Follicular stage-dependent regulation of apoptosis and steroidogenesis by prohibitin in rat granulosa cells

Cell culture media (M199), fetal bovine serum (FBS), penicillin and streptomycin, L-glutamine, sodium pyruvate and trypsin were purchased from Invitrogen (Burlington, Canada). HEPES, Hoechst 33258, equine chronic gonadotropin (eCG), and diethylstilbestrol (DES) were purchased from Sigma (St. Louis, MO). Recombinant human FSH was purchased from National Hormone and Peptide Program (Harbor-UCLA Medical Center, Torrance, CA). Anti-caspase-3 antibody (recognizing both intact and active caspase-3), anti-phospho-Akt (S473) and anti-Akt antibodies were purchased from Cell signaling (Danvers, MA), anti-PHB and anti-β-Actin antibodies were from Abcam (Cambridge, MA). Horseradish peroxidase-conjugated secondary antibodies and reagents for SDS-PAGE were supplied by Bio-Rad (Mississauga, Canada). Enhanced chemiluminescent (ECL) reagent was from Thermo Fisher Scientific (Rockford, IL). Adenoviral-PHB, shPHB and their control particles were obtained from Dr. Winston Thompson (Morehouse School of Medicine, Atlanta). QIAShredder and RNeasy mini kit were purchased from QIAGEN (Mississauga, Canada). Random decamer primers were from Ambion (Austin, TX). Ribonuclease inhibitor and dNTP were from Fermentas (Burlington, Canada). Moloney murine leukemia virus reverse transcriptase was from Promega (Madison, WI). PCR primers were from Invitrogen. All chemical inhibitors were purchased from Calbiochem (Gibbstown, NJ). All other chemicals were of the highest analytical grade available from Sigma.

Granulosa cells from eCG-primed immature rats (10 IU, 48h, s.c.; considered as differentiated granulosa cells) and DES-injected control rats (1 mg/day for 3 consecutive days, s.c.; considered as undifferentiated granulosa cells) were pre-incubated with 6 mM EGTA and 0.5 M sucrose [24] and were released by follicular puncture with a 26.5-gauge needle, washed and centrifuged (900 × g, 5 min). Cell clumps and oocytes were removed by filtering the cell suspensions through a 40-μm nylon cell strainer (BD Biosciences). The viability of granulosa cells was determined by trypan blue exclusion. Granulosa cells (0.9 × 10⁶ per well in 6-well plate) were plated overnight in M199 with 10% FBS under a humidified atmosphere of 95% air and 5% CO₂. After culture overnight in serum-free medium, granulosa cells were treated with FSH (0–200 ng/ml) or STS (1 μM) for a designated duration.

For adenoviral infection, granulosa cells were cultured in serum-free M199 medium containing adenoviral particles for 24 h followed by medium change. Multiplicity of infection (MOI) and duration of infection are detailed in the figures. Equal amounts of adenovirus in each experimental group were achieved by adjusting with an appropriate amount of adenoviral-LacZ (negative control for adenoviral-PHB) or adenoviral-shNeg (negative control for shPHB).

Total RNA of granulosa cells was extracted according to the manufacturer’s instruction, using the QIAGEN RNeasy Mini kit. Two hundred ng total RNA were used to reverse transcribe cDNAs and the mRNA abundance of target genes was analyzed by PCR. The PHB primers used for amplification were a 5^′ forward primer (5^′-TGGCAGCCTGAGTAGACCTT-3^′) and a 3^′ reverse primer (5^′-TCACGGTTAAGAGGGAATGG-3^′). The p450scc primers were a 5^′ forward primer (5^′-ACCCTGAGTCCCAGCGGTTC-3^′) and a 3^′ reverse primer (5^′-CACCCC-TCCTGCCAGCATCT-3^′). The aromatase primers were a 5^′ forward primer (5^′-TGGTCCCG-GAAACTGTGCCT-3^′) and a 3^′ reverse primer (5^′-CCACGCTTGCTGCCGAATCT-3^′). The actin primer were (5^′-CGTCCACCCGCGAGTACAAC-3^′) and a 3^′ reverse primer (5^′- GCCT-CTCTTGCTCTGGGCCT-3^′). The thermal cycling conditions were comprised of an initial denaturation step at 95 C for 10 min followed by 30 cycles amplification for PHB, p450scc and aromatase (20 cycles for actin) at 95 C for 30 sec, 55 C for 30 sec, and 72 C for 30 sec. The PCR products were subjected to 2% ethidium bromide-containing agarose gel and visualized under UV light.

