hedgehog signaling drives radioresistance and stroma...

Therapeutics, Targets, and Chemical Biology

Hedgehog Signaling Drives Radioresistance and Stroma-Driven Tumor Repopulation in Head and Neck SquamousCancers

Gregory N. Gan1, Justin Eagles2, Stephen B. Keysar2, Guoliang Wang2, Magdalena J. Glogowska2,Cem Altunbas1, Ryan T. Anderson2, Phuong N. Le2, J. Jason Morton2, Barbara Frederick1, David Raben1,Xiao-Jing Wang3,4, and Antonio Jimeno2,4,5

AbstractLocal control and overall survival in patients with advanced head and neck squamous cell cancer (HNSCC)

remains dismal. Signaling through the Hedgehog (Hh) pathway is associated with epithelial-to-mesenchymaltransition, and activation of the Hh effector transcription factor Gli1 is a poor prognostic factor in this diseasesetting. Here, we report that increased GLI1 expression in the leading edge of HNSCC tumors is further increasedby irradiation, where it contributes to therapeutic inhibition. Hh pathway blockade with cyclopamine suppressedGLI1 activation and enhanced tumor sensitivity to radiotherapy. Furthermore, radiotherapy-induced GLI1expression was mediated in part by the mTOR/S6K1 pathway. Stroma exposed to radiotherapy promoted rapidtumor repopulation, and this effect was suppressed by Hh inhibition. Our results demonstrate that Gli1 that isupregulated at the tumor–stroma intersection in HNSCC is elevated by radiotherapy, where it contributes tostromal-mediated resistance, and that Hh inhibitors offer a rational strategy to reverse this process to sensitizeHNSCC to radiotherapy. Cancer Res; 74(23); 7024–36. �2014 AACR.

IntroductionHead and neck squamous cell carcinoma (HNSCC) remains

a devastating malignancy with approximately 50% five-yearoverall survival (OS) rates (1). During and immediately aftertreatment, a process called accelerated repopulation occurs inHNSCC whereby a few surviving cells rapidly proliferate lead-ing to tumor recurrence (2) but this process still requireselucidation.

One proposed mechanism for how tumor epithelial cellsbecome resistant to therapy is transformation into a moremesenchymal phenotype known as epithelial-to-mesenchymaltransition (EMT; ref. 3). Recent reports have linked hedgehogpathway (HhP) signaling to EMT induction following (chemo)

radiotherapy in gastric (4, 5), esophageal (6), colorectal (7),ovarian (8), and endometrial (9) cancers. Transcriptional acti-vation of pro-EMT genes by Gli1 can lead to mesenchymalchanges (i.e., spindle cell shape, pseudopodia formation) andincreased tumor invasiveness. Targeting the HhP or its down-stream EMT targets can lead to reversal of the mesenchymal(stem cell "like") phenotype and restoration of chemora-diotherapy sensitivity. A retrospective analysis of RTOG 90-03 demonstrated that patients with HNSCC who expressedhigh levels of Gli1 before treatment had poor local control (LC)rates, more distant metastases, and worse OS; when correlatedwith those patients also expressing high levels of EGFR, thesepatients had the worst outcomes overall (10). We previouslydemonstrated inHNSCC xenografts that EGFR inhibition leadsto upregulation of the HhP in EGFR-dependent, human pap-illoma virus (HPV)-negative HNSCC (11). The interrelatednessof these different pathways suggests that inhibition of bothEGFR and the HhP may be necessary to maximize the effec-tiveness of radiotherapy in HNSCC.

