Bioenergetics and ’s Origins

David Deamer1 and Arthur L. Weber2

1Department of Biomolecular Engineering, Baskin School of Engineering, University of California, Santa Cruz, California 95064 2SETI Institute, NASA Ames Research Center, Mountain View, California 94043 Correspondence: [email protected]

Bioenergetics is central to our understanding of living systems, yet has attracted relatively little attention in origins of life research. This article focuses on resources available to drive primitive and the synthesis of polymers that could be incorporated into molecular systems having properties associated with the living state. The compart- mented systems are referred to as protocells, each different from all the rest and representing a kind of natural experiment. The origin of life was marked when a rare few protocells hap- pened to have the ability to capture energy from the environment to initiate catalyzed heter- otrophic growth directed by heritable genetic information in the polymers. This article examines potential sources of energy available to protocells, and mechanisms by which the energy could be used to drive polymer synthesis.

revious research on life’s origins has for have been activated to assemble into polymers. Pthe most part focused on the chemistry Most biopolymers of life are synthesized when and energy sources required to produce the the equivalent of a water is removed small of life—amino acids, nucleo- to form the ester bonds of nucleic acids, glyco- bases, and amphiphiles—and to a lesser extent side bonds of polysaccharides, and peptide on condensation reactions by which the mono- bonds in . In life today, the removal of mers can be linked into biologically relevant water is performed upstream of the actual polymers. In modern living cells, polymers are bond formation. This process involves the ener- synthesized from activated monomers such as getically downhill transfer of electrons, which is the nucleoside triphosphates used by DNA coupled to either substrate-level oxidation or and RNA polymerases, and the tRNA-amino generation of a proton gradient that in turn is acyl conjugates that supply ribosomes with the energy source for the synthesis of anhydride activated amino acids. Activated monomers pyrophosphate bonds in ATP.The energy stored are essential because polymerization reactions in the pyrophosphate bond is then distributed occur in an aqueous medium and are therefore throughout the to drive most other energy- energetically uphill in the absence of activation. dependent reactions. This is a complex and A central problem therefore concerns mech- highly evolved process, so here we consider anisms by which prebiotic monomers could simpler ways in which energy could have been

Editors: David Deamer and Jack Szostak Additional Perspectives on The Origins of Life available at www.cshperspectives.org Copyright # 2010 Cold Spring Harbor Laboratory Press; all rights reserved; doi: 10.1101/cshperspect.a004929 Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929

1 D. Deamer and A.L. Weber captured from the environment and made avai- proceed toward polymer synthesis if there is a lable for primitive versions of metabolism and way to remove water molecules so that covalent polymer synthesis. Because the atmosphere of bonds can form. Their -catalyzed the primitive Earth did not contain appreciable biosynthesis requires an input of metabolic , this review of primitive bioenergetics is energy, primarily delivered as pyrophosphate limited to anaerobic sources of energy. bonds of ATP, but the linking bonds are also thermodynamically unstable, which means that BACKGROUND can also catalyze spontaneous hydro- lysis of the bonds. The result is that life incorpo- Fundamental Considerations from rates a continuous and controlled synthesis and Thermodynamics and Kinetics breakdown of its polymeric components. Simi- In general, life is characterized by the fact that lar reactions were presumably occurring in the the catalytic and genetic polymers exist in a prebiotic environment, so it is essential to estab- steady state far from equilibrium. It follows lish plausible mechanisms by which polymers that the origin of life can be understood in terms could be synthesized and accumulate despite of a process in which the flow of energy through hydrolytic back reactions. relatively simple systems of molecules produced 3. Because the concentration of reactants a more complex set of polymeric molecules that strongly affects reaction rates, it seems likely had specific physical and chemical properties. that fora prebiotic reaction to proceed at a useful The origin of life occurred when a subset of rate, there must have been concentrating proc- these molecules was captured in a compartment esses such as evaporation or adsorption. Adilute and could interact with one another to produce solution of activated monomers is unlikely to the properties we associate with the living state. form polymers because at low concentrations Energy flow, and the changes it produces, therateofmonomercondensationislessthanthe are described by the fundamental laws of ther- rate of their hydrolytic deactivation. However, modynamics and kinetics. These concepts are adsorption of activated monomers on mineral familiar to most readers, but it is less obvious surfaces could enhance their polymerization by how they can be applied to our understanding increasing their concentration and possibly ori- of the prebiotic environment and the increase enting them in a way that favors condensation. in chemical complexity driven by energy flux. 4. The reactants in a given reaction must Wewill briefly recapitulate them here in relation overcome an energy barrier called activation to the origin of life. energy that limits the rate at which the reaction 1. The amount of energy released as a reac- can occur. Elevated temperatures provide acti- tion proceedstoward equilibrium and is referred vation energy to a potentially reactive system to as free energy, which has components of en- of molecules. The global temperature when thalpy and . Both must be taken into ac- life began is estimated to be in the range of 55 count to understand how systems of molecules and 858C (Kauth and Lowe 2003) but there can become more complex. On the prebiotic would be considerable variability around this Earth, immense numbers and varieties of chem- mean, ranging from cold polar regions to exten- ical reactions were taking place because the sive high temperature vulcanism. It is reason- Earth itself was in a state of thermodynamic dis- able to consider that thermal activation energy equilibrium. Tounderstand the origin of life, it is was likely to be abundant, so that the major essential to sort out which of the many energy hurdles to be overcome in promoting polymer- sources were primary factors in driving the ization reactions would be to concentrate, orga- increasing complexity from which life emerged. nize, and chemically activate monomers. 2. All reactions in principle are reversible 5. Catalysts reduce the activation energy and can approach equilibrium from either barrier so that a reaction can proceed more direction. This means that a potentially destruc- rapidly toward equilibrium. There were no pro- tive reaction such as of polymers can tein enzymes on the prebiotic Earth, so simpler

2 Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929 Bioenergetics and Life’s Origins catalysts like mineral surfaces, metal ions, and H2, thereby producing the electrochemical en- small polymers presumably acted as catalysts ergy that reduces carbon dioxide and yields in primitive metabolic pathways. the organic molecules required by life. Similarly, 6. Chemical kinetics defines the rates at chemolithotrophs transfer high energy elec- which a given reaction occurs, and allows trons from H2 and H2S to lower energy states 2 thermodynamically unstable molecular struc- in electron acceptors such as CO2 and NO3 , tures to exist far from equilibrium. A and use the derived free energy to reduce carbon or nucleic acid in water, for instance, will ulti- dioxide to organic molecules. The organic pro- mately hydrolyze to its component amino acids. ducts of such reactions are subsequently used However,inthe absenceofacatalyst,thisisaslow by heterotrophic life as electron donors in their reaction, so that faster catalyzed reactions of energy metabolism based on either anaerobic biosynthesis can keep up with the slower degra- respiration or fermentation. dative rate of hydrolysis. The difference in In anaerobic photolithotrophy, chemoli- reaction rates is referred to as a kinetic trap. On thotrophy, and respiration, the acquired electro- the early Earth, if there was a relatively fast proc- chemical energy is used to pump protons across ess that could produce chemical bonds between membranes in such a way that an electro- monomers, kinetic traps would allow the result- chemical proton gradient is produced, equiva- ing polymers to have a transient existence even if lent to approximately 0.2 volts of electrical they were thermodynamically unstable. potential. This energy of the proton potential is coupled to ATP synthesis catalyzed by an ATP synthase embedded in the membrane. Overview of Contemporary Bioenergetics The pyrophosphate bond energy in the ATP is The bioenergetic pathways of contemporary then transferred by diffusion to the rest of the anaerobic life are well understood. As depicted cell where it drives a variety of essential meta- in Figure 1, the light energy captured by photo- bolic reactions, motor molecule functions, ion is used to activate and release elec- transport processes, and polymerization reac- o trons from inorganic donors like H2S, S ,or tions of biosynthesis.

