The activation of M2 muscarinic receptor inhibits cell growth and survival in human epithelial ovarian carcinoma

Ovarian cancer is the fifth leading cause of cancer‐related deaths in females. Many ovarian tumor cell lines express muscarinic receptors (mAChRs), and their expression is correlated with reduced survival of patients. We have characterized the expression of mAChRs in two human ovarian carcinoma cell lines (SKOV‐3, TOV‐21G) and two immortalized ovarian surface epithelium cell lines (iOSE‐120, iOSE‐398). Among the five subtypes of mAChRs (M1–M5 receptors), we focused our attention on the M2 receptor, which is involved in the inhibition of tumor cell proliferation. Western blot analysis and real‐time PCR analyses indicated that the levels of M2 are statistically downregulated in cancer cells. Therefore, we investigated the effect of arecaidine propargyl ester hydrobromide (APE), a preferential M2 agonist, on cell growth and survival. APE treatment decreased cell number in a dose and time‐dependent manner by decreasing cell proliferation and increasing cell death. FACS and immunocytochemistry analysis have also demonstrated the ability of APE to accumulate the cells in G2/M phase of the cell cycle and to increase the percentage of abnormal mitosis. The higher level of M2 receptors in the iOSE cells rendered these cells more sensitive to APE treatment than cancer cells. The data here reported suggest that M2 has a negative role in cell growth/survival of ovarian cell lines, and its downregulation may favor tumor progression.


| INTRODUCTION
Ovarian cancer is the fifth leading cause of cancer-related deaths in females. Greater than 95% of ovarian cancers originate in the epithelial cells on the ovary surface (OSE), a derivative of the embryonic coelomic epithelium that takes part in the cyclical ovulatory ruptures and repair.
The mechanisms of epithelial ovarian cancer development have been long debated, and different hypotheses have been proposed. The "Gonadotropin hypothesis" proposes that the hormonal environment of the OSE, as high gonadotropin levels, can have a role in promoting neoplastic transformation by regulating the proliferative activity of OSE cells. 1,2 The high gonadotropin levels found in cysts and peritoneal fluid from ovarian cancer patients support this hypothesis. 3 The second one is the "incessant ovulation hypothesis," first proposed in 1971, 4 according to which repeated episodes of ovulationassociated injury and repair of the OSE cells could induce mutations in these cells and later malignancies.
The ovulatory process itself resembles an inflammatory reaction, with leukocytic infiltration, release of inflammatory cytokines, vasodilatation, DNA repair, and tissue remodeling. Several inflammatory factors, such as cytokines secreted by the infiltrating leukocytes, have been implicated in ovarian carcinogenesis. Epidemiological evidence suggests that ovarian cancer may be related to chronic inflammatory processes (e.g., endometriosis, pelvic inflammatory disease, or mumps viral infection). The alteration in the expression of genes associated with inflammation 5 and the reduction of local inflammation (e.g., tubal ligation and hysterectomy) appear to reduce the risk of ovarian cancer. Moreover, the lower incidence of ovarian cancer has been associated with women taking anti-inflammatory medications. 6 Various studies have reported that the cholinergic system, with its components acetylcholine (ACh), nicotinic and muscarinic receptors (mAChRs), is expressed not only in the brain but can be found in many different organ systems such as cardiovascular, respiratory, digestive systems, and reproductive tracts. ACh's action is mediated by mAChRs, G-protein coupled receptors expressed in neuronal and nonneuronal tissues. 7,8 Five muscarinic subtypes have been cloned in different animal species (M1-M5). These receptors present a high homology degree but significant variability in the carboxy-and amino-terminals and the third cytoplasmic loop. They are involved in controlling cell proliferation by activating pathways such as IP3K and MAPK/ERK kinases.
Numerous groups have described that ACh, via mAChRs, can modulate migration, proliferation, and angiogenesis in several tumor types such as colon, ovarian, lung, breast cancer, 9 melanoma, 10 neuroblastoma 11 and glioblastoma, [12][13][14][15] and inflammatory processes in the lung. 16 In the ovary, studies by Mayerhofer and co-workers 7,8 have demonstrated the presence of the cholinergic autocrine system. Granulosa cells of antral follicles in the rhesus monkey and human ovary contain choline acetyltransferase and produce ACh under follicle stimulating hormone (FSH) stimulation, suggesting cholinergic autocrine/paracrine signals regulating the ovarian function. Moreover, granulosa cells express M1 and M5 receptor subtypes, while the oocytes express the M3 subtype. 17 In the ovarian tumors, a correlation was observed 18 between positive mAChR status of ovarian tumor cell lines and reduced survival time of the patients from whom the tumor cells were derived. Initial studies 19 showed mAChRs in ovarian tumors and tumor cell lines, and the binding profile suggested the presence of M3 receptor subtype. Altogether, these results suggest a significant correlation between high levels of mAChRs and an adverse prognosis of patients with ovarian carcinoma.
Although these studies highlight the influence of mAChRs in ovarian cancer progression, the mAChR subtypes' expression in ovarian cancer cells has been poorly investigated.
The present study aimed to investigate better the mAChR subtypes' expression in ovarian cancer cells and nontumoral iOSE cells. We focused our attention on the M2 receptor, a receptor involved in inhibiting tumor cell proliferation in other cancer cell lines and whose function has never been studied in ovarian cancer.  15,20 mAChR antagonists were used at the final concentration of 10 −6 M for Gallamine (M2 antagonist; CliniSciences), 10 −7 M for Pirenzepine (M1 antagonist; Sigma-Aldrich), and 10 −8 M for 4-DAMP (M3-M5 antagonist; Sigma-Aldrich). 15 The cells were preincubated for 2 h with the antagonists before APE treatment. Cell lines were also treated with Muscarine chloride hydrate (Sigma-Aldrich) an agonist of all mAChRs, at the final concentration of 100 µM, according to previous studies. 21

