Some Papers on Origin of Life Research

A collection of relevant modern research on the study of the origin of life, focusing on the experimental details of prebiotic chemistry.

• The titles of each paper link to the original source.
• My quick summary of each paper is provided for each one, intended to be easy to understand.
• Papers are listed in no particular order.

CONTENTS

• A. Astrobiology, Astrochemistry and Geochemistry
• B. Homochirality
• C. Autocatalysis and Systems Chemistry
• D. Non-Equilibrium Thermodynamics and Information Theory
• E. Synthesis of Small Molecules
• F. Synthesis of Macromolecules
• G. Reactions of Macromolecules
• H. Protocell Models
• I. Synthetic Biology and Molecular Biology

ASTROBIOLOGY, ASTROCHEMISTRY AND GEOCHEMISTRY

Chemical reactions occurring in space, forming the ‘building blocks of the building blocks’ that were likely present at Earth’s formation, or were delivered during the bombardment period that followed. Solid-phase chemistry and transport phenomena enabled by geologic formations may also be relevant.

Chemistry in space

Kwok, Sun. (2009) ‘Delivery of Complex Organic Compounds from Planetary Nebulae to the Solar System.’ International Journal of Astrobiology, 8(3): pp. 161–67. doi:10.1017/S1473550409004492.

IR spectroscopy shows that nebulae (where stars come from) contain complex organic compounds (containing aromatic rings and aliphatic chains), synthesised on timescales of ~1,000 years. These compounds get dispersed throughout space and could have showered the Earth during the bombardment period in the Hadean.

Ciesla, F.J. and Sandford, S.A. (2012) ‘Organic synthesis via irradiation and warming of ice grains in the solar nebula’, Science, 336(6080), pp. 452–454. doi:10.1126/science.1217291.

A model of the outer solar nebula (the dust clouds surrounding the solar system during its formation) found that the ice grains (containing H$_2$O, CO, CO$_2$, NH$_3$ and CH$_3$OH) would have been exposed to UV radiation and temperatures of ~30 K. Lab studies found that these same conditions lead to organic molecule formation including amino acids, lipids, quinones and nucleobases.

Chemistry on the early Earth

Schaible, M. J. et al. (2024). ‘The Astrobiology Primer v3.0’. Astrobiology, 24(S1), doi:10.1089/ast.2021.0129

Discusses a wide variety of geochemical processes that were present on the early Earth, that provided sources of inorganic chemicals and energy gradients.

Haas, S., Tutolo, B. M. and Catling, D. C. (2025). ‘Soda lake phosphorus fluxes controlled by biological uptake imply abundant phosphate in plausible origin-of-life environments’, Geochimica et Cosmochimica Acta, 393, pp.63-74. doi:10.1016/j.gca.2025.01.040.

Phosphorus is required by all life (e.g. in RNA), but it is scarcely bioavailable in natural environments (the ‘phosphorus problem’). It is known that iron-rich volcanic rocks can react with hot water or steam to form a range of phosphates, and that evaporation of resulting solutions can concentrate the phosphorus compounds. In sodium-carbonate (soda)-rich evaporative lakes with nearby volcanic activity, carbonates bind Ca$^{2+}$, preventing it from precipitating phosphate and allowing phosphorus to reach concentrations up to several millimolar. This is a prebiotically plausible solution to the phosphate problem. Phosphate-rich lakes may have preferentially formed on the prebiotic earth because of carbonic acid weathering under CO$_2$-rich atmospheres.

Priye, A. et al. (2017). ‘Synchronized chaotic targeting and acceleration of surface chemistry in prebiotic hydrothermal microenvironments’, Proceedings of the National Academy of Sciences, 114(6), pp. 1275–1280. doi:10.1073/pnas.1612924114.

Synthesis of polymers typically requires monomers to be concentrated in the same place, but the vast volume of the prebiotic ocean would presumably dilute the chemicals within. However, the turbulent flows driven by thermal convection from subsea hydrothermal vents allow for mixing of chemical species in the bulk water while concentrating them at microporous mineral surfaces, enhancing surface reaction kinetics by a factor of ~1000. This effect was studied both by simulation and with microfluidics-based experiments.

Barge, L. M. and Price, R. E. (2023). ‘Diverse geochemical conditions for prebiotic chemistry in shallow-sea alkaline hydrothermal vents’, Nature Geoscience, 15, pp.976-981. doi:10.1038/s41561-022-01067-1.

Shallow pools and hydrothermal vents are often proposed as sites for prebiotic chemistry, and several sites are known where both conditions exist together in the same place. Conditions at these shallow hydrothermal fields include wet-dry cycling, temperature variations and influxes of both saltwater and freshwater. This may remove the need for long-distance mass transport, allowing prebiotic chemistry to overcome the ‘dilution problem’.

Baaske, P. et al. (2007). ‘Extreme accumulation of nucleotides in simulated hydrothermal pore systems, Proc. Natl. Acad. Sci. U.S.A., 104(22) pp.9346-9351, doi:10.1073/pnas.0609592104.

The thermal gradients in the plugged pore systems of hydrothermal vents leads to convection-driven flow patterns. This leads to extreme accumulation of chemical concentrations at the base of the pore. By performing experiments in microfluidics channels, the diffusion and thermophoresis parameters of nucleotides and RNAs were determined and used in a computational model to simulate the concentration distribution in a pore. Accumulation grows exponentially with channel length and oligo-RNA length. With a realistic 30 C temperature difference, 7-fold accumulation is found for nucleotides in a 10:1 aspect channel, rising to several billion-fold for 22-mer ssRNA in a 125:1 aspect channel, over a timescale of hours to days. Tar formation is experimentally found to be precluded by the termination of thermophoresis in the molar range, as well as convective removal of microparticles. This provides a robust solution to the ‘concentration problem’, and can also apply to amino acids and lipids.

Maeda, Y., Tlusty, T. and Libchaber, A. (2012). ‘Effects of long DNA folding and small RNA stem–loop in thermophoresis, Proc. Natl. Acad. Sci. U.S.A., 109(44) pp.17972-17977, doi:10.1073/pnas.1215764109.

A genome-sized strand of DNA (from an extant virus) is added to a 5% PEG aqueous solution. When a temperature gradient is applied, the PEG migrated away by thermophoresis, leaving an osmotic pressure gradient that caused the DNA to form ring structures. Next, the DNA was replaced with short RNAs. For oligo-RNAs without stem-loop secondary structures, such as poly-uracils, no accumulation is observed, leading to depletion. However, for ssRNAs with stem-loop structures, accumulation in the temperature gradient is observed, increasing with stem-loop length. Accumulation was also observed for a 6-mer dsRNA self-acylating ribozyme. The accumulation effect is largest for low salt concentrations. In combination with their previous study where the non-prebiotically-relevant PEG was substituted for large rRNAs yielding the same observations, this provides a mechanism for selective concentration of RNAs by the complexity of their secondary structures, a defining feature of RNA catalytic chemistry, while removing inert homogeneous RNAs.

Mayer, C., Schreiber, U. and Dávila, M. (2017). ‘Selection of Prebiotic Molecules in Amphiphilic Environments’. Life, 7(1), p.3. doi:10.3390/life7010003.

Fluid circulating in tectonic fault zones contains water and supercritical CO$_2$, a well-known solvent for extraction and amphiphile nanoparticle synthesis in organic chemistry. This fluid would be supplied with hydrothermally formed compounds, is protected against dilution effects and destructive UV radiation, provides steady conditions over long timescales, generates structures which are highly selective for amphiphilic compounds such as lipids or peptides composed of hydrophilic and hydrophobic amino acids, and has access to the ‘warm little pond’ environments at the surface.

Chemical compounds from space

McGeoch, J. E. M., & McGeoch, M. W. (2022). ‘Chiral 480 nm absorption in the hemoglycin space polymer: a possible link to replication’, Scientific reports, 12(1), 16198. doi:10.1038/s41598-022-21043-4.

