chapter 1]

GENES, PROTEINS,
AND
EVOLUTION

- we must remember that heredity, development,

and evolution are essentially epigenetic and not preformistic.
We do not inherit from our ancestors, close or remote,
separate characters, functional or vestigial. What we do
inherit is, instead, genes which determine the pattern of
developmental processes. .

T. Dopzuansxy, Evolution, Genetics, and Man.

A, the genes of a species are modified and reshuf-
fled, occasional organisms will appear within a restricted population
having phenotypic characteristics that enable them to explore desir-
able ecological niches which were unattainable by their predecessors.
The individual changes are generally quite small. Many generations
must come and go, during which forays into formerly forbidden ter-
titory by this developing branch of the population become more fre-

GENES, PROTEINS, AND EVOLUTION 211
quent and are of longer duration as the result of further reorganiza-
tion of the gene pool by random mutation and natural selection. In
time the summation of these changes results in a new species, fully at
home in its new environment and sufficiently different in physiology
from its distant ancestors that cross-fertilization is no longer possible.

We have discussed, in several earlier chapters, the techniques used
by the geneticist for the analysis and description of limited portions
of such a chain of events. As long as crosses can be made between
different family lines, phenotypic changes can generally be related to
specific genes and the spread of these genes through a population can
be fairly accurately mapped. Thus, the basic assumptions of evolu-
tionary theory may be directly tested, and with some precision, when
the segment of time under consideration is small, and we are able to
describe the process in terms of changes in genotype. When we deal
with evolution on a larger scale, however, the tools of the geneticist
are no longer applicable. The evolutionist must now rely on the
study of relative morphology and ecology as deduced from the fossil
record, or on the comparative anatomy and physiology of living rep-
resentatives of surviving species.

The principal aim of this book has been to examine the basic prin-
ciples underlying another possible method for the study of evolution.
This method is based on the hypothesis that the individual proteins
which characterize a particular species are unique reflections of the
genes which control their synthesis. The examination of the chem-
istry of a series of homologous proteins is, of course, a purely pheno-
typic approach to the problem. Nevertheless, the evidence available
to us, even at this early date, suggests that the structure of proteins
may be a relatively direct expression of gene structure and that com-
parative protein chemistry may furnish a qualitative view of geno-
typic differences and similarities. If we accept the general hypothe-
sis, we are led to infer, for example, that the “insulin-determining”
genes of the pig and the sperm whale are identical, like the insulins
whose structures they determine. Indeed, should several genes be
concerned with the synthesis of insulin, the same would also be true
for these.

Another interesting potentiality of comparative protein chemistry
is that it might permit us to determine whether the same phenotypic
characteristic, shown by two completely unrelated organisms, is at-
tributable to analogous or to homologous genes. For example, both
bacteriophage T2 and chicken’s eggs contain proteins that have lyso-
zyme activity. The genetic material of both coliphage and chickens
must be said to contain information that can direct the formation of

212 THE MOLECULAR BASIS OF EVOLUTION

proteins with this function. Is it possible that these two organisms
contain nearly identical (that is, homologous) stretches of genetic
material, or are the genes for lysozyme synthesis, and the lysozymes
themselves, entirely different? This would appear to be the sort of
question that might be attacked directly by the comparative study of
protein structure. As we have already seen, in connection with cyto-
chrome c, ribonuclease, hemoglobin, and other proteins, there is ex-
cellent evidence which indicates that many homologous genes do ap-
pear to have survived happily through long periods of time, some
well exceeding the span of the fossil record.

A Biochemical Approach to the Species Problem

The paleontologist, in estimating the rates and directions of evolu-
tion, must depend almost entirely on morphological evidence. Even
with this relatively crude sort of yardstick, he can begin to distin-
guish patterns of change such as we discussed in Chapter 1, in con-
nection with the characteristics of tooth structure in the evolving
horses. He is limited, however, to the results of evolution and can
never hope to elucidate the underlying physiological changes that
participate to produce new phyla.

