tissuetransglutaminasemediatedtumor stroma interaction … · augments oncogenic signaling in...

Biology of Human Tumors

TissueTransglutaminaseMediatedTumor–StromaInteraction Promotes Pancreatic CancerProgressionJiyoon Lee1, Salvatore Condello2, Bakhtiyor Yakubov2, Robert Emerson3,Andrea Caperell-Grant2, Kiyotaka Hitomi4, Jingwu Xie1,5,6, and Daniela Matei1,2,6,7,8

Abstract

Purpose: Aggressive pancreatic cancer is commonly associatedwith a dense desmoplastic stroma, which forms a protective nichefor cancer cells. The objective of the study was to determine thefunctions of tissue transglutaminase (TG2), a Ca2þ-dependentenzyme that cross-links proteins through transamidation and isabundantly expressed by pancreatic cancer cells in the pancreaticstroma.

Experimental Design: Orthotopic pancreatic xenografts andcoculture systems tested the mechanisms by which the enzymemodulates tumor–stroma interactions.

Results: We show that TG2 secreted by cancer cells effectivelymolds the stroma by cross-linking collagen, which, in turn,

activates fibroblasts and stimulates their proliferation. The stifffibrotic stromal reaction conveys mechanical cues to cancer cells,leading to activation of the YAP/TAZ transcription factors, pro-moting cell proliferation and tumor growth. Stable knockdown ofTG2 in pancreatic cancer cells leads to decreased size of pancreaticxenografts.

Conclusions: Taken together, our results demonstrate that TG2secreted in the tumor microenvironment orchestrates the cross-talk between cancer cells and stroma fundamentally affectingtumor growth. Our study supports TG2 inhibition in the pancre-atic stroma as a novel strategy to block pancreatic cancer progres-sion. Clin Cancer Res; 21(19); 4482–93. �2015 AACR.

IntroductionPancreatic ductal adenocarcinoma (PDA) is the fourth cause of

cancer-related death (1). With the currently available treatment,the majority of patients succumb to disease, resistance to chemo-therapy and radiation having been implicated as pivotal obstacles(2). A hallmark of pancreatic tumors is the dense desmoplasticstroma (DS) composed of myofibroblasts and stellate cells whichsecrete collagens and other matrix proteins that serve as a scaffoldon which tumor cells proliferate (3–5). These stromal compo-nents are highly responsive to cytokines (TGFb) and growthfactors (PDGFs, FGFs, sonic hedgehog) abundantly secreted inthe tumor microenvironment (TME; refs. 4–7). The stroma,

believed to protect PDA cells from environmental stress, is hypo-vascular, hypoxic, and presumably less permeable to chemother-apy (4, 5, 8). In support of the latter concept, a recent study usingclinical PDA specimens demonstrated that the delivery of gemci-tabine into human pancreatic tumors is highly heterogenous andstrongly correlated with the existing stromal reaction (9, 10).However, two other recent reports have suggested that somecomponents of the stroma may play a reverse role, by restrainingcancer cell invasion and preventing tumor metastasis (11, 12).The latter studies used analyses in genetically engineered PDAmouse models that may not faithfully mirror the desmoplasticreaction of human tumors. The conflicting outcomes of previousstudies (8, 12) coupled with the continued lack of success treatingthis recalcitrant disease, underscore the complex interplaybetween the diverse components of pancreatic tumors and theinherent limitations of existing experimental models to study andreplicate the diverse traits of the human disease (13). The datapresented here highlight the cross-talk between cancer and stro-mal cells mediated by tissue transglutaminase (TG2) and thesignificance of the cross-linking reactionmediated by this enzymein the matrix to PDA progression.

TG2 is a TGFb inducible protein (14) secreted in the extracel-lular milieu that mediates the transfer of acyl groups betweenglutamine and lysine residues, causing protein cross-linking (15).Highly expressed in pancreatic cancer cells and tumors, TG2 wasreported to activate NF-kB orchestrated survival mechanisms andto contribute to chemotherapy resistance, independently of itsenzymatic function (16, 17). Whether and how TG2 promotespancreatic tumor growth remains unclear. On the basis of theknowledge that the enzyme is secreted and acts as a potent cross-linking factor for collagens andmatrix proteins (18), and that theenzymatic function of TG2 is augmented in the extracellular

1Department of Biochemistry and Molecular Biology, Indiana Univer-sity School of Medicine, Indianapolis, Indiana. 2Department of Med-icine, Indiana University School of Medicine, Indianapolis, Indiana.3Department of Pathology, Indiana University School of Medicine,Indianapolis, Indiana. 4Department of Basic Medicinal Sciences,Graduate School of Pharmaceutical Sciences, Nagoya University,Nagoya, Japan. 5Department of Pediatrics, Indiana University Schoolof Medicine, Indianapolis, Indiana. 6Indiana University Melvin andBren Simon Cancer Center, Indianapolis, Indiana. 7Department ofObstetrics and Gynecology, Indiana University School of Medicine,Indianapolis, Indiana. 8Richard L. Roudebush VA Medical Center,Indianapolis, Indiana.

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Corresponding Author: Daniela Matei, Indiana University School of Medicine,Joseph E. Walther Hall, Room C218D, 980 W. Walnut Street, Indianapolis, IN46202. Phone: 317-278-0070; Fax: 317-278-0074; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-15-0226

�2015 American Association for Cancer Research.

ClinicalCancerResearch

Clin Cancer Res; 21(19) October 1, 20154482

on July 26, 2021. © 2015 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 3, 2015; DOI: 10.1158/1078-0432.CCR-15-0226

http://clincancerres.aacrjournals.org/

matrix (ECM)by factors such as hypoxia and tissue injury (19),wehypothesized that TG2 secreted by pancreatic cancer cells into thetumormilieu alters the composition of the stroma, which, in turn,augments oncogenic signaling in cancer cells facilitating tumorprogression.

By using pancreatic orthotopic xenografts, we demonstrate thatTG2 knockdown in PDA cells inhibits tumor formation andgrowth. We show that TG2 is enzymatically active in pancreaticcells and tumors, and is secreted in the TME where it cross-linkscollagen and stimulates the proliferation of myofibroblasts. Thestroma enriched in cross-linked collagen and fibroblasts activatesthe Yes-associated protein (YAP) and transcriptional coactivatorwith PDZ-bindingmotif (TAZ) transcription factors in cancer cellsand promotes cell proliferation. This is the first report describingthe role of TG2 in the pancreatic DS and demonstrating its role inthe activation of YAP/TAZ transcription factors in cancer cells.Taken together, our data support that by facilitating the cross-talkbetween cancer and stromal cells, TG2 promotes PDAprogression.

