Melatonin Promotes Ubiquitination of Phosphorylated Pro-Apoptotic Protein Bcl-2-Interacting Mediator of Cell Death-Extra Long (BimEL) in Porcine Granulosa Cells

Melatonin (N-acetyl-5-methoxytryptamine) is found in ovarian follicular fluid, and its concentration is closely related to follicular health status. Nevertheless, the molecular mechanisms underlying melatonin function in follicles are uncertain. In this study, melatonin concentration was measured in porcine follicular fluid at different stages of health. The melatonin concentration decreased as the follicles underwent atresia, suggesting that melatonin may participate in the maintenance of follicular health. The molecular pathway through which melatonin may regulate follicular development was further investigated. The pro-apoptotic protein BimEL (Bcl-2-interacting mediator of cell death-Extra Long), a key protein controlling granulosa cell apoptosis during follicular atresia, was selected as the target molecule. BimEL was downregulated when porcine granulosa cells were cultured in medium containing 10−9 M melatonin and isolated cumulus oocyte complexes (COCs) or follicle stimulating hormone (FSH). Interestingly, ERK-mediated phosphorylation was a prerequisite for the melatonin-induced decline in BimEL, and melatonin only promoted the ubiquitination of phosphorylated BimEL, and did not affect the activities of the lysosome or the proteasome. Moreover, the melatonin-induced downregulation of BimEL was independent of its receptor and its antioxidant properties. In conclusion, melatonin may maintain follicular health by inducing BimEL ubiquitination to inhibit the apoptosis of granulosa cells.


Introduction
During mammalian follicular development, only a limited number of follicles are selected to ovulate, with the remainder undergoing atresia at different stages. Follicle atresia is triggered by the apoptosis of granulosa cells, and several apoptotic-signaling molecules, such as hormones/growth factors/cytokines, the death ligand-receptor system, and B cell lymphoma/leukemia 2 (Bcl-2) family members, are involved in this process [1][2][3][4]. The Bcl-2 family, which includes both anti-apoptotic (Bcl-2, B cell lymphoma/leukemia X (Bcl-X)) and pro-apoptotic (Bcl-2 interacting domain (Bid), Bim, Bax, Bak) proteins, are key regulators of apoptosis, and members of the Bcl-2 protein family play pivotal roles in follicular growth and atresia [2,5,6].
Bim (Bcl-2 interacting mediator of cell death), a BH3-only family member, binds with high affinity to anti-apoptotic Bcl-2 family members and regulates apoptotic signaling through Bax and Bak [7]. Moreover, Bim can be phosphorylated by several MAP kinases to regulate its activity [8]. Bim EL is the predominant isoform of Bim in several mammalian tissues, as determined using Western blotting analysis [9]. Our previous studies showed that Bim EL participates in porcine follicular atresia through

Melatonin Downregulates Bim EL in Porcine Granulosa Cells
The Bim EL protein in the surrounding cumulus granulosa cells decreased significantly after porcine COCs were treated with 10 −9 M melatonin for 42-44 h ( Figure 1A). However, there was no change in the Bim EL level in primary granulosa cells after treatment with 10 −9 M melatonin for 24 h ( Figure 1B). Interestingly, melatonin significantly decreased the Bim EL level when primary granulosa cells were cocultured with COCs ( Figure 1B). Similar results were obtained after granulosa cells were treated with FSH ( Figure 1C). Furthermore, COCs exposure and FSH treatment resulted in a Bim EL protein mobility shift ( Figure 1A-C), and this shift was inhibited when the lysates from FSH-treated primary granulosa cells were incubated with λ phosphatase ( Figure 1D), indicating that the observed Bim EL mobility shift was caused by its phosphorylation. These results also indicated that COCs or FSH could induce Bim EL phosphorylation in granulosa cells. Thus, these data suggest that the function of melatonin in the downregulation of Bim EL may depend on its phosphorylation level.

