Meiotic recombination is essential for chromosome segregation at meiosis and fertility. It is initiated by programmed DNA double-strand breaks (DSBs) introduced by Spo11, a eukaryotic homologue of an archaeal topoisomerase (Topo VIA)1. Here we describe previously uncharacterized Spo11-induced lesions, 34 to several hundred base pair-long gaps, which are generated by coordinated pairs of DSBs termed double DSBs. Isolation and genome-wide mapping of the resulting fragments with single base-pair precision revealed enrichment at DSB hotspots but also a widely dispersed distribution across the genome. Spo11 prefers to cut sequences with similarity to a DNA-bending motif2, which indicates that bendability contributes to the choice of cleavage site. Moreover, fragment lengths have a periodicity of approximately (10.4n + 3) base pairs, which indicates that Spo11 favours cleavage on the same face of underwound DNA. Consistently, double DSB signals overlap and correlate with topoisomerase II-binding sites, which points to a role for topological stress and DNA crossings in break formation, and suggests a model for the formation of DSBs and double DSBs in which Spo11 traps two DNA strands. Double DSB gaps, which make up an estimated 20% of all initiation events, can account for full gene conversion events that are independent of both Msh2-dependent heteroduplex repair3,4 and the MutLγ endonuclease4. Because non-homologous gap repair results in deletions, and ectopically re-integrated double DSB fragments result in insertions, the formation of double DSBs is a potential source of evolutionary diversity and pathogenic germline aberrations.
This is a preview of subscription content
Subscription info for Chinese customers
We have a dedicated website for our Chinese customers. Please go to naturechina.com to subscribe to this journal.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Essential scripts are available at GitHub repository https://github.com/FranzKlein-lab/dDSB_tools.
Bergerat, A. et al. An atypical topoisomerase II from Archaea with implications for meiotic recombination. Nature 386, 414–417 (1997).
El-Deiry, W. S., Kern, S. E., Pietenpol, J. A., Kinzler, K. W. & Vogelstein, B. Definition of a consensus binding site for p53. Nat. Genet. 1, 45–49 (1992).
Martini, E. et al. Genome-wide analysis of heteroduplex DNA in mismatch repair-deficient yeast cells reveals novel properties of meiotic recombination pathways. PLoS Genet. 7, e1002305 (2011).
Marsolier-Kergoat, M. C., Khan, M. M., Schott, J., Zhu, X. & Llorente, B. Mechanistic view and genetic control of DNA recombination during meiosis. Mol. Cell 70, 9–20 (2018).
Keeney, S., Giroux, C. N. & Kleckner, N. Meiosis-specific DNA double-strand breaks are catalyzed by Spo11, a member of a widely conserved protein family. Cell 88, 375–384 (1997).
Robert, T. et al. The TopoVIB-like protein family is required for meiotic DNA double-strand break formation. Science 351, 943–949 (2016).
Vrielynck, N. et al. A DNA topoisomerase VI-like complex initiates meiotic recombination. Science 351, 939–943 (2016).
Liu, J., Wu, T. C. & Lichten, M. The location and structure of double-strand DNA breaks induced during yeast meiosis: evidence for a covalently linked DNA-protein intermediate. EMBO J. 14, 4599–4608 (1995).
Keeney, S. & Kleckner, N. Covalent protein-DNA complexes at the 5′ strand termini of meiosis-specific double-strand breaks in yeast. Proc. Natl Acad. Sci. USA 92, 11274–11278 (1995).
Neale, M. J., Pan, J. & Keeney, S. Endonucleolytic processing of covalent protein-linked DNA double-strand breaks. Nature 436, 1053–1057 (2005).
Hunter, N. in Homologous recombination (eds Aguilera, A. & Rotshtein, R.) 1–62 (Springer-Verlag, 2007).
Mimitou, E. P., Yamada, S. & Keeney, S. A global view of meiotic double-strand break end resection. Science 355, 40–45 (2017).
Pan, J. et al. A hierarchical combination of factors shapes the genome-wide topography of yeast meiotic recombination initiation. Cell 144, 719–731 (2011).
Lichten, M. & Goldman, A. S. Meiotic recombination hotspots. Annu. Rev. Genet. 29, 423–444 (1995).
