miR-135a Suppresses Granulosa Cell Growth by Targeting Tgfbr1 and Ccnd2 during Folliculogenesis in Mice

The success of female reproduction relies on high quality oocytes, which is determined by well-organized cooperation between granulosa cells (GCs) and oocytes during folliculogenesis. GC growth plays a crucial role in maintaining follicle development. Herein, miR-135a was identified as a differentially expressed microRNA in pre-ovulatory ovarian follicles between Large White and Chinese Taihu sows detected by Solexa deep sequencing. We found that miR-135a could significantly facilitate the accumulation of cells arrested at the G1/S phase boundary and increase apoptosis. Mechanically, miR-135a suppressed transforming growth factor, beta receptor I (Tgfbr1) and cyclin D2 (Ccnd2) expression by targeting their 3′UTR in GCs. Furthermore, subcellular localization analysis and a chromatin immunoprecipitation-quantitative real-time PCR (ChIP-qPCR) assay demonstrated that the TGFBR1-SMAD3 pathway could enhance Ccnd2 promoter activity and thus upregulate Ccnd2 expression. Finally, estrogen receptor 2 (ESR2) functioned as a transcription factor by directly binding to the miR-135a promoter region and decreasing the transcriptional activity of miR-135a. Taken together, our study reveals a pro-survival mechanism of ESR2/miR-135a/Tgfbr1/Ccnd2 axis for GC growth, and also provides a novel target for the improvement of female fertility.


Introduction
Folliculogenesis, from primordial follicle activation to oocyte release, is a complex and dynamic process that relies on synchronization between the oocyte maturation and the neighboring granulosa cell growth. During this process, ovarian granulosa cells (GCs) provide necessary nutrients and steroids to the oocytes, and therefore play vital roles in ovarian follicle development [1]. In the awakening stage of mammalian oocytes, the flattened GCs from primordial follicles first proliferate and differentiate into cuboidal granulosa cells. This is followed by the dramatic growth of oocytes [2]. Subsequently, GCs rapidly proliferate to form an antral structure composed of multilayer granulosa cells and differentiate into mural GCs and cumulus GCs [3,4]. Only a subset of the antral follicles responds to LH and further releases fertilizable oocytes. The remaining GCs in the raptured follicle will undergo terminal differentiation and form the corpus luteum, which secretes progesterone necessary to maintain pregnancy [5]. GC growth is necessary for follicular maturation and ovulation, whereas GC apoptosis would lead to follicular atresia and degradation [6][7][8].
MicroRNAs (miRNAs) are small, noncoding RNAs that negatively regulate gene expression post-transcriptionally [9]. Currently, miRNAs have been suggested to regulate GC functions and thus closely participate in folliculogenesis, oogenesis, and steroidogenesis [10]. MiR-143 plays a critical role in follicular atresia by regulating cell apoptosis and steroidogenesis [11], and miR-126-3p promotes cell proliferation in the porcine granulosa cells [12]. MiR-16 promotes ovarian GC proliferation and inhibits GC apoptosis in polycystic ovarian syndrome [13]. Notably, miR-135a (as a tumor suppressor) is involved in regulating cell cycle and cell proliferation [14]. The upregulation of miR-135a promotes cell apoptosis and inflammation, along with inhibited cell proliferation and decreased macrophage autophagy [15]. Furthermore, miR-135a participates in the regulation of cyclin E1 (CCNE1) expression and the accumulation of cells arrested at the G1/S phase boundary [16]. In addition, hsa-miR-135 is associated with the survival of patients who have serous ovarian carcinoma, and thus may serve as a potential prognostic biomarker [17]. Our previous work identified that the expression of miR-135a was significantly different in pre-ovulatory ovarian follicles of Large White (LW) and Taihu sows detected by Solexa deep sequencing [18]. Moreover, miR-135a promotes apoptosis and the DNA damage response in GCs in polycystic ovary syndrome [19]. Therefore, investigating the role of miR-135a in female reproduction is indispensable Transforming growth factor (TGF)-β signaling activates its own membrane serine/ threonine kinase receptors TGFBR2 (type II receptor) and TGFBR1 (type I receptor) to promote the binding of SMAD family member (SMAD) 2/3 intracellular signaling to SMAD4 in the nucleus. Then, the SMAD complex regulates transcription by binding to the targeted promoter regions referred to as SMAD-binding elements (SBEs) [20]. TGF-β signaling pathway is crucial for ovarian granulosa cell growth and female fertility [21]. FSH and LH could regulate natriuretic peptide C (NPPC) levels via regulating the levels of TGFB1, TGFBR2, and TGF-β downstream SMAD proteins in GCs and controlling the process of oocyte meiosis [22]. On the other hand, miRNAs can regulate GC function and follicle development by targeting TGF-β signaling pathway. MiR-424/503 cluster members modulate bovine granulosa cell proliferation and cell cycle by targeting SMAD7 [23]. MiR-181a regulates porcine GC apoptosis by targeting TGFBR1 [24], and miR-130a/TGF-β1 axis is involved in sow fertility by regulating granulosa cell apoptosis [25].
In our study, we determined the role of miR-135a in mGC growth retardation and identified Tgfbr1 and Ccnd2 as the targets of miR-135a. In addition, estrogen-induced upregulation of Tgfbr1 and Ccnd2 mRNA levels attributed to the binding of ESR2 to canonical estrogen response element (ERE) in miR-135a promoter. In conclusion, our study provides new insight into understanding the network of ESR2/miR-135a/TGFBR1/CCND2 in mGC growth and follicle development.

