Origin and fate of pseudogenes in Hemiascomycetes: a comparative analysis

Abstract

Background: Pseudogenes are ubiquitous genetic elements that derive from functional genes after mutational inactivation. Characterization of pseudogenes is important to understand genome dynamics and evolution, and its significance increases when several genomes of related organisms can be compared. Among yeasts, only the genome of the S. cerevisiae reference strain has been analyzed so far for pseudogenes.

Results: We present here the first comparative analysis of pseudogenes within the fully sequenced and annotated genomes of eight yeast species, spanning the entire phylogenetic range of Hemiascomycetes. A total of 871 pseudogenes were found, out of which mutational degradation patterns and consequences on the genetic repertoire of each species could be identified. We found that most pseudogenes in yeasts originate from mutational degradation of gene copies formed after species-specific duplications but duplications of pseudogenes themselves are also encountered. In all yeasts, except in Y. lipolytica, pseudogenes tend to cluster in subtelomeric regions where they can outnumber the number of functional genes from 3 to 16 times. Pseudogenes are generally not conserved between the yeast species studied (except in two cases), consistent with their large evolutionary distances, but tend to be conserved among S. cerevisiae strains. Reiterated pseudogenization of some genes is often observed in different lineages and may affect functions essential in S. cerevisiae, which are, therefore, lost in other species. Although a variety of functions are affected by pseudogenization, there is a bias towards functions involved in the adaptation of the yeasts to their environment, and towards genes of unknown functions.

Conclusions: Our work illustrates for the first time the formation of pseudogenes in different branches of hemiascomycetous yeasts, showing their limited conservation and how they testify for the adaptation of the yeasts functional repertoires.

Figure 1

Boxplot [67]of the sequence divergence between pseudogenes and their closest functional homolog. P-distance (ordinate) is expressed as the fraction of non-identical nucleotides at the third positions of codons (see Methods). Left panel: p-distance of pseudogenes whose closest functional homolog (bestmatch) is in the same genome (paralog), central panel: p-distance of pseudogenes whose bestmatch is in another genome, right panel: p-distance of pairs of functional paralogs in the same species. The number of pairs analyzed is indicated in parenthesis (data in Table II).

Figure 2

Conserved pseudogenes at syntenic locations. Each column represents a set of orthologous sequences in a region of synteny conservation. Vertical dashed lines separate the different regions. Rectangles represent annotated genes, dashed rectangles represent pseudogenes detected in this analysis. All these pseudogenes have no paralog in the genome. The topology of the species phylogeny [68] is given on the left of the figure (branch lengths ignored).

Figure 3

Possible origin of pseudogenes. See text for explanations. The diamonds correspond to distinctive criteria and rectangles to deduced origin.

Figure 4

Scenario for the multiplication of pseudogenes in Y. lipolytica. a) Rectangle represents the functional gene, dashed rectangles represent its corresponding pseudogenes. The tree topology is obtained by maximum likelihood reconstruction [69] based on the aligned nucleic acid sequences (branch lengths ignored). The emergence of frameshift mutations (!) and in-frame stop-codons (*) are indicated above corresponding branches. b) Alignment of the translation products of YALI0A14927g and its pseudogenes obtained by MUSCLE. frameshift mutations (!) and in-frame stop-codons (*) are boxed.

Figure 5

Pseudogenes in pairs of ohnologs in Saccharomycetaceae. Same legend as Figure 2. Pairs of ohnologs, i.e. paralogs originating from the whole-genome duplication [70], are linked by brackets.