molecular imaging of tgfb-induced smad2/3 phosphorylation ...tgfb signaling pathway regulates growth...

Imaging, Diagnosis, Prognosis

Molecular Imaging of TGFb-Induced Smad2/3Phosphorylation Reveals a Role for Receptor TyrosineKinases in Modulating TGFb Signaling

ShyamNyati1,2,KatrinaSchinske2,DipankarRay2,MukeshNyati2,BrianDaleRoss1,3, andAlnawazRehemtulla1,2

AbstractPurpose: The dual modality of TGFb, both as a potent tumor suppressor and a stimulator of tumor

progression, invasion, andmetastasis,make it a critical target for therapeutic intervention in human cancers.

The ability to carry out real-time, noninvasive imaging of TGFb-activated Smad signaling in live cells and

animal models would significantly improve our understanding of the regulation of this unique signaling

cascade. To advance these efforts, we developed a highly sensitive molecular imaging tool that repetitively,

noninvasively, and dynamically reports on TGFBR1 kinase activity.

Experimental Design: The bioluminescent TGFbR1 reporter construct was developed using a split firefly

luciferase gene containing a functional sensor of Smad2 phosphorylation, wherein inhibition of TGFbreceptor1 kinase activity leads to an increase in reporter signaling. The reporter was stably transfected into

mammalian cells and used to image in vivo and in vitro bioluminescent activity as a surrogate formonitoring

TGFBR1 kinase activity.

Results: The reporter was successfully used to monitor direct and indirect inhibition of TGFb-inducedSmad2 and SMAD3 phosphorylation in live cells and tumor xenografts and adapted for high-throughput

screening, to identify a role for receptor tyrosine kinase inhibitors as modulators of TGFb signaling.

Conclusion: The reporter is a dynamic, noninvasive imaging modality for monitoring TGFb-inducedSmad2 signaling in live cells and tumor xenografts. It has immense potential for identifying novel effectors

of R-Smad phosphorylation, for validating drug–target interaction, and for studying TGFb signaling in

different metastasis models. Clin Cancer Res; 17(23); 7424–39. �2011 AACR.

Introduction

TGFb signaling pathway regulates growth inhibition andtumor suppression through control of cell proliferation,differentiation, apoptosis,migration, andadhesion(1).Con-sequently, when the TGFb signaling pathway is disrupted,either through genetic mutations of key signaling compo-nents or due to loss of its growth inhibition capabilities,tumorigenesis occurs. It has been shown that TGFb plays adual role in cancer development (2, 3). During early stages oftumorigenesis, when the tumor cells are premalignant, TGFbsuppresses cellular outgrowthwhereas at later stages, follow-ing alteration of TGFb responsiveness and increased ligand

expression, it promotes tumor progression, invasion, andmetastasis. Tumor cells that have lost TGFb-mediated tumorsuppression but maintain intact components of its coresignalingpathwayareespeciallyaggressive.Thesecells exploitTGFb for pro-oncogenic activities: epithelial-to-mesenchy-mal transition (EMT), tumor invasion, mitogen production,metastatic dissemination, and immune evasion (2, 3).

TGFb cellular activity is mediated through a receptorserine–threonine kinase (RSK) complex. TGFb initiates theinteraction of TGFb type-1 and type-2 receptors (TGFBR1and TGFBR2, respectively), which phosphorylate theC-terminal SXS motif of receptor-regulated Smads (R-Smads), Smad2, and Smad3. The phosphorylated R-Smadscomplex with Co-Smad4 translocate into the nucleus and,along with cell-specific transcriptional cofactors, interactwith Smad-binding elements (SBE) in the DNA to regulatetranscription of target genes. The C-terminal phosphoryla-tion of R-Smads is a critical event in TGFb signaling andtranscriptional regulation. It is necessary for Smad complexassembly and disassembly, nuclear translocation, and tran-scriptional activity and stability required for Smad-mediat-ed cellular responses (4).

Despite the importance of TGFb-mediated Smad signal-ing in the progression of human cancers, methods formonitoring in vivo and in vitro Smad phosphorylation are

Authors' Affiliations: 1Center for Molecular Imaging; and Departments of2Radiation Oncology and 3Radiology, University of Michigan, Ann Arbor,Michigan

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

Corresponding Author: Alnawaz Rehemtulla, Center for Molecular Imag-ing, Department of Radiation Oncology, University of Michigan MedicalCenter, 109 Zina Pitcher Place, BSRB Level A528, Ann Arbor, MI 48109.Phone: 734-764-4209; Fax: 734-615-5669; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-11-1248

�2011 American Association for Cancer Research.

ClinicalCancer

Research

Clin Cancer Res; 17(23) December 1, 20117424

on February 10, 2020. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst September 26, 2011; DOI: 10.1158/1078-0432.CCR-11-1248





http://clincancerres.aacrjournals.org/



limited. Indirect molecular imaging techniques have beendeveloped using fluorescent-tagged proteins in the TGFb–Smad signaling pathway to study cellular localization ofSmad proteins in breast cancer cells (5, 6) and protein–protein interaction of critical elements of TGFb signaling inXenopus embryogenesis (7). More direct measures of TGFbsignaling have been designed using a firefly luciferasereporter linked to SBE, to measure transcription of targetgenes in breast cancer metastasis (8, 9), skin wound healing(10), neurodegeneration, (11), and response to injury (12).These imaging methods are limited in that they do notmeasure the key upstream event in TGFb signaling, thephosphorylation of R-Smads.In this study, we developed amolecular imaging reporter

for noninvasive, repetitive monitoring of TGFb activityusing the critical R-Smad phosphorylation step in the TGFbsignaling pathway. The reporter showed higher sensitivityto changes in Smad phosphorylation than measured byconventional Western blot analysis and provided newinsight into pharmacodynamics of TGFb inhibitorSB431542 in vivo. It was successfully adapted for high-throughput screening and utilized to delineate key mod-ulators of Smad2 signaling. The reporter shows immensepotential for identifying novel effectors of R-Smad phos-phorylation, for validating drug–target interactions and forstudying TGFb signaling in different metastasis models.

Materials and Methods

Construction of the bioluminescent TGFbR reporterTo image TGFBR1 kinase activity, we constructed a

recombinant chimeric reporter (Fig. 1A and B) consistingof a 15 amino acid target peptide sequence derived fromthe C-terminus of Smad2 containing the SXS motif, aa453–467, a TGFBR1 substrate (13). The bioluminescent

TGFbR reporter (BTR) reporter was constructed throughfusion of the Smad2 peptide sequence and the FHA2 phos-pho-serine-threonine binding domain of Rad53, whichwere flanked by the amino-(N-Luc) and carboxyl-(C-Luc)terminal domains of the firefly luciferase reporter molecule(14). The cloning primers were designed with a linkersequence (GlyGlyAlaGlyGly) on both ends and XbaI/XmaIprecut sites. The BTR-WT was constructed using the appro-priate primers (CTA GAG GAG GAA GTG GAG GGT TAACTCAGATGGGATCCCCTTCAGTGCGTTGCTCAAGCATGTCACandCCGGGTGACATGCTTGAGCAACGCACTGAAGGGGATCCCATCTGAGTTAACCCTCCACTTCCTCCT). The mutant reporter BTR-MUT was generated bysubstituting both serines of the SXS motif with alanineusing a single-primer mutagenesis protocol, with minormodifications as described earlier (15), using the followingprimer: CCT TCA GTG CGT TGC TCA GCC ATG GCA CCCGGG GGA GGA GCT GGA GGA TCC GGT TAT GT. All theclones were sequence verified.

Cell culture and transfectionThe human lung carcinoma cell line A549 [American

Type Culture collection (ATCC)] was maintained in RPMI-1620 media supplemented with 10% heat-inactivated FBS,1% glutamine, and 0.1%penicillin/streptomycin/gentamy-cin (GIBCO-Invitrogen). Cell cultures were grown in ahumidified incubator at 37�C and 5% CO2. Breast cancercell line 1833 derived from MDA-MB-431 (16) was kindlyprovided by Dr. Joan Massague and maintained in Dulbec-co’smodified Eagle’smediummedia in the same conditionsmentioned above. Cell cultures were grown in a humidifiedincubator at 37�C and 5% CO2. ATCC cell line was testedroutinely for Mycoplasma and purity. All ATCC lines wereexpanded immediately upon receipt, and multiple vials oflow passage cells were maintained in liquid N2. No vial ofcells was cultured for more than 1 to 2 months. Stable celllines were developed by transfecting the BTR reporter plas-mids into A549 and 1833 cells using Fugene (Roche Diag-nostics), and resulting clones were selected using 500 mg/mL (A549) or 250 mg/mL (1833) G418 (Invitrogen) con-taining media. Twenty-four stable clones were selected forboth wild-type and mutant cell lines. The BTR-WT expres-sing clones were analyzed by bioluminescent imaging fol-lowing treatment with 10 mmol/L SB431542 (CaymanChemical). Single-cell subclones were generated and main-tained in 250 mg/mL G418 containing media.

Antibodies and reagentsRabbit polyclonal antibodies to phospho(S465/467)

Smad2, Smad2, p(S423/425)Smad3, Smad6, TGFbR1/ALK5, TGFbRII, C-Myc, C-Jun, p(Y845)EGFR, epidermalgrowth factor receptor (EGFR), p(Y1248)HER2, p(T180/Y182)P38MAPK, P38MAPK, p(P85-Y458/P55-Y199)PI3-K,PI3-K, p(T183/Y185)JNK, c-jun-NH2-kinase (JNK), andglyceraldehyde-3-phosphate dehydrogenase (GAPDH)were obtained from Cell Signaling Technology; anti–E-cadherin and N-cadherin from BD Biosciences; anti-pSer,Smad3, and SMAD7 from Invitrogen; Smad4 (Santa Cruz

Translational Relevance

The response of cells to TGFb stimulation is highlycontextual in cancer. Loss or mutational inactivation ofgenes that mediate the TGFb pathway early in carcino-genesis is indicative of its role as a tumor suppressor. Yet,paradoxically, TGFb also modulates processes such ascell invasion and modifications of the tumor microen-vironment that are critical for metastasis and tumorestablishment at distant sites. Although therapeuticagents have been developed targeting this pathway, theirrational use will require a better understanding of theregulation of its signaling. In this effort, we in this articledescribe the development of a reporter molecule where-in TGFb receptor I kinase activity can be imaged dynam-ically and noninvasively in living subjects. We alsoprovide preliminary results that TGFb signaling activityis impacted by seemingly unrelated signaling cascades,providing potential explanation for the contextual dif-ferences in TGFb downstream signaling events in cancer.

