Cancer Chemother Pharmacol DOI 10.1007/s00280-015-2739-2

ORIGINAL ARTICLE

Cisplatin resistance in human cervical, ovarian and lung cancer cells Jianli Chen1,2 · Charalambos Solomides3 · Hemant Parekh4 · Fiona Simpkins5 · Henry Simpkins1,2 

Received: 31 October 2014 / Accepted: 24 March 2015 © Springer-Verlag Berlin Heidelberg 2015

Abstract  Purpose  This study was performed to determine whether or not in cervical, ovarian and lung cancer cell lines, free radicals (ROS) play a role in cisplatin cytotoxicity and activation of the mitochondrial and JNK/p38 pathways. The role of the enzyme, dihydrodiol dehydrogenase (DDH1), in the activation/deactivation of this pathway and how this may be related to the development of resistance was also investigated.

* Henry Simpkins [email protected] Jianli Chen [email protected] Charalambos Solomides [email protected] Hemant Parekh [email protected] Fiona Simpkins [email protected] 1

The Feinstein Institute for Medical Research, NS-LIJ Health System, 350 Community Drive, Manhasset, NY 11030, USA

2

Department of Pathology and Laboratory Medicine, Staten Island University Hospital, 475 Seaview Avenue, Staten Island, NY 10305, USA

3

Department of Pathology, Anatomy and Cell Biology, Jefferson Medical College, Thomas Jefferson University, 132 S. 10th Street, 260E Main, Philadelphia, PA 19107, USA

4

Department of Pathology and Laboratory Medicine, Temple University School of Medicine, 3401 North Broad Street, Philadelphia, PA 19140, USA

5

Jordan Center for Gynecologic Oncology, Penn Perelman Center for Advanced Medicine, 3400 Civic Center Blvd., 3rd floor West, Philadelphia, PA 19104, USA









Methods  Mitochondrial membrane potential and ROS analysis were performed by flow cytometry, P-JNK and P-p38 by western blotting and mRNA by RT-PCR. Dihydrodiol dehydrogenase (DDH1) and thioredoxin knockdowns were prepared by standard techniques. Results  Cisplatin treatment of a cervical cancer cell line resulted in ROS production with mitochondrial membrane depolarization and phosphorylation of JNK and p38. N-acetyl-cysteine, a free radical scavenger, ameliorated these effects. Treatment of the sensitive cells with H2O2 produced similar effects but at shorter incubation times. Similar results were observed with an ovarian cell line. Downregulation of dihydrodiol dehydrogenase in the cisplatin-resistant cervical and lung cancer cell lines resulted in increased drug sensitivity with detectable production of ROS and activation of the JNK/p38 pathways; however, downregulation of thioredoxin in the cervical cells had minimal effect. Conclusion  Dihydrodiol dehydrogenase appears to play a role in cisplatin resistance in cervical, ovarian and lung cancer cells which includes mitochondrial membrane depolarization, ROS production and activation of the JNK pathway. However, its mode of action cannot be mimicked by an ROS scavenger so its mechanism of action is more complex (a not unexpected finding considering its role in xenobiotic activation/countering oxidative stress). Keywords  Cisplatin resistance · Reactive oxygen species · Dihydrodiol dehydrogenase · JNK/p38 pathways · Human cervical, ovarian and lung carcinoma cells Abbreviations DDH Dihydrodiol dehydrogenase THR Thioredoxin 1 ROS Reactive oxygen species NAC N-acetyl-cysteine

13



H2O2 Hydrogen peroxide DCF 5-(and 6-)-chloromethyl-2′, 7′-dichlorodihydrofluorescein diacetate, acetyl ester DHE Dihydroethidium JC1 5′, 6, 6′-Tetrachloro-1, 1′, 3, 3′-tetraethylbenzimidazolycarbocyanine iodide