Melatonin Downregulates Bim EL Depending on ERK Activation and Bim EL Phosphorylation
Phosphorylation of Bim EL by extracellular signal-regulated kinase 1/2 (ERK1/2) has been shown to induce the degradation of Bim EL [27], so we first determined the status of ERK activation. As shown in Figure 2A,C, ERK1/2 in granulosa cells was activated after treatment with COCs or FSH, but it was not induced by melatonin, revealing that melatonin could not directly promote the degradation of Bim EL by phosphorylation. To determine the precise role of melatonin in this process, primary granulosa cells cultured in the presence of COCs or FSH were pretreated with U0126, an MEK1/2 inhibitor, before melatonin treatment. As expected, U0126 abolished the induction of Bim EL phosphorylation by COCs or FSH in parallel with the abrogation of Bim EL downregulation caused by melatonin ( Figure 2B,D). These results confirmed that the phosphorylation of Bim EL induced by ERK was a prerequisite for Bim EL reduction induced by melatonin. To further confirm whether this phenomenon exists in follicles in vivo, the lysates from granulosa cells obtained from healthy or atretic follicles were subjected to SDS-PAGE to detect ERK activation and Bim EL expression. As shown in Figure 2E, the level of activated ERK1/2 was higher, whereas the Bim EL level was lower, in granulosa cells of healthy follicles compared to atretic follicles. Furthermore, melatonin concentration decreased with the atresia of porcine ovarian follicles. The concentrations of melatonin in healthy, slightly atretic, and atretic follicles were 47.47 ± 6.03 ng/L, 41.97 ± 5.66 ng/L, and 36.50 ± 2.84 ng/L, respectively, and the difference between healthy follicles and slightly atretic or atretic ones was significant (p < 0.05, Figure 2F). These results suggest that ERK activation is responsible for the induction of Bim EL phosphorylation by COCs or FSH, and it promotes melatonin-induced Bim EL downregulation in porcine granulosa cells. This process is likely to play a vital role in maintaining follicle health.

Melatonin Downregulates BimEL in Porcine Granulosa Cells
The BimEL protein in the surrounding cumulus granulosa cells decreased significantly after porcine COCs were treated with 10 −9 M melatonin for 42-44 h ( Figure 1A). However, there was no change in the BimEL level in primary granulosa cells after treatment with 10 −9 M melatonin for 24 h ( Figure 1B). Interestingly, melatonin significantly decreased the BimEL level when primary granulosa cells were cocultured with COCs ( Figure 1B). Similar results were obtained after granulosa cells were treated with FSH ( Figure 1C). Furthermore, COCs exposure and FSH treatment resulted in a BimEL protein mobility shift ( Figure 1A-C), and this shift was inhibited when the lysates from FSH-treated primary granulosa cells were incubated with λ phosphatase ( Figure 1D), indicating that the observed BimEL mobility shift was caused by its phosphorylation. These results also indicated that COCs or FSH could induce BimEL phosphorylation in granulosa cells. Thus, these data suggest that the function of melatonin in the downregulation of BimEL may depend on its phosphorylation level.

Melatonin Downregulates BimEL Depending on ERK Activation and BimEL Phosphorylation
Phosphorylation of BimEL by extracellular signal-regulated kinase 1/2 (ERK1/2) has been shown to induce the degradation of BimEL [27], so we first determined the status of ERK activation. As shown in Figure 2A,C, ERK1/2 in granulosa cells was activated after treatment with COCs or FSH, but it was not induced by melatonin, revealing that melatonin could not directly promote the degradation of BimEL by phosphorylation. To determine the precise role of melatonin in this process, primary granulosa cells cultured in the presence of COCs or FSH were pretreated with U0126, an MEK1/2 inhibitor, before melatonin treatment. As expected, U0126 abolished the induction of BimEL

Post-Translational Pathway Is Involved in Melatonin-Induced Downregulation of BimEL
The molecular mechanism of melatonin-induced downregulation of BimEL was systemically investigated using porcine adherent granulosa cells with the experimental protocol shown in Figure  3A. After 12 h of serum withdrawal, a significant increase in phosphorylated BimEL was observed

