Senataxin plays an essential role with DNA damage response proteins in meiotic recombination and gene silencing

Abstract

Senataxin, mutated in the human genetic disorder ataxia with oculomotor apraxia type 2 (AOA2), plays an important role in maintaining genome integrity by coordination of transcription, DNA replication, and the DNA damage response. We demonstrate that senataxin is essential for spermatogenesis and that it functions at two stages in meiosis during crossing-over in homologous recombination and in meiotic sex chromosome inactivation (MSCI). Disruption of the Setx gene caused persistence of DNA double-strand breaks, a defect in disassembly of Rad51 filaments, accumulation of DNA:RNA hybrids (R-loops), and ultimately a failure of crossing-over. Senataxin localised to the XY body in a Brca1-dependent manner, and in its absence there was incomplete localisation of DNA damage response proteins to the XY chromosomes and ATR was retained on the axial elements of these chromosomes, failing to diffuse out into chromatin. Furthermore persistence of RNA polymerase II activity, altered ubH2A distribution, and abnormal XY-linked gene expression in Setx⁻/⁻ revealed an essential role for senataxin in MSCI. These data support key roles for senataxin in coordinating meiotic crossing-over with transcription and in gene silencing to protect the integrity of the genome.

Figure 1. Targeted disruption of the mouse Setx gene.

A. Diagram of the Setx wild type allele (WT), targeting vector, and mutant alleles (neo+ and KO). Primers used for PCR genotyping (In3F, In3R, LoxPR) and the length of the PCR fragments obtained for WT (In3F and In3R yielding a 600 bp product) and KO (In3F and LoxpR yielding a 339 bp product) are indicated. E, exon; I, intron. NeoR represents the neomycin cassette, and triangles the loxP sites. B. Representative image of PCR genotyping using In3F, In3R and LoxPR primers. Wild type (+/+), heterozygotes (+/−) and knockout (−/−) alleles generate PCR products of 600 bp, 600 bp and 339 bp, and 339 bp, respectively. A negative control for the PCR reaction (−ve) is also shown. M, 100 bp marker. C. RT-PCR of 35 day-old mice testes samples using primers specific to Setx cDNA indicates the absence of Setx expression in KO testes. GAPDH was used as an internal standard. D. Immunoprecipitation of senataxin using anti-human senataxin antibodies (Ab-1/Ab-3) from 35 day-old mouse testes extracts confirmed the absence of the protein in the Setx^−/−. Immunoprecipitation of senataxin from human lymphoblastoid cell extracts from normal (C3ABR) and an AOA2 patient (SETX2RM) confirmed the similar size of senataxin in both species. A species-matched non-specific serum (NSIg) was used as a negative control in for the IP experiments. As expected, no senataxin protein was brought down from Setx ^+/+ testes following the IP with the non-specific serum (NSIg).

Figure 2. Spermatogenesis is disrupted in Setx^−/− mice.

A. Testes from 35-day-old Setx^+/+ and Setx^−/− mice. B–C. Hematoxylin and eosin (H&E)-stained sections of testes from adults. Scale bar, 100 µm. D–E. Enlarged images of regions in (B) and (C). Asterisks in (C) and (E) show vacuolated seminiferous tubules in which both spermatozoa and spermatids are absent. Scale bar, 100 µm. F–G. H&E sections of epididymis from Setx^+/+ and Setx^−/− adult mice. While there are numerous spermatozoids (Spm) in Setx^+/+epididymis, no sperm is present in Setx^−/−. Scale bar, 100 µm. H. TUNEL-stained sections of testes from adult Setx^+/+ and Setx^−/− mice. Many TUNEL-positive cells are observed in Setx^−/− testes. Scale bar, 50 µm. (1) and (2) are magnifications of a seminiferous tubule. I. Increased number of apoptotic tubules in Setx^−/− (Tubules with ≥8 TUNEL-positive cells per tubule). Data is plotted as the mean±standard deviation, n = 1330. J. Block of meiosis at pachytene stage in Setx^−/− mice. Meiotic stage distribution (leptotene, zygotene, pachytene, diplotene) in Setx^+/+ and Setx^−/−. While a similar number of leptotene, zygotene and pachytene stage spermatocytes were found in both testes, no diplotene stage spermatocytes were found in Setx^−/− testes indicating a block at pachytene in KO animals. Data plotted as the mean±standard deviation obtained from 3 mice, 2000 spermatocytes were counted in total for both Setx^+/+ and Setx^−/−. K. Severe reduction in the number of pachytene spermatocytes in Setx^−/− compared to Setx^+/+ during the first meiotic division. No significant change in the percentage of pachytene cells was observed in Setx^+/+ from day 16 to 22. However, a significant reduction in the percentage of pachytene spermatocytes was observed at day 22 in Setx^−/− compared to Setx^+/+ (Student's t-test, p<0.01). Data plotted as the mean±standard error, n = 6000. * indicates p<0.01.

