ESSAYS ON NUCLEIC ACIDS

ESSAYS
ON NUCLEIC ACIDS

BY

ERWIN CHARGAFF

tA

ELSEVIER PUBLISHING COMPANY

AMSTERDAM / LONDON / NEW YORK
1963

SOLE DISTRIBUTORS FOR THE UNITED STATES AND CANADA
AMERICAN ELSEVIER PUBLISHING COMPANY, INC.
52 VANDERBILT AVENUE, NEW YORK 17, N.Y.

sole distributors for great britain

elsevier publishing company limited

12b, rippleside commercial estate

ripple road, barking, essex

library of CONGRESS CATALOG CARD NUMBER 63-8569

WITH 13 ILLUSTRATIONS AND 39 TABLES

ALL RIGHTS RESERVED
THIS BOOK OR ANY PART THEREOF MAY NOT BE

REPRODUCED IN ANY FORM,

INCLUDING PHOTOSTATIC OR MICROFILM FORM,

WITHOUT WRITTEN PERM ISSION FROM THE PUBLISHERS

TO THE MEMORY OF MY TEACHER, KARL KRAUS

Preface

Few recent advances have, for better or for worse, had such an
impact on biological thinking as the discovery of base-pairing in
nucleic acids. These complementariness principles do not only
underlie current ideas on the structure of the nucleic acids, but
they form the foundation of all speculations, more or less well-
founded, on their physical properties (denaturation, hypochromic-
ity, etc.), on the transfer of biological information from deoxy-
ribonucleic acid to ribonucleic acid, and on the role of the latter
in directing the synthesis of specific proteins. They form the basis
of present explanations of the manner in which the amino acids
are activated before being assembled to make a protein; they are
being invoked incessantly in attempts to unravel the nucleotide
code which is thought to be responsible for specifying the amino
acid sequence of proteins.

It will, perhaps, surprise many readers, into whose ears a
different version has been drummed for years, to learn that the
first announcement of base-pairing in nucleic acids was made in
an article, published early in 1950, which forms Chapter 1 of
this book; the statement itself will be found on page 13. Those
to whom it may sound unusually modest or restrained are asked
to consider that at that time the new science of molecular
biology did not yet exist.

Chapters 2 to 9 are also drawn from essays published pre-
viously at different times and at different places, some not easily
accessible. They are all reprinted here without change, with the
exception of Chapter 9 which underwent extensive revision.
Place and year of publication are indicated in each case; and I

VIII PREFACE

should like to use this occasion to thank the several publishers
for the permission to reprint these articles.

The last two chapters, Nos. 10 and 11, have not been pubHshed
before and are of an entirely different kind, stepping forth, as
they do, without the stately periwig of a list of references. They
should be considered as a divertimento, though not without pas-
sages in a minor key; specimens of a sort that rarely finds its
way into a book dealing with scientific matters. Chapter 10 is a
specimen of a recent lecture, as it was prepared for actual
delivery, without undergoing the normal editing for publication
which consists mostly in the substitution of references to the
literature for any critical or unusual remarks that may have been
made. Chapter 11 is a specimen of many conversations that I
have participated in over the last few years; it is, of course, a
composite of many such talks, a collage, as it were: no single
person could be so dim.

There will be some, I am certain, that will find the application
to scientific problems of the means of humor, of satire, and even
of puns, these metaphysical hiccups of language, most unbe-
coming and frivolous. But there are many levels at which
criticism ought to be exercised; and the critique of some of the
concepts of modern science, and especially of its aberrations, has
virtually disappeared at a time when it is more necessary than
ever; at a time when the polarization of science has gone so far
that one now "runs" for scientific awards as for a political office;
that scientific lectures begin to sound like keynote speeches at
political conventions; that scientific reporting has replaced the
intimate gossip from Hollywood; that the persuasiveness of truth
has been replaced by the strength of the acclamation; in other
words, that cliques are surrounded by claques. The emergence
of a Scientific Establishment, of a power elite, has given rise to
a remarkable phenomenon: the appearance of what is called
dogmas in biological thinking. Reason and judgment are inclined
to abdicate when faced with a dogma; but they should not. Just
as in political life, a stiff upper lip often conceals a soft under-
belly. It is imperative that the most stringent criticism be ap-

PREFACE IX

plied to tentative scientific hypotheses that disguise themselves
as dogmas. This criticism must come from within; but it can
only come from an outsider at the inside.

If the title of the last chapter requires an explanation, I may
quote Webster's New Collegiate Dictionary: ''amphisbaena — a
fabled serpent with a head at each end, moving either way".
Whether strand separation was observed in the Middle Ages, is
not recorded.

Columbia University Erwin Chargaff

New York, N.Y.
December, 1962

"^7\

Contents

V

Preface vii

Chapter 1. Chemical Specificity of Nucleic Acids and Mechanism

of Their Enzymic Degradation 1

1. Introduction 1

2. Identity and diversity in high-molecular cell constituents . 2

3. Purpose 5

4. Preparation of the analytical material 6

5. Separation and estimation of purines and pyrimidines . • 7

6. Methods of hydrolysis 10

7. Composition of deoxypentose nucleic acids 11

8. Composition of pentose nucleic acids 15

9. Sugar components 18

10. Depolymerizing enzymes 18

11. Concluding remarks 20

References 23

Chapter 2. Structure and Function of Nucleic Acids as Cell

Constituents 25

1. Biological significance of nucleic acids 26

2. Deoxypentose nucleic acids 27

3. Pentose nucleic acids 35

4. Final remarks 35

References 37

Chapter 3. Deoxypentose Nucleoproteins and Their Prosthetic

Groups 39

1. Introduction 39

2. Deoxypentose nucleoproteins 41

3. Deoxypentose nucleic acids 50

References 60

Chapter 4. The Chemistry and Function of Nucleoproteins and

Nucleic Acids 62

1. Introduction 62

2. Nucleoproteins 63

3. Deoxypentose nucleic acids 65

4. Fractionation of deoxypentose nucleic acids 66

S2672

XII CONTENTS

5. Differences between pentose nucleic acids 69

6. Remarks on functions 70

References 75

Chapter 5. The Very Big and the Very Small: Remarks on

Conjugated Proteins 77

Chapter 6. Of Nucleic Acids and Nucleoproteins 82

1. Introduction 82

2. The significance of isolated cell constituents 83

3. Two types of nucleic acid 84

4. Deoxypentose nucleic acids 86

5. Pentose nucleic acids and nucleoproteins 92

6. The meaning of regularities 93

7. Remarks on nucleoproteins 96

References 98

Chapter 7. Nucleic Acids as Carriers of Biological Information . 100

1. Is there a hierarchy of cellular constituents? 100

2. On biological information chemically conveyed .... 102

3. Invariability of nucleic acids 104

4. Diversity of nucleic acids 105

5. Regularity of nucleic acids 106

6. Concluding remarks 107

References 108

Chapter 8. First Steps towards a Chemistry of Heredity . . . 109

1. Introduction 109

2. Chemical basis of cellular specificity HI

3. On the alphabet of the cell and some language difficulties 113

4. Biochemical space 115

5. On the code-script of biological high polymers . . . . 117
References 125

Chapter 9. The Problem of Nucleotide Sequence in Deoxy-
ribonucleic Acids 126

1. Introduction 126

2. Remarks on the conceptual basis of sequence analysis . . 127

3. Remarks on nomenclature 131

4. Early attempts at sequence investigation 132

5. Sequence studies through differential distribution analysis 134

6. Summarizing remarks 147

References 159

Chapter 10. A Few Remarks on Nucleic Acids, Decoding, and the

Rest of the World 161

Chapter 11. Amphisbaena 174

Index 201

CHAPTER 1

Chemical Specificity of Nucleic Acids and Mechanism
of Their Enzymic Degradation"^

1. INTRODUCTION

The last few years have witnessed an enormous revival in in-
terest for the chemical and biological properties of nucleic acids,
which are components essential for the life of all cells. This is
not particularly surprising, as the chemistry of nucleic acids
represents one of the remaining major unsolved problems in
biochemistry. It is not easy to say what provided the impulse for
this rather sudden rebirth. Was it the fundamental work of Ham-
marsten^ on the highly polymerized deoxyribonucleic acid of
calf thymus? Or did it come from the biological side, for instance,
the experiments of Brachet^ and Caspersson^? Or was it the
very important research of Avery^ and his collaborators on the
transformation of pneumococcal types that started the avalanche?
It is, of course, completely senseless to formulate a hierarchy
of cellular constituents and to single out certain compounds as
more important than others. The economy of the living cell
probably knows no conspicuous waste; proteins and nucleic acids,
lipids and polysaccharides, all have the same importance. But
one observation may be offered. It is impossible to write the
history of the cell without considering its geography; and we
cannot do this without attention to what may be called the
chronology of the cell, i.e., the sequence in which the cellular con-

* This article is based on a series of lectures given before the Chemical
Societies of Zurich and Basle (June 29th and 30th, 1949), the Societe de
chimie biologique at Paris, and the Universities of Uppsala, Stockholm,
and Milan. (Reprinted with permission from Experientia, 6 (1950) 201-209).

References p. 23

2 CHEMICAL SPECIFICITY OF NUCLEIC ACIDS

stituents are laid down and in which they develop from each
other. If this is done, nucleic acids will be found pretty much at
the beginning. An attempt to say more leads directly into empty
speculations in which almost no field abounds more than the
chemistry of the cell. Since an ounce of proof still weighs more
than a pound of prediction, the important genetical functions,
ascribed — probably quite rightly — to the nucleic acids by many
workers, will not be discussed here. Terms such as "template"
or "matrix" or "reduplication" will not be found in this lecture.

2. IDENTITY AND DIVERSITY IN HIGH-MOLECULAR CELL
CONSTITUENTS

The determination of the constitution of a complicated com-
pound, composed of many molecules of a number of organic sub-
stances, evidently requires the exact knowledge of the nature and
proportion of all constituents. This is true for nucleic acids as
much as for proteins or polysaccharides. It is, furthermore, clear
that the value of such constitutional determinations will depend
upon the development of suitable methods of hydrolysis. Other-
wise, substances representing an association of many chemical
individuals can be described in a qualitative fashion only; precise
decisions as to structure remain impossible. When our laboratory,
more than four years ago, embarked upon the study of nucleic
acids, we became aware of this difficulty immediately.

The state of the nucleic acid problem at that time found its
classical expression in Levene's monograph^. (A number of
shorter reviews, indicative of the development of our conceptions
concerning the chemistry of nucleic acids, should also be men-
tioned®-^ ^) The old tetranucleotide hypothesis — it should never
have been called a theory — was still dominant; and this was char-
acteristic of the enormous sway that the organic chemistry of
small molecules held over biochemistry. I should like to illustrate
what I mean by one example. If in the investigation of a disac-
charide consisting of two different hexoses we isolate 0.8 mole of
one sugar and 0.7 mole of the other, this will be sufficient for

HIGH-MOLECULAR CELL CONSTITUENTS 3

the recognition of the composition of the substance, provided its
molecular weight is known. The deviation of the analytical results
from simple, integral proportions is without importance in that
case. But this will not hold for high-molecular compounds in
which variations in the proportions of their several components
often will provide the sole indication of the occurrence of dif-
ferent compounds.

In attempting to formulate the problem with some exag-
geration one could say: The validity of the identification of a
substance by the methods of classical organic chemistry ends
with the mixed melting point. When we deal with the extremely
complex compounds of cellular origin, such as nucleic acids,
proteins, or polysaccharides, a chemical comparison aiming at
the determination of identity or difference must be based on the
nature and the proportions of their constituents, on the sequence
in which these constituents are arranged in the molecule, and on
the type and the position of the hnkages that hold them together.
The smaller the number of components of such a high-molecular
compound is, the greater is the difficulty of a decision. The oc-
currence of a very large number of different proteins was recog-
nized early; no one to my knowledge ever attempted to postulate
a protein as a compound composed of equimolar proportions of
18 or 20 different amino acids. In addition, immunological in-
vestigations contributed very much to the recognition of the
multiplicity of proteins. A decision between identity and dif-
ference becomes much more difficult when, as is the case with
the nucleic acids, only few primary components are encountered.
And when we finally come to high polymers, consisting of one
component only, e.g., glycogen or starch, the characterization of
the chemical specificity of such a, compound becomes a very
complicated and laborious task.

While, therefore, the formulation of the tetranucleotide con-
ception appeared explainable on historical grounds, it lacked an
adequate experimental basis, especially as regards "thymonucleic
acid". Although only two nucleic acids, the deoxyribose nucleic
acid of calf thymus and the ribose nucleic acid of yeast, had been

References p. 23

4 CHEMICAL SPECIFICITY OF NUCLEIC ACIDS

examined analytically in some detail, all conclusions derived from
the study of these substances were immediately extended to the
entire realm of nature; a jump of a boldness that should astound
a circus acrobat. This went so far that in some publications the
starting material for the so-called "thymonucleic acid" was not
even mentioned or that it was not thymus at all, as may some-
times be gathered from the context, but, for instance, fish sperm
or spleen. The animal species that had furnished the starting
material often remained unspecified.

Now the question arises: How different must complicated sub-
stances be, before we can recognize their difference? In the mul-
tiformity of its appearances nature can be primitive and it can
be subtle. It is primitive in creating in a cell, such as the tubercle
bacillus, a host of novel compounds, new fatty acids, alcohols,
etc., that are nowhere else encountered. There, the recognition
of chemical pecuharities is relatively easy. But in the case of the
proteins and nucleic acids, I believe, nature has acted most subtly;
and the task facing us is much more difficult. There is nothing
more dangerous in the natural sciences than to look for harmony,
order, regularity, before the proper level is reached. The harmony
of cellular life may well appear chaotic to us. The disgust for the
amorphous, the ostensibly anomalous — an interesting problem in
the psychology of science — has produced many theories that
shrank gradually to hypotheses and then vanished.

We must realize that minute changes in the nucleic acid, e.g.,
the disappearance of one guanine molecule out of a hundred,
could produce far-reaching changes in the geometry of the con-
jugated nucleoprotein; and it is not impossible that rearrange-
ments of this type are among the causes of the occurrence of
mutations.*

The molecular weight of the pentose nucleic acids, especially
of those from animal tissue cells, is not yet known; and the
problem of their preparation and homogeneity still is in a very
sad state. But that the deoxypentose nucleic acids, prepared

For additional remarks on this problem, compare Ref. 12.

PURPOSE 5

under as mild conditions as possible and with the avoidance of
enzymic degradation, represent fibrous structures of high molec-
ular weight has often been demonstrated. No agreement has as
yet been achieved on the order of magnitude of the molecular
weight, since the interpretation of physical measurements of
largely asymmetrical molecules still presents very great difficul-
ties. But regardless of whether the deoxyribonucleic acid of calf
thymus is considered as consisting of elementary units of about
35,000 which tend to associate to larger structures'^- '^ or whether
it is regarded as a true macromolecule'^ of a molecular weight
around 820,000, the fact remains that the deoxypentose nucleic
acids are high-molecular substances which in size resemble, or
even surpass, the proteins. It is quite possible that there exists a
critical range of molecular weights above which two different
cells will prove unable to synthesize completely identical sub-
stances. The enormous number of diverse proteins may be cited
as an example. Duo non faciunt idem is, with respect to cellular
chemistry, perhaps an improved version of the old proverb.

3. PURPOSE

We started in our work from the assumption that the nucleic
acids were complicated and intricate high-polymers, comparable
in this respect to the proteins, and that the determination of their
structures and their structural differences would require the
development of methods suitable for the precise analysis of all
constituents of nucleic acids prepared from a large number of
different cell types. These methods had to permit the study of
minute amounts, since it was clear that much of the material
would not be readily available. The procedures developed in our
laboratory make it indeed possible to perform a complete con-
stituent analysis on 2-3 mg of nucleic acid, and this in six paral-
lel determinations.

The basis of the procedure is the partition chromatography on
filter paper. When we started our experiments, only the quali-
tative application to amino acids was known'^. But it was

References p. 23

6 CHEMICAL SPECIFICITY OF NUCLEIC ACIDS

obvious that the high and specific absorption in the ultraviolet
of the purines and pyrimidines could form the basis of a quanti-
tative ultra-micro method, if proper procedures for the hydrolysis
of the nucleic acids and for the sharp separation of the hydrolysis
products could be found.

4. PREPARATION OF THE ANALYTICAL MATERIAL

If preparations of deoxypentose nucleic acids are to be subjected
to a structural analysis, the extent of their contamination with
pentose nucleic acid must not exceed 2-3%. The reason will
later be made clearer; but I should like to mention here that all
deoxypentose nucleic acids of animal origin studied by us so far
were invariably found to contain much more adenine than
guanine. The reverse appears to be true for the animal pentose
nucleic acids: in them guanine preponderates. A mixture of ap-
proximately equal parts of both nucleic acids from the same tis-
sue, therefore, would yield analytical figures that would cor-
respond, at least as regards the purines, to roughly equimolar
proportions. Should the complete purification — sometimes an
extremely difficult task — prove impossible in certain cases, one
could think of subjecting preparations of both types of nucleic
acid from the same tissue specimen to analysis and of correcting
the respective results in this manner. This, however, is an un-
desirable device and was employed only in some of the prepa-
rations from liver which will be mentioned later.

It is, furthermore, essential that the isolation of the nucleic
acids be conducted in such a manner as to exclude their
degradation by enzymes, acid or alkali. In order to inhibit the
deoxyribonucleases which require magnesium^'^, the preparation
of the deoxypentose nucleic acids was carried out in the presence
of citrate ions^^. It would take us here too far to describe in detail
the methods employed in our laboratory for the preparation of
the deoxypentose nucleic acids from animal tissues. They
represent in general a combination of many procedures, as de-
scribed recently for the isolation of yeast deoxyribonucleic acid^^.

PURINES AND PYRIMIDINES 7

In this manner, the deoxypentose nucleic acids of thymus, spleen,
liver, and also yeast were prepared. The corresponding compound
from tubercle bacilli was isolated via the nucleoprotein^o. The
procedures leading to the preparation of deoxypentose nucleic
acid from human sperm will soon be published^i. All deoxypen-
tose nucleic acids used in the analytical studies were prepared
as the sodium salts (in one case the potassium salt was used);
they were free of protein, highly polymerized, and formed ex-
tremely viscous solutions in water. They were homogeneous
electrophoretically and showed a high degree of monodispersity
in the ultracentrifuge.

The procedure for the preparation of pentose nucleic acids
from animal tissues resembled, in its first stages, the method of
Clarke and Schryver-^. The details of the isolation procedures
and related experiments on yeast ribonucleic acid are as yet
unpublished. Commercial preparations of yeast ribonucleic acid
also were examined following purification. As has been men-
tioned before, the entire problem of the preparation and homo-
geneity of the pentose nucleic acids, and even of the occurrence
of only one type of pentose nucleic acid in the cell, urgently
requires re-examination.

5. SEPARATION AND ESTIMATION OF PURINES AND PYRIMIDINES

Owing to the very unpleasant solubility and polar characteristics
of the purines, the discovery of suitable solvent systems and the
development of methods for their quantitative separation and
estimation-^' ^^ presented a rather difficult problem in the solution
of which Dr. Ernst Vischer had an outstanding part. The pyrim-
idines proved somewhat easier to handle. The choice of the
solvent system for the chromatographic separation of purines and
pyrimidines will, of course, vary with the particular problem. The
efficiency of different solvent systems in effecting separation is
illustrated schematically in Fig. 1. Two of the solvent systems
listed there are suitable for the separation of the purines found
in nucleic acids, i.e., adenine and guanine, namely (1) n-butanol^

References p. 23

CHEMICAL SPECIFICITY OF NUCLEIC ACIDS

I 23 4 5 6 7 8 9 10 II 12 13

Fig. 1. Schematic representation of the position on the paper chromato-
gram of the purines and pyrimidines following the separation of a mixture.
A = adenine, G = guanine, H = hypoxanthine, X = xanthine, U = uracil,
C = cytosine, T = thymine. The conditions under which the separations
were performed are indicated at the bottom: a = acidic, n = neutral, B =
n-butanol, M = morphohne, D = diethylene glycol, Co = collidine, Q =
quinohne. (Taken from E. Vischer and E. Chargaff24.)

morpholine, diethylene glycol, water (column 5 in Fig. 1); and
(2) ^z-butanol, diethylene glycol, water in an NH3 atmosphere
(column 11). The second system Usted proved particularly con-
venient. The separation of the pyrimidines is carried out in
aqueous butanol (column 1).

Following the separation, the location of the various adsorption
zones on the paper must be demonstrated. Our first attempts
to bring this about in ultraviolet light were unsuccessful, probably
because of inadequate filtration of the light emitted by the lamp
then at our disposal. For this reason, the expedient was used of
fixing the separated purines or pyrimidines on the paper as

PURINES AND PYRIMIDINES 9

mercury complexes which then were made visible by their con-
version to mercuric sulfide. The papers thus developed served
as guide strips for the removal of the corresponding zones from
untreated chromatograms that were then extracted and analyzed
in the ultraviolet spectrophotometer. The development of the
separated bases as mercury derivatives has, however, now become
unnecessary, except for the preservation of permanent records,
since there has for some time been available commercially an
ultraviolet lamp emitting short-wave ultraviolet ("Mineralight",
Ultraviolet Products Corp., Los Angeles, California). With the
help of this lamp it is now easy to demonstrate directly the
position of the separated purines and pyrimidines (and also of
nucleosides and nucleotides^^) which appear as dark absorption
shadows on the background of the fluorescing filter paper and
can be cut apart accordingly. (We are greatly indebted to Dr. C.
E. Carter, Oak Ridge National Laboratory, who drew our at-
tention to this instrument*.)

The extracts of the separated compounds are then studied in
the ultraviolet spectrophotometer. The measurement of complete
absorption spectra permits the determination of the purity of the
solutions and at the same time the quantitative estimation of their
contents. The details of the procedures employed have been
pubhshed^*. In this manner, adenine, guanine, uracil, cytosine,
and thymine (and also hypoxanthine, xanthine, and 5-methyl-
cytosine) can be determined quantitatively in amounts of 2-40
jLig. The precision of the method is ±: 4% for the purines and
even better for the pyrimidines, if the averages of a large series
of determinations are considered. In individual estimations the
accuracy is about d= 6%.

Procedures very similar in principle served in our laboratory
for the separation and estimation of the ribonucleosides uridine
and cytidine and for the separation of deoxyribothymidine from
thymine. Methods for the separation and quantitative determina-
tion of the ribonucleotides in an aqueous ammonium isobuty-
rate — isobutyric acid system have likewise been developed^^- 27.
* A similar arrangement was recently described^s.

References p. 23

10 CHEMICAL SPECIFICITY OF NUCLEIC ACIDS

6. METHODS OF HYDROLYSIS

It has long been known that the purines can be spUt off com-
pletely by a relatively mild acid hydrolysis of the nucleic acids.
This could be confirmed in our laboratory in a more rigorous
manner by the demonstration that heating at 100° for 1 hour in
A^ sulfuric acid effects the quantitative liberation of adenine and
guanine from adenylic and guanylic acids, respectively. The
liberation of the pyrimidines, however, requires much more
energetic methods of cleavage. Heating at high temperatures with
strong mineral acid under pressure is usually resorted to. To
what extent these procedures brought about the destruction of
the pyrimidines could not be ascertained previously owing to the
lack of suitable analytical procedures. The experiments sum-
marized in Table 1, which are quoted from a recent paper^^,
show that the extremely robust cleavage methods with mineral

TABLE 1

RESISTANCE OF PYRIMIDINES TO TREATMENT WITH STRONG ACID*

Exper-
iment
No.

Acid

Heat-
ing
time

(min)

Concentration shift,
% of starting
concentration

Uracil

Cytosine

Thymine

1

HCl (10%)

90

+ 62

— 63

+ 3

2

lOiV HCOOH

[ +

60

4- 3

— 5

0

3

TV HCl (1 :

1)

120

+ 24

— 19

0

4
5

HCOOH (98-

100%)

60
120

0
0

— 1

+ 2

— 2

+ 1

* A mixture of pyrimidines of known concentration was dissolved in the
acids indicated below and heated at 175° in a bomb tube. The concen-
tration shifts of the individual pyrimidines were determined through a
comparison of the recoveries of separated pyrimidines before and after
the heating of the mixture.

acids usually employed must have led to a very considerable
degradation of cytosine to uracil. Uracil and also thymine are

DEOXYPENTOSE NUCLEIC ACIDS 11

much more resistant. For this reason, we turned to the hydrolysis
of the pyrimidine nucleotides by means of concentrated formic
acid. For the Hberation of the purines, N sulfuric acid (100°,
1 h) is employed; for the liberation of the pyrimidines, the
purines are first precipitated as the hydrochlorides by treatment
with dry HCl gas in methanol and the remaining pyrimidine
nucleotides cleaved under pressure with concentrated formic acid
(175°, 2 h). This procedure proved particularly suitable for the
investigation of the deoxypentose nucleic acids. For the study of
the composition of pentose nucleic acids a different procedure,
making use of the separation of the ribonucleotides, was devel-
oped more recently, which will be mentioned later.

7. COMPOSITION OF DEOXYPENTOSE NUCLEIC ACIDS

It should be stated at the beginning of this discussion that the
studies conducted thus far have yielded no indication of the oc-
currence in the nucleic acids examined in our laboratory of
unusual nitrogenous constituents. In all deoxypentose nucleic
acids investigated the purines were adenine and guanine, the
pyrimidines cytosine and thymine. The occurrence in minute
amounts of other bases, e.g., 5-methylcytosine, can, however,
not yet be excluded. In the pentose nucleic acids uracil occurred
instead of thymine.

A survey of the composition of deoxyribose nucleic acid ex-
tracted from several organs of the ox is provided in Table 2
(Ref. 29). The molar proportions reported in each case represent
averages of several hydrolysis experiments. The composition of
deoxypentose nucleic acids from human tissues is similarly il-
lustrated in Table 3 (Ref. 30). The preparations from human
liver were obtained from a pathological specimen in which it was
possible, thanks to the kind cooperation of Dr. M. Faber, to
separate portions of unaffected hepatic tissue from carcinom-
atous tissue consisting of metastases from the sigmoid colon,
previous to the isolation of the nucleic acids.

In order to show examples far removed from mammalian

References p. 23

12 CHEMICAL SPECIFICITY OF NUCLEIC ACIDS

TABLE 2
COMPOSITION OF DEOXYRIBONUCLEIC ACID OF OX

(in moles of nitrogenous constituent per mole of P)

Thymus

Spleen

Constituent

T i\>/}r

Prep. 1

Prep. 2

Prep. 3

Prep. 1

Prep. 2

Adenine

0.26

0.28

0.30

0.25

0.26

0.26

Guanine

0.21

0.24

0.22

0.20

0.21

0.20

Cytosine

0.16

0.18

0.17

0.15

0.17

Thymine

0.25

0.24

0.25

0.24

0.24

Recovery

0.88

0.94

0.94

0.84

0.88

TABLE 3

COMPOSITION OF DEOXYPENTOSE NUCLEIC ACID OF MAN

(in moles of nitrogenous constituent per mole of P)

Constituent

Spi

?rm

Thymus

Liver

Prep. 1

Prep. 2

Normal

Carcinoma

Adenine
Guanine
Cytosine
Thymine

Recovery

0.29
0.18
0.18
0.31

0.96

0.27
0.17
0.18
0.30

0.92

0.28
0.19
0.16
0.28

0.91

0.27
0.19

0.27
0.18
0.15
0.27

0.87

organs, the composition of two deoxyribonucleic acids of micro-
bial origin, namely from yeast^^ and from avian tubercle bacilli^^,
is summarized in Table 4 (Ref. 31).

The very far-reaching differences in the composition of
deoxypentose nucleic acids of different species are best illustrated
by a comparison of the ratios of adenine to guanine and of
thymine to cytosine as given in Table 5. It will be seen that in
all cases where enough material for statistical analysis was
available highly significant differences were found. The ana-
lytical figures on which Table 5 is based were derived by com-
paring the ratios found for individual nucleic acid hydrolysates
of one species regardless of the organ from which the preparation

DEOXYPENTOSE NUCLEIC ACIDS

13

TABLE 4

COMPOSITION OF TWO MICROBIAL DEOXYRIBONUCLEIC ACIDS

Constituent

Yeast

Avian

tubercle

bacilli

Prep. 1

Prep. 2

Adenine
Guanine
Cytosine
Thymine

Recovery

0.24
0.14
0.13
0.25

0.76

0.30
0.18
0.15
0.29

0.92

0.12
0.28
0.26
0.11

0.77

was isolated. This procedure assumes that there is no organ
specificity with respect to the composition of deoxypentose
nucleic acids of the same species. That this appears indeed to
be the case may be gathered from Tables 2 and 3 and even better
from Table 6 where the average purine and pyrimidine ratios in
individual tissues of the same species are compared. That the
isolation of nucleic acids did not entail an appreciable fraction-
ation is shown by the finding that when whole defatted human
spermatozoa, after being washed with cold 10% trichloroacetic
acid, were analyzed, the same ratios of adenine to guanine and
of thymine to cytosine were found as are reported in Tables 5
and 6. It should also be mentioned that all preparations, with
the exception of those from human liver, were derived from
pooled starting material representing a number, and in the case
of human spermatozoa a very large number, of individuals.

The deoxypentose nucleic acids extracted from different
species thus appear to be different substances or mixtures of
closely related substances of a composition constant for different
organs of the same species and characteristic of the species.

The results serve to disprove the tetranucleotide hypothesis.
It is, however, noteworthy — whether this is more than accidental,
cannot yet be said — that in all deoxypentose nucleic acids
examined thus far the molar ratios of total purines to total
pyrimidines, and also of adenine to thymine and of guanine to
cytosine, were not far from 1 .

References p. 23

14

CHEMICAL SPECIFICITY OF NUCLEIC ACIDS

53

"Q

s:

a

1^

.c

^

5;

a

c

^

a

St

5

"^

g

^

a ^

it

2 "^

II

o ex

^ 5^ s:

Tf r-- On Tf

T-I ^ ^' d

vo «o r^ »-i

m 00

^ O <N

o o q

d d o

^ d

vo m n

f*^ ts »-• ^

~ ^ c<j .2
O ^ > <

S O

.2 ^

-a ""

S I

T3 *^

5 ^

CO .2

q 2

> c
.5 a

o .5

22 u,
C o

•S >

•a .>

-O C3

(U

B I

^ o

>

J3

73
c«

"3

2i

c«

a
>»

•o

c

cd

03
O

a

u
<u

a.

c«

« o

^ £ 6 c
.2 -c .2

"» ^ 5>

On

££

PENTOSE NUCLEIC ACIDS 15

TABLE 6

MOLAR PROPORTIONS OF PURINES AND PYRIMIDINES IN DEOXYPENTOSE
NUCLEIC ACIDS FROM DIFFERENT ORGANS OF ONE SPECIES

Species

Organ

A/G

T/C

Thymus

1.3

1.4

Ox

Spleen

1.2

1.5

Liver

1.3

Thymus

1.5

1.8

Sperm

1.6

1.7

Man

Liver (normal)

1.5

1.8

Liver (carcinoma)

1.5

1.8

Abbreviations: A = adenine, G = guanine, T = thymine, C = cytosine.

8. COMPOSITION OF PENTOSE NUCLEIC ACIDS

Here a sharp distinction must be drawn between the prototype
of all pentose nucleic acid investigations — the ribonucleic acid
of yeast — and the pentose nucleic acids of animal cells. Nothing
is known as yet about bacterial pentose nucleic acids. In view
of the incompleteness of our information on the homogeneity of
pentose nucleic acids, which I have stressed before, I feel that
the analytical results on these preparations do not command the
same degree of confidence as do those obtained for the deoxy-
pentose nucleic acids.

Three procedures, to which reference is made in Tables 7
(Refs. 25, 28) and 8, were employed in our laboratory for the
analysis of pentose nucleic acids.

In Procedure 1, the pentose nucleic acid was hydrolyzed to
the nucleotide stage with alkaU, at pH 13.5 and 30°, and the
nucleotides, following adjustment to about pH 5, separated by
chromatography with aqueous ammonium isobutyrate-isobutyric
acid as the solvent. Under these conditions, guanylic acid shares
its position on the chromatogram with uridylic acid; but it is pos-
sible to determine the concentrations of the two components in
the eluates by simultaneous equations based on the ultraviolet

References p. 23

16 CHEMICAL SPECIFICITY OF NUCLEIC ACIDS

TABLE 7
COMPOSITION OF YEAST RIBONUCLEIC ACIDS

(in moles of nitrogenous constituent per mole of P)

Preparation 1

Preparation 2

Preparation 3

Constituent

Proc.

Proc.

Proc.

Proc.

Proc.

Proc.

Proc.

Proc. Proc.

1

2

3

1

2

3

1

2 3

Adenylic acid

0.29

0.26

0.26

0.11

0.24

0.25

0.23 0.24

Guanylic acid

0.28

0.29

0.26

0.25

0.25

0.26

0.28 0.26

Cytidylic acid

0.18

0.17

0.24

0.20

0.21

0.21

Uridylic acid

0.20

0.20

0.08

0.18

0.19

0.20

0.25

Recovery

0.95

0.92

0.84

0.90

0.92

0.97

absorption of the pure nucleotides^^' ^^. The very good recoveries
of nucleotides obtained in terms of both nucleic acid phosphorus
and nitrogen show the cleavage by mild alkali treatment of
pentose nucleic acids to be practically quantitative.

In Procedure 2, the purines are first Uberated by gaseous HCl
in dry methanol and the evaporation residue of the reaction
mixture is adjusted to pH 13.5 and then treated as in Procedure 1.
In this manner, uridylic and cytidylic acids, adenine and guanine
are separated and determined on one chromatogram.

The determinations of free purines and pyrimidines in acid
hydrolysates of pentose nucleic acids, following the methods
outlined before for the deoxypentose nucleic acids, are listed as
Procedure 3. It will be seen that it is mainly uracil which in this
procedure escapes quantitative determination. This is due to the
extreme refractoriness of uridylic acid to complete hydrolysis by
acids, a large portion remaining partially unsplit as the nucleoside
uridine. As matters stand now, I consider the values for purines
yielded by Procedures 1 and 3 and those for pyrimidines found
by Procedures 1 and 2 as quite reliable.

A survey of the composition of yeast ribonucleic acid is pro-
vided in Table 7. Preparations 1 and 2, listed in this table, were
commercial preparations that had been purified in our laboratory
and had been subjected to dialysis; Preparation 3 was isolated

PENTOSE NUCLEIC ACIDS 17

from baker's yeast by Magasanik in this laboratory by procedures
similar to those used for the preparation of pentose nucleic acids
from animal tissues and had not been dialyzed. It will be seen
that the results are quite constant and not very far from the
proportions required by the presence of equimolar quantities of
all four nitrogenous constituents.

An entirely different picture, however, was encountered when
the composition of pentose nucleic acids from animal cells was
investigated. A preliminary summary of the results, in all cases
obtained by Procedure 1, is given in Table 8. Here guanylic acid
was the preponderating nucleotide followed, in this order, by
cytidylic and adenylic acids; uridylic acid definitely was a minor
constituent. This was true not only of the ribonucleic acid of
pancreas which has been known to be rich in guanine^^-^^, but
also of all pentose nucleic acids isolated by us from the livers of
three different species (Table 8).

TABLE 8

COMPOSITION OF PENTOSE NUCLEIC ACIDS FROM ANIMAL TISSUES

Constituent

Calf
liver

Ox

liver

Sheep
liver

Pig
liver

Pis

pancreas

Guanylic acid

16.3

14.7

16.7

16.2

11.$

Adenylic acid

10

19

10

10

10

Cytidylic acid

11.1

10.9

13.4

16.1

9.8

Uridylic acid

5.3

6.6

5.6

1.1

4.6

Purines :pyrimidines

1.6

1.4

1.4

1.1

2.3

In the absence of a truly reliable standard method for the
isolation of pentose nucleic acid from animal tissue, generali-
zations are not yet permitted; but it would appear that pentose
nucleic acids from the same organ of different species are more
similar to each other, at least in certain respects {e.g., the ratio
of guanine to adenine), than are those from different organs of
the same species. (Compare the pentose nucleic acids from the
liver and the pancreas of pig in Table 8.)

References p. 23

18 CHEMICAL SPECIFICITY OF NUCLEIC ACIDS

9. SUGAR COMPONENTS

It is deplorable that such designations as deoxyribose and ribose
nucleic acids continue to be used as if they were generic terms.
Even the "thymus nucleic acid of fish sperm" is encountered in
the literature. As a matter of fact, only in a few cases have the
sugars been identified, namely, J-2-deoxyribose as a constituent
of the guanine and thymine nucleosides of the deoxypentose
nucleic acid from calf thymus, D-ribose as a constituent of the
pentose nucleic acids from yeast, pancreas, and sheep liver.

Since the quantities of novel nucleic acids usually will be in-
sufficient for the direct isolation of their sugar components, we
attempted to employ the very sensitive procedure of the filter
paper chromatography of sugars^^^^^ for the study of the sugars
isolated from minute quantities of nucleic acids. It goes without
saying that identifications based on behavior in adsorption or
partition are by no means as convincing as the actual isolation,
but they will at least permit a tentative classification of new
nucleic acids. Thus far the pentose nucleic acids of pig pancreas^^
and of the avian tubercle bacillus^^ have been shown to contain
ribose, the deoxypentose nucleic acids of ox spleen^^, yeast and
avian tubercle bacilli^^ deoxyribose. It would seem that the free
play with respect to the variability of components that nature
permits itself is extremely restricted where nucleic acids are con-
cerned.

10. DEPOLYMERIZING ENZYMES

Enzymes capable of bringing about the depolymerization of both
types of nucleic acids have long been knovv'n; but it is only during
the last decade that crystalline ribonuclease^^ and deoxyribonu-
clease^^ from pancreas have become available thanks to the work
of Kunitz. Important work on the latter enzyme was also done
by McCartyi^.

We were, of course, interested in applying the chromato-
graphic micromethods for the determination of nucleic acid

DEPOLYMERIZING ENZYMES 19

TABLE 9

ENZYMIC DEGRADATION OF CALF THYMUS DEOXYRIBONUCLEIC ACID

Distribution

Composition

of fractions

Diges-

Dialysis

of

{molar proportions)

tion
{hours)

{hours)

fractions -

{%of
original)

A/G

T/C

A/C

Pu/Py

Original 0

0

100

1.2

1.3

1.6

1.2

Dialysate 6

6

53

1.2

1.2

1.2

1.0

Dialysis

residue 24

72

7

1.6

2.2

3.8

2.0

Abbreviations: A = adenine, G = guanine, T = thymine, C = cytosine,
Pu = purines, Py = pyrimidines.

constituents to studies of enzymic reaction mechanisms for which
they are particularly suited. The action of crystalline deoxy-
ribonuclease on calf thymus deoxyribonucleic acid resulted in
the production of a large proportion of dialyzable fragments
(53% of the total after 6 hours digesdon), without liberarion of
ammonia or inorganic phosphate. But even after extended
digestion there remained a non-dialyzable core whose com-
position showed a significant divergence from both the original
nucleic acid and the bulk of the dialysate. The preliminary
findings summarized in Table 9 (Ref. 40) indicate a considerable
increase in the molar proportions of adenine to guanine and
especially to cytosine, of thymine to cytosine, and of purines to
pyrimidines. This shows that the dissymmetry in the distribution
of constituents, found in the original nucleic acid (Table 2), is
intensified in the core. The most plausible explanations of this
interesting phenomenon, the study of which is being continued,
are that the preparations consisted of more than one deoxy-
pentose nucleic acid or that the nucleic acid contained in its chain
clusters of nucleotides (relatively richer in adenine and thymine)
that were distinguished from the bulk of the molecule by greater
resistance to enzymic disintegration.

In this connection another study, carried out in collaboration
with Zamenhof, should be mentioned briefly that dealt with the

References p. 23

20 CHEMICAL SPECIFICITY OF NUCLEIC ACIDS

deoxypentose nuclease of yeast cells^^-^^ xhis investigation af-
forded a possibility of exploring the mechanisms by which an
enzyme concerned with the disintegration of deoxypentose
nucleic acid is controlled in the cell. Our starting point again was
the question of the specificity of deoxypentose nucleic acids; but
the results were entirely unexpected. Since we had available a
number of nucleic acids from different sources, we wanted to
study a pair of deoxypentose nucleic acids as distant from each
other as possible, namely that of the ox and that of yeast, and
to investigate the action on them of the two deoxypentose
nucleases from the same cellular sources. The deoxyribonuclease
of ox pancreas has been thoroughly investigated, as was men-
tioned before. Nothing was known, however, regarding the exist-
ence of a yeast deoxypentose nuclease.

It was found that fresh salt extracts of crushed cells contained
such an enzyme in a largely inhibited state, due to the presence
of a specific inhibitor protein. This inhibitor specifically inhibited
the deoxypentose nuclease from yeast, but not that from other
sources, such as pancreas. The yeast enzyme depolymerized the
deoxyribose nucleic acids of yeast and of calf thymus, which dif-
fer chemically, as I have emphasized before, at about the same
rate. In other words, the enzyme apparently exhibited inhibitor
specificity, but not substrate specificity. It is very inviting to as-
sume that such relations between specific inhibitor and enzyme,
in some ways reminiscent of immunological reactions, are of
more general biological significance. In any event, a better under-
standing of such systems will permit an insight into the delicate
mechanisms through which the cell manages the economy of its
life, through which it maintains its own continuity and protects
itself against agents striving to transform it.

11. CONCLUDING REMARKS

Generalizations in science are both necessary and hazardous;
they carry a semblance of finality which conceals their essentially
provisional character; they drive forward, as they retard; they

CONCLUDING REMARKS 21

add, but they also take away. Keeping in mind all these reser-
vations, we arrive at the following conclusions. The deoxy-
pentose nucleic acids from animal and microbial cells contain
varying proportions of the same four nitrogenous constituents,
namely, adenine, guanine, cytosine, thymine. Their composition
appears to be characteristic of the species, but not of the tissue,
from which they are derived. The presumption, therefore, is that
there exists an enormous number of structurally different nucleic
acids; a number, certainly much larger than the analytical
methods available to us at present can reveal.

It cannot yet be decided, whether what we call the deoxy-
pentose nucleic acid of a given species is one chemical individual,
representative of the species as a whole, or whether it consists of
a mixture of closely related substances, in which case the con-
stancy of its composition merely is a statistical expression of the
unchanged state of the cell. The latter may be the case if, as
appears probable, the highly polymerized deoxypentose nucleic
acids form an essential part of the hereditary processes; but it
will be understood from what I said at the beginning that a
decision as to the identity of natural high polymers often still is
beyond the means at our disposal. This will be particularly true
of substances that differ from each other only in the sequence,
not in the proportion, of their constituents. The number of pos-
sible nucleic acids having the same analytical composition is
truly enormous. For example, the number of combinations ex-
hibiting the same molar proportions of individual purines and
pyrimidines as the deoxyribonucleic acid of the ox is more than
10^^, if the nucleic acid is assumed to consist of only 100
nucleotides; if it consists of 2,500 nucleotides, which probably
is much nearer the truth, then the number of possible "isomers"
is not far from lO^^^^^.

Moreover, deoxypentose nucleic acids from different species
differ in their chemical composition, as I have shown before; and
I think there will be no objection to the statement that, as far
as chemical possibilities go, they could very well serve as one of
the agents, or possibly as the agent, concerned with the transmis-

References p. 23

22 CHEMICAL SPECIFICITY OF NUCLEIC ACIDS

sion of inherited properties. It would be gratifying if one could
say — but this is for the moment no more than an unfounded
speculation — that just as the deoxypentose nucleic acids of the
nucleus are species-specific and concerned with the maintenance
of the species, the pentose nucleic acids of the cytoplasm are
organ-specific and involved in the important task of differen-
tiation.

I should not want to close without thanking my colleagues who
have taken part in the work discussed here; they are, in alpha-
betical order, Miss R. Doniger, Mrs. C. Green, Dr. B. Magasanik,
Dr. E. Vischer, and Dr. S. Zamenhof.

CHEMICAL SPECIFICITY OF NUCLEIC ACIDS 23

REFERENCES

1 E. Hammarsten, Biochem. Z., 144 (1924) 383.

2 J. Bracket, in Nucleic Acid, Symposia Soc. Exptl. Biol, 1 (1947) 207;

cf. J. Brachet, in Nucleic Acids and Nucleoproteins, Cold Spring
Harbor Symposia Quant. Biol, 12 (1947) 18.

3 T. Caspersson, in Nucleic Acid, Symposia Soc. Exptl. Biol., 1 (1947)

127.

4 O. T. Avery, C. M. MacLeod and M. McCarty, J. Exptl. Med., 79

(1944) 137.

5 P. A. Levene and L. W. Bass, Nucleic Acids, Chemical Catalog Co.,

New York, 1931.

6 H. Bredereck, Fortschr. Chem. org. Naturstoffe, 1 (1938) 121.

7 F. G. Fischer, Naturwissenschaften, 30 (1942) 377.

8 R. S. TiPSON, Advances in Carbohydrate Chem., 1 (1945) 193.

^ J. M. Gulland, G. R. Barker and D. O. Jordan, Ann. Rev. Biochem.,
14 (1945) 175.

10 E. Chargaff and E. Vischer, Ann. Rev. Biochem., 17 (1948) 201.

11 F. ScHLENK, Advances in Enzymol., 9 (1949) 455.

12 E. Chargaff, in Nucleic Acids and Nucleoproteins, Cold Spring Har-

bor Symposia Quant. Biol., 12 (1947) 28.

13 E. Hammarsten, Acta Med. Scand., Suppl. 196 (1947) 634.

14 G. Jungner, L Jungner and L.-G. Allgen, Nature, 163 (1949) 849.

15 R. Cecil and A. G. Ogston,7. Chem. Soc, (1948) 1382.

16 R. CoNSDEN, A. H. Gordon and A. J. P. Martin, Biochem. J., 38

(1944) 224.

17 F. G. Fischer, L Bottger and H. Lehmann-Echternacht, Z. phy-

siol. Chem., Hoppe-Seyler's, 111 (1941) 246.