Post-Translational Pathway Is Involved in Melatonin-Induced Downregulation of Bim EL
The molecular mechanism of melatonin-induced downregulation of Bim EL was systemically investigated using porcine adherent granulosa cells with the experimental protocol shown in Figure 3A. After 12 h of serum withdrawal, a significant increase in phosphorylated Bim EL was observed ( Figure 3B), accompanied by a robust activation of ERK1/2, which was similar to that in primary granulosa cells treated with COCs or FSH. To determine whether melatonin could downregulate the Bim EL protein in porcine adherent granulosa cells, cells were treated with melatonin at different concentrations (0, 10 −11 , 10 −9 , 10 −7 M) for 24 h. As shown in Figure 3C, the levels of Bim EL and Cleaved Caspase3 significantly decreased after 10 −9 M melatonin treatment, and this effect was evident within 3 h after treatment ( Figure 3D).  Figure 3B), accompanied by a robust activation of ERK1/2, which was similar to that in primary granulosa cells treated with COCs or FSH. To determine whether melatonin could downregulate the BimEL protein in porcine adherent granulosa cells, cells were treated with melatonin at different concentrations (0, 10 −11 , 10 −9 , 10 −7 M) for 24 h. As shown in Figure 3C, the levels of BimEL and Cleaved Caspase3 significantly decreased after 10 −9 M melatonin treatment, and this effect was evident within 3 h after treatment ( Figure 3D).  Because the Bim EL protein expression level can be regulated by transcriptional and post-translational pathways, our next experiments aimed to determine the mechanism responsible for this change. As shown in Figure 3E, there was no difference in the mRNA expression of bim 3 h post melatonin treatment compared to the control group. Therefore, we hypothesized that the downregulation of Bim EL was controlled by post-translational modifications. To address this, porcine adherent granulosa cells were incubated with cycloheximide (CHX) alone or co-treated with CHX and melatonin for indicated time periods ( Figure 3F). The combination of CHX and melatonin induced a rapid decrease in the Bim EL level within 3 h after treatment compared with CHX alone. These results indicate that the melatonin-mediated Bim EL decline is regulated at the post-translational level, and Bim EL is actively degraded.

Melatonin Promotes Bim EL Ubiquitination
Proteasomes and lysosomes comprise two major intracellular proteolytic systems in mammalian cells. Hence, we investigated whether Bim EL was increasingly degraded by lysosomes or proteasomes following melatonin treatment. First, porcine adherent granulosa cells were treated with melatonin in the presence of a potent inhibitor of the vacuolar type H + -ATPase, Bafilomycin A1, or chloroquine, which accumulated in lysosomes and raised the intralysosomal pH value. The inhibitors induced the accumulation of LC3-II (data not shown) to manifest its inhibition of lysosomal proteolysis. As shown in Figure 4A, Bafilomycin A1 and chloroquine failed to block Bim EL downregulation by melatonin, indicating that the proteasomal pathway was responsible for Bim EL degradation. Protein phosphorylation is required to prepare Bim for ubiquitination and proteasomal degradation. Porcine adherent granulosa cells were treated with melatonin in the presence of the proteasome inhibitor MG132 for 3 h. However, MG132 could also not counteract the Bim EL repression by melatonin ( Figure 4B). To confirm whether melatonin could influence the proteasomal activity in granulosa cells, we measured chymotrypsin-like activity using a commercially-available proteasome 20S assay kit. Melatonin did not change the proteasomal activity in granulosa cells ( Figure 4C). These results indicate that neither lysosomes nor proteasomes directly participate in Bim EL reduction by melatonin. Because the BimEL protein expression level can be regulated by transcriptional and posttranslational pathways, our next experiments aimed to determine the mechanism responsible for this change. As shown in Figure 3E, there was no difference in the mRNA expression of bim 3 h post melatonin treatment compared to the control group. Therefore, we hypothesized that the downregulation of BimEL was controlled by post-translational modifications. To address this, porcine adherent granulosa cells were incubated with cycloheximide (CHX) alone or co-treated with CHX and melatonin for indicated time periods ( Figure 3F). The combination of CHX and melatonin induced a rapid decrease in the BimEL level within 3 h after treatment compared with CHX alone. These results indicate that the melatonin-mediated BimEL decline is regulated at the post-translational level, and BimEL is actively degraded.

