Whole-exome sequencing identifies GPSM1 as a susceptibility gene for premature ovarian insufficiency


 Background Genetic causes of premature ovarian insufficiency (POI) account for approximately 20~25% of patients. So far, only a few genes have been identified. Results Here, we first identified the c.1840C>A on G-protein signaling modulator 1 (GPSM1) as a susceptibility locus for POI in 10 sporadic POI patients by whole-exome sequencing. The frequency of GPSM1 c.1840C>A was then verified as 3/20 in a POI sample of 20 patients (including the above 10 patients) by Sanger sequencing. RT-PCR and western blot analysis showed the expression of GPSM1 in rat ovaries was increased in the large antral follicle stage compared to the primordial follicle stage (P<0.01). The cell proliferation assay (CCK8) and flow cytometry suggested that the small-interfering RNA-induced silencing of Gpsm1 significantly increased apoptosis and decreased proliferation of rat ovarian granulosa cells (GCs) (P<0.01). Furthermore, suppression of Gpsm1 in GCs reduced levels of cAMP, PKAc, p-CREB as well as the ratio of Bcl-2/Bax, and increased the expression of Caspase-3 and Cleaved Caspase-3 (P<0.01). Conclusions In summary, this study identified a susceptibility variant GPSM1 c.1840C>A of POI for the first time. Gpsm1 was related to rat follicle development, and silencing increased apoptosis and decreased proliferation in rat GCs, possibly through inhibition of the cAMP-PKA-CREB pathway. These findings facilitate the development of the early molecular diagnosis of POI.


Introduction
Premature ovarian insufficiency (POI) is a clinical syndrome characterized by the loss of ovarian activity before the age of 40 years, which results in hypergonadotropic hypogonadism.(1) It affects approximately 1% women under 40 years old (2). The opportunity to preserve fertility in women with POI is small, which has a negative impact on both the psychological and reproductive health of women of childbearing age. As women are marrying at a later age, infertility caused by POI is becoming a greater concern. Early detection of the high-risk population would allow for increased treatment options, and perhaps the POI-related infertility could be avoided. Research into the pathogenesis of POI is particularly important for molecular diagnosis and prevention.
Although most POI cases are idiopathic, genetic factors are considered to be one of the main causes 3 of POI(3), accounting for 20-25%(4). Currently, only several genes have been confirmed to be associated with the pathogenesis of POI, including FOXL2, BMP2, NOBOX, FIGLA, and GDF9 (5). The screening and validation of POI candidate genes is a large task. Currently, most of the research on POI-pathogenesis genes is focused on pedigree studies (6,7). However, POI presents with high genetic heterogeneity; each family or individual patient seems to be unique in the pathogenesis and there is no significant definitive pathogenic gene for POI. If common pathogenic mutations can be sought out in sporadic POI populations, more insight into the genetic changes of POI patients can be gained. In recent years, with the development and maturity of high-throughput sequencing technologies, wholeexome sequencing (WES) has been widely applied to explore new pathogenic genes of hereditary disease, and has been verified as an effective tool for research into genetic etiology (8)(9)(10).
In this study, we used WES to initially identify a POI susceptibility gene in unrelated Chinese women affected by POI. A functional study was then carried out using rat ovarian granulosa cells (GCs). The outcome of this study will provide insight into the potential application of genetic sequencing for molecular diagnosis, allowing for better health management (e.g., early detection and childbearing) for carriers of the deleterious susceptibility gene mutation.

WES and Sanger sequencing identified heterozygous GPSM1 mutation in 3 POI patients
We performed WES on 10 POI patients to. The mean coverage of the target region is 99.8%. More than 98.76% of the target was covered at 20 × depth. We applied the following exome filtration procedure: total variants called in 10 patients and variants that are absent or with minor allele frequency less than 1% in the population public databases (dbSNP, 1000Genomes, and ExAC). After these filters, a total of 12 genes including 22 variants were retained: OR2T29 (c.26A > G), ANKRD36C Details of all variants are listed in Table 1. Of these candidate genes, GPSM1 (G-protein signaling 4 modulator 1) sparked our interest. For GPSM1 (c.1840C > A), the ExonicFunc. refGene, rs number, ExAC, 1000G(all), SIFT score, Ployphen2 score were missense mutation, rs539775258, 0.000008446, 0.0002, 0.001, and 0.998, respectively. Protein alignments revealed the GPSM1 p. Glu614Lys variant, located in the third G-protein regulatory (GPR) motif of the protein, affected an amino acid highly conserved among species (Fig. 1). Sanger sequencing confirmed that GPSM1 (c.1840C > A) were heterozygous in 3/20 patients.

