| Protein domains are the basic structural and functional units in a protein. Interestingly, although most protein domains are structured, a few disordered protein domains have been reported. So, it's important to calculate the number of disordered protein domains encoded by the genome and comprehensively explore their evolutionary and functional features, which will enrich our understanding of the evolution of protein repertoire.Here, we focus on the structural disorder ratio(SDR) of protein domains of six species-- H. sapiens, M. musculus, D. rerio, D. melanogaster, S. cerevisiae and E. coli. We found that about 3.8-22.5% protein domains are highly disordered. Interestingly, the proportions of structured/unstructured protein domains change obviously among the species of different evolutionary lineages. We found that the lower species, such as E. coli and S. cerevisiae, contain more completely structured protein domains, while the higher eukaryotes, such as M. musculus and H. sapiens, contain more highly disordered protein domains. It indicates the proportions of disordered domain may have been changed during evolution. In the view of domain age, we found the younger domains consist of higher fraction of highly disordered domains than the older domains, especially in higher eukaryotes species, such as D. rerio, M. musculus and H. sapiens, in which the highly disordered domains occupy almost half of the young domains.So what are the functional features of disordered domains and structured domains? We found that structured domains tend to localize at membrane, and take part in the biological processes related to transport, GPCR signaling pathway, oxidation reduction, metabolic process(especially for lipid and carbohydrate metabolic process). On the contrary, highly disordered domains tend to localize at intracellular membrane-bounded organelle, anticipate the regulation of transcription, DNA-dependent, cell adhesion, cellular component organization, reproductive process, cell communication, developmental process, etc. It suggests that the increasing of requirement of complex functions during evolution maybe the cause of the emergence of disordered domains.We also found the repeating protein domains are different in DSDR(Domain Structured Disordered Ratio). The factor influencing the variation of DSDR may be the hydrophobic property of amino acids residues within the domain regions and the amino acid residues closed to the domain regions. We found that the variation of DSDR(VDSDR) of the repeating domains which had low sequence similarity were mostly low, which indicates sequences of repeating domains are not conserved, but their structures are conserved. Low sequence similarity can avoid the aggregation of same domains so that the proteins won't be mis-folded. While the low variation of DSDR makes the proteins form correct structures. Comparing the VDSDRs of domains within the same proteins and the domains among different proteins, we found that the repeating domains within the same proteins have much higher proportions of no-variation DSDRs than that of same domains among different proteins. And this trend is much weaker in higher species, which suggests there are more diversifications of the structure and function of the repeating domains within one some protein in higher species.The variation of DSDR of protein domains closely relates to their functions. Proteins with no-variation repeating domains tend to take part in the biological processes related to collagen catabolism, transport, signal transduction and tend to localize at extracellular region, membrane, mitochondrion. While proteins with high-variation repeating domains tend to take part in nitrogen compound metabolism, negative regulation of protein polymerization, transcription, regulation of transcription, cell-cell adhesion, etc and they tend to localize at cytoskeleton, nucleus and non-membrane organelles These results reveal the functional constrains for the variation of DSDR during evolution. |
PDF Full Text Request