Effect of Mst1 on Endometriosis Apoptosis and Migration: Role of Drp1-Related Mitochondrial Fission and Parkin-Required Mitophagy

Endometriosis is a chronic, painful, and debilitating disease in which endometrium-like glandular and stromal cells grow outside the uterine cavity. It is an inflammatory and estrogen-dependent disease that affects 6–10% of women during their reproductive years and up to 50% of women receiving fertility treatments [1, 2]. The endometrium is comprised of stromal cells and epithelial compartments. Notably, epithelial cells normally undergo anoikis, a mechanism of programmed cell death, upon detachment from the extracellular matrix [3]. Accordingly, the survival and migration of endometrial stromal cells (ESCs) outside the uterine cavity is the major cause for the development of endometriosis. Current therapeutic alternatives consist of various hormone treatments, such as GnRH-agonists (GnRH-a), aimed at increasing the apoptotic rate in eutopic and ectopic endometrial cells from women with endometriosis [4].

Mitochondria are vital for cell survival and energy metabolism. In addition, mitochondria are an important target of estrogen. The estrogen receptor-β (ERβ) localizes to the mitochondria in a ligand-dependent or ligand-independent manner. A reduction in mitochondrial function via decreasing circulating estrogen is the standard treatment for endometriosis [5]. These data indicate that mitochondria homeostasis could be considered a therapeutic target for the treatment of endometriosis. Notably, recent studies show that mitochondrial dynamics, especially Drp1-related mitochondrial fission, happen at the early stage of cellular apoptosis, such as in liver cancer, cardiomyocytes and neurocytes [6, 7]. Excessive mitochondrial fission initiates caspase 9-related mitochondrial apoptosis, leading to cell death [8, 9]. However, no evidence is available to establish the role of mitochondrial fission in the progression of endometriosis. Apart from mitochondrial fission, Parkin-required mitophagy is another housekeeper of mitochondrial homeostasis. A growing body of evidence suggests that cell migration undergoes a drastic metabolic reprogramming that requires the well-orchestrated mitophagy to remove the damaged mitochondria and sustain the mitochondrial quantity and quality [10, 11]. However, an imbalance of mitochondrial degradation processes results in alteration of cellular homeostasis, which, in turn, may limit the migration and survival of ESCs outside the uterine cavity. Based on this information, we hypothesized that mitochondrial fission and mitophagy might be involved in the development of endometriosis. If this is the case, then the molecular signaling that links fission and mitophagy to ESCs apoptosis and migration should be explored [12].

The Hippo pathway was first discovered through genetic screens in Drosophila, and many components of the Hippo pathway are highly conserved from Drosophila to mammals [13], including mammalian STE20-like kinase 1/2 (MST1/2), salvador homolog 1 (SAV1), large tumor suppressor 1/2 (LATS1/2), Yes-associated protein (YAP) and its paralog, transcriptional co-activator with PDZ-binding motif (TAZ), all of which form a kinase cascade [14]. As a transcriptional co-activator, YAP or TAZ could combine with the TEA domain family member 1 (TEAD1) in the nucleus to promote the expression of target genes. Recent studies reveal that the Hippo pathway, especially Mst1, is implicated in the regulation of tissue regeneration, adult stem cells proliferation and cancer development [15, 16]. However, the role of Mst1 in endometriosis remains unknown. Notably, careful studies from several researchers indicate that the Hippo-pathway plays a role in mitochondrial function via Bcl-xL[17, 18], demonstrating the possible relationship between Mst1 and mitochondrial homeostasis. Furthermore, accumulating evidence describes the regulatory action of Mst1 on p53. Thus, together with the confirmed role of p53 in regulating mitochondrial fission and mitophagy [19], this evidence raises the possibility that Mst1 may govern fission and mitophagy via p53.

Given the important role of Mst1 in tissue regeneration, cancer development and mitochondrial homeostasis, the regulatory machinery of Mst1 activation has gained much attention. MiR-181c is identified as the transcriptional repressor of Mst1. The overexpression of miR-181c induces the inactivation of Mst1, leading to pancreatic cancer cell survival and migration. However, whether miR-181c is also involved Mst1 regulation in endometriosis is incompletely understood. Accordingly, this study was undertaken to explore the effect and mechanism of Mst1 in the etiology of endometriosis, particularly focusing on Drp1-related mitochondrial fission, Parkin-required mitophagy, the p53 pathway and the upstream regulatory signaling for Mst1.