The role HhP plays in radiotherapy resistance and tumorrepopulation in HNSCC remains unknown. We hypothesizethat radiotherapy similarly modulates the HhP in EGFR-dependent tumors such as HNSCC and could be a mechanismfor resistance. Our work examines intratumor regional differ-ences in HhP signaling, how radiotherapy modulates the HhP,and the effect of HhP inhibition with cyclopamine on HhPgene expression, cell survival, and EMT phenotype in thecontext of radiotherapy. We propose an alternative mecha-nism where radiotherapy-induced Gli1 nuclear (nGLI1) trans-location and GLI1 gene expression is mediated through themTOR pathway. Finally, we show that the tumorigenicity of

1Department of Radiation Oncology, University of Colorado School ofMedicine, Aurora, Colorado. 2Division of Medical Oncology, Departmentof Medicine, University of Colorado School of Medicine, Aurora, Colorado.3Department of Pathology, University of Colorado School of Medicine,Aurora, Colorado. 4Charles C. Gates Center for Stem Cell Biology, Uni-versity of Colorado School of Medicine, Aurora, Colorado. 5Department ofOtolaryngology, University of Colorado School of Medicine, Aurora,Colorado.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Current address for G.N. Gan: UNM Cancer Center, Internal Medicine,Division of Radiation Oncology, Albuquerque, NM.

CorrespondingAuthor:Antonio Jimeno, Professor ofMedicine/Oncologyand Otolaryngology, University of Colorado School of Medicine, 12801East 17th Avenue, Room L18-8101B, Aurora, CO 80045. Phone: 303-724-2478; Fax: 303-724-3892; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-14-1346

�2014 American Association for Cancer Research.

CancerResearch

Cancer Res; 74(23) December 1, 20147024

on August 13, 2019. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 8, 2014; DOI: 10.1158/0008-5472.CAN-14-1346

http://cancerres.aacrjournals.org/

serially cotransplanted tumor and stroma cells increases afterradiation and can be abrogated by HhP inhibition.

Materials and MethodsCell lines and drugsHN11 and TU167 HNSCC cell lines have been previously

described (11–17). siRNA (AKT, S6K1) was completed inserum-free media using 1.3 mL/mL Dharmafect 1 (ThermoScientific) and 100 nmol/L siRNA. GLI1 silencing was com-pleted using doxycycline-inducible (0.5 mg/ml) pTRIPZ lenti-viral constructs (RHS4696-99636732) that introduced shRNAas previously described (11). Cells were treated with 1 mmol/Lcyclopamine (LC Laboratories), 1 mmol/L Rapamycin (LCLaboratories), or 1 mmol/L PF-4708671 (Sigma-Aldrich) for aminimum of 12 hours before irradiation. Generation of chron-ically irradiated cell lines were performed by irradiating cells at50% to 60% confluency daily at 2 Gy/day for 5 days. Cells wereallowed to recover for 1 to 2 weeks before initiating the secondirradiation cycle. Cells were allowed 1 to 2 weeks for recoverybefore initiating any studies.

IrradiationCells and animals were irradiated using a RS-2000 (Rad

Source Technologies, Inc). Cells were irradiated with 2 or 10Gy using a 160 KVp source, at 25 mAmp, and at a dose rate of1.15 Gy/minute. Mice were irradiated using a customizedanimal restraint exposing only the head/neck region of themouse. Radiation dose and dose rate were calculated empir-ically for the RS-2000.

Colony formation assaySerial dilutions were performed and single cells were plated

on 6-well plates. Cells were allowed to adhere for 16 hoursbefore drug treatment and irradiated with a single dose (0–10Gy) at least 8 hours after drug treatment. Cells were washed 3days after initial drug exposure and grown for an additional 7 to10 days until visible colonies on the control plate could bemeasured. Cells were fixed in 10% formalin solution, stainedwith 0.5% crystal violet and washed with cold tap water.Threshold for positive proliferative colonies were �50 cells.Experiments were replicated either two or three times togenerate the radiotherapy dose response curves or the radio-therapy–cyclopamine cell sensitivity assay curves, respectively.