Mohibullah, N. & Keeney, S. Numerical and spatial patterning of yeast meiotic DNA breaks by Tel1. Genome Res. 27, 278–288 (2017).
Borde, V. et al. Histone H3 lysine 4 trimethylation marks meiotic recombination initiation sites. EMBO J. 28, 99–111 (2009).
Zhu, X. & Keeney, S. High-resolution global analysis of the influences of bas1 and ino4 transcription factors on meiotic DNA break distributions in Saccharomyces cerevisiae. Genetics 201, 525–542 (2015).
Sommermeyer, V., Beneut, C., Chaplais, E., Serrentino, M. E. & Borde, V. Spp1, a member of the set1 complex, promotes meiotic dsb formation in promoters by tethering histone h3k4 methylation sites to chromosome axes. Mol. Cell 49, 43–54 (2012).
Lichten, M. Meiotic chromatin: the substrate for recombination initiation. Genome Dyn. Stab. 3, 165–128 (2008).
Buhler, C., Borde, V. & Lichten, M. Mapping meiotic single-strand DNA reveals a new landscape of DNA double-strand breaks in Saccharomyces cerevisiae. PLoS Biol. 5, e324 (2007).
Garcia, V., Gray, S., Allison, R. M., Cooper, T. J. & Neale, M. J. Tel1(ATM)-mediated interference suppresses clustered meiotic double-strand-break formation. Nature 520, 114–118 (2015).
Zhang, L., Kim, K. P., Kleckner, N. E. & Storlazzi, A. Meiotic double-strand breaks occur once per pair of (sister) chromatids and, via Mec1/ATR and Tel1/ATM, once per quartet of chromatids. Proc. Natl Acad. Sci. USA 108, 20036–20041 (2011).
Wu, T. C. & Lichten, M. Factors that affect the location and frequency of meiosis-induced double-strand breaks in Saccharomyces cerevisiae. Genetics 140, 55–66 (1995).
Fan, Q. Q., Xu, F., White, M. A. & Petes, T. D. Competition between adjacent meiotic recombination hotspots in the yeast Saccharomyces cerevisiae. Genetics 145, 661–670 (1997).
Xu, L. & Kleckner, N. Sequence non-specific double-strand breaks and interhomolog interactions prior to double-strand break formation at a meiotic recombination hot spot in yeast. EMBO J. 14, 5115–5128 (1995).
Ohta, K., Wu, T. C., Lichten, M. & Shibata, T. Competitive inactivation of a double-strand DNA break site involves parallel suppression of meiosis-induced changes in chromatin configuration. Nucleic Acids Res. 27, 2175–2180 (1999).
McKie, S. J., Maxwell, A. & Neuman, K. C. Mapping DNA topoisomerase binding and cleavage genome wide using next-generation sequencing techniques. Genes (Basel) 11, E92 (2020).
Spitzner, J. R. & Muller, M. T. A consensus sequence for cleavage by vertebrate DNA topoisomerase II. Nucleic Acids Res. 16, 5533–5556 (1988).
Buhler, C., Lebbink, J. H., Bocs, C., Ladenstein, R. & Forterre, P. DNA topoisomerase VI generates ATP-dependent double-strand breaks with two-nucleotide overhangs. J. Biol. Chem. 276, 37215–37222 (2001).
Gittens, W. H. et al. A nucleotide resolution map of Top2-linked DNA breaks in the yeast and human genome. Nat. Commun. 10, 4846 (2019).
Jang, Y. et al. Selection of DNA cleavage sites by topoisomerase II results from enzyme-induced flexibility of DNA. Cell Chem. Biol. 26, 502–511 (2019).
Murakami, H. & Nicolas, A. Locally, meiotic double-strand breaks targeted by Gal4BD-Spo11 occur at discrete sites with a sequence preference. Mol. Cell. Biol. 29, 3500–3516 (2009).
Claeys Bouuaert, C. et al. Structural and functional characterization of the Spo11 core complex. Nat. Struct. Mol. Biol. 28, 92–102 (2021).
Tokino, T. et al. p53 tagged sites from human genomic DNA. Hum. Mol. Genet. 3, 1537–1542 (1994).
Nagaich, A. K. et al. p53-induced DNA bending and twisting: p53 tetramer binds on the outer side of a DNA loop and increases DNA twisting. Proc. Natl Acad. Sci. USA 96, 1875–1880 (1999).
Lam, I. & Keeney, S. Nonparadoxical evolutionary stability of the recombination initiation landscape in yeast. Science 350, 932–937 (2015).
Carballo, J. A. et al. Budding yeast ATM/ATR control meiotic double-strand break (DSB) levels by down-regulating Rec114, an essential component of the DSB-machinery. PLoS Genet. 9, e1003545 (2013).
Hu, B. et al. Biological chromodynamics: a general method for measuring protein occupancy across the genome by calibrating ChIP-seq. Nucleic Acids Res. 43, e132 (2015).
Liu, L. F. & Wang, J. C. Supercoiling of the DNA template during transcription. Proc. Natl Acad. Sci. USA 84, 7024–7027 (1987).
Achar, Y. J., Adhil, M., Choudhary, R., Gilbert, N. & Foiani, M. Negative supercoil at gene boundaries modulates gene topology. Nature 577, 701–705 (2020).
Wendorff, T. J. & Berger, J. M. Topoisomerase VI senses and exploits both DNA crossings and bends to facilitate strand passage. eLife 7, e31724 (2018).
Zechiedrich, E. L. & Osheroff, N. Eukaryotic topoisomerases recognize nucleic acid topology by preferentially interacting with DNA crossovers. EMBO J. 9, 4555–4562 (1990).
Heldrich, J., Sun, X., Vale-Silva, L. A., Markowitz, T. E. & Hochwagen, A. Topoisomerases modulate the timing of meiotic DNA breakage and chromosome morphogenesis in Saccharomyces cerevisiae. Genetics 215, 59–73 (2020).
Schwacha, A. & Kleckner, N. Interhomolog bias during meiotic recombination: meiotic functions promote a highly differentiated interhomolog-only pathway. Cell 90, 1123–1135 (1997).
Johnson, D. et al. Concerted cutting by Spo11 illuminates meiotic DNA break mechanics. Nature https://doi.org/10.1038/s41586-021-03389-3 (2021).
Fowler, K. R., Sasaki, M., Milman, N., Keeney, S. & Smith, G. R. Evolutionarily diverse determinants of meiotic DNA break and recombination landscapes across the genome. Genome Res. 24, 1650–1664 (2014).
Tiemann-Boege, I., Schwarz, T., Striedner, Y. & Heissl, A. The consequences of sequence erosion in the evolution of recombination hotspots. Phil. Trans. R. Soc. Lond. B 372, 20160462 (2017).
Lukaszewicz, A., Lange, J., Keeney, S. & Jasin, M. De novo deletion mutations at recombination hotspots in mouse germlines. Preprint at https://doi.org/10.1101/2020.06.23.168138 (2020).
Xaver, M., Huang, L., Chen, D. & Klein, F. Smc5/6-Mms21 prevents and eliminates inappropriate recombination intermediates in meiosis. PLoS Genet. 9, e1004067 (2013).
Gao, R. et al. Proteolytic degradation of topoisomerase II (Top2) enables the processing of Top2·DNA and Top2·RNA covalent complexes by tyrosyl-DNA-phosphodiesterase 2 (TDP2). J. Biol. Chem. 289, 17960–17969 (2014).
Cortes Ledesma, F., El Khamisy, S. F., Zuma, M. C., Osborn, K. & Caldecott, K. W. A human 5′-tyrosyl DNA phosphodiesterase that repairs topoisomerase-mediated DNA damage. Nature 461, 674–678 (2009).
Mendoza, M. A., Panizza, S. & Klein, F. Analysis of protein-DNA interactions during meiosis by quantitative chromatin immunoprecipitation (qChIP). Methods Mol. Biol. 557, 267–283 (2009).
Haruki, H., Nishikawa, J. & Laemmli, U. K. The anchor-away technique: rapid, conditional establishment of yeast mutant phenotypes. Mol. Cell 31, 925–932 (2008).
Yue, J. X. et al. Contrasting evolutionary genome dynamics between domesticated and wild yeasts. Nat. Genet. 49, 913–924 (2017).
