www.elsevier.com/locate/ybbrc

Biochemical and Biophysical Research Communications 331 (2005) 1245–1252

BBRC

Curcumin enhances Vinorelbine mediated apoptosis in NSCLCcells by the mitochondrial pathway

Sudip Sen, Himani Sharma, Neeta Singh *

Department of Biochemistry, All India Institute of Medical Sciences, New Delhi, India

Received 4 April 2005Available online 19 April 2005

Abstract

Elderly lung cancer patients and those with poor performance status/co-morbid conditions are deprived of chemotherapybecause of high toxicity of multidrug regimens. Human squamous cell lung carcinoma H520 cells treated with Curcumin were sen-sitized to the cytotoxicity caused by chemotherapeutic agent, Vinorelbine. Both caused apoptosis by increasing the protein expres-sion of Bax and Bcl-xs while decreasing Bcl-2 and Bcl-XL, releasing apoptogenic cytochrome c, and augmenting the activity ofcaspase-9 and caspase-3. Expression of Cox-2, NF-jB, and AP-1 was also affected. 23.7% apoptosis was induced in the H520 cellsby treatment with Curcumin while Vinorelbine caused 38% apoptosis. Pre-treatment with Curcumin enhanced the Vinorelbineinduced apoptosis to 61.3%. The findings suggest that Curcumin has the potential to act as an adjuvant chemotherapeutic agentand enhance chemotherapeutic efficacy of Vinorelbine in H520 cells in vitro. Thus, Curcumin offers the prospect of being beneficialin the above-mentioned patient groups.� 2005 Elsevier Inc. All rights reserved.

Keywords: Curcumin; Vinorelbine; NSCLC; Chemotherapy; Apoptosis; Mitochondrial pathway

Lung cancer is a leading cause of cancer deaths allover the world. It has the notorious distinction of beingthe most common cancer in males in developed as wellas developing nations [1]. The role of chemotherapy inthe treatment of patients with advanced non-small celllung carcinoma (NSCLC) has been defined in a meta-analysis [2] wherein cisplatin based polychemotherapyis generally considered the most advisable frontline ap-proach in these patients. However, some concerns stillpersist about the use of polychemotherapy especiallywith the inclusion of Cisplatin in particular subsets ofpatients with NSCLC such as older patients or thosewith poor performance status [3]. Physiological reduc-tion of functional reserve and the presence of co-morbidconditions make elderly patients unsuitable for Cisplatin

0006-291X/$ - see front matter � 2005 Elsevier Inc. All rights reserved.

doi:10.1016/j.bbrc.2005.04.044

* Corresponding author. Fax: +91 11 26588663.E-mail address: singh_neeta@hotmail.com (N. Singh).

based multi-agent chemotherapy because of their highpotential toxicity. [4–6]. Elderly Lung Cancer Vinorel-bine Study Group (ELVIS), Multicenter Italian LungCancer in the Elderly Study (MILES), and some otherstudies concluded that a new single-agent like Vinorel-bine which is a semi-synthetic vinca alkaloid is as effec-tive and less toxic than the earlier Cisplatin basedcombinations and can be considered for elderly and un-fit patients [7]. The benefit of chemotherapy to patientshaving poor performance status is still in doubt as sometrials purposefully do not include such patients [8].

Curcumin (diferuloyl methane), a polyphenolic phy-tochemical, is a primary component of the dietary spice,turmeric. The pharmacological safety of Curcumin iswell demonstrated by the fact that people in certaincountries have consumed Curcumin as a dietary spicefor centuries in amounts in excess of 100 mg/day with-out any side effects [9]. Curcumin is a well-known

1246 S. Sen et al. / Biochemical and Biophysical Research Communications 331 (2005) 1245–1252

chemopreventive that blocks tumor initiation, promo-tion, and carcinogenesis [10,11]. The anticarcinogenicmechanisms of Curcumin action are not fully under-stood. It appears that Curcumin can suppress the cellgrowth pathway by inhibiting cellular protein kinasessuch as PKC, JNK, and the EGF receptor kinase, lead-ing to growth inhibition [12–14]. It also has the ability toblock the NF-jB cell survival pathway [15] and inhibitc-jun/AP-1 function [16]. Very few studies have lookedinto Curcumin�s role as an adjuvant chemotherapeuticin cancer. This study is a step in this direction and thefindings suggest that it has the potential to be used asa chemo-adjuvant in elderly patients or patients withpoor performance status without increasing toxicity ofthe chemotherapeutic regimen.

