at SciVerse ScienceDirect

Biochimie 95 (2013) 1567e1573

Contents lists available

Biochimie

journal homepage: www.elsevier .com/locate/biochi

Research paper

Augmented sensitivity to methotrexate by curcumin inducedoverexpression of folate receptor in KG-1 cells

Sugapriya Dhanasekaran, Bijesh K. Biswal 1, Venil N. Sumantran, Rama S. Verma*

Department of Biotechnology, Indian Institute of Technology Madras, Chennai, 600 036, TN, India

a r t i c l e i n f o

Article history:Received 21 August 2012Accepted 12 April 2013Available online 23 April 2013

Keywords:Folate receptorCurcuminMethotrexateKG-1 cellsRadioligand FR binding assay

* Corresponding author. 201, Department of BioteTechnology, Chennai 600 036, TN, India. Tel.: þ4422574102.

E-mail addresses: sughaphd@yahoo.com (S. Dhrediffmail.com (B.K. Biswal), venil.sumantran@gmvermars@iitm.ac.in (R.S. Verma).

1 Present address: Room 522, Department of PhDartmouth Medical School, 7650 Remsen, Hanover, N

0300-9084/$ e see front matter � 2013 Elsevier Mashttp://dx.doi.org/10.1016/j.biochi.2013.04.004

a b s t r a c t

Folate receptors are targets of various strategies aimed at efficient delivery of anti-cancer drugs. Folatereceptors also play a role in the uptake of antifolate drugs which are used for therapeutic intervention inleukemia. Therefore, it is important to identify compounds which regulate expression of folate receptorsin leukemic cells. The present study examined if curcumin could modulate the uptake and cytotoxicity ofthe antifolate drug methotrexate, in KG-1 leukemic cells. This is the first report to show that curcumin(10e50 mM) causes a significant, dose-dependent, 2e3 fold increase in uptake of radiolabelled folic acidand methotrexate into KG-1 cells both at 24 h and 48 h of treatment. Interestingly, pre-treatment of KG-1leukemic cells with curcumin (10 mM and 25 mM) also caused a statistically significant enhancement inthe cytotoxicity of methotrexate. We performed Real Time Quantitative RT-PCR to confirm the upregu-lation of FRb mRNA in curcumin treated cells. Immunocytochemistry and Western blotting showed thatcurcumin caused increased expression of folate receptor bin KG-1 cells. Our data show that the mech-anism of curcumin action involves up-regulation of folate receptor b mRNA and protein in KG-1 cells.Therefore, combination of non-toxic concentrations of curcumin and methotrexate, may be a viablestrategy for therapeutic intervention for leukemias using a folate receptor-targeted drug delivery system.

� 2013 Elsevier Masson SAS. All rights reserved.

1. Introduction

Methotrexate is a well-known antifolate drug which has beenwidely used as a chemotherapeutic agent for leukemias since theearly 1950s. However, side effects associated with methotrexatetoxicity limit its therapeutic efficacy [1]. The present study inves-tigated the role of curcumin on the cytotoxicity of methotrexate inKG-1 human leukemic tumor cells. The mechanism of curcumin’saction was further examined by analyzing effects of curcumin onthe expression and function of folate receptors in these cells. Folateenters human cells via two distinct transport systems, the reducedfolate carrier pathway (RFC) which has low affinity for folate, andthe folate receptor pathway which has high-affinity for folate. TheRFC is the predominant carrier used for folate uptake by normal

chnology, Indian Institute of91 4422574109; fax: þ91

anasekaran), bijeshbijesh3@ail.com (V.N. Sumantran),

armacology and Toxicology,H 03755, USA.

son SAS. All rights reserved.

cells. In contrast, rapidly dividing cancer cells over-express the highaffinity folate receptor (FR), which can internalize folic acid andconjugates of folate with equal efficiency [2]. Conjugates of folicacid taken up via folate receptor-mediated endocytosis, provide amechanism for targeted delivery to cells expressing FR [3]. Severalstudies have shown that folate receptor (FR), when expressed athigh levels, could offer the preferred uptake route of novel classesof antifolate drugs that target glycineamide ribonucleotide for-myltransferase and thymidylate synthase [4e6]. Two major formsof folate receptors (FRa and FRb) are expressed in cells [7]. Thus,FRa receptor is over-expressed in human tumors of epitheliallineage such as ovarian carcinomas, and is a useful marker forepithelial cancers [8]. FRb is a neutrophilic lineage marker which isdifferentially expressed in activated macrophages and myeloidleukemia [9]. Thus, FR-beta serves as a marker for myeloid leuke-mia and chronic inflammatory diseases such as rheumatoidarthritis (RA) [10].

