Sidney W. Fox ^HfiHAl Kaoru Harada ^|9K9M Gottfried Krampitz ^SHGu George Mueller xHrall Life as we know it is an association of many functions. Attempts to answer questions of the origin of life, accordingly, might most profitably proceed along the lines of the analysis of life into its individual functions. Instead, therefore, of asking the general question of how did life be- gin, we might more fruitfully inquire: • How did the first enzymes arise when there were no enzymes to make them? • How did the first cells arise when there were no cells to make them? • How did ordered sequences of amino acid residues arise in polypeptides when there were no macromolecules to di- rect them? • How did the first membranes arise to separate individ- ual cells from the environment? After such analytical thinking led to experiments in the second third of this century, new emphases emerged. One of these was an increased attention to the primitive as com- pared to the contemporary. The experiments have led toward definitions of primordial life. Such definitions may yield deeper insights into contemporary life. This set of priorities, which deals with evolution of the primitive to the contemporary has, in turn, led to a new emphasis on constructionistic, or "synthetic," experiments. The traditional reductionist, or "analytical,,, emphasis of biology has been able to tell us more and more about what is. Reductionism can hardly be expected to tell us how "what is" emerged from "what was," since "what was" can in the beginnings of research only be inferred. Although the statement may be trivial, the existence of life as we know it would be unbelievable were it not that our senses tell us it is here. The basically unique virtue of the con- structionistic approach is that it is in the same mode as the evolutionary processes that led to such a fantastically di- verse biota. Another emphasis that has emerged from experiments is that which focuses on systems instead of merely on mole- cules. The association of functions is found in the sys- tems; systems have many properties that could not have been predicted from knowledge of molecules alone. The experiments demonstrate, in addition, that systems and their complex macromolecular precursors could have arisen 80 C&EN JUNE 22, 1970 by exceedingly simple processes, although the resultant products and properties are remarkably complex. The answers that have been obtained in the manner de- scribed are more rigorous than were anticipated on first reflection. For the macromolecular and cellular ori- gins discussed here, the conditions used in the laboratory are found in abundance on the contemporary earth. The primitive earth had also those conditions beyond any significant doubt—moderate heat and water in that order. The thinking rests predominantly on the assumption that chemistry is chemistry wherever and whenever it occurs or occurred, an assumption that increasingly seems safe for at least our solar system. When we examine in more detail the sequence of main events suggested by the experiments, we find a flowsheet for the coherent model from amino acids to proteinoid mi- crospheres. Most of this article will deal with the chemis- try of transition between inanimate and animate models, but it will also touch on related topics of primitive nucleic acids, contemporary nucleic acids, lipids, and other sub- jects for which partial models are available. History In the 19th century, a number of experiments designed to answer the age-old question of how life began were performed. According to some historians of that period, and others since, Pasteur's famous experiments of the early 1860's dealt a mortal blow to the concept of spontaneous generation. What Pasteur's experiments did in reality was to disprove conclusions by others who believed that their experiments supported the concept. Disproof by Pasteur of the validity of the evidence then offered for spontaneous generation was of course not proof of the invalidity of the general theory. In 1878, Pasteur in fact reaffirmed the possibility of spontaneous generation. The general think- ing, however, remained in a confused and negative state until after Oparin published in Russian his first book "The Origin of Life." Oparin explained in outline, in 1924, how life could have come into existence through spontaneous terrestrial processes of organic chemistry. The first experiments interpreted in the context of chemi- Sidney W. Fox Kaoru Harada Gottfried Krampitz George Mueller rUFlftlCAL ORIGINS OF^ELLà 25652562556256255625 Very simple reactions can yield complex products and properties Starting material Amino acids Reaction Heat Product Polyamino acids (proteinoids) Water > Polyamino acids (proteinoids) Proteinoid microspheres cal pathways to life appear to be those of Groth and Suess, published in 1938 in Die Naturwissenschaften, and brought to our attention by Prof. Klaus Dose. These authors con- verted carbon dioxide and water to formaldehyde and glyoxal with ultraviolet radiation of short wavelength. Subsequently W. M. Garrison, D. H. Morrison, J. G. Hamil- ton, A. A. Benson, and M. Calvin reported in 1951 similar results with high-energy radiation in the Berkeley cyclotron. In 1953, S. Miller published experiments in which he pro- duced amino acids by electric discharges in a mixture of ammonia, hydrogen, methane, and water. Miller em- ployed a highly "reducing atmosphere," in accord with the concepts of Oparin and of Urey, and obtained amino acids. In 1966 this type of reducing atmosphere was criticized by P. H. Abelson as geologically contraindicated. Unfortu- nately, controversies on the nature of the primitive geo- chemical situation can hardly be resolved solely by ex- periments in the laboratory. A demonstration of the formation of amino acids was, however, dramatic since these compounds represent a first step to the proteins, hav- ing unique functions as structures, catalysts, and in mem- branes of the living cell. The development of the modern history of the subject coincides largely with the activities of the National Aero- nautics and Space Administration, which purposely fostered research in this area beginning in 1960. The mis- sion of this agency recognized a need to learn how life could begin on any planet, including the earth. Key properties of product Molecular order Enzymelike activities Self-assembly to cell-like units Structure, including ultrastructure Selective retention of large molecules Ability to proliferate by a kind of "budding" are often very different from the conditions under which carbon compounds are formed. The two sets of require- ments need to be analyzed separately; they represent dif- ferent stages in evolution. Another kind of error that has been committed often is that of employing reactions in closed flasks as an indication of what would occur in the open atmosphere of the earth. The correspondence is inexact, especially when fugacious hydrogen or volatile water is present in the flask. Organized microsystems have emerged from the products of some of the organic chemical experiments. These have been studied extensively. In examining terrestrial and ex- traterrestrial units much reliance has been placed on mor- phological characteristics. Morphology may be regarded as a necessary criterion of life, but not a sufficient one. When analyses and experiments are based largely on a given kind of morphology the interpretation is necessarily shallow. Morphology, however, is the first observation by Modes of thinking Each area of science has its own challenges and difficul- ties, and each develops its own methodology and modes of interpretation. In the area of physical and conceptual models of the origins of life, criteria of discipline and rigor used in other subjects do not always serve well. Much of this kind of difficulty can be traced to the fact that experi- ments being performed in laboratories in the 20th century are used to interpret what occurred 2 to 3 billion years ago in the terrestrial realm. The geophysical conditions that existed on the early earth may indeed exist at the present time, as uniformitarian geologists believe, and as Dar- win suggested. Although some investigators relate their findings and thinking to narrow sets of conditions, others believe that the terrestrial realm permitted a wide variety of possibilities. These particular interplays of concepts are related to the choice between two premises. One postulate