Discussion
From our study of 10 POI patients, we identified a novel POI susceptibility gene, GPSM1. The GPSM1 gene was reported in public expression databases to be expressed in multiple types of tissue and cell, especially in ovary (https://www.ncbi.nlm.nih.gov/gene/26086). It encodes activator of G protein signaling 3 (AGS3), which was identified as an evolutionarily conserved protein (13,14) and to be associated with cell division (15), cell proliferation (16), differentiation (17), autophagy (18) and so on.
AGS3 has a modular domain structure consisting of seven tetratricopeptide repeats (TPRs) and four G-protein regulatory (GPR) motifs. It was a regulatory accessory protein of G-protein signaling which could bind preferentially to inactive Gα/o subunit complexed with guanine dinucleotide phosphate (GDP) at multiple GPR motif repeats (15,19,20 Early studies (21,22) have shown that at 24 h following PMSG treatment, the ovaries of rats presented with multiple follicular growth, and at 48 h, the follicular antrum was extremely large and the parietal granulosa cell layer became very thin. Here, we first explored the expression pattern of GPSM1 at different stages of ovarian growth in rats. The results showed that the expression of GPSM1 was significantly increased in the large antral follicle stage compared to the primordial follicle stage. This result indicated that GPSM1 might play an important role in follicular growth in rats.
The apoptosis of GCs and follicular atresia have been shown to play key roles in the pathogenesis of POI (3,23). The mechanism by which GPSM1 regulates heterotrimeric G-proteins in GCs has yet to be elucidated. GPSM1, encoding AGS3, could compete with free G subunits for binding to Gα/o-GDP subunits, regulating downstream signal transduction pathways by inhibiting the Gi subunit. The Gi subunit directly inhibits adenylyl cyclase (AC) activity (24). That is how GPSM1 conducts signals, by activating AC to promote the synthesis of cAMP. Changes in cAMP levels are a common observation in the growth and maturation of GCs (25,26). Thus, we initially postulated GPSM1 in GCs might play an important role in apoptosis by regulating the activation of cAMP pathways.
To verify this hypothesis, we first demonstrated the silencing of Gpsm1 induced GC apoptosis and inhibited proliferation. Furthermore, it was confirmed that the generation of cAMP was correlated to the expression of GPSM1 in rat GCs. Next, we explored the downstream signaling molecules of cAMP 9 in GCs. There was no decline in PKA levels, but PKAc decreased when GPSM1 was down-regulated, which suggested the subsequent signal transduction mediated by cAMP might not be transmitted by the change of PKA level, but probably by the level of PKAc, which is an active subunit decomposed from PKA. Concomitant with the reduction of PKAc, CREB, a target molecule of PKA, as well as an important mediator of multiple signal transduction pathways in GCs (27,28), was found to be less phosphorylated in Gpsm1-knockdown cells. Bcl-2 is a well-known target of CREB (12), and was also reported as one of the important apoptosis suppressors in GCs (29). Bax-dependent apoptosis is a common pathway of cell death and the balance of Bcl-2 and Bax is a key determinant of the survival or death of GCs. As expected, the level of Bcl-2 and the ratio of Bcl-2/Bax were found to be markedly reduced after down-regulation of GPSM1. Finally, as apoptosis markers, Caspase-3 and cleaved Capase-3 showed an uptrend in Gpsm1-knockdown cells.

Conclusions
In summary, we first identified GPSM1 as a novel susceptibility gene for POI, and found it to be associated with follicular development in rats. Then, we confirmed the anti-apoptotic and proliferative functions of GPSM1 in rat GCs, and found that the possible mechanism of action might be through regulation of the Bcl-2/Bax ratio through cAMP-PKA-CREB signaling, which affects the activation of apoptosis protein, Caspase-3, ultimately determining the survival or death of GCs. While further studies are required to verify the results of this study, our findings shed light on the etiology of POI by providing new candidates and pathways to explore.

Induction of follicle development and isolation of ovaries in rats
To stimulate follicular growth, immature female Sprague-Dawley (SD) Rats (24-25 days) were intraperitoneally injected with pregnant mare serum gonadotropin (PMSG, 40 IU/rat) (Ningbo Second Hormone Plant, Zhejiang, China). The experimental group (PMSG group, n = 3) was treated as above.
The control group (NS group, n = 3) was intraperitoneally injected with an equal volume of normal saline (NS). Rats were sacrificed 48 h later, and the ovaries were removed immediately and cleaned with phosphate-buffered saline (PBS) for subsequent assays. One ovary of each rat (n = 6) was used for RT-PCR, and one for western blotting.

Immunohistochemistry
Ovaries of mature female SD rats (12 weeks) were removed, formalin fixed, paraffin embedded, sectioned (4 µm), deparaffinized, and rehydrated. Endogenous peroxidase activity was blocked by incubation of the sections with 3% H 2

RNA isolation and RT-PCR
Total RNA was isolated from ovarian tissues or cultured GCs using TRizol reagent (TaKaRa, Japan) according to the manufacturer's instructions. The first-strand cDNA for total RNA was synthesized  Flow cytometry-based annexin/ propidium iodide (PI) assay 13 GCs seeded in 6-well plates were transfected with siRNA and cultured for 48 h. Then 1 × 10 6 cells were collected, washed twice with ice-cold PBS, and resuspended in binding buffer containing Annexin V-FITC and PI. After incubating for 30 min in the dark, cells were analyzed using a BD FACSVerser flow cytometer system (BD Biosciences, USA) equipped with BD FACSuite software.

Statistical analysis
All data were analyzed using SPSS software (Version 22.0, SPSS Inc., Chicago, USA), and the results are presented as mean ± SEM using at least 3 independent experiments. An unpaired Student t-test was performed when comparing two groups and one-way ANOVA was performed when comparing more than two groups. A P value of 0.05 or less was considered statistically significant (*P < 0.05, **P < 0.01).

Consent for publication
Written informed consent for publication was obtained from all participant.

Availability of date and materials
The raw data of WES required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study. Single nucleotide variants and indels were annotated using ANNOVAR (http://www.oponbioinformatics.org/annovar/). The variants were analysed using dbSNP

Competing interests
The authors declare that they have no competing interests.

Finding
This study was supported by the Science and Technology Planning Project of Guangdong province,

Authors' contributions
Xuzi Cai and Xuefeng Wang conceived and designed the study. Xuzi Cai and Huijiao Fu performed the experiments. Yan Wang and Qiwen Liu analyzed and interpreted the patient data regarding POI. Xuzi Cai wrote the paper. Xuefeng Wang reviewed and edited the manuscript. All authors read and approved the manuscript.

Figure 2
The expression of GPSM1 in ovaries before and after multi-follicle development A.
Morphological changes of rat ovaries: a. Ovaries from NS-treated immature rats were light pink and normal in size; b. Ovaries from PMSG-primed immature rats were significantly

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