Animal models

This study was performed under a protocol approved by the Animal Care Committee of the University of Ottawa and conducted in accordance with the guidelines of the Canadian Council on Animal Care. Mice were housed under controlled environmental conditions with 12-h alternating light/dark cycles, with free access to water and food. To generate Brg1 cKO in GCs and oocytes, we obtained B6;129S2-Smarca4^tm1.2Pcn/Mmnc (hereafter, Brg1^fl/fl) mice from Dr David Lohnes, with the permission of Dr Pierre Chambon [16]. We also obtained double KO (dKO) with the following genotype Brg1^f/f/Brm^+/– from Dr Courtney T. Griffin, with the permission of Dr Moshe Yaniv [17]. B6;129S7-Amhr2^tm3(cre)Bhr/Mmnc (identification number 14245-UNC; hereafter, Amhr2-Cre) was obtained from the Mutant Mouse Regional Resource Center, with the permission of Dr Richard Behringer [18]. Tg(Cyp19a1-cre)1Jri (hereafter, Cyp19-Cre) was obtained from Dr Derek Boerboom (Université de Montreal, Montreal, Canada) [19], and Tg(Gdf9-icre)5092Coo/J (hereafter, Gdf9-Cre) from Dr Teresa Woodruff (Northwestern University), with the permission of Dr Austin Cooney [20]. Tissue-specific mutant mice with depletion of Smarca4/Brg1 were generated by crossing Brg1^fl/fl mice with transgenic mice expressing tissue-specific Cre recombinase. The resulting male mice Brg1^fl/+;Cre^+/wt were then crossed with female mice to obtain experimental Brg1^fl/fl;Cre^+/wt female mice (or Brg1^fl/fl;Gdf9-icre+, in the Gdf9-Cre mutant line), whereas Brg1^fl/fl;Cre^wt/wt littermate female mice were used as controls. Data originating from control mice are combined between experimental mouse lines as they had the same litter sizes and no difference in phenotypes. Wild-type (WT) male C57BL/6J mice were obtained from The Jackson Laboratory. Genotyping assessments were determined by PCR amplification of mouse ear snip DNA samples. Primers and cycle conditions for each genotyping assessments used in this study are summarized in  Supplementary Table 1 (supplementaltable1.pdf).

Fertility trials and ovary weights

To assess female fertility, mating experiments were conducted using 8-week-old female mice with the following genotypes: Brg1^fl/fl, Brg1^fl/fl;Amhr2-Cre, Brg1^fl/fl;Cyp19-Cre, and Brg1^fl/fl;Gdf9-Cre, which were mated with 8-week-old WT male mice. Each mating pair was monitored daily and the date and number of pups were recorded at birth for 6 months. Adult males were removed from the cage after 6 months, and the experiment was terminated 22 days after male removal to record the final litter. Ovary weights were obtained from adult cycling mice, aged 6–8 months, and from 21 to 26-day-old mice 44–48 h after intraperitoneal (i.p.) injection with eCG (5 IU, Folligon: Intervert, Whitby, ON, Canada).

Follicle counting, immunohistochemistry, and immunofluorescence

To assess ovarian histopathology, intact ovaries were collected from Brg1^fl/fl, Brg1^fl/fl;Amhr2-Cre, and Brg1^fl/fl;Gdf9-Cre groups at 2 and 8 months of age. These time points allowed us to determine if fertility is impacted at the start and near the end of their reproductive capacity. Ovaries were fixed in 4% (w/v) paraformaldehyde for 24 h at room temperature and transferred to 70% ethanol to be paraffin embedded. Serial sections were prepared at a thickness of 5 µm, and every fifth section was stained with hematoxylin and eosin (H&E). For follicle assessment on H&E-stained sections, follicles with a visible oocyte nucleus were counted, classified according to Pedersen's system, and scored as healthy or atretic, as previously described in Pedersen and Peters [21]: oocyte surrounded by a layer of flat GCs = primordial, Pedersen Class 3 = primary, Pedersen Classes 4–5 = secondary, and Pedersen Classes 6–8 = antral. Atresia was determined for each follicle based on histological criteria using a weighted scoring system as previously described by Boyer et al. [22]. Briefly, the primary criteria of atresia scoring were: (1) presence and degree of pyknosis (condensed GCs or cells with condensed nuclei), (2) presence and degree of loss of GC attachment to oocytes or loss of cumulus cells, (3) presence of polymorphonuclear neutrophils or lymphocytes, and (4) presence of pyknotic oocytes (primary follicles only). Secondary criteria were GC vacuolation, sparse or missing GCs, and integrity of the basement membrane.

Immunohistochemical analysis of histological sections was performed using an antibody for a marker of cell proliferation, Ki67 (ab16667; Abcam, Cambridge, UK). Sections were incubated with primary antibody at a 1:200 dilution overnight at 4°C. Subsequently, tissue sections were incubated in anti-rabbit horseradish peroxidase-labeled polymer (Dako) for 1 h at room temperature and developed using diaminobenzidine. Sections were lightly counterstained with hematoxylin before mounting with permount (Fisher Scientific). Images were acquired using ScanScope CS2 (Leica Biosystems, Concord, Canada) or using a Zeiss AxioScan Z1 (Zeiss, Oberkochen, Germany). For IHC semi-quantification, images were analyzed using the open-source Orbit Image Analysis software [23]. Briefly, the software uses a machine learning object classification that is built by manually annotating objects to create a classification model. That model was then used for all images to calculate the ratio of positive versus negative KI67-stained cells with the background removed resulting in a ratio relative to WT mice.

Immunofluorescence analysis was performed on fresh frozen ovaries with reproductive tracts from WT female mice at 2 months of age and from control and mutant mice at 8 weeks of age. Ovarian tissues embedded in OCT (Thermofisher, Waltham, MA, USA) were sectioned at a thickness of 10 µm and dried for 30 min on slides at room temperature. Sections were fixed in 4% paraformaldehyde followed by incubations in 0.2% Triton X-100 diluted in PBS for 15 min at room temperature. Slides were rinsed three times PBS and incubated in blocking solution [10% goat serum (G9023, Sigma-Aldrich, St. Louis, MS, USA) diluted in PBS] for 60 min at room temperature followed by an overnight incubation at 4°C in primary antibodies [rabbit polyclonal anti-Brg1 (NBP2–41270, Novus Biologicals, Littleton, CO, USA) at 5 µg/mL diluted in blocking buffer]. Slides were rinsed three times in PBS and incubated in 4 µg/mL dilution of secondary antibodies in blocking buffer [goat anti-rabbit IgG AlexaFluor 488 (A-11006, Invitrogen, Carlsbad, CA, USA)] for 1 h at room temperature and wash three times with PBS. The slides were mounted with ProLongTM Diamond Antifade Mountant with DAPI (P36962, Thermofisher, Waltham, MA, USA) and coverslipped. We used the Zeiss Axioskop 2 MOT (Oberkochen, Germany) fluorescence imaging microscope at ×40 magnification. Intensity thresholds were set according to the highest intensity image, and tissues from a minimum of three mice were analyzed (n= 3).

Oocyte chromatin configuration

Germinal vesicle (GV) oocytes were obtained from 8 to 9 weeks old female Brg1^fl/fl and Brg1^fl/fl;Amhr2-Cre mice, 44–46 h post-eCG (5 IU i.p.) by placing the ovaries in PBS and using a 26-gage needle to puncture the ovaries. Only oocytes fully enclosed by cumulus cells were collected and washed in PBS with gentle agitation until denuded. Oocytes were immediately incubated in 4% paraformaldehyde for 20 min at 37°C, washed three times in PBS, and stained with Hoechst 33342 diluted at 1/300 in PBS for 15 min. Stained oocytes were mounted on slides using mounting media (Immu-mount, Thermofisher) and observed using the Zeiss Axioskop 2 MOT (Oberkochen, Germany) fluorescence imaging microscope at ×40 magnification. Assessment of GV chromatin configuration with non-surrounded nucleus (NSN) or surrounded nucleus (SN) was performed and reported as proportions of SN and NSN oocytes from a minimum of three mice (n= 3).