states that life need have arisen only once, whereas the other proposes that life arose on many, or innumerable, oc- casions. The conditions under which life thrives, however, ! i These photomicrographs of microfossils by Barghoorn, on the left, compared to proteinoid microspheres, on the right, in- dictate obvious morphological similarities 82 C&EN JUNE 22, 1970 Extraterrestrial molecules and moieties could have been the raw materials for molecular evolution7 Location Interstellar space Red giant stars Atmosphere of sun Atmosphere of Venus Atmosphere of Mars Material CH, HCHO, CN, NH3, H, H20,0, (carbon particles, CO, H2, OH, NH, 0.) CH, CH2, CO, CN, H2 CH, CH2, CH4, CO, C02, CN, H2, SiH, 02, N2, N20, S2, (PH) C02, H20, Oi, (CO, HCHO, 0 „ N2) C02, (carbon dust, CH4, CF4, H20, NH3, 02, 0 „ N2, NO) Atmosphere of Saturn, Jupiter CH4, NH3, (H2, N2) Atmosphere of Uranus, Neptune CH4, (H2, N2) Atmosphere of Titan CH4 Atmosphere of comets C2, CH, CH+, CH2, CO, CN, OH, NH, N2, (C,H, C2H2, C2, H3, CS, 0H+, NH2, NH3, SH, S2) Tails of comets CH+, C0+, N2+ o Parentheses indicate uncertainty of their presence. which we recognize the existence of the cell as a unit of life. Although the organic chemistry must be sound for meaningful experiments, and the conditions must be rele- vant to the geological, evolutionary thinking also is a necessary part of the armamentarium of those who choose to relate chemical studies to the origins of life. The num- ber of laboratories that devote a major part of their atten- tion to this problem has been and is relatively small. This is probably because living systems and component macro- molecules are exceedingly complex. From this judgment it is easily possible to deduce that the processes by which these molecules arid systems arose were likewise complex. If they were to occur in the geological realm, however, one is led to infer that they would necessarily have had to be pro- cedurally simple. The experiments suggest, moreover, that evolution of molecular complexity was capable of occurring from simple beginnings very rapidly—in days or less. The evolutionary emphasis in these laboratories led to producing polymers containing 20 monomers, the a-amino acids common to protein. Basically, evolution has em- ployed an array of macromolecules such that specific and precise variations could exist within the borders of this array. Especially was this array needed for the vast num- ber of enzymic activities that the evolving cell has ex- ploited during evolution of organisms. Accordingly, the meaningful beginning of the principal research program described here is often marked by the first successful at- tempts to combine in a simple and geologically plausible process all of the 20 kinds of amino acids common to con- temporary protein. Raw materials The raw materials for these amino acids and other sim- ple biologically significant molecules were simpler com- pounds. The six elements that constitute the bulk of the organisms—carbon, hydrogen, oxygen, nitrogen, phosphorus, and sulfur—are all abundant in the cosmos. The table above summarizes what has been inferred from spectros- copy about the occurrence of simple compounds in the solar system and beyond. With the gradual condensation and cooling of a given celestial body, the substances tabulated condensed into the "carbospheres," through the action of ionizing radiation, electrical discharges, cosmic rays, radioactivity, and heat. The earth is believed to have been condensed from dust about 4.5 billion years ago. The evidence for the oldest life (or prelife) is dated as of 3.1 billion years. The bio- logically essential organic compounds, therefore, would have formed on the primitive earth abiotically 3.0 to 4.7 billion years ago. At the Apollo 11 conference in Houston, an age of 4.7 billion years was reported also for the moon. During the condensation of planetary dust clouds, espe- cially in the later stages of the process, organic compounds would have been synthesized by thermal reactions cata- lyzed by dust and metal (iron) particles. Radiative en- ergy from the sun could not reach the condensing proto- earth because the dust cloud filled the planetary space. Most of the organic compounds would have escaped into this space. However, some of the organic compounds syn- thesized in the final stage of the earth's condensation proc- ess would have been retained on the surface of the primitive earth. When dust disappeared from the solar system, uv radiation from the sun could become an active force. Volcanic activity on the primitive earth was likely much greater than that of today. Electric discharge and radio- activity would also have energized reactions on the primi- tive earth. However, radioactivity would have been largely confined to the crust of the earth rather than being effec- tive in the atmosphere. The primitive atmosphere must have served as a source of organic substances on the earth. The so-called "primi- tive atmosphere" of the earth is now regarded as of secon- dary origin, that is, as having arisen by outgassing from the JUNE 22, 1970 C&EN 83 interior of the earth. The composition of the atmosphere would be quite different from that which would be ex- pected from the cosmic abundance of elements in the uni- verse. The primitive atmosphere was presumably strongly reducing 4.5 billion years ago. Methane, ammonia, and water could have been the hydrogenated forms of carbon, nitrogen, and oxygen. The reducing primitive atmosphere would be converted gradually to the present day oxidizing atmosphere by the formation of oxygen (2H 2 0 -> 2H2 + 0 2 ) by photodissociation of water in the upper atmosphere of the earth. The rate of oxygenation was accelerated later by the photosynthetic processes of plants. Laboratory experiments in simulation have, therefore, been carried out by using mixtures of representatives of the group of CH4, CO, C0 2 , NH3 , N2, H2S, H2, and H 2 0 under overall reducing conditions. Energies used have been uv radiation, ionizing radiation, electric discharge, and heat. These are recognized as simulations of solar radiation, ra- dioactivity of the crust, lightning in the atmosphere, and magmatic activity, respectively. Volcanic lava represents a minute fraction of the heat which has been available from magma. During 1969, new information from microwave sensing indicates that interstellar matter includes significant pro- portions of formaldehyde, ammonia, and water in addition to diatomic hydrogen. Should this be confirmed, the stage of molecular evolution in the universe in general is more advanced than has been credited, and some radical changes in thinking are engendered. Proteins and nucleic acids are of course the most signifi- cant materials among all biomolecules. Attempts to syn- thesize their monomelic constituents, the amino acids, pu- rines, pyrimidines, and sugars were mostly carried out at first. Development of chromatographic techniques, espe- cially the development of the automatic analyzer, made the analyses of amino acids and nucleic acid moieties more convenient and accurate. Terrestrial fractionation The earth fractionates the products of its syntheses on a large scale. Although organic chemists characteristically incorporate fractionation into their processes, this aspect has been largely ignored in the interpretation of organic geochemistry in our context. Recent geochemical observations by Mueller indicate that processes of hydrogenation, polymerization, condensa- tion, and fractionation resulting from hot spring (hydro- thermal) and pneumatolytic (fumarole) activity may par- tially fill the evolutionary gap that separates the hetero- geneous primary carbonaceous complex from the highly specialized composition of the organism. In an unpub- lished study, the polar phase in a Derbyshire, England, carbonaceous complex has been found to contain one or- der of magnitude more amino acids and two orders of mag- nitude more organic acids than the mean for the carbona- ceous complex. Similar differentiated carbonaceous com- plexes have been found in association with hydrothermal veins from southwest Africa, Ontario, Canada, California, and New York. Amino acids Stanley Miller was the first to synthesize organic com- pounds containing nitrogen in a simulation experiment. He used a mixture of methane, ammonia, water, and hydro- gen, and employed electric discharge. Among the organic compounds identified were glycine, alanine, aspartic