Implications of tissue transglutaminase expression in malignant … · Implications of tissue transglutaminase expression in malignant melanoma Jansina Y. Fok,1 Suhendan Ekmekcioglu,1

Implications of tissue transglutaminase expressionin malignant melanoma

Jansina Y. Fok,1 Suhendan Ekmekcioglu,1

and Kapil Mehta1,2

1Department of Experimental Therapeutics, The University ofTexas M.D. Anderson Cancer Center and 2Cancer BiologyProgram, Graduate School of Biomedical Sciences, The Universityof Texas, Houston, Texas

AbstractHuman malignant melanoma is a highly aggressive form ofcancer; the 5-year survival rate in patients with stage III orIV disease is <5%. In patients with metastatic melanoma,systemic therapy becomes ineffective because of the highresistance of melanoma cells to various anticancertherapies. We have found previously that developmentof the drug resistance and metastatic phenotypes in breastcancer cells is associated with increased tissue trans-glutaminase (TG2) expression. In the study reported here,we investigated TG2 expression and its implications inmetastatic melanoma. We found that metastatic melano-ma cell lines expressed levels of TG2 up to 24-fold higherthan levels in radial growth phase of primary melanomacell lines. Activation of endogenous TG2 by the calciumionophore A23187 induced a rapid and strong apoptoticresponse in A375 cells and A23187-induced apoptosiscould be blocked by TG2-specific inhibitors. These find-ings indicated that activation of endogenous TG2 couldserve as a strategy for inducing apoptosis in malignantmelanomas. Importantly, tumor samples from patientswith malignant melanomas showed strong expression ofTG2, suggesting that TG2 expression is selectively up-regulated during advanced developmental stages ofmelanoma. We observed that 20% to 30% of TG2 proteinwas present on cell membranes in association with B1 andB5 integrins. This association of TG2 with cell surfaceintegrins promoted strong attachment of A375 cells tofibronectin-coated surfaces, resulting in increased cellsurvival in serum-free medium. Inhibition of TG2 by smallinterfering RNA inhibited fibronectin-mediated cell attach-ment and cell survival functions in A375 cells. Overall, our

results suggest that TG2 expression contributes to thedevelopment of chemoresistance in malignant melanomacells by exploiting integrin-mediated cell survival signalingpathways. [Mol Cancer Ther 2006;5(6):1493–503]

IntroductionMelanoma is the deadliest form of skin cancer and itsincidence is rapidly growing (1). In stage I and II disease,complete surgical excision of the primary tumor isassociated with a success rate of >95%. However, in stageIII (lymph node infiltration) and stage IV (distant metas-tasis) disease when melanoma has disseminated to multi-ple organ sites, including brain, lungs, liver, and bone,surgical interventions are of limited use. Moreover,systemic therapy has minimal success because of the highintrinsic resistance exhibited by melanoma cells againstvarious anticancer therapies.A large number of genetic, functional, and biochemical

studies suggest that melanoma cells become resistant tochemotherapy by exploiting their intrinsic resistance toapoptosis (2). Alterations that contribute to the develop-ment of resistance to apoptosis can enable cancer cells notonly to survive under stressful conditions (e.g., duringmetastasis) but also to develop resistance to drugs (3, 4).Therefore, elucidation of novel proteins and pathways thatcontribute to the development of resistance to apoptosismay reveal promising molecular targets for effectivetreatment of melanomas.We have found previously that development of the drug

resistance phenotype in several types of cancer cells isassociated with increased expression of tissue transgluta-minase (TG2; refs. 5–9). TG2 (EC2.3.2.13), also called thecytosolic, type II, or liver transglutaminase, is a uniquemember of the transglutaminase enzyme family. Inaddition to catalyzing the calcium-dependent proteincross-linking reactions (10–12), TG2 can catalyze calcium-independent hydrolysis of GTP and ATP (10–12), proteindisulfide isomerase reactions (13, 14), and serine/threoninekinase activity (15–17). Although, it is predominantly acytosolic protein, TG2 also can be secreted outside the cell(17, 18), can translocate to the nucleus with the help ofimportin-a3 protein (19), and can be expressed on the cellmembrane in association with h members of the integrinfamily of proteins (20–22). Cell surface TG2 promotesadhesion and spreading of cells, enhances focal adhesions,and amplifies adhesion-dependent phosphorylation offocal adhesion kinase (5, 20, 21).The integrins, in association with which TG2 can be

expressed on the cell surface, can strongly influence theability of neoplastic cells to migrate, proliferate, undergoapoptosis, and mediate invasion and metastasis (20–26).Integrins differ from other cell surface receptors in that

Received 2/14/06; accepted 4/13/06.

Grant support: PSOCA 093459 SPORE grant in melanoma from the NIH.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.

Requests for reprints: Kapil Mehta, Department of ExperimentalTherapeutics, The University of Texas M.D. Anderson Cancer Center,Unit 362, 1515 Holcombe Boulevard, Houston, TX 77030.Phone: 713-792-8140; Fax: 713-745-4167.E-mail: [email protected]

Copyright C 2006 American Association for Cancer Research.

doi:10.1158/1535-7163.MCT-06-0083

1493

Mol Cancer Ther 2006;5(6). June 2006

Research. on September 2, 2020. © 2006 American Association for Cancermct.aacrjournals.org Downloaded from

http://mct.aacrjournals.org/

they bind their ligands in the extracellular matrix with alow affinity. However, in response to certain stimuli,integrins can cluster (in focal contacts), and their combinedweak affinities then give rise to a spot on the cell surfacethat has enough adhesive capacity (avidity) to promotestable attachment to the extracellular matrix. Alternatively,certain proteins can bind directly to integrins and enhancetheir affinity for the extracellular matrix ligands, therebypromoting cell signaling (27, 28).In the present study, we investigated TG2 expression and

its implications in metastatic melanoma. We showed thatmalignant human melanoma cells and cell lines expresshigh basal levels of TG2. Activation of endogenous TG2 bythe calcium ionophore A23187 led to amassive cross-linkingof cellular proteins and spontaneous apoptosis in these cells.TG2 was also localized on the cell surface of malignantmelanoma cells in association with h1 and h5 integrins. Cellsurface TG2 played an important role in promotingattachment and protecting cells from apoptosis when cellswere cultured on fibronectin or its 110-kDa fragment. Takentogether, these results suggest that activation of constitu-tively expressed TG2 is an effective strategy for selectivelyinducing apoptosis in malignant melanoma cells and thatTG2 expression contribute to the development of chemo-resistance in malignant melanoma cells by exploitingintegrin-mediated cell survival signaling pathways.

Materials andMethodsCell Lines and Patient SamplesMelanoma tissue samples from patients with various

stages of disease were surgically removed from patientsenrolled in institutionally approved trials. Formalin-fixed,paraffin-embedded tissue sections were obtained fromthe Melanoma and Skin Cancer Core Laboratory of TheUniversity of Texas M.D. Anderson Cancer Center (Hous-ton, TX) for use in immunohistochemical labeling of tumortissues.Metastatic melanoma cell lines, A375 and A375-S2, were

obtained from American Type Culture Collection (Mana-ssas, VA). Primary melanoma cell lines, WM35 (radialgrowth phase) and WM793 (vertical growth phase), werekindly provided by Dr. Robert Kerbel (Sunnybrook HealthScience Center, Toronto, Ontario, Canada). and MeWo(metastatic) cell line was provided by Dr. Elizabeth A.Grimm (The University of Texas M. D. Anderson CancerCenter). All the cell lines were maintained in a log phaseof cell growth by culturing in RPMI 1640 supplementedwith 10% FCS, 0.2% normocin (Invivogen, San Diego, CA),2 mmol/L L-glutamine, and 10 mmol/L HEPES at 37jC inthe CO2 incubator. Normal human epidermal melanocyteswere obtained from Clonetics Corp. (San Diego, CA) andcells were maintained according to the manufacturer’sinstructions. Samples of formalin-fixed, paraffin-embeddedhuman benign nevi were retrieved from the Department ofSurgical Pathology.

Measurement ofTG2 EnzymeActivity in Cell LysatesCells grown to 90% confluence were collected in a

minimal volume (100-300 AL) of buffer A [20 mmol/L

Tris-HCl (pH 7.4) containing 1 mmol/L EDTA, 150 mmol/LNaCl, 14 mmol/L 2-mercaptoethanol, and a 1:100 dilutionof protease inhibitor cocktail (Sigma-Aldrich, St. Louis,MO)] and lysed in the same buffer by probe sonication(8-10 pulses of 10 seconds, each). Protein contents of thecell lysates were determined using dye reagent (Bio-Rad, Richmond, CA). Cell lysates were assayed for TG2activity by determining the Ca2+-dependent incorporationof [3H]putrescine (specific activity, 14.3 Ci/mmol; Amer-sham Pharmacia, San Francisco, CA) into dimethylcaseinas described previously (6). The enzyme activity wasexpressed as nanomoles of putrescine incorporated perhour per milligram of total lysate protein.

Western BlottingCell lysate protein (30 Ag) was separated by SDS-PAGE

on a 7.5% gel and electrophoretically transferred onto anitrocellulose membrane. The membrane was probed withanti-TG2 monoclonal antibody (CUB7401; Neomarkers,Fremont, CA) or anti-caspase-3 antibody (Santa CruzBiotechnology, Santa Cruz, CA). Antigen-antibody reactionwas detected using an enhanced chemiluminescencedetection system (Pierce, Rockford, IL). All the membraneswere stripped and reprobed with an anti-h-actin antibody(Sigma-Aldrich) at a dilution of 1:4,000 to ensure evenloading of proteins in different lanes.

Measurement ofTG2 Activity in Intact CellsTG2 activity in intact cells was determined by preincu-

bating cells with 1 mmol/L 5-(biotinamido) pentylamine(BPA), a competitive inhibitor of TG2-catalyzed cross-linking reactions, in 2% FCS overnight at 37jC. To induceactivation of endogenous TG2, cells were treated with thecalcium ionophore A23187 (4 Amol/L, 4 hours). Cells werethen washed and lysed by sonication in 500 AL buffer A.Equal amounts of cell lysate proteins were fractionated bySDS-PAGE on an 8% gel and electrophoretically transferredonto a nitrocellulose membrane. The membrane wasprobed with horseradish peroxidase (HRP)–conjugatedstreptavidin (Sigma-Aldrich) and then with enhancedchemiluminescence reagent.Similarly, the ability of endogenous cellular proteins to

serve as substrates for endogenous TG2-catalyzed cross-linking reactions was tested as described previously (29).Briefly, cell extracts containing equal amounts (30 Ag) ofproteins were incubated in a total volume of 200 AL bufferA containing 1 mmol/L BPA and 3 mmol/L of eitherCaCl2 or EDTA (background control). The reaction mixture(30 AL) was removed at different time points andimmediately mixed with 3� sample buffer to stop thereaction. Reaction mixtures (equivalent to 30 Ag cellprotein) were fractionated by SDS-PAGE and transferredonto nitrocellulose membrane. The membrane was probedwith HRP-conjugated streptavidin and then with enhancedchemiluminescence reagent.

Measurement of Cell Growth and AttachmentNinety-six-well plates (Corning/Costar, Rochester, NY)

were coated with 20 Ag/mL fibronectin (Sigma-Aldrich) orits 110- or 42-kDa fragment (both fragments were kindlyprovided by Dr. Alexey Belkin, University of Maryland,

TG2 Expression in Melanoma1494




Baltimore, MD) or 0.1% bovine serum albumin (BSA;Sigma-Aldrich) in PBS. The nonspecific binding sites wereblocked with 2% BSA. The cells were grown in T-75 flasksto 80% to 90% confluence isolated by trypsinization,washed once with RPMI 1640, and resuspended at 5 �104 cells/mL in serum-free RPMI 1640. Aliquots (200 AL) ofthe cell suspension were added to fibronectin- or BSA-coated wells in quadruplicate and incubated at 37jC for48 hours. At the end of the incubation period, the numberof viable cells remaining in the well was determined bymeasuring their ability to reduce 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2H-tetrazolium into sol-uble formazan.In some experiments, A375 cells were transfected with

TG2-specific [small interfering RNA (siRNA) 1 and 2] orcontrol (scrambled) siRNAs using RNAiFect reagent(Qiagen Sciences, Germantown, MD) in accordance withthe manufacturer’s instructions. After 48 hours of transfec-tion, cells were harvested with 2 mmol/L EDTA, washedtwice in RPMI 1640, and plated (5 � 103 per 0.2 mL/well inserum-free medium) in quadruplicate to each well of 96-well plates that had been precoated with fibronectin orBSA. After 48 hours of incubation, cells were examinedunder the light microscope and tested for cell viability.For attachment assays, control and siRNA-transfected

A375 cells (2 � 104 per well; 0.2 mL serum-free RPMI1640) were plated in fibronectin- or BSA-coated 96-wellplates. After 1-hour incubation at 37jC, cells were viewedunder microscope for morphology. Nonadherent cells wereremoved by washing with PBS and adherent cells were fixedwith 3.7% paraformaldehyde for 1 hour, washed twice withPBS, and stained with 0.1% crystal violet for 40 minutes. Thestained cells were washed with water and lysed in 0.5%Triton X-100, and the absorbance was read at 540 nm.

