Cell. Mol. Life Sci. DOI 10.1007/s00018-014-1594-3

Cellular and Molecular Life Sciences

Research Article

Prevention of cisplatin‑induced ototoxicity by the inhibition of gap junctional intercellular communication in auditory cells Yeon Ju Kim · Jangho Kim · Chunjie Tian · Hye Jin Lim · Young Sun Kim · Jong Hoon Chung · Yun‑Hoon Choung 

Received: 21 August 2013 / Revised: 2 February 2014 / Accepted: 21 February 2014 © Springer Basel 2014

Abstract  Cis-diamminedichloroplatinum (cisplatin) is an effective chemotherapeutic drug for cancer therapy. However, most patients treated with cisplatin are at a high risk of ototoxicity, which causes severe hearing loss. Inspired by the “Good Samaritan effect” or “bystander effect” from gap junction coupling, we investigated the role of gap junctions in cisplatin-induced ototoxicity as a potential therapeutic method. We showed that connexin 43 (Cx43) was highly expressed in House Ear Institute-Organ of Corti 1 (HEI-OC1) cells, mediating cell–cell communication. The viability of HEI-OC1 cells was greatly decreased by cisplatin treatment, and cisplatin-treated HEI-OC1 cells showed lower Cx43 expression compared to that of untreated HEI-OC1 cells. In particular, high accumulation of Cx43 was observed around the nucleus of cisplatin-treated cells, whereas scattered punctuate expression of Cx43 was observed in the cytoplasm and membrane in normal cells, suggesting that cisplatin may interrupt the normal gap junction communication by inhibiting the trafficking of Cx43

Electronic supplementary material The online version of this article (doi:10.1007/s00018-014-1594-3) contains supplementary material, which is available to authorized users. Y. J. Kim · C. Tian · H. J. Lim · Y. S. Kim · Y.-H. Choung (*)  Department of Otolaryngology, Ajou University School of Medicine, San 5 Woncheon‑dong, Yeongtong‑gu, Suwon 443‑721, Republic of Korea e-mail: [email protected] J. Kim · J. H. Chung  Department of Biosystems & Biomaterials Science and Engineering, Seoul National University, Seoul 151‑742, Republic of Korea

to cell membranes in HEI-OC1 cells. Interestingly, we found that the inhibition of gap junction activity reduced cisplatin-induced apoptosis of auditory hair cells. Cx43 siRNA- or 18α-GA-treated HEI-OC1 cells showed higher cell viability compared to control HEI-OC1 cells during cisplatin treatment; this was also supported by fluorescence recovery after photobleaching studies. Inhibition of gap junction activity reduced recovery of calcein acetoxymethyl ester fluorescence compared to control cells. Additionally, analysis of the mechanisms involved demonstrated that highly activate extracellular signal-regulated kinase and protein kinase B, combined with inhibition of gap junctions may promote cell viability during cisplatin treatment. Keywords  Gap junction · Connexin 43 · Bystander effect · Cisplatin · Ototoxicity · HEI-OC1 Abbreviations 18α-GA 18 alpha-glycyrrhetinic acid Akt Protein kinase B ATP Adenosine triphosphate cAMP Cyclic adenosine monophosphate cGMP Cyclic guanosine monophosphate Cisplatin  Cis-diamminedichloroplatinum Cx Connexin ERK Extracellular signal-regulated kinase FRAP Fluorescence recovery after photobleaching HEI-OC1 House Ear Institute-Organ of Corti 1 InsP3 Inositol trisphosphate MAPK Mitogen-activated protein kinases NOX3 NADPH oxidase 3 PKC Protein kinase C ROS Reactive oxygen species SLDT Scrape load dye transfer

