Comparison of Exon-boundary Old and Young Domains during Metazoan Evolution

Abstract

Domains are the building blocks of proteins. Exon shuffling is an important mechanism accounting for combination of a limited repertoire of protein domains in the evolution of multicellular species. A relative excess of domains encoded by symmetric exons in metazoan phyla has been presented as evidence of exon shuffling, and symmetric domains can be divided into old and new domains by determining the ages of the domains. In this report, we compare the spread, versatility, and subcellular localization of old and new domains by analyzing eight metazoan genomes and their respective annotated proteomes. We found that new domains have been expanding as multicellular organisms evolved, and this expansion was principally because of increases in class 1-1 domains amongst several classes of domain families. We also found that younger domains have been expanding in membranes and secreted proteins along with multi-cellular organism evolution. In contrast, old domains are located mainly in nuclear and cytoplasmic proteins. We conclude that the increasing mobility and versatility of new domains, in contrast to old domains, plays a significant role in metazoan evolution, facilitating the creation of secreted and transmembrane multidomain proteins unique to metazoa.

Exon-phase symmetry and intrinsic structural disorder promote modular evolution in the human genome

“…A similar difference can be seen when symmetric and asymmetric exons in this latter class are compared, with an excess of symmetric exons in the short length range (Figure 3B). These differences might infer that exons of order represent more uniform building blocks [domains, in accord with prior inferences (3,5,6)], whereas exons of disorder are much more spread out because they represent three different types of functional modules (motifs, linkers and domains). Although comparison of the motif and domain ‘content’ of exons of different lengths is not conclusive enough, the overrepresentation of certain motif and domain types points to this direction (see later in the text).…”

“…Interestingly, the level of disorder of regions of symmetric exons strongly correlates with the overall level of disorder of the protein ( P < 0.0001) (Supplementary Figure S1). Actually, disorder corresponding to all/asymmetric exons also strongly correlates with the overall level of disorder of the protein ( P < 0.0001), but the average disorder is always higher in the case of symmetric exons, which may infer that in the evolutionary construction of ordered proteins, mostly ordered symmetric exons have been used (5,6), whereas in the evolutionary construction of disordered proteins, mostly disordered symmetric exons have been preferred.…”

“…Because inclusion/exclusion of symmetric exons does not impair the translation reading frame, they are favoured in shuffling reactions. Further, their incorporation into a recipient gene may result in a protein product of altered/improved functionality, provided they encode for autonomous structural/functional units of proteins (domains) (5,6). …”

“…In the case of folded proteins, this dilemma is thought to be solved if the exon encodes for a domain (or secondary structure element) inserted at an appropriate point (most often in a loop) (3,5,6). In the case of disordered proteins (regions), this is not that much of an issue because they can easily accommodate multiple conformations (9,52) imposed by different end-point positions in the host protein.…”

Evolutionary Transitions to Multicellular Life

“…Such novel architecture would then be found in one, Fig. Domain shuffling appears to be an ongoing feature of the evolution of cell surface and extracellular proteins underlying cell adhesion, communication and self-nonself recognition (Lee 2009; for example see Patel et al 1987;Degnan 2010, 2012;. A possible mechanism for the evolution of an early allorecognition locus, and corresponding proteins, is shown.…”