Phosphorylation of the src Epithelial Substrate Trask Is TightlyRegulated in Normal Epithelia but Widespread inMany Human Epithelial CancersChing HangWong,1Frederick L. Baehner,2 Danislav S. Spassov,1Deepika Ahuja,1DonghuiWang,3

Byron Hann,3 Jimmy Blair,4 Kevan Shokat,4,5 Alana L.Welm,6 andMarkM.Moasser1

Abstract Purpose:The frequently elevated activities of the c-src and c-yes products in human epithelialtumors suggest that these activated tyrosine kinaseshave tumorigenic functions analogous to thev-src and v-yes oncogene products. Studies of v-src ^ transformed fibroblasts have identifiedmany of the effectors of this potent oncogene; however, because c-src and c-yes lack themutational and promiscuous activities of their retroviral oncogene homologues, their presumptivetumorigenic functions in human epithelial tumors are more subtle, less well-defined, and awaitidentification of possible effectors more directly relevant to epithelial cells.Experimental Design:We recently identified a transmembrane glycoprotein namedTrask that isexpressed in epithelial tissues but not fibroblasts and is phosphorylated by SRC kinases inmitoticepithelial cells. In this study, we have surveyed the expression and phosphorylation of Trask inmany human epithelial cancer cell lines and surgical tissues and tumors.Results: Trask is widely expressed in human epithelial tissues, but its phosphorylation is tightlyregulated and restricted to detachedmitotic cells or cells undergoingphysiologic shedding.How-ever, abberant Trask phosphorylation is seen in many epithelial tumors from all stages includingpreinvasive, invasive, and metastatic tumors. Trask phosphorylation requires SRC kinases, and isalso aberrantly hyperphosphorylated in the SRC-activated PyMT mouse epithelial tumors anddephosphorylated by the SRC inhibitor treatment of these tumors.Conclusions:The widespread phosphorylation ofTrask in many human epithlelial cancers iden-tifies a new potential effector of SRC kinases in human epithelial tumorigenesis.

The oncogenic potential of Src kinases has been recognized formore than three decades stemming from the identification ofthe v-src tyrosine kinase oncogene as the tumorigenic driver ofthe Rous sarcoma virus (reviewed in ref. 1). v-src or engineeredactivating mutants of c-src are highly transforming in experi-mental models (2, 3). Although the mutational activation of Srckinases is extremely rare in human tumors, Src and Yes show

increased activity in many human epithelial cancers includingcancers of the colon, breast, pancreas, and lung (reviewed inref. 4). This is recapitulated in mouse models of epithelialcancer where Src and Yes kinases are activated in PyMT-inducedand in Neu-induced mammary epithelial tumors (5, 6). Theessential role of Src in the PyMT model is confirmed as thesetumors are suppressed in a src-null background (7). However,the mechanisms that lead to the activation of Src kinases inhuman tumors remain undefined, and an essential role for Srckinases in human cancers remains presumptive at this pointand awaits further definition.

Numerous lines of evidence suggest that the function of Srckinases may be particularly important for invasion and meta-stasis in human cancers. The increased invasive and metastaticproperties of ErbB2-induced epithelial tumors are associatedwith increased expression and activity of Src and this can bereverted by pharmacologic inhibitors or dominant-negativemutants of Src kinase (8). Increased invasive properties aresimilarly conferred to intestinal epithelial cells by overexpres-sion of c-src (9). Treatment of cancer cells with Src-selectiveinhibitors reduces their invasive and migratory propertieswith much less effect on their proliferative attributes (10, 11).In mouse orthotopic models of human epithelial cancer, theinactivation of Src kinases identifies a function more importantto invasion and metastasis than proliferative growth (12, 13).Cellular substrates of Src that are thought to mediate theinvasive and migratory properties conferred by overactive Src

Human Cancer Biology

Authors’Affiliations: Departments of 1Medicine, 2Pathology, 3ComprehensiveCancer Center, 4Cellular and Molecular Pharmacology, 5Howard Hughes MedicalInstitute, and 6G.W.Hooper Foundation, University of California, San Francisco,San Francisco, CaliforniaReceived 10/1/08; revised 12/25/08; accepted 12/30/08; published OnlineFirst3/24/09.Grant support: NIH grant CA113952 and the American Cancer Society RSG-0213901CDD. D.S. Spassov is supported by a Susan G. Komen for the Curepostdoctoral fellowship. C.H.Wong is supported by a California Breast CancerResearch Program Postdoctoral fellowship.The costs of publication of this article were defrayed in part by the payment of pagecharges.This article must therefore be hereby marked advertisement in accordancewith18 U.S.C. Section1734 solely to indicate this fact.Note: Supplementary data for this article are available at Clinical Cancer ResearchOnline (http://clincancerres.aacrjournals.org/).C.H.Wong and F.L. Baehner contributed equally to this work.Requests for reprints:Mark M. Moasser, University of California, San Francisco,University of California at San FranciscoBox 0875, San Francisco, CA94143-0875.Phone: 415-476-0158; Fax: 415-502-5665; E-mail: mark.moasser@ucsf.edu.

F2009 American Association for Cancer Research.doi:10.1158/1078-0432.CCR-08-2533

www.aacrjournals.org Clin Cancer Res 2009;15(7) April 1, 20092311

Research. on July 21, 2021. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

include adhesion signaling proteins including certain integrins,focal adhesion complex proteins, and certain extracellularand membrane proteases (reviewed in ref. 14). Much of thisevidence comes from the analysis of cellular changes inducedby the highly transforming v-src oncogene product.

Although the evidence that Src and Yes are functionallyimportant in human tumors is compelling, mechanisticexploration of this function has proven to be complex andchallenging. A large body of evidence has been derived frommechanistic studies of cell transformation by v-src in fibroblastmodels (reviewed in refs. 14, 15). However, the v-src oncogeneproduct has considerable structural and functional differencesfrom the c-src and c-yes gene products of human tumors andmuch more potent transforming activity compared with thehuman proto-oncogenes. In fact, c-src is not transforming,even when overexpressed, and its mutational activation is veryrarely seen in human cancers (reviewed in ref. 16). Therefore,the tumor promoting functions of c-src and c-yes may be muchmore subtle than their viral oncogene homologues. In addition,the fibroblast models do not faithfully represent the fullspectrum of human cancers. In fact, the increased activity ofSrc kinases is mostly seen in the very common humanepithelial cancers as well as the much rarer mesenchymaltumors that are modeled by experimental fibroblast systems.Epithelial cells differ significantly from mesenchymal cells intheir interactions with each other and with the extracellularmatrix. They form semipermanent attachment structures suchas hemidesmosomes and tight junctions as well as unidirec-tional attachment to basement membrane proteins. Theseattributes define the epithelial tissue architecture and are notpresent in mesenchymal cells. Cell migration and mitosisrequire dissociation of these and other structures and special-ized mechanisms are most likely involved to allow epithelialcell detachment and migration. As such, experimental epithelialcell models may provide mechanistic insight into the role ofSrc kinases in epithelial tumors not appreciated from fibroblastmodels, and there may be important Src effectors in humanepithelial tumors not evident or even expressed in fibroblasts.

