Knockdown of C-MET Induced Apoptosis in ABCB1-Overexpressed Multidrug-Resistance Cancer Cell Lines

ORIGINAL ARTICLE Knockdown of c-MET induced apoptosis in ABCB1-overexpressed multidrug-resistance cancer cell lines

T-H Hung1, Y-H Li2, C-P Tseng1,2,3, Y-W Lan1, S-C Hsu4,5, Y-H Chen6, T-T Huang7, H-C Lai1,2, C-M Chen8,9,10, K-B Choo11 and K-Y Chong1,2,3

Inappropriate c-MET signaling in cancer can enhance tumor cell proliferation, survival, motility, and invasion. Inhibition of c-MET signaling induces apoptosis in a variety of cancers. It has also been recognized as a novel anticancer therapy approach. Furthermore, reports have also indicated that constitutive expression of P-glycoprotein (ABCB1) is involved in the HGF/c-MET- related pathway of multidrug resistance ABCB1-positive human hepatocellular carcinoma cell lines. We previously reported that elevated expression levels of PKCδ and AP-1 downstream genes, and HGF receptor (c-MET) and ABCB1, in the drug-resistant MES-SA/Dx5 cells. Moreover, leukemia cell lines overexpressing ABCB1 have also been shown to be more resistant to the tyrosine kinase inhibitor imatinib mesylate. These ﬁndings suggest that chemoresistant cancer cells may also develop a similar mechanism against chemotherapy agents. To circumvent clinical complications arising from drug resistance during cancer therapy, the present study was designed to investigate apoptosis induction in ABCB1-overexpressed cancer cells using c-MET-targeted RNA interference technology in vitro and in vivo. The results showed that cell viability decreased and apoptosis rate increased in c-MET shRNA- transfected HGF/c-MET pathway-positive MES-SA/Dx5 and MCF-7/ADR2 cell lines in a dose-dependent manner. In vivo reduction of tumor volume in mice harboring c-MET shRNA-knockdown MES-SA/Dx5 cells was clearly demonstrated. Our study demonstrated that downregulation of c-MET by shRNA-induced apoptosis in a multidrug resistance cell line.

Cancer Gene Therapy (2015) 22, 262–270; doi:10.1038/cgt.2015.15; published online 24 April 2015

INTRODUCTION Studies have shown that the inhibition of c-MET expression by 17 Multidrug resistance (MDR) is a major clinical obstacle in cancer RNA interference reduces growth rates and metastasis in glioma 18,19 therapy. P-glycoprotein 1 (permeability glycoprotein, abbreviated or enhanced chemosensitivity in multiple myeloma. However, as P-gp or Pgp), also known as MDR protein 1 (MDR1) or ATP- c-MET shRNA was significantly affected by cells proliferation, binding cassette subfamily B member 1 (ABCB1) or cluster of survival, invasiveness and metastasis in glioma, rhabdo- differentiation 243 (CD243), is a glycoprotein that in humans is myosarcoma, laryngeal carcinoma and breast ductal carcinoma – encoded by the ABCB1 gene. Overexpression of ABCB1 causes cells.17,20 23 The HGF/c-MET pathway is recently been considered effluxes of therapeutic drugs from cancer cells that prevent as a therapeutic target. A series of c-MET kinase inhibitors have effective treatment. As ABCB1 expression in normal tissues leads been developed for studies in cancer therapy include K252a,24 to higher toxicity in cancer treatment with an ABCB1 inhibitor,1,2 SU-1127425 and PHA665752.12 To date, several c-MET inhibitors, Researchers have focused on the molecular pathway of MDR such as cabozantinib and foretinib, are being used in clinical cancer cells as a strategy in the development new molecular trials.26 targets, including Pregnane X receptor,3 tyrosine kinase,4 cyclic It has been shown that constitutive expression of ABCB1 is guanosine monophosphate (cGMP)-specific phosphodiesterase involved in the HGF/c-MET-related pathway in ABCB1-positive type 5 (PDE5),5 and Wnt5A.6 human hepatocellular carcinoma cell lines.10,27 We previously The hepatocyte growth factor (HGF)-related pathway is involved reported that elevated expression levels of PKCδ and the AP-1 in cancer metastasis,7 proliferation,8 drug resistance9 and cell cycle downstream genes, HGF receptor (c-MET) and ABCB1 in the drug- progression.10 c-MET is a HGF receptor tyrosine kinase that is resistant MES-SA/Dx5 cells.28 Furthermore, leukemia cell lines abnormally expressed in different types of tumors.11–13 Aberrant overexpressing the ABCB1 gene have also been shown to be more binding of HGF to c-MET induces the activation of matrix metallo- resistant to the tyrosine kinase inhibitor imatinib mesylate.29,30 proteinases and urokinase plasminogen activator14,15 in numerous These findings suggest that chemoresistant cancer cells may types of cancers, including bladder, renal, cervical, colon, breast, also develop a similar mechanism against c-MET inhibitors. ovary, lung, esophagus, gastric and head and neck cancers.16 The present study focused on evaluating HGF/c-MET-related

