THE genetic code used by all life today may have evolved from two simpler codes in the distant past. If so, traces of ancient genes using these earlier codes might be lurking in the plethora of apparently useless 鈥渏unk鈥� DNA that litters almost all genomes.
The genetic language is made up of three-letter 鈥渨ords鈥�, known as codons, which each specify a particular protein building block, or amino acid. Since each letter in the codon can be any of four different bases, there are 64 possible combinations (see Table). But the code specifies only 20 amino acids, and this has led geneticists to puzzle over how this system evolved.
The curious pattern of redundancies in the code makes sense if the modern triplet code evolved from a codon in which only the middle base specified an amino acid, with the two flanking bases acting merely as spacers, says Jean van den Elsen, an evolutionary biologist at the University of Bath, UK. He and his colleagues suggest that this earliest code may therefore have involved just four amino acids, each determined by which of the four bases was in the middle of the codon.
Advertisement
The code could then have expanded to include sometimes the first and middle bases and sometimes the middle and third bases, giving primordial life forms a bigger tool kit of amino acids to assemble more diverse proteins. These two codes finally adapted to allow still more amino acids by recognising the remaining base in the codon, and the triplet code was born (Journal of Molecular Evolution, DOI: 10.1007/s00239-004-0224-3).
This scenario explains much about the genetic code. 鈥淚t鈥檚 the only way to get a code that looks very similar to our genetic code,鈥� says van den Elsen. For instance, the third base is still irrelevant in specifying some amino acids, such as valine (Val), proline (Pro) and threonine (Thr). For these he thinks the old doublet codons persist essentially unchanged 鈥� indeed, the enzymes that link these amino acids to adaptor molecules called tRNA only match the first two bases of the codon. Even the original single-letter code leaves its mark in the fact that amino acids that share the same base in second position tend to have similar chemical properties.
Rob Knight, a specialist in the evolution of the genetic code at the University of Colorado, Boulder, says the hypothesis is plausible, but should be regarded as only one of several possible scenarios. Van den Elsen is hoping for further confirmation, and plans to scan the reams of 鈥渏unk鈥� DNA in most genomes in search of messages 鈥� ancient genes 鈥� that can be read using the older, doublet codes.