Sedlazeck, F. J., Rescheneder, P. & von Haeseler, A. NextGenMap: fast and accurate read mapping in highly polymorphic genomes. Bioinformatics 29, 2790–2791 (2013).
Brar, G. A. et al. High-resolution view of the yeast meiotic program revealed by ribosome profiling. Science 335, 552–557 (2012).
Cheng, Z. et al. Pervasive, coordinated protein-level changes driven by transcript isoform switching during meiosis. Cell 172, 910–923 (2018).
Duroc, Y. et al. Concerted action of the MutLβ heterodimer and Mer3 helicase regulates the global extent of meiotic gene conversion. eLife 6, e21900 (2017).
Poplin, R. et al. Scaling accurate genetic variant discovery to tens of thousands of samples. Preprint at https://doi.org/10.1101/201178 (2017).
McKenna, A. et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 20, 1297–1303 (2010).
Pedregosa, F. et al. Scikit-learn: machine learning in Python. JMLR 12, 2826–2830 (2011).
Benjamini, Y. & Speed, T. P. Summarizing and correcting the GC content bias in high-throughput sequencing. Nucleic Acids Res. 40, e72 (2012).
Mancera, E., Bourgon, R., Brozzi, A., Huber, W. & Steinmetz, L. M. High-resolution mapping of meiotic crossovers and non-crossovers in yeast. Nature 454, 479–485 (2008).
Panizza, S. et al. Spo11-accessory proteins link double-strand break sites to the chromosome axis in early meiotic recombination. Cell 146, 372–383 (2011).
Claeys Bouuaert, C. et al. DNA-driven condensation assembles the meiotic DNA break machinery. Nature 592, 144–149 (2021).
Mosconi, F., Allemand, J. F., Bensimon, D. & Croquette, V. Measurement of the torque on a single stretched and twisted DNA using magnetic tweezers. Phys. Rev. Lett. 102, 078301 (2009).
Lipfert, J., Kerssemakers, J. W., Jager, T. & Dekker, N. H. Magnetic torque tweezers: measuring torsional stiffness in DNA and RecA-DNA filaments. Nat. Methods 7, 977–980 (2010).
We thank S. Zich and M. Galova for technical assistance, S. Keeney and K. Nasmyth for sharing strains, P. Schloegelhofer, N. Kleckner, M. Xaver, J. Marko, J. Roca for stimulating discussions, and N. Kleckner, M. Xaver, C. Campbell for comments on the manuscript. We thank M. Neale for discussing results ahead of publication. Furthermore, we thank A. v. Haeseler and the Center for Integrative Bioinformatics of Vienna (CIBIV) for hosting D. C. and providing the computer cluster infrastructure. This work was funded by SFB grant F34 and by DK W1238-B20 from the Austrian Science Foundation.
The authors declare no competing interests.
Peer review information Nature thanks Bernard de Massy and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer review reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
Extended Data Fig. 1 DNA fragments released in meiosis are characterized with single-nucleotide precision.