An abiotic polypeptide has been detected identically in 6 different meteorites that landed on Earth, originally misidentified as ‘hemolithin’. Hemoglycin contains 18 glycine (the simplest amino acid) and 4 hydroxyglycine (an abiotic, chiral amino acid) residues, and these were enantioenriched towards the D enantiomer. The residues formed a beta-pleated sheet and the terminal residues were bonded to iron(II) centres. Template self-replication by hydrogen bonding is proposed as the cause for the commonality between multiple meteorites. Hemoglycin has also been found in the sea foams which collect cosmic dust infall, with isotopic abundances indicating extraterrestrial origin.

McCoy, T. J., Russell, S. S., Zega, T. J. et al. (2025). ‘An evaporite sequence from ancient brine recorded in Bennu samples’, Nature 637, 1072–1077. doi:10.1038/s41586-024-08495-6.

Salt minerals were found on the asteroid Bennu, including sodium phosphates, carbonates, sulfates, chlorides and fluorides, forming a frozen brine. These salts were present as components of calcite, magnetite, magnesite, apatite, dolomite and phyllosilicate clay minerals. Many of these minerals are relevant for catalysis of organic molecules.

Glavin, D. P. et al. (2025). ‘Abundant ammonia and nitrogen-rich soluble organic matter in samples from asteroid (101955) Bennu’, Nature Astronomy. doi:10.1038/s41550-024-02472-9.

Samples from the asteroid Bennu (from the same mission as [A5]) found several amino acids (glycine most abundant at 44 nmol/g, plus 19 other non-protein amino acids) and all five nucleobases, among a range of other relevant organics. Biological contamination was ruled out. All chiral amino acids observed were near-racemic.

HOMOCHIRALITY

All chiral biogenic molecules are enantiomerically pure, existing as L-amino acids and D-sugars. A prebiotically plausible ‘symmetry breaking’ mechanism is therefore required to explain how racemic reactants were resolved. This may have occurred at the monomer or the polymer level.

Viedma, C. (2001) ‘Enantiomeric Crystallization from DL-Aspartic and DL-Glutamic Acids: Implications for Biomolecular Chirality in the Origin of Life’, Origins of Life and Evolution of the Biosphere, , 31(6), pp. 501–509. doi:10.1023/a:1012790523136.

Supersaturated solutions of racemic aspartic acid (Asp) and glutamic acid (Glu) will crystallise as enantiopure crystal grains (conglomerates), since once an amino acid has nucleated, the crystal growth favours having the same enantiomer. This process increases the ee of the supernatant.

Tarasevych, A.V. et al. (2013) ‘Deracemization of amino acids by partial sublimation and via homochiral self-organization’, Origins of Life and Evolution of Biospheres, 43(2), pp. 129–135. doi:10.1007/s11084-013-9333-6.

If enantiopure Asn/Asp/Glu/Ser/Thr and additional racemic amino acids form a supersaturated solution, which is then evaporated to form a solid, which is then sublimed at ~100 C, the resulting gas phase contains the enantioenriched amino acids i.e. the deracemisation effect in (Viedma, 2001) is extended to the other amino acids. The ee’s obtained were highly variable due to the random nature of nucleation.

Kojo, S. et al. (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, (19), p.2146. doi:10.1039/b409941a.

The deracemisation mechanism of amino acids in (Viedma, 2001) is extended to all other chiral amino acids. When a racemic solution of asparagine (Asn) and other amino acids co-crystallises, the resulting crystals were enantioenriched in both amino acids, with their ee’s strongly correlated. The range of ee’s obtained remains random and can be very high (up to 100% in some trials) in either direction.

Blackmond, D.G. and Klussmann, M. (2007) ‘Spoilt for choice: Assessing phase behavior models for the evolution of homochirality’, Chemical Communications, (39), p. 3990. doi:10.1039/b709314b.

Partially enantioenriched amino acid solutions (like those formed in (Kojo, 2004)) can recrystallise again, forming racemate crystals while the supernatant tends to the eutectic composition at thermodynamic equilibrium, which can have much higher ee (e.g. Ser > 99%, His = 93%, Leu = 87%). This process can also occur exploiting differential volatility using sublimation/deposition (solid ↔ gas) instead of dissolution (solid ↔ aqueous). Also, amino acids with low eutectic ee’s (e.g. Pro = 50%, Val = 47%) can have these eutectic ee’s increased by co-crystallising with a small achiral hydrogen bond donor molecule (e.g. CHCl$_3$), rising to >90% ee. Suitable prebiotic environments proposed include warm pools of water containing amino acids with cycles of rainfall and evaporation, or in sublimating organic space dust brought to Earth.

Deng, M., Yu, J. and Blackmond, D.G. (2024) ‘Symmetry breaking and chiral amplification in prebiotic ligation reactions’, Nature, 626(8001), pp. 1019–1024. doi:10.1038/s41586-024-07059-y.

Amidonitriles react with amino acids in a ligation (joining) reaction, catalysed by a prebiotic achiral thiol, with faster reaction when the reactants have opposite chirality (promotes heterochiral products). When enantioenriched solutions of both reactants were used, the homochiral product predominated. Also, due to the difference in solubility of the homochiral and heterochiral products, the crystallisation effects in (Kojo, 2004) led to a solution of only homochiral products.

Yu, J. et al. (2024) ‘Prebiotic access to enantioenriched amino acids via peptide-mediated transamination reactions’, Proceedings of the National Academy of Sciences, 121(7). doi:10.1073/pnas.2315447121.

A variety of similar kinetic resolutions to (Deng, Yu & Blackmond, 2024) are listed. Pyridoxamine and pyruvate react reversibly to form alanine and pyridoxal. When catalysed by dipeptides containing proline at low ee, the alanine becomes enantioenriched.

Hazen, R. M., Filley, T. R., & Goodfriend, G. A. (2001). ‘Selective adsorption of L- and D-amino acids on calcite: Implications for biochemical homochirality’, Proceedings of the National Academy of Sciences, 98(10), pp. 5487–5490. doi:10.1073/pnas.101085998.

When calcite (CaCO$_3$) is immersed in a racemic aspartic acid solution, the L and D enantiomers adsorb onto different faces of the asymmetric crystal structure, giving ee’s up to ~10%. The microscopically localised ee values may be far higher than reported due to experimental artefacts.

Michaelian K. (2018). ‘Homochirality through Photon-Induced Denaturing of RNA/DNA at the Origin of Life’. Life (Basel, Switzerland), 8(2), 21. doi:10.3390/life8020021.

UV radiation from sunlight can be scattered and totally internally reflected at a water-air interface to form ~5% circular polarised radiation during late afternoon near the sea surface. At the higher sea surface temperatures in the afternoon, this radiation could melt RNA/DNA duplexes, with faster kinetics for strands containing more D-nucleotides due to the polarisation. Strands with D-nucleotides would become more available for template replication, selecting for more homochiral RNA/DNA. L-tryptophan also complexes enantioselectively with D-RNA, also increasing the ee of the tryptophan.

Globus, N., Fedynitch, A. and Blandford, R. (2021). ‘Polarized Radiation and the Emergence of Biological Homochirality on Earth and Beyond’, The Astrophysical Journal, 910(85). doi:10.3847/1538-4357/abe461

Cosmic rays form spin-polarised muons (due to the weak force’s parity violation) that reach the surface. These could cause enantioselective mutagenesis in RNA/DNA or serve as another source of circularly polarised radiation.

Chen, Y., & Ma, W. (2020). ‘The origin of biological homochirality along with the origin of life’, PLOS Computational Biology, 16(1), pp. e1007592. doi:10.1371/journal.pcbi.1007592

Homochirality of RNA could arise at the polymer level rather than the monomer level, since polynucleotides preferentially polymerise with nucleotides of the same chirality, amplifying small initial imbalances. This was verified by computer simulation.

Hein, J. E.; Tse, E.; Blackmond, D. G. (2011). ‘A Route to Enantiopure RNA Precursors from Nearly Racemic Starting Materials’, Nature Chemistry, 3 (9), pp. 704–706. doi:10.1038/nchem.1108.