Human xenograft therapeutic studiesHN11 and TU167 parental cell lines were implanted into

athymic nude mice using an established floor-of-the mouth(FOTM) method as previously described (18). Seventy-fivethousand cells were resuspended 1:1 in a DMEM/FBS andMatrigelmixture and kept on ice until tumor implantation. A 4-arm study was performed to generate efficacy data. Arms usedfor the parental tumors were: (i) control; (ii) cyclopamine 40mg/kg by oral gavage everyday for 15 days excludingweekends;(iii) radiotherapy 5 Gy once per week for 3 weeks; or (iv) both.When the average tumors reached 100 mm3, mice were dis-tributed into their respective groups. Tumor sizewas evaluatedtwice per week using the following formula: volume ¼ [length� width2]/2. For pharmacodynamic analysis on experimentalday 10, tumorswere extracted 1 hour after drug administration

and/or 48 hours after radiotherapy and divided into one of twogroups: formalin fixation or FACS.

We performed a pharmacodynamic study by implantingHNSCC into FOTM and assigning them to one of the fourtreatment groups (Supplementary Fig. S3A). After receiving theprescribed therapy, tumors were collected from each group,mouse stroma (anti-mouse H2Kd, BioLegend) and humantumor cells (anti-Epcam; Cell Signaling Technology) wereseparated by FACS, recombined in a 1:3 ratio (5,000:15,000)with untreated tumor cells, and implanted onto the flanks ofathymic nude mice. Tumor regrowth was assessed at 3 and 6weeks after implantation and all tumors were excised at 6weeks for analysis.

Statistical analysisTwo-tailed, student paired t test (GraphPad Prism) was used

to compare groups for both in vitro and in vivo experiments. Amajority of cell-based experiments were replicated three timeswith a minimum of three replicates per experiment. In vivoanimal experiment statistics were performed on aminimumof5 animals per group for comparison. Statistical comparisonswere performed by group. Statistical values were shown only ifP < 0.05.

Additional RNA, protein, and immunohistochemical meth-ods are discussed in Supplementary Materials and Methods.

ResultsHhP upregulation in the growing edge of HNSCC tumors

Utilizing our established patient-derived xenografts (PDX)model system (18), mouse stroma, tumor margin, and tumorcenter were isolated by laser capture microdissection (LCM)from control or irradiated flash-frozen flank tumors (Fig. 1A),and evaluated for species-specific GLI1 expression. In nonir-radiated tumors, hGLI1 at the tumor margin was 4.7 timeshigher compared with the tumor center (P ¼ 0.004; Fig. 1B).When comparing irradiated tumor to nonirradiated tumor,normalized hGLI1 was significantly upregulated at the margin(8.8�; P ¼ 0.02) compared with the center (2.7�; P ¼ 0.2).Human sonic hedgehog (SHH), was also significantly upregu-lated at the tumor margin by 9.2 times in irradiated tumors (P¼ 0.02). Furthermore, mGli1 demonstrated a nonsignificanttrend toward upregulation (4.1�) in the mouse stroma(Fig. 1C). Results were performed from a single experiment intriplicate.We examinedGli1 expression following radiotherapyby immunohistochemistry (IHC; Fig. 1D) at a single time pointthat demonstrated upregulation of Gli1 protein but was notdifferentially increased between the margin and center.

Radiotherapy upregulates the HhPWe used two established human epithelial HPV-negative

HNSCC cell lines collected from patients with FOTM SCC (19),HN11 from a primary tumor, and TU167 from a relapsedsubject to confirm our in vivo observations. We determinedwhole transcriptome expression differences quantitatively andfound dissimilar gene expression levels in the hedgehog, epi-thelial cell markers, and mTOR pathways (Supplementary Fig.S1). Consistent with being a relapsed, more transformed celltype, TU167, had decreased epithelial cell markers and

Tumor and Stromal Dependence on Hedgehog Pathway Signaling

www.aacrjournals.org Cancer Res; 74(23) December 1, 2014 7025




increased HhP and mTOR pathway expression. After radio-therapy, GLI1 as well as downstream EMT genes ZEB2 andSNAI1 were acutely and significantly upregulated at 24 and 48hours (Fig. 2A).

