Differential retention and divergent resolution of duplicate genes following whole-genome duplication

Abstract

The Paramecium aurelia complex is a group of 15 species that share at least three past whole-genome duplications (WGDs). The macronuclear genome sequences of P. biaurelia and P. sexaurelia are presented and compared to the published sequence of P. tetraurelia. Levels of duplicate-gene retention from the recent WGD differ by >10% across species, with P. sexaurelia losing significantly more genes than P. biaurelia or P. tetraurelia. In addition, historically high rates of gene conversion have homogenized WGD paralogs, probably extending the paralogs' lifetimes. The probability of duplicate retention is positively correlated with GC content and expression level; ribosomal proteins, transcription factors, and intracellular signaling proteins are overrepresented among maintained duplicates. Finally, multiple sources of evidence indicate that P. sexaurelia diverged from the two other lineages immediately following, or perhaps concurrent with, the recent WGD, with approximately half of gene losses between P. tetraurelia and P. sexaurelia representing divergent gene resolutions (i.e., silencing of alternative paralogs), as expected for random duplicate loss between these species. Additionally, though P. biaurelia and P. tetraurelia diverged from each other much later, there are still more than 100 cases of divergent resolution between these two species. Taken together, these results indicate that divergent resolution of duplicate genes between lineages acts to reinforce reproductive isolation between species in the Paramecium aurelia complex.

Global deceleration of gene evolution following recent genome hybridizations in fungi

“…Ohno (1970) proposed an important theory that explains the enhanced evolutionary capability of polyploid genomes, stating that the emergence of functionally redundant genes would relax their evolutionary constraints and accelerate divergent evolution of their functions. This theory has been confirmed in a wide range of eukaryotes from ancient Vertebrata (Brunet et al 2006), Saccharomycetaceae fungi (Kellis et al 2004;Byrne and Wolfe 2007), Paramecia (McGrath et al 2014), and Brassicaceae plants (Arabidopsis Genome Initiative 2000; Vision et al 2000). Nonetheless, although these studies established long-term benefits of polyploidization, early consequences and mechanisms of polyploidization remain largely unexplored.…”

“…However, we did not observe any significant relationship between the decelerated genes and gene ontology annotations or numbers of protein interaction partners (S. cerevisiae gene annotation and protein-protein interaction data sets were used) (Methods). An alternative explanation could be that gene conversion-a mechanism capable of preserving and homogenizing homeolog sequences and functions (Takuno et al 2008;McGrath et al 2014)-removed nonsynonymous mutations from these genes. Because gene conversions create long identical sequence tracts between homeologs, we utilized these signatures to probe for evidence of gene conversions among hybrid homeologs (Methods).…”

“…Recent analyses of several fungal hybrids have revealed mechanisms such as extensive loss of heterozygosity Louis et al 2012;Stukenbrock et al 2012) and gene expression reprogramming (Tirosh et al 2009) that may help hybrid genomes stabilize. Gene conversion, which is capable of homogenizing redundant paralogs (Takuno et al 2008;McGrath et al 2014), can also play a role in resolving incompatibility between the two parental genomes of a hybrid. Despite these discoveries, to gain further insights into molecular consequences of post-polyploidization evolution, a comprehensive data set that permits comparisons between closely related polyploid and nonpolyploid genomes is absolutely critical.…”

Compensatory Drift and the Evolutionary Dynamics of Dosage-Sensitive Duplicate Genes

“…In a WGD that happened in the ancestor of teleost fish about 300 million years ago, many duplicate pairs persisted for over 200 million years before a member of the pair was lost (Blomme et al 2006;Brunet et al 2006;Sato et al 2009). Delayed loss of duplicates long after a WGD is also seen in Paramecium species (McGrath et al 2014b). Together these patterns indicate that many duplicates after WGDs are dosage sensitive and evolve in two phases: an initial prolonged phase where both duplicates evolve under selection that conserves function and a later phase in which a duplicate is lost.…”

“…(Figure 4 shows the impact of increasing D* on the time until duplicate loss.) Both yeast and paramecia show just this pattern: there is a positive correlation between expression levels and the longevity of duplicated genes following WGD (Seoighe and Wolfe 1999;Aury et al 2006;Gout et al 2010;McGrath et al 2014b). To explain this pattern, Gout et al (2010) argued that stabilizing selection on total expression is stronger on dosage-sensitive duplicates that have high levels of expression.…”