RNA extraction and qRT-PCR

GCs from Brg1^fl/fl and Brg1^fl/fl;Amhr2-Cre genotypes were obtained from immature ovaries 48 h post-eCG. GCs were isolated by placing the ovaries in PBS and using a 26-gage needle to puncture the ovaries to extract GCs. GCs were filtered by a 40 µm cell strainer to separate GCs and oocytes. GCs were collected and centrifuged at 2000 RPM and frozen at –80°C until RNA extraction. Oocytes from the Brg1^fl/fl and Brg1^fl/fl;Gdf9-Cre genotypes were obtained by collecting the cumulus oocyte complexes (COCs) after ovarian puncture. Hyaluronidase was used to remove the cumulus cells immediately after collection, and denuded oocytes were collected and frozen at –80°C until RNA extraction. Total RNA from the GCs and oocytes was extracted using the RNeasy Mini kit (#74106, Qiagen) according to the manufacturer's protocol. Reverse transcription was performed with 200 ng GC RNA and 30 ng oocyte RNA using an iScript cDNA Synthesis Kit (Bio-Rad, Mississauga, ON, Canada). Real-time PCR was performed with Fast SYBR Green Master Mix (Invitrogen, Carlsbad, CA, USA). Each PCR reaction consisted of 5.5 µl of Fast SYBR Green Master Mix, 2.5 µl of RNAse free water, 1 µl of cDNA sample, and 0.5 µl (10 pmol) of Smarca4specific primers ( Supplementary Table 1 (supplementaltable1.pdf)). The qPCR was run using the 7500 Fast system assays (Applied Biosystems). To quantify relative gene expression, the cycle threshold (Ct) of target gene amplification was normalized to the expression level of a housekeeping gene (Actin) using the delta–delta Ct method.

Assessment of ovulation

To determine the natural ovulation rates, 8–10-week-old females of experimental and control genotypes were housed with WT males of proven fertility and monitored daily for the presence of a vaginal plug. For assessment of the number of COCs after superovulation, immature mice were injected with eCG for 48 h prior to hCG injection (5 IU i.p., Chorulon; Intervet) and COCs were collected after 14 h. For both natural matings and superovulation, the animals were euthanized and the oviducts were removed and placed in sterile saline under a dissection microscope. COCs were released by tearing open the ampullae of the oviducts with forceps and were counted.

Steroid hormone measurement

Blood samples were collected by cardiac puncture before euthanasia. Estradiol (E2) and progesterone (P4) levels in the serum were measured by Calbiotech and IBL ELISA, respectively. All assays were performed by the Ligand Assay and Analysis Core Laboratory of the University of Virginia (Charlottesville, VA, USA).

Embryo culture and imaging

Embryos were obtained from female Brg1^fl/fl and Brg1^fl/fl; Amhr2-Cre mice (age: 8–10 weeks) that had been naturally mated with a fertile WT male. Two-cell embryos were flushed from excised oviducts ∼1.5 days after mating (after observation of vaginal plug) with fresh washing media (KFHM). Two-cell embryos were cultured in embryo culture media (KSOM) overlaid with mineral oil at 37°C in 5% CO₂ in air (designated as Day 1), and the number of four-cells or greater on Day 2, morula on Day 3, and blastocyst formation on Day 4 was recorded. For imaging, embryos were fixed with 4% (w/v) paraformaldehyde for 30 min at 37°C and permeabilized (0.25% triton-X 100). They were then stained with phalloidin 488 and Hoechst followed by imaging with a Zeiss 880 confocal microscope (Cell Biology and Image Acquisition core, University of Ottawa). Embryos were imaged with a ×40 oil immersion objective. Z-slices of 2 µm were captured across the whole embryo and scored using ImageJ (Fiji) software.

