Identification and potential value of candidate microRNAs in granulosa cells of polycystic ovary syndrome

Abstract

BACKGROUND:

Polycystic Ovary Syndrome (PCOS) is a major cause of anovulatory infertility. Some studies showed that miRNAs were used as diagnostic/prognostic biomarkers for various diseases.

OBJECTIVE:

To identify candidate miRNAs in Granulosa Cells (GCs) of PCOS and evaluate their potential values for PCOS diagnosis.

METHODS:

We screened differentially expressed miRNAs in GCs between PCOS and controls by the microarray data from the GEO database. GCs were collected from 21 controls and 24 PCOS. The candidate miRNAs were verified by qRT-PCR. The correlation was investigated between candidate miRNAs and clinical characteristics in participants. Diagnostic value of candidate miRNAs was analyzed by receiver operating characteristic (ROC) curve.

RESULTS:

Seven miRNAs were differentially expressed in PCOS compared with controls. Furthermore, the validation results demonstrated that hsa-miR-3188 and hsa-miR-3135b showed higher levels in GCs with PCOS patients ( $p<$ 0.05). In addition, the expressions of hsa-miR-3188 and hsa-miR-3135b were negative correlated with FSH and hsa-miR-3188 was positive correlated with BMI ( $p<$ 0.05). ROC analysis indicated that hsa-miR-3188 and hsa-miR-3135b could differentiate PCOS from controls, and the hsa-miR-3188/3135b improved the predictive accuracy for PCOS.

CONCLUSIONS:

The expressions of hsa-miR-3188 and hsa-miR-3135b in human GCs were significantly associated with PCOS. Moreover, the hsa-miR-3188/3135b has certain diagnostic value for distinguishing PCOS.

Keywords

Granulosa cells hsa-miR-3188 hsa-miR-3135b polycystic ovary syndrome

1. Introduction

Polycystic Ovary Syndrome (PCOS) is a common reproductive endocrine disease and a major cause of anovulatory infertility in reproductive-aged women [1]. PCOS usually begins during adolescence, and which is characterized by hyperandrogenism, oligomenorrhoea, and/or polycystic ovaries. As a major public health issue, PCOS is not only one of the main causes of infertility in women, but it is also associated with long-term health risks, including obesity, metabolic syndrome, type 2 diabetes and cardiovascular disease [2]. Due to its complexity and heterogeneity, the exact etiology and pathogenesis of PCOS are still unclear. Researches showed that it may be related to heredity, environment and internal embryonic factors [3].

MicroRNAs (miRNAs) are small non-coding RNAs, which regulate gene expression by mRNA degradation or translational repression [4]. miRNAs are involved in many cellular processes, including cell growth, differentiation, apoptosis, immune reaction, and tumorigenesis [5, 6]. They can regulate up to 30% of human genes and participate in the occurrence and development in diseases by regulating the expression of target genes at the posttranscriptional level [7]. miRNA is also expressed in reproductive tissues including uterus, fallopian tube, ovary and so on, and takes part in the regulation of various physiological and pathological pathways, such as follicular development, fertilization, implantation and early embryo development [8]. Recently, some researches indicated that miRNAs were involved in the occurrence and development of PCOS. Some research data indicated that there were significant differences of miRNAs in the whole blood, follicular fluid and GCs between PCOS patients and normal population [9, 10]. In PCOS patients, differential expressed miRNAs participate in some signaling pathways including amino acid metabolism, cell differentiation, hormone regulation and so on [11, 12]. Thus, it would provide a new idea and direction that abnormal expression of miRNAs was used to investigate the pathogenesis of PCOS.

With the rapid development of genome sequencing and bioinformatics, it is helpful to comprehensively understand the expression profile of miRNAs in the GCs of PCOS. However, Liu et al. found that there was no overlap of the differentially expressed miRNAs at all compared with that of the previous results in their laboratory [13, 14]. This result indicated that PCOS is a multi-factorial and heterogeneous disease. Hence, a larger scale of patients will aid in helping to further understand the underlying molecular mechanism of PCOS by identifying differentially expressed miRNAs. Considering that miRNA plays a critical role in PCOS, we conducted the present study to analyze the expression profiles of miRNAs in PCOS patients and evaluate their potential values for PCOS diagnosis.