When 1% ee L-proline is enantioenriched through the crystallisation process in [B3], it is able to catalyse the reaction between racemic glyceraldehyde and 2-amino-oxazole to form enantioenriched amino-oxazolines (AOs) at 20-80% ee. Another crystallisation gives enantiopure AOs. These AOs can be converted to enantiopure RNA nucleotides by the prebiotically plausible reactions (Sutherland-Powner type syntheses).

S. F. Ozturk and Sasselov, D. D. (2022). ‘On the origins of life’s homochirality: Inducing enantiomeric excess with spin-polarized electrons’. Proceedings of the National Academy of Sciences of the United States of America, 119(28). doi:10.1073/pnas.2204765119.

Solar UV light can irradiate magnetite deposits to produced spin-polarised photoelectrons due to the spin-aligned magnetic domains. These helical electrons can carry out enantioselective redox reactions due to the chiral-induced spin selectivity (CISS) effect. This process is well-suited to carrying out cyanosulfide chemistry, forming simple sugars and hydroxy-aldehydes.

Lee, C. et al. (2022). ‘Chirality in Organic and Mineral Systems: A Review of Reactivity and Alteration Processes Relevant to Prebiotic Chemistry and Life Detection Missions’. Symmetry. 14(3):460. doi:10.3390/sym14030460

A review of the wide range of reactions catalysed by (or involving adsorption to) minerals with enantiomorphic crystal surfaces, such as (Hazen, Filley & Goodfriend, 2001).

Banerjee-Ghosh, K. et al. (2018). ‘Separation of enantiomers by their enantiospecific interaction with achiral magnetic substrates’. Science, 360(6395), pp.1331-1334. doi:10.1126/science.aar4265.

Due to the CISS effect, α-helical oligopeptides and dsDNA oligonucleotides, as well as chiral amino acids, have enantiospecific differences in initial adsorption rates on ferromagnetic surfaces, depending on the direction of magnetisation. In combination with (Ozturk & Sasselov, 2022), it complements the mechanism for CISS-mediated deracemisation.

Ozturk, S. F. et al. (2023). ‘Chirality-induced avalanche magnetization of magnetite by an RNA precursor’. Nature Communications, 14, 6351(2023). doi:10.1038/s41467-023-42130-8.

Starting with racemic ribose-aminooxazoline (RAO, a relevant pyrimidine ribonucleoside precursor) and natural near-unmagnetised magnetite, a simple crystallisation and dissolution led to enantiopure RAO and the strong magnetisation of the magnetite substrate. This occurs due to the CISS effect, where spin-aligned domains preferentially absorb one enantiomer and promote domain growth in the magnetite, setting up a feedback loop. The resulting magnetic coercivity of the magnetite was 30 times the modern geomagnetic field (stronger than in the Hadean).

AUTOCATALYSIS AND SYSTEMS CHEMISTRY

Studies on biochemical reactions in which the catalyst is also a product of the reaction, or another reaction in a closed set.

Hanopolskyi, A.I. et al. (2020) ‘Autocatalysis: Kinetics, Mechanisms and Design’, ChemSystemsChem, 3(1). doi:10.1002/syst.202000026.

A review of the general topic of autocatalysis. For an autocatalytic reaction $ A + B \rightleftharpoons 2 B $ (A: substrate, B: product and catalyst), the kinetics follow $ \frac{\mathrm{d} [B] }{\mathrm{d} t} \propto [B]^{1/2} $ when $ [A] $ is held constant (square root law). Includes a discussion of the autocatalytic nature of short RNAs (ribozymes), autocatalytic sets and autocatalytic cycles, and mentions the concept of hypercycles proposed by Eigen.

Hordijk, W. (2017). ‘Autocatalytic confusion clarified’, Journal of Theoretical Biology, 435, pp. 22–28. doi:10.1016/j.jtbi.2017.09.003.

Gives the definitions of autocatalytic sets and cycles, and discusses the relevance of autocatalytic sets for origin of life regarding self-replication, including both RNA-first and metabolism-first hypotheses. It is known that the metabolism in E. coli comprises a large autocatalytic set (and a specific type called a ‘reflexively autocatalytic food-generated (RAF) set’), satisfying rigorously-defined properties.

Xavier, J.C. et al. (2020). ‘Autocatalytic chemical networks at the origin of metabolism’. Proceedings of the Royal Society B: Biological Sciences, 287(1922), p.20192377. doi:10.1098/rspb.2019.2377.

Using a database of known biochemical reactions, it is found that the core metabolism of extant life contains a large autocatalytic set (RAF set, introduced in (Hordijk, 2017)) that can be driven by simple inorganic molecules (H$_2$, H$_2$S, CO$_2$, NH$_3$, phosphate) as seen in extant thermophilic autotrophs. The autocatalytic set is proposed to have arisen before enzymes, catalysed instead by mineral surfaces or small molecule cofactors.

Mauksch, M. et al. (2009). ‘Spontaneous Mirror Symmetry Breaking in the Aldol Reaction and its Potential Relevance in Prebiotic Chemistry’. Orig Life Evol Biosph 40, pp.79-91. doi:10.1007/s11084-009-9177-2.

An example of asymmetric organoautocatalysis, slightly more prebiotically plausible than the related Soai reaction (Soai, 1995), which is the aldol condensation of acetone and p-nitrobenzaldehyde. An organic solvent (DMSO) was used, but aldol reactions are known to be feasible in water too. It is most relevant for homochirality of sugars, and it is proposed that such a reaction may be a non-enzymatic ‘chemical ancestor’ of the gluconeogenesis pathway.

Roszak, R. et al. (2024). ‘Emergence of metabolic-like cycles in blockchain-orchestrated reaction networks’. Chem, 10(3), pp.952–970. doi:10.1016/j.chempr.2023.12.009.

A network of 4.9 billion prebiotically plausible reactions on small molecules was surveyed using distributed blockchain computing for cycle formation. About ~100 self-replicating cycles were identified. It is concluded that cycles were unlikely at the small molecule level, instead predominating at the macromolecule level.

Goldman, A.D. and Kacar B. (2021). ‘Cofactors are Remnants of Life’s Origin and Early Evolution’. Journal of Molecular Evolution. 89(3):127-133. doi:10.1007/s00239-020-09988-4.

Many cofactors of extant enzymes are chemically similar to nucleotides (e.g. acetyl CoA, S-adenosyl methionine, ATP, NADH, FAD, cAMP…), supporting the RNA first hypothesis, as they could have been catalytic active sites of ancient ribozymes which became cofactors when the surrounding ribozyme scaffolds were replaced by protein apoenzymes during the evolution of translation. Similar ideas hold for iron-sulfur cluster cofactors as remnants of geologic activity (e.g. iron-sulfur world hypothesis).

NON-EQUILIBRIUM THERMODYNAMICS AND INFORMATION THEORY

Relevant concepts in the complexification of prebiotic chemistry, as suitable systems chemically evolve towards life.

Schneider, E.D. and Kay, J.J. (1994) ‘Life as a manifestation of the second law of Thermodynamics’, Mathematical and Computer Modelling, 19(6–8), pp. 25–48. doi:10.1016/0895-7177(94)90188-0. 

Describes the ‘4th law of thermodynamics’: an open system far from equilibrium will evolve in time so as to maximise the generation of entropy production in the surroundings. This makes low entropy self-organization an inevitable consequence if the non-equilibrium state is maintained, as the system seeks to dissipate gradients in Gibbs free energy imposed upon it. A common non-biological example is the formation of convection cells in confined fluids with strong applied thermal gradients.

Koonin, E.V. and Novozhilov, A.S. (2008) ‘Origin and evolution of the Genetic Code: The universal enigma’, IUBMB Life, 61(2), pp. 99–111. doi:10.1002/iub.146.

The genetic translational code (codon table) is robust to translational errors due to redundancy, but many more possible robust codes exist, so the standard code could have arisen randomly, by a few steps of random codon series reassignments. The ‘frozen accident hypothesis’ is supported by the observation that some specific clades of extant life, as well as mtDNA, uses a range of genetic codes, all slightly differing in different ways.

Michaelian, K. (2017) ‘Microscopic dissipative structuring and proliferation at the origin of life’, Heliyon, 3(10). doi:10.1016/j.heliyon.2017.e00424.

