP in CRF-Induced ACTH Secretion R. Portanova, W. J. Brattin 

Ketamine as an Anesthetic for Obtaining Plasma for Rat Pro- H. Y. Meltzer. D. Stanisic, M. Simo- 

lactin Assays novic. V. S. Fang 
Effects of Thyroxine, Epinephrine and Cold Exposure on Li- 

polysis in Genetically Obese (ob/ob) Mice S. W. Thenen, R. H. Carr i 
Apomorphine-Induced Inhibition of Episodic LH Release in 

Ovariectomized Rau with Complete Hypothalamic Deaf- 

ferentation G. W. Arendash, R. V. Gallo 1 
Does Time of Exposure to Estradiol and LHRH Effect LH 

Release from Bovine Pituitary Cells? V. Padmanabhan, E. M. Convey 1 

Effects of Administration of a LH-RH Inhibitory Analogue on J. A. Vilchez-Martinez, E. Pedroza, D. 

Stages of the Rat Estrous Cycle H. Coy, A. Arimura, A. V. Schally i 


Suppressed Dietary Inducibility of Glucose 6-Phosphate De- 
hydrogenase and Elevated Cyclic AMP in Acute Hepatic K. Taketa, A. Watanabe, M. Ueda, M. 
Injury Kobayashi I 


Superoxide Dismutase in Bovine Fetal Ductus Arteriosus, Tho- 
racic Aorta, and Pulmonary and Umbilical Arteries P. D. Frazer, F. O. Brady 

Polybrominated Biphenyls in Chicken Eggs vs. Hatchability D. Polin, R. K. Ringer I 

Blood Volume Changes during the First Week after Birth in 

the Beagle and Pig S. I. Deavers, R. A. Huggins, H.-P. Sheng I 


Shape Change and the Percentage of Sialic Acid Removed by 

Neuraminidase from Human Platelets E. I. Peerschke, M. B? Zucker 

Stimulation of Erythropoietin Secretion by Single Amino Acids A. Anagnostou, S. G. Schade, W. Fried 


Enhanced Granulocyte Mobility Induced by Chemotactic Fac- T. Tono-oka, M. Nakayama, S. Matsu- 
tor in the Agarose Plate moto 

Protein-Calorie Malnutrition Impairs the Anti-Viral Function 

of Macrophages L. C. Olson, D. R. Sisk, E. Izsak 

Exometabolites of Leishmania donovani Promastigotes. I. Iso- 
lation and Initial Characterization L. H. Semprevivo 

Table of Contents ix 


Adsorption to Clostridium botulinum Cultures of Phage Con- 
trolling Type C Botulinum Toxin Production K. Oguma, H. Sugiyama 61 

Endotoxin Induced Metabolic Alterations in BCG Infected 

(Hyperreactive) Mice V. C. Senterfitt, J. W. Shands, Jr 69 

The Effect of Leukocyte Hydrolases on Bacteria. XI. Lysis by 
Leukocyte Extracts and by Myeloperoxidase of a Staphy- 
lococcus aureus Mutant Which is Deficient in Teichoic 

Acid, and the Inhibition of Bacteriolysis by Lipoteichoic M. N. Sela, I. Ofek, M. Lahav, I. Gins- 
Acid BURG 126 


Thymidine Kinase and DNA Polymerase Activity in Normal 

and Zinc Deficient Developing Rat Embryos J. R. Duncan, L. S. Hurley 39 

L-Histidine-Induced Hypercholesterolemia: Characteristics of 

Cholesterol Biosynthesis in Rat Livers J. K. Solomon, R. L. Geison 44 

L-Histidine-Induced Facilitation of Cholesterol Biosynthesis in A. A. Qureshi, J. K. Solomon, B. Eichel- 

Rats MAN 57 


Suppression of Chemical (DEN) Carcinogenesis in SWR/J 
Mice by Goat Antibodies Against Endogenous Murine 
Leukemia Viruses R. Pottathil, R. J. Huebner, H. Meier 65 

Meubolism of Acyclic and Cyclic A^-Nitrosamines by Cultured 

Human Colon H. Autrup, C. C. Harris, B. F. Trump 111 

Ornithine Decarboxylase Activity in Cells Acutely and Chron- 
ically Transformed by Murine Sarcoma Virus L. J. Kilton, A. F. Gazdar 142 


Pyrazinoic Acid and Urate Transport in the Rat S. J. Frankfurt, E. J. Weinman 16 

Tlte Effects of Indomethacin and Meclofenamate on Estrogen D. Mueller, B. Stoehr, Jr., T. Phernet- 

Induced Vasodilation in the Rabbit Uterus ton, J. H. G. Rankin 25 

Effect of Cholera Toxin on Renal Tubular Reabsorption of R. M. Friedler, S. Tuma, A. Koffler, S. 

Glucose and Bicarbonate G. Massry 48 

Hypophysectomy Alters the Diurnal Food Intake Patterns in 

Rats L. L. Bellinger, V. E. Mendel 80 

Intraerythrocyte pH and Physiochemical Homogeneity J. Warth, J. F. Desforges 136 


The Effect of Heparin on Growth of Manunalian Cells in Vitro T. K. Yang, H. M. Jenkin 88 


I^ecreased Antiviral Effect of Phosphonoacetic Acid on the 

Poikilothermic Herpesvirus of Channel Catfish Disease R. W. Koment, H. Haines 21 

Mouse Hepatitis Vims (MHV) Infection in Thymectomized 

CaH Mice P. Sheets, K. V. Shah, F. B. Bang 34 

taunune Interferon Activates Cells More Slowly Than Does F. Dianzani, L. Salter, W. R. Fleisch- 

Vinis-Induced Interferon mann, Jr., M. Zucca 94 


s of Testosterone Induced-Cholesterol Synthesis In Rat Ventral Prostate (40271) 


n Institute of Microbiology, Rutgers-The State University of New Jersey, New Brunswick, New Jersey 08903 

IS been well established that under 
conditions testosterone administration 
gnificantly affect the rates of DNA, 
nd protein synthesis in the rat ventral 
e (1-4). Testosterone also maintains 
rphology and secretory activity of the 
e gland, both in in vitro and in vivo 
I (5-7). On castration, there is rapid 
ion of the rat ventral prostate includ- 
; cessation of secretory function (8). 
terol has been found to be one of the 
constituents of the prostate secretion 

In this paper, we are reporting the 
ion of cholesterol synthesis by testos- 
in the rat ventral prostate. Kinetics of 
srol synthesis in the ventral prostate 
Qg testosterone administration to cas- 
ats was studied in relation to prostate 
gain, DNA and protein synthesis. 
trials and methods. Animals, Groups of 
It male intact and castrated Wistar rats 
)0 g) were maintained on Purina rat 
ind water ad libitum and were kept 
iltemating 12-hr light and 12-hr dark 
ie. At necropsy final body weights 

inistration of testosterone to castrated 
r. Castrated animals were injected sub- 
)usly with 2 mg of testosterone propi- 
lissolved in sesame oil (10 mg/ml), at 
le time every day for different periods 
3 14 days. 

tro incorporation of radioactive precur- 
to cholesterol, proteins and DNA by 
prostate tissues. At various time inter- 
to 14 days animals were anesthetized 
traperitoneal injections of sodium bar- 
id sacrificed by exsanguination. The 
»es of the ventral prostate gland were 

free of the fat covering. The tissues 
linced and weighed immediately in 
eflon test tubes and kept in ice until 
use. Approximately 25-35 mg sam- 
minced tissues were used to study the 
ration of radioactive precursors into 
irol, proteins and DNA. 

The radioactive precursors, 2-[^*C]acctate 
(sp. activity 50.3 mCi/mmol), 4,5-^H-L-leu- 
cine (sp. activity 5 Ci/mmol), and ^H-methyl- 
thymidine (sp. activity 6.7 Ci/mmol) were 
used in these studies to determine their incor- 
poration into cholesterol, protein and DNA, 
respectively. Tissues were incubated with 2 
ml of Hank's Balanced Sah solution supple- 
mented with 0.2% glucose and either 1 
/iCi/ml of 2-[^^C]acctate or 1 /iCi/ml of ^H- 
leucine or 3 /iCi/ml of ^H-thymidine (pre- 
gasscd with 95% O2 and 5% CO2) at 37*' for 
2 hr on a constant speed shaker. At the end 
of the incubation period, the reaction was 
terminated by instant freezing of the tubes in 
a dry ice-acetone bath. The radioactivity of 
cholesterol, protein and DNA in the tissues 
was then determined. 

Analysis of radioactivity in cholesterol. The 
tissues were saponified by the addition of 
alcoholic KOH to a final concentration of 
10% KOH and 50% ethanol (95%) at 75^ for 
75 min. Unsaponified lipids were pooled by 
repeated extractions with ;i-hexane. The hex- 
ane extracts were evaporated under nitrogen 
and digitonin precipitation was carried out 
according to the procedure of Sperry (11). 
The cholesterol-digitonin complex was dis- 
solved in 1 ml of methanol and 0. 1 ml aliquots 
were counted in duplicate for [^^CJactivity in 
a Packard Scintillation Counter. The rates of 
synthesis were expressed as counts per minute 
per /ig of prostatic DNA. 

Analysis of radioactivity in protein and 
DNA. The tissues were homogenized with a 
Brinkmann polytron and crude protein or 
DNA was precipitated with 5 ml of 6% tri- 
chloroacetic acid (TCA) at 0*". After 10 min, 
the samples were centrifuged at 4^, and the 
precipitates were washed twice with 5 ml of 
6% cold TCA. The precipitates were then 
extracted repeatedly with 95% ethanol: 
chloroform (3:1 v/v) to remove lipids. For 
radioactivity counting in proteins, the 
ethanol-chloroform extracted precipitates 
were dissolved in 2 ml of 10% NaOH and 0.2 



Copyright O 1978 by the Society for ExperimentAl Biology and Medidne 

All rights reserved. 


ml aliquots were counted in duplicate for in- 
activity. To measure the incorporation of ^H 
thymidine into DNA, ethanol-chloroform 
extracted precipitates were dissolved in 2 ml 
of 0.3 N KOH at 3V for 60 min. Proteins and 
DNA were then reprecipitated from super- 
natants with 8 ml of 6% TCA. The KOH 
extraction and the TCA precipitation were 
repeated. The final TCA insoluble fraction 
was treated with 2 ml of 16% perchloric acid 
(PCA) for 20 min at TO*", followed by ccntrif- 
ugation. Aliquots (0.2 ml) of the acid-soluble 
fraction were counted in duplicate for deter- 
mination of radioactivity in the DNA. All 
tritium determinations were made in a xylene 
based scintillation cocktail (aquasol-2. New 
England Nuclear) and counted in a Packard 
Scintillation Counter. 

Colorimetric determinations. The amount of 
cholesterol was quantitated by first saponi- 
fying the tissues and the unsaponified frac- 
tions were used for digitonin precipitation. 
The cholesterol-digitonin complex was used 
for colorimetric determination by the method 
of Parekh and Jung (12). 

DNA and proteins were extracted in simi- 
lar manner described in the section above 
and colorimetric analyses were carried out 
employing the method of Abraham et al (13) 
for DNA and the Biuret procedure (14) for 
protein assay. 

Results, As expected, the data in Table I 
confirm that on castration the prostate weight 
declines to about 12% of the normal rat pros- 
tate weight. Body weights are not signifi- 
cantly affected. Amounts of cholesterol, pro- 
tein and DNA in rat prostate gland, quanti- 
tated colorimetrically, also decline to 12%, 
13% and 25% of their respective normal val- 
ues. Rates of synthesis of cholesterol and 
DNA per /ig prostatic DNA also decline to 

about 8% and 5%, respectively, in ca 
animals. Contrary to the decreases 
rates of synthesis of DNA and chol 
the rate of protein synthesis per fig p: 
DNA remains constant in the castrat 
mals even though the total amount of 
present in the prostate gland of castrai 
is significantly lower than in normal a 
This may be due to the synthesis of hy( 
enzymes that would hydrolyze the p 
present in the normal gland. The res 
dicate that testosterone produced by tfa 
is essential for the maintenance of gh 
cholesterol synthesis in the prostate 
among other prostatic functions. 

To examine whether testosterone 
restore the prostate cholesterol levels 
as the levels of macromolecules, 2 
testosterone propionate in sesame 
mg/ml) was injected daily subcutanec 
castrated rats for varying periods of t 
to 12 days. Prostate weight as well as 
terol, DNA and protein content in tfa 
tate gland were quantitated and the 
are presented in Fig. 1. Administra 
testosterone to castrated rats increa 
amount of cholesterol in the prostate 
DNA and protein content also inc 
Prostate weights and the amounts of 
terol and protein increased more sign! 
after 2 days of testosterone admmis 
The amount of DNA remained cons 
to 2 days and increased significantly 
days of testosterone injections. All f 
rameters tested increased almost equ; 
tween 2 and S days of testosterone tre 
The amount of protein increases sharp 
S days of treatment which is followed 
days by steep increases in prostate wei 
the amount of cholesterol. Contrar) 
creases in prostate weight and in the j 

TABLE \. Effect of Castration on the Synthesis of Prostate Cholesterol, Protein and Dl 

Wcl pros- 

Prostate Cholesterol 

Prostate Protein 

Prostate E 



Total con- 

Rate of 

Total con- Rate of 

Total con- 




tent synthesis 

tent s 






(mgr (cpm)* 
5.83 108.91 








± 11.5 



± 11.98 

± 0.983 ± 70.49 

± 57.97 r 






0.728 102.62 


± 16.1 

± 1.26 



±0.106 ±33.07 

± 12.43 

° All the total contents are expressed in terms of per 100 g body wt. 

* The rate of synthesis are expressed as cpm/^g of prostate DNA isolated. 

'^ Rats were castrated for 7 days. 


fcholesteToU the slopes of curves for DNA 
nd protein at 12 days of testosterone admin- 
itration approached the steady state. 
In Fig. 2 the ratios of cholesterol, protein 
nd DNA content of the prostate glands from 

I 2 345676910 II 12 
Doys of TNtMltroR* Administrotion 

Fig. 1. The effect of testosterone administration to 
the castrated lats on prostate weight and the contents of 
cboksterol, DNA and protein in the ventral prostate. 
The mean values are obtained from groups of six rats. 

testosterone treated castrated rats are pre- 
sented. The ratios of both cholesterol/DNA 
and protein/DNA increase on the adminis- 
tration of testosterone. This would be ex- 
pected since de novo synthesis of enzymes for 
cholesterol synthesis pathway would be re- 
quired for an increase in cholesterol content. 
Figure 3 shows the rate of synthesis of 
cholesterol, DNA and proteins at various pe- 
riods of testosterone administration up to 14 
days. The rates of synthesis of protein and 
cholesterol peak 2 days after testosterone in- 
jection, whereas DNA synthesis peaks after 
4 days of treatment. The two peaks in protein 
and cholesterol synthesis after 2 days and 
again after S days might indicate the synthesis 
of structural components followed by synthe- 
sis of secretory components. After 5 days, 
synthesis of DNA, protein and cholesterol 
decreases and remains at a steady state for 
the remainder of the 14 days of testosterone 
treatment. Despite the fact that the rate of 
cholesterol synthesis per microgram of pros- 
tatic DNA reaches a steady state, the sharp 
increase in cholesterol content upon testoster- 
one administration at day 12 can be ac- 
counted for by the increased prostate weight 
as seen in Fig. 1. 


■ ^ ^ 4 4 A » 4 » t& II 

Ooys of Testosttroot Administrotion 


Fig. 2. The effect of testosterone administration to 
^ castrated rats on the content ratios of choles- 
iCTol/DNA and protein/DNA in the ventral prostate. 

4 6 8 10 12 

Ooys of Ttstostfrone Administrotion 

Fig. 3. The effect of testosterone administration to 
the castrated rats on the rates of synthesis of cholesterol 
protein and DNA in the ventral prostate. The mean 
values are obtained from groups of six rats. 


Discussion. Swyer (IS) reported an increase 
in the cholesterol content of the adenomatous 
portion of enlarged human prostate glands as 
compared to normal glands. Braunstein (16) 
reported the presence of retractile and doubly 
refractile crystals as well as a positive Schultz 
reaction in the cytoplasm of human prostatic 
carcinoma cells indicative of a high content 
of cholesterol. Leav and Ling (17) reported 
the similar findings on tissues derived from 
neoplastic canine prostate gland. 

Since the discovery that the hypocholester- 
olemic polyene macrolides (18) by the oral 
route decreased the size of the enlarged pros- 
tate glands of dogs (19) and hamsters (20), 
there has been increasing clinical evidence 
(21-27) that these drugs affect the symptoms 
of prostatism caused by benign prostatic hy- 
perplasia. Other hypocholesterolemic agents 
such as cholestyramine (28), colestipol (20), 
simfibrate (29) and ^-sitosterol (30) have now 
also been reported to affect the prostate 
gland. Considering that hypocholesterolemic 
drugs in general appear to afTect the choles- 
terol-containing enlarged prostate gland and 
realizing the importance of cholesterol in this 
organ, it became necessary to study choles- 
terol metabolism in the prostate gland and its 
possible regulation by testosterone, a recog- 
nized mediator of other prostatic functions. 

It is very evident from these current studies 
that testosterone is a major factor in the 
synthesis of cholesterol in the prostate gland. 
On testosterone administration to castrated 
rats the amount of cholesterol increases be- 
fore an increase in DNA content. Liao et al. 
have shown that the RNA polymerase activ- 
ity from the prostate of castrated rats is en- 
hanced within a few hours of single injection 
of testosterone. This may mean that the initial 
increase in cholesterol content is more likely 
due to increased RNA and protein synthesis. 
Following the initial cholesterol curve there 
is an increase in DNA content and then 
another increase in cholesterol content. 

The observed two different phases in the 
amount of prostate cholesterol, the first of 
parallel increase with protein from day and 
the second of a sharp increase in cholesterol 
between day 8 and 12 can be explained on 
the basis of cholesterol having a dual function 
in the gland. In the first phase, it is likely that 
only structural or membrane cholesterol is 

synthesized. After 8 days of testosterone ad- 
ministration when the gland approaches the 
normal state, since cholesterol is an important 
secretory product of the prostate gland, 
greater amounts of cholesterol-synthesizing 
enzymes might be produced as indicated by 
the large increase in protein content. This 
would be followed by the synthesis of a large 
amount of secretory cholesterol. Prostate 
weight rises in parallel with the amount of 

The sharp increases in the synthesis of 
cholesterol, DNA and protein is followed by 
a sharp decrease on continuous testosterone 
administration. This may be due to a shift in 
testosterone metabolism in the prostate gland 
where testosterone may be converted to in- 
active or less active metabolites as compared 
to the conversion to a highly active metaboUte 
such as dihydrotestosterone (31). This indi- 
cates that testosterone may be acting both as 
a positive and negative regulator of choles- 
terol synthesis in the prostate gland. 

Summary, The absolute cholesterol content 
and rate of cholesterol synthesis was com- 
pared in rat ventral prostates obtained from 
adult normal and castrated rats. Cholesterol 
content and synthesis reduces to about 8-12% 
in the ventral prostate of castrated animals as 
compared to normal rats. Daily testosterone 
injections to castrated rats elicits a sharp in- 
crease in cholesterol content which correlates 
with an increase in prostate weight. The rate 
of cholesterol synthesis per microgram of 
prostatic DNA increases steeply 2 days after 
testosterone administration and then goes 
down and reaches a steady state after 5 days. 

1. Coffey, D. S., Shimazaki, J., and WilUams-Ashman, 
H. G., Arch, of Biochem. Biophys. 124, 184 (I%8). 

2. Liao, S., Leininger, K. R., Sagher, D., and Barton, 
R. W., Endocrinology 77, 763 (1965). 

3. Liang, T., and Liao, S., Proc. Nat. Acad. Sci. U.S.A. 

4. Williams-Ashman, H. G., Liao, S., Hancock, R. L., 
Jurkowitz, L., and Silverman, D. A, Recent Prog. 
Horm. Res. 20, 247 (1964). 

5. Baulieu, E. E., Lasnitzki, L, and Robel, P., Nature 
(London) 219, 1155(1968). 

6. Robel P., Lasnitzki, L, and Baulieu, E. £., Biochimie 

7. Baulieu, E. E., Le Goascogne, C, Groyer, A., Feyel- 
Cabanes, T., and Robel, P., Vitamins Horm. 33, 1 


OS, C, and Clark, P. J., J. Exp. Med. 72, 747 

T., The Biochemistry of Semen*, p. 130, 
en and Co., London (1954). 
T., NaU. Cancer Inst. Monogr. 13, 235 (1963). 
, W. M.. J. Lipid Res. 4, 221 (1963). 
I, A. C, and Jung, D. H., Anal. Chem. 42, 

am, G. N., Scaletta, C, and Vaughan, J. H., 
, E., Methods Enzymol. 3, 450 (1957). 
, G. L. Cancer Res. 2, 372 (1942). 
stein, H., The Amer. J. Clin. Pathol. 41, 44 

[., and Ling, G., Cancer 22, 1329 (1968). 

her, C. P., and Gordon, H. W., Proc. Nat. 


n, H. W., and Schaffner, C. P., Proc. Nat. 

Sd. 60, 1201 (1968). 

G. M., and Schaflher, C. P., Invest. Urol. H 


, G., Polichnico Sez. Med. 77, 285 (1970). 

, G.. La Chirurgia Generale 19, 1 (1970). 

23. Kljucharev, B. V., Berman, N. A., Ivanov, N. M., 
Margolin, A. M., and Mikhailets, G. A., Voprosy 
Onkologii (Leningrad) 18, 36 (1972). 

24. Keshin, J. G., Int. Surg. 58, 1 16 (1973). 

25. Kljucharev, B. V., Mikhailets, G. A, Berman, N. A., 
Ivanov, N. M., and Margolin, A. M., Urologiia i 
Nefrologiia (Moscow) 38, 40 (1973). 

26. Orkin, L. A., Urology 4, 80 (1974). 

27. Sporer, A, Cohen, S., Kamat, M. H., and Seebode, 
J. J., Urology d, 298 (1975). 

28. Addleman, W., New England J. Med. 287, 1047 

29. Yamanaka, H., Shimazaki, J. Koya, A, Mayuzumi, 
T., Imai, K., Yoshikazu, I., and Shida, K., Endocri- 
nol. Japan 24, 213 (1977). 

30. Effmghaus, K. D., and Baur, M. P., Z. Allg. Med. 

31. Bruchovsky, N., and Lesser, B., in "'Cellular Mech- 
anisms Modulating Gonadal Hormone Action**, (R. 
L. Singhal and J. A. Thomas, eds.) p. 1, University 
Park Press (1976). 

Received February 8, 1978. P.S.E.B.M. 1978. VoL 159. 


The Role of Cyclic AMP in CRF-lnduced ACTH Secretion^ (40272) 


Department of Physiology, School of Medicine, Case Western Reserve University, Cleveland, Ohio 44106 

Experiments in this (1) and other (2-4) 
laboratories have shown that cyclic-3',5'- 
adenosine monophosphate (cyclic AMP) and 
its derivatives stimulate the secretion of 
ACTH. Recently, we have reported that the 
stimulation of ACTH secretion by hypotha- 
lamic median eminence-corticotrophin re- 
leasing factor (HME-CRF) is associated with 
a concomitant increase in adenylate cyclase 
activity; however, cordycepin (3'-deoxyadcn- 
osine) at sufficient concentration to reduce 
adenylate cyclase activity to undetectable lev- 
els, reduces but does not abolish the HME- 
CRF induced secretion of the hormone (5). 
These data suggest that while cyclic AMP 
may be involved in CRF-stimulated ACTH 
secretion, the cyclic nucleotide may not act as 
an obligatory intermediate, but rather may 
act to potentiate secretion. The experiments 
described in the present communication were 
designed to provide further information on 
this hypothesis. 

Materials and methods. The techniques 
used in the preparation and incubation of 
isolated pituitary cells have been described in 
detail elsewhere (6, 7). In brief, anterior pi- 
tuitary glands were removed from male 
Sprague-Dawley rats which had been adre- 
nalectomized 14-28 days prior to sacrifice, 
and were maintained after adrenalectomy on 
0.9% saline drinking solution without steroid 
hormone replacement. Cells were dispersed 
from the glands by mechanical agitation in 
Krebs-Ringer bicarbonate (KRB) buffer con- 
taining 0.2% glucose and 0.25% trypsin. After 
dispersion, cells were collected by centrifu- 
gation and resuspended in KRB buffer con- 
taining 0.2% glucose and 0.5% bovine serum 
albumin (KRBGA), plus 0.1% lima bean 
trypsin inhibitor. Aliquots (0.9 ml) of cell 
suspension were incubated for various times 
together with appropriate combinations of 
HME-CRF, N^O'-dibutyryl-cyclic AMP 
(DBC), corticosteronc, or vehicle (controls). 

' This work was supported by USPHS Grant No. 
AM- 1 3820-08. 

At the end of the incubation period, oeUs 
were removed by centrifugation, and the in- 
cubation medium was acidified, appropri- 
ately diluted, and assayed for ACTH. In most 
cases, the samples were bioassayed according 
to the isolated adrenal cortex cell technique 
described by Sayers et al (8), using synthetic 
ACTH 1-24 (Cortrosyn, Organon) as stan- 
dard. In one experiment (employing concen- 
trations of DBC greater than 1 mA/, see Fig. 
5), in order to circumvent the problem of 
direct DBC stimulation of steroidogenesis bv 
isolated adrenal cells, pituitary cell incuba- 
tion medium was assayed for ACTH by a 
radioimmunoassay (RIA) technique. Rabbit 
anti-human ACTH serum was purchased 
from Burroughs-Wellcome, and ^I-ACTH 
1-24 was obtained from Amersham. Samples 
or standards (ACTH 1-24, Cortrosyn, Orga- 
non) were incubated with inmiune serum in 
0.1 M sodium phosphate (pH 7.4), for 20 hr 
(4^), at which time '^I-ACTH was added 
and the incubation was continued for 6 ad- 
ditional hr. Un-bound ^^I-ACTH was ad- 
sorbed to charcoal, collected by centrifuga- 
tion, and counted in a Packard auto-ganmia 
spectrometer. The method appears valid as 
judged by several criteria: (a) both extracts of 
pituitary cells and samples of pituitary cell 
incubation media gave log dose-displacement 
curves parallel to synthetic ACTH 1-24; (b) 
a number of polypeptides, including ACTH 
5-10, ACTH 5-13, and a-MSH, showed no 
significant cross-reactivity; and (c) analysis of 
samples of pituitary cell extracts or incuba- 
tion media by bioassay and RIA gave essen- 
tially identical values. In all experiments, the 
ACTH content of control incubates was de- 
termined and subtracted from that of incu- 
bates receiving test substance(s). In each ex- 
periment, data obtained from incubates re- 
ceiving identical treatments were pooled, and 
means and standard errors of the means 
(SEM) were calculated. Statistical signifi- 
cance was assessed by means of Student's t 

Extracts of rat hypothalamic median emi- 

^037-9727/78/ 1 59 1 -000650 1 .00/0 

ipynghi ® 1978 by the Society for Experimental Biology and Medicine 
\ right* reserved. 


nence tissue (HME-CRF) were prepared by 
homogenizing freshly excised ventral hypo- 
thalamic-median eminence tissue in 0.2 M 
acetic acid. Insoluble material was removed 
by centrifugation (20,0(X)g, IS min), and was 
reextracted twice with 0.2 M acetic acid. The 
extracts were combined and stored frozen. 
For use, a portion of the extract was adjusted 
to pH 7.0, appropriately diluted (with 
KRBGA) and added to the incubates in a 
volume of 0.1 ml. Doses of HME-CRF are 
expressed as tissue equivalents (i.e., fractions 
3f an HME), which in these experiments had 
Si wet weight of approximately IS mg. Corti- 
;x)sterone (Sigma) in 0.9% saline plus 2.5% 
methanol, was added to appropriate incu- 
bates in a volume of 10 /il. DBC (Sigma) was 
sidded to appropriate incubates in a volume 
3f 0.1 ml of KRBGA. 

Results. Both HME-CRF and DBC stim- 
ulate the secretion of ACTH by isolated pi- 
tuitary cells (Fig. 1), and at the concentrations 
tested (0.2 HME/ml, 1 mM DBC) the ACTH 
secretory responses are nearly identical (ISO 
pg/min/lO* cells). This concentration of 
DBC (1 mM) in the medium did not interfere 
in the subsequent steroidogenic bioassay for 
\CTH, as shown by the fact that addition of 
DBC at the end of the incubation period with 
HME-CRF does not significantly alter the 
-espouse from that of HME-CRF alone. 
tVhen pituitary cells are exposed to DBC 
:hroughout the exposure to HME-CRF, the 
-ate of ACTH secretion is markedly en- 
lanced. The rate of hormone secretion in the 
presence of HME-CRF plus DBC (S75 
jg/min/ltf^ cells) is abnost twice that ex- 
pected if the response to the two agents were 
limply additive. As shown in Fig. 2, the po- 
entiating effect of DBC on HME-CRF in- 
luced ACTH secretion occurs without an 
)bvious time-lag and persists throughout the 
luration of a 4S min incubation. The data in 
r ig. 3 indicate that the exposure of pituitary 
:ells to DBC potentiates HME-CRF induced 
\CTH secretion, even if the cyclic nucleotide 
s removed prior to addition of HME-CRF. 
[n these experiments cells were preincubated 
br IS min in the presence or absence of DBC 
1 mM) and then challenged with HME-CRF 
n the presence or absence of DBC (1 mM). 
HIME-CRF induced ACTH secretion by cells 
exposed to DBC was more than twice that of 

600 r 

«M soo > 


i 300 

< 200 





Fig. 1. Interaction of CRP and DBC on ACTH 
secretion. Isolated pituitary cells prepared from adrenal- 
ectomized rats were incubated for 35 min. Substances 
added, and their time of addition during this interval, 
are indicated below each bar H, HME-CRP (.2 
HME/ml); D, DBC (1 mM). Secretory rates are for the 
30 min-period following the addition of HME-CRF; 
vertical lines represent combined SEM of pituitary and 
adrenal assays (N « 8). 


Fig. 2. Interaction of CRF and DBC on ACTH 
secretion; time course. Pituitary cells were incubated for 
indicated times in the presence of: DBC (1 mM), HME- 
CRF (.2 HME/ml), or DBC (1 mM) plus HME-CRF 
(.2 HME/ml). The ACTH content of control incubates 
(920 ± 42 pg/10^ cells, mean ± SEM, N - 10) did not 
change from 15 to 45 min, and has been subtracted from 
the experimental values presented. Vertical lines repre- 
sent combined SEM of pituitary and adrenal assays (N 

cells which were not exposed to DBC, irre- 
spective of whether the cychc nucleotide was 
present during the preincubation only, the 
incubation with HME-CRF only, or both the 


z _ 




z - 


Fig. 3. Effect of time of addition of DBG on CRF 
induced ACTH secretion. Isolated pituitary cells were 
preincubated (15 min) in the presence or absence of 
DBC (1 mM); the cells were collected by centrifugation, 
washed with KRBGA, recollected by centhfugation, and 
resuspended in KRBG A. Aliquots of both types of cells 
were then incubated (30 min) with HME-CRF (.2 
HME/ml) in the presence or absence of DBC (I mM). 
The presence of DBC (D) during the preincubation and 
incubation periods is indicated beneath each bar. Data 
are expressed as the percentage of the secretory rate of 
cells which were not exposed to DBC (control); vertical 
lines represent SEM of the normalized secretory rates (N 

preincubation and the incubation. 

The experiments described above demon- 
strate the interaction of submaximal doses of 
DBC and HME-CRF. In order to determine 
whether these secretagogues also interact at 
maximal dose levels, two experiments were 
performed. First, isolated pituitary cells were 
exposed to graded doses of HME-CRF in the 
presence or absence of DBC (Fig. 4). In the 
absence of DBC, maximum ACTH secretion 
is noted at a concentration of about 1.8 HME- 
CRF/ml. In the presence of DBC (1 mM), 
the secretory response to each dose of HME- 
CRF is increased more than twofold, even at 
maximum doses of HME-CRF. In the second 
experiment, isolated pituitary cells were ex- 
posed to graded doses of DBC in the presence 
or absence of HME-CRF (Fig. 5). In the 
absence of HME-CRF, maximum ACTH se- 
cretion is produced at a concentration of 
about 10 mM DBC. In the presence of HME- 
CRF (0.4 HME/ml), the secretory response 
is more than doubled at each dose of DBC, 
including the maximal doses. 

Previous findings in our laboratory have 

i.a 34 

Fig. 4. Effect of DBC on ACTH secretion in re- 
sponse to graded doses of CRF. Isolated pituitary cells 
were incubated for 10 min in the presence or absence of 
DBC (1 mM); graded doses of HME-CRF were then 
added and the incubation was continued for an addi- 
tional 30 min. Data presented are net pg ACTH secreted; 
vertical lines represent combined SEM of pituitary and 
adrenal assays (N » 8). 

shown that the secretion of ACTH by isolated 
pituitary cells in response to a variety of 
secretagogues, including DBC, is inhibited by 
corticosterone (9). We therefore carried out 
an experiment to determine if the potentiat- 
ing effect of DBC on HME-CRF stimulated 
ACTH secretion is also inhibited by steroid. 
Pituitary cells were incubated for 30 min in 
the presence or absence of corticosterone (0.1 
/ig/ml) and were then stimulated (for 30 ad- 
ditional min) with either HME-CRF (0.2 
HME/ml), DBC (1 mM), or HME-CRF (0.2 
HME/ml) plus DBC (1 mM). The data in 
Fig. 6 show, as expected, that in the absence 
of exposure to corticosterone, both HME- 
CRF and DBC stimulate the secretion of 
ACTH, and HME-CRF stimulated secretion 
is potentiated by DBC. When the cells are 
exposed to corticosterone, ACTH secretion 



S 10 

OtC (mM) 
Fig. 5. Effect of HME-CRF on ACTH secretion in 
poase to graded doses of DBC. Isolated pituitary cells 
re incubated for 30 min with various doses of DBC in 
presence or absence of HME-CRF (0.4 HME/ml). 
TH was determined by RIA; data presented are net 
ACTH secreted. Vertical lines represent combined 
A of pituitary and radioimmune assays (N » 6). 

nulated by either agent acting singly is 
irly abolished; ACTH secretion stimulated 
HME-CRF and DBC in combination is 
markedly inhibited, but is still signifi- 
itly greater than that induced by HME- 
\F (P < .01) or DBC (P < .01), acting 

DiscussiorL Several lines of evidence sug- 
t an involvement of cyclic AMP in the 
racellular mechanisms which regulate 
TH secretion. Cyclic AMP and its deriv- 
/es have been found to stimulate the secre- 
D of ACTH both in vivo and in vitro (1-4). 
ubitors of cyclic nucleotide phosphodies- 
ase stimulate the secretion of ACTH (4) or 
synergistically with other secretagogues of 
: hormone (2), presumably elevating the 
racellular level of cyclic AMP. Recently, 
observed that addition of HME-CRF to 
pensions of isolated pituitary cells pro- 
xs an increase in adenylate cyclase activ- 



- FiLrL 

Dmin - ■ - • 

SOmin N N 

N^O N*0 

Fig. 6. Interaction of CRF and DBC on ACTH 
secretion; effect of corticosterone. Isolated pituitary cells 
were incubated for 60 min. Substances added, and their 
time of addition during this interval, are indicated below 
each bar; H, HME-CRF (.2 HME/ml); D, DBC (1 mA/); 
B, corticosterone (0. 1 /ig/ml). Secretory rates are for the 
30 min-period following the addition of HME-CRF; 
vertical lines represent combined SEM of pituitary and 
adrenal assays (N « 8). 

ity, concomitant with an increase in the rate 
of ACTH secretion (5). A stimulation of rat 
pituitary adenylate cyclase activity has also 
been reported in response to crude extracts of 
ovine hypothalamus (10) and vasopressin (4), 
an agent which is distinct from hypothalamic 
CRF but nevertheless stimulates the secretion 
of ACTH (11). These observations are all 
consistent with the notion that cyclic AMP is 
involved in the process which mediates 
ACTH secretion, but they provide no infor- 
mation as to the role of the cyclic nucleotide 
in this process. In this regard it is well to 
stress that although a large number of secre- 
tagogues of ACTH are known, no "authen- 
tic" hypothalamic CRF is yet available in 
pure form. Experiments employing crude ex- 
tracts of hypothalamus (including those re- 
ported above) are limited in that responses 
observed may be the consequence of the in- 
teraction of several factors. Indeed, it is for 
this reason that little is known of the cellular 
and molecular processes which mediate 
ACTH secretion or the biochemical pathways 
by which these processes are regulated. 

If the role of cyclic AMP in ACTH secre- 
tion is that of an obligatory "second messen- 



ger**, then it would be expected than an agent 
which inhibits adenylate cyclase activity 
would interfere with secretion of the hor- 
mone. Cordycepin has been found to be an 
inhibitor of adenylate cyclase activity in fat 
cell membranes (12) and guinea pig lung (13). 
Experiments in our laboratory (S) have 
shown that a dose of cordycepin sufficient to 
reduce adenylate cyclase activity to undecta- 
ble levels in isolated pituitary cells only par- 
tially reduces the rate of HME-CRF induced 
ACTH secretion. We interpreted these data 
to mean that while cyclic AMP may indeed 
be involved in CRF-stimulated ACTH secre- 
tion, the cyclic nucleotide may not act as an 
olbigatory intermediate but rather may act to 
potentiate secretion. Sundberg et al (14) have 
advanced a similar proposal with respect to 
the role of cyclic AMP in the secretion of 
several other adenohypophysial hormones. 
The data presented in this communication 
arc consistent with this view. DBC potentiates 
HME-CRF induced ACTH secretion both at 
submaximal and maximal doses of HME- 
CRF (Fig. 4), and HME-CRF potentiates 
DBC induced ACTH secretion both at sub- 
maximal and maximal doses of DBC (Fig. 5). 
The mechanism of the interaction between 
HME-CRF and DBC is unknown. Potentia- 
tion occurs without an apparent lag period 
and persists for at least 30-45 min (Fig. 2). 
Significantly, prctrcatment of isolated pitui- 
tary cells with DBC (foUowed by removal of 
the cyclic nucleotide prior to exposure to 
HME-CRF) potentiates the secretory re- 
sponse to HME-CRF to as great a degree as 
does exposure to HME-CRF in the presence 
of DBC (Fig. 3). This finding does not rule 
out the possibility that cyclic AMP has been 
sequestered within the cells during the pre- 
treatment period, and subsequently poten- 
m!fp ^^"'^one secretion during exposure to 
HME-CRF. Alternatively, this finding is con- 
sistent with the view that the potentiating 
ellect following DBC pretreatment may rep- 
flh^^'^\f P*^y^*^^l and/or chemical change in 
me cell which is exerted after the cyclic nu- 
"^^^^5*^^ has been removed, 
the ^^^ ^ ^*8 ^ dramatically Ulustrate 

on Ar-Ttl ^^it^itory effect of corticosterone 
ue/mu ^cretion. At concentrations (0.1 

Steroid ^*^*^ ^^^ physiological range, the 
Markedly suppresses hormone secre- 

tion in response to HME-CRF ar 
acting singly or in combination. Th 
ings indicate that whatever the role 
AMP in ACTH secretion, the sit 
inhibitory action of the steroid is dis 
appearance of the cyclic nucleotide. 

In conclusion, the data of the pres 
munication support the hypothesis tl 
AMP acts within corticotropin to p 
CRF-induced ACTH secretion. I 
our previous findings (S) indicate 
increased level of cyclic AMP is not 
for ACTH secretion to occur. Taken 
these data suggest that CRF has (at 1 
actions on the corticotroph: (a) the 
of the series of events which even 
secretory granule exocytosis, and (I 
vation of cyclic AMP levels withir 
which then facilitates (through i 
mechanisms) the secretory process. 

Summary. ACTH secretion by is< 
tuitary cells is stimulated both by HI 
and DBC, and when given in com 
the two secretagogues interact syner] 
Although the mechanism of this in 
is unknown, the potentiating effect 
is displayed without an apparent tim 
persists after removal of the cyclic ni 
Corticosterone inhibits ACTH seci 
duced by HME-CRF and DBC, act 
or in combination. The implication 
findings are discussed. 

The authors are grateful to Beth Wiblir 
and James Roe for their expert technical ass 

1. Portanova, R., and Sayers, G., in "Brai 
Adrenal Interrelationships** (A. BrodisI 
Redgate, eds.) p. 319. S. Karger, Basel 

2. Fleischer, N., Donald, R. A., and Bute 
Amer. J. Physiol. 217, 1287 (1969). 

3. Hedge, G. A., Endocrinology 89, 500 {V 

4. Vale, W., and Rivier, C, Fed. Proc. 3d, : 

5. Bratiin, W. J., and Portanova, R., Mol. 

6. Portanova, R., Smith, D. K.. and Sayei 
Soc. Exp. Biol. Med. 133, 573 (1970). 

7. Portanova, R., Proc. Soc. Exp. Biol. M< 

8. Sayers, G., Swallow, R. L., and Giorad 
Endocrinology 88, 1063 (1971). 

9. Sayers, G., and Portanova, R., in "H 
Physiology** (R. O. Greep and E. B. Ast 
Sect. 7, Vol. 6, p. 41 (1975). 


., Kaneko, T., Schnieder, H. P. G., McCann, 13. Weinryb, I., and Michel I. M., Biochem. Biophys. 

and Field, J. B., J. Biol. Chem. 245, 2883 Acu 334, 218 (1974). 

14. Sundberg. D. K., Fawcett, C. P., and McCann, S. 

ova, R., and Sayers, G., Proc. Soc. Exp. Biol. M., Proc. Soc. Exp. Biol. Med. 151, 149 (1976). 

. N., Pointer. R. H., and Ward. W. F.. J. Biol. — — - 

247, 6866 (1972). Received January 3. 1978. P.S.E.B.M. 1978, Vol. 159. 


Ketamine as an Anesthetic for Obtaining Plasma for Rat Prolactin Assays (402 


Departments of Psychiatry,^ Pharmacology and Physiological Sciences,^ and Medicine,^ University ofChicagi 

School of Medicine, Chicago, Illinois 60637 

Various procedures have been utilized to 
obtain blood from laboratory rats for assay 
of plasma prolactin levels. Because of the 
effect of stress and general anesthetic agents 
on plasma prolactin levels (1-4), blood sam- 
pling procedures which do not themselves 
affect plasma prolactin levels are limited. 
Dohler et al (5) recently compared the influ- 
ence of four methods of blood collection un- 
der three anesthetics, ether, chloroform and 
pentobarbital, to decapitation on the release 
of prolactin; in all instances they observed an 
increase in plasma prolactin ranging from 2- 
to 13-fold. 

Lawson and Gala (4) reported that keta- 
mine, which is a dissociative anesthetic, not 
a general anesthetic (6), produced no effect 
on plasma prolactin levels at 10, 30, 60 and 
120 min after intraperitoneal (ip) or intra- 
arterial (i.a.) injection in ovariectomized rats 
with indwelling catheters. Lawson and Gala 
(7) subsequently reported that ketamine, 1(X) 
mg/kg, ip, also produced no change in 
plasma prolactin levels in catheterized, ovari- 
ectomized, estrogen-treated Sprague-Dawley 
rats, with sampling at 10, 30, 60 and 120 min 
after injection. However, ketamine, 50 
mg/kg, i.a., significantly decreased plasma 
prolactin levels at 10, 60 and 120 min. They 
suggested that ketamine differed from other 
anesthetics in its effects on prolactin secretion 
because it induced only stage II anesthesia. 

We were interested in determining what 
effects if any, ketamine had on plasma pro- 
lactin levels in male rats. Since ketamine has 
been shown to inhibit both dopamine and 
serotonin uptake (8, 9), two neurotransmitters 
which have a profound effect on rat prolactin 
secretion (10, 1 1), it was of further interest to 
see if ketamine affected baseline prolactin 
levels or the reserpine, a-methylparatyrosine 
(AMPT)- or 5-hydroxytryptophan (5-HTP)- 
induced increase in prolactin secretion. Drugs 
which inhibit 5-HT uptake will promote the 

increase in prolactin produced by 

Methods, Male Sprague-Dawle 
(Sprague-Dawley, Inc., Madison, Wi; 
ing 2(X)-22S g were housed for at leasl 
in a temperature-controlled (25*^) an 
controlled (6 am-8 pm light period) 
room. They received food and water 
turn. Ten groups of five rats each ha< 
ters placed in the right jugular veil 
pentobarbital anesthesia (13). The c 
were kept patent with flushing with s< 
alternate days. These rats were hand 
quently and accustomed to the procc 
withdrawal of 0.3 ml blood. These 
ceived no anesthesia at the time o 
withdrawal. Another ten groups of i 
each were administered ketamin 
mg/kg, ip Immediately after they 
unresponsive to toe pinch (usually 2 
blood was withdrawn from the inferi 
cava. Finally, ten groups of five ra 
rapidly decapitated. 

Reserpine, 5 mg/kg, ip, was given 
min or 3 hr 30 min before ketami 
mg/kg, ip, or saline. Rats were sacri 
4 hr after reserpine. AMPT, 100 mg 
was given IS min before ketamii 
mg/kg, ip and rats were sacrificed 
later. To determine the effects of ketai 
5-HTP-induced increases in plasma 
tin, ketamine, 25, SO, and 1(X) mg/kg, 
given 30 min before 5-HTP, 30 mg 
For comparison purposes, one group 
was pretreated with fluoxetine (Lilly 1 
a known 5-HT reuptake blocker (14), 
was pretreated with saline, followec 
HTP, as described for the ketami 
treated rats. 

Following sacrifice, plasma sampl 
frozen and assayed later for prolact 
modification of a double antibody r 
munoassay originally developed for 
prolactin assay (15). Prolactin levels 



Copyrighi O 1978 by the Society for ExpehmentAl Biology and Medicine 

AU rights reserved. 



I in terms of NIAMDD-rat prolactin- 
\11 samples utilized in this report were 
i together. The sensitivity of the assay 
ig/ml. The intra-assay variation is less 

letermine if there was a difference in 
)rolactin levels between types of sacri- 
e means for the 10 groups of each type 
ompared with a one way analysis of 
cc (ANOVA). To examine for differ- 
in variance within each of the three 
ents, a completely randomized hierar- 
analysis of variance was performed 
*he effect of drugs on the increase in 
tin produced by 5-HTP was deter- 
by an ANOVA. 

imine HCl was generously supplied by 
Davis- Wamer-Chilcott, Inc., Ann Ar- 
[ich. Alpha-methylparatyrosine meth- 
and 5-hydroxytryptophan methylester 
urchased from Sigma, Inc., St. Louis, 
leserpine was obtained from Ciba- 
Corp., Summit, NJ. Fluoxetine was a 
Eli Lilly, Co., Indianapolis, IN. All 
oses refer to the salt form. 
tits. Prolactin levels for the various 
)f sacrifice are summarized in Table I. 
edian, range and coefficient of varia- 
v.) were calculated utilizing the mean 
>r each group of five rats, 
results of an ANOVA indicated there 
• significant difference between any of 

the three methods of blood collection. How- 
ever, five of the ten groups of catheter sam- 
ples had mean levels that exceeded the high- 
est mean of the ketamine groups (10.2 
ng/ml). Only one of the decapitated groups 
had a mean plasma prolactin which exceeded 
10.2 ng/ml. The ketamine-treated group had 
the lowest prolactin levels and the smallest 
coefficient of variation of the three types of 

Ketamine did not significantly affect the 
increase in plasma prolactin levels produced 
by reserpine or AMPT (Table II). 

5-Hydroxytryptophan, 30 mg/kg, or fluox- 
etine, 10 mg/kg, did not increase plasma 
prolactin levels (Table III). Fluoxetine, to- 
gether with this dose of 5-HTP, produced a 
very significant increase in plasma prolactin. 
However, none of the three doses of ketam- 
ine, plus 5-HTP had any effect on plasma 
prolactin levels. Fluoxetine plus ketamine, 
100 mg/kg, also did not augment plasma 

Discussion, The results of the studies in 
untreated male rats strongly indicate that an- 
esthesia with ketamine does not affect plasma 
prolactin levels. Blood obtained from the in- 
ferior vena cava within 3 min of administra- 
tion of ketamine has levels of prolactin not 
significantly different from that obtained 
from decapitated rats or from rats with in- 
dwelling venous catheters. The latter method 

Rat Plasma Prolactin Levels Following Ketamine, Guillotining and from Indwelling 




Mean ± SEM 



Mean coefficient of 
variation {%) 




5 rats, xlO 
5 rats, xlO 
5 rats, XlO 

IS of each group of 5. 

6.0 ± 0.8 
6.4 ± 1.2 
9.4 ± 1.9 








ABLE U. Effect of Ketamine on 

Plasma Prolactin Levels Following Reserpine or AMPT. 


prolactin (ng/ml)* 

Dose (mg/kg) 


Ketamine p 

eseipine (A) 5 
eseipine (B) 5 
MPF too 

18.5 ±3.7 
21.7 ±2.7 
15.4 ±4.6 

25.0 ±3.1 NS 
21.3 ±1.7 NS 
15.2 ±3.9 NS 

in ± SEM Keumine, 100 mg/kg ip or saline was given 3 hr 55 min (A) or 3 hr 30 min (B) following 
: and 15 min following AMPT. Rats were sacrificed by decapitation 5 min (A) or 30 min (B) after ketamine 
serpine-pretrcated rats, and 15 min after ketamine in the AMPT-pretreated rats. All groups consisted of 5 



TABLE III. Effect of Ketamine and Fluoxetine 
ON Increase in Prolactin Produced by 5-HTP. 




Pretreat- Dose 
mcnt (mg/kg) 

Treat- Dose 
mcnt (mg/kg) 





6.4 ± 1.5 





8.4 ± 1.5 





7.5 ± 1.4 





38.7 ± 4.6 





6.4 ± 1.3 





6.7 ± 2.3 





8.6 ± 1.3 





7.0 ± 1.4 

* Mean ± SEM. The first injection was given 60 min 
before the second injection. Groups of five rats were 
sacrificed by decapiution 15 min after saline or 5-HTP. 

of blood sampling tended to produce the 
highest levels and the greatest variance within 
a given group of S rats, the usual size of our 
control groups. These results indicate that 
where a sinde blood sample is required from 
a given male rat, ketamine anesthesia is ac- 
ceptable. For studies in which anesthetized 
rats might be desirable, ketamine is clearly 
preferable to other anesthetics which them- 
selves afTect prolactin secretion. The reported 
ability of ketamine, SO mg/kg, i.a., to lower 
prolactin levels, in ovariectomized estrogen- 
treated rats (7), if confirmed, would indicate 
that ketamine might afTect the estrogen-stim- 
ulated prolactin secretion process and thus be 
less suitable for use in studies with female 
rats than it appears to be for male rats. The 
lack of efTect of ketamine on prolactin secre- 
tion further documents the difTerence be- 
tween the anesthesia produced by this agent 
and classical general anesthetics. 

The inability of ketamine to reverse the 
increase in plasma prolactin levels produced 
by reserpine or AMPT is strong evidence that 
ketamine does not have direct dopamine ag- 
onist effects in vivo at the pituitary dopamine 
receptors which regulate prolactin secretion. 
Direct dopamine agonists such as apomor- 
phine, bromcryptine or lysergic acid diethyl- 
amide readily reverse the increase in prolactin 
produced by reserpine or AMPT (17, 18 and 
unpublished data from this laboratory). Sim- 
ilarly, the inability to reverse the reserpine or 
AMPT-induced increase in prolactin indi- 
cates ketamine differs significantly from d- 
amphetamine, which has been shown to re- 
verse the increase in prolactin secretion pro- 

duced by reserpine or AMPT (19), presum- 
ably by increasing the release of dopamine 
from tubero-infundibular dopamine neurons 
or blocking its uptake. Previous studies of the 
effect of ketamine on dopaminergic mecha- 
nisms have been in vitro and have dealt with 
the nigro-striatal dopaminergic pathway. 
These differences may account for the differ- 
ences between the results of those studies and 
this one. 

The ability of fluoxetine but not ketamine 
to potentiate the effects of a subthreshold 
dose of S-HTP on prolactin secretion indi- 
cates that ketamine is not an effective inhib- 
itor of serotonin reuptake in vivo at those 
neurons which release the serotonin that po- 
tentiates prolactin secretion. These are be- 
lieved to be the median raphe serotonergic 
neurons (20). However, an effect of ketamine 
on uptake of serotonin by other serotonergic 
neurons is not excluded. 

The lack of effect of ketamine on the re- 
serpine-, AMPT- and 5-HTP-induced in- 
crease in prolactin secretion indicates the suit- 
ability of ketamine for anesthesia in studies 
of the effect of dopaminergic and serotonergic 
drugs on prolactin secretion. 

Summary. Mean plasma prolactin levels 
obtained from male rats following anesthesia 
with ketamine, decapitation or via indwelling 
venous catheters were not significantly differ- 
ent although a larger variance was found in 
the samples obtained via catheters. Ketamine, 
at anesthetic doses, did not affect the in- 
creases in prolactin produced by reserpine or 
a-methylparatyrosine. Ketamine, at various 
doses, did not potentiate the effect of sub- 
threshold doses of 5-hydroxytryptophan on 
prolactin secretion. Thus, ketamine would 
appear to be a suitable anesthetic for use in 
studies of prolactin secretion in male rats. 
Further studies in female rats are required. 

Supported in part by grant nos. USPHS MH 30938, 
29206, and RCSA MH 47808 to HYM. 

1. Krulich, L., Hcfco, E., lllncr. P., and Read, C. B., 
Ncurocndocrinology 16, 293 (1974). 

2. Wuttke, W., Gclato, M., and Mcitcs, J., Endocrinol- 
ogy 89, 1191(1971). 

3. Ncill, J. D., Endocrinology 87, 1 192 (1970). 

4. Lawson, D. M., and Gala, R. R., J. Endocrinol. 62, 



5. Dohlcr, K.-D., Von Zur Muhlcn, A., Gartner, K., 
and Dohier, U., J. Endocrinol. 74, 341 (1977). 

6. Winters, W. D., Ferrar-Allado, T., Guzman-Florcs, 
C. and Alcaraz, M., Neuropharmacology 11, 303 

7. Lawson. D. M., and Gala, R. R., J. Endocrinol. €6, 

8. Azzaro. A. J., and Smith, D. J., Neuropharmacology 

9. Smith, R. C. Meltzer, H. Y., Arora, R. C, and 
Davis. J. M., Biochem. Pharmacol. 26, 1435 (1977). 

10. Meites, J., and Clemens, J., Vitam. Horm. 30, 165 

11. Kambcri, I. A., Mical, R. S., and Porter, J. C 
Endocrinology 88, 1288 (1971). 

II Clemens, J. A.. Sawyer, B. D., and Ccrimclc, B., 

Endocrinology 100, 692 (1977). 
13 Weeks, J. R.. and Davis, J. D., J. Appl. Physiol. 19, 


14. Fuller, R. W., Perry, K. W., and Molloy, B. B., J. 
Pharmacol. Exp. Ther. 193, 796 (1975). 

15. Hwang, P., Guyda, H., and Friesen, H., Proc. Nat. 
Acad. Sci. U.S.A., 68, 1902 (1971). 

16. Kirk, R. E., in "Experimental Design: Procedures 
for the Behavioral Sciences,*' pp. 229, Brooks Cole 
Publishing Co., Belmont, California (1968). 

17. Horowski, R. and Graf, K.-J., Neuroendocrinology 

18. Meltzer, H. Y., Fessler, R. G., Simonovic, M., and 
Doherty, J., Psychopharmacology 54, 39 (1977). 

19. Meltzer, H. Y., Simonovic, M., Fessler, R., and Fang, 
V. S., Neurosci. Abs. Ill, 351 (1977). 

20. Advis, J. P., Simpkins, J. W., Bennett, J., and Meites, 
J., Endocrine Soc. Abs., p. 254, Chicago (June, 1977). 

Received March 16, 1978. P.S.E.B.M. 1978, Vol. 159. 


Pyrazinoic Acid and Urate Transport in the Rat (40274) 


Renal Section, Department of Medicine, Veterans Administration Hospital and Baylor College ofMedicit 

Houston, Texas 77211 

The decrease in urinary excretion of urate 
following the administration of pyrazinamide 
or its active metabolite, pyrazinoic acid 
(PZA), has been extensively utilized as a 
pharmacologic aid in dissecting out the con- 
tribution of secreted urate to the urinary ex- 
cretion of uric acid (1, 2). As originally pro- 
posed, the use of the "Pyrazinamide Suppres- 
sion Test" was based upon the assumptions 
that this compound was a specific and per- 
haps complete inhibitor of urate secretion 
and was without effect on the urate reabsorp- 
tive processes (3, 4). Indirect evidence has 
been presented, however, that neither of these 
assumptions is totally valid (5-8). Published 
studies on the separate effects of PZA on 
urate reabsorption and secretion, however, 
have been limited and somewhat conflicting 
(8-1 1). The current studies were designed to 
examine the effect of PZA, in varying dos- 
ages, on net urate transport and on the urate 
reabsorptive and secretory mechanisms in the 

Methods. Male Sprague-Dawley rats with 
free access to food and water until the time 
of study were used in all experiments. Anes- 
thesia was induced with Inactin (Promonta, 
Hamburg, Germany), 0.5-0.6 mA//kg body 
wt injected intraperitoneally. After a trache- 
ostomy, the right and left jugular veins were 
cannulated and the urinary bladder cathet- 
erized. In the clearance experiments, the left 
femoral artery was cannulated for collection 
of blood samples. In the microinjection and 
precession studies, the left kidney was pre- 
pared for micropuncture as previously de- 
scribed (12, 13). The ureter of the left kidney 
was catheterized with PE-50 tubing to permit 
separate urine collections from each kidney. 
Only animals in which the urine flow rate of 
the left kidney was at least 85% of that from 
the contralateral kidney were included for 
study. In all animals, surgical losses of fluid 
were replaced with a volume of isotonic saline 
equal to 1% of body wt. Body temperature 


was maintained at 37°. Pyrazinoic aci 
dissolved in a solution of sodium hydi 
(0. 1 M)\ the pH was then adjusted to 1.* 
either hydrochloric acid or sodium bica 
ate. In all control periods, the diluent 
was infused to control for the efTect, i 
of diluent infusion. 

Clearance studies. Clearance studies 
performed in diuretic rats receiving 5% 
nitol in isotonic saline at a rate of 12.0 
so as to reproduce the protocol of the mi 
jection studies which require high urin< 
rates. A priming dose of 50 jnCi of [met 
^H] inulin in one ml of isotonic salin 
infused followed by a sustaining infusi 
isotonic saline containing 25 jLiCi/ml of | 
oxy-^H] inulin at a rate of 1.2 ml/hr. A 
90-min equilibration period, two 20-m 
ine collections were obtained. 1.5 ml ol 
rial blood was obtained at the midpo 
each clearance period and was replacec 
the same volume of blood from a don< 

After collection of samples in the c 
periods, pyrazinoic acid in a dose of 
0.40, 0.80, or 1.6 mA//kg body wt (50, 1 
2(X) mg/kg body wt respectively) was ir 
intravenously as a bolus followed by the 
dose infused per hour. After a 90-min 
ibration period, two or three additional 
ance periods were obtained. In order t< 
trol for possible changes in renal fui 
over the time course of these experii 
five rats were studied under the same pr 
but received no infusion of drug. At th 
elusion of all experiments, the kidneys 
removed, stripped of perirenal fat and c 
and weighed in a Mettler analytic b\ 
(Mettler Instrument Corp., Princeton 

Microinjection studies. Microinjection 
ies were performed in animals receivi 
mannitol in isotonic saline at a rate o 
ml/hr. Inulin was not infused system 
After preparation for study, separate g 
of animals received either diluent infus 
a bolus infusion of pyrazinoic acid oi 

Copyright €> 1978 by the Society for ExperimentMi Biology and Medicine 
^^ rights reaerved. 



1.6 mAf/kg body wt followed by the 
)se per hour. An equilibration period 
nin was permitted to elapse before 
microinjections. Intratubular mi- 
:tions were performed with a solution 
ing [2-^^C]urate (50 fiCi/ml) and 
r^'- H]inulin (100 /iCi/ml) adjusted to 
•7.4 with a solution of NaHCOa (0.357 
er). The concentration of uric acid in 
il solution was 0.24 mAf /liter. Tripli- 
Dplets of 12-20 nl were prepared, one 
:h was utilized for the microinjection 
; other two counted directly for total 
tivity. Microinjections were per- 

into early or late proximal tubular 
er a 60-90 sec interval and total urine 
ons obtained sequentially from both 
nd left kidneys. The procedures for 
ijection, localization of microinjection 
d the calculations of the recovery rates 
ientical to those of Kramp, Lassiter 
)ttschalk (8) and have been described 
il from this laboratory previously (12, 

let studies. Animals were prepared as 
microinjection studies except that 5% 
ol in isotonic saline was infused at 
iflicient to increase the urine flow rate 
150 fil/min per kidney. 100 nanoliters 
[2-^^C]urate and [methoxy'^H]inulin 
1 were placed upon the surface of the 
Iney as a droplet and urine collected 
dally in 15-30 sec aliquots from both 
id left kidneys. A sample of the droplet 
1 was counted directly with each ex- 
nt to determine the ratio of ^^C counts 
x>unts. Droplet studies were obtained 
I control animals and in animals in- 
/ith PZA in doses of 0.40, 0.80, or 1.6 
5 body wt/hr as previously indicated. 
empt was made to quantitate total 

ytical methods. Radioactivity of blood, 
and microinjection and droplet sam- 
as determined in Biofluor (New Eng- 
uclear Corp., Boston, MA) in a Pack- 
ri-Carb Uquid scintillation counter 
rd Instruments Co., Downers Grove, 
ith appropriate corrections for *^C 
appearing in the ^H channel. Counts 
1 were converted to disintegrations per 
fter correction for quench, crossover, 
iciency of counting each isotope. The 

urate concentrations of the serum and urine 
were determined by a uricase method using 
the polarographic sensor in a glucose analyzer 
(Beckman Instruments, Fullerton, CA) as 
previously described (12). The clearances of 
inulin (Cinuiin) and urate (Cumte) are expressed 
as /il/min/g kidney wt and are calculated 
from standard formulae. 

All data are expressed as the mean ± SE 
of the mean. P values were calculated by the 
Fisher / test or the Student / test where ap- 

Results. Clearance studies (Table I). Fol- 
lowing the infusion of PZA in a dose of 0.40 
mA//kg body wt/hr, there was no change in 
the glomerular flltration rate, plasma urate 
concentration or in the clearance of urate. 
The fractional excretion of urate, therefore, 
was unchanged and averaged 21.0 ± 1.3 and 
24.0 ± 2.3% (P = NS) in control and experi- 
mental periods respectively. By contrast, the 
infusion of PZA in a dose of 0.80 mAf/kg 
body wt/hr resulted in signiflcant decreases 
in urate clearance from 276.0 ±25.1 to 210.7 
± 20.6 jLil/min/g kidney wt (P < 0.005) and 
in the fractional excretion of urate from 24.4 
± 2.6 to 19.4 ± 2.4% (/^ < 0.01). The plasma 
concentration of urate increased from 58.3 
± 4.2 to 86.8 ± 5.4 /iA//liter (P < 0.001). The 
infusion of PZA in a dose of 1 .6 mM/kg body 
wt/hr resulted in no change in plasma urate 
concentration, the glomerular filtration rate, 
or the clearance of urate. 

In order to control for the time course of 
these experiments, animals studied in identi- 
cal fashion but not receiving an infusion of 
PZA, had no significant change in the glo- 
merular fihration rate, the plasma urate con- 
centration, or the clearance of urate. 

Microinjection studies (Fig. J). To assess the 
effects of varying dosages of PZA on the 
urate reabsorption process and to localize the 
nephron site of altered reabsorption, intratu- 
bular microinjections were performed into 
early or late portions of the proximal tubule. 
Only samples in which inulin recoveries were 
95% or greater were included for analysis. 
Delayed recoveries ranged from to 6% with 
no significant differences between the groups 
of animals. Accordingly, the results are ex- 
pressed as total urate recoveries and are sum- 
marized on Fig. 1. Recoveries from early 
proximal tubule sites averaged 73 ± 2% in 


TABLE I. The Effects of PZA on the Clearance of Uric Acid." 

C»rft, iil/muk/g kw 

Serum Uric Add iiM/httr 

Cu^ M*/«nin/g kw 


Doic of PZA infused 

NoPZA(ii-S) 1016 ±89.3 9S3 ± S6.4 67.8 ±6J 70.2 ± S.9 


0.40 mJlZ/ks/hr (M - 8) 1069 ± S4.7 1007 ±65.2 70.8 ± 3.0 67.2 ± 3.0 


203.0 ±25.0 217.2 ±22.7 19.2 ±1.7 23.1 ±2J 

221.6 ±12.2 237.5 ±24.7 21.0 ±1J 24.0 ±2J 

0.80inJl//ks/hr(ii-6) 1169 ±75.8 1175 ±92.8 58J±4.2 86.8 ± 5.4 276.0 ±25.1 210.7 ± 20.6 24.4 ± 2.6 19.4 ±14 

p NS <o.ooi <o.oo5 <aoi 

1.60inJl//kg/hr(ii-9) 967 ± 36.4 1054 ± 62.9 61.9 ±4.8 68.4 ± 5.9 230.6 ±25.7 260.6 ± 27.4 24.6 ± 3.5 25.5 ± 3X) 

* Valuei expressed as mean ± SEM. FEu,m. - fractional excretion of uric add; C - control periods; E - experimental periods; NS - not sigmficani; 
(n) «" number of animals studied. 

the left kidney (P < 0.05) and no significant 
change in the right kidney. Compared to 
controls, PZA in a dose of 0.80 m3f Ag body 
wt/hr resulted in a significant decrease in the 
ratio of counts from 1.79 ± 0.04 to 1.19 ± 
0.12 (P < 0.05) and 0.79 ± 0.04 to 0.57 ± 
0.07 {P < 0.05) in the left and right kidneys 
respectively. The largest dose of PZA tested 
(1.6 mA//kg body wt/hr) resulted in a 38% 
decrease in the ratio of counts in the experi- 
mental left kidney (P < 0.05) but no signifi- 
cant change in the right kidney. 

Discussion, The presence of active mecha- 
nisms for the bidirectional transport of urate 
by renal tubular cells has made it difficuh to 
assess the individual contribution of urate 
reabsorption or secretion to the urinary ex- 
cretion of urate by classical clearance tech- 
niques. Pyrazinamide or its active metabolite, 
pyrazinoic acid (PZA), has been extensively 
utilized in man and in the intact animal as a 
pharmacologic aid in assessing the magnitude 
of each of these transport processes (1-3). 
The use of PZA in such studies was based 
upon the observation that, following its ad- 
ministration, the urinary excretion of urate 
was markedly reduced, an effect ascribed to 
an inhibition of urate secretion (1-3). More 
recently, doubt has been cast upon the results 
of studies utilizing the PZA-induced decrease 
of urate excretion as an index of urate secre- 
tion (4-6). 

Prior studies from this and other laborato- 
ries have attempted to estimate urate reab- 
sorption and urate secretion utilizing intra- 
tubular microinjection and droplet precession 
techniques, respectively. The rationale be- 
hind these techniques has been previously 
discussed (8, 9, 12-15). PZA in a dose of 0.40 
mM/kg body wt/hr did not affect the frac- 
tional excretion of urate or the rate of urate 

Early Prolmal Tubule Late Prnimal Tubule 


Fig. I. Per cent of toul [2-'*Cl-uralc recovered fol- 
lowing microinjections in early and late proximal tubule 
sites. •/><0.0l. 

controls. Following infusion of PZA in doses 
of 0.40, 0.80, or 1.6 mM/kg body wt/hr, 
recoveries from early proximal tubule sites 
were 73 ± 1, 64 ± 2, and 7 1 ± 1% respectively. 
The urate recoveries after infusion of 0.80 
mA//kg body wt/hr PZA (64 ± 2%) were 
significantly lower than those obtained in 
controls and in animals infused with PZA in 
doses of either 0.40 or 1.6 mA//kg body 
wt/hr. There were no difTerences in urate 
recoveries following microinjections in late 
proximal tubule sites between any of the 
groups of animals. 

Droplet studies (Table II). Urate secretion 
was considered to be present when the ratio 
of [2-^^C]urate to lmethoxy'^H]\nuhn in the 
first urine sample to contain inulin divided 
by the ratio of "C/^H in the droplet solution 
was greater than one. In control animals not 
receiving PZA, the *^C/^H urine-to-droplet 
ratio of counts averaged 1.79 ± 0.10 in the 
experimental left kidney and 0.79 ± 0.07 in 
the contralateral kidney. The infusion of PZA 
in a dose of 0.40 mAf/kg body wt/hr resulted 
in an 11% decrease in the ratio of counts in 




Precession Droplet Studies." 

LeU kidney 

Right kidney 



')f Change 






'liter (II - 4) 

1.79 ±0.10 
1.97 ±0.07 
1.79 ±0.14 
1.86 ±0.12 

1.79 ± 0.02 
1.76 ±0.06 
1.19 ±0.12 
1.16 ±0.09 



0.79 ± 0.07 
0.79 ± 0.05 
0.79 ± 0.04 
0.69 ± 0.02 

0.79 ± 0.06 
0.76 ± 0.04 
0.57 ± 0.07 
0.67 ± 0.03 



mtrol. E - expcrimenul. Values (mean ± SEM) represent the 'XV 'H urine/droplet ratios or counts in the Hrst urine sample to contain inulin. 

ies following intratubular microinjec- 
I did, however, have a small but meas- 
efTect on urate secretion as assessed by 
)plet studies. This apparent discrep- 
lay indicate that either the degree of 
on of secretion was not physiologically 
ant, or that it could not be detected by 
arance or microinjection techniques, 
ontrast to the 0.40 mM dose, PZA in 
Df 0.80 mM/kg body wt/hr resulted in 
ase in the fractional excretion of urate. 
crease in urate excretion could be the 
)f either inhibition of urate secretion, 
ement of urate reabsorption, or a com- 
n of the two. The results of the preces- 
idies confirm that PZA inhibits urate 
m, the degree of inhibition being 
with the 0.80 mM dose than with the 
M dose. The intratubular microinjec- 
idies indicate that urate absorption is 
cd. The mechanism by which PZA 
ihance urate absorption is unknown, 
eral possibilities might be considered. 
; hand, the decrease in urate recoveries 
represent a direct pharmacologic en- 
lent of urate reabsorption from the 
tal tubule. This suggestion has previ- 
)cen proposed from clearance experi- 
[6, 7). On the other hand, the decreased 
rial recovery of urate following intra- 
r microinjection in animals receiving 
I a dose of 0.80 mM may be due to an 
ion of peritubular uptake of urate 
- urate secretion alone. It is possible 
hibition of urate uptake at the antilu- 
x)rder of the renal tubular cells reduces 
1 concentration of urate, thereby cre- 
more favorable lumen-to-cell gradient 
te. Moreover, inhibition of secretion of 
nto the tubular lumen would increase 
jcific activity of the microinjected [2- 
ite. Prior studies from this laboratory 
idicated that reducing the specific ac- 
)f isotopically labeled urate in the tu- 
lumen does not affect the fractional 

rate of [2-^^C]urate absorption (13). The effect 
of increasing the specific activity, however, 
has not been examined directly and, thus, the 
expected changes in specific activity of 
[2- C]urate microinjected into the tubular 
lumen can not be excluded as a possible 
mechanism, at the present time. The current 
studies do not permit us to differentiate be- 
tween a direct pharmacologic effect of PZA 
on the urate absorptive mechanisms and an 
effect of PZA solely on the secretory process 
with a secondary change in the absorptive 
process, but the results of studies using PZA 
in a dose of 1.6 mAf/kg body wt/hr suggest 
that the latter is the more likely explanation, 
namely that PZA in a dose of 0.80 mM 
enhances urate absorption, primarily by in- 
hibition of the secretory process. With the 
largest dose of PZA tested, fractional urate 
excretion and fractional urate recoveries fol- 
lowing microinjections were similar to control 
values. This dose of PZA also significantly 
inhibited urate secretion. It seems likely that 
PZA, 1.6 mM/kg body wt/hr, not only in- 
hibits secretion, but also inhibits reabsorption 
and, at this dose, secretion and reabsorption 
were inhibited to an equal extent. When 
viewed from this perspective, PZA appears to 
inhibit both urate secretion and urate reab- 
sorption, and the inhibition of these processes 
is dose-dependent, but not necessarily of 
equal sensitivity. It was unfortunate that, due 
to an unacceptably high death rate of the 
animals, higher doses of PZA could not be 

Three previously published studies on the 
eiTect of pyrazinamide or PZA on the renal 
handling of urate in the rat bear directly on 
the results in the present study. A significant 
decrease in urate reabsorption has been re- 
ported by Kramp et ai when single bolus 
doses of PZA of either 10, 50, or 100 mg/kg 
body wt/hr were infused (8). The differences 
between their results and those of the current 
study can not be readily reconciled. In a series 



of clearance studies, Boudry observed a small 
antiuricosuric effect of PZA, an effect which 
became more pronounced when the plasma 
urate concentration was increased (16). In a 
more recent study by Abramson and Levitt, 
there was an increase in net reabsorption by 
the end of the proximal tubule following PZA 
administration, a result ascribed to inhibition 
of secretion (11). Also observed in that study 
was a significant reabsorptive flux of urate in 
the loop of Henle following PZA infusion. In 
the current study, recoveries from late proxi- 
mal tubule sites were lower than controls 
following PZA administration, but the 
changes were not statistically significant. 
Thus, we can not confirm or deny, at this 
time, an effect of PZA in nephron sites be- 
yond the proximal convoluted tubule. 

The use of pyrazinoic acid depression of 
urate excretion as an index of urate secretion 
has been based upon the assumptions that 
PZA inhibits urate secretion and is without 
effect on urate reabsorption. The results of 
the present studies confirm that PZA inhibits 
urate secretion, and thereby may secondarily 
enhance urate absorption. In high doses, 
however, PZA has the additional effect of 
inhibiting urate reabsorption. To the degree 
that PZA may affect both urate secretion and 
reabsorption, any conclusions derived from 
the use of PZA as to the magnitude of the 
contribution of secreted urate to the urinary 
excretion of urate Can not be considered 

Summary. These results indicate that urate 
secretion is inhibited by PZA and that the 
degree of inhibition is dose dependent. In the 
highest dose tested (1.6 mA//kg body wt/hr), 
PZA not only inhibits secretion but also in- 
hibits urate absorption. Thus, PZA appears 
to inhibit both urate secretion and reabsorp- 
tion. The inhibition of these processes is dose 
dependent but not necessarily of equal sen- 

Federation for Clinical Research, Southern 
meeting. New Orleans, LA, January 27-29, 19 
national meeting, Washington, D.C., April 30- 
1977, and have appeared in abstract form in CI 
24: 4 16 A, 1976. The studies were supported in p 
Clinical Investigator award to Dr. Weinman 
Associate Investigator award to Dr. Frankfurt f 
Veterans Administration, and were performed w 
Frankfurt was a Fellow in Nephrology of Baylor 
of Medicine. Pyrazinoic acid was kindly supplier 
George Fanelli of the Merck Institute. The autho 
fully acknowledge the technical assistance of ] 
lock, L. Hawk and S. Sansom and the secretari; 
ance of P. Dunham. Dr. Wadi N. Suki provided 
guidance and advice. 

1. Steele, T. H., and Rieselbach, R. E., Am. J. ! 
868 (1%7). 

2. Yii, T. F., Bcrger, L., Stone, D. J., Wolf, 
Gutman, A. B., Proc. Soc. Exp. Biol. Med. 

3. Gutman, A. B., Yii, T. F., and Berger, L., / 
Med. 47, 575(1969). 

4. Steele, T. H., and Boner, G., J. Clin. Invest. : 

5. Diamond, H. S., and Paolino, J. S., J. Clin 
52, 1491 (1973). 

6. Fanelli, G. M., Jr.. Bohn, D., and Sufford, S 
J. Physiol. 218, 627 (1970). 

7. Fanelli, G. M., Jr., Bohn, D., and Rcilly, S 
J. Physiol. 220, 613 (1971). 

8. Kramp, R. A., Lassiter, W. E., and Gottsc 
W., J. Clin. Invest. 50, 35 (1971). 

9. Kramp, R. A., and Lenoir, R. H., Amer. J. 
229, 1654(1975). 

10. Roch-Ramel, F., and Weiner, 1. M., Amer. 
iol.229, 1604(1975). 

1 1. Abramson, R. G., and Levitt, M. F., Amer. 
iol.230, 1276(1976). 

12. Weinman, E. J., Eknoyan, G., and Suki, V 
CUn. Invest. 55,283(1975). 

13. Weinman, E. J.. Steplock, D., and Ekno^ 
Clin. Res. 24,4 16A( 1976). 

14. Weinman, E. J.. Knight, T. F., McKenzie, 
Eknoyan, G., Kidney Int. 10, 295 (1976). 

15. Weinman, E. J., Steplock, D., Suki, W. 
Eknoyan, G., Amer. J. Physiol. 231, 509 (19 

16. Boudry, J. F., Pfliigers Arch. 328, 279 (1971 

These studies were presented in part to the American Received February 27, 1978. P.S.E.B.M. 1978, \ 


:reased Antiviral Effect of Phosphonoacetic Acid on the Poikilothermic Herpesvirus 
of Channel Catfish Disease (40275) 


Departments of Dermatology, Microbiology and Pathology, University of Miami, School of Medicine, 

Miami, Florida 33152 

xcntly, a characteristic sensitivity to the 
phosphonoacetic acid (PAA) has been 
3nstrated for representative herpesvi- 
; of mammalian (1-8) and avian (9) spe- 
In each reported system virus expression 
3een significantly inhibited in the pres- 
of 100 /ig/ml or less concentration of 
. This mode of inhibition has been de- 
ined to be interference of virus-coded 
i polymerase activity (10, 11) and due to 
specificity the therapeutic aspects of this 
in mammalian herpesvirus systems cur- 
y appear quite promising (12). 
e have investigated PAA in a cold- 
led (poikilothermic) herpesvirus system 
le eventual possibility of disease control, 
inel catfish herpesvirus (CCV) is the eti- 
c agent (13, 14) of an economically dev- 
ing disease well known to the commer- 
aquaculture industry (IS). We found 
expression in cell culture to be inhibited 
\A. However, 10-20 times the drug con- 
ation was required compared to that 
int necessary to inhibit warm-blooded 
leothermic) herpesvirus systems. 
iterials and methods. Viruses, cell, re- 
ts. Channel catfish virus strain Auburn 
/a) originally received from Dr. John 
lb (Auburn University, Auburn, AL) was 
ired at 25° in a continuous cell line of 
n bullhead catfish (BB) cells. Channel 
ih virus strain Homestead (CCVh) was 
ted from an epizootic of chaimel catfish 
disease which occurred in South Florida 
Loment, unpublished). This strain differs 
the Auburn strain in its plaque mor- 
>gy and complete lack of syncytial cell- 
ing cytopathic efTects in BB cell culture. 

orrent address; Dr. Roger Koment, Department of 
biology. The University of South DakoU, School 
didne, Vermilion, South DakoU 57069. 
ddress reprint requests to Dr. Harold Haines, De- 
ent of Pathology, University of Miami School of 
inc, P. O. Box 520875, Miami, Florida 33152. 

BB cells were grown at 25° in 75cm^ plastic 
tissue culture flasks under Eagle's medium 
supplemented with 10% fetal calf serum, 
0.075% sodium bicarbonate, 100 units/ml of 
penicillin and 100 jLig/ml streptomycin. 

Stocks of herpes simplex viruses (HSV) 
type 1 (HSV- 1) strain 2bb and herpes simplex 
virus type 2 (HSV-2) strain 196 were prepared 
in human embryo lung cell cultures (Flow 
2000). Primary rabbit kidney (pRK) and 
baby hamster kidney (BHK) cells were cul- 
tured at 37° under the same growth medium 
as described above for BB cell cultures. 

Disodium phosphonoacetate was obtained 
from Abbott Laboratories (Chicago, IL). Di- 
lutions were prepared in either maintenance 
medium (Eagle's medium supplemented with 
2% fetal calf serum, 0.075% sodium bicarbon- 
ate, 100 units/ml of penicillin and 100 jLtg/ml 
of streptomycin) or overlay medium (Eagle's 
medium supplemented with 0.5% methylcel- 
lulose, 5% fetal calf serum, 0.23% sodium 
bicarbonate, 100 units/ml of penicillin and 
100 /ig/ml of streptomycin). 

Virus plaque assay, plaque reduction by 
PAA. A standard virus plaque assay was de- 
veloped for channel catfish virus in BB cells 
under Eagle's medium containing 0.5% meth- 
ylcellulose. This was with modifications 
based on procedures previously described for 
the in vitro assay of herpes simplex virus (16). 
Briefly, tenfold serial dilutions of CCV were 
prepared and inoculated onto confluent 
monolayers of BB cells in 35 mm plastic 
dishes. After 1 hr. incubation at 25° to allow 
virus adsorption, 2 ml of overlay medium was 
added per dish and cultures incubated at 25° 
in a 5% CO2 atmosphere. Afler 72 hr the 
overlay medium was removed, monolayers 
washed once with phosphate buffered saline 
and stained with 1% crystal violet. Plaques 
formed by HSV at 37° were stained at 48 hr 
afler inoculation. All plaques were counted 
with the aid of a stereomicroscope. 



Copyright ® 1978 b^ the Sodci^ tot lji^T>xcAik\i\%vo\o^ %iA>K«&>r«* 

Ail rights reserved. 



To determine plaque reduction a known 
number of plaque forming units (PFU) was 
inoculated onto cell monolayers in 35 mm 
dishes and overlay medium containing in- 
creasing concentrations of PAA was added. 
The average number of plaques counted on 
replicate cultures without PAA was regarded 
as the 100% value of plaques formed. 

Inhibition of virus by PAA-containing me- 
dium. For multiplicity of infection (MOI) 
studies BB cells were grown in 16 X 125 mm 
tissue culture tubes and monolayers were in- 
oculated with different multiplicities of 
CCVa. Maintenance medium containing in- 
creasing amounts of PAA was added, 1 ml 
per tube. Inoculated control tubes contained 
no PAA. Cultures were maintained at 25° for 

I week with daily observation for cytopathic 
efTect (CPE). We define effective concentra- 
tion of PAA as that amount of drug which 
completely inhibited the induction of detect- 
able virus CPE. 

Results. Virus plaque reduction by PAA. 
CCV in amounts of 200, 100 or 50 plaque 
forming units in separate experiments was 
inoculated onto confluent monolayers of BB 
cells in 35mm dishes. Concentrations of PAA 
ranging from 50 to 2000 /ig/ml in overlay 
medium was applied for 72 hr. The resulting 
data listed in Table I indicates that greater 
than 95% of CCVa plaques were inhibited at 
a final drug concentration of 1000 /ig/ml. 
This relationship remains the same whether 
cultures were infected with 200, 100 or 50 
virus plaque forming units. Likewise, the 
wild-type isolate, CCVh, was similarly in- 
hibited in the plaque r^uction assay. How- 
ever, plaques of this strain were reduced 100% 
by concentrations of 500 /ig PAA/ml, half the 
amount required for the laboratory adapted 
CCVa strain. 

In similar experiments using HSV, 200 
PFU were inoculated onto either BHK or 
pRK cell cultures and concentrations of PAA 
in overlay medium applied for 48 hr. Table 

II indicates that in all cases 97% or more of 
both HSV- 1 and HSV-2 plaques were in- 
hibited at a final PAA concentration of 50 

Effect of PAA on host cell viability. The 
efTect of PAA in high concentrations on BB 
cells was determined as follows. At the begin- 
ning of each experiment viable cell counts, as 

TABLE I. Channel Catfish Virus Plaque 
Reduction by PAA. 






PAA Cone* 



CCVa 200 PFU 



















100 PFU 



















50 PFU 


















CCV„ 200 PFU 

















50 PFU 

















° Channel catfish virus strains Auburn (CCVa) and 
Homestead (CCVh). 

* In fig/ml final concentration. 

•■ Average of four plates per PAA concentration. 

calculated by trypan blue dye exclusion, were 
done on BB cells grown in 3S mm dishes. 
Representative cultures were randomly se- 
lected. Overlay medium containing PAA in 
final concentrations of 0, SOO, and 2000 /ig/ml 
was added to cell cultures containing no virus, 
and at 72 hr viable cell counts were done. 
The data in Table III demonstrate that the 
total number of viable cells was the same in 
PAA treated and untreated BB cell cultures. 
This indicates that no drug toxicity occurred 
during the 72 hr-CCV assay period. In addi- 
tion, parallel BB cell cultures containing 
either 0, 500, or 2000 Mg/ml of PAA were 



^LE II. Herpes Simplex Virus Plaque 
Reduction by PAA. 


PAA No. 

Conc'^ plaques'' 

% Plaque 

1 BHK 

































pes simplex vims type 
^SV.2) strain 1%. 

I (HSV-1) strain 2bb and 

y hamster kidney (BHK) cells and 

sy (pRK) cells. 

g/ml fmal concentration. 

primary rab- 

rage of four plates per 

PAA concentration. 

1, trypsinized and successfully subcul- 
wicc under PAA-free growth medium. 
\tionship of PAA to multiplicity ofinfec- 
9 determine if a PAA dose dependency 

for CCV similar to that reported (2) 
5V, CCVa was prepared in various 
ns and inoculated onto BB cells grown 
le culture tubes. These virus dilutions 
x>nded to multiplicities of infection of 
1.1, 1.0, and 6.0 plaque forming units 
U. Maintenance media containing the 
^AA concentrations as listed in Table 

added to each MOI group of inocu- 
IB cell cultures. Viral CPE for all cul- 
lid not progress beyond 4 days after 
ition, but cultures were observed for a 
of 1 week. Results of these experiments 
:ed that a direct relationship does in- 
xist between PAA concentration and 

MOI. For every tenfold increase in 
nput a twofold increase of drug was 
d for total inhibition of virus cytopath- 

This ranged from 500 /xg PAA/ml 
- 0.01 PFU/cell) to more than 2000 
\/ml (MOI - 6.0 PFU/ceU). The tox- 
vel of PAA in BB cells was evident at 
/ig PAA/ml of culture medium. 
ussion. The herpesviruses are widely 
ed throughout animal phylogeny (17). 
gh they infect a range of species the 

resultant interaction may vary subtly from 
subclinical infection to severe disease to on- 
cogenicity. For many reasons those herpes- 
viruses that parasitize homeothermic animals, 
the mammals and birds, have received most 
research attention. It has been consistently 
found that PAA in amounts of 100 /xg or less 
inhibits the expression of each herpesvirus 
tested. Likewise, our results agree with the 
results of others (2, 4) whereby HSV-1 and 
HSV-2 expression at 37° is inhibited by less 
than 100 /ig PAA/ml. 

The data presented in this report support 
the developing contention that susceptibility 
to inhibition by PAA is a new characteristic 
of the herpesviruses. Furthermore, this char- 
acteristic is apparent in poikilothermic as well 
as homeothermic animal-virus systems. Our 
findings indicate, however, that up to 20 times 
the amount of drug required for other her- 
pesvirus systems is necessary to inhibit CCV. 

Currently the precise mode of virus inhi- 
bition which occurs in our system is unclear. 
In homeothermic systems PAA has been 
shown to interfere with enzymes of viral 
DNA replication (10, 1 1). In view of the vast 
phylogenetic distance between the mamma- 
lian and teleostean cell however, there may 
be difTerences in metabolic reactions to anti- 
viral drugs. If the mode of action is similar 
then the action of PAA may be dependent 
upon either temperature or, relatedly, the 
physiology and metabolic rate of the host cell. 
It is well known that enzyme-substrate reac- 
tions can be directly influenced by tempera- 
ture, and the importance of temperature as a 
catalytic mechanism has been demonstrated 
in the regulation of many life functions of 
poikilothermic species (18). The importance 
of host cell physiology is also suggested by 
the increased tolerance of BB cells to PAA. 
We have observed drug toxicity to occur at 

TABLE in. Viable BB Cell Counts After 
Exposure to PAA. 

PAA cone" 


Viable cell count*^ 

1.4 X 10*^ 


1.9 X 10*^ 



1.9 X 10^ 



2.3 X 10* 

" In fig/ml fmal concentration. 

* In hours. 

*" Trypan blue dye 


total number of cells per 




or about the 2500 /xg/ml level as determined 
by loss of monolayer integrity with concur- 
rent decrease in viable cell counts. 

An alternative hypothesis is that the poi- 
kilothermic virus is itself responsible for the 
increased amount of drug required for inhi- 
bition of virus expression. One means to re- 
solve this question would be a determination 
through a range of temperatures of PAA 
levels inhibiting homeothermic herpesviruses 
in BB cells or CCV in homeothermic cells. 
Unfortunately, these experiments are not now 
possible as BB cells will not support the rep- 
lication of those broad host range homeo- 
thermic herpesviruses tested (HSV, pseudo- 
rabies virus) and CCV will only replicate in 
selected cells of catfish origin. 

The investigation of anti- viral drugs serves 
a twofold purpose: The realization of poten- 
tial for control of acute viral disease and the 
attainment of a further understanding of the 
mechanisms of virus host-cell interaction. A 
clearer insight into both these objectives may 
be obtained by study of the mechanism by 
which poikilothermic channel catfish herpes- 
virus is less sensitive than homeothermic her- 
pesviruses to PAA. 

Summary. Both the laboratory adapted 
Auburn strain and a recently isolated wild- 
type strain of channel catfish herpesvirus 
(CCV) were found to be inhibited by phos- 
phonoacetic acid (PAA) when replicated in 
catfish cell cultures. The inhibition of virus 
cytopathic effect by PAA exhibited a direct 
relationship between the multiplicity of infec- 
tion and amount of drug required. However, 
in this poikilothermic system up to 20 times 
the amount of PAA required for inhibition of 
homeothermic herpesvirus systems was found 
necessary to inhibit CCV cytopathology. 

Nordeen, C. W., Ovcrby, L. R., Rodenck, W. R., 
Schleicher, J. B., and Von Esch, A. M., Appl. Micro- 
biol. 2d, 264(1973). 

2. Overby, L. R., Robishaw, E. £., Schleicher, J. B.. 
Rueter, A., Shipkowitz, N. L., and Mao, J. C. H.. 
Antimicrob. Agents Chemother. 6, 360 ( 1974). 

3. Huang, E-S., J. Virol. 16, 1560 (1975). 

4. Duff, R. G., and Overby, L. R., Bad. Proc. p240 

5. Summers, W. C, and Klein, G., J. Virol. 18, 151 

6. Yajima, Y., Tanaka, A., and Nonoyama, M., Virol- 
ogy 71, 352 (1976). 

7. Barahona, H., Daniel, M. D., Bekesi, J. G., Fraser, 
C. E. O., King, N. W., Hunt, R. D., Ingalls, J. K., 
and Jones, T. C, Proc. Soc. Exp. Biol. Med. 154,431 

8. May, D. B., Miller, R. L., and Rapp. F., Iniervirol. 

9. Lee, L. F., Nazerian, K., Leinbach, S. S., Reno, J. 
M., and Boezi, J. A., J. Nat. Cancer Inst. 56, 823 

10. Mao, J. C. H., and Robishaw, E. E., Biochemistry 

1 1. Uinbach, S. S., Reno, J. M., Lee, L. F., IsbeU, A. P., 
and Boezi, J. A., Biochemistry 15, 426 (1976). 

12. Hay, J., Brown, S. M., Jamieson, A. T., Rixon, F. J., 
Moss, H., Dargan, D. A., and Subak-Sharpe, J. H., 
J. Antimicrobiol. Chemother. 3, 63 (1977). 

13. Fijan, N. N., Wellborn, T. L., Jr., and Naftel, J. P.. 
U. S. Dept. Interior Bur. Sport Fish. Wildlife Tech. 
Paper 43, 1(1970). 

14. Wolf, K., and Darlington, R. W., J. Virol. 8, 525 

15. Plumb, J. A., Proceedings Ann. Conf. Southeastern 
Assoc. Game and Fish Commissioners p489 (1971). 

16. Koment, R. W., and Rapp, F., J. Virol. 15, 812 

17. Nahmias, A. J., in "Pathobiology Annual" (H. L 
loachim, ed.) p 153, Appleton-Century-Crofts, New 
York (1972). 

18. Swan, H., "Thermoregulation and Biocnergetics", 
pp430. American Elsevier Publishing Co., Inc., New 
York (1974). 

1. Shipkowitz, N. L., Bower, R. R., Appell, R. N., Received December 12, 1977. P.S.E.B.M. 1978, Vol. 159. 


iffects of Indomethacin and Meclofenamate on Estrogen Induced Vasodilation in 
the Rabbit Uterus' (40276) 



rtmenis of Physiology and Gynecology-Obstetrics, University of Wisconsin Medical School and Wisconsin 
Perinatal Center, Madison General Hospital, Madison, Wisconsin 53715 

y studies have shown that estrogen 
es the blood flow to the pregnant and 
egnant uterus (1-3), but the mecha- 
y which this vasodilation occurs has 
;n determined. 

is been postulated that prostaglandins 
lediate this vasodilation. Prostaglan- 
ive been shown to play a role in the 
ion of blood flow in the pregnant 
Items (4) and some studies have indi- 
hat prostaglandins affect blood flows 
•pregnant uteri (S). Some investigators 
Iso reported flnding increased prosta- 
1 synthesis in uterine tissue following 
n treatment (6, 7). 

following experiment was designed to 
he response of the uterine vasculature 
ogen treatment in the rabbit and to 
ine if prostaglandins are involved in 
sponse through the use of the known 
^landin synthetase inhibitors, indo- 
nn and meclofenamate. 
erials and methods. Non-pregnant fe- 
"^ew Zealand white rabbits weighing 
• kg were used in this study. Surgery 
erformed under Nembutal (Abbott 
sedation supplemented by local xylo- 
Astra Pharm.). A left ventricular cath- 
d. 0.0288 mm) was placed via the left 
I artery and a second polyvinyl catheter 
serted 8-10 cm into the left femoral 
The femoral catheter was then led to 
:k incision via a subcutaneous tunnel, 
theters were secured to a packet made 
;ical tape and attached to the rabbit's 
Experiments were performed the fol- 

day with the awake animal resting 

in a restraining cage. 

mean arterial blood pressure of the 
ivas monitored with a Statham P23Db 

x>rted by Grant No. HEX)6736. 

transducer attached to the femoral catheter. 
Records were made with an R411 Beckman 
recorder with an EO-18 oscilloscope display. 

Blood flows were determined by the left 
ventricular injection of IS micron micro- 
spheres (3M Co., New England Nuclear) la- 
belled with either ^^'Gd, ^'Sn, ^Sr or '^Sc. 
The spheres were prepared as a suspension in 
10% Dextran in saline. Each microsphere in- 
jection had a volume of 0.1-0.2 ml and con- 
tained approximately O.S million spheres. 

Withdrawal. The microspheres were in- 
jected into the left ventricle while simultane- 
ously withdrawing an integrated arterial 
blood sample from the femoral catheter at a 
rate of 2.06 ml/min for l.S min, starting from 
the time of sphere injection. 

Response to estrogen. In Ave animals, the 
control organ blood flows were determined. 
A solution of 1 mg/ml beta estradiol diacetate 
(Sigma) in 95% ETOH was then administered 
at a dosage of 100 fig/kg of body weight via 
the left ventricle. A second determination of 
the organ blood flows was made 2 hr after 
the estrogen treatment. 

Effect of indomethacin pretreatment. In this 
series seven rabbits were pretreated with a 
100 mg/ml solution of indomethacin dis- 
solved in dimethyl sulfoxide at a dosage of 
20 mg/kg of body wt. Indomethacin was 
given 30 min prior to the control blood flow 
measurement, and again 30 min before the 
final measurement of blood flow. The eflect 
of estrogen on the uterine blood flow was 
measured as described above. 

Effect of meclofenamate pretreatment. In 
this series meclofenamate was administered 
to eight rabbits as a 20 mg/ml saline solution 
in a dosage of 20 mg/kg of body wt. The 
meclofenamate was given 30 min prior to the 
control blood flow measurement and again 
30 min before the final measurement of blood 


0037-9727/78/ 159 l-0025$0l .00/0 

Copyright ® 1978 h^ the So6k\^ tot ^^ienxEATiVa\%vc>V»^ vBA>Kt&vsAgt 




flow. The effect of estrogen on the uterine 
blood flow was measured as described above. 

Assay. Upon completion of the experiment, 
the animal was sacriflced and the uterus, 
kidneys and lungs were removed. Care was 
taken to dissect free any adipose or connec- 
tive tissue from the organs. The uterus was 
dissected into five separate samples, and the 
kidneys into three samples each. Two lung 
samples were also taken, one sample coming 
from each main lobe of the lungs. Lung sam- 
ples were taken for assay to determine that 
no shunting of microspheres across the vas- 
cular bed had occurred. The tissues were 
weighed and placed in counting vials. No 
sample vial contained tissue which extended 
more than 1 cm above the bottom of the vial. 

Standard vials were used in assaying the 
samples. Each standard vial contained a 
known number of spheres of one of the iso- 
topes used in the experiment embedded in 
wax approximately O.S cm from the bottom 
of the vial. All measurements of radioactivity 
were made with a three-channel, well-type, 
automatic y counting system (Nuclear Chi- 
cago, model 118S). A standard pattern of 
counting the samples was used in which the 
standard vials were followed by the blood 
samples, obtained during the integrated ar- 
terial withdrawal, followed by the tissue sam- 
ples. The data were printed on paper tape 
which was fed into a Univac 1110 computer 
via an interactive terminal. The data were 
then processed through programs developed 
by our laboratory. The spillovers of each 
isotope into the other channels was deter- 
mined from the standard vials and the counts 
per minute per sphere were also calculated at 
this time. Data were reduced to counts per 
minute and the number of spheres in each 
sample. Organ blood flows and vascular re- 
sistances per gram of tissue and the ratios of 
test resistance to control resistance (T/C) for 
each tissue sample were also calculated. All 
results are expressed as the mean ± the stan- 
dard error of the mean. Statistical analysis 
included paired and un-paired / tests (where 
appropriate) to compare control and test ob- 

Dosage and vehicle. The vehicle for the 
estrogen was ethanol. The dosage adminis- 
tered was small (<.35 ml) and the measure- 
ments of blood flow were made 2 hr after the 

administration of this substance. It is unlikely 
that the presence of ethanol was a significant 
factor in these experiments because ethanol 
was present in both the control (estrogen 
only) studies and in the studies using prosta- 
glandin synthetase inhibitors. The indometh- 
acin was administered with <1 ml of DMSO 
(dimethyl sulfoxide) and our observations 
were made after a delay of 30 min. We have 
examined the cardiovascular effects of 
DMSO in the sheep and have observed no 
significant cardiovascular responses to this 
agent 30 min after its administration. 

The dose levels of indomethacin and mc- 
clofenamate were selected to ensure some 
degree of prostaglandin synthetase inhibition. 
Ryan et al. (5) used 20 mg/kg/day of meclo- 
fenamate and 5 mg/kg/day of indomethacin. 
Venuto et al. (8) have shown that 2 mg/kg of 
indomethacin or meclofenamate both reduce 
uterine venous prostaglandin E2 levels in 
pregnant rabbits. 

Results. Part 1. Responses to estrogen. The 
results obtained in five rabbits are presented 
in Table I. Organ blood flows were measured 
before (Control) and 2 hr after (Test) treat- 
ment with 0. 1 mg/kg estrogen. Mean arterial 
blood pressures were not affected by the es- 
trogen treatment. In each of the five animals, 
the vascular resistance of the uterus decreased 
in response to estrogen. The change in mean 
resistance of 192.96 ± 32.5 in the control stale 
to 36.92 ± 8.5 mm Hg X min/ml x g after 
estrogen, was significant {P < .003). The 
renal vascular resistance was not affected by 
estrogen treatment. 

Part 2. Pretreatment with indomethacin. The 
organ blood flows in seven rabbits which had 
been pretreated with 20 mg/kg indomethacin 
were measured both before (Control) and 
again 2 hr after (Test) estrogen treatment. 
The results are presented in Tables II and III. 
Mean arterial blood pressures were not af- 
fected by indomethacin pretreatment. Uter- 
ine vascular resistance was also not signifi- 
cantly affected by the indomethacin. The 
renal vascular resistance increased from a 
mean control value of 24.63 ± 3.0 to 38.51 
± 5.2 mm Hg x min/ml X g (Table III). This 
was a significant increase {P < .04) due to 
indomethacin pretreatment. 

Following indomethacin pretreatment, 
mean arterial blood pressures were not af- 



fected by estrogen treatment. Uterine vascu- 
lar resistance decreased from a control value 
of299.13 ± 69.1 to 137.90 ± 47.3 mm Hg X 
min/ml X g (P < .004) after pretreatment 
with estrogen (Table II). The renal vascula- 
ture was not afTected by the estrogen treat- 

Comparisons were made of the resistance 
ratios (T/C) between normal rabbits and rab- 
bits which had been pretreated with indo- 
methacin to determine any afTect which in- 
domethacin might have on the vascular re- 
sponse to estrogen treatment (Table IV). The 
untreated uterus had a mean T/C value of 



Blood pressure (mm 

Resistance (mm 

Hg X min)/ml x g 

Uterine resistance 

Renal resistance 







C T 







13.37 25.48 







29.26 24.66 







23.09 15.11 







28.32 27.53 







29.09 30.77 








24.62 24.71 








±3.0 ±2.6 





' The uterine and renal vascular resistance 

per gram of tissue of five rabbits before (C) and 2 h after (T) the 

administration of 0. 1 m 

g/kg estrogen. Mean arterial blood pressures and resistance ratios (T/C) are alsc 




Blood pressure 
(mm Hg) 

Resistance (mm 

Hg X min)/ml x g 

Uterine resistance 

Renal resistance 







C T 







23.65 77.01 







33.50 22.59 







30.45 30.35 







41.3S 44.44 







65.52 85.73 







31.89 21.71 







43.17 38.56 








38.51 45.77 








±5.2 ±9.9 






*The uterine and renal vascular resistances per gram of tissue of seven rabbits pretreated with 20 /xg/kg 
indomethacin before (C) and 2 hs after (T) the administration of 0. 1 mg/kg estrogen. Mean arterial blood pressures 
tnd resistance ratios (T/C) are also given. 


Resistance (mm Hg x min)/ml x g 

Blood pressure (mm Hg) 

Uterine resistance 

Renal resistance 





N P 













24.63 38.51 

±3.0 ±5.2 

5 7 


" A comparison of the uterine and renal vascular resistance per gram of tissue during the control period of five 
Bormal (N) rabbits and seven rabbits pretreated (P) with 20 mg/kg indomethacin. A comparison of the mean arterial 
blood pressures is also provided. 



0.182 which differed significantly (P < .05) 
from the pretreated T/C value of 0.429. In- 
domethacin depressed the uterine response to 

Part 3. Pretreatment with meclofenamate. 
The organ blood flows in eight animals pre- 


Resistance ratios 



acin pre- 

amate pre- 











0.182 0.429 

±0.02 ±0.11 
5 7 


0.182 0.158 

±0.02 ±0.03 
5 8 


" The effect of 0. 1 mg/kg estrogen on the uterine 
vasculature of five normal rabbits and rabbits pretreated 
with either 20 mg/kg indomethacin or 20 mgAg meclo- 
fenamate. The data are expressed as ratios (T/C) of the 
resistance 2 hr after estrogen treatment (T) to that seen 
before administration of the estrogen (C). 

treated with 20 mg/kg meclofenamate were 
measured both before (Control) and again 2 
hr after (Test) estrogen treatment. The results 
are presented in Tables V and VI. Meclofen- 
amate pretreatment had no effect on the 
mean arterial blood pressure. The uterine 
vascular resistance increased from a mean 
control value of 192.96 ± 32.5 to 416.42 ± 
72.6 mm Hg x min/ml X g (P < .02) follow- 
ing the meclofenamate treatment. The renal 
vascular resistance significantly increased 
from 24.63 ± 3.0 to 40.33 ± 6.1 nmi Hg X 
min/ml X g (P < .04) after pretreatment with 
meclofenamate (Table VI). 

Following pretreatment with meclofena- 
mate the mean arterial blood pressure was 
not affected by estrogen treatment. The uter- 
ine vascular resistance significantly decreased 
from a mean value of 416.42 ± 72.6 before 
estrogen to 69.58 ±21.8 mm Hg x min/ml 
X g (P < .001) after estrogen (Table V). The 
renal vasculature was not affected by the 



Blood pressure 

Resistance (mm 

Hg X min)/ml X g 

(mm Hg) 

Uterine resistance 

Renal resistance 


































































































" The uterine and renal vascular resistances per gram of tissue of eight rabbits pretreated with 20 MgAg 
meclofenamate before (C) and 2 hr after (T) the administration of 0.1 mg/kg estrogen. Mean arterial blood pressures 
and resistance ratios (T/C) are also given. 


Blood pressure (mm Hg) 
N P 

Resistance (mm Hg 

X min)/ml X g 

Uterine resisunce 
N P 

N P 



90 88 

±7 ±3 
5 8 


192.96 416.42 
±32.5 ±72.6 
5 8 

24.63 40.33 
±3.0 ±6.1 
5 8 

" A comparison of the uterine and renal vascular resistance per gram of tissue during the control period of five 
normal (N) rabbits and eight rabbits pretreated (P) with 20 mg/kg meclofenamate. A comparison of the mean 
arterial blood pressures is also provided. 



tn treatment. Meclofenamate did not 
the uterine vascular response to estro- 

lission. It has been postulated that es- 
induced uterine vasodilation is me- 
via a biochemical chain of events ini- 
by estrogen receptor binding and con- 
g to the synthesis of new mRNA and, 
[uently, the synthesis of new protein 
illam et al (10) have descritNsd the 
of estrogen on the sheep uterus and 
ndicated a possible release of acetyl- 
e or histamine as the intermediate step 
chain of events. Clark et al, however, 
letermined that the administration of 
line receptor antagonists has no effect 
trogen induced increases in uterine 
volume (11). Resnik et al (12) also 
oncluded that acetylcholine, isoproter- 
nd histamine are not mediators of the 
Lse to estrogen and proposed the release 
aall polypeptide such as bradykinin or 
mzyme which has a role in the biosyn- 
of adenosine and its release (2). Several 
> have indicated an increase in the syn- 
of uterine prostaglandins after estrogen 
ent (5-7). Ryan et al. have shown that 
glandins exhibit properties concurrent 
be hypothesis that prostaglandins me- 
^trogen induced hyperemia (5). They 
[low that blocking prostaglandin syn- 
with both indomethacin and meclofen- 
depressed the uterine response to es- 
in rats. Castracane and Jordan, how- 
lave found that inhibiting protein syn- 
and thereby blocking the biological 
of events leading to the hyperemic re- 
;, had no effect on the estrogen induced 
tion of prostaglandins (13). They con- 
that the production of prostaglandins 
estrogen may be a function of estrogen 
ted to its function as an initiator of 
:n induced hyperemia. 
» study was designed to determine 
er prostaglandin synthesis is a neces- 
:ep in the mediation of the estrogenic 
Lse. The experimental data presented in 
per indicate that estrogen induced vas- 
on is not mediated by prostaglandin 
ion. The vasoconstriction shown to be 
ing in the kidneys following treatment 
ither indomethacin or meclofenamate 
es that prostaglandin synthesis block- 
burred in concordance with the study 

by Malik and McGifTon prostaglandin mod- 
ulation of vascular resistance in rabbit kid- 
neys ( 14). The uterus showed no vasoconstric- 
tion due to indomethacin so that the vasocon- 
striction seen after meclofenamate may have 
been due to a side effect of the drug. The fact 
that indomethacin depressed the uterine re- 
sponse to estrogen is in concordance with the 
Uterature, but must be examined in view of 
the fact that meclofenamate did not produce 
a similar response. It is our conclusion that 
the indomethacin induced depression of the 
response to estrogen was not due to the block- 
ade of prostaglandin synthesis, but due to a 
side effect of indomethacin or its vehicle. 
Therefore, prostaglandin synthesis does not 
appear to be essential to estrogen induced 
vasodilation in the rabbit uterus. 

1. Anderson, S. G., and Hackshaw, B. T., Amer. J. 
Obslct. Gynecol. 119, 589 (1974). 

2. Resnik, R.. Killam, A. P., BalUglia, F. C, Ma- 
kowski, E. L., and Meschia, G., Endocrinology !M, 

3. Roscnfeld, C. R., Morriss, F. H.. Jr.. Ballaglia, F. C, 
Makowski, £. L., and Meschia, G., Amer. J. Obstet. 
Gynecol. 124,618(1976). 

4. Rankin, J. H. G., and Phcmetion, T. M.. Amer. J. 
Physiol. 231, 754 (1976). 

5. Ryan, M. J., Clark, K. E., VanOrdcn, D. E., Farley, 
D., Edvinsson, L., Sjoberg, N. O., VanOrden III, L. 
S., and Brody, M. J., ProsUglandins 5, 257 (1974). 

6. Ham, E. A., Cirillo, V. J., Zanetti, M. E., and Kuehl, 
F. A., Jr., Proc. Nat. Acad. Sci. U.S.A. 72, 1420 

7. Saksena, S. K., and Lau, I. F., ProsUglandins 3, 317 

8. Venuto, R. C, O'Dorisio, T., Slein, J. H., and Ferris, 
T. F., J. Clin. Invest. 55, 193 (1975). 

9. Gorski, J., Toft, D., Shyamala, G., Smith, D., and 
Notides. A., Rec. Progr. Hormone Res. 24,45 (1968). 

10. Killam, A. P., Rosenfeld, C. R., Baltaglia, F. C, 
Makowski, E. L., and Meschia, G., Amer. J. Obstet. 
Gynecol. 115, 1045(1973). 

11. Clark, K. E., Farley, D. B., VanOrden, D. E., and 
Brody, M. J., Proc. Soc. Exp. Biol. Med. 156, 411 

12. Resnik. R., Killam, A. P.. Barton, M. D., BatUglia, 
F. C, Makowski, E. L., and Meschia, G.. Amer. J. 
Obstet. Gynecol. 125, 201 (1976). 

13. Castracane, V. D., and Jordan, V. C, Prostaglandins 

14. Malik, K. U., and McGifT. J. C, Circ. Res. 36, 599 

Received March 10, 1978. P.S.E.B.M. 1978, Vol. 159. 


Superoxide Dismutase in Bovine Fetal Ductus Arteriosus, Thoracic Aorta, and 
Pulmonary and Umbilical Arteries (40277) 


Division of Biochemistry, Physiology and Pharmacology, The University of South Dakota School of Medicine, 

Vermillion, South Dakota 57069 

Soon after birth the lumens of the ductus 
arteriosus and the umbilical artery are oblit- 
erated. Some researchers have suggested that 
oxygen toxicity, resulting from the increased 
arterial oxygen tension occurring after birth 
and the development of an "active hypersen- 
sitivity" reaction to oxygen prior to birth, is 
the cause of widespread intracellular and ex- 
tracellular destruction noted in the subintimal 
regions of the media (muscular layer), as well 
as the intimal layer itself in the ductus arter- 
iosus (1, 2). This is similar to the explanation 
offered for the closure of premature infants' 
retinal arteries and consequent retrolental fi- 
broplasia and blindness from exposure to 
excessively high oxygen levels in the hyper- 
baric chamber (1, 3). Other workers (4-6) 
have presented histological and other evi- 
dence that the ductus arteriosus of guinea pig, 
rabbit, rat, and mouse fetuses allowed to 
breathe showed widespread intracellular and 
matrix destruction in the histological regions 
previously mentioned; such changes were not 
noted in fetuses frozen without permitting 

A suitable explanation for the temporal 
relationship between increasing arterial oxy- 
gen tensions and cellular degeneration could 
certainly be superoxide radicals, hydroper- 
oxides, and hydroxyl radicals, all powerful 
oxidants which are destructive in biological 
systems of cells existing in aerobic conditions 
(3, 7, 8). It has been demonstrated that rats 
exposed to gradually increasing levels of O2 
in their environment had significantly greater 
levels of SOD in lung tissue and survived for 
significantly longer periods of time after ex- 
posure to toxic levels of O2 than did control 
rats (7, 8). We postulated that an overpro- 
duction of hydroxyl radicals could result after 
an increase in oxygen tension in tissues such 
as the ductus arteriosus and umbilical artery, 
if such tissues possessed lower levels of SOD, 
as compared with such permanent tissues as 

the pulmonary artery and thoracic aorta. Su- 
peroxide dismutase catalyzes the dismutation 
of two molecules of the superoxide anion 
forming one molecule each of oxygen and 
hydrogen peroxide, while the superoxide an- 
ion and hydrogen peroxide can react to form 
the hydroxyl radical: 

(a) 2O2- + 2H-^ '- 

O2 + H2O2 

• (Superoxide Dismutase Reaction) 

(b) O2" + H2O2 ^ HO" + HO 

-k- O2 (Haber- Weiss Reaction) 

SOD functions to remove one of the reactants 
of the Haber- Weiss reaction, and this enzyme 
has been extensively studied by McCord and 
Fridovich (3), as well as by others. 

In this study we chose to examine the levels 
of SOD in several tissues of bovine fetuses, 
using an enzymic activity assay and a radial 
inmiunodiffusion assay. There were two 
groups of two tissues used for comparative 
purposes; the two fetal blood vessels which 
obliterate after birth, the ductus arteriosus 
and umbilical artery, and the two blood ves- 
sels which do not obliterate after birth, the 
pulmonary artery and the thoracic aorta. Re- 
ported herein are the results of this study 
which we feel are in support of our hypoth- 

Materials and methods, A local meat pack- 
ing firm allowed us access to fetal calves 
approximately forty minutes following the 
killing of the mother. Gestational ages of the 
calves were estimated using such criteria as 
crown-rump length, body hair patterns, and 
the presence of erupted incisor teeth (9). 
Eighty percent of the calves in this study were 
"full-term" by the criteria mentioned. The 
thoracic cavity was then entered and an en 
bloc excision of the heart, great vessels, and 
the entire length of the thoracic aorta was 
performed; additionally, a small segment of 


0037-9727/78/ 1591 -0030$0 1 .00/0 

Copyright <D 1978 by the Society for Experimental Biology and Medicine 
nifJ^iM rettrved. 



ical artery was obtained from the um- 
. cord. The great vessels were then iden- 
dissected free, excised, washed three 
in 0. 1 S A/ NaCl, immediately frozen on 
e and stored for 2 weeks at -30"*. The 
s were then thawed, washed again three 
in 0.1 S A/ NaCl, homogenized in 4 vol 
S M NaCl with a Tenbroeck glass ho- 
lizer, and centrifuged at 100,000g for 
n at 2^. The supernatant was then used 
lalysis of SOD activity. If there is sig- 
nt blood in the prepared tissue, this wUl 
bate to the total SOD activity of the 
e as erythrocytes do possess significant 
of SOD. Whole blood was obtained 
Four fetal calves in the study. The eryth- 
» were lysed with an equal volume of 
ized water and the solution then re- 
l to 0. 15 A/ NaCl. The lysed erythrocyte 
re was then centrifuged and the super- 
t was diluted with 0.15 M NaCl to 
I solutions with hemoglobin concentra- 
in ranges comparable to the tissue su- 
tants ft-om the blood vessel prepara- 
Hemoglobin levels of the lysed eryth- 
; and blood vessel supematants were 
measured at 24.7 cm~^ using a Cary 
spectrophotometer. SOD activity of the 
erythrocyte supematants was measured. 
D activity was measured using the xan- 
oxidase-cytochrome c assay of McCord 
ridovich (10). Bovine erythrocyte SOD 
»urified to electrophoretic homogeneity 
the method of McCord and Fridovich 
This preparation was used to prepare 
xlics in rabbits. Immune rabbit y-glob- 
were isolated as previously described, 
hese were used to determine the levels 
D using a radial immunodiffusion assay 
Bovine xanthine oxidase was purified to 
geneity from raw cream (12). Protein 
ntrations were determined by the 
xl of Lowry et al. (13). 
I ' HCl and base and cytochrome c were 
led from the Sigma Chemical Com- 
Agar was obtained from Difco Com- 
All other chemicals were reagent grade 


ndts and discussion. The results of en- 
: and immunochemical assays for SOD 
ir tissues from thirteen bovine fetuses 
resented in Table I. In all individual 
ds the levels of SOD determined in the 

four tissues indicated that the ductus arterio- 
sus and umbilical artery were always lower 
than the pulmonary artery and thoracic aorta 
although a comparison between animals did 
not always follow this pattern. 

The data were compared using the ""t" test 
of significance and the results of such com- 
parisons are shown in Table II. As can be 
seen, in nearly all comparisons the levels of 
SOD in the ductus arteriosus and umbiUcal 
artery were statistically significantly lower 
than those found in the pulmonary artery and 
thoracic aorta. The level of SOD in the ductus 
arteriosus and the umbiUcal artery were not 
statistically significantly different from each 
other. Likewise, the levels of SOD in the 
pulmonary artery and thoracic aorta were not 
statistically significantly different from each 

Erythrocytes do contribute to the SOD ac- 
tivity of tissue extracts although this contri- 
bution is negligible if it is possible to wash 
the tissues relatively free from blood (7). In 
this study the hemoglobin in the tissue super- 
natants was in the range of 1-2 x 10" A/. 

TABLE 1. Bovine Fetal Superoxide Dismutase. 


enzyme units" 

mg protein 

Radial im- 


mg protein 

Ductus Arteriosus 
Umbilical Artery 
Pulmonary Artery 
Thoracic Aorta 

2.32 ± 0.33" 
1.97 ±0.16 
3.64 ± 0.32 
3.45 ±0.31 

77.8 ± S.%' 
80.2 ± 7.8 
94.7 ± 6.7 
113.9 ±8.3 

° Determined with xanthine oxidase-cytochrome c 
assay, expressed per mg cytosolic protein. 

^ Expressed as /ig superoxide dismutase per mg cyto- 
solic protein. 

'^ Values are expressed as the mean ± SE for thirteen 
samples run in duplicate. 


*Y* Test of Significance." 

Paired tissues 

Enzyme assay RID 


0.001 0.01 
0.05 O.Ol 
0.001 0.05 
0.001 0.02 


0.4 0.7 
0.7 0.2 

" The data of Table I were analyzed by pairing the 
indicated tissues. The confidence levels are indicated for 
the two types of assay, enzymic and inmiunochemical 
(RID, radial immunodiffusion). 



The SOD activity of the lysed erythrocyte 
supematants in this hemoglobin concentra- 
tion range was negligible (less than one per- 
cent). Additional evidence to discount the 
contribution of erythrocytes in this study is 
noted in that the hemoglobin concentrations 
varied randomly in the tissue samples and 
did not correlate with the differences between 
the SOD activity of the blood vessel prepa- 

Undoubtedly the etiology of ductus arter- 
iosus closure is multifactorial, and it is not 
possible here to elaborate the numerous 
mechanisms proposed (14-16). It is helpful to 
view ductus arteriosus closure as both a phys- 
iological and anatomical event; that is to say, 
the ductus arteriosus responds to varying ox- 
ygen tensions and hemodynamic changes by 
changing its lumen size in situ, and it under- 
goes obliterative fibrotic changes to ulti- 
mately become the ligamentum arteriosum in 
the usual case. The in vitro responsiveness of 
this vessel to varying oxygen tensions has 
been consistently reported in the literature. 
The role of prostaglandins in the closure of 
the ductus arteriosus is also of current interest 

This study suggests that a deflciency of 
SOD could contribute to the degenerative 
cellular changes presumed to occur as part of 
the obliterative process in the bovine ductus 
arteriosus and umbilical arteries. Further 
studies need to be conducted to determine if 
the rise in arterial oxygen tension at parturi- 
tion is sufficient to create the oxidant stress 
this study is proposing. In addition we are 
not capable of ascertaining the distribution of 
SOD across the wall of the tissues we have 
examined, which might be of importance in 
the degeneration of the ductus arteriosus and 
umbilical artery. Perhaps in the future a his- 
tological stain for SOD of sufficient sensitiv- 
ity will be developed and can be used to 
answer such questions. 

The levels of SOD seen in the ductus ar- 
teriosus and in the umbilical artery are 
54-67% (activity assay) and 68-84% (RID 
assay) of the levels found in the pulmonary 
artery and in the thoracic aorta. Michelson et 
al (20) have suggested that "levels of less 
than 50% of the normal mean for superoxide 
dismutase are more or less lethal due to the 
increased toxicity of uncontrolled superox- 
ide." This contention was based on a survey 

of SOD activities in a cross section of the 
human population in France, comparing nor- 
mal and abnormal populations. Extremely 
low levels of SOD correlated in several cases 
with associated physical and mental prob- 
lems. The ability of a newborn to handle an 
increased flux of superoxide, consequent to 
exposure to increased oxygen tensions, may 
reflect the absolute and quantitative amounts 
of SOD present in particular tissues. Those 
with high levels of SOD will survive, and 
those with low levels of SOD will degenerate. 
Summary. Soon after birth the lumens of 
the ductus arteriosus (DA) and umbilical ar- 
tery (UA) are obliterated. It has been sug- 
gested that oxygen toxicity, resulting from an 
increased oxygen tension, is the cause of this 
destruction with superoxide radicals and hy- ' 
droxyl radicals being implicated as mediators. 
A deflciency of superoxide dismutase (SOD) ' 
in these tissues was hypothesized as being 
responsible for an increase in the levels of 
superoxide and hydroxyl radicals. SOD levels 
were determined enzymatically and inmiu- 
nochemically in four tissues obtained from 
thirteen bovine fetuses. SOD levels in the DA 
and UA were found by both assays to be 
statistically signiflcantly lower than that 
found in such permanent vessels as the pul- 
monary artery and thoracic aorta. These data 
are in support of the hypothesis that a lower 
level of SOD in the ductus arteriosus and 
umbilical artery may contribute to the rapid 
deterioration of these tissues upon exposure 
to greatly increased oxygen tensions. 

The kind assistance of Iowa Beef Processors of Dakota 
City, Nebraska in obtaining calf fetuses is acknowledged. 
The technical assistance of two fellow medical students. 
Curt Bucholz and Ron Thune, and of an undergraduate 
student, Mark Martin, is appreciated. Supported by Gen- 
eral Research Support Grant from the National Insti- 
tutes of Health (USPHS 01 RR 05421-14). Frank O. 
Brady is a Research Career Development Awardee of 
the National Institute of Environmental Health Sciences, 
NIH (ES 00022). 

1. Bor, I., and Guntheroth, W. G., Can. J. Physiol. 

2. Stingle, S., Vozehova. S., Krisha, M., Folia Morphol. 

3. Fridovich, I., Adv. Enz. 41, 35 (1974). 

4. Homblad, P. Y., Acta Physiol. Scand. 76, 49 (1969). 

5. Homblad, P. Y., Acta Physiol. Scand. 76, 58 (1969). 

6. Record, R. G.. and McKeown, T., Clin. Sci.l4, 213 




Crapo, J. D.. and Ticracy, D. F., Amer. J. Physiol. 


