Embryonic poly(A)-binding protein (EPAB) is required for oocyte maturation and female fertility in mice

Abstract

Gene expression during oocyte maturation and early embryogenesis up to zygotic genome activation requires translational activation of maternally-derived mRNAs. EPAB [embryonic poly(A)-binding protein] is the predominant poly(A)-binding protein during this period in Xenopus, mouse and human. In Xenopus oocytes, ePAB stabilizes maternal mRNAs and promotes their translation. To assess the role of EPAB in mammalian reproduction, we generated Epab-knockout mice. Although Epab(-/-) males and Epab(+/-) of both sexes were fertile, Epab(-/-) female mice were infertile, and could not generate embryos or mature oocytes in vivo or in vitro. Epab(-/-) oocytes failed to achieve translational activation of maternally-stored mRNAs upon stimulation of oocyte maturation, including Ccnb1 (cyclin B1) and Dazl (deleted in azoospermia-like) mRNAs. Microinjection of Epab mRNA into Epab(-/-) germinal vesicle stage oocytes did not rescue maturation, suggesting that EPAB is also required for earlier stages of oogenesis. In addition, late antral follicles in the ovaries of Epab(-/-) mice exhibited impaired cumulus expansion, and a 8-fold decrease in ovulation, associated with a significant down-regulation of mRNAs encoding the EGF (epidermal growth factor)-like growth factors Areg (amphiregulin), Ereg (epiregulin) and Btc (betacellulin), and their downstream regulators, Ptgs2 (prostaglandin synthase 2), Has2 (hyaluronan synthase 2) and Tnfaip6 (tumour necrosis factor α-induced protein 6). The findings from the present study indicate that EPAB is necessary for oogenesis, folliculogenesis and female fertility in mice.

Figure 1. Generation of Epab-deficient mice

(A) Schematic representation of: the genomic organization of mouse Epab (top panel); the targeting construct engineered in pEZ-Flox (middle panel); and the targeted Epab allele (bottom panel). Exons are indicated by numbered and filled boxes. Expected sites of homologous recombination are shown with straight lines. * indicates the 3′-probe used for Southern blot analysis. Arrows show the location of PCR primers. Arrowheads depict LoxP sites. Restriction sites are indicated as C for ClaI, B for BamHI, S for SalI and X for XhoI. Neo, neomycin gene; TK , thymidine kinase gene. (B) Southern blot analysis of WT (+/+ , first three lanes) and Epab^+/− (+/− , last three lanes) ES cells. BamHI digestion and hybridization with the exon 3 probe detected a 9.5 kb band and an 11.7 kb band for the WT and mutant (Mut) alleles respectively. (C) PCR analysis of genomic DNA extracted from ES cells. A 5.6 kb fragment is amplified from the mutant (Mut) allele using the 4F primer located in the Epab gene and the P-1R primer located in the targeting vector. (D) Epab RT–PCR analysis in WT (+/+, Epab^{+ /−} (+/−), and Epab^−/− (−/−) mouse ovaries. PCR with primers 1F (exon 1) and 9R (exon 9) amplified a 1.2 kb and a 1 kb fragments from the WT and the mutant (Mut) alleles respectively. PCR with primers on exons 1–2 only amplified a fragment from the WT allele. Actin RT–PCR was used as an internal control.

Figure 2. Histomorphometric evaluation of Epab^−/− ovaries

Follicle development was assessed in ovaries of unstimulated mature (10–12 weeks old) WT and Epab^−/− mice. (A) Representative low-magnification micrographs of ovaries from 12-week-old WT (+/+) and Epab^−/− (−/−) mice. Scale bars represent 10 μm. (B) Representative high-magnification micrographs of follicles from 12-week-old WT (Epab^+/+) and Epab^−/− mice at different developmental stages. Prim: primordial; Pr: primary; Sec: secondary; E Ant: early antral; Ant: antral follicles. Scale bars represent 10 μm. (C) Follicular count of unstimulated ovaries from 10–12-week-old WT (black bars) and Epab^−/− (grey bars) mice. Follicle counts were conducted using six ovaries of each genotype. Data represent means ± S.E.M. The number of secondary follicles was significantly higher in Epab^−/− mice; **P < 0.01.