Purine nucleobases absorb in the UVB/C spectrum strongly and undergo rapid non-radiative decay, making them strong candidates for free energy dissipation. A non-equilibrium thermodynamic model is given for UV-light-mediated prebiotic reactions relevant to these purines, including isomerisations and tautomerisation required for their synthesis from HCN, nucleoside phosphorylation, and melting of dsRNA duplexes.

Michaelian K. (2023) ‘The Non-Equilibrium Thermodynamics of Natural Selection: From Molecules to the Biosphere’. Entropy (Basel, Switzerland), 25(7), 1059. doi:10.3390/e25071059.

The non-equilibrium thermodynamic model is extended to all organised systems with a continuous free energy input, and can provide a physical basis for both chemical and biological (natural) selection.

Hazen, M. et al. (2007) ‘Functional information and the emergence of biocomplexity’. Proceedings of the National Academy of Sciences, 104(suppl_1), pp. 8574-8581. doi:10.1073/pnas.0701744104.

Defines ‘functional information’ in biologically-useful context in terms of Shannon information entropy (roughly, how ‘surprised’ we are when observing a particular outcome from a distribution). It is defined for a given system and function (within the context of the system) as $ I(E_x) = -\log_2 F(E_x) $, where $ E_x $ is the degree of function and $ F(E_x) $ is the fraction of configurations in system with degree of function $ \geq E_x $. Different functions of the same biomolecule are available in different systems.

Adami, C. (2015) ‘Information-Theoretic Considerations Concerning the Origin of Life’. Origins of Life and Evolution of Biospheres, 45, pp. 309–317. doi:10.1007/s11084-015-9439-0.

The probability of spontaneous molecular self-replication is a function of the information contained in the replicator. The probability to discover a self-replicator by random chance depends exponentially on the relative rate of formation of the monomers, and if rate at which monomers are formed is somewhat similar to the rate at which they would occur in a self-replicating polymer, the likelihood to discover such a replicator by chance is drastically increased (~64 orders of magnitude). This is verified by computer simulation.

Barbieri, M. (2018) ‘What is code biology?’. BioSystems, 164, pp. 1-10, doi:10.1016/j.biosystems.2017.10.0050303-2647.

A comprehensive review of the concept of ‘code biology’, exploring the origin of the genetic code and the RNA to protein translation system. One idea is that life originally used much fewer amino acids, but the utility of proteins drove selection for an expanded set of amino acids, rising to the 20 in use today.

Frank, S.A. (2013) ‘Topics in the theory of natural selection’. Journal of Evolutionary Biology.

A multi-part series of papers exploring natural selection using the mathematical models of population genetics and information theory. Among the conclusions, it is shown that natural selection maximises the Fisher information metric, as well as increasing the Shannon information when purifying.

Further commentary

In the view of (Jaynes, 1957), thermodynamic entropy $ S $, as explained by statistical mechanics, should be seen as an application of Shannon’s information theory: the thermodynamic entropy is interpreted as being proportional to the amount of further Shannon information needed to define the detailed microscopic state of the system, that remains uncommunicated by a description solely in terms of the macroscopic variables of classical thermodynamics. This is captured in Boltzmann’s original formula $ S = k_B \ln \Omega $: the information entropy of a system is the amount of “missing” information needed to determine a microstate, given the macrostate.

SYNTHESIS OF SMALL MOLECULES

Feasible reaction schemes based on experimental evidence to produce amino acids, sugars and nucleotides, with enantioselectivity and regioselectivity where relevant.

Synthesis of amino acids

Wagner, A. J. et al. (2017). ‘Chiral Sugars Drive Enantioenrichment in Prebiotic Amino Acid Synthesis’. ACS Central Science, 3(4), doi:10.1021/acscentsci.7b00085.

Racemic aminonitriles undergo the Strecker synthesis at pH 7-10 to form aminoamides. When catalysed by D sugars, the products are enantioenriched. With pure D-sugars (D-ribose, D-xylose, D-2-deoxyribose), the product is D-enriched, but with mixtures of D-sugars (more prebiotically relevant), the product is L-enriched, with ee’s of 15-50%.

Pulletikurti, S. et al. (2022). ‘Prebiotic synthesis of α-amino acids and orotate from α-ketoacids potentiates transition to extant metabolic pathways’. Nature Chemistry, 14, pp.1142-1150. doi:10.1038/s41557-022-00999-w.

The Strecker synthesis is the common prebiotic way to make amino acids, but extant life uses transamination of α-ketoacids. Here, amino acids (Ala, Glu, Asp, Gly) are synthesised prebiotically from α-keto acids (pyruvate, α-ketoglutarate, oxaloacetate, glyoxalate) using diamidophosphate (DAP) and cyanide ions at pH 8.5. Wet-dry cycling can also form orotate (a pyrimidine precursor) in low yield from the intermediates. Reaction of the α-ketoacid condensation products with ammonia sources produces compounds found in the Krebs cycle and its secondary metabolites, forming various pathways with relevance to systems chemistry.

Synthesis of sugars

Colón‐Santos, S., Geoffrey and Cronin, L. (2019). ‘Taming the Combinatorial Explosion of the Formose Reaction via Recursion within Mineral Environments’, ChemSystemsChem, 1(3). doi:10.1002/syst.201900014.

The formose reaction is autocatalytic, starting from formaldehyde, but produces a messy range of sugar molecules (trioses, tetroses, pentoses). By repeatedly sampling from the product mixture and adding it to a new reactant mixture, in the presence of minerals, the range of products decreases over multiple cycles. This increases the proportion of the ‘useful’ sugar products, including ribose.

Breslow, R. and Cheng, Z.-L. (2010). ‘L-amino acids catalyze the formation of an excess of D-glyceraldehyde, and thus of other D sugars, under credible prebiotic conditions’, Proceedings of the National Academy of Sciences,107(13), pp.5723–5725. doi:10.1073/pnas.1001639107.

Reaction of formaldehyde with glycolaldehyde to form glyceraldehyde in up to 20% ee in the D-enantiomer at 20 C when enantiopure L-amino acids were present in small amounts. This would lead to D-sugar formation (e.g. D-ribose) in formose reactions. The enantiopurity of the glyceraldehyde can be increased by evaporation, due to differential solubility. Ribose does not have this differential solubility property, although ribose-containing nucleosides do, permitting their amplification.

Synthesis of nucleotides

Kim, H.-J. and Benner, S.A. (2021). ‘Abiotic Synthesis of Nucleoside 5′-Triphosphates with Nickel Borate and Cyclic Trimetaphosphate (CTMP)’. Astrobiology, 21(3), pp.298–306. doi:10.1089/ast.2020.2264.

CTMP reacts with nucleosides to nucleotides, with regioselectivity towards phosphorylation at the 5’ position in the presence of nickel(II) borate. Due to the borate ion complexing at the 2’ and 3’ -OH groups of ribose, aldopentoses are selectively stabilised by borates.

Gibard, C. et al. (2017). ‘Phosphorylation, oligomerization and self-assembly in water under potential prebiotic conditions’. Nature Chemistry, 10(2), pp.212–217. doi:10.1038/nchem.2878.

Diamidophosphate (DAP) is a prebiotically plausible phosphorylating agent produced from trimetaphosphate. DAP efficiently phosphorylates nucleosides, amino acids and lipid precursors in water. Oligonucleotides, peptides and liposomes are also formed under the same conditions, a one-pot synthesis of small polymers.

Becker, S. et al. (2019). ‘Unified prebiotically plausible synthesis of pyrimidine and purine RNA ribonucleotides’. Science, 366(6461), pp.76–82. doi:10.1126/science.aax2747.

A sequence of prebiotically plausible reactions starting from inorganics and D-ribose leads to nucleoside precursors. A catalytic cycle mediated by iron(II/III) and boric acid converts these to nucleosides. Wet-dry cycling with phosphate minerals (lüneburgite) leads to nucleotides with 5’ regioselectivity.

Ruiz-Bermejo, M., Zorzano, M. P. and Osuna-Esteban, S. (2013). ‘Simple Organics and Biomonomers Identified in HCN Polymers: An Overview’. Life, 3(3), pp.421-448. doi:10.3390/life3030421.