Kimball, R. E.. Reddy, K., Pierce, T. H., Schwartz, 

L. W., Mustafa, M. G.. and Cross, C. E., Amer. J. 

Physiol. 230, 1425(1976). 

Benescfa, F., and Wright, J. G., Veterinary Obstetrics, 

Williams and Wilkins Co., Baltimore (1951). 

McCord, J. M., and Fridovich, 1., J. Biol. Chem. 

244, 6049 (1%9). 

Fahey, J. L., and McKelvey, E. M.. J. Immunol. H 


Waud, W. R., Brady, F. O., WUcy, R. D., and 

RajagofNdan, K. V., Arch. Biochem. Biophys. 169, 


Lowry, O. H., Rosebrough, N. J., Farr, A. J., and 

Randall, R. J., J. Biol. Chem. 193, 265 (1951). 

Heymann, M. A., and Rudolph, A. M., Physiol. Rev. 


15. Ikeda, M., Rubinstein, £. H., and Sonnenscl 
R., Proc. Soc. Exp. Biol. Med. 143, 354 (197- 

16. Oberhansli- Weiss, I., Heymann, M. A., Rude 
M., and Mehnon, K. L., Pediat. Res. 6, 693 

17. Friedman, W. F., Hirschklau, M. J., Printz, 
PitUck, P. T., and Kirkpatrick, S. E., N. I 
Med. 295, 526 (1976). 

18. Heymann, M. A., Rudolph, A. M., and Silv 
N. H., N. Engl. J. Med. 295, 530 (1976). 

19. Clyman, R. I., Heymann, M. A., and Rudo 
M., Prostaglandins 13, 219 (1977). 

20. Michelson, A. M., Puget, K., Durosay, P., Be 
J. C. and Ropartz, C. in ^^Biochemical and 9 
Aspects of Active Oxygen** (O. Hayaishi ; 
Asada eds.). p. 247, University Park Press, Ba 

Received January 5, 1978. P.S.E.B.M. 1978, Vol. 


Mouse Hepatitis Virus (MHV) Infection in Thymectomized C3H Mice (40278) 


Department of Paihobiology, School of Hygiene and Public Health, The Johns Hopkins University, Baltimc 

Maryland 21205 

The macrophage plays a crucial role in the 
genetic susceptibility of mice to develop fatal 
hepatitis when infected with mouse hepatitis 
virus (MHV) (1). The aduh C3H mice do not 
die after inoculation of MHV grown in 
Princeton mice (MHV(PRI)) and their mac- 
rophages do not support MHV(PRI) multi- 
plication. On the other hand, infant C3H and 
infant and adult Princeton (PRI) mice de- 
velop a fatal infection after inoculation with 
MHV(PRI) and their macrophages support 
the multiplication of this virus and are de- 
stroyed by it (2). Intermediate susceptibility 
of macrophages in vitro is associated with 
virus persistence in vivo (3). A variety of 
treatments which depress the cell mediated 
immune response makes the MHV infection 
pathogenic for the genetically resistant 
mouse. Adult C3H mice develop fatal MHV 
infection after neonatal thymectomy (4) or 
after treatment with cortisone (S), Cytoxan (6) 
or preinfection with Eperythrozoon coccoides 
(7) and adult A strain mice are rendered 
susceptible to fatal MHV infection by a va- 
riety of treatments such as x-irradiation, ad- 
ministration of antilymphocyte serum and 
neonatal thymectomy (8, 9). 

We report here that although the outcome 
of MHV(PRI) inoculation in aduh PRI and 
C3H mice is very diflerent, both strains are 
infected by the same minimal infectious dose 
of MHV(PRI). In addition, although 
MHV(PRI) infection is fatal both for adult 
neonatally thymectomized C3H mice and the 
genetically susceptible PRI mice, the course 
of infection in these two strains is quite dis- 
similar. These fmdings suggest that the rou- 
tine recovery of adult C3H mice from 
MHV(PRI) infection requires both virus re- 
sistant macrophages and normal thymic func- 

Materials and methods. Virus. MHV-2 
strain of virus originally obtained from Dr. 
John Nelson (10) was maintained in our lab- 
oratory by intraperitoneal (ip) inoculation of 

4 week old PRI mice. This strain is ref 
to as MHV(PRI). A variant j 
MHV(C3H) which was derived 
MHV(PRI) but which is lethal for both 
C3H and aduh PRI mice (11) was maint 
by ip inoculation of 4-week old C3H 
The stock virus preparations were 10^ 
mogenates of livers from virus infected 
ibund mice. Titrations were performe 
inoculation of 0.2 ml of serial tenfold 
tions in each of three tubes of cultured 
toneal macrophages from PRI mice pre] 
as described previously (12), but maint 
in Eagle's minimum essential me 
(Earle's salts) supplemented with 20% 
calf serum (PCS). The cuhures were obs< 
for viral cytopathic effect (CPE) for 8 
The 50% tissue culture infectious 
(TCID50) was calculated by the meth< 
Reed and Muench (13). 

Mice. Three strains of inbred mice, 
C3H and C3HSS were used (12, 14, 15) 
C3HSS strain is congenic with the C3H 
but is susceptible to fatal infection 
MHV(PRI) and its macrophages su 
MHV (PRI) multipUcation. It was deve 
by introducing the PRI gene for suscepti 
to MHV(PRI) into C3H mice (15). Mice 
infected by the ip route. Thymectomy 
performed on C3H mice within 24 hr of 
Both left and right sections of the tfa 
were removed by gentle suction. Thyn 
mized mice were infected at 4-6 weeks c 
Sham operated mice served as controls 

Immunofluorescence. Anti-serum wa 
pared in vaccinated PRI mice. PRI mice 
inoculated for four successive weeks w 
propiolacetone inactivated vaccine (6). 
challenged with Uve virus and bled one 
later. The serum was stored at —20°. 
sections were cut on a cryostat at 4 mi 
Fluorescein conjugated goat anti-mous< 
concentration 1:10 (Meloy Laboratorie: 
Springfield, VA) was used with a counte 
of Evans Blue prepared as a 0.5% sto< 


0037-9727/78/ 1 59 1 -0034$0 1 .00/0 

Copyright <£> 1978 by the Society for Experimental Biology and Medicine 
^/frights reserved. 



and used at a 1:8 concentration. 
tology. Sections of liver were placed in 
lI buffered formalin, cut and stained 
lematoxylin and eosin for histopathol- 

ults, Infectivity and pathogenicity of 
(PRI) for PRI and C3H mice. MHV- 
was titrated in 4-6 week old PRI and 
nice using four mice per dilution and 
oculated mice were observed for 7 days 
Drtality. On day 7, the surviving mice of 
titrations were challenged ip with 2.6 
units of MHVCCsH) virus. Ability of a 
: to resist MHV(C3H) challenge was 
as evidence that it was previously in- 
with MHV(PRI). 

5 MHV(PRI) had a LD50 titer of 10^^ 
RI mice (Table I). None of the PRI 
'ors of the titration resisted MHV(C3H) 
nge indicating that MHV(PRI) did not 
ice a nonfatal immunizing infection in 
nice. In contrast MHV(PRI) produced 
ortality in C3H mice inoculated with 
-10" dilutions of the virus and all but 
f the survivors of this titration resisted 
inge with MHV(C3H). The immunizing 
•f tfie virus was 10*^ for C3H mice. These 
s indicate that PRI and C3H mice were 
ly susceptible to infection with 
(PRI) but that the infection was uni- 
y fatal in PRI and uniformly nonfatal 
H mice. 

urse of MHV(PRI) infection in thymec- 
ed C3H mice, Neonatally thymecto- 
I or sham operated C3H mice were in- 
ted ip with 10'° TCID50 of MHV(PRI) 
observed for mortahty, virus titers in 
md liver pathology. All thymectomized 
were checked for the completeness of 
tion at the time they died or were sac- 
1 and animals with thymus remnants 
excluded from the study, 
the 13 mice infected with MHV(PRI) 
Iso completely thymectomized, the mor- 
was 100% (Fig. 1). This was in contrast 

BLE I. Infectivity and Pathogenicity of 
(PRI) FOR 4-6 Week Old PRI and C.,H Mice. 

MHV(PRl) virus 



Lethal doscwi 




Fig. I . Cumulative mortality of PRI and C3H thy- 
mectomized mice infected with MHV(PRI) virus. Data 
on PRI mortality were pooled from several experiments 
and include over 100 mice which were inoculated with 
similar virus dilutions. Data on CsH-thymectomized 
mice are based on 13 completely thymectomized mice. 

to the 0% mortality in intact C3H mice which 
were inoculated with MHV(PRI). Deaths oc- 
curred between day 5 and 10 after inoculation 
of virus with an average survival time of 
between 7 and 8 days. This timing of mortal- 
ity was very different from that in PRI mice 
which, while they also have a mortality of 
100%, survive only 2-3 days postinfection 
(Fig. 1). 

Virus titers in livers of thymectomized and 
sham operated C3H mice are shown in Fig. 2. 
Until day 7 postinfection, the titers in the two 
groups were comparable and ranged between 
10^® and 10'^. Exceptions to this were two 
sham operated mice which had titers between 
10^^ and lO^"" on day 6. After day 7, there 
was a marked reduction in liver titers of sham 
operated mice. Of eight livers titrated be- 
tween days 8 and 12, seven were negative for 
virus and the eighth had a titer of <10*^. In 
contrast, virus titers in completely thymecto- 
mized mice continued to remain high; all of 
four livers harvested from this group between 
days 8 and 10 had titers between 10^^ and 

The pathologic lesions in sham operated 
and thymectomized C3H mice were very sim- 
ilar until day 5. By day 4 the Uvers showed 
general architectural disruption with coagu- 
lative change with diffuse and focal inflam- 
mation in which polymorphonuclear leuko- 
cytes were most prominent. Eosinophilic bod- 
ies as described by Ruebner and Miyai (16) 
could be seen in areas of necrosis. The Uver 



i • 

I * 

s * 





Fig. 2. Virus titers in livers from virus infected sham 
operated C3H and thymectomized C3H mice on PRI 
macrophage cultures. This graph includes the results of 
several experiments. O « liver from one C3H sham 
operated mouse; x >■ liver from one C3H thymectomized 
mouse, TR » Trace of virus, only one of three cultures 
inoculated with lowest dilution affected. 

sections from sham operated C3H mice taken 
on day 6 showed perivascular infiltration of 
mainly mononuclear cells with small foci of 
mononuclear cells on top of necrotic paren- 
chymal cells. The livers of C3H thymecto- 
mized mice of the same time period had large 
areas of acute fulminating lesions with tissue 
hemorrhage, necrotic debris, and the presence 
of polymorphonuclear leukocytes and mono- 
nuclear cellis. On day 8, liver sections of sham 
operated C3H mice showed a few focal areas 
of resolving lesions, while liver sections from 
C3H thymectomized mice showed foci of de- 
generation with larger foci showing centers of 
liquefaction. Sections of livers of sham oper- 
ated C3H mice from 9 through 14 days were 
normal with the exception of two mice whose 
livers showed ocassional resolving lesions. In 
summary, liver sections from sham operated 
C3H controls showed focal hepatitis with sub- 
sequent recovery, whereas liver sections form 
thymectomized C3H mice showed focal hep- 
atitis progressing to diffuse hepatitis with no 
recovery. The sham operated mice described 
above had more severe pathologic lesions and 
higher virus titers in livers than what is ordi- 
narily found in normal C3H mice infected 
with MHV(PRI). The reason for this was not 

We also compared the pattern of viral mul- 
tiplication in livers of PRI, C3H and CsHss 
mice by histopathology, immunofluorescence 
and viral titrations. Mice were infected ip 
with lO^'^ TCID50 of MHV(PRI) and were 

sacrificed at 3, 6, 10, 24, 48 and 72 hr 
infection. Liver sections from PRI 
showed increased cellular infiltration 
hr which progressed to necrosis of pare 
mal cells with eosinophilic bodies by 
and extensive tissue destruction with h< 
rhage by 72 hr. Immunofluorescence w 
tected and observed to spread as the I 
and cellular destruction grew. In C3H 
an infiltration of mononuclear cells w; 
tected as early as 6 hr postinfection, 
necrotic foci were observed by 48 hr. In 
liver sections very little fluorescence 
noted in the first 10 hr. Small fluoresce] 
of necrosis containing eosinophilic I 
and KupfTer cells were apparent by ' 
There was a striking difference in 
growth between the susceptible (PRI, C 
and resistant (C3H) mice (Fig. 3). Virus 
in PRI livers were higher than those ii 
livers by 2 logio units by 24 hr an< 
difference increased to 6 logio units by 
The C3HSS mice resembled PRI mice 
respect to both virus titers in liver and p 
of mortality. 

Growth of MHV(PRI) in macrophaf, 
turesfrom thymectomized and nonthymt 
ized C3H mice. Earlier work has show 
the pathogenic effect of MHV(PRI) c 
mouse was closely correlated with the s 
of the peritoneal macrophage of that i 
to support multiplication of the virus (! 
It was therefore of interest to see if cu 
of macrophages derived from thymecto 
C3H mice supported growth of MHV{ 

5 7 
o 6 


o ' 

^ 3 

t 2 

6 24 4« 72 


Fig. 3. Virus tilers in livers of PRI, C3H an<] 
mice in the first 3 days after infection with MH^ 
O - PRI mice. X - C3H mice. A - CsHss mice. 



•phages were removed from 4 to 6 week 
ymectomized and nonthymectomized 
Only those thymectomized mice which 
^ had no grossly visible thymus rem- 
were used as macrophage donors. In- 
n containing 10®° infective units of vi- 
s allowed to adsorb for 30 min and the 
^ were then washed and fresh medium 
. The tubes of cultures were observed 
al CPE and harvested at various time 
s after infection. They were stored fro- 

-70° until they were titrated on mac- 
jes from PRI mice. 

virus did not produce CPE in either 
r C3H thymectomized macrophage cul- 
Eind there was no diflerence between 
nacrophages in their abiUty to support 
TRI) multiplication (Fig. 4). Low titers 
IS were recovered from both kinds of 
phages through the observation period 
lys. This pattern of virus growth in C3H 
phages was markedly difTerent from 
i PRI macrophages (Fig. 4), in which 
us grows rapidly with complete lysis of 
lis in 48 hr. 

lission. Our studies confirm the pre- 
reports of Stutman and Yunis (4) and 
Provost and his colleagues (8, 9) that 
ctomy increases the pathogenicity of 
in genetically resistant mice. 
TRI) produced no mortality in the in- 
3H mouse but it infected this strain as 
f as it did the PRI mouse. Thymectomy 
sed the mortality in the C3H mouse 


M 34 41 7) 9* 120 144 1M t«9 

4. Growth of MHV(PRI) in PRI. C3H and ihy- 
ized CaH macrophage cultures. Data for PRI 
lages taken from Shif and Bang 1970 (i 1). A » 
:rophages. O - C3H macrophages, x = thymec- 
C3H macrophages. 

from to 100%. The pathologic studies as 
well as virus titers in the liver clearly indi- 
cated that the death of the thymectomized 
C3H mouse was due to its inabiUty to resolve 
the early hepatic lesions which occurred in 
both thymectomized and nonthymectomized 
animals. These lesions progressed to fulmi- 
nant hepatitis in the thymectomized C3H 
mouse resuhing in death about 6-10 days 
after inoculation of virus, but were com- 
pletely resolved in the intact C3H mouse. This 
requirement of thymic function for the recov- 
ery of C3H mice from MHV hepatitis appears 
to be similar to that described by Blanden 
(17) for the resolution of ectromelia infection 
of mice. 

Although the MHV(PRI) infection was 
uniformly fatal in thymectomized C3H mice 
as well as in PRI mice, the course of the 
disease was very different in these two strains. 
In PRI mice the virus multiplies very rapidly 
leading to death in 2-3 days whereas in the 
thymectomized C3H mice mortality occurred 
later and over a longer period. This diflerence 
very likely reflects the fact that the PRI mac- 
rophages support very well the multiplication 
of MHV(PRI) while the macrophages of thy- 
mectomized C3H mice do not. The C3HSS 
mouse resembled the PRI mouse in its sus- 
ceptibility to MHV(PRI). Differences in sur- 
vival time after MHV infection has been 
shown even among susceptible strains of mice 
(18). This diflerence in susceptibihty was re- 
lated to the varying ability of the macro- 
phages as the primary targets of the virus to 
support viral growth. These observations in- 
dicate that resistance to MHV(PRI) infection, 
as in the intact adult C3H mouse, requires 
both a resistant macrophage which limits the 
spread and multiplication of the virus and an 
intact thymic function which is necessary for 
the resolution of focal hepatic lesions. In mice 
that have susceptible macrophages, namely, 
infant C3H, infant and adult PRI, and the 
congenic C3HSS strain (15), the infection is so 
overwhelming that it kills the mouse before 
it has had a chance to develop an effective T 
cell response. In experiments not described 
here, transfer of inmiune C3H spleen cells to 
C3HSS animals with susceptible macrophages 
failed to confer resistance to challenge with 
MHV(PRI) (19). MHV(PRI) infection of 
C3H mice can also be made more pathogenic 



by treatment of these mice with cortisone (5) 
or Cytoxan (6). The mechanism by which 
these drugs bring about this effect is not clear 
but it could be by their destruction of T cells 
per se, or as suggested by Weiser and Bang 
(20), by release of lymphokines which alter 
macrophage susceptibiUty. LeBlond et al (21) 
have shown that both macrophages and T 
cells are necessary in the transfer of resistance 
to MHY to infant mice. 

Summary, MHV(PRI) virus produced a 
non-fatal immunizing inifection in adult C3H 
mice over a greater than 6.0 logio unit range 
but a uniformly fatal infection in adult 
Princeton (PRI) mice. Neonatally thymecto- 
mized 4-6 week old C3H mice died by day 10 
after inoculation with MHV(PRI). Intact and 
thymectomized C3H mice had comparable 
virus titers in their livers until day 7 postin- 
fection after which time virus was undetect- 
able in intact C3H mice but remained at high 
titers in thymectomized C3H mice. The liver 
pathology was similar in both groups until 
day 6 post infection after which time resolv- 
ing lesions were seen in livers of intact C3H 
mice whereas thymectomized C3H mice de- 
veloped ftilminant fatal hepatitis. In in vitro 
tests, the macrophages of the thymectomized 
C3H mice did not support growth of 
MHV(PRI) virus to any greater extent than 
the macrophages of nonthymectomized C3H 

Although mfection with MHV(PRI) was 
fatal for both PRI and thymectomized C3H 
mice, the course of infection was much more 
rapid in PRI mice. CsHss mice which are 
congenic with C3H mice but have macro- 
phages which support growth of MHV(PRI) 
responded to MHV(PRI) infection with a 
rapidly fatal illness in the same way as PRI 
mice. These data suggest that macrophages 
resistant to viral multiplication and intact 
thymic function are both necessary for resist- 
ance to the lethal effects of MHV(PRI) virus. 

This research was supported in part by 
Science Foundation Grant No. GB 31479. 

1. Bang, F. B. in **Microbial Pathogenicity ir 
Animals** (M. Smith and J. M. Pearce, e 
posium 22, p. 415, Cambridge Univen 

2. Bang. F. B., and Warwick. A.. Virology 9, " 

3. Virelizier, J. L., and Allison. A. C, Arch. 
279 (1976). 

4. Stutman. A., and Yunis. E. J.. Amer. J. 1 
81a (1970). 

5. Gallily. R.. Warwick. A., and Bang. F. B., 
Acad. Sci. U.S.A. 51, 1158 (1954). 

6. Willenborg. D. O.. Shah, K. V., and Ba 
Proc. Soc. Exp. Biol. Med. 142, 762 (1973 

7. Lavelle, G. C, and Bang. F. B., Arch. C 
forsch. 41, 175(1973). 

8. LeProvost, G., Levy-Leblond, E., Virelia 
and Dupuy, J. M., J. Immunol. 114, 221 ( 

9. Dupuy, J. M.. Levy-Leblond, E.. and LeP 
J. Immunol. 114,226(1975). 

10. Nelson, J. B., J. Exp. Med. %, 293 (1952) 

11. Shif, 1., and Bang, F. B.. J. Exp. Med. 

12. Bang, F. B., and Warwick, A., Proc. Nat. 
U.S.A. 46, 1065 (1960). 

13. Reed, L. T., and Muench, H., Amer. J. H^ 

14. Kantoch. M., Warwick, A., and Bang, F. 
Med. 117, 781 (1%3). 

15. Weiser, W., VelUsto, L, and Bang, F. B., 
Exp. Biol. Med. 152, 499 (1976). 

16. Ruebner, B., and Miyai, K., Amer. J. Path 

17. Blanden, R. V., J. Exp. Med. 133, 1074 (1 

18. Fumikiro, T., Hirano, N., Kiuchi, V., and 
K., Japan. J. Microbiol. 20, 293 (1976). 

19. Sheets, P., Doctor of Science Thesis, 1 
Hopkins School of Hygiene and PubUc H 
timore, Maryland, 1975. 

20. Weiser, W., and Bang, F. B., J. Exp. Mec 

21. Levy-Leblond, E., and Dupuy, J. M., J. 
118, 1219(1977). 

Received February 13, 1978. P.S.E.B.M. 1978 


hymidine Kinase and DNA Polymerase Activity in Normal and Zinc Deficient 
Developing Rat Embryos (40279) 


Department of Nutrition, University of California, Davis, California 95616 

pidly dividing cells, DNA synthesis is 
^ by zinc deficiency (1-S). It has been 
ed that the severe teratogenesis result- 
m maternal dietary zinc deficiency in 
ly arise as a consequence of impaired 
ynthesis during fetal organogenesis (2, 
lies on regenerating rat Uver (6), tumor 
7), and rat connective tissue (8) have 
ed that the decrease in DNA synthesis 
tult of zinc deficiency may be linked to 
Kdation of zinc with one or more of the 
spendent enzymes involved in DNA 
sis. The activity of two regulatory en- 
in DNA synthesis, thymidine kinase 
NA polymerase (9), was found to be 
d in zinc deficient rats (7, 8). Thymi- 
inase is of particular importance since 
^ficiency produced a significant reduc- 
activity of this enzyme in regenerating 
;r within 10 hr after partial hepatec- 
whereas DNA polymerase activity, 
synthesis, and protein synthesis were 
ected until some hours later (10). Re- 
Dreosti and Hurley (11) found the 
f of thymidine kinase to be signifi- 
lower in embryos taken at 12 days of 
Dn from zinc deficient dams than in 
rom controls. 

he present study, the efTect of zinc 
ncy on the activity of DNA polymerase 
vestigated in 12-day embryos. In ad- 
since a previous report suggested that 
:mbryos are relatively less sensitive to 
jficiency than are older embryos (12), 
tivity of both thymidine kinase and 
x)lymerase was measured in 9, 10, and 
embryos from zinc deficient and con- 
s to determine whether zinc was acting 
ilar sites at early stages of gestation, 
r studies were also undertaken to test 
ect of in vitro supplementation with 
id other divalent metal ions on the 
' of DNA polymerase. Similar data 
; to thymidine kinase have been re- 
prcviously (11). 

Materials and methods. Materials. (Methyl- 
^H) thymidine (spec. act. 2 Ci/mA/) and 
(methyl-^H) thymidine 5'-triphosphate (spec, 
act. 15 Ci/mA/) were purchased from Amer- 
sham/Searle Corporation, Arlington Heights, 
Illinois. All other chemicals were purchased 
from Sigma Chemical Company, St Louis, 
MO. Whatman DEAE ceUulose (DE 23) filter 
paper circles were obtained from Reeve An- 
gel, 9 Bridewell Place, Clifton, NJ. 

Animals and diets. Virgin female Sprague- 
Dawley rats weighing 210 ± 10 g were bred 
overnight with stock fed males. On day zero 
of gestation, as determined by the presence 
of sperm in the vaginal smear, the animals 
were placed individually in stainless steel 
cages. The animals were fed a zinc deficient 
diet ad libituniy or a control diet ad libitum, or 
a control diet in amounts limited to the mean 
daily food intake of the deficient group (re- 
ferred to as "restricted intake"). 

The zinc deficient diet contained less than 
O.S ppm zinc as measured by atomic absorp- 
tion spectroscopy. The control diet was the 
same purified diet as the zinc deficient diet 
except that it was supplemented with zinc as 
zinc carbonate to a level of ICX) Mg/g. The 
composition of the diet has been described 
previously (13). In addition, all animals re- 
ceived vitamins in glucose three times per 

Collection of samples. On day 9, 10, 11, or 
12 of gestation, the animals were killed and 
embryos were removed by caesarean section. 
In order to obtain sufficient tissue for the 
enzyme assay it was necessary to pool litters 
of embryos. Five litters were pooled for each 
9-day sample, three litters for each 10-day 
sample, two Utters for each 11 -day sample, 
one litter for each 12-day sample. 

Enzyme assays. Pooled embryos were ho- 
mogenized in 12 vol of chilled 0.25 N 
Tris-HCl buffer, pH 8.0, and an enzyme 
solution was prepared for use in the subse- 
quent assays as described by Witschi (14). 


0037-9727 11%! \59\-Q0^9ViV .0^/^ 

Copyright © 197« b^ ibe So6cily tox ^p«hn«xi\A V\oVs(« MA>Msfts«M» 
All righu Teserved. 



Thymidine kinase was assayed by a modi- 
fied procedure described by Witschi (14). The 
reaction mixture contained in a final volume 
of 0.5 ml, 0.25 N Tris-HCl buffer (pH 8.0), 
5.5 jiiA/ ATP, 6.6 jiiA/ 3-phosphoglyceric acid, 
5.5 jtiA/ MgCl2 and 2.5 jtiA/ (5.0 jiiCi) (methyl- 
■^H) thymidine and 0.1 ml of the enzyme 
extract. The reaction mixture was incubated 
at 37® for 15 min and the reaction was 
stopped by immersing the assay tubes in boil- 
ing water for 1 min. After cooling and cen- 
tr^ugation at \000g for 10 min, 50 jul aliquots 
of the protein-free supematants were spotted 
onto DEAE cellulose fiher paper discs and 
the papers were washed in 1.0 mA/ ammo- 
nium formate, water, and 95% ethanol. Ra- 
dioactivity on the dried paper discs was mea- 
sured in a Nuclear Chicago Mark I liquid 
scintillation spectrophotometer. 

DNA polymerase was determined by a 
modified procedure described by Witschi (14) 
and Lehman et al (15). The reaction mixture 
contained in a final volume of 0.5 ml, 0.25 N 
Tris-HCl buffer (pH 8.0), 0.05 /lA/ d-ATP, 
0.05 yM d-CTP, 0.05 jtiA/ d-GTP, 1.5 iiM 
MgCl2, 1.5 jLtAf KCl, 0.05 juAf 2-mercaptoeth- 
anol, 50 /ig heat denatured DNA (70° for 15 
min), 0.05 juA/ (5 jnCi) dTTP and 0.1 ml of 
the enzyme extract. After incubation at 37° 
for 1 hr, the reaction was stopped by the 
addition of 0.1 ml cold 1.0 M HCIO4. The 
precipitate was washed twice with 0.5 M 
HCIO4, dissolved in 0.3 M KOH (3 ml), 
incubated for 60 min at 37°, reprecipitated 
with cold 0.5 M HCIO4, and washed once 
more. The pellet was dissolved in 1 M NaOH 

(2 ml) and 0.5 ml aliquots were withdrawn 
for radioactivity determinations. 

Metal ion supplementation. In certain DNA 
polymerase assays, supplementary zinc and 
other metal ions (0.01-0.2 mM) were added 
to the incubation mixture before addition of 
^H-dTTP. All metal salts used were spectro- 
photometrically pure and, except for the zinc 
sah, contained less than 0.05 jug zinc/g. 

Protein assay. The concentration of protein 
in the fetal homogenates was determined by 
the method of Lowry et al (16). 

Statistical analysis. Mean ± SEM are re- 
ported. The statistical significance of differ- 
ences between means was tested by Student's 
"t" test. 

Results. The activity of thymidine kinase 
was significantly lower in 9, 10, 11, and 12 
days embryos taken from females fed a zinc 
deficient diet than in embryos from either ad 
libitum fed (P < 0.05) or restricted intake (P 
< 0.05) controls (Table I). However, the per- 
centage decrease in activity in the zinc defi- 
cient animals when compared with restricted 
intake controls was not as great in early em- 
bryos as in the 12-day embryos. In addition, 
the activity of thymidine kinase increased 
with increasing age of embryos in all three 
dietary groups (Table I). The percentage in- 
crease in activity was greatest at early stages 
of gestation. Activity in the 9-day groups was 
only twice that of background values. 

DNA polymerase activity was also signifi- 
cantly lower in 9, 10, 11, and 12 day embryos 
from dams fed the zinc deficient diet than in 
embryos from either the ad libitum fed (P < 

TABLE I. Effect of Zinc Deficiency and Day of Gestation on Activity of Thymidine Kinase in Rat 



Day of 

Control ad libitum 

Control restricted intake 

Zinc deficient 


in activ- 
ity (%) 


in activ- 
ity (%) 


in activ- 
ity (%) 

" Thymidine kinase activity expressed as pA/ ^H-thymidine incorporated/mg protein/hr. 
^ P < 0.05 compared to ad libitum and restricted intake controls. 
♦♦ P < 0.05 compared to activity in I -day older embryos in the same group. 
••• Data from Dreosti, I. E.. and Hurley, L. S., Proc. Soc. Exp. Biol. Med. 150, 161 (1975). 




79 ± 9»* 

81 ± I4*» 




323 ± 66»* 


351 ±52*' 






665 ±95 


659 ± lor* 


372 ± 76* 




729 ± 101 


950 ± 48 


356 ± 19* 





0.01) or restricted intake (P < 0.01) controls 

(Table II). The percentage decrease in the 

zinc deficient groups when compared with 

rcstriaed intake controls was similar at all 4 

days of gestation. DNA polymerase activity 

also increased with increasing age of embryos 

in all three dietary groups, but the percentage 

daily increase was not as great as that found 

with thymidine kinase (Table II). Even at 9 

days of gestation, embryos had appreciable 

levels of DNA polymerase activity. 

Addition of zinc, as zinc chloride, to the 
assay medium (at levels between 0.01 mA/ 
and COS mA/) had Uttle efTect on the activity 
of DNA polymerase in extracts from zinc 
deficient and control embryos at 12 days of 
gestation (Table III). However, supplemen- 
ution of these extracts with a higher level, 0.2 
mM zinc, resulted in a statistically significant 
depression of activity in extracts from both 
zinc deficient and control embryos (19% and 
21%, respectively). 

In a further experiment (Table IV), addi- 
tion of Cu'-', Cd*-', Mn"-', Mg"-', Co'^, and 
Fe^* had no efTect on the activity of DNA 
polymerase when added to the medium at 
concentrations of 0.01 mM or 0.2 mA/. 

Discussion, The low activity of thymidine 
kinase and DNA polymerase in embryos 
from zinc deficient dams confirms previous 
reports of reduced activity of these enzymes 
in zinc deficient mammaUan tissues (6-8). It 
further suggests that impaired DNA synthesis 
and teratogenesis associated with zinc defi- 
ciency may be related to reduced activity of 
these enzymes during organogenesis. 

The thymidine kinase salvage pathway is 

important for DNA synthesis only in rapidly 
dividing cells, not in normal adult cells where 
the de novo pathway of DNA synthesis is 
predominant (9, 17). Therefore, the thymi- 
dine kinase pathway may be of critical im- 
portance in the developing embryo. Since the 
effect of zinc deficiency on cell division is 
most manifest in rapidly proUferating tissues, 
it is reasonable to suppose that thymidine 
kinase may be involved. In contrast, while 
the activity of certain DNA polymerase en- 
zymes is enhanced in rapidly dividing cells, 
these enzymes, unlike thymidine kinase, are 
also important in DNA synthesis in normal 
resting cells (18, 19). 

Further support for the idea that thymidine 
kinase, and possibly DNA polymerase, are 
possibly primary sites of action of zinc in 
embryonic tissue is provided by the finding 
of decreased activity of both enzymes with 
decreasing age of embryos. Hurley et ah (12) 
have found a low incidence of congenital 

TABLE III. Effect of Supplementary Zinc on the 

Activity of DNA PoLY^fERASE in 12-Day Rat 



Percent of original DNA polymerase 

Zinc supple- 
mented control 

Zinc deficient 


100 (±3.5) 
106 (±2.7) 
114 (±6.3) 
79 (±3.7)^ 

100 (±3.2) 
113 (±4.5) 
121 (±7.3) 
81 (±3.1)* 

" DNA polymerase activity expressed as nM ^H-TTP 
incorporated/mg protein/hr. 

* P < 0.05 compared to extracts with no zinc added. 

TABLE II. Effect on Zinc Deficiency and Day of Gestation on Activity of DNA Polymerase in Rat 



Control ad libitum 

Control restricted intake 

Zinc deficient 

Day of 


in activ- 
ity (%) 


in activ- 
ity (%) 


in activ- 
ity (%) 

** DNA polymerase activity expressed as nA/ ^H-TTP incorporated/mg protcin/hr. 
* /* < 0.01 compared to ad libitum and restricted intake controls. 
** P< 0.05 compared to activity in 1-day older embryos in the same group. 









2.68 ± 0.25^» 


2.51 ±0.22 






2.% ± 0.28 


2.62 ± 0.42 






3.06 ± 0.35 


2.59 ± 0.38 

1.97 ±0.26^ 





TABLE IV. Effect of Supplementary Metal Ions 

ON THE Activity of DNA Polymerase in 12-Day 

Embryos from Zinc Deficient Dams." 

Percent of original 

Metal ion 


DNA polymerase ac- 


tion (mM) 




100 (±3.5) 



92 (±3.6) 


85 (±6.8) 



95 (±4.9) 


90 (±2.8) 



98 (±3.7) 


98 (±5.7) 



100 (±8.0) 


99 (±4.7) 



100 (±5.8) 


96 (±3.9) 



97 (±9.1) 


93 (±4.0) 

" DNA polymerase activity expressed as nA/ ^H-TTP 
incoqx>rated/mg protein/hr. 

abnormalities in rats fed a zinc deflcient diet 
from days to 8 of pregnancy. The incidence 
of malformations increased when the animals 
were fed a zinc deficient diet for longer pe- 
riods during gestation or for the same length 
of time but at a later stage of gestation. The 
very low activity of thymidine kinase in rat 
embryos at 9 days of gestation, together with 
the relatively smaller decrease in enzyme ac- 
tivity in 9-day embryos than in 12-day em- 
bryos from zinc deflcient animals, may there- 
fore make the early embryo relatively less 
sensitive to zinc deflciency than are later 
embryonic stages. 

The failure of zinc added at the time of 
assay to restore the activity of DNA polym- 
erase in zinc deflcient enzyme extracts con- 
flrms earlier observations with extracts from 
regenerating rat hver (10) and suggests that 
zinc may not be associated with the enzyme 
as a readily dissociable cofactor. This flnding 
is similar to observations with thymidine ki- 
nase and may be explained by a lack of 
incorporation of zinc into the enzyme at the 
time of synthesis. The possibility of reduced 
synthesis of the enzyme as a result of general 
reduced protein synthesis in the zinc deflcient 
animals is unlikely since it has been shown in 
regenerating rat liver that protein synthesis 
was not affected by zinc deficiency until 
10-20 hr after a change in DNA polymerase 
activity was noted (10). 

The inhibitory effect of a high level of zinc 
(0.2 mA/) on DNA polymerase activity in 

vitro was similar to that reported by Dreosti 
and Hurley (11) for thymidine kinase. Such 
inhibition of activity of these two enzymes 
may account for the reduced DNA synthesis 
produced by high levels of zinc in cultured 
rat lymphocytes (20) and transplanted rat 
tumors (21). 

Unlike thymidine kinase, which was rela- 
tively sensitive to Cd^"*" and Cu^"*", addition of 
various metal ions at both low (0.01 mA/) and 
high (0.2 mA/) concentrations had little effect 
on the activity of DNA polymerase in vitro. 
This observation supports the data of Spring- 
gate et al (22) using a zinc free apoenzyme 
and suggests that DNA polymerase is specif- 
ically zinc dependent. 

In conclusion, the flndings reported here 
indicate that the teratogenic effects of zinc 
deflciency in rats may arise from impaired 
activity of fetal thymidine kinase and DNA 
polymerase after day 8 of gestation and that 
the primary effect may be on the regulatoiy 
enzyme, thymidine kinase. The in vitro addi- 
tion of metal ions to zinc deflcient enzyme 
extracts suggests that zinc may not be asso- 
ciated with DNA polymerase as a readily 
dissociable cofactor and that DNA polymer- 
ase is speciflcally zinc dependent. 

Summary, Thymidine kinase and DNA po- 
lymerase activities were signiflcantly {P < 
0.05 and P < 0.01, respectively) lower in 9, 
10, 11, and 12-day embryos taken from dams 
fed a zinc deflcient diet than in those from ad 
libitum fed and restricted intake controls. An 
additional flnding was that of increased activ- 
ity of both thymidine kinase and DNA po- 
lymerase with increasing age of embryos. As 
previously found with thymidine kinase, ad- 
dition of zinc and other divalent metal ions 
in vitro had little effect on restoration of DNA 
polymerase activity from zinc deflcient ex- 
tracts when added at concentrations of 0.0 i 
and 0.05 mA/. When added at a level of 0.2 
mA/, zinc, but not other metal ions, had aO 
inhibitory effect on DNA polymerase activ- 
ity. These findings support the hypothesis 
that the teratogenic effects of zinc deficiency 
are associated with the enzymes involved iX^ 
DNA synthesis. 

This research was supported in part by NIH Rcscar^^ 
Grant No. HD-01743 from the National Institute ^^ 
Child Health and Human Development. 



. M., and Underwood, E. J., Aust. J. Biol. Sci. 


ion, H., Shradcr, R., and Hurley, L. S., Sci- 

6, 1014(1969). 

, I. E., Grey, P. C, and Wilkins, P. J.. S. Afr. 

46, 1585(1972). 

1, J. R., and Dreosti, I. E., S. Afr. Med. J. 48, 

ck, K. H., Fawccu, D. W., and VaUec, B. L., 
Sci. 17,57(1975). 
1. J. R., and Dreosti, I. E., J. Comp. Pathol. 


1, J. R., and Dreosti, I. E.. S. Afr. Med. J. 50, 

A. S., and Oberleas, O., J. Lab. Clin. Med. 


on, J. N., The Biochemistry of the Nucleic 
7th ed., pp. 607, Methuer and Co. Ltd., Lond. 

1, J. R., and Dreosti, I. E., Proc. S. Afr. 

m.Soc. 1,52(1975). 

, I. E., and Hurley, L. S., Proc. Soc. Exp. Biol. 

50, 161 (1975). 

, L. S., Gowan, J., and Swenerton, H., Tera- 

tology. 4, 199 (1971). 

13. Mutch, P. B., and Hurley, L. S., J. Nutr. 104, 828 

14. Witschi, H. P., Biochem. J. 120, 623 (1976). 

15. Lehman, I. R., Bressman, M. H., Simms, E. S., and 
Konnbcrg, A., J. Biol. Chem. 233, 163 (1958). 

16. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and 
RandaU, R. J., J. Biol. Chem. 193, 265 (1951). 

17. Bresnick, E., Thompson, U. B., Morris, H. P., and 
Liebelt, A. G., Biochem. Biophys. Res. Commun. 

18. Chang, L. M. S., and Bollum, F. J., J. Biol. Chem. 

19. Basil, E. F., Brown, D. E., Jenkins, M. D., and 
Laszlo, J., Biochem. 10, 1981 (1971). 

20. Duncan, J. R., and Dreosti, I. E., Agrochemophysica 
7, 1 (1975). 

21. Duncan, J. R., and Dreosti, I.E., J. Nat. Cancer Inst. 
55, 195 (1975). 

22. Springgate, C. F., Mildvan, A. J., Abramson, R., 
Engle, J. L., and Loeb. L. A., J. Biol. Chem. 248, 
5987 (1973). 

Received January 31, 1978. P.S.E.B.M. 1978, Vol. 159. 


L-Histidine-lnduced Hypercholesterolemia: Characteristics of Cholesterol Biosyntt* 

in Rat Livers^ (40280) 


Biomedical Research Unit, Waisman Center, and the Department of Nutritional Sciences, University of IVisci 

Madison, Wisconsin 53706 

Dietary enrichment with high levels of sin- 
gle amino acids induced decreased food in- 
take and growth suppression in young ani- 
mals (1). The effects depend upon the kind 
and concentration of amino acid supple- 
mented. Waisman and his colleagues have 
described in Rhesus monkeys a marked hy- 
perlipemia associated with the dietary admin- 
istration of L-histidine (2, 3). Histidine was 
the only amino acid of nine studied which 
induced this hyperlipemia. The hyperlipemia 
involved all circulating lipids. Serum phos- 
pholipids increased twofold, cholesterol two- 
to threefold and triglycerides three to eight- 
fold. Geison and Waisman fed 5% and 8% 
excess L-histidine diets to rabbits for 4 weeks 
and found a 30% increase in the plasma 
cholesterol level (4). Phospholipid levels did 
not change. The effect in rabbits was less 
pronounced than that observed in monkeys. 

Our report presents the effect of dietary l- 
histidine supplementation in rats. We observe 
alterations in the incorporation of [2-^^C] ac- 
etate or [1-^^C] octanoate into lipids, studied 
in liver slices taken from rats fed a diet sup- 
plemented 5% with L-histidine. 

Materials and methods. The basic diet fed 
in all experiments was Purina Formulab 
Chow containing 23% protein, 6.5% fat, 0.58% 
histidine, carbohydrate, vitamins and min- 
erals. L-histidine (free base) was purchased 
from Ajinomoto, Co., Tokyo. [2-^^C] acetate 
(specific activity 53.3 mCi/mmole), [1-^^C] 
octanoate (specific activity 3.5 mCi/mmole), 
Aquasol (scintillation solution) and Protosol 
were purchased from New England Nuclear 

' This investigation was supported by a grant, 5-TOl- 
HD-00131-08, from the National Institutes of Health, 
United States Public Health Service. A* preliminary re- 
port of this work was presented at the 61st Annual 
Meeting of the Federation of American Societies of 
Experimental Biology, Chicago, Illinois, April 1-8, 1977. 

^ Present Address: Sigma Chemical Co.. P. O. Box 
14508, St. Louis, MO 63178. 

Corp., Elmhurst, IL. DNA standard 
Salmon testes) was purchased from ! 
Chemical Co., St. Louis, MO. Diphen 
nine reagent was obtained from Allied C 
ical. Palatine, IL. It was purified by r 
talizing from boiling hexane to obtain a 
crystalline product. Bovine serum all 
(Nutritional Biochemicals, Cleveland, 
was used as protein standard. For hoi 
nization, a motor-driven Potter-Elvehje 
mogenizer or a Polytron (Brinkman I 
ments) was used. Liver slices were made 
a Mcllwain tissue chopper (The Micklc 
oratory Engineering Co., England). In 
tion of liver slices was performed in a 
noff Metabolic Shaking Incubator. All 
activity measurements were obtained ^ 
Nuclear Chicago scintillation countei 

Male albino rats from Holtzman Ra 
Madison, WI, were obtained at 21 d\ 
age and weighed 55 ± 5 g. They were h 
in individual wire-bottom cages w 
light-dark cycle changing every three 
throughout all experiments. Rats in th< 
trol group were fed ground Purina Fon 
Chow ad lib. In the histidine-treated { 
L-histidine constituted 5% of the di 
weight. It was added to the ground cho 
and fed to the rats ad lib. Since hist 
treated rats eat less than untreated co: 
a second control group (pair-fed co] 
was used. This group was fed the 
amount of food eaten by the histidine-t 
group. After 4 days of feeding, the ratj 
killed by decapitation. Livers were is 
and either homogenized with 9 vol of di 
water for DNA and protein determin 
or sUced for the in vitro experiments. 

Liver protein was estimated by the m 
of Lowry et al. (5) using bovine serum 
min as standard. DNA estimation was 
using the method of Schneider (6) \ 
DNA standard from Salmon testes. 


Copyright e 1978 by the Society for Experimental Biology tnd Medicine 
AJJ rights reaerved. 



bation of liver slices with [2-^^C] ace- 
[1-"C] octanoate was done under op- 
conditions suggested by Dietschy and 
rry (7). CO2 released was trapped by 
9/ in a cup hanging above the incuba- 
nixture and the radioactivity was 
d by dropping the cup into a scintilla- 
d containing AquasoL Liver slices were 
i with 0.9% NaCl three times and ho- 
ized in distilled water using the Poly- 
rhe homogenate was extracted with 
form-methanol in a 2:1 ratio. The ex- 
ras dried and dissolved in a small vol- 
f chloroform, then spotted on a silica 
plate (0.25 mm thick). The plate was 
ped in a solvent system which con- 
heptane-cther-acetic acid in a 75:25:5 
Lipid fractions were visualized by 
ag the plate with 0. 1% 2',7'-dichloroflu- 
in in methanol. Each band was assayed 
iioactivity and the rate of synthesis is 
sed as nanomoles of the labeled pre- 
incorporated into the product per gram 
r per hour. The percent deviation from 
1 in each fraction was calculated. All 
cal analysis was done by using the 
led Student's t test. 

ilts. [2-^^C] Acetate and [1-^^C] octa- 
were found to incorporate into every 
•n of liver Upids and the released CO2 
ous rates. L-histidine primarily affected 
corporation of the labeled substrates 
bolesterol and triglycerides. When [2- 
cetate was used as substrate, the in- 

crease in its incorporation compared to con- 
trols was found to be 107% for unesterified 
cholesterol and 100% for cholesterol esters 
(Fig. 1-A). Both increases were statistically 
significant, P < 0.02 and P < 0.01 respec- 
tively. The opposite effect was observed in 
the case of triglycerides, which showed a 
significant (P < 0.05) decrease of 36%. 

When [2-^^C] octanoate was used as sub- 
strate, there were significant (P < 0.02) in- 
creases in the incorporation of the labeled 
substrate of 90% and 71% for unesterified 
cholesterol and cholesterol esters, respec- 
tively, compared to controls (Fig. 1-B). The 
incorporation into triglycerides was signifi- 
cantly (P < 0.01) decreased by 39%. 

Histidine did not significantly alter the in- 
corporation of the labeled substrates into 
other liver Upids such as phospholipids, free 
fatty acids, monoglycerides and diglycerides. 
Histidine did not alter the activity of the 
tricarboxylic acid cycle, as indicated by the 
insignificant change of the incorporation of 
the labeled substrates into released CO2. 

The effects of excess dietary L-histidine on 
cell size as estimated by Uver DNA and pro- 
tein contents are shown in Table L The DNA 
to protein ratio in histidine-treated rat livers 
was significantly lower than the ratio ob- 
served in both ad lib, {P< 0.05) and pair-fed 
{P < 0.01) controls. The decrease in the ratio 
was 19.4% compared to pair-fed controls. 
Liver DNA and protein contents in histidine- 
treated rats were significantly {P < 0.01) 

I. Percent change from control of the incorporation of ['*C) acetate (A) and ['*C) octanoate (B) into liver 
ctions and CO2 by liver due to L-histidine supplementation. The conditions of the incubation and separation 
fractions arc described in the methods. Asterisks indicate P values for comparison with ad lib, controls: •/* 
••/>< 0.02 and •••/»< 0.01. 




Effect of 5% Dietary l-Histidine Supplementation on DNA and Protein Contents 



DNA (mg/100 g liver) 

Protein (g/lOOg liver) 

DNA ^ 

AD LIB: 95% Chovf-^ 

5% L-HisUdine (5) 
AD LIB: Chow (5) 
Pair-Fed: Chow (5) 

147.7 ± 11.9* 

166.0 ± 3.3 
224.5 ± 3.0^ 

18.3 ± 0.5* 

17.1 ±0.6 
22.6 ± 0.5' 

8.0 ± 

9.8 ± 
10.0 ± 

" Results are expressed as mean ± S.E.M. 
* Significantly different from pair-fed controls, P < 0.01. 
*" Significantly different from ad lib. controls, P < 0.001. 
** Significantly different from ad lib. controls, P < 0.05. 
' (N) — number of rats per group. 

lower than levels in the pair-fed controls. 

Discussion, L-histidine induces in young 
rats a hypercholesterolemia which occurs 
after a brief period of feeding (4 days). His- 
tidine-treated rats are smaller than controls, 
have larger livers and 30-40% higher levels 
of plasma cholesterol (8). In the present study, 
the incorporation of [2-^*C]acetate into cho- 
lesterol by Uver was found to increase by 
100% with the feeding of an L-histidine en- 
riched diet. However, Dietschy and McGarry 
(7) have shown that the acetyl-CoA available 
for cholesterol synthesis in the cytosol is not 
in isotopic equilibrium with the intramito- 
chondrial pool. In order to verify the result, 
[1-^^C] octanoate was used as substrate under 
the same conditions. Octanoate is incorpo- 
rated into cholesterol by the cytosolic biosyn- 
thesis pathway only after its intramitochon- 
drial oxidation to acetyl-CoA. In this way the 
C2 units entering the cholesterol biosynthetic 
pathway were in isotopic equilibrium with 
the intramitochondrial C2 pool. Increases in 
the incorporation of the [1- C] octanoate into 
unesterified cholesterol and cholesterol esters 
by 90% and 71%, respectively, were observed. 

The second significant effect of histidine 
on liver in this study was the 36-39% decrease 
in the incorporation of the labeled substrates 
into triglycerides. Kerr et al (3) showed that 
histidine-induced hyperlipemia in monkeys 
was easily detected by the appearance of a 
"creamy" serum reflecting the predominant 
presence of triglyceride-laden chylomicrons. 
The decrease in triglyceride synthesis ob- 
served in our experiments was in accord with 
the absence of "creamy" serum in the rat. 

Dietschy and McGarry have shown the 
concentrations of the labeled substrates used 
in these experiments (4 mAf for acetate and 

1.1 mAf for octanoate) to be saturai 
the metabolic process under study (7 
also have reported that octanoate was 
efficient precursor than acetate for ste 
thesis. This contrasts with the data o 
in the present study where we obser> 
in Uver the rate of cholesterol synthes 
[2-^^C] acetate was not significantly d 
from that obtained with [1-^^C] octan 

Changes in the liver DNA to prote 
support an increase in hepatic cell sizi 
correlates well with previous results s 
a 100% increase in liver glycogen com 
In sunmiary, the present study demo 
that dietary enrichment with L-histi( 
duces specific effects in cholesterol \ 
glyceride synthesis in weanling rats 
effects might represent the regulation 
cific enzymes in cholesterol biosynth< 
lipogenesis such as )8-hydroxy-4-mel 
taryl coenzyme A reductase and fat 

Summary, A diet supplemented 5% 
histidine caused a 100% increase in th 
poration of [2-^^C] acetate or [1-^^C] 
ate into cholesterol in liver slices of w 
rats after four days of feeding. The ii 
ration of the labeled substrates into 
erides decreased 38%. The hepatic D 
protein ratio decreased 19% with h 
feeding, suggesting an increase in hep 

We wish to thank Dr. Burr Eichelman for hi 
the preparation of this manuscript. 

1. Daniel, R. G., and Waisman, H. A., Growtl 

2. Kerr, G. R., Wolf, R. C, and Waisman, H. 
Soc. Exp. Biol. Med. 119, 561 (1965). 

3. Kerr, G. R., Wolf, R. C, and Waisman. 


••Symposia of the Zoological Society of London" (R. (Colowick, S. P. and Kaplan, N. O., cds.), V( 

N. T-W Ficnncs. cd.)» No. 17, p. 371, Academic p. 680, Academic Press, New York (1957). 

Press, London/New York (1966). 7. Dictschy. J. M., and McGarry, J. D., J. Biol. ( 

4. Gcison, R. L., and Waisman, H. A., Proc. Soc. Exp. 249, 52 (1974). 

BioL Med. 133, 234 (1970). 8. Solomon, J. K., and Geison, R. L., Fed. Pn 

5 Lowry, O. H., Rosebrough, N. J., Farr, A. L., and 1157 (Abstr.) (1977). 

RandaU, R. J., J. Biol. Chem. 193, 265 ( 195 1). 

6. Schneider, W. C, in "Methods in Enzymology" Received May 3, 1978. P.S.E.B.M. 1978. Vol. 159 


Effect of Cholera Toxin on Renal Tubular Reabsorption of Glucose and Bicarbo 



Division of Nephrology and the Department of Medicine, University of Southern California School of Med 

Los Angeles, California 90033 

Cholera toxin (CT) produces fluid and 
electrolyte secretion in the small intestine due 
to stimulation of adenylate cyclase and in- 
creased production of adenosine 3'S' cyclic 
monosphosphate (cAMP) (1-6). Other stud- 
ies have shown that the adenylate 
cyclase-cyclic AMP system is stimulated by 
CT in a variety of tissues such as liver, thy- 
roid, adrenal, fat and leukocytes with no dem- 
onstration of any other major structural or 
enzymatic changes (7-13). Thus, CT may 
provide a pharmacologic tool for the study of 
the effects of stimulating adenylate 
cyclase-cyclic AMP systems (10, 11). 

We have previously shown that the infu- 
sion of cholera toxin into one renal artery of 
dogs is followed by decreased net tubular 
reabsorption of sodium, potassium, calcium, 
magnesium and phosphate with interrelation- 
ships similar to those observed during expan- 
sion of the extracellular fluid volume with 
saline (14). Further studies from our labora- 
tory have demonstrated that expansion of the 
extracellular fluid with a Ringer bicarbonate 
solution is accompanied by increased net pro- 
duction of cyclic AMP by the kidney sug- 
gesting a role for cyclic AMP in the reabsorp- 
tion of these various ions (IS). 

Since extracellular fluid volume expansion 
is accompanied by decreased tubular reab- 
sorption of glucose (16, 17) and bicarbonate 
(18-20) due to suppression in their reabsorp- 
tion, which occurs mostly in the proximal 
tubule (20-25), this study was designed to 
evaluate whether the stimulation of renal ade- 
nylate cyclase with CT also affects the reab- 

' This work was supported by a grant of the United 
States Public Health Service AM 19181, and an Investi- 
gative Group Award GR from the American Heart 
Association, Greater Los Angeles Affiliate. 

^ Dr. Tuma is a Fellow of the American Heart Asso- 
ciation, Greater Los Angeles Affiliate. 

sorption of these two substances in an 
to further document a relationship b 
renal cAMP and tubular reabsorptivc 

Material and methods. Twelve exper 
were carried out in female mongre 
weighing from 18 to 27 kg, anesthetize 
pentobarbital (30 mg/kg). The dog: 
ventilated through a cuffed endotrache 
with a Harvard Respirator. Both uretei 
cannulated through bilateral flank in 
and a curved 23 gauge needle was pU 
the left renal artery in the direction ol 
flow. Isotonic saline was infused in 
renal artery at a rate of 1 ml per min th 
out the studies. A catheter was placed 
aorta through a femoral artery to 
blood samples and to measure arteria 
sure with an aneroid manometer. All • 
ments were started at least 60 min afte 
pletion of surgery. Glomerular filtrati< 
(GFR) was measured using the clean 
exogenous creatinine with standard p 
dose and constant infusion technique, 
collections of 10 min duration were ol 
throughout the studies with blood ol 
at the midpoint of each period. Aft 
control periods purified cholera 
(Schwarz-Mann, Orangeburg, NY 
added to the renal arterial infusion to 
8 jLtg/min for 180 min. 

The effect of CT on glucose reabs< 
was evaluated in five dogs. An intra 
solution containing glucose (10-15%), j 
(23 mEq/1), potassium (10 mEq/1) an 
ride (33 mEq/1) was given at a rat 
ml/min in order to attain a stable hig 
of blood glucose at the time of the m 
effect of CT on tubular transport of < 
Ivtes which usually occurs 100-140 mi 
the administration of CT (14). Both 
and urine samples were collected in i 


0037-9727/78/ 1 59 1 -0048$0 1 .00/0 

Copyright (D 1978 by the Society for ExperimenUl Biology and Medicine 
AJJr^ts reserved. 



test tubes and triplicate determinations of 
glucose were performed. 

The effect of CT on bicarbonate reabsorp- 
tion was studied in seven dogs. In order to 
raise the blood bicarbonate to a stable level 
of 33-37 mEq/1, the animals received pulse 
injections of bicarbonate 50-90 mEq at the 
beginning of the study and every 40 min 
thereafter and a constant infusion of a solu- 
tion containing bicarbonate (240 m£q/l), so- 
dium (263 mEq/1), potassium (10 mEq/1), 
and chloride (33 mEq/1) at a rate of 4 
ml/min. The rate of respiration was adjusted 
by the Harvard Respirator to keep PCO2 
suble around 40 mm Hg. Urine was collected 
anaerobically under mineral oil from the ure- 
teral catheters and blood samples were ob- 
tained anaerobically in syringes containing 

These protocols allowed us to compare tu- 
bular reabsorption of glucose (TRG) and bi- 
carbonate (TRHCO3) by both kidneys when 
all variables other than the infusion of CT 
into one renal artery were equal. 

The concentration of creatinine in the 
blood and urine samples were determined 
with Technicon autoanalyzer (Tarrytown, 
NY), sodium and potassium with Instrumen- 
tation Laboratory flame photomoter (Lexing- 
ton, MA), chloride with CMT 10 chloridom- 
eter (Radiometer, Copenhagen), glucose with 
Beckman glucose analyzer (Beckman Instru- 
ments Incorporated, Palo Alto, California) 
which utilizes glucose oxidase (26), and pH 
and PCO2 with a Radiometer acid base ana- 
lyzer. Model BMS 3-PHM71 (Radiometer, 
Copenhagen). The concentration of bicar- 
bonate in plasma and urine were calculated 
from the Henderson-Hasselbach equation 
utilizing the following factors: Solubility coef- 
ficient for CO2 in plasma and urine of 0.0301 
and 0.0309, respectively; a pk of 6.10 for 
plasma and a pk for urine calculated from its 
ionic strength according t o the formula, pKa 
« 633 - 0.5 VNa"*" + K"*" with the concentra- 
tions of Na and K given in equivalents per 
liter (27). Paired data analysis was used to 
evaluate the statistical signiflcance of the re- 
sults which are expressed as mean ± SEM. 
Resiilis. Effect of CT on glucose reabsorp- 
tion (TRG). The effect of the infusion of CT 
on GFR, fractional excretion of sodium 
(FEn.) and glucose reabsorption are given in 

Table I and Fig. 1. There were no significant 
differences among these parameters between 
both kidneys prior to the infusion of glucose 
and CT. Renal TRG after 100-140 min of 
CT was 80.1 ± 20.2 mg/min, a value signifi- 
cantly (P< .05) lower than that observed for 
the contralateral kidney (98.7 ± 20.7 
mg/min). Renal TRG per 100 ml GFR was 
254 ± 32.7 mg, a value significantly (P< .01) 
lower than that observed in the opposite kid- 
ney (363 ± 43.5 mg per 100 ml GFR). The 
FEns increased significantly from both kid- 
neys but it was markedly higher (P < .01) 
from the kidney receiving CT (11.2 ± 
2.82%) than the contralateral kidney (4.62 ± 

The values for TRG per 100 ml GFR in 
all measurements made from both kidneys 
during the period of 100-140 min after the 
initiation of the infusion of CT and when 
filtered glucose ranged between 700-1900 mg 
per 100 ml GFR are shown in Fig. 1. For any 
given level of filtered glucose, TRG per 100 
ml GFR was lower in the kidney infused with 

Effect of CT on bicarbonate reabsorption. 
The effects of CT infusion on GFR, FEns, 
TRHCO3/GFR and the urinary excretion of 
sodium, chloride and bicarbonate are given 
in Table II and Figs. 2 and 3. Again, there 
were no significant differences between these 
parameters prior to the infusion of bicarbon- 
ate and CT. Renal TRHCO3 after 100-140 
min of CT was not different between both 
kidneys while TRHCO3/GFR x 100 by the 
infused kidney was 2.09 ± .06 mEq per 100 
ml GFR, a value significantly lower (P < .01) 
than that observed in the contralateral kidney 
(2.53 ± .06 mEq per 100 ml GFR). Figure 2 
provides data on TRHCO3/GFR for all 
measurements obtained during the maximal 
effect of CT and a filtered bicarbonate rang- 
ing between 2.8 to 4.1 mEq per 100 ml GFR. 
Again, TRHCO3/GFR x 100 for any given 
level of filtered carbonate was lower under 
the effect of CT. 

The FEns increased in both kidneys but 
was significantly higher (P < .01) in the 
kidney receiving CT (15.9 ± 0.74%) than that 
of the contralateral kidney (7.1 ± .26%). The 
increments in urinary sodium in the CT kid- 
ney were due to both NaCl diuresis (40%) 
and NaHCOs excretion (60%) while the ex- 



TABLE I. Effects of Cholera Toxin on Renal Tubular Reabsorption of Glucose/ 



< 100 



X 100 
















1. Control 










CT-f glucose 










2. Control 










CT + glucose 










3. Control 










CT-t- glucose 










4. Control 










CT + glucose 










S. Control 










CT-t- glucose 










Control, mean 




















CT + glucose, mean 


























CT-f glucose 





Control vs CT + glucose 






" Each point represenu the mean or three to five consecutive collections. The results obtained during cholera toxin (CT) and glucose infusion represent 
the mean of three to five consecutive 10 min collections during the maximum response to CT and stable high plasma glucose. Ccr - clearance of exogenous 
creatinine. ChJCc. x 100 - fraction of filtered sodium excreted. PG - plasnu glucose. TRG/Ct > x 100 - renal tubular reabsorption of glucose per 100 
ml of glomerular flltration. L > left kidney infused with cholera toxin 8 /ig per min: R - right noninfused kidney: Control - coUeaions obtained of prior 
to the infusion of cholera toxin and glucose. CT •¥ glucose > collections obtained at peak cfTects of chotera toxin ( 100-180 min) and stabte levels of high 
plasma glucose. 





100 - 

o o o o oo ft 
^ o 

• . • \ • 


900 — iioo — isie — iste — irte" 

GLUCOSE filtered (mg/iOOml 6FR) 

Fig. 1. The relationship between tubular reabsorp- 
tion of glucose and filtered load of glucose in dogs 
receiving cholera toxin into the left renal artery. Data 
from the left kidney receiving cholera toxin infusion are 
shown in black dots and from the right kidney not 
receiving cholera toxin infusion are presented in open 
circles. Data is expressed as mg per 100 ml GFR. 

cretion of NaCl comprised only 8% of urinary 
sodium from the contralateral kidney with 
the rest (92%) being NaHCOa (Fig. 3). 

Discussion. The results of the present study 
demonstrate that the infusion of CT into one 
renal artery is accompanied by a decrease in 
the renal tubular reabsorption of both glucose 

and bicarbonate by the infused kidney. 

Changes in glomerular flltration rate and 
alterations in status of extracellular fluid vol- 
ume (ECF) are known to influence tubular 
reabsorption of glucose (16, 17, 20, 28, 29). 
Thus, when the absolute amount of glucose 
reabsorbed is plotted against GFR in animals 
in which the rates of sodium reabsorption 
were unchanged, a direct linear relationship 
was found (17). In our studies, GFR was 
either unchanged or modestly increased in 
the infused kidney at a time when TRG was 
lower. This observation clearly excludes 
changes in GFR as the cause for the reduced 
TRG by CT. 

During expansion of ECF, there is an in- 
verse relationship between tubular reabsorp- 
tion of glucose per unit GFR and the fraction 
of filtered sodium excreted (17) suggesting 
that the mechanism responsible for the tu- 
bular reabsorption of glucose and sodium 
may be related. Furthermore, factors that 
inhibit renal transport of sodium such as 
ouabain or acetylstrophantidin suppress the 
reabsorption of glucose as well (30). Changes 
in the status of ECF could not account for 



TABLE II. Effects of Cholera Toxin on Renal Tubular Reabsorption of Bicarbonate." 




X 100 



















1. Control 










CT + NaHCO, 










2 Control 










CT + HCO, 










J. Control 










CT + HCO, 










4 Control 










CT + HCO, 










5 Control 










CT + HCO. 










6 Control 










CT + HCO . 










7 Control 










CT + HCO. 










Control mean 




















CT + NaHCO^ mean 


























CT + NaHCO. 





C onirol vs CT + NaHCO, 






" Each point represents the mean or three to Tive consecutive collections. The results during cholera toxin and bicarbonate infusion represent the mean 
or three to Ave consecutive collections during the peak efTect of cholera toxin ( 100-180 min) and stable high plasma bicarbonate. Ccr * clearance of 
creatinine; CWCr x 100 - fraction of Tiltered sodium excreted: PHCOi - plasma bicarbonate; TRHCOn - tubular reabsorption of bicarbonate: TRHCOi/ 
C( r X 100 - renal tubular reabsorption of bicarbonate per 100 ml of glomerular nitration: L - left kidney infused with cholera toxin. 8 ^ per min; R - 
right noninfused kidney; control - collections obtained prior to the infusion of cholera toxin and bicarboante: CT •¥■ HCOi - collections obtained during 
maximum effect of cholera toxin and during stable high levels of serum bicarbonate. 

8 so 

kJ 7 




2.5 3.0 3.5 4.0 


Fig. 2. The relationship between tubular reabsorp- 
tion of bicarbonate and filtered load of bicarbonate in 
dogs receiving cholera toxin into the left renal artery. 
Data from the left kidney receiving cholera toxin infusion 
are shown in block dots and from the right kidney 
without infusion are presented in open circles. Data is 
expressed as mEq per 100 ml GFR. 

our observation since both kidneys were sub- 
jected to the same conditions of ECF. How- 
ever, CT produced a greater degree of natri- 
uresis in the infused kidney and this may 



Fig. 3. Urinary excretions of sodium, bicarbonate 
and chloride during cholera toxin and bicarbonate in- 
fusion. Values are the mean and SEM of all experiments 
during maximal effects of cholera toxin. 

directly or indirectly have influenced glucose 

The tubular reabsorption of bicarbonate is 
affected by at least five factors. These include 
GFR (31), PCO2 (32, 33), potassium metab- 
olism (34), status of ECF (18, 19), and car- 
bonic anhydrase activity (33). Most of these 



factors could not account for the differences 
in TRHCO3/GFR from both kidneys. Acute 
reductions in GFR are associated with a fall 
in absolute TRHCO3 but TRHCO3/GFR re- 
mains unchanged (31), thus one would antic- 
ipate that a rise in GFR should be associated 
with a proportional increase in TRHCO3 with 
TRHCO3/GFR remaining constant. In our 
bicarbonate infusion studies, the GFR in in- 
fused kidney was higher than the control 
kidney but TRHCO3/GFR was lower. The 
blood levels of PCO2, potassium, as well as 
the state of ECF could not provide explana- 
tion for the unilateral decrease in 
TRHCO3/GFR since both kidneys were ex- 
posed to the same conditions. 

The mechanism(s) through which CT af- 
fects glucose and bicarbonate reabsorption 
are not evident. Several possibilities should 
be considered. First, CT most probably af- 
fects renal tubular transport processes by the 
stimulation of a renal adenylate cyclase with 
increased production of cyclic AMP (14, 35). 
Several lines of evidence exist indicating that 
cyclic AMP reduces reabsorption of various 
ions in the proximal tubule (36-38). The stud- 
ies of Lorentz (39) and Jacobson (46) suggest 
this effect of cAMP is mediated by an in- 
crease in tubular permeability allowing aug- 
mented back flux. It is therefore, plausible 
that the decrease in glucose and bicarbonate 
reabsorption during CT infusion is secondary 
to enhanced back flux of reabsorbate pro- 
duced by cAMP. The observation of Karlin- 
sky et al (41) who showed that the infusion 
of dibutyryl cyclic AMP reduced the tubular 
reabsorption of bicarbonate provides further 
support for the role of CT induced cAMP 
production in the genesis of reduced 

Second, cholera toxin may directly affect 
the tubular transport of glucose and bicar- 
bonate. In the ileum, CT enhances bicarbon- 
ate secretion (1) but there is no evidence for 
an effect of CT on glucose transport by the 
gut (42). Finally, CT may inhibit carbonic 
anhydrase activity and result in reduced reab- 
sorption of bicarbonate; there is no evidence 
as yet supporting such a contention. 

Most of glucose (21, 24, 25) and bicarbon- 
ate (22, 23) reabsorption occur in the proxi- 
mal tubule. Our present observations of de- 
creased TRG and TRHCO3 and natriuresis 

are consistent with an effect of CT in the 
proximal tubule. However, marked decreases 
in proximal tubular reabsorption of sodium 
may not be followed by substantial natriuresis 
unless distal reabsorption of sodium is also 
reduced (43, 44). It seems, therefore, that CT 
should have an effect on tubular reabsorption 
in more distal portions of the nephron as well. 
Indeed, the observation that during bicar- 
bonate loading 40% of the natriuresis in the 
infused kidney was due to NaCl as opposed 
to only 8% in the control noninfused kidney 
(Fig. 3) suggests that CT may have an effect 
on tubular reabsorption of Na at more distal 
sites of the nephron where Na is reabsorbed 
mostly as NaCl. 

The present results together with our pre- 
vious observations (14) have shown certain 
analogies between the natriuresis of expan- 
sion of ECF and that induced by CT: (a) 
Both are accompanied by depressed tubular 
reabsorption of glucose and bicarbonate, 
phosphate, calcium, magnesium and sodium 
chloride and (b) the relations between the 
fraction of Altered Na excreted and that of 
calcium and magnesium are similar in both 
conditions. Since extracellular fluid volume 
expansion with saline is accompanied by in- 
creased renal production of cyclic AMP (15) 
and the renal effects of CT are presumably 
mediated by stimulation of a renal adenylate 
cyclase-cyclic AMP system (14, 35), it could 
be postulated that at least part of the reduc- 
tion in the tubular reabsorption of these var- 
ious substances which occur during expan- 
sion of ECF may be mediated by increased 
production of cycUc AMP. 

Summary. Cholera toxin (CT) reduces tu- 
bular reabsorption of Na, CI, Ca, Mg and P 
most probably through stimulation of a renal 
adenylate cyclase-cyclic AMP system, and it 
is possible that an increased production of 
nephrogenous cyclic AMP during extracel- 
lular fluid volume expansion may be partly 
responsible for the observed natriuresis. In 
order to further evaluate the role of renal 
cyclic AMP in renal tubular transport, we 
studied the effect of CT on glucose (TRG) 
and bicarbonate reabsorption (TRHCO3). 

During the period of maximal effect of CT 
on tubular transport (100-140 min of CT 
infusion into one renal artery) both the TRG 
and TRHCO3 were lower in the infused kid- 


oey than in the contralateral noninfused kid- 
ney; TRG as mg per 100 ml GFR was 254 
± 32.7 vs 363 ± 43.5 (P < .01), and TRHCO3 
as mEq per 100 ml GFR was 2.09 ± 0.06 vs 
153 ± 0.06 (P < .01). The data indicate that 
CT suppresses glucose and bicarbonate reab- 
sorption together with that of sodium and as 
such assign to role for renal cyclic AMP in 
the regulation of the tubular transport of 
these substances. 

The authors wish to thank Mr. Barry Gammel and 
Mrs. Verginia Barbarian for their technical assistance, 
ind Ms. Melinda Ayers, Ms. Jamie Jimenez, Ms. Alice 
Moomjean and Ms. Alberta Ward for their secretarial 


1. Carpenter, C. C. J., Sack, R. B., Feely, J. C, and 
Steenberg, R. W., J. Clin. Invest. 47, 1210 (1968). 

2. Carpenter, C. C. J. Jr., and Greenough W. B. Ill, J. 
Clin. Invest. 47, 2600 (1968). 

3. Carpenter, C. C. J. Jr., Amer. J. Med. 50, 1 (1971). 

4. Field, M., N. Eng. J. Med. 284, 1137 (1971). 

5. Guerrant, R. L., Chen, L. C, and Sharp, G. W., J. 
Infect. Dis. 125,377(1972). 

6. Chen, C. C, Rhode, J. E.. and Sharp, G. W. G., J. 
Clin. Invest. 51,731(1972). 

7. Vaughan, M., Pierce, N. P., and Greenough, W. B. 
HI. Nature (London) 226, 658 (1970). 

8. Pierce, N. P., Greenough, W. B. Ill, and Carpenter, 
C. C. J. Jr., Bacteriol. Rev. 35, 1 (1971). 

9. Gorman, R. and Bitensky, M. W., Nature (London) 

10. Ucfatenstein, L. M.. Henny, C. S., Bourne, H. R.. 
and Greenough, W. B. Ill, J. Clin. Invest. 52, 691 

Bourne. H. R., Lehrer, R. L., Lichtenstein, L. M., 
Wetssman, G., and Zurier, R.. J. Clin. Invest. 52, 698 

II Mashiter, K., Mashiter. G. D., Hauger, R. L., and 
PiekL J. B., Endocrinology 92, 541 (1973). 

13. Donta. S. T., King, M., and Sloper. K., Nature (New 
Biol.) 243, 246 (1973). 

14. Priedler. R. M., Kurokawa, K.. Cobum, J. W., and 
Massry. S. G.. Kidney Int. 7, 77 (1975). 

15. Friedler, R. M., Descoeudres, C, Kurokawa, K., 
Kreusser. W. J., and Massry, S. G., Clin. Science 53, 

16. Robson, A. M., Srivastra, P. L., and Bricker, N. S., 
J. Clin. Invest. 47, 329 (1968). 

17. Kurtzman. N. A., White, M. G., Rogers, P. W., 
Flynn, J. J. Ill, J- Clin. Invest. 51, 127 (1972). 

IS. Purkerson. M. L., Lubowitz, H., White. R. W., and 

Bricker, N. S., J. Clin. Invest. 48, 175^ 

19. Kurtzman, N. S.. J. Clin. Invest. 49, 5 

20. Schultze, R. G., and Berger. H., Kidi 

21. Walker, A. M., Bott, P. A.. Oliver, 
Dowell, M. C, Amer. J. Physiol. 134, 

22. Gottschalk, C. W., Lassiter, W. E., a 
Amer. J. Physiol. 198, 581 (1960). 

23. Rector, F. C, Carter, N. W., and Sel 
Clin. Invest. 44, 278 (1965). 

24. Tune, B. M., and Burg, M. B., Amer. J 

25. Ullrich, K. G.. Rumrich, G., and Klos< 
Arch. 351, 35 (1974). 

26. Kadish, A. H.. Lille, R. L., and Ste 
Clin. Chem. 14, 116(1968). 

27. Hastings, A. B., and Sendroy, J. Jr., J 

28. VanLiew, J. B.. Deeijen, P., and B 
Pfluegers Arch. 295, 232 ( 1967). 

29. Keyes, J. L., and Swanson, R. E., Am 
221, 1 (1971). 

30. Csaky, T., Prachnabmoli, K., Eismani 
D. M., J. Pharmacol. Exp. Thep. 150, 

31. Lemieux, G., Methot, A. L., Gagnan- 
Plante, G. E., Kobreh, M., and Charghi 

32. Kurtzman, N. A., Amer. J. Physiol. 21! 

33. Pitts, R. P., Physiology of body flui 
Medical Publishers Inc., Chicago, 179 

34. Kurtzman, N. A., White, M. G., and I 
Metabolism 22, 481, 1973. 

35. Kurokawa, K., Friedler, R. M., and > 
Kidney Int. 7, 137(1975). 

36. Agus, Z. S., Gardner, L. B., Beck, L. I 
berg, M., Amer. J. Physiol. 224, 1 143 ( 

37. Agus, Z. S., Puschett, J. B., Senesky, 1 
berg, M., J. Clin. Invest. 50, 617 (1971 

38. Gill, J. R., and Casper, A. G.. J. Ch 
1231 (1971). 

39. Lorentz, W. B. Jr., J. Clin. Invest. 53, 

40. Jacobson, H. R., Clin. Res. 25, 436A ( 

41. Karlinsky, M. L., Sager, D. S., Kurtzm; 
Pillay. V. R. G., Amer. J. Physiol. 227 

42. Rohde, J. E., and Cash, R. A., J. Inf. 

43. Howards, S. S., Davis, B. B., Knox, F. 
S., and Berliner, R. W., J. Clin. ln\ 

44. Stein, J. H., Osgood, R. W.. Boonja 
Ferris, T. F., J. Clin. Invest. 52, 2313 ( 

Received March 23, 1978. P.S.E.B.M. I97j 


Shape Change and the Percentage of Sialic Acid Removed by Neuraminidase from 

Human Platelets' (40282) 


Department of Pathology, New York University Medical Center, New York, New York 100 J 6 

iV-acetylneuraminic acid is the only sialic 
acid found in human platelets (1) and 7- IS 
/ig is found per mg of platelet protein (2). 
Bacterial neuraminidases from Clostridium 
perfringens and Vibrio cholera, which cleave 
the a-ketoside linkage between sialic acid and 
the penultimate galactose or galactosamine, 
liberate 40-60% of sialic acid from human 
platelets (1-5). This is thought to be derived 
from surface membrane glycoproteins. 

In a recent study, Motamed et al (6) con- 
clude that platelet surface sialic acid increases 
after shape change. This report is based on 
data obtained from platelets fixed in plasma. 
A nonspecific attachment of plasma proteins 
to the platelet surface is observed after fixa- 
tion (7). As many of these plasma proteins 
contain sialic acid, and the platelet surface 
area increases after shape change (6), it is 
difficult to establish on the basis of these 
studies whether the increase in the amount of 
sialic acid removed by neuraminidase repre- 
sents increased platelet surface sialic acid, or 
merely an increase in the amount of plasma 
proteins fixed onto the platelet membrane. 

Ku and Wu (4, S), using washed platelets, 
reported that thrombin-, collagen- and ADP- 
induced platelet activation increased the 
amount of neuraminidase-removable sialic 
acid. The increases observed after thrombin 
and collagen treatment were inhibited by as- 
pirin, suggesting that material containing 
sialic acid is released during platelet activa- 
tion, and that the increase in sialic acid ob- 
served after activation does not represent sur- 
face sialic acid. 

This study was designed to determine 
whether ADP-induced shape change alters 
the amount of sialic acid on the surface of 
unfixed, aspirin-treated, gel-filtcred platelets. 

Materials and methods. C. perfringens neur- 

' This research was supported by USPHS Grant No. 
HL- 15596 from the National Heart, Lung and Blood 

aminidase (Type VI, Sigma) was purified 
according to the method of Hatton and Re- 
goeczi (8) and shown to be free of proteolytic 
activity against radioactive tosylarginine 
methyl ester (TAMe) (9) and fibrinogen (10). 
The activity of the purified enzyme was de- 
termined according to the method of Warren 
(11), and one unit is defined as the amount 
of enzyme required to liberate 1 ;anole of A^- 
acetyhieuraminic acid from bovine submax- 
illary mucin (Sigma) per min at 37°, pH 5.0. 

Nine ml of blood from normal volunteers 
was collected into 1.5 ml of acid-citratc- 
dextrose (ACD) solution (12) and 0.5 ml of 
1 mM acetylsalicylic acid (Merck) to inhibit 
the ADP-induced release reaction (13). After 
15-min incubation at 37°, platelet-rich 
plasma (PRP) was prepared by centrifugation 
at 280g for 8 min at room temperature. PRP 
was removed and centrifuged at 2100g for 20 
min. The resulting platelet button was resus- 
pended in M» the PRP volume with Tyrode's 
solution containing no added calcium (14), 
but with 0.2% bovine serum albumin (Sigma), 
and 0. 1 mg/ml potato apyrase (Sigma puri- 
fied Grade 1). The buffer was adjusted to pH 
7.4 and the platelet concentrate gel-filtered 
(IS) through Sepharose 2B (Pharmacia) 
equilibrated with Tyrode's solution contain- 
ing albumin but no added glucose, calcium 
or apyrase. The platelets were eluted with the 
same buffer and their ability to aggregate 
with 5 jLiAf ADP was established on a small 
aliquot after adding 0.5 mg/ml fibrinogen 

The suspension, containing 0.5-1.0 X 10® 
platelets/jbd, was divided into four 0.9 ml 
aliquots and brought to 37°. Two samples 
were treated with 0.1 ml 50 fiM ADP, and 
two with 0. 1 ml 0. 1 5 A/ NaCl. The tubes were 
gently inverted twice, and a small aliquot of 
each was fixed in 1% formalin for examina- 
tion under the phase contrast microscope. 

Approximately 1 min after the addition of 
ADP, the pH of the samples was adjusted to 


0037-9727/78/ 1 59 1 -0054$0 1 .00/0 

Copyright (£> 1978 by the Society for Experimental Biology and Medicine 
A// rights reserved. 



ith 0.1 N HCl. One ADP-treated and 
aline control sample than received 40 
neuraminidase (0.21 U/ml). The other 
treated and saline control sample re- 
1 40 fd of ammonium acetate buffer. All 
es were incubated for 15 min at 37°. 
incubation, phase microscope observa- 
howed that the saline control platelets 
stained their discoid shape and the sus- 
)n exhibited the characteristic "swirl" 
agitated. In contrast, the ADP-treated 
es were spiny spheres and failed to swirl 
shaken. A portion of the platelet sus- 
)ns not treated with neuraminidase was 
:ted to mild acid hydrolysis (0.1 N 
4, 1 hr, 80°) and total sialic acid was 
nined. Platelets were counted in another 
•n with a Coulter Counter. The remain- 
f the control samples as well as the 
minidase-treated samples were then 
fuged in a Serofuge (Clay Adams) for 
n. The supematants were removed and 
acid measured without hydrolysis (11). 
irder to determine whether sialic acid is 
sd from platelets after treatment with 
bin and connective tissue as reported 
I and Wu (4), 9 ml of blood was col- 
into 1.5 ml of ACD containing 10 jitl 
rotonin (Amersham, 8 jitCi/ml) (16) and 
»sed as previously described. Fifty jkI of 
ispension of *^C-serotonin-labeled gel- 
d platelets was placed in glass counting 
containing 10 ml Aquasol (New Eng- 
4uclear) and counted in a Packard Liq- 
cintiUation Counter. Another aliquot, 
, was used to determine total sialic acid 
mild acid hydrolysis. The remaining 
ision was divided into three equal ali- 
and treated with 1 U/ml (final concen- 
i) of highly purified human thrombin 
y given to us by Dr. John W. Fenton 
jw York State Department of Health, 
y), a suspension of ground human sub- 
Mus connective tissue (given to us by 
. Lackner, New York University Med- 
cnter), or isotonic saline. After gentle 
g, the suspensions were incubated at 
r 10 min. The samples were then chilled 
sntrifuged at 4°C at 2100 g for 20 min 
let the platelets. The sialic acid was 
ired on 190 jkI of each supernatant after 
icid hydrolysis (11) and 50 jitl of each 
latant was used to measure ^^C. 

Results, The total sialic acid content of the 
ADP-treated and saline control samples did 
not differ. Hence, both values were included 
in the average, which was 60 nmoles/10® 
platelets (Table I). Neuraminidase removed 
47% of sialic acid from both control and 
ADP-treated platelets. There was no sialic 
acid in the supematants of samples treated 
with buffer instead of enzyme, even after 

Two experiments were carried out in which 
platelets were incubated with thrombin or 
connective tissue for 10 min without shaking. 
The platelets lost their "swirl" but no aggre- 
gates were seen on gross inspection. *^C-ser- 
otonin release in the saline control platelets 
was 13.6% and 4.2% (Table II). This material 
is presumably released when platelets come 
into contact with Sepharose beads during gel 
filtration (17). Platelets treated with thrombin 
released 89.1% and 81.2% of their ^^C-sero- 
tonin, and 30% and 43% of their total sialic 
acid, respectively. Platelets treated with col- 
lagen released 57.6% and 33.2% of their ^^C 
serotonin, and 30% and 31% of their total 
sialic acid. 

Discussion. ADP-induced shape change 
does not cause the release of material con- 
taining sialic acid from aspirin-treated plate- 
lets, as none was detected in the supematants 
even after hydrolysis. Thus the sialic acid 
measured in the supematant of suspensions 
treated with neuraminidase presumably rep- 
resents sialic acid cleaved from membrane 

The amount of sialic acid removed by 
neuraminidase is not altered by ADP-in- 
duced shape change. This finding agrees with 
the results of Bunting and Zucker (3) who 
demonstrated that the same amount of trit- 
ium was incorporated into the siahc acid of 
discoid platelets and platelets which had un- 
dergone shape change after exposure to ADP. 

Like others (2, 18), we found that collagen 
and thrombin release material containing 
sialic acid from human platelets. As neura- 
minidase will cleave sialic acid from both the 
platelet membrane and the released material, 
it is essential to prevent the release reaction 
in order to quantitate changes in membrane 
sialic acid during platelet stimulation. 

Since we fmd no difference in the amount 
of sialic acid removed by neuraminidase from 



TABLE I. The Effect of Shape Change on the Amount of Sialic Acid Removed by Neuraminidase from 

Human Platelets. 

Total sialic acid 
(nmoles/lO* platelets) 

Sialic acid removed by neur- 

Change in sialic 
acid removed 

Expt. No. 





































































' t s 0.382, not statistically significant. 

TABLE II. Percent Release of '*C Serotonin and 
Sialic Acid by Stimulated Platelets. 


Sialic acid 

Expt. No. 

Agent added 

(% of To- 

in super- 
natant (% 
of Total) 



















discoid platelets and spiny spheres, we con- 
clude that ADP-induced shape change does 
not alter platelet surface sialic acid. In con- 
trast, sialic acid appears to be lost from plate- 
lets which have been aggregated with ADP 
and disaggregated (3). 

Summary. The siahc acid of human gel- 
filtered platelets was studied before and after 
ADP-induced shape change. Neuraminidase 
cleaved 47% of the total sialic acid from both 
discoid control platelets and platelets that had 
become spiny spheres after treatment with S 
fiM ADP. 

We are grateful to the following colleagues for mate- 
rials used in these experiments: Thomas H. Finlay, Ph.D., 
for the radiolabeled TAMe, Joel U. Harris, Ph.D.. for 
the radiolabeled fibrinogen, Henriette Lackner, M. D., 
for the connective tissue suspension, and John Fenton, 
11, Ph.D. for the purified human thrombin. 

1. MadofT, M. A., Ebbc, S., and Baldini, M., J. Clin. Received March 6, 1978. P.S.E.B.M. 1978, Vol. 159. 

Invest. 43, 870 (1964). 

2. Greenberg, J., Packham, M. A., Cazenave, J. -P., 
Reimers, H.-J., and Mustard, J. P., Lab. Invest. 32, 

3. Bunting, R. W., and Zucker, M. B., Blood 50, Suppl. 

4. Ku, C. S. L., and Wu, K. K., Blood 50, Suppl. 1, 244 

5. Wu, K. K., and Ku, C. S. L., Clin. Res. 26, 359A 

6. Motamed, M., Michal, P., and Bom, G. V. R., 
Biochem. J. 158,655(1976). 

7. Zucker, M. B., Excerpta Medica Int. Congr. Series 
No. 415,280(1978). 

8. Hatton. M. W. C, and Regoeczi, E., Biochim. Bio- 
phys. Acta 327, 114(1973). 

9. Roffman, S., Sanocka, U., and Troll, W., Anal. 
Biochem. 36, 11(1970). 

10. Martinez, J., Palascak, J., and Peters, C, J. Lab. 
Clin. Med. 89,367(1977). 

1 1. Warren, L., J. Biol. Chem. 244, 4406 (1%9). 

12. Aster, R. H., and Jandl, J. H., J. Clin. Invest. 43, 843 

13. Zucker, M. B., and Peterson, J., Proc. Soc. Exp. Biol. 
Med. 127,547(1968). 

14. Walsh, P. N., Brit. J. Haematol. 22, 105 (1972). 

15. Tangen, O., Berman, H. J., and Marfey, P., Thromb. 
Diath, Haemorrh. 25, 168, (1971). 

16. Valdorf- Hansen, J. P., and Zucker, M. B., Amer. J. 
Physiol. 220, 106(1971). 

17. Tangen, O., Andrae, M. L., and Nilsson, B. E.. 
Scand. J. Haemtol. 11, 241 (1973). 

18. Hagen. 1., Biochim. Biophys. Acta 273, 141 (1972). 


L-Histidine-lnduced Facilitation of Cholesterol Biosynthesis in Rats^ (40283) 

Laboratory of Behavioral Neurochemistry, Waisman Center, University of Wisconsin; and the Veterans 
Administration, Madison, Wisconsin 53706 

tritional loading of amino acids, partic- 
phenylalanine, has been used to study 
Q metabolic disorders. During such ex- 
entation Waisman and his colleagues 
histidine-supplemented diets to infant 

> monkeys and noted a marked hyper- 
ia (1, 2). Later Geison and Waisman (3) 
^ and 8% excess L-histidine diets to 4- 
old rabbits and induced a 30% increase 
^ma cholesterol levels. Our investiga- 
have been pursued in rats, attempting 
>vide a more accessible animal model 
iidying dietary histidine supplementa- 

s fed a diet supplemented 5% with l- 
ne develop large livers and hypercho- 
>lemia (4). There is an increase in the 
K-ration of cholesterol precursors into 
terol in liver slices from rats fed excess 
ne (4). This finding prompted further 
igation to determine the effect of histi- 
upplementation on cholesterol biosyn- 
in the 5,000g supernatant solution of 
er homogenate. Mature rats were used 
s study because cholesterol and fatty 
letabolism in weanling rats is unstable, 

> the change of diet from milk to chow 
'evious studies have demonstrated that 
I decreases the rate of synthesis of cho- 
)1 from acetate (6, 7). When fasted ani- 
vere rcfed a normal diet, the synthesis 
>lesterol from acetate returned to nor- 
dthin three days (8). When they were 
a a fat-free diet, cholesterol synthesis 
ed to its normal level within three days 
len declined to a very low level (8). This 
igation studied the effects of histidine 
^mentation on the rate of synthesis of 

I investigation was supported by grants from the 
Research Service of the Veterans Administra- 
ipiul and the University of Wisconsin. A prelim- 
port of this work was presented at the 61st 
Meeting of the Federation of American Societies 
rimental Biology, Chicago, Illinois, April 1-8, 
Ktract No. 2782 

cholesterol and cholesterol precursors from 
acetate and mevalonate in both normal and 
fat-free diets. All measurements were ob- 
tained during high and low diurnal levels of 
cholesterol synthesis. 

Materials and methods. Experimental ma- 
terials were obtained from the following 
sources: [2-^^C] acetate (specific activity 53.3 
mCi/mmole), [2-*^C] RS-mevalonic acid, 
iV,iV'-dibenzylethylene diammonium salt 
(specific activity 40.2 mCi/mmole, and Aqua- 
sol (scintillation solution) from New England 
Nuclear Corp., Elmhurst, IL; glucose-6-phos- 
phate, NAD, NADP, dithiothreitol, digi- 
tonin, and nicotinamide from Sigma Chemi- 
cal Co., St. Louis, MO; EDTA from Fisher 
Scientific Co., Itasca, IL; L-histidine (free 
base) and bovine serum albumin from Nutri- 
tional Biochemical Corporation, Cleveland, 
OH. All other chemicals used were of analyt- 
ical grade. The fat-free diet (Wooley and 
Sebrell), Mod. TD-71125 was from Teklad 
Test Diets, Madison, WI. The normal diet 
was ground Purina Formulab Chow. In the 
histidine-supplemented diets, L-histidine con- 
stituted 5% of the diets by weight. A standard 
fitting Potter-Elvehjem homogenizer was 
used for homogenization. All radioactivity 
countings were done in a Nuclear Chicago 
Scintillation Counter, Isocap/300. 

Male albino rats weighing 30-60 g each 
were obtained from Holtzman Rat Co., Mad- 
ison, WI. Animals were divided into groups 
of four and fed normal and experimental 
diets ad lib. for 18 days after they were re- 
ceived. All rats, excluding the control group, 
then fasted for 2 days and were then refed 
experimental diets ad lib. for three days. This 
provided 21 days of experimental diet as used 
in previous studies of amino acid feeding (9). 
Rats were housed singly in stainless steel 
cages. The light cycle was from 7am to 

Preparation of rat liver homogenate. Rats 
were sacrificed by decapitation, at 2pm or 


0037-9727/78/ l59\-O0SlSO\.Wi/Q 

Copyright © 1978 by Ihe Socxdy tot ^^jcumttvVsX^voNo^ Mv^>\«AL\«c\fc 
All rights reserved. 



lOpM, and the livers were removed quickly 
and placed on ice. Each liver was weighed, 
minced, and then homogenized in a 0.1 M 
potassium phosphate buffer, pH 7.4, contain- 
ing 0.004 M MgClz, 0.001 M EDTA, and 
0.002 M dithiothreitol, with five strokes of a 
Potter-Elvehjem homogenizer. The volume 
of buffer used was 2 nU/g of liver. The ho- 
mogenate was centrifuged for 10 min at S,- 
OOOg. The volume of the supernatant solution 
was recorded. Protein concentrations were 
measured by a modification of the biuret 
procedure (10) using bovine serum albumin 
as standard. 

Assays for the conversion oj acetate and 
mevahnate to NSF^ and DPF^. The rates 
of conversion of I2^^C] acetate and [2^^C] 
mevalonate to NSF and DPF were measured 
by a slight modification of the procedure of 
Slakey et al (11). With acetate as the sub- 
strate, the incubation mixture contained 125 
yX (approximately 5.0 mg protein) of the 5,- 
QOQg supernatant solution diluted to 0.5 ml 
with homogenizing buffer plus cofactors and 
[2-"C] acetate (2.5 /xmoles and 4 x 10* dpm 
per /xmole). With mevalonate as the substrate, 
the incubation mixture contained 75 /il (ap- 
proximately 3.0 mg protein) of the 5,000^ 
supernatant solution diluted to 0.5 ml with 
homogenizing buffer plus cofactors and [2- 
^^C] RS-mevalonate (2.5 /mioles and 2 x 10* 
dpm per /imole). The NSF was counted in a 
toluene scintillation solution and the DPF 
was counted in Aquasol 

Results. Acetate to NSF and DPF. The 
incorporation of [^^C] acetate into the NSF 
and DPF of the 5000g supernatant solution 
of rat liver homogenate is shown in Fig. 1. 
The labeled substrate was incorporated nine 
times more into the NSF of rats which were 
refed a histidine-supplemented chow diet 
than in those of the control group (Fig. lA). 
This increase is statistically significant {P < 
0.001). Refeeding chow, fat-free, or a histi- 
dine-supplemented fat-free diet did not sig- 
nificantly affect the NSF synthesis activity. 
Refeeding of the histidine-supplemented 
chow diet induced a seven- to eightfold in- 
crease in the incorporation of the labeled 

^ NSF = Nonsaponifiable fraction: sterols, squalcnc, 
and terpenols. 

^ DPF = Digitonin-prccipitable fraction: sterols. 

Fig. i. Effects of L-histidine supplementation on the 
rate of conversion of [2-'^C] acetate to the nonsaponifi- 
able (A) and digitonin-precipitable (B) fractions in the 
5000^ supernatant solution of liver homogenates of rats 
maintained in different nutritional states: continuously 
fed, chow (A); fastcd-refed, chow (B); fasted-rcfcd, 95% 
chow + 5% L-histidine (C); fasted-refed, fat-free (D); and 
fasted-refed 95% fat-free +L-histidinc (E). Vertical bars 
represent standard deviations with four rats in each 

substrate into the DPF (Fig. IB). This differ- 
ence is also significant (P < 0.001). Histidine 
supplementation to the fat-free diet did not 
cause a significant increase in the DPF syn- 
thesis activity. 

Mevalonate to NSF and DPF. Effects of 
feeding excess histidine on the incorporation 
of [^^C] mevalonate into the NSF are shown 
in Fig. 2A. A 7.5-fold increase in total syn- 
thesis activity over the matched control was 
observed when refeeding the histidine-sup- 
plemented chow diet (P < 0.001). Refeeding 
of chow, fat-free diet, and a histidine-supple- 
mented fat-free diet did not significantly af- 
fect the NSF synthesis activity. The amount 
of [*^C] mevalonate incorporated into the 
DPF was seven times higher with the refed 
histidine-supplemented diet than with the 
refed chow diet (P < 0.001), as shown in Fig. 
2B. Synthesis activity was 1.6 times higher in 



□ 2PM 




A B C E 



2. Effects of L-histidine supplementation on the 
the conversion of [2-'^C]mevalonate to the non- 
lable (A) and digitonin-precipitable (B) fraction 
5000^ supernatant solution of liver homogenates 
maintained in different nutritional states de- 
in Fig. 1. Vertical bars represent standard devia- 
ith four rats in each group. 

rhich were refed the fat-free diet than in 
mtinuously fed control (P< 0.05). His- 
; supplementation to a fat-free diet did 
ause any significant increase over the 
led control. 

cussion, L-histidine or a histidine metab- 
sfTectively stimulates sterol synthesis in 
irhen added to chow diet. The marked 
ISC (seven- to ninefold) in the incorpo- 
1 rate of labeled substrates into both 
and DPF in this study is considerably 
IX than the increase in plasma choles- 
(30% over normal) which occurred in 
ler histidine supplement study (4). A 
taneous increase in the degradation of 
sterol in the liver may be responsible for 

olesterol synthesis varies diumally (9, 
lowever, none of the enzyme activity 
;x>nverts mevalonate to squalene does 
In this investigation, sterol synthesis 
either acetate or mevalonate at the high 

point of the day was 1.3-1.6 times greater 
than at the low point. 

Under a variety of experimental conditions 
which reduce the conversion rate of acetate 
to cholesterol, the conversion rate of meva- 
lonate to cholesterol does not change or 
changes much less dramatically than that of 
acetate (7, 14). However, in the case of stim- 
ulation of sterol biosynthesis by histidine, a 
similar rate increase was obtained when 
either acetate or mevalonate was used as the 
labeled substrate. The result suggests that 
histidine probably has a significant effect on 
an enzyme or enzymes in the synthesizing 
pathway between mevalonic acid and choles- 
terol. It will be interesting to investigate the 
activities of these enzymes in future studies. 

Refeeding of either a chow or fat-free diet 
did not cause a marked change in sterol and 
squalene synthesis (1.2- to 1.8-fold increase 
over controls). This agrees with results ob- 
tained by Craig et al (8) which show that the 
cholesterol synthesis activity rises from fast- 
ing levels to normal levels within three days 
after refeeding either chow or fat-free diet. 
Histidine supplementation to the fat-free diet 
did not cause a substantial change in the rate 
of sterol and squalene synthesis from acetate. 
This contrasts with the marked increase in 
sterol and squalene synthesis in the histidine- 
treated chow fed group. 

The livers from rats fed fat-free diets, re- 
gardless of histidine treatment, were deep 
yellow due to fat accumulation. This proba- 
bly resulted from a higher rate of fatty acid 
synthesis. If this is true, acetyl-CoA, a com- 
mon precursor for these two divergent path- 
ways (cholesterol and fatty acid synthesis), 
could be exhausted from an endogenous pool 
with long-term feeding, thus impeding histi- 
dine's stimulation of cholesterol synthesis 
from acetate in rats fed a fat-free diet. How- 
ever, the conversion rate of mevalonate into 
sterols and squalene in rats which were fed a 
longterm fat-free diet also did not change 
when histidine was added to their diet. TIus 
result might not be anticipated if the absence 
of acetyl-CoA accounted solely for the lack 
of a histidine effect in rats fed a fat- free diet. 
The next step in the study of these processes 
will be to measure the actual activities of the 
specific enzymes, such as )8-hydroxy-)8-meth- 
ylglutaryl CoA reductase and fatty acid syn- 



Summary. A diet supplemented 5% with 
L-histidine induces hypercholesterolemia in 
rats. To examine the mechanism involved, 
L-histidine was added to either a chow or fat- 
free diet and fed to rats for 18 days. After 2 
days of fasting, the rats were refed the same 
diet for three days. There was a ninefold 
increase in the incorporation of [^^C] acetate 
into the nonsaponifiable fraction in the S,- 
OOOg hepatic fraction of histidine-supple- 
mented chow-fed rats compared to controls. 
The increase in the incorporation of the la- 
beled substrate into the digitonin-precipitable 
fraction was seven- to eightfold. The incor- 
poration of [^"^Clmevalonate was increased by 
sevenfold in both the nonsaponifiable and 
digitonin-precipitable fractions. Longterm 
histidine supplementation to fat-free diet did 
not affect the incorporation of either [^^C] 
acetate or [^^^C] mevalonate into these frac- 

Wc wish to thank Ms. Cynthia Birch for her technical 

1. Kerr, G. R.. Wolf, R. C, and Waisman, H. A., Proc. 
Soc. Exp. Biol. Med. 119, 561 (1965). 

2. Kerr, G. R., Wolf, R. C, and Waisman, H. A., in 
"Symposia of the Zoological Society of London" 

(R.N.T.-W. Fiennes, ed), no. 17, p. 371, Academic 
Press, London/New York (1966). 

3. Geison, R. L., and Waisman, H. A., Proc. Soc. Exp. 
Biol. Med. 133,234(1970). 

4. Solomon, J. K., and Geison, R. L., Fed. Proc. 36, 
1157 (Abslr.) (1977). 

5. McNamara, D. J., Quackenbush, F. W., and Rod- 
well, V. W., J. Biol. Chem. 247, 5805 (1972). 

6. Tomkins, G. M., and Chaikoff, I. L.. J. Biol. Chem. 

7. Bucher, N. L. R., McGarrahan, K., Gould, E., and 
Loud, A. v., J. Biol. Chem. 234, 262 (1959). 

8. Craig, M. C, Dugan, R. E., Muesing, R. E., Slakey, 
L. L., and Porter, J. W., Arch. Biochem. Biophys. 

9. Daniel, R. G., and Waisman, H. A.. Growth 32, 255 

10. Gomell, A. G., BardawilK C. J., and David, M. M., 
J. Biol. Chem. 177,751(1949). 

11. Slakey, L. L., Craig, M. C, Beytia, E., Bricdis, A., 
Feldbruegge, D. H., Dugan, R. E., Qureshi, A. A., 
Subbarayan, C, and Porter. J. W., J. Biol. Chem. 

12. Back, P., Hamprecht, B., and Lyncne, F., Arch. 
Biochem. Biophys. 133, 11 (1969). 

13. Dugan, R. E., Slakey, L. L., Briedis, A. V., and 
Porter, J. W., Arch. Biochem. Biophys. 152, 21 

14. Gould, R. C, and Popjack, G., Biochem. J. 66, 51 

Received February 27, 1978. P.S.E.B.M. 1978. Vol. 159. 


)tion to Clostridium botulinum Cultures of Phage Controlling Type C Botulinum 
Toxin Production (40284) 


Research Institute and Department of Bacteriology, University of Wisconsin, Madison, Wisconsin 53706 

production by Clostridium botulinum 
and D is closely associated with 
phage infection. Cultures become 
»enic when cured of a specific tem- 
»hage while nontoxigenic strains can 
erted to toxigenicity when infected 
; phage (3-6, 8-10). However, the 
on does not occur with all combina- 
^ phage and cultures; only certain 
ulture pairings are productive of con- 
The needed specificity was explained 
; due to differences in the adsorption 
» to cells when three antigenic groups 
mtified among the converting phages 
e possibiUty of other explanations has 
sed by a later report (6) which shows 
same culture can be made toxigenic 
;enically distinct phages, 
present communication further ex- 
the phage-culture specificity needed 
;enicity conversion by comparing the 
ion of one of the type C toxin-con- 
phages to several C. botulinum types 
b cultures and their nontoxigenic de- 

rials and methods. Table I shows the 
►f the cultures used (9-11). Nontoxi- 
:)-A02 and (D)-139 can be lysogen- 
itinely with c-st phage (from C-Stock- 

as to produce type C toxin. 

ires were maintained in Bacto- 

Meat Medium (Difco Lab., Detroit, 

►r the tests, they were grown in LYG 

1 of pH 7.2 made of 1% lactalbumin 
Chemical Co., St. Louis, MO), 2% 
tract (Difco), 0.5% glucose and 0.15% 
-HCl. Plating medium was Bacto- 
eart Infusion Agar (Difco) containing 
v) whole human blood dbtained from 
bank. Plated cultures were incubated 
-obic jars but other cultures were in- 

ment address: Department of Bacteriology, 
Medicine. Hokkaido University, Sapporo, Ja- 

cubated in tightly closed screw-capped tubes. 
All cultures were incubated at 37°. 

Filtrates of C-Stockholm cultured over- 
night in LYG contained c-st phage titer of 
10 plaque forming units (pfu)/ml when 
plated with indicator culture (C)-A02. The 
phage was purified by three successive cycles 
of incubating a transferred plaque for 4 hr 
with (C)-A02 actively growing in 5 ml of 
LYG, filtering culture lysate through Milli- 
pore membrane of 450 nm pores, and replat- 

Two ml of the broth culture from the third 
passage were added to 15 ml of a young 
culture (A620 = 0.2) of (C)-A02 and the cul- 
ture incubated until lysis occurred during the 
next 3-4 hr. The lysate, clarified by centrifu- 
gation and subsequent filtration through a 
Millipore membrane, had a titer of about 10^ 
pfu/ml but the titer decreased during storage 
of more than one week at 4®. The titer was 
regained when the phage stock was treated as 
in the last passage used in its preparation. 

Adsorption tests were done in T2 buffer 
made of 0.4% NaCl, 0.5% K2SO4, 0.15% 
KH2PO4, 0.3% Na2HP04, 1 mM MgS04, 0.1 
mM CaCl2, and 0.001% gelatin (7). This 
buffer was used because of convenience 
rather than superiority over other media. Pre- 
liminary tests have shown that c-st phage 
adsorbs to (C)-A02 equally well in systems 
using LYG, T2 buffer, or T2 buffer contain- 
ing 40 jug tryptophane/ml. 

Cells for adsorption tests were collected 
from overnight incubated cultures by centrif- 
ugation at 3000g for 10 min and washed three 
times with T2 buffer. Suspensions of 1 x 10® 
cells/ml were made on the basis of counts 
made on a Petroff-Hausser counting cham- 

The phage preparation was diluted 1:10 
with T2 buffer. After holding separately in 
an ice water bath for 5 min, 1.8 ml cell and 
0.2 ml phage preparations were combined 
and held at 4®. After the desired adsorption 


0037-9727 /IS/ \59\-0%\Vi\ .«i/Vi 

Copyright © \n% b^ iV« Socicv^ fox Exveuttv«tA»Wvo\n%>j wA>K.t^>sM>fe 
All rights reserved. 



time, the suspension was centrifuged for 10 
min at 4° and 6000g. Unadsorbed phage was 
quantified by plating 0. 1 ml of the decimal 
dilution series of the supernatant with the 
indicator strain. 

The frequency of conversion was deter- 
mined by examining isolated colonies. After 
cell-phage contact, the cells were collected by 
low speed centrifugation (lOOOg, 10 min) and 
plated to obtain isolated colonies. Of the 
colonies developing during 2 days incubation, 
20 random picks were subcultured separately 
for 3 days in tubes of Cooked Meat Medium. 
The presence of type C botulinum toxin in 
these cultures were determined by challeng- 
ing mice ip with O.S ml of culture fluid. 

Results, Adsorption curves of c-st phage 
reacting with (C)- A02 were not different from 
those reported for most other phage systems. 
Phage adsorption depended on the multiplic- 
ity of infection (MOI): starting with 4.0 x 10^ 
pfu and 1.8 x 10® cells/ml, 98% of phage was 
adsorbed in 10 min while 50% was adsorbed 
when the cell concentration was 5.5 x 10^/ 

Adsorption of c-st phage to cells of differ- 
ent cultures during 20 min contact at 4° is 
shown in Table II. Several controls showed 
the reduction in free phage was due to specific 
adsorption. As part of the first experiment of 

TABLE I. Culture Strains Used. 

TABLE 11. Absorption of c-st Phage to type C ani> 

D Strains and Nontoxigenic Strains Derived 

FROM Toxigenic Parents. [ x \(f CEixs/ml; Free 

pfu After 20 min Cell-Phage Contact at 4°C. 

pfu/ml of supernatant fluid 

Type toxin 






wild type 



wild type 



(C)-A02 infected with c-st 



(D)-139 infected with c-st 



AG* trealmenl of C- 



NG* treatment of C- 



spontaneously from wild 



spontaneously from wild 



AG treatment of D-1873 



AG treatment of D- 1873 


spontaneously from wild 
type D-South African 


Expt. 1 

Expt. 2 

Expt. 3 

No cells 

4.3 X 10* 

1.8 X 10* 

1.2 X lO' 


4.0 X 10' 

6.0 X 10" 

3.4 X lO' 


1.7 X 10* 


1.5 X 10* 


7.0 X 10" 


2.8 X lO' 


1.4 X 10" 


1.5 X lO' 


4.0 X 10' 


1.0 X lO' 


3.8 X lO' 


1.3 X lO' 

" For toxigenics, letter indicate type of toxin produced; 
letter in ( ) indicates toxin type of parent from which 
nontoxigenic was derived. 

* AG = acridine orange; NO « nitrosoguanidine. 

Table II, possible adsorption to a nonproteo- 
lytic C botulinum type B culture (QC strain) 
and a type E (Morai strain) was examined. 
The respective titers of 4.5 x 10^ and 4.0 X 
10^ pfu/ml after the adsorption treatments 
showed that c-st did not adsorb to these cells 
nor was it adversely affected by them. Phage 
inactivating factors were not produced by 
cells since titers of 5.2 x 10^ and 4.4 x l(r 
pfu/ml, respectively, were found after treat- 
ing the phage suspension with cell-free cul- 
ture fluids of (D)- 151 and (D)-1873. 

Some quantitative differences were found 
in retesting the same cultures, but the conclu- 
sion can be drawn that c-st adsorbed to all 
cultures except (C)-6813 and (C)-6814. The 
phage adsorbed best to indicator strain (C)- 
A02, parent toxigenic C-Stockholm, and con- 
verted (C)-A02(c-st). The phage adsorbed to 
a slightly less degree to D toxin producer D- 
1873 and non toxigenics derived from parents 
producing this type of toxin. 

As reported previously (11), the phage 
lysed broth cuhures of only (C)-A02 and (D)- 
139. When the lysates were subcultured in 
Cooked Meat Medium, type C toxin was 
formed (12). The phage produced plaques on 
lawns of these two cultures but not on those 
of others. However, by degrees of clearing of 
broth cultures and numbers of plaques 
formed, the phage was more overtly active 
against (C)-A02 than (D)-139. 

(C)-A02 and (D)-139 differed also in their 
rates of conversion to toxigenicity (Table III); 
with optimum conditions of cell-phage con- 



II. Conversion Rates OF (C)-A02 AND (D)- 
Strains by c-st Phage with Optimum 
iTiON Times (4 hr) and in Presence of 2% 







tion min^ 

20 tested 

i — 




i — 




I + 






















plicity of infection. 

>ccll contact time at 37*^. 

OI = 0.1, 4 hr), the conversion to 
city of (C)-A02 was significantly 
than (D)-139. Raising MOI to 5.0 
d the conversion rate for (D)-139. 
% NaCl in the adsorption system in 
s to increase conversion to toxigenic- 
did not favor greater conversion of 

erted isolates of the two culture 
produced approximately the same lev- 
^ LDso/ml) of toxin. When nontoxi- 
olates from the first treatment were 
d to a second conversion test, the 
ion rate of (D)-139 was again lower 
• (C)-A02. 

it was made of the possibility that 
I converted strain (D)-139(c-st) might 
xlified form that could convert (D)- 
a higher rate than the c-st phage 
1 directly from C-Stockhohn. The 

filtrate of an overnight incubated 
(c-st) culture was added to separate, 
growing cultures of (D)- 139 and (C)- 
\tT 4-hr cell-phage contact, the mix- 
ere plated and 20 resulting colonies 

for toxicity tests. None of the (D)- 
ates produced toxin although 13 of 
A02 subcultures had been converted 

ssion, (C)-A02 and (D)-139 were both 
id to type C toxigenicity by c-st 
although the conversion frequency 
lificantly higher for (C)-A02. The 
::e in the conversion rates is related to 
e effective phage adsorption to (C)- 
neasured by comparative adsorption 

results and lysis of broth cultures. 

Several reasons are involved in only certain 
phage-cell pairings being productive of con- 
version to toxigenicity. Included are cases 
where the cells lack receptors for phage at- 
tachment. This situation is illustrated by (C)- 
6813 and (C)-6814 to which c-st phage did 
not adsorb. 

Since c-st adsorbed to some extent to all 
other cultures used, the conversion or non- 
conversion of these cultures is not determined 
by phage adsorption only. (C)-N71 is already 
lysogenized by a nonconverting phage. Since 
this phage has the same host spectrum and 
antigenicity as c-st (11), its presence in the 
cells would confer immunity against the con- 
verting c-st phage. The result would be non- 
conversion to toxigenicity in spite of adsorp- 
tion of c-st to the cells. 

This nonconverting phage could not be 
demonstrated in the remaining cultures to 
which c-st phage adsorbs without converting 
to toxigenicity. It is possible that some of 
these cultures carry a defective phage that 
confers inmiunity against c-st phage; in oth- 
ers, host controlled restriction (1, 2) may be 
important in preventing conversion. 

Summary. C-st phage which governs pro- 
duction of type C botulinum toxin was mixed 
at 4° with cells of C. botulinum type C and D 
cultures and nontoxigenics derived from 
them. The phage adsorbed to all three cul- 
tures producing type C toxin, the one type D 
toxin producer, 2 of 4 nontoxigenics from 
type C parents and the three nontoxigenics 
originating from type D toxin producers. The 
phage adsorbed to some cultures without con- 
verting to toxigenicity. The two nontoxigenic 
which could be converted to toxigenicity dif- 
fered in degrees of phage adsorption and 
conversion rates. 

Research was supported by the College of Agricul- 
tural and Life Sciences, University of Wisconsin, Madi- 
son, by NIH Grant No. FD00712, and contributions 
from food companies. 

1. Arber, W., Ann. Rev. Microbiol. 19, 365 (1965). 

2. Arber, W., and Linn, S., Ann. Rev. Biochcm. 38, 467 

3. Eklund. M. W., Poysky, F. T., Reed, S. M., and 
Smith, C. A., Science 172, 480 (1971). 

4. Eklund, M. W., Poysky, F. T., and Reed, S. M., 
Nature (London) New Biol. 235, 16 (1972). 



5. Eklund, M. W., and Poysky, F. T., Appl. Microbiol 

6. Hariharan, H., and Mitchell W. R., Appl. Environ. 
Microbiol. 32, 145(1976). 

7. Hcrshcy, A. D., and Chase, M.. J. Gen. Physiol 36, 

8. Inoue, K., and lida, H., Jap. J. Microbiol 14, 87, 

9. Inoue, K., and lida, H., Jap. J. Med. Sci. Biol. 24, 53 


10. Oguma, K., lida, H.. and Inoue, K., Jap. J. Micro- 
biol. 17,425 (1973). 

11. Oguma, K., lida, H., and Inoue, K., Jap. J. Micro- 
biol. 19, 167(1975). 

12. Oguma, K., lida, H., Shiozaki, M., and Inoue, K.. 
Infect. Iraraun. 13, 855 (1976) 

Received February 17, 1978. P.S.E.B.M. 1978. Vol. 159. 


ession of Chemical (DEN) Carcinogenesis in SWR/J Mice by Goat Antibodies 
Against Endogenous Murine Leukemia Viruses^ (40286) 


on Laboratory, Bar Harbor, Maine 04609 and The National Cancer Institute,^ Bethesda, Maryland 20014 

ition of spontaneous leukemia in 
ce has been successful following pas- 
lunization with IgGs raised against 
ous ecotropic murine leukemia vi- 
uLVs), specifically the radiation leu- 
irus (RadLV) (1, 2). Also, passive 
y against 3-methylcholanthrene-in- 
rcomas in weanling C3H/f mice was 

using anti-RadLV IgG. Price, et al. 
ive shown that Fischer rat embryo 
5S required preinfection with MuLVs 
to be transformed by chemical car- 
, and that the requirement of viral 

and replication could be fulfilled by 
tropic (RLV) and xenotropic (AT 124 
•31) MuLVs; inhibition of viral rep- 
3y specific antiviral antibodies effec- 
ocked cell transformation (5). Thus 
-s that the expression of endogenous 
is a major determining factor in the 
reactions following carcinogen treat- 
th in vivo and in vitro. 
ircinogenic effect of nitrosamines in 
lice is well documented (6-8). Dieth- 
mine (DEN) treatment of SWR/J 
ulted in a high incidence of lung 
IS and adenocarcinomas (72% versus 
untreated controls) 29 weeks after 
it (9). Since endogenous MuLV 
)n in inbred strains of mice generally 
i with age (10) and upon chemical 
en treatment (1 1), we decided to test 
I-SWR/J system as a model for de- 
g the involvement, if any, of endog- 
uLVs in chemical carcinogenesis in 

J mice lack both infectious ecotropic 
lotropic MuLVs but express the 
ecific antigen (p30) in both spleens 

udy was supported by Coniracl No. NOl CP 
in the Virus Cancer Program of the National 
;titute. The Jackson Laboratory is fully ac- 
^ the American Association of Laboratory 

and thymuses (10). In the following commu- 
nication we report that lung-tumorigenesis 
induced by diethylnitrosamine (DEN) in 
SWR/J mice is significantly delayed by treat- 
ments with antiviral antibodies against both 
RadLV and AT124. 

Materials and methods. Antiviral antibodies. 
Goat IgGs raised against RadLV (Pool #3 
NIH C5682) and AT124 (Pool #1 NIH 
C4928) were obtained from the Laboratory 
of RNA Tumor Viruses, NCI, Bethesda, MD 
20014. These IgG preparations had neutral- 
izing antibody titers of 1 :800- 1 : 1600 based on 
70-100% inhibition of 60-70 AKR-XC 
plaques or 50-60 MSV (AKR) foci on SC-1 
cells (12). 

Mice and treatments. Twenty 8-week-old 
female SWR/J mice were pre- and post- 
treated with each goat anti RadLV and goat 
anti-AT124 IgG, and DEN according to the 
following schedule: 


0.1 ml anti- 

0. 1 ml anti 


ATI 24 IgG 

RadLV IgG 

Day 4 

0. 1 ml anti- 

0.15 ml anti 

AT124 IgG 

RadLV IgG 


0. 1 ml anti- 

0.15 ml anti 

(10 AM) 

AT124 IgG 

RadLV IgG 

Day 7 (5 PM) 

DEN (90 mg/kg) 

DEN (90 mg/kg) 

Day 10 

0. 1 ml anti- 

0.2 ml anti 

AT124 IgG 

RadLV IgG 

Day 14 

0.1 ml anti- 

0.2 ml anti 

ATI 24 IgG 

RadLV IgG 

Two groups of twenty control mice each 
received only the DEN treatment. DEN was 
freshly prepared in trioctanoin (Eastman Ko- 
dak) at a concentration of 10 mg/ml on each 
day of treatment. For each intraperitoneal 
injection the dose was 90 mg/kg (9). 

The mice were set aside for tumor devel- 
opment and killed only when moribund. At 
necropsy lung nodules were counted and the 
lungs weighed to obtain a measure of tumor 
sizes. Lungs together with all other visceral 
organs were processed for histopathological 
evaluation according to standard procedures. 


0037-9727 /7%/ \59\-OW>S%Vi\ .^|^ 

Copyrighl© V91%b^ iXvcSocwV'j fox ^xvftT\ttv*twVB\«\cM5j wA>K«*i«cKt 
All rights reserved. 



While these experiments were under way, 
another group of 15 SWR/J mice received 
normal goat IgGs (Microbiological Associ- 
ates, Bethesda) and DEN as per schedule 
used for anti-RadLV IgGs. 

Statistical evaluation of data. All graphics 
and statistical analyses were done on a Tek- 
tronix microcomputer (Model 4051). Means, 
standard errors (S.E.) and analysis of vari- 
ance were done according to Winner (13). F- 
tests were performed with 95% confidence 
intervals. The observed latency periods of 
AT 124- and RadLV-IgGs treated groups of 
mice were compared with the corresponding 
untreated controls. The data on the normal 
goat IgG-treated group of mice was com- 
pared separately with ail the other groups. 

Results and discussion, DEN treated SWR/ 
J mice passively immunized with goat anti- 
ATI 24 IgG survived up to 60 weeks following 
carcinogen treatment. Fifteen of 20 treated 
mice died from histologically confirmed lung 
tumors. In this group, 50% mortality because 
of lung tumors occurred at 54 weeks post- 
treatment. Thirteen of 20 untreated control 
mice died with multiple lung tumors and 32 
weeks after treatment and the 50% mortality 
occurred by the 29th week (Fig. 1); seven 
mice had pneumonia. 

Anti RadLV IgG inmiunized mice sur- 
vived up to 64 weeks after DEN treatment. 
Eighteen of 20 mice developed histologically 
confirmed lung tumors by 64 weeks, with the 



' 70 

ATI24 I9G treated ( 20 mice) 
Control untreated ( 20 mice) 

10 15 20 25 30 35 40 45 50 55 60 
Weeks after DEN treatment 

Fig. I. Eight week-old SWR/J mice were pre- and 
posttreated with anti AT 124 goat IgG and DEN as per 
schedule (see Materials and Methods). The mice were 
sacrificed when moribund and lungs and other viscera 
processed for histologic evaluation. Anti AT 124 IgG 

treated mice (O O) exhibited statistically significant 

(Table 1) prolongation of their survival times when com- 
pared to untreated controls (• •). 

50% mortality incidence from lung i 
occurring at 58 weeks. Control mice 1 
group survived only to 38 weeks with 
(12/20) lung tumor incidence (Fig. 2) 
1). Five mice in this group died prem; 
by a water bottle accident and the rem 
three suffered from pneumonia. Sin 
three mice receiving normal goat serui 
from injection accidents. 
Eight of 12 normal goat IgG treate 


^ 90- 
I 80- 
f 70 
£ 60 


Z 30- 

e 20- 

"^ .0- 

■ RoOLV IgG treoied (20 mice) 
• Corttroi untreated (20 mice) 

5 10 15 20 25 30 30 40 45 50 55 60 
Weeks after DEN treatment 

Fig. 2. Eight week-old SWR/J mice were gi^ 
tiple doses of anti RadLV goat IgG and D 
Materials and Methods for details). Control i 
ceived only DEN treatments. Anti RadLV IgC 

mice (O O) showed a significantly prolongc 

latency period (Table I) when compared to u 
control mice (•—•). 

TABLE L Effects of Antiviral Antibod 
Lung Tumorigenesis in Den Treated SWR; 


Mean sur- 


vival period 



in weeks 






29.8 (±1.83) 



52.0 (±1.70) 

goat IgG 



35.7 (±1.57) 



57.33 (±1.28) 

goat IgG 

Normal goat 


34.87 (±2.23) 


" Procedure for immunization is given in i 
and methods. 

^ Standard error in parenthesis. 

" F-tests were done according to Winner (10 

'^ Fa critical value. 

' F value in comparison with untreated cont 

^ Level of significance <0.0001 when com 
untreated controls. 

^Normal IgG treated group when compa 
other groups; the F-values for the different groi 
38.46 (AT 124 IgG), 3.15 (untreated control 
(RadLV IgG) and 0.047 (untreated controls). 



