Research Article

Factors influencing meiotic recombination revealed by whole-genome sequencing of single sperm

See allHide authors and affiliations

Science  22 Mar 2019:
Vol. 363, Issue 6433, eaau8861
DOI: 10.1126/science.aau8861

You are currently viewing the abstract.

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

Sequencing and the single sperm

During meiosis, homologous chromosomes undergo doublestrand breaks in DNA that can cross over, shuffling genetic material. However, not every double-strand break resolves in a crossover event. Hinch et al. wanted to determine the rules governing DNA recombination. They developed a method to sequence individual mouse sperm and applied it to mice carrying two different alleles of a protein involved in mammalian crossovers. A high-resolution genetic map revealed the relationships between the distribution of crossovers, proteins involved in recombination, and specific factors determining whether a double-strand break becomes a crossover.

Science, this issue p. eaau8861

Structured Abstract


In diploid organisms, the two chromosomes in each homologous pair act independently of each other during most cellular functions. An exception occurs in meiosis, in which the pair of chromosomes must first locate each other in the cell nucleus and then physically exchange genetic material through recombination and crossing over. This physical exchange is mechanistically essential for proper chromosomal segregation in meiosis. Along with mutation, it also shapes patterns of genetic variation in natural populations, providing the substrate on which natural selection acts. Recombination is initiated by the formation of programmed DNA double-strand breaks (DSBs). Repairing these breaks entails a search for the matching sequence in the homologous chromosome. Although DSBs are predominantly repaired using the homolog as template, only a small proportion result in the formation of crossovers.


Most DSBs occur in narrow regions called recombination hotspots, defined in mice and many other species by the DNA binding specificities of the protein PRDM9. Despite recent progress, much remains unknown about the molecular processes occurring during meiosis and the factors affecting repair outcomes for individual DSBs. We have developed and applied a method for whole-genome amplification and DNA sequencing of single sperm to provide genome-wide maps of crossovers with unprecedented resolution. We combined this with molecular assays for various meiotic stages: H3K4me3 (which measures PRDM9 binding), SPO11-oligos (which count DSBs), and DMC1 on single-stranded DNA (which measures the number and persistence of DSBs).


We report single-cell sequencing of 217 sperm from a hybrid mouse. We inferred 2649 crossovers genome-wide, resolved to a median resolution of 916 base pairs (bp), with 386 crossovers resolved within 250 bp.

By comparing our high-resolution crossover map with stage-specific molecular measures of recombination, we identify four factors that strongly increase the chance that a particular DSB will resolve as a crossover: (i) whether PRDM9 has bound the uncut template chromosome (the chromosome used for repair) at the site of the DSB, (ii) the proximity of the hotspot to the telomere, (iii) local GC content, and (iv) the Prdm9 allelic type of the hotspot. We show that each of these four factors also consistently decreases homolog engagement time, specifically the time until single-stranded DNA at the DSB site has located and invaded the DNA duplex of its homolog.

We show that the precise location of the breakpoint in a crossover—the switchpoint from one homolog to the other—is also affected by whether PRDM9 has bound the template. Crossover breakpoints are modulated by the chromatin environment of the template chromosome, avoiding positions occupied by nucleosomes. We also find that the pseudoautosomal region, which must have a crossover in males, is likely determined by cis-acting factors and has a higher than expected use of hotspots that are activated independently of PRDM9.


Our work identifies several additional roles for PRDM9 in meiosis beyond positioning DSBs. We show that the breaks that are fastest to engage their homolog are more likely to repair as crossovers. Each of the contributing factors we identified also suggests mechanisms that could facilitate the otherwise seemingly intractable challenge of homology search.

Identification of crossovers in sperm and factors affecting the repair of double-strand breaks in meiosis.

(Top) Meiotic cells undergo programmed double-strand breaks, some of which resolve as crossovers. DNA sequencing of single sperm identifies sites of crossover. (Bottom) A double-strand break is more likely to engage its homolog quickly (and to resolve as a crossover) if PRDM9 binds the same location on the homologous chromosome, if it is near the telomere, or if the local GC content is high.


Recombination is critical to meiosis and evolution, yet many aspects of the physical exchange of DNA via crossovers remain poorly understood. We report an approach for single-cell whole-genome DNA sequencing by which we sequenced 217 individual hybrid mouse sperm, providing a kilobase-resolution genome-wide map of crossovers. Combining this map with molecular assays measuring stages of recombination, we identified factors that affect crossover probability, including PRDM9 binding on the non-initiating template homolog and telomere proximity. These factors also influence the time for sites of recombination-initiating DNA double-strand breaks to find and engage their homologs, with rapidly engaging sites more likely to form crossovers. We show that chromatin environment on the template homolog affects positioning of crossover breakpoints. Our results also offer insights into recombination in the pseudoautosomal region.

View Full Text