1Graduate Institute of Biomedical Sciences, Division of Biotechnology, College of Medicine, Chang Gung University, Tao-Yuan, Republic of China; 2Department of Medical Biotechnology and Laboratory Science, College of Medicine, Chang Gung University, Tao-Yuan, Republic of China; 3Molecular Medicine Research Center, College of Medicine, Chang Gung University, Tao-Yuan, Republic of China; 4Cancer Molecular Diagnostic Laboratory, Chang-Gung Memorial Hospital, Lin-Kou Medical Center, Tao-Yuan, Republic of China; 5Department of Pathology, Chang-Gung Memorial Hospital, Lin-Kou Medical Center, Tao-Yuan, Republic of China; 6Graduate Institute of Pharmaceutical Sciences and Graduate Institute of Clinical Pharmacy, College of Medicine, National Taiwan University, Taipei, Republic of China; 7Center for Molecular and Clinical Immunology, College of Medicine, Chang Gung University, Tao-Yuan, Taiwan, Republic of China; 8Department of Life Sciences, National Chung Hsing University, Taichung, Republic of China; 9Agricultural Biotechnology Center, National Chung Hsing University, Taichung, Republic of China; 10Rong-Hsing Translational Medicine Center, National Chung Hsing University, Taichung, Republic of China and 11Department of Preclinical Sciences, Faculty of Medicine and Health Sciences and Centre for Stem Cell Research, Universiti Tunku Abdul Rahman, Selangor, Malaysia. Correspondence: Professor K-Y Chong, Department of Medical Biotechnology and Laboratory Science, College of Medicine, Chang Gung University, 259 Wen- Hwa 1st Road, Gui-Shan, Tao-Yuan, Taiwan 333, Republic of China. E-mail: [email protected] Received 15 March 2014; revised 10 February 2015; accepted 10 February 2015; published online 24 April 2015 c-MET shRNA-induced apoptosis in MDR cancer cells T-H Hung et al 263 pathway expression and applying the c-MET shRNA techniques to shRNA clone (CCTAAGGTTAAGTCGCCCTCGCTC) in a red fluroscence induce apoptosis in MDR cancer cells. In this work, two multidrug- protein-containing expression vector TRC009 were purchased from the resistant cancer cell lines, the human uterine sarcoma cell line National RNAi Core Facility (Academia Sinica, Taipei, Taiwan, Republic of MES-SA/Dx5 and the breast adenocarcinoma cell line MCF-7/ China). ADR2, were cultured in the presence of doxorubicin; ABCB1 was also overexpressed for analysis of the effects in this study. We Viral production and viral transduction noted that the expression level of HGF and its downstream genes, Virus stocks were prepared by cotransfecting the pLenti-LucEGFP, c-MET HGF receptor (c-MET) and urokinase plasminogen activator (uPA) shRNA plasmid or control scramble shRNA plasmid with three packaging were upregulated in drug-resistant MDR cancer cells. We also plasmids, pMDLg/pRRE, CMV-VSVG and RSV-Rev, into 293 T cells. The viral showed that cell viability was decreased in c-MET shRNA- supernatants were collected 36–48 h later, filtered and centrifuged at transfected MES-SA/Dx5 cells in a dose-dependent manner both 20 000 × g for 90 min. The viral titer was determined by the method of end point dilution through counting the number of infected red cells at × 100 in vivo and in vitro, suggesting that HGF/c-MET signaling has an fi fl important role in MDR cancer cells. magni cation under a uorescence microscope (Nikon, Tokyo, Japan) 96 h after infection to 293 T cells. Titer in transducing units was computed as follows: (TU)/ml = (the numbers of red fluorescent cells) × (dilution factor)/ (volume of virus solution). Viral particle was quantified by HIV quantifica- MATERIALS AND METHODS tion ELISA kit. MES-SA/Dx5 cells were seeded in 12-well plate and the cell Cell culture was transduced with equal viral particle of pLenti-LucEGFP virus particle The multidrug-resistant cell line MES-SA/Dx5 (CRL-1977; ATCC, Manassas, following the method of Chan et al.33 and the stably transducted cells were VA, USA) was established from the human sarcoma cell line MES-SA designated MES-SA/Dx5-LG. (CRL-1976; ATCC) in the presence of increasing doxorubicin concentrations as described previously.31 All the cell lines were continuously maintained ’ Transient transfection in McCoy s 5A medium supplemented with 10% fetal bovine serum and 5 2mM antimycotics (Invitrogen Corp., Carlsbad, CA, USA) under 95% air For transient transfection, cells were plated at a density of 1 × 10 cells per and 5% CO at 37 °C. The MES-SA/Dx5 cell line was maintained in the well in a 24-well plate and were transfected with DNA of indicated 2 μ continuous presence of 1.7 μM doxorubicin (Sigma-Aldrich, St Louis, MO, concentrations (Promega, Madison, WI, USA) premixed with 2 l Lipofecta- USA). The multidrug-resistant cell line MCF-7/ADR2 was established from mine 2000 (Invitrogen) for 48 h. At the end of the incubation, cells were the human mammary gland adenocarcinoma cell line MCF-7 (ATCC HTB22) either lysed by the passive lysis buffer (Promega) to perform caspase 3/7 in the presence of increasing doxorubicin concentrations followed by assays and apoptosis analysis, or were assayed by cell viability assay and previously described.32 The MCF-7/ADR2 cell line was maintained in the colony formation. continuous presence of 0.25 μM doxorubicin (Sigma-Aldrich). Both cell lines were be grown in Dulbecco’s modified Eagle’s medium supplemented with Quantitative real-time RT-PCR 10% fetal bovine serum and 2 mM antimycotics (Invitrogen). The RT-PCR was performed as previously described.34 Briefly, total RNA (2 μg) was prepared from the transfectants and was treated with DNase I. Western blot analysis The RNAs were reverse transcribed into cDNAs at 42 °C for 60 min using Total cellular proteins were isolated from cell lines by the PRO-PREP Protein Moloney Murine Leukemia Virus Reverse Transcriptase (Invitrogen). After Extraction Solution (Intron Biotechnology, Kyonggi-do, Korea). Nuclear the oligo (dT)-primed reverse transcription reaction, real-time PCR was and cytoplasmic proteins were isolated using the NE-PER Nuclear and performed in a LightCycler 480 (Roche, Mannheim, Germany) in 96-well Cytoplasmic Extraction Reagents (Pierce, Rockford, IL, USA). Western blot plates. The reaction mixture was 1 μl RT product, 15 μl RealQ-PCR Master was performed as described previously.28 Briefly, approximately 25 or Mix Kit (Ampliqon AqS, Odense, Denmark), 1 μl each of 10 μM forward and 50 μg total proteins was loaded onto each lane and the proteins were reverse primers, and the complete total volume was adjusted to 10 μl with separated in sodium dodecyl sulphate-polyacrylamide gel electrophoresis nuclease-free water. The real-time PCR program was: preincubation at 50 °C gels. After electrophoresis, the resolved proteins were transferred to a for 2 min, initial denaturation at 95 °C for 7 min and 45 cycles at 95 °C for PVDF membrane (Millipore, Billerica, MA, USA) and subsequently stained 10 s, 63 °C for 15 s and 72 °C for 30 s. The program was terminated by a with Ponceau S (Sigma-Aldrich) to confirm complete protein transfer. The final extension at 60 °C for 1 min and cooling at 40 °C for 5 min. For membranes were blocked with 5% nonfat dry milk in PBS-T (0.1% Tween- normalization, the mRNA level of the β-actin gene in each RNA preparation ΔΔ 20 in phosphate-buffered saline (PBS), pH7.4) for 1 h and probed overnight was determined. Relative gene expression was determined by the Ct with the following antisera at appropriate dilutions: 1:1000 dilution of the method, where Ct = threshold cycle. The relative targeted mRNA levels anti-MDR-1 (sc-13131, Santa Cruz Biotechnology, Dallas, TX, USA), a 1:1000 were normalized to the mRNA level of the reference β-actin gene. The dilution of the anti-HGF(sc-1357, Santa Cruz Biotechnology), a 1:1000 melting curve of the amplification product was always checked to ensure a dilution of the anti-c-MET (1996-1, Epitomics, Burlingame, CA, USA), a single clean peak that represented good-quality real-time PCR data. 1:1000 dilution of the anti-PARP and anti-Cleaved-PARP (MAB3290 and MAB3565, (Millipore), and a 1:10 000 dilution of the anti β-actin (MAB1501, – fi Cell viability MTT assay Millipore) antiserum in PBS Tween-20. Identi cation of each protein was 5 achieved with the Western Lighting Plus Reagent (Perkin Elmer, Waltham, Cells were plated at a density of 1 × 10 cells per well in a 24-well plate. The MA, USA) using an appropriate alkaline phosphatase-conjugated second- cells were then treated with the indicated concentrations of various drugs ary antibody. The level of each protein in the western blot analysis was in different sets of experiments. After 48 h of drug incubation, the medium detected by the LAS-3000 chemiluminescence detection device (Fujifilm, was removed and PBS was used to wash the cells. Thiazolyl Blue μ Valhalla, NY, USA). To adjust for loading differences, the optical density of Tetrazolium (Sigma-Aldrich; 200 l) was added to each well and was μ each protein was normalized to that of the β-actin band. incubated with the cells at 37 °C for 2 h. Subsequence, 400 l dimethyl sulfoxide was added to each well and incubation at 37 °C was continued for 20 min. Absorbance of the mixture was read at 540 nm using a Plasmid construction Microplate Reader (VersaMax, Molecular Devices, Sunnyvale, CA, USA). Cell A cDNA encoding the fusion protein of Luciferase-EGFP, pL3-TRE-LucGFP-2L, viability (%) was calculated as the ratio of the surviving cells in each drug- was obtained from Addgene (Cambridge, MA, USA). The 2.5-kb fragment treated experiment set to that of the control. containing the complete coding region of Luciferase-EGFP was released from pL3-TRE-LucGFP-2L by SacIl and EcoRV double digestion and was fi Colony-formation assay lled in. The DNA fragment was then ligated into of lentiviral vector, pLenti 5 6.2 (Life Technologies), and designated pLenti-LucEGFP. The cells were plated at a density of 5 × 10 cells per well in a 6-well plate and grown for 24 h. After c-MET shRNA transfection, the cells were grown in the replaced fresh medium for 24 h. The cells were then replated at a shRNA density of 100 cells per 10 cm2 dish and grown for 7–8 days until discrete The c-MET shRNA clone (TRCN0000121248; CCTTCAGAAGGTTGCTGAGTA) colonies were visualized. After washing with PBS, the colonies were stained targeted at the human MET transcript, and the control scramble with 0.5% crystal violet and counted. After cell counting, cell number was