Materials and methods

Cell culture and treatments. Human NSCLC (squamous cell carci-noma) cell line NCI-H520 was maintained in DMEM (Sigma) sup-plemented with 10% fetal calf serum and antibiotics in a humidifiedatmosphere of 5% CO2 in air at 37 �C [17]. Logarithmically growingcells were treated with Curcumin (Sigma), Vinorelbine (a gift fromDabur Oncology Division, New Delhi, India), and a combination ofthe two for various time periods. Curcumin stock solutions were madeup in dimethyl sulfoxide (DMSO) (Sigma); stored at �20 �C; and finalworking concentration used was 0.1% DMSO. Vinorelbine solutionswere made in normal saline (0.9% w/v) just prior to each experiment.Combination treatment was given at optimized dose of Curcumin for24 h followed by Vinorelbine for 24 h.

MTT assay. H520 cells were grown in 96-well microtiter plates andtreated with Curcumin, Vinorelbine, and a combination of the two inreplicate. Each experiment was repeated three times. One hundredmicroliters of MTT (3-[4,5-dimethyl thiazol-2yl]-2,5-diphenyltetrazo-lium bromide) (Sigma) solution (5 mg/ml) was added to each wellfollowed by incubation for 4 h at 37 �C. The formazan crystals formedwere dissolved in DMSO and the absorbance was measured at 570 nmusing an ELISA reader [18].

In vitro drug dosage was standardized using the MTT assay,serum maximum levels (Cmax), and area under curve (AUC) values.The concentration of Curcumin and Vinorelbine to be administeredwas chosen from a dose–response experiment showing moderatetoxicity to the cells (data not shown). The lowest possible non-toxicdosage was used. A dose of 25 lM was fixed for Curcumin while forVinorelbine 0.1 lg/ml was optimized. Duration of treatment foreach drug was 24 h. Curcumin treatment alone (25 lM for 48 h) wasused to assess the efficacy of the combination of Curcumin andVinorelbine.

Fluorescence microscopy. H520 cells were grown on coverslips.After treatment with Curcumin, Vinorelbine, and a combination of thetwo for the designated time periods, cells were washed with PBS, fixedin 80% methanol, stained with 5 lg/ml propidium iodide (Sigma), andobserved under a fluorescent microscope (Nikon Microphot FXA),and apoptotic cells were identified and counted [18].

Flowcytometric analysis. For flowcytometry, the cells were tryp-sinized, fixed in 70% ethanol, and incubated in the staining solution(20 lg/ml propidium iodide, 50 lg/ml RNase, 0.1% Triton X-100, and0.1 mM EDTA) for 1 h at 4 �C in the dark. The cellular DNA contentwas analyzed using 488 nm for excitation and fluorescence measured at600 nm by EPICS XL-MCL flowcytometer (Coulter Electronics, FL,USA). For each analysis 10,000 events were counted and the data wereanalyzed using Win MDI 2.8 software [18].

Immunocytochemical analysis. Protein expression and intracellularlocalization of transcription factors were assessed by this method.Cells were grown on coverslips and treated with Curcumin alone for24 h, and for 48 h, Vinorelbine alone for 24 h and Curcumin (24 h)followed by Vinorelbine (24 h). The cells were fixed with 4% para-formaldehyde. Endogenous peroxidases were blocked with hydrogenperoxide in PBS containing 70% methanol and non-specific bindingblocked using 5% bovine serum albumin (BSA). The cells were thenincubated with antibodies against NF-jB and c-jun/AP-1 (SantaCruz Biotechnology, CA) at a dilution of 1:100 for 24 h at 4 �C.Immunodetection was achieved by an avidin–biotin horseradishperoxidase-based colorimetric method (Vectastain Elite Kit fromVector laboratories, USA) with 3,3 0-diaminobenzidine (DAB) as achromogen and H2O2 as the substrate, followed by light counterstaining with hematoxylin and examination under a microscope. Theprotein expression was determined semiquantitatively. Specimenswere considered as immunopositive if at least 5% of the cells dis-played distinct immunostaining. Scoring of immunopositivity wasdone on the basis of percentage of cells stained as well as theintensity of staining.

Western blot analysis. The cells were lysed in RIPA lysis buffercontaining protease inhibitors. Equal amounts of protein extracts wereelectrophoresed on 10–15% SDS–polyacrylamide gels and electro-transferred to nitrocellulose membrane. The membrane was thenincubated in 5% BSA for 3 h followed by overnight incubation withantibodies against rabbit Bcl-2, Bcl-XL, Bcl-xs, p53, Bax, Cox-2, andpoly ADP ribose polymerase (PARP) and b-actin (Santa Cruz Bio-technology, USA). b-Actin was used as a loading control. Afterwashing, anti-rabbit alkaline phosphatase, conjugated antibody wasadded and incubated for 2 h. After washing, color development wasdone using 5-bromo-4-chloro-3-indolyl phosphate/nitroblue tetrazo-lium (BCIP/NBT) substrate from Promega Corporation, USA. Thebands were analyzed and quantitated using scanning densitometer(Bio-Rad) [19].