Molecular Imaging of TGFb Kinase Activity

www.aacrjournals.org Clin Cancer Res; 17(23) December 1, 2011 7425




SMAD2

453-467

BioluminescenceBioluminescence

N-L

uc

C-L

uc

N-Luc

C-L

uc

P

FHA2

domain

SMAD2 453-467

(phospho-Ser465/67)

TGFβ

N-Luc FHA2 C-Luc SMAD2453-467

Linker

LTQMGSPSVRCSSMS BTR-WT

TGFβ inhibitors

Phosphatase

LTQMGSPSVRCSAMA BTR-MUT

A

B

C′

E F

D

C

Time (min)

0 μmol/L0.1 μmol/L1 μmol/L5 μmol/L10 μmol/L25 μmol/L

Fo

ld in

du

ctio

n

0

2

4

6

8

10

12

0 60 120 180 240 300 360

SB431542 + TGFβ SB431542 10 μmol/L +TGFβ

pSMAD2

tSMAD2

Luciferase

GAPDH

Mo

ck

1 μ

mo

l/L

5 μ

mo

l/L

10

μm

ol/L

25

μm

ol/L

0.1

μm

ol/L

TG

Fβ

Mo

ck

0.5

h

1 h

2 h

4 h

6 h0.2

5h

TG

Fβ

0

1

2

3

4

5

6

7

8

9

10

0 60 120 180 240 300 360

BTR-WT

BTR-MUT

Time (min)

Fo

ld in

du

ctio

n

0

2

4

6

8

10

12

0 0.1 1 5 10 25

SB431542 (μmol/L)

Fo

ld in

du

ctio

n

r = 0.88

0

2

4

6

8

10

12

0 5 10 15 20 25

Rep

ort

er F

old

act

ivat

ion

SB431542 concentration

r = -0.985

PSmad2 levelfold change

Reporter foldinduction

P = 9.04 E-6

P = 1.25 E-6

P = 0.0003

P = 0.01

P = 4.037 E-6

R 2 = 0.773

1 7.9 7.4 5.7 1.7 0.8 0.1

R 2 = 0.9703

FHA2

domain

Nyati et al.

Clin Cancer Res; 17(23) December 1, 2011 Clinical Cancer Research7426




Biotechnology); and anti-HER2 and anti-firefly luciferaseantibodies from Millipore. The horseradish peroxidase(HRP)-conjugated secondary antibodies were from JacksonLaboratories. TGFBR1/ALK5 inhibitor SB431542, AG1296,Tyrphostin AG1478, and SP600125 were obtained fromCayman Chemical; pan-RTK (receptor tyrosine kinase)inhibitor PP2 from Calbiochem-EMD chemicals; andSB203580 (Invivogen). Lapatinib in tablet form (Tykerb)was purchased from the University of Michigan pharmacy.Erlotinib was a kind gift from Genentech. D-Luciferin waspurchased from Promega Corp. An 84-compound kinase-specific library was obtained from Timtek Drug Discovery.SBE4-Luc reporter plasmid (17) was a kind gift of Dr BertVogelstein (Addgene plasmid #16495).

Bioluminescence reporter assay and live-cell imagingFor the reporter assay, A549 and 1833 BTR cell lines were

seeded in 96-well (5 � 103 cells per well), clear bottom,white-walled plates (Corning, Inc.) 48 hours before assay-ing. Cells were treated in serum-free media with variedconcentrations of test compounds for indicated time per-iods. Luminescence was read with an Envision 2104 multi-label plate reader (PerkinElmer) for 5 repeats after additionof D-luciferin (100 mg/mL final concentration) to the cellmedium; each experiment was done in triplicate. For thehigh-throughput screen, 10 mL of an intermediate dimethylsulfoxide (DMSO) stock of each compound in the kinaseinhibitor library was added to the cell medium using aBeckmanBiomekNXP Laboratory AutomationWorkstation(BeckmanCoulter Inc.), yielding a final assay concentrationof 10 mmol/L for each compound; all appropriate controlswere included and cells incubated for 4 hours beforeD-luciferin was added and bioluminescence measured.Live-cell luminescent imaging was achieved by adding D-

luciferin (100 mg/mL final concentration) to the growthmedium on cells seeded in black-walled, clear-bottom, 96-well plates. Photon counts for each conditionwere acquired5 minutes after incubation with D-luciferin using an IVIS200 imaging system (Xenogen).

siRNA transfection analysisTGFBR1 and TGFBR2 siRNA, an siGENOME Smart Pool,

and nonsilencing siRNA (NSS) were synthesized by Dhar-macon Research. Knockdown experiments were done

by transfecting A549 and 1833 BTR-WT reporter cells with100 nmol/L siRNA for TGFBR1 and TGFBR2, as well asfor nontargeted siRNA (NSS) as a negative control usingDharmafect1 (Dharmacon Research). The transfected cellswere incubated for 60 hours, serum starved overnight, and10 ng/mL TGFb added 1 hour before evaluating reporteractivity with bioluminescent imaging and Western blotanalysis. Fold change in reporter activity was calculatedover change in activity in NSS-treated cells.

Western blot analysis and immunoprecipitationWestern blot analysis was carried out using standard

protocols. A549 and 1833 BTR cell lines were grown inculture dishes, treated with select compounds for desig-nated time periods, and cell lysates resolved on SDS-PAGE gels and transferred to polyvinylidene difluoride(PVDF) membranes. Membranes were probed againstspecific primary antibodies followed by HRP-conjugatedsecondary antibodies, then visualized using the EnhancedChemiluminescence (ECL) Western Blotting System(GE Healthcare).

Immunoprecipitation of the BTR reporter moleculewas carried out by incubating cell lysate (400 mg protein)with 5 mg luciferase-specific antibody for 1 hour. Theimmune complex was captured using 20 mL slurry ofprotein A/G-coupled Sepharose beads (GE Healthcare)for 1 hour, washed 3 times with radioimmunoprecipita-tion assay buffer. The resulting pellet was resolved bySDS-PAGE and transferred to PVDF membrane for West-ern analysis.

To study the effect of various inhibitors on Smad2/3complex formations with Smad4, coimmunoprecipitationwas carried out using a Smad3 antibody. A549 cells wereserum starved and treated with 10 mmol/L inhibitors in thepresence of 10 ng/mLTGFb for 1 hour. Lysatesweremade innative lysis buffer (50mmol/LTris pH7.4, 1%NP40, 0.25%deoxycholate sodium salt, 150mmol/LNaCl, 10%glycerol,and 1 mmol/L EDTA) supplemented with 1� PhosStop(Roche), 1� protease inhibitor cocktail (Roche), sodiumortho vanadate, sodium fluoride, phenylmethylsulfonyl-fluoride, andb-glycerol phosphate. The sampleswere gentlyrocked for 2 hours at 4�C. Immune complexes were cap-tured using 40 mL slurry of protein A/G-coupled Sepharosebeads (GE Healthcare) for 1 hour. The resulting pellet was

Figure 1. Construction and validation of bioluminescent TGFb reporter (BTR) with TGFBR1 kinase inhibitor, SB431542. A, the domain structure of the BTRconstruct: Two versions of the BTR construct were developed; the BTR-WT construct contains the wild-type Smad2 C-terminus, aa 463–467, and the BTR-MUT construct in which alanines were substituted for both serines of the SXSmotif. B, themechanism of action of the BTR reporter involves TGFBR1 kinase–dependent phosphorylation of the Smad2 target peptide which results in its interaction with the FHA2 domain. In this form, the reporter has minimalbioluminescence activity. In the absence of the kinase activity, the target peptide no longer interacts with the FHA2 domain allowing the N-Luc and C-Lucdomains to reassociate, restoring bioluminescence activity. C, A549 cells stably expressing theBTR-WT reporterwere treatedwith increasing concentrationsof SB431542 (0.1, 1, 5, 10, and 25 mmol/L) in presence of TGFb (10 ng/mL) and bioluminescence was measured serially from baseline through 6 hours. Insertdisplays a representative bioluminescent image of BTR reporter response with increasing concentrations of SB431542-treated cells. The change inbioluminescence activity over mock treatment (DMSO) levels was calculated and plotted as fold induction. Bioluminescent measurements are in triplicates;error bars denote SEM. C', reporter fold induction with SB431542 concentration is plotted to illustrate dose-dependent increase in the reporter activity.Pearson correlation coefficient (r), goodness of fit (R2), and statistical significance (P) are calculated and shown on the plot. D, cell lysates were prepared andanalyzed by Western blot using anti–phospho-(S465–S467) Smad2, total Smad2, luciferase, and GAPDH antibodies. Level of pSmad2 was measured usingimageJ. E, Pearson correlation coefficient was calculated between pSmad2 level change and reporter fold induction. A very strong negative correlation(r ¼ �0.985) was observed. F, A549 cells stably expressing BTR-WT and BTR-MUT reporters were treated in triplicate with 10 mmol/L SB431542 andbioluminescence was measured serially for 6 hours. The change in bioluminescence activity over vehicle control was calculated and plotted as fold change.






washed and resolved by SDS-PAGE and transferred to PVDFmembrane and probed using a human anti-Smad4antibody.

Transcriptional response assayA549 cells were transiently transfected with SBE4-Luc

reporter plasmid along with Gaussia luciferase plasmid(pEF-GLuc) as internal control. Transfected cells were incu-bated with 10 mmol/L inhibitors in the presence or absenceof 10 ng/mL TGFb in media containing 1% FBS. SBE4-Lucreporter activity was measured 22 hours posttreatment.Each condition was done in triplicate.