Melatonin Promotes BimEL Ubiquitination
Proteasomes and lysosomes comprise two major intracellular proteolytic systems in mammalian cells. Hence, we investigated whether BimEL was increasingly degraded by lysosomes or proteasomes following melatonin treatment. First, porcine adherent granulosa cells were treated with melatonin in the presence of a potent inhibitor of the vacuolar type H + -ATPase, Bafilomycin A1, or chloroquine, which accumulated in lysosomes and raised the intralysosomal pH value. The inhibitors induced the accumulation of LC3-II (data not shown) to manifest its inhibition of lysosomal proteolysis. As shown in Figure 4A, Bafilomycin A1 and chloroquine failed to block BimEL downregulation by melatonin, indicating that the proteasomal pathway was responsible for BimEL degradation. Protein phosphorylation is required to prepare Bim for ubiquitination and proteasomal degradation. Porcine adherent granulosa cells were treated with melatonin in the presence of the proteasome inhibitor MG132 for 3 h. However, MG132 could also not counteract the BimEL repression by melatonin ( Figure  4B). To confirm whether melatonin could influence the proteasomal activity in granulosa cells, we measured chymotrypsin-like activity using a commercially-available proteasome 20S assay kit. Melatonin did not change the proteasomal activity in granulosa cells ( Figure 4C). These results indicate that neither lysosomes nor proteasomes directly participate in BimEL reduction by melatonin.  It has been shown that activation of the ERK pathway promotes the phosphorylation of Bim EL , which serves to mark Bim EL for ubiquitination [28]. Immunoprecipitation of Bim EL , followed by immunoblot analysis using an anti-ubiquitin antibody, demonstrated that melatonin enhanced the level of Bim EL poly-ubiquitination in porcine adherent granulosa cells supplemented with MG132 ( Figure 5A, left lane). We found that the level of ubiquitinated Bim EL increased, whereas levels of Bim EL , Bim L , and Bim S decreased after melatonin treatment ( Figure 5B, right lane). Western blot analysis of whole-cell lysates showed that melatonin also elevated the overall level of ubiquitination in both porcine primary and adherent granulosa cells ( Figure 5B,C). Taken together, our data suggest that melatonin activates the ubiquitination of Bim EL . It has been shown that activation of the ERK pathway promotes the phosphorylation of BimEL, which serves to mark BimEL for ubiquitination [28]. Immunoprecipitation of BimEL, followed by immunoblot analysis using an anti-ubiquitin antibody, demonstrated that melatonin enhanced the level of BimEL poly-ubiquitination in porcine adherent granulosa cells supplemented with MG132 ( Figure 5A, left lane). We found that the level of ubiquitinated BimEL increased, whereas levels of BimEL, BimL, and BimS decreased after melatonin treatment ( Figure 5B, right lane). Western blot analysis of whole-cell lysates showed that melatonin also elevated the overall level of ubiquitination in both porcine primary and adherent granulosa cells ( Figure 5B,C). Taken together, our data suggest that melatonin activates the ubiquitination of BimEL.

Reduction of Bim EL by Melatonin Does Not Associate with Its Receptor or Antioxidant Properties
To classify the underlying mechanisms of the melatonin-induced decrease in Bim EL , we investigated the potential role of melatonin receptors because various physiological effects of melatonin can be mediated by its two G-protein-coupled MT1 and MT2 receptors. Porcine adherent granulosa cells were treated with melatonin in the presence of luzindole, a melatonin receptor antagonist, and then Bim EL protein was examined. As shown in Figure 6A, luzindole failed to block the Bim EL downregulation by melatonin. In addition, melatonin is known to be a powerful antioxidant. To determine the relevance of antioxidant activity on Bim EL degradation, we tested the effects of two other antioxidant reagents, N-acetylcysteine (NAC) and ascorbic acid (AA), on the Bim EL protein in porcine adherent granulosa cells. The western blot showed no change in the level of Bim EL after treatment compared to the control ( Figure 6B,C). Based on these data, we conclude that melatonin-mediated Bim EL downregulation is independent of the melatonin receptor-mediated pathway or its antioxidant function.

Reduction of BimEL by Melatonin Does Not Associate with Its Receptor or Antioxidant Properties
To classify the underlying mechanisms of the melatonin-induced decrease in BimEL, we investigated the potential role of melatonin receptors because various physiological effects of melatonin can be mediated by its two G-protein-coupled MT1 and MT2 receptors. Porcine adherent granulosa cells were treated with melatonin in the presence of luzindole, a melatonin receptor antagonist, and then BimEL protein was examined. As shown in Figure 6A, luzindole failed to block the BimEL downregulation by melatonin. In addition, melatonin is known to be a powerful antioxidant. To determine the relevance of antioxidant activity on BimEL degradation, we tested the effects of two other antioxidant reagents, N-acetylcysteine (NAC) and ascorbic acid (AA), on the BimEL protein in porcine adherent granulosa cells. The western blot showed no change in the level of BimEL after treatment compared to the control ( Figure 6B,C). Based on these data, we conclude that melatonin-mediated BimEL downregulation is independent of the melatonin receptor-mediated pathway or its antioxidant function.