Oped lung tumors by 40 weeks after 
treatment. The mean survival period 
ignificantly different from antiviral IgG 
id groups. 

e distribution pattern of lung weights of 
reated and untreated mice is depicted in 
). Although immunized mice died with 
tumors after a long latency period, they 
larger lung tumors (Fig. 3); obviously, 
tumors had more time to grow than in 
ol mice which died early. All lung tu- 
were either alveolar adenomas or ade- 
-cinomas as described previously (9). 
indeed the presence of endogenous 
Vs is required for in vivo cell transfor- 
m by chemical carcinogens virus sup- 
ion should either prevent or prolong the 
ss of chemical carcinogenesis. The pres- 
ata clearly indicate a very significant 
ngation of the survival period of DEN 
id mice probably because of a slowed 
r growth in the antiviral IgG treated 
)s (Fig. 4). Antiviral IgG treated mice 
/ed to an expected average life-span of 
le SWR/J mice (9). Normal goat IgG 
lo beneficial effect on the survival time 
EN-treated SWR/J mice, 
e data on lung tumor weights suggest 
the antiviral IgGs had apparently very 

I ■ Normol Gool I9G 
2* ATI 24 I9G 

o<. Anti-ATl24 IqO treoled mica ( solid curve ) 
I. Anti RodLV IqG treated mice (dotted curve) 
• .ijntreoted Control mice (Mlid line) 




20 25 30 35 40 45 SO 55 60 65 
Weeks after DEN treatment 

i. 3. Eight week-old SWR/J mice were given anti 
i or anti RadLV IgG (see Materials and Methods 
;ails) and DEN. Control mice received only DEN 
ents. Wet lung weights of these mice were obtained 
x>psy. Antiviral IgG treated mice developed larger 
lunors after a significantly prolonged latency pe- 
rhe lung-weight distribution curves for AT 124 
curve), RadLV (dotted curve) and untreated con- 
solid Une) were obtained and plotted using a 
c computer (Tektronix-4051). 



r 40- 


3- RodLV I9G 

4 • Untreoted Control for ATI 24 

5<Unireoicd Control for RodLV 

Fig. 4. Effect of antiviral IgGs on the latency period 
of DEN-induced lung tumors in SWR/J mice. The mean 
survival times after carcinogen was plotted against each 
treatment group. The bars (i-#H) represent limits of 
variation. Both AT 124- and RadLV IgG treated groups 
reveal a longer tumor latency period when compared to 
the control groups. 

little effect, if any, on the growth rate of the 
tumors. Obviously by the time tumors ap- 
peared in most mice, the heterologous goat 
IgGs had long been eliminated. 

Goat antiserum against MuLV gp7 1 as well 
as FeLV was shown to prevent oncornavirus 
induced sarcomas in cats (14). Thus it seems 
that the effectiveness of anti AT 124 as well 
as RadLV IgGs against DEN carcinogenesis 
in SWR/J mice might be explained by the 
major homology between the two classes of 
MuLVs. Although the mechanism of the ob- 
served suppression of carcinogenesis is not 
clearly established, the data presented here 
tend to definitely indicate a viral involvement 
in chemical carcinogenesis. Presumably the 
antiviral IgGs partly suppress and delay the 
manifestations of chemical lung carcinogen- 
esis in SWR/J mice. 

Summary. We pre- and posttreated SWR/ 
J mice given 90 mg/kg of DEN with goat 
anti RadLV and AT 124 IgGs and studied 
their effects on the induction and latency of 
lung tumors. The results of these experiments 
tend to indicate a role of MuLVs in the 
etiology of the chemically induced lung tu- 
mors of SWR/J mice. The rate of tumor 
occurrence was greatly reduced in IgG 
treated mice and their survival time was sig- 
nificantly prolonged over nontreated mice. 
These findings require consideration of both 
ecotropic and xenotropic virus classes or their 
structural protein as cofactors in the chemi- 



cally induced lung carcinogenesis process. 
Similar conclusions were drawn previously 
by others in an in vitro chemical transforma- 
tion system (5). 

1. Huebner, R. J., in '* Analytical and Experimental 
Epidemiology of Cancer'' (W. Nakahara, T. Hira- 
yama, K. Nishioka, and H. Sugano, eds.). p. 345. 
University of Tokyo Press, Tokyo (1973). 

2. Huebner, R. J., Proc. Vih International Conference 
on Cancer, Perugia, Italy, in press (1978). 

3. Price, P. J., Suk, W. A., and Freeman, A. E., Science 
177, 1003(1972). 

4. Price, P. J., Bellow, T. M., King, M. P., Freeman, A. 
E., Gilden, R. V., and Huebner, R. J., Proc. Nat. 
Acad. Sci. U.S.A. 73, 152 (1976). 

5. Price, P. J., Suk, W. A., Peters, R. L., Gilden, R. V., 
and Huebner, R. J., Proc. Nat. Acad. Sci. U.S.A. 74, 

6. Issenberg, P., Fed. Proc. 35, 1322 (1976). 

7. Magee, P. N., and Barnes, J. M., Advan. Cai 

8. Rogers, A. E., Sanchez, O., Feinsod, F. 
Newbeme, P. M., Cancer Res. 34, 96 (1974 

9. Diwan, B. A., and Meier, H., Cancer Lettc 

10. Myers, D. D., Meier, H., and Huebner, R 
Sci. 9, 1071(1970). 

11. Meier, H., and Myers, D. D., Haematol. 

12. Huebner, R. J., GUden, R. V., Toni, R., H 
and Trimmer, R. W., in Proc. 3rd Int. S 
"Detection and Prevention of Cancer," p. A 
eel Dekker, New York (1976). 

13. Winner, B. J., in "Statistical Principles in 
mental Design", p. 47, McGraw Hill, N( 

14. Noronha, F. de, Baggs, R., Schafer, W.. an 
nesi, D. P., Nature (London) 267, 54 (1977; 

Received January 5, 1978. P.S.E.B.M. 1978. Vc 


(in Induced Metabolic Alterations in BCG Infected (Hyperreactive) Mice (40286) 


ment of Immunology and Medical Microbiology, College of Medicine, University of Florida, Gainesville, 

Florida 32610 

oxin given to laboratory animals 
loss of glucose homeostasis which is 
I by hypoglycemia and death (1, 2). 
a accumulated to date suggest that 
Dr factor responsible for hypoglyce- 
Tailure of gluconeogenesis rather than 
e glucose consumption, although the 
IS been reported (3). Endotoxin in- 
lypoglycemia has been most exten- 
Lidied in mice by Berry and coworkers 
hese workers found that a key liver 
phosphoenolpyruvate carboxyki- 
iPCK) which regulates gluconeogen- 
3 longer susceptible to glucocorticoid 
n after endotoxin poisoning (4, 6, 7). 
ive stressed the importance of this 
in endotoxin hypoglycemia presum- 
cause this enzyme has a relatively 
ilf life, i.e. 2 hr in rats while other 
3genic enzymes have longer half lives 
he published data indicate that three 
Dgenic enzymes, glucose-6-phospha- 
ctose-l,6-diphosphatase and PEPCK 
at normal or elevated activities in 
om endotoxin poisoned mice until 
I hr. Thereafter, the activities fall (8, 
drop in enzyme activity corresponds 
; development of hypoglycemia. A 
cenario for endotoxin induced hypo- 
i based on these observations is as 
Mice given endotoxin fail to eat, and 
e, do not assimilate exogenous car- 
itcs. Gluconeogenic enzymes are not 
by endogenous steroids elaborated 
ase to stress. As their level falls during 
:umover, particularly that of PEPCK, 
Dgenesis fails and hypoglycemia re- 
infected with Mycobacterium bovis 
come hyperreactive to endotoxin and 
sd approximately 1/ 1000th of the 
^50 (10). These mice have a remark- 
breviated clinical course with aug- 
clinical manifestations. One tenth of 
^am of endotoxin in BCG infected 

mice often causes profound hypoglycemia in 
2 hr, and death with convulsions frequently 
occurs within 4 hr (11). This is unlike the 
response of normal mice which die after 
17-24 hr and in which the hypoglycemia 
occurs later and is less severe. The response 
of the BCG mouse is, therefore, a caricature 
of that of the normal mouse. 

The exaggerated responses and shortened 
time course of BCG mice provide a suitable 
model to study metabolic abnormalities 
caused by endotoxin. We studied the BCG 
mouse model previously and found that 
endotoxin induced hypoglycemia is largely 
due to defective gluconeogenesis (2). Where 
in the gluconeogenic pathway the defect lies 
is unknown. In addition, the rapidity with 
which profound hypoglycemia occurs in 
BCG mice given endotoxin (2 hr vs 17 hr for 
normal mice) suggests that failure of enzyme 
induction and normal enzyme turnover may 
not account for this abnormality. 

In this paper we report experiments de- 
signed to determine if the hypoglycemia in 
endotoxin poisoned BCG mice is due to a 
selective defect in the gluconeogenic pathway 
or if there is a general perturbation of the 
pathway. The studies were performed be- 
tween one and 2 hr after endotoxin (prior to 
profound hypoglycemia) to avoid the poten- 
tial secondary effects of hypoglycemia and 
shock. Additional studies were performed to 
determine the effect of glucocorticoid and 
stimulation of gluconeogenesis by fasting on 
endotoxin hypoglycemia and mortahty. 

Materials and methods. Animals. Pathogen 
free, CD-I female mice weighing 20-25 g 
were obtained from Charles River Breeding 
Laboratories, North Wilmington, Massachu- 
setts. They were fed and watered ad libitum 
and housed in air conditioned quarters fully 
accredited by the American Association of 
Laboratory Animal Care. Unless otherwise 
indicated, all animals were fasted 18-24 hr 
prior to experimentation. The mice were ren- 



Copyright © 197» b^ lYvc SocitV^ tox tx^T\ttv«oMW\<J«o^ MA>Kt^vc«* 
All rights reserved. 



dered hyperreactive to endotoxin by a sys- 
temic infection with Mycobacterium bovis 
BCG given intravenously 13-16 days prior to 
use according to the method of Suter and 
Kirsanow (10). 0.2 ml of a 10-14 day culture 
of BCG in Dubos Liquid Broth (BBL) was 
injected via tail vein into unanesthetized, re- 
strained mice. 

Endotoxin, The endotoxin was prepared 
from a smooth strain of Salmonella typhimu- 
rium. The bacteria were grown in glucose 
minimal salts medium (M-9) supplemented 
with 0.1% Casamino Acids (Difco). At the 
stationary phase of growth they were killed 
with 0.2% formaUn, harvested, and extracted 
by the phenol water procedure of Westphal 
et al (12). Endotoxin challenge was by the 
intravenous route in 0.2 ml saUne. 

Metabolic studies. Glucose determinations 
were performed using the "Glucostat" 
(Worthington Biochemicals) micromethod. A 
20 jLil sample of blood obtained from the 
retroorbital plexus was added to 1.0 ml dis- 
tilled water and deproteinized with 0.5 ml 
1.8% Ba(OH)2-8H20 and 0.5 ml 2.0% 
ZnS04-7H20 solutions. One ml of the result- 
ing supernatant fluid was added to one ml 
"Glucostat" reagent at room temperature. A 
standard curve was prepared for each series 
of reactions. 

Glucose production in vivo was estimated 
by the net increase in blood glucose twenty 
minutes following an intraperitoneal injec- 
tion of 100 jLtM glycerol or fructose. Endo- 
toxin was given intravenously one hour be- 
fore glycerol or fructose. In one experiment, 
the incorporation of '"^C into glucose from ^^C 
glycerol (3 jnCi in 100 [iM) was determined 
by measuring the cpm/mg glucose in blood 
obtained via cardiac puncture. The glucose 
was separated from 1.0 ml whole blood by 
passage through mixed bed resin columns as 
described by Corridor et al. (13). The effluent 
was qualitatively checked chromatographi- 
cally to insure that the radiolabel resided with 
the glucose. 

Substrate oxidation in vivo in mice was 
measured by methods previously described 
(2). The mice were adapted to a gas train in 
such a way that all expired air was bubbled 
through 5 ml NCS'"' (Nuclear Chicago Corp.) 
to collect COo. Aliquots (0.5 ml) were re- 
moved at 15-min intervals and counted in a 

Packard liquid scintillation spectrom 
determine the activity of CO2. 
amounts of l-[^^C]glycerol, 6-[^^C]g 
1-^^C glucose, or l-["C]pahnitate (Ne 
land Nuclear) were injected intravenc 
control BCG infected mice and in B< 
fected mice one hour after 1.0 /ig end 
The isotopes (specific activities 4.6- 
per mM) were injected in 0.2 cc saline 

The free fatty acid concentration 
sera of inividual mice was measured 
metrically at 440 nm and compare 
similarly treated standards of palmil 
dissolved in chloroform. The free fatt 
were extracted from the sera by mix 
ml in 2.0 ml 0.2 M phosphate buffer (] 
and 6.0 ml chloroform. The mixtu 
shaken 2 min and after settling 15 r 
upper layer was removed by aspiratic 
chloroform layer was filtered into clea 
roform rinsed glass stoppered tubes tc 
3.0 ml Cu-triethanolamine reagent wa 
and mixed. The color was developed 
addition of two drops of sodium dieth 
ocarbamate reagent before reading th< 

Results, Previous experiments have 
that glucose production from pyruvat 
creased in BCG infected mice as earl) 
to 2 hr after endotoxin challenge. T 
coneogenic pathway from pyruvate to 
involves all of the key gluconeogenic e 
including phosphenolpyruvate car 
nase. The pathways from fructose an 
erol do not. Therefore, if the endotc 
duced defect in gluconeogenesis is th 
of the loss of a specific enzyme at the 
ning of the pathway one would exp 
glucose production from glycerol an 
tose might be unaffected. The resu 
sented in Tables I and II, however, i 
that the metabolic lesion is not limite< 
loss of phosphoenolpyruvate carbox) 
since glucose production in vivo was i 
tained from glycerol or fructose. The ii 
of exogenous fructose and glycerol c 
blood glucose in the control mice, but 
prevent a decrease in blood glucos< 
given to BCG infected mice one hoi 
endotoxin. In a similar experiment, 1 
beled ^^C glycerol was used to insure 
actual decrease occurred in the incorp 



TABLE 1. The Effect of Exogenous Fructose on 
Blood Glucose Concentration in BCG Infected 
Mice Before and After Endotoxin. 

Mean blood glucose, mg 
percent ± SE 



after fruc- 

BCG infected 



BCG infected mice 1 hr 
ifter 1.0 jug endotoxin 

91 ±3 


109 ±2 

65 ±3 

' 100 plM fructose injected ip. ( ) Indicates the num- 
ber of mice per group. 

TABLE II. Glucose Production from Glycerol in 
BCG Infected Mice Before and After Endotoxin. 

Mean blood 


mg percent ± SE 












BCG infected 


ooatrol mice 


142 ±5 



Exp 2 


158 ±9 

9.6 ± 1 



BCG infected 


mice 1 hr after 

76 ±10 

56 ±5 

l/ig endotoxin 

Exp 2 

91 ± 12 

66± 18 

3.6 ± 1 




* ( ) Indicates the number of mice per group. 

of ^^C label into blood glucose. The results 
shown in Table II indicate that the incorpo- 
ration of ^^C into blood glucose in the exper- 
imental group was only about one-third the 
incorporation which occurred in the control 
mice not given endotoxin. 

Because an increased oxidation of glycerol 
might account for its decreased incorporation 
into glucose, the in vivo oxidation of glycerol 
was measured between one and 2 hr after 
endotoxin by collecting expired ^^C02 after 
^*C glycerol injection. Figure 1 shows the 
cumulative counts per minute of ^^C02 col- 
lected from a group of BCG infected mice 
and a group of BCG mice given endotoxin. 

The results show decreased glycerol oxidation 
after endotoxin. 

Since an increased oxidation of glucose 
could result in an apparent decrease in incor- 
poration of ^^C into blood glucose by its loss 
as expired ^^C02, the in vivo oxidation of 1- 
[^^C]glucose and 6-[^^C]glucose was measured 
in mice after endotoxin. The results presented 
in Fig. 2 show that endotoxin caused de- 
creased oxidation of both l-[^^C]glucose or 6- 
[^^C]glucose. Oxidation of the 6-["C]glucose 
was depressed more than that of l-[ C]glu- 

Endotoxin LDso's were determined in 
fasted and fed BCG infected mice to deter- 
mine if the fasting state, which enhances glu- 
coneogenesis via endogenous steroids, influ- 
enced survival after endotoxin challenge. 
Fasted mice would have a stimulated gluco- 
neogenic pathway and little stored carbohy- 
drate while fed mice would have less active 
gluconeogenesis and much stored carbohy- 
drate. The results presented in Table III show 
no difference in the responses of fasted and 
fed mice. 

The effect of treatment of mice pre- and 
postchallenge with pharmacologic doses of 




15 30 45 

TIME, Minutes 

Fkj. I. Glycerol oxidation in vivo in BCG infected 
mice before and after endotoxin challenge. The endo- 
toxin treated mice were given one /ig endotoxin iv one 
hour prior to the injection of |'^C|glycerol. Each point 
represents the mean cumulative counts per minute of 
expired '"^CO.' from five individual mice. Vertical bars 
indicate ± SD. 




i GlJc Glucose 
^^ 1 ?C Glucofte 


i^ I l!5c Glucose 

yy / 

^ V 

yy /^ 


^ /-^ ^.i 6i5c Glucose 

/I ' / 



1/ f' 

^ 3- 


/// ^^ 


JJ / / # o Control 

I ^• 



If J / A A Endotoxin 

l/r / 


// f f 







TIME. Minutes 

Fig. 2. Glucose oxidation in vivo in BCG infected 
mice before and after endotoxin challenge. The endo- 
toxin treated mice were given one /ig endotoxin iv 1 hr 
prior to the injection of ['*C]glucose. Each point repre- 
sents the mean cumulative counts per minute of expired 
'^C02 from five individual mice. The vertical bars indi- 
cate ± SD. 

TABLE III. Endotoxin LDm's in Fed and Fasted 
BCG Infected Mice. 













































" LDw s were obtained by the method of Reed and 

hydrocortisone was studied to evaluate pro- 
tection against endotoxin in BCG mice. The 
results presented in Fig. 3 show blood glucose 
concentrations and mortality after 1.0 fig 
endotoxin when 3 mg hydrocortisone (Solu 
Cortef, Upjohn) was administered either be- 
fore or after endotoxin. Cortisone reduced 
mortahty significantly only in the group given 
cortisone prior to endotoxin. However, the 
rate at which blood glucose fell was dimin- 
ished quickly i.e. within 2 hr, in all groups 

receiving steroids. In a similar experiment 
the mice were challenged with less endotoxin 
(0. 1 fig). The results were similar except that 
mortality was also reduced in the group given 
cortisone 30 min after endotoxin. These ex- 
periments show that cortisone given before or 
after endotoxin challenge rapidly lessens the 
rate at which blood glucose falls and, de- 
pending on the timing, prolongs survival or 
prevents death. 

Fatty acid oxidation is important in pro- 
viding energy and reducing equivalents to 
drive the gluconeogenic pathway. Therefore, 
palmitate oxidation was measured in BCG 
infected mice and in similar mice between 1 
and 2 hr after endotoxin challenge. The re- 
sults shown in Fig. 4 show that the in vivo 
oxidation of palmitate was reduced about 
50% in the endotoxin poisoned mice. This 
apparent reduction in palmitate oxidation, 
however may be due to an in vivo pool size 
difference. Table IV, showing the serum free 
fatty acid levels in BCG infected mice before 
and after endotoxin, indicates that endotoxin 
caused a 77% increase in circulating free fatty 
acids. The effective specific activity of the 
injected isotope would therefore be decreased 
in mice given endotoxin, and this could ac- 
count for decreased ^^C02 evolution even 
though the rate of fatty acid oxidation is 


8 60 



Ml 09V» 



4 h«Mrt 









Fi(i. 3. Blood glucose concentration and mortality in 
BCG infected mice given I.O /xg endotoxin (A) and 3.0 
mg hydrocortisone 30 min before endotoxin (B), and 30 
min and 1 hr after endotoxin (C and D). Each point 
represents the mean blood glucose of 10 mice. 




TIME, Minutes 
. 4. Palmitate oxidation in vivo in BCG infected 
efore and aAer endotoxin challenge. The endo- 
reated mice were given one /ig endotoxin iv one 
rior to the injection of l-['*C]palmitate. Each point 
nts the mean cumulative counts per minute of 
I '^COj from five individual mice. The vertical 
dicate ± SD. 

inged. On the other hand, a similar 
nation for the reduction in glucose oxi- 
1 is untenable since the concentration of 
. glucose is less in the endotoxin poi- 
1 BCG infected mice than in control 
not given endotoxin. 
cussion. The data in this paper show 
ndotoxin causes a general derangement 
: gluconeognic pathway in BCG-mice. 
x)isoned animals were unable to make 
sc efficiently from glycerol or fructose, 
ously we showed that glucose produc- 
from pyruvate was also impaired (2). 
these substrates enter the gluconeogenic 
vray at different levels, a single lesion is 
ely to be responsible for the abnormal- 
n addition, the data also show that the 
rent decreased incorporation of the sub- 
:s into glucose could not be caused by 
crated catabolism. The rates of oxidation 
th glucose and glycerol were diminished, 
rticosteroids in pharmacological doses 
ctcd BCG-mice from the lethal effect of 
toxin when preadministered. Even when 
. as long as 30 min after endotoxin, 
ids exerted a rapid sparing effect on 
1 glucose. The rapidity with which this 
ng effect occurred i.e. within 2 hr, raises 
estion as to whether the effect was by 
icorticoid induced production of gluco^ 

neogenic enzymes. Increased enzyme produc- 
tion in response to glucocorticoids is a rela- 
tively slow process. The rise in enzyme is slow 
and usually preceeded by a lag of 2-3 hr (16, 
17). It seems more likely in this setting that 
the steroid was preventing some of the toxic 
effects of endotoxin and thereby lessening 
hypoglycemia or, alternatively, was activating 
gluconeogenic enzymes. This argument is 
also supported by the observation that pro- 
tection against endotoxin requires pharma- 
cological doses of corticosteroids while only 
physiological doses are sufficient for enzyme 

The study of fasted and fed mice given 
endotoxin also raises questions about the fail- 
ure of enzyme induction by corticosteroids as 
a cause of hypoglycemia in BCG mice. Fasted 
animals have an active gluconeogenic path- 
way with elevated levels of gluconeogenic 
enzymes (7), while fed animals have high 
stores of carbohydrate but low gluconeogenic 
activity. When challenged with endotoxin fed 
animals rapidly deplete their glycogen stores 
and then have to depend on their low gluco- 
neogenic activity. One might think that the 
fasted animal with high gluconeogenic activ- 
ity might have the advantage in survival. 
However, in spite of this increase in gluco- 
neogenesis, the outcome is the same. Stimu- 
lation of gluconeogenesis by endogenous 
physiological amounts of glucocorticoid, 
therefore, offers no protection. 

The data also suggest that abnormal sub- 
strate oxidation may also be partially respon- 
sible for endotoxin induced hypoglycemia. 
The o^^idation of fatty acids are required for 
the production of energy and reducing equiv- 
alents to drive the gluconeogenic pathway. 
Palmitate oxidation was diminished in BCG- 
mice given endotoxin. However, because of 
the increase in free fatty acids in the blood of 

TABLE IV. The Effect of Endotoxin on Serum 
Free Fatty Acid Level in BCG Infected Mice. 


Free Fatty Acids 
(/icq/ml serum) ±SE 

BCG infected control mice 

BCG infected mice 2 hr aAer 

.97 ± .03 

1.72 ± .03 

' ( ) Indicates the number of mice. 



BCG-mice after endotoxin a correction has 
to be made for fatty acid pool size. Fatty 
acids increased about 77%. Palmitate oxida- 
tion decreased by about the same amount. 
The conclusion is that there was no real 
change in fatty acid oxidation. However, dur- 
ing normal homeostasis a profound fall in 
blood glucose should result in an increase in 
fatty acid oxidation. The failure of fatty acid 
oxidation to increase suggests that loss of 
homeostatic regulation after endotoxin in- 
cludes lipid as well as carbohydrate metabo- 

Summary. The cause of hypoglycemia in- 
duced by endotoxin in BCG infected mice 
was investigated. The major abnormaUty, 
known to be defective gluconeogenesis, was 
studied to determine whether a specific point 
in the gluconeogenic pathway is involved or 
whether the derangement is more general. 
The inability of endotoxin poisoned mice to 
synthesize glucose from glycerol and fructose 
in addition to pyruvate indicated that the 
entire pathway was in disarray. The in vivo 
oxidation of glucose, glycerol and palmitate 
to CO2 was reduced, indicating that enhanced 
aerobic oxidation was not responsible for the 
hypoglycemia. This decrease in substrate ox- 
idation also suggests that impaired glucone- 
ogenesis may be due to decreased energy 
available to maintain the gluconeogenic path- 
way. Pharmacologic doses of glucocorticoids 
were protective in endotoxin poisoned BCG 
infected mice. The rate of development of 
hypoglycemia was rapidly lessened, and mor- 
tality reduced. The data suggest that steroids 
confer protection by preventing or interfering 
with some of the toxic effects of endotoxin or 
perhaps by activating glyconeogenic en- 

zymes. It is unlikely that glucocorticoid me- 
diated enzyme induction plays an anti-endo- 
toxin role in this model. 

1. Berry, L. J., D. S. Smythe, and L. G. Young., J. Exp. 
Med. 110,389(1959). 

2. Shands, J. W. Jr., V. MiUer, H. Martin, and V. 
Senterfitt., J. Bacteriol. 98, 494 (1%9). 

3. Filkins, J. P., and B. J. Buchanan., Proc. Soc. Exp. 
Biol. Med. 155(2), 216 (1977). 

4. Shlascl, T. F., and L. J. Berry., J. Bacterid. 97, 1018 

5. Rippe, D. P., and L. J. Berry., Infect, and Immun. (», 

6. Berry, L. J., in ^'Microbial Toxins V" (S. Kadis, G. 
Weinbaum, and S. J. Ajl, eds.), p. 165 Academic 
Press, New York (1971). 

7. Berry, L. J., D. S. Smythe, and L. S. Colwell., J. 
Bacteriol. 92, 107 (1966). 

8. McCallum, R. E., and L. J. Berry., Infect. Immun. 

9. Elliott, L. P., and 1. S. Snyder., Proc. Soc. Exp. Biol. 
Med. 141,253(1972). 

10. Suter, E., and E. M. Kirsanow., Immunol. 4, 354 

11. Shands, J. W. Jr., V. MiUer, and H. Martin., Proc. 
Soc. Exp. Biol. Med. 130,413 (1%9). 

12. Westphal, O., O. Luderitz, and F. Bister., Z. Natur- 
forsch. 7b, 148(1952). 

13. Corredor, C, K. Brendel, and R. Bressler., Proc. 
Nat. Acad. Sci. U.S.A. 58, 1 199 (I%7). 

14. Itaya, K., and M. Ui., J. Lipid Res. d, 16 (1%5). 

15. Reed. L. J., and J. Muench., Amer. J. Hyg. 27, 493 

16. Levinc, R., and D. E. Haft., N.E.J.M. 283, 237 

17. Ashmorc, J., and G. Weber, in **Carbohydrate Me- 
tabolism and Its Disorders'*, (F. Dickens, P. J. Ran- 
dle, and E. J. Whelan, eds.), p. 335 Academic Press, 
New York (1968). 

Received March 29, 1978. P.S.E.B.M 1978, Vol. 159. 


j Granulocyte Mobility Induced by Chemotactic Factor in the Agarose Plate 



Department of Pediatrics, Hokkaido University School of Medicine, Sapporo, Japan 

. Keller et al. proposed a definition 
iated to the locomotion of leuco- 
>ther cells (1). In their proposal, 
ses to chemotactic (and/or che- 
factor are classified into two types 
is, namely, chemokinesis and 
. Although the definition of these 
if reactions is clear, it is not easy 
rate these reactions separately in 
; several kinds of methods, some 
duated chemokinesis and chemo- 
but it is not certain as yet whether 
pes of reactions can be recognized 
itially separated phenomena, 
tudy we analyzed chemokinesis 
from chemotaxis in the agarose 

r and methods. Leucocyte prepara- 
inized adult blood was mixed with 
volume of a 2% methyl cellulose 
akarai Chemicals, Japan) and was 
settle at room temperature for 30 
!UCocytes in the supernatant were 
by centrifugation at 2S0g for 10 
and the cells were washed in ice- 
solution. The mononuclear cells 
eparated from the leucocyte prep- 
a modification of the method of 
. The leucocyte suspensions in 
tion were layered on top of Ficoll 
)-Hypaque (Winthrop) solution 
% Ficoll + 10 parts 33.9% Hy- 
centrifuged at 2S0g for 45 min at 
•nonuclear cell layer from the top 
lent and the Ficoll-Hypaque so- 
gently aspirated and discarded, 
button was resuspended in 0.2 ml 
K)lution. Contaminating erythro- 
disrupted by hypotonic shock, 
ng twice with Hanks solution, the 
resuspended in Medium 199 
itaining 10% heat inactivated fetal 
at 1 X 10® cells per ml. For che- 
assays, chemotactic factor was 
e medium at a final concentration 

of 10%. The purified granulocytes contained 
less than 1% of other cells including platelets. 

Chemotactic factor preparation: Bacterial 
chemotactic factor was produced by over- 
night growth of Escherichia coli in heart in- 
fusion broth at 37°. The culture broth was 
passed through a 0.22 /im filter and the filtrate 
was then stored at —70° until use. 

Assay of leucocyte mobility. This was per- 
formed by a minor modification of Nelson's 
method (6). The agarose plate was prepared 
by placing 3 ml of 1% agarose (Behringwerke) 
solution in Medium 199 containing 10% heat 
inactivated fetal calf serum into 35 mm x 35 
mm Falcon plastic dishes. When chemoki- 
nesis, namely the enhancement of random 
mobility by chemotactic factor was assayed, 
the chemotactic factor was added uniformly 
to agarose. After the agarose gelled, 3 mm x 
3 mm wells were made by a stainless steel 
punch in the agarose plate, and 10 yX of cell 
suspensions and chemotactic factor were 
placed as shown in Fig. 1. Afler various 
periods of incubation in a 5% CO2 incubator 
at 37°, the distance the cells moved was mea- 
sured under an inverted microscope with an 
ocular grid. Four measurements were aver- 
aged from the margin of the well to the line 
of migration in four quadrants of each well. 
All experiments were carried out in duplicate 
or triplicate. 

Preparation of cells for morphological ex- 
amination. After incubation, the cells were 
fixed with agarose in place by flooding the 
plates with 4 ml of 10% formalin for 48 hours. 
After fixation, the agarose was gently re- 
moved, and the cells were stained by Giemsa 

Results. Random mobility and chemokinesis 
in the agarose plate. The random mobility 
and chemokinesis of normal adult granulo- 
cytes assayed by the agarose plate method are 
shown in Table I. Granulocytes stimulated 
by 10% £". co//-derived chemotactic factor 
added uniformly in agarose showed a 2.6 ± 


0037-9727/78/ 1 59 1 -007550 1 .00/0 






cell suspension 


Fig. I. Agarose plate method employed in the ex- 
periments. Three types of granulocyte mobility were 
measured in this method. The oblique lines means the 
presence of chemotactic factor derived from E. coli 

0.3 (Af ± SE) fold enhanced mobility com- 
pared to nonstimulated granulocytes. Micro- 
scopic appearance of the preparations stained 
by Giemsa solution is shown in Figs. 2 and 
3. Cells moving under the influence of chem- 
otactic factor tend to be more irregular in 
outline than those moving without the influ- 
ence of factor, and formation of blebs or 
pseudopodium-like structures can be ob- 
served. There is no regular orientation of cell 

Then we assayed the two types of granu- 
locyte mobility as a function of time (Fig. 4). 
Cells under the influence of chemotactic fac- 
tor moved rapidly up to 3 h, after which time 
no marked increase in distance was observed. 
On the other hand, in the absence of the 
bacterial factor movement of granulocytes 
increased linearly. However, even after 19 
hours these cells had not moved as far as 
those stimulated by chemotactic factor. 

Relationship between the concentrations of 
chemotactic factor and the degree of chemo- 

kinesis. As shown in Fig. 5, granuh 
bility increased linearly in proporti 
concentration of chemotactic fact 
more than 2.5% of chemotactic i 
stimulation of migration diminisl 
thermore with more than 10% of ch 
factor, granulocyte mobility tend< 
crease, although a significantly enh£ 
bility could be observed when < 
against random mobility. 

Granulocyte mobility under a con 
gradient of chemotactic factor Fir 
amined the influence of a negative c 
tion gradient. In order to form a 
concentration gradient of chemota< 
in the area surrounding the well, gr 
suspension in medium containing 1 
otactic factor was placed in an ag; 
which did not contain chemotactic 
shown in Table I, granulocyte mobi 
a negative concentration gradient ( 
tactic factor increased significan 
compared to mobility without facte 

Next we examined the influence 
tive concentration gradient toward 

TABLE 1. Granulocyte Mobility Und 
Conditions of Chemotactic Stimi 



No factor (random mobihty) 

22.9 ± 6. 

Negative gradient 

33.0 ± 7 

Uniform gradient (10% che- 

54.5 ± 5 


Positive gradient (chemotaxis) 

69.4 ± i: 

" Expressed as x 40 mm. 


Under uniform 

Under posit i\ 

Fk;. 2. Three types of granulocyte mobility after three hour culture. Cells were stained by Giems;) 
random mobility, B: chemokinesis, C: chemotaxis, chemotactic gradient was made at the leA side. 




5 = 

— 3 



2 3 

Fig. 4. The time-course of leucocyte mobility. Cells 
under the influence of 10% chemotactic factor (#) moved 
rapidly, whereas cells not under the influence of the 
factor (O) showed a gradual and linear increase. 

well in which cell suspensions were placed. 
Cells exposed to a chemotactic gradient ap- 
pear to extend further than those under any 
of the uniform concentrations tested (Table 
I and Fig. 5). This fact suggests that chemo- 
taxis is also occurring in this condition. 

Effect of preincubation with chemotactic fac- 
tor on granulocyte mobility. Table II shows 
the effect of preincubation of granulocytes 
with Medium- 199 containing 0, 2.5, or 10% 
of chemotactic factor at 37° for 1 hr, followed 
by washing with Hanks solution, and resus- 
pension in Medium- 199 containing heat in- 
activated 10% of fetal calf serum. The cells 
were then placed into the wells in the agarose 
plate with or without chemotactic factor. 
Preincubation with chemotactic factor of 
three concentrations had no influence on che- 
mokinesis observed in agarose containing 5% 
of chemotactic factor. Furthermore granulo- 
cytes preincubated with 10% of chemotactic 
factor did not show enhanced mobility, 
namely chemokinesis, when the granulocytes 
were placed into the wells in agarose not 
containing chemotactic factor. 

Discussion, Random mobility and chemo- 
kinesis could be observed separately by the 
agarose plate method. Granulocytes under a 

uniform concentration of E. co/i-derived 
chemotactic factor moved at a significantly 
higher rate than in the absence of factor. 
Morphologically differences were also appar- 
ent. Cells showing chemokinesis tend to be 
irregular in outline, whereas those showing 
random mobility tend to be rounded. 

Nelson et al and Cutler reported that the 
distance the cells moved toward the outer 
well in which chemotactic factor was placed, 
was determined by chemotaxis of granulo- 
cytes (6-8). But from the results obtained in 
our experiments, we conclude that this dis- 
tance may be based partially on chemokine- 

The time course of cell mobility triggered 
by chemokinesis is analogous to that in re- 
sponse to chemotaxis as reported by Nelson 
et al and Cutler (6, 7). Thus the distance of 
cells showing chemokinesis as well as those 
showing chemotaxis increases with the pas- 
sage of time. In our experiments, there was a 
dose response relationship between chemo- 

80 1 


25 5 10 25(%) 


Fig. 5. Relationship between the concentrations of 
chemotactic factor and the degree of chemokinesis. Leu- 
cocyte mobility increased linearly in proportion to the 
concentrations of cher.otactic factor under less than 2.5% 
of the factor. 

TABLE 11. Effect of Preincubation with Chemotactic Factor on Granulocyte Chemokinesis." 

Concentration of chemotactic factor with which cells were preincubated 




Mobility in agarose containing 

5% factor (M ± SD) 
Mobility in agarose containing no 

factor (M ± SD) 

50.9 ± 2.2 

22.9 ± 6.3 
rt = 9 

51.3 ±3.8 
n = 3 

53.4 ± 3.3 

22.3 ± 11.7 


' Expressed as x 40 mm. 



kinesis and the concentration of chemotactic 
factor, and the degree of chemokinesis was 
determined by the final concentration of 
chemotactic factor with which granulocytes 
came in contact. Thus cells coming in contact 
with higher concentration of chemotactic fac- 
tor move more rapidly at random, and can 
increase their chances of coming in contact 
with chemotactic factor of higher concentra- 
tion, because of the enhanced mobility. If this 
does not occur, namely when they move away 
from the chemotactic factor, the cells remain 
in the same location and may regain random 
movement regardless of whether they have 
been stimulated or not. Thus the number of 
granulocytes in the area of lower concentra- 
tion of chemotactic factor decreases, whereas 
the number in an area of higher concentration 
increases. This hypothesis may be supported 
by another finding in the present experi- 
ments. Some enhancement of mobility under 
a negative concentration gradient of chemo- 
tactic factor, which can not be explained by 
the concept of chemotaxis, suggests that che- 
mokinesis occurs also under a chemotactic 
gradient. As far as E. co//-derived chemotactic 
factor is concerned, besides a real chemotactic 
response ( 1-4, 9, 10), a chemokinetic response 
may also account for the effect of chemotactic 
gradient on trapping of cells at the site of 

The phenomenon of "deactivation" which 
was shown by Ward and co-workers' study 
concerning chemotaxis (11), could not be ob- 
served in chemokinesis induced by E. coli- 
derived chemotactic factor. It is uncertain 
whether "deactivation" can be observed also 
in chemokinesis induced by other chemotac- 
tic factor such as complement-derived factors. 
However, if such a phenomenon occurs in 
chemokinesis induced by some chemotactic 
(or chemokinetic) factor, the trapping of cells 
in an area where high concentration of chem- 
otactic factor is present may be performed 
more effectively. 

We believe that chemokinesis in addition 
to chemotaxis plays an important role in the 
defense mechanisms in vivo. Further investi- 
gation is required to better understand the 
basic mechanisms involved in the chemotac- 
tic (or chemokinetic) response of granulo- 

The authors gratefully acknowledge the 
helpful advice of Dr. Paul G. Quie. 

Summary. Human granulocyte mobility 
under various conditions of chemotactic stim- 
ulus was studied using the agarose plate 
method. Enhanced mobiUty was observed 
when granulocytes were incubated in the aga- 
rose plate containing chemotactic factor gen- 
erated from E. coli. A dose response type 
relationship was observed between the degree 
of enhanced mobility and the concentrations 
of chemotactic factor in a range of less than 
10%. The rate of mobility was rapid up to 3 
hr, afler which time it was very slow. Prein- 
cubation of granulocytes with chemotactic 
factor of various concentrations did not have 
any influence on granulocyte mobility as- 
sayed afler preincubation. The degree of mo- 
bility tends to be determined by the final 
concentration of chemotactic factor coming 
in contact with granulocytes. Thus granulo- 
cytes under a negative concentration gradient 
also showed an enhanced mobility. On the 
basis of these findings, we propose the hy- 
pothesis that the accumulation of granulo- 
cytes at the site of inflammation can be in 
part explained by chemokinesis, i.e. enhanced 
random mobility. 

1. KcUcr. H. U., Wilkinson, P. C, Abcrcrombic. M.. 
Becker, E. L., Hirsch, J. C, MiUer, M. E., Scott 
Ramsey, W., and Zigmond, S. H., Clin. Exp. Im- 
munol. 27, 377 (1977). 

2. Keller, H. U., and Sorkin, E., Immunology 10, 409 

3. Zigmond, S. H., and Hirsch, J. C, J. Exp. Med. 137, 

4. Anderson, R., Glover, A., and Rabson, A. R., J. 
Immunol. US, 1690(1977). 

5. Boyum, A., Scand. J. Lab. Clin. Invest. Suppl. 97, 

6. Nelson, R. D., Quie, P. C, anv^ Sinmions, R. L., J. 
Immunol. 115, 1650(1975). 

7. Cutler, Jim E., Proc. Soc. Exp. Biol. Med. 147, 471 

8. Nelson, R. D., Fiegel, V. D., and Simmons, R. L., J. 
Immunol. 117, 1676(1976). 

9. Zigmond, S. H., J. CeU. Biol. 75, 606 (1977). 

10. Malech, H. L., Root, R. K., and Gallin, J. 1., J. Cell 
Biol. 75,666(1977). 

11. Ward, P. A., and Becker, E. L., J. Exp. Med. 127, 

Received February 22, 1978. P.S.E.B.M. 1978, Vol. 159. 


Hypophysectomy Alters the Diurnal Food Intake Patterns in Rats (40288) 


Department of Physiology, Baylor College of Dentistry, Dallas, Texas 75246 and Department of Animal Scie 
University of California, Davis, California 95616 

Several hypothalamic nuclei have been 
shown to influence rhythmic physiological 
processes. Destruction of the suprachiasmatic 
(1), dorsomedial hypothalamic (DMN) (2) 
and the ventromedial hypothalamic nuclei 
(VMN) (3) alter the normal diurnal food 
intake pattern of rats. Ablation of these nuclei 
also disrupts the natural corticosterone 
rhythm of rats (4) (5). Recently several papers 
(4, 6-9) have concerned themselves with the 
possible relationship of pituitary hormone 
rhythms and feeding and drinking patterns. 

It has been known for some time that 
hypophysectomy decreases food consump- 
tion of rats. More recently Stephan and 
Zucker (10) reported that rats did not display 
a normal diurnal food intake following hy- 
pophysectomy combined with ovariectomy. 
They noted that the nocturnal rhythms in 
eating and drinking were greatly attenuated 
following hypophysectomy-ovariectomy. 
However, since they (10) measured the ani- 
mals* food intake only at the start and end of 
the light-dark cycle their data do not reveal 
the effect of hypophysectomy on meal pat- 

The present study investigated the individ- 
ual meal patterns of hypophysectomized rats 
in order to determine frequency, duration 
and distribution of the meals. 

Materials and methods. Male hypophysec- 
tomized and nonoperated Sprague-Dawley 
rats were purchased from Simonsen Labora- 
tories Inc., Gilroy, CA. The rats were housed 
individually under a lightidark (L:D) ratio of 
12:12 with lights on at 0600 hr. During the 
experiment the rats were given a purified diet 
consisting of: 15% vitamin-free casein, 0.3% 
L-methionine, 1.0% vitamin premix, 5% salt 
mixture, 49.05% com starch, 24.55% sucrose, 
5% com oil and 0.1% of a choline chloride 
solution. This diet was selected because a 
powered diet had previously given more re- 
producible results than a chow-type diet 
when measured by the automatic food intake 

recorder. The feeder was designed sc 
food could be readily obtained from a a 
fiber glass cup designed to prevent spi 
An event marker recorded each time 1( 
of food was removed from the cup 
Rogers and Leung (1 1) for additional : 

To prevent disturbance of the rats, en 
their room was restricted to 1730-18C 
The animals were allowed a 7 to 1< 
period to adjust to the room, purifiec 
and food intake monitoring apparatus 
to recording food intake patterns. At th 
of this period, which was approximate 
days after the animals were hypophyj 
mized, food intake patterns were reo 
continuously for 3 days. 

Any period of food intake in which 
was not more than a 20-min time laps 
tween recordings was defined as a 
Thus, the number of daily meals cou 
determined. For statistical analysis th< 
mals were considered hypophysectomi 
they did not show weight gains (weig 
arrival 108 ± 1.9 g and at end of study 
± 2.7 g) and after histological examin 
showed no pituitary renmants in the 
turcica. This yielded a population of 
hypophysectomized and seven control 

The data were analyzed using Studc 
test, Mann- Whitney U test and Chi-Sc 

Results. The controls ate 97.6 ± 1.09 
the hypophysectomized (hypox) rats 7 
5.5% of their food during the dark pha 
= 375; P < 0.001) (Fig. 1). When the 
number of meals consumed per 24 hr b) 
groups was compared no significant c 
ences were found (controls 7.9 ± 0.6; 1 
7.4 ± 0.8). However, the light:dark dis 
tion of the meals was significantly dif] 
(Light phase: controls 0.7 ± 0.2 vs hyp< 
± 0.2, P < 0.001; Dark phase: controls 
0.5 vs hypox 5.5 ± 0.7, P < 0.05). The av 
daily intake of the controls was higher 
that of the hypophysectomized rats (1 


0037-9727/78/ 1 59 1 -008050 1 .00/0 

^yright ® 1978 by the Society for Experimenul Biology and Medicine 
i||hu reserved. 



Control Animals 
11 ▲ A^_L^^ 

Hypophysectomized Animals 
L J ii_ ^. _L _ I 

1 _i_.__^__i_ l_l _ 

i ^ 


- I 




H - - 

8 - 


-. J L L 

L J_ 

^ ^ _ _*_ 


-1 - 







Fig. L Twenty four food intake profiles on a representative control animal and seven hypophysectomized rats. 
The animals* food intake patterns are shown for 3 consecutive days. The dark bar corresponds to the dark period. 

0.9 g vs 6.7 ± 1.0 g, P < 0.001). The differ- 
ences in food consumption between the two 
groups was probably associated with the fact 
that the controls continued to grow during 
the adjustment period while growth was ar- 
rested in the hypophysectomized animals. 
The data revealed that both groups ate ap- 
proximately the same number of meals each 
day. Thus, the hypophysectomized rats ap- 
peared to reduce their daily food intake 
through a reduction in average meal duration 
(controls 15.1 ± 1.1 min vs hypox 4.5 ± 1.0 
min, P < 0.001) and meal size (controls 2.7 
± 0.4 g/meal vs hypox 1.1 ±0.2 g/meal, P 

Inspection of the daily feeding patterns 
(Fig. 1) reveals a pronounced diurnal food 
intake rhythm in the control animals and an 
altered pattern in the hypophysectomized an- 
imals. While hypophysectomy has statisti- 
cally altered the animals normal diurnal eat- 
ing rhythm several of the rats (Fig. 1, rats 
#1, 2, S and 7) still appear to be strongly 
influenced by the photoperiod. The remain- 
ing three hypophysectomized rats showed a 
much greater altered eating rhythm. The na- 

ture of the difference waits to be resolved. A 
further indication that hypophysectomy has 
altered the rats normal feeding rhythm is 
shown in the day to day variation in the 
percentage of food consumed during the light 
and dark phases. The day to day variation 
were computed on each group of animals and 
then summed as to the period of feeding and 
analyzed. Control rats showed very little day 
to day variation in the percentage of food 
they consumed during the dark and light 
phase while the hypophysectomized rats 
showed a great deal of variation (Dark phase: 
controls, x"^ = 3.53, P > 0.99 hypox, x^ = 
100.55, P< 0.001). 

Discussion. Stephan and Zucker (10) 
showed that a combination of hypophysec- 
tomy and ovariectomy altered the normal 
diurnal food intake pattern of rats. The pres- 
ent study revealed that hypophysectomy 
alone can modify the normal diurnal feeding 
pattern of rats. 

Hypophysectomy arrests growth and de- 
presses the animals daily food consumption 
(12). The data presented here reveal that, 
while the 24-hr meal frequency of the hy- 



pophysectomized rats was similar to nono- 
perated controls, the light-dark distribution 
of meals was significantly altered by hypoph- 
ysectomy. Furthermore, total daily food in- 
take was reduced by hypophysectomy be- 
cause meal duration and size were greatly 
reduced in the hypophysectomized rats. 
Thus, hypophysectomized rats appear to de- 
crease their food intake through a reduction 
in meal size and not meal frequency. 

It has been proposed (12) that because 
hypophysectomy decreases daily food con- 
sumption, pituitary hormones might be di- 
rectly involved in the regulation of feeding 
behavior. This was challenged (13) on the 
grounds that hypophysectomy causes a drop 
in basal metabolic rate and the decrease in 
food consumption is responding to that de- 
creased basal metabolic need. However, the 
lower basal metabolism cannot readily ac- 
count for the fact that the hypophysectomized 
rats of Stephan and Zucker's study (10) or of 
the present one, displayed an altered diurnal 
feeding pattern. One possible explanation is 
that hypophysectomy removed one or more 
of the pituitary hormones which help deter- 
mine the food intake rhythm. In support of 
the pituitary playing a role in the mainte- 
nance of certain consummatory rhythms is 
the finding that the posterior pituitary hor- 
mone, antidiurectic hormone has been shown 
to be important in maintaining the light-dark 
distribution of drinking in rats (9). 

Rats eat the majority of their food at night, 
conmiencing shortly after the start of the dark 
phase. Increases in several pituitary hor- 
mones (6-8, 14) appear to coincide with the 
onset of normal feeding in rats (10, IS). Cor- 
ticosterone (14, IS) and prolactin (6, 7) peak 
prior to onset of the normal feeding period 
and if rats are fed for only 2 hr per day, the 
natural rhythms of these hormones are mod- 
ified (6, 16, 17). Within a short time both 
hormones show peaks prior to the start of the 
new feeding period (6, 16, 17) with a reduced 
peak remaining at the end of the light period. 
Both the natural corticosterone and prolactin 
rhythms persist in fasted rats (7). Interest- 
ingly, when rats are refed, even after periods 
of fasting up to 36 hr (18, 19), they consist- 
ently consume more when presented with 
food during the dark phase than when refed 
during the light phase. Also noteworthy, is 

the finding that lesions of the DM 
VMN disrupt the natural diurnal feedi 
tern of mature and weanling rats (2, 3 
also altering the normal diurnal cort 
one rhythm (4). In the present study, h 
ysectomy, which would necessarily 
prolactin and alter corticosterone 
seemingly modifies the natural food 
patterns of the animals. However, a 
study of ours (Bellinger et a/., unpu 
observations) indicates that adrenal 
does not alter the diurnal feeding rh) 
rats. This indicates that the cortico 
rhythm is not the cause but only incid< 
the rats feeding rhythm. Thus som< 
pituitary factor(s) may be responsible i 
tial maintenance of the diurnal 

Since on the average the food consu 
of the hypophysectomized rats was stil 
enced by the photoperiod, it appears, 
pituitary hormone(s) or some other p 
factor(s) can only be paritally respons 
maintenance of the normal diurnal 
rhythm. Finally, it must be consider 
hypophysectomy does alter the anim£ 
tabolism and this might possibly moc 
food intake pattern of the animal. 

Summary. Hypophysectomy alters t 
mal diurnal feeding patterns of rats 
the average the hypophysectomized r 
consume the greatest percentage of th( 
during the dark phase. Compared to o 
hypophysectomized rats eat a similia 
ber of meals each day, however, the i 
of food consumed and the duration 
meals are reduced. The pituitary glanc 
to be one of the factors involved in sus 
the natural diurnal feeding rhythm 

The authors wish to thank Drs. Roger and I 
the use of the food intake monitors, and we ex 
appreciation for the expert technical assistance 
Friend, and to S. Turley for typing the manusc 

1. van den Pol A., 7th Annual Meeting Soc. 
roscience. Vol. HI, p. 516. (1977). 

2. Bemardis, L., Physiol Behav. 10, 855 (197: 

3. Kakolewski, J., Deaux, E., Christensen, J., s 
B., Amer. J. Physiol. 221(3), 711 (1971). 

4. Bellinger, L., Bemardis, L., and Mendel, 
rocndocrinology 22, 216 (1976). 

5. Moore, R., and Eichler, V., Brain Res. 42, 2( 

6. Bellinger, L., Moberg, C, and Mendel, V 



>1. Res. 7,43(1975). 

,er, L., Mendel, V., and Moberg, G., Horm. 
Res. 7, 132(1975). 

oto, Y., Arisue, K., and Yamamura, Y., Neu- 
0, F., and Zucker, 1., Neuroendocrinology 14, 
n, F., and Zucker, 1., Physiol, Behav. 8, 315 

. Q., and Uung, P., Fed. Proc. 32, 1709 

D., Ishibashi, T., and Turner, C, Proc. Soc. 

iol. Med. 119, 1238 (1965). 

gnen. J., in **Progrcss in Physiological Psy- 

chology" (E. Stellar and J. Spraguc, cds). Vol IV, p. 
203. Academic Press, New York (1971). 

14. Critchlow, V. in ** Advances in Neuroendocrinology** 
(A. Nalbandov, ed) p. 377. Univ. Illinois Press, 

15. Le Magnen, J., and Devos, M., Physiol. Behav. 5, 

16. Krieger, D., Endocrinology 95, 1195 (1974). 

17. Moberg, C, Bellinger, L., and Mendel, V., Neuroen- 
docrinology 19, 160(1975). 

18. Bare, J., J. Comp. Physiol. Psychol. 52, 129 (1959). 

19. BeUinger, L., and Mendel, V., Physiol. Behav. 14, 43 

Received March 3, 1978. P.S.E.B.M. 1978. Vol. 159. 


Protein-Calorie Malnutrition Impairs the Anti-Viral Function of Macrophages^ (40289) 

Department of Microbiology, Indiana University School of Medicine, Indianapolis, Indiana 46202 

Human malnutrition is accompanied by 
decreased resistance to at least certain infec- 
tious diseases (1). This enhanced susceptibil- 
ity may be mediated by such factors as altered 
complement metabolism (2), deficient cellu- 
lar immune responses (3, 4) and decreased 
production of secretory immunoglobulins (5). 
Little information is available as to whether 
macrophage antimicrobial mechanisms are 
also affected by malnutrition. As an impor- 
tant member of primary host defenses and as 
an effector cell for many of the cellular im- 
mune processes, the macrophage plays a crit- 
ical role in infectious diseases (6). Conse- 
quently any degree of impairment of the 
efliciency with which this cell performs these 
roles might be expected to result in consid- 
erable reduction in host resistance. 

Douglas and Schopfer (7) reported that 
monocyte phagocytic indices are not altered 
by severe protein-calorie malnutrition. Pas- 
swell et al (8), however, noted that phago- 
cytic capacities of macrophages do seem to 
be impaired in mice which were protein-de- 
prived. Nevertheless, neither of these reports 
described the total microbicidal capability of 
the macrophage. Keusch et al have recently 
reported (9) that macrophages from mice 
with kwashiorkor kill Staphylococcus aureus, 
E, coli and salmonella normally in vitro. 
These authors infer that in vivo however, it 
appears likely that macrophage contributions 
to host defenses are impaired. 

We have recently described a model in 
mice which demonstrates that age-specific 
resistance to ip infection with Wesselsbron 
virus (WBV) is macrophage-mediated (10). 
Resistance by mice is essentially complete by 
age 2-3 weeks, and represents the acquired 
ability of peritoneal macrophages to phago- 
cytose and to destroy infectious virus. The 

' Supported by a grant from the Rockefeller Foun- 

" Present address. The Children's Mercy Hospital, 
Kansas City, MO 64108. 

present report represents studies on the re- 
sistance of protein-calorie malnourished mice 
to WBV infection and whether macrophage 
antiviral function is concomitantly affected. 

Materials and methods. Mice. Random- 
bred 3-week old white mice were individually 
caged and allowed free access to water and 
food. Experimental mice were placed on pro- 
tein-depletion diet USP XV composed of 84% 
white dextrin, 9% com oil, 4% salt mixture, 
2% agar, 1% cod liver oil and vitamin supple- 
ment (ICN Nutritional Biochemicals). Con- 
trol mice were fed a normal protein diet 
containing 27% casein. Under these condi- 
tions normal mice showed a mean increase in 
body weight of 5.3% afler 5 days while pro- 
tein-depleted mice lost a mean of 12.9% body 
wt during the same period. Unless otherwise 
noted resistance to infection was determined 
in mice that had been fed protein-depletion 
diet for S days and continued on this diet 
during the observation period. 

Virus. The source and preparation of WBV 
was as previously published (10). Infectivity 
was assayed by inoculating 1-day old mice 
intracerebrally (i.e.); serial tenfold dilutions 
were made and each dilution was inoculated 
into one litter (9-14 mice). End-points were 
determined by summarizing mortality 14 
days later and calculated by the method of 
Reed and Muench (11). 

Results. Of 21 normal mice inoculated ip 
with WBV (10® LDso as assayed in suckling 
mice), none developed signs of illness nor 
died. In contrast, each of 1 1 protein-depleted 
mice developed symptoms of encephalitis; 
three of these mice were sacrificed on day 5 
and brain tissue was assayed for WBV while 
the other 8 mice succumbed within 7 days of 
inoculation. WBV, 10^^ LD50 per 0.1 g tissue 
was recovered from each of the brains tested. 
One of 12 protein-depleted mice observed as 
uninoculated controls died after S days but 
no virus could be recovered from its brain. 

To study how rapidly susceptibility 10 
WBV developed, a series of mice were inoc- 


0037-9727/78/ 1 59 1 -0084$0 1 .00/0 

'nyright © 1978 by the Society for Experimental Biology and Medicine 
'^ts reserved. 



i ip with WBV at various times before 
er switching them to the protein-deple- 
iiet. The details and results of this ex- 
tent are shown in Table I. Mice that 
inoculated with WBV 3 days or longer 
s initiating protein depletion showed 
al resistance to virus. However, suscep- 
y to ip infection with WBV rapidly de- 
ed in relation to protein depletion such 
;ven animals inoculated one day previ- 
showed decreased resistance to infec- 
Signs of encephalitis appeared 5-7 days 
inoculation in all mice succumbing to 

^se data implied that some event oc- 
1 relatively early in the initial stage of 
ion that was sensitive to protein deple- 
To support this idea it was necessary to 
nine how soon after ip inoculation 
could be detected in the central nervous 
1. Groups of mice on normal diet or 
> days on depletion diet were inoculated 
h l& LDso of WBV as before, and brain 
from two animals of each group was 
^d daily thereafter (Table II). In normal 
very small amounts of WBV were re- 
^ 1-3 days postinoculation, while sam- 
oUected on days 4-7 contained no de- 
le virus. Simultaneously collected sam- 
f blood obtained by section of the axil- 

lary vessels contained similar concentrations 
of virus and presumably represented the 
source of virus present in the brain samples. 
In contrast, much greater concentrations of 
WBV could be detected in blood from pro- 
tein-depleted mice. Brain samples also had 
concentrations of virus of similar magnitude 
on days 3 and 4, although the presence of 
viremia made the origin of this virus uncer- 
tain. By day 5, however, the titer of virus in 
the brain significantly exceeded that in the 
blood and by day 6 the animals had devel- 
oped encephalitis. 

Unstimulated peritoneal macrophages 
were collected from normal and from PCM 
mice and the susceptibility of WBV to inac- 
tivation by these cells was studied by the 
methods previously described (10). Briefly, 
cells were collected by washing the peritoneal 
cavity with phosphate-buffered saline (PBS, 
pH 7.2). The cell suspension was inoculated 
into glass bottles and allowed to attach for 2 
hr at 37°. Nonadherent cells were removed 
by repeated vigorous washing with PBS. The 
adherent cells were resuspended by scraping 
and cultured in medium 199 at a concentra- 
tion of 10® cells per ml. WBV was added at 
a multiplicity of infection of one and allowed 
to adsorb 1 hr at 37°. The cultures were then 
washed 3 times and fresh medium 199 was 

E I. Development 


Susceptibility to WBV in Mice Placed on Protein-Depletion Diet Before or 
After Virus Inoculation. 

WBV inoculated on day" 



-I +1 

+3 +5 

o. mice inoculated 
o. mice surviving 



10 10 

10 10 

ice were switched from normal to protein-depletion diet on day 0. WBV, lO^LDw (suckling mouse assay) was 
ted intraperitoneally on days indicated before (minus days) or after (plus days) switching diets. Mortality for 
oup of mice was summarized 14 days aAer inoculation with WBV. 

LE II. WBV Titer in Blood and Brain Tissue After Intraperitoneal Inoculation of Normal and 

PCM Mice. 

Days postinoculation 








nal: blood" 









I: blood" 












•gio LDso per 0.05 ml serum, 
igio LDso per 0. 1 g tissue. 



added at time 0. At intervals some cultures 
were rapidly frozen and thawed 3 times, de- 
bris was removed by centrifugation and the 
supernatant virus content was assayed. The 
experiments were run in triplicate and assay 
results were pooled by summing mortality of 
the individual titrations. In no instance did 
the end-points of individual titrations disa- 
gree by as much as one logio dilution. The 
results of these experiments are illustrated in 
Fig. 1. Whereas infectious WBV had com- 
pletely disappeared by 24 hr in macrophages 
obtained from normal animals there was an 
obvious difference in the ability of macro- 
phages from PCM mice to inactivate WBV. 
This suggests that the extraperitoneal dissem- 
ination that occurred to a much greater extent 
in PCM mice (Table II) resulted from the 
decreased capacity of local defense mecha- 
nisms to inactivate and contain the infectious 
inoculum. Presumably, macrophages were a 
major contributor in the population of cells 

Discussion, The efTects of malnutrition on 
host resistance to virus infections may be 
variable. Measles infections are a classic ex- 
ample of the increased susceptibility to severe 
and often fatal effects of disease occurring in 
the malnourished host (12). 

Experimentally, malnourished mice have 



Fig. I. infection of macrophages with WBV. Titer 
of infectious virus present at intervals after inoculating 
cultures of macrophages obtained from normal and pro- 
tein-deficient mice. 

been shown to have decreased resistance to 
coxsackie B3 virus (13) but apparently in- 
creased resistance to pseudorabies virus (14). 
Host resistance to different viruses represents 
a complex interaction of many mechanisms, 
each affected to varying degrees by malnutri- 
tion. Thus, contrasting results with individual 
viruses with distinctive pathogenetic schemes 
is not surprising. The model employed here 
attempts to isolate the effect of the macro- 

Although mice normally develop age-spe- 
cific resistance to WBV the target organ (cen- 
tral nervous system) remains susceptible to 
infection and disease (10). Thus, the data 
presented here on the loss of resistance to 
peripheral (ip) infection in protein-calorie 
malnourished mice implies that local resist- 
ance is impaired, and that moreover the 
events mediating resistance are rapidly sen- 
sitive to the deleterious effects of this mal- 
nutrition. In normally nourished mice only 
limited amounts of virus gain access to the 
circulation after ip inoculation, and this is 
apparently below the threshold required to 
initiate inifection in the central nervous sys- 
tem. Local restriction of the virus inoculum 
was not effective with protein-calorie mal- 
nutrition; large amounts of virus ap{>eared in 
the circulation and encephalitis ensued. 

The direct interaction of WBV and mac- 
rophages cultured in vitro suggested that a 
primary effect of protein-calorie malnutrition 
was on the ability of macrophages to restrict 
WBV infection. Significant levels of infec- 
tious virus presisted throughout the time pe- 
riod studied and in fact WBV may have 
replicated in the macrophages from malnour- 
ished mice. This must at least in part account 
for the susceptibility of these mice to WBV 
infection. In this regard they would be similar 
to newborn mice which are susceptible to 
infection for similar reasons (10). These re- 
sults suggest that impaired macrophage func- 
tion is an additional feature of protein-calorie 
malnutrition that contributes to the suscepti- 
bility of such individuals to certain virus dis- 
eases. Since the mechanism by which mac- 
rophages exert antiviral effects is not under- 
stood the cellular basis of the defect is ob- 

Summary. Mice which are normally resist- 
ant to infection with Wesselsbron (WBV) 



IS became rapidly susceptible to disease 
to this agent after being placed on pro- 
-depletion diet. After ip inoculation large 
)unts of virus appeared in the circulation 
3wed by fatal encephalitis. In normally 
mice only small amounts of virus could 
detected in blood and no disease devel- 
d. This suggested that local defense mech- 
ims which normally restrict the extent of 
ction was sensitive to the early effects of 
:ein-calorie malnutrition. That this was 
at least in part to impaired antiviral 
:tion of macrophage under these condi- 
s was confirmed by in vitro macrophage 
lies. Over the course of 24 hr infectious 
V disappeared after inoculation into cul- 
s of normal macrophages whereas infec- 
y persisted at high titers in macrophages 
Q protein-depleted mice. 

bis work was supported by a grant from the Rock- 
:r Foundation. 

^rimshaw, N. S., Taylor, C. E., and Gordon, J. E., 
inter. Nutr. Inf. WHO Monograph Scries No. 57 
Sirisinha, S., Edelman, R., Suskind, R., Champa- 

tana, C, and Olson, R. E., Lancet 1, 1016 (1973). 

3. Edelman, R. R., Suskind, R. E., Olson, R. E., and 
Sirisinha, S., Lancet 1, 506 (1973). 

4. Neumann, C. G., Lawlor, G. J., Stiehm, E. R., 
Swendseid, M. E., Newton, C, Herbert, J., Anunann, 
A. J., and Jacob, M., Amer. J. Clin. Nutr. 28, 89 

5. Sirisinha, S., Suskind, R., Edelman, R., Asvapaka, 
C, and Olson, R. E., Pediatrics 55, 166 (1975). 

6. Silverstein, S., Semin, Hematol. 7, 185 (1970). 

7. Douglas, S. D., and Schopfer, K., Clin. Exper. Im- 
munol. 17, 121(1974). 

8. Passwell, J. H., Steward, M. W., and Soothill, J. F., 
CUn. Exp. Immunol. 17,491 (1974). 

9. Keusch, G. T., Louglas, S. D., Hammer, G., and 
Urgubil, K., in '^Malnutrition and the Immune Re- 
sponse" (R. M. Suskind, ed.), pp. 277. Raven Press, 
NewYork, N.Y. (1977). 

10. Olson, L. C, Sithisam, P.. and Djinawi, N. K., J. 
Infect. Dis. 131, 119(1975). 

11. Reed, L. J., and Muench, H., Amer. J. Hyg. 27, 493 

12. Scrimshaw, N. S., Taylor, C. E., and Gordon, J. E., 
Amer. J. Med. Sci. 237, 367 (1959). 

13. Woodruff, J. F., and Kilboume, E. D., J. Infect. Dis. 

14. Cooper, W. C, Good, R. A., and Mariani, T., Amer. 
J. Clin. Nutr. 27,647(1974). 

Received September 19, 1977. P.S.E.B.M. 1978, Vol. 159. 


The Effect of Heparin on Growth of Mammalian Cells in Vitro^ (40290) 


The Hormel Institute, University of Minnesota, Austin, Minnesota 55912 

Heparin has been used primarily as a ther- 
apeutic anticoagulant agent (1), and clinically 
used to treat inflammatory and allergenic 
diseases (2). It has also been shown to accel- 
erate recovery of bum patients and promote 
wound healing in humans and animals (3-S). 
The mechanism and process of bum and 
wound healing from these observations is still 
not well elucidated. It is, therefore, of interest 
to investigate whether heparin can stimulate 
proUferation of human skin diploid cells in 
vitro which might be related to the mode of 
action of the healing process. 

The effect of heparin and other acid mu- 
copolysaccharides on the growth of various 
cell types, mainly maUgnant cells, has been 
studied by a number of investigators. The 
results obtained often have been controver- 
sial. Some investigators fmd inhibitory effects 
on cell growth (6-8), some stimulatory (8-10), 
and some report morphological changes (1 1). 
Therefore, this study was carried out in an 
attempt to provide more information about 
the nature of the effect of heparin on the 
growth of cultured mammalian cells. 

Materials and methods. Chemicals. Amino 
acids, vitamins and newbom calf semm were 
purchased either from Intemational Scientific 
Industries, Inc., Cary, IL, or Grand Island 
Biological Co., Grand Island, NY; predniso- 
lone-2l-sodium-succinate (PSS) and ^-2-hy- 
droxyethylpiperazine -N'-l- ethane -sulfonic 
acid (HEPES) from Sigma Chemical Co., St. 
Louis, MO; S-210 medium from Grand Is- 
land Biological Co.; Waymouth 752/1 dry 
powder medium from Schwarz/Mann Inc., 
Orangeburg, NY; fatty acid-free bovine se- 
rum albumin (FAF-BSA) from Miles Labo- 
ratories, Inc., Elkhart, IN; oleic acid from 
Nu-Chek Prep, Inc., Elysian, MN; various 
forms of heparin were kindly supplied by 
Riker Co., Division of 3M Co., St. Paul, MN; 

' This work was supported in part by the Office of 
Naval Research, Contract Nos. N000I4-75-C-0903, 
NR202-O71, and by The Hormel Foundation. 

Calbiochem, La Jolla, CA, Upjohn Co., Ka- 
lamazoo, MI; and highly purified heparin 
was a gift from Dr. J. A. Cifonelli, University 
of Chicago, Chicago, IL. 

Cell cultures. The sources of the cells and 
the methods used for cultivation were the 
same as described previously. Monkey kidney 
(MK-2) cells were cultivated as monolayers 
in Eaglets minimum essential medium 
(MEM) supplemented with 5% newbom calf 
serum (MEMs) (12). Novikoff hepatoma cells 
were grown in shaker culture in S-210 me- 
dium (13). Human prepuce cells were grown 
as monolayers in Eagle's MEM medium (14) 
supplemented with 10% newbom calf serum 
(MEMio) and baby hamster kidney cells 
(BHK-2 1) were grown in shaker culture using 
a modified Waymouth 752/1 medium (15). 

Growth of cells in the presence of heparin 
Prepuce cells were grown in Eagle's MEM 
supplemented with 4% newbom calf serum 
(MEM4) or MEM 10 in the presence of a wide 
range of heparin. Hanks' balanced salt solu- 
tion (BSS) was used as a base (16). The cells 
were used at an initial density of 2.0-3.0 x 
10^ cells/flask in a volume of 4 ml and were 
placed in 25 cm^ cell culture flasks (Coming 
Glass Works, Coming, NY). The cells were 
incubated at 37° for 6-10 days and were 
enumerated at varying intervals of time after 
trypsinization with the aid of a Coulter 

Thirty ml of BHK-2 1 cells were suspended 
in modified Waymouth 752/1 medium (15) 
containing 2.5% newbom calf semm and dif- 
ferent amounts of heparin. The initial cell 
population contained 3.5 x 10^ cells/ml and 
were incubated at 37° in a New Bmnswick 
gyratory shaker. At 0, 24, 48, 72 and 96 hr 
the cells were enumerated with a Coulter 

An initial population of 3.0 X 10^ MK-2 
cells in 4 ml of modified Waymouth 752/1 
medium (15) was added to 25 cm^ flasks. In 
order to estabhsh monolayers of cells, the 
medium was supplemented with 1% newborn 


0037-9727/78/ 1 59 1 -008850 1 .00/0 

CkJffyri^t £> 1978 by the Society Tor Experimental Biology and Medicine 
MrigJjts reserved. 



a. After 24-hr incubation at 37°, the 
was discarded and the cells rinsed 
i. Fresh modified Waymouth me- 
lout the serum supplement contain- 
ent amounts of heparin were added 
Is. The cells were incubated at 37° 
1 were enumerated at varying inter- 

>fr hepatoma cells were grown in 
Iture in S-210 medium in the pres- 
ifferent amounts of heparin. A start- 
:y of 2 X 10^ cells/ml was incubated 
r 4 days. Cell numbers were deter- 
ery 24 hr. 

ted effect of heparin and PSS on the 
'BHK'2I and prepuce cells. Varying 
of PSS and heparin were added in 
[)mbinations to the growth medium 
te BHK-21 cells in shaker cultures 
•uce cells in monolayers. The cell 
as measured at varying intervals of 
i procedures were the same as de- 
x)ve for testing the effect of heparin 

imum of two independent experi- 
;re performed for all studies. Each 
: was carried out in triplicate, and 
counts were made on each sample, 
ts were analyzed for significant dif- 
)y using a student's t test. 
. The effect of Riker's hog mucosal 
►n the growth of prepuce cells culti- 
MEMio is shown in Table I. There 
immediate differences observed in 
stween control and heparin-treated 
Qg the first 3 days after incubation. 
, the cells cultivated in the medium 
g S and 10 /iig/ml of heparin had an 
)f 30% and 23%, respectively, in cell 
over that of the control cells. There 
dest increase in the number of cells 

grown in medium containing IS and 20 /iig/ml 
of heparin, whereas the cells treated with 80 
/ug/ml of heparin had a 21% decrease in cell 
population when compared to control cells. 
The population of cells treated with 0, S, 10 
and 20 /ug/ml of heparin began to decline 
after day S, whereas cells treated with IS and 
80 /ug/ml of heparin continued to increase in 
cell number. 

When prepuce cells were cultivated in 
MEM4 after initially incubating the cells in 
MEMio for 24 hr, no differences in the growth 
between heparin-treated and untreat^ cells 
were observed until 8 days after incubation 
(Table II). On day 8, cells treated with S 
/[ig/ml of heparin showed a 90% increase in 
cell numbers over that of the untreated cells. 
Cells treated with higher concentrations of 
heparin which were less stimulatory than the 
cells treated with S /ig/ml of heparin showed 
an increase of about 3S% in population. On 
day 10, the cells treated with S, 10 and IS 
|ig/ml of heparin all showed about a 30% 
increase in cell number over the untreated 
cells. Cells treated with 80 /iig/ml of heparin 
had about the same growth rate as that of the 
untreated cells. 

Various heparins with different anticoag- 
ulant activity obtained from Upjohn Co. and 
Wilson Labs and further purified by J. A. 
CifoneUi showed similar stimulatory effects 
on the growth of prepuce cells (Table III). 
Each of the three heparins at a concentration 
of S /ig/ml increased the number of cells 
about 30-S0% from day S to 8 after incuba- 

Heparins from different sources at a con- 
centration of S /ig/ml showed similar stimu- 
latory effect on the growth of prepuce cells 
(Table IV), except there was slightly higher 
cell population when the cells were grown in 

i. Effect of Riker*s Hog Mucosal Heparin on Growth of Human Prepuce Cells in Eagle's 
Minimum Essential Medium Supplemented with 10% Newborn Calf Serum. 








1.33 ±0.03" 
2.97 ± 0.09 
4.70 ±0.10 
4.20 ± 0.29 

1.27 ±0.03 
3.40 ±0.21 
6.10 ±0.35* 
5.67 ± 0.35* 

1.30 ± 0.00 
3.20 ±0.10 
5.77 ± 0.22* 
5.30 ± 0.47 

1.27 ±0.07 
2.60 ±0.17 
5.03 ± 0.07 
6.20 ±0.31* 

1.17 ±0.03 
2.70 ± 0.27 
5.30 ± 0.20 
4.53 ± 0.29 

1.30 ±0.12 
2.63 ±0.18 
3.73 ± 0.70 
4.83 ± 0.46 

e cell number x lO'VHask (25 cm^) ± SEM from three flasks each counted in triplicate. 

antly diflerent from control {P < 0.05). These data are typical results from a minimum of three 

t experiments. 


TABLE U. Effect of Riicer*s Hog Mucosal Heparin on Growth of Human Prepuce Cells in Eagle's 
Minimum Essential Medium Containing 4% Newborn Calf Serum. 










1.50 ±0.07" 
2.00 ± 0.06 
2.77 ± 0.03 
2.83 ± 0.03 
2.83 ± 0.03 
4.87 ± 0.09 

1.93 ±0.09 
2.90 ±0.10 
3.17 ±0.35 
5.40 ± 0.32* 
6.43 ± 0.23' 

1.67 ±0.07 
2.73 ± 0.07 
3.17 ±0.19 
3.60 ± 0.36 
6.40 ± 0.46*^ 

1.93 ±0.15 
2.73 ± 0.22 
3.03 ±0.18 
4.10 ±0.72 
6.30 ±0.15* 

1.93 ±0.07 
2.53 ± 0.07 
3.37 ± 0.34 
3.83 ± 0.52 
5.77 ± 0.09* 

2.23 ± 0.07 
2.37 ± 0.03 
2.60 ± 0.06 
3.27 ± 0.43 
4.83 ±0.19 

" Average cell number x lO'Vflask (25 cm^) ± SEM from three flasks each counted in triplicate. 
* Significantly different from control (P < 0.01). 

' Significantly different from control ( P < 0.05). These data are typical of results from three independent 

TABLE 111. Effect of Heparins (5 ^ig/ml) with Different Specific Activities on Growth of Human 
Prepuce Cells in Eagle's Minimum Essential Medium Containing 4% Newborn Calf Serum. 








1.37 ±0.09^ 
3.07 ±0.15 
3.17 ±0.09 
4.03 ±0.12 
5.20 ±0.15 

3.27 ± 0.20 
4.57 ± 0.23' 
6.00 ± 0.40' 
7.10 ±0.32' 

3.17 ±0.13 
4.33 ±0.19' 
5.37 ± 0.02' 
6.80 ± 0.46' 

3.47 ± 0.29 
4.27 ± 0.22' 
5.63 ± 0.30^ 
6.63 ± 0.09^ 

" A: Beef lung heparin from Upjohn Co. further purified by gel filtration on Sephadex G-75 by J. A. Cifonelli 
(specific activity of 144 1/i/mg). 

* B: Beef lung heparin from Wilson Labs further purified by J. A. Cifonelli (specific activity of 180 Ifi/mg). 

' C: Beef lung heparin from Wilson Labs (specific activity of 1 10 1/x/mg). 

'' Average cell number x lO'^/flask (25 cm^) ± SEM from three flasks each counted in triplicate. 

' Significantly different from control {P< 0.05). 

^Significantly different from control (P < 0.01). These data are typical from three independent experiments. 

TABLE IV. Effect of Various Heparins (5 /ig/ml) on Growth of Human Prepuce Cells in Eagle's 
Minimum Essential Medium Containing 10% Newborn Calf Serum. 









1.17 ±0.03* 
1.90 ±0.06 
2.67 ±0.13 
3.17 ±0.18 
3.23 ± 0.09 

1.17 ±0.03 
2.30 ±0.10^ 
3.13 ±0.24 
3.73 ±0.19 
4.10 ±0.17 

1.27 ±0.03 
2.23 ± 0.09 
3.27 ± 0.35 
5.13 ±0.39^ 
4.07 ± 0.03'' 

1.10 ±0.06 
2.57 ± 0.20* 
3.00 ± 0.06 
4.23 ± 0.23' 
4.03 ± 0.28 

1.23 ±0.03 
2.90 ±0.10^ 
3.77 ± 0.22^ 
4.10 ±0.25*^ 
4.40 ± 0.06" 

" A: Control, B: Riker*s hog mucosal heparin, C: Riker's beef lung heparin, D: Riker*s crude hog mucosal heparin, 
E: Calbiochem*s hog mucosa heparin. 

* Average cell number x 10 '/flask ± SEM from three flasks each counted in triplicate. 

' Significantly different from control (P < 0.05). 

'' Signiflcantly different from control (P < 0.01). These data are typical from three independent experiments. 

the medium containing heparin from porcine of a wide range of concentrations of different 

intestinal mucosa from Calbiochem. heparins. 

Heparin showed little or no effect on the PSS at concentrations of 0.5 /iig/ml and 2 

growth of MK-2, BHK-21 and Novikoff rat /ig/ml inhibited the growth of BHK-21 

hepatoma cells, when the cells were cultivated (50%) and prepuce (25%) cells, respectively 

in the medium containing several concentra- (Figs. 1 and 2). In an attempt to reverse this 

tions of newborn calf serum in the presence inhibitory effect of PSS on cell growth, hog 



Combined effect of prednisolone-21-sodium- 
?SS) and heparin on growth of BHK-21 cells 
n shaker cultures. 30 ml of a cell suspension 
J Waymouth 752/1 medium containing an 
ilation of 3.5 X 10^ cells/ml was placed in 125 
eyer screw-cap flasks. LSS alone or a combi- 
'SS and heparin was added to the medium, 
rere incubated at 37 ^^ and enumerated with a 

inter at varying intervals of time. # •, 

, heparin; O O, 0.5 /xg/ml PSS; ▲ ▲, 

PSS + 5 Mg/ml heparin; A A, 0.5 /ig/ml 

/ig/ml heparin. Vertical bars represent stan- 
of mean. 

heparin was added into the culture 
It was found that cells grown in the 
containing both PSS and heparin 
it the same cell numbers as those in 
> containing no PSS (Figs. 1 and 2), 
reversal of heparin on the inhibitory 

sion. Previous results about the effect 
in and other acid mucopolysaccha- 
cell growth have been equivocal. 
17) first claimed inhibition of mitosis 
art fibroblasts and concentrations of 
varying from 20-500 /ig/ml. Cos- 
) reported cytotoxic action of heparin 
[XX) /xg/ml using Syrian hamster sar- 
lls. Lippman (8) demonstrated that 
at SO fig/ml inhibited growth of 
L" cells. King el al (18) found that 
it 1-1000 /Ig/ml showed little effect 
ivision of mouse "L" cells grown in 
)n cultures. 

Our data showed that heparin appears to 
promote the growth of prepuce cells but did 
not stimulate growth of BHK-21, MK-2 or 
Novikoff rat hepatoma cells. Takeuchi (10) 
noted that acid mucopolysaccharides have 
some promoting activity on tumor growth. 
Ozzello et al. (19) reported the growth pro- 
moting activity of acid mucopolysaccharides 
on a strain of human mammary carcinoma 
cells. They ascribed this action to the negative 
electric charge and the viscosity of acid mu- 

The controversy about the action of hepa- 
rin on the cell growth is presumably due in 
part to different dosages of heparin and the 
cell types used. Heparin in high concentration 
can be inhibitory to the growth of cells culti- 
vated in vitro. If the amount of heparin is 
maintained at a dose that just inhibits coag- 
ulation (2 /Ig/ml), it seems to be relatively 
noncytotoxic (II), and perhaps even stimu- 
latory to cell growth. Zakrezewski (20) 
claimed that the Jensen sarcoma in tissue 
culture was inhibited by heparin, but empha- 


Fig. 2. Combined effect of prednisolone-21-sodium- 
succinate (PSS) and heparin on growth of prepuce cells 
cultivated in Eagle's minimum essential medium supple- 
mented with 10% newborn calf serum (MEMio). Prepuce 
cells at an initial density of 2.3 x 10*^' cells/flask in 4 ml 
of MEM 10 medium were placed in 25 cm^ cell culture 
flasks. PSS alone or a combination of PSS and heparin 
was added to the medium. The cells were incubated at 
37" for 7 days and were enumerated with a Coulter 
counter at varying intervals of lime after trypsinization. 

• •, /ig/ml PSS, heparin; O O, 2 /xg/ml PSS; 

▲ ▲, 2 Mg/ml PSS + 5 Mg/ml heparin; A A, 2 

/xg/ml PSS + 20 /xg/ml heparin. Vertical bars represent 
standard error of the mean. 



sized that this drug was much less effective 
on normal embryonic tissue. 

Medium supplemented with low amounts 
of serum (4%) was used in one set of experi- 
ments to hopefully show a growth stimulatory 
effect when supplemented with heparin. 
Growth of prepuce cells might then be mag- 
nified when cells were subUminally starved 
(21). From the results shown in Table II, 
about 90% increase of cells was observed 
compared to the control in the presence of S 
|ig heparin/ml at day 8. Medium containing 
2% serum was also tested. No stimulatory 
effect of heparin was noted when prepuce 
cells were grown in this medium. Takeuchi 
hypothesized (10) that acid mucopolysaccha- 
rides did not serve as a nutritional component 
for cell growth but protects the cell surface 
and promotes the exchange of various metab- 
olites. Our observation indirectly further sup- 
ports this hypothesis. 

Cell populations from monolayer cultures 
were found to drop in the first 24-hr incuba- 
tion. This probably is attributed to cell lysis 
during the trypsinization process. Therefore, 
the baseUne data for all experiments was best 
interpreted after 24 hr cultivation. 

It has been reported that heparin in animal 
experiments could interact with steroid hor- 
mones(2). Our data demonstrated that hepa- 
rin reversed the inhibitory effect of PSS on 
the growth of prepuce and BHK-21 cells 
cultivated in vitro. This test system could be 
used to indirectly show heparin effects on cell 
growth when little or no activity was noted 
by heparin directly. This observation con- 
firms the hypothesis of Dougherty and Do- 
lonitz (2). 

A question had been raised whether trace 
metal contaminants or other unknown con- 
taminants of heparin might be responsible for 
its activity in aiding bum repair. When crude, 
commercial grade and highly purified hepa- 
rins from hog mucosa and/or beef lung 
sources were tested for promoting cell growth, 
no differences were found in the activity, 
which seemed to negate the role of heparin 
contaminants in the cell cuhure detection 
systems used 

Since heparin and heparin-like compo- 
nents are normal constituents of the blood 
and cells of higher animals, it is not surprising 
to find that heparin at a physiological level is 

harmless and even stimulatory to tt 
growth in vitro. 

Summary. The effect of heparin ( 
growth of four cell types cultivated i 
has been investigated. The results suggc 
heparin appears to have some growt 
moting effects on prepuce cells, w! 
showed little effect on the growth of Nc 
hepatoma, monkey kidney and baby h; 
kidney cells. Heparin reversed the inh 
effect of prednisolone-21-sodium-su( 
on the growth of prepuce and baby h< 
kidney cells. 

The authors thank Gregg Jorgenson and Chri 
son for their excellent techincal assistance. The ( 
and statements contained herein are the private 
the authors and are not to be construed as of 
reflecting the views of the Navy Department of t^ 
Service at large. 

1. Jeanloz, R. W., in "The Carbohydrates" (W. 
and D. Horton, eds.). Vol. II, pp. 609-617, A< 
Press, New York (1970). 

2. Dougherty, T. F., and Dolonitz, D. A., fi 
Cardiol. 14, 18(1964). 

3. McCleery, R. S., Schaffarzick, W. R., and L 
A., Surgery 26, 548(1949). 

4. Fenton, H., and West, G. B., Brit. J. Pharmi 

5. Saliba, M. J., Jr., Dempsey, W. C, and Kn 
L., J. Amer. Med. Assoc. 225, 261 (1973). 

6. Heilbrunn, L. V., and Wilson, W. L., Proc. S 
Biol. Med. 70, 179(1949). 

7. Costachel, O., Fadei, L., and Nachtigal, N 
Cell Res. 34,542(1964). 

8. Lippman, M., in "Epithelial-Mesenchymal 
tions", 18th Hahnemann Symposium (R. 
mayer and R. E. Billingham, eds.), p. 208, \ 
Sl Wilkins Co., Baltimore (1968). 

9. Morrison. L. M., Murata, K., Quilligan, J 
Schjeide, O. A., and Freeman, L.. Proc. S( 
Biol. Med. 118,770(1965). 

10. Takeuchi, J., Cancer Res. 26, 797 (1966). 

11. Abro, A., and Abraham, K. A., Experientia '. 

12. Jenkin, H. M., and Anderson, L. E., Exp. C 

13. Wennerstrom, D. E., and Jenkin, H. M., B 
Biophys. Acta 431, 469 ( 1976). 

14. Sandok, P. L., Knight, S. T., and Jenkin, H 
Clin. Microbiol. 4, 360 (1976). 

15. Guskey, L. E., and Jenkin, H. M., Appl. Mi 

16. Hanks, J. H., and Wallace, R. E., Proc. Sc 
Biol. Med. 71, 196(1949). 


17. Fischer, A., Proloplasma 2d, 344 (1936). 20. Zakrzewski, Z., Klin. Wochenschr. 11, 1 13 (1932). 

18. King, D. W., Bensch, K. G., and Simbonis, S., 21. Westermark. B., Biochem. Biophys. Res. Commun. 
Cancer Res. 18, 382 (1958). 69,304(1976). 

19. Ozzello, L., Lasfargeus, E. Y., and Murray, M. R., 

Cancer Res. 20, 600 (1960). Received May 4, 1978. P.S.E.B.M. 1978, Vol. 159. 


Immune Interferon Activates Cells More Slowly Than Does Virus-Induced Interfere 



Department of Microbiology, University of Texas Medical Branch, Galveston, Texas 77550 

Three antigenically difTerent types of inter- 
feron have been found: (a) a 27000-30000 
MW protein produced by somatic cells (fi- 
broblast interferon) stimulated by viruses, (b) 
an interferon produced by leukocytes (leu- 
kocyte interferon) also stimulated by viruses 
(VIF), and (c) immune interferon (IIF), pro- 
duced by lymphocytes following activation 
by mitogens or by specific antigenic stimula- 
tion (1-3). While the biochemical and biolog- 
ical properties of VIF have been extensively 
explored and many aspects of its mechanisms 
of production and action have been clarified, 
the properties of IIF, especially the mecha- 
nisms of activation of the cells, are as yet 
poorly understood. 

Several differences in function between IIF 
and VIF have been noted. It has been re- 
ported that: preparations of IIF exert higher 
antitumor and immunoregulatory activity as 
compared with VIF (4, 5), VIF immunosup- 
pressive action but not mitogen induced IIF 
action is blocked by mercaptoethanol (6), and 
IIF showed, at least in one system, a de- 
creased ability to inhibit virus yield relative 
to VIF (7). It seems then reasonable that, 
since the difTerent types of IF have different 
mechanisms of induction and show differ- 
ences of biologic activity, they are likely to 
manifest important differences in molecular 
and cellular reactivity (8-10). Since infor- 
mation on this subject could lead to a better 
understanding of the mechanism of action 
(antiviral, antitumor, immunoregulatory) of 
the different types of IF, we have undertaken 
a comparative study on cellular activation by 
VIF and IIF. 

In previous studies (1 1, 12) we showed that 
a very brief reaction (minutes) between VIF 
and cells at 37° rapidly results in cellular 
activation which, after removal of IF, is fol- 

' This investigation was supported by DHEW Grant 
No. 500170 and by CNR Progetto Finalizzato Virus. 

lowed 30 min later by the transcription 
translation of mRNA for the antiviral pre 
responsible for the cellular antiviral state, 
present study is a comparison of these kin< 
of cellular activation using IIF and VIF. 

Materials and methods. Human leuko 
interferon (10^ units/mg protein), induce 
Sendai virus, was obtained from the Anti^ 
Substances Program, NIAID, NIH, and 
produced by methods previously descr 
(13). Mouse fibroblast interferon was 
tained from the mouse C243 cell line indi 
with Newcastle Disease Virus as previc 
described (10^ units/mg protein; 14). Hu 
imQiune interferon (lO^'* units/mg proi 
was obtained from normal lymphocyte 
tures stimulated for 4 days with staphylo 
cal enterotoxin A (SEA). Mouse immuni 
terferon (10^ units/mg protein) was obta 
from mouse splenic cell cultures stimul 
with SEA (15). Virus-type interferons ' 
shown to be resistant to 5 days exposui 
pH 2 and completely neutralized by spc 
antibody. Immune interferons were inst 
at pH 2 and not significantly neutralize< 
antibody to virus-type interferon (15). Ii 
feron and interferon-induced antiviral a< 
ity were measured by the inhibition of 
yield of Sindbis virus (human interferon 
GD7 virus (mouse interferon) hemagglut 
in a single cycle yield assay (16) emplo 
tube cultures of human diploid foreskin c 
HF 19, or mouse L cells, strain CCL-l. Ir 
feron titers are expressed as human or m( 
reference units. Temperature control at 
for short periods of time was effected 
waterbath and longer incubations were 
formed in a 37° incubator containing 4% 
as previously described (11). 

Results. Development of antiviral resistt 
in cells treated with virus-induced or imn 
interferon. The results of a representative 
periment carried out with human leuko 
and immune IFs are shown in Fig. 1. I 


Copyright €> 1978 by the Society for Experimentsl BioJogy and Medicine 
An lights reserved. 




lOOUMt/M vir 

^ — 


5 J 






1 3 














Fig. I. Development of antiviral activity in human 
liploid foreskin cells treated for various periods of time 
^ith too units of either virus-induced or mitogen-in- 
Juced human interferon. 

:ypes of IF were applied at a concentration 
3f 100 reference units per ml as previously 
lescribed (11). At preestablished times the 
interferon was removed and the cultures were 
ivashed 4 times and challenged with virus 
[multipUcity of infection, 10). After 1 hr for 
^iral adsorption the cultures were washed 3 
times and incubated for 18 hr before titration 
of viral yield. A control titration of the level 
of IF activity was included in every experi- 
ment. It may be seen that the cell cultures 
treated with VIF developed substantial re- 
sistance after a 5-min treatment and that the 
degree of resistance to Sindbis virus replica- 
tion continued to rise thereafter so that 1 hour 
later it was greater than the measurable level. 
However, IIF did not induce detectable re- 
sistance over 2 hr, and marginal resistance 
was produced only after 4 hr. The expected 
degree of antiviral activity was induced after 
24 hr treatment. 

Similar results were obtained for mouse L 
cells treated with 300 reference units of either 
virus induced or immune mouse IF and chal- 
lenged with GD-7 (Fig. 2). Additionally the 
same type of kinetics was observed for two 
more virus-cell systems: vesicular stomatitis 
virus (human cells, multiplicity of infection, 
10) and mengovirus (L cells, multiplicity of 
infection, 0.1). 

Since IIF preparations had a much lower 
specific activity as compared with VIF prep- 
arations, the possibility that some contami- 
nant present in IIF could affect the rate of 
cellular activation was examined. Specifically 
cell cultures were treated either with 100 units 
of VIF, 100 units of IIF or 100 units of VIF 
plus 100 units of IIF. The cultures were then 
challenged after 15 min, and 4 hr to deter- 

mine whether protection in the cultures 
treated with both types of IF developed ac- 
cording to the kinetics of development of VIF 
or IIF. The results (Table I) showed that the 
rapid kinetics of development of the antiviral 
state induced by VIF was not slowed by the 
presence of IIF. The same results were ob- 
tained either when the two types of interferon 
were mixed before addition to the cells or 
when either type was added immediately be- 
fore the other. 

Binding of interferon to cells at 37. It has 
been shown that cells treated with VIF bind 
interferon molecules very rapidly (17-21). 
However data on cellular binding of IIF are 
not yet available. Since the lack of rapid 
cellular activation by IIF, as compared with 
VIF, could be due to different kinetics of 
cellular binding, experiments were designed 
to estabUsh the extent of binding of two types 
of IF under the conditions of rapid activation. 

-"-• '>— • — i{r=»« 

Fig. 2. Development of antiviral activity in mouse L 
cells treated for various periods of time with 300 units of 
either virus induced or mitogen induced mouse inter- 

TABLE 1. Induction of the Antiviral State by 

Virus-Induced Interferon, Immune Interferon, or 

A Combination of Both. 

Species of in- 

Type of inter- 

Inhibition of virus 
yield* after treat- 
ment for 

15 min 









VlF + llF 













* Logio inhibition of Sindbis virus PFU yield (human 
interferons) or log2 inhibition of GD-7 virus HA yield 
(mouse interferons). 



Specifically, 0.5 ml of medium containing 
1000 reference units per ml of either virus- 
induced or immune mouse IF were applied 
to duplicate tube cultures of L cells main- 
tained at 37°. After different periods of time, 
one group of cultures was washed 6 times 
with Earle's balanced solution, refed with 0.5 
ml of Eagle's medium, and frozen-thawed 3 
times to release cell-associated IF. The fluids 
were then assayed for IF. The results of a 
representative experiment are shown in Table 

It may be seen that binding of both VIF 
and IIF was essentially maximal after 5 min- 
utes incubation at 37° and that the amount 
of bound IF remained unchanged thereafter. 
There was no significant difference between 
the amount of VIF and IIF bound at each 
time. The IF associated with the cells was 
approximately 0.6-1.2% of the total IF ap- 
plied to the cultures. This finding substan- 
tially agrees with previous observations on 
cellular binding of VIF (16-21) and shows 
that binding of IIF occurs at a similar rate 
and to a similar degree. 

Discussion. It has been previously shown 
that the development of the antiviral state in 
cells treated with VIF is triggered immedi- 
ately by a very brief interaction between IF 
and cells and continues when the IF is re- 
moved from contact with the cells by washing 
and antibody inactivation (22). This finding 
has been confirmed for VIF by the data 
presented in this paper. However IIF, assayed 
under identical experimental conditions, 
failed to immediately activate cells after brief 
contact. Thus in both the human and mouse 
systems, detectable antiviral resistance was 
induced by IIF only after several hours of 
incubation at 37°. The different kinetics of 
cellular activation by the two types of IF may 
be due to: (a) Difference of availability of 
cellular receptors, (b) the presence in the IIF 
preparation, and not in the VIF preparations, 
of substances capable of retarding expression 
of interferon activity under the present ex- 
perimental conditions, and (c) a different 
mechanism of activation of the antiviral state. 

Studies of cellular binding of the two types 
of IF did not show any significant difference 
between their binding activity, suggesting 
that differences of association by the two IFs 
with the cell may not play an important role 

TABLE 11. Cellular Binding of Virus-Induc 
Immune Mouse Interferon. 

Type of Inter- 

Units of interferon associi 
with cells after (min) 

5 10 15 30 


6 12 10 12 
8 10 10 12 

in establishing the different kinetics of 
vation. However, it should be borne in i 
that this relatively durable binding whi 
usually measured (17-21) may not reflec 
transient cell-activating event by whic 
induces rapid resistance (22). Specifica 
has been shown previously that firm bir 
to the cell surface is not required for the i 
induction of the antiviral state by VIF 
the present finding of different kineti 
activation despite equal kinetics of bir 
further supports that conclusion. 

The hypothesis that a component o 
IIF preparation could interfere with the i 
action of the VIF molecule seems less 1 
since the presence of the slow acting 
preparation did not inhibit the rapid a( 
tion by VIF. However this experiment 
not eliminate the possibility of the prei 
of a substance which only inhibits the 
action of IIF. Fui:ther studies with pu; 
preparations of IIF could test this possil 

The hypothesis that VIF and IIF ma 
duce the antiviral state through diff 
mechanisms appears likely and deserve: 
ther study. If the same biological activity 
be evoked through different activation 
esses, the finding may provide a useful \ 
ing model for studying several critica 
activities, such as cellular regulation of 
expression, regulation of gene products 
cell membrane receptor functions. Addi 
ally, the different kinetics of activation c 
antiviral state by the two types of IF 
provide a simple and rapid method to d 
entiate them. 

Summary. The kinetics of activation c 
antiviral state by virus induced interferoi 
by mitogen-induced immune interferon 
been studied comparatively. It has 
found that both human and murine > 
induced interferons are able to activat 
antiviral state after a brief (minutes) co 
with the cells. In contrast, several hours 



iquired for both human and mouse immune 
iterferons to induce a comparable level of 
itiviral resistance. Experiments measuring 
le binding of the two interferons to celk 
lowed that there was no significant differ- 
ice in the rate and degree of binding, sug- 
^sting that a different total association of 
iterfcron with cells could not account for the 
ower kinetics of activation by immune in- 
irferon. Additionally, the possibility that 
>me contaminants present in the immune 
iterferon preparation could nonspeciflcally 
iterfere with the rapid induction phenome- 
on is not supported by the finding that the 
ipid kinetics of cell activation by virus-in- 
uced interferon was not modified by the 
resence of immune interferon. The interest- 
ig possibility which remains is that the two 
Iterferons may activate cells by different 

The authors are greatly indebted to Dr. G. Georgiades 
ad Mr. M. Langford for the generous giAs of immune 
Iterferon and to Dr. S. Baron for helpful suggestions 
nd criticism. 

1. Whelock, E. F., Science 149, 310 (1965). 

2. Green, J. A., Cooperband. S. R., and Kibrick, S., 
Science Ids, 1415 (1%9). 

3. Salvin, S. B., Youngner, J. S., and Lederer, W. H., 
Infect. Immun. 7, 68 (1973). 

4. Salvin, S. B., Youngner, J. S., Nishio, J., and Neu, 
R., J. Nat. Cancer Inst. 55, 1233 (1975). 

5. Sonnenfeld, G., MandeK A. D., and Merigan, T. C 
Fed. Proc. March 1977, Vol. 4, 1977; Cell Immunol. 


6. Johnson, H., Cell. Immun. 36, 220 (1978). 

7. Fleischmann, W. R. Jr., Texas Rep. Biol. Med. 35, 

8. Paucker, K., Texas Rep. Biol. Med. 35, 23 (1978). 

9. Maehava, N., Ho, M., and Armstrong, J. A.. Infect. 
Immun. 17,572(1977). 

10. Chadha, K. C, Sclair, M., Sulkowsky, E., and Carter, 
W. A., Biochem. 17, 196 (1978). 

11. Dianzani, F.. and Baron, S., Nature (London) 257, 

12. Dianzani, F., Levy, H. B., Berg, S., and Baron, S., 
Proc. Soc. Exp. Biol. Med., 152, 533 (1976). 

13. Cantell, K., Strander, H., Hadazi, G., and Nevaliima, 
H. R., in "The Interferons" (G. Riu, ed.), p. 223, 
Academic Press, New York (1968). 

14. Oie, H. K., Gazdar, A. F., Buckler, C. E., and Baron, 
S., J. Gen. Virol. 17, 107(1972). 

15. Johnson, H. M., Stanton, G. J., and Baron, S., Proc. 
Soc. Exp. Biol. Med. 154, 138 (1977). 

16. Oie, H. K., Buckler, C. E., Uhlendorf, C, Hill, D. 
A., and Baron, S., Proc. Soc. Exp. Biol. Med. 140, 

17. Levine, S., Proc. Soc. Exp. Biol. Med. 121, 1041 

18. Friedman, R. M., Science 156, 378 (1967). 

19. Berman, B., and Vilcek. J., Virology 57, 378, (1974). 

20. Kohno, S., Buckler, C. E., Levy, H. B., and Baron, 
S., in "Effects of Interferon on Cells, Viruses and the 
Immune System" (A. Geraldes, ed.), p. 123, Aca- 
demic Press, New York (1975). 

21. Stewart, W. I., II., in "Effects of Interferon on Cells, 
Viruses and the Immune System** (A. Geraldes, ed.), 
p. 75, Academic Press, New York (1975). 

22. Dianzani, F., and Baron. S., Proc. Soc. Exp. Biol. 
Med. 155,562(1977). 

Received April 24, 1978. P.S.E.B.M. 1978, Vol. 159 


Competition Binding Assay Using o-Methyl-{^H}-Demethyl-y-Amanitin for Study of RNA 

Polymerase B (40292) 


Dep€urtment of Microbiology, Graduate School of Public Health and Department of Medicine, U.S. V.A. Hospital and 
University of Pittsburgh, Pittsburgh, Pennsylvania 15240 

The understanding of RNA synthesis and 
processing, and the enzymes and control 
mechanisms involved is of central importance 
in biology. The simplest and most reliable 
criterion for classification of eukaryotic RNA 
polymerases is their sensitivity to the fungal 
toxin a-amanitin. a-Amanitin and other nat- 
urally occurring amatoxins, as well as their 
synthetic derivatives, are of particular interest 
as molecular probes in the study of transcrip- 
tion. They bind very tightly to the polymerase 
molecule at a site separate from that which 
binds to the template DNA and product 
RNA (I, 2). This interaction does not affect 
the stabiUty of the transcription complex 
formed between the enzyme and template 
nor does it interfere with the binding of pre- 
cursor nucleotide triphosphates (1, 2). 

A radioactive derivative of a-amanitin was 
synthesized by Wieland and Fahrmeir for use 
in their structural studies of the molecule (3). 
The method of synthesis as reported used 
large amounts of the parent compound and 
employed various destructive analytic tech- 
niques to study placement of functional 
groups in non-radioactive intermediate com- 
pounds and in the labeled end product. To 
permit the synthesis of a radioacive derivative 
of a-amanitin from a small amount of com- 
mercially available starting material their 
procedure was modified, and several new 
methods for analyzing the unlabeled inter- 
mediate compounds and the end product 
were introduced. This should enable more 
biologists to avail themselves of this powerful 

To demonstrate radiochemical purity and 
ensure reactivity of the labeled end product, 
a competition assay was developed. The assay 
demonstrates that the labeled derivative, o- 
methyl-[^H]-demethyl-Y-amanitin, binds to 
the same site as a-amanitin when reacted 
with either purified or crude preparations of 
RNA-polymerase B. Since this technique es- 

tablishes that the labeled and unlabeled com- 
pounds are essentially interchangeable, it al- 
lows the study of amatoxin binding over 
wider ranges of ligand concentration than 
heretofore possible when radioactive material 
was used alone. 

Methods and materials. To synthesize the 
first intermediate, {^-methyl-a-amanitin, 5 mg 
of a-amanitin were dissolved in 4 ml of an- 
hydrous methyl alcohol. This was added to 3 
ml of an etheric solution of diazomethane 
generated from A^-methyl-A^'-nitro-A^-nitro- 
soguanidine (MNNG) in the outer vessel of 
an MNNG diazomethane generator. The ves- 
sel was stoppered immediately and the reac- 
tion mixture was kept at room temperature 
for 2 hr. The solvents were then evaporated 
in vacuo, the remaining residue was resus- 
pended in a small volume of methanol/HaO 
(1:1) and chromatographed on a column of 
Sephadex LH-20 (1.8 x 1(X) cm) with meth- 
anol/HaO (1:1) (3). Column efiluants were 
monitored at 310 nm with an Isco UA-5 uv 
monitor. Fractions with uv absorbing mate- 
rial were collected, solvents evaporated in 
vacuo, and the remaining residues redissolved 
in 1 mg H2O. Concentration of the products 
was determined spectrophotometrically (1,3, 

The second non-radioactive intermediate, 
{?-methyl-aldoamanitin, was synthesized via 
periodate oxidation of {7-methyl-a-amanitin 
(3). This was accomplished by the addition 
of 0.48 mg sodium periodate dissolved in 1 
ml water to 2.08 mg of o-methyl-a-amanitin 
dissolved in 1.5 ml H2O. The mixture was 
stirred for 5 min at room temperature fol- 
lowed by reduction of the excess periodate by 
the dropwise addition of 1.1 ml of 0.09 A^ 
sodium bisulfite. This mixture was chromat- 
ographed on a column of Sephadex LH-20 
(1.8 X 100 cm) using H2O as the solvent. 

Synthesis of o-methyl-[^H]-demethyl-Y- 
amanitin was accomplished by reduction of 


'^27/78/ 1 59 1 -009850 1 .00/0 

C 1978 by the Society for Experimental Biology and Medicine 



hyl-aldoamanitin with sodium boro- 
lydride (3). One and two tenths mg of 
m boro-[^H]-hydridc (209 mCi/mg) was 
I to 0.94 mg {7-methyl-aldoamanitin dis- 
1 in l.S ml of H2O. The reaction mixture 
ontinuously stirred at 0° for 90 min at 

time the mixture was acidified by the 
on of 1 ml of 0. 1 N HCl. After an 
onal 15 minutes the mixture was neu- 
td with 0. 1 N NaOH. The reaction mix- 
vas chromatographed on a Sephadex 
} column using methanol/H20 (1:1) as 
bed above. Fractions found to have 
i proper uv-spectrum (4) and containing 
ictivity were rechromatographed on a 
in of Sephadex G-50 (0.9 x 15 cm) 
H2O as the solvent. Peak fractions were 
ted, uv-absorbance and counting rates 
determined and the specific activity of 
hyl-[^H]-demethyl-Y-amanitin was cal- 
id. The specific activity was verified by 
ition of the labeled derivative with pu- 

wheat germ RNA polymerase in a 
ig assay described below. 
n layer chromatography of a-amanitin 
derivatives. In order to identify various 
on products and assess their purity, ali- 

of peak fractions were concentrated 
len studied by thin layer chromatogra- 
m Silica Gel-OF plates. Two solvent 
OS were employed, selected for their 
f to separate the intermediate com- 
Is. Chromatograms were visualized by 
ag with Erlich's solution or transcin- 
Idehyde/HCl (3, 5). a-Amanitin was 
as a reference standard against which 
igration of the intermediates was com- 
. Radioactive products were located by 
ng 0.5 cm squares from moist plates 
^ing chromatography. The resultant ma- 
was then digested overnight in Nuclear 
go Solubilizer (NCS) at 45° and 
ed in nonaqueous, toluene based scin- 

"a-red spectrophotometry of a-amanitin 
on-radioactive derivatives. As an addi- 
proof of the proper placement of func- 
groups in the amanitin molecule infra- 
>ectra were obtained for a-amanitin, o- 
4-a-amanitin and o-methyl-aldoaman- 
imall amounts of each compound (ca. 
g) were dissolved in H2O and lyophi- 
KBr pellets were prepared for each 

sample using a Wilk's mini press. The pellets 
were scanned from 4000 to 600 cm" on a 
Beckman Acculab 4 Infra-red Spectropho- 
tometer. The reference beam was attenuated 
to permit adjustment of the baseline. Chro- 
matographically pure a-amanitin was used as 
a reference compound. 

Amatoxin competition binding assay. The 
method used to demonstrate the binding of 
[^H]yamanitin was based on the procedure of 
Cochet-Meilhac et al (1, 2). Purified wheat 
germ RNA polymerase or the enzyme present 
in crude homogenates of baby mouse kidneys 
was used as a substrate for binding the radio- 
active ligand. Crude homogenates were pre- 
pared by grinding whole kidneys of baby 
mice (8-lOd) in a Potter-Elvehjem tissue 
grinder in homogenizing buffer (50 mAf Tris 
HCl pH 7.4; 0.1 mM EDTA; 0.1 mA/ dithio- 
threitol and glycerol 30% v/v). Aliquots of 
100 /il of the crude homogenates or 100 /il of 
the purified enzyme in binding buffer (1.63 
X 10"® A/) were incubated in an assay mix- 
ture containing 500 /il binding buffer (80 mA/ 
Tris HCl pH 7.9; 0.1 mA/ EDTA; 0.1 mA/ 
dithiothreitol; 150 mA/ (NH4)2S04; 0.2 
mg/ml bovine serum albumin; 0.4 mg/ml 
rabbit gamma globulin and 30% (v/v) glyc- 
erol), 10 /il [^H]-Y-amanitin (9.13 x 10"^ 
/imoles ca. 2.0 Ci/mmole) and 10 /il of unla- 
beled a-amanitin in varying concentrations. 
Controls for [^HJ-y-amanitin binding con- 
tained 10 /il binding buffer in place of a- 
amanitin. Samples were incubated at 4^ for 
18 hr. Afler 18 hr 1 vol of (NH4)2S04 solution, 
saturated at 4^, was added to the reaction 
mixture and samples were kept at 4^ for an 
additional hour. Free and unlx)und amanitin 
were then separated by centrifugation at 
39,000g for 20 min. The supernatant was 
discarded and the pellet was redissolved in 1 
ml of binding buffer; an equal volume of 
saturated (NH4)2S04 was again added and 
the samples were incubated at 4° for 30 min, 
at which time they were recentrifuged as 
described above. This suspension-reprecipi- 
tation step was repeated two additional times. 
Finally the pellet was dissolved in 200 /d of 
H2O, digested overnight in NCS at 45° and 
counted in non-aqueous toluene based scin- 
tillant. Counting efficiency was approxi- 
mately 85% of that obtained for unquenched 
samples. Values obtained with the highest 



concentration of a-amanitin were found to 
correspond to the background samples con- 
taining no RNA polymerase when the puri- 
fied enzyme was used. When the assay was 
done using the crude homogenate as a source 
of RNA polymerase the values obtained at 
the highest concentrations of a-amanitin were 
assumed to represent nonspecific binding of 
the labeled derivative. This value did not 
exceed 6% of the total label bound and was 
used to correct experimental values obtained 
with crude homogenates. 

[^H] Amanitin saturation assay. To verify 
the specific activity of the [^HJ-y-amanitin as 
determined by the ratio of radioactiv- 
ity/absorbance at 3 10 nm an experiment was 
done to ascertain the amount of purified 
wheat germ RNA polymerase required to 
saturate a fixed amount of the radioactive 
ligand. Each sample contained [^H]-y-aman- 
itin (1.47 X 10"® M) and variable concentra- 
tions of RNA polymerase from 2.63 X 10"® 
M to 5.26 X 10"® A/. Concentration of [^H]- 
y-amanitin at saturation was based on the 
50% end point. Conditions for the assay are 
identical to those described above for [^H]- 
y-amanitin binding controls. 

Materials, The materials used for these ex- 
periments were obtained from the following 
suppUers: ^-methyl-A/^'-nitro-A^-nitrosogua- 
nidine, /?-dimethyl-aminobenzaldchyde and 
MNNG diazomethane generator, Aldrich 
Chemical Co. Milwaukee, WI; a-Amanitin, 
Boehringer-Mannheim Biochemicals Indi- 
anapolis, IN; Nuclear Chicago Solubilizer 
(NCS) and sodium boro-[^H]-hydride, Amer- 
sham-Searle Co., Arlington Heights, IL; 
transcinnamaldehyde, Eastman Organic 
Chemical Rochester, NY; wheat germ RNA 
polymerase. Miles Laboratories, Elkhart, IN; 
Sephadex LH-20 and G-50, Pharmacia Fine 
Chemicals, Piscataway, NJ; Silica Gel OF 
TLC plates. New England Nuclear, Boston, 
MA; rabbit IgG and ultra pure ammonium 
sulfate Schwartz Mann, Orangeburg, NY; so- 
dium periodate and sodium metabisulfate, 
Sigma Chemical Co., St. Louis, MO. 

Results. Synthesis of a radioactive derivative 
of a-amanitin. Chromatography of the meth- 
ylation product of a-amanitin on Sephadex 
LH-20 resulted in two peaks absorbing at 3 10 
nm (Fig. 1). Thin layer chromatography of 
the material in fraction 13 in butanol/ 
acetone/HaO (30/3/5) yielded a band co-mi- 






II 15 20 » M 19 

FrKtiM Ninbtr 

Fig. 1. Products of methylation of c 
chromatographed on a column of Sephs 
X 100 cm) using methanol/HiO 1:1. The 
was monitored at 310 nm. Fractions ol 
7.5 ml were collected at 40 min intervals 

grating with the a-amanitin n 
0.34). With methanol/HzO (4:1 
vent an effective separation resi 
amanitin migrating further than 
ated derivative (/?/ = 0.86 vs. 0.: 

Infra-red spectrophotometry \ 
the differences in the methyla 
and the parent compound. Th< 
grating methyl derivative (an ar) 
phenolic hydroxyl group of th< 
moiety) was expected to exhibit < 
the regions of 1300-1 180 cm"^ ar 
cm"*. The comparison spectra c 
and (?-methyl-a-amanitin (Fig. 
demonstrate changes betweer 
cm"* and 1125-1025 cm"'. 

The results of Sephadex LH-2i 
raphy of the periodate oxidatio 
o-methyl-a-amanitin are presen 
Only the major peak was four 
typical uv-spectrum for an amat( 
layer chromatography using 
tone/H20 and stained with 
hyde/HCl showed that the rea( 
migrates slower than the a-ama 
{Rf = 0.30) and stains reddish 1 
than violet. Infra-red spectr 
showed changes at 2800 cm"* ai 
consistant with the introductic 
phatic aldehyde group into the ai 
ecule (Fig. 2c). 

Following reduction of o- 
amanitin with sodium boro-[^H 
reaction products were separate 
dex LH-20. Three peaks absorbi 
were eluted (Fig. 4). The ma 
leading peak did not possess a < 


Wavelengtfi in Microns 

** 3 3.> 4 4.> > S> t %.i 7 ?■> I • M 11 12 14 II 

- I 111 r 1 I I I I T 1—1 — t— I 


WsvMNinibor cm 

12t0 IMt Mt Mt 

. A comparison of the infra-red spectra of a-amanitin (2a)o-methyl-a-amanitin (2b) and o-melhyl-aldo- 
(2c). Samples were prepared as KBr pellets and scanned at slow speed using the normal slit program of a 
Acculab 4 Infra-red Spectrophotometer. The reference beam was attenuated to obtain a suitable base line. 

Lnim of an amatoxin and while the 
I in the center peak did resemble an 
in by uv spectrophotometry the rela- 
>unt of incorporated radioactivity was 
IV. Only the material in the trailing 
d both an amatoxin uv spectrum and 
icant amount of incorporated label, 
jor peak of radioactivity was unasso- 
nih any amatoxin containing fraction 
s assumed to be unreacted. Fractions 
40 were pooled, concentrated in vacuo 
•chromatographed on a column of 
ex G-50 to ensure complete removal 
unreacted radioactivity. The column 
is presented in Fig. 5. The major 
of radioactivity coincided with the 
' uv-absorbance in fraction 11. Very 
itaminating radioactivity was present. 
f of the material eluted from Sepha- 
iO was verified by thin layer chroma- 
y in butanol/acetone/H20. The mi- 

gration of the radioactive derivative was com- 
pared to the marker, a-amanitin, which was 
detected by staining with Erlich's reagent. 
The results of the thin layer chromatography 
are presented in Fig. 6. The radioactive de- 
rivative migrates as a single band (Rf = 0.40) 
ahead of the marker {Rf = 0.34). Neither 
infra-red spectrophotometry nor destructive 
analytic techniques were employed to verify 
the chemical structure of the end product 
because of the small amount recovered. The 
end product is assumed to be o-methyl-[^H]- 
demethyl-y-amanitin since only the aldehyde 
formed in the previous step would be availa- 
ble for borohydride reduction. 

Saturation of wheat germ RNA polymerase 
with t^H]-y-amanitin. The specific activity of 
[^H]-Y-amanitin was determined by two in- 
dependent methods. Based on the uv absorp- 
tion and counting rates of several small sam- 
ples, the material in fraction 1 1 (Fig. 5) was 





N n 


Fig. 3. Products of sodium periodate oxidation of o- 
methyl-aldoamanitin were chromatographed on a col- 
umn of Sephadex LH-20 (1.8 x 100 cm) using H2O as 
the solvent. Fractions contained approximately 6.25 ml. 
Column monitoring and fraction collection were accom- 
plished as described in Fig. 1. 

was calculated to be 1.47 X 10"* M and the 
specific activity 1.88 Ci/mmole. Saturation 
data are presented in Fig. 7. 

Amatoxin competition assay, A competition 
assay was designed to test the hypothesis that 
[^H]-Y-amanitin bound to the same site as a- 
amanitin and with approximately the same 
affinity. The concentration of the radioactive 
derivative was constant at 1.47 X 10"® 3/ 
(approximately 12,000 cpm) and the concen- 
tration of the competing, unlabeled a-aman- 
itin was varied from 8.27 x 10"^* to 2.62 x 
10"* A/. The assays were carried out at four 
to five times the concentration of [^H]-y- 
amanitin required to saturate the amount of 
RNA polymerase present. Controls for non- 

• 32 





/I h 



1 III 


1 1.12 

^ , 



• , 

• M 


^ I 


>^ ..— ^ 

*,«*.*r.*^- • 



Fig. 4. The reaction products of sodium boro-[^H]- 
hydride reduction of o-methyl-aldoamanitin were chro- 
matographed on a column of Sephadex LH-20 (1.8 x 
100 cm) using methanol/H20 (1:1) as the solvent. Flow 
rate, column monitoring and fraction size are described 
in Figure 1. 50 /xl aliquots of each fraction were used to 

approximate the total radioactivity. (# •) absorb- 

ance 310 nm; (A A) ['H] cpm. 

estimated to contain 7.25 x 10"^ /mioles/ml 
{?-methyl-[^H]-dcmethyl-Y-amanitin with a 
specific activity of 2.50 Ci/mmolc. Fraction 
12 was found to contain 7.23 x 10"^ 
jLimoles/ml, and have a specific activity of 
2.16 Ci/mmole. Specific activity estimates 
based on the saturation of [^H]-y-amanitin 
with wheat germ RNA polymerase agreed 
well with those obtained by instrumental 
methods. The fifty percent maximum binding 
of RNA polymerase was found to occur at 
7.35 X 10"® M, Assuming that the reaction 
was at equilibrium, had a very small Kd 
(approximately 10"^^ A/, see ref. 1, 2) and that 
the purified enzyme contained a single bind- 
ing site, the concentration of [^HJ-y-amanitin 









X^v. _ 

' ^" 


- 1 

. *,*^-^-^ 








Fig. 5. o- methyl -[^H]-demethyl-Y- amanitin con- 
tained in fraction 39 (Fig. 4) was chromatographed on a 
colunm of Sephadex G-50 (0.9 x 15 cm) using H2O as 
the solvent. Each fraction contains 0.75 ml; flow rate 
0.38 ml/min. Five microliter aliquots of each fraction 
were assayed for radioactivity (#- — #) absorbance 3 10 
nm; (A A) ('HI cpm. 

Fig. 6. Thin layer chromatography of o-methyl- 
|^H]-demethyl-y-amanitin. A 5 /il aliquot of the peak 
fraction eluted from Sephadex G-50 was chromato- 
graphed in buUnol/acetone/H20 (30:3:5). Migration of 
a-amanitin indicated by arrow. Each point represents a 
migration of 0.5 cm. 



8 S 

WkMt Gtnii UNA Ratyntrast ncIO* 

. 7. Saturation of wheat genn RNA polymerase 
> - methyl - [•^H] - demethyl - y- amanitin. o - methyl- 
;methyl-Y-anianitiii (1.47 x 10"** A/) was incu- 
!br 18 hr in the presence of increasing concentra- 
f wheat germ RNA polymerase (2.63 x 10* M 
• X 10"" My Assay mixture was the same as 
cd for the competition assay except unlabeled a- 
in was omitted. 

ic binding did not contain the enzyme 
ound less than 0.8% of the total input, 
'esults of competition assays for both 
ed wheat germ RNA polymerase and 
Lzyme present in crude homogenates are 
itcd in Fig. 8. The percent bound [^H]- 
initin was determined by calculating the 
irtion of counts bound for each concen- 
a of a-amanitin to the counts bound in 
es containing no unlabeled competing 
initin. The ideal curve is based on the 
ise in specific activity of the total ama- 
concentration at saturation, assuming 
I single binding species is present and 
oth compounds compete equally for the 
ig site. The experimental results for 
enzyme preparations closely approxi- 
the ideal curve. 

cussion. Previous studies have demon- 
d o-methyl-[^H]-demethyl-y-amanitin 
a powerful tool in studying the eukar- 
nucleoplasmic RNA polymerase (1, 2, 
However, the unavailability of this com- 
1 or of large enough amounts of a- 
itin to synthesize this derivative by the 
)usly reported method has restricted the 
pread application of this technique. 
; procedures presented in this paper 
es the synthesis of small amounts of 
^amanitin from readily available quan- 
3f starting materials by the introduction 
V nondestructive analytic techniques to 

ensure proper placement of functional groups 
in nonradioactive intermediates. Use of 
shorter columns of Sephadex LH-20 did not 
affect the desired resolution and the intro- 
duction of a short column of Sephadex G-SO 
ensures complete removal of any unreacted 
radioactivity in the end product. This is ver- 
ified by thin layer chromatography of the 
radioactive product and further substantiated 
by the saturation curve of [^H]-y-amanitin 
with purified wheat germ RNA polymerase 
B in which less than 3% of the total input 
remained unbound. The competition assay 
conclusively demonstrates that the final prod- 
uct binds to RNA polymerase in essentially 
the same manner as the unreacted parent 
compound and at the same site. 

Although the competition assay was de- 
signed as a test for reactivity and radiochem- 
ical purity of the final product, the usefubiess 
of this assay exceeds this purpose. Currently, 
studies of amanitin resistant RNA polymer- 
ase B have reUed on the inhibition of enzyme 
activity by various concentrations of ama- 
toxin to characterize wild type or mutant 
enzymes (6, 7). This technique cannot be 
applied to crude cell homogenates as the 
normally resistant RNA polymerases A and 
C as well as RNase would complicate the 
kinetic analysis. 