a, In a rad50S mutant about one-third of the dDSB fragments are robustly resistant to all Exo V concentrations analysed. qPCR of dDSB signals at the YCR047 hotspot (dark blue) or the YCR011 cold region (light blue) using primer pairs PP7 (248 bp) or PP2 (130 bp), respectively (blue bars represent mean values, error bars denote s.d. of three independent experiments, black dots indicate individual values). Digests were performed at 37 °C except for one negative control (4 °C). b, dDSB signals peak during DSB formation in wild-type synchronized meiotic cultures. qPCR as in a (mean and s.d. of two independent experiments are shown, individual values are represented as circles and diamonds). Signals were depleted in the untagged control, the catalytically inactive spo11Y135F mutant and the DSB formation-impaired rec102∆ mutant. c–e, Internal quality controls show single-nucleotide precision and striking signal-to-noise ratio of Protec-seq. c, Diagram explaining the Spo11 signature and the representation of dDSB fragments as arcs. A strong break site produces dDSB fragments in both orientations (grey and green) with a 1-nt offset between Watson (W) and Crick (C) 5′ ends reflecting the 2-nt overhang produced by Spo118 (Spo11 signature). Differently oriented fragments arise from independent cleavage events at different rates. Thus fragments originating at the same break site but ending at opposite sides may show different depths (dDSB asymmetry). d, ‘Cut site fingerprints’ demonstrate single-nucleotide precision in Protec-seq. Cut site fingerprints are single-nucleotide resolution patterns with a window of a few nucleotides around a dominant, isolated Watson signal. Inaccurate determination of 5′ ends would result in a spill over from the dominant signal. Left, averaged fingerprint from 11 cut sites on the Watson strand selected for strength and isolation, with Crick signals serving as independent readout. Watson signals are depicted as filled columns with positive values (percentage of the dominant Watson signal), Crick signals as empty columns with negative values (percentage of the dominant Crick signal). Right, filled and empty columns as in the left panel, but averaged over eight different experiments, involving six different mutant backgrounds at the same cut site (chromosome IV, 824,137). e, Genome-wide detection of the Spo11 signature for assessment of the Protec-seq accuracy. Quantification of the Watson–Crick offset for strong and isolated peaks (Methods) show a high percentage of the 1-nt Watson–Crick offset in the dDSB samples. Wild-type (n = 3,272), tel1∆ rec114-8A (n = 3,726), mre11S (n = 1,155), rad50S (n = 1,947), com1/sae2∆ (n = 1,674), and for comparison Spo11-oligos (n = 1,964, derived from ref. 15) are shown (n denotes the number of peaks that fulfil the selection criteria). f, The architecture of dDSB hotspots is well conserved across resection mutants and the tel1∆ checkpoint mutant. Arcs represent dDSB fragments at a hotspot around 824,100 on chromosome IV, with grey shades corresponding linearly to the fragment depths. Plots in the bottom right show dDSB fragments starting at a single position (116,869, chromosome IV) for wild-type and rad50S strains. Orange bars represent Spo11-oligo 5′ ends. All samples are from t4.
a, Biological repeats as well as samples with similar phenotype show high correlation at 10-nt resolution. 5′ end log2-transformed counts are shown. The corresponding Pearson correlation coefficients (R) are indicated in the bottom right corner of the scatter plots. The compared samples and time points are indicated at the left and bottom axes. Top right, to test whether the differences between rad50S and wild-type are mainly due to differential degradation, sub-samples of rad50S dDSB fragments were selected according to the dDSB length distribution of wild-type cells. The resulting improvement of R from 0.88 (for unselected rad50S versus wild-type dDSB) to 0.93 supports this assumption. b, Positional overlaps of dDSB 5′ ends with Spo11-oligo 5′ ends with 1 nt tolerance. OO, both dDSB fragments ends overlap; ON, one end overlaps; NN, both ends do not overlap. Filled bars represent total fragment counts as a percentage of the total; empty bars represent deduplicated counts. The last panel shows the percentages of the Spo11-oligo sample overlapping with wild-type dDSB (O, overlap; N, no overlap). All samples are from 4 h in SPM (t4). c, dDSB fragment and Spo11-oligo profiles from chromosome III. Genome-wide Pearson correlation coefficients with the wild-type t4 profile at a resolution of 10 nt are indicated on the right side of each panel. d, Left, frequency of (d)DSB sites of certain depths (log10 bins) as a percentage of total recovered break sites. Right, the frequency of (d)DSBs of certain depths (log10 bins) as a percentage of total (d)DSBs. The depths are binned into 1, 2–10, 11–100, 101–1000, 103+1–104, 104+1–105. All samples are from t4.