Figure 3. Defective meiotic recombination and crossover formation in infertile Setx^−/− males.

A. Initiation and repair of programmed DNA DSB as shown by γH2AX staining of spermatocytes spreads of Setx^+/+ and Setx^−/− adult mice. At pachytene, γH2AX staining is restricted to the XY chromosomes (circle) in Setx^+/+ spermatocytes, whereas some γH2AX foci remained on asynapsed autosomes indicating persistence of unrepaired DSB in Setx^−/−. Normal γH2AX staining of the XY chromosomes (circle) was observed in both Setx^+/+ and Setx^−/− pachytene stage spermatocytes. Scale bar, 20 µm. XY, sex chromosomes. B. Persistence of Rad51 foci at pachytene stage in Setx^−/− spermatocytes indicating the presence of unrepaired DSBs (compare 1 and 2). Scale bar, 20 µm. C. Quantitation of Rad51 foci revealed a 6-fold increase in the number of Rad51 foci at pachytene stage in Setx^−/− as compared to Setx^+/+ (Student's t-test, n = 50), * indicates p<0.05. D. Formation of chiasmata at pachytene stage in Setx^+/+ spermatocytes as marked by Mlh1 staining. No Mlh1 foci were detected in Setx^−/− pachytene cells indicating that crossovers do not occur in Setx^−/−. 1 and 2 represent magnification of autosomes. Scale bar, 20 µm. SCP3 or SCP1 were used to identify the meiotic stages. E. Defect in senataxin leads to R-loop structures accumulation in germ cells. Staining with S9.6 antibody (R-loops) on adult spermatocytes revealed an increased formation of R-loops in Setx^−/− germ cells. Scale bar, 20 µm. F. Number of pachytene spermatocytes showing none-faint, medium, and strong R-loop staining intensities for Setx^+/+ and Setx^−/−.

Figure 4. Accumulation of R-loops correlates with apoptosis.

A. R-loop and TUNEL co-staining of histological cross-sections of adult Setx^−/− mice revealed apoptosis in germ cells containing R-loops. Hoechst 33342 was used to stain for DNA. Scale bar, 50 µm. B. Pre-treatment of consecutive testes sections from the same animal with RNAse H dramatically reduced R-loop signal intensity in Setx^−/−. Similar results were obtained for consecutive sections from different Setx^−/− animals (data not shown). C. Quantitation of R-loop positive tubules in Setx^+/+ and Setx^−/−. The Y-axis represents the % of tubules containing at least one R-loop positive cell within the tubule. D. Higher number of R-loops positive germ cells per tubule (1–5 R-loop-positive cells/tubule, 6–10 R-loop-positive cells/tubule, and more than 11 R-loop-positive cells/tubules) in Setx^−/− compared Setx^+/+ confirming the role of senataxin in resolving R-loops in vivo. E. Correlation of apoptosis (TUNEL) with R-loop accumulation in Setx^−/− seminiferous tubules uncovers an essential role for senataxin to resolve R-loops and prevent germ cell apoptosis. Graphic representation of the number of tubules that contain TUNEL-only positive germ cells (white) and TUNEL and R-loop co-stained germ cells (black).

Figure 5. Senataxin localises to the sex chromosomes during meiosis.

A. Staining of spermatocytes spreads with an anti-senataxin antibody (Ab-1) revealed that senataxin localised in majority to the XY chromosomes (1) at pachytene stage in Setx^+/+. Some background staining was also observed on the autosomes. No senataxin was detected in Setx^−/− spermatocytes confirming the specificity of our senataxin antibody. (1) and (2) are magnification of the XY chromosomes. B. Double staining of Setx^+/+ pachytene spermatocytes with senataxin (Ab-1) and SCP3 revealed a diffuse localisation of senataxin to the XY body. Scale bar, 20 µm. C. Partial co-localisation of senataxin with the XY chromosome marker Brca1. While Brca1 stains exclusively the unsynapsed axis of the XY chromosomes, senataxin staining is more diffuse. D. Brca1 staining of the XY chromosomes in Setx^+/+ and Setx^−/− pachytene spermatocytes (day 20). In Setx^+/+, the unsynapsed axis of the XY chromosomes is entirely stained with Brca1 while an incomplete covering (white arrow) of the XY chromosomes is observed in Setx^−/−. Scale bar, 5 µm. E. Lack of senataxin recruitment to XY chromosomes in Brca1^Δ11/Δ11 p53^+/− as compared to Brca1^+/+ p53^+/−. F. Brca1 localised to only part of the unsynapsed axis of the XY chromosomes in Brca1^Δ11/Δ11 p53^+/− while Brca1 coated the entire unsynapsed axis of the XY chromosomes in Brca1^+/+ p53^+/−. G. Lack of evidence for an in situ direct endogenous interaction between senataxin and Brca1 on the XY chromosomes as revealed by negative Proximity Ligation Assay (PLA) results on pachytene spermatocyte spreads. Immunostaining of Setx^+/+ pachytene spermatocytes with Brca1 and Setx individually with SCP3 is also shown. H. Endogenous interaction between ATR and Brca1 was confirmed by PLA. Here, we reveal for the first time a direct endogenous interaction between Brca1 and ATR in situ over the XY chromosomes.

Figure 6. Defective localisation and diffusion of DNA damage response proteins in Setx^−/−.

A. Absence of ATR diffusion over the XY chromatin domain in Setx^−/− compared to Setx^+/+. Scale bar, 5 µm. B. Incomplete diffusion of MDC1 over the XY chromatin domain in Setx^−/−, as indicated by the white arrow. Scale bar, 5 µm. C. Reduced intensity and diffusion of γH2AX staining on the XY chromosomes in Setx^−/− compared to Setx^+/+. D. Altered XY chromosomes structure and formation in Setx^−/− as shown by SCP3 staining. Scale bar 5 µm. E. Percentage of Setx^+/+ and Setx^−/− pachytene spermatocytes at days 16, 20 and 22 with clearly distinguishable XY chromosomes. At every time point, a significant higher percentage of distinguishable XY chromosomes was observed in Setx^+/+ (p<0.01 according to Student's t-test, n = 3600). *,**,*** indicates p<0.01.

Figure 7. Aberrant meiotic sex chromosome inactivation in Setx^−/− mice.

A. Relative levels of transcript of XY-linked germ cell specific genes from Setx^+/+ and Setx^−/− testes was determined by RT-PCR to assess MSCI as previously described . Levels of Actb and Gapdh mRNA were used as positive controls for ubiquitous gene expression. Levels of Dazl transcripts were determined as a positive control for previously documented autosomal meiotic expression . Levels of Ube1x and Ube1y transcripts were included as positive controls for previously documented X- and Y-linked gene expression during spermatogenesis, respectively . X-linked Usp26, Fthl17 and Tktl1, and Y-linked Rbmy genes have previously shown evidence of MSCI . RT-PCR control without template is also shown. B. Immunostaining for the transcriptionally-active form of RNA Pol II (phospho-S2) revealed active transcription of the XY chromosomes in Setx^−/− spermatocytes (circle). In contrast, no signal for active Pol II was visible over the XY chromosomes in Setx^+/+ spermatocytes (circle) confirming the silencing of the XY chromosomes. Scale bar, 20 µm. C. Immunostaining for ubiquitinated histone H2A (ubi-H2A) in Setx^+/+ and Setx^−/− spermatocytes reveals a specific staining on the XY chromosomes (circle) in Setx^+/+. In contrast, while ubi-H2A staining was still present on the XY chromosomes in Setx^−/− (circle), it was also distributed on the autosomes. Scale bar, 20 µm. D. Model depicting the consequences of senataxin disruption on spermatogenesis. In wild type spermatocytes, senataxin resolves R-loops structures that may form during pachytene following the resumption of transcriptional activity, thus allowing effective meiotic recombination to proceed and spermatogenesis to be completed. Senataxin also localizes to the XY chromosomes in a Brca1-dependant manner thus enabling the localization of DNA damage response proteins MDC1, ATR and TopBP1 to the entire XY chromatin domain. This ensures that MSCI takes place and germ cells develop and mature properly. On the other hand, in Setx^−/− spermatocytes, R-loops accumulates at pachytene stage interfering with meiotic recombination and the repair of DNA DSB. This leads to the absence of crossing-over, the arrest of meisois at pachytene stage and the elimination of Setx^−/− spermatocytes by apoptosis. The absence of senataxin also results in impaired XY body formation, defective localization of DNA damage response proteins on XY chromatin and failure to undergo MSCI. The combined defects of aborted meiotic recombination and MSCI failure are responsible for the infertility observed in Setx^−/− males.