EMT is a potential mechanism explaining why LC can becompromised due to accelerated repopulation following pro-longed genotoxic stress. We explored whether chronic irradi-ation modulated HhP and EMT-associated genes. Chronicallyirradiated HN11 demonstrated upregulation in HhP (GLI1,GLI2) and EMT (ZEB2, VIM) genes compared with the parentalline. In TU167, we observed upregulation of the EMT pathwaygenes GLI2 and TWIST1 (Fig. 2B). FACS of chronically irradi-ated HN11 demonstrated increased VIM expression (�4% ofthe total population), indicating that a small fraction of thesecells have assumed a mesenchymal phenotype (Fig. 2C). VIMwas higher in parental TU167 compared with HN11 and wasnot significantly upregulated by sustained radiotherapy.

Effect of cyclopamine on radiotherapy-mediated GLI1upregulation

To test whether radiotherapy-induced GLI1 upregulationcould be suppressed through HhP inhibition, we used theSmoothened receptor inhibitor cyclopamine. Pretreatmentwith 1 mmol/L cyclopamine followed by radiotherapy signifi-cantly suppressed radiotherapy-induced GLI1 in HN11 but notin TU167 (Fig. 3A). Using shGLI1, we documented partial GLI1suppression in HN11 but not in TU167 following radiotherapy(Supplementary Fig. S2). Cells pretreated with 1 mmol/L cyclo-pamine demonstrated no significant change in nGLI1 contentat 1, 2, and4hours.Despite inhibitionwith cyclopamine,we stillobserved a 2-fold increase in nGli1 in HN11 and TU167 (Fig. 3B,first and thirdpanels) after radiotherapy. Similarly, cytoplasmicGli1 (cGli1) was similarly increased following irradiation andsuppressedwithcyclopaminepretreatment. Interestingly, cGli1levels were unable to be fully suppressed following irradiation

Figure 1. Radiotherapy (RT) inducesGLI1 expression in vivo. A, schemadescribing tumor sections ofinterest that were laser capturemicrodissected. B, the amount ofnonirradiated human GLI1 mRNAexpression at the tumor margincompared with the tumor centerwas determined by qRT-PCR fromLCM tissue. C, irradiated andnonirradiated LCM tissue wasanalyzed by qRT-PCR usingspecies specific probes for GLI1and SHH to determineradiotherapy-induced gene foldexpression over baseline. D, twodifferent PDX � radiotherapy.Tumors were harvested, formalinfixed, sectioned, and stained withanti-EGFR or anti-Gli1 for IHC(�20; bar; 100 mm). Statisticallysignificant findings are denoted:�, P < 0.05.

Gan et al.

Cancer Res; 74(23) December 1, 2014 Cancer Research7026




despite pretreatment with cyclopamine. This effect was morepronounced inTU167comparedwithHN11(Fig. 3B, secondandfourth panels). Treating with higher (but not clinically achiev-able) concentrations of cyclopamine (10 mmol/L) led to signif-icant inhibition of radiotherapy-induced GLI1 expression sug-gesting a dose–response effect (data not shown).Using colony formation assay (CFA), the primary cell line

HN11 was significantly more radiosensitive than TU167 atlower radiotherapy doses (Fig. 3C). We then evaluated whethercyclopamine inhibition could enhance cytotoxicity at singleradiation dose. Combinatorial therapy with cyclopamine þradiotherapy demonstrated a significant effect compared withsingle modality therapy or no treatment (control vs. 1 mmol/Lþ radiotherapy: TU167, P ¼ 0.003; HN11, P ¼ 0.0006) with thegreatest response seen in HN11 (Fig. 3D) compared withTU167 (Fig. 3E). Chronically irradiated HN11 cells exposed to1 mmol/L cyclopamine for 12 hours demonstrated a trendtowards suppressingGLI1 (1.7� 0.3–1.5� 0.2, n.s.),GLI2 (1.7�0.03–1.1 � 0.02, P ¼ 0.0025), and VIM (3.0 � 0.2–1.1 � 0.1, P ¼0.0043). No trend was observed for TU167 for GLI1, GLI2, orVIM. These findings demonstrated that chronically irradiatedHN11 cells that induce HhP- and EMT-associated genes wereresponsive to cyclopamine inhibition.