acid, glutamic acid, and ^-alanine. Miller proposed a Strecker-type mechanism for the amino acid syntheses. "R-CHO + HCN 4- N ^ - ^ R - C H - C N R-CH-CN + 2HzO - ^ R-CH-C0CT+ NH3 NIH2 NH3 + However, Ponnamperuma later found that aminonitriles could be synthesized without water during electric dis- charge. This suggested that the amino acids could be formed without aldehydes. The pattern of the amino acid composition in the Miller experiment is similar to that of amino acids synthesized by the oligomerization of hydrogen cyanide, which was found to be a primary product of the spark experiment. Therefore, one reasonable mechanism is oligomerization of hydrogen cyanide and subsequent hy- drolysis. The electric discharge methods were further studied by Abelson (1953-54), by Heyns (1957), by Pavlovskaya and Pasynskii (1959), by Grossenbacher and Knight (1965), and by Ponnamperuma (1966). Ultraviolet light could also be used to form amino acids or their precursors. Abelson found that glycinonitrile re- sulted when a solution of ammonium formate, ammonia, sodium cyanide, and ferrous ion was irradiated with uv light. In 1960 Groth and von Weyssenhoff reported the formation of glycine, alanine, and aminobutyric acid by irradiation with uv light of methane or ethane with am- monia and water. An early experiment employing ionizing radiation was that of Dose and Rajewsky. Irradiation with x-rays and 7-rays of a mixture of methane, ammonia, water, hydrogen, carbon dioxide, and nitrogen resulted in the formation of several amino acids. In 1957 Hasselstrom identified gly- cine, aspartic acid, and diaminosuccinic acid from irradia- tion of ammonium acetate with /3-rays. Palm and Calvin, in 1962, produced glycine and alanine by exposing a mix- ture of methane and ammonia to 5-MeV electrons. In 1955, Fox, Johnson, and Middlebrook prepared as- partic acid by heating ammonium fumarate or ammonium malate and hydrolyzing the products. Nine years later Harada and Fox reported the thermal synthesis of amino acids from methane, ammonia, and water. This gas mix- ture was passed through silica gel in a fast vapor phase re- action in a quartz tube that was heated to about 1000 °C. The reaction product was collected in aqueous ammonia and then was hydrolyzed. Amino acid analysis indicated the formation of most of the amino acids found in protein. Those identified were aspartic acid, threonine, serine, glu- tamic acid, proline, glycine, alanine, valine, isoleucine, leu- cine, tyrosine, and phenylalanine. In this experiment, large proportions of /^-alanine were formed, but little of un- natural amino acids was observed. Formation of phenyl- alanine is especially interesting as a hydrogen-poor amino acid arising from a hydrogen-poor atmosphere. In 1966 Taube confirmed these results in large part and found ly- sine also. Oro performed a similar thermal experiment (1965) but without any catalyst, and obtained largely glycine and alanine. The formation of glycine from hydrogen cyanide has been known for many years. Revival of this type of study was initiated by Oro in 1961. Glycine, alanine, and as- 84 C & E N J U N E 22, 1970 partic acid were synthesized from an aqueous solution of ammonium cyanide. Several investigators—Abelson, Mat- thews, Harada—have emphasized the role of oligomeriza- tion of hydrogen cyanide in the abiotic formation of amino acids. Abelson reported that uv light accelerates the oli- gomerization of hydrogen cyanide and that the oligomer yielded glycine, alanine, serine, aspartic acid, and glutamic acid. Harada has demonstrated the formation of amino acids, similar to the results of Oro and Abelson, by thermal decomposition of form amide. Because of the formation of hydrogen cyanide in the thermal decomposition of forma- mide, formation of amino acids could also be due to the oligomerization of hydrogen cyanide. In 1966 Sanchez, Ferris, and Orgel suggested cyanoacetylene as a possible intermediate for the formation of aspartic acid. Purines and pyrimidines Hydrogen cyanide figures significantly not only in the formation of amino acids, but also as an intermediate in the synthesis of purines and pyrimidines. The first such synthesis was carried out by Oro (1960); he produced adenine from an aqueous solution of hydrogen cyanide. 4-Aminoimidazole-5-carboxamidine and formamidine were probable intermediates in the reaction. The adenine is a pentamer of hydrogen cyanide. N=C-NH2 By using anhydrous hydrogen cyanide Wakamatsu (1966) greatly increased the yield and produced an industrial proc- ess. Adenine was synthesized by Ponnamperuma, Lem- mon, and Calvin (1963) by electron irradiation of methane, ammonia, and water. Since hydrogen cyanide was identi- fied, the mechanism of adenine formation again could have been oligomerization of hydrogen cyanide. In the study of Ponnamperuma, Lemmon, and Calvin, the formation of adenine was enhanced by the absence of hydrogen. This suggests the interpretation that the principal accumulation of bio-organic molecules on the primordial earth might have taken place when most of the hydrogen had escaped from the primitive atmosphere. Under these conditions, one can more readily rationalize the appearance of hydrogen-poor compounds such as phenylalanine and heterocyclic com- pounds, as in the thermal experiments mentioned earlier. In 1967 Sanchez, Ferris, and Orgel demonstrated the role of aminomalononitrile in the synthesis of adenine. Amino- malononitrile, HCN trimer, reacts with hydrogen cyanide to form the tetramer of hydrogen cyanide, which is con- verted photochemically to 4-amino-5-cyanoimidazole. This compound reacts with hydrogen cyanide to form adenine. Uracil was prepared from urea and malic acid by heating with polyphosphoric acid. Ferris, Sanchez, arid Orgel (1968) synthesized cytosine from cyanoacetylene and cyanate. In the middle of the last century, Butlerow (1861) dem- onstrated the formation of sugars from an aqueous solution of formaldehyde in the presence of alkali. More recent de- velopment of separative techniques made it possible to analyze Butlerow's sugar. Formation of ribose from for- maldehyde was reported by Pfeil and Ruckert in 1961 and by heating formaldehyde with kaolinite by Gabel and Ponnamperuma in 1967. In 1962 Oro and Cox explained the origin of deoxyribose in a Federation Proceedings note. Of-C 5HCHO —^H 2 coH(HCoH) 3 CHO When adenosine was treated with orthophosphoric acid, no detectable amount of adenylic acid was formed under uv irradiation. However, when polyphosphate ester, prepared from P 2 0 5 and ethyl ether, was used instead of phosphoric acid, adenylic acid was synthesized. In addition to yielding adenylic acid, the artificial phosphorylation reagent led to detectable amounts of adenosine diphosphate (ADP) and adenosine triphosphate (ATP). Ponnamperuma and Mack (1965) found that heating nucleosides with sodium dihy- drogen phosphate at 160 °C led to the formation of various monophosphates. Waehneldt and Fox demonstrated the phosphorylation of nucleosides with polyphosphoric acid under mild reaction conditions (0 to 22 °C) . Monophos- phates, diphosphates, and triphosphates were each formed in substantial yields (25 to 45% total). The formation of other organic compounds, such as fatty acids and porphyrinlike substances, has been studied under presumably primitive terrestrial conditions. Several at- tempts have been made to synthesize porphyrinlike sub- stances. Tetraphenylporphyrin was isolated from the reac- tion mixture of pyrroles and benzaldehyde after y irradia- tion by Szutka in 1959 and later, in 1965, by using electric discharge. Hodgson and Baker (1967) have shown that porphyrins can be formed from pyrrole and formaldehyde. Macromolecules The origins of macromolecules, such as proteins or nu- cleic acids, seemed earlier to constitute a most formidable aspect of the total problem. As stated, this judgment arises easily from an examination of the structural complexity of biomacromolecules. In an operational sense, however, polymerization of monomers occurs with great ease. This facility is such that, in a number of experiments reporting the production of monomers, hydrolysis of the polymers formed has been necessary—to permit recovery of the mono- mers. The largest measure of success in abiotic production of proteinlike macromolecules has been achieved by pyrocon- densing dry amino acids. Numerous reports of models of prebiotic condensation of amino acids have