Hydrogen cyanide (HCN) is found throughout the universe and can form a wide range of molecules in prebiotically relevant conditions. The nucleobase adenine is a pentamer of HCN, formed by UV light in water. Among the purines, adenine has the highest resonance energy per pi electron, making it preferable for incorporation into RNA due to (Michaelian, 2017). Other purines, pyrimidines and urea also form with NCO$ ^-$ and C$ _2$N$ _2$. Acid hydrolysis of the mixture yields hydantoins (cylic carbamyl amino acids) and pteridines (similar to cofactors like riboflavins and pterins). Radiolysis of the mixture yields important carboxylic acids (oxalic acid, malonic acid, fumaric acid, succinic acid, citric acid, adipic acid, etc.) and carbonyls (formaldehyde, acetaldehyde, methylglioxal, acetone).

SYNTHESIS OF MACROMOLECULES

Feasible reactions to produce macromolecules such as RNA, DNA, proteins and carbohydrates with enantioselectivity and regioselectivity where relevant.

Polypeptides (prebiotic protein syntheses)

Chen, F. and Yang D. (2006). ‘Condensation of amino acids to form peptides in aqueous solution induced by the oxidation of sulfur(IV): An oxidative model for prebiotic peptide formation’, Origins of Life and Evolution of Biospheres, 37(1), pp. 47-54. doi:10.1007/s11084-006-9012-y

Carboxylic acids and amines undergo condensation reactions to form polypeptides in the presence of sulfur(IV) (e.g. SO$_2$, NaHSO$_3$) and oxidant (K$_3$[Fe(CN)$_6$]). This functionality also applies to amino acid polymerisation, including ligation to small polypeptides. The yield is about 2% relative to the amount of sulfur used.

Singh, J. et al. (2022). ‘Prebiotic Catalytic Peptide Ligation Yields Proteinogenic Peptides by Intramolecular Amide Catalyzed Hydrolysis Facilitating Regioselective Lysine Ligation in Neutral Water’. Journal of the American Chemical Society, 144(23), pp.10151–10155. doi:10.1021/jacs.2c03486.

Primary thiols catalyse the ligation of amino acids, amides, and peptides with amidonitriles in water. At pH 7, this reaction is regiospecific to the correct functional groups, leaving the unprotected side chains unaltered in all amino acids.

Greenwald, J., Friedmann, M.P. and Riek, R. (2016). ‘Amyloid Aggregates Arise from Amino Acid Condensations under Prebiotic Conditions’, Angewandte Chemie International Edition, 55(38), pp.11609–11613. doi:10.1002/anie.201605321.

Some amino acids (alanine, valine, aspartic acid, glycine) can polymerise into peptides mediated by carbonyl sulfide in water, and assemble into ordered amyloid fibres with a cross-beta-sheet quaternary structure. These amyloids are highly resistant to hydrolysis. A range of pHs and temperatures allow amyloid formation.

Stolar, T. et al. (2021). ‘Mechanochemical Prebiotic Peptide Bond Formation’, Angewandte Chemie, 60(23), pp.12727–12731. doi:10.1002/anie.202100806.

Mechanical impacts from meteorites and geochemical phenomena can drive mechanochemistry. Under ball milling, solid glycine can polymerise with and without water present. The length of the polypeptides increased with temperature, as well as cyclic dimer formation. In the presence of TiO$_2$, the dimers served as a source of glycine allowing formation of peptides up to 10-mers.

Sumie, Y. et al. (2023). ‘Boron-assisted abiotic polypeptide synthesis’. Communications Chemistry, 6, p.89. doi:10.1038/s42004-023-00885-7.

Up to 39-mers of polyglycine were formed by simple heating of glycine, catalysed by aqueous boric acid at pH 6 - 8 and temperatures of 90 - 130 C, with negligible side reactions. Borate is known to be present in the mineral tourmaline on the prebiotic earth, and can also help to form RNA by (Kim & Benner, 2021).

Holden, D. T., Morato, N. M. and Cooks, R. G. (2022). ‘Aqueous microdroplets enable abiotic synthesis and chain extension of unique peptide isomers from free amino acids’. Proceedings of the National Academy of Sciences, 119(42), e2212642119. doi:10.1073/pnas.2212642119.

Spray ejection of micron-sized droplets of liquid neutral water containing free amino acids (glycine, L-alanine) results in peptide formation (up to 6-mer) at the air-water interface. The surface of the droplets acts as a dry environment that shifts the equilibrium of the reaction towards condensation. This occurs at ambient conditions without any other reagents, catalysts, or radiation.

Polynucleotides (prebiotic RNA syntheses)

Deamer, D. (2021). ‘Where Did Life Begin? Testing Ideas in Prebiotic Analogue Conditions’. Life, 11(2), p.134. doi:10.3390/life11020134.

A variety of experiments studying reactions and processes in ‘prebiotic soups’ (mixtures of glycine, L-alanine, L-aspartic acid, L-valine, nucleobases, sodium phosphate, glycerol, myristic acid, dodecanoic acid…) in geologically active regions. Observations include adsorption of molecules on smectite-kaolinite clay particle suspensions, soapy froth formation, polymerisation and encapsulation of polymers within lipid particles, amphiphile self-assembly and enzyme-free nucleotide polymerisation with wet-dry cycling.

Olasagasti, F. and Rajamani, S. (2019). ‘Lipid-Assisted Polymerization of Nucleotides’, Life, 9(4), pp. 83–83. doi:10.3390/life9040083.

A mixture of lipids and nucleotides led to polynucleotide formation under wet-dry cycling.

Jerome, C.A. et al. (2022). ‘Catalytic Synthesis of Polyribonucleic Acid on Prebiotic Rock Glasses’. Astrobiology, 22(6), pp.629–636. doi:10.1089/ast.2022.0027.

Nucleoside triphosphates form polynucleotides, with a strong bias towards 3’-5’ linkage regioselectivity, in the presence of natural silicate glasses. It is already known that 2’-5’ linkages in ssRNA hydrolyse at pH 7 around 3 times faster than 3’-5’ linkages, providing a potential recycling mechanism that favours higher 3’-5’ content.

Jheeta, S. and Joshi, P. (2014). ‘Prebiotic RNA Synthesis by Montmorillonite Catalysis’. Life, 4(3), pp.318–330. doi:10.3390/life4030318.

Nucleotides form polynucleotides in the presence of montmorillonite clay and simple salts (e.g. NaCl).

Rout, S. K. et al. (2025). ‘Amino acids catalyse RNA formation under ambient alkaline conditions’. Nature Communications, 16, 5193(2025). doi:10.1038/s41467-025-60359-3.

Starting with ribonucleoside 2′,3′-cyclic phosphates (an easily formed prebiotic molecule), addition of aliphatic hydrophobic amino acids (such as L-valine, L-leucine, L-isoleucine, L-alanine, glycine) at pH 9-10 at room temperature, resulted in formation of up to 7-mer oligo-RNA. The cyclic nucleoside phosphate is more stable than 5’-nucleotides, making it more prebiotically plausible. The NMR data showed a bias towards 3’-5’ linkages (about 60%), the correct one for optimally functional/extant RNA. Polymerisation occurred in the dried state. No [Mg]$ ^{2+} $ ions were required. These conditions could be found in cooler volcanic environments, whose alkalinity is also correlated with phosphate enrichment, allowing wet-dry cycling to occur. Thermal gradients (heat fluxes) also selectively enrich Val/Leu/Ile as required. This completes the ‘missing link’ in the coevolution of RNA and proteins, as it shows that amino acids can promote RNA formation while RNA can promote protein formation.

Dass, A. V. et al. (2022). ‘RNA Oligomerisation without Added Catalyst from 2′,3′-Cyclic Nucleotides by Drying at Air-Water Interfaces’. ChemSystemsChem, 5(1). doi:10.1002/syst.202200026.