Materials and MethodsChemicals and reagents

Antibodies for TG2 (NeoMarkers), YAP1 (Abnova), p-YAP(Ser127) and YAP/TAZ (Cell Signaling Technology), and GAPDH(Meridian Life Science, Inc.) were used for Western blotting.Recombinant enzymatically active or mutant (C277S) TG2 werepurified as previously described (20). Unless stated otherwise, allreagents were from Sigma-Aldrich.

Cell cultureAsPC1, BxPC3, and Paca2 cellswere obtained from the American

Type Culture Collection, human pancreatic normal epithelial(HPNE) andPanc1 cellswere a gift fromProf.MurrayKorc [IndianaUniversity Melvin and Bren Simon Cancer Center (IUSCC), India-napolis, IN] (21). The normal humandermalfibroblasts (NHF544)were provided byDr. Dan Spandau (IUSCC, Indianapolis, IN), andGFP-expressing neonatal dermal fibroblasts (GFP-HNDFs) werepurchased from Angioproteomie. Human pancreatic stellate cells(hPSC) were from ScienCell Research Laboratories, and the meso-thelial cell line, LP9,was from theNIAAgingCell Repository. All cell

lines were tested forMycoplasma and authenticated by the IUSCC InVivo Therapeutics Core. AsPC1 and BxPC3 cells were cultured inRPMI-1640 medium (Cellgro) supplemented with 10% fetalbovine serum (FBS; Cellgro) and 1% antibiotics. Panc1, Paca2,NHF544, GFP-HNDFs, LP9, and hPSCs were grown in Dulbecco'sModifiedEagleMedium(DMEM;Cellgro) supplementedwith10%FBS and 1% antibiotics. Cells were grown at 37�C under 5% CO2.Conditionedmedia (CM)was collected after 24-hour incubation of5 � 105 PDA cells in serum-free RPMI media. Coculture experi-mental details are provided in the Supplementary Data.

ImmunohistochemistryImmunohistochemistry (IHC) was performed as previously

described (see the Supplementary Data; ref. 22).

Cell transfectionTG2 was stably knocked down in AsPC1 and Panc1 cells by

transducing MISSION shRNA Lentiviral Transduction Particles(Sigma), containing short-hairpin RNAs (shRNA) targeting TG2(shTG2) in the presence of polybrene (8 mg/mL). As control(shCtrl), cells were transducedwithMISSIONpLKO.1-puro Emp-ty Vector control Transfection Particles (Sigma). Both shCtrl andshTG2 particles were transduced at 2.5 multiplicity of infection.Pooled stable clones were collected after selection with puromy-cin at 1mg/mL concentration, andTG2knockdownwas verifiedbyWestern blotting. Transient transfection of short interfering RNA(siRNA) used sequences targeting YAP and TAZ (Santa CruzBiotechnology) and Lipofectamine 3000 (Life Technologies).siGENOME Control Pool (Dharmacon) was used as a control.

Western blot analysisWestern blot analysis followed a standard protocol (see the

Supplementary Data).

Orthotopic pancreatic xenograftsAsPC1 and Panc1 cells (1.5 � 105), stably transfected with

shTG2 or shCtrl were injected into the pancreatic tail of 7- to8-week-old female nudemice (Harlan) using a 27.5-gauge needleunder isoflurane anesthesia. Three (Panc1 cells) or 4 weeks(AsPC1 cells) after injection, mice were euthanized and tumorswere weighted and measured. Tumor volumes were calculated as4/3 � p � (L/2) � (W/2) � (H/2), where L is length, W is width,and H is height. Metastatic implants were counted. Animalexperiments were approved by the IU Animal Care and UseCommittee, in compliance with federal regulations.

ImmunofluorescenceImmunofluorescence (IF) used antibodies for TG2 (Neomar-

kers), a-smooth muscle actin (a-SMA; Sigma), collagen 1(Abcam), and YAP/TAZ (Cell Signaling Technology). To detectmatrix-bound TG2, cells were not permeabilized before incuba-tion with the TG2 antibody, as previously described (23). Proce-dures are detailed in the Supplementary Data.

Proliferation assaysProliferation assays were performed as described in the Sup-

plementary Data.

Reporter assayTo measure YAP/TAZ transcriptional activity, PDA cells

were transfected with pGal4-TEAD4, pGal4-Luciferase, and

Translational Relevance

Aggressive and chemotherapy-resistant pancreatic cancer iscommonly associated with a dense stroma, which forms aprotective niche for cancer cells. We hypothesized that tissuetransglutaminase (TG2), a transamidating enzyme secreted bypancreatic cancer cells, promotes pancreatic cancer progres-sion by protein cross-linking in the stroma. Our data presentthe first evidence that TG2 is abundantly secreted and enzy-matically active in the pancreatic tumor milieu, where it cross-links collagen and stimulates fibroblast proliferation. Wedemonstrate that TG2 orchestrates the development of acharacteristic fibrotic inflammatory reaction that enablestumor growth. Our data point to a novel mechanism involv-ing the tumor microenvironment by which TG2 promotespancreatic tumor progression. The findings have importantconsequences for developing strategies to block TG2 enzymat-ic activity and control pancreatic cancer growth.

TG2 and the Desmoplastic Pancreatic Stroma

www.aacrjournals.org Clin Cancer Res; 21(19) October 1, 2015 4483




pTK-Renilla at a ratio of 1:10:1 using DreamFect (Oz Biosciences,Inc.). Plasmids were a gift from Dr. Clark Wells (IUSCC, India-napolis, IN; ref. 24). The Dual-Luciferase Assay System Kit (Pro-mega) determined TEAD4 reporter activity after 48 hours oftransfection. To control for transfection efficiency, the luciferaseactivity was normalized to Renilla.

Reverse-transcription PCRReverse-transcription PCR (RT-PCR) and quantitative real-time

PCR (qRT-PCR) procedures are described in the SupplementaryData.

Collagen cross-linkingCollagen 1 solution (STEMCELL Technologies) was treated

with TG2 (Sigma) in the presence of 1 mmol/L DTT, 5 mmol/LCaCl2, and 100 mmol/L Tris–HCl (pH 8.0), as previouslydescribed (25) for 1.5 hours at 37�C.

Picrosirius red stainingParaffin-embedded slides were deparaffiinized and incubated

with Picrosirius red (PSR) solution containing 0.1% of Direct red80 powder (Sigma) dissolved in picric acid (Sigma) for 1 hour.Nuclei were counterstained with hematoxylin. The slides werewashed in 0.5% glacial acetic acid and dehydrated in ethanol andxylene. Imaging with circular polarizing light used an invertedNikon Diaphot 200 microscope (Nikon). The proportion ofdifferently colored collagen fibers was quantified with Meta-Morph (Molecular Devices, see the Supplementary Data; 26).

Soluble collagen assaySoluble collagen in xenografts (n ¼ 6 per group) was deter-

mined by using the Sircol Soluble Collagen Assay Kit (BiocolorLife Science Assays, Northern Ireland; see the SupplementaryData).

TG2 activityPDA cells and fibroblasts were incubated with 500 mmol/L

5-(Biotinamido) pentylamine (5-BP; Thermo Scientific) for 1 hourat 37�C. After fixing in methanol, slides were incubated with AlexaFluor 647 streptavidin (LifeTechnologies) for 1hour. Toassay in situactivity of TG2 in tumor tissue, 10-mm cryosections were incubatedat 37�C in a buffer containing 5 mmol/L CaCl2, 100 mmol/LTris–HCl (pH 8.0), in the presence or absence of 1 mmol/L DTT,and 0.001 mmol/L T26 or T26QN (negative control), as describedpreviously (27–29). As another negative control, 5 mmol/L EDTAwas added to the buffer. Imaging used a LSM 510 META confocalmicroscope (Carl Zeiss, Inc.) under UV excitation.

Statistical analysisThe Student t test compared measurements. P < 0.05 was

significant.

ResultsTG2 is abundantly expressed and enzymatically active in PDAcells and stroma

We used IHC to measure TG2 expression and cellular locali-zation in PDA specimens and in normal pancreas. Patient char-acteristics are presented in the Supplementary Data (Supplemen-tary Table S1). No immunostaining was recorded in the stroma ofnormal pancreas (n ¼ 3), and faint (1þ) staining was noted in

normal ducts. In contrast, strong (2þ to 3þ) TG2 cytoplasmicimmunostaining was recorded in 36 out of 52 (69%) PDA speci-mens, supporting that TG2 expression is increased in PDA com-pared with normal duct epithelium. TG2 immunostaining wasalso recorded in the stroma of 44 out of 52 specimens (84%; Fig.1A), involving both the cellular (fibroblasts) and extracellularcompartments. To determine whether TG2 was enzymaticallyactive in the stroma, 20 additional tumors identified through theIUSCC Tissue Bank as PDA specimens associated with significantdesmoplasia were stained for TG2 and for isopeptide, a covalentbond resulting from TG2 mediated transamidation. Concordantstrong (2þ to 3þ) TG2 and isopeptide staining were recorded in19out of 20 specimens (Fig. 1A), supporting that TG2 is expressedand active in the pancreaticDS. Isopeptide stainingwas detectablein the matrix and the basal membrane.

TG2 expression levels in cell lysates and CM from PDA, HPNE,stellate cells, and fibroblasts were examined by using Westernblotting. Abundant TG2 expression was detected in BxPC3 andAsPC1 cells and in the CM, confirming that it is secreted by PDAcells (Figs. 1B). TG2 expression was detectable in HPNE cells butlower than in most cancer cell lines. IF determined TG2 cellularlocalization and enzymatic activity bymeasuring incorporationof5-(Biotinamido) pentylamine (5-BP) and FITC-labeled T26 pep-tide, known TG2 substrates (Fig. 1C; Supplementary Fig. S1). TG2was expressed in the cytosol and the plasmamembrane of AsPC1and Panc1 cells, and its enzymatic activity was detectable in thecytoplasmof both cell types. In contrast, TG2was present, butwasenzymatically inactive in fibroblasts (Fig. 1C) and in LP9 normalmesothelial cells (Supplementary Fig. S1B), suggesting that theenzymatic activity may be differentially regulated in cancer versusnormal cells. Specificity is supported by lack of IF signalwhen cellswere incubated with the mutant T26QN peptide (not a TG2substrate) or in the presence of EDTA, which chelates Ca2þ

(Supplementary Fig. S1A). In addition, extracellular TG2 wasdetectable in the matrix deposited by AsPC1, BxPC3, and Panc1cells (Supplementary Fig. S2), supporting the concept that PDAcells secrete TG2 into the ECM.

TG2 knockdown inhibits PDA tumor growth in vivoTo proceed with functional studies, TG2 was knocked down in

AsPC1 andPanc1 cells by stably transducing shRNA targeting TG2(shTG2). TG2 protein and mRNA expression levels were signifi-cantly downregulated in shTG2-transduced cells compared withcells transduced with scrambled shRNA (shCtrl; Fig. 1D). TG2secretion in the CM was also decreased in shTG2 compared withshCtrl-transduced PDA cells (Fig. 1D, right). However, in vitro, nosignificant differences in the proliferation and ability to formclones betweenPDAcells�shTG2were observed (SupplementaryFig. S3), suggesting that any effects of TG2 on tumor growth arenot explained by direct effect on cancer cells.

To measure the effects of TG2 on tumor formation and pro-gression, orthotopic pancreatic xenografts derived from AsPC1andPanc1 cells inwhichTG2was stably knockeddownwereused.Average tumor volume and weight were significantly decreasedin xenografts derived from AsPC1þshTG2 compared withAsPC1þshCtrl cells (Fig. 2A, left and middle). There was nosignificant difference in the number ofmacro-metastases betweengroups (Fig. 2A, right). Stable knockdown of TG2 inAsPC1þshTG2 xenografts comparedwith controls was confirmedby IHC and was quantified by comparing H scores (see theSupplementary Data; Fig. 2B). Similar decrease in average tumor

Lee et al.

Clin Cancer Res; 21(19) October 1, 2015 Clinical Cancer Research4484




Figure 1.TG2 is expressed and active in pancreatic cancer cells and tumors. A, immunostaining for TG2 in normal pancreas (left, top; �40, bottom;�200); immunostainingfor TG2 (top right) and isopeptide (bottom right) in PDA specimens (�200). B, Western blotting for TG2 in PDA, HPNE, stellate (hPSC) cells and NHF544fibroblasts (left) and in CM (right, 40 mL/lane) collected from PDA cells. C, IF for TG2 (Alexa Fluor 488, green) and 5-BP (streptavidin, Alexa Fluor 647, red) in AsPC1,Panc1, and NHF544 cells (�600). Nuclei are visualized by DAPI. D, TG2 expression measured by RT-PCR (left) and Western blotting (middle and right) inAsPC1 and Panc1 cells stably transduced with shRNA targeting TG2 (shTG2) or control shRNA (shCtrl).






Figure 2.TG2 knockdown inhibits tumor growth in an orthotopic pancreatic xenograft mouse model. A, average tumor volume (left, P ¼ 0.02), weight (middle, P ¼ 0.04),and number of metastases (right, P ¼ 0.70) of xenografts derived from AsPC1 cells transduced with TG2 or control shRNA (shCtrl; n ¼ 11, shTG2; n ¼ 12).B, IHC for TG2 in AsPC1þshCtrl and AsPC1þshTG2 xenografts (left, �100). H scores quantitated TG2 expression in tumors (right, n ¼ 10 per group, P ¼ 0.004).C, average tumor volume (left, P ¼ 0.01), weight (middle, P ¼ 0.006), and number of metastases (right, P ¼ 0.03) derived from Panc1þshTG2 or þshCtrlcells (n ¼ 7 per group). D, IHC for TG2 in Panc1þshCtrl and PanC1þshTG2 xenografts (left, �100). H scores quantitated TG2 expression in tumors (right, n ¼ 3per group, P ¼ 0.007). Bars represent average measurements � standard error (SE); �, P < 0.05.

Lee et al.





volume and weight was observed in Panc1þshTG2 versusPanc1þshCtrl xenografts (Fig. 2C, left and middle). The numberofmacro-metastases was significantly decreased in Panc1þshTG2tumor-bearing mice compared with controls (Fig. 2C, right).Stable knockdown of TG2 was confirmed by IHC (Fig. 2D).Together, these data demonstrate that TG2 expression levels inPDA cells strongly correlate with in vivo tumor growth.

TG2 is enzymatically active and cross-links collagen in PDAtumors

Under normal physiologic conditions, TG2 is enzymaticallyinactive in vitro and in vivo. Various stressors, such as injury orhypoxia, are required to activate TG2 (30). To determine whetherTG2 is activated in the pancreatic TME, we measured incorpo-ration of the FITC-labeled T26 peptide, which is specifically cross-linked to ECM proteins by TG2 (29). Enzymatically active TG2was detectable both intra- and extracellularly in AsPC1 and Panc1xenograft sections (Fig. 3A), suggesting that the enzyme is active inthe pancreatic milieu. Incubation with the T26 peptide in thepresence of EDTA and with the mutant T26QN peptide, which isnot a TG2 substrate, were used as negative controls to provespecificity. TG2 activity was detectable even in the absence ofDTT,a known TG2 activator (Supplementary Fig. S4; ref. 31), support-ing that the enzyme is activated in situ.

One of the known TG2 substrates is collagen 1, abundant in thepancreatic DS. PSR staining was used to determine whethercollagen 1 was cross-linked by TG2 in the pancreatic TME.Birefringence of PSR staining under polarizing microscope ishighly specific for collagen fibers, which appear red, orange,yellow, or green in order of decreasing thickness (26). Collagenfibers stained mostly bright orange, red, and yellow inAsPC1þshCtrl tumors, while fibers were mostly green and yellowin AsPC1þshTG2 tumors (Figs. 3B, left and middle), suggestingthat collagen fibers are thicker in TG2-expressing tumors. More-over, the area of birefringence corresponding to collagen wasmore extensive in AsPC1þshCtrl compared with AsPC1þshTG2xenografts (Fig. 3B, right), suggesting increased collagen deposi-tion in TG2-expressing tumors, most collagen being depositedaround ducts and at the edge of tumors. To verify that thiscorrelation is relevant to human pancreatic tumors, PSR stainingof TG2-high and TG2-low human tumors was carried out (n ¼ 6per group). TG2 expression levels strongly correlated with thepercentage of cross-linked collagen (P < 0.001; Fig. 3C).

In addition, the sircol and the hydroxyproline assays were usedto measure soluble and total collagen, respectively, in proteinextracts from xenografts expressing TG2 or not. The amount ofsoluble collagen was increased in AsPC1þshTG2 compared withAsPC1þshCtrl tumors (Fig. 3D, left) and in Panc1þshTG2 com-pared with Panc1þshCtrl xenografts (Fig. 3D, right), while totalcollagen content was similar (Supplementary Fig. S5). Collective-ly, these data suggest that TG2 secreted from PDA cells promotescollagen cross-linking in the pancreatic stroma.

TG2-mediated cross-linked collagen and TG2-expressing PDAcells promote the growth and activation of fibroblasts

We next determined the functional significance of native andTG2-cross-linked collagen to the proliferation of fibroblasts andcancer cells. The proliferation of NHF544 fibroblasts wasincreased on collagen coated compared with uncoated plates(Fig. 4A, top left) and on TG2-cross-linked collagen comparedwith native collagen (Fig. 4A, bottom left), supporting that

collagen and cross-linked collagen in the matrix promote fibro-blast proliferation.

Next,we studied the influence of TG2 secretedby cancer cells onstromal cell proliferation. Direct coculture of TG2-expressingAsPC1 cells significantly promoted GFP-labeled fibroblast(GFP-HNDF) proliferation over a 14-day period compared withcoculture of fibroblasts with AsPC1þshTG2 cells (Fig. 4A, right).Fibroblast proliferation in coculture with cancer cells expressingand secreting low levels of TG2 (AsPC1þshTG2)was not differentfrom proliferation of fibroblast cultured alone, supporting thatsecretion of TG2 in coculture models augments the fibroticresponse. However, in the absence of cocultured PDA cells,addition of recombinant TG2 to the growth media did not alterfibroblast or stellate cell proliferation (Supplementary Fig. S6),suggesting that TG2 acts in concert with other factors secreted inthe matrix.

To measure fibroblast activation in the presence of cancer cells,expression of a-SMA was measured by IF. Coculture of NHF544cells with AsPC1 and Panc1 cells induced expression of a-SMA(Fig. 4B, left). To determine whether TG2 secreted by PDA cellscontributes tofibroblast activation, NHF544 cells were coculturedwith PDA cells �shTG2, and IF staining measured collagendeposition. Increased collagen 1 staining was observed inNHF544 fibroblasts cocultured with AsPC1 and Panc1 controlcells, compared with fibroblasts cocultured with AsPC1þshTG2and Panc1þshTG2 cells (Fig. 4B, right). Likewise, in tumors, thenumber of a-SMA-expressing (activated) myofibroblasts wasincreased in AsPC1þshCtrl compared with AsPC1þshTG2 xeno-grafts (Fig. 4C). Collectively, these data show that in thepancreaticTME, TG2-secreting PDA cells promote the growth, activation,and collagen deposition by myofibroblasts.

We next tested whether these stromal alterations converselyregulate the growth of PDA cells. Proliferation of AsPC1 andPanc1 cells was promoted when cells were cultured on collagencoated versus uncoated plates (Fig. 4D, top) and on a feeder layerof hPSCs versus uncoated plates (Fig. 4D, bottom). Taken togeth-er, these results demonstrate that PDA cells secreting TG2 in thetumor milieu promote fibroblast activation and proliferation aswell as collagen cross-linking, leading to stromal alterations,which, in turn, promote tumor growth.

Matrix-induced YAP/TAZ activation promotes PDA cellproliferation

As the pancreatic stroma becomes infiltrated by collagen andfibroblasts, its stiffness increases (32). One of the mechanismsactivated by the physical properties of the ECM is the YAP/TAZpathway (33), which regulates cancer cell growth. We measuredactivation of the pathway by IF staining for the two transcriptionfactors in PDA cells plated on collagen versus uncoated plates.Increased nuclear YAP/TAZ staining was observed in Panc1 andAsPC1 cells on collagen (Fig. 5A). Western blotting showeddecreased levels of phospho-YAP (p-Yap, inactive) and increasedratio of total-YAP/p-YAP were recorded in PDA cells plated oncollagen-coated versus uncoated plates (Fig. 5A, right), indicatingYAP activation by the collagen matrix.

Next, IF staining determined the effect of TG2 on YAP/TAZactivation in PDA cells. Increased nuclear localization wasrecorded in TG2-expressing control Panc1 and AsPC1 cells com-pared with PDA cells transduced with shTG2 (Fig. 5B). Decreasedlevel of p-YAP and increased ratio of total-YAP/p-YAP wereobserved in Panc1þshCtrl compared with Panc1þshTG2 cells






Figure 3.TG2 is enzymatically active and cross-links collagen in the PDA stroma. A, IF using FITC-labeled T26 and EDTA (negative control), and T26QN (negative control) inAsPC1 and Panc1 xenografts (�200). B, PSR staining determines deposition and thickness of collagen fibers in AsPC1þshCtrl and AsPC1þshTG2 tumors; top,bright-field transmission (BFT) microscopy; bottom, polarizing light microscopy (�200, left). Quantification of collagen fiber thicknesses based on differences inbirefringence (middle) and quantitative assessment of collagen content based on area staining for collagen per field (right, n¼ 11 per group, P¼ 0.001). C, PSR stainingmeasures deposition and thickness of collagen fibers in TG2 high and low human tumors; top, IHC for TG2 (�100); bottom, polarizing light microscopy (�200, left).Quantification of collagen fiber thicknesses based on differences in birefringence (middle) and quantitative assessment of collagen content based on areastaining for collagen per field (right, n¼ 6 per group). Bars represent fold changes� SE, P < 0.0001. D, soluble collagenmeasured by the Sircol assay in AsPC1� shTG2(left, n ¼ 6 per group, P ¼ 0.001) and Panc1�shTG2 tumors (right, n ¼ 6 per group, P ¼ 0.005). Bars represent average measurements � SE; � , P < 0.05.

Lee et al.





Figure 4.TG2 promotes fibrosis in the PDA stroma. A, proliferation of NHF544 cultured on collagen-coated or uncoated plates (left, top, P < 0.04) and on native orTG2-cross-linked collagen (left, bottom, P < 0.03). Proliferation of GFP-HNDF in coculture with AsPC1þshCtrl or AsPC1þshTG2 cells or grown alone for 14 days(right, P < 0.03). B, IF for a-SMA (left, Alexa Fluor 488, green) in NHF544 cultured alone (left) or cocultured with AsPC1 (middle) or Panc1 cells (right).Nuclei are visualized by DAPI. IF for collagen 1 (right, Cy5, red) of NHF544 cocultured with AsPC1þshCtrl or AsPC1þshTG2 (left) or with Panc1þshCtrl andPanc1þshTG2 cells (right). C, IHC for a-SMA identifies myofibroblasts in AsPC1þshCtrl and AsPC1þshTG2 xenografts (top, �100). H scores quantify fibrosis(bottom, shCtrl; n¼ 10, shTG2; n¼ 7,P<0.05). D, proliferation ofAsPC1 andPanc1 cells on collagen-coated or uncoated plates (top,P <0.03), and on uncoatedplatesor a feeder layer of hPSCs (bottom, P < 0.006). A and D, bars represent average of four replicate measurements for the groups � SE; � , P < 0.05.






Figure 5.TG2 and collagen activate YAP/TAZ in PDA cells. A, IF for YAP/TAZ (left, Alexa Fluor 488, green) in Panc1 and AsPC1 cells grown on collagen-coated oruncoated slides (�600). Pie charts (right, top) show percentage of nuclear YAP/TAZ, and Western blotting (right, bottom) shows total- and p-YAP in Panc1and AsPC1 cells plated on collagen or uncoated plates. Bars represent the ratio of total-YAP/p-YAP. B, IF for YAP/TAZ (left, Alexa Fluor 488, green) inPanc1þshCtrl or Panc1þshTG2 and AsPC1þshCtrl or AsPC1þshTG2 cells grown on collagen (�600). (Continued on the following page.)

Lee et al.





(Fig. 5B, right). We next measured the transcriptional activity ofYAP in Panc1 cells plated on collagen by using the TEAD4 reporterassay and by measuring the mRNA levels for CTGF, a knownYAP/TAZ target gene. TEAD4 activity (Fig. 5C, top) and CTGFmRNA levels (Fig. 5C, bottom) were increased in Panc1þshCtrlcompared with Panc1þshTG2 cells, confirming activation of thepathway in TG2-expressing cells.

To determine the significance of YAP/TAZ to PDA cell prolif-eration, transient siRNA transfection targeting both transcriptionfactors was used. YAP and TAZ knockdown by siRNA confirmedby Western blotting (Fig. 5D, top) caused decreased proliferationof AsPC1 and Panc1 cells grown on collagen (Fig. 5D, bottom).Collectively, these data suggest that the transcription factors YAPand TAZ are activated in TG2-expressing PDA cells and outline anew mechanism by which cancer cell proliferation is modulatedby TG2 through signals reflected from the tumor into the sur-rounding stroma (Fig. 6).

DiscussionOur findings outline how changes in the pancreatic cancer

stroma induced by TG2 secreted from PDA cells alter tumorgrowth. Although overexpression of TG2 in epithelial cancers(34, 35) including pancreatic cancer (36, 37) has been previouslyreported, and its direct effects on cellular oncogenic signaling havebeen dissected (37, 38), this is the first report describing the effectsof TG2 on the pancreatic TME, and its subsequent consequenceson cancer cell proliferation and tumor growth. We demonstratethat TG2-expressing PDA cells form larger pancreatic tumors,enriched in thick collagen fibers and inmyofibroblasts. The densestroma of TG2-expressing tumors conveys signals leading to YAP/TAZ activation in cancer cells promoting cell proliferation andtumor growth. This report has several important consequences.