Although most of the ancient species disappeared, representatives
of almost all the phyla escaped extinction by adapting to their new
environments, thus perpetuating large parts of heredity. We have
available to us, then, a contemporary sample of the life of the past
from which we should be able to deduce a great deal about the fac-
tors that were decisive in phylogenesis long ago. The study of
“biochemical evolution” has already been of considerable value in the
establishment of biological interrelationships. For example, the oc-
currence of melanocyte-stimulating, oxytocic, and vasopressor hor-
monal activity in extracts of the neural gland if tunicates furnishes
strong evidence in support of the assignment of this subphyllum, the
Urochordata, to the direct pathway between the invertebrates (which
lack MSH activity) and vertebrates. The presence of both arginine
phosphate (an invertebrate phosphagen ) and creatine phosphate (the
typically vertebrate phosphagen) in tunicates adds additional support
to this assignment.

We shall not attempt to discuss here the numerous contributions
of this sort that biochemistry has made to evolutionary theory. The
reader will find this material summarized in a number of comprehen-
sive essays and books.!:** Our present concern is primarily with

GENES, PROTEINS, AND EVOLUTION. 213
 

Figure 96. An electron photomicrograph of collagen fibrils from bovine skin.
Magnification 42,000. Obtained through the kindness of Dr. Jerome Gross,
Massachusetts General Hospital, Harvard University Medical School.

214 THE MOLECULAR BASIS OF EVOLUTION

the biochemical changes in protein molecules that are much nearer,
in metabolic terms, to the genes themselves than are such products
of enzymatic action as the phosphagens, or the eye pigments of
Drosophila. As Wald has put it “It is a truism in biochemistry that
each species of animal and plant possesses specifically different pro-
teins.” The full understanding of speciation must, almost certainly,
be sought in the structure of proteins. To expand this point a bit,
let us consider two elegant examples of speciation that are demon-
strably related to changes in protein structure.

The protein collagen is largely responsible for the physical prop-
erties of such structural tissues as skin and cartilage. When collagen
fibrils (Figure 96) are exposed to heat they change markedly in in-
ternal structure and yield the molecular form known as gelatin.
Now recent studies by X-ray crystallography have shown that the
collagen molecule is very likely composed of three strands of poly-
peptide, cross-linked through a system of hydrogen bonds of con-
siderable strength". The amino acid sequence, Gly.Pro.Hypro, ap-
pears fairly frequently along the chains, and the hydroxyl groups on
the hydroxyproline residues are presumably major contributors to the
hydrogen bond network. When collagen is heated in solution, the
hydrogen bonds become ruptured at a critical temperature, known
as the “shrinkage temperature,” and the organized structure is quickly
disoriented to form the more globular and amorphous gelatin struc-
ture. Although the exact mechanism of this rearrangement is not
known, it is possible, on the basis of the results of current work on
the properties of synthetic polyproline and of mixed polymers of
proline and glycine, that the shrinkage may be associated with a
cis-trans isomerization at proline-proline or proline-hydroxyproline
bonds,* in conjunction with ordinary “entropic” denaturation.

On the basis of these chemical and physical observations we might
suppose that collagen molecules, suited to either cold) or warm
habitats, could be devised by nature through the introduction or de-
letion of hydroxyproline residues. Animals living in climates tending
to be very warm would do well to utilize collagens with high shrink-
age temperatures, and those living in cold climates could do with
considerably fewer sites for cross-linkage and with lower shrinkage
temperatures.

The studies of K. H. Gustavson and of T. Takahashi on the col-
lagens of fishes suggest that this is precisely the mechanism which
has been employed." The shrinkage temperatures of cold-water fishes
are always lower than those of warm-water fishes, and an amazingly
linear relationship exists between shrinkage temperature and the con-

GENES, PROTEINS, AND EVOLUTION 215
=
w

 

 

 

 

 

 

 

 

 

 

 

 

 

5 ©i11
oS 10
212 en)
oo
S
vi
9
Bu o
Los) ©8
o
S
_ 10
o
m 9
5 ®% ©
< 495
8 3 20
= @3
S
a
z 7p
Ss
n>
=x
6
30 40 50 60 70

Shrinkage temperature, 7,, °C

Figure 97, The relationship between the hydroxyproline contents of the collagens
of various fishes and their “shrinkage temperature.”

tent of hydroxyproline (Figure 97), although the vertebrate collagens
are otherwise extremely similar in composition. It is a provocative
fact that collagen shrinkage temperatures seem to fall about 15 or
20° above the highest temperatures likely to be encountered by a
species, as though this margin of safety were adequate in the ordi-
nary course of climatic events.