Introduction Cisplatin and carboplatin are among the most widely used drugs for the treatment of solid organ cancers and are curative for most patients with germ cell tumors [1]. They are currently used in standard chemotherapy protocols for the treatment of patients with ovarian, bladder, cervical, head and neck and lung cancers [2]. It should also be stressed that although cervical cancer has decreased in the USA, worldwide it is the third largest cause of cancer in women with 275,000 deaths [3]. The main action site of cisplatin is thought to be DNA, primarily the N7-position of guanosine leading to inter- and intra-strand cross-links [4] as opposed to oxaliplatin, which produces DNA breaks [5]. However, intrinsic and/or acquired resistance to platinum-based chemotherapy has increasingly become a limiting factor in the treatment of many patients. Many studies have attempted to decipher the mechanisms of cisplatin resistance employing different human carcinoma cell lines. The human cervical cancer cell line ME180 and its resistant counterpart ME180R, which exhibits stable resistance to cisplatin (9–15-fold) and was previously designated as an ovarian cell line (C13), exhibited differences when compared to the parental sensitive cell line (2008). The resistant cell line showed decreased intracellular accumulation of cisplatin [6], increased replicative bypass of cisplatin-DNA adducts [7], reduced expression of membrane-associated beta tubulin [8], decreased expression of cytokeratin 18 [9], mitochondria with aberrant morphology and hypersensitivity to lipophilic cations [10] as well as activation of the protein kinase C and cyclic AMP signal transduction pathways [11], and cDNA microarrays demonstrated that the cytoplasmic enzyme, dihydrodiol dehydrogenase (DDH), was upregulated in the resistant cells [12]. Subsequent work showed that transfection of DDH (primarily the DDH1 isoform) into a whole series of human cancer cell lines produced cisplatin resistance [13] and its downregulation sensitized the resistant cancer cells to cisplatin [14] and resistance to irradiation, and adriamycin as well as cisplatin was reported to be mediated by this enzyme in seven lung cell lines [15]. Clinical studies with human ovarian [16] and lung cancer specimens [17, 18] employing immunohistochemistry also showed that DDH expression may be associated with chemotherapy resistance. Recently, secreted DDH was detected by proteomics

13

Cancer Chemother Pharmacol

in NSCLC patients [19]. DDH belongs to a superfamily of monomeric cytosolic NADPH-dependent oxidoreductases which are involved in the activation/inactivation of several xenobiotics [20]. Most interestingly, under conditions of oxidative stress and drugs that deplete GSH (e.g., ethacrynic acid), DDH1 can be induced tenfold, and it has been hypothesized that this induction may be a component in an effective counter-response to oxidative stress [21]. It has also been implicated in resistance to 2-deoxytubercidin in renal carcinoma cells [22], and its downregulation produced an increase in intracellular ROS levels [14]. A study, therefore, was initiated with the cisplatin-resistant cell line to confirm its tumor cell origin and to investigate whether cisplatin treatment produces free radicals and whether this is related to cisplatin cytotoxicity, mitochondrial membrane depolarization (i.e., mitochondrial pathway of apoptosis) and activation of the JNK/p38 signaling pathway (since this has been implicated in cisplatin cytotoxicity) [23, 24]. The roles of dihydrodiol dehydrogenase 1(DDH1) and thioredoxin (TRX—a free radical scavenger) were then investigated employing DDH1 and thioredoxin knockdowns of the cervical cisplatin-resistant cell line (ME180R) and a DDH1 knockdown of a highly resistant lung cancer cell line (A549).

Materials and methods Cell culture reagents and drugs Cell culture reagents and gentamicin were obtained from Cellgro (Herndon, VA) and RNAzol B from Tel-Test Inc. (Friendswood, TX). The cisplatin-sensitive and cisplatinresistant ME180, ME180R cell lines and SKOV3 were grown in RPMI-1640 medium supplemented with 10 % fetal bovine serum and gentamicin at a final concentration of 10 µg/ml. The A549 cells were grown in Ham’s F-12 fetal calf serum and gentamicin (10 µg/ml). The ME180 cells were obtained from Dr. P. Andrews (Georgetown University, DC) in 1991 and were designated as ovarian in origin (2008 and C13 cell lines). Recently, their origin has been disputed [25]. The A549 and SKOV3 were obtained from the ATCC (Manassas, VA). Cisplatin and N-acetylcysteine (NAC) were purchased from Sigma-Aldrich Chemical Co. (St. Louis, MO). Immunocytochemistry The cells in logarithmic growth phase were centrifuged on slides in a CytoSpin, fixed in 90 % cold ethanol for 10 min at room temperature. Mouse monoclonal antibodies to pancytokeratin, P16, inhibin, CA125, CK5/6 and TTF were obtained from Ventana Medical Systems (Phoenix, AZ), CK7