| Cell growth
To analyze cell growth, we plated the cells in 12-well plates at a density of 3 × 10 4 cells/well. Twenty-four hours after plating, the cells were treated with increasing APE concentrations for 1-3 days. At the end of the treatments, adherent and floating cells were combined and stained with Trypan Blue. Viable cells were evaluated using a hemocytometer.

| Evaluation of cell death by propidium iodide (PI) staining
The cells were plated in 35-mm dishes at a density of 7 × 10 4 cells/dish. The following day, cells were treated with APE (50 and 100 μM) for 48 h. At the end of the treatment, cells were trypsinized and resuspended in phosphate-buffered saline (PBS)/FBS 1:1. Then the samples were incubated with 2 μg/ml PI (Sigma-Aldrich) for 30min at room temperature (RT) and subjected to FACS analysis using CyAn ADP flow cytometer. Samples were analyzed with FlowJo software, version 10.5.3.

| Annexin V staining for apoptosis detection
Annexin V analysis for apoptosis detection was performed using eBioscience™ Annexin V-APC Apoptosis Detection Kit (Thermo Fisher Scientific) according to the manufacturer's instructions. Cells were treated with APE 100 μM at different times (4-8-24 h) at the density of 1.5 × 10 5 cells/dish. At the end of the treatment, cells were trypsinized, washed twice with PBS1X, and incubated with staining solution for 15 min at RT in the dark. Samples were run on the BD FACSCanto II and analyzed with BD FACSDiva Software (BD Bioscences).
2.6 | Total RNA preparation, reverse transcription polymerase chain reaction (RT-PCR), and real-time PCR Total RNA was extracted using TRI-Reagent (Sigma-Aldrich). Total RNA was reverse-transcribed in a final volume of 20 μl using the M-MLV Reverse Transcriptase kit (Invitrogen) according to the manufacturer's instructions. The expression of human mAChR subtypes was evaluated by semiquantitative RT-PCR using One-Step RT-PCR Kit (GeneDireX). RPLP0 was used as the housekeeping gene. The presence of transcripts for mAChRs and apoptotic markers was also evaluated by SYBR Green Real-Time PCR on an Applied Biosystems Real-Time PCR System using SYBR Green Universal PCR Master Mix (EuroClone), following the manufacturer's recommendations. Each sample was normalized to its RPS29 or RPLP0 content. The primers utilized are described in the Supporting Information: Table S1. Primer's specificity was confirmed by melting curves.

