Compensated Pathogenic Deviations: Analysis of Structural Effects

Abstract

Pathogenic deviations (PDs) in humans are disease-causing missense mutations. However, in some cases, these disease-associated residues occur as the wild-type residues in functionally equivalent proteins in other species and these cases are termed 'Compensated Pathogenic Deviations' (CPDs). The lack of pathogenicity in a non-human protein is presumed to be explained in most cases by the presence of compensatory mutations, most commonly within the same protein. Identifying structural features of CPDs, and detecting specific compensatory events, will help us to understand traversal along fitness landscape valleys in protein evolution.We divided mutations listed in the OMIM database into PD and CPD datasets and performed two independent analyses: (i) we searched for potential compensatory mutations spatially close to the CPDs and (ii) using our SAAPdb database, we examined likely structural effects to try to explain why mutations are pathogenic, comparing PDs and CPDs. Our datasets were obtained from a set of 245 human proteins of known structure and contained a total of 2328 mutations of which 453 (from 85 structures) were seen to be compensated in at least one functionally equivalent protein in another (non-human) species.Structural analysis results confirm previous findings that CPDs are, on average, 'milder' in their likely structural effects than uncompensated PDs and tend to be on the protein surface. We also showed that the residues surrounding the CPD residue in the folded protein are more often mutated than the residues surrounding an uncompensated mutation, supporting the hypothesis that compensation is largely a result of structurally local mutations.

Genes, mutations, and human inherited disease at the dawn of the age of personalized genomics

“…CPDs tend to be less severe in terms of the difference in physicochemical properties between the substituted and substituting amino acids than is normally the case for pathological mutations [Barešić et al, 2010;Ferrer-Costa et al, 2007]. In the context of human disease, Suriano et al [2007] have provided a good example of the influence of compensated and compensating mutations in the OTC gene.…”

Triangulation of the human, chimpanzee, and Neanderthal genome sequences identifies potentially compensated mutations

“…In evolution, compensatory mutations are unlikely to occur singly; indeed, Poon et al [2005] have suggested that an average of 11.8 compensatory mutations may interact epistatically with a given deleterious mutation so as to restore wild-type levels of fitness. CPDs tend to be less severe in terms of the physicochemical difference between the substituted and substituting amino acids than normally pathological mutations [Barešić et al, 2009;FerrerCosta et al, 2007]. Barešić et al [2009] have also shown that amino acid residues surrounding the compensated residue in the folded protein are mutated more often than residues surrounding an uncompensated mutation, consistent with the view that compensation relies upon structurally local mutations.…”

Compensated pathogenic deviations

“…However, while it is theoretically possible that a single mutation could do so, compensation of a clash is most likely to be achieved by making a number of smaller changes. Both Ferrer-Costa et al (33) and Barešić et al (10) found that CPDs have a higher average solvent accessibility. In other words, they are much more likely to be found on, or near, the protein surface.…”

“…Restoring protein stability and hence regulated activity, will often need compensatory mutations that restore stability to the acceptable range of free energies. From a structural analysis perspective, compensated mutations are preferentially on the protein surface (10,33). As shown by Poon et al (24), compensatory events are most commonly intragenic, so the surface location is likely to be a result of it being easier to accumulate compensatory events (probably before the CPD mutation occurs) rather than it being anything to do with interactions with other proteins.…”

“…As shown by Poon et al (24), compensatory events are most commonly intragenic, so the surface location is likely to be a result of it being easier to accumulate compensatory events (probably before the CPD mutation occurs) rather than it being anything to do with interactions with other proteins. In addition CPDs have 'milder' effects on the protein structure than uncompensated mutations (10,33) and tend to be more conservative in nature (33).…”