The present study was conducted in accordance with the Declaration of Helsinki and the guidelines of the Ethics Chinese PLA general Hospital, Beijing, China. The all experimental protocol was approved by Ethics Committee of Chinese PLA general Hospital, Beijing, China.

The human endometrial stromal cells (ESCs) were purchased from the American Type Culture Collection (ATCC) Company (Manassas, VA, USA) and cultured in RPMI medium (Thermo Scientific, Rockford, IL) supplemented with 10% fetal bovine serum (FBS, HyClone, USA), 1% L-glutamine and 0.5% gentamycin (Sigma, Cat. No. G1272) at 37 °C in an atmosphere of 5% CO₂ [20]. To inhibit mitochondrial fission, Mdivi1 was used for 2 hours. To reduce or enhance the activity of miR-181c, the pharmacological blocker and mimic of miR-181c were used, respectively.

Cell migration was analyzed using a transwell chamber assay (24-wells, 8-µm pore size with polycarbonate membrane) as previously described [21]. For the scratch assay, cells were cultured in a serum-free medium for 24 h and then disrupted with pipette tips. Wound healing was observed for 48 h, and photographs were taken every 24 h [22].

For the analysis of the protein levels, cells were lysed with RIPA supplemented with PMSF. The protein concentration was analyzed using the BCA protein assay. The protein (50 µg) was separated by 10% SDS-PAGE and then transferred to PVDF [23]. The membranes were blocked with 5% nonfat milk for 1 h at room temperatures and then incubated with primary antibodies: GAPDH (1: 2000), caspase3 (1: 2000), XIAP (1: 1000), caspase9 (1: 1000), survivin (1: 1000), Bax (1: 1000), Bad (1: 1000), Parkin (1: 1000), p62 (1: 1000) and LC3II (1: 1000) from Cell Signaling Technology, U.S.A.; Beclin1 (1: 1000), Atg5 (1: 1000), p53 (1: 1000), VDAC (1: 1000) and F-actin (1: 1000) from Abcam, USA) overnight at 4°C. The membranes were washed in TBST for 15 min and were then incubated with horseradish peroxidase-conjugated secondary antibody (Santa Cruz Biotechnology, U.S.A.) for 1 h at room temperature. Blots were detected through an enhanced chemiluminescence (ECL) substrate kit (Thermo USA) [24].

The mitochondrial transmembrane potential was analyzed using a JC-1 Kit (Beyotime, Shanghai, China) according to the manufacturer’s protocol [25]. The cellular ATP levels were measured using a firefly luciferase-based ATP assay kit (Beyotime). The opening of the mPTP was visualized as a rapid dissipation of tetramethylrhodamine ethyl ester fluorescence as described in a previous study [26].

TUNEL assay was used to detect the apoptosis of ESCs, per the manufacturer’s instructions [27]. TUNEL staining was performed with fluorescein-dUTP (Invitrogen, U.S.A.) for apoptotic cell nuclei, and DAPI was used to stain all cell nuclei. Cells in which the nucleus was stained with fluorescein-dUTP were defined as TUNEL positive. The slides were then measured under a confocal microscope [28].

Caspase3 activity kit (Beyotime Institute of Biotechnology, China) was used to detect the activity of caspase3 according to the manufacturer’s protocol. The assay was repeated 3 times. MTT assay was used to examine cell viability [29]. ESCs were seeded in a 96-well plate. Then, 20 µl MTT (5 mg/ml) was added to the medium for 4 h. Next, the supernatant was discarded, and 100 µl DMSO was added to each well for 10 min. Finally, the optical density (OD) was measured at A490 nm [30].

After treatment, the cells in all groups were firstly washed under PBS for about three times, and then permeabilized for 15 min at with 0.1% Triton X-100 at the room temperature. Then, the cells were washed three times in PBS again [31]. Subsequently, the primary antibodies were used to label the cellular proteins overnight at 4°C. Then, cells were washed under PBS, which were loaded with secondary antibodies for about 1 h at room temperature [32]. The primary antibodies used in the present study were as follows: Tomm20 (1: 500, Abcam, #ab78547), LAMP1 (1: 200, Abcam, #ab24170), HtrA2/Omi (1: 500, Abcam, #ab133504), p53 (Ser15) (1: 500, Cell Signaling Technology, #9284), Mst1 (1: 500, Abcam, #ab3336), Parkin (1: 500, Abcam, #ab15954), F-actin (1: 500, Abcam, #ab205), Tubulin (1: 500, Abcam, #ab18207), LC3II (1: 500, Abcam, #ab48394) and Drp1 (1: 500, Abcam, #ab184247) [33].

