Sirt1/Nrf2 pathway is involved in oocyte aging by regulating Cyclin B1

Reduced Nrf2 expression is detected in aged mouse oocytes

Transcription factor Nrf2 is a key regulator of the antioxidant defense system, aging-associated diseases and inflammation [14,15]. Therefore, we checked whether Nrf2 expression in oocytes was accordingly changed in response to maternal age. The Nrf2 protein levels in young oocytes (isolated from 6-8 week mice) and old oocytes (isolated from 8-10 month mice) were compared, and a decrease in the Nrf2 level was detected in the old oocytes (P < 0.05; Fig. 1), suggesting that such a decrease may contribute toward the occurrence of observed meiotic defects in old oocytes.

Figure 1. Nrf2 decline in old mouse oocytes. Western blot analysis revealed a reduced Nrf2 expression in mouse oocytes from aged females compared with those from young controls. Actin served as a loading control throughout. Band intensity was calculated using ImageJ software, the ratio of Nrf2/Actin expression was normalized and values are indicated. Data are expressed as the mean ± SD, *P<0.05 vs. control.

Cellular distribution of Nrf2 during oocyte meiosis

To explore the potential involvement of Nrf2 in oocyte maturation, we first examined Nrf2 distribution at different developmental stages (Fig. 2A). Immunostaining clearly showed that Nrf2 was expressed in mouse oocyte. The fluorescence signals reside in the entire immature oocytes, and appear to be accumulated in the germinal vesicles. When the oocytes enter metaphase, Nrf2 localized around the spindle region in the course of spindle formation. During MII, Nrf2 continued to associate with the spindle region. Using a double staining method, we confirmed the co-localization of Nrf2 and α-tubulin (Fig. 2B). Such a dynamic distribution pattern suggested that Nrf2 may have a function in the formation or stability of meiotic spindle, or in the regulation of meiotic progression.

Figure 2. Cellular distribution of Nrf2 during oocyte meiosis. (A) Immunofluorescent staining was employed to show the subcellular localization of Nrf2. Green, Nrf2; Blue, chromatin. A total of 30 oocytes were examined for each group. Scale bar, 20 μm. (B) Immunofluorescent staining for co-localization of α-tubulin with Nrf2 in mouse oocytes. This shows the co-localization of Nrf2 with spindles in mouse oocyte. Red, Nrf2; green, tubulin; blue, chromatin. Scale bar, 20 μm.

Nrf2 depletion adversely affects maturational progression in mouse oocytes.

To investigate the function of Nrf2 in oocyte meiosis, we injected Nrf2 siRNA into fully grown oocytes. Then the oocytes were arrested at the GV stage by milrinone treatment for 20 hours to promote mRNA degradation. This led to a significant reduction in the expression of both Nrf2 mRNA (Fig. 3C) and protein (Fig. 3A and B) levels (P < 0.05), while other gene products were not directly affected. We then analyzed how Nrf2-knockdown (Nrf2-KD) affects oocyte maturation (Fig. 3C).

Figure 3. Effects of Nrf2 knockdown on oocyte maturation. Fully-grown oocytes injected with Nrf2-siRNA were arrested at the GV stage with milrinone for 20 hours, and were then cultured in milrinone-free medium to evaluate the maturational progression. Negative control siRNA was injected as a control. (A) Western blot analysis showing the reduced expression of Nrf2 after siRNA injection. (B) Immunostaining showing the loss of Nrf2 in oocytes with siRNA injection. Scale bar, 20 μm. (C) The relative mRNA level of Nrf2, Nrf1 and Keep1 were determined by RT-qPCR in control- and Nrf2-siRNA-injected oocytes. mRNA levels in control oocytes were set to 1. (D) Quantitative analysis of the GVBD rate and Pb1 extrusion rate in control and Nrf2-siRNA oocytes. Data are expressed as the mean ± SD, *P<0.05 vs. control.

After culturing for 3 hours, the germinal vesicle breakdown (GVBD) rate decreased in the Nrf2-knockdown group compared with that in the control group (P < 0.05; Fig. 3D). Moreover, the ratio of first polar body (Pb1) extrusion was significantly reduced (P < 0.05) in Nrf2-KD oocytes after 14-hour in vitro maturation (Fig. 3D). GVBD means oocytes entry M phase, and the extrusion of Pb1 marks the oocyte maturation. These results, similar as our previous study [16], indicate that Nrf2 participates proper meiotic divisions.