a, Spo11 cleavage preference and internal GC accumulation in S. cerevisiae as in Fig. 2a, but for 64-nt dDSB fragments (wild-type, t4; n = 164,104, with n being the number of independently identified fragments of a given length). The GC content of 47.8% is lower than for the 34-nt fragments (57.9%), but markedly above the genome average (38.3%). The GC enrichment is not caused by preferential resistance to degradation (data not shown), or by preferential PCR amplification, as the enrichment is robust against deduplication, and GC-rich DNA is known to be underrepresented in over-amplified libraries62. Separate analysis of the 5′ and the 3′ ends of dDSB fragments reveals an asymmetry of base preferences relative to the cleavage axes. The nucleotide preferences inside the fragment (at positions 1, 2 and 13 relative to the cleavage axes) are stronger than their counterparts that flank the dDSB fragment (at positions −1, −2 and −13). All features are robust to deduplication (data not shown). b, c, Cleavage preference and internal GC accumulation in S. kudriavzevii for dDSB fragment lengths of 64 nt (b; 50.4% GC, n = 12,004) and 34 nt (c; 59.4% GC, n = 853) are very similar to their counterparts in S. cerevisiae. Preferential excision of GC-rich sequences could help to limit GC accumulation at yeast hotspots63. d, In tel1∆ rec114-8A double mutants, the periodicity of dDSB fragment lengths was detectable for up to 335 nt. Peak lengths are indicated by pink bars and numbers on the top; valleys are indicated by blue bars. Peaks and valleys were called automatically by a custom algorithm after smoothing with bandwidth = 3 and resolution = 1 (R ‘ksmooth’). e, S. kudriavzevii com1/sae2∆t6 displays a similar distribution of dDSB fragment lengths as S. cerevisiae resection mutants and the same 10.4n + 3 nt periodicity as all S. cerevisiae dDSB samples. f, Quantification of dDSB signal from ref. 15 (gel separation). Left, on the basis of their length distribution, signals are interpreted either as DSB oligonucleotides (dark blue, A1) or dDSB fragments (cyan, A2). Right, diagram illustrating the rationale for estimation of the dDSB fraction F. Each single DSB is represented by two oligonucleotides, one (approximately 30 bp) gives rise to the A1 signal, the other one (14 bp long) is not visible on this gel. dDSB fragments smaller than 70 bp produce two long oligonucleotides with periodic lengths and two with Spo11-oligo lengths that can contribute to the 30-nt peak (A1) (Methods). g, Periodicity does not require nucleosomes. The length periodicity is observed with dDSB fragments lying with both ends in nucleosome-occupied (NN) or in nucleosome-depleted (DD) regions. The nucleosome map is derived from ref. 13 (Methods). h, Table of peak fragment lengths <108 bp averaged from 21 dDSB samples of various genetic background, resulting in a mean peak to peak distance of 10.4 ± 0.36 nt and helix lengths ranging from 11.33 bp (for 34-nt long fragments) to 10.7 bp (for 107 nt).
Extended Data Fig. 4 Top2 ChIP–seq profiles indicate accumulation of topological stress at hotspots at the time of DSB formation.
a, (d)DSB 5′ ends located within ±1 kb of TSSs from assembled transcripts from previously published RNA-seq data of 3–6 h of meiosis (n = 7,913), stratified by downstream high (H), medium (M) and low (L) transcriptional stress (expression level times length). Wild-type dDSB, rad50S dDSB, Spo11-oligos, and spo11Y135F dDSB are shown. The corresponding Spearman correlation coefficients (rho) are shown. b, dDSB signals of rad50S (left) and spo11Y135F (right) located within ±1 kb of a Top2 peak (n = 10,689; from Top2 ChIP–seq experiment at t4, with depth >100) were stratified into 12 groups according to the corresponding Top2 peak intensities. The Spearman correlation coefficients (rho) are indicated. c, Examples of calibrated Top2 ChIP–seq signals accumulating at several kilobase-wide hotspot clusters. Top2 t4 (dark green), untagged control (grey), Spo11-oligos (orange bars), Watson transcripts and open-reading frames (ORFs) (black), Crick transcripts and ORFs (green) are shown. d, Top2–Myc peaks at TSSs, stratified by high, medium and low transcriptional stress (expression level times length). From left to right: all TSS (n = 7,913), divergent (n = 3,930), tandem (n = 2,109) and convergent sites (n = 3,900). The corresponding Spearman correlation coefficients (rho) are shown. e, Typical examples of Top2 occupancy at divergent, tandem or convergent sites with corresponding DSB signals. Colour labels as in c.