Cells were washed with cold PBS and incubated on ice in lysis buffer (Beyotime, China) for 30 minutes. The cells were subsequently scraped, and the homogenates were spun at 800 x g for 5 minutes at 4°C. The supernatants were centrifuged at 10, 000 x g for 20 min at 4°C to acquire the pellets, which were spun again [6, 24]. The final pellets were suspended in lysis buffer containing 1% Triton X-100 and were noted as mitochondrial-rich lysate fractions [34].

Mitochondrial respiration was initiated by adding glutamate/malate to a final concentration of 5 and 2.5 mmol/L, respectively. State 3 respiration was initiated by adding ADP (150 nmol/L); state 4 was measured as the rate of oxygen consumption after ADP phosphorylation. The respiratory control ratio (state 3/state 4) and the ADP/O ratio (number of nmol ADP phosphorylated to atoms of oxygen consumed) were calculated as previously described [35, 36].

The contents of [Ca²⁺]c was imaged with Fluo-2 (Molecular Probes). Samples were then directly examined by confocal microscopy using the ×40 1.42 NA oil immersion objective. For quantification of the concentrations change of [Ca²⁺]c, flow cytometry was used [37]. For analysis of [Ca²⁺]m, Rhod-2 (Molecular Probes) was used and the images were captured by confocal microscopy. Fluorescence intensity of Furo-2 and Rhod-2 was measured by excitation wavelengths of 340 and 550 nm, and emission wavelengths of 500 and 570 nm, respectively. Data (F/F0) were obtained by dividing fluorescence intensity (F) by (F0) at resting level (t = 0) which was normalized by control groups [9, 19].

After treatment, cellular RNA was isolated via Trizol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. Reverse transcription of RNA was performed with Transcriptor First Strand cDNA Synthesis Kit (Roche Applied Science, Switzerland) [38]. About one µg of RNA was reversely transcribed into cDNA for each sample. Synthesized cDNA was amplified on the LightCycler® 480 System with SYBR Green I Master (Roche Applied Science, Mannheim, Germany) [39]. The cycling conditions were as follows: 95°C for 10 min, followed by 40 cycles of 95°C for 15 sec and 72°C for 35 sec, for telomere PCR. The experiments were repeated three times with triplicates of each sample. Fold change of Mst1 and Parkin mRNA expression was normalized by GAPDH as an internal control [40].

To carry out the gain-of function assay about Mst1 and Parkin, we overexpressed the Mst1 and Parkin in ESCs via adenovirus-based technology. Firstly, the pCDH-mCMV-Mst1 plasmid was purchased from Vigene Bioscience. Then, the plasmid was transfected into 293 T cells via lipofectamine 2000 for about 48-96 hours in an incubator with 5% CO₂ at 37°C [41]. Subsequently, collection of the supernatant to obtain the adenovirus which was identified via PCR. After amplification of the obtained Mst1-adenovirus (Ad-Mst1), the ESCs were transfected by Ad/Mst1. The transfection and overexpression efficiency were measured via western blots.

To evaluate the functional role of p53, Mst1 and HtrA2/Omi, siRNAs were used to reduce their expression. The selective siRNAs were purchased from Yangzhou Ruibo Biotech Co., Ltd. After treatment, cells were washed under Opti-Minimal Essential Medium without serum or antibiotics [42]. Then, the cells were cultured for about 1-2 days to 30-40% confluence. Then, the siRNA was transfected into cells via Lipofectamine 2000 transfection reagent according to the manufacturer’s protocol [43]. After 48-96 hours, the cells were collected and the knockdown efficiency were determined via western blot analysis [44].

All results were expressed as mean±SD. We performed statistical analysis using one-way ANOVA test. A value of P< 0.05 was considered statistically significant.