RNA sequencing data retrieval

RNA sequencing data from a previous study of human folliculogenesis assessing gene expression in GCs and oocytes (GSE107746) were retrieved from the Gene Expression Omnibus website ( https://www.ncbi.nlm.nih.gov/geo). Expression levels of SMARCA4 were directly extracted from a transcriptomic study of the human ovary and are provided in fragments per kilo-million mapped reads (FPKM) plus one then log2 transformed from GCs and oocytes originating from the same follicular stage [24]. RNA sequencing data from cow early embryonic development were retrieved from EmbryoGENE profiler website ( https://emb-bioinfo.fsaa.ulaval.ca/IMAGE/). Briefly, transcript abundances in FPKM were normalized against the spiked synthetic transcriptome ERCC abundances in each library to obtain an absolute estimate of gene expression, which can be compared between developmental stages [25]. RNA-seq data sets were not reanalyzed but were statistically evaluated for differences in expression between follicle stages using an ANOVA with Tukey's multiple comparison test.

Statistical analyses

All statistical analyses were performed using Prism 9 software (GraphPad Software Inc., La Jolla, CA, USA). Data sets were compared using t-test, two-tailed t-test with Mann–Whitney post hoc, or ANOVA (with Tukey's or Dunnett's multiple comparisons posttest) when specified. All data are presented as means ± SEM.

Smarca4 is expressed in GCs and oocytes throughout folliculogenesis and early embryogenesis

To investigate the expression pattern of SMARCA4, we used data from a previous RNA sequencing analysis of human GCs and oocytes and verified the expression during folliculogenesis and oogenesis [24]. SMARCA4 is expressed at all stages of human folliculogenesis in both GCs and oocytes with significantly increased expression in GCs with stage of follicle progression and a sharp decrease in oocytes in secondary follicles relative to the other stages (Figure 1A and B). Using data from a previous RNA sequencing study on bovine early embryonic development [26], we noted that SMARCA4 tends to decrease before the ZGA, which occurs at the eight-cell stage in cows [27]. Interestingly, SMARCA4 expression is significantly increased in the expanded blastocyst compared with eight-cells and morula (Figure 1C). These analyses show that SMARCA4 is expressed early during folliculogenesis and oogenesis and is also strongly expressed in the embryo following the ZGA, supporting its role in the embryonic development in cows and women. We then stained WT mouse ovaries for BRG1 by immunofluorescence to assess the spatial expression within the ovary. BRG1 was detectable in GCs at all stages of follicular development, similar to other species, but was also expressed in the stroma, theca cells, and in the luteinized GCs. Interestingly, BRG1 expression was also detected specifically in the epithelium of the oviduct and uterus (Figure 1D).

Smarca4 expression is successfully reduced in cKO mutant mice

To delete the Smarca4/BRG1 from GCs and oocytes, we used conditional recombination of Brg1 floxed/floxed mice crossed with mice heterozygous for a “knock-in” allele containing a Cre-Neo cassette inserted into Amhr2 [18], denoted as Amhr2-Cre or crossed with transgenic mice with Cyp19 or Gdf9 promotor ligated to iCre cDNA [19, 20], denoted Cyp19-Cre and Gdf9-Cre, respectively. Cre-mediated recombination occurs at different time points during folliculogenesis, depending on the timing of activation of the allelic promoters. Briefly, Cre activity in Amhr2-cre mice was reported as early as embryonic day e11.5 in female urogenital ridges and at embryonic day e12.5 in gonads and Müllerian ducts [28]. In postnatal ovaries, Amhr2-Cre activity was reported in GCs of all secondary and antral follicles, with lower to no detection in primordial and primary follicles. However, low Cre activity was also found in theca cells, oocytes [28], and in the ovarian surface epithelium (OSE) [29]. In the postnatal ovaries, Cyp19-Cre activity was reported in GCs of all antral follicles and most luteal cells with undetectable expression in primordial/primary follicles, theca cells, oocytes [19], or OSE [29]. Cre activity in GDF9-cre mice was reported as early as the primordial follicle stage in 3-day-old mice and older with no expression in somatic cells [20]. The conditional recombination of Brg1 by Cre is performed using a Cre-LoxP system for which LoxP sites had been inserted in the introns located upstream of exon 17 and downstream of exon 18 [16]. Using RT-qPCR, we designed primers specific to Exons 17 and 18 of the Smarca4 gene ( Supplementary Table 1 (supplementaltable1.pdf)) and confirmed the reduced expression of Smarca4 transcripts in GCs from the Amhr2-Cre and Cyp19-Cre mutant mice at 48 h after eCG and in oocytes from the Gdf9-Cre mutant compared with the appropriate control mice (Figure 2A–C). We performed immunof luorescence and confirmed the depletion of BRG1 in GCs from antral follicles from the Amhr2-Cre and Cyp19-Cre mutant mice and found no expression in oocytes from the Gdf9-Cre mutant (  Supplementary Figure 1 (supplementalfigure1_revised.pdf)).

Conditional KO of BRG1 using Amhr2-Cre and Gdf9-Cre in female mice results in sterility and subfertility, respectively

To investigate the effect of maternal BRG1 on female fertility, we conducted a breeding study over a period of 6 months. Eight-week-old WT (C57BL/6) males were mated with 8-week-old control mice (Brg1^fl/fl), somatic cell mutants (Brg1^fl/fl;Amhr2-Cre and Brg1^fl/fl;Cyp19-Cre), and germ cell mutants (Brg1^fl/fl;Gdf9-Cre). Number of litters and size were documented (Figure 3A). None of the six female Amhr2-Cre mutant mice had any pups and were classified as sterile. The six Gdf9-Cre mutant females had a total of 26 litters with only 102 pups for an average of 4.02 pups per litter, which was significantly fewer than control mice. Linear regression on the number of pups per litter over time shows no significant change in the slope, but did show a significant difference in the intercept compared with control mice (Figure 3B), reflecting an earlier onset of the decline in fertility. Interestingly, the fertility of Gdf9-Cre female mice did not have the increase in litter size normally seen during the first 2–3 months following pairing with a fertile WT male. Instead, their fertility decreased from the start.

Fertility of female mice with Cyp19-Cre recombination was not significantly different from control mice. Since it has been previously reported that BRM might compensate for BRG1 loss [8], we tested this possibility by crossing the Brg1^fl/fl;Cyp19-Cre mouse model to Brm null mice. Female mice with dKO for both BRM and BRG1, noted as Brm^–/–;Brg1^fl/fl;Cyp19-Cre, showed no differences in fertility compared with control mice (Figure 3C). Henceforth, data from Cyp19-Cre will not be shown because of lack of any apparent phenotype. Our results suggest that depleting BRG1 in GCs early during folliculogenesis negatively impacts fertility, whereas later depletion, during the preovulatory stage, has no impact on fertility. Since deleting BRG1 early in oocyte growth decreased by half the number of pups, rendering these mice sub-fertile, we further explored the stages of development where defects became apparent.

BRG1-deficient oocytes are associated with fewer ovulations

To investigate whether the reduced reproductive performance of Brg1 cKO female might be explained by an ovulation deficiency, we counted the number of ovulated COCs in mutant mice (Brg1^fl/fl;Amhr2-Cre and Brg1^fl/fl;Gdf9-Cre) following either natural mating or superovulation. Naturally mated female mice (2 months old) were euthanized after observation of the vaginal plug and COCs in the ampulla were counted. In naturally mated mice, there was no difference in total number of COCs per mouse in Amhr2-Cre mutants (mean 7.60 ± 0.67) compared with control mice (mean 7.16 ± 0.87) but a significant reduction to about half the number of COCs from Gdf9-Cre mutant mice (mean 3.70 ± 0.62) (Figure 3D). Similar results were found in the superovulated immature mice, where the Amhr2-Cre mutant mice showed no difference (mean 38.50 ± 6.94) in the total number of COCs compared with controls (mean 36.50 ± 2.59), but the number of COCs was significantly reduced in the Gdf9-Cre mutant mice (mean 14.00 ± 2.30) (Figure 3E). The results of these experiments suggested that the defect in Amhr2-Cre mutant mice occurs after ovulation, whereas Gdf9-Cre mutant mice have deficiencies in follicular growth and/or ovulation.

Figure 1.

SMARCA4/BRG1 is expressed during female reproduction. SMARCA4 expression level in FPKM mapped reads plus 1 then log2 transformed from normalized RNA-seq data of human GCs (A) and oocytes (B) originating from primordial, primary, secondary, antral, and preovulatory follicles from women. (C) Time course expression of SMARCA4 in cow early embryo development in relative expression, total FPKM abundances normalized against the ERCC synthetic transcriptome abundances (Life Technologies). (D) Immunofluorescence of BRG1 (FITC) and DAPI in ovaries and reproductive tracts from WT mice at 2 months of age. Separate channels and a merge are shown at ×40Xmagnification, scale bar is 55 µm and a minimum of three samples of each genotype have been analyzed (^*P <0.05, ^**P <0.01, ^***P <0.001).

BRG1 cKO female mice have ovarian histologic anomalies

To identify whether the ovulation anomalies in Gdf9-Cre mutant mice are associated with poor follicular development or failed ovulation, we compared the ovaries of control mice with those of Amhr2-Cre and Gdf9-Cre mutant mice. Mutant mouse ovaries were significantly heavier compared with control mice in both immature (21–26 days) and adult cycling mice (6–8 months) ( Supplementary Figure 2A (supplementalfigure2.pdf)). We also noted the occasional presence of luteinized follicles in immature ovaries of Gdf9-Cre mutant mice only ( Supplementary Figure 2B (supplementalfigure2.pdf)).

The number of primordial follicles showed no obvious differences between the groups at the two different ages assessed, 2 and 8 months (Figure 4A and B). However, the number of primary follicles was significantly higher in younger mutant mice and significantly lower in older mutant mice compared with controls. The same tendency occurred for the number of secondary follicles but not for the tertiary follicles. Female mice from the Gdf9-Cre genotype had decreased numbers of tertiary (antral) follicles at both ages and increased numbers of atretic follicles. Although similar in trend, that difference was not significant for female mice of the Amhr2-Cre genotype. Overall, the histologic analysis showed that Gdf9-Cre mutant mice, despite having more follicles activated to grow, are disrupted late in follicle development, resulting in fewer healthy antral follicles and therefore fewer ovulations. Interestingly, Amhr2-Cre mutant mice also activated more follicles but retained enough to become healthy antral follicles with no apparent impairment in ovulation.

Figure 2.

Brg1 is effectively knocked down in mutant mice. (A–C) Expression of Brg1 was determined by RT-qPCR in GCs and oocytes isolated from immature eCG-treated mice of the indicated genotypes, Brg1^fl/fl;Amhr-Cre, Brg1^fl/fl;Cyp19-Cre and Brg1^fl/fl;Gdf9-Cre models (n = 5 animals/time point). Data are means ± SEM. ^* P < 0.05, ^** P < 0.01.

Figure 3.

Brg1 mutant mice have altered reproductive performances and ovulation capability. (A) Fertility data of the control group Brg1^fl/fl, and Brg1^fl/fl;Amhr2-Cre, Brg1^fl/fl;Cyp19-Cre and Brg1^fl/fl;Gdf9-Cre mice are presented. Total litters and total pups are the total of n= 6 pairs per group. Each female mouse was bred with a fertile WT male for a period of 6 months and the number of pups was counted after each birth. Litter size refers to the average number of pups per litter. Differences in litter size compared with control group were analyzed using one-way ANOVA with Dunnett multiple comparison test. Data are means ± SEM. (B) Fertility rate gradually decreased over fertility trial period in Brg1^fl/fl;Gdf9-Cre mice compared with control group. Simple linear regressions are shown in dashed black and red lines with their respective equation. Intercepts are significantly different between groups. (C) Fertility rate in Brm^–/–;Brg1^fl/fl;Cyp19-Cre is not different from single-Brg1^fl/fl;Cyp19-Cre knockout and Brg1^fl/fl control mice. (D, E) Number of COCs counted after ovulation in naturally cycling females and after superovulation from three female mice per genotype. Data are presented as means ± SEM and analyzed using ANOVA with Dunnett's multiple comparison test (^*P <0.05, ^**P <0.01, and ^****P <0.0001, ns = not statistically different).

Figure 4.

Brg1 mutant mice show variations in the proportions at each follicle stage. Follicle numbers at different stages in the ovaries of Brg1^fl/fl (black bars), Brg1^fl/fl;Amhr2-Cre (white bars) and Brg1^fl/fl;Gdf9-Cre (gray bars) at (A) 2 months and (B) 8 months were counted. Left ovaries from each indicated genotype (n= 3) were serially sectioned, and all follicles with visible nucleus from every fifth section were counted, categorized (primordial, primary, secondary, and tertiary), and scored (healthy or atretic). Data were analyzed using one-way ANOVA with Dunnett's multiple comparison test to compare significance against the control group. Data are means ± SEM (^*P <0.05, ^**P <0.01, ^***P <0.001).

BRG1 plays a role in female steroidogenesis

Since follicular growth is abnormal in mutant mice, we investigated if steroid hormone production was affected by BRG1 cKO. Serum hormone assays revealed a significant decrease in E2 production in Gdf9-Cre mutant mice compared with control mice, but no difference in Amhr2-Cre mutant mice neither after eCG nor after injection of both eCG and hCG. In contrast, both mutants showed a significant increase in progesterone levels following eCG and hCG compared with control mice (Figure 5A). To determine if there might be an association between steroid hormone levels in response to gonadotropin stimulation and changes in Smarca4 expression in normal ovaries, we performed RT-qPCR on GCs from WT mice and found that the post-hCG decrease in E2 levels was associated with increased expression of Smarca4 (Figure 5B).

BRG1 null GCs lead to poor oocyte quality

The follicle assessment shows that there are no differences between control and Amhr2-Cre mice in terms of healthy and atretic antral follicles and ovulations. We also found no significant difference in GV chromatin configuration between the control and Amhr2-Cre mutant mice ( Supplementary Figure 3 (supplementalfigure3.pdf)). To further investigate whether the infertility phenotype of the Amhr2-Cre mice might be explained by an embryonic phenotype, we collected and cultured two-cell embryos from 2-month-old Brg1^fl/fl control mice and Brg1^fl/fl;Amhr2-Cre mutant mice after natural mating with a WT male mouse. Figure 6A shows images of each developmental stage between two-cell embryos to the formation of blastocysts. There was no difference in the number of collected two-cell embryos between both groups, but the number of healthy blastocysts in the Amhr2-Cre group (mean 0.6 ± 0.40) was significantly reduced compared with the control group (mean 5.16 ± 0.47) (Figure 6B). We also noted that 83.7% of two-cell embryos collected from control mice reached blastocysts but only 7.89% of two-cell embryos from mutant mice developed into blastocysts (mean 7.60 ± 0.67 versus mean 0.60 ± 0.40). We then collected and cultured embryos from both control and Amhr2-Cre mutant mice to assess any anomalies under confocal microscopy. Figure 6C shows images of morulas stained for DNA and actin to better identify mitotic rates by counting cells in metaphase division and to assess micronuclei (mn), which are extra-nuclear bodies that contain damaged chromosome fragments because of mitotic defects. We discovered that embryos originating from mutant female mice had higher numbers of micronuclei per embryo (mean 4.82 ± 0.5 versus mean 2.68 ± 0.51) and also had higher cell counts compared with controls (mean of 50 ± 2.53 versus mean 41 ± 3.39). Although the mitotic index was slightly lower and the number of micronuclei per cell was slightly higher, neither difference was significant (Figure 6D–G). The greater abundance of extra-nuclear DNA suggests mitotic division errors, which may contribute to the failure to form viable blastocysts. Since it is known that accelerated follicle growth contributes to reduced oocyte quality [30], we sought to compare the rates of GC proliferation in association with follicle state in the GC mutant mice. As expected, the rate of proliferation of GCs in Amhr2-Cre mutant mice was higher in both primary and antral follicles, as shown by immunostaining (IHC) for Ki67 (Figure 7).