The cohesin subunit RAD21L functions in meiotic synapsis and exhibits sexual dimorphism in fertility
2011; Springer Nature; Volume: 30; Issue: 15 Linguagem: Inglês
10.1038/emboj.2011.222
ISSN1460-2075
AutoresYurema Herrán, Cristina Gutiérrez‐Caballero, Manuel Sánchez‐Martín, Teresa Hernández, Alberto Viera, José Luís Barbero, Enrique de Álava, Dirk G. de Rooij, José Á. Suja, Elena Llano, Alberto M. Pendás,
Tópico(s)Nuclear Structure and Function
ResumoArticle8 July 2011free access The cohesin subunit RAD21L functions in meiotic synapsis and exhibits sexual dimorphism in fertility Yurema Herrán Yurema Herrán Instituto de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno S/N, Salamanca, Spain Search for more papers by this author Cristina Gutiérrez-Caballero Cristina Gutiérrez-Caballero Instituto de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno S/N, Salamanca, Spain Search for more papers by this author Manuel Sánchez-Martín Manuel Sánchez-Martín Departamento de Medicina, Campus Miguel de Unamuno S/N, Salamanca, Spain Search for more papers by this author Teresa Hernández Teresa Hernández Instituto de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno S/N, Salamanca, Spain Search for more papers by this author Alberto Viera Alberto Viera Unidad de Biología Celular, Universidad Autónoma de Madrid, Madrid, Spain Search for more papers by this author José Luis Barbero José Luis Barbero Departamento de Proliferación Celular y Desarrollo, Centro de Investigaciones Biológicas (CSIC), Madrid, Spain Search for more papers by this author Enrique de Álava Enrique de Álava Instituto de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno S/N, Salamanca, Spain Search for more papers by this author Dirk G de Rooij Dirk G de Rooij Center for Reproductive Medicine, Academic Medical Center, University of Amsterdam, The Netherlands Search for more papers by this author José Ángel Suja José Ángel Suja Unidad de Biología Celular, Universidad Autónoma de Madrid, Madrid, Spain Search for more papers by this author Elena Llano Corresponding Author Elena Llano [email protected] Instituto de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno S/N, Salamanca, Spain Departamento de Fisiología, Campus Miguel de Unamuno S/N, Salamanca, Spain Search for more papers by this author Alberto M Pendás Corresponding Author Alberto M Pendás [email protected] Instituto de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno S/N, Salamanca, Spain Search for more papers by this author Yurema Herrán Yurema Herrán Instituto de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno S/N, Salamanca, Spain Search for more papers by this author Cristina Gutiérrez-Caballero Cristina Gutiérrez-Caballero Instituto de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno S/N, Salamanca, Spain Search for more papers by this author Manuel Sánchez-Martín Manuel Sánchez-Martín Departamento de Medicina, Campus Miguel de Unamuno S/N, Salamanca, Spain Search for more papers by this author Teresa Hernández Teresa Hernández Instituto de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno S/N, Salamanca, Spain Search for more papers by this author Alberto Viera Alberto Viera Unidad de Biología Celular, Universidad Autónoma de Madrid, Madrid, Spain Search for more papers by this author José Luis Barbero José Luis Barbero Departamento de Proliferación Celular y Desarrollo, Centro de Investigaciones Biológicas (CSIC), Madrid, Spain Search for more papers by this author Enrique de Álava Enrique de Álava Instituto de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno S/N, Salamanca, Spain Search for more papers by this author Dirk G de Rooij Dirk G de Rooij Center for Reproductive Medicine, Academic Medical Center, University of Amsterdam, The Netherlands Search for more papers by this author José Ángel Suja José Ángel Suja Unidad de Biología Celular, Universidad Autónoma de Madrid, Madrid, Spain Search for more papers by this author Elena Llano Corresponding Author Elena Llano [email protected] Instituto de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno S/N, Salamanca, Spain Departamento de Fisiología, Campus Miguel de Unamuno S/N, Salamanca, Spain Search for more papers by this