Measurement of ApoptosisCells (1 � 106) were seeded into 25-cm2 tissue culture

flasks. Two days later, the cultures were incubated withRPMI 1640 (containing 2% FCS) alone or medium contain-ing A23187. At predetermined time points, cells werevisualized under phase-contrast microscope and photo-graphed. In some experiments, total (floating and adherent)cell populations were collected and centrifuged (800 � g ,5 minutes), and the pellets were immediately processedfor further experimentation. To determine the effect of TG2inhibition on A23187-induced apoptosis, cells were treatedwith 1 mmol/L BPA overnight before treatment withA23187.In some experiments, apoptosis was also determined by

using the ApoAlert Annexin kit (BD Biosciences, SanDiego, CA). Briefly, after appropriate treatment, cells werewashed in PBS and suspended (1 � 106/mL) in bindingbuffer. For each 150 AL cell suspension, 5 AL Annexin V Cy5and 10 AL propidium iodide were added and incubated for15 minutes in the dark at room temperature. Ten thousandevents were counted by flow cytometry.

Flow CytometryCells were detached with 2 mmol/L EDTA, washed, and

resuspended (2 � 106/mL) in PBS containing 0.1% BSA.

Cell suspensions (0.1 mL) were incubated with primaryantibodies specific to various integrins (anti-integrin h1,MAB 1987Z; h3, MAB 1957; h5, MAB 1926; av, MAB 1980; ora5, MAB 1956Z; all from Chemicon, Temecula, CA) or TG2(anti-TG2 monoclonal antibody CUB7401) on ice for 30minutes and then washed twice with ice-cold PBS. Thefluorochrome-labeled secondary antibodies (goat anti-mouseimmunoglobulin G Alexa 546 or goat anti-rabbit IgG Alexa488; both from Molecular Probes, Eugene, OR), whichindicated antigen-antibody reaction, were detected using aFACScan flow cytometer. For the control setting for each celltype, isotopic IgG along with secondary antibody was used.

ImmunoprecipitationCells were lysed in extraction buffer containing 20

mmol/L Tris-HCl (pH 7.6), 100 mmol/L NaCl, 0.5%NP40, and 5 mmol/L EDTA supplemented with proteaseinhibitors. Total cell lysate protein (400 Ag) was preclarifiedby incubation with protein G-Sepharose beads. Thepreclarified lysates were incubated overnight at 4jC withanti-h1 or anti-h5 integrin antibody (Chemicon). The nextday, antigen-antibody complexes were removed by incu-bating the solutions with either anti-mouse IgG or anti-rabbit IgG (1 hour at 4jC) and then protein G-Sepharosebeads (1-3 hours at 4jC). Beads were washed with theextraction buffer and eluted with 2� sample buffer. Boundproteins were analyzed with Western blotting. To detectTG2 in the immunoprecipitates, membranes were firstprobed with anti-TG2 (M-300; Neomarkers) antibody andthen with anti-mouse IgG-HRP. The membrane was thenstripped and reprobed with anti-h1 or anti-h5 integrinantibodies.

Immunofluorescence StainingCells (2� 105) were cultured on glass coverslips in six-well

plates, rinsed thrice with PBS, fixed with 3.7% paraformal-dehyde for 15 minutes, and blocked with 5% normal goatserum for 1 hour. The cells were immunostained by usingprimary antibodies specific to various integrins and TG2.Goat anti-mouse IgG Alexa 488 or Alexa 546 or goat anti-rabbit IgGAlexa 488 or Alexa 546 was used as the secondaryantibody. The stained coverslips were mounted on glassmicroscope slides in mounting medium (80% glycerol plus20% PBS). Images were taken under a Nikon Eclipsefluorescence microscope (Melville, NY) using MetaFluorsoftware (Universal Imaging Corp., Downingtown, PA).Similarly, paraffin-embedded tissue sections from select-

ed tumor samples were deparaffinized and immunostainedwith primary rabbit anti-TG2 and mouse anti-h1 or anti-h5integrin antibodies. Goat anti-mouse IgG Alexa 488 andgoat anti-rabbit IgG Alexa 546 were used as the secondaryantibodies. Stained sections were mounted with mountingmedium and viewed under a light microscope (Nikon).Appropriate controls (mouse and rabbit IgG in place of theprimary antibodies and either primary antibody alonealong with both of the secondary antibodies) were includedto determine the specificity of the reaction.

ImmunohistochemistryFormalin-fixed, paraffin-embedded tumor sections (5 Am

thick) were immunostained for TG2. Briefly, after antigen

Molecular Cancer Therapeutics 1495




retrieval, tissue sections were incubated with anti-TG2monoclonal antibodies overnight at 4jC and then incubatedfor 30 minutes each with biotinylated secondary antibodyand peroxidase-labeled streptavidin. Antigen-antibodyreactions were detected by exposure to 3,3¶-diaminobenzi-dine and hydrogen peroxide chromogen substrate (VectorLaboratories, Burlingame, CA). Slides were counterstainedwith hematoxylin and mounted. The negative controlswere incubated with nonimmune mouse IgG in place ofthe primary antibody.

ResultsTG2 Expression in Normal Melanocytes andMelano-ma Cell LinesWe first determined the TG2 expression in epidermal

melanocytes isolated from normal skin and in fivemalignant human melanoma cell lines representing variousstages of disease progression: WM793 (vertical growthphase), WM35 (early stage), MeWo (metastatic), A375-S2(metastatic), and A375 (highly metastatic amelanotic). Themelanocytes isolated from normal human skin showedcomplete lack of TG2 as determined by Western blotting(Fig. 1A, lane 6) and enzymatic activity (Fig. 1B). Similarly,WM35 cells representing early-stage disease (radial growthphase) showed little TG2 expression (Fig. 1A and C) orenzyme activity (2.6 F 0.4 nmol/h/mg). A375 cells, on theother hand, representing highly metastatic amelanoticmalignant melanoma, showed a 20- to 25-fold increase inTG2 expression or enzyme activity compared with normalepidermal melanocytes and WM35 cells (Fig. 1). The othercell lines representing advanced-stage and metastaticmelanoma also had high TG2 expression and enzymeactivity.

TG2 Expression and Sensitivity to Chemotherapy inMelanoma Cell LinesNext, we determined whether the level of expression of

TG2 in melanoma cell lines corresponded with theirsensitivity to the cytotoxic effects of cisplatin and dacarba-zine, the two most commonly used drugs for treatingmalignant melanoma. There was no significant correlationbetween TG2 levels and sensitivity of melanoma cell linesto cisplatin (Fig. 2A). However, TG2-low WM35 cells weremore sensitive to dacarbazine than were the TG2-express-ing cell lines (Fig. 2B). In general, all five cell lines testedexhibited relative high resistance to both drugs comparedwith the responses of other cancer cell lines, such as MCF-7breast cancer cells (data not shown).

