Available online at www.sciencedirect.com

ScienceDirect Journal of Nutritional Biochemistry xx (2014) xxx – xxx

Curcumin induces apoptosis and inhibits growth of orthotopic human non-small cell lung cancer xenografts☆,☆☆ Shahar Lev-Ari a,⁎, Alex Starr b , Sara Katzburg a , Liron Berkovich a , Adam Rimmon a , Rami Ben-Yosef a , Akiva Vexler a , Ilan Ron a , Gideon Earon a b

a Laboratory of Herbal Medicine and Cancer Research, Institute of Oncology, Tel-Aviv University, Tel-Aviv, Israel Department of Pulmonology, Tel-Aviv Sourasky Medical Center, Tel-Aviv, Israel, affiliated to the Faculty of Medicine, Tel-Aviv University, Tel-Aviv, Israel

Received 11 September 2013; received in revised form 19 March 2014; accepted 19 March 2014

Abstract Non-small cell lung cancer (NSCLC) is the leading cause of cancer-related mortality. Curcumin is involved in various biological pathways leading to inhibition of NSCLC growth. The purpose of this study was to evaluate the effect of curcumin on expression of nuclear factor κB-related proteins in vitro and in vivo and on growth and metastasis in an intralung tumor mouse model. H1975 NSCLC cells were treated with curcumin (0–50 μM) alone, or combined with gemcitabine or cisplatin. The effects of curcumin were evaluated in cell cultures and in vivo, using ectopic and orthotopic lung tumor mouse models. Twenty mice were randomly selected into two equal groups, one that received AIN-076 control diet and one that received the same food but with the addition of 0.6% curcumin 14 days prior to cell implantation and until the end of the experiment. To generate orthotopic tumor, lung cancer cells in Matrigel were injected percutaneously into the left lung of CD-1 nude mice. Western blot analysis showed that the expressions of IkB, nuclear p65, cyclooxygenase 2 (COX-2) and p-ERK1/2 were down-regulated by curcumin in vitro. Curcumin potentiated the gemcitabine- or cisplatin-mediated antitumor effects. Curcumin reduced COX-2 expression in subcutaneous tumors in vivo and caused a 36% decrease in weight of intralung tumors (P=.048) accompanied by a significant survival rate increase (hazard ratio=2.728, P=.036). Curcumin inhibition of COX-2, p65 expression and ERK1/2 activity in NSCLC cells was associated with decreased survival and increased induction of apoptosis. Curcumin significantly reduced tumor growth of orthotopic human NSCLC xenografts and increased survival of treated athymic mice. To evaluate the role of curcumin in chemoprevention and treatment of NSCLC, further clinical trials are required. © 2014 Elsevier Inc. All rights reserved. Keywords: Curcumin; NSCLC; Cisplatin; Apoptosis; Synergistic effect

1. Introduction Lung cancer is the second most common cancer in both genders and the leading cause of cancer death, accounting for 28% of all cancer deaths expected to occur in 2013 [1]. About 85% to 90% of lung cancers are non-small cell lung cancer (NSCLC), with a 5-year survival rate of only 16% [1]. Standard treatment regimens for NSCLC depend on the stage of the disease. Early-stage tumors are treated primarily with surgery or radiotherapy [2]. Radiation may also be used postoperatively when surgical margins are close or positive. More advanced cancers often require multimodality therapy consisting of surgery, radiation and chemotherapy. Chemotherapy for NSCLC uses a ☆ Financial support: This study was supported by The Edmond Benjamin, De Rothschild Foundation and Chaya and Kadish Shermeister Endowment. ☆☆ Potential conflicts of interest: None. ⁎ Corresponding author at: Laboratory of Herbal Medicine And Cancer Research, Institute of Oncology, Tel-Aviv Sourasky Medical Center, 6 Weizmann Street, Tel-Aviv 64239, Israel. Tel.: +972 3 697 4833; fax: +972 3 697 4832. E-mail address: [email protected] (S. Lev-Ari).

http://dx.doi.org/10.1016/j.jnutbio.2014.03.014 0955-2863/© 2014 Elsevier Inc. All rights reserved.

combination of two drugs, most frequently combination of carboplatin–taxol or gemcitabine–cisplatin [2]. Both antitumor agents gemcitabine and cisplatin interfere with DNA replication and repair and induce cell apoptosis [3,4]. Cisplatin induces its cytotoxic properties through binding to nuclear DNA and subsequent interference transcription and/or DNA replication mechanisms [4]. Kim et al. [5] found that activation of NF-κB is required for cisplatin-induced apoptosis in HNSCC cell lines. Cisplatin was shown to induced IkBα degradation and NF-κB-dependent transcriptional activation prior to cell death [5]. The transcription factor nuclear factor κB (NF-κB) is a crucial regulator in oncogenesis. It promotes proliferation, inhibits apoptosis and, by doing so, maintains the balance between normal cell division and cell death. NF-κB, a multifunctional transcription factor, is activated by numerous extracellular stimuli, including cytokines, growth factors, carcinogens and tumor promoters that lead to the expression of defined target genes for diverse biological functions [6]. NF-κB is a heterodimeric protein composed of five subunits [RelA (p65), RelB, c-Rel, NF-κB1 (p50 and p105) and NF-κB2 (p52) proteins], and it is retained in the cytoplasm by the inhibitory subunit, IkBa [6]. The extracellular stimulus that initiates activation of NF-κB is dependent on the degradation of IkBa proteins, which is

