lyoo ©00 @[f [l [f · neously in meat. he and his contemporaries thought that the spontaneous...

lYOO� ©00�� @[f [L�[f�� MOLECULES AND NATURAL SELEOION

Copyright© 1973, by John Wiley & Sons, Inc.

All rights reserved. Published simultaneously in Canada.

No part of this book may be reproduced by any means, nor transmitted, nor translated into a machine language without the written permission of the publisher.

Library of Congress Cataloging in Publication Data

Orgel, Leslie E The origins of life.

Bibliography: p. 1. Life-Origin.

I. Title. [DNLM: 1973] QH325.068 ISBN 0-471-65692-5

2. Chemical evolution. 1. Biogenesis. QH 325 0680

577 72-10534

ISBN 0-471-65693-3 (pbk)

Printed in the United States of America

10 9 8 7 6 5 4 3 2

PREFACE

This book is not written for professional b iologists or chemists, but rather for col lege or advanced h igh school students and general readers who have a l im ited backg round in chemistry or b iology. I have t ried to show that studies of the origins of l ife have recently progressed to a point at which i t i s possible to p ropose plausible mechanisms for most of the steps i n the evolut ion of liv ing organisms from the i norganic constituents of the pr imit ive earth. There are, of cou rse, enormous gaps in our knowledge, but I bel ieve that the or igi ns of l ife can now be d iscussed fru itfu l ly with i n the framework of modern chemistry and evolutionary bio logy. The extension of the d iscussion to l ife on other p lanets is straightforward, but necessari ly very speculative.

I have adopted a n um ber of p rocedu res which wou ld be out of p lace i n a special ized monograph. Only the main current of thought on the orig i ns of l ife is d iscussed and l itt le mention is made of alternatives. I have not g iven references to the or ig i nal l iterature, but instead have included a short b ibl iography that should enable the reader to trace ideas to their sources. I emphasize that I c la im no priority for the i deas that I present; they are the work of many hands.

The sequence of chapters may occasion some surprise, for I have not a lways developed subjects systematically, as i n a textbook. The log ical structure o f mo lecular b io logy, for example, is more eas i ly g rasped by the nonspecial ist than the chemistry that underl ies it . I have, therefore, inverted the

v

trad it ional order and d iscussed the consequences of basepair ing, coding , etc. before d iscuss ing the chemi stry i nvolved. A detailed d iscussion of the optical activity of l iv ing systems h as been relegated to an appendix l ate i n the book, s ince I reg ard this topic as a special aspect of b iochem ical specificity.

Although the sub-title of th is book is " Molecules and Natural Selection," I have l i ttle to say about n atural selection and the or igi n of species. References to books deal i ng with that subject are g iven in the bibl iog raphy.

I am indebted to D r. B. Chu. D r. F. H. C. C rick, Dr. M. J. Dowler, Dr. J. R. Holum, Dr. H . A. Orgel, Mr. R. M. O rgel, Dr. C. Sagan, and M r. R. Sti leman for valuable comments on the or ig ina l d raft of my manuscript. I also wish to acknowledge my g ratitude to the John Simon G uggenheim Memorial Foundation for a g rant which g reatly faci l itated the completion of the work.

San Diego, California

vi Preface

L. E. Orgel

PART ONE

'U'[][][g �li\'D'illOO[g @[}' 'D'OO� !?OO®OO!L[g�

TI Historical Background 3

� The Fossil Record 1 7

® Molecu lar Biology 33

� Topics in Biochemistry 61

@ The Biochemical Record (with an Appendix on Panspermia) 87

PART TWO

�'IT�!?� 'D'®Wli\0000 ill �®!Lill'D'O®� @ History of the Earth, Atmosphere,

and Oceans 99

w Sources of Energy 1 1 3

ffi Prebiotic Synthesis 1 23

® The Formation of Polymers 1 33

ll® Replicating Molecules and Natural Selection (with an Appendix o n Optical Activity) 1 45

llll From Repl icati ng Polymers to Cells 1 69

[]� Natural Selection 1 77

PART THREE

� lX 'D'illill 'IT� ill 00 [g�'U'ill Oill!L !LO [}'� TI® What is Life? 1 87 []� Extraterrestrial Organic Chemistry 1 99 TI® Life i n the Solar System 207

[]@ Intelligence in the Universe 21 9

Summary of the Main Argument 229 Bibliography 233 Index 235

PART

ONE

Historical Background

Introduction

The disti nction between the l ivi ng and the i nani mate was one of the fi rst to be drawn by man at the dawn of his cu ltu ral

· development. Most societies, whether primitive or advanced, have regarded the l iv ing world with special reverence. It is not su rpris ing, therefore, that the orig in of l ife has always been a subject of profound phi losphical interest. If many of the opin ions that were widely held even i n comparatively recent t imes now seem naive, it is not because the people who held them were un i nte l l igent. Although faced with problems that were beyond the technolog ical capacity of thei r t imes, the early ph i losophers and biologists often gave remarkably farsig hted answers to the fundamental questions about the natu re and orig in of l ife.

The last hundred years or so have seen an explosive ex-

3

pansion of our u nderstand ing of chemistry and bio logy. For the fi rst t ime the problem of the origins of l ife is coming with in range of the experi mental and theoretical tools at our disposal . Later in this book we shal l see how chemistry, b iology, geology, and astronomy are beg inn ing to throw exciti ng new l ig ht on the orig ins of l ife on earth. F i rst, however, i t is instructive to exam ine some of the opin ions he ld in the past. This should at least prevent us from underesti mat ing the problem ; no doubt many of our own ideas will seem naive in the futu re.

Spontaneous Generation

Most scientists would ag ree that d iscussions of the orig ins of l ife shou ld deal with processes that occu rred long ago on the pri mitive earth and which led, after m i l l ions of years, to the emergence of l iv ing cel ls from l i feless start ing materials. However, this point of view is a com paratively recent one. Most early cu ltures accepted that a god had created man and certain of the h igher animals. They also thought that other organisms, such as insects and mice, were sti l l bei ng generated spontaneously from mud or decayi ng organic matter.

The theory of spontaneous generation was developed and systematized by the Greeks and received its most inf luential treatment in the writings of Aristotle (384-322 B.C.).

For two thousand years from the t ime of Aristotle, educated men did not question the theory of spontaneous generation even in its si mplest form. Aristotle's authority was not challenged in a serious way unti l the seventeenth centu ry, and it was not unt i l the m i ddle of the n i neteenth century that Pasteu r f inal ly showed that spontaneous generation does not occu r. How could a theory that now seems obviously wrong and even absurb have survived so long ?

Fi rst we m ust remember that the biological science of the Greeks was based very largely on casual observations, unaided by i nstru ments such as the m icroscope. Just as i mportant was the fai l u re of the Greeks to appreciate that properly designed bio logical experi ments are far more instructive

4 The Nature of the Problem

than chance observations. To the casual observer it indeed appears obvious that young h umans develop wit hin their mothers, that birds eme rge ful ly made from eggs, and that maggots form spontaneously in decaying meat. I t req uires a great deal more than casual observation to discover that the third conclusio n is incorrect.

With a l l our modern backg round in bio logy and medicine it is hard to realize that until the seventeenth centu ry intel l igent, wel l -trained professional men, doctors and scientists, would h ave held al l three conclusions to be equal ly self-evident. I t required men of genius to question the theory of spontaneous generation, and it req uired the establishment of new standards for the design and interpretation of experiments final ly to discredit it.

Aristotle, despite his unquestio ned intel lectual preeminence, was not a lways a careful observer. If he could state that men have more teeth than women, it is not too surprising that he thought that frogs could form from damp earth. I t is a little more unexpected that van Helmont ( 1 580-1 644), one of the foreru nners of scientific chemistry and bio logy, published a recipe for producing mice from soiled cloth ing and a little wheat. He was not too su rprised when he found that these mice, des pite their singu lar origin, mated successfu l ly with normal animals.

It was an I talian bio logist, Redi, who in a crucial series of experiments published in 1 668, opened the attack on the theo ry of spontaneous generation . In retrospect Redi's experiments seem simple. I t was wel l known that meat, when left to decay in the open , b reeds maggots. Redi covered the meat with fine meshed muslin. He found that under these circ umstances no maggots appeared. A careful examination of the muslin revealed the presence of eggs that were too large to pass through the mesh . After that it was not difficult to show conclusively that so long as insects and their eggs are kept away fro m decaying meat no maggots appear.

This demonstration did not deal a fatal b low to the theory of spontaneous generation , since, at that time, it was not c lear that resu lts obtained fo r a sing le species could be generalized to the whole living world. Redi himself never c laimed to have disproved the theory of spontaneo us gener-

5 Historical Background

'Pil.tt Je · ""$� C11�, Je rttril1riJ� .tlmui.s }lllr: 1J.':PI1rl1tltt-in_F� .

Figure 1.1 . The end of the prescientif ic period. Doves and anthropomorphic flowers g rowing on orchids. An apparently normal dove is seen just after detaching itself from the plant on the l eft. What happened to the l ittle men is not clear. This f igure is taken from the Mundus Subterraneus of Athanasius Kircher, Amsterdam, 1665.

Figure 1.2. The beginning of contro l led experimentation on generation. A typical life cycle worked out by Redi (for the Elder Fly). The figure is taken from Esperienze lntorno alia Generazione degl'lnsetti, Florence, 1668. It fi rst appeared only three years after the original of Figu re 1 . 1 .

::;:

ation; he maintained o n ly that maggots d id not arise spontaneously i n meat. He and h is contemporaries thought that the spontaneous generation of other insects probably d id occur. Nonetheless, it is u n l i kely that the theory of spontaneous generation would have survived long had i t not been for the work i n it iated by Anthony van Leeuwenhoek (1632-1723), one of the fi rst microscopists.


During the latter part of the seventeenth centu ry Leeuwenhoek pub l ished a series of detai led descriptions of the "ani malcu les" which he found to be very widely d istri buted, for example, i n materials as different as rai nwater and dung. Today, in p lace of "animalcu le," we use the word microorganism. Leeuwenhoek seems to have discovered representatives of most of the major classes of m icroorganisms, inc luding bacteria and yeasts. He showed that numerous "animalcu les" could be seen whenever organic extracts were al lowed to stand for long periods in contact with air and were then examined under the microscope.

Microorganisms are so small that they pass through musl in and the other materials which Redi had used to keep f l ies and the ir eggs away from meat. Thus it often happened that m icroorganisms penetrated barriers which, i n Red i's t ime, were thought adeq uate for thei r exclusion. When th is happened , new microorganisms seemed to appear spontaneo usly.

Since, at that time, many people sti l l bel ieved that insects arose spontaneously, it is hardly su rpris ing that the spontaneous generation of these newly d iscovered m icroorganisms was often taken for g ranted. However, thanks to the work of Red i , th is point of view was not held un iversally. Leeuwen hoek h i mself bel ieved that microorganisms fel l i nto h is solutions from the air. In an attempt to settle the matter Louis Job lot, i n 1 71 8, carried out the fi rst experiment of a type that was to be repeated with increasi ng attention to detai l fo r the next centu ry and a half.

Joblot boi led plant extracts fo r several m inutes and then divided these steri le i nfusions into two portions. He left one of them in an open vessel and the other one in a vessel covered with parchment. M icroorganisms appeared i n the open vessel but not in the closed one. Next, Joblot removed the parchment covering and showed that th is previously steri le so lution soon developed its own population of microorganisms. Joblot co ncl uded, correctly, that the microorganisms were not generated spontaneously.

Joblot's work did not co nvince his col leagues, and the conf l ict proved s ingu larly diff icult to resolve. In retrospect it is clear that many scientists of the e ighteenth and n ineteenth


centu ries real ized neither the diff icu lty of exclud ing bacteria completely from an organic sol ut ion, nor the resistance of certai n organisms and thei r spores to heat steri l ization .

John de Turbevi l le Needham (1 71 3-1 781 ) carried out a very extensive series of experiments from which he concluded that, even when al l possi b le precautions are taken , m icroorganisms do appear spontaneously i n previously steri l ized solutions. Spal lanzani carried out better experiments in 1 765 and reached the opposite conclusion. He critic ized Needham's experimental techniq ues; in particular he claimed that heat steri l ization of the covered containers had been i nadequate in Needham's experiments. Needham counterattacked by sayi ng that Spal lanzani had spoi led the broth and the air in his f lask by heat ing them too i ntensely; m icroorganisms did not develop because no adequate nourishment remai ned. There seems l itt le doubt that Spallanzan i won this contest by means of bri l l iantly executed exper�ments, but the theory of spontaneous generation was so fi rmly entrenched that he fai led to convi nce all of his contem poraries.

The technical problem of obtain ing steri le sol utions was aggravated by the real ization that oxygen from the ai r has a profound effect on the decay of organ ic matter and is requ i red for the growth of many types of microorgan isms. The suppo rters of spontaneous generation qu ite leg it imately i nsisted that ord inary "unspoi led" air be present along with the nutrients in any test of the theory. They clai med that a i r is spoi led by i ntense heat ing , so the ir opponents were forced to devise tech niques for steri l iz ing a stream of a ir without exposi ng it to h igh temperatures. Th is d i ff icu lt problem was f inal ly solved by one of the g reatest of a l l experimenters, Louis Pasteur.

The French Academy offered a prize for the most convincing experiments shedd ing l ight on the orig ins of l iv ing creatures. I n 1 862, Louis Pasteur won the prize for a series of experiments that f inal ly d iscredited the theory of spontaneous generation. He analyzed the fau lts of previous experiments and then designed an experimental program of h is own to e l im inate them. His f i rst step was to d raw a stream of air through a pad of gu ncotton. The accumulated dust was


extracted with an organic solvent and the solid that remained was examined under the microscope. It proved to contain microorganisms in great number and variety. Thus the possibility that his solutions became contaminated by airborne organisms could not be questioned.

Next Pasteur devised a number of ingenious procedures for sterilizing a stream of cool air. Once he had solved this problem he was able to show that no germs were formed in previously boiled solutions. In one very beautiful series of experiments, Pasteur partially filled a flask with broth and then drew the narrow neck of the flask into an""· He next boiled the contents of the flask long enough to sterlize them and then allowed the flask and its contents to cool. Although the flask was freely open to unheated air, its contents remained sterile for, as Pasteur had anticipated, all airborne particles had been trapped on the curved surfaces of the--. When the--shaped neck was cut off, the broth soon began to decompose, thus showing that it was still able to support the growth of microorganisms once they got in. Since the air in the vessel was never heated, Pasteur's opponents could not claim that it was "spoiled."

(a) (b)

Figure 1.3. Louis Pasteur's flasks. The end of the theory of spontaneous generation. The contents of the flasks are boiled and then allowed to cool. (a) No microorganisms appear because airborne particles cannot get past the ""-shaped bend in the tube. (b) Once the bend is cut off, the solution develops a large population of microorgan isms. There is no q uestion of "spoi l ing" the air or the broth, since nothi ng is done to the body of the flask or to its contents in going from (a) to (b).


These and many si m i lar stud ies were of enormous i mportance for the development of bacteriology. They received great attention from the scientific commun ity and were very widely believed to have shown that spontaneous generation is not a un iversal , contemporary process. They helped to raise the question of the orig ins of l ife i n its modern form for the fi rst time.

Recent Evolutionary Theories of the Origins of Life

Unti l the latter half of the n ineteenth centu ry it was general ly bel ieved in Western societies that God had created the hig her animals, once and for all , "according to thei r ki nd ." Pasteur's work showed that no l iving organisms, not even bacteria, come into existence except as the descendants of s imi lar organisms. At earl ier periods, the theory of special creation could perhaps have been extended to inc lude microorgan isms - God might have created them along with the h igher an imals but then decided not to mention them in Genesis. The development of Darwi nian evolutionary theory soon made such explanations unacceptable.

Theories of the orig ins of l ivi ng organ isms have always been strongly influenced by contemporary rel ig ious and phi l osoph ical opin ions. I n the period immediately fol lowing Pasteur's work on spontaneous generation a very b itter pub l ic controversy was rag ing i n England between orthodox Christians and Charles Darwin 's more enth usiastic disciples. This gave a special flavor to subsequent d iscussions of the orig ins of l ife.

Darwin in his theory of evolut ion through natural selection proposed that species are not i nvariant, but change slowly with the passage of ti me. Darwin was not the fi rst to advance a theory of evolution , but he was the fi rst to support his theory with a mass of evidence sufficient to convince most men of science of his day. He and Alfred Russel Wal lace were also the f irst to propose an acceptable explanation of evolution - natural selection or the survival of the fittest.

We shal l return to this i dea at greater length elsewhere.

1 1 Historical Background

Crudely, it was based on two generalizations. Firstly, in any species there is a good deal of variation from individual to individual. Secondly, children tend to resemble their parents more closely than they resemble members of the population chosen at random. Darwin argued that those individuals best adapted to their environment-the fittest-would survive to reproductive age more often than would less fit individuals. Hence the fittest individuals would, on the average, contribute more children to the next generation than would the less fit. Since children tend to resemble their parents, it follows that the children's generation would contain a higher proportion of fit individuals than had the parental generation. Thus, with the passage of time, the species would gradually change in the direction of better adaptation -the fittest would win out.

Darwin's theory differs radically from previous views on evolution, since it attributes change to the operation of chance. Variation within a species arises in a haphazard way; some changes are beneficial and others detrimental. Natural selection eliminates the detrimental variations and preserves those that are beneficial. This is a very different point of view from that put forward in the evolutionary theories current before the time of Darwin and Wallace. According to some of these theories evolution is willed; a species improves because parents pass on to their descendants "information" that they have acquired from the environment during their lifetime. In other pre-Darwinian theories of evolution, the environment is supposed to act directly on the members of a species and to bring about inheritable changes in them.

Most of Darwin's supporters were prepared to generalize his theory and to assume that living organisms could have arisen from the inorganic world by an evolutionary process. Many Christians found this hard to accept, since it implied that the evolution of primitive organisms from inorganic matter and of man from primitive organisms could have occurred without the intervention of any supernatural creative act. Thus the problem of the origins of life became a central issue in the wide-ranging controversy between religion and biology. More than a hundred years later the State


Board of Ed ucation i n Cali fornia takes essential ly the same posit ion as Darwin 's opponents: b io logy textbooks used i n schools in Cal iforn ia are req u i red t o g ive equal wei ght to evo lutionary theories of the or ig ins of l ife and to the story of creation recounted in the B ib le.

Darwi n h i mself was cautious in his pub l ished writi ngs, but there is no doubt that he bel ieved that l i fe cou ld have or ig i nated spontaneously on the pr im itive earth . The fo l lowing quotation from one of h is letters makes th is q uite c lear.

It is often said that all co nd it ions for the fi rst prod uction of a l iv ing o rgan ism are present, which would ever have been present But if (and oh, what a b ig if) we cou ld conceive in some warm l i ttle pond , with al l sorts of ammonia and phosphoric salts, l i ght , heat, electr ic ity, etc., present, that a protei n compound was chemically formed ready to un dergo sti l l more complex changes, at the present day such matter wo ul d be i nstantly devou red or absorbed, which wou ld not have been the case before l iv ing creatu res were fo rmed.

Whi le Darwin and many other n ineteenth and early twentieth centu ry scientists bel ieved that l ife might have evolved from i norgani c materials on the prim itive earth, they appreciated that organi c chemistry had not yet advanced to a point at which an exper imental attack on the prob lem of chem ical evolution co uld prove usefu l . The reaction of the Swed ish chemist , Svente Arrhen i us, to the fal l of the doctrine of spontaneous generation is typical of that of another g roup that inc luded many dist ingu ished chemists and phys i c ists. He bel ieved that, s ince l ife was no longer appeari ng afresh on the earth , i t must have been introduced from another planet. Arrhen i us, in a series of scientific papers and popular books, developed a theory cal led Panspermia accord ing to which organisms, aided by the pressure of rad iat ion, could have made the long journey across space from another solar system.

Neither this idea, nor the somewhat related notion that l ife was carried to the earth on a meteorite, i s rid ic ulous, although both are open to serious objections. However, even if correct they do not solve the p roblem of the or ig ins of l i fe


but merely transfer the problem to another planet To overcome this criticism, Arrhenius and many of his supporters argued that life must be eternal. I n this way they abolished the problem by decree-since life is ete rnal the q uestion of its origin does not arise.

Today this view seems mystical. Pres u mab ly, each planet has evolved from inorganic dust and simple gases f loating freely in space. Hence, it seems that the spontaneous creation of living organisms on any p lanet by other than evo lutionary processes wou ld requ ire su pernatu ral intervention. Arrhenius' theory shifts the site of the origin of life to another planet, but does not provide a mechanism for the origin of l ife.

In summary, little important work on the origins of life was carried out in the period immediately fol lowing the revol utionary p u bl ications of Pasteur and Darwin. Experimental work was largely restricted to f ruitless attempts to prove that microorganisms can be produced in special environments, for example, in organic material subjected to the inf luence of radium. Although some scientists believed that life had evolved from inorganic matter long ago, they also believed, correctly, that experimental approaches to the problem were premature. A large and infl uential g roup did not be lieve that life cou ld have evolved spo ntaneously and accepted either explicity or impl icity the need fo r supernatu ral intervention.

It may not be an accident that the next major step forward was taken in a society that had deliberately adopted a materialistic philosophy and was actively antireligious. I n 1 923, i n Russia, A. I . Oparin suggested that the atmosphere of the earth, long ago, was very different from the present atmosphere. In particu lar, it did not contain oxygen but rather hydrogen and other reducin g compounds, such as methane and ammonia. Oparin proposed that the organic chemicals on which life depends formed spontaneously in such an atmosphere, u nder the infl uence of sun light, lightning, and the high temperatu res existing in volcanos. A similar suggestion was made independently by anothe r materialist, J. B. S. Haldane, in Eng land.

These two p roposals infl uenced most of the authors whose speculations about the origins of life were published

1 4 The Nature of the Problem

J--10 CM--1

Figure 1 .4. Stanley Mil ler's flask. The beginning of prebiotic chemistry. (See Chapter 8 for a ful l description of these experiments.) Reproduced with permission from J. Am. Chem. Soc., 77, 2352 (1955).

d u ring the 1 930s and 1 940s. It is surprising that experimental confirmation was delayed fo r so long, since the theoretical notions had such wide cu rrency. It was not u ntil 1 953 that Stanley Miller, working with Harold U rey in Chicago, demonstrated that impo rtant biochemicals are indeed fo rmed in su rprising ly large amounts when an electric discharge is passed through an atmosphere of the kind proposed by Oparin and Haldane.


These experiments were the beg inn ing of the most recent phase i n the study of the orig ins of l ife. They led to a detailed consideration of the conditions that must have existed at the surface of the prim itive earth and to attempts to reconstruct i n the laboratory the chemistry that would have occu rred under those conditions. Describ ing this work is one of the objectives of the present book.

There has, however, been a second major inf luence on our i deas about the orig ins of l ife. In 1 953, the Watson-Crick structure of DNA was annou nced and s ince then molecular biology has developed at an accelerat ing pace. Recent discoveries concern ing the structu re of the genetic system and the way in which it operates have sharpened enormously our understanding of what needs to be explained by a successful theory of chemical evo lution.

1 6 The Nature of the Problem

The Fossil Record

Geological Dating

The fossi l record is our on ly ::o urce of i nformation about the antiqu i ty of l ife on earth . Yet, unti l recently, i t was not possible to determ ine the ages of fossi ls. With i n the last twenty years a l l th is has changed. A new tech nique called rad ioisotope dati ng has been i ntrod uced which al lows geologists to determ ine the ages of foss i ls with considerable accuracy. Rel iable est imates of the ages of some very early forms of l ife can now be g iven .

Before the introduction of rad ioisotope dati ng , geologists were sometimes ab le to determine the relative ages of d i fferent geological strata by making use of two very s imple pr incip les. When layers of rock have been formed by sed imentation , that is by deposit ion at the bottom of oceans or lakes, the lower layers must clearly be o lder than the ones that l ie above them. This pr incip le does not apply to layers

1 7

(a) (b) Figure 2.1. Ordering of rocks according to their age: (a) the strata get o lder as one goes downward through a sedimentary column; (b) the i ntrusion must be younger than any of the sedi mentary layers that it cuts.

of igneous rock, that is to layers of rock which have been formed by the cooling of molten material extruded from the interior of the earth. Here, however, the second simple principle was sometimes applicable. If a layer of one type of rock cuts through a stratum of quite a different kind, then the layer that is unbroken must be the more recent.

Using this kind of information it was often possible to order the strata in a locality according to their relative ages, even when the strata had been disturbed by earth movements and mountain building. It is important to notice, however, that these methods do not provide any evidence about the absolute ages of the strata, nor do they help to determine the relative ages of rocks found in widely separated places.

Relative ages could sometimes be determined by an examination of the fossil record. The fossils found in sedimentary rocks represent the creatures that lived when the rocks were being deposited. Sometimes fossils are nothing more than molds of the harder parts of organisms. Sediment often fills the interior of clam shells, for example, and then gradually hardens. The shells later dissolve away leaving


Figure 2.2. Internal mou lds in Portland limestone. The moulds are gastropod and bivalve shel ls. (Reproduced with permission from The Elements of

Palaeontology by R. M. Black, Cambridge University Press, New York, 1970.)

beh ind i nternal molds. External molds develop i n a s im i lar way; some represent the exterior of the complete ani mal whi le others are restricted to as small a detai l as a footpri nt or a worm's burrow.

The most fami l iar types of fossi ls are those formed by replacement. Most shel ls are made of calc ium carbonate. I n t he course o f t ime s l ightly so lub le s i l icates contained i n ground water often rep lace the more soluble calci u m carbonate. The calci u m phosphate which makes up the bones of vertebrates can be replaced by s i l i cate in a s imi lar way. Remarkably faithfu l rep l icas of the harder parts of long-dead

Figure 2.3. A wel l-preserved fossi l Echinoderm (sea-urchin). (Reproduced with permission from The Elements of Palaeontology by R. M. Black, Cambridge University Press, New York, 1970.)


creatures are formed by this process. Since silicate fossils are much more d u rable than shel ls or bones, they outlast the remn ants of the orig in al animals.

To make use of fossils in determin ing the relative ages of rocks, we must assu me that if two rocks contain the same kinds of fossil they are eq ual ly o ld . This amo unts to supposing that each species persisted through roug hly the same period of earth history in all of its habitats. Original ly this assum ption was questioned, but once it had been accepted it enabled geologists to establish relations between strata in widely different localities. Thus if fossils of the same species were found in strata in Europe and North America, the strata were ded uced to be equal ly o ld .

By applying these simple ideas with g reat ingenuity and industry, geologists and palaeontologists were able to piece together an extraordinarily detailed picture of the seq uence of animals that have inhabited the earth. Their evidence was one of the foundatio ns on which Darwin based his theory of evolution. The problem of absol ute ages, however, remained to be solved ; work on the order in which species have succeeded each other did not provide information about the time periods involved. Early estimates of the ages of fossils were based on indirect evidence and were very far from correct.

The next major step forward in geology depended on radioiostope dating, a tech niq ue made possib le by advan ces in the chemistry and physics of radioactive elements. Radioactive elements are, of course, unstable. They emit hig henergy radiation and are thereby transformed into new elements. The products of radioactive decay usual ly have chemical p roperties completely different from those of the materials from which they are formed. The gas, argo n, for exam ple, is fo rmed by the decay of the meta l , potassium.

The study of radioactivity is compl icated, main ly because most chemical elements occur as several different isotopes. By this we mean that several kinds of atom, each having a different mass {weig ht) occur together in samples of the chemical element. Carbon, fo r example, occurs as fou r different isotopes, uc, 12C, 13C, 14C, with masses 1 1, 12, 13, and 14, respectively. One of these isotopes, 12C, is much

21 The Fossil Record

more abundant than the others. Although al l forms of carbon have the same chemical properties, the rare isotope 14C is radioactive, wh i le the abundant isotope 12C is stable.

From our poi nt of view, the most i m portant characterist ic of a radioactive isotope is its half-l ife. Each rad ioactive atom has a fixed probabi l ity of decaying per unit ti me. This probabi l ity is unaffected by the environ ment of the atom and by the time that the atom has already survived. Atoms get older, but they do not age. We define the half-l ife of an isotope as the length of time needed to g ive an atom a one-i ntwo chance of decaying. Equivalently, it can be defi ned as the t ime it takes for half of the atoms in a large sample to decay. Half-l ives vary from fractions of a second to b i l l ions of years.

The idea on which rad ioisotope dating is based is easi ly understood. Let us overs impl ify and suppose that the isotope of potass ium, 4°K, decays to an argon isotope, 40Ar, with a half-l ife of 1 .26 b i l l ion years. Then , if we found that a rock specimen contained equal numbers of 4°K and 40Ar atoms and if we cou ld be sure from the nature of the rock that all of the 40Ar must have been formed by the decay of 4°K, we could deduce that the rock was one half- l ife old, that is 1 .26 b i l l ion years old. If we found three 40Ar atoms for each 4°K atom we wou ld know that th ree quarters of the potassium had decayed and hence that the rock was two half-l ives old , that is 2.52 bi l l ion years o ld , and so on. To determine the age of the rock al l we would need to know is the relative abundances of 4°K and 40Ar.

I Z I

� I

u 1/4 ----L----<l I I e: 118-----�---- .:. ----

1116 ---- 1- ---- + ----- t- - --Figure 2.4. The decay of a radioactive element as a function of its half-life.


Real elements do not behave as s imply as we have supposed. 4°K decays to 40Ar, with a half-l ife of 1 .26 b i l l ion years, but it a lso forms a calc ium isotope 4°Ca. Diff icu lties of this k ind are not serious; they compl icate the arithmetic but they do not l i mit the appl icabi l ity of the method . Major d ifficu lties are usually encountered when the radioactive element or its decay products cou ld have been lost from the speci men. Soluble compounds are often washed out when water percolates through a rock, and volati le com pounds escape when the rock is heated. Consequently, porous rocks that have been exposed to water, or rocks that have been heated, are often unsuitable for dat ing in this way.

Th ree i mportant methods of radioisotope dating have been appl ied to old sed iments. They are based on the decay of u ran ium to lead, rub id ium to stronti um , and potassi um to argon, respectively. In every case the abundance of the radioactive element and of its decay products can be measured accu rately. By choosing rock speci mens with care and by employing more than one method of dating whenever this is feasib le, i t has been possible to establ ish the ages of many of the oldest geological formations.

The Classical Fossil Record

Geology and palaeontology were wel l-developed sciences long before absolute dates could be assigned to fossi l remains. This led to the development of unsystematic nomenclature , which is baffl ing , except to the geologist. In F igure 2.5 we g ive the geological eras and periods together with the ages which have been assig ned to them by radioisotope dati ng. The terms cenozoic , mesozoic , and palaeozoic are self-explanatory, at least to Greeks and a few others. They refer to su pposedly recent, m iddle, and ancient periods in the development of l ife. The names of the periods have varying derivations; many are named after local ities where thei r formations are particu larly wel l -represented.

The oldest fossi ls of largish ani mals date from the early Cambrian, about 580 mi l l ion years ago. These creatu res were marine invertebrates ; by the end of the Cambrian period about 500 m i l l ion years ago, sponges, jel lyf ish , starfish,


THE GEOLOGIC TIME SCALE

RELATIVE

DURATION IN DURATIONS

MILLIONS OF MILLIONS OF OF MAJOR YEARS AGO YEARS GEOLOGICAL

ERA

CENOZOIC

PERIOD EPOCH {RECENT QUARTERNARY

PLEISTOCENE

{"'""" MIOCENE

TERTIARY OLIGOCENE

EOCENE

PALEOCENE

} (APPROX.I (APPRO X.)

0-1 1

1-13 13

13-25 12

25-36 1 1

36-58 6

58-63

63-135 72

MESOZOIC JURASSIC { '"'"""�

135-1B1 46

TRIASSIC 1B1-230 49

PERMIAN 230-280 50

PENNSYLVANIAN 280-310 30

MISSISSIPPIAN 310-345 35 PALEOZOIC DEVONIAN 345-405 60

SILURIAN 405-425 20

ORDOVICIAN 425-500 75

CAMBRIAN 500-600 100

Although many local subdivisions are recognized, no world-wide system has been evolved. The Precambrian

PRECAMBRIAN MIDDLE r· lasted for at least 2% billion years. Oldest dated rocks are at least 2 ,700 miiJion, possibly 3,300 million, years old.

LOWER

Figure 2.5. The geolog ical t ime scale showing the standard d ivisions based on the fossi l record and the lengths of t ime obtained from dating rocks. Wi l l iam Lee Stokes, Essentials of Earth History,

2nd ed . , © 1966. Reprinted by permission of Prentice-Hal l , Inc . , Eng lewood Cl iffs, N. J .

INTERVALS

CENOZOIC

MESOZOIC

PALEOZOIC

and marine worms were a l l com mon. However, the domi nant animals were the tri lobites ; some species were as small as a p in head and others as much as eighteen inches long. The tr i lobites in t ime became ext inct , but many of the other Cambrian organisms are c learly re lated to animals that can sti l l be seen today in t idepools.


Fish were the f i rst verteb rates to appear. Fragments of "bony" coverings of fish- l ike creatu res are found i n deposits that are more than 420 m i l l ion years old, but fossi ls of fish . much more than 400 mi l l ion years old a re uncom mon. It seems that a remarkably rapid evolution of a g reat variety of fishes occurred i n the Devonian period about 380 mi l l ion years ago. The new species i ncluded some true fishes, such as the sharks , and also scaly fishes with l u ngs and l i mb- l ike fins. These latter were the ancestors of the amphib ians.

The fi rst amphibians were qu ite smal l , but by 300 mi l l ion years ago g iant salamander- l ike creatures had al ready taken to the land. The earliest land p lants evolved from seaweed or

Figure 2.6 A spinose trilobite from Czechoslovakia (x4). (Reproduced by permission of Professor H. B. Whitti ngton of the Sedgwick Museum, Cambridge.)


other algae about 400 m i l l ion years ago; 1 QO m i l l ion years later the land was covered with ferns and pr imitive trees. I t was in th is environ ment, perhaps 280 mi l l ion years ago, that the fi rst repti les appeared and then g radually displaced the amphib ians. The earl iest remains of modern insects date from about the same time.

The period from 180 to 60 mi l l ion years ago was the age of the repti les. These inc luded the giant d i nosaurs and flying pterosaurus as well as more fami l iar l izards, turt les, and crocodi les. The f i rst mammals evolved from repti les about 180 mi l l ion years ago, but they did not become dominant for more than another 100 mi l l ion years. A qu ite different evolutionary b ranch led fro m the repti les to the modern b i rds.

The mammals began to take over the land a l i tt le less than 60 mi l l ion years ago. All of the modern groups of mammals appeared with in the next 25 m i l l ion years. Thirtyfive m i l l ion years ago, pri mitive horses, pigs, rodents, and monkeys were a l ready establ ished.

The mammals conti nued to evolve and d iversify. Palaeontological evidence suggests that man's ancestors d iverged from other g roups of apes about 20 mi l l ion years ago , b ut biochemical evidence suggests a much shorter t ime s ince the separat ion, perhaps on ly five m i l l ion years. Very recently the ice ages, which ended 10-15,000 years ago, caused the exti nction of many plants and of the g iant mammals. They also resulted i n great changes in the d istr ibution of s u rviving species. S i nce then the environ ment has been comparatively stable, except where it has been transformed by agricu lture or other human activit ies.

Fossil Microorganisms

The creatu res of the Cambrian period were the oldest to leave foss i ls only because they were t he fi rst to form hard shells. These fossi ls represent compl icated , m u lt ice l lu lar organisms which must have had a long evolutionary history. Fortu nately, foss i ls of m icroorgan isms f rom earl ier t imes have survived. The examination of Pre-Cambrian microfossi ls began early in th is century, but has on ly recently begun to


produce detai led i nformation about these earliest forms of l i fe. I n part, this rapi d i ncrease i n our knowledge is due to the use of the electron microscope.

One of the most com pletely studied groups of PreCambrian microfossi ls is that from the Gunfl int formation located on the north shore of Lake Mich igan in Canada.

Figure 2.7. Representative organisms from the Gunfl int Chert. (Reproduced with permission from J. W. Schopf, Bioi. Rev., 45, 319, (1970).)


Although the rocks in which these fossi ls are found are about 1 .8 b i l l ion years old, many of the foss i ls resem ble very c losely species of b lue-green algae which are sti l l l iv ing today.

Foss i l remains from earl ier periods are few in n u m ber and less wel l-preserved. This i s because extremely old, unmod ified sed imentary rocks are rare. In the course of t ime most of the oldest sedi m ents have spent periods i n the inter ior of the earth where they have been heated to such an extent that no fossi l s have survived. Very few sed i mentary rocks more than 3 b i l l ion years old are known. The s impl icity of the structures of the oldest microfossi ls is another sou rce of d ifficu lty, s ince it is sometimes impossible to dist inguish fossi ls from in organ ic artifacts.

These problems have led to occasional controversy amongst experts. Nonetheless, it i s generally accepted that microfossi ls resembl ing contemporary b lue-green algae are present in rocks th ree b i l l ion years old or somewhat older. The microfossi ls i n the Fig Tree series from Africa {3. 1 b i l l ion years o ld) are perhaps the best preserved. It is also c lear that the earl iest Pre-Cam b rian fossi ls are the s implest, and that microorganisms of increasing complex ity evolved throughout the later Pre-cambrian. Thus the c lassical foss i l record despite i ts enormous variety, corresponds on ly to the most recent 20% of a conti nuous record covering more than 3 b i l l ion years.

We have emphasized that Pre-Cambrian fossi ls and modern algae look very much al i ke. How do we know that the biochemistry of Pre-Cam brian organisms was s im i lar to modern biochem istry? A new and rapi dly developing branch of palaeontoloy which deals with chemica l foss i ls attempts to answer th is question . Chemical fossi ls are organic substances that are found in fossi l-bearing rocks and that derive from the or ig inal o rganisms. Several i mportant biochem icals have been found in association with fossi l a lgae in very old sed i ments. Unfortunately much of the evi dence is sti l l very co ntroversial , so wh i le on the whole it su pports the idea that biochem istry has changed l i tt le since the early PreCambrian, it would be wise to reserve judgment u nt i l more evi dence i s avai lable.


Figure 2.8. A simple bacterium-l ike particle about 3 b i l l ion years old from the Pre-Cambrian of South Africa (Fig Tree Series). Notice how difficult it is to be sure that a particle of this kind is really the fossi l of a l iving organism. (Reproduced with permission from E. S. Barghoorn and J. W. Schopf, Science, 152, 758 (1966).)

The evidence that we have d iscussed shows that i t is very probable that si ng le-celled organisms resembl ing modern b lue-green algae were already present on the earth 3.0 to 3.5 b i l l ion years ago. It does not i n any way prove that this was the ti me when they first appeared. The earth is about 4.5 b i l l ion years old and we know virtually noth ing about the first part of its h istory. The most that we can say is that l ife appeared at some ti me duri ng the first b i l l ion or so years after the formation of the earth.

In order to develop a fee l ing for the vast ti mes that have passed s ince the beg inn ing of l ife on the earth, it is helpfu l to th i n k in terms of the number of elapsed generations. There have been about 1 00 hu man generat ions s ince the flowering of Greek civ i l ization and 500-1 ,000 since the last ice age. About 50,000 generations have gone by s ince the orig in of man. By contrast, much more than 10 m i l l ion generations separate us from the earl iest mammals. The l i neage of modern insects is· even longer, runn ing to hundreds of m il l ions of generations.

Liv ing organ isms s imi lar to bacteria and algae h ave existed on the earth for three b i l l ion years or more. We do not k now how long the most primiti ve organisms took to divide; si n ce modern bacteria u nder ideal condit ions take about 20 m inutes to reproduce, let us assume an average generatio nt ime of two or three hours. Modern bacteria are then calculated to be the product of some 1 0,000,000,000,000 generations of evo lut ion. Even if our assumptions are i ncorrect, th is esti mate co uld hardly be off by more than a factor of ten. It is astonish ing that , as we shal l see, much of the chemical organization of cells is so stable that it seems to h ave remained unchanged throughout this enormous number of generations.

Questions Posed by the Fossil Record

The most obvious questions raised by the fossi l record concern the nature of the evolutionary processes that led to the formation of new species. How d id complex organisms


evolve from the very simple algae and bacteria that l ived more than three b i l l ion years ago ? What condit ions favored the evol ut ion of the mammals? Which creatures are the d i rect ancestors of man? The foss i l record also raises a second set of questions concerned with the or igins of l i fe itself. What happened in the fi rst bi l l i on years of earth h istory that led to the appearance of organisms simi lar to modern bacteria? Was there one origin of l ife, or were there many?

Most elementary introductions to biology deal with the origin of species. They show that the theory of natural selection is adequate to explain the evol ution of com plex organisms from simpler ones. This subject is not of primary interest to us, since it deals with a very late stage in the h istory of l ife. We shal l concentrate on the evoluti on of the most pri m itive cel ls from the inorganic constituents of the primitive earth. Natu ral selection wi l l have to be discussed extensively, but for the most part only in this special ized context. The reader interested in the more conventional appl ications of the theory should consult the books cited in the b ib l iog raphy for an account of natural selection and the or ig in of species.

Living o rganisms represent the ult imate in m iniatu rization; the machinery of l ife is constructed on the atomic scale. The si mplest l iving things are incredib ly compl i cated; they are so m uch more compl icated than anything in the nonliving world that even the largest modern industrial complexes seem relatively s imple when compared with the smal lest l iving cel ls. If we have interpreted the fossi l record correctly, equally minute and co mparably co mpl icated organisms had evolved on the earth three b i l l ion years ago. It is the enormous gap that must be bridged between the most compli cated inorg anic objects and these s implest l iving organisms that provides most of the intel lectua l chal lenge of the problem of the orig ins of l ife.

Before we can beg in to discuss the crucial transition from relative ly si mple inorganic systems to l iving organisms,

. we shall need to describe in some deta il the end produ ct of the or ig ins of l ife -the most pri mitive l iv ing organisms. We know very l i ttle d i rectly about the chemistry of the organisms


that lived on the earth three billion years ago, but we may infer a great deal from the behavior of modern organisms. The next two chapters describe the growth and reproduction of the simplest modern organisms, and in Chapter 5 this information is used to infer as much as possible about the behavior of the corresponding Pre-Cambrian organisms.

32 The Nature of t h e Problem

Molecular Biology

Cells and Proteins

Clearly we cannot hope to u nderstand how l i fe began without knowin g someth ing about the way in which l iving th ings work. In this chapter we shal l su rvey some modern ideas on the structure and fu nction of the simplest l iv ing organisms.