Figure 3. Epab-deficient female mice do not generate embryos or mature (MII) oocytes

(A) GV-stage oocytes were obtained from the ovaries of 10–12-weeks-old WT (Epab^+/+), Epab^+/− or Epab^−/− mice 44 h after stimulation with 5 IU of PMSG (n = 6 for each group). There was no difference between Epab^+/+ (black bar), Epab^+/− (dark grey bar) and Epab^−/− (light grey bar) mice in the number of GV-stage oocytes obtained. The results represent means ± S.E.M. (B) Mature (MII) oocytes were collected from the oviducts of superovulated 10–12-week-old WT (Epab^+/+), Epab^+/− or Epab^−/− mice (n = 15 for each group). Epab^−/− mice (light grey bars) had significantly lower MI and MII oocytes compared with Epab^+/+ (black bars) or Epab^+/− (dark grey bars) mice. In addition, the total number of oocytes found in the oviducts of Epab^−/− mice was significantly lower. Results are presented as means ± S.E.M.; ***P < 0.001 for Epab^−/− compared with Epab^+/− or Epab^+/+ mice. (C) Two-cell embryos were collected from the oviducts of superovulated 10–12-week-old WT (Epab^+/+), Epab^+/− or Epab^−/− female mice (n = 10 for each group) mated with 12-week-old fertile WT males. Epab^−/− mice did not produce two-cell embryos, whereas the number of two-cell embryos collected from Epab^+/− mice (dark grey bar) was similar to Epab^+/+ (black bar). Results are presented as means ± S.E.M.; ***P < 0.001. (D) Assessment of GVBD (consistent with metaphase I stage). GV-stage oocytes were collected from PMSG-primed WT and Epab^−/− mice (n = 4 mice for each genotype) and cultured under in vitro maturation conditions. A total of 135 Epab^+/+ and 225 Epab^−/− oocytes were assessed. The results are presented as means ± S.E.M. At 18 h, only 16.5 % of Epab^−/− oocytes completed GVBD compared with 74.5 % of Epab^+/+; ***P < 0.001. (E) Assessment of maturation [consistent with metaphase II (MII) stage] described as GVBD and the appearance of a polar body in WT and Epab^−/− oocytes cultured under in vitro maturation conditions (as described for D). At 18 h, 0 % of Epab^−/− oocytes had completed maturation compared with 41.9 % of Epab^+/+; ***P < 0.001.

Figure 4. Epab is required for meiotic division and chromosome alignment

GV-stage oocytes were collected from PMSG-primed WT and Epab^−/− mice (n = 4 mice for each genotype). Oocytes were analysed at baseline (0 h) (A), or after 9 h (B) or 18 h (C) of culture under in vitro maturation conditions. Column 1, DAPI (blue); Column 2, anti-α-tubulin antibody (green); Column 3, merged images of DAPI and anti-α-tubulin staining. (A) At baseline, both WT (Epab^+/+) and Epab^−/− oocytes have intact nuclear membranes, consistent with GV stage. (B) At 9 h, most WT (Epab^+/+) oocytes underwent GVBD, with chromosomes aligned on the spindle, consistent with MI stage, whereas most Epab^−/− oocytes remained at GV stage. In the small number of Epab^−/− oocytes with GVBD, microtubule-like structures (stained with anti-α-tubulin) could be visualized. (C) At 18 h, those WT (Epab^+/+) oocytes that reached MII had their chromosomes aligned on the spindle within the oocyte and in the polar body. Most Epab^−/− oocytes remained at GV stage, whereas some had disseminated chromosomes, and others showed chromosomes that remained in the centre of the oocytes without microtubule formation.

