The expression of Y-linked Zfy2 in XY mouse oocytes leads to frequent meiosis 2 defects, a high incidence of subsequent early cleavage stage arrest and infertility

Abstract

Outbred XY(Sry-) female mice that lack Sry due to the 11 kb deletion Sry(dl1Rlb) have very limited fertility. However, five lines of outbred XY(d) females with Y chromosome deletions Y(Del(Y)1Ct)-Y(Del(Y)5Ct) that deplete the Rbmy gene cluster and repress Sry transcription were found to be of good fertility. Here we tested our expectation that the difference in fertility between XO, XY(d-1) and XY(Sry-) females would be reflected in different degrees of oocyte depletion, but this was not the case. Transgenic addition of Yp genes to XO females implicated Zfy2 as being responsible for the deleterious Y chromosomal effect on fertility. Zfy2 transcript levels were reduced in ovaries of XY(d-1) compared with XY(Sry-) females in keeping with their differing fertility. In seeking the biological basis of the impaired fertility we found that XY(Sry-), XY(d-1) and XO,Zfy2 females produce equivalent numbers of 2-cell embryos. However, in XY(Sry-) and XO,Zfy2 females the majority of embryos arrested with 2-4 cells and almost no blastocysts were produced; by contrast, XY(d-1) females produced substantially more blastocysts but fewer than XO controls. As previously documented for C57BL/6 inbred XY females, outbred XY(Sry-) and XO,Zfy2 females showed frequent failure of the second meiotic division, although this did not prevent the first cleavage. Oocyte transcriptome analysis revealed major transcriptional changes resulting from the Zfy2 transgene addition. We conclude that Zfy2-induced transcriptional changes in oocytes are sufficient to explain the more severe fertility impairment of XY as compared with XO females.

Keywords: First cleavage; Meiosis 2 defects; Mouse; Preimplantation failure; XY female infertility; Y gene expression in oocytes; Zfy2.

Fig. 1.

The gene content of the mouse Y chromosome short arm and that of the Y^Sry- and Y^d-1 chromosomes. The Y^Sry- chromosome has an 11 kb deletion removing the testis-determining gene Sry. The Y^d-1 chromosome has a deletion (estimated at 1-4 Mb) removing most of the Rbmy cluster and leaving an estimated two copies of Rbmy (Y^d-3 has an estimated three copies of Rbmy). XY^d-1 and XY^d-3 mice develop as females because Sry is transcriptionally inactivated.

Fig. 2.

Reproductive lifetime breeding performance of XO, XY^Sry- and XY^d females. (A) Mean number of litters. (B) Mean number of offspring. The number of females of each genotype is shown. The planned comparisons show that addition of Y^Sry- causes a marked reduction in fertility, and that addition of the Y^d chromosomes has a significantly reduced effect on fertility. ^*P<0.05, ^**P<0.01, ^***P<0.001. Error bars indicate s.e.m.

Fig. 3.

Oocyte numbers per ovary in single X genotypes and XX controls. (A) Total oocytes at 5 dpp. (B) Pool oocytes at 28 dpp. (C) Growing oocytes at 28 dpp. (D) Growing oocytes at 35-56 dpp. The number of females is shown for each point. The data plots serve to emphasise the substantial reduction in oocyte numbers with a single X chromosome and the fact that counts for the 1X and 2X genotypes covary. The latter is due to a substantial maternal or litter effect on oocyte numbers (see supplementary material Table S1A). No differences in oocyte numbers were identified between the 1X genotypes that were sufficient to explain the markedly poorer fertility of XY^Sry- females. P values relate to the X-dosage effect and derive from ANOVA with ‘Cross’ and ‘X dosage’ as factors.

Fig. 4.