Measurement of cytochrome c release. Cytosolic extracts of the cellswere prepared as described [20]. The cytosolic fraction was collectedby centrifugation (12,000 rpm for 30 min at 4 �C). One hundred andfifty micrograms of protein was electrophoresed on a 15% SDS–polyacrylamide gel and transferred to a nitrocellulose membrane.Cytochrome c was detected by Western blotting using mouse mono-clonal anti-cytochrome c antibody (Santa Cruz Biotechnology, USA)(1:350 dilution) for 4 h. b-Actin was used as a loading control. Afterwashing, anti-rabbit alkaline phosphatase-conjugated antibody wasadded and incubated for 2 h. After washing, color development wasdone using 5-bromo-4-chloro-3-indolyl phosphate/nitroblue tetrazo-lium (BCIP/NBT) substrate from Promega Corporation, USA. Thebands were analyzed and quantitated using scanning densitometer(Bio-Rad) [19].

TUNEL assay. Apoptotic cells were visualized by the terminaldeoxynucleotidyl transferase mediated dUTP nick end labeling (TU-NEL) technique using the Dead End Colorimetric Cell Death Detec-tion kit (Promega, USA) as described earlier [19]. The apoptotic index(AI) (number of apoptotic cells in 500 cells) was determined bymicroscopic examination of randomly selected fields containing at least500 cells.

Caspase-3, -8, and -9 activity assay. Caspase-3, -8, and -9 activitieswere measured by using synthetic fluorogenic substrate (Ac-DEVD-AMC, substrate for caspase-3; Pharmingen, Germany; Ac-LETD-AFC, substrate for caspase-8, and Ac-LEHD-AFC, substrate forcaspase-9; Genotech, USA) as described previously [17]. Amounts offluorogenic AMC/AFC moiety released were measured using a spec-trofluorimeter (ex. 380 nm, em. 420–460 nm for caspase-3; ex. 400 nm,em. 490–520 nm for caspase-8 and -9).

Statistical analysis. Results are expressed as means and standarderror of mean (SEM). Statistical analyses were performed with Stu-dent�s two-tailed paired t test using SPSS (Windows version 7.5).Values of p < 0.05 were considered statistically significant.

Fig. 1. MTT assay; flowcytometric evaluation of H520 cells in variousstages of cell cycle and TUNEL staining in control—AA0; cells treatedwith Curcumin (25 lM for 24 h)—BB 0; cells treated with Curcumin(25 lM for 48 h)—CC0; cells treated with Vinorelbine (0.1 lg/ml for24 h)—DD0; and cells treated with Curcumin (25 lM for 24 h)followed by Vinorelbine (0.1 lg/ml for 24 h)—EE0. F, bar diagramcomparing the percentage cytotoxicity/apoptosis in control—bar 1;cells treated with Curcumin (25 lM for 24 h)—bar 2; cells treated withCurcumin (25 lM for 48 h)—bar 3; cells treated with Vinorelbine(0.1 lg/ml for 24 h)—bar 4; and cells treated with Curcumin (25 lMfor 24 h) followed by Vinorelbine (0.1 lg/ml for 24 h)—bar 5. Theresults are means ± SEM of three individual experiments.

S. Sen et al. / Biochemical and Biophysical Research Communications 331 (2005) 1245–1252 1247

Results

Pre-treatment of H520 cells with Curcumin enhances

the apoptosis induced by Vinorelbine

MTT assay showed that Curcumin alone (24 h) in-duced 29.8 ± 2.1% (p = 0.013) cytotoxicity, Curcumin(48 h) induced 30.5 ± 2.2% (p = 0.023) cytotoxicitywhile Vinorelbine alone induced 36 ± 2.16%(p = 0.005) cytotoxicity. Priming the cells with 25 lMCurcumin for 24 h prior to treatment with 0.1 lg/mlVinorelbine (for 24 h), increased the cytotoxicity to64.4 ± 3.2% (p = 0.010 with Curcumin alone (48 h)and p = 0.003 with Vinorelbine alone) (Fig. 1F). Dis-tinct morphological changes characterized by membraneblebbing, chromatin condensation, and formation ofapoptotic bodies were also observed. The above findingswere confirmed by flowcytometry (Fig. 1) which showedthat the apoptosis induced by treating H520 cells withCurcumin followed by Vinorelbine [61.3 ± 1.7%(p = 0.003 with Curcumin alone (48 h) and p = 0.038with Vinorelbine alone)] is more than individual treat-ment with Curcumin (24 h) [23.7 ± 1.4% (p = 0.006)],Curcumin (48 h) [28.8 ± 2.94% (p = 0.015)], or Vinorel-bine [37.9 ± 3% (p = 0.007)] and 5.6 ± 1.6% observed inthe vehicle treated (0.1% DMSO) control cells. TUNELassay showed control cells having an apoptotic index of0.053 ± 0.01, AI of cells treated with Curcumin (24 h)0.216 ± 0.018, AI of Curcumin (48 h) 0.27 ± 0.008, AIof Vinorelbine (24 h) 0.36 ± 0.007, and AI of Curcuminfollowed by Vinorelbine being 0.6 ± 0.02 (Fig. 1).