Starting with 2’,3’-cGMP (cyclic guanine nucleotides) in warm water, oligomerisation up to 10-mer RNA was observed at the air-water interface within 24 hours on drying, with no catalyst required. This occurred in a variety of conditions: alkaline (pH 7 - 12), hot water (40 - 80 C: optimal temperature 40 C), with salt ions (20 mM Na$ ^+$, 40 mM K$ ^+$, slightly enhanced by K$ ^+$). Up to 15-mers were found in trace amounts. The RNA linkages were in equal proportions 2’-5’ and 3’-5’. This serves as yet another model reaction of wet-dry cycling processes forming RNA, this time at the water’s surface rather than at deep sea environments.

Lipids (prebiotic membrane syntheses)

Santos, T. and Futerman, A. (2022). ‘The fats of the matter: Lipids in prebiotic chemistry and in origin of life studies’. Progress in Lipid Research, 92(101253), doi:10.1016/j.plipres.2023.101253.

A review of prebiotic lipid chemistry, and their capacity to enable compartmentalisation of chemically evolving systems. Lipids can be formed from ammonium salts of fatty acids and glycerol on hot kaolinite clay by esterification. Other alcohols can be used, including derivatives of amino acids with -OH side chains, found in extant lipids.

Fiore, M. et al. (2022). ‘Synthesis of phospholipids under plausible prebiotic conditions and analogies with phospholipid biochemistry for origin of life studies’. Astrobiology, 22(5), pp.598-627. doi:10.1089/ast.2021.0059.

A review summarising modern prebiotic lipid synthesis and reactions, including their phosphorylation, esterification and ligation with choline. Lipids are formed by Fischer-Tropsch-type reactions of simple gaseous precursors (CO, H$_2$O, H$_2$, CO$_2$), which in turn form from the thermal decomposition of formic acid and oxalic acid. This can form a diverse range of n-alkanes, straight/branched alkenes, fatty alcohols, fatty acids and fatty formyl esters. Triacyanocuprate and tetracyanocuprate can act as electron donors. Glycerol can be phosphorylated in a deep eutectic solvent of urea and choline chloride. Glycerol can form mono-, di- and triglycerides with hot fatty acids and various mineral catalysts. Since bilayer stability against Mg$ ^{2+}$ ions increases with phospholipid-to-lipid ratio, and Lipids form micelles while phospholipids form bilayers, this provides a selection mechanism for phospholipid membranes.

Bonfio, C. et al. (2019). ‘Length-Selective Synthesis of Acylglycerol-Phosphates through Energy-Dissipative Cycling’, Journal of the American Chemical Society, 141(9), pp.3934-3939. doi:10.1021/jacs.8b12331.

Medium-to-long chain carboxylic acids and glycerol phosphates are both prebiotically relevant (made by photocatalysis/Fischer-Tropsch reactions to generate a complex mixture, then phosphorylation). Activation of the carboxylic acids with imidazole followed by reaction of glycerol 2-phosphate led to phospholipid formation. The medium-chain amphiphiles were found to self-assemble into vesicles stable across a wide range of conditions and capable of retaining nucleotides and oligonucleotides inside. Hydrolysis after acylation was higher in short-chain amphiliphiles, providing a recycling mechanism for selection of longer chains.

REACTIONS OF MACROMOLECULES

Studies on the reactions that macromolecules can undergo in prebiotically feasible conditions, typically leading to the development of self-replicators and other autocatalytic activity.

Chemical activation of RNAs and polypeptides

Song, E.Y. et al. (2020) ‘Prebiotically plausible RNA activation compatible with ribozyme‐catalyzed ligation’, Angewandte Chemie International Edition, 60(6), pp. 2952–2957. doi:10.1002/anie.202010918.

Diamidophosphate (DAP) reacts with nucleotides to protect the 2’ and 3’ -OH groups, activating them for regioselective self-replication with a ribozyme.

RNA self-replicators (chemical evolutionary dynamics of ribozymes)

Voytek, S.B. and Joyce, G.F. (2009) ‘Niche partitioning in the coevolution of 2 distinct RNA enzymes’, Proceedings of the National Academy of Sciences, 106(19), pp. 7780–7785. doi:10.1073/pnas.0903397106.

Ribozymes with two different sequence identities were used for self-replication with 5 alternative substrates. The subsequent mutations in each ribozyme led to enhanced catalysis of each ribozyme for a different substrate, mirroring Darwinian evolution at the chemical level.

Lincoln, T.A. and Joyce, G.F. (2009) ‘Self-sustained replication of an RNA enzyme’, Science, 323(5918), pp. 1229–1232. doi:10.1126/science.1167856.

An RNA ligase was prepared that self-replicates given only small oligo template RNAs. Amplification occurs with a doubling time of about 1 hour and can be continued indefinitely. The RNA ligase was ~100 nt long but only small parts of its sequence were critical to catalytic activity.

Vaidya, N. et al. (2012) ‘Spontaneous network formation among cooperative RNA Replicators’, Nature, 491(7422), pp. 72–77. doi:10.1038/nature11549.

Mixtures of RNA fragments that self-assemble into self-replicating ribozymes spontaneously form cooperative catalytic cycles and networks. When such cooperative networks are competed directly against selfish autocatalytic cycles, the former grow faster, indicating an intrinsic ability of RNA populations to evolve greater complexity through cooperation.

Mizuuchi, R., Furubayashi, T. and Ichihashi, N. (2022). ‘Evolutionary transition from a single RNA replicator to a multiple replicator network’. Nature Communications, 13(1), p.1460. doi:10.1038/s41467-022-29113-x.

A ribozyme is studied for its long-term evolution capacity. The RNA diversifies into multiple coexisting ‘host’ and ‘parasite’ ’lineages’, whose frequencies in the population initially fluctuate and gradually stabilise. The final population, comprising five RNA lineages, forms a replicator network with diverse interactions, including cooperation to help the replication of all other members.

Engelhart, A., Powner, M. & Szostak, J. (2013). ‘Functional RNAs exhibit tolerance for non-heritable 2′–5′ versus 3′–5′ backbone heterogeneity’. Nature Chem 5, pp. 390–394. doi:10.1038/nchem.1623.

Polyribonucleotides do not require 3’-5’ linkage regioselectivity to have ribozyme functionality. The 2’-5’/3’-5’ heterogeneity lowers the melting point of the RNA duplexes, making thermal strand separation easier and permitting continued replication.

Le Vay, K. et al. (2021). ‘Enhanced Ribozyme-Catalyzed Recombination and Oligonucleotide Assembly in Peptide-RNA Condensates’. Angew Chem Int Ed Engl. 6;60(50): pp. 26096-26104. doi:10.1002/anie.202109267.

Ribozyme activity is greatly enhanced by charge-mediated phase separation with poly-L-lysine. This shifts the reaction equilibrium from cleavage in solution to ligation in peptide-RNA co-aggregates and coacervates. This compartmentalization enables robust isothermal RNA assembly over a broad range of conditions.

Papastavrou, N., Horning, D. P. & Joyce, G. F. (2024). ‘RNA-catalyzed evolution of catalytic RNA’. Proc. Natl. Acad. Sci. U.S.A. 121(11) e2321592121, doi:10.1073/pnas.2321592121.

An RNA enzyme with RNA polymerase activity was used to replicate and evolve an RNA enzyme with RNA-cleavage activity. The fidelity of the polymerase is sufficient to maintain heritable information over the course of evolution, with a succession of variants of the RNA-cleaving RNA enzyme arising that have progressively increasing fitness.

Polypeptide self-replicators

Maury, C.P.J. (2018). ‘Amyloid and the origin of life: self-replicating catalytic amyloids as prebiotic informational and protometabolic entities’. Cell. Mol. Life Sci. 75, pp. 1499–1507. doi:10.1007/s00018-018-2797-9.

The self-assembly of short peptides into β-sheet amyloids leads to structural stability. These amyloids can also have catalytic properties allowing for self-replication and information transfer (by scaffolding as a template for replication). The interactions between amyloid, RNA, and protein are both complex and cooperative.

Functional small RNAs

Mizuuchi, R. and Ichihashi, N. (2023). ‘Minimal RNA self-reproduction discovered from a random pool of oligomers’. Chemical Science, 14, pp.7656-7664. doi:10.1039/D3SC01940C.