The antifolate drug Methotrexate is a potent inhibitor of dihy-drofolate reductase (DHFR), an enzyme responsible for maintainingthe cellular pool of reduced folates. MTX can cause cell cycle arrestby inhibiting synthesis of dTMP and purine precursors for DNAsynthesis [11e13]. Cellular resistance to methotrexate (MTX) inleukemic cells is reported to be due to reduced MTX transport by

S. Dhanasekaran et al. / Biochimie 95 (2013) 1567e15731568

RFC [9,14]. However, folate receptors can play an important role inMTX resistance since these receptors can transport MTX in differentcell types. Thus, synovial macrophages of RA patients have highexpression of folate receptor b (FR-b) and the capacity for uptake ofMTX and FR-b specific drugs have been developed and character-ized for their therapeutic potency in synovial inflammation [10].Overexpression of folate receptors in MCF-7 breast cancer cells wasalso reported to result in increased sensitivity to antifolates [14].Clinical studies show that severe side-effects of drugs such as MTX,limits the application and therapeutic index of these antitumordrugs [15]. Therefore, it would be useful to find ways to increaseuptake of MTX via FRb mediated transport in leukemic tumor cells.Curcumin, a bioactive constituent of the rhizomes of Curcumalonga, possesses remarkable chemopreventive and anti-inflammatory properties [16e18]. The cytotoxic properties of cur-cumin have been well documented in cultured tumor cells derivedfrom various types of leukemias and lymphomas [19e21]. Inter-estingly, in vitro studies also suggest that curcumin enhanced theanti-proliferative and pro-apoptotic activities of anticancer agentssuch as bortezomib and bleomycin in multiple myeloma cells andhuman malignant testicular germ cells, respectively [22,23].

This study examined if curcumin couldmodulate the uptake andcytotoxicity of MTX in KG-1 leukemic cells. This is the first report toshow that curcumin enhances cytotoxicity of MTX via increasedMTX uptake in KG-1 cells. We show that the mechanism of cur-cumin action involves up-regulation of folate receptor bmRNA andprotein in KG-1 cells. These data can contribute toward formulationof an optimum dose of curcumin, which in combination withreduced methotrexate concentrations, could effectively kill tumorcells while reducing the cytoxicity associated with methotrexatetreatment in leukemic patients.

2. Materials and methods

2.1. Chemicals

Curcumin was purchased from Himedia Laboratories, India.Methotrexate and Anti-Rabbit IgG-FITC conjugated secondaryantibody were purchased from SigmaeAldrich Co., USA. FolateReceptor-Rabbit polyclonal primary antibody was purchased fromSanta Cruz Biotechnologies (Santa Cruz, CA). [30, 50, 7, 9-3H] Folicacid (FA) (35 Ci/mmol) and [30, 50, 7-3H (N)] methotrexate (MTX)(50.8 Ci/mmol) were purchased from Moravek Biochemicals (Brea,CA). Folate free RPMI 1640 medium, penicillin streptomycin, L-glutamine and FBS were obtained from GIBCO-BRL Life Technolo-gies Inc., USA. MTT 3-(4,5-Dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide was purchased from Sisco Research Labora-tories (SRL), India. All other chemicals used were extra pure, tissueculture grade, and analytical grade.

2.2. Cell culture and treatments

KG-1 is a human bone marrow-derived, acute myeloid leukemiccell line which expresses FRb. KG-1 cells were obtained from Na-tional Center for Cell Science (NCCS), Pune, India. Cells werecultured in Folate free RPMI 1640 media supplemented with 4 mML-Glutamine, 100 U/mL penicillinestreptomycin and 10% FBS at37 �C in 5% CO2. Curcumin was dissolved in DMSO at a concentra-tion of 100mM as a stock solution. The stock solutionwas diluted incell culture medium to the required concentration immediatelybefore use. The final DMSO concentration in themediumwas 0.10%,which was not cytotoxic to cells. 5 mM Methotrexate was freshlyprepared in 250 mM sodium carbonate buffer (pH 9.5) as a stocksolution and used immediately.

2.3. Cell viability studies

KG-1 cells (1 � 104/well) were seeded in 96-well micro-titerplate with 60% confluence. After initial-6 h culture, cells wereincubated for 12 h, 24 h, and 48 h; at 37 �C in 5% CO2 withvarious concentrations of curcumin (5 mMe100 mM) in folatefree RPMI 1640 medium, supplemented with 100 U/mL peni-cillin, 100 mg/mL streptomycin 4 mM L-glutamine and 10% FBS(which contains <1 nM folate as the only source of folate) [24].For treatment with MTX, cells were washed with low pH buffer(10 mM sodium acetate, pH 4, 150 mM NaCl, 7 mM glucose) at4 �C to remove membrane bound folate. Cells were thenwashed twice with PBS and treated with MTX (10 nMe100 mM)in folate free RPMI 1640 medium with 10% FBS for 24 h, 48 hand 72 h at 37 �C in 5% CO2. The experiments were performedin triplicates.

