A missense in HSF2BP causing primary ovarian insufficiency affects meiotic recombination by its novel interactor C19ORF57/BRME1

Primary Ovarian Insufficiency (POI) is a major cause of infertility, but its etiology remains poorly understood. Using whole-exome sequencing in a family with three cases of POI, we identified the candidate missense variant S167L in HSF2BP, an essential meiotic gene. Functional analysis of the HSF2BP-S167L variant in mouse showed that it behaves as a hypomorphic allele compared to a new loss-of-function (knock-out) mouse model. Hsf2bpS167L/S167L females show reduced fertility with smaller litter sizes. To obtain mechanistic insights, we identified C19ORF57/BRME1 as a strong interactor and stabilizer of HSF2BP and showed that the BRME1/HSF2BP protein complex co-immunoprecipitates with BRCA2, RAD51, RPA and PALB2. Meiocytes bearing the HSF2BP-S167L variant showed a strongly decreased staining of both HSF2BP and BRME1 at the recombination nodules and a reduced number of the foci formed by the recombinases RAD51/DMC1, thus leading to a lower frequency of crossovers. Our results provide insights into the molecular mechanism of HSF2BP-S167L in human ovarian insufficiency and sub(in)fertility.


Introduction
The process of gametogenesis is one of the most complex and highly regulated differentiation programs. It involves a unique reductional cell division, known as meiosis, to generate highly specialized cells: the gametes. Indeed, the outcome of meiosis is the production of oocytes and spermatozoa, which are the most distinctive cells of an adult organism and are essential for the faithful transmission of the genome across generations.
The meiotic division is an orderly process that results in the pairing and synapsis of homologous chromosomes and crossover (CO) formation, which ultimately enable homologous chromosomes segregation (Hunter, 2015;Loidl, 2016;Zickler and Kleckner, 2015). In mammals, pairing of homologs is dependent on the repair of self-induced double-strand breaks (DSBs) during prophase I by homologous recombination (Handel and Schimenti, 2010) and it leads to the intimate alignment of homologous chromosomes (synapsis) through the zipper-like synaptonemal complex (SC) (Cahoon and Hawley, 2016). The SC is a proteinaceous tripartite structure that provides the structural framework for DSBs repair , as epitomized by the tight association of the recombination nodules (RNs, multicomponent recombinogenic factories) and the axial elements of the SC (Zickler and Kleckner, 2015).
Meiotic DSBs repair is an evolutionarily conserved pathway that is highly regulated to promote the formation of at least one CO per bivalent. This chromosome connection between bivalents through chiasmata is required for a correct reductional division. As other DNA repair processes, proper meiotic recombination is essential for genome stability and alterations can result in infertility, miscarriage and birth defects (Geisinger and Benavente, 2017;Handel and Schimenti, 2010;Webster and Schuh, 2017).
Infertility refers to failure of a couple to reproduce and affects 10-15% of couples (Isaksson and Tiitinen, 2004). Infertility can be due to female factors, male factors, a combination of both or to unknown causes, each category representing approximately 25% of cases (Isaksson and Tiitinen, 2004;Matzuk and Lamb, 2008). There are several underlying causes and physiological, genetic and even environmental and social factors can play a role. Forward and reverse genetic analyses in model organisms have identified multiple molecular pathways that regulate fertility and have allowed to infer reasonable estimates of the number of protein-coding genes essential for fertility (de Rooij and de Boer, 2003;Schimenti and Handel, 2018).
Primary ovarian insufficiency (POI) is a major cause of female infertility and affects about 1-3% of women under 40 years of age. It is characterized by cessation of ovarian function before the age of 40 years. POI results from a depletion of the ovarian follicle pool and can be isolated or syndromic. Genetic causes of POI account for approximately 20% of cases (Rossetti et al., 2017). Although infertility-causing pathogenic variants are inherently unlikely to spread in a population, they can be observed within families, especially when there is consanguinity. Such cases provide crucial insights into the function of the genes and molecular mechanisms that they disrupt. Over the last decade, causative variants in several genes have been found using whole exome sequencing in 'POI pedigrees'. In particular, pathogenic variants in genes involved in DNA replication, recombination or repair, such as STAG3, SYCE1, HFM1, MSH5 and MEIOB have been formally implicated in this condition by ourselves and others (Caburet et al., 2014;Caburet et al., 2019a;de Vries et al., 2014;Guo et al., 2017;Primary Ovarian Insufficiency Collaboration et al., 2014).
In this study, we have identified in a consanguineous family with POI the candidate S167L missense variant in HSF2BP, an essential yet poorly studied meiotic gene. HSF2BP encodes an interactor of the heat-shock response transcription factor HSF2 (Yoshima et al., 1998). During the course of this work and, in agreement with our results, two independent groups showed that HSF2BP is essential for meiotic recombination through its ability to interact with BRCA2 (Brandsma et al., 2019;Zhang et al., 2019). Here, we report that the introduction of the missense variant HSF2BP-S167L in mouse leads to subfertility and DNA repair defects during prophase I. In addition, we identified a protein complex composed of BRCA2, HSF2BP, and the as yet unexplored C19ORF57/BRME1 (meiotic double-stranded break BRCA2/HSF2BP complex associated protein) as a key component of the meiotic recombination machinery. Our studies show that a single substitution (S167L) in HSF2BP leads to a reduced loading of both BRME1 and HSF2BP at the RNs. Furthermore, our results suggest that meiotic progression requires a critical threshold level of HSF2BP/BRME1 for the ulterior loading of the recombinases to the RNs.