2

S. Lev-Ari et al. / Journal of Nutritional Biochemistry xx (2014) xxx–xxx

mediated through the activation of the IkBa kinase (IKK) complex that subsequently releases the active NF-κB to translocate into the nucleus where it regulates numerous gene expressions, including cyclooxygenase 2 (COX-2) [7]. Since NF-κB is involved in cancer development, modulating NF-κB activation pathways has important implications in cancer prevention and therapy. NF-κB, which may play an important role in COX-2 induction, is mediated through activation of the canonical NF-κB pathway [8]. Charalambous et al. [9] have shown that upregulation of COX-2 was accompanied by increased expression of NF-κB-p65 and IκB-kinase-alpha (IKKα) in human colorectal cancer epithelial cells. Those authors suggested that these were early postinitiation events involved in tumor progression [8]. COX-2 is induced by proinflammatory or mitogenic stimuli and is overexpressed in a variety of human cancers, including NSCLC [10–12]. Elevated tumor COX-2 expression is associated with increased angiogenesis, tumor invasion and promotion of tumor cell resistance to apoptosis [11,12]. COX-2 was found to be up-regulated at most stages of tumor progression in lung cancer [13]. Takahashi et al. [14] demonstrated an association of up-regulation of COX-2 with tumor invasion and metastasis in human lung tumors. Moreover, the proportion of adenocarcinoma cells with marked COX-2 expression is reportedly much greater in lymph node metastases than in the corresponding NSCLC primary tumors [10]. Other studies have shown that elevated expression of COX-2 is associated with poor prognosis and a worse overall survival rate in NSCLC [11]. Consistent with these findings, several epidemiologic studies have shown a positive association between consumption of nonsteroidal anti-inflammatory drugs and the incidence of lung cancer [11,15]. Selective COX-2 blocking (COXIB) agents and celecoxib in particular have a strong potential for the chemoprevention of human lung cancer [16]. Long-term use of COXIBs has, however, been associated with increased risk of serious cardiovascular events [17], and clinical trials have shown that the addition of celecoxib to chemotherapy failed to benefit the survival of NSCLC patients [13]. Taken together, these findings establish the need to find a better therapeutic strategy for NSCLC. Curcumin (diferuloyl methane) is a natural yellow-pigmented polyphenol component of the spice turmeric, derived from the roots of the Curcuma longa plant indigenous to Southeast Asia. Curcumin has been used as an anti-inflammatory agent in traditional Indian Ayurvedic medicine for centuries [18]. Numerous in vitro and in vivo studies have shown that curcumin possesses anticancer activities via its effect on a variety of biological pathways. Additionally, curcumin affects growth factor receptors and cell adhesion molecules involved in tumor growth, angiogenesis and metastasis, all of which are relevant to cancer. Curcumin also has antioxidant and anti-inflammatory properties and is able to modulate various signaling mechanisms among which is the ability to inhibit COX-2 at the transcriptional level [19], leading to the suppression of prostaglandin synthesis [20–23]. Moreover, curcumin is an inhibitor of the transcription factor, NF-κB, and downstream gene products, such as COX-2. Curcumin blocks the IκK-mediated phosphorylation and degradation of IκBα, thus NF-κB remains bound to IκBα in the cytoplasm and is not able to enter the nucleus to activate transcription, thereby constraining the expression of the COX2 gene [24]. Curcumin can potentiate the antitumor effects of gemcitabine on human cancer cells [25–27]. It has been shown to potentiate the antitumor effects of gemcitabine in an orthotopic model of human bladder cancer through suppression of proliferative and angiogenic biomarkers [26]. It enhances the effect of cisplatin via inhibition of IKKβ protein of the NF-κB pathway [28]. Combined therapy of curcumin and cisplatin by means of differing mechanisms exhibited a synergistic effect leading to cell death and growth suppressive effect [29]. While curcumin exerted its effect through the inhibition of cytoplasmic and nuclear IKK, leading to inhibition of NF-κB activity, cisplatin mediated its effect via increased expression of p16 and p53, leading to cellular

senescence [29]. Curcumin's inhibitory effect on carcinogenesis has been demonstrated in several animal models of various tumor types [18,25,30–32]. In vitro studies of NSCLC have shown that curcumin activated cell cycle arrest and apoptosis through different pathways [18]. In the present study, we investigated the effects of curcumin on survival and apoptosis of human NSCLC carcinoma cell lines and sought to determine whether this effect is associated with downregulation of COX-2, p-ERK1/2 and EGFR expression. We further investigated the pattern of growth and metastatic processed and the response to curcumin of subcutaneous and orthotopic NSCLC tumors. To the best of our knowledge, this is the first study to assess the effect of curcumin in an orthotopic intralung NSCLC model. This study may pave the way to the implementation of curcumin treatment in combination with chemotherapy for patients with NSCLC. 2. Materials and methods 2.1. Cell culture and reagents The human NSCLC cell lines H1975, H358 and H1299 were obtained from the American Type Culture Collection (ATCC). The human lung carcinoma PC-14 cell line was kindly provided by Prof. I. Fidler (M. D. Anderson Cancer Center, Houston, TX, USA). All cell lines were grown and maintained in Dulbecco's modified Eagle's medium (DMEM; Biological Industries, Beit HaEmek, Israel) supplemented with 10% fetal calf serum, L-glutamine, sodium pyruvate, 1% penicillin and 1% streptomycin (full medium) at 37 °C, in an atmosphere of 95% oxygen and 5% CO2. Curcumin (97% purity) was purchased from Merck (Whitehouse Station, NJ, USA) and gemcitabine from Eli Lilly (Indianapolis, IN, USA). Specific antibodies against p65, IκBα and COX-2, EGFR and p-EGFR were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA), against phosphoIκBα from Cell Signaling Technology (USA) and against actin from MP Biomedicals (USA). 2.2. Cell viability assay Cell viability was evaluated by XTT assay as described previously [31]. The cells (1– 2×103 cell/well) were seeded in 96-well microwell plates, incubated at 37 °C for 24 h and then treated with the tested drugs. After 72 h, cell viability was assessed by the ability of metabolically active cells to reduce the tetrazolium salt to colored formazan compounds. The optical density was read at 450 nm. Each variant of the experiment was performed in triplicates. The data were expressed as the mean values of at least three different experiments. Cell survival following treatment was expressed as a percentage of viable cells relative to control value. 2.3. Synergism between curcumin and chemotherapy To determine whether the addition of curcumin to gemcitabine or cisplatin is synergistic, additive, or antagonistic, cytotoxicity data were analyzed using the combination index (CI) values calculated by the CalcuSyn software (Biosoft, Ferguson, MO, USA), which is based on multiple-drug effect equations [33]. 2.4. Flow cytometry analysis Cell cycle and apoptosis were assessed by flow cytometry. The cells were plated at a density of 0.5×106 per 10-cm dish. The tested drugs were added 24 h later, at selected concentrations. Cells (1–2×106) were washed in phosphate-buffered saline (PBS), and the pellet was fixed in 3 ml ethanol for 1 h at 4 °C. The cells were pelleted, resuspended in 1 ml PBS and incubated for 30 min with 0.15 mg/ml RNAse at 37 °C, then stained with 5 μg/ml propidium iodide for 1 h before flow cytometry analysis. Data acquisition was performed on a FACScan and analyzed by CellQuest software (Becton Dickinson Immunocytometry Systems, San Jose, CA, USA). Data for at least 10,000 cells were collected for each data file. Necrotic cells that had been detected by counting cells following staining with trypan blue before fixation were excluded from the calculation of apoptotic cells. 2.5. Protein extraction and Western blot COX-2, p-ERK1/2, EGFR and p-EGFR expression was evaluated by Western blot analysis. Exponentially growing cells were collected, washed three times in ice-cold PBS and resuspended in lysis buffer [20 mM Tris-HCI pH 7.4, 2 mM EDTA, 6 mM βmercaptoethanol, 1% NP-40, 0.1% sodium dodecyl sulfate (SDS) and 10 mMNaF, plus the protease inhibitors, leupeptin 10 μg/ml, aprotinin 10 μg/ml and 0.1 mM phenylmethylsulfonyl fluoride]. The nuclear extracts of H1975 cells that had been cultured for 72 h in curcumin-supplemented media were prepared according to the manufacturer's protocol with NucBuster Protein Extraction Kit (Novagen, EMD Biosciences). The protein concentration of each sample was estimated using the Bio-Rad protein assay (Bio-Rad Laboratories, Hercules, CA, USA). Actin expression was used to verify that equal amounts

