Control of meiotic double-strand-break formation by ATM: local and global views

Figures & data

Figure 1. (A) Meiotic DSB formation. The SPO11-TOPVIBL complex induces DSBs at genomic locations enriched for H3K4me3 and H3K36me3, histone modifications deposited by PRDM9. DSBs activate the ATM kinase, which suppresses SPO11 from further DSB formation by an as yet unknown mechanism. In response to DSBs, ATM phosphorylates histone H2AX (γ-H2AX) in surrounding chromatin. (B) Meiotic DSB processing and detection. SPO11, covalently attached to DNA, is removed by endonucleolytic cleavage by the MRE11 complex. This reaction releases SPO11 bound to short oligonucleotides (SPO11-oligo complexes) and exposes short single-stranded DNA (ssDNA) tails at DSB sites. These tails are further extended by resection and become substrates for DMC1 and RAD51 strand exchange proteins, which mediate homologous recombination. SPO11 oligos, ssDNA, and DMC1/RAD51-coated ssDNA can be analyzed by deep sequencing to map genome-wide DSB positions (see main text for references)

Figure 2. (A) Local DSB patterning. (top) Examples of SPO11-DSB hotspots on chromosome 1 in Atm wild type (wt) and Atm null (data from ref [17].). Note the emergence of a new hotspot in the absence of ATM. (bottom) Overlays of Atm wt and Atm null hotspot profiles show a wider distribution of SPO11 oligos around the center of hotspot B in Atm null. Methods: As previously described [17], hotspots were defined as regions in smoothed SPO11-oligo maps with uniquely mapped reads exceeding 50 times the genome average. The hotspot center was defined as the coordinate of the maximum SPO11-oligo value within the hotspot. Here, SPO11-oligo maps were smoothed with a 101-bp Hann filter. RPM denotes reads per million.(B) Schematic of MRE11-independent formation of SPO11 oligos by introduction of nearby DSBs (see main text)

Figure 3. Large-scale DSB patterning

(A) Schematic illustrating ATM-mediated DSB distributions across chromosomal domains. (left) Weaker DSB suppression in a DSB-rich domain. In a wild-type cell, this domain contains 10 potential sites for DSB formation (DSB hotspots), of which 5 experience DSBs. In an Atm-null cell, new hotspots emerge (see main text and panel C below), so this domain contains 20 DSB hotspots. Of these, 10 experience DSBs, which gives a 2-fold increase in DSB numbers when ATM is missing. (right) Stronger DSB suppression in a DSB-poor domain. In a wild-type cell, this domain contains 3 DSB hotspots, of which 1 experiences a DSB. In an Atm-null cell, this domain contains 9 DSB hotspots, of which 4 experience DSBs. This gives a 4-fold increase in DSB numbers when ATM is missing. Note that the ratio of new hotspots to matched hotspots (shared between Atm null and Atm wt) is higher in the DSB-poor domain. Strengths of individual hotspots are not indicated (see main text).(B) Representative SPO11-oligo maps of a DSB-rich (left) and DSB-poor (right) domain at the distal end of chromosome 18 (data from ref [17].). The subtelomeric region of chromosome 18 was reported to display frequent meiotic crossing over [141]. In the absence of ATM, the sum of normalized SPO11-oligo counts increases more in the DSB-poor domain than in the DSB-rich domain, 2.4-fold and 1.3-fold respectively. This is because the existing weak hotspots in the DSB-poor domain increase more than the existing stronger hotspots in the DSB-rich domain, on average 5.8-fold in poor and 1.5-fold in rich domain. In addition, more new weak hotspots appear in the DSB-poor domain, such that the ratio of new to matched hotspots in the poor domain is higher (DSB-poor domain: 22 new to 10 matched hotspots; DSB-rich domain: 38 new to 34 matched hotspots). SPO11-oligo maps were smoothed with a 10001-bp Hann filter. RPM denotes reads per million.(C) Behavior of 5-Mb domains of chromosome 18 in relation to SPO11-oligo counts and DSB hotspot numbers in Atm wt and Atm null. (top left) DSB-poor domains (lower SPO11-oligo counts, colder domains) contain fewer hotspots. (top right) In the absence of ATM, strengths of domains with fewer hotspots in wild type tend to increase more. (bottom left) In the absence of ATM, normally colder domains tend to show more new hotspots relative to the number of matched hotspots in the domain. (bottom right) Domains with higher ratio of new to matched hotspots tend to be more strongly suppressed in the wild type by ATM. r represents Pearson’s correlation.