To investigate the functional relevance of Mst1 during endometriosis, we performed qPCR and western blots to detect the level of Mst1 in ectopic and normal endometrium from women with endometriosis. In this study, 8 cases of tissue samples were collected and detected using Mst1 primers or an anti-Mst1 antibody. The Mst1 mRNA and protein expressions were significantly lower in the ectopic endometrium compared to the normal endometrium (Fig. 1A-C). Furthermore, we conducted immunohistochemistry to observe the change in Mst1. Mst1 immune expression was distributed strongly in the nucleus and cytoplasm of the epithelial and stromal cells in the normal endometrium, whereas Mst1 protein expression was weak in the ectopic endometrium (Fig. 1D-E). These results suggested that Mst1 was downregulated in endometriosis.

Based on the poor expression of Mst1 in endometriosis and its well-established ability to regulate cell apoptosis and migration, we hypothesized that Mst1 may control the survival and mobilization of endometrial cells. To test this hypothesis, we performed an adenovirus-based overexpression (Ad/Mst1) of Mst1 in a human endometrial stromal cell line (ESCs). The overexpression efficiency is shown in Fig. 1F-H. Then, the transfected ESCs were subjected to MTT and TUNEL assays. The data showed that the Ad/Mst1-ESCs had a reduced cellular viability compared to the control group (Fig. 1I-K). Moreover, we also conducted a transwell assay to evaluate the role of Mst1 on ESCs migration. As illustrated in Fig. 1L-M, the overexpression of Mst1 alleviated the number of migrated cells when compared to the control group. Altogether, the above data demonstrated that Mst1 was dysregulated in ectopic endometrium, and the recovery of Mst1 expression impaired ESCs migration and viability.

Mitochondrial fission is necessary for cell apoptosis and migration. In our experiments, we used the RFP-mitochondria probe to label mitochondria and observed the mitochondrial fission under Mst1 overexpression. As shown in Fig. 2A, compared to the control group, the Ad/Mst1-treated ESCs had more mitochondrial fragmentations. However, this tendency was inhibited by Mdivi1, an inhibitor of fission. To quantify the fission, we measured the length of the mitochondria. The average mitochondrial length in the control group was 7.91±2.13 µm (Fig. 2B). However, the overexpression of Mst1 reduced the mitochondrial length to 2.09±0.96 µm. Notably, Mdivi1 reversed the length of the mitochondria despite transfection with Ad/Mst1.

Given that fission is activated by Drp1, which shuttles from the cytoplasm on the surface of the mitochondria in the process of mitochondrial fission, we therefore assessed whether Mst1 regulates the Drp1 translocation on mitochondria. First, western blots was used to detect the cellular location of Drp1. Ad/Mst1 enhanced the expression of mito-Drp1 and reduced the content of cyto-Drp1 compared to the control group (Fig. 2 C-E). These phenotypic changes were rescued by Mdivi1. To provide more solid evidence for the Drp1-related mitochondrial fission, we co-stained the mitochondria and Drp1 at the same time. As shown in Fig. 2F-G, Ad/Mst1 stimulated the overlap of Drp1 and mitochondria. Moreover, in the amplified panel of Fig. 2G, the fragmented mitochondria were labeled by more Drp1. Notably, Mdivi1 blocked the action of Ad/Mst1 on Drp1-related mitochondrial fission.

To determine whether Mst1-mediated fission was associated with ESC apoptosis and migration inhibition, we first evaluated the mitochondrial apoptosis pathway. Mitochondrial apoptosis is characterized by the mitochondrial potential dissipation, mPTP opening and pro-apoptotic factor liberation from the mitochondria into the cytoplasm, which are followed by the activation of caspase 9 and caspase 3. As shown in Fig. 3A-B, Ad/Mst1 reduced the membrane potential compared to the control group. Moreover, Ad/Mst1 also increased the mPTP opening rate (Fig. 3C). However, the inhibition of fission via Mdivi1 reversed the mitochondrial potential and reduced the mPTP opening.

As a consequence of the mPTP opening, the mitochondria-contained pro-apoptotic factors (cytochrome c, second mitochondria-derived activator of caspases and HtrA2/Omi) are released into the cytoplasm and even into the nuclear space, where they interact with and activate caspase 9/3. Through an immunofluorescence assay of HtrA2/Omi, we found that Ad/Mst1 evoked HtrA2/Omi diffusion into the cytoplasm and even into the nucleus (Fig. 3D-E). This effect was concomitant with an increase in the expression of the down-stream pro-apoptotic proteins (Fig. 3F-L). However, Mdivi1 cancelled these effects. Altogether, the above data confirmed that Mst1-mediated fission was responsible for ESCs apoptosis via the activation of the caspase 9-involved mitochondria apoptosis pathway.