18 M. McCarty, J. Gen. Physiol, 29 (1946) 123.

19 E. Chargaff and S. Zamenhof, J. Biol. Chem., 173 (1948) 327.

20 E. Chargaff and H. F. Saidel, J. Biol. Chem., Ill (1949) 417.

21 S. Zamenhof, L. B. Shettles and E. Chargaff, Nature, 165 (1950)

756.

22 G. Clarke and S. B. Schryver, Biochem. J., 11 (1917) 319.

23 E. Vischer and E. Chargaff, J. Biol Chem., 168 (1947) 781.

24 E. Vischer and E. Chargaff, /. Biol Chem., 176 (1948) 703.

25 E. Chargaff, B. Magasanik, R. Doniger and E. Vischer, J. Am.

Chem. Soc, 71 (1949) 1513.

26 E. R. Holiday and E. A. Johnson, Nature, 163 (1949) 216.

27 E. Vischer, B. Magasanik and E. Chargaff, Federation Proc, 8

(1949) 263.

28 E. Vischer and E. Chargaff, /. Biol Chem., 176 (1948) 715.

29 E. Chargaff, E. Vischer, R. Doniger, C. Green and F. Misani,

J. Biol Chem., Ill (1949) 405.

30 E. Chargaff, S. Zamenhof and C. Green, Nature, 165 (1950) 756.

31 E. Vischer, S. Zamenhof and E. Chargaff, J. Biol Chem., Ill (1949)

429.

24 CHEMICAL SPECIFICITY OF NUCLEIC ACIDS

32 E. Hammarsten, Z. physiol. Chem., Hoppe-Seyler's, 109 (1920) 141.

33 p. A. Levene and E. Jorpes, /. Biol. Chem., 86 (1930) 389.

34 E. Jorpes, Biochem. J., 28 (1934) 2102.

35 S. M. Partridge, Nature, 158 (1946) 270.

36 S. M. Partridge and R. G. Westall, Biochem. J., 42 (1948) 238.

37 E. Chargaff, C. Levine and C. Green, /. Biol. Chem., 175 (1948) 67.

38 M. KUNITZ, J. Gen. Physiol., 24 (1940) 15.

39 M. KUNITZ, Science, 108 (1948) 19.

40 s. Zamenhof and E. Chargaff, J. Biol. Chem., 178 (1949) 531.

41 S. Zamenhof and E. Chargaff, Science, 108 (1948) 628.

^2 S. Zamenhof and E. Chargaff, /. Biol. Chem., 180 (1949) 727.

CHAPTER 2

Structure and Function of Nucleic Acids as
Cell Constituents'^

It is safe to say that living systems require the presence of both
types of nucleic acid or, in the case of parasitic systems, the
presence of at least one. If, to use a designation of Schr6dinger\
one refers to the chromosomes as "the hereditary code-script",
the great biological importance of all components of these nuclear
structures, viz., nucleic acids, proteins, and, perhaps, lipids, is
obvious, unless one assumes that one or the other of these com-
ponents has been added by nature as a meaningless and purely
decorative flourish. This is, however, not likely.

In animals and higher plants the deoxypentose nucleic acids
(DNA) are exclusively or almost exclusively situated in the
nucleus. Pentose nucleic acids (PNA) are present in the nucleoli
and the various cytoplasmic elements, e.g., the mitochondria,
submicroscopic particles, etc. Recent work with Elson^ on the
nucleotide composition of PNA in different fractions of rat liver
cells has provided preUminary evidence of differences in com-
position between nuclear and cytoplasmic PNA. It is not yet
known whether this will be generally true; but results, recently
reported by Magasanik^, seem to point to the presence in pig
liver PNA of two differently composed fractions.

Whether DNA really is hmitedto the nucleus is not entirely
certain, since the available cytochemical or cytophysical methods
presumably require the presence of a compact mass of DNA
and may not reveal its occurrence in a diffusely distributed form,
as could be the case in the cytoplasm of egg cells (c/. Ref. 4).

* Reprinted with permission from Federation Proc, 10 (1951) 654-659.
References p. 37

26 NUCLEIC ACIDS AS CELL CONSTITUENTS

The bacterial cell represents a special case. Whether the micro-
bial nucleus is the sole repository of DNA in microorganisms
cannot yet be decided with certainty.

1. BIOLOGICAL SIGNIFICANCE OF NUCLEIC ACIDS

The discussion of the biological significance of a ubiquitous cell
constituent is, strictly speaking, superfluous. But there exist a few
important instances pointing to a direct involvement of nucleic
acids; and some of them will be listed here briefly.

All virus preparations so far described are, or contain,
nucleoproteins (cf. the recent survey of Davidson^). The same
seems to be true of intracellular parasites, such as rickettsiae^- '^
or paramecin^.

Specific DNA preparations are known which are able to in-
duce the transformation of bacterial types. This extremely im-
portant phenomenon, first discovered in pneumococci®, has later
been shown to operate also in E. coli^^ and in Hemophilus
injluenzae^^ {cf. also Ref. 12). The possibility that reactions of
this kind are of more general biological importance and not
limited to the field of bacterial transformations cannot be reject-
ed. What appears particularly remarkable is that it is here the
free nucleic acid and not a nucleoprotein (as in the case of viruses)
that is able to impose its own synthesis on the receptor cell,
whereas in general nucleic acids seem to occur in cells only in
the form of conjugated nucleoproteins. The mechanisms through
which these transformations take place and the chemical features
distinguishing these biologically active DNA specimens are com-
pletely obscure. There is, however, httle doubt that it is the
bacterial DNA itself, or a particular DNA fraction present in the
transforming preparations, which is the carrier of activity. Recent
work on the agent operative in the transformation of H. influenzae
has shown that highly purified DNA preparations from two types
are active in extremely low concentrations: 0.0004 //g of DNA
per ml in type b; 0.01 /^g in type c^^.

The investigation of the relative efficiencies of different wave-

DEOXYPENTOSE NUCLEIC ACIDS 27

lengths of ultraviolet light has yielded curves closely resembling,
but not entirely identical with, the UV absorption spectrum of
nucleic acids. This work, mainly due to Stadler and Uber, Hol-
laender and Emmons, and Knapp and Schreiber, has been
reviewed by Lea^^.

A certain degree of constancy, within the same species, of the
DNA concentration per diploid nucleus, and of roughly one-half
this amount per haploid sperm nucleus, has been discovered by
Boivin and his collaborators*- ^^' ^^ and confirmed in other labo-
ratories^^- ^^.

DNA is in its composition identical in different tissues of the
same species^^. Moreover, in the few cases where comparison
was possible, no chemical differences have been observed be-
tween the composition of the DNA from the sperm cells and
from differentiated tissues of the same species, in contrast to the
very different composition of nuclear proteins in such instances.

2. DEOXYPENTOSE NUCLEIC ACIDS

All DNA preparations that have been studied in detail have

several features in common. They are asymmetrical molecules of

high molecular weight (around 10^), yielding extremely viscous

solutions. They appear to contain the same deoxy sugar, namely

2-deoxyribose. They contain two purines, adenine and guanine,

and two or three pyrimidines, viz. thymine and cytosine, and in

several cases^^ also 5-methylcytosine. The deoxyribonucleotides

released enzymically from calf thymus DNA appear to be the
5'-phosphates-i'^2^

The conclusions to which our work has led us have been sum-
marized recentlyi^' 2^, and I shall limit myself here only to the
main points.

a. DNA is in its composition characteristic of the species from
which it is derived. This can in many, but not in all, cases be
demonstrated by determining the ratios in which the individual
purines and pyrimidines occur. There will, however, be very
many borderhne cases in which such differences in composition

References p. 37

28 NUCLEIC ACIDS AS CELL CONSTITUENTS

are not sufficiently significant to permit their use as the sole
criterion of differentiation. The most important question of the
sequence in which, in a particular nucleic acid, the nucleotides
follow each other has so far barely been approached. The
elaboration of methods for sequence analysis is, perhaps, one
of the most urgent problems in nucleic acid chemistry, since
differences in nucleotide sequence may very well be among the
determinants of chemical and biological specificity.

b. No differences in composition have so far been found in
DNA from different tissues of the same species. This provisional
conclusion refers only to the over-all composition. It is in this
connection noteworthy that no chemical differences appear to
exist between the composition of DNA from normal human
tissue^* and that of preparations from human cancer tissue^s.

c. The tetranucleotide hypothesis is incorrect.

d. There exist a number of regularities. Whether these are
merely accidental cannot yet be decided. In almost all DNA
preparations studied until now the ratio of total purines to total
pyrimidines never was far from 1 . Similarly the ratios of adenine
to thymine and of guanine to cytosine were near 1.

e. There appear to exist two main groups of DNA, namely
the "AT type", in which adenine and thymine predominate, and
the "GC type", in which guanine and cytosine are the major
constituents. The latter has so far been found only in certain
microorganisms^^.

Data to support these conclusions have, in the main, been
presented in previous publications and were summarized recent-
lyi9,23 J should like to limit myself here to the discussion of a
few as yet unpublished results. The study of the composition of
the DNA of salmon sperm^^ has brought out some of the regu-
larities mentioned before particularly clearly (Tables 10 and 11).
This substance belongs to the "AT type".

Another investigation, undertaken in collaboration with G.
Brawerman, deals with the DNA of wheat germ and the course
of its degradation by crystalline pancreatic deoxyribonuclease.
Previous work in our laboratory on the enzymic disintegration

DEOXYPENTOSE NUCLEIC ACIDS

29

TABLE 10

SALMON SPERM DNA; PROPORTIONS

(in moles of nitrogenous constituent per mole of P in hydrolysate)

Constituent

Mean
proportion

Standard
error

Adenine

0.280

0.005

Guanine

0.196

0.004

Cytosine

0.192

0.006

Thymine

0.274

0.005

TABLE 11

SALMON SPERM DNAl MOLAR RELATIONSHIPS

Adenine to guanine
Thymine to cytosine
Adenine to thymine
Guanine to cytosine
Purines to pyrimidines

P accounted for as percentage of P in hydrolysate

Average number of gram-atoms N per mole constituent

Atomic N : P ratio in DNA preparations

1.43

1.43

1.02

1.02

1.02

95.8 (± 1.6)

3.7

3.6, 3.7

TABLE 12

WHEAT GERM DNA; INTACT PREPARATION AND ENZYMICALLY PRODUCED CORES

(in moles of nitrogenous constituent per mole of P in hydrolysate)

Constituent

Intact
DNA

19%
core

8%
core

Adenine

0.27

0.33

0.35

Guanine

0.22

0.20

0.20

Cytosine

0.16 ^

0.12

0.10

5-Methylcytosine

0.06

0.04

0.04

Thymine

0.27

0.26

0.23

Total purines

0.49

0.53

0.55

Total pyrimidines

0.49

0.42

0.37

Recovery

0.98

0.95

0.92

References p. 37

30

NUCLEIC ACIDS AS CELL CONSTITUENTS

.1 ^

cog

o

t-(

?^

X

•i:

<ii

«i)

.*»

5

a

1

"S:

?^

K

C

<4)

^

s:

s*

c

O

s»

<;;

.^

"Q

^

60

Tf<^»OTj-Tt(^«ovor^vor>.

a\ (N On

^OOsOONOOO^OsO

OO'-HfNlt^O'-'OSOOvoON
OOa\OCT\(N|ONONOOOOO

o o

OOOOOOOOOOv'-^

mu-)0\moofN|'^>o

-5 -^

C3 ^

'(73

-2 ^

■(-« 00

3 •-

C 9

C "-M

O OS

c<o C

3 O

DEOXYPENTOSE NUCLEIC ACIDS 31

of calf thymus DNA^^ had shown it to proceed according to a
peculiar and complex pattern. The action of the enzyme resulted
in the formation of dialyzable fragments and of a dialysis residue
(core). The latter was characterized by greatly increased ratios
of adenine to guanine, thymine to cytosine, purines to pyrimi-
dines, and by greater resistance to enzymic attack. Preliminary
results of studies of this type on wheat germ DNA are presented
in Table 12. The analytical findings on the intact DNA are in
good agreement with analyses recently reported for the same
nucleic acid from two other laboratories^^- ^^. The figures given
here for the "19% core" and the "8% core" refer to the dialysis
residues recovered when 81 and 92% of the DNA respectively
had been converted by the enzyme to dialyzable fragments. The
trend of degradation appears similar to that observed with calf
thymus DNA.

The molar ratios found in the DNA specimens studied in our
laboratory are compared in Table 13. The tendency toward
certain regularities will be observed. The figures for hen DNA
(chicken erythrocytes) and for the DNA from the K-12 strain of
E. coli must be considered as preliminary. Both components were
studied in collaboration with B. Gandelman. The DNA of
Hemophilus influenzae, type c, was studied by S. Zamenhof. A
few points are noteworthy. In the case of wheat germ DNA,
methylcytosine and cytosine apparently must be considered
together, if the regular ratios, observed in most other instances,
are to be obtained. Another remarkable fact is that the regu-
larities are maintained even in the nucleic acids of the "GC type"
despite the complete inversion in individual ratios.

As was already mentioned before, it is almost impossible to
decide at present whether these regularities are entirely fortuitous
or whether they reflect the existence in all DNA preparations of
certain common structural principles, irrespective of far-reaching
differences in their individual composition and the absence of an
easily recognizable periodicity. It may be assumed that the
nucleic acids as we know them today, as for that matter also the
proteins, are the result of an age-long selection process in the

References p. 37

32 NUCLEIC ACIDS AS CELL CONSTITUENTS

course of which many less suitable or less stable components
must have been eliminated. One could speak of the survival of
the fittest nucleic acids. Such macromolecules will exhibit diver-
sity and uniformity at the same time, since they are called upon
to perform, in diverse species, the same tasks. It is, therefore,
perhaps not astonishing that the nucleic acids will share certain
features that may be directly connected with their stability or
their ability to form conjugated nucleoproteins. One property
which is quite striking is the uniform absorption spectrum in the
ultraviolet of all highly polymerized DNA specimens, both with
respect to the position and the intensity of the absorption
maximum. The center of absorption fluctuates only between 257
and 261 m/i and the (e) P is around 6600. Another surprising
feature is the balance between amino groups and enolic hydroxyls
in all DNA preparations examined by us. (See last column of
Table 13.) Even in the core preparations, this ratio changed
only very little.

But let us return for a moment to the other outstanding char-
acteristic of nucleic acids, viz., their diversity. If we accept the
evidence of the existence of species-specific DNA, then there
arise many new questions, both of a biological and chemical
nature. DNA presumably is an important part of the chro-
mosomes and may be surmised to be involved in their biological
functions. Does this mean that a cell contains as many different
DNA individuals as it contains genes? Or can one and the same
species-specific DNA form so many three-dimensional structures,
in connection with the proteins to which it is attached, that the
genie requirements are fulfilled? (For a more detailed discussion
of some of these points, cf. Ref. 31.) This question, as so many
others in this field, cannot yet be answered. No way has as yet
been found to fractionate a family of very similar macromolecules
which may differ in no more than the sequence of a few of their
component nucleotides. But one could perhaps say that the more
regular the arrangement of nucleotides is in a given DNA, the
less the chance of its forming many different specific structures.
And this brings us again to the very important question of

DEOXYPENTOSE NUCLEIC ACIDS 33

nucleotide sequence and periodicity in a nucleic acid chain. Any
simplified assumption with respect to periodicity has been dis-
proved by the studies on the course of action of deoxyribonu-
clease on the DNA of calf thymus-^ and of wheat germ. The
composition of both the dialyzable degradation products and the
dialysis residues, the "cores", exhibited continuous and charac-
teristic changes with respect to the distribution of purines and
pyrimidines. One must conclude that the sequence is highly
aperiodic and that it is not inconceivable that the same cellular
DNA could give rise to many different nucleoproteins, depending
upon the shape and configuration of the particular protein.

It must, moreover, be understood that the recognition of
periodicity, i.e., the presence of recurring units, will be particu-
larly difficult in a macromolecule of the type of DNA. If, for
instance, in a chain composed of 3000 nucleotides a particular
sequence of 100 consecutive nucleotides were repeated 30 times,
this periodicity could not be recognized, unless we had a method
producing cleavage only at the points where these repeating units
are joined. In other words, the perception of periodicity would
require the proper distance for a bird's-eye view which will not
be easy to attain.

Another approach to the problem of sequence analysis in
DNA may be seen in the study of its controlled chemical degra-
dation. That the purines can be detached from a nucleic acid with
much greater ease than can the pyrimidines has long been known;
but the resulting end product, thymic acid, was rather non-
descript. In collaboration with M. E. Hodes and C. Tamm it has
been possible to develop procedures, soon to be published, in
which all purines could be cleaved from a DNA preparation,
leaving behind a non-dialyzable product in a yield of about 94%
that retained all the pyrimidines of the original DNA in unchanged
proportions. We have designated preparations of this type as
apurinic acid. These compounds may prove of interest for struc-
tural studies on DNA, since the position in the DNA chain of the
initially present purine nucleotides is now marked by reactive
aldehydo groups. The properties of a typical preparation of

References p. 37

34

NUCLEIC ACIDS AS CELL CONSTITUENTS

TABLE 14

APURINIC ACID FROM CALF THYMUS DNA

DNA

Apurinic
acid

N
P.

N/P

Absorption maximum, m/i*
Absorption maximum, e (?)
Optical rotation**
Adenine, moles per mole P
Guanine, moles per mole P
Cytosine, moles per mole P
Thymine, moles per mole P
Thymine to cytosine

14.9

6.0

9.1

10.7

3.6

1.2

258.5

268

6,600

4,600

102°

+ 50°

0.28

0

0.21

0

0.21

0.18

0.26

0.24

1.3

1.3

* In M phosphate buffer, pH 7.1.
** In 0.1 M phosphate buffer, pH 7.1.

TABLE 15

ACTION OF RIBONUCLEASE ON PNA

Source Fraction

Nucleotide composition
{as moles 1 100 M nucleotide in starting material)

Guanylic Adenylic Cy tidy lie Uridylic j . j
acid acid acid acid

Yeast

Pig liver

Starting

30

Dialyzable

(total nu-

cleotides)

18

Dialyzable

(mononu-

cleotides)

0

Core

66

Starting

36

Dialyzable

(total nu-

cleotides)

18

Dialyzable

(mononu-

cleotides)

0

Core

71

21
15

0
18

19

14

0
15

23

19

14
8

29

26

18
0

26 100

26 78

19 33

8

16 100

15 73

10 28

14

FINAL REMARKS 35

apurinic acid are contrasted in Table 14 with those of the calf
thymus DNA specimen that served as the starting material.

3. PENTOSE NUCLEIC ACIDS

Time does not permit an adequate discussion of the chemistry
of PNA. In the past few years, significant contributions to this
field were made by GuUand, Kerr, Schmidt, Allen, Loring, Cohn,
and many other workers. I should Uke to make brief mention of
some recent work done by Magasanik in our laboratory in which
the course of action of crystalUne ribonuclease on PNA from
yeast and from pig liver was studied with the use of chromato-
graphic and spectroscopic procedures. About 60-70%, or in
some cases somewhat more, of the nucleotides present in the
initial substrates were liberated by enzymic action as rapidly
dialyzable nucleotides of low molecular weight. This fraction
consisted of free cytidylic and uridylic acids, which comprised
a high proportion of the total pyrimidine nucleotides, and of
combined purine and pyrimidine nucleotides. All PNA prepara-
tions yielded a non-dialyzable residue, resistant to enzymic at-
tack, which was found to consist to about two thirds of guanyUc
acid and of varying amounts of the other nucleotides. A selection
of these experiments is presented in Table 15. These and other
findings have led us to certain conceptions concerning the
specificity of ribonuclease and the structure of PNA; but refer-
ence must be made to a more detailed article^.

4. FINAL REMARKS

It is fitting to conclude this all toa sketchy survey with a confes-
sion of ignorance. What our studies have taught us more than
anything else is how Httle we know as yet about the chemistry
of nucleic acids. The chemical specificity of macromolecules and
the interactions between them through which the organization of
the cell is maintained can only partly be understood in terms of
our present knowledge. In the approach to a scientific problem

References p. 37

36 NUCLEIC ACIDS AS CELL CONSTITUENTS

two principles are operative: generalization and simplification.
Both are necessary and both dangerous. It is obvious that we can
learn more geometry from the illustrations in a textbook on
projective geometry than from the beautiful pictures in Sir
D'Arcy Thompson's On Growth and Form. But it is difficult to
say where the danger line lies beyond which oversimplification
will produce a dogmatic ignorance. Should we stress the multi-
formity of nature, which makes us forget the simplicity of its
basic designs; or should the essential shape win over the ac-
cidental form? In Wycherley's The Country Wife a quack is ad-
dressed as follows: "Doctor, thou wilt never make a good chemist,
thou art so incredulous and impatient". If patience and credulity
were all the chemist needed, the problem of the nucleic acids —
still so baffUng and elusive — would have been solved a long time
ago.

NUCLEIC ACIDS AS CELL CONSTITUENTS 37

REFERENCES

1 E. ScHRODiNGER, What is Life?, Cambridge University Press, London,

1945, p. 19.

2 D. Elson and E. Chargaff, Federation Proc, 10 (1951) 180.

3 B. Magasanik and E. Chargaff, Biochim. Biophys. Acta, 7 (1951)

396.
-* C. Vendrely and R. Vendrely, Compt. rend. Soc. Biol., 143 (1949)
1386.

5 J. N. Davidson, The Biochemistry of the Nucleic Acids, Methuen, Lon-

don, 1950, p. 136.

6 S. S. Cohen and E. Chargaff, J. Biol. Chem., 154 (1944) 691.

7 S.. S. Cohen, J. Immunol, 65 (1950) 475.

8 W. J. VAN Wagtendonk, J. Biol. Chem., 173 (1948) 691.

9 O. T. Avery, C. M. MacLeod and M. McCarty, /. Exptl. Med., 79

(1944) 137.
i<^ A. Boivin, a. Delaunay, R. Vendrely and Y. Lehoult, Experientia,
1 (1945) 334; 2 (1946) 139.

11 H. E. Alexander and G. Leidy, J. Exptl. Med., 93 (1951) 345.

12 M. U. Dianzani, Boll. ist. sieroterap. milan., 29 (1950) 161.

13 S. Zamenhof, G. Leidy, H. E. Alexander, P. L. FitzGerald and E.

Chargaff, Arch. Biochem. Biophys., 40 (1952) 50.
1^ D. E. Lea, Actions of Radiations on Living Cells, Cambridge Univer-
sity Press, London, 1947, p. 181.

15 A. Boivin, R. Vendrely and C. Vendrely, Compt. rend., 226 (1948)

1061.

16 R. Vendrely and C. Vendrely, Experientia, 4 (1948) 434; 5 (1949)

327.

17 A. E. MiRSKY and H. Ris, Nature, 163 (1949) 666.

18 C. Leuchtenberger, R. Vendrely and C. Vendrely, Proc. Natl. Acad.

Sci. U.S., 37 (1951) 33.

19 E. Chargaff, Experientia, 6 (1950) 201. [See Chapter 1 of this book.]

20 G. R. Wyatt, Nature, 166 (1950) 237.

21 E. VoLKiN, J. X. Khym and W. E. Cohn, J. Am. Chem. Soc, 73 (1951)

1533.

22 C. E. Carter, J. Am. Chem. Soc, 73 (1951) 1537.

23 E. Chargaff, J. Cellular Comp. Physiol, 38, Suppl. 1 (1951) 41.

24 E. Chargavf, S. Zamenhof and C. Green, Nature, 165 (1950) 756.

25 E. ChargaPf, Abstr. 5th Intern. Cancer Congr., Paris, 1950, p. 101.

26 E. Chargaff, S. Zamenhof, G. Brawerman and L. Kerin, /. Am.

Chem. Soc, 72 (1950) 3825.

27 E. Chargaff, R. Lipshitz, C. Green and M. E. Hodes, J. Biol. Chem.,

192 (1951) 223.

28 S. Zamenhof and E. Chargaff, J. Biol Chem., 178 (1948) 531; 187

(1950) 1.

38 NUCLEIC ACIDS AS CELL CONSTITUENTS

29 M. M. Daly, V. G. Allfrey and A. E. Mirsky, J. Gen. Physiol, 33

(1950) 497.

30 G. R. Wyatt, Biochem, J., 47 (1950) vii.

31 E. Chargaff, Cold Spring Harbor Symposia Quant. Biol, 12 (1947)

28.

CHAPTER 3

Deoxypentose Nucleoproteins and Their
Prosthetic Groups^

1 . INTRODUCTION

Some time in 1800, the German poet Kleist wrote a letter to a
friend in which he mentioned an idea that had occurred to him
when passing through an arched gateway: "Why, I thought, does
the vault not collapse, though entirely without support? It stands,
I replied, because all the stones want to fall down at the same
time." Similarly, one could say that the cell lives because all its
components desire to die simultaneously. They survive, in other
words, by getting into each other's way.

That branch of chemistry that is, among many other things,
entrusted with conducting a traffic survey of the Uving cell is
comparatively young. At the present time, it is true of cytochem-
istry, as essentially of many other sciences, that we can recognize
the very general and the very special; but what lies between — and
that is the greater part of the world — is completely obscure.

In trying to compare the state of biochemistry, for example,
of fifty years ago with that of today, one could use a metaphor.
In the old times, the knowledge of biology was perhaps similar
to what could be made out in a very large, very dark house.
Many objects could be more felt than seen with equal dimness,
once the eyes got used to the darkness; and scientists were con-
scious of the limiting conditions under which they worked. In
our time, however, a few very powerful and very narrow beams
of Hght have been thrown into a few corners of this dark house.

* Reprinted with permission from Symposia Soc. Exptl. Biol., 9 (1955)

32-47.

References p. 60

40 DEOXYPENTOSE NUCLEOPROTEINS AND PROSTHETIC GROUPS

and several things can be seen in a clarity and illumination that
almost distort their significance. But at the same time we have
lost our dark-adaptation; and since we all have a tendency to fol-
low the light, we have moved into these cosy corners, to the
detriment of the rest, which still is, by far, the major part of
nature. In pointing this out one runs the risk of being accused of
trying to spread darkness.

As concerns the subject of the present conference, it will
become clear that the biological significance of fibrous proteins
can best be defined in those instances to which I have referred
above as very special, for instance, in muscular contraction or in
blood coagulation. With regard to other conjugated proteins,
among which must be counted the nucleoproteins, it is their very
ubiquity that has made difficult the recognition of their function,
unless one is satisfied with the trite assignment of an essentially
mechanical task; but this amounts to no more than saying that
these proteins are, because they are. What could, of course, be
questioned is whether a purely teleological search for "functions"
is likely to prove adequate for the future development of our
science. While substances and reactions endure, interpretations
often perish with unseemly rapidity; and few things have as short
a half-life time as the premature assignment of a function.

One more introductory remark: When speaking of nucleo-
proteins as fibrous proteins, one must specify that this desig-
nation refers principally to the deoxypentose nucleoproteins,
though some fibrous pentose nucleoproteins are known. Further-
more, there is little indication that the carrier proteins themselves
are fibrous. This character is conferred on the conjugated protein
by the shape of the prosthetic group it contains, namely, the
deoxypentose nucleic acids. And, strictly speaking, why these
nucleic acids, or for that matter many other macromolecules of
natural occurrence, exist as fibers of so marked an anisometry
cannot yet be explained satisfactorily on chemical grounds.

In the following pages a brief survey of the chemistry of those
nucleoproteins that can be classified as fibrous, namely, the
deoxypentose nucleoproteins, will be followed by the consider-

DEOXYPENTOSE NUCLEOPROTEINS 41

ation of their prosthetic groups, the deoxypentose nucleic acids.
It will not be possible to do more than mention some of the cur-
rent problems. For a fuller account, compare a previous survey^.
Reviews of the chemistry of the pentose nucleoproteins have
recently been published^- ^.

2. DEOXYPENTOSE NUCLEOPROTEINS

a. Classification

The deoxynucleoproteins are usually defined as conjugated pro-
teins in which the deoxypentose nucleic acid, functioning as
a prosthetic group, is linked to the protein by electrostatic at-
traction or by secondary valence forces. It is, however, extremely
difficult to distinguish, in macromolecules, between these two
types of combination. It is, in addition, equally difficult to decide
whether the nucleoprotein isolated in a given case really pre-
existed in the cell or whether it was the result of a fortuitous
combination of solutes present in the same extract. A genuine
nucleoprotein, which in some of its properties ought to differ
from a mere mixture of its components, must be regarded as a
geometrically unique combination of two giant polyampholytes.
It is only recently that attempts have been made to define more
sharply the distinction between a nucleoprotein and a protein
nucleate artifact^.

Since the deoxypentose nucleic acid fraction of a given nucleus
is composed of a large number of differently constituted individu-
als^"'^, a nucleoprotamine or a nucleohistone must comprise many
different conjugated proteins. A search of the literature will lead
to the conclusion that the number of genuine deoxypentose
nucleoproteins isolated until now is small. It can, moreover, not
yet be stated whether the major part, or all, of the deoxypentose
nucleic acids present in the cell occurs in the form of nucleo-
proteins.

As regards the protein moiety, three principal groups of deoxy-
nucleoproteins may be distinguished: (1) the nucleoprotamines
occurring in ripe spermatozoa of many fish genera; (2) the

References p. 60

42 DEOXYPENTOSE NUCLEOPROTEINS AND PROSTHETIC GROUPS

nucleohistones mostly investigated in the form of calf thymus
nucleohistone; (3) the nucleoproteins in the proper sense of this
term in which the nucleic acid is combined with a protein lacking
the basic properties of the protamines and histones^. We are,
however, very far from a satisfactory classification of the types
of deoxynucleoprotein occurring in cell nuclei, especially with
regard to changes in the composition of the protein moiety taking
place during development and in the various phases of the mitotic
cycle.

b. Isolation

The principal methods for the preparation of nucleoprota-
mines and nucleohistones, i.e., complexes in which the protein
partner carries a positive charge, are based on (a) extraction with
strong salt solutions^; (b) extraction with solutions of low ionic
strength^^. The first-mentioned procedure, widely used as the
first step in the isolation of deoxypentose nucleic acids, is ac-
companied by an, at least partial, dissociation and may lead to
artifacts. It has, on the other hand, the advantage that the activity
of deoxyribonucleases is suppressed in strong salt solutions. By
extraction with solutions of low ionic strength the exposure of
the conjugated protein to dissociating conditions in the course of
its preparation is avoided. There exists, however, the danger of
a partial enzymic degradation of the nucleic acid brought about
by the release of nucleases. As examples of the isolation of
nucleohistone at a low electrolyte concentration the studies on
calf thymus nucleohistone in Refs. 4 and 11 may be cited; the
isolation of a nucleoprotamine from sea-urchin sperm is discussed
in Ref. 12. Preparations afforded by extraction with strong
salt solutions are described, for instance, in Ref. 13. The isolation
from avian tubercle bacilli of a nucleoprotein with entirely dif-
ferent solubility properties has been reported^.

c. Properties

A real decision on the native state or the attributes of intact-
ness of a nucleoprotein will hardly be possible before suitable

DEOXYPENTOSE NUCLEOPROTEINS 43

biological test procedures have been found. The nucleoproteins
of calf thymus and of fish sperm remain the only readily acces-
sible deoxynucleoproteins, and most of our knowledge is derived
from the study of these compounds. Nucleohistone is soluble at
a very low ionic strength. As the electrolyte concentration in-
creases, the solubility first drops rapidly, reaching a minimum
in isotonic solutions, and then rises gradually. Though variations
are found, freshly prepared nucleohistone is, in general, soluble
in NaCl solutions stronger than 0.7 or 0.8 M. Small amounts of
phosphorylated impurities can be extracted from nucleohistone
by 0.02-0.15 M NaCl solutions; but they usually lack the typical
nucleic acid spectrum (see Table 16). The solubility properties
of nucleoprotamine seem to be essentially similar.

The absorption spectrum in the ultraviolet of a nucleoprotein
is, in general, identical with that of its nucleic acid moiety (maxi-
mum around 260 mu). No depression of the extinction coef-
ficient of the nucleic acid component is noticeable, nor does the
extent of extinction change materially under conditions of ahnost
complete dissociation (compare Table 16). This is of interest,
since the extinction of a nucleoprotein could have been expected
to be less than that of the free nucleic acid, if the purines and
pyrimidines of the latter were involved in the architecture of the
intact conjugated protein in such a manner as to bring about the
suppression of some of the chromophores^^.

The nucleic acid content of different nucleoproteins varies
from about 35 to 60% of the dry weight. In most nucleohistone
preparations it is around 50%. In trout and herring nucleopro-
tamines a phosphorus to arginine ratio of 1 has been found^^.

Although there exists considerable evidence that deoxypentose
nucleoproteins, or at any rate thenucleoprotamines and nucleo-
histones, occur in strong salt solutions in a largely dissociated
state, in which a far-reaching separation between nucleic acid
and protein has taken place, it would probably be wrong to
regard such solutions as representing no more than a mixture of
these components.

References p. 60

44 DEOXYPENTOSE NUCLEOPROTEINS AND PROSTHETIC GROUPS

I

o

I

•Si <3 *

11

OS O O
'-I »o to

On

r^ o o

o6 o o

r-l O >0

Tf O O

^ o o

r-^ o o

'-H >o O

On r-^

so"

so O O

^ o o

\0 so

t^ o o

^ o «o

'^ vq

vo"

o o
>o o

oo O O

l-H o «o

'I ^

c

8.S

I »o

On

so o O

'— I O »0
so -^^
T^*" so"

'oo o
O C

(^ > W

-3 o

o ^

^ a

o '^

tH 4)
O

o

s

u

o
s a

£2 O

S 6

^t3 (S

.2 <L)

(1) ^

3 O
WJ O

DEOXYPENTOSE NUCLEOPROTEINS

45

d. Dissociation and cleavage

Before the discussion of the nucleic acids proper a few words
should be said about the cleavage and dissociation of the nucleo-
proteins, especially the nucleohistones and nucleoprotamines. The
cleaving effect of high electrolyte concentrations was first ob-
served by Bangle ; it is of great importance for the isolation of
reasonably intact deoxypentose nucleic acids.

A systematic study in our laboratory of the behavior of
nucleoproteins towards electrolytes has opened a promising ap-
proach to the problem of the fractionation of deoxypentose
nucleic acids. When calf thymus nucleohistone is exposed to
increasing NaCl concentrations under conditions permitting the
removal of both undissociated nucleoprotein and liberated
protein, increasing quantities of sodium nucleate will be found
in solution^' ^^. Similar results are obtained when the time of
contact with the salt solution is varied rather than the concen-
tration of the latter^^. When the removal of the non-dissociated
portion is brought about in the presence of a denaturing phase,

Fig. 2. Fractional dissociation of calf thymus nucleohistone (curve A)
and of wheat germ nucleoprotein (curve B) in the presence of a CHCI3
phase. The quantities of nucleic acid P (as percentage of P in the starting
material) extracted at the NaCl concentrations indicated on the abscissa
are plotted as the ordinate.

References p. 60

46 DEOXYPENTOSE NUCLEOPROTEINS AND PROSTHETIC GROUPS

0.4 0.6 0.8 1.0 2.6
MNaCI

Fig. 3. Fractional dissociation, in the presence of a CHCI3 phase, of
artificial complexes between the deoxyribonucleic acids of calf thymus
(curves A and B) and pig liver (curve C) with globin (curve A) or histone
(curves B and C). Co-ordinates as in Fig. 2. (Taken from Crampton

et fl/.6.)

such as chloroform-pentanol, a large number of distinct nucleic
acid fractions can be separated, as will be discussed below. The
fractional dissociation of calf thymus nucleohistone and of the
nucleoprotein of wheat germ, based on previous work^ and on
unpublished experiments with Miss Rakoma Lipshitz, is shown
in Fig. 2. These are curves taken from single experiments; but
the typical shape can be reproduced easily.

e. Behavior of artifacts

The behavior of artificially prepared complexes of nucleic acids
with positively charged substances has also been studied. Such
nucleate compounds have been prepared not only with histone
and globin^, but also with synthetic polyelectrolytes, e.g., polyly-
sine or polyvinylamine^^. The fractional dissociation of deoxy-
ribonucleate complexes with globin or histone is illustrated in

DEOXYPENTOSE NUCLEOPROTEINS

47

Fig. 3. In general, globin nucleates appear to undergo dissociation
more readily than histone nucleates. The relative proportions of
histone and nucleic acid and the state of polymerization of the
latter are of great importance for the properties of the artifacts
(Fig. 4). It can be seen that the ease of dissociation of the com-
plexes increases with the relative amount of nucleic acid present
and decreases very markedly with the viscosity of the nucleic
acid. It may be mentioned that under appropriate conditions
nucleohistone is rendered soluble at a low ionic strength by the
addition of large amounts of degraded deoxyribonucleic acid^*^.

0.5 0.8 1.0 2.6 0.5

Af NaCI

0.8 1.0 2.6

Fig. 4. Successive extraction with NaCl solutions, in the presence of a
CHCI3 phase, of artificial histone complexes with calf thymus deoxy-
ribonucleic acid. In (a) the amount of histone varied, in (b) the degree of
polymerization of the nucleic acid. The weight ratio of histone to nucleic
acid P was 23 (curve 1), 18 (curve 2), 14 (curve 3), 4.5 (curve 4), 17
(curves 5-7). The viscosities, as r]^^ (P), i.e., the specific viscosity divided
by the molarity of the solution with respect to P, were: curve 5, 1000
(intact preparation); curve 6, 420 (0.1% solution heated at 75° for 5 min);
curve 7, 240 (0.1% solution heated at 75° for 40 min). Co-ordinates as
in Fig. 2. (Taken from Crampton et ciI.^k)

The cationic species used also influences the progress of the dis-
sociation. The remarkable effect of an admixture of as little as
0.06 M MgClo to the NaCl solutions is shown in Fig. 5. When

References p. 60

48 DEOXYPENTOSE NUCLEOPROTEINS AND PROSTHETIC GROUPS

solutions of MgCb alone are employed, the liberation of nucleic
acid from its complex with histone occurs at even lower salt
concentrations, as much as 70% of the nucleic acid P being
detached by 0.3 M MgClo and almost the entire amount by a
0.6 M solution.

Fig. 5. Effect of magnesium ions on the fractional dissociation, in the
presence of a CHCI3 phase, of artificial histone complexes with calf
thymus deoxyribonucleic acid. The weight ratio of histone to nucleic acid
P was 17. The successive extractions were performed with NaCl solutions
of the indicated molarity in the absence (curve 1) and in the presence of
0.06 M MgCl2 (curve 2). Co-ordinates as in Fig. 2. (Taken from Cramp-
ton et al.^.)

In these experiments the histone nucleate complex was formed
at the lowest NaCl concentration employed for fractionation. The
conditions under which the complex is formed are, however, not
without influence on its subsequent behavior. This is brought
out by the experiments assembled in Fig. 6. Two differently
treated nucleohistone preparations are compared in curves 1 and
2. Before being precipitated by dialysis against 0.15 M NaCl, one
portion was in contact with 3 M NaCl (curve 2), the other portion
remained in distilled water (curve 1). The precipitates then were
extracted successively with salt solutions of increasing strength
in the absence of a denaturing chloroform phase. Similarly,
artificial histone nucleates were formed in the absence of salt

DEOXYPENTOSE NUCLEOPROTEINS

49

100

A

J^ 5

90

71'

80

^^

o.

f «

< 70

i .i-^

^ 60

■1 ^-^^

•i 50

w

f-

?■ If.

-hi'

i II

^ 30

!• ■ 1

• /■

n/

20

-^v

10

,

^« , , , , .

0.5 1.0 1.5 2.0 2.5 3.0
MNaCI

Fig. 6. Successive extraction with NaCl solutions of increasing strength,
in the absence (curves 1-4) and in the presence (curves 5 and 6) of a
CHCI3 phase, of calf thymus nucleohistone (curves 1 and 2) or of artificial
histone complexes (curves 3-6) with calf thymus deoxyribonucleic acid.
Compare the text for experimental details. Co-ordinates as in Fig. 2.
(Taken from unpublished experiments with Miss Rakoma Lipshitz.)

(curve 3) and in 3 M NaCl (curve 4) and extracted successively.
In another series, similarly produced complexes (curve 5 without,
curve 6 with 3 M NaCl) were extracted stepwise in the presence
of CHCls-pentanol. These experiments permit a comparison of
the behavior of nucleohistone and of histone nucleate com-
plexes, prepared at a low electrolyte concentration (curves 1 and
3), towards extraction with salt solutions in the absence of
chloroform with that of preparations that had initially been in
contact with a 3 M NaCl solution (curves 2 and 4). The effect
of chloroform on the rate of appearance of nucleic acid P in the
extracts is demonstrated by a comparison of curve 3 with 6 and
of curve 4 with 5. Certain characteristic differences between
nucleohistone and histone nucleate are brought out when curve
1 is compared with 3 and curve 2 with 4.

References p. 60

50 DEOXYPENTOSE NUCLEOPROTEINS AND PROSTHETIC GROUPS

/. Isolation of deoxypentose nucleic acids

The first step in the preparation of a deoxypentose nucleic
acid is, irrespective of whether this is done consciously or not,
the isolation of a nucleoprotein. Once the nucleoprotein or a
mixture of nucleic acid and protein has been extracted, the
removal of the protein is carried out by treatment with chloro-
form, saturation with sodium chloride, or with the aid of an
anionic detergent. The several methods have been reviewed
critically in a recent survey^.

3. DEOXYPENTOSE NUCLEIC ACIDS

One of the surest signs that one has reached a stage of truth is
that its affirmation begins to bore the audience. I shall, therefore,
limit myself to a brief consideration of the principal conclusions
regarding the composition and structure of nucleic acids, many
of which already had been formulated six or seven years ago^^
i.e., in what were the early days of modern nucleic acid research.

a. Is there more than one deoxypentose nucleic acid in nature?

All evidence points to the existence of a very large number
of differently constituted nucleic acids; a number probably ex-
ceeding by far what can be revealed by constituent analysis. For
it should be remembered that, while differences in nucleotide
composition signify divergences in nucleotide sequence, an iden-
tical composition does not prove an identical sequence. An early
estimate of the possible number of isomers having the com-
position of calf thymus deoxyribonucleic acid, which assumed
a chain of 2,500 nucleotides, gave the figure of lO^^^^ Although
the differentiation between nucleic acids on the basis of dif-
ferences in the contents of their nitrogenous constituents is a
relatively crude expedient, since it does not permit a distinction
between sequence variations that are not accompanied by a
change in total composition, it can be shown in many cases that
nucleic acids of different origin differ in composition (compare

DEOXYPENTOSE NUCLEIC ACIDS 51

the survey in Ref. 1). To mention only the most extreme examples:
the ratio of the sum of adenine and thymine to that of guanine
and cytosine is about 1 .9 in the deoxyribonucleic acid of Echino-
cardium cordatum, but about 0.4 in that of avian tubercle bacilli.
It is, in fact, possible to make a gross distinction between three
types of deoxypentose nucleic acid: (a) the AT type, in which
adenine and thymine are the major constituents; (b) an inter-
mediate type, in which all bases are present in nearly equimolar
proportions (found only in Escherichia coli); and (c) the GC
type, in which guanine and cytosine predominate. While, therefore,
the question whether there exists more than one deoxypentose
nucleic acid must be answered in the affirmative, a decision
whether different tissues of the same host yield entirely iden-
tical nucleic acid specimens is more difficult. For the moment,
it would appear that differences in the composition of prepara-
tions from different organs, if they exist at all, are of very
doubtful significance.

b. Do there exist recognizable repeating units?

The tetranucleotide hypothesis is incorrect. One must conclude
that there is no subunit of recognizably recurrent structure larger
than a mononucleotide. If repeating polynucleotide units occur
in the nucleic acid chain, we have no way as yet of demonstrating
their existence. The nucleic acids must be regarded as complicated
and intricate high polymers of undetermined and largely ar-
rhythmic nucleotide sequence.

c. Regularities

The mistakes of the present are made by trying to avoid the
mistakes of the past. Though the danger of the premature im-
position of regularities is well exemplified by the tetranucleotide
hypothesis, it would be an error not to draw generahzing con-
clusions from experimentally established facts. The apodictic
manner of the following enumeration is chosen to conserve space.
Despite wide divergences in their nucleotide composition and
sequence all deoxyribonucleic acids appear to be characterized

References p. 60

52 DEOXYPENTOSE NUCLEOPROTEINS AND PROSTHETIC GROUPS

by the following regularities^' ^i* ^^i (a) The sum of the purine
nucleotides equals that of the pyrimidine nucleotides, (b) The
molar ratio of adenine to thymine equals 1. (c) The molar ratio
of guanine to cytosine (+ methylcytosine) equals 1. (d) The
number of 6-amino groups (adenine + cytosine + methylcyto-
sine) equals that of 6-keto groups (guanine + thymine). The last-
mentioned regularity, and in some cases also the others, but with
uracil taking the place of thymine, applies also to certain pentose
nucleoproteins^^.

It is, in fact, because of these relationships that it is possible
to state that, for instance, in the deoxypentose nucleic acids of
wheat germ 5 -methylcytosine substitutes for a part of the
cytosine^^' 2^ or that in certain bacteriophage nucleic acids 5-
hydroxymethylcytosine replaces cytosine entirely^^. It is, in-
cidentally, remarkable that so far no purine satellites have been
discovered.

But nature is inexorably approximate; and one could suspect
that occasional deviations from fit are more significant than the
fit itself. This may still turn out to be true of such giant molecules
as the nucleic acids, but I do not believe it will. In any event, I
should like to cite here a recent survey in which the results of
all published analyses, carried out in several laboratories, of
deoxypentose nucleic acid specimens from about fifty different
genera were tabulated^ The average values of the molar relation-
ships that bring out the complementariness principles mentioned
before were as follows.

Adenine to thymine 1.009

Guanine to cytosine (or its analogues) 1.001

Purines to pyrimidines 1.000

6-Amino to 6-keto groups 1.008

If it is considered that the nucleic acids varied very widely in
their nucleotide composition (ratios of adenine + thymine to
guanine + cytosine between 2 and 0.4), the agreement of the
1 : 1 relationships is most impressive, especially in view of the
diversity of sources, preparations and procedures. No authentic
case of an appreciable deviation has yet come to my attention.