a, Specific and complete nuclear depletion of Rec104 within 30 min by the ‘anchor away’ technique53. In situ staining of wild-type cells at t4.5 stained with an anti-haemagglutinin (HA) antibody to visualize Rec104-FRB–HA3. The left two panels show nuclei stained either with DAPI or anti-HA antibody in yeast cells without rapamycin, and Rec104 is localized in the nucleus. The right two panels are as in the left two panels, but after the addition of rapamycin (1 μg ml−1) for 30 min. Rec104 is in the cytoplasm and nuclei are visibly depleted of Rec104 (blue arrows). The experiment was repeated six times with n = 100 nuclei inspected. b, Spo11 is retained in the nucleus during the anchor-away technique. In situ staining of Rec104-FRB-HA3 with an anti-HA antibody (red), Spo11–Myc18 (green) or DAPI (cyan) in rad50S cells without rapamycin (left) or with rapamycin added at t4 for 30 min (right). The experiment was repeated six times with n = 100 nuclei inspected. c, dDSB fragments from wild-type cells with one end at a single position (399,594) on chromosome VII before (grey) and after (red) 30 min of DSB inhibition by Rec104 depletion. Almost all fragments greater than 100 bp are lost, whereas fragments smaller than 70 bp persist. d, Length-independent decay of dDSB fragments in rad50S Rec104-FRB-HA3 cells. The fold change of (calibrated) fragment length levels 60 min versus 0 min without rapamycin (dark cyan), 30 min versus 0 min plus rapamycin (light violet), and 60 min versus 0 min plus rapamycin (dark violet) are shown. For one ratio (60 min versus 0 min minus rapamycin), the joint fragment depths are shown to indicate the sample number per length (light grey). A fold change of one indicates stability; a fold change of 0.5 indicates a half-life of 30 min. Fragments with disfavoured lengths (‘valleys’) are slightly less stable overall.
Extended Data Fig. 6 dDSB fragments reveal preferential gap formation at sites of Msh2-independent full gene conversions.
a, Raw data of publicly available S288C/SK1 hybrid octads were re-aligned to the parental reference genomes and re-analysed to create a high-precision and high-confidence recombination event map, with focus on full gene conversions (‘6:2’ events) (Methods, Supplementary Table 4). Average number of crossover (CO), non-crossover (NCO) and double crossover (dCO) events of msh2∆ (four octads with on average 211.8 ± 32.8 events), msh2∆ mutLγ∆ (four octads, 189.0 ± 8.8 events), and msh2∆ exo1∆ (two octads, 148.5 ± 29.0) octads are depicted by doughnut plots, in which the outer ring represents all events, and the inner circle denotes the 6:2-containing events (with a radius proportional to their fraction). Among 6:2 events, crossovers are markedly overrepresented in msh2∆ mutants (P = 4.3 × 10−12), but not in msh2∆ mutLγ∆ double mutants (P = 0.54 Fisher’s exact test) as the MutLγ-independent full gene conversion events show an only slightly enhanced tendency for crossovers (29.1% of 6:2 versus 25.5% of total events). Because the preferential association of 6:2 events with crossovers disappears when MutLγ is defective, the crossover-associated 6:2 events seem to be generated by the MutLγ pathway, not by dDSBs. Deletion of another MutLγ component, Exo1, in conjunction with Msh2, yields even fewer full gene conversion events (14.8%), because it also affects the early stages in both DSB and dDSB processing, including the interhomologue bias44, and thus affects 6:2 formation from all pathways. b, c, As in Fig. 4f, g, but with msh2∆ octad data. Msh2-independent 6:2 events (n = 349 from four octads) have significantly higher dDSB (gap) coverage than randomly distributed events. In b, the cumulative fraction of Msh2-independent 6:2 events is plotted against the number of identified gaps that completely overlap them. Plots based on gap distributions from (left to right) wild-type, rad50S and spo11Y135F strains are shown. Results from distributing 6:2 events 1,000× randomly are included (light blue dots). Values above random indicate a positive correlation between the incidences of 6:2 events and gap probabilities. P values were determined by Wilcoxon rank test and Bonferroni multiple-test correction. In c, the dDSB fragment coverage of wild-type (top) and rad50S (bottom) strains per observed or 1,000× randomly distributed (light blue) 6:2 events are plotted. Median values of wild-type and rad50S were significantly different from random (Mood’s median test, P = 7.0 × 10−14 and 4.6 × 10−30, respectively). The central lines indicate the median, the boxes indicate the interquartile range, and whiskers span maximally 1.5 times the interquartile range.