| 5-Bromo-2′-deoxyuridine (BrdU) cell proliferation assay
The cells at the exponential phase were treated with increasing APE concentrations for 48 h. At the end of treatment, BrdU (Sigma-Aldrich) was added to the medium at a final concentration of 45μM for 20 min. Samples were fixed with 4% PFA/PBS for 10min at 4°C. Partial DNA denaturation was performed using 3.5 N HCl for 3 min at RT. The cells were permeabilized in 0.1% TritonX-100 for 10 min at RT; the reduction of nonspecific background signal was obtained as described for immunofluorescence. Samples were then incubated with rat anti-BrdU (1:50; Abcam, ab6326) for 1 h at RT. After the washes, the cells were incubated with species-specific secondary antibody for 1 h at RT (Donkey Anti-Rat IgG (H+L)-FITC; Jackson Immuno Research, #712-545-153, diluted 1:200). In the end, the nuclei were stained with 1 μg/ml Hoechst-33342. After two washes for 5 min each in PBS, the slides were mounted with buffered glycerol and examined using a Zeiss Axioscope Imager 2 fluorescence microscope (Carl Zeiss SpA). Pictures were acquired at a magnification of 10×. Total (an average of 3000 cells) and BrdU positive cells were counted in 20 randomly chosen fields, and the percentages of total BrdU positive cells out of the totals were quantified.

| Nuclear staining
For nuclear morphological analysis, both floating and adherent cells were fixed in 4% PFA/PBS. Floating cells were cytocentrifuged onto a glass slide at 1000 rpm for 20 min. After fixation the cells were washed with PBS and then incubated for 12 min with 1 µg/ml Hoechst-33342. Staining was visualized on a Zeiss Axioscope Imager 2 fluorescence microscope (Carl Zeiss SpA).

| Statistical analysis
Statistical analyses were performed using one-way analysis of Variance followed by the Student-Newman-Keuls method to compare multiple groups. Values with p < 0.05 were considered statistically significant.

| Characterization of mAChRs in ovarian tumor cell lines and nontumor immortalized cell lines
RT-PCR and real-time PCR determined mAChR transcripts in two ovarian cancer cell lines, SKOV-3 and TOV-21G, and two immortalized nontumorigenic human OSE cell lines iOSE398 and iOSE120. Except for M4 (not shown), all the other receptors were found in the four cell lines ( Figure 1A). We focused our attention on M2 receptor. The real-time PCR analysis confirmed that M2 mRNA levels were significantly higher in the two iOSE cells than cancer cell lines ( Figure 1B). When reverse transcriptase was omitted, no bands were detectable, indicating the absence of genomic DNA contamination (data not shown). The higher levels of M2 in OSE cells were also confirmed at the protein level by western blot analysis ( Figure 1C).

| M2 receptor is involved in the inhibition of ovarian cancer cell proliferation
To investigate the possible role of M2 receptor in the control of cell growth and survival, we treated the cells with a selective M2 agonist, APE, at different concentrations, ranging from 25 to 100 μM, and at different culture times. We observed a significant decrease in cell number in response to APE treatment, and the effect was dose and time-dependent in all cell lines ( Figure 2A). Moreover, as evidenced by IC50 values (Figure 2A), iOSE cells were more sensitive to the drug than cancer cells.
To confirm that APE effect was mediated by M2 receptor activation, we performed pharmacological competition experiments. The co-treatment for 48 h with APE (50-100 μM) and the M2 antagonist Gallamine, demonstrated the ability of Gallamine to counteract the APE effect on all cell lines ( Figure 2B). To exclude a possible contribution of M1, M3, and M5 receptor subtypes in the cell growth inhibition-APE mediated, we used Pirenzepine and 4-DAMP, respectively M1 and M3/M5 antagonists. As shown in Figure 2B, treatment with these antagonists could not counteract APE effects on cell survival, suggesting that the odd mAChRs are not involved. As control of mAChR antagonists' efficiency, we analyzed cell growth in the presence of Muscarine (100 μM), a nonselective mAChR agonist. Muscarine treatment significantly improved cell growth in all cell lines with respect to control conditions ( Figure 2C). When we blocked the inhibitory receptor M2 with Gallamine, the addition of Muscarine induced a further improvement in cell proliferation. However, this increase was statistically significant only in the iOSE-120. Conversely, in the presence of the odd mAChR antagonists, we found, as expected, that Muscarine, through M2 receptor, caused a substantial decrease of cell number in all cell lines ( Figure 2C).

