Inhibition of ANO1/TMEM16A Chloride Channel by Idebenone and Its Cytotoxicity to Cancer Cell Lines

The expression levels of anoctamin 1 (ANO1, TMEM16A), a calcium-activated chloride channel (CaCC), are significantly increased in several tumors, and inhibition of ANO1 is known to reduce cell proliferation and migration. Here, we performed cell-based screening of a collection of natural products and drug-like compounds to identify inhibitors of ANO1. As a result of the screening, idebenone, miconazole and plumbagin were identified as novel ANO1 inhibitors. Electrophysiological studies showed that idebenone, a synthetic analog of coenzyme Q10, completely blocked ANO1 activity in FRT cells expressing ANO1 without any effect on intracellular calcium signaling and CFTR, a cAMP-regulated chloride channel. The CaCC activities in PC-3 and CFPAC-1 cells expressing abundant endogenous ANO1 were strongly blocked by idebenone. Idebenone inhibited cell proliferation and induced apoptosis in PC-3 and CFPAC-1 cells, but not in A549 cells, which do not express ANO1. These data suggest that idebenone, a novel ANO1 inhibitor, has potential for use in cancer therapy.


Introduction
Calcium-activated Clchannels (CaCCs) are widely expressed in various cell types and tissues, and they are implicated in many physiological activities such as epithelial fluid secretion, smooth muscle contraction, and sensory signal transduction [1][2][3]. CaCCs were first described over 3 decades ago but the molecular identity of CaCCs has recently been identified [4]. In 2008, three independent research groups reported that anoctamin-1 (ANO1, TMEM16A) gene encodes a CaCC, showing calcium-activated Clcurrents when it was expressed in oocytes and mammalian cells [5][6][7]. ANO1 is expressed in various cell types including tracheal, intestinal, and glandular epithelia, smooth muscle cells, intestinal pacemaker cells, sensory neurons, and several tumors [5,[7][8][9]. ANO1 was also known as discovered on GIST-1 (DOG1), tumor amplified and overexpressed sequence 2 (TAOS2), and oral cancer overexpressed 2 (ORAOV2) [10,11]. DOG1, TAOS2 and ORAOV2 are named so because ANO1 is strongly overexpressed in gastrointestinal stromal tumours (GIST) and oral squamous cell carcinomas. ANO1 is mapped to the chromosomal band 11q13 that is frequently amplified in a variety of human carcinomas including head-and-neck squamous cell carcinoma (HNSCC), GIST, breast and prostate cancer. Recent evidence suggests ANO1 involvement in cell proliferation, cell migration, tumorigenesis and cancer progression [12,13]. For instance, inhibition of ANO1 expression in prostate cancer PC-3 cells significantly reduced proliferation, metastasis and invasion, and blocked tumor growth in a xenograft mouse model [14]. Pharmacological inhibition of ANO1 by T16A inh -A01, a selective ANO1 inhibitor, reduced proliferation of interstitial cells of Cajal (ICC) and CFPAC-1 pancreatic cancer cells expressing endogenous ANO1 [15]. In breast cancer cells, down-regulation of the ANO1 gene expression reduced proliferation, provoked apoptosis, and inhibited tumor growth in a xenograft model. In addition, pharmacological inhibition of CaCC activity of ANO1 reduced cell viability in HNSCC, esophageal squamous cell carcinoma (ESCC) and breast cancer cells via inhibition of epidermal growth factor receptor (EGFR) and calmodulin-dependent protein kinase II (CAMKII) signaling [16].
For the identification of novel ANO1 inhibitors, we performed a cell-based screening with a collection of natural products and drug-like compounds using a cell based high-throughput screening assay established for the identification of ANO1 inhibitors in previous study [19]. We found some drug-like compounds and natural products showing potent ANO1 inhibitory activity, and investigated the effect of the hit compounds on growth inhibition of cancer cell lines, which express ANO1 endogenously.