Our studies have focused on epithelial cell models andepithelial cancers with wild-type Src family kinases. These mod-

els may reveal mechanisms and markers not readily apparentin fibroblast models of v-src transformation, yet possibly highlyrelevant to the most common human malignancies. In thiseffort, we previously identified Trask, a substrate of Src kinasesthat is phosphorylated during epithelial cell mitosis (17). Traskis a type 1 transmembrane glycoprotein seen in multicellularorganisms from vertebrates to mammalians, unique in thehuman genome with no homologous family members, and itscellular or biochemical functions are not readily predictable bycomparative homology. Its structure contains a large extracel-lular domain containing two CUB domains and a smallerintracellular region containing a proline rich region. Trask isexpressed as a 140 kDa membrane glycoprotein but can becleaved by the membrane serine protease MT-SP1 within theextracellular region to yield a smaller 85 kDa Trask protein (17).The functional relevance of this cleavage is not yet understood,but transcriptional profiling analysis of numerous normal andtumor tissues reveals that the expression of Trask and MT-SP1 istightly linked and particularly elevated in tumors (18, 19). Traskwas also identified through independent efforts as CDCP1 inmicroarray analyses of RNA transcripts overexpressed in coloncancers compared with adjacent normal tissues (20, 21). Inanother line of study, a subtractive immunization screendesigned to identify antibodies against more metastatic variantsof HEp3 carcinoma cells identified SIMA135, identical to Trask/CDCP1, as a putative marker of metastasis, although furtheranalysis in panels of tumor cell lines found no correlation withmetastatic activity (22). We identified Trask/CDCP1 by virtue ofit being the predominant tyrosine phosphorylated and Srcinhibitor suppressible protein in detached mitotic epithelialcells (17). Trask is a substrate of Src and Yes kinases in vitro andin cells, and its phosphorylation can be inhibited by severalclasses of Src-selective tyrosine kinase inhibitors (17). Ininducible transfection models, we found that Trask over-expression and constitutive phosphorylation stabilize a de-tached cellular phenotype and poor ability to spread on plastic.

In the current study, we report a much broader context for thephysiologic phosphorylation of Trask. Although Trask isexpressed in many epithelial tissues, its phosphorylation istightly regulated and only seen in mitotic cells or in cellsundergoing physiologic shedding. However, Trask phosphory-lation is seen in most cancer cell lines in the anchorage-independent state, and widely seen in human preinvasive,invasive, and metastatic epithelial cancers. The widespreadphosphorylation of Trask in cancers identifies a new SRC-drivenpathway, linked with the anchorage-independent attribute ofepithelial tumorigenesis.

Materials andMethods

Cell culture and reagents. All cell lines were obtained from theAmerican Type Culture Collection. Cells were grown in a 1:1 mixtureof DMEM:F12 media supplemented with 10% heat-inactivated fetalbovine serum, 100 U/mL penicillin, 100 Ag/mL streptomycin, and4 mmol/L glutamine, and incubated at 37 C in 5% CO2. MCF10A cellswere grow in DMEM:F12 media supplemented with 5% donor horseserum, 0.5 Ag/mL hydrocortisone, 10 Ag/mL insulin, 20 ng/mL epider-mal growth factor, 100 U/mL penicillin, 100 Ag/mL streptomycin, and4 nmol/L glutamine. To force cells into suspension, cells were washedin PBS and exposed to a 0.05% solution of trypsin or a 2 mmol/Lsolution of EDTA in Hank’s buffer. When required to maintain cellsin suspension and prevent spreading and attachment, cells were spun

Translational Relevance

Src inhibitors are being studied in patients with epithelialmalignancies. This class of antitumor agents presentsunique challenges in clinical testing and development,principally because the tumorigenic functions and down-stream molecular effectors of SRC, and the expected bio-logical outcomes of SRC inhibitor therapy are not yetwell-described. In this article, we present a novel Srcsubstrate that is abberantly phosphorylated in epithelialcancers. Trask phosphorylation is uniquely linked with Srckinases, relatively unique to epithelial tissues, and phy-siologically relevant to the anchorage-independent state inepithelial cells. These attributes makeTrask an importantnewbiomarker thatmaynotonly be aneffector of activatedSRC kinases in epithelial tumors but may have predictiveor pharmacodynamic utility in the clinical development ofSRC inhibitors in epithelial cancers.

Human Cancer Biology

www.aacrjournals.orgClin Cancer Res 2009;15(7) April 1, 2009 2312

Research. on July 21, 2021. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

down, resuspended in growth media, and cultured in ULC plates(Corning) for growth. To harvest lysates, cells were rapidly scraped onice at fixed time points and lysed in radioimmunoprecipitation assaybuffer. Antiphosphotyrosine antibodies (PY99) were purchased fromSantaCruz Biotechnology, Inc. Polyclonal anti-Trask antibodies weregenerated by immunizing rabbits with a recombinant full-length Traskintracellular domain. Monoclonal anti-Trask antibodies were generatedby immunizing mice with a recombinant full-length Trask extracellulardomain and recognize both cleaved and uncleaved forms of Trask.Anti–phospho-Trask antibodies were generated against a phospho-peptide immunogen containing sequences centered around phosphor-ylated tyrosine 743 of Trask in rabbits and affinity purified on aphospho-peptide column. PP1 was purchased from EMD-Calbiochem.Dasatinib was purchased from Bristol-Myers Squibb and purified forin vitro and in vivo use (see Supplementary Methods). pcDNA4-MycTrask was constructed as previously described (17). Total cellularlysates were harvested in modified radioimmunoprecipitation assaybuffer [10 mmol/L Na phosphate (pH 7.2), 150 mmol/L NaCl,0.1% SDS, 1% NP40, 1% Na deoxycholate, protease inhibitors, and1 mmol/L sodium orthovanadate]. For Western blotting, 50 Ag of eachlysate were separated by SDS-PAGE, transferred to membrane, andimmunoblotted using appropriate primary and secondary antibodiesand enhanced chemoluminescence visualization. Immunoprecipita-tions were done by incubating 300 Ag of cellular lysate with theindicated antibodies overnight at 4jC. Immune complexes wereprecipitated by incubation with protein-G sepharose beads, washedseveral times in radioimmunoprecipitation assay buffer, and boiled insample buffer.

