Brain-derived neurotrophic factor promotes human granulosa-like tumor cell steroidogenesis and proliferation by activating the FSH receptor-mediated signaling pathway

Brain-derived neurotrophic factor (BDNF) and FSH receptor (FSHR) are expressed in ovarian granulosa cells, and play important roles in regulating follicle growth and oocyte maturation. Studies have linked the BDNF-associated signaling pathway to FSHR mRNA expression in the regulation of follicle development, but the mechanisms remain unknown. In the current study, we found that BDNF stimulated the secretion of estradiol and progesterone, and increased the proliferation of KGN cells (human granulosa-like tumor cell line). BDNF treatment also increased phosphorylated and ubiquitinated FSHR, and activated cAMP/PKA/CREB signaling pathway. Moreover, inhibition of BDNF expression by siRNA markedly reduced the estradiol secretion and down-regulated FSHR, aromatase and phosphorylated CREB; meanwhile, FSH treatment partly alleviated the effects of BDNF siRNA on KGN cells. These findings suggested that BDNF modulates graunlosa cell functions and the action probably mediated by FSHR-coupled signaling pathway, to affect aromatase-mediated steroidogenesis. These results provide an alternative target to optimize ovarian granulosa cell function.


KGN cells secrete BDNF and the secretion is enhanced by FSH treatment.
In the current study, we first determined BDNF production in KGN cells by ELISA. BDNF was detected both in lysates (349.3 ± 13.9 pg/ ml) and cell culture supernatants (63.2 ± 9.2 pg/ml), suggesting that BDNF was produced and secreted by KGN cells (Fig. 1). Previous research showed that gonadotrophin increased BDNF transcript level of non-stimulated granulosa cells 22 . KGN cells were treated with FSH, and increased BDNF protein level was found in lysates (427.4 ± 18.9 pg/ml) and cell culture supernatants (102.8 ± 11.9 pg/ml) ( Fig. 1), indicating that BDNF secretion was stimulated by gonadotrophin. These results demonstrated that KGN cells have common characteristics of normal human granulosa cells, i.e. production and secretion of BDNF.

Steroidogenesis is promoted by BDNF.
Steroidogenesis is one of the major physiological functions of granulosa cells. We next assessed the effects of BDNF treatment on steroidogenesis in KGN cells. The results showed that compared with control group (196 ± 22.5 pM), treatment with FSH (225 ± 15.7 pM) or BDNF (230 ± 4.5 pM) alone had no notable effects on estradiol level, while combined treatment with BDNF and FSH (274 ± 26.9 pM) markedly increased estradiol secretion ( Fig. 2A). For progesterone production, compared with control group (0.36 ± 0.03 nM), BDNF alone (0.66 ± 0.07 nM) did not enhance the secretion in granulosa cells, while FSH (2.80 ± 0.57 nM) and BDNF plus FSH (4.36 ± 0.59 nM) both promoted secretion, with higher level obtained for the latter treatment (Fig. 2B). These results suggested that combination of FSH and BDNF significantly elevated hormone secretion in KGN cells.

Discussion
It is well known that ovarian BDNF functions as a regulator of follicular development and oocyte maturation 25,26 . In the current study, we demonstrated that KGN cells produced BDNF, and FSH treatment increased this production. Furthermore, BDNF treatment and BDNF knockdown were used to determine how BDNF affects the expression and modifications of FSHR, as well as the activities of FSHR-coupled cAMP/PKA/CREB signaling pathway. Through these explorations, the molecular mechanisms underlying the effects of BNDF on hormone secretion were elaborated.
