cience shows us that the universe evolved by self-organization of matter towards more and more complex structures. Atoms, stars and galaxies self-assembled out of the fundamental particles produced by the Big Bang. In first-generation stars, heavier elements like carbon, nitrogen and oxygen were formed. Aging first-generation stars then expelled them out into space – we, who consist of these elements, are thus literally born from stardust. The heaviest elements were born in the explosions of supernovae. The forces of gravity subsequently allowed for the formation of newer stars and of planets. Finally, in the process of biological evolution from bacteria-like tiny cells (the last universal common ancestor) to all life on earth, including us humans, complex life forms arose from simpler ones.
Upon considering this self-organization of material structures in the realm of philosophy, one may conclude that it happens either because the underlying laws of nature simply are the way they are, or because they were designed by God for this purpose. Since we know that the laws of nature are so self-sufficient that, based on them, the complexity of the entire physical universe evolved from fundamental particles, and further, complex life forms from simpler ones during biological evolution, we can reasonably extrapolate that they would also allow life itself to originate spontaneously, by evolution of complex structures – regardless if we believe these laws are designed or undesigned. Therefore, we should expect an origin of life by natural causes from both theistic and atheistic philosophical perspectives.
After critical study of the scientific literature I conclude that advances in our knowledge, with particularly exciting findings in the last decade, have now made the spontaneous origin of life a plausible assumption. I will first discuss the scenario where the scientific results obtained so far satisfy me most, the “gene-first” model, and then move on to other hypotheses. Even though in my view the fundamental problem of reaction specificity currently puts the “metabolism-first” scenario at a disadvantage (see below), it might very well become the favorite model in the field if results from future experiments would convincingly address the issue. Alternatively, it might turn out that the origin of life will best be explained as a cooperative phenomenon between emergent metabolism and genetics. This alternative is outlined in Gen-e-sis: The Scientific Quest for Life’s Origins (2005) by Robert Hazen, p. 243 (referred to here as Gen-e-sis). The book, which I strongly recommend, gives a good overview of data and models in origin-of-life research so far, and also provides fascinating looks behind the scenes, all in a compelling narrative style (the author himself has contributed substantially to this research).
Compared to science of evolution, the science of abiogenesis (origin of life) is still seriously underdeveloped in its explanatory power, despite the recent progress. As science writer Richard Robinson notes in his article: “Give biologists a cell, and they’ll give you the world. But beyond assuming the first cell must have somehow come into existence, how do biologists explain its emergence from the prebiotic world four billion years ago?”
Indeed, it is one thing that we know all the chemical building materials of life, and that the functioning of life can be fully explained by their collaboration in an extremely complex system. Yet it is another thing entirely how, at the origin of life, they could have formed an initial organization by themselves step by step (via whatever intermediary processes and building blocks). At first glance, evolution from bacteria-like organisms (the last universal common ancestor) to humans may seem child’s play in comparison: it started from an already tremendously complex, entirely self-sufficient, biochemical machinery and bit by bit simply made it even more complex.
One cannot lose out of sight that the most elementary cells we currently know, which are not permanently dependent on host-metabolism, the bacterium Mycoplasma genitalium, have 482 protein-coding genes (most bacteria, such as E. coli, encode for more than 2000 different proteins), from which, according to the probably best experimental study to date (Glass et al. 2006), the essential ones are 387. The likely most accurate hypothetical study (Gil et al. 2004), puts the minimal number of genes at 216. All the proteins produced from these genes are involved in a maze of pathways of metabolism, replication, as well as building and maintenance of structure, which is of bewildering complexity.
In fact, how else than through such a minimum amount of complexity, could even a primitive cell have met the just mentioned basic demands? How could such a vastly complex network of more than 200 proteins have arisen by itself? One might ask: would it not have to have arisen at once?
Essential to the spontaneous origin of life was the availability of organic molecules as building blocks. The famous “prebiotic soup” experiment by Stanley Miller (Miller 1953, Miller-Urey experiment) had shown that amino acids, the building blocks of proteins, arose among other small organic molecules spontaneously in the lab by sparking a mixture of methane, hydrogen, ammonia and water. These conditions were assumed to simulate those on the primitive earth. Already in 1922 Oparin had proposed that the early Earth had such a reducing atmosphere (in his classic “The Origin of Life” from 1936 he expanded on these ideas). Observations of Jupiter and Saturn had shown that they contained ammonia and methane, and large amounts of hydrogen were inferred to be present there as well (it is now known that hydrogen is the main atmospheric component of these planets). These reducing atmospheres of the giant planets were regarded as captured remnants of the solar nebula and the atmosphere of the early Earth was assumed by analogy to have been similar.
Only in a reducing atmosphere like this, synthesis of organic molecules – also sugars and organic bases, building blocks of nucleotides – would have been possible in large amounts (Chyba, Sagan 1992).
However, new calculations indicate that hydrogen escaped from the early atmosphere at a much slower rate than previously thought, yielding an atmosphere where hydrogen was a major component (about 30%) and which was therefore highly reducing (Tian F. et al. 2005; see also press release). The authors measured the production of organic molecules through UV photolysis under those conditions, and conclude that at 1010 kg/year it “would have been orders of magnitude greater than the rate of either the synthesis of organic compounds in hydrothermal systems or the exogenous delivery of organic compounds to early Earth”.
Another new study supports an early reducing atmosphere as well. Chondrites are primitive material from the solar nebula and are generally believed to be the building blocks of the Earth and other rocky planets, asteroids and satellites. During and after planet formation, gases escape from the chondritic material due to high temperature and pressure. Systematic, detailed calculations on what these gases must have been show that they are mainly the highly reducing hydrogen, methane and ammonia – the same gases as in the Miller-Urey-type experiments (Schaefer, Fegley 2006; see also press release). The composition of the gases varied with temperature only to a moderate extent, and was found to be largely independent of the actual pressure under which outgassing may have occurred, which appears to support robustness of the conclusions.