2.4. MTT assay

Cell survival studies were carried out with the MTT assay. Afterincubation of KG-1 cells with or without curcumin or MTX in serumcontaining, folate free media. 0.5 mg/mL of MTT in PBS was addedto each well and incubated at 37 �C for 4 h in the dark. The reactionwas stopped with 100 ml of lysis solution (20% SDS in 50% Dime-thylformamide). Absorbance was measured at 570 nm with back-ground subtraction at 650 nm using an ELISA plate reader (BioRadUSA). Percent viability was calculated as (Cell viability with testcompound/cell viability of control) � 100. Results are expressed aspercent viability of cells treated with curcumin or MTX relative tountreated controls.

2.5. Uptake of FA and MTX by radiolabelled ligand-binding assay

Cells (1 � 105/well) grown in suspension culture were centri-fuged (300� g) and washed with the HBS buffer (135 mM NaCl,5 mM KCl, 2.5 Mm CaCl2, 0.6 mM MgSO4, 6 mM glucose and10 mM Hepes, pH 7.4). For uptake studies in controls, 1 � 105 cellswere incubated with 10 pmol [3H] Folic acid or 10 pmol [3H]Methotrexate at 37 �C for 1 h. Uptake of [3H] FA and [3H] meth-otrexate was also measured in cells after treatment with curcumin(10e100 mM) for 24 h and 48 h. The uptake assay was performedas described [25], and results are expressed as fold increase inuptake of [3H] FA or [3H] MTX in curcumin treated cells relative tountreated controls.

2.6. Determination of FRb mRNA expression by RT-PCR

KG-1 cells (0.2 � 106/well) were seeded in 6-well plates. After a6 h pre-culture in RPMI 1640 with 10% FBS medium, cells weretreated with different concentrations of curcumin (0, 5, 10, 25, 50,100 mM) for 12 h, 24 h and 48 h. Total RNA was isolated by guani-dinium thiocyanateephenolechloroform extraction method asdescribed earlier [26]. Briefly, 2 mg of total RNA was reverse tran-scribed to cDNA using MuLV-Reverse transcriptase (Finzymes,USA), and PCR amplification was performed with cDNA. Primers ofFRb- Forward 50-ATG CCA CTT CTG CTG CTT CT-30 and Reverse 50-AGT GAC TCC AGA GGC CTT CA-30 (552 bp). Primers of GAPDH-Forward 50-GAA GGT GAAGGT CGGAGT-30 and Reverse 50-GAAGATGGT GAT GGG ATT TC-30 (188 bp) were used. PCR conditions wereas follows: 94 �C for 5 min, followed by 35 cycles of 94 �C for 45 s,60 �C for 45 s, and 72 �C for 45 s, followed by a final extension of72 �C for 5 min per product. Bands were visualized in 0.8% poly-acrylamide gels.

Fig. 1. Effects of curcumin and methotrexate on viability of KG-1 cells. Cytotoxic effectsof various concentrations of curcumin and methotrexate on KG-1 cells (a) curcumininduced cell death at 12 h, 24 h, and 48 h. (b) Methotrexate induced cell death (0e100 mM) at 72 h. IC50 of methotrexate was calculated as 50 nM. All values aremean � SE of three replicates.

Fig. 2. Effects of curcumin (10 mM and 25 mM) with and without methotrexate on cellviability at 12 h, 24 h, and 48 h. After curcumin treatment, cells were washed andtreated with methotrexate (50nM for 72 h). Results are expressed as percent cellviability relative to controls which are normalized to 100%. All values are Mean � SEof three replicates. Viability of cells treated with curcumin and methotrexate wassignificantly lower than viability of cells treated with curcumin alone [(Comparisonbetween 10 mM curcumin versus 10 mM curcumin with 50 nM MTX (*a), and 25 mMcurcumin versus 25 mM curcumin with 50 nMMTX (*b)]. The symbol (*c) indicates thatviability of cells treated with curcumin and methotrexate was also significantly lowerthan viability of cells treated with methotrexate alone. Symbol * represents statisticalsignificance at p < 0.01.

Fig. 3. Curcumin enhances uptake of [3H] folate and [3H] methotrexate in KG-1 cells.Uptake of [3H] FA and [3H] methotrexate was measured in untreated controls and incells treated with curcumin (10e100 mM) for 24 h and 48 h. Levels of uptake in controls

S. Dhanasekaran et al. / Biochimie 95 (2013) 1567e1573 1569

2.7. Determination of FRb protein expression byimmunocytochemistry

KG-1 cells (0.2 � 106/well) were seeded in 6-well plates for 6 hin RPMI 1640 with 10% FBS medium, and treated with curcumin (0,10, 25, 50, 100 mM) for 12 h, 24 h and 48 h. After incubation, cellswere harvested, washed with PBS and low pH buffer (10 mM so-dium acetate, pH 4, 150 mM NaCl, 7 mM glucose at 4 �C), to removemembrane bound folate. Cell suspensions were prepared as smearson poly lysine coated glass slides, and fixed with 4% para-formaldehyde. Cells were washed with PBS and incubated with 5%FBS for 1 h for blocking. After washing with PBS, cells were incu-bated with primary FR polyclonal antibody (1:100 dilution) (Santacruz Biotechnology (FL-257) sc-28997) for 2 h. Cells were thenwashed and incubated with secondary anti-rabbit IgG-Phycoery-thrin (IgG-PE) (1:200 dilution) for 30min. Cells weremounted with50% glycerol and FRb protein expression was visualized under thefluorescence microscope (Nikon Tie, Japan).