Clinical cases
The parents are first-degree cousins of Israeli Arab origin. Of the five daughters, three are affected with POI and presented with early secondary amenorrhea. They had menarche at normal age (at 13-14) but with irregular menses that stopped around 25. Only one of the patients affected by POI could have a child with the help of a fertility treatment (see pedigree in Figure 1). In order to identify the genetic basis of this familial POI case, we performed whole exome sequencing on genomic DNA from two POI patients, III-2 and III-3, and their fertile sister III-10 (Supplementary file 1a). Variants were filtered on the basis of (i) their homozygosity in the patients, (ii) their heterozygosity or absence in the fertile sister, (iii) their absence in unrelated fertile in-house controls and (iv) a minor allele frequency (MAF) below 0.01 in all available databases (Supplementary file 1b). This filtering process led to the identification of a missense substitution located in the HSF2BP gene: rs200655253 (21:43630396 G > A, GRCh38). The variant lies within the sixth exon of the reference transcript ENST00000291560.7 (NM_007031.2:c.500C > T) and changes a TCG codon into a TTG (NP_008962.1:p.Ser167Leu). It is very rare (Variant Allele Frequency/VAF 0.0001845 in the GnomAD database and 0.0005 in the GME Variome dedicated to Middle-East populations) and absent in a Figure 1. Pedigree of the consanguineous family with the variant HSF2BP-S167L. III-1 and III-2 are monozygotic twins, who appear phenotypically dizygotic. Clinical investigation confirmed POI, with normal 46, XX karyotype (500 bands and SKY spectral karyotyping). Year of birth and age of menarche are indicated when known. III-1 became amenorrheic at age 24 and III-2 at age 25, both after irregular menstruations since menarche. III-1 presents with a short stature (152 cm, within the 3-5 percentile), a normal neck, cubitus valgus and metacarpal shortening of 4-5. Ultrasound investigation showed normal uterus and ovaries. Her g-banding karyotyping was normal 46, XX (500 bands) and variants in FMR1 gene were ruled out. III-2 displays a normal secondary sexual development with no dysmorphic sign. Clinical investigation confirmed POI, with normal 46, XX karyotype (500 bands and SKY spectral karyotyping). The elder sister III-3 was also diagnosed with POI, with no further clinical information. She is 160.5 cm. She had one normal pregnancy with the help of 'fertility treatment', and a second unsuccessful attempt. The two fertile sisters, III-6 and III-10 had their menarche at 14-15 and 13-14 respectively, with regular menstruations ever since. They are respectively 150 cm and 151 cm, with no clinical sign, and each one had several children without difficulties. The fertile brother III-7 is 171 cm and shows no health or fertility problem. He developed frontal baldness since the age of 30. The genotype of each individual at the variant genomic position in HSF2BP is shown in red, as determined by Sanger sequencing for available DNAs (See Figure 1-figure supplement 1). The online version of this article includes the following figure supplement(s) for figure 1:   homozygous state from all available databases. The variant was verified by Sanger sequencing and was found to segregate in a Mendelian fashion within the family: the affected twin III-1 was homozygous for the variant and both parents and fertile siblings were heterozygous carriers (Figure 1-figure supplement 1). Therefore, there was no homozygous males identified in this family, preventing the analysis of the impact of this variant on male fertility. Serine 167 is a highly conserved position and the S167L variant is predicted to be pathogenic or deleterious by 11 out of the 18 pathogenicity predictors available in dbNSFP 3.5. ( Mice with the HSF2BP S167L variant show a partial reduction of fertility During the course of this work, two independent groups showed that HSF2BP is essential for meiotic recombination through its ability to interact with the armadillo repeats of BRCA2 (Zhang et al., 2019). Both groups showed that genetic disruption of Hsf2bp in mouse leads to the accumulation in the chromosomes axes of DNA repair proteins such as gH2AX (ATR-dependent phosphorylation of H2AX marks DSBs) and the single stranded-DNA binding protein RPA, a strong reduction of the recombinases DMC1 and RAD51 at the RNs and a lack of COs as labelled by MLH1 (Baker et al., 1996). The end result is male sterility (Brandsma, 2006;Brandsma et al., 2019;Zhang et al., 2019). However, loss of HSF2BP in female mice showed a milder meiotic phenotype (Zhang et al., 2019 and our own data, see below) and a weak albeit non-statistical significant reduction of fertility (Brandsma et al., 2019) despite all of the mutants are nulls though in different genetic backgrounds.
In order to confirm the causality of the S167L variant in this POI family, we generated a knock-in mouse Hsf2bp S167L/S167L by genome editing (Figure 2-figure supplement 1a). We also generated a loss-of-function model (Hsf2bp -/-) for direct comparison (Figure 2-figure supplement 1b-d).
Hsf2bp S167L/S167L male and female mice were able to reproduce but females showed a significant reduction in the number of litters (Figure 2a), whilst males only showed a slight non-significant reduction in fertility (Figure 2a), suggesting that the S167L variant impacts murine fertility.
Histological analysis of Hsf2bp S167L/S167L ovaries revealed no apparent differences in the number of follicles in comparison to wild-type (WT) animals ( . Testes from Hsf2bp S167L/S167L mice displayed a reduced size (21% reduction compared to WT mice; testis/body weight ratio: S167L 0.26% ± 0.07 (n = 12) vs 0.33% ± 0.05 for WT controls (n = 14), **p<0,01, Figure 2d and Figure 2-figure supplement 2b) and this reduction was stronger in Hsf2bp -/testes (70% reduction compared to WT, testis/body weight ratio: Hsf2bp -/-0.10% ± 0.005 (n = 6) vs 0.33% ± 0.05 for WT controls (n = 14) ****p<0,001, Figure 2d and Figure 2-figure supplement 2b). Histological analysis of adult Hsf2bp S167L/S167L testes revealed seminiferous tubules with a partial arrest with apoptotic spermatocytes (meiotic divisions) and their epididymis exhibited scarcer spermatozoa (Figure 2e). Consistent with these results, Hsf2bp S167L/S167L males showed increased numbers of meiotic divisions positive for TUNEL staining ( Figure 2f) and a reduction in the number of spermatozoa in the epididymis (3.3 Â 10 6 in the Hsf2bp S167L/S167L mutant vs 4.3 Â 10 6 in the WT; Figure 2g). During mouse spermatogenesis, the 12 stages of the epithelial cycle can be distinguished in seminiferous tubule sections by identifying groups of associated germ cell types (Ahmed and de Rooij, 2009). Following these criteria, the seminiferous epithelium of Hsf2bp -/mice showed a stage IV arrest, characterized by a massive apoptosis of zygotene-like spermatocytes occurring at the same time that In spermatogonia divide into B spermatogonia (Figure 2e). The presence of spermatogonia, spermatocytes, Sertoli and Leydig cells was not altered in any of the mutants (Figure 2e). These results suggest that mice bearing the POI-causing variant only partially phenocopy the human disease.
Next, we analyzed whether the POI-inducing variant affects the loading/stability of HSF2BP by immunolabeling meiocytes from Hsf2bp S167L/S167L mice. We observed a striking reduction of HSF2BP staining at the axes during prophase I in both spermatocytes and oocytes (Figure 3a-b). Western blot analysis of WT, Hsf2bp S167L/S167L and Hsf2bp -/in whole testis extracts from 13 days post-partum (dpp) animals ( Figure 3c) revealed that the reduced labeling observed by immunofluorescence correlated with a reduced protein expression level, suggesting that the mutation leads to a reduced expression and/or stability.
Given that HSF2BP is essential for DNA repair, we carried out a comparative staining analysis of gH2AX, the ssDNA-binding protein RPA, the recombinases RAD51 and DMC1, the ssDNA-binding protein SPATA22 (complexed to RPA during resection) and CO formation in meiocytes from Hsf2bp S167L/S167L , Hsf2bp -/and WT animals ( In accordance with the reduction of COs, we observed the presence of univalents in the XY pair at pachynema as well as univalents in metaphase I spermatocytes from Hsf2bp S167L/S167L mice ( Figure 6d and Our analysis in females showed accumulation of gH2AX staining ( Figure 4c) but no accumulation in RPA labeling in Hsf2bp S167L/S167L and Hsf2bp -/oocytes ( Figure 4d and Figure 4-figure supplement 1b). Similar to the spermatocytes, DMC1 and RAD51 staining showed a reduction in both Hsf2bp S167L/S167L and Hsf2bp -/oocytes (Figure 5c-d and Figure 5-figure supplement 1c-d). SPATA22 labeling in females showed a clear accumulation in Hsf2bp -/but only a trend towards accumulation in Hsf2bp S167L/S167L oocytes ( Figure 6a and Figure 6-figure supplement 1a). In Figure 2 continued of follicles. Bar in panels 100 mm. (c) Quantification of the number of follicles (primordial, primary, secondary and antral follicles) per ovary in Hsf2bp +/+ , Hsf2bp S167L/S167L and Hsf2bp -/females at 5 dpp and 6 weeks of age showing no differences between Hsf2bp +/+ and Hsf2bp S167L/S167L but a strong reduction in the oocyte pool in Hsf2bp -/females. Ovaries: five dpp/6 weeks = 5/5 ovaries from Hsf2bp +/+ and Hsf2bp S167L/S167L and 4/3 from Hsf2bp -/-. Two-tailed Welch's t-test analysis: ns, no significant differences, **p<0.01, ***p<0.001. (d) Testis size of Hsf2bp S167L/S167L (left, 21% reduction) and Hsf2bp -/mice (right, 70% reduction) in comparison with their WT counterparts. See Figure 2-figure supplement 2b for the quantification. (e) PAS and Hematoxylin stained testis sections. The S167L variant leads to a partial spermatogenic arrest with an elevated number of apoptotic meiotic divisions (blue asterisks) and a reduction of the number of spermatozoa in the epididymides in comparison with the WT control (Hsf2bp +/+ ). The null allele (Hsf2bp -/-) showed a complete spermatogenic arrest at epithelial stage IV and absence of spermatozoa. Massive apoptosis of spermatocytes is indicated (blue arrowheads). Bar: upper panels 10 mm, lower panels 20 mm. (St) Seminiferous tubules, (Ep) Epididymides. (f) Immunohistochemical detection of apoptotic cells by TUNEL staining showing an increase of apoptotic meiotic divisions in stage XII tubules from Hsf2bp S167L/S167L males (magnified panel). Plot under the panel represents the quantification. Mice: n = 3 adult mice for each genotype. Two-tailed Welch's t-test analysis: ***p<0.001. Bar in panel, 25 mm. (g) Quantification of epididymal sperm in Hsf2bp +/+ and Hsf2bp S167L/S167L adult mice. Epididymides: n = 8 for each genotype. Two-tailed Welch's t-test analysis: *p<0.05. The online version of this article includes the following figure supplement(s) for figure 2: Figure supplement 1. Generation and genetic characterization of Hsf2bp S167L and Hsf2bp-deficient mice. Figure supplement 2. Fertility defects in Hsf2bp S167L/S167L mice. agreement with the lower presence of recombinases, the number of COs (measured as interstitial CDK2 foci) was also reduced in Hsf2bp S167L/S167L oocytes, and a stronger reduction was observed in Hsf2bp -/oocytes ( Figure 6c and Figure 6-figure supplement 1d). Overall, male and female Hsf2bp S167L/S167L mice share alterations in the meiotic recombination pathway although with different reproductive outcome.
We next sought to understand how the HSF2BP pathogenic variant was mediating the observed meiotic alteration. HSF2BP has been shown to bind BRCA2, an essential protein for meiotic homologous recombination (Martinez et al., 2016;Sharan et al., 2004), by a direct interaction that involves Arg200 in HSF2BP and the Gly2270-Thr2337 region within the C-terminal fragment of BRCA2 Figure 3. Meiocytes from Hsf2bp S167L/S167L mice show a decrease in the expression of HSF2BP. (a-b) Double immunofluorescence of HSF2BP (green) and SYCP3 (red) in Hsf2bp +/+ and Hsf2bp S167L/S167L (a) spermatocyte and (b) oocyte spreads showing a strong reduction in the labeling of HSF2BP at the chromosome axis. Plots under the panels show the quantification. Nuclei analyzed: 30 zygonemas and 40 pachynemas from two adult male mice of each genotype. In females 38/39 zygonemas and 37/35 pachynemas from two 17.5 dpc embryos of Hsf2bp +/+ and Hsf2bp S167L/S167L , respectively. Twotailed Welch's t-test analysis: ****p<0.0001. (c) Western blot analysis of protein extracts from 13 dpp WT, Hsf2bp S167L/S167L and Hsf2bp -/testes using polyclonal antibodies against HSF2BP. Tubulin was used as loading control. Graph on the right represents the relative quantification of the immunoblotting. Mice: n = 2 Hsf2bp +/+ , Hsf2bp S167L/S167L and Hsf2bp -/-. Two-tailed Welch's t-test analysis: *p<0.05. Bar in panels a-c, 10 mm. The online version of this article includes the following figure supplement(s) for figure 3:  . DNA repair in Hsf2bp S167L/S167L mice. (a, c) Double labeling of gH2AX (green) and SYCP3 (red) in (a) spermatocyte and (c) oocyte spreads from WT, Hsf2bp S167L/S167L and Hsf2bp -/mice. (a) Males display an accumulation of gH2AX patches in Hsf2bp S167L/S167L pachynemas and a strong accumulation in the whole nucleus in Hsf2bp -/zygotene-like arrested cells. Plots on the right of the panel represent the percentage of pachynemas with gH2AX labeling (Nuclei: 364 Hsf2bp +/+ and 376 Hsf2bp S167L/S167L from three adult mice) and the quantification of gH2AX intensity on autosomes at Figure 4 continued on next page (Brandsma et al., 2019). Given the impossibility to detect endogenous BRCA2 by immunofluorescence in mouse spermatocytes, we carried out co-localization/interaction assays in a heterologous system by transfecting BRCA2-C (i.e. its C-term) and HSF2BP in U2OS/HEK293T. Our results showed that BRCA2-C co-immunoprecipitates with both HSF2BP-WT and HSF2BP-S167L in similar ways (Figure 6-figure supplement 2a). In single transfections, HSF2BP localized in the nucleus and cytoplasm whereas BRCA2-C showed nuclear localization. This pattern changed drastically to a nuclear dotted pattern when co-transfected ( Figure 6-figure supplement 2b). This re-localization was independent of the HSF2BP variant, suggesting that the HSF2BP variant effects are not directly mediated by BRCA2 delocalization.