author Alberto M Pendás Corresponding Author Alberto M Pendás [email protected] Instituto de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno S/N, Salamanca, Spain Search for more papers by this author Author Information Yurema Herrán1,‡, Cristina Gutiérrez-Caballero1,‡, Manuel Sánchez-Martín2, Teresa Hernández1, Alberto Viera3, José Luis Barbero4, Enrique de Álava1, Dirk G de Rooij5, José Ángel Suja3, Elena Llano *,1,6 and Alberto M Pendás *,1 1Instituto de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno S/N, Salamanca, Spain 2Departamento de Medicina, Campus Miguel de Unamuno S/N, Salamanca, Spain 3Unidad de Biología Celular, Universidad Autónoma de Madrid, Madrid, Spain 4Departamento de Proliferación Celular y Desarrollo, Centro de Investigaciones Biológicas (CSIC), Madrid, Spain 5Center for Reproductive Medicine, Academic Medical Center, University of Amsterdam, The Netherlands 6Departamento de Fisiología, Campus Miguel de Unamuno S/N, Salamanca, Spain ‡These authors contributed equally to this work *Instituto de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno, 37007 Salamanca, Spain. Tel.: +34 92 329 4809; Fax: +34 92 329 4743; E-mail: [email protected] de Biología Molecular y Celular del Cáncer (CSIC–USAL), Campus Miguel de Unamuno, 37007 Salamanca, Spain. Tel.: +34 92 329 4809; Fax: +34 92 329 4743; E-mail: [email protected] The EMBO Journal (2011)30:3091-3105https://doi.org/10.1038/emboj.2011.222 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The cohesin complex is a ring-shaped proteinaceous structure that entraps the two sister chromatids after replication until the onset of anaphase when the ring is opened by proteolytic cleavage of its α-kleisin subunit (RAD21 at mitosis and REC8 at meiosis) by separase. RAD21L is a recently identified α-kleisin that is present from fish to mammals and biochemically interacts with the cohesin subunits SMC1, SMC3 and STAG3. RAD21L localizes along the axial elements of the synaptonemal complex of mouse meiocytes. However, its existence as a bona fide cohesin and its functional role awaits in vivo validation. Here, we show that male mice lacking RAD21L are defective in full synapsis of homologous chromosomes at meiotic prophase I, which provokes an arrest at zygotene and leads to total azoospermia and consequently infertility. In contrast, RAD21L-deficient females are fertile but develop an age-dependent sterility. Thus, our results provide in vivo evidence that RAD21L is essential for male fertility and in females for the maintenance of fertility during natural aging. Introduction Structurally, the somatic cohesin complex consists of four subunits: two members of the family of proteins responsible for the structural maintenance of chromosome (SMC1α and SMC3) that heterodimerize, one kleisin subunit that closes the ring (Scc1/RAD21) and is the substrate for the protease separase, and a HEAT repeat domain protein (SA1/STAG1 or SA2/STAG2). In vertebrates, during prophase, most of the cohesins are dissociated from the chromatid arms by phosphorylation of the STAG1/2 subunit by PLK1 (polo-like-kinase 1) (Losada et al, 2002; Sumara et al, 2002). The remaining centromeric cohesins are released from chromosomes at the onset of anaphase by the cleavage of RAD21 by separase (Musacchio and Salmon, 2007). During meiosis, two rounds of chromosome segregation follow a single round of replication to generate haploid gametes. The first meiotic division differs from mitosis in that homologous chromosomes pair, synapse, recombine and segregate to opposite poles as a result of their mono-orientation. The second meiotic division is similar to mitosis since the two recombined chromatids segregate to opposite poles (bi-orientation). During the onset of anaphase I, loss of sister chromatid arm cohesion occurs following separase-dependent cleavage of REC8, that replaces RAD21 during meiosis (Kudo et al, 2006, 2009). However, centromeric cohesion is maintained by the protective action of shugoshin-like-2 preventing separase-mediated cleavage of REC8 (Llano et al, 2008). This mechanism enables bi-orientation of recombined homologues. Once chromosomes have congressed at the metaphase II plate, separase is reactivated