-n -IS -u -IS -10 -n -n -n -to -ss -so -4S -40 

Ln Mtiir Cmk. AMNtin 

Fig. 8. Competition binding assay. The ideal curve 

(O O) represents the percent of the total amatoxin 

present as o-methyl-[^H]-demethyl-Y-amanitin. The as- 
say was carried out in the presence of a constant amount 
of o-melhyl-['H]-demethyl-Y-amanitin (1.47 x 10* M) 
previously determined to be in excess required to saturate 
either the purified wheat germ RNA polymerase (2.63 
X \0~^ M) or the enzyme present in 0.0189 g of mouse 
kidney homogenate. Percent saturation was calculated 
from the total radioactivity bound in samples containing 
no unlabeled a-amanitin; wheat germ RNA polymerase 
( A ■ • • A); crude mouse kidney homogenate (# #). 



The amanitin competition assay could pro- 
vide a new method for studying the interac- 
tion of amanitin with resistant RN A polym- 
erase B enzymes. The assay is essentially free 
from interference by other RN A polymerases 
and is unaffected by RNase, therefore, crude 
homogenates as weU as purified enzyme prep- 
arations can be studied. In addition, the assay 
provides a means of direct measurement of 
dose-response over a wide range of concen- 
tration and could provide additional insight 
into possible mechanisms of amanitin resist- 

Summary, An improved method permitting 
the synthesis of a radioactive derivative of 
a-amanitin from a small amount of the com- 
mercially available parent compound has 
been developed. The labeled derivative was 
used in an amatoxin competition binding as- 
say designed to detect eukaryotic RNA po- 
lymerase B in either purified form or in crude 
homogenates. Both compounds are shown to 
compete for the same binding site and with 
approximately the same affinity. The com- 
petition assay proves to be both sensitive and 
highly selective for RNA polymerase B and 

provides a new, direct method for stu 
the enzyme-amanitin interaction o\ 
much broader range of concentration 
previously reported. 

We greatly appreciate the assistance of Ms. 
Abdou in preparation of illustrations. Ms. Donna 
bers for secretarial assistance, and Drs. John A 
strong, Patricia A. Craven and Mary Edmonds fc 
cally reviewing the manuscript. This work was sup 
through the Medical Research Service of the 
Veterans Administration. 

1. Cochet-Meilhac. M., and Chambon, P. Bi( 
Biophys. Acta. 353, 160 (1974). 

2. Cochet-Meilhac, M., Nuret, P., Courvalin, J. ( 
Chambon, P., Biochem. Biophys. Acta 35 

3. Wieland, Th.. and Fahrmeir. A.. Leibigs Ann. 

4. Wieland, Th., Pure Appl Chcm. 9, 145 ( 1964 

5. Dalgliesh, C. E., J. Clin. Pathol. 8, 73 (1955). 

6. Somers, D. G., Pearson, M. L., and Ingles, ( 
Biol. Chem. 250,4825(1975). 

7. Guialis, A., Beatty, B. G., Ingles, C. J., and ' 
M.M., Cell 10, 53 (1977). 

Received March 13, 1978. P.S.E.B.M. 1978, Vol. 


)metabolites of Leishmania donovani Promastigotes. I. Isolation and Initial 
Characterization (40293)^ 


neni of Zoology and Bureau of Biological Research, Rutgers University, Piscataway, New Jersey 08854 

3n and characterization of parasitic 
n exometabolites is of importance 
of the possible role these may play 
ost-parasite relationship. Since the 
of intracellular parasites do become 
id contaminated with host cell sub- 
solation, purification, and character- 
)f these parasite products becomes 

nania provides an ideal system where 
ence of exometabolites produced by 
itive intracellular protozoan parasite 
; host may readily be studied. Leish- 
jrganisms have two morphological 
le amastigote, an obligative intracel- 
m infecting vertebrates, and the pro- 
e, which exists extracellularly in the 
ctor and will grow readily in culture, 
ms have been reported to produce 
ase exometabolites which demon- 
tigenic identity (1, 2). 
ts to date dealing with substances 
imulate in the media in which pro- 
es are metabolizing (metabolized 
I have involved either undefined 

media containing blood proteins (1, 
t solutions (4, 5). In order to deter- 
iromastigote substances were present 
>olized Senekji's medium after pro- 
e growth, Clinton et al (3) utilized 
electrophoretic procedures, reacting 
zed medium against antiserum 

rabbits to the homologous promas- 
3ne band formed between the anti- 
id a substance from the metabolized 

ork was supported in part by a grant from the 
liversity Research Council and a Charles and 
jsch Memorial Fund Award to the Bureau of 
Research of Rutgers University. Dr. Nicholas 
is thanked for his continuing support and 
suggestions and advice. Dr. William B. Foster 
for his reading of the manuscript. 
. address: Department of Zoology, Morrill 
^ersity of Massachusetts, Amherst. Mass. 

medium. No reaction was observed when 
nonmetabolized medium was tested. Schnur 
et al (1) utilized metabolized Feinberg and 
Whittington's medium and reacted this with 
rabbit antipromastigote hyperimmune serum 
and demonstrated multiple bands (termed 
EF) by diffusion in gels. Since Schnur et al 
obtained their metabolized medium from cul- 
tures of promastigotes in log phase, they con- 
cluded that the EF substances were exome- 
tabolites and not products of lysis. In addi- 
tion, the molecular weight of the EF sub- 
stances was within the range of 25,000 and 
70,000, but they were not immunogenic when 
injected into rabbits. Decker and Honigberg 
(6), however, reported successful induction of 
antibodies in mice to the exometaboUte. Re- 
sults utilizing less defined media suggest that 
promastigotes of Leishmania produce exome- 
tabolites, but there is no agreement as to their 
number and immunogenicity (1, 3, 6). The 
lack of agreement in the data may be attrib- 
uted to the different media used to culture 
the promastigotes. 

Media used in in vitro culture should be 
defined and protein free to facilitate recovery 
of exometabolites more closely resembling 
the native form released from the parasite. 
Greenblatt and Glaser (4) used Locke's so- 
lution with glucose at 37° to maintain pro- 
mastigotes and found a variety of molecules 
including various amino acids, hypoxanthine, 
guanosine, uracil, and ribose in the medium. 
They did not detect any large molecules and 
concluded that the low molecular weight sub- 
stances found in the metabolized medium 
resulted from leakage and not gross lysis. On 
the other hand. Decker and Janovy (5) in a 
similar study detected not only small mole- 
cules but also proteins and RNA. Thus, while 
salt solutions may be ideal for recovery of 
leakage products from promastigotes, they 
may not adequately support complete metab- 
olism of the organisms. Measurable quanti- 
ties of larger molecular weight excretion- 


0037-9727 /!%/ \S9\-0\QSSQ\ .QKil^ 
W\ righu reserved. 


Leishmania donovani exometabolites, i. 

secretion metabolic products may not accu- 
mulate. On the other hand, the higher molec- 
ular weight products detected could be the 
result of lysis. 

More, recently Slutzky and Greenblatt (7) 
isolated a substance by degradative isolation 
techniques (boiling and 33% trichloroacetic 
acid solution) from proteid L. tropica metab- 
olized medium. The substance isolated was 
initially associated with medium protein, was 
immunologically active and carbohydrate 
rich. The isolated entity did not pass through 
a 30,000 mol wt exclusion membrane. Little 
or no protein was reported to be associated 
with the isolated entity. 

The object of this study was to isolate and 
characterize the metabolic by-products of L. 
donovani promastigotes in their native form. 
To accomplish this log phase promastigotes 
of L. donovani were maintained in protein 
free tissue cuhure medium to minimize the 
interference of lytic by-products. The metab- 
olized medium was then fractionated and 
examined spectrophotometrically and sero- 

Materials and methods. Amastigotes uti- 
lized to initiate promastigote cultures were 
obtained from the spleens of hamsters in- 
fected with the 3S strain of L. donovani (8). 
Spleens were homogenized in sterile phos- 
phate-buflered (pH 7.0) physiological saline 
and amastigotes isolated by diflerential cen- 
trifugation (9). All cultures were initiated at 
a density of S X 10^ organisms per ml and 
subcultures were made when the density of a 
culture reached 2x10^ promastigotes per ml. 
Promastigotes utilized to generate metabo- 
lized culture medium were never less than 4 
nor more than 15 subcuhures removed from 
the initial amastigote-seeded culture. All cul- 
tures were incubated with an atmosphere of 
5% CO2 in air at 25 ±0.1°. 

The culture medium utilized to grow pro- 
mastigotes (growth medium) consisted of 9 
parts Medium 199 with Hanks' salts (Gibco) 
and 1 part whole defibrinated rabbit blood 
(Pel Freeze). The blood was centrifuged (4 hr 
at 2000g) before inclusion into the medium 
to separate serum from cells. Serum was in- 
activated at 56° for 30 min and stored at 
—20° until used. Cells were washed in excess 
Hanks' balanced salt solution (Gibco) 5 times 
and lysed in a volume of double distilled 

water equal to lOx the packed cell vol 
Cell ghosts were removed by centrifug 
(24 hr at 200g) and the supernatant uti 
in the medium. To prepare 1 liter of gr 
medium, 100 ml of Medium 100 (1 Ox] 
added slowly to 500 ml of lysate and a : 
cient amount of double distilled water a 
to bring the volume to 950 ml. The pH 
maintained at 7.2 by addition of NaHC* 
needed. Serum (50 ml) was then added 
the medium sterilized by filtration throi 
0.22 /iMillipore filter. 

Medium used to maintain promasti 
(maintenance medium) consisted of Me 
199 with Hanks' salts (Gibco) and 25 
Hepes bufler (Sigma). The pH was adj 
to 7.2 with 1 N HCl or NaOH. The mi 
nance medium was sterilized as desc 

Promastigotes were allowed to metat 
both growth and maintenance media. Gi 
medium cuhures were initiated at a dc 
of 5 X 10^ organisms per ml. Whei 
promastigote density reached 8 X 10*^ p 
(mid log phase), the cultures were centriJ 
(20 min at 2000g) separating promasti 
from the medium. Organisms were wasl 
times in excess Hanks' balanced salt sol 
and resuspended in maintenance medii 
a density of 10^ promastigotes per ml. 
tures in maintenance medium were incul 
8 hr at 25° with a 5% CO2 in air atmosp 
Promastigotes were removed by centri 
tion (1 hr at 2000g) and the metabc 
medium was fihered through a 0.22 /il 
pore filter, concentrated lOx by lyoph 
tion and stored at -20°. 

Two ml aliquots of lOx concentrate< 
tabolized maintenance medium were 
tionated on a column (1.6 x 80 cm) of 
fine grade Sephadex G25 (Pharmacia), 
volume was 54 ml, bed volume was 1( 
and flow rate was 7 ml per hr. The e 
used was a 5% acetic acid solution in dis 
water. The colunm was characterized i 
ing a-melanocyte stimulating hormone 
wt 1910; Brady kinin, mol wt 1204; and g 
pentapeptide, mol wt 768, all purchased 
Calbiochem. Elution values were 94, 12( 
145 ml respectively. Each standard wa 
phed to the column as a 1 ml vol conta 
50 jug peptide. Elution volume was < 
mined from the maximum of the el 

Leishmania donovani exometabolites, i. 


Llution values for the standards 
against the log of their molecular 

approximate a straight line. 

nl fractions were collected from the 
and analyzed on a Beckman DB 

24 spectrophotometer. Absorption 

were obtained between 190 and 350 

mount of peptide present in a fraction 
nated photometrically by the method 
er and Miller (10). Fractions were 
Eed to dryness and redissolved in sol- 
IS M NaF in glass double distilled 
The blank contained solvent only, 
ince was measured at 193 nm and a 
i curve generated using bovine serum 
I, a-mclanocyte stimulating hormone, 
^entapeptide and Bradykinin (Calbi- 
The standard curve developed here 
istinguishable from that presented by 
and Miller with 11 /ig/ml protein 
an absorbance value of 0.7. Direct 
onality between concentration and 
nee was applicable for all standards 
absorbance of 0.7. Since all fractions 
1 protein amount was estimated had 
nee values greater than 0.7, aliquots 
3ns were diluted with solvent until an 
nee value of 0.7 was attained. The 
of protein in a fraction was calculated 
plying the dilution factor x 1 1 /ig/nil. 
imount of sugar present in fractions 
^rmined by the procedure of Dubois 
1). A standard curve for D-galactose 
lerated which was indistinguishable 
at presented by Dubois et al The 
value for triplicate samples contain- 
g of D-galactose was 0. 1 1 absorbance 
490 nm. 

fic chemical tests for tryptophan and 
were performed on selected fractions 
procedure of Fischl (12) and the 
of Udenfriend and Cooper (13) as 
i by Massin and Lindenberg (14) re- 
ly. Controls were composed of 50 
olutions of tryptophan, tyrosine and 

L donovani promastigote immune se- 
\ raised in rabbits by injecting a ho- 
le composed of freeze-thawed pro- 
es in saline and FCA (1:1). Each 
jceivcd a total volume of 1 ml, con- 
21 mg ^ (determined by Kjeldahl 

procedure [Campbell et al, 15]) delivered in 
0. 1 ml aliquots at one time to 8 sc sites on the 
back and 2 im sites in the hind legs. The 
animals were bled 30 days after immuniza- 
tion. Serum was recovered by centrifugation 
(1 hr at 2000g) and stored at -20°. 

Test antigens were prepared by mixing 
(6:1) metabolized maintenance medium (free 
of serum) with nonmetabolized growth me- 
dium (containing serum) and concentrating 
tenfold by lyophilization. Control antigens 
were nonmetabolized growth medium and 
nonmetabolized maintenance medium pre- 
pared in the same manner. 

The microsolutes in each sample were ex- 
changed by diafUtration (16) and standard- 
ized using a 500 mol wt cutofT ultrafiltration 
membrane (UM 05) with a Model 12 stirred 
cell (Amicon). Five sample volumes of bar- 
bital buffer (17) were exchanged with a pre- 
dicted 994-% complete exchange of micro- 
solute (16). 

Gel diffusion plates were prepared by 
pouring 10 ml melted agar solution ( 1% Difco 
Bacto Agar in barbital buffer [17] with 0.1% 
sodium azide) into a 9 cm-diameter petri dish. 
Wells (5 mm O.D.) were cut in the agar 7.5 
mm apart (center to center). After the wells 
were filled with either antiserum or antigen 
solution, the plates were incubated 48 hr in a 
humid atmosphere at 25°. Precipitin lines 
appeared within I to 2 days but were allowed 
to develop for a total of 4 to 7 days. Gels 
were washed free of nonreacting protein with 
barbital buffer (17) for 48 hr (4 changes of 
buffer) and stained wet with a saturated so- 
lution of picric acid in 1% acetic acid. 

Results, When promastigote metabolized 
and nonmetabolized growth media were 
tested against rabbit antipromastigote im- 
mune serum by gel diffusion, the metabolized 
medium reacted forming multiple precipitate 
bands. This confirmed earlier reports that 
exometabolites were present in the metabo- 
lized growth medium and would react with 
specific antiserum (1, 3). When promastigote 
metabolized maintenance medium was tested 
against the same antiserum, no reaction oc- 
curred. This suggested that the presence of 
serum protein was necessary for the exome- 
tabolite to react with antibody. 

To determine whether serum protein was 
indeed essential for formation of specific pre- 


Leishmania donovani exometabolites, i. 

cipitates, metabolized and nonmetabolized 
maintenance media were mixed with non- 
metabolized growth medium, the microsolute 
environment standardized, and reacted with 
immune serum. The mixture containing me- 
tabolized maintenance medium yielded mul- 
tiple precipitate bands identical to the ones 
observed when metabolized growth medium 
was used as the reacting antigen (Fig. 1). No 
reaction occurred with the nonmetabolized 

When metabolized maintenance medium 
was fractionated, spectrophotometric analysis 
at 274 nm revealed two major fractions (A 
and B) (Fig. 2) with elution values of 101 and 
122 ml respectively. Ultraviolet absorption 
spectra of these major fractions from 190 to 
350 nm are shown in Fig. 3. None of the 
fractions from nonmetabolized maintenance 
medium demonstrated either of the major 
peaks shown in Fig. 2. 

When all fractions collected after column 
chromatography of either metabolized main- 
tenance medium or nonmetabolized mainte- 
nance medium were mixed with nonmetabo- 
lized growth medium and tested against an- 
tipromastigote immune serum, only Fraction 
B reacted to form precipitate bands (Fig. 1). 
These precipitate bands demonstrated reac- 
tions of identity with those formed against 
antipromastigote immune serum using pro- 
mastigote metabolized growth medium as the 

Fig. 1. Gel diffusion plate depicting reactions of pro- 
mastigote metabolized growth medium (well A), concen- 
trated promastigote metabolized maintenance medium 
mixed with nonmetabolized growth medium (well B), 
concentrated promastigote metabolized maintenance 
medium (well C) and concentrated nonmetabolized 
maintenance medium mixed with nonmetabolized 
growth medium (well D) against rabbit antipromastigote 
immune serum (well E). 



90 110 



Fig. 2. ScphadexG25 gel filtration profile of metab- 
olized maintenance medium at 274 nm. 





340 a*o ato 

WAVillNOfH IN nm 

Fig. 3. Ultraviolet absorption spectra of gel filtration 
Fraction A ( ) and Fraction B ( ), pH 7.2. 

reacting antigen. When Fraction A (10 ab- 
sorbance units) was mixed with the antiserum 
prior to reaction with Fraction B, no evidence 
of neutralization was observed. 

Fraction A and B samples with absorption 
values of 3.1 (at 293 nm) were estimated to 
contain approximately 48 fig peptide and 10 
jbtg sugar per ml. Fractions from the column 
which eluted both immediately before and 
after Fractions A and B were determined not 
to contain sugar. 

Discussion. The results suggest that at least 
two low molecular weight substances arc re- 
coverable from promastigote metabolized 
protein free medium during the log phase 
growth of the organisms. No high molecular 
weight substances were detected as might 
have been anticipated if the recovered sub- 
stances were the result of promastigote lysis. 
Microscopic examination of log phase cul- 
tures revealed no lysed organisms suggesting 
that recovered substances are indeed exome- 
tabolites and not products of autolysis. Spec- 

Leishmania donovani exometabolites, i. 


ita shown in Fig. 3 suggest the presence 
(tide bonds (10) with tyrosine present in 
on A and tryptophan in Fraction B 
rhc presence of these amino acid resi- 
was confirmed by colorimetric proce- 
Detection of sugar in Fractions A and 
jests that the substances may be glyco- 
les. Since the molecules appear to be of 
olecular weight and the ratio of protein 
;ar is approximately 5:1, the carbohy- 
cntity is most likely composed of only 

) molecular weights of the substances in 
ons A and B appear to be in the range 
1-1900 in that their elution values were 
lediate between those of gastrin penta- 
le (mol wt 768) and a-melanocyte stim- 
g hormone (mol wt 191 1) (see Andrews 
It is premature at this time to assign a 
precise molecular weight. The estimated 
ular weight of recovered substances 
Its glycopeptides composed of from 5 to 
lino acid residues. Peptides of this size 
1 be expected to act as simple haptens 

lerally low molecular weight substances 
t induce an imune response unless con- 
d to a larger carrier molecule (21). The 
gation of low molecular weight material 
>rotein carrier endows that conjugated 
n with multivalency with respect to the 
aic moiety (20). The exometabolites ap- 
:o act as monovalent haptenic groups, 
lata suggest that the simple substance 
^d in Fraction B attaches to sites on the 
n molecule making the conjugated mol- 
multivalent with respect to that site and 
iblc to form precipitates when reacted 
mtipromastigote immune serum. This 
rotation is supported by the fact that 
inds formed with the promastigote me- 
zed growth medium are identical to 
observed when the substance in Frac- 
I is mixed with protein. The substance 
iction A did not form precipitates when 
against antipromastigote immune se- 
This may have occurred because no 
ilent entities formed or because there 
isufficient antibody present specific for 

J exometabolite produced by L. tropica 
^n reported to be a carbohydrate-rich 
ince that does not pass through a 30,(XX) 

mol wt exclusion membrane (7); however, it 
has been demonstrated to be adsorbed ini- 
tially to medium proteins. While it is not 
impossible that L. donovani and L. tropica 
produce physically distinct exometabolites, 
the major differences reported may result 
from the method of isolation. The L. donovani 
exometabolite reported here was isolated by 
gentle procedures under mild conditions 
while Slutzky and Greenblatt utilized more 
harsh procedures. 

Fraction B exometabolite is released by 
both amastigotes and promastigotes as evi- 
denced by the fact that reactions of identity 
occur when promastigote metabolized growth 
medium and amastigote infected spleen ho- 
mogenate supernatant react with antipromas- 
tigote immune serum (2). Leishmania dono- 
vani promastigote metabolized growth me- 
dium has been used as a vaccine and induced 
specific protection against amastigote chal- 
lenge (22). If the protective substance in me- 
tabolized medium is a conjugated antigen, 
then Fraction B exometabolite may be the 
antigenic determinant responsible for the 
protection. Work is proceeding to determine 
if Fraction B, after conjugation to a protein 
carrier, will act as an immunogen and induce 
specific protection. 

Summary. Two exometabolites have been 
demonstrated to accumulate in protein free 
culture medium in which log phase promas- 
tigotes of L. donovani dire metabolizing. These 
molecules demonstrate gel filtration charac- 
teristics suggesting a molecular weight in the 
range of 8(X)-19(X). The ultraviolet absorption 
spectra of the exometabolites suggest the 
presence of peptide bonds with tyrosine pres- 
ent in one and tryptophan in the other. Sugar 
was demonstrated to be associated with both 
Fractions A and B, suggesting the exome- 
tabolites are glycopeptides. The exometabo- 
lite in Fraction B did not react with specific 
antibody to form precipitates unless it was in 
combination with serum protein. The data 
strongly suggest that the exometabolite con- 
jugates with protein forming a multivalent 

1. Schnur, L. F., Zuckerman, A., and Greenblall,C. L., 
IsraelJ. Med. Sci. 8,932(1972). 

2. Scmprcvivo, L. H., Ph.D. Thesis, Rutgers University, 
New Brunswick, New Jersey (1975). 


Leishmania donovani exometabolites. 

3. Clinton, B. A.. Palczuk. N. C, and Stauber, L. A., J. 
Immunol. 108, 1570(1972). 

4. Greenblatt, C. L., and Glaser, P., Exp. Parasitol. Ki, 

5. Decker, J. E., and Janovy, J., Jr., Comp. Biochem. 
Physiol. 498,513 (1974). 

6. Decker, J. E., and Honigberg, B. M., J. Parasitol. 

7. Slutzky, G. M., and Greenblatt. C. L., FEBS Lett. 

8. Stauber, L. A., Exp. Parasitol. 18, 1 (1966). 

9. Clinton, B. A., Ph.D. Thesis, Rutgers University, 
New Brunswick, New Jersey ( 1969). 

10. Mayer, M. M., and Miller, J. A., Anal. Biochem. 3<i, 

11. Dubois. M., Gilles, K. A., Hamilton, J. K., Rebers, 
P. A., and Smith, F., Anal. Chem. 28, 350 (1956). 

12. Fischl, J., J. Biol. Chem. 235, 999 (I960). 

13. Udenfriend, S., and Cooper, J. R., J. Biol. Chem. 
1%, 227 (1952). 

14. Massin, M., and Lindenberg, A. B., Bull. Soc. Chim. 

Biol. 39, 1201(1957). 

15. Campbell, D. H., Garvey, J. S., Cremer, N. E., and 
Sussdorf, D. H., ''Methods in Immunology," 2nd 
ed., 455 pp. W. A. Benjamin, Inc., Massachusetts 

16. Blatt, W. F., Robinson, S. M., and Bixler, H. J.. 
Anal. Biochem. 26, 151 (1968). 

17. Alberty, R. A., in *The Proteins" (H. Neurath and 
K. Bailey, eds.). Vol. I, Part A, p. 461. Academic 
Press, New York (1953). 

18. WeUaufer, D. B., Adv. Prot. Chem. 17, 303 (1962). 

19. Andrews, P., Meth. Biochem. Anal. 18, I (1970). 

20. Landsteiner, K., 'The Specificity of Serological R^ - 
actions,*' 330 pp. Charles C. Thomas, Springfield, « 
Illinois (1936). 

21. AbramofT, P.. and La Via, M. F., "Biology of the 
Immune Response," 492 pp. McGraw-Hill Book ; 
Company, New York (1970). 

22. Semprevivo, L. H., J. Parasitol. 63(Suppl.),43 (1977). 

Received January 9, 1978. P.S.E.B.M. 1978, Vol. 159. ; 


m of Acyclic and Cyclic A/-Nitrosamines by Cultured Human Colon ^ (40294) 

e Studies Section, Experimental Pathology Branch, Carcinogenesis Program, National Cancer Institute, 
farmland 20014 and Department of Pathology, School of Medicine, University of Maryland, Baltimore, 

Maryland 21201 

o compounds are a major class of 
ircinogens which are candidates to 
lan cancers (I). ^-Nitrosamines 
detected in ambient air over cer- 
areas (2), in tobacco smoke (3), 
rinks (4, 5). Furthermore, they can 
in vivo by the reaction of nitrite 
ible amines under acid conditions, 
he stomach (6). They may also be 
enteric bacteria e.g. E. coli in situ 
osamines require metabolic acti- 
xert their mutagenic and carcino- 
ity (8-10). This requirement of 
X)uld, in part, explain the organo- 
:inogenicity of the N-nitrosamines 
rthermore could affect an individ- 
)tibility to the carcinogenic action 

ental systems to study carcinogen- 
y in human epitheUa are being 
(11, 12). We have previously 
; cultured human colon can acti- 
rcinogens from several chemical 
;. polycyclic aromatic hydrocar- 
ylhydrazine, N-nitrosamines, into 
> which bind to cellular macro- 
(13). We now report metabolic 
everal aliphatic N-nitrosamines in 
iman colonic mucosa. 
s and methods. Non-tumorous hu- 
c tissues were collected at the time 
rgery or "immediate" autopsy (14) 
al of 11 patients; 7 with and 4 
icer of the colon. The tissues were 
y put in sterile containers on ice 
sed in L-IS medium within 15 min 
^al from the patient and kept at 4° 
hr until cultured. The specimens 
nto squares (0.5 x 0.5 cm) and 
previously described (13). 
hrs in culture, one of the following 

i in part by Grant No. NO I CP 43237 from 
dancer institute. 

[*^C]labeled N-nitrosamincs (New England 
Nuclear, Boston, MA) was added to the cul- 
ture media to give a concentration of 100 /xA/: 
[*^C]Dimethylnitrosamine [35 mCi/nmiole; 
prepared on NCI contract NOl-CP-55677 
and purified by the method of den Engelse et 
al (15); N'Y^C- 1 -ethyl]diethyhiitrosaminc 
(14.5 mCi/mmole); ^-pC-2,6]nitrosopiperi- 
dinc (18.8 mCi/mmole); ^-[^^C-2,5]nitroso- 
pyrrolidine (16.2 mCi/nmiole); ^-[^H-3,4]- 
nitrosopyrroUdine (5 mCi/nmiolc). NyN' 
[*^C(U)]dinitrosopiperazine (16.5 mCi/mole); 
N' [ pyrrolidine -^^ C - 2] nitrosonornicotinc 

Five explants per experimental variable in 
three sterile 60 mm plastic Petri dishes (Fal- 
con Plastics, Oxnard, CA) were placed on a 
rack in a closed container (Nalgene plastic 
jar, 500 ml) which was modified with two 
ports for replacing air with 95% ©2-5% CO2 
(16). The containers were placed on a rocker 
platform and rocked approximately 10 cycles 
per minute for 24 hr. In order to remove "C- 
CO2 formed by the metabolism of the N- 
nitrosamine the containers were flushed with 
N2 for 5 min and the CO2 absorbed in two 
tubes each containing 8 ml 0.2 A/ Ba(OH)2. 
After removal of the explants, I ml 3M phos- 
phoric acid (pH 3) was added to each culture 
dish to release CO2 dissolved in the media. 
After 4 hr at 37°, the containers were then 
flushed with N2 for another 5 min. 

The tissue culture medium was transferred 
to a reaction flask (Kontes Glassware, Vine- 
land, NJ) the sidearm of which contained a 
small vial with 0.5 ml 4N KOH, and oxidized 
by HgCl2 (100 mg/ml) at 90° for I hr (15). 
The KOH-solution was added to the 
Ba(OH)2-solution. The precipitate was col- 
lected on Whatman GF/C filters and washed 
with absolute ethanol until the count in the 
washing solution was negligible. Medium 
without explants of colon served as control. 
The precipitate and filter were suspended in 


0037-9727 /7%/ \S9\4i\\\Vi\ ."^JlWi 

Copyright ® V91% b^ t!h« Sooiei^ \ot ^xv«\tcwesAa\%vc:ft<Q^ wcA>Kt«o«B 
AU hghu reserved. 



3 ml water and 10 ml Aquasol liquid scintil- 
lation cocktail (New England Nuclear, Bos- 
ton, MA) and counted. 

The mucosa was scraped from the explant, 
and DNA and protein isolated by the phenol 
extraction procedure. DNA was purified on 
a CsCl-gradient and the binding level mea- 
sured as previously described (17). Binding to 
protein was also assayed (13). One explant 
from each variable was fixed in 3% glutaral- 
dehyde buffered with 0. 1 M s-coUidine (pH 
7.4) and prepared for light microscopy (18). 

DNA, isolated from a total of 54 explants 
(pooled from three cases), was hydrolysed 
with 0.1 M HCl at 70° for 1 hr and bases 
were isolated by high-pressure liquid chro- 
matography (Column: Durrum DC 1-A; 15 
X 0.21 cm; Durrum Chemicals, Sunnyvale, 
CA; Solvent: 0. 1 M ammonium formate, pH 
4.5; Flow rate: 0.6 ml/min). Markers for N-7 
and 0-6 methylguanine were added to the 
hydrolyzed DNA; the elution was monitored 
at 254 mA/ and 0.4 ml fractions were col- 

lected. The radioactivity was measured by 
liquid scintillation methods. The material 
eluting in the void volume (90% of the radio- 
activity) was treated with cone, perchloric 
acid at 1(X)° for 1 hr and methanol removed 
by vacuum-distillation and the radioactivity 
was determined. 

Results. Formation of *^C-C02 after incu- 
bation of A^-nitrosamines with human colon 
indicates that cultured human colonic mu- 
cosa is able to metabolize both acyclic N- 
nitrosamines (Table I), such as dimethylnitro- 
samine (DMN) and diethylnitrosamine 
(DEN), and cyclic ^-nitrosamines (Table II). 
Variation in the ability to metabolize cyclic 
A^-nitrosamine was observed among individ- 
uals. Under these test conditions only N-m- 
trosopyrrolidine (NPy) was metabolized by 
all cases studied, iV,W-dinitrosopiperazinc 
(DNP) by five cases and A^-nitrosopiperidinc 
only by one case. No ^^C-C02 was formed 
from A^-nitrosonomicotine possibly due to 
the chemical structure (the C-14 labeled atom 

TABLE I. Metabolism of yv,iV-DiALKYLNiTROSAMiNES by Cultured Human Colon." 



















































• Colonic explants were cultured in chemically defined media for 24 hrs and the ('*C]labellcd yv-niirosamines 
were added at a concentration of 100 fiM to groups of five explants for 24 hr. 

*/?moles nitrosamine bound per mg of either DNA or protein, single determination, 
ymoles '^C-COi formed per mg DNA. 
•^N.D. « not detectable. 

TABLE 11. Metabolism of Cyclic ^V-Nitrosamines by Cultured Human Colon." 







COr forma- 






































































































"Colonic explants were cultured in chemically defined media for 24 hr and the ("Cllabelled ^-ni^ro^amines were added at a concentration of IOf> 
;imolcs to group> of five explants for 24 hr. 

'dpm per 100 /tg of either DNA or protein. Mngle determination. 

■ pmoles "C-CO.. formed per mg DNA. 

'' Incubated 

• N.D. - none detectable. 



ly one C-H bond)— but nonlabelcd 
iild have been formed from other car- 
)ms in the pyrrolidine ring. Only 
ind NPy consistently formed alkylat- 
ctics which reacted with cellular DN A 
:ascs. DMN, DEN, NPy, and DNP 
:o protein; when compared to the other 
samines high binding levels of DNP 
isLT protein were observed. The bind- 
a in Table II is given as either dpm 
I jug DNA or dpm per 100 /ig protein 
xact chemical structure of the adducts 
between the iV-nitrosamines and the 
nolecules are unknown at the present, 
ive correlation (r = 1.00) was found 
a alkylation of DNA by DMN and 
rmation, while NPy did not show any 
tion (r = 0.24, /? > 0. 1). No correlation 
a DMN and NPy binding to protein 
32-formation was found (r = 0.14, /? 
and r = 0.41, /? > 0.1, respectively), 
alkylated DNA in both N-7 and 0-6 

I of guanine (Table III). However, 
r the radioactivity was associated with 

II in the initial peak. Treatment of this 
il with cone, perchloric acid released 
OH (40% of radioactivity). The mor- 
Y of the explants, as monitored by high 
on light microscopy, showed good 
ation in all the reported cases. 
ission. A^-Nitroso compounds induce 
in many animal species (10) and have 
nplicated in causing human cancers 
•Nitrosamines rarely induce colonic 
in experimental animals. However, N- 
mides such as ^-methylnitrosourea 
id iV-methyl-A^'-nitro-A^-nitrosoguani- 
(0), caused colo-rectal carcinoma in 
; when applied intrarectally. 
arcinogens require metabolic activa- 

exert their carcinogenic efTect (21). 
:inogens, impUcated in colon carcino- 
, could be activated enzymatically in: 
ins other than the colon and reach the 

III. Methylation of Human Colonic DNA 
BY ['*C]DMN. 




Initial peak 


ibers in parentheses, percentage of the total 
>f dpm added to the column. 

target tissue via the blood circulation (22); (b) 
the intestinal lumen by deconjugation of me- 
tabolites by the microflora (23, 24); and (c) 
the intestinal mucosa by various enzymes e.g. 
the mixed-function oxidases (24). We have 
previously shown that both human and rat 
colonic mucosa in culture can activate pro- 
carcinogens into metabolites that bind to 
DNA; explants of human colon can metabo- 
lize DMN, 1,2-dimethylhydrazine and benzo- 
[fllpyrene (B?) (13, 25). This observation sug- 
gests the importance of the third pathway 
described above. 

A 50-fold inter-individual variation was 
found in the binding of DMN to human 
colon DNA, lower than the 100-fold variation 
observed in the binding levels of BP to DNA 
in cultured human colon (26) and the 75-fold 
variation in the binding levels of BP to DNA 
in cultured human bronchus (27). Several 
factors for this variation were considered. The 
/>t/r£z-individual variation due to the meth- 
odology was minimal, i.e., coefficient of var- 
iation 0.1 (13). The viabiHty of the tissue as 
monitored by high-resolution microscopy 
was good in all the reported cases; however, 
changes in cellular physiology could, in part, 
account for some of the observed differences. 
There is a positive correlation between the 
level of radioactivity associated with DNA 
and C02-formation. Alkylation took place at 
both the 0-6 and N-7 position giving a ratio 
of 0.5. However, this radioactivity only ac- 
counted for a small part of the total radioac- 
tivity. Treatment of the material in the initial 
peak with strong acid, released about 40% of 
the radioactivity in form of methanol, indi- 
cating that the major alkylation site could 
either be the phosphate groups or the oxygens 
in thymidine and/or cytosine. This finding 
however requires further investigtion. Incor- 
poration of *"*€ from *'*C-C02 in the purine 
ring of the nucleic acids by de novo synthesis 
could also account for some of the radioactiv- 
ity associated with DNA (13). Human liver 
slices (28) and human bronchus (29, 30) are 
also able to metabolize DMN into CO2 and 
alkylating species which reacted with DNA. 
DMN has been shown mainly to alkylate the 
0-6 and N-7 positions of guanine in DNA 
(31); the ratio of methylation of 0-6 to N-7 
being nearly 1.1 in cultured human bronchus 
(30), while a lower ratio was found in animal 
experiments (32). 



The ability of the colon to metabolize the 
diflerent N-nitrosamines varies among indi- 
viduals. While colon from all investigated 
cases could metabolize DMN, only two cases 
could metabolize DEN into metabolites 
which reacted with DNA. Since the [*^C]- 
atom is located at the two-position of the 
ethyl group the alkylating moiety can be 
deduced as being an ethyl group. NPy was 
also metabolized by colon from all the cases. 
Binding of both ^H- and ['^C]NPy suggests 
that an adduct(s) is formed between a metab- 
olite of NPy and DNA. Opening of the ring 
in NPy indicated by C02-formation suggests 
that several possibilities for alkylating species 
exist. Lack of correlation between alkylation 
of DNA by NPy and C02-formation could 
also implicate a more complex pattern of 
metabolism. It has been suggested that two of 
the reaction-products between NPy and nu- 
cleic acids are 7-(2-carboxy)ethylguanine 
and/or 7-methylguanine (33). However, a re- 
cent observation indicates that the alkylation 
species could be 3-formyl-l-propanediazo- 
hydroxide (34). The molecular structure of 
the DNA adduct in human colon is under 
investigation. Formation of *^C-C02 in vivo 
by rats injected with either 2,5-[^^C]NPy or 
3,4-[^^C]NPy shows that ring oxidation occurs 
at both two and three positions (33). DNP 
had a high binding level to protein, while 
binding to DNA was only observed in one 
case. This observation of a high level of pro- 
tein binding is similar to our results from 
cultured human bronchus (16). 

N-nitrosamines may reach the colonic mu- 
cosal epithelial cells by several routes, where 
they could be metabolically activated. DMN 
has been detected in the blood of people 
ingesting both spinach and bacon; spinach is 
recognized as a rich source of nitrate/nitrite 
(35). N-nitrosamines have also been detected 
in the feces of human subjects, whose diet did 
not contain any detectable N-nitrosamines 
indicating that the compounds were formed 
in situ (36). 

The etiology of human colonic cancer is a 
complex problem. No exogenous chemical 
compounds have been so far proven to cause 
this carcinoma in the human. Our observa- 
tions, that human colonic mucosa can acti- 
vate several types of procarcinogens {e.g. BP, 
7, 1 2-dimethylbenz[a]anthracene, 1 ,2-dimeth- 

ylhydrazine and aliphatic N-nitrosamines) 
into forms that bind to DNA, suggests that 
the colon should be added to the list of organs 
which are likely to be susceptible to the car- 
cinogenic action of these compounds. 

Summary. Cultured human colon mucosa 
was found to metabolize both acyclic and 
cyclic N-nitrosamines as measured by '^C- 
CO2 formation and reaction of the activated 
moieties with cellular macromolecules. Di- 
methylnitrosamine and N-nitrosopyrrolidinc 
were metabolized by explants from all pa- 
tients studied. A positive correlation between 
binding of dimethylnitrosamine to DNA and 
C02-formation was observed. DMN alkyl- 
ated DNA in both 0-6 and N-7 position of 
guanine. However, most of the radioactivity 
was associated with an acid labile compound. 
High binding levels of N,N'-dinitrosopipcra- 
zine to protein without concomitant binding 
to DNA were detected. Inter-individual var- 
iation in both binding level to DNA and 
ability to metabolize the different N-ni- 
trosamines was observed. 

We would like to thank Drs. U. Saffiotti. G. Stoner. 
and T. Bowden for valuable comments. Ms. R. Schwartz 
for technical assistance, and Mrs. M. Bellman for secre- 
tarial assistance. 

1. Lijinsky, W., and Epstein. S. S., Nature (London) 
225,21 (1970). 

2. Fine, D. H., Rounbehler, D. P., Belcher, N. M., and 
Epstein, S. S., Science 192, 1328 (1976). 

3. McCormick, A.. Nicholson. M. J., Baylisma. M. A., 
and Underwood, J. C, Nature (London) 244, 237 

4. Warthesen, J. J., Bills, D. D., Scanlan, R. A., and 
Libbery, L. M., J. Agr. Food Chem. 24, 892 (1976). 

5. Sen, N. P.. Seaman, S.. and Miles, W. F., Fed. 
Cosmet. Toxicol. 14, 167 (1976). 

6. Sander, J., Arch Hyg. 151, 23 (1967). 

7. Hashimoto. S., Yokokura, T., Kawai. Y.. and Mulai. 
M., Fed. Cosmet. Toxicol. 14, 553 (1976). 

8. Magee, P. M., Monlesano, R., and Preussmann. R.. 
in "Chemical Carcinogenesis" (C. Searle, ed). p. 
449, American Chemical Society. Washington, DC. 

9. Monlesano, R., and Bartsch, H.. Mut. Res. 32, 179 

10. Druckrey, H., Preussmann, R., Ivankovic, S.. 
Schmahl, D., Afkham. J., Blum, C, Mennel. H. D.. 
Muller. M., Peiropoulos, P., and Schneider. H., Z 
Krebsforsch. 69, 103(1967). 

11. Harris, C. C, Beitr. Pathol. 158, 389 (1976). 

12. Harris, C. C, Aulrup, H., Stoner, G. D., and Trump. 



'Pathogenesis and Therapy of Lung Can- 

:. Harris, ed.), p. 559. Marcel Dekker, Inc., 


i., Harris, C. C, Stoner, G. D., Jesudason, 

rnimp, B. F., J. Nat. Cancer Inst. 59, 351 

i. F., McDowell, E., Barrett, L. A., Frank, 

id Harris, C. C, in "Experimental Lung 

(E. Karbe and J. J. Park, eds.) p. 548, 

Vcriag, New York (1974). 

else, L., Gebbink, M., and Philippus, E.,, 133(1975). 

. C, Autrup, H., Stoner, G. D., McDowell 

ump, B. F., and Schafer, P., J. Nat. Cancer 


"., Genta, V., Frank, A., Kaufman, D., 

L.. McDowell, E., and Trump, B., Nature 


., Wright, E., and Harris, C. C, in "Methods 

hysiology" Vol. 8 (D. Prescott, ed.), p. 277, 

c Press, New York (1974). 

I, T., and Weisburger, J. H., Proc. Soc. Exp. 

, 166(1975). 

I, T., Sato, T., Hayakawa, M., Sakuma, A., 

ano, H.,Gann62,231(1971). 

A., Cancer Res. 30, 559 (1970). 
1, R. L., Singh, D. V., and Nigro, N. D., 
es. 35, 1369(1975). 

A. G., and Drasar, B. S., Nature (London) 
;cr, J. H., Dis. Col. Res. 16, 431 (1973). 

25. Autrup, H., Harris, C. C, Fugaro, S., and Selkirk, 
J., Chem.-Biol. interactions, 18, 337 (1977). 

26. Autrup, H., Barrett, L. A., Jackson, F. E., Jesudason, 
M. L., Stoner, G., Phelps, P., Trump, B. F., and 
Harris, C. C. Gastroenterology, 74, 1248 (1978). 

27. Harris, C. C, Autrup, H., Connor, R., Barrett, L. A., 
McDowell, E. M., and Trump, B. F., Science 194, 

28. Montesano, R., and Magee, P. M., Nature (London) 
228, 1773(1970). 

29. den Engelse, L., Gebbink, M., and Emmelot, P., 
Chem.-Biol. inter. 11, 534 (1975). 

30. Harris, C. C, Autrup, H., Stoner, G. D., McDowell, 
E. M., Trump, B. F., and Schafer, P., Cancer Res. 

31. Lawley, P., in "Screening Tests in Chemical Carci- 
nogenesis" (R. Montesano, H. Bartsch, L. Tomatis, 
eds.) p. 181, International Agency for Research on 
Cancer, Lyon (1976). 

32. Margison, G., Margison, J., and Montesanno, R., 
Biochcm. J. 157,627(1976). 

33. Kruger, F. W. and Bertram, B., Z. Krebsforsch. 83, 

34. Hecht, S. S., Chen, C-B. B., and Hoffmann, D. 
Cancer Res. 8, 215 (1978). 

35. Fine, D. H., Ross, R., Rounbchler, D. P., Silvergleid, 
A., and Song, L., Nature (London) 265, 753 (1976). 

36. Vargheesc, A. J., Land, P., Furrer, R., Bruce. W. R., 
Proc. Amer. Assoc. Cancer Res. 18, 80 (1977). 

Received September 1, 1977. P.S.E.B.M. 1978, Vol. 159. 


Effects of Thyroxine, Epinephrine and Cold Exposure on Lipolysis In Genetically Obese 

(ob/ob) Mice' (40296) 


Department of Nutrition, Harvard School of Public Health, Boston, Massachusetts 02 US 

Mayer and Barmett (1) observed that ge- 
netically obese (ob/ob) mice were unable to 
withstand exposure to a cold environment 
and that the administration of thyroid hor- 
mone before cold exposure slightly prolonged 
their survival. More recent studies provided 
evidence that ob/ob mice were hypothyroid 
(2, 3) and that thyroid hormone administra- 
tion corrected the observed hypothermia dur- 
ing cold exposure (4). Experimentally pro- 
duced hypothyroidism in rats prevents nor- 
mal epinephrine-stimulated lipolysis (5), and 
in vitro studies using adipose tissue from 
ob/ob mice have demonstrated a similar re- 
duction in epinephrine-stimulated free fatty 
acid (FFA) release (6-8). These observations 
support the hypothesis that hypothyroidism 
in ob/ob mice results in defective lipolysis, 
thus limiting FFA as a substrate for thermo- 
genesis during cold exposure. However, in 
vivo studies in ob/ob mice at ambient tem- 
perature failed to show defective lipolysis 
either in response to catecholamines (9) or 
during fasting (10). 

In order to examine this apparent discrep- 
ancy between the in vivo and in vitro data in 
the literature and to study the metabolic ef- 
fects of cold stress in ob/ob mice more pre- 
cisely, an experiment was designed to inves- 
tigate the hormonal influences on lipolysis 
and the relevant parameters of carbohydrate 
metabolism during cold exposure in these 
animals. Specifically, this study measured the 
effects of pharmacological doses of thyroxine 
( Ta) on both in vivo and in vitro FFA release 
in cold-exposed ob/ob mice in comparison to 
the effects in non-obese mice. In addition, the 
effect of Ta treatment on epinephrine-stimu- 
lated FFA release from adipocytes was as- 

' Research support was provided by USPHS National 
Institutes of Health Grant Nos. AM-02911 and AM- 
00106 and the Fund for Research and Teaching, De- 
partment of Nutrition, Harvard School of Public Health. 

Materials and methods, Male weanling 
mice of the obese strain, C57BL/6J-ob were 
purchased from Jackson Laboratories, Bar 
Harbor, ME. The obese (genotype, ob/ob) 
and non-obese (genotypes, -f /ob and -f /+) 
mice were fed an experimental diet contain- 
ing 20% casein, 32% glucose, 32% sucrose, 
10% com oil, 5% salts (11), 0.5% vitamin mix 
(12) and 0.2% choline chloride. At 18 weeks 
of age and prior to Ta treatment and cold 
exposure, nonfasting blood samples were 
taken from the retro-orbital sinus in heparin- 
ized tubes for basal glucose and insulin de- 
terminations. At 24 hr before cold exposure, 
half of the obese and nonobese mice were 
injected ip with 100 /ig L-thyroxine (Sigma 
Chemical Co., St. Louis, MO) in 0.25 ml of 
0.9% NaCl adjusted to pH 12 with NaOH, \ 
and the remaining half were injected with \ 
alkaline NaCl alone. These injections were \ 
repeated immediately before cold exposure, i 
After 90 min at 4°, animals were killed by j 
decapitation and blood collected in heparin- \ 
ized tubes for determinations of plasma glu- ; 
cose (13), insulin (14), and FFA (15). k 

In vitro lipolysis was measured in prepa- i 
rations of adipocytes isolated from 1 g per- i 
tions of epididymal adipose tissue by the ;: 
method of Rodbell (16). The washed fat ceUs I 
were suspended in 9 ml of Krebs-Ringer hi- s 
carbonate buffer containing 3% fatty acid- ^ 
free albumin. Three ml of this suspension was 
used to determine DNA (17). For determi- 
nation of FFA release, 0.9 ml samples of fat 
cell suspension were incubated in duplicate 
vials for two hours at 37° in 2.1 ml Krebs- 
Ringer bicarbonate buffer containing 3% 
fatty acid-free albumin with and without 1.1 
X 10"^ A/ epinephrine (Fisher Scientific Co., 
Fairlawn, NJ). At the end of the 2-hr incu- 
bation period, these test samples containing ' 
fat cells and incubation medium were cooled 
to 4°, homogenized and extracted for lipid 
(18). Zero time samples were prepared in 
duplicate by adding 0.9 ml portions of fat cell 

0037^9727/78/1591-01 I6$0l.00/0 
Copyright (D 1978 by the Society for Experiittetital Biology and Medicine 
^^ rights reserved. 




suspension to 2.1 ml of buffer at 4°. They 
were homogenized without incubation and 
the lipid was immediately extracted (18). The 
FF A content of lipid extracts from zero time 
and test samples was determined (IS) and the 
total FF A content of the homogenates of fat 
cells plus incubation medium was calculated. 
The FFA in zero time samples was measured 
to provide an index of intracellular levels of 
FFA after cold exposure, as well as for cal- 
culation of FFA release from triacylglycerol 
during the incubation period, since the FFA 
content of the zero time samples was sub- 
tracted from that of the incubated test sam- 
ples. FFA release was expressed as /xeq 
FFA//ig DNA/hr. Statistical comparisons 
were made by Student's / test (19). 

Results. The effects of T4 treatment on 
plasma glucose, insulin and FFA after cold 
exposure for 90 min are presented in Table I. 
T4 treatment had no statistically significant 
effect on any of these parameters in either 
obese or nonobese mice. However, there was 
a tendency toward higher plasma FFA values 
in r4-treated mice, particularly for the obese. 

Under non-fasting conditions at ambient 
temperature, obese mice had plasma glucose 
values of 228 ± 20 (mean ± SE) mg/dl and 
insulin values of 96 ± 8 juU/ml, while non- 
obese mice had glucose values of 187 ± 19 
mg/dl and insulm values of 24 ± 1 /iU/ml. 
In comparison to these basal values, the ele- 
vated glucose and depressed insulin values 
shown in Table I indicate the response to the 
stress of cold exposure in both obese and 
nonobese mice. The higher plasma glucose 
and insulin values of the obese in comparison 
to nonobese mice under basal conditions are 
characteristic of this genotype. These same 
differences were observed in cold-exposed 

obese and nonobese mice. 

The intracellular concentrations of FFA in 
isolated adipocytes are also shown in Table 
I. Adipocytes from obese mice treated with 
T4 had a significantly higher concentration of 
FFA than those from untreated obese mice. 
In contrast, fat cells from nonobese mice had 
similar FFA concentrations regardless of 
treatment, and these were not significantly 
different from the mean value for untreated 
obese mice. This elevated zero time FFA 
concentration only in r4-treated obese mice 
suggests an increased in vivo lipolytic re- 
sponse to 7*4 in these animals. 

Figure 1 illustrates the results of the meas- 



Fig. I. Effect of T4 treatment in vivo on production 
of FFA by hydrolysis in fat cells isolated from epididy- 
mal adipose tissue of cold-exposed mice and incubated 
with (+) and without (-) 1.1 X 10"^ M epinephrine 
(EPI). Number of animals are indicated on each column, 
bars represent SEM, and P values compare differences 
between paired column means. 

TABLE I. Effect of Thyroxine Treatment on Obese and Nonobese Mice Exposed to the Cold (4°) for 90 




Plasma elucose 

Plasma insulin 

Plasma FFA 

Adipocyte FFA 
(/xeq/Mg PNA) 




379 ± 54 (6)" 
375 ± 64 (6) 


24 ± 5 (6) 

23 ± 3 (6) 


722 ± 142 (6) 
855 ± 208 (6) 


0.25 ± 0.08 (6) 

1.33 ±0.43 (6) 

P < 0.05 


256 ± 58 (6) 
237 ± 11(7) 


17 ± 1 (6) 

18 ± 1 (7) 


697 ± 75 (4) 
716 ±65 (5) 


0.37 ± 0.09 (3) 
0.41 ±0.05(7) 


* Values are mean ± SE. Number of mice sampled in parentheses. 



urements of in vitro lipolysis after cold expo- 
sure for 90 min. Lipolysis in isolated adipo- 
cytes is presented as /xeq FFA released per 
/ig DNA per hr of incubation in order to 
express the results in terms related to cell 
number rather than cell mass. The data from 
the untreated control mice show that adipo- 
cytes from cold-exposed obese mice had re- 
duced lipolytic activity in comparison to 
those from nonobese mice. The addition of 
epinephrine to the incubation medium of fat 
cells from both obese and nonobese control 
mice increased FFA release, but the values 
were not significantly different from those 
under non-stimulated conditions. The epi- 
nephrine-stimulated release of FFA from the 
fat cells of untreated obese mice remained 
significantly (P< 0.05) lower than the release 
from fat cells of untreated nonobese mice. 

T4 treatment of obese mice before cold 
exposure had a striking effect on in vitro 
lipolysis in contrast to the small but not sig- 
nificant effect in nonobese mice (Fig. 1). 
Adipocytes from r4-treated obese mice re- 
leased significantly more FFA both in the 
presence (P< 0.05) and in the absence (P< 
0.01) of epinephrine in comparison to adi- 
pocytes from corresponding untreated obese 
mice. The response to epinephrine of adipo- 
cytes from r4-treated obese mice was more 
than three times greater than that of fat cells 
from untreated control obese mice. However, 
adipocytes from nonobese mice showed no 
increase in FFA release when treated with T4 
prior to cold exposure, although there was a 
significant (P < 0.05) rise in FFA release in 
response to epinephrine in r4-treated non- 
obese mice. 

Discussion, From the data presented it is 
apparent that the failure of ob/ob mice to 
survive during cold exposure was not attrib- 
utable to insufficient circulating FFA since 
plasma values in obese mice were comparable 
to those in nonobese mice after cold exposure 
for 90 min at 4°. Other studies in this labo- 
ratory (20) showed that a more prolonged 
cold exposure (up to 4 hr) also resulted in 
similar plasma FFA values in obese and non- 
obese mice. The plasma FFA values obtained 
in the present experiment were similar to 
those found by Abraham et al. (9) for obese 
and nonobese mice after norepinephrine ad- 
ministration and after a 24-hr fast. Their 

study and our study during cold stress si 
normal in vivo lipolysis for ob/ob mice 
roid hormone treatment, which is knc 
alleviate the hypothermia in ob/ob m 
also did not significantly alter FFA val 

The in vitro results indicated an inhi 
of basal and epinephrine-stimulated lif 
in adipocytes from ob/ob mice aftei 
stress, a condition in which lipolysis s 
be maximally stimulated. This was sim 
the inhibition of FFA release from a< 
tissue of ob/ob mice found at ambien 
perature by other investigators (6-8), { 
that Marshall and Engel (6) did no 
inhibition under basal conditions (w 
epinephrine). In addition, Herberg et a 
reported increased release of FFA froi 
didymal adipose tissue under both ba& 
epinephrine-stimulated conditions. Ho^ 
these latter investigators pre-incubate< 
pose tissue in Krebs-Ringer bicart 
buffer with albumin and glucose (no in 
before measuring lipolysis. It is possibi 
insulin was ^Vashed out" by this prcin 
tion and no longer exerted its known i 
tory effect on lipolysis (22). Otto et ^ 
found elevated lipolysis as measured b) 
erol release from adipose tissue of ob/ol 
under basal conditions, but reduced sei 
ity to epinephrine and thyroid hormoi 
ministration. Although FFA release s 
parallel glycerol release during lipolysi 
may not occur under these conditions ( 
ing from ours, in which adipose tis! 
contrast to adipocytes were incubated 
presence of glucose. 

The coexistence of our in vivo results 
ing similar plasma FFA after cold ex| 
in both r4-treated and untreated obes 
nonobese mice and the in vitro results 
ing variable FFA release from fat c 
possible for several reasons. First, wh 
cells from untreated obese mice releas< 
FFA on a cell number basis, the ino 
number of fat cells in these obese mic 
could be sufficient to maintain plasnu 
at similar concentrations to those in noi 
mice. Also, in vitro conditions are not : 
gous to those in vivo. For example, it i 
sible that a more rapid turnover of circi 
FFA or an inhibition of FFA release ir 
blood occurs in vivo, FFA determin 



Iso made on homogenates of fat cells 
nibation medium and, therefore, rep- 
d fatty acid release from triacylglyc- 
it hot necessarily release from fat cells, 
an et al (24) have shown that under 
iicumstances intracellular FFA con- 
ions increased without increasing FFA 
from the cell. Since the intracellular 
oncentration in the r4-treated obese 
/as significantly higher than in any 
^oup, as determined from the zero 
mples, it is possible that all of the FFA 
d during lipolysis in vivo were not re- 
into the circulation, 
significant increase in FFA release 
pid stores in adipocytes in response to 
lirine in both nonobese and obese mice 
with T4 is in agreement with obser- 
; in other rodents in which Ta poten- 
the action of epinephrine (25) and in 
the lipolytic response of adipocytes in 
as afTected by the in vivo thyroid status 
animal (5, 26). However, the accen- 
epinephrine-stimulated lipolysis in 
mice treated with T^ as compared to 
«r effect in nonobese mice suggests an 
ed sensitivity to epinephrine in obese 
ice the hypothyroid status is corrected, 
ough this study did not directly test 
;>othesis that decreased thermogenesis 
ob mice during cold exposure was a 
)f decreased FFA availability, the evi- 
of reduced lipolysis by adipocytes 
was reversed by Ta treatment supports 
fpothesis. Ahhough circulating FFA 
trations were not significantly affected 
:reatment, it is possible that Ta poten- 
the rate of FFA release from adipose 
in vivo as well. A similarly increased 
FFA uptake and oxidation could allow 
reased thermogenesis, while maintain- 
sma FFA concentrations constant. 
mary. Treatment of ob/ob mice with 
ir prior to cold exposure did not alter 
I concentrations of glucose, insulin and 
luring cold exposure although ob/ob 
emained hyperglycemic and hyperin- 
nic when compared to nonobese mice, 
ontent of and FFA release from iso- 
idipocytes were significantly elevated 
reated obese mice after cold stress as 
red to untreated obese mice. Ta treat- 
ilso produced a marked increase in 

epinephrine-stimulated FFA release from fat 
cells of obese mice in vitro. 

These results indicate that correction of the 
hypothyroid status of ob/ob mice with phar- 
macological doses of Ta improved the in vitro 
lipolytic response of fat cells, but did not alter 
the circulating concentrations of important 
energy sources for thermogenesis in vivo. 

The technical assistance of Ms. Teresa Pasquine is 
gratefully acknowledged. 

1. Mayer, J., and Barrnett, R. J., Yale J. Biol. Med. 24i, 

2. Joosten, H. F. P., and van der Kroon, P. H. W., 
Metabolism 23, 425 (1974). 

3. Otto, W., Taylor, T. C, and York, D. A., J. Endo- 
crinol. 71, 143(1976). 

4. Ohtake, O., Bray, G. A., and Azukizawa, M., Amer. 
J. Physiol. 233, Rl lO-Rl 15 (1977). 

5. Goodman, H. M., and Bray, G. A., Amer. J. Physiol. 
210, 1053(1966). 

6. Marshall, N. B., and Engel, F. L., Lipid Res. 1, 339 

7. Leboeuf, B., Lochaya, S., Lcboeuf, N., Wood, F. C. 
Jr., Mayer, J., and Cahill. G. F., Amer. J. Physiol. 
201, 19(1%1). 

8. Steinmetz, J., Lowry, L., and Yen, T. T. T., Dia- 

9. Abraham, R. R., Dade, £., Elliot, J., and Hems, D. 
A., Diabetes 20, 535 (1971). 

10. Marliss, E. B., Cuendel, G., Balant, L., Wollheim, 
C. B., and Stauffacher, W., in *'Lipid Metabolism, 
Obesity and Diabetes: Impact Upon Atherosclero- 
sis.*' Horm. Metab. Res. Supplement Series, No. 4, 
p. 93, Academic Press, New York (1974). 

11. Thenen, S. W., and Mayer, J., J. Nutr. 107, 320 

12. Thenen, S. W., and Mayer, J., Proc. Soc. Exp. Biol. 
Med. 153,464(1976). 

13. Fales. F. W., in "Standard Methods of Clinical 
Chemistry" (D. Seligson, ed.). Vol. 2, p. 101. Aca- 
demic Press, New York (1963). 

14. Herbert, V., Lau. K., Gottlieb, C. W., and Bleicher, 
S. J., J. Clin. Endocrinol. 25, 1375 (1965). 

15. Itaya, K., and Ui. M., J. Lipid Res. 6, 16 (1965). 

16. Rodbell, M., J. Biol. Chem. 239, 375 (1964). 

17. Burton, K., Biochemistry 62, 315 (1955). 

18. Bligh, E. G., and Dyer, W. J., Can. J. Biochem. 37, 
911 (1959). 

19. Mendenhall, W. M., "Introduction to Probability 
and Statistics, 3rd Ed.," Wadsworth Publishing Co., 
Belmont, CA( 1971). 

20. Thenen, S. W., and Carr, R. H., Fed. Proc. Fed. 
Amer. Soc. Exp. Biol. 36, 1 150 (1977). 

21. Herberg, L., Gries, F. A., and Hesse- Wortman, C, 
Diabclologia 6, 300 (1970). 


22. Jungas, R. L., and Ball, E. G., Biochemistry 2, 586 25. Swanson, H. E., Endocrinology 60, 205 (1957). 
(1963). 26. Dcbons, A. F., and Schwartz, I. L., J. Lipid Res. 2. 

23. Johnson, P. R., and Hirsch, J., J. Lipid Res. 13, 2 86 (1%1). 

24. Cushman, S. W., Heindel, J. J., and Jenrenaud, B., 

J. Lipid Res. 14, 632 (1973). Received February 8, 1978. P.S.E.B.M. 1978. Vol. 159. 


>rphine-lnduced Inhibition of Episodic LH Release in Ovariectomized Rats with 
Complete Hypothalamic Deafferentation^ (40296) 


iversily of California, San Francisco School of Medicine, Department of Physiology, San Francisco. 

California 94143 

iboratory recently reported that apo- 
e, a drug that stimulates dopamine 
s, caused a transient (50-60 min) but 
inhibition of the episodic pattern of 
^se normally observed in ovariecto- 
its (1, 2). This effect is mediated by 
•n of dopamine receptors since pi- 
md d-butaclamol, agents which block 
:eptors, prevent the inhibitory effect 
The present study was designed to 
le if this inhibition is mediated by an 
ictivation of dopamine receptors 
le hypothalamic-pituitary unit (3, 4), 
le of it in some other region of the 
ith a significant dopaminergic input 
the neostriatum (3). Therefore, the 
of apomorphine on episodic LH 
ivere determined in ovariectomized 
/iously subjected to complete hypo- 
: deafiferentation in order to isolate 
lal basal hypothalamus (MBH)-pitui- 
: from the rest of the brain. 
ials and methods. Adult female 
-Dawley rats (Simonsen Laborato- 
roy, CA) weighing 260-280 g were 
led on a lighting schedule of 14 hr 
hr darkness (light on 0500-1900 hr) 
Lab chow and water ad libitum. Daily 
imears were taken and only those rats 
two or more consecutive 4-day es- 
cles were used for experimentation, 
srentation of the MBH was per- 
with a small double-edged Halasz- 
fe (5) of bayonet shape (dimensions: 
2.0 nmi, radius 1.6 mm). Under so- 
:ntobarbital anesthesia (35 mg/kg 
animal's head was placed in a ster- 
nstrument with the ear bars 2.4 mm 
e level of the tooth bar. After drilling 

rted by grants from the National Institutes of 
»5577 and AM06704). 
It address: Department of Anatomy, UCLA 
Medicine, Los Angeles, California. 

a hole in the skull, the knife was lowered 
through the superior sagittal sinus to the base 
of the skull 8.3 mm anterior to the interaural 
line. The knife was first rotated to the right 
90"", and then 180'' to the left (to maximize 
the probability for completeness of the ante- 
rior section of the cut). The blade was next 
stereotaxically moved 3 nmi posteriorly, and 
then rotated 180'' to the right. It was then 
moved anteriorly 3.3 mm (to assure complete- 
ness). Finally, the blade was rotated 90'' to- 
ward the starting position, and removed from 
the brain at the point of entry. Following 
deafferentation, vaginal smears were taken 
for 3 to 6 weeks after which time only those 
rats having shown either constant vaginal 
estrous or diestrous smear patterns for three 
weeks or more were ovariectomized. 

Six weeks following ovariectomy a poly- 
ethylene cannula was inserted into the exter- 
nal jugular vein and used for collecting blood 
samples the following day. An additional 
cannula was placed subcutaneously in the 
animal's back for later drug administration. 
The next day, after an iv injection of 200 
units heparin, unanesthetized, unrestrained 
animals were bled continuously through a 
piece of flexible tubing, one end of which was 
connected to the animal's cannula and the 
other end through a peristaltic pump to a 
microliter syringe kept on ice for the collec- 
tion of blood samples. Fifty or 100 /il whole 
blood were collected every 5 or 10 min, re- 
spectively, and added directly to assay tubes 
(kept in an ice bath) containing 400 or 450 
/xl of phosphate buffered saline with 0.1% 
gelatin. After collecting blood samples for a 
XVi- to 2 hr-control period, animals were in- 
jected with apomorphine hydrochloride (a 
selective stimulator of dopamine receptors (6, 
7), Merck Chem., Rahway, NJ, 1.5 mg/kg in 
saline) through the indwelling sc cannula. 
Bleeding was then continued for an addi- 
tional I to I V2-hr period. Whole blood sam- 


0037-9727 11%! \S9\J0\l\Vi\ X»|^ 
AU hghu reserved. 



pies were analyzed for LH by a slight modi- 
fication (8) of the ovine-ovine rat LH double 
antibody radioimmunoassay of Niswender et 
al (9). LH values (ng/ml whole blood) are 
expressed in terms of the NIAMDD Rat LH- 
RP-1 preparation which has a biological po- 
tency equivalent to 0.03 x NIH-LH-Sl. 

Following experimentation, rats were per- 
fused with 10% formalin plus 1% calcium 
chloride. The extent of hypothalamic deaffer- 
entation was determined both by visual ex- 
amination of the cut at the base of the brain 
as well as by close histological examination 
after sectioning brains at 50 /xm in the trans- 
verse plane and staining with Nissls stain 
using basic fuchsin. 

Results. Forty-five of 53 animals (85%) 
showed persistently leucocytic (constant dies- 
trous) vaginal smear patterns for at least 3 
weeks following hypothalamic surgery. The 
remaining eight rats (15%) exhibited persist- 
ent vaginal comification (constant estrous) 
during this same period of time. No hypotha- 
lamic necrosis was observed in the great ma- 
jority of animals subjected to deafferentation 
and later used for experimentation. The ne- 
crosis that was seen in a few rats involved 
only the extreme rostral or caudal sections of 
the deafferented tissue and never involved 
the arcuate nucleus-median eminence region. 
The pituitary gland's of all experimental ani- 
mals were not damaged by the knife. Addi- 
tionally, no apparent histological differences 
with regard to the extent of deafferentation 
were discernible between constant estrous 
and constant diestrous animals (see Fig. 1). 
The deafferented tissue included all of the 
arcuate nucleus and median eminence, much 
of the ventromedial nucleus, and variable 
amounts of the dorsomedial nucleus. The 
posterior part of the suprachiasmatic nucleus 
was included within one side of the hypotha- 
lamic island in 2 of 8 constant estrous and 3 
of 12 constant diestrous rats. 

Twelve of the 45 rats displaying a persist- 
ently leucocytic smear pattern following hy- 
pothalamic deafferentation were randomly 
selected for bleeding 6 weeks after ovariec- 
tomy. In all 12 animals pulsatile LH release 
was absent and LH levels were very low (<28 
to <1 10 ng/ml). The rat in constant diestrus, 
depicted in Fig. 1, had <28 ng LH/ml whole 
blood during a 3 hr bleeding period. Of the 

eight completely deafferented, constant es- 
trous animals bled 6 weeks after ovariectomy, 
five exhibited pulsatile LH release during a 
1 Vi- to 2 hr-control period of bleeding (Fig. 
2), though at somewhat reduced levels when 
compared with pulsatile LH release normally 
seen in ovariectomized rats. In the remaining 
three rats, problems occurred during the 
bleeding procedure in one, while the other 
two animals displayed ei^er nonepisodic, 
low blood LH levels or only one LH pulse in 
the control period. 

Apomorphine caused a stereotyped gnaw- 
ing behavior pattern in ovariectomized rats 
with complete hypothalamic deafferentation, 
much as it does in intact or ovariectomized 
animals not subjected to hypothalamic sur- 
gery (1,2, 6). This agent was administered to 
eight rats with complete hypothalamic deaf- 
ferentation which previously had shown con- 
stant vaginal estrous smear patterns before 
ovariectomy. In the five rats having well de- 
fined episodic LH release patterns during the 
control period, apomorphine caused an inhi- 
bition (four rats) or reduction (one rat) of 
pulsatile LH secretion lasting at least 40-90 
min. Three examples are given in Fig. 2. The 
extent of the cut in the middle animal repre- 
sented in Fig. 2 is shown in the top of Fig. 1. 
The response to apomorphine could not be 
determined in the remaining 3 rats because 
of the reasons cited above. 

Discussion. This study demonstrates that 
apomorphine, a specific dopamine receptor 
stimulating agent (6, 7), can exert an inhibi- 
tory effect on episodic LH release in ovari- 
ectomized rats previously subjected to com- 
plete hypothalamic deafferentation. We have 
previously observed this inhibition in ovari- 
ectomized animals not subjected to complete 
hypothalamic deafferentation (1,2) and have 
shown that the sc injection of saline (1) or 
distilled water (2) into ovariectomized rats 
had no effect on episodic LH release. Fur- 
thermore, the sc injection of apomorphine 
into animals with hypothalamic deafferenta- 
tion was accomplished through the use of an 
indwelling sc cannula connected to a suffi- 
cient length of flexible tubing to extend out 
of the animal's cage. Thus, the animals were 
unaware of any injection procedure. It ap- 
pears from these and our previous data that 
the inhibition of episodic LH release caused 


Constant estrus 


Constant diestrus 

I mm 

Fig. 1. Representative sagittal reconstructions indicating the extent of complete hypothalamic deafTerentation in 
rats subsequently showing constant estrous or constant diestrous vaginal smear patterns. Actual brain cross sections 
for each of these animals are shown to the left. The arrows indicate the location of the knife cut. 

by apomoq>hme is a result of activation of 
dopamine receptors within the medial basal 
hypothalamus (MBH) and/or pituitary 
gland, and not outside this region. 

The postsynaptic dopamine receptors re- 
sponsible for inhibition of episodic LH 
release are probably associated with neurons 
innervated either by dopaminergic neurons 
originating in the arcuate nucleus or within 
the substantia nigra, and both these areas 
send axonal projections to the median emi- 
nence (10-13). In this regard, the median 
eminence contains high concentrations of 
LHRH (14, 15), apparently within the ter- 
minals of LHRH neurons. It is possible that 
activation of dopamine receptors on these 
LHRH neurons may result in an inhibition 
of LHRH release. A hypothalamic site of 
action for apomorphine is suggested by the 

evidence thai portal vein infusion of dopa- 
mine had no effect of LH release (16), while 
the in vitro pituitary secretion of LH was 
inhibited by dopamine only when the median 
eminence was included in the incubation (17). 
Alternatively, a pituitary site of action cannot 
be ruled out since dopamine receptors are 
present there (4). 

It should be emphasized that the inhibition 
of episodic LH release by apomorphine could 
only be tested in those few hypothalamic- 
deafferented animals showing a constant vag- 
inal estrous smear pattern, since only in these 
rats was episodic LH release present after 
ovariectomy. The vast majority of deaffer- 
ented rats (85%) exhibited a constant vaginal 
diestrous smear pattern and in this type ani- 
mal LH levels were very low and nonpulsatile 
after ovariectomy. Blake and Sawyer (18) 



Time (hours) 

Fig. 2. Three examples ofthe effect of apomorphine 
(APO; 1.5 mg/kg sc) on episodic LH release in ovariec- 
tomized rats with complete deafferentation ofthe medial 
basal hypothalamus. 

indicated that S of 1 1 animals subjected to 
complete deafferentation of the MBH had 
constant vaginal estrous smear patterns and 
episodic LH release after ovariectomy. These 
authors suggested on the basis of these ani- 
mals that pulsatile LH secretion may possibly 
be inherent to the MBH-pituitary unit. In 
only a small percentage of the rats in the 
present report was the deafferented hypotha- 
lamic tissue capable of maintaining episodic 
LH secretion. In agreement with Blake and 
Sawyer (18), complete MBH deafferentation 
also had produced constant vaginal comifi- 
cation in these rats. Inclusion of the supra- 
chiasmatic nucleus within the deafferented 
region has been suggested to account for the 
persistence of LH secretion and this constant 
vaginal estrous smear pattern (19). In the 
present study the suprachiasmatic nucleus 
was anterior to, or destroyed by the knife cut 
in the large majority of rats in both groups. 
Moreover, even when a portion of this nu- 
cleus was included within the deafferented 
hypothalamic tissue in a few constant dies- 
trous rats, very low, nonepisodic blood LH 
levels still resulted. Thus, the reason why 
some rats should continue to show episodic 
LH release while others do not, when the 

extent of hypothalamic deafferents 
pears similar in both groups, is not 
present. The absence of pulsatile L 
tion following MBH deafferentatioi 
due to severing the axons of LHRH 
whose cell bodies lie outside the MB! 
interrupting fibers stimulating LH 
thesis and/or release. Complete dea 
tion results in a large decrease it 
content in the rat MBH (20, 21). 
more, norepinephrine has been sug 
play an excitatory role in the regu 
LH secretion (1, 22-25) and the cc 
norepinephrine in the MBH is to 
pleted by deafferentation (26). Nev< 
afferent input to the MBH seems I 
quired in the great majority of rats t 
episodic LH secretion. Moreover, d 
receptors within the MBH-pituitar 
seem responsible for mediating the ii 
effect of apomorphine on pulsatile L 

Summary, Complete neural dea 
tion of the MBH in 53 rats resul 
constant vaginal diestrous smear p; 
85% of the rats, and in this type anii 
low blood LH levels and absence of 
LH release followed ovariectomy, 
maining 15% had a constant vagina 
smear pattern, and most demonstra 
satile LH secretion following ovar 
Thus, afferent input to the MBH see 
required in most rats to sustain epis 
secretion. Administration of apomoi 
dopamine receptor stimulator, to 
wtuch were in constant estrus before 
tomy, resulted in inhibition of puis 
secretion, suggesting that this apom< 
induced inhibition is a result of acti> 
dopamine receptors within, rather tl 
side, the MBH-pituitary unit. 

We would like to thank Brad Garibaldi. Bo 
Veronica Lickova and Cindy Voytek for th< 
these studies, and Annette Lowe and Linda D 
preparation of the figures and typing the n 
Thanks also to Dr. G. Niswender for anti-ovii 
15, Dr. H. Papkoff for highly purified ovin 
G3222B for iodination. Dr. J. Garcia for goat 
against rabbit gammaglobulin, and Dr. A. F. F 
the NIAMDD for the rat LH used as a refen 

1. Drouva, S. V., and Gallo, R. V., Endocrii 



a, S. v.. and Gallo. R. V., Endocrinology 100, 


. J. Z., Reisine, T. D., and Yamamura, H. I., 

Res. 13^,578(1977). 

I, M., Roberts, J., and Wcincr, R., Fed. Proc. 

J (1977). 

:, B., and L. Pupp, Endocrinology 77, 553 

u N. E., Rubcnson, A., Fuxc, K., and Hokfcll, 

Phannacol. 19, 627 (1967). 

A. M., Psychopharmacologia 10, 316 (1%7). 

i, R. B., Gallo, R. V., and WiUiams. J. A., 

Tinology%, 1210(1975). 

nder, G. D., Midglcy, A. R., Jr., Monroe, S. 

d Reichert, L. E., Jr., Proc. Soc. Exp. Biol. 


K.. Acta. Physiol. Scand. 64 (Suppl.) 247, 37 

und. A., Moore, R. Y., Nobin, A., and Slencvi, 
ainRes. 51, 171(1973). 
ats, M., Brownstein, M., Saavedra, J. M., and 
Ki, J., Brain Res. 77, 137 (1974). 
J. S., Palkoviis, M., and Brownstein, M. J., 
Res. 106,363(1976). 

its, M.. Arimura, A., Brownstein, M., Schally, 
and Saavedra, J. M., Endocrinology 95, 554 

15. Wheaton, J. E., Knilich, L., and McCann, S. M., 
Endocrinology 97, 30 (1975). 

16. Kamberi, 1. A., Mical, R. S., and Porter, J. C, 
Endocrinology 87, 1 (1970). 

17. Miyachi, Y., Mecklenburg, R. S., and Lipsett, M. B., 
Endocrinology 93, 492 (1973). 

18. Blake, C. A., and Sawyer, C. H., Endocrinology 94, 

19. Hayashi, S., Mennin, S. P., and Gorski, R. A., Neu- 
roendocrinology 14, 321 (1975). 

20. Weiner, R. I., Patlou, E., Kerdelhue, B., and Kordon, 
C. Endocrinology 97, 1597 (1975). 

21. Brownstein, M. J., Arimura, A., Schally, A. V., Pal- 
kovits, M., and Kizer, J. S., Endocrinology 98, 662 

22. Sawyer, C. H., Markee, J. E., and Hollinshead, W. 
H., Endocrinology 41, 395 (1947). 

23. Ojeda, S. R., and McCann, S. M., Neuroendocrinol- 
ogy 12, 295 (1973). 

24. Kalra, S. P., and McCann, S. M., Neuroendocrinol- 

25. Krieg, R. J., and Sawyer, C. H., Endocrinology 99, 

26. Weiner, R. I., Shrync, J. E., Gorski, R. A., and 
Sawyer, C. H., Endocrinology 90, 867 (1972). 

Received February 14, 1978. P.S.E.B.M. 1978, Vol. 159. 


The Effect of Leukocyte Hydrolases on Bacteria. XI. Lysis by Leukocyte Extra 

by Myeloperoxidase of a Staphylococcus aureus Mutant Which is Deficient in ' 

Acid, and the Inhibition of Bacteriolysis by Lipoteichoic Acid^ (40297) 


Department of Medical Microbiology, School of Medicine, and the Department of Oral Biology, Hebrew i 
Hadassah School of Dental Medicine, founded by the Alpha Omega Fraternity, Jerusalem, Israt 

In previous publications, we have shown 
that Staph, aureus, which had been harvested 
from the logarithmic phase of growth, was 
readily lysed by human leukocyte extracts 
(ENZ) and by myeloperoxidase (MPO). On 
the other hand, bacteria obtained from the 
stationary phase of growth were highly re- 
sistant to degradation by these agents (1-8). 
It was further demonstrated that the lysis of 
the bacteria by the leukocyte factors was 
probably caused by the activation of autolytic 
systems and not by the direct effect of lyso- 
somal hydrolases on the bacterial walls (8). It 
was suggested that one of the reasons for the 
resistance to degradation of the stationary 
phase bacteria was not due to the lack of 
autolytic enzymes in the old cells, but to the 
much thicker cell walls of such cells (7, 8). 

It is well established that Gram positive 
bacteria possess teichoic acid (TA) as an in- 
tegral part of the cell wall and a membrane- 
associated lipoteichoic acid (LTA) (9). Since 
TA was claimed to deter the interaction of 
lysozyme with the peptidoglycan, thus con- 
ferring resistance to bacteriolysis (10), and 
since LTA has been implicated in the inhi- 
bition of autolytic enzymes in bacteria (11, 
12), it was of interest to test the effect of TA 
and LTA on bacteriolysis induced by leuko- 
cyte factors. The data presented show that a 
mutant of Staph, aureus, which completely 
lacks ribitol TA but nevertheless possesses 
membrane-associated LTA, is much more 
susceptible to lysis by ENZ and by MPO than 
the parent strain. It will also be shown that 
while LTA strongly inhibited the lysis of 
Staph, aureus by ENZ and by MPO, TA was 

' This study was supported by Research Grants ob- 
tained from The Carol and Ella Reisfeld Fund; The 
Chief Scientist, the Ministry of Health; and by the Max 
Bogen Research Fund obtained through the Friends of 
the Hebrew University in the United States of America. 

not inhibitory. 

Materials and methods. Microoi 
The following Staph, aureus strains ^ 
ployed: The parent strain SH (Str^ 
mutants 52A5 and 52A2. The mutar 
lacks ribitol teichoic acid in the cell 
no ribitol phosphate polymer was dc 
any other cell fraction or in the s| 
dium. The lack of ribitol teichoic ac 
cell wall is caused presumably by soi 
in the membrane or in some unknot 
required in the polymerization or ati 
step of the teichoic acid to murein. 
tant 52A2 lacks 7V-acetylglucosamii 
cell wall ribitol teichoic acid. All the 
are also known to be deficient in \ 
(for details see reference 13). The? 
were kindly supplied by Dr. D. I 
from the Department of Biophj 
Weitzmann Institute, Rehovoth, Is 
In addition we have employed Stap 
strain Cowan I which is known to I 
protein A. The bacterial strains w 
vated either in Brain Heart Infusi( 
broth (Difco Laboratories, Detroit, 
BHI which contained 0.5 /xCi/ml c 
mally labeled [*^C]D-glucose, specifi 
150-250 mCi/mmol (New England 
Boston, MA) as described (2). All t 
rial cells were harvested either fron 
arithmic phase of growth (after 3 hi 
bation, OD = 280 Klett units at 54< 
Klett Summerson colorimeter) or s 
phase of growth (after 18 hr of in 
OD = 620 Klett units at 540 /xm). 
were washed several times in saline 
resuspended in distilled water. 

Lipoteichoic acid (LTA). Lipoteic 
(LTA) was isolated from Strep, mi 
1895/74), group A streptococc 
C203S) and from Staph, aureus (S 
52A2 and cowan I) by phenol (Ma 
Inc., St. Louis, MO) or by lysozym 


Copyright e> 1978 by the Society for Experimeaul Biology and Medicine 
trijghu reserved. 



deal Co., St. Louis, MO) (14). The bac- 
extracts were dialyzed with six changes 
tilled water and were then lyophilized, 
preparation did not contain any traces 
enol. Teichoic acid (TA) was isolated 
Staph, aureus SH by TCA (BDH, Eng- 

according to a method described by 
mo and Slade (16). 

jcylation of LTA, was performed as de- 
d by Knox and Wicken (9). 
opolysaccharide (LPS), LPS from £. coli 
J8 was purchased from Difco Labora- 
(Detroit, MI). The LPS was dissolved 
ine to the desired concentration. 
duction of anti'LTA serum. Antibodies 
5t LTA were prepared by immunizing 
5 either with Strep, mutans SE 1895/74, 
^taph. aureus Cowan I or with group A 
\ streptococci, according to a procedure 
bed in detail (14). 

ermination of LTA activity. LTA was 
nined quantitatively by its ability to 
ize human RBC to agglutination in the 
ice of a standard anti-LTA serum as 
bed (16). 

tenolysis. The lysis of staphylococci was 
•med as described in detail (2). Briefly, 
ibeled bacterial suspensions containing 

10^ cpm/lOO Klett units per ml were 
ited for 18 hr at 37° in 0.1 A/ acetate 

pH 5.0 either with freeze and thaw 
ts of human blood leukocytes contain- 
►-500 /xg/ml of protein or with purified 
peroxidase which was kindly supplied 
r. L Olsson from the Department of 
al Medicine University of Lund, Swe- 

den, as well as with nuclear histone (Sigma 
Chemical Co., St. Louis, MO). MPO was 
used here as a cationic protein and not as a 
bactericidal agent which in collaboration with 
H2O2 and halide is a strong bactericidal agent. 

The degree of lysis was determined by 
measuring the percentage release of soluble 
radioactivity from the standard labeled bac- 
terial suspension (2). 

The inhibition of bacteriolysis. Radio-la- 
beled bacteria were incubated for 15 min at 
37° with various amounts of LTA (phenol or 
lysozyme extract) with deacylated LTA (9) 
with TA or with LPS. Following incubation, 
the cells were lysed by leukocyte factors, and 
the degree of inhibition of lysis was deter- 
mined as described (2). The results were ex- 
pressed as the percentage of inhibition of the 
release of radioactivity from a standard sus- 
pension of [*^C]labeled staphylococci. 

Results. The lysis of staphylococci by leu- 
kocyte factors. Table I shows that when sta- 
tionary bacteria were employed, only strain 
52A5 (which is deficient in TA) underwent 
massive lysis following treatment with human 
leukocyte extracts (ENZ). On the other hand, 
all the bacterial strains employed were 
equally susceptible to lysis when harvested 
from the logarithmic phase of growth (young 
cells). It is important to note that identical 
results were obtained when a purified prep- 
aration of myeloperoxidase (MPO) or histone 
were used instead of the leukocyte extracts 
(not shown). Table I also shows that strain 
52A5 looses a somewhat higher percentage of 
radioactivity when incubated in buffer alone 

TABLE I. The Lysis of Different Staphylococcal Strains by Leukocyte Extracts. 

% release of radioactivity from" 

Presence or absence of 


logarithmic phase bacte- 

stationary phase bacteria 


Protein A 




































diolabeled bacteria ( 100 Klett units/ml 540 /un) suspended in 0. 1 A/ acetate buffer pH 5.0 were incubated for 

t 37'' with 100 ^g/ml of human leukocyte extracts (ENZ) and the soluble radioactivity was determined as 

sd in Materials and Methods. Similar results were obtained with MPO or histone. The data are the mean of 


A was extracted from the bacteria either with phenol or by lysozyme as described in Materials and Methods. 

is mutant lacks ^-acetylglucosamine in its TA. . 



(spontaneous lysis) as compared with the 
other strains. Since all the bacterial strains 
employed were found to possess LTA (Table 
I), it is postulated that TA, but not the mem- 
brane-associated LTA, may play an impor- 
tant role in the protection of old bacteria 
against lysis by leukocyte factors. 

The inhibition by LTA of the lysis of staph- 
ylococci. LTA was recently shown to be a 
potent inhibitor of autolytic enzymes in Strep, 
faecalis (11) and Diplococcus pneumoniae 
(12). Since we have recently postulated (8) 
that the lysis of Staph, aureus by leukocyte 
extracts and by membrane-damaging agents 
like MPO and Phospholipase A2, was due to 
the activation of autolytic enzymes, it was of 
interest to examine the possibility that LTA 
will also inhibit bacteriolysis induced by leu- 
kocyte factors and by MPO. Table II shows 
that when LTA (derived either from staphy- 
lococci or from streptococci) was added to 
staphylococci (SH and 52A5) in the presence 
of an inducer of lysis like ENZ or MPO, a 
strong inhibition of lysis occurred. It is also 
seen that H2O2 did not modify either the lytic 
effect of MPO or the inhibitory effect of LTA 
on bacteriolysis induced by MPO. The Table 
also shows that neither deacylated LTA nor 
TA nor LPS had any inhibitory property. It 
is also shown that none of the inhibitors 
employed lysed the bacteria. In other exper- 
iments (not shown) we found that the lysis of 
staphylococci by ENZ could not be inhibited 
by cytoplasmic fractions or cell walls derived 
from group A streptococci, when used at 
similar concentrations. 

Discussion. The data on the higher suscep- 
tibility to lysis of the T A-deficienl mutant by 
leukocyte factors and by MPO and the inhi- 
bition of bacteriolysis by LTA, further con- 
tribute to the understanding of the possible 
role which may be played by TA and LTA in 
the biology of the staphylococci. 

Since TA was claimed to deter the inter- 
action of lysozyme with the peplidoglycan 
(10) it may be postulated that the lack of this 
wall component from the mutant 52A5 ren- 
dered the cell more susceptible to bacterioly- 
sis. Since however, the lysis of Staph, aureus 
by leukocyte enzymes was found not to be 
lysozyme-dependenl (3, 5) and since the TA- 
less mutant was not more susceptible to lysis 
by lysozyme than the parent strain (Table II), 

TABLE 11. The Effect of LTA, Deacylated LTA. 


Leukocyte Factors. 

% Release of radioac- 

tivity after IShrfrom:^ 


Reaction mixture" 

Strain SH 


Buffer alone 



Lysozyme 100 /ig 



Leukocyte extracts lOO/ig 



MPO !00/ig 



LTA' 250 jug 






Leukocyte extracts + H2O2 0.3 




Leukocyte extracts + LTA 150 




Leukocyte extracts + LTA 250 




Leukocyte extracts + LTA 500 




Leukocyte extracts + LTA 500 



Mg + H2O2 0.3 Mg 

Leukocyte extracts + Deacyl- 



ated LTA 150 Mg 

Leukocyte extracts + Deacyl- 



ated LTA 250 Mg 

Leukocyte extracts + TA 250 




Leukocyte extracts + LPS 500 




MPO+H2O2 0.3Mg. 



MPO + LTA 250 Mg 



MPO + LTA 500 Mg 



MPO + LTA 500 Mg + H2O2 



0.3 Mg 

" The reaction mixtures were added to the labele 
bacteria (logarithmic phase) in 0. 1 A/ acetate buffer pi 

** Lysis was determined as the percentage of the soli 
bilized radioactivity as described in Materials and metl 
ods. The data are the average of five experiments. 

' All LTA preparations (see Materials and method 
behaved similarly. 

''ND— Not done. 

one should seek other explanations for th 
higher susceptibility of the mutant to lysis b 
leukocyte factors. 

It may be postulated that since old stapt 
ylococci (shown to be resistant to degrade 
tion) (Table I), possess much thicker cell wal 
(17) and since TA forms the bulk of tl 
staphylococcus cell wall, it is possible that tl 
lack of TA from the mutant renders the *thii 
ner" cell wall of these microorganisms mo 
susceptible to degradation by the autolyt 
enzymes, which are activated by the leuk 
cyte factors (8). Thus TA may be essential f 



ilization of the cell wall not only 
ysozyme but also against the autoly- 

has been shown to be a potent inhib- 
utolytic enzymes in a variety of bac- 
>ecies (11, 12). The findings that ex- 

LTA can inhibit the lysis, by ENZ 
'O (Table II) of staphylococci which 
endogenous LTA (Table I) is intri- 
o explain this phenomenon one may 
e that since leukocyte extracts, lyso- 
id histone were shown to remove the 

LTA from bacterial cells (14), the 
the staphylococcus cells by leukocyte 
nay involve, first the removal of en- 
is LTA from the bacterial cells by the 
te factors, then the release from inhi- 
f the autolytic enzymes, and finally 
itation of the activity of the autolytic 

3wn in Table II lysis of staphylococci 
rganisms known to produce H2O2) 
J induced by MPO. These results are 
est, since neither KCN nor NaNa, 
re hemeprotein inhibitors, could in- 
:teriolysis by MPO (unpublished ob- 
is). It thus points to the possibiUty 
^O (a cationic substance) like other 
ne-damaging agents, (e.g. LCP, his- 
lospholipase A2, polymyxin B, coli- 

may interact with the protoplast 
ne and through perturbation, acti- 

membrane-associated autolytic sys- 
rough the removal of LTA (14). This 
o explain teleologically why PMN 
large amounts of MPO. 
act that H2O2 did not modify the 
7 effect of LTA, on the lysis of staph- 
by MPO (Tabic II), further supports 
tnption that MPO in this system acts 
lie protein. 

masons for the use of acid buffers in 
eriolytic system are based on our 

findings (S, 18) that optimal killing 
5 of staphylococci by leukocyte ex- 
ad histone took place at pH 5.0, 
only a slight effect was obtained at 
aterrelationships among TA, LTA, 

systems and leukocyte factors in 
to the degradation of microbial cell 
stituents in inflammatory sites merit 

Summary. A Staph, aureus mutant (52A5) 
which is deficient in wall teichoic acid (TA) 
was found to be highly susceptible to lysis by 
leukocyte extracts (ENZ) and by myeloperox- 
idase (MPO) when harvested from the sta- 
tionary phase of growth. On the other hand, 
a staphylococcus mutant, which is deficient 
in 7V-acetyl glucosamine in its TA (52A2), the 
parent strain SH and a protein A rich strain 
Cowen I, could be lysed by the leukocyte 
factors only when harvested from the loga- 
rithmic phase of growth. 

The lysis of all the bacterial strains by ENZ 
or by MPO was strongly inhibited by Upotei- 
choic acid (LTA) derived either from staph- 
ylococci or from streptococci. On the other 
hand, deacylated LTA, TA, LPS, cytoplasmic 
or cell wall components derived from strep- 
tococci had no inhibitory effect on bacteriol- 
ysis. It is concluded that TA may be impor- 
tant in the protection of old bacterial cells 
against degradation by leukocyte factors, and 
that LTA may be involved in the control of 
autolytic enzymes in staphylococci. The role 
of MPO in bacteriolysis is also discussed. 

1. Nccman, N.» Lahav, M., and Ginsburg, I., Proc. Soc. 
Exp. Biol. Med. 146, 1137 (1974). 

2. Lahav, M., Neeman, N., James, J., and Ginsburg, 1., 
J. Infec. Dis. 131, 149(1975). 

3. Ginsburg, 1., Neeman, N.. Duchan, Z., Sela, M. N.. 
James, J., and Lahav, M., Inflammation 1,41 (1975). 

4. Sela, M., Lahav, M., Neeman, N., Duchan, Z., and 
Ginsburg, I., Inflammation 1, 57 (1975). 

5. Neeman, N., Duchan, Z., Uhav, M., Sela, M. N., 
and Ginsburg, 1., Inflammation 1, 261 (1976). 

6. Ginsburg, I., and Sela, M. N., Critical Rev. Micro- 
biol. 4, 249 (1976). 

7. Efraii, C, Sacks, T., Neeman, N., Lahav, M., and 
Ginsburg, I., Inflammation 1, 371 (1976). 

8. Lahav, M., and Ginsburg, I., Inflammation 2, 165 

9. Knox, K. W., and Wicken, A. J., Bacteriol. Rev. 37, 

10. Morse, S. I., Ann. N.Y. Acad. Sci., 128, 191 (1965). 

11. Cleveland, R. F., Wicken, A. J., Daneo-Moorc, L., 
and Shockman, G. D., J. Bacteriol. 126, 192 (1976). 

12. Holtje, J., and Tomasz, A., Proc. Nat. Acad. Sci. 72, 

13. Shaw, D. R., Mirclman, D., Chatteijee, A. N., and 
Park, J. T., J. Biol. Chem. 245, 101 (1970). 

14. Sela, M. N., Lahav, M., and Ginsburg, I., Inflam- 
mation 2, 151 (1977). 

15. Malsuno, T., and Slade, H. D., J. Bacteriol. 102, 747 


16. Neeman, N.. and Ginsburg, 1., Israel J. Med. Sci. 8, 18. Klebanoff, S. J., in "Phagocytic Cells in Host Resis 
1799 (1972). ancc" (J. A. Bellanti and D. H. Dayton, cds.) p. 4 

17. Suganuma, A., in "The Staphylococci" (J. O. Cohen. Raven Press. New York (1975). 

cd.) p. 21, Intersciencc, New York (1972). Received April 21, 1978. P.S.E.B.M. 1978. Vol. 159. 


Polybrominated Biphenyls in Chicken Eggs vs. Hatchability^ (40298) 


Michigan State University, East Lansing, Michigan 48824 

; fall of 1973, polybrominated biphen- 
B) were accidentally introduced into 
' poultry and livestock within Michi- 
illions of chickens and thousands of 
ere destroyed to lessen the conlami- 
3f PBB into the food-chain (1). Fries 
I had reported that eggs from hens fed 
PBB averaged 21.5 ppm of hexabro- 
enyl (6-BB) plus heptabromobi- 
(7-BB), and that 7 weeks after with- 
of the PBB diet eggs contained 2.2 
' these compounds. Fat from these 
ntained 69.5 and 62.4 ppm of these 
t the respective times of 9 weeks on 
Its and 7 weeks after their withdrawal, 
idues were reported to be 1.5x the 
level of PBB after 4 weeks of feeding 
ets (3). This steady state effect was 
ed to occur as early as 10 days (4, 5). 
•n data obtained from feeding PBB to 
), calculations (4) revealed a relation- 
PBB in eggs to be 1.3x the level in 
:, similar to the 1.5 value calculated 
le dose-response curve (5) and the 
1 value by Cecil et al. (3). 
70 mortality as a percent of fertile 
eraged 6.2% and 3.9% for diets con- 
either 20 ppm PBB or no PBB, re- 
ly (7). Ringer and Polin (8) showed 
tchability declined and chicks from 
inated eggs were less viable when hens 
d 125 ppm PBB in the diet, but not 
•d 25 ppm (8) or 30 ppm (4). Quail 
tched normally when PBB was fed at 
, but failed to hatch when 100 ppm 

tudy reported herein will establish the 
ship between PBB in eggs vs. hatch- 
by two approaches, which will be 
;o disagree on the extent of this rela- 
>. The impHcations of this incompali- 
lay indicate that analysis for 6-BB 
t be a definitive approach to assess 

al Article No. 8433. Michigan Agricultural 
nt Station. 

PBB toxicity. 

Materials and methods. Adult female White 
Leghorn chickens 10 months in production 
(about 60 weeks of age) were assigned at 
random into one of 7 treatments, or to a 
control group. Twenty-four hens were in each 
group. PBB, as Firemaster FF-l was used in 
this experiment. This compound differs from 
that of Firemaster BP-6 used in other studies 
(2, 3, 6, 7) in that FF-l has anli-caking sub- 
stances added and had been milled to obtain 
a free-flowing compound. In other words, 
Firemaster BP-6 was an intermediate prod- 
uct. Firemaster FF-l was the final product 
sold commercially, the one used in this study, 
and the chemical involved in the contamina- 
tion of Michigan's livestock and poultry. It is 
reported (4) to contain 62.8% 6-BB and 13.8% 
7-BB, as compared to 79.2% 6-BB and 14.3% 
7-BB for the Firemaster BP-6 (2). 

The details of the materials and methods 
used in the experiment, as well as the proce- 
dure for the analysis of PBB in eggs were 
reported (5). Briefly, the hens were fed FF-l 
in the diet at 0.2, I, 5, 25, 125, 625 or 3125 
ppm for 5 weeks, then fed feed without FF-l 
for 8 weeks to obtain data on withdrawal 
effects. Eggs require about 9-10 days to be 
completely formed, 8-9 of which are for yolk 
formation. Thus, sampling of eggs started on 
day 9 of the experiment, and was on every 
7th day thereafter until the 37th day after 
withdrawal (Table I). The experiment started 
June 17, 1974. Starting on June 18th, eggs 
were saved. So that the time for egg sampling 
coincided with the middle of a 7-day collec- 
tion period, the first setting of eggs in the 
incubator were those collected in the first 5 
days on the experiment. All subsequent set- 
tings were from 7-day collections. The mid- 
point of the 1st collection period was day 9 
on and represented equally days 6 through 8 
for the accumulation phase and days 10 
through 12 of the steady-state phase. Thus, 
the hatch value for a week's collection would 
represent the hatch value for the midpoint of 


All righu reserved. 




Relationship Between Hatch and Hexabromobi phenyl (6-BB) in eggs 




5.0 25.0 125.0 

















Day for 
































Day 9 on 


(90) 92 


(88) 93 


(100) 93 


(101) 91 


(60) 28 






16 ON 


(91) 95 


(97) 92 


(105) 97 


(90) 87 


(60) 5 






23 ON 


(96) 88 


(101) 84 


(85) 89 


(76) 96 


(57) 2 






30 ON 


(89) 96 


(98) 89 


(93) 95 


(94) 85 







2 OFF 


(62) 94 


(93) 93 


(86) 94 


(82) 84 


(50) 18 





9 OFF 


(83) 92 


(81) 91 


(81) 94 


(76) 92 


(66) 74 






16 OFF 


(74) 92 


(75) 93 


(61) 90 


(61) 92 


(58) 86 






23 OFF 


(75) 93 


(67) 84 


(71) 94 


(60) 87 


(67) 85 




(14) 43 


30 OFF 


(64) 91 


(60) 88 


(52) 92 


(57) 90 


(55) 91 


(18) 33 


(17) 35 


37 OFF 


(56) 95 


(56) 96 


(53) 91 


(41) 83 


(55) 78 


(19) 21 


(14) 21 


44 OFF 


(53) 93 


(56) 100 


(40) 98 


(38) 92 


(40) 93 


(30) 37 


(8) 63 


51 OFF 


(65) 94 


(59) 95 


(44) 91 


(41) 83 


(49) 88 


(43) 19 


(6) 67 

( ) " Number fertile; Z hatch - (Number match/numbe« fertile) x 100 
a " ff-1 mithdramn 7 days sooner^ therefore add 7 days to 'day for egg sample' 

Hatchability of eggs from White Leghorn chickens fed diets with polybrominated biphenyl Firemaster FF-l. and 
the hexabromobiphenyl levels (6-BB) analyzed in eggs representative of each hatch. 

that week; in this case, day 9 on. The eggs 
from the mid-point of the week were opened, 
pooled, and analysed for 6-BB by the Mich- 
igan Department of Agriculture, as previ- 
ously described (5). Hens were artificially 
inseminated once a week with semen col- 
lected from males housed in a separate room 
and fed diet without FF-1. 

6-BB was assayed by gas liquid chromatog- 
raphy using one or both procedures employ- 
ing a '^H-foil electronic detector at a temper- 
ature of 220^ in the column and detector, and 
250° in the injector, or a *^'^Ni-delector at 
temperatures of 270° in the column, 310° in 
the detector and 300° in the injectorport. The 
important aspect for this experiment was that 
FF-l was assessed from chromalograms by 
reading the peak height of the 6-BB peak 
using Firemaster BP-6 (Lol-#5 143) as a stan- 
dard. Subsequent comparison of this stan- 
dard with those used by the Food and Drug 
Administration (FDA) showed comparable 

patterns. The standard BP-6 was obtained 
from Michigan Chemical Company, the for- 
mer manufacturer of FF-l. Linear and cur- 
vilmear regression and analysis of variance 
were applied to the data (9), after converting 
percentage values of hatch to arcsin yffo (9). 
Results and discussion. Table I contains the 
weekly hatchability data (number hatch per 
number fertile) and the 6-BB levels of eggs 
representative of the day and week that the 
eggs were collected. Not included in Table I 
were the hatchability data for the first 5 days 
on the experiment for which no egg samples 
were obtained. These hatch values were 95.3, 
80.0, 83.1, 88.1, 88.5, 90.0, 69.7 and 61.5% for 
the eggs from FF-1 levels of 0, 0.2, 1, 5, 25, 
125, 625 and 3125 ppm in the diet, respec- 
tively. When these values were considered 
with those of Table I for the first 5 weeks that 
FF-1 was fed, hatchability of control eggs 
averaged 89.9 (± 4.1)%, mean (± SD), and 
91.2 (± 3.1)% for the entire 13 weeks of the 



experiment. Hatchability during 5 weeks of 
feeding FF-1 at 0.2, 1, 5 and 25 ppm were 
90.6, 88.5, 92.8, and 89.9%, respectively; none 
of these values were significantly different, P 
< .05, from the control value. On the other 
hand, poor hatches were obtained when 125 
ppm FF-1 was fed, but not until the hatch 
representing day 9 on (actually days 6-12) 
was obtained. Within the first 5 days of feed- 
ing FF-1 at 625 and 3125 ppm, hatchability 
was significantly {P < .05) below normal. 
None of the eight fertile eggs hatched that 
were obtained from the hens fed 625 ppm 
PBB and representing day 9 on; and no eggs 
were laid by those hens fed 3125 ppm FF-1. 
A subsequent experiment revealed (4, 5) that 
during the steady-state phase of days 9 to 35 
for feeding FF-1, the minimum effective level 
for FF-1 in the diet to produce a significant 
effect on hatchability was between 30 and 45 

6-BB was detected in whole egg samples, 
based on the dose-response curve for steady 
state values (5), at 0.3, 1.5, 7.4, 43.4, and 215 
ppm for the treatment levels of 0.2, 1, 5, 25 
and 125 ppm FF-1 in the diet, respectively. 
The latter level produced a high mortality in 
chick embryos (Table I) during the last few 
days of hatch. Edema of the abdominal and 
cervical regions was the prevalent pathologi- 
cal sign observed in embryos and newly 
hatched chicks from FF-1 treatment (Fig. la 
and lb). The clinical signs resemble those of 
embryos from polychlorinated biphenyl 
treatment (7). The edema was the only side 
effect to be observed that was increased in 
incidence above abnormalities detected in 
control embryos. 

Estimated XVi values were obtained from 
the 6-BB data in Table 1 and found to be 10 
and 21 days for depletion time from prior 
treatment with FF-1 at 0.2-1.0 ppm, and 

Fig. I. Edematous condition of embryo (lb) and chick (la) from feeding polybrominaled biphenyl, Firemaster 
FF-1. to hens at dietary levels higher than 42 ppm. Note the accumulation of fluid typically seen in abdominal and 
head region of embryos, and cervical area of chicks. 



TABLE 11.' 

Level of FF- 
1 withdrawn 

Depletion curve 


0.2 ppm 

1.0 ppm 

5.0 ppm 

25.0 ppm 

125.0 ppm 

Y- -0.2024-0.03 17X 

Y - -0.0647-0.0286X 

Y - 0.3287-O.0i37X 
Y- 1.1870-0.0131X 

Y - 1.881 8-0.0 158X 



" Dose-response lines based on depletion curves for 6- 
BB from chicken eggs after removal of diets with FF-1. 
The relationship is Y « a + bx, where x « days of 
withdrawal starting at day 9 off, and Y * log ppm 6-BB 
in whole egg. 

5-125 ppm, respectively (Table II). Thus, 
higher levels of FF-1 treatment required a 
longer time for depletion based on 37 days of 
measurements during withdrawal. Further- 
more, factorial analysis of these depletion 
data revealed a significant linear and quartic, 
but not cubic, effect. This suggests that the 
slopes of the depletion curves are flattening 
to some extent and that with depletion be- 
yond 37 days the iVz values will be greater. 

As previously pointed out (5), the relation- 
ship between FF- 1 in the diet and 6-BB levels 
in eggs during the steady-state phase, was 
expressed by the equation Y = 0.1763- 
1.012X, where X = log ppm of FF-1 in the 
diet, Y = log ppm 6-BB in whole egg. This is 
in good agreement with the data by other 
investigators (3, 6). The response of hatcha- 
bility to FF-1 in the diet (3), during the 
steady-state phase, was estimated to be Y = 
297.14-140.74X, where X - log ppm FF-1, 
and Y = arcsin y/% of hatchability. 

The algebraic summation of these two 
regressions derived to relate log ppm 6-BB in 
egg, as X, to arcsin \/% hatch, as Y, is given 
in Figure 2, line "a", along with a plot of the 
values from Table I for treatments with FF- 
1 at 25, 125 and 625 ppm. The regression 
lines under comparison are: (a) the line based 
on the derived steady-state values, (b) the line 
based on the depletion phase ("off" data), (c) 
the line based on the steady-state phase ("on" 
data), (d) the line representing both the 
steady-state and depletion phase of the data 
from Table I, and (e) a regression line based 
on egg residues between 30 and 85 ppm of 6- 

The lines for "b", "c", and "d" above were 
calculated on the basis of Y = a -»- bx -I- cx^. 
From these regressions, a linear regression 

can be calculated to represent the a] 
linear portion of these curvilinear line: 
ing the data into response lines for eitl 
"c", or "d" revealed slopes and ini 
very unUke the derived equation, " 
whose linear slopes were very similar 
legend). The Une calculated for "e" 
and a non-significant, P > 0.05, slop 
cated by the lack of correlation (r = 
between hatchability and 6-BB levels ( 
85 ppm in eggs. 

Considering all of these comparison 
derived regression, the conclusion n 
reached that the derived equation i 
hatchability to egg residues is not va 
terestingly, the range of 6-BB levels f 
to 85 ppm supposedly covered a ra 
hatchability from no effect down to < 
of 64%, based on the derived equatioi 

Fries et al (10) reported that 7-BB d 
more rapidly in chicken eggs than 6-B 
withdrawal of diets with BP-6. Als< 
noted that the concentrations of thes 
ponents in fat of hens being fed 20 pp 
4 times that of diet for 6-BB and 1.: 
that of diet for 7-BB. Thus, there is ei 


Fig. 2. Relationship between hexabromo 
(6-BB) in whole egg and the hatchability of eggs 
during and after the feeding of diets with 25. 12 
3125 ppm Firemaster FF-1, a polybrominated I 
Where X « log ppm 6-BB and Y « hatchability 
x/%« the response lines are: (a) A derived equa 
- 321.66-139.07x; (b) a response line for data 
depletion phase where Y » 24.0 + 95.8x-48.1> 
linear portion is described by Y = 160.9-68.1 
response line for data from the steady-state pha 
Y - 96.8 + 10.7x-18.8x^ whose linear porti< 
scribed by Y = 173.4-65.8x; (d) summation < 
"b" and "c" where Y = 56.8 + 37.6x-23.3> 
linear portion is Y = 167.2-67.3x, and; (e) a 
line for egg samples with 6-BB between 30- 
where Y - 127.6-43.6x. 



for differential metabolism of the isomers that 
comprise BP-6, and thus FF-1. The supposi- 
tion to consider is that not all isomers of these 
PBBs gave equivalent toxicity and that this 
would account for our inability to find a close 
correlation between 6-BB and embryo toxic- 
ity over that wide range of 6-BB levels in 
eggs, and the incompatibility between derived 
and actual curves of 6-BB in eggs vs. hatch- 
ability. On this basis, these hatch and residue 
data indicated that analysis for FF-1 based 
on the analysis of only the 6-BB peak was not 
a definitive approach toward assessing toxic- 
ity of BP-6 or FF-1. 

Other isomers, and their metabolites will 
have to be considered in the overall relation- 
ship of ingested PBBs to the residues in tissues 
and their toxicity of the compounds. 

Summary, A relationship between poly- 
brominated biphenyl, Firemaster FF-1 (FF- 
1), in the diet, and eggs, as monitored by 
hexabromobiphenyl (6-BB), and embryo tox- 
icity, as measured by hatchability, were ex- 
amined. The minimum dietary level of FF-1 
for an effect on hatchability was estimated at 
42 ppm, which produced an egg residue esti- 
mated at 65.9 ppm 6-BB. Generally, as FF-1 
in the diet increased, egg residues increased 
and hatchability decreased. Regression equa- 
tions were established for these relationships. 
After withdrawal of FF-1 in the diet, hatch- 
ability returned to normal when FF-1 treat- 

ments had been <625 ppm. Over a range of 
30-85 ppm 6-BB in eggs there was poor cor- 
relation to an effect on hatchability. These 
latter data were discussed in terms that the 6- 
BB peak may not be a defmitive approach to 
assess FF-1 toxicity. 

The authors express their appreciation for assistance 
in this project to Mr. Sulo Hulkonen, Mr. Edward Ko- 
waleski, Mr. Dennis Dodson, and Ms. Melinda Neff. 

1. Carter, L. J., Science 192, 240 (1976). 

2. Fries, G. F., Smith, L. W., Cecil, H. C, Bitman, J., 
and Lillie, R. J., Presented at 165th Mtg., Amer. 
Chem. Soc., Paper #52, Pesticides Chem. (1973). 

3. Cecil, H. C, and Bitman, J., Poultry Sci. 57, 1027 

4. Ringer, R. K., and Polin, D., Fed. Proc. 36, 1894 

5. Polin, D., and Ringer, R. K., Environ. Health Per- 
spect. 283(1978). 

6. Babish, J. G., Gutenmann, W. H., and Stoewsand, 
G. S., J. Agric. Food Chem. 23, 879 (1975). 

7. Cecil, H. C, Bitman, J., LiUie, R. J., Fries, G. F., 
and Verrett, J., Bull. Environ. Conum. Toxic. 11, 

8. Ringer, R. K., and Polin, D., Poultry Sci. 54, 1810 

9. Snedecor, G. W., and Cochran, W. G., Sutistical 
Methods. 6th ed. Iowa State Univ. Press (1%8). 

10. Fries, G. F., Cecil, H. C, Bitman, J., and Lillie, R. 
J., Bull. Environ. Contam. and Toxic. 15, 278 ( 1976). 

Received February 10, 1978. P.S.E.B.M. 1978, Vol. 159. 


Intraerythrocyte pH and Physiochemical Homogeneity^ (40299) 


Blood Research Laboratory, New England Medical Center Hospital and the Department of Medicine, Tufts V 
School of Medicine, 171 Harrison Avenue, Boston, Massachusetts 021 1 1 

Intraerythrocytic pH is a major determi- 
nant of glycolytic metabolism, membrane 
function, and oxygen dissociation. We have 
investigated the possibiUty that in a physi- 
ochemical sense these functions are con- 
trolled by separate pH environments within 
the red cell. 

Caldwell (1), Adler et al (2), and Waddell 
and Bates (3), have demonstrated that the 
results of the determination of intracellular 
pH (pHi) by the distribution of a weak base 
differs from that determined by the distribu- 
tion of a weak acid in a heterogeneous system. 
Using a weak acid, the pHi closer to the 
highest pH of the various intracellular com- 
partments is recorded, while a weak base 
reflects a value closer to the lowest pH of the 
compartments. If the pH^ as determined by a 
weak acid and a weak base are identical, then 
the cell interior is Ukely to be homogeneous 
(1). Using rat diaphragm muscle Adler (4) 
showed that the pHi measured by the 
weak acid 5,5-dimethyloxazolidine-2,4- 
dione (DMO) was signiflcantly higher than 
the pHi measured by the weak base, nicotine. 
Physiochemical inhomogeneity, therefore, 
was demonstrated, as one might expect on 
morphological grounds. Using the same weak 
acid and weak base, we have investigated the 
pHi of the human erythrocyte. 

Methods. Venous blood was mixed with 
14.3 jLtg of heparin per ml, centrifuged, and 
the buffy coat removed. The erythrocytes 
were resuspended in their own plasma. 

Radioactive DMO, 5,5-dimethyloxazoli- 
dine-2,4-dione-[2-*^C] (New England Nu- 
clear, Boston, MA), specific activity 11 

' These studies were supported by grant number HL- 
15157 from the National Institute of Heart and Lung 

■ Current Address: James Warth, M.D., Assistant Pro- 
fessor of Medicine, Wayne Stale University, Department 
of Hematology, Harper-Grace Hospitals, 3990 John R., 
Detroit, Michigan 48201. 

mCi/mmole, was added to give a fin; 
centration of 0.00066 mg/ml of blood. \ 
active nicotine, nicotine-methyl-[^^C] 
linckrodt Corp., St. Louis, MO), speci 
tivity 2.41 mCi/mmole, was added to 
ond aliquot of blood to give a final c 
tration of 0.01 1 mg/ml. Both tubes of 
were incubated at 37° for 20 min, a ti 
determined experimentally as adequi 
both DMO and nicotine to reach equili 
The pH, pHe, of each aliquot was me 
to the nearest 0.01 unit using a Coming 
165 pH blood gas analyzer (Coming 
tific Instmments, Medfield, MA). Tl 
quots were centrifuged, the plasma re 
and a microhematocrit corrected for ti 
plasma [1.31% (5)] determined on the i 
packs. One ml of packed cells or one 
plasma containing either nicotine-^ 
DMO-^^C was added to 5 ml of dei 
water. Each sample was prepared in 
mplicate. One ml of each mixture was 
ted into 2.0 ml of 10% trichloraceti 
(TCA), centrifuged, and 1 ml of the 
natant counted in a Packard Tri-Carb 
Scintillation Counter Model 3320. Cou 
minute were converted to disintegratic 
minute by use of a quench curve. 

Binding studies. To investigate the p 
binding of DMO to human plasma, the 
mel-Dryer technique (6) was utilized 
ml of normal heparinized (14.3 / 
plasma were incubated with 0.01 ml of 
Y^C\ at 37° for 30 min and an aliquot 1 
on to a G-25 Sephadex column with a 
void volume after balancing the colum 
DMO-['^C] in heparinized (15 /ig/ml) 
phate buffered sahne. Aliquots of eacl 
tion were counted and the optical < 
determined at 280 nm. 

Binding studies of nicotine to 1 
plasma and erythrocytes were also cone 
Ten ml of heparinized venous blooc 
incubated with 0.3 mg of nicotine-f^'^C] 
ml of 0.9% sodium chloride while anot 


0037-9727/78/ 1 59 1 -0 1 36$0 1 .00/0 

Copyrighx © 1978 b> the Society Tor Experimental Biology and Medicine 
AJJ rights reserved. 



ml were allowed to incubate with the same 
amount of radioactive nicotine plus 0.85 mg 
of cold nicotine contained in 50 jul of 0.9% 
sodium chloride. Three determinations were 
made from each tube of the ratio of intracel- 
lular to extracellular dpm per g of water. 

Calculations. The external dpm for both 
DM0 and nicotine determinations were con- 
verted to dpm/g of plasma water using 0.94 
as the fraction of solvent water in plasma (7). 

The internal dpm for both DMO and nic- 
otine determinations were converted to 
dpm/g of erythrocyte water after correction 
for the trapped plasma in the erythrocyte 
pack, utilizing 0.59 as the fraction of solvent 
water in the erythrocyte (8-1 1). 

The dpm/g of plasma water and cell water 
are then entered into the appropriate formula 


p#r. + log 

(DMO). (10P""-P^'- -hi)- (DMO)e 

(DMO)e J 

pH, « p#r. - log 
"(nico tine), (IQp^'-p"^ -hi)- (nicotine)e l 
(nicotine)e J 

The p#r« of nicotine was chosen as 7.85, 
the value used by Effros and Chinard (13). 

The pHe used was the average of the two 
measured values, which never differed by 
more than 0.01 of a pH unit. There was no 
significant difference between the pHe values 
measured in the tube containing DMO versus 
the tube containing nicotine. 

Results. The DMO and plasma binding 
study showed no rise or subsequent fall in 
dpm occurring in association with the collec- 
tion of the plasma protein peak. Thus, there 
was no evidence of binding of DMO to hu- 
man plasma. 

The studies done to evaluate the possibility 
of nicotine binding to human plasma or 
erythrocytes showed that the ratio of internal 
to external dpm per g of water was identical 
at the two widely different concentrations of 
nicotine. This is strong evidence against bind- 
ing or active transport of nicotine as well as 
evidence against the permeability of the 
erythrocyte membrane to the nicotine ion 

Table I shows the results of the pHi deter- 

minations using both DMO and nicotine. 
Four determinations using each indicator 
were carried out for each normal sample. The 
range of the four values falls within 0.16 of a 
pH unit. There is no significant difference by 
analysis of the 5 pairs of DMO and nicotine 

Discussion. Caldwell (1) stated that if the 
values for pH, obtained from multiple differ- 
ent indicator methods are in agreement it is 
probable that the cell interior is "reasonably 
uniform". Waddell and Bates (3), using a 
current operational definition of pH, stated 
that in an inhomogenous system, pHi calcu- 
lated from the distribution of a weak acid 
yields a pH value closer to the higher value, 
and that pH^ calculated from the distribution 
of a, weak base yields a result closer to the 
lower pH value in that inhomogenous system. 

Accurate determinations of pHi using such 
indicators depend upon the absence of bind- 
ing, the absence of active transport, and the 
impermeability of the cell membrane to the 
ionic species of the weak acid or weak base 
used. Waddell and Butler (IS) demonstrated 
that DMO is not significantly bound to bo- 
vine serum albumin. Calvey (16) showed that 
DMO is not bound to, or actively transported 
by, rabbit erythrocytes. Bromberg et al. (17) 
showed that human erythrocytes do not bind 
DMO. We have demonstrated that DMO is 
not bound to human plasma and that nicotine 
meets the three criteria for accurate indicator 
compounds set forth above. 

We have chosen 0.59 gm H2O/IOO ml of 

TABLE 1. pHi (Average Appears Above the 

Range) Determined by DMO and Nicotine in 

Human Erythrocytes. 



































No sign 

ficant difference" 

" Calculations based on paired data analysis on five 



cells as the fraction of solvent water in the 
erythrocyte. This value is amply supported 
(10-13). However, values up to 0.72 gm 
H2O/IOO ml of cells (9) can be defended. Use 
of the latter value would assume no bound 
water and would produce a DMO value 0. 1 
of a pH unit lower and a nicotine value 0. 1 of 
a pH unit higher than the average values we 
calculated. Such a resuh would be at variance 
with the theoretical considerations presented 
by Waddell and Bates (3). 

In performing the above experiments we 
used erythrocytes of various ages. Assuming 
that each cell is homogeneous but that they 
vary somewhat in intracellular pH as a func- 
tion of age, the pH, determined by DMO 
should be equal to the pHi measured by 

There are only two results possible if a 
weak acid and a weak base are used to deter- 
mine the pHf. Either the weak acid gives a 
higher pHi reading than the weak base, in 
which case the system is heterogeneous, or 
the readings are equal, in which case the 
system is homogeneous. Our experiments uti- 
lized erythrocytes from four normal donors 
and the same indicator compounds, DMO 
and nicotine, that were used to demonstrate 
physiochemical heterogeneity of skeletal 
muscle cells (4). Our results show no signifi- 
cant difTerence between the pH, value ob- 
tained by DMO and that obtained by nico- 
tine. It is possible that in a heterogeneous 
system pH, values determined by these com- 
pounds could, fortuitously, turn out identical, 
as has been suggested by Carter (18) in his 
report of equal pH, values determined by 
DMO and nicotine in barnacle muscle, an 
apparently heterogeneous system. We have 
no proof that this has not occurred here but 
it would seem unlikely given the difference 
between the erythrocyte and barnacle muscle 
fiber microscopically. Further, Bone et al 
(19) using a single donor showed no signifi- 
cant difference between the hydrogen-ion 
concentration in erythrocytes as determined 
by the weak acid DMO and the weak base 

We conclude that the internal pH of the 
normal human erythrocyte is uniform 
throughout the cell and that this cell is phys- 
iochemically homogeneous. This information 
supports the concept that in the human eryth- 
rocyte hydrogen-ion dependent processes 

such as glycolytic metabolism, mei 
function and oxygen dissociation an 
lated by a single value for each cell. 

Summary. In order to determine tb 
iochemical homogeneity of the hum 
cell, intraerythrocyte pH was simultai 
measured using the weak acid S,5-dii 
oxazolidine-2,4-dione (DMO) and th 
base nicotine. If a cell is homogeneou 
measurements will yield the same resi 
cell is heterogeneous, the DMO readi 
be closer to the highest pH in the cell : 
while the nicotine will read closer 
lowest pH. The results show no sigi 
difference between the intracellular | 
tained by either of these methods (ave 
DMO = 7.10, by nicotine = 7.06 at a 
age external pH of 7.33). We conclu 
the human erythrocyte is physioche 

1. Caldwell, P. C, in 'international Review i 
ogy" (G. H. Bourne and J. F. Danielli, cds. 
Academic Press, New York (1956). 

2. Adler. S., Roy, A., and Relman. A. S., J. Clii 

3. Waddell, W. J., and Bates, R. G., Phys. Re> 

4. Adler, S., J. Clin. Invest. 51, 256 (1972). 

5. Garby, L., and Vuille, J. C, Scand. J. L 
Invest. 13,642(1961). 

6. Ackers, G. K., in "Methods in Enzymolog 
Colowick and N. O. Kaplan, eds.). Vol. 2' 
Academic Press, New York (1973). 

7. Dittmer, D. S., in "Blood and Other Body F 
19. Federation of American Societies for Ex 
tal Biology, Washington, D.C. (1961). 

8. Agostoni, A., Berfasconi. C, Gerli, G. C, 
M., and Rossi-Bernardi, L., Science 182, 3G 

9. Drabkin, D. L., J. Biol. Chem. 185, 231 (19 

10. LeFevre, P. G., J. of Gen. Phys. 47, 585 (I* 

11. Savitz, D., Sidel, V. W., and Solomon, A. 
Gen. Phys. 48, 79 (1964). 

12. WaddeU, W. J., and Butler, T. C, J. Clin, li 

13. Effros, R. M., and Chinard, F. P., J. Clin. Ii 

14. Butler, T. C, Waddell, W. J., and Poole, D 
Proc. 26, 1327(1967). 

15. Waddell, W. J., and Butler, T. C, Proc. i 
Biol. Med. %, 563 (1957). 

16. Calvey, T. N., Experientia 26/4, 385 (1970; 

17. Bromberg. P. A.. Theodore, J., Robin. E. 
Jensen, W. N.. J. Lab. Clin. Med. 66, 464 ( 

18. Carter, N. W., Kidney International I, 341 

19. Bone, J. M., Verth, A., and Lambie, A. T., • 
andMol. Med. SI, 189(1976). 

Received April 10, 1978. P.S.E.B.M. 1978, Vol. 


Stimulation of Erythropoietin Secretion by Single Amino Acids (40300) 


tments of Pathology and Medicine, Abraham Lincoln School of Medicine and Medical Service, Veterans 
tration West Side Hospital, Chicago, Illinois 60612, and Department of Medicine. Michael Reese Hospital, 

Chicago. Illinois 60616 

in deprivation in rats results in a de- 
>f the amount of erythropoietin (Ep) 
id in response to hypoxic stimulation 
s effect is rapidly reversed if the pro- 
irived animals are fed a single protein 
>nsisting of albumin or hemoglobin 
)r to or shortly after the onset of the 

stimulus (2). The study detailed be- 
sides evidence that in protein de- 
rats erythropoietin secretion can be 
led by individual amino acids. 
rials and Methods. Female Sprague- 

rats weighing 100-200 g were used, 
al diet in pellet form containing less 
)% protein (*'protein-free"), but oth- 
lutritionally complete, was purchased 
irina-Ralston Co., Missouri. Solutions 
I amino acids in 3 cc of distilled water 
ven with a blunt-ended #18 needle 

per OS into the rat stomach. To dis- 
e amino acids, the pH of the solution 
bt changed at times from moderately 
H 3.0) to strongly alkaline (pH 9.0). 
^r, in each individual experiment, the 
he water fed to control rats was ad- 
3 that of the test amino acid solutions. 
[Terence in the pH of the various so- 
fed to the rats was not found to afTect 
asma Ep levels. 

iropoietin production was stimulated 
ing the animals in a hypobaric cham- 

exposing them to 0.5 atmosphere for 
;. Immediately afterwards, rats were 
linated by cardiocentesis and the 

obtained from each experimental 
4-5 rats) was pooled and assayed for 
nation of Ep levels in posthypoxic 
hemic mice by the method of Gordon 
jintraub (3) (6-8 assay mice each re- 
).5 ml of pooled plasma). It should be 
hat minor variations in the spring 
ism of the regulatory valve of our 
ric chamber result in some inconsis- 
1 the chamber pressure from experi- 

ment to experiment. Therefore, a control 
group of rats fed water adjusted to the pH of 
the amino acid preparations were always in- 
cluded in each experimental trial. The statis- 
tical significance of the differences was deter- 
mined by the Student's t test. 

Results. Effect of single L-amino acids (Ta- 
ble I). Rats fed a protein-free diet for 6 days 
were fed 1(X) mg of an L-amino acid and were 
immediately afterwards exposed to hypoxia. 
Table I shows the mean plasma Ep levels of 
rats fed various amino acid solutions. The 
data indicate that L-methionine, L-cystine 
and L-leucine produced the most intense and 
consistent stimulation of Ep production (P 
< 0.001). L-Tyrosine and L-asparagine pro- 
duced a small but still significant (P < 0.05) 
increase in plasma Ep levels whereas the rest 
of the amino acids had no significant effect 

Effect of various doses of L-amino acids 
(Tables II and III). Protein-deprived rats fed 
from 25 to 100 mg L-methionine or L-cystine 
prior to hypoxia had significantly higher 
plasma Ep levels compared to the control 
group. Increasing the amount fed to 800 mg 
did not produce a further increase in Ep 
production and may have been inhibitory. 
When histidine or glycine was fed in doses of 
10 to 400 mg per rat, no increase in the 
posthypoxic plasma Ep levels was detected. 

Discussion. Decreased Ep production oc- 
curs in the presence of protein deficiency (1). 
This decrease has been related by some to the 
depression of basal metabolism associated 
with starvation (4). We have recently dem- 
onstrated that a single feeding of protein 
(hemoglobin or albumin) to protein deprived 
rats produces an immediate enhancement of 
Ep production which is dose related and 
which does not correlate with changes in the 
oxygen consumption of the animals (2). We 
concluded that the production of erythropoi- 
etin depends not only on oxygen supply vs 


003*7 -9ra n%l \'i>^\-^\')ms\ .^ 1^ 



TABLE I. Effect of Feeding a Single l-Amino 

Acid on Posthypoxic Plasma Ep Levels of Protein 

Deprived Rats. 

% 59Fc uptake into RBC's of 
assay mice (mean ± 1 SEM) 

Amino acid fed 
(100 mg) 

Control (H2O) 

AA group 

Alanine (3) 
Arginine (3) 
Valine (3) 
Serine (4) 
Methionine (7) 
Cystine (4) 
Tyrosine (3) 
Tryptophane (2) 
Phenylalanine (2) 
Leucine (5) 
Isoleucine (2) 
Histidine (4) 
Asparagine (4) 
Glycine (6) 
Lysine (4) 
Glutamic acid (4) 
Aspartic acid (4) 
Threonine (4) 
Proline (5) 
Cysteine (6) 

2.83 ± 0.65 
2.83 ± 0.66 

4.75 ± 1.10 
4.32 ± 0.79 
3.08 ± 0.33 
2.79 ± 0.73 
1.96 ±0.16 

2.55 ± 0.87 
2.86 ± 0.56 
2.06 ±0.12 
2.27 ± 0.03 
3.16 ±0.55 
2.18 ±0.07 
3.98 ±1.01 

3.76 ± 1.12 

3.56 ± 1.25 
3.56 ± 1.25 
4.12 ± 1.00 
4.93 ± 1.17 
3.50 ± 0.69 

3.94 ± 1.00 
4.04 ± 1.27 
5.97 ± 1.70 
3.00 ± 0.98 

11.00± 1.19- 
10.10 ±0.60" 
4.74 ± 0.98* 

1.61 ±0.18 
2.41 ±0.28 
4.52 ± 0.44" 
4.29 ± 1.11 
3.34 ± 0.32 
4.84 ±0.51* 

5.95 ± 1.58 
4.29 ± 1.38 
6.99 ±2.10 

5.62 ± 1.53 
6.90 ± 1.99 
8.77 ± 2.92 
5.34 ± 0.99 

Numbers in parentheses signify the number of exper- 
imental trials conducted. 

TABLE II. Effect of Feeding Various Doses of l- 

Methionine and l-Cystine on Posthypoxic Plasma 

Ep Levels of Protein Deprived Rats. 

% 59Fe Incorporation 
assay mice (mean 

i into RBC's of 
± 1 SEM) 

Amount fed 




25 mg 
50 mg 
75 mg 
100 mg 


100 mg 
400 mg 
800 mg 

2.09 ± 0.25 

8.55 ±4.61 

8.41 ±3.39 

10.02 ± 3.90 

10.43 ± 1.96 

3.68 ± 0.74 
12.67 ± 1.05 
12.15 ± 1.09 

5.70 ± 0.27 

2.74 ± 0.23 
4.20 ± 1.41 
7.00 ± 2.05 
4.91 ± 1.07 
5.79 ± 0.23 

1.50 ±0.21 
9.19 ±0.94 
6.56 ± 1.41 
5.41 ±0.32 

demand of the Ep producing sites, but also 
on the continuous supply of amino acids (2). 
The present experiments were done to deter- 
mine whether individual amino acids were 
important for the biosynthesis of erythropoi- 
etin as occurs with other polypeptide hor- 
mones (insulin, growth hormone) (5). Only 
three amino acids, (methionine, leucine and 
cystine) had a significant efTect in raising the 

TABLE III. Effect of Feeding Various Doses of ] 

Histidine or l-Glycine on Posthypoxic Plasma E; 

Levels of Protein Deprived Rats. 

% 59Fe Incorporation into RBCs of 
assay mice (mean ± 1 SEM) 

Amount fed 




4. II ±0.47 

3.18 ±0.27 

10 mg 

4.00 ± 0.60 

3.66 ± 0.23 

50 mg 

2.51 ±0.58 

4.63 ± 0.56 

100 mg 

4.24 ± 0.50 

4.51 ±0.75 

400 mg 

3.32 ± 0.40 

2.18 ±0.21 

plasma Ep levels of the protein deprived rats. 
Other amino acids had minimal or no efTect. 
There is a parallel to this finding in the 
studies which show a great variation in the 
ability of single amino acids to stimulate se- 
cretion of insulin or growth hormone (6, 7). 

Some amino acids which had no efTect at 
the 100 mg dose level were tested at smaller 
doses to determine whether their dose re- 
sponse curves were maximal at the lower 
levels. The results were negative. The data 
also suggest that higher doses of cystine and 
methionine may be inhibitory. We have no 
explanation for this possibility, although 
large doses of amino acids may suppress the 
transport of other amino acids across cell 
boundaries (8). 

Summary. Protein deficiency in rats results 
in decreased ability to produce erythropoietin 
after hypoxic stimulation. This defect can be 
reversed by a single protein feeding at the 
time of exposure to hypoxia. The present 
experiments show that feeding of methionine, 
leucine or cystine also corrected the defect in 
erythropoietin production. These amino acids 
may serve to signal the adequacy of protein 
reserves and permit the synthesis of erythro- 
poietin. Other single amino acids had mini- 
mal or no effect. 

Study has been supported by University of lUinois 
BRSG 7612, Veterans Administration Research Funds, 
funds from the American Cancer Society, lUinois Divi- 
sion. Leukemia Research Foundation and the Hematol- 
ogy Research Foundation. The authors wish to thank 
Mrs. Jan Rone for her technical assistance. 

1. Reissman, K. R., Blood 23, 146 (1964). 

2. Anagnostou, A., Schade. S., Ashkinaz, M., Barone. 
J., and Fried, W., Blood 50, 1093 (1977). 

3. Gordon. A. S., and Weintraub, A. H., in "Erythro- 
poiesis*' (L. O. Jacobson and M. Doyle, eds.) p. 1 
Grune and Stratton, New York (1962). 



4. Krantz, S. B., and Jacobson, L. C, "Erythropoietin C, Guntsche, E. M., and RuU, J. A., J. Clin. Endo- 
and the Regulation of Erythropoiesis," p. 17. Uni- crinol. Metabol. 25, 1140(1965). 

versity of Chicago Press, Chicago, (1970). 8. Christensen. H. N., in "Free Amino Acids and Pep- 

5. Eisensiein, A. B., Amer. J. Clin. Nutr. 21,467 (l%8). tides in Tissues, Vol I", (H. N. Munro, ed.) p. 105. 

6. Floyd, J. C. Fajans, S. S., Conn, J. W., Knopf, R. Academic Press, New York, 1964. 

F., and RuU. J. A., J. Clin. Invest. 45, 1479 (1966). 

7. Knopf, R. F., Conn. J. W.. Fajans, S. S., Floyd, J. Received April 17. 1978. P.S.E.B.M. 1978, Vol. 159. 


Ornithine Decarboxylase Activity in Cells Acutely and Chronically Transformed by 

Murine Sarcoma Virus (40301) 


NCI' VA Medical Oncology Unit, National Cancer Institute, Bethesda, Maryland 20014 and Veterans Administration 

Hospital, Washington. DC 20422 

Polyamine biosynthesis is one of the earli- 
est events occurring during cellular prolifer- 
ation (1). Ornithine decarboxylase (ODC), 
which catalyzes the formation of putrescine 
from ornithine, is the rate limiting enzyme in 
polyamine biosynthesis (2). Resting cells have 
low, stable ODC levels which increase rapidly 
upon the onset of growth (3, 4). We have 
demonstrated that increased ODC activity 
follows infection of cultured mouse Balb/3T3 
(B/3T3) cells with murine sarcoma virus 
(MSV) (5). The increase in ODC activity is 
independent of the population doubling time 
and commences immediately prior to mor- 
phological transformation. Elevation of ODC 
levels also precedes morphological transfor- 
mation by Rous sarcoma virus (6). 

Transforming stocks of MSV consist of 
mixtures of defective transforming virus and 
non-transforming murine leukemia virus 
(MuLV) (7). The MuLV is usually present in 
great excess, and dual infection of mouse cells 
with both viruses is required for MSV repli- 
cation. Cells infected with the transforming 
virus alone undergo transformation, and re- 
tain the sarcoma genome, but do not release 
infectious virus. Two such classes of trans- 
formed cells have been described: (a) Non- 
producer (np) which do not release virus 
particles (8), and (b) sarcoma virus positive, 
leukemia virus negative (S-l-L— ) cells which 
release noninfectious virus particles and have 
MuLV gs antigen (8). Superinfection of both 
of these transformed cell classes with MuLV 
results in release of infectious transforming 
and nontransforming viruses. However, su- 
perinfected S-l-L— cells undergo further mor- 
phological alteration (thus providing a focus 
assay for MuLV). Superinfection of np cells 
does not result in morphological alteration. 

' Send rcprinl requests to A. F. Gazdar, MD, NCI- 
VA Medical Oncology Unit, VA Hospital Washington, 
D.C. 20422. 

In this communication we describe exper- 
iments studying the relationship between el- 
evated ODC activity, virus induced morpho- 
logical transformation, virus production, and 
rates of cellular division. We compare pro- 
ducer, np and S-l-L— derivatives of a single 
murine cell clone. 

Materials and methods. Cell lines. B/3T3, 
clone A3 1, is a contact inhibited, 'flat* non- 
virus releasing cell (9). It becomes trans- 
formed after MSV infection, but productive 
infection with MuLV does not induce mor- 
phological change. D245E6 is a S+L- B/3T3 
clone selected for its relative 'flatness'. After 
MuLV superinfection, its morphology be- 
comes more transformed (10). KA31 is a 
Kirsten MSV transformed np clone of B/3T3 
(11). MuLV superinfection of KA31 results 
in release of transforming and nontransform- 
ing viruses without morphological change. 
B/3T3 and KA31 cells were obtained from 
Dr. Stuart Aaronson and D24SE6 cells from 
Dr. Robert Bassin. Cells were maintained in 
75 mm flasks in 5% CO2 atmosphere at 37°. 
Fluids were changed at 24 or 48 hr intervals. 
Cells were grown in Eagle's essential medium 
(D245E6) or Dulbecco's modification of it 
(B/3T3 and KA31). Medium was supple- 
mented with 10% heat inactivated (56**, 30 
min) fetal bovine serum, penicillin (100 
units/ml) and streptomycin (1(X) /xg/ml). 

Viruses and virus assays. (3z-MSV, a mix- 
ture of transforming and nontransforming 
viruses, was recovered from the supernatant 
fluids of acutely infected B/3T3 cells, and 
had a titer of 2 x 10^ focus forming units/ml 
(12). Rauscher leukemia virus, a strain of 
MuLV, was obtained by concentrating the 
supernatant fluids of chronically infected 
BALB/c JLSV-9 cells, and had a titer of 1 
X 10^ plaque forming units/ml. Infectious 
center assays were modifications of the com- 
monly used methods for assays of MSV and 
MuLV (8, 13). Fifty or 100 mitomycin C 


0037-9727/78/ 1591-01 42$0 1 .00/0 

Copyright © 1978 by the Society for Experimental Biology and Medicine 
AJJnghis reserved. 



d (25 fig, 1 hr) test cells were added onto 
^ly seeded indicator cells. For MSV as- 
the indicator cells were B/3T3, and foci 
Tactile MSV transformed cells were enu- 
;ed 5 days later. For MuLV assays, 
- Al-2 cells (14) were used as indicator 
and plaques (consisting of supertrans- 
id cells that had lysed or floated away) 
erated 5 days later. Colony forming ef- 
cy (CFE) in semi-solid medium was 
nined by suspending 1x10^ viable cells 
^% agarose over a 0.9% agarose base. 

1, 8 and IS days, another layer of 
se was placed over the cell containing 

Colonies were counted 18 days after 

ler Methods, ODC activity was assayed 
easuring enzyme released CO2 as de- 
id previously (S). Replicate plates were 
I twice with saline and frozen (-20°) 
assayed. Cells were gently scraped into 
r, freeze-thawed three times, and centri- 
(4S00g for 10 min). Supernatant fluids 
ml) were incubated with 50 /d [^^C]- 
line in plastic tubes equipped with a 
;r stopper supporting a polyethylene 
r well. After incubation (37°, 45 min), 
J hydroxide of hyamine was added to 
weU. After a further incubation of IS 
0.2 ml of perchloric acid was added to 

each well. Tubes were agitated for IS min to 
release bound CO2, the center wells were 
removed, and their radioactivities deter- 
mined. Protein was determined by the Lowry 
method (12). Cells were counted with a he- 
mocytometer, and viability determined by 
trypan blue exclusion. 

Results, Properties of the cell lines used are 
presented in Table I. Uninfected B/3T3 cells 
were epithelioid and contact inhibited, did 
not release virus, and failed to grow in soft 
agarose. Productive infection with MuLV did 
not alter its morphology. Within 48 hr of 
MSV infection, B/3T3 cells became round or 
spindle shaped, were highly refractile and 
adhered poorly to the substrate. Morpholog- 
ical transformation was accompanied by 
release of transforming and nontransforming 
viruses, and the ability to grow in soft agarose 
at low efficiency. Uninfected D24SE6 cells 
were large polygonal cells with slight overlap- 
ping of their edges, which grew in soft agarose 
but did not release infectious virus. On su- 
perinfection with MuLV, D24SE6 cells un- 
derwent ftirther morphological transforma- 
tion, and closely resembled MSV infected 
B/3T3 cells. The superinfected cells released 
both MSV and MuLV, but their ability to 
grow in soft agarose decreased. Uninfected 
KA31 cells were small and highly refractile. 

TABLE 1. Characteristics of Control and Virus Infected Cells. 


% Infectious Centers* 

% CFE in soft 


OIX: activity 











n -»- MuLV 





r3 -»- MSV 












> + MuLV 






sfer 0) 

S -»- MuLV 






sfer 4) 






-»- MuLV 






sfer 0) 
-»- MuLV 






sfer 4) 

msformed morphology arbitrarily graded as follows: O contact inhibited, nonrefractile cells similar to parent 
-»- nonrefractile cells with some cellular overlapping; +-!■ refractile cells with formation of dense cellular 
and -f +-1- highly refractile cells with scant cytoplasm and poor anchorage dependency, the cells tended to 

to the supernatant fluid prior to reaching confluency. 

t«nt foci 5 days after plating control or infected cells on Al-2 (MuLV assays) or B/3T3 monolayers (MSV 

jiimum ODC activity is the highest measured level of enzyme activity, usually occurring four days after 



with short spindly processes. They grew in 
soft agarose before and after superinfection 
with MuLV, but released MuLV and MSV 
only after superinfection. 

Growth characteristics of the cells em- 
ployed are shown in Fig. 1. The growth rates 
of uninfected B/3T3 and D245E6, and 
MuLV infected B/3T3 were similar, while 
uninfected and infected KA31 cells grew 
more rapidly and reached a higher cell den- 
sity at day 7. As noted previously (5) MSV 
transformed B/3T3 cells grew slower than 
uninfected cells, although the differences 
were not marked in the present experiment, 
when tissue culture grown virus stocks were 
substituted for animal tumor harvests. MuLV 
superinfection of D24SE6 cells resulted in a 
considerable increase in the population dou- 
bling time, but had no effect on the growth 
of K A3 1 cells. Trypan blue exclusion studies 
revealed less than 2% nonviable cells in con- 
trol and virus infected cell lines at all obser- 
vation points. 

ODC levels of cell lines were measured 1, 
3, 4, 5 and 7 days after seeding. The highest 
levels measured (usually occurring 4 days 
after seeding) are shown in Table I, and the 
entire curves are presented in Fig. 2. Rela- 
tively low ODC levels occurred after seeding 
and at confluence. Comparable data were 
obtained when ODC activity was expressed 
as a function of cell number or of cellular 
protein. Control B/3T3 and D245E6, and 
MuLV infected B/3T3 had relatively low 
'maximum' levels (ie. the highest levels mea- 
sured). MSV infection of B/3T3 and MuLV 
infection of D245E6 cells resulted in 20- and 
sevenfold increases respectively in maximum 
ODC activities. Uninfected KA31 cells had 

a higher baseline ODC activity than the other 
cell lines, but superinfection resulted in a 
twofold increase only. 

The temporal relationships between eleva- 
tion of ODC activity, morphological trans- 
formation and virus production were also 
studied. After four passages MSV trans- 
formed B/3T3 and MuLV superinfected 
D245E6 and KA31 cells had not further al- 
tered morphologically, or in their ability to 
release transforming and non transforming 
viruses, or grow in soft agarose (Table I). 
However, ODC levels of MSV infected 
B/3T3 cells fell rapidly after transfer (Fig. 2), 
while the drop in ODC levels of superinfected 
D245E6 cells was smaller and took longer. 
ODC levels of superinfected KA3 1 cells fell 
only slightly, but the baseline levels were high 
and the initial rise on superinfection was 

Individual clones of B/3T3 and MSV 
transformed B/3T3 cells were selected after 
isolation in liquid or semi-solid media. Char- 
acterization of the 13 clones selected and their 
ODC data are presented in Table II and Fig. 
3. Uninfected B/3T3 clones (numbers 1-5) 
had a flat morphology, did not release viruses, 
failed to grow in soft agarose, and had low 
maximum ODC levels. MSV transformed 
clones 6-12 had higher ODC levels, which 
appeared related to the degree of morpholog- 
ical transformation. Although isolated from 
MSV infected cells, clone 13 released only 
MuLV, had a flat morphology, did not grow 
in soft agarose, and had low ODC activity. 
Presumably this clone arose from a cell in- 
fected with the nontransforming component 
of MSV only. With one exception, all clones 
had maximum ODC activities 4 days after 

13 5 7 

Days after Seeding 

13 5 7 
Days after Seeding 

13 5 7 
Days after Seeding 

Fig. I. Growth curves of uninfected and virus-infected cell lines. Cells were infected in suspension with MuLV 
at a multiplicity (MOI) of 3:1 or MSV (MOI 10:1) at 37° for I hr prior to seeding. 


Ssl^Q D24S KA31 






I too 




•«■ Bai>/3T3 





200 ' 






*iBl>245 800 

DA o MuLV 44lt*Ct*d 





Fig. 2. ODC activity of control and transformed Balb/3T3 cells. In the upper panel, cells were infected 
mediately prior to seeding and harvested on days 1, 3M. 5, and 7. ODC activities in bar graphs (lower panel) 
nsent the oiaximum levels measured during weekly cell passages. 

TABLE II. Characteristics of Control, Transformed and Virus-Infected Balb/3T3 Clones.' 

% Infectious centers 

Maximum ODC 

1 ranstormed mor- - 

% Crt. m doit 


Clone # 





(pmol/ 10** cells) 









































































"Uninfected B/3T3 clones (#1-5) were isolated from liquid medium. MSV transformed clones were obuined 
om liquid (#7-12) or semi-solid media (#6 and 13). Clones were transferred 18 days after seeding, and analyzed 
)dtys later. See also legend of Table 1. 

«ding. The exceptional clone divided slower 
lan the others and was still in exponential 
owth phase at day 7. 
Discussion, While previous studies have in- 

dicated that virus induced transformation re- 
sults in increased intracellular ODC levels, 
the relationship is complex. Our present ex- 
periments further define the association by 





4 7 

Days aftar Infaction 

Fki. 3. ODC Activity of control and MSV infected 
B«lb/3T3 Clones. The data of the control clones with 
the highest and lowest ODC values are displayed. The 
remaining three control clones had intermediate values 
lying within the shaded areas. One clone isolated from 

the MSV infected culture (O O) was nontransformed 

and only released MuLV. The ODC values of one MSV 
transformed clone fell within the shaded area and is not 

Studying several parameters, including mor- 
phology, growth rate, virus production and 
time, ODC levels of transformed and non 
transformed cells alter with cell growth, the 
highest levels occurring during logarithmic 
growth. Elevation of ODC levels (during cell 
growth) accompanies cellular morphological 
change to a transformed or more transformed 
phenotype. Following infection, B/3T3 man- 
ifests considerable changes in both morphol- 
ogy and ODC activity, D245E6 more modest 
alterations, and KA31 essentially none. The 
elevated ODC levels accompan\ing transfor- 
mation cannot be explained by increases in 
cell growth rate^ doubling times (B/3T3 and 
D245E6) arc lengthened or unahered (K,A3n 
at>er virus infection. ODC elevation is also 
IKM related to release of transforming or non 
iransfortning viruses The cloning experi- 
nvuils indicate that the relatively few trans- 
formed clones so obtained have higher ODC 
ItN^ls (during cell grv>wih) than non trans- 
fom^ed cK>ncs. While translormed clones 
ha\^ a wKk renage of ODC activities, acute 
vims transtomiation is cv>nsisiently accom- 
panied b\ a very high ekxaiion. With oell 
passage. OIX" k\^ls of acuteh transformed 

cells return towards baseline levels, perha| 
because most acutely transformed cells fail ] 

Our findings that elevated ODC levels a( 
company acute virus induced morphologia 
transformation may be explained by the re 
cent report of Isom (16). She found tha 
infection of fibroblasts by potentially onco 
genie human cytomegalovirus (CMV) rapid!; 
induced a multiplicity dependent increase ii 
ODC activity. Isom's experiments indicat 
that CMV infection overrides end produc 
repression of ODC by putrescine. Thus th 
oncogenic potential of a virus may be relatei 
to its ability to interfere with normal regula 
tory functions of key cellular metabolic en 

Summary. Ornithine decarboxyasc (ODC 
activity increases when cells are acutely trans 
formed with murine sarcoma virus (MSV) 
Three contact inhibited or MSV transfonnc( 
clones of Balb/3T3 were transformed or su 
pertransformed by MSV or its accompanyinj 
non-transforming 'helper' virus (MuLV), anc 
the relationships between ODC activity, mor 
phology, virus production and growth ratcj 
were examined. Clones isolated from lhes< 
lines were also studied. AU of the virus in 
fected lines released both MSV and MuLV 
ODC activities could not be correlated wilt 
differences in growth rates. The only consist 
ent relationship was between elevated ODC 
activity and acute morphological transfor- 
mation, suggesting that polyamine metabo- 
lism plays a crucial role in the transfonnatioc 
process. With time, the elevated ODC activ 
ities returned towards baseline levels. Thu.< 
ODC activity does not appear to be a useful 
marker for chronic infection or transforma- 
tion by type C viruses. 

The authors thjmk John Minna. Harold StuU. Herber 
Oie, Eduard RusseUL Pathda Hefei and Theresa Gre 
gorio for suggestKXis and assistance. 

I . Bachrach. L" .. 'Functions of Naturally Occuhng Pol 
N^amines* Academic Press^ New York ( 1973). 

: RussclL D , and Sn\tler. S H. Proc NaU. Acad Sd 
I S.A ^LU^OU^ie^SV 

.^ LemKich. K J . Biochun. Biophy. Acu 35i ^ 

4 Ho^an. B L . Biochent Biopliy. Res. Coounun. ^ 

5 Gaidar. A F. SiulL H B. KOhoiL L. J. and B«cb 



rach, U., Nature (London) 2«, 696 (1976). 

6. Don« S., Weiner, H., and Bachrach, U., Cancer Res. 

7. Hartley, J., and Rowe, W., Proc. Natl. Acad. Sci. 
U.S.A. 55, 780 (1966). 

8. Aaronson, S. A., Bassin, R. H., and Weaver, J. Virol. 

9. Aaronson. S. A., and Todaro. C, Science 68, 1024 

10. Gisselbrecht, S., Bassin, R. H., Gerwin, B. I., and 
Rein, A., int. J. Cancer 14, 106 (1974). 

11. Aaronson, S. A., and Rowe, W. P., Virology 42, 9 

12. Gazdar, A. F., Chopra, H. C, and Sarma, P. S., Int. 
J. Cancer 9, 219 (1972). 

13. Bassin, R. H., Tuttle, N., and Fischinger, P. J., 
Nature (London) 229, 564 (1971). 

14. Gazdar. A. F., Russell, E. K., and Minna, J. D., 
Proc. Soc. Exp. Biol. Med. 149, 688 (1975). 

15. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and 
Randall, R. J., J. Biol. Chem. 193, 265 (1951). 

16. Isom, H. C, Proc. Amer. Assoc. Cancer Res. 19, 24 

Received January 19. 1978. P.S.E.B.M. 1978, Vol. 159. 


Suppressed Dietary Inducibility of Glucose 6-Phosphate Dehydrogenase and El 
Cyclic AMP in Acute Hepatic Injury^ (40302) 


The First Department of Internal Medicine, Okayama University Medical School, Okayama 700, Jap 

Glucose 6-phosphate dehydrogenase 
(G6PD, EC is a key enzyme of the 
pentose phosphate pathway and induced by 
dietary glucose and amino acids, but not by 
either alone (1-3). Thus, the dehydrogenase 
level in rat liver is under a dual dietary 
control, i.e. transcriptional and posttranscrip- 
tional regulations; a glucose-dependent step 
of the induction being sensitive to actinomy- 
cin D (3) and blocked by increasing cychc 
3',5'-adenosine monophosphate (cycUc AMP) 
level (4). An entirely different type of G6PD 
induction could be brought about by intoxi- 
cation of rat with carbon tetrachloride and 
other hepatotoxins (5-7). Ahhough the syn- 
thesis de novo of the enzyme protein is in- 
volved in the hepatotoxin-induced increase 
in G6PD activity, it does not require newly 
synthesized RNA (5) and is insensitive to 
manipulations to raise hepatic cyclic AMP 
level (8). We found in our preliminary exper- 
iments with acute thioacetamide intoxication 
of rat that the dietary induction of G6PD was 
markedly depressed in the injured Uver (9). 

A further study of this observation, re- 
ported in the present communication, re- 
vealed that the reduced dietary inducibility 
of G6PD in the acute hepatic injury could be 
explained at least by a dietary unresponsive 
increase in cyclic AMP level in the injured 
Uver. Thioacetamide was chosen in this study 
to produce an acute Uver damage with ele- 
vated G6PD activity, because the intoxication 
with thioacetamide, unlike carbon tetrachlo- 
ride, caused no reduction in dietary intake. 

Materials and methods. Male Sprague- 
Dawley rats, weighing 130-150 g, were de- 
prived of food for 24 hr before intraperitoneal 
injection of 20 mg thioacetamide (Merck Co., 
Darmstadt, Germany; dissolved in saline) per 

' This work was supported by a Grant-In- Aid for 
Scientific Research (B) from the Ministry of Education, 
Science and Culture. 

100 g body weight. The animals were 
fasted for 24 hr and divided into the fc 
three groups: GC, refed on a glucose 
(7:3 in weight) mixture; G, placed oi 
glucose diet; and S, starved for additi 
hr. Control rats received equivalent a 
of saline in place of the thioacetamii 
tion and treated identicaUy with res 
the dietary change. 

The animals, each group consisting 
rats, were kiUed at indicated time inte 
ether overdose (10), which gives 
cycUc AMP and cyclic 3',5'-guanosin( 
phosphate (cyclic GMP) values both 
to those obtained by a freezing mett 
1 1). A smaU portion of the liver was 
removed, weighed (10) and extracte 
by fixing with trichloroacetic acid 
(12). After removal of TCA with eth 
the aqueous extract was evaporated 
under nitrogen and reconstituted in > 
give an original volume. The cycUc nu 
concentrations were determined, aftei 
priate dilution and succinylation, b 
dioimmunoassay method using ' 
cycUc AMP and cyclic GMP kits f 
Shoyu Co., Chiba, Japan) (13). 

The activities of G6PD, low-Km 
nase (EC and glucokina: in Uver supematants and of 
aminotransferase (GPT, EC 
and the contents of glycogen in livei 
were determined as described previo 
7, 14). The enzyme activities and cy 
cleotide concentrations in Uver w 
pressed on the basis of unit supematj 
tein, because the Uver weight increase* 
edly in refed groups of rat due to an e: 
glycogen deposition (Table I). AU th< 
are given as means and standard erroi 
means for each group. Histological 
were made on Uver specimens by a h< 
ylin-eosin staining. 

Results. Alterations of G6PD acti\ 

Copyright (£> 1978 by the Society for Expehmenul Biology and Medicine 
^^ nghts reserved. 


g6pd and cycuc amp in uver injury 


cleotide contents in liver following 
nd thioacetamide treatments are il- 
in Fig. 1 and those of other param- 
Lcpatic injury and dietary effect are 
zed in Table I. The specific activity 

in control animals increased mark- 
a refeeding with glucose and casein, 
dth glucose alone. In thioacetamide- 
Its, G6PD activity increased signifi- 

24 and 48 hr of the hepatotoxin 
I even if the animals were starved 
neration and necrosis of hepatocytes 
dent histologically. These data are 
t with our previous findings (1-3, 5). 
e starved and thioacetamide-treated 
placed on a glucose-casein diet, only 

further increase in G6PD activity 
uccd by the refeeding (GC vs. G or 
trast with the steep rise found in the 
ats. There was no difference in the 
»f diet consumed between the injured 
rol groups. The results apparently 



12 3 

Doyt afttr trtatmtnt 
Activities of G6PD and levels of cyclic AMP 
3MP in livers of control and thioacetamide- 
placed on different dietary regimens. The 
;:ate the time of administration of thioaceta- 
-•) or saline (0..~.. 0).feQaQQQ g, period 
>r all the groups; and ryxx/yx^. period of 
r Groups G and GC and of continued fasting 

indicated that the dietary induction of G6PD 
was impaired in the injured liver despite the 
fact that the enzyme activity was increased 
by hepatic injury itself. That the extent of 
hepatic injury per se was not affected by the 
different dietary treatments was evident from 
the similar increases in low-Km hexokinase 
activity in the three different dietary groups 
of injured rats (Table I). The hepatic level of 
this enzyme increases by liver injury (6, 9, 14) 
and is unresponsive to dietary change as the 
data for control groups given in the same 
table reveal. There were no significant differ- 
ences either in serum GPT activity, a sensitive 
marker of liver injury, among the three in- 
jured groups (Table I). The activity of glu- 
cokinase, another dietary inducible enzyme 
(1, IS), was reduced by thioacetamide intox- 
ication and the induction of this enzyme by 
glucose or glucose-casein refeeding was also 
diminished in the injured livers as may be 
seen in the table. 

The values of cyclic AMP obtained with 
livers of well-fed rats (0-day value in Fig. 1) 
fell in the range of reported values (10-12, 
16). The hepatic cyclic AMP level increased 
significantly in 3 days of starvation in both 
thioacetamide-injured and control groups, al- 
though the extent of the increase was slightly 
larger in the injured group than in the control. 
An important result of this experiment is that 
the rise in hepatic cyclic AMP content on 
prolonged fasting of injured rats could not be 
suppressed by refeeding glucose-containing 
diets in contrast with the rise in the control 
animals. The hepatic levels of cyclic GMP in 
control groups agreed well with the reported 
values (11, 12) and changed little by dietary 
alteration. In thioacetamide-treated rats, 
however, the cyclic GMP content increased 
significantly upon prolonged starvation. The 
increase was much less, although above the 
control levels, in the refed groups of intoxi- 
cated rats. A possibility of overestimating 
cyclic GMP level in the presence of high 
concentrations of cyclic AMP was neglected 
by obtaining constant values with different 
dilutions of liver extract in radioimmunoas- 

In thioacetamide-treated rats, the amount 
of glycogen deposited in the liver after re- 
feeding was significantly less than in un- 
treated rats, even though dietary intakes were 


g6pd and cyclic amp in liver injury 

TABLE 1. Activities of GPT in Sera and of Glucokinase and Low-Km Hexokinase in Livers and 
Contents of Glycogen and Protein in Livers of Control and Thioacetamide-Treated Rats Placed on 

Different Diets. 

Experimental conditions 

Enzyme activities 




Tissue consuiuenis 





~ kinase 


Dietary manipulation 



(nmoles/min/mg protein) 

(mg/g liver) 



32 ±2 

20.3 ± 4.2 

3.2 ± 1.3 

42.0 ± 7.7 


1-day starved 


24 ± 1 

8.9 ± 0.4 

4.4 ± 0.3 

3.6 ± 3.6 


1-day and 5 hr starved 


27 ±2 

10.7 ± 1.8 

3.4 ± 0.4 

1.9 ± 0.8 


+ (5 hr) 

24 ±4 

9.1 ±0.6 

5.0 ± 0.5 

1.1 ±0.5 

108 ±2 

2-day starved 


21 ±3 

10.6 ± 1.8 

3.2 ± 0.2 

0.2 ±0.1 


+ (lday) 

224 ± 53 

6.2 ± 1.7 

10.2 ± 0.4 

0.2 ±0.1 

103 ±3 

3-day starved 

32 ±6 

2.5 ± 0.6 

3.0 ± 0.2 

0.2 ±0.1 

123 ±3 

+ (2 days) 

261 ±66 

0.4 ± 0.3 

20.1 ± 1.4 

0.2 ±0.1 

102 ±2 

2-day starved and 1- 


27.5 ± 2.2 

3.1 ±0.2 

69.4 ±7.1 


day refed on G 

+ (2 days) 

323 ± 100 

9.0 ± 2.2 

21.8 ± 1.3 

35.8 ± 8.4 

88 ±2 

2-day starved and 1- 


29 ±4 

41.9 ±2.7 

3.9 ± 0.2 

80.0 ± 9.8 


day refed on GC 

+ (2 days) 

229 ± 50 

7.6 ± 2.7 

20.9 ± 0.6 

31.5 ± 16.1 


similar in both groups of rat and almost no 
ingested dietary mass remained in the gas- 
trointestinal tracts at the time of sacrifice. 

Discussion. G6PD is a unique enzyme in a 
sense that a single molecular species is in- 
volved in a wide variety of inductive re- 
sponses; such as those to dietary, hepatotoxic 
and neoplastic changes (17). Thus, tjfie induc- 
tion mechanism of this enzyme appears to be 
different depending on the type of inductive 
stimuU. The dietary induction of G6PD re- 
quires de novo RNA synthesis at a low cycUc 
AMP level (3, 5), whereas carbon tetrachlo- 
ride-induced increase of G6PD synthesis ob- 
Ugates neither of them (7). The latter mech- 
anism would also apply to the increased he- 
patic G6PD level in thioacetamide-injured 
rat (17). Accordingly, the impairment of die- 
tary induction of G6PD in injured hver is 
possibly at the level of transcription. The 
block at this step could be accounted for at 
least by the high hepatic level of cyclic AMP 
observed in the thioacetamide-injured rats 
refed on glucose and casein. The increased 
level of cyclic AMP appears to be also re- 
sponsible for the reduced accumulation of 
hepatic glycogen in the injured refed rats. 
Incidentally, the low hepatic cycUc AMP level 
alone is not sufficient to induce this enzyme, 
since in control rats a sole glucose diet low- 
ered the cyclic AMP level without inductive 

Whether the dietary unresponsive increase 
in cyclic AMP level by thioacetamide treat- 
ment is due to a sustained hyperglucagone- 
mia or an altered adenylate cyclase-phospho- 
diesterase system is to be solved in future 
studies. Although an increased portal level of 
glucagon is reported in acute ethionine intox- 
ication of rat, glucose infusion has been 
shown to decrease the hepatic cyclic AMP 
content (16). Prostaglandin may well be an 
attractive candidate for such a stimulant as 
to the dietary insensitive elevation of cyclic 
AMP in injured liver. 

A reduced dietary response of G6PD in 
regenerating liver following partial hepatec- 
tomy (18) could be similarly explained by 
elevated cyclic AMP levels in the remnant 
liver (10). Since, however, the thioacetamide- 
induced hepatic degeneration and necrosis is 
also followed by a rise in DNA synthesis (7), 
some conditions associated with cell division 
may serve as another common underlying 
mechanism for the suppression of dietary 
induction of G6PD. The small increases in 
hepatic cycUc GMP content found in the late 
stage of thioacetamide injury might be more 
or less related to the regenerative process of 
the injured liver (7, 19), although a direct 
effect of the carcinogen can not be excluded 

It is of some interest to note that another 
dietary inducible and cyclic AMP-sensitive 

o6pd and cycuc amp in liver injury 


ne, glucokinase (1, 13), was also shown 
\ less responsive to glucose-containing 
in the injured liver. Since G6PD and 
Cm hexokinase could be induced by he- 
injury itself (5, 6, 9, 14), the decrease in 
(kinase activity may also represent a spe- 
metabolic response of hepatocyte to the 
/ rather than a mere destructive process 
neral protein synthesis. Thus, in hepatic 
/, the induction of more differentiated 
enzymes is suppressed and that of prim- 
or fundamental enzymes is enhanced, 
ting in an undifferentiated enzyme pat- 
6, 14). A similar loss of dietary response 
SPD and other carbohydrate-metaboliz- 
nzymes in preneoplastic livers has been 
»nstrated by Poirier and others (21). An 
jd inducibility of some enzymes of 
o acid metabolism in chronic adminis- 
in of carbon tetrachloride and a noncar- 
;enic azo dye is also reported from their 
atory (22). Although thioacetamide is a 
tocarcinogen, its acute intoxication, as 
oyed in the present experiment, could be 
3reted better as a hepatic injury, which 

little significance as precancerous le- 
Elucidation of the mechanisms of al- 

enzyme induction in acute hepatic in- 
vould provide a clue for the understand- 
)f undifferentiated gene expression in 
K)plastic livers and in turn hepatomas. 
mmary. The dietary induction of liver 
3 was found to be markedly impaired in 
icute hepatic injury of rat caused by 
cetamide intoxication. The level of 
: AMP in the injured liver was increased 
ould not be reduced by glucose-contain- 
iets. The results indicated that the sup- 
ed dietary inducibility of G6PD in he- 
injury is accounted for at least by the 
ry unresponsive increase in cyclic AMP 
in the injured liver. 

;ot, H. C, Peraino, C, Pries, N., and Kennan, A. 
Adv. Enz. Rcgul. 2, 237 (1964). 

2. Taketa, K., Kaneshige, Y., Tanaka, A., and Kosaka, 
K., Biochem. Med. 4, S31 (1970). 

3. Taketa, K., Kaneshige, Y., and Kosaka, K., Enzyme 

4. Watanabe, A., Takesue, A., and Taketa, K., Enzyme 

5. Watanabe, A., and Taketa, K., J. Biochem. (Tokyo) 

6. Taketa, K., Watanabe, A., and Kosaka, K., in 
**Onco-Developmental Gene Expression** (W. H. 
Fishman and S. Sell, eds.), p. 219. Academic Press, 
New York (1976). 

7. Watanabe, A., Miyazaki, M., and Taketa, K., Gann 

8. Watanabe, A., Takesue, A., and Taketa, K., Enzyme 

9. Taketa, K., Tanaka, A., Watanabe, A., Takesue, A., 
Aoe, H., and Kosaka, K., Proc. Symp. Chem. Phys- 
iol. Pathol. II, 30 (1971). 

10. MacManus, J. P., Franks, D. J., Youdale, T., and 
Braceland, B. M., Biochem. Biophys. Res. Commun. 

11. Kimura, H., Thomas, E., and Murad, F., Biochim. 
Biophys. Acta 343, 519 (1974). 

12. Steiner, A. L., Pagliara, A. S., Chase, L. R., and 
Kipnis, D. M., J. Biol. Chem. 247, 1 1 14 (1972). 

13. Honma, M., Satoh, T., Takezawa, J., and Ui, M., 
Biochem. Med. 18, 257 (1977). 

14. Taketa, K., Tanaka, A., Watanabe, A, Takesue, A., 
Aoe, H., and Kosaka, K., Enzyme 21, 158 (1976). 

15. Ureta, T., Radojkovic, J., and Niemeyer, H., J. Biol. 
Chem. 245,4819(1970). 

16. DeRubertis, F. R., and Craven, P., Meubolism 25, 

17. Watanabe, A., and Taketa, K., Arch. Biochem. Bio- 
phys. 158,43 (1973). 

18. Potter, v. P., and Ono, T., Cold Spring Harbor 
Symp. Quant. Biol. 2d, 355 (1%1). 

19. Miura, Y., Iwai, H., Sakata, R., Ohtsuka, H., El- 
hanan, E., Kubota, K., and Fukui. N., J. Biochem. 
(Tokyo) 80, 291 (1976). 

20. Vesely, D. L., and Levey, G. S., Proc. Soc. Exp. Biol. 
Med. 155,301(1977). 

2 1 . Poirier, L. A., Poirier, M. C, and Pilot, H. C, Cancer 
Res. 29,470(1969). 

22. Poirier, L. A., and Pilot, H. C, Cancer Res. 29, 475 

Received March 31, 1978. P.S.E.B.M. 1978, Vol. 159. 


Blood Volume Changes during the First Week after Birth in the Beagle and Pig (40303) 


Department of Physiology, Baylor College of Medicine, Houston, Texas 77030 

Birth marks the end of the parasitic and 
aquatic life of the fetus and the beginning of 
numerous physiological adjustments which 
adapt the newborn to a new and different 
environment. Among the adjustments which 
occur in different species at birth are those 
involving the circulatory system, and within 
this system are the changes in red cell and 
plasma volumes and venous hematocrit. 
However, data available for the newborn hu- 
man over the first few days following birth 
present no clear pattern of change in the 
plasma volume, red cell volume, blood vol- 
ume, or hematocrit. A portion of the varia- 
bility in these data may be the resuh of early 
or late clamping of the umbilical cord (1, 2). 
But even if analysis of the data is restricted to 
those investigations where the cord is 
clamped early, the results are contradictory. 
Plasma volume, for example, is reported to 
remain constant over the first 24 hr following 
birth (3), to increase significantly within 3-5 
hr (4, 5), to decrease in the first 2Yi hr (6), or 
to increase over the period of 4-24 hr after 
birth (1). Changes reported for red cell vol- 
ume, blood volume, and hematocrit are 
equally varied, although in most of these 
investigations blood volume and red cell vol- 
ume are calculated from the measured plasma 
volume and hematocrit. 

There are relatively few studies examining 
the changes in blood volume immediately 
after birth for species other than the human. 
In the pig, McCance and Widdowson (7) 
report a 30% increase in plasma volume 24 
hr following birth, while Ramirez et at. (8) 
report a small but significant increase in 
blood volume during the first 12-hr period 
following birth. In the rat there is a small 
reduction in plasma volume between days 4 
and 14 (9), but, contrary to the results of these 
authors, in the same species Garcia (10) re- 
ports a rise in plasma volume from birth to 
13 days of age, and Constable, no significant 
change (11). 

In the present article, data are presented 
for both the pig and the beagle for the period 
between birth and day 7 following birth, and 
the changes in plasma, red cell, and blood 
volumes and hematocrits are examined. 

Materials and methods. The beagles used in 
this investigation were from the colony main- 
tained at the Wynne Unit of The Texas De- 
partment of Correction in Huntsville. A de- 
scription of the physical facility and the rou- 
tine procedures used for breeding, immuni- 
zation, and diet was published previously 
(12). A pig colony for research purposes was 
established while one of us (R.H.) was serving 
as acting chairman of the Department of 
Physiology at Mahidol University in Bang- 
kok, Thailand, and was maintained at Kaset- 
sart University by the courtesy of university 
officials and The Rockefeller Foundation. 
Details of the management of this colony also 
have been published (13). Standard proce- 
dures, modified for small animals, were used 
to measure red cell volume with ^^Cr (14) and 
plasma volume either with *^^I-albumin or 
the dye T-1824. There was no statistically 
significant difference between the plasma vol- 
umes measured with ^^^I-albumin or T-1824 
(12, 15). Hematocrits were measured by the 
micro method; no correction was made for 
trapped plasma. Beagle pups up to 3 hr after 
birth were not sedated, while those older than 
4 hr were given 0.5-1 mg of morphine sulfate, 
injected subcutaneously. The pigs were anes- 
thetized with 5-10 mg/kg of pentobarbital 
sodium, administered intravenously. Differ- 
ent animals, and usually from the same litter, 
were used for the collection of data for each 
of the time periods after birth. 

Results. The data for the beagles arc pre- 
sented in Table I. For day (day of birth) 
data were available from 10 min to 18 hr 
following birth, and because there was evi- 
dence of significant changes within this time, 
the data were divided into three 6-hr periods. 
The average age of the pups was 2.5 hr for 


0037-9727/78/ 1 59 1-01 52$01 .00/0 

"opyright 1978 by the Society for Experimenul Biology and Medicine 



TABLE 1. Red Cell and Plasma Volume Changes in Newborn Beagles. 





Blood vol- 

Venous he- 



Body weight kg 



ume mlAg 

matocrit % 


BVR cells* 

0.190 ±0.02*' 

49.2 ± 2.6 

46.4 ± 1.5 

95.6 ± 4.9 

56.0 ± 1.9 

51.0 ± L9 

0.906 ± 0.02 










0.244 ± 0.005 

38.7 ± 2.3 

45.3 ± 0.3 

84.0 ± 2.6 

52.3 ± 2.6 

45.9 ±L3 

0.881 ±0.02 


(4) ^ 







0.266 ± 0.01 

40.5 ± 1.0 

53.3 ± 1.3 

93.6 ± 2.0 

47.5 ± L5 

43.2 ± 0.4 

0.912 ± 0.03 



P< 0.001 

P < 0.05 





0.275 ±0.01 

45.4 ± 1.9 

48.2 ± 1.1 

91.3 ± 2.5 

53.5 ± 1.4 

47.1 ± 1.2 

0.899 ± 0.01 










0.251 ±0.01 

56.7 ± 4.5 

62.2 ±3.1 

118.9 ±2.5 

52.1 ±2.7 

47.3 ± 2.9 

0.892 ± 0.01 



P < 0.02 

?< 0.001 

?< 0.001 





0.273 ± 0.006 

46.2 ± 4.0 

59.1 ±2.2 

105.3 ± 4.6 

46.4 ± 2.6 

43.3 ± 2.3 

0.933 ± 0.02 



P < 0.05 


P < 0.05 





0.436 ± 0.21 

35.6 ± 2.2 

62.2 ± 2.6 

97.8 ±3.1 

38.8 ± 1.5 

36.4 ± 2.2 

0.930 ± 0.01 


?< 0.001 



i>< 0.001 



cell volumc/(rcd cell volume + plasma volume). 

ulatory hematocrit/venous hematocrit. 




iber of animals. 

lue for difference from 

previous value. 

it 6-hr period, 8.5 hr for the second 
and 16.5 hr for the last 6-hr period, 
mean red cell volume for beagles 2.5 
was 49.2 ± 2.6 ml/kg. In pups 8.5 hr 
: red cell volume was significantly less 
.05), and did not change again during 
It 6-hr period. The plasma volume of 
.5 hr old was 46.4 ± 1.5 ml/kg, with 
nge during the next 6 hr, however, it 
;nificantly higher (P < 0.(X)1) in pups 
r old. Blood volume was 95.6 ± 4.9 
in pups 2.5 hr after birth and decreased 
s 8.5 hr old due to the decrease in red 
lume. The blood volume, as the result 
igniflcant increase in plasma volume 
n 8.5 and 16.5 hr, was only sUghtly 
16.5 hr than at birth. The venous 
xnrit decreased during the successive 6- 
lods, resulting in a significantly (P < 
ower hematocrit in pups 16.5 hr old 
>r those at 2.5 hr. The trend for the 
tory hematocrit was the same as that 
venous hematocrit; consequently, the 
r circulatory to venous hematocrit was 
illy unaltered. 

cell and plasma volumes were signifi- 
higher for the day-1 than for the day- 

pups, using the pooled data for the 20 
beagles on day 0. This increase in plasma and 
red cell volumes resulted in a significant in- 
crease in blood volume (P < 0.(X)1) for the 
day-1 pups. On day 2 there was a significant 
decrease in red cell volume and blood volume 
(P < 0.05), but only a sUght reduction in 
plasma volume. On day 7 there was a further 
significant decrease (P < 0.00 1) in red cell 
volume, while the decrease in blood volume 
was not significant due to an increase, al- 
though not significant, in plasma volume. 

Changes in venous and circulatory hema- 
tocrits reflected those of cell and plasma vol- 
umes throughout the period of study. Since 
the increases in red cell and plasma volumes 
between day and day 1 were of the same 
magnitude (20-23%), neither circulatory nor 
venous hematocrit changed signiflcantly, and 
the ratios of the two hematocrits (BVRceiia) 
remained the same. From day 1 to day 2 both 
venous and circulatory hematocrits de- 
creased, but not significantly; however, be- 
tween days 2 and 7 there was a further sig- 
nificant decrease in venous hematocrit ac- 
companied by a similar change in circulatory 
hematocrit. The BVRceiia remained relatively 



constant over the first 7 days after birth (0.899 
± 0.01 on day and 0.93 ± 0.01 on day 7), 
indicating that there was no shift in the dis- 
tribution of red cells and plasma in the cir- 
culation during the 7-day period. 

For the pig (Table II), the exact times at 
which red cell and plasma volumes were mea- 
sured on day were not known; therefore, 
only the mean value was calculated. The 
principal changes observed in the pig were an 
increase in plasma volume (P < 0.05) be- 
tween days 1 and 2 and a decrease in red cell 
volume between days and 2 and days 2 and 
7, with the decrease on day 7 significant when 
compared with day (P < 0.05). Blood vol- 
ume decreased progressively, and on day 7 it 
was significantly less than that measured on 
day (P < 0.05). Venous hematocrit de- 
creased between days and I (P < 0.05), 
with no further signiflcant change on day 2 
or day 7. The values for BVRceUa were 0.85 
on day and 0.79 on day 7. 

Discussion. During the first few days after 
birth there are changes in both the red cell 
and plasma volumes in the beagle and the 
pig, but the pattern of the changes is different 
for the two species. What the red cell and 
plasma volume changes are in the human 
over the first few days after birth is uncertain 
at present because of the diversity of the data. 
However, data by Usher et aL(\) suggest that 
in the human neonate, as in the pig and 
beagle, there is over the first few days follow- 
ing birth an increase in plasma volume and 
a decrease in red cell volume and venous 

hematocrit, although there are differences in 
the time relationship at which the changes 

In the beagle the increase in plasma vol- 
ume, which may occur as early as 12 hr 
following birth, is accompanied by an in- 
crease in plasma protein concentration. The 
increase in total protein concentration is due 
to an increase in the globuUn fractions, while 
the albumin concentration remains stable, so 
that the albumin-globulin ratio decreased sig- 
nificantly (15). Thus the expansion of plasma 
volume can be explained by a shift of fluid 
into the circulation due to an increase in 
plasma protein. This shift of fluid among 
body compartments is substantiated further 
by the fmding in the beagle that between 
days and 1 there is a significant increase in 
the volume of extracellular fluid, at the ex- 
pense of intracellular fluid, while total body 
water remains constant (16). 

An increase in plasma volume similar to 
that in the beagle occurs in the pig, although 
the increase is between days 1 and 2. Accord- 
ing to McCance and Widdowson (7), who 
first observed an increase in plasma volume 
within hours following birth of the pig the 
increase is the result of absorption of colos- 
trum through the gut with a marked increase 
in the globulin portion of total plasma protein 
concentration. This mechanism is suggested 
also as an explanation of the plasma volume 
expansion in the beagle. 

In the immediate neonatal period, red cell 
volume in the beagle, but not in the pig, is 

TABLE II. Red Cell and Plasma Volume Changes in Newborn Pigs. 



Body weight 

Red cell vol- 

Plasma vol- 

Blood vol- 

Venous he- 


after birth 


ume ml/kg 

ume ml/kg 

ume ml/kg 

matocrit % 


BVR cells* 


1.7 ± O-OS*^ 

24.8 ±2.1 

63.2 ±2.1 

88.0 ± 3.2 

33.1 ± 1.3 

28.1 ± 1.4 

0.85 ± 0.04 








Day 1 

1.6 ±0.05 

58.0 ± 2.5 

27.1 ±2.7 

P < 0.05' 

Day 2 

1.5 ±0.14 

20.2 ± 1.4 

66.7 ± 3.3 

86.9 ± 3.4 

26.6 ± 1.3 

23.0 ± 1.2 

0.86 ± 0.03 



P < 0.05 






2.4 ±0.11 

19.7 ± 1.0 

64.4 ± 1.5 

82.7 ± 1.6 

29.9 ± 0.8 

23.8 ± 1.9 

0.79 ± 0.04 








° Red cell volume/(red cell volume + plasma volume). 

* Circulatory hemalocrit/venous hematocrit. 

' Mean ± SE. 

** Number of animals. 

*" P value for difference from previous value. 



le than the plasma volume, and, 
la volume, the changes are diffi- 
lin. At 2.5 hr after birth the red 
is 49.2 ± 2.6 ml/kg in the beagle, 
^nificantly over the next 6-18 hr, 
iring the next 24 hr, then de- 
^een 48 and 72 hr. The venous 
eflects the changes in red cell and 
imes fairly consistently. For ex- 
day to day 1 there is a signifi- 
e in both red cell and plasma 
d as the percentage increase for 
same, there is no change in the 
atocrit. On day 2 there is a sig- 
(19%) in red cell volume and a 
\ in plasma volume, and these 
accompanied by an 11% fall in 
atocrit. Inasmuch as the venous 
[S affected by shifts in both red 
sma volumes, it cannot be used 
mate changes in either red cell or 
me. Shifts in the circulatory he- 
^hole body hematocrit) follow 
; venous hematocrit, so that the 
two hematocrits remains approx- 
same. Therefore, for the beagle 
ay period after birth, it is possible 
either red cell or plasma volume 
>y use of this ratio (BVRceiis) and 
hematocrit, estimate the other 
ible accuracy. 

d volume for newborn mongrel 
sr, 135 ml/kg (17), than the 95.6 
ured for the beagle. This differ- 
data for the newborn dog results 
)m two factors. The first factor is 
il (17) measured plasma volume 
ed red cell volume from the ve- 
ocrit. The latter represents the 
hematocrit and therefore overes- 
red cell and blood volumes. The 
)r may be even more important 
it in the estimation of blood vol- 
;ral species including the dog, the 
ce of both *^^I-tagged albumin 
dye from the circulation is more 
newborn than in the aduh (15). 
ly, with either of these tags, when 
volume is calculated from a single 
wvas done in the newborn mon- 
iter the time of samphng after 
the tag on day 0, the greater will 
in the plasma volume measure- 

ment. For instance, in pups on day a sample 
taken 15 min after injection of the tag may 
overestimate plasma volume by 15%, result- 
ing also in an overestimation of red cell and 
blood volumes. 

The fluctuations in red cell volume in the 
beagle during the first week after birth pose 
several questions; one of them is the possible 
sites from which the red cells can be seques- 
tered or released. The volume of red cells 
shifting into and out of the circulation is 
relatively large: between 2.5 and 8.5 hr after 
birth the circulating red cell volume decreases 
by 22%, while between 18 and 48 hr after 
birth red cell volume increases by 28%. 
Changes in red cell volume of a similar mag- 
nitude for the human over the first 5-hr pe- 
riod after birth are reported by Sisson and 
Whalen (18). The difference between their 
results (17) and those reported for other new- 
bom humans (6) may be explainable on the 
basis of the time of cord clamping during 
birth. This explanation, however, does not 
appear to be applicable to the changes seen 
in the beagle. Sisson and Whalen (18) also 
postulated, as an explanation for the changes 
in red cell volume in the newborn human, 
"an initial temporary sequestration of blood 
in the viscera and caudal end of the body," 
and the blood was later "introduced into the 
general circulation as vascular and pulmo- 
nary patterns were stabilized." 

The spleen and bone marrow are suggested 
also as blood reservoirs capable of signifi- 
cantly increasing blood volume in the human 
during the first 24 hr after birth (19). In the 
aduh dog both the spleen and the liver are 
known to be active red cell reservoirs (20), 
but whether this is true also in the newborn 
pup can be inferred only from indirect data. 
The unit red cell volume (ml/ 1 00 g) of all 
organs in the beagle decreases between days 
and 1 (21, 22). The combined red cell 
volume of the heart, lungs, kidneys, spleen, 
stomach, skeletal muscle, intestines, and skin 
is 28% of the total red cell volume on day 0, 
but decreases to 16% on day I. The decrease 
in the volume of red cells in these tissues 
coincides with an increase in the circulating 
red cell volume on day 1. These data, while 
providing no information on the mechanisms 
concerned in the relatively rapid fluctuations 
in red cell volume of the newborn beagle, do 



provide tentative support to the idea that 
there may be reservoirs of red cells in the 
circulation of the newborn and that red cells 
move in and out of these reservoirs under the 
control of unknown stimuli. 

Summary, During the first week of post- 
natal life, there were significant changes in 
red cell volume, plasma volume, and venous 
and circulatory hematocrits in both the beagle 
and the pig. In beagle pups the mean red cell 
volume decreased between 2.3 and 8.3 hr 
after birth, then increased at 16.3 hr, with a 
further increase on day 1. Between days 2 
and 7, red cell volume decreased. There was 
evidence of a release of red cells into the 
circulation from red cell reservoirs. In the 
newborn pig, red cell volume decreased be- 
tween days and 2, but was not significantly 
different on day 7 from day 2. In the beagle 
the mean plasma volume did not change 
during the first 12 hr following birth; it in- 
crease between 12 and 24 hr after birth and 
remained unchanged through day 7. In the 
pig, plasma volume decreased between day 
and day 1, increased on day 2, and was not 
significantly different on day 7 from day 2. 
The increase in plasma volume was the result 
of an increase in plasma protein, which 
caused a redistribution of fluid among the 
various fluid compartments. In the beagle, 
blood volume decreased between 2.3 hr and 
8.3 hr, increased at 16.3 hr with a further 
increase on day 1, then decreased on day 2, 
with no further change on day 7. The blood 
volume in the pig decreased progressively 
between day and day 7. The changes in 
venous and circulatory hematocrits for both 
the beagle and pig reflected those of red cell 
and plasma volumes throughout the first 
week of life. The BVRceiis did not change 
significantly, indicating that there was no 
sluft in the distribution of red cells and 

plasma in the circulation over this 7-day pe- 

1. Usher, R., Shephard, M., and Lind, J., Acta Paedial 

2. Whipple, G. A., Sisson, T. R. C, and Lund, C. J., 
Obslel. Gynecol. 10, 603 (1957). 

3. Mollison, P. L., Veall, N., and Cutbush, M., Arch. 
Dis. Child. 25, 242 (1950). 

4. Gairdner, D., Marks, J., Roscoe, J. D., and BretteU, 
R. O., Arch. Dis. Child. 33, 489 (1950). 

5. Sisson, T. R. C, Lund, C. J., Whalen. L. E., and 
Telek. A., J. Pediat. 55, 163 (1959). 

6. Steele, M. W., Am. J. Dis. Child. 103, 42 (1962). 

7. McCance, R. A., and Widdowson, E. M., J. PhysioL 

8. Ramirez, C. G., Miller, E. R., UUrey, E. D., and 
Hoefer, J. A., J. Anim. Sci. 22, 1068 (1963). 

9. Travnickova, E., and Heller, J., Physiol. Bohemos- 

10. Garcia, J., Amer. J. Physiol. 190, 19 (1957). 

11. Constable, B. J., J. Physiol. ItfT, 229 (1963). 

12. Huggins, R. A., Deavers, S., and Smith, E. L., Pediat. 
Res. 5, 193 (1971). 

13. Setiabudi, M., Sheng, H. P., and Huggins, R. A., 
Growth 40, 127(1976). 

14. Huggins, R. A., Smith, E. L., Deavers, S., and Over- 
ton, R. C, Amer. J. Physiol. 189, 249 (1957). 

15. Deavers, S., Huggins, R. A., and Smith, E. L., Amer. 
J. Vet. Res. 32, 1169 (1971). 

16. Sheng, H. P., and Huggins, R. A., Proc. Soc. Exp. 
Biol. Med. 139,330(1972). 

17. Lee, P., Brown, M. E., and Hutzler, P. T., Amer. J. 
Vet. Res. 37, 561 (1976). 

18. Sisson, T. R. C, and Whalen, L. E., J. Pediat. 56,43 

19. Low, J. A., Kerr, N. D., and Cochon, A. R., Amer. 
J. Obstet. Gynecol. 86, 886 (1963). 

20. Kraintz, L., DeBoer, J., Smith, E. L., and Huggins. 
R. A., Amer. J. Physiol. 195, 628 (1958). 

21. Deavers, S., Huggins, R. A., and Smith, E. L. 
Growth 36, 195 (1972). 

22. Smith, E. L., Deavers, S., and Huggins, R. A., Proc. 
Soc. Exp. Biol. Med. 140, 285 (1972). 

Received April 17, 1978. P.S.E.B.M. 1978, Vol. 159. 


ime of Exposure to Estradiol and LHRH Effect LH Release From Bovine Pituitary 

Cells?' (40304) 


mal Reproduction Laboratory, Department of Dairy Science, Michigan State University, East Lansing, 

Michigan 48824 

nizing hormone releasing hormone 
) induced increase in serum LH is 

coincident with periods of increased 
1 secretion in cows (1), ewes (2), 

(3, 4) and female rats (5, 6). Exoge- 
Togens also increase magnitude of LH 

by LHRH in cows (7), ewes (8), 

(9) and female rats (10). However, 
s to demonstrate direct effects of es- 
on LH secretion in vitro have yielded 
; results, i.e. estradiol increased (11, 
reased (13-15) or did not change (16, 
Qtity of LHRH induced LH release in 

:periments reported, we investigated 
)f estradiol and LHRH on LH secre- 
bovine pituitary cells in primary cul- 
iriables were dose and time of expo- 
cells to E2 and LHRH alone or in 

rials. Medium for culture was Dul- 
minimal essential medium^ supple- 
with essential and non-essential 
icids and buffered as in reference 12. 
olutions of synthetic LHRtf, pre- 
1 0.1% knox gelatin:0.05 M phosphate 
1 saline, were added to cultures in 10 
Estradiol- 17j8 (E2)^ in 10% ethanol, 
led in volumes such that final concen- 
Df ethanol in medium was 0.1%. 
mltures. Bovine pituitary cell cultures 
epared (12). Briefly, bovine anterior 
ies were sliced (si MM), diced (si 
nd resulting pieces washed thrice with 

shed with approval of the Michigan Agricul- 
^riment Station as journal article no. 8508. This 
vas supported in part by NIH Grant No. HD- 

tiom all correspondence should be sent. 
ecco*s medium from Difco Labs, Detroit, MI; 
lurtesy of Dr. R. Rippel, Abbott, N. Chicago, 
diol np and collagenase (type 1-150 /ii/mg) 
la, Chicago, ill.; Violcase from GIBCO, Grand 
5w Yorlc. 

medium. Pituitary cells were dispersed from 
these pieces by stirring in 0.3% collagenase^ 
for 45 min then 0.25% Viokase^ for 15 min. 
Washed cells were suspended (s5 x 10* 
cells/ml) in medium containing 10% bovine 
serum^ and 1 ml of suspension transferred to 
each well of multiwell culture plates. Pituitary 
cells were in culture for 5 days with medium 
changed at 24-hr intervals beginning at 48 hr. 
On day 5 cells were washed 4 times with 
serum free medium and treatments begun. 
Medium did not contain serum during treat- 

Experimental design. Experiment J. The ob- 
jective was to determine effects of varying 
time of exposure and concentration of LHRH 
on quantity of LH released. Treatments were 
arranged as a five x six factorial experiment 
with concentration of LHRH (0, 0.1, 1.0, 10 
and 100 ng/ml) and time (.75, 1.5, 3, 6, 12 
and 24 hr) as main effects. There were six 
replicates per treatment combination (n — 

Experiment 2. The objective was to deter- 
mine effects of varying time of exposure and 
concentration of estradiol on quantity of LH 
released. Treatments were arranged as a three 
X four factorial experiment with concentra- 
tion of estradiol (0, 5 and 50 ng/ml) and time 
(3, 6, 12 and 24 hr) as main effects. There 
were 12 replicates per treatment combination 
(/I = 144). 

Experiment 3. The objective was to exam- 
ine the interaction of estradiol and LHRH on 
LH release over time. Treatments were ar- 
ranged as a four x two x five factorial ex- 
periment with concentrations of estradiol (0, 
0.5, 5.0, and 50 ng/ml) and LHRH (0 and 
100 ng/ml) and time of exposure to estradiol 
and LHRH (1.5, 3, 6, 12, and 24 hr) as main 
effects. There were four replicates per treat- 
ment combination (/i = 160). 

Within each experiment, treatments were 
begun concurrently and medium collected 


0037-9121 /l%/\S9\4i\SlV^\.^l^ .>^^v6«. 



and frozen after the prescribed interval of 
treatment. Medium was assayed for LH by 
methods described in 18. 

Statistical analysis of data. In instances 
where data, hormone concentrations or time 
were not normally distributed, statistical 
analysis were performed after logarithmic 
transformation of values. Data from each 
experiment were analyzed by analysis of var- 
iance appropriate to factorial experiments 
(19). Significant differences due to main ef- 
fects were determined by Dunnett's t-test 
(20). Additionally, data were subjected to 
polynomial regression analysis (19) to evalu- 
ate change in LH release over time or con- 
centration of hormones tested. 

Results. Experiment 1. Effects of varying 
time of exposure and concentration of LHRH 
on LH concentration in medium are in Fig. 
1. In the absence of LHRH, LH accumulated 
in medium during 24 hr and this increase was 
curvilinear (P < 0.001) i.e. rate of accumu- 
lation increased with time. Within time pe- 
riods, increase in LH release by LHRH over 
the range 0.1 to 100 ng/ml, was linear (P < 
0.001) when exposure was for .75, 1.5, 3, 6, or 
24 hr but curvilinear (P < 0.001) when for 12 
hr. Dose-response slopes generated from data 
normalized by logarithmic transformation 
were not different among times i.e. with the 
exception of 12 hr, LH release induced by 
100 ng LHRH/ml was twice that of compa- 
rable control values. However, the actual in- 
crease in amount of LH release (ng/ml) over 
controls, induced by each concentration of 
LHRH, increased with increasing time of 


Experiment 2. Effects of varying time of 
exposure and concentration of estradiol on 
quantity of LH in medium are in Tabic 1. 
Estradiol did not affect concentration of LH 
in medium when present for 3 hr but in- 
creased (P < 0.001) LH relative to controls 
when present for 6, 12 or 24 hr. Both concen- 
trations of estradiol tested increased LH ac- 
cumulation in medium and magnitude of 
increase was dependent on the dose of E2 i.e. 
50 ng E2 released more LH than 5 ng (? < 

Experiment 3, Effects of varying time of 
exposure and concentration of estradiol on 
LHRH induced LH release are in Fig. 2. 
Within each combination of LHRH and es- 
tradiol, accumulation of LH in medium was 
curvilinear (P < 0.001) and greater (P < 
0.001) for cuhures incubated with LHRH 
than for comparable controls. Estradiol, pres- 
ent for 1.5 or 3 hr, did not affect LH concen- 
tration in medium of cultures incubated with 
or without LHRH. However, when estradiol 
was present 6, 12 or 24 hr LH accumulation 
in medium was increased (P< 0.001) relative 
to controls. This was true for cultures incu- 
bated with or without LHRH. In addition, 
magnitude of LH release, within these time 
periods, was linearly (P < 0.001) related to 
concentration of estradiol used. A compari- 
son of cultures incubated with and without 
LHRH, within time, revealed that slopes of 
estradiol dose-response were not different (P 

Discussion. Results confirm our previous 






liO' a 1.0 
o lo.a 

I to- ■ lOO.O 



nnaiii nilll »lil 

Tim* of Incubotion with LHRH(hr.) 
Fig. 1. LH concentration in medium following incubation of bovine pituitary cells with 0, 0.1 
LHRH/ml media for .75, 1.5, 3, 6, 12 or 24 hr. Values are means ± SE. 

1, 10 or 100 ng 



tion that LHRH causes LH release 

»vine pituitary cells in culture and that 

^ of LH released is linearly related to 

ration of LHRH over the range 0. 1 to 

ml (12). This result was demonstrable 

me of exposure to LHRH was as short 

dn or as long as 24 hr. Additionally, 

increase in LH release relative to 

» induced by each dose of LHRH was 

idem of time LHRH was present. This 

• the conclusion that ability of LHRH 

;:e LH release appears to be consistent 

east 24 hr. Resolution of effects of 

on increasing LH concentration, as 

ned by difference in LH concentra- 

control cultures and those given 

increased markedly with time. For 

ison, time of exposure to LHRH 

than 3 hr may be desirable. 

nt results confirm our previous obser- 

hat estradiol when present for 24 hr 

d basal and LHRH mduced LH 

from bovine pituitary cells (12). These 

lents provide evidence that estradiol 

5 present for more than 3 hr before 

.ory effects on LH release are demon- 

Our failure to demonstrate an effect 

diol at .75 or 3 hr agrees with results 

rs using rat pituitary cells in culture 

ds lag period may represent time re- 

for estradiol to exert biological 

in gonadotrophs that result in in- 

LH release. Inhibitors of protein syn- 

»lock the stimulatory effect of low 

'estradiol on LH release (17). Failure 

t an effect of estradiol during the first 

treatment may reflect time required 

ein synthesis. Alternatively, tlus lag 

an artifact of the culture system. 

I. Effect of Estradiol- 17)8 and Time of 
RE to Estradiol on Medium Luteinizing 
Hormone Levels. 




8.6 ± .7- 9.2 ± .6- 

4.3 ± .9- 21.3 ± .9* 
2.5 ±3.7- 32.3 ±3.6* 

4.4 ±2.4- 80.7 ±2.7* 



9.2 ± .5- 9.0 

23.5 ± .9* 16.4 

41.7 ±2.8" 32.2 

100.8 ± 4" 82.0 


s within time periods with different super- 
significantly different at P < 0.05. Values are 
;E (II - 12). 



• as 

A 0.0 
■ 90.0 

TliM of Ineubotlondir.) 

Fig. 2. LH concentration in medium of bovine pi- 
tuitary cultures incubated with 0, .5, 5 or 50 ng/ml E2 
for 1.5, 3. 6. 12, or 24 hr with LHRH (0 or 100 ng/ml) 
present through out the incubation period. Dashed lines 
represent data obtained when LHRH was present. 

Considering that rate of accumulation of LH 
in medium accelerated during the 24 hr ex- 
perimental period, failure to detect an effect 
of LH release at 45 min and 3 hr may be 
because LH release at this time is very low 
and gonadotrophs not receptive to this stim- 
ulus. An argument against the latter view is 
that LHRH was equally efficacious in causing 
LH release at all times tested. Our results also 
demonstrate that once estradiol aflects LH 
release, this eflect remains quantitatively sim- 
ilar at least to 24 hr in cultures incubated 
with and without LHRH. LH release by rat 
pituitary cells was increased by 500 ng/ml 
estradiol for 6 or 24 hr (15) or 0.27 ng/ml for 

Results of experiments designed to inves- 
tigate in vivo effects of estradiol on LHRH 
induced LH release revealed a biphasic effect 
i.e. estradiol first decreased, then increased 
magnitude of LHRH induced increase in se- 
rum (21-23). In these in vitro experiments, 
estradiol did not inhibit basal or LHRH in- 
duced LH release suggesting the initial inhib- 
itory effect in vivo is not mediated via a direct 
effect on the pituitary. 

Summary, Time course of 17-^8 estradiol 
and luteinizing hormone-releasing hormone 
effect on LH release was studied using bovine 
pituitary cells on day 5 of culture. LHRH at 
concentrations of .1, 1, 10 and 100 ng/ml 



increased LH in medium linearly with in- 
creasing log concentration of LHRH when 
present for .75, 1.5, 3, 6 and 24 hr and the 
percent increase over controls was same at 
each time period. In addition, estradiol (pres- 
ent for 6, 12 or 24 hr) at .5, 5, and 50 ng/ml 
also increased LH release linearly both in the 
presence or absence of LHRH. We conclude 
that the stimulatory effect of LHRH on LH 
release from bovine pituitary is consistant 
over 24 hr and the stimulatory effect of E2 on 
both basal and LHRH induced LH release 
may be mediaed at least in part directly on 
the pituitary. 

The authors acknowledge Dr. R. R. Neitzel and L. T. 
Chapin for valuable assistance with computer program- 
ming and P. Harkins and C. Wallace for technical help. 

i. Zolman, J., Convey, E. M., and Britt, J. H., J. Anim. 
Sci. 39, 355 (1974). 

2. Reeves, J. J., Arimura, A., and Schally. A. V.. J. 
Anim. Sci. 32, 123(1971). 

3. Yen, S. S. C, Vandenburg, C, Rebar, R., and Ehara, 
Y.. J. Clin. Endocrinol. Metabol. 35, 931 (1971). 

4. Thomas, K., Cardon, M., Donnez, J., and Ferin, J., 
Contraception 7, 289 (1973). 

5. Gordon, J. H., and Reichlin, S., Endocrinology 94, 

6. Zeballos, C, and McCann, S. M., Endocrinology 96, 

7. Hausler, C. L., and Malven, P. V., J. Anim. Sci. 42, 

8. Coppings, R. J., and Malven, P. V., Proc. Soc. Exp. 

Biol. Med. 148,64(1975). 
9. Yen, S. S. C, Vandcnburg, G., Siler, T. M.. J. Clin. 
Endocrinol. Metabol. 39, 170 (1974). 

10. Cooper, K. J., Fawcett, C. P., and McCann, S. M, 
Proc. Soc. Exp. Biol. Med. 145, 1422 (1974). 

1 1. Drouin, J., Lagaoe, L., and Labrie, F., Endocrinology 

12. Padmanabhan, V., Kesner, J. S., and Convey, E. M., 
Biol. Rep. 18,608(1978). 

13. Schally, A. V., Redding, T. W.. and Arimunu A., 
Endocrinology 93, 893 (1973). 

14. Tang, L. K. L., and Spies, H. G., Endocrinology %, 

15. Steinberger, A., and Chowdhury, M., Endocrine Res. 
Commun. 1,389(1974). 

16. Piacsek, B. E., and Meites, J., Endocrinology 79,432 

17. Schneider, H. P. G., and McCann, S. M., Endocri- 
nology 84, 330 (1970). 

18. Convey, E. M., Beal, W. E., Seguin, B. E., Tannen, 
K. J., and Lin, Y. C, Proc. Soc. Exp. Biol. Med. 151, 

19. Steel, R. G., and Torrie, J. H., "Principles and 
Procedures of Statistics.'* McGraw-Hill, New Yofk 

20. Kirk, R. E., in "Experimental Design: Procedures 
for the Behavioral Sciences** Wadsworth, Behnont, 
CA 94 (1968). 

21. Vilchez-Martinez, J. A., Arimura, A., Debeljuk, L., 
SchaUy. A. V., Endocrinology 94, 1300 (1974). 

22. Libertun, C, Cooper, K. J., Fawcett, C. P., and 
McCann, S. M.. Endocrinology 94, 1518 (1974). 

23. Keye, W. R., and Jaffe, R. B., J. Clin. Endocrinol 
Metabol. 41, 1003(1975). 

Received April 27, 1978. P.S.E.B.M. 1978, VoL 159. 


Df Administration of a LH-RH Inhibitory Analogue on Stages of the Rat Estrous 

Cycle' ^ (40305) 


of Medicine, Tuiane University School of Medicine, and Endocrine and Polypeptide Laboratory, Veterans 
Administration Hospital, New Orleans, 70146 

been demonstrated that synthetic 
f analogues of LH-RH can block 
ttory surges of gonadotropins and 
in hamsters (1) and rats (2, 3). The 
sous administration of 750 /xg of 
)-Leu®]-LH-RH four times on proes- 
moon in hamsters, produced an 80% 
Dn of the LH surge and a 30% block- 
ulation (1). In rats, 6 mg of [D-Phe^ 
.H-RH, injected in several doses on 
loon ojf proestrus, brought about a 
bition of ovulation (2) whereas a 
se of 1.5 mg of [D-Phe^ Phe^ d- 
-RH at noon on proestrus strongly 
LH and FSH surges and suppressed 
by 85% (3). Recently, Beattie et al 
ted that [D-Phe^ D-Ala^]-LH-RH 
itly inhibited ovulation when it was 
on days other than estrus in rats. 
vivo assays, such as inhibition of 
nduced LH release in inmiature 
, and blockade of ovulation in nor- 
[D.Phe^ Phe^ D-Phe^-LH-RH is 
ent and longer acting than [D-Phe^, 
.H-RH (3, 5), which in turn is more 
an [D-Phe^ D-Ala^J-LH-RH (5). We 
'efore investigated the effects of [d- 
5^ D-Phe®]-LH-RH on ovulation in 
1 injected at different stages of the 
^cle or daily during estrus (E), dies- 
1) and diestrus 2 (D2). 
als and methods. Adult female rats 
River CD strain), weighing 200-250 
laintained under conditions of cou- 
nting (14 hr Ught and 10 hr dark- 
temperature (22°). Following a one 

led in pan by NIH Contract Nos. HD-72- 
lD-6-2841; USPHS Grant Nos. AM-07467 
$55; and by the Veterans Administration, 
ed in part at the 6(Hh Annual Meeting of 
jsiract #2722, Anaheim, CA, April 11-16, 

week period of adjustment to the animal 
house, their estrous cycles were determined 
by inspection of daily vaginal smears. Only 
those animals presenting at least two succes- 
sive, regular 4-<iay cycles were used. 

In the first experiment, the animals were 
injected s.c. with a single 1.5 mg dose of [d- 
Phe^ Phe^ D-Phe^l-LH-RH in 0.5 ml of ve- 
hicle or with vehicle alone (20% propylene 
glycol/saline solution) at noon of either E, 
Dl, D2 or proestrus. Another group was in- 
jected at 9 AM on proestrus. On the following 
estrus, the animals were sacrificed and ovu- 
lation was checked by counting the number 
of ova under a dissecting microscope. The 
number of rats which ovulated compared to 
the total number of rats was considered an 
index of the antiovulatory activity of the an- 

In a second experiment, rats were injected 
s.c. with a 1.5 mg dose of [D-Phe^, Phe^ d- 
Phe^]-LH-RH twice a day (9 am and 4 pm; 
total dose: 3 mg/day) during E, Dl and D2. 
No injection was given on proestrus. Control 
rats were injected with 0.5 ml of vehicle alone. 
On the following E, ovulation of both control 
and experimental animals was checked as 
described above. At 4 pm on each day of 
treatment, a blood sample from the jugular 
vein of control and experimental animals was 
collected within 20-30 sec under light ether 
anesthesia. The blood was centrifuged and 
sera separated and stored at -20*^ until as- 
sayed for LH and steroids. Some ovaries from 
control and [D-Phe^ Phe^ D-Phe^l-LH-RH 
treated rats were removed at the time ovula- 
tion was being determined. The ovaries were 
fixed in Bouin's solution and then stained 
with hematoxylin-eosin (Bay Histology Ser- 
vice, San Rafael, CA). Vaginal smears were 
also examined daily during the period of 
treatment. Serum LH was determined by the 
double antibody radioimmunoassay method 


AU h%hl& T«sierv«d. 



of Niswender et al (6) as described elsewhere 
(7, 8). NIH-LH-Sn was used as the standard 

Estradiol and progesterone were measured 
in duplicate by the method of Abraham et al 
(9) with slight modifications. About 1000 
dpm of 2,4,6,7-'H-17j8 estradiol (SA:91, 3 
Ci/HLA/) and of 1,2,6,7-^H progesterone (SA 
103 Ci/mM) were added to one ml of plasma 
to estimate the recovery of the steroids. Each 
sample was extracted twice with anesthetic 
ether (Mallinckrodt). The ether extract was 
evaporated and the steroids were then resus- 
pended in 1 ml of isooctane and were chro- 
matographed on celite micro-columns. Pro- 
gesterone was eluted with isooctane and es- 
tradiol with isooctane: ethyl acetate (3:2). The 
estradiol fraction was diluted in 0.3 ml of 0.1 
M phosphate buffer (pH 7.8) containing 0.14 
M NaCl, 0.01 M EDTA, 0.015 M sodium 
azide, and 0.1% gelatin. After an aliquot was 
taken to estimate steroid recovery, 0.2 ml 
of the solution was incubated with 2,4,6,7-^H- 
17)8 estradiol (0.1 ml/40,000 dpm) and with 
estradiol antiserum (0. 1 ml at 1/100,000). The 
antiserum (S-310) was obtained from Abra- 
ham's laboratory. The estradiol recovery was 
80%. The sensitivity of the assay was 2.5 
pg/tube with an interassay coefficient of 

The progesterone fraction was diluted with 
1 ml of phosphate buffer; an aliquot was 
taken to estimate recovery and another (50 

jbd) was incubated with 1,2,6,7,^H progester- 
one (0.1 ml/40,000 dpm) and with progester- 
one antiserum (0.1 ml at 1/7,000). The anti- 
serum 3-oxime-BSA cross reacted with the 
following steroids: Testosterone and 20a-OH- 
progesterone less than 1%, and 17j8-OH-pro- 
gesterone and deoxycorticosterone 2%. The 
recovery was 86%, the interassay coefficient 
of variation was 10%, and the sensitivity was 
25 pg/tube. The free and bound hormones 
were separated using 0.2 ml of dextran-coated 

Duncan's new multiple range test (10) was 
used to analyze the significance of the diiffer- 
ences in LH serum levels among the groups. 
The results from the ovulatory test were ex- 
pressed as binomial data using one for ovu- 
lation and for no ovulation; they were 
subjected first to analysis of variance (1 1, 12) 
and then compared by Duncan's new multi- 
ple range test (10) as described previously 
(5-8). The LH-RH analogue was prepared in 
our laboratory by the solid phase method (5). 
Its purity was confirmed by TLC and amino 
acid analysis. 

Results, Table I shows the effect on ovula- 
tion of a single dose of [D-Phe^ Phe^ D-Phc*] 
-LH-RH injected at different days of the 
estrous cycle. It can be seen that when the 
analogue was injected at noon of proestrus, 
a 100% blockade of ovulation was observed. 
The degree of ovulation blockade decreased 
to 33% and 17% when the analogue was 

TABLE 1. Blockade of Ovulation in the Rat by [D-Phc^ Phc\ D-Phc*l-LH-RH (Analogue) Administerh) 

AT Different Stages of the Estrous Cycle." 

# of rats 

ovulaled/lotal # of 

% of blockade of 

Mean ± SE of ova in 




ovulating rats 

A. Proestrus (noon) 

1. Vehicle 


12.2 ± 0.5 

2. Analogue 




B. Proestrus (9 am) 

3. Vehicle 


13.0 ±0.6 

4. Analogue 



10.7 ± 0.9 

C. Dieslrus 2 (noon) 

5. Vehicle 


12.7 ± 0.2 

6. Analogue 



9.8 ± 1.1 

D. Dieslrus 1 (noon) 

7. Vehicle 


12.0 ± 0.4 

8. Analogue 


11.6 ±0.2 

E. Esirus (noon) 

9. Vehicle 


11.0 ±0.4 

10. Analogue 


10.2 ± 0.6 

" Dose of analogue: 1.5 mg/rai at the lime shown in parenthesis. Duncan's new multiple range test: • Significantly 
different from the respective control value, t Significantly different from the value of Group 2. 



d at 9 AM on proestrus and on D2, 
tively. Ovulation was not blocked after 
ng the analogue on Dl and the pre- 
E (Table I). 

effect of daily injections for three days 
I, and D2) of [D-Phe^ Phe^ D-Phe^- 
H is presented in Table II. Two out of 
en rats treated with the analogue ovu- 
(86% blockade of ovulation), one of 
fully (12 ova) and the other partially (4 
On the other hand, only one of the 
•1 rats failed to ovulate (Table II). 
ire 1 shows the efTect of daily injections 
Phe^ Phe^ D-Phe^-LH-RH on the LH 

levels. In the analogue-treated rats, 
LH was significantly lower (P < 0.01) 
lie control group, when they were com- 
on the evening of proestrus. This dem- 
tes that the injection of [D-Phe^, Phe^, 
^-LH-RH during E, Dl and D2 in- 
1 the LH surge that was seen in the 
1 animals on the afternoon and evening 

ire 2 shows the effects of [D-Phe^ Phe^ 
^]-LH-RH on serum levels of steroids. 

estradiol in the treated rats was not 
;d, progesterone was higher on Dl and 
on the afternoon of proestrus in those 
Moreover, the animals treated with the 
;ue showed a normal vaginal smear 
a during the period of observation. 

ovaries of the anovulatory animals 
1 with [D-Phe^ Phe^ D-Phel-LH-RH 
d a histological pattern similar to ovu- 

untreated animals; normal follicular 
»pent including antral follicles were 
It, although corpora lutea hemorrhagica 
ibsent in the latter animals. 
nission, Beattie et al (4) reported that 
e^ D-Ala®]-LH-RH significantly in- 
1 ovulation when it was injected on 

E 11. Blockade of Ovulation in the Rat by 
>-Phc^ Phc^ D-Phc^-LH-RH (Analogue) 
Administered on E, Dl and D2." 

# of rats %of 

ovulated/ blockade Mean ± SE 

total # of of ovula- of ova in ovu- 

oup rats tion lating rats 

11.8 ±0.3 
8.0 ± 4.0^^ 




•se of analogue: l.S mg/rat twice a day (total dose: 
laily). VOne rat ovulated four ova. Duncan's 
e range test: * Significantly different from the 
f the control group. 

Fig. 1. Effect of the administration of (o-Phe^ Phe^ 
D-Phe'^l-LH-RH on serum LH levels. Animals were in- 
jected s.c. with either the analogue (l.S mg) or vehicle at 
9 AM and 4 pm on E, Dl and D2. Blood was taken at 4 
PM. Each point represents the mean ± SE of six rats. 


■r^pTiVM^ C:P^«> 




Fig. 2. Effect of administration of [o-Phe^ Phe^ d- 
Phe*l-LH-RH on serum steroid levels. Animals were 
injected s.c. with either the analogue (1.5 mg) or vehicle 
at 9 AM and 4 pm on E, Dl and D2. Blood was taken at 
4 PM. Each point represents the mean ± SE of six rats. 

different days of the estrous cycle in 4-day 
cycling rats. Ovulation was inhibited by 
97%, 87% and 79% after proestrus, D2 and 
Dl injection of the analogue, respectively. 
Using a more potent analogue, [D-Phe^, Phe^, 
D-Phe^]-LH-RH (5), we were able to inhibit 
ovulation considerably when this analogue 
was injected either on the morning or at noon 
of proestrus. Only a 17% inhibition of ovu- 
lation was observed when [D-Phe^, Phe^, d- 
Phe^]-LH-RH was injected at noon on D2 
and no inhibition of ovulation was seen when 
it was injected on the previous E. Apparently, 
the length of the action of [D-Phe , Phe^ d- 
Phe®]-LH-RH is not sufficient to block ovu- 
lation when it is injected before Dl. Further- 
more, the injection of 1.5 mg twice a day at 
9 AM and 4 pm (total dose of 3 mg/rat/day) 



on E, Dl and D2, brought about almost 
complete suppression of the LH surge on the 
next day (Proestrus) and an 86% blockade of 
ovulation on the following estrus morning, 
without altering the normal vaginal smear 
pattern. This might be due to unaltered serum 
levels of estradiol after the treatment (Fig. 2). 
Discrepancies between our results and tihose 
obtained by Beattie et al (4) might be due to 
the different schedule of treatment and doses 

It is interesting to point out the changes in 
serum progesterone levels observed in the 
animals treated with the analogue on E, Dl 
and D2 throughout the experiment. They 
were higher on Dl and lower on proestrus 
when they were compared with those of the 
control rats. Because the peak of serum LH 
levels in the rats of our colony occurs between 
3 and 4 pm, the low proestrous afternoon 
levels of progesterone could be due to a 
blockade of LH release produced by direct 
effects of the analogue on the pituitary and 
hypothalamus. The lack of LH release could 
have impaired the subsequent ovulation and 
luteinization. On the other hand, high pro- 
gesterone levels during Dl might have con- 
tributed to the blockade of the LH surge and 
ovulation. It has been demonstrated that ad- 
ministration of progesterone or synthetic an- 
alogues early in the cycle depresses proestrus 
serum LH and FSH and delays ovulation 

In conclusion, using antagonist analogues 
of LH-RH it is possible to block ovulation 
without affecting the rat estrous cycle. Thus, 
the possibility exists to develop an even more 
potent analogue which can be used in humans 
without altering plasma estrogen levels. 

Summary, [D-Phe^ Phe^ D-Phe®]-LH-RH, 
a potent antagonist of LH-RH, was injected 
during the different stages of the estrous cycle 
in rats at a dose of l.S mg/rat. When it was 
administered at noon or proestrus, a 100% 
blockade was observed. This decreased to 
33% and 17% when the analogue was injected 
at 9 AM on proestrus and diestrus 2, respec- 
tively. No blockade of ovulation was ob- 
served after the injection of the analogue on 
or on the previous estrus. The 
administration of the analogue 
on E, Dl and D2, brought about 
\ suppression of LH surge on 

proestrus and an 86% blockade of o\ 
without altering the cyclic vaginal smt 
tern. In this case, serum levels of e 
were not modified but progesteron< 
were significantly lower on proestr 
higher on diestrus 1 in the analogue 
group as compared to control rat 
higher level of progesterone on die 
might account in part for the inhibitio 
LH-surge and blockade of ovulation 
inhibitory analogues of LH-RH. 

We thank Mrs. J. Gauthier and Mrs. J. ' 
their valuable technical assistance; Dr. G. N 
Dr. Ward and NIAMDD-Rat-Pituitary Horn 
gram for the gifts of materials used in radioin 

1. de la Cruz, A., Coy, D. H., SchaUy, A. V., C 
de la Cruz, K. G., and Arimura, A., Proc. ! 
Biol. Med. 149,576(1975). 

2. Corbin, A., and Beattie, C. W., Endoa 
Commun. 2, 1 (1975). 

3. de la Cruz, A., Coy, D. H., VUchez-Martic 
Arimura, A., and Schally, A. V., Science 

4. Beattie, C. W., Corbin, A., FoeU, T. J., Gi 
Rees, R. W. A., and Vardely, J., Contracc 
341 (1976). 

5. Vilchez-Martinez, J. A., Coy, D. H., Cc 
Arimura, A., and Schally, A. V., Endoci 
Commun. 3, 231 (1976). 

6. Niswender, G. D., Midgley, A. R., Monn 
and Reichert, L. E., Proc. Soc. Exp. Biol. I 

7. Vilchez-Martinez, J. A., Arimura, A., and 
A. v., Acta Endocrinol. 81, 73 (1976). 

8. Nishi, N., Coy, D. H., Coy, E. J., Arimura 
SchaUy, A. V., J. Reprod. FertU. 48, 1 19 (1 

9. Abraham, G. E., Hopper, K., Tulchinsky, 1 
dloff, R. S., and OdeU, W. D., Analyt. Utl 

10. Steel, R. G. D., and Torrie, H. J., "Princ 
Procedures of Statistics", McGraw-Hill, N 

1 1 . Hsu, T., and Feldt, L. S., Amer. Res. J. 6, 5 1 

12. Seeger, P., and Gabrielsson, A., Psychol. 

13. Everett, J. W., Endocrinlogy 43, 389 (1948] 

14. Schwartz, N. B., Rec. Prog. Horm. Res. 25, 

15. Zeihnaker, G. H., Acta Endocrinol. (KbH 

16. Redmond, W. C, Endocrinology 83, 1013 i 

17. Beattie, C. W., and Corbin, A., Endocrim 

Received December 19, 1977. P.S.E.B.M. 1978, 



Annual Report of the Secretary-Assistant Treasurer and Managing Editor for 
the Year Ending December 31, 1977 

finance. The following is an abbreviated financial report prepared by Leo Kaden, C.P.A., of Padell, Kaden, 
lell and Co. 

Balance of cash in banks at January 1. 1977 $29,153 


From "Proceedings": 





Publication charges 






Other receivables 


Total from "Proceedings" 






Excess of investments sold over investments purchased 


Dividends interest and income 


Sundry receivables 


Monies received from Internal Revenue Service 


Due from Michigan Section 




Refunds of Honorariums 


Total receipts for period 


Toul funds available 


From "Proceedings": 

Academic Press 249,078 

Printing 2,042 

Total for "Proceedings" 25 1 , 1 20 

Office supplies, postage, and telephone 10,386 

Subscriptions, refunds, and Honorarium 2,650 

Meetings and travel 3,862 

Pension 4.519 

Salaries and payroll taxes 4 1 .048 

Professional fees 1,250 

Bank charges and miscellaneous expenses 150 

National Society for Medical Research 1,500 

Total disbursements for the period 3 16,484 

Balance of cash in bank at December 31, 1977 $58,056 


0037-9121 /!%/ \SS\-bmvVi\S»l^ 

M\ h^hls Ttserved 



Editors. The Editorial and Publication Committee 
consists of: Drs. M. Zucker, Chairperson; I. Clark, M. 
Hilleman, S. I. Morse, and S. SeiAer. 

The following editors will be concluding their term of 
office in the Society. They have elected to serve for 
another 3-ycar period. They are: Drs. E. Alpen, W. McD. 
Armstrong, L. M. N. Bach, C. A. Barraclough, A. L. 
Barron, K. G. Brand, E. D. Bransome, Jr., S. M. Cain, 
H. F. Clark, I. Clark, P. P. Foa, Z. N. Gaut, J. Genest, 
E. C. Gotschlich, N. S. Hahni, J. J. Holland, C. Howe, 
S. Klahr, S. Koletsky, C. A. Krakower, M. Kuschner, C. 
Lenfant, R. D. Leverc, L. C. McLaren, J. Meitcs, C. R. 
Morgan, S. I. Morse, P. J. Mulrow, P. N. Patil, 1. 
Rothchild, W. H. Sawyer. R. A. Schemmel R. Schmid, 
N. J. Schmidt, J. H. Shaw, J. G. Stevens, G. J. Thorbecke, 
D. B. Zilversmit. 

The following editors will be retiring. They are: Drs. 
J. J. Bourgoignie, S. Chien, I. Green, P. D. Harris, S. G. 
Korenman, E. H. Lennette, C. S. Lieber. H. Liebhaber, 
H. R. Morgan, M. B. A. Oldstone, J. H. Oppenheimer. 
R. R. Paradise, R. J. Peanasky, H. J. Weiss, M. Winick. 

Read, E. M. Shcvach, J. L. Vaitukaitis, R. M. Welsh, Jr. 

National Membership Committee. Drs. A. K. Weiss, 
Chairman; A. W. Bemheimer, S. Chien, E. Farber. E. 
D. Frohlich, L. S. Hurley, H. W. Overbcck. M. R. 

In Memorium. P. Bard, C. W. Fishel, J. H. Gast, B. 
N. Horwitt, 1. Innvcrfield, T. D. Kinney, J. KoUias, R. 
S. Ual, T. A. McCoy, M. S. Rabcn, C. S. Stulbcrg. W. 
H. Summerskill. 

Election. The mail ballot has resulted in the election 
of Dr. R. W. Berliner, as President-Elect for 2 years 
beginning January I, 1978; Dr. G. W. Siskind. Treasurer, 
and Dr. M. R. Nocenti, Secreury-Assistant Treasurer, 
for a similar period. 

The following were elected members of the Council 
for a period of 4 years: Drs. A. H. Briggs, H. F. DeLuca. 
J. P. Gilmore, M. W. Orsini, E. E. Selkurt, and D. B. 

The following have been elected to the Editorial 
Board for a term of 3 years. They are: Drs. D. R. Boggs, 
N. S. Bricker, N. E. Cremer, R. R. Gala, M. Greenwood, 
D. Horrobin, E. D. Jacobson, P. L. LaCelle, J. H. Laragh, 
G. J. MacDonald, T. C. Mcrigan, Jr., F. N. MiUer, S. 
Oparil, R. Ostwald, D. D. Porter, J. A. Ramaley, W. O. 

Tellers. Drs. M. Blank and R. Emmers. 

Miss Felice M. 0*Grady has been in charge of the 
duties of the Society's National Office. 



Washington, D.C. 


Ohio Valley 





1. C. Munro 
M C. Davison 
J. C. Hampton 
R. W. Gardicr 
M. Mizell 
L. P. Gangarosa 
L. Hinshaw 




pnent to Reviewers 

laging Editor and the members of the Editorial Board thank the following scientists 
ssistance with the reviewing of manuscripts submitted for publication in the 
ng the year 1977. Their service to the Society and the biomedical and scientific 
is appreciated and commended. 


M. Bissel 

O. A. Carretero 

J. P. Davis 


L. Z. Bito 

K. J. Catt 

J. W. Davis 


H. S. Black 

W. F. Chambers 

V. DeFendi 


B. C. Black-Schaffer 

R. M. Chanock 

F. Deinhardt 


C. A. Blake 

C. P. Channing 

P. Del Soldato 

R. L. Blakley 

E. A. Chaperon 

H. F. DeLuca 

G. Blobel 

P. W. Choppin 

P. DeWeer 


E. I. Bloomquist 

E. I. Ciaccio 



D. E. Bockman 

F. Clark 



P. Bodel 

J. H. Clark 

G. F. DiBona 

R. J. Bonney 

J. W. Clarkson 

N. R. DiLuzio 


L. Borella 

K. H. Clifton 

J. A. DiPaola 


H. B. Bosmann 

D. L. Coffin 

J. DiSalvo 


R. K. Boutwell 

L. A. Cohen 

F. L. Dixon 


W. C. Boyd 

P. P. Cohen 

H. C. Doku 

S. Bradley 

R. Cole 

E. Domino 


M. G. Brattain 

J. E. Coligan 

D. M. Donaldson 

G. A. Bray 

D. G. CoUey 

B. H. Douglas 


B. M. Breckenridge 

H. R. Colten 

R. G. Douglas 

R. Bressler 

B. Combes 

S. D. Douglas 


C. M. Brinkman 

J. D. Conner 

T. P. Dousa 


D. A. Brodie 

M. E. Conrad 

J. W. Drysdale 


M. J. Brody 

P. Constantinides 

P. Dukes 

T. M. Brody 

M. A. Come 

K. Eakins 


F. Bromer 

J. A. Cook 

S. Ebbe 

A. M. Brown 

K. E. Cooper 

R. Edelman 


R. G. Brown 

M. Costlow 

T. S. Edgington 


J. S. Buckner 

R. Couch 

P. Eggena 


R. L. Bums 

A. W. Cowley 

A. B. Einstein 


P. Calabresi 

J. E. Craighead 

R. J. Elin 

F. T. Caldwell 

G. Cudkowicz 

L. Ellenbogen 

J. B. Campbell 

H. M. Cunningham 

S. A. Ellison 


D. M. Carlson 

M. F. Dallman 

W. K. Elwood 


J. R. Carlo 

P. R. Dallman 

J. W. Ensinck 


R. A. Carey 

J. C. Daniel 

A. Epstein 

C. B. Carpenter 

W. H. Dantzler 

A. Erslev 


W. A. Carter 

J. D. Davis 

J. L. Evans 

0037-9721 /l%/\S9\-bmuV^\a^|^ 

Copyright (D \r\% b\ vV\c Socv«\^ to\ txvtt\twttvv%\^\cJ«%^ wA>K^^ 
AU rights rc^rved. 



J. H. Exton 
A. S. Fauci 
R. E. Feency 
J. D. Feinblatt 
P. FcUg 
M. Femandes 
J. P. Filkins 
J. Fine 
J. Fischberg 

D. Fitzgerald 
J. P. Flatt 

A. Flynn 
G. E. Foley 
J. S. C. Fong 
R. M. Forbes 
W. Y. Forbes 

E. L. Forker 
R. E. Forster 
J. G. Forte 

H. H. Freedman 

D. W. Frederiksen 
M. V. 


E. Freese 
R. Freter 
G. H. Fried 

P. A. Friedman 
L. T. Frobish 
J. C. Frolick 
H. A. Gaafar 
R. Ganguly 
G. E. Gantner 
J. Gardner 
W. F. Geber 
M. Gelato 
J. E. Gerich 
G. C. Gerritsen 
R. K. Gershon 
1. 1. Geschwind 
A. M. Geumi 
D. J. Giard 
T. P. Gibson 
R. E. Giles 
T. M. Glenn 
G. Glick 
J. A. Glomset 

F. C. Goble 
A. Goldin 
M. Goldin 

A. L. Goldberg 
J. Goldman 
A. Goldstein 
I. J. Goldstein 
J. L. GoUan 
C. J. Goodner 
R. A. Gorski 
A. A. Gottlieb 
A. M. Gotto 
H. L. Greene 
G. S. Greenwald 

E. Grim 

J. B. Grogan 

A. C. Groom 
M. 1. Grossman 
C. J. Guber 

C. Gumey 

F. J. Haddy 
H. D. Hafs 
S. Hakomori 
K. A. Halmi 

K. J. Hammerman 

B. Hampar 

P. C. Hanawalt 
R. I. Handin 

B. F. Harland 
P. E. Hartman 
M. R. Haussler 
R. J. Havel 

J. P. Hayslett 
O. M. Hechter 
F. W. Heggeness 
J. P. Heggers 

C. Heidelberger 

D. D. Heistad 
K. E. Hellstrom 
H. G. Hempling 

E. Henderson 
R. 1. Henkin 
E. C. Henson 
V. Herbert 

J. Hill 

B. L. Hillcoat 

P. Hinkle 

L. B. Hinshaw 

A. J. W. Hitchman 

A. F. Hofmann 

W. W. Hofmann 

W. Hollander 

J. Holcenberg 
R. T. Holman 
H. Holtzer 
W. B. Hood 
E. C. Homing 
S. L. Hosowitz 
C. F. Howard 
J. B. Howard 
S.-S. Huang 
K. Hubel 
G. A. Hudson 
M. J. Hughes 
L. S. Hurley 
Y. Israel 
K. J. Isselbacher 
H. D. Itskovitz 
E. J. Jacobson 
W. B. Jakoby 
J. D. Jamieson 
L. S. Jensen 
A. G. Johnson 
L. R. Johnson 
W. J. Johnson 
K. C. Jones 
J. G. Joshua 
R. A. Jungman 
P. J. Kadowitz 
M. L. Kakade 
G. Kaley 
M. A. Kaliner 
A. M. Kaplan 

E. L. Kaplan 
A. Kappas 
S. Karpatkin 
A. J. Kastin 
A. Katz 

R. L. Katz 
H. M. Katzen 
S. Kaufman 

F. Kavaler 
H. R. Keiser 
J. W. Kendall 
P. Kezdi 

C. V. Kies 

D. G. Kilbum 
Y. S. Kim 

V. E. Kinsey 
D. M. Kipnis 

M. Klagsbrun 
J. I. Kitay 
C. R. Kleemar 
J. L. Kleinerm 
L. M. Klevay 
J. J. Kocsis 
S. S. Koide 
P. E. Kolattuk 
J. H. Kom 
J. H. Komby 
H. Koprowski 
H. S. Koren 
J. H. Kom 
J. Kowal 
P. A. Kramer 
S. Krantz 
E. L. Krawitt 
G. Kreibich 
L. Kremizner 
M. E. Kriebcl 
G. A. Krishna 
S. C. Kriss 
S. C. Kukreja 
I. Kupfermann 
K. Kurokawa 
P. E. Lacy 
P. Lalezari 
C. E. Lane 
J. Laragh 
P. C. Laris 
P. C. Larson 
S. A. Latt 
A. M. Lawrcnc 
H. S. Lawrcna 
C. Y. Lee 
A. M. Lefer 
H. A. Lester 
J. Levin 
G. M. Levinc 
L. Levine 
S. B. Levy 
J. G. Levy 
L. Levy 
T-K. Li 
L. M. Lichtcns 
J. Lieberman 
M. W. Liebcnr 
L E. Liener 



I. M. Lipton 
I. W. Littlefield 
J. Locb 
J. Lostrob 
B. Lucchesi 
L. N. Lukens 
J. D. Lutton 
S. E. Lux 
D. Macfarlane 
B. MacMahon 
D. F. Magee 
L. E. Mallette 
J. MaUon 
R. C. Manok 
G. M. Maniatis 
G. R. Marchand 

F. I. Marcus 
R. Marcus 

H. S. Margolis 
H. Markowitz 
J. M. Marsh 
R. F. Marsh 
J. Martin 
J. R. Martinez 

G. R. Marshall 
G. T. Mannering 

D. T. Mason 
B. G. Massry 
L. C. Maxwell 
M. W. McBumcy 
R. McCuskey 

J. D. McGarry 
L. M. McKenzie 
J. M. McKenzie 
J. F. McKelvy 
J. M. McLaughlin 
I. F. McMurtry 
P. McLaughlin 
H. Megel 
J. H. Menna 
K. M. J. Menon 
T. C. Mcrigan 

E. Merler 

H. J. Mersmann 
G. Meschia 
R. A. Meyer 
W. L. Meyer 
J. G. Michael 

F. Milgrom 

F. N. Miller 

G. Miller 

J. N. Miller 
W. L. Miller 
J. A. Mills 

T. Mohanakumar 
H. W. Moon 
A. S. Moore 
M. Morad 
H. P. Morris 
T. Q. Morris 
N. E. Morrison 

E. E. Morse 
H. D. Mosier 
M. Moss 

W. J. MueUer 

A. E. Munson 
R. A. Murphy 
S. J. Mustafa 
R. D. Myers 
C. F. Nathan 
M. Navia 

W. A. Neill 
H. N. NeUans 
L. Nelson 
P. M. Newbome 
H. D. Niall 

F. H. Nielsen 
K. D. Nolph 
M. J. Novy 
J. A. Oates 

C. Oliver 
J. A. Olson 

B. W. O'MaUey 
S. Oparil 

R. A. Ormsbee 
J. E. Osbom 

D. Osoba 

J. V. Osterman 
D. L. Oxender 
J. S. Pagano 
R. H. Painter 
H. C. Palfrey 
H. Papkoff 
A. B. Pardee 

C. W. Parker 
J. C. Parker 
F. E. Payne 

C. M. Pearson 
R. A. Pederseh 
R. S. Pekarek 
J. N. Periera 
B. Peterkofsky 
R. G. Petersdorf 
E. R. Pfefferkom 
B. B. Pharriss 

J. V. Pierce 

D. J. Pillion 
M. C. Pinsker 
J. J. Pisano 
M. Poe 

R. Pollack 
W. G. Pond 
D. D. Porter 
J. C. Porter 

B. Postic 

M. D. PouUk 
M. C. Powanda 
A. E. Powel 
A. S. Prasad 
K. N. Prasad 

D. Pressmann 
T. G. P. Pretlow 
H. G. Preuss 

J. P. Price 
O. S. Privett 

E. M. Rabellino 
G. S. Rachelefsky 
J. M. Radley 

D. Rabinowitz 
J. Rahkin 
R. G. Rahwan 
N. Raica 
W. T. Rainey 
L. G. Raisz 
J. A. Ramaley 
V. D. Ramirez 
H. N. Ranney 

C. E. Rapela 

F. Rapp 

P. Rathnam 
W. E. Rawls 
C. G. Ray 
L. Raymond 
J. D. Regan 
W. S. Rehm 
A. I. Reid 

R. A. Reisfeld 
K. Reissmann 
R. Relkin 
H. Y. Reynolds 
J. S. Rhim 

A. Rider 

G. D. Riegle 
R. Rifkind 
T. R. Riggs 
J. A. Rillema 
J. M. Ritchie 
J. M. Rivers 
R. Rivlin 
R. S. Roberts 
W. C. Roberts 
D. F. Rochester 

B. Roizman 
S. Roseman 
D. A. Roth 
J. A. Roth 

A. B. Rothballer 

B. Rubin 
R. J. Rubin 
M. N. Runner 
P. S. Russell 
R. J. Rutter 

J. W. Ryan 
J. M. Saavedra 
T. M. Saba 
J. Sacher 

B. Sachs 
G. Sachs 

O. Z. Sallinger 
H. H. Sandstead 
G. Sayers 
H. ^chachter 

C. D. Scher 
R. J. Schimke 

D. Schlessinger 
H. G. Schneider 
R. S. Schnitzer 

B. A. Schottelius 
R. W. Schrier 

C. J. Schwartz 
H. L. Schwartz 
M. K. Schwartz 
C. E. Schwerdt 
P. Seeman 

E. E. Seifen 



L. B. Senterfit 

J. W. Severinghaus 

R. E. Shade 

S. S. Shapiro 

R. K. Sharma 

L. Shear 

W. T. Shearer 

S. S. Shefer 

R. Sheldon 

C. J. Shellbarger 
G. K. Sherer 
M. E. Shils 

R. S. Sherwin 

D. A. Shriver 
H. Shwachman 

E. J. Simon 

D. P. Simpson 
L. L. Skeggs 

A. Small 
R. Smeby 

B. T. Smith 

C. A. Smith 

F. A. Smith 
L. L. Smith 

C. T. Snowden 
M. S. Soloff 
M. K. Song 

E. H. Sonnenblick 
S. Solomon 

T. L. Sourkes 
E. Spaziani 
N. Sperelakis 
H. G. Spies 
J. L. Spinak 
L. M. Sreebny 

S. K. Srivastava 
R. L. Stambaugh 
H. C. Stanton 
B. Stavric 
S. P. Steamer 
D. J. Stechschulte 
R. A. Steeves 
J. H. Stein 
D. Steinberg 
M. Steinberg 
M. Steiner 

B. G. Steinetz 
M. Stem 

C. D. Stiles 
B. D. Stollar 
G. D. Stoner 
J. B. Storer 
M. J. Straus 
R. M. Stroud 
G. P. Studzinski 
M. Subbish 

F. F. Sun 

G. Y. Sun 

R. M. Suskind 

J. W. Suttie 

K. Suzuki 

L. SweU 

J. V. Taggart 

Y. Takeda 

J. F. TaUman 

K. R. Tanaka 

G. S. Tannenbaum 

S. R. Tannenbaum 

R. D. Tang 

R. N. Taub 

J. C. Taylor 
S. M. Tenney 
R. C. Theuer 
G. J. Thomas 
J. C. Thompson 

F. G. Toback 
R. A. Tobel 

G. J. Todaro 
Y. T. Topper 
G. L. Tritsch 
M. D. Trousdale 
H. A. Tucker 

S. A. Turkania 

A. R. Twardock 

D. E. Uphoff 

A. C. Upton 

W. W. Vale 

S. Varon 

H. Valtin 

J. J. Vogel 

R. R. Wagner 

M. A. Wainberg 

R. H. Waldman 

M. C. Walker 

R. I. Walker 

S. F. WaUner 

R. W. Wannemacker 

P. A. Ward 

N. L. Wamer 

T. E. Webb 

G. Weber 

W. Weber 

R. P. Wedeen 

W. O. Weigle 

M. L. WeU 

H. R. Weiss 

B. B. Wentworth 
D. G. Wenzcl 
S. L. Werner 

D. Westmoreland 

B. Wexler 
J. A. White 
J. G. White 
H. B. White 
R. P. White 

C. F. Whitfield 
W. G. Weist 
J. T. Willerson 
J. A. Williams 
W. J. WiUiams 
B. W. Wilson 
J. G. Wilson 
R. B. Wilson 
S. M. Wolff 
K. Wong 

K. D. Wuepper 
J. R. Wundcriich 
R. E. Wuthicr 
J. B. Wyngaardcn 
R. S. Ualow 

E. V. Yamada 
K. Yoshinaga 
O. T. Y. Yu 
A. A. Yunis 

E. Yunis 
W. Sawalich 

F. J. Zemen 
V. A. Ziboh 
T. W. Zieglcr 

G. Zins 

". . . invaluable reading for all biologists." -NATURE 


edited by G. H. BOURNE and J. F. DANIELLI 


". • . mainfalns the tradition and set-up of the previous volumes 
and certainly provides up-to-date data on varied aspects of 
cytology ... a valuable acquisition to any library/' 


". . . in keeping with the high standards set by the editors . . . 
this series is a significant contribution to a science that im- 
pringes on many fields/' 


". . . contains several excellent reviews with a substantial 
an^ount of informafion on each topic .... the volume will prove 
a valuable addition to your library/' —ASM NEWS 

Complete information on each volume in the series is available on request. 

Take advantage of the convenience of our Continuation Order Plan: 

Your CONTINUATION ORDER authorizes us to ship and bill each volume 
automatically, immediately upon publication. This order will remain in effect 
until cancelled. Please specify volume number with which your order is to 
begin. Please direct all inquiries and orders to the Sales Department. 

AP 7403 


A Subsidiary of Harcouri Brace Jovanovich, Publishers 



". . . this series is invaluable for workers at all levels of 

cell biology/' -NATURE 



METHODS IN CELL BIOLOGY (formerly Methods in Cell Physiology) presents 
detailed descriptions of methods and techniques now in use in the field of 
cell biology research— techniques which have not been published in full de- 
tail elsewhere in the literature. Each volume features state-of-the-art 
reviews that bring the reader up to date on the most current innovations 
and refinements of prevailing techniques, as well as discussions of stan- 
dardized methods that frequently reveal a more efficient means of studying 
current problems. Descriptions of each technique are so complete that even 
cell biologists with little or no experience in a particular area can apply 
the technique to their work. 

"The editor, D. M. Prescott, is to be congratulated on the production of a 
thoroughly useful volume." — NATURE 

". . . will be gladly received not only by those working in the field of phy- 
siology of the cell but . . . will become a useful manual for all cytologists 
and biologists." —FOLIA MORPHOLOGICA 

"Something new or hard to find is included in nearly every chapter. Alto- 
gether the volume makes quite a useful collection and is recommended for 
advanced graduate students, research scientists, and any library which tries 
to keep an up-to-date working collection for cell biologists or microbiologists." 


Complete information on each volume in the series is available on request. 

Take advantage of the convenience of our Continuation Order Plan: 

Your CONTINUATION ORDER authorizes us to ship and bill each volume 
automatically, immediately upon publication. This order will remain in effect 
until cancelled. Please specify volume number with which your order is to 
begin. Please direct all inquiries and orders to the Sales Department. 



A Subsidiary of Harcourt Brace Jovanovich, Publishers 




ojftke Societtf 

Our members are requested to note the following list of Friends. We 

wish to express our thanks to each of them. Their contributions help 

materially In meeting the very sharp increase in costs of publication. 


Burroughs Wellcome Co.. Inc. 

Schering Corporation 

Ciba Pharmaceutical Products. Inc. 

G. D. Searle and Co. 

E. 1. du Pont de Nemours & Co. 

Sharp and Dohme. Division of Merck & Co.. Inc. 

Hoffmann-LaRoche, Inc. 

Smith Kline and French Laboratories 

Eli Lilly and Co. 

Squibb Inst, for Medical Research 

Mead Johnson & Co. 

Sterllng-Winthrop Research Institute 

Merrell-National Uboratories 

Syntex USA, Inc. 

Chas. Pfizer and Co.. Inc. 

Wallace Laboratories. Division of Carter 

Ortho Pharmaceutical Corporation 

Products. Inc. 

Rogoff Foundation 

Warner-Lambert Laboratories 

Damon Runyon Memorial Fund 

Wyeth Laboratories 

Sandoz Pharmaceuticals 

General Instractions 

Manuscripts should be written in clear, concise and grammatical English, and should conform to the general 
style of the Journal and the specific instructions listed below. Manuscripts which are not adequately prepared will 
be returned to the authors, since it is not feasible for the Editors to undertake extensive revision or rewriting of 
manuscripts submitted. Contributors, particularly those unfamiliar with English usage, are encouraged to seek the 
help of colleagues in the preparation and review of manuscripts prior to submission. This practice will usually 
reduce the time required for review and will avoid delays in the publications of the manuscript. 


All manuscripts should be submitted to Dr. M. R. Nocenti, 630 W. 168th Street, New York, N.Y. 10032. 


Authors submitting manuscripts containing data from experiments involving recombinant DN A moleculas must 
provide a statement for each of the two reviewers which certifies that their experiments complied with the NIH 
guidelines on physical and biological containment procedures. 

1 . Only original papers will be considered. Manuscripts are accepted for review with the understanding that the 
same work has not been and will not be published nor is presently submitted elsewhere, and that all persons listed 
as authors have given their approval for the submission of the paper; further, (hat any person cited as a source of 
personal communications has approved such citation. Written authorization may be required at the Editors 
discretion. Articles and any other material published in the Proceedings of the Society for Experimenial Biology 
and Medicine represent the opinions of the author(s) and should not be construed to reflect the opinions of the 
Editor(s), the Society, or the Publisher. 

Authors submitting a manuscript do so on the understanding that if it is accepted for publication, copyright in 
the article, including the right to reproduce the article in all forms and media, shall be assigned exclusively to the 
Society. The Society will not refuse any reasonable request by the author for permission to reproduce any of his or 
her contributions to the journal. Send requests for permission to reproduce items published in Proceedings of the 
Society for Experimental Biology and Medicine to: Dr. Mero R. Nocenti, Managing Editor, Society for Experi- 
mental Biology and Medicine, 630 W. 168th St., N.Y., N.Y. 10032. 

A manuscript rejected by the PSEBM should not be re-submitted. All manuscripts will be given a quality rating 
by the two reviewing editors; those manuscripts with low priority ratings will jgot be accepted even though they 
have been classed as generally acceptable. Split decisions will be decided on the basis of the two priority ratings. 

2. a. Manuscripts should be as concise as possible, yet sufficiently detailed to permit critical appraisal. 

b. Manuscripts (including tables, legends, and footnotes) should be double or triple spaced. 

c. The first page of the manuscript should contain the complete title of the paper, category for the *Tableof 
Contents'' (select from list in item 20). names of authors (without degrees), affiliations (including Zip 
Codes), and a running title consisting of no more than 40 characters (including spaces). The second page of 
the manuscript should give the name and complete address of the author to whom ALL correspondence 
should be sent. Please include Zip Code. 

d. Units of weights, measures, etc., when used in conjunction with numerals, should be abbreviated and un- 
punctuated, e.g., 6 R, 3 g, 5 ml, 8% (see No. 20 below). 

3. Manuscripts oi nonexperimental researches, or those with inadequate controls, are not acceptable. 

4. Unnecessary subdivision of a research into several manuscripts is not acceptable. 

5. a. Manuscripts devoted to improvement of procedure or of apparatus may be accepted when a new principle 
is involved or when decidedly superior biological results are obtained. Evidence of such superiority should 
be given. 

b. Confirmatory or negative results will not be accepted unless they are of obvious biological significance. 

6. Length of manuscripts should average 3 printed pages, including tables, charts, and references. The maximal 
length allowed is 7 printed pages. All manuscripts exceeding 17 typed pages (including tables, charts and 
references) will be returned to authors. 

7. Title should be limited to 15 words. Manuscripts should contain an Introduction, Materials and Methods. 
Results, Discussion and a Summary. 

8. Conclusions should be based upon experimental data submitted. 

9. Figures. All figures should be cited consecutively by Arabic numerals in the text with figure legends typed on 
a separate sheet. These should contain sufficient experimental detail to permit the figure to be interpreted without 
reference to the text. Units should be clearly indicated in the figures themselves. Wherever possible, curves should 
be combined into a single figure in order to keep the number of illustrations to a minimum. 

PLEASE NOTE: All figures and illustrations are to be submitted in such form as to permit photographic 
reproduction without retouching or redrawing. This includes the lettering, which is reproduced as part of the 
photoengraving and is not set in type. Line drawings should be carefully drafted with black India ink on white 
drawing paper or blue drafting cloth, no larger than 8.5 x 1 1 .5 inches overall (21 x 27.5 cm). The lettering should 
be large enough to allow a reduction of two-thirds off. High quality glossy prints are acceptable. 

10. Tables. These should be numbered with Roman numerals and cited con^cutively in the text. Each table 
should be titled and typed double-spaced on a separate sheet. Refer to current issues of the Proceedings for the 
acceptable style of tables. The title of each table should clearly indicate the nature of the contents, and sufBcient 
experimental detail should be included in footnotes to the entries to permit the reader to interpret the results. 
vita must be clearly indicated for each of the entries in the table. To save space, repetition of similar experinnents 
^ - - ( which can be calculated from other entries in the table should be avoided wherever possible. 

Footnotes. Footnotes in the text should be identified by superscripts consisting of Arabic numerals and 
d be typed on separate sheet; footnotes in the tables should be identified with superscript lower-case letters a, 
etc.. and placed at the bottom of the Ubie. 

a. References. Only essential citations should be submitted, and they should be arranged nunierically 
at the end of the manuscript. References to the literature should be cited in the text by Arabic numerals 
in parentheses, set on the text line. 

b. Abbreviations of journal titles should follow the style used in Chemical Abstracts (Vol. ACCESS, 
Key to the Source Literature of the Chemical Sciences, 1969 Edition. Please note the style of capitaliza- 
tion and punctuation for journal articles, books, and edited books in the following examples: 

1. Ludens, J. H., Bach, R. R., and Williamson, H. E., Proc. Soc. Exp. Biol. Med. 130, 1 156 (1969). 

2. Abramson, D. I., "Circulation in the Extremities," 557 pp. Academic Press, New York (1967). 

3. Newell, A., and Simon, H. A., in "Computers in Biochemical Research" (R. W. Stacy and B. 
Waxman,eds.), Vol. 2, p. 154. Academic Press, New York (1965). 

c. "Personal communication," "unpublished," "submitted" and numerous abstracts should be ex- 
cluded from the reference list. If the manuscript has been accepted for publication, include it in the ref- 
erence list, giving journal, year, etc. If not accepted do not include it in the reference list. 
Tratk or popular name or abbreviation of a chemical may be used only when preceded by the chemical or 
Lfic name; thereafter, any of these names or abbreviations may be used. Trade names should begin with a 
1 letter. 

Structural formulas of chemicals should be used only when absolutely necessary. 
The Proceedings is copyrighted by the Society for Experimental Biology and Medicine, 

a. Authors are expected to discourage circi^tion of reprints for commercial purposes. 

b. Reprints are limited to 1300. 

Changs in galley proof shouUi be absolutely minimal. Authors will be charged for excessive changes. 
Sponsor is held responsible for non-member's manuscript. The sponsor should write stating (a) how 
ar he is with the author and his research, (b) whether the author is scientifically reliable, (c) that the manu- 
conforms to Notice to Contributors, (d) that he has critically examined the manuscript, (e) wherein the 
script is a significant contribution to science, and (0 that he assumes financial responsibility involved. The 
or should be in the same laboratory as the author. Where not of the same or recent past laboratory, he 
d specify why such letter conits from a member at another laboratory. 

The authors are required to pay a part of the cost of publication in the form of a page charge of $20.00 

tify the General Secretary at least one month before change of address. 

Biochemistry, Endocrinology, Enzymology, Growth and Development, Hematology, Immunology, 
>biology, Nutrition, Oncology, Pathological Physiology, Pathology, Pharmacology, Physiology, Radio- 
Sy, Tissue Culture, Virology. 
Abbreviations. Contributors are requested to use the following abbreviations: 





counts per minute 


cubic centimeter 




degree Cekius (Centigrade) 


degree Fahrenheit 










inside diameter 













spell out 











millimeter mm 

milliosmole mOsm 

minute min 

molal (concentration) m 

molar (concentration) M 

mole spell out 

molecular weight mol wt 

nanogram ng 

nanometer nm 

normal (concentration) N 

osmole Osm 

ounce oz 

outside diameter o.d. 

parts per million ppm 

percent % 

picogram pg 

revolutions per minute rpm 

second sec 

specific activity sp act 

square centimeter cm* 

square meter m* 

subcutaneous sc 

volt V 

volume vol 

pease stand up? 



»w^ A. 

*ii. *r 

It has. For more than 40 years of 
critical pharmacological and toxi- 
cological research. Plus almost every ^ "' 
other area of study for which mice are used. 

The Charles River CF-l's were initiated by Carworth. In recent 
years, they've made you even happier. Because we've made them 
healthier. Through our COBS* (Caesarean-Originated, Barrier- 
Sustained) techniques that have become the standard of exceflence 
for all mice, rats, guinea pigs and rabbits., A major reason why 
we've sold more than 10,000,000 CF-l's' 
alone in the past 4 years. 

From major phamiaceutical companies 
and government n.i^encies to the top univer- 
sity medical •.eiUei-s «'uid private research 
labor-'iioric •. 'iv C!-"-] s ^laiu] up. .Ajid 
slan<: si:, v. ;•' '-i; • 'i.'S .■).'i.''.'^lnr orderinjjj 
orni<.r.- ii;!':- •• •::•;•. !'> I'-l! 'l.v ii'.n];. ouv 

CF-'l i:- i'-i' .\:/: i>r.:. 

Cliarles tiWerjQ> 


Members can help considerably by mentioning our Proceedings 
when communicating with or ordering supplies from our advertisers. 




(MTERESTEO IN PEPTIDES? R-pfus has few equats when it com» to OUALITY or SELECTION. 


■ SECRETIN, pcntaatetate,svnth. (The complete hormone) ■ SECRETIN 21-27 

(15-LEUCINE) HUMAiy GASTRlJy l,svnth. iHG-t7) ■ TETRAGASTRIN. synth. 

(15-METH*0NINE) HUMAN GASTRIN I, synth. (HG-17)** 

HUMAN GASTRIN 2-17. synth," • (15-NORLEUCINEl HUMAN GASTRIN I, «ynth. (H6-17) 

THYROCALCITONm 23-32, synth. • THYROCALCITONIN 27-32, synth, 







ASP-TYR-MET~GLY-TRP-MET'ASP-PHE NH2 (Carboxy termmal ends of Cholecystoklnin 




LEU-TRP-MET-ARG-PHE-ALA (Model peptide substrate) ■ LYS-HIS-LYS 


PHE-ASP-ALA-SER-VAL iC-terminal penfapeptide of RNase) 

PHE-VAL-GLN^TflP-LEU-MET-ASN-THR (Glucagon 21-29) 

SER-GLY-ALA-GLY-ALA-GLY (sequence of silk) 


Currenily in -synthesis. 

STEROIDS' CATALOG, NOW IN PRODUCTION, Thi^ catalog includes antibodies to steroids as well as 
ftf roid antigaM and to the best of our knowledge the only fif m offering antibodies to selected 




Catalogs and literature 

are available 

free upon request. 

Research Pius Labofatones, loc. Bck 571 Denvdfe, IMJ. 07834 
(2011823 3592 ♦ (201) 823-3699 

Council 1978-79 

President, DeWitt Stetten, Jr. 
National Institutes of Health 

President- Elect, Robert W. Berliner 
Yale University 

Past President, Dennis W. Watson 
University of Minnesota 

Treasurer, Gregory W. Siskind 
Cornell Medical Center 

Secretary and Ass' t Treasurer, Mero R. Nocenti 
Columbia University 


Univ. of Kansas 

A. H. Briggs '81 
University of Texas 

H. F. DeLuca '81 
University of Wisconsin 

P. P. Foa '79 
Sinai Hosp. of Detroit 

I.J. Fox '81 
Univ. of Minnesota 

J. P. GiLMORE '81 

Univ. of Nebraska 

M. Orsini '81 
Univ. of Wisconsin 

R. J. Peanasky '79 
Univ. of South Dakota 

E. E. Selkurt '81 
Univ. of Indiana 

M. D. Siperstein '79 
Univ. of California 

D. B. Zilversmit '81 
Cornell Univ. 


'Membership in the Soeietv for Experimental Biology and Medicine is open to all individuals who 
have independently published original meritorious investigations in experimental biology or experimen- 
tal medicine and who are actively engaged in experimental research. In general, applicants should be 
beyond a supervised post-doctoral experience in order to be able to demonstrate the ability to conduct 
independent investigations. 

Application forms may be obtained fn)m the Office of the Secretary, SiKicty for Experimental 
Biology and Medicine. 630 W. I68ih St., N.Y., N.Y. 10032." 

Board of Editors 


Managing Editor 

630 W. 

168th Street 

New York, N. Y. 10032 



R. R. Gala 

C. S. Lieber 

R. B. Roberts 

N, Jr. 

R. C. Gamo 

C. W. Lloyd 

R. Ross 

Z. N. Gaut 




G. L. Gebber 

C. C. Lushbaugh 



J. Genest 

G. J. MacDonald 




D. F. Malamud 

B. B. Saxena 



L Mandl 

A. J. Sbarra 


M. Greenwood 

A. J. Marcus 




A. Mazur 



N. S. Halmi 

S. M. McCann 



C. G. Harford 

L. C. McLaren 

N. J. Schmidt 

P. C. Harpel 

J. Meites 

H. A. Schneider 



T. C. Merigan, Jr. 

E. E. Sei.kurt 



F. H. Meyers 

J. H. Shaw 


J. J. Holland 

F. N. Miller 

E. M. Shevach 




N. Shock 



C. R. Morgan 

M. M. SiGEL 



S. L Morse 



E. D. Jacobson 


N. E. Sladek 


H. D. Janowitz 


A. A. Spector 


D. C. Johnson 

D. Nathan 

R. S. Spiers 

R. C. Johnson 


J. G. Stevens 


T. J. Kindt 

S. Oparil 

A. Stracher 


S. Klahr 

P. Y. Paterson 

E. D. Thomas 


S. Koletsky 

P. N. Patil 

G. J. Thorbecke 


C. A. Krakower 

W. E. Paul 

M. L. Tyan 


L. C. Krey 

M. J. Peach 

J. L. Vaitukaitis 



V. A. Pedrini 

C. M. Veneziale 

P. L. LaCei.t.e 

G.L. Plaa 

C. S. Vestling 


B. N. LaDu 

S. A. Plotkin 

S. R. Wagle 


M. E. Lamm 

D. D. Porter 

M. E. Weksler 


C. A. Lang 

A. S. Rabson 

J. M. Weller 


J. H. Laragh 

J. A. Ramaley 

R. M. Welsh 

C. Lenfant 

M. M. Rapport 



C. E. Leroy 

W. D. Reid 



R. Levere 

J. A. Resko 



J. V. Levy 



Editorial and Publications Committee 

M. ZucKER, '82, Chairperson; L Clark, '80; M. Hilleman, '82; 

S. L Morse, '78; S. Seifter, '82. 

ITie President, President-Elect and Secretary 


Volume 159, Number 2, November 1978 

Copyright © 1978 by the Society for Experimental Biology and Medicine 
All Rights Reserved 

No part of this publication may be reproduced or transmitted in any form or by any means, electronic 
or mechanical, including photocopy, recording, or any information storage and retrieval system, 
without permission in writing from the copyright owner. 

The appearance of the code at the bottom of the first page of an article in this journal indicates the 
copyright owner's consent that copies of the article may be made for personal or internal use, or for the 
personal or internal use of specific clients. This consent is given on the condition, however, that the 
copier pay the stated per copy fee through the Copyright Clearance Center, Inc. (Operations Staff. 
P.O. Box 765, Schenectady, New York 12301) for copying beyond that permitted by Sections 107 or 
108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as 
copying for general distribution, for advertising or promotional purposes, for creating new collective 
works, or for resale. Copy fees for pre- 1978 articles are the same as those shown for current articles. 

Proceedings of The Society for Experimental Biology and Medicine, Vol. 159. No. 2, November 1978. Published monthly 
except August by Academic Press, Inc., 1 1 1 Fifth Avenue, New York. N. Y. 10003. Second class postage paid at New 
York. N. Y. and at additional mailing offices. 1978: Subscription per year $48.00 U.S.A.; $62.00 outside U.S.A. All prices 
include postage. Send notices of change of address to the Office of the Publisher at least 6-8 weeks in advance. Please 
include both the old and new addresses. Copyright (S> 1978 by the Society for Experimental Biology and Medicine. 




-*- '^^j 


When growing \ 
cells in culture is a \ 

1 'xJii^'^^ 

problem, our Growth ^^^'^^'^^^ 

) o^-^• -^ 

Factors may be the ^\^ 

answer. In many cases, they ^X—^^ 

stimulate cell division, modulate ^*> / 

growth and function, and j / 


influence migration and organization If 

1/ ^ 

of cells \n vitro. Collaborative Research fJ 

IL y^"^^ 

Growth Factors now available from CS^ 


stock include: 

Epidermal Growth Factor 


Fibroblast Growth Factor 


Multiplication Stimulating Activity 


Nerve Growth Factor 


Endothelial Cell Growth Supplement 


Human Thrombin 


In addition, we have [^"1] labelled derivatives and antisera for EGF, 
FGF. MSA and NGF. 

To find out more about our Growth Factors and what they can mean 
to your research, call or write for our free technical bulletins. Or for im- 
mediate delivery of these products call our Order Department 61 7-899-1 1 33. 

Pof Research Use Only 

Research Products Division 

Collaborative Research, inc. 

1365 Main Street. Waltham. Mass. 0215 
Tel. (617) 899-1133 TWX 710-324-76C 

Flodent Cages and Sland 


mann has a most complete selection of labo- 
y animal cages and support items, and every 
Is custom crafted to meet or exceed the rigid 
iards mandated for today's animal research 


nann is a privately owned company with its 

operation being one of cage manufacturing 

iervlce. Our people know caging. They know 


experience of our staff has been combined 

first class facilities to enable us to provide 

superior service to our customers. 
Write now for our new complete catalog. See \ 
sixty-eight years of specializing can mean to ; 
laboratory animal housing requirements. 

10 West Aylesbury Road, Timonium, Md. 21 C 
Telephone: (301) 252-2000 


From the Company that is SPECIA 
for its Customers' Benefit 

Members can help considerably by mentioning our Proceedings 
when communicating with or ordering supplies from our advertisers. 



Osborne / Mendel 



Certified Pathogen Fiiee 

Caesarean derived Barrier reared 

iperlativc is the word that best describes the 
w CAMM Certified Pathogen Free Rats 

d you can now get them in Sprague-Dawley. 
iborne Mendel or Wistar strains. 

Jsearchers and scientists from facilities all 
ound the country have been writing to us 
ling us how pleased they are with these new 
^MM rats. They are available now. Ready for 
ipment today by company trucks or Direct 
)ute Air Express. Please contact us with your 
luirements. Ask for our price list. 


By Company Truck and 

Direct Route Air Express 



414 Black Oak Ridge Road, Wayne. New Jersey 07470 

Conventional methods of concen- 
trating macromoiecules from bioiogi- 
cai extractions or separations are 
often a time-consuming problem, 
especially with dilute solutions. Sam- 
ples may be contaminated, inactiva- 
ted, or partially lost. And working 
with small samples Is difficult. 

The I SCO Model 1750 Elect rophor- 
etic Sample Concentrator of fers you a 
new way to concentrate macromoie- 
cules. It's fast, convenient, inexpen- 
sive, non-disruptive, and adaptable 
to a wide variety of samples. 

The Model 1750 will handle 1 to 10 
ml samples, normally produces 50- 
fold concentrations in 90 minutes or 
less, and allows quicl^ and easy 
recovery of concentrates. It uses no 
pumps or gases- samples are con- 
centrated by a combination of 
electrophoresis and filtration. 

The instrument is suitable 
for concentrating proteins. 

nucleic acids, certain whole cells and 
fragments, or any other water soluble 
macromolecule larger than 3500 MW 
and having elect rophoretic mobility. 
During concentration, samples may 
also be separated from non-ionic 
media such as polyacrylamlde gels or 
sugar solutions obtained from density 

Setup time is minimal and four or 
more samples can be concentrated 
simultaneously. Cleaning is simple, 
too -a quick rinse with deionized 
water is all that's necessary. 

Learn more about the versatile 
Model 1750 Elect rophoretic Sample 
Concentrator and other ISCO Instru- 
ments. Send for your copy of our 
catalog today: Instrumentation 
Specialties Company, P.O. Box 5347, 
Lincoln, Nebraska 68505. Or dial 
direct, toll free (800) 228-4250 
(continental U.S.A. except 

Instruments with a difference 


Members and subscribers are requested to cooperate with our advertisers. 

Camm Research vii 

Charles River Cover 4 

CoUabocmUve Research v 

ISCO viii 

Research Plus Lab i 

Wahmann Manufacturing Co vi 


fects of Ethanol on the Absorption and Retention of Lead 
fferential Centrifugation Studies of Guinea Pig Lung Pro- 

Qthesis of Rat Liver Mitochondrial Proteins after the Admin- 
istration of a Nonlethal Dose of Cydoheximide 

J. C. Barton, M. E. Conrad 2 
L. G. Ferren, W. T. Stauber, G. Kalnit- 

SKY 2 

J. J. Ch'ih, P. A. Froman, T. M. Devlin : 


'ect of Ethanol on Parathyroid Hormone and Calcitonin 
Secretion in Man 

mulation of Growth Hormone Release by Intraventricular 
Administration of 5HT or Quipazine in Unanesthetized 
Male Rats 

eraction of Ethanol and Thyroxine on Hepatic Oxygen Con- 

ect of a Phosphodiesterase Inhibitor, 3-lsobutyl 1-Meth- 
ylxanthine, upon the Stimulatory Effect of Human Follicle- 
Stimulating Hormone and Human Luteinizing Hormone 
upon Cyclic Adenosine 3':5'-Monophosphate Accumula- 
tion by Porcine Granulosa Cells 

ictuations of Human Pancreatic Polypeptide in Plasma: Ef- 
fect of Normal Food Ingestion and Fasting 

>lactin Receptors in Mouse Liver: Species Differences in 
Response to Estrogenic Stimulation 

ect of Long-Term Administration of Epinephrine and Pro- 
pranolol on Serum Calcium, Parathyroid Hormone, and 
Calcitonin in the Rat 

idence for Maternal and Fetal Differences in Vitamin D 

G. A. Williams, E. N. Bowser, G. K. 

Hargis, S. C. Kukreja, J. H. Shah, N. 

M. VoRA, W. J. Henderson 1 

E. VuAYAN, L. Krulich, S. M. McCann : 

S. P. Singh, A. K. Snyder 

A. M. LiNDSEY, C. P. Channing 

M. L. ViLLANUEVA, J. A. Hedo, J. Marco : 

S. Marshall, J. F. Bruni, J. Meites 

A. N. Harney, S. C. Kukreja, G. K. Har- 
gis, P. A. Johnson, E. N. Bowser, G. A. 


G. E. Lester, T. K. Gray, R. S. Lorenc 


Lcreatic Secretory Isoenzyme of Alkaline Phosphatase 

W. P. Dyck, a. M. Spiekerman, C. R. 
Ratliff I 


ht Scatter Characteristics of Erythroid Precursor Cells Stud- W. M. Grogan, R. B. Scott, J. M. Col- 

ied in Flow Analysis lins 

nopoicsis in Diffusion Chambers in Strontium-89 Marrow- S. S. Adler, F. E. Trobaugh, Jr 

Ablated Mice 


i-Idiotypic Response of BALB/c Mice to a Myeloma Protein R. Tungkanak, S. Sirisinha 
of BALB/c Origin 

urc of Thymopoietin, Ubiquitin and Synthetic Serum D. Martinez, A. K. Field, H. Schwam, A. 
Thymic Factor to Restore Immunocompetcncc in T-Ccll A. Tytell, M. R. Hilleman 

I^ficicnt Mice 

Table of Contents xi 


Absence of Cytotoxic Effect of Selected Pathogens on HLA B27 D. Dilley, P. T. Fan, R. Bluestone 184 

Positive Fibroblasts 

Depressed Splenic T Lymphocyte Numbers and Thymocyte W. H. Brissette, R. M. Coleman, N. J. 

Mif(ratory Patterns in Murine Malaria Rencricca 317 


Effect of Magnesium Deficiency on Intestinal Calcium Trans- H-F. Chou, R. H. Wasserman, R. 

port in Rats Schwartz 171 

Effect of Diet on Adhesion and Invasion of Microflora in the G. G. Untawale, A. Pietraszek, J. 

Intestinal Mucosa of Chicks McGinnis 276 

Influence of DieUry Fat, Fasting, and Acute Premature Wean- D. R. Romsos, K. L. Muiruri, P-Y. Lin, 

ing on in Vivo Rates of Fatty Acid Synthesis in Lactating G. A. Leveille 308 



9-/3-D-Arabinofuranosyladenine Inhibition of Chemically In- P. J. Price, P. C. Skeen, C. M. Hassett 253 

duced Rat Embryo Cell Transformation 
S-Adenosylhomocysteine MeuboUsm in Rat Hepatomas J. D. Finkelstein, B. J. Harris, M. R. 

Grossman, H. P. Morris 313 


Inhibition ofRenal Prostaglandin Synthesis and MeuboUsm by R. J. Roman, M. L. Kauker, N. A. Ter- 

Indomethacin in Rats ragno, P. Y-K. Wong 165 

Effects of Indomethacin and Tolmetin on Furosemide-Induced B. Noordewier, M. D. Bailie, J. B. Hook 180 
Changes in Renin Release 

Salbutamol as a Topical Anti-Inflanmiatory Drug R. J. Seely, E. M. Glenn 223 

The Effects of Ethanol on Cerebral Regional Acetylcholine T. H. Parker, R. K. Roberts, G. I. Hen- 
Concentration and Utilization derson, A. M. Hoyumpa, Jr., D. E. 

Schmidt, S. Schenker 270 


Siruaural Determinants of the Renal Tubular Activity of Vi- J. Winaver, J. B. Puschett 204 

tamin Ds Derivatives: Studies with la-Hydroxy, 24/{,25- 

Dihydroxy and la,24/{,25-Trihydroxy Vitamin Da 

Effect of Big and Little Gastrins on Pancreatic and Gastric J. E. Valenzuela, R. Bugat, M. I. Gross- 

Secretion man 237 

Mechanism of Prosuglandin £2 Stimulation of Renin Secretion J. L. Osborn, B. Noordewier, J. B. Hook, 

M. D. Bailie 249 

The Effect of Prostaglandin E2 and Indomethacin on the Pla- A. Berssenbrugge, D. Anderson, T. 

cental Vascular Response to Norepinephrine Phernetton, J. H. G. Rankin 281 

Relation of Vitamin D-Dependent Intestinal Calcium-Binding R. H. Wasserman, G. F. Combs, Jr 286 

Protein to Calcium Absorption during the Ovulatory Cycle 

in Japanese Quail 
Glutaminase-Y-Glutamyltransferase: Subcellular Localization T. C. Welbourne 294 

and Ammonia Production in Acidosis 

Accumulation of Latex in Peyer's Patches and Its Subsequent M. E. LeFevre, R. Olivo, J. W. Vander- 

Appearanoe in Villi and Mesenteric Lymph Nodes hoff, D. D. Joel 298 


An Inactivated Hepatitis A Virus Vaccine Prepared from In- P. J. Provost, M. R. Hilleman 201 

fected Marmoset Liver 


)ition of Renal Prostaglandin Synthesis and Metabolism by Indomethacin in Rats 


:hard J. Roman; michael l. kauker, norberto a. terragno, 


iment of Pharmacology, University of Tennessee, Center for the Health Sciences, Memphis, Tennessee 38163 

)inethacin, because of its potency as an 
or of prostaglandin biosynthesis in vi- 
, is widely employed as a pharmaco- 
igent to investigate the renal actions of 
enous prostaglandins. Evidence for in- 
m of renal prostaglandin production in 
IS usually been established by demon- 
g a reduction of prostaglandin release, 
owering of prostaglandin (PG) concen- 
i in renal venous blood or a decreased 
y excretion of prostaglandins. How- 
here are few studies in which the ability 
omethacin to inhibit renal prostaglan- 
nthesis has been systematically evalu- 
uid none of the studies has been done 
» surgically prepared for acute experi- 
tion. Furthermore, release of prosta- 
ns by the kidney probably reflects the 
tivity of enzymes that synthesize and 
le prostaglandins. Assessment of pros- 
din release may not be adequate to 
line the extent of inhibition of prosta- 
n synthesis by indomethacin in vivo 
ndomethacin has been reported to in- 
n vitro the primary prostaglandin cata- 
ig enzymes: IS-hydroxyprostaglandin 
rogenase (PGDH) and prostaglandin 
cetorcductase (9-KRD). Indeed, Ter- 
ei al (2) have recently shown that 
lethacin does not inhibit renal release 
staglandin E2 in conscious dogs. In the 
It study, the effect of indomethacin (2 
;) on prostaglandin release, synthesis 
ital>olism was investigated in anesthe- 
nondiuretic rats. 

erials and methods. Male Wistar rats 
ing between 200-400 g were anesthe- 

sent Address: Biotechnology Resource in Elec- 
obe Analysis LHRRB, Harvard Medical School, 

Massachusetts 021 IS. 
id Reprint Requests to: Department of Pharma- 

University of Tennessee, Center for the Health 
s, 800 Madison Avenue, Box CR-301, Memphis, 
see 38163. 

tized with ip Inactin, 100 mg/kg of body wt. 
After tracheostomy, cannulas were placed in 
the right external jugular vein for infusions 
and the right carotid artery for recording of 
blood pressure. The left kidney was exposed 
and a polyethylene cannula (PE-50) was 
placed in the left ureter to allow for urine 
collections (3). The following drugs were used 
in the present study: indomethacin (Merck, 
Sharp and Dohme), meclofenamate (Parke 
Davis & Co.), phenylbutazone (Geigy Co.), 
RO 20-5720 (Hoffman La Roche, Inc.). The 
following three types of studies were carried 

(a) Prostaglandin bioassay. In each experi- 
ment, two rats were surgically prepared as 
described above and, after one nour equili- 
bration, l>oth members of the pair received 
either indomethacin (2 mg/kg), meclofena- 
mate (2 mg/kg), RO 20-5702 (2 mg/kg), 
phenylbutazone (50 mg/kg) or 3 mAf sodium 
carl>onate vehicle alone. Drugs were infused 
iv at a rate of 40 /il/min in an approximate 
total volume of 0.2 ml/ 100 g body wt. After 
30 min, a 5 ml blood sample was collected 
from the left renal vein over a 1 to 2 min 
period. Blood samples from the two rats were 
pooled and injected into ice-cold ethanol. 
Samples were bioassayed for prostaglandin 
E2-like activity after an acidic lipid extraction 
as described previously (4). Since the extracts 
of blood samples were not chromatographed 
to separate the various prostaglandins, the 
reported values represent total prostaglandins 
and are expressed as the concentration of 
PGE2-like substance in the original samples 
without correction for losses (10-15%) that 
occur during the extraction procedure (4). 

(b) Prostaglandin synthesis. In each of these 
experiments, two rats were prepared as al>ove. 
After a 1 hr equilibration, urine flow and 
blood pressure were recorded during two 
clearance periods of 10 min each. The rats 
were then infused with either indomethacin 


0037-9721 n^/YSSl^iV^SViXS^I^ 
AU hg^U T«iferv«d. 



(2 mg/kg, 4 experiments) or vehicle (3 exper- 
iments) as described above. Urine flow and 
blood pressure were again measured during 
two experimental clearance periods after a 
30-min equilibration. The kidneys from the 
two rats were removed and the renal papillae 
were quickly excised and homogenized in ice- 
cold 0.05 M KH2PO4 buffer, pH 7.4, with a 
Polytron homogenizer. Aliquots of papillary 
homogenates equivalent to SO mg of wet tis- 
sue were incubated at 37^ for 30 min in 2 ml 
of O.OS M KH2PO4 buffer containing 0.4 /iCi 
of l-[^^C]arachidonic acid and 2 mM reduced 
glutathione (S). The reaction was stopped by 
acidification with 1 M citric acid (final pH 
3.0). The reaction mixture was extracted 3 
times with 6 ml of ethylacetate. The com- 
bined extract was evaporated under nitrogen. 
The resulting residue was dissolved in 100 /il 
of chloroform: methanol (1:1, v/v), quanti- 
tatively spotted on thin-layer chromato- 
graphic plates, and separated by chloro- 
form:methanol:acetic acid:water (90:9:1:0.65, 
v/v) as the solvent system. Assays were run 
in duplicate. Prostaglandin production in 
boiled tissue controls was subtracted to cor- 
rect for nonenzymatic formation (5). 

(c) Prostaglandin metabolism. Eight addi- 
tional rats were prepared and infused with 
indomethacin or vehicle as in the prostaglan- 
din synthesis studies. In each experiment, the 
kidneys were removed after the experimental 
clearance periods, the renal cortex and outer 
medulla were excised and homogenized as 
described. The soluble enzyme fractions con- 
taining the PG metabolic enzymes were ob- 
tained by high speed centrifugation 
(105,000^). The fractions thus acquired were 
used to determine the effect of indomethacin 
on the activity of 9-KRD and PGDH (both 
NAD* and NADP"*^ dependent) using pro- 
cedures described previously (6, 7). In brief, 
PGDH activity was assayed by incubating 
aliquots of the high speed supernatant at 37^ 
for 10 min with NAD* or NADP* (4 mM), 
3H-PGE2 (0.56 yLM\ 300,000 dpm, NEN, Bos- 
ton, MA) and 0.05 M KH2PO4 buffer, pH 
7.4, in a final volume of 1 ml. The reaction 
was stopped by acidification with 1 M citric 
acid to pH 3.0. Authentic PGE2 and 15-keto 
PGE2 standards were added to the assay mix- 
ture and extracted 3 times with 2 ml of ethyl- 
acetate. The extract was dried under a stream 

of nitrogen. The residue was redissol 
100 fd of chloroform:methanol (1:1, v/ 
aliquot of 50 /d of the extract was app 
a thin-layer chromatographic plate 
plate, 0.25 mm thick, 20 x 10 cm, sil 
precoated plastic sheets, Brinkman, K! 
separated in iso-octane: ethyl acetate: 
addrwater (25:55:10:50, v/v). PGE2 a 
15-keto metabolite were located by ex 
the TLC plate to iodine vapor, follov 
spraying the plate with 10% phosp 
lybdic acid in ethanol. Areas corresp< 
to authentic PGE2 and 15-keto PGE: 
dards were cut out and suspended in IC 
0.4% Omnifluor toluene liquid scinti 
fluid and counted in a Nuclear Chicagc 
II liquid scintillation spectrometer. T 
served cpm were converted to dpm u 
quench correction curve and externa 
dard channel ratios. The results are exp 
as p moles of 15-keto PG formed per n 
mg of protein. 

9-KRD activity was determined 
presence of an NADPH generating : 
(7) containing: NADPH, 0.15 mM„ gl 
6-phosphate, 3.5 mM; 2 units of glu< 
phosphate dehydrogenase; ^H-PGE2 ai 
M KH2PO4 buffer (pH 7.4), and the s 
enzyme fraction in a final volume ol 
After 10 min incubation at 37^, the re 
was stopped by acidification with 1 A 
acid to pH 3.0. Samples were extracts 
separated by thin-layer chromatogra] 
described above. Areas corresponding 
thentic PGE2 and PGFaa standards w* 
out and the radioactivity was determi 
before. Protein concentration was detei 
by the method of Lowry et ai (8) using 
serum albumin as standard. All assay 
carried out in triplicate and controls w< 
simultaneously using boiled supemata 
suits are presented as the mean ± S 
significance was determined by Stud 
test. P < 0.05 was considered signified 

In order to establish the relations 
different in vitro doses of indomethacii 
inhibition of renal cortical enzyme a 
the effect of increasing concentrations 
domcthacin (0-50 /ig/ml) on three maj 
tical metabolic enzymes were invest 
Assay procedures were similar to th< 
scribed above, different concentration 
domethacin were added to the incuba 



indicated (Fig. 1). 

Results. Urine flow and blood pressure 
were measured in these studies to obtain an 
indication of the physiologic state of the rats 
under the experimental conditions. Control 
urine flows were similar in both vehicle and 
indomethacin treated rats, averaging 1.70 ± 
0.26 and 2.07 ± 0.51 /J/min/lOO g b wt 
respectively. After indomethacin urine flow 
decreased 41% (P < 0.05), whereas after in- 
fusion of an equal volume of vehicle alone it 
increased 61% (P < 0.05). Mean systemic 
blood pressure was unchanged after admin- 
istration of indomethacin (ft-om 1 16 ± 4 to 
113 ± 4 mm Hg, P > 0.1) or vehicle (from 
124 ± 4 to 123 ± 3 nmi Hg, P > 0.2). 

The concentration of prostaglandin Ea-like 
substance in renal venous blood of vehicle 
pretreated rats (Table I) was approximately 
17-fold greater than levels measured in arte- 
rial blood of two additional pairs of animals 
(66 ± 6 pg/ml, P < .01), indicating that 
prostaglandin found in the venous blood of 
these rat kidneys was of renal origin. Mean 
renal venous blood prostaglandin levels were 
significantly lowered, by 69% and 90%, re- 
spectively, in rats inftised with indomethacin 
or meclofenanate. Similarly, in single exper- 
iments 2 other nonsteroidal anti-inflanmia- 
tory drugs (NSAID), phenylbutazone and 


• 9-K R C 



Fig. 1. Dose-response relationship of indomethacin 
on renal cortical prostaglandin metabolic enzyme activ- 
ities in vitro. The effect of indomethacin was expressed 
as per cent of enzyme activity after correction for the 
control. Each point represents the mean of duplicate 
determinatioos; controb were run without indomethacin. 

RO 20-5702 appeared to reduce renal pros- 
taglandin release (Table I). 

The effect of indomethacin pretreatment 
on prostaglandin synthetase activity of renal 
papillary homogenates was also studied in 
vitro, Pretreatment with indomethacin, 2 
mg/kg, significantly reduced the synthesis of 
prostaglandins E2, D2 and ¥20 from their pre- 
cursor arachidonic acid (Table II). Renal 
papillary PGE2 production was inhibited 97% 
by in vivo indomethacin pretreatment. Addi- 
tion of indomethacin, 5 /xg/ml, to incubations 
of renal papillary homogenates obtained 
from vehicle pretreated rats also diminished 
prostaglandin production. The degree of 
prostaglandin synthetase inhibition produced 
by addition of indomethacin in vitro (5 fig/ml) 
and pretreatment with indomethacin in vivo 
(2 mg/kg) was similar. 

Indomethacin pretreatment also interfered 
with renal prostaglandin metabolism in the 
present studies. The effect of indomethacin 
on the key prostaglandin metabolic enzymes 
is shown in Table III. Treatment with indo- 
methacin inhibited renal cortical-medullary 
9-KRD activity by 61% {P < 0.05). NAD^- 
dependent PGDH activity was also dimin- 
ished by 46%, however this decrease was not 
statistically significant. The enzyme NADP"*^- 
dependent PGDH was not affected by indo- 

The effect of indomethacin pretreatment 

TABLE I. Effect of Indomfthacin and Other 
NSAID" ON THE Concentration of Prostaglandin 
E-LiKE Substance in Renal Venous Blood of Rats. 



pg/ml PGErlike 


Vehicle (7) 

1102 ± 167 

Indomethacin, 2 mg/kg (6)' 

343 ± 87' 

Meclofenamate, 2 mg/kg (3) 

108 ± 17" 

Phenylbutazone, 50 mg/kg (1) 


RO 20-5702, 2 mg/kg (1) 


Mean values ± SE are presented. 

° NSAID B nonsteroidal anti-inflammatory drugs. 

* PGEtrlike material was assayed on a cascade of rat 
stomach strip, rat colon and chick rectum. 

' P < 0.005, sutistically different from vehicle treated 

'* P< 0.001, statistically different from vehicle treated 

' Numbers in parentheses « number of samples as- 
sayed. Each sample contained two 5 ml samples of renal 
venous blood obtained from two rats. 



TABLE IL Prostaglandin Biosynthesis by Homogenates of Renal Papillae from Rats Pretreated wfth 

Vehicle or Indomethacin. 

Rate of prostaglandin biosynthesis" 




Total PC 

Vehicle pretreated (3)*' 
Indomethacin pretreated (2 mg/kg) (4) 
Indomethacin added in vitro (5 /ig/ml) (4) 

2.65 ± 0.48 
0.08 ± 0.04' 
0.18 ±0.08' 

0.68 ± 0.20 
0.06 ± 0.01* 
0.15 ±0.12 

0.17 ±0.02 

0.02 ± o.or 

0.05 ± 0.03* 

3.49 ± 0.48 
0.38 ±0.18' 

Mean data ± SE are presented. 

" Values expressed as picomoles of prostaglandin formed/min per mg wet wt of tissue. 

* P < 0.05, compared to vehicle pretreated. 

' P < 0.005, compared to vehicle pretreated. 

'' Numbers in parentheses » number of experiments. 

on the PG metabolic enzymes were also con- 
firmed by the in vitro experiments. Indometh- 
acin at a dose of 5 /xg/ml in vitro produced 
marked inhibition of PG 9-KRD but was less 
effective on NAD^-dependent PGDH. At a 
dose of 25 /ig/ml 9-KRD was inhibited 95% 
while NAD^-dependent PGDH activity was 
lowered only 15%. However, at this dose 
range indomethacin produced little or no ef- 
fect on NADP^-dependent PGDH (Fig. 1). 
Discussion. In the present investigation, in- 
hibition of renal prostaglandin synthetase 
after administration of 2 mg/kg indometha- 
cin to anesthetized nondiuretic rats was as- 
sessed by two methods. These experiments 
demonstrated the following: (a) the concen- 
tration of a PGE-like substance in the renal 
venous blood was reduced 69% by indometh- 
acin; (b) indomethacin pretreatment de- 
creased, by greater than 90%, the conversion 
of radiolabeled arachidonic acid to various 
prostaglandins (PGE2, ¥2^ and D2) by renal 
papillary homogenates; (c) NSAID other 
than indomethacin were also effective in low- 
ering renal venous prostaglandin levels. In- 
domethacin in vivo reduced, but did not com- 

pletely abolish, net renal prostaglandin out- 
put in anesthetized rats prepared for acute 
experimentation. Associated with an inhibi- 
tion of prostaglandin production was a sig- 
nificant decline in urine flow, which is con- 
sistent with the proposal that prostaglandins 
affect tubular handling of water by attenuat- 
ing the antidiuretic action of ADH (9). 

The extent of renal prostaglandin synthe- 
tase inhibition by indomethacin, as deter- 
mined by the decline in renal venous prosta- 
glandin levels (69%), was lower than that 
estimated by in vitro prostaglandin produc- 
tion by papillary homogenates (97%). The 
dissimUar degree of inhibition indicated by 
the two methods may reflect inherent differ- 
ences in the experimental procedures, Ho- 
mogenization of the renal papallae in the 
tissue incubation studies, for example, may 
have allowed indomethacin greater access to 
the enzyme cyclo-oxygenase thus producing 
a more complete blockade of prostaglandin 
synthesis than that which occurred in vivo. 
On the other hand, the present studies pro- 
vide evidence suggesting an alternative expla- 
nation; i.e., the degree of prostaglandin syn- 

TABLE III. Metabolism of Prostaglandin E2 by the Soluble Enzyme Fraction of Renal Cortex and 
Outer Medulla from Rats Pretreated with Vehicle or Indomethacin. 


NAD* dependent NADP* dependent NADPH dependent 

Vehicle pretreated (4/ 
Indomethacin pretreated, 2 mg/kg (4) 

2.12 ±0.66 0.98 ±0.12 0.88 ± 0.18 
I.l4±0.3r 1.04 ±0.12^ 0.34 ±0.12^ 

Mean data ± SE are presented. 

Values are expressed as picomoles of PGF20 or 15-keto PGE2 formed/min per mg protein. 

" 15-PGDH « 15-hydroxyprostaglandin dehydrogenase activity. 

" PG-9-KRD « Prostaglandin E2 9-ketoreductase aaivity. 

'' Not significant P > 0.05. 

** P< 0.05 compared to vehicle pretreated. 

" Numbers in parentheses » number of experiments. 



thesis inhibition after indomethacin may not 
have been accurately reflected by the decline 
in renal prostaglandin release because the 
drug impaired prostaglandin metabolism as 
well as synthesis. Such a conclusion is sup- 
ported by our finding that 9-KRD activity of 
renal cortico-meduUary homogenates was re- 
duced by 61% after indomethacin pretreat- 
ment. Additionally, although a significant 
difference was not detected in the present 
prostaglandin metabolism study, the decline 
of 46% in mean NAD^-dependent PGDH 
activity after indomethacin is consistent with 
the view that indomethacin aflects both pros- 
taglandin synthesis and metabolism. The 
finding that indomethacin inhibited the sol- 
uble enzyme, 9-KRD, after systemic admin- 
istration implies that this compound gained 
access to sites located in the intracellular 

Inhibition of renal cyclo-oxygenase, 9- 
KRD and PGDH by indomethacin and other 
NSAID in vitro has been reported previously 
(10, 11). The concentrations used for half- 
maximal inhibition of PG synthesis were of 
the same order of magnitude as the concen- 
tration shown to produce half-maximal inhi- 
bition of prostaglandin metabolic enzymes. 
The present observations, however, provide 
the first evidence that a standard in vivo dose 
of indomethacin, 2 mg/kg* producing an es- 
timated unbound plasma concentration of 5 
/ig/ml, interferes with prostaglandin metab- 
olism. The effect on the PG metabolic en- 
zymes was confirmed by the in vitro experi- 
ments which indicated that indomethacin in- 
deed affected the major metabolic route of 
PCs in the kidney, llie additional observa- 
tion, both in vitro and in vivo, that 9-KRD 
activity was markedly reduced by NSAID 
especially by indomethacin whereas the en- 
zyme NADP^-dependent PGDH was unaf- 
fected, suggests tlutt these enzymes may have 
different active site(s) even though they have 
been reported to be identical (12). 

In conclusion, indomethacin, meclofena- 
mate and other NSAID markedly reduced 
net renal prostaglandin production in rats 
surgically prepared for acute experimenta- 
tion. It appears from the data reported here 
that indomethacin, after in vivo administra- 
tion, may have a complex action to impair 
both synthesis and metabolism of renal pros- 

taglandins. Differential inhibition of the en- 
zymes involved in net prostaglandin produc- 
tion and alterations in the types of prosta- 
glandins formed in various parts of the kid- 
ney compUcate the interpretation of data ob- 
tained during indomethacin treatment. The 
usefulness of this agent to evaluate the role 
of prostaglandins in the regulation of renal 
function may thus be limited. However, due 
to species differences which exist with respect 
to prostaglandin degradation, the conclusion 
of this study may not be extrapolated to other 

Summary, The effect of indomethacin and 
other NSAID on renal prostaglandin synthe- 
sis and metabolism was studied in nondiuretic 
rats prepared for acute experimentation. 
Thirty minutes after the administration of a 
2 mg/kg iv dose of indomethacin, the con- 
centration of prostaglandin in renal venous 
blood as determined by bioassay was reduced 
69%. In addition, conversion of radiolabeled 
arachidonic acid to prostaglandin E2 in vitro 
by the renal papillae of indomethacin pre- 
treated rats was inhibited 97%. 

Pretreatment with indomethacin also in- 
hibited renal cortical-medullary prostaglan- 
din E2 9-ketoreductasc activity by 61%. 
N AD^-dependent 1 S-hydroxy-prostaglandin 
dehydrogenase activity was diminished 46%; 
however, this inhibition was not statistically 
significant. NADP"*^-dependent 15-hydroxy- 
prostaglandin dehydrogenase activity was 
unaffected by pretreatment. It is concluded 
that indomethacin alters net renal prostaglan- 
din production by inhibiting both prostaglan- 
din synthesis and its metabolism. 

This work was supported in part by research grants 
from the USPHS: HL-22075. AM-17711 and HL- 
1922801 HED; and the American, Tennessee and Mem- 
phis Heart Associations. Prostaglandins and their metab- 
olites were gifts of Dr. Udo Axen of the Upjohn Com- 
pany and RO 20-5702 was kindly supplied by Dr. J. R. 
Paulsrud of Hoffman and La Roche, Inc. We also would 
Uke to thank Misses Judy Early and Pat Goldstein, and 
Mr. Grant Barr for their assistance in these experiments. 

1. Flower, R. J., Pharmacol. Rev. 26, 33 (1974). 

2. Terragno, N. A., Terragno, D. A., and McGiff, J. C, 
Circ. Res. 40, 590 (1977). 

3. Roman, R. J., and Kauker, M. L., Circ. Res. 38, 185 

4. McGifT. J. C, Crowshaw, K., Terragno, N. A., and 
Lonigro, A. J., Circ. Res. 27 (Suppl. 1), 121 (1970). 



5. Wong, P. Y-K, Majeska, K. J., and Wuthier, R. D., 
ProsUglandins 14, 839 (1977). 

6. Wong, P. Y-K, and McGifT, J. C, Biochim. Biophys. 
Acu 500, 436, 1977. 

7. Wong, P. Y-K, Terragno, A.. Terragno, N. A., and 
McGiff, J. C. ProsUglandins 13, 1103 (1977). 

8. Lowry, O. H., Roscbrough, N. J., Fair, A. L., and 
Randall R. J., J. Biol. Chcm. 193, 265 (1951). 

9. Anderson, R. J., Berl, T., McI>onald, K. M. and 

Schricr, R. W., J. Clin. Invest. 5^ 420 (1975). 

10. Stone, K. J., and Hart, M., Prostaglandins 12, 197 

11. Wright, J. T., Jr., Cordcr, C. N.. and Taylor, R., 
Biochem. Pharmacol. 25, 1669 (1976). 

12. Hassid, A., and Levine, L., ProsUglandins 13, 503 

Received February 8, 1978. P.S.E.B.M. 1978, Vol. 159. 


N4agnesium Deficiency on Intestinal Calcium Transport in Rats (40307) 


Division of Nutritional Sciences, Cornell University, Ithaca, New York 14853 

Ca absorption in rats has been 
)c increased (1-4), unaffected (5, 
ased (7) as a result of magnesium 
Possible factors accounting for 
nces are the degree of magnesium 
ngth of the depletion period, and 
cts of inanition or alterations in 
3f the deficient animals. Another 
lakes it difficult to compare find- 
fTerent laboratories is the partic- 
ue used to determine Ca absorp- 
le exception of two studies (1, 2), 

Ca absorption was measured by 
method, previous investigations 

Ca transport in magnesium de- 
were carried out by means of 
itro techniques (3-7). To obtain 
somewhat more applicable to the 
Eignesium depletion in the intact 
t avoiding the inaccuracies of the 
;:edure, we used the ligated loop 
perfusion techniques to estimate 
icross the small intestine. Both 
re carried out in vivo and permit 
It of Ca fluxes in a defined intes- 
it. In situ perfusion has the addi- 
itage of allowing estimation of Ca 
ady state, using serial perfusions 
animal with solutions of varying 

and methods. Male Sprague- 

weighing 120-130 g, were pair- 
:rial which lasted for either 14 or 
th the magnesium deficient and 
s were made from a basal diet 
een described in detail elsewhere 

quantities of magnesium sulfate 
to the basal diet to final concen- 
ppm Mg^^ for the deficient diet 
m Mg^^ for the control diet. The 

tigation was supported in part by N.I.H. 
(-04652 and the Cornell Agricultural Ex- 
jn, Ithaca, New York, 
presented here were submitted in partial 
he requirements for the Ph.D. degree. 

diets contained 20% casein, 0.6% Ca and 0.4% 
P. Deionized water was allowed ad libitum. 

Experiment 1. The rate of Ca absorption 
across the duodenum and ileum was deter- 
mined in rats depleted of magnesium for 14 
and 28 days. All transport estimates were 
carried out after an overnight fast. Ca trans- 
port was determined by the in vivo ligated 
loop technique described by Wasserman and 
Taylor (9). Ten centimeters of duodenum and 
IS cm ileum were ligated in each rat. Each 
loop was injected with O.S ml of a dosing 
solution containing 0. 1 /xCi ^^Ca (Cambridge 
Nuclear, Billerica, MA) in 5 mM CaCb and 
150 mM NaCl, pH 7.2. Five, 15, 30, 60, or 90 
min after dose injection into the loop, the rats 
were anaesthetized with ether and the ligated 
loops were excised. Blood samples, obtained 
by cardiac puncture, were collected in tubes 
containing Na2EDTA. The left tibiae were 
removed and cleaned of connective tissue and 
muscle. Radioactivity measurements were 
made in all intestinal loops and cleaned bones 
by use of a Nuclear Chicago well-type Nal 
crystal automatic ganmia counter. Ca absorp- 
tion was calculated by subtracting percent 
injected ^'Ca remaining in the loop from 
100%. The difference was designated % ^'Ca 
transferred to the body. 

Experiment 2. The in situ perfusion tech- 
nique was used in rats magnesium depleted 
for 14 days to study Ca fluxes in the duo- 
denum. The procedure has been described in 
detail by Wasserman et al (10). Each rat was 
perfused in sequence with five dosing solu- 
tions, each containing 150 mAf NaCl, 0.1-0.4 
/iCi ^'Ca/ml, and 0.5, 1.0, 2.0, 10.0 or 20.0 
mM Ca. All solutions were adjusted to pH 
7.2-7.4. The solutions were perfused into the 
ligated duodenal loop in order of increasing 
Ca concentration at a rate of 4-5 ml/hr using 
a motorized syringe pump (Harvard Appa- 
ratus Co., Dover, MA). An initial equilibra- 
tion period of 40 minutes was allowed for 
each concentration of Ca perfused, and the 
outflow solution collected during this period 


0037-9721 /1%/\S914i\l\Vi\.«^|^ 



was discarded. During the subsequent perfu- 
sion, carried out for 30 min, the outflow 
solutions were collected in graduated tubes. 
Volumes were measured to obtain estimates 
of water inflow and outflow rates. Ca influx 
was calculated as follows (10): Ca infux (lu- 
men to blood) - ("'CaO (Wj) - (*^Cao) 
(Wo)/[(SAi + SAo)/21 X L where: 'Xa = 
radiocalcium content of fluids (cpm/ml); S A 
- specific activity of fluid Ca (cpm/mole); W 
= rate of water flow (ml/hr); L = length of 
duodenal segment (cm); i,o » superscripts 
referring to inflowing and outflowing solu- 
tions respectively. 

Previous studies showed that calcium is 
absorbed by two processes, one process is 
saturable and the other has the characteristics 
of diflusion (11). In the present series, the 
rates of passive and saturable diflusion were 
calculated as follows. A straight line parallel 
to the diflusional component of each influx 
curve was drawn through the origin. At 0.5, 
1, 2, 10 and 20 mAf Ca concentrations the 
values for passive diflusion were subtracted 
from the corresponding values for Ca influx. 
The differences, representing the portion of 
influx due to saturable transport, were plotted 
against luminal Ca concentration. 

Ca and Mg content of blood and tibia. The 
tibiae were dried at 90° for 3-4 days and 
ashed at 550° for 16-20 hr in a Thermolyne 
muffle furnace. The ash was dissolved in 3 
ml concentrated HCl. Suitable dilutions of 
tibia ash or of plasma were analysed for Mg 

and Ca by atomic absorption using a P 
Elmer atomic absorption spectrophoto: 
Model 290 B. 

Statistical analysis. All analyses were 
using the paired / test. Levels of signifi 
were based on the differences between 
of Mg deficient and control rats (13). 

Results. Response to magnesium dep 
Weight gains were similar in magnesiu 
ficient and control rats for the initial 1( 
of Mg depletion. By day 14, howeve 
body weights of the magnesium deficiei 
were significantly below those of thei 
fed controls. By this time plasma Mg cc 
trations were markedly reduced an 
mained at this low level throughout t 
days of depletion (Table I). Plasma Ca 
showed some variability. Mean values 
significantly elevated in the rats Mg de] 
for 14 days in experiment 2, when there 
seven animals per group. No statistical 
nificant differences were seen in rats de] 
for 14 or 28 days in experiment 1 when 
were only three rats per group. The m 
sium content of the tibiae was signifii 
reduced and the calcium content sUghtl 
significantly increased after 28 days 
depletion. (Table II). 

Intestinal Ca transport. Intestinal Ca 
port (% ^'Ca transferred from the lun 
the blood) was consistently less in the 
nesium deficient rats than in their pa 
controls. The difference between th< 
groups was significant after 2 weeks of 

TABLE I. Body Weights and Concentration of Plasma Ca and Mg" in Rats Depleted of Mag nest 

14 AND 28 Days. 

Days of de- 


Experiment 1 

Experiment 2 


Mg depleted 




Body wt (g) 

198 ± 2.9 

192 ± 2.5* 

224 ± 4.7 



Plasma Mg (mg%) 

2.54 ± 0.08 

1.21 ±0.07* 

2.64 ±0.14 

1.48 ± 


Plasma Ca (mg%) 

10.9 ± 0.4 

11.3 ±0.5 

10.5 ± 0.2 







Body wt (g) 

251 ±3.1 

220 ± 3.0* 



Plasma Mg (mg%) 

2.43 ± 0.07 

1.04 ±0.08* 


Plasma Ca (mg%) 


10.4 ± 0.3 



10.4 ± 0.3 




" Values are means ± SEM; figures in parentheses represent number of rats in each group. 

* ' ' Significantly different from control values, P < 0.001, P < 0.005 and P < 0.01 respectively. 

** Plasma calcium was measured only in three rats which were not used for calcium absorption measuremt 



nesium depletion and further increased after 
4 weeks (Fig. 1). Uptake of radioactivity by 
the tibiae generally reflected differences in 
intestinal Ca transport (Fig. 2). Almost com- 
plete transfer of the injected dose had oc- 
curred in the duodenal loop 60 mmutes after 
the dose had been injected; less than 60% had 
been transferred from the ileum in 90 min. 
Unidirectional calcium fluxes at dfiferent 
levels of luminal Ca concentration obtained 
by the in situ perfusion method are shown in 
Fig. 3. Ca influx was consistently less in 
magnesium depleted rats than in their pair 
fed controls at all Ca concentrations. How- 

TABLE II. CoMPOsmoN of the Tibia in Rats 

Magnbsium Depleted for 28 Days afo) Their Pair 

Fed Controls.* 



Magnesium de- 

Water (% 
Mg (meq/tibia) 
Ca (meq/tibia) 
Mg and Ca (meq/g 
dry wt) 

0.55 ± O.OIO 
0.34 ± 0.006 
38.49 ±0.65 
0.098 ± 0.002 
0.30 ± 0.004 
3.394 ± 0.06 
10.15 ±0.13 
10.44 ±0.13 

0.60 ± 0,010* 

0.34 ± 0.004 

42.70 ± 0.58* 

0.047 ±0.001* 

0.14 ±0.002* 

3.60 ± 0.054" 

10.45 ± 0.086^ 

10.58 ± 0.86 

* Values are means ± SEM of 23 rats in each group. 
^^'Significantly different from control values, P < 
0.001, P < 0.01 and i* < 0.05 levels respectively. 

Fig. 1. Effect of Mg deficiency on %^^Ca transferred 
to body with time. Each point represents mean ± SEM 
of Mg deficient rats and pair fed controls, (a) 2 weeks 
Mg depletion, 2-4 rats/time point; (b) 4 weeks depletion, 
^8 rats/time point. •— — # duodenum, control; 

O duodenum, Mg deficient; A A ileum, con- 

irol; A A ileum, Mg deficient. Overall rate of absorp- 

tK)Q was significantly reduced in the duodenum of 14- 
^y depleted rats (P < 0.05). The decrease was not 
significant in the ileum. After 28 days the decrease was 
^gnificant in both segments, P < 0.01 in the duodenum, 
^< 0.001 in the ileum. 

20 40 60 80 20 40 60 

Absorption Period (min) 

Fig. 2. Effect of Mg deficiency on % ^^Ca upuke by 
the whole tibia with time. Each point represents mean 
± SEM of Mg deficient rats and pair fed controls, (a) 2 
weeks Mg depletion, 3-4 rats/time point, (b) 4 weeks Mg 

depletion, 6-8 rats/time point. • • control; 

O O Mg deficient. Overall **Ca uptake was signifi- 
cantly reduced in Mg deficient rats, P < 0.01 after 2 
weeks, P < 0.00 1 after 4 weeks of depletion. 

5 10 IS 

Introlummol Co Concentration (mM) 

Fig. 3. Effect of 14 days of Mg depletion on Ca 
transferred from lumen to blood using the in situ perfu- 
sion technique. Each point represents the mean obtained 
in 4-9 rats (mean values ± SEM and the number of rats 
for each point are shown in Table HI). # # Pair fed 

control, total *'Ca transferred A A Mg deficient, total 

*'Ca transferred Pair fed control, linear portion 

— • • — Mg deficient, linear portion Mg deficient 

and pair fed controls, curvilinear portion with plateau. 

ever, the mean values for the data points used 
to construct Fig. 3, shown in Table III, show 
statistically significant differences only at lu- 
minal Ca concentrations of 0.5, 10 and 20 
mM. The relationship between Ca influx and 
luminal Ca concentration was in agreement 
with the pattern described by Dumont et al 
(13), suggesting that Ca absorption in the rat 
duodenum was comprised of at least two 
components. The curvilinear portion of the 
transport-concentration curve, at the lower 
concentration of calcium, suggests the pres- 



TABLE III. Mean ± SEM of Ca Transferred from 
Lumen to Blood (/iMole/Hr cm"*). 

Ca cone. Pair fed control Mg deficient 

(mA/) N •— • A— A 






0.22 ± 0.02 
0.33 ± 0.03 
0.42 ±0.06 
1.30 ± 0.22 
2.64 ± 0.20 

0.11 ±0.01" 
0.27 ± 0.03 
0.38 ± 0.05 
0.97 ±0.16" 
1.82 ± 0.25* 

'Significantly different from control values (P < 

ence of a saturable, carrier mediated mecha- 
nism and the linear portion, at higher calcium 
concentrations, passive diffusion. With this 
as reference, magnesium depletion appeared 
to depress passive diffusion of Ca across the 
duodenal mucosa with also a significant effect 
at the lowest calcium concentration, 0.5 mM 
(Table III). 

Discussion. In the present investigation the 
depression in Ca transport seen in Mg de- 
pleted rats appeared to be entirely due to a 
decrease in passive diffusion. This finding is 
in conflict with several previous reports which 
suggested either an increase or no change in 
intestinal Ca transport of magnesium-defi- 
cient rats (1-S). Of the two previous studies 
that had shown a decrease in Ca transport (6, 
7), one (7) showed an increase in active Ca 
transport after 10 days of Mg depletion and 
a significant decrease when Mg depletion was 
prolonged for 19 days. The transport data in 
the latter investigation were obtained by an 
in vitro procedure using a modified Ussing 
apparatus (7). Rats fed adequate magnesium 
diets showed comparably decreased rates of 
Ca transport following thyroparathyroidec- 
tomy. The authors suggested that both mag- 
nesium deficiency and thyroparathyroidec- 
tomy depressed Ca transport by alterations in 
vitamin D metabolism, presumably at the 
level of regulation of the hydroxylation of 25- 
OH-D3 to the 1 -hydroxy- or the 24-hydroxy 

The data reported here do not indicate that 
Mg depletion of the magnitude or duration 
applied in this investigation substantially al- 
tered vitamin D metabolism. A significant 
decrease in l,25-(OH)2D3 should have de- 
creased intestinal Ca absorption by both pas- 
sive diffusion and saturable transport. While 
variability may have obscured the signifi- 
cance of differences in Ca transport of defi- 

cient and control rats at the luminal Ca con- 
centrations of 1 and 2 mM (Fig. 3), the overall 
decrease in intestinal Ca transport seen in 
Mg-depleted animals was small. The Mg de- 
ficient diet used in this investigation (SO ppm 
Mg) was chosen to avoid marked differences 
in body weights of Mg-depleted and pair fed 
control rats. Walling et al (7) used a Mg-free 
diet which probably caused acceleration and 
enhancement of Mg-depletion and its mani- 
festations, possibly including disturbances in 
vitamin D metabolism. 

Recent findings in this laboratory ( 14) sug- 
gest an explanation for the data reported here 
which would support the observation of Wall- 
ing et al (7) that parathyroidectomy and Mg 
deficiency had similar effects on intestinal Ca 
transport. Microscopic examination of para- 
thyroid sections removed from Mg deficient 
rats at intervals from 2 to 2 1 days of depletion 
showed progressive manifestations of hy- 
poactivity (14). The same rats consistently 
exhibited hypercalcemia comparable to that 
seen in the present investigation in 14 day 
Mg depleted rats (Table I). Reduction of 
parathyroid hormone activity is an appropri- 
ate response to hypercalcemia. One of the 
consequences of parathyroid hypoactivity 
would be depression in intestinal Ca trans- 

In conclusion, the decreased rate of intes- 
tinal Ca absorption in Mg deficient rats ob- 
served in this investigation appears to be due 
primarily to reduction in the rate of passive 
Ca diffusion. Among several consequences of 
magnesium deficiency likely to depress intes- 
tinal Ca transport is hypoactivity of the para- 
thyroid glands. This aspect of magnesium 
deficiency is now under investigation. 

Summary, Calcium transport across the 
duodenum and ileum was measured by an in 
vivo ligated loop technique in Mg depleted 
rats and rats pair fed a magnesium adequate 
diet. Intestinal Ca transport and tibial ^'Ca 
uptake were consistently decreased in mag- 
nesium depletion. Analysis of Ca fluxes, car- 
ried out by in situ perfusion, showed a signif- 
icant decrease in passive diffusion, with less 
consistent effects on the saturable transport 
component. Both bone and plasma showed 
markedly decreased Mg concentration. Tibia 
Ca levels were slightly but significantly in- 
creased and plasma levels were either normal 



or slightly, but significantly elevated. The 
basis for the decrease in Ca transport of Mg 
depleted rats observed in this investigation is 
not clear. The data suggest a general altera- 
tion in mucosal membrane transport rather 
than a specific effect on Ca transport per se. 

I. Akock, N., and Maclntyre, I., Biochem. J. 7^ 19 

I Alcock, N., and Maclntyre, I.. Clin. Sci. 22, 185 


3. Kessner, D. M., and Epstein, R. H., Proc. Soc. Exp. 
Biol. Med. 122,721(1966). 

4. Morehead, R. M., Jr.. and Kessner, D. M., Amer. J. 
Physiol. 217, 1607 (l%9). 

5. Lifshitz, F., Harrison, H. C, and Harrison, H. E., 
Proc. Soc. Exp. Biol. Med. 125, 19 (1%7). 

6. Krawitt, £. L., Proc. Soc. Exp. Biol. Med. 141, 569 

7. Walling, M. W., Favus, M. J., and Kimberg, D. V., 
Proc. Soc. Exp. Biol. Med. 148, 1038 (1975). 

8. Greger, J. L., and Schwartz, R.. J. Nutr. 104, 1610 

9. Wasserman, R. H., and Taylor, A. N., J. Nutr. 103, 

10. Wasserman, R. J., Kallfelz, F. A., and Comar, C. L., 
Science 133, 883(1%!). 

11. Wasserman, R. H., and Taylor, A. N. in "Mineral 
Meubolism" (C. G. Comar and F. Bronner, eds.). 
Vol. Ill, Chpt. 5, Academic Press, New York (1969). 

12. Snedecor, G. W. and Cochran, W. G., **Sutistical 
Methods,** Sixth Edition, The Iowa Sute University 
Press, Ames, Iowa (1967). 

13. Dumont, P. A., Curran, P. F., and Solomon, A. K., 
J. Gen. Physiol. 43, 631 (1960). 

14. Jones, J. E., Schwartz, R.. and Krook, L. Fed. Proc. 

Received May 31, 1977. P.S.E.B.M. 1978, Vol. 159. 


Anti-ldiotypic Response of BALB/c Mice to a Myeloma Protein of BALB/c Origi 


Department of Microbiology, Faculty of Science, Mahidol University, Bangkok, Thailand 

It has been demonstrated that a BALB/c 
myeloma protein with anti-DNP activity 
(Ftotein-31S) can stimulate anti-idiotypic re- 
sponse in several strains of inbred mice, in- 
cluding the strain from which the plasmacy- 
toma MOPC-31S was originally induced (1, 
2). Antibodies produced were found to be 
specific for the antigen-binding site of Pro- 
tein-31S (1-3). Tungkanak and Sirisinha (3) 
also reported that the Fc fragment of Protein- 
315 was not required for the induction of 
anti-idiotypic response in BALB/c mice. The 
anti-idiotypic antibody produced in response 
to stimulation by the Fab-fragment of Pro- 
tein-31S was indistinguishable from that pro- 
duced in response to undigested protein (3). 
The purpose of the present study was to 
follow the development of an anti-idiotypic 
antibody response of BALB/c mice to Pro- 
tein-31S, particularly with regard to the abil- 
ity of these anti-idiotypic antibodies to com- 
pete with the hapten for an antigen-binding 
site on Protein-31S. The results showed that 
the susceptibiUty of anti-idiotypic antibody to 
inhibition by excess hapten (DNP-lysine) de- 
pends largely on the immunization procedure 
used, i.e., the anti-idiotypic antibodies pro- 
duced following a single booster injection 
showed a marked increase in the ability to 
compete with DNP-lysine for the antigen- 
binding site of Protein-315. Evidence availa- 
ble suggests that this change was associated 
with an increase in the affinity of the anti- 
idiotypic antibody produced after a booster 

Materials and methods. Antigens. Protein- 
315 and its peptic product (Fv-315) were 
prepared and purified as described previously 

* Supported in part by the Rockefeller Foundation. 

^ A portion of this study was submitted by R. T. to 
the Faculty of Graduate Studies, Mahidol university, in 
partial fulfillment of the requirements for the Degree of 
Doctor of Philosophy. 

^To whom all correspondence and reprint requests 
should be made. 

OO37-9727/78/1592-01 76$Ol.0O/0 
<^offyr:ifhtCJ978 by ate Society for ExperimeatalBkjk^ 

(3). Myeloma sera from BALB/c mice < 
ing MOPC-315, MOPC-460, MOPC 
Adj.PC-22A, J504, and S176 tumors 
kindly provided by Dr. Herman N. 
(Massachusetts Institute of Technology 
ton, MA). 

Immunization schedule. BALB/c mi 
both sexes used in this study were orig 
obtained from Jackson Laboratory, Ba 
bor, Maine. Adult mice were immunizec 
time with 2(X) /ig of purified Protein-3] 
tributed at the same two front footpac 
four other sites along the back. The pr 
course of immunization consisted of 3 v 
injections of immunogen in con 
Freund's adjuvant, in incomplete Fn 
adjuvant, and in potassium phos] 
buffered saline (PBS) pH 7.2, respec 
The animals were bled from orbital v 
plexus one week after the third injectic 
at weekly intervals thereafter. A 
booster injection of 2(X) /ig of immuno; 
PBS was given 1 week after the mic 
received a full course of primary imm 
tion had been bled 4 times. These ai 
were bled again during the 4 suca 
weeks. A similar second booster injectic 
given to some of these mice and Uie ai 
were thereafter bled as described. Indi 
sera from the same group (5-10 mi< 
group) were pooled and kept frozen 

Analysis of anti-idiotypic antibody. Ad 
otypic antibody to Protein-315 was 
mined by radioimmunoassay using ^ 
beled Protein-315 or Fv-315 as antigc 
Pooled sera obtained at weekly interval 
analyzed for their antigen-binding ca| 
susceptibility to inhibition by excess h 
and cross-reactivity with five other my 
A sera of BALB/c origin exactly as dts> 
by Tungkanak and Sirisinha (3). 

Results. The antigen-binding capac 
BALB/c antisera, as determined by thei 
ity to react with ^I-labeled antigen (Pi 



315 or Fv-31S), could be detected as early as 
1 week after completion of the primary course 
of immunization with Protein-31S (Fig. 1). 
Neither the antigen-binding capacity nor the 
sensitivity to inhibition by excess hapten al- 
tered much during the 13 weeks of observa- 
tion period. There was also no demonstrable 
change in specificity as the hapten inhibition 
values obtained when either Protein-31S or 
Fv-31S was used as antigen in the assay sys- 
tem were similar (Fig. 1). 

The antigen-binding capacity of these an- 
tisera was enhanced following a single 
booster injection with 200 jLig of Protein-315. 
As shown in Fig. 2, the quantity of labeled 
antigen precipitated by 10 /xl of antiserum 
increased from less than 40% to more than 
60% one week after boosting. It is more inter- 
esting however to find that the ability of these 
post-boosting sera to compete with excess 
hapten for the antigen-binding site of Protein- 
315 increased markedly, i.e., within one week 
the hapten inhibition value decreased from 
more than 80% to less than 10%, regardless of 
the type of antigen used in the assay system. 
Although the susceptibility to inhibition by 
hapten gradually increased during the next 
few weeks, the inhibition value did not quite 
return to the pre-boosting level. Similar but 

8 10 12 


Fig. 1. Antigen-binding capacity and susceptibility 
to inhibition by excess hapten (750 nanomoles of DNP- 
iysine) of BALB/c antiserum to Protein-315 from the 
non-boosting group. Bleeding time represents time after 
the last injection of the primary course of immunization. 
Both Protein-315 and Fv 315 were used as antigens in 
the assay system. 

less obvious changes were observed following 
a second booster injection. 

Selected samples of the pre-boosting (week 
4) and post-boosting (weeks 7, 9 and 12) 
antisera were diluted with pooled normal 
BALB/c senmi and then retested for their 
susceptibility to inhibition by excess hapten. 
The results showed that the hapten inhibition 
values gradually increased as the antisera 
were being diluted (Table I). The effect of 
dilution on the hapten inhibition value was 
independent of the type of the antigen used 
in the assay system. 

Despite a marked change in sensitivity to 
inhibition by hapten of the antisera obtained 

8 10 


Fig. 2. Antigen-binding capacity and susceptibility 
to inhibition by excess hapten (750 nanomoles of DNP- 
lysine) of BALB/c antiserum to Protein-315 from the 
boosting group. The animals were boosted at weeks 5 
and 10 (arrows). See legend to Fig. 1 for other explana- 

TABLE I. Effect of Antiserum Dilution on 
Hapten Inhibition* 



(%) Maximum hapten inhibi- 
tion (by 750 nmoles DNP-lys) 


lute 1:5* 1:10 1:20 1:40 






77 87 92 84 90 

3 53 65 68 68 

36 77 81 82 82 

12 55 69 73 73 

" Excess hapten (750 nanomoles) was mixed with 0.5 
/ig of '^Mabeled Protein-315 before 10 /d of antiserum 
(or its dilution) was added. Thereafter the reaction mix- 
ture was treated as described in Materials and methods. 

* Diluted with pooled normal BALB/c serum. 



after boosting, the specificity of these antisera 
remained unchanged. This was evident from 
the results of a cross-reactivity study using 
five other myeloma A sera to inhibit the 
idiotypic reaction. Like the results obtained 
with the pre-boosting antisera, the myeloma 
sera from mice carrying MOPC-460, Adj.PC- 
22 A, and SI 76 tumors failed to inhibit the 
anti-idiotypic activity of these postboosting 
antisera while those from MOPC-292 and 
JS04 mice demonstrated slight inhibition (less 
than 20%). 

Discussion. The present observations con- 
firm and extend the original report of Siri- 
sinha and Eisen (1) that under appropriate 
conditions anti-idiotypic antibody response 
to a BALB/c myeloma protein with anti- 
DNP activity can be induced in BALB/c 
mice. The anti-idiotypic antibody produced 
is directed largely, if not exclusively, to the 
antigen-binding site of Protein-31S, as evi- 
dent from the observations that the antisera 
were highly sensitive to inhibition by excess 
hapten and they cross-reached only slightly, 
if any, with five other myeloma A proteins 
available for testing. 

The interesting feature of the anti-idiotypic 
response is that a single booster injection not 
only increased the total antigen-binding ca- 
pacity of these sera but also markedly de- 
creased their susceptibility to inhibition by 
excess hapten (Fig. 2). The results obtained 
following a booster injection are markedly 
different from those obtained after the pri- 
mary course of immunization. In the non- 
boosting group, the antisera obtained at 
weekly intervals throughout the 13 weeks of 
observation were equally sensitive to inhibi- 
tion by hapten and their antigen-binding ca- 
pacity decreased only slightly during this pe- 
riod (Fig. 1). 

The change of the hapten inhibition value 
obtained after boosting was much larger than 
can be explained on the basis of a quantitative 
increase of antibody production by these an- 
imals. The reduction of the hapten inhibition 
value must therefore be attributable to 
changes in other parameters, e.g., affinity and 
specificity. Athough an increase in affinity of 
the anti-idiotype produced after a single 
booster injection is consistent with the general 
characteristic of a secondary antibody re- 
sponse (S), the possibility that this could also 

be associated with a shift in specificity a 
be completely ruled out. Circumstantia 
dence, however, supports the possibility 
a marked reduction in the sensitivity to 
ten inhibition of these postboosting s< 
more likely associated with an increa 
affinity of antibody. Firstly, the hapten 
bition value of the postboosting ser 
creased when they were diluted prior tc 
ing (Table 1). This interpretation is cons 
with the explanation of Sher and Coli 
who employed the phosphorylcholine s] 
in their study. Secondly, both the bii 
capacity and hapten inhibition were si 
when either Protein-315 or Fv-315 wa 
ployed as antigen in the assay system 
2), suggesting that the reaction is prin 
restricted to the Fv region. Lastly, the 
terns of cross-reactivity of the pre-bo< 
and the post-boosting antisera with 
myeloma A proteins were indistinguis 
from one another (unpublished observa 
The possibility that the observed decree 
susceptibility to inhibition by hapten 
laboratory artifact associated with the 
system employed is unlikely as there ^ 
gradual return of these values towan 
preboosting level within a few weeks a 
booster injection. Likewise, the possi 
that there was insufficient labeled antig 
the test system to react with the anti 
produced after boostering is also un 
because under the condition used fo 
assay of antibody, there was excess ar 
left in the supernatant fluid. In additi 
this evidence, we may add that a \x 
injection of antigen under identical con< 
to other strains of mice (C57BL/6J and 
failed to cause any reduction of the h 
inhibition value (unpublished observat 
It appears from these observations thei 
that the insusceptibility to inhibition b 
cess hapten of the anti-idiotype proc 
following a booster injection is more 
associated with an increase in affinity i 
than a shift in specificity of these antib< 
Summary. The anti-idiotypic respon 
BALB/c mice to myeloma protei 
BALB/c origin (purified Protein-315 ft 
plasmacytoma MOPC-315) was analyze 
Its antigen-binding capacity and suscef 
ity to inhibition by excess hapten (DN 
sine). The results showed that the ani 



otypic antibody that is sensitive to inhibition 
by hapten could be detected for at least 3 
I months after completion of the primary 
course of immunization. Following a single 
booster injection, there was an increase of the 
antigen-binding capacity and the susceptibil- 
ity of these post-boosting antisera to inhibi- 
tion by hapten was markedly reduced (from 
more than 80% to less than 10% under the 
assay system employed). However, the hap- 
ten inhibition value gradually returned to- 
ward the preboosting level within a few 
weeks. The data obtained suggest that the 
change in the hapten inhibition value after 
boosting is associated with increased affmity 
rather than a shift in specificity. 

myeloma sera used in this study and for the valuable 
suggestions and criticisms during the course of this study. 
Encouragement from Dr. P. Matangkasombut is also 
greatly appreciated. 

1. Sirisinha, S., and Eisen, H. N., Proc. Nat. Acad. Sci. 

2. Granato, D., Braun, D. G., and Vassalli, P., J. Im- 
munol. 113,417 (1974). 

3. Tungkanak, R., and Sirisinha. S. J. Immunol. 117, 

4. Lieberman, R.. and Potter, M., J. Mol. Biol. 18, 516 

5. Eisen, H. N., and Siskind» G. W., Biochemistry, 3, 

6. Sher, A., and Cohn, M., J. Immunol. 109, 176 ( 1972). 

7. Herzenberg. L. A., McDevitt, H. O., and Herzenberg, 
L. A., Ann. Rev. Genetics 2, 209 ( l%8). 

The authors are grateful to Dr. H. N. Eiscn for the Received March 27, 1978. P.S.E.B.M. 1978, Vol. 159. 


Effects of Indomethacin and Tolmetin on Furosemide-lnduced Changes in Renin 

Release' (40309) 


Departments of Pharmacology, Physiology, and Human Development, Michigan State University, East Lansing, 

Michigan 4S824 

The diuretic furosemide increases renin 
release by the kidney, an effect independent 
of volume depletion which accompanies di- 
uresis (1). The stimulus for renin release by 
furosemide appears to be related to both 
changes in renal arteriolar resistance (2, 3) 
and a direct tubular efTect subsequent to 
blockade of sodium and chloride reabsorp- 
tion prior to or at the macula densa (1,2). 

Furosemide-induced renin release is 
blocked by the prostaglandin synthetase in- 
hibitor, indomethacin, by an undefined 
mechanism (4). After indomethacin, the abil- 
ity of furosemide to increase renal blood flow 
is blunted, while the natriuretic efTect is un- 
affected (4-6). 

Calcium has recently been suggested to 
play a role in renin secretion (7, 8). Although 
furosemide is acutely calciuretic, the signifi- 
cance of this effect with respect to renin 
release has not been evaluated. The purpose 
of these experiments was to determine if the 
blockade of furosemide induced renin release 
by prostaglandin synthetase inhibitors, indo- 
methacin and tolmetin, was correlated with 
changes in the calciuretic response to furo- 

Materials and methods. Surgical. Male 
mongrel dogs, 15-25 kg, were used in all 
experiments. The animals were anesthetized 
with sodium pentobarbital (30 mg/kg), intra- 
venously, and a cuffed endotracheal tube was 
inserted. The dogs were artificially ventilated 
with a Harvard respirator. Catheters were 
placed in the left femoral artery and in both 
femoral veins. Normal saline (0.9% NaCl) 
was infused into one femoral vein to replace 
fluid losses from surgery and to hydrate the 
animal until total urine flow was 0.5-2.0 

' This study was supported by USPHS Grant No. 

^ Dr. Noordcwicr was supported by a fellowship from 
the Michigan Kidney Foundation. 

'lyright (D 1978 by the Society for ExperimctitMi Biology and Medkane 
ifhtg reterved. 

ml/min. The saline infusion was then re- 
duced to equal urine flow. Inulin was infused 
into the other femoral vein at a rate calculated 
to maintain plasma inulin concentration be- 
tween 30-50 mg/dl. Arterial blood pressure 
was monitored with a Statham P23AC trans- 

Experimental protocols. I. Effect of indo- 
methacin on renal responses to intravenous 
furosemide. Glomerular filtration rate esti- 
mated by the clearance of inulin (Cin), uri- 
nary excretion of Na, K, and Ca and plasma 
renin concentration (PRC) were measured 
during two control 10-min clearance periods 
and during the intravenous infusion of furo- 
semide (2 mg/kg/hr). Following the furose- 
mide clearance periods, each dog received 
increasing doses of indomethacin (0.01, 0.05, 
0.1, 0.5 mg/kg, iv). Furosemide infiision con- 
tinued during the administration of indo- 
methacin. Twenty minutes after each dose of 
indomethacin, two clearance periods were 
obtained. In addition to the dogs treated with 
indomethacin, three dogs were injected with 
saline instead of indomethacin in an experi- 
mental protocol identical to that described 
above (4 injections at 40 min intervals). These 
dogs are referred to as "time" control dogs. 
II. Effect of indomethacin or tolmetin pre- 
treatment on renal response to intrarenal fu- 
rosemide. After two control clearance pe- 
riods, furosemide was infused (15 jLigAg/min) 
into the renal artery of the experimental kid- 
ney, and two clearance periods obtained. The 
infusion of furosemide was stopped and 30 
minutes were allowed for urine flow to return 
toward control. Two additional control pe- 
riods were then run and tolmetin (5 mg/kg) 
or indomethacin (2 mg/kg) was administered 
intravenously. After 20 min, two more clear- 
ances were taken and furosemide again in- 
fused intrarenally. Two additional clearance 
periods were obtained during furosemide in- 
fusion. Excretion of Na, K, and Ca, as well 



f, and PRC were measured during each 
nee period. The dose of furosemide in 
3f the tolmetin treated dogs was S 
/min but since the results did not differ 
the results from the two dogs given IS 
/min the data were pooled. 
ilytical Urinary and plasma Na and K 
determined by flame photometry and 

atomic absorption spectroscopy. Inulin 
ntrations were determined by the 
)d of Walser et al (9). Plasma renin 
ntration was estimated by incubating 
a with excess homologous renin sub- 
, The amount of angiotensin I generated 
hen determined by radioimmunoassay 
rhe data were analyzed utilizing anal- 
\f variance with a randomized block 
I. The 0.05 level of probabiUty was used 

criterion of significance. 
Jilts, Furosemide infused intravenously 
/kgAr) increased the urinary excretion 
lium and calcium (Table I). Plasma 

concentration (PRC) was also in- 
d. All values remained elevated 
^out drug administration (Table I), 
erular filtration rate was not affected by 
;mide. Increasing doses of indometha- 
oduced dose related decreases in PRC 
alcium excretion (Table II). Sodium 

excretion and GFR were not changed when 
indomethacin was given during furosemide 
infusion (Table II). Potassium excretion (not 
shown) also increased after furosemide and 
was not affected by indomethacin. 

Furosemide, infiised into the renal artery, 
also increased urinary excretion of sodium, 
potassium and calcium (Figs. 1 and 2). PRC 
also increased when furosemide was given. 
Electrolyte excretions and PRC return^ to- 
ward control when the infusion of furosemide 
was stopped. Indomethacin (Fig. 1) and tol- 
metin (Fig. 2) had little effect on electrolyte 
excretion although each parameter tended to 
be lower than the previous control. Similarly, 
PRC tended to decrease after indomethacin 
or tolmetin. A second infusion of furosemide 
increased sodium, potassium, and calcium 
excretion but PRC was not afifected by furo- 
semide after administration of indomethacin 
(Fig. 1) or tolmetin (Fig. 2). 

Discussion. Although the role of calcium in 
renin release is still obscure, there is increas- 
ing evidence that movement of this ion within 
the juxtaglomerular cell may be an important 
regulatory mechanism. Addition of calcium 
to kidney slices incubated in calcium free 
media produces an immediate, large increase 
in renin release (7). Similarly, the isolated 

TABLE I. Effect of Time on Furosemide-induced Changes in Renal Functign 

Saline dose 


ml/min) S£ 39.9 ± 12.0 
ng Al/ml/hr) 14.9 ± 3.2 

Control Furosemide" 

/lEq/min) SB 
fiEq/min) SE 

152 ± 32 
0.61 ± 0.07 

29.3 ± 9.7 
38.9* ± 13.1 

538* ±61 
11.2*± 1.9 

31.6 ± 10.0 
32.2* ± 13.6 

649* ± 47 
15.1* ±3.6 

32.7 ± 10.8 
30.0* ± 6.6 

848* ±92 
12.6* ± 1.6 

31.6 ±9.4 
30.4* ±7.1 

924* ± 150 
10.9* ± 0.9 

35.1 ±9.5 
29.7* ±3.0 

7 17* ±36 
11.9* ±3.2 

rosemide was infused at a rate of 2 mg/kg/hr, iv. 
^nificantly different than control ( P < .05). 

ILE II. Effect gf Increasing Doses of Indomethacin on Furosemide-induced Changes in Renal 


Indomethacin dose (mgAg) 








S Al/ml/hr) 

Eq/min) SE 
Eq/min) SE 

43 ±4 
16.3' ±7.1 

130^ ± 33 

39 ±4 
40.1* ±9.6 

844* ±96 
17.6* ±3.2 

37 ±5 
27.6*'^ ± 6.8 

779* ± 56 
13.4* ± 1.9 

36 ±6 

17.4' ±4.7 

876* ± 20 
14.1* ±2.8 

38 ±7 
14.5' ± 4.5 

34 ±5 
13.6' ± 5.5 

854* ±22.1 656* ±59 
10.8*' ± 2.0 9.8*' ± 2.2 

osemide was infused at a rate of 2 mg/kg/hr, iv. 
lificantly different than control (P< .05). 
lificantly different than furosemide {P< .05). 



^ so 

Gm Fure Con 

Fig. 1. Effect of indomethacin on renal response to 
furosemide. Excretion rates of sodium (UNaV), potassium 
(UkV), and calcium (Uc«V) and plasma renin concentra- 
tion were measured. After a control period (CON), fu- 
rosemide was infused into the renal artery (FURO). 
After the furosemide infusion was stopped, control val- 
ues were measured (CON), and indomethacin was given 
iv (INDO). Forty minutes after indomethacin, furosem- 
ide was infused again (INDO + FURO). The mean and 
1 SE arc given (AT - 4). • different from previous 
clearance period (P< .05). 

perfused kidney of the cat releases renin in 
response to calcium only after prior exposure 
to calcium free perfusate (8). These data in- 
dicate that an increase in intracellular free 
calcium may be involved in renin release. 

The present experiments demonstrate that 
blockade of furosemide-induced renin release 
by indomethacin or tolmetin does not depend 
on alterations in net tubular transport of so- 
dium, potassium or calcium. The major stim- 
ulus for renin release during furosemide ad- 
ministration appears to be inhibition of so- 
dium (or chloride) flux at the macula densa 
similar to that observed in the cells of the 
thick ascending limb of the loop of Henle ( 1 ). 
Since the prostaglandin synthetase inhibitors 
failed to alter the urinary excretion of sodium 
in this study or in previous work (4), it is 
unlikely that the effect of indomethacin or 
tolmetin on renin release could involve 


Fig. 2. Effect of tolmetin on renal response to furo- 
semide. Excretion rates of sodium (Un«V), potassium 
(UkV) and calcium (UcaV) and plasma renin concentra- 
tion (PRC) were measured. After a control period 
(CON), furosemide was infused into the renal artery 
(FURO). After the furosemide infusion was stopped, 
control values were measured (CON), and tolmetin was 
given iv (TOL). Forty minutes after tolmetin, furosemide 
was infused again (TOL + FURO). The mean and 1 SE 
are given (AT - 4). • different from previous clearance 
periods (P<. 05). 

changes in sodium transport at the macula 

Similarly, alterations in calcium load to the 
macula densa do not appear to be important 
in the action of tolmetin or indomethacin. 
Although there was a small dose related de- 
crease in calcium excretion after indometha- 
cin, calcium excretion rate was well above 
control even after the highest dose of indo- 
methacin tested (Table I). In contrast, PRC 
had decreased dramatically. In addition, pre- 
treatment with neither indomethacin nor tol- 
metin altered the increase in calcium excre- 
tion to intrarenal furosemide, while both 
drugs blocked any increase in PRC (Figs. 1 
and 2). Thus, blockade of ftirosemide-in- 
duced renin release by prostaglandin synthe- 
tase inhibitors does not require an alteration 
in the calciuretic effect of furosemide. Lester 
and Rubin also found extracellular calcium 



was not a determinant in the release of renin 
following furosemide (8). Since prostaglandin 
synthetase inhibitors, such as indomethacin 
or tolmetin, do not appear to affect sodium 
or calcium load at the macula densa, their 
site of action is probably subsequent to the 
signal perceived by the macula densa. 
Whether their action involves alterations in 
the state of intracellular calcium remains to 
be investigated. 

Summary. Prostaglandin synthetase inhib- 
itors, indomethacin and tolmetin, blocked fu- 
rosemide-induced increase in renin secretion 
whether the furosemide was given intrave- 
nously or into the renal artery. Tolmetin and 
indomethacin did not affect the natriuretic, 
kaliuretic or calciuretic response to furose- 
mide. Therefore, blockade of furosemide-in- 
duced renin release does not appear to require 
an alteration in sodium or calcium load at the 
macula densa. Thus, the site of action of 
prostaglandin synthetase inhibitors on renin 
release is probably subsequent to the signal 
perceived by the macula densa. 

Laboratories, Inc., Fort Washington, Pennsylvania for 
supplying the tolmetin. The authors also acknowledge 
the technical assistance of Mrs. Peggy Wagner, Mr. Keith 
Crosslan, and Mr. Terry Steele. 

1. Vander, A. J., and Carlson, J., Circ. Res. 25, 145 

2. Corsini, W. A., Hook, J. B., and Bailie, M. D., Circ. 
Res. 37, 464 (1975). 

3. Osbom, J. L., Hook, J. B., and Bailie, M. D., Circ. 
Res. 41, 481 (1977). 

4. BaiUe, M. D., Crosslan, K., and Hook, J. B., J. 
Pharmacol. Exp. Ther. 199, 469 (1976). 

5. Williamson, H. E., Bourland, W. A., and Marchand, 
G. R., Proc. Soc. Exp. Biol. Med. 148, 164 (1975). 

6. Bailie, M. D., Barbour, J. A., and Hook. J. B., Proc. 
Soc. Exp. Biol. Med. 148, 1173 (1975). 

7. Chcm, D. S., and Poisner, A. M., Proc. Soc. Exp. 
Biol. Med. 152,565(1976). 

8. Lester, G. E., and Rubin, R. P., J. Physiol. 269, 93 

9. Walser, M., Davidson, D. G., and OrlofT, J., J. Clin. 
Invest. 34, 1520(1955). 

10. Haber, E., Koemer. T., Page, L. B.. Kliman, B. and 
Pumodc, A., J. Clin. Endocrinol 29, 1349 (1969). 

The authors wish to Uiank Dr. R. Z. Gussin of McNeil Received December 13, 1977. P.S.E.B.M. 1978, Vol. 159. 


Absence of Cytotoxic Effect of Selected Pathogens on HLA B27 Positive Fibrol 


Woiisworth Veterans Administration Hospital, Los Angeles, Calfomia 90073 

There is strong suggestive evidence that at 
least one of the seronegative spondyloarthro- 
pathies, Reiter's disease (RD), follows genital 
exposure to certain infectious agents includ- 
ing Chlamydia trachomatis and Ureaplasma 
urealyticum (1). In addition, postdysenteric 
RD and similar forms of acute reactive ar- 
thritis are known to follow enteric infections 
with Enterobacteriaceae such as Salmonella 
and Yersinia (2). Eventually, some patients 
with these acute post-infective arthropathies 
may develop chronic sequelae identical to 
those seen in ankylosing spondylitis (AS) (1). 

It is now firmly established that Uiese 
same seronegative spondyloarthropathies are 
strongly associated with the B-locus histo- 
compatibility antigen HLA B27 (3). Thus it 
appears that exposure to certain specific mi- 
crobial agents in a genetically susceptible host 
may be a prerequisite for the development of 
this spectrum of acute-to-chronic rheumatic 

Several possibilities emerge from this con- 
cept. Firstly, the B27 antigen present on cell 
surfaces might facilitate microbial attachment 
and invasion. Secondly, the cell surface anti- 
gen B27 might share antigenic similarities 
with the microbial agents initiating these dis- 
eases. Under this circumstance the body's 
host defense mechanism may not recognize 
the agents as foreign and antigenic. Thirdly, 
the cbo-omosomal locus which codes for HLA 
B27 is located within the major histocompat- 
ibility complex (MHC) region of the sixth 
autosomal chromosome. There is strong evi- 
dence in other species and suggestive evi- 
dence in man that immune response genes 
are also located within the same genetic com- 
plex, and that such genes may be linked to 
the HLA loci (4). Conceivably, immune re- 
sponsiveness controlled by HLA-linked genes 
may be responsible for the development 
and/or propagation of connective tissue in- 
flammation typified as RD and AS. 

A first step in elucidating the potential role 

of MHC gene products in RD and i 
explore their influence on cell surface 
tibility to implicated pathogens. Usi 
diocytotoxidty assay previously stand 
by cell counts and correlated to dye ex 
cytotoxicity of implicated pathogens 
man cells was investigated. The expc 
reported here indicate that HLA B^ 
no such direct role in initiating th 
lesion of the seronegative spondyloai 

Material and methods. Target cells, 
skin fibroblasts were cultured from 4 
depth punch biopsies of normal and 
tient volunteers with RD or AS who h 
typed for the absence or presence < 
B27. It has been demonstrated that fib 
of B27-positive individuals retain the 
surface markers for at least 12 weeks 
(S, 6). Explants ^mm^, devoid of ep 
and subcutaneous fat, were secured ir 
culture flasks (Falcon, Oxnard, CA) 
face tension of the culture medium. ( 
were established in Eagle's BME 
salts) (Gibco, Grand Island, NY) 
mented with penicillin 1(X) units/ml, 
mycin sulfate 1 1 /xg/ml, L-glutamine 
and 15% unheated fetal calf serum 
(Gibco, Grand Island, NY), and kcp 
CO2 at 37^. Medium was changed we 
2 weeks, twice weekly thereafter. A 
first change, all further culture was 
absence of antibiotics. At 4-S weeks, t 
fluent fibroblast monolayers were try] 
(0.25% in Hanks (HBSS; Gibco, Gi 
land, NY), pH 8.2, 5' at 37**) and subc 
in 1:2 splits. After four subcultures, t 
were assayed for bacterial and myco 
contamination. Fibroblasts for cytc 
targets were harvested at late log 
phase and used only within Sth-20tl: 

Pathogens. The following pathog< 
tained from sources indicated, were < 
by standard methods. At least three 



right e 1978 by the Society fix ExperimenuJ Biology and Medidne 



tures preceded the final effector organisms 
used to assay cytotoxicity. Identity of later 
subcultures was reconfirmed by source labs. 

are expressed as fractional ^^Cr release. In 
Fig. 2, the *^Cr release effected by dH20 is 
illustrated as specific cytotoxicity by relating 

Yersinia enierocoliiica 
Salmonella minnesota 
Herpes simplex virus 
Vreaplasma urealyticum 

Mycoplasma hominis 

Chlamydia trachomatis 
Chlamydia irachomaiis 







UW-3 (type D) 
UW-5 (type E) 

Dr. R. Weaver, Center for Disease Control, Atlanta, GA. 

Dr. G. Kalmansen, Wadsworth VA, Los Angeles. 

Dr. M. Fiala, Harbor General Hospital, Los Angeles; 

Dept. of Infectious Diseases. 
National Institute of Allergy and Infectious Diseases, Be- 

thesda, MD. 
Dr. M. Shepard, Naval Medical Field Research Lab, 

Camp Lejeune, N.C. 
Dr. S. p. Wang, Univ. of Washington; Dept. of Pathobiol- 


Cytotoxicity assay. Fibroblast cultures were 
trypsinized into fresh culture medium and 
adjusted to 1.0 x 10^ cells per ml. 1 /xCi/ml 
Na2"Cr04 (ICN, Irvine, CA) was added and 
the suspension distributed in 1 ml aliquots to 
sterile flat-bottomed glass tubes (Cal Gass 15- 
lOS) which could sul^equently be inserted in 
a well-type Nuclear Chicago y-counter. Gas 
phase of each tube was equilibrated with 5% 
CO2 in air, and the tubes were capped and 
incubated at 37^. At 16 hr the adherent fibro- 
blast monolayers were washed with 1 ml/tube 
HBSS containing 10% FCS. After 1 more hr 
the wash was repeated, and the monolayers 
were covered with 1 ml of fresh culture me- 
dium. 0. 10 ml of selected log-phase pathogen, 
adjusted to proper multiplicity of infection 
(MOI), was added to each tube. Positive con- 
trols received 6N HCl; negative controls re- 
ceived medium only. At intervals after the 
addition of pathogens, supematants were 
transferred to separate y-counter tubes, the 
remaining monolayers were gently rinsed 
with 5% FCS in HBSS, and the rinses were 
pooled with supematants. Tubes with media 
and tubes with cells were counted for y-emis- 
sion. Fractional *^Cr release (FR) for each 
cell-medium pair was expressed as 

^^(medium)A** VUs) + ''^''(medium) 

and specific cytotoxicity of each pathogen as 

"^(pathogen) - 

Figures 1 and 2 illustrate the course of a 
prototype assay, used to determine suitable 
levels of target cell label. In Fig. 1, sponta- 
neous, intermediate, and maximum ^^Cr 
release are achieved in culture medium, 70% 
distilled water (dH20), and 0.1 M sodium 
dodecyl sulfate (SDS) respectively. Results 

it to a baseline represented by spontaneous 

Results, Table I shows the mean fractional 
^^Cr release of three B27-negative and three 
B27-positive fibroblast strains exposed to 
each pathogen. In Table II the specific cyto- 
toxicity of each pathogen on these target fi- 
broblast strains has been calculated. Over the 
range of dilutions used (10\ 10^ and 10"^ 
MOI), paired / tests revealed no significant 
difference in the cytotoxic effect of any one 
pathogen on B27-positive compared to B27- 
negative fibroblasts with the apparent excep- 
tion of Ureaplasma urealyticum. However, the 
differential killing for this organism is almost 
certainly not significant since it only repre- 

♦_/< — 01 M SOS 

70% dHjO 

12 24 

HoJS Afwr miury 

Fig. 1. Fractional **Cr release. Cytotoxic efTect of 
70% dH20 on normal human skin fibroblasts. Data 
points represent mean ± SEM of three determinations. 

^ 100 


o SO 

► 7D\ dHgO 


Hours Affcr Injury 

Fig. 2. Specific cytotoxicity. Results of Fig. 1 ad- 
justed to spontaneous release baseline. Data points rep- 
resent FR (X) ' FR (Spontaneous)/FR (Maximum). 


TABLE I. Mean Fractional '*Cr Release of B27- and B27 Fibroblasts in Presence of Patho 

Cytotoxicity " 
read tl boon 

Mean fractioiul ' 

"Cr release ± SEM" 

10* MO! 


lO* MO 

ViaUc orunisin 






Y. enterocoiiiica type 8 
S. minnesoU S9S 


.626 ± .057 
.929 ± .017 
.177 ±.030 

.683 ± .074 
.876 ± .004 
.180 ±059 

.444 ± .062 
.913 ± .013 
.166 ±0.29 

.578 ± .092 
.891 ± .016 
.119 ±018 

.411 ±072 

.912 ± .022 .1 

.194 ±026 



.162 ±033 

.154 ±036 

.194 ±045 

.110 ±005 

.234 ± .060 

U. urealyticuin T-960 


.213 ± .063 

.164 ±008 

.2 19 ±.080 

.172 ± .009 

.211 ±027 

M. hoininU#IOOI 

C. irachomatis UW-3 (D) 

C. trachomatis UW-5 (E) 


.245 ± .081 
.245 ± .024 
.242 ± .030 

.170 ±004 
.217 ±.013 
.246 ± .027 

.183 ±024 
.246 ± .024 
.246 ± .039 

.187 ±025 
.244 ± .01 1 
.214 ± .008 

.169 ±019 
.255 ± .029 
254 ± .036 

" Three experiments were performed. 

^ One experiment performed, due to technical difTtculties. 

TABLE II. Specific Cytotoxicity of Pathogens on B27- and B27 Fibroblasts. Significanc 

read at 

Specific cytotoxicity 






Viable organism 








Paired T 

Y. enterocolitica type 8 









S. minnesoU 59S 









CMV AD- 169 









HSV-2 #1000 









U. urealyticum T-690 









M. hoininis#1001 









C. trachomatis UW-3 (D) 









C. trachomatis UW-S (E) 









sents a difference between 1 and 2% nonspe- 
cific versus 3 and 4% specific cytotoxicity. 
The / value is large because the two ranges 
were very small and did not overlap. The 
different pathogens damaged targets to dif- 
ferent extents, but none truly differentiated 
between B27-negative and B27-positive fibro- 

Discussion. Over 90% of patients with RD 
or AS bear B27 on their cell surface. Con- 
ceivably, the presence of this membrane poly- 
peptide might render the cells more vulnera- 
ble to direct attack by certain pathogens, 
notably those temporally incriminated with 
patients with RD and other post-infective 
arthropathies. Such susceptibility would per- 
mit rapid attachment, invasion, and cell de- 
struction before host immune recognition and 
defense were fully mobilized. However, our 
study shows that B27-positive skin fibroblasts 
do not differ from B27-negative skin fibro- 
blasts with respect to their susceptibility to 
damage by selected pathogens. 

Other relevant organisms assayed in other 
in vitro systems may possibly demonstrate an 
enhanced and differential cytotoxicity toward 

B27-positive target cells, but this w< 
pear to be unlikely. 

Summary. A sensitive index of in 
damage has been used to investigate 
sibility that HLA B27-positive fibrot 
peculiarly susceptible to those ii 
agents incriminated in the seronegati 
dyloarthropathies. No evidence for < 
tial susceptibility related to the pre 
absence of the B27 antigen could be 

We thank Dr. P. I. Terasaki for HLA typi 
donors, Linda Prince for secretarial assistan« 
Goldberg for preparation of the manuscript. 

1. Bluestone, R., and Pearson, C. M., Adv. 
22, 1 (1977). 

2. Aho, K., Ahvonen, P., Lassus, A., Sievei 
Tiilikainen, A., Arthr. Rheum. 17, 521 (T 

3. Bluestone, R., Hosp. Pract. 10, No. 4, 131 

4. McDevitt, H. O., and Benacerraf. B., Adv. 

5. Miggiano, V. C, Nabhold, M., and Bodn 
in ''Histocompatibility Testing*' (P. L Ter. 
p. 623, Munksgaard, Copenhagen (1970). 

6. Batchelor, J. R.. Ann. Rheum. Dis. 34, 4 1 

Received October 11, 1977. P.S.E.B.M. 1978. 


Effect of Ethanol on Parathyroid Hormone and Calcitonin Secretion in Man (4031 1) 




Section of Endocrinology, Departments of Medicine and Nuclear Medicine, VA West Side Hospital and UniversUy of 
Illinois College of Medicine, Chicago, Illinois 60612 

Peng et al (1, 2) showed that ethanol can 
induce hypocalcemia in dogs and in intact 
and parathyroidectomized rats, which could 
not be prevented by exogenous parathyroid 
hormone (PTH). Ramp et al (3) observed 
ethanol-induced hypoodcemia in chickens. 
Subsequent studies in the rat from our labo- 
ratory (4) showed that ethanol caused a dose 
related hypocalcemia and an increase in PTH 
secretion which, however, was not sufiicient 
to correct the hypocalcemia. The present 
study (a) evaluated the effect of ethanol on 
PTH and calcitonin (CT) secretion in vivo in 
normal man, and (b) evaluated the mode of 
ethanol effect on PTH secretion by studying 
its effect on bovine parathyroid tissue in vitro. 
Maierials and methods. Human studies. 
Normal male subjects aged 25-50 years, on 
normal diets and with no evidence of renal, 
calcium (Ca) metabolic or other endocrine 
abnormalities underwent the alcohol inges- 
tion test. Informed consent was obtained 
from each subject to undergo this procedure, 
which had b^n approved by the Human 
Investigation Committee of this institution. 
The subjects were fasted for 12-15 hr before 
the procedure and were recumbent during 
the procedure. A scalp vein needle was placed 
in an antecubital vein and attached to an 
infusion set via a 3-way stopcock for the slow 
administration of normal saline (0.5 ml/min) 
and withdrawal of serial blood specimens. 
After a 30 min rest period, blood specimens 
were obtained for plasma immunoreactive (i) 
CT and for serum iPTH and Ca at -10 and 
-5 min for baseline values. The subject then 
drank ethanol (0.8 gAg) in the form of 86 
proof bourbon whiskey (one fourth of total 
dose at 0, 20, 40 and 60 min). Additional 
blood specimens were obtained at V^ 1, IV^, 

' Supported by the Medical Research Service of The 
Vecerans Administration. 

2, 3, 4 and 4¥i hr from the time ethanol 
ingestion was begun. A control group of five 
normal male subjects underwent a procedure 
which was similar except that ingestion of a 
volume of tap water approximating the vol- 
ume of ethanol was substituted for the inges- 
tion of ethanol. 

A portion of each blood specimen was 
placed in a chilled heparinized tube, centri- 
fuged in a refrigerated centrifuge and the 
plasma separated and frozen immediately for 
subsequent analysis of iCT. The other portion 
of the blood specimen was placed in a plain 
tube, allowed to clot for 1 hr, centrifuged, 
and the serum separated and frozen for sub- 
sequent iPTH and Ca determination. 

Serum iPTH was determined by a method 
developed in this laboratory (5) using a 
guinea pig antibovine PTH antiserum, puri- 
fied bovine PTH (Wilson Laboratories, lot 
147865) for tracer, and dilutions of a pool of 
human parathyroid tissue culture medium for 
standards. This antiserum detects both the 
intact molecule and the amino terminal frag- 
ment of bovine iPTH, its molar affinity for 
bovine PTH 1-34 being approximately one 
half that for the intact PTH 1-84 molecule in 
the utilized portion of the standard curve. 
This antiserum has a high affinity for human 
and monkey iPTH, and detects dilutions of 
human serum and purified bovine PTH with 
superimposable displacement curves over a 
60-fold dilution span. The normal mean 
value for human serum iPTH ^ 6.2 fd eq 
standard human parathyroid tissue culture 
medium/ml (pi eq/ml) with a normal range 
(mean ± 2 SD) of 3.8-8.6 pd cq/ml. 

Plasma iCT was determined by a method 
developed in this laboratory (6, 7) using a 
goat antihuman synthetic CT antiserum and 
human synthetic CT (N.V. Organon, batch 
#SC 30) for standard and tracer. The normal 
mean value for human adult male plasma 


0037-9721 n%/\S914i\^Vi\S»l^ 
AH rig^ ittr vcd. 



iCT is 218 pg/ml with a normal range (mean 

Ethanol was added to serum and plasma 
specimens to a concentration of 1 .6%, allowed 
to incubate at 4^ for 2 hr and then assayed 
for iPTH and iCT respectively to determine 
whether ethanol may cause any degradation 
of these hormones or modification of the 
displacement of tracer, which would modify 
the assay-detected concentrations of iPTH or 

Serum Ca concentration was determined 
by a modification of the method of Hill (8). 
The normal mean value is 9.2 mg/dl with a 
normal range of 8.2-10.2 mg/dl. 

In vitro studies. Fresh bovine parathyroid 
tissue slices were incubated for 4 hr in Eagle 
Minimal Essential Medium with 10% calf 
serum by the technique previously described 
from this laboratory (9). The medium was 
completely aspirated and replaced by fresh 
medium hourly. During the first 2 hr the 
medium in all flasks contained 1.25 mM Ca 
(considered to approximate the ionized Ca 
concentration of normal plasma). The first hr 
of incubation was considered an equilibration 
period and this medium was discarded. The 
iPTH in the medium removed at the end of 
the next hr was considered to represent the 
control or zero-time baseline secretion of the 
tissue in that flask. The composition of the 
medium was then modified to contain either 
a high (3.0 mAf) or a low (0.75 mAf) Ca 
concentration or to contain either 0.05% or 
0.3% ethanol, and incubation was continued 
for 2 additional hr. The iPTH concentration 
of each hourly medium sample was deter- 
mined by radioimmunoassay as previously 
described (9), using purified bovine PTH for 
standard and tracer. The concentration of 
iPTH in pg/mg wet wt of parathyroid tissue 
in the zero-time baseline medium sample of 
each flask was designated as 100%. The iPTH 
concentration in the medium harvested at the 
end of each hr for the next 2 hr was then 
expressed as a percent of this zero-time base- 
line value for that flask (9). At least three 
control flasks containing 1.25 mM Ca during 
the entire incubation period were included 
with each group of incubation flasks to eval- 
uate uniformity of secretion with time. The 
percent of zero-time baseline values obtained 
on hours 1 and 2 with the control flasks were 

then adjusted to 100%, and the data 1 
of the other flasks corrected to this i 
baseline. Also, aliquots of media in< 
without tissue were assayed to de 
whether the ethanol had any nonspe 
fects on the immunoassay results. 

In all studies the mean and SE 1 
time period were calculated from the 
ual percent values for the time pe 
each subject (in vivo studies) or each 
vitro studies). Statistical tests of sign 
were carried out with Student's / test 

Results. Human studies. The mean 
values (mean ± SE) for the six huni 
jects were: iPTH-6.6 ± 0.30 /il eq s 
human parathyroid tissue culture 
um/ml, iCT-269 ± 24.5 pg/ml, Ca 
0.1 1 mg/dl. The direct addition of et] 
serum or plasma caused no change i 
or iCT concentrations from those obs 
the serum or plasma without addi 
ethanol. As indicated in Fig. 1, infi 
normal saline and ingestion of water 
mal subjects caused no change in iPI 
or Ca. However, as shown in Fig. 2, ii 
of ethanol caused a significant (P 
increase in iPTH to 107.2 ± 2. 1 1% of 
by 30 min, at a time when only hal 
ethanol had been ingested. The iPTH 
tration continued to increase, reachinj 
value of 138.9 ± 4.44% of baseline (P « 
at 2 hr, with gradual decrease there 
106.0 ± 8.10% of baseline at 4¥i hr. 






Norma I Saline i v 

12 3 4 

Tim* in Hours 

Fig. 1. Effect of iv infusion of normal s 
ingestion of tap water on serial serum iPTH, pi 
and serum Ca concentrations during a AVi-hi 
normal man. Values (mean ± SE) at each tii 
are expressed as percent of the baseline pre 
values (designated as \00%). N » 5. Baselii 
iPTH - 6.4 ± 0.28 /il eq/ml, iCT - 240 ± 22 
Ca- 9.3 ±0.12 mg/dl. 





^Konol0 29<^9 orollyi4 

M f 1 

2 3 

Tim* in Hours 

Fig. 2. Effect of ethanol ingestion on serial serum 
iPTH, plasma iCT and serum Ca concentrations during 
the following 4)^ hr in normal man. Values (mean ± SE) 
at etch time period are expressed as per cent of the 
baseline pre-ingestion values (designated as 100%). N *- 
6. Baseline values: iPTH - 6.6 ± 0.30 /il cq/ml, iCT - 
269 ± 24.5 pg/ml Ca - 9.4 ± O.l 1 mg/dl. 

iCT concentration increased more slowly, 
showing a significant (P < 0.05) rise to 1 15.0 
± 6.07% of baseline at 1 V^ hr, reaching a peak 
value of 137.8 ± 7.13% of baseline (/>< 0.001) 
at 3 hr and then decreasing to 101.9 ± 1.90% 
of baseline at 4V^ hr. Serum Ca did not sig- 
nificantly change at any time tested. 

In vitro studies, Aliquots of medium (with 
or without added ethanol) which had been 
incubated without parathyroid tissue re- 
vealed no modification of the trace B/F ratio, 
indicating that neither the medium nor 
ethanol had any nonspecific effects on the 
immunoassay results. Changes in in vitro se- 
cretion of iPTH, related to medium changes 
in Ca ion concentration or to addition of 
ethanol, are portrayed in Fig. 3. Hourly iPTH 
secretion revealed only minimal variation 
when medium containing 1.25 mA/ Ca was 
used during the entire incubation period: 
(iPTH = 325 ± 13.8, 311 ± 22.4 and 319 ± 
14.0 pg/mg wet wt of parathyroid tissue/hr 
at 0, 1 and 2 hr respectively). Each value was 
designated at 100% for that hr. Medium con- 
taining low (0.75 mA/) Ca caused a significant 
(/*< 0.001) increase in iPTH release to 142.5 
± 9.41% and 240.2 ± 8.10% of baseline at the 
first and second hr of incubation respectively. 
Medium containing high (3.0 mA/) Ca caused 
a significant (P < 0.001) decrease to 57.3 ± 
4.63% and 42.7 ± 4.23% of baseline at the 
fust and second hr of incubation respectively. 
Addition of two concentrations of ethanol to 
1.25 mA/ Ca medium caused increases in 


Timt in Hours 

Fig. 3. Effect of Ca concentration and of concentra- 
tions of ethanol on hourly iPTH secretion in vitro by 
normal bovine parathyroid tissue. Medium prior to zero 
time contained 1 .25 mM Ca and no ethanol in all flasks. 
When medium containing 1.2S mM Ca was continued 
during the entire incubation period (control flasks), the 
iPTH values (mean ± SE) were: 325 ± 13.8, 31 1 ± 22.4 
and 319 ± 14.0 pg/mg wet wt parathyroid tissue at 0, 1 
and 2 hr respectively. Each of these values was desig- 
nated as 100% baseline for that hr. Each value is ex- 
pressed as per cent of the zero time iPTH secretion for 
that flask, (corrected for the minimal variation of secre- 
tion in control flasks for that hr). Each point is the mean 
± SE for six flasks. 

iPTH secretion. At a concentration of 0.05% 
ethanol, the increase to 105.1 ± 6.10% of 
baseline at 1 hr was not significantly different 
from baseline, but the iPTH increase to 122.1 
± 6.74% at 2 hr was significantly (P < 0.02) 
increased. The 0.3% concentration of ethanol 
caused increases in iPTH secretion to 124.1 
± 6.35% at 1 hr and to 166.3 ± 11.26% of 
baseline at 2 hr, both being significantly (P 
< 0.02 and P < 0.001 respectively) increased 
from baseline secretion. 

Discussion. Our initial studies of the effect 
of ethanol on Ca metabolism in the rat (4) 
suggested that decrease in serum Ca was the 
primary event and the observed increase in 
serum iPTH was in response to the hypocal- 
cemic effect of ethanol. However, this com- 
pensatory increase in iPTH did not prevent 
or fully correct the hypocalcemia. We pro- 
posed (through without supporting data) that 
ethanol may induce a decrease in bone re- 
sorption, leading to hypocalcemia and a rel- 
ative skeletal resistance to the resorptive ac- 
tion of PTH. However, the studies of Peng et 
al (2) suggest that decreased bone resorption 
does not occur and that there may be a 
shifting of Ca from extracellular fluid into 
tissues to explain the hypocalcemia. This ex- 



planation is strengthened by the observations 
of Ramp et al. (3) that adding ethanol to the 
organ culture medium enhanced mineral ac- 
cretion by embryonic chick bone. 

The present study indicates that, in normal 
man, ethanol induces an increase in both 
serum iPTH and iCT without detectable 
change in plasma Ca. This observation could 
be explained by ethanol-induced decrease in 
bone resorption or increase in bone accretion, 
but without skeletal resistance to PTH. In this 
situation very minimal hypocalcemia would 
induce increased PTH secretion, with rapid 
bone resorption and restoration of serum Ca 
to normal, so that hypocalcemia was never 
detectable. However the in vitro observations 
indicate that a primary change in serum Ca 
is not the total explanation of the changes in 
serum iPTH. In this situation ethanol had a 
direct stimulatory effect on the parathyroid 
cell which was dose-related. It is therefore 
possible that ethanol has both an indirect (via 
induced hypocalcemia) and a direct effect on 
PTH secretion. The effect of ethanol on the 
C cell of the thyroid was not studied in vitroy 
but is inferred to also be direct, leading to 
increase in CT secretion. The simultaneous 
increase in PTH and CT secretion may at 
least partially explain the lack of changes in 
serum Ca in the present study. 

Other investigators have reported that 
ethanol can stimulate CT secretion in patients 
with medullary carcinoma of the thyroid, and 
have proposed ethanol ingestion as a CT 
secretagogue (along with calcium infusion 
and pentagastrin injection) as a diagnostic 
test for this tumor (10-14). Initially, it could 
not be demonstrated that ethanol affected CT 
secretion in normal subjects (13), but a sub- 
sequent study, using a more sensitive assay 
method, demonstrated that some normal sub- 
jects do show a CT response to ethanol (14). 
In the present study using a larger dose, 
ethanol elicited a CT response in all six nor- 
mal subjects tested. We are not aware of a 
previous report of the effect of ethanol on 
PTH secretion. 

The lower media concentration (0.05%) of 
ethanol in the in vitro study is comparable to 
the average blood ethanol concentration 
achieved in social drinking situations, and the 
0.3% media concentration of ethanol is com- 
parable to the blood ethanol level achieved 

by severely intoxicated subjects (IS 
doses of ethanol (0.8 gAg) ingested 
human subjects in the present study rei 
them only moderately intoxicated. Th< 
ethanol, in amounts often ingested by 
drinkers, increases both PTH and CT 
tion, and therefore may modify Ca h 

Summary. Ingestion of 0.8 g/kg etlu 
1 hr by normal man caused si^iifia 
creases in both serum PTH and plasi 
concentrations, with peak values of V. 
baseline at 2 hr for PTH and of 138% 
for CT. Serum Ca did not change duri 
period of observation. Incubation of 1 
parathyroid slices in 1.2S mM Ca Bag 
dia with 0.05% or 0.3% ethanol caused 
icant increases in PTH secretion to 122 
166% of baseline respectively. Therefc 
in vitro, ethanol can be demonstrated 
rectly stimulate PTH secretion, (2) l 
ethanol ingestion induces an increase h 
without detectable hypocalcemia, sugj 
(a) prompt PTH secretion and action t 
pensate for a hypocalcemic effect of c 
so that actual hypocalcemia is not dete 
and/or (b) direct parathyroid stimu 
Though the exact mechanisms are u 
the data indicate that ethanol, in ar 
often ingested by social drinkers, in< 
both PTH and CT secretion, and th< 
may modify Ca homeostasis. 

The authors are grateful to Bertha Jackson, 
Kawahara and Patricia Johnson for technical a^ 
and to Barbara Lovett for secretarial assistance. 

1. Peng, T. C, and Gitelman, H. J., Endocrine 

2. Peng, T. C, Cooper, C. W., and Munsor 
Endocrinology 91, 586 (1972). 

3. Ramp, W. K., Murdock, W. C, Gonnermar 
and Peng, T. C, Calc. Tiss. Res. 17, 195 (19 

4. Shah, J. H., Bowser, E. N., Hargis, G. K., ^ 
awat, N., Baneijee, P., Henderson, W. J., a 
hams, G. A., MeUbolism 27, 257 (1978). 

5. Hargis, G. K., Williams, G. A., Reynolds, 
Kawahara, W., Jackson, B., Bowser, E. N. 
R. M., Clin. Chem. 23, 1989 (1977). 

6. Hargis, G. K., Williams, G. A., Reynolds, 
Chertow, B. S., Kukreja, S. C, Bowser, E. 
Henderson, W. J., Endocrinology 102, 745 

7. Hargis, G. K., Reynolds, W. A., Williams 
Kawahara, W., Jackson, B., Bowser, E. 
Pitkin, R. M., Clin. Chem. 24, 595 (1978). 



8. Hill, J. B.. Clin. Chem. 2, 122 (196S). 

9. Williams, G. A.. Hargis, G. K., Bowser, E. N., 
Henderson, W. J., and Martinez, N. J., Endocrinol- 
ogy W, 687 (1973). 

10. Cohn, S. L., Grahame-Smith, D., Maclntyre, 1., and 
Walker, J. G., Lancet 2, 1 172 (1973). 

11. Wells, S. A., Jr., Cooper, C. W., and Ontjes, D. A., 
Metabolism 24, 1215(1975). 

II Milhaud, G., Riberiro, F. M., Calmettes, C, Tabou- 
let, J., Coutris, G., and Moukhtar, M. S., Nouv. 

13. Dymling, J. F., Ljungbcrg. O.. Hillyard, C. J., Green- 
berg, P. B., Evans, 1. M. A., and Maclntyre, 1., Acta 
Endocrinol. 82, 500 (1976). 

14. HiUyard, C. J., Cooke, T. J. C, Coombes, R. C, 
Evans, 1. M. A., and Maclntyre, I., Clin. Endocrinol. 

15. Mendelson, J. H., in 'Textbook of Medicine** (P. B. 
Beeson and W. McDermott, eds.), p 597. W. B. 
Saunders Co., Philadelphia (1975). 

Received February 17. 1978. P.S.E.B.M. 1978, Vol. 159. 


Pancreatic Secretory Isoenzyme of Alkaline Phosphatase (4031 2) 


Section of Gastroenterology, Department of Internal Medicine and Section of Biochemistry, Department of Clinical 
Pathology, Scott and White Clinic, Temple, Texas 76501 

Alkaline phosphatase exists in a wide va- 
riety of tissues in different molecular forms. 
Characterization of these isoenzymes is pos- 
sible on the basis of their resistance to various 
physical and chemical manipulations. As 
early as 1944, Nothmann (1) reported that 
ligation of the pancreatic duct in dogs pro- 
duced an increase in serum alkaline phospha- 
tase, but there have been few attempts to 
measure this enzyme in pancreatic juice. 
Wames and Buhner (2) demonstrated the 
presence of alkaline phosphatase in the duct 
system, islet cells, and acini of the human 
pancreas. Wames et al (3) extracted alkaline 
phosphatase from normal human pancreas 
and pancreatic tumors and showed that these 
enzymes have distinctive isoenzyme charac- 
teristics when compared with the enzymes of 
the small intestine and of normal serum. 

The present study was designed to examine 
the isoenzyme characteristics of alkaline 
phosphatase in canine pancreatic secretory 
fluid. The availability of pure human pan- 
creatic juice from a patient with a traumatic 
fistula allowed us to conduct similar obser- 
vations in this fluid. 

Methods. Six adult mongrel dogs, weighing 
14-18 kg, were previously prepared with gas- 
tric and pancreatic flstulas fltted with 
Thomas cannulas in the stomach and duo- 
denum (4). Animals were not studied until 
3-4 weeks after this operation and were de- 
prived of food but not water for approxi- 
mately 18 hr prior to each study. A continu- 
ous iv infusion of 0.15 M sodium chloride 
was given at a rate of 50 ml/hr. Observations 
were carried out in conscious animals during 
continuous intravenous infusion of secretin, 
0.5 U/kg per hr. The secretin used in these 
studies was from a single batch purchased 
from the Gastrointestinal Hormone Research 
Unit, Karolinska Institute Chemistry Depart- 

' Reprint requests to: Dr. W. P. Dyck, 2401 South 
3 1 St Street, Scott and White Clinic, Temple, Texas 7650 1 . 

OOJ7-9727/78/I592-0l92S0\ .00/0 
Copyright €> 1978 by the Society for ExperimentMl Biology and Medicine 
^Ml rights reserved. 

ment, Stockholm, Sweden. The gastric can- 
nula was kept open during all observations to 
prevent the entry of add into the duodenum. 
The duodenal cannula was opened and a 
glass cannula was inserted into the pancreatic 
duct under direct vision. Pancreatic secretion 
was collected continuously as 10-min speci- 

Pancreatic juice also was collected from a 
patient who had an established posttraumatic 
pancreatic fistula that was draining clear, al- 
kaline juice, 400-600 ml/day, with a bicar- 
bonate concentration of 68 meq/liter and an 
amylase concentration of 120,000 Somogyi 
U/100 ml. Fluid was collected by direct can- 
nulation of the fistula with a sterile catheter 
after appropriate skin cleansing to minimize 
the likUhood of bacterial contamination. 

Alkaline phosphatase, expressed in inter- 
national units, was determined by the method 
of Roy (5) with thymoiphthalein monophos- 
phate as the substrate. 

Isoenzyme characterization, based on dif- 
ferent susceptibilities of alkaline phosphatase 
isoenzymes to inhibition by urea and L-phen- 
ylalanine (6-8) and heat inactivation (9), was 
performed in all specimens. The method of 
Kind and King (10) was used for alkaline 
phosphatase measurements in these isoen- 
zyme studies. 

Isoenzymes present in the human pan- 
creatic fistula fluid were examined by acryl- 
amide gel electrophoresis and compared to 
the electrophoretic behavior of alkaline phos- 
phatase of known human origin from liver, 
bone, and intestine. Liver alkaline phospha- 
tase was obtained from the serum of patients 
with known liver disease and intestinal al- 
kaline phosphatase was purchased from 
Dade Corporation. Bone alkaline phospha- 
tase was obtained from shavings of bone 
extracted with butanol to remove insoluble 
material and break the protein-lipid bond. 
The alkaline phosphatase obtained from the 
pancreatic fistula fluid was concentrated ten 




efore electrophoresis. All samples to be 
3phoresed were dialyzed for twelve 
against two changes of electrophoresis 
'. Alkaline phosphatase isoenzymes 
^parated by Raymond's method of con- 
is polyacrylamide gel electrophoresis in 

alts. At the low dose of secretin infusion 
d in the canine studies, pancreatic se- 
y volumes varied from 5 to 10 ml/ 10 
rhe mean (± SEM) alkaline phospha- 
oncentration in specimens from all six 
Us (87 collections) was ISA ± 1.1 
nl. Alkaline phosphatase concentration 
icreatic fistula fluid collected from the 
It was 17.8 mU/ml. 

are 1 shows the percentage of alkaline 
hatase remaining in pancreatic juice 
zach of the six dogs after incubation of 
lens with urea or phenylalanine or after 
lactivation. There was relative uniform- 
long the animals in that the isoenzyme 
ited relative resistance to phenylalanine 
tion, intermediate inhibition by urea, 
larked thermal lability, 
enzyme characteristics of alkaline phos- 
ie in canine and in human pancreatic 
3ry fluid are compared in Fig. 2. The 
alkaline phosphatase activity remain- 

ing after phenylalanine inhibition was 67% in 
canine pancreatic juice compared to 85% in 
human juice. The mean values after urea 




Fig. 2. Mean percent of alkaline phosphatase re- 
maining in canine pancreatic juice and the percentage of 
enzyme remaining in human pancreatic fistula fluid after 
incuhation with urea or L-phenylalanine or after heat 
inactivation. Bars at left represent mean ± SEM of all 
collections from six dogs. 





DOG 1 

DOG 2 

DOG 3 

DOG 4 

DOG 3 

DOG 6 


(n = 18) 





I. Percentage alkaline phosphatase remaining in pancreatic juice after incubation with urea or L-phenylal- 
after heat inactivation. Each bar represents the mean ± SEM of all 10-min collections in a single animal 
oatinuous intravenous infusion of secretin, O.S UAg per hr. n « number of observations in each mean. 



inhibition were 30% and 34%, respectively, 
and after heat inactivation were 7.5% and 
21% respectively. 

Figure 3 shows the electrophoretic mobility 
of the alkaline phosphatase isoenzyme in the 
human pancreatic fistula fluid compared to 
mobilities of isoenzymes derived from other 
human tissue sources. The pancreatic enzyme 
exhibited a pattern of mobility clearly differ- 
ent from that of any of the isoenzymes of 
other tissues sources. 

Discussion, When Nothmann (1) found 
that ligation of the pancreatic duct in dogs 
resulted in a progressive increase in serum 
alkaline phosphatase activity, he assumed 
that this increased activity was of pancreatic 
origin. Subsequent studies (12, 13) have 
shown a significant increase in alkaline phos- 
phatase concentration in duodenal juice after 
CCK-pancreozymin stimulation. The dem- 
onstration, by histochemical techniques, of 
this enzyme in various cellular components 
of the human pancreas (2) and the subsequent 
identification of distinctive isoenzyme char- 
acteristics of pancreatic alkaline phosphatase 
(3) are consistent with the presence of this 
enzyme in pancreatic secretory fluid. 

Our data are in agreement with the find- 
ings of Wames et al (3) who showed that 
pancreatic alkaline phosphatase was much 
more sensitive to heat inactivation and urea 
inhibition than was the enzyme from the 
small intestine, but, in contrast, was largely 
unaffected by L-phenylalanine. 

The question of whether increased serum 

Fig. 3. Electrophoretic patterns (acrylamide gel) of 
alkaline phosphatase isoenzymes from human sources: 
1, bone; 2, mixed liver and intestine; 3, intestine; and 4, 
pancreatic fistula fluid. Vertical electrophoresis in pH 
9.0 Trismalein acid buffer (0.283 M and 0.019 M, re- 
spectively) at 4**; 300 V for 3 hr, suined with sodium a- 
naphthyl acid phosphate (1 hr). 

total alkaline phosphatase values i 
times reflect a predominant increase 
pancreatic isoenzyme remains unac 
and must await isoenzyme charactei 
studies in subjects with acute inflam 
as well as neoplastic disease of the p£ 
Additional techniques, such as acrylan 
electrophoresis, will doubtless aid in 
refining our means of identifying the 
of different serum isoenzymes (14). 
Summary, Alkaline phosphatase 
was measured in hormonally stimulat 
creatic juice from six dogs and in pai 
fistula fluid from a human subject. Iso< 
characterization studies, based on d 
susceptibilities to urea and L-phenyl 
inhibition and to heat inactivation in 
similarities between canine and hiuns 
creatic secretory alkaline phosphatase 
pared to intestinal alkaline phosphat 
pancreatic isoenzyme was much mor 
tive to heat inactivation and urea in! 
but much more resistant to L-phenyl 
inhibition. The electrophoretic mob 
the enzyme present in human pancreai 
was different from that of human I 
bone, or intestinal alkaline phosphata 

1. Nothmann, M. M., Proc. Soc. Exp. Biol. M 

2. Wames, T. W., and Buhner, D. J., Anat. 

3. Wames, T. W., Timperley, W. R., Hinc, P., 
G., Gut 13, 513 (1972). 

4. Thomas, J. E., Proc. Soc. Exp. Biol. Med 

5. Roy, A. v., Clin. Chem. 16, 431 (1970). 

6. Bahr, M., and Wilkinson, J. H., Clin. Ch 

7. Home, M., Cornish, C. J., and Posen, S 
Clin. Med. 72, 905 (1968). 

8. Kreisher, J. H., Close, V. A., and Fishmai 
Clin. Chim. AcU 11, 122 (1965). 

9. RalUfF, C. R., Hall, F. F., Culp, T. W., Gc 
E., and Westfall, C. L., Amer. J. Gastroec 

10. Kind, P. R. N.. and King, E. J., J. Clin. 1 

11. Raymond, S., Ann. NY Acad. Sci. 121, 35( 

12. Wames, T. W., Hine, P., and Kay, G., Gui 

13. Oyck, W. P., Martin, G. A., and Ratlii 
Gastroenterology 64, 599 (1973). 

14. Smith, I., Lightstone, P. J., and Perry, J. 
Chim. Actel9,499(l%8). 

Received January 16, 1978. P.S.E.B.M. 1978, \ 


e of Thymopoietin, Ubiquitin and Synthetic Serum Thymic Factor to Restore 
Immunocompetence in T-Cell Deficient Mice (40313) 


of Virus and Cell Biology Research, Merck Institute for Therapeutic Research^ ^est Point, Pennsylvania 
the * Department of Medicinal Chemistry, Merck Sharp and Dohme Research Laboratories, West Point, 

Pennsylvania 19486 

le of the thymus in lymphocyte ho- 
(1, 2) and in conversion of precur- 
hocytes to thymus-dependent lym- 
(T cells) has been the subject of 
investigation during the last 20 
sence of the thymus leads to diverse 
ogic deficiencies that can be re- 
' thymus grafts or by thymus grafts 
in cell-impermeable chambers (3-7), 
g that the thymus might induce mat- 
f T lymphocytes through production 
>le factor(s). Candidate thymic fac- 
; been prepared by several investi- 
he biological activities of these fac- 
; been assessed mainly by in vitro 
of T cell markers (e.g. thy-1 anti- 
^mphocyte populations (8-12), with 
to attention given to whether there 
oration of thymus-dependent im- 
ictions measurable in vivo, 
esent study was undertaken to eval- 
Tal substances that induce T cell 
for their abiUty to restore thymus- 
it immunocompetence in thymecto- 
8 mice. The substances tested were 
letin and ubiquitin, prepared by Dr. 
itein (10, 11), and synthetic serum 
ctor, defmed by Bach et ah (13). 
lis and methods. Mice. C58/J mice 
lined from the closed colony main- 
r the Merck Sharp & Dohme Rc- 
boratories by Buckshire Farms, Per- 
or were purchased from The Jack- 
ratory. Bar Harbor, Maine. 
il evaluated. Partially purified thy- 
1 II (10) and ubiquitin (11) were 
as aliquoted lyophilized prepara- 
)r. Gideon Goldstein, Sloan-Ketter- 
:er Research Center, New York, 
o be tested were dissolved in phos- 
ifTered sahne (PBS) immediately 
ise. Mice were injected intraperito- 

neally with the substances in 0. 1 ml. Synthetic 
serum thymic factor (Pyroglu-Ala-Lys-Ser- 
Glu-Gly-Gly-Ser-Asn), defmed by J-F Bach 
et al (13), was synthesized by the peptide 
synthesis group (R. G. Strachan, W. J. Pale- 
veda, S. J. Bergstrand, R. F. Nutt, R. Hirsch- 
mann, F. W. Holly, and D. F. Veber, manu- 
script in preparation), of Merck Sharp and 
Dohme Research Laboratories, and found 
positive for in vitro biological activity by Dr. 
J-F Bach, Nccker Hopital, Paris, France. 
Dosages, formulations and frequencies of 
treatment employed in these experiments 
were based on recommendations made by 
Dr. G. Goldstein for thymopoietin and ubi- 
quitin, and by Dr. J-F Bach for serum thymic 

An unrelated pentapeptide, Asp-Scr-Asp- 
Pro-Arg (14), and a decapeptide, Val-His- 
Lcu-Ser-Ala-Glu-Glu-Lys-Glu-Ala (15), that 
failed to show in vitro biological activity in 
tests performed by Dr. J-F Bach were used 
for control purpose. 

Thymectomy. Mice that were 0-2 days old 
were anesthetized by cooling at -20° (16) for 
5-8 min (depending on size), thymectomized 
according to the method of Sjodin et al (17), 
and then warmed under an infrared lamp 
(35 °C) for 30 minutes. Young adult (4- to 6- 
week-old) mice were anesthetized by a single 
intraperitoneal injection (62.5 mg/kg) of so- 
dium pentobarbital (Nembutal, Abbott Lab- 
oratories, North Chicago, Illinois), and thy- 
mectomized according to the method of Dar- 
denne and Bach (18). Sham thymectomized 
mice were treated surgically in the same man- 
ner except that the thymic lobes were not 
removed. At the appropriate time, all thy- 
mectomized mice were examined for presence 
of thymic remnants with the aid of a dissect- 
ing microscope. Mice found to have thymic 
remnants were excluded from the study. 



Copyright © \91% b^ lYvt Socvev^ Vox ^xvtTvmt,tv\»\%va\«i^ %sw^\K<&«>s*. 

AU rights reserved. 



Anti'thymocyte serum treatment. Heat-in- 
activated rabbit anti-mouse thymocyte and 
normal rabbit sera were purchased from Mi- 
crobiological Associates, Bethesda, Mary- 
land. A single 1 ml injection of serum was 
given intraperitoneally 3-4 days after thy- 
mectomy. Certain lots of anti-thymocyte sera 
were toxic for the mice and were not used. 

Preparation and administration of h cell 
suspensions. The challenge inoculum was 
made by mixing equal volumes of viable (2 
X 10^/ml) and irradiated (2 X lOVml) line lb 
leukemic cell suspensions. Suspensions of vi- 
able (C58 mouse-derived) lb cells were pre- 
pared in Hanks balanced salt solution as 
described previously (19). To prepare y-irra- 
diated lb cells, suspensions of viable cells were 
exposed to 10,000R in a Model 109 Co®^ 
Irradiator (J. L. Shepherd and Assoc., Glen- 
dale, CA) that delivered 62,000R/min. Mice 
were injected intraperitoneally with 1 ml of 
the lb cell mixture (10^ viable plus 10* irra- 
diated cells). The mice were observed for 21 
days and gross examination of the viscera of 
all mice that died was made to assure that 
deaths were due to leukemia. 

Mitogenic responses. To test for capability 
to respond to mitogens, spleen and mesenteric 
lymph node cell suspensions were prepared 
in medium RPMI 1640 (Grand Island Bio- 
logical Co., Grand Island, NY) containing 
5% fetal calf serum (Microbiological Associ- 
ates). Five replicate cell suspensions, each 
containing 4 x 10^ cells in 0.2 ml, were pre- 
pared for testing the response to concanavalin 
A (Con A) (Miles Laboratories, Kankakee, 
IL) and phytohemagglutinin P (PH A) (Difco, 
Detroit, MI) in final concentration of 0.4 
/ig/ml and 1: 1000 dilution, respectively. After 
3 days of incubation (37°, 5% CO2), 1 /iCi of 
tritiated thymidine (New England Nuclear, 
Boston, MA) in 0.025 ml was added to each 
cell preparation, and incubation was contin- 
ued for an additional 4 hr. The cells were 
harvested, washed to remove residual free 
fluids, and dissolved in 10 ml of Scintisol 
(Isolab, Akron, OH). The counts per minute 
were determined, and the mean cpm was 
calculated for the 5 replicate cultures in each 

Results. Failure of thymopoietin to restore T 
cell mitogen responses of lymphocytes from 
neonatally thymectomized C58 mice. Findings 

in preliminary experiments indicated 
both spleen and lymph node cells froi 
treated or PBS-treated neonatally thyc 
mized CSS mice failed to be stimulated 
cell mitogens. In fact, incubation with ( 
or PHA generally resulted in decreases 
midine incorporation compared to tt 
control cells that were not treated with 

To test for ability of thymopoietin 
store T cell mitogen responses, neon 
thymectomized CSS mice were treated 
with thymopoietin for 4 weeks startin 
week of age. The animals were sacrific 
cervical dislocation 1 day following tb 
injection. Spleen and lymph node cells 
removed from the animals and teste 
mitogenic responses to Con A and PH 
shown in Table I, treatment with thym< 
tin did not restore normal responsiven 
the spleen and lymph node cells to the 

Failure of thymopoietin and ubiquitin 
store resistance to line h leukemia in 
thymectomized C58 mice. It was demons 
in previous studies that normal adul 
mice develop an immune response (sur 
when simultaneously vaccinated and 
lenged with a mixture of viable and 
line lb leukemic cells, whereas inmiun 
pressed mice do not (19, 20). This im 
response is highly dependent on ftinc 
maturity of the T-lymphocytes (21). A 
periment was carri^ out in which adul 
mectomized and sham thymectomized 
trol animals were treated with rabbit 
thymocyte serum to reduce the populati 
competent lymphocytes in the periphery 
animals were then challenged with the 
preparation described above. As sho^ 
Fig. 1, sham thymectomized animals 
initially highly susceptible to challenge 
lb cells but their immunologic responsi^ 
was regained within 4 weeks after seru: 
ministration. By contrast, animals tha 
been thymectomized did not regain th( 
sistance. Similar differences in regenei 
of T cell mitogen responses (22) and 1 
bearing lymphocytes (23) were observe 
tween adult thymectomized and shan 
mectomized mice given anti-thymocy 

An attempt was made to restore th 



TABLE I. Failure of Thymopoietin to Restore T Cell Mitogen Responsiveness in Neonatally 

Thymectonozed C58 Mice. 

In Vivo treatments" 

Thymectomy Substance 

In Vitro mitogen responses^ 

Spleen Cells 

Lymph Node Cells 

Control Con A stim- Control Con A stim- 

cpm (No. PHA stimu- ulation in- cpm (No. PHA stimu- ulation in- 
mice) lation index dex mice) lation index dex 




























" Neonatally thymectomized (NTx) C58 mice were treated ip with 1 /ig thymopoietin (TP) or with PBS 5x/week 
for 4 weeks (20 treatments) starting at 1 week of age. 

* Averages of individual determination obtained from the indicated numbers of mice. Stimulation Index -■ 
Mitogen Stimulated cpm/control culture cpm 

* Data from two animals. 

1 234567 89 10 II 12 

Fig. 1. Effect of adult thymectomy on recovery of 
the immune response to line lb leukemic cells following 
anti-thymocyte serum treatment. Groups of thymecto- 
mized (#) and sham thymectomized (O) mice were 
injected with an admixture of 10^ viable Ih cells and 10^ 
y-irradiated lb cells at the indicated times after the 
injeaion of 1 ml rabbit anti-mouse thymocyte serum. 
Each point represents 20 mice. Groups of thymectomized 
mice given normal rabbit serum survived the injection 
of lb cells. 

munologic responsiveness of anti-thymocyte 
serum-treated adult thymectomized CS8 mice 
by administration of thymopoietin or ubiqui- 
tin. Such mice were injected intraperitoneally 
with 1 jLig thymopoietin or ubiquitin S times 
per week for 5 weeks prior to challenge with 
lb cells. Neither thymopoietin nor ubiquitin 
restored the resistance of serum-treated thy- 

mectomized animals to line lb leukemia (Ta- 
ble II). On the other hand, control serum- 
treated sham thymectomized mice were re- 
sistant to challenge. 

Failure of serum thymic factor to restore 
resistance to line h leukemia in adult thymec- 
tomized CSS mice. In similar experiments to 
those described above, anti-thymocyte se- 
rum-treated adult thymectomized CSS mice 
were injected 3 times per week for 8 weeks 
with synthetic serum thymic factor (0.1 ng) 
prepared with carboxymethylcellulose as de- 
scribed by M-A Bach (24). T cell inmiuno- 
competence was measured in terms of the 
survival rates of animals challenged with line 
lb leukemia. As shown in Table III, resistance 
to challenge was not restored to serum-treated 
thymectomized mice by treatment with serum 
thymic factor. Control animals that had been 
sham thymectomized and treated with anti- 
thymocyte serum were resistant to challenge. 

In the experiments just described, repeated 
injections of carboxymethylcellulose, in 
which the test preparations were suspended, 
was toxic; causing skin nodules, ulceration 
and death in roughly half the animals during 
the 8 week period of treatment. To avoid this, 
adult thymectomized animals that had been 
given anti-thymocyte serum were treated 5 
times per week for 8 weeks with 1 jug serum 
thymic factor in PBS and then challenged 



TABLE II. Lack of Effect of Thymopoiehn and Ubiquitin on Resistance of Anti-Thymocyte Serum- 
Treated Adult Thymectomized Mice to Challenge with Line Ik Leukemia. 

Treatment of mice 

Response to line Ih leukemia'' 

Pre-Therap)^ T Cell De- 

Subsunce Tested'^ 

No. of surviving/total 

Average time of death in 





Thymectomized test animals 
Thymopoietin 0/ 16 (0) 

Ubiquitin 1/13 (8) 

Thymectomized control animals 
PBS 0/12(0) 

None 0/13(0) 

Nonthymectomized control animals 
PBS 11/12(92) 

PBS 20/20(100) 

10.69 ± 0.60 
10.69 ± 0.78 

10.75 ± 1.06 
10.92 ± 1.75 

" C58 mice were adult thymectomized (ATx) or sham operated (STx) and given rabbit anti-mouse thymocyte 
serum (RAMTS) 3 days later. 

^ Treated mice received 1 yi% of thymopoietin or ubiquitin ip 5x/week for 5 weeks. 
^ All mice were challenged with a mixture of 10^ viable and 10^ irradiated lb cells. 
* One mouse died on day 15. 

TABLE 111. Lack of Effect of Cohplexed Serum Thymic Factor on the Resistance of Anti-Thymocyte 
Serum-Treated Adult Thymectomized Mice to Challenge with Line Ik Leukemia. 

Treatment of mice 

Pre-therapy T 
cell depletion" 

Substance tested^ 

Deaths following treatment with the 
test substimces in CMC 

No. of mice Survival following 

challen£e with line lb 

Final Survival (%) leukemia (%) 


Thymectomized test animals 
ATx, RAMTS Serum thymic factor 15 7 (46) 

ATx, RAMTS Decapeptide 15 12 (80) 

Thymectomized control animals 
Atx, RAMTS Buffered saline solution 15 8 (53) 

Sham thymectomized control animals 
STx, RAMTS Buffered saline solution 15 9 (60) 

1/7 (14) 
4/11 (36) 

0/7 (0) 

9/9 (100) 

" C58 mice were adult thymectomized (ATx) or sham operated (STx) and given rabbit anti-mouse thymocyte 
serum (RAMTS) four days later. 

^ All test subsunces were conuined in carboxymethylcellulose (CMC) that was highly toxic causing deaths in the 
animals. Treated mice received 0.5 ml sc containing 0.1 ng of serum thymic factor or control decapeptide combined 
with 27 mg CMC. Treatment was surted 6 days following RAMTS treatment and continued three times per week 
for a total of 20 injections. CMC and total volume of treatment were reduced to 5 mg and 0. 1 ml, respectively, aAer 
seven injections. 

with lb cells. The findings given in Table IV 
show that the thymic factor in PBS, as in 
carboxymethylcellulose, failed to restore im- 
munocompetence to the mice. 

Discussion, The main criterion used to clas- 
sify materials as thymic hormones has been 
their capacity to induce T cell surface mem- 
brane markers on lymphocytes. However, the 
induction of such cell markers seems not to 
reflect a maturation event specifically in- 
duced by thymic hormone, since many im- 
related substances, including nonthymic tis- 

sue extracts, ubiquitin, poly A:U, endotoxin, 
prolactin, glucagon, prostaglandin E and his- 
tamine, all have the ability to induce the same 
cell surface markers (11, 25, 26). Therefore, 
it is of value to test candidate thymic hor- 
mones in more discriminating assays; assays 
that would measure the effect on immune- 
competence. The present studies were carried 
out to measure the ability, if any, of thymo- 
poietin, ubiquitin, and serum thymic factor 
to restore immimocompetence in vivo in T 
lymphocyte deficient CSS mice. Daily admin- 



.E IV. Lack of Effect of Uncomplexed Seruh Thymic Factor on the Resistance of Anti- 
40CYTE Serum-Treated Adult Thymectomized Mice to Challenge with Line Ik Leukemia. 

Treatment of Mice" 

Deaths following treatment with the 
test substances 


Substance tested 

No. of mice 

irvival (%) 

Survival following 

challenge with line lb 

leukemia (%) 

Surt Final Su 




Thymectomized test animals 
Serum thymic factor 13 9 
Penupeptide 13 12 

Thymectomized control animals 
Buffered saline solution 13 13 

Sham thymectomized control animals 
Buffered saline solution 13 12 




2/9 (22) 



ere treated as indicated in Table III except that each mouse received 0.1 ml PBS sc containing 
ic factor or control pentapeptide five times per week for a total of 36 injections. 

of thymopoietin to neonatally thy- 
xd C58 mice for 4 weeks did not 


he responsiveness of their lymph 
1 spleen cells to T cell mitogens, 
daily treatments of anti-thymocyte 
ated adult thymectomized CS8 mice 
rmopoietin, ubiquitin, or serum 
ctor were inefTective in restoring the 
)f these mice to resist challenge with 
iikemia. By contrast, sham thymec- 
CSS mice, that had received anti- 
e serum, recovered their resistance 
cell challenge spontaneously within 
eeks after serum treatment. Thus, 
ietin and serum thymic factor, in the 
used, did not mimic the restorative 
provided by the intact thymus, 
ilure of the candidate thymic factors 
^estigation to restore inmumocom- 
a T cell deficient mice might be due 
insufficient exposure of the precur- 
to these substances or to total irrel- 
f the substances to the cell matura- 
:ess. The more likely explanation, 
as suggested by A. L. Goldstein et 
s that inmiunologic maturation is a 
ivolving a number of steps and that 
factor, initiating a single cellular 
ght not be reflected in any meaning- 
Lnologic activity. Scheid et al, for 
have demonstrated that in vivo (28) 
7 (26) treatment with thymopoietin 
TL and thy-1 antigens on murine 
Ics. However, lymphocytes that 
TL marker are known to be inmia- 
mmunoincompetent (29). Similarly, 
r/. (8, 24) have shown that serum 

thymic factor can maintain a normal level of 
short-lived thy-1 -positive lymphocytes in 
adult thymectomized mice. Yet, the responses 
of adult CS8 mice to T cell mitogens and line 
lb leukemia are dependent on a long-lived 
population of thy-1 -positive lymphocytes. 
Perhaps the evaluation of only one substance 
at a time would inevitably result in failure to 
induce immimocompetence. 

Abstract. Thymopoietin, ubiquitin, and se- 
rum thymic factor, all of which induce T cell 
markers on lymphocytes, have been evalu- 
ated for their capacity to induce thymus-de- 
pendent activities in vivo. Multiple treatments 
over a period of weeks failed to restore either 
resistance to line lb leukemia or responses to 
T cell mitogens in T cell-deficient C58 mice. 
The findings suggest that these substances are 
ineffective in inducing thymus-dependent im- 
mimocompetence that is meaningful in the 
intact animal. 

Excellent technical assistance was provided by P. A. 
Dennison, M. E. Davies, C. DeWitt, and S. Michelson. 

1. Metcalf, D., Brit. J. Cancer 10, 442 (1956). 

2. Comsa, J., Lc Sangrc 27, 838 (1956). 

3. MiUer. J. F. A. P., Lancet tt, 748 (1961). 

4. Miller, J. F. A. P., Basten, A., Sprent, J., and Cheers, 
C. CeUular Immunol. 2, 469 (1971). 

5. Good, R. A.. Biggar, W. D.. and Park, B. H., in 
''Progress in Immunology, F* (B. Amos, ed.), p. 699. 
Academic Press, New York (1971). 

6. Stutman, O., Yunis. E. J., and Good, R. A., J. Nat. 
Cancer Inst. 43, 499 (1969). 

7. Osoba, D., J. Exp. Med. 122, 633 (1965). 

8. Bach, J-F, and Dardenne, M., Immunol. 25, 353 



9. Goldstein, A. L.. Slater, F. D., and White. A.. Proc. 
Nat. Acad. Sci. 56, 1010 (1966). 

10. Goldstein, G., Nature 247, 11 (1974). 

11. Goldstein, G., Scheid, M., Hammerling, U., Boyse, 
E. A., Schlesinger, D. H., and Niall. H. D.. Proc. 
Nat. Acad. Sci. 72, II (1975). 

12. Bach, J-F, and Camaud, C. Prog. Allergy 21, 342 

13. Bach, J-F. Dardenne, M., Pleau. J-M, and Rosa, J.. 
Nature (London) 266, 55 (1977). 

14. Hamburger. R. N.. Science 189, 389 (1975). 

15. Veber, D. F., Bennett, C. D.. Milkowski, J. D., Gal, 
G., Denkewalter. R. G.. and Hirschmann, R., Bio- 
chem. Biophys. Res. Commun. 45, 235 (1971). 

16. East, J., and Parrott, D. M. V., J. Endocrinol. 24, 249 

17. Sjodin, K., Dahnasso, A. P.. Smith, J. M., and Mar- 
tinez, C. Transplanution 1, 521 (1963). 

18. Dardenne, M., and Bach, J-F, in Biological Activity 
of Thymic Hormones (D. W. van Bekkum, ed.), p. 
235. Kooyker Sci. Pub., Rotterdam (1975). 

19. Lukasewycz, O. A., Martinez, D., and Murphy, W. 
H., J. Immunol. 114, 1491 (1975). 

20. Martinez, D., Lukasewycz, O. A., and Murphy, W. 
H., J. Immunol. 115, 724 (1975). 

21. Lukasewycz, O. A., DufTey, P. S., and Murphy, W. 
H.. J. Immunol. 116, 976 (1976). 

22. Martinez, D., Field, A. K., TyteU, A. A., and HiUe- 
man, M. R., Abst. Amer. Soc. Microbiol., p. 106 

23. Phelps, A. H.. Martinez, D.. and Field. A. K.. Fed. 
Proc. 36, 1062 (1977). 

24. Bach. M-A, J. Immunol. 119, 641 (1977). 

25. Singh, U.. and Owen, J. J. T., Eur. J. Immunol. 6, 59 

26. Scheid, M. P., Goldstein, G., Hanunerling, U.. and 
Boyse, E. A.. Anal. N. Y. Acad. Sci. 249, 531 (1975). 

27. Goldstein, A. L., Low. T. L. K.. McAdoo. M., 
McClure, J., Thurman. G., Rossio, J.. Lai, C-Y, 
Chang. D.. Wang. S-S, Harvey. C, Ramel A. H., 
and Meinhofer. J.. Proc. Nat. Acad. Sci. 74, 725 

28. Scheid, M. P.. Goldstein. G.. and Boyse. E. A., 
Science 190, 1211 (1975). 

29. Boyse, E. A., Old, L. J., and Stockert, E., in Immu- 
nopathol. Int. Symp. 4th ed. by P. Grabar and P. 
Miescher, Grune and Stratton, New York, p. 23 

Received May 12, 1978. P.S.E.B.M. 1978, Vol. 159. 


lactivated Hepatitis A Virus Vaccine Prepared from Infected Marmoset Liver 



Division of Virus and Cell Biology Research, Merck Institute for Therapeutic Research, 
West Point, Pennsylvania 19486 

olation of the CR326 strain of human 
A virus in mystax marmosets was 
from these laboratories (1, 2) in 1973 
/irus was shown to be inactivated by 
;hyde (3). It was demonstrated, sub- 
y, that the livers of white-mous- 
nd rufiventer marmosets (5. mystax 
\abiatus, respectively) infected with 
irus contained large amounts of hep- 
/iral antigen (3-6). This made possi- 
development of the first practical 
M hepatitis A virus antigen and anti- 
the complement fixation and im- 
Iherence (lA) methods. The present 
bowed that CR326 strain hepatitis A 
ccine, purified from infected mar- 
vtx and inactivated with formalde- 
juced homologous lA antibody and 
1 marmosets against hepatitis A virus 

ials and methods. Marmosets. Wild- 
rufi venter (5. labiatus) marmosets 
d. The animals were conditioned and 
led as described previously (1). 
:. Isocitric dehydrogenase (ICD) as- 
B performed, as previously described 
the marmoset plasmas collected at 
itervals. Values of 1500 Sigma units 
;r obtained for two or more consec- 
eks were considered to be indicative 
ion. Assays for hepatitis A antibody 
armoset sera were by the lA proce- 
cribed earlier (5). 

Itis A vaccine. A rufiventer marmoset 
:ted intravenously with 25th rufiven- 
loset passage CR326 hepatitis A vi- 
liver was perfused in situ with phos- 
ifTered salme solution (PBS) and re- 
rom the marmoset at the time that 
noset first showed pronounced ICD 
i on the 14th day after inoculation. 
:ted liver was homogenized with PBS 
mortar and pestle with alundum to 
0% suspension. The supemate was 

collected following low speed centrifugation 
and was diluted further to give a 5% liver 
extract. The extract was then heated at 60° 
for 30 min after which it was further clarified 
by centrifiigation at 2500 rpm for 30 min 
yielding an amber-colored supemate that was 
slightly opalescent. Formalin in a final con- 
centration of 1:40(X) was added to the super- 
nate, and the mixture was incubated with 
continuing agitation at 35.5° for 4 days. The 
formalin was then partially neutralized with 
sodium bisulfite to give a final concentration 
of 10 /ig/ml formaldehyde. This was the vac- 
cine, and it was stored at 4°. The viral particle 
content per ml was 1.4 x 10^° as measured by 
electron microscopy and the hepatitis A an- 
tigen titer was 1:8 by lA. The liver from a 
noninfected rufiventer marmoset was proc- 
essed in an identical way to produce vaccine 
for control purpose. 

Vaccination, Rufiventer marmosets were 
employed, and all were initially devoid of 
human hepatitis A virus antibody. Eight mar- 
mosets were each given 1 ml amounts of 
hepatitis A vaccine subcutaneously at bi- 
weekly intervals for 14 weeks (eight injec- 
tions). An additional eight animals were in- 
jected subcutaneously at the same time with 
normal marmoset liver vaccine. Six more an- 
imals were each given an intravenous injec- 
tion of 1 ml of hepatitis A vaccine on a single 
occasion for testing for absence of live hepa- 
titis A virus in the vaccine. 

Marmoset challenge. All marmosets were 
challenged intravenously 17 weeks after the 
first vaccine injection, with 1 ml of a 10"® 
dilution of CR326 hepatitis A virus contain- 
ing approximately ICr fifty percent marmoset 
infectious doses of virus. 

Results. I A antibody responses. Serum an- 
tibody titrations were performed on plasma 
samples collected at weekly intervals during 
the 17-week immunization regimen and the 
9-week period following challenge. Figure 1 


0037-9721 /l^/YS^l-Ol'CiWiVS^I^ 
AU hglhu reserved. 



shows that all eight animals displayed lA 
antibody after the sixth vaccine dose had 
been given (by 12 weeks), at least three of the 
animals having responded after the fifth dose. 
The titers ranged from 1:20 to 1:320. None of 
the animals given control vaccine developed 
hepatitis A antibody. One of the six animals 
in the viral inactivation test group that re- 
ceived a single dose of vaccine intravenously 
developed antibody by the 12th week after 
injection. Table I shows that none of the 
animals in any group developed positive ICD 
enzyme elevations prior to challenge indicat- 
ing that the materials given to the animals 
did not contain live hepatitis A virus. 





'-I — I — I — I—I — I — r-i — r— I — I — r — I — I — r— I — r— i — i — r— i 

6 7 • 9 10 II 12 13 14 \t 16 l7 10 19 20 2l 22 23 24 29 26 


Fig. I. Hepatitis A antibody responses in rufiventer 
marmosets as measured by immune adherence during 
the immunization and challenge regimens. 

Protective efficacy. The marmosets in all 
three groups were challenged intravenously 
with live CR326 hepatitis A virus during the 
17th week after vaccination was initiated, and 
the findings are shown in Fig. 1 and Table I. 
All eight of the marmosets that received the 
normal liver vaccine showed elevations in 
ICD and all developed lA antibody with 
titers ranging from 1:640 to 1:2560 or greater. 
By contrast, none of the animals given hep- 
atitis A vaccine showed ICD elevations and 
none showed more than twofold increase in 
antibody titer. Interestingly, only two of the 
six animals that were given a single dose of 
vaccine intravenously showed elevations in 
ICD, and these two animals developed lA 
antibody. One other animal developed pro- 
nounced lA antibody, without an ICD ele- 
vation. All of the three remaining animals 
appeared to be protected even though only 
one had lA antibody prior to challenge. 
These fmdings indicated that the vaccine 
given subcutaneously in multiple injections 
was highly effective in preventing experimen- 
tal hepatitis on challenge in marmosets and 
that even a single dose of vaccine given intra- 
venously afforded protection to live virus 
challenge in some animals. 

Discussion. The work on which the present 
findings are based represents the first dem- 
onstration that inactivated hepatitis A virus 
can afford protection against live hepatitis A 
virus challenge. Vaccine was given in eight 
divided aqueous doses, and it seems likely 
that protection might have been afforded fol- 
lowing fewer doses, especially if an immmu- 
nologic adjuvant had been employed. This 
vaccine might prove equally effective in pre- 
venting hepatitis A in man and might, there- 
fore, be of extreme importance in the control 
of the disease. The limited availability of 
marmosets and the lack of ability, to date, to 
achieve practical replication of the virus in 

TABLE I. Antibody and Enzyme Determinations in a Controlled Study of Human Hepatitis A Vaccine 

IN Marmosets. 



Before hepatitis A 

virus challenge 

After hepatitis A virus challenge 

MarmcHtct group 

Antibody responsie 
No. PoR./Total 

Enzyme elevation 
No PoR./Total 

Antibody response En/yme elevation 
No. Pos /Total No Pos./Total 

Protective eflkacy of vaccine 

Normal liver vaccine 
Hepatitis A vaccine 
Virus inactivation test 



8/8 8/8 
0/8 0/8 
3/6 2/6 

(partial protection by i.v. vaccine 



the laboratory precludes any substantial 
progress toward routine immunization in 
man at the present time. 

Summary, Human hepatitis A virus, par- 
tially purified from the liver of a rufiventer 
marmoset infected with CR326 strain virus, 
was inactivated with formalin and was shown 
to be highly potent in stimulating homolo- 
gous antibody in marmosets when adminis- 
tered subcutaneously at bi-weekly intervals 
in eight divided doses. The vaccine was 
shown to prevent hepatitis A in all marmosets 
when chidlenged with Uve hepatitis A virus 
in a controlled study. 

We gratefully acknowledge the technical assistance of 
Frank Banker, W. P. M. Fisher, Paula Giesa, Marilyn 
Johnston, and Paul Koser. Electron microscopy was 
performed by Dr. Bohdan Wolanski. 

1. Mascoli, C. C, Ittensohn, O. L., Villarejos, V. M., 
Arguedas, G., Arguedas, J. A., Provost, P. J., and 
Hilleman, M. R., Proc. Soc. Exp. Biol. Med. 142, 276 

2. Provost, P. J., Ittensohn, O. L., Villarejos, V. M., 
Arguedas, G., Arguedas, J. A., and Hilleman, M. R., 
Proc. Soc. Exp. Biol. Med. 142, 1257 (1973). 

3. Provost, P. J., Wolanski, B. S., Miller. W. J., Itten- 
sohn, O. L., McAleer, W. J., and Hilleman, M. R., 
Proc. Soc. Exp. Biol. Med. 148, 532 (1975). 

4. Provost, P. J.. Ittensohn, O. L., Villarejos, V. M., and 
Hilleman, M. R., Proc. Soc. Exp. Biol. Med. 148, 962 

5. MiUer. W. J., Provost, P. J., McAleer, W. J., Itten- 
sohn, O. L., Villarejos, V. M., and Hilleman, M. R., 
Proc. Soc. Exp. Biol. Med. 149, 254 (1975). 

6. Provost, P. J., Villarejos, V. M., and Hilleman, M. 
R., Proc. Soc. Exp. Biol. Med. 155, 283 (1977). 

Received May 24, 1978. P.S.E.B.M. 1978, Vol. 159. 


Structural Determinants of the Renal Tubular Activity of Vitamin D3 Derivatives: 
with la-Hydroxy. 24R,26-Dihydroxy. and 1a.24R.26-Trihydroxy Vitamin D3 


Renal- Electrolyte Section, Department of Medicine, Allegheny General Hospital and the University of P 
School of Medicine, Pittsburgh, Pennsylvania 15212 

Previous studies from this and other labo- 
ratories have documented an acute afTect of 
vitamin D3 and its major metabolites on renal 
tubular electrolyte transport (1, 2). The infu- 
sion of the biologically active metabolites of 
vitamin D3, 25-hydroxyvitamin D3 (25-hy- 
droxycholecalciferol, 25-HCC) and 1,25-di- 
hydroxyvitamin Da (l»25-dihydroxycholecal- 
ciferol, 1,25-DHCC), have been shown to 
produce an enhancement of phosphate, cal- 
cium, and sodium reabsorption both in the 
dog and rat (1-6). Recently, substantial prog- 
ress has been made in the identification and 
biochemical synthesis of other naturally oc- 
curring vitamin D metabolites as well as 
structural analogs. These advancements have 
provided us with the opportunity to study the 
renal tubular effects of additional metabolites 
and analogs of the parent compound and to 
identify the structural requirements of these 
vitamin D3 derivatives with regard to their 
transport actions. The data demonstrate that 
in order for an antiphosphaturia to occur, the 
derivative must contain a hydroxyl group in 
the 1 position. Furthermore, no effect on 
calcium or sodium transport is evident unless 
the compound possesses a 25-hydroxyl group 
which is sterically unhindered. 

Methods. Acute clearance studies were per- 
formed in female mongrel dogs weighing 16 
to 23 kg which had been thyroparathyroid- 
ectomized (TPTX) at least 48 hr prior to the 
experiment. Details of the surgical procedures 
and clearance technique have been reported 
elsewhere (1, 3). Completeness of parathy- 
roidectomy was verified by comparing the 
serum calcium concentration 2 to 4 days post- 
operatively to those just before the procedure. 
Animals with at least a 30% reduction in 
serum calcium were selected for study. Thy- 
roid replacement was accomplished by oral 
administration of 0. 1 to 0.2 mg of synthyroid 
(Flint) daily. The animals were fasted and 
thirsted for 16 hr before the study and re- 

ceived 5 U of vasopressin (Pitressi 
in oil, Parke, Davis and Company 
ning prior to the study. The dogs v 
thetized with 25 mg/kg of sodium 
bital with subsequent intermitten 
mental doses as required. A cuffed 
cheal tube was inserted and the anii 
ventilated with a Harvard respiratc 
ters were inserted into a hindlimb 
infusion of saline and into the extei 
lar vein for blood sampling. Primin] 
inulin and /7-aminohippurate (PAH 
jected and a sustaining infusion of ( 
stances was administered at a i 
ml/min in physiological saline 
Aqueous vasopressin was added to 
tion in an amount calculated to c 
mU/min. Volume expansion was p 
by infusing a 0.9% saline solution c 
1.0 to 1.5 mEq/liter of calcium { 
The total amount of saline infusec 
proximately 2.5% of animal bod 
after which urinary losses were re] 
adjusting the rate of infusion so 
expansion was sustained. Urine c 
were begun and after a steady 
achieved, one of the following exp 
maneuvers was performed. In f 
(group A, control animals), 0.25 i 
vehicle (propylene glycol) was injec 
venously as a bolus. The experiment 
continued for approximately 2 h 
which 10 to 12 clearance periods of 
mately 10 min each were obtainec 
other dogs (group B) 0.625 jig of la- 
vitamin D3 (la-hydroxycholecalcil 
HCC)^ dissolved in 0.25 ml of \ 
glycol was given according to the s 
tocol as described for the contn 
Group C consisted of five dogs wl 

' The la-HCC utilized in this study was 
supplied by Dr. Jack Hinman, Upjohn Con 
amazoo, Mich. 


Copyright 1978 by the Society for ExperimenuJ Biology and Medicine 



1 0.625 /ig of 24/?,25-dihydroxyvitamin 
(24/?,25-dihydroxycholecalciferol, 24/?, 
>HCCf dissolved in 0.25 ml of propylene 
)1. The animals in experimental group D 
ved 0.625 jug of la,24/?,25-THCC dis- 
td in 0.25 ml of propylene glycol. Blood 
drawn at the beginning of the study, at 
)Iateau of each steady state, before each 
rimental maneuver, and every 30 min 
ighout the study. Blood and urine were 
/zed for inulin, PAH, phosphorus, cal- 
if and sodium by methods previously de- 
ed from this laboratory (1). Serum ul- 
itrates were obtained by centrifuging se- 
handled anaerobically through CF-50 
ifuge cones (Amicon Corp., Lexington, 
i.). Statistical evaluation of the data was 
)nned by paired t test. 
jsults. Table I summarizes the data ob- 
;d in the control experiments (group A) 
sU as those in which the synthetic analog 
tamin D, la-HCC (group B) or the vi- 
n D metabolites, 24/?,25-DHCC (group 
r la,24/?,25-THCC (group D), were ad- 
stered intravenously. In the control ani- 
the intravenous administration of pro- 
ne glycol did not cause any changes 
;r in the absolute urinary excretion of 
5 ions or in their percentage excretion 
(Figs. 1-3, group A, Table I). Neither 
renal hemodynamics nor serum ultra- 
able calcium concentration (SUFca) al- 
I in any consistent manner. The acute 
inistration of 0.625 /xg of 24/?,25-DHCC 
no effect on either the absolute or per- 
ige excretion rates of phosphate, calcium, 
dium (group C, Table I). The mean delta 
0^ was -3.6 ± 3.9% (P > 0.40). However, 
la,24/?,25-THCC and la-HCC, when 
1 in the same amount (0.625 jug), induced 
nificant decline (by 28 and 30%, respec- 
y) in the percentage excretion of phos- 
e (/» < 0.01, < 0.05). The mean changes 
losphate excretion were —6.6 ± 1.6 and 
± 1.4%, respectively. These decrements 
accompanied by reductions in the ab- 
e excretion rates of phosphate of 18 and 
respectively, which were also statisti- 
significant {P < 0.05, < 0.02, Fig. 1). 
dgnificant change in either absolute or 

indly provided by Dr. Milan Uskokovic, Roche 
mtories, Nutley, N.J. 

percentage calcium or sodium excretion was 
observed in the animals receiving la,24/{,25- 
THCC and la-HCC (groups C and D, Table 
I, Figs. 2 and 3). Glomerular fihration rate 
and effective renal plasma flow as measured 
by the clearances of inulin and PAH, respec- 
tively, were unaltered. There was no statisti- 
cally signiflcant change in the level of serum 
sodium concentration in any of the groups. 
SUFca decreased slightly but consistently in 
group C (from 1.76 ± 0.08 to 1.70 ± 0.07 
mmole/liter) after 24/^,25-DHCC was ad- 
ministered. This change did not aflect either 
Altered load or excretion rate. In all of the 
other groups there was no significant change 
in either SUFc« or SUF phosphate. 

Discussion. Previous studies of the biologi- 
cal activities of 24/?,25-DHCC and la, 
24/?,25-THCC have been limited to an eval- 
uation of these substances in the skeleton and 
gastrointestinal tract. Furthermore, not only 
is there only a single study of the effects of 
la-HCC on the kidney (8) but none of the 
studies involving these compounds have been 
performed in the dog. The vitamin D3 deriv- 
atives utilized in this study were evaluated, 
therefore, with the following objectives in 
mind. First, it was our intention to attempt to 
establish their respective capacities to alter 
renal electrolyte transport and to compare 
these experimental observations with those 
previously obtained with 25-HCC and 1,25- 
DHCC (1,2). Second, the availability of these 
agents provided us with the opportunity to 
investigate what might be the structural de- 
terminants of vitamin D metabolites as re- 
gards their ability to alter electrolyte reab- 
sorption at the renal tubular level. 

These newly described derivatives of the 
parent vitamin have recently been shown to 
have substantial activity in stimulating intes- 
tinal calcium and phosphate transport (9-12). 
They are also active in elevating serum cal- 
cium and phosphorus and in the mobilization 
of calcium from bone in rachitic rats (12). 
However, in the latter systems, the response 
to 24/?,25-DHCC occurs after a considerable 
time lag (13). Furthermore, nephrectomy and 
a high calcium diet abolished the effect of 
24/^,25-DHCC on intestinal calcium trans- 
port (9, 13). This finding suggests that renal 
conversion of this metabohte to la,24/?,25- 
THCC or some other more polar metabolite 












a 3 2 2 ! 

•H -H -H -H 

00 CO Oh m < 

al^ Oh d ' 

2-^2 § 2 2 

I I V ■ > 

0,^0, ft, a, a,, 

S«^ OH (A ; 

<n -« -« < 

S « 








5 c. - *. ! 

r4 «*» IN IN < 

-H -H -H -H 

en 00 O O < 

00 HO d en ' 
2^2 2 2 2 

0,^0, a, a, a,, 

CM* f-. ^ «^ • 

m eo n n < 
-H -H -H -H 
r^ tn m 00 • 

p Si 2 d : 

a 3 

•H -H -H -H 

O Oh — 1^ < 

OH d oi i*J * 
2*^2 2 2 2 
I I V ■ > 

<N ^ ^ < 

•H -H 



•H -H -H -H 

•A O rn Oh « 

r^' 00' d 00 < 

■* CO 2 5« ^ 5^ ^' 

2 2 2 2 
I I I I 

0,^0, ft, ft, 
O f>j o o < 

>o <N ^ oi < 
-H -H -H -H 

1^ >o ^ ^ < 

s 1 




32 - 





24 - 



1. 1 ^^**^ 



▼ -^ 





Ti : 





i i 

12 - 




2*.29 0MCC 

4 - 


l.2*.l9 TMCC 

I. The effects of vitamin Da raeuboliles and la- 
1 the absolute excretion rate of phosphate. Note 
lificant reduction in phosphate excretion follow- 
administration of both la-HCC and la,24/?,25- 
24/?,25-HCC was without effect on urinary phos- 
xcretion. Data points represent the mean values 
or all dogs before (C) and after (E) administration 
: compounds. 







4 - 

A l«MCC 

• 24.ZS0HCC 



C E 

2. The effect of vitamin Da metabolites and \a- 
1 urinary calcium excretion. All three compounds, 
DHCC, la,24/?.25-THCC. and la-HCC, were 

effect on the renal tubular handling of calcium. 

iircd for its biological activity to be 
scd (13). Indeed, radioactive 24/?,25- 
Z has been shown to be further metab- 
in vivo as well as in vitro to la,24/?,25- 
' (14). It seems likely that hydroxyl- 
it the 1 position is a prerequisite step in 
etabolism of 24/?,25-DHCC for this 
mnd to become biologically active, 
results on renal phosphate excretion 
31 the observations obtained in the in- 
il transport system that hydroxylation 












; 7 










J. T 





500 - 

A i«NCC 

• t4.29 0MCC 



C E 

Fig. 3. The effects of vitamin D3 metaboUtes on the 
urinary excretion of sodium. Note that none of the agents 
(la-HCC, 24/?,25-DHCC, or la,24/?,25-THCC) changed 
urinary sodium excretion. 

at the 1 position is necessary for the manifes- 
tation of its biological activity. Thus, a com- 
mon feature of all the compounds which in 
the present study effected a reduction in phos- 
phate excretion, was the presence of the la- 
hydroxy configuration. Since glomerular fil- 
tration rate and effective renal plasma flow, 
as well as the filtered load of phosphorus 
were unchanged (Table I), it is reasonable to 
conclude that the changes observed were due 
to a direct action of these compounds on the 
renal tubule. Furthermore, we conclude from 
these findings that the action of vitamin D 
metabolites on the tubular transport of phos- 
phate depends upon the presence of the la- 
hydroxyl configuration. Since, in earlier stud- 
ies from this laboratory, 25-HCC was likewise 
very effective in reducing phosphate excre- 
tion (1), we infer that the latter metabolite 
was converted in vivo to another vitamin D3 
derivative containing a la-hydroxyl group. 
This most probably means the formation of 
la,25-DHCC or some other "tissue active" 
substance, as yet unidentified (2). The fact 
that 25-HCC does not act immediately on 
renal electrolyte transport (1) and requires 
the "permissive" effect of either parathyroid 
hormone (4, 7) or vasopressin (6) for its renal 
tubular effects to become evident, are sup- 
portive of this thesis. 

Perhaps the most important synthetic an- 
alog of la,25-DHCC currently available, at 
least from a therapeutic standpoint, is la- 



HCC. This compound is almost as potent as 
la,2S-DHCC in stimulating intestinal cal- 
cium transport in the chicken (IS) and has 
approximately two to five times the activity 
of vitamin D3 on calcification of the skeleton 
and in stimulating gut absorption of calcium 
in the rat (16). It has recently been reported 
that the intravenous administration of la- 
HCC to the rachitic rat produces an enhance- 
ment of intestinal calcium absorption within 
1 hr of its infusion (17). This extremely rapid 
onset of action suggested to the investigators 
that la-HCC might act directly on the cell 
membrane transport of calcium. However, 
studies by Zerwekh et al (18) suggest that the 
action of la-HCC requires prior conversion 
to la,2S-DHCC. In addition, it has been dem- 
onstrated that tritiated la,2S-DHCC appears 
in the intestine and bone within 2 hr after 
intravenous administration of la-[6-^H]hy- 
droxy vitamin D3 (19). The design of the 
acute clearance studies presented in this re- 
port was such that observations were made 
for only 2 hr following the administration of 
the vitamin Da derivatives. Thus, since fur- 
ther metaboUc conversion of both la-HCC 
and 24/{,2S-DHCC appears to require longer 
than 2 hr, we presume that the observed 
changes in renal transport were due to the 
action of the unchanged compounds. As re- 
gards the effects of la-HCC, our results con- 
firm the observations of Pechet and Hesse (8) 
and Toflfolon et al (17) that la-HCC has a 
very rapid onset of action. Of course, we 
cannot rule out the possibility of some (more 
rapid) metabolism of these substances to an 
as yet unidentified 'tissue-active" metaboUte 
or metaboUtes. 

Unlike 25-HCC and 1,25-DHCC (1, 2) 
none of the vitamin D3 derivatives examined 
in the current study (la-HCC, 24/^,25- 
DHCC, or la,24/^,25-THCC) were effective 
in altering either sodium or calcium excretion 
when given acutely. While no explanation of 
these observations is conclusively provided 
by the data, the fmdings could be explained 
as follows. The la-hydroxylated compounds 
(la-HCC and la,24/?,25.THCC) may act at 
different sites within the nephron or other 
receptor molecules than those affected by 25- 
HCC and la,25-DHCC. Alternatively, it may 
be that in order for a compound to alter 
calcium and sodium reabsorption, it must 

have a hydroxyl group in the 25 pc 
in both the 1 and 25 positions. Furtl 
it appears that the 25-hydroxyl group 
also be sterically unhindered. Inde< 
et al recently presented evidence tha 
DHCC was rather less potent than c 
HCC or 24/^-HCC in its ability to ef 
resorption (20). They proposed that i 
factors, steric hindrance, or an exce 
drophilic groups in this region ma] 
the decreased activity of this compoi 
ther study will be required to elucid^ 
of the above postulated mechanisms 
ers) explain the experimental obsc 

Summary. The acute effects of 
DHCC, la,24/?,25-THCC, and la- 
the renal handling of phosphate, calc 
sodium were evaluated in the TF 
which had been mildly volume expai 
infused with vasopression to estabUs! 
phaturia. Both la-HCC and \c 
THCC when given intravenously in 
of 0.625 jLig produced a significant de 
urinary phosphate excretion. P< 
phosphate excretion decreased by 
28%, respectively (P < 0.05, < 0.0 
there was no alteration in renal hem 
ics or in the filtered load of this ion, 
suggest a direct action of these coi 
on renal tubular transport mechan 
change in the urinary excretion of c 
cium or sodium was observed folio 
administration of the two vitamin E 
tives. 24/?,25-DHCC was without < 
the renal handling of all three ions. 

When previous experimental fin 
garding the renal actions of 25-M 
la,25-DHCC arc considered, the dat 
that the 1 -hydroxyl grouping is req 
the metabolites of vitamin D to ; 
phosphate transport at the renal tubi 
It appears that a sterically unhind 
hydroxyl group is necessary in ordc 
vitamin D^ derivatives to act on the 1 
tion of either calcium or sodium. 

This work was supported by grants from tl 
Institutes of Health (AM 17,575 and RR I 
National Science Foundation (PCM 77-0905 
Western Pennsylvania Heart Association, and 
ing grant from the U.S. Public Health S< 
05905). We appreciate the technical assista 
Kenneth Swint, Mrs. Roberta ShefTer, and 



t, J. B., Moranz, J., and Kumick, W. S., J. 

vest. 51,373(1972). 

U J. B., Fernandez, P. C, Boyle, I. T.. Gray, 

Amdahl, J. L., and DeLuca, H. F., Proc. Soc. 

ol. Med. 141,379(1972). 

I, J. B., Beck, W. S., Jelonek, A., and Fer- 
P. C, J. Clin. Invest. 53, 756 (1974). 

«r, M. M., Robinetle, J. B., DcLuca, H. F., 

lick, M. F., J. Clin. Invest. 53, 913 (1974). 

». L. S., Sheehe, P. R., and Weiner, I. M., 

. Physiol. 22d, 1490(1974). 

1 1., Szramowski, J., and Puschett, J. B., Min. 

clab. 1,48(1978). 

t, J. B.. Beck, W. S., and Jelonek, A., Science 


M. M., and Hesse, R. H., Araer. J. Med. 57, 


, M. W., Hartenbower, D. L., Cobum, J. W., 

rman, A. W., Arch. Biochem. Biophys. 182, 


4. L., Norman, A. W., Taylor, A. N., and 

owcr, D. L., J. Nutr. 106, 724 (1976). 
Y., DeLuca, H. F., Ikekawa, N., Morisake, 
Koizumi, N., Arch. Biochem. Biophys. 170, 


12. Miravet, L., Redel, J., Carre, M., Queille. M. L., and 
Bordicr. P.. Calcif. Tiss. Res. 21, 145 (1976). 

13. Boyle, I. T., Omdahl, J. L., Gray, R. W., and De- 
Luca, H. F., J. Biol. Chem. 248, 4174 (1973). 

14. Holick, M. F., Kleiner- Bossaller. A.. Schnoes, H. K., 
Kasten, P. M., Boyle, I. T., and DeLuca, H. F., J. 
Biol. Chem. 248, 6691 (1973). 

15. Haussler, M. R., Zerwkh, J. E., Hesse, R. H., Riz- 
zardo, E., and Pechet, M. M., Proc. Nat. Acad. Sci. 
USA 70, 2248 (1973). 

16. Holick. M. F., Kasten-Schraufrocel, P., Tavela, T., 
and DeLuca, H. F., Arch. Biochem. Biophys. 166, 

17. Toffolon, E. P., Pcchct, M. M., and Issclbachcr, K.. 
Proc. Nat. Acad. Sd. USA 72, 229 (1975). 

18. Zcrwckh, J. E.. Brumbaugh, P. F.. Haussler, D. H., 
Cork, D. J., and Haussler, M. R., Biochemistry 13, 
4097 (1974). 

19. Holick, M. F.. Tavela, T. E.. Holick, S. A.. Schnoes, 
H. K.. DeLuca, H. F.. and Gallagher, B. M., J. Biol. 
Chem. 251, 1020(1976). 

20. Stem, P. H., DeLuca, H. F., and Ikekawa, N., Bio- 
chcm. Biophys. Res. Commun. 67, 965 (1975). 

Received January 3, 1978. P.S.E.B.M. 1978, VoL 159. 


Stimulation of Growth Hormone Release by Intraventricular Administration of 5HT or 
Quipazine in Unanesthetized Male Rats^ (40316) 


Department of Physiology, University of Texas Health Science Center at Dallas, 5323 Harry Hines Boulevard, 

Dallas, Texas 75235 

Although there is considerable evidence in 
favor of a stimulating role of the central 
serotoninergic system on the secretion of GH 
in man and nonhuman primates (1-4), little 
information is available in other species (5). 
CoUu et al (6) reported that intraventricular 
injection of serotonin stimulated GH secre- 
tion in rats anesthetized with urethane and 
that the effect was abolished by phcnoxy- 
benzamine, an a receptor blocker. Since ex- 
periments on animals in deep urethane an- 
esthesia are opened to some criticism, we 
have investigated, in the present work, the 
effect of intraventricular administration of 
serotonin in unanesthetized unrestrained rats 
and compared them with the effect of intra- 
ventricular administration of the serotonin 
receptor agonist quipazine (7). 

Materials and Methods, Adult male rats of 
the Sprague-Dawlcy strain (Simonsen Lab- 
oratories, Gilroy, California) were used. They 
were housed under controlled conditions of 
lighting (light on from 0500 to 1900 hr) and 
temperature (24 ± 1°) with free access to 
food and water. After 2 weeks of adaptation 
in our animal facility, a 23-gauge stainless- 
steel cannula was implanted into the third 
ventricle and 1 week later the animals were 
fitted with Silastic intravenous catheters as 
described earlier (8, 9). On the day of the 
experiment, usually 2 days after implantation 
of the intravenous cannulas, the animals were 
transferred in their cages into a quiet labo- 
ratory and polyethylene extension tubes 
(PESO, 12 in. long) filled with a solution of 
heparin in 0.9% NaCl were attached to the 
distal end of permanent iv cannulas. Thirty 
to sixty minutes later a preinjection blood 
sample (0.6-0.8 ml) was withdrawn; then the 
intraventricular injection was performed and 
postinjection samples (0.6-0.8 ml) were taken 

' Supported by Grants AM 10073 and HD09988. 
^ On leave of absence from Delhi University, India. 

at 10, 30, and 60 min into heparinized sy- 
ringes. The volume of each sample was re- 
placed immediately after each bleeding by an 
equal volume of 0.9% saline. 

The intraventricular injections were per- 
formed according to the procedure described 
earlier (8, 9): Serotonin (serotonin creatinine 
sulfate complex, Calbiochem) or quipazine 
maleate (gift of Miles Labs, Inc.) were freshly 
dissolved in 0.9% NaCl; the pH was adjusted 
to 5.5 and then administered into the ventricle 
in a volume of 2 /xl. The dosage of 5HT is in 
terms of the free base. Controls for the 5HT- 
treated animals received 40 jug of creatinine 
sulfate while controls to quipazine were in- 
jected with 0.9% NaCl. In all cases the intra- 
ventricular injection was given over a period 
of approximately 60 sec. 

In two experiments the animals were in- 
jected with serotonin receptor blocker, meth- 
ysergide maleate (gift of the Sandoz Labora- 
tories), 10 mg/kg ip, 60 min before the intra- 
ventricular administration of either 5HT or 

After centrifugation of the heparinized 
blood samples, plasma was collected and 
stored frozen at —20° until assay. Concentra- 
tion of GH in the samples was determined by 
the NIAMDD radioimmunoassay system for 
rat GH.^ All samples were measured in du- 
plicates at two different dilutions. The results 
are expressed in nanograms per milliliter in 
terms of the RP-l GH standard provided 
with the kit. 

The statistical significance of the results 
was evaluated by the paired / test for sequen- 
tial changes within the same group and by 
Student's / test for differences between two 
groups for a particular time. 

Results and Discussion, Intraventricular in- 
jection of 5HT, 4 or 20 /xg, caused a significant 

^ Kits for determination of GH were provided through 
the NIAMDD-NIH Pituitary Hormone Program. 


^78 by the Society for ExpenmentA\ Biologv and Medicine 




a of plasma GH levels, which was 
t at 10 min and persisted throughout 
ition of the experiment (Table I). The 
)onse was related to the dose of SHT 
he peak levels at 30 min were signif- 
digher in animals receiving 20 jug of 
m in the animals injected with 4 jug. 
1 of creatinine sulfate to control ani- 
I not influence plasma GH. Pretreat- 
' the rats with methysergide had no 
e on the preinjection GH levels, but 
letely aboUshed the GH-stimulating 

intraventricular SHT. 
izine also induced elevation of 
GH. However, in comparison with 
ct of SHT, the secretory responses 
layed and a dose-related increase ap- 
)nly at 30 min after administration of 
g which persisted until the 60-min 
i of the experiment (Table II). The 
ory effect of quipazine was abolished 
eatment of the animals with methy- 

Intraventricular administration of 
iCl in the group of controls had no 
1 plasma GH. 
Iiink that our results provide strong 

evidence that activation of the central sero- 
toninergic system promotes secretion of GH 
in the rat. This conclusion is most directly 
supported by the GH-stimulating effect of 
intraventricular administration of SHT. Sup- 
pression of the effect of SHT with serotonin 
receptor blocker, methysergide, lends addi- 
tional support to this conclusion. 

There is considerable evidence that quipa- 
zine activates the central serotoninergic sys- 
tem (10-12) probably by a combination of 
several effects, which include activation of 
serotonin receptors, inhibition of serotonin 
reuptake by serotoninergic nerve terminals, 
and possibly enhanced release of serotonin 
(7, 13-15). It is, therefore, highly probable 
that the stimulation of GH secretion follow- 
ing intraventricular administration of quipa- 
zine originated in the activation of the central 
serotoninergic system. The similarity between 
the effect of SHT and quipazine as well as 
the fact that the effect of both drugs was 
suppressed by methysergide also speaks for 
this conclusion. 

Difficult to explain is our observation that 
the GH-stimulating effect of quipazine was 

E I. Third Ventricular Injection of Serotonin or Systemic Administration of Methysergide 
ED BY Intraventricular Serotonin on Plasma GH Levels (Nanograms per Milliliter of Plasma). 

Time after injection (min) 

Treatment and dose 





;reatinine sulfate, 40 /xg (7)° 


gidc.* 10 mg/kg, -♦■ serotonin, 


33.2 ± 1.5 

31.3 ±3.2 
27.8 ± 4.4 
24.5 ± 4.6 

28.9 ± 4.0 
54.7 ± 9.0» 
17.5 ± 6.3 

30.8 ± 2.8 
54.6 ± 6.0* 
103.6 ± 6.5» 
25.8 ± 3.2 

34.0 ± 2.0 
55.6 ± 12.0» 

55.6 ± 3.4» 

38.7 ± 7.5 

ber of rats per group. 

ysergide was given ip in a volume of 0. 1 ml of saline 1 hr before third ventricular injection. 

).05 vs preinjection level. 

0.001 vs preinjection level. 

E II. Third Ventricular Injection of Quipazine or Systemic Administration of Methysergide 
-LOWED BY Intraventricular Quipazine on Plasma GH Levels (Nanograms per Milliliter). 

Time After injection (min) 

ment and dose 





.-2 Ml 



gide, 10 mg/kg, + 


27.0 ± 3.3 
31.0 ± 1.5 

33.5 ± 2.3 

30.6 ±2.1 

30.6 ± 1.1 

27.2 ±2.1 

26.3 ± 2.3 
23.6 ± 1.2 

29.8 ± 3.2 
62.6 ± 8.9» 
105.4 ± 2.6»* 
26.0 ± 0.6 

30.3 ± 0.8 
67.2 ± 4. 1» 
80.6 ± 9.2*» 
27.0 ± 0.6 

Table I. 

ysergide was given as in Table 1 
).05 vs preinjection level. 
0.(X)1 vs preinjection level. 



delayed as compared with the effect of 5HT. 
This delay is probably not caused by different 
pharmacodynamic properties of quipazine, 
because both drugs induce activation of pro- 
lactin secretion, attaining peak levels 10 min 
after intraventricular administration (unpub- 
lished results). It is, therefore, possible to 
speculate that quipazine, in addition to acti- 
vation of the central serotoninergic system, 
may have a short-lasting effect of another 
kind which is inhibitory to GH secretion. 

Our results obtained with the intraventric- 
ular administration of SHT in unanesthetized 
free-moving animals confirm the earlier work 
of CoUu et al, (6) on animals anesthetized 
with urethane. To our knowledge this is the 
first report on the GH-releasing effect of 

Summary. Intraventricular injection of 
SHT (4 and 20 /xg) in unanesthetized, unre- 
strained male rats fitted with permanent in- 
trajugular cannulas for withdrawal of blood 
samples caused a dose-related elevation of 
plasma GH levels. Similar effects were also 
observed following intraventricular injection 
of the serotonin receptor agonist, quipazine. 
The GH-releasing effect of both drugs was 
abolished by a serotonin receptor blocker, 
methysergide. It is concluded that activation 
of the central serotoninergic system stimu- 
lates GH secretion in the rat. 

The authors wish to thank Mrs. Shirlee Barnes for 
secretarial assistance. 

1. Imura, H., Nakai, Y., and Yoshimi, T., 
Endocrinol Metab. 36, 204 ( 1973). 

2. Miiller, E. E., Brambilla, F.. Cavagnini, F., 
M., and Panerai, A., J. Clin. Endocrinol. ^ 
1 (1974). 

3. Smythe, G. A., and Lazarus, L., J. Clin. Ii 

4. Bivens, C. H., Lebovitz, H. E.. and Feldm; 
N. Engl. J. Med. 289, 236 (1973). 

5. Martin, J. B., in "Frontiers in Neuroendoc 
(L. Martini and W. F. Ganong, Eds.), Vol. 
Raven Press, New York (1976). 

6. Collu, R., Fraschini, F., Visconti, P., and M 
Endocrinology 90, 1231 (1972). 

7. Fuller, R. W., Snoddy, H. D., Perry. K. W. 
B. W., Mollay, B. B., Bymaster, F. P., and 
T., Life Sci. 18,925(1976). 

8. Vijayan, E., and McCann, S. M., Neurocr 
ogy25, 150(1978). 

9. Vijayan, E., and McCann, S. M., Neurocr 
ogy 25, 21 1(1978). 

10. Rodriguez, R., Rojas-Ramirez, J. A., and 
Colin, R. R., Eur. J. Pharmacol. 24, 164 (I 

1 1. Medon, P. J., Leeling, J. T., and Phillips, B 
Sci. 13,685(1973). 

12. Samamin. R., Bemasconi, S., and Quatt 
Psychopharmacology 46, 219 ( 1976). 

13. Jacoby, J. H., Howd, R. A., Levin. M 
Wurtman, R. J., Neuropharmacology 15, 5 

14. Green, A. R., Youdim, B. H.. and Graha 
D. G., Neuropharmacology 15, 173 (1976) 

15. Hamon, M., Burgoin, S., Enjalbert, A., Bo 
and Hery. F., Nannyn-Schmiedebert*s Af 
macol. 249,99(1976). 

Received June 9, 1978. P.S.E.B.M. 1978, Vol. I 


Effects of Ethanol on the Absorption and Retention of Lead (4031 7) 


of Hematology and Oncology, University of Alabama in Birmingham, Birmingham, Alabama 35294, and 
Veterans Administration Hospital, Birmingham, Alabama 

equent clinical association of plumb- 
increased alcohol intake has sug- 
[lat ethanol may augment lead ab- 
and toxicity. This investigation was 
:en to determine the effects of acute 
>nic ethanol administration on lead 
3n and excretion. 

ials and methods. Male albino rats of 
5en-free Wistar strain weighing 200 
at the time of absorption measure- 
intravenous lead injection were used 
periments. The principles of labora- 
nal care as promulgated by the Na- 
esearch Council were observed. All 
were housed in polypropylene cages 
ag absorbent bedding in a room pro- 
ith automatically controlled temper- 
id lighting. The rats were given a 
[ pelleted laboratory chow (Wayne 
X, Allied Mills, Inc.) fed ad libitum, 
ralized deionized water was supplied 
imals except in some experiments in 
yjh ethanol (v/v) was substituted for 

absorption studies were performed 
urement of total body radioactivity 
ill animal whole-body liquid scintil- 
jtcctor (Packard-ARMAC). The ra- 
pes utilized were obtained from New 
Nuclear as ^^Pb acetate (sp act 
iCi/mg of lead) or ^*^Pb nitrate (sp 
iCi/mg of lead). Because of its half- 
l years, ^*°Pb was selected for use in 

1 studies only; ^^Pb, having a half- 

2 days, was used for all other exper- 
All measurements of radioactivity 
Tccted for radiodecay by comparison 
»propriate standard afler subtraction 
;round radioactivity. Lead absorption 
ents were performed in rats fasted 
It from food but not fluids. Under ip 
rbital anesthesia (4 mg/100 g), the 
was tied with a silk suture to prevent 
loss of absorbed lead. A laparotomy 
formed, the small intestine was iso- 

lated proximally and distally with umbilical 
tape, and the bile duct was ligated with silk 
suture. One milliliter of radiolabeled lead- 
containing test solution was injected into the 
isolated intestinal segment. Injections were 
accomplished by entering the gut lumen 
proximal to the proximal Ugature with a 21- 
gauge hypodermic needle, passing it intralu- 
minally through the ligature loop, tightening 
the ligature, and then injecting the test dose 
into the isolated segment with subsequent 
withdrawal of the needle and tying of the 
ligature. In one experiment, animals were 
administered test doses through an oroeso- 
phageal tube following laparotomy. The ab- 
domen was then closed with stainless-steel 
clips and the tats were placed in 1 -quart 
vented cardboard ice cream containers. Total 
body radioactivity was measured in a whole- 
body detector and compared to a 250-ml 
water-filled plastic bottle containing a test 
dose equal to that injected into the animals. 
Four hours after administration of the test 
dose, each animal was killed by cervical dis- 
location. Isolated intestinal segments were 
excised from the carcass and whole-body ra- 
dioactivity was again quantified and com- 
pared to the original whole-body radioactiv- 

To assess the effects of chronic ethanol 
ingestion on lead absorption, a group of eight 
animals was given 10% ethanol (v/v) as their 
exclusive source of fluids for 3 weeks while 
eight controls received water. The rats took 
food and fluids readily. While animal weights 
in experimental and control groups were ini- 
tially the same (140 ± 5 and 141 ± 7 g, 
respectively), weight gain in ethanol-treated 
animals was less than that of controls (62 ± 
5 vs 84 ± 15 g, /? < 0.05); isocaloric pair 
feeding also results in a similar diminution in 
weight gain in animals receiving ethanol (1). 
Animals were given a test dose of i jug of Pb 
and absorption was determined. Specimens 
of duodenum and liver were taken from ad- 


AU hghu reserved. 



ditional similarly prepared experimental and 
control animals for light and electron micro- 
scopic studies. The influence of acute alcohol 
ingestion was studied by the quantification of 
lead absorption in groups of animals receiv- 
ing 1 ml of the following solutions in 50% 
ethanol: (1) 1 Mg of Pb; (2) 10 /ig of Pb; (3) 
100 /ig of Pb; and (4) 1 mg of Pb. Controls 
received the same quantities of lead in 
aqueous solutions. Segments of duodenum 
from similarly prepared experimental and 
control rats were examined by light and elec- 
tron microscopy. 

To determine whether the diminished lead 
absorption from ethanol solutions was due to 
a direct effect on the intestine, 16 rats were 
given 1 pig of Pb in the isolated intestinal 
segment. Half the animals simultaneously re- 
ceived 1 ml of 50% ethanol above the pyloric 
ligature by oroesophageal intubation. Control 
animals received 1 ml of saline. In an addi- 
tional experiment, intestinal loops with open 
distal ends were injected with 50% ethanol 
followed after IS min by lavage with 0.5 ml 
of air and 1 ml of saline and subsequent tying 
of the cecal ligature. Intestinal loops in con- 
trols were pretreated with saline followed by 
similar washing. Lead absorption experi- 
ments were then performed using 1 /ig of Pb. 
In a final study to assess the role of the site of 
absorption in lead absorption, six groups of 
eight animals were subjected to laparotomy 
with bile duct and cecal ligation. The rats in 
each group received 1 /ig in water or 50% 
ethanol (pH 4) by the following means: (1) 
aqueous and (2) alcoholic lead via oroeso- 
phageal tube, the solution confined to the 
stomach by a pyloric ligature; (3) aqueous 
and (4) alcoholic lead via oroesophageal tube 
without pyloric ligation; (5) aqueous and (6) 
alcoholic lead in isolated gut loop. Lead ab- 
sorption was then quantified as previously 

To study the effects of aqueous and alco- 
holic solutions on lead solubility, 100 ml of 
each of the following solutions were prepared 
as controls: (1) 1 Mg of Pb/ml; (2) 10 /ig of 
Pb/ml; (3) 100 /ig of Pb/ml; and (4) 1 mg of 
Pb/ml. Similar solutions of lead in 50% 
ethanol were also prepared. One microCurie 
of ^Pb was added to each 100-ml solution 
and the pH was adjusted to 2.0. After mixing, 
1 ml of each of the resulting solutions was 

removed as a standard. Each solut 
titrated against 0. 1 N NaOH to pH I 
2-ml samples being removed at each 
pH value. Similar samples were tak 
returning to pH 2 with 0.1 N HCl. 
taken from each titration were centri 
3000 rpm x 30 min and 1 ml of sup 
was removed from each for quantifi< 
radioactivity in a Packard auto-gami 
trometer. Solubility of aqueous and i 
lead was expressed as percentage of: 
as a function of pH. Supematar 
ethanol solutions varying from aque 
trols by more than 5% were applied 
adex G-25 columns equilibrated m 
ethanol at the appropriate pH. Colu 
tions were counted successively t 
peaks of radioactivity which would 
the presence of soluble lead-containi 

In lead excretion experiments, ex] 
tal animals were given 10% ethanol fi 
ing for 3 weeks prior to injection and 
received water. Ethanol was continu 
experimental group throughout ( 
measurements. While similar weight 
perimental and control rats at the 
fluid conditioning (143 ± 7 g, 141 ± : 
again noted, ethanol-treated animals 
less at the time of injection (205 ± I 
± 9 g, /7 < 0.05). Each animal was 
sodium pentobarbital anesthesia (4 
g) to facilitate injection of 1.0 /iCi oi 
0.5 ml of 0.9% NaCl (pH 7.4) into tl 
vein of the penis. Whole-body cou 
obtained immediately after dosing a 
tervals thereafter. Body retention of r 
was calculated by comparison ti 
counts with correction for decay by 
ison to a standard. At the termin 
excretion studies 4 weeks after dosin 
whole-body radioactivity was measu 
sues for electron microscopic stud 
fixed in 2% glutaraldehyde, postfix< 
osmic acid, and embedded in Arald 
tions 150 to 200 A thick were staii 
saturated uranyl acetate and lead cit 
examined using an RCA EMU4 elec 
croscope. Thick sections (1 /im) wer 
with toluidine blue. Additional sped 
light microscopy were fixed in 10% i 
paraffin embedded, and stained wit 
toxylin and eosin. 



absorption and excretion experiments 
of eight animals were used. Except as 
ibove for animals receiving ethanol 
ally, there were no differences in 
inimal weights among the various 
compared in this study. For absorp- 
dies, all rats received I ml of a lead- 
ing solution adjusted to pH 4.0, which 
zs the pH of gastric contents and 
ns lead solubility. All lead quantities 
ressed as grams of elemental lead as 
1 /iCi of ^Pb or 2^^Pb was used as a 
(topic label for each rat. Data are 
;d as means and standard errors of 
in. Statistical comparisons were made 
tudent's two-tailed / test for unpaired 

h. Chronic ethanol ingestion signifi- 
educed the absorption of a single dose 
ous lead. While animals receiving wa- 
i fluid source for 3 weeks absorbed 
1.6% of a test dose of 1 jug of Pb, those 
d with 10% ethanol for 3 weeks ab- 
only 15.2 ± 2.9% (p < 0.05). While 
rom animals receiving alcohol for 3 
tiowed moderate fatty change, no light 
opic or uhrastructural changes were 
1 duodenal mucosa of the same ani- 
s illustrated in Fig. 1, the absorption 
from aqueous solutions was signifi- 
jreater than that from ethanol solu- 
i concentrations of 1 and 10 jug of 
[p < 0.005, p < 0.005). At lead con- 
ons of 100 and 1 mg of Pb/ml, ab- 
I from alcoholic solutions appeared 
greater than from controls but the 
ices were not significant (p = 0.20, p 
. Duodenal mucosa from animals re- 
50% ethanol acutely with or without 
owed disruption of villous tips, pyk- 
uclei, and increased villous invasion 
lonuclear cells. By electron micros- 
^struction of microvilli, mitochondrial 
5, and irregularity of mitochondrial 
e noted in addition. No abnormaUties 
ted in animals given saline or aqueous 
lutions. The solubility of ^Pb in 
s solutions is shown in the upper half 
2. Lead is more soluble in acid solu- 
d increasing amovmts are precipitated 
ncreases. As illustrated in the lower 
Fig. 2, little change in radiolead sol- 
Kxurs when 50% ethanol is used as a 



rO pg roo pg 
Test Dose 
Fig. 1. The acute effects of ethanol administration 
on the absorption of a single dose of lead chloride. 

carrier. Application of supematants obtained 
in these experiments to Sephadex G-25 col- 
unms revealed no evidence of lead-containing 

Since animals receiving ethanol on both an 
acute and chronic basis appeared to have 
diminished lead absorption unattributable to 
reduced lead solubility or macromolecule for- 
mation in the presence of alcohol, additional 
experiments were performed to determine 
whether at least part of this inhibitory effect 
was due to a direct effect of ethanol on intes- 
tinal mucosa. Rats with a pyloric ligature 
simultaneously administered 1 /ig of Pb in the 
intestinal loop and 50% ethanol in the stom- 
ach showed lead absorption (Fig. 3) which 
did not significantly vary from that observed 
in control animals. Absorption of lead in 
animals with ethanol-pretreated intestinal 
loops, however, was significantly less than 
that seen in rats with saline-pretreated gut 
loops (3.9 ± 0.5 vs 13.5 ± 1.5% control, p < 
0.0005). As shown in Table I, only small 
quantities of aqueous or alcoholic lead were 
absorbed by the stomach (2.5 ± 0.7 and 2.1 
± 0.4%, respectively). The absorption of lead 
in aqueous solution by the intestine (30.6 ± 
1.5%) was significantly higher than that of 
lead in alcoholic solution (8.2 ± 0.8%, p < 
0.005) and is similar to findings shown in Fig. 
1. When alcoholic lead solutions were given 
via oroesophageal intubation such that both 
stomach and intestine could act as absorptive 
sites, lead absorption increased to 22.4 ± 
3.2%. This value was less, however, than lead 






2468 10 8642 
100 M PbOi/n* H2O 

2468 K)8642 

\mqnC\t/n* HjO 

pH pH 

Fio. 2. The solubility of ^Pb in PbCl2 as affected by pH. Upper diagrams indicate solubility in aqueous 
solutions; lower diagrams indicate solubility in 50% ethanol. 

uptake from aqueous solutions (28.6 ± 1.7% 
p < 0.05). The findings suggest that a gastric 
factor, perhaps ethanol-stimulated gastric 
acid, may act to modify lead absorption al- 
though in these experiments alcoholic lead 
uptake remained significantly less than con- 

As seen in Fig. 4, the excretion of lead in 
fltiimak chronically receiving 10% ethanol 
did not significantly vary from control ani- 
mals at any time during the experiment. Both 
groups showed an initial rapid phase of lead 
elimination during the first week after injec- 
tion, in which time about one-half of the 
administered dose was excreted. This was 
followed by a slower phase of lead loss from 
the body. By using a best fit slope derived 
from mean-square analysis plotted on semi- 
logarithmic graph, the half excretion time for 
lead remaining after Day 7 was approxi- 
mately the same for each group, about 160 

Discussion, A variety of clinical reports of 
lead poisoning in heavy consumers of alcohol 
(2-4) has suggested that ethanol may enhance 
lead accumulation and potentiate its toxic 
manifestations. While Mahaffey et fl/. ( 1) con- 
cluded that there was little synergism of 
ethanol and lead as measured by morphologic 
and biochemical parameters of lead toxicity, 
no studies to date have directly measured the 
effects of alcohol on lead absorption or reten- 
tion. The results of this study indicate that: 
(i) the acute and chronic administration of 
ethanol inhibits the ability of the rat small 
intestine to absorb lead; (2) the effect does 
not seem attributable to diminished lead sol- 
ubility in alcohol; (3) the inhibitory effect 
may be related, in part, to the direct toxicity 
of ethanol on intestinal mucosa; and (4) 
chronic ethanol ingestion does not appear to 
alter the excretion of lead given as a single 
intravenous dose. While the mechanism of 
lead absorption is unknown, Krawitt (S, 6) 



ided that acute or chronic ethanol ad- 
tration inhibited calcium transport in 
d rat gut sacs and that this effect was 
ated with direct mucosal toxicity. Since 
ice exists that one or more intestinal 
ns important in calcium mucosal bind- 
id transfer may participate in lead ab- 
on (7), a similar direct toxic effect on 
inal mucosa may be responsible for the 
ished lead absorption found in these 
Lments. Whether the anatomic damage 
xlenal mucosa observed in these exper- 
s after acute ethanol administration is 
isible for the diminished lead absorp- 
ifter acute or chronic ethanol adminis- 
a cannot be determined at present. The 
of obvious mucosal damage in rats 
ically fed ethanol suggests that acute 
hronic alcohol exposure may diminish 
absorption by different mechanisms, 
workers, however, have noted ultra- 
ural changes in small intestinal mucosa 
ring more prolonged low-level ethanol 
ure (8). While there is evidence that a 
c factor may modify the absorption of 

Gostric Injection 

Gut Loop 
Pre -treatment 

3. The intestinal absorption of a single dose of 
ioride in rats without prior ethanol exposure as 
oed by gastric injection of saline or 50% ethanol 
id by pretreatment of the intestinal loop by saline 
ethanol (right). 

alcoholic lead from stomach and intestine, 
the lead absorption does not exceed that ob- 
served in aqueous lead control animals. 

Factors enhancing the susceptibility to lead 
poisoning have been reviewed (9). Several 
dietary deficiencies common among heavy 
alcohol users have been established as capa- 
ble of potentiating the manifestations of lead 
toxicity. While protein deficiency reduces 
lead absorption (1 1), it produces greater sus- 
ceptibility to lead toxicity (12, 13). Dietary 
calcium deficiency increases lead retention 
(14, IS) ard potentiates morphological and 
biochemical parameters of lead poisoning 

(16) but does not alter lead absorption (7). 
Iron deficiency both enhances lead toxicity 

(17) and increases lead absorption (10, 11). 
The effects of ascorbic acid, pyridoxine, and 
other micronutrients on lead metabolism and 
toxicity are not known with certainty (9). 

Since these experiments indicate that acute 
or chronic ethanol exposure does not increase 
lead absorption, particularly at concentra- 
tions commonly seen in lead-containing 
"moonshine" whiskey (1-10 /ig of Pb/ml) 
(18), the apparent synergism of lead and 
ethanol reported in alcoholics may be related 
to increased lead exposure (lead-contami- 
nated illicit whiskey or industrial environ- 
ments) and/or nutritional deficiencies as pre- 
viously concluded (1). These studies suggest 
that chronic ethanol ingestion does not alter 


Fio. 4. The whole-body retention of lead following 
a single intravenous dose of lead-210 in rats chronically 
ingesting 10% ethanol. 

\BLE I. Effect of Absorptive Site on Absorption of Lead from Aqueous 


AND Alcoholic Lead 

Site of absorption 


Stomach and intestine 


jlution Aqueous AlcohoUc 
bforpUon (%) 2.5 ± 0.7 2. 1 ± 0.4 

Aqueous Alcoholic 
28.6 ± 1.7 22.4 ± 3.2 

Aqueous AlcohoUc 
30.6 ±1.5 8.2 ±0.8 



the elimination of small quantities of lead 
administered as a single intravenous dose. 
Although there are no previously published 
reports of the effects of ethanol on lead ex- 
cretion, the variety of renal lesions seen in 
plumbism and the known augmentation of 
lead-induced renal abnormaUties by alcohol 
(1) suggest that diminished excretion may be 
of significance only when large quantities of 
lead are involved. 

Summary. To determine the effects of acute 
and chronic ethanol ingestion on the absorp- 
tion of lead, experiments were performed us- 
ing an in vivo isolated gut loop technique. 
Acute administration of S0% ethanol signifi- 
cantly reduced the absorption of lead at con- 
centrations of 1 and 10 jLig of Pb/ml. This 
effect appears to be independent of lead sol- 
ubility in alcohol and is associated with struc- 
tural changes in intestinal mucosa, suggesting 
toxicity. Absorption of a single dose of lead 
was also diminished in animals chronically 
exposed to ethanol. Elimination of a single 
intravenous dose of lead was not affected by 
chronic alcohol ingestion. These findings sug- 
gest that the clinically reported synergism of 
lead toxicity and ethanol is related not to 
increased lead absorption or diminished lead 
excretion but to nutritional deficiencies and 
increased lead exposure among some alco- 

1. Mahaffcy. K. R., Goycr, R. A., and Wilsc 
Arch. Environ. Hlth. 28, 217 (1974). 

2. GilfilUn, S. C. J. Occup. Med. 7, 53 (1965 

3. Owen, C, Dodson, W. H., and Hammack 
Med. J. 60, 44 (1967). 

4. Cramer, K., AcU Med. Scand. Suppl. 445, 

5. Krawitt, E. L., J. Ub. Clin. Med. 85, 665 ^ 

6. Krawitt, E. L Proc. Soc. Exp. Biol. Med 

7. Barton, J. C, Conrad, M. E., Harrison 
Nuby, S., J. Ub. Clin. Med. 91, 366 (1978 

8. Rubin, E., Rybak, B. J., Lindenbaum, J., C 
D., Walker, G., and Lieber, C. S., Gastroc 

9. Goyer, R. A., and Mahaffey, K. R., Envi 
Perspect. 2(1962). 

10. Ragan, H. A., J. Ub. Clin. Med. 90, 700 ( 

11. Conrad, M. E., and Barton, J. C, Gastroc 

12. Baemstein, H. D., and Grand, J. A., J. P 
Exp.Ther. 74, 18(1942). 

13. Gonizea, I., et al. Arch. Sci. Physiol. 18, 2 

14. Lec;rer, L. G., and Bing, F. C, J. A.M. A. 

15. Shields, J. B., and Mitchell, N. H., J. Nui 

16. Six, K. M., and Goyer, R. A., J. Ub. CUn 

17. Six, K. M., and Goyer, R. A., J. Ub. CUn 

18. Morris, C. E., Heyman, A., and Pozefaky 

Received December 12, 1977. P.S.E.B.M. 1978 


: Scatter Characteristics of Erythroid Precursor Cells Studied in Flow Analysis 



Ttments of Biochemistry, Medicine, and Pathology and the Cancer Center, Medical College of Virginia, 

Richmond, Virginia 23298 

analysis is a powerful new tool to 
laracteristics of individual cells in sus- 
(1). The presently available flow in- 
ts measure either scatter of mono- 
ic light by cells or fluorescence of 
irome-labeled cell structures. In ad- 
ight scatter (LS) or fluorescence can 
criminating parameter by which se- 
;lls can be separated electronically for 
lal study. 

marrow is a complex mixture of cells 
al developing myeloid lines (granu- 
erythrocytes, megakaryocytes, and 
tes) as well as lymphocytes. Effective 
f these cell types requires samples 
nriched with an individual cell type, 
flow instruments to sort a cell type, 
1 based on either LS or fluorescence 
made. Fluorescence usually requires 
ble alteration and killing of cells. 
S study is cell-sparing, sorted cells 
»e chemically unaltered and possibly 
ally active. Studies of bone marrow 
flow analysis have shown the possi- 
of separating selected cell types of 
I, 3). In previous reports from this 
»ry (3), emphasis has been placed on 
ng granulocyte precursors of varying 
)f development. In this report, the 
ty of separating erythrocytes and 
d precursors in relatively pure form 
^ssed. This was accomplished by a 
lary isopycnic fractionation of rabbit 
on density gradients followed by flow 
of the gradient fractions, using LS 
to determine sorting parameters. 
ials and methods. Preparation of mar- 
pensions. Rabbit marrow was removed 
s long bones, Altered, and washed as 
jly described (4) except that hypo- 
sis of erythroid precursors was not 

ty gradient fractionation of bone mar- 
ished bone marrow cells were sus- 

pended in Ficoll/Hypaque solution and dis- 
persed in a linear density gradient formed 
with the Beckman gradient former. The mix- 
ing solutions had densities of 1.0478 and 
1.1579 g/ml. Gradients were formed in 13-ml 
tubes for the SW41 rotor and the cells were 
separated isopycnically during a 40-min cen- 
trifiigation at 4300g. 

Flow analysis and electronic cell sorting. 
Fractions from the preliminary isopycnic sep- 
aration of cells were analyzed with a Coulter 
Electronics Company TPS-1 sorter. Cells 
were analyzed at a flow rate of 1000 to 3000 
cells per second and LS histograms were gen- 
erated as described previously (3). The dis- 
tinct and reproducible distributions in the LS 
histograms were used to set electronic sort 
windows by which 100,000 cells were sorted 
in each of two windows simultaneously. 

Sorted cells were collected in fetal calf 
serum and collected on microscope slides in 
a Shandon cytocentrifiige. Diflerential cell 
counts were performed after staining with 
Wright's stain. 

Results, Light scatter profiles of bone mar- 
row cells at differing buoyant densities. Blood 
cell precursors of the bone marrow, both 
erythroid and myeloid, increase in buoyant 
density as they mature. Thus a preliminary 
separation of bone marrow cells by isopycnic 
sedimentation in Ficoll/Hypaque gradients 
allows collection of gradient fractions near 
the top of the gradient which are rich in 
immature cells and fractions of increasing 
maturity progressing to the bottom of the 
gradient (4). 

The cells recovered from each density gra- 
dient fraction (I-X) were subjected to flow 
analysis. The LS histograms, with the cell 
numbers on the ordinate and increasing LS 
intensity on the abscissa, are shown in Fig. 1. 
For clarity, 4 of the 10 gradient fractions 
which best illustrate the typical changes in 
the profiles from top to bottom of the gra- 


0037-9727/78/ 1 592-02 1 9$0 1 .00/0 

Copyright © W]% b^ \h€ SocxtVj fex ^xvtt\twtTvV«\%voV3(^ *sA>K^Ac«»r 




Fig. 1. Light scatter (LS) histograms of rabbit mar- 
row cells. Cell number is plotted on the ordinate and 
increasing LS is indicated on the abscissa. Arabic nu- 
merals refer to limits of windows for electronic sorting 
of cells. Roman numerals refer to LS patterns of fractions 
derived from preliminary density gradient fractionation 
of the marrow (fraction I at top, fraction X at bottom of 
the density gradient). 

dients are shown. The Arabic numerals on 
the abscissa designate sort fractions and in- 
dicate the segments under the LS profile cho- 
sen for individual electronic sorts. In gradient 
fraction X, too few cells were available for a 
fourth sort fraction. 

The pattern consists of distributions rep- 
resenting distinct classes of cells with similar 
LS properties. The peak included in sort frac- 
tion 1 dominates in terms of cell number. 
Each curve is adjusted to show the peak at 
the top of the histogram so that the relative 
proportions of the sort fractions can be com- 

In gradient fraction 1 (top of the density 
gradient) the peak included in sort fraction 2 
is more prominent than the third peak (sort 
fraction 3), but both peaks are similar in 
gradient fractions closer to the bottom of the 
density gradient, and both are small com- 
pared to the peak in sort fraction 1 at the 
bottom of the gradient. 

In terms of total cells, 46% of the cells in 
gradient fraction I are erythroid, over half of 
which are nucleated. Erythrocytes make up 
84% of the cells of the gradient fraction X, 
but only 3% of these are nucleated. 

The differential counts of individual sort 
fractions in Fig. 2 show the distribution of 
cell classes in several sort fractions derived 

from each density gradient fraction. The dif- 
ferential cell counts from each sort fraction 
are reported in three categories, represented 
by the three bars under each sort designation 
in Fig. 2. The "lymph" bar indicates lympho- 
cytes and smudged nuclei (shaded portion), 
which may sort with lymphocytes. The 
"RBC" bar indicates erythrocytes and eryth- 
roid precursors. The latter are indicated by 
shading and "nRBC". The third bar ("gran") 
in each sort indicates granulocytes, mature 
polys, and their precursors. 

Sort 1 is predominantly an erythroid frac- 
tion, sort 2 is enriched with nucleated red 
cells, and sorts 3 and 4 are primarily granu- 
locyte fractions. If sort fraction 2 of gradient 
IV (mid-gradient) is chosen, a sample of cells 
is obtained which is over 90% erythroid, 64% 
of the cells being nucleated erythroid precur- 

An example of this fraction is shown in 
Fig. 3. This photomicrograph shows a group 
of nucleated erythrocyte precursors and one 
larger cell which may be lymphoid. The gran- 
ulocytes chiefly responsible for the LS peak 
in sort fraction 3 are more mature than those 
which predominate in sort fraction 4. The 
larger less mature granulocytes scatter more 
light than the more mature cells (3). 

Discussion. The scatter by cells of an inci- 
dent beam of light is determined in part by 
the size of the cell, but also depends on 
reflection from cell surfaces, phase-shift in 
light passing around or through the cell, and 
diffraction of light by internal structures 
within the cell. The instrument involved in 
this study utilizes a light detector which col- 
lects light scattered 2^ to 20*" from the inci- 
dent beam. It is known that cell size is the 
most important determinant at low angles of 
scatter (2° to 5°) and presumably internal 
structure plays a greater role in determining 
the intensity of scatter at larger angles (S). 

It is clear from these studies that cell size 
is not the only determinant of scatter. The 
mature erythrocytes and reticulocytes scatter 
less light than any other cell type in the 
marrow, and there is a distinct separation of 
peaks of nucleated and non-nucleated eryth- 
rocytes. Since the size of the maturing red 
cell precursors decreases in a continuous fash- 
ion, if size were the major determinant of LS, 
there would be one broad, continuous peak 






i M 



U □LlIU 
s ^ i s 

5 I " I 










8 ^ 

I 30 

& <0 







Fig. 2. Differential cell counts of cells sorted from density gradient fractions whose LS profiles are seen in Fig. 
Etch sort fraction contains cells obtained from the abscissa in Fig. 1. Each sort fraction is reported as lymphs 
mphocytes and smudged nuclei). RBC (erythrocytes and nuclear RBC), and gran (neutrophils and their 
!Cursors). In gradient fraction X only three sort fractions were collected. 




Fig. 3. Photomicrograph of sort fraction 2 of density gradient fraction IV. Cells shown are nucleated erythrocyte 
ecursors and one probable lymphoid cell. Original magnification, x ICXX). 




of erythroid LS. Instead, discrete distribu- 
tions were observed (Fig. 1). This was also 
evident in other LS studies from our labora- 
tory (3) in which it was shown that lympho- 
cytes of various sizes were found to have very 
similar LS properties. It is evident that the 
character of the nucleus is an important LS 

Granulocytic cells tend to scatter more 
light than erythrocytes, normoblasts, or lym- 
phocytes. This is due no doubt to both greater 
cell size and much greater complexity of cy- 
toplasmic organelles. 

It is evident that LS of cells, especially 
when combined with separation based on 
buoyant density differences, is a useful means 
of isolating erythrocyte precursors for study. 

Summary, Light scatter (LS) differences 
among cells of rabbit marrow was studied by 
flow analysis using a Coulter two-parameter 
cell sorter. A preliminary fractionation of the 
marrow into samples enriched with cells of 
varying degrees of maturation was accom- 
plished in Ficoll/Hypaque density gradients. 

Subsequent study of each of these c< 
pies in flow analysis demonstrated un 
profiles which distinguished erytl 
from nucleated erythroid precursc 
granulocyte precursors. The combine 
ration procedures made it possible 
fractions of erythroid precursors with 
as 90% erythroid cells, two-thirds o 
were nucleated precursors. 

This research was supported by Research C 
08482, CA 16059, and CA 17177 from the 
Institutes of Health. 

1. Horan, P. K., and Wheclcs, L. L., Sciena 

2. token, M. R.. Sweet. R. G., and Herzenb< 
J. Histochcm. Cytochem. 24, 284 (1976). 

3. Scott, R. B., Grogan, W. McL., and Colli 
Blood 51, 1137(1978). 

4. Scott. R. B.. Eanes, R. Z., Cooper, L. W. 
L. L., and Eastment. C. A.. Brit. J. Haen 

5. Bninsting, A., J. Histochem. Cytochem. 

Received March 29, 1978. P.S.E.B.M. 1978, Vo 


Salbutamol as a Topical Anti-inflammatory Drug (40319) 

Upjohn Company, Department of Hypersensitivity Diseases Research, Kalamazoo, Michigan 49001 

the initial events in acute inflam- 
the release of histamine from mast 

response to tissue injury or 
ntibody complexes. Histamine 
ation and increased permeability of 
i. Local reddening and edema ap- 
)wed by secondary characteristics of 
pain (1). Drugs that inhibit hista- 
%se prevent or reduce tissue inflam- 
ihibition of histamine release is ac- 
id partly by increasing the cellular 
cycUc adenosine monophosphate 
2). Anti-inflammatory steroids stim- 
nyl cyclase to convert adenosine 
ate to cAMP, and /J-adrenergic ag- 
nulate adenyl cyclase at the )8-ad- 
eceptor (3). 

ortisone ( 1 7a-hydroxycorticoster- 
isol) is used effectively to reduce 
[animation; however, salbutamol 

several distinct advantages. Salbu- 
hydroxy) is a relatively specific jS- 
c agonist and selectively stimulates 
gic receptors (4). The cardiovascu- 
ntral nervous system effects of other 
mimetic amines are caused in part 
s on the receptors which are preva- 
>se tissues. 

e report the local anti-inflammatory 
!* salbutamol when applied topically 
^d rat ears. 

lis and Methods, This method of 
ocal inflammation in rat ears by 

is essentially that of Tonelli et al. 
(v/v) croton oil solution in absolute 

applied by micropipet to the outer 
* both ears (0.05 ml each). The ears 
dematous in 3 to 6 hr and remain 
d for up to 48 hr. Inflammation 
s measured by cutting off the ears 

address: The Great Western Sugar Company 
id Development Lab, Loveland, CQlorado 

m reprint requests should be addressed. 

at S.S hr and weighing them. Drugs are usu- 
ally applied simultaneously in the croton 
oil-ethanol mixture. In some cases, as noted, 
drugs are applied after the croton oil. Male 
Sprague-Dawley rats (200-240 g) are used. 
Untreated control rats provide the weight of 
normal nonedematous ears. Croton oil- 
treated rats demonstrate the extent of inflam- 
mation in the absence of drugs. Hydrocorti- 
sone (1%), serving as a positive control, con- 
sistently inhibits inflammation by 80 to 
100%. Data are expressed as milligrams of 
edema of both ears, that is, the increase in 
weight of both ears over the untreated con- 
trols. The weights in each group are averaged 
and the standard error of the mean is calcu- 
lated (depicted by vertical line extensions on 
the graphs). 

Results, Local inflammation is inhibited 
totally by hydrocortisone and salbutamol 
when they are appUed topically to the ears at 
1 to 2% (w/v) in the croton oil solution (Fig. 
1). Croton oil causes the ears to gain an 
average of 155 mg in the absence of any anti- 
inflammatory agent. Drug concentrations of 
0.1% reduce the edema by 80%. When drugs 
are applied to a distant shaven area of the 
back, anti-inflammatory activity still occurs 
but higher concentrations are required (Fig. 

Hydrocortisone and salbutamol reduce lo- 
cal edema even when applied after the in- 
flammation reaction is in progress (Fig. 2). In 
the case of salbutamol, significant reduction 
of inflanmiation is obtained when given up 
to 2 hr after appUcation of the croton oil. 
Hydrocortisone is not as effective when given 
this late in the development of acute inflam- 

Salbutamol is found to be inactive orally 
in our model (Fig. 3). Doses of up to 35 
mg/kg body wt, delivered orally by stomach 
tube, failed to significantly inhibit ear edema. 

Propranolol (a )8-adrenergic receptor 
blocking agent) interferes with the ability of 
salbutamol to inhibit inflammation, but di- 


0037-9727/78/ 1 592-0223$0l .(W/Q 

AU n^\& TtacTvtd. 









• I II 21 11 II 21 I 21 II III 21 II Ml 


Fio. 1. Local and systemic anti-inflammatory activ- 
ity of hydrocortisone and salbutamol. The drugs are 
applied directly to the ears (A) or to a shaven area on 
the back (B). In both A and B, the croton oil was applied 
to the ears to induce inflammation. In this and subse- 
quent graphs the averages of five animals per group are 
presented, and the vertical line extensions represent the 
standard errors of the mean. 


Fig. 2. The effects of salbutamol and hydrocortisone 
on local inflammation when they are administered dur- 
ing the course of the inflammation reaction. Croton oil 
was applied to the ears to induce inflammation, while 
salbutamol (1%) and hydrocortisone (1%) were also ap- 
plied but at various times after the croton oil. 

benamine (an a-adrenergic receptor blocking 
agent) has no influence (Fig. 4). Neither pro- 
pranolol nor dibenamine prevents the anti- 
inflammatory activity of hydrocortisone. 

Discussion, The need exists for a locally 
active anti-inflammatory drug that can be 
applied direaly. Salbutamol (Ventolin, Allen 
and Hansbury) is used in foreign countries in 
the management of asthma (6). Green (7) has 
reported that salbutamol, injected ip, could 
reduce inflammation both in the mouse per- 
itoneum induced by acetic acid and in the rat 
hindpaw edema induced by carrageenin. He 
also demonstrated that the activity is not 

mediated by release of adrenal cortico: 

Salbutamol is very effective in the prev 
tion of local inflammation. Although it is 
orally active in our model of inflammati 
salbutamol is effective if applied directl} 
the inflamed site or at a remote site. 1 
suggests that the drug is readily absorbed i 
the circulatory system; however, larger o 
centrations are required if the drug is 
applied at the site of inflanmiation. 






0.1 10 2.0 

1.3 3.0 3S0 

ORAL (M|/ko) 

Fig. 3. Topical and oral activity of salbutamol 
local inflammation. Salbutamol was applied directly 
the ears, or given orally by stomach tube, at varic 
doses, 30 min prior to the croton oil. 




eturaiH nn m§ 

JU. T 

r pno [ 





Fig. 4. Effects of 4% propranolol (pro) and 4% 
benamine (dib) on the local anti-inflammatory acti^ 
of hydrocortisone and salbutamol. Croton oil, propr 
olol, and dibenamine were given independently ( 
controls and in combination with the drugs (simulta 
ous application, including croton oil). 


ion of salbutamol as a bronchodi- 
e treatment of asthma is mediated 
drenergic receptors in the bronchus 
3-adrenoceptors in the mast cells 
volved in the action of salbutamol 
nation. The )8-blocking agent pro- 
Qterferes with the ability of salbu- 
nhibit inflammation. Dibenamine, 
ing agent, has no effect. Hydrocor- 
in a different manner since neither 
3I nor dibenamine block the effect 

able cardiac side effects are ex- 
be minimal, because salbutamol is 
selective for )82-receptors and has 
t on Pi receptors which predomi- 
: heart. The minimal adverse effects 
Lg compared with other /J-agonists 
t control of asthma are discussed 
1 (8) and by Dochom (9). We have 
-inflanmiatory activity of other ag- 
t salbutamol was pursued because 
nost effective and because of its 
yr." Morrison and Farebrother (10) 
rted a case of salbutamol overdose 
be the physiological and cardiovas- 
its that occur. Further studies are 
but it appears safe to attempt to 
I inflammatory conditions of the 

Summary, Using croton oil-induced 
edema, hydrocortisone and salbutamo 
anti-inflanunatory activity when appli( 
ically. Both drugs act to some exten 
when applied after the inflammation re 
is in progress. Both drugs are also activ< 
applied to a shaven area of the back, 
remote from the ear inflammation. Si 
mol acts by a different mechanism tha 
inflammatory steroids. The advantages 
butamol are discussed and it appears 1 
useful adjunct in the treatment of infli 
tory dermatoses. 

1. Mclmon, K. L.. and MorrcUi, H. F. (eds). 
Pharmacology, p. 382. Macmillan (1972). 

2. Lichtenstein, L. M., and Margolis, S., Sciei 
902 (l%8). 

3. Brittain, R. T.. Jack, D., and Ritchie, A. i 
Drug Res. 5, 197 (1970). 

4. Zsoter, T. T., and Epstein, S. W., Chest 

5. Tonelli, G., Thibault, L., and Ringler, I.. ] 

6. Rcbuck, A. S.. Drugs 7, 344 (1974). 

7. Green, K. L., Brit. J. Pharmacol. 45, 322 (1< 

8. Brittain, R. T., Proc. Roy. Soc. Med. 65, 75< 

9. Dockhom, R. J., Ann. AUergy 29, 539 (1971 
10. Morrison, G. W., and Farebrother, M. J. B. 

2,681 (1973). 

Received January 11. 1978. P.S.E.B.M. 1978. V< 


Interaction of Ethanol and Thyroxine on Hepatic Oxygen Consumption^ (40321 

Medical Research Service, Veterans Administration Hospital and the Department of Medicine, Chicago Me< 

School, North Chicago, Illinois 60064 

Chronic feeding of ethanol to rats has been 
shown to stimulate respiration by liver slices 
through an increase in conversion of ATP to 
ADP by the (Na + K)-ATPase system (1,2). 
The calorigenic effect of thyroid hormones 
also involves stimulation of (Na + K)- ATP- 
ase (3). However, some studies have sug- 
gested that the availability of mitochondrial 
substrate and not ADP may determine the 
rate of respiration and that thyroxine (T4) 
enhances the availability of the substrate for 
mitochondrial oxidation (4). 

The present study was done to investigate 
interrelationship between the effects of 
chronic ethanol ingestion and T4 treatment 
on O2 consumption by rat liver slices and 
isolated mitochondria. Further, the influence 
of the available oxidizable substrate for the 
ethanol and T4 effects on respiration of rat 
liver slices was studied. 

Materials and Methods. Thirty-two 
Sprague-Dawley male rats weighing 150 to 
200 g were divided equally into four groups 
at random. Group A received tap water and 
group B received 20% (v/v) ethanol as the 
only drinking solution ad libitum. Group C 
was rendered thyrotoxic by daily ip injection 
of I-T4 (150 jiig/100 g body wt) for 14 days. 
Group D received 20% (v/v) ethanol as 
drinking solution and T4 treatment as out- 
lined for group C. All animals were housed 
in individual cages, fed regular Purina Chow 
ad libitum, and weighed at regular intervals. 
Animals in group D lost considerable weight 
(see Table I) and appeared sick, although 
none died. In eight relatively young rats, 
average weight 100 g, a 25% mortality rate 
was observed during 20% ethanol 4- T4 treat- 
ment and therefore present studies involved 
relatively larger animals. 

' This work was supported by the Medical Research 
Service, Veterans Administrationand was presented, in 
part, at the American Physiological Society 28th Annual 
Fall Meeting, October 1977, Hollywood, Florida. 

After 14 days the animals were fast 
18 hr and then sacrificed by decapit 
Blood was collected for the estimati 
serum T4 levels (5). Livers were remove 
placed immediately in ice-cold oxygc 
medium containing 135 mA/ NaCl, t 
KH2PO4, 0.5 mAf MgCk, 5 mAf Tris 
and 10 mAf glucose, pH 7.4. Liver slic 
mm thick were prepared and their respi 
was determined in a Warburg apparatui 
cision Scientific). Each Warburg flasl 
tained approximately 60 mg of tissue ii 
of the oxygenated mediimi mentioned i 
Respiration was measured for three o 
utive 30-min periods. Thereafter 50 /jd 
M succinate was added to the mediunc 
the side arm to give a final concentrat 
7 mA/ and respiration of the liver slia 
estimated for three additional lO-min jh 

To determine oxygen consumption ( 
lated mitochondria instead of liver slio 
tochondria were isolated from the same 
according to the technique of Johnso 
Lardy (6). An aUquot, 0.05 ml, of the 
chondrial suspension was placed in a 
burg flask containing 3 ml of incubatic 
dium which contained 62.5 mAf si 
185.5 mAf mannitol, 10 mAf KCl, II 
Tris-HCl, pH 7.4, 5 mAf K2HPO4, : 
MgCb, 0.2 mAf EDTA, 7 mAf succinat 
83.3 fiM ADP. Respiration was measui 
three consecutive 10-min periods. 

Respiration estimations for liver slio 
mitochondria were done in triplicate fc 
liver. The protein content of the livei 
and of each mitochondrial suspensio 
determined by the Lowry method (7 
data were expressed as microUters of C 
sumed per minute per milligram of f 
and statistically analyzed by Student's 

Results. Table I shows mean ± SEM 
of body weight and serum thyroxine le 
rats receiving ethanol, thyroxine, or a 
bination of Uiese two substances (n " 
each group of animals). Rats that re 


0037-9727/78/ 1592'0226SO\. 00/0 
Copyright <S> 1978 by the Society for Experimenui Biology and Medicine 
^^^ r^is reserved. 



TABLE I. Effect of Ethanol Ingestion on Body Weights and Serum T* Levels of Normal and 

Thyroxine-Treated Rats. 


Initial body weight 

Final body weight 



Serum thyroxine 

A. Saline 

B. Ethanol 

C. T4 

D. Ethanol + T4 

166 ± 8 
179 ± 8 
176 ± 10 
179 ± 7 

238 ± 8 
194 ± 9 
229 ± 12 
132 ± 7 

72 ±6 
15 ±4* 
53 ±8 
-47 ± 4* 

4.6 ± 0.4 
3.9 ± 0.3 
16.6 ± 2.4* 
14.1 ±2.9* 

• ^ < 0.001 compared to controls, (saline). 

either ethanol or T4 gained significantly less 
weight than controls (/? < O.OS): Simultaneous 
treatment with ethanol and thyroxine pro- 
duced a marked loss in body weight as com- 
pared to controls (176.2 ± 7 vs 127 ±l\p< 
0.01). Serum T4 levels were significantly 
higher (p < 0.01) in T4-treated animals and 
ethanol ingestion exerted no discernible effect 
on serum T4 values. 

Effect of ethanol on O2 consumption by liver 
slices of euthyroid and thyrotoxic rats. Figure 

1 shows that chronic ethanol ingestion de- 
creased the rate of oxygen utilization from a 
control value of 0.098 ± 0.004 to 0.082 ± 
0.004 /ig of 02/min/mg of protein in hver 
sUces of euthyroid rats. The results were sig- 
nificant at the 2% level. Addition of succinate 
to liver slices produced a marked increase in 

02 consumption to 0.183 ± 0.01 and 0.243 
± 0.01 /il of 02/min/mg of protein in controls 
and ethanol-treated animals, respectively. 
Furthermore, with succinate as oxidizable 
substrate, ethanol pretreatment produced an 
increase (p < 0.001) in the rate of respiration 
instead of a depression of respiration ob- 
served with glucose as the substrate. 

Figure 2 shows that in thyrotoxic rat hver 
sUces the O2 consumption was 50% greater 
than in euthyroid rat hver shces (/? < 0.001). 
Chronic ethanol ingestion decreased O2 con- 
sumption by nearly 50% from 0.147 ± 0.004 
to 0.07 1 ± 0.005 id of 02/min/mg of protein 
(p < 0.001) with glucose as substrate. Addi- 
tion of succinate increased respiration of T4- 
treated rat Uver shces and chronic ethanol 
ingestion enhanced the rate of respiration 
further firom 0.282 ±0.015 to 0.367 ± 0.028 
nl of 02/min/mg of protein {p < 0.02). 

Effect of ethanol and thyroxine on O2 con- 
sumption of isolated mitochondria of rat liver. 
As shown in Fig. 3, mitochondria isolated 
from euthyroid rat Uver showed no significant 
difference in respiration after chronic ethanol 
treatment as compared to control values. 









Fig. 1. Mean ± S£M (n » 8) oxygen consumption 
rate by liver slices of rats fed 20% ethanol as drinking 
solution or tap water (controls) for 14 days. The rate of 
O2 consumption was estimated with liver slices in media 
containing glucose before and after the addition of suc- 


w 03 






T, O0H*Ti 

Fig. 2. Mean ± S£M (n « 8) oxygen consumption 
by liver sUces of T4-treated rats that received 20% ethanol 
or tap water ad libitum for 14 days. Oxygen estimation 
was done as described under Fig. 1 and T4 injections 
were given as described in the text. 

However, in mitochondria isolated fi-om thy- 
rotoxic rat hvers it was observed that chronic 
ethanol treatment enhanced O2 consumption 
significantly fi-om 1.27 ± 0.032 to 1.57 ± 
0. 1 18 jiil of 02/min/mg of protein (/? < 0.05). 
Discussion. Previously it has been shown 
that daily ingestion of ethanol (35% calorie- 
wise) for 21 to 27 days enhanced oxygen 
consumption by rat hver shces. The under- 




a. to 



1 , 








, ^ _: 






Fig. 3. Mean ± SEM (n « 8) oxygen consumption 
by mitochondria isolated from livers of euthyroid and 
T4-treated rats. Both groups of euthyroid or T4-treated 
rats were given 20% ethanol or tap water £ui libitum for 
14 days. Injections of T4 were given as described in the 

lying mechanism was reported to be an in- 
creased activity of the (Na -»- K)ATPase ac- 
tivity (1, 2). The present data derived from 
rats consuming relatively less ethanol (i.e., 
20% (v/v) as drinking solution ad libitum for 
14 days) show that the ethanol effect on 
respiration of rat Uver shces is dependent on 
the available oxidizable substrate. The O2 
consumption by Uver shces was increased in 
medium containing succinate as substrate but 
decreased when glucose was used instead of 

Substrates can provide electrons to the res- 
piratory chain at the beginning (the level of 
NADH dehydrogenase), at the middle (ubi- 
quinone level), and at the terminus (cyto- 
chrome c level). Succinate which is flavin- 
linked provides electrons at the cytochrome 
Z^-ubiquinone segment and therefore by- 
passes energy coupling site I at the level of 
NADH dehydrogenase. The utilization of 
electrons from glucose is partly NAD-linked 
and thus involves energy coupling site I. The 
present results might be explained by an in- 
hibitory effect of ethanol on coupling site I 
or on some steps prior to it. In fact, Ceder- 
baum et al (8) have shown that chronic 
ethanol ingestion (36% caloriewise) depresses 
mitochondrial respiration by damaging cou- 
pling site I. 

Whereas ethanol enhanced O2 consump- 
tion in rat Uver sUces incubated with succi- 
nate, it did not exhibit a similar effect when 
isolated mitochondria from the same Uvers 
were studied. Other studies have reported a 

depression of mitochondrial respi 
chronic ethanol ingestion and asc 
effect to a damage to the respiratc 
Furthermore, structural changes in i 
dria including swelling, disfigurati 
rientation of cristae, and intramito 
crystalline inclusion are observ 
chronic ethanol treatment (9). Coi 
fat infiltration of hepatocytes has 
shown (10). In the present study rat 
less ethanol, and any morphologies 
in mitochondria, although not doc 
were perhaps insufficient to depres 
tion. In fact, the O2 consumption 
chondria isolated from thyrotoxic 
was enhanced by chronic ethanol 
Therefore, it is unlikely that the r< 
chain was damaged by ethanol as 
rats in this study. 

The calorigenic effect of thyroid '. 
on Uver is ascribed to an increase 
production due to stimulation of (! 
ATPase activity (3). However, Prii 
Buchanan (4) showed O2 consumpt 
Uver sUces was greater with succi 
with glucose and suggested that tl 
biUty of oxidizable substrate rather 1 
controls the rate of O2 consump 
present data show that chronic eths 
ment decreased O2 consumption oft 
rat Uver sUces when glucose was a\ 
substrate but a converse effect occui 
succinate was added. It seems ths 
inhibition of energy coupling site 
cussed above, was sufficient to bloci 
on Uver sUces respiration in a glu 
taining medium. On the other han 
crease in succinate-supported resp 
the same Uver sUces might be rela 
creased (Na -»- K)- ATPase activity 

It needs to be emphasized that rat 
D receiving ethanol -»- T4 lost coi 
weight and appeared sick. A decrea 
intake might have resulted in a Umit 
of substrate for O2 utilization. Thi 
conceivable effect of malnutrition ii 
to that of ethanol should be cons 
interpretation of data derived fron 
rats. Nonetheless, individual rat U^ 
its own control in terms of compari 
rates of O2 utilization during two 
substrates, i.e., glucose and succina 

Summary, Interrelationship bet 



of chronic ethanol ingestion and T4 
cnt on O2 consumption by rat liver 
ind isolated mitochondria was investi- 
The data showed that ethanol influ- 
»n O2 consumption by Uver shces was 
lent on the available oxidizable sub- 
as it was decreased when estimated in 
containing glucose but increased in 
containing succinate as oxidizable sub- 
The respiration of thyrotoxic rat Uver 
was altered by ethanol in a manner 
r to that observed with euthyroid rat 
»lices. Whereas ethanol ingestion en- 
1 succinate-supported respiration of eu- 
i and thyrotoxic rat Uver sUces, it pro- 
a similar effect in isolated mitochon- 
f thyrotoxic rat Uvers but not of euthy- 
it Uvers. 

authors wish to thank Mrs. Ruth M. Bonovich 
[ig this manuscript. 

1. Israel, Y., Videla, L., and Bernstein, J., Fed. Proc. 

2. Israel, Y., Videla, L., MacDonald, A., and Bernstein, 
J.. Biochem. J. 134, 523 (1973). 

3. Ismail-Beigi, F., and Edelman, S., Proc. Nat. Acad. 
Sci. USA 67, 1071(1970). 

4. Primack, P. M., and Buchanan, J. L., Endocrinology 

5. T4 RIA (PEG) Diagnostic Kit, Abbott Laboratories, 
North Chicago, 111. 

6. Johnson, D., and Lardy, H., in **Methods in Enzy- 
mology," Vol. 10, p. 94. Academic Press, New York 

7. Lowry, O. H., Rosebrough, A. J., Farr, A. L., and 
RandaU, R. J., J. Biol. Chem. 193, 265 (1951). 

8. Cederbaum, A. I., Leiber, C. S., and Rubin, E., Arch. 
Biochem. Biophys. 165, 560 (1974). 

9. Leiber, C. S., and DeCarli, L. M., in '*MeUbolic 
Aspects of Alcoholism** (C. S. Leiber, Ed.), p. 31. 
Univ. Park Press, Baltimore (1977). 

10. Gordon, E. R., J. Biol. Chem. 248, 8271 (1973). 

Received January 16, 1978. P.S.E.B.M. 1978, Vol. 159. 


Effect of a Phosphodiesterase Inhibitor, 3-lsobutyl 1 -methylxanthine, upon the 
Stimulatory Effect of Human Follicle-Stimulating Hormone and Human Luteinizing 
Hormone upon Cyclic Adenosine 3':5'-Monophosphate Accumulation by Porcine 

Granulosa Cells^ (40321 ) 


Department of Physiology, University of Maryland School of Medicine, 660 West Redwood Street, 

Baltimore, Maryland 21201 

A mechanism of polypeptide hormone ac- 
tion on target cells is to stimulate formation 
of cAMP which subsequently acts as an in- 
tracellular mediator of hormone action. Intra- 
cellular cAMP levels are the result of given 
rate of synthesis combined with a given rate 
of degradation or extracellular release. The 
cyclic nucleotide is believed to be hydrolyzed 
to 5 '-AMP by one or more cyclic nucleotide 
phosphodiesterases ( 1 ). Methylxanthines 
have been shown to exert inhibitory effects 
on the action of phosphodiesterase (2, 3). We 
have shown previously that LH and FSH can 
stimulate cAMP accumulation by porcine 
granulosa cells (GC) and that the amount of 
cAMP accumulated in response to the two 
gonadotropins differs according to the stage 
of maturation of the follicle (4). In addition, 
observations from previous studies (4) suggest 
that the phenomenon of cAM P accumulation 
by porcine GC in response to the stimulatory 
effects of the gonadotropins occurs over time. 
For GC from small and medium follicles the 
intracellular cAMP accumulated in response 
to FSH was not observed to decline signifi- 
cantly in incubations of 30 min or less. The 
decline occurred between 30- and 60-min 
periods of incubation and it was during this 
time interval that the increase in cAMP ac- 
cumulation in the incubation medium was 
observed to occur. For GC from large follicles 
the intracellular cAMP accumulated in re- 
sponse to LH was not observed to decline 
with 30- nor with 60-min incubations; how- 
ever, a significant increase in the cAMP ac- 
cumulation in the incubation medium oc- 

' Supported by Research Grant Hd 08835 and Train- 
ing Grant HD 00435 from the National Institute of Child 
Health and Human Development. 

^ Author to whom reprint requests should be ad- 

curred between 30- and 60-min periods of 
incubation. The present studies were d^ 
signed to investigate the influence that phos- 
phodiesterase may exert on the cAMP accu- 
mulation phenomenon previously observed 
in porcine GC in response to the stimulatoiy 
effects of FSH and LH. In the present studies 
the phosphodiesterase influence was exam- 
ined indirectly using a potent phosphodies- 
terase inhibitor. 

The effects of phosphodiesterase inhibition 
upon cAMP accumulation by porcine GC 
previously have not been adequately exam- 
ined. The influence of methylxanthine upon 
the stimulatory effects of purified hFSH and 
hLH on porcine GC intracellular cAMP ac- 
cumulation and upon cAMP accumulation in 
the incubation medium was investigated. 
These studies enabled the determination of 
the relative approximate contribution of syn- 
thesis, degradation, and extracellular release 
to cAMP levels occurring in porcine GC dur- 
ing various stages of follicular maturation in 
response to hFSH and hLH. 

Materials and methods. Granulosa cell har- 
vest. Porcine ovaries were obtained from a 
local meat packing plant within IS to 20 min 
of sacrifice of the animals. Granulosa cells 
were harvested from small (1-2 mm), me- 
dium (3-5 mm), and large (6-12 mm) follicles 
according to the method of Channing and 
Ledwitz-Rigby (5). Using dye exclusion as an 
indication of cell viability, the cells were 
counted in a hemocytometer in 0.06% trypan 

Hormones and chemicals. Highly purified 
hLH, LER-1705, having a potency of 3800 
lU/mg and an FSH activity of 3 lU/mg, and 
hFSH, LER-1577^ having an FSH potency 
of 880 lU/mg were used. These two hormone 
preparations were provided by Dr. L. E. 
Reichert, Jr. The FSH preparation as sup- 



"^pyrigbt (£> 1978 by the Society for Experimenttd Biology and Medicine 
f ngAis reserved. 



Dr. Reichert had been pretreated 
motrypsin to inactivate the contam- 
LH selectively (6). The residual LH 
5 reported to be 5.7 lU/mg using the 
ascorbic acid depletion assay (7). Ac- 
to Amir et al (8), controlled chymo- 
ligestion does not destroy FSH activ- 
etennined by the Steehnan-Pohley 

;lls were incubated in the absence or 

of the hormones in Eagle's medium 
ag Earle's salts (pH 7.4; Grand Island 
al Co., Grand Island, N.Y.), 25 mM 

buffer (Calbiochem), 2.2 g/hter 
3 (Grand Island Biological Co.), and 
ae serum albumin (BSA) fraction V 
Chemical Company). This was des- 
Eagle's medium plus 1% BSA. 3-Iso- 
-methylxanthine (MIX) was pur- 
rom the Aldrich Chemical Company 
kee) and was diluted in Eagle's me- 
js 1% BSA. The final concentration 
d for incubations with the cells and 
ropins was 0.2 mM. Both [^H]cAMP 
14 Ci/mmol) and nonlabeled cAMP 
chased from Schwartz Bio-Research, 

losa cell incubations and experimental 
es. Granulosa cells from small and 
follicles were suspended in Eagle's 
plus 1% BSA and dispensed in ali- 
2 X 10^ cells. Cells from large foUi- 
e dispensed in ahquots of 5 x 10^ 
libations were carried out in Packard 
itillation vials containing the appro- 
ormone. When 3-isobutyl 1-meth- 
ae (MIX) was used it was added to 
containing the appropriate hormone 
ials containing no hormone prior to 
of the cells. The final incubation 
per vial was 1.0 ml. Three to five 
aliquots of cells were used for each 
in each experiment. Incubations 
Tied out for 30 and 60 min under 
as previously described (5). The re- 
as arrested by placing the vials im- 
y in ice. The cells were separated 
; incubation medium by centrifuga- 
: incubation medium was decanted 
en for later assay of cAMP content, 
aining cell pellets were subjected to 
jm acetate extraction and following 
;ation the clear supernatant was de- 

canted and frozen for later assay of intracel- 
lular cAMP. 

Cyclic AMP assay. Cyclic AMP accumu- 
lation was determined by a competitive pro- 
tein binding assay (10) with modifications (5, 
1 1). Using 1.25 pmol of cAMP as a standard 
after every 10 unknown samples, the intra- 
assay coefficient of variation was less than 
16% and for 30 randomly selected assays the 
between assay coefficient of variation was 
less than 15%. 

Results, Effect of MIX upon intracellular 
cAMP accumulation. The presence of 0.2 mM 
MIX in the incubation medium did not sig- 
nificantly alter the control levels of intracel- 
lular cAMP in GC harvested from small, 
medium, and large follicles following 30- or 
60-min incubation periods (Table I). Addi- 
tion of 1.0 and 10 /ig of hFSH resulted in an 
increase in intracellular cAMP accumulation 
in GC from small, medium, and large follicles 
(Table I). In the case of cells from small 
follicles, addition of 10 /ig of hFSH led to a 
greater than 13-fold increase (p < 0.001) in 
intracellular cAMP levels following 30-min 
incubations and a greater than 22-fold in- 
crease (/? < 0.001) following a 60-min incu- 
bation period (Tables I and II). In contrast, 
addition of 10 jLig of hFSH to cells from large 
follicles led to less than a 3-fold increase 
above control levels after either 30- or 60-min 
incubation periods {p < 0.001 and /? < 0.01, 
respectively). A small nonsignificant (p > 
0.05) potentiating effect of 0.2 xxiM MIX upon 
the stimulatory effect of 1.0 and 10 jug of 
hFSH upon intracellular cAMP accumula- 
tion was observed (Table I). 

Addition of hLH stimulated intracellular 
cAMP accumulation in GC (Table I). The 
stimulation was greater in the case of GC 
harvested from large compared to medium 
and small follicles. Addition of 0.2 mM MIX 
exerted a small nonsignificant (p > 0.05) 
potentiating effect upon the LH stimulation 
of intracellular cAMP levels in cells from all 
three types of follicles (Table I). If GC were 
incubated for 60 rather than 30 min addition 
of 0.2 mM MIX still had no significant effect 
upon hFSH and hLH stimulation of intracel- 
lular cAMP levels (Table II). 

Effect of MIX upon cAMP released into the 
incubation medium. The presence of 0.2 mM 
MIX in the incubation medium did not sig- 



TABLE I. Comparison of Effect of 0.2 mM 3-Isobutyl 1-Methylxanthine upon hFSH and hLH 
Stimulation of Intracellular cAMP Accumulation in Porcine GC during 30-min Incubations." 

IntraceUular cAMP (pmol/5 x 10^ cells) 

Source of GC and 

Small follicle 
0.1 /ig hFSH 
1.0 Mg hFSH 
10.0 /ig hFSH 

0.01 /ig hLH 
1.0 /ig hLH 

Medium Follicle 
0.1 /ig hFSH 
1.0 /ig hFSH 
10.0 /ig hFSH 

0.01 /ig hLH 
1.0 /ig hLH 

Large follicle 
0.1 /ig hFSH 
1.0 /ig hFSH 
10.0 /ig hFSH 

0.01 /ig hLH 
1.0 /ig hLH 

3-Isobutyl 1-methylxanthine 
Absent Present 

8.1 ±0.7 

10.4 ± 1.3 

62.7 ± 3.7 

91.5 ±9.4 

8.6 ± 2.2 

9.5 ± 1.9 

12.7 ±2.1 

8.9 ± 1.6 

8.0 ±0.3 

38.3 ±6.1 

51.6 ±4.2 

13.2 ± 3.6 

23.5 ± 3.4 

30.8 ± 1.4 

92.9 ±6.2 

85.9 ± 14.8 

184.5 ± 10.3 

204.3 ± 9.9 

134.3 ± 10.0 

236.5 ± 12.0 

278.1 ± 12.4 

8.9 ±0.9 

14.2 ± 2.8 

71.0 ±5.6 

99.3 ±9.3 

8.7 ± 1.3 

12.7 ± 1.9 

15.4 ±2.1 

8.4 ± 1.2 

9.9 ±0.5* 

37.6 ±4.5 

40.4 ± 6.9 

11.3 ±1.0 

30.6 ± 3.2 

36.9 ± 3.5 

109.6 ± 8.2 

95.0 ± 17.7 

191.0 ±21.8 

231.8 ±20.5 

149.4 ±31.0 

272.1 ± 13.9 

314.5 ± 13.4 

' Data are expressed as the means ± SE of four observations. Granulosa cells harvested from small, medium, and 
large porcine follicles were incubated for 30 min with hFSH or hLH in the absence or presence of 0.2 mM MIX and 
the intracellular cAMP levels were determined. Students t test was used to compare results (MIX present vs MIX 
absent). The differences were not statistically significant (p > 0.05) unless indicated. 


nificantly alter control levels of cAMP re- 
leased into the incubation medium by GC 
from any size follicle during 30- or 60-min 
incubation periods (Tables III and IV). Ad- 
dition of 10 jiig of hFSH to GC from small 
follicles led to a 16- and 4S-fold increase in 
incubation medium cAMP levels following 
30- and 60-min incubation periods, respec- 
tively (Tables III and IV). In the case of GC 
from small and medium follicles, addition of 
0.2 mM MIX in the presence of 1.0 (data not 
shown) and 10 jLig of hFSH led to a significant 
increase in incubation medium cAM P con- 
tent (Tables III and IV). In contrast, the 
presence of MIX did not significantly poten- 
tiate the effect of hFSH upon cAMP accu- 
mulation in the incubation medium by GC 
from large follicles (Tables III and IV). 

The presence of MIX brought about a 
significant potentiation of the stimulatory ef- 
fect of 1.0 jiig of hLH upon cAMP released 
into the incubation medium by GC from 
small and medium follicles following 30- and 
60-min incubations (Tables III and TV). In 
the case of GC from large follicles the poten- 
tiating effect of MIX upon hLH stimulation 
of cAMP accumulation in the incubation me- 
dium was not significant (p > O.OS) during 
30- or 60-min incubations (Tables III and 

After a 60-min incubation period with 
either 10 /ig of hFSH or 1.0 /ig of hLH the 
incubation medium cAMP levels were con- 
sistently greater than the intracellular levds 
in the case of cells from all three follicle types 
(Tables II and IV). 



11. Comparison of Effect of 0.2 mM 3- 
. 1-Methylxanthine upon hFSH and hLH 


lULATioN IN Porcine GC during 60-min 

Intracellular cAMP (pmol/5 x 

10' cells) 


3-Isobutyl 1-methylxanthine 


Absent Present 


3.1 4.3 

±0.2 ±0.7 


71.0 75.8 

±4.8 ±3.7 


10.8 13.6 












are expressed as the means ± SE of four 
ns. Granulosa cells harvested from small, me- 

large porcine follicles were incubated for 60 
iFSH or with hLH in the absence or presence 

MIX and the intracellular cAM P levels were 
1. Student's t test was used to compare results 
ent vs MIX absent). The differences were not 
f significant (p > 0.05). 

sion. The lack of a significant poten- 
ffect of MIX on intracellular cAM P 
ation by porcine GC in response to 
>0-min periods of incubation with 
^SH or hLH could indicate that en- 
hydrolysis of cAMP by a phospho- 
e(s) is not a major mechanism re- 
: for controlling the intracellular lev- 
e cyclic nucleotide. Alternatively, it 
le that this methylxanthine does not 
lermeate the GC plasma membrane 
essfuUy inhibit phosphodiesterase or 
x)ncentration employed was not suf- 
> inhibit GC intracellular phospho- 
e(s). It is evident from the findings 
investigators that concentrations of 
iging from 0.01 to 1.0 mAf have 

potentiating effects on cAMP accumulation. 
Methylxanthine has been observed to poten- 
tiate the effect of ACTH upon cAMP levels 
in rat adrenal homogenates and quarters (13) 
and in isolated fat cells (14). Mendelson et al. 
(12) reported that the sensitivity of isolated 
rat testis interstitial cells to hCG stimulation 
was significantly enhanced with the presence 
of 0. 1 mAf MIX and in the absence of MIX, 
cAMP accumulation in response to hCG was 
reduced in magnitude by about 60%. These 
investigators used the sonicated incubation 
mixture for assay of cAMP; thus their re- 
ported findings reflect inclusion of both the 
intracellular and incubation medium cAMP 
content and the site of the potentiating effect 
remains obscure. Channing (IS) observed 

TABLE III. Comparison OF Effect OF 0.2 mA/ 3- 
IsoBUTYL 1 -Methylxanthine vpos hFSH and hLH 

Stimulation of cAMP Accumulation in the 

Incubation Medium by Porcine GC during 30-min 


Incubation medium cAMP 

(pmol/5 X 10' cells) 

Source of GC 
and treatment 

3-Isobutyl 1-methyUanthine 
Absent Present 

SmaU follicle 


7.0 6.9 

±0.7 ±0.8 

10.0 /ig hFSH 

115.3 143.2 

±3.2 ±3.5*** 

1.0 /xg hLH 

6.2 13.4 

±0.7 ±1.4*** 

Medium follicle 


6.5 7.1 

±0.6 ±0.8 

10.0 /xg hFSH 

25.1 49.3 

±2.9 ±3.4*** 

1.0 fig hLH 

10.8 19.8 

±1.6 ±1.4*** 

Large follicle 


26.3 27.6 

±4.9 ±4.6 

10.0 /ig hFSH 

110.2 158.0 

±19.5 ±26.4 

1.0 /ig hLH 

210.9 269.1 

±62.8 ±66.3 

" Data are expressed as the means ± SE of eight 
observations. Granulosa cells harvested from small me- 
dium, and large porcine follicles were incubated for 30 
min with hFSH or with hLH in the absence or presence 
of 0.2 tnM MIX and the incubation medium cAMP 
levels were determined. Student*s t test was used to 
compare results (MIX present vs MIX absent). 

•••/>< 0.001. 



TABLE IV. Comparison of Effect of 0.2 mM 3- 


Stimulation of cAMP Accumulation in the 

Incubation Medium by Porcine GC during 60-min 


Incubation medium cAMP 

(pmol/5 X 

10' cells) 

Source of GC 

3-Isobutyl 1-methylxanthine 

and treatment 



Small follicle 






10.0 Mg hFSH 





1.0 fig hLH 





Medium follicle 






10.0 /xg hFSH 





1.0 /xg hLH 





Large follicle 






10.0 Mg hFSH 





1.0 /ig hLH 





' Data are expressed as the means ± SE of four 
observations. Granulosa cells harvested from small, me- 
dium, and large porcine follicles were incubated for 60 
min with hFSH or with hLH in the absence or presence 
of 0.2 mA/ MIX and the incubation medium cAMP 
levels were determined. Student's / test was used to 
compare results (MIX present vs MIX absent). 



•••/?< 0.001. 

that in 20-min incubations of porcine GC 
from medium-sized follicles, addition of 3.0 
mM aminophylline to incubation medium 
containing either FSH or LH significantly 
increased the concentration of intracellular 
cAMP when compared to the effect of FSH 
or LH alone. The difference in these fmdings 
and the results observed in the present studies 
could be due to differences in the effect of the 
two inhibitors on GC phosphodiesterase ac- 
tivity; it is possible that aminophylline has a 
synergistic effect with the gonadotropins in 
stimulating cAM P production. In another se- 
ries of experiments, addition of theophylline 
alone without gonadotropins to incubations 
of isolated prepubertal rat ovaries resulted in 

a stimulation of cAM P accumulation signifi- 
cantly above control levels in both the tissue 
and in the incubation medium (16). The ef- 
fects of theophylline could have b^n due to 
the indirectly mediated inhibitory influence 
upon protein synthesis or due to a direct 
inhibition of phosphodiesterase (17). 

If cAMP is protected from the hydrolytic 
action of phosphodiesterase by subcellular 
compartmentalization in GC, inhibition of 
the degradative enzymatic activity by meth- 
ylxanthine would not be significantly appar- 
ent. Cheung (18) has shown that cAMP 
bound to the protein kinase regulatory sub- 
unit is not susceptible to phosphodiesterase 
activity and only is degraded when disso- 
ciated from the protein. It was concluded that 
the rate of hydrolysis of cAMP is governed 
by its rate of dissociation from the protein 
kinase regulatory subunit. In the present 
studies it is possible that the lack of a signif- 
icant potentiating effect of methylxan^e 
upon gonadotropin stimulation of intracellu- 
lar cAMP accumulation could have resulted 
from cAM P being bound to the protein ki- 
nase regulatory subunit during the time in- 
tervals examined. Means et al, (19, 20) ob- 
served that when testis were incubated for I 
hr with FSH, the protein kinase remained 
maximally active following an additional 2 
hr of incubation without the gonadotropin 
present. Similar compartmentalization of in- 
tracellular cAMP may occur in porcine GC 
and explain the lack of a significant poten- 
tiating response of the phosphodiesterase in- 

The finding that methylxanthine has a sig- 
nificant potentiating effect upon cAMP con- 
tent in the incubation medium in response to 
either hFSH or hLH stimulation could be 
due to the presence of plasma membrane 
fragments in the incubation medium which 
makes the phosphodiesterase more accessible 
to the inhibitory action of MIX. Alterna- 
tively, it is possible that an extracellular phos- 
phodiesterase may exist and have a role in 
the degradation of cAMP released from the 
GC. It is apparent from these and previous 
studies (4) that significant concentrations of 
cAMP are released extracellularly by porcine 
GC in response to the stimulatory action of 
the gonadotropins. Enzymatic degradation of 
extracellular cAMP has been reported for 



ions of prepubertal rat ovaries using 
ippearance of labeled cAMP as well 
uction of labeled products of cAMP 
ition, indicating that cAMP released 
I incubation medium was undergoing 
liular degradation by a phosphodies- 

)rs influencing the intracellular loca- 
e extracellular release, plasma mem- 
permeability, and metabolism of 
in porcine GC require more defmitive 
before the questions posed can be 

possibility does exist that the MIX 
lave side effects other than inhibition 

nary. In order to examine a possible 
phosphodiesterase in mediation of the 
of LH and FSH upon granulosa cell 
levels, porcine (GC) from small (1-2 
nediiun (3-5 mm), and large (6-12 
>llicles were incubated with human 
iFSH) and LH (hLH) for 30 and 60 
the absence or presence of 3-isobutyl 
yrlxanthine (MIX), a potent phospho- 
ise inhibitor. Subsequently, the intra- 
and incubation medium cAMP con- 
ere determined by a protein binding 
During a 30-min incubation, 10 jLig of 
done brought about an 1 1-fold, 5-fold, 
fold increase in intracellular cAMP 
ilation and a 16-fold, approximately 
and 4-fold increase in incubation me- 
AMP levels in GC from small, me- 
ind large follicles, respectively. Addi- 
0.2 mAf MIX exerted a nonsignificant 
).05) potentiating effect upon hFSH 
tion of intracellular cAMP accumu- 
n cells obtained from the three types 
:les. In the case of cells obtained from 
ind medium foUicles, addition of 0.2 
IX in the presence of 10 jug/ml hFSH 
ig/ml hLH led to 41 to 69% potentia- 
< 0.001) of the effect of the FSH and 
on cAMP accumulation in the incu- 
medium. This was evident after a 30- 
•min incubation period. In the case of 
tained from large follicles, addition of 
f MIX had a nonsignificant potentiat- 
«t (/? > 0.05) on cither hFSH or hLH 
tion of cAMP accumulation in the 
ion medium, 
ly be concluded that probably there 

are low levels of intracellular phosphodiester- 
ase in porcine granulosa cells and that gonad- 
otropins act to stimulate the generation of 
cAMP rather than alter the rate of destruction 
of cAMP. The findings support the existence 
of an extracellular phosphodiesterase which 
may act to regulate or modulate the extracel- 
lular levels of cyclic AMP. 

We thank Dr. Leo Reichert and the National Pituitary 
Agency for provision of the purified human pituitary 

1. Cheung. W. Y.. in "Role of CycUc AMP in CeU 
Function** (P. Greengard and E. Costa, eds.). Vol. 
58, p. 51. Raven Press, New York (1970). 

2. Butcher, R. W., and Sutherland, E. W., J. Biol. 
Chem. 237, 1244(1962). 

3. Beavo, J. A., Rogers, N. L., CrofTord. O. B., Hard- 
man, J. G., Sutherland, E. W., and Newnuui, E. V., 
Mol. Pharmacol. 6, 597 (1970). 

4. Lindsey, A. M., and Channing, C. P., Biol. Reprod. 
in press (1978). 

5. Channing, C. P., Ledwitz-Rigby, F., in "Methods in 
Enzymology** (J. G. Hardman and B. W. 0*Malley, 
eds.). Vol. 39, p. 183. Academic Press, New York 

6. Reichert, L. E., Jr., J. Clin. Endocrinol. Metab. 27, 
1065 (1967). 

7. Parlow, A. F., in "Human Pituitary Gonadotropins** 
(A. Albert, ed.), p. 301. Charles C Thomas, Spring- 
field, lU. (1961). 

8. Amir, S. M., Barker, S. A., Butt, W. R., and Crooke, 
A. C, Nature (London) 209, 1092 (1966). 

9. Steelman, S. L., and Pohley, F. M., Endocrinology 

10. Gilman, A. G., Proc. Nat. Acad. Sci. USA 67, 305 

1 1. Mashiter, K., Mashiter, L., Hauger, R. L., and Field, 
J. B., Endocrinology 92, 541 (1973). 

12. Mendelson, C, Ehifau, M., and Catt, K., J. BioL 
Chem. 250, 8818 (1975). 

13. Peytremann, A., Nicholson, W. E., Liddle, G. W., 
Hardman, J. G., and Sutherland, E. W., Endocri- 
nology 92, 525 (1973). 

14. Beavo, J. A., Rogers, N. L., CrofTord, O. B., Baird, 
C. E., Hardman, J. G., Sutherland, E. W., and 
Newman, E. V., Ann. N.Y. Acad. Sci. 185, 129 

15. Kolena, J., and Channing, C. P., Endocrinology 90, 

16. Selstam, G., Rosberg, S., Liljekvist, J., Gronquist, 
L., Perklev, T., and Ahren, K., Acu Endocrinol. 
(Kbh) 81, 150(1976). 

17. Ryan, R. J., Bimbaumer, L., Lee, C. Y., and Hun- 
zicker-Dunn, M., in "International Review of Phys- 
iology, Reproductive Physiology, IF* (R. O. Grecp, 



cd.). Vol. 13, p. 85. Univ. Park Press, Baltimore, Md. 

18. Cheung, W. Y., Biochem. Biophys. Res. Commun. 

19. Means, A. R., in **Receptors for Reproductive Hor- 
mones, Advances in Experimental Medicine and 
Biology'* (B. W. O'Malley and A. R. Means, eds.). 

20. Means, A. R., Fakunding, J. L., and Tindall, D. J., 
BioLReprod. 14,54(1976). 

21. Selstam, G., Studies on regulatory mechanism of the 
cyclic AMP system in the ovary, Ph.D. dissertation. 
Department of Physiology, University of Gotebotg, 
Goteborg, Sweden (1975). 

Vol. 36, p. 431. Plenum Press. New York (1973). Received June 1, 1978. P.S.E.B.M. 1978, Vol. 159. 


ffect of Big and Little Gastrins on Pancreatic and Gastric Secretion (40322) 


VA Wadsworth Hospital Center and UCLA School of Medicine, Los Angeles, California 

rin exists in several molecular forms, 

which, big gastrin (G34) and little 
(G17), account for most of the gastrin 
:irculation (1). The molar concentra- 

G34 in blood plasma is about twice 
G17. Infusion of equimolar doses of 
Dus G34 and G17 produces approxi- 
equal gastric acid secretory responses 
ds to molar blood concentrations of 
»out five to seven times greater than 
^fleeting the slower removal of G34 
le circulation. 

not known whether the different mo- 
forms of gastrin have different relative 
es for various target organs. To ex- 
this question we studied simultane- 
le gastric acid and pancreatic protein 
ry responses to G34 and G17 in dogs 
istric and pancreatic fistulas. The dog 
rable for such studies since in this 
the doses of gastrin needed to stimu- 
ncreatic protein secretion and gastric 
:retion are in the same range (2). 
'rials and Methods. Natural human un- 
1 little gastrin (G 17-1) and natural por- 
Ifated big gastrin (G34-II) were kind 

Professor R. A. Gregory and Doctor 
Tracy, University of Liverpool, Eng- 
holecystokinin (CCK), 20% pure, was 
sed from the G.LH. Research Unit, 
iska Institutet, Stockholm, Sweden. 
lals. Four dogs weighing 20 to 24 kg 
repared with a Thomas gastric fistula 
id a pancreatic fistula (PF) by a mod- 
lerrera technique (3). Studies were 
no sooner than 4 weeks after surgery. 
»ut not water was withheld for 18 hr 
each test. The interval between tests 
least 48 hr. 

'riments. NaCl (0.15 M) was infused 
nously into a leg vein at 30 ml hr"\ 
ptides were added to the saline infu- 

give the required doses (25, 50, 100, 
K), 800, and 1600 pmol kg"' hr"' of 
5 and 53, 106, 213, 425, and 851 pmol 
r"^ of CCK). Each dose was given 

during 45 min starting with the lowest dose 
and doubling it until the highest dose was 
given. Gastric and pancreatic juices were col- 
lected continuously and separated into 15- 
min samples. Volumes were measured to the 
nearest 0. 1 ml. Acid concentration was deter- 
mined by titrating 0.2-ml samples with 0.2 M 
NaOH to pH 7 on an automatic titrator (Ra- 
diometer, Copenhagen). Total protein con- 
centration was measured spectrophotometric- 
ally at 280 nm, using bovine serum albumin 
as standard. The responses were expressed as 
the mean of the last two 15-min collections 
from each dose. Two tests were done with 
each stimulant in each of three dogs and a 
fourth dog had one test with each stimulant. 
Basals were subtracted from each 15-min 
sample, and results of the two tests in each 
dog were averaged. Before averaging, the 
square root of acid output was computed and 
used in all analyses to make variances more 
uniform and straighten out the response 

Results. G34-II and G17-I were found to 
be approximately equipotent in stimulating 
gastric acid secretion (Fig. 1), confirming ear- 
lier studies (6). The relative potency of G17 
with respect to G34 was 0.7 with 95% limits 
of 0.4 to 1.5 using doses 100, 200, and 400 for 
G 17 and 50, 100, and 200 for G34. 034-11 
and G 17-1 appeared to differ from each other 
in potency in stimulating pancreatic protein 
secretion (Fig. 2). Relative potency of G17 
with respect to G34 was about 0.3 to 0.4, 
depending on the doses used, with limits of 
about 0.1 to 0.6. The response to CCK is 
shown for comparison. CCK did not stimu- 
late acid secretion. Relative potency of CCK 
to G17 was 1.5 (0.99 to 2.4) and to G34 was 
0.5 (0.3 to 0.7). 

The data do not, however, show a signifi- 
cant difference in selectivity for gastric acid 
and pancreatic protein secretion between G17 
and G34. Comparison of the relative potency 
of G 17 to G34 for acid secretion to that for 
protein secretion was made by computing 


0037-9721 /!%/ \!)SlJWiyiVi\.^l^ 



potency of G17 to G34 for each dog sepa- 
rately for acid and for protein. The mean 
differences ± SE for relative potency for acid 
secretion minus relative potency for protein 
secretion were 0.49 ± 0.29 and 0.58 ± 0.30 
depending on the G34 doses used for esti- 
mating protein potency. These differences 
were not significant by paired / test. As a 
further comparison, we computed the equa- 
tion: protein = a -^ b (acid)* for each dog for 
each test. The slopes were similar for G17 
and G34. Figure 3 shows means for pan- 
creatic protein response plotted against gas- 
tric acid response. 

Discussion. These studies show that the 
potency of G34 relative to G 17 is not signif- 
icantly different for gastric acid and pan- 




1 0.25 

y ^ 


!| 0.16 



2 0.09 










^ / 

25 50 100 200 400 


Fig. 1. Acid secretion in response to graded doses of 

I 20 



/ / 



25 50 100 




53 106 213 
DOSE (pmoJ kg-^h-b 



• Q34 










• / 

/ . ' 


S 8 










0.01 0.04 0.09 0.25 0.36 0.49 
ACID SECRETION (iNnel mmC^. M|uai« root Koto) 

Fig. 3. Linear regression of protein secretion on acid 

creatic protein secretion, indicating that one 
of these gastrins is not more selective than the 
other for these targets. Although the present 
results do not show a large difference in 
selectivity, further studies with other gastrins 
or other targets or in other species might 
reveal such differences. 

Summary, In dogs with gastric and pan- 
creatic fistulas the potency of porcine big 
gastrin (G34-II) relative to human little gas- 
trin (G 17-1) was not significantly different for 
stimulation of gastric acid and pancreatic 
protein secretion. 

This work was supported by a Veterans Administra- 
tion Senior Medical Investigatorship (MIG) and bv 
Grant 20971 from the National Institute of Arthritis. 
Metabolism and Digestive Diseases. Doctor ValenzueU 
held a Fogarly International Fellowship (5 F05 TW2197) 
from the National Institutes of Health. Doctor Bugat 
held a fellowship from the French Ministry of Foreign 

We thank Janet Elashoff for statistical assistance. 
Ruth Abercrombie for drawing the figures, and Kuwa 
Chou for typing the manuscript. 

Fig. 2. Pancreatic protein secretion in response to 
graded doses of G34. G 17, and CCK. 

1. Walsh, J. H.. and Grossman, M. 1., N. Engl. J. Med. 
292, 1324 (1975); 292, 1377 (1975). 

2. Stening, G. F., and Grossman, M. 1., Amer. J. Phys- 
iol. 217, 262 (1969). 

3. Hcrrcra, F.. ei al. J. Appl. Physiol. 25, 207 ( 1%8). 

4. Winer, B. J., "Statistical Principles of Experimental 
Design." McGraw-Hill, New York (1962). 

5. Draper. N. R., and Smith, H.. "Applied Regression 
Analysis." Wiley, New York (1966). 

6. Walsh, J. H., Dcbas. H. T., and Grossman, M. I.. J. 
Clin. Invest. 54,477(1974). 

Received June 12, 1978. P.S.E.B.M. 1978, Vol. 159. 


)ifferential Centrifugation Studies of Guinea Pig Lung Proteases (40323) 

rtments of Biochemistry and Physiology and Biophysics, College of Medicine, The University of Iowa, 

Iowa City, Iowa, 52242 

t literature has indicated the pres- 
serveral cathepsins in lung tissue, 
ude preparations, Otto (1) reported 
:ral rat organs, including lung, con- 
ithepsins Bl and B2. McDonald and 
:rs (2), using aqueous extracts of a 
of rat tissues and employing highly 
synthetic substrates, reported that 
Ltained dipeptidylpeptidase I (cath- 

II, III, and IV. Finally, cathepsin D 

isolated and purified from extracts 
»genized rabbit and beef lung (3). 
ly little is known about the properties 
proteolytic enzymes in this tissue, or 
eir distribution among the subcellu- 
ons of lung. A preliminary study of 
ng (4) indicated that the subcellular 

prepared were heterogeneous, that 
zyme markers were widely distrib- 
3ng the fractions, and that classical 
il enzymes appeared to be distributed 
ly in lung than in liver. With lung 
ig several different kinds of cells, one 
Lpect a heterogeneity of organellar 
consequently a wider distribution of 
enzymes in particles of varying size, 
terest in proteolytic enzymes in lung, 
»rmal (S, 6) and pathological (7) con- 
las prompted us to examine the dis- 

of 18 enzymes among five subcel- 
:tions prepared by differential pellet- 
m marker enzymes and eleven pro- 
;nzymes were examined to lay a basis 
»re detailed examination of lung ly- 
and lysosomal proteases and pepti- 

\als and methods. Disruption and sub- 
fractionation. Lungs were obtained 
l-g short-haired, outbred guinea pigs 
K:al colony. The lungs were perfused 
puhnonary artery) with 200 to 300 
ine to remove all blood, minced, and 

rted in part by a grant from The National 
f Health (HL 16920). 

then suspended in cold 0.25 M sucrose solu- 
tion (pH 7.2). The suspended material was 
disrupted by brief homogenization, achieved 
by five up-and-down strokes of a motor- 
driven (1000 rpm) Potter-Elvehjem homog- 
enizer (clearance, 4-6 /un). This homogenate 
protocol consistently resulted in high yields 
of intact lysosomes. The homogenate was 
brought to 10% (w/v) with 0.25 M sucrose 
and was filtered through cheesecloth prior to 
centrifugation. Fractionation of the homoge- 
nates was achieved by 5-fraction differential 
centrifugation following the procedure for 
liver (8) without modification. These frac- 
tions were: nuclear, N (5 lOg x 10 min); heavy 
mitochondrial, M (10,000g x 5 min); light 
mitochondrial, L (40,000g x 10 min); micro- 
somal, P (100,000g X 45 min); and soluble, S 
(nonsedimentable). Total activity of 18 en- 
zymes was investigated in each of these frac- 
tions after treatment with Triton X-100 
(Sigma). The overall concentration of Triton 
X-lOO was 0.2% (w/v). It was used to release 
membrane-bound enzyme activities. This low 
level of Triton did not affect any of the 
enzyme assays. 

Enzyme analysis. All concentrations given 
are final concentrations in the assay mixture. 
Cytochrome oxidase and cathepsin D were 
assayed as described by Canonico and Bird 
(9). Lactate dehydrogenase was detected us- 
ing Sigma Kit 500. 7V-Acetyl-^-glucosamini- 
dase was detected using the (0.02 A/) /?-nitro- 
phenyl derivative (Sigma) in 0.1 M acetate 
buffer, pH 5.0. The reaction was stopped with 
1.25 N NaOH and filtered (Whatman No. 
42), and the absorbance was read at 440 nm 
on a Gilford spectrophotometer. Acid p-ni- 
trophenylphosphatase was assayed as de- 
scribed by Bosmann and Hemsworth (10). 
Alkaline /?-nitrophenylphosphatase was de- 
termined by the procedure of Garen and 
Levinthal (11) except that the pH was held at 
8.8 (where the color is somewhat more in- 
tense) rather than at 8.0. Succinate dehydro- 


0037-9727 /78/ VS91-QIiSM^\ S» |^ 
Ml hglhu rcieTvtd. 



genase was measured using the method of 
Pennington (12) in which the dye 2-(/?-iodo- 
pheny 1 - 3 -/? - nitropheny 1) - 5 - phenyl - tetrazo - 
lium was reduced by succinate to produce 
formazan which was extracted into ethyl ac- 
etate and read at 490 nm. Glucose 6-phos- 
phatase was determined by the method of 
Nordlie and Arion (13) using the sodium 
cacodylate buffer, pH 6.5. Inorganic phos- 
phorus was determined by the method of 
Chen et al (14). Cathepsin A was measured 
using the method of lodice et al. (15) with N- 
carbobenzoxy-a-glutamyl-L-tyrosine (Cyclo 
Chemical Co., Los Angeles, California) as 
substrate. The rate of production of free 
amino groups was monitored with the nin- 
hydrin reagent of Moore and Stein (16). To 
analyze the cathepsin Bl, the method of Bar- 
rett (17) was employed. Dipeptidylpeptidases 
I, II, III, and IV were determined using the 
method of McDonald et al (18). The sub- 
strates used for the fluorimetric assays were 
as follows: dipeptidylpeptidase I, 0.1 mM 
Gly-Arg-)8-naphthylamide (2) in 5 mM 
NaCl-7.5 mM 2-mercaptocthanol-70 mM 
sodium succinate, pH 5.0; dipeptidylpepti- 
dase II, 0.2 mM Lys-Ala-)8-naphthylamide 
in 10 mM 3,3-dimethylglutaric acid, pH 5.5; 
dipeptidylpeptidase III, 0.03 mM Arg-Arg- 
^-naphthylamide in 62.5 mM Tris-HCl, pH 
9.0; dipeptidylpeptidase IV, 0.17 mM 
Gly-Pro-)8-naphthylamide in 20 mM 
Tris-HCl, pH 7.8. Calibration was carried 
out with known standards of )8-naphthyla- 
mine. All )S-naphthylamides were purchased 
from Bachem (Torrance, California). For the 
analysis of elastolytic esterase, the method of 
Visser and Blout (19) was used. In this pro- 
cedure, 0.33 mM /?-nitrophenyl N-tert-bu- 
tyloxycarbonyl-L-alanate (Sigma) was used 
as substrate in 0.05 M sodium phosphate-3% 
acetonitril, pH 6.5. Dipeptidase was assayed 
using the titrimetric assay of Bryce and Rabin 
(20). Glycyl-L-leucine was used as the sub- 
strate. A radiometer litrigraph Type SBR 2c 
was used to keep the pH constant at 8.4 by 
adding standardized acid. Neutral and alka- 
line protease activities were measured on 1% 
heat-denatured casein solutions at pH 7.0 and 
8.5, respectively, similar to the method of 

^ Cyclo Chemical Company's inventory has been pur- 
chased by Vega- Fox Biochemicals, Tuscon, Arizona. 

Kunitz (2 1 ). After 30 min, 10% trichloroacetic 
acid was used to precipitate proteins and 
large peptide fragments. The absorbance of 
the supernatant at 280 nm was used as an 
indication of protease activity. 

Protein was determined by the Biuret 
method of Gornall et al. (22) using bovine 
serum albumin Fraction V (Sigma) as stan- 

Presentation of results. To simplify con- 
struction of tables and graphs, the following 
symbols were used: N = nuclear fraction; M 
= heavy mitochondrial fraction; L = light 
mitochondrial or lysosomal fraction; P = mi- 
crosomal fraction; S = final supernatant or 
cytoplasmic fraction. 

The percentage of an enzyme in any one 
fraction was determined by dividing the ac- 
tivity in that fraction by the total activity 
obtained in the five fractions x 100. The 
percentage recovery was determined by di- 
viding the sum of an enzyme*s activity in the 
five fractions, N, M, L, P, and S, by the 
activity determined on a sample of homoge- 
nate prior to centrifugation, X 100. 

The relative specific activity in each frac- 
tion was obtained as follows: percentage of 
total activity/percentage of total protein x 
10, according to de Duve et al. (8). 

The distributions of the enzymes' activities 
after differential centrifugation are presented 
by plotting the mean relative specific activity 
against the protein content of each fraction. 
The area of each block represents the per- 
centage of the total activity recovered in that 
fraction, and the height corresponds to the 
degree of purification achieved (8). 

Enzyme specific activities are presented in 
milliunits per milligram of protein where I 
unit equals 1 jLimole of substrate hydrolyzed, 
or 1 unit of absorbance released at 280 nm, 
per minute at 37°C. The units for cytochrome 
oxidase are calculated according to Cooper- 
stein and Lazarow (23). 

Results. Enzyme distribution following de- 
ferential centrifugation. The distribution of 18 
enzymes and of the lung protein following 
differential centrifugation are presented io 
Table I, along with the percentage of each 
enzyme recovered. Despite the heterogeneity 
of the lung cell populations, the distribution 
recorded for the various enzymes paralleled 
that found in liver. For example, the major 



TABLE I. Percentage of Total Activity in Tissue Fractions." 


Percentage en- 







iC oxidase (1) 






12.1 ±6.8 

39.2 ± 22.4 

25.5 ± 15.6 

12.7 ± 12.1 

10.5 ± 12.6 

121.7 ± 48.3 

lydrogenue (3) 

8.6 ± 6.2 

3.7 ± 3.4 

3.1 ±0.6 

7.7 ± 1.9 

76.9 ±8.1 

79.2 ± 8.6 

^-glucoMitiinidiffff ( 1 ) 







ophenylphosphauue (3) 

22.6 ± 10.9 

14.5 ± 5.7 

21.5 ± 5.5 

19.8 ± 5.3 

21.5 ± 1.8 

96.0 ± 6.4 


4.8 ±5.1 

13.6 ± 5.7 

20.9 ± 4.5 

29.3 ± 5.6 

31.1 ±9.2 

69.9 ±31.9 

phosphatase (3) 

13.2 ± 6.0 

9.8 ±4.1 

18.3 ± 4.8 

25.8 ± 1.7 

32.8 ± 8.6 

198.6 ±131.0 

A (3) 

19.0 ± 13.6 

5.2 ± 5.2 

31.6 ±14.1 

4.6 ± 6.3 

39.7 ± 3.7 

62.3 ± 24.2 


24.5 ± 13.9 

15.2 ± 21.0 

39.5 ± 24.4 

3.5 ± 4.8 

17.2 ± 8.4 

362.5 ±417.7 


17.5 ± 4.8 

20.5 ± 5.8 

20.8 ± 5.7 

7.6 ± 2.8 

33.6 ± 4.4 

142.2 ± 29.9 

peptidase I (3) 

13.9 ±17.1 

8.9 ± 1.6 

25.6 ± 4.8 

3.4 ± 2.9 

48.2 ± 12.8 

68.4 ± 12.8 

peptidase II (3) 

18.7 ± 8.8 

24.1 ±9.0 

20.2 ± 5.6 

4.2 ± 3.2 

32.7 ± 9.9 

1 14.5 ± 33.6 

peptidase III (3) 

1.8 ± I.I 

5.9 ± 9.0 

4.4 ± 4.7 

2.6 ± 0.7 

85.3 ± 15.2 

147.5 ± 12.2 

peptidase IV (3) 

16.6 ± 14.9 

9.4 ± 3.8 

22.1 ±4.9 

33.5 ± 13.8 

18.4 ± 3.3 

113.2 ±16.2 

esterase (3) 

11.9 ±7.3 

9.2 ± 3.7 

15.5 ± 3.4 

10.5 ± 2.7 

52.9 ± 6.5 

99.2 ± 15.8 


3.9 ± 3.8 

5.5 ± 3.8 

11.3 ±2.0 

29.7 ± 13.1 

49.5 ±21.5 

139.2 ± 30.2 


2.8 ± 3.9 

5.3 ± 4.6 

11.4 ± 1.0 

31.1 ± 13.2 

49.4 ± 20.3 

151.2 ±16.3 








21.3 ± 12.2 

13.2 ± 3.3 

9.9 ± 4.0 

11.5 ±4.5 

44.2 ± 7.2 

104.5 ± 4.7 

indicate the mean percentage ± the standard deviation of the mean. Numbers in parentheses indicate the number of experimenu. The 
enzyme activity and percenuge enzyme recovered were calculated as described in the section under Experimental. 

* the activities of both cytochrome oxi- 
id succinate dehydrogenase is found in 
ivy mitochondrial fraction, and of lac- 
ihydrogenase in the supernatant frac- 
s expected. 

percentage recovery of nine enzymes 
irome oxidase, succinate dehydrogen- 
ctate dehydrogenase, N-acetyl-)8-glu- 
nidase, acid-/?-nitrophenylphospha- 
ipeptidylpeptidase II and IV, elasto- 
terase, and dipeptidase) and of protein 
ood (i.e., 79-122%); the recoveries of 
jnzymes (alkaline /?-nitrophenylphos- 
e, cathepsin A, and dipeptidylpepti- 
) were low (62-70%) whereas six en- 
(glucose 6-phosphatase, cathepsins Bl 
, dipeptidylpeptidase III, and neutral 
kaline protease) showed significantly 
total activity in the sum of the frac- 
han in the whole homogenate (Table 
s possible that fractionation removed 
[ihibitor of these enzymes and allowed 
atcr expression of total activity in the 
ns. This has already been demon- 
in our laboratory, where the addition 
tail aliquot of the supernatant fraction 
light mitochondrial fraction strongly 
ed cathepsin Bl activity, as measured 
e hydrolysis of benzoyl-arginyl-^- 
lylamide (24). 

itive specific activities. The relative spe- 
ctivity of each enzyme in each of the 

five tissue fractions is presented in Table II. 
These values were plotted vs the percentage 
protein in each fraction to give the graphs 
which are presented in Fig. 1. 

Cytochrome oxidase and succinate dehy- 
drogenase, two mitochondrial markers, were 
enriched in the heavy mitochondrial fraction, 
M, and to a lesser extent in the light mito- 
chondrial fraction, L. 

7V-Acetyl-^-glucosaminidase, acid /?-nitro- 
phenylphosphatase, dipeptidylpeptidase I, di- 
peptidylpeptidase II, cathepsin A, cathepsin 
Bl, cathepsin D, and elastolytic esterase all 
showed greatest enrichment in the light mi- 
tochondrial fraction, L. Among these en- 
zymes there appeared to be two separate 
patterns of distribution. Cathepsin A, cath- 
epsin Bl, and dipeptidylpeptidase I appeared 
to distribute so that the light mitochondrial 
fraction was greatly enriched over the neigh- 
boring fractions. On the other hand, 7V-acetyl- 
^-glucosaminidase, cathepsin D, dipeptidyl- 
peptidase II, acid /?-nitrophenylphosphatase, 
and elastolytic esterase distributed through- 
out the fractions such that the light mitochon- 
drial fraction, L, was only slightly enriched 
over the neighboring fractions. In this second 
class of enzymes, the distribution throughout 
the fractions seemed to be broader than the 
first class. 

The microsomal fraction, P, was enriched 
in glucose 6-phosphatase, alkaline /?-nitro- 



TABLE 11. Relative Specihc Activities in Tissue Fractions.* 








Cytochrome oxidase ( 1 ) 





Succinate dehydrogenase (3) 

0.9 ± 0.8 

3.8 ± 1.4 

3.5 ±1.1 

1.5 ± 1.2 

0.3 ±0.4 

1.1 ±0.3 

0.6 ± 0.2 

1.0 ±0.3 

2.0 ± 0.3 

5.1 ±0.6 

A^- Acetyl-/3-glucosaminidase ( 1 ) 






Acid ;?-nitrophenylphosphatase 

Alkaline ;?-nitrophenylphospha- 

1.6 ± 0.7 

1.7 ±0.8 

3.3 ± 0.4 

2.6 ± 0.4 

0.7 ± .03 

0.5 ± 0.5 

1.4 ±0.2 

3.2 ± 0.4 

3.9 ± 0.5 

1.0 ±03 


Glucose 6-phosphatase (3) 

1.1 ±0.6 

1.3 ±0.8 

2.9 ± 0.2 

3.7 ± 0.9 

1.1 ±0.3 

Cathepsin A (3) 

1.7 ± 1.6 

0.6 ± 0.5 

5.5 ± 0.8 

0.6 ± 0.7 

1.6 ±0.5 

CathepsinBl (3) 

2.0 ± 1.7 

1.2 ± 1.5 

5.7 ± 1.7 

0.4 ± 0.5 

0.6 ± 0.3 

Cathepsin D (3) 

1.5 ±0.4 

2.6 ± 1.0 

3.6 ± 0.5 

1.1 ±0.4 

1.2 ±0.3 

Dipeptidylpeptidase I (3) 

0.7 ± 0.5 

1.4 ±0.7 

5.2 ± 0.8 

0.5 ± 0.3 

2.1 ±0.8 

Dipeptidylpeptidase II (3) 

1.6 ±0.4 

3.1 ±0.7 

3.6 ± 0.5 

0.6 ± 0.3 

1.2 ±0.3 

Dipeptidylpeptidase III (3) 

0.3 ± 0.3 

0.8 ± 0.9 

1.5 ± 1.6 

0.8 ± 0.2 

6.6 ± 2.6 

Dipeptidylpeptidase IV (3) 

1.0 ±0.2 

1.0 ±0.1 

3.4 ± 0.7 

5.1 ±2.4 

0.6 ± 02 

Elastolytic esterase (3) 

1.1 ±0.4 

1.4 ±0.8 

3.3 ± 0.8 

1.9 ± 0.5 

2.3 ± 0.5 

Neutral protease (3) 

0.5 ± 0.5 

0.8 ± 0.7 

2.1 ±0.5 

4.5 ± 0.7 

2.0 ± l.l 

Alkaline protease (3) 

0.4 ± 0.5 

0.9 ± 0.8 

2.2 ± 0.6 

4.6 ± 0.7 

2.0 ± 0.9 







" Values are the mean relative specific activity ± standard deviation of the mean. Numbers in parentheses are the 
number of experiments. 

phenylphosphatase, neutral protease, dipep- 
tidylpeptidase IV, and alkaline protease. For 
these enzymes considerable activity was also 
found in the light mitochondrial fraction. 

Three enzymatic activities were found in 
the cytosol: lactate dehydrogenase, dipepti- 
dase, and dipeptidylpeptidase III. The spe- 
cific activities for the 18 enzymes in guinea 
pig lung are presented in Table III. 

Discussion. This study of differential cen- 
trifugation, combined with the biochemical 
analysis of marker enzymes, satisfies the cri- 
teria of de Duve et al.y (8) for separation of 
organelles. The fact that the mitochondrial 
enzymes, cytochrome oxidase and succinate 
dehydrogenase, the lysosomal enzyme, ^- 
acetyl-)3-glucosaminidase, the microsomal 
enzyme, glucose 6-phosphatase, and the cy- 
tosol enzyme, lactate dehydrogenase, were 
enriched in the fractions M, L, P, and S, 
respectively, indicated that the experimental 
procedure employed was capable of resolving 
to some degree the designated subcellular 
organelles. The somewhat broad distributions 
observed with these markers indicated that 
the fractions were heterogeneous in the or- 
ganelles they contained. This was confirmed 

20 40 60 80 2C 40 60 80 


Fig. 1. Distribution patterns of enzymes after differ- 
ential centrifugation. Fractions are N, nuclear. M, heavy 
mitochondrial; L, light mitochondrial; P, microsomal; 
and S, nonsedimentable. Enzyme abbreviations are 

DAP, dipeptidylaminopeptidase or dipeptidylpeptidase; 
PNP-Pase, ;?-nitrophenylphosphatase. 



III. Specific Activities of 

Various Lung 


ac oxidase (5) 

50.2 ± 7.8 

dehydrogenase (4) 

12.3 ± 2.2 

A (4) 

46.3 ± 20.4 


3.4 ± 0.6 

Ipeptidase I (2) 



11.7 ±2.7 

esterase (4) 

31.7 ±2.0 

Ipeptidase II (2) 


3-glucosaininidase (5) 

5.2 ± 1.7 

rophenylphosphatase (5) 

5.3 ± 1.7 



■phosphatase (2) 


Ipeptidase IV (2) 


rotcase (2) 


otease (2) 


Ipeptidase III (2) 




hydrogenase (2) 


alues (obtained with the whole homogenate) 
lits per milligram of protein ± the standard 
e mean. The numbers in parentheses are the 

>scopic examination. The broad dis- 
LS seen in guinea pig lung were also 
in rabbit lung (4). 

ither enzymes were used as secondary 
Acid /'-nitrophenylphosphatase 
id as a lysosomal marker in spite of 
that isoenzymes of the true acid 
tase exist in different parts of the cell 
lile the distribution of /'-nitrophen- 
latase activity was broad, it did cor- 
to that of a lysosomal enzyme. Al- 
7-nitrophenylphosphatase was also 
a secondary marker of the '*micro- 
since this enzyme has generally been 
I as being a component of the plasma 
ne (26). Fragmented plasma mem- 
>f rat kidney (27) and liver (28) have 
md to sediment with the microsomal 

istribution of the dipeptidylpeptidase 
illed dipeptidylaminopeptidase) en- 
1 lung parallels their distributions in 
sues. Dipeptidylpeptidase I (or dipep- 
inopeptidase I or cathepsin C) has 
calized in rat liver (2) and bovine 
2). A cytosol distribution was noted 
ptidylpeptidase III from bovine an- 
ituitary (2). Dipeptidylpeptidase IV 
1 shown to have a microsomal distri- 
n porcine kidney (2) and rat liver (2). 
nber of proteolytic enzymes has been 

localized in subcellular fractions in tissues 
other than lung, largely using the technique 
of differential centrifugation. Cathepsin A 
and another carboxypeptidase-like enzyme 
appeared to be found in the heavy mitochon- 
drial fraction (29); cathepsins B (29) and D 
(30) were lysosomal in origin; and di- and 
tripeptidases in different tissues have been 
variously reported as being in the supernatant 
(29) and in the microsomal fractions (31, 32). 
Lung cathepsins A, Bl, and D all appeared 
to be lysosomal in nature. The lung dipepti- 
dase activity was found to be clearly associ- 
ated with the cytosol fraction, as assayed with 
Gly-L-Leu, Gly-Gly, Gly-DL-Phe, or 
Gly-DL-Ser. Of eight dipeptides tested with 
this enzyme, the most effective substrate was 
Gly-L-Leu, the data for which are reported 
here. No tripeptidase activity against 
Gly-Gly-Gly or L-Leu-Gly-Gly was noted. 

Elastolytic esterase was determined by the 
rate of breakdown of a synthetic substrate, /?- 
nitrophenyl ^'-tert-butyloxycarbonyl-L-alan- 
ate. The enzyme present could not be de- 
tected using the orcein-elastin assay (33). The 
failure to react with the latter substrate could 
have been due to extremely low levels of 
elastase or to the fact that this enzyme was 
not a true elastase. Such an enzyme has been 
recently characterized from human pancreas 
(34). We have, therefore, chosen to call the 
enzyme measured, elastolytic esterase. This 
enzyme was enriched most in the light mito- 
chondrial fraction, L, but was also present in 
the cytosol in sizeable quantities. The possi- 
bilities of a dual location of the same enzyme 
or of two different enzymes remain for con- 

Neutral and alkaline proteases distributed 
with the microsomal enzyme markers. The 
similarity of distribution and the method of 
assay of the two proteases would leave open 
the possibility that the same enzyme is being 
measured at two different pH values. 

The distribution of enzymes noted in this 
work does not differ significantly from the 
distributions of similar enzymes in other tis- 
sues. Further work performed on guinea pig 
lung using isopycnic-zonal centrifugation to 
obtain better resolution of fractions will be 

Summary. Five subcellular fractions were 
isolated from guinea pig lung homogenates 
by differential centrifugation. These fractions 



were defined biochemically by the analysis of 
18 enzymes representing different subcellular 
compartments. Succinate dehydrogenase and 
cytochrome oxidase distributed with the 
heavy mitochondrial fraction, while ^'-acetyl- 
)3-glucosaminidase, acid /?-nitrophenylphos- 
phatase, cathepsins A, Bl, and D, dipeptidyl- 
peptidases I and II, and elastolytic esterase 
distributed with the light mitochondrial frac- 
tion. Alkaline /?-nitrophenylphosphatase, glu- 
cose 6-phosphatase, dipeptidylpeptidase IV, 
neutral protease, and alkaline protease all 
demonstrated a ''microsomal" enrichment. In 
the cytosol were found lactate dehydrogen- 
ase, dipeptidylpeptidase III, and a dipepti- 
dase. The lung subcellular fractions were het- 
erogeneous with cross-contamination be- 
tween the heavy mitochondrial, light mito- 
chondrial, and ''microsomal'* fractions. The 
enzyme distributions noted were similar to 
those found in other tissues. 

1. Otlo» K.. in *Tissuc Proteinases" (A. J. Barrett and 
J. T. Dingle, eds.)« p. 1. American Elsevier, New 
York (1971). 

2. McDonald, J. K., CaUahan, P. X., Ellis, S., and 
Smith, R. E., in *Tissue Proteinases" (A. J. Barrett 
and J. T. Dingle, eds.), p. 69. American Elsevier, 
New York (1971). 

3. Rojas-Espinosa, O., Dannenberg, A. M., Murphy, P. 
A., Straat, P. A.. Huang, P. C, and James, S. P., 
Infect. Immunity 8, I(X)0 (1973). 

4. Hook, G. E. R., Bend, J. R., Hoel D., Fouts, J. R., 
and Gram, T. E., J. Pharmacol. Exp. Ther. 182, 474 

5. Ihnen, J., and Kabiitsky, G., in '^Intracellular Protein 
Caubolism, 11" (V. Turk and N. Marks, eds.), p. 
259. Plenum Press, New York (1977). 

6. Singh, H., and Kahiitsky, G., J. Biol. Chem., in 

7. Kahiitsky, G., Singh, H., Kuo, T., and Richerson, H. 
B., Fed. Proc. 3d, 1091(1977). 

8. de Duve, C, Pressman, C, Gianetto, R., Wattiaux, 
R., and Appehnans, F., Biochem. J. 60, 604 (1956). 

9. Canonico, P. G., and Bird, J. W. C. J. Cell Biol. 45, 
321 (1970). 

10. Bosmann, H. B., and Hemsworth, B. A.. Physiol. 

Chem. Phys. 2,249(1970). 

11. Garen, A., and Levinthal, C, Biochim. Biophys. 
Acu 38, 470 (1960). 

12. Pennington, R. J., Biochem. J. 80, 649 (1%1). 

13. NordUe, R. C, and Arion, W. J., in "Methods in 
Enzymology," Vol. IX, p. 619. Academic Press, New 
York (1966). 

14. Chen, P. S., Toribara, T. Y., and Warner, H., Anal. 
Chem. 28, 1756(1956). 

15. lodice, A. A., Leong, V., and Weinstock, I. M., Arch. 
Biochem. Biophys. 117, 477 (1966). 

16. Moore, S., and Stein, W. H., J. Biol. Chem. 211,907 

17. Barrett, A. J., Anal. Biochem. 47, 280 (1972). 

18. McDonald, J. K., Zeitman, B. B., Reilly, T. J., and 
Ellis, S., J. Biol. Chem. 244, 2693 (1%9). 

19. Visser, L., and Blout, E. R., Biochim. Biophys. AcU 

20. Bryce, G. F., and Rabin, B. R., Biochem. J. 90, 509 

21. Kunitz, M., J. Gen. Physiol. 30, 291 (1947). 

22. Gomall, A. G., Bardawill, C. J., and David, M. M., 
J. Biol. Chem. 177, 751 (1949). 

23. Cooperstein, S. J., and Lazarow, A., J. Biol. Chem. 

24. Singh, H., and Kahiitsky, G., J. Biol. Chem. 253, 

25. Neil, M., and Homer, M. W., Biochem. J. 92, 217 

26. Bosmann, H. B., Hagopian, A., and Eylar, E. H., 
Arch. Biochem. Biophys. 128, 51 (1968). 

27. GrifTm, M. J., and Cox, R. P., Nature (London) 204, 

28. Amar-Costesec A., Beaufay, H., Wibo, M., Thines- 
Sempoux, D., Feytmans, E., Robbi, M., and Bertbet, 
J., J. Cell Biol. 61, 201 (1974). 

29. Rademaker, W. J., '*De localizatie van enige pro- 
teasne in der levercel," Kemink en zoon, Ph.D. 
thesis, Utrecht (1959). 

30. Beaufay, H., Bendall, D. S., Baudhuin, P., Wattiaux, 
R., and de Duve, C, Biochem. J. 73, 628 (1959). 

31. Binkley. F., J. Biol. Chem. 23d, 1075 (1961). 

32. Weiss, B., J. Biol. Chem. 205, 193 (1953). 

33. Sachar, L. A., Winter, K. K., Sicher, N., and Frankel 
S., Proc. Soc. Exp. Biol. Med. 90, 323 (1955). 

34. Mallory, P. A., and Travis, J., J. Biochem. 14, 722 

Received March 23, 1978, P.S.E.B.M. 1978. Vol 159. 


tuations of Human Pancreatic Polypeptide In Plasma: Effect of Normal Food 
Ingestion and Fasting^ (40324) 


Ctmica Puerto de Hierro, Universidad Autonoma de Madird, Madrid 35, Spain 

^cretion of human pancreatic poly- 
hPP) is stimulated by food ingestion 
nee this response persists for several 
, it could be predicted that under the 
dietary habit of three meals a day 
lPP would be elevated above fasting 
ring most of the daytime. Thus, we 
imined the daily fluctuations of cir- 
hPP in normal individuals subjected 
a meal schedule and in an inverse 
, i.e., during prolonged fasting. In 
, the effect of the ingestion of a low- 
bulky meal as well as tap water on 
etion was examined. 
ials and methods. Healthy, nonobese 
participated in this study. Their ages 
rom 20 to 24 years. Informed consent 
lined. In a group of seven male vol- 
iPP plasma levels were measured at 
ntervals from 8:30 to 24:00 hr while 
)Uowing meal schedule: breakfast at 
> cup of coffee with milk and two 
s); lunch at 13:00 hr (300 g of boiled 
i, 200 g of grilled beef, and one pear); 
t 20:00 hr (vegetable salad, 200 g of 
ake, SO g of white bread, and one 
1 a second group of 12 volunteers 
tales and four females) fasting was 
led for 84 hr. They received water ad 
ind 40 mEq of K"^ daily. Upon ter- 
i of the experiments, a body weight 
.S±0.2 kg was recorded. Blood sam- 
e obtained 12, 18, 24, 36, 42, 48, 60, 
ind 84 hr after the last meal, which 
;n in the evening (21:00 hr) prior to 
rvation period. Volunteers were ad- 
3 our clinical research center on the 
n preceding the experiments. In fur- 
eriments, six volunteers (three males 
e females) were given either 400 g of 
e salad (250 g of asparagus and 150 
uce) or 500 ml of tap water on two 

•rted in part by a grant (12-130-77) from the 
lacional de Prevision, Spain. 

different days. These tests were performed 
after an overnight fast. 

The collection and processing of blood 
samples has been previously described (3). 
Plasma glucose was determined by means of 
a commercial glucose-oxidase preparation 
(Biochemica Test Combination, Boehringer 
Mannheim GmbH). Radioimmunoassay was 
used to estimate insulin (4), glucagon (5), and 
hPP (6). Results are expressed as 
means±SEM. Differences between values 
were calculated for significance by paired / 
test analysis. 

Results. Figure 1 shows the daily fluctua- 
tions of plasma hPP levels in a group of seven 
subjects kept on a conventional meal sched- 
ule. Mean fasting hPP concentration was 61 
± 15 pg/ml. Ingestion of each meal was 
followed by a sustained hPP elevation. After 
breakfast plasma hPP rose to 158±35 pg/ml 
at 1 1:30 hr (/? < 0.01) while lunch and dinner 
elicited more marked increases (551 ±131 
pg/ml at 15:00 hr, /? < 0.01; 640±153 pg/ml 
at 20:30 hr,/? < 0.01, respectively). It is note- 
worthy that between meals circulating hPP 
did not return to basal values. As expected, 
following each meal the concentrations of 
glucose and insulin in plasma increased in a 
parallel fashion. 

In view of the apparent association of hPP 
secretion with the consumption of food, we 
tested the effect of a low-calorie, bulky meal 
on plasma hPP (Fig. 2). This meal elicited a 
sixfold increase of hPP concentration with 
only a small rise of plasma insulin and glu- 
cose. The ingestion of even 500 ml of tap 
water (Fig. 3) more than doubled the levels 
of circulating hPP. 

In Fig. 4 are depicted the mean hPP, glu- 
cagon, insulin, and glucose plasma levels for 
a group of 12 volunteers subjected to 84 hr of 
fasting. Basal (after a 12-hr overnight fast) 
hPP concentration was 61 ± 16 pg/ml. Pro- 
longing of fasting resulted in an increase of 
circulating hPP, which became statistically 


0037-9721 n%l \S91-014!>Vi\.^l^ 
AU h|^U reaerved. 



significant 24 hr after the last meal (181 ±53 
pg/ml, p < 0.02) and persisted elevated dur- 
ing the remainder of the experimental period. 
It is remarkable that for each day plasma 
hPP showed a distinct pattern, with a pro- 
gressive rise from 9:00 to 21:00 hr and a 



II I H.7 




09 K) 11 12 13 U IS )d 17 18 19 20 21 22 23 24 

Fig. 1. Daily fluctuations of plasma pancreatic poly- 
peptide in normal subjects under conditions of normal 
food ingestion (mean±SEM). The large dots represent 
statistically signiflcant differences from the baseline val- 

subsequent decline during the ni 
overall curve, however, exhibited a 
ing trend. Finally, during fasting pL 
cose and insulin declined while glue 

Discussion. The foregoing data co 
stimulatory effect of food intake 
creatic polypeptide secretion in man 
that the ingestion of a fiber-rich me 
as plain water provokes hPP release 
that the hPP response to food rep] 
part a nonspecific effect, perhaps t 
quence of gastric distention as point 
Schwartz et al (7). Furthermore, o 
demonstrate that under normal die 
ditions, the successive postprandia 
circulating hPP maintain its levels a 
ing values throughout the daytime, 
the physiological role of pancres 
peptide remains enigmatical, it is 
to have the category of a digestive 
since the administration of the bovir 
in dogs modifies gastric and pancre< 
tion as well as gastrointestinal and 
motility (8). Contextually, the persi 
vation of plasma hPP may be thougl 
a tonic influence on some of these 
On this basis, in conditions of food 
tion a decrease of circulating hPP 


I I I I I f I I I 1 1 1 I 1 1— — 

20 40 60 80 100 120 MO 160 180 


Fig. 2. Effect of ingestion of a vegetable meal on pancreatic polypeptide plasma levels in norm 
(mean±SEM). The large dots represent statistically significant differences from the baseline values. 




500 ml 

^^ I Glucose 
90 i mg/dl 

80-1 ^ ^ 



4 H H- ^ 

-20 -10 



40 50 60 

x:t of up water ingestion on pancreatic polypeptide plasma levels in normal subjects (mean±SEM). 
represent statistically significant differences from the baseline values. 






09 15 21 
12 18 24 

09 15 21 
36 42 48 

09 15 2« 
60 66 72 

creatic polypeptide plasma levels during 
ng in normal subjects (mean±SEM). The 
-esent sutistically significant differences 
ne values. 

J reported for gastrin (9). How- 
ged fasting resulted in a progres- 
; of this factor in blood, an obser- 
;reement with that of Floyd and 
I). Moreover, in the absence of 
plasma hPP showed circadian 
vith higher concentrations in the 

late evening than in the preceding and sub- 
sequent morning. A similar pattern was ob- 
served by the above-mentioned authors with 
determinations at 8:00 AM and 4:00 PM. In 
interpreting the rise of plasma hPP during 
fasting, the concomitant decline of glycemia 
should be considered, since even a modest 
fall of blood sugar provokes hPP secretion (1, 
6). Also, as described for glucagon (10), the 
possibility of diminished metabolic clearance 
of hPP should be contemplated. However, 
either of these alternatives fails to explain the 
circadian oscillations of hPP. Current evi- 
dence indicates that parasympathetic stimu- 
lation induces hPP secretion (11, 12) and, 
thus, changes in vagal tonus may afTect cir- 
culating hPP. Accordingly, the reduction of 
vagal tonic activity associated with sleep (13) 
could be responsible for the low hPP plasma 
levels found in the morning. In man, during 
a 24-hr fast a circadian rhythm of gastric acid 
secretion, with greater output in the evening 
than in the morning has been documented 
(14). The relationship between this phenom- 
enon and the parallel changes of hPP remains 

In any case, the understanding of the par- 
adoxical rise of plasma hPP in both anabolic 
(feeding) and catabolic (fasting) situations 
awaits a better knowledge of the biological 
activity of this putative hormone. 



Summary, In this work wc have examined 
the daily fluctuations of circulating hPP in 
normal individuals subjected to a conven- 
tional meal schedule (breakfast, lunch, and 
dinner) as well as during food deprivation for 
84 hr. In addition, we have tested the effect 
of ingestion of a low-calorie, fiber-rich salad 
as well as SOO ml of tap water on hPP secre- 

Ingestion of each meal was followed by a 
sustained hPP elevation. Between meals, cir- 
culating hPP did not return to basal values. 
Both the vegetable meal and the water load 
evoked hPP release, suggesting that the hPP 
response to food intake is partially a nonspe- 
cific effect. In the fasted group, plasma hPP 
rose significantly 24 hr after the last meal and 
persisted elevated for the remainder of the 
experimental period. Moreover, in this con- 
dition hPP showed circadian variations, with 
higher values in the late evening than in the 
preceding and subsequent morning. 

Since pancreatic polypeptide is suspected 
to possess gastrointestinal functions, its ele- 
vation in plasma throughout the daytime in 
conditions of normal feeding may be thought 
to exert a tonic influence on some digestive 
process. On this basis, the increase of hPP 
during prolonged fasting appears paradoxical 
and, indeed, the explanation of this phenom- 
enon awaits a better knowledge of the biolog- 
ical activity of this peptide. 

Ms. Pilar Garcia, and Ms. Begona Samper is gr 

1. Floyd, J. C. Jr.. Fajans, S. S.. and Pck, S. 
Assoc. Amer. Physicians 89, 146 (1976). 

2. Schwartz, T. W., Rchfeld, J. F., Sudil, F., 
L.-I., Chance, R. E., and Moon, N., Lancet 

3. Marco, J., Hedo, J. A., MartineU, J., CaUe, 
Villanueva, M. L., J. Clin. Endocrinol. Mc 
215 (1976). 

4. Herbert, V., Uu, K.-S., Gottlieb, C. \ 
Bleicher, S. J., J. Clin. Endocrinol. Metab. i 

5. Faloona, G. R., and linger, R. H.. in "Met 
Hormone Radioimmunoassay** (B. M. JafTc 
R. Behrman, eds.), p. 317. Academic Pre; 
York (1974). 

6. Marco, J., Hedo, J. A., and Villanueva, N 
Clin. Endocrinol. Metab. 46, 140 (1978). 

7. Schwartz, T. W., and Rehfeld, J. J., Lance 

8. Lin, T. M., Evans, D. C, Chance, R. E., ani 
G. F., Amer. J. Physiol. 232(3), E31 1 (1977) 

9. Lichtenberger, L. M., Lechago, J., and Johi 
R., Gastroenterology 68, 1473 (1975). 

10. Fisher, M., Sherwin, R. S., Hendler, R., an 
P., Proc. Nat. Acad. Sci. USA 73, 1735 (197i 

11. Schwartz, T. W., Hoist, J. J., Fahrenknig. J 
kaer, S., Nielsen, O. V., Rehfeld, J. F., Schaf 
O. B., and Sudil, F., J. Clin. Invest. 61, 781 

12. Hedo, J. A., Villanueva, M. L., and Marc 
Clin. Endocrinol. Meub., 47, 366 (1978). 

13. Uichnetz, G. R., Exp. Neurol. 35, 194 (1972 

14. Moore, J. G., and Englert, E., Jr., Nature (1 
226, 1261 (1970). 

The expert technical assistance of Ms. Ana Ramirez, Received May 16, 1978. P.S.E.B.M. 1978, Vol. l: 


Mechanism of Prostaglandin E2 Stimulation of Renin Secretion (40326) 


Departments of Human Development, Physiology, and Pharmacology, Michigan State University, 

East Lansing, Michigan 48824 

'enal infusion of prostaglandins or the 
andin precursor, arachidonic acid, 
en shown to stimulate renin secretion 
(1), rats (2), and rabbits (3). In addi- 
dbition of prostaglandin synthesis de- 
endogenous renin secretion (3), renin 
1 in response to hemorrhage (4), and 
ide-stimulated renin secretion (S). In 
which renin release has been blocked 
methacin, infusion of prostaglandin 
^) significantly increased the release 
I beyond the original control values 

nechanism by which PGE2 increases 
cretion may involve one or a combi- 
->{ three factors. First, the hormone 
vre a direct effect upon the juxtaglo- 
apparatus. Second, PGE2 may acti- 
vascular baroreceptor mechanism by 
;tation of the renal vasculature (7). 
*GE2 may stimulate a tubular macula 
ceptor since in addition to decreasing 
distance, PGE2 infusion also increases 
klium excretion (6). In the present 
ents, the mechanism by which PGE2 
es renin secretion was evaluated by 
son of the effect of PGE2 on renin 
1 with the vasodilating agents acetyl- 
bradykinin and eledoisin. 
ials and methods. Male mongrel dogs 
esthetized with sodium pentobarbital 
kg iv). Following insertion of a cuffed 
:heal tube, dogs were artificially ven- 
Harvard Apparatus, Inc.). A femoral 
id two femoral veins were cannulated 
recording of arterial blood pressure, 
of inulin (3% solution at 1 ml/min), 
ision of saline. Blood pressure was 
i with a strain gauge pressure trans- 
Itatham P23AA) and a direct writing 
aph (Grass polygraph). The left kid- 
exposed via a flank incision and the 
)f both kidneys were cannulated with 
f^lene tubing. A noncannulating elec- 
letic flowmeter probe (Carolina Med- 

ical Electronics) was placed on the renal ar- 
tery and renal blood flow was recorded on 
the oscillograph. Renal venous blood samples 
were collected by placing a curved 20-gauge 
needle attached to polyethylene tubing into 
the renal vein. A curved 22-gauge needle 
attached to polyethylene tubing was inserted 
into the renal artery distal to the flow probe 
for the intrarenal infusion of PGE2, acetyl- 
choline, bradykinin, and eledoisin. Each dog 
was hydrated prior to the experiment with a 
solution containing 140 mEq/liter sodium 
chloride and 3.0 mEq/liter potassium chlo- 
ride, infused at S.O ml/min until the total 
urine flow rate reached 0.5 to 1.5 ml/min. 
The infusion rate was then decreased to equal 
the urine flow rate. Experiments were begun 
1 hr following the completion of surgery. 

In each experiment, two control clearance 
periods of 10 min duration each were fol- 
lowed by the infusion of one of the vasodi- 
lating agents. The rate of infusion of the drug 
was adjusted to increase renal blood flow 20 
to 40%. Two additional clearance periods 
were obtained. Systemic arterial and renal 
venous blood samples were collected at the 
midpoint of each clearance period. Drug in- 
fusion was stopped and a 30-min period en- 
sued during which renal blood flow returned 
to control levels. Drug metabolism was as- 
sumed to be complete when RBF was stable 
again and the next control period and drug 
treatment were begun. The order of admin- 
istration of acetylcholine, bradykinin, and 
prostaglandin E2 was randomized throughout 
the experiments. Eledoisin was always ad- 
ministered last, due to its presumed slower 
rate of metabolism. The range of doses of 
each vasodilator used were as follows: acetyl- 
choline, 210 to 420 ng/kg/min; bradykinin, 
7 to 21 ng/kg/min; eledoisin, 15 to 32 
ng/kg/min; and prostaglandin E2, 14 to 60 

Analytical and statistical procedures. 
Plasma and urine inulin concentration were 


0037-9727 /7%/ VS^l-OlA^^I^^I^ 



determined by the diphenylamine method 
described by Walser et al (8). GFR was 
estimated by the clearance of inulin. Plasma 
renin concentration was determined by ra- 
dioimmunoassay for the generated angioten- 
sin I (9). Hematocrit was measured on all 
arterial blood samples by the micromethod. 
Renal plasma flow was calculated from the 
renal blood flow and hematocrit. Sodium and 
potassium concentration of both plasma and 
urine were determined by flame photometry 
and the electrolyte excretion rates were cal- 
culated. Renin secretion was calculated as the 
product of the renal venous-arterial renin 
concentration difference and renal plasma 
flow. Renal blood flow and renin secretion 
mean differences were tested by a paired / 
analysis. Sodium and potassium excretion 
was calculated as the percentage increase 
from control and treatments were compared 
by one-way analysis of variance. The 0.05 
level of probability was used as the criterion 
of significance. 

Results. Infusion of PGE2 significantly in- 
creased renal blood flow (Fig. 1). The in- 
crease in renal blood flow was associated with 
an increase in renin secretion from a control 
value of 925±327 to 1710±486 ng/min (Fig. 
1). Eledoisin also increased renal blood flow 
but did not change renin secretion (Fig. 1). 

Both acetylcholine and bradykinin increased 
renal blood flow but neither drug affected 
renin secretion (Fig. 2). 

Renal vasodilation with acetylcholine, bra- 
dykinin, PGE2, or eledoisin increased both 
sodium and potassium excretion of the 
treated kidney (Table I). The percentage in- 
creases following each drug were not signifi- 
cantly different from each other. The sodium 
and potassium excretion of the contralateral 
kidney was not affected by drug infusion. 
The ^omerular filtration rate of the treated 
and untreated kidneys did not change during 
drug infusion. Unilateral renal vasodilation 
did not alter the mean systemic blood pres- 

Discussion, The secretion of renin may be 
altered by a vascular mechanism located in 
the afferent glomerular arteriole (7) since de- 
creases in renal resistance stimulate the 
release of renin (10). The present experiments 
demonstrate that PGE2 increased both ipsi- 
lateral renal blood flow and renin secretion 
(Fig. 1) while not affecting mean systemic 
blood pressure or contralateral venal func- 
tion. Renal vasodilation due to acetylcholine, 
bradykinin, or eledoisin, however, did not 
affect renin secretion (Figs. 1 and 2). Since 
the increase in renal blood flow was similar 
following infusion of all drugs, a vascular 

C , ELE C ^ ELE 



Fig. I. Eficcl of prostaglandin E2 (PGE2) and eledoisin (Ele) on renal blood flow (RBF) and renin secretion 
(RS). C « control. Values are expressed as means ± SEM. n»().*p< 0.05. 



inism does not appear to be primarily 
isible for the increase in renin secretion 
ing PGE2. 

arenal infusion of bradykinin has been 
I to increase renal PGE secretion (11). 
fore, bradykinin may affect renin secre- 
1 a manner similar to that of PGE2. In 
csent experiments, the dose of brady- 
which increased renal blood flow ap- 
nately 20 to 40% of control was less 
the dose of bradykinin previously re- 
I to increase PGE release (11). Thus, 
lasma or tissue PGE2 concentration 
ed in response to bradykinin in these 
ments may not have been sufficient to 
I response similar to that produced by 
fusion of PGE2. 

ubular mechanism located at the mac- 
^nsa region of the distal nephron also 

E 1. Effect of Acetylchoune, Bradykinin, 




Na excretion" 

K excretion" 


301.2 ± 135.2 
385.7 ±212.1 
121.4 ± 36.6 
478.0 ±231.3 

26.2 ± 12.3 
75.9 ± 30.6 
22.5 ± 12.3 
118.8 ±51.6 

ues expressed as percent increase. 

affects renin secretion by sensing changes in 
tubular sodium or chloride transport (12). 
Acetylcholine and PGE2 have been shown to 
decrease proximal tubular sodium reabsorp- 
tion (13). Similarly, bradykinin decreased 
proximal tubular sodium reabsorption by a 
mechanism related to vasodilation of the 
renal vasculature (14). The present data dem- 
onstrate that intrarenal infusion of acetylcho- 
line, bradykinin, PGE2, or eledoisin increased 
sodium and potassium excretion (Table I) to 
a similar degree in all experiments without 
afTecting GFR. Since changes in tubular so- 
dium reabsorption or changes in potassium 
excretion following infusion of acetylcholine, 
bradykinin, eledoisin, or PGE2 are similar, 
the changes in electrolyte excretion do not 
account for the PGE2-induced increase in 
renin secretion. PGE2 has been shown to 
increase renin release in vitro (IS). Although 
Weber et al did not report PGE2 to increase 
renin release, arachidonic acid, PGE2, and 
endoperoxide I and II all increased renin 
release in vitro (16). In the present experi- 
ments, both the hemodynamic and tubular 
responses produced by PGE2 appear to be 
similar to those elicited by bradykinin, ele- 
doisin, and acetylcholine in vivo. Thus, PGE2 
may increase renin secretion by a direct ac- 

C ,^, ACH 











C ^ BOK 



2. Effect of bradykinin (Bdk) and acetylcholine (Ach) on renal blood flow (RBF) and renin secretion (RS). 
ntrol. Values are expressed as means ± SEM. n^ {).*p< 0.05. 



tion on the vascular juxtaglomerular cells or 
a component of the juxtaglomerular appara- 

Summary. Intrarenal infusion of acetylcho- 
line, bradykinin, eledoisin, and PGE2 in- 
creased renal blood flow to a similar degree. 
Sodium and potassium excretion were simi- 
larly affected by each vasodilator. Renin se- 
cretion increased following PGE2 but was 
unaffected by acetylcholine, bradykinin, or 
eledoisin. It is suggested that PGE2 increases 
renin secretion by a direct effect on the jux- 
taglomerular apparatus. 

We express our appreciation to Mr. Keith Crosslan, 
Mrs. Peggy Wagner, and Mr. Terry Steele for their 
excellent technical assistance and to Miss Diane Hummel 
for the preparation of the manuscript. This research was 
supported by USPHS Grants AM 10913 and HD 06290. 

1. Wcming, C. Vetter. W., Weidmann, P., Schweikcrt, 
H. U., Stiel, D., and Sicgcnthalcr, W., Amcr. J. 
Physiol. 220,852(1971). 

2. Weber, P.. Holzgreve, H., Stephan, R., and Herbst, 
R., Eur. J. Pharmacol. 34, 299 (1975). 

3. Larsson, C, Weber, P., and Anggard, E., Eur. J. 
Pharmacol. 28,391 (1974). 

4. Romero, J. C, Dunlap, C. L., and Strong, C. G.. J. 
Clin. Invest. 58, 282 (1976). 

5. BaUie, M. D., Crosslan, K., and Hook, J. B., J 
Pharmacol. Exp. Ther. 199, 469 (1976). 

6. Yun, J.. Kelly, G., Bartter, F. C, and Smith, H., Jr., 
Circ. Res. 40, 459 (1977). 

7. Skinner, S. L., McCubbin, J. W., and Page, 1. H., 
Science 141, 814 (1%3). 

8. Walscr, M., Davidson, D. G.. and Orloff. J.. J. Clin. 
Invest. 34, 1520(1955). 

9. Haber, E., Koemer, T., Page, L. B., Kliman, B., and 
Pumode, A., J. Clin. Endocrinol. 29, 1349 (1969). 

10. Gotshall, R. W., Davis, J. O., Blaine, £. H., Musac- 
chia, X. J., Braverman, B., Freeman, R., and John- 
son, J. A., Amer. J. Physiol. 227, 251 (1974). 

11. McGifT, J. C, Terragno, N. A., Malik, K. U., and 
Lonigro, A. J., Circ. Res. 31, 36 (1972). 

12. Vander, A. J., and Carlson, J.. Ciic. Res. 25, 145 

13. Martinez-Maldonado, M., Tsaparas, N., Eknoyan, 
G., and Suki, W. N., Amer. J. Physiol. 222, 1147 

14. WiUis, L. R., Ludens, J. H., Hook, J. B., and WU- 
Uamson, H. E., Amer. J. Physiol. 217, 1 (1969). 

15. Dew, M. E., and Michelakis, A. M., Pharmacologist 
16, 198(1974). 

16. Weber, P. C, Larsson, C, Anggard, E., Hamberg. 
M., Corey, E. J., Nicolaou, K. C, and Samuelsson. 
B., Circ. Res. 39, 868 (1976). 

Received February 1, 1978. P.S.E.B.M. 1978, VoL 159. 


•D-Arabinofuranosyladenine Inhibition of Chemically Induced Rat Embryo Cell 

Transformation (40326) 


iiological Associates" Torrey Pines Research Center, 2945 Science Park Road, La Jolla, California 92037 

intileukemic chemotherapeutic drug, 
rabinofuranosylcytosine (ara-C) was 
sly shown to be an in vitro transform- 
it (1) for an established line of Fischer 
ryo cells, which had previously been 

be an accurate and sensitive indi- 
chemicals having carcinogenic prop- 
, 3). We were interested in using this 
f^stem to examine the transforming 
il of 9-)3-D-arabinofuranosyladenine 

an analog of ara-C which is also 
sed clinically as a cancer chemother- 
and antiviral agent in humans (4, 5). 
itumor and antiviral activities of both 
nd ara-C appear to be derived from 
iiibition of DNA synthesis (6-8). We 
lere that unlike ara-C, ara-A is not a 
ming agent for Fischer rat embryo 
1706). Further, nontoxic levels of ara- 
ct the cells from transformation in- 
y the known polycycUc hydrocarbon 
;en, 3-methylcholanthrene (MCA). 
nals and methods, (A) Toxicity testing. 
on in plating efficiency relative to a 

1 control was used to determine the 
of ara-A. Five hundred cells (F1706 
S ml of the complete growth medium 

minimum essential medium in 
►alts supplemented with 10% fetal bo- 
um, 2 mM L-glutamine, 0. 1 mM non- 
I amino acids, 100 units of penicillin, 
• /ig of streptomycin/ml) were added 
60-mm plastic cell cuhure dish (Lux), 
hes were incubated overnight at 37° 
nidified 5% C02-in-air incubator. The 
»ming the medium was decanted and 
I with a fresh medium containing se- 
tions of ara-A which had been diluted 

into the growth medium. Five days 
I dishes were fixed and stained (meth- 
iue-carbol fuchsin) and macroscopic 

were counted. 

ransformation assay. In two separate 
ents run concurrently by two differ- 
stigators, F1706 D95 cells were inoc- 

ulated into 75-cm^ plastic cell culture flasks 
(Lux) at a concentration of 10,000 cells/ml 
and 14 ml per flask. On Days 2 and 5, cultures 
were refed with either growth medium alone 
or growth medium containing either 0.01 or 
1.0 jug/ml ara-A. On Day 6, the cells from 
each group were transferred to fresh cultures 
in their respective media at a concentration 
of 1000 cells/ml and 10 ml per flask. The 
next day, 10 ml of growth medium was added 
to one-half the cultures from each group 
(without decanting the old media), and 10 ml 
of medium containing 0.4 /ig/ml of MCA to 
the othe half MCA was diluted in acetone to 
1000 /ig/ml and was further diluted in the 
growth medium. After an additional 2 days 
of incubation, the medium was decanted, and 
the cultures were washed with growth me- 
dium and refed with growth medium still 
supplemented with ara-A, but no longer con- 
taining the MCA. Three days later the cul- 
tures were again refed, but now with a growth 
medium void also of ara-A. The next day, 
new cultures were initiated at 500 cells/ml. 
This treatment schedule resulted in the fol- 
lowing duplicate sets of cultures: media only 
(negative control), 0.2 /ig/ml MCA (positive 
control), 0.01 /ig/ml ara-A, 1.0 /ig/ml ara-A, 
0.0 1 jLtg/ml ara-A plus 0.2 /ig/ml MCA, and 
1.0 jug/ml ara-A plus 0.2 /ig/ml MCA. At 
each subculture following the initial treat- 
ment, one set of flasks was set aside to be 
held without subdivision (holding series), and 
the other set subdivided 1:2 weekly to provide 
two new sets of cultures, one for the holding 
series and one for subdivision. Transforma- 
tion was determined by the appearance of 
foci of cells lacking contact inhibition and 
orientation and by the formation of macro- 
scopic colonies in semisolid agar (9). Tumor- 
igenicity was determined by subcutaneous 
inoculation of 5 x 10'^ cells into newborn 
Fischer rats (F344/f Mai). 

Results. We routinely test each compound 
for oncogenic potential at approximately the 


0037-9721 /l%/\S91-«l^^Vi\.^|^ 
AU hg^U Ttaerved. 



LD30 (concentration reducing the relative 
plating efficiency by approximately 30%) and 
at the highest concentration resulting in no 
reduction in relative plating efficiency 
(MNTD or maximum nontoxic dose). For 
ara-A these levels were 1.0 and 0.01 jug/ml, 
respectively (Table I). 

At neither level did ara-A, itself, induce 
cell transformation of F 1706 cells. However, 
as expected, cells treated with 0.2 /ig/ml MCA 
were phenotypically transformed by the third 
vertical subculture (D -h 3), and when tested 

TABLE I. Toxicrrv of ara-A" as Determined by 
Reductign in Plating Efficiency of F1706 D95*. 


Relative plating 
efficiency "^ (%) 













" 9-/3-D-Arabinofuranosyladenine. 

^ A serial line of Fischer rat embryo cells in its 95th 
population doubling. 

** The percentage of cells giving rise to macroscopic 
colonies, relative to the media only control in which the 
absolute plating efficiency was arbitrarily set at 100%. 
The absolute plating efficiency of the control was 20% 
(108 colonies out of 500 cells plated). 

at D + 6 produced macroscopic colonies in 
semisolid agar. When tested at D + 3, all 
cultures were negative for growth in agar. 
Cultures treated with MCA in the presence 
of either level of ara-A were still phenotypi- 
cally normal at the termination of the exper- 
iment 8 subcultures after treatment and fsuled 
to grow in semisolid agar when tested at D 
-h 3 and D -h 6. When inoculated into the 
newborn Fischer rats at D -h 8, the cultures 
treated with MCA alone were tumorigenic. 
The first tumor was found 52 days postinoc- 
ulation and by the 82nd day, 1 1 of the 14 rats 
were positive. In contrast, a total of 45 rats 
inoculated with cells from cultures treated 8 
subcultures earlier with either ara-A or MCA 
in the presence of ara-A were still tumor free 
when the experiment was terminated 94 days 
postinoculation (Table II). 

Discussion. Many drugs used in cancer 
chemotherapy are transforming agents (1, 10, 
11), mutagens (12), and oncogens (13, 14). 
One such agent, ara-C, had previously been 
found to induce transformation in mass cul- 
tures of secondary hamster embryo cells (IS). 
This observation was later confirmed using a 
quantitative hamster transformation system, 
as well as the F1706 cells used in the present 
study (1). Subsequently, it was demonstrated, 

TABLE II. MCA "-Induced Transformation of F1706* and Protection from Transformation by ara-A. 

Tumor results,' No. pos- 

itive/No. inoculated 


GrowUi in 

(days to 1st tumor-days 

Trcauncnt (per ml) 

agar (06)" 

to last tumor) 

Media control 




Media control 




0.2 /ig/ml MCA 

+ (+3) 



0.2 /ig/ml MCA 

+ (+3) 



1.0 /ig ara-A 




1.0 /ig ara-A 




0.01 /ig ara-A 




0.01 /ig ara-A 




1.0 /ig ara-A + 0.2 /ig MCA 




1.0 /ig ara-A + 0.2 /ig MCA 




0.01 /ig ara-A + 0.2 /ig MCA 




0.01 /xg ara-A + 0.2 /ig MCA 




" 3-Mcthylcholanthrcnc. 

* A serial line of Fischer rat embryo cells. 

'^ Newborn Fischer rats inoculated with 5x10* cells (0.05 ml) from D + 8. Rats without tumors were held 94 days 
and then sacrificed. 

'' Triplicate agar dishes were each inoculated with 50,000 cells from cultures at D + 6 (6 population doublings 
after removal of the MCA), held 4 weeks at 37'' in a humidified 5% CO2 incubator, and screened for the appearance 
of macroscopic colonies. 

' Not done. 

^Twelve rats inoculated, 10 killed by mother. 



using the C3H/IOTI/2 mouse embryo cells 
(16), that oncogenic transformation took 
place maximally in the S phase of the cell 
cycle (17). We know from double-bUnd stud- 
ies that 90% of the chemicals which transform 
these cells are also oncogenic for mice and 
rats (2). Since it is possible that tumor induc- 
tion in the rodent may be relevant to tumor 
induction in man, it seems wise to avoid 
where possible the use of chemotherapeutic 
agents which transform rodent cells. Ara-C is 
a transforming agent. Ara-A did not trans- 
form the F1706 rat cells, and at nontoxic 
doses protected the cells from transformation 
induced by the potent carcinogen, MCA. 

We have previously used this in vitro sys- 
tem (F1706) to show that several antiviral 
antibiotics, i.e., streptonigrin (18), cordycepin 
(19), and geldanamycin (20), could protect 
the cells from chemically induced transfor- 
mation. We suggested that this protection was 
due to the ability of the antibiotic to inhibit 
endogenous oncoma virus expression, since 
each drug also inhibited the '*tum-on'' of 
endogenous virus by halogenated pyrimi- 
dines. This explanation, however, is not ap- 
plicable to ara-A protection of MCA-induced 
cell transformation, since ara-A did not in- 
hibit transient virus induction by halogenated 
pyrimidines under similar conditions. 

These studies suggest that in vitro cell trans- 
formation assays may have value, not only as 
a prescreen for potentially oncogenic chemi- 
cals, but also for compounds having antican- 
cer properties. 

Summary, The cancer chemotherapeutic 
and antiviral agent 9-)3-D-arabinofuranosyl- 
adenine (ara-A) was examined for potential 
oncogenicity, using a serial Une of Fischer rat 
embryo cells, which was previously shown to 
be a sensitive and accurate indicator of chem- 
icals carcinogenic for rodents. We report here 
that at the concentrations tested, ara-A was 
not a transforming agent. Further, ara-A pro- 
tected the cells from transformation induced 
by the known carcinogen, 3-methylcholan- 

The authors thank Dr. Aaron E. Freeman for his 
technical assistance, and Ms. Joan Owens for assistance 
in preparation of this manuscript. This work was sup- 
ported by Contract NOl-CP-43240 within the Virus 
Cancer Program of the National Cancer Institute. 

1. Kouri, R. E., Kurtz, S. A., Price, P. J., and Benedict, 
W. R., Cancer Res. 35, 2413 (1975). 

2. Freeman, A. E., Weisburger, E. K., Weisberger, J. 
H., Wolford, R. G., Maryak, J. M., and Huebner, R. 
J., J. Nat. Cancer. Inst. 51, 799 (1973). 

3. Freeman, A. E., Igel, H. I., and Price, P. J., In Vitro 

4. Le Page, G. A., Khalig, A., and Gottlieb, J. A., Drug 
Metab. Disposition 1, 756 (1973). 

5. Schabel, F. M., Chemotherapy 13, 321 (1968). 

6. York, J. L., and LePage, G. A., Canad. J. Biochem. 

7. Furth, J. J., and Cohen, S. S., Cancer Res. 27, 1528 

8. Furth, J. J., and Cohen, S. S., Cancer Res. IS, 2061 

9. MacPherson, I., in "Soft Agar Techniques in Tissue 
Culture— Methods and Application** (P. F. Kruse, 
Jr., and M. K. Patterson, Jr., eds.), p. 276. Academic 
Press, New York (1973). 

10. Price, P. J., Suk, W. A., Skcen, P. C, Chirigos, M. 
A., and Huebner, R. J., Science 187, 1200 (1975). 

11. Benedict, W. F., Baneijee, A., Gardner, A., and 
Jones, P. A., Cancer Res. 37, 2202 (1977). 

12. Benedict, W. F., Baker, M. S., Haroun, L., Choi, E., 
and Ames, B. N., Cancer Res. 37, 2209 (1977). 

13. Harris, C. C, Cancer 37, 1014 (1976). 

14. Sieber, S. M., and Adamson, R. H., Advan. Cancer 
Res. 22, 57 (1975). 

15. Jones, P. A., Taderera, J. V., and Hawtrey, A. O., 
Eur. J. Cancer 8, 595 (1972). 

16. Reznikoff, C. A., Brandow, D. W., and Heidelberger, 
C, Cancer Res. 33, 3231 (1973). 

17. Jones, P. A., Baker, M. S., Bertram, J. S., and 
Benedict, W. F., Cancer Res. 37, 2214 (1977). 

18. Price, P. J.. Suk, W. A., Spahn, G. J., Chir^os, M. 
A., Lane, J. A., and Huebner, R. J., Proc. Soc. Exp. 
Biol. Med. 145, 1197(1974). 

19. Price, P. J.. Suck, W. A., Peters, R. L., Martin, C. E., 
Bellew, T. M., and Huebner, R. J., Proc. Soc. Exp. 
Biol. Med. 150,650(1975). 

20. Price, P. J., Suk, W. A., Skeen, P. C, Spahn, G. J., 
and Chirigos, M. A.. Proc. Soc. Exp. Biol. Med. 155, 
461 (1977). 

Received May 11, 1978. P.S.E.B.M. 1978, Vol. 159. 


Prolactin Receptors in Mouse Liver: Species Differences In Response to Estrogenic 

Stimulation' (40327) 


Department of Physiology, Neuroendocrine Research Laboratory, Michigan State University, East Lansing, Michipm 


Specific prolactin (PRL) receptors have 
been demonstrated in the liver of many spe- 
cies, including rats and mice (1-3). Ovariec- 
tomy (OVX) decreased and estrogen replace- 
ment increased PRL binding sites. The in- 
ductive effects of estrogen on PRL binding in 
the liver was dose related in OVX rats, and 
anti-estrogens reduced PRL receptors in the 
liver of female rats (4). One mechanism 
whereby estrogen induced PRL receptors is 
by stimulation of pituitary PRL release, re- 
sulting in induction of hepatic PRL binding 
sites in the liver (5, 6). However, since very 
low doses of estrogen increased PRL binding 
in the liver without altering serum PRL levels 
(4), and since the PRL-inhibiting ergot drug 
CB-1S4 did not decrease the estrogen-in- 
duced increase in hepatic PRL binding sites 
(4), it is possible that estrogen may act directly 
on the liver to increase PRL receptors. 

All of the above studies were performed in 
rats. To determine whether the estrogen effect 
on PRL receptors was observable in other 
species, we examined the effects of estrogen 
on PRL receptors in the liver of mice. The 
results indicate that estrogen inhibits induc- 
tion of PRL receptors in the Uver of female 
mice, in contrast to its stimulation of PRL 
receptors in the liver of male and female rats. 

Materials and methods. Adult male and 
female Swiss- Webster mice were obtained 
from Spartan Research Animals, Haslett, 
Michigan. Mice were housed in a tempera- 
ture-controlled (25 ± 1°) and artificially il- 
luminated room (lights on from 0500 to 1900 

' Published with the approval of the Michigan Agri- 
cultural Experiment Station as Journal Article No. 8523. 

^ Aided in part by NiH Research Grant AM04784 
from the National Institute for Arthritis. Metabolism 
and Digestive Diseases. 

^ We wish to thank Dr. Y. N. Sinha. Scripps Clinic 
and Research Foundation, La JoUa, California for the 
mouse prolactin RIA reagents used in this study. 

hr daily) and received food and water ad 

Experiment 1. Female mice were OVX on 
Day 1 and were injected sc daily with 2 /lig of 
estradiol benzoate (EB) in 50 ^ of com oil 
on Days 8 through 14. On the 15th day aH 
OVX were killed together with a group of 
intact females which were similarly injected 
daily on Days 8 through 14 with vehicle 

Experiment 2. Female mice, OVX 14 days 
prior to estrogen treatment, were given daily 
sc injections of either 1, 10, 20, or 50 /ig of 
EB in 50 /xl of com oil. Mice were then killed 
after 12 days of treatment, together with 
groups of intact and OVX controls which 
were injected with vehicle alone. Additional 
treatment groups given daily injections of 20 
/ig of EB were killed after 6 or 9 days of 

Experiment 3. Male mice were given a sin- 
gle 2-jLig EB sc injection in 50 /il of com oil 
and killed 7 days later. Controls were injected 
with vehicle alone. 

At the end of each experiment the mice 
were anesthetized with ether and decapitated, 
and the blood obtained from the cervical 
wound was allowed to clot at 4®. The scrum 
was separated by centrifugation and stored at 
—20® for later serum PRL measurements. 
Livers were removed and a microsomal mem- 
brane fraction was obtained by differential 
centrifugation as described previously (1). 
PRL was iodinated by a lactoperoxidase 
method (1) and the binding of [^^I]iodo-PRL 
to Uver membranes was determined. Incuba- 
tions with membrane protein and [^^I]iodo- 
PRL were performed at 4° for 60 hr, in the 
presence of excess ( 1 /ig) unlabeled PRL and 
in its absence. Livers from female mice were 
assayed for PRL binding, using 300 /xg of 
membrane protein per tube, whereas for male 
livers 1000 /ig per tube was used. Specific 
binding refers to the difference in radioactiv- 



r the Society for Expcriinenul Biology and Medicine 



bound to membranes after incubations 
\h and without unlabeled PRL, and for 
se of representation is expressed as a per- 
itage of the total counts added. PRL bind- 
l to liver membranes from mice has been 
3wn to be both time and temperature de- 
ndent, and specific for lactogenic hormones 
. Mouse PRL was measured by a double 
tibody radioimmunoassay using the mate- 
Is and methods of Sinha et al (7). The 
>logical potency of the mouse PRL stan- 
rd was 2S.0 lU/mg. The data in Expts 1 
d 2 were treated by an analysis of variance 
* unequal sample size, followed by a Stu- 
nt-Neuman-Kuels test for comparison of 
^ans among groups. Student's / test was 
^ to determine significance in Expt 3. P 
D.OS was consider^ to be significant. 
Results. Figure 1 shows that OVX signifi- 
lUy increased (P < 0.01) ['^IJiodo-PRL 
idmg to mouse liver membranes and that 
s enhanced binding could be decreased to 
act control values by estrogen replacement, 
lien this experiment was repeated (Fig. 2) 
Lh various doses of EB and longer treat- 
»t times, similar results were obtained. 
rK increased (P < 0.05) specific [^^IJiodo- 
:L binding from 14.48 ± 0.85% in the intact 
itrols to 19.93 ± 0.60%, Replacement by 
ecting 1 and 10 /ig of EB for 12 days 
luced PRL binding to 11.84 ± 0.53 and 
90 ± 0.81%, respectively, which were not 
nificantly different from intact control val- 



S 16- 

Fio. 1. Effects of OVX and OVX with EB replace- 
nt on specific [*^I]iodo-PRL binding to liver mem- 
ne preparations from female mice. For each tissue 
iple, incubations were performed in tripUcate at 4° 
60 hr, using 300 ftg of membrane protein per tube. 
5 amount of [***I]iodo-PRL per tube was I.O x 10* 
0. The line above each bar represents 1 SEM, and the 
nbers in white indicate the number of observations 
group. **P< 0.01 when compared to inuct controls. 


'* i = i i 

.° i i - 3 s i s s 

_tay II 








i.i Mil 

OVI ♦ l»a| il 


Fig. 2. Serum PRL levels and [^'^IJiodo-PRL bind- 
ing to Uver membranes from intact and OVX mice and 
OVX mice given daily injections of different doses of 
EB. For each tissue sample, incubations were performed 
in triplicate at 4° for 60 hr, using 300 fig of membrane 
protein per tube. The amount of [^^I]iodo-PRL per tube 
was 1.0 X lO^cpm. The line above each bar represents 1 
SEM, and the numbers in white indicate the number of 
observations for each group. *P < O.OS as compared to 
inuct controls. **P < 0.01 as compared to inuct controls. 

ues, whereas 20 and SO /ig of EB significantly 
reduced binding to below intact levels. Serum 
PRL was reduced from 12.0 ± 1.5 ng/ml 
(intact controls) to 8.09 ± 2.0 ng/ml in the 
OVX rats. All estrogen-treated groups had 
serum PRL values significantly higher than 
those in intact controls. 

Figure 3 demonstrates the effects of a single 
injection of 2 /xg of EB on specific PRL 
binding sites in liver membranes obtained 
from male mice. PRL binding increased (P 
< 0.01) from 22.61 ± 1.16 to 33.72 ± 1.29% 
at 7 days postinjection. Since PRL binding 
sites on male liver membranes were measured 
using 1000 /ig of membrane protein rather 
than 300 /ig of membrane protein (as used in 
quantitating PRL receptors in the liver of 
females), specific binding is higher in the 
livers of females than in the livers of males 
when compared on a milligram of protein 
basis. This is in agreement with the data of 
Posner (3). 

Discussion, The presence of specific PRL 







Fig. 3. Specific binding of [*^I]iodo-PRL to liver 
membranes from male mice 7 days after a single injection 
of EB. For each tissue sample, incubations were per- 
formed in triplicate at 4® for 60 hr, using 1000 fig of 
membrane protein per tube. The amount of [^^I]iodo- 
PRL per tube was 1.0 x 10^ cpm. The line above each 
bar represents 1 SEM, and the number in white indicates 
the number of observations for each group. **P < 0.01 
as compared to male controls. 

receptors in liver membranes of female mice 
agrees with the findings of other investigators 
(3, 8). However, our results indicate that OVX 
results in an increase of hepatic PRL recep- 
tors in female mice, whereas estrogen treat- 
ment over a large dose range reduced PRL 
binding to intact or below intact values. These 
data in female mice represent a striking con- 
trast to the effects of OVX and estrogen 
replacement on PRL receptors in liver of 
female and male rats. 

In female rats the effects of estrogen on 
increasing hepatic PRL receptors was con- 
vincingly demonstrated to be mediated 
throu^ stimulation of pituitary PRL release 
(S, 9). However, other data suggest a direct 
effect of estrogen on the liver to modulate 
PRL binding sites (4). In the present study, 
all doses of estrogen significantly increased 
serum PRL levels in female mice. The in- 
crease in PRL, however, is not believed to 
have altered hepatic PRL receptors since 
other investigators have reported that neither 
the high levels of endogenous PRL during 
pregnancy, nor exogenous PRL injections to 
female mice, influenced PRL binding sites in 
the liver (3, 8). Therefore, a direct effect on 
the liver appears likely, although an indirect 
effect of estrogen cannot be excluded. 

In male mice a single injection of 2 /ig of 
EB was able to signihcantly increase PRL 
binding sites in the liver. Since estradiol val- 
erate has been reported to stimulate PRL 

binding sites in the liver of male rats (S) 
apparent that both male rats and male 
respond similarly to the stimulatory acti* 
estrogen on hepatic PRL receptors. This 
contrast to the opposite effects of estrog< 
hepatic PRL binding sites of female rat 

Although the physiological signiflcan 
these results is not known at this time, 
has been shown to have numerous effec 
liver function of various species. Thus, 
was reported to regulate free fatty acid 
thesis in dog (10) and rat (1 1) livers, stim 
hepatic RNA synthesis in dwarf mice 
modulate ornithine decarboxylase activ 
the liver of rats (13), and increase soi 
medin release from rat livers (14). Ho^ 
in order for PRL to exert an effect on a t 
cell, it must first bind to a stereosp 
plasma membrane receptor to induce i 
cellular changes. Consequently, rec 
modulation could provide a mechanist 
altering the sensitivity of target orgai 
circulating PRL. Therefore, determ 
which hormones can alter PRL receptor 
the direction of these changes are impc 
for clarifying the physiological actioi 
PRL on liver function. 

The present data clearly demonstrat 
important species difference between fc 
rats and mice in estrogenic control of he 
PRL receptors and may have several i 
cations. Thus, the use of the rat as a n 
for investigating factors modulating PR 
ceptors in the liver cannot be considered 
for other species. Moreover, the functio 
PRL on liver function may be difTerec 
tween males and females of even the 
species, since control of PRL recepto 
liver of male and female mice are difG 
Our data indicate that estrogen inhibits 
binding sites in the female, whereas i 
male, binding is stimulated. Thus, th 
sponse of hepatic PRL receptors to esti 
is both species and sex dependent. The c 
anisms of action by which these effect 
mediated remain to be clarified. The d 
ential findings in these two species need 
considered when designing and interpi 
studies on the effects of PRL on liver 

Summary, Serum PRL and hepatic 
receptors were measured in intact and < 



and OVX mice given several closes of 
3VX significantly increased PRL bind- 
1 the liver of female mice, and EB re- 
1 receptors to intact or below intact lev- 
t was concluded that estrogen decreases 
receptors in the liver of female mice, 
is a striking contrast to the stimulatory 
: of estrogen on hepatic PRL receptors 
de and female rats. EB elevated serum 
in OVX mice, but since other investi- 
s reported that PRL does not alter he- 
PRL receptors in female mice, it appears 
^ that estrogen reduced PRL binding 
by a direct effect on the liver. However, 
direct effect cannot be excluded. In male 
estrogen increased PRL receptors in the 
as in male rats. 

e present data demonstrate important 
6S differences between female rats and 
le mice in estrogenic control of hepatic 
receptors. Moreover, the inhibitory ef- 
)f estrogen in female mice, and its stim- 
ry action in male mice, suggest that the 
•nse of hepatic PRL receptors to estrogen 
be sex dependent in different species, 
mechanisms of action by which these 
'S are mediated remain to be clarified. 

Campbell, G. A., and Meites, J., Endocrinology %, 

2. Posner, B. I., Kelly, P. A., Shiu, R., and Friesen, H., 
Endocrinology 95, 521 (1974). 

3. Posner. B. I., Endocrinology 98, 645 (1976). 

4. KeUy, P. A.. Ferland, L., Labrie, P., and Delean, H., 
in "Hypothalamus and Endocrine Function" (F. La- 
brie, J. Meites, and G. Pelletier, eds.), p. 321. Plenum 
Press, New York (1976). 

5. Posner, B. I., Kelly, P. A., and Friesen, H., Science 

6..Costlow, M. E., Buschow, R. H., and McGuire, W. 
D.,LifeSci. 17, 1457(1975). 

7. Sinha, Y. N., Selby, F. W., Uwis, U. J., and Van- 
derlaan, W. P., Endocrinology 92, 1045 (1972). 

8. Knazek, R. A., Liu, S. C, and Gullino, P. M., 
Endocrinology 101, 50 (1977). 

9. Frantz, W. L., Mann, L. C, and Welsch, C. W., 
IRCSJ. Med.Sci.5,32(1977). 

10. Winkler, B., Rathgeb, L, Stelle, R., and Altszlur, N., 
Endocrinology 88, 1349 (1971). 

11. Macleod, R. M., Bass, M. D., Huang, S. C, and 
Smith, M. C, Endocrinology 82, 253 (1968). 
Chen, H. W., Hamer, D. H., Heiniger, H., and Meier, 
H., Biochim. Biophys. Acu 287, 90 (1972). 
Richards, F. F., Biochem. Biophys. Res. Commun. 
d3, 292 (1975). 

Francis, M. J. O., and Hill, D. J., Nature (London) 
255, 167 (1975). 




sUto, M., Marshall S., Boudreau, M., Bruni, J. F., Received April 13, 1978. P.S.E.B.M. 1978, Vol. 159. 


Hemopoiesis in Diffusion Chambers in Strontium-89 Marrow-Ablated Mice^ (40328) 


Rush-Presbyterian-St, Luke*s Medical Center, Chicago, Illinois 60612 

Hemopoiesis can be evaluated by studying 
the proliferation of hemopoietic cells in dif- 
fusion chambers (DCs) implanted into the 
peritoneal cavities of animals. During the 
initial several hours after seeding and implan- 
tation, the number of cells recoverable from 
the inoculum declines by 40 to 60% (1). 

The predominant hemopoietic precursor 
cell responsible for the enlarging hemopoietic 
cell population in DCs appears to be a gran- 
ulocyte/macrophage committed precursor (2, 
3). Multipotent hemopoietic stem cells (CFU- 
S), however, also have been shown to play a 
role in DC hemopoiesis (2, 4-7) and the 
number of these cells reaches its peak in DC 
cultures before the total number of hemo- 
poietic elements reaches its maximum (8, 9). 

If the hemopoietic cell inoculum consists 
of steady state cells, such as cells from mar- 
rows of normal mice, proliferation in DCs 
begins after a lag of about 18 hr (1). A 
number of investigators have found that in 
DCs the number of cells harvested, at least 
from Days 4 to 7 after implantation, is related 
linearly to the number of cells in the inocu- 
lum, suggesting that there is little or no sig- 
nificant cell-cell interaction (10-12). Niska- 
nen and his colleagues (13, 14), on the other 
hand, found that as the numbers of cells 
seeded in DCs were increased, the growth of 
both differentiated granulocytes as well as 
CFU-S was inhibited. In addition, prolifera- 
tion of cells in DCs is modified substantially 
by pretreatment of the host animals with 
agents such as irradiation (8, 10, IS, 16) or 
cytotoxic drugs (2, 4, 10, 13, 15, 17), both of 
which perturb the hemopoietic state of host 

' This work was supported by grants CA-041 14 and 
CA-22736 from the National Institutes of Health, NCI, 
Bethesda, Md.. and by a grant from the Leukemia Re- 
search Foundation, Chicago, 111. 

^ Recipient of Research Fellowship No 6F22-CA9452 
from the NCI, Bethesda, Md., during part of the time 
this work was performed. 

Elevated levels of colony-stimulating activ- 
ity (CSA), i.e., glycoproteins required for the 
growth of granulocyte/macrophage precur- 
sors in vitro, have been found in the serum of 
animals treated with whole-body irradiation 
(13, 16, 18-20). In whole-body-irradiated an- 
imals, hemopoiesis in DCs has been found to 
parallel the increase in serum CSA levels (IS, 
16); this relationship is expected, as the DC 
technique primarily assesses granulopoiesis. 

The bone-seeking radionuclide, ®®Sr, can 
be used to ablate marrow hemopoiesis selec- 
tively (21-24). By 10 days after *Sr injection 
(4 /iCi/g body wt) the marrows of mice arc 
aplastic and contain less than 2% of the nor- 
mal number of CFU-S (24). The spleens of 
^Sr-treated mice support marked compensa- 
tory-hemopoiesis and these mice develop 
only a mild anemia but a more severe leu- 
kopenia; with the passage of time, hemo- 
poiesis is gradually restored in the marrows 
of these mice (21, 24, 25). In a previous study, 
we did not detect an elevation in serum CSA 
levels in ®®Sr-treated mice (26). 

In an attempt to evaluate the presence of 
a humoral stimulus for hemopoiesis in ^Sr- 
treated mice, at various times after ^Sr treat- 
ment, we implanted into such mice DCs con- 
taining 1 X 10^ marrow cells from normal 
mice. We evaluated the total number of cells, 
proportions of the various cellular elements, 
and the number of CFU-S in the DCs 72 hr 
after implantation. 

Materials and methods. Pathogen-free fe- 
male CAFi (Balb/c x A/He) mice (Cumber- 
land View Farms, Clinton, Tenn.), 14 to 16 
weeks old, were housed in cages with dispos- 
able plastic bottoms; a maximum of 10 mice 
were housed per cage. The mice were per- 
mitted food and acidified (pH 3.2) water ad 
libitum. On Day the mice were given ip 
injections of ^^SrCk (Oak Ridge National 
Laboratories, Oak Ridge, Tenn.), 4 /xCi/g 
body wt, in 0.25 ml of a solution buffered to 
pH 5 to 6; control mice were injected with a 
comparable amount of cold ^rCl2. On days 


Copyright €> 1978 by the Society for ExperimenuJ Biology and Medicine 
^^ r^ts reaerved. 



}, and 39 after the Sr injections, DCs 
planted into the peritoneal cavities of 
:ized (sodium pentobarbital) mice 
IT later the DCs were removed. Prior 
fice, blood was obtained from each 
>use by bleeding it from the lateral 

plexus into heparinized capillary 
i microhematocrit determination, to- 
sated cell count (by hemacytometer), 
[)-cell differential count were per- 
on the blood from each mouse. For 
ic studied, five radio-®®Sr-treated and 
1-^Sr-treated mice were studied. Each 
vas implanted with two DCs, one into 
t side and the other into the left side 
>eritoneal cavity; the chambers were 

for identification prior to implanta- 

ivere constructed by gluing deionized 
)re membranes (Nuclepore Corp., 
ton, Calif.) which had 0.22-fmi pores, 
wo sides of plastic rings (Millipore 
Vfedford, Mass.) with Millipore MF 

the DCs were tested for leaks by 
1 with air under water and then ster- 
i 70° dry heat for 16 hr. They were 
vith 1 X 10® marrow cells pooled from 
iirs of three CAFi mice; the cells were 
ed in 0.1 ml of Hanks' balanced salt 

(HBSS). The holes used to fill the 
rs were occluded with plugs of dental 
ior to implantation, the DCs were 
^ in a solution of penicillin and strep- 
I. After 72 hr in the mice, the cham- 
re removed and placed into a solution 
ing 0.5% grade B Pronase (Calbi- 
San Diego, Calif.) and 5% Ficoll (Lit- 
^netics Lab Products, Kensington, 

which they remained for 90 min at 
mperature; they were agitated contin- 
The wax plugs were removed and the 
s of the DCs were removed by aspi- 
hrough the filling hole by means of a 
;e needle attached to a tuberculin sy- 
he chambers were washed thrice with 
if HBSS; the last wash was performed 
e removal of one of the Nuclepore 
mes. At each time studied, the con- 
* the five chambers which were im- 
into the right sides of the mice were 
ed to form one suspension of pooled 
id those from the five chambers from 
sides another. These two suspensions 

of pooled cells were counted and assayed 
separately. Cytospin centrifiige (Shandon 
Southern Instruments, Inc., Sewickley, Penn.) 
slides were prepared from each suspension of 
pooled cells and a 4(X)-cell differential count 
was performed on each of the suspensions. 
The criteria of Benestad (27) were used to 
classify proliferative and nonproliferative 
granulocytic elements. Duplicate nucleated 
cell counts were performed on each cellular 
suspension by means of a hemacytometer. 

The CFU-S content of each cellular sus- 
pension was assayed by the surface spleen 
colony technique of Till and McCullough 
(28). The pooled cellular suspensions were 
diluted so that the equivalent of V^th or Vioth 
of the contents of a single chamber was con- 
tained in 0.5 ml of HBSS which was then 
injected into a lateral tail vein of an assay 
mouse which had been exposed to 900 rad of 
whole-body irradiation provided by a ^^'Cs 
source (Gamma Cell 40, Atomic Energy of 
Canada, Ltd., Ottawa, Canada) within the 
previous 3 hr. We used 12 to IS mice to assay 
each suspension of cells. 

The results of the studies performed on the 
chambers implanted into the right and left 
sides of the mice were evaluated separately; 
as the results from the two groups were vir- 
tually identical we will report only the pooled 
data. We had control studies at each time 
interval and report the results of the cell 
counts and CFU-S assays individually. The 
differential counts performed on the contents 
of the chambers implanted into the control 
mice were very similar at the four times eval- 
uated; this is to be expected, as the control 
mice, injected with SrCl2 were "normal" 
animals at all times. To simplify the reporting 
of differential counts of the DC cells, we have 
reported the differential counts from the cells 
implanted into the normal mice as means ± 
SB obtained from all the time intervals stud- 
ied; the results from the experimental mice 
are reported separately for each time. 

Student's / test was used to evaluate the 
statistical differences between the results ob- 
tained from the ^Sr and ®®Sr groups. As there 
are only two values (obtained from the right 
and left chamber suspensions) for the total 
numbers of cells per chamber and for the 
differential counts of the cells for each group 
at each time studied, we did not analyze these 



Statistically (Fig. IC; Table I). 

Results, The ®^Sr-trcatcd host mice were 
significantly anemic only during the period 
in which the second group of chambers were 
implanted, i.e., 10 to 13 days after ®®Sr injec- 
tion (Fig. lA), but these mice were granulo- 
cytopenic at all times studied (Fig. IB). 

The total number of nucleated cells har- 
vested from the DCs implanted into the ^Sr- 
treated mice was greater than that harvested 
from the DCs housed in the control mice at 
all times studied (Fig. IC); the largest differ- 
ences occurred at the 10- to 13-day and 18- 
to 21 -day time periods when the ratios be- 
tween the cell contents of the DCs from the 
^Sr and those from the ®®Sr control mice were 
1.8 and 2.4, respectively (Fig. IC). In addi- 
tion, at the first three times studied, the num- 
bers of cells harvested from the chamber 
housed in the ^^Sr-treated mice exceeded the 
numbers (1 x 10® cells) in the original inoc- 

In general, the proportion of the various 
cellular elements in the DCs of the ^Sr and 

^Sr mice were quite similar (Table I). There 
was, however, a slight increase in the propor- 
tion of blasts in the DCs from the ^Sr mice 
during the first three times studied (Table 1). 
In addition, in the 10- to 13-day DCs from 
the ^Sr-treated mice there was a modest in- 
crease in the proportion of nucleated red 
blood cells (Table I); this was the only time 
during which the ®®Sr-treated mice were sig- 
nificantly anemic (Fig. lA). 

The inoculum contained about 340 CPU- 
S. The numbers of CFU-S harvested from 
the DCs housed in the ®®Sr-trcatcd mice were 
significantly greater than those from the DCs 
housed in the ®®Sr control mice. The greatest 
difference between the numbers of CFU-S in 
the two groups occurred in those chambers 
implanted during the second (10-13 days) 
and third (18-21 days) intervals studied (Fig. 
ID); these were the same times during which 
the largest differences were found in total 
numbers of nucleated cells per chamber. The 
second-interval-chambers, implanted 10 days 
after ^Sr injection, contained more than 


Fig. 1. (A) Packed red cell volume (as a percentage) and (B) granulocyte counts per cubic millimeter of bloo<l 
from diffusion chamber (DC) host mice; and (C) nucleated cell counts and (D) numbers of CFU-S of DO- All 
chambers were in mice for 72 hr. Days indicated are numbers of days after Sr injections which also were days on 
which blood counts were performed and DCs harvested. Means ± SE. N.S.» not significant; *p < 0.01; **p < 0.001 



vice as many CFU-S as did the inoculum. 

Discussion, The larger number of CFU-S 
i DCs cultivated in ®®Sr-treated mice as com- 
Eued to that in DCs from control mice sug- 
:sts that in ^Sr-treated mice there is a hu- 
loral mechanism(s) which effects either 
lore rapid proliferation of CFU-S or a 
lortening of the preproliferative lag period 
r both. The early (by 72 hr) substantial 
Lcrease in the number of differentiated blood 
lements in DCs from ^Sr-treated mice sug- 
tsts that there also is a stimulus for the 
roliferation of conmiitted precursor cells. 

In Table II we have sunmiarized studies 

from the literature on the growth of cells in 
DCs implanted into hemopoietically stressed 
mice in which both cell numbers and CFU-S 
were studied early after DC implantation. 
The increase in DC contents above that in 
the inoculum in our ^Sr-treated mice oc- 
curred as early (Day 3) as a similar increase 
in 800-rad whole-body-irradiated (WBI) mice 
(8) (Table II); on Day 3, the magnitude of the 
increase in DCs from ®®Sr-treated mice may 
even have been slightly greater than that in 
DCs from 800-rad WBI mice (8). Moreover, 
if cell density does influence the growth rate 
of cells in DCs, the increase noted in our 

TABLE I. Differential Counts, as Percentages, of Cells from Mice Used to Inoculate Diffusion 
Chambers (DCs) and of Cells Harvested from DCs Cultivated in **Sr- or **Sr-TREATED Mice. 


tive granu- 


and macro- 

Red cell 





2.1 ±0.3 
1.0 ±0.7 


11.1 ± 1.3 
19.6 ± 3.3 


30.3 ± 3.0 
42.0 ±2.1 


0.6 ± 0.2 
26.1 ± 1.6 


30.0 ± 1.4 
1.0 ±0.9 


20.6 ± 0.6 
7.2 ± 1.7 


5.3 ± 1.2 
3.1 ± 1.8 


' This category includes: basophils, eosinophils, plasma cells, megakaryocytes, and cells in mitoses. 

^ For each interval, cells pooled from the femurs of three normal CAFi mice were used to inoculate the DCs. 

'^ Data for DCs cultivated in ^r-treated mice are values pooled from all four times studied; means ± SE. 

'' Dau for DCs cultivated in ^Sr mice are averages of dau obtained from two groups (right and lefl) of chambers 
t each time. 

' Day after injection of ^Sr, this was the day on which chambers were harvested. All chambers were in mice for 

^ There were more red cell precursors in DCs implanted into ^Sr-treated mice on Day 10 and harvested on Day 
3; Day 13 was the only time at which ^Sr mice were substantially anemic (Fig. 1 A). 

TABLE IL Review from the Literature of Studies in Which Both Numbers of Cells and CFU-S Were 

Assayed in DCs Implanted into Hemopoietically Stressed Mice: (a) The First Day After DC 

Implantation on Which the Cell Population (Total and CFU-S) Exceeded that of the Inoculum and 

(B) The Magnitude of This Very Early Increase. 

Author (method used to stress mouse hemopoiesis) and size of 



Day; magnitude of 

increase over input* 

Day; magnitude of 
increase over input* 

Niskanen et a/. (13) (Cyclophosphamide, 350 mg/kg) 

1 X 10^ nucleated cells 

5 X 10^ nucleated cells 
Shuhnan and Robinson (9) (500 R WBH 

1 X 10^ nucleated cells 
Boyum et al. (8) (800 R WBI) 

7 X 10* granulocytes 
Adler and Trobaugh (present study) (4 /iCi/g of ^Sr) 

1 X 10* nucleated cells 



5; 1.6X 

Not done 

4; 3.4X 

6; 1.4X 

3; 1.2X 

3; 1.2X 

3; up to 


3; up to 2x 

* For the study of Boyum ei aL (8) data are granulocytes/chamber rather than nucleated cells/chamber. 

* In some cases the magnitude