Immunohistochemical studies. Deparaffinized sections were rehy-drated and antigen retrieval was done by 15-min incubation in warmtrypsin followed by microwave in 10 mmol/L citrate buffer for total of10 min in 1-min intervals. Slides were then washed and blocked with3% H2O2 followed by blocking in goat serum and primary incubationat 4jC overnight. Secondary staining was done using biotinylated goatanti-rabbit antibodies (Vector Laboratories) and colorized usingVectastain ABC kit (Vector Laboratories) and 3,3¶-diaminobenzidine–H2O2 substrate (Sigma). Slides were then counterstained withhematoxylin, dehydrated through graded alcohols and xylene, andmounted. Slides were studied and imaged under brightfield microsco-py. All staining procedures included positive and negative controls.Controls were prepared from formalin-fixed paraffin-embedded cellbuttons of cell lines with well-defined expression and phosphorylationof Trask by immunoblotting techniques. For Trask immunostains, thepositive control was MDA-MB-468 cells and the negative control wasMCF-7 cells. For phospho-Trask immunostains, the positive control was

MDA-MB-468 cells and the negative controls were MCF-7 cells anddasatinib-treated MDA-MB-468 cells. MDA-468 cells have abundantexpression and phosphorylation of Trask by Western blotting. MCF-7cells have no expression of Trask by Western blotting or by reverse

Fig.1. The expression and phosphorylation ofTrask were determined and compared in a diverse panel of epithelial cancer cell lines.Total cell lysates were immunoprecipitatedwith rabbit anti-Trask antibodies (R6674) and immunoblotted with mouse anti-Trask antibodies (M19) or anti-phosphotyrosine antibodies.The MDA-231cell lysate wasincluded in all blots as a basis to allow comparison ofTrask expression and phosphorylation across different blots.The MCF10A cell lysate is also included at the far right asan example of untransformed epithelial cells. Arrows, 140- and 85-kDa forms ofTrask. IP, immunoprecipitation; IB, immunoblot.

Fig. 2. A, the HCT116 andDLD-1 (colon cancer),MiaPaCa2 (pancreatic cancer),and PC3 (prostate cancer) epithelial cancer cells were harvested either whileadherent (A) or 2hafterbeing forcedinto suspensionbyEDTA(S). Cell lysateswereanalyzed for the expression andphosphorylationofTrask as indicated.Arrows,140- and 85-kDa forms ofTrask.B, the expression andphosphorylationofTraskwasdetermined and compared in L3.6pl-luc pancreatic cancer cells under differentin vitro and in vivo circumstances.These L3.6pl-luc cells express luciferase for usein in vivo imaging. All immunoprecipitates are from equal amounts of cell lysate.Lane1and2, correspond to lysates fromadherent and suspendedcells, respectively,growing in vitro.These cells were also grownorthotopically within the pancreas ofnudemice and their growthmonitoredby in vivo imaging as described inMaterialsandMethods.Well-establishedpancreatic tumors from two different sacrificedmicewere harvested, snap frozen, and their lysates used in lanes3 and 4. A peritonealmetastasis fromone of themicewas removed and lysate used in lane 5.These datacomparing the same cancer cells in vitro and in vivo show that despite themorerestricted phosphorylationofTrask in the in vitro growthmodel, their tumorigenicgrowth invivo is characterizedbymaximalandconstitutivephosphorylationofTrask.

Phosphorylation of Trask in Epithelial Cancers

www.aacrjournals.org Clin Cancer Res 2009;15(7) April 1, 20092313

Research. on July 21, 2021. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

transcription-PCR. Staining intensity was scored and agreed upon bytwo investigators according to the following definitions: 0 indicates novisible expression (similar to negative controls); 1+ indicates expressionthat is faint but is identifiable and above background and negativecontrols; 2+ indicates moderate expression that is clearly evident butalso clearly less intense than positive controls; 3+ indicates intenseexpression similar to positive controls. Only intensity was used forscoring. Percentages of cells were not a parameter used for scoringbecause Trask phosphorylation has focal characteristics in tissues that isrelevant to cell behavior, and some degree of focal variability may alsobe due to uneven fixation artifacts. Immunohistochemically stainedtissue sections were viewed and imaged using an Olympus BX41brightfield microscope fitted with a DP70 digital camera. Images wereacquired using the Olympus DP Controller software and g adjusted foroptimal representation.

Animal studies. Animals were handled according to protocolsapproved by the University of California, San Francisco, InstitutionalAnimal Care and Use Committee. Ex vivo infection of mouse mammaryepithelial cells was done as previously described (23). Briefly, primarymammary epithelial cells were harvested from 10- to 12-wk-old donor

mice and infected in vitro with pMIG-PyMT on 2 successive days andtransplanted into cleared mammary fat pads of 3-wk-old recipient mice.Tumors developed in f3 mo. Tumor-bearing mice were treated withdasatinib or vehicle control administered by Alzet Osmotic Pumps at arate of 0.4 mg/d of dasatinib. Mice were sacrificed and tumors rapidlydissected and immediately fixed in formalin for immunohistochemicalstudies or snap frozen for Western blotting studies. Pancreaticorthotopic models were generated as follows. The mouse pancreaswas surgically exposed through an abdominal excision under anesthe-sia, and 250,000 luciferase-expressing L3.6pl tumor cells implanteddirectly into the pancreas, and the abdomen closed. Tumors werevisible by in vivo imaging within 2 to 3 wk and grossly palpable by 4 to5 wk with a very high take-rate at which time the mice were euthanizedand tumors rapidly excised and snap frozen for Western blotting.

Results

Trask expression and frequent phosphorylation in human cancercell lines. The expression and phosphorylation of Trask was

Table 1.

Breast tissue and cancer

Case Tissue Disease subtype Trask score P-trask score

1 Breast Normal 2+* 02 Breast Normal 2+* 03 Breast Normal 2+* 04 Breast Normal 2+* 05 Breast Normal 2+* 06 Breast Normal 2+* 07 Breast Normal 2+* 08 Breast Normal 2+* 09 Breast Normal 2+* 2+ some areas, hyperplasia

10 Breast LCIS 0/1+ No tumor but 2+ in nl&ADH11 Breast DCIS+LCIS 2+ LCIS, no DCIS 3+ LCIS, also 3+ col hyp, also in nl12 Breast DCIS+LCIS 1+ DCIS, 1+ LCIS 013 Breast DCIS+IDC+ILC 2+ DCIS, 2+IDC, 2+ LN met 0 DICS, 0 IDC, no tumor in LN14 Breast IDC+DCIS 3+ DCIS, 3+ IDC, 2+ LN met 0 DCIS, 0 IDC, 0 LN met15 Breast IDC+DCIS 1+ DCIS, 1+ IDC, 2+ LN met 0 DCIS, 2+ IDC, 0 LN met16 Breast IDC+DCIS 2+ DICS, 2/3+ IDC, 1+ LN met 0 DCIS, 0 IDC, 0 LN met17 Breast IDC+DCIS 2+ IDC 2+ IDC (in emboli)18 Breast IDC+DCIS 1+ DCIS, 1+ IDC 2+ DCIS (basal layer), 0 IDC19 Breast IDC+DCIS+IDC+DCIS 3+ DCIS, 3+ IDC, 3+ LN met 3+ DCIS (focal), 3+ IDC, 0 LN met20 Breast IDC+DCIS+LCIS 2+ DICS, 1/2+ IDC 3+ DCIS, 2+ LCIS, 2+ IDC21 Breast IDC+LCIS 1+ LCIS, 1+ IDC 2+ IDC (invading), 0 LCIS22 Breast ILC 0 023 Breast ILC+LCIS 2+ LCIS, 2+ IDC, 2+ LN met 3+ DCIS, 2+ LCIS, 3+ IDC, 0 LN met24 Breast IDC+ILC 0 ILC, 1+ IDC 2+ in migrating ILC, 2+ IDC25 Breast IDC 2+ 026 Breast IDC 3+ 2+27 Breast LN mets 0/1+ 2+ LN met28 Breast LN mets 1+ Very rare 2+29 Breast Liver mets 2/3+ 1+30 Breast Liver mets 0 3+31 Breast Lung met 2+ 3+32 Breast Liver mets 3+ Insuff. spec33 Breast Liver mets 3+ 034 Breast Liver mets 3+ 035 Breast Liver mets 3+ 036 Breast Liver mets 3+ 0