S. Lev-Ari et al. / Journal of Nutritional Biochemistry xx (2014) xxx–xxx

of protein were loaded. Samples containing 50 μg of total cell lysate was loaded onto a 10% SDS-polyacrylamide gel and subjected to electrophoresis. Proteins were transferred to “Hybond-C” membranes (Amersham, Arlington Heights, IL, USA) in transfer buffer (25 mM corresponding Tris, 190 mM glycine, 20% methanol) using a Trans Blot transfer apparatus (Bio-Rad Laboratories) at 70 mA for 12–18 h at room temperature. The membranes were blocked with blocking buffer (PBS/0.2% Tween-20/0.5% gelatin) for 1 h at room temperature and subsequently washed three times for 5 min in washing buffer (PBS/0.05% Tween-20). The membranes were incubated with corresponding monoclonal human antibody for 1 h at room temperature, then washed as described above and incubated with antigoat secondary antibodies (1:2000) for 1 h at room temperature. Immune detection was performed using the ECL Western blot detection system (Amersham).

2.6. NoShift transcription factor Assay for NF-κB activity The enzyme-linked immunosorbent assay was carried out using NoShift transcription factor assay kit integrated with NoShift NF-κB reagents (Novagen, EMD Biosciences) according to the manufacturer's protocol. A total of 62 μg of nuclear extracts was incubated on ice for 30 min with biotinylated NF-κB consensus WT DNA (10 pmol), together with 500 ng salmon sperm DNA and 0.01U Poly(dI-dC) in 4× NoShift binding buffer. HeLa nuclear extract was used as a positive control, while as an additional control, a competition assay was simultaneously performed with a 50-fold molar excess (500 pmol) of nonbiotinylated NF-κB or mutant nonbiotinylated NF-κB, which was added before the addition of extracts. Samples, along with 1× NoShift binding buffer (1:4) were transferred into a freshly washed streptavidin-coated microassay plate, sealed and incubated for 60 min at 37 ºC. The contents of the wells were then removed, and the wells were washed and added mouse mAb anti-NF-κB (p65). The plate was sealed and incubated for 60 min at 37 ºC. The contents of the wells were again removed, the wells were washed and added goat antimouse IgG HRP-conjugated and incubated for 30 min at 37 ºC. The contents of the wells were removed once more, the wells were washed again and tetramethylbenzidine (TMB) substrate was added. The plate was sealed with aluminum foil and incubated for 30 min at room temperature. The reaction was stopped by adding 1 N HCl, and the absorbance (450 nm) was measured using a microplate reader.

2.7. In vivo studies 2.7.1. Animals Athymic 4- to 6-week-old CD-1 nude mice were obtained from the Harlan Animal Production Area (Jerusalem, Israel). The animals were housed in a laminar flow cabinet under pathogen-free conditions in standard vinyl cages with air filter tops. Cages, bedding and water containers were autoclaved before use. The Ethics Committee for Accreditation of Laboratory Animal Care approved all facilities in accordance with the current regulations and standards of the Israeli Ministry of Health. The same protocol of treatment was used in both ectopic and orthoptopic tumor models. In each experiment, 20 mice were randomly selected into two equal groups, one that received AIN-076 control diet and one that received the same food but with the addition of 0.6% curcumin. The treatment begun 14 days prior to cell implantation and lasted until the end of experiment.

2.7.2. Ectopic tumor model Subconfluent H1975 cells were harvested by a brief treatment with trypsin/EDTA and resuspended in DMEM with 10% fetal bovine serum. Cell viability was determined by trypan blue exclusion. Only single-cell suspensions with N90% viability were used for injections. Before implantation, the cells were washed with cold PBS by centrifugation, resuspended in PBS and kept on ice before used. Tumor cells (1×106 cells in 0.2 ml PBS) were injected subcutaneously into the flank area. Tumor size was measured twice weekly by a digital caliper. The mean tumor size at each time point was calculated from the pooled data. At the end of the experiment, the animals were sacrificed with a lethal dose of chloroform, necropsied, and their tumor tissues were harvested, fixed in formalin and proceeded for histological and immunohistochemical (IHC) evaluation.