The relationship between the visual pigments of marine fishes and
the depths of their habitats is another dramatic example of adapta-
tion through modification of protein structure. Denton and War-
ren,’ Munz,* Wald and his colleagues,® and many other investigators
have studied the chemical structure and the spectral properties of a
variety of fish rhodopsins. Rhodopsin, composed of a vitamin A
derivative complexed with a protein, opsin, constitutes the light-
sensitive element of the retinal rods. The vitamin A-like prosthetic
group, retinene, responsible for light absorption, has been found to
be identical in all the species listed in Table 17. Since opsins do
not themselves absorb light in the spectral interval between 480 and
503 mp, the shifts in the position of the absorption maxima shown
in Figure 98 must be attributed to the effects of the opsins on the

216 THE MOLECULAR BASIS OF EVOLUTION

TABLE 17
Spectral Properties of Rhodopsins from Various Fishes

Summer

Range of

Depth, Amaxs

Species fathoms my E540/ Emax
Summer flounder (Paralichthys dentatus
Linnaeus) 2-10 503 0.695

Scup (Stenotomus versicolor Mitchel!) 1-20 498 0.586
Butterfish (Poronotus triacanthus Peck) 1-30 499 0.610
Barracuda (Sphyraena borealis DeKay) 1-10 498 0.575
Cod (Gadus callarias Linnaeus) 5-75 496 0.530
Cusk (Brosme brosme Miiller) 10-100 494 0.455
Lancet-fish (Alepisarus ferox Lowe) >200 480 0.250

spectral properties of retinene. The mechanism by which conjuga-
tion with opsin can induce a change in the spectral properties of
retinene is quite obscure. We have previously discussed a related in-
stance of a spectral shift, where the absorption characteristics of the

 

12 |
Lancetfish

— Flounder
1.0,

0.8 |

0.6 |-

Extinction

0.4 |-

0.2 |-

 

 

 

 

 

450 500 550 600
Wave length, mu

Figure 98. Absorption spectra of rhodopsins of marine fishes in 2 per cent aqueous
digitonin solution. The maximum absorption (Amax) shifts toward shorter wave-
lengths in rough correlation with the depth of habitat. See Table 17.

GENES, PROTEINS, AND EVOLUTION 217
tyrosine residue are modified by hydrogen bonding of the hydroxyl
group. The shift in this case was small, of the order of a few milli-
microns. In the rhodopsin absorption system maxima differ by as
much as 23 mp as we move from the summer flounder to the lancet
fish. This large shift implies a major change in the nature of the
interaction between protein and prosthetic group.

The biological observation that makes all this of special interest is
the fact that a correlation is observed between the mean depth of
habitat of the various species of fishes and the spectral properties of
their visual pigments. The correlation is not at all exact and, as the
data in Table 17 show, a wide spread of X,,,, exists at all depths.
Nevertheless, the information available is sufficient to form the basis
of a strong hypothesis. Over twenty years ago G. L. Clarke observed
that the increasing blueness of light with depth in the ocean raises
“the question of the possibility of a shift in the sensitivity of the eye
of a deep-water fish toward the blue end of the spectrum.” This
possibility is realized in the spectroscopic observations just listed, and
it is now possible to apply the techniques of protein chemistry to the
elucidation of the details of this fascinating chapter in biochemical
ecology. The study of the structural modifications in opsin which
have taken place during the evolution of the fishes will be especially
interesting since, as we have seen for the insulins and cytochromes c,
large spans of evolutionary time may pass without too extensive a
change in a particular protein molecule. The changes in the opsin
molecule may be so cleverly contrived and so incisive that extensive
alterations in sequence and folding have been unnecessary. On the
other hand, if alterations have been extensive, we shall be required
to rationalize a very complex set of interactions between protein and
prosthetic group. Both alternatives are intriguing, to say the least,
and the study of the chemistry of the opsins should make a most
valuable contribution to the understanding of evolution at the molecu-
lar level.

The Rate of Evolution

As G. G. Simpson has pointed out, the question “How fast has evo-
lution occurred?” is meaningless without the addition of the qualifica-
tions, “the evolution of what organisms, of which of their structures,
and at what time in their history.” The opossum, for example, has
changed relatively little in the past 80,000,000 years, whereas the

218 THE MOLECULAR BASIS OF EVOLUTION

evolution of the horses during the past 60,000,000 years has involved
at least eight distinct genera.