Animals and Collection of Ovaries
All mice were housed in a pathogen-free environment at 20-22 • C, 50-70% relative humidity, and under a 12 h/12 h light/dark cycle. All mice had ad libitum accesses to a standard chow diet. Twenty-one-day-old female Kunming mice (Center for Disease Control; Hubei, China) were injected intraperitoneally (i.p.) with 10 units of pregnant mare serum gonadotropin (PMSG; Ningbo Second Hormone Factory, Ningbo, China). After 48 h, mice were killed by cervical dislocation, and their ovaries were harvested for in vitro experiments. All animal experiments were conducted in accordance with the guidelines of the Animal Care and Ethics Committee of Huazhong Agricultural University.

Luciferase Reporter Assay
Each recombinant construct plasmids were transfected with pRL-TK (E2241, Promega, Wisconsin, USA) into mGCs. After 24 h, cells were collected and luciferase activities were measured using the Dual-Luciferase Reporter Assay System (E1910, Promega, Wisconsin, USA) according to the manufacturer's instructions. The experiments were repeated at least three times, and the results were expressed as the means ± SD.

Quantitative Real-Time PCR
Total RNA and cDNA were prepared from cultured mGCs as described previously [18]. Quantitative real-time PCR was performed using the iTaq TM Universal SYBR Green Super Mix (172-5121, Bio-Rad, Hercules, CA, USA) and analyzed on CFX384 Touch™ Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). The primers are listed in Table S1.

Chromatin Immunoprecipitation
pCMV-N-HA-ESR2 was constructed and transfected into mGCs for a Chromatin immunoprecipitation (ChIP) assay. ChIP was performed using the EZ-ChIP Kit (Millipore, MA, USA) according to the manufacturer's instructions. The AVCX130 system (Sonics & Materials, Newtown, CT, USA) was used for cell sonication. Anti-HA (ab9110, Abcam), anti-SMAD3 (ab208182, Abcam) and normal anti-mouse-IgG (ab6789, Abcam, Cambridge, UK) were used for the immunoprecipitation reactions. DNA fragment from ESR2-immunoprecipitated complex was quantified via qPCR. The primer sequences are described in Table S1.

Flow Cytometry Analyses
Fluorescence-activated cell sorting (FACS) was used to measure cell cycle and apoptosis. For the analysis of cell cycle, cells stained with Propidium Iodide (PI) according to the manufacturer's manual. The apoptosis experiments were performed according to the manufacturer's protocol for the Annexin V-FITC Apoptosis Detection Kit (AD10, Dojindo, Shanghai, China).

Cell Viability Assay
Cell viability was determined using the Cell Counting Kit-8 (CCK-8; RM02823, Abclonal, Wuhan, China) assay according to the manufacturer's instructions.

