A novel intra-molecular protein-protein interaction code based on partial complementary coding of co-locating amino acids.
Proteins are assumed to contain all the information necessary for unambiguous folding and specific interaction with each other. However, ab initio structure prediction is often not successful because the amino acid sequence itself is simply not sufficient to guide between endless folding possibilities. It seems to be logical to try to find the "missing" information in nucleic acids, in the redundant codon. Statistical analyses of approximately 35K amino acid co-locations in 80 different protein structures indicate the existence of a weak intra-molecular protein-protein interaction code. Co-locating amino acids are preferentially coded by codons which are complementary in reverse orientation to each other at the 1st and 3rd codon positions, but not necessarily at the 2nd. This code, called D-1 X 3/RC-3 X 1, limits the number of preferred amino acid pairs from 20 to 10.3+/-0.8 (SEM, n=20) and emphasizes the importance of "strictly" defined amino acids (those having less synonymous codons). The existence of this code does not by any means violate the known physicochemical rules of protein folding or interaction. It is suggested that the biological source of preferential (specific) amino acid co-locations is the partial complementarity of their codons. This special coding of co-locating amino acids is important to better understanding of some fundamental biochemical processes and observations such as: (a) protein folding; (b) specific and high affinity protein-protein interactions; (c) the role of the wobble bases; (d) the significance of the redundant genetic code; (e) the origin of specific protein-protein interactions. Furthermore it might be useful even in protein design.