Figure 1. TG2 expression and enzyme activity in normal melanocytesand melanoma cell lines. A, TG2 expression was determined by Westernblotting using the anti-TG2 monoclonal antibody CUB7401 as a probe. Thenitrocellulose membrane was stripped and reprobed with anti-h-actinantibody to ensure even loading of lanes. Cell extracts (30 Ag each) wereloaded as follows: lane 1, A375; lane 2, A375-S2; lane 3, MeWo; lane 4,WM35; lane 5, WM793; lane 6, normal melanocytes. B, TG2 enzymeactivity was determined in the cell extracts by studying Ca2+-dependentincorporation of [3H]putrescine into dimethylcasein as described inMaterials and Methods. Columns, mean of six values from twoindependent experiments; bars, SD. C, immunofluorescence microscopyimages of WM35 and A375 cells immunostained with TG2-specific antiIgG1.

Figure 2. Sensitivity of melanoma cells to cisplatin and dacarbazine. Inquadruplicate, wells in 96-well plates containing 2,000 cells per well in0.2 mL RPMI 1640 were either left untreated or treated with the indicatedconcentrations of cisplatin (A) or dacarbazine (B). Forty-eight hours aftertreatment, viable cells remaining in wells were determined by 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2H-tetrazolium assay,and the percentage viability was calculated. Experiments were repeatedat least three times with similar results. Points, mean of quadruplicatevalues from a representative experiment; bars, SD.





ActivationofTG2 as aTarget for InducingApoptosis inMelanoma CellsBecause it is well known that activation of endogenous

TG2 can induce apoptosis in various cell types (10–12, 30),we next determined whether activation of endogenous TG2could serve as a strategy for killing malignant melanomacells. Treatment with calcium ionophore A23187, whichactivates TG2, induced significant cell killing in all the celllines as determined by 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2H-tetrazolium assay (Fig. 3A).The TG2-rich cell lines A375 and A375-S2 were moresensitive to A23187-induced cell killing than the TG2-low

WM35 cells (Fig. 3A). The morphologic changes induced byA23187 seemed to be characteristic of changes seen in cellsundergoing apoptosis and included condensed nuclei,ruffled membranes, and appearance of prominent apoptoticbodies (Fig. 3B). The A23187-induced changes were muchmore pronounced in TG2-rich A375 cells than in TG2-lowWM35 cells (Fig. 3B).To determine the nature of the cell death induced by

A23187 in A375 and WM35 cells, we studied Annexin Vlabeling of cells treated with A23187 (4 Amol/L, 8 hours).The results (Fig. 4A) indicated that cell death in response toA23187 was apoptotic and that TG2-rich A375 cells were

Figure 4. Calcium ionophore A23187-induced apoptosis in A375 and WM35 melanoma cells. A, cells grown to 80% confluence were incubated ineither medium alone or medium containing 4 Amol/L A23187. Twenty h later, cells were harvested and stained for Annexin V and analyzed by flowcytometry to determine the number of apoptotic cells as described in Materials and Methods. B, cells were left untreated or treated with 4 or 8 Amol/LA23287 for 20 h. After the treatment, cells were harvested and analyzed for Annexin V staining by flow cytometry. Representative experiment repeatedtwice with <20% SD. C, after 24- and 48-h treatment with 2 Amol/L A23187, cells were harvested and the cell lysates (30 Ag/lane) were analyzed forcaspase-3 activation by Western blotting. Cells incubated in medium alone for 48 h (�) served as control. The nitrocellulose membrane was treated withstrip buffer and reprobed with anti-h-actin antibody to ensure even loading of proteins in different lanes.

Figure 3. Effect of treatment withthe calcium ionophore A23187 onmalignant melanoma cells. A, dose-dependent cytotoxicity of A23187against malignant melanoma cellsas determined by 3-(4,5-dimethylth-iazol-2-yl)-5-(3-carboxymethoxy-phenyl)-2H-tetrazolium assay after48 h of treatment. Points, mean ofquadruplicate values from a repre-sentative experiment; bars, SD.Experiments were repeated at leasttwice with similar results. B, photo-micrographs showing morphologicchanges induced in response to treat-ment of WM35 and A375 melanomacells for 24 and 48 h with A23187(2 Amol/L).





significantly more susceptible to A23187-induced apoptosisthan were TG2-low WM35 cells (Fig. 4A and B). TheAnnexin V labeling data also correlated with caspase-3activation; under identical conditions, caspase-3 activationwas more rapid and more pronounced in A375 cells than inWM35 cells (Fig. 4C).Next, we determined whether A23187-induced apoptosis

in A375 cells was related to activation of endogenous TG2.To test this, we first looked for cellular proteins in A375 andWM35 cell lysates that could serve as substrates forendogenous TG2. Equal amounts of cell proteins wereincubated in the presence of BPA (a competitive inhibitor ofTG2-catalyzed protein cross-linking reactions) in thepresence of either 5 mmol/L Ca2+ or EDTA (backgroundcontrol). Reaction mixtures were subjected to immunoblot-ting and the membranes were probed with HRP-streptavi-din as described in Materials and Methods. The resultsshowed numerous proteins that in a calcium-dependentmanner could serve as substrates for endogenous TG2in both cell types (Fig. 5A). MCF-7 cells that lack TG2expression (6) did not show any BPA labeling in presenceor absence of calcium (Fig. 5A, lanes 1 and 4). Moreimportantly, treatment of A375 and WM35 cells withA23187 in the presence of BPA resulted in significantlabeling of cellular proteins (Fig. 5B). Cells treated withBPA alone in the absence of A23187 failed to show anylabeling of the cellular proteins, suggesting that the labelingof cellular proteins is mediated by activated TG2 inresponse to increased calcium levels induced by A23187treatment. In general, A23187-induced incorporation ofBPA was much more pronounced in TG2-rich A375 cellsthan in TG2-low WM35 cells (Fig. 5B).To confirm direct involvement of endogenous TG2 in

A23187-mediated killing of A375 cells, we next studiedthe effect of BPA on A23187-induced apoptosis. A375 andWM35 cells were incubated with 4 Amol/L A23187 in thepresence or absence of 1 mmol/L BPA for 20 hours andanalyzed for accumulation in sub-G1 phase of cell cycle(apoptotic fraction) using flow cytometry. Treatmentwith A23187 resulted in f32% apoptosis in A375 cells(Fig. 5C). However, the presence of BPA during A23187treatment significantly attenuated the extent of apoptosisin these cells (P < 0.001). These results suggested thatTG2-catalyzed protein cross-linking reactions play anessential role in execution of A23187-induced apoptosis.Interestingly, in TG2-low WM35 cells, the extent ofA23187-induced apoptosis was significantly less than inA375 cells and BPA failed to rescue these cells fromapoptosis (Fig. 5C).

TG2 Expression inTumor Samples from Patients withMelanomaNext, we determined the status of TG2 expression in

tumor samples from a small cohort of patients (n = 12)with various stages of tumor progression. In normalskin, TG2 expression was restricted to the basal layeronly (Fig. 6A), whereas both primary (Fig. 6B) andmetastatic (Fig. 6C) melanomas showed high levels ofTG2 expression.

Figure 5. A23187-induced apoptosis results from the activation ofendogenous TG2. A, endogenous protein substrates in MCF-7 (lanes 1and 4 ), A375 (lanes 2 and 5), and WM35 (lanes 3 and 6) cells wereanalyzed by incubating total cell lysate proteins in the presence of 1 mmol/LBPA and 5 mmol/L CaCl2 (lanes 1–3) or EDTA (lanes 4–6). At theindicated times, 3� sample buffer was added to stop the reaction, andthe reaction mixtures (30 Ag/lane) were subjected to SDS-PAGE on 8%gel and analyzed for TG2-catalyzed conjugation of BPA into proteins byimmunoblotting using HRP-conjugated streptavidin as a probe. B, in situactivation of TG2 was studied by preincubating A375 (lanes 1 and 3 )and WM35 (lanes 2 and 4) cells with 1 Amol/L BPA for 12 h. A23187(4 Amol/L) was then added (lanes 3 and 4) to cultures 4 h beforeharvest, and the incorporation of BPA into proteins was determined byimmunoblotting using HRP-conjugated streptavidin. C, A375 and WM35cells were incubated overnight in medium alone or medium containing1 mmol/L BPA. The cultures were either continued under the sameconditions or treated with 4 Amol/L A23187 for an additional 20 h. Atthe end of treatment, cells were harvested and analyzed by flowcytometry for accumulation in the sub-G1 phase of cell cycle asdescribed in Materials and Methods.





Association ofTG2with Cell Surface IntegrinsBecause it has been shown previously that TG2 can be

expressed on the cell membrane in association with hmembers of the integrin family of proteins (5, 20–22),we wished to determine whether that was the case forTG2 in melanoma cells. We first compared the cell sur-face expression of various integrins in WM35 and A375cells. The surface expression of h1 integrin on TG2-richA375 and TG2-low WM35 cells was similar (Fig. 7).However, the expression of other integrins was consid-erably higher (2- to 5-fold) in A375 cells than in WM35cells (Fig. 7). In addition, the surface expression of TG2was much stronger in A375 cells than in WM35 cells(Fig. 7).Based on these results and the results published earlier

(5, 20, 21), we hypothesized that TG2 may closely associatewith integrins in A375 cells. To test this contention,immunoprecipitates from A375 and WM35 cells wereisolated using anti-h1 or h5 integrin antibody and testedfor the presence of TG2 protein. The results clearlyestablished that anti-h1 and anti-h5 integrin antibodyeffectively pulled down TG2 protein in addition tointegrins (Fig. 8A and B). These results suggested thatTG2 protein is closely associated with h integrins on cellsurface membranes of A375 cells.Association of TG2 with h1 and h5 integrins was further

supported by colocalization studies using fluorescencemicroscopy. As shown in Fig. 8C, TG2 (green fluores-cence) colocalized with h1 and h5 integrins (red fluores-cence) in A375 cells as evidenced by the yellowfluorescence in the merged images. Similarly, TG2colocalized with these integrins in a limited number ofpatient samples (n = 3) that were tested for this purpose(Fig. 8D). Previously, we observed a similar association ofTG2 with h1 and h5 integrins in drug-resistant MCF-7breast cancer cells (5).Because TG2 has high binding affinity for the 42-kDa

gelatin-binding domain of fibronectin (31, 32) and is closelyassociated with integrins on the cell surface, we nextexamined whether TG2 expression could promote fibro-nectin-mediated cell attachment and signaling in malignant

melanomas. Indeed, incubation of TG2-positive A375 cellson fibronectin-coated surfaces resulted in stronger adher-ence than did incubation of A375 cells on BSA-coatedsurfaces (Fig. 9A). WM35 cells, which express low levels ofTG2 protein, showed weak attachment to both fibronectin-and BSA-coated surfaces (data not shown). The 110-kDafragment of fibronectin promoted strong adherence of A375cells (Fig. 9A) but not of WM35 cells (data not shown). The42-kDa fragment of fibronectin, however, failed to supportthe attachment of either A375 or WM35 cells; adherence ofcells on surfaces coated with 42-kDa fragment was similar

Figure 6. TG2 expression in normal human skin (A) and primary (B) and metastatic (C) melanoma tumor samples. Paraffin-embedded tissues frommelanoma patients were processed for immunohistostaining as described in Materials and Methods. Representative results.

Figure 7. Expression of TG2 and integrins on the surface of A375 andWM35 cells. Flow cytometry was used to determine the cell surfaceexpression of the h1, h3, h5, av, and a5 integrins and TG2 protein. Shadedhistograms, fluorescence intensity of cells incubated with isotypic controlIgG; open histograms, fluorescence intensity of the indicated integrin orTG2 as revealed by immunostaining of cells with specific antibodyfollowed by fluorochrome Alexa 488–conjugated secondary antibody asdescribed in Materials and Methods.





to that on BSA-coated control (Fig. 9A; data not shown).Two antibodies against TG2 (CUB7401 and rabbit poly-clonal antibody) did not affect the attachment of A375 orWM35 cells on surfaces coated with fibronectin or the 110-or 42-kDa fragments thereof (data not shown). In contrast,the function blocking anti-h1 integrin (JB1A) stronglyblocked the adhesion of A375 cells to fibronectin and its110-kDa fragment (data not shown).To further analyze the involvement of TG2 in fibronectin-

mediated cell attachment, we used a siRNA-basedapproach (5). We first tested the ability of siRNAs todown-regulate TG2 protein in A375 cells. The transfectionefficiency of siRNAs as determined by fluorescein-labelednonspecific siRNA was consistently 70% to 90%. Thetransfection with control siRNA (scrambled) had noappreciable effect on TG2 level (Fig. 9B). However, specificknockdown of TG2 with individual siRNAs (siRNA1 andsiRNA2) suppressed the protein level by >70% (Fig. 9B).We then analyzed the effect of TG2 knockdown onfibronectin-mediated cell attachment. Inhibition of TG2 bysiRNA effectively blocked the fibronectin-mediated attach-ment and spreading of A375 cells (Fig. 9C and D). No sucheffects were observed when the cells were transfected withcontrol nonspecific (scrambled) siRNA. These resultssuggested that TG2 plays an important role in promotingthe attachment and spreading of A375 melanoma cells tofibronectin-coated surfaces.In a parallel experiment, untreated and siRNA-trans-