As mounting evidence stresses similarities between the Spo11 complex and type IIB topoisomerases1,33,41 our ‘topological trap’ model strives to explain how Spo11 induces (d)DSBs at DNA crossings. To explain the encounter between the complex and a DNA crossing, we suggest that Spo11 traps a second strand in the tip of bent DNA, a principle that if corroborated, could explain how type II topoisomerases arrive at crossings in general. a, When axis-anchored64 Spo11 encounters a DNA strand (G-segment, grey), the tethering of the G-strand to the presumed Mer2, Mei4, Rec114 dependent condensate65 forms a physical barrier, which can capture an additional strand. Spo11 bends the DNA at the binding site, but does not cleave the DNA. This will draw a second strand (T-segment, brown) into the tip of the bend, and provoke the formation of a DNA crossing. After the arrival of the second strand (T-segment), Spo11 undergoes a conformational change, and entraps both strands. This movement accompanies the cleavage reaction as for topoisomerase VI41, which results in the breakage of both single strands of the G-segment, setting the T-segment free. b, If the G-segment interacts with two or more Spo11 complexes before it encounters a T-strand, the formation of a gap and the corresponding release of a dDSB fragment can occur. Spo11 prefers motifs that mediate DNA bending in defined angles35. Two such bending angles will add up to form a flat U-shape (second panel from top), only if these bending motifs are in phase, resulting in cleavage products of periodic length. DNA between the two tethering Spo11 complexes is in close proximity to the condensate—an additional constraint that may cause the observed GC preference in short fragments. The ensuing steps are completely analogous to single DSB formation in a, with one additional requirement. A dDSB will only occur when a T-strand crosses both G-strand tethers. As shown in the second and third panel, it is important that the T-strand enters below both G-strand arms, which triggers the conformational change of the Spo11 complexes to clamp down the crossings and cleave the G-segment in a concerted fashion. If the T-segment crosses below one, but above the other, grey arm, only a single crossing is trapped and the structure resolves with a single DSB. The two DNA crossings between the G- and T-strands generate the dDSB irrespective of how many Spo11 complexes tether the G segment. Because of the torsional stiffness of DNA at that scale, measured by the twist persistence length, phosphodiester bonds are presented at the same angle to Spo11 at periodic distances (top panel in b). The twist persistence length for dsDNA ranges from 120 to 300 bp, increasing with the stress acting on DNA66,67. These numbers are in good agreement with the maximal range (around 335 bp) over which we could detect periodicity (Extended Data Fig. 3d).
This file contains Supplementary Discussions 1-3.
Supplementary Table 1 S. cer. SK1 ASM205788v1 gene annotation (gtf format). Sequence data were downloaded from SGD (R64-2-1, version 17. 1. 2017, with non-coding and other features) and the Ensembl database (release 98, R64-1-1), and subsequently re-aligned and merged.
Supplementary Table 2 S. cer. SK1 ASM205788v1 transcript annotation as obtained from re-alignment and assembly of RNA-seq data from GEO GSE34082 and GSE108778, including transcript start sites, lengths and expression levels from 3 hrs – 6 hrs after initiation of meiosis.
Supplementary Table 3 S. cer. SK1 ASM205788v1 Spo11-oligo map, wild type t4. Raw data was downloaded from GEO GSE84696 (SRR3942949, replicate 2), re-processed and re-aligned (unique alignments only). 5’-end counts of Watson and Crick strands are reported separately.
Genotype and recombination event predictions with S. cer. ASM205788v1 and R64 coordinates for S288C/SK1 hybrid octads with msh2Δ (4 octads; raw data from SRA accession SRP075437), msh2Δ/mlh1Δ or msh2Δ/mlh3Δ, and msh2Δ/exo1Δ (2 octads each; SRP111430) background. Data for each octad (in total 10) are presented in separate Excel sheets.
About this article
Cite this article
Prieler, S., Chen, D., Huang, L. et al. Spo11 generates gaps through concerted cuts at sites of topological stress. Nature 594, 577–582 (2021). https://doi.org/10.1038/s41586-021-03632-x