NOTE. Sections from the indicated archival tumor tissue sources were prepared and stained with anti-Trask and anti–phospho-Trask antibodies.The staining intensity was scored by two observers according to procedures and definitions described in Materials and Methods.All immunostains and their analyses were conducted using well-defined positive and negative controls as described in Materials and Methods.Abbreviations: LCIS, lobular carcinoma in situ; DCIS, ductal carcinoma in situ; IDC, infiltrating ductal cancer; ILC, infiltrating lobular cancer;LN, lymph node; met, metastasis.*Luminal > basal.

Human Cancer Biology

www.aacrjournals.orgClin Cancer Res 2009;15(7) April 1, 2009 2314

Research. on July 21, 2021. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

surveyed in a panel of human epithelial cancer cell lines(Fig. 1). Trask is widely expressed in most of these cancer celllines, although some lung cancer cell lines lack expression ofTrask. Interestingly, many cancer cell lines show aberrantphosphorylation of Trask, some at lower levels (such as MDA-231 or A549), and some at high levels (such as MDA-468 or

Colo205). Cancer cell lines vary in their morphologic character-istics in tissue culture. Interestingly, we found a correlationbetween Trask phosphorylation and the spread or suspendedmorphology of cells. Certain epithelial cancer cells, which,when cultured in vitro, grow naturally in suspension or in semi-suspension show high phosphorylation of Trask (examples:

Table 2.

Colon tissue and cancer

Case Tissue Disease subtype Trask score P-trask score

1 Colon Normal 2+* 0 crypts (3+ mitoses) /2+ apex2 Colon Normal 2+* 0 crypts (3+ mitoses)3 Colon Normal 2+* 0/1 + in crypts/2+ apex4 Colon Normal 2+* 0 crypts (3+ mitoses) /2+ apex5 Colon Normal 2+* 0 crypts/some 2+ apex6 Colon Normal 2+* 0 crypts/1+ apex7 Colon Normal 2+* 0 crypts/2+ apex8 Colon Normal 2+* 0 crypts/some 2+ apex9 Colon Normal 2+* 0 crypts (3+ mitoses) /2+ apex

10 Colon Normal 2+* 0 crypts/2+ apex11 Colon Normal 2+* 0 crypts/2+ apex12 Colon Normal 2+* 0 crypts/2+ apex13 Colon Normal 2+* 0 crypts/2+ apex14 Colon FAP 3+ patchy 3+ patchy15 Colon FAP 3+ apical/1+ crypts/nl colon = 3+ 3+ patchy16 Colon Tubular adenoma 1+ 3+17 Colon Tubulovillous adenoma 1+ focal 3+ patchy18 Colon Tubular adenoma 2+ 3+19 Colon Tubular adenoma 2+ 3+20 Colon Tubular adenoma 2+ 3+ scattered21 Colon Tubular adenoma 2+ 3+22 Colon Tubular adenoma 3+ 1+23 Colon Tubular adenoma 3+ 1+24 Colon Villous adenoma Patches of 0, areas of 2+ 3+ patchy25 Colon Tubular adenoma Patches of 0, areas of 2+ 3+ patchy26 Colon Tubular adenoma + cancer 2+ 3+ in cancer, 0 in adenoma27 Colon Tubular adenoma + cancer 1/2+ 0 cancer, 3+ adenoma28 Colon Cancer 3+ 1+ focal29 Colon Cancer 2+ 2+ focal30 Colon Cancer 2+ 3+ focal31 Colon Cancer 1+ 3+32 Colon Cancer 1/2+ 3+33 Colon Cancer 2+ 3+ focal34 Colon Cancer + LNs 1/2+ 2+35 Colon Cancer + LNs 3+ tumor/2+ LN 2+ tumor/2+ LN36 Colon Cancer + LNs 1+ tumor/1+ LN Mostly 0, scattered 3+37 Colon Liver mets 2+ 3+38 Colon Liver mets 1+ 3+39 Colon Liver mets 1+ 040 Colon Liver mets 2+ 3+ only very focal41 Colon Liver mets 1+ 042 Colon Liver mets 1+ 3+43 Colon Liver mets 1+ 3+44 Colon Liver mets 3+ 1 = focal45 Colon Liver mets 2+ 2+46 Colon Liver mets 2+ 0/rare 3+ cells47 Colon Liver mets 2+ 048 Colon Lung mets 2+ 3+ focal49 Colon Lung mets 3+ 2+50 Colon Lung mets 1+ 051 Colon Lung mets 2+ 2+ focal52 Colon Lung 3+ 053 Colon Liver met 1+ 3+54 Colon Liver met 3+ 3+55 Colon Liver met 2+ 3+

Abbreviation: FAP, familial adenomatous polyposis.*Uniform throughout.

Phosphorylation of Trask in Epithelial Cancers

www.aacrjournals.org Clin Cancer Res 2009;15(7) April 1, 20092315

Research. on July 21, 2021. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

MDA-468, Du4475, Colo205, H69, H510, H740). MCF10A-untransformed breast epithelial cells have abundant expressionof Trask but undetectable phosphorylation.Tumor growth in vivo induces Trask phosphorylation. Alth-

though constitutive phosphorylation of Trask is not seen inmany cancer cell lines when grown in monolayer, Traskphosphorylation can be induced in these cancer cells whendeprived of anchorage. The phosphorylation of Trask is shownfor four different types of cancer cells cultured in suspension(Fig. 2A). Because growth in vitro in monolayer is very differentthan growth in vivo , we compared and contrasted the in vitroand in vivo phosphorylation of Trask in L3.6pl pancreaticcancer cells. In these cells, Trask is minimally phosphorylatedwhen grown in monolayer spread onto plastic but phosphor-ylated when grown in suspension (Fig. 2B, lanes 1 and 2).When these cells are grown as an orthotopic tumor model inthe mouse pancreas in vivo , Trask is phosphorylated, reflect-ing the adhesion-independent nature of tumorigenic growthin vivo (Fig. 2B, lanes 3 and 4). Trask is phosphorylated in theprimary tumor as well as in a metastatic tumor (Fig. 2B, lanes 3,4 , and 5).Trask expression and phosphorylation in human tissues and

tumors. To study the expression and phosphorylation of Traskin vivo , we raised anti-Trask and anti–pY743-Trask polyclonalantibodies. Conditions for immunohistochemical staining ofparaffin-embedded tissues were established, and the specificityof these immunostains for Trask and phospho-Trask wereconfirmed using formalin-fixed paraffin-embedded positiveand negative controls from cell lines that express or do notexpress Trask, and from cell lines with constitutively phosphor-ylated Trask or dephosphorylated Trask due to Src inhibitorpretreatment (Supplementary Materials and Methods). Intyrosine mutation studies, we identified that Trask is phos-phorylated on multiple tyrosine residues simultaneously,including Tyrosines 707, 734, and 743 (data not shown). Ourattempts to raise anti–p-Trask antibodies were most success-ful against Tyr743. The expression and phosphorylation ofTrask was surveyed in a panel of archival human epithelialcancer surgical or biopsy specimens and a number of controlnormal biopsies or excisions by immunohistochemical studies.Trask is widely expressed in many normal epithelial tissues butnot detectable by immunohistochemical analysis in mesenchy-mal tissues or central nervous sytem tissues (data not shown).The expression and phosphorylation of Trask in tissues andcancers of the breast, colon, and lung are shown in Tables 1, 2,and 3.

Anti–phospho-Trask immunostains showed no evidence ofTrask phosphorylation in normal epithlelial tissues (Tables 1, 2,and 3). However, detached mitotic cells have intense Traskphosphorylation. This is best seen in colonic crypts where thereis high mitotic activity (Fig. 3A). The mitotic nature of thesedetached and hyperphosphorylated cells was confirmed byimmunostaining with anti–phospho-histone H3 antibodies,and their colon epithelial identity verified by CDX2 staining(Fig. 3A). In addition, there is frequent nonmitotic phosphor-ylation of Trask at the apices of intestinal villi (Fig. 3B. Apicalcells and 3B). Apical cells are frequently shed into the lumen,and the apical cells with Trask phosphorylation are likely cellsthat are about to be shed. Intestinal epithelial cells arecontinuously replaced by mitotic activity within the cryptsand exfoliation at the apex.

Trask expression and frequent phosphorylation in humancancers. Trask is expressed in most epithelial cancers that wesurveyed. In addition, Trask phosphorylation is seen in manycancers. Trask phosphorylation is seen in early preinvasivecancers, such as tubular adenomas of the colon and rare casesof ductal carcinoma in situ of the breast (Figs. 4A and Figs. 5D),as well as invasive primary tumors, and tumors at sites ofdistant metastases (Fig. 5A-D). In some tubular adenomas,there is abundant tumor cell shedding with high Trask phos-phorylation (Fig. 4A). These Trask-phosphorylated detachedcells are not mitotic cells as confirmed by p-HH3 staining(Fig. 4B), and they are not at the apices of the villi where phy-siologic shedding is seen in normal colons. Rather, they revealan abnormal interphase detachment and shedding in thesepreinvasive cancers. In some tumors, Trask phosphorylation isseen in focal areas and in a patchy distribution, whereas othertumors show a more widespread phosphorylation of Trask.

Table 3.

Lung tissue and cancer

Case Tissue Diseasesubtype

Trask score P-trask score

1 Lung Normal lung 3+* 02 Lung Normal lung 2+* 0, 3+ detached/

mitotic cells3 Lung Normal lung 2+* 04 Lung Normal lung 3+* 0, 3+ rare

detaching cells5 Lung Normal lung 2+* 0, 3+ in mitoses6 Lung Normal lung 3+* 07 Lung Normal lung 3+* 08 Lung Normal lung 2+* 09 Lung Normal lung 3+* 010 Lung Normal lung 2+* 011 Lung Adenoca 3+ Patchy 1+12 Lung Adenoca 2+ 013 Lung Adenoca 2+ 3+14 Lung Adenoca 3+ 015 Lung Adenoca 2+ 0/1+16 Lung Adenoca 3+ Patchy 2+17 Lung Adenoca 3+ 018 Lung Adenoca/BAC 2+ 019 Lung BAC 2+ 020 Lung BAC 0/1+ 021 Lung BAC 0/1+ 022 Lung Larger cell 3+ 2+23 Lung Squamous cell 3+ 024 Lung Squamous cell 1+ 1+25 Lung Squamous cell 3+ Patchy 3+26 Lung Squamous cell 3+ 3+27 Lung Squamous cell 3+ Patchy 3+28 Lung Squamous cell 2+ 029 Lung Squamous cell 3+ 3+30 Lung Small-cell ca 1+ 031 Lung Small-cell ca 1+ 032 Lung Small-cell ca 1+ 033 Lung Small-cell ca 2+ 2/3+34 Lung Small-cell ca 1+ 035 Lung Small-cell ca 1+ 3+36 Lung Small-cell ca 1+ 3+

Abbreviations: Adenoca, adenocarcinoma; BAC, bronchoalveolarcarcinoma; small-cell ca, small-cell carinoma.*Bronchioles.

Human Cancer Biology

www.aacrjournals.orgClin Cancer Res 2009;15(7) April 1, 2009 2316

Research. on July 21, 2021. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

Trask phosphorylation is also seen in lymphatic tumor emboli(Fig. 5D). The widespread phosphorylation of Trask seen intumors is not due to the mitotic phase, as evidenced by thecounterstain that shows that most of the tumor cells are ininterphase. Intense phosphorylation of Trask is seen in somecancers (Fig. 5).Trask is a unique Src substrate. The phosphorylation of

Trask is tightly linked with Src kinases, and in contrast to manyof the well-known substrates of Src kinases, Trask may be aunique subtrate of Src kinases or at minimum, uniquelydependent on Src kinases for tyrosine phosphorylation.Although the transfection and overexpression of Trask in allcells that we have tested induces its phosphorylation, itsoverexpression in SYF (Src/Yes/Fyn null) cells shows noevidence of phosphorylation (Fig. 6A), suggesting that Srckinases are essential for Trask phosphorylation. The cotrans-fection of c-src or c-yes into SYF cells restores Trask phosphor-ylation, further confirming the essential role of src kinases inTrask phosphorylation (Fig. 6A).

To further confirm the link between Src and Traskphosphorylation in vivo , we studied the phosphorylation ofTrask in a model of Src-induced epithelial tumorigenesis.Mouse mammary tumors induced by the expression of the

Polyoma middle T (PyMT) antigen are driven by the PyMT-induced activation of Src (7). We established an orthotopicmodel of PyMT-driven mouse by ex vivo PyMT retroviralinfection of mouse mammary epithelium and reimplantationinto the mammary fat pad. Trask expression and phosphory-lation were studied in these tumors and their lung meta-stases by immunohistochemical methods. Trask is expressedin the normal mammary epithelium but not phosphorylated(Fig. 6B). However, PyMT-induced mammary tumors showsignificant phosphorylation of Trask, particularly in lung meta-sases (Fig. 6B). The phosphorylation of Trask in these tumorsis due to the activation of Src kinases by PyMT and can besuppressed by treating the mice with Src inhibitors (Fig. 6B).The phosphorylation of Trask is not seen in all parts of thetumor but rather in a patchy and focal distribution. The growthof these tumors is highly suppressible by Src inhibitors asexpected from their Src-driven biology (Fig. 6C).