BDNF is critical to the regulation of cell function and survival. The BDNF protein is first synthesized as a precursor protein, proBDNF, which is then cleaved to generate mature BDNF (mBDNF); only mBDNF is considered to be biologically active. There are probably two secretory pathways of proBDNF 27 . It was proposed that proBDNF processing into mBDNF occurs extracellularly, and is dependent on the extracellular proteases, as found in cultured hippocampal neurons 28 . Yang et al. also indicated that proBDNF is secreted 29 . An alternative pattern was proposed that proBDNF produced by neurons is rapidly converted to mBDNF intracellularly for storage, and released by excitatory input 30 . According to the above studies, whether proBDNF cleavage occurrs intracellularly or extracellularly, mBDNF has extracellular biological functions. Therefore, in the current study, BDNF was added in the medium was to increase extracellular BDNF directly, while BDNF knockdown by siRNA aimed to decrease BDNF secretion. Estradiol and progesterone are major hormones that regulate ovarian growth and physiological function [31][32][33][34] . In mammals, circulating estrogens are mainly produced by the granulosa cells, and required for the development of secondary sex characters in females 35 . Estradiol mainly enhances the response of granulosa cells to gonadotropins 36 and plays essential roles in follicular development 37 . Progesterone secretion is critical for ovulation, as well as establishment and maintenance of pregnancy. According to the classic "two-cell, two-gonadotropin" model, estradiol is transformed from androgens which is synthesized from progesterone 38 . Our data showed a dramatically increased estradiol and progesterone after combined treatment with BDNF and FSH (Fig. 2), suggesting synergistic effects of BDNF and FSH in inducing hormone secretion in KGN cells. Accordingly, increased estradiol and progesterone levels induced by BDNF and FSH (Fig. 2) provided alternative options for promoting steroidogenesis in granulosa cells.
Besides promoting steroidogenesis, BDNF also increased KGN cell proliferation. We found a statistically significant increase in EdU-positive cells upon BDNF and BDNF plus FSH treatments (Fig. 3), which indicated the inductive effect of BDNF on granulosa cells proliferation. Similar biological effect was observed in the central nervous system. BDNF was shown to stimulate the differentiation of neural progenitor cells, regulating neurogenesis 39 . Moreover, BDNF administration or its increased expression promotes hippocampal neurogenesis and increases neuronal activity 40,41 . Therefore, BDNF not only regulates the neurons, but also promotes growth and physiological functions (steroidogenesis) in granulosa cells.
To elicit ovarian steroidogenesis, functional FSHR is required 16 . Our findings indicated FSHR protein level decreased by combined treatment (Fig. 4). FSHR activity is regulated by multiple patterns, including mutations and post-translational modifications 20 . FSHR mutation in female animals induces ovarian dysfunction, including smaller ovaries, and no corpora lutea and mature Graafian follicles, suggesting that FSHR mutation could block follicular maturation 42 . In addition, post-translational modifications of the FSHR protein, such as phosphorylation, have pivotal roles in a variety of processes 43 . Receptor phosphorylation commonly occurs on serine and threonine, and is linked to FSHR desensitization and internalization 14 . Desensitization is an important physiological feedback mechanism that protects against receptor overstimulation. GPCR desensitization also plays essential roles in integrating biological signals through second messenger protein kinase-dependent phosphorylation 44 . Desensitization and internalization are prominent to the GPCR in the regulating downstream signaling events and recycling back to the cell surface 44,45 . It is thought that after internalization, one of the optional locations of the receptor is the lysosome 45 . However, another degradation pathway has been proposed, namely the ubiquitin-proteasome pathway (UPP). Previous findings demonstrated that FSHR can be ubiquitinated and degraded by the proteasome, since treatment with the proteasome inhibitor MG132 increases FSHR level 46 . Accordingly, decreased FSHR protein level, and increased phosphorylation and ubiquitination of FSHR induced by BDNF plus FSH treatment (Figs 4 to 7) suggested that BDNF probably facilitates FSHR internalization and degradation of (might be mainly through UPP), by inducing its phosphorylation and ubiquitination.
