The molecular evolution of genes and proteins: a tale of two serines

The advent of techniques for cloning and rapidly sequencing DNA has produced an explosive increase of sequence information for nucleic acids and their inferred proteins. Careful study of this large store of data might give us new insights into the relations between the linear sequences of genes and their functions embodied in the three-dimensional structure of proteins, and also illuminate the origin and evolution of the structural complexity of present-day proteins. Here I argue from such a study that the active site sequences of enzymes that have analogous essential serine residues lie in fact on two lines of descent from an ancient ancestral enzyme which had a cysteine instead of serine in its active site. This is based on the assumption that the two codon types which define the separate lines of descent and which have different bases in two positions could not interconvert by single mutations.