2. Methods

2.1 Microarray analysis

The miRNA expression microarray data for PCOS were obtained from the Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo/), and found GSE84376 (Affymetrix Multispecies miRNA-3 Array platform). In GSE84376, it was included 15 PCOS samples and 13 control samples. Differentially expressed miRNAs were analyzed using the online software GCBI (https://www. gcbi.com.cn/gclib/html/index). In this study, differentially expressed miRNAs were screened and selected by the thresholds of Q value $<$ 0.05, $p$ value $<$ 0.05 and fold change $>$ 1.2.

2.2 Study subjects

In this study, 21 controls and 24 patients with PCOS were included, who were recruited at the Children’s Hospital of Shanxi and Women Health Center of Shanxi. The diagnosis of PCOS was based on the 2003 Rotterdam criteria [15]. Control subjects were selected for in vitro fertilization (IVF) due to male factor infertility. All subjects underwent their first cycle of IVF and did not take any medication interfering with sex hormone secretion, lipid and glucose metabolism at least three months before joining the study. In addition, participants with a history and evidence of endocrine diseases, premature ovarian failure and endometriosis were excluded. This study was approved by the Children’s Hospital of Shanxi and Women Health Center of Shanxi Ethical Committee and all participants provided informed consent.

2.3 Purification of GCs

GCs were isolated from the follicular fluid aspirate of patients who had been admitted for IVF. Follicular fluid was centrifuged at 2000 g for 5 min, and the supernatant was removed. The pellet was washed by 1 $\times$ phosphate buffer saline (PBS) and supernatant was removed. The pellet was resuspended in 1 $\times$ PBS and pipetted onto a 50% Percoll gradient. The three layers could be distinguished by centrifuged at 5000 g for 20 min, then the middle ring-like layer was collected. The pellet was resuspended with the same volume 1 $\times$ PBS and red blood cells were removed using red blood cell lysing buffer (RLB) according to the volume 1:3. It was centrifuged at 2000 g for 5 minute after stand for 10 minutes and washed three times with 1 $\times$ PBS. Collected GCs were centrifuged at 2000 g for 5 min, supernatant was discarded, the pellet was snap-frozen in liquid nitrogen and stored at $-$ 80 ${}^{\circ}$ C.

2.4 Validation of candidate miRNAs by quantitative real-time PCR (qRT-PCR)

Total RNA of purified cells was extracted using miRNeasy Mini Kit (Qiagen, Germany; Cat. No. 217004) according to the manufacturer’s protocol. Total RNA was reverse transcribed using a miScript II RT Kit (Qiagen, Germany; Cat. No. 218161). All samples were amplified in triplicate using a miScript SYBR Green PCR Kit (Qiagen, Germany; Cat. No. 218073). U6 small nuclear RNA (U6-snRNA) was used as a miRNA internal control. miRNAs primer and U6 were purchased from iGeneBio Co. (Guangzhou, China).

2.5 Statistical analysis

All data were presented as mean $\pm$ standard error of mean (SEM). $p$ value $<$ 0.05 was considered statistically significant. SPSS for Windows version 16.0 (SPSS Inc., Chicago, IL, USA) was used for all statistical analyses. Student’s $t$ -test was used to investigate the significant differences between the two groups. Pearson correlation analysis was conducted to assess the correlation between miRNAs and biochemical parameters. Receiver operating characteristic (ROC) curve was performed to estimate for the diagnostic value of candidate miRNAs using GraphPad Prism 6.0 (GraphPad Software, Inc., La Jolla, CA, USA). All statistical charts were also conducted using GraphPad Prism 6.0.

3. Results

3.1 Expression profiles of miRNAs

We obtained publicly available microarray dataset GSE84376 from the GEO database. According to the screening criteria, a total of 7 differently expressed miRNAs were identified in the GCs of PCOS. Through the comparison of 7 candidate miRNAs between two groups, 7 up-regulated miRNAs in PCOS group were identified. No down-regulated miRNAs were detected. The differentially expressed miRNAs in GCs is shown in Fig. 1.