The apoptotic response of Curcumin and Vinorelbine

is mediated by increasing the protein expression of

pro-apoptotic members and downregulating the

anti-apoptotic members

It was investigated whether the individual drugs or acombination of the two could affect the expression ofp53 and Bcl-2 family members. The protein expressionof p53 was increased 2.16-fold by Curcumin alone(24 h) (p = 0.014), 2.2-fold by Curcumin (48 h)(p = 0.004), and 2.6-fold by Vinorelbine alone(p = 0.003) as compared to the control whereas theircombination caused a 3.2-fold increase (p = 0.011 withCurcumin alone (48 h) and p = 0.024 with Vinorelbinealone) (Fig. 2B). Antiapoptotic Bcl-2 was decreased1.69-fold (p = 0.001) by Curcumin (24 h), 1.7-fold(p = 0.006) by Curcumin (48 h), and 1.15-fold(p = 0.017) by Vinorelbine alone, while their combina-tion caused a 1.86-fold reduction in the same protein(p = 0.432 with Curcumin alone (48 h) and p = 0.004with Vinorelbine alone) (Fig. 2A). Curcumin decreasedantiapoptotic Bcl-XL by 1.56-fold (24 h) (p = 0.027),1.55-fold (48 h) (p = 0.010) and Vinorelbine 1.8-fold(p = 0.011), whereas their combination caused a 2.5-fold

decrease (p = 0.013 with Curcumin alone (48 h) andp = 0.058 with Vinorelbine alone) (Fig. 2A). Proapop-totic Bcl-xs was increased 1.84-fold (p < 0.001) by

Fig. 2. Western blot analysis in H520 cell lysates for (A) Bcl-XL, Bcl-2, Cox-2, and poly ADP ribose polymerase (PARP), and (B) Bcl-xs, p53, Bax,and cytochrome c (in cytosolic fraction) in control—bar 1; cells treated with Curcumin (25 lM for 24 h)—bar 2; cells treated with Curcumin (25 lMfor 48 h)—bar 3; cells treated with Vinorelbine (0.1 lg/ml for 24 h)—bar 4; and cells treated with Curcumin (25 lM for 24 h) followed by Vinorelbine(0.1 lg/ml for 24 h)—bar 5. With representative blots. The results are means ± SEM of three individual experiments.

1248 S. Sen et al. / Biochemical and Biophysical Research Communications 331 (2005) 1245–1252

Curcumin (24 h), 2.2-fold (p = 0.027) by Curcumin(48 h), and 2.1-fold (p = 0.004) by Vinorelbine whiletheir combination caused a 2.9-fold increase (p = 0.047with Curcumin alone (48 h) and p = 0.059 with Vinorel-bine alone) (Fig. 2B). Bax, another pro-apoptotic pro-tein, was up-regulated by 1.4-fold (p = 0.031) byCurcumin alone (24 h), 1.46-fold (p = 0.013) by Curcu-min alone (48 h), 1.9-fold (p = 0.022) by Vinorelbinealone, while their combination caused a 2.6-fold upreg-ulation (p = 0.010 with Curcumin alone (48 h) andp = 0.007 with Vinorelbine alone) (Fig. 2B). Cox-2 wasdecreased by 1.65-fold (p = 0.001) by Curcumin alone(24 h), 1.9-fold (p = 0.024) by Curcumin alone (48 h),and 1.1-fold (p = 0.01) by Vinorelbine treatment whiletheir combination decreased it by 1.65-fold (p = 0.426with Curcumin alone (48 h) and p = 0.007 with Vinorel-bine alone) (Fig. 2A).

Curcumin and Vinorelbine induced cytochrome c release

from the mitochondria

In an effort to analyze the involvement of mito-chondria in the apoptosis induced by Curcumin andVinorelbine, we looked at the cytosolic levels of cyto-chrome c. Compared to the control, the level of cyto-chrome c in the cytosol of the cells treated withCurcumin alone (24 h) was increased by 1.08-fold(p = 0.020), with Curcumin alone (48 h) by 1.09-fold(p = 0.023), and 1.11-fold (p = 0.019) on treatmentwith Vinorelbine alone. The combination of Curcuminand Vinorelbine increased it by 1.37-fold (p = 0.005with Curcumin alone (48 h) and p = 0.009 with Vino-relbine alone) (Fig. 2B). Release of cytochrome c fromthe mitochondria was validated with a positive control(UV irradiated cells).