Notably, several studies have reported that HtrA2/Omi inhibits cancer migration via impairing F-actin-based lamellipodium. The arrangement and dynamic behavior of F-actin filaments supply the protrusive force for cell movement via forming lamellipodium, which drives cell migration in some direction. To explore whether Mst1-mediated fission blunted ESCs migration via HtrA2/Omi, siRNA technology was used to silence HtrA2/Omi in the Ad/ Mst1-treated ESCs. Then, the wound healing assay was utilized to evaluate ESCs mobilization. First, the Ad/Mst1-treated ESCs showed reduced mobility compared to the control group (Fig. 3M-N). However, the knockdown of HtrA2/Omi reversed the capacity of the Ad/Mst1-treated ESCs to migrate. Considering the important role of F-actin-based lamellipodium in ESC migration, an immunofluorescence assay of F-actin was carried out. In addition, to demonstrate whether F-actin was specifically regulated by Mst1 and fission, we also co-stained tubulin, another element of the cytoskeleton, which makes up the microtubule and promotes intracellular transport but has no role in cellular migration. As shown in Fig. 3O-Q, after the overexpression of Mst1, F-actin was unclear, inordinate and fragmented, which was accompanied by an obvious decline in the ratio of F-actin to tubulin, and this was indicative of F-actin degradation. Furthermore, Ad/Mst1 also reduced the length of the lamellipodium (Fig. 3O-Q). However, these configuration changes were rescued by the knockdown of HtrA2/ Omi or application of Mdivi1. Notably, no changes were observed in the tubulin structure. These data indicated that Mst1 governed the lamellipodium-based migration of ESCs in a fission-dependent manner.

Apart from fission, mitophagy is also involved in mitochondrial quantity and quality control, which have the ability to block cellular apoptosis via the removal of damaged mitochondria. Therefore, we wanted to know whether Mst1 played a role in mitophagy. Western blot assays interrogating the mitophagy parameters revealed that Ad/Mst1 reduced mito-LC3II, Beclin1, p62 and Atg5 (Fig. 4A-F). Moreover, Ad/Mst1 also strongly repressed the expression of Parkin, which is a receptor or activator of mitophagy (Fig. 4A-B). To provide more solid evidence for the action of Mst1 on mitophagy, we co-stained the mitochondria and lysosome at the same time. As shown in Fig. 4G-H, in the control cells, spindle mitochondria were partly tagged by lysosomes. However, after the overexpression of Mst1, the mitochondria divided into several fragmentations, which were separated from the lysosomes, indicative of mitophagy inhibition. Considering the inhibitory role of Mst1 on Parkin expression, we conducted a Parkin rescue experiment using an adenovirus-based overexpression (Ad/Parkin) of Parkin. The transfection efficiency is shown in Fig. 4A-B. As expected, re-introduction of Parkin promoted the lysosome interaction with the mitochondrial debris, which also sustained the network of the mitochondria. These data indicated that Mst1 controlled mitophagy via Parkin.

Mitophagy arrest results in mitochondrial dis-homeostasis. Energy production is the central role of the mitochondria, in our study, we found that Mst1-mediated mitophagy inhibition caused a significant drop in ATP production, which was reversed by Parkin overexpression (Fig. 5A). In addition to the energy disorder, mitophagy suppression was also associated with cellular oxidative stress as evidenced by higher ROS production (Fig. 5B-C). This effect was concomitant with a decline in the contents of anti-oxidant factors, such as SOD and GSH (Fig. 5D-F). In contrast, the lipid peroxide MDA increased under Mst1 overexpression, whereas regaining Parkin reversed the redox balance. These data indicated that Mst1-mediated mitophagy suppression contributed to cellular injury, which might be involved in ESC apoptosis.