Cells are the basi c un it of l ife. Bacteria, the least com pl i cated of l iv ing organisms, consist of a s ing le cel l ; h igher animals may contain thousands of b i l l ions of cells. Whereas bacteria cells are small and s i mple, animal cel ls are often large and compl icated. Nerve cells, for example, are sometimes many feet long. (See F igure 3.1 ).

We shall see in C hapter 5 that al l modern species are descended from bacteria or algae that l ived on the earth b i l l ions of years ago. In consideri ng the o ri g i ns of l i fe , therefore, it is natu ral to take the bacterial cell as the prototype of

33

/Dendritl"'5 Microvilli • ._ (facinK sinus spa(_£>) Nucleus� ,·� / Nucll'i ·� �U M1tochondna Endopl.asmK

, ,�•:. reticulum� 1r Glycogen .n . I f fY

1 Axon� �-'• 'i I "'-\ ·��M;'"' hond,;J

·,, ·,! Myel;nsheathJ\ ·,,

A. Striated muscle cell '· '•'\.

B. Motor neuron \

� Motor

end plates

C. liver parenchymal cell

D. Kidney cell (prox1mal convoluted tubule)

Ta1l filaments Mitochondrtil Nucleus-�:- . �- � ..,_,.CEtcoodlcz,G o ., co-.,., em 0 •• 'CZir-.

E. Sperm cell

Figure 3.1. Some typical an imal ce l ls. The diagrams are not drawn to scale. They demonstrate the great d ifferences between d ifferent k inds of ce l ls i n complex organisms. Each ce l l type is adapted to its own specific function. (Reproduced with permission from A. G. Loewy and P. Siekevitz, Cell Structure and Function, 2nd ed. , Holt, R inehart and Winston, Inc., New York, New York, 1969).

al l l iving cells. Asking how l ife began is equivalent to aski ng how the f irst u n icel l u lar organisms evolved. We cannot formulate that q uestion correctly unti l we know something about modern un ice l lu lar organisms.

The chemical processes that enable bacteria to g row


and d ivide are so compl icated that they cannot be descri bed in the way that one describes the operation of a s imple machi ne. Instead we must fol low a more roundabout route, f i rst descr ib ing the broad strategy of bacterial g rowth and later coming back to fi l l in some of the detai ls.

The bacterium, Escherichia coli (E. coli), has been studied more extensively than any other microorganism. S ince it is a relatively s imple un icel l u lar organism, it may conveniently be used to i l lustrate most aspects of bacterial growth. E. coli cel ls are rods 2-3 x 1 o-4 em long and 5 x 1 o-s em in d iameter; more than 1 ,000,000,000,000 cel l s cou l d be packed i nto a vol u me of one cubic em.

The bacteri u m is enclosed by a rig id cel l wal l and by a ce l l membrane which is situated j ust with in the wal l . The cel l wal l serves t o protect the membrane from damage. Cel ls g rown in the presence of penic i l l i n lack cel l wal ls and are consequently very fragi le. The role of the membrane is i n part passive; i t acts as a barrier t o keep essential molecules i nside the cel l and to prevent the entrance of harmful molecu les. The cel l membrane also plays a more active part in bacterial g rowth, for i t contains a series of "pumps" which select usefu l molecu les from the outside and concentrate them withi n the cel l . Thus the membrane mai ntains the correct balance between the contents of the cel l and the external environment.

Figure 3.2. An electron micrograph of cell d iv is ion in E. coli. (Reproduced with permission from S. F. Conti and M. E. Gettner, J. Bacteriol. 83, 544

( 1962).)

35 Molecular Biology

The main business of the cel l is carried on within the membrane. Here, a seri es of some hundreds of coordinated chemical reactions take p lace. The so le purpose of a l l this activity i s to use nutrients in the environment to make more E. coli. This task is accompl ished with im pressive effic iency; under ideal conditions a co mplete cycle of rep l ication takes only twenty m inutes. Thus one E. coli cel l could produce 272(1 021 -1 022) descendants in a sing le day.

The indivi dual chem ical reactions that take p lace with in the ce l l are qu ite s imi lar to those carried out in the laboratory. However, in the ce l l they al l take place in aq ueous solut ion at room temperatu re. The organic chemist can make use of speci al solvents and carry out reactions at very h igh or very low tem peratu res. Even so he cannot yet match the synthetic abi l ity of the cel l .

The rem arkable transformations ach ieved within l iv ing cel ls are made possib le by a set of molecu les cal led enzymes. Enzymes are catalysts; that is, they are molecules which speed up chem ical reactions without themse lves being changed in the process. They are able to function repeatedly. So, one enzyme molecule, i f g iven suff ic ient time, can transform many times its own mass of material . The su bstances on which enzymes act are cal led substrates. Each minute, a typica l enzyme transforms a few thousand substrate mo lecules into products.

Enzymes are proteins. The proteins constitute a fam i ly of substances whose molecules are very large and are made up of "bu i ld ing b locks" cal led amino acids. Thus the proteins are polymers ("po ly-" many; " -meros", parts) having am ino acids as the i r monomers (" mono-" sing le ; "-meros", parts). We can compare a protein molecule to a long word made up of a number of letters (the monomers) some of which may be used many times. Just 20 kinds of amino acids a re used in the construction of proteins.* (A few more are formed occasional ly by the modification of the twenty standard amino acids after they have been incorporated into proteins.)

* For a discussion of the optical activity of proteins and nucleic acids see the appendix to Chapter 1 0.


Alanine (Ala)

Votlnt (\/of)

ltoituelne ( Ha)

Strlu l Str )

Tbr•onlne (Thr}

Cysteine ( Cyt)

i l H -c - c

I \H NHz

H 0 I II cH5 -� -\

NH2 OK

CH5 H 0 \ I II c11-c - c I I \

CHs NH2 OH

H 0 I II CH5- CH2 - CH-C - C I I \

CHa NH2 OH

H 0 I II HO - CH2-� -\ NH2 OH

II H 0 I I I cH - c-c - c ' I I \

OH NH2 OH

H 0 I II HS- Cifo- �- \

NH2 OH

H 0 luthlonlno {Mol) I 1/ CH5- s - CH2- CH2 -�- \

NH2 OH

H 0 Lyolno {Lyt) I II

NH2-CHz-CHz-CH2-cH2-�- C\

NH2 OH

H 0 At;lnlno ( Arg l I II NH2 - c -NH-CH -CH -CH - c - C I • • • I \

NH NH2 OH

Jbpartic acid

(Alp}

Alporagint

(Aan)

Glutomle oeld

(Giul

Glutamine

(Gin)

PhtFI)'Ialanine

( P.a)

Tr;-ptophon

{ Tty)

Hlttldlne (Hit)

Proline (Pro}

\ 'i / c-cH2-c-c I I \

HO NH2 OH

0 H 0 \ I II

C-CH - c - c I 2 I \ NH:2 NH2: OH

0 H 0 \ I II 1

c -cH2-cH2 - �- c\

HO NH2 OH

0 H 0 ' I II 1

c - CHz-CHz-�- C\

NH2 NHz OH

H 0 o-CH-�-� I \ NH2 OH

r·

-yH2 l ..... N,..CH -\

I OH H

Figure 3.3. The twenty naturally occurring amino acids. Standard abbreviations are g iven in parentheses.

II II II II II II

I I I I I I - - N-C - C - N -C-C - N - C - C - N - C - C -N - C - C - N -C-C - - -

1 I II I I II I I II I I II I I II I I II H H 0 H H 0 H H 0 H H 0 H H 0 H H 0

Figure 3.4. The standard method of joi n i n g amino acids to form peptides. A can be any one of the twenty side groups shown in Figure 3.3. From Molecular Biology: Genes and the Chemical Control of Living Cells by J. M. Barry, Prentice-Hal l , Inc. , Eng lewood C liffs, N. J . , pp. 3 and 4.

The u n i q ueness of a word or sentence depends on the choice of letters and the seq uence in which they are put together. I n a s imi lar way, the u n i q ueness of the prote i n depends on the c h o i c e o f a m i n o acids and the sequence i n which they are joined. J ust as a single alteration can c hange the meani n g of a sentence, so, chang i n g a single amino acid may a lter the p roperties of a protein. Conseq uently, the health and wel l-bei ng of an organism depen ds on havi ng the correct prote i ns made exactly r ight every ti me. Getting the seq uences exactly r ight is largely the job of another fam i ly of polymers in a cel l , the nuc leic acids - the chemicals of h e redity. We shal l d i scuss the n u c l e i c acids i n the next section.

Protei n s are the most versati le of biological molecu les, for a l l chemical reactions i n cells are catalyzed by protei ns. In addit ion, proteins are essential for the o peration of the pumps in the cel l membrane. Most E. coli strains h ave f ibers cal led f l age l l a attached to their membranes. These fibers, which enable the organ isms to "swi m , " are made of protein.

Lys Val Phe G ly Arg Cys G l u Leu Ala Ala Ala (G iy Leu Ser Tyr G ly Arg Tyr Asn Asp Leu Gly

Asn Trp Val Cys Ala Ala Lys Phe G l u Ser Asn (G iy Tyr Asp Thr Ser G ly Asp Thr Asn Arg Asn

l l u Leu Gin l l u Asn Ser Arg Trp Trp Cys Asn (Cys Pro l l u Asn Cys Leu Asn Arg Ser G ly Pro

Ser Ala Leu Leu Ser Ser Asp l l u Thr Ala Ser (Asn Met G ly Asp G ly Asp Ser Val l lu Lys Lys

Ala Trp Val Trp Arg Asn Arg Cys Lys G l y Thr Len Arg Cys Gly Arg l l u Trp

Met Lys) His Arg Thr G i n) Th r Ala Asp Gly) Thr Arg Val Asn) Ala Cys Asp Val ) Ala G i n-'

Figure 3.5. Amino acid sequence of chicken lysozyme. Note the complexity of a typical protein. For abbreviations see Fig u re 3.3.


In h igher ani mals the contractile parts of muscle are made up of protein and so are i mportant parts of nerve cel ls. A l l dynamic processes in cells are mediated by proteins. (The energy needed to br ing about these processes is provided by sugars, fats, starches, and oi ls in the d iet.)

Each enzyme performs a unique function in the cell . Several h undred enzymes carry out routine production steps ; many more can be b rought into action under specia l c i rcu mstances, for example, when an essential substance in the environment begins to run out and an alternative so urce of it m ust be found. In many cases the activity of an enzyme is regulated by the intracel lu lar envi ronment. In this way the rates of the various processes going on in the cell are adjusted to make g rowth and maintenance as efficient as pos-si ble. As in a modern assembly plant, raw materials a re transformed into finished products by the coordination of large numbers of s imple operations.

Nuc leic Acids and Protein Synthesis

The analogy between a cel l and a facto ry is a usefu l one, but it fai ls to draw attention to the most remarkable property of l iving systems, their abi l ity to reproduce. Since a cel l can d ivide to g ive two identical dau ghter cells, there must be a mechanism for dup licating the cel lu lar mach inery, that is, for making new p rotein molecu les with exactly the same sequ ences as those or iginal ly p resent in the parental cel l . At f irst s ight, it m ight seem that the easiest way to do this would be to copy each ind ividual protein molecu le, letter by letter. In fact, the synthesis of the correct protein molecu les is achieved by a qu ite d ifferent method.

Proteins are not dupl icated d i rectly; instead, the information needed to specify a protein seq uence is encoded in the sequence of a tota l ly d ifferent polymeric molecule, a nucleic acid . The nucleic acids are i m po rtant because they are the only molecules in the cel l that can, with the help of appropr iate enzymes, repl icate d i rectly. As we shal l see, the d i rect repl ication of nucleic acids, that is the process in which a nucleic acid molecu le di rects the synthesis of a new molecule with exactly the same sequence, is the essential process that g uarantees that daughter cel ls resemble the i r


parents so closely. In higher organisms, also, nucleic-acid replication is responsible for all aspects of biological inheritance.

There are two kinds of nucleic acids in every cell deoxyribonucleic acid (DNA} and ribonucleic acid (RNA). Each time a cell divides, the genetic nucleic acid, DNA, is duplicated accurately and one copy is passed on to each daughter cell. The DNA carries all of the information needed to direct the synthesis of new cellular proteins. However, DNA takes no direct part in protein synthesis. Instead, DNA functions as a master-copy from which RNA subcopies are made and these RNA sub-copies are the ones that carry the genetic information to the protein-synthesizing system. It is as though the DNA was a benevolent dictator sending "doubles" to represent itself in situations where it might otherwise be damaged.

The nucleic acids never act as ordinary catalysts. If the proteins are thought of as the machinery of the cell, the genetic nucleic acids must be thought of as the blueprints. There is a division of effort within the cell; proteins are responsible for most cellular activity while nucleic acids make possible the storage and transmission of genetic information.

DNA is a polymer made up of four types of small molecules called deoxynucleotides. To understand how the nucleic acids fulfil their dual function in the cell, we must look rather closely at their chemical structures. Each deoxynucleotide consists of a deoxyribose phosphate molecule attached to one of four bases. It is the nature of the base that distinguishes one deoxynucleotide from another. The deoxynucleotides that make up DNA are usually designated as A,

DNA I I I I I ._ _ _ _ _ - - - - - - -1

r - -----, 1 RNA : L.. - - - - --.J

No direct route

Protein

: I I I

t----- -- - - - -"''

Figure 3.6. The role of R NA as an intermediate in protein synthesis. DNA does not take a d irect part in protein synthesis, although it ultimately controls the nature of the protein synthesized.


Deoxyadenosine Deoxythymidine

.------ NuciJoside -----.,

Purine base = adenine H H....._ / N �N H�Jl.3AH N N

Sugar = deoxyribose

Pyrimidine base = thymine

0 CH3�N,..H (H) )l_3_A H � 0

Deoxyguanosine

I Purine base = guanine

0 :C _..H � 1N H� �N/H I H

Deoxycytidine

I Pyrimidine base = cytosine H /H 'N H�N

Jl.3.Jlo H N

Nucleotide -- -- ' I I

Deoxyadenosine 5' -phosphate Deoxythymidine 5' -phosphate Deoxyguanosine 5' -phosphate Deoxycytidine 5' -phosphate

Figure 3.7. The four deoxynucleotides occurring i n DNA. I n RNA there is an extra OH group of each sugar, and the methyl group of thymine is replaced by hydrogen to give urac i l . These diffe rences are both i nd icated in parentheses i n the f igure. ( Reproduced with pe rmission from J . D. Watson, The Molecular Biology of the Gene, W. A. Benjamin , I nc. , New York, 2nd ed. , 1970.)

T, G, and C,* which are abbreviations for adenylic acid, thymidylic acid, guanylic acid, and cytidylic acid, respectively. Since the deoxynucleotides that make up a nucleic acid are joined together in a regular way, we can represent any DNA molecule as a sequence of these four letters, for example, TCATTGTd We include an arrow to show that the direction of reading is important, just as it is in an English word or sentence.

The four components of RNA are very similar to the deoxynucleotides that made up DNA; they are called ribonucleotides. The sugar deoxyribose is replaced by a closely related sugar, ribose, and one of the four bases, T, is replaced by uracil (U)* ( Figure 3.7). The chains are joined up in the same way in RNA and DNA.

The replication of DNA and RNA or the copying of one from the other depends on a very remarkable structure that can be formed from polynucleotide chains. A pair of polynucleotide chains fits together to form a beautifully regular double-helix, but only if the sequences of ribonucleotides or deoxynucleotides in the two chains are complementary. (See Figure 3.9) It is necessary that, after the direction of one chain has been reversed, the two chains can be lined up so that each A is opposite to a T (A opposite U in RNA) and each G is opposite to a C (see Figure 3. 1 0) . The sequences ACTAAGC and GCT-TAGT match after the second sequence is reversed, while the sequences ACT AAGC and GGTTACT fail to match in the second and sixth positions. The rules, A = T and G = c. that govern the matching of bases are known as the WatsonCrick pairing rules. In the next chapter we shall see how they come about.

The genetic material, DNA, is always stored in cells in the form of a double-helix. When it replicates the two strands of DNA behave independently and each strand directs the synthesis of a complementary strand. In this way two new double helices are formed, each identical with the original doublehelix. (See Figure 3. 1 1 ) Mistakes are very rare, because if the wrong base does go into one of the newly synthesized strands, the regularity of the corresponding double-helix is disturbed. Then DNA replication stops until the incorrect base is removed.

* Adenine (A) and guanine (G) are purine bases. Thymine (T). uracil (U), and cytosine (C) are pyrimidine bases.


5' end

I 0 I O=�-o-H.c

H, � N

o-

0 I O=P-0-H C I •

o-

H0N JL. -� Cytosine

H N 0

Guanine

0 H3c0wH

)L_ -� Thymine

H N 0

0 I

0

Figure 3.8. A section from a DNA chain showing the sequence ACGT. (Reproduced with permission from J_ D. Watson, The Molecular Biology of the Gene, W. A. Benjamin , Inc. , New York, 2nd ed. , 1 970. ) )

(a) Figure 3.9. Two representations of the DNA double-helix. (Reproduced with permission from (a) J. D. Watson and F. H. C. Crick, Nature, 1 71 , 737(1953.))

The mechanism of rep l ication described above is . the on ly one that is important in cells. In certain vi ruses, however, RNA is used instead of DNA as a genetic molecule. The RNA repl icates d i rectly and also fu nctions d i rectly in protein synthesis. It should also be noted that many vi ruses contain sing le-stranded nucleic acids. They repl icate in al most the same way as ind ividual strands of double-helical DNA. (See F igure 3.1 2)

No other polymeric mo lecu les are known which are able


(b) Is reproduced with pe rmission from F. H. C. Crick, in Molecular Basis of Life, W. H. Freeman, San Francisco, 1968.

to combine together to g ive compact two-chain com plementary structures. The Watson-Crick pairin g ru les, which form the basis of the molecular theory of heredity, are , as far as we know, un ique. S ince it seems un l ikely that any repl icati ng structure could b e made u p from amino acids, we can understand why the information needed to determine the sequence of a protein must be encoded in a non-protein polymer, such as a nucleic acid ; there is no structu ral basis for the d i rect letter-by-letter repl ication of protei ns.


C ··· G

G ··· C

A ··· T

A ··· T

T ··· A

C . .. G

A .. · T

(a) (b)

Figure 3.1 0. (a) Two complementary strands of DNA; (b) two DNA strands that fail to match because of GG and CC mispairing.

How Proteins are Made

We have seen that DNA does not act d i rectly i n protei n synthesis. I nstead, one of its strands fu nctions as a template which l i nes u p the components of RNA accord ing to the Watson -Crick pairi n g ru les and thus leads to the synthesis of strands of RNA complementary to the active DNA strand. These newly synthesized RNA molecules are known as messenger RNA, s ince they act as i ntermediaries carryi ng the genetic i nformation stored i n the DNA seq uence to the prote in-synthesizi ng apparatus. The series of operations by means of which the sequence of n u cleotides i n messenger RNA c hains determi nes the sequence of amino acids in protein molecules is ca lled translation. C learly, translation must be qui te compl icated since messages written i n the fourletter alphabet of the n ucleic acids are used to specify the twenty letters which appear in the prote in translations.


Old Old

Old New

Figure 3.1 1 . A diagrammatic representation of the . replication of DNA. The details of the process are not understood at the time of writing. (Repro- · duced with permission from J. D. Watson, Molecolar Biology of the Gene, W. A. Benjamin, Inc. , New York, 2nd ed., 1 970.)

It is perhaps s u rpris ing that the pr inciple i nvolved i n translation i s s imple and fam i l iar. The letters of the n ucleic acid message are always read three at a ti me from a f ixed start ing po int. (See F igure 3.1 3) Each tri p let of nucleotides corresponds either to an amino acid or to a s ignal to stop translat ion. Since there are 64 (4 x 4 x 4) different trip lets and o n ly 20 amino acids, many of the amino acids are represented by two o r more trip lets. I n add ition th ree trip lets corre-


+ + + + +

Replication of

single strand

Repl ication

of double strand

Destruction of

-•• strands

Single +•• strand

Complementary double helix

Two double helices

Figure 3.1 2. The simplest form of repl ication of a single strand RNA or DNA. Most s ingle strand viruses used a more economical method in which several positive strands are made sequential ly on a single negative strand.

+

Two +•• strands

spond to the stop s ign. The detai led rules fo r translati ng nucle ic aci d sequences i nto p rote in sequences have been worked out in the last decade or so. The solution is known as the genetic code. The c racking of the genetic code is one of the triumphs of molecu lar bio logy.

Translation is much more compl icated than repli cation. So, i t i s not surpris ing that the translation apparatus is more compl icated than the repl ication apparatus. More than a hundred protei n and nucle ic acid molecules take part i n t ranslation. Repl ication i s brought about by one, o r at the most a very few, p rotei ns. More detai ls about translation are g iven in Chapter 4.

We are now in a position to see how the parts of the genetic system fit together. Two sets of molecu les, proteins and nucle ic acids, cooperate to al low cel ls to g row and d ivide. Nucleic acids store and transmit genetic i nfo rmation ; p rotei ns do al most everything else in the cel l (except prov ide

l A t

Starting point

Met I Phe I Lys ' Pro ' Phe I I I I I I

u G l U u u: A A Gl C c u: u u C ' · - - ---

Figure 3.1 3. Translation of messenger RNA i nto protein . See Figure 3. 1 4 for the code.


a: w I= w ...J ti a: u:

u

c

A

G

SECOND LETTER

u c A G

uuu Phe

ucu UAU Tyr

UGU Cys uuc ucc UAC UGC

UUA UCA Ser

UAA OCHRE UGA UMBER UUG

Leu UCG UAG AMBER UGG Tryp

cuu ccu CAU His

CGU cue

Leu CCC

Pro CAC CGC

Arg CUA CCA CAA CGA CUG CCG CAG

GluN CGG

AUU] ACU AAU AspN

AGU Ser AUC l ieu ACC

Thr AAC AGC

AUA ACA AAA Lys

AGA Arg AUG Met ACG AAG AGG

GUU GCU GAU Asp

GGU GUC GCC GAC GGC GUA

Val GCA

Ala GAA GGA

Gly

GUG GCG GAG Glu

GGG

Figure 3.14. The genetic code. There are three stop signs that signal the end of a polypeptide chai n - UAA, UAG, UGA. All polypeptides are in i tiated with the AUG codon for meth ionine. The methionine is often removed from the end of the c hain by a special enzyme.

u c A G

u c A G

u c A G

u c A G

-t ;!; :II 0 rm -t -t m :II

energy) . Nucleic acids cou ld not rep l icate or d i rect proteinsynthesis without the help of preformed proteins ; no proteins co u ld be synthesized without the i nformation stored i n preformed nucleic acids. (See F igures 3. 1 5 a n d 3. 1 6). O n e of our major p rob lems i n later chapters is to see how such a "hen and egg" relation cou ld have developed.

Cellular Function

A more detai led account of the way i n which cel ls fu nction must now be presented. For s impl ici ty , bacteria wi l l be d iscussed. Most bacteria derive energy and materia l for g rowth from n utrients in their environment. The membrane contains proteins cal led permeases which select the molecu les the cel l needs from outside and pump them i nto the cel l . I t is


Nucleic acid replication

-- : Nucleic acid

: Protein

Figure 3.15. The logic of ce l l division. All details have been oversimpl ified.

More nucleic acids

} Everything needed to make a cell work

Figure 3.16. � Enzyme functions of proteins; -+H-i# Information-transferring functions of nucle ic acids i n which the sequence of a preformed n ucleic acid determines the sequence of a newly formed nucleic acid or protein .


q uite unclear how many different permeases E. coli can make; the total n umber could be as h igh as one hundred.

E. coli cel ls a re qu ite flexible i n the i r nutritional demands. Many d ifferent organi c compounds are suitable as so urces of food, for example, amino acids, g lycerol , and g lucose. If no suitable nutrient is d i rectly avai lable in the env i ronment, E. coli cells can often prod uce extracel lu lar enzymes which convert complex substances in the envi ronment, for example, starch , into s imple n utrients. Fu rthe rmore , most E. Coli strains are able to detect d istant sources of food and wi l l , u nder appropriate condit ions, move towards them.

Once ins ide the cel l , the raw materials from the environment must be converted into a l l of the smal l molecules requ i red for growth. Clearly, very m any steps are involved in conve rt ing si mple nutrients, such as g l ucose and ammoni a, i nto amino acids, nucleotides, fats, etc. Several h u ndred enzymes that help to synthesize i mportant b iochemi cals have a lready been i dentified ; the total number could exceed one thousand.

Another major cel l u lar activity is the synthesis of polymers {e.g. , proteins, n ucleic acids, starches, etc.). We have al ready seen that the synthesis of proteins requ i res more than a hundred mac romolecules whi le the repl ication of DNA is much s im pler and requ i res very few. These are not the on ly proteins involved i n polymer synthesis. Many scores of

· enzymes are requ i red, for example, to synthesize add i

tional components of cel l walls and ce l l membranes. There i s another group of enzymes with in the cel l that

searches out damaged RNA and protein molecu les and deg rades them so that the i r components can be used agai n. S ince no cel l can reprod uce if i ts genetic material has been destroyed , damaged DNA molecules have to be dealt with in qu ite a d ifferent way; they are repai red rather than degraded. In recent years a g roup of enzymes has been discovered, each of which is able to repai r a particu lar k ind of defect in damaged DNA. Without these repai r activities, ce l ls would often be unab le to d ivide to g ive viable daughter cel ls.

Energy is expended in almost every one of the steps which we have described so far. Different cel ls derive their


energy from d ifferent sources, but a l l of them use a s ingle chemical , ATP, as the pri ncipal i ntermediate i n the uti l izat ion of energy. Photosynthetic bacteria use the energy of sun l ight d i rectly to make ATP, but most other bacteria obtai n the energy needed to make ATP by breaking down nutrients, such as sugars, i nto s impler molecu les. E. coli, for example, can derive energy by convert ing g lucose to carbon d ioxide and water.

Processes of this kind have been described as "contro l led burn ing ," but this ph rase is somewhat deceptive. A typical sugar, such as g l ucose, wi l l burn i n a i r to give carbon dioxide and water. Energy is released in the p rocess as heat (and a l ittle l ight). Although a heat engine cou ld be run on g l ucose, the l iv ing cel l has a different and better way of making use of this and related substances. The cel l employs about twenty enzymes to d ismantle g lucose, step by step. The whole series of reactions is ach ieved mi ld ly at room temperature, and about half the avai lable energy is channel led d i rectly into the synthesis of ATP, without ever appeari ng as heat. Thus, although heat engi nes and cells sometimes bring about the same overal l chemical changes, the oxidation of nutrients by air to carbon dioxide and water, they do it i n d i fferent ways.

Cells produce energy i n many other ways. I n the absence of air, fo r exam ple, they often break down g l ucose to lactic acid . Processes l i ke th is , which do not requ i re ai r , are cal led fermentations; they produce much less ATP for each nutrient molecu le consumed than do oxidations. Other k inds

NH2 I

c N'l- 'C....-N:-. I II "cH

HC.:::::- /C--N1 � 0 0 N

-o- P-o-�-o-�-o-H c� I I I

2

-o -o -o H H H H

OH OH Figure 3.17. The structu re of ATP.


of cel ls are able to derive al l of thei r energy from inorganic sources.

Despite the diversity of energy sou rces used by different k inds of cells, the major energy-uti l iz ing p rocesses are striki ngly s im i lar i n al l cel ls. ATP is used, often ind i rectly, in the synthesis of n ucleic acids and protei ns, in many syntheses of smal l molecules, in the repai r of damaged nucleic acids, in pumping, and so on. Of the major g roups of act ivities wh ich we have discussed so far, only the deg radative reactio ns can conti nue indefi n itely after the supply of ATP is cut off. Cells degenerate without the use of energy, but new synthesis is abso lutely dependent on ATP. In Chapter 4 the structure of ATP and the way in which the energy stored i n ATP is made to do usefu l work are described in g reater detai l.

So far we have dealt with protei ns whi ch catalyze chemical reactions, transport molecu les across membranes, o r act d i rectly i n some other way on the contents of the cel l or on su bstances in its environ ment. A d istinct c lass of proteins is i nvolved in contro l l ing and coord inating al l these activities. The protei ns of th is class are able to use i nfo rmation abo ut the i nternal and external environments of the cel l to contro l the activity of preformed enzymes or the synthesis of new enzymes.

It is worth looking at two typ ical examples. Suppose that cells that have been g rowing on g l ucose are transferred to a medium contain i ng a d ifferent sugar, say lactose. At f i rst no enzymes are p resent that can uti l ize lactose. So, g rowth stops. However, with i n a short t ime new enzymes are synthesized that can act on lactose. Then g rowth resu mes. This adaptation to a new med ium is achieved through the action of a "contro l " p rote in which recognizes that lactose is the best available nutrient in the envi ron ment and then causes the enzymes that act on lactose to be synthes ized.

Suppose next, that after cells have been growi ng fo r some t ime on g l ucose as the sole nutrient, a comp lete m ixture of ami no acids is i n trodu ced i nto the envi ron ment. Clearly it would be u neconomical for the cel l to co ntinue makin g amino ac ids when they a re freely avai lable. Amino acids are, therefore, pum ped in from the outside and the enzymes making amino acids are very qu ick ly shut off. Later a


further economy is achieved by stopping the synthesis of a l l the enzymes requ ired to make am ino acids, s ince these enzymes are not needed as long as the . external supply of amino acids lasts.

It should now be clear that at any g iven t ime the enzymes function ing with in a cell constitute only a fracti on of those which the cel l is able to make. To deal with a chang ing envi ronment, cells must have a reserve capacity to make many proteins that are only occasional ly req u i red.

What is the i nside of a cel l l i ke? We know that water is essential for l ife, but how much of the interior of a cell is left over for water once proteins, nucleic acids, and assorted smal l molecu les have been f itted i n ? In general, cells are f i l led with a concentrated jel ly- l i ke solution of protein. In E. coli, for example, about 80% of the cel l content is water and most of the rest is protei n. Other cel ls are s i m i lar, although the detai led composition of the intracel lu lar med ium varies from organism to o rganism.

Viruses

The s implest cells are self-contained reproduc ing un i ts ; they wi l l g row and d iv ide i n envi ron ments that contain noth ing but su itable n utrients and a few i no rgani c salts. V iruses h ave m uch in common with cells, but they d iffer from them i n one i mportant and clear-cut way. Viruses cannot reproduce without the help of the macromolecu lar machinery i nside the cells of their hosts.

The s implest vi ruses contain a short nucleic acid which codes for a very smal l n u mber of prote ins. Every vi rus is able to code for the one or more p roteins which form a protective coat around its nucleic acid, and most v i ruses also produce an enzyme to repl icate the i r RNA or DNA. Small vi ruses are comp letely dependent on the i r hosts for a su pply of the components needed to b u i ld proteins and nucleic ac ids. More i mportant, they are dependent on the proteinsynthetic apparatus of the i r hosts; they do not carry the complex protei n-synthetic apparatus needed to express the information encoded in their nucleic acid. Larger vi ruses can


carry out some additional functions for themse lves, but they are sti l l dependent on the synthetic capacity of their hosts.

It has sometimes been suggested that the f i rst organ isms were viruses rather than cel ls. This suggestion is based on the mistaken notion that vi ruses are autonomous self-repl icating u nits. In fact, s ince viruses cannot repl icate without the he lp of a host cel l , the evolution of vi ruses cou l d not have occurred before t h e appearance o f cel lu lar forms of l i fe. From this point of view, at least, v i ruses shou ld not be regarded as l ivi ng organ isms.

Molecular Genetics

The genetic materia l of E. coli is physical ly contin uous. It forms a double-helical DNA molecule conta in ing about 4,000,000 base pairs. However, the gen etic materia l is functional ly d iscont inuous in the sense that i t contro ls the synthesis o f some thousands o f d iscrete k inds of protei n molecules. The sequence of DNA respons ib le for the synthesis of a s ingle prote in is cal led a gene. A typical p rote in might contain 200 amino acids ; the corresponding gene wou ld contain 600 base pairs. Genes are arran ged in l inear sequence along the DNA molecule, separated by an e laborate scheme of punctuation marks. The punctuation marks are themselves particular nucleotide sequences, but special ones that are recognized by the translation apparatus to mark the beginn ings or endings of standard messages.

Early in the development of m icrobia l genetics, the relation between the genetic material and the functional e lements of cells was summed u p aptly in the p h rase "one gene- one enzyme . " Since then we have learned a g reat dea l about the d etai led operation of the genetic system, but this general ization remains a val uable one. We analyze the relation between the fu nctional materials and the genetic materials of any organism in terms of the proteins and their corresponding DNA sequences -the enzymes and the genes that specify them.

Our knowledge of genes rests on very strong physical evidence. Genes are no longer mysterious abstractions i n-


Figure 3.1 8. The DNA released from a damaged bacteriophage. DNA in E. coli is about ten ti mes longer. (Reproduced with permission from A. K. Kleinschmidt, Biochim. Biophys. Acta, 61 , 857(1 962).)


voked to explain compl icated facts about in heritance i n p lants a n d animals. Genes can b e seen i n the e lectron m icroscope and their lengths can be measured. The order of hundreds of genes on E. coli DNA is known with certainty. A very s imple gene has been synthesized i n the laboratory. There are sti l l some d ifficu lt p roblems i n bacterial genetics, but the nature of the gene and of its relation to cel l u lar p roteins is understood , at least in b road out l ine . The organization of the genetic material i n h igher organisms, that is, the structu re of chromosomes, is u nderstood less wel l .

The description of the genetic system that we have given so far contai ns an important supposition : DNA is rep l icated without error. Recent studies with E. coli show that in fact errors do occur with a f requency of perhaps one in 1 07 ; that is, on the average, one error occurs for every 1 0,000,000 bases copied. Since E. coli DNA contains about 4,000,000 letters, there is almost a one-in-two chance of a m istake occurring in each rou n d of rep l ication . In the course of many m i l l ions of generations, the accumu lation of mistakes has i mportant consequences.

Mistakes i n DNA repl ication are cal led mutations. The most com mon mutations are substitutions of one nuc leotide for another or additions and deletions of a s ingle nucleotide. More compl icated and extensive errors occur, but less frequently. The usual consequence of a mutation is the production of a prote in with a modified amino acid sequence. Three typical mutations, a substitution , an i nsert ion and a delet ion, are i l l ustrated i n F igure 3.1 9.

Strangely, perhaps , i t is erro rs i n the synthesis of DNA which permit bacterial evo lution. I f no e rrors occurred, each m icroorganism wou l d p roduce descendants with an unchanged DNA seq uence, and hence, with an unchanged

AGAT Substitution

- - - - - - - - ACAT ~ ACGAT Insertion

AAT Deletion

t Figure 3.19. Some typical m utatio ns.


genetic potential ; fam i l ies derived from a s ingle ancestor would remain exactly l i ke that ancestor for generation after generation. The occu rrence of occasional errors in the rep l i cation of DNA i s the on ly source of variety i n those bacteria for which sexual processes are un important. Of course, i n h igher animals the reassortment of genetic material associated with sexual reproduction provides a second, and often more i mportant, source of variation.

Most mutations are harmful or, at best, neutral. A mutated form of an enzyme is usual ly i n active or less eff ic ient than the orig inal form ; a mutated control p rotein usually switches on or off at the wrong times. Occasional ly, mutat ions are advantageous if they p roduce modified proteins better able to carry out an old function , or i f they p roduce a protein able to perform a new and usefu l fu nction.

I t is interest ing to compare m utati ons with errors made in sett ing the type for a book. Most ran dom errors lead to nonsense or, what i s worse, to a deceptive new sense. Rarely, a new and qu ite unanticipated mean i n g may result f rom an error. The change from "the comforts of i mmortality" to "the comforts of i mmoral ity" in a rel ig ious tract is a stri k ing example of a deletion.

One should note that, even with in a s ing le cel l, the effects of a mutation may be out of all p roport ion to the physical change that occ u rs in the DNA A mutation i nvolvi ng a s ingle base, i f it leads to the prod uction of an inactive variant of an essentia l enzyme, can k i l l a whole cel l . There i s no relation between the phys ical extent of the change in the DNA and the im portance of the change for the cel l . Return ing to the analogy with p rinti ng, i t is as though some mutations deleted the words "do not" from an essential section of an instruction manual, whi le others merely shortened the recommendations on the dust cover.

Many mutations can be transmitted unchanged through a large number of generations. Once an error has occu rred, it may be copied unchanged for m i l l ions of generations. Thus, a s i ng le mistake in DNA repl ication can affect the performance, not only of an i ndividual bacteri um, but of a whole species. These arg uments apply in much the same way to h igher organisms. S ickle-cel l anemia, for example, i s a not


unco mmon d isease i n which the oxygen-carryi ng capacity of the b lood i s i m paired. A l l carriers of the d i

.sease are de

scended from one o r a very few ancestors i n whom the mutation occu rred. The degree of ampl if ication involved in an instance l i ke th is is hard to match in the nonbiolog i ca l world. A s ing le event, at the molecu lar level , has affected the l ives of a very large numbe r of i n div iduals.

We can now see why m utations, almost all of which are i nd ividual ly harmful , permit a species to become better adapted to its env i ronment. Organisms that i nherit the d isadvantageous mutations are sooner o r later .eli minated by the competition of their f itter neighbors. On the other hand, organ isms that i nhe rit one of the very few benefic ia l m utations are l ikely to g row at the expense of the i r neighbors and wil l p robab ly d isplace them. This is part of the molecular i n terpretation o f Darwin 's theory of natu ral select ion. The biologica l world is dominated by the l ineages that invent usefu l adaptations; fai l u res are soon forgotten.


Topics in Biochemistry

Introduction

This chapter brings together a nu mber of un related topics in chemistry and molecu lar b iology that wi l l be usefu l i n later chapters. Only the section on base-pai r ing is essential for an understanding of the general argument of the book, but the other sections should prove usefu l to most readers. Those who have d ifficu lty with chemistry are advised to read qu ickly through al l the material except that on base-pai r ing. They may f ind it usefu l to return to other sections when they are referred to later in the text.

As we al l know, chemistry deals with the properties of atoms and molecules. An atom consists of a central positively charged nucleus surrou nded by a cloud of negatively charged electrons just sufficient in number to cancel out the positive charge of the nucleus. The carbon atom, for example, consists of a nucleus of charge +6 surrounded by a c loud of 6 electrons.

61

When atoms are brought together, thei r electronic clouds become deformed. I n many cases the deformation is of such a type that the atoms attract each other strongly and combine to form molecu les. However, when the nuclei get too close, the repulsive force between the i r positive charges always overcomes any attractive force due to the deformation of the electron c louds. Thus, the nuclei in molecu les do not approach each other i ndefin itely closely but settle down at a distance known as the interatomic distance ( but l iteral ly the i ntern uclear distance). The d istance between neighboring nuclei i n a molecule is characteristically about 1 o-s em or 1 A.. A molecular chain 1 em long contains about 1 0S atoms.

For many pu rposes molecules can be treated as sol id objects, each with a shape and size determined by the ru les of valency and stereochemistry. The chemical ru les of va-

Figure 4.1. Molecu lar models of some typical molecu les : (a) water (H20), (b) benzene (C6 H6), (c) adenine (C7N7H7) .


OH I

c H-'j "'H H Figure 4.2. Tetrahedral arrange ment of four groups around carbon i n methanol (methyl alcohol).

Ieney specify the numbers of bonds that can be formed by each type of atom. Additional rules describe the stereochemistry of atoms, that is, the way in which the bonds formed by an atom are arranged in space. A carbon atom, for example, in one of its characteristic envi ronments, is surrounded by four groups at the apexes of a regu lar tetrahedron. Recently far subtler ru les have been discovered that th row additional l ig ht on the spatial arrangements of atoms in complex molecu les, such as proteins.

We do not need to understand these ru les in detai l . For the purposes of this book it is sufficient to real ize that in almost all molecu les the different groups of atoms are arranged in a wel l-defined structu re. Each group of atoms has certain properties that, to a fi rst approximation , are independent of i ts position in the molecule, particu larly if the d ifferent active groups are not too close together. A knowledge of the properties and positions of the groups of atoms in a molecu le is a l l that is needed to expla in many molecu lar properties. We shal l f i rst i l lustrate th is point by describ ing some propert ies of an important group of molecu les cal led surface-active agents.

Surface-Active Agents and Colloids

It is possib le to classify the atoms or groups of atoms that make up organ ic molecules i nto those that tend to enhance and those that tend to d imin ish the solubi l ity of a substance in water (Table 4.1 ) . These groups are called hydroph i l ic and hydrophobic, because they tend, respectively, to seek

63 Topics in Biochemistry

Hydrophobic G roups

I -CH-

Table 4.1

Hydrophi l ic Groups

-OH -NH2

out or to avoid water. Most sugars, for example, are very soluble in water because thei r molecu les contain many hydroph i l ic OH g roups, whi le most petroleum products are i nsolub le because their molecu les are almost exclusively made up o f hydrophobic CH, CH2 , an d CH3 g roups.

When substances contain ing many hydroph i l ic g roups dissolve in water, they form t rue solutions i n _which the dissolved molecu les are isolated from each other by a layer of water. Substances contain ing hydrophobic g roups are usual ly i nsoluble in water because water molecu les tend to stay together and squeeze out hydrophobic i ntruders, whi le hydrophobic molecu les tend to remain together and avoid water. * The reasons why water molecu les and hydrophobic molecu les do not mix can be understood i n terms of physical and chemical theories, but the explanations are too diff icult to be g iven here.

Molecu les that have a hydroph i l ic head and a hydrophobic tai l have qu i te exceptional so lub i l i ty properties, because the heads tend to dissolve in water whi le the tai ls have a tendency to stay i n contact with each other. Such molecu les are cal led surface-active agents. They inc lude soaps and detergents.

Surface-active agents often accumulate at boundaries, for examp le, at the interface between oil and water. In this way both ends of the molecu les of the surface-active agent can be accom modated in favorable envi ron ments : the head

• Many molecules are insoluble for different reasons.


remains in the aqueous phase whi le the tai l d issolves i n the o i l . The i mportance of detergents as c leans ing and emulsifying agents depends on their abi l ity to bridge the inte rface between a hyd rophobic substance and water.

When surface-active agents are shaken with water, they often aggregate to fo rm associations of molecu les in which the hydrophobic tai ls are packed together i nside whi le the hydroph i l ic heads stick out. In th is way the hydrophobic parts of the molecu le are able to rem ai n together wh i le al lowi ng the hydroph i l i c parts to establ ish contact with the aq ueous env iro nment.