Rbmy expression in ovaries and the consequences for fertility. (A) RT-PCR analysis of Yp genes in testis from 8-week-old XY male (positive control) and in ovaries from 8-week-old XY^Sry-, XY^d-1 and XX (negative control) females. Dazl amplification confirms that germ cells are present; β-actin is a loading control. In XY^Sry- ovaries, spermatid-specific transcripts (Zfy2 from Cypt promoter, H2al2y) are not expressed, whereas Zfy1, Uba1y, Kdm5d, Eif2s3y, Uty, Ddx3y, Usp9y, Zfy2 (germ cell) and Rbmy (markedly reduced in XY^d-1) are expressed. (B) Western blot detection of RBMY in 17 dpp testis (lanes 1, 2 and 4 are positive controls, with 50% loading for lane 2; lane 3 is a negative control) and in 7-week-old ovaries (lanes 5-7, Rbmy carriers; lane 8, negative control); actin is a loading control. RBMY is detected in the positive testis controls, although markedly reduced in the XSxr^aO testes (8-fold-depleted Rbmy cluster). No expression could be detected in Rbmy-positive ovaries, including XY^Sry- that has a complete Rbmy complement. (C) Rbmy transgene addition and fertility. The mean number of offspring per female and the number of females of each genotype are shown. Rbmy transgene addition does not alter the fertility of XO or XY^d-1 females. (D) Comparison by qRT-PCR of Rbmy transcript levels (normalised to the germ cell-specific transcript Dazl) in 7-week-old XY^Sry-, XY^d-1 and XO,Rbmy ovaries. XO,Rbmy ovaries have substantially higher Rbmy transcript levels than XY^d-1 ovaries, but they fall far short of the levels in XY^Sry- ovaries. ^*P<0.05, ^***P<0.001; NS, non-significant. Error bars indicate s.e.m.

Fig. 5.

Zfy2 and Zfy1 transgene expression in ovaries and the consequences for fertility. (A) Zfy2 transgene addition and XO fertility. The mean number of offspring per female and the number of females of each genotype are shown. The Zfy2^nf (non-functional) transgene provides a control for disruption of the Hprt locus. Zfy2 transgene addition almost completely sterilises XO females whereas Zfy2^nf has no effect. (B) Comparison by qRT-PCR of Zfy2, Usp9y and Zfy1 transcript levels (normalised to Dazl) in 7-week-old XY^Sry-, XY^d-1 and XO,Zfy2 ovaries, relative to those in 17.5 dpp testes (positive control). The lowest Zfy2 expression is seen in XY^d-1 ovaries. (C) Zfy1 transgene addition and XO fertility. The mean number of offspring per female and the number of females of each genotype are shown. Neither Zfy1^lo (low expressor) nor Zfy1^hi (high expressor) alters XO fertility. (D) Comparison by qRT-PCR of Zfy1 transcripts with and without exon 6 in 7-week-old XX,Zfy1^lo and XX,Zfy1^hi ovaries (normalised to Dazl) relative to those in 15 dpp testis (positive control). Zfy1^hi results in much higher Zfy1 expression than Zfy1^lo, but there was a large excess of the transcript lacking exon 6 in ovaries with either transgene. ^*P<0.05, ^**P<0.01, ^***P<0.001; NS, non-significant. Error bars indicate s.e.m.

Fig. 6.

The impact of Zfy2, Zfy1, Y^Sry- and Y^d-1 on preimplantation development. (A) Percentage of embryos by stage of development after 1 and 4 days of culture (embryos at 2.5 dpc and 5.5 dpc, respectively). There is a striking retarding effect of the Zfy2 transgene in XX and XO mice, and of Y^Sry- in XY^Sry- mice, with many embryos exhibiting 2- to 4-cell arrest. (B) Lifetime offspring production by the various genotypes of females with a single X chromosome is strongly correlated (R²=0.9773) with the proportion of embryos that reach the blastocyst stage in culture. (C) Percentage of embryos by stage of development at 3.75 dpc after development in vivo. Eighteen to 64 embryos were collected per genotype, originating from at least two females per genotype.

Fig. 7.

Zfy2-elicited abnormalities originating before the 2-cell stage. (A) All XO 2-cell embryos have at least one polar body (arrows indicate examples); XO,Zfy2 frequently lack polar bodies (brackets indicate examples) and have blastomeres of unequal size and abnormal shape. (B) Polar body counts from around 25 DAPI-stained 2-cell embryos per genotype. (C) Examples of polar bodies and their nuclear morphology. PB, polar body; l, live; d, dead; d/f, dead/fragmenting. (D) Increased frequency of eccentrically located nuclei, and occasional additional nuclei, in blastomeres of XO,Zfy2 and XY^Sry- females. Around 50 embryos per genotype were assessed. (E) Examples of eccentric and multiple nuclei. (F) Impairment of the second meiotic division assessed after SrCl₂ activation.