Discussion

A principal function of epithelial cells is adherence to tissuelevel architecture. At its most basic level, this architecture con-sists of monolayers of attached cells against a basement

Fig. 3. A, sections from normal colon biopsieswere prepared and stained with anti ^ phospho-Traskantibodies.The staining intensity was scored bytwo observers according to procedures anddefinitions described in Materials and Methods.All immunostains and their analyses were conductedusing well-defined positive and negative controlsas described in Materials and Methods.Traskphosphorylation is seen in occasional mitotic cells.This is shownhere in the mitotically active cryptregions of normal colonic mucosa.The detachingcells in the crypts were confirmed to be epithelial cellsand not extrinsic cells as shown by staining with theintestinal epithelial differentiation marker CDX2 andconfirmed to be mitotic cells by the mitotic markerphospho-histone H3. B, trask phosphorylation is alsoseen in some clusters of cells at the apices of colonicvilli where shedding commonly occurs.

Phosphorylation of Trask in Epithelial Cancers

www.aacrjournals.org Clin Cancer Res 2009;15(7) April 1, 20092317

Research. on July 21, 2021. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

membrane. This highly structured organization in epithelialtissues allows them to serve as protective barriers and functionto interface with the luminal or outside environment. Certaincellular functions require disengagement from substratum,including functions such as cell migration during developmentor wound repair, mitotic cell division, and physiologic shed-ding. Disengagement from substratum requires the activationof autonomous signaling mechanisms not regulated by tissuearchitecture. The activation of such anchorage-independentprograms and the ability to defy tissue-level imposed programsare also important parts of epithelial tumorigenesis. Much ofthe signaling that underlies the anchorage-dependency andindependency of epithelial cells occurs through tyrosine phos-phorylation. Here, we report that a common signaling pathwaythat characterizes the anchorage-independent growth of epi-thelial tumor cells is the Src family kinase phosphorylation oftheir transmembrane substrate Trask.

In untransformed epithelial cells, the phosphorylation ofTrask is tightly regulated and Trask phosphorylation is only

seen during circumstances of mitotic detachment or physio-logic shedding such as seen at the apex of the intestinal villi.Trask phosphorylation can be induced in vitro by forceddetachment and growth in suspension. In contrast to thetightly regulated phosphorylation of Trask in untransformedepithelial cells and tissues, we find that the Src phosphorylationof Trask is commonly seen in many epithelial cancers andcancer cell lines. We see Trask phosphorylation in all stages ofcancer including preinvasive cancers such as tubular adenomasof the colon, as well as invasive, and metastatic cancers. Assuch, the abberant phosphorylation of Trask is not an activityacquired in later stages of invasive or metastatic cancers. It ispossible that Trask phosphorylation in preinvasive cancersidentifies those preinvasive tumors that are at high risk forinvasion and dissemination. Future studies may attempt to testthat hypothesis; however, this may be too simplistic ahypothesis, considering what we know thus far about thefunctions of Trask. It is likely that both the phosphorylationand the dephosphorylation of Trask are important functions in

Fig. 4. A, representative images from immunostains ofcolon tubular adenomas is shown. In contrast to normalcolonic crypts, whereTrask phosphorylation is restrictedto mitotic cells, in adenomas, there is abundantTraskphosphorylation in many clusters and foci of tumor cells.There is also abundant tumor cell luminal shedding.B, the frequent luminal shedding in adnomas is nonmitotic asconfirmed by the absence of phospho-Histone H3 staining.

Human Cancer Biology

www.aacrjournals.orgClin Cancer Res 2009;15(7) April 1, 2009 2318

Research. on July 21, 2021. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

migrating cells and likely also in invading and metastasizingcells and some level of regulation would be optimal for thesetumorigenic functions. As such, Trask phosphorylation may bea marker of cellular activities at certain points in time or certaintumor areas. Consistent with this, we see only focal phosphor-ylation of Trask in many tumors. The spacial focality we seen in

the immunostains may merely reflect a snapshot in timecaptured at the moment of tissue acquisition and fixation, andmay be evidence of a wider temporal plasticity in Traskphosphorylation reflecting the dynamic nature of tumor celladhesion and migration characteristics in vivo that we areunable to appreciate in these snapshot analyses. The cellular

Fig. 5. Representative images from phospho-Trask immunostains are shown for a number of subtypes of epithelial cancers as indicated.

Phosphorylation of Trask in Epithelial Cancers

www.aacrjournals.org Clin Cancer Res 2009;15(7) April 1, 20092319

Research. on July 21, 2021. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

attributes that Trask phosphorylation may be reporting couldbe disrupted cell-stromal or cell-cell interactions, changes incytoskeletal signaling, or activation of anoikis resistance. Wehave evidence that Trask phosphorylation is regulated by cell-

cell and cell-matrix engagement.7 Several studies have reportedthe activation of Src kinases when epithelial cells detach from

Fig. 6. A, SYF cells were transfected with apcDNA4-MycTrask expression vector and either SrcorYes-expressing vectors, and the expressedTraskproteinwas immunoprecipitatedwith antimyc antibodiesand immunoblotted with antiphosphotyrosine oranti-Trask antibodies. Lanes correspond to (1)untransfected cells, (2) cells transfected withpcDNA4-MycTrask and pCMV6 vector, (3) cellstransfected with pcDNA4-MycTrask and pCMV6-src,(4) cells transfected with pcDNA4-MycTrask andpCMV6-yes. Lane 5 contains the same lysate as lane 4but was immunoprecipitated with IgG control. B,PyMT-induced mammary tumors were generated inmice as described inMaterials andMethods and allowedto grow tof2.5 cm in size. Mice were sacrificied andtumor tissues were fixed in formalin, and lungmetastases were dissected, identified, and fixed.Sections of normalmammary tissue from controlmice aswell as sections from PyMT-induced tumors, and theirlung metastases were studied by immunohistochemicalanalyses to determine the expression andphosphorylation ofTrask.Tumors from mice treated withthe Src inhibitor dasatinib for1wk were also harvestedfor analysis.Trask phosphorylation is induced in thistumor model and dephosphorylated by Src inhibitortherapy. C, mice were orthotopically implanted withPyMT-induced syngeneic tumors, and after tumors wereestablished atf1mo, 30 tumor-bearing mice wererandomized to 2 arms and treated with dasatinib orDMSO control, and tumor sizes were measuredtwice weekly.

7 Manuscript in submission.

Human Cancer Biology

www.aacrjournals.orgClin Cancer Res 2009;15(7) April 1, 2009 2320

Research. on July 21, 2021. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

matrix and its function in averting anoikis when detached(24–26). The specific role of Trask/CDCP1 in promotinganoikis resistance was recently shown in a lung cancer cellline (27).