If a sufficient nu mber of molecules assoc iate, we refer to the agg regate as a col loidal particle, and we refer to the mixture of col loidal particles with water as a col lo idal d ispe rsion (solution) or s imply as a col loid. A co l lo id is a m ixtu re which resembles a solution i n that it contai ns more or less uniform particles d ispersed in a solvent in such a way that they do not settle out. However, the solute is present in aggregates much larger than the s ing le organi c molecules present i n a s imple solution. Col loidal d ispe rs ions have a number of specia l properties d ifferent from those of s imple solutions, for example, they scatter l ight in a characteristic way. The part icles in a col loidal dispersion are smal ler than those present in suspensions or preci pi tates.

Su rface-active agents are only one of a large class of molecules that form co l loidal d ispersions. The Russian scientist A. I . Oparin has emphasized the impo rtance of certai n

f t r 1 r 1 t r · t 1 1 1 1 1 1

(a) (b) (c)

Figure 4.3. Some typical arrangements of surface-active molecu les, • hyd ro ph i l ic head , hydrophobic tai l : (a) bridge at a n o i l-water boundary, ( b) a collo idal droplet, (c) a b i layer membrane.


col loidal d ispersions, which he refers to as coacervates, i n the orig i n of l ife. Both h igh mo lecu lar-weight polymers and simple surface-active agents cou ld have contri buted to the formation of these agg regates.

Another important structu re formed by su rface-active agents is the b imolecu lar leaflet. Here sheets of molecu les are brought together with their hydrophobic tai ls i nside and the hyd roph i l ic heads in contact with water. Artif ic ial b i layers of th is k ind form membranes that are impermeable to e lectrically charged atoms and molecu les. S imi lar b i layers are thought to be im portant components of bio logical membranes.

The Structure of Biological Polymers

(a) Base-Pairing and the Structure of N ucleic Acids. The nucleic acids are the molecu les responsi ble for all forms of in heritance. Without them, the evolution of l iv ing organisms s imi lar to those that we know would have been impossi b le. Al most a l l of the functions of n ucleic acids depend on basepai ri ng ; one cannot appreciate the problem of the orig ins of l ife without understand ing this one aspect of structural chemistry.

Fo r nucleic acids to perform their genetic function, it is clearly necessary that one strand of a DNA dou ble-hel ix be able to d i rect the synthesis of the complementary strand. However, it is equal ly important that the two strands should come apart at the r ight t ime. Otherwise, it wou ld be impossib le for further DNA repl ication to occur, or fo r DNA to d i rect the synthesis of messenger RNA. It fol lows that the two strands of. a double-hel ix must be held together by forces that are strong enough to guarantee some stabi l ity, but not so strong as to prevent separation at an appropriate moment.

Ordinary chemical bonds within mo lecu les are very strong. A double-hel ix in which the strands were joi ned by stab le chemical bonds cou ld not be separated qu ickly enough to fu lfi l a genetic fu nction. On the other hand, the forces between molecu les are usual ly so weak and nond i rec-


tional that they wou ld be unable to hold a double-he l ix together i n a wel l-defi ned structu re. DNA is he ld together by special i ntermo lecu lar bonds of i ntermediate strength, known as hydrogen bonds. Each n ucleotide in a base-pai r forms two or th ree such bonds, in such a way that the two members of the pai r are held i n exactly the correct relative configu ration.

The structu res of the base-pai rs are i l l ustrated i n F igure 4.4. To defi ne the structu res adeq uately, two or th ree hydrogen bonds are req u i red in each base-pai r. Analogy with a two or three prong electric p lug is he lpfu l here. If on ly one prong is inserted the p lug can sti l l rotate in the socket, but once two or th ree prongs are inserted, the whole structu re is held rig id ly.

The base-pairs have a remarkab le property - they are geometrically equivalent. In Figu re 4.5 the th ick l i nes repre-

H . H \ I

H-C O. H \ t . ·.. I C-C '·H-N N H H-d N-H . \ - / ""'C.-/ \ I '•, c r. I

N-C ··-... 11 � -N I \ N\ p \

o c=N T

H\

H N-H-. \ I ·· ..

I H A

c-c ··-0 H-d 'N

\ /N""'c.....--H \ 1 ··... c-c 1 N-c ·-.. 1 � -N I \ · H-N\ p \

o.. c=N ··... I C ··H-N I G H

Figure 4.4. The A-T and G-C base pai rs. Note the equ ivalent positions of the bonds jo in ing the bases to the backbone (th ick l i nes).


sent the bonds by which the bases are attached to the backbone of the double-helix. The hydrogen bonds hold the base-pairs in such a way that these two bonds are the same distance apart and have the same orientat ion whether we are deal i ng with an AT, TA, GC, or CG base-pair. It is th is property of geo metrical equ ivalence that permits the construction of a regu lar dou ble-hel ix from fou r different com ponents.

We can now see why base-pai ri ng perm its accurate repl ication to occu r. The enzyme responsib le for jo in ing up the backbone of a new strand of DNA wi l l fu nction only if al l new un its are presented to it i n exactly the same orientat ion. The preformed DNA strand, act ing as a tem plate, presents a

(a) (b)

Figure 4.5. Base-pair equ ivalence and DNA re pl ication. For convenie nce of i l l ustration the doublehelix has been unwound. The enzyme will accept a new base (dotted l i ne) only i n the configuration shown in (a). If m ispai ri ng occurs as in (b), the enzyme recogn izes that the orientation is wrong and rejects the incorrect base. (The ce l l has a second l ine of defense. If a wrong base sl ips through, it is usually cut out again.)


nucleotide monomer in the same orientation whether it is T, C, A, or G provided the new un it is base-pai red correctly- th is is g uaranteed by the property of geometrical equ ivalence. However, if the wrong base attaches itself to the template DNA, it wi l l be presented to the enzyme in a noneq uivalent orientation and consequently be rejected. In this way errors of repl ication are avoided.

I t cannot be emphasized too strongly that the abi l ity to form geometrically equ ivalent base-pai rs is a property of the bases themselves, which does not depend on the presence of enzymes or other biological polymers. A forms a specific base-pai r with U , and G with C, when the components are dissolved in a sim ple organic so lvent, such as chloroform, for example. As we shall see, it has been possi ble to make use of the equ ivalence of the base pai rs to imitate some important aspects of n ucleic acid repl ication in total ly nonbiological systems. It seems l i kely that a re lated nonenzymatic repl ication of nucleic acids was the fi rst "genetic" process to occur on the pr imitive earth. The evolution of l ife as we know it was probably dependent on the geometrical equ ivalence of the Watson-Crick base pai rs.

(b) Proteins. A protein consists of one or more po lypeptide chai ns. The seq uence of amino acids in a protein is said to define its pri mary structu re. However, one can learn very l ittle about the properties of a prote in from the pri mary structu re alone. Usually it is necessary to determine the distribution of the various parts of a protei n in space, the secondary and tertiary structu res, to understand how a protein functions.

In recent years, X-ray crystal lography has been appl ied successfu l ly to the study of proteins. The stru ctu re of a typ ical prote in is i l l ustrated i n F igure 4.6. C learly, on ly a part of such a large and compl icated molecule can be di rectly i nvolved i n the chemical modif ication of a smal l substrate. This reg ion is cal led the active site.

The active site must be confined to a reg ion of space that is not much larger than the part of the substrate that is to be acted on. However, the amino acids that make up the active site are usual ly widely separated in the pri mary struc-


Figure 4.6. (a) Model of Myoglobin showing every atom. The course of the backbone is marked with a white cord. (Reprodu ced with permission from John Kendrew.)

ture of the protein . The polypeptide backbone is i ntricately folded to form a structure which places the amino acids of the active site i n exactly the correct positions.

Enzymes are not on ly efficient but also h igh ly selective ; they act upon their substrates but often have no effect at a l l on c losely related molecu les. The selectivity of an enzyme is contro l led by its specificity site. The g roup of amino aci ds that make up th is site is usual ly d ist inct from those in the active site. They are posit ioned in space i n such a way as to d i rect the appro priate part of the substrate to the active site


(b) Space-f i l l ing model of the same p rotein. This incredibly complex structure is repeated exactly in every myoglobin molecule. ( Reproduced with perm ission from H. C. Watson.)

but to prevent the approach of molecu les other than the su bstrate.

The extraordinary efficiency with which enzymes catalyze chemical reactions is sti l l not u nderstood comp letely, but we do know that structu ral factors are very i mportant. The molecu les synthesized by organic chemists are usual ly qu ite smal l , and their functional g roups tend to point outwards from a central reg ion. On the other hand, p roteins are much larger than the i r substrates and are often folded so that the am ino acids of the active site point i nwards from a cavity around the substrate. This permits enzymes to get a very much fi rmer gr ip on the i r substrates than can be ach ieved by smal l-molecule catalysts.

The amino acids in the active site of an enzyme are essential for its function ; any change i n these amino acids usual ly resu lts i n the inactivation of the enzyme. Changes i n


other parts of an enzyme wi l l often modify its catalyt ic properties, but wi l l rarely cause complete inactivation. The evolut ion of enzymes in modern organisms is thoug ht, i n most cases, to be associated with changes in the specificity site or in other reg ions d isti nct from the active site.

Changes in the specifi city site may either i ncrease or decrease the range of substrates on which an enzyme can act. Microorganisms can adapt to use a new sugar as a source of energy, for example, by chang ing the specif ici ty site of a pre-existing enzyme i n such a way as to accept the new sugar as a substrate. On the other hand, they can become resistant to certain d rugs by modifying thei r enzymes in such a way that they no longer act upon the d rug. Changes of this type have been important i n biochemical evo lution.

Changes i n the backbone of a protei n i nactivate an enzyme completely on ly if they drastical ly alter the active site. Changes on the outside of a protein , far from the active site, are usual ly without effect on enzyme activity or affect the activity only s l ightly. In the cou rse of evolut ion, it is probab ly this type of change that has been responsible for those smal l modifications of protein structure that adapt an enzyme to work under s l ightly changed conditions. The adaptation of an enzyme to work at somewhat h igher or lower temperatures, for example, is bel ieved to occur by the accumulation of changes i n the primary sequence outside and possi bly far away from the active site.

Synthetic catalyst Protein

(� (b)

Figure 4.7. (a) A synthetic catalyst with fu nctional grou ps pointing out. (b) A much larger protein catalyst with fu nctional groups that can encircle a su bstrate.


Chemical Free Energy

Most fami l iar chemical reactions are spontaneous. They are also irreversib le in the sense that they do not proceed spontaneously in the opposite d i rection. A piece of i ron rusts spontaneously, but rust never spontaneously reverts to metal l i c i ron ; coal bu rns to carbon d ioxide and water, but a mixtu re of carbon d ioxide and water has no spontaneous tendency to form organic material agai n. Quantitative studies of systems of this k ind have contributed to the development of a branch of chemistry cal led chemical thermodynamics. The notion of eq u i l ibri u m plays a central part in this subject.

A closed chemical system is said to have reached equil ibr ium if it does not tend to undergo fu rther chemical change. I ron exposed to dam p air, for exam ple, does not reach eq u i l i br ium u nti l it is complete ly converted to rust. Our definit ion impl ies that any closed chemical system wou ld reach equ i l ibri u m if left long enough. It continues to react so long as it has any tendency to react; when it has no further tendency to react it is, by defin ition , in equ i l ibr ium.

This definit ion of eq ui l ibri u m contai ns no reference to ti me. I n some cases the approach to eq u i l ibrium is so slow that it cannot be studied i n the laboratory. Nonetheless , it is always possi ble, in pr incip le, to calculate the properties that a system wou ld have i n equ i l ibri um , if it had time to get there. Alternatively, special methods can be used to accelerate the approach to equi l ibr ium. A mixtu re of hyd rogen and oxygen reacts very s lowly at room temperature, but it equ i l i brates explosively i f exposed to a sma l l spark. Enzymes and other catalysts accelerate the approach to equ i l i br ium, but i n a less dramatic way.

The natu re of equ i l i br ium i n a chemical system is made clearer by consideri ng the behavior of more fami l iar mechanical systems. Water flows spontaneously downh i l l , but never flows spontaneously uph i l l . The state i n which no further motion occurs because the water has reached the lowest level attainable corresponds to the equ i l ibr ium state i n a chemical system. An isolated system contain ing parts that move relative to each other sooner or later comes to rest under the influence of frict ion; however, systems never


speed up under the action of friction (unless external forces are applied). In this case the "equilibrium" state is the one in which there is no tendency for the relative velocities to change - in a system subject to frictional forces it is the state in which the parts are at rest relative to each other.

These ideas lead us naturally to the idea of free energy. Man has learned to exploit many of the spontaneous processes that occur in his environment to help him to do work. Water flowing downhill is used to turn paddlewheels; the motion of the air is harnessed and made to push sail boats; fuel is burned and the heat that is evolved is used to drive engines. We use the term free energy to describe the part of the energy released in spontaneous processes that can, in principle, be used to do work.

The free energy is the upper limit of the amount of work that can be derived from a given spontaneous process. It is possible to build inefficient machines that waste most of the available energy, for example, by converting it to frictional heat, but no machine can be constructed that does more work than that which corresponds to the available free energy. Real machines are always somewhat inefficient owing to friction, etc., but it is usual ly possible to calculate how much work could be done by a hypothetical, perfectly efficient, frictionless machine.

When we do work on a mechanical system we make use of the free energy released in one process as it takes place spontaneously to drive another process away from mechanical equilibrium. In typical cases we might use water power to raise a weight or e lectrical power to accelerate the air in a wind tunnel . In a l l such cases we couple together two systems so that the free energy that is released as one moves spontaneously towards equilibrium drives the other one away from equilibrium.

Now let us return to chemical systems and compare them with the mechanical systems discussed above. Work can be derived from chemical systems if they are capable of reacting spontaneously, that is if they are not already in equil ibrium. The chemical free energy corresponding to any given reaction is the maximum amount of work that can be obtained from the reaction as it proceeds spontaneously to


equi l ib rium. Just as with mechanical systems, the max imum work that can be real ized f rom a chemical reaction can usual ly be cal cu lated, even though it can rarely if ever be real ized.

The f ree energy that is ava i lab le f rom spontaneous chemical processes can be used in a variety of ways. Hydrocarbons can be burned and the heat that is released at h igh temperatures can be used to d rive a steam engine, for example. In b iological systems chemical reactions are almost always coupled together d i rectly so that the f ree energy released i n one reaction as it moves towards equ i l ibr ium drives another reaction away from equ i l ib rium. We shal l see that in cells part of the f ree energy avai lable f rom the oxidation of g lucose is used to synthesize ATP. The free energy that cou ld be derived from the hydro lysis of ATP is then uti l ized in tu rn to bring about the synthesis of p roteins, nucleic acids, and other molecu les.

Heat engines are a lmost always a good deal less than 1 00% efficient, so that the work done is always less than the free energy that is avai lable. The same is true of b iochemical "engines" ; on ly about half of the energy avai lab le f rom the oxidation of g lucose, for example, is converted to ATP. Nonetheless, the enzyme-catalyzed energy conversions that occur i n cel ls are sti l l a good deal more efficient than are s imi lar p rocesses in non livi ng systems.

Free Energy in Biological Systems

We have seen that a l l closed chemical systems u lt imately run downhi l l to equ i l ib rium. Livi ng systems somehow escape th is fate. This once led some authors to bel ieve that l iv ing systems are g u ided by some "vital force" that enables them to evade the laws of thermodynamics. The situat ion is now real ized to be much s imp ler. Livi ng organisms are not closed systems, so the laws of thermodynamics in their standard form do not requ i re that they run down h i l l to equ i l ib ri u m. The fol lowing analogy may be usefu l . An ai rplane with a fixed supply of fuel must u lt i mately make contact with the


earth, but an airplane that is refueled in mid-air can fly forever (leaving aside wear due to friction).

A population of bacteria such as E. coli can div.ide forever, so long as i t i s constantly "refueled" by glucose or some other nutrient in the environment. The bacterium avoids running downhill to equil ibrium as long as it can derive energy from an external supply of organic compounds. There i s no "thermodynamic mystery" about the survival and reproduction of simple living organisms as long as they are adequately nourished.

However, if we consider all of the interdependent organisms that make up the biosphere, we do have to explain how they get their energy. Living organisms, l ike other chemical systems, are constantly dissipating free energy. If this free energy could not be replaced, all l ife would inevitably come to an end. There is only one important process that restores free energy to the biosphere-photosynthesis. Insofar as energy is concerned, the biosphere is living on foreign aid; photosynthesis is the only exchange mechanism that permits light emitted by the sun to contri bute to the maintenance and evolution of life on earth.

A single substance, ATP, is i nvolved in almost all energy transactions within the cell-pursuing the analogy with economics, we may regard ATP as the universal currency of biological energy exchange. Different organisms derive their

NH2 I

c N7 'C--N I II \'cH

HC.::::- /C-..fi N

0 0 0 II A II B II

HO- P- 0- P-0- P-0-H2C I I I

HO HO HO

0

OH OH

Figure 4.8. The structure of ATP. The bonds marked A and B are those which are split when ATP is used to provide energy for biochemical reactions.


energy from d ifferent sou rces, but a l l of them convert it to ATP as a prel i m inary to ut i l iz ing it for g rowth and function.

Among the many biochemical reactions that generate ATP, those that proceed in the absence of oxygen are most interesting to us, si nce they could have occu rred on the primitive earth. These reactions are often referred to as fermentations. In fermentations, the atoms that make up organic molecu les, such as g lucose, are rearranged in a compl icated way to form more stable compounds; energy that would be released as heat during the rearrangement, if it occu rred spontaneously i n a nonbiological system , is used to form ATP by l iv ing organisms.

Higher ani mals use a qu ite d ifferent mechanism, oxidative phosphorylation, to obtai n most of the energy they need. The oxidation of g lucose to carbon d ioxide by oxygen from the air is coupled to the synthesis of ATP. Much more energy is released by the oxidation of g lucose than by fermentation ; so, much more ATP is formed for each molecule of g lucose used up. However, oxidative phosphorylation must be a re latively recent biological in novatio n , s ince, as we shal l see later, there was no oxygen i n the pri m itive atmosphere.

The free energy content of ATP is stored i n the bonds that jo in together the th ree phosphorus atoms. These bonds are, therefore, cal led h igh-energy phosphate bonds. When ATP is left in contact with water, the h igh-energy bonds are broken and the free energy stored i n them is wasted. In the presence of su itab le enzymes, however, these bonds are not hydrolyzed d i rectly; i nstead , ATP undergoes a series of co upled reactions that lead to changes usefu l to the cel l . The final products obtained from ATP are the same as those obtained by d i rect hydrolysis, but part of the energy which would be released as heat i n d i rect hyd rolysis is made to do usefu l work in biological systems.

Free Energy and the Synthesis of

Proteins and Nucleic Acids

We have seen that the survival of l iv ing organisms depends on their abi l ity to use external sources of energy to produce ATP. The steps by which energy is used to synthesize ATP


are poorly understood . We are better informed about the methods that are used t� make ATP do usefu l work, for example, in br ing ing about the synthesis of b io log ical polymers. The fol lowi ng treatment of this compl icated subject is much overs impl ified .

In Chapter 9 we shal l see that b io log ical polymers are never in eq u i l i b rium with water; an aqueo us so lution of a protei n , for example, tends to decompose into amino acids. It fol lows that protein synthesis, because it converts amino acids to proteins, is an uph i l l process and must be d riven by the deco mposition of ATP. How are two such d ifferent reactions as the down h i l l hydrolysis of ATP and the uph i l l synthesis of proteins coup led together?

The general pr inciple i nvolved is fair ly straightfo rward . If ATP is to i nfluence the behavior of amino acids, it must fi rst react with them. Furthermore , during the reaction as much as possible of the free energy or igi nal ly stored in the ATP m ust be conserved , for otherwise no energy would be left to d rive the synthesis of proteins uph i l l . Stated in another way, ATP must fi rst react with the ami no acids to form h ighenergy i ntermediates that preserve most of the free energy released by the decomposition of the ATP.

The necessary reactions are catalysed by a set of enzymes (activat ing enzymes) that are un iversal ly d istri buted in l iv ing organ isms.

0 0 � -� � H2N-CHR-C-OH + ATP ___.____. H2N-CHR-C �PA + P2

Amino acid

(not activated)

(energy

source) Aminoacyl adenylate Pyrophosphate

(activated)

0 II nH2N-CHR-C�PA

Aminoacyl adenylate lmany steps

0 0 � II

H2N-CHR-C-NH-CHR-C··· · · + nAMP Peptide-n-un its Adenylic acid

Net Resu lt: nATP + nAmino Acid --+ nPeptide + nAMP + nPyrophosphate


Once the activated i ntermediates are formed they cou ld , in principle, react with one another d i rectly to form the peptide bonds of proteins. The energy released by the breakdown of the intermediates is sufficient to d rive the synthesis of proteins u ph i l l . In p ractice, the mechan ism of protein synthesis is more complex (see below), but the general princ ip les involved are si m i lar.

It is thought that most, if not a l l , b iological condensation reactions proceed by somewhat sim i lar mechanisms. We know less about prebiotic condensation reactions. It is un l ikely that ATP was the fi rst prebiotic sou rce of energy, but somewhat sim i lar molecu les may have been avai lable on the prim itive earth. In any case, it is almost certain that many prebiotic condensation reactions involved activated intermediates of one k ind or another (Chapter 9).

Photosynthesis

An i mportant consequence of photosynthesis is the formation of organic materials at the expense of sun l ight and carbon d ioxide in the atmosphere. In plant photosynthesis, water is the source of the hydrogen needed to reduce carbon dioxide to sugars ; oxygen is released i nto the atmosphere.

6C02 + 6H20 ____. (CH20)6 + 602 Glucose

Carbon d ioxide can also be reduced by microorganisms at the expense of other substances such as hydrogen su lfide. Since these alternative reduc ing agents may have been abundant on the primitive earth, they may have been s ign if icant for the evolution of photosynthesis.

Plant photosynthesis provides the organic compounds that other organisms use, d i rectly or i nd i rectly, as sou rces of structu ral material and energy. The h igher ani mals, for example, recover the energy stored by plants when they oxid ize the prod ucts of photosynthesis back to carbon d ioxide, using atmospheric oxygen. The oxygen itse lf has, of course, been generated by photosynthesis.


C6 H1 206 + 02 Sugar Oxygen

Plant "Animal" photosynthesis (;�oo

) C02 H20 Energy + Energy

Carbon Water dioxide

Figure 4.9. The biological carbon cycle.

Protein Synthesis

Protein synthesis is a complex process in which amino acids are joi ned in specific sequen ces acco rd ing to instructions encoded in a stretch of DNA. The nucleotides making up the DNA are read in g roups of th ree f rom a fixed starti ng poi nt. The amino acid assigned to any g roup of three nuc leotide bases is specified by the genetic code (F igure 3. 1 4) .

Each seg ment of a DNA doub le-hel ix consists of two comp lementary strands. Of these, on ly one is used to specify the seq uence of amino acids in p roteins. We shal l refe r to this strand as the positive strand. We have seen that the DNA does not take part d i rectly in protein synthesis, but is represented by messenger RNA. Clearly, the messenger RNA m ust have the same sequence as the positive strand of DNA, and hence it must be constructed using the negative strand of DNA as template.

The assembly of proteins takes place in or on a ribosome, a complex structu re composed of protei ns and nuc le ic acids. Messenger RNA and "prepared " amino acids associate with the ri bosome- unchanged messenger RNA and proteins are released.

A remarkable feature of protein synthesis is the " preparation" that precedes that presentation of amino acids to the ribosome. Amino acids do not enter the ribosome in the free form, nor as any s imple, activated derivatives. I nstead , each am ino acid is f i rst attached to its own special type of RNA molecu le . These RNA molecules are cal led tRNA's (short for transfer RNA's) ; the tRNA to which g lycine becomes at-

SO The Nature of the Problem

+ + +

C ··· G G . .. c--i I I

G ··· C C ··· G -{ I I

A ··· T T ··· A --l I I I A ··· T T ··· A ---1 I I T ·· · A

I A ··· U � I I I

C ··· G G · · · C � I I I A ··· T T · · · A --I

I I I I

(a) (b)

Figure 4.10. Solid l ines represent DNA strands; dotted l ine represents messenger RNA: (a) rest ing double-strand of DNA; (b) negative DNA strand d i rects synthesis of positive RNA strand.

tached is written as tRNAg1y. and s im i larly with the other amino acids .

. The attachment of amino ac ids to the i r tRNA's thus constitutes a prel im inary sorti ng of amino acids - g lycine is attached to tRNAg1y, alan ine to tRNAa1a . and so on. This sorti ng is achieved with the help of twenty activati ng enzymes, each of which recognizes both a tRNA and the co rresponding amino acid . The process of attach ing an amino acid to its tRNA is referred to as load ing .

The tRNA's have another remarkable property : they form compl i cated th ree-d i mensional structu res often referred to as cloverleaf structu res, i n which specia l seq uences of three residues are exposed in j ust the right way to form base-pai rs with three nucleotides i n messenger RNA. The choice of these th ree exposed bases on the tRNA's is a key aspect of protein synthesis. Let us see how this works out i n a s imple case.


CD � !. ?!:. '::, 'Jt.'" � \ I lf lll <f

C> III U I I lf lli U l> ll l d, I I C> I ll (.J I I t � ('\-�-- ® e> .�O"j:.' �...- ,, D, ® 1:: �-,;-�-;r \o ·� Q' • - - - - Q� rf G-G-C-G-C ;:)

o-o-o-:>-n.... �e 'G', 1 = = = = ::f U-A -r:.

'.., _.,C-C-G-G-A, I - u -u ? 111 9

C ll l ::'l I I <jl ll l �'l lf ll l <f C ll ll!l I I Cjl ll l l!l Cjl l1 1 � 5' end

3' end t

�}0 t

Activated {:t:�-z alanyl I :t:

residue £ "' w Figure 4.11 . (a) The structure of alani ne tRNA d rawn in the conventional cloverleaf form. ( Reprod uced with permission from Holley et al., Science, 1 47, 1 462 (1 965).)

We know that glyc ine is incorporated i nto a prote in sequence whenever it is signal led by the seq uence GGG in a messenger RNA. This translation is made possib le by a tRNAaiy. i n which the exposed sequen ce is CCC. The tRNA.g1y is f i rst loaded by its activating enzyme; next the exposed CCC sequence of the tRNA.g1y associates t ightly with the GGG sequence i n the m essenger RNA by Watson -Crick basepair ing. This g uarantees that a g lycine is p laced i n the right


T !J! C arm

(b)

DHU arm

oo .A.

0 75 A

0 50 A

0 25 A

O A

(b) One of the proposed ful ly folded structures of tRNA. Note how arms two and three are folded around the waist of the molecu le . (Reproduced with permission from A. L. Leh ninge r, in Biochemistry, Worth Publ ishers, Inc., New York, 1970.)

position to be i ncorporated i nto the protei n opposite the GGG sequence i n messenger RNA.

This example i l lustrates the general princ ip le. Each amino acid can be loaded onto one or more tRNA's. These tRNA's have exposed sequences of bases that pai r with exactly those triplets of bases i n the messenger that code for the amino acid i n question. The exposed trip let of bases on a tRNA is called an anticodon, because it base-pairs specifi-

83 Topics In Biochemistry

cal ly with the cod ing tr ip let of bases (codon) on the messenger RNA. I n the s implest cases on ly Watson-Crick base-pai rs are i nvolved, so that the anticodon is j ust the sequence complementary to the tri p let that codes for the appropriate amino acid . Tryptophan, for example, is specified by the tri p let UGG, so the exposed anticodon of tRNA1ry is CCA. (Note the reversal of the order of bases. This is necessary because the chains joi ned by base-pai rs run in opposite d i rections. )

To see how a peptide beg inn ing with the sequence metgly-try . . . could be formed, consider what happens when a messenger RNA that begins with AUG GGG UGG associates with the ri bosome. Two loaded tRNA's carrying methionine (anticodon CAU) and g lyc ine (anticodon CCC) attach themselves to the two f i rst tri plets of the message. A pepti de bond

mRNA 51-end

l �� A � u __,...... c, G

� g i\..Sl_ gly. m et. G e li() u

��c{W : ��,.y I C.

3'-end Figure 4.1 2. A diagrammatic representation of the formation of a polypeptide which begins with the sequence met-gly-tryp . . . . The fi rst tRNA which carried the methionine has fulfi l led its function and is shown fal l ing off the messen ger. The th i rd amino acid (tryp) is just taking its place on the messenger. In the next stage met-gly wi l l be transferred to the loaded tryp tRNA.


is then formed leaving a tRNAg1y, loaded with met-gly, on the ribosome; at the same time the un loaded tRNAmet fal ls off. Next a loaded tRNA1ry (anticodon CCA) associates with the ri bosome and attaches itself to the UGG sequence; this t ime the peptide met-gly is transferred , leavi ng the peptide metgly-tryp attached to tRNA1ry and free ing the tRNAg1y. A long protein chain can be bu i lt up in a seq uence of sim ple steps of this k ind.

The tRNA's are often referred to as adaptors, because they make it possib le for messenger RNA to order the amino acids i n a prote in , without ever coming into d i rect contact with them . This is an extremely i m portant poi nt. Messenger RNA's are the kind of molecu le that can associate specifically with other RNA molecu les by base-pai ri ng , but they cannot d iscrim i nate d i rectly between amino acids. The use of tRNA's al lows this d ifficu lt d iscri m ination to be carried out by a protein . The messenger RNA has on ly to order the correct tRNA's, for everyth ing else is done by the activat ing enzymes.

At an early stage in the evol ution of l ife, before the development of protein synthesis, no activati ng enzymes with highly specific catalytic activity could have existed. How cou ld polynucleotides have d i rected the synthesis of polypeptides, before the evolution of the activat ing enzymes ? This is one of the great puzzles of evolutionary biology. We shal l return to it i n Chapter 1 0.


The Bio,hemi,al Re,ord

Paleonto logists, by studyin g the foss i l record, have traced back the h istory of l ife some three b i l l ion years. U nfortunately, no corresponding biochemical record is avai lable. Whi le foss i ls often survive for a very long t ime, most o rganic chemicals i n the fossi ls are destroyed q u ickly by heat o r moisture. Thus, although th e study of the organic substances associated with fossi ls is p rogressing, it is un l i kely to tel l us much about the very earl iest phases i n the development of l ife.

We are forced to conclude, therefore, that the on ly evidence that the b io logical sciences are l i kely to provide about biochemical evo l ution must come from the examination of species that are sti l l al ive today. The task of reconstructing the h istory o f the origins o f l ife from the biochemistry o f l iving species can be compared with that of establ ish ing the

87

history of technology by examin ing mach ines constructed since 1 970. How, i n the latter case, could one te l l whether the use of the wheel and the lever were contemporary innovations, or whether one was i ntrod uced into a techno logy that already made use of the other? No doubt, there are questions about the or ig ins of l ife which wi l l turn out to be of this ki nd ; that is, questions that are unanswe rable u n less we can f ind the historical evi dence. Fortunately, some of the questions that are most interesting to us can be answered in other ways.

Valuable i nformation about the early stages i n biochemical evol ut ion can be obtained by compari ng the organic constituents of d ifferent modern organisms. If we f ind the same organic compound in members of two l iv ing speci es, we have to decide between two alternatives. The co mmon constituent may occu r i n the two species because the abi l ity to make it has been i nherited from a common ancestor, or the steps needed to synthesize the su bstance may have developed independently in the two species i n response to some co mmon need. We refer to these two processes as d ivergent and convergent evolution , respectively. Whi le it is someti mes d ifficu lt to decide between these alternatives, in the most important instances the cho ice is obvious.

If , for example, the synthesis of some compl icated organic chem ica l evolved independently in two species, it is un l ikely that the two methods used to make it would be identical. Convergent evolution, therefore, can explain the s im i larities between related chemical pathways in different species only if the degree of s im i larity between the pathways can reasonably be attri buted to c hance. If not, we must assume that d ivergent evo lution has occurred. Com pare the prob lem of decid ing between convergent and divergent evolutio n with that of discoverin g whether o r not the work of two i nventors is i ndependent. Clearly, if the i nventions are very s imi lar and a whole paragraph appears in identical form i n their patent appl ications, we are j ustified i n supposing that one is a plagiarist (or both, i f they have a common "ancesto r"). The identity of passages in the patents cannot be attri buted to chance and, therefore, proves that the i nventors did not "converge" on the same solution to their prob lem.


Very compl icated features of biochemistry which are identical i n sufficient detai l in two species m ust, therefore, have been i nherited from a common ancestor. It fol lows that if certai n com plex features of biochem istry are common to a l l forms of l ife, then there must have been a common ancestor from which all l iving things are descended.

We have a lready seen that the genetic apparatus is a very complex but un iversal feature of a l l l iv ing ce l ls. The components of DNA and RNA, the 20 amino acids, and the genetic code are the same in all organisms. It is inconceivable that two systems that evolved independently could resemb le one another so closely, so we may safely concl ude that there was a common ancestor of a l l l iving things.* We shal l refer to this organism as our last com mon ancestor.

Our conclusion that there must have been a un ique, compl icated ancestor of all l iv i ng thi ngs poses a new problem. Clearly our last common ancestor must have been the product of a long evol utionary process. Why do we not see evidence of other forms of l ife that evolved in paral le l with it? The nature of the problem and an outl ine of an answer are ind icated in F igure 5.1 .

In the course of ti me, any "species" gives rise through mutation and natu ral selection to new and com peting "species ." Some of these u lt imately become extinct, but the successfu l ones go on in their tu rn to produce even more varied forms of l ife. The natu ral conseq uence of evo lution is the production of d i fferent "species," each adapted to its own special environ ment. What happened to the organisms which were competing with our last common an cestor?

To understand the complete dominance of the descendants of any organ ism, we m ust su ppose that it acqu i red so considerable an advantage over a l l com petitors that it was able to e l im i nate them. It is un l i kely that we shal l ever know the nature of the " i nvention" that made this poss ib le. It cou ld , for example, have been an extrace l lu lar enzyme able to attack a l l competitors, or a more efficient method of producing ATP. Possibly it involved the poo l ing of " i nven-

• Strictly, we can deduce only that al l modern organisms are derived from a small inbreeding ancestral popu lation.

89 The Biochemical Record

-- ----- ---:, - - ... _ - --... ... _ _ __ _ ... ... ... --- ! ------ - - -- -

Modern species

e = Common ancestor

• = Last common ancestor

- Denotes extinction of "species"

Figure 5.1. Ou r last common ancestor. It is assumed arbitrari ly that there were three independent "fami l ies" of organisms. Note that several organisms coexisted with out last common ancestor but were f inal ly e l iminated by species that diversified from ou r last common ancestor. (The dotted l ines ind icate that the three fami l ies under consideration cou ld have had a common orig in even fu rther back in the past.)

tions" between two organisms in some form of genetic exchange, rather than a change with i n a s ingle organism. I n any case the biochemistry o f the organism i n which the crucial advance was made u lti mately became the universal form of biochemistry.

If we accept these arguments, we must conclude that


comparative biochemistry alone can te l l us l ittle about the or ig ins of l ife. In the absence of addit ional ev idence or some more subt le theoretical arguments, there is no way of learni n g about any organ ism more pri m itive than our last common ancestor. Al l traces of other forms of l i fe that must have coexisted with it and d iffered s ign if icantly from it have disappeared from the biochemi cal record.

We have at last come face-to-face with the i rreducib le problem of the orig ins of l ife. How d id an organism that already incorporated al l of the u n iversal features of contemporary biochemistry develop from the inorganic co nstituents on the su rface of the pri mitive earth?

I f no trace remai ns of pri mitive forms of l ife dating back to the period before the development of our last common ancestor, what other evidence i s avai lable to throw l ig ht on the earl ier phases of biochemical evol utio n ? The only way i n which we can derive usefu l i nformation wou ld seem to be by studying the organ ization of biochemical reactions in modern cel ls. What we need is a speculative reconstruct ion of the sequence of steps i nvolved in the evo lution of biochemistry and in particular in the evolution of the genetic apparatus.

This rather difficu lt idea is best i l lustrated by a s i mple example. I n a l l present-day organisms the genetic apparatus is made up of p roteins and nucleic acids. What reason is there to bel ieve that the fi rst genetic apparatus conta ined nucleic acids? In Chapters 3 and 1 0, it is emphasized that proteins cannot repl icate and, hence, that a genetic system cannot be constru cted from proteins a lone. Thus, we shal l see that, even in the absence of d i rect foss i l evidence, we can be a lmost certai n that the first organ isms contai ned nucleic aci ds or c losely related polymers.

We should be clear, however, that the study of the o ri g ins of l ife, because i t is a h istorical study, i s attended b y d ifficu lties that do not occur in most other b ranches of science. We are searching for a h istorical reconstruction that must be i mag inative, but not too i mag inative. What constraints are needed to guarantee that specu lation is kept with in reasonable bounds?

The answer to this question is quite straightforward.


Each step in any proposed history of the evolution of life must be compatible with all relevant scientific knowledge. Chemistry, geology, and astronomy all provide information about the primitive earth and the processes that could have occurred on it. Finding a theory that is plausi ble in the l ight of all the relevant information will certainly be very difficult. It seems unlikely that we shall ever f ind two radically different theories, both of which are satisfactory.

Any complete theory of the origins of life must specify a series of reactions that could have occurred under prebiotic conditions and it must be adequate to explain the evolution of our last common ancestor. A theory gains in status with each new demonstration that a proposed step can be simulated plausibly in the laboratory. Since our last common ancestor was composed of organic molecules similar to those occurring in modern organisms, it is first necessary to show how these substances were formed on the primitive earth. Then we must explain how a primitive organism evolved from a mixture of suitable organic molecules.

Unfortunately, the conditions that existed on the primitive earth are very different from those that are usually used by organic chemists. Although many of the constituents of cel ls can be synthesized in the laboratory, a few of the most important of them cannot yet be made under prebiotic conditions. It is even harder to join the simple organic molecules together to form polymers similar to proteins and nucleic acids. Consequently, a great deal of work will have to be done before we can propose a single complete theory of the origins of l ife and show by experiment that each step could have occurred on the primitive earth.

Our Program

It is important at this point to distinguish two aspects of the problem of the origins of life. In the next section of this book we shall be concerned with the origins of life on earth. This is a very difficult subject, but the difficulties are not of a philosophical nature. We have to explain the evolution of cells from inorganic matter; we do not have to define "li fe" in an


abstract way, neither do we have to decide at what point the word " l ivi ng" could fi rst be appl ied to the evolving system of organic molecules on the pri mitive earth.

I n Chapters 1 3-1 6 we shal l specu late more generally about l ife in the Universe. Then we shall have to consider questions that have a phi losoph ical or at least a semantic aspect. What is l ife? Need it be based on the chemistry of carbon ? Need it be based on "chemistry" at a l l ? It is fortunate that we do not have to answer these more elusive questions before we can get started on the top ic that i nterests us most.

For the moment we have only to explain the evolution of our last com mon ancestor. Clearly, l i ke a modern cel l , our last common ancestor was a com pl icated chemical facto ry enclosed i n a l i p id mem brane. Al l i m portant chemical reactions were catalyzed by proteins made of the twenty standard amino acids. Protein synthesis was d i rected by nucleic ac ids ind isti nguishable from those in modern cel ls. Even the complex system of enzymes and nucleic acids that is responsible for modern protein synthesis had ach ieved its f inal fo rm.

The synthetic abi l ity of our last common ancestor was equal ly impressive. Mechanisms had al ready evolved that perm itted the synthesis of the standard amino ac ids, sugars, and nuc leotide bases from very s imple starti ng materials. I n addition, the pathways leading to most other i mportant biochemicals were avai lable. ATP could be generated by the decomposition of sugars and used to support many kinds of synthetic activity. Recent evidence suggests that the sequences of amino acids in many of the enzymes present i n o u r last common ancestor were very s im i lar to those o f the corresponding enzymes of modern organ isms.

It wi l l be convenient to d ivide our d iscussion of the orig ins of l ife into th ree sections, corresponding to th ree major phases in the evolution of our last common ancestor. F i rst we shal l describe what is known about the pri mitive earth and about the prebiotic synthesis of organ ic material on its surface. We shal l be most interested in the co m ponents of the genetic apparatus, the amino aci ds , sugars , and nucleotide bases. This aspect of prebiotic chemistry is d iscussed in


Chapters 6-8. Next, i n Chapter 9, we go on to consider how the small molecules, once they h ad accumulated in sufficient quantities, were jo ined together to form random nonbiologi cal polymers. Final ly, in Chapters 1 0 and 1 1 , we come to the most difficult but also the most i nteresti ng part of the problem: how did a h igh ly o rganized cell evolve from a mixture of random o rganic polymers ?

Appendix to Chapter 5 - Panspermia

The accou nt of the origins of l i fe which we have outl ined is the conventional one. However, a qu ite d i ffe rent theory was popu lar du ring parts of the n ineteenth centu ry. In Chapter 1 we d iscussed briefly a theory called Panspe rmia, according to which l ife did not evolve from inorganic matte r on earth, but reached us fu lly developed in the form of a bacterial spore that had escaped from a distant planet. The theory, as orig inal ly proposed, had mystic overtones, but the idea that l ife evolved elsewhere i n the un ive rse and then i nfected the earth is not in itself u n reasonable. I do not th ink that any theory of th is general kind is l i kely to be correct, but s ince suc h theories have often been d ismissed too dogmatically, they are worth discussing here.

It seems certain that no spores cou ld survive the jou rney from another solar system to the earth. I t is easy to calcu late the amount of radiation that a spore wou ld re ceive du ring the journey, and this is many orders of magnitude greater than that needed to ki l l a terrestrial spore. More sign ificantly, the amount of radiation received would seriously disru pt any organized material made u p of carbon, hydrogen, n itrogen, and oxygen. Thus, the theory of Panspe rmia i n its strictest form cannot be correct. However, the related theory, that the f i rst cel l arrived on the earth within a meteorite, is harder to d isprove.

The frequency with which meteorites reach the earth from distant parts of the galaxy cou ld be calculated if we knew how often chu nks of matter escape from planetary systems. Unfortunately this i nformation is not avai lable. Estimates that have been made recently suggest that escape is extremely improbable and hence that very few meteorites make the journey. The d ifficulty is that one might have been e nough, and no one can prove that in the f irst b i l l ion years of the earth's history no single object f ro m outer space penetrated our atmosphere.

