Simultaneous Targeting of COX-2 and AKT Using Selenocoxib ... · for cancer cells. Selenocoxib-1-GSH reduced development of xenografted tumor by approximately 70% with negligible

Chemical Therapeutics Discovery & Preclinical Development

Simultaneous Targeting of COX-2 and AKT UsingSelenocoxib-1-GSH to Inhibit Melanoma

Raghavendra Gowda1,5,6,7, SubbaRao V. Madhunapantula1,5,6,7, Dhimant Desai1, Shantu Amin1, andGavin P. Robertson1,2,3,4,5,6,7

AbstractMelanoma is a highly metastatic and deadly disease. An agent simultaneously targeting the COX-2, PI3K/

Akt, and mitogen-activated protein kinase (MAPK) signaling pathways that are deregulated in up to 70% of

sporadic melanomas might be an effective treatment, but no agent of this type exists. To develop a single drug

inhibiting COX-2 and PI3K/Akt signaling (and increasing MAPK pathway activity to inhibitory levels as a

result of Akt inhibition), a selenium-containing glutathione (GSH) analogue of celecoxib, called selenocoxib-1-

GSH was synthesized. It killed melanoma cells with an average IC50 of 7.66 mmol/L compared with control

celecoxib at 55.6 mmol/L. The IC50 range for normal cells was 36.3 to 41.2 mmol/L compared with 7.66 mmol/L

for cancer cells. Selenocoxib-1-GSH reduced development of xenografted tumor by approximately 70% with

negligible toxicity by targeting COX-2, like celecoxib, and having novel inhibitory properties by acting as a

PI3K/Akt inhibitor (andMAPKpathway activator to inhibitory levels due toAkt inhibition). The consequence

of this inhibitory activity was an approximately 80% decrease in cultured cell proliferation and an approx-

imately 200% increase in apoptosis following 24-hour treatment with 15.5 mmol/L of drug. Thus, this study

details the development of selenocoxib-1-GSH,which is a nontoxic agent that targets the COX-2 and PI3K/Akt

signaling pathways in melanomas to inhibit tumor development. Mol Cancer Ther; 12(1); 3–15. �2012 AACR.

IntroductionMelanoma remains one of the most invasive and drug-

resistant cancers, making the development of clinicallyeffective therapies a major obstacle (1). Recent U.S. Foodand Drug Administration approval of vemurafenib(PLX-4032) illustrates the drug resistance hurdle facedby melanoma drugs inhibiting single targets. Vemurafe-nib targets mutant V600EB-Raf present in 50% to 60% ofsporadic melanomas, and has response rates of up to80% (2, 3). However, almost all initially respondingpatients developed recurrent resistant disease within ayear (4, 5). Therefore, drugs are needed that can be addedto the current arsenal of compounds for use alone or incombination with agents such as vemurafenib (6). Oneapproach is to improve the therapeutic efficacy of existingagents through chemical modification, enabling them to

target multiple key pathways regulating cancer develop-ment (7, 8).

Celecoxib is a drug that inhibits COX-2 activity (9).COX-2 is a ubiquitously expressed inducible enzyme thatplays an important role in theproduction ofprostaglandinE2 (PGE2; refs. 10, 11). Celecoxib inhibits COX-2, therebyreducing the production of PGE2 (9). PGE2 affects cellularproliferation, motility, invasiveness, angiogenesis, andpromotes survival by inhibiting apoptosis (10, 11). Fur-thermore, PGE2 is a tumor-inducing eicosanoid that pro-motes tumor development and progression, leading tomore invasive disease (12). COX-2 is overexpressed incarcinomas of the colon, breast, lung, prostate, cervix,stomach, andmelanocyte suggesting it could be an impor-tant therapeutic target (13, 14). Initial studies in this reportconfirm that COX-2 expression is elevated in melanomacell lines and in tumor biopsies compared with normalhuman melanocytes and that targeting COX-2 but notCOX-1 in melanoma cell lines using siRNA inhibitedxenografted melanoma tumor development.

Concentrations of celecoxib required to induce apopto-sis of cultured cells are high and ranged from 25 to 100mmol/L and clinical use is associated with cardiovascularside effects at doses of 200 mg per day (15, 16). Therefore,scientists are developing a variety of celecoxib analoguesthat are effective at lower concentrations or have alteredproperties (17, 18). Interestingly, some analoguesmaintainCOX-2 inhibitory potency, whereas others do not, but allseem to decrease the viability of cancer cells in culture tovarying degrees depending on the new properties of the

Authors' Affiliations: Departments of 1Pharmacology, 2Pathology, 3Der-matology, and 4Surgery; 5Penn State Melanoma Center; 6Penn StateMelanoma Therapeutics Program; and 7Foreman Foundation for Melano-maResearch, Pennsylvania State University College ofMedicine, Hershey,Pennsylvania

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Author: Gavin P. Robertson, Department of Pharmacol-ogy, Pennsylvania State College of Medicine, 500 University Drive, Her-shey, PA 17033. Phone: 717-531-8098; Fax: 717-531-0480; E-mail:[email protected].

doi: 10.1158/1535-7163.MCT-12-0492

�2012 American Association for Cancer Research.

MolecularCancer

Therapeutics

www.aacrjournals.org 3

on April 23, 2020. © 2013 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst October 30, 2012; DOI: 10.1158/1535-7163.MCT-12-0492

http://mct.aacrjournals.org/

agents (19–21). Several reports document that seleniumincorporation into the structural backbone of certain com-pounds can enhance the therapeutic potential of the drugby providing it with new inhibitory properties, whichinvolve inhibition of the Akt signaling pathway (22, 23).The Akt pathway is important in melanoma developmentand is activated in up to 70% of sporadic melanomas (24).Therefore, selenium incorporation into celecoxib wouldhave potential to enhance its therapeutic efficacy by pro-viding additional inhibitory properties unrelated to itsinitial COX-2 targeting efficacy. An analogue of celecoxibhas been developed that contains selenium, called seleno-coxib-1, but the drug is toxic to animals, limiting its use as atherapeutic agent (19). Therefore, additionalmodificationsof selenocoxib-1 were needed to maintain cancer cellkilling efficacy, with decreased toxicity of normal cells.

The novel discovery detailed in this report is that a formof selenocoxib-1 called selenocoxib-1-GSH was devel-oped, which does not exhibit the same toxicity on normalcells as selenocoxib-1, making it a potentially useful ther-apeutic agent. Selenocoxib-1-GSH more effectively killedmelanoma cell lines than celecoxib. The new agentretained COX-2 targeting activity and, as predicted, hadnew Akt signaling inhibitory properties. Mechanistically,selenocoxib-1-GSH inhibited melanoma cell survival bytargeting the COX-2 and PI3K/Akt pathways (andincreased pErk1/2 to inhibitory levels due to targetingof the Akt pathway), which decreased cellular prolifera-tion and triggered apoptosis mediated through a G0–G1

block, resulting in fewer cells in S and G2–M phases of thecell cycle. Intraperitoneal administration of selenocoxib-1-GSH retarded the growth of xenografted melanomatumors up to 70% without affecting animal body weightormajor organ functions. Thus, amore effective agent hasbeen developed from a toxic agent that can decreasemelanoma development by targeting key signaling path-ways without causing major organ-related toxicity.