Just as the anatomical organization of some organisms has changed
much more rapidly than that of others, it seems likely that we shall
find a large spread in the rates at which specific protein molecules
have been modified during evolution. Although our basis for discus-
sion of this point is still very thin, it is already evident that some pro-
teins have undergone far greater structural change than others over
an equivalent period. Compare, for example, the somatotropins of
the sperm whale and the sheep with the insulins of these same species.
Although the changes in insulin structure have been restricted to very
minor modifications in a limited part of one polypeptide chain, the
somatotropins are quite markedly modified in molecular weight, in
cystine content, and in the number of polypeptide chains. Equally
striking differences in degrees of modification exist between numerous
others of the examples discussed in Chapter 7.

How are we to plan our experimental approach in attempting to
establish some chemical coherence in the tremendous puzzle of specia-
tionP We must, it would seem to me, begin with the basic assump-
tion that the phenotypic character of a species is primarily determined
by its unique spectrum of proteins. We may then proceed to a
study of the extent to which each of the individual proteins within
any spectrum may be modified without loss of biological function.
As we already know, the degree of “violability” of different proteins
may vary enormously as judged from the results of in vitro studies on
denaturation and chemical modification in relation to function. Even
here, however, many of the observed differences in sensitivity may
be overemphasized and may depend on the choice of methods used
for modification. Even though two proteins may be very similar in
regard to the proportion of their total structure that is essential for
function, one set of reagents may attack critical parts of one and not
seriously alter the other. Amino groups, for example, may be acety-
lated with essentially no effect in pepsin, but at least some of these
Same groups appear to be critical for the activity of lysozyme. A
proper comparison of two biologically active proteins thus must de-
pend on the use of a wide variety of inactivating reagents, and ulti-
mately on the deliberate degradative sort of study that aims to reduce
proteins to their minimum, functionally adequate size.

Since, however, proteins can be modified without loss of function,
it seems certain that the permissible degree of modification, in terms
of fractions of their total structure, will vary somewhat from molecu-

GENES, PROTEINS, AND EVOLUTION 219
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a eetcin sof aivon a does not seem too farfetched to think of SERS
ganism as being subdivisible into those that au & 2 &
have structures quite closely tailored to an essential functional re- = £ s 3 F oe
quirement, those that are designed with only moderate “efficiency” ° 21 BRS 4 * ~
or whose function is relatively dispensable, and those that are inter- 3—|— —_t— Q 9 > a Ba
mediate. ‘ Once again, illustrations come to mind. Several individuals NS g Bg 3 = B8
y slight clinical abnormality have been shown to be ST = ge e° go
completely devoid of serum albumin. These individuals, able to lead 4 an + g _ ° = 8
a normal existence, are living evidence for the dispensability of this a ele a 5 £ 53 3 A
protein under the ecological circumstances peculiar to humans. On x7 3 Boge os
the other hand, no one will question the inability of most species to §—— 5 8 Be SEs
survive in the absence of cytochrome c or of the enzymes necessary = 2 at gs é §
for oxidative phosphorylation. © “ gay Ze ™
. If we accept these subdivisions of the protein spectrum, we may © x | Je a er "
“express” a species in terms of a hierarchy of protein structures rang- S £38 3 E Bi De
ing in violability from none to very much. The further evolution 8 2% 232 as *
of this species, involving the usual mutation and natural selection, «| +e Se e 332
would then be reflected in a change in its proteins, one end of the > gf Jt PSG Bo
spectrum remaining relatively fixed while the other may change con- 2—|— Og Bye 2s
siderably. Thus the cytochrome c molecule, which we might think a x 5 3 . s 5 2
of as a relatively “primitive” protein, and indispensable for most life, ee sf 5 & 4 Org a=
would be stubbornly perpetuated in the evolving phyla with minimal ~__| \) 328 aw Zoe 8s
change, whereas the structure of the serum albumins might fluctuate =—|- =8 2 3 m8 &
with the shifting parameters of natural selection. From time to time SON —t 5 S q5 Za
entirely new protein structures might arise, as in the dramatic ap- ngs \ | _ ES fe Ae g
pearance of insulin and of other hormones at the point in evolution . 1S ease ee
when the protovertebrates and vertebrates appeared. The molecu- —— 2 ge o 2 E =& = i
lar basis for such “explosive” appearance of new protein entities is s—t— |) 28 8 = 23 PS,
of course, completely obscure. Only a thorough understanding of eT “si Ses. 5 : :
the processes of protein biosynthesis and of genetic information trans- < 27 —t— é ie z 3 a
fer will enable us to choose between such alternatives as the de novo s—l— Bs gE i B
creation of a whole new gene as opposed to the fortuitous reshufflin 3 Ss v8 . 4
of already available genetic units. 8 ~ ‘S a3 é £2%
We may safely predict that the patterns of change observed in the = z ; 5 ; 3 g .
protein spectrum by future biochemists will not always be smooth 7 3— \ —_t— . Bs Bee 3
and tidy. The criteria of natural selection will differ greatly from o-—— & 8 = 8 & Ee
species to species, from environment to environment, and from period ~ 3 2 E 23 $s 2
to period, and the survival value of gene mutations in a population my Bp Joe He see ed
and of their images in the phenotype, will be quite varied. | ess 2 YS =
| we ~ eos5i.%
A large number of important aspects of evolution have been omitted P 2 33 a 2 S
from this book. Some of these omissions may be attributed to type- “Sees
THE MOLECULAR BASIS OF EVOLUTION GENES, PROTEINS, AND EVOLUTION
221