Discussion
Melatonin plays a pivotal role in female reproduction, including puberty, ovarian follicle growth, ovulation, and luteinization [29][30][31]. Several studies have demonstrated that melatonin has beneficial effects on oocyte maturation and subsequent embryo development in many species, such as mice [32], cattle [33], and pigs [15]. Except for its antioxidant properties, however, the role of melatonin in female reproduction remains largely unknown. Throughout the reproductive life span, more than 99% of germ cells are eliminated from the ovary through follicular atresia, and granulosa cells play a major role in this process. It is well-established that the initial step of follicular atresia is granulosa cell apoptosis [34]. Our previous study showed that the pro-apoptotic protein BimEL plays an important role in porcine granulosa cell apoptosis [9]. In this study, we demonstrated that

Discussion
Melatonin plays a pivotal role in female reproduction, including puberty, ovarian follicle growth, ovulation, and luteinization [29][30][31]. Several studies have demonstrated that melatonin has beneficial effects on oocyte maturation and subsequent embryo development in many species, such as mice [32], cattle [33], and pigs [15]. Except for its antioxidant properties, however, the role of melatonin in female reproduction remains largely unknown. Throughout the reproductive life span, more than 99% of germ cells are eliminated from the ovary through follicular atresia, and granulosa cells play a major role in this process. It is well-established that the initial step of follicular atresia is granulosa cell apoptosis [34]. Our previous study showed that the pro-apoptotic protein Bim EL plays an important role in porcine granulosa cell apoptosis [9]. In this study, we demonstrated that melatonin decreases the Bim EL protein via inducing its ubiquitination through the post-translational pathway in porcine granulosa cells.
It has been shown that the supplementation of porcine maturation medium with 10 −9 M melatonin is beneficial for in vitro maturation (IVM) of porcine oocytes and subsequent embryo development [15]. Moreover, our previous study showed that Bim EL -mediated apoptosis in cumulus cells accelerates oocyte aging and degeneration [35]. In this study, we showed that melatonin decreased the Bim EL protein in cumulus granulosa cells during IVM. Granulosa cells play an important role in supporting oocyte maturation, which prompted us to focus on the effects of melatonin on Bim EL protein expression in porcine primary granulosa cells. However, melatonin treatment alone failed to decrease the Bim EL level. Conversely, a combination of melatonin with COCs or FSH downregulated the Bim EL protein, although melatonin can induce bim expression in several cancer cell lines [11,25,26]. Thus, these results suggested that the mechanism of regulating bim expression may be different between cancer cells and normal tissue cells or these contradictory results are from using different melatonin concentrations.
Our results demonstrated that ERK1/2 was activated in granulosa cells by COCs and FSH rather than by melatonin. These results were consistent with those of a previous report by Baumgarten et al., who found that FSH could activate the ERK pathway in human cumulus granulosa cells [36]. Oocyte-secreted factors such as GDF9 also activate ERK in human granulosa cells. A recent study reported that melatonin activated ERK1/2 in HEK293 cells in a concentration-dependent manner after 5 min of treatment [37]. Moreover, the activation of ERK1/2 induced by melatonin antagonized UVB-induced apoptosis in U937 cells [38]. Thus, melatonin appears to activate ERK1/2 in different cells. ERK1/2 kinase was responsible for Bim EL phosphorylation because the MEK1/2 inhibitor U0126 blocked COC-and FSH-induced Bim EL phosphorylation. However, U0126 can partially inhibit MEK5 and ERK5 activation [39]; thus, the phosphorylation of Bim EL may also be partially due to ERK5 activation. Moreover, accumulating data indicate that multiple phosphorylated isoforms of Bim EL are regulated by different kinase pathways, including ERK, JNK, and p38 MAP kinases, which may result in different apoptotic end-points [8]. For example, UV-mediated JNK activation results in the phosphorylation of Bim EL on Thr-112, potentiating its apoptotic activity [40], and sodium arsenite-induced apoptosis causes the phosphorylation of Bim EL at Ser-65 by p38 in PC12 cells [41]. In contrast, phosphorylation by ERK on Ser-55/65/73 targeted Bim EL for degradation via the ubiquitin-proteasome pathway and promoted cell survival [28]. In the present study, melatonin-induced downregulation of Bim EL was abolished by U0126, indicating that Bim EL phosphorylation by ERK was essential for this process; however, the specific phosphorylated sites in Bim EL induced by COCs or FSH should be confirmed in a further study.
In this study, we observed correlations among ERK1/2 activation, the Bim EL level, and follicle health status. An inverse relationship between ERK activation and Bim EL level was observed in porcine follicles, with healthy follicles displaying a higher level of phosphorylated ERK than atretic ones. Similarly, higher levels of phosphorylated ERK in dominant follicles were detected compared with subordinate follicles in sheep [42]. These data suggest that ERK activation plays an important role, not only in melatonin-induced downregulation of Bim EL , but also in the process of dominant follicle selection.
We showed that serum starvation of porcine adherent granulosa cells significantly activated ERK and enhanced Bim EL phosphorylation, which provided an ideal model to mimic the effects of COCs and FSH in primary granulosa cells. The same phenomenon of ERK activation by serum starvation also exists in human colon carcinoma cells [43]. Following the activation of ERK, melatonin downregulated the Bim EL protein and Cleaved Caspase 3 level in porcine adherent granulosa cells ( Figure 3C). It is well-established that gene transcription, mRNA stability, and post-translational modifications regulate the Bim EL level by different stimuli [8,44,45]. The transcriptional factors FOXO3a, Runx3, E2F1, c-Jun, and SP1 have been shown to regulate bim transcription. During follicular development, FSH regulates Bim EL expression through FoxO3a in granulosa cells [9]. It was demonstrated that melatonin induces the expression of transcription factors of Sp1 and E2F1, coinciding with the induction of Bim EL in renal cancer Caki cells [11]. However, we did not detect an obvious mRNA change of bim in porcine granulosa cells by real-time PCR after melatonin treatment ( Figure 3E). Moreover, a significant decline of Bim EL was observed when granulosa cells were treated with melatonin in the presence of CHX ( Figure 3F), implying that the downregulation of Bim EL in granulosa cells by melatonin was closely associated with its post-translational modification.
