Limits...
The red queen model of recombination hotspots evolution in the light of archaic and modern human genomes.

Lesecque Y, Glémin S, Lartillot N, Mouchiroud D, Duret L - PLoS Genet. (2014)

Bottom Line: Recombination is an essential process in eukaryotes, which increases diversity by disrupting genetic linkage between loci and ensures the proper segregation of chromosomes during meiosis.However, the reasons for these changes and the rate at which they occur are not known.Surprisingly, however, our analyses indicate that Denisovan recombination hotspots did not overlap with modern human ones, despite sharing similar PRDM9 target motifs.

View Article: PubMed Central - PubMed

Affiliation: Laboratoire de Biométrie et Biologie Evolutive, UMR CNRS 5558, Université Lyon 1, Villeurbanne, France.

ABSTRACT
Recombination is an essential process in eukaryotes, which increases diversity by disrupting genetic linkage between loci and ensures the proper segregation of chromosomes during meiosis. In the human genome, recombination events are clustered in hotspots, whose location is determined by the PRDM9 protein. There is evidence that the location of hotspots evolves rapidly, as a consequence of changes in PRDM9 DNA-binding domain. However, the reasons for these changes and the rate at which they occur are not known. In this study, we investigated the evolution of human hotspot loci and of PRDM9 target motifs, both in modern and archaic human lineages (Denisovan) to quantify the dynamic of hotspot turnover during the recent period of human evolution. We show that present-day human hotspots are young: they have been active only during the last 10% of the time since the divergence from chimpanzee, starting to be operating shortly before the split between Denisovans and modern humans. Surprisingly, however, our analyses indicate that Denisovan recombination hotspots did not overlap with modern human ones, despite sharing similar PRDM9 target motifs. We further show that high-affinity PRDM9 target motifs are subject to a strong self-destructive drive, known as biased gene conversion (BGC), which should lead to the loss of the majority of them in the next 3 MYR. This depletion of PRDM9 genomic targets is expected to decrease fitness, and thereby to favor new PRDM9 alleles binding different motifs. Our refined estimates of the age and life expectancy of human hotspots provide empirical evidence in support of the Red Queen hypothesis of recombination hotspots evolution.

Show MeSH

Related in: MedlinePlus

Derived allele frequency (DAF) spectra of mutations leading to motifs loss in the human branch.DAF of mutations affecting HM (purple bars) and CM (orange bars) along the human branch (green branch in Figure 1). Allele frequencies are extracted from 1000 genomes phase I, using all available populations [31]. Mutations count for each motif (F2 subset) is indicated (N).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4230742&req=5

pgen-1004790-g003: Derived allele frequency (DAF) spectra of mutations leading to motifs loss in the human branch.DAF of mutations affecting HM (purple bars) and CM (orange bars) along the human branch (green branch in Figure 1). Allele frequencies are extracted from 1000 genomes phase I, using all available populations [31]. Mutations count for each motif (F2 subset) is indicated (N).

Mentions: These observations are consistent with the self-destructive dBGC drive model. However, they could also be explained by a possible mutagenic effect of recombination. To distinguish between these two possibilities, we analyzed the derived allele frequency (DAF) spectra of mutations in HM and CM motifs: under the hypothesis that the increased HM loss rate is simply due to a higher mutation rate (and not a fixation bias, like BGC), the two DAF spectra are expected to be identical. We included in these analyses all modern-human mutations detected as polymorphic by the 1000 genomes project [31], as well as fixed ones. We observed that the DAF spectrum of HM mutations is shifted towards higher frequencies compared to CM mutations (Figure 3), with an average mean DAF almost three times higher (13% vs. 5%; Wilcoxon test p = 1.9×10−6). Overall 3.7% of HM mutations detected in the modern human branch are fixed, compared to 0.2% for CM mutations (Proportion test p = 1.6×10−4). This demonstrates that the accumulation of HM losses in the human branch is a consequence of a fixation bias, as predicted by the dBGC model [32].