Table 1
Clinical characteristics of participants

Variables	PCOS group		Control group		$p$ value
	( $n=$ 24)		( $n=$ 21)
Age (year)	28.708	$\pm$ 0.802	29.571	$\pm$ 0.994	0.503
Infertility duration (year)	3.333	$\pm$ 0.457	3.619	$\pm$ 0.428	0.651
Menarche age (year)	13.833	$\pm$ 0.223	13.381	$\pm$ 0.176	0.118
BMI (Kg/m ${}^{2}$ )	25.958	$\pm$ 0.836	22.173	$\pm$ 0.878	0.003
Basal hormone levels
LH (mIU/ml)	7.255	$\pm$ 0.726	5.429	$\pm$ 0.759	0.039
FSH (mIU/ml)	6.865	$\pm$ 0.385	8.191	$\pm$ 0.434	0.027
E2 (pg/ml)	81.448	$\pm$ 8.459	76.800	$\pm$ 4.782	0.635
P (ng/ml)	0.710	$\pm$ 0.140	0.425	$\pm$ 0.037	0.070
T (ng/ml)	47.202	$\pm$ 4.702	46.333	$\pm$ 3.946	0.855
LH/FSH	1.127	$\pm$ 0.129	0.661	$\pm$ 0.071	0.004
BMI, Body mass index. FSH, follicle-stimulating hormone; LH, luteinizing hormone;
P, progesterone; E2, estradiol; T, testosterone. Data were present by the mean $\pm$ SEM.
$p<$ 0.05 was considered statistically significant.

Figure 1.

A list of differentially expressed miRNAs in GCs. (A) Hierarchical clusters of aberrantly expressed miRNAs between PCOS and Control group. (B) Volcano plots of microRNA (miRNA) profiles in a comprehensive analysis. (C) List of significant difference miRNAs in GCs from PCOS compared to controls using microarray analysis. PCOS, Polycystic Ovary Syndrome; GCs, Granulosa Cells.

3.2 Demographic and clinical characteristics

In the present study, we enrolled 45 patients including 24 PCOS and 21 controls. The demographic and clinical characteristics of participants are listed in Table 1. There were no differences in age, infertility duration, menarche age, E2, P, T between PCOS and control groups. Concentration of LH and LH/FSH were significantly increased in PCOS subjected to controls, while concentration of FSH was significantly decreased ( $p<$ 0.05).

Figure 2.

Validation of 6 miRNAs in ovarian GCs between two groups by qRT-PCR. GCs, Granulosa Cells; qRT-PCR, quantitative real-time reverse transcription-polymerase chain reaction. ${}^{**}p<$ 0.01.

3.3 Validation of miRNAs in ovarian GCs between PCOS and controls

In order to verify the expression levels of miRNAs in the ovarian GCs between PCOS and control groups, we performed the qRT-PCR assay (Fig. 2). The results showed that the expression levels of hsa-miR-3135b and hsa-miR-3188 in GCs of PCOS were significantly upregulated compared with controls ( $p=$ 0.0021, and $p<$ 0.001, respectively). However, no significant differences were observed in the levels of hsa-miR-1225-5p, hsa-miR-1587, hsa-miR-4433-3p, hsa-miR-4749-5p ( $p>$ 0.05). In addition, there was no hsa-miR-4417 expression detected in GCs. Therefore, we selected hsa-miR-3135b and hsa-miR-3188 as candidate miRNAs for further study.

3.4 Relationship of candidate miRNAs with demographic and clinical parameters

In this study, we further analyzed the associations of hsa-miR-3135b and hsa-miR-3188 levels with demographic and clinical parameters (Table 2). Interestingly, we found that both hsa-miR-3188 and hsa-miR-3135b in GCs were significantly negative correlated with FSH ( $r=-$ 0.297, $p=$ 0.047; $r=-$ 0.355, $p=$ 0.017, respectively). In addition, we found that hsa-miR-3188 was significantly positive correlated with BMI ( $r=$ 0.333, $p=$ 0.025).