© 2015 Nature America, Inc. Cancer Gene Therapy (2015), 262 – 270 c-MET shRNA-induced apoptosis in MDR cancer cells T-H Hung et al 264 estimated by dissolving the crystal violet in 70% ethanol (4 ml per well) with anti-c-MET (Epitomics) or anti-active caspase antibodies (Sigma- and then optical density values measured at 562 were used as previously Aldrich) using the sensitive Dako-REAL, Alkaline-Phosphatase/RED detec- reported.35 In each group, cells transfected with the control scramble tion system (Dako, Glostrup, Denmark). Hematoxylin was used for shRNA were set at 100%. Results (mean ± s.d.) represent data from counterstaining according to the manufacturer’s protocol. triplicate wells. Statistical analysis Caspase 3/7 assay The surviving fraction and the relative luminescence unit were measured MES-SA and MES-SA/Dx5 cells were plated at a density of 8 × 103 cells per in triplicate samples and were expressed as mean ± s.d. The Student’s t-test well in a 96-well plate. The cells were then treated with the indicated was used for statistical analysis. Po0.05 was considered as statistically dosages of PHA665752 and/or verapamil for 48 h. At the end of the drug significant. treatment, 100 μl Caspase-Glo 3/7 Reagent (Promega) was added to the treated cells in each well. After 30 min of incubation at room temperature, the relative luminescence unit was measured by GLOMAX 20/20 RESULTS Luminometer (Promega) and as an indication of Caspase 3/7 apoptotic Elevated expression of HGF pathway-related genes in the MDR activity. cells The MES-SA/Dx5 and MCF7/ADR2 cells, which overexpressed Flow cytometric analysis of apoptosis ABCB1 but not ABCC1 and ABCG2, were selected from MES-SA In preparation of flow cytometry, pc-MET-shRNA plasmid-transfected cells and MCF7 cell lines by doxorubicin treatment (Figure 1).36,37 Real- were collected at the indicated time points. The cells were then stained time PCR quantification showed that the expression levels of with Annexin V and propidium iodine by using the Annexin V-FITC ABCB1 was increased by 4300-fold in MES-SA/Dx5 (Figure 1a) and Apoptosis Detection Kit (Strong Biotech Corp, Taipei, Taiwan, Republic of 4 China) following the manufacturer’s protocol. by 30-fold in MCF-7/ADR2 (Figure 1b). Western blot analysis showed that ABCB1 expression occurred only in MES-SA/Dx5 cells and MCF-7/ADR2 cells (Figure 2a). To characterize modulation of Tumor xenografts in nude mice the HGF pathway in the drug-resistant cells, expression levels of Eight-week-old male BALB/c nude mice were purchased from the National genes involved in the HGF pathway, including HGF and c-MET, Laboratory Animal Center (Taipei, Taiwan, Republic of China). The were evaluated in MES-SA/Dx5 and MCF-7/ADR2 cell lines by Institutional Animal Care committee of Chang Gung University approved western blot analysis. The data indicated that HGF and c-MET these experiments. The mice were maintained in an air-conditioned animal facility under constant temperature and humidity conditions with a 12:12 expression levels were increased in the MES-SA/Dx5 and MCF-7/ light–dark cycle and the mice were allowed ad libitum in the diet or ADR2 cell lines when compared with the respective parental cells drinking water. Mice were randomly picked to different groups and each (Figure 2a). Real-time PCR quantification showed that the group had at least five or more mice. c-MET shRNA or control scramble expression levels of HGF, c-MET and uPA were increased by shRNA-transfected MES-SA/Dx5 cells (1 × 106) was subcutaneous trans- 43-fold in MES-SA/Dx5 (Figure 2b) and by 41.5-fold in MCF-7/ planted into BALB/c nude mice as previously described.35 Tumor volumes ADR2 (Figure 2c) when compared with the respective parental (V) were calculated as V (mm3)=(L × W2)/2, where L was the largest cells. Thus, these data indicated that the HGF pathway was diameter and W was the diameter perpendicular to L. activated in the drug-resistant cells.