In vivo effect of radiotherapy and cyclopamine in anorthotopic HNSCC modelTo further our in vitro observations, we evaluated tumors in

vivo, combining cyclopamine and radiotherapy. Mice were

implanted in the FOTM (Fig. 4A) with either HN11 or TU167cell lines and randomized to one of the four treatment arms(Fig. 4B). Radiotherapy � cyclopamine led to a reduction intumor volume, compared with control or cyclopamine alone inHN11 and TU167 parental tumors (Fig. 4C and D). Of note,cyclopamine had no effect on inhibiting tumor growth inHN11and, similar to the control arm, the tumor continued to grow. Ithas not been reported in the literature that cyclopaminestimulates tumor proliferation and, given the size of the errorbar compared with the control arm, is unlikely to be suggestiveof the above hypothesis. In HN11, tumor volumes respondeddramatically to radiotherapy (67% reduction) or the combi-nation treatment (77% reduction), which correlated well withtheir in vitro susceptibility to radiotherapy (Fig. 4C). In TU167,we observed smaller tumor sizes in both the radiotherapyalone and combinatorial therapy arms (radiotherapy: 37%reduction compared with control; cyclopamineþ radiothera-py: 49% reduction compared with control; Fig. 4D). TU167-irradiated tumors had a propensity for developing cystic,cavitated lesions (radiotherapy: 3/10 tumors; dual therapy:8/10; Supplementary Fig. S3A) beginning around day 25. Atday 45, representative cystic and noncystic TU167 tumors wereharvested from euthanized mice for analysis. The noncystictumors by hematoxylin and eosin (H&E)were solid throughoutcomparedwith the cystic tumors, whichwere only a thin rim oftissue (Supplementary Fig. S3B). Tumor proliferation evaluat-ed using Ki67 for the noncystic tumors revealed that control(H-score: 80) or cyclopamine-treated (H-score: 80) tumors

HN11A

B C

0 24 48 0 24Time (h)

48 0 24 48 0 24 48 0 24Time (h)

1.0%

Vimentin-FITC

SS

SS

Vimentin-FITC

4.6%

1.9%

* P < 0.05; ** P < 0.01;*** P < 0.001

1.8%

Parental

HN11

TU167

GLI1GLI2

ZEB2

Vimen

tin

Twist

1

Snai1

Chronic RT

48 0 24 48

876543210

9876543210

Rel

ativ

e g

ene

exp

ress

ion

to n

on

irra

dia

ted

cel

ls

5

4

3

2

1

0

Rel

ativ

e g

ene

exp

ress

ion

to p

aren

tal l

ine

TU167

GLI1

HN11 chronically irradiatedTu167 chronically irradiated

ZEB2SNAI1

Figure 2. Radiotherapy (RT) inducesin vitro GLI1 expression. A,mechanistic studies wereperformed on two HNSCC celllines, TU167 and HN11. GLI1,ZEB2, and SNAI1 expression wasevaluated by qRT-PCR at 0, 24,and 48 hours. B, chronicallyirradiated cells comparedwith theiroriginal parental cell lines wereevaluated for HhP or EMT-geneexpression by qRT-PCR. C,expression levels of vimentin wereevaluated between parental andchronically irradiated cells byFACS. Statistically significantfindings are denoted: �, P < 0.05;��, P < 0.01; ��, P < 0.001.