| M2 receptor activation affects cell cycle progression
To investigate whether the M2-dependent decrease in cell number could be ascribed to decreased cell proliferation, the fraction of cells in the S-phase was evaluated by bromodeoxyuridine (BrdU) incorporation. In both cancer cell lines and iOSE-120 cells, we observed a decrease in the percentage of cells in the S-phase following APE treatment, which was dose-dependent ( Figure 3A-C).
We performed PHH3 counts on both cancer cell lines since PHH3 is a marker of cells within the cell cycle's mitotic phase. 23 PHH3 counts showed that in TOV-21G cell line APE treatment induced a slight  Figure 3D). On the contrary, in SKOV-3 cells, APE caused an increase, dose-dependent, of this percentage ( Figure 3E), suggesting the accumulation of SKOV-3 in the M phase.
To further investigate this result, we better analyzed all the cell cycle phases in the two cancer cell lines. The analysis of the cell cycle by FACS confirmed that APE treatment induced in SKOV-3 cells a progressive accumulation in G2/M with the time (Figure 3G), while TOV-21G after a transient slowdown in S-phase slowly recovered their growth rate ( Figure 3F).

| M2 receptor activation reduces cell survival inducing cell death
We analyzed by PI staining the effect of APE treatment on cell death. The data obtained demonstrated increased cell death in all cell lines, even with different sensitivity. iOSE cells were the most sensitive to the treatment  was effective only at the highest concentration (Supporting Information: Figure S1).

| M2 receptor activation induces mitotic catastrophe in SKOV-3 ovarian cancer cells
Considering the accumulation of SKOV-3 cell line in the M phase upon APE treatment, we investigated whether these cancer cells presented abnormal mitosis. We performed an analysis of mitotic spindles of SKOV-3 cell line. Immunofluorescence assays with anti-α-tubulin showed that control cells displayed standard mitotic spindles with chromosomes correctly condensed ( Figure 4A). Conversely, treated cells exhibited abnormal mitotic spindles. We observed monopolar or multipolar spindles with significant defects in chromatin condensation ( Figure 4B,C). Cells displaying aberrant spindles did not progress through the metaphase-anaphase, and the percentage of aberrant mitosis increased in a dosedependent manner ( Figure 4D). To further confirm these data, we performed the same experiment on TOV-21G cell line (p53 positive cells), and we found only a few aberrant mitoses that did not increase after APE treatment ( Figure 4E).
To further investigate the pattern of cell death in the two cancer cell lines, we carried out cytofluorimetric analyses with Annexin V-APC/PI. As shown in Figure 5, with the increase of time, we observed a transient increase of apoptotic cells in TOV-21G cells followed by a secondary increase in necrotic cells. Conversely, in SKOV-3 cells, at the same culture times, we did not observe increasing apoptosis but mainly an increase in necrotic cells.

| M2 receptor activation can induce different cell death pathways in ovarian cancer cells
Considering the different effects observed in two tumor cell lines, we also investigated the possible apoptotic effects downstream of the M2 agonist treatment. Caspase-3 cleavage is a critical step in apoptosis; therefore, we analyzed the presence of cleaved-caspase3 in response to APE treatment in the different cell lines ( Figure 6A-C). In iOSE cell line, we found a dosedependent increase of activated caspase 3. In TOV-21G, as expected, we observed an increase only at the highest concentrations of APE, while SKOV-3 did not show activation of caspase 3.
To determine which molecular pathway was involved in APE-induced cell death, we investigated by real-time PCR the expression of genes of proapoptotic Bcl2 family (PUMA, p53 upregulated modulator of apoptosis, and NOXA, Phorbol-12-myristate-13-acetate-induced protein 1) and an antiapoptotic one (BCL-xL). PUMA expression was induced in TOV-21G and iOSE-120 but not in SKOV-3 ( Figure 5D). Conversely, the expression of NOXA was induced in all cell lines ( Figure 6E). Following APE treatment BCL-xL (B-cell lymphoma-extra-large) was slightly downregulated in iOSE-120, in TOV-21G remained unchanged, while in SKOV-3 cells was significantly upregulated ( Figure 6F).