Table 2
Pearson correlation analysis of the correlation between miRNAs and biochemical parameters of patients

BMI, body mass index.
Variables	hsa-miR-3188		hsa-miR-3135b
	$r$	$p$ value	$r$	$p$ value
BMI	0.333	0.025	0.142	0.363
LH	0.074	0.627	0.110	0.474
FSH	$-$ 0.297	0.047	$-$ 0.355	0.017
LH/FSH	0.220	0.146	0.195	0.200

Figure 3.

ROC curve analysis to evaluate the diagnostic value of hsa-miR-3188 and hsa-miR-3135b in GCs of PCOS. PCOS, Polycystic Ovary Syndrome; GCs, Granulosa Cells; ROC, receiver operating characteristic; AUC, area under curve.

3.5 Diagnostic value of candidate miRNAs

To evaluate the diagnostic value of candidate miRNAs in GCs for discrimination of PCOS patients from control subjects, ROC curves were performed (Fig. 3). ROC analysis revealed that as a single marker, the diagnostic accuracy of hsa-miR-3188 alone of 0.902, sensitivity of 83.3%, and specificity of 85.7% and the diagnostic accuracy of hsa-miR-3135b alone was AUC of 0.760, sensitivity of 62.5%, and specificity of 85.7%. Furthermore, the combination of hsa-miR-3188 and hsa-miR-3135b was found to be a better predictor of PCOS, with AUC of 0.911, sensitivity of 95.8%, and specificity of 76.2%. Therefore, it suggested that the hsa-miR-3188 combined with hsa-miR-3135b should be used as a potential marker for distinguishing PCOS patients from health controls.

4. Discussion

PCOS is one of the endocrine disorders affecting 5% $\sim$ 10% of reproductive-aged women [16]. The pathogenesis of PCOS remains unclear because of its complication so far. Some studies have reported that the abnormal expression of miRNAs was related with PCOS [12, 17]. In order to provide new insights into the pathogenesis and etiology of PCOS, miRNAs have received more attention in recent studies. Ovulation obstacle is one of important clinical features of PCOS. GCs regulate the growth and maturation of oocytes by the gap junction, the dysfunction of GCs disturbed follicular development and maturation in PCOS [18]. Some researchers suggested that the apoptosis of GCs is closely related to follicle atresia [19, 20]. Dicer is the ribonuclease III for synthesis of mature functional miRNAs, was expressed in both oocyte and GCs of follicle. Lei et al. found that conditional inactivation of Dicer1 in the follicular GCs led to an increased primordial follicle, accelerated early follicle recruitment and more degenerate follicles [21]. It followed that the deletion of miRNA in GCs would lead to abnormal oogenesis and follicle growth obstacle.

In the present study, we analyzed the expression profiles of miRNAs retrieved from the GEO database by the online tool GCBI, and a total of 7 miRNAs were differentially expressed in GCs of PCOS. Furthermore, these differentially expressed miRNAs were verified by qRT-PCR and found that the expression of hsa-miR-3135b and hsa-miR-3188 in GCs of PCOS was significantly upregulated. Hsa-miR-3135b and hsa-miR-3188 were two newly identified miRNA. Recent reports about hsa-miR-3135b have shown that it plays an important role in age-related macular degeneration [22], chronic heart failure [23], coronary artery calcification [24]. In addition, some studies suggested that hsa-miR-3188 can regulate nasopharyngeal carcinoma proliferation and chemosensitivity [25], breast cancer proliferation, apoptosis, and migration [26]. However, the roles and molecular mechanisms of newly identified hsa-miR-3135b and hsa-miR-3188 in PCOS have not been reported. Therefore, we speculated that differently expressed of hsa-miR-3135b and hsa-miR-3188 may play a pivotal role in the pathogenesis and progression of PCOS.

The level of LH increased prevalence in PCOS patients. A high level of LH suppress FSH production, so that prematurely luteinized GCs small and antral follicles stasis, and the level of testosterone is high, that could lead to polycystic ovary [27]. In this study, the demographic and clinical characteristics of PCOS were analyzed. The analysis results showed that compared with the control groups, the levels of basic endocrine (including LH, LH/FSH and T) of PCOS group had increased, and FSH expression decreased. Further correlation analysis showed that hsa-miR-3135b and hsa-miR-3188 were negatively correlated with FSH. The results suggested that hsa-miR-3135b and hsa-miR-3188 may be associated with endocrine disorder of PCOS. However, their physiological functions and mechanism with PCOS should be further studied in the future.