Proteasomes or lysosomes play imperative roles in controlling intracellular protein quantity and quality [46]. Choi et al. reported that melatonin increased the degradation of TGFBlp via activating autophagy and counteracted the inhibition of autophagy by Bafilomycin A1 in corneal fibroblasts [47]. Previous studies have shown that melatonin can inhibit proteasome activity [48]. Nevertheless, neither a lysosomal-degradation inhibitor nor a proteasomal inhibitor could inhibit the decline in Bim EL induced by melatonin in our study. Unexpectedly, melatonin markedly increased the ubiquitination of Bim EL . According to our results, melatonin may also have induced the ubiquitination of other proteins in granulosa cells. Therefore, melatonin decreased the Bim EL level by increasing its ubiquitination. Although melatonin was previously regarded as an inhibitor of the ubiquitin-proteasome system [49], our contradictory results may be attributed to different cells with different experimental conditions. Park et al. reported that melatonin could increase the level of Bim EL by inhibiting the activity of proteasomes to induce apoptosis of human renal cancer cells [11]. In their experiment, cancer cells were treated with melatonin at the concentration from 0.1 to 1 mM, but the concentration was just 1 nM in our experiment. Above all, according to previous studies, melatonin presented toxicity to the oocyte during maturation when its concentration reached at 10 −5 M [15]. In addition, the effect of melatonin on proteasome activity may be different between cancer cells and granulosa cells. It has been shown that mitochondrial-associated Bcl-2 family proteins, including Bax, Bcl-2, and Bim, are regulated through ubiquitin/proteasomal degradation during apoptosis [50]. Wan et al. reported that APCCdc20 acts as an E3 ubiquitin ligase to promote Bim ubiquitination and destruction [51]. Additional studies are required to delineate the precise mechanisms involved in melatonin-induced protein ubiquitination in granulosa cells.
Generally, melatonin participates in various physiological processes via its receptors or antioxidant properties [52][53][54]. However, our results showed that melatonin downregulated Bim EL through neither a receptor-mediated nor antioxidant pathway. Therefore, the detailed mechanism of Bim EL downregulation by melatonin in granulosa cells waits further investigation.
A hypothetical model of Bim EL regulation by melatonin in granulosa cells is presented in Figure 7. Although there are some details that need to be further pursued, our results demonstrate that melatonin downregulates phosphorylated Bim EL via a ubiquitin-proteasome pathway in porcine granulosa cells to maintain follicle health. Future studies should investigate how melatonin precisely regulates protein ubiquitination. obvious mRNA change of bim in porcine granulosa cells by real-time PCR after melatonin treatment ( Figure 3E). Moreover, a significant decline of BimEL was observed when granulosa cells were treated with melatonin in the presence of CHX ( Figure 3F), implying that the downregulation of BimEL in granulosa cells by melatonin was closely associated with its post-translational modification. Proteasomes or lysosomes play imperative roles in controlling intracellular protein quantity and quality [46]. Choi et al. reported that melatonin increased the degradation of TGFBlp via activating autophagy and counteracted the inhibition of autophagy by Bafilomycin A1 in corneal fibroblasts [47]. Previous studies have shown that melatonin can inhibit proteasome activity [48]. Nevertheless, neither a lysosomal-degradation inhibitor nor a proteasomal inhibitor could inhibit the decline in BimEL induced by melatonin in our study. Unexpectedly, melatonin markedly increased the ubiquitination of BimEL. According to our results, melatonin may also have induced the ubiquitination of other proteins in granulosa cells. Therefore, melatonin decreased the BimEL level by increasing its ubiquitination. Although melatonin was previously regarded as an inhibitor of the ubiquitin-proteasome system [49], our contradictory results may be attributed to different cells with different experimental conditions. Park et al. reported that melatonin could increase the level of BimEL by inhibiting the activity of proteasomes to induce apoptosis of human renal cancer cells [11]. In their experiment, cancer cells were treated with melatonin at the concentration from 0.1 to 1 mM, but the concentration was just 1 nM in our experiment. Above all, according to previous studies, melatonin presented toxicity to the oocyte during maturation when its concentration reached at 10 −5 M [15]. In addition, the effect of melatonin on proteasome activity may be different between cancer cells and granulosa cells. It has been shown that mitochondrial-associated Bcl-2 family proteins, including Bax, Bcl-2, and Bim, are regulated through ubiquitin/proteasomal degradation during apoptosis [50]. Wan et al. reported that APCCdc20 acts as an E3 ubiquitin ligase to promote Bim ubiquitination and destruction [51]. Additional studies are required to delineate the precise mechanisms involved in melatonin-induced protein ubiquitination in granulosa cells.
Generally, melatonin participates in various physiological processes via its receptors or antioxidant properties [52][53][54]. However, our results showed that melatonin downregulated BimEL through neither a receptor-mediated nor antioxidant pathway. Therefore, the detailed mechanism of BimEL downregulation by melatonin in granulosa cells waits further investigation.
A hypothetical model of BimEL regulation by melatonin in granulosa cells is presented in Figure  7. Although there are some details that need to be further pursued, our results demonstrate that melatonin downregulates phosphorylated BimEL via a ubiquitin-proteasome pathway in porcine granulosa cells to maintain follicle health. Future studies should investigate how melatonin precisely regulates protein ubiquitination.