The authors mention that it had been found that a reducing atmosphere of methane and ammonia is extremely vulnerable to destruction by UV sunlight (Kuhn, Atreya 1979, Kasting et al. 1983). They also point out, however, that recent developments suggest that a reducing atmosphere is more stable than previously believed:
- It was found that hydrogen escape from the Earth’s atmosphere was less efficient than previously thought (referring to the study above).
- Observations of the atmosphere of Titan, Saturn’s moon, which is composed primarily of methane and nitrogen, show that photochemically produced hydrocarbon aerosols form a haze layer in the upper atmosphere that protects the lower atmosphere from photochemical destruction. Such a haze layer could also have been produced on the early Earth from outgassed methane and ammonia (Zahnle 1986, Sagan and Chyba 1997, Pavlov et al. 2000).
Of course, if life arose in deep-sea hydrothermal vents (see below), the composition of Earth’s early atmosphere would become largely irrelevant. To a certain extent, this also holds true for organic building blocks delivered to the earth by interplanetary dust particles and on carbonaceous meteorites.
Although some estimates assume a relatively concentrated prebiotic soup of organic molecules in the earth’s ocean or other waters (e.g. De Duve, Miller 1991 and references therein), others have argued that the prebiotic soup would have been too dilute. However, locally it might have been concentrated by such simple processes as, for example, evaporation in puddles or shallow lakes, possibly with long-term wet/dry cycles. It should be kept in mind for evaluating all chemical scenarios that, due to its nature, the origin of life must have been a very local event; this is also important for the issue of the origin of homochirality of amino acids and sugars, see below.
It is now widely agreed that at the origin of life there was not the current DNA/(RNA)/protein system for gene information on one hand and catalysis, regulation, and structural function on the other. It would beg the question, what came first, protein or DNA? Protein catalysis without gene information, which allows it to be maintained and propagated, is not sufficient in the long term, and DNA gene information without catalysis, necessary for the function of life, would be useless as well.
Instead, it is assumed that RNA acted as a precursor of both protein and DNA, in the sense that it can serve both as catalyst (like protein enzymes) and as carrier of genetic information (like DNA, RNA is a polynucleotide). Even in the modern cell ribozymes (catalytic RNAs) still play a vital, albeit limited, role. In the ribosome, the synthesis of the peptide chains of proteins from RNA code is accomplished by ribozymes. They also catalyze splicing of RNA.
Could this so-called RNA World have offered a good basis for the origin of life? Although this is a commonly held view, it appears to have been made obsolete by ongoing research.
Leslie Orgel is one of the leading figures in origin-of-life research since many years, and he is one of several researchers who independently from each other proposed in the 1960s the RNA world as a precursor of the current DNA/protein world. Gerald Joyce is also a top scientist in the field. The authors argue in a joint article published in The RNA World, 2nd edition (2000), p. 68, on solid chemical grounds that, because of the complex and stereospecific chemistry required, “the de novo appearance of oligonucleotides on the primitive earth would have been a near miracle” (1). After describing a chemically more plausible scenario, PNA (peptide nucleic acid) as a precursor to RNA, the authors point to the enormous difficulties of a transition from PNA to RNA, and to the fact that it yet has to be established that PNA could result in a replicating system. They go on to say that although the presumed RNA World should be considered a milestone and a plateau in the early history of the earth, the concept “does not explain how life originated” (p.74). They conclude (p. 74): “One can sketch out a logical order of events, beginning with prebiotic chemistry and ending with DNA/protein based life. However, it must be said that the details of these events remain obscure and are not likely to be known in the near future.”
I refer the reader to Orgel’s authoritative 2004 review article “Prebiotic Chemistry and the Origin of the RNA World” (Orgel 2004; free full text, you may need to register on the link, which is free). It is a must-read for scientists interested in origin-of-life research, and for those who are otherwise interested in this field and have some basic understanding of chemistry. Orgel reviews a lot of impressive chemistry yielding building blocks for life (with a considerable amount of the experiments over the last few decades having come from his own lab, published in leading journals), but he concludes that even just prebiotic synthesis of nucleotides is “unlikely” (nucleotides are the monomer precursors of oligonucleotides and polynucleotides).
Indeed, even though it is commonly accepted that the RNA World presumably had played an important role in the development of life, it now seems clear that it must have been preceded by other steps, if life were to have arisen spontaneously.
Difficulties remain with a precursor genetic system. In his 2004 review, where he describes the already mentioned PNA system and also a TNA (threose nucleic acid) system – the latter being more promising, and simpler to synthesize than RNA (2) –, Orgel states that “the idea that RNA was ‘invented’ by a simpler genetic system is now a popular one, but no convincing precursor system has been described”. However, it is not implausible that such a system existed.
A highly interesting and chemically appealing, yet still untested idea is the PAH World (PAH = polycyclic aromatic hydrocarbons) which is described extensively in Gen-e-sis: 222 ff and also at Wikipedia and at pahworld.com.
Other difficulties are that a ribozyme (catalytic RNA) that can copy itself completely has not yet been found, but this rather seems just a matter of time – so far, a 200 base ribozyme can copy 14 bases of its sequence. Also, “the formidable problem of separating the double-stranded product of the copying reaction so as to permit a second round of copying would remain to be solved” (Orgel 2004 review). Solutions for these problems of copying a long ribozyme sequence and of double strand separation have been proposed in the article “Synthesizing Life” (Szostak et al. 2001); experiments will have to show if they are feasible.
Yet let us assume that all these issues will be adequately addressed and that a precursor genetic/catalytic system for RNA existed, which could have synthesized nucleotides in an activated form that polymerized upon catalysis by mineral surfaces (cf. Orgel 2004 review), all of which is not inconceivable. In this manner a self-replicating RNA molecule could have arisen out of a large pool of random RNAs.
Since fatty acids could have been available in the environment, a primitive fatty acid membrane could have surrounded the first self-replicating RNA molecules (due to their molecular properties, fatty acids can form vesicles spontaneously) (3); this would not have allowed passage of the RNA polymers so that they would have stayed together, but would have let the much smaller nucleotides through, fed in from a precursor genetic/catalytic system. Such a membrane would have had different characteristics of semi-permeability than modern lipid membranes, where a lot of molecule transfer is regulated through protein channels.