appeared—Oro, Steinman, Lemmon, Calvin, Ponnamperuma—as well as the synthesis and progressive substitution of polyglycine by Akabori. The thermal process is the only one that has been shown to yield simultaneously: • Macromolecules of molecular weight in the thousands. • Some proportion of each of the amino acids common to proteins. • Polymers having a variety of enzymelike activities. • The tendency to form cell-like structures upon contact with water. Such polymers are known as proteinoids. The direct polymerization of amino acids is inhibited by the presence of much water, as the prototype reaction indi- cates: H3NCHRC00" + H ^ N C H R ' C O O " " : ^ : \\\ NCriRCOMHCHRCOoT + H20 A 6 ° - \Boo to 4 0 0 0 The K's calculated from this range of values indicate equi- librium far over on the side of hydrolysis, more than 95%, for example. Only small yields of small peptides in dilute JUNE 22, 1970 C&EN 85 These representative abiotic syntheses demonstrate ways I Compound class Reactants 1 Amino acids Purines, pyrimidines CH4f NH3, H2f H20 Ammonium fumarate C02f IMH3, H2, H20 CH4, NH3f H20, H2f C02, N2 Ammonium acetate Ammonium carbonate CH4f NH3f H20 NH3t HCN, H2 CH4f NH3f H20 CH4f NH 3 F H20 CH=C—CN, HCN, NH3f H20 HCN, NH3, H20 Malic acid, urea, Energy Electric discharge Heat Electric discharge X-ray T-ray /3-ray uv Heat (70 °C) Accelerated electron Heat (>850 °C) Heat (100 °C) Heat (70 °C) Heat (130 °C) polyphosphoric acid CH3, NH3, H20 Accelerated electron Products Amino acids, hydroxy acids, HCN, urea Aspartic acid Amino acids Amino acids Glycine, aspartic acid, diaminosuc- cinic acid Glycine Glycine, alanine Amino acids Glycine, alanine Amino acids Aspartic acid Adenine Uracil Adenine Investigators Miller: 1953; 1955 Fox, Johnson, Middlebrook: 1955 Abelson: 1956 Dose, Rajewsky: 1957 Hasselstrom, Henry, Murr: 1957 Paschke, Chang, Young: 1957 Groth, von Weyssenhoff: 1957 Ord, Kamat: 1961 Palm, Calvin: 1962 Harada, Fox: 1964 Sanchez, Ferris, Orgel: 1966 Ord, Kimball: 1961 Fox, Harada: 1961 Ponnamperuma, Lemmon, Mariner, Calvin: 1963 aqueous solution are permitted by the analysis of the ener- getics, unless the bond formation is coupled with energy- yielding reactions. The fact that some chemists have at- tempted to produce polypeptides in dilute aqueous solution, with agents such as cyanamide and dicyandiamide, has oc- casionally been explained on the basis that organisms are predominantly aqueous. No evidence is at hand, however, that protein biosynthesis occurs in dilute aqueous solution. Ribosomal particulates seem to be used in all, or nearly all, cases. These particles are not composed of dilute aque- ous solution. Also, many key reactions in cells occur in hydrophobic regions of enzymes, lipid layers, and the like. A tendency for small molecules to decompose in water is rather well known to experimental organic chemists and was documented by Hull in 1960. The tendency for large molecules to decompose in water, as we have pointed out, has not always been fully appreciated. Some data show high-energy radiation to be more destructive to macro- molecules when wet than when dry. Theoretically, temperatures above the boiling point of water were visualized as one geologically plausible way of overcoming the thermodynamic barrier to synthesis of mul- tiple peptide bonds in the same macromolecule. The acti- vation of primitive terrestrial organic reactions by energetic conditions other than temperatures above the boiling point of water has often been suggested. Other forms of energy have, according to such analyses, been thousands of times as abundant as heat at such temperatures. Such statements have generally been based on estimates of volcanic lava, an estimate that fails to represent more than a small fraction of available terrestrial heat. Energies such as uv radiation and electric discharge are, moreover, of utterly no avail if macromolecules are decomposed by these agents. As the preparative organic chemist knows, heat can be gentle and can be controlled to avoid severe decomposition. The inclusion of sufficient proportions of aspartic acid, glutamic acid, or lysine has been shown to be necessary for the pyrocondensation of amino acid mixtures. At first, a proportion of at least 25 to 30% of one of these nonneutral amino acids was believed to be necessary, and led to pro- teinoids in yields of 10 to 40%. More recently, experi- ments have shown that smaller proportions of aspartic acid, glutamic acid, lysine can be employed. Equimolar proportions of 18 amino acids have been used. Precise variations are feasible. The ease of the syntheses is dem- onstrated by the fact that 10 precisely varied proteinoids can be produced within seven working days, analyses being performed subsequently. Except for substantial but in- complete decomposition of serine, threonine, and cystine the amino acid contents of thermal proteinoids resemble those of proteins. Molecular weights are in the range 3000 to 11,000. The proteinoids give all the common protein color tests. Depending on the amino acid mixtures used, the proteinoids resemble in solubility the albumins, the globulins, the his tones, and others. They can be salted in and salted out. They are precipitated by protein reagents such as phosphotungstic acid. When prepared from opti- cally active amino acids, the proteinoids are found to be optically active, but some of the monomers are recoverable from hydrolyzates only as the racemates. The ir absorption patterns are like those of contemporary protein except for an aspartoylimide band; these linkages are hydrolyzed 86 C&EN JUNE 22, 1970 small organic compounds could have Compound class Reactants Sugars Nucleotides Hydrocarbons Porphyrins HCHO, CH3CHO; glyceralde- hyde, acet- aldehyde, Ca(OH)2 HCHO Adenosine, polyphosphate ester Nucleoside, phosphate Nucleosides, polyphos- phoric acid Methane Methane Pyrrole, benzaldehyde Pyrrole, formal- dehyde, N i 2 + Cu 2 + CH4f NH3, H20 Energy Heat (50 °C) uv uv Heat (160°C) Heat (22°C) Electric discharge Heat (1000 °C; silica gel) 7-rays Electric discharge arisen spontaneously Products 2-Deoxyribose 2-Deoxyxylose Ribose, deoxy- ribose AMP, ADP, ATP Nucleotides Nucleotides Higher hydro- carbons Higher hydro- carbons Tetraphenyl- porphyrin Porphyrin Porphyrin Invest igators Oro, Cox: 1962 Ponnamperuma: 1965 Ponnamperuma, Sagan, Mariner: 1963 Ponnamperuma, Mack: 1965 Waehneldt, Fox: 1967 Ponnamperuma, Woeller: 1964 Or6, Han: 1966 Szutka, Hazel, Menabb: 1959 Hodgson, Baker: 1967 Hodgson, Ponnamper- uma: 1968 easily to yield peptide bonds, as indicated by ir patterns. The polymers are, as a whole, hydrolyzed by proteolytic enzymes, although quite slowly. The proteinoids have nu- tritive quality, for which they are being further investi- gated. Appropriate thermal polymers possess hormonal (MSH) activity. Others possess enzymelike activity, as explained later. An especially striking feature of the polymers formed by condensing a-amino acids by heat is their limited hetero- geneity. If the proteinoid preparation had been randomly constituted, a horizontal line would have, theoretically, re- sulted in the plot above/below. The variety is, however, internally limited. The significance of this fact to the primordial events is that prebiotic nucleic acids were not needed to specify amino acid residue sequences in the first proteins. According to the experiments, the necessary in- formation could have been fed in by the diverse reacting amino acids. This absence of a need for prior nucleic acids was stated by two of us with Vegotsky in 1959 and several times since. Steinman has recently claimed evidence for selective reactions in producing dipeptides, and has drawn the same inference of self-ordering. The production of polynucleotides, by heating mononu- cleotides, has been reported by Schramm and his coworkers and by Schwartz from this laboratory. Schramm has not presented detailed evidence of natural linkages in his poly- mers prepared with the ungeological ethyl metaphosphate. Schwartz has found that a polymer of cytidylic acid (but not other mononucleotides) can be produced by heating with polyphosphoric acid. The validity of this mineral acid as a primitive terrestrial agent has been debated, but it is more directly defensible than Schramm's reagent, prepared in chloroform and ether. In addition, polyphosphoric acid or salts arise by heating from phosphoric acid or salts. Substantial proportions of Schwartz's polymers are attacked by ribonuclease and by venom phosphodiesterase, indicat- ing natural linkages. A number of biologists have seriously entertained the idea that nucleic acids arose at an advanced stage in evolu- tion. Some experiments are consistent with that concept and others are consistent with Calvin's concept of a simul- taneous origin of protein and nucleic acid for a type more contemporary than we have been discussing. These ex- perimentally based interpretations do not, however, obviate other concepts. Experiments that should deepen our un- derstanding of this aspect are being performed currently by the simultaneous condensation of mixed adenylic acid an- hydrides of the amino acids common to protein. The pos- sible origins of "amino acid adenylates," which are universal intermediates in protein biosynthesis, have been partly ex- plained. The use of this energy-rich form enables the con- densation of amino acids in aqueous solution. 