DEOXYPENTOSE NUCLEIC ACIDS 53

d. Molecular structure

Deoxypentose nucleic acids are polynucleotide chains that are
held together by phosphoric acid diester bridges between the
deoxypentose molecules. Our knowledge of the sugar found in
different preparations rests mostly on chromatographic evidence
which has up to now yielded no indication of the presence of
any but 2-deoxyribose^. The position of the phosphate bridges
has been studied in some detail^^ only in the deoxyribonucleic
acid of calf thymus in which it appears to be 3', 5'. No deter-
minations of end groups have as yet been possible nor has the
existence of a terminal phosphate group been demonstrated.
Strictly speaking, it cannot yet be asserted that the deoxypentose
nucleic acids occur as open and not as closed chains. It is,
however, probably too early to worry about the tail before we
can describe the animal.

The excellent X-ray photographs obtained at King's Col-
\QgQ27, 28 ^jj^ ^jjg discovery of the general unity relationships men-
tioned before have formed the basis of an interesting structural
proposal^®' ^^. This model, which in the recent past has been
unveiled repeatedly and sometimes very amusingly*, assumes a
heUcal dyad in which two coiled strands are held together by a
specific pairing of the nitrogenous constituents showing the unity
relationships mentioned before. This conception has several at-
tractive features; it makes good use of some of the experimentally
established facts and is, on the whole, not incompatible with the
X-ray^2 and part of the chemical evidence. But between what is
plausible and what is true in the natural sciences there lies an
ocean which has seldom been crossed.

The invention of models is an essentially sterile undertaking
unless two things are thereby accomplished: (a) a satisfying ex-
planation of all known and sometimes seemingly contradictory

* "This pairing is likely to be so fundamental for biology that I can-
not help wondering whether some day an enthusiastic scientist will
christen his newborn twins Adenine and Thymine!" (Ref. 31). No doubt,
this too will happen; but is "enthusiastic" the correct epithet?

References p. 60

54 DEOXYPENTOSE NUCLEOPROTEINS AND PROSTHETIC GROUPS

facts; (b) an impulse to further and crucial experimentation.
Therefore, before this structural hypothesis is elevated to the
dignity of a natural law, a few remarks may be in order. The
model provides a satisfactory explanation, though it probably is
not the only possible one, of the unity relationships observed ex-
perimentally. Whether, in its detailed form, it does more than
describe the structure of that portion of the processed nucleic
acid preparation from which the diffraction patterns were ob-
tained, remains, however, to be estabhshed. More work on genuine
nucleoproteins, especially on compounds with proteins other than
histone or protamine, e.g., of the type found in tubercle bacilli^,
is urgently required. The existence of two discrete and com-
plementary chains has not been demonstrated; no deoxypen-
tose nucleic acid preparation has shown noticeable deviations
from the 1 : 1 ratios, not even for the ratio of purines to pyrimi-
dines.

It is, however, the second part of the hypothesis, namely, the
suggestion of a copying mechanism for the mutual reproduction
of the two half-chains, that is more difficult to reconcile with
some of what is known. It is assumed that a specific pairing is
brought about by hydrogen-bond formation between the nitrog-
enous bases of the two chains in such a manner that a base
having a keto group at the 6 position (guanine, thymine) can
bond only with a base carrying a 6-amino group (adenine,
cytosine), and that a purine always bonds with a pyrimidine, thus
maintaining a constant distance between the two chains. It will
be observed that the model is quite strict as regards certain
restrictions governing the size of the partners (purine is linked
to pyrimidine) and the position of the hydrogen bonds (purine- 1
to pyrimidine- 1 ; purine-6 to pyrimidine-6), but does not require
the absence or presence of substituents in other positions. Thus,
5-methylcytosine or 5-hydroxymethylcytosine should be able to
take the place of cytosine, and uracil and 5-methyluracil (thy-
mine) should be similarly exchangeable, both in their positions on
the ''positive" or "negative" or during the automatic replication
of the chains. That the latter replacement does not occur despite

DEOXYPENTOSE NUCLEIC ACIDS 55

the availability of uracil is amply demonstrated by the absence
of uracil from deoxypentose nucleic acids.

As regards the 5-substituted cytosine derivatives, the situation
is even more disturbing. One must surely assume that both
cytosine and 5-methylcytosine are available to the cell during the
formation of certain deoxypentose nucleic acids, since both
pyrimidines are incorporated in what appears to be constant
proportions. For instance, three different groups of workers,
analyzing several independently prepared specimens of the deoxy-
pentose nucleic acid of wheat germ, found the following figures
for the 5-methylcytosine content (as moles per 100 gram-atoms
of phosphorus)^^'^^'^*: 5.8, 6.0, 5.7. Similarly, the corresponding
proportion of 5-methylcytosine in the nucleates from bovine tis-
sues is around 1.3 (see the survey in Ref. 1). This constancy of
incorporation must be due either to a specific selection operating
during the reproduction of the nucleic acids or, much less likely,
to an unexpectedly steady level in the supply of these pyrimidines
or their precursors in the cell. In the latter case, the degree of
incorporation of a pyrimidine and its 5-substituted analogue
would be governed by the laws of probability and there could be
no preferential accumulation of the satellite in certain portions
of the polynucleotide chain or in one or the other of the fractions
into which a total deoxypentose nucleic acid preparation can be
separated. But the available evidence suggests that 5-methyl-
cytosine does not replace cytosine at random and that the scheme
is either incorrect or incomplete in some essential features. This
evidence is twofold.

(1) During the fractionation of calf thymus deoxyribonucleic
acid, which will be mentioned in the concluding section of this
article, very significant divergences in the concentration of 5-
methylcytosine in the several fractions were found (compare
Table 17 which is taken from Ref. 5). Similar results, soon to
be published, have been obtained with fractions of the deoxy-
pentose nucleic acid of wheat germ^^.

(2) The action of pancreatic deoxyribonuclease on deoxy-^
ribonucleic acids results in the production of a considerable;

References p. 60

56 DEOXYPENTOSE NUCLEOPROTEINS AND PROSTHETIC GROUPS

TABLE 17

5-METHYLCYTOSINE DISTRIBUTION IN FRACTIONS OF CALF THYMUS
SODIUM DEOXYRIBONUCLEATE*

Fraction

24.0

29.0

28.7

23.4

19.8

21.3

2.7

1.2

0

8.9

24.2

_

8.7

16.5

—

NaCl molarity in extraction of nucleohistone gel 0.40 0.45 1.7 2.6
Corrected mean proportions:

Thymine 24.2

Cytosine 23.9

5-Methylcytosine 1.9
Molar ratio:

Thymine to 5-methylcytosine 12.7

Cytosine to 5-methylcytosine 12.6

* The mean proportions of each constituent have been corrected on the
assumption that one-half of the nucleic acid P is contributed by the
pyrimidine nucleotides. In one hydrolysis experiment with fraction 4 a
minute amount of 5-methylcytosine (about 0.6 mole) was found.

proportion of dinucleotides which have recently been separated
and analyzed^^. Unless the assumption is made that this enzyme
distinguishes specifically between cytosine and its 5-methyl
derivative, it is significant that the proportions of the various
dinucleotides are far from what randomness of incorporation
would predict. For instance, the frequency of methylcytosine
being Unked to guanine in a dinucleotide as compared with that
of cytosine being so attached is 20 times as high in calf thymus
nucleic acid, and 1 3 times as high in the wheat germ preparation,
than would have been expected for non-selective incorporation.
The unexpectedly high proportion of 5-methyldeoxycytidylic acid
that is linked to deoxyguanyhc acid^^-^^ makes one, in fact,
wonder whether it is not the adjoining nucleotide rather than the
one opposite that directs the incorporation.

Two other findings could also be cited, though their detailed
discussion would take us too far, namely, (a) the incorporation,
from the medium, of 5-bromouracil or 5-iodouracil in the deoxy-
pentose nucleic acids of certain deficient bacteria or of bacterio-
phages in replacement of part of the thymine^^- ^^j and (b) the

DEOXYPENTOSE NUCLEIC ACIDS 57

complete replacement of cytosine by 5-hydroxymethylcytosine in
certain phage nucleic acids-^. The latter synthesis, at any rate,
may be presumed to occur in spite of the presence of the anal-
ogous pyrimidine, cytosine.

e. Is there more than one deoxypentose nucleic acid in a cell?

An answer to this question appeared very difficult for a long
time. On the one hand, many different nucleic acid preparations
from the same genus showed a remarkable constancy of com-
position. On the other hand, it appeared rather unlikely on
general grounds that a cell, which is able to produce so many
different proteins, should give rise to no more than one deoxy-
pentose nucleic acid. But the task of separating a mixture of
closely related polyampholytes of a very high molecular weight
appeared hopeless; and one could ask whether a nucleic acid
preparation comprising a very large number of individuals should
not be treated as one compound. This is, indeed, a problem that
leads directly into the epistemology of chemical identity; and
many have found that it is an ungrateful task to spend one's life
at the frontier of knowledge without the proper passport.

Later, a process for the fractionation of deoxypentose nucleic
acids was discovered which is based on the fractional disso-
ciation of a nucleoprotein, as discussed above (see also Fig. 2).
The procedure, which led to the separation of many deoxypentose
nucleate preparations into a series of fractions of divergent
purine and pyrimidine contents, was first applied to calf thymus
nucleohistone^ and later extended to the fractionation of other
nucleic acids through their complexes with histone, globin^ or
polylysine^^. Subsequently, a similar arrangement was described
in a preliminary form in which the fractionation was carried out
by means of a histone-kieselguhr column^.

The gradual extraction of nucleohistone or of an artificially
prepared protein nucleate with salt solutions of increasing strength
in the presence of a denaturing phase, such as chloroform-pen-
tanol, was found to yield a series of nucleic acid fractions with
diminishing concentrations of guanine and cytosine and rising

References p. 60

58 DEOXYPENTOSE NUCLEOPROTEINS AND PROSTHETIC GROUPS

100

■ Adenine + thymine
13 Guanine + cytosine
m Guanine + hydroxymethyl cytosine
T 1 2 3 4 T 1 2 345

50

T 1 2 3456

0 50 100 0 50 100 0 50 100 0 50 100
Calf thymus Pig liver Human spleen Coliphage 76

Fig, 7. Composition of fractions (in moles per 100 gram-atoms of phos-
phorus) prepared by the fractional dissociation of artificial histone com-
plexes with the indicated sodium deoxypentose nucleate preparations.
Each block represents a fractionation run; the consecutively numbered
bars within each block represent individual fractions, with the NaCl con-
centration of the solutions employed for extraction rising from left to
right. The first column (T) indicates the average composition of the total
deoxypentose nucleic acid preparations. The abscissa indicates the propor-
tion of original nucleic acid P recovered in each fraction. (Taken from

Chargaffi.)

concentrations of adenine and thymine. What should, however,
be stressed is that in all fractions the equimolarity of each pair
of constituents and of total purines and pyrimidines is fully main-
tained. A selection of such fractionation experiments is shown in
Fig. 7.

These findings, fully discussed in the publications cited above,
suggest that the deoxypentose nucleic acid of a given cell is
composed of a very large number of differently constituted in-
dividuals and that it is possible to achieve the resolution of this
entire spectrum of structural gradations into several distinct
"bands", in which all the regularities characteristic of the entire
nucleic acid are maintained.

DEOXYPENTOSE NUCLEIC ACIDS 59

There is little doubt that these fractions should not be con-
sidered as individuals. If we are dealing with a continuous
spectrum, as I have suggested, it may, in fact, be impossible to
isolate what could be considered as a rigorously homogeneous
specimen; it is not easy even to imagine a proper criterion for
such a decision. Conceivably, no two nucleic acid molecules,
within the same nucleus, are entirely identical. But here we clash
against the definition of a macromolecule; and seeking refuge in
statistics, the science of reconcihng the parallels, we may con-
clude that it is hard to fight on a front where friend and enemy
wear the same uniform.

References p. 60

60 DEOXYPENTOSE NUCLEOPROTEINS AND PROSTHETIC GROUPS
REFERENCES

1 E. Chargaff, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. I, Academic Press, New York,
1955, p. 307.

2 R. Markham and J. D. Smith, in H. Neurath and K. Bailey (Eds.),

The Proteins, Vol. 2, Academic Press, New York, 1954, p. 1.

3 B. Magasanik, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. I, Academic Press, New York,

1955, p. 373.
* C. F. Crampton, R. Lipshitz and E. Chargaff, 7. Biol. Chem., 206

(1954) 499.
5 E. Chargaff, C. F. Crampton and R. Lipshitz, Nature, 111 (1953)

289.
^ C. F. Crampton, R. Lipshitz and E. Chargaff, 7. Biol. Chem., 211

(1954) 125.

7 G. L. Brown and M. Watson, Nature, 111 (1953) 339.

8 E. Chargaff and H. F. Saidel, 7. Biol. Chem., Ill (1949) 417.

9 A. E. MiRSKY and a. W. Pollister, Proc. Natl. Acad. Sci. U.S., 28

(1942) 344.

10 E. Hammarsten, Biochem. Z., 144 (1924) 383.

11 J. A. V. Butler, P. F. Davison, D. W. F. James and K. V. Shooter,

Biochim. Biophys. Acta, 13 (1954) 224.

12 M. H. Bernstein and D. Mazia, Biochim. Biophys. Acta, 10 (1953)

600; 11 (1953) 59.

13 A. W. Pollister and A. E. Mirsky, 7. Gen. Physiol., 30 (1946) 101.

14 B. Magasanik and E. Chargaff, Biochim. Biophys. Acta, 1 (1951) 396.

15 K. Felix, H. Fischer, A. Krekels and R. Mohr, Z. physiol. Chem.,

Hoppe-Seyler's, ISl (1951) 224.

16 L Bang, Beitr. chem. Physiol. Pathol, 4 (1904) 331; 5 (1904) 317.

17 E. Chargaff, in Some Conjugated Proteins, New Brunswick, 1953,

p. 36.

18 J. A. Lucy and J. A. V. Butler, Nature, 174 (1954) 32.

19 P. Spitnik, R. Lipshitz and E. Chargaff, 7. Biol. Chem., 215 (1955)

765.

20 C. F. Crampton and E. Chargaff, 7. Biol. Chem., 226 (1956) 157.

21 E. Chargaff, Experientia, 6 (1950) 201. [See Chapter 1 of this book.]

22 E. Chargaff, Federation Proc, 10 (1951) 654. [See Chapter 2 of this

book.]

23 D. Elson and E. Chargaff, Nature, 173 (1954) 1037.

24 G. R. Wyatt, Biochem. J., 48 (1951) 584.

25 G. R. Wyatt and S. S. Cohen, Biochem. J., 55 (1953) 774.

26 D. M. Brown and A. R. Todd, in E. Chargaff and J. N. Davidson

(Eds.), The Nucleic Acids: Chemistry and Biology, Vol I, Academic
Press, New York, 1955, p. 409.

27 M. H. F. WiLKiNS, A. R. Stokes and H. R. Wilson, Nature, 171 (1953)

738.

REFERENCES 61

28 R. E. Franklin and R. G. Gosling, Nature, 171 (1953) 740.

29 J. D. Watson and R H. C. Crick, Nature, 171 (1953) 737, 964; Cold

Spring Harbor Symposia Quant. Biol, 18 (1953) 123.

30 F. H. C. Crick and J. D. Watson, Proc. Roy. Soc. {London) A, 223

(1954) 80.

31 F. H. C. Crick, Sci. American, 191, No. 4 (1954) 54.

32 M. H. F. WiLKiNS, W. E. Seeds, A. R. Stokes and H. R. Wilson,

Nature, 111 (1953) 759.

33 G. Brawerman and E. Chargaff, /. Am. Chem. Soc, 73 (1951) 4052.

34 S. G. Laland, W. G. Overend and M. Webb, 7. Chem. Soc, (1952)

3224.

35 R. Lipshitz and E. Chargaff, Biochim. Biophys. Acta, 19 (1956) 256.

36 R. L. SiNSHEiMER, J. Biol. Chem., 208 (1954) 445.

37 J. D. Smith and R. Markham, Nature, 170 (1952) 120.

38 D. B. Dunn and J. D. Smith, Nature, 174 (1954) 305.

39 S. Zamenhof and G. Griboff, Nature, 174 (1954) 306.

CHAPTER 4

The Chemistry and Function of Nucleoproteins
and Nucleic Acids'^

1. INTRODUCTION

The progress in our knowledge of the chemistry and biology of
the nucleic acids has been truly enormous in the last seven or
eight years. As is customary in the natural sciences, the first con-
ception and formulation of the problems soon was followed by
the invention of the technical means required for their solution;
and thereupon ensued an avalanche of papers of which some
were important and more were trivial.

As there has appeared recently a very comprehensive treatise
on the nucleic acids, which attempts to take stock of our present
knowledge^, there would be little sense in my engaging in a
swimming contest with a leviathan. Hence, instead of citing much
literature I shall refer, wherever possible, to the pertinent chap-
ters of this book.

In the biological sciences, as in many others, we are faced at
present with a silent struggle between those believing in first
causes and looking for a teleological foundation of all occur-
rences and those who think that macrocosm and microcosm are
governed by laws of probability. The clash between these two
schools has found an epigrammatic expression in the anecdote
relating a remark that Einstein is said to have made about the
uncertainty principle. "But the Good Lord does not throw
dice!" — a statement, I should hke to add parenthetically, that
itself severely restricts His omnipotence, since there is no earthly

* Reprinted with permission from Rend. ist. lombardo sci. Pt. I, 89
(1955) 101-115.

NUCLEOPROTEINS 63

or heavenly reason why the Good Lord should not throw dice if
He so desires.

Be that as it may, there is a strong tendency in all of us to
establish some sort of hierarchy and to consider the body or the
cell as operating under a chain of commands. Therefore, a bright
graduate student will be able to tell us what some of the
profoundest intellects in biology would have shuddered to con-
template, namely, "DNA makes RNA, and RNA makes protein".
There is, of course, little ground for such a brash statement;
but this is a field in which the winds of fashion blow strong,
and they have been blowing from this direction for quite some
time. The moment will soon have arrived when it will be pos-
sible to consider the correctness of this generalization.

In the following, I shall be able only to touch upon a few
aspects of a very large theme, as they may occur to a speaker
who attempts to give a comprehensive view, but has little time
in which to do it.

2. NUCLEOPROTEINS

Our knowledge of the nucleoproteins has gone through a few,
rather curious stages-. At the very beginning, i.e., in the times
of Miescher, Hoppe-Seyler, Kossel, etc., there existed little doubt
that the nucleic acids were, in the cell, attached to proteins and
that these nucleoproteins were more than a mere mixture of the
partners. The very scanty experimental means then available did
not permit to go much further. These ancient masters, who
combined a great respect for nature with a great reticence in its
interpretation, were succeeded by much more impetuous and
less cautious generations; and the.nucleoproteins often were con-
sidered as obnoxious obstacles to the purification of the all-
important proteins. It is surprising how little work has been done
on nucleoproteins in the recent past. Even when the term nucleo-
protein appears in the title of a publication, the latter will often
be found to deal either with the nucleic acids themselves or with
so-called model compounds.

References p. 75

64 NUCLEOPROTEINS AND ISfUCLEIC ACIDS

The nucleoproteins are defined as conjugated proteins in which
the nucleic acid functions as the prosthetic group and is Unked
to the protein by electrostatic attraction or by secondary valence
forces, and sometimes by both. It is, of course, necessary to make
a clear distinction between the deoxypentose and the pentose
nucleoproteins. In the deoxypentose nucleoproteins^ the electro-
static contribution often is very pronounced, owing to the fact
that the deoxyribonucleic acids frequently are found in as-
sociation with proteins carrying strong positive charges, such as
the protamines or histones. Such complexes with basic proteins
are dissociated comparatively easily; and it is, for this reason,
sometimes difficult to decide whether such a nucleoprotein really
existed in the living cell or whether its isolation was the result of
a fortuitous combination of solutes that happened to be present
in the same extract.

In decided contrast to the deoxyribonucleohistones and pro-
tamines, most ribonucleoproteins^' ^ are very difficult to untie.
This impedes the isolation of intact ribonucleic acids, since such
preparations are so difficult to free of protein that, when this is
accomplished, they can no longer be regarded as truly represent-
ative specimens. It would seem that electrostatic links are only
of secondary importance in the architecture of most pentose
nucleoproteins. It may be considered as a consequence of this
stubborn resistance to dissociation on the part of the pentose
nucleoproteins that nucleotide analyses carried out on the total
conjugated nucleoprotein have, on the whole, given much more
satisfactory results than those performed on isolated pentose
nucleic acids^.

Of the representatives of the deoxynucleoprotein family it is
the nucleohistone of calf thymus that has been studied more than
any other and that is — probably quite wrongly — considered as
the prototype of the deoxypentose nucleoproteins. Less work has
been done on the nucleoprotamines, as they occur, for instance,
in fish sperm; and very little is known about microbial deoxy-
nucleoproteins. As to the pentose nucleoproteins, cytoplasmic
preparations mostly derived from the microsome fraction"^ and

DEOXYPENTOSE NUCLEIC ACIDS 65

complexes of bacterial origin can be studied. The most widely
investigated nucleoproteins, of both the deoxy and pentose types,
namely, the viruses and phages, probably represent more com-
plex types of organization than are encountered in the conjugated
proteins that I have mentioned above.

Two principal methods are used for the isolation of deoxy-
ribonucleohistone and of similar complexes with protamine,
namely, (a) extraction with strong salt solution; (b) extraction
with solutions of very low ionic strength. The procedures using
strong salt solutions are most widely used, though they would
appear more objectionable, if the dissociation of the nucleo-
protein and the formation of an artifact is to be avoided. These
nucleoproteins are soluble at a very low electrolyte concentration.
As the latter increases, the solubility first drops rapidly, reaching
a minimum in isotonic solutions, and then rises gradually. Though
variations occur, freshly prepared nucleohistone is, in general,
soluble at NaCl concentrations higher than 0.7 or 0.8 M.

The course of dissociation of two nucleoproteins (calf thymus
and wheat germ) at rising NaCl concentrations is illustrated in
Fig. 2 (p. 45, Ref. 8) and is based on experiments of Crampton
et aiy and of Lipshitz and Chargaff^^. The observation that a
progressive separation between protein and nucleic acid takes
place has made possible the fractionation of the latter, as will be
mentioned below.

3. DEOXYPENTOSE NUCLEIC ACIDS

The position of the phosphate bridges holding together the
mononucleotides of a polynucleotide chain has been studied in
detail only in calf thymus deoxyribonucleic acid; the available
evidence points to this link going from the 5' position to the 3'
position on the neighboring nucleosides^. Whether this generali-
zation holds for many, or all, deoxypentose nucleic acids cannot
yet be stated. Wherever the sugar moiety could be identified, it
proved to be 2-deoxyribose^; but here, again, no general state-
ment can be made. As to the nitrogenous constituents, two

References p. 75

66 NUCLEOPROTEINS AND NUCLEIC ACIDS

purines, adenine and guanine, and two pyrimidines, cytosine and
thymine, occur in all deoxypentose nucleic acids, with the ex-
ception of certain E. coli phages in which cytosine is replaced
by 5-hydroxymethylcytosine. A minor, fifth component is often
encountered, viz., 5-methylcytosine. For the structure of the
nucleosides and nucleotides a recent survey may be consulted^^.

The evidence accumulated thus far indicates that there exists
a very large number of different nucleic acids, which can be
distinguished by different proportions, and therefore by differ-
ences in the sequence, of the component nucleotides^. The two
most extreme examples are the deoxyribonucleic acid from a sea
urchin, in which the ratio of the sum of adenine and thymine to
that of guanine and cytosine ("the dissymmetry ratio") is 1.9
("AT type"), and, on the other hand, the deoxyribonucleic acid
of the avian tubercle bacillus in which it is 0.4 ("GC type").
While species specificity of composition can often be demon-
strated, no indications of organ specificity have as yet emerged.

The incorrect tetranucleotide hypothesis has been abandoned;
and it must be concluded that the nucleic acids are complicated
and intricate high polymers, devoid of a recognizable periodicity.
Despite wide divergences in their composition all deoxyribonu-
cleic acid preparations appear to be characterized by several
regularities^ These are: (a) A + G = C + T; (b) A = T; (c)
G = C; and, therefore, (d) A + C = G + T*. Only the last
regularity, with thymine being replaced by uracil — A + C =
G + U — seems to apply to the ribonucleic acids^; in other
words, the molar sum of compounds carrying an amino group
in 6 position equals that of compounds with a 6-keto group in
both types of nucleic acid.

4. FRACTIONATION OF DEOXYPENTOSE NUCLEIC ACIDS

Throughout all our work on the composition of nucleic acids
we were beset by one question: does a nucleic acid preparation.

* Abbreviations: A, adenine; G, guanine; C. cytosine; T. thymine; U,
uracil; or the corresponding nucleotides.

FRACTIONATION OF DEOXYPENTOSE NUCLEIC ACIDS 67

as it is isolated from a given cell, represent one molecular species,
several or many? On the one hand, we had observed a remark-
able constancy in composition, at any rate of the deoxypentose
nucleic acids from the same species. On the other hand, it ap-
peared unlikely on general biological grounds that a cell gave rise
to only one kind of deoxypentose nucleic acid. But the task of
separating a mixture of closely related polyampholytes of a very
high molecular weight appeared hopeless. Often, when I discussed
this problem with my colleagues, I was asked what could be done
about it; and my answer was that, if we were dealing with three
or four different nucleic acids, a separation may be feasible, but
that with a very large number of individuals in the same prepa-
ration we might as well treat them as one compound. This is,
indeed, a problem that leads us directly into very deep waters,
namely, the meaning of chemical identity in macromolecules.

In any event, it occurred to us that a separation, even if partial,
could perhaps be effected at the nucleoprotein stage. As I
mentioned before, the fractional dissociation of nucleohistone by
salt solutions of increasing concentration seemed to offer pos-
sibilities. This hope was realized^^, and the findings were later
extended in our laboratory^^- ^^' ^^, and confirmed in others^^* ^'^.
(For a survey, compare Ref. 3). As an example, I am showing
here the fractionation of a preparation of wheat germ nucleo-
protein which is described in a paper that is in course of
publication^^. The data are assembled in Table 18. The typical
picture can be seen: the fractions are characterized by progres-
sively decreasing proportions of guanine, cytosine and 5-methyl-
cytosine and progressively increasing quantities of adenine and
thymine; but the unity relationships are maintained. There is
noted, moreover, an uneven distribution of methylcytosine with
respect to cytosine.

Numerous fractionation experiments with different nucleic
acids, in many of which artificially prepared complexes with
basic proteins^^' ^^ or polyelectrolytes^^ were employed, have led us
to conclude that a preparation of deoxyribonucleic acid may com-
prise a very large number of differently constituted individuals.

References p. 75

68

NUCLEOPROTEINS AND NUCLEIC ACIDS

o

^ t^

q > •;:

o 5; 2

^OOOOOOOOON

qpoooopppp

0000000000

«o 00 ra

O ON ^

,J, d> ^

T- 1 ^— I rn

^ "^ ^ "^

00 o\ On r^

rj (S <S (N (N (N c^

_ Tf f<J Tf

vo r^ r^ r-^

as ON ^ '^ p
r-^ vd vo vd vo

p in 00
«n >o vd

(Ni r-; vo p r4 00

fvj r^ ^ T^ r-I r-^

«o Ov "-J -^
>o «o •^

ra«or^>o«^pr-4(vjp^

vdlO^t^ONONONOOOO

r-r-~oo»nc<^t-;«na\

10 »o '^ "^ ^

ooooodooo

^ a 5 > >

> > >

f2

O

o

o

03

(J cu

•a >;

Oh -J

PENTOSE NUCLEIC ACIDS 69

5. DIFFERENCES BETWEEN PENTOSE NUCLEIC ACIDS

The demonstration of clear-cut analytical differences in pentose
nucleic acids has met with greater difficulties than was true of
the deoxypentose nucleic acids. This was in part due to their
greater lability, which rendered many of the results obtained on
highly purified specimens doubtful, but it was also due to the
fact that the spread in individual nucleotide proportions is by no
means as great in PNA as in DNA. That there existed great dif-
ferences between the ribonucleic acids of yeast and those of pan-
creas or liver had, however, been recognized early. The entire
problem has been reviewed^. I shall limit myself here to quoting
a few typical results from a recent paper in which total pentose
nucleoprotein fractions were analyzed under conditions that
could not have led to fractionation^. These results will be found
in Table 19.

A comparison between pentose nucleic acids from different

TABLE 19

NUCLEOXroE RATIOS IN PENTOSE NUCLEIC ACIDS*

Source

Pii

Py

A-\-U

6- Am

G + C

6-K

Ox liver

0.80

0.63

1.04

Rat kidney

0.96

0.66

1.00

Cytoplasm of rat liver and kidney

0.99

0.62

0.93

Paracentrotiis lividus, eggs and embryos

1.08

0.77

0.99

Wheat germ

1.07

0.76

1.00

Yeast

1.00

1.12

0.93

E. coli

1.18

0.87

1.00

M. phlei

1.06

0.73

0.93

* Taken from Elson and Chargaff^, with the exception of the results on
wheat germ PNA which are taken from Lipshitz and Chargaffio. — Ab-
breviations: A, adenylic acid; G, guanylic acid; C, cytidylic acid; U,
uridylic acid; Pu, purine nucleotides: Py, pyrimidine nucleotides; 6-Am,
A + C; 6-K, G + U.

References p. 75

70 NUCLEOPROTEINS AND NUCLEIC ACIDS

Species and tissues and from different locations within the cell
has, moreover, demonstrated that while the differentiation
between species and organs on analytical grounds meets with dif-
ficulty, differences between nuclear and cytoplasmic nucleic acids
can be shown^^.

6. REMARKS ON FUNCTIONS

The spectator who has followed the development of our knowl-
edge of nucleic acids in the past few years cannot help feeling
that the stage is set for a grand finale. Many an acrobat has an-
nounced the salto mortale to a gasping audience (hoping to
change it into a salto immortale); and though often only a grace-
ful pirouette could be seen, tamen est laudanda voluntas. But I
beUeve that the deeper we get, the darker it becomes. It is true,
the temptation is almost overwhelming to pour all into one pot:
growth, and its change from benign to malignant; bacterial trans-
formations; the transmission of hereditary properties; the multi-
plication of virus and phage; the synthesis of inducible enzymes;
the differentiation of tissues; the formation of antibodies — every-
where an interplay between nucleic acids and proteins, a spin-
ning wheel in which the thread makes the spindle and the spindle
the thread. But if it is hard to recognize the similarities between
what is ostensibly dissimilar, it is even more difficult to make out
the differences between what appears so similar; and moderation
still is among the tools of our science.

Brushing past, therefore, the heralds of immediate glory and
convinced, as I am, that we are not obliged to solve the secret
of life in 1955, I must conclude that the actual and established
evidence on direct biological functions of the nucleic acids is
rather meager. I should, furthermore, question whether a search
for functions, at any rate in the narrowly anthropomorphic
framework of a vulgar teleology, is what we need most at the
moment.

The relevant information on the biological roles of the deoxy-
pentose nucleic acids^^ and the pentose nucleic acids^ has been

REMARKS ON FUNCTIONS 71

reviewed recently. I shall list here only briefly what to me ap-
pears most significant.

(a) The nucleic acids are ubiquitous and presumably indis-
pensable constituents of living matter. Autarkic entities (cells
and cell communities) appear to require the presence of both
types of nucleic acid, parasitic entities (viruses, phages) the
presence of one. It must be considered as highly important that
the transfer of biological information can be effected, or aided,
either by pentose nucleic acids (plant and, probably, animal
viruses) or by deoxypentose nucleic acids (bacteriophages).
Against the argument of omnipresence the objection could be
raised that this hardly distinguishes the nucleic acids from other
plastic components of living matter, e.g., the proteins or the
Upids, though it is doubtful whether the latter are required by
parasites.

(b) The most direct evidence of an involvement of poly-
nucleotides of the deoxy series in the genetic apparatus of the
cell is due to the epochal work of Avery and his colleagues on
the phenomena of bacterial transformation. There can be little
doubt that our present understanding of the mechanisms that are
brought into play during transformation is at a most rudimentary
stage, but even less doubt that this discovery will be marked as
an important date in the history of biology.

(c) The investigation of the mutagenic action spectrum of
ultraviolet radiation has yielded curves that closely resemble the
absorption spectrum of nucleic acids. Certain chemicals known
to react with nucleic acids exhibit mutagenic effects^^ and others
known to be cytostatic or cytotoxic have been shown to be in-
corporated into nucleic acids^^.

(d) Boivin and his colleagues, pointed out that the deoxy-
pentose nucleic acid content of a diploid nucleus is constant for
a given species and twice the amount found in the corresponding
haploid nucleus^i.

(e) Many lines of evidence suggest a role of pentose nucleic
acids in protein synthesis and in morphogenesis^. The relation-
ship has, however, remained purely formal and generic; no cor-

References p. 75

72 NUCLEOPROTEINS AND NUCLEIC ACIDS

relation between specific sequences in the nucleic acid and
specific synthetic capacities has been established.

(f) The fact that the nucleotide distribution in deoxypentose
nucleic acids is characteristic of the species, and unchanged in
different tissues of the same host, must be contrasted with the
far-reaching changes undergone by the proteins.

But an old question must be asked before many of these ar-
guments: What is a cause, what is a symptom? Are nucleic acids
different because they come from different cells, or is this very
difference one of the causes of biological specificity? We seem to
have lost the ability to say "We don't know". But this is a
privilege that I still wish to retain.

We may draw the following preliminary conclusions from the
work on the chemistry of the nucleic acids. There exists a very
large number of differently composed individuals which are
distinguished by differences in the sequence of their component
nucleotides; for we must remember that different composition
signifies different sequence. No perceptible periodicity, nor a
repeating unit, can be claimed; but the distribution does not ap-
pear to be random. In calf thymus deoxyribonucleic acid the
purine nucleotides have a greater chance to be next to, or near,
other purine nucleotides than next to pyrimidine nucleotides, and
the converse must be true of the latter^^. Similar conclusions can
be drawn from our older work on ribonucleic acids-^, when ac-
count is taken of what is now known about the specificity of
pancreatic ribonuclease-^.

Yet, deoxyribonucleic acids, whether the whole or the sub-
fractions, all show the impressive regularities that I have men-
tioned before in Section 3; and most ribonucleic acids appear to
exhibit at least one regularity (Table 19). There may be some
that will ask whether there is any sense in chasing such regular-
ities or even whether, in view of the mischief worked by the now
defunct tetranucleotide hypothesis, a search for such norms is
not outright deplorable. If this represented an esthetic quest for
pre-established harmony or the music of the spheres, such ob-
jections would, perhaps, be justified. But quite apart from these

REMARKS ON FUNCTIONS 73

observations having grown out of experiments, we should not
forget that regularities of this sort may be the best, or only, means
of recognizing the existence of systems concerned with the
preservation, or the transfer, of information: tasks that we should
like to assign to the nucleic acids or, more probably, to the
nucleoproteins. A man standing at the street entrance of a subway
station may notice that the efflux of passengers, in contrast to
their inflow, showed a very marked periodicity or regularity; he
would rightly conclude that the trains operated according to a
schedule, and this might be the only way in which he could
reconstruct a timetable: the information was conveyed to the
observer by the imposition of a regularity of which the pas-
sengers themselves could have been unaware.

In the speculations on the biological role of the nucleic acids
the nucleoproteins usually are neglected. I believe this to be a
mistake. There is Httle evidence that the nucleic acids occur in
living matter in the free state. Though it is possible that the
conjugated proteins in which nucleic acids are found represent
artifacts resulting from fortuitous combination, this is not likely.
One could, in fact, assume that if specific nucleotide sequences
are the actual genetic determinants, they could give rise, in com-
bination with proteins, to unique geometrical shapes, so that —
through the association of a specific polynucleotide and a specific
polypeptide — another dimension, as it were, is produced. I may
refer to a discussion of some of these problems in which the
inheritance of geometrical peculiarities was considered in anal-
ogy to problems in topology ^^.

In any event, there can be little doubt that methods for the
sequence analysis of nucleic acids are most urgently needed. We
are still very far from it; we cannot yet say whether a nucleic
acid molecule should be considered as the carrier of the infor-
mation assigned to one gene or whether — and this appears more
likely to me — one and the same molecule represented a string
carrying multiple potential information, similar to the knot
writing of ancient Peru, the quipu.

The copying mechanism suggested for deoxyribonucleic acid^^

References p. 75

74 NUCLEOPROTEINS AND NUCLEIC ACIDS

provides an example of a system through which information is
preserved. Our recent suggestion^, in which the regularity ob-
served in ribonucleic acids, namely, the equality in the propor-
tions of 6-amino and of 6-keto compounds (compare Table 19),
is translated into a relationship with protein, whose regularly oc-
curring peptide bonds could, through hydrogen bonding, impose
this regularity upon two polynucleotide chains, is an example of
a system through which information is transferred.

I believe, however, that while the nucleic acids, owing to the
enormous number of possible sequential isomers, could contain
enough codescripts to provide a universe with information, at-
tempts to break the communications code of the cell are doomed
to failure at the present very incomplete stage of our knowledge.
Unless we are able to separate and to discriminate, we may find
ourselves in the position of a man who taps all the wires of a
telephone system simultaneously. It is, moreover, my impression
that the present search for templates, in its extreme mechano-
morphism, may well look childish in the future and that it may
be wrong to consider the mechanisms through which inheritable
characteristics are transmitted or those through which the cell
repeats itself as proceeding in one direction only. With these
words I have perhaps again tied the knot that I tried to open in
the beginning, since it may be that the first causes themselves are
subject to a random priority.

NUCLEOPROTEINS AND NUCLEIC ACIDS 75

REFERENCES

1 E. Chargaff and J. N. Davidson (Eds.), The Nucleic Acids: Chemistry

and Biology, Vols. I and II, Academic Press, New York, 1955.

2 J. N. Davidson and E. Chargaff, in E. Chargaff and J. N. Davidson

(Eds.), The Nucleic Acids: Chemistry and Biology, Vol. I, Academic
Press, New York, 1955, p. 1.

3 E. Chargaff, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. I, Academic Press, New York,
1955, p. 307.

4 B. Magasanik, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. I, Academic Press, New York,
1955, p. 373.

5 J. Brachet, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. II, Academic Press, New York,
1955, p. 475.

6 D. Elson and E. Chargaff, Biochim. Biophys. Acta, 17 (1955) 367.

7 G. H. HOGEBOOM AND W. C. SCHNEIDER, in E. ChARGAFF AND J. N.

Davidson (Eds.), The Nucleic Acids: Chemistry and Biology, Vol. II,
Academic Press, New York, 1955, p. 199.

8 E. Chargaff, Symposia Soc. Exptl. Biol., 9 (1955) 32. [See Chapter 3

of this book.]

9 C. F. Crampton, R. Lipshitz and E. Chargaff, /. Biol. Chem., 206

(1954) 499.

10 R. Lipshitz and E. Chargaff, Biochim. Biophys. Acta, 19 (1956) 256.

11 D. M. Brown and A. R. Todd, in E. Chargaff and J. N. Davidson

(Eds.), The Nucleic Acids: Chemistry and Biology, Vol. I, Academic
Press, New York, 1955, p. 409.

12 J. Baddiley, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. I, Academic Press, New York,
1955, p. 137.

13 E. Chargaff, C. F. Crampton and R. Lipshitz, Nature, 111 (1953)

289.

14 C. F. Crampton, R. Lipshitz and E. Chargaff, /. Biol. Chem., 211

(1954) 125.

15 P. Spitnik, R. Lipshitz and E. Chargaff, J. Biol. Chem., 215 (1955)

765.

16 G. L. Brown and M. Watson, Nature, 111 (1953) 339.

17 J. A. Lucy and J. A. V. Butler, Nature, 174 (1954) 32.

18 D. Elson, L. W. Trent and E. Chargaff, Biochim. B'ophys. Acta, 17

(1955) 362.

19 R. D. HoTCHKiss, in E. Chargaff and J. N. Davidson (Eds.), The

Nucleic Acids: Chemistry and Biology, Vol. II, Academic Press,
New York, 1955, p. 435.

20 G. B. Brown and P. M. Roll, in E. Chargaff and J. N. Davidson

(Eds.), The Nucleic Acids: Chemistry and Biology, Vol. II, Academic
Press, New York, 1955, p. 341.

76 NUCLEOPROTEINS AND NUCLEIC ACIDS

21 R. Vendrely, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. II, Academic Press, New York,
1955, p. 155.

22 C. Tamm, H. S. Shapiro, R. Lipshitz and E. Chargaff, J. Biol. Chem.,

203 (1953) 673.

23 B. Magasanik and E. Chargaff, Biochim. Biophys. Acta, 7 (1951) 396.

24 G. Schmidt, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. I, Academic Press, New York,
1955, p. 555.

25 E. Chargaff, Cold Spring Harbor Symposia Quant. Biol., 12 (1947)

28.

26 J. D. Watson and F. H. C. Crick, Nature, 111 (1953) 964.

CHAPTER 5

The Very Big and the Very Small
Remarks on Conjugated Proteins'^

It is an old experience in the natural sciences that what is poison
for one generation often is honey for the succeeding one. Whether
this indicates, in a given case, the dawn of a better era or an
inurement by frequent exposure to sublethal doses of truth often
cannot be decided without the perspective of centuries. There are
important exceptions, but in general a scientific truth fades every
30 years, to be replaced by another equally evanescent. A well-
designed and well-constructed chair lasts longer.

It is not very long ago that the extreme contempt for the
amorphous and intractable, felt by generations of organic
chemists (or at least by the second-rate specimens), has made
room for the realization that there is little sense in treating living
and growing tissue merely as the starting material for the isolation
of well-behaved crystalline substances. In recent times, the respect
for nature and its multiform manifestations (as such a very
healthy sign) has, in fact, sometimes assumed exaggerated
proportions; and it is occasionally necessary to point out that the
living cell is not simply a macromolecule with a skin, or the
bacteriophage a nucleoprotein with a tail. So-called model ex-
periments often are carried to incredible lengths, prompting one
to say that confusion superimposed on complexity may produce
papers, but not results, and that a skunk dipped into chlorophyll
is not yet an apple tree. The secret of the organization of the cell
will not be found by a clever sleight of hand.

* Reprinted with permission from S. Graff (Ed.), Essays in Biochemis-
try, Wiley and Sons, New York, 1956, pp. 72-76.

78 CONJUGATED PROTEINS

It is, however, becoming clear that organization, as observed
on the macroscopic and microscopic levels, must be matched, on
the submicroscopic and molecular levels, by the existence of pat-
terns in which the varied arrangement of a limited number of
constituents serves to impress individuality and specificity on cells
or cell communities. One could venture the opinion that biochem-
ical evolution is accompanied, or indeed caused, by the for-
mation of macromolecules of ever-increasing complexity com-
posed of an ever-diminishing number of constituents. One
mechanism by which this may be accomplished is that of conjuga-
tion. Many different conjugated proteins are being recognized,
such as the nucleoproteins, the lipoproteins, the mucoproteins,
or the chromoproteins; many enzymes that carry cof actors, dis-
tributed in specific positions on the protein molecule, belong to
one or the other of these groups. But we encounter also lipopep-
tides, mucolipids, etc.; and many more unrecognized compounds
of this type must daily be going down the drains of our labora-
tories than repose in the graveyards of our scientific journals.
When I am sometimes told that biochemistry has "run out of
good problems", I shudder and reply: "Biochemistry has not
even begun!"

The general aspects of the problem of conjugation have rarely
been formulated clearly. This is perhaps not surprising, for this
class of substances has long found itself between two chairs, as it
were: too big to be handled conveniently by the chemist; too
small to be seen consistently by the morphologist. The chemist
strove for the isolation of the smallest unit endowed with homo-
geneity; the biologist attempted the recognition of the simplest
structure endowed with function. The prize went to the loudest
prophet with monomaniac intent. Since scientists in our time
have, on the whole, lost the ability to say "We don't know", it is
usually the man with the premature explanations that brings
home the bacon; and by the time he has been found out he will
have devoured it.

Another obstacle to the recognition of the biological impor-
tance of the conjugated proteins may be seen in their being

CONJUGATED PROTEINS 79

usually considered apart from each other under the headings of
their respective prosthetic groups. I am aware of only one in-
stance, namely, in a symposium held some time ago (1953) at
Rutgers University, in which an attempt was made to consider
the conjugated proteins as a family of substances having more in
common than the small print that they occupy in the current
textbooks of biochemistry. But there can be little doubt that
many, if not all, hfe processes take place on what may be con-
sidered the surfaces of conjugated proteins. It is, perhaps, not
uninstructive very briefly to compare two groups of conjugated
proteins on which my laboratory has spent some effort, viz., the
nucleoproteins and the lipoproteins.

As concerns the nucleoproteins, a strict distinction must be
made between the complexes containing deoxypentose nucleic
acids and those in which a pentose nucleic acid acts as the
prosthetic group. When the conjugated proteins associated with
the same type of nucleic acid are compared, it will be noticed
that their properties are governed by the type of protein they
contain rather than by the composition of their nucleic acid
moiety. This is undoubtedly due to the ability of proteins to differ
much more from each other in their chemical and physical
properties than do the nucleic acids. The deoxynucleoproteins
are, in many cases, complexes whose protein moiety is repre-
sented by a protein of markedly basic properties, such as a
histone or a protamine. Although it is inviting to consider such
compounds as salts between the cationic protein and the anionic
nucleic acid, this would be wrong. It is, for the moment at any
rate, much safer to regard the nucleoprotamines and the nucleo-
histones as specifically conjoined complexes of a complicated
and, thus far, little-understood geometry to which both electro-
static and secondary valence bonds contribute. What all these
compounds, however, appear to have in common is that they are
readily dissociated by high electrolyte concentrations and that
under conditions permitting the removal of the protein moiety,
by precipitation or denaturation, the nucleic acids are liberated.

