Decrease in ovarian reserve through the inhibition of SIRT1-mediated oxidative phosphorylation

Abstract

Objective: To establish an oxidative stress-induced model of premature ovarian insufficiency (POI) and to explore the effect of SIRT1 and mitochondrial oxidative phosphorylation on the ovarian reserve.

Methods: Mice were treated with intraperitoneal injections of 3-nitropropionic acid (3-NPA) at different doses and for different time periods to induce a model of POI. Subsequently, the efficiency of each regimen was evaluated. The expression of SIRT1 in ovarian tissue was examined. Then, SIRT1 was knocked down in human luteinized granulosa cells (GCs), and its function and related receptor and gene expression were examined. Finally, a SIRT1 antagonist and agonist were used to explore the effects of SIRT1 on ovarian function in vivo and on the change in mitochondrial oxidative phosphorylation complexes (OXPHOS).

Results: Decreases in ovarian reserve were successfully induced through the intraperitoneal injection of 40 mg/kg 3-NPA for 3 weeks, and SIRT1 was down-regulated in the model group. The knockdown of SIRT1 impaired the estrogen synthesis capacity of human GCs and decreased the expression of related genes. 3-NPA and SIRT1 antagonist Ex-527 decreased ovarian function and inhibited OXPHOS. In contrast, the SIRT1 agonist resveratrol promoted the recovery of ovarian function in the model group and improved OXPHOS. Additionally, P53, CASPASE 3, and BAX were down-regulated and BCL2 was up-regulated in the 3-NPA and Ex-527 groups; opposite trends were observed after resveratrol treatment.

Conclusions: The intraperitoneal injection of 40 mg/kg 3-NPA for 3 weeks could effectively induce POI. The increase in oxidative stress inhibited SRIT1 and mitochondrial oxidative phosphorylation, inducing follicular apoptosis.

Keywords: 3-nitropropionic acid; mouse model; ovarian reserve; oxidative stress; premature ovarian insufficiency.

Figure 1

Comparison of clinical manifestations at different doses of 3-NPA. (A, B) 3-NPA had no significant effects on body weight. (C) 3-NPA could prolong the mean length of the diestrus phase and decrease the mean length of the estrous phase after two weeks of treatment, especially when used at a dose of 50 mg/kg. (D) 3-NPA could prolong the mean length of the diestrus phase and decrease the mean length of the estrous phase after three weeks of treatment, especially when used at a dose of 40 mg/kg and 50 mg/kg. (E, F) 3-NPA reduced the ovarian index, especially when used at a dose of 40 mg/kg and 50 mg/kg. (G, H) Serum E₂ concentration in different groups; mice treated with 40 mg/kg 3-NPA for 3 weeks and those treated with 50 mg/kg 3-NPA for 2 or 3 weeks showed a significant decrease. (I, J) Serum FSH concentration in the different groups; mice treated with 40 mg/kg or 50 mg/kg 3-NPA for 3 weeks showed a significant increase. (K, L) Serum AMH concentration in different groups; mice treated with 40 mg/kg 3-NPA for 3 weeks and those treated with 50 mg/kg 3-NPA for 2 or 3 weeks showed a significant decrease. (M) In the mice treated with 50 mg/kg 3-NPA for 2 weeks, the number of primordial follicles decreased. (N) In the mice treated with 40 mg/kg or 50 mg/kg 3-NPA for 3 weeks, the number of primordial follicles decreased, the number of atretic follicles increased, and the number of growing follicles did not change significantly. (N=8 in all assays; *P<0.05, **P<0.01, and ***P<0.001 compared with the control group, NS: none significant).