1 R-C-C-O I I ri o 1 "o-P-—c II o Nl NH- l L II X H ^ N / W N CH / JUNE 22, 1970 C&£N 87 The presence of the nitrogen base in the reactive monomer suggests, conceptually, pathways to polynucleotides; ex- periments having this goal indicate conversion of the simple adenylate simultaneously with the polymerization of amino acids. One can thus postulate the simultaneous origins of protein and nucleic acid in a contemporary system. Un- published experiments also indicate selective complexing of condensation products of adenylates by added poly- nucleotide. With the adenylates as with the free amino acids, the as- sembly of amino acid residues is nonrandom. Throughout the discipline of molecular evolution, internally self-pro- moting processes are observed, but nowhere as clearly nor with as extensive documentation as in the condensation of heterogeneous mixtures of amino acids. Origin of enzymes Since contemporary enzymes are formed by pre-existing enzymes, we face the fundamental question (Dixon and Webb, 1952) of: If enzymes are formed only by enzymes, how were the first enzymes formed? An acceptable answer to that question in a model re- quires at least simultaneous weak catalytic activities. As Calvin pointed out years ago, all that was necessary at first were weak activities. These could later be enhanced and specialized by the processes of evolution. Accordingly, a first question to put to any physical model of "urprotein" is whether it has properties resembling those of enzymes. At least seven laboratories have published re- ports of catalytic activities in synthetic proteinoid. So far, five kinds of rate-enhancing or enzymelike activ- ity have been found for thermal proteinoid. These are hydrolysis, decarboxylation, amination, and deamination; 0.15 a recent report by Dose indicates peroxidase activity in hemoproteinoid. The first report of catalysis by thermal proteinoid, by Rohlfing and Fox, was the hydrolysis of the nonorganismic substrate p-nitrophenyl acetate. Systematic deletion of amino acids from the polymers showed that histidine resi- dues were essential for the activity. Heating buffered solu- tions of proteinoid or of copolymers of aspartic acid and histidine residues resulted in loss of activity, which was correlated with hypochromicity. A change in conforma- tion was thus inferred. Usdin, Mitz, and Killos (1968) prepared thermal poly- amino acids from several «-amino acids often implicated in active sites—glutamic and aspartic acids, histidine, serine, and tyrosine—and studied these for the same type of ester- ase activity. They state that the activity per milligram toward p-nitrophenyl acetate was only one order of mag- nitude less than the activity of chymotrypsin measured un- der similar conditions. Michaelis-Menten kinetics were demonstrated. The active polymer was inhibited by di- isopropyl fluorophosphate; it could be restored almost com- pletely by classical reactivators. Oshima (1968) found that proteinoids also catalyze the hydrolysis of p-nitrophenyl phosphate. The activity could be increased by gel filtration. Loss of activity was caused by several enzyme inhibitors or by heating in buffered solu- tion. Thermal proteinoid has also been found to enhance the rate of conversion of more fully biological substrates. The strongest activity of proteinoid that has been identified so far, by Rohlfing (1967), is that of decarboxylation of oxaloacetic acid by lysine-rich proteinoid. Thermal homo- polylysine was about 15 times as active as free lysine, and three to four times as active as polylysine prepared from 0.05 10 20 This elution pattern from DEAE-cellulose of fractionated l:l:l-proteinoidamide is neither symmetrical nor horizontal—the expectation if the preparation were randomly constituted. 30 40 50 To the contrary, an internal self-ordering of amino acids in the polymers is indicated. Many other kinds of evidence yield the same inference —that internal self-ordering occurs 88 C&EN JUNE 22, 1970 Leuchs' anhydride. By this kind of decarboxylation py- ruvic acid is formed. Acetylation resulted in more than 90% loss of activity. Krampitz and Hardebeck found that pyruvic acid also is oxidatively decarboxylated in the presence of thermal proteinoid. The pH optimum for this reaction is 8.3. This reaction yields predominantly carbon dioxide and acetic acid, but also 3-hydroxybutanone, acetaldehyde, and traces of ethanol. The carbon dioxide was found to be de- rived from C-l and not from C-2 or C-3 of pyruvate. The acidic type of proteinoid was more active than the lysine-rich type. Heating proteinoid in a dry state up to 170 °C had no significant effect. However, heating the proteinoid in a buffered solution caused a drastic, albeit not entire, loss of activity. Experiments with simple copolymers of amino acids demonstrated that copolymers of glutamic acid and threo- nine or leucine are many times as active on pyruvate as proteinoid or any of the other thermal polymers tested. Other substrates that are decarboxylated at an enhanced rate of conversion in the presence of acidic proteinoid are: glucuronic acid, a-ketoglutaric acid, pyruvic acid, glycolic acid, glyoxylic acid, and oxalic acid. Each of these de- carboxylation reactions has its own specific pH optimum. The inclusion of metals such as zinc or copper as co- factors in a solution containing lysine-rich proteinoid ac- celerated the amination of various keto acids in the pres- ence of Cu2+ at pH 7.0. This reaction uses urea, or in some cases ammonium salts, as the NH2-group donor. Such amination reactions have been observed for a-keto- glutaric acid, pyruvic acid, phenylpyruvic acid, oxaloacetic acid, and glyoxylic acid. Acidic or neutral proteinoid was not active at all in the amination reactions. Of particular relevance is the fact that proteinoid and Cu2+ must both be present. The in- clusion of pyridoxal phosphate or pyridoxamine in the re- action mixture had no effect. Other substances such as nicotinamide, adenine, adenosine, AMP, ADP, ATP, poly- A, RNA, DNA, and most amino acids gave positive re- sults when used as NH2-group donors in the described reaction, as well as ammonium salts. The amination of keto acids with ammonium salts was reversed by using Cu+ instead of Cu2 + . This yields a fourth type of reaction enhanced by proteinoid, deamina- tion. Glutamic acid is deaminated by basic thermal pro- teinoid; the resulting product is a-ketoglutaric acid. The pH optimum is also 7.0. The mechanism of this reaction, requiring an oxidation step, has not been explained. In each case of activity reported here, mixtures of amino acids showed either no action or much less than the poly- mers that were formed from them. This suggests that only polymers rather' than monomers first functioned as en- zymes. Moreover, only polymers would be retained by cellular membranes. Several studies have shown that the higher the molecular weight of a thermal polymer, the higher is its activity. For each type of substrate, kinetic studies have supported the inference of an intermediate proteinoid-substrate complex. In a few instances, the pro- teinoid has been shown to be a true catalyst by the usual criteria. In most cases, no more than rate enhancement can yet be claimed. The thermal polymers of amino acids promise supple- mentary insight into the structure and the mechanism of action of contemporary enzymes. A constructionist ap- proach may thus be added to the traditional reductionist understanding of the relationship of structure to function. The results