Materials and MethodsCell lines and culture conditions

Human primary melanocytes containing wild-type B-Raf-FOM103 and NHEM 558, and human melanoma celllines harboring mutant V600EB-Raf - WM35, WM115,WM278.1, A375M, and 1205 Lu (provided by Dr. Herlyn;Wistar Institute, Philadelphia, PA) were cultured asdescribed (25). Human fibroblast FF2441 cells were pro-videdbyDr. CraigMyers (Penn StateCollege ofMedicine,Hershey, PA) and metastatic melanoma cell lines UACC903 (V600EB-Raf) were provided by Dr. Mark Nelson (Uni-versity of Arizona, Tucson, AZ). Wild-type B-Raf–con-taining C8161.Cl9 (provided by Dr. Danny Welch, Uni-versity of Kansas, Kansas City, MO) and MelJuSo (pro-vided by Dr. Judith Johnson, Institute for Immunology,Heidelberg, Germany) cell lines were maintained in Dul-becco’s Modified Eagle Medium (DMEM) supplementedwith 10% FBS. Cell lines were authenticated and main-tained in a 37�Chumidified5%CO2atmosphere incubator

and periodically monitored for genotypic characteristicsand tumorigenic potential to confirm cell line identity andphenotypic behavior.

Analysis of human melanoma patient tumorsTumors were pulverized using a mortar and pestle

chilled in liquid nitrogen and protein lysates extracted asreported previously (26). Western blotting was used tomeasure levels of COX-2 protein, normalized toa-enolaseusing ImageJ software.

siRNA efficacy and knockdown studiesTo determine efficacy of siRNA-mediated knockdown,

200pmol of siCOX-2 #1 or siCOX-2 #2,was comparedwithscrambled siRNA or reconstitution buffer followingnucleofection into 1� 106 of 1205 Lu or A375M cells usingan Amaxa nucleofector with solution R/program K-17(1205 Lu) or solution R/program A-23 (A375M). Trans-fection efficiency of viable cells was more than 90%.Following siRNA transfection, cells were reseeded andleft to recover for 2 days followed by replating in 96-wellplates to measure cell viability using the MTS assay(Promega). To show siRNA-mediated protein knock-down in vitro, 1 � 106 of 1205 Lu, UACC 903, andA375M cells were similarly nucleofected with 200 pmolof siCOX-2 #1, siCOX-2 #2, and100pmol/Lof V600EB-RAF,MEK1,MEK2, ERK1, andERK2, scrambled siRNA, recon-stitution buffer, and protein lysateswere harvested at day4 or 6, and analyzed by Western blot analysis. DuplexedStealth siRNA(Invitrogen)wasused for these studies. Thefollowing siRNA sequences were used: COX-2 #1: UCCAGACAAGCAGGCUAAUACUGAU;COX-2 #2:GAGUUA UGU CUU GAC AUC CAG AUC A. siRNAsequences for scrambled V600EB-RAF, MEK1, MEK2,ERK1, and ERK2 were used as previously reported (27).

COX inhibition studiesHuman recombinant COX-2 activity was assayed using

a commercial COX-inhibitor screening assay kit (CaymanChemical) according to the manufacturer’s protocol. Theconcentrations of celecoxib and selenocoxib-1-GSH testedwere 0.2, 2.0, and 20 nmol/L. SC-560 and DuP-697, stan-dard inhibitors for COX-1 and COX-2, respectively, wereused as positive controls. Dimethyl sulfoxide (DMSO)served as a negative control for 100% activity. The assaywas conducted in duplicate and repeated twice.

Synthesis of celecoxib, selenocoxib-1, andselenocoxib-1-GSH

Celecoxib was synthesized as described previously (28).Selenocoxib-1 was prepared as reported (19). Selenocoxib-1-GSH conjugate was prepared by reacting molar equiva-lent selenocoxib-1 with glutathione (GSH) in tetrahydrofu-ran: H2O (2:1) mixture. pH was adjusted to slightly basicconditions to generate selenocoxib-1-GSH conjugate in aquantitative yield as a yellow powder [mp:196–198�C; 1HNMR (DMSO-d6, 500 MHz) d 1.72–1.83 (m, 3H), 1.88–1.98(m, 1H), 2.23 -2.36 (m, 2H), 2.91 and 2.94 (dd, 1H, J¼ 10Hz),

Gowda et al.

Mol Cancer Ther; 12(1) January 2013 Molecular Cancer Therapeutics4




3.16and3.18 (dd, 1H, J¼ 4.5Hz), 3.57–3.72 (m, 3H), 4.18and4.23 (dd, 2H, J¼ 12.5Hzand22Hz), 4.51 (td,1H, J¼ 4.0Hz),6.62 (s, 1H, CH), 7.25 (d, 1H, aromatic, J ¼ 2.5 Hz), 7.26(d, 1H, aromatic, J ¼ 3.5 Hz), 7.35–7.39 (m, 3H, aromatic),7.41 (dt, 2H, aromatic, J ¼ 8.5 Hz and 2.0 Hz), 7.79 (dt,2H, aromatic, J¼ 8.5 Hz and 2.0 Hz), 8.44 (d, 1H, J¼ 7Hz),8.75 (ds, 1H); MS (M/Z, Intensity): 681 (Mþ, 100)]. Iden-tity of compound was confirmed by nuclear magneticresonance (NMR) as well as mass spectrometry (MS), andpurity more than 99% was validated by high-performanceliquid chromatography.

Western blot analysisCell lysates were harvested and processed as described

previously (27). A total of 1.5 � 106 melanoma cells wereplated in 100-mm culture dishes and were treated 48hours later with celecoxib, selenocoxib-1-GSH (5–20mmol/L), PLX-4032 (0.2–20 mmol/L), or U0126 (2.5–50mmol/L) for 6 to 72 hours. Protein lysates were collectedfor Western blotting. Blots were probed with total andpAkt (Ser473), pPRAS40 (Thr246), pErk1/2 (Thr202/Tyr204), total and pMek1/2 (Ser217), and cleaved PARPfrom Cell Signaling Technology. Total PRAS40 wasobtained from Invitrogen. Erk2, cyclin D1, p27, a-enolase,and secondary antibodies conjugated with horseradishperoxidase were purchased from Santa Cruz Biotechnol-ogy. COX-1 and COX-2 antibodies were obtained fromCayman Chemical Company. Immunoblots were devel-oped using the enhanced chemiluminescence detectionsystem from Amersham Pharmacia Biotech.

Cell viability, proliferation, apoptosis, and cell-cycleanalysisViability and IC50 (mmol/L) of normal human melano-

cytes, fibroblast, and melanoma cells following treatmentwith inhibitors were measured using the MTS assay (22,23). In brief, 5� 103 cells per well in 100 mL of media wereplated and grown in a 96-well plate for 36 to 72 hours formelanoma (WM35,WM115, 1205 Lu, andUACC 903) andnormal cell lines (FOM103 and FF2441). Cellswere treatedwith 0.312 to 100 mmol/L of celecoxib, selenocoxib-1, andselenocoxib-1-GSH for 24, 48, or 72 hours with DMSO asvehicle control. IC50 values for each inhibitor (in mmol/L)for respective cell lines were measured from 3 indepen-dent experiments using GraphPad Prism version 4.01from GraphPad Software.Cellular proliferation and apoptosis rates were mea-

sured by seeding 5 � 103 cells in 96-well plates, followedby treatment for 72 hours with celecoxib or selenocoxib-1-GSH. Percentage of proliferating or apoptotic cells wasquantified using a colorimetric assay using a cell prolif-eration ELISA BrdUrd kit from Roche Applied Sciencesor Apo-ONE Homogenous Caspase-3/7 Assay Kit fromPromega (22), respectively.Cells in each population of the cell cyclewere examined

by growing 1205 Lu or UACC 903 melanoma cells in 100-mmculture dishes followed by treatmentwith 12.5 and 25mmol/L of celecoxib and selenocoxib-1-GSH for 72 hours.