Out of a random pool of 20-mer oligonucleotides (total of 10^12 possible sequences), in the presence of phosphate (which formed a bond at the 2’ and 3’ positions) and 100 mM MgCl$_2$, a sequence was found that permitted self-replication by template-directed ligation. The found RNA has a stem-loop secondary structure and contains a mixture of 2’-5’ and 3’-5’ linkages.

Mariani, A. and Sutherland, J. D. (2017). ‘Non-Enzymatic RNA Backbone Proofreading through Energy-Dissipative Recycling, Angew. Chem. Int. Ed., 56, pp.6563–6566. doi:10.1002/anie.201703169.

Starting with a 13-mer oligo-RNA duplex with a 2’-5’ linkage in the middle of one strand (and the rest 3’-5’ linked), mildly alkaline conditions (pH 9.25, 21 C) led to the isomerisation of the 2’-5’ linkage into a 6-mer oligo-RNA 2’-3’ cylic phosphate. Hydrolysis with $ [Mg^{2+}] \leq $ 40 mM opened the cyclic phosphate to give the 6-mer and 7-mer oligo-RNAs, with the 6-mer having either a terminal nucleoside 2- or 3-phosphate. From there, acetylation with N-acetylimidazole and phosphorlyation with N-cyanoimidazole led to regioselective ligation to form 3’-5’ RNA from the 6/7-mers. Only 1% degradation of the all-3’-5’-linked RNA was observed after 24 days. With multiple rounds of this energy-dissipative ‘recycling’ of RNA, all-3’-5’-linked RNA can emerge from a backbone-heterogeneous (2’-5’ and 3’-5’) mixture.

Gianni, E. et al. (2026). ‘A small polymerase ribozyme that can synthesize itself and its complementary strand’. Science, 0, eadt2760, doi:10.1126/science.adt2760.

A variety of polymerase ribozymes (named QT45 and QT51, of length 45 nt and 51 nt respectively) are discovered from random sequence pools, subject to rounds of selection for replication activity. QT45 catalyses general RNA-templated RNA synthesis in either its monomeric or dimeric form, capable of synthesising using nucleotide triphosphates as monomers, dimers or trimers, synthesising both its the complementary strand and its own strand. Defined triplets and one hexanucleotide are needed to complete the synthesis of its self strand. QT51. The 3’-5’ phosphodiester bond was fully regiospecific. QT45 attained 94.1% per-nucleotide fidelity with yields of ~0.2% in 72 days in mildly alkaline (pH 9) eutectic ice at -7 C. The discovery of polymerase activity in a small RNA motif suggests that polymerase ribozymes are more abundant in RNA sequence space than previously thought.

Functional small peptides

Timm, J. et al. (2023). ‘Design of a minimal di-nickel hydrogenase peptide’. Science Advances, 9(10), doi:10.1126/sciadv.abq1990.

A 13-amino acid oligopeptide was designed to exhibit hydrogenase activity. The protein is a redox metalloenzyme with a di-nickel cluster active site that converts solvated protons into hydrogen (2 H$^+$ + 2 e$^- \rightarrow $ H$_2$). This structure is much simpler than extant hydrogenases. The peptide was stable at pH 5.5 - 10 and at temperatures 20 $ ^{\circ}$C - 80 $ ^{\circ}$C. Simple Archean eon life is expected to be dependent on hydrogen gas for metabolism at these conditions.

Rufo, C. M. et al. (2014). ‘Short peptides self-assemble to produce catalytic amyloids’. Nature Chemistry, 6, pp.303-309. doi:10.1038/nchem.1894.

Several 7-mer oligopeptides were found to form a β-strands, in particular those with alternating hydrophilic-hydrophobic primary structure (e.g. LKLKLKL, L: leucine, K: lysine, with many others). When positions 2 and 4 were replaced with histidine (H), the β-strand structure was retained, and some of these peptides self-assembled into an amyloid fibril structure with β-pleated sheets, and exhibited catalytic activity for the hydrolysis of p-nitrophenyl acetate (pNPA) in the presence of Zn(II) ions. These peptides work by binding Zn(II) ions with the histidine residue side chains, a motif found in many extant hydrolase enzymes. The sequence IHIHIQI was especially effective, with $ \frac{k_{cat}}{K_m} = 360 \ M^{-1} \ s^{-1} $ (quite high) at pH 10.3.

RNA-peptide interactions

Xu, D. and Wang, D. (2021). ‘Protein-free ribosomal RNA scaffolds can assemble poly-lysine oligos from charged tRNA fragments’, Biochemical and Biophysical Research Communications, 544, pp.81-85. doi:10.1016/j.bbrc.2021.01.036.

Two RNA sequences, named ptc1a and ptc1b, both around 110 nt long, whose sequences are derived from sections of extant rRNA, are synthesised. When added to tRNA charged with lysine, the RNA dimer acted as a peptide ligase ribozyme, synthesising a oligopeptide 9-mer of polylysine.

Su, M. et al. (2023). ‘Triplet-Encoded Prebiotic RNA Aminoacylation’, J. Am. Chem. Soc., 145(29), pp.15971-15980. doi:10.1021/jacs.3c03931.

Prebiotically plausible syntheses of aminoacyl phosphate anhydrides are already known. Two strands of complementary oligo-RNAs were prepared, one of which is 5’-terminated with an aminoacyl phosphate. These short RNAs anneal together form a duplex, wherein the 3’ and 5’ ends are in close proximity. Transfer of the aminoacyl group occurs between the ends of the RNA with high enantioselectivity and chemoselectivity based on the amino acid, forming self-aminoacylated RNA. For aminoacyl-transfer from a mixed anhydride donor strand, the chemoselectivity and stereoselectivity of aminoacylation depend on the terminal three base pairs of the stem. This provides an origin of extant aminoacyl-tRNA by duplication and the ‘secondary genetic code’, which is necessary for genetically encoded protein synthesis. This mechanism is also compatible with the hypothesis that tRNA arose by duplication of RNA, as it has an antisymmetric cloverleaf structure.

Singh, J. et al. (2025). ‘Thioester-mediated RNA aminoacylation and peptidyl-RNA synthesis in water’, Nature 644, pp.933–944. doi:10.1038/s41586-025-09388-y.

dsRNA dinucleotides react at their 3’-OH termini with aminoacyl-thiols (prebiotically relevant amino acid derivatives with many known syntheses) to form aminoacyl-RNA at pH 6-7 and room temperature in water. 14 different proteinogenic amino were tested and showed strong selectivity and especially high efficiency for arginine via side-chain catalysis. Additionally, a simple change in conditions (thioester to thioacid activation) leads to peptide formation rather than aaRNA formation, both of which are key reactions in the extant ribosome.

Turk, Chumachenko & Yarus (2010). ‘Multiple translational products from a five-nucleotide ribozyme’. PNAS, 107(10), pp.4585-4589. doi:10.1073/pnas.0912895107.

The 5-nt ribozyme GUGGC binds the first three bases of its partially complementary 4-nt RNA GCCU, and then reacts with PheAMP (phenylalanine adenosine monophosphate) to form a bond between Phe and the 2’-OH of the complementary strand. Furthermore, additional Phe residues can react, forming up to RNA-Phe$_3$. The GUGGC/GCCU complex has a simple 3D structure and does not require divalent cations for folding with Mg$^{2+}$ and K$^{+}$ being slightly stimulatory and Na$^{+}$ being slightly inhibitory. This represents prebiotically relevant aminoacylation and peptide bond formation with a single short ribozyme containing a 3-nt active site.

Shi, L. et al. (2026). ‘From peptides to DNA: All required steps can be catalyzed’. Proceedings of the National Academy of Sciences, 123(9), e2534387123, doi:10.1073/pnas.2534387123.

Three hypothetical processes required to convert information stored in proteins into genetic information in RNA or DNA: 1) step-wise degradation of peptides by an amino peptidase, sequentially releasing amino acids, 2) matching the identity of released amino acids to codons by aptazymes (RNA adapters), and 3) ligating codon triplets into longer RNAs that can be reverse-transcribed into DNA, have been experimentally demonstrated.