fected A375 cells were continued for an additional 48 hoursin serum-free medium on fibronectin- and BSA-coatedplates. After 48 hours, the ability of TG2 and fibronectin to

support cell growth and cell survival in the presence ofserum-free medium was tested by determining the numberof viable cells remaining in wells by crystal violet staining.TG2-rich A375 cells cultured on fibronectin-coated surfacescould survive and grow effectively under serum-freeconditions (Fig. 10). However, on BSA-coated surfaces,these cells failed to survive (data not shown). Moreimportantly, knockdown of TG2 in A375 cells with siRNAsmarkedly reduced the survival and growth of these cellseven on fibronectin-coated surfaces (Fig. 10). These resultssuggested that TG2-dependent interaction between malig-nant melanoma cells and fibronectin is critical for inducingcell survival and cell growth signaling.

DiscussionThis study shows that expression of TG2 is up-regulatedduring advanced stages of malignant melanomas. TG2expression promotes cell attachment and integrin-mediat-ed cell survival signaling on fibronectin-coated surfaces.Importantly, activation of endogenous TG2 results inrapid apoptosis, indicating that activation of endogenousTG2 could serve as a strategy for killing malignantmelanomas.Depending on the cell type and the location of TG2

within the cell, TG2 can serve as a proapoptotic or anantiapoptotic protein (7, 33–36). Although predominantly acytosolic protein, TG2 can also localize in the nucleus (19)where it seems to associate with a variety of proteins, suchas pRb, p53, and histones, and can regulate cellularfunctions (15–17). In association with the h subunits of

Figure 8. TG2 closely associateswith integrins h1 and h5 in malignantmelanoma cells. Cell extracts fromA375 and WM35 cells (400 Agprotein each) were immunoprecipi-tated with anti-integrin h1 (A) or h5(B) antibody. The immunoprecipitate(lanes 3 and 4) and total cell extracts(30 Ag protein; lanes 1 and 2) fromA375 and WM35 cells were sub-jected to immunoblotting and probedwith anti-TG2 antibody. The mem-branes were stripped and reprobedwith either anti-h1 (A, right ) or anti-h5 (B, right ) antibodies. A375 cells(C) or tissue samples from melanomapatients (D) were incubated withanti-TG2 and either anti-h1 or anti-h5 integrin antibody and then withAlexa 546– tagged anti-rabbit IgG(red) and Alexa 488–tagged anti-mouse IgG (green ). The two-colorimages were obtained with a fluores-cence microscope using MetaFluorsoftware.





the integrin family of proteins, TG2 can localize to the cellmembrane and facilitate adhesion, spreading, and motilityof cells (22, 37, 38). It is estimated that all TG2 on the cellsurface is present in a 1:1 complex with integrins (20, 21).We recently reported that drug-resistant and metastatic

breast cancer cells exhibit high levels of TG2 (5–9, 39).Although the general consensus is that drug resistance andmetastasis represent different phenotypes, it is well knownthat increased resistance to apoptosis is an importantfeature of both phenotypes (2–4). In view of this, it istempting to speculate that high basal expression of TG2 inmalignant melanomas promotes integrin-mediated signal-ing that affects not only the cell-adhesive, migratory, andinvasive functions of these cells but also their survival andgrowth. Indeed, in the study reported here, we found thatpresence of TG2 promoted strong attachment of melanomacells to fibronectin and its 110-kDa fragment. The attach-ment of cells could be effectively blocked by knockingdown TG2 expression using a siRNA approach (Fig. 9).These results suggest that TG2 expression promotes stableinteraction between integrins and fibronectin, the majorprotein in the extracellular matrix that plays an importantfunction in inducing cell growth and cell survival signaling(40). Several previous studies have underscored this critical

role for fibronectin. For example, culture of a5h1 integrin-expressing cells on fibronectin is associated with increasedexpression of the antiapoptotic protein Bcl-2 and protectionof cells from apoptosis in response to various stresses (41)Similarly, several cancer cell lines have been shown to bemore resistant to chemotherapy- and radiation-induced celldeath when they are cultured on fibronectin-coatedsurfaces (42–44).Based on these observations and the observation that TG2

is closely associated with integrins h1 and h5 in melanomacells (Fig. 8), it is tempting to speculate that TG2 expressioncontributes to chemoresistance and radiation resistance inmalignant melanomas. Indeed, we found that knockdownof TG2 by siRNA in A375 cells strongly influenced theirability to attach to the fibronectin-coated surfaces and tosurvive under serum-free conditions (Fig. 10). In recentyears, evidence supporting a role of TG2 in protecting cellsfrom apoptosis has been emerging (7, 33–36). For example,treatment of breast cancer cells with epidermal growthfactor was shown to induce TG2 expression that in turnrendered the cells resistant to chemotherapeutic drugs (45).Previous studies by our group have shown that irrespectiveof the type and source of cells development of the drugresistance phenotype in cancer cells is associated with

Figure 9. Attachment and spreading of A375 cells on fibronectin-coated surfaces is mediated by TG2. A, A375 cells were seeded in 96-well platescoated with BSA, fibronectin (Fn ), or 110- or 42-kDa fragments of fibronectin (2 � 104 per well; 0.2 mL serum-free medium). After 1-h incubation, cellswere analyzed for attachment after washing and staining with crystal violet as described in Materials and Methods. Columns, averages of six values fromtwo independent experiments; bars, SD. B, A375 cells were transfected with TG2-specific siRNAs (siRNA1 and siRNA2) or control (scrambled) siRNA.After 48 h, cells were harvested and analyzed for TG2 enzyme activity and TG2 protein levels (inset ; lane 1, untreated; lane 2, control siRNA; lane 3,siRNA1; lane 4, siRNA2). Untreated WM35 cells were used as control. C, untreated (1) and siRNA-transfected (2, control siRNA; 3, siRNA1; 4, siRNA2)A375 cells were incubated in quadruplicate on fibronectin- or BSA-coated plates. After 1-h incubation, nonadherent cells were removed and adherent cellswere examined under light microscope or stained with crystal violet (D) for quantitative analysis.