| DISCUSSION
The cholinergic autocrine loop has been demonstrated in the normal ovary. 7,8 Moreover, it has been reported that ovarian cancer cells express mAChRs. 18 Oppitz et al. 18 described a statistically significant correlation between mAChR status in ovarian carcinoma cell lines and poor prognosis of patients from whom the cell lines were established. However, poor information on the type of receptor present in epithelial ovarian cancer are reported. Initial studies on ovarian adenocarcinoma reported a binding profile whose results suggested the presence of M3 receptors. 19 These receptors are the most common ACh receptors in cancers, and their expression level has been frequently correlated with increased cell proliferation. 24 The present study aimed to better characterize the expression of the mAChR subtypes in ovarian cancer cells and nontumoral iOSE cells. Our data showed that except for M4, the studied cell lines express all mAChR subtypes, albeit at different levels. Interestingly, the cancer cell lines showed lower levels of M2 receptors. This receptor is involved in the inhibition of tumor cell proliferation in other cancer cell lines 9,11,15,25,26 and its function has never been studied in ovarian cancer.
The control of cell proliferation is a balance of proliferative/antiproliferative effects and may depend on the levels of expression of the different mAChR subtypes. The simultaneous activation of all mAChRs by muscarine, a no selective orthosteric agonist, increased tumor cell number in all treated cell lines. Interestingly, this increase was more evident in cancer cells ( Figure 2C), suggesting that the lower levels of M2 in cancer cell lines compared with iOSE, might represent a strategy promoted by the tumor cells to favor cancer cell proliferation.
Moreover, the lower levels of M2 rendered cancer cells less sensitive to M2 agonist APE. In fact, APE decreased cell growth in all analyzed cell lines in a doseand time-dependent manner, even though with different sensitivity. The iOSE cells were the most sensitive to the treatment (Figure 2A), and cancer cells were the most resistant.
Interestingly, we found differences in the nuclear morphology after APE treatment. While in the iOSE-120 and the TOV-21G, we observed the presence of fragmented nuclei typically associated with apoptotic processes, in SKOV-3, we found together with cells with fragmented nuclei, the presence of multiple micronuclei and uncondensed chromatin, suggesting that SKOV-3 were dying by mitotic catastrophe as has been recently demonstrated in U251 cells. 27 To better understand the impact of APE on cell proliferation, we have analyzed, by FACS analysis, the cell cycle progression in the two cancer cell lines. While TOV-21G recovered to the normal cycle after an initial These different responses to M2 activation may also depend on the different status of p53 in our cells; in fact, iOSE-120 and TOV-21G express p53 wild-type that was positively regulated by APE treatment (data not shown), while SKOV-3 cells do not express p53 at the protein or mRNA level. 28 Therefore, the absence of p53 protein may render this cell line unable to recover the cell damage induced by APE treatment.
Upon spindle damage, cells become arrested at the metaphase-anaphase transition point, escape from the block (mitotic slippage), and are definitively arrested at the G1 stage. P53 mediates this arrest 29 and, as previously | 1449 observed in other tumor types, 11,13 cells with mutated p53 fail to resolve the mitotic division or DNA damage and may undergo aberrant mitosis and mitotic catastrophe. 30,31 Mitotic catastrophe may cause cell death following abnormal mitosis in cells presenting defects in checkpoint functions. 