Occasionally deoxynucleoproteins are encountered in micro-

80 CONJUGATED PROTEINS

organisms that are exceptional in resembling the ordinary pentose
nucleoproteins rather than the nucleohistones. On the other hand,
certain plant viruses represent exceptions to the behavior of the
most commonly found pentose nucleoproteins. Of the latter, as
they occur in the microsomes and the nucleoli, it may be said
that they exhibit less of an electrostatic character than the deoxy-
nucleoproteins. The bonds holding the ribonucleic acid to the
protein are broken with much less ease; and one gains the im-
pression that in this case the structure of the prosthetic group is
inextricably associated with the structure of the entire conjugated
protein. Whereas the dissociation of a nucleohistone may be com-
pared to the removal of branches from the trunk of a tree, the
separation of the ribonucleic acid and protein moieties of a
ribonucleoprotein resembles much more the disentanglement of
the warp and the woof of a fabric.

All nucleoproteins share, however, one important feature: they
are combinations of two types of giant ampholytes, each of which
can, and undoubtedly does, exhibit innumerable specificities as
regards shape and constituent sequence. Their combination
probably does add a new dimension; but each partner is, in itself,
fully competent to maintain a specific pattern and to convey
intricate information.

Of the lipoproteins, on the other hand, it could be said that it
is only through the attachment of the monomeric lipids to a
protein that a specific pattern of lipid arrangement becomes pos-
sible. It is not improbable that the future will show that certain
lipids can exist in the cell in a polymerized form capable of
exhibiting sequential specificity. But up to the present the lipids
seem to be the only bulk components of tissues that must be as-
sumed to exist principally in a monomeric form. If the establish-
ment of specific arrangements is considered as an attribute of
cellular organization, the formation of specific lipoproteins is one
of the ways in which the lipids can take part in such specific
patterns. Certain lipids probably are attached to the proteins
by a combination of electrostatic and hydrogen bonds; others
may occur as solutions in the lipid moieties of lipoproteins.

CONJUGATED PROTEINS 81

There is little evidence of the existence of covalent links between
lipid and protein.

The two types of conjugated protein considered here very
briefly are, to a certain extent, representative of conjugated
proteins in general. The prosthetic group may be soluble or in-
soluble in water; it may be a monomer of comparatively simple
structure, a mixture of monomers, or a macromolecule having
itself a complicated structure and being capable of an intricate
sequential specificity. As may have been gathered from what I
said before, I consider the principle of conjugation as the main
process through which nature makes big things bigger and small
things big. Size, in compounds participating in the Hfe of the cell,
is probably not an accident. Moreover, such processes of pre-
determined aggregation may be one of the ways in which what
sometimes is stupidly referred to as the "assembly line" is realized
in the living cell. The models of which we can conceive are
probably no more than an absurd caricature of the synthetic
mechanisms, a multiplicity of templates in space and templates
in time, through which the organism maintains patterns of this
high degree of complexity, unless we assume (and there is no
reason for that) that what is duplicated is not really a duplicate.

Romantic deduction has done much harm in the sciences. But
the use, the almost unpredictable use, of iniagination is an es-
sential element in the operations of the human mind. The injunc-
tion not to be astonished — nil admirari — is one of the most stupid
legacies of antiquity. When we consider this ever-repeated giant
throw of dice, this internally regulated cataract of reactions,
sequences, and products, our first response must be a deep
astonishment at a chaotic regularity which has thus far defied
our understanding. Eternal surprise is the engine that drives the
searching intellect. Let us hope that the coming generations will
not have lost the ability to wonder about the many meanings of
these palimpsests of nature.

CHAPTER 6

Of Nucleic Acids and Nucleoproteins^

That each from other differs, first confess;
Next, that he varies from himself no less.

POPE I EPISTLE TO COBHAM I 19-20

1 . INTRODUCTION

A speaker before a group such as I have the honor of addressing
today has one of two choices: he can say very much about very
httle or very little about very much. There are, of course, two
more choices possible; but the usual lecturer — grown fat in, if
not from, the pursuit of science — will be too conceited to admit
the one and too modest to consider the other. In any event,
though it would be nice to be able to say very much about a very
important matter, the title of my talk must have told you that
nothing of the kind is contemplated. In fact, as I go on I may
give the impression of asking questions rather than giving an-
swers. I can only hope that I shall not find myself in the position
of the Indian conjurer who after having fixed the traditional rope
in the air climbed up to hang himself with it.

We live in a time that is drunk with experiments. Dubious
results, dubiously paralleled, serve to estabhsh so-called facts with
a celerity that would make a monkey blush. A scientific Rip van
Winkle, were he to return from only a few weeks' sleep, would
find that in this short time ten people have described a "system
that makes RNA" and ten, a "homogenate that makes DNA";
and twenty others have even "synthesized protein in vitro'\ But
when he takes a closer look at the sort of RNA and DNA and
protein that have been produced, he may decide to go back to
sleep. Perhaps Aristotle would have felt at home, but would

* Reprinted with permission from Harvey Lectures, 1956-1957, 52 (1958)
57-73.

ISOLATED CELL CONSTITUENTS 83

Plato, would Socrates? What Heraclitus would have said, I know.
He would have said: "You do not make the same mess twice".
For, really, our entire conception of the meaning of isolated cell
constituents and of their action under the artificial conditions of
customary experimentation is due for a revision. I, for one,
would not be astonished if thirty or fifty years hence most of the
terms and conceptions with which our science — perhaps then
known as baroque biochemistry — has made us familiar would be
deader than phlogiston is today.

2. THE SIGNIFICANCE OF ISOLATED CELL CONSTITUENTS

Now, let us ask a very silly question*. Can we be sure that the
substances that we isolate from the cell exist in the living cell?
I should say, the answer will have to be: They do and they do
not. A simple monomer — a fatty acid, a sugar, etc. — which we
find in a cell probably has not been produced de novo in the
course of the devastating processes known as careful isolation.
Even in such cases, to decide whether these substances occurred
in the free state is a difficult matter. It becomes immensely more
so when we come to high polymers often endowed, as in the case
of proteins or nucleic acids, with a multiplicity of charges. I have
little doubt that the physical description of the molecular ar-
rangement of such complicated compounds, which must be
drawn, quartered, pickled, or embalmed in order to be studied,
defines the pleasing shape into which they can be put rather than
the form in which they exist in life. For this reason, I look with
esthetic pleasure but with the utmost diffidence and mental reserve
on the various structural models and spiral contortions — beautiful
examples of non-representational sculpture — that have been
proposed for the nucleic acids. I should advise to wait and see.
Models — in contrast to those who sat for Renoir — improve with
age.

* "The 'silly question' is the first intimation of some totally novel
development." L. Price, Dialogues of Alfred North Whitehead (Mentor
Book Edition, New York, 1956, p. 145).

References p. 98

84 NUCLEIC ACIDS AND NUCLEOPROTEINS

The organic chemist, the biochemist, the crystallographer, the
biologist often look at the same object and they may even use
the same term, but they mean different things. Our natural
sciences, which once could find room in a single brain, have
grown diffuse and multilingual. We are faced with a real problem
of translation: from the molecular to the submicroscopic level;
and then, in turn, to microscopic structures, morphological units,
organelles, cells and cell communities. Where and when shall we
find the dictionary that could help us in these translations? I
should say, probably not before biochemistry has earned the first
half of its name. It is, of course, not surprising that such a term
as "nucleic acid" does not carry the same connotation in the
several disciplines. A sculptor and a tailor naturally have dif-
ferent ideas of what makes a man: one looks at the torso, the
other at the pants.

Nevertheless, we must be content with avoiding the avoidable
and use the means that our science has given us. It is time for
me to speak of nucleic acids.

3. TWO TYPES OF NUCLEIC ACID

You all know that the convenient slang terms DNA and RNA
stand, respectively, for deoxyribonucleic and ribonucleic acids*.
If this were a more strictly chemical talk I would point out that
the sugars have been identified in only a few cases and that a
more neutral designation, such as deoxypentose and pentose
nucleic acids, would be safer. But tonight we may disregard that,
as the nature of the sugars occurring in the nucleic acid speci-
mens of which I am going to speak has in almost all cases been
verified.

When about ten or eleven years ago my colleagues, first of all

* The following abbreviations will be used: DNA, deoxypentose nucleic
acid(s); PNA, pentose nucleic acid(s); A, adenine; G, guanine; C, cytidine;
MC, 5-methylcytosine; T, thymine; U, uracil (or the corresponding nucleo-
tides); Pu, purines; Py, pyrimidines; 6-Am, adenine + cytosine; 6-K,
guanine + uracil.

TWO TYPES OF NUCLEIC ACID 85

Dr. E. Vischer and Mrs. C. Green, and I embarked on the study
of the chemistry of nucleic acids, the main known facts were
based principally on the fundamental work of Miescher, Kossel,
Levene, Steudel, Feulgen, Thannhauser, Hammarsten, Jorpes,
Caspersson, Brachet, Bawden, Pirie, Stanley, Avery, and their
collaborators^'-. At that period, "the present stage of our knowl-
edge of nucleic acid chemistry could perhaps be compared to
that in which protein chemistry found itself at the beginning of
this century"^. The composition of only two nucleic acid prepa-
rations, the deoxyribonucleic acid of calf thymus and the ribonu-
cleic acid of yeast, was more or less completely known, and this
only qualitatively. Although nucleic acids, at any rate calf thymus
DNA and the pentose nucleic acids of plant viruses, were known
to exist as high polymers, although viruses had been recognized
as nucleoproteins, although the principle active in bacterial trans-
formation had been identified as deoxypentose nucleic acid — and
I could go on listing a few more disconcerting facts — it was
almost generally assumed that nucleic acids were relatively simple
compounds composed of a series of "tetranucleotides". As no
methods were available to test this structural hypothesis, this
simpUfication contributed considerably to the peace of mind of
the chemist; and it continues to have this soothing effect on many
investigators even now.

We approached the study of nucleic acid chemistry in the
belief that the nucleic acids were complicated macromolecules,
comparable to the proteins in their intricate and specific struc-
ture^. It was evident that for a discussion of the chemical specific-
ity of the nucleic acids to become possible, one question had to
be answered first: How much of what do they contain? This
required, first of all, the elaboration of precise micromethods for
both the qualitative and quantitative determination of all nucleic
acid components — procedures that because of the strong ultra-
violet absorption of the purines and pyrimidines were based on
filter paper chromatography and spectrophotometry in the ultra-
violet^-^.

The principal nitrogenous constituents of the nucleic acids are

References p. 98

86 NUCLEIC ACIDS AND NUCLEOPROTEINS

adenine, guanine, cytosine, and thymine in DNA, and the first
three and uracil in RNA. DNA appears somewhat more varied
in the array of its constituents than RNA; sateUites such as 5-
methylcytosine or 6-N-methyladenine are found occasionally, and
in the even coliphages 5-hydroxymethylcytosine occurs instead
of cytosine. What should be borne in mind is that, in general,
there are two purines and two pyrimidines and that one member
of each of these two groups carries, in its 6-position, an amino
group, the other a hydroxyl or, tautomerically, a keto group.

4. DEOXYPENTOSE NUCLEIC ACIDS

a. Composition and regularities

As early as in 1949, our studies had led to the conclusion that
there existed a very large number of different deoxyribonucleic
acids: different as regards the proportions, and therefore the
sequence, of their component nucleotides, but of a composition
characteristic of the species, though not of the tissue, from which
they were derived^- ^' ^^. These conclusions have received ample
confirmation since that time^^-^^. In Table 20, I have selected a
few analyses from our own work^^- ^^"^^ in which DNA specimens
from various sources, easily seen to be different in composition,
are listed. The examples have been so chosen as to comprise
the entire spread in divergences found so far. The dissymmetry
ratio, i.e., the ratio of the sum of adenine + thymine to that of
guanine + cytosine + methylcytosine, ranges from approximate-
ly 1.9 to 0.4. We have referred^^ to deoxypentose nucleic acids
with a ratio above 1 as the AT type, to those with a ratio below 1
as the GC type; the latter appears to occur only in several micro-
organisms. Escherichia coli belongs to an intermediate class with
almost equal quantities of all components.

It is, of course, not always possible to distinguish between
nucleic acids of very different origin through a study of their
total composition. Two examples of analytically indistinguishable
preparations^^' 20 are listed in Table 21. In such instances, it is
necessary to resort to other methods of differentiation; I shall

DEOXYPENTOSE NUCLEIC ACIDS 87

refer to them later. But one thing should be stressed here. Some
people are bothered by such similarities as I have shown in Table
21, and I have heard it said, and also read it in one of the reviews
of the recent treatise on nucleic acids^ that there is no convincing

TABLE 20

COMPOSITION OF SEVERAL TOTAL DEOXYRIBONUCLEIC ACIDS

Dissym-

Moles per

100 gram

-atoms P

metry

ratio

A

G

C

MC

T

No. Origin

{A-\-T)l
{G + C + MC)

1 Paracentrotus

livid us

32.8

17.7

16.2

1.1

32.1

1.85

2 Man

30.4

19.6

19.2

0.7

30.1

1.53

3 Ox

29.0

21.2

19.9

1.3

28.7

1.36

4 Wheat germ

28.1

21.8

16.8

5.9

27.4

1.25

5 E. coli (K-12)

26.0

24.9

25.2

—

23.9

1.00

6 Avian tubercle

bacillus

15.1

34.9

35.4

—

14.6

0.42

evidence that the deoxypentose nucleic acids of mammaUan tis-
sues differ from one species to another. I should, however, like
to submit an entirely opposite proposition, namely, that once it
is shown that there exist different nucleic acids (Table 20), it is
safer to assume that all nucleic acids are species- specific. For we

TABLE 21

EXAMPLE OF TWO DEOXYRIBONUCLEIC ACID SPECIMENS
INDISTINGUISHABLE ANALYTICALLY

Moles

per 100

gram-atoms

P

Dissymmetry
ratio

Origin

A

G

C

T

(A + T)/{G + Q

Sheep liver
Salmon sperm

29.3
29.7

20.7
20.8

20.8
20.4

29.2
29.1

1.41
1.43

References p. 98

88 NUCLEIC ACIDS AND NUCLEOPROTEINS

must remember that the minimum estimate^ of the number of
possible permutational isomers in which a substance of the com-
position of calf thymus DNA could exist is phantastically high,
namely 10^^^^. Actually, this number will be even much larger,
as a molecular weight of 750,000 for DNA, used in these com-
putations, is too low. How then expect two different organisms
that cannot make identical proteins to be able to dupUcate this
enormous throw of dice?

If we take it for granted that there exist a very large number
of different nucleic acids — a number presumably much larger
than can be shown merely by analysis — we may go on to ask:
Are there any principles that unify what appears so disparate?
No simple or recognizable repeating unit can be made out in
these giant polynucleotide chains; they do not seem to possess a
perceptible periodicity. But we became aware very early in our
work^ of several remarkable regularities apparently characteristic
of all deoxypentose nucleic acids. They are as follows: (a) The
sum of the purine nucleotides equals that of the pyrimidine
nucleotides, (b) The molar proportion of adenine equals that of
thymine, (c) The molar proportion of guanine equals that of
cytosine (+ methylcytosine). (d) The number of 6-amino groups
(adenine + cytosine + methylcytosine) equals that of 6-keto
groups (guanine + thymine). These regularities may be consid-
ered as well established. (See Table XVII in Ref. 13.) Polynu-
cleotide fragments produced by the enzymic degradation of DNA
do not show these regularities^^, a finding that will have to be
taken into account should well substantiated deviations from
these pairing principles be discovered.

b. Fractionation

Having plunged into a whirlpool, though decorated with har-
monious surface ripples, we proceeded to make it even more
chaotic. We asked a question. Granted, we said, that the DNA
of a given species represents a completely arrhythmic, highly
polymeric nucleotide chain, should we consider it as strictly
homogeneous or as consisting of a bundle of different chains, dif-

DEOXYPENTOSE NUCLEIC ACIDS 89

fering from each other in the patterns of nucleotide sequence?
I asked this question in our first paper on the chemistry of DNA^;
but for several years I was resigned to there not being an answer,
for there is nothing more difficult at the present stage of our
knowledge than the fractionation of a family of closely related
polyampholytes of a very high molecular weight. Fortunately, I
had started — first together with Miss Lipshitz and later also with
Dr. Crampton — on a study of calf thymus nucleohistone: its
chemical properties and its behavior on dissociation in strong
salt solutions--. We found that the extent of splitting of the
conjugated protein into nucleic acid and protein was a function
of the electrolyte concentration. When the DNA portions removed
by fractional dissociation were analyzed, they were found to
exhibit a curiously graded composition^^. The stepwise extraction
of nucleohistone or of artificially prepared nucleates of pro-
teins^'''' ^^' -^ or polylysine-^ with salt solutions of increasing
strength under conditions favorable for the denaturation of the
protein moiety yielded a series of nucleic acid fractions with
diminishing concentrations of guanine and cytosine and rising
concentrations of adenine and thymine. The equimolarity of each
pair of constituents and of total purines and pyrimidines was,
however, fully maintained in all fractions^^. The weighted average
composition that could be computed from all fractions cor-
responded very closely to that found for total DNA preparations
of the same species.

One may conclude that the DNA of a cell comprises an entire
spectrum of differently constituted individuals; a spectrum, how-
ever, with "bands" of different position and intensity in different
species. It is not inconceivable that no two nucleic acid molecules,
within the same nucleus, are entirely identicaP^: a prospect that
would seem to condemn us to forced statistics for life.

c. Structure and sequence

The existing evidence makes it appear likely that the nucleic
acids are chains in which nucleosides are linked to each other by
phosphoric acid diester bridges between the 5'- and 3 '-positions-'^.

References p. 98

90

NUCLEIC ACIDS AND NUCLEOPROTEINS

The "backbone", therefore, is represented by a regularly con-
structed polydeoxyribophosphate; and it is in the "superstruc-
ture" of purines and pyrimidines, linked glycosidically to V of
the sugar, that the differences between nucleic acids must reside
(Fig. 8). An alphabet of four or five letters may look meager, but
if the words which these four letters spell out are 25,000-letter
words, the result may be oppressively informative. Even if we
had a rigorously homogeneous nucleic acid and procedures for
its stepwise and orderly dismemberment, it would be hopeless to
undertake the decoding of its nucleotide sequence. By the time
the cryptographer had completed his task and written down the
entire arrangement, evolution would probably have overtaken
him and he could start over again.

A strict sequence analysis is, therefore, not only unattractive,
but probably impossible, at any rate in DNA; an assay of the
distribution density of individual components, i.e., a survey of
tendencies in nucleotide sequence, could, however, be considered.
By survey I mean an attempt at crudely mapping the order in
which the mononucleotides are aligned in a given nucleic acid
chain. If this alignment does not take place at random, prefer-
ential arrangements discovered in the various preparations may
furnish a means for their differentiation in addition to what total
analysis may reveal. We have attempted to achieve this in three
independent lines of investigation, based on the vertical or the
horizontal degradation of the nucleic acids (p. 338 of Ref. 13) or
on both.

D

A n p A A ■ A

Fig. 8. Schematic representation of a DNA segment. The purines are

represented by black, the pyrimidines by white, symbols. There is an equal

number of black and white squares and of black and white triangles. The

dissymmetry ratio of squares to triangles is 1.5.

DEOXYPENTOSE NUCLEIC ACIDS 91

These are the three lines of attack; but I shall deal only with
the third in some detail, (a) The action of crystalline pancreatic
deoxyribonuclease on DNA shows a characteristic trend in the
composition of small fragments formed at different periods; in
addition, it brings about the formation of larger fragments ex-
hibiting a distortion in the original internucleotide ratios^^-^^-^^.
I beheve it was through these studies that the arrhythmic charac-
ter of nucleotide sequence in DNA was first recognized, (b) A
partial horizontal degradation of deoxypentose nucleic acids leads
to the class of compounds designated as apurinic acids, first
studied with Drs. Tamm and Hodes^^. Their formation is based
on the great lability of the glycosidic link of the purines. Apurinic
acids may be considered as large polynucleotides, of at least 60
sugar phosphates and nucleotides in a row^^, that have been
deprived of their purines without distortion of the original inter-
pyrimidine ratios*. Such products are much more labile to chem-
ical, though not to enzymic^^, attack than intact DNA; and the
study of their degradation by alkali led to the conclusion that calf
thymus DNA is rich in chains in which tracts of pyrimidine
nucleotides alternate with stretches in which purine nucleotides
predominate^^. Similar conclusions could later be drawn with
respect to other deoxyribonucleic acids^''^. (c) The third approach
to the problem of nucleotide arrangement undertaken in col-
laboration with Dr. Shapiro^'* (and in part unpublished) has to
do with the production of pyrimidine nucleoside diphosphates.
It was known since the early work of Levene and Thannhauser,
and has also been confirmed more recently^^, that the acid treat-
ment of DNA leads to the formation of some 3',5'-diphosphates
of deoxycytidine and thymidine. The mechanism of this reaction
and its significance were, however, left in doubt. Kinetic studies
on the breakdown of small oligonucleotides prompt the con-
clusion that the first release of nucleoside diphosphates (30 min
at 100° and 0.1 M H2SO4) is due to the liberation of those

* If in Fig. 8 the black symbols are taken to represent purines, the
apurinic acid yielded by this sequence would contain deoxyribophosphate
units in positions 1, 3, 8, 9, and 10.

References p. 98

92 NUCLEIC ACIDS AND NUCLEOPROTEINS

pyrimidine derivatives that were flanked on both sides by purine
nucleotides, whereas more prolonged treatment affects poly-
pyrimidine stretches, with the glycoside fission of cytidine more
readily brought about than that of thymidine. It was possible to
develop methods for the quantitative estimation of the diphos-
phates and other small fragments of degradation. Table 22 gives
a selection of some of the results. It is clear that this sort of study
may be a powerful tool in bringing out variations in the nucleo-
tide arrangement of different nucleic acids; it is possible to
distinguish between specimens that cannot be distinguished by
the total analysis of their constituents. It may be concluded that
the nucleic acids listed in Table 22 represented chains consisting
predominantly of polypurine and polypyrimidine units and that
the arrangement of constituents was far from random.

5. PENTOSE NUCLEIC ACIDS AND NUCLEOPROTEINS

In glancing, more briefly, at the composition of various pentose
nucleic acids^^ one has to state that nothing much could have
been stated as long as isolated preparations were compared. The

TABLE 22

DEOXYCYTIDINE AND THYMIDINE 3 ',5 '-DIPHOSPHATES FROM DNA*

<^0.65 ^0.7 ^t Oq.Io ^1.0 ^''2.6

Molar ra^io^in DNA ^ ^^ ^ ^^ ^3^ ^ ^^ ^ ^^ ^ ^3

mole%'o?totalT 23.5 12.8 15.4 15.9 19.9 16.0

pCp**
mole % of total C

Molar ratio
pTp/pCp

10.8 6.00 9.70 9.20 6.30 14.0

2.24 2.68 2.16 2.58 4.77 1.80

* 30 min, 100°, 0.1 M H2SO4; DNA preparations from ox = O, man =
M, Arbacia = Ar. Subscripts: t = total DNA; the others indicate NaCl
molarity at which fraction was obtained.
** pTp and pCp denote the 3 ',5 '-diphosphates of thymidine and cytidine.

THE MEANING OF REGULARITIES 93

more rigorously purified the specimens were, the less constant
was their composition, even when material from the same source
was compared. In Table 23, a few examples have been chosen from
our own work"^"^' ^^. A complete compilation of analytical results
has appeared recently^^. No generally applicable regularities in
composition seem to impose themselves from such a survey.

TABLE 23

COMPOSITION OF SEVERAL PNA PREPARATIONS FROM YEAST AND LFVER*

Ade-

Giia-

Cyti-

Uri-

nylic

nylic

dylic

dyl ic

Source

No.

acid

acid

acid

acid

PulPy

6-Aml6-K

Yeast

1

30

30

19

21

1.5

1.0

2

30

28

22

20

1.4

1.1

3

26

28

22

24

1.2

0.9

4

26

27

21

26

1.1

0.9

5

21

30

23

26

1.0

0.8

Pig liver

1

16

38

33

13

1.2

1.0

2

19

36

29

16

1.2

0.9

* In moles per 100 moles nucleotide.

It was only after Dr. Elson in our laboratory undertook the
study of the ribonucleotide distribution in extracts of the total
PNA without further purification^^"^^ that one striking regularity
emerged^^' ^^. Whereas none of the other regularities listed before
for DNA applied to PNA in a general fashion, one persisted with
very few exceptions, namely, the equality of constituents having
6-amino groups (adenylic and cytidylic acids) and of those carry-
ing 6-keto groups (guanylic and uridylic acids): 6- Am = 6-K.
(Compare the frequency distribution of nucleotide ratios in 104
PNA samples^^ illustrated in Fig. 9.) I show some typical results
in Table 24 compiled from several recent papers^^' ^^' ^^' ^*.

6. THE MEANING OF REGULARITIES

I hope I have shown that there is one regularity that almost all
nucleic acids, deoxypentose and pentose nucleic acids, appear to

References p. 98

94

NUCLEIC ACIDS AND NUCLEOPROTEINS

have in common, namely, the presence in them of an equal num-
ber of nucleotides carrying an amino group in 6-position and of
those having a 6-keto group. Clearly, if this contention is correct
there must exist an agent imposing this regularity on the poly-

A + U
G + C

40

-

J 6Am

- *

- j

A/U
6/C

FREQUENCY DISTRIBUTIONS OF

NUCLEOTIDE RATIOS IN TOTAL

PNA

104 samples

I j-a.

L

-.i

L-T-H

ABSCISSAS' VALUE OF RATIO
CRDINATES: FREQUENCY OF
OCCURRENCE

0.5

1.0 1.5 2.0

0.5 1.0

1.5 2,0

arrow indicates mean

Fig. 9. Frequency distribution of various nucleotide ratios in 104 PNA

analyses, based on the data of Elson and Chargaff^s. The symbols are

explained in the footnote on page 84.

TABLE 24

NUCLEOXroE RATIOS IN PENTOSE NUCLEOPROTEINS

Source

PulPy

{A + U)/{G + Q

6-Am/6-K

Ox liver

0.80

0.63

1.04

Rat kidney

0.96

0.66

1.00

Cytoplasm of rat liver and

kidney

0.99

0.62

0.93

Paracentrotus lividus.

eggs and embryos

1.08

0.77

0.99

Wheat germ

1.07

0.76

1.00

Yeast

1.00

1.12

0.93

Escherichia coll

1.18

0.87

1.00

Mycobacterium phlei

1.06

0.73

0.93

Azotobacter vinelandii

1.21

0.78

1.00

THE MEANING OF REGULARITIES 95

nucleotide chains. It must select, by attracting or by rejecting,
those components that fit, out of the surplus of innumerable me-
tabolites being offered in the course of the life of the cell. The
screen, the sieve, the template: how many mechanomorphic con-
trivances could one not think of! I believe, however, that this is a
riddle that will not be solved through allegory. It is quite possible
that our biochemical thinking will have to acquire an entire new
dimension before we shall stop talking complete nonsense. But I
hope I shall not be accused of preaching metabiochemistry.

There will be many among us who, having learned from the
often experienced defeat of easy or premature generalizations in
biochemistry, are allergic to attempts to bring order to the seem-
ing chaos of the living cell. They may rightly ask: Is there any
sense in looking for regularities in composition, such as I have
discussed before; to apply interior decoration to a continuum?
They are, of course, correct in protesting against the dictatorial
sway of half-baked hypotheses. But there will always be a time
in the natural sciences — and it will always be too early — when a
summary, a provisional and tentative summary, must be drawn
up. As long as we realize that the experimental sciences operate
under an unwritten statute of limitations, no harm will result.

In the case of the nucleic acids it is particularly important to
ascertain whether any unifying principles governing their con-
struction can be discerned. They are considered by many people
to be the carriers of biological information. This rather ill-defined
notion views the DNA as the holder of the majority vote in the
transmission of hereditary properties and the RNA as the Execu-
tive Vice-President in charge of protein production. Be that as it
may — and there may be some surprises around the corner — bio-
logical information must be preserved; it must be transferred;
and, in the case of mutations, it must be changed. An incredible
amount of male handicraft is being wasted on information models.
But if there is any meaning to the concept of biological infor-
mation, the existence of regularities of the sort pointed out by me
before may furnish us the only chemical clue to systems through
which information is preserved or — and this is more important —

References p. 98

96 NUCLEIC ACIDS AND NUCLEOPROTEINS

transmitted. Compositional regularities in themselves are not suf-
ficient for an explanation; but they will, I believe, have to be
taken into account in any eventual solution of this problem. You
may not be able to reconstruct an animal from its tracks; but you
may succeed in identifying it through its footprints.

7. REMARKS ON NUCLEOPROTEINS

During my entire talk this evening, there has loomed in the
background, without ever making a real appearance, the second
half of my title, the nucleoproteins. Immanently, they must be
present in all our thinking on nucleic acids, for there is little evi-
dence that nucleic acids occur in the free state in the living cell.
If they ever disrobe, it is only temporarily; for the most part they
appear dressed decently in a coat of protein. Why not, therefore,
consider the whole thing, the conjugated proteins; the more so
since there is much evidence of the close connection between
nucleic acid and protein synthesis? In reality, we are not ready
for it despite the merry voices of innumerable biochemical boy
scouts who have pitched their hasty tents all over the field.

But it is probably not an accident that nucleic acids and
proteins are so closely associated in nature. It is hard to believe
that the proteins, in all their variety, have only the function of
an ornament or wrapping material. Even in the plant viruses, I
do not find it easy to entertain the Cytherean vision of a nucleic
acid resting in a merely protective shell of protein.

I have no time to go into all that and can only refer to a few
recent discussions^^- *^-^'^. But I should like to repeat one sugges-
tion^^, namely, that the regularity in PNA, and perhaps also in
DNA, of which I have spoken, 6-Am = 6-K, could be imposed
by an extended polypeptide chain bonded to two polynucleotide
chains so that each peptide carbonyl is linked by a hydrogen bond
to the 6-amino group of adenine or cytosine and each peptide
amino group to the 6-keto group of guanine or uracil. It is rather
unhkely that such nucleoproteins will be found easily^^; the en-
forcement of this stoichiometry may be an essentially dynamic

NUCLEOPROTEINS 97

process that cannot be tested by the static means of isolation.
Other ways may have to be sought.

If I have touched so Ughtly upon so few aspects of an important
problem, I must apologize; and I must apologize even more for
dealing only with matters with which my colleagues and I have
been directly concerned. But what sense would there have been in
trying to outrun a mastodon in two volumes?^ What such a treat-
ment loses in validity, it may gain in intensity. For if there is one
thing that I have learned during a life in science it is that even in
the tiniest splinter there may mirror itself an entire world.

References p. 98

98 NUCLEIC ACIDS AND NUCLEOPROTEINS

REFERENCES

1 E. Chargaff and J. N. Davidson (Eds.), The Nucleic Acids: Chemistry

and Biology, Vols. I and II, Academic Press, New York, 1955.

2 J. N. Davidson and E. Chargaff, in E. Chargaff and J. N. Davidson

(Eds.), The Nucleic Acids: Chemistry and Biology, Vol. I, Academic
Press, New York, 1955, p. 1.

3 E. Chargaff and E. Vischer, Ann. Rev. Biochem., 17 (1948) 201.

4 E. Chargaff, Experientia, 6 (1950) 201. [See Chapter 1 of this book.]

5 E. Vischer and E. Chargaff, J. Biol. Chem., 168 (1947) 781.

6 E .Vischer and E. Chargaff, J. Biol. Chem., 176 (1948) 703.

7 E. Chargaff, B. Magasanik, R. Doniger and E. Vischer, J. Am.

Chem. Soc, 71 (1949) 1513.

8 B. Magasanik, E. Vischer, R. Doniger, D. Elson and E. Chargaff,

J. Biol. Chem., 186 (1950) 37.

9 E. Chargaff, E. Vischer, R. Doniger, C. Green and F. Misani,

J. Biol. Chem., Ill (1949) 405.

10 E. Vischer, S. Zamenhof and E. Chargaff, J. Biol. Chem., Ill (1949)

429.

11 E. Chargaff, J. Cellular Comp. Physiol, 38, Suppl. 1 (1951) 41.

12 E. Chargaff, Federation Proc, 10 (1951) 654. [See Chapter 2 of this

book.]

13 E. Chargaff, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. I, Academic Press, New York,
1955, p. 307.

14 E. Chargaff and R. Lipshitz, J. Am. Chem. Soc, 75 (1953) 3658.

15 E. Chargaff, R. Lipshitz and C. Green, J. Biol. Chem., 195 (1952)

155.

16 B. Gandelman, S. Zamenhof and E. Chargaff, Biochim. Biophys.

Acta, 9 (1952) 399.

17 M. E. HODES and E. Chargaff, Biochim. Biophys. Acta, 22 (1956) 348.

18 R. Lipshitz and E. Chargaff, Biochim. Biophys. Acta, 19 (1956) 256.

19 E. Chargaff, S. Zamenhof, G. Brawerman and L. Kerin, J. Am.

Chem. Soc, 72 (1950) 3825.

20 E. Chargaff, R. Lipshitz, C. Green and M. E. Modes, J. Biol. Chem.,

192 (1951) 223.

21 S. Zamenhof and E. Chargaff, J. Biol. Chem., 187 (1950) 1.

22 C. F. Crampton, R. Lipshitz and E. Chargaff, J. Biol. Chem., 206

(1954) 499.

23 E. Chargaff, C. F. Crampton and R. Lipshitz, Nature, 172 (1953)

289.

24 C. F. Crampton, R. Lipshitz and E. Chargaff, J. Biol. Chem., 211

(1954) 125.

25 P. Spitnik, R. Lipshitz and E. Chargaff, J. Biol. Chem., 215 (1955)

765.

26 E. Chargaff, Symposia Soc. Exptl. Biol., 9 (1955) 32. [See Chapter 3

of this book.]

REFERENCES 99

27 D. M. Brown and A. R. Todd, in E. Chargaff and J. N. Davidson

(Eds.), The Nucleic Acids: Chemistry and Biology, Vol. I, Academic
Press, New York, 1955, p. 409.

28 E. Chargaff and H. S. Shapiro, Exptl Cell Research, Suppl. 3 (1955)

64.

29 M. E. Hodes and E. Chargaff, Biochim. Biophys. Acta, 22 (1956) 361.

30 C. Tamm, M. E. Hodes and E. Chargaff, J. Biol. Chem., 195 (1952)

49.

31 C. Tamm and E. Chargaff, J. Biol. Chem., 203 (1953) 689.

32 C. Tamm, H. S. Shapiro and E. Chargaff, /. Biol. Chem., 199 (1952)

313.

33 C. Tamm, H. S. Shapiro, R. Lipshitz and E. Chargaff, J. Biol. Chem.,

203 (1953) 673.
3-i H. S. Shapiro and E. Chargaff, Federation Proc, 15 (1956) 352.

35 C. A. Dekker, a. M. Michelson and a. R. Todd, J. Chem. Soc,

(1953) 947.

36 B. Magasanik, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. I, Academic Press, New York,
1955, p. 373.

37 E. Chargaff, B. Magasanik, E. Vischer, C. Green, R. Doniger and

D. Elson, /. Biol. Chem., 186 (1950) 51.

38 B. Magasanik and E. Chargaff, Biochim. Biophys. Acta, 7 (1951) 396.

39 D. Elson and E. Chargaff, Phosphorus Metabolism, Johns Hopkins

Univ., McCollum-Pratt Inst., Contrib., 2 (1952) 329.

40 D. Elson, T. Gustafson and E. Chargaff, J. Biol. Chem., 209 (1954)

285.

41 D. Elson, L. W. Trent and E. Chargaff, Biochim. Biophys. Acta, 17
(1955) 362.

42 D. Elson and E. Chargaff, Nature, 173 (1954) 1037.

43 D. Elson and E. Chargaff, Biochim. Biophys. Acta, 17 (1955) 367.

44 A. Lombard and E. Chargaff, Biochim. Biophys. Acta, 20 (1956) 585.

45 J. Brachet, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. 11, Academic Press, New York,
1955, p. 475.

46 E. Chargaff, Rend. ist. lombardo sci., 89 (1955) 101. [See Chapter 4

of this book.]

47 R. D. HoTCHKiss, in E. Chargaff and J. N. Davidson (Eds.), The

Nucleic Acids: Chemistry and Biology, Vol. II, Academic Press,
New York, 1955, p. 435.

48 E. Chargaff, D. Elson and H. T. Shigeura, Nature, 178 (1956) 682.

CHAPTER 7

Nucleic Acids as Carriers of Biological Information"^

1. IS THERE A HIERARCHY OF CELLULAR CONSTITUENTS?

Our time is probably the first in which mythology has penetrated
to the molecular level. I read, for instance, in a recent article by
a very distinguished biologist:

". . . In the early phases of the molecular stage of evolution, only simple
molecules were formed . . . Later more complex molecules, such as
amino acids and perhaps simple peptides, were formed.

"In the more advanced phases of this period it is believed that there
appeared a molecule with two entirely new properties: the ability system-
atically to direct the formation of copies of itself from an array of
simpler building blocks, and the property of acquiring new chemical
configurations without loss of ability to reproduce. These properties, self-
duplication and mutation, are characteristic of all living systems and
they may therefore be said to provide an objective basis for defining the
living state.

"Evidence is accumulating that the nucleic acids of present-day
organisms possess these two properties, and it is perhaps no longer useless
to speculate that the first "living" molecule might have been a simple
nucleic acid, perhaps protected by an associated simple protein . . ."

Thus, what started cosmically with beautiful and profound
legends has come down to a so-called "macromolecule". If poetry
has suffered, precision has not gained. For we may ask whether
a model that merely provides for one cell constituent continually
to make itself can teach us much about life and its origins. We
may also ask whether the postulation of a hierarchy of cellular
constituents, in which the nucleic acids are elevated to a patriar-
chal role in the creation of living matter, is justified. I believe

* Reprinted with permission from "The Origin of Life on the Earth",
I.U.B. Symposium Series, vol. 1, Pergamon Press, London, 1959, p. 297.

HIERARCHY OF CELLULAR CONSTITUENTS 101

there is not sufficient evidence for so singling out this particular
class of substances.

We know that autarkic entities (cells and cell communities)
require the presence of a very large number of different com-
pounds, foremost among which are the ubiquitous plastic con-
stituents, namely: (1) nucleic acids (both of the deoxypentose
and pentose types, DNA and PNA); (2) proteins; (3) lipids;
(4) polysaccharides. Parasitic systems (viruses, phages) apparent-
ly require the presence of nucleic acids (DNA or PNA) and
protein. But such a generic statement probably is no more
meaningful than if we were to say that all machines consist of
iron, copper, nickel, etc. For neither the knowledge of quantity
nor that of quality alone can satisfy our inquiry; and for an
understanding of what is meant by organization an entirely new
dimension will have to be added: a dimension of which barely
the foundations are discernible at present. Moreover, to return
to the example of the machine, it is not likely that we could learn
much about the "origin of the automobile" from an inspection of
the parts of a present-day car; nor could such an examination
help us to decide whether there did not once exist an automobile
made of glass, though we may conclude that, at any rate under
present conditions, it would hardly have survived as the fittest
means of locomotion.

In any event, even if we commit the oversimplifying fallacy of
postulating a virus-like structure as the first "living" molecule,
we are still left to deal with a nucleoprotein: a singular that
conceals a multitude of possible structures. If scientific facts were
subject to a vote, the majority opinion would probably place the
nucleic acids at the top of the hierarchy and let the proteins fol-
low; something like this: DNA -> PNA -» protein. But in my
opinion it would be more honest to confess that we know very
little indeed about these things and to say that the road to the
future should not be uselessly cluttered up with shoddy, and
often entirely baseless, hypotheses. It may, however, not be un-
instructive to say a few words about the chemical connotations of
the concept of biological information.

References p. 108

102 NUCLEIC ACIDS AND BIOLOGICAL INFORMATION

2. ON BIOLOGICAL INFORMATION CHEMICALLY CONVEYED

A biochemist, when asked to consider this problem, probably
would first of all translate it into the concept of chemical speci-
ficity— with which I have dealt in more detail on several oc-
casions^-^— and think of a substance of an elevated molecular
weight, built of a number of chemically different monomers and
possessing chemical or physical features which are preserved
without change within the species, but which serve to distinguish
the particular substance from analogous ones produced by other
species. Specific polysaccharides, specific proteins, specific
nucleic acids are the type of compounds that will come to mind
immediately; and the immunological specificity exhibited by
many representatives of the first two groups perhaps is their most
striking property. The recognition of the existence of specific
nucleic acids required much more time; but their specificity, too,
can now be taken to be an established fact on both chemical and
biological grounds^-^. It is, indeed, only in the fourth group of
principal cell constituents, namely the lipids, that difficulties with
respect to the occurrence of species specificity still are en-
countered. As I have pointed out recently^, one way in which
the cell may be able to render lipids specific is in their arrange-
ment as prosthetic groups of lipoproteins. But there exist in-
dications that nervous tissue contains complex mucolipids of high
molecular weight which could very well exhibit species speci-
ficity 6.

A warning is, however, in order. We cannot yet answer the
question: How identical is identity? In considering cell repUcation,
we hke to assume that the complex and specific cell constituents
are reproduced so as to give entirely invariant structures, with the
hundreds or even thousands of different amino acids or nucleo-
tides always arranged in an unchanged sequence. Actually, very
little is known; the extent of play that nature may allow itself
cannot yet be gauged.

It is, in any event, safe to assume that the real specificity of a
given cell resides in the nature of its plastic constituents rather

BIOLOGICAL INFORMATION CHEMICALLY CONVEYED 103

than in that of its catalytic components, which latter probably are
a symptom rather than a cause. But there is a giant step from the
establishment of sequential specificity in proteins, nucleic acids,
polysaccharides to the formulation of the manner in which the
"information" stored in them is not only preserved, but also
transferred from one class of components to the other. The
polysaccharides and lipids cannot at all be fitted into what little
we know. As regards the nucleic acids and proteins, however,
it would appear to me safer, instead of postulating a hierarchy,
to envisage a triangular relationship, e.g., as follows:

1
DNA

/ \

PNA Protein

2 3

No arrows have been placed in this diagram, for their direction
may conceivably be different in different systems; so that, for
instance, in a plant virus the relation is between 2 and 3, in
bacteriophage duplication and in bacterial transformations
between 1 and 3, in autarkic systems between 1, 2 and 3. It is,
moreover, quite conceivable that the transfer of information often
can proceed in either direction and that the prevailing impression
that it invariably flows from the nucleic acids to the proteins is
mistaken. We may not yet have learned how to isolate suitable
protein preparations; I should not be surprised if it were found
eventually that both halves of a plant virus, the PNA and the
protein, may be "infective". It was, therefore, not without pur-
pose that I used the plural above in speaking of "life and its
origins".

One could, however, ask: Is th^e cell really nothing but a system
of ingenious stamping presses, stencilling its way from life to
death? Is hfe itself only an intricate chain of templates and
catalysts and products? My answer to these and many similar
questions would be No; for I beHeve that our science has become
too mechanomorphic, that we talk in metaphors in order to
conceal our ignorance, and that there are categories in biochemis-

References p. 108

104 NUCLEIC ACIDS AND BIOLOGICAL INFORMATION

try for which we lack even a proper notation, let alone an idea
of their outlines and dimensions.

Regardless, however, of whether we accept the currently fash-
ionable template hypothesis, we must assume that there exist
agents or complexes of agents in the cell that preside over the
selection and specific arrangement of the constituents of all cell-
specific polymers: proteins, nucleic acids, blood group substances,
bacterial polysaccharide antigens, etc. These agents must be able
both to preserve and to transfer whatever codes are stored in the
constituent sequence and the specific physical shapes of these
compounds. How these agents operate is not yet within our
power to describe; but there exists some evidence that the nucleic
acids, both DNA and PNA, are concerned in these operations.
It lies within the scope of these remarks to inquire whether there
are any outstanding features that are common to all nucleic acids,
for this may be of value for our understanding of the chemical
characteristics of such information systems. In dividing the rest
of the discussion under the headings of invariability, diversity,
and regularity, it will be apparent that the first two quahties
probably are common to all specific cell constituents, whereas the
third is a most unusual feature of the nucleic acids which ap-
parently is not shared by the other plastic cell substances.

3. INVARIABILITY OF NUCLEIC ACIDS

This is a property that is, often by tenet rather than by experi-
ence, held to be common to all macromolecular cell components.
How limited our knowledge is, I have already emphasized before;
but it is, in general, assumed that a given protein or specific
polysaccharide or nucleic acid is invariable in structure in a
particular species. Although, as I shall mention below, there
exists impressive evidence that the nucleic acids of a cell com-
prise an entire series of different individuals, the constancy,
with respect to characteristic base composition, of total deoxy-
pentose nucleic acids appears well established^; a constancy that
seems to apply to preparations from all tissues of the same

DIVERSITY OF NUCLEIC ACIDS 105

host' or from different variants of the same microbial species^.
As concerns pentose nucleic acids a decision appeared more
difficult for some time owing to the unsatisfactory nature of
isolated PNA preparations. When methods were developed per-
mitting the analysis of total PNA without previous isolation,
essentially in the form of nucleoprotein, it could, however, be
shown that the nucleotide composition of the total pentose nucleic
acid of a given cell also is nearly constant^- 1^. In more recent
experiments on the pentose nucleic acids of Azotobacter vine-
landii^^ and of E. coli^^ this remarkable invariability of PNA
could also be demonstrated, in the latter case even under a variety
of conditions, for instance, the presence or absence of simul-
taneous protein synthesis and with and without the addition of
an excess of a single nucleoside.

4. DIVERSITY OF NUCLEIC ACIDS

Long before detailed chemical analysis had become possible the
existence of many different proteins and polysaccharides had
been recognized, often through their antigenic or other biological
properties, characteristic physical qualities or some outstanding
chemical components. The specific enzymic properties of many
proteins also were noted very early. Direct biological tests are
not often applicable to nucleic acids. The activity of specific
deoxypentose nucleic acids in bacterial transformations is well
estabhshed. The demonstration of the exclusive role of nucleic
acids in virus growth and enzyme induction is, in my opinion, as
yet far from unequivocal. But on the whole there surely exists
enough evidence to speak of a diversity of nucleic acids on
biological grounds.

The chemical diversity of DNA, as shown by widely different
proportions of the constituent nucleotides in different specimens,
has, since the time of its first discussion^- 1^, been demonstrated
in very many instances^. Of more recent date is the discovery that
a total DNA preparation can be fractionated into a whole series
of differently composed, but regularly graded, fractions^^. The

References p. 108

106 NUCLEIC ACIDS AND BIOLOGICAL INFORMATION

distribution of these fractions in a given cell must, however, be
constant within narrow limits, as shown by the invariability of
total DNA mentioned above.