In vivo bioluminescence imaging of mouse tumormodels

Tumor xenografts expressing BTR-WTwere established byimplanting 2.5� 106 stably transfected BTR-WT A549 cellson eachflankof 4- to 6-week-old female athymicmice of thegenotype CD-1 nu/nu (Charles River Laboratories). Whentumors reached a volume of approximately 40 to 60 mm3,in about 4 weeks, bioluminescence activity was monitored.The mice were anesthetized with 2% isofluorane/airmixture and given a single intraperitoneal injection of150 mg/kg luciferin in normal saline. Image acquisitionwas initiated after injection of luciferin and serial back-ground bioluminescence images acquired for 4 hours priorto drug administration. Mice were injected with a singleintraperitoneal dose of 10 mg/kg body weight withSB431542, a dosing regimen previously reported to inhibitTGFb signaling in vivo (18, 19), or vehicle control (DMSO)and bioluminescent images done before treatment as wellas, at 1, 4, 8, and 24 hours.

In parallel, tumor xenografts were generated in miceusing A549 BTR-WT and MUT reporter expressing cells.Mice were treated as described and bioluminescence wasmeasured 4 hours after treatment. All mice experimentswere approved by the University Committee on the Useand Care of Animals (UCUCA) of the University ofMichigan.

ImmunohistochemistryTumors from mice treated with SB431542 (10 mg/kg)

or vehicle, for 4 hours, were excised and fixed in formalin.The tumors were processed at the UMCCC Tissue CoreFacility and stained with anti-pSmad2 antibody (1:100);micrographs were taken using an Olympus microscopefitted with an Olympus DP-70 high resolution digitalcamera. Cell nuclei stained positive for pSmad2 werecounted in 3 random fields for vehicle-treated (n ¼ 3)and SB431542-treated (n ¼ 3) tumors. A 2-sided student ttest was done to assess statistical significance. Slides wereadjusted for brightness and contrast with Adobe Photo-shop CS2 (Adobe Inc.), but the micrographs underwentno other manipulations.

Data analysesWestern blot signal intensity was measured using the

image processing and analysis program, ImageJ v1.45

(20). Pearson correlation coefficient (r) was estimated toconfirm how well the SB431542 doses correlate to BTRreporter fold induction and pSmad2 levels. Analysis ofstatistical significance (student t test, P values) was doneto estimate the significance of reporter fold inductionwith SB431542 doses. In addition, regression analysis(goodness of fit; R2) was carried out to confirm relation-ships between SB431542 dose, reporter fold induction,and pSmad2 levels. All the statistical analyses were doneon Microsoft Excel 2010. GraphPad Prism v5 (GraphPadSoftware), nonlinear regression analyses, and sigmoidaldose response (variable slope) was used to generateEC50 values.

Results

Construction and mechanism of the bioluminescentTGFbR1 reporter

The key step in TGFb signaling is the phosphorylationof receptor-regulated Smads by active TGFbR1 receptor.To study this essential step, we designed a bioluminescentTGFbR1 reporter using a split firefly luciferase construct(Fig. 1A), consisting of the critical TGFb substrate, the C-terminal SXS motif of Smad2, linked to the phospho-peptide-binding domain FHA2, and flanked by theN-terminal (N-Luc) and C-terminal (C-Luc) domains ofthe firefly luciferase reporter molecule. The functionalbasis of the reporter is shown schematically in Fig. 1B.The reporter exhibits low level luminescence when theSmad2 substrate is phosphorylated allowing it to bind tothe FHA2 domain, thereby sterically preventing reconsti-tution of the luciferase reporter molecule. A reduction inSmad2 phosphorylation through inhibition of TGFbR1kinase activity, or other downstream modulators ofSmad2 phosphorylation, releases the stearic inhibitionallowing reconstitution of the luciferase reporter mole-cule, whose activity can be detected by bioluminescenceimaging.

In addition to the wild-type reporter (BTR-WT), we con-structed aBTRmutant (BTR-MUT) reporter basedon reportsthat mutations of the SXS motif abolish Smad2 phosphor-ylation by TGFbR1 (13). The Smad2 C-terminal serineresidues 465 and 467 were mutated to the neutral aminoacid alanine (Fig. 1A).

Increased BTR reporter activity following treatmentwith TGFbR1 inhibitor SB431542

To evaluate whether the BTR reporter provides a sen-sitive and robust response to inhibition of TGFb-medi-ated Smad2 phosphorylation, we imaged A549 and 1833BTR-WT cells following treatment with varied concentra-tions of SB431542 (21–24) in the presence of 10 ng/mLTGFb. Bioluminescence activity was measured sequential-ly for 6 hours revealing both concentration- and time-dependent increases in reporter activity, reaching peakluminescence within 60 minutes posttreatment (Fig. 1C;Supplementary Fig. S2A). The BTR reporter activityincreased with greater concentrations of inhibitor,

Nyati et al.





reaching a maximum expression of more than 10-foldwith 25 mmol/L SB431542 in A549 (Fig. 1C) and 8-fold in1833 (Supplementary Fig. S2A) maintaining a steady,although lower, bioluminescence for approximately 4hours followed by a gradual decline in signal. EC50 valuesfrom the same datasets (reporter fold induction at 1 hour)were estimated to be 2.09 and 17.67 mmol/L for A549 and1833 cell lines, respectively. The increased BTR reporteractivity correlates with a concomitant decrease inTGFbR1-dependent levels of phosphorylated Smad2 andwith increasing concentrations of SB431542 (Fig. 1D;Supplementary Fig. S2B). EC50 values (pSmad2 levels at1 hour) were calculated to be 1.6 and 0.36 mmol/L forA549 and 1833 cell lines, respectively. A very strongpositive correlation (Pearson correlation coefficient r ¼0.88 in Fig. 1C and r ¼ 0.998 in Supplementary Fig. S2A’)was found between reporter fold induction values andSB431542 concentration. In addition, we also estimatedR2 values (goodness of fit), which determines how closelyour data fit linear regression. R2 values were estimated tobe 0.773 (Fig. 1C’) and 0.996 (Supplementary Fig. S2A’).An analysis of statistical significance between reporterfold induction and SB431542 concentration was done.P values were found to be statistically significant. Simi-larly, a strong negative correlation (r ¼ �0.985 in Fig. 1Eand r ¼ �0.5308 in Supplementary Fig. S2B, bottompanel) between reporter fold induction and pSmad2levels was found, which further confirms the robustnessof the reporter. A549 BTR-MUT cells treated withSB431542 (10 mmol/L) in the presence of TGFb showonly a slight increase in reporter activity, which was notsustained throughout the duration of the experiment(Fig. 1F), confirming the specificity of the BTR-WT report-er activity to TGFBR1 kinase phosphorylation of theSmad2 substrate. These changes in BTR reporter activitywere not the result of changes in levels of total Smad2 orreporter (Fig. 1D; Supplementary Fig. S2B and C).

BTRreporter sensitivity to indirect inhibitionofSmad2phosphorylationStudies have shown that active src-kinase is a down-

streammediator of TGFb-induced cellular responses (25).To study the sensitivity of the BTR reporter assay toindirect effectors of Smad2 phosphorylation, we treatedA549 BTR-WT cells with the src-kinase inhibitor, PP2(26), in the presence of 10 ng/mL TGFb. Inhibition ofsrc-kinase activity results in both a concentration- andtime-dependent increase in reporter response, with peakbioluminescence occurring approximately 2 hours post-treatment (Supplementary Fig. S1A). Western blot anal-ysis confirms that increase in reporter activity correspondsto decrease in pSmad2 levels, 2 hours post-PP2 treatment(Supplementary Fig. S1B), as well as a concentration-dependent decrease in pSmad2 with increasing PP2 con-centration (Supplementary Fig. S1B). Identical experi-ments using the BTR-MUT expressing A549 cell line showminimal effect of PP2 on reporter activity (Supplemen-tary Fig. S1C). Similar to SB431542 treatment, these

results were not due to any significant change in levelsof total Smad2 (Supplementary Fig. S1B).

In vitro A549 BTR-WT cellular responses to TGFbstimulus

Tomonitor the effect of the biological transducer on BTRreporter activity, we treated A549 BTR-WT cells with 10 ng/mL TGFb, revealing a nearly 60% decrease in reporteractivity at 1 hour posttreatment, whereas treatment ofA549 cells expressing the mutant reporter show minimaleffect on reporter response (Fig. 2A). Western blot analysisof similarly treated A549 BTR-WT cells indicate a corre-sponding increase in pSmad2 at 15 minutes that remainedat maximal levels for 2 hours, followed by slightly lowerlevels at 4 and 6 hours (Fig. 2B).

Verifying that TGFb induction phosphorylates the targetpeptide, the reporter molecule was immunoprecipitatedfrom cell extracts with anti-luciferase antibody and immu-noblotted with a phospho-serine–specific antibody. TheBTR protein had a substantial increase in phosphoserinelevels following TGFb treatment as compared with controls(Fig. 2C; Supplementary Fig. S2D). Following treatmentwith SB431542, phosphoserine levels are significantly low-er than control levels (Fig. 2C; Supplementary Fig. S2D)correspondingwith increasedBTR reporter activity, as deter-mined directly by bioluminescence imaging.

To further confirm the link between reporter activity andTGFb signaling, we suppressed the expression of theupstream kinases by transiently transfecting A549 BTR-WTcells with TGFbR1- and TGFbR2-specific siRNA resulting inapproximately 6-fold and 3-fold increases in reporter activ-ity, respectively (Fig. 2D). Cells transfected with nontargetsiRNA (NSS) did not affect reporter activity (Fig. 2D).Similar observationsweremade in1833BTRWTcellswhichexhibited 2.5-fold increase in reporter activity withTGFBR1siRNA (Supplementary Fig. S2E); although, nosignificant activation of the reporter was observed withsiRNA to TGFBR2. Cell extracts prepared from parallelexperiments and probed with antibodies against TGFbR1and TGFBR2 confirm the siRNA-mediated suppression ofthese proteins. A substantial increase in pSmad2 wasobserved in NSS-transfected TGFb-treated cells, whereastheTGFBR1 and TGFBR2 knockdown cells, with or withoutTGFb stimulation, have minimal levels of pSmad2 (Fig. 2E;Supplementary Fig. S2F) which closely correlates withreporter bioluminescence activity (Fig. 2D; SupplementaryFig. S2E).