2.7.3. Orthotopic (intralung) tumor model Cells (as described above for subcutaneous model) were injected intrathoracally with growth factor reduced Matrigel, which anchors the tumor cells thereby preventing their diffusion into the lung. A stock solution of 500 μg of Matrigel in 1 mL of PBS was used for all experiments. Suspensions of equal volumes of cells in icecold Matrigel were prepared. Cells, syringes and needles were kept on ice before the injections. The mice were anesthetized before injection with sodium pentobarbital (50 mg/kg body weight) and placed in the right lateral decubitus position. Cell inoculum was injected percutaneously into the left lateral thorax at the lateral dorsal axillary line using 1-ml tuberculin syringes with 30-gauge hypodermic needles. The experiments were terminated when the mice in the nontreated group become moribund. The animals were necropsied, and their tumor tissues were harvested and weighed.

3

2.8. Immunohistochemistry Subcutaneous tumors were embedded in paraffin for IHC staining with specific anti-COX-2 antibodies. Antigen retrieval was performed at 95 °C in citrate buffer pH 6.0, 6.4 M sodium citrate dehydrate, 1.6 M citric acid monohydrate for 40 min. The slides were cooled at room temperature for 20 min and washed 3×3 min with Tris buffer (pH 7.6), 0.15 M sodium chloride, 0.05 M Trizma HU. The slides were peroxidase-blocked for 5 min and washed as described above. The slides were then incubated for 30 min with the primary antigen, followed by the secondary antigen (visualization reagent), followed by the substrate-chromogen solution (3,3′-diaminobenzidine), and finally counterstained with hematoxylin. For a negative control, the primary antibody was replaced with a nonspecific antibody. Staining was evaluated by licensed pathologists following Dako guide instructions [34]. 2.9. Statistics The results for each variant in the different experimental in vitro designs were represented as an average of two to four experiments (performed in triplicate). Mean values and standard errors were calculated for each point from the pooled normalized data. The one-way analysis of variance (ANOVA) test (SPSS software package, SPSS Inc., Chicago, IL, USA) was used for evaluating the differences in the effect of each drug and their combinations; Statistical significance (Pb.05) was established by the post hoc Tukey's pairwise comparison. Survival studies were assessed using Kaplan–Meier survival curves and analyzed with the Mantel–Cox log-rank test.

3. Results 3.1. Effect of curcumin on NSCLC cell survival in vitro A dose-dependent inhibitory effect of curcumin on survival was found in all human lung carcinoma cell lines tested, i.e., H1299, H358, PC-14 and H1975 (Fig. 1). Half maximal inhibitory concentration (IC50) of curcumin was lower in PC-14 (IC50=10 μM) and H1975 (IC50=15 μM) cells. H1975 cells expressed high levels of COX-2 compared to COX-2-deficient cell lines [H358 (IC50=22 μM) and H1299 (IC50=28 μM)]. The H1975 cell line, which was characterized by constituently activated NF-κB-related proteins, a high level of COX2 and a high degree of tumorigenicity, was chosen for further studies. 3.2. Effect of curcumin on induction of apoptosis in NSCLC cells in vitro Since curcumin treatment resulted in a significant decrease of cell survival, we examined the effect of curcumin on induction of apoptosis. Flow cytometry (FACS) analysis showed a dose-dependent effect of curcumin on the percentage of cells with subdiploid DNA content, characteristic for apoptosis (Fig. 2A and B). We further demonstrated that this treatment resulted in time-dependent caspase-3 activation, as indicated by the increased 17-kDa band (Fig. 2C). Such activation of caspase-3 was accompanied by the cleavage of 118 kDa PARP-1 protein into 87-kDa fragment, another hallmark of cells undergoing apoptosis. 3.3. Effect of curcumin on expression of NF-κB-related proteins NF-κB-related proteins are known to be constitutively activated in lung cancer cells. To investigate whether the inhibitory effect of curcumin on lung cancer cell survival is mediated through the alteration of NF-κB-related proteins, protein extracts of nontreated and curcumin-treated H1975 cells were analyzed by Western blot assay to assess their phospho-IkB and nuclear p65 expression. As shown in Fig. 3A, curcumin treatment decreased the expression of both IkB and nuclear p65 in a dose-dependent manner: a dose of 10 μM decreased both proteins approximately ~50% and at a dose of 25 μM, these proteins were not detectable. Moreover, by using the NoShift Transcription Factor Assay, curcumin treatment was found to significantly inhibit the activity of NF-κB in a dose-dependent manner (Fig. 3B): there was a 70% inhibition at the maximal dose of curcumin used (50 μM).

4

S. Lev-Ari et al. / Journal of Nutritional Biochemistry xx (2014) xxx–xxx

3.4. Curcumin inhibited COX-2 and p-ERK1/2 expression The effect of curcumin on expression of COX-2 and phosphorylated ERK1/2 (p-ERK1/2) was evaluated since COX-2 and phosphorylated ERK1/2 (p-ERK1/2) are important for cancer cell resistance. Western blot analysis showed that curcumin decreased COX-2 expression in H1975 cells in a dose-dependent manner (Fig. 3A). High levels of curcumin inhibited phosphorylation of ERK1/2. Curcumin did not alter the expression of EGFR or p-EGFR (data not shown). 3.5. Combined effect of curcumin and chemotherapy on cell survival Since our H1975 cells were found to be sensitive to curcumin, we further examined the combined effect of curcumin with two chemotherapeutic drugs currently used in clinical oncology, cisplatin and gemcitabine. The doses were chosen within the known range of low inhibitory effect of each drug alone. The cells were treated for 72 h with selected combinations of those agents. The effect of the combined treatment of curcumin with gemcitabine or cisplatin on the survival of H1975 cells was higher than the cytotoxic effect of curcumin or each of the chemotherapeutic drugs alone. CalcuSyn software analysis revealed a synergistic cytotoxic effect on H1975 cells: the CI was less than 1.0 for all studied combinations of curcumin and cisplatin and for most studied combinations of curcumin and gemcitabine (Fig. 4A and B). 3.6. Effect of curcumin on growth of subcutaneous human NSCLC tumor To examine the efficacy of curcumin treatment on the growth of NSCLC cells in vivo, we injected athymic nude mice with H1975 cells subcutaneously. A curcumin diet (0.6%) had been added 14 days prior to inoculation of those cells, and it was continued until the end of the experiment. All of the mice in the untreated and curcumin-treated groups tolerated the sc tumor growth well, as assessed by similar median body weights in the control and treatment groups and the absence of clinical signs of toxicity (data not shown). The mice were sacrificed 5 weeks after cancer cells implantation, when the control mice had become moribund and dyspnea was developed. At the end of the experiment, the mice were necropsied, and the subcutaneous tumors were harvested, fixed in formalin and processed for routine histology and IHC.