reported in the literature on activities of synthetic proteinoid explain how enzymes or their evolu- tionary precursors could have come into existence in the absence of enzymes to make them, on this or other planets. One can also infer that enzymelike activity is not a sure qualitative sign of life as we know it, despite the fact that contemporary forms of life are heavily dependent on en- zymes. To develop the picture for all enzymes, of course, would require much more research. Origins of metabolism Otherwise stated, the results described explain in prin- ciple how enzymes were formed in the absence of living systems. Because of the variety of reactions enhanced or catalyzed by individual thermal proteinoids the question arises as to how these individual reactions could contribute to a model of a "protometabolism." Indeed, some con- temporary proteins are found to be devoid of a given ac- tivity studied in a proteinoid. This fact suggests that a proteinoid type of polymer might have served as "urpro- tein" more successfully than many or all contemporary biopolymers. For the origins of contemporary metabolism we also need conceptually to have a stage of evolution that has already reached the level of organization of macromolecules into structural elements. The spontaneous formation and the properties of organized microparticles have been men- tioned earlier and will be described in detail later. These particles, as models for protocells, are composed of pro- teinoid that can enhance the various reactions already dis- cussed. Our modern view of metabolism calls for a com- partmentalization of reactions within the cell. Such a formation of compartments has been observed within pro- teinoid microspheres. So far, three sequences of biochemical reactions en- hanced by proteinoid can be integrated. The simplest sequence begins with glyoxylic acid, which can be both decarboxylated and aminated. The decarboxylation is en- hanced by the acidic type of proteinoid. The amination is brought about by lysine-rich proteinoid—the acidic and neutral types of the synthetic polymer are entirely inactive. Another route is established on the basis of the various reactions of a-ketoglutaric acid, as a member of the citric acid cycle. The decarboxylation of a-ketoglutaric acid by proteinoid simulates one step of the natural cycle. Fur- thermore, transamination can be visualized in connection with the origin of the Krebs cycle. An even more extended sequence of reactions begins with both the decarboxylation and the amination of oxalo- acetic acid. Pyruvic acid, which plays a key role in cell metabolism, is both oxidatively decarboxylated and ami- nated by different kinds of proteinoid. These sequences, shown on the next page, can be interpreted to explain how keto acids were formed from amino acids. Such examples demonstrate a beginning for the integra- tion of glycolysis, the Krebs cycle, and transamination in an early organism or in the protocell. Much needed is an understanding of how, biochemically, the first cells trapped energy. Many of these reactions occur when the intermediates are heated in aqueous solution. Their facilitation at lower temperatures and specific control can be visualized as aris- ing from increasingly specific and specialized proteinoids. JUNE 22, 1970 C&EN 89 This photomicrograph of proteinoid microspheres shows compartmentalization within the microspheres. Current views of cell metabolism call for compartmentalized reactions within cells, thus these microspheres fulfill this requirement Optical activity The stereospecific nature of the components and reac tions of living units is a fundamental characteristic. The configuration of amino acids in contemporary living things is almost entirely L. This applies to all mammalian pro teins studied and to most, or perhaps all, of the proteins of organisms lower in the evolutionary scale. The mono saccharides found in organisms are mainly of the D con figuration. These configurational biases are found in their most complex manifestations in the nucleoproteins. Nucleic acids containing either D-deoxyribose or D-ribose react with basic proteins, themselves consisting of amino acid residues of the L configuration. At the level of the amino acid, any one DL-amino acid is thermodynamically more stable as the racemate than as ^! O/j&oarific HO cuva- NHi dovtchV ° * C - C ^ WO H PH 1.0 Cu* c-c -c -c C02 ρΗ 5tSH P H 7 . 0 Cu+ CO? PM 8.3 H * o I I // H - C - C - C I ' O H HH% H - c - c - QjlX/dAikdA^ (uuL H - C - C This reaction sequence shows a way keto acids can be formed from amino acids by various kinds of proteinoid and how metabolic pathways might have originated either the L- or D-enantiomorph. At the next higher level, that of the polymer, the structures composed of L-amino acids solely are thermodynamically more stable than those typically polymerized from mixed D- and L-amino acids. The reason for this is largely steric, the residues of a single configuration being most easily accommodated to the pep- tide backbone. This kind of stability is discernible in the more complex systems but is absent from the less complex monomers. In the case of the polymer composed of purely L-residues, the molecule is more highly ordered than other- wise, at the same time that it is thermodynamically more stable. The explanation of the origins and development of op- tical activity of molecules found in living systems includes the older ideas of circularly polarized light and dissym- metric quartz crystals. Another possibility was demon- strated by Garay in 1968; he reported stereoselective destruction of D- or L-tyrosine by irradiation with /?-rays. This is one version of the explanation that the optical ac- tivity may be related to the nonconservation of parity (Fox, 1957; Ulbricht, 1959). According to this thinking, the existence of L-amino acids on the earth and on other other planets in the universe is predetermined because of the dissymmetric nature of matter. If the racemate is a mixture rather than a compound of the two enantiomorphs, the racemic mixture is easily resolved from supersaturated aqueous solution. Ex- periments by Harada (1965) on resolution of DL-amino acids by inoculation with the seed of one configuration have yielded in a maximally simple way L- or D-amino acids of optical purity greater than 90%. On the primitive earth, the necessary seed could have originated by meteoro- logical separation of a D- from an L-crystal formed in a racemate that was also of the mixed type. This sugges- tion was made by one of us in 1956, and elaborated by Northrop in 1957. Prebiological resolution might have occurred spontane- ously during the polymerization of amino acids, or other monomers, to favor a single configuration—as mentioned earlier in the explanation of the stability of proteins com- posed of L-residues. Experiments by Doty and his co- workers using L- and D-Leuchs' anhydrides have shown that the L-form polymerizes faster and further than the DL-type. Other explanations of the origin and develop- ment of optical activity are scattered in the literature. A process originally suggested by Langenbeck was the slow generation-by-generation enrichment of one form from a statistically slightly disproportionate L,D-mixture because of the evolutionary pressure for a single configuration, the latter being more biochemically efficient. In addition, once a degree of configurational bias was established by natural experiment, that one-sidedness would undoubtedly trigger additional one-sidednesses. The relationships between (poly) amino acids and (poly)nucleotides in contemporary genetic systems are stereospecific. This fact suggests that optical activity of key biomolecules was already largely or entirely accom- plished when the contemporary code appeared in organisms that did not contain it until that time. Model experiments are being used to test this concept. First lifelike microsystems A most fundamental aspect of the thermal polyamino acids is their propensity to assemble, on contact with water, 90 C & E N J U N E 22, 1970 Harada found that DL-amino acids could be resolved simply by inoculation under presumed geological conditions Amino acid resolved Aspartic acid Glutamic acid Asparagine Glutamine [«]2D5 Solvent (hydrochloric acid) +24.6 6N -23.6 6N +28.5 6N -27.5 6N +28.6 3.6N -29.7 3.6N +32.8 IN -31.4 IN Optical purity 99% 95 91 88 93 97 100 96 This electron micrograph of microspheres reveals the outward diffusion of polyamino acid from the interior with a pH in- crease. The double-layer character is very evident into uniformly sized microsystems that have many of the properties of contemporary cells. The microsystems thus formed are composed of macromolecules that have internal anhydromonomeric order and enzymelike activities. Many reports on models of cells, or of de novo cells, can be found in an extensive literature. Many of these, like Oparin's coacervate droplets, lack one crucial kind of rele- vance. Inasmuch as they are produced from polymers obtained from plants or animals, they fail to answer the fundamental question of how cells arose on an earth de- void of parental cells. Oparin's well-known coacervate droplets are shown be- low. Besides failing to answer the crucial primordial question, they are neither uniform nor stable. Proteinoid microspheres, however, have these properties and answer the crucial question. They have the same shape, range of size, uniformity, and tendency to associate that are found in many populations of coccoid bacteria. When trans- ferred to solutions hypertonic or hypertonic to those they are made in, they shrink or swell respectively. They can be made to stain gram-negative or gram-positive, the latter response resulting from inclusion of sufficient lysine-rich proteinoid in a system containing primarily acidic pro- teinoid. These microspheres have ultrastructure. When their electron micrographs are viewed, they are often mistaken for simple bacteria. Expert audiences tend to identify sections of Bacillus cereus as sections of proteinoid micro- spheres and vice versa. The electron micrograph, above, shows features that result from an increase in pH from 3.5 to 5.0. This is the diffusion outward through the boundary of polyamino acid, from the interior. The boundary can then be seen to consist of a double layer. This double layer is somewhat thicker than those usually seen in cells. The remarkable aspect is that the double-layered structure was long believed to be highly characteristic of contempo- rary cells. This arrangement is, however, intrinsic to par- ticles produced by a simple, utterly geological process of two steps, the first step having once been thought to be brutally pyrolytic. The double-layered boundary is found to be selective when the pH of the suspension is raised from 3.5 to 4.5 or 5.5. The polymer that diffuses through the boundary to the exterior has composition and a sedimentation co- efficient very similar to those of the polymer composing the boundary. Experiments in which proteinoid microspheres were made in the presence of glucose, fructose, insulin, glycogen, or starch and subsequently washed in a stand- ard way with water showed selective retention of the polysaccharides. Such evidence indicates that some func- tions of a membrane were intrinsic to the self-assembled particle. This property, especially, places an emphasis on the emergence of the first cell. Although the experiments sug- Model cells such as these coacervate droplets of Oparin, left, fail to answer the crucial primordial ques- tion because they are made from polymers obtained from plants or animals—in this instance gelatin, gum, arabic, and ribonucleic acid. Our proteinoid microspheres, right, arise from monomers and exhibit a stability and uniformity lacking in coacervate droplets (microspheres X4200, average diameter, 5 fivn) JUNE 22, 1970 C&EN 91 Proteinoid microparticles possess many properties similar to contemporary cells Stability (to standing, centrifugation, sectioning) Microscopic size Variability in shape but uniform in size Numerousness Stainability Producibility as gram-positive or gram-negative Osmotic type of property in atonic solutions Ultra structure (electron microscope) Double-layered boundary Selective passage of molecules through boundary Catalytic activities Patterns of association Propagation by "budding" and fission Growth by accretion Motility Selective inclusion of polynucleotides with basic proteinoids (particles are composed of nucleoproteinoid not proteinoid) gest that the self-assembly occurred innumerable times and was procedurally simple, rugged, and almost instantaneous, such natural experiments were the momentous events of evolution—the living individual first appeared and the geological surroundings became an environment. The protocell could communicate selectively through this mem- brane with the environment. The development of an ability of such individual proto- cols to multiply, at the systems level, was also intrinsic to the self-assembly of self-ordered polymer. The pro- teinoid microspheres possess intrinsically the capacity to proliferate through a kind of "budding" and growth by accretion, in a manner partly reminiscent of the reproduc- tion of some yeasts and some bacteria. The modeled evolution of this phenomenon is being studied. The microparticles possess in large degree the rate- enhancing activities of the polymer of which they are composed. Early enzymic activities would, however, be enhanced by being associated within the same package. Since microspheres of several kinds tend to compart- mentalize, the evolutionary stage would have been set for the deployment of enzymes that characterize the efficient metabolic pathways of contemporary cells. Conclusion Syntheses of many small bio-organic molecules, of many macromolecules resembling biopolymers, and the self-as- sembly of cell-like microsystems have all been demon- strated. These last two demonstrations have been per- formed in experiments under conditions that are found in abundance on the contemporary earth, and that are also inferable for the primitive earth. For condensation polymerization of amino acids, the necessary conditions are represented by the hypohydrous activating state. One suitable set of conditions is that of terrestrial surface, or close-to-surface, temperatures above the boiling point of water. These need not be volcanic, in fact, volcanic temperatures are too high for most carbon compound chemistry, except that of rapid vapor phase con- versions. A well-known region of appropriate geological character is Yellowstone National Park—no volcanoes. More than a thousand hot spring areas have been cata- logued for the United States alone. On the contemporary earth such zones number in the thousands. Concentration to dryness and polymerization of amino acids in the dried residue can both occur at one temperature. For the step of self-assembly only water is needed. Again, as one example, sporadic rain amply meets this re- quirement. Hydrothermal gradients, abundant on the con- temporary earth, may also be invoked. Those in south- west Africa have recently been intensively studied. The results of amino acid condensation and self-assembly remarkably resemble "organized elements" from the Or- gueil and Ivuna meteorites and some of Barghoorn and his associates' microfossils shown earlier in the discussion. Although the original studies may indeed have turned up fossilized microorganisms, a newer outlook is that such units may alternatively have been self-assembled poly- amino acids or mineral substitution products thereof. Considerable similarity exists between the "organized elements" from the Orgueil meteorite, at the left, and the dividing proteinoid micro- particles, on the right. Such "organized elements" may not be the fossilized microorganisms previously thought but, rather, self-as- sembled polyamino acids Meteorite photo: Nature You can purchase reprints of this C&EN Feature, Chemical Origins of Cells 1 to 49 copies—50 cents each 50 to 299 copies—40 cents each Quantity prices on request TO: REPRINT DEPARTMENT ACS Publications 1155—16th St., N.W. Washington, D.C. 20036 FROM: Name Street City Amount enclosed $_ State ZIP Code Number wanted _ 92 C&EN J U N E 22, 1970 Proliferation of proteinoid microspheres through "budding" and growth by accretion The "buds" on these microspheres appeared spontaneously as the microspheres stood in their mother liquor When the mother liquor was warmed, the "buds" were re- leased from the microspheres These previously released and stained "buds" when trans- ferred to a saturated proteinoid solution and cooled—from 37 to 25 °C—grew by accretion This "bud" is on a second-generation microsphere. All these photomicrographs are X1500 These comparisons, the easy resolution of DL-amino acids, and the simulatory enzymic effects of proteinoids illustrate results