To show that A549 BTRWT cells undergo EMT transitionand exhibit the characteristicmesenchymal phenotypewithTGFb treatment (3, 27, 28), we treated serum-starved cellswith 10 ng/mL TGFb for up to 96 hours and studied thechanges in cell morphology every 24 hours posttreatment.TGFb-treated cells lost cell-to-cell contact and exhibitedcharacteristic elongated, spindle-shaped morphology(Fig. 2F). These samples were prepared for Western blotanalysis and probed against epithelial marker E-cadherinandmesenchymal marker N-cadherin. An increase in levelsof N-cadherin and corresponding decrease of E-cadherin






IP: Luc

IB: Luc

IP: Luc

IB: pSer

pSMAD2

Luciferase

GAPDH

Mock

0.2

5 h

0.5

h

1 h

2 h

4 h

6 h

pSMAD2

SMAD2

Luciferase

GAPDH

c-Jun

A B C

D E G

F H

Input

Mock

TG

Fβ

TG

Fβ

+

SB

431542

1 1.8 0.4

TGFβ 10 ng/mL

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 60 120 180 240 300 360

BTR-WTBTR-MUT

siRNA

0

1

2

3

4

5

6

7

8

NSS

NSS

+TGFβ

TGFB

R1

TGFB

R1

+TGFβ

TGFB

R2

TGFB

R2

+TGFβ

pSMAD2

SMAD2

TGFBR1

TGFBR2

siRNA

TGFBR1

TGFBR2

NSS

TGFβ

+ -++ +

+ +++ +

- ---- ---- --

-- -

Luciferase

Time (min)

Fo

ld in

du

ctio

nF

old

ind

uct

ion

0.5 h 24 h 48 h 72 h

TGFβ

Control

TGFβ 10 ng/mL

N-cadherin

E-cadherin

Luciferase

GAPDHM

ock

0.5

h

24 h

48 h

72 h

96 h

c-myc

c-jun

pSMAD2

c-Jun

c-Myc

SB431542

TGFβ

+++++ +

-- ---- -- --

0 1 3 6 0 1 3 6

Time (h)

SMAD2

Figure 2. BTR reporter activity is dependent on reporter substrate phosphorylation through TGFb signaling. A, A549 cells stably expressing BTR-WT reportersand BTR-MUT treated with TGFb (10 ng/mL) and bioluminescence measured serially for 6 hours. B, cell lysates prepared and analyzed by Westernblot against anti–phospho-(S465–S467) Smad2, total Smad2, luciferase, C-jun, and GAPDH antibodies. C, BTR-WT reporter expressing A549 cells weretreated with TGFb (10 ng/mL) or TGFb and SB431542 (10 mmol/L) for 1 hour; cell lysates were prepared and immunoprecipitated with luciferase-specificantibody. The level of substrate phosphorylation was determined by Western blot analysis with phospho-Ser antibody. The immunoblot wasprobed with luciferase-specific antibody as a control and the input lysate was probed against phospho-(S465–S467) Smad2, total Smad2, luciferase, andGAPDH. D, A549 cells stably expressing the BTR-WT reporter were transfected with 100 nmol/L siRNA for 72 hours, TGFb was added for an additional1 hour, and reporter activity was measured. Fold change in reporter activity was calculated from change in nontargeted siRNA (NSS). Error bars denoteSEM. E, cell lysates were prepared and analyzed by Western blot against anti–phospho-(S465–S467) Smad2, total Smad2, TGFBR1, TGFBR2, andluciferase. F, A549 BTR-WT cells were serum starved and treated with TGFb (10 ng/mL) for 72 hours showing epithelial-to-mesenchymal transition. G, celllysatesofA549-BTRWTcells treatedwith TGFb for 0.5, 24, 48, 72, and96hourswereprobedwith antibodiesagainstN-cadherin, E-cadherin, luciferase, c-jun,c-myc, and GAPDH. H, Western blot analysis of A549 BTR-WT cells treated with 10 mmol/L SB431542 or 10 ng/mL TGFb for 0, 1, 3, and 6 hourswith antibodies against phospho-(S465–S467) Smad2, c-myc, and c-jun.

Nyati et al.





was observed (Fig. 2G), which is a characteristic of cellsundergoing EMT.Two early regulatory genes shown to be downstream

targets of R-Smad activation are the oncogenes c-jun andc-myc (29, 30). Expression levels of c-myc decreased over 96hours and c-jun levels increased during the same timeperiod (Fig. 2G). We further elucidated the connection ofc-myc and c-jun expression to TGFb signaling in A549 cellsby treating with TGFb alone and in the presence ofSB431542. With TGFb, c-myc levels remain constant forup to 6 hours, but when TGFb signaling was inhibited, c-myc levels increased (Fig. 2H). Conversely, c-jun expressionlevels increase within 1 hour after TGFb treatment and areabolished in the presence of SB431542 (Fig. 2H).

In vivo imaging and analysis of TGFb signaling in BTRexpressing tumor xenografts

Having established the specificity and sensitivity of theBTR reporter in vitro, we next investigated reporter utility formonitoring changes in TGFb signaling in a tumor mousemodel. We implanted A549 BTR-WT cells into the flanks ofnude mice to establish tumors. When the tumors reached avolume of 40 to 60 mm3, we monitored bioluminescenceover time in mice treated with TGFBR1 kinase inhibitor orvehicle control. Representative images of control- andSB431542-treated mice are shown in Fig. 3A. Biolumines-cence activity in mice treated with SB431542 increased inthe first hour and reachedpeak activity of greater than5-foldwithin 4 hours posttreatment. Reporter activity remained

Figure 3. Molecular imaging ofTGFb signaling in vivo. A, tumorxenografts expressing BTR-WTwere established on each flank of 4-to 6-week-old female athymicCD-1nu/nu mice. Images of in vivo BTRreporter activity are shownfollowing intraperitoneal injection ofSB431542 (10 mg/kg body weight)or vehicle control (DMSO) at 1, 4, 8,and 24 hours. B, bioluminescenceactivity in tumor-bearing micebefore treatment and in response totreatment was measured andexpressed as fold induction. Valuesrepresent the mean � SEM. C,tumor xenografts were sectioned4 hours posttreatment withSB431542 (10 mg/kg body weight)or vehicle control (DMSO) andimmunohistochemical analysis wasdone using pSmad2(S465–S467)-specific antibody. The scale bar is 2mm and a 10-fold magnification ofselected section. D, quantificationof number of cells staining positivefor nuclear pSmad2 for control-(n ¼ 3) and SB431542-treated(n ¼ 3) tumors in 3 random fields. Adecrease in pSmad2-stained nucleiin SB431542-treated tumors (761nuclei, 37.03%) was observedwhen compared with DMSO-treated tumors (1294 nuclei,62.97%). Statistical analysis:�, P ¼ 0.06 calculated by a 2-sidedStudent t test.

0 4 8 24Time (h)

Control

SB431542

1

2.0

0.5

1.0

1.5

2.5

2.0

0.5

1.0

1.5

2.5

A

10

6 p/s

ec/c

m2/s

r10

6 p/s

ec/c

m2/s

r

Control/DMSO (n = 10)

SB431542 10 mg/kg (n = 10)

Control/DMSO SB431542

0

200

400

600

800

1,000

1,200

1,400

1,600

Control SB431542

pS

ma

d2

-po

sitiv

e n

ucl

ei

62.97%

37.03%

P = 0.06B

C

D*

0

1

2

3

4

5

6

7

0 4 8 12 16 20 24

Fo

ld in

ductio

n

Time (h)






elevated through 8 hours followed by lower, but sustained,reporter activity at 24 hour (Fig. 3A and B). Biolumines-cence was unchanged over the 24-hour period in vehicle-treated mice (Fig. 3A and B).

In a similar experiment, mice (n ¼ 5) bearing xenograftsof A549 BTR-WT and MUT reporters were treated with 10mg/kg SB431542 andDMSO control. Bioluminescence wasacquired 4 hours after treatment and plotted as fold changeover control (Supplementary Fig. S3). BTR-WT expressingtumors showed around 8-fold induction in biolumines-cence activity with SB431542, whereas MUT tumors exhib-ited an attenuated response (around 3.5-fold).

Because BTR reporter bioluminescence representsdecreased levels of pSmad2, we next evaluated pSmad2levels in tumor xenografts harvested 4 hours posttreatmentwhen peak bioluminescence was observed. Immunohisto-logic staining was done using pSmad2-specific antibody(Fig. 3C and D). Total numbers of pSmad2-positive nucleiwere counted for each group and plotted, revealing a sig-nificant decrease (P ¼ 0.06) in levels of pSmad2-stainednuclei in SB431542-treated tumors (761 nuclei, 37.03%) ascompared with control tumors (1294 nuclei, 62.97%; Fig.3D).

High-throughput screening of BTR reporter assayagainst a kinase-inhibitor compound library revealscross-talk with multiple kinase signaling pathways

Having confirmed the sensitivity of the BTR reporter toboth direct and indirect inhibition of TGFb-stimulatedphosphorylation of the C-terminal serines of Smad2, weexamined its suitability for high-throughput screening.Cellular activity of the BTR reporter was screened againsta kinase inhibitor library in the presence of TGFb. Of the 84small molecular weight kinase inhibitors screened, 13compounds increased reporter activity. The active com-pounds included 5protein tyrosine kinase (PTK) inhibitors,(tyrphostin analogs, AG1478, AG 1288, AG 825, andRG14620, and an Erlotinib analog, BML-265); a selectiveJNK inhibitor (SP600125); 2 p38 MAP kinase inhibitors(SB203580, SB202190); a phosphoinositide 3-kinase(PI3K) inhibitors (LY294002); 2 inhibitors of Akt signaling(BML257, triciribine); and finally, staurosporine, an apo-ptosis inducer and potent pan-specific protein kinase inhib-itor, and quercetin, a common dietary flavanoid believed tohave antioxidant and anticancer properties. No significantchange in reporter activity was observed when the screenwas repeated using A549 BTR-MUT cells, confirming assayspecificity for phosphorylation of the C-terminal SXS motif(Fig. 4A). In addition, theHTS screenwas repeated using the1833 BTR-WT reporter cell line, to independently validatethe obtained hits (Fig. 4C). Most of the hits between 2 celllines were similar, although the reporter fold inductionvaried (Fig. 4A andC; Table 1). The EGFR-directed inhibitorAG1478was ahit inA549and1833 cells (8.2-fold and12.7-fold). Not surprisingly, the Akt inhibitor (BML257) and thePI3K inhibitor (LY294002) were also hits in the A549 cells(5.0-fold and 5.4-fold, respectively). In 1833 cells,LY294002 yielded a 12.3-fold increase in bioluminescence,

whereas BML257 yielded 3.1-fold increase. A JNK inhibitor(SP600125) and p38MAPK inhibitors (SB203580 andSB202190), important regulators of TGFb signaling, werealso hits in both cell lines (see Table 1). Interestingly,GW5074 was a hit in 1833 cells but not in the A549 cells.