13

Y. J. Kim et al.

Introduction Cis-diamminedichloroplatinum (cisplatin) is an effective chemotherapeutic drug that has been widely used to treat various types of cancers [1]. However, its clinical use is quite limited due to its severe side effects, such as ototoxicity and nephrotoxicity [2]. In particular, ototoxicity has been reported to occur in about 75–100 % of patients treated with cisplatin [3] and has been shown to have irreversible, cumulative, and bilateral characteristics. However, the mechanism mediating cisplatin-induced ototoxicity has not been clearly elucidated. Histochemical studies on the effects of cisplatin have shown that hair cells and supporting cells in the organ of Corti are major targets, and exposure of these cells to cisplatin leads to cell death or dysfunction [1]. Cisplatin has been reported to accumulate in the spiral ganglion and the lateral wall, including spiral ligaments and stria vascularis [1, 4, 5]. Cisplatin is transferred into hair cells through several transporters, i.e., copper transporter 1 [6] and organic cation transporter 2 [7], and causes increases in the expression of NADPH oxidase 3 (NOX3) and the superoxide generating enzyme [8, 9]. With the rise in NOX3, large amounts of reactive oxygen species (ROS) are produced [2], subsequently activating the mitogen-activated protein kinases (MAPK) pathway [10, 11] and promoting the release of cytochrome C from mitochondria [12]; this pathway can lead to apoptosis via caspase activation [10]. Despite these previous reports, the mechanisms underlying cisplatin-induced ototoxicity are not yet fully understood. Gap junctions are intercellular channels that allow the passage of ions and small molecules less than 1 kDa between adjacent cells [13]. Gap junctions are formed by docking of paired connexons, which are composed of 6 proteins called connexins (Cxs). Twenty-one and 20 different types of Cx proteins have been identified in human and mouse genomes, respectively [14]. In general, gap junction intercellular communication is thought to play important roles in cellular homeostasis, embryonic development, proliferation, differentiation, growth, and physiological modulation. Recent studies have suggested that gap junctions are also critical regulators under pathological conditions [15, 16]. In this regard, some studies have shown that gap junction coupling is controlled by injurious agents, which, in turn, can either weaken or worsen the damage, the socalled “Good Samaritan effect” and “Bystander effect”, respectively [17]. The “Bystander effect” has been particularly well documented in cancer [18]. Cxs are degraded in most cancers, and the upregulation of gap junction communication and overexpression of Cxs have been shown to potentiate anticancer drug efficacy and to reduce drug resistance [19, 20]. In contrast, neuronal cells are more susceptible to toxic exposure or focal ischemia by gap junction

13

blocking agents, suggesting that gap junctions provide a “Good Samaritan effect” under conditions that promote cytotoxicity [21]. Thus, the role of gap junctions may differ according to the cell type, period, and severity of the injury [22, 23]. In the mammalian inner ear, gap junctions are abundant in the cochlear and vestibular regions, where Cx26, Cx30, Cx31, and Cx43 are expressed [24, 25]. These gap junctions play a critical role in the circulation of K+ between the fluids of the inner ear [26]. The circulation of K+ is also essential for the maintenance of specialized ionic composition and positive endocochlear potential of 80–100 mV in endolymph, which are required for normal hearing [27, 28]. Furthermore, gap junctions are thought to be involved in Ca2+ signaling and energy metabolism through controlling the passage of small molecules, such as inositol trisphosphate (InsP3), adenosine triphosphate (ATP), cyclic adenosine monophosphate (cAMP), and cyclic guanosine monophosphate (cGMP) [29–32]. Mutations in human Cx26, Cx30, Cx31, and, Cx43 have been linked to syndromic and nonsyndromic deafness. Moreover, knockout of Cx in the mouse ear causes severe hearing loss [30, 33, 34]. Together, these findings indicate that inner ear gap junctions may be crucial for normal hearing. Guided by these ideas, we propose herein that under pathological conditions, cochlear cells may be affected by altered gap junction communication via the “Good Samaritan effect” or the “Bystander effect.” To test this hypothesis, we investigated the role of gap junctions in cisplatininduced ototoxicity using an auditory cell line.

Materials and methods Cell culture and treatments House Ear Institute-Organ of Corti 1 (HEI-OC1) cells, conditionally immortalized cochlear epithelial cells, were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Gibco-BRL, Grand Island, NY, USA) supplemented with 10 % fetal bovine serum (FBS; Gibco, Milan, Italy) at 33 °C in 10 % CO2. A high density of cells during plating (>4 × 104 cells/cm2) is essential for the formation of gap junctions, and this condition was used for all experiments. To induce cell death, cells were incubated with 10, 25, or 50 μM cisplatin (Sigma-Aldrich, St. Louis, MO, USA) for 24 or 48 h. The gap junction inhibitor 18 alphaglycyrrhetinic acid (18α-GA; Sigma-Aldrich, Germany) was used to block gap junction communication between the cells. Cells treated with dimethyl sulfoxide (DMSO) served as a vehicle control for both cisplatin and 18α-GA. To explore the effects of mitogen-activated protein kinase (MAPK) inhibition, 10 μM LY294002 [a phosphoinositide

