Evidências de mudanças estruturais proteicas em transições macroevolutivas

Abstract

Proteins are linear polymers of amino acids involved in a huge range of different structures and functions within the cell. Its versatility is related to the diversity and ordering of residues found in its side chains. The polymer chain (primary structure) folds into secondary and tertiary structures, and may form multimers, known as quaternary structures. Genomic and structural data allow new discoveries about how proteins probably evolved from a small initial set of domains and domain arrangements. Some protein-coding genes appear to have arisen from random parts of DNA, others seem to exist for billions of years, virtually unchanged. In this work, we investigate if exists a structural bias related to the period of origin of the proteins. Knowing the time of origin of proteins, determined by phylostratigraphic methods, we analyzed data from secondary structure, obtained both from experimental data and prediction data. We found that more recent proteins have generally fewer structured content and, at same time, shows more diverse structural content of alpha helices and beta strands. Differences in the use of amino acids in domains and inter-domains regions may explain these changes. We noticed that the most significant changes occur in proteins originating from Opisthokonta. Functions of the DNA binding and transcription factors are enriched in proteins originated in this clade and in proteins with lower secondary structure composition. Furthermore, we investigated if the observed bias can be extended to understanding processes of novel proteins origins. Predictions of ORFs derived from noncoding regions showed that the secondary structure composition of these sequences are compatible with those found in natural proteins. Thus, the source of new genes is compatible with the readily achieved by translation of these sequences, given support for de novo hypothesis.