increased TG2 expression (5–9, 39). Importantly, we foundin an earlier study that knocking down TG2 protein byTG2-specific siRNA restored sensitivity to doxorubicin indrug-resistant MCF-7 breast cancer cells (5). A similarreversal in sensitivity to doxorubicin was noted by Han andPark (46) in drug-resistant PC-14 lung cancer cells inresponse to TG2 inhibition by an antisense approach. Theseobservations strongly suggest that TG2 expression cancontribute to the development of chemoresistance. Becausemost chemotherapeutic drugs are known to kill cancer cellsby inducing apoptosis (47–49), it is likely that TG2 confersdrug resistance by up-regulating prosurvival and anti-apoptotic pathways. Indeed, culture of TG2-positive cellson fibronectin-coated surfaces has been shown previouslyto induce strong activation of the focal adhesion kinase(5, 20, 21, 50), an upstream event that leads to the activationof various downstream antiapoptotic and cell survivalsignaling pathways (51, 52).In addition to its role in protecting cells from apoptosis

and conferring chemoresistance, TG2 has also been shownto have a proapoptotic role (12–15). The proapoptoticfunctions of TG2 are linked to the ability of TG2 toirreversibly cross-link proteins in the presence of Ca2+. It islikely that under extremely stressful conditions massive

release of Ca2+ from intracellular stores or influx of Ca2+

from outside the cell leads to activation of TG2 to its cross-linking configuration resulting in post-translational modi-fication of key proteins and onset of apoptosis. Indeed, inthe study reported here, we found that TG2-expressingA375 cells exhibited resistance to chemotherapeutic drugs(Fig. 2) but responded well to treatment with the calciumionophore A23187 (Fig. 3). The A23187-induced cell deathof A375 cells was apoptotic (Figs. 4B and 5) and was due tothe protein cross-linking activity of TG2 (Fig. 6A). Thepresence of BPA, a competitive inhibitor of TG2, effectivelyblocked A23187-induced apoptosis in A375 cells (Fig. 6C).Based on these results, we propose that high levels of

TG2 expression in malignant melanoma cells can confer adrug resistance phenotype by promoting integrin-mediatedcell attachment and cell survival signaling pathways. Onthe other hand, high expression of TG2 in melanoma cellscan be exploited as a potential target to kill these hard totreat cancer cells. Our results also indicate that activation ofendogenous TG2 could serve as a strategy for inducingapoptosis in malignant melanomas, indicating that highexpression of TG2 in melanoma cells can be exploited.

Acknowledgments

We thank David M. Harris for the technical support and Stephanie P.Deming for editorial help.

References

1. Stewart BW, Kleihues P. Human cancers by organ sites. WorldCancer Report. In: Stewart BW, Kleihues P, editors. Lyon: IARC; 2003.p. 181–270.

2. Soengas MS, Lowe SW. Apoptosis and melanoma chemoresistance.Oncogene 2003;22:3138–51.

3. Fesik WF. Promoting apoptosis as a strategy for cancer drug discovery.Nat Rev Cancer 2005;5:876–85.

4. Krebel RS, Kobayashi H, Graham CH. Intrinsic or acquired drugresistance and metastasis: are they linked phenotypes? J Cell Biochem1994;56:37–47.

5. Herman J, Mangala SL, Mehta K. Implications of increased tissuetransglutaminase expression in drug resistant breast cancer (MCF-7) cells.Oncogene 2005;25:3049–58.

6. Chen JS, Agarwal N, Mehta K. Multidrug-resistant MCF-7 breastcancer cells contain deficient intracellular calcium pools. Breast CancerRes Treat 2002;71:237–47.

7. Mehta K, Fok JY, Mangala LS. Tissue transglutaminase: from biologicalglue to cell survival cues. Front Biosci 2006;11:163–85.

8. Chen J, Kanopleva M, Multani A, Pathak S, Mehta K. Drug resistantbreast cancer MCF-7 cells are paradoxically sensitive to apoptosis. J CellPhysiol 2004;200:223–34.

9. Mangala SL, Mehta K. Tissue transglutaminase in cancer biology. ProgExp Tumor Res 2005;38:125–38.

10. Fesus L, Piacentini M. Transglutaminase 2: an enigmatic enzyme withdiverse functions. Trends Biochem Sci 2002;27:534–9.

11. Chen J, Mehta K. Tissue transglutaminase: an enzyme with a splitpersonality. Int J Biochem Cell Biol 1999;31:817–36.

12. Lorand L, Graham RM. Transglutaminases: crosslinking enzymes withpleiotropic functions. Nat Rev Mol Cell Biol 2003;4:140–56.

13. Chandrashekar R, Tsuji N, Morales T, Ozols V, Mehta K. Cloning of anovel transglutaminase from dog heart worm with protein disulfideisomerase activity. Proc Natl Acad Sci U S A 1998;95:531–6.

14. Hasegawa G, Suwa M, Ichikawa Y, et al. A novel function of tissuetransglutaminase: protein disulfide isomerase. Biochem J 2003;373:793–803.

15. Mishra S, Murphy LJ. Tissue transglutaminase has intrinsic kinase

Figure 10. Fibronectin-mediated attachment and survival of A375 cellsis dependent on TG2 expression. Control and siRNA-transfected A375cells were incubated in serum-free medium on fibronectin-coated plates.After 48-h culture, cells were examined under a light microscope (A) oranalyzed for cell viability by 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxyme-thoxyphenyl)-2H-tetrazolium assay (B).





activity: identification of transglutaminase 2 as an insulin-like growthfactor-binding protein-3 kinase. J Biol Chem 2004;279:23863–8.

16. Mishra S, Murphy LM. The p53 oncoprotein is a substrate for tissuetransglutaminase kinase activity. Biochem Biophys Res Commun 2006;339:726–30.

17. Mishra S, Saleh A, Espino PS, Davie JR, Murphy LJ. Phosphorylationof histones by tissue transglutaminase. J Biol Chem 2006;281:5532–8.

18. Greenberg CS, Birckbichler PJ, Rice RH. Transglutaminases: multi-functional crosslinking enzymes that stabilize tissues. FASEB J 1992;5:3070–7.

19. Peng X, Zhang Y, Zhang H, et al. Interaction of tissue trans-glutaminase with nuclear transport protein importin-a3. FEBS Lett 1999;446:35–9.

20. Akimov SS, Belkin AM. Cell-surface transglutaminase promotesfibronectin assembly via interaction with the gelatin-binding domain offibronectin: a role in TGFh-dependent matrix deposition. J Cell Sci 2001;114:2989–3000.

21. Akimov SS, Krylov D, Fleishmann LF, Belkin AM. Tissue trans-glutaminase is an integrin-binding adhesion coreceptor for fibronectin.J Cell Biol 2000;148:825–38.

22. Zemskov EA, Janiak A, Hang J, Waghray A, Belkin AM. The role oftissue transglutaminase in cell-matrix interactions. Front Biosci 2006;11:1057–76.