Cancer Chemother Pharmacol

and CK20 from Cell Marque (Rocklin, CA), WT1 from Dako (Carpinteria, CA) and Ber-Ep4 from BioCore (Irvine, CA). They were employed as instructed by the respective manufacturer. Thrombomodulin and mesothelin staining were performed at Integrated Oncology (New York, NY). HPV testing was performed employing the Cervista test (Hologic, Madison, WI). The slides were scored on a 0/4+ scale. Cytotoxicity assays Cell growth assays were performed as described previously [9]. Briefly, 4000 cells were seeded into 96-well plates in triplicate with different concentrations of drugs and incubated for 72 h at 37 °C. Ten microliters of MTT(5 mg/ml) was then added to each well at the end of the incubation period and incubated for a further 5 h. The plate was then scanned at 570 nm in a 96-well plate reader. Each experiment was performed at least three times in triplicate. Colony‑forming assays (CFA) Cells (1 × 105) were seeded in 6-well plates incubated for 24 h and then treated for 4 h with different drug concentrations. The cells were then washed twice with drugfree medium and trypsinized with 0.25 % trypsin-0.2 % EDTA to obtain a single-cell suspension—200 cells were seeded into 60 mm dishes in duplicate and incubated for two weeks in a drug-free complete medium to allow for colony growth. At the end of incubation period, the culture medium was aspirated and the cells fixed and stained with 0.5 % methylene blue in 50 % ethanol for 40 min at room temperature. Thereafter, the plates were gently washed with water and allowed to air-dry. Visible colonies (containing 50 or more cells each) were counted to determine the percent colony formation for each drug treatment. IC50 values were expressed as the mean ± SD (standard deviation) from triplicate experiments. Flow cytometry Mitochondrial membrane potential analysis was performed employing JC1 (5, 5′, 6, 6′-tetrachloro-1, 1′, 3, 3′-tetraethylbenzimidazolycarbocyanine iodide), which was purchased from AnaSpec (Fremont, CA). Cells (1 × 105) were cultured in 500 µl of complete RPMI-1640 medium for 24 h at 37 °C in a 24-well plate. Cisplatin was added, and the cells incubated further for 24 h. JC1 (10 µM) was added 30 min prior to the end of the incubation. The medium was aspirated, the cells trypsinized (0.05 % trypsin–EDTA) and centrifuged, and the pellets washed and resuspended in HBSS buffer and analyzed with a BD FACSVerse or FACSFortessa (excitation wavelength 525 nm: emission wavelength 575 nm).

Intracellular ROS levels were detected using the fluorescent probes dihydroethidium (DHE) (AnaSpec, San Jose, CA), 5-(and 6-)-chloromethyl-2′, 7′-dichlorodihydrofluorescein diacetate and acetyl ester (DCF) (AnaSpec, San Jose, CA). Cells (1 × 105) were cultured for 24 h in complete RMPI-1640 medium and then treated with different concentrations of cisplatin for an additional 24 h. DHE (2 μM) was added for 15 min or DCF (1 μM) for 30 min at 37 °C. The cells were washed and suspended in HBSS buffer, and flow cytometry was performed with a BD FACSVerse or FACSFortessa (excitation wavelength 488 nm: emission wavelengths 525 nm for DCF; 575 nm for DHE). Western blotting analysis Cells (1 × 106/ml) were incubated under normal growth conditions at 37 °C and then washed with PBS at 4 °C (3×), and a whole cell lysate was prepared from each cell line by scraping the cells into a buffer containing 20 mM Tris–HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1 % (v/v) Triton X-100, 0.5 % (v/v) Nonidet P40, 2.5 mM sodium pyrophosphate, 1 mM sodium orthovanadate, 50 mM sodium fluoride and 1× protease inhibitor cocktail and incubated on ice for 30 min. The lysate was then centrifuged at 13,000 g for 10 min, and the supernatant stored at −80 °C until use. Proteins were separated by SDS-PAGE and were transferred to a PVDF membrane. The different antibodies were applied (at concentrations described by the manufacturer) to the PVDF membrane, and the bands identified by enhanced chemiluminescence reagents (Pierce Biochemicals, Rockford, IL). The antibodies utilized were rabbit polyclonal JNK, phosphoJNK, p38, phospho-p38 (Santa Cruz Biotechnology, Santa Cruz, CA), DDH1 and DDH2 mouse polyclonal antibodies (Abnova Corp.,Walnut, CA), mouse monoclonal actin (Calbiochem, San Diego, CA), mouse monoclonal GAPDH (Thermo Fisher Scientific, Rockford, IL) and a DDH3 mouse monoclonal antibody (Abcam, Cambridge, MA). The secondary antibodies were anti-goat, anti-rabbit and anti-mouse (Pierce Biochemical, Rockford, IL) and rabbit anti-mouse (Thermo Fisher Scientific, Rockford, IL). All the western blots were repeated at least three times. Densitometry analysis was performed with UN-SCAN-IT gel 6.1 software and expressed as mean ± SD (standard deviation). Semi‑quantitative PCR analysis Total RNA was isolated from the cells (2 × 106) after a brief wash (2×) with cold PBS (pH 7.4). RNAzol (Teltest, Friendswood, TX) was added followed by chloroform extraction, isopropanol precipitation and a 75 % (v/v) ethanol-DEPC wash. The reverse transcription reaction consisted of 1 µg of RNA, 4 units of Omniscript RT, 1 µM