If a l iving cel l escaped from another planet within a meteorite , there is no reason to doubt that it cou ld have survived the journey


to the earth. It would have been protected from radiation by the sol id material of the meteorite, whi le the low temperature of outer space wou ld have prevented any spontaneous chemical deterioration. Chapter 1 6 presents arguments which suggest that there may be a class of planets with evolutionary histories s imi lar to that of the earth. An organ ism from such a planet might well have found the earth a hospitable place.

The assumption that l ife was brought to the earth from another planet would deserve more attention if it made it easier to understand how l ife evolved, or if it explained some surprising facts about l ife on earth. Is this the case? Clearly if we believe that l ife began on another planet, we are justified in consideri ng a wider range of prebiotic envi ronments than could have existed on the earth. It could be imagined, for example, that l ife began in an ocean of l iquid ammonia, since it is qu ite possible that there are planets which have oceans of this kind. It turns out that this additional freedom in choosing a prebiotic environment would not be very helpfu l. The conditions that we customarily assume to have existed on the prim itive earth are as favorable for the orig ins of l ife as any that can reasonably be postulated for another planet.

The situation is rather different when we come to consider the evolution of biological order. Many molecu lar biologists are unconvinced by the arguments that we have given to explai n why biochemistry is so un iform. They find the universality of the genetic code particu larly surpris ing, since it is hard to see why competing organisms with sl ightly different codes could not coexist. If we could show that l ife evolved on another planet, this difficu lty would be removed.

Suppose that l ife on some other planet had evolved to produce organisms much more diverse than those on earth. On such a planet the genetic code might wel l have been different in different species. However, if a meteorite carried a single organism to the as yet l ifeless earth, that organism would have taken over completely before any competitors could appear. Thus, all terrestrial organisms would have inherited the particu lar code used by thei r unique extraterrestrial ancestor. I n biological terms all terrestrial l ife would consist of a single fami ly derived from a more diverse popu lation on another planet.

I think it more l ikely that a un ique code evolved on the earth than that l ife reached us from another planet. However, the latter point of view is an interesting one and should be entertained by students of the orig ins of l ife, at least on sleepless n ights. Could life have been planted on the earth, deliberately, by the members of a technological society on another planet?


PART

TWO

History of the Earth,

Atmosphere, and Oceans

I ntroduction

A relative ly smal l num ber of sim ple organ ic molecu les are invo lved in the fundamental b iochemical processes of protein synthesis and nucleic acid rep l ication. The same twenty amino acids occur i n the prote ins of a l l o rganisms. The nuc leic acids iso lated from cel ls a lways conta in the same sugars, ribose and deoxyribose, and the same set of nucleotide bases. I f the earl iest organisms were recogn izably related to contemporary organ isms, they too must have made use of some of these compounds. The demonstration that many biological ly important compounds can be formed from inorganic material under conditions si m i lar to those that must have prevai led on the primitive earth is perhaps the m ost i mportant advance we have made so far in our u nderstand ing of the or igins of l ife. Before we can beg i n to describe this work we must outl ine what is known about the

99

early history of the earth, for it was in the atmosphere , oceans, and lakes of the prim itive earth that the chemical constituents of the fi rst l iving organ isms were made.

Origin of the Solar System

It was once wi dely bel ieved that the solar system was formed when another star passed very c lose to the su n ; the pl anets were thought to have been to rn fro m the sun by the very large g ravitational f ield of the nearby star. This theory is no longer ten able. Calculations show that it is necessary to assume an extremely c lose and therefore extremely i mprobable near-co ll is ion to account for the formation of objects as large as the planets. Furthermore, even if objects that large had been formed, they wou ld have had so much i nternal energy that they wo uld have disi nteg rated.

Al l modern theories of the formation of the solar system are based on the Kant-Laplace hypothesis , according to which the sun and planets are derived from a vast cloud of dust and gases. Many such dust clouds have been observed in our own galaxy. A dust cloud, as it rotates, f i rst f latte ns i nto a d isc . lf the gravitational f ie ld is large enou gh, material then beg ins to accumu late at the center of the dust cloud. I n t ime, as more and more material condenses, a central star i s formed.

However, the accum ulation of material at the center of the dust cloud does not conti nue i ndef in itely. The material i n the dust cloud i s rotating around its center; consequently, as the cloud col lapses the rate of rotation must increase. If al l the material fo rmed a s ingle star, the star would spin so qu ickly that its outer layers would be thrown off again. Instead most of the rotational motion in the dust cloud is transferred to a relatively smal l amo u nt of matter which su bsequently moves in orbit around the central star.

It is bel ieved that the material in the solar system which escaped the in it ial condensation and remained i n regions far f rom the sun took part in a series of second ary condensations. So l id objects were formed and g radually swept up al l the dust particles i n concentric r ings around the sun. These

1 00 Steps toward a Solution

Figure 6.1 . An imaginative i l lustration of the later stages i n the condensation of the solar system according to the Laplace hypothesis. (Reproduced with permission from Stars, Planets and Life by Robert Jastrow, Wi l l iam Heineman n, Ltd., London, 1 967.)

objects became the planets. The Kant-Laplace hypothesis explains in a simple way why the planets rotate about the sun in a single plane and why the plane of rotation of the sun coincides with the plane of rotation of the planets around the sun. Once the dust cloud had flattened into a disc, all further rotational motion had to take place within the plane of the disc.

The composition of the planets must have been determined by the density, composition, and temperature of the dust cloud in the regions where they condensed. There is little doubt that the original dust cloud contained a large excess of hydrogen and helium, since these elements are so much more abundant in the universe than are any of the heavier elements. The major planets, Jupiter and Saturn, do indeed contain a great excess of hydrogen and helium, but the composition of the earth is very different. The earth contains large amounts of heavier elements, such as oxygen and iron, and very little hydrogen and helium. Why did hydrogen and helium escape from the earth?

A gas can be retained in the atmosphere of a planet only if the gravitational field of the planet is sufficient to prevent molecules of the gas from escaping into space. Three important factors determine which molecules can escape and which are trapped in the atmosphere. The heavier the planet, the larger the gravitational field. Hence, it is harder for atoms or molecules to escape from larger planets. The colder the atmosphere, the slower the motion of individual atoms or molecules. Hence, it is harder for atoms or molecules to escape from a cold atmosphere. Finally, the heavier the atoms or molecules, the more strongly they are held back by the gravitational field. Hence, light atoms and molecules escape more easily than do heavy ones.

A complicated mathematical theory has been developed to make these ideas more quantitative. It is found that the major planets, Jupiter and Saturn. are large and cold enough to retain all substances in their atmospheres, including hydrogen and helium. On the other hand, Mars and the earth, because they are much smaller and hotter, lose the lightest atoms, hydrogen and helium, although they can retain molecules such as nitrogen, oxygen, and water. The moon, which is much smaller again, cannot retain any atmosphere.

102 Steps toward a Solution

Figure 6.2 The Solar System. (Reproduced with permission from Stars, Planets and Life by Robert Jastrow, Wil l iam Heinemann, Ltd. , London, 1 967.)

History of the Earth

Two methods have been used to estimate the age of the earth. One of them depends on the assumption that the p lanets and the meteorites were formed at abo ut the same time. The ages of a number of meteorites have been determ i ned by rad ioisotope methods. There is general agreement that the meteorites were formed about 4.5 b i l l ion (4.5 x 1 09) years ago. The second estimate of the age of the earth depends on measu rements of lead isotope ratios in various terrestrial rocks. The detai led arg u ments are co mpl i cated, but the conclusion is quite straightforward. The lead isotope ratios also point to an age of about 4.5 b i l l i on years for the earth . In view of the c lose agreement between these two q u ite i ndependent estimates, an age of 4.5 b i l l ion years can be accepted with a good deal of confidence.

The temperature of the surface of the pri mitive earth is a lso of g reat concern to us, but in this case we are less certain about the facts. The widespread bel ief that the earth was once completely molten is based on the theory that the planets were torn from the sun in a near-co l l ision with anothe r star. Si nce the col l ision theory has been abandoned, there is no longer a val i d reason for bel ievi n g that the surface of the earth was ever very hot. It has been c laimed that the presence in the earth's crust of a nu mber of substances which wou l d h ave been d riven out by temperatu res above 300°C proves that the surface of the earth was never very hot. This arg u ment is n ot u n i versally accepted , and some geophysicists believe that the s u rface of the earth was melted by the heat released with i n the earth soon after its formation.

Whi le this controversy remains to be resolved, there are good reasons for bel ieving that vo lcan i c activity on the earth m ust once have been a good deal more extensive than it is now. Vo lcan ic action at present is sustained enti rely by heat generated by the decay of rad ioactive su bstances in the i nterior of the earth. This was also an i mportant sou rce of i nternal energy on the pri mitive earth. The mate rial making u p the pri m itive earth certai nly contai ned larger amounts of the radioactive elements than are present now. It is known, for


example, that most of the radioactive potassi u m isotope, 4°K which was present on the pr imitive earth has already decayed ; so, energy production fro m 4°K was many ti mes g reater on the pri mi tive earth than it is now. The same consi derations apply, more or less strongly, to the production of energy by each of the other radioactive elements.

Since energy released in radioactive decay is the only energy ava ilable to maintain volcanic activity, and si nce much more of this energy was released on the pr imit ive earth than is released now, it may be safely assu med that volcani c activity was once more extensive than i t is now. However, it i s hard to be sure j ust how large an effect th is was. It is l i kely that, despite the much h igher level of vo lcan ic activity, the average temperature of the earth's surface soon settled down with i n the range that p revai ls today. I n al l subsequent d iscussion of the or ig ins of l i fe, it wi l l be assumed that the temperatures on the prim itive earth were s im i lar to those on the contemporary earth.

We have seen that the absence of hydrogen and hel i u m i n the earth's atmosphere i s read i ly explai ned b y the small size and relatively high temperature of our planet. When the abundances of certai n other elements are examined, a paradoxical s ituation i s revealed. Both neon and argon have been lost a lmost completely from the earth's atmosphere, but water, n itrogen and, oxygen have been retai ned. Theory shows clearly that s ince water, n it rogen, and oxygen mole-

Table 6.1 Atomic or Molecular Weig hts of Constituents of a Reducing Atmosphere

Atom o r Molecule Formula Mass

Hydrogen atom H 1 Hydrogen molecule H2 2 Hel i u m He 4 Methane CH4 1 6 Ammonia NH3 1 7 Water H20 1 8 Neon Ne 20 Argon A 40

1 05 History of the Earth, Atmosphere, and Oceans

cules are l ig hter than argon atoms, they should have escaped more eas i ly than argon.

There is only one reasonable ex planation of this u nexpected result. At a t ime when the earth h ad only partial ly condensed, neon and argon were free in the atmosphere and, hence, cou ld escape from the weak gravitational f ield, whi le carbon, n itrogen, and water were present in nonvolat i le materials trapped i n the dust particles. This explanation is a very reasonable one. The outstand ing feature of the chemistry of neon and argon is that these gases do not form any chemical co mpounds at a l l . Un l i ke carbon, n itrogen, and oxygen, they co uld not have been trapped with in the dust partic les i n a chemically combined form.

We are thus forced to a su rpris ing and i m portant conclusion, namely, that the gases in the atmosphere and the water in the ocean were expel led from the interior of the earth. When the earth condensed, most of the volat i le gases i n the d ust cloud escaped. The water that is now i n the oceans and the gases i n the atmosphere were not volat i le at that t ime; they were expel led from the i nterior of the earth at a later date. Soon after the earth had formed , its i nterior began to warm up , for large amounts of energy must have been released by the decay of rad ioactive elements. Any s l ightly vo lati le material i n the i nterior of the earth was, thus, forced to the surface as volcan ic gas. Later, as the temperature of the i nterior rose even h ig her, nonvolat i le materials were destroyed by the intense heat and the products of the i r destruction were fo rced to the surface and added as s imple gases to the atmosphere.

These conclusions are very i m portant when one considers the or ig ins of l ife, for they show that even if , as is l i kely, complex organic substances were p resent in the dust cloud, they would have been largely destroyed in the period i mmed iately after the formation of the earth. Most of the organic compounds needed for the beg inn ing of l i fe must have been synthesized afresh i n the earth 's atmosphere and oceans.

The Primitive Atmosphere

One of the most i mportant and controversial d iscussions connected with the orig ins of l i fe is concerned with the com-


position of the primitive atmosphere. To make c lear the signif icance of th is quest ion, we m ust fi rst explai n what is meant by the terms "oxi diz ing" and " reduci ng." The fo l lowing def in itions are of l i mited appl ication.

Most chemical elements fo rm compounds with hydrogen and with oxygen. The compound that contai ns the h ighest proportion of oxygen is said to be fu l ly oxid ized, whereas the compound that contains the h ighest proportion of hyd rogen is said to be fu l ly reduced. The fu l ly oxidized forms of carbon, n itrogen , and su lfur, for exam ple, are carbon dioxide (C02), nitrogen pentoxide (N205), and su lfur trioxide (803) , respectively. The corresponding fu l ly reduced molecu les are methane (CH4), ammonia (NH3), and hyd rogen su lfide (H2S).

Molecules which contain less oxygen than the fully oxid ized form or less hydrogen than the fu l ly reduced fo rm are sai d to be in an intermediate state of oxidati on. Carbon, for example, forms a series of molecules in intermed iate states of oxidation, namely carbon monoxide (CO), formic acid (HC02H), formaldehyde (CH20), and methano l (CH30H). This series i l l ustrates an im portant point: the oxidation state can be lowered by removing oxygen or by add ing hyd rogen, for molecu les whose compositions differ by one or more water molecules are i n the same oxi dation state. Al l b io logica l ly important organic mo lecu les are i n intermediate oxi dati on states.

The term "oxidizing" is used to descri be co nditions which tend to convert elements or co m pounds into the i r

___.. C=O � H"'-

O=C=O C=O ---_. H-C?0 / H/

Carbon dioxide

-o-H Carbon monoxide Formaldehyde

form�: acid H

, /H

H-C-0- H -- H-C - H H

/ 'H

Methanol Methane

Figure 6.3. Successive stages in the reduction of carbon dioxide. Note that carbon monoxide and formic acid differ by one H20 molecu le and so are equally reduced.

1 07 History of the Earth, Atmosphere, and Oceans

more highly oxidized forms. The simplest oxidizing gas mixtures are those that contain free oxygen. In a simi lar way, reducing conditions are those which tend to convert compounds to more highly reduced forms, and the simplest reducing atmosphere contains free hydrogen.

Consider an atmosphere which contains only hydrogen ( H.�). oxygen (02), and water (H20). Under the influence of ultraviolet light or electric discharges, free hydrogen and free oxygen combine until either the hydrogen or the oxygen is used up.

2H2 + 02 ....- 2Hz0

If hydrogen is initial ly present in excess, the final atmosphere contains hydrogen and water, whereas if oxygen is initial ly present in excess, the final atmosphere consists of oxygen and water. As we have seen, a simple atmosphere of this kind is cal led "reducing" in the former case and "oxidizing" in the l atter. The word " neutral" wil l be used to describe a balanced atmosphere containing water, but no excess of free hydrogen or of free oxygen.

If a smal l amount of an organic compound is introduced into an oxidizing atmosphere containing free oxygen, it is sooner or later converted into carbon dioxide. If a smal l

• + Water Water � + Water Figure 6.4. Combination of hydrogen and oxygen to give water. If the ratio of H2 to 02 is just 2 to 1 , only water is left. Otherwise an excess of either H2 or 02 may remain after the maximum amount of water has been formed.


quantity of carbon dioxide is introduced into a reducing atmosphere containing hydrogen, it is sooner or later converted into methane. If a large quantity of carbon dioxide is introduced into a reducing atmosphere, the excess hydrogen may be used up before a l l of the carbon dioxide is reduced to methane. In this case, compounds in intermediate oxidation states, including organic compounds, may be formed. The term "neutral" wil l be used more general ly than before to describe an atmosphere containing carbon dioxide and water, but no oxygen and no carbon compounds more reduced than C02• Organic compounds are formed only in an atmosphere that contains more hydrogen or less oxygen than does a neutral atmosphere.

In a similar way, if nitrogen is introduced into a reducing atmosphere, it is partial ly converted into ammonia (NH3). In an oxidizing atmosphere molecular nitrogen (N2) and oxides of nitrogen are present. A neutral atmosphere is defined as one that contains N2 and water, but neither ammonia nor oxides of nitrogen.

The oxidizing or reducing character of the primitive · atmosphere is very important for any discussion of the ori

gins of life, because the organic compounds from which al l living things are made are not stable in an oxidizing atmosphere. Furthermore, as wil l be seen in Chapter 7, the synthesis of important biochemical compounds takes place readily in a reducing atmosphere, but not in a neutral or an oxidizing atmosphere. Thus, most modern theories of the origins of life assume that the primitive atmosphere was reducing. It is important to see whether this assumption is consistent with the geological and geochemical evidence.

Today the atmosphere of the earth is strongly oxidizing. It contains about 80% of molecular nitrogen ( N2) and about 20% of free oxygen (02). The smal l amount of carbon in the atmosphere is present as carbon dioxide (C02). The chemical characteristics of the atmosphere are almost entirely determined by the free oxygen that is present in it. No organic compounds are synthesized in our atmosphere, nor are they stable if they are introduced into it. The volcanic gases from which the atmosphere was formed must have contained water and probably carbon dioxide as wel l. However, geolo-

109 History of the Earth, Atmosphere, and Oceans

gists have shown that the composition of the earth is such that l ittle, i f any, free oxygen cou ld have been present. The oxygen which is now found i n the atmosphere m ust, therefore, have been formed by the decomposition of water ; there is no other abundant source from which it cou ld have come.

Two mechanisms are know n that cou ld account for the formation of oxygen from water on the pri miti ve earth. The f irst is the action of h igh-energy u ltravio let l ig ht from the sun on water molecu les in the upper atmosphere. This produces hydrogen i n addit ion to oxygen, but the hydrogen escapes from the earth's gravitat ional f ield leavi ng free oxygen behind. When the amount of oxygen that cou ld have accumu lated i n 4.5 b i l l ion years through the photochemical decomposit ion of water i s calcu lated, it is found to be less than the amount now present in the atmosphere. Thus, although some oxygen has u ndoubtedly been formed in this way, much of i t must have had a d i fferent o ri g in . *

The other process that produces free oxygen is photosynthesis (Chapter 4). At the present ti me, algae and land p lants are p roducing vast amounts of oxygen. S ince fossi l algae that look very simi lar t o modern oxygen-producing photosynthetic algae are found i n Pre-Camb rian rocks, photochemical p roduction of oxygen, has p robably been going on for a very long time. It seems l i kely that most of the oxygen i n o u r atmosphere has been produced by photosynthesis.

Wh i le i t is genera l ly ag reed that the primitive atmosphere did not contain more than a trace of free oxygen , widely differi ng views have been expressed con cerni ng its reducing character. Accordi ng to one extreme view, the atmosphere was made up of completely reduced compou nds- methane (CH4), ammonia (NH3), and water (H20), and pro bably some molecular hydrogen. Accord i n g to the other extreme view, the earth 's early atmosphere was neutral and contained carbon d ioxide, n it rogen, and water. I t i s not possible at present to dec ide where, between these two extremes, the truth l ies.

Authors who bel ieve that the pr imit ive atmosphere was fu l ly reduced argue that s ince hyd rogen is by far the most

• This conclusion is now disputed.

1 1 0 Steps toward a Solution

abundant element in interstellar space, it certainly predominated in the dust cloud from which the solar system was formed. They conclude that it also predominated in the primitive atmosphere. They draw attention to the large amounts of methane in the atmospheres of the outer planets and claim that this also is evidence that large amounts of methane must once have been present on the earth. However, since the atmosphere came from inside the earth, both of these arguments are weak. Hydrogen from the dust cloud may have escaped completely as the earth was forming; the gases expelled from the interior of the earth may have been quite different from the gases making up the primary atmospheres of the outer planets.

The arguments in favor of an oxidizing atmosphere are no more conclusive. The gases evolved from volcanoes are not strongly reducing, and it is clear that an atmosphere derived from contemporary volcanic gases would contain a great deal of carbon dioxide. However, most volcanic gases are formed by the recirculation of water and gases from the earth's crust and do not come from the deeper layers of the earth. Thus, the composition of contemporary volcanic gases tells us little about the composition of the gases which come from the earth's core. Even if we knew the composition of such gases, we could not assume that it is the same as that of volcanic gases on the primitive earth, because great changes have occurred in the interior of the earth during the 4.5 billion years that have elapsed since its formation.

There is some geological evidence of a different kind which favors the view that the early atmosphere was reducing, or at least not oxidizing. Mineral deposits formed under reducing conditions contain iron in what is called the ferrous state, while mineral deposits formed when free oxygen is available always contain ferric iron. Very large amounts of ferrous iron were deposited in early PreCambrian times. This shows that the early Pre-Cambrian atmosphere did not contain much free oxygen, but unfortunately it does not distinguish unambiguously between a neutral and a reducing atmosphere.

We shall see that the success of most prebiotic

1 1 1 History of the Earth, Atmosphere, and Oceans

syntheses which start with simple gases requires that the reaction mixtures be reducing rather than neutral or oxidizing, but it does not depend on the detailed composition . of the mixture. Ful ly reducing mixtures containing methane, ammonia, and hydrogen or partial ly reducing mixtures containing carbon monoxide, nitrogen, and hydrogen behave in very similar ways. Although the nature of the earth's first atmosphere is not yet known in detail, it is almost certain that it was sufficiently reducing to make possibl e the organic syntheses which wil l be discussed in later chapters. Of course, the finer details of the composition of the earth's primitive atmosphere are a subject of great intrinsic interest, but they are perhaps less critical for theories of the origins of life than is sometimes supposed. To explain the formation of organic mol�cules it must be supposed that the atmosphere was reducing and not much more needs to be assumed about its composition.

We know very little about the time that it took to form the atmosphere and oceans. Presumably, water, carbon compounds, and nitrogen or ammonia were released together from the interior of the earth. After a while the atmosphere became saturated with water vapor and rain began to fal l . At first smal l lakes must have formed and then, as more and more water entered the atmosphere, these lakes must have been enlarged to form the oceans. The accumulation of salts in the oceans depended on the weathering of rocks and may have been quite slow.

All this is important because it raises an interesting question about the environment in which life began. If the formation of the oceans was rapid, one is justified in assuming that life began in oceans and lakes similar to those with which we are familiar. If, on the other hand, the oceans formed s lowly, life may have begun before the oceans reached their present size and before they acquired their present content of salts. It is not possible to decide between these a lternatives. In the rest of this book we shal t make the conservative assumption that, when life began, the distribution of water on the earth was much as it is now. If the oceans had been smaller and less sal ine than they are now, some of the difficulties connected with the accumulation of organic materials would have been less severe.


Sources of Energy

Energy Sources

It is generally believed that the very first stage in the origins of life was the accumulation of large amounts of dissolved organic material in the oceans and lakes of the primitive earth. The solution that was formed in this way is often referred to, picturesquely, as the prebiotic soup. It is important to know whether, when life began, the prebiotic soup was hot or cold, thick or thin, and so on. In this chapter the formation of the ingredients of the soup will be discussed in a preliminary way.

A reducing atmosphere of the kind that is thought to have been present on the primitive Earth is stable indefinitely unless it is acted upon by a source of energy. No new organic compounds are formed when a mixture of methane, nitrogen, and water, for example, is left to stand in the dark. Complex organic compounds are formed if and only if the

113

mixtu re i s heated strongly, i rradiated with u ltraviolet l ig ht, acted u pon by an electric discharge, or subjected to the action of some other form of energy.

The chemistry involved in the formation of organic molecules from sim ple gas mixtures is too co mpl icated to be discu ssed extensively here. Instead a single reaction wi l l be descri bed : the formation of hydrogen cyanide (HCN) from nitrogen (N2) and methane (CH4 ). This i s one of the most i mportant p rebiotic reactions and, fortunately, one of the simp lest.

When an electric discharge is passed t h rough an atmosphere contai n ing molecular nitrogen (N2 ), some of the molecules absorb so much energy that they d issociate i nto a pair of atoms.

N2 ____,. 2N

Now, whi le nitrogen mo lecules are q u ite u n reactive, nitrogen atoms attack a l most any other atom or molecule. I n particular, nitrogen atoms react with methane molecules to give hydrogen cyanide and hydrogen .

N + CH4 ---+ HCN + ! H2 Methane Hydrogen

cyanide

It fo l lows that when an electric discharge is passed through a mixture of n itrogen and methane, the n itrogen atoms formed i n the discharge react with methane molecu les in thei r envi ron ment to form hydrogen cyanide. I n a simi lar way, n itrogen molecu les are d issociated into atoms when nitrogen is heated very strongly or i rradiated with h igh-energy u ltraviolet l i g ht. Thus, hydrogen cyan ide cou ld eq ually well have been formed i n the p ri mitive atmosphere from n itrogen and methane by the action of l i ghtning, volcanoes, or u ltraviolet l ight.

Hydrogen cyanide, un l i ke nitrogen and methane, is very so luble in water. Any hydrogen cyanide formed in the pri m itive atmosphere would, therefore, have dissolved i n raindrops and then fou n d its way i nto the oceans. I n the next chapter, the way i n which i m portant organic compou nds are formed from aqueous sol utions of hydrogen cyanide w i l l be


discussed. Here we need recognize only the i mportant general pri nciple that reactive molecules, l i ke hydrogen cyanide, were formed in the prim itive atmosphere. S ubsequently, they dissolved in oceans and lakes, where they reacted with one another, spontaneously, to form the prebiotic sou p.

It i s clearly i mportant to decide which forms of energy contri buted most to the synthesis of organi c compou nds o n t h e primitive earth. This wi l l n o t be an easy matter u n t i l o n e can b e certain abo ut t h e com positio n o f t h e primitive atmosphere. In the meanti me, one can make some i nformed guesses.

The s u n su pplies far more energy to the earth than does any other source. Each year about 260,000 calories of radiant energy are i ncident on each square centi meter of the atmosphere. This amount of energy is suffi cient to boi l away a layer of water twelve feet dee p covering the whole surface of the earth . However, not all of this energy is useful because l ight can bri n g about chemi cal reactions o n ly if it is absorbed by one of the components of the reaction mixtu re; l ight that passes th rough a m i xt u re of gases without bei n g absorbed can h ave n o i nfl uence o n t h e .g ases.

Table 7.1. Present Sources of Energy Averaged over the Earth

Source

Total radiation from sun Ultraviolet l ight beyond

3000 A

2500 A 2000 A 1 500 A

Electric d ischarges Shock waves Radioactivity (to 1 .0 km depth) Volcanoes Cosmic rays

Energy (cal cm-2yr-1)

260,000

3,400 563

41 1 .7 4 1 . 1 0.8 0. 1 3 0.001 5

Figure 7. 1 i l l u strates the way i n which the energy reaching the atmosphere from the sun is distributed over the

1 1 5 Sources of Energy

Figure 7.1 . Solar radiation intensity above the atmosphere at earth's distance from the sun. I n al l an average of about 260,000 calories is i ncident on

each square centimeter of the atmosphere per year. (Reproduced with permission from Source Book on the Space Sciences by Samuel Glasstone, Van Nostrand, Inc., New York 1 965.)

i nfrared, v is ib le, and u lt rav iolet reg ions of the spectrum. Most of the energy is in the visi ble region ; the energy avai lable fal ls off rapid ly in the u ltravio let. Since the gases that were present in the primitive atmosphere do not absorb visible l ight, on ly the fraction of the energy present as u ltraviolet radiation cou ld have been used for the synthesis of o rganic compounds. The precise amount wou ld have depended on the composit ion of the pri mitive atmosphere.

An atmosphere contai n i ng methane, ammonia, n it rogen, and water of the kind that m ight have existed on the p ri m i-


tive earth could have absorbed at most 40 calories per square centimeter per year, that is, less than 0.02% of the total solar energy. This would not have been enough to cause very extensive organi c synthesis. However, i t has been shown that if hydrogen sulfide or formaldehyde were present in sufficient quantities in the atmosphere, they could have absorbed a much larger amount of ultraviolet energy and made it available for the synthesis of organic compounds. At present we have no information about the abundances of formaldehyde and hydrogen sulfide in the primitive atmosphere.

The reader should realize that the amount of ultraviolet light reaching the surface of the earth today is very much less than the amount that reaches the top layer of the atmosphere. The surface of the earth and the lower atmosphere are protected by a layer of ozone (03 ) in the upper atmosphere. Ozone, unlike the other components of the atmosphere, absorbs ultraviolet light strongly and prevents it from reaching the earth. Otherwise, men would be subjected to very harmful doses of ultravi olet light whenever they were exposed to direct sunlight.

I I I I I I I I I I I I 'f

302 ---+ 203 Oxygen Ozone

Visible and UV light

1

/ . from sun � : : : . . J - -�-�-�- -�-�- --i--1 I l I Ozone layer 1

RedJcing 1 - ·r- - --r�-l-�- - - -

ffill l i ·f� Surface of earth Surface of earth

Figure 7.2. Partial shielding of the earth from ultraviolet l ight by the ozone layer formed in an oxidizing atmosphere.

117 Sources of Energy

Certain ly, there was no ozone on the primitive earth because it is formed from 02 a n d can not s u rvive i n a �educing atmosph e re. Many of the photoc hemical reactions occ ur ring in the p ri m itive atmosp here coul d not occu r today near the su rface of the earth because so l i ttle u ltraviolet l ight penetrates the ozone layer.

Electric d ischarges have been used to study prebiotic synthesis more often than any other energy sou rce, main ly because they prod uce orga nic compou nds with h i g h efficiency. At the p resent t ime the q uantity of e lectic energy released on the earth is very much smaller than the total amount of solar energy i ncident on the atmosphere. However, a l l of the electric energy is i n a form that is effective in synthesizing o rganic compounds from a reducing atmosphere, w h i le most of the sun's energy wou l d have passed through such an atmosphere without causing chemical c hange.

We do not know how often t h u nderstorms occu rred in the reducing atmosphere of the p ri mitive earth. Consequently, we do not know how much e lectric energy was avai l ab le. I shal l assume that the amount was not very d ifferent from that avai lable today. However, this may be an u n deresti mate si nce there a r e reasons f o r believing that t h u n derstorms w o u l d b e very frequent i n a reducing atmosphere conta i n i n g ammonia.

Very large amounts of energy were released within the earth through the decay of radioactive s ubstances. We do not bel ieve that this energy was i mportant for the origins of l ife s in ce most of it was l i berated far below the s u rface. Even if organic com pounds were formed, they co u l d never have found their way i nto the pr imitive oceans and lakes.

Volcanic activity, which is due i n d i rectly to the energy l iberated by radioactive decay p rocesses, occurs at the s u rface of the earth and cou l d have contributed to prebiotic synthesis. The amount of energy presently released i n vo lcanoes is q u ite small , but on the p ri mitive earth i t cou l d easi ly have been as much as ten t imes g reater. Nonetheless, it is doubtful that volcanoes were the major site of o rgan ic synthesis, because no synthetic processes a re known which make efficient use of vo lcani c energy.

Molten lava is hot enough to bring about the synthesis


of o rganic compo u nds from a reducing atmosphere, but once the lava comes i n to contact with the atmosphere, its surface cools off very rapi dly and sol idifies. Once this happens, the synthesis of o rgani c compounds becomes inefficient, since only the gases which penetrate through the sol i d su rface of the lava to the hotter material beneath are heated strongl y.

A g reat deal h as been written about the i m portance of volcanoes an d hot springs for the orig ins of l ife, and it would be rash, in the absence of further evidence, to dismiss vo lcan i c energy as u n i m portant. Some organi c com pounds would certainly have been formed at l ocations whe re a red ucing atmosphere made contact with hot lava. A few real istic experiments carried out i n an active volcano might answer a n u m ber of i mportant q u estions - and provide a good deal of excitement for the i nvestigators in the p rocess.

Many other forms of energy have been proposed as causes of prebiotic synthesis, for ex13m ple, cosmic rays, son i c energy generated by ocean waves, and shock waves generated by thunderstorms or by meteorites enteri ng the earth's atmosphere. I n al l cases the amou nts of energy involved are quite smal l , and only i n the case of shock waves is synthesis c laimed to be efficient enough to make up for this disadvantage. If the rather s u rpris i n g claims for the efficiency of synthesis i n shock waves can be confirmed, we will have to consider them as a potenti al ly i m portant source of prebiotic o rganic compo u n ds.

It is u n li kely that al l of the organic compounds synthesized on the pri mitive earth we re formed i n the same way. We need to determ ine how much each energy sou rce contri buted to the synthesis of each class of compound, rather than to f ind the unique energy source that was responsi ble for al l prebiotic syntheses. In my opinion, electric discharges and ultraviolet l i g ht must certainly h ave been i m po rtant, b ut the role of other energy sources is as yet u nclear.

It is perhaps worth em phasizing that few nonbiological organic compou nds are acc u m ulati n g on the earth today. Nonbiologi cal synthesis of organic compou nds does not occur in o u r atmosphere because so much free oxygen i s p resent. Even those compo unds that are i ntrod u ced i nto the

1 19 Sources of Energy

atmosphere as i n d ustrial po l lutants are largely degraded by m icroorganisms. Co mpounds of the k ind that are bel ieved to h ave been formed on the p ri m itive earth are al most al l b iodegradable; the novel prod ucts of the chemical laboratory, fo r example, the ch lori nated hydrocarbons, are the most i m portant ones that escape degradation.

Rates of Accumulation of Organic Compounds

We must now try to get some i d ea about the rates at which o rgan i c compounds were formed i n the atmosphere, and the amounts that accumu lated on the surface of the earth and i n the oceans. Fi rst, i t must be realized that the ti me ava i lable for synthesis and accumu lation of organ i c com po u n ds was very long; so long, i n fact, that it falls outside the range for which we have any i ntuitive feel ing. It seems almost certain that acc u m u lation took more than a m i l l ion years, and perhaps i t took c loser to a bi l l ion years.

Let us suppose, for example, that the amount of o rganic material synthesize d on the earth each year was very small , say a k i logram per square k i lometer, that is about a n ounce fo r every ten acres. Then , if a l l of the material formed i n a b i l l ion years acc u m ulated eve n ly over the surface of the earth , a layer of organic solids three feet deep wou l d h ave resu lted. This example should make it c lear that rapi d and u n reasonable rates of synthesis need not be assu med to justify the existence of a r ich prebiotic sou p.

I n fact the rate of synthesis was probably m u ch g reater than has been assumed i n the above ex ample. If electric energy was avai lable i n amounts comparable with those released on the earth today, and if this energy was used as efficiently as it is in modern laboratory experi ments, about 1 mg of organ i c materi al wou ld h ave been formed each year for each square centi meter of the earth 's s u rface. This is equivalent to ten thousand ki log rams per squ are ki lo meter and wo u l d lead to the accu m u l ation of a l ayer of o rganic materi al three feet deep in as l ittle as a h u n d red thousand years.


It seems u n l i kely that the efficiency of prebiotic synthesis cou l d ever have approached that of laboratory synthesis, but even if it had been only 1 % as efficient, i t wou l d have prod uced vast amou nts of material with i n a period of a few m i l l ion years. Simi lar conclusions are reached when correspond i n g cal culations are carried out for organic synthesis i n itiated by u ltraviolet l i g ht from the sun, if we suppose that hydrogen sulfide or formaldehyde was present i n the atmosphere.

The esti m ates which we have made of the amou nts of crgani c material accu m u lati ng on the pri mitive earth are based on the assu m ption that al l material formed i n the atmosphere would have acc u m u l ated at the surface. We know that this is a n overs i mplif icati o n , because many of the organic compounds formed u nd e r prebiotic co nditions are u nstable and decom pose in much less than a mi l l ion years. However, even when this is taken i nto accou nt, there is every reaso n to bel i eve that the oceans and lakes of the pri mitive earth containe d abundant o rganic material . The p rebiotic soup may easi ly have contained as much as a g ram of o rganic mate rial per l iter of water. This is about a third as concentrated as an average chi cken bou i l lo n . *

.

* The calcu lation was performed for Knorr's Chicken Boui l lon made up according to the supplier's instructions.

121 Sources o f Energy

I I I

Prebioti' Synthesis

Introduction

The term "prebiotic chemistry" is commonly used to describe chemi cal reactions that are carried out in the laboratory with the i ntention of s imulating processes that occu rred on the pri mitive earth. The reactions themselves are often l ittle different from those stu died by chemists whose motivation is quite d ifferent Like t h e man who discovered to h i s surprise that h e h ad been writi ng p rose for years, many organ ic c he mists have been doing prebiotic chemistry for years, without real iz ing it.

One of the most i mportant experi ments in prebiotic chemistry was carried out i n 1 832 by the German chemist , Fried ric h Wohler. When Wohler published h i s work, scientists sti l l believed that t here was an essential difference between the c hemistry of l ifeless materials, such as minerals or rocks, and the chemistry of the constituents of l iving

1 23

organisms. The two subjects were called i norgan ic and organi c chemistry, respectively. These terms are sti l l used, although today the su bject matter of o rganic chemistry includes the properties of almost al l co m pounds of carbon, whether they are of biological origin or not.

Wo hler showed that when an inorgan ic compound, ammon ium cyanate, is heated, an organ ic com pound, urea, is formed. Woh ler's experi ment is im portant because it i s part of a chain of evi dence demonstrating that there is really no unbridgeable gap between the chemistry of l ifeless and li vi ng matter. It is interest ing that Wohler's synthesis of u rea can also be considered as an i mportant prebiot ic reaction in the modern sense, s ince it is l i kely that u rea was formed from ammonia and cyanate on the pri mitive earth.

Many reactions carried out in the late ni neteenth and early twentieth centuries qualify as prebiotic. Glycine, an i mportant ami no acid , was obtained from hydrogen cyanide, for example, and sugars were synthesized from formaldehyde. S ince both hydrogen cyanide and formaldehyde had a lready been obtai ned from inorganic sou rces, these reactions cou ld j ustif iably have been described as prebiotic.

Prebiotic chem istry in the sense of the del iberate s imulation of reacti ons that occurred on the pri m itive earth i s of much more recent orig in . Opari n and Haldane emphasized that i n the period before the f irst l iv ing organisms evolved, the earth's atmosphere must have been reducing. They suggested that a mixture of organic compounds was formed in such an atmosphere and that the fi rst l iv ing organisms were assembled from a se lection of these compounds. We have argued that the results of astronomical and geophysical research are broadly consistent with the hypoth esis that the pr imitive atmosphere was reducing, although we must ad m it that many questions of detai l remai n to be decided.

Synthesis of Ami no Acids, Sugars, and Nucleotide Bases

I n 1 953 Stanley M i l ler, then a student of Harold Urey at the U nivers i ty of Ch icago, subjected a mixture of methane,


ammonia, hydrogen, and water to the action of an electric discharge. He was able to show that, just as Oparin and Haldane had predicted, a mixture of organic compounds including amino acids was formed. These experiments stimulated many related studies and can reasonably be considered to mark the beginning of modern work on prebiotic chemistry.

Prebiotic chemistry is a complicated branch of organic chemistry. The detailed results are of a highly technical character and can be understood only in the framework of mechanistic organic chemistry. In this chapter we will summarize the progress achieved to date without going into detail. Readers who have the necessary background in organic chemistry will find the details in the references cited at the end of the book.

The equipment used by Miller in his first experiment is shown in Figure 8. 1 . The small flask was filled with water, and the rest of the apparatus with a mixture of methane, hydrogen, and ammonia. The gas mixture, together with some water vapor, was caused to circulate past the tungsten electrodes by boiling the liquid water in the small flask. Then a spark discharge was formed between the electrodes by applying a high electric potential across them.

The products formed in the electric discharge dissolved in the water which liquified in the condenser and were then carried down into the small flask. In an apparatus of this kind, volatile products are constantly distilled out of the small flask along with steam and then subjected again to the action of the discharge; nonvolatile products accumulate in the small flask.

Miller's experiment is instructive because it illustrates a number of features common to many attempts to simulate the chemistry that occurred on the primitive earth. Miller chose to work with a mixture of methane, hydrogen, ammonia, and water because, at the time, it seemed probable that the primitive atmosphere contained these gases as its main components. The boiling water in �he small flask was supposed to represent the primitive ocean. Finally, the electric discharge which was passed through the gas was considered as an equivalent to lightning in the primitive atmosphere. In interpreting the resu lts, it was supposed that the

125 Prebiotic Synthesis

r Water}

Ammonia Methane

Hydrogen

Wires carrying electric current

Spark discharge

Figure 8.1 . Apparatus used in Mi l ler's experiments on the formation of amino acids from a reducing gas mixture. Reproduced with permission from Stars, Planets and Life by Robert Jastrow, Wil l iam Heinemann , Ltd., London, 1 967, p. 1 34.

material that acc u m u lated i n the boi l i n g water corresponded to organi c su bstances that would have acc u m u lated in the pri mitive ocean.

Some featu res of M i l ler's experi ments are clearly unreal istic : the oceans, for example, were not boi l i n g at the t ime when l ife evolved. M i l le r chose to work with boi l ing water rather than a cold "ocean " for tec h n i cal reasons; otherwise the gases in the apparatus wo uld not have circulated q uickly


enough. This procedu re h ad another i mportant advantage, for it speeded up a n umber of reactions that would h ave been very slow in a cold solution. The effective l ife of a g raduate student is at most a few years, so it is often necessary to ac hieve with in a few days what must have taken hundreds of years or more on the primitive earth. That is why experiments in prebiotic chemistry rarely take the form of an attem pt to rep roduce exactly the conditions that existed on the p ri mitive earth. More often extrapo lation is used where exact s imu lation would be impractical ; we study chemical reaction under conditions which al low us to work out in a short time what would h ave happened i n a much longer ti me on the pri m itive earth.

Two types of extrapolation are particu larly i mportant. In general , the rates of chemical reactions increase as the temperature increases. It is possib le to make accu rate esti mates of reaction rates at low temperatu res i f the rates are measu red at two or more h igher temperatures. In this way, it is poss ible to learn in days or weeks about reactions which would have taken m i l l ions of years in the pri m itive ocean .

The rates of many chemical reactions also depend in a fairly si mple way on the concentrations of the reactants. This makes it poss ib le to determine how d i lute solutions in the pr imitive ocean would have behaved, by stu dyi ng more concentrated solutions i n the laboratory. This is i mportant, because it is diff icult to work with more than a few l iters of solutions i n an ordin ary chemical laboratory, whi le the volume of the oceans was enormously larger.

We must now return to the res ults of Mi l ler's experime nt. At the end of a week, the spark was turned off and the contents of the apparatus were a l lowed to cool down. The solution i n the smal l f lask was then subjected to a deta i led analysis. The results were extremely surpris ing. As m uch as 1 5% of the carbon which had or ig inal ly been in the "atmosphere " ; was present as identified organic compou nds in the "ocean." About 5% of the carbon had been converted i nto i mportant biochemical compounds. The most strik ing feature of Mi l ler's results was the discovery that several of the natural amino ackls were formed in substantial amou nts.