Pathways of anaerobic energy flow

Autotrophy Heterotrophy Carbon source - CO2 Carbon source - organic substances Photolithotrophy — –1 o +2 –1 Major e donors: H2O, H2S, S , H2, Fe Fermentation — e donors: organic substances –1 –1 electron e acceptors: CO2 e acceptors: organic substances donors and –1 –1 Chemolithotrophy — e donors: H2, H25 Anaerobic — e donors: organic substances acceptors –1 –1 respiration e–1 acceptors: NO–1, SO–2, Fe+3, Mn+4 e acceptors: NO3 , CO2 2 4

Primary Photochemically generated energy source high-energy electrons or proton gradient High-energy electrons of carbon substrates

Fermentation— Anaerobic respiration— Energy Membrane electron transport Substrate-level oxidation of membrane electron transport transduction and proton gradient organic intermed. drives ATP chain and proton gradient system synthesis often via thioesters drives ATP synthesis

Energy NADPH ATP NADPH ATP transfer (electron-pair + (anhydride (electron-pair + (anhydride molecules redox energy) bond energy) redox energy) bond energy)

+ C, N, P, S + C, N, P, S

Biosynthetic Monomers + Polymers + Sugars Monomers + Polymers + Sugars products (Energized-carbon (Energized-carbon storage) storage) Figure 1. Energy sources and processing by anaerobic and . See text for discussion.

Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929 3 D. Deamer and A.L. Weber

In contrast, in fermentation, the electro- and geological electrochemical energy. The sec- chemical energy stored in organic substrates ond is a series of relatively low energy reactions activates the transfer of high energy electrons that incorporate condensation processes by from carbon groups (electron donors) to lower which monomers are assembled into random energy states provided by other carbon groups polymers. These include anhydrous heat, min- (electron acceptors). This substrate-level elec- eral-catalyzed synthesis, and sugar-driven reac- tron transfer oxidizes the electron donor to tions. The third is the energy flow in metabolic create an anhydride intermediate (usually a networks in which an energy source in the local thioester) that in turn drives the synthesis of environment is captured and then transfer- ATP (Gottschalk 1986) red to standardized energy carriers (like ATP The coupling of ATP synthesis to electron and NAD[P]H). These are catalytically coupled transfer and proton gradients is clearly a highly to a series of intracellular reactions that ulti- evolved system. The bioenergetic processes used mately activate monomers required for cata- by the first forms of cellular life must have been lyzed polymerization. very different even though the same sources Table1 showsthe kinds and relative amounts of energy were available, with the exception of energy on today’s Earth. It is a reasonable as- of electron transport to molecular oxygen. To sumption that similar energy sources were avail- understand the origin of life, we need to estab- able 4 billion years ago at the time of life’s origin. lish plausible sources of energy that would drive Light energy is by far the most abundant, and in synthetic reactions. fact photochemistry drives virtually all life to- day. Could light have been a primary energy source for the first forms of life? This is an ob- Energy Sources on the Prebiotic Earth vious possibility, yet there is a major conceptual Three kinds of energy are considered in this problem: In modern life, capturing visible light article. As shown in Figure 2, the first is the rel- requires a pigment system and a mechanism for atively high energy required to synthesize small transducing the energy content of photons into molecules that have the potential to serve as chemical energy to be used in metabolism, and monomers for polymerization reactions and there is as yet no plausible way to do this in a feedstock for a primitive metabolism. These prebiotic scenario. include photochemical energy available in However, ultraviolet light is clearly a poten- ultraviolet light, atmospheric electric discharge, tial source of high energy for small molecule

Prebiotic energy flow

Photochemical energy Geochemical energy Primary energy source UV light Electric discharge Heat Heat Mineral reducing power

Primary Neutral Atmos. Reducing Atmos. products HCHO, HNO HCN, RCHO + NH 3 H2, H2S, CH4, CO

Hydrosphere Sugars, Amino acids, Acetic acid, primary NO–1 Æ NH nucleobases methanethiol products 2 3

Secondary Amino acids, Polypeptides, Acetic acid thioester products thioesters, phosphoanhydrides N-heterocycles, model protocells Figure 2. Possible pathways for synthetic prebiotic reactions. See text for discussion.

4 Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929 Bioenergetics and Life’s Origins

Table 1. Most of the energy flux on the early Earth was about UV flux and composition of the early in the form of light energy from the sun, just as it is atmosphere, these workers calculated that form- today. aldehyde was added to the oceans at a rate of 1011 Energy sources on the early Earth (kilojoules moles/year. It has long been known that formal- 22 21 m yr ) dehyde (HCHO) in alkaline solutions readily re- Solar radiation 24,000 acts to form a variety of , as is (UV,250 nM) discussed later. Cyanide (HCN) is another com- Shock waves from 200 impacts mon product of UV and electrical energy im- Radioactivity 117 pinging on mixtures containing nitrogen and Electrical discharges 2.9 gases such as CO and CO2, and cyanide readily Volcanoes 5.4 reacts with itself to produce other biologically Chemical energy Significant for the origin of relevant molecules. For instance, a few years after life, not yet estimated. the Miller experiment was published, John Oro The value shown in the table is the energy of (1961) discovered that at high concentration, ultraviolet radiation in a wavelength range that HCN could undergo pentamerization to form is absorbed by common organic compounds adenine. such as PAH, and therefore capable of activating Elevated pressures and temperatures asso- photochemical reactions. Table adopted from ciated with geothermal conditions can also estimates given by Chyba and Sagan (1992) and Miller and Urey (1959). There is considerable promote significant chemical reactions. In par- uncertainty in these estimates, which are presented ticular, the Fischer-Tropsch (FT) reaction uses only to illustrate order-of-magnitude approxi- elevated temperatures to provide activation mations for energy flux in meter-scale localized energy for a reaction in which environments. and hydrogen combine to produce long-chain hydrocarbons. Nooner et al. (1987) were among the first to test this reaction as a potential prebi- synthesis, together with the energy released by otic source of hydrocarbon derivatives, using shock waves generated by impacting comets meteoritic iron as a catalyst, which lowers the and asteroid-sized objects (Chyba and Sagan activation energy of the reaction. 1992). The electrical discharge used in the Although it is a reasonable assumption that original Miller-Urey experiments, and elevated the synthesis of organic compounds required temperature and pressures associated with geo- for the origin of life was driven by energy avail- thermal environments (.2008C) can also do able in the prebiotic environment, another chemical work, but these sources represent a source of potential chemical energy was avail- tiny fraction of the total energy flux. Electrical able in organic compounds delivered by comets discharge is meant to simulate lightning in re- and meteoritic infall during late accretion. ducing gas mixtures, and Miller (1953) found Chyba, Sagan, and co-workers (1990, 1992) es- that small amounts of highly reactive HCN timated the total amount of organic carbon and HCHO were produced in the discharge, compounds that could be delivered in this way, which afterwards underwent Strecker reactions and within an order of magnitude, it was in the to produce several amino acids. This experiment range of the amounts estimated to be synthe- revolutionized origins of life research, because sized by Miller-Urey reactions under the most for the first time, biologically relevant molecules favorable conditions. The energy content would were synthesized in a simulated prebiotic envi- be present in the form of reduced carbon com- ronment using a plausible source of energy. pounds that could undergo chemical modifica- Formaldehyde can also be synthesized by tion if they were exposed to mineral surfaces in UVlight interacting with water vapor and carbon geologicalsettings of appropriate fugacity (Shock dioxide in the upper atmosphere, and a plausible 1990) (See also Pizzarello and Shock 2010). photochemical reaction was proposed by Pinto This possibility is largely unexplored and is likely et al. (1980). From reasonable assumptions to be a fruitful direction for future research.

Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929 5 D. Deamer and A.L. Weber

In summary, early investigations, beginning synthesized from cyclic 20,30-AMP when it was with Stanley Miller’s experiments, showed that dried and heated. Usher (1977) suggested that the energy available in electrical discharge, UV the presence of a template should promote such light, and elevated temperatures, when imping- reactions. Earlier McHale and Usher (1976) ing on simple gas mixtures, can produce reactive had shown that 6-mers of adenylic acid could compounds like HCN and HCHO, which in form a phosphodiester bond to give a 12-mer, turn react to form amino acids, purine and if a polyuridylic acid template was present. pyrimidine bases, lipid-like amphiphiles, and Although anhydrous heat would appear to carbohydrates. Assuming that inorganic phos- be a plausible source of energy to drive prebiotic phate was also available, all of the monomers condensation reactions, there are two problems required for the synthesis of major polymers with this approach. The first is obvious: In a dry of life would be available for the next stage state, potential reactants are trapped in a solid of increasing complexity in the prebiotic envi- and diffusion is minimized, so that bond ronment. formation is limited to interactions between neighboring molecules. The second problem is that multiple reactions become activated as Thermal Energy temperature increases, with the result that non- Moderate heating at temperatures below 2008C specific chemical bonds begin to form, ulti- is not a direct source of useful chemical energy. mately producing polymerized tars. But at Heat speeds up the rate at which a reaction temperatures between 608 and 808C, it may be approaches chemical equilibrium, but does possible to organize reactants in such a way not change the position of equilibrium to favor that bond formation is more specific. Cycling one side of a reaction. However, heat can move between hydrated and anhydrous conditions the equilibrium of a reaction if it changes the could then drive the synthesis of specific kinds concentration of the reactants by removing sol- of polymers. Furthermore, if the dry phase of vent, or by evaporating (volatilizing) a product a condensation reaction is cycled repeatedly of one side of a reaction, such as water from through a hydrated phase, the reaction can a condensation reaction. Under these condi- lead to the accumulation of more complex tions, net polymer synthesis becomes favorable products, as is discussed in the next section. because the monomer concentration is in- It seems reasonable to expect that small creased, and once the solvent water has evapo- amounts of organic compounds would occur rated, continued heating favors polymerization in the prebiotic environment, but they would by removing the water produced by condensa- be present as complex mixtures with inorganic tion reactions. anions and cations. The prebiotic environment Early researchers investigating the origin of was likely to have thousands of different organic life attempted, with some success, to drive con- compounds dissolved in a global ocean and densation reactions by drying potential reac- lacustrine bodies of fresh water that accumu- tants such as amino acids and nucleotides at lated on volcanic land masses rising above the elevated temperatures. The advantage of using ocean. The mixture would contain biologically anhydrous heating to drive condensation is relevant compounds such as amino acids that it is by far the simplest way to produce the and sugars, largely as racemic mixtures of their polymers required for life to begin. Although D and L enantiomers. Because the organic the amino acid polymerization studied by compounds would be present as very dilute Fox and colleagues is a prominent example solutions, it is essential to discover processes (Fox and Harada 1958), there were also early by which such compounds could be concen- attempts to drive phosphodiester bond forma- trated and organized to participate in intermo- tion under these conditions. For instance, lecular reactions such as polymer synthesis Verlander and Orgel (1974) found that oligo- and protocell assembly needed for the origin nucleotides up to six nucleotides long could be of life.

6 Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929 Bioenergetics and Life’s Origins

There are two obvious ways in which dilute drive synthetic reactions, ATP is not a primary solutions can be concentrated. The simplest is energy source, but rather is an energy transfer the input of heat energy to evaporate the solu- molecule that picks up energy from an energy tion onto mineral surfaces. For instance, evapo- source and then delivers it to energy-requiring rating a volume of water containing one micro- reactions. This constant resynthesis (cycling) mole of an organic solute such as an amino of ATP inside the cell is revealed by estimates acid evenly spread onto a mineral surface area showing that to synthesize 1 g of cell mass re- of 100 cm2 will form a film approximately 10 quires the energy of about 20–100 g of ATP molecules thick. During the last stages of eva- (Stouthamer 1977). The chemical energy con- poration, the concentration of solutes pass tent of ATP is present in the pyrophosphate through molar concentrations, thereby promot- bonds that link the second and third phosphate ing any chemical reactions that could not occur groups of ATP. These are anhydride bonds, and spontaneously in dilute solutions. Subsequent their chemical energy is released by energetically wetting cycles would release the products into downhill group transfer reactions of the phos- the environment for further processing. phate group to other molecules, an activating The second process, originally proposed by process called phosphorylation. The second Bernal (1951), is that clay surfaces strongly ad- molecule gains chemical energy and can in sorb organic compounds from dilute solutions turn undergo reactions that otherwise will not and thereby concentrate them. Furthermore, occur. Classic examples include the formation the orderly arrangement of charged groups in of aminoacyl-tRNA in protein synthesis, or the crystal structure of the clay can impose order acetyl-CoA in fatty-acid synthesis. on the adsorbed solutes and thereby promote The question is whether pyrophosphate polymerization reactions. A good example is bond energy could have been a significant the polymerization of activated nucleotides into source of chemical energy in the reactions lead- short molecules of RNA, to be discussed later in ing to the origin of life. In fact, phosphate is such this article (for review, see Ferris 1999, 2002). an integral part of all contemporary life that It is interesting to note that simple freezing phosphorylation reactions must have been is also a process that concentrates potential reac- incorporated in primitive metabolic pathways. tants in an otherwise dilute solution. As a solu- Baltscheffsky (1996) has argued that this is plau- tion freezes, the microscopic crystals that sible, in part because pyrophosphate and poly- initially form are nearly pure ice, so that solutes phosphates are readily produced simply by are concentrated into fluid eutectic phases heating inorganic phosphate under anhydrous between the crystals. Freezing has been used to conditions, a process known to occur under promote nucleobase synthesis in frozen cyanide natural conditions. Pyrophosphate-containing solutions (Miyakawa et al. 2002) and Kanavar- minerals, canaphite and wooldridgeite, have ioti et al. (2001) found that oligomers of RNA been discovered in quarries, albeit in minute qu- were synthesized from activated nucleotides antities and in the form of microscopic crystals. frozen at 2188C. (See Orgel 2004 for review.) Furthermore, Baltscheffsky found that the cou- Although extensive ice on a hot early Earth pling membranes of a photosynthetic bacterial seems unlikely, there is still uncertainty about species—Rhodospirrilum rubrum—synthesize actual temperatures available in specific sites, pyrophosphate instead of ATP. The R. rubrum so ice eutectics cannot be dismissed as a means use the pyrophosphate as an energy source, to concentrate dilute reactants in solution. much as other use ATP. Despite the ease of capturing thermal dehy- dration energy as pyrophosphate bonds, a plau- Thermal Energy Storage in Pyrophosphate sible pathway for incorporating it as an energy Bonds source in early life has not yet been established. Most chemical energy in cells today circulates in This is well worth further study, as is discussed the cytoplasm as ATP.Although cells use ATP to in the last section of this article.

Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929 7 D. Deamer and A.L. Weber

Sunlight and Photochemical Energy Geological Electrochemical Energy Light only becomes a useful energy source The electrochemical reactions used by aerobic if there is a pigment to absorb the photons. life today require oxygen produced by plants. When photons are absorbed by a pigment mol- In mitochondria and aerobic bacteria, an elec- ecule, they interact with the electronic structure tron transfer from reduced substrates to mo- of the molecule and add energy to the bonds to lecular oxygen produces 0.2 volts and enough produce an excited state. This seemingly simple protonic current to synthesize ATP. It is im- photochemical reaction is the energetic founda- probable that a complete chemiosmotic system tion of most life on the Earth, because this is was available to the first forms of life, but what happens when a chlorophyll molecule simpler electrochemical reactions are still con- absorbs light. Starting with chlorophyll in its ceivable mechanisms in which donors and ac- ground state, a photon of red light is absorbed ceptors of electrons provide an energy source. and increases the energy content of chlorophyll. Several potential donors were likely to be avail- The added energy causes it to go to an excited able in the prebiotic environment. Perhaps the state that then donates an electron through most plausible is hydrogen gas itself, as well as multiple reactions to carbon dioxide, which hydrogen sulfide (H2S) and methane. A variety ends up after further reactions as a carbohy- of microorganisms today use these gases as a drate. In addition, the transfer of the electron source of electrons, a good example being the to lower energy states is coupled to the genera- abundant bacteria present in hydrothermal tion of a proton gradient across the membrane vents. There is a consensus that little or no oxy- that is used to drive ATP synthesis. In this way, gen was present in the early atmosphere, so what the original light energy is conserved in the electron acceptors might have served instead of form of chemical energy. After it loses the elec- oxygen? A useful list of potential anaerobic tron, chlorophyll is positively charged and the donors and acceptors was presented by Gaidos electron is replaced from a water molecule in et al. (1999), who considered the interesting the “water splitting reaction,” which releases question of energy sources that might be avail- oxygen. This is the source of virtually all of the able in a putative ocean beneath Europa’s icy oxygen in the Earth’s atmosphere. surface. Terrestrial anaerobes use sulfate, nitrate, If photochemistry is essential for life today, iron (III), and manganese (IV) as electron could it also have provided energy for the origin acceptors, with CO, , and hydrogen sul- of life? Perhaps, but there is a problem: What fide as electron donors. Sulfate would be abun- pigment molecules were available? Certainly dant on the early Earth, as it is today. Nitrate not chlorophyll, which is a very complex mole- and nitrite could have been made on the prebi- cule requiring multiple enzymatic steps for its otic Earth from nitrogen gas by photochemical synthesis. Furthermore, even if primitive pig- synthesis and aqueous processes (Summers and ments were present, the capture of light energy Khare 2007) and in high pressure geochemical would not be possible unless there were mem- processes (Brandes et al. 1998). In addition, branes available that could contain the pigment Cody et al. (2000, 2004) have shown how sulfide molecules. Toprocess the absorbed light energy, minerals could have been involved in primitive a primitive cell would also require a system of metabolic pathways. electron transport molecules to transduce the light energy into chemical energy. Future re- Energy and Chemistry search might someday discover a mechanism by which this seemingly complex series of reac- Sulfur chemistry is of particular relevance to tions could arise, but until then it seems plausi- the origin of life, because the minimal amount ble that an energy source other than light was of molecular oxygen in the prebiotic environ- used by the first living cells. In other words, ment would allow sulfur to be abundant as an the first life was likely to be heterotrophic. element, as a reduced gas (hydrogen sulfide),

8 Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929 Bioenergetics and Life’s Origins and as iron sulfide minerals. Takinginto consid- solution, and Maurel and Orgel (2000) showed eration this prebiotic context, Wa¨chtersha¨user the elongation of a decapeptide of oligoglutamic (1988) proposed that life did not arise by assem- acid up to 15 mers in the presence of thiogluta- bly of pre-existing organic compounds, but mic acid. Zepik et al. (2006) extended this reac- instead as two-dimensional synthetic chemistry tion to protocells by encapsulating decaglutamic on an iron sulfide mineral surface called pyrite. acid in liposomes composed of dimyristoyl- According to this idea, when hydrogen sulfide phosphatidylcholine, and found that externally reacts with iron in solution to produce insoluble added thioglutamic acid was able to permeate iron sulfide mineral, the reaction generates the vesicle membrane at a rate sufficient to electrons that can be donated to the bound com- support elongation of the decapeptide within pounds and thereby drive a series of energeti- the vesicles. It is clear that sulfur chemistry has cally uphill chemical reactions that otherwise considerable potential for expanding our under- cannot take place in solution. Wa¨chtersha¨user standing of early bioenergetic processes. sees these reactions as the beginning metabo- lism, which occurred on a mineral surface rather Chemiosmotic Energy Conversion to than in the volume of a cell. To test this idea, Anhydride Energy Huber and Wa¨chtersha¨user (1997) heated a dis- persion of iron and nickel sulfides together with Life today uses either or sub- a source of carbon and analyzed the products. strate-level phosphorylation to convert electro- There was no evidence of a long chain of inte- chemical energy to anhydride chemical energy grated reactions, but using CO and H2S, acetic of ATP (Mitchell 1961, 1966). Was chemios- acid was produced, and from CO and CH3SH mosis an ancient process used by the first cells, the thioester CH3-CO-SCH3 was synthesized. or a later discovery? Arthur Koch (1985) first ad- In later work in which amino acids were added dressed this question and speculated that chem- to the simulation (Huber and Wa¨chtersha¨user iosmosis may have arisen in the first forms of 1998; Huber et al. 2003), the amino acids were primitive cellular life, and this conjecture is observed to be chemically activated and formed worth considering here. Three relevant postu- peptide bonds. lates of chemiosmosis are listed below: DeDuve (1991) has proposed another ver- † Coupling membranes maintain a proton sion of sulfur chemistry as a source of chemical gradient, and the gradient must be of suffi- energy related to the origin of life, which in- cient magnitude to drive ATP synthesis. volves thioesters that could have served to acti- vate energy-dependent steps in primitive † Coupling membranes pump protons by elec- metabolic pathways before the advent of ATP. tron transport using either light or electron- Thioesters are synthesized when a water mole- transfer as an energy source. cule is lost during the reaction of an organic † Coupling membranes contain an ATPase acid and a thiol: R-COOHþR’-SH – . that is also a proton pump. R-CO-SR’.Unlike the relatively low-energy con- tent of an ester bond, the thioester bond has an The lipid composition of biological membranes energy content equivalent to the pyrophosphate today contains specific lipids that are products bond of ATP. Earlier Weber (1984, 1998) had of highly evolved metabolic pathways. A stable shown that a-hydroxy acid and a-amino acid hydrophobic lipid bilayer barrier composed of thioesters could be synthesized by reacting sug- hydrocarbon chains 16–18 carbons in length is ars (or sugar precursors) with ammonia and a required to maintain ionic concentration thiol under plausible prebiotic conditions. gradients, particularly protons in the case of cou- Several investigators have incorporated sul- pling membranes. Lipid bilayers are notoriously fur chemistry in their research. Wieland (1953) leaky to protons (Nichols and Deamer 1980; originally showed that amino acid thioesters Paula et al. 1996), so the first self-assembled spontaneously form peptide bonds in aqueous membranes composed of relatively short-chain

Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929 9 D. Deamer and A.L. Weber amphiphiles would be unable to maintain signif- energy content of sugars is converted to the icant proton gradients, as recently shown by anhydride energy of ATP by substrate-level oxi- Chen and Szostak (2004). Mansy (2010) dis- dation phosphorylation, a process that does not cusses the role of membrane permeability in require the organized membrane structures of primitive cellular systems. Even if a plausible phosphorylation coupled to electron transfer. primitive barrier membrane can be discovered, As discussed later, the energy content and reac- an electron transport system and ATP synthase tivity of sugars also allows them to act as sub- would need to be incorporated in the bilayer strates for chemically spontaneous synthetic for chemiosmosis to be a source of chemical processes that yield many of the molecular energy. In our judgment, the complex require- products required for the origin of life. Such ments for a functioning chemiosmotic system sugar-driven syntheses require no external weighs against the proposition that primitive source of chemical energy (Weber 2000). life used chemiosmosis for converting electro- chemical energy to the anhydride energy of ATP. RECENT EXPERIMENTAL STUDIES