23. Jin H, Varner J. Integrins: roles in cancer development and astreatment targets. Br J Cancer 2004;90:561–5.

24. Giancotti FG, Ruoslahti E. Integrin signaling. Science 1999;285:1028–32.

25. Miranti CK, Brugge JS. Sensing the environment: a historicalperspective on integrin signal transduction. Nat Cell Biol 2002;4:83–90.

26. Guo W, Giancotti HG. Integrin signaling during tumor progression. NatRev Mol Cell Biol 2004;5:816–26.

27. Shimizu Y, Rose DM, Ginsberg MH. Integrins in the immune system.Adv Immunol 1999;72:325–80.

28. Howe A, Aplin AE, Alahari SK, Juliano RL. Integrin signaling and cellgrowth control. Curr Opin Cell Biol 1998;10:220–31.

29. Singh RN, McQueen T, Mehta K. Detection of the amine proteinsubstrates of transglutaminase with 5(biotinamido) pentylamine. AnalBiochem 1995;231:261–3.

30. Mehta K. Mammalian transglutaminases: a family portrait. Prog ExpTumor Res 2005;38:1–18.

31. Hang J, Zemskov EA, Lorand L, Belkin AM. Identification of a novelrecognition sequence for fibronectin within the NH2-terminal h-sandwichdomain of tissue transglutaminase. J Biol Chem 2004;280:23675–83.

32. Jeong JM, Murthy SNP, Radek JT, Lorand L. The fibronectin-bindingdomain of transglutaminase. J Biol Chem 1995;270:5654–8.

33. Milakovic T, Tucholski J, McCoy E, Johnson JVW. Intracellularlocalization and activity state of tissue transglutaminase differentiallyimpacts cell death. J Biol Chem 2004;279:8715–22.

34. Antonyak MA, Singh US, Lee DA, Boehm JE, Combs C, Zgola MM.Effects of tissue transglutaminase on retinoic acid-induced cellulardifferentiation and protection against apoptosis. J Biol Chem 2001;276:33582–7.

35. Yamaguchi H, Wang H-G. Tissue transglutaminase serves as an

inhibitor of apoptosis by cross-linking caspase 3 in thapsigargin-treatedcells. Mol Cell Biol 2006;26:569–79.

36. Boehm JE, Singh U, Combs C, Antonyak MA, Cerione RA. Tissuetransglutaminase protects against apoptosis by modifying the tumorsuppressor protein p110 Rb. J Biol Chem 2002;277:20127–30.

37. Akimov SS, Belkin AM. Cell-surface tissue transglutaminase isinvolved in adhesion and migration of monocytic cells on fibronectin.Blood 2001;98:1567–76.

38. Balklava Z, Verderio E, Collighan RJ, Gross SR, Adams J, Griffin M.Analysis of tissue transglutaminase in the migration of Swiss 3T3fibroblasts: the active-state conformation of the enzyme does not affectcell motility but it is important for its secretion. J Biol Chem 2002;277:16567–75.

39. Mehta K, Fok J, Miller FR, Koul K, Sahin AA. Prognostic significanceof tissue transglutaminase in drug resistant and metastatic breast cancer.Clin Cancer Res 2004;10:8068–76.

40. Pankov R, Yamada KM. Fibronectin at a glance. J Cell Sci 2002;115:3861–3.

41. Zhang Z, Vuori K, Reed JC, Ruoslahti E. The a5h1 integrin supportssurvival of cells on fibronectin and up-regulates Bcl-2 expression. Proc NatlAcad Sci U S A 1995;92:6161–5.

42. Korah R, Boots M, Wieder R. Integrin a5h1 promotes survival ofgrowth-arrested breast cancer cells: an in vitro paradigm for breast cancerdormancy in bone marrow. Cancer Res 2004;64:4514–22.

43. Aoudjit F, Vuori K. Integrin signaling inhibits paclitaxel-inducedapoptosis in breast cancer cells. Oncogene 2001;20:4995–5004.

44. Cordes N, Blaese MA, Plasswilm L, Rodemann HP, Van Beuningen D.Fibronectin and laminin increase resistance to ionizing radiation and thecytotoxic drug Ukraine in human tumour and normal cells in vitro . Int JRadiat Biol 2003;79:709–20.

45. Antonyak MA, Miller AM, Jansen JM, et al. Augmentation of tissuetransglutaminase expression and activation by epidermal growth factorinhibit doxorubicin-induced apoptosis in human breast cancer cells. J BiolChem 2004;279:41461–7.

46. Han JA, Park SC. Reduction of transglutaminase-2 expression isassociated with an induction of drug sensitivity in the PC-14 human lungcancer cell line. J Cancer Res Clin Oncol 1999;125:89–95.

47. Debatin K. Activation of apoptosis pathways by anticancer treatment.Toxicol Lett 2000;112:41–8.

48. Hickman JA. Apoptosis and chemotherapy resistance. Eur J Cancer1996;32A:921–6.

49. Gottesman MM. Mechanisms of cancer drug resistance. Annu RevMed 2002;53:615–27.

50. Belkin AM, Tsurupa G, Zemskov E, Veklich Y, Weisel JW, Medved L.Transglutaminase-mediated oligomerization of the fibrin(ogen) aC domainspromotes integrin-dependent cell adhesion and signaling. Blood 2005;105:3561–8.

51. Sonoda Y, Matsumoto Y, Funakoshi M, Yamamoto D, Hanks SK,Kasahara T. Anti-apoptotic role of focal adhesion kinase (FAK): induction ofinhibitor of apoptosis and apoptosis suppression by the overexpression ofFAK in human leukemic cell line, HL-60. J Biol Chem 2000;275:16309–15.

52. McLean GW, Carragher NO, Avizienyte E, Evans J, Brunton VG,Frame MC. The role of focal adhesion kinase in cancer: a new therapeuticopportunity. Nat Rev Cancer 2005;5:505–15.





2006;5:1493-1503. Mol Cancer Ther Jansina Y. Fok, Suhendan Ekmekcioglu and Kapil Mehta malignant melanomaImplications of tissue transglutaminase expression in

Updated version

http://mct.aacrjournals.org/content/5/6/1493

Access the most recent version of this article at:

Cited articles

http://mct.aacrjournals.org/content/5/6/1493.full#ref-list-1

This article cites 47 articles, 19 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/5/6/1493.full#related-urls

This article has been cited by 9 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] at

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

Permissions

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

.http://mct.aacrjournals.org/content/5/6/1493To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/content/5/6/1493.full#ref-list-1

http://mct.aacrjournals.org/content/5/6/1493.full#related-urls

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

mailto:[email protected]



Documents

Implications of tissue transglutaminase expression in malignant … · Implications of tissue transglutaminase expression in malignant melanoma Jansina Y. Fok,1 Suhendan Ekmekcioglu,1