Materials and solutions
Idebenone, coenzyme Q10, plumbagin, miconazole, and other chemicals, unless otherwise indicated, were purchased from Sigma. Mouse ANO2 was purchased from Origene Technologies Inc. (Rockville, MD, USA, catalog No MC205812). The compound collection used for screening (Spectrum Collection, 2320 compounds) was purchased from MicroSource Discovery Inc. (Gaylordsville, CT). This library consists of human therapeutic drugs or drug-like compounds and natural products. The compounds were diluted with DMSO to reach a concentration of 2.5 mM. This was used as the 100x concentrated stock solution which was treated on the cells.
Cell based screening ANO1/TMEM16A and YFP expressing FRT cells were plated in 96-well black-walled microplates (Corning Inc., Corning, NY) at a density of 20,000 cells per well in F12 medium supplemented with 10% FBS, 100 U/mL penicillin, 100 μg/mL streptomycin. Assays were done using FLUOstar Omega microplate reader (BMG Labtech, Ortenberg, Germany) and MARS Data Analysis Software (BMG Labtech). Each well of 96-well plate was washed 3 times in PBS (200 μL/wash), leaving 100 μL PBS. Test compounds (1 μL) were added to each well at 25 μM final concentration. After 10 min, 96-well plates were transferred to a plate reader for fluorescence assay. Each well was assayed individually for TMEM16A-mediated Iinflux by recording fluorescence continuously (400 ms per point) for 2 s (baseline), then 100 μL of 140 mM Isolution containing 200 μM ATP was added at 2 s and then YFP fluorescence was recorded for 6 s. Initial iodide influx rate was determined from the initial slope of fluorescence decrease, by nonlinear regression, following infusion of iodide with ATP.

Ussing Chamber study
Snapwell inserts containing ANO1-or CFTR-expressing FRT cells were mounted in Ussing chambers (Physiologic Instruments, San Diego, CA). The basolateral bath was filled with Tris-ATP, and 10 HEPES (pH 7.2). Pipettes were pulled from borosilicate glass and had resistances of 3-5 MO after fire polishing. After establishing the whole-cell configuration, ANO1 was activated by ATP (100 μM). Whole-cell currents were elicited by applying hyperpolarizing and depolarizing voltage pulses from a holding potential of 0 mV to potentials between -80 mV and +80 mV in steps of 20 mV. Recordings were made at room temperature using an Axopatch-200B (Axon Instruments, Union City, CA). Currents were digitized and analyzed using a Digidata 1440A converter (Axon Instruments), and pCLAMP 10.2 software (Molecular Devices, Sunnyvale, CA). Currents were low-pass filtered at 1 kHz and sampled at 5 kHz.

Immunoblot
Cell extracts and immunoblotting were prepared as described previously [23]. FRT-ANO1, PC3, CFPAC-1 and A549 cells were lysed with cell lysis buffer (50 mM Tris-HCl, pH 7.4, 1% Nonidet P-40, 0.25% sodium deoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM Na 3 VO 4 , and protease inhibitor mixture). Whole cell lysates were centrifuged at 15,000 g for 10 min at 4°C to remove the cell debris, and equal amounts (80 μg protein/lane) of supernatant protein were separated by 4-12% Tris-glycine precast gel (KOMA BIOTECH, Seoul, Korea) and then transferred onto PVDF membrane (Millipore, Billerica, MA). Membrane was blocked with 5% nonfat skim milk in Tris-buffered saline (50 mM Tris-Cl, pH 7.5, 150 mM NaCl) including 0.1% Tween 20 for 1 hour at room temperature. This membrane was then incubated overnight with primary ANO1 antibody (a generous gift of Young Duk Yang, CHA University). After washing with 0.1% Tween 20 in Tris buffered saline (TBST), the blot was further incubated for 45 min at room temperature with an anti-rabbit secondary antibody (Cell Signaling). The membrane was then washed three times with TBST for 5 minutes and then visualized using the ECL Plus western blotting detection system (GE Healthcare Amersham; Piscataway, NJ).

Wound healing assay
Cell mobility was assessed using a scratch wound assay. PC3 cells were cultured in a 96-well plate until it was confluent. The cell layer was wounded using a 96-Well WoundMaker (Essen BioScience, Michigan, USA) and washed twice with fresh serum free media. The cells were incubated with serum free medium, and images of the wounds were automatically taken every 2 h for 48 h using the IncuCyte ZOOM (Essen BioScience). The images were analyzed by the IncuCyte software package (Essen BioScience).

Statistical analysis
The results of multiple experiments are presented as the means ± S.E. Statistical analysis was performed with Student's t-test or by analysis of variance as appropriate. A value of P < 0.05 was considered statistically significant.

