LinE proteins lead the way in DNA break hotspot determination

From the Smith Lab, Basic Sciences Division

Like two boxers returning to their corners at the end of a round, it is the imperative of duplicated chromosomes to segregate to opposite sides of a cell during cell division. But within this process lies a striking paradox – to faithfully segregate, these chromosomes must first be stuck together. This sticking together provides tension as the chromosomes begin to head to their respective corners, and “tension signals that the centromeres are going to opposite poles of the cell,” describes Dr. Gerry Smith, a professor in Fred Hutch’s Basic Sciences Division and UW/Fred Hutch Cancer consortium member. During mitosis, chromosomes are adhered together as though by glue. During meiosis, the connection is much more intimate – the two chromosomes are broken apart and then spliced together in a process called crossing over that results in their direct physical coupling. In a new research article in the journal Nucleic Acids Research, Dr. Smith’s group examined how these crossover events are regulated in yeast cells.

Crossing over begins with the formation of DNA double-strand breaks (DSBs), which can then be repaired to restore normal chromosome structure or be recombined to generate crossovers. “DSB’s are not uniformly distributed across the chromosomes. Rather, there are sites, called DSB hotspots, at which DSBs occur preferentially”, wrote Dr. Smith’s group. A group of proteins called LinE proteins – namely Rec25, Rec27, and Mug20 – localize to hotspots and are required for DSB formation at these locations. However, a lack of understanding of how LinE proteins are recruited to hotspots has “precluded testing whether LinE protein binding to a chromosomal site is sufficient to create a DSB hotspot,” the authors wrote. To overcome this limitation, they devised a clever new strategy to promote binding of LinE proteins to a site of their choosing. They introduce a specific DNA sequence, called lacO, into a region of the fission yeast genome that lacks hotspot activity, and then fused the LinE proteins to a protein from E. coli, LacI, that binds the lacO sequence. The resulting forced localization of LinE proteins caused large increases in DSB formation and up to a ten-fold increase in recombination at the lacO site. Further, recombination at the lacO site used the same molecular machinery as at endogenous hotspots, indicating that recruitment of LinE proteins is sufficient to create a DSB hotspot. The authors did however, note a discrepancy between the frequencies of DSB formation and recombination at this site – they observed much less recombination than expected for the number of DSBs generated at this site, suggesting that many DSBs were resolved without generating crossovers. This finding also indicated that their artificial hotspot was, perhaps not surprisingly, not a perfect replica of an endogenous hotspot.

The lacO/LacI strategy further allowed the authors to examine another intriguing but poorly understood characteristic of DSB hotspots. “DSB hotspots do not act independently – they interact with neighboring hotspots”, the authors explained. These interactions include DSB competition, in which introduction of a new hotspot reduces the frequency of DSB formation at nearby hotspots, and DSB interference, in which the formation of a DSB at one hotspot decreases the probability of formation of a second DSB at a nearby hotspot. “In cells with wild-type LinEs, DSBs show both competition and interference”, the authors noted. However, in a result they described as “remarkable”, the group found that the lacO hotspot exhibited almost no competition with nearby endogenous hotspots; in fact, one nearby hotspot actually became more active. Conversely, the lacO hotspot did display DSB interference with a nearby hotspot, highlighting the separability of these two forms of hotspot interactions.

Dr. Smith was eager to highlight the level of teamwork involved in this project. “This work started with Josh Cho, a technician, about 8 years ago. He ambitiously made and tested lots of fusions of LacI protein to each of the linear element proteins and made several lacO insertions into a gene (ade6) well-suited to measure recombination and DNA break formation. He found a range of activities and showed they worked as expected. Unexpectedly, postdoc Mridula Nambiar found that there were 10 times more DNA breaks than expected from the recombination frequencies. Research scientist Randy Hyppa then showed, with lots of work using techniques he developed, that the high break:low recombination ratio is due to higher-than-expected repair of DNA breaks with the sister, which cannot give genetic recombinants.  And he showed the DNA break hotspot made by Josh has DNA break interference but not competition.” Following up, Hyppa explained that he is particularly excited by the group’s observation that the lacO hotspot does not exhibit DSB competition. “I think the most significant contribution is the data can help us reveal what factors are important for establishing DNA double-strand break (DSB) hotspot competition. It’s been long known that introducing a strong DSB hotspot will suppress existing nearby hotspots, but it has remained unclear how this works. It seems LinEs that are not loaded by the usual mechanism result in DSB hotspots that do not show competition. This insight will allow us to target the proteins complexes that are involved in DSB competition, which we currently believe to be the DNA cohesin and condensin complexes. We have experiments underway to examine DSB competition in cohesin mutants, and we are also looking at how the cohesin complex interacts with the LinE proteins.”

hotspot competition model
Model for DSB hotspot formation and interference. Image provided by Dr. Gerry Smith

This work was supported by the National Institutes of Health

Fred Hutch/UW Cancer Consortium member Gerry Smith contributed to this work

Hyppa RW, Cho JD, Nambiar M, Smith GR. Redirecting meiotic DNA break hotspot determinant proteins alters localized spatial control of DNA break formation and repair. Nucleic Acids Res. 2022 Jan 25;50(2):899-914. doi: 10.1093/nar/gkab1253. PMID: 34967417; PMCID: PMC8789058.