With the in-depth study of miRNAs, more and more kinds of miRNAs have been applied in the diagnosed and identified of diseases as a biomarker. In PCOS patients, there are also many reports of related miRNAs. Naji et al. evaluated the diagnostic value of differentially expressed miRNAs in follicular fluid (FF) for discrimination of PCOS patients from control subjects, and found that miR-21 had the largest diagnostic value for discriminating PCOS [10]. Sathyapalan et al. suggested that plasma level of miR-93 was a more efficient biomarker than miR-223 for diagnosis of PCOS [28]. Scalici et al. reported that the combination of FF miR-30a, miR-140 and let-7b expression levels can discriminated between PCOS and normal ovarian reserve [29]. Therefore, specific miRNA or combination of miRNAs is attractive as biomarkers for disease diagnosis and monitoring. In this study, we used the ROC curve to evaluate the diagnostic values of hsa-miR-3188 and hsa-miR-3135b in GCs to PCOS. Our results indicated that hsa-miR-3188 alone and hsa-miR-3135b alone have the ability to diagnose PCOS, and diagnostic accuracy could be improved by the combination of hsa-miR-3188 and hsa-miR-3135b.

In conclusion, our present study identified that hsa-miR-3188 and hsa-miR-3135b level were significantly increased in PCOS patients. Moreover, the combination of hsa-miR-3188 and hsa-miR-3135b may be a promising marker for the diagnosis of PCOS. However, more studies focusing on the miRNA-associated pathogenesis of PCOS are required in the future to further understand the complex relationships between PCOS and candidate miRNAs.

Footnotes

Acknowledgments

This work was supported by the project sponsored by the Research Fund of National Key Research and Development Program (grant no. 2018YFC1002103) and Research Project Supported by Shanxi Scholarship Council of China (grant no. 2012-100).

Conflict of interest

The authors declare no conflicts of interest in this work.

References

Costello

Ledger

. Evidence-based lifestyle and pharmacological management of infertility in women with polycystic ovary syndrome. Women’s Health. 2012; 8(3): 277-90.

Orio

Muscogiuri

Nese

Palomba

Savastano

Tafuri

, et al. Obesity, type 2 diabetes mellitus and cardiovascular disease risk: an uptodate in the management of polycystic ovary syndrome. European Journal of Obstetrics, Gynecology, and Reproductive Biology. 2016; 207: 214-9.

Jin

Xie

. Treatment strategies for women with polycystic ovary syndrome. Gynecological Endocrinology: The Official Journal of the International Society of Gynecological Endocrinology. 2018; 34(4): 272-7.

Piva

Spandidos

Gambari

. From microRNA functions to microRNA therapeutics: novel targets and novel drugs in breast cancer research and treatment (Review). International Journal of Oncology. 2013; 43(4): 985-94.

Friedman

Farh

Burge

Bartel

. Most mammalian mRNAs are conserved targets of microRNAs. Genome Research. 2009; 19(1): 92-105.

Lendvai

Kiss

Kovalszky

Schaff

. Alterations in microRNA expression patterns in liver diseases. Orvosi Hetilap. 2010; 151(45): 1843-53.

Munker

Calin

. MicroRNA profiling in cancer. Clinical Science. 2011; 121(4): 141-58.

Carletti

Christenson

. MicroRNA in the ovary and female reproductive tract. Journal of Animal Science. 2009; 87(14 Suppl): E29-38.

Murri

Insenser

Fernandez-Duran

San-Millan

Escobar-Morreale

. Effects of polycystic ovary syndrome (PCOS), sex hormones, and obesity on circulating miRNA-21, miRNA-27b, miRNA-103, and miRNA-155 expression. The Journal of Clinical Endocrinology and Metabolism. 2013; 98(11): E1835-44.

10.

Naji

Aleyasin

Nekoonam

Arefian

Mahdian

Amidi

. Differential expression of miR-93 and miR-21 in granulosa cells and follicular fluid of polycystic ovary syndrome associating with different phenotypes. Scientific Reports. 2017; 7(1): 14671.