BRME1, a novel interactor of HSF2BP
In order to further understand the mechanism underlying the pathogenicity of the HSF2BP-S167L variant, we searched for proteins that interact with the murine HSF2BP through a yeast two hybrid (Y2H) screening. The analysis of the clones with putative interactors revealed that 19 out of 98 analyzed clones matched the uncharacterized gene 4930432K21Rik, which corresponds to human C19ORF57, hereby dubbed BRME1 for Break Repair Meiotic recombinase recruitment factor 1. This HSF2BP interactor consists of 600 amino acids with a high content of acidic residues, has no recognizable functional domains and is intrinsically disordered. The interaction was validated by transiently transfecting plasmids driving the expression of HSF2BP and BRME1. Both HSF2BP-S167L and WT interacted with BRME1 ( Figure 7a). We further validated this interaction in vivo by co-immunoprecipitation (co-IP) of both proteins from mouse whole testis extracts ( Figure 7b). To identify the regions required for this interaction, we split the BRME1 protein into three fragments (N-terminal, central region and C-terminal). We mapped the HSF2BP/BRME1-interacting domain to the C-term fragment of BRME1 (spanning residues 475-600 of the murine protein, Figure 7c). In line with this, the Brme1 D142-472/D142-472 mutant mice, expressing the BRME1 protein devoid of its central part, were fertile and did not show defects in chromosome synapsis or an alteration of HSF2BP loading to axes, further indicating that a large fraction of the coding protein of BRME1 is not essential for BRME1/HSF2BP function in vivo (Figure 7-figure supplement 1).
To determine the role of BRME1 in recombination and DNA repair, we analyzed its cytological distribution pattern in different mutants lacking synapsis/recombination-related proteins. These mutants were the meiotic cohesin REC8 (Bannister et al., 2004) Baudat et al., 2000). HSF2BP staining was also carried out for a direct comparison. We were able to show that none of the recombinationdeficient mutants abrogate BRME1 labeling at zygotene (or the corresponding meiotic stage at which the mutant spermatocytes are arrested), in contrast to its absence of loading in SPO11-deficient mice (Figure 7-figure supplement 6b, left). These results are very similar to those obtained for HSF2BP in these mutants (Figure 7-figure supplement 6b, right) and indicate that SPO11dependent DSBs are essential for targeting HSF2BP/BRME1 to the RNs, and that the heterocomplex can be positioned at early events soon after DSBs generation.
To functionally analyze the role of BRME1 in mouse fertility, we generated a Brme1 -/null mutant by genome editing (Figure 8-figure supplement 1a-d). Brme1 -/females, despite being fertile, showed a strong reduction of the follicle pool ( Figure 8a). Male Brme1 -/mice were infertile, the average size of their testes was severely reduced (76% reduction compared to WT; testis weight/ body weight ratio: Brme1 -/-0,08% ± 0004 (n = 6) vs 0,33% ± 0,05 for WT controls (n = 14), ****p<0.0001, Figure 8b and Figure 2-figure supplement 2b), and lacked spermatozoa ( Figure 8c). Histological analysis showed a meiotic arrest at epithelial stage IV with apoptotic spermatocytes ( Figure 8c). Double immunolabeling of SYCP3 and SYCP1 revealed that spermatocytes were partially synapsed and showed a partner-switch phenotype in which synapsis is not restricted to homologous pairs ( Figure 8d). The arrest corresponds to a zygotene-like stage though a small fraction of cells (3,7% ± 1,9; n = 3) were able to escape this blockage reaching early pachytene. Brme1 -/oocyte spread analysis revealed the presence of a subset of fully synapsed pachynemas but an increased number of cells with different degree of asynapsis (47,9% ± 2,2 vs 12% ± 5,7 in the WT; n = 2 (both genotypes), *p<0,05 Figure 8e and  green text) and western blot with the indicated antibodies (WB: red text) (c) Schematic representation of full-length BRME1 protein and the corresponding deletion (D) constructs (filled boxes) generated to decipher the essential BRME1 region for interacting with HSF2BP (green positive interaction and red no interaction). Western blots under the scheme show the Co-IP experiments. HEK293T cells were transfected with GFP-HSF2BP and the different delta constructs of Flag-BRME1. The D475-600 abolishes the interaction, indicating that the C terminus of BRME1 is the essential region of interaction with HSF2BP. (d) Triple immunofluorescence of BRME1 (green), HSF2BP (red) and SYCP3 (blue) in WT spermatocyte spreads showing high colocalization between BRME1 and HSF2BP at late zygotene and pachytene (See Supplementary file 1d for quantification). Bar in panel, 10 mm. (e) Double immunolabeling of spermatocyte spread preparations with HSF2BP (green) and BRME1 (red) analyzed by Stimulated emission depletion (STED) microscopy. Bar in panel, 5 mm. (f-g) Double immunofluorescence of BRME1 (green) and SYCP3 (red) in (f) spermatocytes and (g) oocyte spreads from Hsf2bp +/+ , Hsf2bp S167L/S167L and Hsf2bp -/showing a strong reduction of BRME1 staining in the S167L mutant and absence in the Hsf2bp knock-out. Plots next to the panel represent the quantification. See also extended Figure 7-figure supplement 4a. Male nuclei: Hsf2bp +/+ / Hsf2bp S167L/S167L /Hsf2bp -/-: n = 16/16/20 early and 15/15/16 late zygonemas, 16/16 /-pachynemas from two adult mice. Female nuclei: n = 18 pachynemas from two embryos (17.5 dpc) of each genotype. Two-tailed Welch's t-test analysis: *p<0.05, ****p<0.0001. Bar in panels, 10 mm. (h) Western blot analysis of protein extracts from 13 dpp WT, Hsf2bp S167L/S167L and Hsf2bp -/testes using an antibody against BRME1. Tubulin was used as loading control. Graph on the right represents the relative quantification of the immunoblotting. Mice: n = 2 Hsf2bp +/+ , Hsf2bp S167L/S167L and Hsf2bp -/-. Two-tailed Welch's t-test analysis: *p<0.05. The online version of this article includes the following figure supplement(s) for figure 7:       interaction between HSF2BP and BRME1, we tested whether HSF2BP localization depended on BRME1 by immunolabeling of HSF2BP in Brme1 -/spermatocytes and oocytes. Our results showed a strong reduction of HSF2BP staining in BRME1-null spermatocytes ( Figure 8f) and a total absence in oocytes ( Figure 8g). Western blot analysis of HSF2BP in 13 dpp testis extracts from Brme1 -/mice showed a strong reduction in comparison with the WT control (Figure 8h), suggesting again that BRME1 is necessary for HSF2BP protein stabilization.