Identification of ANO1/TMEM16A inhibitors
A cell-based screening of a collection of natural products and drug-like compounds was done for the identification of ANO1 inhibitors. ANO1 chloride channel activity was measured using Fischer rat thyroid (FRT) cells stably expressing human ANO1 and the genetically encoded iodide-sensing fluorescent protein, YFP-F46L/H148Q/I152L. As shown in Fig 1A, for the screening to identify ANO1 inhibitors, the FRT cells were pre-incubated with test compounds in PBS prior to addition of an iodide and ATP, an agonist of P2 purinergic receptor which causes an increase in intracellular calcium concentration, containing solution. In the presence of ANO1 inhibitor, YFP fluorescence quenching by iodide intake through ANO1 will be inhibited.
Screening of 2320 compounds yielded 24 compounds that blocked iodide influx by > 70% at 25 μM. We found three novel ANO inhibitors, idebenone, plumbagin and miconazole, from the primary hit compounds. Idebenone, plumbagin and miconazole significantly inhibited ANO1 chloride channel activity in a dose-dependent manner and fully inhibited ANO1 at 30 μM (Fig 1C).

Characterization of idebenone
Idebenone was further studied because idebenone, a synthetic analog of coenzyme Q10 (CoQ10), shows potent and selective inhibition of ANO1, and is widely used as a powerful antioxidant. Apical membrane currents measurement in ANO1-expressing FRT cells gave an IC 50 of 9.2 μM for idebenone and show almost complete inhibition of ANO1 chloride currents by 30 μM idebenone (Fig 2A and 2B). To investigate the effect of idebenone on intracellular calcium signaling, HT-29 and FRT cells were loaded with Fluo-4 NW, a fluorescent calcium sensor. The cells were pretreated with 30 μM idebenone and then ATP applied at a concentration of 100 μM induced transient increase in the cytosolic calcium concentration. The ATP-induced cytosolic calcium increase was not significantly affected by idebenone ( Fig  2C). To elucidate if idebenone alters the other chloride channels, cystic fibrosis transmembrane conductance regulator (CFTR) and ANO2 (TMEM16B), which shares a high amino acid homology to ANO1, we measured apical membrane currents in FRT cells expressing human wild-type CFTR and mouse ANO2 (mANO2). CFTR channel activity was affected only a little by idebenone. It inhibited CFTR by 8.2 ± 0.2% at 30 μM, a concentration that fully inhibits ANO1 (Fig 2D), but, not surprisingly, idebenone strongly inhibited 100 μM ATP-induced activation of mANO2 in a dose dependent manner in FTR-mANO2 cells ( Fig 2E). We measured apical membrane currents to observe the effect of CoQ10 on ANO1 channel activity in FTR-ANO1 cells because idebenone is a short-chain analog of CoQ10. As shown in Fig 2F, 100 μM CoQ10 did not inhibit ATP-induced ANO1 chloride currents. To investigate whether idebenone reversibly inhibits ANO1, we measured apical membrane currents in FTR-ANO1 cell with E act , a specific activator of ANO1. Pretreatment of 100 μM idebenone completely inhibited E act -induced ANO1 activation ( Fig 2G). As shown in Fig 2H, pretreatment of 100 μM idebenone completely inhibited ATP-induced ANO1 activation. However after washout, most of the inhibitory effect of idebenone on ANO1 activation by E act was removed. Idebenone thus reversibly inhibits ANO1 without alterations in intracellular calcium signaling.