Nrf2 is essential for spindle organization and chromosome alignment in oocytes

The above data suggested a possibility that Nrf2 depletion may affect meiotic apparatus in oocytes. To investigate the regulatory mechanism of Nrf2 during meiosis, Nrf2-KD and control oocytes were immunolabeled with anti-tubulin antibody to visualize the spindle and were co-stained with Hoechst 33342 to bind chromosomes. As shown in Fig 4A, most control oocytes at the metaphase stage showed a typical barrel-shaped spindle and well-organized chromosomes at the equator plate. By contrast, oocytes depleted of Nrf2 displayed a high percentage of spindle/chromosome defects, including loose spindles and chromosomal alignment, compared with controls (P < 0.05). These findings imply that the correct assembly of meiotic apparatus in mouse oocytes depends on Nrf2 expression.

Figure 4. Effects of Nrf2 knockdown on spindle structure, chromosome alignment, Cyclin B1 and CDK1 levels of mouse oocytes. (A) Control and Nrf2-KD oocytes were collected and stained for visualizing spindle (green) and chromosomes (blue). Scale bar, 20 μm. (B) Quantification of control and Nrf2-KD oocytes with spindle defects or chromosome misalignment. (C) The relative mRNA level of Cyclin B1 and CDK1 were determined by RT-qPCR in control and Nrf2-siRNA injected oocytes. mRNA levels in control oocytes were set to 1. (D) Western blot analysis showing the reduced expression of Cyclin B1 after Nrf2 siRNA injection. (E) Decreased CDK1 protein levels in Nrf2-KD oocytes. Data are expressed as the mean ± SD, ^*P<0.05 vs. control.

Nrf2 depletion interferes with Cyclin B1/CDK1 expression in mouse oocytes

The effects of Nrf2-KD on meiotic progression prompted us to consider the possible target in mouse oocytes. Therefore, we further analyzed the expression of Cyclin B1 and Cyclin-dependent kinase 1 (CDK1) by qPCR and western blot analysis. The qPCR results shown in Fig. 4C indicated declined Cyclin B1 and CDK1 mRNA levels in Nrf2-KD oocytes (P < 0.05). While there are no significant difference in the expression of Wee1, CDC25A and MYT1, which have been reported can regulate the expression of cyclinB1/CDK1 [17-19], between control and Nrf2-KD oocytes (Fig. S3B). Furthermore, immunoblot analysis showed that the expression of Cyclin B1 and CDK1 proteins in the Nrf2-KD group was significantly reduced (P < 0.05) compared with those in the control group (Fig. 4D-E). The relative protein intensity analysis also confirmed this (loading control: β-actin). Taken together, these results demonstrated that Nrf2 promotes spindle formation and meiotic division in oocytes through regulating Cyclin B1/CDK1.

Sirt1 depletion alters Nrf2 expression in oocytes

To identify a signaling pathway that might account for the maintenance of spindle assembly and meiotic division by Nrf2 in oocytes, we assessed the effects of Sirt1 depletion on Nrf2 levels. For this purpose, we first examined the localization/levels of endogenous Sirt1 by immunofluorescence. Double staining was also performed in GV oocytes and it was confirmed that Sirt1 co-localized with Nrf2 (Fig. 5A).

Figure 5. Sirt1 depletion reduces Nrf2 expression in oocytes. (A) Double staining of GV oocytes with Nrf2 antibody (red) and Sirt1 antibody (green), and counterstaining of chromosome with Hoechst 33342 (blue), confirming the Sirt1 co-localization with Nrf2. (B) Control and Sirt1-KD oocytes were double stained with anti-Sirt1 antibody (green) and Nrf2 antibody (red), and counterstained for chromosomes (blue). (C) The relative mRNA levels of Sirt1, Nrf2, Cyclin B1 and CDK1 were determined by RT-qPCR in control- and Sirt1-siRNA-injected oocytes. mRNA levels in control oocytes were set to 1. (D) Reduced Sirt1 levels after Sirt1-siRNA injection were confirmed by western blot analysis. Actin served as a loading control. Band intensity was calculated using ImageJ. Bars represent the mean ± SD. ^*P<0.05 vs. control. (E-G) Western blot analysis showed the reduced Nrf2, Cyclin B1 and CDK1 protein levels in oocytes following Sirt1 knockdown, with actin as a loading control. Bars represent the mean ± SD. ^*P<0.05 vs. controls.

Then we injected Sirt1 siRNA into fully grown oocytes, which were arrested at the GV stage by milrinone treatment for 20 hours to promote mRNA degradation. As shown in Fig. 6C, the abundance of Sirt1 mRNA was specifically (p<0.05) decreased. A significant reduction of Sirt1 protein level in oocytes was confirmed by immunofluorescence and western blot analysis (p<0.05; Fig. 5B, D).