1 27 Prebiotic Synthesis

H I ,

o H-C-C

I "oH

N H2

Glycine Alanine Aspartic acid

Figure 8.2. Th ree of the amino acids formed in Mi l ler's experiments.

G lycine, alan ine, aspartic aci d , and g l utami c acid were identif ied with certainty.

To u n derstand how su rprisin g this is, one m ust consider the resu lts against the backg rou n d of general organ i c chemistry. Many mi l l ions of organic com po u n ds a re known, and many thousands have struct u res no m o re com pl icated than those of the am i no acids. Before these experiments were reported, most chemists wou l d have antici pated that an electric discharge act ing on a mixture of methane, ammonia, hyd rogen , and water wou ld produce a mixt u re co ntai n i n g smal l q ua ntities of many different s ubstances. I t su rely cannot be a coi nci dence that relatively few substances were obtained and that many of them are i m portant biochem ical compounds fou n d in al l l iv ing organisms.

These observati ons are most easi ly explained if it is supposed that the most primitive organisms were com posed of o rgan i c compounds that had been formed by the action of an e lectric discharge on a redu c i n g atmosphere. S ince mode rn organisms m u st have retai ned many featu res of thei r pr imitive an cestors, we cou l d then u nderstand why cells contain so many of the compounds that M il ler i dentified among h is products.

I n fact the situation i s more compl icated. M i l ler's work stim u lated many further efforts to u nderstand the organi c chemistry that occurred on the primitive earth. I t was soon d i scovered that s imi lar sets of organic molecules, i nc luding amino acids, are formed whenever a reducing mixtu re of g ases is heated strongly enough, i rrad iated with u ltraviolet l i g ht or subjected to any type of electric d ischarge. Fu rtherm ore, it was fou n d that the natu re of the gas mixture is not c ritical, p rovided the mixture is reducing. M i l ler's m ixtu re of


methane, hydrogen, n itrogen, ammonia, and water can be rep laced by a m ixtu re of carbon monoxide, hydrogen, n i trogen, and water, for example. I t may be con cluded that amino acids were avai lable on the pri mitive earth s ince they are formed whenever a reducing g as mixture i s treated violently enough. It is not known which energy sou rce was most i mportant on the earth, although I personal ly bel ieve that electric discharges must have made a major contribution to prebiot ic synthesis.

During the twenty years that have elapsed si nce the publ icati on of M i l ler's origina l paper, nearly al l of the natural ly occurring ami no acids have been identified as products in one or the other of the many prebiotic synthesis that have been reported. The reactions are not all eq ual ly convi ncing, but there is l i tt le reason to doubt that if the atmosphere was red ucing, many of the twenty natural amino acids would have accumulated on the pr imitive earth.

So far it has proved impossible to obtain amino acids from an atmosphere which contai ns free oxygen or from a mixtu re of carbon d ioxide, n itrogen, and water. This is a very i mportant negative resu lt, for it argues strong ly i n favo r of the Oparin-Haldane hypothesis that the prim itive atmosphere was reducing. Life co uld not have got started i n an atmosphere of the type that exists today.

This situation may seem somewhat paradoxical , since today al l h igher and most other forms of l ife are completely dependent on oxygen. Only a few types of bacteria and certain other microorganisms can su rvive under the conditions that were necessary for the origin of l i fe. This is an i mpressive example of adaptation. Most organ isms presu mably began to use oxygen once i t was abundant and after a ti me became absolutely dependent on it. Now, they can no longer su rvive in an oxygen-free env i roment.

When an organi c chemist makes a surprising discovery, h is f i rst i nstinct i s usually to prove that it is not surpris ing at al l . He tries to show that the new facts that he has d iscovered fit in with, and perhaps extend, generally accepted theories. M i l ler was soon able to demonstrate that some of the amino acids among h is products had been formed by a well-known route. One of the f irst syntheses of g lyc ine to be

1 29 Preblotlc Synthesis

reported invo lved heating together hyd rogen cyanide and formaldehyde in an aqueous solution of ammonia. M i l ler showed that hydrogen cyanide and formaldehyde were formed by the electric d ischarge and reacted together i n aqueous sol ution t o g ive g lyci ne. Thus, M i l ler was repeating a wel l -known synthesis of g lycine, but generat ing the start ing materials with an electric d ischarge under prebiotic conditions.

This work focused attention on hydrogen cyanide and led to the next major development in p rebiotic chemistry. Juan Oro was checking whether amino acids could be formed from hyd rogen cyanide and ammonia, in the absence of formaldehyde or si m i lar su bstances. He fou nd that if he warmed ammonia and hydrogen cyanide together i n aqueous solution for a few days, amino acids s imi lar to those d iscovered by M i l ler were formed. He found , i n addit ion, that aden ine was obtained in about 0.5% yield. More recently, a commercial synthesis of aden ine from ammonia and hyd rogen cyanide has been developed i n Japan.

Adenine is , of course, one of the fou r bases present in RNA and DNA. It is also a component of ATP, the i ntermed iate involved in the storage and uti l ization of chemical energy in biological systems. The structu re of adenine is q u ite compl icated , so much so that it is qu ite astonishing that it can be formed i n large amounts from hydrogen cyanide i n so s imple a reaction. It is hard to avoid the conclusion that aden ine occupies a central position in biochemistry because it is one of the few organ ic compounds of this degree of complexity that formed i n large amounts on the pr imitive earth.

The evidence that many i mportant biochemicals are formed with su rpris ing ease under prebiotic conditions has

NH2 I

N�c'c--�, I II �CH HC.::::;N"'C ...... N

H Adenine

Figure 8.3. The formula of aden i ne.


H - C==N H -C==C-C == N Hydrogen cyanide Cyanoacetylene

Figure 8.4. The formulas of hydrogen cyanide and the related molecule, cyanoacetylene.

contin ued to accumu late. By now, syntheses of al most a l l of the monomeric components of the genetic apparatus have been achieved under prebiot ic conditions. Here, the desc ription of o ne more exam ple must suffice. Si nce fou r n ucleotide bases occur together in RNA, it seemed l i kely that they were formed together by s im i lar mec hanisms on the pr imitive earth. It was natu ral , therefore, to look fo r a synthesis of cytosine (C) and urac i l (U) from an i ntermediate s im i lar to hydrogen cyani de.

An exami nation of the structure of cytosine (Fig ure 3.7) suggested that its synthesis cou ld be achieved by start ing with cyanoacetylene, a gas closely related to hydrogen cyan ide. This i ndeed proved to be the case. However, it was su rpri sing to f ind that cyanoacetylene i s an ab undant product in almost al l prebiotic reactions that produce hyd rogen cyanide. When methane and n i trogen are subjected to an electric d ischarge, for example, hyd rogen cyanide is the main n itrogen-conta in ing product but cyanoacetylene is also formed in reasonable amou nts.

So far, noth ing has been said about the prebiot ic synthesis of two very d ifferent components of the geneti c system, the sugars - deoxyribose and r ibose. I n fact, the preb iotic synthesis of r ibose was achieved in the n i neteenth century by a Russian scienti st, Butlerov. In the But lerov reaction, formaldehyde is shaken with chalk or l i me. Th is very s imple proced ure gives a com plex mixture of su gars, i nc luding ri bose. Deoxyribose can be obtained by a modification of this procedure.

We may sum marize th is aspect of prebiotic chemistry as fol lows. Wheneve r a reducin g mixture of gases contain ing carbo n , hydrogen, n itrogen, and oxygen is treated vio lently enough , a s im i lar pattern of smal l , h igh ly reactive mo lecu les is formed. These i n c lude intermediates, such as formaldehyde, hyd rogen cyani de, and cyanoacetylene. In the presence of water and ammonia, the reactive i ntermediates

1 31 Prebiotic Synthesis

combine together to form more compl icated organic molecu les. The products are remarkable in that they inc lude many more of the compounds i mportant in modern biochemistry than can be attributed to chance. To explain this f ind ing it must be supposed that the fi rst organisms evolved from organic com pounds formed by the action of h ighenergy sou rces on a red ucing atmosphere and that modern organisms have evolved from pri mitive ones without much change in chemical co mposition.

It w i l l be seen in Chapter 14 that new evi dence supporti ng these conclusions has come from an unexpected source. Astronomers have found that interste l lar d ust clouds contain large amou nts of organic material . The fi rst organ ic co mpounds to be identified were formaldehyde, hyd rogen cyan ide, and cyanoacetylene, th ree of the most popular p rebiotic intermediates. To some extent this must be a coincidence, but it probably ind icates that our ideas on prebiotic syntheses are broadly correct. There cannot be too many radical ly different fami l ies of organic molecu les that are formed from inorganic start ing materials under prebiotic conditions.


The Formation of Polymers

Concentration Mechanisms

I n the last c hapter we saw that h ighly reactive organic molecu les, once they had formed i n the pri mitive atmosphere, d isso lved i n oceans and lakes, and that they reacted there to form ami no acids and other compounds important for biochemistry. No further progress toward the appearance of l ife was possib le unti l these molecu les h ad combined to form polymers . We can be sure of this, because the operation of the genetic apparatus depen ds on pro perties of polymers that have no counterpart among the properties of smal ler molecu les. This is particu larly clear in the case of the n ucleic acids. The sequence of nuc leotides in a nuc leic acid transmits the i nformation needed to specify a protein sequence. No smal l molecu les cou ld have performed th is fu nction because there is no way i n which they could have carried so m uch information.

1 33

Laboratory experi ments have shown that po lypeptides and polynucleotides are not easi ly formed from ami no acids and n u cleotides in di l ute solut ions. Thus, the prod uction of polymers on the primitive earth was al most certain ly p receded by the formation of concentrated aq ueous sol utions, by the deposit ion of sol i d organic materi al , or by the adsorption of organic compounds on the su rfaces of m i nerals, and so forth. The prebioti c sou p had to thicke n or sol i d ify before the next stage i n the ori g i ns of l ife cou l d begi n . Some of these same concentration mechanisms may have played a part i n the syntheses of su bstances, such as sugars and nucleotide bases, which were di scussed in the previous c hapter.

Evaporation is the most fam i l i ar mechanism that brings about the concentrat ion of d i l ute aq ueous sol utibns. If the process i s conti n ued long e nough, so l id deposits are formed. There are many places on the earth where concentrated solutions or sol i ds have been formed in this way. Desert lakes, such as the Salton Sea in Cal ifornia and the Dead Sea, are satu rated with salt; i n l an d salt-f lats h ave often been formed by the evaporation of l akes.

The case of the Dead Sea is typical . Water h as flowed i nto it from the s u rro u nd i n g mo untains carrying with it salts extracted from the rocks. Si nce the c l imate i s very hot and very d ry, a g reat deal of water has evaporated, leaving the d i ssolved salts beh ind. Consequently, the water of the Dead Sea is satu rated with sod i u m and magnes i u m salts, and l arge amounts of sol i d salt h ave been deposited i n the area. The tou rist visitin g the caves at Sodom can i nspect a p i l l ar of so l i d salt that we bel ieve to have been formed by evaporation, althou g h a different account of its ori g i n is g iven by the local i n h abitants.

Sol id deposits of salt are also formed on rocky seashores. Tidepoo ls, formed by very h i g h ti des, someti mes evaporate to leave behind d eposits of sol i d salt that are not wetted agai n u nt i l the next very h i g h t ide or the next rai nfal l. Salt crusts can also be formed by the evaporation of water that has splashed u p above the l i mit of the tide.

I f we are correct in bel ievi n g that nonvolat i le organic compo u n ds were formed on the primitive earth, they mu st


have been concentrated i n lakes and tide pools, just as salts are today. Some water-soluble pol lutants are concentrati ng i n this way at the present ti me. I t seems almost certai n that concentrated solutions or sol id deposits of organi c material formed by evaporation played an i m portant role in the orig ins of l i fe.

Volati le materials cannot usually be concentrated by evaporation, since they are driven off f i rst when an aqueous solution is heated. All water-soluble su bstances, inc lud ing those which are volati le, can be concentrated from aq ueous sol ution by freez ing. S ince many of the h igh ly reactive molecules formed i n a reducing atmosphere are volat i le, concentration by freezing is l i kely to have been particu larly important in the earl ier stages of chemi cal evolution. Moreover, as we shal l see, a nu mber of late steps i n the o ri g i ns of l ife are a lso l i kely to have occu rred at low temperatures.

If a d i lute aqueous solution is cooled to a tem perature below O"C, i ce beg ins to separate out Si nce organic compounds are not sol u ble i n ice, they are left beh ind in the solution. Thus, as more and more i ce is formed, the solution left behind becomes more and more conce ntrated. Th is process conti nues unt i l the solution is saturated. At th is point the whole mass sol id if ies to g ive a mixture of sol id organic material and ice crystals.

Th is meth od of concentrat ing o rganic solutions is put to good use i n the production of applejack i n Canada and the northern United States. Barrels of cider are left out for long periods in wi nter unt i l most of their contents are frozen. The l iqu id that remains contai ns all of the alcohol that was or ig inal ly present in the cider and almost a l l of the flavori ng materials. This provides an advantageous method of preparing a strong, palatable l i quo r. Most governments col lect taxes on dist i l led l iq uor, but very few, if any, tax l iq uor concentrated by freezing.

A number of reactions that must have been i m portant on the pri m iti ve earth have recently been shown to occur most readi ly in cold or partial ly frozen solutions. The synthesis of aden ine from hydrogen cyanide is an important example; the tem plate reactions to be discussed in Chapter 10 are others.

Other concentration mechanisms that may have operated

1 35 The Formation of Polymers

on the primitive earth depend o n adsorption. Many of the most i m portant prebiotic molecules stick ten aciously to the surface of certain m i nerals. They are extracted from the b u l k o f a n aqueous solution a n d co ncentrated i n a narrow zo ne at the l i q u i d-so l i d i nterface.

The clay mi nerals have very large su rface areas and can, therefore, extract large amou nts of organic material from aqueous sol uti on. Bernal was one of the fi rst scientists to emphasize that clay surfaces may have been the s ite of important prebiotic condensati o n reactions. Recently, as we shal l see, evidence has accu m u l ated to show that i mpo rtant condensation reactions do i ndeed take place very eff iciently on the surface of a clay called Montmori l l o n ite. Other experiments s u ggest that other su bstances i n c l u d i n g the most common phosphate m i nerals, the apatites, catalyze certai n prebiotic condensations.

Most organi c compo u n ds di ssolve in water to g ive true solutions, that i s sol utions that contai n o n ly i so l ated sol ute molecules ; th is i s the case for g l ucose and acetic acid, fo r example. A n u m ber of compo u nds, detergents and soaps, for example, do not dissolve to form true solutions, but instead disperse to give smal l more o r less s pheri cal particles of organic material (see Chapter 4). Opari n cal led these co l loidal dro p lets "coacervates" , a term which is now widely used. Si nce many organ i c s ubstances are soluble in coace r -


W M

Figure 9.2. The difference between (a) a true solution and {b) a colloid. In a true solution, solute molecules are wel l-separated. In a colloidal aggregate the solute molecules cluster together.

vates or are adsorbed on thei r surfaces, Oparin suggested that, on the pri mitive earth , coacervate droplets provi ded the most im portant site for prebiotic condensation reactions.

From many poi nts of view this is an attractive i dea. Organic molecules d isso lved i n a coacervate droplet would exist in a local ly nonaqueous envi ronment, even though the droplet was float ing in a vast excess of water. Many condensation reactions go best in the absence of water and these wou ld be favored if they were carried out in coacervates. So far l i tt le expe rimental ev idence has been produ ced to show that i m po rtant prebi otic condensation reactions do occur under these conditions, but the subject is of considerable interest and deserves further study.

·

Condensation Reactions

The most i m portant b iolog ical polymerizat ion reactions are al l dehydrations, that i s reactions i n which the l ink between neighboring monomers is estab l ished by the removal of water. This is particu larly c lear i n the case of peptide-bond format ion , e .g . ,


0 0 I I ,-------------; II -H O

H2N -CH2-C- lQ_Ij_± __ ljj NH-CH2- C-OH 7 0 0 II II

�N- C�- C- NH-C�-C-OH

It must be admitted from the beg inn ing that the way in which condensation reactions occu rred on the prim itive earth, is not understood. A few condensations are known to proceed under prebiotic conditions, but no efficient synthesis has been discovered that can p lausib ly be considered as a forerunner of prote in synthesis or polyn uc leotide rep l ication.

The most important biochemical condensations take place in a predomi nantly aqueous envi ronment. It seems un l ikely that any form of n ucleic acid repl ication , whether enzymatic or not, could occur in the absence of water. Th is makes it almost certain that some of the earl iest prebiotic condensations took place in sol ution. However, attempts to sim u late these reactions have not been very successfu l ; it is diff icu lt to form a polymer by removing water from monomers if the environment al ready contains a vast excess of water.

It is much easier to dehydrate mixtures of amino aci ds or nucleotides by heati ng them unti l water is d riven off. So me thermal condensations can be made to proceed quite efficiently in this way. Unfortunately, there is no obvious relationship between any of these reactions and modern biological condensations. I f dry thermal reactions were i m portant for the orig ins of l ife, it is hard to understand how they were replaced by sol ution reactions later in the cou rse of biochemical evolution.

I bel ieve that the two types of reaction occurred i n close association on the prim itive earth. Random polymers and reactive intermed iates were formed in efficient thermal reactions. When the m ixtu res of products formed i n this way were wetted, fo r example by rain or dew, they reacted further in so lution. If this pictu re is correct, modern biological condensations evolved from these latter solution reactions.


Of course, there are other poss ib i l ities. Perhaps there are more eff ic ient methods of condensing monomers i n an aqueous environment than any discovered so far. I n that case, the genetic system may have evolved without the intervention of thermal co ndensations. I n view of these uncertainties, both types of reaction w i l l be described.

Thermal Condensations

If a solution of amino acids is heated gently, water is driven off and a so l id cake of organi c material is left beh ind. On stronger heatin g , chemi cal ly bou nd water is e l im inated and, under certain ci rcu mstances, peptides are formed. If , for example, a mixtu re of al l twenty natu ral ly occu rri ng amino acids i s heated, good yie lds of po lypeptides are obtai ned ; they have been cal led protei noids, si nce they are claimed to resemble proteins qu ite closely.

From many points of v iew, this is an i deal prebiotic condensation. The reaction cond itions are very s imple and no reagents are needed other than the amino acids themselves. However, the reaction does not occur at tem peratures substantially below 1 30°C. The h ighest temperatures reached at the surface of the earth today, except in vo lcanoes, is close to 80°C, and it seems un l i kely that substantial ly h igher temperatures occu rred at the s u rface of the pri mitive earth. Thus the thermal polymerization of ami no acids by d i rect heati ng could have occu rred on ly in vo lcanoes.

Reasons fo r question ing that vol canoes were i m portant for the o rig ins of l ife have been g iven al ready. The thermal synthesis of polypeptides, if it was a s ign if icant prebiotic reaction, probably occu rred at lower temperatu res, perhaps in the presence of organic catalysts. However, this is an undecided issue and a number of autho rs, part icu larly the American scientist Dr. Sidney Fox, bel ieve that vo lcanoes p layed a major role in the synthesis of organic polymers on the pr imit ive earth.

So far, attempts to form nucleic acids by dry heati ng have had even less success than the corresponding attempts to synthesize protei ns. Many different steps are involved in


the formation of a n ucleic acid from its com ponents. S ince none of them occurs at a l l eas i ly on heating under p lausible prebiotic condit ions, it m ust be concluded that the synthesis of nucleic acids is un l ikely to have occu rred in this way on the pri mitive earth.

Despite these disappoi ntments, or perhaps because of them, slow progress is being made along s l ightly different l i nes. Livi ng organisms are ab le to bri ng about d ifficu lt syntheses with the help of enzymes. There is no poss ib i l ity that enzymes, i n the sense of polypept ides with precisely defi ned amino acid seq uences, existed on the pri m itive earth before the appearance of l ife. Nonetheless, s imple organic or m ineral catalysts that permitted thermal condensation reactions to proceed at temperatu res as low as 70-80°C, may have been present.

So l itt le is known about catalysis in the sol id state that we cannot predict which m inerals or organ ic molecu les would catalyze thermal condensations. Searching for a prebiotic catalyst is , therefore, rather l i ke looking for a need le i n a haystack. Nonetheless, a few interest ing catalysts have been discovered by trial and error. It was stated i n Chapter 8 that the synthesis of u rea by Wohler can be considered as the f irst prebiotic synthesis. It tu rns out that u rea is an excel lent catalyst for the i ntroduction of inorgan ic phosphate i nto nuc leosides, one of the d i ff icult steps i n the formation of nucleic acids from their components. With the help of u rea, it is possible to add phosphate to n ucleosides and to form short polynucleotides f rom them under condit ions that occur today in many desert areas.

Many different organ ic compou nds must have been formed on the pr imitive earth, and some of them certain ly became concentrated i n rather l im ited geograph ical reg ions. Areas that were rich in catalytical ly active material wou ld have provided sites for later stages in the evolution of l ife. Sites where m i neral catalysts were abundant cou ld also have been i m portant. The search for prebiotic catalysts among compounds that cou ld have accumu lated on the pri mitive earth is sure to become an increasingly i m portant aspect of prebiotic chemistry.


Condensations in Solution

Next, attempts that have been made to synthesize polymers in aq ueous sol utions must be described. Here, we are faced with the d iff iculty that a l l proteins and n ucleic acids are decom posed i nto thei r constituents by water; the protei ns react with water to g ive amino acids whi le the nucleic acids decompose fi rst i nto nuc leoti des and then i nto sugars, bases, and i norganic phosphate (see F igure 3.7) . The reverse reaction, the formation of proteins o r n ucleic acids from monomers, never occurs spontaneously in aq ueous sol ution .

Some of the most i m portant aspects of modern chemistry are con cerned with the concept of equ i l ibr ium (Chapter 4). A system is said to have reached chemica l eq u i l ibri um when there is no longer any tendency for i t to undergo fu rther chemical change. Clearly a solution of a protei n i s not in equ i l ibr ium, s ince proteins in ti me decom pose into amino acids. On the other hand, a solution of amino aci ds is in equ i l ibri um , si nce there is no ten dency for amino acids to react to form peptide bonds. P rote ins and n ucleic acids, i n so l ution , d o not reach eq u i l ibr ium u nti l they have broken down into thei r com ponents.

It was shown i n Chapter 4 that a chemica l system can never be sh ifted from its equ i l i brium position witho ut the expenditure of energy. It fol lows that energy m ust be suppl ied to the monomers if amino acids are to be conve rted to polypeptides, or nuc leotides to polyn ucleotides. I n l iv ing systems the energy i s always provided by the breakdown of ATP. We wou ld l i ke to know where the energy came from on the prim itive earth.

Many experi mental stu d ies, both of b io logical and nonbiolog ica l condensation reactions in so lution, have shown that a s ing le basic mechanism is i nvolved. Monomers always react fi rst with the sou rce of energy to form activated inte rmediates, also cal led h igh-energy i ntermediates. These activated intermediates, because they conta in a large amo unt of chemical energy, react to form polymers remember, the original monomers cou ld not fo rm polymers di rectly. The


seq uences of reactions i nvolved in protein synthesis (Ch apter 4) i l lustrates th is pri nciple.

0 0 I I enzyme II

H2N-CHR-C-OH + ATP -----'---'> H2N -CHR-C .-,....PA + P2 Amino acid

(not activated)

(energy source)

Aminoacyl adenylate (activated)

0 II

n H2N-CHR-C.-...-PA Aminoacyl adenylate

imany steps

0 0 II I I

Pyrophosphate

H2N-CH R-C-NH-CHR-C · · · · · + nAMP Peptide-n-un its Adenylic acid

Net Result : nATP + nAmino Acid ___.,. nPeptide + nAM P + nPyrophosphate

Amino acids react with the primary source of energy, ATP, to gi ve h ig h-energy intermediates cal led aminoacyl adenylates. These are the i ntermediates from which proteins are u ltimately fo rmed.

It seems al most certain that s im i lar mechan isms were involved in any prebioti c condensations that took place in aqueous soluti on. Th is raises_ two connected questions: What were the i m portant h igh-energy i ntermediates on the primitive earth? How were they formed ? Several answers to these questions have been offered, but none of them is total ly convi ncing.

H ig h-energy intermed iates could have fo rmed when d ry mixtu res of so l ids were heated with su itab le catalysts. One attractive scheme i nvo lves h igh-energy phosphate compounds from the very beg inn ing of the evolution of the genetic apparatus. ATP and rel ated compounds could have formed from n ucleotides and inorganic phosphates in the presence of u rea or some other catalyst. The h ig h-energy phosphates formed i n this way wou ld have undergone fur-


ther condensation when the products of the thermal reacti on came i nto contact with water.

A n u mber of spec ia l s ites on the pri mitive earth can be imagined where a seq uence of reactions of this k ind cou ld have occu rred. I n deserts, the surface may be heated to 80°C d u ring the day and then wetted with dew at ni ght. Ti depools and shal low desert lakes a re locations where hot-dry and cool-wet condit ions alternate at intervals of months or even years. It seems qu ite l ikely that the po lymerization phase of the orig ins of l ife too k place at specia l s ites s u ch as these. Alternatively, po lymerization cou ld j ust poss ib ly have occurred near volcanoes, where certain rock surfaces are maintai ned at steady high temperatu res except when drenched with rai n.

A nu mber of completely different schemes i n which activated i ntermediates are formed d i rectly in solution have been proposed. U nfortu nately, they are not easi ly explained without goin g i nto g reat detai l . Here it m ust s uffice to remark that certain hi ghly reactive molecu les that co u ld have formed i n the prim itive atmosphere, for example, cyanoacetylene, react with amino acids and nucleoti des in aq ueous sol ution to form h igh-energy i ntermediates. These intermediates subseq uently react to form polymers . U nfortunately, the react ions are so i neffic ient that it has not been possi b le to synthesize polypeptides or po lyn ucleotides i n reasonable amounts by these methods.

More and more attention is now being g iven to the search for materials that catalyze condensation reactions in aqueous env i ron ments. Some of the most exciti ng resu lts from such wo rk are those being reported from Aharon Katchalsky's labo ratory in Israel. He and h is co l laborators have shown that a co mmon clay minera l , Montmori l lon ite, adsorbs certain activated amino acids and converts them in almost 1 00% yield to long po lypeptide molecu les by comb in ing a n umber of mi neral catalysts. Katchal sky's g roup succeeded in forming polypeptides in good yield d i rectly from amino acids and ATP - th is i s an i m pressive achievement.

The particular processes discovered by Katchalsky may


or may not have been i mportant for the origins of l ife. However, his work i ndicates clearly that polymerization of the type that m ust be postulated to accou nt for the ori g i ns of l ife do take p l ace on the s u rface of s i m pl e mi neral catalysts. They may als o take place on coacervates, but this remai ns to be demonstrated.


Replicating Molecules and

Natural Selection

Molecu les and Natural Selection

In Chapters 7-9 we discusseJ a fo rmati on of si mple organic molecu les on the prim itive earth and the way in which they reacted with one another to fo rm polymers. These react ions ·m ust have set the stage for the o ri g i ns of l ife, but they are very different from characteristical ly b io logical processes, such as DNA repl ication and prote in synthesis. The present chapter deals with the prob lem of the evol ut ion of a selfrep roducing biologica l system from a fami ly of random polymers. This transition was the crucial phase in the o rig ins of l i fe.

The natu re of the c rit ical transit ion to a biological system can be explained on ly after we have seen how the theo ry of n atural selection can be appl ied to the behavior of populations of rep l icati ng macromolecu les. The princi ple of natu ral selection was ori gi na l ly proposed by Darwin and

1 45

Wallace to accou nt for the way in which ani mals evolve. Since we shal l be deal ing with the evolution of molecu les rather than that of ani mals, we shal l have to d iscuss the theory in a somewhat u nconventional way.

The theory of natural selection can be appl ied to the behavior of any popu lation of reproducing entities. Here, these entities are referred to loosely as organ isms, without imp lying that they are normal l iv ing organisms; from this point of view, an organism could be an elephant, but it cou ld equally wel l be a molecule of RNA. The ai m of the theory of natu ral selection is to accou nt for the changes that occur i n popu lations o f organisms as they compete agai nst each other.

To describe a populat ion, one needs to classify its members and then to specify how many ind iv iduals there are i n each of the c lasses. Someti mes, the way in which the characteristics of the members of a class change with ti me is of i nterest. On other occasions, on ly the variation i n the sizes of the classes may be of i nterest. I n considering the human inhabitants of the U.S.A. , one m ight be concerned , for exam ple, with the way in which the average height is changing in occidental and oriental com m un ities. Alternatively, one might be i nterested only i n the proportions of occidentals and orientals in the total popu lation.

When we are i nterested only i n the numbers of members in d ifferent wel l -defi ned classes, the law of natural selection takes a s imple form : the organisms that reproduce most efficiently sooner or later dominate the population , and a l l other closely related organisms are e l imi nated. Th is principle had been recognized long before Darwin developed his complete theory, for exam ple by Malthus at the end of the eig hteenth centu ry.

We have to be careful about the mean ing we g ive to the expression " reproduce efficiently. " An organism that prod uces few offspring all of which reach reproductive age often outg rows a competitor that produces more offspring many of which die i n chi ldhood. It is possi ble to take such co mpl ications i nto account and to derive a number that describes the overa l l rate of growth of a popu lation (the Malthusian parameter). For many purposes, it is more conve-


n ient to specify the rate of growth ind irectly by g iv ing the

t ime that i t takes for a population to double in size. Natural

selection guarantees that the organism that dou bles the s ize of its popu lation in the shortest time sooner or later eliminates al l its competitors.

To see how these ideas apply to the evolution of macromolecular systems, consider two fami l ies of self-repl icating organisms, say RNA molecules, that are competing against each other. Suppose the s lower-growing population takes a

minute to double and the faster-growin g population takes 50 seconds. Then, i n 5 minutes, the two populations go through 5 or 6 doubl ings, respectively; the ratio of the size of the faster-growing population to that of the s lower-growi ng population , therefore, doubles every 5 min utes. After 5 hours, a member of the faster-growi ng popu lation would produce on the average 2:60(1 018) t imes more descendants than a member of the slower-growing popu lation.

When we study natural selection experimental ly, we are never able to maintain very many rounds of rep l ication be

fore the supply of nutrients runs out. I n typical experiments on bacteria or repl icatin g RNA molecules, a smal l i nitial pop

ulation is chosen and permitted to go through about ten rou nds of repl ication. At this point it is necessary to add a small sample from the expanded population to a new supply of nutrients to start the second cycle of growth. This process may be repeated many times.

Let us return to our n umerical example and see how this new procedure wou ld work o ut. I t w i l l be assumed that the g rowth cycles last for five minutes and are a lways in i tiated with samples of 1 ,000 organisms. It wi l l also be assumed that the o rig inal population contains equal numbers of fast- and slow-growing RNA molecu les.

Our very f i rst sam ple wou ld contain about 500 fastgrowing and 500 slow-growing organisms. We have seen that the ratio of the size of the fast-growing population to that of the s low-growing population doubles every five minutes, so o u r second sample wou ld contain about 667 fast-growing and 333 slow-growing o rganisms. It is easy to calculate that the ratio wou ld approximate 800 : 200 (4 : 1 ) at the th ird sampl ing, 889 : 1 1 1 (8 : 1 ) at the fou rth sampl ing , and

1 47 Replicating Molecules and Natural Selection

so on. By the n i nth sampl ing we would expect on ly about 4 slow-g rowing organisms to be left. Th ree fu rther cycles wou ld usual ly be enough to e l im i nate the last slow-growing organisms. At that po int selection would be complete ; no slow-growi ng organisms would be fou nd in any subseq uent cycle. It should be noted that the proportion of slow-growing organisms decreases geometrical ly with ti me; if each growth cyc le were in it iated with a sample contain ing 1 ,000,000 organisms instead of 1 ,000, selection would take only twice as long.

The amount of food avai lable for growth is always l i mited under natu ral conditions, so that expansion of popu lations of organisms is always subject to restrictions si m i lar to those d iscussed above. When a sample contain ing different kinds of organisms is introduced i nto a hospitable envi ronment, the popu lation grows u nti l it is l imited by the supply of nutrients. Then the popu lation size remains constant wh i le the d ifferent types of organ isms in the sam ple compete agai nst each other. Sooner or later, the organisms that repl icate most eff iciently outg row al l their com petitors. If no new types of organisms were produced by mutation, selection would then be at an end.

The next part of the theory of natu ral selection becomes relevant when one considers the way in which new types of organisms evolve. Notice that this part of the theory presupposes that new variants can arise spontaneously. If organisms were always exactly l ike thei r ancestors, one population m ig ht displace another, but no population with novel characteristics cou ld evo lve; some species might become exti nct, but no new species cou ld evolve to replace them.

In Darwi n 's time l i ttle was known about the sources of the variabi l ity that make evol ution possi ble. Darwi n 's own ideas on the subject proved to be i ncorrect. The basis for an adeq uate solution of the problem of variat ion in p lants and ani mals was proposed by Mendel as early as 1 865, but it was so much in advance of its time that it was neglected. Mendel 's laws were not red iscovered unti l the beg inn i ng of the twentieth centu ry.

Mendel ian genetics is a compl icated su bject. The d iff icu lties are due to the manner i n which each parent contributes to the genetic constitution of the offspring in sexual ly


reproducing species. We do not have to worry about any of these d iff icu lties , because we shal l be concerned with evol ution at a period before the development of sexual mechanisms of reprod uction. A "progeny" molecule i s derived from a s ing le parent and, in the absence of mutation , is indi stinguishable from that parent.

Let us now consider a fam i ly of repl icating n ucleic acids competing against each other in a constant enviro ment. (The arg ument wou ld be exactly the same for any other type of repl i cating molecu le.) I n the absence of mutat ion, those molecu les in the popu lati on that repl icate fastest would e l iminate al l others, and after that no further change co uld occu r. However, si nce n ucleic acid rep l ication is i mperfect, mutant molecules would occasiona l ly be produ ced that rep l icated at a rate different fro m that of the standard type. Mutants that rep l icated more slowly wo uld be e l imi nated by se lection in

the usual way, but any mutant that repl i cated faster mi ght take over.

We can now see how a population of repl icati ng molecu les wou l d be l ikely to evolve when transferred i nto a new environ ment. At fi rst, the most efficient type in the orig inal popu lation, F say, would take over. In a new environ ment it is al most certain that a better adapted mutant M 1 would soon appear and displace F. In time M1 would , i n turn, be displaced by a faster-growing mutant M2 , and so on. However, this process could not go on fo rever. Each successive type that dominated the population wou ld have to be better adapted than any of its predecessors. In t ime a mutant MA would appear that was as wel l adapted as possib le ; once this mutant took over, evolut ion wo uld be at an end.*

When we d iscuss the rep l ication of polymers on the primitive earth , we assu me that the env i ronment differed f rom place to place. Some regions were hot and others cold : in some the prebiotic soup was concentrated and in others it was more d i l ute. U nder these c i rcu mstances it i s l i kely that d ifferent rep l icatin g po lymers took over the various regions. A primitive form of "speciation" must have existed from the beginn ing .

The reader may be surprised to real ize that we have just

* This is an oversimpl ification, but adequate for the present purposes.


described, i n a general way, the crucial step i n the origi ns of l ife. Fam i l ies of rep licating molecu les, com peti n g agai nst each other and evolving under the influence of m utation and natu ral selection, m u st have passed by i m percepti ble steps into the richly diversified species that we know today. I n few other cases can such a complicated story have h ad such a s imple beginning.

One may or m ay not ag ree that a fam i ly of repl icati n g macromolecul es o f t h e type described above can b e fairly described as livi ng. However, it i s certain ly true that they exhibit at best a rather uninteresti ng form of l i fe. The next stage in the evolution of l ife must have been the selection of polymers that could do someth ing more i nteresting than repl icate at the expense of preformed monomers.

S ince at this stage, the repl icati n g entit ies were no more complicated than polynucleotides, further adaptation m ust have depended on simple i nteractions between these polymers and other nonrepl icating molecules in thei r environment. F i rst, the polynucleotides " learned" to capture whichever small molecules i n the p rebiotic sou p could help them to repl icate faster. Later they must h ave learned to join together smal l molecules i n the prebiotic soup and to hold onto the polymeric products. G radually the n u cleic acids came to assume the role of d ictators, contro l l i n g the chemistry of their envi ronment for their benefit.

The most i mportant of the interactions between repl icat ing nucleic acids and smaller molecu les i nvolved the amino acids; the genetic code i s the f inal produ ct of the evo lution of nucleic aci d/amino aci d i nteraction. The i nvention of prote i n synthesis permitted nucleic acids to d i rect the synthesis of enzymes and i n this way establ ish a lmost complete control over their chemical environment. The variety of l iving forms i s an expression of the diverse strategies that n ucleic acids h ave evolved i n order to make use of thei r environ ment to favor thei r own repl ication.

These are the ideas that motivate much modern work o n t h e o ri gi ns o f l i fe. T h e poi nt o f view is deli berately l i m ited. We concentrate on the repl ication of n u cleic acids and on the way i n which nucleic acids use thei r envi ro nment to faci l i tate their own rep licatio n . In th is field, the molecular biol-


ogist is l i ke a man who regards an elephant as a co mpl icated device des igned to repl icate elephant DNA; the other point of view, that elephants are interest ing animals in them-selves, i s tempo rari ly forgotten. In the next section we shall concentrate on the repl i cati ng nucleic acid molecu les and the adaptations which the successfu l ones underwent to outg row the i r neighbors.

Nucleic Acids and Proteins

I n the last section we discussed the app l ication of the theory of natural selection to the evolution of fam i l ies of repl icating organ ic polymers. Although the arguments used apply equally well to the evol ution of any fami ly of polymers , i t was assu med that the n ucleic acids were the fi rst genetic molecu les on the primitive earth. S im ilarly, i t was su pposed that polypeptides played an important part i n the early stages of the development of l ife. These si mple and plaus ib le assumptions need to be justified , s ince they h ave often been criticized in the past.

It has been argued that the biochemistry of the f i rst organisms may have been quite d ifferent from contemporary biochemistry, and hence that the f irst genetic appa ratus may not have been composed of nucleic acids and proteins. This argu ment is logical ly sound. The hypothesis that nucle ic acids and p rote ins have been involved from the very beg inn ing of l ife cannot be accepted without some justification.

Amino acids and nucleotides are formed readi ly under prebiotic condit ions, so polypeptides and polyn ucleoti des m ust h ave been amongst the most abundant polymers formed on the primitive earth. F u rthermore , we shal l see that there is evi dence suggesti ng that n ucle ic acids could have repl icated i n the absence of enzymes. These experi mental observations certa in ly make it seem very l ikely that the genetic apparatus has always been composed of proteins and nuc leic acids, b ut they are not sufficient to p rove the point.

There is an additional and very powerf u l arg u ment for


working with polynucleotides and polypeptides. No detai led descri ption of a genet i c system based on polymers other than nucleic acids has ever been pro posed. At the present time, studies of nonenzymat i c repl i cation of macro mo lecul es m ust deal with polyn u cleotides, because we know of no other repl icat ing polymers. S i m ilarly, we know of n o m ate rials other than polypepti des that can serve as models of pri m itive enzymes.

Unt i l recently, theo ri es of the o ri g i ns of l i fe were often based on the ass u mption that the f i rst o rgan i sms to evolve on the earth were made enti rely of protei n. This view is no longer held so widely, because it is thought u n l i kely that any molecule related to present-day proteins could have repl i cated; a n o rganism composed entirely of protei ns cou l d not have evolved by m utati on and natural selection. Polypeptides may have been i m po rtant as catalysts at an early stage i n the o rig ins of l ife, but they co uld not have f unctioned as the fi rst genetic material .

Polyn ucleoti des, u nl i ke proteins, do underg o reacti ons related to residue-by-residue repli cation under prebiotic cond it ions . This has been demonstrated in recent experiments that use artif icial polymers related to nucleic aci ds. Poly U i s a polymer that contains o n l y o n e type of res i d ue, u ridyl ic acid (U) , so it can be consi dered as a very si mple nucleic aci d . Poly U has t h e remarkable property that, when mixed with monomeric derivatives of adenyl ic ac i d (A) at sufficiently low temperatures, it organizes the adenyli c ac i d res idues i nto a helix. Once the A residues are present i n a hel ix, they are more eas i ly joi ned together than w hen they are free i n solut ion. Thus, at low enough tem peratures, poly U d i rects the synthesis of polymers of A. Reacti ons l i ke this are called template-directed reactions, because the polymer acts as a template bri nging together the monomers i n a way that makes it easier to jo in them up.

Other experi ments have shown that ·poly C, a si mple polymer of cyti dyl i c aci d , d i rects the synthesis of polymers of G. On the other hand, poly U has no i nfl uence on the po lymeri zation of derivatives of G, and poly C has no i nf luence on the polymerization of derivatives of A. Thus the Watson-


U ··· A U ··· A l U ··· A U ··· A l U ··· A U "· A l U .. · A u ... A l U .. · A U .. · A l u .. . A u ... Al u ... A u ... A l

(a) {b)

Figure 1 0. 1 . (a) A poly U-po ly A double-hel ix; (b) part of a poly U-mono A triple-helix. The A's are arranged head-to-tai l, ready to be joined up. For simpl icity the second poly U chain is omitted.

Crick rules (Chapter I l l ) , that A pairs with U and G pairs with C, are obeyed in template-d i rected reactions. These reactions al most certai n ly depend on interactions between bases that are the same as those that make enzymatic DNA-repl ication possible (Ch apter IV).

We are sti l l far fro m bei ng able to rep l icate a n ucleic acid without the help of enzymes. No netheless, the i nformation that we have al ready obtai ned is of g reat i mportance. It shows that very crude template-di rected processes corresponding to n ucleic acid rep l i cation could probably have occured on the pri m itive earth . Si nce template-directed reactions are su ccessfu l on ly at low temperature, it suggests also that the f i rst stages i n the evolution of the genet ic apparatus occu rred in a cool prebiotic soup.

If we accept these arguments about the p roperties of polypeptides and polyn ucleotides we can clarify a subject


that has generated a good deal of heated d iscussion -which came f irst, the proteins or the nuc leic acids? To answer the question we need f i rst to d isti ngu ish carefu l ly between two classes of polypeptides.