Fig. 1. Effect of curcumin on survival of lung cancer cell lines (H1975, H358, PC-14 and H1299). Cells were exposed for 72 h to different concentrations of curcumin. Cell viability was measured by XTT assay. Cell survival was expressed as percentage of viable cells relative to control. Points, mean values from three individual experiments done in triplicates; bars, S.D.

As shown in Fig. 5A, curcumin treatment significantly inhibited the growth of H1975 xenografts. The tumors were more necrotic in the curcumin-treated mice compared to the nontreated mice xenografts. IHC analyses, performed by a licensed pathologist, revealed that the tumors from nontreated mice expressed high levels of COX-2, while COX-2 expression was decreased in the curcumin-treated tumors (Fig. 5B). No quantitative measurements were done. 3.7. Effect of curcumin on growth of orthotopic human NSCLC xenografts H1975 cells in Matrigel were implanted into mouse lungs to generate tumors in a natural orthotopic site that provides a biologically appropriate environment for tumor growth, invasion and metastasis. At 5 weeks postinjection, the control group mice started to die from spread disease. The remaining mice were sacrificed between–up to ~9 weeks after cancer cells implantation when all the control mice had become moribund. All the mice had developed intralung tumors. Histological observations discovered that the developed tumors in several mice had progressed into the mediastinal lymph nodes and had spread to the right lung as well as into the pleura and chest wall, while several others formed slow-growing tumors leading to cachexia that was characterized by wasting of the interscapular muscles. The nontreated mice displayed a much more invasive tumor cell phenotype, i.e., metastasis to mediastinal lymph nodes, and invasiveness to the pleura, chest wall muscles and ribs, as well as invasiveness to the spine in several of the mice. Curcumin treatment resulted in a 36% decrease in tumor weight (P=.048; Fig. 6), fewer detectable metastases,and a significant increase in survival rate (hazard ratio=2.728, P=.036). 4. Discussion Our in vivo orthotopic model of NSCLC showed, for the first time, that curcumin inhibited the growth of the intralung tumors induced by H1975 cells. Curcumin induced apoptosis in H1975 cells in vitro in a timedependent manner by activating caspase 3 (Fig. 2C), which proteolytically carried out the cell death program. Curcumin inhibited IkB phosphorylation, reduced p65 levels and inhibited the activation of NF-κB. Curcumin has been shown to suppress NF-κB activation as well as a multitude of other biological signals pertinent to cancer [35]. Treatment of human carcinoma cell lines with curcumin led to the down-regulation of constitutive NF-κB activation, suppression of NF-κB-regulated gene products and inhibition of cell growth associated with apoptosis [36–38]. Here, we studied the role of curcumin in suppressing the growth of NSCLC in vitro, followed by an investigation of NSCLC orthotopic intralung xenografts. We found that curcumin suppressed NF-κB activation. This inhibition of NF-κB activation involved the suppression of IKK activation, which led to the suppression of phosphorylation and degradation of IkBa and consequent p65 nuclear translocation. Curcumin down-regulated all NF-κB regulation targets, such as apoptosis and proliferation. Our results are in line with numerous studies that demonstrated the effects of curcumin on NSCLC cell survival through down-regulation of antiapoptotic survival genes in vitro and in vivo [39]. Our current in vivo experiment demonstrated that curcumintreated animals had a significant reduction in tumor growth as was measured by the decrease in size of both the subcutaneous and orthotopic xenografts (36% and 40%, respectively) compared to untreated animals. Also, the treated subcutaneous xenografts had a less aggressive appearance, while IHC staining detected a high expression of COX-2 that was reduced after treatment with curcumin (Fig. 5B). COX-2, a downstream molecular target of NF-κB, is a known NFkB-dependent mediator of inflammation that promotes tumor growth and vascularization: it has already been identified as a target of chemopreventive compounds. We observed down-regulation of

S. Lev-Ari et al. / Journal of Nutritional Biochemistry xx (2014) xxx–xxx

A

5

B Pre-G1

G1

S

G2/M

Curcumin (µM)

sub-G1

G1

S

G2+M

0

0.50 ± 0.41

57.48 ± 4.92

28.33 ± 1.97

13.28 ± 5.25

10

3.13 ± 1.19*

37.28 ± 4.10**

25.25 ± 2.91

32.63 ± 2.86**

25

14.75 ± 2.62**

29.25 ± 7.99**

34.93 ± 8.98

20.68 ± 3.73**

Control

Curcmin (10 µM)

C Curcmin (25 µM)

Fig. 2. Induction of apoptosis by curcumin in H1975 NSCLC. A, typical flow cytometry (FACS) analysis of cell cycle following treatment with low (10 μM) and high (25 μM) doses of curcumin. B, Table of summarized data of four independent flow cytometry experiments (*Pb.05 and **Pb.01). C, Western blot analysis of caspase-3 activity and cleavage of PARP. Cell suspensions and whole cell extracts were prepared and analyzed as described in Materials and Methods.

the expression of COX-2, whose synthesis is known to be regulated by NF-κB. It is, therefore, most likely that the observed effect of curcumin on the reduction in COX-2 expression is mediated by the attenuation of NF-κB binding, which is needed for COX-2 expression. The enzyme COX-2 is overexpressed in many cancers. It can promote tumor development and progression when activated and has a direct impact on lung carcinogenesis [15]. Components of tobacco smoke and other pulmonary microenvironmental pollutants

can induce COX-2 expression and prostaglandin E2 (PGE2) production [40]. There has been much evidence to suggest that selective inhibition of COX-2 (COXIBs), and celecoxib in particular, reduced the formation and progression of tumors in both animal models and humans [10,12]. Most importantly, these agents were found to inhibit the progression of lung tumors in vivo. Moreover, several studies indicated that COXIBs could enhance the inhibitory effect of tumor growth and the efficacy of chemotherapy agents on NSCLC when used

Fig. 3. Effect of curcumin treatment on expression of NF-κB-related proteins in H1975 cells. A, Expression of IkB, nuclear p65, COX-2 and p-ERK1/2 determined by Western blot analysis. B, NF-κB activity evaluated by NoShift Transcription Factor Assay. Cells were treated with different concentrations of curcumin (0–25 μM) for 72 h and then collected for both assays used as described in Materials and Methods.