and centromeric cohesin complexes are released to allow chromatid segregation. In addition to REC8 (Parisi et al, 1999), a meiotic paralogue of RAD21, there are also meiosis-specific mammalian paralogues of SMC1α, and STAG1–2, that is SMC1β and STAG3, respectively (Prieto et al, 2001; Gruber et al, 2003). Aside from these canonical functions, the cohesin complexes also participate in somatic homologous recombination between sister chromatids allowing the assembly of recombinational repair complexes, as well as recombination between homologous chromatids by assembly of the synaptonemal complex (SC) in meiocytes (Klein et al, 1999; Hartsuiker et al, 2001). The SC consists of a proteinaceous structure, the axial element (AE), allowing the association of each pair of sister chromatids. After pairing, the AEs are called lateral elements (LEs) to which transverse filaments (TFs) associate to give rise to the tripartite SCs. The SC provides the structural framework for synapsis, double-strand break (DSB) repair and exchange between homologues (Henderson and Keeney, 2005). During prophase I, most if not all, cohesin subunits expressed in mammalian spermatocytes colocalize with SYCP3, a structural AE/LE component (reviewed in Suja and Barbero, 2009). In fission and budding yeast, the RAD21/Scc1 α-kleisin of the cohesin complex is replaced by the meiosis-specific REC8 protein. Yeast rec8 mutants exhibit premature sister chromatid separation during prophase I, that are defective for the assembly of the SC (Klein et al, 1999). It has been assumed that most eukaryotes display a dual α-kleisin system (REC8 versus RAD21) similar to the well-studied system of Schizosaccharomyces pombe and Saccharomyces cerevisae, aside from the notable exception of Caenorhabditis elegans (Severson et al, 2009). Very recently, we and two other groups have biochemically characterized a new member of the α-kleisin family of proteins, RAD21L (Gutiérrez-Caballero et al, 2011; Ishiguro et al, 2011; Lee and Hirano, 2011). RAD21L, is a paralogue of RAD21 and it is transcribed more abundantly in testis and has been postulated to be a canonical cohesin subunit. RAD21L interacts with SMC3, SMC1α/β and STAG3 (Gutiérrez-Caballero et al, 2011; Ishiguro et al, 2011; Lee and Hirano, 2011). Consequently, the protein is localized to the AEs/LEs in meiocytes. From the four meiotic-specific cohesins described (REC8, STAG3, SMC1β and RAD21L), loss of function mouse models for REC8 (Bannister et al, 2004; Xu et al, 2005) and SMC1β (Revenkova et al, 2004) have been developed. REC8 mutant male and female mice are sterile and show severe defects in synapsis, and chiasma formation (Bannister et al, 2004; Xu et al, 2005). SMC1β-deficient males show a pachytene arrest whereas mutant females present a premature loss of cohesion at metaphase II that leads to sterility (Revenkova et al, 2004). In this work, we describe the precise localization of RAD21L in mouse spermatocytes and its functional characterization by a gene targeted mutation in the mouse. We provide cytological and in vivo evidence showing that the roles of RAD21L differ from those of RAD21 and REC8, and that RAD21L is as essential as REC8 for driving the initial steps of prophase I in male meiosis. RAD21L-deficient males show a defect in chromosome synapsis at prophase I, which provokes an arrest at a zygotene-like stage leading to total azoospermia. In contrast, RAD21L-deficient females are fertile but develop an age-dependent sterility. Thus, our results demonstrate for the first time that the recently identified RAD21L is a functionally relevant meiotic α-kleisin, which is essential for male fertility and for the maintenance of fertility during natural aging. Results and discussion Immunolocalization of the RAD21L protein The recently identified third member of the α-kleisin protein family in mammals, RAD21L, is expressed in spermatocytes throughout meiosis I (Ishiguro et al, 2011; Lee and Hirano, 2011), with some discrepancies in relation with its time of disappearance (pachytene versus metaphase I). In order to assess the localization of RAD21L, we carried out a detailed analysis of mouse spermatocytes spreads using immunofluorescent (IF) antibodies. RAD21L