are normalized to a value of 1.0. Results are expressed as fold increase in uptake of [3H]FA or [3H] MTX relative to controls. Data represent Mean � SE for three independentexperiments. Symbol (*a) denotes 0 mM versus 10 mM curcumin. Symbol (*b) denotes0 mM versus 25 mM curcumin. Symbol (*c) denotes 0 mM versus 50 mM curcumin, andsymbol (*d) denotes 0 mM versus 100 mM curcumin. Symbol * represents statisticalsignificance at p < 0.01.

2.8. Real-time RT-PCR for gene expression analysis

We validate the RT-PCR Data with quantitative real-time RT-PCR. Primers of FRb- Forward 50-GAC TAC CTG CCC TCA GCT TG-30

and Reverse 50-GCA ACA GCA TGG GTG ACG A-30 (182 bp). Primersof b-Actin Forward 50-AGC GAG CAT CCC CCA AAG TT-30 and Reverse50-GGG CAC GAA GGC TCA TCA TT-30 (280 bp) were used. TheComparative CT Method (2DDCT) was used to determine the rela-tive quantification of target genes, normalized to a reference gene(b-Actin) and relative to a calibrator (fourth passage of differenti-ated MSCs) [27]. A validation experiment was performed to verifythat the target efficiencies and referencewere approximately equal.

Fig. 4. Curcumin up-regulates FRb mRNA expression in KG-1 cells. The figure showsthe FRb mRNA expression profile in KG-1 cells incubated with curcumin for 12 h, 24 h,and 48 h. PCR products of FRb and GAPDH were of 522 kb and 188 kb respectively.Lanes 1e6 depicts levels of FRb mRNA expression in response to curcumin (0, 5, 10, 25,50,100 mM) respectively, at 12 h, 24 h, and 48 h respectively.

S. Dhanasekaran et al. / Biochimie 95 (2013) 1567e15731570

3. Western blot analyses

After treatment, cells were harvested and cell lysates weremadein a cell lysis buffer (50mMTrisHCl,150mMNaCl, pH 8.0),with 0.1%Triton X, 0.01 mg/mL protinin and 0.05 mg/mL phenyl-methylsulfonyl fluoride. The protein content was quantified usingPierce BCA protein assay Kit (Prod#23225-Thermo Scientific, U.S.A).An equal amount of the protein (30 mg/sample) was electropho-retically separated by 12% sodium dodecyl sulfatepolyacrylamidegel electrophoresis. Proteinswere transferred onto PVDFmembranein a Bio-Rad Trans blot apparatus. PVDF matrices were preblockedwith 3%BSA inTBS and0.05%Tween-20 for 1h at roomtemperature.Following TBS-Tweenwashes, preblocked matrices were incubatedwith the anti-rabbit FR polyclonal primary antibody (Santa cruzBiotechnology (FL-257) sc-28997) and anti-mouse b-Actin (Internalcontrol). Appropriate conjugated secondary antibodies were thenadded and color developed with ECL determination.

Fig. 5. Curcumin increases expression of FRb protein in KG-1 cells. Fluorescent images ofantibody, are shown. Panel (A: aee) shows FRb protein expression after 12 h exposure to curafter 24 h treatment with curcumin (0, 10, 25, 50,100 mM), respectively. Panel (C: aee) shorespectively.

4. Results

4.1. Effects of curcumin and methotrexate on viability of KG-1 cells

Effects of curcumin and methotrexate on the viability of KG-1cells were determined in order to calculate the IC50 of both com-pounds. As shown in Fig. 1a, curcumin caused a dose-dependentdecrease in viability of KG-1 cells at different time points. Thus,KG-1 cells exposed to 50 mMcurcumin for 24 h showed a 43% in lossof cell viability. Longer exposure to lower concentrations of cur-cumin (25 mM for 48 h) also inhibited KG-1 cell viability by 33%.These IC50 values of curcumin (25e50 mM) are consistent withvalues reported for anti-proliferative effects of curcumin inleukemic cells [28]. An exposure to 100 mM curcumin causedmaximum cytotoxic effects of curcumin with 38% of loss in cellviability at 12 h, 60% of loss in cell viability at 24 h, and 80% loss ofviability at 48 h in KG-1 cells. Upon treatment with methotrexatefor 24 h and 48 h, there was no significant cell death (data notshown). However, longer exposure of KG-1 cells to methotrexate(50 nM for 72 h) resulted in a 50% decrease in cell viability (Fig. 1b).This concentration of methotrexate (50 nM) was fixed as IC50 forsubsequent experiments. Fig. 1b also shows that KG-1 cells exposedto 100 mM MTX for 72 h showed a maximal 75% loss of viability.Thus, long-term exposure to high concentrations of MTX has adose-dependent, cytotoxic effect on KG-1 cells.