PTMs act as molecular switches and initiate protein-protein interactions. Protein degradation can be controlled by PTM crosstalk, e.g. phosphodegrons in ubiquitin-mediated protein degradation 43 . FSHR Figure 8. Effects of BDNF on cAMP level, PKA and CREB activities, and CREB mRNA expression. KGN cells were treated with FSH (100 ng/ml), BDNF (5 ng/ml) and FSH (100 ng/ml) plus BDNF (5 ng/ml), respectively, for 24 h. (A and B) Cell extracts were subjected to ELISA for the determination of cAMP production and PKA kinase activity in treated and untreated cells. (C) Western blot analysis of protein extracts from treated and untreated cell showing the levels of pCREB (Ser133) and total CREB. (D) Relative CREB activity was derived as pCREB normalized to total CREB. (E) Total cellular RNA was extracted, and CREB mRNA levels were determined by real-time PCR. Relative gene expression levels were normalized to β-actin, and expressed as fold change. *P < 0.05, **P < 0.01 and ***P < 0.001 versus control group. phosphorylation and ubiquitination have been reported, but the association of the two modifications remains largely unclear. Our findings indicated that BDNF affects FSHR expression, and promotes its phosphorylation and ubiquitination. Phosphorylation can serve as a signal for recognition by a binding protein (ubiquitin) that carries out ubiquitination 47,48 . There are many parallels between phosphorylation and ubiquitination. Phosphoproteomic analysis indicates that the majority of proteins in a cell can be phosphorylated, and the number of proteins known to be subject to ubiquitination is steadily increasing 48 . Therefore, increased phosphorylation of FSHR may elicit ubiquitination.
As shown above, increased activity of cAMP/PKA/CREB signaling pathway was induced by BDNF or BDNF plus FSH (Fig. 8). However, FSH alone did not induce the increase cAMP level and PKA activity. CREB phosphorylation at Ser133 is regulated by PKA and other kinases 49 . Nevertheless, CREB is directly phosphorylated by PKA in granulosa cells, not by other kinases 17 . The transcription factor CREB is crucial for stimulus-transcription coupling, which converts the events occuring at cell membrane into an intercellular gene expression; in turn, this ultimately affects the function of individual cells by regulating the expression of certain proteins 49 . CREB regulates genes expression by binding to the cAMP-response element (CRE), a specialized stretch of DNA that within the regulatory region of numerous genes 23 . A point mutation in the DNA-binding domain prevents CREB from binding to CRE, and makes it unable to bind to DNA-binding sites 23 . Thus, the transcriptional activity relies on the combination of CREB and CRE. Analysis of CRE-mediated transcription in KGN cells demonstrated that CRE is essential for aromatase activity 50 .
To further assess the effects of BDNF on hormones biosynthesis and related mechanisms in granulosa cells, BDNF siRNA was used to knockdown BDNF, and down-regulation of estradiol, FSHR, aromatase, and phosphorylated CREB were observed (Fig. 9). Aromatase is an enzyme responsible for a key step in estrogens biosynthesis. FSHR mediates the activation of the cAMP/PKA signaling pathways and subsequently stimulates aromatase expression 51, 52 . Therefore, by triggering cAMP dependent signaling cascades, CYP19A1 transcription is upregulate, which consequently induces estrogen biosynthesis 52 . In addition, CYP19A1 transcription is promoted by CREB 53 . Accordingly, the increased estradiol secretion induced by treatment with BDNF plus FSH (Fig. 2) could be a stimulatory factor for cAMP/PKA/CREB signaling pathway, which mediated by FSHR.
In conclusion, this study demonstrated that KGN cells produced BDNF, and treatment with BDNF plus FSH induced the steroidogenesis and KGN cell proliferation. In addition, FSHR modifications and the FSHR-coupled signaling pathway were also regulated by BDNF and BDNF plus FSH. Moreover, the above effects of BDNF could be inhibited by BDNF siRNA. These findings concerning in KGN cells may be applied for a further understanding of the functional regulation of granulosa cells. Overall, this study has established a molecular link between BDNF and FSHR function, partly revealing the mechanisms by which neurotrophic factors modulate hormone secretionin granulosa cells.