was first detected at the leptotene stage as short threads that colocalized with SYCP3 along developing AEs (Figure 1A–D). During zygotene, RAD21L colocalized with SYCP3 at both the autosomal AEs/LEs, and the unsynapsed AEs of the sex chromosomes (Figure 1E–H). In early pachytene, RAD21L was detected as lines along the autosomal SCs where it colocalized with SYCP3. Further signals for RAD21L were found at the pseudoautosomal region of homology between the sex chromosomes, where their AEs are synapsed. Furthermore, there was some additional staining in the unsynapsed AEs of the XY bivalent (Figure 1I and J). By late pachytene, there was an increase in RAD21L labelling on the sex chromosomal AEs and on the chromatin of the sex body (Figure 1K and L). This localization contrasts with the observed weak staining of REC8 at the AEs of the sex chromosomes at pachytene (see asterisks in Figure 7Q and R). In early diplotene, the intensity of the RAD21L labelling decreased along the desynapsing and still synapsed LEs (Figure 1M–P) to finally disappear by mid-diplotene (Figure 1Q and R). Concomitantly, RAD21L labelling began to accumulate at centromeres (Figure 1O–T) while it was progressively lost from the AEs and the chromatin of the sex chromosomes (Figure 1M–T). During diakinesis, RAD21L was highly enriched at the centromeres of all autosomes and was not detected along the desynapsed LEs. However, there was a faint RAD21L signal at the unsynapsed AEs of the sex chromosomes (Figure 2A and B). This pattern of RAD21L distribution remained during metaphase I (Figure 2C–F). At higher magnification, metaphase I autosomal bivalents show RAD21L signal at their centromeres but the labelling did not completely colocalize with SYCP3 at the inner centromere domain (ICD) (Figure 2G–I). With regards to the metaphase I sex bivalent, a faint RAD21L signal was observed along its interchromatid domain (Figure 2J). In addition, the centromeric RAD21L signal at the Y was larger than that at the centromere of the X chromosome (Figure 2J). The labelling of RAD21L was similar for both metaphase I and anaphase I (Figure 2F and K). During the second meiotic division, RAD21L was detected as a pair of brightly stained spots at the centromeres of metaphase II chromosomes (Figure 2L), and as single spots in segregating chromatids at anaphase II (Figure 2M). Figure 1.Distribution of RAD21L during prophase I. Double immunolabelling of RAD21L (green) and SYCP3 (red) in spread spermatocytes. (A–D) During leptotene, RAD21L appears as a succession of small dots that colocalize with SYCP3 along developing AEs. (E–H) During zygotene, RAD21L and SYCP3 colocalize along AEs/LEs. Sex chromosomes (X, Y) have still not synapsed. (I–L) RAD21L colocalizes with SYCP3 along autosomal SCs and sex chromosomes (XY) AEs. In late pachytene (K, L), RAD21L appears enriched at the chromatin of the sex body (XY) and at their AEs. (M–P) In early diplotene, RAD21L vanishes along desynapsing autosomal LEs, but is still enriched at the sex AEs and at the sex body (XY). (Q–T) In mid and late diplotene, RAD21L appears at the centromeres and along the sex AEs, and faintly at the sex body (XY). Download figure Download PowerPoint Figure 2.Distribution of RAD21L during diakinesis and meiotic divisions. Double immunolabelling of RAD21L (green) and SYCP3 (red) and counterstaining of chromatin with DAPI (blue) in spread spermatocytes. (A–F) During early diakinesis (A, B) and metaphase I (C–F), RAD21L is present at the centromeres of all chromosomes and at the interchromatid domain of the sex bivalent (XY). Arrowheads mark the enlargements of SYCP3 along the X chromosome in diakinesis, and the large agglomerates of SYCP3 in the cytoplasm of metaphase I spermatocytes. (G–I) Enlargements of three selected metaphase I bivalents. RAD21L is enriched at the centromeres but does not completely localize with SYCP3, and is not present at the interchromatid domain where SYCP3 is detected. (J) Selected metaphase I sex bivalent. RAD21L appears as a faint signal along the interchromatid domain and as bright signals at the centromeres. The RAD21L signal at the centromere of the Y chromosome (Y) is larger than that present at the X