The samples were processed as described previously (22,23). Stained cells were analyzed using the FACScan ana-lyzer fromBectonDickinson and the data processedusingModFit LT software from Verity Software House (22, 23).Experiments were replicated twice.

Reactive oxygen species assayThe intracellular reactive oxygen species (ROS) was

monitored according to a published protocol (29). A totalof 1.5� 106melanoma cellswere plated in 100-mmculturedishes and treated 48 hours later with 5 to 20 mmol/Lconcentration of celecoxib, selenocoxib-1, or selenocoxib-1-GSH. After 24-hour treatment, total cells (floating andadherent) were collected in ice-cold PBS and 5 � 103 cellsper well placed in 100 mL of culture media in a 96-wellplate containing 10 mmol/L 20,70-dichlorfluorescein-dia-cetate and incubated at 37�C for 30 minutes. Amount offluorescent 20,70-dichlorfluorescein wasmeasured using aSpectraMax-M2 plate reader. Amount of ROS presentcompared with DMSO vehicle-treated cells was repre-sented in arbitrary units. The assay was conducted twicewith 4 replicates each time.

Tumorigenicity assessments following targeting ofCOX-2 using siRNA

Tumor kinetic studies were undertaken in athymic-Foxn1nu nude mice (Harlan Sprague Dawley). Then,200 pmol/L of siRNA COX-2 #1 or COX-2 #2 werenucleofected into 2 � 106 of 1205 Lu cells and, after 48hours of recovery, 1� 106 cells were collected in 0.2 mL of10% FBS–DMEMand injected subcutaneously above boththe left and right rib cages of 4- to 6-week-old femalemice(5 mice/group; experiments were replicated twice).Dimensions of developing tumors were measured onalternate days up to day 21.5, using calipers by L � W� D (mm3; ref. 22).

Animal studies using selenocoxib-1-GSH fortumorigenicity assessments

Six days after subcutaneous injection of 1� 106 1205 LuorUACC903 cells in 0.2mLofDMEMsupplementedwith10% FBS into 4- to 6-week-old nude mice, when a fullyvascularized tumor (50–75 mm3) had formed (5 mice/group; 2 tumors/mouse). Mice were treated intraperito-neally with selenocoxib-1-GSH (0.127 mmol/L, equivalentto 10 ppmselenium) or celecoxib (0.127mmol/L) inDMSOon alternate days for 4 weeks. Body weight (grams) anddimensions of the developing tumors (mm3) were mea-sured at the time of drug treatment (22, 23).

Toxicity assessmentsFour- to 6-week-old athymic-Foxn1nu nude mice were

treated with either vehicle control or selenocoxib-1-GSH(n¼ 5) as described in tumor kinetics studies.At the endoftreatment, blood was collected from each sacrificed ani-mal in a plasma separator tube with lithium heparin(Microtainer; Becton Dickinson) following cardiac punc-ture and analyzed for alkaline phosphatase, alanine

Targeting Melanoma Using Selenocoxib-1-GSH

www.aacrjournals.org Mol Cancer Ther; 12(1) January 2013 5




aminotransferase, aspartate aminotransferase, total albu-min, total bilirubin, creatinine, blood urea nitrogen, totalcholesterol, total triglyceride, and glucose levels to ascer-tain possible liver, heart, kidney, and pancreas-relatedtoxicity. A portion of vital organs, liver, heart, kidney,pancreas, and spleen, from each animal was formalin-fixed and paraffin-embedded to assess toxicity-associatedchanges in cell morphology and tissue organization fol-lowing hematoxylin and eosin (H&E) staining. In addi-tion, the effect of celecoxib, selenocoxib-1, and seleno-coxib-1-GSH on the survival of mice was determined byintraperitoneally injecting celecoxib (0.127 mmol/L), sele-nocoxib-1 (0.032–0.064 mmol/L), or selenocoxib-1-GSH(0.127–0.254 mmol/L) daily for 7 days (n ¼ 3). Numberof surviving animals or changes in body weight wasrecorded.

Statistical analysisStatistical analysis was carried out using Prism 4.01

GraphPad Software. One-way or 2-way ANOVA wasused for group-wise comparisons, followed by the Tukeyor Bonferroni post hoc tests. For comparison between 2groups, the t test was used. Results represent at least 2 to 3independent experiments and are shown as averages �SEM.Resultswith aPvalue less than 0.05 [95% confidenceinterval (CI)] were considered significant.

ResultsCOX-2 expression is elevated in advanced-stagemelanoma patient tumors and melanoma cell lines

Elevated expression and activity of COX-2 has beenreported in cancers of the prostate, breast, colon, kidney,liver, and skin (13, 30). To confirm the initial report inmelanomas (31), the expression of COX-2 was measuredby Western blotting in a panel of melanoma patienttumors and cell lines representing radial (WM35), vertical(WM115 and WM278.1), and metastatic (A375M, UACC903, and 1205 Lu) stages of development (Fig. 1A). Sev-enty-six percent (19/25) of melanoma patient tumors hadelevated COX-2 expression when compared with normalhuman melanocyte (NHEM) control cells (Fig. 1A, left).Similarly, all melanoma cell lines examined had higherCOX-2 protein content than that observed inmelanocytes,albeit, in varying amounts (Fig. 1A, right). Expression ofCOX-2 in UACC 903, A375M, and 1205 Lu cell lines was43, 76, and 329-fold higher than that observed in melano-cytes, respectively.

Reduction of COX-2 protein levels using siRNAtargeting V600EB-Raf or COX-2–decreasedmelanomacell viability

To determine whether targeting COX-2 would reduceviability of melanomas, metastatic 1205 Lu and A375Mcells that express relatively high levels of protein weretransfected with 2 different siRNAs targeting differentregions of the mRNA and cell viability compared withcontrols nucleofected with a scrambled siRNA, buffer

control, or siRNA targeting V600EB-Raf (Fig. 1B). In bothcell lines, targeting COX-2 reduced melanoma viabilityby 32% to 63%. Targeting mutant V600EB-Raf usingsiRNA reduced COX-2 expression, suggesting proteinexpression was regulated through this pathway (Fig.1B).

siRNA and pharmacologic agents targeting theMAPK pathway confirm that COX-2 expression isregulated through V600EB-Raf signaling inmelanomas

To examine whether siRNA-mediated targeting ofMek1/2 or Erk1/2 downstream of V600EB-Raf woulddecrease COX-2 expression, siRNA or pharmacologicagents were used to decrease protein expression or activ-ity. The 1205 Lu and UACC 903 cells were nucleofectedwith siRNAs inhibiting mutant V600EB-Raf, Mek1/2, orErk1/2. A significant decrease in COX-2, but not COX-1,was observed when each member of the V600EB-Raf sig-naling pathway was targeted (Fig. 1C).