Materials and Methods
All chemicals used in this study were purchased from Sigma Chemical Co. (St. Louis, MO, USA) unless otherwise specified.

Materials and Methods
All chemicals used in this study were purchased from Sigma Chemical Co. (St. Louis, MO, USA) unless otherwise specified.

Classification of Follicles and Measurement of Melatonin in Porcine Follicular Fluid
Follicles were classified as healthy or atretic according to previously established morphological criteria [53]. In brief, healthy follicles had vascularized theca interna and clear amber follicular fluid with no debris. The slightly atretic and atretic follicles had gray theca interna and flocculent follicular fluid of different degrees. The concentration of melatonin in follicular fluid was assessed using the porcine melatonin ELISA kit (Shanghai MLBIO Biotechnology Co., Ltd., Shanghai, China) with a measurement range for melatonin from 1.5-65 ng/L.

In Vitro Maturation of Oocytes
In vitro maturation (IVM) of oocytes was performed as previously described [35]. In brief, porcine ovaries were collected at a local abattoir and transported to the laboratory within 2-3 h after collection. Cumulus oocyte complexes (COCs) were aspirated, and those with several layers of unexpanded cumulus cells were cultured in maturation medium for 42-44 h. The maturation medium was Tissue Culture Medium 199 with Earle's salts (TCM199; Invitrogen, Carlsbad, CA, USA) supplemented with 10% porcine follicular fluid, 10 IU/mL hCG (Chorulon, Intervet Australia Pty Ltd., Victoria, Australia), 10 IU/mL eCG (Folligon, Intervet Australia Pty Ltd.), 10 ng/mL EGF, 0.6 mM cysteine, 75 mg/L penicillin, and 50 mg/L streptomycin. After IVM, cumulus cells were separated from oocytes and lysates were used for Western blotting.