Figure 4. Hypothetical model for ATM-mediated DSB control. (top) Chromatin loops containing DSB hotspots are anchored to the chromosome axis. Before DSB formation, the IHO1-REC114-MEI4 complex and the axis-structural component HORMAD1 assemble along the chromosome axis. The IHO1-REC114-MEI4 complex together with HORMAD1 promotes DSB formation by SPO11. At present, physical interactions between IHO1, REC114, and MEI4 are inferred only from yeast two-hybrid analyzes [118,120], but further support for these interactions comes from extensive two-hybrid and coimmunoprecipitation studies of the orthologous proteins in budding and fission yeasts: Mer2-Rec114-Mei4 [142–145] and Rec15-Rec7-Rec24 [146,147], respectively. (bottom) A DSB is formed preferentially at a DSB hotspot in a chromatin loop tethered to the axis. In response to the DSB, ATM is activated and phosphorylates REC114, HORMAD1, and/or other proteins. Phosphorylation events separately or in combination inhibit additional DSB formation at the same chromatin loop (1) or at the adjacent loop (2) which is pre-activated for DSB formation, by inhibiting or destabilizing of the IHO1-REC114-MEI4 complex. HORMAD1 phosphorylation in the vicinity of the DSB prevents the assembly of the IHO1-REC114-MEI4 complexes and thus DSB formation at other nearby chromatin loop (3)

Lange J, Yamada S, Tischfield SE, et al. The landscape of mouse meiotic double-strand break formation, processing, and repair. Cell. 2016;167:695–708e16. PMID:27745971.

 PubMed Web of Science ®Google Scholar

Froenicke L, Anderson LK, Wienberg J, et al. Male mouse recombination maps for each autosome identified by chromosome painting. Am J Hum Genet. 2002;71:1353–1368. PMID:12432495.

 PubMed Web of Science ®Google Scholar

Stanzione M, Baumann M, Papanikos F, et al. Meiotic DNA break formation requires the unsynapsed chromosome axis-binding protein IHO1 (CCDC36) in mice. Nat Cell Biol. 2016;18:1208–1220. PMID:27723721.

 PubMed Web of Science ®Google Scholar

Kumar R, Bourbon HM, de Massy B. Functional conservation of Mei4 for meiotic DNA double-strand break formation from yeasts to mice. Genes Dev. 2010;24:1266–1280. PMID:20551173.

 PubMed Web of Science ®Google Scholar

Arora C, Kee K, Maleki S, et al. Antiviral protein Ski8 is a direct partner of Spo11 in meiotic DNA break formation, independent of its cytoplasmic role in RNA metabolism. Mol Cell. 2004;13:549–559. PMID:14992724.

 PubMed Web of Science ®Google Scholar

Maleki S, Neale MJ, Arora C, et al. Interactions between Mei4, Rec114, and other proteins required for meiotic DNA double-strand break formation in Saccharomyces cerevisiae. Chromosoma. 2007;116:471–486. PMID:17558514.

 PubMed Web of Science ®Google Scholar

Steiner S, Kohli J, Ludin K. Functional interactions among members of the meiotic initiation complex in fission yeast. Curr Genet. 2010;56:237–249. PMID:20364342.

 PubMed Web of Science ®Google Scholar

Miyoshi T, Ito M, Kugou K, et al. A central coupler for recombination initiation linking chromosome architecture to S phase checkpoint. Mol Cell. 2012;47:722–733. PMID:22841486.

 PubMed Web of Science ®Google Scholar