Next, vemurafenib (PLX-4032), a V600EB-Raf inhibitor,was used to inhibit activity of this pathway. Cells treatedwith PLX-4032 showed decreased COX-2 protein expres-sion, beginning after 12 hours of treatment for 1205 Lucells. In the case of UACC 903, a significant decrease wasseen from 24 hours of treatment (Fig. 1D, top). Similar tosiRNA studies (Fig. 1C), no changes were observed inCOX-1 expression in either cell line following treatment.Adecrease in phosphorylation of Mek1/2 and Erk1/2 pro-teins showed the inhibitory activity of PLX-4032 on theV600EB-Raf pathway. Like PLX-4032, theMek1/2 inhibitorU0126 also reduced levels of COX-2 protein withoutaffecting COX-1 in 1205 Lu and UACC 903 cell lines (Fig.1D, bottom). Therefore, targeting V600EB-Raf or down-stream proteins in the signaling cascade reduced expres-sion of COX-2 in melanomas to decrease the proliferativepotential of the cells. Thus, COX-2 lies downstream ofV600EB-Raf, Mek-1/2, and Erk-1/2 in this important sig-naling pathway.

Development of selenocoxib-1-GSH retaining COX-2 inhibitory efficacy

Concentrations of celecoxib required to trigger apopto-sis in cultured cells range from 25 to 100 mmol/L andclinical use is associated with cardiovascular side-effectsat doses of 200 mg per day (32). To circumvent theseconcerns, an analogue of celecoxib has been created con-taining selenium and is called selenocoxib-1 (Fig. 2A).While selenocoxib-1 inhibited the viability of melanomacells, it was toxic and reduced normal cell growth withsimilar IC50s to that observed for melanoma cells (Table 1& Fig. 2B). To lessen the toxicity of selenocoxib-1 onnormal, but not melanoma cells, the compound wasfurther modified incorporating GSH, generating seleno-coxib-1-GSH (Fig. 2A), which significantly decreased thetoxicity on normal cells butmaintained its melanoma cell-killing efficacy (Fig. 2B). Furthermore, selenocoxib-1-GSHinhibited the growth ofmelanoma cell lines irrespective of

Gowda et al.





Figure 1. COX-2 expression increased in melanomas, regulated through the MAPK pathway. A, left, elevated levels of COX-2 expression in melanomapatient tumors and cell lines; right, COX-2 expression increased in a cell line-based melanoma tumor progression model. Lysates collected from normalhuman melanocytes, radial growth phase (WM35), vertical growth phase (WM115, WM278.1), and metastatic (A375M, UACC, and 1205 Lu) stage cell lineswere subjected to Western blot analysis and probed for COX-2. a-Enolase served as a control for equal protein loading. B, targeting COX-2 usingsiRNAs decreasedmelanoma cell viability. The 1205 Lu andA375Mmelanoma cells were nucleofectedwith 2 nonoverlapping siRNAs against COX #1 and #2using Amaxa nucleofector, programK17. TargetingCOX-2 reducedmelanoma viability by 32 to 63%. siRNA targetingmutant V600EB-Raf served as a positivecontrol. a-Enolase served as a control for equal protein loading. C, siRNA-mediated inhibition of the MAPK pathway decreased COX-2 expressionin melanomas. The 1205 Lu and UACC 903 cells were nucleofected with siRNAs inhibiting mutant V600EB-Raf, Mek1 or Mek2, and Erk1 or Erk2.Compared to cells treated with scrambled siRNA, decreasing protein levels of each member of the MAP kinase pathway led to a decrease in COX-2 but notCOX-1 levels. a-Enolase served as a control for equal protein. D, top, PLX-4032 targeting of V600EB-Raf decreased COX-2 expression. The 1205 Lu orUACC 903 cells were treated with 0.2–20 mmol/L PLX-4032 for 6, 12, 24, and 48 h. Levels of pMek1/2, pErk1/2, and COX-2 decreased after 12 h of drugtreatment. No changes were seen in COX-1 expression. a-Enolase served as a control for equal protein loading. Bottom, U0126 targeting of Mek1/2decreased COX-2 expression. The 1205 Lu or UACC 903 cells were treated with 2.5–50 mmol/L U0126 for 48 h. Levels of pErk1/2 and COX-2 similarlydecreased following drug treatment, with no changes in COX-1 expression. a-Enolase served as a control for equal protein loading.






B-Raf mutation status (data not shown). Toxicity limitingthe potential clinical utility of selenocoxib-1, but notselenocoxib-1-GSH, was corroborated in animal studies(Supplementary Table S1). Celecoxib at 0.127 mmol/L ledto death of all animals following 7 days of treatment,whereas selenocoxib-1 at concentrations of 0.032 to

0.064 mmol/L, led to weight losses resulting in 14% or100% animal mortality after 7 days of treatment. In con-trast, animals receiving 0.127 to 0.254 mmol/L (equivalentto 5–10ppmof selenium)of selenocoxib-1-GSH exhibitednegligible weight loss of approximately 2% and nomortality (Supplementary Table S1). Due to the toxicity

Figure 2. Development ofselenocoxib-1-GSH. A, structureof celecoxib, selenocoxib-1, andselenocoxib-1-GSH. B,selenocoxib-1-GSH kills cancermore effectively that normal cells.Selenocoxib-1 inhibitedmelanomacell viability. It also reduced normalcell growthwith IC50s similar to thatof melanoma cells, indicatingtoxicity. In contrast, selenocoxib-1-GSH, but not selenocoxib-1,inhibited the growth of radialgrowth phase (RGP), verticalgrowth phase (VGP), andmetastatic melanoma cells, with alesser effect on the growth ofnormal human melanocytes orfibroblast cells. C, selenocoxib-1-GSH retained COX-2 inhibitoryactivity. Human recombinantCOX-2 activity was assayed usinga commercial COX-inhibitorscreening assay kit. 0.2, 2.0, and20nmol/L of celecoxib andselenocoxib-1-GSH weretested and shown to retainCOX-2 inhibitory activity.DMSO served as the negativecontrol (100% activity). Theproduct of this enzymaticreaction was determinedspectrophotometrically at 405 nm.Assay was performed in2 independent experiments.

Gowda et al.





associated with selenocoxib-1, subsequent studiesfocused on selenocoxib-1-GSH. Selenocoxib-1-GSH wasnot predicted to have altered COX-2 inhibitory activityby incorporating selenium in the place of sulfur butrather to have Akt inhibitory properties and, as pre-dicted, it retained COX-2 inhibitory activity similar tothat of celecoxib (Fig. 2C).

Selenocoxib-1-GSH inhibited melanoma cellularproliferation and increased apoptosis by arrestingcells in the G0–G1 phase of the cell cycleSelenocoxib-1-GSH, but not celecoxib, inhibited mela-

noma cell viability in a dose-responsive manner (Fig. 3A).At 12.5 mmol/L, selenocoxib-1-GSH led to a 40% to 60%decrease in cell viability compared with control DMSO–vehicle-treated cells. Next, mechanisms leading to cellgrowth inhibition after treatment with selenocoxib-1-GSHwere examined by measuring the level of cellular prolife-ration, apoptosis, and the percentage of cells in the vari-ous phases of the cell cycle. Selenocoxib-1-GSH reducedproliferation of 1205 Lu and UACC 903 melanoma cells(Fig. 3B) and increased caspase-3/7 activity, which is anindicator of apoptosis (Fig. 3C). A significant decreasein caspase-3/7 activity was observed when UACC 903cells were treated with 50 mmol/L selenocoxib-1-GSH,which can be attributed to massive cell death (Fig. 3C).The effect of selenocoxib-1-GSH on cell-cycle distributionwas measured by analyzing propidium iodide–stained1205 Lu and UACC 903 cells using a Becton DickinsonFACScan. Selenocoxib-1-GSH treatment increased thesub-G0–G1 cell population, which is indicative of cellularapoptosis. The sub-G0–G1 cell population increased by6.2- and 4.7-fold, respectively, when 1205 Lu and UACC903 cells were treated with 25 mmol/L selenocoxib-1-GSH(Fig. 3D). In addition, an increase in the G0–G1 cell popu-lation was also observed at 12.5 and/or 25 mmol/Lselenocoxib-1-GSH (Fig. 3D). Thus, selenocoxib-1-GSHinhibited cellular proliferation and triggered apoptosismediated through a G0–G1 block, resulting in fewer cellsin the S and G2–M phases of the cell cycle.

Selenocoxib-1-GSH inhibited Akt signaling, whichactivatesMAPKactivity to reducemelanoma cellularproliferation and promote apoptosis

Selenium incorporation into the structural backbone ofcertain agents can enhance the therapeutic potential of theagent by providing the compound with new inhibitoryproperties (19, 22, 23, 33). Selenocoxib-1-GSH retainedCOX-2 inhibitory activity as predicted (Fig. 2C). To deter-mine whether selenium incorporation into selenocoxib-1-GSH provided the compound with new Akt pathwayinhibitory properties, pAkt levels were examined in mel-anoma cells following treatment. Compared with cele-coxib, selenocoxib-1-GSH treatment inhibited Akt phos-phorylation in a dose-dependent manner (Fig. 4A). Fur-thermore, phosphorylation of the downstream Akt3 sub-strate PRAS40 was significantly inhibited.

The 1205 Lu and UACC 903 melanoma cells haveelevated MAPK activities due to the presence of consti-tutively active V600EB-Raf; however, the levels are mod-erated into a range that promotes rather than inhibitscellular proliferation (34). Akt3 has been shown to phos-phorylate V600EB-Raf to lower MAPK pathway activity topromote cellular proliferation (34). Treatment of melano-ma cells with selenocoxib-1-GSH, which decreased Aktactivity, led to a significant increase in pERK1/2 levels(the indicator of MAPK pathway activity; Fig. 4B), to apoint where it no longer promoted proliferation but led tocell senescence. This was due to decreased phosphoryla-tion and regulation of V600EB-Raf by Akt (34). In addition,selenocoxib-1-GSH inhibited expression of cyclin D1 andincreased levels of p27 (Fig. 4C). Finally, increased cas-pase-3/7 and cleaved PARP levels were observed indi-cating higher levels of apoptosis in selenocoxib-1-GSHcompared with celecoxib-treated cells (Fig. 4D).

Selenocoxib-1-GSH inhibited melanoma tumordevelopment in mice without significant toxicity

Initially, siRNA-targeting COX-2 was used to reduceprotein expression inmelanoma cells tomeasure the effecton melanoma tumor development to serve as a control.

Table 1. Selenocoxib-1-GSH kills melanoma cells more effectively than normal cells

FOM103 FF2441 WM35 WM115 UACC 903 1205 Lu

Celecoxib >100 >100 51.3 � 2.3 54.4 � 3.6 >100 >100 24 hSelenocoxib-1 -GSH 66.3 � 3.0 >100 52.6 � 3.8 30.1 � 3.3 30.9 � 2.8 24.6 � 2.2Celecoxib >100 >100 42.3 � 1.8 45.3 � 2.7 >100 83.6 � 3.3 48 hSelenocoxib-1 -GSH 53.4 � 4.3 75.5 � 5.6 4.1 � 0.8 5.8 � 0.9 20.6 � 1.7 17.2 � 1.6Celecoxib 68.0 � 1.2 65.3 � 3.3 37.9 � 3.3 41.8 � 2.9 76.6 � 4.4 66.1 � 3.2 72 hSeleriocoxib-1 -GSH 41.2 þ 3.2 36.3 � 5.6 2.7 � 0.4 3.1 � 0.4 14.2 � 1.2 10.6 � 2.6

Normal Radial Vertical Metastatic

NOTE:Normal andmelanomacellswere seeded in to a 96-well plate and, after 36 to 72 hours, treatedwith increasing concentrations ofcelecoxib, selenocoxib-1, or selenocoxib-1-GSH for the indicated time period. Number of viable cells was measured using MTS andpercentage decrease in viability calculated. IC50 values for each inhibitor in mmol/L for respective cell lines were measured from 3independent experiments using GraphPad Prism version 4.01 (GraphPad Software, La Jolla, CA).






Inhibition of COX-2 protein expression using siRNAsreduced xenografted melanoma tumor development byan average of 71% after 21 days compared with controls,suggesting COX-2 was a good therapeutic target in mel-anomas (Fig. 5A). Next, the effect of intraperitonealadministration of selenocoxib-1-GSHonxenograftedmel-anoma tumordevelopmentwas examined (Fig. 5B andC).

Decreased xenografted tumor development comparedwith control-treated mice was observed from day 16 in1205 Lu tumors (Fig. 5B). Similarly, a significant decreasewas observed in UACC 903 tumors from day 22 (Fig. 5C).For both cell lines at the end of treatment, up to a 70%decrease in tumor volume was observed following sele-nocoxib-1-GSH treatment comparedwith controls (Fig. 5B

Figure 3. Selenocoxib-1-GSHinhibited melanoma cell growth byreducing cellular proliferation,triggering apoptosis, and arrestingmelanomacells in theG0–G1 phaseof the cell cycle. A–C, selenocoxib-1-GSH, but not celecoxib, inhibitedmelanoma cell proliferation andinduced apoptosis. 1205 Lu andUACC 903 cells were treated withincreasing concentrations ofcelecoxib and selenocoxib-1-GSHfor 72 h and cell viability,proliferation, and apoptosis ratesmeasured by MTS, BrdUincorporation and caspase-3/7assays, respectively. Datarepresent averages of at least 3independent experiments; bars;SEM. D, selenocoxib-1-GSHarrested melanoma cells in theG0–G1 phase of the cell cycle. The1205 Lu and UACC 903 cells weretreated with 12.5 and 25 mmol/L ofcelecoxib, selenocoxib-1-GSH, orvehicle DMSO control for 72 h.Total floating and adherent cellswere collected and stained withpropidium iodide to analyze thedistribution of cells in differentphases of the cell cycle stagesusing a FACScan analyzer.Selenocoxib-1-GSH, but notcelecoxib treatment, inhibited cellcycle progressionby increasing thesub-G0–G1 population andarresting cells in the G0–G1 phasesof the cell cycle. Data representan average of 2 independentexperiments.

Gowda et al.





Figure 4. Selenocoxib-1-GSH inhibitedAkt signaling to reduce the proliferative potential andpromote apoptotic signaling inmelanomacells. A, selenocoxib-1-GSH inhibits the PI3K/Akt signaling pathway. B, selenocoxib-1-GSH activates the MAPK signaling pathway. C, selenocoxib-1-GSH decreased cyclin D1protein levels, indicating a reduction in cellular proliferation. D, selenocoxib-1-GSH increased levels of cellular apoptosis.






and C). No noticeable changes in animal body weightwere observed (Fig. 5B and C; insets). The levels of bloodmarkers for major–organ-related toxicity and analysis ofH&E-stained tissue sections showed negligible differ-ences comparedwith controls at the concentrations exam-ined (Fig. 5D and Supplementary Fig. S1). These datasuggest that selenocoxib-1-GSH can inhibit melanomatumor development without significant organ-relatedtoxicity.

DiscussionIncidence and mortality rates of malignant melanoma

continue to increase annually (35). Although efforts have

been made to design structurally well-defined small-molecular inhibitors that interact with protein targets inmelanoma cells, these efforts have failed due to devel-opment of resistant disease (36). Therefore, the realiza-tion now is that multiple important targets driving thedevelopment of this disease will need to be simulta-neously targeted to most effectively manage melanomaand reduce the probability of resistant disease devel-opment. This may be achievable through the use of drugcocktails or a single drug that simultaneously inhibitsmultiple key signaling pathways implicated in melano-ma development (37). In addition, selection of patientsexpressing proteins targeted by the drug would be a key

Figure 5. Targeting COX-2 inhibited melanoma tumor development with negligible toxicity. A, siRNA-mediated reduction of COX-2 protein levels decreasedmelanoma tumor development in mice. The 1205 Lu melanoma cells were nucleofected with siRNA targeting COX-2. Forty-eight h later, viable cells weresubcutaneously injected into left and right flanks of nudemice. Developing tumors weremeasured on alternate days for 21.5 days. The tumorigenic potentialof COX-2-treated cells decreased by �70% compared to buffer or scrambled siRNA control cells. B and C, selenocoxib-1-GSH treatment decreasedxenografted melanoma tumor development. To measure the effect of selenocoxib-1-GSH on tumor development, mice were s.c. injected with 1�106 1205Lu (B) or UACC 903 (C) melanoma cells. After 6 d, mice were treated i.p. with selenocoxib-1-GSH (0.127 μm, equivalent to 10 ppm selenium) orcelecoxib (0.127 μm) in DMSOon alternate days for 4 weeks. Selenocoxib-1-GSH reduced tumor development by 70.5% and 52% in 1205 Lu andUACC903cells, respectively, compared to DMSO control treated mice (P < 0.001; two-way ANOVA). No significant difference was observed in body weight of micetreated with the drug, indicating negligible toxicity (B and C; inset). No significant difference was observed in body weight of mice treated with thedrug, indicating negligible toxicity (B and C; inset). D, selenocoxib-1-GSH does not affect blood biomarkers indicative of major organ-related toxicity.The levels of bloodmarkers for major organ-related toxicity, ALKP (alkaline phosphatase), ALT (alanine aminotransferase), AST (aspartate aminotransferase),ALB (total albumin), TBIL (total bilirubin), CREA (creatinine), BUN (blood urea nitrogen), CHOL (total cholesterol), TRIG (total triglyceride) and GLU (glucose)were evaluated and levels indicated negligible differences compared to controls at the concentrations examined.

Gowda et al.





factor that could lead to better results in the clinic. Inthis study, COX-2 protein levels are shown to be ele-vated in 76% of melanoma patient tumors and cell lines.Targeting COX-2 using siRNA, inhibited the growth ofmetastatic melanoma cells in culture and retarded thedevelopment of xenografted melanoma tumors in mice,indicating that COX-2 would be a good therapeutictarget.Because siRNA-mediated inhibition of COX-2 expres-

sion decreased melanoma tumor growth, pharmacologi-cal agents inhibiting COX-2 activity with high selectivitymay have therapeutic potential for inhibitingmelanomas.However, the COX-2 selective inhibitor celecoxib has aneffect onmelanoma cell proliferation but only at very highconcentrations, necessitating the development of analo-gues better able to kill these cells at lower concentrations(38, 39). Initially, an analogue of celecoxib was developedthat contained selenium, called selenocoxib-1, but thedrug was toxic to normal cells and lethal to animals,which limited its use as a therapeutic agent (18). Tomanage this concern, a GSH derivative called seleno-coxib-1-GSHwasdeveloped. It killed cultured cancer cellsat doses 5-fold lower than those required to kill normalcells.Addition ofGSH to a compound can be used to increase

bioavailability and reduce cytotoxicity as seen with theGSH conjugate of benzyl selenocyanate for inhibiting ofcolonic preneoplastic lesions and aberrant crypt focidevelopment (40). One mechanism by which GSH con-jugates inhibit cancer involves the reduction of GSHreductase activity, which depletes intracellular reducedGSH, thereby enhancingROS-mediated cell death (41, 42).Others have reported that GSH-depleting agents canselectively sensitize cancer cells to high ROS levels (43).Another mechanism by which GSH-conjugated antican-cer agents inhibit cancer cells growth involves reductionof elevated ROS levels. High ROS levels mediate cancerdevelopment and reduction reverses this process (29).Our data suggest that selenocoxib-1-GSH reduces ROSlevels inmelanoma cellsmore effectively than celecoxib orselenocoxib-1 to kill these cells (Supplementary Fig. S2).Enhanced growth-inhibitory properties of selenocoxib-

1-GSH could also be attributed to the incorporation ofselenium into the structure of celecoxib. Selenium is anantioxidant nutrient reported to inhibit oncogenic Aktand NFkB pathways as well as inducing the expressionof tumor suppressors PTEN, p53, and KLF-4 to mediateapoptotic cell death (19, 44). Incorporation of seleniuminto the structure of drugs has been shown to increase theagent’s potency by inhibiting Akt signaling (22, 23, 45).The selenium-containing analogues of the PBIT andPBITC called PBISe and ISC-4, respectively, inhibitedmelanoma cell growth and suppressed tumor develop-ment in animals more effectively than the sulfur-contain-ing parental compound (22, 23). In addition, because verylow levels of selenium occur in the majority of patientswith melanoma, incorporating selenium may not onlyprovide this micronutrient but also increase tumor cell-

killing efficacy (33). While several reports document lossof COX-2 inhibitory activity when celecoxib was deriva-tized or analogues were synthesized, selenocoxib-1-GSHretained COX-2 inhibitory activity and had new Akt-targeting capabilities. In contrast, 2,5-dimethyl-celecoxiblacked the ability to inhibit COX-2, but still had antitumoractivity (46).

The PI3K/Akt andMAPK signaling pathways are con-stitutively activated in melanoma and play a prominentrole in the development of recurrent resistant disease (22,23, 47). Selenocoxib-1-GSH while retaining COX-2 inhib-itory activity at levels seen with celecoxib, also blockedAkt signaling. Decreasing levels of active pAkt3 increasedV600EB-Raf activity and downstream MAPK-signalingactivity to levels that are inhibitory, inducing cellularsenescence (34, 48). This phenomenon occurs followingselenocoxib-1-GSH treatment of melanoma cells. Seleni-um-containing PBISe treatment acts in a similarmanner todecrease Akt activity, consequently increasing activity ofMAPK pathways to inhibitory levels (23). Other studiesand this one found that increased pErk-1/2 in turn upre-gulated COX-2 protein expression, consistent with COX-2lying downstream in the MAPK pathway (49). HighMAPK pathway activity mediated by selenocoxib-1-GSHor PBISe induced cell senescence arresting cells in G0–G1

phase of the cell cycle. Combined targeting of these path-ways inhibited cell proliferation by lowering cyclin D1and increasing p27 levels, which enhanced rates of cellu-lar apoptosis.