If the seq uence of a polypeptide is determi ned, whol ly or in part, by the seq uence of a preformed polynucleotide, it w i l l be referred to as an informed polypeptide. Otherwise a po lypeptide is called noninformed.* By defi n it ion , i nformed po lypeptides co u ld not have existed in the absence of nucleic ac ids; noninfo rmed polypeptides, on the other hand , could have accumu lated before the fi rst n ucleic acids.

There is an extremely i mportant d isti nction between info rmed and noninformed polypeptides - natu ral selection can operate on the former, but not on the latter. To see this we need on ly recol lect the previous d iscussion. S ince polypeptides cannot rep l icate, there is no known way in which one generation of noninfo rmed polypeptides can i nfl uence the next. I n the absence of some form of "reprod uction" it is un l i ke ly that natu ral se lection could occur. A system contain ing nucleic acids and i nformed polypeptides is subject to natu ral selection ; those nucleic acids that d i rect the fo rmation of useful informed polypepti des are successfu l in e l im inat ing their less "talented" competitors. Notice that natu ral selection does not act on the i nformed po lypeptides d i rectly, but on combi ned systems of i nformed po lypeptides and " i nform ing" polynucleotides.

We can now answer the orig inal question qu ite precisely: natu ral selection could not act on protein sequences unt i l nucleic acid repl ication was underway. On the other hand, noninformed po lypeptides could well have existed from the beg inn ing and they could possibly have catalyzed the fo rmation and repl ication of the fi rst n ucleic acids.

We have no i nformation at present on this last point. It is possib le that pri mitive nucleic acids were formed , without the help of polypeptides, on the surface of some mi neral. Alternatively, coacervates and organic catalysts, inc lud ing noni nformed po lypeptides, may have faci l itated the fo rma-

* The words random and nonrandom are confusing in this context. Noninformed polypeptides are not necessarily random.


tion o r rep l ication of the m ost primitive n ucle ic acids. On the whole, I find the second alternative more convi ncing, because it a l lows more g radual t ransition to be m ade from primitive to contemporary nucleic acid repl i cation.

The Evolution of the Genetic Apparatus

From now on it wi l l be assu med that the most pr imitive genetic p rocess is polynucleotide repl ication and that protein synthesis is the major adaptation that permitted nucleic acids to control their chemical envi ronment. Our next problem is to understand i n more detai l the sequence of steps that occu red on · the primitive earth and led to the evolution of the genetic apparatus in its fi nal form.

At this point it wi l l be useful to review some of the material presented i n Chapter 3. DNA is a regu lar polymer composed of just fou r monomeric com ponents, T, C , A and G . I n a l l l iving organisms DNA functions as a master copy o f the genetic mater ia l ; the repl ication of DNA is essentia l for the propagation of genetic i nformation from one generation to the next. DNA also di rects the synthesis of messenger RNA; messenger RNA is an i ntermediary that carries the information orig inal ly coded in DNA to the protei n-synthesiz ing apparatus.

Protein synthesis is a compl icated process in which messenger RNAs determ ine the sequence of new protein molecules. The relation between the seq uences of nucleotides in a messenger RNA and the sequence of amino acids in the prote in which it specifies i s determi ned by the genetic code. The genetic code ass igns an amino acid to a sequence of three n ucleotides; it is therefore cal led a th ree-letter code. The genet ic code inc ludes triplets of nuc leoti des that s ignal the termination of a prote in chai n as well as those that specify amino acids.

When th ink ing about the evolution of this compl icated system , it is essential to remember that it is a prod uct of natu ral selection and not the construction of a rational biochem ist. It is a mistake to bel ieve that the most prim itive g enetic system was as wel l-defined as the present system, but


sim p ler. On the contrary, the fi rst genetic system is l ikely to have been less wel l-defi ned and possi b ly more com plex.

It is often asked whether the fi rst rep l icati ng nucleic acid was an RNA or a DNA. Probably it was neither; the fi rst nucleic acid cou ld wel l have contained ri bonucleotides, deoxyri bonucleotides, and derivatives of other sugars. Anything in the prebiotic sou p that fitted into a rep l icati ng structure would have been uti l ized. The com ponents of DNA and RNA, if they were both present in the prebiotic soup, could have been d i rected into d isti nct polymers on ly after the evolut ion of enzymes. The fi rst nucleic acids may wel l have contained only two different bases, rather than fou r, but conceivably they cou ld have contai ned more than fou r. In any case, the rep l ication of pri mitive n ucleic acids must have been less accu rate than enzymatic DNA repl ication. S im i larly, the earl iest proteins m ust have contained a com plex mixture of amino acids derived from the prebiotic soup. No doubt some of the amino acids in pri mitive polypeptides are no longer represented i n proteins; converse ly, some of the twenty natu rally occurring amino acids may have been absent from pr imitive polypeptides. It is possib le that compounds other than amino acids occasional ly found the i r way i nto pr im itive proteins.

The modern prote in-synthetic apparatus is made up of more than a hundred proteins and nucleic acids. The most pr imitive form of cod ing can have involved no more than a few partial ly ordered polymers. It m ust, therefore, have lacked almost a l l of the specificity of modern protein synthesis.

The pri mitive protein-synthetic apparatus, for exam ple, cannot have discri m inated between closely related amino acids. It is un l i kely that leucine, isoleucine, and val ine were recognized as different amino acids before the evolution of the protei n-synthetic apparatus was wel l -advanced. The pri mitive genetic code, therefore, specif ied classes of re lated amino acids, rather than ind ividual amino acids. The very fi rst form of cod ing may have differentiated on ly two classes of amino acid . S im i lar arguments suggest that there could have been no eff ic ient stop and start signs in the most prim i tive code; "punctuation" could not be achieved without the


help of wel l-defined proteins. Thus, the length of pr imit ive i nformed polypeptides cannot have been specified accurately.

I t seems c lear, the refore, that the f i rst i nfo rmed polypeptides were fami lies of partly-ordered pepti des of variable length. They may have acted as weak nonspecific catalysts, or they cou ld have performed an even less exacting funct ion, for example, as "g lue" hold ing together coacervates i n the pre bi otic soup. I t wou ld certai n ly b e a mistake to think of pr imitive i nformed polypeptides as thoug h they were modern enzymes.

So far we have emphasized the sloppiness of primitive prote in synthesis. Now we must deal with the one feature of protein synthesis that m ust have been sharply defined from the beg inn ing . The genetic code, from a very early stage i n its evol ution on, m ust have been a th ree- letter code. There i s n o obvious reason why a two-letter or fou r-letter code shou ld not have evolved i n the primitive earth. However, no transit ion from an advanced two- or fo u r-letter code to a three-letter code wou ld have been poss ib le. Such a transition wou ld have led to a d isastrous m is interpretati on of al l the genetic information that had been accum u lated by natural selection.

We do not understand much about the later stages in the evolution of the code. Somehow, an i naccu rate system m ust have pu l led itself up by its bootstrings ; polypeptide seq uences that were poorly defined must have combi ned together to b u i ld new protein-synthetic apparatus capable of

Figure 10.2. The impossib i l i ty of switch ing from a two-letter code to a three-letter code. After evo lving to d i rect the synthesis of a particu lar sequence of 7-8 amino acids , the polynucleotide shown above would have to switch to determin ing a completely unrelated sequence of five amino acids. Clearly the new sequence could not function in the same way as the old one did.


synthesizing sl ightly better defined sequ ences. The code m ust g rad ually have improved in d iscri m i natio n unti l it "crystal l ized" into its final shape.

We do not know whether the structure of the genetic code is a h i storical accident or not. The code may have developed in its present form because of specific i nteractions between amino acids and tri n u cleotides. Perhaps glycine interacts more strongly with the seq uence GGG (o r the complementary sequence CCC) than with any other tri n ucleotide. In that case GGG wou ld probably code for glyci ne if the genetic apparatus evolved a second ti me. Alternatively, the relationshi p between trinucleotides and amino acids specified by the code could have been determi ned by arbitrary factors. Perhaps g lycine and GGG happened to concentrate i n a particular ti depool and this led to their association in the code. If so, the genetic code, i f it evolved agai n , would be u n l i kely to associate GGG with g lycine.*

Theories of the later stages i n the evolution of the genetic code are not described here since they are all so h i gh ly specu lative. Experi mental work o n this problem i s beg i n n i n g in several laboratories. Hopefully, we shal l soon be able to report some p ro gress.

Summary

One m ust i nvoke the principle of natural selection to explain how a b io logical system could have evolved from a p rebiotic sou p which contained only fami l ies of small organic molec u les and the random po lymers that could be made from them. It seems most l i kely that polymers related to nucleic acids were formed in the p rebiotic soup and coul d repli cate, even if only i naccu rately, under p rebi otic conditions. Natu ral selection would then have allowed the polymers that replicated most efficiently to win o ut.

The operation of natu ral selection would have favored those n u cleic aci d-l i ke polymers that could i mprove their

* There are more plausible reasons for believing that historical accidents determined the structure of the code, but explaining them would take us too far afield.


competitive position by making use of smal l molecules in the i r env iron ment. The amino acids were among the molecules used in th is way. The genetic code is the resu lt of an elabo rate series of adaptat ions by means of which polynuc leotides come to cont ro l their environment by d i recting the synthesis of structu ral and catalytically active polypeptides. V i rtually noth ing is known about the successive steps i n this adaptation. This is perhaps the most chal leng ing aspect of the problem of the or ig ins of l i fe .

Appendix to Chapter 1 0 - 0ptical Activity

Mi rror symmetry is fami l iar in everyday l ife. There is a sense in which a left-hand glove and a right-hand g love have the same shape. Nonetheless one can no.t be superimposed on the other, and they interact very differently with an object such as a right hand. Pai rs of objects l i ke this are said to be related as nonsuperimposable mi rror images. Many objects can be superimposed on their

Figure 1 0A.1 . Right- and left-hand gloves are mi rror i mages, but no amount of reorientation permits a right-hand glove to be superimposed on a left-hand g love.


m irror images -ten n is rackets, for example - but most complicated objects found in nature do not possess this poperty.

Molecu les, l ike other objects that have a well-defined spatial structure, may or may not be superimposable on their mirror i mages. It w i l l be obvious from what has been said above that large i rregu larly shaped molecules usually cannot be superimposed on

their mirror-images. The existence of pairs of molecules having the same kind of relation to each other as a right-hand and a left-hand glove has important consequences for organ ic chemistry.

Pasteur discovered that certain , apparently pure, organ ic chemical compounds are deposited from solutions as a mixtu re of two types of crystals. The shapes of the crystals are related as mirror i mages. Pasteur separated by hand the two forms of sodi u m ammonium tartrate, a complex organ ic salt, and showed that solutions of the separated compounds rotated the plane of polarization of l ight in opposite d irections. Any substance that rotates the plane of polarization of l ight is said to be optically active.

It would take us too far into physics to explain what is meant by the plane of polarization of l ight and how the rotation of the plane of polarization is measured. From our poi nt of view it is sufficient to recognize that the detection of optical activity always indicates the presence of a molecule or structure that cannot be superi mposed on its mirror image. The two m irror image forms of an optically active molecule are referred to as enantiomers. They are usually designated as o- and L-. Pasteur's experiment achieved the f i rst separation of optical enantiomers, salts of o- and L-tartaric acid.

The simplest optically active molecu les are those that include a carbon atom surrounded by four d issi milar groups at the apexes of a tetrahedron. The naturally occu rring amino aci ds (except g lyci ne) are of this type. Organic molecules in which each tetrahedral carbon atom is attached to at least two identical g roups (e.g. hydrogen atoms) are optically i nactive since they can always be

Figure 1 0A.2. o- and L-forms of sodium ammonium tartrate.


H I H2Nr.:::....C...,::::�CH3 \

\ COOH L- Alanine

H \ H3C �:::....t.....,:::�NH2 I

I HOOC 0 -Aianlne

Figure 1 0A.3. L- and o-alanine, showing that they are mirror images.

superimposed on their m irror images. G lycine, for example, is optically i nactive, because its central carbon atom is attached to two

hydrogen atoms. Solutions of organic material can be optically inactive for one

of two reasons. In some cases, al l molecu les in the solution may be in herently optically inactive, as in a solution of glycine. Alternatively, the solution may contain optical ly active molecu les but the o- and L-enantiomers may be present in equal amounts so that the rotation produced by the o-enantiomer just cancels out that produced by the L-enantiomer.

All chemical reactions between optically i nactive starting materials in solution give rise to optical ly inactive solutions of products. Sometimes, the product molecules are themselves individual ly inactive, but in the most interestin g reactions optically active o- and L-molecules are produced but in equal numbers. This latter result is easily understood. In a symmetrical environment there is no possible reason why L-molecules should be produced more or less often than o-molecules.

Optically active crystals can sometimes be obtained from op-

H I H2N, C�H

\ \ COOH

Glycine

Figure 10A.4. Unlike alanine, g lycine can be superimposed on its m irror i m age. If the rig hthand molecule is rotated so that the NH2 and COOH g roups take up the same orientations that they have in the left-hand molecule, the two hydrogens of the right-hand molecule wi l l coi ncide in position with those of the left-hand molecule.


tically i nactive solutions, but this does not contradict the result stated above. In such cases, o- and L-crystals are always obtained in roughly equal numbers. This is what Pasteur found when he carried out his classic i nvestigation of sod ium ammonium tartrate crystals.

In a s imi lar way molten inorgan ic substances that are themselves optically inactive sometimes sol idify to form optically active crystals. Optically active crystals of the abundant mi neral, quartz, are formed from an optically inactive melt of si l ica, for example. Whenever a careful count of left-handed and right-handed quartz crystals has been made, they have been shown to be present in almost equal n umbers.

These results lead to a conclusion that is very important in d iscussion of the orig ins of life. Whi le prebiotic synthesis may produce optically active molecu les, they always produce the o- and L-enantiomers in equal number. No net optical activity could have been produced by abiotic reactions on the primitive earth. Some authors have correctly pointed out that this resu lt is no longer strictly true when the magnetic field of the earth, the polarizatio·n of l ight from the sun, and other compl ications are taken into account. Most students of the field th ink that the importance of these factors has been exaggerated ; we shall suppose that the prebiotic sou p was optically inactive.

· Living organisms are un ique i n that they contain a large

nu mber of optically active constituents. Only L-amino acids are present in protei ns, for example, and only o-nucleotides in nucleic acids. Si nce the time of Pasteur, this remarkable feature of the composition of l iv ing things has fascinated biochemists and biolog ists. Pasteur hi mself wrote " I am incl ined to think that l ife, as it appears to us, must be a product of the dissymmetry of the un i verse . . . . " Any complete theory of the orig in of l ife must explain how organisms contain ing optically active biochemicals and polymers evolved from an optically inactive soup.

The solution to this problem is to be found by considering the i nteraction between pairs of optically active objects. We saw that although left-hand and right-hand gloves have the same shape, they interact differently with a right hand. However, each hand fits its own g love, and the two g loved hands that resu lt are mirror images. Simi larly, each hand fai ls to fit the wrong g love, and the resu lting misfits form mirror images.

This example i l l ustrates a general princi ple that is important in organ ic chemistry. When a pai r of optically active molecu les interact, they form one of a set of four possible combinations: if the two molecu les are A and B, the possible products are o-A-o-B, L-A L-B,


Figure 1 0A.5. Althoug a right-hand glove does not fit a left-hand glove and vice versa, the misfits obtained by putting a hand i nto the wrong g love are sti l l mi rror images.

D-A L-8, and L-A o-8. Just as with g loves and hands, o-A o-8 and L-A L-8 are m irror images and so are o-A L-8 and L-A o-8, but L-A L-8 and L-A o-8, for example, have qu ite d ifferent shapes. This argu ment can be extended to complicated systems containing many optically active components. If each component of a complex object is replaced by its mirror image, the mirror i mage of the object is obtained . If every molecule in the reader was repl aced by its mirror image, it would be possible to construct the m i rror i mage of the reader -with the heart on the wrong side, of course. However, if the conf igurations of some molecu les were changed, but not of others, the resu lt would be chaotic.

These argu ments suggest that self-contained chemical systems that are perfect mirror i mages of each other behave identical ly, whereas systems in which some but not all components are mirror images have qu ite different chemical properties. This conclusion is correct and can be proved qu ite rigorously. It fol lows, for example, that if a natu ral ly occu rring enzyme (made up of L-amino acids) is able to synthesize a o-nucleotide, it is certain that the correspond ing artifical enzyme made up of o-amino acids cou ld synthesize the L-nucleotide. On the other hand, a correspond ing polypeptide contain ing both o- and L-am ino acids would almost certa in ly have no enzyme activity.

Figure 1 0A.6. Mirror images and partial mi rror images.


We can now see that in tryin g to decide why l iving systems contain L-ami na acids, o-nucleotides. and so on, we are trying to solve two quite different k inds of prob lems at the same time. To make progress, these problems wi l l have to be separated. First. i t wi l l b e necessary t o explained why a l l the co mponents of a protein or nucleic acid have the same configuration. More generally, the choice of the relative configurations of all the co mponents of a l iving organism must be explained. Next one must elucidate why l iving organisms contain L-ami na acids and o-nucleotides, rather than o-amino acids and L-nucleotides, that is , why a single apparently arbitrary choice was made between the l iving systems that we are fami l iar with and their mirror images.

From what is al ready known about the interactions of optical ly active molecu les i t can be seen that there cou l d be no structural reason for select ing l iv ing organisms of one type of " handedness" rather than those of the other. It may, however, be anticipated that structural reasons wi l l be fo und to account for most, if not al l , of the choices of relative config u ration that were made in the cou rse of chemical evolution .

We do not yet understand what determined the choice of the relative configurations of al l the different constituents of cells, but in some cases plausible explanations can be given. For example, it is qu ite easy to see why o- and L-nucleotides cannot occur together i n the DNA structure. The reg ular double-helix composed of o-nucleotides spirals i n a clockwise d irection. If a DNA- l i ke molecule were synthesized from L-nucleotides, it wou l d differ from DNA only in that the structure wou ld spi ra l i n an anticlockwise d irection. However, a molecule contai n ing both o- and L-n ucleotides could not form a regu lar structure at al l , since the direction i n which the helix turns would b e constantly changing. An i rregu lar structure would have properties qu ite different from those of DNA, so the mechanism that leads to the repl ication of DNA could not operate. A comparison with a spira l stai rcase may be usefu l here. Rig ht-handed or left-handed spira l stai rcases are equally useful , but a stai rcase that was constantly changing its handedness wo uld not be usefu l , except to a rock cl imber. (See Figure 1 0A.7).

At the moment there exists no convincing argument that exp lains why al l amino acids in proteins have the same configuration , nor is it understood why L-amina acids, rather than o-am ino acids, are associated with o-nucleotides. When more is known about the mechan ism of protein synthesis, the answers to these questions should become clear. In the meantime, it may be supposed that, whenever, in the course of biochemical evolution, a new biochem-


Figure 1 0A.7. The consequences of cl imbing a spi ral staircase that c hanges from right-handed to left-handed direction of ascent.

ical compound became im portant, natural selection led to the choice of the enantiomer that functioned best.

To faci l itate the next part of this d iscussion it is conven ient to refer to contemporary l iving organisms as L-organisms (short for organisms with L-amino acids, o-nucleotides, and so on). We have to explain why the earth is populated entirely by L-organsims, whi le mirror i mage o-organlsms are conspicuously absent. We have seen that there is no chemical reason why L-organisms should be more efficient than thei r mirror i mages, nor is there any obvious reason why o- and L-organisms should not coexist.

I believe that the success of L-organisms was a matter of chance; the earth could equally well have been popu lated by oorganisms. Some authors correctly clai m that the influence of the earth 's magnetic field, the rotational motions of the earth or some similar factor could have biased the odds in favor of L-organisms. It is doubtful that any of the biases so far suggested was significant.

Consider two models of the evolution of life on the primitive earth. Fi rst suppose that the evolution of a self-repl icati ng system was a very i mprobable event. Then the fi rst self-repl icating system would have had ample time to eat u p the prebiotic soup before any competitor could have gotten started. Once the prebiotic soup was used up, there was no longer any possibi l ity of a second, independent origin of l ife. According to this model L-organisms inherited the earth because the fi rst organism happened to be an L-organism. The fi rst organism could equally well have been a o

organism ; in that case we wou ld now be wondering where the Lorganisms had gone.

Next consider a completely d ifferent model in which we suppose that, once the prebiotic soup became suff iciently concentrated, self-repl icating systems arose with h igh probabi lity. Then, roughly equal numbers of L-and o-organisms would have evolved on the earth and would have competed against each other. This situation is very l i ke the one that we consider in Chapter V when d iscussing the un iversality of the genetic code. L-organisms cou ld only have el iminated their competitors by "discovering" one or more new adaptions that gave them a very pronounced selective advantage and al lowed them to outg row their mi rror image competitors.

Perhaps we shall never know, as a matter of h istorical fact, just what happened three bi l l ion or more years ago to establish, once and for al l , the predominance of L-organisms, but the problem does not raise any new matters of princ ip le. The evolution of optical specificity was just part of the more general process of the establ ishment of biological order.


We saw in Chapter 5 that the universality of the genetic code forces us to ask how one type of biological organization came to win out over others that must have been very similar to it. The universality of L-organisms carried that q uestion one step further- how did one form or organization win out over another that had exactly the same biological potential ? Neither question can be answered in detai l , but neither raised metaphysical issues.


From Replicating Polymers

to Cells

The Evolution of Cells

Every l iv ing cel l is surrou nded by a selectively permeable membrane that functions, among other th ings, to hold together the components of the genetic apparatus, the biosynthet ic enzymes and the smal l molecules that are uti l i zed dur ing g rowth and divis ion. A s impler structure that played a role s im i lar to that of the cel l membrane m ust have evolved early in the development of l i fe.

Before d iscussing the development of the cel l , it is conven ient to characterize a number of stages lead i ng up to the evolution of self-contai ned biolog ical systems. In the very f i rst stage, ind ividual polymer molecules m ust have formed and repl icated at the expense of material that was already present in the p rebiotic soup. S ince there was, as yet, no requ i re ment for cooperation between molecules, a segregating mechanism could have performed no usefu l func-

1 69

tion. Repl ication must have occu rred i n free solution , on the su rface of rocks, or wherever else conditions happened to be favorable.

We bel ieve that in the next stage of chemical evolut ion, nucleic acids began to d i rect the synthesis of other polymers, inc lud ing polypeptides. As soon as this happened it became important to keep together whole fam i l ies of macromolecu les. At this point a structu re was req u i red that wou ld adsorb or enclose macromolecu les, but there was no need for a bag that was i m permeable to smal l mo lecules. A membrane might, if anyth ing, have been d isadvantageous, s ince it cou ld have excluded usefu l abiotic molecu les from the interior of the "cel l . "

I t was only when prim itive organisms began making biochemical compounds for themselves that it became i mportant to retain smal l molecules. The operation of natu ral selection leaves l ittle room for charity. Most of the advantage that could be derived from the i nvention of new biochemical syntheses wou ld have been lost if organisms had shared the benefits of thei r d iscoveries with their competitors. To prevent this from occu rri ng it was necessary to i nvent a membrane that was i m permeable to useful molecu les made with in the ce l l , but permeable to raw materials from outside. Such a membrane is said to be semi permeable.

Al l that is known about the repl ication of nucleic acids and the synthesis of p rote ins suggests that these processes evolved in a predominantly aqueous envi ron ment. When th inking about the origin of the cel l , therefore, one needs to consider on ly those segregation mechanisms by means of which fam i l ies of polymers are kept together in the presence of a large excess of water. We can propose several mechanisms of this kind that cou ld have operated before the evolut ion of membranes, but it is not certain which of them was important on the pri mitive earth. Two of the most plausible mechan isms wi l l be d iscussed : adso rption to mineral particles and adsorption to col loidal organic material .

Polymers stick very t ightly to the surfaces of a number of common mi nerals. Proteins and n ucleic acids are adsorbed strongly by the very common phosphate mineral, apatite; many organic su bstances, both small molecu les and


polymers , are adsorbed by various c lays. It seems possi b le that the very first "organ isms" consisted of noth ing more than small mi neral g rains to which "genetic systems," that is fami l ies of repl icating polymers, were adsorbed. "Reproduction" would have occu rred whenever a fam i ly of polymers managed to colonize a new grain. If this pictu re is correct, the most pr imitive organisms were more l i ke pi n cushions than bags.

Theories of this kind have one great attraction. It has always been thought that the surface of a mi neral m ig ht act as a primitive catalyst and help to br ing about the orderly polymerization of amino acids and nucleotides. If the same mineral cou ld both catalyze the formation of polymers and hold on to them once they had formed, it wo uld have provided an advantageo us s ite for further evol ution. We know of minerals that catalyze polypeptide fo rmation and of others that adso rb nucleic acids and prote ins, but so far we do not know of any m ineral that has all the properties that are req ui red to favor extensive biological evolution. It i s c learly i mportant to search systemati cally for such a mi neral.

An alternative and equal ly plausible theo ry suggests that cells evolved from some k ind of col lo idal particle, for example a coacervate of the type discussed b riefly in Chapter 9. So l i ttle is known about the composition of the prebiotic soup that it is not possib le to be precise about the chemica l natu re of this agg regate. Nonetheless, i t seems l i kely that col loidal droplets of one k ind or another wou ld have formed in time fro m any suff iciently concentrated and sufficiently complex m ixtu re of organ ic compounds. The cel l , accord ing to th is theory, arose from a microscopic droplet of "o i l " floati ng i n the prebiotic sou p.

A detai led proposal of this k ind involves thermal polypeptides. I f the polymers that are formed by heating a mixture of am ino acids at 1 30°C (Chapte r 9) are boi led with water, free-floati ng microspheres of organic material are formed. These mi crospheres are about the size of bacteria and they adsorb polynuc leotides and many othe r organic molecu les from aqueous solution. Dr. S idney Fox has sug gested that these microspheres are the precursors of l iving cells.

1 71 From Replicating Polymers to Cells

It is doubtfu l that thermal polypeptides formed the major com ponent of the matrix from which cel ls evolved. However, they do provide an i nteresti ng i l l ustration of the kind of structure that could have formed in the prebiotic soup. It seems l i kely that polypeptides and other polymers, no matter how they were formed, wou ld have stuck together to g ive s im i lar col loidal d roplets. Once formed, these d roplets could have col lected a variety of polymers, metal ions, organ ic phosphates, and other molecu les at thei r surfaces. Col loidal agg regates of this kind may indeed have played a part in the evolution of cel ls .

It is not known whether mineral partic les or col loidal organ ic droplets were im portant for the fi rst step in the evolution of the cel l . Nevertheless, in either case, the next step must have been the evol ution of a semi permeable membrane. Biological membranes are made up in large part of a c lass of organ ic molecu les cal led l i pi ds. Under some c i rcumstances, artif ic ial membranes are formed when l i pids are dispersed in water (Chapter 4). If l i pids had been present in the prebiotic soup, membranes cou ld perhaps have formed spontaneously; otherwise it is much harder to envisage the steps that could have led from a col loidal aggregate to a cel l surrou nded by a selectively permeable membrane.

The Evolution of Metabolism

The earl iest organisms g rew in the prebiotic soup by maki ng use of preformed molecu les, such as nucleotides and amino acids. Many modern organisms, on the other hand, synthes ize all biochemical compounds they need from a few very s imple starti ng materials. The bacteri u m E. coli, for example, req ui res on ly g lucose as a sou rce of carbon, whi le plants th rive on carbon d ioxide. This section wi l l describe the stages by which pri mitive organisms became independent of the prebiotic soup.

Livi ng organ isms could not have derived any advantage by synthesizi ng for themselves compounds that were sti l l freely avai lable in the prebiotic soup. However, as soon as the supply of essential biochemical compounds was used


up , alternative sources had to be fou nd. It was at this point that organisms learned to synthesize these compounds from simple starti ng materials. The sequences of synthetic reactions that are used by l iv ing organisms to convert s imple precurso rs to comp lex biochemical compou nds wi l l be referred to as b iosynthetic pathways. B iosynthetic pathways differ from prebiotic pathways because they ut i l ize enzymes. Conseq uently, every step in a biosynthetic pathway m ust proceed in so lution.

Some biochemical pathways may have d iffered l itt le from the corresponding prebiotic pathways; the fi rst enzymes may have helped along processes that were occurri ng inefficiently i n the absence of catalysts. However, this could not always have been the case, as the fo l lowi ng example wi l l show.

Phenylacetylene is read i ly formed from methane in the gas phase and reacts with aqueous ammonia and hyd rogen cyanide, under appropriate co nditions, to g ive the essential amino acid , phenylalanine. On the pri m itive earth it was probable that phenylacetylene formed i n the atmosphere and reacted i n oceans and lakes to form phenylalan ine. However, phenylacetylene can not easi ly be made in so lution , even with the help o f enzymes. When the su pply o f phenylalan ine ran out, it was necessary, therefore, to i nvent a completely new synthesis of phenylalanine that did not i nvolve pl;lenylacetylene.

I t is bel ieved that many biosynthetic pathways developed, paradoxical ly, by reversing the pathways of spontaneous decomposition. This diff icu lt idea is best i l l ustrated by an analogy from modern industry. The Israeli ai r force was at one time largely dependent on France for its su pply of f ighter ai rcraft. After the war of 1 967, France ceased to supply spare parts for these aircraft and Israel was forced to manufacture its own replacements . It has been claimed that it was the need to manufactu re an increasingly wide range of spare parts that provided the incentive for the establ ishment of an autonomous a ircraft industry in Israel.

This example i l l ustrates a general princ ip le : when a comp lex product of techno logy that has previously been suppl ied from outside is in short supply, it is usual ly easier


to improvise the repair of damaged models than to develop a complete production l ine. It i s al most always possib le to improvise repai rs, whereas the man ufacture of a product from raw materials is a majo r undertaking that requ ires extensive plann ing . Once enough repai r capacity is avai lab le, i t may become easier to develop a complete production l i ne.

A s imi lar idea is important i n the context of biochemical evolution , because natural selection must achieve its resu lts without planning. The development of a biosynthetic pathway that involves several steps is i l l ustrated in Figure 1 1 . 1 When the supply of a co mpound B runs out, an ample supply of its fi rst decomposition prod uct 01 wi l l often be avai lable. The fi rst enzyme to evolve (Ed wil l therefore convert 01 to B. In t ime 01 wi l l be used up and i t wi l l be necessary to use 02, the next decomposition product , as a starting point for the synthesis of B . However, since the enzyme E1 is a l ready avai lable, i t is on ly necessary to evolve a sing le enzyme E2 to convert 02 to 01 ; then E1 would f in ish the synthesis by convert ing 01 to B. In a s im i lar way new steps E3, E4 , and E5 cou ld be added u nti l it became possib le to convert some very abu ndant precu rsor P through 04, 03, 02, and 01 to B. The evolution of the pathway woul d then be complete.

P - 04 - 03 - 02 - 01 - B Spontaneous decomposition

01 B 1•' synthetic enzyme develops

02 __§_.,.. 01 B 2"d synthetic enzyme develops

03 � 02 01 __§..,. B an� synthetic enzyme develops

04 ____.§__, 03 02 __§_.,.. 01 __§..,. B 4th synthetic enzyme develops

P � 04 ____.§__, 03 02 _§..,. 01 � B 5th synthetic enzyme develops

P 04 B Useless

P -----=,___;:::...:....:::3..:...::::....:....::!---- B Impossible

Figure 1 1 .1 . The top l ine shows the hypothetical cou rse of decomposition of an important biochemical B. The next five l ines show the "rescue" of successive decomposition products. The last two l ines show that partial synthesis of D4. from P is useless, while the d irect route from P to B could never evolve.


It is i mportant to notice that, i n this scheme, each enzyme, as soon as it evolves, performs a usefu l function . There is no corresponding way of evo lvin g a pathway from P to B i n the forward d i rection, if the transformation cannot be achieved i n a sing le step. An enzyme , for example �. could never be selected alone, because it wou ld be useless without the other enzymes of the pathway. The chance of several enzymes evolving sim ultaneously is neg l ig i ble.

The next phase of biochemical evolution m ust have occurred when even the most abundant com ponents of the prebiotic soup were exhausted. It then became necessary to derive new, water-soluble organic materials from the gaseous components of the atmosphere; the photosynthetic fixation of carbon dioxide (Chapter 4) m ust have evolved at this time. Afterwards photosynthesis became essential for the continuation of l ife on earth. Organi c compounds were constantly being oxidized to carbon dioxide and it was on ly through p hotosynthesis that the supply of organi c carbon compounds cou ld be replen ished. It is l i kely that, at a sti l l l ater stage, n itrogen-fix ing organisms evolved, i n response to a shortage of ammonia in the environment.

The evolution of cells that were surrounded by sem i permeable membranes and were capable of carry ing out a wide range of biosynthetic reactions marked the final stage in the evo lut ion of l ife. The fossi l evidence suggests that it occurred on the prim itive earth at least three b i l lion years ago. The evo lution of m u lt ice l l u lar o rganisms was sti l l to take more than two b i l l ion years, wh i le creatu res with hard shel ls d id not become abundant u nt i l half a b i l lion years ago. Something is known and much more has been g uessed about the h istory of l ife d u ring the long period of specia l ization that marked the transition from primitive cells to complex modern o rganisms, but this material falls outside the scope of the p resent d iscussion . A n um ber of the books recommended for further reading deal with the evolution of h igher forms of l ife.


Natural Selection

I ntroduction

The evi dence for evolution derived, for example, fro m the foss i l record is now overwhel m ing. I n the last decade or so, stud ies of protei n sequences have further confi rmed the tradit ional evolution ists' deductions in remarkable detai l . Biochemists have shown that the d i fference between corresponding proteins of d ifferent species are proportional to the evolutionary d istance between the species. The sequence of amino acids in the hemoglobin of the great apes is a lmost i ndisti ngu ishable from that of h uman hemoglobin . On the other hand , there are substantial d i fferences between human hemoglobin and that of less c losely related mammals, such as m ice or rabb its. The hemoglobin of b i rds is even more different fro m that of man.

I t is possib le to write computer programs that examine the sequences of proteins fro m d i fferent spec ies and deduce

1 77

the evol utionary relation between the species . No information i s g iven to the computer except the amino acid sequences, but the computer prints out a fam i ly tree that matches c losely the one deduced by com parative anatomists and palaeontologists. During the ni neteenth centu ry it was suggested that God had "p lanted " the foss i l record to deceive wicked b io logists ; a supporter of that point of view wou ld now have to suppose that God had also "fixed" the sequences of tens of thousands of proteins in hundreds of thousands of species, for no better reason.

The fact that evol ution has occurred does not te l l us anyth ing d i rectly about the mechanisms that have operated. The work of the pre-Darwi n ian evolutionists should n ot be underestimated. They correctly recognized evo lution as an alternative to special creation. The i r fai lu re to arrive at the correct mechanism is a meas u re of the or ig inal ity of Darwin and Wal lace.

Lamarck bel ieved that animals could pass on to thei r descendants the characte ristics that they had acq u i red d u ring thei r own l ifeti me. The ancestors of the g i raffe, for example, benefited from the advantage of longer necks and somehow i nfl uenced the lengths of the necks of the i r c h i ldren. This idea may not be attractive today but i t was almost i nevitable in the context of pre-Darwinian thought.

The world i s fu l l of animals that are obviously adapted to thei r own particular way of l i fe. If no one created them, how could they be so wel l-adapted ? Before Darwin, it was natu ral to thi nk that someone or someth i n g had d i rected adaptation. I f i t wasn't an external God, something i nternal was the only alternative. This led some Lamarckians to concepts connected with the "striv ing" of species for i mprovement. Just as physicists invented a nonexistent ether because their common sense view of wave motion was i nadequate, so b iolog ists i nvented the i nhe ritance of acq u i red c haracteristics because their common sense ideas on adaptation were inadequate.

It may be worthwhi le to d i g ress at this poi nt to say someth ing more about the i nheritance of acq u i red characteristics. Although the theory h ad mystical i m pl i cations there is nothing mystical about the theory itself- it j ust does not apply to the evol ution of terrestrial p lants or an imals. It is poss i b le to cut the t ips off the tai ls of mice for a few gener�-


30

1 3 29

1 1

1 5

1 1 9

Figure 1 2.1. The relationship between species as determined from sequences of cytochrome C. Note how the computer program places species that are known on other g rounds to be closely related close together in the fam i ly tree. Species marked with an asterisk are m icroorgan isms. (Reproduced with permission from W. M. Fitch and E. Markowitz, Biochem. Genetics, 4, 579, (1 970). )

tions and then to exam ine the young to see whether they are born with shorter tai ls. I n fact they sti l l have normal tai ls, but if they did not, theo retical b io logists wou ld not be at a loss for nonmyst ical explanations.* (This exam ple does not do justice to Lamarck's ideas about evol ut ion it was chosen on ly to show that the theory of the i n heritance of acqu i red characteristics is not r id icu lous.)

We have seen that the idea of natu ral selection is a very s imp le one and that it completely e l im inates the need to postu late any i nternal or external "wi l l " that di rects evolution. Sma l l heritab le vari at ions arise i n i ndividuals by chance; if they are d isadvantageous they are e l im inated, but if they are advantageous they enable the descendants of the fortunate i ndividuals to outgrow their com petitors. Adaptat ion is the conseq uence of the accumu lation of many advantageous variations. ·

Let us now i mag ine the Darwin ian explanation of the length of the g i raffe's neck. Darwi n ians would c la im that, had you measu red the necks of the creatu res from which g i raffes evolved, you would have found that some were shorter than others. You wou ld also have noted that longnecked parents tended to have long-necked ch i ldren , independently of their environment. However, i n env i ronments where long-necked individuals were at an advantage, say where the lower vegetation had been eaten by some other species, tal l wel l -fed parents would have had more ch i ldren that reached reproductive age than shorter ani mals. Needless . to say their ch i ldren wo uld have had longer-thanaverage necks - selection wo uld have begu n . The repetition of such selection, generation after generation, produced the g i raffe's neck.t

Accord ing to Lamarck, evol ution proceeds because the average neck length of the ch i ldren of a given pair of animals is i nf luenced by the parents' experience. Acco rding to Darwi n , evol ution proceeds because long-necked ani mals

• The invention of a molecular mechanism that permits the Lamarckian evolution of tail length in mice is left as an exercise for the advanced student.

t A more plausi ble but less graphic theory claims that the giraffe's legs and neck evolved to allow the animal to escape from lions.


have more of those ch i ldren who, because of their longer necks , can compete effic iently for food and so su rvive to reproductive age. This is not a metaphysical d isti nction : i n pr inciple i t wo u ld o n l y b e necessary to measure necklengths i n a few fami l ies of evolv ing g i raffes to show that Darwin 's theory is correct. In pract ice this k ind of experiment is rarely if ever possible, and we are forced to rely on less d i rect evi dence.

Is There a Paradox about Evolution?

In the m id-1 970s it is d i ff icu lt to understand just how revol utionary the theory of evo lution by natu ral selection seemed in the mid-nineteenth centu ry. Today, to many biologists, the law of natural selection seems almost a tautology - at best it is a piece of theoretical b io logy that could be worked out by anyone able to m ult ip ly. One shou ld not be deceived - it was a very hard discovery to make.

Darwin and Wal lace are rig htly regarded as among the g reatest scientific i n novators because they brought about a conceptual revolut ion. Darwi n 's sol ut ion to the central problem of biology was far from obvious even to h is most bri l l i ant supporters, u nt i l they had d igested the accumulated evidence. T. H . Hux ley was not completel y convi nced about natu ral se lection unti l he read the Origin of Species. When he had f in ished he is reported to have exclai med, " How exceedingly stu pi d not to have thought of that."

Darwi n's wo rk provided the so l ution to the central problem of b io logy- it showed how complex wel l -adapted organisms evolve from organisms that are s impler and less wel l-adapted. Recent experi mental studies fu l ly justify his bel ief that the variat ions that make selection possi b le are the conseq uences of random events. Why do so many phi losophers c la im that there is a paradox about evolution ?

The supposed paradox is concerned with design. It is very hard to avoi d us ing wo rds that suggest purpose when describ ing the wonderful ly adapted structu res that occ u r i n the l iv ing world. I t i s very tempting to describe the r ibosome, for exam ple, as an elaborate structu re des igned to carry out

1 81 Natural Selection

protei n synthesis. Why are we tempted to use these " l oaded" words when describing what. after a l l , are only the consequences of errors i n the process of nucleic aci d repl icatio n ?

One view, t h e o n ly correct one, poi nts out that when bi ologists m ake statements t h at seem to i nvolve desi g n or p u rpose, they are using convenient abbreviations. W h en they say, "the ribosome was des ig ned to c arry out protein synthesis," they are avo i d i ng long expl anations. We wou ld h ave to say, "early i n the evol ution of l ife, a nucleic acid that could direct the synthesis of i nformed polypeptides wou l d have had a selective advantage over . . . . " Before w e h ad f in ished, we wou l d h ave rewritten m u ch of Chapter 1 0. Perhaps we should agree to use the verb "to natu ral ly select" in such situations. The ph rase "the ribosome was naturally selected to carry out protein synthesis" may be inelegant, but it expresses what most biologists mean q u ite precisely.*

Although this i s the formally correct answer, it wi l l not _ sati sfy everyone. It is gen ui nely surprising that an organ ism that has evolved by random m utation and selecti on appears to be desi g ned. The idea is contrary to i ntuition. It is true, but to many it w i l l a lways remain ridiculous.

Several of the greatest discoveries i n science are counte ri ntuitive. How many people feel that the earth i s movi n g a n d t h e sun is a t rest. o r that objects conti n u e i n u n iform motion unti l somethi n g h appens to the m ? I suspect that most people have n ot assi m i l ated the f i rst idea i ntu itively. As you watch the sunset, do you feel the sun s inking i n the West o r you rself spinning away f rom it? The second idea feels more and more reasonable as the space program advances - yo u really do need a retro-rocket to bri n g the astronauts down. Evolution by natural selectio n is perhaps the hardest of the counterintuitive i deas that we are asked to accept. It is not o n ly surprising, but, to some people, down right disagreeable.

Only experience leads to the assimi l ation of correct, cou nterintuitive argu ments. One comes to accept them because one cannot do without them. If you read a lot about

• Darwin h imself accepted this usage.


r ibosomes or b utterf l ies and thi n k hard enough about the way they came to be as they are, you wi l l probably f ind that you are using the i dea of natu ra l selection , without notic ing i t - alternatively, you may g ive up and become a mystic. I t seems to me even more i mportant to consider the o rig ins of l ife i n terms of natural selectio n ; otherwise there is no way of approach ing some of the most i nteresti ng aspects of the su bject.