Cell Culture and Treatment
Porcine primary or adherent granulosa cells were cultured as previously described [9]. In brief, porcine ovaries were collected at a local abattoir and transported to the laboratory within 2-3 h after collection. Ovaries were washed thrice with sterile 0.9% saline (37 • C) containing 100 IU/L penicillin and 100 mg/L streptomycin. Granulosa cells were then isolated by puncturing healthy follicles (2-5 mm in diameter) with a 25-gauge hypodermic needle and gently washing thrice with DMEM/F12 supplemented with 1% fetal bovine serum, 100 IU/L penicillin, and 100 mg/L streptomycin. Primary granulosa cells were selected under a microscope for different treatments according to the experimental design. In addition, granulosa cells were isolated by puncturing healthy follicles (2-5 mm in diameter) with a 25-gauge hypodermic needle and gently washing thrice with DMEM containing 10% fetal bovine serum, 100 IU/L penicillin and 100 mg/L streptomycin. Cells were then incubated at 37 • C in a humidified atmosphere of 5% CO 2 /95% air for 24 h. The cells were passaged upon reaching confluence. The culture medium was replaced with DMEM containing 100 IU/L penicillin and 100 mg/L streptomycin at 12 h after passaging, and the cells were cultured for an additional 12 h. Thereafter, the cells were treated with melatonin and/or other compounds for the indicated time periods. The cells were pretreated with 10 µM LY294002, 1 µM cycloheximide, 10 µM choloroquine, 100 nM Bafilomycin A1, or 5 µM MG132 at 1 h before melatonin treatment.

Western Blotting
The cells were lysed in Laemmli sample buffer (Bio-Rad, Hercules, CA, USA). An equal amount of protein was separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE; 12% acrylamide gel), and proteins were transferred to nitrocellulose membranes (Millipore, Billerica, MA, USA). After blocking with 5% non-fat milk Tris-buffered saline containing 0.1% Tween-20 (TBST), the membranes were incubated with primary antibodies overnight at 4 • C. After washing with TBST, the membranes were incubated with the appropriate secondary antibodies conjugated to horseradish peroxidase at a dilution of 1:2000 for 1 h. The protein bands were visualized using an enhanced chemiluminescence detection system (Applygen Technologies Inc., Beijing, China). The western blotting images were processed using Image J software (National Institutes of Health, Bethesda, MD, USA).

Lambda Phosphatase Treatment
Whole-cell extracts (30 µg) were incubated with or without 1 µL of lambda protein phosphatase (400 U/µL; New England BioLabs, Ipswich, MA, USA) for 2 h at 30 • C. The samples were boiled for 5 min after adding 10 µL of 5× SDS sample buffer.

Proteasome Activity Assay
Chymotrypsin-like protease activity was measured using the Amplite™ Fluorometric Proteasome 20S Assay Kit (AAT Bioquest Inc., Sunnyvale, CA, USA) according to the manufacturer's instructions. In brief, 100 µL of proteasome assay loading solution was added to each well after granulosa cells were treated with 10 −9 M melatonin for 3 h and the tissue culture plate was then incubated at 37 • C for 2 h. The assay was performed by monitoring the fluorescence at 480 nm EX/520 nm EM.

Immunoprecipitation
To detect ubiquitinated Bim in whole-cell extracts, granulosa cells were homogenized in RIPA buffer containing a broad-spectrum protease inhibitor cocktail (Roche, Basel, Switzerland). The protein concentrations of the cell extracts were measured, and equal amounts of protein were incubated overnight at 4 • C with a polyclonal Bim antibody (Cell Signaling Technology, Beverly, MA, USA). The antibody-antigen complex was then incubated with protein A/G PLUS-agarose beads (Santa Cruz Biotechnology, Santa Cruz, CA, USA). The immunoprecipitated proteins were resolved by SDS-PAGE (12% acrylamide gel) and processed for Western blotting using specific antibodies to detect ubiquitin and Bim.

Real-Time quantitative PCR
A reverse transcription polymerase chain reaction was used to determine if melatonin regulated the mRNA expression of bim. First-strand cDNA synthesis was carried out with the SuperScript™ First-Strand Synthesis System for RT-PCR (Invitrogen) according to the manufacturer's instructions. The real-time PCR primers for bim were: 5 -AGGCTGAACCCGCAGATA-3 (forward) and 5 -GCATTAAATTCGTCTCCAATACG-3 (reverse). The real-time PCR primers for β-actin were: 5 -ATCGTGCGGGACATAAG-3 (forward) and 5 -CTCGTTGCCGATGGTGAT-3 (reverse). The mRNA expression of bim was normalized to that of the endogenous control β-actin. Real-time PCR reactions were performed using the ABI 7900 Sequence Detection System (Applied Biosystems, Foster City, CA, USA).

Statistical Analysis
Data are presented as the means ± standard deviation (S.D.) of at least three independent replicates. Data were analyzed by one-way and two-way analysis of variance (ANOVA) and Duncan's test using SAS software (SAS Institute, Cary, NC, USA). Differences were considered statistically significant at p-values < 0.05.