In conclusion, selenium-containing selenocoxib-1-GSHretains COX-2 inhibitory activity and has new PI3K/Aktinhibitory activity to decrease melanoma cell growth byarresting cells in the G0–G1 phase of the cell cycle topromote melanoma cell apoptosis and inhibit cellularproliferation. Thus, a potentially clinically viable drughas been developed from a toxic agent that can decreasemelanoma development by targeting the COX-2 andPI3K/Akt signaling pathways without causing major–organ-related toxicity.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: R. Gowda, S.V. Madhunapantula, D. Desai,S. Amin, G.P. Robertson.Development of methodology: R. Gowda, S.V. Madhunapantula,D. Desai, S. Amin, G.P. Robertson.Acquisition of data: R. Gowda, S.V. Madhunapantula, G.P. Robertson.Analysis and interpretation of data (e.g., statistical analysis, biostatis-tics, computational analysis): R. Gowda, S.V. Madhunapantula, G.P.Robertson.Writing, review, and/or revision of the manuscript: R. Gowda, S.V.Madhunapantula, D. Desai, S. Amin, G.P. Robertson.Administrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): R. Gowda, S.V. Madhunapantula,D. Desai, S. Amin, G.P. Robertson.Study supervision: G.P. Robertson.

AcknowledgmentsThe authors thank Arati Sharma, Omer Kuzu, Virginia Robertson, and

Anton Mulder for technical assistance.






Grant SupportFinancial support for this study was provided by the NIH CA-127892-

01A (G.P. Robertson) and The Foreman Foundation for MelanomaResearch (G.P. Robertson).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby marked

advertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received May 17, 2012; revised September 7, 2012; accepted October 4,2012; published OnlineFirst October 30, 2012.

References1. Oliveria SA, Hay JL, Geller AC, Heneghan MK, McCabe MS, Halpern

AC. Melanoma survivorship: research opportunities. J Cancer Surviv2007;1:87–97.

2. Flaherty KT, Puzanov I, Kim KB, Ribas A, McArthur GA, Sosman JA,et al. Inhibition of mutated, activated BRAF in metastatic melanoma.N Engl J Med 363:809–19.

3. Wolchok JD, Neyns B, Linette G, Negrier S, Lutzky J, Thomas L, et al.Ipilimumab monotherapy in patients with pretreated advanced mela-noma: a randomised, double-blind, multicentre, phase 2, dose-rang-ing study. Lancet Oncol 11:155–64.

4. Smalley KS, Sondak VK. Melanoma–an unlikely poster child for per-sonalized cancer therapy. N Engl J Med 363:876–8.

5. Fedorenko IV, Paraiso KH, Smalley KS. Acquired and intrinsic BRAFinhibitor resistance in BRAF V600Emutant melanoma. Biochem Phar-macol 2011;82:201–9.

6. Villanueva J, Vultur A, Herlyn M. Resistance to BRAF inhibitors:unraveling mechanisms and future treatment options. Cancer Res2011;71:7137–40.

7. Nguyen N, Sharma A, Sharma AK, Desai D, Huh SJ, Amin S, et al.Melanoma chemoprevention in skin reconstructs and mousexenografts using isoselenocyanate-4. Cancer Prev Res (Phila) 2011;4:248–58.

8. Nikolaou VA, Stratigos AJ, Flaherty KT, Tsao H. Melanoma: newinsights and new therapies. J Invest Dermatol 2012;132:854–63.

9. Hawk ET, Viner JL, Dannenberg A, DuBois RN. COX-2 in cancer–aplayer that's defining the rules. J Natl Cancer Inst 2002;94:545–6.

10. Sobolewski C, Cerella C, Dicato M, Ghibelli L, Diederich M. The role ofcyclooxygenase-2 in cell proliferation and cell death in human malig-nancies. Int J Cell Biol 2010:215158.

11. Meric JB, Rottey S, Olaussen K, Soria JC, Khayat D, Rixe O, et al.Cyclooxygenase-2 as a target for anticancer drug development. CritRev Oncol Hematol 2006;59:51–64.

12. Biberstine KJ, Darr DS, Rosenthal RS. Tolerance to appetite suppres-sion induced by peptidoglycan. Infect Immun 1996;64:3641–5.

13. Becker MR, Siegelin MD, Rompel R, Enk AH, Gaiser T. COX-2 expres-sion in malignant melanoma: a novel prognostic marker? MelanomaRes 2009;19:8–16.

14. Denkert C, Kobel M, Berger S, Siegert A, Leclere A, Trefzer U, et al.Expression of cyclooxygenase 2 in human malignant melanoma.Cancer Res 2001;61:303–8.

15. WaskewichC, Blumenthal RD, Li H, Stein R,GoldenbergDM, Burton J.Celecoxib exhibits the greatest potency amongst cyclooxygenase(COX) inhibitors for growth inhibition ofCOX-2-negative hematopoieticand epithelial cell lines. Cancer Res 2002;62:2029–33.

16. Drazen JM.COX-2 inhibitors–a lesson in unexpected problems. NEnglJ Med 2005;352:1131–2.

17. Jawabrah Al-Hourani B, Sharma SK, Suresh M, Wuest F. Cyclooxy-genase-2 inhibitors: a literature and patent review (2009–2010). ExpertOpin Ther Pat 2011;21:1339–432.

18. Desai D, Sinha I, Null K, Wolter W, SuckowMA, King T, et al. Synthesisand antitumor properties of selenocoxib-1 against rat prostate ade-nocarcinoma cells. Int J Cancer 2010;127:230–8.

19. Desai D, Kaushal N, Gandhi UH, Arner RJ, D'Souza C, Chen G, et al.Synthesis and evaluation of the anti-inflammatory properties of sele-nium-derivatives of celecoxib. Chem Biol Interact 2010;188:446–56.

20. ChuangHC,KardoshA,GaffneyKJ, PetasisNA,Schonthal AH.COX-2inhibition is neither necessary nor sufficient for celecoxib to suppresstumor cell proliferation and focus formation in vitro. Mol Cancer2008;7:38.

21. Hwang JT, Kim YM, Surh YJ, Baik HW, Lee SK, Ha J, et al. Seleniumregulates cyclooxygenase-2 and extracellular signal-regulated kinase

signaling pathways by activating AMP-activated protein kinase incolon cancer cells. Cancer Res 2006;66:10057–63.

22. Sharma A, Sharma AK, Madhunapantula SV, Desai D, Huh SJ, MoscaP, et al. Targeting Akt3 signaling in malignant melanoma using iso-selenocyanates. Clin Cancer Res 2009;15:1674–85.

23. Madhunapantula SV, Desai D, Sharma A, Huh SJ, Amin S, RobertsonGP. PBISe, a novel selenium-containing drug for the treatment ofmalignant melanoma. Mol Cancer Ther 2008;7:1297–308.