Figure 5. Epab is required for cytoplasmic polyadenylation of Dazl, Ccnb1 and c-Mos mRNAs

(A) GV-stage oocytes were isolated from WT (Epab^+/+) and Epab^−/− mice and total RNA was isolated from 100 oocytes for each genotype at baseline (0 h) and after 18 h of in vitro maturation. Poly(A)-tail lengths were determined using a PCR-based poly(A) tail assay. In Epab^−/− oocytes, poly(A)-tail lengths of Ccnb1, c-Mos or Dazl mRNAs did not increase upon 18 h of in vitro maturation, whereas a >100 bp increase was observed in WT. ACTB mRNA was polyadenylated prior to oocyte maturation and maintained poly(A)-tail length in both WT and Epab^−/− oocytes. (B and C) GV-stage oocytes were collected from PMSG-primed WT (Epab^+/+) and Epab^−/− mice and cultured for in vitro maturation. Oocytes were collected at baseline (0 h), after 9 h or 18 h of culture under in vitro maturation conditions. CCNB1 and DAZL protein expression were determined with Western blotting and normalized to actin. In Epab^−/− oocytes, CCNB1 and DAZL protein expression did not increase upon 18 h of in vitro maturation compared with a significant increase observed in WT. Results are presented as means ± S.E.M.; ***P < 0.001 for Epab^−/− compared with Epab^+/+ mice.

Figure 6. Microinjection of Epab mRNA into Epab^−/− oocytes does not rescue oocyte maturation

(A) GV-stage oocytes collected from PMSG-primed WT or Epab^−/− mice (n = 4 mice for each genotype) were microinejected with HA-tagged Epab mRNA. Uninjected controls were co-cultured for each group. Following overnight culture in milrinone-containing medium, the next day the oocytes were washed and cultured under in vitro maturation conditions, and evaluated for GVBD at 4 h. Epab mRNA-injected or uninjected Epab^−/− oocytes did not undergo GVBD, compared with 74.3 % and 72.5 % of uninjected and injected Epab^+/+ oocytes, respectively. The results are presented as means ± S.E.M; ***P < 0.001. (B) At 18 h, Epab mRNA-injected and uninjected oocytes were assessed for polar body extrusion (MII stage). Uninjected or injected Epab^−/− oocytes did not demonstrate polar body extrusion, compared with 44.3 % and 40.6 % of uninjected and injected Epab^+/+ oocytes respectively. Results are presented as means ± S.E.M.; ***P < 0.001. (C) Western blot with an anti-HA antibody was performed in uninjected and injected oocytes (n = 10 per sample) of WT and Epab^−/− mice to determine EPAB–HA protein expression. Uninjected oocytes did not express the HA-tagged-protein, whereas a protein of the correct size was detected in injected WT and Epab^−/− oocytes.

Figure 7. Epab is required for normal cumulus expansion

Cumulus expansion was assessed in the ovaries of hyperstimulated 10–12-week-old WT and Epab^−/− mice (n = 4 for each genotype), collected 9 h after the hCG injection. (A) Representative micrographs of antral follicles (upper frame) and COCs (insert and lower frame) from WT (Epab^+/+) and Epab^−/− mice. More compact COCs and a lower number of granulosa cell layers was observed in Epab^−/− mice. Scale bar represents 50 μm. (B) The degree of cumulus expansion was evaluated for COCs in preovulatory follicles of WT (black bars) and Epab^−/− (grey bars) mice as described previously [31]. The results represent the mean percentage of COCs for each score ± S.E.M.; *P < 0.05. (C) The diameter of the antral follicles was measured in the ovaries of WT (black bar) and Epab^−/− (grey bar) mice. The results represent the means ± S.E.M.; ***P < 0.001. (D) The diameter of the oocytes contained within antral follicles was measured in the ovaries of WT (black bar) and Epab^−/− (grey bar) mice. The results represent the means ± S.E.M.; ***P < 0.001.

Figure 8. Epab is required for the expression of genes that regulate cumulus expansion

Ovaries of superovulated 10–12-week-old WT (Epab^+/+) or Epab^−/− mice (n = 5 for each group) were collected 4 h after hCG injection. COCs were obtained by ovarian puncture and cumulus cells were isolated. The expression of Areg, Ereg, Btc, Ptgs2, Has2 and Tnfaip6 was assessed using qRT-PCR. Expression of the target gene was normalized to β-actin levels. The 2^−ΔΔC_t (cycle threshold) method was used to calculate relative expression levels. Results were reported as a fold change in gene expression between WT (black bars) and Epab^−/− mice (grey bars). The results are presented as means ± S.E.M.; *P < 0.05.