c-MET shRNA lentivirus intratumor injection Decreased cell viability in c-MET shRNA-transfected MDR cells Eight-week-old female BALB/c nude mice were purchased from the As the HGF pathway was activated in the drug-resistant MES-SA/ National Laboratory Animal Center (Taipei, Taiwan, Republic of China). Dx5 and MCF-7/ADR2 cells, we further used RNA interference to All the in vivo experimental protocols were approved by the Chung Gung University Animal Center. The mice were anesthetized. Cultured reverse c-MET expression for reducing cell viability in these cells. tumor cells were removed by trypsinization, washed twice with PBS Cell viability MTT assay and colony-formation assay were used to and suspended in McCoy’s 5a medium. MES-SA/Dx5-LG cells (1 × 106) were monitor the viability of the MES-SA/Dx5 and MCF7/ADR2 cell lines mixed with Matrigel (BD Bioscience, San Jose, CA, USA) followed by that were transfected with c-MET shRNA in comparison with subcutaneous injection into the dorsal region near the thigh of the mouse transfection with scrambled shRNA as the control at different as previously described.6 Upon tumor establishment, mice were allocated dosages. The cell viability MTT assay showed that the survival rate to the following groups: (a) scramble shRNA as the control group; (b) c-MET 3 of both cell lines declined with increasing concentration of c-MET shRNA as the experiment group. Upon reaching a volume of 500 mm , shRNA transfection (Figure 3a and b). Compared with MCF7/ADR2 7 μ intratumoral injections of viral particles at a titer of 10 TU in 100 l PBS cells, the survival rate of MES-SA/Dx5 cells was more greatly were initially performed. The tumor volume was measured every 7 days up to 3 weeks, and the volume of a tumor was calculated using the formula affected by c-MET shRNA transfection (Figure 3a and b). V (mm3)=(L × W2)/2 where L and W indicate the length and width of a Furthermore, the colony-formation ability was decreased in the tumor, respectively. On day 21 treatment with doxorubicin, all mice were c-MET shRNA-transfected MES-SA/Dx5 cells (Figure 3c and e) and killed and tumors were resected, weighed and frozen or fixed in formalin MCF7/ADR2 cells (Figure 3d) in a dose-dependent manner and paraffin embedded for immunohistochemical studies. compared with the control set. To verify induction of apoptosis by c-MET shRNA in MDR cells, Bioluminescent IVIS imaging system caspase 3/7 activity assay and detection of annexin V apoptosis fi Xenografted nude mice were kept in a chamber and were anesthetized were performed. c-MET expression was signi cantly reduced in with 2% isofluorane/oxygen mixture and i.p. injection with 20 mg ml− 1 c-MET shRNA-transfected MES-SA/Dx5 cells (Figure 4a), but not in − 1 D-luciferin (Promega) in PBS (200 mg kg ). Mice were sedated using the control using scrambled shRNA-transfected MES-SA/Dx5 cells isoflurane, and live anesthetized mice were imaged using the Biolumines- (data not shown). The cleavage of the caspase 3 substrate PARP cent IVIS Imaging system (Xenogen Corp., Alameda, CA, USA). The resulting increased in c-MET shRNA-transfected MES-SA/Dx5 and MCF7/ light emission was quantified using the LivingImage software (Xenogen ADR2 cells (Figure 4b). Furthermore, caspase 3/7 activities were − − − Corp.). Raw values were reported as photons s 1 cm 2 sr 1. found to increase two- and three-fold in c-MET shRNA-transfected MES-SA/Dx5 cells when transfected with 100 and 200 ng c-MET Immunohistochemical staining studies shRNA, respectively (Figure 4c). The caspase 3/7 activity was also Paraffin sections from c-MET-transfected MES-SA/Dx5 xenograft tumor increased 2- and 2.5-fold in MCF7/ADR2 cells at 100 and 200 ng specimen were microwave heated (750-W, three 5-min cycles) in 10 mM c-MET shRNA, respectively (Figure 4d). Apoptotic cell death citrate, pH 6.0 or 1 mM EDTA, pH 8.0 and immunostaining was performed increased significantly in the c-MET-transfected MES-SA/Dx5 and

Figure 1. ABCB1 was upregulated in MES-SA/Dx5 and MCF7/ADR2 cells. mRNA expression levels of ABCB1, ABCG2 or ABCC1 in (a) MES-SA and MES-SA/Dx5 or (b) MCF-7 and MCF-7/ADR2 cells. Expression levels were normalized to β-actin. Data represent means ± s.d. of three independent experiments. **Po0.01 indicate the differences between the drug-resistant and the parental cells.

