Evolutionary Rate and Duplicability in the Arabidopsis thaliana Protein–Protein Interaction Network

Abstract

Genes show a bewildering variation in their patterns of molecular evolution, as a result of the action of different levels and types of selective forces. The factors underlying this variation are, however, still poorly understood. In the last decade, the position of proteins in the protein–protein interaction network has been put forward as a determinant factor of the evolutionary rate and duplicability of their encoding genes. This conclusion, however, has been based on the analysis of the limited number of microbes and animals for which interactome-level data are available (essentially, Escherichia coli, yeast, worm, fly, and humans). Here, we study, for the first time, the relationship between the position of proteins in the high-density interactome of a plant (Arabidopsis thaliana) and the patterns of molecular evolution of their encoding genes. We found that genes whose encoded products act at the center of the network are more evolutionarily constrained than those acting at the network periphery. This trend remains significant when potential confounding factors (gene expression level and breadth, duplicability, function, and length of the encoded products) are controlled for. Even though the correlation between centrality measures and rates of evolution is generally weak, for some functional categories, it is comparable in strength to (or even stronger than) the correlation between evolutionary rates and expression levels or breadths. In addition, genes encoding interacting proteins in the network evolve at relatively similar rates. Finally, Arabidopsis proteins encoded by duplicated genes are more highly connected than those encoded by singleton genes. This observation is in agreement with the patterns observed in humans, but in contrast with those observed in E. coli, yeast, worm, and fly (whose duplicated genes tend to act at the periphery of the network), implying that the relationship between duplicability and centrality inverted at least twice during eukaryote evolution. Taken together, these results indicate that the structure of the A. thaliana network constrains the evolution of its components at multiple levels.

Evolutionary Rate Heterogeneity of Core and Attachment Proteins in Yeast Protein Complexes

“…2002; Hahn and Kern 2005; Lemos et al 2005; Chakraborty et al . 2010, 2011; Alvarez-Ponce and Fares 2012). Similar trend has been also observed in our core/attachment data set (Spearman’s ρ connectivity vs .…”

“…According to this proposal, it has been demonstrated that essential genes evolve slower than the genes which are dispensable (Jordan et al 2002). Moreover, proteins those interact with a large number of partners evolve at a slower rate compared with the proteins with fewer number of interacting partners (Hirsh and Fraser 2001; Jordan et al 2002; Fraser et al 2003; Hahn and Kern 2005; Lemos et al 2005; Chakraborty et al 2011; Alvarez-Ponce and Fares 2012). However, these results have been questioned by some researchers (Batada et al 2006).…”

Gene Duplicability of Core Genes Is Highly Consistent across All Angiosperms

“…Despite the strong correlation between gene duplicability and gene function observed here, it remains to be further investigated which evolutionary mechanisms are responsible for the observed strong bias in duplicate retention patterns, and it remains to be established whether gene function directly influences gene duplicability or whether biased gene retention could be a by-product of other evolutionary phenomena instead, such as for instance the preservation of intermolecular interactions (dosage balance) or sequence constraints related to high levels of gene expression (Davis and Petrov, 2004;Drummond and Wilke, 2008). In particular, since network structure is often believed to constrain protein evolution and to underlie complex phenotypic traits, future work into this direction might benefit from investigating gene duplicability in a network context (Bekaert et al, 2011;D'Antonio and Ciccarelli, 2011;Alvarez-Ponce and Fares, 2012;Chae et al, 2012;Conant, 2014).…”

Metabolic Flux Is a Determinant of the Evolutionary Rates of Enzyme-Encoding Genes

“…To gain insight into the evolution of molecular systems, much research has been devoted to relating evolutionary properties of the genes, such as their substitution rates or detectable events of selection, with measures of their position and role within the system in which they participate, generally through its representation as a network. This approach of assessing molecular systems as networks has revealed that a significant, although small, part of the variation in the evolutionary rates of the genes can be explained by their position within the network (Fraser et al ; Papp et al ; Hahn and Kern ; Kim et al ; Greenberg et al ; Jovelin and Phillips ; Alvarez‐Ponce ; Alvarez‐Ponce and Fares ). However, biological interpretation of position within networks and its significance from an evolutionary point of view has largely remained obscure.…”