chromosome (X). (K) RAD21L partially colocalizes with SYCP3 at anaphase I centromeres. Arrowhead marks an SYCP3 agglomerate. (L, M) RAD21L appears as a pair of signals at metaphase II centromeres (L) and as single signals at anaphase II centromeres (M). Download figure Download PowerPoint Our results partially agree with those very recently reported on the distribution of RAD21L in mouse spermatocytes by Ishiguro et al (2011) and to a lesser extent with those reported by Hirano's group (Lee and Hirano, 2011). However, there are some differences with respect to the distribution pattern of RAD21L along the AEs/LEs. Besides the remarkable divergence in the timing of disappearance of RAD21L during meiosis I (pachytene versus metaphase I), these two groups described that RAD21L and REC8 localize as discontinuous (mutually exclusive) lines along zygotene AEs/LEs and pachytene SCs. Based on this, Ishiguro et al (2011) have proposed a cohesin ‘barcode’ model where meiosis-specific cohesin complexes with either RAD21L or REC8 have intrinsic and alternating loading sites along the AEs/LEs, which might facilitate homologous pairing. However, our antibodies detected continuous lines along SCs. These discrepancies might be due to differences in image acquisition, different sensitivity of the antibodies used, or dilutions employed. This aspect will need further clarification. The three kinds of cohesin complexes comprised of RAD21, RAD21L or REC8 might have different functions during both meiotic divisions since their distribution and dynamics are different not only during prophase I, but also during metaphase I and metaphase II. For instance, in metaphase II chromosomes, RAD21L appears as two separate signals at each centromere (Figure 2L), which is in contrast to RAD21 that has not been detected at centromeres (Parra et al, 2004), while REC8 appears at the ICD as one spot between sister kinetochores consistent with its role in centromere cohesion (Kudo et al, 2006). While the distribution of REC8 is consistent with its function in maintaining both arm and centromere cohesion during the two meiotic divisions (Kudo et al, 2006, 2009; Tachibana-Konwalski et al, 2010), the localization of RAD21 and RAD21L suggests that they might have different roles. In this regard, the accumulation of RAD21L at centromeres during diplotene, its enrichment at the centromere of the Y chromosome during metaphase I, and its distribution at metaphase II centromeres is strikingly similar to the distribution of the shugoshin-like-2 (Gómez et al, 2007; Llano et al, 2008) and MCAK (Parra et al, 2006) during male mouse meiosis. Thus, the enrichment of RAD21L at centromeres during diplotene might contribute to the assembly of the ICD (Parra et al, 2009). Gene disruption of Rad21l To address the function of RAD21L and to validate genetically that it constitutes a functional subunit of a novel meiotic cohesin, we created a targeted mutation of the murine Rad21l locus by an insertional strategy that disrupts the open reading frame (ORF) of the locus (Supplementary Figure S1A and B; Adams et al, 2004). Heterozygous targeted mice transmitted the mutation to the offspring at Mendelian frequencies (1:2:1). RT–PCR was used to evaluate the interruption of the ORF of the Rad21l gene in the homozygous targeted mice (Supplementary Figure S1C; see Materials and methods). The absence of the protein in these homozygous targeted mice was also validated using two different antibodies, which were specific against RAD21L (Supplementary Figures S1E and S2; Gutiérrez-Caballero et al, 2011). Consequently, spermatocytes from homozygous targeted mice did not show RAD21L immunofluorescence (Supplementary Figure S1D). Heterozygous targeted mice showed neither cellular nor aberrant organismal phenotypes (indicative of the lack of a gain of function). Thus, we concluded that the mutation is functionally a null allele of Rad21l. Histological analysis and male infertility in Rad21l−/− mice Rad21l−/− mice developed normally and displayed no overt phenotype. However, while female mice lacking RAD21L were fertile, males were sterile since they failed to produce offspring. Testes from Rad21l−/− mice were on average 70% smaller than