Figure 2. Upregulation of hepatocyte growth factor (HGF)/c-MET signaling in MDR cells. (a) Protein expression of ABCB1, HGF and MET of MES-SA and MES-SA/Dx5 cells or MCF-7 and MCF-7/ADR2 cells, mRNA expression levels of HGF, c-MET or uPA in (b) MES-SA and MES-SA/Dx5 or (c) MCF-7 and MCF-7/ADR2 cells. Expression levels were normalized to β-actin. Data represent means ± s.d. of three independent experiments. *Po0.05 and **Po0.01 indicate the differences between the drug resistant and the parental cells.

MCF7/ADR2 cells at days 3 and 7 posttransfection (Figure 4e comparison with the control scrambled shRNA-treated group and f). The data indicated that downregulation of c-MET by RNA (Figure 5a) from weekly monitoring of the tumor size in the interference signiﬁcantly reduced cell viability and induced injected mice. As c-MET shRNA was observed to markedly induce apoptosis of the drug-resistant MES-SA/Dx5 and MCF7/ADR2 cells apoptosis in vitro (see Figure 4 above) and that c-MET-depleted in a dose-dependent manner. MES-SA/Dx5 cells showed retarded tumor growth in vivo,we further evaluated the effects of intratumoral injection of c-MET shRNA lentivirus into the MES-SA/Dx5-LG subcutaneous xenograft Reduction of tumor volume in c-MET shRNA-knockdown tumors in the mouse model. At day 35 after the xenograft, there MES-SA/Dx5 cells in vivo were signiﬁcant decreases in the tumor volume of the c-MET To assess the effects of c-MET knockdown on drug-resistant cells shRNA lentiviral vector transduction group when comparison with in an animal model, c-MET shRNA-transfected MES-SA/Dx5 cells the control scrambled shRNA vector-treated group (Figure 5b). were injected subcutaneously into nude mice (n = 5 per group). The IVIS luciferase imaging results further indicated that c-MET c-MET shRNA was found to substantially delay tumor growth in shRNA lentiviral transduction led to slower tumor growth when

Figure 3. Reduction of cell viability in transient c-MET–shRNA-knockdown MDR cells. (a) MES-SA/Dx5 or (b) MCF-7/ADR2 cells were transfected with the indicated concentrations of c-MET–shRNA plasmid for 48 h. At the end of the treatment, cells were collected for measurement of viability by using the MTT assay. (c) MES-SA/Dx5 and (d) MCF-7/ADR2 cells were transfected with the indicated concentrations of c-MET-shRNA plasmid for 48 h. At the end of the treatment, cells were collected for measurement of viability by using the colony-formation assay and colony-formation was scored after 7 days. (e) Representative images of colony-forming assay from MES-SA/Dx5 cells. Data represent means ± s.d. of three independent experiments. *Po0.05 indicate the differences between the c-MET–shRNA-knockdown MDR cells and the respective untreated controls. *Po0.05, the viability are expressed as the mean ± s.d. (n = 3).

compared with the control scramble shRNA (Figure 5c and d). The ABCB1 and the breast cancer resistance protein were upregulated tumor sizes of the xenografts of the c-MET shRNA lentivirus- in the cells. Lasagna et al.27 reported that HGF interacted with its transduced MES-SA/Dx5 cells were reduced 4-fold in the mean receptor c-MET controlled Ets-1 for angiogenesis activity in an relative tumor volume compared to those of the vector control cells ABCB1-positive human hepatocellular carcinoma cell line. Hung 28 at 21 days of c-MET shRNA treatment (Figure 5e and f). Further- et al. reported that FZD1 regulated PKCδ and its signaling more, xenografts of c-MET shRNA lentivirus-transduced MES-SA/ transduction pathway, such as the AP-1 downstream genes, HGF Dx5 cells showed induced caspase 3 activation (Figure 5g). Taken receptor (c-MET) and ABCB1 has an important role in drug together, c-MET knockdown inhibited tumor formation in vivo. resistance in the human uterine sarcoma cell line MES-SA/Dx5. In addition, siRNA knockdown of the ABCB1 gene led to decreased HGF expression.27 Furthermore, ABCB1 was reported to be involved in HGF/MET autocrine loop in celecoxib-induced DISCUSSION autophagy and cell cycle arrest in ABCB1-expressing hepatocel- Involvement of the HGF/c-MET pathway in regulating drug lular carcinoma cell lines.10 In this work, higher expression levels of resistance in MDR cancer cells HGF pathway-related genes were found in the drug-resistant Le Vee et al.38 investigated the expression levels of drug MCF-7/ADR2 and MES-SA/Dx5 cell lines (Figure 2). Therefore, our resistance-related proteins in human hepatocytes and found that data supported that HGF signaling had an important role in MDR

Figure 4. Induction of apoptosis in transient c-MET shRNA-knockdown MDR cells. (a) Protein expression of MET of indicated cells after c-MET shRNA transfection for 48 h in western blot analysis. (b) Expression of PARP and cleaved PARP of indicated cells after c-MET shRNA transfection for 48 h in western blot analysis. MDR cells were transfected with the indicated concentrations of c-MET–shRNA plasmid for 48 h. At the end of the treatment, cells were collected for measurement of caspase 3/7 activities of shRNA-transfected (c) MES-SA/Dx or (d) MCF-7/ADR2 cells using the Caspase-Glo 3/7 assay. The activity was detected by luminescence and was proportional to its intensity which was normalized by cell viability. Annexin V-stained cells of shRNA-transfected (e) MES-SA/Dx5 or (f) MCF-7/ADR2 cells using the Flow cytometric analysis at the indicated time points. Data represent means ± s.d. of three independent experiments. *P o0.05 or **P o0.01 indicate the differences between the transient c-MET shRNA-knockdown MDR cells and the respective untreated controls.