ANO1 inhibitors decrease cell proliferation and migration in adenocarcinoma cells
Three types of human adenocarcinoma cell lines were tested to determine the effect of idebenone on the endogenous CaCCs activity and cell growth and migration. Western blotting revealed that endogenous ANO1 is expressed at high levels in PC3 and CFPAC-1 (human pancreatic ductal adenocarcinoma derived) cells but not in A549 (human lung adenocarcinoma derived) cells (Fig 4A). The YFP fluorescence quenching assay with CFPAC-1 cells expressing the halide sensing mutant YFP revealed that idebenone potently blocked ATP-induced CaCC chloride channel activation in a dose-dependent manner (Fig 4B). To investigate whether the ANO1 inhibitors affect the cell growth, idebenone, miconazole and plumbagin were applied to PC3 cells expressing high levels of ANO1 and A549 cells not expressing ANO1. The results are shown in Fig 4C, and they reveal that idebenone significantly inhibited the cell growth in PC3 cells but not in A549 cells. Miconazole and plumbagin showed strong inhibition of cell growth in both PC3 and A549 cells, but it was not surprising because in several previous investigations, it was shown that miconazole and plumbagin inhibited cell growth of various human cancer cells through different mechanisms [24][25][26][27][28][29]. The effect of ANO1 inhibition by idebenone on cell migration was determined using a wound healing assay in PC3 cells. In the assay, control PC3 cells covered 63.6 ± 2.5% (n = 5) of the wound, whereas T16A inh -A01 (30 μM) and idebenone (10 and 30 μM) treated cells covered 39.6± 2.5, 26.1 ± 1.8 and 13.8 ± 0.8% (n = 5) of the wound at 48h post-wound, respectively (Fig 4D).
To further determine the effect of idebenone on cell proliferation, we observed cell proliferation in response to idebenone using MTS assay and BrdU assay in PC3, CFPAC-1 and A549 cells (Fig 5). In this study, coenzyme Q10 was used as negative control because it does not inhibit ANO1/CaCCs even though idebenone is a synthetic analog of coenzyme Q10 (Fig 2F). The cells were treated with idebenone (3, 10 and 30 μM), coenzyme Q10 (100 μM) and T16A inh -A01 (10 μM), and the cell proliferation was quantitatively estimated after 2 days. In PC3 and CFPAC-1 cells, idebenone inhibited cell viability and BrdU incorporation in a dosedependent manner, but idebenone and T16A inh -A01did not affect cell viability and BrdU incorporation in A549 cells (Fig 5A-5C). Coenzyme Q10 did not block the cell proliferation in all the cell lines. Downregulation of ANO1 induces apoptosis in several cancer cell lines overexpressing ANO1 [14,16,30]. To investigate whether idebenone induces apoptosis in ANO1 expressing cells, TUNEL staining was performed in the adenocarcinoma cell lines. PC3, CFPAC-1 and A549 cells were incubated with 30 μM idebenone for 48 h, and then DNA damage was monitored using the TUNEL assay. TUNEL positive cells were detected in PC3 and CFPAC-1 cells but not in A549 cells (Fig 5D-5F). Discussion ANO1, a recently identified CaCC, is strongly overexpressed in various tumor types including HNSCC, GIST, breast and prostate cancer. Notably, Recent evidence strongly suggests ANO1 involvement in cell proliferation, cell migration, tumorigenesis and cancer progression [12,13,16,31]. Several studies have indicated that ANO1 may be used as a therapeutic target of tumors because down regulation of ANO1 showed potential therapeutic benefits on HNSCC, ESCC, GIST, breast and prostate cancer [14,16,30,32,33]. Thanks to recent identification of ANO1, few small molecule ANO1 inhibitors are available such as CaCC inh -A01, tannic acid, T16A inh -A01, and MONNA [17,[19][20][21], but the pharmacological property and the mechanisms of action of the inhibitors are still unclear. In this study, we performed a cell-based screening to identify novel ANO1 inhibitors with a collection of drug like compounds and natural products. Interestingly, we discovered that idebenone strongly inhibited ANO1.
We showed that idebenone potently inhibited the ANO1 activity in YFP fluorescence quenching assay (Fig 1C) and electrophysiological studies (Figs 2A and 3A). To find out whether antioxidant and electron carrier activity of idebenone affect ANO1 channel function, we observed the effect of CoQ10 on ANO1 channel activity because idebenone and CoQ10 share the same substitution pattern of the quinone moiety and differ only in the alkyl tail attached to the C6-carbon atom of their quinone ring. As shown in Fig 2F, pretreatment of 100 μM CoQ10 slightly (but not significantly) increased ANO1 currents instead of inhibiting ATP-induced ANO1 currents in ANO1 expressing FRT cells. In addition, CoQ10 did not block the cell viability and proliferation in several ANO1 expressing cell lines (Fig 5), and idebenone did not affect CFTR and intracellular