Prevention of cisplatin-induced ototoxicity

3-kinase (PI3K) inhibitor] and 20 μM PD98059 [a MAPK/ extracellular signal-regulated kinase (ERK) inhibitor] were added to the media 2 h prior to treatment with cisplatin. All chemicals for MAPK inhibition were supplied by Calbiochem (San Diego, CA, USA). RNAi experiments Lipofectamine RNAi Max (Invitrogen, Carlsbad, CA, USA) was used for transfection of HEI-OC1 cells with Cx43 siRNA (Santa Cruz Biotechnology, Santa Cruz, CA, USA), a pool of three target-specific 20–25-nucleotide siRNAs designed to knockdown gene expression. For efficient knockdown of Cx43, HEI-OC1 cells were transfected with 20 nM Cx43 siRNA and RNAi Max in a time-dependent manner, as described in the manufacturer’s protocol. A scrambled siRNA oligo was used as a control (Santa Cruz Biotechnology). Knockdown of Cx43 was assessed by Western blotting, reverse transcription-polymerase chain reaction (RT-PCR), and immunocytochemistry. Western blotting HEI-OC1 cells were washed with phosphate-buffered saline (PBS) and lysed using radio-immunoprecipitation assay (RIPA) buffer. Equal amounts of protein from each sample were analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) followed by electro-transfer to polyvinylidene fluoride (PVDF; Millipore, Billerica, MA, USA) membranes. Membranes were blocked for 1 h with 5 % skim milk in PBS containing 0.05 % Tween20 (PBST) and incubated with primary antibodies overnight at 4 °C. Membranes were washed 4 times with PBST and incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies for 1 h at room temperature. After washing, the protein signal was detected using a chemiluminescence solution (GenDEPOT, Barker, TX, USA). The intensity of protein bands was quantified using ImageJ software (Broken Symmetry Software, USA). β-Actin was used as a loading control. Antibodies used were as follows: anti-connexin 43, anti-PARP, anti-phospho-p44/42 MAPK, anti-total-p44/42 MAPK, anti-caspase 3, anti-phosphoAkt, anti-phospho-SAPK/c-Jun N-terminal kinase (JNK), anti-total-SAPK/JNK, and anti-β-actin. All antibodies that are used for Western blotting were purchased from Cell Signaling Technology (Danvers, MA, USA) and diluted as 1:1,000 except β-actin (1:2,000). Immunocytochemistry For immunocytochemical analysis, HEI-OC1 cells were grown on glass coverslips (Marienfeld, LaudaKoenigshofen, Germany). Cells were fixed in 4 %

paraformaldehyde for 20 min, washed with PBS, and permeabilized in 0.2 % Triton-X 100/PBS for 10 min at room temperature. After washing, they were blocked with 1 % bovine serum albumin (BSA; GenDEPOT) in PBS and processed for indirect immunofluorescence using primary antibodies and secondary antibodies coupled with fluorescein isothiocyanate (FITC) or cyanine 3 (Cy3) (Jackson ImmunoResearch Laboratories, West Grove, PA, USA). The primary and secondary antibodies were diluted in a blocking buffer. Nuclei were stained with 1 μg/ml 4′,6′-diamidino2-phenylindole (DAPI; Invitrogen) in PBS for 2 min at room temperature. The cells were visualized with a Zeiss LSM 700 confocal or AxioVision LE 4.5 microscope (Carl Zeiss MicroImaging Inc., Thornwood, NY, USA). 3‑(4,5‑Dimethylthiazol‑2‑yl)‑5‑(3‑carboxymethoxyphenyl)‑ 2‑(4‑sulfophenyl)‑2H‑tetrazolium (MTS) assay Cells were grown in 96-well plates (4 × 103 cells/well) and transfected with Cx43 siRNA. Control cells were transfected with control siRNA. At 24, 48, and 72 h following siRNA transfection, MTS (Promega, Madison, WI, USA) solution was added to each well. After a 4-h incubation at 33 °C in the dark, the absorbance was measured at 490 nm using a Power Wave X Microplate ELISA Reader (Bio-TeK Instruments, Winoski, VT, USA). All experiments were repeated three times in triplicate, and viability was normalized to that of the control. Scrape load dye transfer (SLDT) technique HEI-OC1 cells were plated and grown overnight to confluence in 35-mm culture dishes. Next, 0.05 % Lucifer yellow (LY; Sigma-Aldrich) in PBS was added to cover the cells, and cells were randomly scraped with a sharp blade. After incubating for 5 min at 33 °C, cells were washed with PBS and fixed with 4 % paraformaldehyde. Dye-coupled cell layers were observed using a fluorescence microscope. The numbers of dye-coupled cells were counted to evaluate gap junction intercellular communication. Fluorescence recovery after photobleaching (FRAP) Monolayer HEI-OC1 cells were washed with PBS, and calcein AM (623 Da, Molecular Probes, Eugene, OR, USA) in 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) was loaded for 15 min. Dye-loaded cells were placed in an incubation chamber (21 °C) under a Zeiss LSM 700 confocal microscope (Carl Zeiss MicroImaging Inc.). To test gap junctions, we selected one cell that was adjacent to least three other cells and photobleached the cell with a short intense exposure to an argon laser. After photobleaching, the recovery of fluorescence was monitored every 10 s