The group of Szostak has performed extensive and quite plausible studies that these fatty acid vesicles as containers for RNA would have allowed growth and replication merely by physico-chemical mechanisms, until a more sophisticated membrane machinery, steered by the cell itself and more resembling what is found in current organisms, would have taken their place (3).
Inside the vesicles, the self-replicating RNA could have started copying itself. During copying, various things would have been possible. High-fidelity copies would have yielded the same self-replicating molecule. Copies with errors would mostly have resulted in RNA that was non-functional, but in a minority of cases, they could have yielded RNA that copied itself faster. At this stage, a “batttle of the bubbles” (see press release, Chen et al. 2004) could have started – it has been shown that RNA/vesicle systems that contain more genetic material (which would have resulted from faster RNA replication) develop more internal tension than neighboring vesicles that do not contain as much RNA, and draw membrane material from them. Importantly, this would have allowed for natural selection of vesicles by competition even in the absence of the ability to synthesize their own membrane components and thus to directly control their own growth.
A small portion of other copying errors (again, most substantial errors would probably have resulted in non-functional molecules, but those would have been filtered out by natural selection) could have led to RNA molecules with yet other, entirely different catalytic properties than the copying function. A new property could have allowed the RNA/vesicle system to even better compete for resources: just like in the case of the RNA molecules featuring the better copying function, the RNA would have evolved (4). The copy function of the parent molecules would probably have acted on these daughter molecules as well (like an RNA polymerase enzyme that copies any RNA). RNA/vesicle systems that had the altered RNA molecules with the new beneficial function, in addition to retaining the RNA with the copy function, would have been favored by natural selection. Finally, through reiteration of such processes, a series of new catalytic properties could, for example, have allowed the RNA pool within the vesicles to start making its own nucleotides. Would it then have been self-sufficient? To a certain degree, yes. Would this have been the first primitive cell? Why not?
It is just that in this scenario, the initial metabolism would have been much simpler than today’s metabolism: Among others, energy metabolism could have been replaced by passage of activated building blocks for molecules from the outside environment into the vesicle (in a sense providing a preliminary substitute for modern-day ATP production, a possibility in view of the simple metabolism), and lipid metabolism, building of membrane structure and its regulation during replication would have been replaced by simple vesicles plainly obeying physico-chemical forces.
In other words: the cell would have depended more on the outside world, but for what it was doing, it was to a certain degree self-sufficient (today’s organisms also need sustenance from the outside world of course, in the form of nutrients). On the other hand, the dependence of the cell on the outside world would also have been possible in a more immediate manner. A modern cell cannot, for example, use fatty acids from the outside in a way that they are directly incorporated as membrane elements.
Earlier we had asked: How could a complex network of more than 200 essential proteins, as it is found in today’s most elementary cells, have arisen on its own?
The key to answering this question appears the combination of the above two attributes in the primitive cell: more direct dependence on the outside world than the modern cell, but also a greater ability to show such dependence – by accepting molecular building blocks as such without having to convert nutrients into them. Starting with these characteristics and gradually moving on from there, evolution indeed could have “eased” the cellular system into more complexity.
Perhaps lipid synthesis, in a precursor form of modern synthesis, could have made the system more independent, the RNA system could have, step by step, “invented” protein synthesis – as mentioned, the modern ribosomes still contain ribozymes (catalytic RNA) that catalyze the formation of peptide bonds which eventually result in proteins – and so on. Finally, complex metabolism could have been achieved and the transition to the modern DNA/(RNA)/protein world. The dualism DNA/protein of course is a source of complexity in itself, one that is lacking in an RNA-only organism.
What about the difficult issue of a genome which holds all genes together? It might have been that in the first primitive cells RNAs were ligated “by accident” step by step, one by one, into forming a genome precursor and that each such step conferred an advantage in natural selection over competitor cells, since genes would not have been lost anymore during cell division, and replication would have been synchronized. Over time, an entire small genome potentially could have organized itself in this manner, until mechanisms for internal expansion, like they are found in modern genomes, could have taken over, e.g. gene duplication and variation of the duplicated gene. Going further, an RNA genome could have been replaced, bit by bit, with a DNA genome.
Certainly, reading of a string of ligated RNA as several single genes would have required that the primitive cellular mechanism would have come across a way to recognize beginning and end of a sequence. Things like promoter regions and start/stop codons, in the form they are used in protein genes, would not have been present in a primitive RNA organism.
As for the universal genetic code: currently there is good evidence that the genetic code may not be arbitrary, as a “frozen accident”, but may have been selected for by evolution up until the last universal common ancestor, see for example Knight et al. 1999, Freeland et al. 2003.
Although much of the above scenario of origin of life is still highly speculative, it is not entirely implausible.
In any case, minerals most likely provide the clue to a lot of the answers regarding the origin of life. They have been demonstrated to allow for the prebiotic synthesis of nucleotide precursors that have so far proven elusive, for example, the synthesis of ribose in sufficient purity – borate minerals stabilize ribose (Ricardo, A et al. 2004; see also press release; however, for a possible stereoselective synthesis of D-ribose catalyzed by amino acids, see below). Minerals have also been shown to catalyze polymerization of nucleotide-like molecules (Orgel 2004). Vesicle formation is aided by them as well, and mineral particles could have wound up inside vesicles and there exhibited catalytic properties (Hanczyc et al. 2003, Hanczyc et al. 2006).
In a different putative scenario, minerals also play an interesting role. Instead of in an aqueous “prebiotic soup” on or near the surface of the earth, it has been hypothesized that life may have begun in the depths of the ocean, in the unique environment of deep-sea hydrothermal vents. In high-pressure, high-temperature water as there, organic molecules show a level of (albeit not always particularly specific) chemical reactivity that is usually observed in “normal” aqueous environments only upon speeding-up of reaction rates by enzymes. See for example the review Hazen et al. 2002. For the physico-chemical properties of high-pressure, high-temperature water, see Basset M-P 2003 and Gen-e-sis: 1. Catalysis by minerals, such as those present in deep-sea hydrothermal vents, further enhances chemical reactions in such an aqueous environment. Degradation under these high-temperature and high-pressure conditions of synthesized organic molecules may be prevented by minerals as well – at least this has been shown for amino acids (see Hazen et al. 2002). Fatty acids, as a source of membrane-forming material, might have been synthesized there too (see Orgel 2004).