BRME1 and HSF2BP form a multimeric complex with PALB2 and BRCA2
To further delineate the interactome of BRME1, we immuno-precipitated BRME1 from testis extracts coupled to mass-spectrometry. We identified as expected HSF2BP as the main interactor, but also BRCA2, PALB2, RAD51 and RPA, strongly suggesting that they form a large multimeric complex (Supplementary files 1g-1h). For validation, we transfected the corresponding expression plasmids in HEK293T cells for co-IP analysis. BRME1 co-immunoprecipitated with BRCA2 and HSF2BP when they were all co-transfected, but importantly BRME1 alone did not co-immunoprecitate with BRCA2 (Figure 10a-b). The reciprocal co-IP of BRCA2 with HSF2BP and BRME1 was also positive. We also observed modest but positive co-IP of HSF2BP with RPA, PALB2 and RAD51; and of BRME1 with RAD51 and RPA but not with PALB2 ( Figure 10-figure supplement 1a). These interactions were further analyzed in a cell-free TNT system coupled to co-immunoprecipitation assays. We observed an absence of direct interaction between any of them, with the exception of BRME1 and HSF2BP, as expected from the Y2H analysis ( Figure 10-figure supplement 1b). These results suggest that these proteins belong to a complex (or complexes) in vivo (likely through BRCA2) and that the HSF2BP-S167L variant could be altering BRME1 interaction with partners of major BRCA2-containing recombination complexes.
Finally, given the interaction of BRME1 and HSF2BP, we analysed their interdependence in U2OS cells. Transfected HSF2BP was localized diffusely in the cytoplasm and the nucleus (Figure 10c). mm. (f-g) Double labeling with HSF2BP (green) and SYCP3 (red) of (f) spermatocyte and (g) oocyte spreads from Brme1 -/mice showing faint HSF2BP labeling in spermatocytes and total absence of labeling in oocytes. Plot on the right of (f) panel represents de quantification of HSF2BP foci in Brme1 -/spermatocytes. Nuclei: n = 30 zygonemas/zygonemas-like from two adult mice of each genotype (Brme1 +/+ values from Figure 3a) Two-tailed Welch's t-test analysis: ****p<0.0001. Bar in panels, 10 mm. (h) Western blot analysis of protein extracts from 13 dpp WT, Brme1 -/and Hsf2bp -/testes with a specific antibody against HSF2BP. Tubulin was used as loading control. Graph on the right represents the relative quantification of the immunoblotting. Mice: n = 2 Brme1 +/+ , Brme1 -/and Hsf2bp -/-. Two-tailed Welch's t-test analysis: *p<0.05. The online version of this article includes the following figure supplement(s) for figure 8:  However, when HSF2BP was co-overexpressed with BRME1, its pattern changed to an intense nucleoplasm staining with nuclear invaginations that resemble nucleoplasmic reticulum (Figure 10c; Malhas et al., 2011). Interestingly, such invaginations were reduced when BRME1 was co-transfected with HSF2BP-S167L (Figure 10c). In addition, the intensity of the fluorescence signal of HSF2BP-S167L was lower than for HSF2BP-WT and both intensities increased when HSF2BP was co-expressed with BRME1 ( Figure 10c). Western blot analysis indicated a reduced protein stability of the S167L variant (Figure 10d), in agreement with the results observed in vivo (Hsf2bp S167L/S167L mutant, Figure 3). Interestingly, the protein expression level of transfected HSF2BP increased when co-transfected with BRME1 and was partially dependent on proteasome degradation (Figure 10d), indicating a role of BRME1 in stabilizing HSF2BP. Taken altogether and given the low protein expression of BRME1 in the Hsf2bp S167L/S167L , these results suggest a functional interdependence between BRME1 and HSF2BP that leads to their lower protein stability/expression in mutant meiocytes, which might induce recombination defects.