HN11GLI1

HN11

Fold change(baseline):


nGli1

Lamin B1

cGli1

Actin



nGli1

Lamin B1

cGli1

Actin

10 Gy:

1 mmol/L Cyclo:

Time (h):

A

B

C D E

24 h

0 1 0Concentration (mmol/L)

1 0 1 0Concentration (mmol/L)

1

48 h

24 h48 h8

76543210

Rel

ativ

e g

ene

exp

ress

ion

to

no

nir

rad

iate

d c

ells

0 2 4 6 8 10

Control

Cyclo

Cyclo

+ R

TRT

Control

Cyclo

Cyclo

+ R

TRTRadiotherapy (dose)

1

0.1

0.01

0.001

Su

rviv

ing

fra

ctio

n (

%)

(Lo

g s

cale

)

1.2

1.0

0.8

0.6

0.4

0.2

0.0Su

rviv

ing

fra

ctio

n (

%) 1.2

1.0

0.8

0.6

0.4

0.2

0.0Su

rviv

ing

fra

ctio

n (

%)

9876543210

Rel

ativ

e g

ene

exp

ress

ion

to

no

nir

rad

iate

d c

ells

TU167

HN11TU167

HN11

n.s.

n.s.

TU167

TU167

*P < 0.05; **P < 0.01; ***P < 0.001

Figure 3. Effect of cyclopamine on radiotherapy induced GLI1 gene expression and cytotoxicity. A, cells were pretreated with 1 mmol/L cyclopamine 8 hoursbefore receiving radiotherapy. Cellswere harvested at specific time points andGLI1 expression evaluated by qRT-PCR. B, nuclear extractswere generated aspreviously described from cells � 1 mmol/L cyclopamine � radiotherapy. (Continued on the following page.)

Gan et al.





demonstrated the highest amount of Ki67 staining comparedwith radiotherapy (H-score: 65) or dual therapy–treated (H-score: 40) tumors. The dual therapy–treated tumors demon-strated the most reduced Ki67 staining both in the peripheryand core (Supplementary Fig. S3C). This suggests that therewas both increased necrosis and decreased proliferation inradiotherapy and dual treated TU167 tumors, in spite of smallvolumetric differences. In comparison, the cystic versus non-cystic radiotherapy treatment tumors suggest less proliferat-ing cells in the cystic tumors and growth is restricted to thetumor margin/edge.Tumors were harvested for pharmacodynamic analysis 10

days into treatment and immunohistochemical staining per-formed for EGFR, Gli1, ALDH1, and BMI1. EGFR stainingconfirmed presence of HNSCC cells over mouse stroma. FaintGli1 expression was observed in control cells and this wassignificantly upregulated throughout the tumor followingradiotherapy. Radiotherapy-induced Gli1 expression could bepartially abrogated by cyclopamine treatment [Fig. 5, HN11(left) and TU167 (right)]. When evaluating the mouse stroma,we similarly observed an increase in stromal Gli1 followingradiotherapy and a pronounced decrease with dual therapy by

H-score. In particular, we observed a pronounced effect withcyclopamine alone in suppressing mouse stromal GLI1 inTU167 over HN11 (Supplementary Table S1). Similarly, ALDH1and BMI1 were upregulated following radiotherapy and par-tially inhibited by cyclopamine.

Given our in vitro and in vivo findings, we next sought toexplore: (i) an alternative mechanism that may drive tumorGli1 cytoplasmic-to-nuclear translocation independent of thecanonical HhP and (ii) whether inhibition of HhP in stromalcells can block tumor repopulation in vivo.

mTOR/S6K1 mediates GLI1 cytoplasmic-to-nucleartranslocation

In both cell lines, despite inhibition with significant con-centrations of cyclopamine, radiotherapy was still able toinduceGLI1 gene expression and increase nGli1 accumulation,suggesting that an accessory pathway (independent of canon-ical HhP signaling) may be driving radiotherapy-induced nGli1accumulation. Therefore, we hypothesized that radiotherapycould activate the mTOR/S6K1 pathway and enhance Gli1cytoplasmic-to-nuclear translocation (Fig. 6A). Followingradiotherapy, GLI1 was upregulated by 24 hours in both HN11