In addition, there are hydrothermal vents that have semi-permeable microenvironments (mimicking a lipid membrane), which could retain molecules at high concentrations (a possible solution to the “dilution problem”) (5).
Chemistry in these environments may open up new possibilities for the synthesis of a genetic polymer preceding RNA, required in the “gene-first” scenario.
Some extend the above findings from deep-sea hydrothermal vents to a variant of the “metabolism-first” scenario as opposed to the “gene-first” scenario described above, a hypothesis that claims that complex metabolic cycles could have self-organized, independent of a genetic system, which is capable of providing polymer catalysts. Such complex self-organization putatively could have been the basis for the synthesis of nucleotides and oligonucleotides, among others. In an inventive, original and detailed model, Günter Wächtershäuser, now a prominent name in the field, had suggested that the beginning of life was what has been termed a “flat life”, an elaborate two-dimensional metabolism on mineral surfaces of deep-sea hydrothermal vents; this also addresses the “dilution problem” by concentrating all chemistry on the surface. For his model, see Wächtershäuser 1988, Wächtershäuser 1990 and follow-up publications.
However, in my view this and related scenarios (e.g. Smith, Morowitz 2004) are, given current knowledge in the field of chemistry, unrealistic. It is simply hard to see how the extraordinarily high specificity of chemical reactions, required for complex sequences of reactions and metabolism – for the functioning of life –, would in general be possible without catalytic polymers featuring a three-dimensional substrate pocket. These are only provided by a “gene-first” scenario. I have to agree with Orgel when he says (the reverse citric acid cycle that he mentions lies at the center of the “metabolism-first” scenario proposed by Wächtershäuser):
“There is no agreement on the extent to which metabolism could develop independently of a genetic material. In my opinion, there is no basis in known chemistry for the belief that long sequences of reactions can organize spontaneously – and every reason to believe that they cannot. The problem of achieving sufficient specificity, whether in aqueous solution or on the surface of a mineral (6), is so severe that the chance of closing a cycle of reactions as complex as the reverse citric acid cycle, for example, is negligible. The same, I believe, is true for simpler cycles involving small molecules that might be relevant to the origins of life and also for peptide-based cycles.”
Thus it appears reasonable to assume that the development of metabolic cycles and pathways would have required genetic/catalytic polymers – even though, obviously, opinions are divided on the issue.
Interestingly, Robert Hazen, who apparently favors the “metabolism-first” model, does not even mention the matter of specificity in Gen-e-sis, when he writes on p. 208 ff. about the reverse citric acid cycle. He only points out that reaction speed is an issue without enzymes, something that may be resolved by mineral catalysis (p. 210). Yet without sufficient reaction specificity – in addition to just reaction speed – a cycle could not close. And even if, hypothetically, it could do so for a moment, it would not be able to sustain itself, since it would become “diluted out” after a few rounds. Such specificity is not evident from the experiments described so far.
In this context, however, it is of high interest that recently it has been shown by a group from Stockholm University that such simple organic molecules as single amino acids, which might also be generated in deep-sea hydrothermal vents, can catalyze the stereo-specific synthesis of sugars (building blocks of nucleotides, themselves parts of RNA and DNA) from simple starting materials with enzyme-like specificity, albeit only in organic solvents (Cordova, Ibrahem et al. 2005; see also Richard Robinson’s article, in which Harold Morowitz expresses excitement about the potential of catalysis by small organic molecules for metabolism). For example, either L- or D-enantiomers of certain amino acids (with serine among those, see below) can trigger formation not just of a certain type of sugar with excellent chemoselectivity (i.e. avoiding unwanted side reactions), but also the formation of one out of 16 possible enantiomers of this sugar with approx. 99% stereospecificity.
Even more promising are the findings in a follow-up paper (Cordova et al. 2006). Here the authors report that small peptides (mainly dipeptides tested, or peptides with no more than five amino acids) (7) and amino acid tetrazoles can catalyze aldol reactions, some of them yielding sugars, with great stereospecificity in water, not just in organic solvents.
Could highly specific catalysis by amino acids or other small organic molecules (8) more generally, i.e. not just in this particular reaction, substitute to a certain extent for the three-dimensional substrate pockets of catalytic polymers? Could it close metabolic cycles?
These speculative possibilities, involving specific catalysis by organic molecules, would come closer to the “metabolism-first” scenario which was proposed by Nobel Prize winner Christian De Duve ( American Scientist, Sept./Oct. 1995). He assumes a protometabolism involving a thioester world – providing also energy for molecular reactions – which would have become the basis for the emergence of the RNA world (this scenario was suggested for the “prebiotic soup”, not for chemistry in hydrothermal vents). However, De Duve bases his catalysis on multimers derived from thioesters, “structurally similar to the [first] small catalytic proteins”. It is not easy to see how such relatively large complex catalytic units could have constantly formed with high reproducibility in a spontaneous manner. Such reproducibility would have been required for establishing and maintaining a stable metabolism, or otherwise complex sequences of reactions – of course, it poses no problem for genetic polymers, as provided by the “gene-first” model. Reproducibility (and abundance) of gene-less synthesis of putative small molecule catalysts would likely have been a much lesser issue.
Could the unique high-pressure, high-temperature aqueous environment of deep-sea hydrothermal vents, which produces drastic changes in the reactivity of organic compounds (see above), also cause small organic molecules to act as specific catalysts that would not perform this function in “normal” aqueous solution?
Finally, to solve the main puzzle of the “gene-first” model: could catalysis by small organic molecules even be involved in synthesis of nucleotides and oligonucleotides in the form of the right stereoisomers – an immensely more complex chemistry than just the synthesis of sugars – in the absence of genetic/catalytic polymers (precursors of RNA thus)? Future research will tell us about all these issues.