Figure 6. Maternal age-associated oocyte spindle and chromosome abnormalities suppressed by Nrf2 overexpression. (A) PBS (control group) or Nrf2 plasmid (overexpression group) was microinjected into old GV oocytes, which were arrested for 20 h with milrinone to allow synthesis of new Nrf2 protein. Results indicated that Nrf2 protein was efficiently overexpressed. (B) Representative examples of meiotic spindle and chromosomes at MII stage in young oocytes, old oocytes and old oocytes injected with PBS or Nrf2 plasmid. Arrowheads indicate misaligned chromosomes. (C) Incidence of spindle/chromosome defects in indicated oocytes. Data are expressed as the mean ± SD percentage from 3 independent experiments in which 100 oocytes were analyzed, *P<0.05 vs. control.

Notably, following Sirt1-knockdown, Nrf2 was significantly decreased, based on immunostaining and western blotting (Fig. 5B and D). These data suggested that the expression of Nrf2 during oocyte meiosis is regulated by Sirt1. Due to the limited number of oocytes, we examined the relationship between sirt1 and Nrf2 in ovarian tissues as an alternative. CO-IP was performed to confirm the interaction relationship between Sirt1 and Nrf2 in mouse ovaries (Fig. S1). Additionally, compared with controls, immunoblot and quantitative analysis showed that Sirt1-knockdown in oocytes had marked effects on Cyclin B1 and CDK1 expression (P < 0.05; Fig. 5C, E and F). This suggested that Sirt1 may act as the upstream molecule for Nrf2-Cyclin B1/CDK1 in mouse oocytes.

Nrf2 overexpression ameliorates maternal age-associated oocyte meiotic defects

It has been well documented that female fertility decreases with advanced maternal age due to chromosomal and spindle abnormality in oocytes. We performed overexpression experiments to test whether enhancing Nrf2 expression in old oocytes could rescue their meiotic phenotypes. Nrf2 CRISPR Activation Plasmid was injected into old GV oocytes, which were arrested for 20 h with milrinone to allow synthesis of new Nrf2 protein. Immunoblotting with an anti-Nrf2 antibody confirmed that Nrf2 protein was efficiently overexpressed (P < 0.05; Fig. 6A). Confocal analysis revealed that only 9.4±5.4% of ovulated MII oocytes obtained from young mice exhibited abnormal spindle formation and chromosome alignment; however, 52.3±6.8% of MII oocytes retrieved from aged mice showed spindle defects or misaligned chromosomes. Notably, these abnormalities were only detected in 30.2±4.0% of old oocytes with Nrf2 overexpression, which is significantly decreased as compared with old oocytes injected with PBS (51.6±3.5%, P < 0.05; Fig. 6B and C). Taken together, these results suggested that Nrf2 overexpression reduces the occurrence of maternal age-associated meiotic defects.

Nrf2 is related to age in female granulosa cells

To investigate the consistency of expression trends of Nrf2 in oocytes and granulosa cells, we collected granulosa cells from old and young mice. The result showed the Nrf2 level in old mouse granular cells decreased (Fig. 7A).

Figure 7. The correlation between Nrf2 and age in granular cells. (A) The relative mRNA levels of Nrf2 were determined by RT-qPCR in control and old granular cells. mRNA levels in control granular cells were set to 1. Data are expressed as the mean ± SD. (B) The relative mRNA level of Nrf2 was determined by RT-qPCR in young (22-35 years) and old (36-49 years) ovarian granular cells. Data are expressed as the mean ± SD, *P<0.05 vs. young group. (C) Analysis of correlation between Nrf2 expression and age in human ovarian granular cells. R=-0.5972.

Then, we measured the mRNA level of Nrf2 in ovarian granular cells (from donors aged 22-49 years) by quantitative analysis. A significant decrease in Nrf2 with age was detected, as observed in mouse granular cells (Fig. 7B). The results showed that the expression of Nrf2 was correlated with age in females (r=-0.5972, P < 0.05), and Nrf2 mRNA levels in the young group were significantly higher than those in the old group (P < 0.05; Fig. 7C). The above results suggested that the decreased expression of Nrf2 may be related to the decline of reproductive capacity of older women.

Antibodies

Rabbit polyclonal anti-Nrf2 (cat. no., ab137550), mouse monoclonal anti-Sirt1 (cat. no., ab110304), rabbit monoclonal anti-Cyclin B1 (cat. no., ab181593) and rabbit monoclonal anti-CDK1 (cat. no., ab32384) were purchased from Abcam (Cambridge, MA, USA); mouse monoclonal anti-α-tubulin-FITC antibody (cat. no., 76074) was purchased from Sigma-Aldrich; Merck KGaA; FITC-conjugated goat anti-rabbit IgG was purchased from Thermo Fisher Scientific, Inc. (Waltham, MA, USA).

Granular cells collection

This human study was approved by the Ethics Committee of Nanjing Jinling Hospital, and was performed in accordance with National and International guidelines. A total of 62 Chinese females were recruited (aged 22-49 years). Females with known karyotype abnormalities, previous chemotherapy or radiotherapy, ovarian surgery or autoimmune diseases were excluded. Granulosa cells were derived from the part of the ovaries that had been discarded after routine ovulation. Before initiating the study, written informed consent was obtained from the all participants.