The variation in the composition of different PNA prepa-
rations is marked, though perhaps not quite as striking as in
DNA^^. In addition, there is some evidence of the existence of
different pentose nucleic acids in the same cell; e.g., the nuclear
and the cytoplasmic PNA of rat liver differ in composition^^. A
reliable chemical fractionation of PNA has, however, not yet
been achieved.

5. REGULARITY OF NUCLEIC ACIDS

As I have mentioned before, this is perhaps the most unusual
property distinguishing the nucleic acids from other cell-specific
high polymers. In saying this, I am not referring to the regular
position of the phosphodiester bridges connecting the nucleo-
sides in the nucleic acid chain: they are assumed to be generally
3 ',5 '-bridges; for the regularity of the links holding the monomers
together is common to most biological polymers. What is so
peculiar is the remarkable balance between the several constitu-
ents noticed in all deoxypentose and in almost all pentose nucleic
acids: a type of equipoise that I am not aware of ever having
been encountered in other mixed polymers that do not contain
simple repeating units.

There are several regularities of which three are characteristic
of only DNA^'^. They are: (1) The molar quantity of adenine
equals that of thymine. (2) The molar quantity of guanine equals
that of cytosine (+ methylcytosine). (3) The sum of the purine
nucleotides equals that of the pyrimidine nucleotides. The fourth
regularity, finally, applies to nearly all nucleic acids, DNA and
PNA^O; it is: (4) The molar sum of nucleotides carrying 6-amino
groups (adenylic, cytidylic acids) equals that of nucleotides having
6-keto groups (guanylic, thymidylic or uridylic acids).

In considering these regularities one must take into account that
the nucleic acids are devoid of any perceptible periodicity or of

CONCLUDING REMARKS 107

repeating sub-units larger than a mononucleotide. They are com-
plicated high polymers of a largely arrhythmic nucleotide sequence
which, however, does not appear to be fortuitous^^.

6. CONCLUDING REMARKS

It is inviting to assume that the special biological functions of
the nucleic acids are reflected in those chemical features that
distinguish them from other high-molecular cell components,
namely, in the unusual regularities in nucleotide composition
mentioned above. In which way these functions are exerted cannot
yet be said; but I already have pointed out before that in my
opinion the nucleoproteins rather than the separate moieties of
the conjugated proteins will eventually be found to be the oper-
ative units. A detailed inquiry into the mechanism of this
conjunction of protein and nucleic acid, let alone the construc-
tion of a reasonable model or the attempt at unraveUing the
information code of the cell, is premature; for no truly homo-
geneous nucleoprotein has yet been discovered of which it could
be said that the particular nucleic acid molecule contained in
it, with its unique nucleotide sequence, has given rise to the
particular protein, with its equally unique amino acid sequence,
or vice versa. In other words, we are still lacking the Rosetta stone
of biochemistry.

It is not profitable to speculate on the direction in which the
ancestral roles of the nucleic acids could be approached. But it
can be predicted that further investigations along the following
lines will be of value: (a) virus nucleoproteins; (b) mechanism of
enzyme induction; (c) mechanism of antibody production; (d)
role of priming substances in .the enzymic synthesis of high
polymers.

References p. 108

108 ^fUCLEIC ACIDS AND BIOLOGICAL INFORMATION

REFERENCES

1 E. Chargaff, Experientia, 6 (1950) 201. [See Chapter 1 of this book.]
^ E. Chargaff, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. I, Academic Press, New York,

1955, p. 307.

3 R. D. HoTCHKiss, in E. Chargaff and J. N. Davidson (Eds.), The

Nucleic Acids: Chemistry and Biology, Vol. II, Academic Press,
New York, 1955, p. 435.

4 J. Brachet, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. II, Academic Press, New York,
1955, p. 475.

5 E. Chargaff, in S. Graff (Ed.), Essays in Biochemistry, Wiley and

Sons, New York, 1956, p. 72. [See Chapter 5 of this book.]

6 A. Rosenberg and E. Chargaff, Biochim. Biophys. Acta, 21 (1956)

588.

7 E. Chargaff and R. Lipshitz, /. Am. Chem. Soc, 75 (1953) 3658.

8 E. Chargaff, Symposium sur le metabolisme microhien, Congr. intern.

biochim., 2. Congr., Paris, (1952) 41.

9 D. Elson, T. Gustafson and E. Chargaff, J. Biol. Chem., 209 (1954)

285.

10 D. Elson and E. Chargaff, Biochim. Biophys. Acta, 17 (1955) 367.

11 A. Lombard and E. Chargaff, Biochim. Biophys. Acta, 20 (1956) 585.

12 A. Lombard and E. Chargaff, Biochim. Biophys. Acta, 25 (1957) 549.

13 E. Chargaff, E. Vischer, R. Doniger, C. Green and F. Misani, J.

Biol. Chem., \11 (1949) 405.

14 E. Chargaff, C. F. Crampton and R. Lipshitz, Nature, 172 (1953)

289.

15 D. Elson, L. W. Trent and E. Chargaff, Biochim. Biophys. Acta, 17

(1955) 362.

16 H. S. Shapiro and E. Chargaff, Biochim. Biophys. Acta, 23 (1957)

451.

CHAPTER 8

First Steps towards a Chemistry of Heredity"^

1. INTRODUCTION

An observer of our biological sciences today sees dark figures
moving over a bridge of glass. We are faced with an ever ex-
panding universe of light and darkness. The greater the circle of
understanding becomes, the greater is the circumference of sur-
rounding ignorance. But as our sciences expand, their specific
gravity appears to become lower. Out of swimmers we have all
turned into floaters. A unified and consistent vision of nature
has become impossible in our days, at any rate for working
scientists. Ironically enough, the only universal scientists left are
the publishers of scientific books or the writers of science fiction.
Each science protects itself from its neighbors by a cordon of
slogans and catchwords; and fashion dictates whether this year
we are featuring enzymes or proteins or nucleic acids and whether
we wear the molecules long or short. New journals are bom
every day by Caesarean section performed by skilful publishers;
and as new disciplines are formed, so are new and mutually un-
inteUigible languages: a Tower of Babel made of paper.

The feeling of discouragement before the seemingly chaotic
wave of nature, before this self-duplicating cataract, is not new.
Long before the unjustly famous "Ignorabimus" of Du Bois-
Reymond, innumerable mournful voices could be heard, the most
intense, the most heart-rending, being that of Pascal; for here is

* Reprinted with permission from Intern. Congr. Biochem., 4th Congr.,
Vienna, Austria, 14 (1958) 21-35.

References p. 125

110 FIRST STEPS TOWARDS A CHEMISTRY OF HEREDITY

a Moses who must curse the Promised Land*. One later citation
may stand for many. When "in the late evening of an eventful
life" one of the last, and one of the greatest, of universal scien-
tists, Alexander von Humboldt, published his Kosmos, he had
sworn off the vitalism of his youth. He wrote in his introduction:

". . . Die Mythen von imponderablen Stoffen und von eigenen Lebens-
kraften in jeglichem Organismus verwickeln und triiben die Ansicht der
Natur. Unter so verschiedenartigen Bedingnissen und Formen des Er-
kennens bewegt sich trage die schwere Last unseres angehauften und jetzt
so schnell anwachsenden empirischen Wissens. Die griibelnde Vernunft
versucht muthvoll und mil wechselndem Gliicke die alten Formen zu zer-
brechen, durch welche man den widerstrebenden Stoff, wie durch mecha-
nische Constructionen und Sinnbilder, zu beherrschen gewohnt ist.

Wir sind noch weit von dem Zeitpunkte entfernt, wo es moglich sein
konnte alle unsere sinnlichen Anschauungen zur Einheit des Naturbe-
griffs zu concentrieren. Es darf zweifelhaft genannt werden, ob dieser
Zeitpunkt je herannahen wird. Die Complication des Problems und die
Unermesslichkeit des Kosmos vereiteln fast die Hoffnung dazu. Wenn
uns aber auch das Ganze unerreichbar ist, so bleibt doch die theilweise
Losung des Problems, das Streben nach dem Verstehen der Welter-
scheinungen, der hochste und ewige Zweck aller Naturforschung . . ."

A. V. Humboldt, Kosmos, 1 (1845) 67.

The continuous struggle between vitalists and their adversaries
was built on a solid rock of ignorance of what they really were
fighting about. The best scientific discussions often take place in
a thick mental fog. There can be little doubt: the ever renewed
waves of vitahsm have been repelled successfully; but so many
victories have left our science short of breath. We shall not be
searching here for a quaintly monistic "vital force". But just as
much philosophy was necessary to dispose of the "philosopher's
stone", much life will still have to flow past us, before we begin
to understand what Life is. For have we really come nearer to
the solution of the problem if the age-old question "What is
life?" is replaced by an inquiry into the meaning of the organi-
zation of the cell? And if we no longer accept the God of Laplace,
that most International of all Business Machines, are we so much
better off babbling of "feedbacks"? We laugh at La Mettrie's

* See Pascal on the "vanite des sciences" in Pensees (Brunschvicg, No. 67,
74, 144, 604).

CELLULAR SPECIFICITY HI

"L'homme machine"; but "La cellule machine" could be a con-
temporary title. The old anthropomorphism* of the sciences has
been replaced by a streamlined mechanomorphism, with some
sort of "templates" made of stainless protoplasm. But we may
ask: Was this change really beneficial?

"Lasciate ogni speranza" is, however, hardly a motto that
would appeal to the munificence of research foundations and
government agencies; and a moderate optimism, intimating more
than it ever promises outright, is the fashion of the day. We shall
not dare transgress it, though no immediate vistas of roseate
finality will be painted.

The title I have chosen for my talk is at the same time bold
and modest. It should indicate how great the goal is and how
far we are from it. But, as I recently said on another occasion^,
"there will always be a time in the natural sciences — and it will
always be too early — when a summary, a provisional and ten-
tative summary, must be drawn up. As long as we realize that the
experimental sciences operate under an unwritten statute of
limitations, no harm will result". I shall first consider in what
manner the phenomena of heredity may be subject to chemical
research and what could be said to be the chemical basis of cel-
lular specificity.

2. CHEMICAL BASIS OF CELLULAR SPECIFICITY

Let us start with a dictionary definition. "Heredity — 3. Biol.
The property of organic beings, in virtue of which offspring inherit
the nature and characteristics of parents and ancestors generally;
the tendency of like to beget like. (Often spoken of as a law of
nature.)" Thus the great Oxford^ Dictionary (Vol. V, p. 238);
and it is interesting to note that the term is not very old, the first
example of its use dating from 1863. It is the sum of all the
characteristics of a given cell which could be defined as cell
specificity. But even if we are agreed that life at one level must

* "II ne faut pas juger la nature selon nous, mais selon elle." Pascal,
Pensees (Brunschvicg, No. 457).

References p. 125

112 FIRST STEPS TOWARDS A CHEMISTRY OF HEREDITY

be all chemical, just as on another plane it is all physical, the
translation of the concept of specificity into the terms of our
science remains extremely difficult.

If we compare a Staphylococcus with an E. coli cell or a sea
urchin egg with a mammalian ovum, comparative biochemistry
would probably tell us that many of their metabolic reactions,
many of their enzyme systems, are quite similar. Yet, how dif-
ferent is the whole! It must be confessed that there does not yet
exist a good and all-embracing biochemical explanation of speci-
ficity. If we look at a cat and a mouse, we know that they are dif-
ferent— and they themselves certainly know it even better. But
when we start to analyze them, we shall find much that is similar
and so very little that is different. It is almost as if we tried to
get at the meaning of "time" by taking apart a clock. There can
be little doubt that comparative biochemistry is barely in its
beginning and that, if it has difficulty in defining the differences
between genera, let alone species, it is hardly equipped to deal
with the distinction of individuals^. The biology^ and immunology
of individuality^ are much further advanced. It would be nice if
this were merely another way of saying that we shall have to
descend into nature by a few more decimals. But I am not certain
whether a radical change in the direction of our efforts will not
be necessary.

In any event, a biochemist meditating on the chemical basis of
cellular specificity will be led directly to the concept of chemical
specificity. He will assume — and I believe correctly — that dif-
ferences in the observable behavior of cells must rest on differ-
ences in their chemical make-up and in the relative location of
the interlocking cellular elements within the meshwork of the
cell. As concerns the plastic constituents of the cell (proteins,
nucleic acids, polysaccharides, lipids), the existence of important
specificity characteristics within the groups of proteins and
polysaccharides had been revealed by immunological studies long
before organic chemistry had advanced to a point where chemical
differentiation became possible. It took even longer for the
chemical distinction of nucleic acids of different origin to become

ALPHABET OF THE CELL 113

feasible^. The lipids had to wait longest. Whereas the existence
of many different tissue lipoproteins had been recognized for
quite some time^, it is only recently that the occurrence of specific
lipid polymers is being discussed^- *.

Which types of compounds then, we could ask, are responsible
for the specific character of a cell? The answer is obvious. If we
assume that all chemical and physical features of the cell are
preserved without change within the species, but serve to distin-
guish the particular cell from that of other species, we must say
that it is the aggregate of all cellular constituents that defines
chemical specificity. But this does not mean that all substances
are equally decisive in maintaining hereditary continuity; some
may be causes, others merely symptoms. There is much reason
for the assumption that the directive influence on the mainte-
nance of cell specificity must be sought within the class of highly
polymerized plastic constituents, at any rate, the proteins and the
nucleic acids.

3. ON THE ALPHABET OF THE CELL AND SOME LANGUAGE
DIFFICULTIES

When a science approaches the frontiers of its knowledge, it
seeks refuge in allegory or in analogy. The latter attempt — and I
consider it preferable to the first — has enlisted the support of
modern disciplines, such as cybernetics or information theory^.
A subordinate analogy to the maintenance of cellular specificity,
but one, perhaps, more easily grasped, is that of the manner in
which communication is brought about through language. This
should not be taken to mean that I consider the cell as a system
of phonemes; for we have no right to introduce such concepts as
intelligence or mind into the working of the cell. But the com-
parison with communication through words will, at least, serve
to explain the meaning of a term much used in these days,
namely, biological information. I have dealt with this problem
before^^' ^^ and can be brief here. A very tentative definition of
biological information could describe it as the aggregate of all

References p. 125

114

FIRST STEPS TOWARDS A CHEMISTRY OF HEREDITY

the signals — summa cellulae — that preside over the maintenance
of the hereditary processes, ensuring the unaltered transfer of all
specificity characteristics of the mother cell to the daughter cells.
We could then ask, but without much hope of an immediate and
decisive answer: What is the content of biological information
and what are its carriers; how is it preserved and how is it trans-
ferred; and what are the agents of such information transfers? If
we knew the answers to these questions, we should have made
more than the first steps towards a chemistry of heredity. If, on
the other hand, we continue the criminal nonsense of filling our
world with ever increasing quantities of ionizing radiation, we
may soon be able to speak of the last steps towards a chemistry
of heredity.

Lipid Polysaccharide

Fig. 10.

As to the content of biological information, the answer will
depend upon whether the assumption I made a few minutes ago
proves correct, namely, that the nucleic acids and the proteins,
especially the conjugated proteins^-, are the principal, or even
the sole, agents directing the maintenance of cellular specificity.
In statements of this kind, allowance must be made for the neces-
sity that we shall look as stupid to our successors of 2058 as our
predecessors of 100 years ago look to us. In other words, we may
be able to run away from our past, but never from our future.

The other questions are even more difficult to answer, for they
refer to what may be called the flow of biological information;
and it is here that our real language difficulties begin. There must
exist in the cell a system of extremely rigid controls that supervise

BIOCHEMICAL SPACE 115

the inviolate reproduction of all relevant constituents; a system
presumably indispensable for the cell to maintain itself in a state
of specificity. But the search for the nature of these controls has
been largely unsuccessful. It has, however, led to the construction
of many, often pretty, models which lately, since they are now
made of transparent plastic, have gained much in astral beauty.

Our only refuge can be in vagueness, for instance, by pro-
posing some scheme of triangular relationships which may take
the childishly geometrical form shown in Fig. 10. This is, of
course, only one of many possible schemes for the interrelation
of the plastic cell constituents. No arrows have been placed quite
on purpose^^; and some of the soUd lines should probably be
broken lines or omitted entirely. The central role of ribonucleic
acid is based on what is now beginning to be known of the
metaboUc functions of certain conjugated ribonucleotides. One of
the main difficulties in all such schemes is that we do not yet
know whether the same agent is concerned with the selection and
alignment of the monomers and with the activation of their
linking into polymers.

4. BIOCHEMICAL SPACE

When we speak of the "flow" of biological information^^ or of
the "movement" of a given precursor, such as an amino acid or
phosphoric acid, into the various polymers^^- ^'^ found in the
several cellular elements, such as nuclei, mitochondria, micro-
somes, etc., we adhere subconsciously to a concept that, I believe,
is seldom formulated properly, namely, that of biochemical space.
It is probably a thankless job to estabUsh oneself as the Pindar
of negative attributes; but there are many things in nature that
are better described by a No than by a Yes. The cell, we might
say, is not a machine. It is not a mixture of soUds or even
crystals. It is not a system of solutions. Even in describing it as
an organized community of interfaces, of boundaries, we ap-
proach only a corner of the truth.

What would be necessary for an understanding of biochemical

References p. 125

116 FIRST STEPS TOWARDS A CHEMISTRY OF HEREDITY

space? We are all familiar with the four coordinates that define
the space-time continuum. It is curious that in biochemistry the
time coordinate is the one most easily defined, since through the
use of isotopic tracers the dating of a polymer often can be
achieved. But we know very little of the spatial conditions under
which the cell manufactures its constituents. It is even pos-
sible— and this should not be forgotten in all the talk about
''templates" — that we may be dealing with templates in time
rather than space and that as yet undefinable time sequences
regulate the replication of cellular polymers.

What I mean in speaking of the movement of precursors can,
perhaps, best be illustrated by an example relating to the for-
mation of protein and ribonucleic acid in rat liver^^- ^^ which is
shown in Table 25. It will be seen that in experiments of short

TABLE 25

UPTAKE in vivo OF RADIOACTIVE LEUCINE BY THE PROTEIN AND OF

RADIOACTIVE SODIUM PHOSPHATE BY THE RIBONUCLEIC ACID IN CYTOPLASMIC

FRACTIONS FROM RAT LIVER»

Relative specific activity^

Protein

Ribonucleic acid^

Fraction

Minutes after injection

5

20

5

20

120 1440

Cytoplasmic supernatant fluid 1

Microsomal ribonucleoprotein
particles'^ I 4

II 6

1

2
2

11
1

9 20
1 1

7 10 1
1 1 1

o The values are adapted from recent datai^, i4,

^ In each column the fraction having the lowest specific activity was taken

as 1.

^ The specific activities of the nucleic acid samples were computed through

the summation of the weighted specific activities of the 2' -f 3' nucleotides

produced by alkaUne hydrolysis.

^ This represents the microsome portion insoluble in deoxycholate; the

protein fraction I is insoluble at pH 13, fraction II insoluble at pH 6

(Ref. 15).

CODE-SCRIPT OF BIOLOGICAL HIGH POLYMERS 117

duration the protein precursor appears in the microsomal proteins
before it is found in the supernatant fraction, whereas the op-
posite is true of the ribonucleic acid precursor. I have condensed
the data and rounded off the figures for the sake of simplicity.
If, in addition to the supernatant fraction (Sp), the microsomal
portion soluble in deoxycholate (DS) and the two distinguishable
protein fractions of the deoxycholate-insoluble microsome portion
(RNP I and II) are surveyed, the movement of leucine into the
proteins can be described as follows: RNP II -> RNP I -^ DS
-> Sp. An inverse, but probably correlative, direction is found
for the movement of inorganic phosphate into the ribonucleic
acids, namely, Sp -^ DS -^ RNP. The importance of such find-
ings will emerge more clearly if we consider that not only the pro-
teins, but also the nucleic acids, of the different cell fractions may
be presumed to differ with respect to composition and function.

5. ON THE CODE-SCRIPT OF BIOLOGICAL HIGH POLYMERS

If you permit me to take a bird's-eye view — even if it is the view
of a very shortsighted bird — I could ask: What is the immediate
contribution that chemistry can make to an understanding of
heredity? (Please notice that both here and in the title of my talk
I used the word chemistry, not biochemistry.) I believe, I have
already indicated that in my opinion this contribution must be
sought in the complex of problems that has to do with the preser-
vation, the transfer and the flow of biological information, and
also with what could be called its geometry. We must gain a
clearer understanding of the manner in which the ability to send
the signals that direct the replication of macromolecules is embed-
ded in the very structure of the various high polymers of the cell.
I have often said that I do not believe that there exists a hier-
archy among the cellular constituents; so that we should not say
that, for instance, the nucleic acids or the proteins are more im-
portant than the lipids or the polysaccharides. But we can say
that the minimum of viable organization is represented by a
nucleoprotein, as in the viruses, and the minimum of signaling

References p. 125

118 FIRST STEPS TOWARDS A CHEMISTRY OF HEREDITY

potential by a deoxypentose nucleic acid, as in the transforming
principles. Neither the nucleoprotein nor the nucleic acid can, of
course, stand alone; they require the background of complete
cellular organization. But at least the first step is indicated to us:
we must attempt to read the code of the high polymers, first of
all of the nucleic acids and the proteins. That this cannot take the
shape of a legerdemain, of clever prestidigitation, is obvious.
Little will be gained if the products formed in an uncooked meat
extract or in a so-called homogenate are termed biosynthetic. But
the search for the chemical structure of the decisive polymers,
for the alignment, the sequence of their monomeric components
is open to us.

The sequential analysis of the proteins and the nucleic acids
will in itself not be sufficient; the less so, since no nucleoprotein
has so far been isolated of which it could be assumed that the
protein and nucleic acid moieties of the conjugated protein ex-
hibited a direct relationship as of cause and effect. "We are still
lacking the Rosetta stone of biochemistry^^" But such studies
may indicate the direction that our future efforts are to take. As
concerns the proteins, they will be discussed at this congress by
more competent voices than could be mine. But since it has come
to be that my laboratory has probably been among the first to
place the structure of the nucleic acids into the context of modern
biochemistry, you will permit me to devote most of the rest of my
talk to this subject.

As I mentioned before, there exists some evidence that the
nucleic acids, both the deoxypentose and the pentose nucleic
acids, are concerned in the operations that have to do with the
selection and the specific arrangement of the constituents of
many, if not all, cell-specific polymers, such as proteins, nucleic
acids, and perhaps even blood group substances and bacterial
and other polysaccharides, etc. What then, could we ask, are the
outstanding features that are common to all nucleic acids, but
absent from other specific cell substances? This consideration
may be of immediate importance for an understanding of the
chemical characteristics of the information systems to which I

CODE-SCRIPT OF BIOLOGICAL HIGH POLYMERS 119

have alluded before. In all our following discussions we must
keep in mind that the nucleic acids are devoid of any perceptible
periodicity or of recognizable repeating subunits. They are high
polymers of a largely arrhythmic nucleotide sequence which is,
however, not random.

Until our first results were pubHshed in 1948 and 1949
(Refs. 16, 17) the nucleic acids, within their two known sugar
categories, were considered as analytically indistinguishable; the
problems posed by their analysis were not even seriously dicussed.
Now, the great diversity in the composition of the nucleic
acids is recognized generally. I limit myself here to two exam-
ples (Tables 20, p. 87, and 26). Although occasionally nucleic
acids from widely distant cellular sources are encountered that
cannot be distinguished by constituent analysis, it is safe to
assume that different species never produce identical nucleic
acids.

But do the same cells always produce the same nucleic acids;
in other words, do the nucleic acids exhibit the property of

TABLE 26

NUCLEOTIDE RATIOS IN RIBONUCLEOPROTEINS*

Source

PulPy

{A + U)l
iG + C)

6-Aml
6-K

Ox liver

0.80

0.63

1.04

Rat kidney

0.96

0.66

1.00

Cytoplasm of rat liver and kidney

0.99

0.62

0.93

Nuclei of rat liver

0.66

0.81

1.06

Paracentrotus lividus, eggs and embryos

1.08

0.77

0.99

Wheat germ

1.07

0.76

1.00

Yeast

1.00

1.12

0.93

Mycobacterium tuberculosis (bovine)

1.06

0.65

0.98

Mycobacterium phlei

1.06

0.73

0.93

Azotobacter vinelandii

1.21

0.78

1.00

Escherichia coli

1.27

0.83

1.06

* Taken from previous papersi8-22^ Abbreviations: A, adenylic acid; G,
guanylic acid; C, cytidylic acid; U, uridylic acid; Pu, purine nucleotides;
Py, pyrimidine nucleotides; 6-Am (6-amino nucleotides), A + C; 6-K
(6-keto nucleotides), G + U.

References p. 125

120 FIRST STEPS TOWARDS A CHEMISTRY OF HEREDITY

invariability within the species? This is apparently the case with
respect to the total deoxypentose nucleic acid of a species, and
also with respect to the total pentose nucleic acid of unicellular
organisms. Many examples can be found in the literature-^- ^^' ^^.
The two properties I have mentioned, namely, diversity in
composition with regard to different species and invariability of
composition within the same species, are not unique with the
nucleic acids; they are probably shared by all highly polymerized
cell constituents, although it may be possible to distinguish
between a direct, gene-controlled specificity (nucleic acids,
proteins) and a reflected specificity (polysaccharides, lipids)
which latter is mediated through specific enzymes that are them-
selves subject to control by the genes. But the nucleic acids
show, in addition, a third, most unusual characteristic which
distinguishes them from other cell-specific high polymers. I am
referring to the well-known compositional regularities^ ^ i^. These
are, in the case of deoxyribonucleic acids: (a) the molar sum of
the purine nucleotides equals that of the pyrimidine nucleotides;

(b) the molar proportion of adenine equals that of thymine;

(c) the molar proportion of guanine equals that of cytosine
(+ methylcytosine); (d) there is the same number of 6-amino
nucleotides (adenylic, cytidylic acids, etc.), as of 6-keto nucleo-
tides (guanylic, thymidylic acids). For most of the ribonucleic
acids only the last regularity holds: the molar sum of adenylic
and cytidylic acids equals that of guanylic and uridylic acids.

This remarkable balance between the several nucleic acid
constituents represents a type of equipoise that I do not believe has
ever been encountered in other mixed polymers that do not contain
simple repeating units. That some of the principal beneficiaries
of these discoveries have acknowledged them with great reluctance
is a phenomenon not unusual in the history of the sciences.

One of the greatest discoveries in biology that I have had the
good fortune to witness, namely, the recognition by Avery* and

* In the two recent textbooks of genetics that I have among my books
the name of Avery does not appear. "Habent sua fata" not only the
"libelir but also what they leave out.

CODE-SCRIPT OF BIOLOGICAL HIGH POLYMERS 121

his collaborators of a specific deoxypentose nucleic acid as the
agent responsible for bacterial transformation^^, has placed the
nucleic acids, or at any rate the deoxypentose nucleic acids, very
near the gene of the geneticist. If then these nucleic acids are,
indeed, the genetic determinants, or form part of these systems,
one could ask in what manner their biological functions can be
translated into chemical structure, for translation there must be.
Two possibilities suggest themselves. (1) Each nucleic acid chain,
containing, perhaps, 10,000 or more nucleotides in a row, carries
many single determinants. But then each gene could hardly be
represented by a simple oligonucleotide sequence, say, a tri- or
tetranucleotide, since it is highly improbable statistically that such
sequences could be unique. (2) Less implausibly, each gene is
represented, or determined, by a separate nucleic acid which is a
distinct chemical entity, the quite well established constancy of
composition of the total deoxyribonucleic acid of a cell being
merely the statistical consequence of the unchanging character
of the cell. In this case, a genetic determinant, defined by a
characteristic chemical structure, would be unique if no two
nucleic acid molecules, within the same cell, were identical. I
have mentioned this possibility of nonidentity before^^, and it
actually is not such an unreasonable assumption: a haploid
nucleus containing, let us say, 3 • 10— ^^ g of deoxypentose nucleic
acid^^ would carry 300,000 molecules of nucleic acid of molecular
weight 6-10^ and correspondingly fewer if a higher molecular
weight is taken. The great difficulty in such speculations is to
reconcile the position of people for whom the gene is like a state
of mind with that of others for whom it is a substance. Who
would ask himself how many memories he has of the same event?
It has, in fact, been possible, first in our laboratory^^, to frac-
tionate total deoxyribonucleic acid preparations into a series of
regularly graded fractions of decreasing guanine and cytosine,
and rising adenine and thymine contents, but all showing the
pairing principles that I have mentioned before. The method
used by us was the fractional dissociation, under special con-
ditions, of nucleohistones or histone nucleates. If a preparation

References p. 125

122 FIRST STEPS TOWARDS A CHEMISTRY OF HEREDITY

IS

5 °

S 2

S o

Q f^

^l

o ^

a-

ex

ex

C ~ s^ O -^

■|fi||.2|

^

s

?s

^.^

^

^

^

^

«u

"o

.c

s

^

w

o

•2

5

1

2i

1^

•S

5

5

i

§

e?

"^

o

^

s

<4J

-«

^

00

'O

<L)

On

T— 1

as

I

Tf

»o

s§

ON

od

'^

i2 .

O o

•C c«^

o\

q

Dh 1

o

T— 1

"O -J**

en

C*^

a
o

i^

"cd

(J

U

c

«4-l

c

,o

60 ii

<

z

__^

-^-s

1^

eS

■2

vo

<A

o

(N

a-g

1

"c3

fel

ffi

^

0)

CODE-SCRIPT OF BIOLOGICAL HIGH POLYMERS 123

of deoxyribonucleic acid from mammalian tissue really com-
prises between 100,000 and 300,000 different individuals, the
isolation of "pure" compounds is unattainable; and it is even dif-
ficult to conceive of the criteria that would apply. For my part, I
must say that the significance of the existence of many deoxy-
ribonucleic acid fractions is by no means clear. There is still, I
believe, a great mystery hidden in the structure of the nucleic
acids.

From much of what I have said before it must have been
evident how important it would be to gain an insight into the
order of alignment of the nucleotide monomers in a nucleic acid
chain; for the "information" coded into the nucleic acids must
find its chemical expression in the specific arrangement of the
purines and pyrimidines on the sugar phosphate backbone, that
is, in the nucleotide sequence. I have no time here to discuss our
earUer attempts, such as the stepwise enzymic degradation or the
study of the formation and the breakdown of the apurinic acids^^,
which had served to emphasize the arrhythmic character of the
polynucleotide chain. But I should like to say a few words about
the results of our recent work which is based on the method of
differential distribution analysis developed with Shapiro-^. This
procedure, which consists in the analysis of the breakdown
products formed by the controlled, stepwise acid degradation of
deoxyribonucleic acids, makes use of the great lability towards
acid of the purine glycosides. If we consider a segment of a
polynucleotide chain, we may assume that it will contain
"solitary" pyrimidine nucleotides flanked by purine nucleotides
and vice versa as well as "bunched" stretches of varying length
consisting exclusively of purine or of pyrimidine nucleotides
(compare Fig. 8, p. 90). The first stage of acid degradation, to
which we shall limit ourselves today, results in the removal of
the purines followed by chain rupture in such a manner as to
liberate the "solitary" pyrimidine nucleotides in the form of the
nucleoside 3 ',5 '-diphosphates. These, as well as other more com-
plex breakdown products, can be determined quantitatively,
furnishing a characteristic feature of the particular deoxyribonu-

References p. 125

124 FIRST STEPS TOWARDS A CHEMISTRY OF HEREDITY

cleic acid chain which goes beyond what mere constituent
analysis can do. I show in Table 27 a small selection of results
obtained with three preparations that are indistinguishable ana-
lytically.

It would take me too long to discuss all the implications of this
work, but I may state the principal findings. (1) Analytically in-
distinguishable nucleic acid specimens from different cellular
sources exhibit entirely different sequence characteristics. (2) In
no case is the nucleotide sequence that to be expected from a
random arrangement of monomers. (3) At least 70% of the
pyrimidines, and therefore 70% of the purines, occur in tracts of
at least three of one kind, i.e., three or more purines followed by
three or more pyrimidines, etc. This has a curious consequence
in that it restricts the number of possible permutations, though
this figure is still enormous. For example, a calf thymus deoxy-
ribonucleic acid consisting of 10,000 nucleotides (molecular
weight 3-10^) could exist in more than 10^^^^ isomers. With the,
admittedly oversimplifying, assumption that it consists of trinu-
cleotide tracts of one kind, i.e., adenine-adenine-adenine, etc., the
figure for permutational isomers drops from 10^^^*^ to 10^^^^.
But you will agree even so that the nucleic acids may, in their
sequences, contain enough code-scripts to supply a universe with
information.

But do we have enough intelligence to read the information
thus offered to us? Not yet, at any rate. There are people who
suffer from giddiness when watching the Milky Way for a long
time on a clear night. Such people should not look into the cell
nucleus. Where else is there so much in so little? And beyond the
range of our perception, this millimicrocosm does not end, but
there begin galaxies of invisibility. Still, I say to those who come
after us: Scibitis, you shall know! But only then will the real
search begin. In a short time, in a very short time, our science
has built a ladder into the heavens; but the first two or three
rungs are missing. Let us hope that they will be found in the
next hundred years.

FIRST STEPS TOWARDS A CHEMISTRY OF HEREDITY 125

REFERENCES

1 E. Chargaff, Harvey Lectures, 1956-57, 52 (1958) 57. [See Chapter 6

of this book.]

2 R. J. Williams, Biochemical Individuality, Wiley and Sons, New York,

1956.

3 L. LoEB, The Biological Basis of Individuality, C. C. Thomas, Balti-

more, Md., 1945.

4 P. B. Medawar, Harvey Lectures, 1956-1957, 52 (1958) 144.

5 E. Chargaff, Experientia, 6 (1950) 201. [See Chapter 1 of this book.]

6 E. Chargaff, Advances in Protein Chem., 1 (1944) 1.

7 A. Rosenberg and E. Chargaff, Biochim. Biophys. Acta, 21 (1956)

588.

8 A. Rosenberg and E. Chargaff, J. Biol Chem., 232 (1958) 1031.

9 H. QuASTLER (Ed.), Information Theory in Biology, University of Il-

linois Press, Urbana, 111., 1953.

10 E. Chargaff, Rend. ist. lombardo sci., 89 (1955) 101. [See Chapter 4

of this book.]

11 E. Chargaff, in The Origin of Life on the Earth, U.S.S.R. Academy

of Sciences, Moscow, 1957, p. 188. [See Chapter 7 of this book.]

12 E. Chargaff, in S. Graff (Ed.), Essays in Biochemistry, Wiley and

Sons, New York, 1956, p. 72. [See Chapter 5 of this book.]

13 H. T. Shigeura and E. Chargaff, Biochim. Biophys. Acta, 24 (1957)

450.

14 H. T. Shigeura and E. Chargaff, J. Biol. Chem., 233 (1958) 197.

15 E. Chargaff, D. Elson and H. T. Shigeura, Nature, 178 (1956) 682.

16 E. ViscHER AND E. Chargaff, J. Biol. Chem., 176 (1948) 715.

17 E. Chargaff, E. Vischer, R. Doniger, C. Green and F. Misani, J.

Biol. Chem., Ill (1949) 405.

18 D. Elson and E. Chargaff, Biochim. Biophys. Acta, 17 (1955) 367.

19 R. Lipshitz and E. Chargaff, Biochim. Biophys. Acta, 19 (1956) 256.

20 A. Lombard and E. Chargaff, Biochim. Biophys. Acta, 20 (1956) 585.

21 A. Lombard and E. Chargaff, Biochim. Biophys. Acta, 25 (1957) 549.

22 T. TsuMiTA and E. Chargaff, Biochim. Biophys. Acta, 29 (1958) 568.

23 E. Chargaff, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. I, Academic Press, New York,
1955, p. 307.

24 O. T. Avery, C. M. MacLeod and M. McCarty, J. Exptl. Med., 79

(1944) 137.

25 E. Chargaff, Symposia Soc. Exptl. Biol., 9 (1955) 32. [See Chapter 3

of this book.]

26 R, Vendrely, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. II, Academic Press, New York,
1955, p. 155.

27 E. Chargaff, C. F. Crampton and R. Lipshitz, Nature, 111 (1953)

289.

28 H. S. Shapiro and E. Chargaff, Biochim. Biophys. Acta, 23 (1957)

451; 26 (1957) 596, 608.

CHAPTER 9

The Problem of Nucleotide Sequence in
Deoxyribonucleic Acids'^

1 . INTRODUCTION

Now that some of the most tremendous problems in biochemistry
have, it is said, been solved successfully — such as the nature of
the genetic material, its structure and mode of replication, and
even its biosynthesis — our time, tired of so much exertion, may
take a deep breath and then, perhaps, decide that it all has to be
done over again. I should not dare contradict, for I have always
been more impressed by the enormity of the gap between claim
and achievement than by the magnitude of the former.

In any event, we have set ourselves a much more modest, but
by no means easy, task, namely, the elucidation of some of the
aspects of the primary structure of deoxyribonucleic acids,
especially in regard to the arrangement of the nucleotide con-
stituents. In this connection, and since this talk is given in Stock-
holm, I may be permitted to recall that it was here, and also in
Uppsala, that I had the first opportunity to review our original
observations on deoxyribonucleic acid. This is what I said in
1949 and what was printed a few months later^.

". . . The deoxypentose nucleic acids extracted from different species thus
appear to be different substances or mixtures of closely related sub-
stances of a composition constant for different organs of the same species
and characteristic of the species.

* This chapter has been revised and expanded from a previous publica-
tion reprinted with permission from T. W. Goodwin and O. Lindberg
(Eds.), Biological Structure and Function, Vol. I, Academic Press, London
and New York, 1961, p. 67.

CONCEPTUAL BASIS 127

"The results serve to disprove the tetranucleotide hypothesis. It is,
however, noteworthy — whether this is more than accidental, cannot yet
be said — that in all deoxypentose nucleic acids examined thus far the
molar ratios of total purines to total pyrimidines, and also of adenine to
thymine and of guanine to cytosine, were not far from 1."

You will recognize here, among other things, the first statement
of the well-known pairing principles.

Even the earliest observations on the nucleic acids, which
showed the existence of remarkable similarities and, at the same
time, of outstanding differences in the distribution and, there-
fore, the sequence of the constituent monomers, the nucleotides,
made it appear of great interest to learn something about the
structural principles that came into play. I shall first discuss some
of the basic concepts that must be considered.

2. REMARKS ON THE CONCEPTUAL BASIS OF SEQUENCE ANALYSIS

a. The problem of homogeneity

What determines the composition of a given biological polymer
has been much discussed in the recent past, at any rate in regard
to proteins and nucleic acids. Almost no attention has been paid,
as has been pointed out recently^, to the presumably equally rigid
control of the composition of other cell-specific polymers, such
as the polysaccharides or even certain macromolecular lipids.
The usual expedient, in which I wish I could concur with greater
enthusiasm^, has been the formulation of some sort of template
which obviously makes up in versatility for what it lacks in con-
creteness. This Disneyland of templates and pools and feedbacks,
presided over by never-smiling augurs, will make us the laughing
stock of later times.

What determines the size of a given biological polymer has, on
the other hand, been neither determined nor even much discussed.
The cell will in many instances contain, or at least it will be
made to yield, compounds belonging to the same class, but very
much different in size. Proteins, the best investigated group of
cellular macromolecules, vary in size over a considerable range.

References p. 159

128 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

from something like 12,000 to many millions^. Deoxyribonucleic
acid preparations appear, in contrast, to fall into two principal
groups of molecular weight: (a) of 6 to 8 million; (b) of 12 to 16
million^.* Such distinctions are probably not of great heuristic
value, since the cell, which is not simply a bag containing many
chemicals of different size and quality, must impress on its com-
ponents a pattern of multiple associations which it is not possible
even to define at present.

Dissimilarity in size is, however, less of a problem for the
student of the chemistry of deoxyribonucleic acids than is the
possible heterogeneity of his specimens as regards their com-
position and sequence characteristics. It is, in fact, possible to
divide a preparation of total deoxyribonucleic acid into a series of
fractions of graded composition (starting with a, sometimes ex-
treme, GC type^ and going to a very marked AT type) which,
however, all still exhibit the pairing regularities^; but the real
meaning of such a fractionation is far from clear. It could be,
as was pointed out by us^, that these fractionated preparations
actually represent some form of sub-units of a large aggregate,
though it is not easy to describe the nature of the links that are
disrupted repeatably during the mild fractionation procedure.
In this view, in its extreme form, the nucleus of a given species
could conceivably contain only one type of deoxyribonucleic acid.
At the other extreme would have to be placed the view that a
given nucleic acid preparation may comprise an entire spectrum
of differently constituted individuals so that no two nucleic acid
molecules within the same nucleus would be entirely identicaP^.
That neither of these opposite extremes can as yet be disavowed
entirely demonstrates the deep cleft that still exists between
metabiological assertions and scientific facts.

* An interesting exception — a polydeoxyribonucleotide of mol. wt. 10,000
to 12,000 — has been described recentlyf"' as a component of crystalline cyto-
chrome bey. It is noteworthy that this small polymer of only about 40
nucleotides still exhibits the pairing principlesi mentioned above.

CONCEPTUAL BASIS 129

b. The problem of macromolecular structure

It would, of course, be good if the investigator of the structure
of a deoxyribonucleic acid knew whether he was deaUng with one
chain or with two^^ or even more^- complementary chains; but at
the present, little advanced, state of our knowledge of the nucleic
acids this is not essential. The value of any information on the
sequential characteristics of the nucleic acid is not diminished if
the complementary chains are considered to be joined at one end
or even at both ends, as may well be the case, so as to constitute
an uninterrupted sequential progression. I am not aware of the
demonstration of distinct end groups in undegraded deoxy-
ribonucleic acid.

The maintenance of sequence integrity of the preparations to
be examined is, on the other hand, a problem of great importance.
If — quite apart from the secondary valence forces supporting the
architecture of the secondary or tertiary structures of the deoxy-
ribonucleic acids — there really exist weak links in their primary
structure^^, it may be essential to avoid the rupture of the latter,
since otherwise valuable features of sequential arrangement may
be lost. This appUes, of course, even more to the avoidance of
chemical or enzymic degradation during the isolation.

It is quite likely that very few of the deoxyribonucleic acid
preparations described so far, and none of the models proposed
to describe their structure, are representative of the native state.
The actual operative entity — though not necessarily amenable
to biological testing in vitro — possibly is an aggregate of very
long polynucleotide chains linked to each other, perhaps by
oUgopeptide bridges, and bonded to proteins in a spatially unique
configuration. If this is so, the problem of heterogeneity, men-
tioned in the preceding section, is ostensible rather than actual,
having been introduced as a necessary, but strictly non-biological,
artifact of purification. By this token, a preferred isolation
method for deoxyribonucleic acid would be one that avoided, as
far as possible, denaturation by chemical or physical means.
Nearest to these requirements is perhaps the procedure described

References p. 159

130 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

on p. 326 of a previous survey^ and applied by us to preparations
from calf thymus^^ and Escherichia coli protoplasts^^.

c. The problem of statistical sequence analysis

The possibility that no two nucleic acid molecules within the
same nucleus are entirely identical offers "a prospect that would
seem to condemn us to forced statistics for life''^^. It is this more
than anything else that has made work on the nucleotide sequence
in nucleic acids appear so unattractive. Who would, after all,
undertake to read a book that has been passed through a grinder?
Nevertheless, being rather modest in what I expected to gain
from a perusal of the nucleic acid text, I have never been able to
share these apprehensions completely. I knew that a great deal
can be learned about an unknown language through a study of
its phonemes, their frequency, distribution density, and allo-
phonic relationships.

If the total deoxyribonucleic acid of a given species represents
a text, it is made up of "words" — the individual molecules — that
are composed of a singularly meagre alphabet: four or five letters.
But the words so spelled out are 10,000-letter words, each of
which could occur in a fantastically great number of positional
isomers: between 10^^^^ and 10^^^^, according to how many
restrictions on neighbors are admitted^. The situation facing us
in examining a nucleic acid preparation comprising a large num-
ber of isomers or homologues would, then, be comparable to one
in which all the words in a dictionary are lined up end to end in
a continuous, and essentially arrhythmic and aperiodic, sequence.

It is quite clear that the first attempt at unraveling such a
clutter will have to be based on statistics and that it must limit
itself to the description of tendencies or trends of arrangement.
To give an example: running together the thirteen words making
up the first sentence of King Lear I obtain a monster word of
fifty-seven letters of which twenty-one are vowels. On this word
a number of determinations can be made: (a) the ratio of con-
sonants to vowels; (b) the nature of the individual consonants
and vowels; (c) the relative frequency of each constituent. If I

NOMENCLATURE 131

have a way of removing the vowels without disturbing the rest
of the arrangement, I shall isolate six solitary consonants, eight
pairs of consonants, three bunches of triple consonants, and one
cluster of five consonants in a row. Each of these units, I would
conclude, was originally flanked on both sides by vowels. Other
words would yield other combinations, with the unambiguity of
distinction increasing with the length of the consonant clusters. In
very long words composed of only two vowels and two or three
consonants, unique clusters can be expected only very rarely; but
the relative frequency of the various combinations of consonants
(runs of 1, 2, 3, etc.) will be a means of unique differentiation,
even though it will not yet make it possible to reconstruct the
entire text.

This is what we have been doing with many deoxyribonucleic
acid preparations.

3. REMARKS ON NOMENCLATURE

a. Definition of the term "homotope"

If one monomeric constituent can take the position of another in
a definable segment of a polymer, we propose to designate it as a
homotope (from the Greek for occupying the same place). If
in the A-chain of insulin positions 8 to 10 are occupied by
Ala.Ser.Val in the ox^"^, but by Ala.Gly.Val in the sheep^^, I
would say that serine and glycine are homotopic with respect to
this sequence. The importance of this term for a consideration
of nucleic acid structure will become clear presently.

b. Definition of the term ''pleromer"

If in the total composition of polymers that are characterized by
balances such as the well-known pairing principles in deoxy-
ribonucleic acid^'"^, and in a more limited way in ribonucleic
acids^^, one constituent can ostensibly replace another in respect
to these balances, we propose to designate it as a pleromer (from
the Greek for filling up the measure). Thus, if in the deoxy-
ribonucleic acid of wheat germ^^ the molar quantity of guanine

References p. 159

132 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

equals that of the sum of cytosine and 5-methylcytosine, I would
say that these two 6-amino pyrimidines are pleromeric.

c. Solitary and bunched nucleotides

A pyrimidine nucleotide that, within the polynucleotide chain of
a nucleic acid, is contiguous only to purine nucleotides, will be
referred to as "solitary"; pyrimidine nucleotides that occur in

A A A A A A 'A A 'y\

/

/

/

/

VV.\/V.VVV.