To evaluate the effect of these inhibitors on TGFb-induced Smad2 phosphorylation, select active compounds,alongwith 2 clinically important receptor kinase inhibitors,erlotinib and lapatinib, underwent further analysis. A549BTR-WT cells were treated with inhibitors in the presence ofTGFb, lysateswere prepared, andWestern blots probedwithpSmad2 antibodies. The results (summarized in Table 1)confirm the previously described inhibition of TGFb-acti-vated Smad2 phosphorylation by the direct TGFbR1 inhib-itor SB431542 and indirect inhibition with src-kinaseinhibitor, PP2. Of the selected kinase inhibitors, lapatinib,AG1478, and LY294002 completely block TGFb-inducedSmad2 phosphorylation, whereas erlotinib, AG825,SP600125, SB203580, SB202190, and staurosporine showonly moderate reduction in measurable pSmad2 levels.BML257 and quercetin treatments resulted in no measur-able reduction in pSmad2 levels despite increased reporteractivity (Fig. 4B). In addition Western blots were probedwith pSmad3 antibody to access the potential role of non-TGFb receptor kinase inhibitors in inhibiting TGFb-medi-ated Smad3 phosphorylation. Lapatinib and AG1478mod-erately block TGFb-induced Smad3 phosphorylation,whereas SB203580, SB202190, and staurosporine showonly marginal reductions in measurable pSmad3 levels.

To delineate other signaling events affecting TGFb recep-tor activity, we probed the Western blot with antibodies totarget proteins (see Table 1 for references). The PTK inhi-bitors erlotinib, lapatinib, and tyrphostin AG1478 reducelevels of pHER2 andpEGFR,while increasing the expressionof the inhibitory-Smad, Smad7. The tyrphostin analogAG825 had no measurable effect on activation of HER2 orEGFR kinase. AG825 is an ATP-competitive inhibitor ofHER2 and EGFR kinase activity in cell-free assays but, due tohigh intracellular ATP levels, it has minimal activity inwhole cells (31). Treatment with JNK inhibitor SP600125resulted in a moderate decrease in TGFb-mediated Smad2phosphorylation but levels of pJNK, asmeasuredwith pJNK(T180/Y182) antibody, remain unchanged with TGFb orSP600125 treatment. TGFb treatment increases phosphor-ylated p38 MAP kinase (p-p38MAPK) levels which isunchanged following treatment with the p38MAPK inhibi-tors SB203580 and SB202190; p-p38MAPK levels are atten-uated by tyrphostin AG1478, SB431542, and PP2 treat-ment. Levels of phosphorylated PI3Kwere partly reduced byLY294002, a PI3K inhibitor, as well as, with lapatinib,tyrphostin AG1478, and SP600125. A slight decrease inluciferase expression was observed with PP2, erlotinib,lapatinib, tyrphostin AG1478, staurosporine, andLY294002 treatments with no corresponding decrease inGAPDH levels, suggesting that results are not due to differ-ences in protein loading (Fig. 4B).

Staurosporine treatment caused only a small increase inBTR activity and a moderate decrease in Smad2

Nyati et al.





phosphorylation, while completely blocking activation ofPI3K and causing a marked increase in p-p38MAPK, alongwith reduced expression of most of the other proteinsmeasured, including luciferase. It has been reported thatstaurosporine treatment of Mv1Lu cells, 100 ng/mL,increased pSmad2 and enhanced TGFb-stimulated apopto-sis (32) suggesting that a 5-hour exposure to a 10-foldhigher concentration of staurosporine, as seen in our exper-iment, most likely initiated cell death, thus decreasingprotein expression (Fig. 4B).

Quercetin, a common dietary flavanoid believed to haveantioxidant and anticancer properties, was observed todecrease levels of HER2 protein and inhibit downstreamPI3K signaling in breast cancer cells (33) and suppress basalexpression of TGFBR1/2, Smad2/3/4, and pSmad2/3 after72-hour treatment of keloid fibroblasts (34). In this study,quercetin treatment stimulated BTR reporter biolumines-cence and showed a moderate inhibition of TGFBR1 andpEGFR levels but did not show any measurable inhibitionof pSmad2 levels, PI3K signaling, or increased expression of

TGFBR1

pSMAD2

SMAD2

pHer2 (Y1248)

Her2

pEGFR (Y845)

EGFR

pPI3K

(P85-Y458/P55-Y199)

pP38MAPK

(T180/Y182)

pJNK

(T183/Y185)

Luciferase

GAPDH

Mock

Tyr

AG

14

78

SB

20

21

90

BM

L257

SP

600125

Sta

uro

sp

ori

ne

Quer

dih

ydra

te

LY2

94

00

2

TG

Fβ

on

ly

SB

43

15

42

PP

2

Erl

otin

ib

La

pa

tin

ib

AG

82

5

SB

20

35

80

TGFβ 10 ng/mL

Fold

induction

0

1

2

3

4

5

6

7

8

9

0 10 20 30 40 50 60 70 80 90

BTR-WT

BTR-mutSB203580

Tyrphostin AG1478

LY294002

SB202190BML265

SP600125 BML257Quercetin

dihydrate

AG825 Tyrphostin AG1288

STS Triciribine

Compound number

SMAD6

SMAD7

A

C

B

0.0 1.0 0.0 0.1 0.6 0.0 0.1 1.1 0.8 0.7 0.5 1.1 0.6 0.1 1.0

RG14620SMAD3

pSMAD3

c-Myc

c-Jun

PI3K

P38MAPK

0

2

4

6

8

10

12

14

16

18

0 10 20 30 40 50 60 70 80 90

BTR-WT

Compound number

Fold

induction

Tyrp

hostin A

G12

88

STS

Tyrp

hostin A

G14

78

LY29

4002

SP6001

25

BML2

65

Triciri

bineSB20

2190

SB2035

80GW

5074

Hyp

ericine

Palm

itoyl-D

L-ca

rnitine

PP2Ty

rpho

stin 5

1

0.0 1.0 0.0 0.0 1.2 0.4 0.5 1.1 1.2 0.9 0.8 1.1 0.7 1.0 1.4

Figure 4. High-throughput screening of BTR reporter assay against a small molecular weight library of kinase inhibitors reveals TGFb signaling cross-talk withmultiple kinase signaling pathways. A, A549 cells stably expressing the BTR-WT and BTR-MUT reporters were incubated with compounds from an 84-compound small molecular weight kinase inhibitor library (10 mmol/L) for 4 hours, TGFb (10 ng/mL) was added for 1 hour, and reporter activity measured andplotted as fold change over DMSOcontrols. Compoundswhich exhibit 3 folds or higher induction in the reporter activity aremarked in the plot. All assayswererepeated at least 3 times. Values from one representative experiment are shown. B, cell lysates from similarly treated A549 BTR-WT cells were probed withantibodies against phospho-(S465–S467) Smad2, total Smad2, phospho-(S423–425) Smad3, total Smad3, c-Myc, c-Jun, Smad6, Smad7, TGFBR1,p(Y1248)-HER2, HER2, p(Y845)-EGFR, EGFR, p-(P85-Y458/P55-Y199)-PI3K, total PI3K, p(T180-Y182)-P38MAPK, total P38MAPK, p(T183/Y185)-JNK,luciferase, andGAPDH.Densitometric analysis of pSmad2 level for all treatmentswasmeasuredandcalculatedas fold changeover TGFb-treated samples.C,kinase inhibitor library screen as described in Fig. 4A was carried out with 1833 cells stably expressing the BTR-WT reporter.






inhibitory Smad7. These results suggest that the reporterresponse may be a result of an immeasurable change inpSmad2 level due to moderate changes in TGFBR1 andpEGFR expression (Fig. 4B). Lysates were also probed withc-Jun and c-Myc antibodies to evaluate the effect of eachinhibitor on TGFb/Smad-mediated transcription.

Two RTK inhibitors which abrogated TGFb-inducedactivation of Smad2 (lapatinib and AG1478) were select-ed for further analysis. Lapatinib activated the BTR-WTreporter at a dose of 1 mmol/L (�4-fold) but failed toactivate the mutant reporter significantly at doses as highas 10 mmol/L (Fig. 5A). Tyrphostin AG1478 also activat-ed the WT reporter similarly and did not alter the activityof the MUT reporter substantially (Fig. 5B). These datashow that lapatanib and AG1478 were valid hits in thescreen.

To further evaluate the validity of the 2 RTK inhibitorsas modulators of TGFb signaling, their impact onSmad2/3–Smad4 complex formation was evaluated. For-mation of a Smad2/3–Smad4 complex is a critical eventfor nuclear translocation of Smad2/3. Lysates from A549cells treated with SB431542, PP2, AG1478, and lapatinib

in the presence of TGFb were immunoprecipitated usinga Smad3 antibody and coimmunoprecipitation of Smad4was evaluated (Fig. 5C). SB431542 and PP2 completelyblocked Smad3–Smad4 complex formation comparedwith TGFb-treated cells, whereas AG1478 substantiallyand lapatinib marginally blocked complex formationbetween Smad3–Smad4 in A549 cells. The reduced Smadcomplex formation was due to blockade of TGFb-induced phosphorylation and activation of Smad2/3 asseen in the input lysates (Fig. 5C). In addition to eluci-date whether inhibition of Smad2/3–4 complex forma-tion by PP2 and RTK inhibitors have any effect ondownstream transcriptional responses mediated byTGFb, we carried out transcriptional response assay usinga TGFb-responsive reporter SBE4-Luc (ref. 17; Fig. 5D).TGFb strongly activated the SBE4-Luc reporter in A549cells which was significantly blocked by the ALK5 inhib-itor SB431542 and moderately blocked by PP2, AG1478,and lapatinib. These results suggest that PP2, AG1478,and lapatinib partially inhibit the activation of Smadpathway resulting in inhibition of TGFb-induced tran-scriptional activity.

