In organisms that reproduce sexually, haploid gamete cells are generated during meiosis, a type of cell division during which parental chromosomes (homologs) separate from each other. This process requires a physical connection, known as a “crossover”, between homologs. Crossover events result in a reciprocal exchange of genetic material between the maternal and paternal chromosomes and are a major source of genetic variation in the nuclei produced by meiosis. This swapping of genetic information can be detected as linkage alterations with the use of appropriate DNA markers. In 1915, Alfred Sturtevant found, while constructing genetic maps of Drosophila, that the positions of multiple crossovers along a chromosome were not random with regard to each other. In fact, the presence of one crossover in a given chromosome interval tends to prevent the occurrence of another crossover in a nearby interval. This phenomenon was accordingly termed “interference” by Hermann Muller, who made the early observation. Interference has subsequently been shown to occur in nearly all eukaryotes investigated; however, the basis of how it works remains unknown. What is known is that crossovers arise from DNA breaks, which occur at “hotspots” with high frequency. DNA double-strand breaks (DSBs) and the consequent crossovers have to be tightly orchestrated for meiosis to occur successfully.
Why has it been so challenging to understand how interference works? Basic Sciences Division Member Dr. Gerry Smith thinks that some of the hurdles were “first a technical challenge – the difficulty of precisely measuring low-frequency events – and second a conceptual challenge – there were no good molecular models to study this. Only in 1990 did King and Mortimer propose a model for crossover interference, one involving polymerization of an unstated protein along chromosomes. And since then the models for crossover interference have not involved known molecules and made testable predictions.” Fast-forward to 2018, Dr. Smith and his laboratory discovered a physical mechanism for control of DSBs between distant hotspots through the 3D clustering of hotspots bound by their determinant proteins. From these findings, they propose a physical mechanism for crossover interference, whose mechanism has remained elusive for more than 100 years. Their findings are published in a recent issue of the Proceedings of the National Academy of Sciences.
The authors picked the fission yeast Schizosaccharomyces pombe to study the mechanism of meiotic crossover interference. In this organism, DSB hotspots are approximately 30-50 kb apart, and almost all are bound with high specificity by Rec25, Rec27, Mug20 and Rec10 proteins. Together with other chromosomal components, these proteins form linear elements (LinEs) that dictate the formation of DSBs and chromosomal structures. The authors first investigated how DSB hotspots along a chromosome communicate with each other by determining the effect of adding or deleting a hotspot on the frequency of DSBs at nearby hotspots. Adding a hotspot reduces DSBs at a nearby hotspot, and deleting a hotspot increases nearby hotspot DSBs; they call this "DSB competition." DSBs are not made independently at nearby hotspots. Instead, one DSB interferes with formation of another, such that doubly-broken DNA molecules are rarer than predicted from independence; they call this "DSB interference." Competition and interference are limited to ~200 kb regions. Additionally, a homolog of the human Ataxia-Telangiectasia mutated (ATM) protein, the Tel1 DNA damage-response protein kinase, was found to be necessary for both DSB and crossover interference, which had not previously been reported in S. pombe.
To explain how DSBs at one hotspot reduce DNA breaks at another hotspot within a limited chromosomal region, the authors reasoned it might be mediated by physical interactions between hotspots in the region. They assayed the proximity or clustering of hotspots bound by LinE proteins using a modified chromosome-conformation-capture (3C) technique, a method related to chromatin interaction analysis by paired-end tag sequencing (ChIA-PET). Kyle Fowler, a former research technician in the Smith lab and now a graduate student at UCSF, pioneered this effort. He modified the technique to allow the relatively infrequent physical interactions between hotspots to be detected. Across 603 hotspots, the authors found a preferential ligation of hotspots to sites <100 kb away, with the most frequent interactions between hotspots spaced 10-20 kb apart. Indeed, DSB hotspots communicate along chromosomes by forming physical interactions with other DSB hotspots bound by their determinants, the LinE proteins. The authors propose that limiting the number of breaks within a cluster produces both hotspot DSB interference and competition, which suggests a mechanism for how DSB formation can be limited within hotspot clusters. Specifically, they propose that only one DSB is made per cluster; in the absence of Tel1 kinase, DSB formation continues after the first DSB is made, resulting in loss of DSB and crossover interference. Because DSBs give rise to crossovers, their work “gives a plausible, supported solution to a 100-year-old problem central to genetics -- meiotic crossover interference” explained Dr. Smith. It also provides a means of communication along a chromosome.
When asked about immediate next steps, Dr. Smith revealed: “Our next (current) goal is to gather more evidence that crossover interference requires clustering of the LinE proteins.” Randy Hyppa, a Research Scientist in the Smith lab, added that in the long run: “It would be interesting to find out why the physical interactions are limited to 100-200 kb, and not between homologs.” Perhaps this solution to an intellectual problem would also operate within higher organisms? Dr. Smith hopes this study can lead the way. For now, he says: “As academics, we’re satisfied with finding a solution.”
Fowler KR, Hyppa RW, Cromie GA, Smith GR. 2018. Physical basis for long-distance communication along meiotic chromosomes. Proc Natl Acad Sci U S A Oct 2;115(40):E9333-E9342.
Funding was provided by the National Institutes of Health.