Discussion
Using exome sequencing, we identified the S167L missense variant in HSF2BP in a consanguineous family with three cases of POI with secondary amenorrhea. All affected family members are homozygous for the variant, and the healthy relatives are heterozygous carriers. The causality of the HSF2BP-S167L variant is supported by the meiotic phenotype and the subfertility observed in Hsf2bp S167L/S167L female mice. Furthermore, the DNA repair defects in murine Hsf2bp S167L/S167L meiocytes, displayed by the reduced number of RAD51/DMC1 foci on DSBs and the subsequent reduction in the number of COs, provide evidence that this missense variant alters meiotic recombination. This conclusion was further supported by the comparative analysis of the S167L allele with the Hsf2bp null allele, which revealed that the missense variant can be considered as a hypomorphic allele. This is in agreement with the secondary amenorrhea observed in the patients, and the residual (medically-assisted) fertility in one of the affected sisters. Our identification of HSF2BP as a gene implicated in POI is in line with recent reports of POI-causing variants in genes that are required for DNA repair and recombination, such as MCM8, MCM9, SYCE1, MSH4, PSMC3IP, FANCM or NBN (AlAsiri et al., 2015;Carlosama et al., 2017;de Vries et al., 2014;Fouquet et al., 2017;He et al., 2018;Tenenbaum-Rakover et al., 2015;Tucker et al., 2018;Wood-Trageser et al., 2014;Zangen et al., 2011).
Meiotic mouse mutants often exhibit sexually dimorphic phenotypes (Cahoon and Libuda, 2019). These differences can have a structural basis, given that the organization of the axial elements is known to be different between sexes. This is supported by the difference in length of the axes and  Figure 5c). Two-tailed Welch's t-test analysis: ***p<0.001, ****p<0.0001. Bar in panels, 10 mm. (c-d) Double immunofluorescence of RAD51 (green) and SYCP3 (red) in Brme1 +/+ and Brme1 -/-(c) spermatocytes and (d) oocytes showing a reduction in the number of RAD51 foci in the Brme1 -/in comparison to the WT. Plots under each panel represent the quantification. See also extended panels on Figure 9-figure supplement 2c-d. Male nuclei for RAD51: Brme1 +/+ /Brme1 -/n = 39/39 early and 37/45 late zygonemas/zygonemas like from two adult mice of each genotype (Brme1 +/+ values from Figure 5b). Female nuclei for RAD51: Brme1 +/+ /Brme1 -/n = 35/31 zygonemas and 42/40 pachynemas from two embryos (17.5 dpc) of each genotype (Brme1 +/+ values from Figure 5d). Two-tailed Welch's t-test analysis: ****p<0.0001. (e-f) Quantification of SPATA22 intensity in (e) spermatocytes and (f) oocyte spreads from Brme1 +/+ and Brme1 -/-. See immunofluorescences in Figure 9-figure supplement 2e-f. Nuclei: Males, n = 20 cells from two adult mice of each genotype (Brme1 +/+ from Figure 6a). Females, Brme1 +/+ /Brme1 -/n = 41/40 zygonemas and 40/40 pachynemas from two embryos of each genotype (17.5 dpc) (Brme1 +/+ from Figure 6a). Two-tailed Welch's t-test analysis: ****p<0.0001. (g) Double immunofluorescence of MLH1 (green) and SYCP3 (red) in Brme1 +/+ and Brme1 -/spermatocytes showing the absence of MLH1 labeling in the knock-out. (h) Double labeling of CDK2 (green) and SYCP3 (red) in oocyte spreads from 17.5 dpc Brme1 +/+ and Brme1 -/embryos. During meiotic prophase I, CDK2 localizes to the telomeres of chromosomes from leptotene to diplotene. However, around mid-pachytene additional interstitial CDK2 signals appear at CO sites, colocalizing with MLH1. As a measure of COs, just interstitial CDK2 foci (non-telomeric) have been counted. Brme1 -/females show a strong reduction in the number of COs. Plot under the panel show the quantification. Nuclei: Brme1 +/+ /Brme1 -/n = 79/49 from three embryos (17.5 dpc) in WT and two embryos in Brme1 -/-(Brme1 +/+ from Figure 6c). Two-tailed Welch's t-test analysis: ***p<0.0001. Bar in all panels, 10 mm. The online version of this article includes the following figure supplement(s) for figure 9:   In the presence of HA-HSF2BP (triple co-transfection) BRCA2-C and BRME1 coimmunoprecipitate (co-IPs between HSF2BP and BRCA2-C are shown in Figure 6-figure supplement 2a). (c) Transfected U2OS cells with plasmids encoding Flag-HSF2BP (WT or S167L) and EGFP-BRME1 alone or together were immuno-detected with antibodies against Flag (red) and EGFP (green). Transfected HSF2BP (WT and S167L) labels the whole cell (S167L less intense) whereas BRME1 shows nuclear localization. When co-expressed, BRME1 and HSF2BP change their patterns and form nuclear invaginations that resemble nucleoplasmic reticulum. This phenotype is milder in the presence of HSF2BP-S167L than with the WT (graph under the panel: quantification of the number of cells showing a nucleoplasmic reticulum pattern). n > 400 cells from two independent transfections of each condition. Two-tailed Welch's t-test analysis: *p<0.05. Bar in panel, 20 mm. (d) HEK293T cells were transfected with Flag-HSF2BP (WT and S167L) alone or with GFP-BRME1. Additionally, cells transfected with Flag-HSF2BP were treated with the proteasome inhibitor (MG132, 10 mM) and analyzed by western blot. Cherry was used as transfection efficiency control. HSF2BP-WT was expressed at higher levels than HSF2BP-S167L and their detection (both the WT and the S167L variant) was increased when co-transfected with BRME1. The increase was greater for the HSF2BP-S167L variant in comparison with the WT. Incubation with MG132 increased the detection levels of transfected HSF2BP mimicking the effect of co-transfecting BRME1. n = 3 independent transfections for each condition. Two-tailed Welch's t-test analysis: **p<0.01, ***p<0.001, ****p<0.0001. The online version of this article includes the following figure supplement(s) for figure 10: by the essential role that the meiotic cohesin subunit RAD21L plays in males but not in females (Herrán et al., 2011). In general, meiotic recombination mutants appear to proceed further in female than in male, because in males the asynapsis of the sex bivalent leads to a loss of silencing of the Y chromosome, and perhaps also because of the presence of less stringent checkpoints in oogenesis (Hunt and Hassold, 2002). A sexually dimorphic phenotype is also observed here in our HSF2BP-S167L mutant mice. Both sexes show a significant decrease in the number of COs and consequently an increase in the frequency of meiocytes showing bivalents without CO. This leads to a significant reduction of spermatozoa in the epididymis, while the number of oocytes and their distribution in the follicular pool are not affected in females. Our failure to detect any significant impact of the variant on male fertility could be due to the high variability in litter frequency and size in our study. However, it is known that a strong reduction of the spermatozoa count (up to 60%) does not affect male mouse fertility (Schürmann et al., 2002), which would explain the normal fertility of male mice bearing the HSF2BP-S167L variant despite the presence of overt meiotic alterations. By contrast, female mice with the HSF2BP-S167L variant show a mild sub-fertility phenotype with a reduction of litter frequency. This can be due to the very much fewer gametes available for fertilization in females in comparison to males but also to molecular differences in the meiotic recombination process in both sexes (Cahoon and Libuda, 2019), as displayed by the absence of RPA accumulation or the more pronounced decrease in DMC1 foci observed in HSF2BP-S167L oocytes.
Although we cannot exclude that a POI-like phenotype would appear over time in HSF2BP-S167L female mice, the sub-fertility observed in HSF2BP-S167L females appears to be milder in comparison to the phenotype of the human patients. This could be explained by a lower sensitivity of mice to hypomorphic alleles compared to humans, in a similar manner to the known lower gene dosage sensitivity of the former in the context of genes that are haploinsufficient in human (Veitia, 2003). In addition, the initial events of human female meiosis appear to be more error-prone than in mice, or even than human males, as evidenced by the increased incidence of synaptic defects in the human oocytes or the fact that MLH1 foci appear much earlier in prophase I (Hassold et al., 2007). Furthermore, it has recently been shown that human oocytes exhibit a specific CO maturation inefficiency . Indeed, despite a higher total number of COs in women, the frequency of bivalents without CO is paradoxically higher in women than in men. Altogether, these observations could explain the stronger phenotype of the human POI patients compared to the Hsf2bp S167L/S167L female mice.
The absence of homozygous male carriers in the consanguineous family studied here prevents the direct comparison of the impact of the HSF2BP-S167L variant on fertility phenotypes between human and mouse males. Although future studies might identify infertile men homozygous for the S167L variant, it is interesting to note that another variant of HSF2BP (G224*) was shown to affect recombination rate in males and that two siblings homozygous for this HSF2BP variant in the analyzed Icelandic population were healthy but without descendants, suggesting they were infertile (Halldorsson et al., 2019). This reinforces a conserved function of HSF2BP in human male fertility.
We have shown through a biochemical analysis that BRME1 immunoprecipitates mainly with its partner HSF2BP (constituting or belonging to a complex) but also with PALB2, RAD51, RPA and BRCA2 (in testis extracts). These interactions could possibly be mediated by the multidomain hub protein BRCA2 (Siaud et al., 2011) as HSF2BP interacts directly with BRCA2 (Brandsma et al., 2019) and with BRME1 (this work). In addition, BRCA2 also directly interacts with the DSBs recruiter PALB2, with the recombinases RAD51 and DMC1 (through different specific domains), and with DNA (Siaud et al., 2011). This BRCA2-containing complex participates in the orderly orchestration of events at DSBs such as the initial binding of RPA to the resected DNA, the exchange of RPA by RAD51/DMC1, and the loading of the MEIOB-SPATA22 complex to the RPA complexes (Martinez et al., 2016;Zhao et al., 2015). Interestingly, genes with recently identified variants in POI patients are implicated in the repair of induced DSBs at the early stages of meiosis and encode BRCA2-interacting factors, such as MEIOB, DMC1 or BRCA2 itself (Caburet et al., 2019a;Caburet et al., 2020;Caburet et al., 2019b;He et al., 2018). This highlights the crucial importance and the high sensitivity of this particular meiotic step, and the hub role of BRCA2 as a tightly regulated platform for correct meiotic recombination.
We have also shown by several complementary approaches that the proteins HSF2BP/BRME1 constitute in vivo a functional complex in which both subunits are essential for meiotic recombination and for their mutual protein expression and/or stability in vivo. Accordingly, the genetic depletion of BRME1 or HSF2BP leads to similar if not identical phenotypes in which oogenesis is altered with severe defects in chromosome synapsis that promotes premature loss of ovarian follicles and spermatogenesis is arrested at zygotene-like stage resulting in a lack of spermatozoa. These meiocytes are not able to load the recombinases RAD51/DMC1, impairing the proper repair of DSBs leading to the generation of no COs or very few in males and females, respectively. As a consequence, zygonema-like spermatocytes accumulate the single strand binding proteins SPATA22 and RPA, whereas oocytes accumulates only SPATA22. During the course of the reviewing of this work, three Brme1 knockouts have been described Shang et al., 2020;Takemoto et al., 2020). All the described male mutants show strong fertility defects and similar molecular alterations although with different severity. In females, the two works that address their analysis (Shang et al. and Takemoto et al.), describe normal fertility which is in contrast with the strong reduction in the follicle pool and meiotic defects observed in our Brme1 -/females. The higher severity of our male and female mutants could be explained on the basis of the different genetic background of the mice given that all of them are apparently similar.
The S167L human recessive POI variant behave as a hypomorphic allele in mice, which results in a reduction of the protein expression/stability of itself and of its partner BRME1 in vivo and in transfected cells. As a consequence, both male and female Hsf2bp S167L/S167 mice show a similar but milder phenotype than that of the Hsf2bp -/or Brme1 -/-, consisting in a reduction in the spermatozoa count while being fertile and a subtle reduction in female fertility. Molecularly, the reduction observed in the meiocytes of the mutant Hsf2bp S167L/S167L mice of RAD51/DMC1, the reduction of COs (both in males and females), the accumulation of RPA (only in males) and SPATA22 (in males and females) are also weaker than in the null mutant. The observed accumulation of RPA in males is likely to occur at the early stages of recombination because SPATA22 loading to the DSBs is also increased in the mutants of HSF2BP and BRME1. RPA, as part of a trimeric replication protein complex (RPA1-RPA2-RPA3), binds and stabilizes ssDNA intermediates that form during DNA repair. In meiosis, RPA is also forming a complex with two other essential meiotic players MEIOB (homologue of RPA) and SPATA22. However, the loading of this complex to DSBs is RPA-independent (Shi et al., 2019). It has been postulated that RPA functions in meiosis at two different stages; (i) during the early recombination stages when the DSBs ends are resected by the MRN complex and (ii) during the strand invasion into the homologous duplex that is carried out by RAD51/DMC1 and ssDNA is generated at the displacement loops (Shi et al., 2019). The observed lack of DNA-binding ability of HSF2BP/BRME1 points towards a model in which the absence of the complex HSF2BP/ BRME1 through a direct interaction with BRCA2 impairs the replacement of RPA by RAD51/DMC1 in the foci that form on the DSBs of the spermatocytes. Similarly, the reduced expression at the protein level of HSF2BP/BRME1 as a consequence of the POI variant, which does not affect their heterodimerization, would make them less proficient in replacing RPA in the spermatocytes by the recombinases RAD51/DMC1 leading to a lower frequency of COs. Given the unknown function that RPA plays in vivo during oogenesis (Shi et al., 2019), it is tempting to speculate that the role of RPA in mediating the replacement of RAD51/DMC1 in female meiosis would be carried out by another protein complex such as SPATA22/MEIOB in a HSF2BP/BRME1-dependent manner.
Very recently, a high-resolution genome-wide recombination map revealed novel loci involved in the control of meiotic recombination and highlighted genes involved in the formation of the SC (SYCE2, RAD21L, SYCP3, SIX6OS1) and the meiotic machinery itself as determinants of COs (Halldorsson et al., 2019). Within the second category, variants of the SUMO ligase RNF212 and the ubiquitin ligase HEI10 have been largely documented as genetic determinants of the recombination rate in humans and, importantly, so were variants of HSF2BP. Consequently, gene dosage of RNF212 and HEI10 affects CO frequency through their activity in CO designation and maturation (Lake and Hawley, 2013;Reynolds et al., 2013). We found that both BRME1 and HSF2BP localization are unaffected in the loss-of-function mouse mutants of Rnf212 and Hei10 (Ccnb1ip1). This observation together with the proper co-localization of BRME1/HSF2BP with RPA allows us to map these proteins upstream in the recombination pathway.
It is worth noting that some of the genes affecting the recombination rate have also been described as 'fertility genes', such as SYCP3, HFM1 and HSF2BP (Geisinger and Benavente, 2017;Primary Ovarian Insufficiency Collaboration et al., 2014 and this work). Altogether, we propose that different variants of the same meiotic gene (alleles responsible for mild or strong phenotypes) can give rise to either an altered genome-wide recombination rate with no detrimental effect, or cause infertility when the decreased recombination rate falls below the lower limit of one COs per bivalent. In the present POI family, the S167L variant in HSF2BP seems to be below that limit. To our knowledge, HSF2BP is one of the very few human genes with variants known to affect both the genome-wide recombination rate in the human population and meiotic chromosome missegregation (fertility) through a reduction of the recombination rate (Halldorsson et al., 2019). Along similar lines, it is conceivable that variants with additive effects (Schimenti and Handel, 2018) can lead to a genome-wide reduction of the recombination rate and thus to aneuploidy and infertility. Specifically, variants in genes involved in meiotic recombination and SC constituents could be responsible for a large fraction of genetic infertilities. These variants should be under purifying selection and would be removed or substantially reduced from the population. However, this is not the case for genes with sexual phenotypic dimorphism (Gershoni and Pietrokovski, 2014) as is apparent for a wide number of meiotic genes (Cahoon and Libuda, 2019), including HSF2BP and BRME1, where individuals of one of the sexes are fertile carriers.
In summary, we describe for the first time a human family where POI co-segregates with a genetic variant in HSF2BP (S167L) in a Mendelian fashion. Humanized mice reveal that the HSF2BP variant is a hypomorphic allele that promotes the lower protein expression and/or stability of the HSF2BP/ BRME1 complex and phenocopy in a milder manner the meiotic defects observed in mice lacking either HSF2BP or its direct interactor BRME1.

Whole exome sequencing
Written informed consent was received from participants prior to inclusion in the study and the institutions involved. Genomic DNA was extracted from blood samples by standards protocols.
For individuals III-3 and III-10, library preparation, exome capture, sequencing and initial data processing were performed by Beckman Coulter Genomics (Danvers, USA). Exon capture was performed using the hsV5UTR kit target enrichment kit. Libraries were sequenced on an Illumina HiSEQ instrument as paired-end 100 bp reads. For individual III-2, library preparation, exome capture, sequencing and data processing were performed by IntegraGen SA (Evry, France) according to their in-house procedures. Target capture, enrichment and elution were performed according to manufacturer's instructions and protocols (SureSelect Human All Exon Kits Version CRE, Agilent). The library was sequenced on an Illumina HiSEQ 2500 as paired-end 75 bp reads. Image analysis and base calling was performed using Illumina Real Time Analysis (RTA 1.18.64) with default parameters.
Variant filtering was performed on the following criteria: . minimum depth at variant position of 10, . correct segregation in the family, on the basis of homozygosity by descent: variants should be homozygous in both affected sisters III-2 and III-3, and heterozygous or homozygous for Reference allele in the fertile sister III-10, . absence in unrelated in-house fertile controls, . Minor Allele Frequency (MAF) below 1% in global and in each population in the GnomAD database, . presence in the coding sequence (i.e not in UTRs, introns, intergenic,.) . high predicted functional impact on the protein. Impact was evaluated based on the predictors included in dbNSFP3.5 (Cahoon and Libuda, 2019;Carlosama et al., 2017;Cox and Mann, 2008) (considered as pathogenic when the majority of the predictors agreed).
The number of variants fulfilling those criteria is provided in Supplementary file 1b. Visual inspection of the variant was performed using the IGV viewer.