The term chiral is used to describe an object which is non-superimposable on its mirror image. The mirror image forms are the enantiomers which were just mentioned. Life almost exclusively synthesizes L-amino acids and D-sugars. This homochirality is essential for the functioning of proteins as amino acid polymers, and for the structure of DNA and RNA, which requires incorporation of D-sugars.
How did this homochirality of amino acids and sugars arise? It is a question that has puzzled origin-of-life researchers for decades, yet a series of recent findings appears to address it astonishingly well.
Above experiments with stereospecific amino acid catalysis of the synthesis of sugars may provide a good start for an answer. The prebiotic stereospecific catalysis of D-sugar synthesis could have occurred by strongly enriched enantiomers of amino acids or small peptides. But how would these have been present in sufficient purity?
The answer to this question appears to be enrichment in two or three of the following steps:
How could a greater presence of one enantiomeric form of an amino acid over the other (enantiomeric excess, abbreviated: ee), be it ever so slight, have arisen at all? After all, typical synthesis of an amino acid in the laboratory results in an exact 1:1 ratio of the L and D enantiomer – a so-called racemic mixture.
One possible source are meteorites. On the Murchison meteorite, a well-studied example, the L-form of some of the amino acids found is present in up to 9% ee (Cronin, Pizzarello 1997). These ee’s may have been induced by circularly polarized UV light (Bailey et al. 1998).
Yet while chemical evolution of early life could very well have built upon meteoritic material, sources of slight or even pronounced enantiomeric excesses of amino acids could have arisen in numerous places on the prebiotic earth.
Robert Hazen and colleagues found (Hazen et al. 2001) that crystals of the common rock-forming mineral calcite (CaCO3) can preferably adsorb D- or L-forms of aspartic acid (and in preliminary experiments, D- or L-forms of alanine as well) depending on the chirality of the crystal surface. Average ee was a few percentage points. The work is also described, with a vivid and personal look behind the scenes, in Hazen’s Gen-e-sis, chapter “Left and Right”.
Even though such slight local enantiomeric excesses may be enough to trigger a series of amplifying events (see below), another study found much larger preferences of adsorption of amino acids to mineral surfaces (Wedyan, Preston 2005). While in their study quartz, kaolin and montmorillonite showed slight preferences for adsorption of enantiomers, ordinary sediments from estuaries exhibited strong selectivity. At the slightly acidic pH 4, typical D/L ratios were hugely different from 1, reaching at their most extreme up to 100 for serine. At pH 9 effects were still significant, even though less pronounced. Sediments were ashed in order to remove organic matter which could introduce chiral bias. The authors are cautious: “The possibility that the ashing process actually etches and activates the mineral surface as it burns off natural (chiral?) organic matter cannot be discounted.” However, they also note that it is remarkable that such strong selectivity should occur at all.
An entirely different, attractive mechanism to achieve enantiomeric excess has been reported which, it seems, is robust (Kojo et al. 2004). It is based on the internal properties of amino acid mixtures. Most racemic amino acids (i.e. featuring a 1:1 ratio of the L and D enantiomer) form crystals that are also racemic. However, when racemic D, L-asparagine forms crystals (which may, for example, happen upon cooling of a hot solution, or slow evaporation of a solution), they are not racemic, but show varying degrees of excess of either the L-enantiomer or the D-enantiomer. When other amino acids are present, either their L- or their D-form preferentially co-crystallizes with the enantiomeric form (L or D) of asparagine that confers an excess during formation of the particular crystal at hand.
The internal consistency of the results of the study was quite amazing. If there was a bias towards L-asparagine in one crystal, practically all the other 12 co-crystallized amino acids showed a bias towards the L-form as well. The degree of bias towards L-asparagine varied between crystals; when the bias was more pronounced, the bias towards the L-form in the other amino acids was also more pronounced. The ee’s observed could be very high. If in yet another crystal there was a bias towards the D-form of asparagine, the other co-crystallized amino acids exhibited a bias towards their D-enantiomer as well.
Of course, the overall enantiomeric balance of the entire amino acid mixture (the sum of all crystals and the liquid phase above them) was still racemic. However, if, these kinds of crystals were formed from a solution on the prebiotic earth, and some of the crystals were then physically separated from others – which could have happened by many ordinary processes – and later redissolved, the amino acids in solution would automatically have exhibited an ee. Depending on the particular crystal where the solution derived from, this ee could have been high.
Simultaneous independent studies by different groups have shown that, once an initial ee in a mixture of amino acids exists, even if it is just very slight, it can have an enormous effect. This effect can occur when solid and dissolved amino acids coexist in equilibrium, something that could well have happened in a prebiotic landscape – all that is needed is an aqueous environment where the concentration of a particular amino acid is so high that it partially falls out of solution and forms a solid phase. This could, for example, take place by change of the aqueous environment (temperature, salt, pH etc.) or simply by limited evaporation of an aqueous solution.
A detailed study was performed by the British group of Donna Blackmond (Klussmann et al. 2006). When an amino acid mixture shows an excess of one enantiomer, in most cases an equilibrium of solid and dissolved amino acids will consist of the following two or three components:
- racemic crystals (crystals with a 1:1 ratio of the L- and D-enantiomers, no enantiomeric excess either way)
- pure crystals of the enantiomer in excess (Whether just one of these two solid phases is present, or both coexist, will depend on the overall ee.)
- an amino acid solution in equilibrium with the solid phase(s) which also exhibits a certain ee.
Yet as it turns out, for several amino acids this ee in solution is much higher than the overall ee of the total mixture (the solid material, dominated by racemic crystals, thus shows correspondingly less ee). At the most extreme, serine provides an almost enantiopure solution (> 99% ee) in water from a nearly racemic sample (only about 1% ee) under solid-liquid equilibrium conditions.