PROTOCELL MODELS

Studies on the behaviour of lipid vesicles to test for their validity as the first chemical structures used to isolate cells from their environment.

Kurihara, K. et al. (2011). ‘Self-reproduction of supramolecular giant vesicles combined with the amplification of encapsulated DNA’. Nature Chemistry, 3(10), pp.775–781. doi:10.1038/nchem.1127.

A protocell model membrane is made from cationic surfactant molecules. When DNA and DNA polymerase is contained within, the resulting protocells undergo division and continued self-replication of the DNA in each protocell.

Damer, B. and Deamer, D. (2020). ‘The Hot Spring Hypothesis for an Origin of Life’. Astrobiology, 20(4). doi:10.1089/ast.2019.2045.

Experiments show lipid-encapsulated polymers can be synthesized by cycles of hydration and dehydration to form protocells. Protocells cycling through wet, dry, and moist phases will subject polymers to combinatorial selection and create structural and catalytic functions out of initially random sequences, including structural stabilisation, pore formation, and primitive metabolic activity.

Kwiatkowski, W. et al. (2021). ‘Prebiotic Peptide Synthesis and Spontaneous Amyloid Formation Inside a Proto‐Cellular Compartment’. Angewandte Chemie, 60(10), pp.5561–5568. doi:10.1002/anie.202015352.

Chemical gradients in simple mixtures of activated amino acids and fatty acids form of amyloid-like peptide fibres that get contained within lipid vesicles. The lipid vesicles act both as a semi-permeable membrane, allowing selective passage of activated amino acids, while blocking diffusion of amyloidogenic peptides that form spontaneously within.

Cho, C.J. et al. (2025). ‘Protocells by spontaneous reaction of cysteine with short-chain thioesters’. Nat. Chem. 17, pp.148-155 (2025). doi:10.1038/s41557-024-01666-y

Cysteine reacts with a 8-carbon thioester of choline in water at pH 8 to form a two-tailed amphiphile (a diacylcysteine), which forms membrane bilayer vesicles in the presence of a silica substrate. This occurred at concentrations far below the critical micelle concentration (CMC). Peptides with an N-terminal cysteine residue were also capable of reaction and vesicle formation, with faster kinetics and lower CMC. The vesicles were stable in divalent cation solution up to 2 mM Mg(II) and 10 mM Ca(II). Aminoacylating ribozymes (using biotinyl-Tyr(Me)-oxazolone as their substrate) encapsulated within the vesicles retained their functionality.

Cohen, Z. R. et al. (2024). ‘Stabilization of Prebiotic Vesicles by Peptides Depends on Sequence and Chirality: A Mechanism for Selection of Protocell-Associated Peptides’. Langmuir, 40(17), pp.8971-8980. doi:10.1021/acs.langmuir.4c00150.

Lipid bilayers encapsulating homochiral Leu-Leu, Leu-Gly and Leu-Ser dipeptides grew to larger sizes and were more stable to salt-induced flocculation than those encapsulating heterochiral dipeptides, which led to leaky vesicles. This provides a selection mechanism for homochirality in vesicle-bound dipeptides.

Katla, S.K., Lin, C. and Pérez-Mercader, J. (2025). ‘Self-reproduction as an autonomous process of growth and reorganization in fully abiotic, artificial and synthetic cells’. Proc. Natl. Acad. Sci. U.S.A., 122(22), e2412514122, doi:10.1073/pnas.2412514122.

Abiotic vesicles were created in water from a mixture of 1) PEG terminated with a hydrophobic chain transfer agent (an amphiphile), 2) monomers such as acrylonitrile (photopolymerisable to form block copolymers with PEG), 3) zinc tetraphenylporphyrin (a photocatalyst), in the presence of visible light. This represents a far-from-equilibrium system with robust dynamics. The polymers assemble into micelles enclosing the reactive monomers. These vesicles evolve in time through feedback between polymerisation, degradation and chemiosmotic gradients, resulting in self-reproduction. As vesicles grow, they polymerise their contents, leading to their partial release and their reproduction into new vesicles, exhibiting a loose form of heritable variation, competitive exclusion and Darwinian selection.

SYNTHETIC BIOLOGY AND MOLECULAR BIOLOGY

Experiments on cells and macromolecules derived from extant biology. These are usually not prebiotically relevant, but can inform us about the minimum viable complexity of LUCA and later life.

Pelletier, J.F. et al. (2021). ‘Genetic requirements for cell division in a genomically minimal cell’. Cell, 184(9), pp.2430-2440. doi:10.1016/j.cell.2021.03.008.

Building on Craig Venter’s 2010 work creating synthetic cells derived from the bacteria Mycoplasma mycoides, a new type of synthetic minimal cell (JCVI-syn3.0) is created. These cells contain ~500 genes and undergo normal cell division with spherical cell shape.

Sahakyan, H. et al. (2025). ‘In silico evolution of globular protein folds from random sequences’, Proc. Natl. Acad. Sci. U.S.A. 122 (27) e2509015122. doi:10.1073/pnas.2509015122.

Stable, globular protein folds can evolve from random amino acid sequences with ease, resulting from selection acting on a realistic number of amino acid replacements. About half of the in silico evolved proteins resemble simple folds found in nature, whereas the rest are unique.

Kellock et al. (2016). ‘Peptides Comprised of Alternating L- and D- Amino Acids Inhibit Amyloidogenesis in Three Distinct Amyloid Systems Independent of Sequence’, J Mol Biol., 428(11), pp.2317-2328. doi:10.1016/j.jmb.2016.03.013.

Oligopeptides need not be homochiral to have biochemical functionality. The paper describes a 23-mer oligo whose primary structure consisted of alternating L and D amino acid residues (syndiotacticity), forming an α-hairpin secondary structure and a function for inhibiting amyloid aggregation. Scrambling the amino acid content (but retaining the syndiotacticity) did not affect this function. A similar oligopeptide is the antibiotic gramicidin, another syndiotactic 15-mer that forms a complete alpha helix, and also forms a dimeric quaternary structure in micelles. It acts as an ionophore, another prebiotically relevant function.

Wang, M., Hoegler, K. and Hecht, M. (2019). ‘Unevolved De Novo Proteins Have Innate Tendencies to Bind Transition Metals’, Life, 9(8). doi:10.3390/life9010008.

De novo proteins (never before seen in nature) with no evolutionary history can still bind metal ions such as Zn$^{2+}$, Co$^{2+}$, Cu$^{2+}$ and Ni$^{2+}$. This shows that metal binding is an intrinsic property of polypeptides, and may have been important in the origin of life in developing metalloenzymes.

Raanan, H. et al. (2020). ‘Small protein folds at the root of an ancient metabolic network’, Proc. Natl. Acad. Sci. U.S.A., 117(13) pp.7193-7199. doi:10.1073/pnas.1914982117.

Short proteins (less than 70 amino acids) with simple folds (mainly α-helices) can bind cofactors such as iron-sulfur clusters, cobalamin and flavins, and exhibit redox activity. These proteins are found in all three domains of life, indicating their ancient origin. This suggests that short proteins with simple folds could have played a key role in the emergence of early metabolic networks as simple oxidoreductases for electron transfer reactions.

Poole, Penny & Sjöberg (2001). ‘Confounded cytosine! Tinkering and the evolution of DNA’, Nature Reviews Molecular Cell Biology, 2, pp.147-151. doi:10.1038/35052091.

Sometime after abiogenesis, thymine (T) replaced uracil (U) as RNA transitioned to DNA, thus solving a short-term problem for storing genetic information (point mutation of C to U via deamination). Any engineer would have replaced C, but evolution is a tinkerer, not an engineer. By keeping C and replacing U, the problem was never eliminated, returning once again when DNA methylation was harnessed for epigenetics, as 5-methylcytosine deaminates to T instead.

Restrepo Sierra, A. M. et al. (2026). ‘A synthetic cell with integrated DNA self-replication and lipid biosynthesis’. Nature Communications. doi:10.1038/s41467-026-69531-9.

…