Sanger sequencing analysis
To confirm the presence and segregation of the variant, direct genomic Sanger DNA sequencing of HSF2BP was performed in the patients, the parents and non-affected siblings using specific primers: HSF2BP-ex6F: 5'-ctagaatcttctgtatcctgca-3' and HSF2BP-ex6R2: 5'-ggtctggaagcaaacaggcaa-3'. The resulting chromatograms are shown in Figure 1-figure supplement 1.

Predictions of pathogenicity and sequence conservation
The S167L variant was predicted to be pathogenic or deleterious and highly conserved by 11 out of the 18 pathogenicity predictors available in dbNSFP 3.5 (Supplementary file 1c). Upon verification, it appears that the conflicting interpretation of this variant might stem from the single occurrence of a Leu at this position in zebrafish. As the change in zebrafish is the variant that we have in the human family, we checked all the available sequences (Ensembl Release 99, January 2020, removing the one-to-many relationships). Ser167 is very highly conserved in mammals, birds and reptiles and fish and is present in 208 of 212 orthologous sequences (

Generation of CRISPR/Cas9-edited mice
For developing all the mutant mice models (Hsf2bp -/-, Hsf2bp S167L/S167L , Brme1 D142-472/D D142-472 , Brme1 -/-, Spo11 -/-, Rnf212 -/and Hei10 -/-) the different crRNAs were predicted at https://eu.idtdna. com/site/order/ designtool/index/CRISPR_CUSTOM. The crRNAs, the tracrRNA and the ssODNs were produced by chemical synthesis at IDT (crRNAs and ssODNs sequences are listed in Supplementary files 1i-1j). For the Hsf2bp S167L we introduced a mutation in the mouse counterpart residue (p.Ser171Leu) of the POI mutation found in the clinical case (p.Ser167Leu). However, for the shake of simplicity, on this manuscript we refer to the mutant allele by the acronym of the human mutation (S167L). The ssODN contains the mutation on the corresponding position of the mouse sequence (c.512C > T, p.Ser171Leu, see character in red in Supplementary file 1j) and the PAM mutations avoiding amino acid changes (see characters in bold in the Supplementary file 1j). For the Spo11 -/mice generation, the ssODN contains the mutations in the active site (TACTAC >TTCTTC p.YY137-138FF, see Supplementary file 1j) and the PAM mutations (bold characters in Supplementary file 1j). In all cases the crRNA and tracrRNA were annealed to obtain the mature sgRNA. A mixture containing the sgRNAs, recombinant Cas9 protein (IDT) and the ssODN (30 ng/ml Cas9, 20 ng/ml of each annealed sgRNA and 10 ng/ml ssODN) were microinjected into B6/ CBA F2 zygotes (hybrids between strains C57BL/6J and CBA/J) (Singh et al., 2015) at the Transgenic Facility of the University of Salamanca. Edited founders were identified by PCR amplification (Taq polymerase, NZYtech) with primers flanking the edited region (see Supplementary file 1k for primer sequences). The PCRs products were direct sequenced or subcloned into pBlueScript (Stratagene) followed by Sanger sequencing, selecting the founders carrying the desired alelles. The selected founders were crossed with wild-type mice to eliminate possible unwanted off-targets. Heterozygous mice were re-sequenced and crossed to give rise to edited homozygous. Genotyping was performed by analysis of the PCR products produced from genomic DNA extracted from tail biopsies. The primers and the expected amplicon sizes are listed in the Supplementary file 1k. Mouse mutants for Rec8, Six6os1 and Psma8 have been previously described (Bannister et al., 2004;Gó mez-H et al., 2019;Gó mez-H et al., 2016).

Ethics statement
All the experiments were approved by the Ethics Committee for Animal Experimentation of the University of Salamanca (USAL) and the Ethics committee of the Spanish Research Council (CSIC) under protocol #00-245. Accordingly, all the mouse protocols used in this work have been approved by the Animal Experimentation committees mentioned above. Specifically, mice were always housed in a temperature-controlled facility (specific pathogen free, spf) using individually ventilated cages, standard diet and a 12 hr light/dark cycle, according to EU law (63/2010/UE) and the Spanish royal law (53/2013) at the "Servicio de Experimentació n Animal, SEA. In addition, animal suffering was always minimized, and we made every effort to improve animal welfare during the life of the animals. The mice analysed were between 2 and 4 months of age, except in those experiments where the age is indicated.

Histology
For histological analysis, after the necropsy of the mice their testes or ovaries were removed and fixed in Bouin´s fixative or formol 10%, respectively. They were processed into serial paraffin sections and stained with haematoxylin-eosin (ovaries) or Periodic acid-Schiff (PAS) and hematoxylin (testes). The samples were analysed using a microscope OLYMPUS BX51 and images were taken with a digital camera OLYMPUS DP70. For TUNEL assay, sections were deparaffinized and apoptotic cells were detected with the In Situ Cell Death Detection Kit (Roche) and counterstained with DAPI.

Follicle counting
The inner third of each ovary was serially sliced into 5 mm thick sections and follicles were counted every five sections and classified into four stages (primordial, primary, secondary and antral). Only those follicles in which the nucleus of the oocyte was clearly visible were counted.

Epididymal sperm count
The epididymides were removed, minced and incubated in 1,5 ml of KSOM for 30 min at 37˚C to release sperm into the medium. The suspension was incubated for 10 min at 60˚C and the total sperm count was quantified by using a hemacytometer.

Fertility assessment
Hsf2bp +/+ and Hsf2bp S167L/S167L males and females (8 weeks old) were mated with WT females and males, respectively, over the course of 4-12 months. six mice per genotype (seven mice for Hsf2bp S167L/S167L females) were crossed. The presence of copulatory plug was examined daily and the number of pups per litter was recorded.