32 This hypothesis was confirmed by a high number of aberrant mitosis in APEtreated SKOV-3 and their absence in TOV-21G and by the different profiles obtained by cytofluorimetric analyses with Annexin V-APC/PI. Moreover, the expression levels of gene involved in cell death were differently modulated in the three cell lines. APE treatment induced PUMA (p53 upregulated modulator of apoptosis), and NOXA in both iOSE-120 and TOV-21G, while NOXA but not PUMA was induced in SKOV-3 cells, which do not express endogenous p53. Interestingly, in APE-treated cells, cleaved caspase-3 was readily induced in TOV-21G and iOSE-120 but not in SKOV cells, while induction of BCL-xL was observed in SKOV but not in TOV-21G and iOSE-120. These data are in line with published results showing that BCL-xL is already expressed in SKOV-3, and its expression increases upon stimulation with BPR0L075, a novel synthetic indole compound that inhibits tubulin polymerization. 33 Interestingly, in SKOV3, BPR0L075 induced an alternative, caspase-3 independent, cell death by mitotic catastrophe. 33 Caspase-3 is considered a key effector enzyme in inducing cell apoptosis. Bcl-xL is an inhibitor of the Bcl-2 family of apoptotic proteins and is associated with resistance to antitumoral agents in many tumor types. 34 In addition to its involvement in apoptosis, Bcl-xL overexpression increases the aggressiveness of cancer cells by positively regulating invasion and migration, promoting tumor angiogenesis, and maintaining cancer stem cell phenotype. 35 In the literature, a relation between caspase-3 and Bcl-xL activities has been described. It has been shown that Bcl-xL could block the activation of caspase-3, 36 while in polymorphonuclear neutrophils, caspase-3 may regulate Bcl-xL levels by direct proteolytic cleavage. 37 The induction of BCL-xL in SKOV-3, but not in the other cell lines, may explain the different cell death responses upon APE treatment in the different cell lines. In fact, in hepatocellular carcinoma, it has been shown that even though BCL-xL overexpression effectively blocked doxorubicin-induced apoptosis, however, it did not prevent doxorubicin-induced cell death through mitotic catastrophe. 38 In every ovulatory cycle, OSE cells are injured, and subsequently, through repair mechanisms, the cells undergo cyclical ovulatory ruptures and repair, leading to a predisposition to developing mutation and, later, malignancies. 4 OSE cells can be influenced by several factors and hormones present in their environment. 39,40 It has been demonstrated that gonadotropins (FSH and luteinizing hormone [LH]) may regulate OSE cells proliferation and selected ovarian tumors either directly through LH-and FSH-receptors or by stimulating the expression of growth factors such as KGF, HGF, and KL. [41][42][43] Moreover, granulosa cells produce ACh in response to FSH, 8 and the presence of mAChRs on OSE cells renders OSE cells responsive to ACh. Therefore, a correct balance of the different mAChR subtypes might be relevant for controlling OSE cell proliferation/death. The data showing the prevalence of M3 receptors in ovarian cancer 19 and the lower levels of M2 receptors, as demonstrated in the present work, further support the significant correlation between mAChR status and poor prognosis in patients with ovarian cancer. 18 In conclusion, our results, according to the results obtained in other tumor types, 9,12,14,15,44 confirm the inhibitory role of M2 receptors on tumor cell proliferation and survival.
Although the nature of the effects produced by APE in ovarian cell lines requires further investigations, these results indicate that the M2 mAChR could represent a new promising tool to explore in ovarian cancer development and therapy. The identification of novel compounds able to overcome drug resistance that can be used as adjuvant along with chemotherapeutic drugs are needed to improve the survival of ovarian cancer patients. Moreover, the development of novel M2 subtypes specific agonists with higher affinity compared to APE that could be used at lower concentrations than APE may also represent new interesting therapeutic perspectives for the ovarian cancer treatment. 44,45 AUTHOR CONTRIBUTIONS Marilena Taggi, Andjela Kovacevic, Chiara Capponi, Marta Falcinelli, and Veronica Cacciamani performed experiments; Rita Canipari, Elena Vicini and Ada Maria Tata conceived, supervised the study, and analyzed data; Marilena Taggi and Rita Canipari wrote the manuscript; Elena Vicini and Ada Maria Tata revised the manuscript.
All the authors read and approved the final version of the manuscript.