6-8-week-old female ICR mice were injected with human chorionic gonadotrophin (hCG), approximately 46-48 h after injection of 5 IU Pregnant Mares Serum Gonadotropin (PMSG). After 12 h, cumulus-oocyte complex (COC) were harvested from oviduct ampullar, oocytes were removed by hyaluronidase. Granulosa cells were collected for the following assays.

Oocyte collection and culture

Approximately 46-48 h after injection of 5 IU PMSG, fully-grown immature oocytes were harvested from the ovaries of 6-8-week-old female ICR mice. Enclosed cumulus cells were removed by repeatedly pipetting, and then oocytes were cultured in M2 medium under mineral oil at 37˚C in a 5% CO₂ incubator. At appropriate time points, oocytes were selected for the following assays.

siRNA-knockdown and overexpression

Microinjection of siRNA or plasmid was used to knock down Nrf2/Sirt1 or to overexpress Nrf2 with a Narishige microinjector. The Nrf2-siRNA and Sirt1-siRNA sequences are shown in Table S1.

For knockdown experiments, siRNA was diluted with water to give a stock concentration of 30 mM, and then 2.5 picoliter solution was injected into the cytoplasm of fully-grown immature oocytes. A siRNA negative control was injected as a control. For overexpression experiments, Nrf2 CRISPR Activation Plasmid was diluted with water to give a stock concentration of 30 ng/μL, and then 2.5 picoliter solution was injected into oocytes. PBS was injected as a control.

After injections, oocytes were arrested at the GV stage with 2.5 µM milrinone for 20 hours, and were then cultured in milrinone-free M2 medium for maturation.

Quantitative real-time PCR

Total RNA was isolated from granulosa cells or oocytes using an RNA Isolation Kit (cat. no., KIT0204; Invitrogen; Thermo Fisher Scientific, Inc.), and cDNA was quantified by RT-qPCR using a Roche Light Cycler 96 Real-time PCR system (Roche Diagnostics, Basel, Switzerland). The fold-change in gene expression was calculated using the 2^-∆∆Ct method with the house keeping gene, GAPDH, as the internal control. Primer sequences are listed in Table S2.

Western blotting

Approximately 100 oocytes were lysed in Laemmli sample buffer containing protease inhibitor, and boiled for 5 min before being subjected to 10% SDS-PAGE. A PVDF membrane was used to transfer the separated proteins, and was then blocked in TBST (TBS containing 0.1% Tween 20) and 5% skimmed milk for 1 hour. Then the PVDF membrane was separated and incubated overnight at 4˚C with the following primary antibodies: Rabbit anti-Nrf2 antibody (1:1000), anti-Sirt1 antibody (1:750), anti-Cyclin B1 antibody (1:1000), anti-CDK1 antibody (1:1000) and anti-actin antibody (1:2000). After being washed 3 times in TBST, the membranes were incubated with HRP-conjugated secondary antibodies for 1 hour at room temperature, prior to being processed using an ECL Plus Western Blotting Detection System. ImageJ software (National Institutes of Health, Bethesda, MD, USA) was used to quantify fluorescence intensity as previously described [39,40].

Immunofluorescence and confocal microscopy

For staining of Nrf2 and Sirt1, oocytes were fixed with 4% paraformaldehyde for 30 minutes and then permeabilized with 0.5% Triton X-100 for 20 minutes. After 1 hour of blocking in 1% BSA-supplemented PBS, samples were incubated overnight at 4˚C with the following primary antibodies: Anti-Nrf2 antibody, anti-Sirt1 antibody and FITC-conjugated anti-tubulin antibody. For tubulin immunofluorescence staining, GV oocytes were extracted for 10 min with 0.1% Triton X-100 before fixation, and then stained with the anti-tubulin antibody. After three washes in PBS, oocytes were labeled with Alexa Fluor 488 goat-anti mouse IgG at room temperature for 1 hour. Hoechst 33342 (blue) was used for chromosome staining. Oocyte samples were mounted on anti-fade medium (Vectashield, Burlingame, CA, USA), and were then examined under a Laser Scanning Confocal Microscope (LSM 710, Zeiss, Germany) equipped with the x40 objectives.

Statistical analysis

Statistical analysis was performed using SPSS 19.0 software (IBM Corp., Armonk, NY, USA). Correlations were assessed using Pearson’s method. Data are presented as the mean ± standard deviations (SD), unless otherwise indicated. The mean and SD of the data were calculated and statistically analyzed by Student’s t-test and analysis of variance (ANOVA) when appropriate (P<0.05).