(Continued.) nGLI1 accumulationwas evaluated by immunoblot and temporal fold changes described. C, CFA radiotherapy dose curveswere generated fromserially diluted cells grown on 6-well plates treated 24 hours after adhesion with 0–10 Gy irradiation. Cells were fixed, stained, and counted approximately8–10 days after radiotherapy. CFAs were performed to assess the effect of cyclopamine and 2 Gy radiotherapy on cell growth and survival on HN11 (D) andTU167 (E). Eight hours before irradiation, cellswere pretreatedwith 1mmol/L cyclopamine in lowserummedia. Lowserummediawas chosenas itmorecloselyrepresents the tumor milieu. Furthermore, use of low serum media before drug treatment will prevent exogenous growth factors from competing with theeffects of cyclopamine. Statistically significant findings are denoted: �, P < 0.05; ��, P < 0.01; ��, P < 0.001. n.s., nonsignificant.

Athymic nude miceFOTM tumors

Control

Cyclopamine(40 mg/kg x15 days)

RT (5 Gy x3 treatment)

Dual therapy

Tumors treated with cyclopamine daily except for weekends

BA

C HN11

175

150

125

100

75

50

25

0

% T

um

or

no

rmal

ized

to c

on

tro

l

n.s.

*****

n.s.

TU167D

** n.s.

*

*

* P < 0.05; ** P < 0.01; *** P < 0.001

ControlCyclopamineRTDual therapy

150

125

100

75

50

25

0

% T

um

or

no

rmal

ized

to c

on

tro

l

Figure 4. In vivo effects ofcyclopamine and radiotherapy (RT).A, figure describing location oforthotopic FOTM tumor implantedin athymic nude mouse. B, schemadescribing cyclopamine and/orradiotherapy dose and duration foranimal tumor growth studies.Tumor growth for orthotopicallyimplanted FOTM HN11 (C) andTU167 (D) parental tumors wasassessed at 28 days posttreatmentinitiation.






Figure 5. Radiotherapy (RT)-induced GLI1,ALDH1, BMI1 expression. IHC was performedon formalin-fixed tumors and stained witheither H&E, EGFR, GLI1, ALDH1, or BMI1.Differences in IHC (�20; bar 100,mm)expression were evaluated by H-score.

Gan et al.





and TU167 (Fig. 6B), and pretreatment with 1mmol/L ZSTK474(AKT inhibitor) was able to significantly suppress GLI1 expres-sion inHN11 following radiotherapy but not in TU167 (data notshown). This resistance was consistent with the observedoverexpression of the mTOR pathway in TU167, and use ofa more specific inhibitor was warranted. Following pretreat-ment with 1 mmol/L rapamycin, we observed suppression ofboth GLI1 mRNA and Gli1 cytoplasmic-to-nuclear proteinaccumulation following radiotherapy in TU167 (Fig. 6C). Weobserve a similar increase in cGli1 following irradiation and theirradiation effect can similarly be suppressed with rapamycinpretreatment. In contrast, HN11 inhibition with rapamycin didnot prevent radiotherapy-induced nGli1 accumulation (Sup-plementary Fig. S4) and is in contrast to ourmRNA results. Ourtranscriptome resultsmay offer an answer for this discrepancy,where TU167 has greater mTOR gene expression compared

with HN11 and may be more sensitive to the effects of mTORinhibition, which may translate into diminished Gli1 accumu-lation. Cellular Gli1 content is known to be regulated both atthe transcriptional and proteasomal level (20). The suppres-sion of GLI1may suggest that the role of rapamycin may be toregulate gene expression and perhaps is irrespective of nuclearprotein accumulation.