Statistical Analysis
All results are presented as the mean ± SD. Each treatment had three replicates. Twotailed t-test was used when two groups were compared. Significant differences were evaluated using an independent-samples t-test. p < 0.05 was considered statistically significant.

MiR-135a Inhibits Cell Cycle and Proliferation in Murine GCs
Our previous study has identified that the lower level of miR-135a in pre-ovulatory ovarian follicles of Taihu sows with characteristics of high ovulation rate and large litter size, compared with that of LW sows by Solexa deep sequencing ( Figure S1A) [18]. To investigate the role of miR-135a in mGC growth, miR-135a overexpression or inhibition was performed in mGCs (Figure S1B,C). Ki67 staining and CCK-8 assay demonstrated that miR-135a overexpression in mGCs resulted in a reduction of Ki67-positive cells and a decline of cell vitality, which was in accordance with the observation of high proliferative activity and cell vitality in miR-135a inhibited mGCs ( Figure 1A-C). Moreover, fluorescenceactivated cell sorting (FACS) analysis revealed that miR-135a overexpression obviously increased the cell population of the G1 phase, accompanied by an elevated rate of G1/S and a decreased proliferation index (PI, PI = (G2 + S)/G1) in mGCs ( Figure 1D). However, miR-135a inhibition reduced the number of GCs in the G1 phase and improved the number of mGCs in the S phase, accompanied by decreased G1/S arrest and increased PI in mGCs ( Figure 1D). Subsequently, western blot analysis showed that miR-135a overexpression significantly repressed the protein levels of cyclin-dependent kinase 4 (CDK4), cyclin E1 (CCNE1) and proliferating cell nuclear antigen (PCNA) that were the important regulators of cell cycle and cell proliferation ( Figure 1E), while miR-135a inhibition could upregulate these protein levels ( Figure 1F). These results suggest that miR-135a may be involved in cell cycle arrest at the G1/S phase. On the other hand, cell apoptosis assay demonstrated a higher apoptosis rate in miR-135 overexpressed mGCs and a lower apoptosis rate in miR-135 inhibited mGCs ( Figure 1G). Consistently, miR-135a overexpression increased BAX protein level and decreased BCL2 protein level, which was inverse in mGCs with miR-135a inhibition ( Figure 1H,I). These results suggest that miR-135a can repress cell proliferation and accelerate apoptosis in mGCs.

MiR-135a Modulates mGC Proliferation by Targeting Tgfbr1
To investigate the effect of TGFBR1 on mGC growth, the Tgfbr1 overexpressed vector was transfected into mGCs. Ki-67 staining and CCK-8 assay demonstrated that TGFBR1 could increase cell viability and the population of Ki-67 positive mGCs ( Figure 3A,B).
Then, FACS analysis demonstrated that Tgfbr1 overexpression elevated the cell number in S and G2 phases accompanied by decreased G1/S arrest and increased PI in mGCs ( Figure 3C). Consistently, CDK4, CCNE1, and PCNA protein levels were increased in Tgfbr1-overexpressed mGCs compared with control mGCs ( Figure 3D). Moreover, inhibition of endogenous miR-135a expression could increase cell viability, but this effect disappeared after Tgfbr1 knockdown in mGCs ( Figure 3E). In addition, we found that CCND2 enhanced cell viability ( Figure 3F) and Ccnd2 knockdown abolished the function of miR-135a inhibitor on cell viability ( Figure 3G). These results suggest that miR-135a can repress mGC proliferation via Tgfbr1 and Ccnd2. In addition, Tgfbr1 overexpression resulted in a decrease of apoptotic cell rate with 1.5-fold reduction of BAX protein level and 1.3-fold elevation of BCL2 protein level, compared with the control group ( Figure 3H,I).