those from wild-type mice, and their epididymides lacked spermatozoa (Figure 3A and B). Histopathological analysis revealed an absence of postmeiotic cell types despite of the presence of spermatogonia, and Sertoli and Leydig cells (Figure 3A). Within a mouse testis, the seminiferous epithelium contains a mixture of germ cells at various developmental stages. Staging of each tubule section is defined (from I to XII) according to the group of associated germ cell types that are present (Russell, 1990). Following this criteria, mutant mice appeared to be arrested at stage IV of the epithelial cycle (Figure 3A). FACS analysis of whole cells from seminiferous tubules was carried out and sustained the prophase I arrest by the absence of the haploid compartment in Rad21l−/− testes (Figure 3B). In order to rule out proliferation defects in spermatogonia, PCNA immunostaining of wild-type and Rad21l−/− tubules was performed and no differences in the basal layer of PCNA-positive cells were found (Figure 3C). Given the lack of spermatozoa, we carried out TUNEL staining and showed that the prevalence of apoptotic cells in Rad21l−/− tubules was higher than in wild type (Figure 3C). Finally, studying the histology of the testis, it became clear that spermatogenesis proceeds apparently normal up to prophase I. Then, in stage IV, there is a massive apoptosis of spermatocytes. Extensive apoptosis was also observed at 19 days of age (Figure 3D), indicating that spermatocytes of the first wave of spermatogenesis were already affected. Thus, we conclude that RAD21L is essential for spermatogenesis in the mouse and its deficiency provokes total azoospermia that leads to infertility. Figure 3.The absence of RAD21L provokes azoospermia. (A) The deficiency of RAD21L promotes a complete block of mouse spermatogenesis. Genetic ablation of Rad21l leads to a reduction of the testis size, and an arrest of spermatogenesis in epithelial stage IV, identified by the presence of intermediate spermatogonia (arrows) about to divide into B spermatogonia. Massive apoptosis of spermatocytes (asterisks) can be seen. The spermatogenic arrest leads to empty epididymides and azoospermia. Bar in upper panels, 100 μm and in lower panels, 25 μm. (St) Seminiferous tubules. (Ep) Epididymides. (B) Abnormal ploidy of Rad21l−/− spermatocytes. FACS analysis of cells from seminiferous tubules showing the absence of the haploid compartment in Rad21l−/− testes. (C) Immunohistochemical detection of proliferating cells with anti-PCNA and apoptotic cells by TUNEL staining show the absence of proliferative defects and an increase of apoptotic cells in Rad21l−/− seminiferous tubules, respectively. Bar in both panels, 25 μm. (D) Tubule degeneration in juvenile mice (13 days postpartum (d.p.p.) and 19 d.p.p.) lacking RAD21L and spermatogenic arrest prior to pachytene studied by histology of testes from Rad21l+/+ and Rad21l−/− males. At 13 d.p.p., spermatogenesis has reached to late zygotene and at 19 d.p.p. to late pachytene. Spermatocyte apoptosis (asterisks) was first seen in 19 d.p.p. Download figure Download PowerPoint SC morphology and synapsis in mutant spermatocytes To functionally analyse the infertility in the mutant mice and to more precisely characterize the meiotic arrest, we first studied the assembly of the SC. Spermatocytes spreads were studied and staged by staining for SYCP3. It appeared that in the absence of RAD21L, synapsis between homologues was not completed (Figure 4A and B). To determine the extent of the disruption of synapsis, we monitored the distribution of the TF protein SYCP1 as colabelling of SYCP3 and SYCP1 highlights regions of synapsis in wild types. Mutant spermatocytes did not proceed beyond zygotene-like stage (Figure 4A). This blockade was further supported by the absence of immunolabelling for the mid-pachytene-specific histone variant H1T (Supplementary Figure S3), supporting the observed arrest at epithelial stage IV as determined by histological analysis (Figure 3A). Using SYCP3 staining of the zygotene-like spermatocytes from Rad21l−/− mice, we observed discontinuous/fragmented stretches of AEs that did not progress to the expected 19 fully synapsed autosomal bivalent chromosomes (Figure 4A and B). Furthermore, a fraction of the arrested spermatocytes displayed ring-like structures (Figure 4B, arrowhead) and synapsis between non-homologous chromosomes occurred (Figure 4A and B, arrows). To further analyse the synaptic defects, we investigated the centromere distribution by immunolabelling with a human anti-centromere antibody (ACA) (Figure 4B). In wild-type leptotene spermatocytes, the number of centromere signals never exceeded 40. As synapsis progressed, these centromeric foci diminished to 21 (19 signals from synapsed autosomes + 2 signals of the XY bivalent) at pachytene when homologous pairing of autosomes is complete and their centromeres are very closely juxtaposed (Figure 4B). In Rad21l−/− zygotene-like spermatocytes, we scored on average 30±3.5 foci (30 nuclei analysed). This result also points to a deficient synapsis between homologues, at least at their centromeric regions (Figure 4B). In order to further study this failure of synapsis, we stained spermatocytes for the kinase ATR and the DNA-binding protein TOPBP1 as these reliably stain the unsynapsed AEs/LEs at leptotene–zygotene and the unsynapsed AEs and the chromatin of the sex body at pachytene (Perera et al, 2004). Moreover, TOPBP1 and ATR also accumulate at the unsynapsed AEs of mutant spermatocytes with a meiotic arrest such as Dmc1−/− and Msh5−/− spermatocytes (Barchi et al, 2005). Our immunolabelling results on wild-type spermatocytes revealed that ATR and TOPBP1 appeared as foci along the unsynapsed leptotene and zygotene AEs/LEs, whereas at mid-pachytene both proteins were restricted to the sex body (Figure 4C and D). RAD21L-deficient and wild-type spermatocytes showed a similar number of ATR and TOPBP1 foci at leptotene and zygotene AEs. However, as meiosis arrested at a zygotene-like stage in Rad21l−/−, these foci also persisted and were not eliminated (Figure 4C and D). In summary, RAD21L deficiency in mouse spermatocytes leads to abnormal AEs/LEs, which are fragmented and poorly aligned/synapsed (a large number of AEs are kept individually), some stretches of AEs and LEs are decorated with SYCP1 and synapsis between non-homologous chromosomes occurs. Figure 4.Rad21l−/− spermatocytes show defects in synapsis. (A) Double labelling of SYCP3 (red) and SYCP1 (green) showing fragmented AEs/LEs with aberrant synapsis and with patches of SYCP1 in mutant spermatocytes (arrow) as compared with their wild-type control. (B) Double immunolabelling of SYCP3 (red) and kinetochores (anti-centromere autoantibody, ACA (green)) in Rad21l+/+ and Rad21l−/− spermatocytes. In wild-type spermatocytes, the number of ACA signals is reduced from 40 to 21 between zygotene to pachytene stage. These signals localize at one end of the AEs/SCs. In Rad21l−/− spermatocytes, synapsis is incomplete and the number of ACA signals is always higher than 21. The presence of some centromeres along the same synapsed region (blue arrow), some unsynapsed AEs between synapsed regions (arrow) and the presence of ring structures formed by chromosomes with two neighbouring centromeres (arrowhead) are indicatives of non-homologous synapsis. (C, D) Double immunolabelling of SYCP3 (red) and ATR or TOPBP1 (green) in wild-type or Rad21l−/− spermatocytes. In wild-type spermatocytes, ATR (C) and TOPBP1 (D) proteins localize to unsynapsed AEs. At pachytene, these proteins only appear at the sex body. In Rad21l−/− zygotene-like spermatocytes, these proteins remain accumulated at AEs. *Sex body (XY). Download figure Download PowerPoint Defective DSB processing in the mutant spermatocytes The absence of REC8 leads to severe defects in DSB processing in yeast and to a lesser extent in mouse meiosis (Klein et al, 1999; Xu et al, 2005). Taking into account these data and the arrest observed in the Rad21l mutant, we studied whether RAD21L deficiency promotes a deficit in the repair of the programmed DSBs generated by the nuclease SPO11 at early leptotene, a frequent cause of meiotic arrest (Viera et al, 2009). Thus, we first monitored the formation of DSBs and analysed the presence of γ-H2AX histone variant, which is phosphorylated at prophase I in response to the SPO11-induced DSBs in an ATM-depen
Referência(s)