6

S. Lev-Ari et al. / Journal of Nutritional Biochemistry xx (2014) xxx–xxx

Fig. 4. Effect of curcumin treatment in combination with gemcitabine (A) or cisplatin (B) on H1975 lung cancer cell survival. Cells were exposed for 72 h to different treatments. Cell viability was measured by XTT assay. The mode of combined effect was analyzed by CalcuSyn software.

alone and in combination with chemotherapy [41]. Supported by these findings, phase II–III clinical trials were commenced, incorporating COX-2 inhibitors into therapies for advanced NSCLC, but they yielded controversial results. Although chemotherapy plus COX-2 inhibitors seems to be the right strategy designed for the treatment of NSCLC, the abundant preclinical and clinical studies failed to show any survival benefit for patients with NSCLC. Moreover, the prolonged use of COXIBs was found to be associated with severe cardiovascular side effects (for review, see Ref. [42]). Wang et al. [15] recently found that celecoxib promoted cell invasion leading to metastasis and chemoresistance through the induction of epithelial-mesenchymal transition in NSCLC cells, indicating that the use of Cox-2 inhibitors, especially celecoxib, may jeopardized NSCLC patients by causing cancer metastasis and chemoresistance. We have shown previously that

curcumin inhibitory effect on cell survival was mediated through a mechanism that probably involved inhibition of the COX-2 pathway in vitro and other non-COX-2 pathways. This conclusion was based on the evaluation of the effect of curcumin on COX-2 protein expression, COX-2 mRNA expression and PGE2 production levels in the cells expressing high and low levels of COX-2 [21]. It is obviously crucial to develop a better and effective chemopreventive agent with minimal toxicity that can successfully inhibit tumor development and progression. When chemotherapy is considered for a patient, cisplatin combined with gemcitabine may be considered as first-line chemotherapy for the treatment of NSCLC [2]. Here, we showed that curcumin significantly enhanced the antitumor efficacy of each of those drugs. Our implementation of CalcuSyn software to calculate

Fig. 5. Effect of curcumin on growth of subcutaneous tumors following implantation of H1975 lung cancer cells in nude mice. A, Tumor size. B, COX-2 expression. Cells (1×106/0.2 ml PBS) were injected to the flank of CD-1 nude mice. Experiments were terminated when the mice became moribund. Mice were autopsied, and their tumor tissues were harvested and fixed in formalin for histological evaluation. Tumor size was measured twice weekly by calipers. Each point represents average pooled data from 8 to 10 tumors. The expression of COX-2 in tumor tissues from nontreated and treated mice was evaluated by licensed pathologist.

S. Lev-Ari et al. / Journal of Nutritional Biochemistry xx (2014) xxx–xxx

7

Fig. 6. Effect of curcumin treatment on the growth of intralung tumors following implantation of H1975 lung cancer cells in nude mice. A, Lung tumor weight. Columns, mean values of lung tumor weight from three individual experiments; bars, S.D. B, Survival rate of mice after cancer cells implantation. Cells (1×106/0.2 ml PBS) were injected percutaneously with Growth Factor Reduced Matrigel (500 μg of Matrigel in 1 mL of PBS) into the left lung with 30-gauge hypodermic needles. The experiments were terminated when the mice became moribund. Mice were autopsied, and their tumor tissues were harvested, weighed and fixed in formalin for histological evaluation. A and B, Differences in tumor growth after exposure to curcumin were determined using the one-way ANOVA test. ⁎Significant differences (Pb.05).

the effect of combined therapy of curcumin with gemcitabine or cisplatin yielded results that clearly indicated a synergistic interaction between curcumin and these therapeutic agents in certain concentrations. Platinum-based two-drug chemotherapy is a standard first-line treatment of patients with stage IIIB or IV NSCLC. This therapeutic approach is associated with response rates ranging from 19% to 37%, median survival time ranging from 7 to 10 months, and a 1-year survival rate of b45% [42]. Treatment with gemcitabine or cisplatin, combined or alone, improved survival by only few weeks (average= 5.1 months [42]). Our results are in accord with other studies by showing that curcumin can potentiate the efficacy of gemcitabine [43] and of cisplatin [44] in the treatment of NSCLC. The results of the current study demonstrate that curcumin can effectively suppress NF-κB activity and COX-2 expression, as well as cell proliferation/survival in the setting of NSCLC. Several clinical phase I–III trials have shown that patients can tolerate up to 12 g per day when free curcumin is ingested [30,45]. Other studies have shown that curcumin is a potent radiosensitizer of human tumor cells [46–48]. The data that emerged from our current study support the concept of a potential role of curcumin in chemoprevention and treatment of NSCLC and as an effective way to improve the anticancer activity of currently available chemodrugs and in offering an effective way to improve the anticancer activity of currently available chemodrugs. The advantage of a nontoxic preventive therapy for cancer patients who have received adjuvant therapy and face the risk of relapse would be substantial.