/

■■&

A /

/

V

/

V

V.

Fig. 11. Schematic representation of a fragment of a double strand of a

deoxyribonucleic acid. The purines (A, adenine; G, guanine) are depicted

in black, the pyrimidines (C, cytosine; T, thymine) in white.

tracts of two or more, flanked on both sides by purine nucleotides,
will be designated as "bunched''^^. In the diagram shown in Fig.
11, soUtary thymidylic acid, for instance, appears in positions 3, 9
and 8', bunched pyrimidine nucleotides are seen in positions 5-7.

4. EARLY ATTEMPTS AT SEQUENCE INVESTIGATION

Since 1947, when in an early review on deoxyribonucleic acids
(p. 32 of Ref. 22), I first discussed the possible importance of
variations in the nucleotide sequence for the biological specificity
of a deoxyribonucleic acid, this problem has remained among the

EARLY SEQUENCE INVESTIGATION 133

interests of my laboratory. Our first chemical evidence on the
existence of different deoxyribonucleic acids^^-^^ made it clear
that whatever "code" was carried by the nucleic acids must be
imprinted on the sequential arrangement of the monomeric con-
stituents and prompted us to search for ways to approach this
difficult task.

The general problem is similar to that of the structure of
proteins, but immensely more forbidding. The very great length
of the chains composed of a much smaller number of different
monomers raises obstacles that render improbable an unam-
biguous solution by existing methods. Moreover, the available
procedures are not yet as refined as in the case of the proteins:
few specific enzymes, no generally applicable method for marking
the end groups. On the other hand, the remarkable difference in
the stability of the glycosidic bonds holding the purines and the
pyrimidines offered a novel possibility that we have exploited
fully.

Our first attempts in the direction of utilizing a specific enzyme
were concerned with the problem of a recognizable repeating unit.

a. Stepwise degradation by deoxyribonuclease

When crystalline pancreatic deoxyribonuclease acts on a highly
polymerized preparation of a deoxyribonucleic acid, it is possible
to conduct the experiment in such a manner as to collect the
products detached gradually in a dialyzable form and to separate
them from the enzyme-resistant core^-^. The latter is charac-
teristically different in composition from the intact polymer; there
is, moreover, a significant trend in the composition of the
products produced by the stepwise enzymic digestion. No indi-
cations of any regularity in the release or the composition of frag-
ments were observed; no sub-units of recognizably recurrent
structure were seen: deoxyribonucleic acids apparently exhibit a
largely arrhythmic nucleotide sequence.

b. Apurinic acid
This has proved a very useful compound; it is, in fact, the first

References p. 159

134 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

intermediate product in all procedures involving acid degrada-
tion^^. When a deoxyribonucleic acid is exposed to mildly acidic
conditions (pH 1.6 at 37°) all the purines are cleaved off gradual-
ly, leaving behind a polymer representing the original poly-
nucleotide, but composed of deoxyribophosphate units, at the
places of the previous purine nucleotides, and of the pyrimidine
nucleotides in unchanged ratio and at the same position as in the
native starting material. For instance, in the segment represented
in Fig. 11, positions 3, 5-7, 9, T, 2', 4', 8', 10' would remain
unchanged, whereas in the remaining places the purines would
now have made room for the reactive aldehydo groups of the
deoxy sugar. The presence of the aldehydo group^^ and, there-
fore, of a free hydroxyl at 4' brings about a remarkable labili-
zation of the polymer: it is broken not only by alkali, but even by
buffers (pH 8.6) containing primary amino groups^^. It is likely
that this susceptibility of the poly-sugar-phosphate backbone of
apurinic acid to amines is involved in the degradation reactions
employing diphenylamine under acidic conditions'^.

Studies on the arrangement of the pyrimidine nucleotides in
apurinic acid'^ were, in fact, the first that made possible an ap-
proach to the problem of nucleotide sequence in deoxyribonucleic
acid. The information so obtained was limited, since it was only
qualitative, but it showed that a considerable portion of the
pyrimidine nucleotides, and therefore also of the purine nucleo-
tides, was arranged in the form of tracts of several nucleotides of
one kind.

5. SEQUENCE STUDIES THROUGH DIFFERENTIAL
DISTRIBUTION ANALYSIS

a. Mechanism

The important discovery that among the fragments produced by
the acid degradation of deoxyribonucleic acids there are found
the 3',5'-diphosphates of deoxycytidine and thymidine is due to
the work of Levene, Thannhauser and their collaborators^^"^^.
A more recent re-investigation^* by improved techniques, which

DIFFERENTIAL DISTRIBUTION ANALYSIS

135

confirmed the occurrence of these diphosphates, prompted a dis-
cussion of the possible bearing of this finding on the problem of
nucleic acid structure (p. 366 of Ref. 7). It was clear that for
these fragments to serve as a tool in research on nucleotide
sequence the mechanism of their release had to be better under-
stood. It was also indispensable to be able to study them under
rigorously controlled conditions and on a quantitative basis.

The kinetics of the liberation of the pyrimidine nucleoside
diphosphates were first studied, in collaboration with Shapiro, on
a series of models, namely, the several deoxyribo dinucleotides all
having cytidylic acid as one of their components'^. These inves-
tigations served as the basis of a well-controlled, differential
hydrolysis procedure which was usually performed in three
stages^i' '^. The diagnostically most valuable results are obtained
in stage I (0.1 M H2SO4, 35 min, 100°), when the release of
pyrimidine nucleoside diphosphates can be taken to reflect direct-
ly the abundance of these units as solitary pyrimidine nucleotides
in the polynucleotide chain. In the present survey I shall limit
myself to the findings based on this stage of the differential
hydrolysis procedure. The results yielded in stages II and III,
which involve longer periods of hydrolysis, are mainly indicative
of the secondary cleavage of bunched pyrimidine sequences; they
are of great value as further means of differentiation between
analytically indistinguishable nucleic acids of different cellular
origin. This will be exemplified in the following section.

CHO

CH,

C^i-

CHOH
.0 — CHo

OH

P

ll\

o o-

Py

CH-

I

CH,

I

CH-

I

CH-

I

CHc

-O OH

\/

P

CHO

CHOH

-O OH

\/

P

o o..

In its simplest form the cleavage of a polynucleotide chain and
the formation of the diphosphates of the pyrimidine nucleosides
(Py) may be regarded as a series of ^ elimination reactions^^.
After the liberation of the purines the first elimination pre-

References p. 159

136 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

sumably occurs at the broken line A. Since no extraneous sugar
fragment is found in ester linkage with the resulting nucleoside
diphosphate, a subsequent hydrolysis or elimination must take
place at the other flanking sugar (broken line C). It is likely that
this cleavage follows a fj aldehyde elimination (broken line B),
which has left a double bond between the second and third
carbon atom of the sugar. Similar considerations apply to the

TABLE 28

DIFFERENTIAL DISTRIBUTION ANALYSIS OF THREE OTHERWISE
INDISTINGUISHABLE DEOXYRIBONUCLEIC ACID PREPARATIONS*

Source:

Ox

Man

Arbacia lixula

DNA fraction:

(0.75 M)

{1.0 M)

(2.6 M)

T C TC

T C

TC

T C TC

Total

pyrimidine,

as mole % P 29.9 20.0 29.9 19.8 30.9 19.6

Solitary
pyrimidine,
as mole % of
total constitu-
ent in DNA 15.9 9.2 19.9 6.3 16.0 14.0

Solitary or
bunched
pyrimidines, as
mole % P 4.75 1.84 1.72 5.95 1.25 1.71 4.94 2.74 1.66

Total T/C,

molar ratio 1.50 1.51 1.58

Solitary T/C,

molar ratio 2.58 4.76 1.80

* The figures are taken from previous papers-i'SC. The deoxyribonucleic
acid fractions are described by the molarity of NaCl employed for the
dissociation of the histone nucleate^. T designates thymine or thymidylic
acid, C cytosine or deoxycytidylic acid, TC the two isomeric dinucleoside
triphosphates comprising cytidine and thymidine (reported as moles of
dinucleoside triphosphate per 100 gram-atoms of nucleic acid phosphorus).

DIFFERENTIAL DISTRIBUTION ANALYSIS 137

release of bunched pyrimidine sequences. A more than tentative
formulation of the reaction mechanism will, however, have to
await a better understanding of the fate of the free deoxyribose
residues liberated in the initial stage of the removal of purines by
acid.

b. Nucleotide arrangement in analytically indistinguishable
deoxyribonucleic acids of different cellular origin

It happens quite often that deoxyribonucleic acids of taxonomi-
cally entirely different origin exhibit identity of composition as
regards the distribution of the nitrogenous constituents (compare
the example of the deoxyribonucleic acids of the sheep and the
salmon cited previously^ *^). In such cases, the method of differen-
tial distribution analysis discussed in this survey is destined to be
of great value. It permits the examination of detailed features of
the nucleotide arrangement, far beyond the possibilities of dis-
tinction offered by total constituent analysis. I have selected, for
inclusion in Table 28, three examples of nucleic acid fractions
that cannot be distinguished analytically. There can be little
doubt about the entirely different sequential plans governing the
structure of these polymers if the values for the abundance of
solitary pyrimidine units are compared.

c. Nucleotide arrangement in different deoxyribonucleic
acids of animal origin

Table 29 compares the distribution findings obtained with total,
unfractionated preparations of the deoxyribonucleic acids from
calf thymus, human spleen, and sperm preparations from two sea
urchin species. These data do not exhaust the information gathered
by us on different animal nucleic acids^^, but are presented
here to indicate the scope of the method.

d. Nucleotide arrangement in the total deoxyribonucleic acid
of rye germ and in a complete series of fractions

The use of a deoxyribonucleic acid from a plant source offers
several advantages. These nucleic acids have been shown to con-

References p. 159

138 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

I

E

. c
.5 3

•-2 c«

^2

•^ o
o

<N T^ '-H

ri 00

^ o ^

i-i a>

o

C/5

o
E

U

a

SH

«u c

-S

<s

DIFFERENTIAL DISTRIBUTION ANALYSIS 139

tain a fifth nitrogenous constituent, 5-methylcytosine, in appreci-
able quantity^' -^. In the specimens isolated from both wheat and
rye germ the concentration of this pyrimidine amounts to 5.9
mole % (Table 30). This means that nearly a quarter of the
cytosine has been "replaced" by methylcytosine. Since the molar
concentration of guanine equals the sum of these two pyrimidines,
one may consider them as pleromeric in the sense defined above.
Minor constituents of deoxyribonucleic acid (or, for that matter,
of ribonucleic acid) are destined to play an important role in
future work on the elucidation of nucleic acid structure, as they
will serve as additional markers in the array of nucleotides con-
stituting the polymer chain. Furthermore, the presence of a pyri-
midine sharing with cytosine the property of being a 6-amino
derivative and with thymine that of being a 5-methylpyrimidine
is of the greatest interest for a better understanding of the
mechanisms of selection, incorporation, and replication that are
at work.

TABLE 30

COMPOSITION OF DEOXYRIBONUCLEIC ACIDS OF RYE GERM AND WHEAT GERM

Rye germ* Wheat germ**

Molest 100 gram-atoms P

Adenine

27.7 28.1

Guanine

22.6 21.8

Cytosine

16.4 16.8

Methylcytosine

5.9 5.9

Thymine

27.4 27.4

Molar ratios

Adenine + thymine to guanine +

cytosine + methylcytosine

1.23 1.25

Purines to pyrimidines

1.01 1.0

Adenine to thymine

1.01 1.03

Guanine to cytosine + methylcytosine

1.01 0.96

Cytosine to methylcytosine

2.78 2.85

* Taken from a previous publications^.
** Taken from a previous publication's.

References p. 159

140 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

Before the study of the nucleotide arrangement in the deoxy-
ribonucleic acid of rye germ could be undertaken, it was neces-
sary to examine the properties of deoxymethylcytidine 3',5'-
diphosphate whose production is indicative of the frequency of
this pyrimidine as a solitary nucleotides^. At the same time, the
two dinucleoside triphosphates comprising 5-methylcytidine and
either cytidine or thymidine were also studied.

The investigation of the deoxyribonucleic acid of rye germ^^
offered us, for the first time, the opportunity of comparing the
pyrimidine distribution in the total preparation with that found
in a complete series of seven fractions isolated through the frac-
tional dissociation of the histone nucleate. These fractions were,
in their distribution and composition, remarkably similar to those
previously isolated by the fractionation of the deoxyribonucleic
acid from wheat germ^^. The availability of such a complete
series of fractions permitted, moreover, a test of the validity of
the results given by the differential distribution analysis; a
weighted average of the values recorded for the individual frac-
tions could be compared with those given by the total nucleic
acid preparation that had not been subjected to fractionation and
showed an excellent agreement^^.

The principal results of this study are, as regards the frequency
of soUtary pyrimidine nucleotides, shown schematically in Fig.
12. In this histogram, we give a brief description of the total
pyrimidine composition of all specimens examined — total deoxy-
ribonucleic acid and seven fractions — and depict the quantities
of soUtary nucleotide of each of the three pyrimidines as per cent
of the total constituent in the preparation. 5-Methylcytosine, it
will be seen, has a surprisingly high tendency, when compared
with the other 6-aminopyrimidine, cytosine, to occur as a solitary
unit, similar in this respect to thymine. I shall return to this point
later.

e. Nucleotide arrangement in a microbial deoxyribonucleic
acid of the GC type

Our survey has so far all been dealing with deoxyribonucleic

DIFFERENTIAL DISTRIBUTION ANALYSIS

141

acids that belong to the AT type. It was of interest to see
whether the same considerations could be applied to a nucleic
acid of the GC type, i.e., one in which guanine and cytosine
preponderated. We first discovered this type in tubercle bacilli^^;
but it has since that time been found in many micro-organ-
isms^'^^. The distribution analysis of the deoxyribonucleic acid
of an avirulent variant (BCG) of bovine tubercle bacilli is shown
in Table 31. In the nucleic acids of the AT type, as was shown
before, the quantity of thymidine 3',5'-diphosphate released

SOLITARY PY NUCLEOTIDES
BASES, MOLE % IN DNA

TOTAL RGDNA
RGDNA-0.60
RGDNA- 0.65
RGDNA- 0.70
RGDNA- 0.75
RGDNA- 0.80
RGDNA- 0.90
RGDNA- 1.70

IN RYE GERM DNA & FRACTIONS

DIPHOSPHATES, % OF BASES
5 10 15 20 25

27.4
16.4
5.9

26.3
17.0
6.8

27.1
16.2
6.4

28.2
16.6
6..

27.8
16.6
5.6

28.0
16.8
6.1

28.6
»6.5
5.4

29.2
15.8
4v7

T 1 1 I I

///////////////////A

c

M

_

////////////////\

C

M

T

(////// /////////a

C

M

_

''■//////////////////A

C

M

T

•///////////////////A

C

M

'////////////////A

C

M

T

''///////////////////////A

C

M

T
C

V///yy//////////////A

M

0,T D , C ■ , M

Fig. 12. Pyrimidine distribution analysis of the total deoxyribonucleic acid
of rye germ and its fractions. The first column indicates the NaCl molarity
at which dissociation of the fraction from its histone salt occurred. The
second column summarizes the composition of the specimens in terms of
the molar concentrations (per 100 gram-atoms of nucleic acid P) of thy-
mine (T), cytosine (C), and 5-methylcytosine (M). The histogram shows the
frequency of the solitary nucleotides, isolated as the nucleoside 3',5'-
diphosphates, relative to the total concentration of the constituent in the
nucleic acid. Based on previously published data^^.

References p. 159

142 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

during hydrolysis exceeded that of deoxycytidine 3 ',5 '-diphos-
phate: this indicated the occurrence of larger quantities of soUtary
thymine than cytosine, but the ratios of the soHtary residues were
not those that would have been predicted statistically from the
total abundance of these pyrimidines in the polymers. In the
nucleic acid of the GC type an inversion in the rates of liberation
of the nucleoside diphosphates is observed, as expected; but in
this instance, the ratios of sohtary and of total pyrimidines are
not widely different: 0.59 v^-. 0.52.

TABLE 3 1

DIFFERENTIAL DISTRIBUTION ANALYSIS OF DEOXYRIBONUCLEIC ACID OF BCG*

T C CC

Total pyrimidine, as mole % P 16.7 32.3

Solitary pyrimidine, as mole % of total

constituent in DNA 13.6 12.0

Solitary or bunched pyrimidines, as

mole % P 2.27 3.88 1.13

Total T/C, molar ratio 0.52

Solitary T/C, molar ratio 0.59

* The figures are taken from a previous paper^o. CC designates dicytidine
triphosphate. Compare Table 28 for other explanations.

/. Nucleotide arrangement in the deoxyribonucleic acid of E. coli
grown in the absence and presence of 5-bromouracil

The deoxyribonucleic acid of E. coli is unusual, being composed
of nearly equimolar proportions of the four nucleotide con-
stituents^'^^ No obvious distinctions appear, even when different
variants and mutants are examined, although it is not unlikely
that a careful differential distribution analysis would reveal some
differences. Such a study has, however, not yet been undertaken.
It appeared of great interest to study the effect of a fraudulent
analogue on the sequence characteristics of a deoxyribonucleic

DIFFERENTIAL DISTRIBUTION ANALYSIS

143

acid. For this reason, the nucleic acid produced by thymine
auxotrophs of E. coli, grown in the presence of thymine, was
investigated and compared with that formed by the same organ-
isms when suppUed also with 5-bromouracil. This 6-keto-
pyrimidine is known to be accepted by certain bacterial cells as
an analogue of thymine^^'^^. The incorporation of the halo-
pyrimidine, which is essentially limited to cells unable to syn-
thesize thymine, is accompanied by many biological consequen-
ces. The cells suffer a gradual loss of viabiUty^^ and exhibit a
considerable increase of the rate of formation of mutants^^,
sometimes with preferred sites of particularly high mutability^^.
It has also been shown that bacteriophage T2 particles may retain
their infectivity, despite the uptake of a high proportion of 5-
bromouracil^^.

The results of our first study with the thymine auxotroph,
strain I, were quite surprising^*: they seemed to indicate a very
severe distortion of the sequence characteristics of the deoxy-
ribonucleic acid in which about 35 mole % of thymine had been
replaced by 5-bromouracil*^. Further work with this and another
mutant strain has, however, failed to bear out this conclusion;

Fig. 13. Chromatography of hydrolysates of deoxyribonucleic acid of
E. coli, mutant strain I (Preps. 3 and 4 in Table 32): (a) grown in the
absence; (b) in the presence of 5-bromouracil. Elution from DEAE-cel-
lulose with an increasing linear gradient of LiCl in lithium acetate buffer
(pH 5); other techniques as in Ref. 51. Total volume of eluent 1,350 ml;
7-ml fractions. Taken from a recent paper^o.

References p. 159

144 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

+

I IM ■U

& o 3 PQ

3 O

W5 c/5 r- . ^

I

VO

1

>o

1

>o

medium
wo cycl
h or wi
thymin

1

>o

1

■^

1

r^

•^ Z 1 H

q

1

q

q

§

§

^

"

"

1 a syr
uracil
thymi
cytosi

m

o

^

^

5 o «^

q

q

q

q

q

q

15t-arg- was grow
and 10 /iig of 5-br
presence of 2 /^g/m

enine; G, guanine;

ents (G, T, B).

oo

•o

^

00

00

m

<y\

On

On

ON

On

q

d

d

d

d

d

^

_ (U (D "2 =3

^ c ^ ^ .ti

a,-- *-• -3

e|.H<i

O 43 „ .. O

1

1

n

1

q

On

rhe thymine aux
with 1 /<g of t
n for 18 h at 37'
e. Abbreviations
C); 6-K, 6-keto

1

vo

1

in

1

OO

Tt

-^

d

^

f<^

Tf

NO

fN

(N

'"'

(N

^

r-;

NO

-^

>n

^

w-l

ir^

>/^

»n

>n

»o

ON

in

q

• o 1 -S <

3 >. rt a 5

(N

(N

fN

(N

<N

fNJ

q

^

NO

0\

OO

en

d res
of th
simil
halo
:onstii

»o

"^

Tt

^

^

in

(N

r^^

(N

(N

(N

fN

«J ^ ^ "^

m

oa

PQ

H

+

H

+

H

+

•5-« o 1

H

H

H

reviously p
r ml, with
37'; strain
f 10 «g/ml
; 6-Am, 6-a

'"H

(S

m

Tf

»o

NO

d. oj O -^

c "^ ^ c -g

1

o — - c o u

1

1^

HH

H^

Based
ementec
pHcatio
e additi
bromou

>0

1— 1

* -E 2^6^

DIFFERENTIAL DISTRIBUTION ANALYSIS

145

and I shall here consider the situation as it appears in the light
of our most recent evidence^^. When the results of the total
analysis of six nucleic acid specimens from two mutant strains
are compared (Table 32), it will indeed be seen that 5-bromo-
uracil and thymine are pleromeric with respect to the equality of
the molar quantities of purines and pyrimidines, of 6-amino and
6-keto nucleotides, and of 6-aminopurine (adenine) and 6-keto-
pyrimidines (thymine + 5-bromouracil).

The improved methods for the determination of the distribution
of pyrimidine clusters^^ were applied to Preparations 1-4 (Table
32). The elution diagram showing the separation of the pyri-
midine clusters released from Preparations 3 and 4 is shown in
Fig. 13, the phosphorus distribution in all four preparations of
nucleic acid in Table 33. It may be seen (Fig. 13) that the nucleic

TABLE 33

DISTRIBUTION OF PYRIMIDINE NUCLEOTIDE CLUSTERS IN DEOXYRIBONUCLEIC

ACID OF TWO THYMINE AUXOTROPHS OF E. CoU GROWN IN THE ABSENCE AND

PRESENCE OF S-BROMOURACIL^O

Strain:

ements:

15 t-

-arg-

/

Growth suppl

T

T + B

T

T + B

DNA Prep. No.:

1

2

3

4

Fraction No.

Number of
pyrimidine
nucleotides

Pro port >

ion of DNA

phosphorus, %

in
fragment (n)

I

1

13.7

13.4

13.9

13.9

II

2

12.1

12.3

12.2

11.7

III

3

7.7

7.6

7.4

7.6

IV

4

5.4

5.5

5.1

5.6

V

5

3.4

3.6

3.7

4.0

VI

6

2.2 \

2.3 )

VII
VIII

7
8

1.3
1.1 I

8.1**

1.4 (
1.2 (

6.5**

IX

3.3* )

3.5* )

Total

50.2

50.5

50.7

49.3

* Based on
** Based on

an average chain length of n
an average chain length of n

= 11.

= 8.

References p.

159

146 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

TABLE 34

FREQUENCY OF SOLITARY AND COUPLED PYRIMIDINE NUCLEOTIDE UNITS IN

DEOXYRIBONUCLEIC ACID OF A THYMINE AUXOTROPH OF E. CoU t-arg"

IN THE ABSENCE AND PRESENCE OF 5-BROMOURACIL*

Growth supplements:
DNA Prep. No.:

T + B
2

Moles of pyrimidine per
100 gram-atoms of
DNA phosphorus

Solitary units

TP2

BP2

CP2

Total T or B

Coupled units

T2P3

TBp3

B2P3

TCP3

BCP3

C2P3

Total T or B

3.81 (0.05)

1.60 (0.05)

—

2.02 (0.08)

4.53 (0.17)

4.84 (0)

3.81

3.62

2.42 (0.01)

0.66 (0.10)

—

1.32 (0.24)

—

0.97 (0.05)

5.42 (0.06)

2.50 (0.14)

—

3.18 (0.10)

3.44 (0.02)

3.27 (0.11)

5.13

5.79

B/(T + B) • 100 in solitary units
B/(T + B) • 100 in coupled units
B/(T + B) • 100 in total DNA

56
56
56

* The nucleic acid preparations are numbered as in Table 32. The same
abbreviations as in Table 32 are used, with p indicating the number of
phosphate groups in the fragments having the general formula Py„p„+i
(Py = pyrimidine). No distinction is made between sequential isomers of
the coupled units. The figures are the averages of the results of two
separate hydrolysis experiments, each chromatographic component being
determined twice; the deviations from the averages are given in paren-
theses. This table is taken from a recent publication^o.

acid specimens produced in the absence of the halopyrimidine
yielded at least eight distinct fractions, going from solitary units
to octanucleotide sequences, whereas with the preparations con-
taining 5-bromouracil the longest runs separated clearly were
those of five pyrimidines in a row. It may be that this often
observed difference is not without significance.

SUMMARIZING REMARKS 147

In all nucleic acid specimens the frequency of solitary and
coupled pyrimidine nucleotide units — and in some cases also that
of triplets — was determined. An example of such an analysis is
shown in Table 34. It will be seen that — in contrast to the first
observations^^ — no evidence emerged of a gross non-randomness
in the ability of the fraudulent analogue to substitute for thymine
within the polynucleotide chain.

6. SUMMARIZING REMARKS

a. Sequence characteristics common to all deoxyribonucleic acids

Though the first application of the procedure for the differential
analysis of pyrimidine distribution served to demonstrate that
deoxyribonucleic acids from different species, even those show-
ing identical base composition, were vastly different with regard
to the alignment of their nucleotide constituents (Table 28), the
more important and interesting uses of this method bear on other
aspects of the sequence problem. These relate, in fact, to the
existence of sequential patterns that all deoxyribonucleic acids
examined heretofore appear to have in common.

The first consequence of our studies may be the statement
that in no case can the nucleotide sequence be predicted from
the knowledge of the frequency of the individual constituents.
The discarded tetranucleotide hypothesis is a primitive example
of an attempt to make such a prediction; I have dealt with this
a long time ago^. This is, incidentally, quite similar to what one
would observe in a meaningful text: a certain letter may be par-
ticularly frequent; but this does not mean that it wiU occur, with
commensurate frequency, as a doublet. We shall probably find,
in general, that the symbols constituting a text are not arranged
at random, if this term is understood to refer to an alignment
governed only by the relative frequency of the elements, once
the requirement of large numbers is fulfilled.

This brings up the problem of the non-randomness of the
nucleotide sequence in deoxyribonucleic acids which we have
discussed fully in a previous paper^i. Whether strict randomness

References p. 159

148 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

is conceivable in a living system capable of orderly replication,
at any rate on the level of macromolecular events, is a question
which it would be out of place to discuss here. The entire com-
plex of randomness in biology would certainly deserve a more
profound examination than I am equipped to perform. In any
event, the conclusion that in no case investigated the pattern of
nucleotide distribution was that to be expected from a random
arrangement of constituents-^* ^^' ^^' ^^ may contribute to the as-
surance that in studying the nucleotide sequence in a deoxy-
ribonucleic acid we are, indeed, dealing with some sort of mean-
ingful message. If the array of constituents in a biologically
functional polynucleotide has been compared to a giant throw
of dice, we must conclude that it is a throw of loaded dice.

Another consequence of our studies on deoxyribonucleic
acids of animal and plant original' ^'^-^^ is the conclusion that at
least 60% of the pyrimidines occur as oligonucleotide tracts
containing three or more pyrimidines in a row; and a correspond-
ing statement must, owing to the equality relationship, apply also
to the purines.

b. Possibilities of cluster analysis

It is quite clear that the various soUtary nucleotides and the small
bunched oligonucleotides, such as doublets or triplets, will vary
in frequency, i.e., in relative quantity, according to the plan
governing the nucleotide sequence of the particular deoxy-
ribonucleic acid; but they are most unhkely to exhibit uniqueness
of quality and will, therefore, not be sufficient to define a se-
quence unambiguously. Uniqueness may, however, be expected, at
least occasionally, when very large clusters can be included in
the survey. We have made a beginning with regard to clusters of
pyrimidine nucleotides^^. When a deoxyribonucleic acid is cleaved
by acid under moderate conditions (0.1 M H2SO4, 35 min, 100°)
and the fragments are fractionated on DEAE-cellulose by means
of a hthium chloride gradient, the phosphorylated pyrimidine
ohgonucleotides can be separated in order of increasing chain
length (compare Fig. 13). Further separation within the different

SUMMARIZING REMARKS 149

size groups then is based on two-dimensional chromatography on
filter paper.

In this manner, essentially the entire pyrimidine phosphorus
of a nucleic acid can be accounted for (see the examples in Table
33). With the amounts usually employed for hydrolysis (20 mg
of nucleic acid) the isolation of sequences longer than nine is
difficult; but there is no doubt that with more material and larger
columns the separation of even longer units will be possible,
especially if by the use of radioactive labels the sensitivity of the
detection of fractions could be extended. What will become,
however, increasingly harder or even impossible, is the separation
of very large clusters into individual components, e.g., the dis-
crimination between Tispie and T14CP16, etc., let alone the
preparation of the many sequential isomers of this type.

One gains the impression that almost all possible combinations
of pyrimidine nucleotide units are encountered. But there is one
important, and possibly quite significant, exception. There seems
to exist a certain bias in the arrangement of homologous pyrimi-
dine clusters: many nucleic acids appear to contain much longer
all-thymine than all-cytosine units. This apparent trend may be
exemplified by the instance of the deoxyribonucleic acid of
E. coli which contains nearly equimolar quantities of thymine
and cytosine (compare Table 32). One should have expected
similar proportions of the stretches of varying length comprising
only one or the other of these pyrimidines. Actually, no cytosine
units longer than C3 have been observed in the hydrolysates,
whereas T4 and T5 units have invariably been found, together
with the various mixed fragments such as T3C, T2C2, TC3, T4C,
T3C2 and T2C3. In general, cytosine is found in a relatively
greater abundance among the smaller units; the converse is true
of thymine.

These and related methods, employing specific enzymes, will
doubtless have to be refined considerably, before more definitive
answers to the sequence problem and its biological meaning
become possible. It would, for instance, be of the greatest interest
to determine the nature of the lethal sequence of a parasitic

References p. 159

150 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

deoxyribonucleic acid that enables it to impose itself upon, and
to paralyze, as it were, the bacterial host.

c. What is meant by ''replacement"?

The platitude that "nobody is irreplaceable" — if it ever holds
true — certainly cannot hold for the deoxyribonucleic acids, if
they carry the biological functions ascribed to them currently.
Nevertheless, one often finds the statement in the literature that
an analogue of a purine or pyrimidine, not normally occurring in
nature, is, when offered to the cell, incorporated into a deoxyribo-
or ribonucleic acid "in the place" of a particular natural con-
stituent of the nucleic acid. Such statements are, of course,
derived from the existence of the pairing principles^, i.e., they
are based on the fact that certain nucleic acid constituents are
pleromers, to use the nomenclature suggested above. Good
examples of such pleromerism can be seen in Table 30 with
respect to cytosine and 5-methylcytosine and in Table 32 with
respect to thymine and 5-bromouracil; a more comprehensive
selection of such instances will be given at the end of this paper.
What does the assertion that X can be "replaced" in part by Y
actually signify? (1) It means that Y is incorporated into the
deoxyribonucleic acid chain and that the cell, hence, must pos-
sess, or acquire, the enzymic apparatus necessary for the produc-
tion and utihzation of the derivative of Y that is the direct
precursor of the polynucleotide chain. (2) If X and Y share the
same functional group {e.g., 6-keto or 6-amino) through which
alone the selection of the members of the newly forming chain
takes place — as foretold by a mechanism postulating the replica-
tion of the complementary halves of a double strand^^ — it follows
that every X molecule along the chain has an equal chance of
being "replaced" by a Y molecule.

d. Selection of a natural analogue

When we deal with a natural analogue, such as the 5-methyl
derivative of cytosine, it is, of course, meaningless to speak of the
replacement of one by the other. Such a statement acquires

SUMMARIZING REMARKS 151

meaning only in reference to the possible mechanism of the
biosynthesis of the polynucleotide chain or to the existence of
such principles of pleromerism as those mentioned before. In
other respects, in polymer chains that are not constructed haphaz-
ardly, each monomeric constituent must have its preordained
place; a place that it must have entered, at the time of poly-
merization, through some form of a specific selection mechanism.
A replication mechanism of this nature is, indeed, offered by the
scheme^- postulating a specific pairing of a 6-aminopurine
(adenine) with a 6-ketopyrimidine (thymine) and of a 6-keto-
purine (guanine) with a 6-aminopyrimidine (cytosine).

This postulate and the fact that 5-methylcytosine is pleromeric
with cytosine could have led to the expectation that these two
pyrimidines could replace each other at random in the deoxy-
ribonucleic acid chain, selection having regard only to the
presence of a 6-amino group in the pyrimidine. The amazingly
constant quantity of methylcytosine incorporated into the deoxy-
ribonucleic acids of wheat and rye germ (Table 30, and compare
also the tabulation given in a previous paper^^) should, however,
render such a conclusion most questionable. In what manner
could a guanine residue in the preformed chain exercise a
preference for pairing with cytosine, rather than with methyl-
cytosine, in the newly forming counterpart chain? There are
surely limits to the amount of foresight that we can attribute to
an ostensibly automatic process.

Another observation, made repeatedly^- ^^' ^^, has to do with
the trend of the distribution of 5-methylcytosine in consecutive
fractions obtained by the fractionation of deoxyribonucleic acid
from calf thymus, wheat germ, and rye germ. In all instances, the
relative abundance of 5-methylcytosine with respect to cytosine
varied in such a manner that the percentage contribution of the
minor pyrimidine was higher in the earher fractions. In no case
could one have claimed that the two 6-aminopyrimidines replaced
each other at random within the polynucleotide chain. I have
discussed, in more detail, the significance of the finding of a
disproportionate distribution of cytosine and 5-methylcytosine in

References p. 159

152 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

view of current replication hypotheses at a previous occasion^'^.
An even more impressive difference between the ways in which
methylcytosine and cytosine are selected for insertion into the
polynucleotide emerges from the study of the frequency of these
pyrimidines as solitary and bunched nucleotides^"^. As will be
seen in Fig. 12, a remarkably large proportion of methylcytosine,
and a surprisingly small proportion of cytosine, exist as solitary
units. These two 6-aminopyrimidines can hardly share a common
pattern of nucleotide sequence. What is, however, remarkable is
that the mole fractions of 5-methylcytosine and thymine ap-
pearing as solitary units are nearly equal in most instances. One
must conclude that, whereas there is evidence that the two 6-
aminopyrimidines are not treated indiscriminately by the selection
mechanism, there is some indication that no such distinction is
exercised in regard to the two 5-methylpyrimidines. In the terms
defined above, cytosine and 5-methylcytosine are pleromers, but
not homotopes; thymine and 5-methylcytosine appear to be
homotopic with respect to the sequence purine-pyrimidine-
purine, although they are not pleromeric.

e. Selection of a fraudulent analogue

It is striking that almost all extraneous purine or pyrimidine
analogues that can, by the living cell, be incorporated massively
into nucleic acids — e.g., 5-bromo-, 5-chloro- or 5-iodouracil into
deoxyribonucleic acid, 5-fluorouracil or 8-azaguanine into
ribonucleic acid — belong to the 6-keto type. In the instance in-
vestigated in the greatest detail, viz., the incorporation of 5-
bromouracil into the deoxyribonucleic acid of E. coli, the problem
of replacement can be posed directly. In the case of a natural
satelUte, such as 5-methylcytosine, the complement appears to be
fixed rigidly by the cell, whereas no obvious cellular mechanism,
except the death of the cell, can be discerned that would limit the
extent of incorporation of the halopyrimidine. One would expect
the cell to discriminate between a foreign intruder and the normally
required thymine; but the auxotroph apparently can be inveigled
into accepting the substitute, very much in contrast to the wild type.

SUMMARIZING REMARKS 153

The inspection of Table 34 will show that in the nucleotide
sequences under comparison the relative contribution of 5-
bromouracil to the sum of 6-ketopyrimidines equalled the relative
abundance of the analogue in the total nucleic acid. For instance,
in Prep. 2 the halopyrimidine accounted for 56% of the sum of
the 6-ketopyrimidines of the nucleic acid (Table 32) and the
same figure was found for its relative frequency in the solitary
and dinucleotide units (Table 34). Similar results were obtained
with the other mutant strain, not listed here in detail; and in this
case the agreement extended even to the trinucleotides^^. In ad-
dition, the frequencies of thymine in the various pyrimidine
nucleotide clusters released from the nucleic acid specimens
synthesized in the presence of thymine alone were quite similar
to the frequencies of the sum of thymine and 5-bromouracil in
the coiTesponding fractions from preparations produced in the
presence of the analogue.

/. The concepts of pleromerism and homotopy and the neighbor

problem

The definitions of the suggested terms "pleromer" and *'homo-
tope" have been given above. The compilation in Table 35, by
no means complete, gives a few examples of such pleromeric
relationships in deoxyribonucleic acids. There can be little doubt
that in all these instances the established pairing relationships
would have been destroyed, had the minor pleromer been omitted
from the computations. Only one satellite normally occurring in
nature, viz., 5-methylcytosine, could be included in the table.
Another minor component — if it can be classified as a normal
constituent — namely, N-methyladenine, plays a baffling role. It
has been reported to occur, among other instances, in the deoxy-
ribonucleic acid of a thymine auxotroph of E. coli in not in-
considerable quantity ^^. Under special conditions of growth, its
concentration increases, somewhat in measure with a decrease in
the amount of thymine. It is, however, at the moment difficult to
see through what mechanism a 6-ketopyrimidine and a 6-methyl-
aminopurine could function as pleromer s.

References p. 159

154 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

I to

IP

ill

Q i: c

o
03

ON O

O T^

ao 5

o r^

o o

ON On

r^ ^ d> d>

B B B B

^ M xi fi
H ^ b-^ i-^

O '-^ ^
ON M- MD

0

0

0

T)

0

cd

C3

>.

>v

u

t-1

u

0

0

3

3

c3

3

I'

2

i

P

0

0

s

2

'^

^

^

m

I-H

«

1

"o

"0

"0

d)

0

0

0

^

>»

^

P<

UJ

UJ

1*^

Oh

6 o

o.S

XI c

<u o

5i) ^

o a
1^

o o

C/3 C3

.a "Sh

>>X1

OJ

il

>. o

^^
xj

a o
a * g

NO -y^ 3

t+-i of a
O g -c

•^ .5 "S
aj ^ "^

c3 .a o
^ I

* NO (/J

3 3

II

c -S

2 S

^ 2
1 ^-1

'^ -T-l

O «

3 o

o a

o o

-S

I?

SUMMARIZING REMARKS 155

Pleromeric relationships in ribonucleic acids are much more
difficult to estabhsh, as the only complementariness* useful for
scrutiny is the equality of the 6-amino and 6-keto nucleotides^^.
Of the three examples listed in Table 36 only the one deahng
with the incorporation of 5-fluorouraciP^ carries conviction. It
is not impossible that, when more careful analyses become
available, 8-azaguanine and guanine will also emerge as plero-
mers.

The usefulness of the concept of homotopy is, for the time
being, more evident with respect to the amino acid sequence in
proteins than to the nucleotide arrangement in nucleic acids.
More and more instances of such homotopic relationships are
being revealed as knowledge on specific sequences in proteins
increases. As concerns the deoxyribonucleic acids, we are largely
limited to the consideration of simple sequences, such as purine-
pyrimidine-purine, or of small bunches of pyrimidine nucleotides.
The development of the study of larger clusters, mentioned
before, will doubtless contribute much to our knowledge. I have
emphasized above a possible instance of homotopy, namely, the
remarkable similarity in distribution, as solitary nucleotides, of
the two 5-methylpyrimidines (thymine and 5-methylcytosine) in
rye germ deoxyribonucleic acid^^.

At the present stage of our knowledge statements on the
presence or absence of homotopy can be valid only in the most
general statistical terms. Specific homotopic replacements may,
of course, have taken place even where no all-embracing posi-
tional replacement can be affirmed. As concerns the replacement
of thymine by a foreign analogue, discussed above in more detail,
it ought to be understood that the observation of an ostensibly
random replacement of thymine by 5-bromouracil does not ex-
clude the possibility that the analogue may be involved in certain
non-random preferences for neighbors. For instance, no infor-

* This regularity — in the absence of the other pairing regularities
characteristic of deoxyribonucleic acids — may, formally, be taken to mean
that in ribonucleic acids adenine can pair not only with uracil but also
with guanine, and cytosine not only with guanine but also with uracil.

References p. 159

156 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

•S ^

1 1

5 --

00

0

d

6

^3

Ti 3 3

'u 'u 'C

00^

^ 00

r4 as

■^ >»

>. 3

^ ^ iZ

> --^ is

lij lil

C3

D

cd

3

^

\3

60

0

^

(4-1
0

73

3

4^

c

(-1

0

<u

0

M

VO

CO

Ta

(D

M

'5

c«

bp

_o

ti

0

>.

TJ

0

a>

T3

T3

3

c

oi

'2

^

TS

<u

C«

0

*©

S

s

^

3

%_•

C/5

73

tH

iS

'0

cd

"o

e

0

0

13

■^

JS

,0

0

'^I'

X)

C^

0^

3

(5

0

*

*

rt

*

SUMMARIZING REMARKS 157

mation is available about the relative frequencies of pTpBp and
pBpTp among the pyrimidine doublets or about the relative
abundance of adenine-bromouracil-adenine as compared with
that of adenine-thymine-adenine among the solitary pyrimidine
units. Until much more is known it would be safer to assume that
in the deoxyribonucleic acids pleromers are not necessarily
homotopes, and homotopes not necessarily pleromers.

This brings me to the last topic of the present survey, namely,
the neighbor problem. It was, I believe, first stated in the following
terms^*^. 'The frequency of methylcytosine being linked to
guanine in a dinucleotide as compared with that of cytosine being
so attached is twenty times as high in calf thymus nucleic acid,
and thirteen times as high in the wheat germ preparation, than
would have been expected for non-selective incorporation. The
unexpectedly high proportion of 5-methyldeoxycytidylic acid that
is linked to deoxyguanylic acid^^'^^ makes one, in fact, wonder
whether it is not the adjoining nucleotide rather than the one op-
posite that directs the incorporation." This remarkable phenom-
enon— the high tendency of methylcytosine to be next to guanine
— was again touched upon in our study on the sequence charac-
teristics of the deoxyribonucleic acid of rye germ^*. Several
analogous observations also were discussed there of which one
may be mentioned. When, in this nucleic acid, the coupled
dipyrimidine units are surveyed, it is found that methylcytosine
has a greater tendency to be linked to either thymine or cytosine
than the latter has to associate with itself or with thymine; and
there is a definite bias in favor of methylcytosine-cytosine units.

Somewhat similar observations have also been made in the
case of the ribonucleic acids in which the analysis of neighboring
tendencies is easier because of th.e facility of obtaining any given
nucleoside as either the 3'- or the 5'-phosphate. An examination
of this type has, for instance, been made with ribonucleic acid
preparations from different cellular portions of rat liver^^. A
particularly instructive example of the neighbor principle is of-
fered by the "soluble" ribonucleic acids of the cytoplasm con-
cerned with the specific transfer of amino acids^^, in which the

References p. 159

158 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

terminal sequence adenine-cytosine-cytosine is required; a se-
quence that can apparently be enforced by a specific enzymic
mechanism.

The manner in which a growing deoxypolynucleotide chain
could guarantee the quality of its continuation is not yet a topic
encouraging speculation*. Our present ideas about the enzymes
taking part in the biosynthesis of specific polymers wiU probably
have to undergo a stringent revision before the directed synthesis
of comphcated sequences will begin to be understood. For this,
more than a biological form of Scotch tape is required. Certain
sequential restrictions could, incidentally, be explained if, under
circumstances, runs of more than one nucleotide were the im-
mediate precursors of the polymer chain. We have, for instance,
discussed the evidence, in rye germ nucleic acid, of the two
dinucleotides cytosine-guanine and methylcytosine-guanine being
able to substitute each other at random^''^.

In considering the problem of the nucleotide sequence in
deoxyribonucleic acids we have barely turned the comer. There
is a long road before us; and we shall not see its end.

* "Whereof one cannot speak, thereof one must be silent." Wittgenstein,
Tractatus Logico-Philosophicus.

NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACID 159

REFERENCES

1 E. Chargaff, Experientia, 6 (1950) 201. [See Chapter 1 of this book.]

2 E. Chargaff, in The Origin of Life on the Earth, Intern. Union

Biochem. Symposium Series, VoL 1, Pergamon Press, London, 1959,
p. 297. [See Chapter 7 of this book.]

3 E. Chargaff, Trans. Intern. Congr. Biochem, 4th Congr., London,

14 (1959) 21. [See Chapter 8 of this book.]

4 J. T. Edsall, in H. Neurath and K. Bailey (Eds.), The Proteins, Vol.

1, Academic Press, New York, 1953, p. 549.

5 C. L. Sadron, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. Ill, Academic Press, New York,
1960, p. 1.

6 M. D. IMoNTAGUE AND R. K. MoRTON, Nature, 187 (1960) 916.

7 E. Chargaff, in E. Chargaff and J. N. Davidson (Eds.), The Nucleic

Acids: Chemistry and Biology, Vol. I, Academic Press, New York,
1955, p. 307.

8 E. Chargaff, C. F. Crampton and R. Lipshitz, Nature, 172 (1953)

289.

9 C. F. Crampton, R. Lipshitz and E. Chargaff, J. Biol. Chcm., Ill

(1954) 125.
It) E. Chargaff, in Fibrous Proteins and Their Biological Significance,
Symposia Soc. Exptl. Biol., 9 (1955) 32. [See Chapter 3 of this book.]

11 J. D. Watson and F. H. C. Crick, Nature, 111 (1953) 737.

12 L. F. Cavalieri, B. H. Rosenberg and J. F. Deutsch, Biochem.

Biophys. Research Communs., 1 (1959) 124.

13 G. Bernardi, M. Champagne and C. Sadron, Nature, 188 (1960) 228.

14 C. F. Crampton, R. Lipshitz and E. Chargaff, J. Biol. Chem., 206

(1954) 499.