A smaller study, independently conducted around the same time, reports similar findings (Breslow, Levine 2006). Slow evaporation of an aqueous solution of phenylalanine at just 1% ee of the L-enantiomer led to a solution of this amino acid with 40% ee of the L-enantiomer above solid material. If, in turn, such a solution was allowed to evaporate, the resulting solution in equilibrium with the solid material had a 90% ee. Similar results were found with the D-enantiomer.
In both these studies the findings are discussed also in terms of amino acid catalysis of aldol reactions, which include sugar synthesis.
If all the above mechanisms were not enough, and an almost enantiopure amino acid solution, which could catalyze the stereospecific synthesis of sugars, was still not at hand in the local environment where primitive life first was shaped, the final boost towards an enantiopure sugar could have occurred the following way:
The group from Stockholm University showed that in the catalysis of sugar synthesis by amino acids a significant amplification of enantiomeric excess can occur (Cordova, Engquist et al. 2005). For example, a reaction mediated by the amino acid proline with an ee as low as 40% still yielded almost enantiopure hexose sugar. The reaction catalyzed by proline at 10% ee furnished the sugar with 33% ee.
In summary, a plausible scenario for the stereospecific synthesis of D-sugars (up to > 99% ee) on the prebiotic earth can be envisioned from all the findings described:
- Origin of local enantiomeric excesses – slight or pronounced – of amino acids that can catalyze the reaction
- In case these ee’s are just slight, a tremendous amplification by solid phase-liquid phase equilibria; these can also enhance already high ee’s
- If necessary, further chiral amplification during amino acid catalysis of sugar synthesis
[If specific enantiomers of small peptides were the catalysts (see above), they could have arisen by conjugation of amino acids after enantiomeric enrichment.]
Once enantiopure D-sugars were synthesized, and also had found their way into RNA, which incorporates D-ribose, the step towards L-amino acids in proteins might have been an automatic one.
Protein synthesis requires the aminoacylation of RNA, and it is in this step that L-amino acids could have been selected for, as an elegant study shows (Tamura, Schimmel 2004).
In the study, an RNA minihelix was used that recapitulates the domain within transfer RNAs which harbors the amino acid attachment site. Such RNA minihelices are thought to be the progenitors of modern transfer RNAs, which present amino acids bound to them to the ribosomal machinery that makes proteins. The authors showed that the RNA minihelix was aminoacylated by activated amino acids with a clear preference for L- as opposed to D-amino acids. A mirror-image RNA system showed the opposite selectivity.
Three amino acids – alanine, leucine and phenylalanine – were tested, and the observed selectivity was 4-fold. The authors point out in a follow-up paper, which studies the mechanism (Tamura, Schimmel 2006), that such a 4-fold effect, repeated under selective pressure many times, can lead to an overwhelming preference for an L-amino acid in a biological system.
On the other hand, peptide chains could also have acquired chirality independent of RNA, see for example:
Saghatelian et al. 2001; see also press release, Hitz, Luisi 2004, Weissbuch et al. 2004, Plankensteiner et al. 2005 (this last article, together with a series of related ones, is also available in “The Amino Acid and Peptide World and the Origin of Life” (PDF). <!– [http://www-c724.uibk.ac.at/theochem/staff/kpl/Dissertation.pdf]; you may need to copy the URL into the URL window of your browser). –>
Of course, any such mechanisms in peptides could have been aided as well by an enantiomeric excess of amino acids as achieved in the above described scenarios.
Whatever the precise sequence of events at the origin of life may have been, the cumulative strength of all of the above data indicates that the “mystery” of the origin of life’s chirality will have a perfectly natural explanation after all.
The issue of chirality, among others, has been touted by creationists as a “huge problem” for the concept of an origin of life by natural causes. Allegedly, only a miraculous intervention by God could have solved the problem. Yet the above findings are a typical example for why the “God-of-the-gaps” concept does not work: science rapidly closes the gaps that previously might have been thought to be reserved for miraculous intervention.
This is exactly what should be expected if either the material world is all there is, or if the world was created by a God who, as primary cause, chose to create through secondary causes – precisely those natural causes that science studies. In fact, creationists should seriously ask themselves if their concept of God is not a belittling one: the Intelligent Designer as “tinkerer” who is forced to break his own created laws of nature once in a while because they are insufficient to achieve certain stages in the development of the material world. From a theistic philosophical perspective, the actual findings of science suggest a much grander idea of God: the Designer who laid out an elegant and self-sufficient set of laws of nature that accomplish the unfolding of his creation by inducing self-organization of the material world. This idea is easily compatible with the concept of God of many mainstream religions, including most Christian ones.
Obviously the “gaps” are closing everywhere: for example, another proudly recurring theme on creationist websites is the “enormous difficulty” of prebiotic synthesis of ribose, since ribose would have been very unstable under the alkaline conditions thus far believed to be necessary for its synthesis. Yet newer research, which the creationists apparently are unaware of, has made this objection irrelevant: as mentioned above, already quite a while ago this problem had been reported to be resolved – borate minerals stabilize ribose (Ricardo, A et al. 2004; see also press release). In addition, when it comes to stereospecific sysnthesis of D-ribose, catalysis of the reaction by amino acids or small peptides – probably under less extreme conditions of pH – appears a distinct possibility, as we have seen.
A recurring theme in the literature about the origin of life have been hypercycles, as an overarching organization of autocatalytic sets. They were proposed as a model for the origin of life and are explained at Principia Cybernetica Web.
However, complex hypercycles lead their colorful lives only as computer simulations. Thirty-five years after the introduction of the hypothesis by Eigen et al., there is no experimental evidence whatsoever for possible complex, prebiotic hypercycles (9). This makes them still nothing more than mere speculation, and the hypothesis is not frequently mentioned in more recent scientific literature on the origin of life.
Even though recent, exciting research has provided plausible scenarios for the origin of life and has answered many questions, it is clear that a lot of research remains to be done, since much of the origin-of-life scenarios is still hypothesis. Experimental models are needed that are both realistic and of some appreciable complexity. Were it possible, for example, to show that a primitive RNA organism could be built in the laboratory, as the Szostak lab plans to do, it would be a significant step forward. For this, see Carl Zimmer’s article; there also the hope is expressed that evolution of such an organism might be observable on the lab bench. I would agree that ethical issues are practically non-existent since the organism would not be able to live in the present-day outside world.