Once one gets used to the i dea of natural selection, one f inds it helpfu l i n th ink ing about the development of many systems other than l iv ing organisms. One should not underestimate the i m portance of trial and e rror in the development of tec hnology, for example. The develo pment of the revolve r is an object lesson in evolution. Those c lumsy guns with revolvi ng barrels are the d inosaurs: there were many small successes and many g reat fai l u res on the way to the Peacemaker, and for every company that presently makes revolvers, there m ust be ten that have been e l im inated.

There is no paradox about evo l ution by natural selection. The i dea is a subtle one and it takes a l ittle t ime to get used to it. No doubt that there is m uch more to be learned about the mechanism of evo lut ion, but most b io logists wou ld agree that two things are certain - biological adaptation is made poss i b le by n atu ral select ion, and a l l evol ut ion depends on random events. The replacement of "wi l l" by "chance" as the mediator of b io logical change has transformed o u r view of man's relat ion to the rest of the U niverse. For better or worse, that transformation is un li kely to be reversed.

In a precisely paral le l way we know that o n ly processes i nvolv ing natural selection cou ld have led to the development of l i fe on earth. We have even more to learn about the or ig in of l ife than about the or ig in of species. We cannot even be q uite sure that molecu les l ike n ucleic acids were the f i rst materials on which selection acted. Nonethel ess, I th ink we are j ustified i n rejectin g any theory of the origins of l ife that fai ls to explain at least i n pr inci ple, how natural select ion cou ld have led to the development of b io logica l comp lexity.

1 83 Natural Selection

PART

THREE

What is Life?

Introduction

In Chapters 1 5 and 1 6 a set of related problems concerned with the existence of l ife on other planets wi l l be d iscussed. Does l ife exist elsewhere in the solar system? Could l ife on another planet be based on si l icon compounds rather than on carbon ? Are there i ntel l i gent societies elsewhere in the universe? Problems of this kind i nevitably raise questions about the natu re of l ife. In this chapter this age-o ld question wi l l be taken up .

The word " l ife" has many d i fferent but overlapping meanings that have developed i ndependently of systematic biology. Hence, it is not wel l-adapted to use in techn ica l , scientific d iscussions. Unfortunately, there are no precisely defi ned Eng l ish expressions to substitute for it ; we cannot d iscuss l ife on other worlds without saying someth ing about " l ife. "

1 87

The fol lowing exam ples i l l ustrate the k inds of confusion that can arise. As far as the layman is concerned , a horse is either dead or al ive, and that is the end of the matter. The bio log ist's point of view is more complex. Of course he wou ld not deny the fundamental difference between a dead horse and a l ive horse but he wou ld regard the former as sti l l part of the l iv ing world , as long as most of its ce l ls were fu nction ing normal ly. In a qu ite d i fferent context, many elementary introductions to bio logy state that al l l iv ing things must be able to reproduce, but, if i nterpreted l iteral ly, this req u i rement wou ld "k i l l " the m u les, s ince mu les are steri le. Even biologists themselves are l iable to argue about vi ruses, si nce vi ruses can reproduce, but on ly with the help of a l iving host. We shal l see that these d ifficu lt ies arise because many of the q uestions that are asked about the nature of l i fe are not scientific, but are concerned with Eng lish usage.

Today most scientists would agree that "What is l i fe?" is not a good scientific question. Scientists l i ke to deal with q uestions that can be answered by experi ment in an impersonal way. "What is l ife?" is not a question of this k ind , for it has someth ing in common with questions l i ke "What is freedom?" It is qu ite possi ble for two microbiolog ists to reach identical conclusions about the structure and function of a virus, for example, and yet to d isagree as to whether it is l iving or not. S im i lar d iff icu lties are l i kely to arise when we try to decide whether a society is free or not.

To help the reader to d isti ng u ish more clearly between the scientific and the semantic issues involved , consider the fo l lowing hypotheti cal situation . Su ppose we came upon a strange "organ ized body" on Mars and had to g ive an account of it to our scientif ic col leagues. We wou ld fi rst examine the structure of the object and how it behaved. Once we cou ld describe the object and its behavior in complete detai l , we wou ld have f in ished our work as "b io log ists ." If the object were a very strange one, we mi ght f ind ou rselves trying to decide whether or not to cal l it l iv ing. In that case, we wou ld not be carrying on a normal scientific enquiry, but trying to settle the way in which the Engl ish words " l ivi ng" and " non l iv ing" shou ld be used i n a novel context.

This suggests that when we discuss the forms that l ife might take on other planets, we should concentrate on the

1 88 Extraterrestrial Life

structure and behavior of strange objects rather than the ir status as " l iv ing" or "non l iv ing" bei ngs. Let us start by examin ing the attributes of terrestrial l iv ing organisms. This wi l l lead us to characterize a broad c lass of objects that inc ludes al l those that are considered al ive on intu itive g rou nds. However, th is class also inc l udes a number of s imp ler, se lf-reproducing objects that wo u ld p uzzle the layman because he would not know whether to cal l them l iving or not. For reasons that wi l l beco me c lear al l s u ch objects wil l be referred to as CITROENS.

Terrestri al Biology

Most elementary i ntroductions to b io logy contai n a section on the nature of l i fe. It is usual in such d iscussions to l i st a n u m ber of properties that d istingu ish l iv ing f rom non l iving th ings. Reproduction and metabo l ism, for example, appear i n all of the l ists ; the abi l ity to respond to the env i ron ment i s another old favorite. T h i s approach extends somewhat the chef's defi n ition "If it qu ivers, it's al ive." Of course, there are also many characteristics that are restricted to the l iv ing world but are not common to al l forms of l ife. P lants cannot pursue the i r food; ani mals do not carry out photosynthes is ; lowly organisms do not behave i nte l l igently.

It is possib le to make a more fu ndamental d i sti nction between l iv ing and non l iv ing th ings by exami n ing their molecular structu re and molecular behavior. In brief, l iv ing organisms are d isti ngu ished by thei r specified co mplexity. * Crystals are usual ly taken as the prototypes of s imp le , wel l specified structures, because they consist of a very large n u m ber of i dentical molecu les packed together in a un iform way. Lu mps of granite or random m ixtu res of polymers are exam ples of structu res which are complex but not specified. The crysta ls fai l to qual ify as l iv ing because they lack complexity; the mixtures of polymers fai l to qua l ify because they lack specif icity.

• It is impossible to find a sim ple catch ph rase to capture this complex idea. "Specified and, therefore, repetitive complexity" gets a little closer (see

later).

1 89 What Is Life?

These vague i deas can be made more precise by introducing the idea of i nformation. Roughly speaking, the information content of a structure is the m in imum number of instructions needed to specify the structure. One can see intuitively that many i nstructions are needed to specify a comp lex structure. On the other hand, a si mple repeating structure can be specified in rather few instructions. Comp lex but random structures, by defi nition , need hardly be specif ied at al l .

These differences are made clear by the fo l lowing example. Suppose a chemist ag reed to synthesize anyth ing that could describe accu rately to h im . How many instructions would he need to make a crystal , a m ixtu re of random DNA- l ike polymers or the DNA of the bacteri um E. coli?

To describe the crystal we had i n m ind , we wou ld need to specify which substance we wanted and the way in which the mo lecu les were to be packed together i n the crystal. The fi rst req u i rement could be conveyed i n a short sentence. The second would be almost as brief, because we could describe how we wanted the f i rst few molecules packed together, and then say "and keep on doing the same. " Structural i nformation has to be g iven on ly once because the crystal is reg u lar.

I t wou ld be almost as easy to tel l the chem ist how to make a m ixtu re of random DNA- l ike polymers. We would fi rst specify the proportion of each of the four nucleotides in the mixtu re. Then, we would say, "Mix the nucleotides in the req u i red proportions, choose nucleotide molecu les at random from the m ixtu re, and jo in them together in the order you f ind them." In this way the chemist would be sure to make polymers with the specif ied composition , but the sequences wou ld be random.

It is q u ite i mpossible to produce a corresponding si mple set of i nstructions that wou ld enable the chemist to synthesize the DNA of E. coli. In this case, the sequence matters; on ly by specifyi ng the sequence letter-by-letter (about 4,000,000 i nstructions) cou ld we te l l the chem ist what we wanted h im to make. The synthetic chemist wou ld need a boo k of instructions rather than a few short sentences.

It is i mportant to notice that each polymer molecule in a random m ixtu re has a sequence j ust as defi n ite as that of E.


coli DNA. However, in a random mixture the sequences are not specified , whereas in E. coli, the DNA sequence is crucial. Two random mixtu res contain q u ite d ifferent polymer sequences, but the DNA sequences in two E. coli ce l ls are identical because they are specified. The polymer sequences are complex but random ; although E. coli DNA is also complex, it is specified in a un ique way.

The structure of DNA has been emphasized here, but s imi lar arguments wou ld apply to other polymeric materials. The protein molecu les in a cel l are not a random mixtu re of polypeptides ; a l l of the many hemoglobin molecu les in the oxygen-carrying blood cel ls, for example, have the same sequence. By contrast, the chance of gett ing even two identical sequences 1 00 amino acids long i n a sample of random polypeptides is neg l ig ib le. Agai n , sequence i nformation can serve to disti ngu ish the contents of l iv ing cells f rom random m ixtu res of organic polymers.

When we come to consider the most i mportant functions of l iving matter, we agai n f ind that they are most easi ly d ifferentiated from inorganic processes at the molecu lar level . Cel l division, as seen under the microscope, does not appear very d ifferent from a nu mber of processes that are known to occur in co l lo idal solutions. However, at the molecular level the d ifferences are unmistakable: cell division is preceded by the repl i cation of the cel lu lar DNA. It is this genetic copyi ng process that d isti nguishes most c learly between the molecular behavior of l ivi ng organisms and that of non l iving systems. In bio logical processes the number of i nformation-rich polymers is increased du ri ng g rowth ; when col loidal d roplets "d ivide" they just break up i nto smal ler droplets.

CITROENS

We have seen that it is d ifficu lt, if not i mpossib le, to f ind a r ig id defi nit ion of l ife that incorporates al l of our i ntuitive ideas. Instead , we can make a l ist of the attributes that help us to decide whether or not a system is l ivi ng - reproduction ,

1 91 What Is Life?

metabol ism, excitabi l ity, and so on - and ag ree to cal l an organism al ive if it possesses a su itable selection of these attri butes. This is a usefu l approach i n introductory d iscussions of terrestrial b iology, but it is not so usefu l when one d iscusses al ien forms of l ife. I n the latter case, i t is too d ifficu lt to complete the l ist ; it is impossible to enumerate al l the types of behavior that m ight characterize nonterrestria l forms of l ife.

,

The d iscussion in the previous section suggests a more usefu l proced ure. When we talk about l ife on other p lanets, we are not th ink ing about s imple objects that behave in a s imple way. The structu re and behavior of an object wou ld need to be nonrandom and reasonably compl icated to interest the student of extraterrestria l l ife. We have al ready seen that a great deal of information is needed to specify the structu re of a compl icated nonrandom object. It may be conclu ded that anything that we would want to ca l l " l ivi ng" wou ld have a high information content. This apparently s imple req u i rement has far-reach ing consequences.

It fol lows i mmed iately that any " l iv ing" system m ust come into existence either as a consequence of a long evolut ionary process or a m i racle. It can be shown that the probabi l ity of · a stru ctu re arisi ng spontaneously decreases very rapid ly as the information content of the structu re increases. A ch impanzee sitting at a typewriter m ight easi ly compose a word of Shakespeare, but wou ld be un l i kely to compose a l ine ; if a ch impanzee wrote a whole play, the event could legitimately be cal led a m i racle. In a s imi lar way, the formation of a structure complex enough to be cal led " l iving" , in a s ingle event, wou ld be a m i racle.

S ince, as scientists, we must not postulate mi racles we must suppose that the appearance of " l ife" is necessari ly preceded by a period of evo lution. At f i rst, rep l icati ng structu res are formed that have low but non-zero i nformation content. Natural selection leads to the development of a series of structures of increasing complexity and information content, unt i l one is formed which we are prepared to cal l " l iving" .

Th is conclusion, i n turn, has i mportant consequences. Si nce selection cannot occur without reproduction, no


system of " l iv ing" organ isms c an evolve except f ro m p ri mitive repli cat i n g structu res.* This does not mean that every advanced o rganism mu st be able to rep ro d u ce, b ut the persistence of " life" req u i res t h at some of them can. Every " l i vi n g " creature m ust be the descendant of other related creatu res.

Final ly, dur ing reproduction, the i n fo rmation that spec ifies the stru cture of the parent m ust be passed on to each c h i l d ; otherw ise the c h i ldren wou l d not resemble the parents. The transm issi on of this genet i c i nformation m ust be reasonably accu rate, for if too many errors occu rred the chi ldren wou ld not be able to s u rvive. I t fol lows that genetic i nformation m u st be stored in a stru ct ure that i s stable t h roughout the reproductive l ifet i me of the parent.

These considerations lead us to the fol l owi n g req u i rements that a re necessary and s uffi cient to q ualify a struct u re as "al ive " :

1 . T h e object i s com plex a n d yet well-specified. 2. The object is able to rep roduce. t

These conditions, as we have seen, i m p ly that:

(a) the object is a prod uct of natu ral select ion.t (b) t h e i nformatio n needed to specify the object is

stored in a struct u re that i s stab le fo r the reproductive l ifet ime of the object.

A new term fo r such " l iv ing" o rganisms, whether terrestrial or not, m u st now be i ntro duced. They are Co mplex Information-Transformi ng Reproducing Objects that Evolve by Nat u ral Selectio n - CITROENS. (NOTE t h i s n o u n h as no s i ng u lar form - o n e cannot have a citroen , only a fami ly of CITROENS).

Any o bjects that we wo u l d i nt uit ively call al ive would

• Such structu res are not necessarily based on stable molecules. although these structures are easier to th ink about than any others (see below). The stable structures could conceivably consist of patterns of chemical reactions, for example. t Alternatively, the object may be the descendant of related objects that can reprod uce, even if it is itself "steri le." t Unless it is a product of human technology-see p. 1 96.

193 What Is Life?

have to be CITROENS. However, DNA- l ike molecu les that were capable of replicati ng , say on the surface of a catalytic m ineral , wou ld also be exam ples of CITROENS. All CITROENS are interesti ng because they can evolve by natu ral selection ; even if they are themselves re latively s imple they are able, under su itable conditions, to generate systems of great complexity. Any inte l l igent form of l ife on any p lanet must have evolved in the past from a system of s imple CITROENS.

Novel CITROENS

These ideas lead us natural ly to a useful c lassifi cation of possible types of biological organ ization. If the i nformation content of a fam i ly of CITROENS is stable, it must be stored i n some stable structu re. We are fami l iar with on ly one type of stable "genetic memory", the n ucleic acid system. The l iterature of science fiction abounds with alternative "geneticmemory banks". A nu mber of poss ib i l i ties are suggested by computer technology; a genetic memory m ight be based, for exam ple, on a system of electric cu rrents in a superconductor, or on the pattern of magnetization of a magnetic tape. Another favorite theme is that of submicroscopic "nuclear l ife" , based on some stable reprod ucing col lection of protons, mesons, neutrons, and other fundamental particles. Hoyle has described a l iving organism that has a continuous cloud-l i ke structure.

The reader should appreciate that these ideas, and many others l i ke them, remai n in the f ield of science fiction. It is not c lear that there are any stable, self-reproducing "genetic-memory banks," except those based on stable chemical structu res l ike nucleic acids. On the other hand, we can not prove that they are impossib le. Maybe, on so me other planet, the local scientists, while conceding the possib i l ity of genetic systems based on molecu les l i ke nuc leic acids, doubt that they would real ly work.

It w i l l be left as an exercise for the reader to work out the detai ls of a few of the more bizarre systems. Alternatively, he m ight l i ke to read the science fiction referred to i n


the b ib l iography. Let us now d i rect our ideas toward earthl i ke CITROENS, by restricting further d iscussion to " geneticmemory ban ks" based on stable chemical structu res.

For a polymeric molecu le to act as a genetic materia l , i t must b e stable th roughout the l ife of the o rganism and must also be able to d i rect the synthesis of identical or nearly i dentical geneti c molecu les fro m s ubstances in its envi ronment. We are fam i l iar with a single example - the nuc le ic acids - but q u ite different systems may be possib le." I can see no reason why genetic systems based on materials that are stable at h ig her temperatu res, perhaps s i l i cates, should not function on planets as hot as Mercu ry. S im i larly, materials that would react explos ively under terrest rial condit ions might form the basi s for a stable genet ic system operating at very low temperatu res. We are not yet in a posit ion to say whether high-temperatu re CITROENS cou ld be constructed from derivatives of s i l icon or boron , for example, or if lowtemperature CITROENS could evolve and thri ve at very low temperatures in an ocean of l iqu id hyd rogen.

I t i s on ly when we come to CITROENS that are based on carbon chemistry and that survive in an aqueous envi ronment, that we are on more fam i l iar g rou nd. Even then, there i s no reason to bel ieve that an i n dependent form of l ife of th is k ind wou ld resemble terrestr ia l l ife very closely. It i s not c lear that a l l genetic systems need be based on polymers contai n ing phosphorus, for exam ple, or that a l l genetic mo lecu les need to contai n n i t rogen i n ad dit ion to carbon , hydrogen , and oxygen. Perhaps, a s i n g l e fami ly of macromolecules might act both as the geneti c structu re and the enzymatic machinery of nonterrestr ial CITROENS.

We do, however, have experimenta l evidence that tel l s u s someth ing about the probabl e composit ion of earth-l i ke CITROENS. We know that the range of organic compounds formed abu ndantly u nder prebiotic co ndit ions is qu ite narrow. It seems probable that earth-l i ke l ife, if it exists elsewhere, wi l l l ikely be based on the kinds of com pound that a re formed i n reduc ing planetary atmospheres. I n the l ight of the evidence presented in Chapters 6-8, it wou ld be sur-

• One such system is discussed in detai l in the book by Cairns-Smith cited in the bibliography.

1 95 What is Life?

prisi ng, therefore, if am i no acids and sugars fai led to p lay some role i n the chemistry of other earth- l ike forms of l ife.

It would be rash to d raw more far-reach ing conclusions. It is always tempting to bel ieve that the detai ls of terrestrial biochem istry are the i nevitable consequences of the laws of chemistry and physics and depend l itt le on h istorical accident. This temptation should be resisted . We can safely predict that if we ever encounter extraterrestrial CITROENS, they wi l l have many u nearthly features.

The Products of Human Technology

Wil l iam Paley (1 743-1 805) in h is Natural Theology, or the Evidence of the Existence and Attributes of the Deity from the Appearance of Nature drew attention to a subtle and interest ing problem. Paley fi rst remarks that if we d iscovered a watch lying on the ground, we would not question that it had a maker. We co u ld not bel ieve that a compl icated object so obviously adapted to tel l i ng the ti me is the product of the b l ind forces of inorganic natu re. Paley then argues, by analogy, that the l iv ing world must also have a maker, God, s ince it shows even g reater evidence of design .

We have al ready seen that the operation of natural selection, a completely random process, leads to the evo lut ion of organisms that do i ndeed seem to have been designed. Paley made the mistake of th ink ing that there is a single way i n which complex, wel l -adapted objects came into existence - creation. In fact there are two that we now know about- natural selectio n , and fabrication by man. However, Paley was r ight to emphasize the need for special explanations of the existence of objects with h igh i nformation content, for they cannot be formed in no nevo lutionary, i n organic processes.

We are fami l iar with many prod u cts of technology that fai l to be CITROENS on ly because they do not reproduce autonomously. There is, thus, an exception to the rule that objects of high information content m ust be the d i rect product of natu ral selection ; they may be the products of human ingenuity. This exception does not weaken the argu ment,


since the i nte l l igent "creators" i n th is case are themselves the products of natu ral selection.

The creation of l ife i n the laboratory is a subject that fascinates the layman. If we are prepared to count the s implest CITROENS as al ive, I th ink that the creation of totally new forms of l ife wi l l be possible in the future, perhaps with in a hundred years. Of cou rse, any man-made CITROENS wi l l be far removed from the l itt le g reen men of the more s implemi nded science fi ction.

1 97 What is Life?

Extraterrestrial Organic Chemistry

The Murchison Meteorite

Many sol id objects fal l from space into the earth 's atmosphere each year. The heat generated as they are slowed down by the frictional resistance of the a ir is sufficient to make them white hot. The smal ler objects are vaporized completely i n the atmosphere , but so me of the largest ones reach the surface of the earth. Incandescent objects that vaporize co mplete ly in the atmosphere are cal led meteo rs (shooti ng stars) ; objects that reach the su rface of the earth are cal led meteorites.

Large meteo rites are qu i te uncommon ; it is est imated that objects with d iameters much greater than a meter (a yard) reach the su rface of the earth only about once in ten years. A number of large craters formed by the i mpact of meteo rites are known ; they show that a few massive objects must have struck the earth long ago. I n more recent ti mes a

1 99

meteorite that fel l i n Siberia caused an explosion powerful enough to be detected al l over the earth .

We do not know, for su re, where meteorites come from, nor even if they al l come from the same place. I t is most l ikely that they or iginate in the asteroid belt, a col lection of massive rocks and smal ler debris that orbits around the sun between Mars and Jupiter. The material that makes up the asteroid belt is sometimes assumed to be the remains of a smal l p lanet that disi ntegrated long ago, but it is by no means certain that the asteroids were formed i n this way.

The meteorites that are commonly d isplayed in muse-

... � - �� ·

Figure 14.1 . Barringer's Crater in Arizona. Nearly a m i le in d iameter i t was produced about 20,000 years ago by an i ron-nickel meteorite weighing many thousands of tons. (Reproduced with permission from Physics of the Earth by T. F. Gaskel l , Funk and Wagnalls Publ ish i ng Co. , New York, 1970.)

200 Extraterrestrial Life

urns are made of metal l ic i ron and have a very characterist ic appearance. The stony meteorites resemble ord i nary terrestrial rocks. They are, therefore, much harder to f ind and they make much less i m pressive exh ibits. Although stony meteorites are not fam i l iar to the publ ic , they are, in fact, the most common variety.

In this chapter we shal l be concerned with the carbonaceous chondrites, a subc lass of the stony meteorites. The chondrites are named after the smal l spheres (chondru les) of s i l icate materials that they contain . The carbonaceous chondrites are un iquely i nteresti ng because up to 4% of thei r mass is carbon, some of i t i n the form of organ ic compounds.

The carbonaceous chondrites conduct heat poorly. Thus, although the i r su rfaces are heated br iefly to i ncandescence, the i r i nteriors remain cool as they fal l through the atmosphere. In pr inciple, therefore, it should be poss ib le to identify the org ani c compounds native to a carbonaceous chondrite by removi ng the th in fusion crust and analyz ing the unmodified material i n the interior.

The Swedish chemist Berze l ius attempted precisely such an analysi s as early as 1 834. At that t ime the techniques of organic chemistry were n ot suff ic iently developed to permit Berzel ius to identify part icular organi c compounds. He d id , however, conclude that nonterrestrial o rgani c material was present. Berze l ius was carefu l to point out that the presence of organic material should not be i nterpreted to mean that l iv ing organisms necessari ly existed at the site of origi n of the meteorite.

Si nce Berzel ius' ti me, organi c chemists have tried to improve on h is analysis. Although many organic co mpounds have been i dentified in one meteorite or another, none of the earl ier analytical results can be accepted without reservation. The major problem has always been terrestria l contam inati on. Carbo naceous chondrites, because they are permeable to water, read i ly pick up organi c compounds such as amino acids from the soi l . Some of the most famous meteorites have been kept i n museums and laboratories for years, and have p icked up further contami nants dur ing storage.

F rom time to t ime, a good deal of excitement is gen-

201 Extraterrestrial Organic Chemistry

erated by c laims that the remains of l iving organisms have been identified i n a meteorite. Sometimes, the "extraterrestrial organisms" turn out to be spores picked up on the g round; in other cases the organisms are probably i norganic artifacts. I n no case is there evidence adequate to establish the presence of l iving organisms or thei r remains i n a meteorite.

Although much of the organic material reported in earl ier analyses of meteorites was picked up on the earth, the carbonaceous chondrites undoubtedly contai n ind igenous organ ic compounds. No doubt, some of these com pounds have been identified correctly. However, unti l uncontaminated meteoritic material became avai lable, it was very difficu lt to be su re whether an organ ic compound in a meteorite was extraterrestrial or not.

On September 28th, 1 969, a large carbonaceous chondrite fel l and broke up in an arid region close to the town of Murchison i n Austral ia. Several p ieces of the meteorite were col lected with in a few days, and many more with in six months. The danger of contamination was fu l ly appreciated , and every precaution was taken to avoid contact between the specimens and sources of terrestrial organic matter. By 1 970 an extensive analysis of frag ments of the Murch ison meteorite was under way in two i ndependent laboratories in the U .S.A.

Both laboratories confi rmed that the Murchison meteorite contains, in addition to other organ ic compounds, large quantities of amino acids. The natu ral amino acids, glyci ne, alanine, g lutamic acid , val ine, and prol ine were identified. In add it ion, some simple amino acids that do not occur in the proteins of l iv ing organisms were found in the meteorite.

We have seen that in this i nstance the c i rcu mstances surrounding the col lection of the speci mens were such that the danger of contamination was m in im ized. The fo l lowi ng three independent l ines of evidence tend to confirm that the amino acids are extraterrestrial in orig in .

Fi rst, the amino acids isolated from the meteorite i nc lude several that are absent f rom l ivi ng organisms and others that are u ncommon. These amino acids are the ones that, on theoretical g rounds, would be expected i n a mixtu re


of prebiotic chemicals. I n one o r two cases, they had been identif ied among the products of prebiotic reactions before they were found in the M u rchison meteorite. In more recent

experiments all of the amino acids identified in the meteorite have been obtai ned by the action of an electric discharge on a "prebiotic" red ucing atmosphere.

The second observation is the most convincing. We saw in the Appendix to Chapter 1 0 that the amino acids present in l iv ing organisms are optical ly active; they are almost a lways in the L-form. Amino acids in the soi l are always of b iologi cal o rig in and, therefore, are optical ly active. On the other hand, prebiotic amino acids are optically inactive; oand L-molecules are present in eq ual nu mbers. The amino ac ids in the M u rch ison meteorite are a l l optical ly i nactive and hence cannot be terrestria l , biological contami nants picked up from the soi l .

The final observation is concerned with isotope ratios. In brief, a l l terrestrial organic material of bio logical or ig in is characterized by a particu lar value of the ratio of the abundances of two carbon isotopes, 12C and 14C. The material on the meteorite is clai med to have a quite d ifferent ratio of 12C

and 14C abundances. If the 12C : 14C isotope ratio is indeed anomalous for the amino aci ds i n the M u rchison meteorite, they m ust be extraterrestrial in orig in .

The M u rchison meteorite contains a number of other organ i c com pou nds in addition to the amino acids, but they are not closely related to i mportant biochemical compounds.

The hydrocarbons found in the meteorite are of the kind that one wou l d expect to form under prebiotic conditions, and they are quite d ifferent from those found in contemporary organisms. It is i nteresting also that i mportant c lasses of biochemical compounds are absent from the meteorite ; a carefu l search fai led to detect adenine, for exam ple. Although

pyri mid ines were found, they are not the ones that occur in nucleic ac ids.

The d iscovery of large amo unts of several natural ly occurri ng amino acids in the M u rchison meteorite c learly establ ishes that these compounds are formed spontaneously somewhere else in the solar system. However, u nt i l we know more about the origin of the meteorite, it is not possible to


draw any more precise conclusions. There is no guarantee that the amino acids found i n the meteorite were formed u nder any of the prebiotic condit ions that have been used in the laborato ry. We cannot, therefore, c laim that the new results support any detai led theory of prebiotic synthesis. On the other hand, they do encou rage the bel ief that our general i deas about prebiotic synthesis have some val id ity.

Molecules in Space

The detection of organi c molecules in regions of space very far from the earth has been made possible by recent developments i n rad ioastronomy. Many molecu les emit (or absorb) rad io waves at characteristic sets of freq uencies. Each molecule h as, so to speak, a un ique rad iofreq uency "signature" or spectrum that d isti ngu ishes it from a l l other molecules. Thus, if we can show that a l l of the frequencies i n the "signatu re" of a molecule are among the freq uencies emitted by a d istant rad io sou rce, we can be sure that the molecule is present in the sou rce. The improved rad iotelescopes that have become avai lable during the last decade or so have made it possib le to apply radiofreq uency spectroscopy to some d istant astronomical objects.

Using this techn ique, many smal l molecu les have been shown to be present i n large amou nts in i nterstel lar dust clouds. The f i rst molecules to be detected were water and ammonia, two molecu les that were expected to be abu ndant. The next compounds to be identified came as a su rprise to the astronomers. They were hydrogen cyan ide, formaldehyde, and cyanoacetylene, a group of reactive organ ic molecules. S i nce then a number of other si mple organi c substances have been detected.

We are al ready fam i l iar with hydrogen cyanide, formaldehyde, and cyanoacetylene; they are bel ieved to have been amongst the most i mportant prebiotic i ntermediates that were formed i n the prim itive reduci ng atmosphere of the earth. It is known that hydrogen cyanide, formaldehyde, and cyanoacetylene react in aqueous so lut ion to g ive a variety of important biochemical compounds; sugars can be derived


from formaldehyde, amino acids and adenine from hyd rogen cyanide, and pyrim id ine bases from cyanoacetylene. The presence of hydrogen cyanide and formaldehyde in the d ust clouds is perhaps not too su rprising , si nce they are among the s implest of organ ic molecu les. However, cyanoacetylene is a rather unfami l iar substance, and its presence in the dust cloud was qu i te unexpected.

I t is probably a coi nc idence that the f i rst three organic molecu les to be detected i n space are i m po rtant prebiotic inte rmediates. A number of further molecules have now been identified. Not al l of these substances are known to be i nvolved in the p rebiotic synthesis of i m po rtant biochemical compounds. Neverthe less, these novel d iscoveries in rad ioastronomy p rovide another h i nt that rather few classes of organic compounds are formed u nder p rebiotic cond itions.

The d ust clouds in which organic molecu les have been detected are bel ieved to be the s ites of formation of new stars. Cou l d sim i lar com pounds in the d ust c loud from which the solar system evolved have contributed to the o rig ins of l ife on earth? It was argued i n Chapter 6 that p reformed organic compounds i n the d ust c loud a re u n l i kely to have been very i m portant for the ori gins of l ife, because they wou ld have been distribu ted u n i fo rm ly throughout the earth .

Formaldehyde

0 / "'-. H H

Water

H - C = N

Hydrogen cyanide

Ammonia

H - C = C C =: N

Cyanoacetylene

Figure 1 4.2. The fi rst five molecu les (other than H2 and CO) d iscovered by radio-astronomers in space. Also five of the most i m po rtant prebiotic molecules.


However, the organic matter that formed in the dust cloud and accumulated at the surface of the earth, may have played some part i n the orig ins of l i fe.

The new results described i n this chapter tel l us l ittle about the detai ls of prebiotic synthesis on the prim itive earth, but they are very im portant in a wider context. We now know that molecu les that can be synthesized u nder prebiotic condition in the laboratory and are thought to have been i mportant for the or ig ins of l ife on earth occur in two q u ite d ifferent extraterrestrial materials. This suggests that they are l i kely to occu r at many p laces in the un iverse. Thus, when we come to consider the possib i l i ty that l ife has evolved on other planets, we shall be able to assume with some confidence that amino acids and related biochemical compounds were formed e lsewhere in the u niverse.


Life in the Solar System

Historical I ntroduction

The possib i l ity that there are J iv ing beings e lsewhere in the un iverse has fascinated men for at least two thousand years. The Greek writer Lucien, who J ived i n the second centu ry A.D. , sent h is adventurers on voyages to the moon and stars. He was not i nterested in problems of propulsion ; the wing of an eag le and the wing of a vulture were al l that his heroes needed to make their jou rneys. Lucien founded a l i terary trad ition in which the behavior of the fictional inhabitants of the moon and planets is used to parody h uman behavior.

Fo r Giordano Bruno, one of the most i mportant scientists of the sixteenth centu ry, extraterrestrial l ife was not a subject for l ig ht-hearted treatment. He wrote, "There are innumerable suns and i nn umerable earths c i rc l ing round their suns just as our seven planets c i rcle round our sun. Living th ings dwel l on these worlds. " Such speculations sti l l

207

sound daring - i n Bruno's day they hel ped lead to the stake. He was burned by the Inquisit ion in the year 1 600.

The fi rst modern science f iction tale was written not long afterwards by Kepler, probably a l ittle before 1 635. Its pub l ication was not thought prudent unti l after Kepler's death. Kep ler is, of course, famous fo r having d iscovered the laws descr ib ing the motion of the p lanets around the sun. H is Somnium or Dream, a fantasy written l ate in l ife, is less wel l known, although it i s a h i g h ly or iginal work. I t i ntrodu ces for the f i rst t ime many of the elements that characterize the best of modern science f iction.

The moon which Kepler describes is the moon which he could see through his telescope. Kepler mai ntai ns the plausi bi l ity of h is narrative by the authenticity with which he descri bes the jou rney to the moon and the condit ions that h is hero, Durocatas, fi nds there. Kepler real ized that the i n habitants of the moon must have adapted to the extremes of the lunar c l imate; h is g igantic l izards g row to maturity and die in a si ng le lu nar day.

After the pub l ication of Kepler's Somnium, the "p lu ral ity of worlds" became a po pu lar topic of discussion i n Europe, and has remai ned so to the present day. Joh n Wi lk ins ( 1 61 4-1 672) , one of the founders of the Royal Society of Lo ndon , examined the possibi l ity that there are i ntel l i gent beings on the moon and predicted that men wo uld one day travel there. Christian Huygens ( 1 629-1 695), the Dutch physic ist, considered eac h of the planets in turn as a possi ble habitation for l ife. Athanasius K i rc her (1 601 -1 680) concerned h imself with quite different questions. He wondered , for example, whether water on the planets should be used for carrying out baptisms.

The fi rst c lai m to have detected an extraterrestrial cu l tu re appeared at the end of the n ineteenth century. Percival Lowel l , an American astrono mer, identified a series of l i near featu res on the surface of Mars as canals bu i lt by the Mart ians. H is wo rk excited a g reat deal of publ ic i nterest. In t ime i t became clear that much of the Martian surface structure that Lowel l descri bed is i maginary, and that the rest can be ex plai ned in a quite ordinary way.

The last twenty years have seen a g reat expansion of


Figure 1 5.1 . The canals of Mars accord ing to Percival Lowel l . No such features are revealed by more recent examinations e ither from the Earth or from orbiting space-c raft. (From Mars as the Abode of Life by P. Lowel l , Macmil lan, New York, 1 909.)

scientific inte rest in the possib i l ity of extraterrestr ia l l ife. The bi ggest s ingle factor has been the space program. The Vik ing mission to Mars i n 1 975, for examp le, is specifical ly designed to search for evi dence of l ife. S i nce it represents an i nvestment of th ree-quarters of a b i l l ion dol lars, it is c lear that many people take the possib i l ity of life on Mars seriously.

The development of rad ioastronomy has also been im-

209 Life In the Solar System

Figure 15.2. A Model of the Vik ing Spacecraft that wi l l land on Mars i n 1976. (Reproduced with permission of the National Aeronautics and Space Administration.)

portant. It is now possi ble to search for rad iocommunications from planets i n other solar systems. Astronomers, at least, seem to th ink that there is a chance that an extensive search would be rewarded ; more than half of those who answered a recent questionnai re thought that it wou ld be sensib le to use large radiotelescopes to search for technologically advanced extraterrestrial cu ltures.

Life on the Moon and Planets

Our nearest neighbor is, of cou rse, the moon. The moon is too l ig ht to retain an atmosphere, and there can be no l iqu id water on its surface. For these reasons almost everyone ag reed that l i fe was not l ikely to exist there. However, this did not prevent the development of a very strange controversy. Most biolog ists were prepared to ignore the possi-


bi lity that organ isms could survive on the moon. A small minority felt that although i t was un l i kely that there was l i fe on the moon , l unar organ isms, if they existed, m ight do so much damage to the population of the earth that extreme precautions shou ld sti l l be taken to prevent return i n g Apollo astronauts from i nfect ing the earth.

Faced with this d ifficu lt choice, the National Aeronauti cs and Space Admin istration (NASA) decided on caution. An elaborate and expensive space q uaranti ne was bu i l t i n Texas, and return irig astronauts and their equi pment were isolated u nt i l shown to be free of dangerous l unar organ isms. We now know that the surface of the m oon is steri le. The reg ions exami ned so far are not o n ly l ifeless, they are also almost free of organi c carbon. Almost everyone ag rees that there is no chance that l ife exists anywhere on the surface of the Moon.

With the advantage of h indsig ht, we can now see that the lunar q uaranti ne was unnecessary. At the t ime it was very d ifficu lt to k now what to do about a contingen cy that was extremely improbable, but might j ust possibly have been total ly disastrous for the h uman race.

It is generally believed that the evolution of earth- l ike l ife would not be possible at h igh temperatures in the absence of l iqu id water. Hence it is thoug ht extremely u n l i kely that the planet Mercury is inhabited s ince Mercury is very hot and has no atmosphere. It is harder to be certain about the lower l im it of the temperature range in which l i fe is possible, but it is l i kely a lmost certa in that the outer planets, U ranus, Neptune, and Pluto are too cold to support l i fe.

The surface of Venus is very hot (- 480"C). No l iqu id water could exist at th is temperatu re; i n fact few if any organ ic com pounds cou l d survive, even in the sol id state. We can, therefore, be confident that there is no earth- l ike l ife at the surface of Venus. Since the top layer of the atmosphere is at a much lowe r tem perature (- 40°), there is certainly a reg ion in the atmosphere where the temperature is opti mal for the existence of l ife. It has been suggested that this region may be i nhabited. I feel that the d ifficu lties i mpl ici t in the evolution of l i fe i n a reg ion far from a sol id o r l iqu id su rface are s o considerable that this i s u n l ikely, but it

21 1 Life In the Solar System

is certain ly not i mpossible. It is also j ust possi ble that l ife evolved at the su rface long ago, when conditions were more favorab le.

S im i lar arg u ments apply to the two major planets , Jupiter and Saturn. These planets are com posed in the main of hyd rogen and he l i um . They do not have so l id surfaces, so if l ife evolved at a l l it m ust have been in the atmosphere. The tem perature at the top of the atmosphere of Jupiter and Satu rn are much lower than the one on Venus, but deep in the atmospheres there are bel ieved to be reg ions at temperatu res that are su itable for l ife. Agai n , the possib i l ity that l ife has evolved in these reg ions can not be d ismissed enti rely, but I th ink the probabi l ity is smal l .

This leaves Mars as the most promis ing p lanet on which to search for evidence of l i fe. I n the remainder of th is chapter we shal l su rvey what is known about su rface conditions on Mars and the l i ke l ihood that l ife cou ld exist there. This subject is particu larly interesti ng at the moment, si nce in 1 976 we shal l have the opportun ity to land scientif ic instru ments on the surface of Mars to look for evidence of l i fe.

It is bel ieved that the red co lor of Mars is due to large quantities of red, i ron-contain ing m inerals present on its su rface. Photographs taken by a Mariner spacecraft orbit ing the planet show that the surface is cratered and looks very m uch l ike that of the moon in many places. However, because Mars is the site of intensive volcan ic activity and also because it has an atmosphere, the su rface of the p lanet shows much more evidence of active, geological processes. Ground-based observations have detected dust storms that someti mes obscure large areas of the p lanet's surface, and pictu res taken by Mari ner spacecraft have confi rmed these observations in a spectacular way and revealed reg ions where craters have been fi l led in rapid ly, presumably by d rifti ng d ust.

The temperatu res on Mars are more extreme than are those on earth ; but, certain ly, they are not such as to preclude the existence of earth- l ike l ife. At the equator, temperatu res as h igh as 32°C (90°F) are reco rded in the summer, but the temperature fal ls to - 1 00°C at the coldest period of the Martian n ight. Tem peratures at the poles range from - 1 0°C to - 1 30°C.


The pressu re of the atmosphere at the Martian su rface is on ly about 1 % of that on the earth . The very th in atmosphere is com posed very largely of carbon d ioxide. I n addit ion i t contains a smal l amount of carbon monoxide. I t is not surpris ing that there is on ly very l ittle oxygen i n the atmosphere , but the recent f inding that n i t rogen is absent came as a g reat su rprise. The Mart ian atmosphere does contain some water, but very l ittle.

The white polar caps are very p romi nent in wi nter. They appear, to the eye, as though they were made of snow or ice.

Figure 1 5.3. Mars as photographed by the Mariner VII Spacecraft at a d istance of 534,398 mi les from the Martian surface. The south polar cap is c learly visible at the bottom of the photograph. (Reproduced with permission of the National Aeronautics and Space Admin istration.)

21 3 Life in the Solar System

Apparently, this is not the case. The main co m ponent of the Martian ice cap is known to be sol i d carbon di oxide. Water is a m i nor but i mportant constituent.

The most i ntri g u i n g changes that occu r on the Martian su rface are associated with the recession of the ice caps i n sum mer. Each year, as the i ce caps vaporize, a "wave of darkening" is in itiated in the polar regions and spreads in

Figure 15.4. The "wave of darkening" that spreads towards the equator as the "ice-cap" melts. (Reproduced with permission from Stars, Planets and Life by Robert Jastrow, Wi l l iam Heinemann Ltd. , London, 1 967, p. 1 03.)

214 Extraterrestrial life

the d i rection of the equator. It has been suggested that the wave of darken ing is caused by the seasonal g rowth of vegetation. However, there are a number of less exciti ng possib i l ities : the darken ing coul d be caused, for example, by drifti ng dust.

Recent photographs of Mars taken f rom Mariner spacecraft, although of excel lent q ual ity, are not adequate to establ ish or rule out the presen ce of l iv ing organisms. Equivalent photog raphs of the Earth wou l d not reveal unambiguous evidence of terrestrial l i fe . Thus, there is no d i rect expe ri mental evidence for o r against the existence of l ife on Mars. The most recent ind i rect evidence shows that the atmosphere on M ars d iffers more from our atmosphere than was thought previously. This perhaps makes it s l ightly less l ikely th at earth - l ike forms of l ife are present than seemed probable a few years ago. However, the odds have not chan ged very much.

Three characteristics of the Martian atmosphere seem at f i rst s ight to make it unsu itable for the evol ution of l ife. I t contains l i tt le water and no detectable n i trogen. In addition the atm os phere is not sufficiently reduc ing to make possib le prebiotic syntheses of the k ind discussed i n Chapter 7. However, none of these observations argues strongly against the existence of l ife on Mars.