As mentioned above, KG-1 cells were not sensitive to MTX priorto the 72 h time point (Fig. 1b). Therefore, we examined whetherpre-treatment of KG-1 cells with non toxic concentrations of cur-cuminwould alter the cytotoxicity of MTX in these cells. Thus, KG-1cells were treated with curcumin (10 mM and 25 mM for 24 h or48 h). After removal of curcumin containing media, cells weretreated with MTX (50 nM for 72 h), and cell viability was measured.Fig. 2 shows that 10 mM and 25 mM curcumin alone caused 13% and21% loss of cell viability respectively, at 24 h. However, cells pre-treated with 10 mM and 25 mM curcumin for 24 h, followed by MTXtreatment; showed a statistically significant 68% and 77% loss of cellviability respectively, when compared with cells treated with10 mM and 25 mM curcumin alone (p< 0.01). At the 48 h time point,cells treated with 10 mM and 25 mM curcumin treatment alone,showed 21% and 33% loss of cell viability respectively. However,

KG-1 cells stained with FR antibody and FITC-conjugated anti rabbit IgG secondarycumin (0, 10, 25, 50,100 mM), respectively. Panel (B: aee) shows FRb protein expressionws FRb protein expression after 48 h treatment with curcumin (0, 10, 25, 50,100 mM),

Fig. 7. Curcumin augments the expression of FRb protein in KG-1 cells. Western blotanalysis showing expression of FRb in KG-1 cells before and after treatment withcurcumin for 24 in various concentrations (0, 10, 25, 50,100 mM). Curcumin upregulatedFRb expression after 24 h treatment to compared with the control cells. b-Actin used asan internal control.

Fig. 6. Curcumin increases the mRNA expression of FRb in KG-1 cells. Figure shows theFRb mRNA expression profile in KG-1 cells incubated with curcumin for 24 h, and 48 h.FRb and b-Actin were examined by real time RT-PCR. The graph depicts the foldchanges of FRbmRNA transcript expression in response to curcumin concentrations (0,5, 10, 25, 50,100 mM) in 24 h and 48 h respectively when compared with control(untreated) KG-1 cells. Relative quantification was done with b-Actin expression andvalues were normalized.

S. Dhanasekaran et al. / Biochimie 95 (2013) 1567e1573 1571

cells pre-treated with 10 mM and 25 mM curcumin for 48 h followedby MTX treatment, showed a statistically significant 78% and 84%loss of cell viability respectively, when compared with cells treatedwith 10 mM and 25 mM curcumin alone (p < 0.01). Thus, Fig. 2clearly demonstrates that pre-treatment of KG-1 leukemic cellswith non-toxic concentrations of curcumin (10 mM and 25 mMcurcumin) caused a statistically significant enhancement in thecytotoxic effects of MTX.

4.2. Effects of curcumin on uptake of [3H] FA and [3H] MTX in KG-1cells

The enhanced cytotoxicity of MTX in curcumin treated cells(Fig. 2) could be due to increased uptake of MTX into KG-1 cells.Therefore, we examined whether curcumin could modulate thetransport and uptake of FA and MTX by KG-1 leukemic cells. Fig. 3shows that non-toxic concentrations of curcumin (25 mM) for 24 h,induced a significant 1.37 and 1.87 fold increase in uptake of FA andMTX uptake respectively (p < 0.01). Higher concentrations of cur-cumin (50 mM) for 24 h, also induced a statistically significant 2.31and 2.78 fold increase in uptake of FA and MTX respectively(p < 0.01). A maximum 3.40 fold increase in MTX uptake wasobserved in KG-1 cells treated with 100 mM curcumin for 24 h. Astatistically significant augmentation of uptake of radiolabelledmethotrexate was also observed in KG-1 cells treated with curcu-min (25 mMe100 mM) for 48 h. Thus, Fig. 3 clearly demonstratesthat pre-treatment of KG-1 leukemic cells with non-toxic concen-trations of curcumin (10 mM and 25 mM) caused a significant in-crease in the uptake of MTX by these cells.