First, we demonstrate that TG2 knockdown in pancreatic cancercells inhibited tumor growth. This effect wasmore pronounced inPanc1 compared with AsPC1 cells, possibly because of the shorterdoubling times of Panc1 cells, which are more aggressive andsemi-mesenchymal. Interestingly, the smaller tumor volumeswere not accounted for by decreased proliferation or clonogenicpotential of thePDAcells, but by substantial stromal alterations. Aprevious study using siRNA targeting TG2 delivered as liposomesinto the peritoneal cavity noted inhibition of pancreatic tumorgrowth along with decreased tumor vascularity and decreasedcancer cell proliferation (36). In that study, TG2 knockdownaffected all TG2-expressing cells exposed to the carrier liposomes.It is, therefore, possible that TG2 knockdown in endothelial cellsled to the antiangiogenic effect reported there and the subsequentinhibition of cell proliferation (36). Contrasting to that study andconsistent with TG2 knockdown restricted to PDA cells in ourmodel, we did not observe changes in multi-vessel densitybetween tumors expressing TG2 or not (not shown). However,we report a significant decrease in the number of activatedmyofibroblasts and thick collagen fibers in tumors derived fromshTG2-transduced cells. We show that these stromal alterations

directly affect proliferation of fibroblasts and cancer cells. Theprotective role of the pancreatic stroma to cancer cells has beenpreviously recognized and linked to the resistance of pancreatictumors to chemo- and radiotherapy (39–41). Our data implicatefor the first time TG2 as a modulator of the pancreatic TMEthrough its protein cross-linking properties.

Second, we report that TG2 is enzymatically active in PDA cellsand in the matrix where it cross-links collagen, and possibly otherECM substrates. It is known that the functions of TG2 are tightlycontrolled by environmental factors, which cause allosteric struc-tural changes and alter the accessibility of the catalytic site. HighCa2þ concentrations in the ECM allow the protein to adopt anopen conformation that exposes the enzymatic core, while highintracellular GTP forces a closed, enzymatically inactive structure(19). It is, therefore, accepted that TG2 is enzymatically inactive inthe intracellular environment, under normal physiologic condi-tions, as we have also observed here by analyzing human fibro-blasts and mesothelial LP9 cells. However, cross-linking activitywas detectable in PDA cells and theirmatrix as assessed in vitro andin vivo, suggesting that the enzyme is activated in transformed cellsand in the pancreatic matrix. Although the factors causing TG2activation in this context remain unknown, it is tempting tospeculate that hypoxia or other types of tissue injury inflicted bytumorigenesis may be inducing transglutaminase activity in vivo(30, 42). The involvement of metabolic or physical factors thatcould be expressed or function differently in transformed versusnormal cells (e.g., oxidative stress, calciumhomeostasis) cannot beexcluded. Interestingly, a previous report suggested that IkBa, theNF-kB inhibitory subunit, is a direct substrate of TG2 in PDA cells(16), although direct evidence of TGase activity was not provided.

We also show that collagen 1 was cross-linked by TG2 in thepancreatic milieu leading to formation of a dense matrix thatstimulated the growth of fibroblasts and cancer cells. Increasedcollagendepositionwasnoted infibroblasts coculturedwith TG2-expressing PDA cells. Our findings are consistent with a previous

(Continued.) Pie charts (right, top) show the percentage of nuclear YAP/TAZ, and Western blotting (right, bottom) shows total- and p-YAP in Panc1þshCtrl orPanc1þshTG2 grown on collagen. Bars represent the ratio of total-YAP/p-YAP. C, TEAD4 reporter assay in Panc1þshCtrl or Panc1þshTG2 grown on collagen(top, n ¼ 5, P ¼ 0.009). Luciferase activity was normalized to Renilla. Bars represent averages of 5 replicates/group � SE; � , P < 0.05. qRT-PCR measuresCTGFmRNA levels in Panc1þshCtrl or Panc1þshTG2 grown on collagen (bottom,P¼0.04). D,Western blotting for YAP and TAZ inAsPC1 andPanc1 cells transfectedwith siRNA targeting YAP and TAZ (top). Proliferation of AsPC1 (bottom, left) and Panc1 (bottom, right) cells transfected with siRNA targeting YAP andTAZ (n ¼ 4 per group, P < 0.01). Bars represent average of 4 replicates � SE; �, P < 0.05.

Figure 6.Proposed mechanism by which TG2 promotes tumor growth.






report demonstrating that collagen cross-linked by TG2 stimu-lated the proliferation of dermal fibroblasts (25) and with studiesdemonstrating the involvement of TG2 in wound healing and inpathologic fibrotic responses in the lung or liver (15, 43–46). Wesuspect that the enzyme has other substrates in the pancreaticECM and that their modifications through cross-linking couldcontribute to the distinct architecture of the DS, which, in turn,fosters tumor progression.

Third, we link here for the first time TG2 to activation of thetranscriptional regulators, YAP and TAZ, in cancer cells. Thispathway, strongly implicated in cell survival and proliferationand activated in epithelial cancers (47, 48), is known to beregulated by the physical properties of the ECM and to translatemechanical cues into relevant biologic signals (33). Recent reportshave implicated YAP signaling in pancreatic tumor progressiondownstream of mutant Ras or independent of it (49, 50). Wepropose that the pancreatic matrix rendered denser by TG2through collagen cross-linking and stimulation of a fibroticresponse, provides positive proliferative feedback to the cancercells by stimulating YAP/TAZ signaling. Interestingly, TG2 expres-sion levels did not alter cancer cell proliferation in the absence ofthe matrix. However, we cannot exclude that signals other thantissue stiffness, directly modulated by TG2 or by the matrix,contribute to YAP activation in this context.

These observations led us to posit that the stimulatory growthsignals conveyed by TG2 require the stroma, which "deflects"mechanical or biologic cues from the tumor milieu into an activeoncogenic program. In all, our findings support the involvementof enzymatically active TG2 in the progression of PDA and futurestudies targeting the cross-linking activity of TG2 to disrupt tumorprogression.

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

Authors' ContributionsConception and design: J. Lee, D. MateiDevelopment of methodology: J. Lee, S. Condello, B. Yakubov, K. Hitomi,J. Xie, D. MateiAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): R. Emerson, D. MateiAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): J. Lee, S. Condello, B. Yakubov, D. MateiWriting, review, and/or revision of the manuscript: J. Lee, S. Condello,K. Hitomi, D. MateiAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): A. Caperell-Grant, K. Hitomi, J. Xie, D. MateiStudy supervision: D. MateiOther (experimental procedures): J. Lee

AcknowledgmentsThe authors thank Drs. Dan Spandau, Murray Korc, and Clark Wells for

reagents; Bhadrani Chelladurai and Dr. Malgorzata Kamocka for technicalassistance; and Dr. Kenneth Nephew for helpful comments.

