Oocyte-specific deletion of Hdac8 in mice reveals stage-specific effects on fertility

in Reproduction
Correspondence should be addressed to J L Gerton or F E Duncan; Email: jeg@stowers.org or f-duncan@northwestern.edu
Restricted access

Eighteen histone deacetylases exist in mammals. The class 1 histone deacetylases HDAC1 and HDAC2 are important for oogenesis and fertility in mice, likely via their effects on histones. The reproductive function of HDAC8, another class 1 enzyme, has not been explored. One key target of HDAC8 is the SMC3 subunit of cohesin, an essential complex mediating sister chromatid cohesion and chromosome segregation. In current models, HDAC8 activity is required for SMC3 recycling, but this function should be dispensable in oocytes since cohesion is established during pre-meiotic S phase and maintained until meiotic resumption during ovulation. Whether other oocyte-specific HDAC8-mediated deacetylation events are required for oogenesis and female fertility is unknown. We used two Cre drivers to remove Hdac8 at specific stages of oocyte development to address whether HDAC8 is required for female fertility in mice. When HDAC8 was knocked out in oocytes in primary and later stage follicles (Zp3-Cre), oogenesis and folliculogenesis appeared normal and mice were fertile. However, females were subfertile when HDAC8 was knocked out prior to pre-meiotic S phase and cohesion establishment (Vasa-Cre). This subfertility was independent of chromosome segregation errors during meiosis but rather appeared to be the result of defects in oogenesis that resulted in smaller fully grown oocytes with a reduced ability to resume meiosis. In all cases, we did not observe compensatory changes in HDAC1, HDAC2 and HDAC3 levels. Thus, although oocyte-specific expression of HDAC8 is not essential for mouse oogenesis after meiotic S phase, it contributes to optimal fertility. We infer that oocyte-specific expression of the deacetylase HDAC8 is required early in oogenesis for optimal fertility.

Downloadable materials

  • Supplemental Figure 1 HDAC8 antibodies are not validated for immunofluorescence. Representative images of prophase I-arrested oocytes from Hdac8(floxed/floxed) and Zp3-Cre:Hdac8(floxed/floxed mice that were immunostained with A) Santa Cruz (sc-11405) and B) Abcam (ab187139) antibodies. Scale bar, 50m.

 

    Society for Reproduction and Fertility

Article Information

Metrics

All Time Past Year Past 30 Days
Abstract Views 62 62 62
Full Text Views 19 19 19
PDF Downloads 11 11 11

Altmetrics

Related Articles

Figures

  • View in gallery

    HDAC8 levels in the oocyte are reduced in conditional knockout animals without a compensatory upregulation of HDAC1-3. (A) Immunoblot analysis of HDACs in oocyte protein extracts from Zp3-Cre:Hdac8 (floxed/floxed) and Hdac8 (floxed/floxed) mice. (B) Densitometry analysis of HDAC protein levels were performed and values were normalized to γ-tubulin levels. Each experiment was performed in duplicate. (C) Immunoblot analysis of HDACs in oocyte protein extracts from Vasa-Cre:Hdac8 (floxed/ex) and Hdac8 (floxed/ex) mice. (D) Densitometry analysis of HDAC protein levels were performed and values were normalized to γ-tubulin levels. Each experiment was performed in duplicate. The arrow highlights the HDAC8 immunoreactive band, whereas the asterisks highlight a non- specific band. Data are expressed as mean ± s.e.m. (*P < 0.05).

  • View in gallery

    Zp3-Cre-mediated loss of oocyte-specific HDAC8 does not affect follicular dynamics. (A) Histologic sections of ovaries stained with periodic acid–Schiff (PAS) from Zp3-Cre:Hdac8 (floxed/floxed) and Hdac8 (floxed/floxed) mice at 18 days of age. (B) Follicle quantification in ovaries from 18-day old mice (n = 3 mice from each genotype). (C) Hematoxylin and Eosin (H&E) stained histologic sections of ovaries from 7-week-old mice of each genotype. (D) Follicle quantification in ovaries from 7-week-old mice (n = 3 mice from each genotype). (E) H&E stained histologic sections of ovaries from 7-month-old mice of each genotype. (F) Follicle quantification in ovaries from 7-month-old mice (n = 3 mice from each genotype). A student t-test was performed to examine statistical significance between genotypes at each stage of follicle development. No statistically significant differences were observed. Data are expressed as mean ± s.e.m. (*P < 0.05). The asterisks highlight follicles with antral cavities, and the arrows highlight corpora lutea. EA, early antral follicle; LA, large antral follicle; PF, primary follicle; PmF, primordial follicle; SF, secondary follicle.

  • View in gallery

    Vasa-Cre-mediated loss of oocyte-specific HDAC8 does not affect follicular dynamics. (A) H&E stained histologic sections of ovaries from Vasa-Cre:Hdac8 (floxed/ex) and Hdac8 (floxed/ex) mice at 18 days of age. (B) Follicle quantification in ovaries from 18-day-old mice (n = 3 mice from each genotype). (C) H&E stained histologic sections of ovaries from 7-month-old mice of each genotype. (D) Follicle quantification in ovaries from 7-month-old mice (n = 3 mice from each genotype). A student t-test was performed to examine statistical significance between genotypes at each stage of follicle development. No statistically significant differences were observed. Data are expressed as mean ± s.e.m. (*P < 0.05). The asterisks highlight follicles with antral cavities, and the arrows highlight corpora lutea. EA, early antral follicle; LA, large antral follicle; PF, primary follicle; PmF, primordial follicle; SF, secondary follicle.

  • View in gallery

    Oocyte-specific loss of Hdac8 results in subfertility in a stage-specific manner. (A) Zp3-cre-mediated deletion of Hdac8 did not affect the number or size of litters in a standard 6 months breeding trial (n = 5 mice for both the Zp3-Cre:Hdac8 (floxed/floxed and Hdac8 (floxed/floxed) genotypes). Each point on the graph represents the litter size from an individual female; females produced different numbers of litters. (B) Zp3-cre-mediated deletion of Hdac8 does not affect the average litter size from all the females in a standard 6 months breeding trial. (C) Mice with the Vasa-cre-mediated deletion of Hdac8 had a tendency to have smaller litter sizes in the first 4 litters in a standard 6 months breeding trial, and this was significant for the second litter (n = 5 mice for the Vasa-Cre:Hdac8 (floxed/ex) genotype and n = 4 mice for the Hdac8 (floxed/ex) genotype). Each point on the graph represents the litter size from an individual female; females produced different numbers of litters. (D) Vasa-cre-mediated deletion of Hdac8 showed reduced average litter size from all the females in a standard 6 months breeding trial. A student t-test was performed to examine statistical significance between genotypes and data are expressed as mean ± s.e.m. (*P < 0.05).

  • View in gallery

    Vasa-Cre-mediated loss of oocyte-specific HDAC8 results in smaller oocytes. Transmitted light images of fully grown oocytes collected from (A) Hdac8 (floxed/floxed) and Zp3-Cre:Hdac8 (floxed/floxed) mice and (B) Hdac8 (floxed/ex) and Vasa-Cre:Hdac8 (floxed/ex) mice. (C) Vasa-cre-mediated deletion of Hdac8 in oocytes affects oocyte diameter, n = 3 mice for Hdac8 (floxed/floxed) (n = 115 oocytes), n = 3 mice for Zp3-Cre:Hdac8 (floxed/floxed) (n = 76 oocytes), n = 3 mice for Hdac8 (floxed/ex) (n = 80 oocytes) and n = 3 mice for Vasa-Cre:Hdac8 (floxed/ex) (n = 63 oocytes). A student t-test was performed to examine statistical significance between genotypes and data are expressed as mean ± s.e.m. (P < 0.001). Scale bar, 50 µm.

  • View in gallery

    Oocyte-specific loss of HDAC8 does not affect spindle morphology or chromosome segregation. (A, B, C and D) Representative images of MI oocytes from Hdac8 (floxed/floxed) (A), Zp3-Cre:Hdac8 (floxed/floxed (B), Hdac8 (floxed/ex) (C) and Vasa-Cre:Hdac8 (floxed/ex) (D) mice showed normal spindle morphologies and chromosome alignment. Oocytes were immunostained with anti-α-tubulin antibody to visualize spindles and counterstained with DAPI to visualize chromosomes, n = 5 mice for Hdac8 (floxed/floxed) (n = 68 oocytes), n = 4 mice for Zp3-Cre:Hdac8 (floxed/floxed) (n = 31 oocytes), n = 4 mice for Hdac8 (floxed/ex) (n = 42 oocytes) and n = 5 mice for Vasa-Cre:Hdac8 (floxed/ex) (n = 30 oocytes). The number of oocytes indicates those cells that contained MI spindles oriented parallel to the image plane which could be accurately evaluated for chromosome alignment. Scale bar, 50 µm. (E) Chromosome spreads from Hdac8 (floxed/ex) (n = 101 spreads from 8 mice) and Vasa-Cre:Hdac8 (floxed/ex) (n = 46 spreads from 6 mice) mice showed normal chromosome number. Spreads were immunostained with anti-CREST antibody to visualize centromeres and counterstained with DAPI to visualize chromosomes. (F) Quantification of percentage of spreads with normal (n = 20 pairs of sister chromatids) and abnormal chromosome numbers.

PubMed

Google Scholar