References [1] American Cancer Society I. Global facts & figures surveillance research; 2013. [2] NCCN guideline version for NSCLC 2.2013. NCCN National comprehensive cancer network inc; 2013. [3] Hertel LW, Boder GB, Kroin JS, Rinzel SM, Poore GA, Todd GC, et al. Evaluation of the antitumor activity of gemcitabine (2′,2′-difluoro-2′-deoxycytidine). Cancer Res 1990;50(14):4417–22. [4] Ra Lohman. Cisplatin: synthesis, antitumour activity and mechanism of action. Pharm Weekbl Sci 1985;7(5):173–80. [5] Kim SB, Kim JS, Lee JH, Yoon WJ, Lee DS, Ko MS, et al. NF-kappaB activation is required for cisplatin-induced apoptosis in head and neck squamous carcinoma cells. FEBS Lett 2006;580(1):311–8.

[6] Lin YBL, Chen W, Xu S. The NF-kappaB activation pathways, emerging molecular targets for cancer prevention and therapy. Expert Opin Ther Targets 2010;14 (1):45–55. [7] Ducut Sigala JL, Bottero V, Young DB, Shevchenko A, Mercurio F, Verma IM. Activation of transcription factor NF-kappaB requires ELKS, an IkappaB kinase regulatory subunit. Science 2004;304(5679):1963–7. [8] Charalambous MP, Lightfoot T, Speirs V, Horgan K, Gooderham NJ. Expression of COX-2, NF-kappaB-p65, NF-kappaB-p50 and IKKalpha in malignant and adjacent normal human colorectal tissue. Br J Cancer 2009;101(1):106–15. [9] Charalambous MP, Maihöfner C, Bhambra U, Lightfoot T, Gooderham NJ. Colorectal Cancer Study Group. Upregulation of cyclooxygenase-2 is accompanied by increased expression of nuclear factor-kappa B and I kappa B kinase-alpha in human colorectal cancer epithelial cells. Br J Cancer 2003;88:1598–604. [10] Hida T, Kozaki K, Ito H, Miyaishi O, Tatematsu Y, Suzuki T, et al. Significant growth inhibition of human lung cancer cells both in vitro and in vivo by the combined use of a selective cyclooxygenase 2 inhibitor, JTE-522, and conventional anticancer agents. Clin Cancer Res 2002;8(7):2443–7. [11] Krysan KMF, Zhu L, Dohadwala M, Luo J, Lin Y, Heuze-Vourc'h N, et al. COX-2dependent stabilization of survivin in non-small cell lung cancer. FASEB J 2004;18 (1):206–8. [12] Masferrer JL, Leahy KM, Koki AT, Zweifel BS, Settle SL, Woerner BM, et al. Antiangiogenic and antitumor activities of cyclooxygenase-2 inhibitors. Cancer Res 2000;60:1306–11. [13] Groen HJ, Sietsma H, Vincent A, Hochstenbag MM, van Putten JW, van den Berg A, et al. Randomized, placebo-controlled phase III study of docetaxel plus carboplatin with celecoxib and cyclooxygenase-2 expression as a biomarker for patients with advanced non-small-cell lung cancer: the NVALT-4 study. J Clin Oncol 2011;29(32):4320–6. [14] Takahashi T, Kozaki K, Yatabe Y, Achiwa H, Hida T. Increased expression of COX-2 in the development of human lung cancers. J Environ Pathol Toxicol Oncol 2002;21(2):177–81. [15] Wang ZL, Fan ZQ, Jiang HD, Qu JM. Selective Cox-2 inhibitor celecoxib induces epithelial-mesenchymal transition in human lung cancer cells via activating MEK–ERK signaling. Carcinogenesis 2013;34(3):638–46. [16] Harris RE, Beebe-Donk J, Alshafie GA. Reduced risk of human lung cancer by selective cyclooxygenase 2 (COX-2) blockade: results of a case control study. Int J Biol Sci 2007;13(3):328–34. [17] McGettigan P, Henry D. Cardiovascular risk with non-steroidal anti-inflammatory drugs: systematic review of population-based controlled observational studies. PLoS Med 2011;8(9). [18] Darvesh AS, Aggarwal BB, Bishayee A. Curcumin and liver cancer: a review. Curr Pharm Biotechnol 2012;13(1):218–28. [19] Goel A, Boland CR, Chauhan DP. Specific inhibition of cyclooxygenase-2 (COX-2) expression by dietary curcumin in HT-29 human colon cancer cells. Cancer Lett 2001;172(2):111–8. [20] Zhang F, Altorki NK, Mestre JR, Subbaramaiah K, Dannenberg AJ. Curcumin inhibits cyclooxygenase-2 transcription in bile acid- and phorbol estertreated human gastrointestinal epithelial cells. Carcinogenesis 1999;20 (3):445–51. [21] Lev-Ari S, Strier L, Kazanov D, Madar-Shapiro L, Dvory-Sobol H, Pinchuk I, et al. Celecoxib and curcumin synergistically inhibit the growth of colorectal cancer cells. Clin Cancer Res 2005;11(18):6738–44.

