Theoretical and experimental work has been conducted about possible prebiotic syntheses of cysteine. Acti-

The Strecker reaction of aldehydes with ammonia and hydrogen cyanide first leads to α-aminonitriles, which are then hydrolyzed to α-amino acids. However, before reacting with water, these aminonitriles can be trapped by aminothiols, such as cysteine or homocysteine, to give 5- or 6-membered ring heterocycles, which in turn are hydrolyzed to dipeptides. We propose that this two-step process enabled the formation of thiol-containing dipeptides in the primitive ocean. These small peptides are able to promote the formation of other peptide bonds and of heterocyclic molecules. Theoretical calculations support our experimental results. They predict that α-aminonitriles should be more reactive than other nitriles, and that imidazoles should be formed from transiently formed amidinonitriles. Overall, this set of reactions delineates a possible early stage of the development of organic chemistry, hence of life, on Earth dominated by nitriles and thiol-rich peptides (TRP).

Life developed on Earth probably about 3.8 billion years ago, on a planet that was already largely covered by oceans and where the atmosphere was very humid. The reactions, which may have led to the formation of the first polymers, particularly to the first peptides and nucleic acids, must have been compatible with these conditions. This is the case of the reaction of nitriles with aminothiols, such as cysteine and homocysteine. Since aminonitriles are the probable precursors of amino acids, this condensation reaction has been able to rapidly yield dipeptides, tripeptides, oligomers and even true polymers, each containing thiol functions. These thiol-rich peptides (TRP's) would then have assumed the various catalytic roles that the peptides containing cysteine residues play today. They allowed a rapid bloom of life in the primitive ocean. In this scenario, RNA's are not the first polymers, but have been synthesized, like DNA's, thanks to the catalytic properties of thiols in a mostly TRP world. In this world, due to its ability to form a thiolactone, homocysteine may have played the leading role in enabling the previously formed oligomers to be stappled together, thus accelerating the formation of long peptide chains.

Homocysteine is a non-proteinogenic sulfur-containing amino acid. Like cysteine, it can form disulfide bridges and complex metallic cations. It is also closely related to methionine, the first amino acid in the synthesis of all contemporary proteins. Furthermore, its cyclized form, a five-membered ring thiolactone, is stable in acidic and neutral water. Here, we demonstrate that this thiolactone may have been formed in the primitive ocean directly from the Strecker precursor of homocysteine, an aminonitrile. Even though it is poorly reactive, this thiolactone may be open by some amines, yielding amides which, in turn, could be the precursors of longer peptides.