As for the role of mitophagy in ESC mobilization, we focused on the balance of cytoplasmic calcium ([Ca²⁺]c) and mitochondrial calcium ([Ca²⁺]m). Cellular migration depends on a moderate [Ca²⁺]c concentration. A [Ca²⁺]c overload hinders cellular contraction and mobilization. Thus, we first detected the synchronous change in [Ca²⁺]m and [Ca²⁺]c fluorescence in Ad/Mst1-treated ESCs. As shown in Fig. 5G, Ad/Mst1 elevated the fluorescence intensity of [Ca²⁺]m and [Ca²⁺]c compared to the control group. However, the recovery of Parkin reduced [Ca²⁺]m and [Ca²⁺]c fluorescence. Furthermore, the calcium map of [Ca²⁺]c was also in agreement with the above findings that Mst1 amplified the calcium signal in a mitophagy-dependent manner (Fig. 5H-I). At last, we observed a change in F-actin, which is a necessary element for cell migration. As shown in Fig. 5J-K, Ad/Mst1 reduced the F-actin expression, which was reversed by Parkin overexpression, suggesting that mitophagy maintained the F-actin homeostasis.

To explain the mechanism by which Mst1 manipulates fission and mitophagy, we focused on p53. The Mst1-Hippo pathway is the activator of p53, which regulates fission and mitophagy. In the current study, we found that Mst1 overexpression elevated the expression of phosphorylated p53 at Ser15 (Fig. 6A-B), which was indicative of p53 activation. This change was accompanied with an increase in mito-Drp1 and a drop in cyto-Drp1 (Fig. 6A-C). However, with the knockdown of p53 via siRNA, the expression of mito-Drp1 was reduced and cyto-Drp1 was increased (Fig. 6A-C). These data indicated that Mst1 regulated Drp1 via p53. Furthermore, Drp1 translocation to the mitochondria was determined via a post-transcriptional modification. Drp1 has two key phosphorylation sites, including Ser616 and Ser637. Drp1 is activated by phosphorylation at Ser616 but is inactivated by phosphorylation at Ser637. The balance of these two phosphorylation sites determines the activity of Drp1. We found that Ad/Mst1 primarily induced the phosphorylation of Drp1 at Ser616. (Fig. 6A, D-E). However, this effect was nullified by p53 silence. Notably, no change was observed in p-Drp1^Ser637 regardless of Mst1 overexpression or p53 knockdown (Fig. 6A, D-E). These data indicated that p53 controlled Drp1 activity via a post-transcriptional modification.

To further explain whether p53 controlled Parkin expression, we focused Parkin gene transcription. We found that Parkin mRNA expression decreased in response to Mst1 overexpression (Fig. 6F) but was increased with the knockdown of p53. These data indicated that p53 was the transcriptional repressor of Parkin. Furthermore, we conducted a luciferase assay of Parkin with Mst1 overexpression and p53 knockdown. As shown in Fig. 6G, Mst1 repressed the transcriptional activity of Parkin, which was enhanced by p53 silencing.

To determinee whether p53 was also involved in fission and mitophagy, we used co-immunofluorescence to stain the mitochondria and LC3II. As shown in Fig. 6H-J, in the control group, the spindle mitochondria are partly marked by dotted or aggregated LC3II, indicative of mitochondrial fission. However, Mst1 overexpression induced mitochondrial mass division into several fragmentations. In addition, Ad/Mst1 also inhibited LC3II accumulation, which is indicative of autophagy flux arrest. However, the knockdown of p53 maintained the mitochondrial network and reversed the LC3II accumulation. These data confirmed that Mst1 controlled fission and mitophagy via p53. At the molecular level, p53 regulated Drp1 and Parkin at the post-transcriptional and mRNA levels, respectively.

Lastly, we investigated the upstream signaling for Mst1 downregulation. A recent study suggests that mi-R181c [45] is the transcriptional repressor of Mst1 in pancreatic cancer. In the present study, we found that miR-181c was also increased in the ectopic endometrium compared to the normal endometrium (Fig. 7A). Upon inhibition of miR-181c via its inhibitor, Mst1 expression was reversed (Fig. 7B-C). Notably, the miR-181c mimics slightly limited the expression of Mst1 in the ESCs (Fig. 7B-C). However, no significant difference was observed between the miR-181c mimics and the control group. These data indicated that in the normal ESCs, Mst1 was strongly inhibited by miR-181c, and thus, the application of miR-181c mimics could not further reduce the expression of Mst1.

Furthermore, we wanted to understand the regulatory signaling for the miR-181c upregulation. A recent study identified that endometriosis progression was associated with impaired endometrial decidualization, which is regulated by the orphan nuclear receptor subfamily 4 group A (NR4A). The decreased expression of NR4A disrupted endometrial decidualization, representing a novel mechanism to explain subfertility, adenomyosis and endometriosis. In the present study, we found that the NR4A agonist cytosporone B (CsnB) reduced the expression of miR-181c (Fig. 7D), suggesting that NR4A plays a role in inhibiting the expression of miR-181c. Through immunofluorescence, we found that the activation of NR4A increased the expression of Mst1 (Fig. 7E), similar to the results in the miR-181c inhibition, suggesting that NR4A is the upstream activator of Mst1 via repressing miR-181c.

Next, to determine whether NR4A/miR-181c is involved in mitophagy and fission, we co-stained Parkin. As shown in Fig. 7E, the inhibition of miR-181c or the activation of NR4A via CsnB reduced the expression of Parkin. However, these effects were inhibited by the knockdown of Mst1 via siRNA, suggesting that NR4A/miR-181c regulation of Parkin required Mst1. Similarly, the miR-181c inhibitor or the NR4A agonist induced more mitochondrial fragmentations, and these effects were nullified by Mst1 siRNA (Fig. 7F-G).

Furthermore, we also found that the miR-181c inhibitor or the NR4A agonist induced F-action dysfunction and impaired lamellipodium formation (Fig. 7H-J), which were reversed by Mst1 siRNA. Similarly, the caspase 3 activity assay in Fig. 7K also suggested that NR4A/ miR-181c influenced ESC apoptosis via Mst1.

Ample evidence suggests the involvement of Mst1 in the inhibition of hepatocellular carcinoma metastasis, pancreatic cancer relapse, gastric cancer survival, cardiac reperfusion injury and metabolic liver injury. However, little is known about the role of Mst1 in the protection of endometriosis. In the present study, we found the following: (1) Mst1 was significantly downregulated in the ectopic endometrium of endometriosis; (2) the overexpression of Mst1 was closely associated with the inability of ESCs to migrate and survive; (3) mechanistically, regaining Mst1 powerfully enhanced Drp1 post-transcriptional phosphorylation at Ser616 and repressed Parkin transcription activity via p53, leading to mitochondrial fission activation and mitophagy inhibition; (4) Drp1-related fission greatly induced ESC apoptosis and migration arrest in a HtrA2/Omi-dependent manner; (5) additionally, defective Parkin-required mitophagy dramatically evoked cellular oxidative stress, energy metabolism and calcium overload; (6) through excessive mitochondrial fission and aberrant mitophagy, Mst1 launched caspase 9-related mitochondrial apoptosis and abrogated F-actin/lamellipodium-dependent cellular migration; and (7) finally, we identified that NR4A/miR181c was the upstream signal for the Mst1 dysfunction in endometriosis. As far as we know, this study comprehensively described the important role of the NR4A-miR181c-Mst1 pathway in endometriosis apoptosis and migration via regulation of Drp1-related mitochondrial fission and Parkin-required mitophagy for the first time (Fig. 8).

Endometriosis, a benign gynecological disease defined as the presence of endometrial cells outside the uterus, seriously impacts the quality of life and reproductive ability of 6–10% of reproductive-aged women and is strongly associated with chronic pelvic pain and infertility. The endometrium of the uterus is lined by a columnar epithelium that is supported by a stromal cell foundation, which consists of a variety of cell types [46]. The endometrial epithelial cells and ESCs are the main two cell types, as they have the absolute predominance in quantitative terms. The ESCs actively produce numerous growth factors and cytokines that affect epithelial function. According to Sampson’s hypothesis, ESCs migrate to and survive in ectopic locations, which is the key feature of endometriosis. Accordingly, the molecular mechanism of ESCs survival and mobilization should be considered, with a goal of discerning the rationale for therapeutically targeting endometriosis

In the present study, we found that Mst1 dysregulation was responsible for the excessive survival and migration of ESCs. Mst1, a critical regulator of the hippo pathway, is closely related to evolutionarily conserved serine/threonine protein kinases that play a pivotal role in cell proliferation, survival, morphology and motility and organ size [47]. Recent studies in conditional Mst1 ablation and transgenic Mst1/2 double-knockout mice exhibit abrogation of Yap phosphorylation with intense nuclear accumulation, leading to the development of hepatocellular carcinoma, pancreatic cancer, cardiac reperfusion injury and metabolic liver disease [48, 49]. However, our data introduced Mst1 into the etiology of endometriosis. We demonstrated that Mst1 was downregulated in the ectopic endometrium, which negatively correlated with ESC survival and migration. Once Mst1 activity was recovered in the ESCs, the ability of the ESCs to survive and mobilize was impaired. These data indicate that Mst1 acts as a negative regulator of endometriosis, dictating ESCs to excessive apoptosis and migration inhibition. Therefore, our findings unveil the neglected role of Mst1 in endometriosis, highlighting a novel therapeutic target of endometriosis, involving ESC apoptosis and migration. However, more clinical evidence is needed to support our notion.

One mechanism by which Mst1 impacts ESCs apoptosis and migration is through the activation of Drp1-related mitochondrial fission by p53 [50]. p53 amplifies Drp1 activity via a post-transcriptional modification at Ser616. Excessive Drp1-related fission forces the mitochondria to liberate HtrA2/Omi into the cytoplasm where HtrA2/Omi activates the caspase 9-related mitochondrial apoptosis pathway. Meanwhile, HtrA2/Omi abolishes ESC migration by minimizing the stability of F-actin, leading to lamellipodium collapse. Another critical regulatory mechanism of Mst1 in ESC apoptosis/migration is Parkin-required mitophagy. Mst1 blocks Parkin transcription and expression via p53, resulting in a mitophagy delay. The worn-out mitophagy fails to remove the damaged or fragmented mitochondria, evoking cellular oxidative stress, energy disorder and a calcium overload. Thus, through excessive fission and deficient mitophagy, Mst1 enhanced ESCs apoptosis and limited its migration.

The concept that fission or mitophagy modulates cell apoptosis is supported by a plethora of studies [8, 51, 52]. With respect to cell migration, several researchers reasoned that fission requires F-actin to form the potential contraction point at the surface of the mitochondria, which extensively consumes F-actin stress fibers, a necessary element for cell migration [53]. Moreover, oxidative stress and energy deletion, as a result of bad-structured mitophagy or fission, have the ability to induce the F-actin degradation via the regulation of cofilin, a depolymerization agent of F-actin [52]. However, in the present study, we identified that HtrA2/O mi was the pro-apoptotic signal for ESC apoptosis induced by fission. Meanwhile, HtrA2/Omi induced F-actin collapse, which impaired the formation of lamellipodium. HtrA2/Omi is a serine protease and chaperone protein, and it binds to and cleaves the inhibitor of apoptosis proteins (IAPs), leading to cellular apoptosis [54]. Additionally, previous studies have also reported that the F-actin cytoskeleton is regulated by p53 via the activation of HtrA2/Omi [55]. These findings are in agreement with our results. Importantly, our data established the links between Mst1, Drp1-related mitochondrial fission and Parkin-required mitophagy with respect to cell migration. To the best of our knowledge, this is the first study to explore the interactions of the Mst1-Hippo pathway and mitochondrial dynamics [56].

Finally, we found that Mst1 dysregulation was controlled by NR4A and miR181c. Nuclear receptors represent a family of transcription factors responsible for the regulation of many intracellular pathways, such as cancer, metabolic and proliferative diseases [57, 58]. They are termed orphan because their endogenous ligands have not yet been identified. The orphan nuclear receptor subfamily 4 group A (NR4A) is progressively decreased in adenomyosis, which impairs endometrial decidualization [59]. In the present study, we found that the activation of NR4A reversed Mst1 expression via miR-181c, suggesting that NR4A/miR-181c could be involved in the development of endometriosis. Ample evidence identifies miR-181c as the transcriptional repressor of Mst1 [45]. Moreover, numerous studies have profiled the role of miR-181c in mitochondrial protection [60-62]. Furthermore, a high expression of miR-181c is considered a predictive marker of the recurrence in colorectal cancer, suggesting an inhibitory effect of miR-181c on cell survival and proliferation. These findings are in accordance with our results. However, the question of how NR4A regulates miR-181c expression remains unanswered. More studies are needed to explore the underlying mechanism.

In summary, our report illustrates that Mst1 functions as a novel endometriosis suppressor involving ESC viability and mobility. Mechanically, Mst1 is triggered by NR4A/ miR-181c and regulates Drp1-related mitochondrial fission and Parkin-required mitophagy via p53. Thus, strategies to regulate the balance of the Mst1-Hippo pathway, Drp1-related mitochondrial fission and Parkin-required mitophagy could be a therapeutic target with respect to endometriosis.

This study was funded by the National Natural Science Foundation of China (Grant Number: 81571411). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

The authors have declared that they have no conflicts of interest.