4.3. Effect of curcumin on expression of FR-b mRNA and protein

Having shown that curcumin caused a dose-dependent andstatistically significant increase in cytotoxicity (Fig. 2), and uptakeof methotrexate into KG-1 leukemic cells (Fig. 3); we investigatedthe mechanism of curcumin’s actions by examining the effects ofcurcumin on expression of folate receptor b. Fig. 4 shows thatcurcumin caused a significant, dose-dependent increase in FRbmRNA expression at 12 h and 24 h. Therewas maximum FRbmRNAexpression at 24 h and 48 h in KG-1 cells exposed to 10 mM and25 mM curcumin when compared to internal control GAPDH. Asexpected, higher, cytotoxic concentrations of curcumin (50e100 mM) reduced expression of FRb mRNA at 48 h. Fig. 5 displaysexpression of FRb protein in KG-1 cells after exposure to increasingconcentrations of curcumin (10e25 mM) for 12 h, 24 h, and 48 h.When compared with controls, curcumin induced a dose-dependent increase in levels of FRb expression in KG-1 cells.Maximal FRb expression was achieved when KG-1 cells weretreated with low concentrations of curcumin (10 mMe25 mM) for12 h and 24 h. Thus, KG-1 leukemic cells exposed to non toxicconcentrations of curcumin for 12 h and 24 h, show a dose-dependent increase in expression of FRb mRNA (Fig. 4) and FRbprotein (Fig. 5).Curcumin’s ability to induce expression of FRbmRNA and FRb protein is consistent with the enhanced uptake of FAand MTX in curcumin treated cells (Fig. 3).

4.4. Upregulation of FRb in mRNA transcript (real time PCR) in KG-1cells

In order to confirm the results in Figs. 4 and 5, we examined theeffects of curcumin on FRb mRNA transcript level by quantitativereal-time PCR analyses. Compared to control, the curcumin treatedcells exhibited a significant, dose-dependent increase expression inFRb mRNA expression at 24 h and 48 h (Fig. 6). Notably, expressionof FRb, was enhanced at 25.7-fold (10 mM) and 9.6-fold (25 mM) at

24 h. At 48 h, Fig. 6 shows an average 3.5-fold increase in FRbmRNAlevels in KG-1 cells exposed to 5 mM and 10 mM curcumin compa-rable to internal control b-Actin. Maximal FRb expression wasachieved when KG-1 cells were treated with low concentrations ofcurcumin (10 mMe25 mM) in 24 h and (5 mMe10 mM) in 48 h. Thus,KG-1 leukemic cells exposed to non toxic concentrations of cur-cumin for 24 h and 48 h, show a dose-dependent increase inexpression of FRb mRNA.

4.5. Enhancement of FRb protein (western blot) in KG-1 cells

The immunocytochemistry data in Fig. 5 showed that curcumincaused a significant, dose-dependent increase in the levels of FRbprotein levels at 24 h. In order to verify these results, we assessedthe effects of curcumin on levels of FRb protein by western blotanalysis. Thus, Fig. 7 showed that treatment of KG-1 cells with 10,25 and 50 mM curcumin, caused a significant, dose-dependent in-crease in expression of FRb protein relative to the b-Actin control.These results confirm our immunocytochemistry data, and clearlyshow that non-toxic concentrations of curcumin increase theexpression of FRb protein in KG-1 cells.

Together, the data in Figs. 4e7 prove that upregulation of theFolate B receptor FRb by curcumin, provides a mechanism for theenhanced uptake (Fig. 3) and cytotoxicity of MTX in KG-1 cells(Fig. 2). Based on these observations, we hypothesize that admin-istering optimal concentrations of curcumin in combination with

Fig. 8. Proposed mechanism of curcumin action in KG-1 cells. Schematic representations show that curcumin enhances uptake and cytotoxicity of methotrexate by augmentinglevels of FRb in KG-1 cells. (Methotrexate (MTX) and b-folate receptor (FRb)).

S. Dhanasekaran et al. / Biochimie 95 (2013) 1567e15731572

low-dose of MTX, could be a promising strategy for treatment ofleukemia, because it could potentially reduce the genotoxic effectsassociated with MTX treatment (Fig. 8).

5. Discussion

Curcumin is well known for its antioxidant, anti-inflammatory[29], and anticarcinogenic properties [30,31]. There are manystudies showing that curcumin modulates signaling pathways,enzymes, and transcription factors, which promote proliferation ofcancer cells [32]. However, there are few reports on the effects ofcurcumin on receptors and transporters, which regulate tumor cellgrowth. Thus, curcumin was shown to inhibit growth of humancolon and bladder cancer cells by suppressing expression of theEGFR gene [33,34]. Few compounds have been reported to modu-late expression of folate receptors. Thus, Wang et al. showed thatFR-beta expression was elevated 20-fold by all-trans retinoic acid(ATRA) in KG-1 leukemic cells in a dose dependent and reversiblemanner in the absence of cell growth inhibition [35]. Our study issignificant because it is the first report to show that non-toxicconcentrations of curcumin induce expression of folate receptor bin KG-1 leukemic cells (Figs. 4e7). Notably, curcumin alsoaugmented folate receptor b function as a transporter for radio-labelled folic acid and methotrexate in these cells (Fig. 3). We alsofound that pre-treatment of KG-1 leukemic cells with non-toxicconcentrations of curcumin enhanced the cytotoxicity of Metho-trexate, an important antifolate chemotherapeutic drug in thesecells (Fig. 2).

6. Conclusion

This is the first report to show that curcumin caused a signifi-cant, dose-dependent, 2e3 fold increase in uptake of radiolabelledfolic acid andmethotrexate into KG-1 leukemic cells. Pre-treatmentof KG-1 cells with non-toxic concentrations of curcumin caused astatistically significant increase in the cytotoxicity of methotrexatein these cells. Our data also show that the mechanism of curcuminaction involves up-regulation of folate receptor b mRNA and FRbprotein in KG-1 leukemic cells.

Acknowledgment

Part of this work is supported by the grant no: SR/WOS-A/LS-205/2009.

References

[1] H. Zachariae, Methotrexate side-effects, Br. J. Dermatol. 36 (1990) 127e133.[2] B.A. Chabner, T.G. Roberts Jr., Chemotherapy and the war on cancer, Nat. Rev.

Cancer 5 (2005) 65e72.[3] X. Zhao, H. Li, R.J. Lee, Targeted drug delivery via folate receptors, Expert Opin.

Drug Deliv. 3 (2008) 309e319.[4] G. Jansen, G.R. Westerhof, I. Kathmann, G. Rijksen, J.H. Schornagel, Growth-

inhibitory effects of 5,10-dideazatetrahydrofolic acid on variant murine L1210and human CCRF-CEM leukemia cells with different membrane-transportcharacteristics for (anti)folate compounds, Cancer Chemother. Pharmacol. 28(1991) 115e117.

[5] G.R. Westerhof, G. Jansen, N. Van Emmerik, I. Kathmann, G. Rijksen,A.L. Jackman, J.H. Schornagel, Membrane transport of natural folates andantifolate compounds in murine L1210 leukemia cells: role of carrier- andreceptor mediated transport systems, Cancer Res. 51 (1991) 5507e5513.

[6] G.R. Westerhof, J.H. Schornagel, I. Kathmann, A.L. Jackman, A. Rosowsky,R.A. Forsch, J.B. Hynes, F.T. Boyle, G.J. Peters, H.M. Pinedo, G. Jansen, Carrierand receptor-mediated transport of folate antagonists targeting folate-dependent enzymes: correlates of molecular-structure and biological activ-ity, Mol. Pharmacol. 48 (1995) 459e471.

[7] W. Yan, M. Ratnam, Preferred sites of glycosylphosphatidylinositol modifica-tion in folate receptors and constraints in the primary structure of the hy-drophobic portion of the signal, Biochemistry 34 (1995) 14594e14600.

[8] K.R. Kalli, A.L. Oberg, G.L. Keeney, et al., Folate receptor alpha as a tumor targetin epithelial ovarian cancer, Gynecol. Oncol. 108 (2008) 619e626.

[9] J.F. Ross, H. Wang, F.G. Behm, et al., Folate receptor type beta is a neutrophiliclineage marker and is differentially expressed in myeloid leukemia, Cancer 85(1999) 348e357.

[10] N. Nakashima-Matsushita, T. Tomma, S. Yu, T. Matsuda, N. Sunahara,T. Nakamura, et al., Selective expression of folate receptor-beta and itspossible role in methotrexate transport in synovial macrophages from pa-tients with rheumatoid arthritis, Arthritis Rheum. 42 (1999) 1609e1616.

[11] J.J. McGuire, Anticancer antifolates: current status and future directions, Curr.Pharm. Des. 9 (2003) 2593e2613.

[12] G.J. Peters, G. Jansen, Resistance to antimetabolites, in: R.L. Schilsky,G.A. Milano, M.J. Ratain (Eds.), Principles of Antineoplastic Drug Developmentand Pharmacology, Marcel Dekker, New York, 1996, pp. 543e585.

[13] M. Tsurusawa, M. Niwa, N. Katano, T. Fujimoto, Flow cytometric analysisby bromodeoxyuridine/DNA assay of cell cycle perturbation ofmethotrexate-treated mouse L1210 leukemia cells, Cancer Res. 48 (1988)4288e4293.

[14] K.N. Chung, Y. Saikawa, T.H. Paik, K.H. Dixon, T. Mulligan, K.H. Cowan,P.C. Elwood, Stable transfectants of human MCF-7 breast cancer cells withincreased levels of the human folate receptor exhibit an increased sensitivityto antifolates, J. Clin. Invest. 91 (1993) 1289e1294.

[15] L.R. Espinoza, L. Zakraoui, C.G. Espinoza, F. Gutierrez, L.J. Jara, L.H. Silveira,M.L. Cuellar, P. Martinez-Osuna, Psoriatic arthritis: clinical response and sideeffects to methotrexate therapy, J. Rheumatol. 19 (1992) 872e877.

[16] Duvoix, R. Blasius, S. Delhalle, M. Schnekenburger, F. Morceau, E. Henry,M. Dicato, M. Diederich, Chemopreventive and therapeutic effects of curcu-min, Cancer Lett. 223 (2005) 181e190.