Grant SupportThis work was made possible by funding from the National Cancer

Institute (Award CA152502), a Project Development Team within the ICTSINIH/NCRR (Grant Number UL1TR001108), and the IUPUI Signature CenterInitiative.

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 to indicatethis fact.

Received January 29, 2015; revised May 7, 2015; accepted May 24, 2015;published OnlineFirst June 3, 2015.

References1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin

2015;65:5–29.2. Willett CG, Czito BG, Bendell JC, Ryan DP. Locally advanced pancreatic

cancer. J Clin Oncol 2005;23:4538–44.3. Mahadevan D, Von Hoff DD. Tumor-stroma interactions in pancreatic

ductal adenocarcinoma. Mol Cancer Ther 2007;6:1186–97.4. Miyamoto H, Murakami T, Tsuchida K, Sugino H, Miyake H, Tashiro S.

Tumor-stroma interaction of human pancreatic cancer: acquired resistanceto anticancer drugs and proliferation regulation is dependent on extracel-lular matrix proteins. Pancreas 2004;28:38–44.

5. Muerkoster S,Wegehenkel K, Arlt A,WittM, Sipos B, KruseML, et al. Tumorstroma interactions induce chemoresistance in pancreatic ductal carcinomacells involving increased secretion and paracrine effects of nitric oxide andinterleukin-1beta. Cancer Res 2004;64:1331–7.

6. OmaryMB, Lugea A, Lowe AW, Pandol SJ. The pancreatic stellate cell: a staron the rise in pancreatic diseases. J Clin Invest 2007;117:50–9.

7. Muerkoster SS, Werbing V, Koch D, Sipos B, Ammerpohl O, Kalthoff H,et al. Role of myofibroblasts in innate chemoresistance of pancreaticcarcinoma—epigenetic downregulation of caspases. Int J Cancer 2008;123:1751–60.

8. Olive KP, JacobetzMA,DavidsonCJ,GopinathanA,McIntyreD,HonessD,et al. Inhibition of Hedgehog signaling enhances delivery of chemotherapyin a mouse model of pancreatic cancer. Science 2009;324:1457–61.

9. Koay EJ, Baio FE, Ondari A, Truty MJ, Cristini V, Thomas RM, et al. Intra-tumoral heterogeneity of gemcitabine delivery and mass transport inhuman pancreatic cancer. Phys Biol 2014;11:065002.

10. Koay EJ, Truty MJ, Cristini V, Thomas RM, Chen R, Chatterjee D, et al.Transport properties of pancreatic cancer describe gemcitabine deliveryand response. J Clin Invest 2014;124:1525–36.

11. Ozdemir BC, Pentcheva-Hoang T, Carstens JL, Zheng X, Wu CC, SimpsonTR, et al. Depletion of carcinoma-associated fibroblasts and fibrosis

induces immunosuppression and accelerates pancreas cancer with reducedsurvival. Cancer Cell 2014;25:719–34.

12. Rhim AD, Oberstein PE, Thomas DH, Mirek ET, Palermo CF, Sastra SA,et al. Stromal elements act to restrain, rather than support, pancreatic ductaladenocarcinoma. Cancer Cell 2014;25:735–47.

13. Swartz MA, Iida N, Roberts EW, Sangaletti S, Wong MH, Yull FE, et al.Tumor microenvironment complexity: emerging roles in cancer therapy.Cancer Res 2012;72:2473–80.

14. Cao L, Shao M, Schilder J, Guise T, Mohammad KS, Matei D. Tissuetransglutaminase links TGF-beta, epithelial to mesenchymal transitionand a stem cell phenotype in ovarian cancer. Oncogene 2012;31:2521–34.

15. Verderio EA, Johnson T, Griffin M. Tissue transglutaminase in normal andabnormal wound healing: review article. Amino Acids 2004;26:387–404.

16. VermaA,WangH,Manavathi B, Fok JY,MannAP, Kumar R, et al. Increasedexpression of tissue transglutaminase in pancreatic ductal adenocarcinomaand its implications in drug resistance and metastasis. Cancer Res 2006;66:10525–33.

17. Cheung W, Darfler MM, Alvarez H, Hood BL, Conrads TP, Habbe N, et al.Application of a global proteomic approach to archival precursor lesions:deleted in malignant brain tumors 1 and tissue transglutaminase 2 areupregulated in pancreatic cancer precursors. Pancreatology 2008;8:608–16.

18. Belkin AM. Extracellular TG2: emerging functions and regulation. Febs J2011;278:4704–16.

19. PinkasDM, StropP, Brunger AT, Khosla C. Transglutaminase 2 undergoes alarge conformational change upon activation. PLoS Biol 2007;5:e327.

20. Yakubov B, Chelladurai B, Schmitt J, Emerson R, Turchi JJ, Matei D.Extracellular tissue transglutaminase activates noncanonical NF-kappaBsignaling and promotes metastasis in ovarian cancer. Neoplasia 2013;15:609–19.

21. Shannon HE FM, Xie J, Gu D, McCarthy BP, Riley AA, Sinn AL, et al.Longitudinal Bioluminescence Imaging of Primary versus Abdominal

Lee et al.





Metastatic Tumor Growth in Orthotopic Pancreatic Tumor Models inNOD/SCIDy(-/-) Mice. Pancreas 2015;44:64–75.

22. Satpathy M, Cao L, Pincheira R, Emerson R, Bigsby R, Nakshatri H, et al.Enhanced peritoneal ovarian tumor dissemination by tissue transgluta-minase. Cancer Res 2007;67:7194–202.

23. Faye C, Inforzato A, Bignon M, Hartmann DJ, Muller L, Ballut L, et al.Transglutaminase-2: a new endostatin partner in the extracellularmatrix ofendothelial cells. Biochem J 2010;427:467–75.

24. Adler JJ, Johnson DE, Heller BL, Bringman LR, RanahanWP, Conwell MD,et al. Serum deprivation inhibits the transcriptional co-activator YAP andcell growth via phosphorylation of the 130-kDa isoform of Angiomotin bythe LATS1/2 protein kinases. ProcNatl Acad Sci U S A2013;110:17368–73.

25. Chau DY, Collighan RJ, Verderio EA, Addy VL, Griffin M. The cellularresponse to transglutaminase-cross-linked collagen. Biomaterials 2005;26:6518–29.

26. Rich L, Whittaker P. Collagen and picrosirius red staining: a polarizedlight assessment of fibrillar hue and spatial distribution. J Morphol Sci2005;22:97–104.

27. Fukui M, Kuramoto K, Yamasaki R, Shimizu Y, Itoh M, Kawamoto T, et al.Identification of a highly reactive substrate peptide for transglutaminase 6and its use in detecting transglutaminase activity in the skin epidermis.FEBS J 2013;280:1420–9.