S. Sen et al. / Biochemical and Biophysical Research Communications 331 (2005) 1245–1252 1249

Curcumin and Vinorelbine induced apoptosis is mediated

by caspase-9 and -3

Cytochrome c in the presence of apaf-1 activates cas-pase-9, which in turn results in the activation of down-stream caspase-3. We investigated whether Curcuminand Vinorelbine also induce the activation of these casp-ases. Fig. 3 shows Curcumin alone (24 h) activated cas-pase-9 by 2.27-fold (p = 0.004) while Curcumin (48 h)increased the activity by 2.4-fold (p = 0.008). Vinorel-bine alone increased it by 2.98-fold (p = 0.005) whiletheir combination was responsible for bringing about a4.45-fold increase (p = 0.003 with Curcumin alone(48 h) and p < 0.001 with Vinorelbine alone) in this cas-pase. Caspase-8 activity in controls and treated cells re-mained unaffected (Fig. 3). Increased caspase-8 activitywas observed in cells treated with TNF-a (positive con-trol). Downstream effector caspase-3 was increased 2.74-fold (p = 0.001) by Curcumin alone (24 h), 7.15-fold(p = 0.017) by Curcumin alone (48 h), and 11.3-fold(p = 0.002) by Vinorelbine alone while their combina-tion was responsible for a 17.9-fold increase in this effec-tor caspase (p = 0.010 with Curcumin alone (48 h) andp = 0.007 with Vinorelbine alone) (Fig. 3). Since theactivation of downstream caspase-3 by most agentscauses the cleavage of PARP (poly ADP ribose polymer-ase), we examined its expression by Western blotting.The 116 kDa uncleaved fragment of PARP was de-creased 1.39-fold (p = 0.02) by Curcumin alone (24 h),1.5-fold (p = 0.029) by Curcumin alone (48 h), and 2.7-fold (p = 0.011) by Vinorelbine alone while their combi-nation caused a 12-fold decrease (p = 0.006 with Curcu-min alone (48 h) and p = 0.032 with Vinorelbine alone)(Fig. 2A).

Fig. 3. Caspase-8, -9, and -3 activity in H520 cells in control—bar 1;cells treated with Curcumin (25 lM for 24 h)—bar 2; cells treated withCurcumin (25 lM for 48 h)—bar 3; cells treated with Vinorelbine(0.1 lg/ml for 24 h)—bar 4; and cells treated with Curcumin (25 lMfor 24 h) followed by Vinorelbine (0.1 lg/ml for 24 h)—bar 5. Theresults are expressed in terms of arbitrary fluorescence units (AFU) permg protein. The results are means ± SEM of three individualexperiments.

Curcumin and Vinorelbine induced apoptosis is associated

with inhibition of transcription factors NF-jB and AP-1

The expression and nuclear translocation of both thetranscription factors were examined and it was foundthat individually Curcumin and Vinorelbine downregu-late their activity but the combination of Curcuminand Vinorelbine remarkably decreased the expressionand nuclear translocation of both NF-jB and AP-1(Fig. 4).

Fig. 4. Expression and localization of transcription factors, NF-jBand AP-1, as assessed by immunocytochemistry in NSCLC (H520)cells in control—A; cells treated with Curcumin (25 lM for 24 h)—B;cells treated with Curcumin (25 lM for 48 h)—C; cells treated withVinorelbine (0.1 lg/ml for 24 h)—D; and cells treated with Curcumin(25 lM for 24 h) followed by Vinorelbine (0.1 lg/ml for 24 h)—E.(200· magnification). Note the decreased expression and reducednuclear localization of both the transcription factors on treatment withCurcumin, Vinorelbine, and their combination.

1250 S. Sen et al. / Biochemical and Biophysical Research Communications 331 (2005) 1245–1252

Discussion

Search for new chemopreventive and antitumoragents that are more effective but less toxic has kindledgreat interest in phytochemicals. Curcumin, derivedfrom the plant Curcuma longa, is one such compoundwhich was used in this study. It has been shown to inhi-bit the growth of a wide variety of tumor cells in multi-ple experimental model systems [21] but little is knownabout its potential as an adjuvant chemotherapeuticagent. It has been reported that Curcumin, an antioxi-dant, inhibits chemotherapy induced apoptosis of thosechemotherapeutic agents which generate ROS and acti-vate JNK [22]. We have studied the sensitization effect ofCurcumin in Vinorelbine induced apoptosis in humanlung squamous cell carcinoma cell line, H520. The re-sults of this study indicate that priming the cells withCurcumin before treatment with the chemotherapeuticdrug Vinorelbine can enhance the latter�s cytotoxicityin vitro.

Certain chemotherapeutic agents are known to in-duce apoptosis through induction of death receptors[23]. Depending on the cell type, Curcumin inducesapoptosis through both mitochondrial [24] and mito-chondria-independent mechanism [25]. Release of cyto-chrome c, activation of caspase-9 and -3, and cleavageof PARP by both Curcumin and Vinorelbine have beendemonstrated in the present study.