Figure 2

40 mg/kg 3-NPA significantly induced manifestations similar to those of POI. (A) 40 mg/kg 3-NPA could prolong the mean length of the diestrus phase and decrease the mean length of the estrous phase after three weeks of treatment. (B) 40 mg/kg 3-NPA could decrease the number of primordial follicles and increase that of atretic follicles. (C–F) 40mg/kg 3-NPA decreased the ovarian index (C), increased serum FSH levels (D), decreased serum E₂ levels (E), decreased serum AMH levels (F), and decreased mouse fertility (G). (H) Although the 40 mg/kg 3-NPA group had fewer pups, the difference was not significant. (I, J) The offspring of the mice treated with 40 mg/kg 3-NPA showed lower weight than the control group (N=8 in all assays; *P<0.05, **P<0.01).

Figure 3

Comparison of ovarian oxidative stress-related changes between the control and 3-NPA groups. (A–U) SOD expression was lower (A–G), GSH-Px expression was lower (H–N), and MDA expression was higher (O–U) in GCs from 3-NPA mice than in those from control mice (Triplicates slides in each group, N=8 in all assays; *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001).

Figure 4

SIRT1 is involved in the function of human GCs. (A) SIRT1 was downregulated in the 3-NPA model group in immunohistochemistry (Triplicates slides in each group). (B) SIRT1 was downregulated in the 3-NPA model group in western blot (Quadruplicates samples in each group). (C, D) SIRT1 was knocked down using a lentivirus. (E) Estrogen synthesis decreased after SIRT1 knockdown. (F) Genes related to estrogen synthesis were down-regulated after SIRT1 knockdown. (G) Functional receptors were down-regulated in GCs after SIRT1 knockdown (*P<0.05, **P<0.01, ***P<0.001, ****P<0.0001).

Figure 5

Res improves ovarian function by activating SIRT1. (A) No changes in body weight. (B) Ex-527 and 3-NPA prolonged the mean length of the diestrus phase and decreased the mean length of the estrous phase, while Res had the opposite effect. (C) Ex-527 and 3-NPA decreased the ovarian index and Res increased the ovarian index. (D) Ex-527 and 3-NPA increased serum FSH levels, and Res decreased these levels. (E) Ex-527 and 3-NPA decreased serum AMH levels and Res increased these levels. (F) Ex-527 and 3-NPA decreased the number of primordial follicles. (G) Ex-527 and 3-NPA decreased SIRT1 activity, while Res increased it. (H, I) Ex-527 and 3-NPA could decrease SOD (H) and GSH-Px (I) activity, while Res could increase SOD (H) and GSH-Px (I) activity. (J) Ex-527 and 3-NPA increased MDA activity, while Res decreased this activity (N=8 in all assays; *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001).

Figure 6

Mitochondrial oxidative phosphorylation-induced apoptosis may result in oxidative stress-induced POI. (A) Granulosa cells in ovarian tissue from the control group. (B) In the 3-NPA model group, GCs apoptosis and nuclear pyknosis were obvious, the chromatin margin was set, organelle structure was destroyed, and empty shots appeared. (C) Obvious granulocyte apoptosis was observed in ovarian tissue from mice in the Ex-527 intervention group; chromatin agglutination and vacuoles were also observed. (D) Ovarian granulocyte apoptosis was significantly lower in the resveratrol intervention group than in the model group. (E) Mitochondrial morphology in ovarian tissue from mice in the control group was largely normal, and the number of mitochondria was high. (F) In the 3-NPA group, mitochondrial morphology in ovarian tissue was destroyed, stromal edema was obvious, and the endoplasmic reticulum was dilated and damaged. (G) Ovarian mitochondria in mice from the Ex-527 intervention group were obviously swollen and cavitated. (H) Ovarian mitochondria in the resveratrol intervention group showed significantly improved morphology; the mitochondrial crest was clearly visible, and the mitochondrial matrix was uniform. (I) OXPHOS expression was down-regulated in the Ex-527 and 3-NPA groups, while it was up-regulated in the Res group. (J) P53, BAX, and CASPASE3 were up-regulated in the Ex-527 and 3-NPA groups and down-regulated in the Res group. BCL2 was down-regulated in the Ex-527 and 3-NPA groups, while it was up-regulated in the Res group. Red arrow: granular nucleus; Blue arrow: mitochondria.