of cancer cells. As several c-MET inhibitors have been approved for Furthermore, Bcr-abl-positive leukemic and ABCB1-expressing cells clinical trials, selected components of the HGF pathway might be were resistant to the tyrosin kinase inhibitor, imatinib.29,30 Brendel effective therapeutic targets in MDR cancers. et al.39 found that resistance to imatinib and nilotinib in ABCG2- transduced K562 cells occurred via P-CRKL downregulation. More- over, CGP74588, which is a metabolite of the tyrosine kinase inhibitor Involvement of the P-glycoprotein in regulating drug resistance Gleevec, has lower apoptosis and antiproliferation abilities in resistant against the c-MET inhibitor in MDR cancer cells K562/Dox cells,40 indicating that drug resistance of tyrosine kinase Many c-MET kinase inhibitors have been developed for inhibiting inhibitors occurred in ABCB1-expressing cells. Moreover, the 12,24,25 the HGF/MET pathway in cancers, but drug resistance to kinase combination of verapamil and tyrosine kinase inhibitor raises the inhibitors was discovered only in recent years. Drugresistanceof risk of cumulative cardiac toxicity.41 In addition, the evidences kinase inhibitors was also reported in several leukemic cell lines or indicated that tyrosine kinase inhibitor geﬁtinib resistance was clinical cases resistant to treatment with tyrosine kinase inhibitors. found in ABCG2-transduced human lung cancer PC-9 cells.42

Figure 5. Reduction of tumor formation ability in c-MET shRNA-knockdown MES-SA/Dx5 cells in vivo.(a) Tumor growth curve at indicated time points of c-MET shRNA or control scramble shRNA vector-transfected MES-SA/Dx5 cells were subcutaneous transplanted into nude mice. (b) Intratumoral injections of c-MET shRNA lentiviral particles into the MES-SA/Dx5-LG xenografted nude mice. (c) Tumor formation and (d) tumor luciferase activity of c-MET shRNA (left) and control scramble shRNA (right) after shRNA lentivirus intratumor injection was monitored by BLI of c-MET shRNA and control scramble shRNA cells after shRNA lentivirus intratumor injection monitored by using BLI at day 21 and 35. (e) tumor and (f) tumor weight of c-MET shRNA and control scramble shRNA-transduced MES-SA/Dx5 cells. (g) Immunohistochemistry staining of c-MET and activated caspase 3 in c-MET shRNA- and control scramble shRNA-transduced MES-SA/ Dx5 cells.

Enhancement of apoptosis and reduction of tumorigenicity in proliferation, adherence and invasiveness of human multiple c-MET shRNA-treated MDR cells in vivo and in vitro myeloma U266 cells as well as increased the chemosensitivity of c-MET inhibition by RNA interference or speciﬁc inhibitors as a these cells to doxorubicin. Moreover, our results demonstrated therapeutic approach has been reported for several types of that c-MET shRNA in MDR cells induced apoptosis and increased cancers. Such an application in NSCLC and SCLC cells has been the accumulation of cleavage PARP and Caspase 3/7 activities shown to lead to reduced viability and proliferation in vitro and (Figure 4). We have also shown that c-MET shRNA-induced in vivo.43,44 Several reports have demonstrated that the use of apoptosis and reduced tumorigenicity in MDR cells in vivo (Figure 5). Taken together, our data showed that the c-MET either viral or nonviral vector of c-MET shRNA delivery system 45 shRNA-induced silencing of HGF/c-MET signaling shows greater inhibited proliferation, tumorigenicity, migration and invasion potential in MDR cancer therapy. abilities of cancer cell lines.46,47 Likewise, c-MET shRNA also induced apoptosis17,48 as well as inhibited c-MET-mediated Erk signaling.49 Furthermore, Que et al.19 observed that the down- CONFLICT OF INTEREST regulation of c-MET by short-hairpin RNA inhibited the The authors declare no conﬂict of interest.