Bcl-2 family proteins are important regulators ofapoptotic signaling, acting to either inhibit (Bcl-2, Bcl-XL) or promote (Bax, Bcl-xs) cell death [26]. Amongthe different types of NSCLC, Bcl-2 expression was de-tected more frequently in squamous cell carcinoma(25–35%) than in adenocarcinoma (�10%) [27]. Bcl-2and Bcl-XL are known to inhibit apoptosis by regulatingmitochondrial membrane potential and cytochrome c

release needed for activation of caspase-9 [28]. Our re-sults further indicate that Curcumin and Vinorelbineby themselves induce apoptosis by downregulatinganti-apoptotic Bcl-2 and Bcl-XL, and upregulatingpro-apoptotic Bcl-xs and Bax, thus leading to cyto-chrome c release from the mitochondria and activatingcaspase-9 and -3. In our study, there appears a directactivation of the mitochondrial pathway withoutinvolvement of caspase-8 contrary to a few prior reports[29,30] but other groups clearly state that activation ofcaspase-9 need not always depend on activation of cas-pase-8 [31,32].

The p53 tumor suppressor gene plays a critical role atthe G1/S cell cycle transition, where it can either blockentry into S phase or activate apoptosis in response toDNA damage. Though an earlier report [33] states thatp53 mutations do not predict survival or response toanti-cancer therapy in patients with NSCLC, this studyshows that both Curcumin and Vinorelbine by them-selves and their combination induce cell death by the

p53 apoptotic pathway. Since pro-apoptotic Bax isknown to be a p53 downstream target, our result ofthe expression of Bax being affected supports p53involvement. The observed increase of p53, Bax, andBcl-xs, and decrease of Bcl-2, Bcl-XL perhaps confersensitivity to apoptosis or favor the onset of apoptosiswhen Curcumin pretreatment is followed by Vinorelbinecompared to that induced by Curcumin/Vinorelbinealone.

Curcumin can affect cellular proliferation throughmodulation of various other signaling pathways liketranscription factors, NF-jB and AP-1; mitogen acti-vated protein kinases; and growth factor receptor ki-nases and cyclooxygenases [21]. Both NF-jB and AP-1translocate to the nucleus on activation and have beenimplicated in the cell survival and proliferation path-ways [34,35]. Curcumin has also been shown to down-regulate the transcription factors, NF-jB and AP-1[36]. It is possible that the decrease in expression and nu-clear translocation of the transcription factors, NF-jBand AP-1, observed in our study may contribute to/bepartially responsible for the chemotherapeutic enhanc-ing effect of Vinorelbine by Curcumin because activationof NF-jB and AP-1 has been demonstrated to play ananti-apoptotic role [34]. Thus, the suppression of NF-jB and AP-1 by both Curcumin and Vinorelbine andtheir combination may explain their antiproliferativeand apoptosis inducing effects.

It is known that NF-jB regulates the expression ofCOX-2 gene transcription [36,37] and ectopic expressionof COX-2 has been shown to block apoptosis whereassuppression of COX-2 enhances apoptosis [38]. It is inter-esting to note that Curcumin alone, but not Vinorelbine,downregulates the expression of COX-2 thus accountingfor the decreased expression of this protein in the combi-nation therapy. Proteins such as COX-2 and Bcl-2 showthat though Curcumin and Vinorelbine cause apoptosisthrough similar mechanisms, there are certain pathwayswhich are not equally affected by both and thus give a ba-sis for the potentiating role of Curcumin on Vinorelbine.

In conclusion, it appears that pre-treatment with Cur-cumin enhances the apoptotic effect of Vinorelbine by themitochondrial pathway in H520 cells by increasing theprotein expression of pro-apoptotic members anddecreasing the anti-apoptotic members; releasing cyto-chrome c and augmenting the activity of caspase-9 anddownstream caspase-3 and cleavage of its substratePARP. Further studies to elucidate bioavailability andmetabolism of Curcumin in vivo are ongoing. The identi-ficationofCurcumin as an agent having apoptotic activityand potential synergy with standard cytotoxic drugs willallow reduction of doses as well as multidrug regimensof toxic chemotherapeutic agents while maintaining oreven enhancing chemotherapeutic efficacy. This is a preli-minary step towards developing promising investiga-tional regimens for the treatment of NSCLC and will

S. Sen et al. / Biochemical and Biophysical Research Communications 331 (2005) 1245–1252 1251

help in aiding reign this killer disease especially in theelderly and immunocompromised patients in the future.

Acknowledgments

The authors are grateful to Department of Biotech-nology, New Delhi, India, for financial support and Da-bur (Oncology division), New Delhi, India, forproviding Vinorelbine.

References

[1] Park. Epidemiology of chronic non-communicable diseases andconditions, in: Park�s Textbook of Preventive and Social Medicine17th edn. M/s Banarsidas Bhanot Publishers, Jabalpur, India,2002, pp. 285–294.

[2] Non-Small Cell Lung Cancer Collaborative Group: Chemother-apy in non-small cell lung cancer: A meta-analysis using updateddata on individual patients from 52 randomised clinical trials. Br.Med. J. 311 (1995) 899–909.