13



Cancer Chemother Pharmacol

Table 1  Immunohistochemical characterization of the cell lines Cell line ME180 ME180 (R) SKOV3

Pancyto 4+ 4+ 2+

CK7 – – 2+

CK20 – – –

CK 5/6 4+ 3+ –

p16 4+ 4+ –

WT1 −/1+ −/1+ –

Inhibin – – –

CA125 – – –

Ber-Ep4 ND ND ND

Thrombo ND ND ND

Meso ND ND ND

TTF ND ND ND

A549

3+

4+

–

–

–

ND

ND

ND

–

–

–

1+

Reagents and methods of detection are described in the “Materials and methods” section Thrombo thrombomodulin, Meso mesothelin

oligo-dT primer, 0.5 mM dNTP, 10 units of RNase inhibitor and 1 × RT buffer. Reverse transcription was performed at 37 °C for 1 h and inactivated at 93 °C for 3 min. The cDNA was then amplified by PCR using gene-specific primer pairs. Each PCR consisted of 1 × PCR buffer, 1.5 mM MgCl2, 200 µM dNTP, 2.5 units of Taq polymerase and 0.2 mM gene-specific forward (F) and reverse (R) primers. The PCR conditions were as follows: an initial denaturation at 94 °C for 15 s, followed by 55 °C for 30 s and 72 °C for 30 s for a specific number of cycles (optimized for each primer) to ensure that the product intensity fell within the linear phase of amplification. A final elongation step was performed for 10 min at 72 °C. RT-PCR amplification of GAPDH was used as internal control to verify that equal amounts of RNA were used from each cell line. The PCR products were separated on a 1.5 % agarose gel stained with ethidium bromide (0.35 µg/ml) by electrophoresis in 1 × Tris borate EDTA buffer. A HaeIII digest of Ф-X174 DNA was used as a standard marker. The primer sequences are: DDH1: forward 5′-CTAACCAGGCCAGTGACAGA-3′, reverse 5′-CTCATGCAATGCCCTCCATG-3′. DDH-2: forward 5′-GCTAACCAGGCCAGTGACAGAAATG-3′, reverse 5′-CTTCTGGCAGACCTCATGCAATG-3′. DDH3: forward 5′-CCCATTGTTTTTGTAATCTCTG-3′, reverse 5′-TTATTTCAAAATGATAAAAATTTATTG-3′. GAPDH: forward 5′-GAAGGTGAAGGTCGGAGTC-3′, reverse 5′-GAAGATGGTGATGGGATTC-3′. RNA Knockdown siRNA corresponding to the DDH1 gene was designed as described in the pSilencer neo instruction manual (Ambion, Austin, TX). The hairpin siRNA template oligonucleotides were designed by entering siRNA target sequence into the web-based converter(www.ambion.com/techli/ misc/psilencer_converter.html), synthesized by Ambion and inserted into a pSilencer 3.1 neo vector (Ambion, Austin, TX) following pSilencer™ neo manual instruction(cat # 5764,5770) and referred to as pSilencer-DDH1 (gene sequences available on request). All siRNA-associated plasmids were analyzed by restriction of endonuclease digestion and DNA sequencing [14].

13

Thioredoxin shRNA (ID 7295) was obtained from Origene (Rockville, MD). The resistant ME180R or A549 cells were aliquoted into six-well plates at a density of 2 × 105 cells per well. After a 24-h incubation period, pSilencer-DDH1 or thioredoxin 1, 1 µg gene-specific shRNA or 1 µg shRNA cloning vector (as control) was then transfected into the different cell lines with Lipofectamine 2000 in serum-free DMEM medium as described by the supplier (Invitrogen, Carlsbad, CA). The next day, the transfected cells were trypsinized with 0.05 % trypsin–EDTA, and aliquots of 200 cells were seeded into 100-mm culture dishes containing 10 ml of complete RPM-1640 medium with puromycin (final concentration 1 µg/ml) or geneticin (G418 sulfate, final concentration 700 μg/ml). After 3–4 weeks, clones were selected and grown in RPMI-1640 with puromycin or geneticin and the transfectants were then frozen in liquid nitrogen. Statistical analysis The linear regression analysis and paired t test were performed using the SigmaStat Statistical Analysis System, version 1.01. P  0.05 when statistically compared with the group treated with 50 μM of cisplatin was marked with #; groups indicated with *, **, ***, **** were co-treated with NAC and have a statistical P