- Read: given current data, a scientifically rather unsustainable proposition; oligonucleotides are short pieces of RNA or DNA.
- After Orgel’s review was written, the group of Szostak showed that an existing DNA polymerase is able to replicate the simpler TNA with high fidelity (Ichida et al. 2005).
- For availability of fatty acids in the prebiotic environment and vesicle studies, see Hanczyc et al. 2003.
For an overview of vesicle studies by diverse groups, see Hanczyc, Szostak 2004.
For availability of prebiotic fatty acids, see also Orgel 2004.
- Outside a cellular context, RNA evolution in the laboratory is routine now – RNA engineering according to mechanisms somewhat resembling evolution has been shown to produce many RNA catalysts. Certainly, all the sophisticated procedures used in the laboratory would not have been available to an ancient precursor of a cell. On the other hand, nothing in the test tube quite resembles the possibilities that would arise from the competition of replicating cells.
- See Richard Robinson’s article and Russell, Martin 2004. The hypothesis that life might have originated in hydrothermal vents appears, to some extent, attractive also due to the fact that metal sulfides, found in those locations and capable of catalysis of simple organic reactions, are still found in the catalytic centers of central metabolic enzymes such as ferredoxin, succinate dehydrogenase and bacterial acetyl-CoA synthase – possibly hinting at evolutionary preservation of primitive metal sulfide catalysis.
- Wächtershäuser presents interesting theoretical arguments why surface metabolism, as opposed to reactions in solution, should address reaction specificity in a sufficient manner, yet his viewpoint has also been criticized by Christian De Duve and Stanley Miller in a joint article (De Duve, Miller 1991). Experimental evidence for the specificity of a sequence of reactions strictly confined to a mineral surface (not just catalyzed by it) would be hard to obtain, and also Wächtershäuser’s own experimental publications, as well as those of others testing his “metabolism-first” scenario, thus far do not really address this issue so central to his hypothesis, but instead concentrate on other, simpler aspects of his model.
- Several examples of how prebiotic oligomerization of amino acids to small peptides could have occurred:
- Highly specific primarily in terms of chemoselectivity; in terms of stereospecificity only where applicable. This specific catalysis could also take place in conjunction with complex formation on mineral-surfaces as found in deep-sea hydrothermal vents, or organic molecules might form complexes with metal solubilized in this high-temperature, high-pressure environment (cf. Gen-e-sis: 119).
- To my knowledge, so far only one, very simple, hypercycle has been generated almost 10 years ago in the lab with two self-replicating peptides by the group of Ghadiri (Lee et al. 1997). However, these authors posted a correction where they state: “Although the kinetic data suggest the intermediary of higher-order species in the autocatalytic processes, the present system should not be referred to as an example of a minimal hypercycle in the absence of direct experimental evidence for the auto-catalytic cross-coupling between replicators.”There is a single report on a naturally occurring hypercycle (Eigen M et al. 1991). But this example, relating to the infection cycle of an RNA bacteriophage, obviously is not prebiotic – the cycle evolved on the complexity of living beings as template.
Bailey J, Chrysostomou A, Hough JH, Gledhill TM, McCall A, Clark S, Menard F, Tamura M. 1998. Circular polarization in star-formation regions: implications for biomolecular homochirality. Science 281: 672-674 [PubMed]
Basset M-P. 2003. Is high-pressure water the cradle of life? Journal of Physics: Condensed Matter 15: 1353-1361 [Abstract, free full text available there]
Breslow R, Levine SM. 2006. Amplification of enantiomeric concentrations under credible prebiotic conditions. Proceedings of the National Academy of Sciences USA 103:12979-12980 [PubMed]
Chyba C, Sagan C. 1992. Endogenous production, exogenous delivery and impact-shock synthesis of organic molecules: an inventory for the origins of life. Nature 355:125-132 [PubMed]
Chyba CF. 2005. Rethinking Earth’s early atmosphere. Science 308: 962-963 [PubMed]
Cordova A, Ibrahem I, Casas J, Sunden H, Engquist M, Reyes E. 2005. Amino acid catalyzed neogenesis of carbohydrates: a plausible ancient transformation. Chemistry 11: 4772-4784 [PubMed]
Cordova A, Engquist M, Ibrahem I, Casas J, Sunden H. Plausible origins of homochirality in the amino acid catalyzed neogenesis of carbohydrates. Chemical Communications (2005): 2047-2049 [PubMed]
Cordova A, Zou W, Dziedzic P, Ibrahem I, Reyes E, Xu Y. 2006. Direct asymmetric intermolecular aldol reactions catalyzed by amino acids and small peptides. Chemistry 12: 5383-5397 [PubMed]
Cronin JR, Pizzarello S. 1997. Enantiomeric excesses in meteoritic amino acids. Science 275: 951-955 [PubMed]
Eigen M, Biebricher CK, Gebinoga M, Gardiner WC. 1991. The hypercycle. Coupling of RNA and protein biosynthesis in the infection cycle of an RNA bacteriophage. Biochemistry 30:11005-11008 [PubMed]
Freeland SJ, Wu T, Keulmann N. 2003. The case for an error minimizing standard genetic code. Origin of Life and Evolution of the Biosphere 33: 457-477 [PubMed]
Glass JI, Assad-Garcia N, Alperovich N, Yooseph S, Lewis MR, Maruf M, Hutchison CA 3rd, Smith HO, Venter JC. 2006. Essential genes of a minimal bacterium. Proceedings of the National Academy of Sciences USA 103: 425-430 [PubMed] [Free full text]
