Conditional inactivation of the DNA damage response gene Hus1 in mouse testis reveals separable roles for components of the RAD9-RAD1-HUS1 complex in meiotic chromosome maintenance

Abstract

The RAD9-RAD1-HUS1 (9-1-1) complex is a heterotrimeric PCNA-like clamp that responds to DNA damage in somatic cells by promoting DNA repair as well as ATR-dependent DNA damage checkpoint signaling. In yeast, worms, and flies, the 9-1-1 complex is also required for meiotic checkpoint function and efficient completion of meiotic recombination; however, since Rad9, Rad1, and Hus1 are essential genes in mammals, little is known about their functions in mammalian germ cells. In this study, we assessed the meiotic functions of 9-1-1 by analyzing mice with germ cell-specific deletion of Hus1 as well as by examining the localization of RAD9 and RAD1 on meiotic chromosomes during prophase I. Hus1 loss in testicular germ cells resulted in meiotic defects, germ cell depletion, and severely compromised fertility. Hus1-deficient primary spermatocytes exhibited persistent autosomal γH2AX and RAD51 staining indicative of unrepaired meiotic DSBs, synapsis defects, an extended XY body domain often encompassing partial or whole autosomes, and an increase in structural chromosome abnormalities such as end-to-end X chromosome-autosome fusions and ruptures in the synaptonemal complex. Most of these aberrations persisted in diplotene-stage spermatocytes. Consistent with a role for the 9-1-1 complex in meiotic DSB repair, RAD9 localized to punctate, RAD51-containing foci on meiotic chromosomes in a Hus1-dependent manner. Interestingly, RAD1 had a broader distribution that only partially overlapped with RAD9, and localization of both RAD1 and the ATR activator TOPBP1 to the XY body and to unsynapsed autosomes was intact in Hus1 conditional knockouts. We conclude that mammalian HUS1 acts as a component of the canonical 9-1-1 complex during meiotic prophase I to promote DSB repair and further propose that RAD1 and TOPBP1 respond to unsynapsed chromatin through an alternative mechanism that does not require RAD9 or HUS1.

Figure 1. Conditional Hus1 inactivation in the mouse testis results in reduced testis size and sperm count.

A. Western blot analysis of HUS1 protein in adult (12-week old) control and Hus1 CKO testes. B, E. Photographs of testes from 12-week old Hus1 CKO and control mice from Stra8-Cre and Spo11-Cre crosses, respectively. In these representative images, the Stra8-Cre control is Cre-positive Hus1^+/flox, whereas the Spo11-Cre control is Cre-negative Hus1^flox/Δ1. C, F. Testis weights of 17-day and 12-week old adult Hus1 CKO and control mice from both Stra8-Cre and Spo11-Cre crosses, shown as the mean testis weight relative to body weight ± SEM. D, G. Epididymal sperm counts from Stra8-Cre and Spo11-Cre Hus1 CKO 12-week old males, shown as the mean ± SEM. Statistically significant differences between Hus1 CKO and the respective control as determined by Student's t-test are indicated (* at p<0.001; ^‡ at p<0.01).

Figure 2. Hus1 loss results in germ cell depletion.

A. 100× images of H&E-stained histological sections from 12-week old control (left; Cre+ Hus1^+/flox) and Stra8-Cre Hus1 CKO (right) testes. B. GCNA1 staining of germ cells in control and Stra8-Cre Hus1 CKO testes indicating germ cell loss. C. TUNEL staining of control and Stra8-Cre Hus1 CKO adult testes indicating germ cell apoptosis. D. Quantification of TUNEL staining shown in C, shown as the mean ± SEM. Asterisk indicates statistically significant difference between Hus1 CKO and control animals (p<0.05, Student's t-test). E. Higher magnification (400×) images of TUNEL staining in Stra8-Cre Hus1 CKO adult testes. Arrows highlight TUNEL-positive cells that appeared to be pre-meiotic (left), mid-prophase I (middle), or at meiotic metaphase (right).

Figure 3. Hus1 inactivation results in abnormal accumulation of γH2AX on autosomes, an extended sex body domain, inclusion of autosomes within the sex body, and X-autosome fusions.

A–F. Immunofluorescence staining for γH2AX and SYCP3. γH2AX staining persisted in a perpendicular pattern on synapsed autosomes in pachytene (B) and diplotene stage (C), and in clouds surrounding unsynapsed chromosomal regions (D). The sex body domain marked by γH2AX was extended in Stra8-Cre Hus1 CKOs (B,F), and often contained whole or partial autosomes (B,F). Some Stra8-Cre Hus1 CKO nuclei exhibited apparent X chromosome-autosome end-to-end fusions (E). G,H. Western blot analysis of CHK1 (G) and CHK2 (H) in total testis extracts from mice of the indicated genotypes. In the CHK2 immunoblot, pre-B cells that were untreated (U) or treated with 5Gy ionizing radiation (IR) are included as controls. β-actin is shown as a loading control. In Hus1 CKO samples, total CHK1 levels were increased on average 2.9-fold over controls, and phosphorylated CHK1 (S345) levels were increased 4.1-fold. CHK2 phosphorylation was not detected in the absence of Hus1.