calcium signaling (Fig 2C and 2D). Together, the above results suggest that the mechanism of action of idebenone on ANO1 inhibition may be direct rather than through its antioxidant and electron carrier ability. There is still a possibility that idebenone affect cell proliferation and apoptosis via the other pathways. However, idebenone showed potent inhibition of ANO1 and cell proliferation in ANO1 expressing cells, and an ANO1 specific inhibitor T16A inh -A01 showed similar response on cell proliferation (Fig 5). This result suggests that idebenone induced ANO1 inhibition is at least partially involved in the cytotoxic effect of idebenone. In Fig 4D, we showed that strong inhibition of cell migration into a wound area by idebenone. In wound healing assay, we used serum-free cell culture media to reduce the cell growth because both cell proliferation and migration affect the rate of wound healing. Serum deprivation resulted in~54% growth inhibition (data not shown) and the inhibitory effect of idebenone on wound healing was stronger than that of idebenone on cell proliferation (Figs 4D and 5A). These results suggest that idebenone inhibits cell migration even though growth inhibition by idebenone can affect the result.
In clinical trial, patients have been treated with idebenone up to 2,250 mg/day. When healthy male subjects received a single oral dose of 750 mg of idebenone three times per day, the maximum plasma concentrations (C max ) of parental and total idebenone were 22.4 and 8,158 ng/mL, respectably [39]. The very low plasma concentrations of parent idebenone (~66 nM) due to rapid metabolism and the high plasma concentrations of total idebenone (~24 μM) including idebenone and its metabolites suggest that further studies on the effect of idebenone metabolites on ANO1 activity are needed to evaluate the role of ANO1 inhibition in the various biological effects of idebenone.
In this study, we also showed that miconazole and plumbagin potently inhibited ANO1 channel activity without altering the intracellular calcium concentration in FRT and HT-29 cells (Figs 1C and 2C). These two compounds strongly decreased cell growth of PC3 cells expressing high levels of ANO1 at 30 μM concentration showing near complete inhibition of ANO1 (Fig 4C). Miconazole, an imidazole antifungal agent, inhibits ergosterol biosynthesis and is commonly applied topically to treat superficial mycoses [40]. Recent studies have shown that miconazole inhibits cell growth by modulation of p53-associated signaling pathway and suppression of HIF-1α protein synthesis, but the underlying mechanism is still not well understood [24,25]. Plumbagin, a quinonoid obtained from the roots of Plumbago zeylanica, has been found to possess highly potent biological activities including anticancer properties. Plumbagin reportedly has chemopreventive properties against several cancer cells such as inhibition of cell growth in breast cancer cells, inducing apoptosis in pancreatic cancer cells, and inhibition of invasion and migration in liver cancer cells via various pathways including inhibition of AKT and downstream targets, but the underlying molecular mechanism is still unclear. [26][27][28][29]. In previous studies, miconazole and plumbagin showed dose-dependent inhibition of cell growth in human cancer cells with IC 50 in the range 10-20 μM and 3-10 μM, respectively [24,[26][27][28]. In Fig 1C, we showed that 10-20 μM miconazole and 3-10 μM plumbagin can significantly inhibit ANO1 activity. Thus, ANO1 inhibition by miconazole and plumbagin may be one of possible mechanisms of cytotoxic effects in cancer cells. To the best of our knowledge, this is the first work that shows idebenone, miconazole and plumbagin inhibit ANO1 chloride channel, and inhibition of ANO1 channel activity by these compounds may be, at least partially, involved in inhibition of cell proliferation.
In Fig 4A, ANO1 expression was analyzed by Western blotting, and it is revealed that the size and thickness of ANO1 bands were different in FRT-ANO1, PC3, CFPAC-1 cells. In the present study, we did not clarify the difference of ANO1 bands in the three cell lines. The alternative splicing of ANO1 [6], different ANO1 expression levels, and post-translational modifications such as glycosylation may create different sized ANO1 proteins.
In conclusion, this study shows that idebenone is a novel potent inhibitor of ANO1, and it did not affect intracellular calcium signaling and CFTR chloride channel activity. Idebenone potently inhibited ANO1/CaCC channel activity in PC-3 and CFPAC-1 cells, and significantly reduced cell proliferation, inhibited cell migration and induced apoptosis in these cells. These data suggest that idebenone may be a powerful candidate in the development of novel therapeutic agents for the prevention and treatment of cancer.