8

S. Lev-Ari et al. / Journal of Nutritional Biochemistry xx (2014) xxx–xxx

[22] Lev-Ari S, Strier L, Kazanov D, Elkayam O, Lichtenberg D, Caspi D, et al. Curcumin synergistically potentiates the growth-inhibitory and pro-apoptotic effects of celecoxib in osteoarthritis synovial adherent cells. Rheumatology (Oxford) 2006;45(2):171–7. [23] Lev-Ari S, Starr A, Vexler A, Karaush V, Loew V, Greif J, et al. Inhibition of pancreatic and lung adenocarcinoma cell survival by curcumin is associated with increased apoptosis, down-regulation of COX-2 and EGFR and inhibition of Erk1/2 activity. Anticancer Res 2006;26(6B):4423–30. [24] Jobin C, Bradham CA, Russo MP, Juma B, Narula AS, Brenner DA, et al. Curcumin blocks cytokine-mediated NF-kappa B activation and proinflammatory gene expression by inhibiting inhibitory factor I-kappa B kinase activity. J Immunol 1999;163(6):3474–83. [25] Kunnumakkara AB, Guha S, Krishnan S, Diagaradjane P, Gelovani J, Aggarwal BB. Curcumin potentiates antitumor activity of gemcitabine in an orthotopic model of pancreatic cancer through suppression of proliferation, angiogenesis, and inhibition of nuclear factor-kappaB-regulated gene products. Cancer Res 2007;67(8):3853–61. [26] Tharakan ST, Inamoto T, Sung B, Aggarwal BB, Kamat AM. Curcumin potentiates the antitumor effects of gemcitabine in an orthotopic model of human bladder cancer through suppression of proliferative and angiogenic biomarkers. Biochem Pharmacol 2010;79(2):218–28. [27] Lev-Ari S, Vexler A, Starr A, Ashkenazy-Voghera M, Greif J, Aderka D, et al. Curcumin augments gemcitabine cytotoxic effect on pancreatic adenocarcinoma cell lines. Cancer Invest 2007;25(6):411–8. [28] Duarte VM, Han E, Veena MS, Salvado A, Suh JD, Liang LJ, et al. Curcumin enhances the effect of cisplatin in suppression of head and neck squamous cell carcinoma via inhibition of IKKbeta protein of the NFkappaB pathway. Mol Cancer Ther 2010;9(10):2665–75. [29] Wilken R, Veena MS, Wang MB, Srivatsan ES. Curcumin: a review of anti-cancer properties and therapeutic activity in head and neck squamous cell carcinoma. Mol Cancer 2011;10:12. [30] Dhillon N, Aggarwal BB, Newman RA, Wolff RA, Kunnumakkara AB, Abbruzzese JL, et al. Phase II trial of curcumin in patients with advanced pancreatic cancer. Clin Cancer Res 2008;14(14):4491–9. [31] Shpitz B, Giladi N, Sagiv E, Lev-Ari S, Liberman E, Kazanov D, et al. Celecoxib and curcumin additively inhibit the growth of colorectal cancer in a rat model. Digestion 2006;74(3–4):140–4. [32] Killian PH, Kronski E, Michalik KM, Barbieri O, Astigiano S, Sommerhoff CP, et al. Curcumin inhibits prostate cancer metastasis in vivo by targeting the inflammatory cytokines CXCL1 and -2. Carcinogenesis 2012;33(12):2507–19. [33] Chou T-C, Talalay P. Analysis of combined drug effects: a new look at a very old problem. Trends Pharmacol Sci 1983;4:450–4. [34] Wulff S, editor. Guide to special stains. Carpinteria, California, USA: DakoCytomation; 2004.

[35] Wilken R, Veena MS, Wang MB, Srivatsan ES. Curcumin: a review of anti-cancer properties and therapeutic activity in head and neck squamous cell carcinoma. Mol Cancer 2011;10. [36] Bharti AC, Donato N, Singh S, Aggarwal BB. Curcumin (diferuloylmethane) downregulates the constitutive activation of nuclear factor-kappa B and IkappaBalpha kinase in human multiple myeloma cells, leading to suppression of proliferation and induction of apoptosis. Blood 2003;101(3):1053–62. [37] Shishodia S, Potdar P, Gairola CG, Aggarwal BB. Curcumin (diferuloylmethane) down-regulates cigarette smoke-induced NF-kappaB activation through inhibition of IkappaBalpha kinase in human lung epithelial cells: correlation with suppression of COX-2, MMP-9 and cyclin D1. Carcinogenesis 2003;24 (7):1269–79. [38] Hussain AR, Ahmed M, Al-Jomah NA, Khan AS, Manogaran P, Sultana M, et al. Curcumin suppresses constitutive activation of nuclear factor-kappa B and requires functional Bax to induce apoptosis in Burkitt's lymphoma cell lines. Mol Cancer Ther 2008;7(10):3318–29. [39] Li L, Braiteh FS, Kurzrock R. Liposome-encapsulated curcumin: in vitro and in vivo effects on proliferation, apoptosis, signaling, and angiogenesis. Cancer 2005;104(6):1322–31. [40] Moraitis D, Du B, De Lorenzo MS, Boyle JO, Weksler BB, Cohen EG, et al. Levels of cyclooxygenase-2 are increased in the oral mucosa of smokers: evidence for the role of epidermal growth factor receptor and its ligands. Cancer Res 2005;65(2):664–70. [41] Williams CS, Tsujii M, Reese J, Dey SK, DuBois RN. Host cyclooxygenase-2 modulates carcinoma growth. J Clin Invest 2000;105(11):1589–94. [42] Manegold C, vZN, Szczesna A, Zatloukal P, Au JS, Blasinska-Morawiec M, et al. A phase III randomized study of gemcitabine and cisplatin with or without PF-3512676 (TLR9 agonist) as first-line treatment of advanced non-small-cell lung cancer. Ann Oncol 2012;23(1):72–7. [43] Rocks N, Bekaert S, Coia I, Paulissen G, Gueders M, Evrard B, et al. Curcumin– cyclodextrin complexes potentiate gemcitabine effects in an orthotopic mouse model of lung cancer. Br J Cancer 2012;107(7):1083–92. [44] Tsai MS, Weng SH, Kuo YH, Chiu YF, Lin YW. Synergistic effect of curcumin and cisplatin via down-regulation of thymidine phosphorylase and excision repair cross-complementary 1 (ERCC1). Mol Pharmacol 2011;80(1):136–46. [45] Anand P, Kunnumakkara AB, Newman RA, Aggarwal BB. Bioavailability of curcumin: problems and promises. Mol Pharm 2007;4(6):807–18. [46] Chendil D, Ranga RS, Meigooni D, Sathishkumar S, Ahmed MM. Curcumin confers radiosensitizing effect in prostate cancer cell line PC-3. Oncogene 2004;23(8):1599–607. [47] Orr WS, Denbo JW, Saab KR, Ng CY, Wu J, Li K, et al. Curcumin potentiates rhabdomyosarcoma radiosensitivity by suppressing NF-kappaB activity. PLoS ONE 2013:e51309. [48] Sreekanth CN, Bava SV, Sreekumar E, Anto RJ. Molecular evidences for the chemosensitizing efficacy of liposomal curcumin in paclitaxel chemotherapy in mouse models of cervical cancer. Oncogene 2011;30(28):3139–52.