28. Yamane A, Fukui M, Sugimura Y, Itoh M, Alea MP, Thomas V, et al.Identification of a preferred substrate peptide for transglutaminase 3and detection of in situ activity in skin and hair follicles. FEBS J 2010;277:3564–74.

29. Johnson KB, Petersen-Jones H, Thompson JM, Hitomi K, Itoh M, BakkerEN, et al. Vena cava and aortic smooth muscle cells express transglutami-nases 1 and 4 in addition to transglutaminase 2. Am J Physiol Heart CircPhysiol 2012;302:H1355–66.

30. Siegel M, Strnad P,Watts RE, Choi K, Jabri B, OmaryMB, et al. Extracellulartransglutaminase 2 is catalytically inactive, but is transiently activated upontissue injury. PloS One 2008;3:e1861.

31. Jin X, Stamnaes J, KlockC,DiRaimondoTR, Sollid LM,KhoslaC. Activationof extracellular transglutaminase 2 by thioredoxin. J Biol Chem 2011;286:37866–73.

32. Alvarez R,MusteanuM,Garcia-Garcia E, Lopez-Casas PP,MegiasD,GuerraC, et al. Stromal disrupting effects of nab-paclitaxel inpancreatic cancer. Br JCancer 2013;109:926–33.

33. Dupont S, Morsut L, Aragona M, Enzo E, Giulitti S, Cordenonsi M, et al.Role of YAP/TAZ in mechanotransduction. Nature 2011;474:179–83.

34. Mehta K, Fok J, Miller FR, Koul D, Sahin AA. Prognostic significance oftissue transglutaminase in drug resistant andmetastatic breast cancer. ClinCancer Res 2004;10:8068–76.

35. Condello SMC, Nagdas S, Cao L, Turek J, Hurley TD,Matei D. Beta-cateninregulated ALDH1A1 is a target in ovarian cancer spheroids. Oncogene2015;34:2297–308.

36. Verma A, Guha S, Diagaradjane P, Kunnumakkara AB, Sanguino AM,Lopez-Berestein G, et al. Therapeutic significance of elevated tissue trans-

glutaminase expression in pancreatic cancer. Clin Cancer Res 2008;14:2476–83.

37. Verma A, Guha S, Wang H, Fok JY, Koul D, Abbruzzese J, et al. Tissuetransglutaminase regulates focal adhesion kinase/AKT activation by mod-ulating PTEN expression in pancreatic cancer cells. Clin Cancer Res2008;14:1997–2005.

38. Shao M, Cao L, Shen C, Satpathy M, Chelladurai B, Bigsby RM, et al.Epithelial-to-mesenchymal transition and ovarian tumor progressioninduced by tissue transglutaminase. Cancer Res 2009;69:9192–201.

39. Erkan M. Understanding the stroma of pancreatic cancer: co-evolution ofthe microenvironment with epithelial carcinogenesis. J Pathol 2013;231:4–7.

40. Garrido-Laguna I, Uson M, Rajeshkumar NV, Tan AC, de Oliveira E,Karikari C, et al. Tumor engraftment in nude mice and enrichment instroma-related gene pathways predict poor survival and resistance togemcitabine in patients with pancreatic cancer. Clin Cancer Res 2011;17:5793–800.

41. Gore J, Korc M. Pancreatic cancer stroma: friend or foe? Cancer Cell2014;25:711–2.

42. Filiano AJ, Bailey CD, Tucholski J, Gundemir S, Johnson GV. Transgluta-minase 2 protects against ischemic insult, interacts with HIF1beta, andattenuates HIF1 signaling. FASEB J 2008;22:2662–75.

43. Collighan RJ, Griffin M. Transglutaminase 2 cross-linking of matrix pro-teins: biological significance and medical applications. Amino Acids2009;36:659–70.

44. Shweke N, Boulos N, Jouanneau C, Vandermeersch S, Melino G, DussauleJC, et al. Tissue transglutaminase contributes to interstitial renal fibrosis byfavoring accumulation of fibrillar collagen through TGF-beta activationand cell infiltration. Am J Pathol 2008;173:631–42.

45. Olsen KC, Epa AP, Kulkarni AA, Kottmann RM,McCarthy CE, JohnsonGV,et al. Inhibition of transglutaminase 2, a novel target for pulmonaryfibrosis, by two small electrophilic molecules. Am J Respir Cell Mol Biol2014;50:737–47.

46. Olsen KC, Sapinoro RE, Kottmann RM, Kulkarni AA, Iismaa SE, JohnsonGV, et al. Transglutaminase 2 and its role in pulmonaryfibrosis. Am JRespirCrit Care Med 2011;184:699–707.

47. Cordenonsi M, Zanconato F, Azzolin L, Forcato M, Rosato A, Frasson C,et al. The Hippo transducer TAZ confers cancer stem cell-related traits onbreast cancer cells. Cell 2011;147:759–72.

48. Zeng Q, Hong W. The emerging role of the hippo pathway in cell contactinhibition, organ size control, and cancer development in mammals.Cancer Cell 2008;13:188–92.

49. Kapoor A, Yao W, Ying H, Hua S, Liewen A, Wang Q, et al. Yap1 activationenables bypass of oncogenic Kras addiction in pancreatic cancer. Cell2014;158:185–97.

50. Zhang W, Nandakumar N, Shi Y, Manzano M, Smith A, Graham G, et al.Downstream of mutant KRAS, the transcription regulator YAP is essentialfor neoplastic progression to pancreatic ductal adenocarcinoma. Sci Sig-naling 2014;7:ra42.






2015;21:4482-4493. Published OnlineFirst June 3, 2015.Clin Cancer Res Jiyoon Lee, Salvatore Condello, Bakhtiyor Yakubov, et al. Promotes Pancreatic Cancer Progression

Stroma Interaction−Tissue Transglutaminase Mediated Tumor

Updated version

10.1158/1078-0432.CCR-15-0226doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2015/06/04/1078-0432.CCR-15-0226.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/21/19/4482.full#ref-list-1

This article cites 50 articles, 15 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/21/19/4482.full#related-urls

This article has been cited by 7 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://clincancerres.aacrjournals.org/content/21/19/4482To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-15-0226

http://clincancerres.aacrjournals.org/content/suppl/2015/06/04/1078-0432.CCR-15-0226.DC1

http://clincancerres.aacrjournals.org/content/21/19/4482.full#ref-list-1

http://clincancerres.aacrjournals.org/content/21/19/4482.full#related-urls

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/21/19/4482


tissuetransglutaminasemediatedtumor stroma interaction … · augments oncogenic signaling in...

Documents