Figure 4. RAD51 foci persist in late pachytene and diplotene spermatocytes in the absence of Hus1.

A–E. Immunofluorescence staining for RAD51/DMC1 from zygotene through diplotene stage of meiosis in control (Cre-negative Hus1^{flox/ Δ1}) and Stra8-Cre Hus1 CKO males. Arrows indicate persistent foci, and asterisks indicate symmetrical foci located on either side of chromosomal cores. F–I. Quantification of RAD51 foci at the indicated stages. ** indicates p<0.01; * indicates p<0.05.

Figure 5. Conditional Hus1 knockout meiotic chromosomes display synapsis defects and ruptures in the synaptonemal complex.

Meiotic chromosome spreads from control (Cre-negative Hus1^{flox/ Δ1}) and Stra8-Cre Hus1 CKO mice were stained for SYCP1 and SYCP3, as well as with DAPI. A. Normal synapsis of pachytene chromosomes in control males, as indicated by SYCP1 and SYCP3 immunofluorescence. B. Chromosomal asynapsis in pachytene-stage Stra8-Cre Hus1 CKO nuclei, with unsynapsed chromosomal regions devoid of SYCP1. C. Diplotene Stra8-Cre Hus1 CKO chromosomes with ruptures in the SC lateral elements, as indicated by arrows.

Figure 6. TOPBP1 localization to and RNA Pol II exclusion from the sex body domain remain unperturbed following Hus1 inactivation.

A–E. Immunofluorescence staining for TOPBP1 (A–C) and RNA Polymerase II (D–E) in control (Cre-negative Hus1^{flox/ Δ1}) and Stra8-Cre Hus1 CKO pachytene spermatocytes. The asterisk in B indicates an autosome partially included within the sex body domain and coated with TOPBP1. Arrows in C indicate abnormal TOPBP1 staining in a perpendicular pattern on autosomes in pachytene Hus1 CKOs.

Figure 7. RAD9 localizes to meiotic chromosomes during early prophase I and colocalizes with a subset of RAD51 foci.

A–C. Meiotic chromosome spreads from control animals were stained for RAD9 and SYCP3. In wild-type adult males, RAD9 localized along the synaptonemal complex of synapsed and unsynapsed chromosomes during zygotene (A) and pachytene (B, C). RAD9 localized to autosomes and the sex chromosomes in early pachytene (B), was confined primarily to the X chromosome by mid-late pachytene (C), and was absent by diplotene. D. Meiotic chromosome spreads from control animals were stained for RAD9 and RAD51. RAD9 colocalized with a subset of RAD51 foci, particularly along the X chromosome in pachytene-like nuclei.

Figure 8. RAD9 and RAD1 localize in overlapping yet distinct patterns on meiotic chromosomes.

A. Meiotic chromosome spreads from control animals were stained for RAD9 and RAD1. RAD9 colocalized with a subset of RAD1 foci in control spermatocytes. A region containing one X-Y pair is shown at higher magnification. B–E. Meiotic chromosome spreads from Slx4^mut/mut (B, C) or Msh4^−/− (D, E) mice were stained for SYCP3 and RAD1 (B, D) or RAD9 (C, E). RAD1 localized more continuously along sex chromosome cores and autosomes and to asynaptic sites in Slx4 and Msh4 mutants, whereas RAD9 formed fewer, more punctate foci along asynapsed chromosome cores. Chromosome spreads in panels B–E were additionally stained using human CREST serum, which marks centromeric regions. CREST signal is detected in the red channel (middle column) and in the merged images.

Figure 9. RAD1 and RAD9 localization to meiotic chromosomes is differentially affected by Hus1 loss.

Meiotic chromosome spreads from Stra8-Cre Hus1 CKO mice were stained for SYCP3 and RAD1 (A, C, D) or RAD9 (B, E). A,C,D. RAD1 continued to localize to sex chromosomes and aberrant meiotic chromosomes in the absence of Hus1. Most notably, RAD1 localization to asynapsed autosomes and chromosomes with the sex body domain persisted in cells from Hus1 CKO mice. By contrast, RAD9 localization to both normal and aberrant chromosome structures was abolished following Hus1 loss (B,E).