Cancer Gene Therapy (2015), 262 – 270 © 2015 Nature America, Inc. c-MET shRNA-induced apoptosis in MDR cancer cells T-H Hung et al 269 ACKNOWLEDGMENTS 20 Taulli R, Scuoppo C, Bersani F, Accornero P, Forni PE, Miretti S et al. Validation of This work was supported in part by the Taiwan Ministry of Science and Technology, met as a therapeutic target in alveolar and embryonal rhabdomyosarcoma. – NSC 95-2320-B-182-028-MY3 and MOST 103-2320-B-182-021 to KYC, NSC 99-2632- Cancer Res 2006; 66: 4742 4749. B-182-001-MY3 to CPT, NSC 102-2815-C-182-062-B to YHL. The Chang Gung Memorial 21 Phillip CJ, Stellrecht CM, Nimmanapalli R, Gandhi V. Targeting MET transcription as a therapeutic strategy in multiple myeloma. Cancer Chemother Pharmacol 2009; Hospital Grant (CMRPD 34012, CMRPD 180133, CMRPD1B0472, CMRPD1B0473) and 63:587–597. Taiwan Ministry of Education (EMRPD1C0121). The IVIS imaging facilities were 22 Ji CY, Xie ZN, Chen JC, Wang YN, Guan LL, Li HT et al. [RNA interference targeting supported by the SPF Animal Center Animal Image Laboratory, Chang Gung c-Met inhibits proliferation of human laryngeal carcinoma Hep-2 cell line in vitro University. The tissue slide imaging by HistoFAXS of Molecular Imaging Center, Chang and in its xenografts in nude mice]. Zhonghua Er Bi Yan Hou Tou Jing Wai Ke Za Gung Memorial Hospital, Linkou Medical Center, TaoYuan. Taiwan. Zhi 2010; 45:61–67. 23 Lee SJ, Seol HJ, Lee HW, Kang WY, Kang BG, Jin J et al. Gene silencing of c-Met leads to brain metastasis inhibitory effects. Clin Exp Metastasis 2013; 30: REFERENCES 845–854. 24 Morotti A, Mila S, Accornero P, Tagliabue E, Ponzetto C. K252a inhibits 1 Lum BL, Gosland MP. MDR expression in normal tissues. Pharmacologic implica- the oncogenic properties of Met, the HGF receptor. Oncogene 2002; 21: tions for the clinical use of P-glycoprotein inhibitors. Hematol Oncol Clin North Am 4885–4893. 1995; 9: 319–336. 25 Arena S, Pisacane A, Mazzone M, Comoglio PM, Bardelli A. Genetic targeting of 2 Thomas H, Coley HM. Overcoming multidrug resistance in cancer: an update on the kinase activity of the Met receptor in cancer cells. Proc Natl Acad Sci USA 2007; the clinical strategy of inhibiting p-glycoprotein. Cancer Control 2003; 10: 104: 11412–11417. 159–165. 26 Sharma N, Adjei AA. In the clinic: ongoing clinical trials evaluating c-MET- 3 Vadlapatla RK, Vadlapudi AD, Pal D, Mitra AK. Mechanisms of drug resistance in inhibiting drugs. Ther Adv Med Oncol 2011; 3:S37–S50. cancer chemotherapy: coordinated role and regulation of efflux transporters and 27 Lasagna N, Fantappie O, Solazzo M, Morbidelli L, Marchetti S, Cipriani G et al. metabolizing enzymes. Curr Pharm Des 2013; 19: 7126–7140. Hepatocyte growth factor and inducible nitric oxide synthase are involved in 4 Wang YJ, Zhang YK, Kathawala RJ, Chen ZS. Repositioning of tyrosine kinase multidrug resistance-induced angiogenesis in hepatocellular carcinoma cell lines. inhibitors as antagonists of atp-binding cassette transporters in anticancer drug Cancer Res 2006; 66: 2673–2682. resistance. Cancers (Basel) 2014; 6: 1925–1952. 28 Hung TH, Chen CM, Tseng CP, Shen CJ, Wang HL, Choo KB et al. FZD1 activates 5 Shi Z, Tiwari AK, Shukla S, Robey RW, Singh S, Kim IW et al. Sildenafil reverses protein kinase C delta-mediated drug-resistance in multidrug-resistant MES-SA/ ABCB1- and ABCG2-mediated chemotherapeutic drug resistance. Cancer Res Dx5 cancer cells. Int J Biochem Cell Biol 2014; 53:55–65. 2011; 71: 3029–3041. 29 Chen L, Wang JM, Xu XP, Gao L, Fei XH, Lou JW et al. [The reverse effect on drug- 6 Hung TH, Hsu SC, Cheng CY, Choo KB, Tseng CP, Chen TC et al. Wnt5A regulates resistance against tyrosine kinase inhibitor STI571 in mdr1 and bcr-abl positive ABCB1 expression in multidrug-resistant cancer cells through activation of the leukemic cells]. Zhongguo Shi Yan Xue Ye Xue Za Zhi 2003; 11:600–603. non-canonical PKA/β-catenin pathway. Oncotarget 2014; 5: 12273–12290. 30 Mahon FX, Belloc F, Lagarde V, Chollet C, Moreau-Gaudry F, Reiffers J et al. MDR1 7 Locatelli A, Lofgren KA, Daniel AR, Castro NE, Lange CA. Mechanisms of HGF/Met gene overexpression confers resistance to imatinib mesylate in leukemia cell signaling to Brk and Sam68 in breast cancer progression. Horm Cancer 2012; 3: line models. Blood 2003; 101: 2368–2373. – 14 25. 31 Chong KY, Hsu CJ, Hung TH, Hu HS, Wang TH, Wang CH et al. Wnt pathway 8 Mueller KL, Madden JM, Zoratti GL, Kuperwasser C, List K, Boerner JL. Fibroblast- activation and ABCB1 expression account for attenuation of proteasome secreted hepatocyte growth factor mediates epidermal growth factor receptor Inhibitor-mediated apoptosis in multidrug-resistant cancer cells. Cancer Biol Ther tyrosine kinase inhibitor resistance in triple-negative breast cancers through 2015; 16:149–159. paracrine activation of Met. Breast Cancer Res 2012; 14: R104. 32 Ke W, Yu P, Wang J, Wang R, Guo C, Zhou L et al. MCF-7/ADR cells (re-designated 9 Chen CT, Kim H, Liska D, Gao S, Christensen JG, Weiser MR. MET activation NCI/ADR-RES) are not derived from MCF-7 breast cancer cells: a loss for breast fi mediates resistance to lapatinib inhibition of HER2-ampli ed gastric cancer cells. cancer multidrug-resistant research. Med Oncol 2011; 28: S135–S141. Mol Cancer Ther 2012; 11:660-669. 33 Chan AW, Chong KY, Martinovich C, Simerly C, Schatten G. Transgenic monkeys 10 Mazzanti R, Platini F, Bottini C, Fantappie O, Solazzo M, Tessitore L. Down- produced by retroviral gene transfer into mature oocytes. Science 2001; 291: regulation of the HGF/MET autocrine loop induced by celecoxib and mediated by 309–312. P-gp in MDR-positive human hepatocellular carcinoma cell line. Biochem Phar- 34 Huang TT, Chong KY, Ojcius DM, Wu YH, Ko YF, Wu CY et al. Hirsutella sinensis – macol 2009; 78:21 32. mycelium suppresses interleukin-1β and interleukin-18 secretion by inhibiting 11 Li Y, Chen CQ, He YL, Cai SR, Yang DJ, He WL et al. Abnormal expression of both canonical and non-canonical inflammasomes. Sci Rep 2013; 3: 1374. E-cadherin in tumor cells is associated with poor prognosis of gastric carcinoma. 35 Chong KY, Lai CC, Su CY. Inducible and constitutive HSP70s confer synergistic – J Surg Oncol 2012; 106: 304 310. resistance against metabolic challenges. Biochem Biophys Res Commun 2013; 430: 12 Crosswell HE, Dasgupta A, Alvarado CS, Watt T, Christensen JG, De P et al. 774–779. PHA665752, a small-molecule inhibitor of c-Met, inhibits hepatocyte growth 36 Calcagno AM, Ambudkar SV. Molecular mechanisms of drug resistance in factor-stimulated migration and proliferation of c-Met-positive neuroblastoma single-step and multi-step drug-selected cancer cells. Methods Mol Biol 2010; 596: cells. BMC Cancer 2009; 9: 411. 77–93. 13 Berthou S, Aebersold DM, Schmidt LS, Stroka D, Heigl C, Streit B et al. The Met 37 Calcagno AM, Salcido CD, Gillet JP, Wu CP, Fostel JM, Mumau MD et al. Prolonged kinase inhibitor SU11274 exhibits a selective inhibition pattern toward different drug selection of breast cancer cells and enrichment of cancer stem cell char- receptor mutated variants. Oncogene 2004; 23: 5387–5393. acteristics. J Natl Cancer Inst 2010; 102: 1637–1652. 14 Pepper MS, Matsumoto K, Nakamura T, Orci L, Montesano R. Hepatocyte growth 38 Le Vee M, Lecureur V, Moreau A, Stieger B, Fardel O. Differential regulation of drug factor increases urokinase-type plasminogen activator (u-PA) and u-PA receptor transporter expression by hepatocyte growth factor in primary human hepato- expression in Madin-Darby canine kidney epithelial cells. J Biol Chem 1992; 267: cytes. Drug Metab Dispos 2009; 37: 2228–2235. 20493–20496. 39 Brendel C, Scharenberg C, Dohse M, Robey RW, Bates SE, Shukla S et al. Imatinib 15 Jeffers M, Rong S, Vande Woude GF. Enhanced tumorigenicity and invasion- mesylate and nilotinib (AMN107) exhibit high-affinity interaction with ABCG2 on metastasis by hepatocyte growth factor/scatter factor-met signalling in human primitive hematopoietic stem cells. Leukemia 2007; 21:1267–1275. cells concomitant with induction of the urokinase proteolysis network. Mol Cell 40 Mlejnek P, Dolezel P, Faber E, Kosztyu P. Interactions of N-desmethyl imatinib, an Biol 1996; 16: 1115–1125. active metabolite of imatinib, with P-glycoprotein in human leukemia cells. Ann 16 Organ SL, Tsao MS. An overview of the c-MET signaling pathway. Ther Adv Med Hematol 2011; 90:837–842. Oncol 2011; 3:S7–S19. 41 Haouala A, Widmer N, Duchosal MA, Montemurro M, Buclin T, Decosterd LA. Drug 17 Chu SH, Feng DF, Zhang H, Chen ET, Duan ZX, Li XY et al. c-Met-targeted RNA interactions with the tyrosine kinase inhibitors imatinib, dasatinib, and nilotinib. interference inhibits growth and metastasis of glioma U251 cells in vitro. J Neu- Blood 2011; 117:e75–e87. rooncol 2009; 93: 183–189. 42 Katayama K, Shibata K, Mitsuhashi J, Noguchi K, Sugimoto Y. Pharmacological 18 Que W, Chen J, Chuang M, Jiang D. Knockdown of c-Met enhances sensitivity to interplay between breast cancer resistance protein and gefitinib in epidermal bortezomib in human multiple myeloma U266 cells via inhibiting Akt/mTOR growth factor receptor signaling. Anticancer Res 2009; 29:1059–1065. activity. APMIS 2012; 120:195–203. 43 Ma PC, Jagadeeswaran R, Jagadeesh S, Tretiakova MS, Nallasura V, Fox EA et al. 19 Que W, Chen J. Knockdown of c-Met inhibits cell proliferation and invasion and Functional expression and mutations of c-Met and its therapeutic inhibition with increases chemosensitivity to doxorubicin in human multiple myeloma U266 cells SU11274 and small interfering RNA in non-small cell lung cancer. Cancer Res 2005; in vitro. Mol Med Report 2011; 4:343–349. 65: 1479–1488.