that enlarge our judgments of the meaning of tests for life to be performed on Mars and other planets from a distance. At the level of macromolecule and protocell, research has, for the first time, provided in experimentally repeat- able detail answers, in principle, to the following questions: • How did enzymes come into existence without enzymes to make them? Appropriate mixtures of diverse amino acids and appropriate geophysical conditions were needed. • How did cells come into existence without cells to make them? On suitable simple treatment with water, self-ordered poly amino acids assembled spontaneously into protocells. • How did membranes arise? The necessary selective and other properties were intrinsic to the self-assembled microsystems. Proteinoid itself has been shown to have some properties of lipids. • How did replication begin? One mode could have been, at the systems level, proliferation through "budding" of proteinoid microspheres, and physical growth of the "buds." With further evolutionary development, this could have resembled self-reproduction more closely; the contact might have yielded a basis for transmission of in- formation to offspring. • How did "information" arise in macromolecules with- out the complex contemporary code? The first information could have arisen in polymers formed by the condensation of mixtures of diverse amino acids. The order indicated by experiments modeling such events makes unnecessary the concept of prior nucleic acid for the enzymelike and cell-like behavior described. Information in biopolymers thus originated from the monomers, and those internal limitations in the individual came from the environment. As explained in detail in other publications, this result is gratifyingly consistent with the second law of thermody- namics, and helps to invalidate the order-out-of-chaos con- cept for living systems. Most striking, we think, is that these five answers are not independent ones. They all trace to one material, the kind of variegated polyamino acid that could have, or per- haps must have, arisen spontaneously. We can ask anew the old question of whether the whole is greater than the sum of the parts, when we as chemists examine a system that has been assembled in laboratory experiments. The answer can be analyzed into two an- swers, structural and functional. Structurally the whole is the sum of the components, but at least a little more than that. Hydrogen bonds, for example, exist between the as- sembled components, whereas those bonds did not exist when the components were separate. Functionally, the JUNE 22, 1970 C&EN 93 whole is much more, crucially so, than the sum of the parts. For example, the assembled particulate unit has a membranelike structure that did not exist when the indi- vidual macromolecules of that unit were in solution. The structural unit that emerged can screen molecules that are too large to pass its boundaries. This is a function missing from the molecular components in solution. This system can also participate in proliferation in a primitive way. Striking examples of the whole being greater than the sum of the parts are found also in some contemporary enzymes, such as pyruvate decarboxylase, which in the assembled complex has activity absent from the polypeptide subunits. Contemporary living systems are undoubtedly in many ways more complex chemically than proteinoid micro- spheres and they possess the ability to synthesize their own polymers and to convey this ability to their offspring. The fact that we can to some degree define the differences be- tween contemporary cells and models of a protocell having many essential properties, however, indicates that the evolutionary analyses have led to a partial retracing of the evolutionary sequence. Those differences include energy- trapping mechanisms, nucleic acids, protein biosynthesis, and the genetic code. We now have chemical and geological reasons to believe that molecules evolved to primitive lifelike systems through rugged reactions simply, quickly, often, and in many ter- restrial locations. The answers so far available are simpler than those generally anticipated. The chemistry and phys- ics of systems are found to transcend the chemistry and physics of molecules. The research has shown that the problem can be approached through chemical discipline; it need no longer be regarded as imponderable. Dr. Sidney W. Fox is director of the Institute of Molecular Evolution and professor of biochemistry at the Uni- versity of Miami. He holds a Ph.D. in biochemistry from Caltech, which he received in 1940; he studied under Prof. Hugh M. Huffman. He has been chairman of the Division of Biological Chemistry of ACS and a national councilor in ACS, a former associate editor of Chemical Reviews, and is currently an advisory editor of Currents in Modern Biology. In spring 1969 he was a U.S.-U.S.S.R. interacademy lecturer on the origin of life, in exchange with A. I. Oparin. His earlier interests included sequence determinations in peptides, incorporation of amino acid analogs in bacteria, proteolytic enzymes, plant hormones, and evolution of protein molecules; cur- rent ones include the origin of the genetic code and related problems. Dr. Fox was born in 1912 in Los Angeles, Calif. Dr. Kaoru Harada is senior research scientist in the Institute of Molecular Evolution and an adjunct associate professor in the department of chem- istry of the University of Miami. He obtained his Ph.D. in organic chem- istry under Prof. T. Kaneko at Osaka University in 1960. He has been ac- tive in prebiological chemistry since the mid-1950's, having developed this interest from studies of the chemistry of HCN. He has published more than 60 research papers, with an emphasis on stereochemical aspects of peptide and protein chemistry. Dr. Harada was born in 1927 at Osaka, Japan. ADDITIONAL READING 1. Berkner, L. V. and Odishaw, H., eds., "Science in Space," McGraw Hill Book Co., Inc., New York, 1961. 2. Bernal, J. D., "The Origin of Life," The World Pub- lishing Co., Cleveland, 1967. 3. Calvin, M., "Chemical Evolution," Oxford University Press, New York, 1969. 4. Fox, S. W., ed., "The Origins of Prebiological Systems and of their Molecular Matrices," Academic Press, New York, 1965. 5. Fox, S. W., Self-ordered polymers and propagative cell-like systems, Naturwissenschaften, 56, 1-9, (1969), in English. 6. Kenyon, D. H. and Steinman, C , "Biochemical Pre- destination," McGraw Hill Book Co., New York, 1969. 7. Keosian, J., "The Origin of Life," 2nd ed., Reinhold Publishing Corp., 1968. 8. Oparin, A. L, Pasynskii, A. G., Braunshtein, A. E., and Pavlovskaya, T. E., eds., "The Origin of Life on the Earth," Pergamon Press, New York, 1959. 9. Pittendrigh, C. S., Vishniac, W., and Pearman, J. P. T. "Biology and the Exploration of Mars," National Academy of Sciences-National Research Council Pub- lication 1291, Washington, D.C., 1966. 10. Ponnamperuma, C. and Gabel, N. W., Current status of chemical studies on the origin of life, Space Life Sciences, 1, 64-96 (1968). Dr. Gottfried Krampitz is an asso- ciate professor at the University of Bonn, where he has worked in nutri- tion, protein analysis, and more re- cently on enzymelike activities in thermal proteinoids. He obtained his Ph.D. at the University of Bonn, West Germany, in 1954. He has twice spent a year at the Institute of Molec- ular Evolution, and participated in the writing of this article on his second visit. Dr. Krampitz was bom in 1927 at Oberrathen, Germany. Dr. George Mueller is professor in the Institute of Molecular Evolution, and visiting professor at the University of Conception, Chile, where he organ- ized the department of geology. Dr. Mueller is author of more than 50 publications, and contributor to sev- eral books, two of which are on or- ganic geochemistry. Dr. Mueller is a pioneer in the study of carbonaceous meteorites. He has published predictions of a basaltic surface of the moon and other aspects that were subse- quently verified by the Surveyor and Apollo programs. He obtained his Ph.D. in mineralogy under Prof. S. E. Holling- worth at the University of London in 1951. He was born in 1916 at Budapest, Hungary. 94 C&EN JUNE 22, 1970 CHEMlCAL ORIGINS OF CELLS History Modes of thinking Raw materials Terrestrial fractionation Amino acids Purines and pyrimidines Macromolecules Origin of enzymes Origins of metabolism Optical activity First life like microsystems Conclusion