[3] F. Oshita, T. Kurata, T. Kasai, M. Fakuda, N. Yamamoto,Y. Ohe, et al., Prospective evaluation of the feasibility ofcisplatin-based chemotherapy for elderly lung cancer patientswith normal organ functions, Jpn. J. Cancer Res. 86 (1995)1198–1202.

[4] I.S. Fentiman, U. Tirelli, S. Monfardini, M. Schneider, J. Festen,F. Cognetti, et al., Cancer in the elderly: why so badly treated?Lancet 28 (1990) 1020–1022.

[5] S. Monfardini, R. Yancik, Cancer in the elderly: meeting thechallenge of an aging population, J. Natl. Cancer Inst. 85 (1993)532–538.

[6] J.C. Soria, J.Y. Douillar, J.L. Pujol, Prognostic analysis ofsurvival in the European randomized trial comparing navelbine(NVB) vs. navelbine cisplatin (NVB-P) vs. vindesine cisplatin(VDS-P), Proc. Am. Soc. Clin. Oncol. 18 (1999) 491a.

[7] P.A. Bunn Jr., Chemotherapy for advanced non-small-cell lungcancer: who, what, when, why? J. Clin. Oncol. 20 (2002) 23S–33S.

[8] R.O. Dillman, S.L. Seagren, K.J. Propert, J. Guerra, W.L. Eaton,M.C. Perry, et al., A randomized trial of induction chemotherapyplus high-dose radiation vs. radiation alone in stage III non-small-cell lung cancer, N. Engl. J. Med. 323 (1990) 940–945.

[9] H.P. Ammon, M.A. Wahl, Pharmacology of Curcuma longa,Planta Med. 57 (1991) 1–7.

[10] M.T. Huang, Z.Y. Wang, C.A. Georgiadis, J.D. Laskin, A.H.Conney, Inhibitory effects of Curcumin on tumor initiation bybenzo[a]pyrene and 7,12-dimethylbenz[a]anthracene, Carcinogen-esis 13 (1992) 2183–2186.

[11] M.T. Huang, R.C. Smart, C.Q. Wong, A.H. Conney, Inhibitoryeffect of Curcumin, chlorogenic acid, caffeic acid and ferulic acidon tumor promotion in mouse skin by 12-O-tetradecanoylphor-bol-13-acetate, Cancer Res. 48 (1998) 5941–5946.

[12] M.C. Jiang, H.F. Yang-Yen, J.J. Yen, J.K. Lin, Curcumininduces apoptosis in immortalized NIH 3T3 and malignant cancercell lines, Nutr. Cancer 26 (1996) 111–120.

[13] Y.R. Chen, T.H. Tan, Inhibition of the c-Jun N-terminal kinase(JNK) signaling pathway by Curcumin, Oncogene 17 (1998) 173–178.

[14] L. Korutla, J.Y. Cheung, J. Mendelsohn, R. Kumar, Inhibitionof ligand-induced activation of epidermal growth factor receptortyrosine phosphorylation by Curcumin, Carcinogenesis 16 (1995)1741–1745.

[15] S. Singh, B.B. Aggarwal, Activation of transcription factor NF-kappaB is suppressed by Curcumin (diferuloyl-methane), J. Biol.Chem. 270 (1995) 24995–25000.

[16] T.S. Huang, S.C. Lee, J.K. Lin, Suppression of c-Jun/AP-1activation by an inhibitor of tumor promotion in mouse fibroblastcells, Proc. Natl. Acad. Sci. USA 88 (1991) 5292–5296.

[17] N. Singh, N. Khanna, H. Sharma, S. Kundu, S. Azmi, Insightsinto the molecular mechanism of apoptosis induced by TNF-alpha in mouse epidermal JB6-derived RT-101 cells, Biochem.Biophys. Res. Commun. 295 (2002) 24–30.

[18] V.G. Reddy, N. Khanna, N. Singh, Vitamin C augmentschemotherapeutic response of cervical carcinoma HeLa cells bystabilizing P53, Biochem. Biophys. Res. Commun. 282 (2001)409–415.

[19] H. Sharma, S. Sen, M. Mathur, S. Bahadur, N. Singh, Combinedevaluation of expression of telomerase, survivin, and anti-apop-totic Bcl-2 family members in relation to loss of differentiation andapoptosis in human head and neck cancers, Head Neck 26 (2004)733–740.

[20] G.D. Diaz, Q. Li, R.H. Dashwood, Caspase-8 and apoptosis-inducing factor mediate a cytochrome c independent pathway ofapoptosis in human colon cancer cells induced by the dietaryphytochemical chlorophyllin, Cancer Res. 63 (2003) 1254–1261.

[21] B.B. Aggarwal, A. Kumar, A.C. Bharti, Anticancer potential ofCurcumin: preclinical and clinical studies, Anticancer Res. 23(2003) 363–398.