We must d istingu ish between conditions that permit the evolution of l ife and those that permit the persistence of l ife once it has evolved. It is doubtfu l that there is enough free water on Mars at p resent to permit the evol ution of l i fe to occur. However, it is l i kely that l i qu id water was once abundant, for the most recent Mariner p ictures reveal surface features which are hard to i nterpret as anyth ing other than river beds of comparatively recent orig in . I f so, l ife may have evolved on Mars i n a wet period and gone underground (h ibernated ?) . Recently a g reat deal of chemical ly bound water has been detected on m inerals at the Martian surface, and there is no obvious reason why l iv ing organisms should not use it.

I f there is no n itrogen on Mars, i t is un l i kely that earthl i ke forms of l i fe can exist. However, Martian organisms may be able to survive i n an environment contain ing much less

21 5 Life In the Solar System

Figure 1 5.5 A 700 ki lometer long sinuous val ley on the surface of Mars. This va l ley may just possibly have been produced by the col lapse of the roof over a surface lava flow. On the other hand it could have been formed by erosion - if so, water must once have been abundant on the Martian surface. (Reproduced with the permission of the National Aeronautics and Space Administration.)

n itrogen than is present on the earth. We can not exc lude the poss ib i l ity that there is a percent or so of nitrogen in the Martian atmosphere and that this is enough to supply the needs of Martian organ isms. Alternatively, some nitrogen may be present in the soi l as nitrate, a form of nitrogen which could probably be uti l ized more read i ly than could molecular n itrogen.

The low abu ndance of red ucing compounds in the Martian atmosphere was thought unti l recently to prove that the synthesis of organic molecu les cou ld not occur there. However, recent experi ments have shown that this conclusion is incorrect. An atmosphere of the same composition as the Martian atmos!Jhere was i rrad iated with u ltravio let l ight in the presence of a nu mber of inert powders. I n every case, the carbon monoxide in the atmosphere was converted i nto organ ic compounds; formaldehyde, a potential precursor of sugars, was the most abu ndant product.

In the past, the atmosphere of Mars may have been more red ucing. I n that case a wider range of organic compounds wou ld have been avai lable. If l ife evolved under such conditions, there is no reason why i t should not have adapted itself as the atmosphere became more oxid iz ing . We know that j ust such an adaptation has occurred on the earth.

Many of the outstand ing questions abo ut Mars shou ld be settled i n 1 976 when two Vik ing spacecraft are expected to land instrument packages on the su rface. One g roup of i nstruments is designed to detect Martian microorgan isms, if there are any in the neighborhood of the spacecraft. Another one wi l l carry out a com plete analysis of the organ ic constituents of the Martian surface. If no l ife is present, we should get an overa l l p icture of the prebiotic organ ic chemistry of Mars. If we are fortunate enough to d iscover l iv ing organ isms, a new perspective in bio logy wi l l open up for us. I bel ieve that the Vik ing Project is the most i mportant scientific venture in the enti re Space Prog ram.

21 7 Life in the Solar System

Intelligence in the Universe

In this chapter we shal l consider the possi b i l ity that there are intel l igent extraterrestrial societies, and the chance that, if they exist, we shal l be able to commun icate with them. We shal l talk about earth-l i ke forms of l ife on ly, because we know noth ing about any other k ind . This subject is wel l worth d iscussing , even if everything that can be sai d is h igh ly speculative.

The probabi l ity that technological societies based on earth- l ike l ife exist outside the solar system can be d iscussed on ly after the problem has been broken down i nto a number of parts, namely:

1 . How many planets (N) are there in the Universe? 2. What proportion of them (FL) could support l ife? 3. What is the probabi l ity (Fp) that l ife evolves on a

p lanet that can support l ife?

21 9

4. What is the probab i l ity (FT) that a technological society develops on an in habited p lanet?

5. How long, on t he average, do tech nological societies survive ?

I f t h e answers t o these questions were known i t wou ld be . possible to use very elementary argu ments to est imate the number of technolog ical societies in the un iverse. Let us start with a very simple problem that i l l ustrates the p rinc ip le that is i nvolved . Suppose that there are 1 000 islands in a certain reg ion of the Pacif ic Ocean, and that one tenth of them are in habitable. Then it is obvious that there are about 1 000 x 1 /1 0 = 1 00 inhabitab le islands. In exactly the same way if there are N p lanets in the universe and a fraction F2 are of a type that can su pport l ife, then there are about N x FL planets able to support l ife. The same type of argument can be used to show that N x FL x Fp planets not on ly can support l ife, but actual ly do become in habited , and that N x FL x Fp x FT planets sooner or later come to harbor technolog ical civi l izations.

If we are interested in commu nication with other planets, we need to know what proport ion would be occupied by inte l l igent civi l izations at the moment of contact. Clearly th is would be less than the p roport ion of p lanets that are occupied at one t ime or another in their h istory si nce we might attem pt to make contact with the planet before an i ntel l igent society had become establ ished or after it had become exti nct. Agai n the example of is lands i n the Pacif ic is instructive. Suppose that the supply of foodstuff on an is land is adequate for on ly one year, but that it takes ten years to replace the supply once it has been used up. Then the chance of f inding anyone on an inhabitable island on the occasion of a random visit wou ld be at most one i n ten. I n the same way i f a n average planet survives for 1 0,000,000,000 years, but is usually made permanently uni nhabitab le with in 1 0,000 years by tech nological development, then the chance of fi nd ing a resident technological society at the t ime of f i rst contact would be 1 0,000/1 0,000,000,000 that is on ly one i n a m i l l ion. More general ly the number of technological societies i n the un iverse at any ti me is l i kely to be


N x FL x Fp x FT x L/Lp,* where L1 is the average l ifet ime of a technological society and Lp is the average l ifeti me of a planet.

Stars occur in widely separated clusters known as galaxies. Our own galaxy contai ns about 1 01 1 stars ; the number of stars in the universe is about 1 020• Unfortu nately, even the nearest stars are so far away that astronomers would be unable to detect planets as smal l as the earth d i rectly with the telescopes that are avai lable today. Esti mates of the number of planets in the universe must be based on i nd i rect evidence.

A nu mber of theories of star formation have been developed recently. They lead to the conclusion that planets no larger than the solar planets and also heavier p lanet-l i ke objects (dark com panions) are common. To some extent these predictions have been verified experi mental ly. Observations on nearby stars have shown that a surprisi ngly large proportion do have com panions heavier than the largest solar planets. S ince the large planets have been fou nd, it is reasonable to bel ieve that the smal l planets predicted by the theory are also there.

Another arg ument for the existence of a large nu mber of planets is provided by studies of the rotational (sp inn ing) motion of stars. The material i n the interste l lar dust clouds, from which stars are formed , rotates about the center of the cloud. A well -known law of physics, the law of the conservation of angular momentum, shows that this rotational motion cannot stop when the dust cloud condenses into a star. However, d i rect observation of stars si m i lar to the sun shows, that they rotate very slowly.

A number of mechan isms have been proposed to account for the miss ing rotational motion. In the most plaus ib le theory, it is suggested that most of the rotational motion has been transferred from the central star to a group of earth- l ike planets. Si nce we know that the planets possess most of the rotational motion in the solar system , this explanation is particu larly attractive. However, we cannot be sure that it is correct; some other theories accou nt for the loss of angular momentum without i nvolving planets at al l . * This i s a simpl ification o f a formu la given by F . Drake.

221 Intelligence in the Universe

NGC 4565 Viewed edge-on

NGC 421 6 Tilted 15 degrees

NGC 7331 Tilted 30 degrees

Figure 16.1. Some typical galaxies seen at various orientations. (Reproduced with permission from Stars, Planets and Life by Robert Jastrow, Wil l iam Heinemann, Ltd. , London, 1 967, p. 27.)

Most astronomers agree that planets are l ike ly to be very nu merous. There could wel l be as many as 1 010 in our own galaxy and 1 019 in the un iverse. How many of them could support earth-l ike l ife?

S ince l ife s imi lar to that on earth cou ld exist on ly in the presence of l iqu id water, earth-l ike l ife is improbable except in regions where the temperatu res are moderate. Astronomers have tried to esti mate the proportion of planets that have maintai ned a su itable range of temperature for a suff iciently long ti me to permit the evolution of l ife. They have concluded that such planets are quite common ; there cou ld be as many as 1 Q9 i n our galaxy, and a correspondi ngly larger number in the un iverse.

Next we must esti mate the probabi l ity that l ife does evolve, g iven a suitable envi roment. This is an extremely controversial topic. Some scientists believe that l ife i nevitably occu rs where conditions are favorable, but others bel ieve that the evolution of l ife is extremely un l ikely even under the best conditions. I bel ieve it to be i mpossible to make a meaningful estimate, at the present t ime.

The evidence presented in Chapters 7 and 8 makes it probable that a rich prebiotic sou p accum u lates on a considerable proportion of earth-l i ke planets. It is m uch harder to estimate the probabi l ity that l iving th ings evolve in such a soup. The evidence avai lable at present does not indicate whether the evolution of b io logical systems is l ikely or not. We wi l l not be able to decide unti l we know much more about the fi rst steps i n the evolution of l ife : the repl ication of polymers under prebiotic conditions.

Many writers have taken a more opti mistic view of this problem, and some have clai med that l ife is almost certain to evolve i n a su itable envi roment. They bel ieve that, s ince l ife has evolved on earth, it is almost sure to have evolved on other earth-l i ke planets. However, this seems to be an i ncorrect concl usion , based on a m isunderstanding of probab i l ity theory.

The remain ing questions that m ust be answered before we can esti mate the probable number of tech nological societies in the un iverse are difficu lt to answer for s imi lar reasons. So far, on ly one ani mal has developed sufficient i ntel l igence to construct a tech nological society on earth. It is


not poss i b le , fro m this s ingle example, to est imate how l i kely it i s that i nte l l i gence wou l d develop o n some other i n habited planet, or whether a really i nte l l i gent speci es wo u l d bother to develop a h i g h ly sophi sti cated technology.

E stimates of the time for which technolog ical societies are l i kely to survive depend more on the temperament of the writer than on relevant i nformation. Optimi sts bel ieve that the future of man k i nd is u n l i mited ; pessimists bel ieve that our chance of survivi ng for a further h u n d red years is poor. It is clear that, d u ri ng the l ast h u n d red years, technological prog ress on the earth has been trau matic. We do not know whether this is a matter of h istorical accident, or whether the potenti al ly self-destructive powers of techn ology always present a serious problem to advanced soci et ies. Clearly, we are not i n a positi o n to esti mate the average l i feti me of a technological society.

It may be concl uded that we cannot make mean i ngful esti mates of some of the q u antities that appear in our form u l a for the n u m ber of inte l l i gent societies i n the u niverse. There may be many of them or we may be alone. At an early stage in the development of a science, ign orance of this k ind i s not u nusual and i s certain ly no cause for embarrassement. The problem of the " p l u rality of i nh abited worlds" i s part of our i ntel l ectual h eritage. It is exciting to think that we are at l ast beg i n n i n g to understand the chemical and b i ologi cal princi ples that are involved in the evo l ution of l ife, and to b u i l d the telescopes with which to com m u n icate with extraterrestrial civi l izations, if they exist.

Communication with Extraterrestrial Civi lizations

Even if extraterrestrial civi l izations do exist, i t may prove di ff icult to com m u n icate with them if they are very far away. To appreciate this point it is necessary to have some feel ing for the enormous d istances that are involved. Light travels at the speed of 1 86,000 mi les per second, or ·5,880,000,000,000 m iles per year. The d i stance even to the nearest stars i s


measured in l ight years, that is in un i ts of 5 ,880,000,000,000 mi les.

Our own galaxy is a f lattened d isc contain ing about 1 01 1 (a hundred b i l l ion) stars ; it is about 1 00,000 l ight years across and 2,000 l ight years thi ck. The galaxy c losest to our own is about two mi l l ion l ight years away; astronomical objects more than five b i l l ion l i ght years away have been detected.

No radio signal or sol id object can travel faster than the speed of l ight. This makes it difficu lt to communicate with distant p lanets. It is i mpossible to establ ish two-way commun ication with any planet more than 35 l i ght years away in less than 70 years, for example. Only a few hundred stars fal l with i n 3 5 l ight years o f the earth , so on ly a t iny proportion of p lanets are close enough to be contacted in a hu man l ifetime.

When objects are accelerated to speeds approach ing that of l ight, the i r est imates of t ime are l i kely to be d ifferent from our own. A travel ler on a spacesh ip that accelerated to almost the velocity of l ight, s lowed down , and then retu rned home, would find that a longer t ime had elapsed on earth than on h is spacesh ip . He might retu rn as a young man, to f ind a l l h is contemporaries had been dead for centu ries. * I n a s im i lar way, a n astronaut on a n accelerati ng spacesh ip cou ld travel very long distances i n ti mes that appeared qu ite short to h im , although they wou ld appear much longer to us. None of these paradoxes affect the conclusion that we cannot receive a reply from a d istant planet i n less than twice the t ime that l ight takes to travel there.

We are not yet in a position to travel to other solar systems, but in time this should be possi ble. Sooner or later we shal l be able to make retu rn journeys to the few hundred stars with in 30 l ight years of the earth. Unfortunately, such journeys wi l l probably not take place in our l i feti mes. For future generations, space travel may provide the most d i rect form of two-way communication with extraterrestrial civi l izations, if there are any close enough.

Two-way commu nication using rad io or l ight waves is

• Most theoretical physicists believe that th is is a legitimate deduction from the special theory of relativity, and l i kely to be true.


restricted i n the same way by the f in ite speed of l i g ht. If we sent o ut a message to one of the m ost d i stant stars i n our own galaxy, i t wou l d take at l east 1 00,000 years to get a reply. Two-way com m u n i cation with a civi l ization i n another galaxy wo u l d certai n l y take at least 4 m i l l i o n years. Even the prospect of conversing with the nearest stars, a few l i g ht years away, i s daunti ng.

The prospects of picking up a message sent out spo ntaneously f rom another soc iety are m u ch better. We cou l d , i n pri nciple, receive messages f ro m any part of the u n iverse - a message (or a spaces h i p) trave l l i n g at al most the velocity of l ight that was sent out fro m the nearest galaxy two m i l l ion years ago , for examp le, should reach us about now. Of cou rse, the i n d iv iduals who sent the message wou ld probably h ave d i ed long ago.

Our abi l i ty to detect s ign als from other planets woul d depend o n th ree factors. The d istance away of t h e sender, the power of h i s signal , and the sharpness with w h i ch he beamed h i s message to the earth . A g reat deal of thought has been g iven to the problem of detecti n g weak s ignals from o utside the solar system , and certai n g eneral co nclusions have been reached.

The radiotelescopes that are used at the present t ime are not sensitive enou g h to permit us to eavesdro p on conversations between civi lizations whose power consumption is s imi lar to our own. Eavesdropping wou l d be possible only if the sig nals that are used on other planets are much more powerful than any that can be generated on earth.

The detection of beamed s ignals would be much si mpler. I f any extraterrest rial civi l ization has set up a beacon to make its presence known, we might wel l be able to detect it. The fi rst meeting of C ETI, a society whose objective it i s t o com m u n icate with extraterrestrial i ntell igence, was held at the Byurakan Observatory i n Soviet Armenia d u ri n g September, 1 971 . A gro u p of American, Eu ropean, and Soviet scienti sts, i nc lud ing two sceptical Nobel Prize w i n ners, ag reed that a l i mited search for extraterrestrial s ignals is j ustified at the present t i me.

What wo u l d messages of extraterrestri al or ig in be l i ke ?

226 Extraterrestrial Ufe

Almost surely the technology on a planet that sent such messages would be more advanced than our own. The senders, therefore, would be at least as fami l iar with mathematics and physics as we are. They would also real ize that anyone able to receive thei r messages m ust belong to a society capable of bu i ld ing telescopes. It seems probable, therefore, that the messages would i nclude sim ple texts deal ing with mathematics and technology. A g reat deal of work has already been done on the decipheri ng of texts of this kind. We cou ld probably understand them once they had been identified.

It is hard to know what other i nformation extraterrestrial soc ieties would send us. A fasci nati ng but g loomy prognosis is g iven in Hoyle's novel A for Andromeda. Some optimistic writers bel ieve that the senders would be benevo lent m issionaries, determined to improve the lot of mankind. It wi l l be left to the reader to write h is own science f ict ion, and to decide for h imself whether it wou ld be wiser to answer the fi rst messages we receive or to keep very qu iet.

It is i mpossible to est imate the probabi l ity of our receiv ing intel l igent messages from space. It may be foo l ish to take such a poss ib i l ity seriously, or it may be un i maginative to do otherwise. The detection of another i ntel l igent society i n the un iverse would be one of the most i m portant and exciti ng events i n our history ; if I had control of a radiotelescope, I wou ld be wi l l ing to gamble a few percent of the watching-ti me to search for evi dence of extraterrestrial inte l l igence.


Summary of the Main

Argument

The earth was fo rmed from a cloud of dust and gases about fou r and a ha lf b i l l ion years ago. The cloud was made up very largely of hydrogen and hel i um , but these gases, together with most other vo lat i le material , escaped duri ng the condensat ion process. Our atmosphere and oceans are derived fro m gases that were expel led from the i nterio r of the earth at a later date.

The pr imit ive atmosphere was reducing. It co ntained water, methane, carbon dioxide, and n i trogen ; in addit ion, some ammonia and hydrogen may have been present. We know that the atmosphere was sti l l red ucing a b i l l ion and a half years later, by which t ime organ isms s im i lar to algae or bacteria had evolved on the surface of the earth. When we talk about the orig ins of l ife, we mean the series of processes which occu rred in the atmosphere, oceans, and lakes of the primit ive earth and led , more than three b i l l ion years ago, to the appearance of the fi rst l iv ing organisms.

We have no geolog ical record of the events that occu rred so long ago. However, i t seems very l i kely that the f i rst step

229

i n the orig ins of l i fe occu rred when very s imple organ ic molecules were formed in the earth's reducing atmosphere by the act ion of l i ghtni ng, u lt ravio let energy from the sun, shock-waves, and so forth. These compou nds were washed i nto oceans and lakes where they reacted to form a complex mixture of o rgani c substances, inc lud ing many that were to form part of the most prim it ive l iv ing organisms.

The mixture of chem ical compounds that was formed on the pr imit ive earth is referred to as the p rebiotic soup. Prebiotic chem istry is concerned with labo ratory experiments that s imu late the p rocesses bel ieved to have been i nvo lved in the formation of the prebiotic soup. The most strik ing success achieved so far in the study of the or ig ins of l ife is the demonstrat ion that the components of l ivi ng . o rganisms are part icu larly abundant i n the m ixtu re of o rgan ic chemicals formed in prebiotic react ions. Most of t he important components of modern o rgan isms. for example, amino acids, sugars, and nu cleic-ac id bases, have been synthesized in the laboratory u n der condit ions that cou ld have p revai led on the primit ive earth.

The next stage in the orig ins of l ife m ust have been the concentrat ion or "thicken ing" of the prebiotic soup. Evaporation was al most surely impo rtant, but freez ing and adsorption on the surface of rocks or co l loi dal parti cles may also have played a part Once the p rebioti c sou p was suffi ciently concentrated, polymers resembl ing proteins and n uc leic acids cou ld be formed. Not m u ch success has been attai ned in s imu lat ing these polymerizat ion reactions in the labo ratory, but some prog ress has been made.

The major i ntel lectual problem presented by the o rig ins of l i fe i s concerned with the next stage, the evo l ut ion of biologi cal o rganization. How did a co mplex self-repl icati ng o rganism evolve from an unorganized m ixtu re of polymeri c molecu les? Little experimental evidence is avai lable, s o one is forced to attempt a speculative reconstruction of this phase i n the o ri g i ns of l ife.

The key to the u nderstand ing of the evolut ion of biological organizat ion is the theory of natu ral selection . Before the evolut ion of compl icated self-repl i cati ng o rganisms, natu ral selection must have acted o n somethi ng m uch si m pler,

230

Extraterrestrial Life

probably on polymeric molecu les resemb l ing nu cleic acids. It is bel ieved that nucle ic acid-l i ke molecu les were formed i n the prebiotic soup and were able to reproduce without the help of enzymes. The theory of natural selection then shows that those molecu les that cou ld repl icate fastest wou ld have become domi nant i n the prebi otic soup.

As the competit ion became fiercer, the more successful fami l ies of self-repl icati ng molecules must have " learned" to make use of smal l molecu les in their environ ment to help them to rep l icate even faster. The most i mportant of these adaptations i nvolved the ami no acids; u ltimately a fami ly of self-repl icating n ucleic acids evolved to the point where they could beg in to contro l the synthesis of po lypeptide sequences that had usefu l catalytic properties. This adaptation led u lti mately to the evo lution of protein synthesis and the genetic code.

The evol ution of prote in synthesis is not u nderstood in detai l . One of the g reat chal lenges of the problem of the orig i ns of l ife is to demonstrate in the laboratory how polynuc leotides, without the help of preformed enzymes, could have repl i cated and begu n to control the synthesis of pept ides with determi ned sequences. Once th is has been done we shal l be wel l on our way to un derstand ing the orig ins of the f i rst l iv ing cel ls.

Unt i l recently, theories of the orig in of the prebiotic soup were based enti rely on laboratory experiments. With in the last few years d ramatic developments in astronomy have changed th is situat ion. A n umber of small organic molecules have been detected i n interstellar dust clouds far from the earth. The most abundant of these i nc lude formaldehyde, hydrogen cyanide, and cyanoacetylene. These three molecu les had been proposed previously as three of the most important precursors of the p reb iotic soup. Thus radioastronomy has provided evidence su pport ing strong ly our general ideas about prebiotic synthesis.

The d iscovery of many of the twenty natural ly occurring amino acids in the Murchison meteorite, a meteorite that cou ld not have been contami nated with terrestrial organic material , i s equal ly i m portant. In fact the mixture of amino ac ids in the meteorite matches almost exactly the mixtu re of

231 Summary

amino acids formed by the action of an electric discharge on a reducing gas mixture. Clearly, amino acids are formed i n large amou nts somewhere in the solar system far from the earth.

These new f ind ings suggest that the range of organic compounds that can be formed u nder prebiot ic condit ions is smal l . It is not u n l i kely, therefore, that there are other planets on which rich prebiotic sou ps have accumu lated. We do not know how l i kely it is that l i fe would evolve on such planets, but we can see no good reason why earth- l i ke l i fe should not ex ist elsewhere i n the u niverse. I t is also possi b le that total ly al ien fo rms of l i fe have evolved on other planets, but we are not as yet in a posit ion to say much about this possi bi l ity.

Studies of the ori g i ns of l i fe have reached a part icu larly excit ing po i nt. We hope with in a few decades to u nderstand how l ife on earth evolved from a random mixture of o rganic compounds. It may take a l i ttle longer to create novel selfrepl i cat i ng systems, but i t is l i kely that this wi l l be achieved with i n , say, a hundred years. Once we understand more about the evolut ion of b io log ical o rganization, we should be able to say something quantitative about the probabi l i ty that l ife exists elsewhere i n the un ive rse. Perhaps we shal l not have to wait so long - a systematic search for s igns of i ntel l igent l ife elsewhere i n the un iverse is l i kely to beg in i n the near future. If we succeed in maki ng contact with an extraterrestrial c ivi l ization, o u r view of man's place i n the universe wi l l certa in ly change.

232


Bibliography

Origin of Life

Oparin, A. 1 . , The Origin of Life on Earth, Academic Press, Inc., New York, 1 957. Some of the detailed argu ment has now been superceded , but a classic and sti l l a very i nteresting book.

Oparin, A. 1., Life, Its Origin, Nature and Development, Academic Press, Inc. , New York, 1 964. A more recent but in many ways less satisfactory treatment.

Kenyon , D. H . , and G. Stei nman, Biochemical Predestination, McGraw-H i l l Book Co., New York, 1 969. An excel lent advanced text.

Mi l ler, S . , and L. E. Orgel, The Origins of Life, Prentice-Hall , Eng lewood Cl iffs, New Jersey, 1 973. A book at the advanced undergraduate level emphasizing chemical problems.

Bernal, J. D., The Origins of Life, World Publ ishing Co., New York, 1 965. A dictionary on the origins of l ife. Some of the discussion is now out of date. -

Cai rns-Smith, A. G. , The Life Puzzle, U niversity of Toronto Press, Toronto, 1 971 . An unorthodox view of the orig ins of l ife.

Molecular Biology

Watson, J. D., The Molecular Biology of the Gene, 2nd Ed. W. A. Benjamin , New York, 1 970. Perhaps the standard treatment of molecular biology at the g raduate or advanced undergraduate level.

Monod, J., Chance and Necessity, Alfred A. Knopf, New York, 1 971 . A very elementry account of molecular biology that concentrates on phi losoph ical impl ications.

Woese, C., The Origins of the Genetic Code, Harper and Row, New York, 1 967. A detailed d iscussion of the genetic code and how i t evolved.

233

Evolution

Darwin, C., The Origin of Species, The origin of species (sixth ed. 1 872) and The Descent of Man, Modern Library, New York, also reprint of 1 st ed. (1 859). Phi losophical Library, New York. The most important book in the l i terature of bio logy and sti l l excellent reading .

Eiseley, L. , Darwin's Century, Victor Gol lancz, London, 1 959. A detai led but always interesting account of the development of evo lutionary theory.

Stebbings, G. L., Processes of Organic Evolution, Prentice-Hal l , Eng lewood Cl iffs, New Jersey, 1 966.

The Fossil Record

Eicher, D. L., Geological Time, Prentice-Hal l , Eng lewood Cl iffs, New Jersey, 1 968.

Schopf, J. W., "Precambrian fossils and evolutionary events prior to the origi n of vascular plants," Biological Reviews 45, 3 1 9 (1 970). A review of recent work on the earl iest microfossi ls .

Astronomy and Geophysics

Jastrow, R., and A. G. W. Cameron, The Origin of the Solar System,

Academic Press, Inc . , New York, 1 963.

Rittman, A., Volcanoes and their Activity, lnterscience, New York, 1 962.


Shklovsk i i , I. S. , and C. Sagan, Intelligent Life in the Universe,

Holden-Day, San Franciso, 1 966. An exciting treatment of the orig ins of l ife and of intel l igent societies.

Hoyle, F., Of Men and Galaxies, University of Washington Press, Seattle, Washington, 1 964.

Science Fiction

Hoyle, F., The Black Cloud, a Signet Science Fiction paperback edit ion, The New American Library, New York, 1 959. An outstandingly imaginative account of a "foreign" intel l igence.

· Hoyle, F. , Andromeda Breakthrough, Harper & Row, New York, 1 964.

234 Bibliography

INDEX

Activated intermediates, 1 41 -42 Active site, 69 Adenine, 1 30 Adsorption, 1 36, 1 70-71 Algae, blue-green, 28, 30 Amino acids

in meteorites, 202-204 naturally occurring, 36-39 prebiotic synthesis of, 1 24-29

Anticodon, 83-85 Applejack, 1 35 Aristotle, 4-5 Arrhenius, Svente, 1 3-1 4 Artificial l ife, 1 97 Asteroid belt, 200 Atmosphere

contemporary, 1 09-1 0 primitive, 1 06-1 2 reducing, 1 4-15, 1 07-1 2, 1 1 3

ATP, 52-53, 75-79, 1 41 , 1 42

Bacteria, 33-36 fossil , 30

Base pairing. See Watson-Crick pairing

Bernal, 1 36 Berzel ius, 201 Biosynthetic pathways, 1 73-74 Bruno, Giordano, 207 Butlerov, 13 1

California State Board of Education, 1 2-13

Carbonaceous chondrites, 201 Catalysts, 36

prebiotic, 1 40, 1 43 Cell membrane, 35, 49, 1 69-70 Cells. 33-36

evo lution of, 1 69-75 functioning of, 49-54

Cell wall, 35 CETI, 226 CITROENS, 1 9 1-97 Clays, 1 36 Cloverleaf structure, 81 Coacervates, 66, 1 36-37, 1 44, 1 54,

1 57, 1 71 Codon, 84-85 Colloids, 63-66, 1 72, 19 1 Complementarity, 42-45 Concentration

by adsorption, 1 36-37 by evaporation, 1 34-35 by freezing, 1 36-37

Condensation reactions, 78-79, 1 37-44

i n solution, 1 41 -44 thermal, 1 39-40

Cyanoacetylene, 1 31 , 1 32, 204

Darwin, Charles, 1 1 -14 , 1 45-46, 1 48, 1 78, 1 8 1

Deoxyribonucleotides, 40-42 Deserts, 1 40, 1 43 Design, 1 8 1 -82, 1 96

235

DNA, 39-46, 80-85. See also Nucleic acids

Double helix, 42-45, 66-69 Drake's formula, 21 9-21 Dust clouds, 1 00-1 02, 204-206

Earth age of, 1 04 primitive, temperature of, 1 04

E. col i . See Bacteria Electric discharge, 1 5, 1 1 4, 1 1 8,

1 25 Enantiomers, 1 60-67 Energy, sources of 1 1 3-21 Enzymes, 36-39, 55, 69-72

repair, 51 Equi l ibrium, 73-74, 1 4 1 Evolution, 1 77-83

as explanation of origin of l ife, 1 1 -1 4

biochemical, 87-89, 91 convergent and divergent, 88 molecular, 1 45-55

Extraterrestrial civi lizations, 21 9-27

Fermentation, 52-53, 77 Formaldehyde, 1 1 7, 1 24, 1 30, 1 31 ,

1 32, 204 Fossil microorganisms, 26-30 Fossil record, classical, 23-26 Fossils, 1 8-21

chemical, 28 Fox, Sidney, 1 39, 1 7 1 -72 Free energy, 73-79

Genes, 55-57 Genetic apparatus, 1 5 1

evolution of, 1 55-58 Genetic code, 48-49, 1 50, 1 55

evolution of, 1 57-58 primitive, 1 56 and protei n synthesis, 80-85

Genetic material, 39-46. See also DNA, Nucleic acids

Genetics, molecular, 55-59 Geolog ical dating, 1 7-23 Gravitational escape, 1 06

Haldane, J. B. S., 1 4-1 5, 1 24 Half-life, 22-23 Helmont, van, 5 Hemoglobin, 1 77, 1 9 1 Hoyle, 227 Huxley, T. H., 1 8 1 Huygens, Christian , 208 Hydrogen bonds, 67-68 Hydrogen cyan ide, 1 1 4, 1 24,

1 30-32, 204 Hydrophi l ic molecu les, 64-66 Hydrophobic molecules, 64-66

I nformation, 1 90-93 Isotope dating, 1 7, 21 -23

Joblot, Louis, 8 Jupiter, 21 2

Kant-Laplace hypothesis, 1 00 Katchalsky, Aharon, 1 43 Kepler, 208 Kircher, 208

Lamarck, 1 78, 1 80 Last common an cestor, 89, 93 Leeuwenhoek, Anthony van, 7-8 "Life," 1 87-89 Lowell , Percival , 208 Lucien, 207

Malthus, 1 46 Mari ner spacecraft, 2 1 2 Mars, 208-209, 21 2-1 7 Martian canals, 208 Mendel, 1 48 Mercury, 21 1 Metabolism, evolution of, 1 72-75 Meteorites, 1 3, 94, 1 99-204

Murchison , 202-204 Mi l ler, Stanley, 1 5 , 1 24-29 Molecular forces, 66-67 Molecular structure, 62-63 Molecules, self-replicati ng, 1 47-55 Montmorillonite, 1 36, 1 43 Moon, 2 1 0-1 1 Mutations, 57-59, 89, 1 49

236 Index

Natural selection, 1 77-83 chemical record and, 89 CITROENS and, 1 92-94 design and, 1 96 h istorical background, 1 1 -1 2 molecules and, 1 45-55 mutations and, 59

Needham, John de Turbeville, 9 Neptune, 21 1 Nucleic acids, 39-46, 66-69,

1 51 -59 primitive, 1 56 replication of, 39-40, 42-45,

66-69 Nucleotide bases, prebiotic

synthesis of, 1 30-31 Nucleotides. See Ribonucleotides,

Deoxyribonucleotides

Oparin, A. 1., 1 4- 15, 65, 1 24-25, 1 36 Optical activity, 1 59-67 Organic compounds, accumulation

of, 1 20-21 Oro, Juan, 1 30 Oxidative phosphorylation, 77 Oxygen, atmospheric, 1 09-10 Ozone, 1 1 7- 18

Paley, Wil l iam, 1 96 Panspermia, 1 3, 94-95 Pasteur, Louis, 4, 9-1 0, 1 60, 1 62 Permeases, 49-51 Photosynthesis, 76, 79, 1 1 0 Planets, 1 00-1 02, 21 9-221 Pluto, 21 1 Polymers, 36 Polynucleotides, 1 50, 1 51 -59. See

also Nucleic acids Polypeptides 1 42, 1 43, 1 51 -59. See

also Proteins formation of, 1 37-38 informed and non-informed, 1 54 thermal, 1 71 , 1 72

Prebiotic chemistry, 1 23-32 Prebiotic soup, 1 1 3, 1 34, 1 50, 1 53

optical activity of, 1 62 Proteins, 36-39, 69-72, 15 1 -59 Protein synthesis, 46-49, 80-85

primitive, 1 56

Rad ioactivity, 21 -23 Radioisotope dating. See Isotope

dating Redi , 5, 8 Replication, of molecules, 1 46-47 Ribonucleotides, 42 Ribosomes, 80, 85 RNA, 39-46, 80-85. See also

Nucleic acids messenger, (mRNA), 46-49,

83-85 transfer, (tRNA), 80-85

Salt flats, 1 34 Saturn, 2 12 Solar system, 1 00-1 03 Space travel , 225-26 Spallanzani , 9 Specificity site, 70, 72 Spontaneous generation, 4-1 1 Stereochemistry, 63 Sugars, prebiotic synthesis of, 13 1 Surface active agents, 63-66 Symmetry, mirror, 1 59-60

Technological societies, 21 9-224 Templates, 68, 1 52-53 Tide pools, 1 34, 1 43 Time paradoxes, 225 Translation, 46-49 Trilobites, 24

Uranus, 21 1 Urea, 1 24, 1 40 Urey, Harold, 1 5, 1 24

Valency, 63 Venus, 21 1 -21 2 Viking spacecraft, 209, 21 7 Vi ruses, 44, 54-55 Volcanoes, 1 04-1 05, 1 1 1 , 1 1 8-1 19,

1 39, 1 43

Wallace, Alfred Russell, 1 1 -1 2, 1 46, 1 78, 1 81

Watson-Crick pairing, 42-45, 66-69, 1 52

Wilkins, John, 208 Wohler, Friedrich, 1 23-24, 1 40

237 Index

()� oF' li (}() 7,) -u CF..u_ s tttudi.Mc vliJ().�61i--/J 3q?'� Clil...l..-S ME �e-e- y (tPJ/J1�� 5t AJd'tf

I 3 '11Jif-

fS�Hi;euu-- C� IS � �'r()YV/'£ J'llllif-31.//7 ?SfJ

� f;Jor6AJ> fis� MJn�rm �C?/n� ?�t/,r -- >er LrJ v�/':5 11\v L n ft.,Z: ��Vtr>B Po VIVUitv tt fVriJ t/'flf}AP 5 3'J/'z.. Y?J..I/.-'If/1 �· {�;fiZ.fiJ)HOJ!} Lft-1../ f/JIJl-UI' r9c--cO ilJo I rl- /'#.P'rfF;A/ !;fS' L-� y�?I(.M-tt¥AJ i'11Airtl£ {t?VI1f?ucrl#'� � /4£?Nc4JI£W L.f$Pl,

vY1D y;cvl.-£ C ONVp'Lt; r fJI'\) �I;:> 5 C � tl-fi/YltfL >J NL/' GQ"u.. !hVM&>y ;?�AJe-r-totU 72-f?L. - 5"3�2.

lif/o of COIL I S flk?ftN I �0/'o w.M'M.. t)if'3 VJilV>f.) t.J o 'r �itONOVVUAP� ;2{;/t...l t:� S"1NLJ1- 5.>"/'2. f-� (O!Ufct1EW 911-WN1-¥{� c;j-/'2- - )q[/� NI.J<-t.£1 L liCHOi, ,v&.c.. F'l#'f_ r_ff€. � /<;1J{)I.AhV b�3 S�P1 of lAIIi fYW > I (�IM£ Jli/A,trA s.fi/"4 '106£� f:wt/

i.3As(. llfttl- &f\.o,tVt�j!(...{c,L Ew1£Y�c::-£ 67-65 if �fNtY.:�'r7(. llt£"/Lic '1- Y7rM/ Of Nv t..t,..£! L /i-l1r1� I s r

hf:.I\J6:r(( /Mx.-£ $ J bq fl.. (',Ly/£,#rtiJE bA�IJoN£ IS �;;p 7iJ/'(

* J(.v�c,y /�C 65StrlJ b IN c£LL-5 1s-flz_-j/3 f'/1-t.'rr; 5'jA/'t1'/651$ £.5-5/iv'nfi& /f'r� II#: vA.J/I/£4yfL £Nt,.t v y' cull/tiNcf 76'-1 - 7112 f�T(JN >YtVrf'IC5fS flitxvt/l£j �f/)' 7�/L fttf NO!' ; sv- filffito'rl<.. ;o.fl.t_e_ Or E.NU&'j 7 1f2_

1\'e 'f<l<; V N l1#' �'> �!lmJtA- 7'1 /"I{ IW4' V() Mccvv '>rtivt-1" (yt{{;/11} oP lA y::'£ f'� EXt� r;l()b t.&

'$ 7---81/3

Vfll>{ co(ll'f'uu#fZIJ M-�5 /lW�'t" tMv'fZ 6i):tW llfl /f€.lt-t 11i/.) 91 fll ( t3uv- lWd Y� )

' OM!t!T'! AvLO?� T/f£ mE> 'D/flZ--vut LW ro t<\vfN.f /!kEN' ArJ fvo j/lkttl� <if'f /3 f/kTf-IAH CM'I\JOI ;Zi/L{c� C(lfllf

f'_f.A.viX ftW sr g,[ a!'l-'ti/:IN s ;u:_ 10 J-14.Ji �tJ.tt:cr fH'1rle1/llt{;4L f4AP1tV7tJ£. �s/1'11-� !Db el-s 1 02-if ·'/IJCIJ6£VJ "· £-9-£fiA.J€O Wrl'z_ � . �( c()h1/b...W;9J AJD'rsrMc...f" ;,v t!Je1Przu1Jb 1�1'5 �IL. �� [Ovf/!U �� J l{)if o!l&nvvK r�t> (!}vvt£ F/ZD1At ;11)¢6/UtJ/{_ JLJ{�2_ /VIJv.-fZ./l k.xlftu'v�FM/1 vN�11 (.. jt-b //Jl/ - I'"Z- 7/1

c 15M �:rAw·, ... t- 3 s-o81 tK>f '* -w"') A�lilJO'? f/tt?Nrk;P 13'1 frl� I Ybf/

N �r IAf€. /b1NI� rv!J!YA/ /Lft);? �Y6� /1..91'<-( �ltVf �� t3t>{2.. P,t.-tf't/!l(Jt} PtbW] /V-n'l;vO ;9?tp_s 1 � 3tvL-f' �1\1� �C-I{{)N11 S{�lf� f'1f-f'L /l..i/t.AL¥fflAIG t1JvCl.iftL � PJ 1'1�f'2-

AJ!k.££on JG� � �� A4,1Jfl" I ')O (l?; u,..lfi - t.Ml? t>of2 - t s-oNl/* 1-c�NLf I IIJo /h;rwv F41Z '> r 1 :; -ufz_

!a-.ylfil'n..fJf> plJflf( wwMFZ- l r;'1f3 r?N� /St't /ot.,yfVIf/L I t;r)f'3 L S ( &lf..vt:n{ ?ys r� �t£. &;IM/r__&' J 'ib'l/ �� fl/-oVhitV S'fN�>; j : t oO f,top;;tltJ> 't' JJVLl.£1C pi(A1JJ filttY! tr?V€. a>/)£ � No •• f!)tV {"/J.1f:JJ01/ � J� b Lf- 1�'-ffl't -- U�'TT£,.v)_ U41£Tt- ov?f_ fl'l,.> sr" 1-M.vC 6£t,v > vt�> #B-vlflfO) lfifr v, t'tl&tn�£o lbo<el-c.:r· �Nf.d JN f{_(),JI+�;IIJWVfl';i/ly l blf3 I l]!ltiJC

L�;No l!l£f/f1Al), D wu .. fo'vrl£:, AJ.A- . /12-f� � -tftMJfftJ IJvctfonotf> �J >//Mll"-'6 /6515 'l ''./vjv-So '' o;,,.-[Y'Z'£J co�rpAJb} 1171'Y-/5'1 T?o/.5 -/j'jp �'-- !YIE-,m6i/MIL AlE.cc.. ! /(p1 PI - J 7oi'J ---- )

� �;:rr..vl11 1t.J€«. , /!'J ;A.!� t1 �" 17DPJ ' 7 z..,P; -1-?z...

Yvtt{/ZP':Jifnfi,-1[ /3{6tAhV/AJ' / 7 / lf J ]]jlf/- j;i)/'1 f:,Q. ://l�;o;lc sou./' 1 72 1/J L-P :;..:::---

({ol..[ &€n c� f1't..o111 G�vws£ 1 77..AJt.J7 J lll#vt-r,qU. (tJYI1/LlYt TY J 1�fll Gv�{)/il �ot.unm 1{1),:?2 Dr;s.l&tt.1 t9tV lt.NtJSdJft..l I 'biL..-f?- I �3�/ Y¥c.tiH-fiJ ti�Yt'rY Nri/ -Nt/'c_tf

L JN��N ��?N@�� l-ffW lil€(J

ffinil. ,fy � ::-fW 1,v s'fll-vc'f'i�AJ/. lfoiiJlf iF t. (f)[{ 15 lf/18 � lftJII)iffr !1nJy uwltJ& {lhV & /l4s l'lz&lf J;f)fD {PIV!fi,v! /12/2.._ 1 ,qbltllf l.-1 fC.. b£i\J£SIS : £vo tM »1� � /f/lf3

50/i;vv-rsT!. c � /o!.tv£..'-4-r£ .f'Y'\ �z..£.> I�VfVLflf('..k fV�u CWJJI/mf)A}> fix- tfRi.. Jot3

lyoo ©00 @[f [l [f · neously in meat. he and his contemporaries thought that the spontaneous...

Documents