At the end of the culture period, floating cells and attached cells (recovered by 0.05% trypsin treatment) were pooled and centrifuged (1000 × g, 10 min). For protein extraction, cell pellets were resuspended in a lysis buffer (PBS, pH 7.4) containing NaCl (150 mM), SDS (0.1%), sodium deoxycholate (0.5%), Nonidet P-40 (1%), and the protease inhibitor cocktail (Sigma) and kept on ice for 30 min. Cell lysates were sonicated and centrifuged (12,000 × g, 5 min, 4°C) to remove insoluble material. Supernatant was recovered and stored at −20°C until further processing. Protein concentrations in each sample were determined by the Bradford assay (Bio-Rad Laboratories). Twenty μg of protein of cell lysates were subjected to SDS-PAGE with 4.5% stacking and 15% separating gels. Proteins were electrophoretically transferred onto nitrocellular membrane (NC, Bio-Rad), blocked at room temperature with 5% skim milk in TBST [0.05% Tween-20 in Tris (10 mM) and NaCl (0.15 M), pH7.4 (TBS)] for 1 h and then incubated overnight at 4°C with diluted primary antibodies (1:1000) in TBST with constant agitation. The membranes were then treated with a secondary antibody (1:2000 to 1:10,000 based on different primary antibody). After washing three times with TBST, immunoreactive bands were visualized with ECL according to the manufacturer’s instruction. Intensity of bands of the exposed X-ray film was determined by densitometrically scanned, quantified, using AlphaEaseFC (Alpha Innotech, CA) and normalized with β-Actin.

Apoptotic cells were identified morphologically by Hoechst-33258 (bisBenzimide, Sigma) staining as previously reported [25]. At the end of the culture period, suspended cells were collected by centrifugation and attached cells were trypsinized. The two cell fractions were then pooled, pelleted, and suspended in 10% phosphate-buffered formalin containing Hoechst 33258 (6.25 μg/ml; room temperature, 2 h), Cells were then spotted on slides and assessed for typical apoptotic nuclear morphology. To quantify the number of apoptotic cells, healthy and apoptotic cells were counted (counter was “blinded to sample identity”) and the apoptotic cells were expressed as a percentage of total cells. A minimum of 400 cells were counted in each treatment group.

Spent medium from granulosa cell cultures were collected, centrifuged (900 × g, 5 min) and kept in −80°C for hormone analysis. 17β-estradiol and progesterone concentrations in spent medium were measured using enzyme immunoassay kit (EIA; Enzo Life Sciences, Farmingdale, NY) according to the manufacturer’s instruction. The detection limitation of estradiol was 28 pg/ml, and the intra- and inter-assay coefficients of variation were 8 and 6%, respectively. The detection limitation of progesterone was 8.5 pg/ml, and the intra- and inter-assay coefficients of variation were 7 and 6%, respectively.

To examine whether PHB is expressed in vivo in a follicular stage-dependent manner, differentiated and undifferentiated granulosa cells were collected and the gene expression of PHB was analyzed by traditional PCR. As shown in Figure 1A, differentiated granulosa cells exhibited increased expression of p450scc and aromatase, steroidogenic enzymes known to be associated with granulosa cell differentiation. The PHB expression in differentiated granulosa cells was lower than that in undifferentiated granulosa cells.

To explore whether FSH regulates PHB expression in vitro, undifferentiated and differentiated granulosa cells were cultured with FSH (100 ng/ml) for designated time (0–24 h) and the contents of PHB in two types of cells were examined. As shown in Figure 1B, FSH had no effect on PHB content in undifferentiated granulosa cells (p > 0.05). In the contrast, FSH up-regulated PHB expression in differentiated granulosa cells (p < 0.001). While there was an apparent gradual increase in PHB content with the duration of culture in the presence of FSH, significant upregulation was not evident until 24 h. Both undifferentiated (p < 0.0001) and differentiated (p < 0.001) granulosa cells exhibited a rapid increase in phosphorylated Akt content; however, Akt activation in the former was stronger (6-fold vs. 2-fold change at 0.25h) and sustained longer (over basal level at 4-8h) compared with that in the latter.

We then examined whether FSH regulates PHB contents in differentiated granulosa cells in a concentration-dependent manner. Undifferentiated and differentiated granulosa cells were cultured with different FSH concentration (0–200 ng/ml) for 24 h and the content of PHB at the two follicular stages was examined. As shown in Figure 1C, FSH had no effect on PHB content in undifferentiated granulosa cells (p > 0.05); however FSH significantly up-regulated the content of PHB (p < 0.001) at a concentration range of 10–100 ng/ml in differentiated granulosa cells. There was no significant difference between the FSH (200 ng/ml) and the control groups. Since we have reported that Akt and PHB could regulate each other [23], we also assessed the phosphorylated and total Akt levels in this experiment. As shown in Figure 1C, although there was a trend of decrease, FSH didn’t significantly reduce Akt phosphorylation in undifferentiated granulosa cells (p > 0.05). In the contrast, phosphorylated Akt levels in granulosa cells in differentiated granulosa cells decreased in the presence of FSH (Figure 1C, p < 0.05), which was reversely related to PHB contents.