Next, we evaluated whether inhibition of the downstreammTOR effector S6K1 could prevent nGli1 accumulation. Sim-ilarly, GLI1 expression was significantly reduced in the pres-ence of S6K1 inhibitor (1 mmol/L PF-4708671) following radio-therapy (Fig. 6B). We attempted to determine whether loss ofS6K1 could similarly suppress GLI1 expression and preventnuclearGli1protein accumulation. Four siRNA constructswerescreened and their ability to suppress S6K1 expression bymRNA and immunoblot were assessed. The two best candidate

Figure 6. Radiotherapy (RT)-induced mechanism for nuclear GLI1 accumulation. A, schema proposing a mechanism for radiotherapy mediatedGLI1 cytoplasmic-to-nuclear translocation. B, targeted inhibitors against mTOR and S6K1 were used on cells pretreated with drug 12 hours beforeradiotherapy to determine whether GLI1 gene expression could be inhibited. Cells were harvested 24 hours after irradiation and GLI1 gene expressionanalyzed by qRT-PCR. C, cytoplasmic and nuclear extracts were generated as previously described from TU167 treated with �1 mmol/L rapamycin� radiotherapy. cGli1 and nGli1 accumulation was evaluated by immunoblot and temporal fold change was described. Statistically significant findingsare denoted: �, P < 0.05; ��, P < 0.01.






constructs were then selected. Both siRNA constructs sup-pressed expression of S6K1. ppS6K1 was still observed despiteknockdown and was completely absent following radiothera-py. Our results suggest that inhibition with one siRNA con-struct was able to suppress nGli1 accumulation in both HN11and TU167. However, the other siRNA construct gave contraryresults and does not appear to suppress radiotherapy-inducedGli1 nuclear accumulation (Supplementary Fig. S5). Our find-ings suggest a possible link between mTOR/S6K1 and nGli1accumulation but must be interpreted with caution.

In vivo tumor repopulation following radiotherapyLocoregional failure due to tumor repopulation following

radiotherapy or chemoradiotherapy remains a significantproblem. In vitro and in vivo studies have implicated thatgenotoxically stressed tumor or fibroblast cells may play arole in tumor repopulation. In vivo, tumor and tumor-asso-ciated fibroblasts (stroma) are closely admixed with oneanother. On the basis of our PDX and TU167 in vivo resultsdemonstrating increased mouse stromal Gli1 following radio-therapy, subsequent suppression by pretreatment with cyclo-pamine and the significant response of TU167 tumors to dualtherapy, we hypothesized that cyclopamine may also targetthe tumor stroma and given its combined affect on TU167 invivo, may inhibit tumor repopulation. We treated orthotopi-cally implanted TU167 tumors with vehicle, cyclopamine,radiotherapy, or the combination, surgically resected them,separated tumor and stroma compartments by flow cytome-try, and coimplanted these in different permutations withuntreated tumor cells (Supplementary Fig. S6A). Thisaddressed whether the irradiated tumor or stroma couldenhance tumor repopulation and whether inhibition of theHhP could abrogate repopulation. We observed a significant10-fold difference in tumor regrowth at 3 weeks favoring theirradiated stroma combined with fresh tumor cell lines (SRT:T) compared with the nonirradiated control stroma:tumor(SC:T) pairing (P ¼ 0.01). When stroma cells pretreated withcyclopamine and radiotherapy were combined with freshtumor (SD:T), we observed a significant reduction in tumorregrowth rates at 3 weeks when SD:T was compared againstSRT:T (P¼ 0.007) implicating that HhP following radiotherapymay be involved in stroma-mediated tumor repopulation (Fig.7A). We also evaluated TRT:T pairings and observed a similarthough nonsignificant trend in tumor regrowth at 3 weeks.(Fig. 7A; Supplementary Table S2). When we evaluated thesame pairings between SRT:T versus TRT:T, there was greaterthan a 4-fold difference in size favoring the SRT:T over thetumor:tumor pairing (P ¼ 0.03; Supplementary TableS2). Figure 7B shows representative mice with S:T and T:Tpairings at 6 weeks. At 6 weeks, the TRT:T pairings continuedto demonstrate a significant trend towards smaller tumorsizes (267 � 27 mm3 vs. 579 � 102 mm3) compared with theSRT:T pairings (P ