Exosomal miR-143-3p Derived from PCOS Follicular Fluid Induces Granulosa Cell Apoptosis by Targeting BMPR1A and Suppression of Smad1/5/8 Signaling

Polycystic ovary syndrome (PCOS) is a common endocrine and metabolic disorders disease in women of reproductive age. The anovulation caused by abnormal follicular development is still the main characteristic of infertile patients with PCOS. Granulosa cells (GCs), an important component of follicular microenvironment, affect follicular development through GCs dysfunction. Increasing evidence indicates that exosomal miRNAs derived from follicular uid (FF) of patients play critical roles during PCOS. However, which and how follicular uid derived exosomal miRNAs play a pivotal role in controlling granulosa cells function and consequently follicular development remain largely unknown. Herein, we showed that miR-143-3p is highly expressed in follicular uid exosomes of PCOS patients and can be delivered into granulosa cells. Furthermore, the functional experiments showed that the translocated miR-143-3p promoted granulosa cell apoptosis, which are important in follicle development. In terms of mechanism, we demonstrated that BMPR1A was identied as a direct target of miR-143-3p. Overexpression of BMPR1A reversed the effects of exosomal miR-143-3p on GCs apoptosis and proliferation by activating Smad1/5/8 signaling pathway. These results demonstrate that miR-143-3p-containing exosomes derived from PCOS follicular uid promoted granulosa cell apoptosis by targeting BMPR1A and blockading Smad1/5/8 signaling pathway. Our ndings provide a novel mechanism underlying the roles of exosomal-miRNA in follicular uid of PCOS and facilitate the development of therapeutic strategies for PCOS.


Introduction
Polycystic ovary syndrome (PCOS) is a common endocrine and metabolic disorders disease of reproductive age women, characterized by chronic anovulation, polycystic ovarian morphology and hyperandrogenism 1,2 . The etiology is complex and the pathogenesis remains unclear. However, previous studies revealed that PCOS patients exhibited abnormal follicular development, which is the main cause of anovulatory infertility 3 .
The follicular microenvironment is essential for follicular development, oocyte maturation and quality 4 . It has been reported that the oocyte and surrounding granulosa cells (GCs) display an interdependent relationship in the stage of follicle development via direct gap junctions. It was also found that there was nutrient exchange and biological signal transmission between GCs and oocytes in the follicular uid microenvironment through paracrine or autocrine manners which regulated the growth, development and maturation of follicles 5,6 . Thus, the dysregulation of GCs affects the ovarian follicular microenvironment, which might impair folliculogenesis and nally lead to poor reproductive outcome in PCOS patients.
Therefore, exploring the function of GCs can explain abnormal follicular development of PCOS patients. Previous studies have reported that the apoptosis of GCs affected follicles development, oocyte growth and maturation, and triggered follicular atresia during the early phase of follicular development 7 .
Increasing evidence indicates that the apoptosis of GCs in PCOS patients was signi cantly higher than those in healthy controls group 8,9 . However, the underlying mechanism of abnormal apoptosis of GCs involved in the process of abnormal follicular development in PCOS remains unclear.
Follicular uid is the critical microenvironment for follicular development and contains a wide variety of biologically active molecules, including exosomes that have a lipid bilayer membrane structure with diameter of 30-150 nm 10 . The exosomes contain cell-speci c proteins, lipids, and nucleic acids which act as a carrier passes biological information into target cells 11 . MicroRNAs (miRNAs) are small single stranded non-coding RNA molecules with 21-24 nucleotides, which can regulate target gene expression by binding to the 3' untranslated regions (3' UTR) of target gene mRNAs to silence these genes at a posttranscriptional level 12 . A growing number of studies report that follicular uid and the exosomes derived from follicular uid of PCOS patients (PCOS-FF exosomes) have many differentially expressed nucleic acids especially miRNAs [13][14][15] . However, the effect of FF-exosomal miRNA on follicular development of PCOS patients remains unclear.
In the present study, our study mainly focused on exosomal-miR-143-3p and explored the effect of PCOS-FF derived exosomal-miR-143-3p on the apoptosis and proliferation of ovarian GCs, and then analyzed its underlying molecular mechanisms involved in follicular dysplasia in PCOS. Our ndings will hopefully afford perspectives to understand the progression of PCOS.

Results
Comparison of clinical general information, laboratory data and apoptosis related indexes of granulosa cells between PCOS and normal controls A total of 146 patients were enrolled in this study, including 80 healthy controls and 66 PCOS patients. As shown in Table 1, there were no substantial differences of patients' age, infertility years, the basal levels of E2, prolactin (PRL) and days of stimulation in two groups (p>0.05). Body mass index (BMI), the basal levels of LH, testosterone (T) and antral follicle count (AFC) in the PCOS group were signi cantly higher than these in the controls group (p<0.05). Meanwhile, the basal levels of FSH and the Gn dosage (IU) were signi cantly lower in PCOS group than that in the healthy control group (p<0.05). Laboratory data analysis indicated that the number of retrieved oocytes increased signi cantly, while MII oocytes rate, 2PN fertilization rate and high-quality embryos rate in the PCOS group decreased signi cantly compared with the healthy controls group (p<0.05). Consistently, we found that the pro-apoptotic gene Bax mRNA levels were signi cantly increased, and the anti-apoptotic gene Bcl-2 mRNA levels was signi cantly decreased in the PCOS group compared with the healthy controls group (Supplementary Fig.S1). Isolation and characterization of follicular uid-derived exosomes Exosomes derived from FF of healthy controls group and PCOS patients were isolated and characterized according to the above method. TEM analysis revealed that exosomes derived from FF of PCOS patients were round membrane-bound vesicles with a diameter of 30-150 nm, which were consistent with exosomes from controls (Fig. 1A). Western blotting showed that exosome marker protein TSG101 and HSP70 were enriched in these exosomes as expected (Fig. 1B). To further con rm whether FF-exosomes can be taken up by primary GCs, we labeled the FF-exosomes with PKH67 uorescent dye and then coincubated with GCs for 12 h. Immuno uorescence staining revealed that the PKH67-labeled exosomes were e ciently absorbed by GCs and transferred to the cytoplasmic compartment (Fig. 1C). Overall, these results imply that FF-exosomes can be successfully isolated and can be internalized by GCs.

PCOS-FF derived exosomes promote granulosa cells apoptosis and inhibit proliferation in vitro
To determine the functional role of PCOS-FF derived exosomes on GCs apoptosis and proliferation, we cultured primary granulosa cells and KGN cells with normal-FF derived exosomes and PCOS-FF derived exosomes for 48 h, respectively. As shown in Fig. 2A and B, the mRNA and protein expression levels of pro-apoptosis gene Bax were upregulated in GCs exposed to PCOS-FF derived exosomes, while the mRNA and protein expression levels of anti-apoptosis gene Bcl-2 were downregulated in GCs exposed to PCOS-FF derived exosomes. As expected, the apoptotic rates of primary GCs and KGN cells were signi cantly enhanced in the PCOS-FF derived exosomes treated group as determined by PE Annexin V assay ( Fig. 2C and D). We further evaluated the effects of PCOS-FF derived exosomes on granulosa cell proliferation in vitro via CCK8 assay, PCOS-FF derived exosomes treatment signi cantly inhibited growth capacity of primary GCs and KGN cells compared with PBS and normal-FF derived exosomes treatment ( Fig. 2E and F). Taken together, our ndings show that PCOS-FF derived exosomes promote apoptosis and inhibit proliferation of granulosa cells.

miR-143-3p is enriched in PCOS-FF derived exosomes rather than in normal-FF derived exosomes
Previous studies have shown that serum, follicular uid as well as granulosa cell release large amounts of microvesicles containing both coding and non-coding RNAs, including miRNAs with multiple functional properties [16][17][18][19] . Thus, we hypothesized that PCOS-FF-exosomes might enhance apoptosis of granulosa cells by transferring speci c miRNAs. Hence, we screened the differentially expressed miRNAs in follicular uid derived exosomes between normal population and PCOS patients according to the GRA001999 database of National Genomics Data Center, and found that miR-143-3p was enriched in exosomes of PCOS-FF, and the result was con rmed by the expression levels of miR-143-3p in normal-FFexosomes and PCOS-FF-exosomes as detected by qRT-PCR analysis ( Fig. 3A and B). In addition, the miR-143-3p levels was signi cantly increased in primary granulosa cells and KGN cells treated with PCOS-FFexosomes compared with PBS and normal-FF-exosomes ( Fig. 3C and D). Thus, our results suggest that miR-143-3p is enriched in PCOS-FF derived exosomes.

Exosomal-miR-143-3p promotes granulosa cells apoptosis and suppresses proliferation in vitro
To further clarify the biological roles of the miR-143-3p on granulosa cell apoptosis and proliferation, we successfully overexpressed or silenced miR-143-3p expression in KGN cells by using a miR-143 mimic or inhibitor, respectively (Fig. 4A). Western blotting results revealed that the protein levels of pro-apoptotic gene Bax was signi cantly increased by miR-143-3p mimic transfection and decreased by miR-143-3p inhibitor transfection. However, the protein levels of anti-apoptosis gene Bcl-2 was signi cantly decreased by miR-143-3p mimic transfection and increased by miR-143-3p inhibitor transfection. (Fig. 4B). Cell apoptosis assay showed that miR-143-3p overexpression signi cantly promoted cell apoptosis, whereas miR-143-3p inhibition showed an anti-apoptotic phenotype ( Fig. 4C and D). Meanwhile, CCK8 assay showed that miR-143-3p overexpression signi cantly enhanced cell proliferation, while miR-143-3p inhibition suppressed cell proliferation ( Fig. 4E and F). These results reveal that PCOS-FF derived exosomal miR-143-3p plays a crucial role in enhancing granulosa cell apoptosis and reducing proliferation.
BMPR1A is a direct target gene of miR-143-3p According to the online bioinformatics prediction based on Target Scan, a binding sequence of miR-143-3p was found in the 3'UTR of BMPR1A (Fig. 5A). Subsequently, dual-luciferase reporter gene assay was applied to verify their interaction. PGL-3-BMPR1A-wt plasmids and PGL-3-BMPR1A-mut plasmids were constructed and co-transfected with miR-143-3p mimic into the KGN cells. The results displayed that luciferase activity was reduced in KGN cells co-transfected with miR-143-3p mimic and wild-type BMPR1A 3'UTR reporter, while no luciferase activity change was found in KGN cells co-transfected with miR-143-3p mimic and mutant BMPR1A 3'UTR reporter (Fig. 5B).
To investigate the relationship between BMPR1A and miR-143-3p, we examined BMPR1A mRNA and protein expression in KGN cells transfected with the miR-143-3p mimic or inhibitor. As shown in Fig. 5C and D, the mRNA and protein expression of BMPR1A were decreased by miR-143-3p mimic transfection and increased by miR-143-3p inhibitor transfection. These data indicate that BMPR1A is a direct target of miR-143-3p.

BMPR1A overexpression attenuates the effects of miR-143-3p on cell apoptosis and proliferation in KGN cells
To explore whether BMPR1A was involved in the regulation of miR-143-3p on cell phenotype, we cotransfected miR-143-3p mimic and pcDNA-BMPR1A plasmid in KGN cells. Cell apoptosis and CCK8 assays indicated that the combined transfection of miR-143-3p mimic and BMPR1A expression plasmid showed the effect of inhibiting cell apoptosis and promoting cell proliferation ability compared with miR-143-3p mimic transfection alone ( Fig. 6A  processes, such as cell proliferation, apoptosis 26, 27 . In this study, over-expression of miR-143-3p signi cantly enhanced the apoptosis rate of GCs, accompanied by the inhibition of cell proliferation. Therefore, it is suggested that exosomal miR-143-3p in PCOS follicular uid may play an important role in PCOS follicular dysplasia by affecting the biological functions of GCs. Bone morphogenetic proteins (BMPs) belong to the transforming growth factor β (TGFβ) superfamily. Their biological effects are mediated by BMP-speci c type I (BMPR1A and BMPR1B) and type II (BMPR2) serine/threonine kinase receptors as well as SMAD-related proteins 28 . BMPs have been demonstrated required for normal folliculogenesis in the ovary by regulating several key biological processes including cell proliferation, differentiation, apoptosis, and steroidogenesis [29][30][31][32] . Thus, it is suggested that disruption of BMPs system may be involved in folliculogenesis disorders.
BMPR1A, is a bone morphogenetic protein receptor activin-like kinase and a key signaling molecule in the BMP signaling pathway 32 . In this study, the binding sites between miR-143-3p and BMPR1A were predicted by bioinformatics analysis, and luciferase reporter assays further veri ed the direct binding between miR-143-3p and BMPR1A in KGN cells.
Our results veri ed that treatment of KGN cells with the miR-143-3p promoted cell apoptosis and suppressed proliferation via directly targeting BMPR1A. Moreover, over-expression of BMPR1A protein partially conferred protection against miR-143-3p induced cell apoptosis and reversed the inhibitory effect of miR-143-3p on granulosa cell proliferation. In this study, we also found that miR-143-3p mediated cell apoptosis and proliferation in KGN cells were related to the phosphorylation of Smad1/5/8.
In conclusion, our research revealed for the rst time that miR-143-3p-containing exosomes derived from PCOS follicular uid promote granulosa cell apoptosis by targeting BMPR1A and blockading Smad1/5/8 signaling pathway. Therefore, these ndings provide novel insights into GCs dysfunction in PCOS and suggest that exosomal-miR-143-3p derived from follicular uid of PCOS patients may be a promising molecular target for the treatment of PCOS.

Methods
Clinical samples, ovarian stimulation Characterization of FF-exosomes was con rmed by transmission electron microscope (TEM): 5-10 µL exosomes were dropped onto the carbon-coated copper grids at room temperature for 3-5 min, and then absorb the excess liquid with absorbent paper. Then10 µL 2% phosphotungstic acid was pipetted on the grids for staining for 2-3 min, the excess uid was removed, and the grid was dried at room temperature. Finally, the copper grid was detected under TEM at 80KV (HITACHI). Exosome-speci c marker TSG101 and EV-associated protein marker HSP70 were analyzed by western blotting.

Isolation and culture of granulosa cells from follicular uids
On the day of oocyte retrieval, the rst tube of serum-free follicular uid was collected, and then centrifuged at 2000 g for 10 min, the supernatant was stored at -80°C for further experiments. The cell pellets were resuspended in phosphate-buffered saline (PBS) and 1:1 added onto coll solution, then centrifuged for 15 min at 1500 g. Cells at the interface were removed and washed twice with PBS, the cell pellets were resuspened in DMEM/F-12 (Hyclone, logan, UT, USA) medium containing 10% Fetal Bovine Serum (FBS) (Gibco, Carlsbad, CA, USA), supplemented with 100 U/mL penicillin and 100 µg/mL streptomycin at 37°C in 5% CO 2 cell culture incubator.