Immunocytology and antibodies
Testes were detunicated and processed for spreading using a conventional 'dry-down' technique or for squashing (Gó mez-H et al., 2016). Oocytes from fetal ovaries (E16.5, E17.5 and E19.5 embryos) were digested with collagenase, incubated in hypotonic buffer, disaggregated and fixed in paraformaldehyde. Rabbit polyclonal antibodies against HSF2BP and BRME1 were developed by Protein-techTM against a fusion protein of poly-His with full length HSF2BP or BRME1 (pUC57 vector) of mouse origin. Two antibodies (named R1 and R2) were generated against each protein (HSF2BP or BRME1) by immunization of two different host rabbits. Rabbit polyclonal antibody against DMC1 was developed by ProteintechTM against a DMC1 peptide (EESGFQDDEESLFQDIDLLQKHGINMA-DIKKLKSVGICTIKG). The primary antibodies used for immunofluorescence were rabbit aHSF2BP R2

Image acquisition and analysis
Slides were visualized at room temperature using a microscope (Axioplan 2; Carl Zeiss, Inc) with 63 Â objectives with an aperture of 1.4 (Carl Zeiss, Inc). Images were taken with a digital camera (ORCA-ER; Hamamatsu) and processed with OPENLAB 4.0.3 and Photoshop (Adobe). The slides from the different genotypes used for comparative analyses were all freshly prepared in parallel and immunofluorescence were also carried out in parallel with the same freshly prepared cocktail of antibodies. Slides were not frozen to avoid differences in the background and antigen reactivity. All the images acquired were taken with constant exposure times for comparison. Quantification of foci and fluorescence intensity were performed using Image J software. Only the axis-associated foci were counted. For colocalization analysis, the same nucleus was quantified without rotation (experiment) and after rotating 90 degrees one of the images. This condition allows to determine non-specific colocalization (random). Background was subtracted for intensity quantification. Squashed preparations were visualized with a Delta vision microscopy station. Stimulated emission depletion (STED) microscopy (SP8, Leica) was used to generate the super-resolution images. Secondary antibodies for STED imaging were conjugated to Alexa 555 and 488 (Invitrogen) and the slides were mounted in Prolong Antifade Gold without DAPI.

Y2H assay and screening
Y2H assay was performed using the Matchmaker Gold Yeast Two-Hybrid System (Clontech) according to the manufacturers' instructions. Mouse Hsf2bp cDNA was subcloned into the vector pGBKT7 and was used as bait to screen a mouse testis Mate and Plate cDNA library (Clontech Laboratories Inc). Positive clones were initially identified on double dropout SD (synthetic dropout)/-Leu /-Trp/Xa-Gal/Aureobasidin A plates before further selection on higher stringency quadruple dropout SD /-Ade /-His /-Leu /-Trp/X-a-Gal/Aureobasidin A plates. Pray plasmids were extracted from the candidate yeast clones and transformed into Escherichia coli. The plasmids from two independent bacteria colonies were independently grown, extracted and Sanger sequenced.
Doublestranded DNA annealing was carried out in 50 mM NaCl, 25 mM Tris-HCl, pH 7.5 buffer with complementary sequences at molecular equivalence by a denaturing step (5 min at 95˚C) and a slow return to room temperature. DNA was immobilized onto Dynabeads M-280 Streptavidin (Dynal) following the manufacturer instructions (0.2 pmol per 1 mg of beads). Protein extracts were obtained from in vitro coupled transcription/translation systems (TNT T7 Coupled Reticulocyte Lysate Systems, Promega) according to manufacturer's protocol. 15 ml of Flag-tagged proteins from TNT assays were pre-incubated on ice for 10 min in modified DBB (DBB: 50 mM Tris HCl, 100 mM NaCl, 10% (w/v) glycerol, Complete Protease inhibitor, 1 mM 2-mercaptoethanol pH 7,4 modified with 25 mM Tris-HCl, 1 mM EDTA plus 5 mg/ml BSA). After this preincubation 500 mg Dynabeads with immobilized ss-or ds-DNA were added and incubated for 1 hr at 4˚C under agitation. Then the beads were washed three times (5 min rotating at RT) in 700 ml of modified DBB without BSA, before being washed once in 700 ml of rinsing buffer (modified DBB with 150 mM NaCl). Finally, DNA-binding proteins were eluted by resuspending the beads in 30 ml of Laemmli buffer boiling the samples for 5 min. The samples were analyzed by western blot.

Cell lines and transfections
HEK293T and U2OS cell lines were obtained from the ATCC and transfected with Jetpei (PolyPlus) according to the manufacturer protocol. Cell lines were tested for mycoplasma contamination using the Mycoplasma PCR ELISA (Sigma).

Testis immunoprecipitation
Testis extracts were prepared in 50 mM Tris-HCl (pH8), 500 mM NaCl, 1 mM EDTA 1% Triton X100. 4 mg of protein were incubated with 10 mg of the specific antibody against the protein to be immunoprecipitated for 2 hr at 4˚C rotating. Then 50 ml of sepharose beads (GE Healthcare) were added to the protein-Ab mixture and incubated overnight at 4˚C with rotation. After that, the proteinbounded beads were washed four times with 500 ml of the extraction buffer by centrifugating 1 min at 10,000 rpm and 4˚C. Finally, the co-immunoprecipitated proteins were eluted from the beads by resuspending the beads in 50 ml Laemmli buffer and boiling for 5 min. The samples were analyzed by western blot.
Testis immunoprecipitation coupled to mass spectrometry analysis 200 mg of antibodies R1 and R2 against BRME1 (two independent IPs) and IgG from rabbit (negative control) were crosslinked to 100 ul of sepharose beads slurry (GE Healthcare). Testis extracts were prepared in 50 mM Tris-HCl (pH8), 500 mM NaCl, 1 mM EDTA 1% Triton X100. 20 mg of protein extracts were incubated o/n with the sepharose beads. Protein-bound beads were packed into columns and washed in extracting buffer for three times. Proteins were eluted in 100 mM glycine pH3 and analysed by Lc-MS/MS shotgun in LTQ Velos Orbitrap at the Proteomics facility of Centro de Investigació n del Cá ncer (CSIC/University of Salamanca).

Mass spectrometry data analysis
Raw data were analysed using MaxQuant v 1.6.2.6 (Cox and Mann, 2008) against SwissProt Mouse database (UP000000589, Oct, 2019) and MaxQuant contaminants. All FDRs were of 1%. Variable modifications taken into account were oxidation of M and acetylation of the N-term, while fixed modifications included considered only carbamidomethylation of C. The maximum number of modifications allowed per peptide was of 5. Proteins were quantified using iBAQ (Schwanhäusser et al., 2011). Potential contaminants, reverse decoy sequences and proteins identified by site were removed. Proteins with less than two unique peptides in the R1 and R2 groups were not considered for ulterior analysis. Proteins with less than two unique peptides in the control group and more than two in both groups R1 and R2 were selected as high-confidence candidates (group R1 and R2 only). An additional group of putative candidates was selected for those proteins with two or more unique peptides in one of the R1 or R2 groups and no unique peptides in the control sample (groups R1 only and R2 only, respectively).

Statistics
In order to compare counts between genotypes, we used the Two-tailed Welch's t-test (unequal variances t-test), which was appropriate as the count data were not highly skewed (i.e., were reasonably approximated by a normal distribution) and in most cases showed unequal variance. We applied a two-sided test in all the cases. Asterisks denote statistical significance: *p-value<0.05, **p-value<0.01, ***p-value<0.001 and ****p-value<0.0001.

Ethics
Animal experimentation: All the experiments were approved by the Ethics Committee for Animal Experimentation of the University of Salamanca (USAL) and the Ethics committee of the Spanish Research Council (CSIC) under protocol #00-245. Accordingly, all the mouse protocols used in this work have been approved by the above mentioned Animal Experimentation committees. Specifically, mice were always housed in a temperature-controlled facility (specific pathogen free, spf) using individually ventilated cages, standard diet and a 12 h light/dark cycle, according to EU law (63/ 2010/UE) and the Spanish royal law (53/2013) at the "Servicio de Experimentació n Animal, SEA. In addition, animal suffering was always minimized and we made every effort to improve animal welfare during the life of the animals.  . Transparent reporting form Data availability All data generated or analysed during this study are included in the manuscript and supporting files.