© 2015 Nature America, Inc. Cancer Gene Therapy (2015), 262 – 270 c-MET shRNA-induced apoptosis in MDR cancer cells T-H Hung et al 270 44 Wang ZX, Lu BB, Yang JS, Wang KM, De W. Adenovirus-mediated siRNA targeting 47 Xie B, Xing R, Chen P, Gou Y, Li S, Xiao J et al. Down-regulation of c-Met expression c-Met inhibits proliferation and invasion of small-cell lung cancer (SCLC) cells. inhibits human HCC cells growth and invasion by RNA interference. J Surg Res JSurgRes2011; 171: 127–135. 2010; 162:231–238. 45 Zhang SZ, Pan FY, Xu JF, Yuan J, Guo SY, Dai G et al. Knockdown of 48 Li Y, Zhang S, Tang Z, Chen J, Kong W. Silencing of c-Met by RNA interference c-Met by adenovirus-delivered small interfering RNA inhibits hepato- inhibits the survival, proliferation, and invasion of nasopharyngeal cellular carcinoma growth in vitro and in vivo. Mol Cancer Ther 2005; 4: carcinoma cells. Tumour Biol 2011; 32: 1217–1724. 1577–1584. 49 Leelawat K, Leelawat S, Tepaksorn P, Rattanasinganchan P, Leungchaweng A, 46 Xie B, Tang DG, Dong JH. Effects of c-met-siRNA on the growth and invasion of Tohtong R et al. Involvement of c-Met/hepatocyte growth factor pathway in hepatocellular carcinoma MHCC97-H cells. Zhonghua Gan Zang Bing Za Zhi 2006; cholangiocarcinoma cell invasion and its therapeutic inhibition with small inter- 14: 499–504. fering RNA speciﬁc for c-Met. J Surg Res 2006; 136:78–84.