Exosome uptake assay
The puri ed exosomes were labeled with PKH67 green Fluorescent Cell Liner Kit (Sigma-Aldrich, St. Louis, MO, USA) according to the manufacturer's instructions. Brie y, 20 µg exosome-XE was added to 0.5mL Dilution C, 1µL PKH67 was added to 0.5mL Dilution C, then fully mixed and incubated with each other at room temperature for 3 min. 1mL 1% bovine serum albumin was then added to neutralize the excess dye. Then exosome-PKH67 was extracted with exoEasy Maxi Kit following the manufacturer's instructions (Qiagen, Germany). The labeled exosomes were then co-incubated with KGN cells for 24 h. Then the cells were xed with 4% paraformaldehyde at room temperature for 10min, washed twice with PBS, the nuclei were stained with DAPI. Finally, the signal was observed under confocal microscope.

RNA extraction, reverse transcription and RT-PCR analysis
Total RNA was extracted with Trizol reagent (Invitrogen, Carlsbad, CA, USA), and then 1µg of RNA was reverse-transcribed into cDNA. QRT-PCR was performed using the SYBR Premix Ex Taq kit in accordance with the manufacturer's protocol. The exosomal RNA was extracted using exoRNeasy Serum/Plasma Starter Kit (Qiagen, Germany). miRNA was then reverse-transcribed into cDNA by Bulge-Loop TM miRNA RT Primer (RiboBio Co., Ltd Guangzhou, China). U6 small nuclear RNA was used as the internal reference of miRNAs. Ct values were indicated by using 2 −ΔΔCt method.
To quantify apoptosis related gene Bax and Bcl-2 mRNA expression, we used oligo d(T) 18 primers to reverse transcribe total RNA into cDNA. Then, qRT-PCR was performed by using SYBR Green dye and speci c primers for Bax, Bcl-2 and β-actin. The primer sequences we used are listed in Table S1.
Western blotting SDS lysis buffer freshly mixed with a protease and phosphatase inhibitor cocktail (Thermo Scienti c, Rockford, Cambridge, MA) was used to isolate proteins from cells. Western blotting assay was performed as described previously 33 . All the antibodies we used are listed in Table S2.

Apoptosis and proliferation assay
Cell apoptosis assay was conducted as described PE AnnexinV Apoptosis Detection Kit (BD Biosciences, San Jose, CA, USA) according to the manufacturer's instructions. The cells were pretreated with Exosomes or miR-143-3p mimic or inhibitor (RiboBio Co., Ltd Guangzhou, China), and then 48 h after treatment, cells were collected and diluted into the density of 1x 10 5 which resuspended in binding buffer, then 5 µL PE AnnexinV and 7-AAD were introduced. The cells were incubated at room temperature in darkness for 15 min. Apoptosis cells were observed by ow cytometry (BD Biosciences, Franklin Lakes, NJ, USA).
The cell proliferation assay was conducted according to manufacturer's instructions. KGN  Vector construction and dual-luciferase reporter assay MiRNAs that target BMPR1A were predicted by online software programs, TargetScan (TargetScan Human 7.1). The full-length of 3'UTR BMPR1A containing the predicted wild-type or mut-type miR-143-3p binding sites were ampli ed by PCR and then cloned into the PGL3-Basic reporter vector. PGL-3-BMPR1Awt plasmids and PGL-3-BMPR1A-mut plasmids co-transfected with miR-143-3p mimic into the KGN cells along with pRL-TK vector using lipo2000 reagent (Invitrogen, Carlsbad, CA, USA). After 24 h of transfection, luciferase activity was determined using dual-luciferase reporter assay system (Promega, Madison, WI, USA) according to the manufacturer's protocol, the Renilla luciferase activity was regarded as normalization.

Statistical analysis
Clinical data was analyzed by SPSS 17.0 software (SPSS Inc, Chicago, IL, USA). The measured data was presented as average and standard deviation of three independent experiments. Quantitative data analysis was performed using Graph Pad Prism software version 7.0. The differences of statistics signi cance between different treatments were evaluated by one-way ANOVA or Student t-test. P values lower than 0.05 was considered statistically signi cant.    Quantitative PCR analysis of the expression levels of miR-143-3p in primary granulosa cells and KGN cells after normal-FF-exos or PCOS-FF-exos treatment for 48 h. All data were presented as mean ± SD.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.