The red queen model of recombination hotspots evolution in the light of archaic and modern human genomes.

Lesecque Y, Glémin S, Lartillot N, Mouchiroud D, Duret L - PLoS Genet. (2014)

Derived allele frequency (DAF) spectra of mutations leading to motifs loss in the human branch.DAF of mutations affecting HM (purple bars) and CM (orange bars) along the human branch (green branch in Figure 1). Allele frequencies are extracted from 1000 genomes phase I, using all available populations [31]. Mutations count for each motif (F2 subset) is indicated (N).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4230742&req=5

pgen-1004790-g003: Derived allele frequency (DAF) spectra of mutations leading to motifs loss in the human branch.DAF of mutations affecting HM (purple bars) and CM (orange bars) along the human branch (green branch in Figure 1). Allele frequencies are extracted from 1000 genomes phase I, using all available populations [31]. Mutations count for each motif (F2 subset) is indicated (N).
Mentions: These observations are consistent with the self-destructive dBGC drive model. However, they could also be explained by a possible mutagenic effect of recombination. To distinguish between these two possibilities, we analyzed the derived allele frequency (DAF) spectra of mutations in HM and CM motifs: under the hypothesis that the increased HM loss rate is simply due to a higher mutation rate (and not a fixation bias, like BGC), the two DAF spectra are expected to be identical. We included in these analyses all modern-human mutations detected as polymorphic by the 1000 genomes project [31], as well as fixed ones. We observed that the DAF spectrum of HM mutations is shifted towards higher frequencies compared to CM mutations (Figure 3), with an average mean DAF almost three times higher (13% vs. 5%; Wilcoxon test p = 1.9×10−6). Overall 3.7% of HM mutations detected in the modern human branch are fixed, compared to 0.2% for CM mutations (Proportion test p = 1.6×10−4). This demonstrates that the accumulation of HM losses in the human branch is a consequence of a fixation bias, as predicted by the dBGC model [32].

Bottom Line: Recombination is an essential process in eukaryotes, which increases diversity by disrupting genetic linkage between loci and ensures the proper segregation of chromosomes during meiosis.However, the reasons for these changes and the rate at which they occur are not known.Surprisingly, however, our analyses indicate that Denisovan recombination hotspots did not overlap with modern human ones, despite sharing similar PRDM9 target motifs.

View Article: PubMed Central - PubMed

Affiliation: Laboratoire de Biométrie et Biologie Evolutive, UMR CNRS 5558, Université Lyon 1, Villeurbanne, France.

ABSTRACT
Recombination is an essential process in eukaryotes, which increases diversity by disrupting genetic linkage between loci and ensures the proper segregation of chromosomes during meiosis. In the human genome, recombination events are clustered in hotspots, whose location is determined by the PRDM9 protein. There is evidence that the location of hotspots evolves rapidly, as a consequence of changes in PRDM9 DNA-binding domain. However, the reasons for these changes and the rate at which they occur are not known. In this study, we investigated the evolution of human hotspot loci and of PRDM9 target motifs, both in modern and archaic human lineages (Denisovan) to quantify the dynamic of hotspot turnover during the recent period of human evolution. We show that present-day human hotspots are young: they have been active only during the last 10% of the time since the divergence from chimpanzee, starting to be operating shortly before the split between Denisovans and modern humans. Surprisingly, however, our analyses indicate that Denisovan recombination hotspots did not overlap with modern human ones, despite sharing similar PRDM9 target motifs. We further show that high-affinity PRDM9 target motifs are subject to a strong self-destructive drive, known as biased gene conversion (BGC), which should lead to the loss of the majority of them in the next 3 MYR. This depletion of PRDM9 genomic targets is expected to decrease fitness, and thereby to favor new PRDM9 alleles binding different motifs. Our refined estimates of the age and life expectancy of human hotspots provide empirical evidence in support of the Red Queen hypothesis of recombination hotspots evolution.

Show MeSH
Related in: MedlinePlus