11.

Imbar

Eisenberg

. Regulatory role of microRNAs in ovarian function. Fertility and Sterility. 2014; 101(6): 1524-30.

12.

Sorensen

Wissing

Salo

Englund

Dalgaard

. MicroRNAs Related to Polycystic Ovary Syndrome (PCOS). Genes. 2014; 5(3): 684-708.

13.

Liu

Zhang

Shi

Lin

Chen

, et al. Altered microRNAs expression profiling in cumulus cells from patients with polycystic ovary syndrome. Journal of Translational Medicine. 2015; 13: 238.

14.

Zhang

Tong

Liu

. Characterization of microRNA profile in human cumulus granulosa cells: Identification of microRNAs that regulate Notch signaling and are associated with PCOS. Molecular and Cellular Endocrinology. 2015; 404: 26-36.

15.

group REA-SPcw. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS). Human Reproduction. 2004; 19(1): 41-7.

16.

Orio

Vuolo

Palomba

Lombardi

Colao

. Metabolic and cardiovascular consequences of polycystic ovary syndrome. Minerva Ginecologica. 2008; 60(1): 39-51.

17.

Sorensen

Udesen

Wissing

Englund

ALM

Dalgaard

. MicroRNAs related to androgen metabolism and polycystic ovary syndrome. Chemico-Biological Interactions. 2016; 259(Pt A): 8-16.

18.

Hasegawa

Yanaihara

Iwasaki

Mitsukawa

Negishi

Okai

. Reduction of connexin 43 in human cumulus cells yields good embryo competence during ICSI. Journal of Assisted Reproduction and Genetics. 2007; 24(10): 463-6.

19.

Cataldo

Dumesic

Goldsmith

Jaffe

. Immunolocalization of Fas and Fas ligand in the ovaries of women with polycystic ovary syndrome: relationship to apoptosis. Human Reproduction. 2000; 15(9): 1889-97.

20.

Tilly

. Apoptosis and ovarian function. Reviews of Reproduction. 1996; 1(3): 162-72.

21.

Lei

Jin

Gonzalez

Behringer

Woodruff

. The regulatory role of Dicer in folliculogenesis in mice. Molecular and Cellular Endocrinology. 2010; 315(1-2): 63-73.

22.

Ghanbari

Erkeland

Colijn

Franco

Dehghan

, et al. Genetic variants in microRNAs and their binding sites within gene 3’UTRs associate with susceptibility to age-related macular degeneration. Human Mutation. 2017; 38(7): 827-38.

23.

Wang

Kwee

Grass

Neely

Gregory

Fox

KAA

, et al. Whole blood sequencing reveals circulating microRNA associations with high-risk traits in non-ST-segment elevation acute coronary syndrome. Atherosclerosis. 2017; 261: 19-25.

24.

Liu

Ling

Sun

Liu

Zhong

, et al. Circulating microRNAs correlated with the level of coronary artery calcification in symptomatic patients. Scientific Reports. 2015; 5: 16099.

25.

Zhao

Luo

Liu

Gao

, et al. miR-3188 regulates nasopharyngeal carcinoma proliferation and chemosensitivity through a FOXO1-modulated positive feedback loop with mTOR-p-PI3K/AKT-c-JUN. Nature Communications. 2016; 7: 11309.

26.

Chen

. MiR-3188 Regulates Cell Proliferation, Apoptosis, and Migration in Breast Cancer By Targeting TUSC5 and Regulating the p38 MAPK Signaling Pathway. Oncology Research. 2017.

27.

Chang

Cook-Andersen

. Disordered follicle development. Molecular and Cellular Endocrinology. 2013; 373(1-2): 51-60.

28.

Sathyapalan

David

Gooderham

Atkin

. Increased expression of circulating miRNA-93 in women with polycystic ovary syndrome may represent a novel, non-invasive biomarker for diagnosis. Scientific Reports. 2015; 5: 16890.

29.

Scalici

Traver

Mullet

Molinari

Ferrieres

Brunet

, et al. Circulating microRNAs in follicular fluid, powerful tools to explore in vitro fertilization process. Scientific Reports. 2016; 6: 24976.