The apparent deregulation of Trask in preinvasive epithelialtumors is interesting but not altogether surprising. Indeed,increased activity of both Src or Yes is an early event in colonneoplasia and seen very frequently in preinvasive lesions ofthe colon (28–30). The finding of Trask phosphorylation inpreinvasive tumors suggests that some of the molecular andpossibly phenotypic attributes of invasive or metastaticepithelial cancers are already present in preinvasive tumors.Consistent with this, the activation of Src and Yes in preinvasivetumors of the colon is seen predominantly in tumors at highestrisk for progression to invasive cancer (28, 29). The deregula-tion of Trask in preinvasive tumors could have phenotypicconsequences, but these would likely be clinically undetectablewith little symptomatic consequence. These could includephenotypes such as shedding of tumor cells. However, due tothe intact nature of the basement membrane in preinvasivetumors, the shedding would be entirely into the epitheliallumen, which would be of little clinical or symptomaticconsequence to affected patients. Indeed, we do find significantinterphase tumor cell shedding in tubular adenomas, which wedo not see in normal colon tissues (Fig. 4A). Trask phosphor-ylation could also increase the survival of luminally shed cells.Consistent with this, exfoliated tumor cells are detected muchmore succesfully in the stools of patients with tubularadenomas or colon cancers compared with the stools ofpatients with inflammatory diseases of the colon or normalcolons (31, 32), and exfoliated breast epithelial cells aredetected in nipple aspirates of women with preneoplasticlesions of the breast (33). Although not clinically significant inpreinvasive disease, the phenotypic consequence of deregulatedTrask phosphorylation may be much more significant afterprogression to invasive disease because the ability to surviveand migrate within connective tissues, lymphatics, and thebloodstream can be mediated through Trask phosphorylation.Consistent with this, we do see Trask phosphorylation in tumoremboli (Fig. 5D).

When we use the same tumor cells to compare in vitro andin vivo growth, we find that Trask phosphorylation in the in vivomodel resembles the anchorage-deprived state of the in vitromodel, not the adherent monolayer growth (Fig. 2B). This isnot surprising because it is widely recognized that the growth oftumor cell lines in flattened monolayers on tissue culture–treated plasticware represents an artificial state with manydifferences from their growth in vivo.

The mechanisms by which Trask may promote survival insuspended epithelial cells are currently unknown. Trask/CDCP1 was found to bind PKC y in a phosphorylation-

dependent manner through the C2 domain of PKC y (34). PKCy is thought to regulate apoptosis, although its role may becomplex and both proapoptotic and antiapoptotic functionshave been attributed to PKC y (35). The functions of PKC y maybe context dependent. In murine mammary epithelial cells,PKC y activation promotes the ability to survive and grow inan anchorage-independent manner (36). Almost surely Traskregulates functions other than survival because we havepreviously shown that the forced overexpression and over-phosphorylation of Trask retains detached epithelial cells insuspension and prevents them from respreading on substrate,which suggests that Trask phosphorylation affects cytoskeletalsignaling as well (17). We have been studying the role of Traskin cytoskeletal signaling and have identified mechanistic linksbetween Trask and components of the actin cytoskeleton,although this evidence is only preliminary at this point.

The fact that the phosphorylation of Trask is entirelydependent on src kinases may suggest its use as an in vivomarker of the activity of Src kinases. Its phosphorylation inPyMT-induced tumors and in many human cancers isconsistent with this. However, it is unlikely that the level ofTrask phosphorylation faithfully reflects the activity of Srckinases. In the tumor cell panel reported here, the level ofTrask phosphorylation varies widely and does not parallel themuch more uniform activity of Src kinases previously publishedby ourselves and others (37, 38). In addition, althoughepithelial cells undergoing mitosis or detachment show modestincreases in Src kinase activity (39), the abundant phosphor-ylation of Trask seen in these circumstances greatly exceedsincreases in Src kinase activities (26). In addition, Traskphosphorylation seems to be partly dependent on Traskexpression, and its phosphorylation can be induced by itsoverexpression (17).8 Clearly, there are many variables otherthan the steady-state activity of Src kinases that regulate the Src-mediated phosphorylation of Trask.

In summary, Trask is a recently identified Src substrate thatmay be highly relevant to epithelial tumorigenesis. Theevidence indicates that Trask is a principal marker ofanchorage-independent signaling in epithelial cells and itsSrc-driven phosphorylation is tightly regulated in epithelialcells. The deregulation of Trask phosphorylation is a commonand early event in epithelial tumorigenesis, and furthermechanistic exploration of its signaling functions should beof high priority.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

We thank the Immunohistochemical Core, the Tissue Core, and the MousePathology Core facilities of the University of California at San Francisco Compre-hensive Cancer Center for their services in this work, and in particular, the technicalassistance of Lorretta Chan of the immunohistochemical core.8 Manuscript in preparation.

References1. Martin GS. The road to Src. Oncogene 2004;23:7910^7.2. Cartwright CA, EckhartW, Simon S, Kaplan PL. Celltransformation by pp60c-src mutated in the carboxy-terminal regulatory domain. Cell 1987;49:83^91.

3. Jove R, Hanafusa H. Cell transformation by the viralsrc oncogene. Annu Rev Cell Biol 1987;3:31^56.4. Irby RB, Yeatman TJ. Role of Src expression andactivation in human cancer. Oncogene 2000;19:5636^42.

5.Muthuswamy SK, MullerWJ. Activation of Src familykinases in Neu-induced mammary tumors correlateswith their association with distinct sets of tyrosinephosphorylated proteins in vivo. Oncogene 1995;11:1801^10.

Phosphorylation of Trask in Epithelial Cancers

www.aacrjournals.org Clin Cancer Res 2009;15(7) April 1, 20092321

Research. on July 21, 2021. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

6.Guy CT, Cardiff RD, MullerWJ. Induction of mamma-ry tumors by expression of polyomavirus middleToncogene: a transgenic mouse model for metastaticdisease. Mol Cell Biol 1992;12:954^61.7. Guy CT, Muthuswamy SK, Cardiff RD, Soriano P,Muller WJ. Activation of the c-Src tyrosine kinase isrequired for the induction of mammary tumors intransgenic mice. Genes Dev1994;8:23^32.8. Tan M, Li P, Klos KS, et al. ErbB2 promotes Srcsynthesis and stability: novel mechanisms of Src acti-vation that confer breast cancer metastasis. CancerRes 2005;65:1858^67.9. Pories SE, Hess DT, Swenson K, et al. Overexpres-sion of pp60c-src elicits invasive behavior in rat colonepithelial cells. Gastroenterology1998;114:1287^95.10. Vultur A, Buettner R, Kowolik C, et al. SKI-606(bosutinib), a novel Src kinase inhibitor, suppressesmigration and invasion of human breast cancer cells.Mol CancerTher 2008;7:1185^94.11. Ito H, Gardner-Thorpe J, Zinner MJ, Ashley SW,Whang EE. Inhibition of tyrosine kinase Src sup-presses pancreatic cancer invasiveness. Surgery2003;134:221^6.12. Yezhelyev MV, Koehl G, Guba M, et al. Inhibition ofSRC tyrosine kinase as treatment for human pan-creatic cancer growing orthotopically in nude mice.Clin Cancer Res 2004;10:8028^36.13. Boyer B, BourgeoisY, Poupon MF. Src kinase con-tributes to the metastatic spread of carcinoma cells.Oncogene 2002;21:2347^56.14. Frame MC. Newest findings on the oldest onco-gene; How activated src does it. J Cell Sci 2004;117:989^98.15. Frame MC, Fincham VJ, Carragher NO,Wyke JA.v-Src’s hold over actin and cell adhesions. [Review][119 refs]. Nat Rev Mol Cell Biol 2002;3:233^45.16. Ishizawar R, Parsons SJ. c-Src and cooperatingpart-ners inhumancancer. Cancer Cell 2004;6:209^14.17. Bhatt AS, Erdjument-Bromage H, Tempst P, CraikCS, Moasser MM. Adhesion signaling by a novelmitotic substrate of src kinases. Oncogene 2005;24:5333^43.