[17] D. Karunagaran, R. Rashmi, T.R. Kumar, Induction of apoptosis by curcuminand its implications for cancer therapy, Curr. Cancer Drug Targets 5 (2005)117e129.

[18] B.B. Aggarwal, A. Kumar, A.C. Bharti, Anticancer potential of curcumin: pre-clinical and clinical studies, Anticancer Res. 23 (2003) 363e398.

S. Dhanasekaran et al. / Biochimie 95 (2013) 1567e1573 1573

[19] S. Shishodia, H.M. Amin, R. Lai, B.B. Aggarwal, Curcumin (diferuloylmethane)inhibits constitutive NF-kB activation, induces G1/S arrest, suppresses pro-liferation, and induces apoptosis in mantle cell lymphoma, Biochem. Phar-macol. 70 (2005) 700e713.

[20] Duvoix, F. Morceau, M. Schnekenburger, S. Delhalle, M.M. Galteau, M. Dicato,M. Diederich, Curcumin-induced cell death in two leukemia cell lines: K562and Jurkat, Ann. N. Y. Acad. Sci. 1010 (2003) 389e392.

[21] S. Uddin, A.R. Hussain, P.S. Manogaran, K. Al-Hussein, L.C. Platanias,M.I. Gutierrez, K.G. Bhatia, Curcumin suppresses growth and inducesapoptosis in primary effusion lymphoma, Oncogene 24 (2005) 7022e7030.

[22] J. Park, V. Ayyappan, E.K. Bae, C. Lee, B.S. Kim, B.K. Kim, Y.Y. Lee, K.S. Ahn,S.S. Yoon, Curcumin in combination with bortezomib synergisticallyinduced apoptosis in human multiple myeloma U266 cells, Mol. Oncol. 4(2008) 317e326.

[23] A. Cort, M. Timur, E. Ozdemir, E. Kucuksayan, T. Ozben, Synergistic anticanceractivity of curcumin and bleomycin: an in vitro study using human malignanttesticular germ cells, Mol. Med. Rep. 6 (2012) 1481e1486.

[24] H. Matsue, K.G. Rothberg, A. Takashima, B.A. Kamen, Folate receptor allowscells to grow in low concentrations of 5-methyltetrahydrofolate, Proc. Natl.Acad. Sci. U S A 89 (1992) 6006e6009.

[25] B.K. Biswal, R.S. Verma, Differential usage of the transport systems for folicacid and methotrexate in normal human T-lymphocytes and leukemic cells,J. Biochem. 146 (2009) 693e703.

[26] P. Chomczynski, N. Sacchi, Single-step method of RNA isolation by acidguanidiniumthiocyanate-phenol-chloroform extraction, Anal. Biochem. 162(1987) 156e159.

[27] J.S. Yuan, A. Reed, F. Chen, C.N. Stewart Jr., Statistical analysis of real-time PCRdata, BMC Bioinform. 7 (2006) 85, http://dx.doi.org/10.1186/1471-2105-7-85.

[28] B.M. William, A. Goodrich, C. Peng, S. Li, Curcumin inhibits proliferation andinduces apoptosis of leukemic cells expressing wild-type or T315I-BCR-ABLand prolongs survival of mice with acute lymphoblastic leukemia, Hematol-ogy 6 (2008) 333e343.

[29] R. Motterlini, R. Foresti, R. Bassi, C.J. Green, Curcumin, an antioxidant and anti-inflammatory agent, induces heme oxygenase-1 and protects endothelial cellsagainst oxidative stress, Free Rad. Biol. Med. 28 (2000) 1303e1312.

[30] C.V. Rao, A. Rivenson, B. Simi, B.S. Reddy, Chemoprevention of colon carci-nogenesis by dietary curcumin a naturally occurring plant phenolic com-pound, Cancer Res. 55 (1995) 259e266.

[31] H.P. Ammon, M.A. Wahl, Pharmacology of Curcuma longa, Planta Med. 57(1991) 1e7.

[32] J.K. Lin, Molecular targets of curcumin, Adv. Exp. Med. Biol. 595 (2007)227e243.

[33] A. Chen, J. Xu, A.C. Johnson, Curcumin inhibits human colon cancer cell growthby suppressing gene expression of epidermal growth factor receptor throughreducing the activity of the transcription factor Egr-1, Oncogene 25 (2006)278e287.

[34] G. Chadalapaka, I. Jutooru, R. Burghardt, S. Safe, Drugs that target specificityproteins downregulate epidermal growth factor receptor in bladder cancercells, Mol. Cancer Res. 5 (2010) 739e750.

[35] H. Wang, X. Zheng, F.G. Behm, M. Ratnam, Differentiation-independent reti-noid induction of folate receptor type beta, a potential tumor target inmyeloid leukemia, Blood 96 (2000) 3529e3536.