Table 1. Summary of kinase inhibitors and their effects on bioluminescent TGFBR1 reporter activity andphosphorylated Smad2 levels

Kinase-inhibitors(10 mmol/L)

Reporterbioluminescence(Fold response)

1833(Fold response)

pSmad2levels

Kinase-inhibitortarget(s)a

References: inhibitorsand C—terminalpSmad2/3

BTR-WT BTR-MUT BTR-WT

Vehicle (DMSO) 1.0 1.0 1 0.0TGFb only 0.4 1.0 — 1.0SB431542 10.0 2.0 3.9 0.0 TGFBR1 kinase (21, 22, 41, 50)PP2 4.0 1.0 4.1 0.1 Src family nonreceptor

tyrosine kinases(25)

AG1478 8.2 0.9 12.7 0.1 PTK inhibitor selectivefor EGFR

(37)

AG825 3.6 0.9 0.89 1.1 PTK inhibitor selectivefor EGFR

(37)

SP600125 4.8 1.2 16.8 0.8 c-Jun N-terminalkinase (JNK)

(41, 42, 45)

SB203580 8.0 1.1 4.0 0.7 p38 MAP kinase activity (41, 42, 44)SB202190 6.2 0.8 6.4 0.5 p38 MAP Kinase (43)BML257 5.0 1.1 3.1 1.1 Akt1/2/3 serine–threonine

protein kinasesStaurosporine 3.0 0.5 4.2 0.6 Pan-specific (32)LY294002 5.4 0.8 12.3 0.1 Phosphoinositide 3-kinase

(PI3K)(41, 47)

Quercetin 4.6 1.1 2.0 1.0 EGFR and Her-2 tyrosinekinase, PI3K,

(34)

Erlotinib n/a n/a n/a 0.6 EGFR tyrosine kinase (37)Lapatinib n/a n/a n/a 0.0 EGFR and Her-2

tyrosine kinase(37)

aTarget information from product inserts.

Nyati et al.





Discussion

The dual modality of TGFb, both as a potent tumorsuppressor and a stimulator of tumor progression, invasion,

and metastasis, make it a critical target for therapeuticintervention in human cancers. The ability to conductreal-time, noninvasive imaging of TGFb-activated Smadsignaling in live cells andanimalmodelswould significantly

A B

C D

0

1

2

3

4

5

6

7

8

9

0 0.1 1 2 5 10

WT

MUT

Fo

ld in

duct

ion

Tyrphostin AG1478 (μmol/L)

0

1

2

3

4

5

6WT

MUT

Fo

ld in

duct

ion

0 0.1 1 2 5 10

Lapatinib (μmol/L)

0

Moc

kTG

F

SB4315

42PP2

AG14

78

Lapa

tinib

1

2

3

4

5

6

SB

E4-

Lu

c fo

ld in

du

ctio

n

Treatments

IB: Smad4

IP: S

mad3

tSmad2

pSmad2

Smad4

tSmad3

pSMAD3

IInput

Mock

AG

1478

TG

Fβ

only

SB

431542

Lapatinib

TGFβ 10 ng/mL

PP

2

GAPDH

Figure 5. A, A549 cells stably expressing the BTR-WT andMUT reporter were treated with increasing concentrations of RTK inhibitor lapatinib (0.1, 1, 2, 5, and10 mmol/L) and tyrphostin AG1478. B, bioluminescence was measured after 1 hour. The change in bioluminescence activity over mock treatment(DMSO) levels was calculated and plotted as fold induction. Bioluminescent measurements were done in 8 wells. C, A549 cells were treated with TGFb(10 ng/mL) or TGFb and SB431542, PP2, AG1478, and lapatinib (10 mmol/L) for 1 hour, cell lysates were prepared and immunoprecipitated withSmad3-specific antibody. The level of Smad2/3–Smad4 complex formation was determined byWestern blot analysis with Smad4 antibody. The input lysatewas probed against pSmad2, total Smad2, pSmad3, total Smad3, Smad4, and GAPDH. D, A549 cells were transiently transfected with SBE4-Lucreporter plasmid and GLuc plasmid as internal control. Cells were treated with 10 mmol/L SB431542, PP2, AG1478, and lapatinib in presence of 10 ng/mLTGFb for 22 hours. Firefly luciferase activity was normalized to Gaussia luciferase, and fold induction (over mock/DMSO treated) is expressed as themean � SEM of triplicate measurements.






improve our ability to develop targeted therapeutics forcancer and to monitor their effects in vivo. To study TGFb-mediated phosphorylation of the C-terminal serines of R-Smads, we constructed a TGFBR1 bioluminescent molecu-lar imaging reporter, described in Fig. 1A and B, based onsplit-luciferase technology (14).We validated the specificityand responsiveness of the BTR reporter through directinhibition of TGFBR1 kinase activity and, indirectly,through inhibition of src-kinase, a downstream modulatorof TGFb-stimulated Smad2 phosphorylation (25). Reporterbioluminescence activity shows a robust response followinginhibition of TGFBR1 kinase activity with SB431542. Time-and concentration-dependence of SB431542 inhibition ofTGFBR1 kinase was confirmed by increased reporter activityand a concomitant decrease in levels of pSmad2 (Fig. 1Cand E; Supplementary Fig. S2).

The EC50 derived using the reporter (for SB431542) wasestimated to be 2.09 mmol/L, whereas the pSmad2Westernsyielded an EC50 of 1.6 mmol/L in A549 cells. In 1833 cells,the values were 15.68 mmol/L using the reporter and 0.36mmol/L using pSmad2Westerns. Strong positive correlation(Fig. 1C’; Supplementary Fig. S2A’), goodness of fit, andstatistical significance between reporter fold induction andSB431542 concentration clearly show the robustness of theBTR reporter. A strongnegative correlationbetween reporterfold induction and pSmad2 levels (Fig. 1E; SupplementaryFig. S2B, bottom panel) further substantiated the effective-ness of the reporter. Inhibition of downstreammodulationof pSmad2 levels by src-kinase also showed time- andconcentration-dependent increases in reporter biolumines-cence, alongwith a decrease inmeasurable levels of pSmad2(Supplementary Fig. S1). These results strongly support therelationship between increased reporter activity and bothdirect and indirect inhibition of Smad2 phosphorylation.

Direct confirmation that TGFb is a biological transducerof BTR reporter activity is shownby the reductionof reporterbioluminescence following TGFb treatment of A549 BTR-WT cells (Fig. 2A) and the corresponding increase inpSmad2 (Fig. 2B). The sustained decrease in reporter activ-ity and prolonged elevation of pSmad2 are consistent withcontinuous shuttling of Smadproteins between cell nucleusand cytoplasm shown to exist during TGFb signaling (23).Immunoprecipitation of the reporter, followed by analysisof phosphoserine content, confirms that the reporter wasphosphorylated in a TGFb-dependent manner (Fig. 2C;Supplementary Fig. S2D). In addition, suppression of theupstream kinases by siRNA knockdown of TGFbR1 andTGFbR2 verifies the direct link between reporter activity andTGFb receptor signaling. Knockdown TGFBR1 and TGFBR2cells have reduced levels of pSmad2 and show a significantincrease in reporter bioluminescence (Fig. 2D and E; Sup-plementary Fig. S2E and F).

TGFb signaling contributes to tumor progression andmetastasis through activation of EMT, a process in whichepithelial cells convert to fibroblastoid-like cells whichgenerate an invasive and metastatic phenotype associatedwith poor clinical outcomes for cancer patients (35). TGFb-activation of Smad2 signaling induces EMT in A549 BTR-

WT cells, resulting in the loss of cell-to-cell contact and thecellular conversion to an elongated, spindle-shaped mor-phology (Fig. 2F). Protein analysis reveals the characteristicEMT downregulation of E-cadherin expression, a criticalcaretaker of epithelial integrity, and a corresponding upre-gulation of N-cadherin (Fig. 2G; ref. 35). Activated Smad2has also been shown to regulate expression of the proto-oncogenes c-myc (30) and c-jun (29). TGFb downregula-tion of c-myc is critical to growth inhibition (30) in whichthe upregulation of c-jun is important for cell survival (29).TGFb stimulation of A549 BTR-WT cells suppresses c-mycexpression with a corresponding upregulation of c-jun(Fig. 2G). Cell treatment with TGFBR1 kinase inhibitorSB431542 results in increased expression of c-myc andeliminates c-jun expression (Fig. 2H). A TGFb increase intotal c-jun levels is consistent with previous findings show-ing an effective downstream signaling cascade initiated byactivated R-Smads in response to TGFb (29).

Several imaging techniques have been designed to mon-itor TGFb signaling in various in vivomodels (5–12, 36), butnone directlymeasure the phosphorylation of R-Smads, thekey event in the TGFb signaling pathway. Accordingly, high-throughput screening of a kinase inhibitor library identifiedthe existence of a complex cross-talk between TGFb signal-ing and tyrosine kinase receptors, as well as p38MAPK, JNK,and PI3K pathways through attenuation of R-Smad phos-phorylation at the C-terminal serines (Summarized in Sup-plementary Fig. S4).

Research has shown growth factors which signal throughRTKs such as, hepatocyte growth factor and epidermalgrowth factor (EGF), mediate Smad2 activity independentof TGFbR1 activation, suggesting intricate cross-talkbetween RTK and RSK signaling pathways (37, 38). Ourresults show that EGF receptor (EGFR) and human epider-mal growth factor receptor 2 (HER2) kinase inhibitorserlotinib, lapatinib, and tyrphostin AG1478 block TGFb-mediated increases in pSmad2 (Fig. 4B). Lapatinib andtyrphostin AG1478 marginally reduce TGFb-mediatedphosphorylation of Smad3, whereas erlotinib had no sig-nificant effect (Fig. 4B).The reduction in pSmad2 levelscoincides with an equally pronounced decrease in expres-sion of the active kinases, phosphorylated HER2 (pHER2),and phosphorylated EGFR (pEGFR). Treatment of A549BTR-WT cells with SB431542 also reduces basal levels ofpEGFR, further supporting signaling cross-talk betweenRTKand RSK pathways.