TGFBR1-SMAD3 Signaling Pathway Promotes mGC Proliferation by Regulating Ccnd2 Expression
To investigate whether TGFBR1 is involved in cell cycle and cell proliferation via TGFβ signaling pathway, western blot analysis was used to detect TGFBR1-mediated

TGFBR1-SMAD3 Signaling Pathway Promotes mGC Proliferation by Regulating Ccnd2 Expression
To investigate whether TGFBR1 is involved in cell cycle and cell proliferation via TGFβ signaling pathway, western blot analysis was used to detect TGFBR1-mediated phosphorylation of SMADs (p-SMADs). Tgfbr1 overexpression obviously increased p-SMAD3 protein level in mGCs ( Figure 4A), whereas miR-135a expression could decrease p-SMAD3 protein level in mGCs ( Figure 4B). It has been suggested that p-SMAD3 can translocate into the nucleus from the cytoplasm and then activate the transcriptional expression of downstream genes [28]. Subsequently, immunofluorescence assay showed Tgfbr1 overexpression promoted the nuclear distribution of SMAD3, while miR-135a overexpression mainly resulted in nuclear exclusion of SMAD3 ( Figure 4C). Furthermore, a SMAD3 binding site (−794 bp to −782 bp) was predicted in the Ccnd2 promoter region by BIOBASE software. ChIP-qPCR analysis revealed that Tgfbr1 overexpression significantly increased the binding of SMAD3 to the promoter of Ccnd2 ( Figure 4D). Indeed, the mRNA and protein levels of Ccnd2 were increased in Tgfbr1-overexpressed mGCs ( Figure 4E,F). These results suggest that TGFBR1 regulates the transcription of Ccnd2 by facilitating the recruitment of SMAD3 to Ccnd2 promoter. phosphorylation of SMADs (p-SMADs). Tgfbr1 overexpression obviously increased p-SMAD3 protein level in mGCs ( Figure 4A), whereas miR-135a expression could decrease p-SMAD3 protein level in mGCs ( Figure 4B). It has been suggested that p-SMAD3 can translocate into the nucleus from the cytoplasm and then activate the transcriptional expression of downstream genes [28]. Subsequently, immunofluorescence assay showed Tgfbr1 overexpression promoted the nuclear distribution of SMAD3, while miR-135a overexpression mainly resulted in nuclear exclusion of SMAD3 ( Figure 4C). Furthermore, a SMAD3 binding site (−794 bp to −782 bp) was predicted in the Ccnd2 promoter region by BIOBASE software. ChIP-qPCR analysis revealed that Tgfbr1 overexpression significantly increased the binding of SMAD3 to the promoter of Ccnd2 ( Figure 4D). Indeed, the mRNA and protein levels of Ccnd2 were increased in Tgfbr1-overexpressed mGCs ( Figure 4E,F). These results suggest that TGFBR1 regulates the transcription of Ccnd2 by facilitating the recruitment of SMAD3 to Ccnd2 promoter.

Transcription Factor ESR2 Negatively Regulates miR-135a Expression
To identify the regulatory elements of miR-135a expression in mGCs, a series of truncated miR-135a promoters were used to drive luciferase gene expression to detect promoter activity. Notably, the luciferase assay showed that pGL3-miR-135a-D5 had the highest luciferase activity, suggesting that the fragment from -250 bp to -100 bp in the 5 flanking region contained negative regulatory elements for miR-135a promoter activity ( Figure 5A). Subsequently, two ESR2 binding sites (−197 bp~−183 bp, −141 bp~−127 bp) were predicted in the miR-135a promoter region by the BIOBASE software ( Figure 5B). To identify ESR2 binding sites, the mutant pGL3-miR-135a-D5 vectors were constructed by site-directed mutagenesis. As a result, compared with the wild-type pGL3-miR-135a-D5, two single ESR2-element mutants and the double ESR2-element mutant showed obvious increases in promoter activity, which suggests that ESR2 could bind to the miR-135a promoter at two sites (−197 bp~−183 bp, −141 bp~−127 bp) and repress miR-135a promoter activity in mGCs ( Figure 5C). ChIP-qPCR analysis also demonstrated that ESR2 could specifically bind to the miR-135a promoter region ( Figure 5D). and Ccnd2 expression in estrogen-treated mGCs. The means ± SD were calculated from three independent experiments. * p < 0.05, ** p < 0.01, *** p < 0.001.