220
writer fatigue. Most of them, however, have been purposely omitted
because of the lack of adequate factual material for discussion, and
this book is already well supplied with speculation. We might, for
example, have taken up the question of the spatial organization of
genes in relation to function. The recent studies on the mapping of
genes related to histidine biosynthesis in Salmonella (Figure 99) by
Hartmann, Demerec, and others indicate that the various cistrons
associated with the series of intermediate enzymes occur in the same
region of the genetic strand and that these genetic determinants are
arranged on the gene map in the same order as the reaction sequence
itself. A schematic representation of this linkage map is given in
Figure 99. The evolutionary implication that linked biochemical steps
have been added, successively, in sequence along the chromosome is
a very exciting one but is clearly not general. In Neurospora, for ex-
ample, genetic loci for closely related enzymatic steps are scattered
at random throughout the chromosomal apparatus.

We might, also, have spent some time on the question of cytoplasmic
heredity, which we know to be of importance in many biological sys-
tems. Here again the scarcity of published information is a limiting
factor. The study of inheritance of traits in a non-Mendelian fashion
is likely to be difficult and confusing, and the genetic or biochemical
study of such traits might receive disproportionately little attention.
As Nanney has recently suggested, “It is perhaps only natural that in-
vestigations of ‘messy’ characteristics are discontinued before publica-
tion and that investigators move on to traits more readily analyzed.”
The omnivorous reader will find Nanney’s review’ of the subject of
cytoplasmic heredity in The Chemical Basis of Heredity excellent
reading.

The list of omissions can be extended. The chemistry of RNA and
its genetic properties, the rearrangements of genes within the chromo-
some and the phenotypic consequences of such rearrangements, the
problem of polyploidy, the interactions of nonallelic genes—many of
these might, even now, be discussed with some intelligence in bio-
chemical terms.

The relationships between genotype and phenotype will, predict-
ably, become a major preoccupation of more and more “pure” and
medical scientists during the coming years. This book has grown out
of my own attempts to arrive at some sort of appreciation of the
potentialities of chemical genetics and the evolutionary approach.
It will have been well worth the effort if it can help to stimulate
the growing interest in evolution as the central theme in the life
sciences,

222 THE MOLECULAR BASIS OF EVOLUTION

REFERENCES

1. J. B. S, Haldane, The Biochemistry of Genetics, Allen & Unwin, London,
second impression 1956.

2. M. Florkin, Biochemical Evolution, Academic Press, New York, 1949.

3. G. Wald, in Modern Trends in Physiology and Biochemistry, Academic
Press, New York, 1952.

. A. Rich and F. H. C. Crick, Nature, 176, 915 (1955).

. W. F. Harrington, Nature, 181, 997 (1958).

. K. H. Gustavson, The Chemistry and Reactivity of Collagen, Academic
Press, New York, 1956. The analytical studies of Takahashi are discussed
on pages 224-227.

. E, J. Denton and F, J. Warren, Nature, 178, 1059 (1956).

F. W. Munz, Science, 125, 1142 (1957).

- G, Wald, P. K. Brown and P. S. Brown, Nature, 180, 969 (1957).

D, L. Nanney, The Chemical Basis of Heredity (W. D. McElroy and

B. Glass, editors), Johns Hopkins Press, Baltimore, 1957.

Ooo

Sn an

GENES, PROTEINS, AND EVOLUTION 223