As a followup to Why 20 amino acids instead of 64? and What is the smallest number of amino acids required for life?, I am trying to understand the prehistory of amino acids in cells.

All living cells now use 20 (or 22 if you include selenocysteine and pyrrolysine) amino acids. Perhaps in the past there were fewer amino acids used in cells, especially if the theory of a 2-nucleotide codon is to be believed.

Is there any theory as to the order in which these amino acids were incorporated into cells?

This paper (by Akanuma, Kigawi, Yokoyama, "Combinatorial mutagenesis to restrict amino acid usage in an enzyme to a reduced set", http://www.pnas.org/content/99/21/13549.full) suggest that the full complement of 20 amino acids are not strictly necessary by reducing the number of amino acids in a certain enzyme to 13, yet still retain functionality.

  • $\begingroup$ the quippiest answer might be ZERO since its been shown that RNA can reproduce itself, at least in theory and probably at some point in biological history as well. $\endgroup$
    – shigeta
    Commented Jan 16, 2012 at 17:33
  • 1
    $\begingroup$ If there ever was a 2 bp code - i think it would have not produced proteins as we know them - there simply isn't enough variety to make a structure as we know it. Proteins would have done something else at that point. Not sure what to make of the question at that point. Maybe it was all keratin... $\endgroup$
    – shigeta
    Commented Jan 17, 2012 at 1:45
  • $\begingroup$ @shigeta: check the paper I've added as a reference $\endgroup$
    – John Smith
    Commented Jan 18, 2012 at 1:41

1 Answer 1


Edward N. Trifonov is a professor at the Institute of evolution at University of Haifa, Israel. One of the main research topics in his group is the reconstruction of the origins of life. In one of his papers, The triplet Code From First Principles, he proposes the the chronological appearance of the 20 amino acids. There are a lot of hypothesis and testable predictions, among which is the theory of S. Miller that some biological compounds have been abiotically produced in the atmosphere. This prediction has been confirmed by imitation experiments, yielding ten different amino acids, as glycine and alanine being the prevalent ones.

The author tries to predict the chronological order in which the 20 amino acids appeared by taking into account various criteria (studies, hypothesis and viewpoints) that propose some order, as each one of them contributes to the consensus order to a certain extent. This contribution is described in the form of rank. He compiles a list of 60 rankings of 60 different criteria to derive the consensus order. Some of the criteria are:

N1’. Criteria based on various evaluations of complexity of amino-acids. N1, N34, N35, N37 – as in (5), and N44 (6). G, A, S, P, V, C, D, T, N, L, K, I, E, (MQ), H, R, F, Y, W

N2’. Criteria based on evolution of amino-acyl-tRNA synthetases. N2, N7 (5). (AG), (DFHKNPST), (CEILMQRV), (WY)

N4’. Criteria based on amino-acid compositions of various sets of presumably ancient proteins. N4, N26 (5), N41 (7), N45 (8), N47 (9), N48 (10), and N58 (11).

The first ten (earliest) consensus amino acids are the same as in Millers’s experiments. The amino acids appeared last have their respective codons borrowed from already existing ones. The derived consensus order is G, A, D, V, P, S, E, (L, T), R, (I, Q, N), H, K, C, F, Y, M, W.

The author also proposes a reconstruction of the chronological order in which the 64 codons had appeared.

  • $\begingroup$ This is an interesting paper. I've not picked through it in detail, but it seems to assume we started with the triplet code immediately and did not pass through a doublet code as other people have proposed ( imb-jena.de/~sweta/genetic_code_and_evolution/evolution.html ) $\endgroup$
    – John Smith
    Commented Jan 20, 2012 at 14:44
  • $\begingroup$ Yeah, the assumption is that the earliest amino acids (ala ang gly) were encoded by triplet codons. Perhaps you can look at Trifonof's eariler literature to see whether he discusses this. $\endgroup$ Commented Jan 20, 2012 at 18:04

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .