Collagens achieve highly biased amino acid composition not only by random mutation but also by slanted assignment of the genetic code table

Abstract

Collagens are proteins that are ubiquitous in animal bodies. They have unique triple-helix domains mainly consisting of glycines, prolines, and hydroxyprolines, resulting in collagens being highly biased amino acid compositions. Traditionally, we have considered that only the accumulation of random mutations develops the structures and compositions of proteins. However, the results of the present study suggest that it is not only random mutations but also the slanted assignment of the genetic code table that assisted in forming the biased amino acid composition of collagens.

In my previous paper, I showed that the synonymous codon usage selections in all proteins of various 23 bacteria species primarily offset the influences of guanine and cytosine (GC) content variation on their amino acid compositions. In this report, I did the same analysis on all the human proteins and found two things. First, the human proteins' coding sequences have a broader GC content range than each bacteria species. Second, most human proteins' synonymous codon usage selections offset the influences of GC content variation like bacteria, but those of some proteins like collagens did not. Instead, synonymous codon usage selections of these proteins emphasized the feature of their high GC content.

These findings suggest that the slanted genetic code table assignment assisted animals in forming collagens, highly biased amino acid composition proteins, and assisted us be multicellular organisms in our evolutions.