Chemical Energy of Organic Substrates: Sugar-driven Prebiotic Synthesis Carbohydrates In addition to being the sole energy and carbon The environment of the prebiotic Earth was far source of fermentative organisms today, sug- from equilibrium, so that a variety of chemical ars have chemical properties that make them reactions were occurring simultaneously. The very attractive substrates for synthetic processes problem is to gain some understanding of which needed for the origin of life. First, sugars can of these was relevant to the origin of life, and be synthesized under plausible prebiotic con- how they were incorporated. Living systems ditions from formaldehyde and glycolaldehyde today use chemical reactions to release energy by the formose reaction (Schwartz and de in small steps called metabolism, which can be Graaf 1993; See also Benner et al. 2010). Second, defined as a series of chemical reactions linked sugars are reactive and contain considerable self- in a molecular system that provides energy and transformation energy, properties that allow small molecules required for growth. Each step them to react with ammonia, yielding many is catalyzed by a specific enzyme, and the reac- types of molecules needed for the origin of life. tion rates are controlled by feedback loops in These sugar-driven syntheses require no addi- which a product is an allosteric inhibitor of tional source of chemical energy (Weber 2000). the enzyme to be regulated. If the first life was The synthetic versatility of sugars is shown by heterotrophic, what might have been their spontaneous reactions in the presence of available as a source of chemical energy? ammonia that yield catalytic amines, biomono- Of all the organic substrates, sugars are by mers (amino acids), metabolites (pyruvate, gly- far the most attractive organic energy substrate colate), energy molecules (hydroxy and amino of primitive anaerobic life, because they are acid thioesters), alternative nucleobases (2-pyra- able to provide all the energy and carbon needed zinones that resemble uracil), heterocyclic mole- for the growth and maintenance of a fermenta- cules (furans, pyrroles, imidazoles, pyridines, tive metabolism. In fact, the sugars that are the and pyrazines), polymers (polypyrroles and pol- first substrates of the glycolytic pathway can be yfurans), and cell-like organic microspherules considered to be optimal biosynthetic sub- (Weber 2001-2008, refs. therein). Sugars have strates because they contain mainly alcohol also been shown to drive the prebiotic synthesis groups that have maximum self-transformation of ammonia from nitrite. Remarkably, these pre- energy, and a single carbonyl group (aldehyde biotic synthetic processes based on sugar chemis- or ketone) that makes them reactive and able try can evolve directly into modern sugar-driven to form covalent adducts to enzyme active sites biosynthesis without violating the principle of (Weber 2004). Moreover, in fermentation, the evolutionary continuity.

10 Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929 Bioenergetics and Life’s Origins

Finally, sugar synthesis from formaldehyde stainless steel chamber, producing a mixture and glycolaldehyde, and the subsequent conver- of fatty acids and alcohols as products. At elev- sion of sugar products to carbonyl-containing ated temperatures, oxalic breaks down into a products can be catalyzed by small molecules mixture of CO, CO2, and H2, the reactive gases (ammonia and amines including amino acids required for FTT synthesis. In a later paper, and peptides). In fact, small L-dipeptides (the Simoneit et al. (2006) found that if stoichiomet- isomer found in proteins) stereoselectively cata- ric glycerol was present in the mixture along lyzed the formation of D-ribose (Pizzarello with a variety of fatty acids, the same conditions and Weber 2010). These ammonia and amine- produced good yields of monoglycerides. catalyzed reactions yielded aldotriose (glyceral- These results are significant because the dehyde), ketotriose (dihydroxyacetone), aldote- majority of membrane-forming lipids today troses (erythrose and threose), ketotetrose are glycerol esters. Furthermore, monoglycer- (erythrulose), pyruvaldehyde, acetaldehyde, ides by themselves can assemble into lipid bi- glyoxal, pyruvate, glyoxylate, and several un- layers, and when mixed with fatty acids are able identified carbonyl products. The uncatalyzed to form stable membranes (Monnard et al. 2002; control reaction yielded no pyruvate or glyoxy- Mansy and Szostak 2008). Future research in this late, and only trace amounts of pyruvaldehyde, area should be directed toward characterizing acetaldehyde, and glyoxal. With L-alanine, the such membranes in terms of long-term stability rates of triose and pyruvaldehyde synthesis and capacity for maintaining ionic concentra- were about 15-times and 1200-times faster, re- tion gradients of protons and other cations. spectively, than the uncatalyzed reaction (Weber 2001). Because amines are also products of sug- Phosphodiester Bond Synthesis Driven by ar–ammonia reactions, these studies suggested Anhydrous Cycles that the sugar–ammonia reaction could be au- tocatalytic. This possibility was tested in a later Clay mineral is a common example of an organ- study, which showed that reaction of the triose izing surface. Clays have a surprisingly large sugar (glyceraldehyde) with ammonia yielded crystalline surface area, over 100 m2/g, and a crude product mixture capable of catalyzing have a multilamellar structure that seems likely a 10-fold acceleration of the same sugar–am- to adsorb, concentrate, and organize potential monia reaction that produced the catalytic monomers on and between the lamellae. Ferris products (Weber 2007). and coworkers (1996) have made extensive studies of montmorillonite clay as an organ- izing agent, and established that activated Amphiphile Synthesis Using Geothermal monomers of mononucleotides can in fact Energy polymerize on clay surfaces into RNA-like pol- Franz Fischer and Hans Tropsch discovered in ymers up to 50 nucleotides in length containing the 1920s that a gaseous mixture of carbon both 30 –50 and 20 –50 phosphodiester bonds. monoxide and hydrogen, when passed over a However, the resulting polymers are tightly hot iron catalyst, produced excellent yields of bound to the clay surfaces, and if they are to hydrocarbons. Oro and coworkers (Nooner participate in the origin of cellular life, the et al. 1987) first showed that the Fischer-Tropsch RNA products must somehow be associated type synthesis (FTT) also worked with mete- with membranous vesicles to form a protocellu- oritic iron-nickel as a catalyst, and proposed lar system. Significantly, Hanczyc et al. (2003) that long-chain fatty acids may have been pro- found that clay particles with bound RNA duced this way on the early Earth. McCollom were readily encapsulated in fatty-acid vesicles. et al. (1999) and Rushdi et al. (2001) found It is not generally realized that lipids also that the FTT reaction also worked simply by form multilamellar structures that have the treating oxalic acid to elevated temperatures potential to organize and concentrate mono- 1508–2508C and corresponding pressures in a mers. The most common image of lipids is in

Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929 11 D. Deamer and A.L. Weber the form of microscopic vesicles bounded by a Carbonyl Sulfide as a Plausible Prebiotic lipid bilayer membrane, usually referred to as Condensing Agent liposomes. However, in the dry state, lipids are At some point, polymerization reactions must present as multilamellar matrices consisting of have evolved from simple processes driven by stacked bilayers, or less often as hexagonal an input of physical energy to a more complex phases in which the lipids form indefinitely mechanism involving primitive versions of acti- long tubes that tend to have a hexagonal pack- vation occurring in an aqueous environment. ing, rather than lamellar (Reiss-Husson and For many years, research has focused on discov- Luzzati 1967; Deamer et al. 1970). Furthermore, ering a plausible condensing agent that can per- if lipid vesicles undergo drying, the bilayers fuse form this feat, and recent discoveries suggest into the multilamellar phase, and any solutes that carbonyl sulfide is a likely candidate. Car- present are trapped and concentrated between bonyl sulfide (COS) is a reactive compound the lipid head groups of bilayers. Unlike the sol- that has been detected in volcanic gas and min- id matrix of clay surfaces, the bilayers are liquid eral ash (Rasmussen et al. 1982), along with its crystals, which means that trapped solutes have chemical relatives, carbon disulfide and carbon diffusional mobility. dioxide. Leman et al. (2004) found that if COS is Rajamani et al. (2008) investigated the possi- present in an aqueous solution of amino acids, bility that the order imposed on mononu- di- and tripeptides are synthesized with yields cleotides by multilamellar lipid matrices could up to 80%. In a second paper, Leman et al. promote polymerization. The nucleotides used (2006) reported that amino-acyl phosphate an- were ordinary 50-ribonucleotides that had not hydrides up to 30% yields were synthesized in been chemically activated. Instead, the chemical mixtures of amino acids, phosphate, and COS. potential for ester bond formation was provided These results offer a strong clue to the man- by cycling through anhydrous conditions at ner in which phosphate was initially incorpo- moderately elevated temperatures in the range rated into primitive metabolic pathways, partic- of 608–908C. Thermal cycling of the mononu- ularly those leading to peptide bond formation cleotides in mass ratios from 2:1 to 1:2 with the and the synthesis of small oligopeptides that phospholipids was shown to yield relatively may have served as catalysts and structural com- long chains of linear RNA-like polymers, ranging ponents of early life. from 20 to 100 nucleotides in length. The linear- ity was determined by nanopore analysis of the products, and the RNA-like character was estab- Novel Prebiotic Synthesis of Nucleotides lished by 32P-end labeling with T7 RNA kinase. A variety of phospholipids could promote the The nucleotide monomers of nucleic acids con- polymerization, but in the absence of lipids, sist of a purine or pyrimidine linked through only short nucleotide oligomers were detected. a nitrogen–carbon bond to a pentose sugar, These results show that there is sufficient which in turn is phosphorylated through an chemical potential available in anhydrous con- ester bond on the 50 hydroxyl group. Each of the ditions to drive phosphate ester synthesis as component molecules of nucleotides was pre- long as the nucleotide monomers are concen- sumably present in organic mixtures on the pre- trated and organized within a fluid liquid crys- biotic Earth, but must be linked into the more talline matrix that permits diffusional mobility. complex molecular structure of nucleotides be- Furthermore, at the end of the reaction when fore they can be incorporated into nucleic acid the lipid matrix is rehydrated, the products of polymers. One might imagine that phosphory- the reaction are encapsulated in lipid vesicles. lation of a ribose sugar could occur, because This appears to be a plausible pathway to pro- ester bonds are not difficult to synthesize by tocells, which could be generated at the edges simple condensation reactions, but synthesis of volcanic geothermal pools where wet–dry of the C-N bond between a sugar and a nucleo- cycles would be common. base has been much more challenging.

12 Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929 Bioenergetics and Life’s Origins

A recent paper by Powner et al. (2009) (see studies of conditions that can add order to reac- also Sutherland 2010) reported a remarkably tive monomers are likely to reveal new clues to efficient series of reactions that leads to mono- plausible polymerization processes that could nucleotide synthesis. Instead of attempting to occur in the prebiotic environment. produce C-N bonds between an existing pyri- An alternative to anhydrous heat is a con- midine such as cytosine and ribose, the reaction densing agent such as carbonyl sulfide, which uses sequential additions of cyanamide, cya- can activate peptide bond and phosphanhy- noacetylene, glycolaldehyde, and glyceralde- dride bond synthesis in aqueous solutions. hyde, all in the presence of phosphate. Under This discovery certainly deserves further atten- these conditions, spontaneous reactions first tion. So far, only short oligomers have been pro- form arabinose aminooxazoline and anhy- duced using COS and dilute monomers, but dronucleoside intermediates, which then add conditions that can concentrate and organize phosphate and condense into cytosine mono- the reactants are likely to produce much longer phosphate. Although it is unlikely that such a polymers. Related to carbonyl sulfide as a complex mixture of reactants and phosphate condensation agent is the energy available in might occur in the prebiotic environment, the thioester bonds (DeDuve 2005). Although mo- fact that the reaction can occur at all in aqueous lecularoxygen was virtually absent from the pre- solution, using only the chemical energy of the biotic atmosphere, its neighbor in the periodic reactants, opens a new direction for future in- table—sulfur—was abundant, both as an ele- vestigations that may reveal a simpler process. ment and in common gases such as H2S and COS. Therefore, it seems probable that sulfur was incorporated into a variety of organic mol- CHALLENGES AND FUTURE RESEARCH ecules, and further studies of the thioester bond DIRECTIONS as a plausible intermediate in primitive meta- bolic pathways should be fruitful. Identifying a Source of Condensation Energy A third source of energy for polymerization In this article, we briefly touched on processes processes is the chemical potential of simple car- by which certain kinds of energy could have bohydrates. Sugars, when heated near 708Cin driven chemical reactions related to the origin the presence of ammonia or amines, are known of life. There is a consensus that electrical dis- to produce polymers containing furan and charge and ultraviolet light could drive the syn- pyrrole residues, and cell-like microspherules thesis of reactive molecules like cyanide and (Weber 2005, refs. therein). Sugars also drive formaldehyde, which in turn would react to the synthesis of a-hydroxyacid thioesters and produce more complex molecules that could a-amino acid thioesters that are known to oligo- serve as potential monomers for primitive merize forming polyesters and peptides, respec- forms of life. However, a plausible energy source tively (Weber 1998, refs. therein). In addition, for polymerization remains an open question. in his description of the sugar-driven synthesis Condensation reactions driven by cycles of of uracil-like 2-pyrazinone, Weber (2008) pro- anhydrous conditions and hydration would posed that sugars also have the potential to form seem to be one obvious possibility, but seem pyrazinone monomers with a-hydroxycarbonyl limited by the lack of specificity of the chemical side-chain groups (CH2OH-CO-CH2-pyrazi- bonds that are formed. On the other hand, there none) that could spontaneously polymerize to may be conditions yet to be discovered that give oligomers joined by enol ether linkages or organize monomers in such a way that the for- dehydrated aldol linkages. In contrast to other mation of specific chemical bonds is promoted. prebiotic syntheses, the previously cited sugar- The organizing effect of clay mineral surfaces is driven processes are “one-pot” reactions that one well-known example, and incorporation of yieldreactivemonomers capableofspontaneous monomers into orderly lipid matrices also pro- oligomerization without being coupled to an motes specific polymerization reactions. Future additional source of chemical energy, like ATP.

Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929 13 D. Deamer and A.L. Weber