18. Bhatt AS,Welm A, Farady CJ,Vasquez M,WilsonK, Craik CS. Coordinate expression and functionalprofiling identify an extracellular proteolytic signal-ing pathway. Proc Natl Acad Sci U S A 2007;104:5771^6.19. Bhatt AS, Takeuchi T, Ylstra B, et al. Quantitationof membrane type serine protease 1 (MT-SP1) intransformed and normal cells. Biol Chem 2003;384:257^66.20. Scherl-MostageerM, SommergruberW, Abseher R,Hauptmann R, Ambros P, Schweifer N. Identificationof a novel gene, CDCP1, overexpressed in human co-lorectal cancer. Oncogene 2001;20:4402^8.21. Perry SE, Robinson P,MelcherA, et al. Expressionofthe CUB domain containingprotein1 (CDCP1) gene incolorectal tumour cells. FEBSLett 2007;581:1137^42.22. Hooper JD, Zijlstra A, Aimes RT, et al. Subtractiveimmunization using highly metastatic human tumorcells identifies SIMA135/CDCP1, a 135 kDa cell sur-face phosphorylated glycoprotein antigen. Oncogene2003;22:1783^94.23.Welm AL, Sneddon JB,Taylor C, et al. The macro-phage-stimulating protein pathway promotes metas-tasis in a mouse model for breast cancer and predictspoor prognosis in humans. Proc Natl Acad Sci U S A2007;104:7570^5.24.WindhamTC, Parikh NU, SiwakDR, et al. Src activa-tion regulates anoikis in human colon tumor cell lines.Oncogene 2002;21:7797^807.25.Wei L,YangY, Zhang X,Yu Q. Altered regulation ofSrc upon cell detachment protects human lung adeno-carcinoma cells from anoikis. Oncogene 2004;23:9052^61.26. Loza-Coll MA, Perera S, ShiW, FilmusJ. A transientincrease in the activity of Src-family kinases inducedby cell detachment delays anoikis of intestinal epithe-lial cells. Oncogene 2005;24:1727^37.27.UekitaT, Jia L, Narisawa-SaitoM,YokotaJ, KiyonoT,Sakai R. CUB domain-containing protein 1 is a novelregulator of anoikis resistance in lung adenocar-cinoma. Mol Cell Biol 2007;27:7649^60.28. Cartwright CA, MeislerAI, EckhartW. Activation of

the pp60c-src protein kinase is an early event incolonic carcinogenesis. Proc Natl Acad Sci U S A1990;87:558^62.29. Pena SV, Melhem MF, Meisler AI, Cartwright CA.Elevated c-yes tyrosine kinase activity in premalignantlesions of the colon [see comments]. Gastroenterolo-gy1995;108:117^24.30. Iravani S, MaoW, Fu L, et al. Elevated c-Src proteinexpression is an early event in colonic neoplasia. LabInvest 1998;78:365^71.31.Villa E, Dugani A, Rebecchi AM, et al. Identificationof subjects at risk for colorectal carcinoma through atest based on K-ras determination in the stool. Gastro-enterology1996;110:1346^53.32.Ahlquist DA, SkoletskyJE, Boynton KA, et al. Colo-rectal cancer screening by detection of altered humanDNA in stool: feasibility of a multitarget assay panel.Gastroenterology 2000;119:1219^27.33.Wrensch MR, Petrakis NL, Miike R, et al. Breastcancer risk in women with abnormal cytology in nip-ple aspirates of breast fluid. J Natl Cancer Inst 2001;93:1791^8.34. Benes CH,Wu N, Elia AE, Dharia T, Cantley LC,Soltoff SP. The C2 domain of PKCy is a phosphotyro-sine binding domain. Cell 2005;121:271^80.35. Brodie C, Blumberg PM. Regulation of cell apopto-sis by protein kinase c y. Apoptosis 2003;8:19^27.36. Grossoni VC, Falbo KB, Kazanietz MG, de KierJoffe ED, Urtreger AJ. Protein kinase C y enhancesproliferation and survival of murine mammary cells.Mol Carcinog 2007;46:381^90.37.MoasserMM, SrethapakdiM, Sachar KS, KrakerAJ,Rosen N. Inhibition of Src kinases by a selective tyro-sine kinase inhibitor causes mitotic arrest. Cancer Res1999;59:6145^52.38. BiscardiJS, Belsches AP, Parsons SJ. Characteriza-tion of human epidermal growth factor receptor andc-Src interactions in human breast tumor cells. MolCarcinog1998;21:261^72.39. Park J, Meisler AI, Cartwright CA. c-Yes tyrosinekinase activity in human colon carcinoma. Oncogene1993;8:2627^35.

Human Cancer Biology

www.aacrjournals.orgClin Cancer Res 2009;15(7) April 1, 2009 2322

Research. on July 21, 2021. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

2009;15:2311-2322. Clin Cancer Res   Ching Hang Wong, Frederick L. Baehner, Danislav S. Spassov, et al.   Many Human Epithelial CancersTightly Regulated in Normal Epithelia but Widespread in Phosphorylation of the src Epithelial Substrate Trask Is

  Updated version

  http://clincancerres.aacrjournals.org/content/15/7/2311

Access the most recent version of this article at:

  Material

Supplementary

 

http://clincancerres.aacrjournals.org/content/suppl/2009/03/24/1078-0432.CCR-08-2533.DC1Access the most recent supplemental material at:

   

   

  Cited articles

  http://clincancerres.aacrjournals.org/content/15/7/2311.full#ref-list-1

This article cites 39 articles, 11 of which you can access for free at:

  Citing articles

  http://clincancerres.aacrjournals.org/content/15/7/2311.full#related-urls

This article has been cited by 10 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

  .pubs@aacr.orgDepartment 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://clincancerres.aacrjournals.org/content/15/7/2311To request permission to re-use all or part of this article, use this link

Research. on July 21, 2021. © 2009 American Association for Cancerclincancerres.aacrjournals.org Downloaded from