Discussion
The success of female reproduction relies on high quality oocytes which is determined by the well-organized process of follicle development. In the whole process, granulosa cells are mainly responsible for protecting and supporting the oocyte through providing essential nutrients, growth factors, and steroids [29,30]. Therefore, normal GC Then, the luciferase assay showed that ESR2 could significantly decrease pGL3-miR-135a-D4-driven luciferase activity, and further downregulate endogenous miR-135a expression in mGCs ( Figure 5E,F). As ESR2 can regulate gene expression via canonical estrogen-dependent transcriptional activation, we investigated the effect of estrogen on miR-135a, Tgfbr1 and Ccnd2 expression. Upon estrogen stimulation, miR-135a level was significantly decreased by 3-fold, and Tgfbr1 and Ccnd2 mRNA levels were increased by 1.6-and 2.1-fold, respectively ( Figure 5G). These results suggest that estrogen promotes the binding of ESR2 to the miR-135a promoter region and thus negatively regulates its endogenous expression to improve Tgfbr1 and Ccnd2 expression.

Discussion
The success of female reproduction relies on high quality oocytes which is determined by the well-organized process of follicle development. In the whole process, granulosa cells are mainly responsible for protecting and supporting the oocyte through providing essential nutrients, growth factors, and steroids [29,30]. Therefore, normal GC growth plays a crucial role in maintaining follicle development. In our study, we revealed that miR-135a was a proliferation repressor, resulting in cell cycle arrest at G1 phase in mGCs. Mechanically, we confirmed that miR-135a could target the 3 UTR of Tgfbr1 and Ccnd2. Moreover, TGFBR1-SMAD3 pathway enhanced endogenous Ccnd2 expression by promoting the transcriptional activity of Ccnd2. Furthermore, ESR2 could negatively regulate miR-135a expression via binding ERE element in miR-135a promotor region in mGCs ( Figure 6). promoting the transcriptional activity of Ccnd2. Furthermore, ESR2 could negatively regulate miR-135a expression via binding ERE element in miR-135a promotor region in mGCs ( Figure 6).

