Stable recombination hotspots in birds

Abstract Although the DNA-binding protein PRDM9 plays a critical role in the specification of meiotic recombination hotspots in mice and apes, it appears to be absent from many vertebrate species, including birds. To learn about the determinants of fine-scale recombination rates and their evolution in natural populations lacking PRDM9 , we inferred fine-scale recombination maps from population resequencing data for two bird species, the zebra finch Taeniopygia guttata , and the long-tailed finch, Poephila acuticauda , whose divergence is on par with that between human and chimpanzee. We find that both bird species have hotspots, and these are enriched near CpG islands and transcription start sites. In sharp contrast to what is seen in mice and apes, the hotspots are largely shared between the two species, with indirect evidence of conservation extending across bird species tens of millions of years diverged. These observations link the evolution of hotspots to their genetic architecture, suggesting that in the absence of PRDM9 binding specificity, accessibility of the genome to the cellular recombination machinery, particularly around functional genomic elements, both enables increased recombination and constrains its evolution.