Neutral and non-neutral evolution of duplicated genes with gene conversion

Abstract

Gene conversion is one of the major mutational mechanisms involved in the DNA sequence evolution of duplicated genes. It contributes to create unique patters of DNA polymorphism within species and divergence between species. A typical pattern is so-called concerted evolution, in which the divergence between duplicates is maintained low for a long time because of frequent exchanges of DNA fragments. In addition, gene conversion affects the DNA evolution of duplicates in various ways especially when selection operates. Here, we review theoretical models to understand the evolution of duplicates in both neutral and non-neutral cases. We also explain how these theories contribute to interpreting real polymorphism and divergence data by using some intriguing examples.

Figure 1

Simple gene conversion model for a pair of duplicated genes (or paralogous genes). Converted tracts will be pasted to the corresponding part of the paralog, while there is no change in the original gene. Conversion can result in the novel mutation (shown in red) being spread or reversed. Although converted tracts will become identical in sequence, conversion will create novel haplotypes that are chimeras of the two genes (shown in purple and yellow).

Figure 2

Illustration of the sequence evolution after gene duplication. (a) Paralogs in species X and Y that originated by a duplication before the divergence of X and Y. The region in blue and red are evolving without and with gene conversion, respectively. (b) Gene tree without gene conversion, which reflects the true evolutionary history. (c) Gene tree with frequent gene conversion.

Figure 3

Typical behavior of the paralogous divergence d since the duplication event. d first increases to an equilibrium value represented by the broken line, then drifts around it. During this phase, the duplicates are considered to be undergoing concerted evolution. When concerted evolution is terminated, d starts increasing roughly linearly.

Figure 4

Hypothetical polymorphism data from a pair of duplicated genes with sample size 5. Positions where the nucleotides are in red contain individuals that have the same (shared) substitutions in both copies, referred to as shared polymorphisms. For instance, in position 1, haplotype 2 contains the substitution of C to A in both copies. The same applies to haplotype 3 in positions 5 and 6 (AT to TC), and haplotype 4 in position 9 (A to T).

Figure 5

Sequence divergence of the red- and green-opsin genes in human and macaque. The graph shows the divergence between the human red-opsin (long-wave: LW) and green-opsin (medium-wave: MW) genes of intron 4, exon 5, and intron 6. The divergence data was taken from [46]. The gene structure of the region is shown below with the introns 4 and 6 represented by black lines and the exon 5 represented by a gray box. Regions where the similarity is higher between paralogs of the same species than between orthologs of other species (i.e., the introns) due to frequent gene conversion is in a light red background, whereas regions where the similarity is higher between the orthologs than the paralogs (i.e., the exon) are in a light blue background. The two blue arrows indicate the positions of the two fixed amino acid substitutions that are largely responsible for the difference in color sensitivity [53]. The amino acid sequences of the red-opsin (LW) and green-opsin (MW) genes of human and macaque [49] around the two functionally significant substitutions are shown below. The sites in blue are sites where the difference is fixed between either of the orthologous pair, and the sites of the two functionally significant substitutions are in bold.

Figure 6

Different possible scenarios of the evolution of the RNase genes in colobine monkeys. Douc langur (Asian colobine) and Guereza (African colobine) both contain duplicates of the RNase1 gene, designated here as RNase1B. (a) Examples of sites containing amino acid substitutions that are shared across species but not within species, and those that are shared within species but not across species are shown in blue and red, respectively. The sequences are taken from [76]. Although [76] shows two duplicated genes in Guereza, designated as 1βand 1γ, the sites shown here are all sites that are identical in both 1β and 1γ. The exon-intron structure is shown above with the UTRs in open boxes and the coding region in a gray box. The positions of the sites containing substitutions shared across species but not within species are indicated by blue arrows, and those shared within species but not across species are indicated by red arrows. Regions where the similarity is higher between sequences of the same species than between sequences of other species is in a light red background, whereas regions where the similarity is higher between the sequences across species than sequences within species are in a light blue background. (b) One possible scenario—that the duplication occurred after the divergence of the two species. (c) Another possible scenario—that the duplication occurred prior to the divergence of the two species. One example each of amino acids shared across species but not within species (R and W in blue), and shared within species but not across species (P and S in red) are shown. The ancestral state is shown at the root of the tree and the branches in which the substitutions would have occurred according to each scenario are indicated. The arrow indicates gene conversion, although the direction is hypothetical and could have occurred in either direction.