13

Y. J. Kim et al.

for 400 s. The fluorescence intensity of the bleached region was measured using LSM confocal software. The percent recovery was calculated according to the following formula [35]: Recovery (%) = (Ft − F0/Fb) × 100 (Ft fluorescence at each time point after photobleaching, F0: fluorescence at 0 s after photobleaching, Fb: fluorescence before photobleaching). To compare recovery (%) between cells, we normalized the fluorescence intensity of postbleaching to 0. Additionally, to assess tendencies of fluorescence kinetics in an image time series, we analyzed images by kymography using METAMORPH software (Universal Imaging Corp., Downington, PA, USA). We defined a horizontal rectangular region of interest (ROI), and intensities were measured within each image. Statistical analysis Statistical significance was calculated using the Mann– Whitney U test in SPSS software (version 12.0, SPSS Inc., Chicago, IL, USA). P values of 0.05 or less were considered significant.

Results Auditory cells expressed functional gap junctions Since ototoxicity causes apoptosis and gap junction communication has been described as either a pro- or an antiapoptotic mediator [36], we sought to study the effects of gap junction activity on the auditory cell line, HEI-OC1. To investigate whether Cx43 was expressed in these cells and whether gap junction channels could form to facilitate communication between adjacent cells, HEI-OC1 cells were grown to 90 % confluence on cover slips. As shown in Fig. 1a, we observed diffuse Cx43 expression in the cytoplasm and punctuate Cx43 expression in the plasma membranes between HEI-OC1 cells. Western-blot analysis also revealed that HEI-OC1 cells expressed Cx43 (Fig.  1a). In order to assess the function of gap junctions in HEI-OC1 cells, we used two parallel techniques: FRAP and SLDT assay. To test whether gap junctions could mediate intercellular communication, we compared contact cells with noncontact sequestered HEI-OC1 cells. HEI-OC1 cells were incubated in calcein AM, which passes readily through gap junctions. The cells were photobleached by exposure to about 5 s of intense light from an argon laser. The recovery of fluorescence intensity was monitored every 10 s for 400 s. As shown in Fig. 1b and c, contact with adjacent HEI-OC1 cells permitted recovery of about 42 % of the prebleach signal 400 s after photobleaching, whereas noncontact sequestered HEI-OC1 cells tended to exhibit a decrease in the rate of recovery by 16 %. In

13

Fig. 1  Expression of the gap junction protein Cx43 between adjacent ▸ HEI-OC1 cells and gap junction-mediated intercellular communication analysis. a Representative immunofluorescence staining of Cx43 (green) and DAPI (blue) in the auditory cell line HEI-OC1, human cervical epithelial carcinoma cell line HeLa (negative control), and human lung epithelial carcinoma cell line A549 (positive control). White arrows indicate the presence of gap junction plaques. a Western-blot analysis of Cx43 expression in HEI-OC1, HeLa, and rat heart cells (positive control). Equal protein loading was verified by β-actin expression. Gap junction activity was measured by either fluorescence recovery after photobleaching (FRAP) assay (b–d) or scrape loading dye transfer (SLDT) assay (e–f). b HEI-OC1 cell monolayers were incubated in calcein AM (2 μM, 10 min, green) and photobleached with a laser. Recovery of fluorescence in the photobleached cell was monitored every 10 s for 400 s. The region of bleaching is indicated by dotted lines with arrows. Representative confocal micrographs obtained prior to beaching (0 s, left panel), immediately after bleaching (25 s, middle panel), and after recovery (400 s, right panel) in HEI-OC1 cells in noncontact, contact, and treated with the gap junction inhibitor, 18α-GA (25 μM, 2 h). Scale bar 20 μm. c Summary of time course of FRAP data for many individual cells. For every bleached cell, the fluorescence signals present in each cell immediately prior to and immediately after photobleaching were normalized to 100 and 0 %, respectively. d Kymograph analysis of the movement of calcein AM dyes in HEI-OC1 cells in noncontact, contact, and treated with the gap junction inhibitor, 18α-GA. Time-lapse images were collected at 10-s intervals before and after photobleaching in the selected ROI. The horizontal lines of the kymograph represent the time of the image series. e Lucifer yellow was loaded by scrape line in monolayers of HEI-OC1 cells in noncontact, contact, and treated with the gap junction inhibitor, 18α-GA (25 μM, 2 h). Cells were monitored by fluorescence (lower) and phase-contrast (upper) microscopy. f Bar graph showing the percentage of dye-coupled cells in contact, noncontact, and the group treated with the gap junction inhibitor, 18αGA. Data are presented as the mean ± SD, p