KGN cell culture and treatment. KGN cells were seeded in culture dishes and cultured in the DMEM/F12 medium supplemented with 10% fetal bovine serum (Gibco, Thermo Fisher Scientific, USA), penicillin (100 U/ ml) and streptomycin (100 μg/ml) in a humidified incubator containing 5% CO 2 at 37 °C. Four cell groups were set up: control (PBS); FSH (FSH, 100 ng/ml); BDNF (BDNF, 5 ng/ml); FSH and BDNF (FSH at 100 ng/ml and BDNF at 5 ng/ml) treatment. After 24 h of incubation, the culture medium and whole-cell extracts were obtained for biochemical analyses.
Hormone Assays. After treatment, the culture medium from control and treated cells was collected and centrifuged at 1000 g for 3 min, and supernatants were moved to fresh tubes for the assay. The concentrations of estradiol and progesterone were determined on a chemiluminescent immunoassay system (Backman Coulter, Inc., Brea CA, USA). The intra-assay and inter-assay coefficients of variation for estradiol and progesterone were 5.13% and 6.23%, and 8.18% and 7.89% respectively.
EdU assay. The effect of BDNF on cell proliferation was assessed with the Cell-Light EdU DNA cell proliferation kit (RiboBio, Guangzhou, CHN), according to the manufacturer's instructions. Briefly, KGN cells (4 × 10 4 cells per well) were seeded in 96-well plates. Then, the cells were incubated with 50 μM of EdU for 2 h at 37 °C. After two washes with PBS, the cells were fixed with 4% formaldehyde for 30 min at room temperature. Next, the cells were treated with 0.5% Triton X-100 for 10 min at room temperature and incubated with 1 × Apollo reaction cocktail (100 μl/well) for 30 min. After two washes with 0.5% Triton X-100 for 10 min at room temperature, DNA was stained with Hoechst 33342 (100 μl/well) for 30 min and visualized by fluorescence microscopy. EdU-positive cells were obtained from fluorescent images; confluent cells were randomly selected from each analysis. Cells were counted with ImageJ, and the relative positive ratios were averaged from four values per group.
ELISA for assessing BDNF and cAMP levels, and PKA kinase activity. BDNF and cAMP levels in cell lysates and culture supernatants were determined using commercially available kits, according to the manufacturer's protocols in triplicate. Briefly, BDNF levels in cell lysates were adjusted to the amount of protein in the corresponding culture well (pg BDNF/mg protein). Cell culture supernatants and lysates were added to a plate that pre-coated with human BDNF specific antibody, and incubated at 37 °C for 90 min. This was followed by incubation with biotinylated anti-human BDNF antibody at 37 °C for 60 min. After addition of avidin-biotin-peroxidase-complex for 30 min at 37 °C and color development, absorbance was measured at 450 nm on a multi-mode microplate reader (Synergy HTX, BioTek Instruments, INC., USA). The BDNF assay had a linear over a range of 31.2-2000 pg/ml. For cAMP assay, the primary antibody solution was added for 1 h; then, cAMP conjugate and lysates were added to the assay plate; cAMP levels were measured accoding to the manufacturer's recommendations at 450 nm on a multi-mode microplate reader. The number of moles of cAMP was calculated by comparing with a curve derived from the standards provided with the kit. The mean minimum detectable dose of cAMP was 1.5 pmol/ml.
For PKA kinase activity evaluation, the wells were soaked with kinase assay dilution buffer for 10 min at room temperature. Then, standards and KGN cell lysates were added to the wells with ATP to initiate the reaction. After incubation at 30 °C for 90 min, the wells were emptied to terminate the kinase reaction, and the phosphospecific substrate antibody was added for 60 min at room temperature. HRP conjugated anti-rabbit IgG was added at room temperature for 30 min prior to a wash step. Subsequently, the TMB substrate solution was added at room temperature for 30-60 min. OD of each well was immediately determined at 450 nm.