PgmNr P363: Secreted Proteins evade the Expression – Rate Anticorrelation.

Authors:
Felix Feyertag; Patricia M. Berninsone; David Alvarez-Ponce


Institutes
Univ. of Nevada, Reno, NV.


Abstract:

The evolutionary rate of proteins can vary by orders of magnitude, both between and within species. A primary factor influencing protein evolutionary rate is protein expression level. A strong negative correlation between expression level and evolutionary rate exists in all studied organisms (the so-called E−R anticorrelation). The translational robustness hypothesis proposes that the main underlying cause of the E−R anticorrelation is the deleterious effect of proteins misfolding due to mistranslation. According to this hypothesis, highly abundant proteins are under strong selective pressure to fold correctly despite translation errors. The misinteraction avoidance hypothesis proposes that highly expressed proteins are under stronger selective pressures in order to avoid unspecific protein-protein interactions. In this study we investigate the E−R anticorrelation in secreted proteins, and find that, in contrast to intracellular proteins, the anticorrelation is weaker and/or non-existent in secreted proteins. We calculated the evolutionary rate of all human proteins from comparison with their mouse orthologs, and divided them into secreted (N = 2279) and non-secreted (N = 13567). In accordance with previous studies, we observed a significant E−R anticorrelation for non-secreted proteins (ρ = -0.259, p = 4.56×10181); however, for secreted proteins this anticorrelation was not observed (ρ = 0.038, p = 0.084). We confirmed this apparent contradiction to the E−R anticorrelation amongst the class of secreted proteins by repeating our analyses using protein abundance data for 6 individual human tissues, and mRNA abundance data for 32 human tissues. In all cases, the E−R anticorrelation was weaker among secreted proteins. Furthermore, we made similar observations in other species, finding that the E−R anticorrelation is weaker in secreted proteins than in non-secreted proteins in C. elegansD. melanogasterS. cerevisiaeA. thaliana and E. coli. We will discuss how the secretory pathway may reduce the selective pressures of misfolding and misinteraction, thus providing possible explanations as to why the E−R anticorrelation is observed in non-secreted proteins but not in secreted proteins.