[22] S. Somasundaram, N.A. Edmund, D.T. Moore, G.W. Small,Y.Y. Shi, R.Z. Orlowski, Dietary Curcumin inhibits chemother-apy-induced apoptosis in models of human breast cancer, CancerRes. 62 (2002) 3868–3875.

[23] M. Muller, S. Strand, H. Hug, E.M. Heinemann, H. Walczak,W.J. Hofmann, et al., Drug-induced apoptosis in hepatoma cellsis mediated by the CD95 (APO-1/Fas) receptor/ligand system andinvolves activation of wild-type p53, J. Clin. Invest. 99 (1997) 403–413.

[24] D. Morin, S. Barthelemy, R. Zini, S. Labidalle, J.P. Tillement,Curcumin induces the mitochondrial permeability transition poremediated by membrane protein thiol oxidation, FEBS Lett. 495(2001) 131–136.

[25] K. Piwocka, K. Zablocki, M.R. Wieckowski, J. Skierski, I. Feiga,J. Szopa, et al., A novel apoptosis-like pathway, independent ofmitochondria and caspases, induced by Curcumin in humanlymphoblastoid T (Jurkat) cells, Exp. Cell Res. 249 (1999) 299–307.

[26] J.C. Reed, Bcl-2 family proteins, Oncogene 17 (1998) 3225–3236.[27] R. Silvestrini, A. Costa, C. Lequaglie, C. Mochen, S. Veneroni,

M. Leutner, et al., Bcl-2 protein and prognosis in patients withpotentially curable non-small-cell lung cancer, Virchows Arch.432 (1998) 441–444.

[28] J. Yang, X. Liu, K. Bhalla, C.N. Kim, A.M. Ibrado, J. Cai,et al., Prevention of apoptosis by Bcl-2: release of cytochrome c

from mitochondria blocked, Science 275 (1997) 1129–1132.[29] R.J. Anto, A. Mukhopadhyay, K. Denning, B.B. Aggarwal,

Curcumin (diferuloylmethane) induces apoptosis through activa-tion of caspase-8, BID cleavage and cytochrome c release: itssuppression by ectopic expression of Bcl- 2 and Bcl-xl, Carcino-genesis 23 (2002) 143–150.

[30] J.A. Bush, K.J. Cheung Jr., G. Li, Curcumin induces apoptosis inhuman melanoma cells through a Fas receptor/caspase-8 pathwayindependent of p53, Exp. Cell Res. 271 (2001) 305–314.

[31] A. Klopfer, A. Hasenjager, C. Belka, K. Schulze-Osthoff, B.Dorken, P.T. Daniel, Adenine deoxynucleotides fludarabine andcladribine induce apoptosis in a CD95/Fas receptor, FADD andcaspase-8-independent manner by activation of the mitochondrialcell death pathway, Oncogene 23 (2004) 9408–9418.

[32] H. Ding, C. Han, J. Zhu, C.S. Chen, S.M. D�Ambrosio, Celecoxibderivatives induce apoptosis via the disruption of mitochondrial

1252 S. Sen et al. / Biochemical and Biophysical Research Communications 331 (2005) 1245–1252

membrane potential and activation of caspase 9, Int. J. Cancer 113(2005) 803–810.

[33] H. Safran, T. King, H. Choy, A. Gollerkeri, H. Kwakwa, F.Lopez, et al., p53 mutations do not predict response to paclitaxel/radiation for nonsmall cell lung carcinoma, Cancer 78 (1996)1203–1210.

[34] B.B. Aggarwal, Apoptosis and nuclear factor-kappa B: a tale ofassociation and dissociation, Biochem. Pharmacol. 60 (2000)1033–1039.

[35] D. Karunagaran, B.B. Aggarwal, Transcription factors as targetsfor the drug development, in: G. Keri, I. Toth (Eds.), MolecularPathomechanisms and New Trends in Drug Development, Har-wood Academic Publisher, New York, 2002, pp. 76–91.

[36] S.M. Plummer, K.A. Holloway, M.M. Manson, R.J. Munks,A. Kaptein, S. Farrow, et al., Inhibition of cyclo-oxygenase2 expression in colon cells by the chemopreventive agentCurcumin involves inhibition of NF-kappaB activation via theNIK/IKK signalling complex, Oncogene 18 (1999) 6013–6020.

[37] F. Zhang, N.K. Altorki, J.R. Mestre, K. Subbaramaiah, A.J.Dannenberg, Curcumin inhibits cyclooxygenase-2 transcription inbile acid- and phorbol ester-treated human gastrointestinalepithelial cells, Carcinogenesis 20 (1999) 445–451.

[38] C.S. Williams, M. Tsujii, J. Reese, S.K. Dey, R.N. DuBois, Hostcyclooxygenase-2 modulates carcinoma growth, J. Clin. Invest.105 (2000) 1589–1594.