Mutational activity of meiotic recombination

Recombination occurs in small localized regions 0.5-2 kilobases in size known as hotspots. Hotspot activity is highly variable among different individuals. Models about recombination hotspots are incomplete and controversial with several bizarre findings such as the hotspot paradox, rapid hotspot turnover during evolution and what appear to be heritable shifts in how hotspots are recruited for recombination. Not much is known about what drives hotspot activity, but new evidence supports that Prdm9, a zinc finger protein that acts as a histone methyltransferase, is a trans-activating factor regulating mammalian hotspot usage. The Prdm9 gene is highly variable especially in its Zinc fingers contacting the DNA and can accommodate changes in the DNA recognition sequence. This finding is very relevant because it suggest first, that sequence determinants play a pivotal role in meiotic recombination, and second, that there are mechanisms in place to balance variation in these sequence determinants. But why are sequence determinants that drive hotspot usage so variable and what are the main factors influencing this evolution? It has been argued that recombination leads to de novo mutations. Mutations could be an important driver of sequence evolution and influence hotspot activity. The observation that recombination is mutagenic is based on indirect sequence comparisons that have been quite controversial. The aim of this proposal is to experimentally investigate the mutagenic activity of meiosis in humans. Specifically, the number of new mutations introduced in recombination products will be measured and counted. The study includes the study of mutations in both crossovers and conversion products. These measurements will provide for the first time experimental verification whether recombination is mutagenic in humans. The results will be integrated into models of hotspot turnover and sequence evolution. This proposal addresses a small piece of the complexity of recombination hotspots, but it is crucial for understanding the bigger picture of what factors might influence recombination activity and its evolution.

Mutations are changes in our DNA that affect future generations if they occur in our germline. The factors driving new mutations are not very well known, but meiosis could be a potentially important source of new germline mutations. Meiosis is a key process in our sexual reproduction localized in recombination hotspots. Interestingly, the DNA in recombination hotspots is changing quickly, yet so far it is not fully understood why. The aim of this proposal was to analyze if the process of crossing over occurring in meiotic recombination leads to new mutations. For this purpose, we developed a highly sensitive methodology which counts mutagenic events in single crossover products. This approach also allowed us to characterize the frequency and type of the mutations within recombination hotspots and gain further insight about the mutagenic mechanisms. By sequencing a large number of single recombinant molecules obtained from human sperm for different donors and two hotspots, we found more new mutations in molecules with a crossover than in molecules without a recombination event, demonstrating for the first time that meiosis is mutagenic. Although, recombination had been previously suspected to be mutagenic, inferences made from sequence comparisons have not found strong evidence for a mutagenic effect of recombination. Thus, our data is an important contribution proofing that recombination is a source of new germline mutations. Moreover, based on the enrichment of mutations at methylated dinucleotides (CpG sites), it is likely that single-stranded DNA processing occurring during crossing over might be an important driver of mutagenesis. In addition to mutations, our large data set of crossovers also provided new evidence that GC alleles are transmitted more often than AT alleles at polymorphic sites, leading to a non-Mendelian segregation of alleles. This phenomenon, known as biased gene conversion, was previously demonstrated experimentally only in yeast. We showed that this biased transmission is a strong driver of hotspot sequence evolution opposing mutation, consistent with the idea that this biased transmission might be an adaptation to counteract the mutational load of recombination. Finally, many theories in evolutionary biology consider the process of recombination to be independent of mutagenesis, a premise that needs to be reconsidered in light of our latest findings.