Hanczyc, MM, Fujikawa, SM, Szostak, JW. 2003. Experimental models of primitive cellular compartments: Encapsulation, growth and division. Science 302:618-622 [PubMed] [full article on the publication website of the Szostak lab]
Hanczyc MM, Szostak JW. 2004. Replicating vesicles as models of primitive cell growth and division. Current Opinion in Chemical Biology 8:660-664 [PubMed] [full article on the publication website of the Szostak lab]
Hanczyc MM, Mansy SS, Szostak JW. Mineral Surface Directed Membrane Assembly. Origin of Life and Evolution of the Biosphere 2006 Aug 15; [Epub ahead of print] [PubMed] [full article on the publication website of the Szostak lab]
Hazen, RM, Filley TR, Goodfriend GA. 2001. Selective adsorption of L- and D-amino acids on calcite: Implications for biochemical homochirality. Proceedings of the National Academy of Sciences USA 98:5487-5490 [PubMed] [Free full text]
Hazen RM, Boctor N, Brandes JA, Cody GD, Hemley RJ, Sharma A, Yoder Jr HS. 2002. High pressure and the origin of life. Journal of Physics: Condensed Matter 14: 11489-11494 [Abstract]
Hitz TH, Luisi PL. 2004. Spontaneous onset of homochirality in oligopeptide chains generated in the polymerization of N-carboxyanhydride amino acids in water. Origin of Life and Evolution of the Biosphere 34:93-110 [PubMed]
Huber C, Eisenreich W, Hecht S, Wächtershäuser G. 2003. A possible primordial peptide cycle. Science 301: 938-940 [PubMed]
Ichida JK, Horhota A, Zou K, McLaughlin LW, Szostak JW. 2005. High fidelity TNA synthesis by Therminator polymerase. Nucleic Acids Research 33:5219-5225 [PubMed] [full article on the publication website of the Szostak lab]
Kasting JF, Zahnle KJ, Walker JCG. 1983. Photochemistry of methane in the earth’s early atmosphere. Precambrian Research 20: 121-148 [Abstract]
Klussmann M Iwamura H, Mathew SP, Wells DH Jr, Pandya U, Armstrong A, Blackmond DG. 2006. Thermodynamic control of asymmetric amplification in amino acid catalysis. Nature 441: 621-623 [PubMed]
Knight RD, Freeland SJ, Landweber LF. 1999. Selection, history and chemistry: the three faces of the genetic code. Trends in Biochemical Sciences 24: 241-247 [PubMed]
Kojo S, Uchino H, Yoshimura M, Tanaka K. 2004. Racemic D,L-asparagine causes enantiomeric excess of other coexisting racemic D,L-amino acids during recrystallization: a hypothesis accounting for the origin of L-amino acids in the biosphere. Chemical Communications (2004): 2146-2147 [PubMed]
Kuhn WR, Atreya SK. 1979. Ammonia photolysis and the greenhouse effect in the primordial atmosphere of the earth. Icarus 37: 207-213 [Abstract]
Lee DH Severin K, Yokobayashi Y, Ghadiri MR. 1997. Emergence of symbiosis in peptide self-replication through a hypercyclic network. Nature 390:591-594) [PubMed]
Leman L, Orgel LE, Ghadiri MR. 2004. Carbonyl sulfide-mediated prebiotic formation of peptides. Science 306: 283-286 [PubMed]
Orgel LE. 1998. The origin of life – a review of facts and speculations. Trends in Biochemical Sciences 23: 491-495 [PubMed]
Orgel LE. 2004. Prebiotic Chemistry and the Origin of the RNA World. Critical Reviews in Biochemistry and Molecular Biology 39: 99-123 [PubMed] [ free full text, you may need to register on the link, which is free]
Plankensteiner K, Reiner H, Rode BM. 2005. Stereoselective differentiation in the salt-induced peptide formation reaction and its relevance for the origin of life. Peptides 26:535-541 [PubMed]
Pavlov AA, Kasting JF, Brown LL, Rages KA, Freedman R. 2000. Greenhouse warming by CH4 in the atmosphere of early Earth. Journal of Geophysical Research 105: 11981-11990 [PubMed]
Ricardo, A Carrigan MA, Olcott AN, Benner SA. 2004. Borate minerals stabilize ribose. Science 303:196 [PubMed]
The RNA World, 2nd edition (2000), Cold Spring Harbor Laboratory Press, Raymond F. Gesteland, Thomas R. Czech, John F. Atkins eds.
Second edition links: [Amazon] [Powell’s] [Barnes and Noble] <!–[Find a Library]–>
Third edition links: [Amazon] [Powell’s] [Barnes and Noble] [Find a Library]
Sagan C, Chyba C. 1997. The early faint sun paradox: organic shielding of ultraviolet-labile greenhouse gases. Science 276: 1217-1221 [PubMed]
Saghatelian A, Yokabayashi Y, Soltani K, Ghadiri MR. 2001. A chiroselective peptide replicator. Nature 409:797-801 [PubMed]
Schaefer L, Fegley Jr. B. 2006. Outgassing of ordinary chondritic material and some of its implications for the chemistry of asteroids, planets, and satellites. Icarus, in press
[preprint on the publication page of the Planetary Chemistry Laboratory in St. Louis]
Tian F, Toon OB, Pavlov AA, De Sterck H. 2005. A hydrogen-rich early Earth atmosphere. Science 308: 1014-1017 [PubMed]
Tamura K, Schimmel P. 2004. Chiral-selective aminoacylation of an RNA minihelix. Science 305:1253 [PubMed]
Tamura K, Schimmel P. 2006. Chiral-selective aminoacylation of an RNA minihelix: Mechanistic features and chiral suppression. Proceedings of the National Academy of Sciences USA 103:13750-13752 [PubMed]
Wedyan M, Preston MR. 2005. Isomer-selective adsorption of amino acids by components of natural sediments. Environmental Science & Technology 39:2115-2119 [PubMed]
Weissbuch I, Bolbach G, Leiserowitz L, Lahav M. 2004. Chiral amplification of oligopeptides via polymerization in two-dimensional crystallites on water. Origin of Life and Evolution of the Biosphere 34:79-92 [PubMed]
Zahnle KJ. 1986. Photochemistry of methane and the formation of hydrocyanic acid (HCN) in the earth’s early atmosphere. Journal of Geophysical Research 91: 2819-2834 [Abstract]