Phosphate Reactions An alternative scenario is that pigments existed in the environment that partitioned There are good reasons why phosphate is central into lipid bilayers of membranes in such a way to energy metabolism today. These were de- that light energy could be transduced into scribed by Westheimer (1987) in an excellent chemical energy. Were organic molecules pre- review that should be required reading for stu- sent in the prebiotic environment that might dents interested in the origins of life. An impor- serve as membrane-bound pigments? Polycyclic tant, still unanswered question concerns how aromatic hydrocarbons (PAH) are abundant phosphate might have first become involved in forms of organic carbon, and most PAH species life processes. Phosphate today is mostly present absorb light in the near UV and blue region of in the form of a mineral called apatite, the same the spectrum. After accepting photons, the combination of calcium and phosphate that excited states can act as reducing agents and composes tooth enamel and bones. Apatite has release protons, thereby generating chemios- a very low solubility, so what was the original motic pH gradients (for review, see Deamer source of a soluble form of phosphate? There et al. 1994). Another interesting photochemical are no convincing explanations yet, but it is pos- reaction of PAH derivatives is photocarboxy- sible that life began in a low pH environment lation. For instance, when exposed to near UV similar to that of volcanic hydrothermal springs. light, phenanthrene reacts directly with carbon CalciumphosphatereadilydissolvesatacidicpH dioxide to produce phenanthrene carboxylic rangesto release phosphateanions. The presence acid, probably the simplest possible example offreephosphateinsolutionmayhavepermitted of carbon dioxide fixation (Tazuke et al. 1986; incorporation into organic compounds as phos- See Deamer 1997 for review). These properties phate esters, followed bya second set of chemical of PAH compounds have yet to be explored reactions that initiated primitive metabolic thoroughly and surely represent a fruitful area pathways involving phosphate. For instance, for future research. Prabahar and Ferris (1997) reported that ade- nine itself is able to activate phosphate under certain conditions. Identifying a plausible mec- hanism for prebiotic phosphorylation repre- REFERENCES sents an important problem for future research. Baltscheffsky H. 1996. Origin and evolution of biological energy conversion. New York: Wiley VCH. Benner SA, Kim H-J, Kim M-J, Ricardo A. 2010. Planetary organic chemistry and the origins of Biomolecules. Pigments and Cold Spring Harb Perspect Biol 2: a003467. Sunlight drives virtually the entire biosphere Bernal JD. 1951. The physical basis of life. London: Routledge and Kegan Paul. today, but when and how did life first begin to Brandes JA, Boctor NZ, Cody GD, Cooper BA, Hazen RM, capture light as an energy source? Although Yoder HS. 1998. Abiotic nitrogen reduction on the early today’s photosynthetic process seems much Earth. Nature 395: 365–336. too complex to have been a source of energy Chen IA, Szostak JW. 2004. A kinetic study of the growth of for early life, there must have been some sort of fatty acid vesicles. Biophys J 87: 988–998. Chyba CF, Sagan C. 1992. Endogenous production, exoge- pigment available that could begin capturing nous delivery and impact-shock synthesis of organic light energy in a useful way and initiate the evo- molecules: An inventory for the origin of life. Nature lutionary path toward modern photosynthesis. 355: 125–113. Pteridines are examples of relatively simple Chyba CF,Thomas PJ, Brookshaw, Sagan C. 1990. Cometary delivery of organic molecules to the early Earth. Science pigment molecules that can be generated from 249: 366–373. amino acids exposed to anhydrous heat. This Cody GD. 2004. Transition metal sulfides and the origin of reaction and potential roles of pteridines in the metabolism. Ann Rev Earth Planetary Sci 32: 569–599. origin of life have been extensively explored by Cody GD, Boctor NZ, Filley TR, Hazen RM, Scott JH, Sharma A, Yoder HS. 2000. Primordial carbonylated Kritsky and coworkers and deserve further study iron-sulfur compounds and synthesis of pyruvate. (for review, see Kritsky and Telegina 2004). Science 289: 1337–1340.

14 Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929 Bioenergetics and Life’s Origins

Deamer DW. 1991. Polycyclic aromatic hydrocarbons: extreme environments and astrobiology. J. Seckbach, Ed. Primitive pigment systems in the prebiotic environment. Springer Netherlands. Adv Space Res 12: 183–189. Leman L, Orgel L, Ghadiri MR. 2004. Carbonyl sulfi- Deamer DW. 1997. The first living systems: A bioenergetic de-mediated prebiotic formation of peptides. Science perspective. Microbiol Mol Biol Rev. 61: 239–262. 306: 283–286. Deamer DW, Harang-Mahon E, Bosco G. 1994. Self- Leman LJ, Orgel LE, Ghadiri MR. 2006. Amino acid de- assembly and function of primitive membrane pendent formation of phoshate anhydrides in water structures. In: Early life on Earth. Nobel Symposium mediated by carbonyl sulfide. J Am Chem Soc 128: 20–21. No. 84. Bengston S. ed. Lohrmann R, Bridson PK, Orgel LE. 1980. Efficient metal- Deamer DW, Leonard R, Tardieu A, Branton D. 1970. ion catalyzed template-directed oligonucleotide synthe- Lamellar and hexagonal lipid phases visualized by freeze- sis. Science 208: 1464–1465. etching. Biochim Biophys Acta 219: 47–60. Mansy SS. 2010. Membrane transport in primitive cells. DeDuve C. 1991. Blueprint for a cell: The nature and origin of Cold Spring Harb Perspect Biol 2: a002188. life. New York: Neil Patterson Publishers. Mansy SS, Szostak JW. 2008. Thermostability of model DeDuve C. 2005. Singularities: Landmarks on the pathway of protocell membranes. Proc Natl Acad Sci 105: life. Cambridge University Press. 13351–13355. Ferris JP.1999. Prebiotic synthesis on minerals: Bridging the Maurel M-C, Orgel LE. 2000. Oligomerization of prebiotic and RNA worlds. Biol Bull 196: 311–314. a-thioglutamic acid. Orig Life Evol Biospheres 30: Ferris J. 2002. Montmorillonite catalysis of 30–50 mer 423–430. oligonucleotides: Laboratory demonstration of potential McCollom TM, Ritter G, Simoneit BRT.1999. Lipid synthe- steps in the origin of the RNA world. Orig Life Evol sis under hydrothermal conditions by Fischer-Tropsch- Biospheres 32: 311–332. type reactions. Orig Life Evol Biospheres 29: 153–166. Ferris JP, Hill AR, Liu R, Orgel LE. 1996. Synthesis of long Miller SL. 1953. Production of amino acids under possible prebiotic oligomers on mineral surfaces. Nature 381: 59. primitive Earth conditions. Science 117: 528–529. Fox SW, Harada K. 1958. Thermal copolymerization of amino acids to a product resembling protein. Science Miller AL, Urey HC. 1959. Organic compound synthesis on 128: 1214. the primitive Earth. Science 130: 245–51. Gottschalk G. 1986. Bacterial metabolism. New York: Mitchell P. 1961. Coupling of phosphorylation to electron Springer-Verlag, New York, pp. 1–11, 210–317. and hydrogen transfer by a chemi-osmotic type of mech- anism. Nature 191: 144–148. Gaidos EJ, Nealson KH, Kirschvink JL. 1999. Life in ice-covered oceans. Science 284: 1631–1633. Mitchell P. 1966. Chemiosmotic coupling in oxidative and photosynthetic phosphorylation. Biol Rev 41: 445–501. Hanczyc MM, Fujikawa SM, Szostak JW. 2003. Experi- mental models of primitive cellular compartments: Miyakawa S, Cleaves HJ, Miller SL. 2002. The cold origin Encapsulation, growth, and division. Science 302: of life: Implications based on pyrimidines and purines 618–622. pruduced from frozen ammonium cyanide solutions. Hargreaves WR, Deamer DW. 1978. Liposomes from ionic, Orig Life Evol Biosph 32: 209–218. single-chain amphiphiles. 17: 3759–3768. Monnard PA, Apel CL, Kanavarioti A, Deamer DW. 2002. Huber C, Wa¨chtersha¨user G. 1997. Activated acetic acid Influence of ionic inorganic solutes on self-assembly by carbon fixation on (Fe,Ni)S under primordial con- and polymerization processes related to early forms of ditions. Science 276: 245–247. life: Implications for a prebiotic aqueous medium. Astrobiology 2: 139–152. Huber C, Wa¨chtersha¨user G. 1998. Peptides by activation of amino acids with CO on (Ni,Fe)S surfaces: Implications Nichols JW, Deamer DW. 1980. Net proton-hydroxide for the origin of life. Science 281: 670–672. permeability of large unilamellar liposomes measured Huber C, Eisenreich W, Hecht S, Wa¨chtersha¨user G. 2003. by an acid-base titration technique. Proc Natl Acad Sci A possible primordial peptide cycle. Science 301: 77: 2038–2042. 938–940. Nooner DW, Oro J. 1987. Synthesis of fatty acids by a closed Kanavarioti A, Monnard P-A, Deamer DW. 2001. Eutectic system Fischer-Tropsch process. Adv Chem 178: phases in ice facilitate nonenzymatic nucleic acid synthe- 159–171. sis. Astrobiology 1: 271–281. Orgel L. 2004. Prebiotic adenine revisited: Eutectics and Knauth LP,Lowe DR. 2003. High Archean climatic temper- photochemistry. Orig Life Evol Biospheres 34: 361–369. ature inferred from oxygen isotope geochemistry of Oro J. 1961. Mechanism of synthesis of adenine from hydro- cherts in the 3.5 Ga Swaziland Supergroup, South Africa. gen cyanide under possible primitive Earth conditions. GSA Bulletin 115: 566–580. Nature 191: 1193–1194. Koch A. 1985. Primeval cells: Possible energy-generating Paula S, Volkov AG, VanHoek AN, Haines TH, Deamer DW. and cell-division mechanisms. J Mol Evol 21: 270–77. 1996. Permeation of protons, potassium ions, and small Koch AL, Schmidt TM. 1991. The first cellular bioenergetic polar molecules through phospholipid bilayers as a func- process: Primitive generation of a proton motive force. tion of membrane thickness. Biophys J 70: 339–348. J Mol Evo. 33: 297–304. Pinto JP, Gladstone GR, Yung YL. 1980. Photochemical Kritsky MS, Telegina TA. 2004. Role of nucleotide-like coe- production of formaldehyde in Earth’s primitive atmos- nzymes in primitive evolution. In Cellular origin, life in phere. Science 210: 183–185.

Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929 15 D. Deamer and A.L. Weber

Pizzarello S, Shock E. 2010. The organic composition of Szostak JW, Bartel DP, Luisi PL. 2001. Synthesizing life. carbonaceous meteorites: The evolutionary story ahead Nature 409: 387–390. of biochemistry. Cold Spring Harb Perspect Biol 2: Tazuke S, Kazama S, Kitamura N. 1986. Reductive photo- a002105. carboxylation of aromatic hydrocarbons. J Org Chem Pizzarello S, Weber AL. 2010. Stereoselective syntheses of 51: 4548–4553. pentose sugars under realistic prebiotic conditions. Usher DA. 1977. Early chemical evolution of nucleic acids: a Orig Life Evol Biosph (in press). theoretical model. Science 196: 311–313. Powner MW, Gerland B, Sutherland JD. 2009. Synthesis of Usher DA, McHale AH. 1976. Nonenzymic joining of activated pyrimidine ribonucleotides in prebiotically oligoadenylates on a polyuridylic acid template. Science plausible conditions. Nature 459: 239–242. 192: 53–54. Prabahar KJ, Ferris JP. 1997. Adenine derivatives as Verlander MS, Orgel LE. 1974. Analysis of high molecular phosphate-activating groups for the regioselective weight material from the polymerization of adenosine formation of 3’,5’-linked oligoadenylates on montmoril- 0 0 lonite: possible phosphate-activating groups for the cyclic 2 ,3-phosphate. J Mol Evol 3: 115–120. prebiotic synthesis of RNA. J Am Chem Soc 119: Wachtershauser G. 1988. Before enzyme and template: 4330–4337. Theory of surface metabolism. Microbiol Rev 52: Rajamani S, Vlassov A, Benner S, Coombs A, Olasagasti F, 452–484. Deamer D. 2008. Lipid-assisted synthesis of RNA-like Walde P, Wick R, Fresta M, Mangone A, Luisi PL. 1994. polymers from mononucleotides. Orig Life Evol Biosphere Autopoietic self-reproduction of fatty acid vesicles. 38: 57–74. J Am Chem Soc 116: 11649–11654. Rasmussen RA, Khalil MAK, Dalluge RW,Penkett SA, Jones Weber AL. 1998. Prebiotic amino acid thioester synthesis: B. 1982. Carbonyl sulfide and carbon disulfide from the Thiol-dependent amino acid synthesis from formose eruptions of Mount St. Helens. Science 215: 665–667. substrates (formaldehyde and glycolaldehyde) and Reiss-Husson F, Luzzati V. 1967. Phase transitions in lipids ammonia. Orig Life Evol Biospheres 28: 259–270. in relation to the structure of membranes. Adv Biol Weber AL. 1984. Nonenzymatic formation of “energy-rich” Med Phys 11: 87–107. lactoyl and glyceroyl thioesters from glyceraldehyde and a Rushdi AI, Simoneit BRT.2001. Lipid formation by aqueous thiol. J Mol Evol 20: 157–166. Fischer-Tropsch-type synthesis over a temperature range Weber AL. 2000. Sugars as the optimal biosynthetic carbon of 100 to 4008C. Orig Life Evol Biosphere 31: 103–118. substrate of aqueous life throughout the universe. Orig Schwartz AW,de Graaf RM. 1993. The prebiotic synthesis of Life Evol Biospheres 30: 33–43. carbohydrates: A reassessment. J Mol Evolution 35: Weber AL. 2001. The Sugar Model: Catalysis by amines and 101–106. amino acid products. Orig Life Evol Biospheres 31: 71–86. Shock E. 1990. Geochemical constraints on the origin of Weber AL. 2004. Kinetics of organic transformations under organic compounds in hydrothermal systems. Orig Life mild aqueous conditions: Implications for the origin of Evol Biospheres 20: 331–367. life and its metabolism. Orig Life Evol Biosphere 34: Simoneit BRT, Rushdi AI, Deamer DW. 2006. Abiotic 473–495. formation of acylglycerols under simulated hydrothermal Weber AL. 2005. Growth of organic microspherules in conditions and self-assembly properties of such lipid sugar-ammonia reactions. Orig Life Evol Biospheres 35: products. Adv Space Res 11: 1649–1656. 523–536. Sleep NH, Zahnle K, Kasting JF, Morowitz HJ. 1989. Anni- hilation of ecosystems by large asteroid impacts on the Weber AL. 2007. The Sugar Model: Autocatalytic activity early Earth. Nature 342: 139–142. of the triose-ammonia. Orig Life Evol Biospheres 37: 105–111. Stouthamer AH. 1977. Energetic aspects of the growth of microorganisms. In Microbial Energetics. BA Haddock Weber AL. 2008. Sugar-driven prebiotic synthesis of and WAHamilton, Eds. Cambridge University Press Lon- 3,5(6)-dimethylpyrazin-2-one: A possible nucleobase of don. pp. 285–315. a primitive replication process. Orig Life Evol Biospheres 38: 279–292. Stribling R, Miller SL. 1987. Energy yields for hydrogen cyanide and formaldehyde synthesis: The HCN and ami- Westheimer FH. 1987. Why nature chose phosphates. no acid concentration in the primitive ocean. Orig Life Science 235: 1173–1178. Evol Biospheres 17: 261–273. Wieland T,Bokelmann E, Bauer L, Lang HU. 1953. Polypep- Summers DR, Khare B. 2007. Nitrogen fixation on early tide syntheses. 8. Formation of sulfur containing peptides Mars and other terrestrial planets: Experimental demon- by the intramolecular Liebigs. Ann Chem 582: 129–149. stration of abiotic fixation reactions to nitrite and nitrate. Zepik HH, Rajamani S, Maurel MC, Deamer D. 2007. Astrobiology 7: 333–341. Oligomerization of thioglutamic acid: encapsulated Sutherland JD. 2010. Ribonucleotides. Cold Spring Harb reactions and lipid catalysis. Orig Life Evol Biospheres Perspect Biol 2: a005439. 37: 495–505.

16 Cite this article as Cold Spring Harb Perspect Biol 2010;2:a004929