An important componentof cross-talk betweenTGFb andRTK signaling is the increased expression of inhibitory-Smad (I-Smad) proteins (39). The I-Smads, Smad6, andSmad7, modulate signaling of TGFb and, another TGFbisoform, bonemorphogenetic protein (BMP) by binding tothe activated type-1 receptor, thereby preventing recruit-ment and phosphorylation of R-Smads (38, 39). In somecell types, TGFb or BMP induce I-Smad expression resultingin a negative feedback mechanism to regulate further TGFbor BMP signaling. Besides induction by TGFb or BMP,I-Smad expression can also be induced in a cell-dependentmanner by other cytokine signaling pathways and stress

Nyati et al.





induction (39, 40). In A549 cells, expression levels ofSmad6 and Smad7 did not increase following TGFb treat-ment. However, treatment with EGFR and HER2 kinaseinhibitors significantly increased expression of Smad7 inA549 BTR-WT cells resulting in interference with recruit-ment and phosphorylation of Smad2 by active TGFbR1(Fig. 4B). These data suggest that, at least in part, the RTKsignaling pathways increase activation of R-Smads by sup-pressing Smad7 expression and subsequent interferencewith Smad2 phosphorylation, providing evidence of theimportance of RTK and RSK cross-talk in regulating TGFb-induced Smad2 signaling.In addition to cross-talk between RTK and RSK signaling,

other kinase pathways have been shown to modulateSmad2 phosphorylation. BTR reporter activity wasincreased by inhibitors of PI3K, p38MAPK, and JNK sig-naling pathways. Martin and colleagues (41) showed thatkinase inhibitors against PI3K (LY294002), p38MAPK(SB203580), and JNK (SP6000125) activity attenuateTGFb-mediated phosphorylation of Smad2 in lung fibro-blasts, in which treatment with inhibitors for other kinasesignaling pathways, MEK1 (PD98059), PKC (Ro31-8425),and Erk/MEK (U0126), had no effect. These observationsare supported by our screening results, BTR reporter activitywas increased for inhibitors shown to reduce R-Smadactivation (LY294002, SB203580, and SP600125; Fig.4A) and was inactive when treated with inhibitors to kinasepathways that do not modulate TGFb-stimulated phos-phorylation of C-terminal serines of Smad2/3 (PD98059,Ro31-8425, and U0126; Data not shown). A parallel screenusing the 1833 cell line stably expressing the BTR-WTreporter yielded hits that were overlapping with the hitsobtained from the A549 cells (Fig. 4C). Published studiessupport some of these hits, for example, inhibitors for JNK(SP600125) and p38MAPK (SB203580) have been shownto downregulate phosphorylation of the C-terminal serinesin rat mesangial cells (42), and the p38MAPK inhibitor(SB202190) abolishes TGFb activation of Smad-dependentpromoters and attenuate pSmad2 levels in glioblastomacells (43).However, the extent of signaling cross-talk between TGFb

andmitogen-activatedproteinkinase (MAPK)pathwayshasbeen shown to be dependent on cell type and cell context.Where the previous studies show JNK and p38MAPK mod-ulation of pSmad, TGFb-induced p38MAPK activity inmouse mammary epithelial (NMuMG) cells was shown tomediate TGFb responses independent of Smad phosphor-ylation (44). The cell-specific differences in cross-talkbetween MAPK and Smad pathways are also evident withthe JNK pathway. Where active JNK enhances or maintainsTGFb-mediated expression in fibroblasts (41) and mesan-gial cells (42), through phosphorylation of C-terminalSmad2, inhibiting JNK activity in embryonic lung explantsincreases both endogenous and TGFb-induced Smad phos-phorylation (45), suggesting that active JNK antagonizesTGFb-induced pSmad2. In yet another study, TGFb-activat-ed JNKwas shown toonlyphosphorylate Smad3outside theC-terminal SXS motif in Mv1Lu cells (46).

Besides cross-talk between p38MAPK and JNKwith TGFbsignaling pathways, our data suggests that the PI3K–Aktsignaling pathway may also provide significant regulationof TGFb-induced cellular responses. Treatment with thePI3K inhibitor LY294002 results in a robust activation ofthe BTR reporter and a nearly complete inhibition of TGFb-mediated phosphorylation of the C-terminal serines ofSmad2 (Fig. 4A and B). Inhibition of Akt1/2/3, a PI3Ktarget protein, with BML257 also activates the BTR reporterbut did not yield a measurable decrease in pSmad. PI3Ksignaling has been shown to be critical for Smad2-depen-dent activity in fibroblasts (41) and transcriptionalresponses, EMT and cell migration of NMuMG cells (47).

Although C-terminal SXS phosphorylation by the TGFbtype-1 receptor is the key event in TGFb signaling, phos-phorylation of the linker region and, to a lesser extent, theMH1 domain of Smad proteins by intracellular proteinkinases can also positively and negatively regulate R-Smadactivity (2). Inhibitors for many of these kinases werescreened with the BTR reporter and, as expected, did notincrease bioluminescence, confirming the specificity of theBTR reporter assay to phosphorylation of the critical SXSmotif.MEK1 inhibitor, PD98059, which inhibits RTK phos-phorylation of Smad2 through activation of ras proteinsdoes not affect TGFb-mediated Smad2 phosphorylation(37) and did not activate the BTR reporter in our kinaseinhibitor screen. Protein kinase inhibitors, known to mod-ulate pSmad in the linker region, were also inactive in thehigh-throughput screen, including the Ca2þ-calmodulin–dependent protein kinase II (CaMKII)-specific inhibitorKN93 (48) and inhibitors of ROCK, PKA/PKG, and MLCK(Y027632, HA-1077, and ML-7, respectively; ref. 49).

In summary, we have designed and validated a sensitiveand highly specific bioluminescence imaging tool to mon-itor TGFb-mediated phosphorylation of the C-terminalserines of R-Smads in live cells and tumor xenografts. TheBTR reporter is adaptable to high-throughput screening inwhich it can be used to aide in identification and evaluationof therapeutic compounds targeting TGFb–Smad2 signal-ing. Following compound identification, this biolumines-cent imaging technique provides a repetitive, noninvasive,dynamic method to monitor in vivo TGFb signaling toevaluate drug–target interaction and pharmacodynamicsof therapeutic compounds that directly or indirectly effectTGFb-stimulated Smad2 phosphorylation providing newinsights into TGFb signaling and help in identifying newdrug therapeutics for cancer treatment.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

Barbara Olson for critical reading of the manuscript, Dr. J. Massague forproviding the 1833 cell line, and Dr. Bert Vogelstein for providing the SBE4-Luc reporter plasmid. The authors thank Genentech for a kind gift ofErlotinib. Light microscopy was done at the Microscopy and Image AnalysisLaboratory at the University of Michigan. The tumors were processed at theUMCCC Tissue Core Facility.






Grant Support

This work was supported by the following grants: P01 CA85878,R01CA129623, R21CA131859, and P50CA093990.

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

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

Received May 17, 2011; revised September 1, 2011; accepted September12, 2011; published OnlineFirst September 26, 2011.

References1. Schmidt-Weber CB, Blaser K. Regulation and role of transforming

growth factor-beta in immune tolerance induction and inflammation.Curr Opin Immunol 2004;16:709–16.

2. Massague J. TGFbeta in cancer. Cell 2008;134:215–30.3. Meulmeester E, Ten Dijke P. The dynamic roles of TGF-beta in cancer.

J Pathol 2011;223:205–18.4. Ross S, Hill CS. How the Smads regulate transcription. Int J Biochem

Cell Biol 2008;40:383–408.5. Kang Y, HeW, Tulley S, Gupta GP, Serganova I, Chen CR, et al. Breast

cancer bone metastasis mediated by the Smad tumor suppressorpathway. Proc Natl Acad Sci U S A 2005;102:13909–14.

6. Giampieri S,ManningC,Hooper S, Jones L, Hill CS, Sahai E. Localizedand reversible TGFbeta signalling switches breast cancer cells fromcohesive to single cell motility. Nat Cell Biol 2009;11:1287–96.

7. Hida N, Awais M, Takeuchi M, Ueno N, Tashiro M, Takagi C, et al.High-sensitivity real-time imaging of dual protein-protein interac-tions in living subjects using multicolor luciferases. PLoS One 2009;4:e5868.

8. Korpal M, Yan J, Lu X, Xu S, Lerit DA, Kang Y. Imaging transforminggrowth factor-beta signaling dynamics and therapeutic response inbreast cancer bone metastasis. Nat Med 2009;15:960–6.

9. Serganova I, Moroz E, Vider J, Gogiberidze G, Moroz M, Pillarsetty N,et al. Multimodality imaging of TGFbeta signaling in breast cancermetastases. FASEB J 2009;23:2662–72.

10. Chong AK, Satterwhite T, Pham HM, Costa MA, Luo J, Longaker MT,et al. Live imaging of Smad2/3 signaling in mouse skin wound healing.Wound Repair Regen 2007;15:762–6.

11. Luo J, Lin AH, Masliah E, Wyss-Coray T. Bioluminescence imaging ofSmad signaling in living mice shows correlation with excitotoxicneurodegeneration. Proc Natl Acad Sci U S A 2006;103:18326–31.

12. Lin AH, Luo J, Mondshein LH, ten Dijke P, Vivien D, Contag CH, et al.Global analysis of Smad2/3-dependent TGF-beta signaling in livingmice reveals prominent tissue-specific responses to injury. J Immunol2005;175:547–54.

13. Macias-SilvaM, Abdollah S, Hoodless PA, Pirone R, Attisano L,WranaJL. MADR2 is a substrate of the TGFbeta receptor and its phosphor-ylation is required for nuclear accumulation and signaling. Cell1996;87:1215–24.

14. Luker KE, Smith MC, Luker GD, Gammon ST, Piwnica-Worms H,Piwnica-Worms D. Kinetics of regulated protein-protein interactionsrevealed with firefly luciferase complementation imaging in cells andliving animals. Proc Natl Acad Sci U S A 2004;101:12288–93.

15. Nyati S, Ranga R, Ross BD, Rehemtulla A, M.S. B. Molecular Imagingof GSK3b and CK1a kinases. Anal Biochem 2010;405:246–54.

16. Kang YB, Siegel PM, Shu WP, Drobnjak M, Kakonen SM, Cordon-Cardo C, et al. A multigenic program mediating breast cancer metas-tasis to bone. Cancer Cell 2003;3:537–49.