24. LoPiccolo J, Blumenthal GM, Bernstein WB, Dennis PA. Targeting thePI3K/Akt/mTOR pathway: effective combinations and clinical consid-erations. Drug Resist Updat 2008;11:32–50.

25. Satyamoorthy K, DeJesus E, Linnenbach AJ, Kraj B, Kornreich DL,Rendle S, et al. Melanoma cell lines from different stages of progres-sion and their biological and molecular analyses. Melanoma Res1997;7(Suppl 2):S35–42.

26. Madhunapantula SV, Sharma A, Robertson GP. PRAS40 deregulatesapoptosis in malignant melanoma. Cancer Res 2007;67:3626–36.

27. Sharma A, Tran MA, Liang S, Sharma AK, Amin S, Smith CD, et al.Targeting mitogen-activated protein kinase/extracellular signal-regu-lated kinase kinase in the mutant (V600E) B-Raf signaling cascadeeffectively inhibits melanoma lung metastases. Cancer Res 2006;66:8200–9.

28. Penning TD, Talley JJ, Bertenshaw SR, Carter JS, Collins PW, DocterS, et al. Synthesis and biological evaluation of the 1,5-diarylpyrazoleclass of cyclooxygenase-2 inhibitors: identification of 4-[5-(4-methyl-phenyl)-3-(trifluoromethyl)-1H-pyrazol-1-yl]benze nesulfonamide(SC-58635, celecoxib). J Med Chem 1997;40:1347–65.

29. SomasundaramS,EdmundNA,MooreDT,SmallGW,ShiYY,OrlowskiRZ. Dietary curcumin inhibits chemotherapy-induced apoptosis inmodels of human breast cancer. Cancer Res 2002;62:3868–75.

30. Ghosh N, Chaki R, Mandal V, Mandal SC. COX-2 as a target for cancerchemotherapy. Pharmacol Rep 2010;62:233–44.

31. Flockhart RJ, Armstrong JL, Reynolds NJ, Lovat PE. NFAT signalling isa novel target of oncogenicBRAF inmetastaticmelanoma.Br JCancer2009;101:1448–55.

32. Senior K. COX-2 inhibitors: cancer prevention or cardiovascular risk?Lancet Oncol 2005;6:68.

33. Gowda R, Madhunapantula SV, Desai D, Amin S, Robertson GP.Selenium-containing histone deacetylase inhibitors for melanomamanagement. Cancer Biol Ther 2012;13:756–65.

34. Cheung M, Sharma A, Madhunapantula SV, Robertson GP. Akt3 andmutant V600E B-Raf cooperate to promote early melanoma develop-ment. Cancer Res 2008;68:3429–39.

35. Gray-Schopfer V, Wellbrock C, Marais R. Melanoma biology and newtargeted therapy. Nature 2007;445:851–7.

36. Kitano H. A robustness-based approach to systems-oriented drugdesign. Nat Rev Drug Discov 2007;6:202–10.

37. Lasithiotakis KG, Sinnberg TW, Schittek B, Flaherty KT, Kulms D,Maczey E, et al. Combined inhibition of MAPK and mTOR signalinginhibits growth, induces cell death, and abrogates invasive growth ofmelanoma cells. J Invest Dermatol 2008;128:2013–23.

38. Bundscherer A, Hafner C, Maisch T, Becker B, Landthaler M, Vogt T.Antiproliferative and proapoptotic effects of rapamycin and celecoxibin malignant melanoma cell lines. Oncol Rep 2008;19:547–53.

39. Wilgus TA, Breza TS Jr, Tober KL, Oberyszyn TM. Treatment with 5-fluorouracil and celecoxib displays synergistic regression of ultravioletlight B-induced skin tumors. J Invest Dermatol 2004;122:1488–94.

40. Hasegawa T, Okuno T, Nakamuro K, Sayato Y. Identification andmetabolism of selenocysteine-glutathione selenenyl sulfide (CySeSG)in small intestine ofmice orally exposed to selenocystine. Arch Toxicol1996;71:39–44.

Gowda et al.





41. Estrela JM, Ortega A, Obrador E. Glutathione in cancer biology andtherapy. Crit Rev Clin Lab Sci 2006;43:143–81.

42. Burg D, Mulder GJ. Glutathione conjugates and their syntheticderivatives as inhibitors of glutathione-dependent enzymesinvolved in cancer and drug resistance. Drug Metab Rev 2002;34:821–63.

43. Gouaze V, Mirault ME, Carpentier S, Salvayre R, Levade T, Andrieu-Abadie N. Glutathione peroxidase-1 overexpression preventsceramide production and partially inhibits apoptosis in doxorubi-cin-treated human breast carcinoma cells. Mol Pharmacol 2001;60:488–96.

44. Bureau C, Hanoun N, Torrisani J, Vinel JP, Buscail L, Cordelier P.Expression and function of Kruppel like-factors (KLF) in carcinogen-esis. Curr Genomics 2009;10:353–60.

45. Das A, Bortner J, Desai D, Amin S, El-Bayoumy K. The seleniumanalog of the chemopreventive compound S,S0-(1,4-phenylenebis[1,2-ethanediyl])bisisothiourea is a remarkable inducer

of apoptosis and inhibitor of cell growth in human non-small celllung cancer. Chem Biol Interact 2009;180:158–64.

46. Schonthal AH,Chen TC,Hofman FM, Louie SG, Petasis NA.Celecoxibanalogs that lackCOX-2 inhibitory function: preclinical development ofnovel anticancer drugs. Expert Opin Investig Drugs 2008;17:197–208.

47. Tran MA, Gowda R, Sharma A, Park EJ, Adair J, Kester M, et al.Targeting V600EB-Raf andAkt3 using nanoliposomal-small interferingRNA inhibits cutaneous melanocytic lesion development. Cancer Res2008;68:7638–49.

48. Madhunapantula SV, Hengst J, Gowda R, Fox TE, Yun JK, RobertsonGP. Targeting sphingosine kinase-1 to inhibit melanoma. Pigment CellMelanoma Res 2012;25:259–74.

49. Elder DJ, Halton DE, Playle LC, Paraskeva C. The MEK/ERK pathwaymediates COX-2-selective NSAID-induced apoptosis and inducedCOX-2 protein expression in colorectal carcinoma cells. Int J Cancer2002;99:323–7.






2013;12:3-15. Published OnlineFirst October 30, 2012.Mol Cancer Ther Raghavendra Gowda, SubbaRao V. Madhunapantula, Dhimant Desai, et al. Selenocoxib-1-GSH to Inhibit MelanomaSimultaneous Targeting of COX-2 and AKT Using

Updated version

10.1158/1535-7163.MCT-12-0492doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2012/10/26/1535-7163.MCT-12-0492.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/12/1/3.full#ref-list-1

This article cites 45 articles, 14 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/12/1/3.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mct.aacrjournals.org/content/12/1/3To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-12-0492

http://mct.aacrjournals.org/content/suppl/2012/10/26/1535-7163.MCT-12-0492.DC1

http://mct.aacrjournals.org/content/12/1/3.full#ref-list-1

http://mct.aacrjournals.org/content/12/1/3.full#related-urls

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mct.aacrjournals.org/content/12/1/3


Simultaneous Targeting of COX-2 and AKT Using Selenocoxib ... · for cancer cells. Selenocoxib-1-GSH reduced development of xenografted tumor by approximately 70% with negligible

Documents