Although PHB has been reported to be anti-apoptotic in undifferentiated granulosa cells [19, 20], whether it plays a similar role in granulosa cells after differentiation is not known. To address this question, comparison experiments were performed using undifferentiated and differentiated granulosa cells. Granulosa cells were infected with adenoviral-PHB (MOI = 40, adenoviral-LacZ as control, 24 h) to over-express or knockdown PHB, and then cultured with the apoptosis inducer staurosporine (STS, 1 μM, 2 h). As shown in Figure 2A, whereas PHB over-expression had no effect on basal level of apoptosis in undifferentiated granulosa cells, it suppressed STS-induced caspase-3 cleavage (Figure 2A, PHB, p < 0.05; STS, p < 0.0001; PHB × STS, p < 0.05) and apoptosis (PHB, p < 0.05; STS, p < 0.0001; PHB × STS, p < 0.05). In contrast, overexpression of PHB in differentiated granulosa cells had no obvious effect on STS-induced caspase-3 cleavage and apoptosis (Figure 2B), suggesting that PHB is anti-apoptotic during preantral follicular growth but this property is lost as the cells differentiate during follicle transition into the antral stage.

Similar experiments were performed using adenoviral-shRNA to knockdown PHB in granulosa cells derived from two preparations. As shown in Figure 3, STS-induced caspase-3 cleavage and apoptosis were enhanced after knockdown of PHB in undifferentiated granulosa cells (Figure 3A, for cleaved caspase-3, shPHB, p < 0.01; STS, p < 0.0001; shPHB × STS, p < 0.05. For apoptosis, shPHB, p < 0.01; STS, p < 0.0001; shPHB × STS, p < 0.05). However, there was no effect of PHB knockdown on STS-induced caspase-3 cleavage and apoptosis in differentiated granulosa cells (Figure 3B).

We have previously reported that PHB suppressed FSH-induced estradiol and progesterone secretion and p450scc/aromatase expression in undifferentiated granulosa cells [23]. As the regulatory role of PHB in granulosa cell apoptosis is dependent on the state of cellular differentiation, we then examined whether it regulate steroidogenesis differently in differentiated granulosa cells by comparing the steroidogenic responsiveness of granulosa cells at the two differentiative states. Undifferentiated and differentiated granulosa cells from two preparations were cultured with FSH (0–100 ng/ml) ± testosterone (T, 0.5 μM, 24 h), which served as a substrate of aromatase, and then the levels of estradiol and progesterone in spent medium were measured. We have previously demonstrated that T enhanced FSH-induced progesterone and estradiol secretion in undifferentiated granulosa cells [23]. We observed that FSH-induced progesterone production in these cells was dramatically increased in the presence of T (5.49 ± 0.4 vs. 0.63 ± 0.13 ng/ml) and the estradiol secretion exhibited similar effect (48.7 ± 5.7 vs. 0.91 ± 0.16 ng/ml). We concomitantly tested the effect of T and FSH on differentiated cells (Figure 4A) and observed that T also enhanced FSH-stimulated progesterone (7.79 ± 0.63 vs. 1.22 ± 0.98 ng/ml) and estradiol secretion (94.4 ± 10.23 vs. 0.88 ± 0.14 ng/ml) in differentiated granulosa cells (Figure 4A, for progesterone: FSH, p < 0.001; T, p < 0.001; FSH × T, p < 0.001. for estradiol: FSH, p < 0.001; T, p < 0.001; FSH × T, p < 0.001). The basal levels of estradiol and progesterone at the two cellular differentiative states were similar; however production of these steroids in the presence of FSH and T was about 2-fold higher in differentiated granulosa cells than in undifferentiated granulosa cells.

Next we examined whether PHB suppresses FSH-induced steroidogenesis in granulosa cells and if its response is dependent on the differentiated state of the cells. Granulosa cells from two preparations were infected with adenoviral-PHB (adenoviral-LacZ as control) and then cultured with FSH (0–100 ng/ml) in the presence of T (0.5 μM) for 24h. The levels of estradiol and progesterone in spent medium were measured by EIA. Exogenous PHB suppressed FSH-induced steroid production and the expression of steroidogenic enzymes p450scc and aromatase in undifferentiated granulosa cells [23]. FSH-induced progesterone production in these cells was dramatically suppressed by exogenous PHB (5.45 ± 0.47 vs. 3.14 ± 0.35 ng/ml) and the estradiol secretion exhibited similar effect (55.6 ± 5.0 vs. 21.76 ± 7.72 ng/ml). Concomitant studies with differentiated granulosa cells indicate that PHB similarly inhibited FSH-induced progesterone (7.70 ± 0.52 vs. 4.70 ± 0.78 ng/ml) and estradiol (95.56 ± 11.17 vs. 48.96 ± 6.7 ng/ml) secretion (Figure 4B, for progesterone: PHB, p < 0.01; FSH, p < 0.001; PHB × FSH, p < 0.01. for estradiol: PHB, p < 0.05; FSH, p < 0.001; PHB × FSH, p < 0.05) in vitro. The contents of p450scc and aromatase induced by FSH were also suppressed by PHB (Figure 4C).