Additions, losses, and rearrangements on the evolutionary route from a reconstructed ancestor to the modern Saccharomyces cerevisiae genome

Abstract

Comparative genomics can be used to infer the history of genomic rearrangements that occurred during the evolution of a species. We used the principle of parsimony, applied to aligned synteny blocks from 11 yeast species, to infer the gene content and gene order that existed in the genome of an extinct ancestral yeast about 100 Mya, immediately before it underwent whole-genome duplication (WGD). The reconstructed ancestral genome contains 4,703 ordered loci on eight chromosomes. The reconstruction is complete except for the subtelomeric regions. We then inferred the series of rearrangement steps that led from this ancestor to the current Saccharomyces cerevisiae genome; relative to the ancestral genome we observe 73 inversions, 66 reciprocal translocations, and five translocations involving telomeres. Some fragile chromosomal sites were reused as evolutionary breakpoints multiple times. We identified 124 genes that have been gained by S. cerevisiae in the time since the WGD, including one that is derived from a hAT family transposon, and 88 ancestral loci at which S. cerevisiae did not retain either of the gene copies that were formed by WGD. Sites of gene gain and evolutionary breakpoints both tend to be associated with tRNA genes and, to a lesser extent, with origins of replication. Many of the gained genes in S. cerevisiae have functions associated with ethanol production, growth in hypoxic environments, or the uptake of alternative nutrient sources.

Figure 1. Inferring ancestral gene content and gene order.

(A) Phylogenetic relationships among the species considered (not to scale; topology from ref. [89]), and the position of the ancestor whose genome was reconstructed. The dot indicates the whole-genome duplication. Numbers of genomic rearrangements (reciprocal translocations and inversions) shared by post-WGD clades are shown above the branches (Table S2). The numbers on the right show the number of genomic blocks shared by each species and the reconstructed ancestor. Asterisks indicate block numbers that are probably overestimates because the corresponding genome sequences are fragmented into many contigs. (B,C) Principles of the parsimony method for ancestral genome reconstruction. Colors represent continuous chromosomal regions. For simplicity the diagrams show only one non-WGD species and one post-WGD species, but in practice we used all the species shown in panel A. We infer that a gene was present in the ancestor if it is present in at least one non-WGD species and one track of a post-WGD species, or if paralogs are present on both post-WGD tracks. Genes found only on one post-WGD track, or only in non-WGD species, cannot be inferred to have been present in the ancestor and are marked with red crosses. Two types of scenario for gene order rearrangement can exist . In each case the inferred ancestral order is shown on the right. (B) Single break of synteny. Gene order is conserved between a non-WGD species and one of the two tracks in a post-WGD species, due to a rearrangement on the other track after WGD. (C) Double break of synteny. The two tracks from the post-WGD species agree with each other but disagree with the non-WGD species. The ancestor is inferred to have the gene order present in the post-WGD species.

Figure 2. Synteny relationship between the reconstructed ancestral genome and the modern S. cerevisiae genome.

Each colored block represents a region in S. cerevisiae that is colinear with a region of the ancestral genome. The 182 double-conserved-synteny blocks in S. cerevisiae map onto the ancestor in a 2∶1 pattern. Colors correspond to the 16 modern S. cerevisiae chromosomes as shown at the bottom.

Figure 3. Example of a simple reciprocal translocation in S. cerevisiae.

Parts of two ancestral chromosomes, ANC5 and ANC2, are shown at the top. After WGD, these formed four chromosomes (labeled Post5A, Post5B, Post2A, Post2B), each of which retains a subset of the ancestral gene sets. Parts of S. cerevisiae chromosomes XI and XIV are derived from chromosomes Post5A and Post2A, respectively, without further rearrangement. A reciprocal translocation between chromosomes Post5B and Post2B gave rise to part of S. cerevisiae chromosomes XV and IX.

Figure 4. Reciprocal translocations that formed the S. cerevisiae genome from the ancestral genome.

Each point on the circle represents a breakpoint, and names two genes (separated by a | symbol) that were adjacent in the ancestral genome but became separated by reciprocal translocation on the S. cerevisiae lineage. These breakpoints form the ends of the synteny blocks shown in Figure 2. The genes at the breakpoints are arranged according to their current positions on the S. cerevisiae chromosomes, so each breakpoint appears twice in the circle (once for each end, usually on different chromosomes). Green backgrounds join the names of pairs of breakpoints that were formed by simple reciprocal translocation events. As an example, the pink dots highlight the new junctions on chromosomes IX and XV that were formed by the simple reciprocal translocation shown in Figure 3, involving the breakpoints YOR084W|YIL143C and YOR085W|YIL142W. In cases of breakpoint reuse, the genes on one side of the pair of breakpoints are adjacent in S. cerevisiae, but the genes on the other side are not. By iteratively linking each of the non-matching genes to the gene that is adjacent to it in the S. cerevisiae genome, we can describe groups of 3–5 reciprocal translocation events with breakpoints that have been used more than once (colored arcs). We observed one event where a breakpoint created by a telomeric translocation was reused (dashed gray line). Telomeric translocation events are indicated by gray backgrounds on breakpoint names. This diagram is an adjacency graph applied to a genome halving context ,.

Figure 5. The 124 non-telomeric genes that were gained on the S. cerevisiae lineage since WGD.

Colored backgrounds indicate genes that are adjacent and may have been gained simultaneously.

Figure 6. The YPR071W gene family in S. cerevisiae.

All members of this family were gained by S. cerevisiae since WGD and all are located near tRNA genes. (A) T-Coffee multiple alignment of the five proteins. Black and grey backgrounds show residues that are identical or similar, respectively, in ≥3 sequences. (B) Maps of the genomic regions around each gene. Red, YPR071W family members; orange, tRNA genes; blue, other genes in the gained set; white, Ty elements and long terminal repeats; gray, genes in the ancestral set (only the first gene on each side is shown). Tick marks indicate intervals of 1 kb.