Figure 6.
A schematic model for miR-135a regulating granulosa cell growth. The schematic diagram depicting the mechanism that miR-135a regulates murine granulosa cell growth. First, miR-135a binds to 3′UTR of Tgfbr1 and Ccnd2 to repress their expression. Then, downregulated TGFBR1 fails to facilitate nuclear distribution of SMAD3, which could not enhance Ccnd2 expression. Finally, ESR2 functions as a transcription factor to directly bind to the miR-135a promoter region, decreases the transcriptional activity of miR-135a, and eventually suppresses miR-135a-mediated mGC growth suppression.
It has been suggested that miR-135a is closely related to the negative regulation of TGFβ-SMAD signaling pathway. miR-135a suppresses TGF-β-mediated epithelial-mesenchymal transition by targeting SMAD3 [31] and SMAD5 [32]. Moreover, miR-135a-5p is a key regulator of the TGFBR1/TAK1 pathway, resulting in the attenuation of vascular in- Figure 6. A schematic model for miR-135a regulating granulosa cell growth. The schematic diagram depicting the mechanism that miR-135a regulates murine granulosa cell growth. First, miR-135a binds to 3 UTR of Tgfbr1 and Ccnd2 to repress their expression. Then, downregulated TGFBR1 fails to facilitate nuclear distribution of SMAD3, which could not enhance Ccnd2 expression. Finally, ESR2 functions as a transcription factor to directly bind to the miR-135a promoter region, decreases the transcriptional activity of miR-135a, and eventually suppresses miR-135a-mediated mGC growth suppression.
It has been suggested that miR-135a is closely related to the negative regulation of TGFβ-SMAD signaling pathway. miR-135a suppresses TGF-β-mediated epithelial-mesenchymal transition by targeting SMAD3 [31] and SMAD5 [32]. Moreover, miR-135a-5p is a key regulator of the TGFBR1/TAK1 pathway, resulting in the attenuation of vascular inflammation in rats with chronic kidney disease [33]. Our study identified that miR-135a could suppress the expression of Tgfbr1 via targeting its 3 UTR. Previous studies reported that TGFBR1 was regulated by several miRNAs [24,34,35]. Both miR-1343 and miR-181a could increase GC apoptosis by targeting Tgfbr1 [11,34]. miR-140-5p downregulates Tgfbr1 expression, causing cell cycle arrest at the G1/S phase [35]. Consistently, our study found that miR-135a could result in G1/S arrest and cell proliferation suppression in mGCs, and miR-135a could regulate TGFBR1-mediated cell proliferation. These results provide plenty of evidence to verify that miR-135a represses cell cycle and cell proliferation via targeting Tgfbr1 in mGCs.
CCND2 is a specific cell cycle regulator during G1/S transition [36] and a proliferation regulator during folliculogenesis [37]. The CCND2-CDK4 complex phosphorylates retinoblastoma-associated protein and regulates the cell cycle during G1/S transition. Phosphorylation of RB1 allows dissociation of the transcription factor E2F from the RB/E2F complex and subsequently activates transcription of E2F-targeted genes which are responsible for cell cycle [38]. CCND2 expression is also regulated by many microRNAs. MiR-206 inhibits cell proliferation by targeting Ccnd2 and retarding G1/S phase transition in human laryngeal squamous cells [39]. MiR-373-3p inhibits cell propagation and boosts apoptosis in gemcitabine resistance pancreatic carcinoma cells by targeting Ccnd2 [40]. In this study, we found that miR-135a could repress Ccnd2 expression by targeting its 3 UTR and regulate CCND2-mediated mGC proliferation. As CCND2 is widely known as a cyclin, our study did not further explore its roles in cell cycle and proliferation.
Binding of TGF-β to its receptor leads to activation of the transcription factor SMAD3, and SMAD3 translocates into the nucleus where the factors can induce transcriptional expression of targeted genes [41][42][43]. Our results showed Tgfbr1 overexpression significantly increased but miR-135a overexpression obviously decreased the expression level of p-SMAD3 and nuclear distribution of SMAD3. It has been reported that SMAD3 coprecipitates with upstream DNA sequences of Ccnd2 transcription start sites in mGCs [44]. Indeed, our study found Tgfbr1 overexpression could enhance endogenous Ccnd2 expression via binding of SMAD3 to the Ccnd2 promoter, whereas the function of TGFBR1 was further suppressed in miR-135a overexpressed mGCs. These data suggest that miR-135a inhibits Ccnd2 expression via the targeted mRNA inhibition and suppression of TGF-β signaling. miRNA expression is also regulated by transcription factors in granulosa cells. Steroidogenic factor-1 suppresses miR-383 transcription and then mediates estradiol release in GCs [45]. TGF-β1 enhances the binding of p53 and NF-kB to the promoter region of the miR-244 host gene, promotes pri-miR-244 transcription, and affects GC proliferation and estradiol synthesis in GCs [46]. ESR2 can regulate its targeted genes through the canonical estrogen response element (ERE)-dependent transcriptional activation or through regulatory pathways independent of binding to the cognate response element [47]. Loss of ESR2 in mutant mice, resulting in diminished GC-responsiveness to gonadotropins, reduced estrogen production associated with impaired follicle maturation and ovulation failure [48]. Our study identified that ESR2 negatively regulated miR-135a expression through binding to two ERE elements in miR-135a core promoter region, and verified estrogen-induced negative regulation of miR-135a expression. Previous studies of miR-135a were overexpressed in GCs from PCOS patients, and promoted GC proliferation and repressed GC apoptosis via repressing ESR2 expression [49]. Therefore, we provide a protection mechanism of estrogen-mediated TGFBR1 and CCND2 mRNA away from miR-135a mediated degradation. Meanwhile, these results may suggest that miR-135a has a different expression pattern during folliculogenesis compared with the pathological process. Under physiological or pathological accommodation, miR-135a and ESR2 may determine the state of GC growth via repressing each other's expression (which merits further investigation).

Conclusions
In conclusion, our data document the roles of murine miR-135a in cell cycle retardation and cell proliferation suppression by suppressing Tgfbr1 and Ccnd2 expression. Our study suggests a pro-survival mechanism of ESR2/miR-135a/TGFBR1/CCND2 axis, which may provide a new method for improvement of female fertility.
Author Contributions: F.L. and L.W. designed the experiments; L.W. performed the experiments; L.W. and F.L. wrote the manuscript; L.W., Y.C., S.W., J.T. and G.C. contributed reagents/materials/ analysis tools. L.W., Y.C. and S.W. analyzed the data. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement:
The data that support the findings of this study are available from Fenge Li upon reasonable request.