17. Zawel L, Dai JL, Buckhaults P, Zhou S, Kinzler KW, Vogelstein B, et al.Human Smad3 and Smad4 are sequence-specific transcription acti-vators. Mol Cell 1998;1:611–7.

18. Chen Y, Kam CS, Liu FQ, Liu Y, Lui VC, Lamb JR, et al. LPS-inducedup-regulation of TGF-beta receptor 1 is associated with TNF-alphaexpression in human monocyte-derived macrophages. J Leukoc Biol2008;83:1165–73.

19. Segawa S, Goto D, Yoshiga Y, Sugihara M, Hayashi T, Chino Y, et al.Inhibition of transforming growth factor-beta signalling attenuatesinterleukin (IL)-18 plus IL-2-induced interstitial lung disease in mice.Clin Exp Immunol 2010. 160;394–402.

20. Abramoff MD, Magalhaes PJ, Ram SJ. Image processing with Image.J Biophotonics International 2004;11:36–42.

21. Callahan JF, Burgess JL, Fornwald JA, Gaster LM, Harling JD, Har-rington FP, et al. Identification of novel inhibitors of the transforminggrowth factor beta1 (TGF-beta1) type 1 receptor (ALK5). J Med Chem2002;45:999–1001.

22. Laping NJ, Grygielko E, Mathur A, Butter S, Bomberger J, Tweed C,et al. Inhibition of transforming growth factor (TGF)-beta1-inducedextracellular matrix with a novel inhibitor of the TGF-beta type Ireceptor kinase activity: SB-431542. Mol Pharmacol 2002;62:58–64.

23. Inman GJ, Nicolas FJ, Hill CS. Nucleocytoplasmic shuttling of Smads2, 3, and 4 permits sensing of TGF-beta receptor activity. Mol Cell2002;10:283–94.

24. Matsuyama S, Iwadate M, Kondo M, Saitoh M, Hanyu A, Shimizu K,et al. SB-431542 andGleevec inhibit transforming growth factor-beta-induced proliferation of human osteosarcoma cells. Cancer Res2003;63:7791–8.

25. Zhang X, Arnott JA, Rehman S, Delong WG Jr, Sanjay A, Safadi FF,et al. Src is a major signaling component for CTGF induction by TGF-beta1 in osteoblasts. J Cell Physiol 2010;224:691–701.

26. Hanke JH, Gardner JP, Dow RL, Changelian PS, Brissette WH, Wer-inger EJ, et al. Discovery of a novel, potent, and Src family-selectivetyrosine kinase inhibitor. Study of Lck- and FynT-dependent T cellactivation. J Biol Chem 1996;271:695–701.

27. Kim JH, Jang YS, Eom KS, Hwang YI, Kang HR, Jang SH, et al.Transforming growth factor beta1 induces epithelial-to-mesenchymaltransition of A549 cells. J Korean Med Sci 2007;22:898–904.

28. Kasai H, Allen JT, Mason RM, Kamimura T, Zhang Z. TGF-beta 1induces human alveolar epithelial to mesenchymal cell transition(EMT). Resp Res 2005;6:56.

29. Huang Y, Hutter D, Liu Y, Wang X, Sheikh MS, Chan AM, et al.Transforming growth factor-beta 1 suppresses serum deprivation-induced death of A549 cells through differential effects on c-Jun andJNK activities. J Biol Chem 2000;275:18234–42.

30. Yagi K, Furuhashi M, Aoki H, Goto D, Kuwano H, Sugamura K, et al.c-myc is a downstream target of the Smad pathway. J Biol Chem2002;277:854–61.

31. Osherov N, Gazit A, Gilon C, Levitzki A. Selective inhibition of theepidermal growth factor andHER2/neu receptors by tyrphostins. JBiolChem 1993;268:11134–42.

32. Solovyan VT, Keski-Oja J. Proteolytic activation of latent TGF-betaprecedes caspase-3 activation and enhances apoptotic death of lungepithelial cells. J Cell Physiol 2006;207:445–53.

33. Jeong JH, An JY, Kwon YT, Li LY, Lee YJ. Quercetin-induced ubiqui-tination and down-regulation of Her-2/neu. J Cell Biochem 2008;105:585–95.

34. Phan TT, Lim IJ, Chan SY, Tan EK, Lee ST, Longaker MT. Suppressionof transforming growth factor beta/smad signaling in keloid-derivedfibroblasts by quercetin: implications for the treatment of excessivescars. J Trauma 2004;57:1032–7.

35. Wendt MK, Allington TM, Schiemann WP. Mechanisms of the epithe-lial-mesenchymal transition by TGF-beta. Future Oncol 2009;5:1145–68.

36. Luo J, Wyss-Coray T. Bioluminescence analysis of Smad-dependentTGF-beta signaling in live mice. Methods Mol Biol 2009;574:193–202.

37. de Caestecker MP, Parks WT, Frank CJ, Castagnino P, Bottaro DP,Roberts AB, et al. Smad2 transduces common signals from receptorserine-threonine and tyrosine kinases. Genes Dev 1998;12:1587–92.

38. DerynckR, ZhangYE.Smad-dependent andSmad-independent path-ways in TGF-beta family signalling. Nature 2003;425:577–84.

39. Park SH. Fine tuning and cross-talking of TGF-beta signal by inhibitorySmads. J Biochem Mol Biol 2005;38:9–16.

Nyati et al.





40. Moustakas A, Souchelnytskyi S, Heldin CH. Smad regulation in TGF-beta signal transduction. J Cell Sci 2001;114:4359–69.

41. MartinMM,Buckenberger JA, Jiang J,MalanaGE, Knoell DL, FeldmanDS, et al. TGF-beta1stimulateshumanAT1 receptor expression in lungfibroblasts by cross talk between theSmad, p38MAPK, JNK, andPI3Ksignaling pathways. Am J Physiol Lung Cell Mol Physiol 2007;293:L790–9.

42. JiangW,ZhangY,WuH, Zhang X,GanH, Sun J, et al. Role of cross-talkbetween the Smad2 and MAPK pathways in TGF-beta1-induced col-lagen IV expression in mesangial cells. Int J Mol Med 2010;26:571–6.

43. Dziembowska M, Danilkiewicz M, Wesolowska A, Zupanska A,Chouaib S, Kaminska B. Cross-talk between Smad and p38 MAPKsignalling in transforming growth factor beta signal transduction inhuman glioblastoma cells. Biochem Biophys Res Commun 2007;354:1101–6.

44. Yu L, Hebert MC, Zhang YE. TGF-beta receptor-activated p38 MAPkinase mediates Smad-independent TGF-beta responses. EMBO J2002;21:3749–59.

45. Wu S, Kasisomayajula K, Peng J, Bancalari E. Inhibition of JNKenhances TGF-beta1-activated Smad2 signaling in mouse embryoniclung. Pediatr Res 2009;65:381–6.

46. Engel ME, McDonnell MA, Law BK, Moses HL. Interdependent SMADand JNK signaling in transforming growth factor-beta-mediated tran-scription. J Biol Chem 1999;274:37413–20.

47. Bakin AV, Tomlinson AK, Bhowmick NA, Moses HL, Arteaga CL.Phosphatidylinositol 3-kinase function is required for transforminggrowth factor beta-mediated epithelial to mesenchymal transition andcell migration. J Biol Chem 2000;275:36803–10.

48. Abdel-WahabN,WicksSJ,MasonRM,Chantry A.Decorin suppressestransforming growth factor-beta-induced expression of plasminogenactivator inhibitor-1 in human mesangial cells through a mechanismthat involves Ca2þ-dependent phosphorylation of Smad2 at serine-240. Biochem J 2002;362:643–9.

49. Meyer-ter-Vehn T, Sieprath S, Katzenberger B, Gebhardt S, Grehn F,Schlunck G. Contractility as a prerequisite for TGF-beta-inducedmyofibroblast transdifferentiation in human tenon fibroblasts. InvestOphthalmol Vis Sci 2006;47:4895–904.

50. Inman GJ, Nicolas FJ, Callahan JF, Harling JD, Gaster LM, Reith AD,et al. SB-431542 is a potent and specific inhibitor of transforminggrowth factor-beta superfamily type I activin receptor-like kinase(ALK) receptors ALK4, ALK5, and ALK7. Mol Pharmacol 2002;62:65–74.






Correction

Correction: Molecular Imaging of TGFb-InducedSmad2/3 Phosphorylation Reveals a Role forReceptor Tyrosine Kinases in Modulating TGFbSignaling

In this article (Clin Cancer Res 2011;17:7424–39), which was published in theDecember 1, 2011, issue ofClinical Cancer Research (1), themiddle initial of oneof the coauthorswas omitted. The correct listing isMukeshK.Nyati. The authorsregret this error.

Reference1. Nyati S, Schinske S, Ray D, Nyati M, Ross BD, Rehemtulla A. Molecular imaging of TGFb-

induced Smad2/3 phosphorylation reveals a role for receptor tyrosine kinases in modulatingTGFb signaling. Clin Cancer Res 2011;17:7424–39.

Published OnlineFirst March 13, 2012.doi: 10.1158/1078-0432.CCR-12-0480�2012 American Association for Cancer Research.

ClinicalCancer

Research

www.aacrjournals.org 2115

2011;17:7424-7439. Published OnlineFirst September 26, 2011.Clin Cancer Res Shyam Nyati, Katrina Schinske, Dipankar Ray, et al. Signaling

βReveals a Role for Receptor Tyrosine Kinases in Modulating TGF-Induced Smad2/3 PhosphorylationβMolecular Imaging of TGF

Updated version

10.1158/1078-0432.CCR-11-1248doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2011/09/26/1078-0432.CCR-11-1248.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/17/23/7424.full#ref-list-1

This article cites 50 articles, 17 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/17/23/7424.full#related-urls

This article has been cited by 6 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/17/23/7424To 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-11-1248

http://clincancerres.aacrjournals.org/content/suppl/2011/09/26/1078-0432.CCR-11-1248.DC1

http://clincancerres.aacrjournals.org/content/17/23/7424.full#ref-list-1

http://clincancerres.aacrjournals.org/content/17/23/7424.full#related-urls

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/17/23/7424


molecular imaging of tgfb-induced smad2/3 phosphorylation ...tgfb signaling pathway regulates growth...

Documents