Melatonin suppresses serum starvation-induced autophagy of ovarian granulosa cells in premature ovarian insufficiency

Abstract

Objectives: Premature ovarian insufficiency (POI) refers to the decline and cessation of ovarian functions in women under 40 years of age. Melatonin (MT) acts as a protective for the ovary. This study elucidated the role of MT in autophagy of granulosa cells (GCs) in POI via modulating the phosphatidylinositol-3-kinase (PI3K)-Akt-mammalian target of rapamycin (mTOR) pathway.

Methods: The expression levels of microRNA (miR)-15a-5p, signal transducer and activator of transcription 3 (Stat3), and relevant hormones in the clinically collected serum samples of POI patients and healthy controls were examined. Human ovarian granulosa-like tumor cells (KGN) underwent serum starvation (SS) treatment to induce POI cell models and then received MT treatment. The expression levels of miR-15a-5p, Stat3, p-PI3K/PI3K, p-Akt/Akt, and p-mTOR/mTOR in KGN cells were tested via quantitative real-time polymerase chain reaction and Western blotting. KGN cell viability was assessed by MTT assay and the protein levels of autophagy-related markers Beclin-1, microtubule-associated protein light chain 3 II/I, and p62 were detected by Western blotting. The binding relation between miR-15a-5p and Stat3 was verified via the dual-luciferase reporter gene assay. Functional rescue experiments were performed to probe the underlying role of miR-15a-5p/Stat3/the PI3K-Akt-mTOR pathway in KGN cell autophagy.

Results: miR-15a-5p was increased whilst Stat3 was decreased in the serum of POI patients and SS-induced KGN cells. MT inhibited miR-15a-5p and Stat3, activated the PI3K-Akt-mTOR pathway, and repressed cell autophagy in SS-induced KGN cells. miR-15a-5p targeted and repressed Stat3 expression. Upregulation of miR-15a-5p or downregulation of Stat3 or the PI3K-Akt-mTOR pathway promoted KGN cell autophagy.

Conclusion: MT suppressed miR-15a-5p and activated Stat3 and the PI3K-Akt-mTOR pathway, finally impeding SS-induced autophagy of GCs.

Keywords: Autophagy; Granulosa cells; Melatonin; PI3K-Akt-mTOR pathway; Premature ovarian insufficiency; Serum starvation; Stat3; miR-15a-5p.

Fig. 1

MT limits SS-induced KGN cell autophagy and downregulates miR-15a-5p. A The expression levels of miR-15a-5p in the serum of the healthy controls (n = 40) and POI patients (n = 32) were examined using qRT-PCR. KGN cells were treated with serum starvation (SS) to induce autophagy and then treated with melatonin (MT). B KGN cell viability was detected via MTT assay; C The expression levels of autophagy-related proteins Beclin 1, LC3 II/I, and p62 in KGN cells were examined via Western blotting; D The expression levels of miR-15a-5p in KGN cells were examined via qRT-PCR. Experiments were performed in 3 independent repetitions; experimental data were presented as mean ± SD; data in figure A were examined via the t-teat; data in figure B were examined via two-way ANOVA; data in figures C-D were examined via one-way ANOVA, followed by Tukey’s post-hoc test; ** p < 0.01

Fig. 2

miR-15a-5p upregulation boosts SS-induced KGN cell autophagy. KGN cells were transfected with plasmid miR-15a-5p-mimic (miR-mimic), with mimic-NC as a control. A The transfection efficiency was examined via qRT-PCR. KGN cells with miR-15a-5p-mimic were combined with MT treatment. B KGN cell viability was detected via MTT assay; C The expression levels of autophagy-related proteins Beclin 1, LC3 II/I, and p62 in KGN cells were examined via Western blotting. Experiments were performed in 3 independent repetitions; experimental data were presented as mean ± SD; data in figure A were examined via Student’s t-test and data in figures B-C were examined via one-way ANOVA, followed by Tukey’s post-hoc test; * p < 0.05 and ** p < 0.01

Fig. 3

miR-15a-5p binds to Stat3. A The binding association between miR-15a-5p and Stat3 was verified via dual-luciferase reporter gene assay; B-C Stat3 expression levels in KGN cells were examined via qRT-PCR and Western blotting; D The expression levels of Stat3 in the serum of the healthy controls (n = 40) and POI patients (n = 32) were examined using qRT-PCR; E The correlation between serum miR-15a-5p expression and serum Stat3 expression was analyzed by Pearson correlation analysis. Experiments were performed in 3 independent repetitions; experimental data were presented as mean ± SD; data in figure A were examined via two-way ANOVA and data in figures B-C were examined via one-way ANOVA, followed by Tukey’s post-hoc test; ** p < 0.01

Fig. 4

Silencing Stat3 neutralizes the repression of MT on SS-induced KGN cell autophagy. KGN cells were transfected with si-Stat3-1 and si-Stat3-2, with si-NC as a negative control. A The transfection efficiency was examined via qRT-PCR. si-Stat3-1 with better silencing effect was chosen for combined treatment with MT. B The protein levels of Stat3 in KGN cells were determined via Western blotting; C KGN cell viability was detected via MTT assay; D The expression levels of autophagy-related proteins Beclin 1, LC3 II/I, and p62 in KGN cells were examined via Western blotting. Experiments were performed in 3 independent repetitions; experimental data were presented as mean ± SD; data in figure C were examined via two-way ANOVA and data in figures A-B and D were examined via one-way ANOVA, followed by Tukey’s post-hoc test; ** p < 0.01

Fig. 5

MT activates the PI3K-Akt-mTOR pathway via miR-15a-5p/Stat3. The activation of the PI3K-Akt-mTOR pathway in KGN cells was detected via Western blotting. Experiments were performed in 3 independent repetitions; experimental data were presented as mean ± SD and analyzed via one-way ANOVA, followed by Tukey’s post-hoc test; ** p < 0.01

Fig. 6

Silencing the PI3K-Akt-mTOR pathway promotes SS-induced KGN cell autophagy. KGN cells were treated with PI3K inhibitor (LY294002) and combined with MT treatment. A The activation of the PI3K-Akt-mTOR pathway in KGN cells was detected via Western blotting; B KGN cell viability was detected via MTT assay; C The expression levels of autophagy-related proteins Beclin 1, LC3 II/I, and p62 in KGN cells were examined via Western blotting. Experiments were performed in 3 independent repetitions; experimental data were presented as mean ± SD; data in figure B were examined via two-way ANOVA and data in figures A and C were examined via one-way ANOVA, followed by Tukey’s post-hoc test; ** p < 0.01