15 E. Chargaff, H. M. Schulman and H. S. Shapiro, Nature, 180 (1957)

851.

16 E. Chargaff, Harvey Lectures, 1956-1957, 52 (1958) 57. [See Chap-

ter 6 of this book.]

17 A. P. Ryle, F. Sanger, L. F. Smith and R. Kitai, Biochem. J., 60

(1955) 541.

18 H. Brown, F. Sanger and R. Kitai, Biochem. J., 60 (1955) 556.

19 D. Elson and E. Chargaff, Biochim. Biophys. Acta, 17 (1955) 367.

20 G. R. Wyatt, Biochem. J., 48 (1951) 584.

21 H. S. Shapiro and E. Ch arg aff,,, B/oc/i/m. Biophys. Acta, 26 (1957)

608.

22 E. Chargaff, Cold Spring Harbor Symposia Quant. Biol, 12 (1947)

28.

23 E. Chargaff, E. Vischer, R. Doniger, C. Green and F. IVIisani, /.

Biol. Chem., Ill (1949) 405.

24 E. Vischer, S. Zamenhof and E. Chargaff, J. Biol. Chem., Ill (1949)

429.

25 M. E. HoDES and E. Chargaff, Biochim. Biophys. Acta, 22 (1956) 361.

160 NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS

26 C. Tamm, M. E. Hodes and E. Chargaff, ]. Biol Chem., 195 (1952)

49.

27 C. Tamm and E. Chargaff, J. Biol. Chem., 203 (1953) 689.

28 C. Tamm, H. S. Shapiro, R. Lipshitz and E. Chargaff, J. Biol. Chem.,

203 (1953) 673.

29 K. Burton and G. B. Petersen, Biochem. J., 75 (1960) 17.

30 p. A. Levene and W. a. Jacobs, J. Biol. Chem., 12 (1912) 411.

31 S. J. Thannhauser and B. Ottenstein, Z. physiol. Chem., 114 (1921)

39.

32 p. A. Levene, J. Biol. Chem., 48 (1921) 119.

33 S. J. Thannhauser and G. Blanco, Z. physiol. Chem., 161 (1926) 116.

34 C. A. Dekker, a. M. Michelson and a. R. Todd, J. Chem. Soc,

(1953) 947.

35 H. S. Shapiro and E. Chargaff, Biochim. Biophys. Acta, 26 (1957) 596.

36 H. S. Shapiro and E. Chargaff, Biochim. Biophys. Acta, 23 (1957) 451.

37 H. S. Shapiro and E. Chargaff, Biochim. Biophys. Acta, 39 (1960) 68.

38 R. Lipshitz and E. Chargaff, Biochim. Biophys. Acta, 19 (1956) 256.

39 H. S. Shapiro and E. Chargaff, Biochim. Biophys. Acta, 39 (1960) 62.

40 T. TsuMiTA AND E. Chargaff, BiocMm. Biophys. Acta, 29 (1958) 568.

41 A. N. Belozersky and A. S. Spirin, in E. Chargaff and J. N. David-

son (Eds.), The Nucleic Acids: Chemistry and Biology, Vol. Ill,
Academic Press, New York, 1960, p. 147.

42 F. Weygand, a. Wacker and H. Dellweg, Z. Naturforsch., 7b (1952)

19.

43 D. B. Dunn and J. D. Smith, Nature, \1A (1954) 305.

44 S. Zamenhof and G. Griboff, Nature, 174 (1954) 306.

45 D. B. Dunn and J. D. Smith, Biochem. J., 67 (1957) 494.

46 R. M. LiTMAN AND A. B. PARDEE, Nature, 178 (1956) 529.

47 S. Benzer and E. Freese, Proc. Natl. Acad. Sci. U.S., 44 (1958) 112.

48 R. M. LiTMAN AND A. B. PARDEE, Biochim. Biophys. Acta, 42 (1960)

117.

49 H. S. Shapiro and E. Chargaff, Nature, 188 (1960) 62.

50 R. Rudner, H. S. Shapiro and E. Chargaff, Nature, 195 (1962) 143.

51 J. H. Spencer and E. Chargaff, Biochim. Biophys. Acta, 51 (1961)

209.

52 J. D. Watson and F. H. C. Crick, Nature, 111 (1953) 964.

53 D. B. Dunn and J. D. Smith, Biochem. J., 68 (1958) 627.

54 R. Lipshitz and E. Chargaff, Biochim. Biophys. Acta, 42 (1960) 544.

55 D. B. Dunn, J. D. Smith and P. F. Spahr, J. Mol. Biol, 2 (1960) 113.

56 J. Horowitz and E. Chargaff, }\uiut^, i >4 i>59) 1213.

57 J. D. Smith and R. Markham, Nature, 170 (1952) 120.

58 R. L. SiNSHEiMER, J. Biol Chem., 208 (1954) 445.

59 H. T. Shigeura and E. Chargaff, J. Biol Chem., 233 (1958) 197.

60 M. B. HoAGLAND, in E. Chargaff and J. N. Davidson (Eds.), The

Nucleic Acids: Chemistry and Biology, Vol. Ill, Academic Press,
New York, 1960, p. 349.

CHAPTER 10

A Few Remarks on Nucleic Acids, Decoding, and
the Rest of the World *

Men naar Contexten er bleven meningsl0s, saa hjcelper det
ikke med store Overskuelser, saa gj^r man bedst i at tage
Talens enkelte Dele frem; naar Munden slaaer Sladdcr, saa
hjcelper det ikke at ville holde et sammenhcengende Fore-
drag, men man gj0r bedst i at tage hvert Ord for sig**

KIERKEGAARD | EN LITERAIR ANMELDELSE | p. 115

One of our great men is reported to have said: When in doubt
tell the truth. This is an excellent and not much followed precept.
But our science has not been in doubt for quite some time. It
has learned that it is profitable to outsmart nature.

Many of us, when we were young, thought that it was the task
of the natural scientist to understand the ways of nature. But it
has come about in our times that what man desires — or at any
rate what some men desire — is to change these ways. It all started
as a search for truth; but hundreds of thousands, at Hiroshima,
at Nagasaki, paid with their lives for such lovable inquisitiveness.
What an ivory, what a tower!

So I have learned to watch these clever fingers closely and not
to beheve the song of the sirens, even if they are sirens with a
Ph.D. When they tell me: "As soon as we have mastered the
genetic code, we shall know how to cure cancer", my answer is:
"Maybe; but before this, you will probably succeed in converting
humanity into a bunch of mongoloid idiots, by putting just the
right amount of base analogues into the DNA".

* This is the text of a lecture given, in a much abbreviated form, at a
symposium on "Basic Problems in Neoplastic Disease" held at Columbia
University on March 12, 1962.

** But when the context has become meaningless, grandiose surveys
are of no use; it is best to bring out the individual parts of the speech.
When the mouth only babbles, it is of no use to undertake a coherent
lecture; it is best to take each word in itself.

162 NUCLEIC ACIDS, DECODING, ETC.

A few years ago, in 1953, I had the task of introducing a sym-
posium on conjugated proteins, and I should like to cite a portion
of my introduction.

". , , Scientists, like little fishes, swim in schools. When we open one of our
scientific journals these days, we find a very uneven distribution of topics.
Some important fields are almost entirely neglected, others seem to ex-
plode into bursts of unbelievable mediocrity. Really valuable contributions
in the fields most in vogue at present probably are just as scarce as those
dealing with the stepchildren of present-day biochemistry. But not all
disciplines make it so easy to call each mush a "homogenate", each soup
a "partially purified extract", and to speak — when you have nothing
whatever — of a "system". There is a real danger that our science may
suffocate in its own excrements."

This, I fear, has come about. Prepackaged concepts have become
widely used; a form of collage makes it easy to give a pleasingly
abstract shape to what many of us were wont to consider the
most concrete thing in the world. Instant Science: just take a
teaspoonful and add hot air.

Conscience — that safety valve of human hybris — has no longer
a place in science. A spurious, artificially ordered universe permits
the bestial grinding out of facts, many of them actually pseudo-
facts, micro-idols of an Easter Island of the mind. The vox
humana may still be a stop on the organ; but where is it found
in our scientific papers? We are being invited to dig deep holes
to hide from the consequences of our atomic discoveries; but
there will be no hole deep enough, once we have learned to
manipulate the genetic material. Goethe's "Zauberlehrling" has
become the prototype of us all. It will also be nice to watch the
consequences of the attempt to tickle the Universe, as in the
projected atomic explosions in outer space: Daedalus sending up
Icarus to provoke the sun. These infinite spaces of which Pascal
dreamed, is it not time that they came to the end of their infinity?

DNA now is a magic name, the philosopher's stone of our days,
the quintessence of contemporary alchemy. Since I have been
able to follow the development of this remarkable creature from

NUCLEIC ACIDS, DECODING, ETC. 163

before the time when it went to nursery school — it was almost
still in its baby carriage — a few words may serve to paint a
backdrop. For decades, the sciences of biochemistry and of
genetics had each gone their own way, stepchildren in turn of
two older sciences: biochemistry of chemistry, genetics of biology.
What to the one was the molecule, the gene was to the other.
There was no bridge between the two, and they were on the
whole quite happy. It may be remarked that also in the Tower of
Babel the many nations got along quite nicely as long as they did
not talk to each other; bedlam broke loose when togetherness set
in. This point arrived for us between 1937 and 1944 when plant
viruses were shown to be ribonucleoproteins, when the trans-
forming principle in pneumococci was shown to be deoxyribonu-
cleic acid. At about the same time — and I had my share of the
guilt — the concept of "biological information" raised its head and
began to sport a multicolored beard which has become ever more
luxurious despite numerous appUcations of Occam's razor. What
must be remembered is that the entire idea of a chain of com-
mand, of a transmission belt of biological information could
hardly have been formulated, had not the study of viruses made
it possible to place a dateline on the start of development of a
biological system. It is of course possible that we have been
foolish in applying the principle of the unity of nature to such
biological primitives and in extending the results from these
studies to the entire realm of life. There are many regions in
which the unity of nature cannot be invoked. By the way, it is
quite astonishing that there is practically no evidence for the
existence of such chains of biological information in higher forms
of life, in multicellular communities, as have been derived from
the study of a few viruses and micro-organisms. Even the loudly
proclaimed "universal genetic code" rests, for the time being at
least, mostly on highly brain-washed E. coli.

It can, however, not be denied that very impressive advances
have been made in the last fifteen years in the chemistry, the

164 NUCLEIC ACIDS, DECODING, ETC.

physics, and the biology of the nucleic acids, and even more so
in what might be called their metabiology or, perhaps better,
their parabiology: a field that could best be described as the
borderline between genetics and mythology. Some of the alle-
gorical paintings illustrating our views of the role of the nucleic
acids compare favorably with the products of the minor masters
of the rococo.

Perhaps the first advance that was made in the chemistry of
the nucleic acids was the elaboration of sufficiently precise micro-
methods that permitted the quantitative analysis of very small
amounts. This was accomplished between 1946 and 1948 and it
truly formed the basis of all later work; for it served to demon-
strate the existence of a very large number of different nucleic
acids and did away with the old tetranucleotide hypothesis. But
before going on, I should pause for a moment and remember the
names of those that came before, names that our time — long in
bibliography, but short in memory — likes to forget: Miescher,
Hoppe-Seyler, Piccard, Kossel, Altmann, Neumann, Jones, Steu-
del, Feulgen, P. A. Levene, Thannhauser, Hammarsten, Jorpes,
Gerhard Schmidt, and Dische. Great work was done by them and
by others, and even without the time-saving expedient of silly
abbreviations.

The next thing that the chemical investigation of the nucleic
acids accomplished was to demonstrate the species specificity of
DNA, i.e., to show that the DNA of a given species had a
characteristic and unchanging nucleotide composition; a com-
position that appeared to be characteristic of the species, but not
of any particular organ. And the third thing was the observation
made in 1949 that there existed in all DNA specimens very
pecuUar base-pairing regularities. These observations, together
with the X-ray studies four years later, had as a direct conse-
quence the formulation of DNA as a regularly paired double
structure.

Many of you will undoubtedly be aware of the numerous
speculations about the GC content of DNA and its meaning for
the physical and biological properties of the polymer, but few

NUCLEIC ACIDS, DECODING, ETC. 165

will know that the distinction between the AT types and the
GC types of DNA was first made here at Columbia, just as was
the introduction of such much used terms as the hypochromicity,
the denaturation, the heterogeneity of nucleic acids and many
other things.

Now for some of the problems posed by the chemical structure
of a single- or double-stranded polymer as it is assumed to exist
in DNA; and much of what I have to say appUes also to RNA.
Glancing at an arbitrary segment from such a very highly poly-
merized macromolecule, we see that the nucleotides are linked 5'
to 3', that their arrangement is arrhythmic, but that the com-
plementariness regularities are maintained. The lowest estimate
of molecular weight of such a polymer is about 6-10^, except
for a few small viruses; the highest estimate has no limit: the
latest quotations on the market were around 180-10^. Even with
only ten thousand nucleotides per strand there are 10^^^^ isomers.
We deal with four or five components; but the "alphabet" is not
as meager as it may appear, for these few letters spell out
"words" that are very long indeed: ten thousand letters or more.
There could, therefore, be plenty of information for the unhappy
cryptographer. But what is impUed in such a statement as "DNA
is the repository of biological information"? This is, of course,
a very convenient slogan and may be taken as another example
of the motorized anthropomorphism that disfigures much of our
biological reasoning at present. The concept postulates the exist-
ence of a chain of information in the cell which is thought to be
represented by a series of signals, as it were, emanating from a
specific DNA and transmitted by a specific RNA messenger of
matching composition to the unspecific ribonucleoproteins of the
ribosomes, thus instructing the latter to proceed to the manufac-
ture of specific proteins. A simple-minded chemist could, of
course, ask himself whether this was more than a terminological
scaffold conceaUng our ignorance, something of the order of an
allegory, pretty but awfully vague.

But there can be little doubt: sequential specificity of high
polymers of cellular origin is, as we have learned in the recent

166 NUCLEIC ACIDS, DECODING, ETC.

past, one of the principal marks of the specificity of the cell. The
approach to the problem of the sequence of nucleotides in a
nucleic acid is, however, much more difficult than in the case of
the proteins. The main reason is that we still lack a sufficient
arsenal of different enzymes breaking predictable internucleotide
bonds in a specific manner. So far, we have been compelled to
rely mostly on the differences in stabiUty of the purine and
pyrimidine nucleosides when exposed to chemical degradation.
It is indeed very regrettable that amazingly little work is being
carried out at present on the search for new and specific
nucleases.

The basic fact which is at the origin of all our sequence studies
on DNA is the great difference in stability towards acid exhibited
by the purine and the pyrimidine nucleosides. Whereas the
glycosidic link between purine and sugar is cleaved with little
difficulty, pyrimidine nucleosides are quite stable under these
conditions. For this reason, it is relatively easy to deprive a
DNA polymer chain of its purine complement without disturbing
the arrangement of the pyrimidine nucleotides. We have called
degradation products of this type, i.e., chains which contain free
sugar phosphates at the places where purines were, apurinic acid.
Additional, slightly more energetic, degradation will now release
the various pyrimidine clusters occurring in the original DNA.
We refer to pyrimidines that find themselves interjacent between
purines as "solitary" pyrimidine nucleotides; longer pyrimidine
runs can be designated as "bunched", or more particularly as
"couples", "triplets", etc. For purposes of such an analysis, which
we have referred to as differential pyrimidine distribution anal-
ysis, it is not necessary to distinguish between the two strands
of a double helix, if it occurs, since such a DNA polymer could
be viewed as being joined at one end, or even at both ends, so
as to produce one continuous polynucleotide. The analysis is
based on the differential release of pyrimidine nucleoside 3',5'-
diphosphates by /^-elimination and hydrolysis.

The principal conclusions from a large series of such studies
are: (1) The method permits the distinction of DNA specimens

NUCLEIC ACIDS, DECODING, ETC. 167

that are indistinguishable by the analysis of their total purine and
pyrimidine composition. Nucleic acids having identical total
nucleotide contents, though derived from different organisms,
differ widely in sequence characteristics. (2) A large proportion
of the pyrimidines — and therefore also of the purines — occurs as
tracts of three or more in a row. All-thymine sequences of five or
even longer occur in many instances, but cytosine has so far not
been seen in units larger than triplets. (3) The nucleotide ar-
rangement appears not to be random. This could be taken as an
indication that a DNA chain could indeed represent a meaningful
text. But is a text without a reader still a text? "What is the sound
of one hand clapping?" asked the Zen buddhists.

Which are, one might ask, some of the immediate problems in
the chemistry of the nucleic acids? First, a better understanding
of what is meant by a DNA molecule. Is the apparent heteroge-
neity of DNA ostensible or real? Is there one giant molecule of
DNA per chromosome — of a molecular weight of hundreds of
milHons — or are there many, different both chemically and func-
tionally?

Secondly, if we really deal with giant chains, do they consist only
of polynucleotides or are many tracts, each consisting of several
hundreds of nucleotides, held together by bridges that are not of
nucleotide character? The fact that DNA is depolymerized by
alkali to smaller polynucleotides, but still of respectable size,
would seem to indicate the second possibility. I have been inter-
ested in this problem for years, but with inconspicuous success.
There is an insuperable difficulty in demonstrating, among
hundreds or thousands of components of one kind, one of another.
It is very hard to distinguish between a trace constituent and a
trace contamination. One could think of many types of possible
bridging compounds. A derivative of a hydroxyamino acid, such as
phosphoserine, or a bifunctional amino acid, e.g., glutamic acid,
could form a diester, or an oUgopeptide could do the same.
If something of the order of glycerophosphate, ribophosphate, or
even a ribonucleotide, were inserted into the chain at infrequent,
but possibly predetermined, intervals, regions of reactivity dif-

168 NUCLEIC ACIDS, DECODING, ETC.

ferent from the bulk of the polymer would be created. Alternative-
ly, it would be sufficient for a purine or pyrimidine to be missing
occasionally, so that at these places a deoxysugar phosphodiester
occurred; in this manner less stable links would be introduced at
which the chain could cleave via elimination or cyclization. The
problem of the existence of discontinuities in the DNA chains is
not without general interest, since these could serve to explain
how such a polymer, and even a circular structure, could rupture
at fixed and predictable positions, as has sometimes been assumed
in bacterial and phage genetics.

And thirdly, and perhaps most importantly, if there really is
a transcription belt starting from DNA, as I have mentioned
before, what are the chemical mechanisms that are at play? How
are the precursors selected and held in place, how are nearest-
neighbor preferences exercised? There is a real problem in the
question by what forces the precursors of a polymer — nucleotides
or amino acids — are held onto the "template" before they are
polymerized enzymically. Such silly metaphors as "the zipper"
will not do. Anyone who has had trouble with this useful con-
traption knows how important the rigid aUgnment of "precur-
sors" is for its proper functioning.

This brings me to the last topic I wanted to mention. Majority
opinion — which means, as usual, a small minority with a strong
voice — postulates that DNA or RNA carries a code which is, in
the last resort, translated into the specific amino acid sequence
of a protein. This code is said to be non-overlapping, composed
of nucleotide triplets that are read from a fixed starting point,
and "degenerate", i.e., several signs may have the same meaning.
It is also thought to be universal.

"Code - 2. Any system of principles or rules; as, a code of
ethics. 3. A system of signals for communication by telegraph,
flags, etc.; also a system of words or other symbols arbitrarily
used to represent words; as, a secret code."" {Webster's New
Collegiate Dictionary). — Number 2, I decided, obviously does

NUCLEIC ACIDS, DECODING, ETC. 169

not apply, so it must be No. 3. But are we concerned with words
here, and is arbitrariness what we would ascribe to nature? The
understanding of words, the translation of ciphers, even the
reception of signals, they all require the intervention of intel-
ligence, either directly or indirectly. I had been interested in this
problem for quite a while, long before I read Teilhard de Chardin
from whose book {Le Phenomene Humain) the old saying of
TertuUian seemed to arise in a new and disturbing version:
Animal naturaliter christianum. He found consciousness, i.e.,
mind, everywhere, though in different degrees, in the mineral and
the protein, in the virus and in man. Now, this is a dangerous, a
slippery path; and I am afraid to take even the first step on it.
For if I fear anything, it is the mystic with the shde rule. Poets
look better without their laboratory coats. The vapor pressure of
pneuma should remain unmeasured. To study the effect of mental
concentration on reaction velocity I leave to the next century.

In any event, when the word "code" is used in the context of
cellular specificity, most people will know what it signifies in a
loose sense, though many may be baffled by its concrete meaning
and even more by the manner in which it is supposed to operate.
For this reason, we were all elated — though perhaps with dif-
ferent degrees of fervor — by the announcement, a few months
ago, that the amino acid code had been broken through the dis-
covery that several uridylic acids in a row, perhaps three or more,
coded for phenylalanine. These findings were soon confirmed and
extended in other laboratories, so that it would now seem that
the coding units for all amino acids are known, though there still
are some uncertainties, especially with regard to the sequence
within the triplets. There were, of course, some very baffUng
circumstances, e.g., the unexpected preponderance of uracil in all
coding units and also the observation that poly-U was enormously
more effective in incorporating phenylalanine than were any of
the other coding units for their respective amino acids.

We had a very direct interest in all these questions, for we
were determining the frequency of the all-pyrimidine runs in
various species of DNA; and there should have been a direct

170 NUCLEIC ACIDS, DECODING, ETC.

relationship between their abundance — and also between the
DNA type, AT or GC — and the coding efficiency of a DNA.

But before engaging in much work, I thought the situation
over. And a short time ago I underwent a sort of conversion
which lasted for one week. I decided to believe everything — but
everything — that I read in the literature. I swallowed all bio-
chemical snapshots, preliminary communications, all press re-
leases that punctuated the, more or less, friendly competition
between different laboratories — in other words, the entire sour
grapevine. I followed the international spelling bee with open-
mouthed admiration. I became what you might call a molecular
fundamentalist. And this is what I encountered during my brief
pilgrimage through the bible-belt of molecular biology. The
dogma stated: DNA makes "messenger" RNA, makes protein.
"Fine", I said; but though temporarily a True Believer, I had
not stopped being a chemist. So I reasoned: "Messenger" RNA
is said to be a complete image of DNA, mirroring its base-pairing
and other regularities; A = U; G = C, and so on. The amino
acid composition of many proteins is known precisely, and in
many instances also the amino acid sequence, thanks to the excel-
lent work of the past few years. The amino acid code has been
published, at least in the newspapers. So let us reconstruct the
messenger coding for a given protein. This scheme was the more
inviting, since the determination of the nucleotide sequence of an
RNA by direct means is beset with very great difficulties.

TABLE 37

BOVINE PANCREATIC RIBONUCLEASE

Number of amino acids: 124

Calculated composition
of -messenger" RNA ^^ ^^^' ^^

Ratios

A63 A 16.9
G59 G 15.9
C,7 C 20.7
U,,3 U 46.5

Pu/Py
6-Am/6-K

A/U
G/C

(A + U)/(G + C)

0.49
0.60
0.36

0.77
1.7

NUCLEIC ACIDS, DECODING, ETC. 171

First, I chose pancreatic ribonuclease, and the results are
shown in Table 37.

You will notice that there are many surprising features as
regards this fictitious RNA. It does not look at all like anything
that we are familiar with. Almost one half consists of uridylic
acid; the pyrimidines amount to two thirds of the whole; the
ratio of 6-amino to 6-keto nucleotides is only 0.6. One could not
be farther away from an image of DNA. It turned out that one
could; for I chose as my next example a protein closely associated
with DNA in a sperm cell of which it constitutes a considerable
proportion.

I took salmine, the protamine that, next to DNA, is the prin-
cipal component of salmon sperm (Table 38).

This reconstructed "messenger" looks even more unreal. It
contains only one adenylic acid in 174 nucleotides. Even if

TABLE 38

SALMINE

Number of amino acids: 58

Calculated composition

of "messenger"

RNA

As mole %

Ratios

Ai

A 0.6

Pu/Py

0.40

G49

G 28.2

6-Am/6-K

0.49

C56

C 32.2

A/U

0.01

Ues

U 39.1

G/C

(A + U)/(G + C)

0.87
0.66

TABLE 39

SALMON DNA

Mole %

Ratios

A

29.7

Pu/Py

L02

G

20.8

6-Am/6-K LOO

C

20.4

A/T

L02

T

29.1

G/C

1.02

(A + T)/(G + C) 1.43

172 NUCLEIC ACIDS, DECODING, ETC.

one assumed that the "messenger" is produced in the form of
two, presumably antiparallel, strands as an image of the particular
segment of DNA, of which only one carries the "coded" infor-
mation, a sequence of salmon DNA which in a tract of 174
nucleotides contained only one adenylic acid looks most unreal
on many grounds. For instance, salmon DNA is of a quite
marked AT type (Table 39).

You will notice that in salmon DNA almost every third
nucleotide has a good chance of being adenylic acid. How
probable is it then that this nucleotide occurs only once in such
a long stretch? Moreover, even a glance at the published "code"
will convince us that there can be very few proteins that would
demand a less implausible messenger. We are, hence, left with
the assumption that each double segment of DNA can specify,
via two RNA chains, two different amino acid sequences of
which only one is realized. But what does the other half of "mes-
senger" do and what is its fate? Surely, nobody would like to
put forward the inane slogan: One gene, two enzymes.

Another explanation, but not a very good one, of these dif-
ficulties may be seen by some in the assumption that the code is
"degenerate", i.e., more than one triplet may specify one amino
acid. It should, however, be pointed out that if the "degeneracy"
took the form of having one triplet coding for a given amino acid
in one species, but another coding unit for the same amino acid
in another species, the proclaimed universality of the code would
vanish. And then there is, as always, still another possibility,
namely, that the whole thing is wrong.

In any event, when I saw aU this I was very baffled. I concluded
that what I had tried to do will not be a shortcut towards a
solution of the problem of nucleotide sequence in RNA and
DNA. I went home and lost my faith.

5

Most scientists like to believe that we live in a cosmos, i.e., in
a universe having order, harmony, purpose. Many feel that in this
world all hangs together, though few will presume to explain how

NUCLEIC ACIDS, DECODING, ETC. 173

it does so. "Pound St. Paul's Church mto atoms, and consider
any single atom; it is, to be sure, good for nothing: but, put all
these atoms together, and you have St. Paul's Church." (Johnson
in Boswell's Life, July 20, 1763). You can see, Samuel Johnson
was what you might call a Premature Molecular Biologist; other-
wise, he could have considered the possibility that reassembly
may yield Liverpool Station. Whether his Umpid eighteenth-
century mind would have enjoyed the gnostic stew that is being
served up at present remains, however, questionable.

We have become insensible to the incredible amount of noise
and tribal incantations that accompany scientific research in our
days. "Tolle, lege!" said a voice to St. Augustine. But if in our
time he had, following this injunction, picked up and read one of
our foremost picture magazines devoted to illustrating the more
appealing and intimate sides of our public personalities, and
therefore hard up for copy, this is what he would have seen as a
headline in a recent issue: "Will man discover *the molecular
structure of God'?" If anything can turn the stomach of a saint,
this is it. Has our time forgotten how heavy a word can be?

CHAPTER 11

Amphisbaena

. . . et tu in te manes, nos autem in experimentis volvimur?

AUGUSTINUS I CONFESSIONES | IV, 5

'Tis all in peeces, all cohaerence gone;
All just supply, and all Relation.

DONNE I AN ANATOMIE OF THE WORLD
THE FIRST ANNIVERSARY, 213-4

Nay, so devoted are we to this principle, and at the same
time so curiously mechanical, that a new trade, specially
grounded on it, has arisen among us, under the name of
"Codification" , or code-making in the abstract; whereby any
people, for a reasonable consideration, may be accom-
modated with a patent code; — more easily than curious
individuals with patent breeches, for the people does not

need to be measured first.

CARLYLE I SIGNS OF THE TIMES (1829)

Two men sit on a bench, in August 1961, an Old Chemist (O)
and a Young Molecular Biologist (Y).

o:

Now that I have you here for a few minutes of quiet talk, I shall
start by saying: The cell is not a machine.

y:
But what kind of a start would that be? Do you want me to agree?
This would be surely stupid, for there are all sorts of machines,
and you have probably never heard of the theory of automata.
I should much rather turn your saying around and offer: A
machine is a cell.

o:
We are then already in the middle of model building, the favorite
occupation of modern biophysics. It is all done in front of mir-

AMPHISBAENA 175

rors, with wire and plastic, glue and papier-mache; the knowledge
of a child combined with the naivete of the grown-up.

y:

But really, why are you so much against machines?

o:

Of course, I am not; some of my best friends are machines. But
I am very much against a strictly mechanomorphic view of
living nature. A machine is a deterministic construction; some-
one— an inteUigence — has had to make it; and even if it could be
"programmed" to make itself, who did the programming? A self-
reproducing machine that was not built by a primordial engineer
is an abomination thrown up by turbulent and sick times which,
in chiliastic dreams mixing superpower and impotence, have
created a mythology of a most maculate conception. I shall not
formulate the secret of life sitting on this bench on a summer day;
but I can say that we shall not have a satisfactory theory of
biology before we have learned to combine, in one concept, the
dialectic play between determinism and accident — a sort of
random nonrandomness — which seems to characterize the living
and reproducing cell.

y:
It would seem to me that you are simply an angry old man
preaching some sort of dialectical biochemistry in which the
observer is walking simultaneously on both sides of the street in
opposite directions. One could call this also schizophrenic.

o:
Well, as to being an angry old man, there is plenty to be angry
about, and it makes more sense for an old man to be angry than
for a young one; like most things, anger must be earned. King
Lear was an angry old man. We are no longer used to passion
in the natural sciences; it has been replaced by ambition. Our
young geniuses are passionately ambitious instead of being pas-

176 AMPHISBAENA

sionately passionate; and it has become very difficult to distin-
guish between what is an ardent search for truth and what is a
vigorous promotion campaign. What started as an adventure of
the highest has become the survival of the slickest or the quickest.
"Cloak and dagger" has changed to "cloak and suit". We now
have DNA tycoons and others have "made a kilHng" in RNA.
A generation of scientific quiz kids knowing the answer to every-
thing. But to throw pearls before young molecular biologists is
not the purpose of this talk. Let's return to the beginning and I
will ask you "What is molecular biology?" Now, if physics is the
science of the states, and chemistry that of the conversions, of
matter and biology comprises the application of their laws to
animate nature, what could be meant by molecular biology?
You can, of course, apply any adjective to any noun, but the
results are most often bizarre.

y:
Here you go again. I could, of course, be witty, as you think you
are, and say that molecular biology is what is published in the
journal carrying that name. All academic gowns carry, strictly
speaking, the color of their respective payrolls —

o:

Here I interrupt already. Your jocular definition cannot hold,
since the irregular rate of appearance of this admirable journal is
vastly inferior to that at which molecular biologists are produced
nowadays. My definition, incidentally, would be that molecular
biology is essentially the practice of biochemistry without a license.

y:

This is, as you must know, a completely frivolous definition.
There is naturally much more to molecular biology. In order to
prevent you from going on endlessly, I shall agree that the
naming of a new science — or a new name for an old one — has
also its practical reasons. Symposia, congresses, new journals,
more money to be had more easily . . .

AMPHISBAENA 177

o:

And the feeling to be a pioneer at no extra cost. And how the
publishers love these new handles by which to extract the old
money. In fact, this is not the end: soon we shall have molecular
sociology, molecular history, and a little later perhaps molecular
theology. The fragmentation of the sciences proceeds via adjec-
tives.

y:

I believe, even among your age group you are a horrible excep-
tion.

o:

Very likely. In a time in which everything that is new is true,
the older the fogy the more he must galvanize himself into being
able to participate in the exuberant initiation dances that greet
each new molecule before it is replaced by an even newer one.
But I must say, you can't escape senility by trying to become a
juvenile delinquent.

y:
I shall ignore this. What I said a moment ago about the advan-
tages of having a new science was not meant satirically. This is
the way sciences grow in our day; they have become a mass
movement, a corporation in which the majority of the population
carry stock. Their development must be studied by the methods
of sociology.

o:

Another stew of a science.

y:
Let's slaughter one at a time. To return to our subject — if we
still have one — ^this is, I believe, the way our new science came
about. It started with the recognition of the importance of
macromolecules in biology, of their chemically distinct and pre-

178 AMPHISBAENA

cisely describable structures, and of the specific functions which
very often could be clearly assigned to them. Take the en-
zymes . . .

o:

I will take them. But who did the assigning? You isolate a
protein, you ask it a very limited number of questions by bringing
it together with a few substances. If it happens to react with one
or the other, you call them "substrates" and you proceed to
assign a specific function to this particular protein. Did you
notice the man that passed us a moment ago? He limped; and
you would rightly say that this was because he had one shorter
leg. But how could I refute someone who claimed that this man
had one leg too short in order to limp? You say that this enzyme
is present in the cell, in order to perform this reaction. Hie
Rhodus, hie salta! But maybe this is not Rhodes or maybe he
does not jump, or maybe he jumps somewhere else. How sure
can you be, for instance, that all the various phosphatases that
can be fished out of a cell mush really act as such in the living
cell, that this is their "function"? Ours — and yours — still is a
post mortem science; we are forced to destroy the overriding, the
overpowering category of life.

y:
Don't tell me you are a vitalist.

o:

Of course, not. But a discussion between us about the meaning
of life would be stupid: I am too rigid and you are too flexible.
All I can say is: Life is what's lost in the test tube. Better tell me
what you think about what I said concerning the assigning of
functions.

y:
I think there are such enzymes and such. Many have the functions
that we have given them, others may not be called upon to act

AMPHISBAENA 179

under normal circumstances; they may be a memory that the cell
has retained of past events. But surely even you cannot deny that,
for instance, the energy relationships of the cell and their en-
zymic basis are well understood and are the best example of the
concept of the unity of biochemistry?

o:

This I will not deny, though I cannot say that I am particularly
fond of this concept of unity. It has done a lot of mischief. Life,
as we know it, seems to be characterized by two antithetical
principles: unity and diversity. And it is very difficult to decide,
in a concrete case, with which of these principles we are dealing.
The unity of nature is usually laboriously pieced together by
combining, in a spuriously serial or cyclic fashion, snips and bits
taken from the most widely distant organisms. It is a sort of
biomontage. The fact that we all live and die in one and the same
world should not conceal another fact: that we are all different.
Even life is only one, and a minor, form of nature: a tiny foam
on the crystals of the earth. For an old man who is in love with
what goes on around him and with the multiformity of its ap-
pearances there can be only one battle cry: Vive la difference!
Is it really so important that both torturer and victim enjoyed the
same hearty meal and digested it in the same manner? Besides,
similar mechanisms do not always foreshadow similar functions.
The combustion oven designed by Liebig and the combustion
ovens constructed in Auschwitz, though based on the same scien-
tific principles, can hardly be taken as a proof of the unity of
nature. But your little speech about the processes of parturition
that resulted in the forceps delivery of molecular biology stopped
at the enzymes. Perhaps you go on.

y:

What you like to call my little speech did not stop, it was stopped.
In any event, the recognition that polypeptides of high
molecular weight could show great specificity as enzymes in
metabolic reactions and as antigens or antibodies in immunology

180 AMPHISBAENA

led to many studies of their physical and chemical structures.
Ultracentrifugation, X-ray diffraction, electrophoresis, electron
microscopy, the many wonderful separation methods carrying the
name of chromatography: all these were instrumental in bringing
about these exciting times. It became possible not only to analyze
proteins completely but even to determine their amino acid
sequence in many cases. The triumphs of structure determination
were so great that everywhere young people were flocking to the
banner of the helical nature of things.

o:

You are spiraling a little too fast for me. Even Lucretius had to
stop for breath. Your dithyrambic tone reminds me that not long
ago I have heard a gushing voice on the radio speaking of "DNA,
this miracle monocule".

y:
What of it? Some people have difficulty pronouncing "molecule";
but this will change. Just as there were times when people found
it easy to say "transubstantiation". As to DNA, I shall soon come
to it. Let me continue. The advances I mentioned a minute ago
were, of course, not the only ones. Even before they had come
to fruition, studies had begun on other fronts. Certain viruses had
been purified sufficiently so as to warrant their chemical exami-
nation and had been recognized as nucleoproteins. First came
the plant viruses which were found to contain RNA, as do also
most of the animal viruses. The bacteriophages, on the other
hand, contain mostly DNA. But perhaps the most exciting thing
that happened was the discovery that microbial transformation —
a phenomenon known for quite some time — was due to specific
forms of DNA. This had the effect of placing DNA very near to
what the geneticists at that time were still calHng the gene.

o:
If you don't mind let us still stick to this hoary term. I know, the
packaging has been changed lately; but since it still is the same

AMPHISBAENA 181

old thing, why not conserve the convenient brand name? After
all, if you want to describe a syndrome you must first give it a
name. The one advantage, for instance, that I can see in using
the term "molecular biology" is that it puts nearly all that is
unknown in biochemistry into one convenient corner. But I
wanted to say a word about transforming DNA. I well remember
the great excitement with which I followed the first discoveries
concerning specific pneumococcal deoxyribonucleic acid. Then
we still spoke of "desoxyribonucleic". In fact, this was probably
the main reason for my becoming interested in the nucleic acids
at about that time, 1944 or 1945. But there has been a consid-
erable period since then and not so very much has happened in
this field; and I must confess that my misgivings about the
justification of expanding these few observations on micro-
organisms to the entire realm of life have been growing at an ever
increasing rate. Generalization has its definite uses in science;
without it we should all soon be without a job. But at the same
time, there is a great danger of its gliding into glibness. It was in
1889 that the great Swiss historian Jacob Burckhardt wrote a
letter to a friend in which he warned of the oncoming of what
he called "les terribles simplificateurs". Just as the locusts, once
they are through with a field, have simplified it horribly, could
we not say that this is also true of some of the great generali-
zations in biology? Color and variety, the pulsations of accident
and fate, the tremendous urges and instincts, the pendulum of
birth and death: all have disappeared and we are left with what
I once called "a plantation of match sticks". So, when I Usten to
the arguments that microbial transformation proves the gene
character of DNA, I must ask: is this discovery one of the
features of the unity of nature or one of the facets of its diversity?
That's where the dialectics has to come in to which we referred
at the beginning of our little disputation.

y:

I can see, you want to keep your cake and sell it. You are an
intolerable mystic, and I have had the feeling that, while you

182 AMPHISBAENA

speak of nature, there should be a slight music of harps. Why do
you keep plucking the feathers out of the goose that lays golden
eggs for all of us? After all, you are one of the classics in this
field.

o:

Thank you, I should rather not be. In your definition, a classic
in science is a man who does not have to be quoted any longer.
For the pickpocket, the man with the widest pockets is a classic.
And it has been remarked that Banquo is seldom quoted in
Macbeth's papers.

y:

All right, then I shall call you a Cassandra in long pants. But my
story was not yet entirely finished. I had been talking about trans-
forming DNA; and if you had not broken in, I should have con-
tinued by mentioning other instances in which nucleic acids were
assigned direct roles in the determination of hereditary properties.
For instance, there are the bacteriophages attacking E. coli which
seem to do it by injecting their specific DNA into the bacterial
cell. This is enough to set off an entire chain of events terminating
in the production of many phage particles and the rupture of the
host cell. And then we have the plant viruses; they contain
specific types of RNA which are infective; that means, the
nucleic acid itself, when applied to the plant, can give rise to the
formation of innumerable complete virus particles. These are all
definite and specific molecules exerting definite and specific
biological effects; and here you have the quintessence of our new
science, molecular biology. But this is not all.

o:
I am sure, it is not. Even to sell soap nowadays, you need an
a cappella choir. What baffles an old-fashioned chemist in all this
is what has become of the chemical concept of a molecule. I have
heard people of a still older generation claim that this concept
ends with the applicability of Avogadro's law. But never mind.

AMPHISBAENA 183

In any event, it must have been a pleasant surprise for some
biologists to learn that at last they were dealing with molecules.
I should have thought that this was what they had been doing all
the time. It reminds me a bit of Monsieur Jourdain in MoUere's
Bourgeois Gentilhomme who is astonished to hear that what
he has been speaking all his life is called "prose".

y:
Surely, you must have heard the term "molecular disease"?

o:

Alas, I have; and for a time I thought that this was a disease to
which molecules were particularly prone, a sort of molecular
measles. But I soon learned that this was another symptom of
the general sloganification of science, everything — like in a poor
cartoon — having a pithy label hanging from the mouth. Some of
these slogans may have been convenient or useful at one time,
such as "energy-rich phosphate bonds", "dynamic state of body
constituents"; of others, for instance, "messenger RNA", I am
less sure. But there are so many of them, and they are so glit-
tering, so gUb! The more I hear them, the less I enjoy them. The
pecuUar relationship between names and understanding has often
been discussed, best perhaps by Mephistopheles.

y:

Great concepts require great names.

o:

Or perhaps great names can substitute for great concepts. But I
beUeve, you had not yet finished.

y:
This is correct, for I now have to introduce the crowning con-
cept, namely, "biological information".

184 AMPHISBAENA

o:

Do you mean to say that life itself has now acquired a press
agent?

y:

Of course not; but there has arisen all over the world a group of
young, militant and successful practitioners — evangelists, you
might call them — who are spreading the new knowledge with
devotion and perseverance. All seems to fit.

o:

Or at least you disregard what doesn't. Such must have been the
atmosphere when Phlogiston went strong: all seemed to fit until
a little balance was brought into play. Even the great Ockham
with his immortal razor would no longer succeed in shaving the
multicolored beard that is now sprouting all over the beautiful
face of biochemistry. Opening one issue of a journal, I am as-
sailed by an abysmal cackle of terms, ill-fitting, yet strident,
garish and banal. The shockate, the grindate and the sonicate,
the suicide and the abortion, the fingerprints and the hot spots,
the repressors and the co-repressors, the feedbacks, the pools
and the templates, the regulators, the operators and the operons;
and floating over the entire allegorical cesspool, the mysterious
messengers, angelic or diabolic in their evanescent every-
whereness. Have we really arrived at the stage of non-objective
biochemistry, of molecular action painting?

y:
You have left out the hybrids and many other things; but I am
glad to see that you are fairly familiar at least with our nomen-
clature. Let me say, however, that to make a scientific revolution
one must break many eggheads. Your objections are ineffective
and they will not count in the end. Biological or, if you prefer,
genetic information is an extremely important and useful con-
cept which even a Shakespearean fool cannot laugh off. Before
I go on, I should like to ask you a question. When trying to

AMPHISBAENA 185

understand the life of the cell and the functions of its individual
parts, don't you believe in the division of labor?

o:

Frankly, I don't; at least not in the crudely mechanomorphic
way — the wheels and the gears and the levers — in which this is
usually done. Even if Nature were one gigantic servo-mechanism,
I am afraid the beards of the cybemeticists, entrusted with
servicing it, would get into the way of the feedbacks. In the living
cell there must be a way in which quantity — or, better, density
or compression — regulated on an as yet undescribable time scale,
becomes a new and unique quality, namely, that of life.

y:

I am afraid, you are, after all, a vitalist. Returning now to this
business of information and disregarding all you have said, this
is what it amounts to. We do believe in the division of labor in
the cell, with each part, yes, with each molecule, having a definite
and recognizable function. And we believe in the existence of a
strict hierarchy.

o:

I know, I know. Mix anything with everything in the right propor-
tions and the resulting puree will say: Papa! But tell me, since
you mention the hierarchy of the cell, reading the recent literature
I get the impression that the cell is a society of slaves that have
no master.

y:

Not at all. In the beginning was DNA . . .

o:
I hear the start of a new apocryphal gospel with DNA as the
logos of our times.

y:

Is it possible that you don't believe in DNA?

186 AMPHISBAENA

o:

If I don't believe that the moon is made of green cheese, does
this necessarily mean that I don't believe in the moon?

y:

Anyway, DNA is the genetic material which in the last resort is
responsible for the maintenance and the transmission of the
hereditary properties of the cell. We know about the mechanism
of its replication which is an ingenious deduction from the
generally accepted structural model of DNA: a double helix
composed of two intertwined polynucleotide chains. After the
separation of these strands which is very easy to explain, as it
offers no thermodynamic difficulties and can be carried out with
a model costing less than a dollar, each strand now proceeds to
the production of its counterpart.

o:
How general is general? And without wanting to offend your
sense of economy, thermodynamical or otherwise, may I ask
whether you mean this beautiful scheme to apply to meiosis as
well as to mitosis?

y:

I am not interested in diploids.

o:

I wish, your parents had felt the same way. And would you say
that during the splitting of a chromosome the other constituents,
the proteins, the lipids, undergo similar processes, or do they
simply follow the leader out of family affection?

y:
Let's not waste our time with triviahties. Who cares about lipids
anyway? The smart cookies leave them alone. I shall give you
here a purely formal scheme which you can take or leave; but
you would be well advised to take it.

AMPHISBAENA 187

o:

I know, scientific dogmas are phagocytes which eat only what is
good for them. They cannot be refuted or dethroned; but they
vanish eventually owing to the fickleness of subsequent genera-
tions who lose interest in them. In fact, the more absurd a hypoth-
esis is, the stronger must be the belief in it.

y:

I continue. DNA is, as I said, the primary genetic determinant
carrying a code, as yet unbroken, through which in the last resort
the composition of RNA and of the proteins is specified. Before
this code can be expressed, the two strands composing the double
helix of the DNA must be separated, unwound, perhaps enzymat-
ically. I believe, it was you who called this hypothetical enzyme
an "unscrewase". You will observe that, because of the com-
plementary structure of the two strands, the information stored
in either one really is sufficient.

o:
Have the two strands been shown to exist, let alone to have a
complementary structure?

y:
Well, yes and no. But you should be the last to ask such a question.

o:

Sometimes I wake up in the dark of the night and I begin to think
of all these claims and discoveries and models, of aU this molec-
ular prestidigitation; and I ask myself: Is this all a confidence
game? They are all so brilliant; why are they so shallow? Why
did the manna of the heavens tu|Ti into porridge? Why does the
liquidation of a science begin at the top; why do its greatest
triumphs turn into its worst disasters?

y:
I don't believe you expect an answer. AU you have to do is to
take a spectra of a DNA before and after heating.

188 AMPHISBAENA

o:

Now I can see that you are a modem young scientist. The use of
these false singulars, a spectra, a media, a bacteria, a phenomena
or even a phenomenum, etc., is this not the summa cum laude of
the new generation?

y:
I have never found Latin useful in my work. If you, too, had
studied more mathematics and physics, instead of wasting your
time, you would have accomplished more.

o:

This may very well be so. I have felt for some time that your
Ph. D. should be spelled pH D.

y:

In any event, this is how it works. There is no doubt at all, the
direction is: DNA to RNA to protein; but never backwards. The
genetic signals emanating from the DNA go first to a special kind
of RNA which turns over very rapidly, never amounts to more
than a few per cent of the total RNA of the cell, and mimics,
in its composition, that of the DNA. Since it carries the message
contained in the DNA, we call it the messenger RNA.

o:

This small and short-lived Corps of Commissionnaires, has
anyone seen it?

y:

What do you mean by "seen it"? There is plenty of indirect
evidence that such an RNA is formed.

o:

No, I mean the message, has anyone shown that it exists? Is it
not possible that the entire imposing terminological scaffold is
nothing but a suitcase for the emperor's new clothes? Is it not

AMPHISBAENA 189

possible that there is no message, no messenger, that the entire
question is asked, and therefore answered, wrongly?

y:

If you deny that there is such a thing as "information", you will
still have to get the same fellow, but call it by a different name.

o:

You want to say what is usually said when one has no answer,
that this is merely semantics?

y:

Yes. Incidentally, messenger RNA is not the only name; others
have referred to it as informational RNA or pulse RNA. The
men that really discovered the stuff gave it no name and will
therefore be rightly forgotten.

o:

This is true, never be a pioneer; not who comes first counts in
science, but who comes last. But let us stay with these charming
conceits for a Htde while and try to think things through. I know,
in our days what cannot be done indirectly, isn't being done at
all any more. But chemistry is, after all, the science of substances;
and if all this is true there must be a substantial basis to it. You
said that it all starts with DNA. This DNA, I take it, exists in the
cell as a double-stranded heUx, as what people with a brilliant
gift for vulgarization have called "the coil of life". The two
strands run in opposite directions with respect to their terminal
phosphates and have a complementary nucleotide sequence, so
that purines can pair with pyrimidines, adenine with thymine,
and guanine with cytosine. And you might say that this scheme
was confirmed before it was devised, since this had been shown
to be true of the composition of DNA. It has been forgotten
completely that the specific base-pairing in nucleic acids was dis-
covered in an old-fashioned chemical laboratory.

190 amphisbaena

y:

Well, I happen to remember, but who cares? What did these
people whom you call old-fashioned do to make their findings
known?

o:
They published them.

y:

Published? Are you being jocular? Is this what you call aggres-
sive scholarship? Did they send out mimeographed copies of
their papers long before publication? Did they found base-pairing
clubs? Did they distribute neckties with suitable emblems?

o:

But who would want to spend his life dancing a minuet before
assembled science reporters? There still are people who do not
wish to join the noise boys; they have other concerns. When you
start on something new, you are all alone and it is so terribly
dark; and then, suddenly, you may come face to face v^th the
Winding whiteness of reality. There is nothing more exquisite,
nothing rarer in the world. Afterwards you have a choice: You
stay in the laboratory, hoping that it will happen again; but it
seldom does. Or you begin to travel through the country giving
minstrel shows.

y:

Anyway, even if the composition of DNA had been found entirely
different, we should have doubted the analysis, not our concept.

o:

I can see, you are a True Believer. It is with such deductions
that the road to the paradise of scientists is paved. But really, I
did not wish now to attack what one is pleased to call the "cen-
tral dogma", for I know that the mythopoeic urge of humanity
does not stop at the door of the laboratory. But let me continue,

AMPHISBAENA 191

and I shall assume that the existence of two complementary
strands, each having two ends, a head and a tail as it were, has
been proved.

y:
Well, if it hasn't, it will be.

o:
This DNA, as I understand it, has two functions: it has to make
itself — that is easy: unscrew, assemble, polymerize, recombine.
This has now become a so-called project for the so-called Science
Fairs of our so-called high-school boys. But it also has to make
the rest of the cell; and this is not so easy. For in this DNA
must be contained the quintessence of all this tremendous life of
our earth, the flagellum and the spirillum, but also the brain that
invented the St. Matthew Passion. "Dinanzi a me non fuor cose
create ..."

y:
Don't talk Latin at me.

o:

Never mind. What I wanted to point out is that in our days
DNA plays the role of a self-replicating philosopher's stone. First
we have an extremely vicious circle: DNA makes DNA makes
DNA, etc.; a dreary, tragic desert, an Yves Tanguy landscape.
But simultaneously — and by what delegation of functions even
you could not tell me — the Dr. Jekyll side of DNA gets into the
act and makes A and B and C. I am always being told that all
biological information resides, in the last resort, in DNA. But
organisms contain many types of. specific molecules, quite apart
from the proteins and the nucleic acids: the mysterious conjugated
proteins, some of the lipids, cell wall substances, blood group
substances, specific polysaccharides.

y:

These polysaccharides contain no information.

192 AMPHISBAENA

o:

How do you know? Did you talk with them lately? But let me
return to our argument. When DNA gives play to the friendlier
part of its nature, the first thing it does, I am told, is to preside
over the manufacture of messenger RNA. This RNA is said to
show the composition regularities of the entire DNA. If the latter
really consists of two complementary strands, this must mean
that two complementary RNA strands, or a mixture of shorter
pieces amounting in the aggregate to two such strands, have been
made. To put it more concretely, a triple uracil in one messenger
polynucleotide would have to correspond to a triple adenine in
the complementary RNA structure. Since a ribosome — "they also
serve who only stand and wait" — can make a protein only when
it is "programmed" by a messenger RNA which on the whole
cannot be a double strand, as it must be able to engage in specific
hydrogen-bonding, the conclusion would have to be that any
given section of a DNA dyad should give rise to two entirely
different proteins. Would you consequently subscribe to the
revised slogan "one gene, two enzymes"?

y:
Well, I don't know. There are hundred ways out.

o:
But should a scientist behave like a tracked cockroach? I know,
almost anything you can write on a piece of paper will eventually
be realized in a so-called system and then it will form a "fact".
I cannot help feeling, however, that not all facts are equally
worth knowing. I thought it was the task of the natural sciences
to discover the facts of nature, not to create them.

y:
You are just an obscurantist.

o:
I have often been accused of spreading darkness. And I cannot

AMPHISBAENA 193

deny that the dazzUng light being thrown on a few spots to the
exclusion of the rest has distorted all the proportions of our
science. How many frantic about-faces someone of my age has
had to live through! And the yelping is not at an end; in fact, it
is getting worse. Miss Molecule of 1962 remains to be crowned;
and I fear — and you hope — there will be many more. The next
candidate will no doubt be messenger RNA. And some time
later it will be the operon buffo: "Figaro qui, Figaro qua!" But
to continue with what I was saying before. You may, of course,
ascribe to my advancing years the difficulties I experience in
understanding what is claimed to go on. Even if we accept the
existence of a short-lived species of RNA that is made under the
direct control of DNA and by mirroring the composition of the
latter transfers the information from DNA to the proteins and if
we add the relatively small amount of so-called soluble RNA, we
are left with the bulk, almost 85 per cent, of the cellular RNA
for which neither function nor mode of formation is apparent.
You will, of course, with your usual originality tell me that Rome
was not built in one day, to which I shall reply that less credit
should be given to those who fail to solve great problems than to
those that succeed in solving small ones. There is nothing easier
than to fall off the Mount Everest.

y:

But you cannot deny the existence of a rapidly turning-over RNA
that does mirror the composition of DNA, in base-pairing, etc.?

o:

Most of the analyses that I have seen are far from convincing.
A lot of good will or, better, lack of experience is necessary for
such enthusiastically sweeping claims. But never mind, many
people have made a good living in science by selling the emperor's
new clothes. De nihilo nihil does not hold for molecular biology.
Only where there is nothing, all is possible. If you consider a
cell, and the amount of packing or compression it must require,
the traffic problems become so enormous as to demand the for-

194 „ AMPHISBAENA

mulation of a new dimension for which we lack even the sHghtest
intimation; instead of which we are being fed terminological cant.
DNA slowly reproducing itself and at the same time rushing to
make hundreds of different messengers; all these messengers mill-
ing about madly in search of ribosomal easy chairs on which to
procreate and die; proteins peeling off and proceeding to their
respective posts; unborn lipids and polysaccharides crying for
non-existing templates: a molecular Walpurgis Night, a Feder-
ation Meeting of the universe, only even less comfortable than
in Atlantic City. It would all be very funny if it did not corrupt
our youth. Who, when he knew it when it was so small, would
have thought of DNA as the demiurgos of a Manichaean world?

y:

You are using too many long words to say nothing. The fact that
you knew DNA when it had a molecular weight of 800,000,
whereas now it is up to 160 milUon, is of no significance, except
to show that you did not know how to invest wisely.

o:

Of course, this is not at all what I meant. But even if we reconcile
ourselves to this bizarre situation and accept the notion that
everything that happens in a cell is under the ultimate control of
DNA, we encounter other difficulties. Many viruses are essential-
ly ribonucleoproteins; and in several cases their RNA itself has
been shown to be infective. You mentioned it yourself before.
This RNA is presumably able to repUcate itself; but where is the
specific DNA that presides over this replication?

y:

I can give you two different answers. I could assume that there is
something special about viral RNA that enables it to act as its
own template, independently of any control exercised by DNA.

o:

But this is a self- strangling argument. Shouldn't in this case viral

AMPHISBAENA 195

RNA show complete base-pairing? Besides, from the chemical
point of view, there does not seem to be anything special about
the RNA of viruses. Incidentally, since in this reasoning all roles
are reversed, I expect to hear almost any day about a "messenger
DNA".

y:
I have an alternative answer. Let us assume that all RNA, viral
as well as ribosomal, transfer or messenger, is made under the
control of DNA, but only in the case of messenger are both DNA
strands involved; the other RNA species reflect the composition
of a single strand of DNA. By now you must credit me with
enough intelligence to propose hundred plausible schemes to
explain this.

o:

I would not go so far; but for n problems you will always have
« + 1 mutually exclusive, but equally irrefutable, explanations.

y:
Never mind. So far as virus RNA is concerned, a second as-
sumption must, of course, be made, namely, that for a cell to be
able to support a certain virus it must contain, as part of its
genome, a stretch of DNA that is normally non-operative, but
becomes active under the influence of the invading viral RNA
molecule; and it is this DNA portion that is reflected by the
infective RNA.

o:
I feel like Peer Gynt in the cave of the trolls, refusing to undergo
a simple eye operation that would permit him to see as straight
what looks crooked to him. You, too, will undoubtedly grow into
one of our synthetic geniuses — these fake celebrities, glued
together with the spittle of Madison Avenue — of whom there are
so many that they clutter up the place. And yet, the brain power
of all these cytopractors combined would not fill the inkwell of

196 AMPHISBAENA

Pascal. To them, everything is so simple. The hypnotic effect of
repeating nonsensical statements over and over has produced a
general trance that is mistaken for a view of nature. Altogether,
I was taught that it is the task of the natural sciences to under-
stand, not to outwit, nature. I am often told that this or that is
an "educated guess" — a truly nasty expression. Much would be
gained if the guessers were educated instead. Some of the dis-
cussions on microbial heredity and chemical genetics that I have
heard sounded like a bunch of midwives deliberating on the
immaculate conception. And what can be done to stem the ever-
growing avalanche of rubbish being published? I can think of
only one way: to publish all papers anonymously, without
authors' names.

I
y:
By the way, grapevine has told me that the code has been broken.

o:

I hope, someone keeps the pieces; one may need them again.

y:

How can you be facetious before such an achievement? Don't
you see that we have entered a new era?

o:

Thus, slowly, slowly, step by step, scream by scream, drum roll
by drum roll, gold medal by gold medal, do you expect to recon-
struct the fingerprints of God! But what good will it do you?
You could not read them, you could not classify them. All the
gimmicks in the world — to use the terminology with which you
are familiar — will not help you.

y:
You talk as if we were still in the early middle ages.

o:
Maybe the natural sciences will always be in the early middle

AMPHISBAENA 197

ages. There goes a deep crack through our porcelain world; and
even theoretical physics, perhaps the most highly developed
science, feels, I am told, much malaise and intellectual discom-
fort. Could it be that molecular biology is the last refuge of the
scientific optimist?

y:
Well, we have much to be optimistic about. This has been a
marvelous period for the biological sciences, a true renaissance.
We have learned more about life and heredity in the last five
years than in the preceding fifty; and for this reason we can
afford to disregard most of the older literature. Even you, who
has difficulty in reconciUng yourself to all our discoveries, will
not be able to deny the tremendous upsurge.

o:
There is so much to get reconciled to. In fact, that such very bad
times as ours have given rise to so much good science, does this
not speak against science?

y:

Not at all. You seem to have the romantically foolish idea that
only a good man can be a good scientist.

o:

It is always dangerous to use the argument ad hominem, and you
should not judge from yourself. But is it not a desperate situation
when an old proverb must be reversed to read: Wherever the
fish stinks there is its head? It is getting late, though, and I had
not quite finished with what I was saying before. Even if the
correct code is found and the flow^of so-called information takes
place as postulated by the admirers of biological automation, very
little of what occurs in a living cell is really clarified. What
determines the specific character of a cell, which is perpetuated
in a hereditary fashion, is constituted of a very large number of
different compounds, many of them situated specifically within
the cell; and these substances, once we break the cell and isolate

198 AMPHISBAENA

and separate its constituents, will be recognized as very many
different species of molecules, such as proteins, lipids, polysac-
charides, nucleic acids, etc. Most of those — and not only the first
and last — have a highly specific composition and structure; but
how they exist, interact and are held in definite places in the
functioning cell is completely obscure. I am certainly no vitaHst
in the sense in which this is usually meant; but I cannot stomach
people who claim that they have understood and explained
"Hamlet" by telHng me how often the word "and" appears in the
first act. And I do object to the tremendous noise that is being
made about trivial and often meaningless observations. In con-
trast to previous times, I believe, many of our reputedly great
discoveries are entirely unearned. Also, the morass of alleged
facts in which we are suffocating has brought it about that those
who may speculate cannot do so any longer. Look at the enor-
mous variety in the shapes of organisms, organs, even cellular
components — ^where is the biochemistry of specific shape? Where
is the biochemistry of cell differentiation? Is there a separate
nucleotide code for your fingerprints which are different from
mine? Is it a pairing error in position 79 which has produced the
visions of Blake? It is above all against this shabby mechanization
of our scientific imagination, which kills all ability to notice the
unforeseen, that I protest, against this mat finish over a chaos
of unrecognized ignorance, this butcher-like brutality with things
that cry for gentle caution. Our young people are being brought
up in the belief that "they never had it so good". They — and
especially the best ones — are being condemned to a future of
disillusion and discouragement.

y:

That's what you say. I am not at all discouraged; quite the contrary.
I only have to think of the unprecedented possibilities that are
opening before us. When we know the Universal Code, we shall
soon learn how to interfere with certain nucleotide sequences in
DNA, how to change them specifically and thereby to produce
desirable genetic changes. Artificial insemination with the stored

AMPHISBAENA 199

Sperm cf dead geniuses has already been proposed by eminent
authorities. Two Uttle Einsteins in every middle class household,
what a vista!

o:

But can the centipede survive under a duodecimal system? The
more you talk about breeding geniuses the less Hkely you are to
get them.

y:

Again one of your unscientifically mystical remarks. I only
mentioned this proposal as a first modest feeler into the great
future. Later on, we shall be able to look up the nucleotide
sequence of every DNA; and each purine and pyrimidine will
have a number; and we shall know of each what happens when
we change it. And boy, will we change it!

o:

And then you will get the real "human engineering". Once you
can alter the chromosomes at will, you will be able to tailor the
Average Consumer, the predictable user of a given soap, the
reUable imbiber of a certain poison gas. You will have given
humanity a present compared with which the Hiroshima bomb
was a friendly caster egg. You will indeed have touched the
ecology of death. I shudder to think in whose image this new
man will be made.

y:

Well, you may not be around to see it. Anyway, it was nice to
have talked with you. I must still run up to the lab to turn off
the Spinco.

o:

is, the evening has come. I shall go home.

/""

They leave, in opposite directions.

Index

Acid degradation of DNA, lability
of purine glycosides to acid, dif-
ferential distribution analysis,
123, 124

Adenine and guanine, content in
animal DNA's and in animal
RNA's, 6

Analogues, selection of fraudulent
— in formation of DNA, 152, 153

— , — of natural — in formation
of DNA, 150-152

Apurinic acid from calf thymus
DNA, composition and properties
(table), 33, 34

— , definition, 33

— in sequence analysis, 33, 91, 123,
133-134, 166

Arbacia lixiila, pyrimidine 3 ',5 '-di-
phosphates from DNA, differen-
tial distribution analysis, 92, 122,
138

Arrhythmic character of nucleotide
sequence in DNA, 91

AT type DNA, 28, 51, 66, 86, 128,
141, 165, 170, 172

Autarkic entities (cells and cell
communities), and viruses and
phages, requirements in — , 71,
101, 103

Avian tubercle bacilli, composition
of DNA (tables), 13, 14, 87

— , ratios in DNA, 30, 50, 66, 87

8-Azaguanine, incorporation into
DNA, 152, 155

Azotobacter vinelandii, invariability

of total RNA, 105
— , nucleotide ratios in ribonucleo-

proteins, 105, 119

— RNA, 94, 105, 119

"Backbone" of nucleic acids, 90

Bacterial transformation by a specif-
ic DNA, 71, 121

Base-pairing regularities in DNA,
13 and passim

Biochemical evolution, 78

— and formation of macromole-
cules, 78

Biochemical space, 115-117
Biochemistry, genetics, and nucleic

acids, 163, 164
Biological information, 73, 95, 113-

115, 163, 184
— , chemically conveyed, 102-105
— , DNA as repository of, 104, 165
— , flow of—, 103, 113-115
— , linguistic analogies, 104, 113,

130, 165

— and nucleic acids, 73, 74, 100-
107

— and regularities in nucleic acids,
95, 121

— , sequential specificity in pro-
teins, nucleic acids, polysaccha-
rides and preservation and trans-
fer of information, 103

— , triangular relationship between
DNA, RNA and proteins in pre-
servation and transfer of — , 103

202

INDEX

Biological polymer, what determines

the size of a — ?, 127
Biological role of nucleic acids, 26,

70-72
, participation of proteins in,

26, 73
5-Bromouracil, incorporation into

DNA, biological consequences,

143-146
— , — of E. coli thymine auxo-

troph, 152-154
Bunched and solitary nucleotides

(pyrimidine), 123, 148, 166
, schematic representation of

a fragment of a double strand of

DNA, 90, 132

Calf liver, composition of RNA's
(table), 17

Calf thymus DNA, 12, 138

— , apurinic acid from — , 34

— , enzymic degradation, 19, 31,
33

— , fractional dissociation from his-
tone and globin complexes, 45—
50

— , method of isolation, 129

— , 5-methylcytosine distribution in,
56, 151

Cell constituents, spatial conditions
under which the cell manufac-
tures its constituents, 116

Cell and machine, 174, 175

Cellular specificity, chemical basis,
102-104, 112, 113

— , directive influence of proteins
and nucleic acids in maintenance
of_, 104

— and heredity, 1 1 1

Chemical concept of a molecule and
molecular biology, 182

Chicken erythrocyte DNA, 30-32

5-Chlorouracil, incorporation into
DNA, 152, 154

Citrate ions, use in isolation of nu-
cleic acids for inhibition of deoxy-
ribonucleases, 6

Clusters of nucleotides in DNA, 19

, analysis of sequence, longer

all-thymine than all-cytosine units
in many DNA's, 149

Code-script of biological high poly-
mers, 117-124

Coding of information and order of
alignment of the nucleotide mon-
omers in a nucleic acid, 124

— for phenylalanine, 169

— units for amino acids, critical
remarks, 168-172

Comparative biochemistry, 112

Compositional regularities in nucleic
acids 13, 28, 30, 51-52, 54, 66,
72, 86-88, 93-96, 106, 120, 127,
128, 131, 155, 164, 170

Conjugated proteins, 78-81

Conjugation in the formation of
macromolecules, role in biochem-
ical evolution, 78

Cybernetics, 113

Cytidine, isolation and estimation, 9

Cytochrome b2, polydeoxyribonu-
cleotide component of — , 128

Cytoplasm of rat liver and kidney,
nucleotide ratios in ribonucleo-
proteins, 94, 119

Cytosine, decomposition to uracil
by heating with strong acid, 10

Cytostatic and cytotoxic chemicals
which are incorporated into nu-
cleic acids, 71

Deoxy cytidine and thymidine 3',5'-
diphosphates released from DNA
of various organisms (table), 92,
136, 138, 141, 142

Deoxymethylcytidine 3 ',5 '-diphos-
phate, 140

Deoxypentose nucleic acid(s), see
also deoxyribose nucleic acid(s),
deoxyribonucleic acid(s), DNA

— , sugar in—, 18, 53, 65, 84

Deoxypentose nucleoprotein(s) of
calf thymus and wheat germ,
course of dissociation at rising
NaCl concentrations, 46, 65

INDEX

203

Deoxypentose nucleoprotein(s) (con-
tinuation)

— , classification according to the
protein moiety, 41

— , dissociation and cleavage, be-
havior towards electrolytes and
fractionation of DNA, 45, 57, 58,
65

— , fibrous structure, 40

— , isolation by extraction with so-
lutions of low ionic strength, 42,

65, 89

— , — with strong salt solutions,
42, 65, 89

— and their prosthetic groups, 39-
60

Deoxyribonuclease from pancreas,
action on calf thymus DNA, for-
mation of non-dialyzable core,
18, 19, 28, 31, 33

— , stepwise degradation of DNA,
19,91, 133

— of yeast, inhibitor specificity, 19,
20

Deoxyribonucleohistone, isolation,
65

Deoxyribonucleoprotamine, isola-
tion, 65

Deoxyribose nucleic acid(s) (deoxy-
ribonucleic acids, DNA, deoxy-
pentose nucleic acids), see also
nucleic acid(s)

— , absence of organ specificity in
composition of DNA's of the
same species, 13, 14

— , — of periodicity in structure,
33, 51

— , arrhythmic character of nucleo-
tide sequence, 91

— , AT type and GC type, 28, 51,

66, 86, 128, 141, 165, 170, 172
— , balance between amino groups

and enolic hydroxyls in various
DNA preparations, 32, 52
— , balance of 6-amino groups and
of 6-keto groups, 32, 52

— of calf thymus, degradation by
deoxyribonuclease (table), 19

Deoxyribose nucleic acid(s) (con-
tinuation)

— , common features, 27

— , composition, 11-13

— , — of DNA fractions obtained
by fractional extractions of wheat
germ nucleoprotein gels (table),
68

— . — in human tissues (table), 12

— , — in normal human tissue and
in cancer tissue, 28

— , — and regularities, 86-88

— , — in several organs of the ox
(table), 12

— , — of several total — (table), 87

— , — in yeast and avian tubercle
bacilli (table), 13

— content per diploid and haploid
nucleus, 27, 71

— and copying mechanism, 73

— , differential analysis of pyrimi-
dine distribution in three analyti-
cally indistinguishable — (table),
122

— , dissymmetry ratios, 86

— , diversity, 32, 57

— , equality of sum of purine nu-
cleotides and sum of pyrimidine
nucleotides, 52, 66

— , evidence for the existence of a
very large number of — in na-
ture, 50, 51

— , examples of pleromeric rela-
tionships (table), 154, 156

— , fractionation, 57-59, 88-89

— , — through their artificial com-
plexes with histone, globin or po-
lylysine, 57

— , — on histone-kieselguhr col-
umn, 57

— , — at the nucleoprotein stage,
66-68

— , genes and specificity of — in
chromosomes, 31

— , identity of composition in dif-
ferent tissues of the same species,
27,28

204

INDEX

Deoxyribose nucleic acid(s) (con-
tinuation)

— , induction of transformation of
bacterial types, 26

— , inhibition of degradation by
deoxyribonuclease with citrate
ions, 6

— , involvement in the genetic ap-
paratus of the cell, 71

— , isolation of nucleoproteins as
first step in isolation of — , 50

— , location in cells, 25, 26

— , maintenance of sequence integ-
rity, 129

— , molar ratios in — of different
origin (table), 30

— , — of adenine to thymine and
of guanine to cytosine (+ methyl-
cytosine), 13, 51, 52

— , molecular structure, discussion
of Watson-Crick model, 53-56

— , molecular weight, 5, 88, 128,
165

— , position of the phosphate bridg-
es in calf thymus — , 65

— , presence outside the cell nucleus,
25

— , purines and pyrimidines, 65, 66

— , regularities in composition, see
base-pairing regularities

— , schematic representation of seg-
ment, 90, 132

— , selection of fraudulent and of
natural analogues in formation,
150-153

— , sequence analysis, see sequence
analysis of DNA

— , species differences, 13, 14, 86-
88

— , species specificity and organ
specificity, 13, 66

— , specimens, uniformity of ab-
sorption spectrum in the ultra-
violet, 32

— , structure and sequence of nu-
cleotides, feasibility of sequence
analysis, 90

Deoxyribose nucleic acid(s) (con-
tinuation)

— of thymus, spleen, liver, tuber-
cle bacilli, human sperm; isola-
tion, 7

Deoxyribothymidine, separation
from thymidine, 9

Depolymerizing enzymes of nucleic
acids, 18-20

Differential distribution analysis of
DNA, characteristics of nucleo-
tide sequence, 122-124, 134-148

, mechanism, 134-137

Differential pyrimidine distribution
analysis, principal conclusions,
166, 167

Dinucleoside triphosphates compris-
ing 5-methylcytidine and either
cytidine or thymidine, 140

Dissymmetry ratio of nucleic acids,
definition, 66, 86

Diversity and identity in high-
molecular cell constituents, 2-5

— of nucleic acids, chemical diver-
sity in DNA of various species
and in total cell DNA, 50, 57, 66,
88, 105-106

, — in RNA of various spe-
cies and in RNA's of the same
cell, 105-106

isolated from a given cell,

66

Division of labor in the life of the
cell, 185

DNA, see also deoxyribose nucleic
acid(s) (deoxyribonucleic acid),
deoxypentose nucleic acid(s)

— , is there evidence for the exist-
ence of DNA chains for biologi-
cal information in higher forms
of life?, 163

DNA and RNA, critical remarks
about present conceptions of func-
tions of — , 185-195

— , remark about the terms, 84

Dynamic state of the body constitu-
ents, 183

INDEX

205

Echinocardiiim cordatum, ratio of
adenine + thymine to cytosine
+ guanine in DNA, 50

End groups in undegraded DNA,
absence of — ?, 53, 129

Escherichia coli, DNA produced by
thymine auxotrophs in the pres-
ence of bromouracil, 143-147

— , equimolarity of the bases in
DNA, 51, 86

— , invariability of total RNA, 105

— , K-12 strain DNA, 30, 31, 86,
87

— , longer all-thymine than all-
ey tosine stretches in DNA, 149

— , molar ratios of bases present in
DNA, 51, 86, 87

— , nucleotide ratios in nucleopro-
tein, 94, 119

— , — in RNA, 69, 119

— , pleromeric relationships in
DNA, 154

— protoplasts, method of isolation
of DNA, 130

— , transformation by DNA, 26

5-Fluorouracil, incorporation into
DNA, 152, 156

Fractionation of deoxypentose nu-
cleic acids at the nucleoprotein
stage, 57, 58, 66-69, 121, 123

GC type DNA, 28, 31, 51, 66, 86,
128, 141, 142, 164, 165, 170

Genes and specificity of DNA in
chromosomes, 32

Genetic apparatus of the cell and
DNA, 71

Genetic determinants and nucleic
acid chains, many determinants in
one DNA chain or one DNA
chain for each determinant, 121

Genetics, biochemistry, and nucleic
acids, 163, 164

Hemophilus influenzae (continua-
tion)
— , type b and c, DNA, 26
— , type c, DNA, 30, 31
Heredity, definition. 111
— , first steps towards a chemistry

of _, 109-124
Heterogeneity of nucleic acids as

regards composition and sequence

characteristics, 128
Hierarchy, existence of — in the

cell, 1, 63, 100, 101, 117, 185
High-molecular cell constituents,

identity and diversity, 2-5
Homogeneity of nucleic acids in the

cell, 127-128
Homotopes and pleromers in

DNA's, 157

— in polymer constitution, defini-
tion, 131

Homotopy and pleromerism and
the neighbor problem in forma-
tion of nucleic acids, 153-158

— of thymine and 5 -methylcy tosine
in rye germ DNA, 155

Human tissues, composition of DNA
from various — (table), 15

— , differential distribution analy-
sis of DNA of spleen (table),'
138

— , isolation of sperm DNA, 7

— , molar ratios of nucleotides in
DNA of spleen, liver and other
tissues, 12, 14, 30, 87, 92

— , pyrimidine 3 ',5 '-diphosphates
from DNA, differential analysis,
pyrimidine distribution, 92, 122,
138

Hydrolysis of the nucleic acids, 10-
11

5 -Hydroxy methylcy tosine, 86

— in bacteriophage nucleic acids,
52

— in DNA of E. coli phages, 66

Harmony of cellular life, 4
Hemophilus influenzae, transforma-
tion by DNA, 26

Identification of the macromolecu-
lar compounds of cellular origin,
3, 4, 104

206

INDEX

Identity and diversity in high-mo-
lecular cell constituents, 2-5

, is there a critical range

of molecular weights as for abili-
ty of cells to synthesize complete-
ly identical substances?, 5

Information, see biological informa-
tion

Information theory, 113

Invariability of nucleic acids, 104-
105

— within the species, 120
5-Iodouracil, incorporation into

DNA, 152, 154
Isolation of nucleic acids, exclusion
of degradation, 6

Lipoproteins, nature of bonds be-
tween lipid and protein, 80
— , specific patterns, 80, 113
Liver, nucleic acids, 11, 12, 14, 15,
17, 93

Macromolecular structure, the prob-
lem of the — of the native state
of DNA, 129, 130

Magnesium ions, effect on fractional
dissociation of artificial complex-
es of histone with calf thymus
DNA, 48

— requirement by deoxyribonucle-
ases, 6

Mechanomorphic concept in bio-
chemical conceptions of cell life,
103

— of living nature, 111, 175
6-N-Methyladenine, 86, 153
Methylcytosine-cytosine units in rye

germ DNA, 157
5-Methylcytosine, amount in DNA
of wheat and rye germ (table),
139

— distribution in fractions of calf
thymus DNA (table), 56

— in DNA of E. coli phages, 66
— , frequency of — and cytosine as

solitary and bunched nucleotides
in DNA, 152

5-Methylcytosine, frequency of —
(continuation)

— , — and thymine as solitary nu-
cleotides in DNA, 152

— , high proportion linked to gua-
nine in a dinucleotide as compared
with cytosine in various DNA's,
56, 157

— , occurrence, 86

— , percentage contribution to ami-
nopyrimidine content in various
fractions of DNA, 151

— in structure analysis of nucleic
acids, 139-141

— in wheat germ DNA, 52, 55, 139
Minimum of signaling potential in

biological information, 117

— of viable organization, 117
Models and model experiments, re-
flections on — , 53, 77, 83, 95,
174, 186, 187

Molecular weights of DNA, 5, 88,
128, 165

Morphogenesis, role of RNA, 22, 71

Mutagenic action spectrum of ultra-
violet radiation and absorption
spectrum of nucleic acids, 71

Mutagenic effects of chemicals
which react with nucleic acids, 71

Mycobacterium phlei, nucleotide
ratios in RNA, 69, 94, 119

Mycobacterium tuberculosis (bo-
vine), nucleotide ratios in RNA,
119

Native DNA, preservation of native
state, 129

Native state of nucleoproteins, cri-
teria, 42, 43

Neighbor principle, high tendency
of methylcytosine to be next to
guanine in DNA's, 157

— in nucleotide sequence in nucleic
acids, 153-158

— in soluble RNA of liver cyto-
plasm, 157

— in structure of calf thymus DNA,
72

INDEX

207

Nomenclature, remarks on — in
sequence analysis of nucleic acids,
131-132

Nuclei of rat liver, nucleotide ratios
in ribonucleoprotein, 119

Nucleic acid(s), see also deoxyribose
nucleic acid(s) and pentose nucleic
acid(s)

— , absence of repeating polynucle-
otide units, 51, 72

— , biological role of — and of
nucleoproteins, 26, 73

— , brief listing of relevant informa-
tion on the biological roles of
the — , 71-72

— , as carriers of biological infor-
mation, 73, 100-107

— , depolymerizing enzymes, 18-20

— , diversity, nucleotide sequence
and periodicity in — chain, 31,
32, 33, 105-106, 119

— , exclusion of degradation by
enzymes, acid or alkali, in isola-
tion, 6

— , invariability, 104-105, 120

— , isolation, 6

— , meaning of regularities in com-
position, 93-96

— , methods for analysis of nucleo-
tide composition, 6-11

— , methods of hydrolysis, 10-11

— , principal constituents, 85, 86

— , reaction with chemicals and
mutagenic effects, 71

— , reflections on possible role of
DNA in transmission of inherited
properties and of pentose — in
differentiation, 21, 22

— , remarks on functions, 70-74

— , remarks on nomenclature in
sequence analysis, 131-132

— , role of proteins in function
of _, 73, 74

— , separation at the nucleoprotein
stage, 68

— , some immediate problems in the
chemistry of — , 167, 168

Nucleic acid(s) (continuation)

— , structure and function of — as

cell constituents, 25-36, 70-74
— , — and sequence of nucleotides,

90
— , sugar components, 18, 53, 65,

84
— , ultraviolet absorption spectrum

and mutagenic action spectrum

of ultraviolet radiation, 71
— , urgency of sequence analysis, 73
Nucleohistone(s), 41, 64, 79
— , bonds between nucleic acid and

histone, 41

— of calf thymus, isolation, 42

— , fractional dissociation of calf
thymus — and of wheat germ
nucleoprotein, 46, 89

— , fractionation of nucleic acids
in — by exposure to increasing
NaCl concentrations, 44, 45, 65

— , nucleic acid content in — , 43

— , properties of calf thymus —
solutions at different NaCl con-
centrations (table), 44

— , solubility, 43

Nucleoprotamines, 41

— , bonds between nucleic acid and
protamine, 79, 80

— , dissociation in strong salt solu-
tions, 43, 45

— , phosphorus to arginine ratio, 43

— from sea-urchin sperm, isola-
tion, 42

Nucleoproteins, see also ribonucleo-
protein(s) and deoxypentose nu-
cleoprotein(s)

— , absorption spectrum in the ul-
traviolet, 43

— , artificially prepared complexes
of nucleic acids with positively
charged histone, globin and syn-
thetic polylysins and polyvinyl-
amine, 45-49, 57, 58

— from avian tubercle bacilli, iso-
lation, 42

— , bonds between nucleic acid and
protein, 64, 79, 80

208

INDEX

Nucleoproteins (continuation)
— , deoxypentose — and their pros-
thetic groups, 39-59
— , distinction between pentose —
and deoxypentose — , 64, 79, 80

— as fibrous proteins, 40

— , fractional dissociation of artifi-
cial complexes, 46, 47

— , function of the nucleic acid and
of the protein part in — , 107

— , native state, 42, 43

— , nucleic acid content, 43

— , remarks on — , 96

— and RNA's, 92-96
Nucleotide arrangement in different

DNA's of animal origin (table),
137, 138

— in the DNA of E. coli grown in
the absence and presence of 5-
bromouracil, 142-146

— in a microbial DNA of the GC
type (table), 140-142

— in the total deoxyribonucleic
acid of rye germ and in a com-
plete series of fractions, 137-140

Nucleotide ratios in ribonucleopro-

teins, 69, 94, 119, 120
Nucleotide sequence in DNA (see

also sequence analysis), 126-158

Order of alignment of the nucleo-
tide monomers in a nucleic acid
chain and coding of information,
123

Organ specificity of composition of
nucleic acids, 12, 66

Ox tissues, composition of DNA of
thymus, spleen, liver, 12, 14, 15,
30,87

— , molar ratios in DNA, 14, 15,
30, 87, 92

— , nucleotide ratios in ribonucleo-
proteins of liver, 94, 119

— , — in RNA of liver, 17, 69

— , pyrimidine 3',5'-diphosphates
from DNA, differential analysis,
pyrimidine distribution, 92, 122,
138

Paper chromatography of purines
and pyrimidines, 5, 7-9, 149, 180

Paracentrotus lividus, differential
distribution analysis of sperm
DNA, 138

— DNA, 87

— , eggs and embryos, nucleotide
ratios in ribonucleoproteins, 69,
94, 119

Pentose nucleic acid(s) (PNA, ribo-
nucleic acids, RNA)

— , action of ribonuclease, 34, 35

— , composition, 15-17, 93

— , — in animal tissues (table), 17

— , — in several preparations from
yeast and liver (table), 93

— , — in yeast (table), 16

— , differences between — of the
same organ of different species
and of different organs of the
same species, 17

— , — in composition between nu-
clear and cytoplasmic RNA in
liver, 25

— , — between RNA's from differ-
ent locations within the cell,
69-70

— and differentiation, 22

— , equality in the proportion of
6-amino and 6-keto compounds
and system for transfer of infor-
mation, 52, 66, 74, 93, 106

— , frequency distribution of vari-
ous nucleotide ratios in total —
of 104 samples, 94

— , guanine content of RNA of
pancreas, 17

— , isolation from animal tissues, 7

— , location in cells, differences be-
tween nuclear and cytoplasmic — ,
25

— , methods of analysis, 15, 16

— , — of hydrolysis, 15, 16

— , molecular weight, 4

— and nucleoproteins, 92

— , nucleotide ratios in various or-
ganisms and organs (table), 69
— , preparation, 7

NDEX

209

Pentose nucleic acids (continua-
tion)
— , regularities in composition, 52,

66, 74, 93, 106
— , role in protein synthesis and in

morphogenesis, 71
— , soluble, for transfer of amino

acids in liver cytoplasm, 157
— , species and organ specificity, 17
— , sugar component in — , 18,

84
Pentose nucleoproteins, fibrous, 40
— , links between nucleic acid and

protein, 64, 80
— , occurrence, 64, 65
Periodicity, absence in nucleic acids,

31, 33, 91
Phage Tor+ DNA, pleromeric rela-
tionships, 154
Phosphoric acid bridges in DNA,

53, 65, 89
Pig tissues, action of ribonuclease

on PNA, 34
— , fractional dissociation of liver

DNA from histone and globin

complexes, 46
— , RNA of liver, pancreas, 17
Pleromer, definition of the term —

in polymer constitution, 131
Pleromeric relationships in DNA's

(table), 154
— in RNA's, 155, 156
Pleromerism and homotopy and the

neighbor problem in nucleic

acids, 153-158
Pleromers and homotopes in DNA's,

157
Pneumococcus, transformation by

DNA, 1, 26, 71, 121, 181
Precursors, movement of — into

various polymers in cellular ele-
ments, 115, 116
Prosthetic groups in conjugated

proteins, 81
Protein synthesis, role of nucleic

acids, 71, 95, 96
Purines, location on paper chroma-
tograms in ultraviolet light, 9

Purines (continuation)

— , molar proportions in DNA's
from different organs of one spe-
cies (table), 15

— , from different species

(table), 14

— , release from nucleic acids by
mild acid hydrolysis, 9, 10

— satellites in DNA's, 52

— , schematic representation of the
position on paper chromatogram,
8

— , separation and estimation, sol-
vent systems, 7-9

— , spectrophotometric determina-
tion, 9

Pyrimidine(s) distribution of the
total DNA of rye germ and wheat
germ and its fractions, 139-141

— , location on paper chromato-
grams in ultraviolet light, 9

— , molar proportions in DNA's
from different organs of one spe-
cies (table), 15

— , from different species

(table), 14

— nucleoside diphosphates, forma-
tion by acid treatment of DNA,
use for study of the nucleotide
sequence, .91, 92, 123, 134, 135

— nucleotides, all-thymine and all-
cytosine units in DNA, 149

, liberation of the pyrimidines,

10, 11

, solitary and bunched — in

DNA, 123, 148, 166

— , occurrence in DNA as oligonu-
cleotide tracts containing three
or more pyrimidines in a row, 148

T—, release from nucleotides by hy-
drolysis with concentrated formic
acid, 11

— , resistance to treatment with
strong acid (table), 10

— , schematic representation of the
position on paper chromatogram,
8

— , separation and estimation, 7-9

210

NDEX

Pyrimidine(s) (continuation)
— , spectrophotometric determina-
tion, 9

Rat tissues, nucleotide ratios in ri-
bonucleoproteins of liver and
kidney and cytoplasm of liver and
kidney, 69, 94, 119

— , pleromeric relationships in liver
DNA, 156

— , uptake in vivo of radioactive
leucine by liver cytoplasm, 116

Reflections on present trends in bio-
logical sciences, 100-102

Regularities in composition of nu-
cleic acids, meaning, 93-96

of RNA's and DNA's, 13,

28, 30, 51-52, 54, 66, 72, 86-88,
93-96, 106, 120, 127, 128, 131,
155, 164, 170

Replacement, meaning of the term,
— of natural purine or pyrimi-
dine in nucleic acids by analogues,
150

Ribonuclease, action on pentose
nucleic acid of yeast and pig liver,
composition of dialyzable nucleo-
tides and core (table), 34

— from pancreas, 18, 34, 35
Ribonucleic acid, see pentose nucleic

acids, RNA, PNA
Ribonucleoprotein(s), links between
nucleic acid and protein, 64, 80

— of the ribosomes, 165

Ribonucleotides, methods for sepa-
ration and quantitative determi-
nation in an aqueous ammonium
isobutyrate-isobutyric acid sys-
tem, 9

Ribosomes, 165

RNA (see also pentose nucleic
acids, ribonucleic acids, PNA)

— and DNA, remark about the
terms — , 84

Rye germ DNA, comparison of the
pyrimidine distribution in total
DNA with that in DNA fractions,
139-141

Rye germ DNA (continuation)
— , pleromeric relationships, 154

Salmon DNA, identity of composi-
tion as regards distribution of ni-
trogenous constituents with sheep
DNA, 137

— , molar ratios, 30

Salmon sperm DNA, composition,
29, 30, 87

Sea urchin DNA, dissymmetry ra-
tio, 66

— sperm DNA, 42, 66, 138
Sequence analysis of DNA, 31, 33,

90,91

— of nucleic acids, lines of attack,
91,92

, possibilities of cluster anal-
ysis, 148-150

— , remarks on the conceptual basis
of _, 127-128

Sequence characteristics common to
all DNA's, 147-148

— of DNA, occurrence of at least
70% of the pyrimidines in animal
and plant DNA as oligonucleo-
tide tracts containing three or
more pyrimidines in a row, 148

— of nucleic acids, 124
Sequence investigation, apurinic

acid, 133, 134
— , early attempts, 132

— of nucleic acids through differ-
ential distribution analysis, see
differential distribution analysis

— , the problem of statistical analy-
sis, 130-131

— , stepwise degradation of DNA
by deoxyribonuclease, 133

Sequence of nucleotides in DNA,
the problem of non-randomness,
147, 148

Sequence studies of DNA, differ-
ence in stability to acid of purine
and pyrimidine nucleosides in
DNA, 135, 166

INDEX

211

Sequential specificity of high poly-
mers as mark of the specificity of
the cell, 165, 166

Serratia marcescens DNA, 30

Sheep DNA, 87, 137

Sheep liver, composition of RNA's,
17

Solitary and bunched nucleotides,
in DNA, 166

, schematic representation of

a fragment of a double strand of
DNA, 132

Soluble RNA's of liver cytoplasm
for transfer of amino acids, 157

Solvent systems for separation of
purines and pyrimidines, 7-8

Species diversity and invariability
of composition within the same
species of highly polymerized cell
constituents, 120

Species specificity of composition of
DNA, 13, 15, 66, 120, 164

of RNA, 120

Specific arrangement of the consti-
tuents of cell-specific polymers,
role of nucleic acids, 118

Specificity, direct gene-controlled
and reflected — of highly poly-
merized cell constituents, 120

Spermatozoa, ratios of adenine to
guanine and of thymine to cyto-
sine in deoxypentose nucleic acids
of human—, 12, 13, 14, 15

Sub-units in DNA, absence of — ,
133

Sugar components of nucleic acids,
18, 53, 65, 84

Templates, 2, 74, 81, 95, 104, 111,

116, 127, 184, 194
— in time for manufacturing of cell

constituents, 81, 116
Terminal phosphate group in DNA,

53
Tetranucleotide hypothesis, 2, 3, 13,

28, 51, 85, 127

Thymic acid, 33

Thymonucleic acid, 3

Thymus, nucleic acids of — , 3, 12,
14, 15

Transformation, bacterial, by DNA,
26,71, 121, 181

Tubercle bacilli, distribution analy-
sis of DNA of avirulent variant
(BCG) of bovine — , 141, 142

— , isolation of DNA via the nu-
cleoprotein, 7

Ultraviolet absorption spectra of
DNA specimens, 32

— of nucleoproteins, 43

Ultraviolet radiation, mutagenic ac-
tion spectrum and absorption
spectrum of nucleic acids, 71

Unity of biochemistry, the concept
of — , 179

Unity of nature, the concept of — ,
179

Uracil, instead of thymine in RNA's,
11

Uridine, isolation and estimation, 9

Uridylic acid, refractoriness to hy-
drolysis by acids, 16

Vitalism, 110

Wheat germ DNA, 29, 30, 31, 33,
46, 68, 87, 139, 154

, composition of intact prepa-
rations and of enzymically pro-
duced cores, 28, 29

— , nucleotide ratios in pentose nu-
cleoproteins, 94, 119

— RNA, 69

Yeast DNA, composition (table),
14, 30

— RNA, action of ribonuclease, 34
, composition, nucleotide ra-
tios, 16, 69, 93, 94, 119

X-ray investigations of DNA, 53

<^^y^^Al

PRINTED IN THE NETHERLANDS
BY VONK & CO'S DRUKKERIJ, ZEIST

)

imm\

iL..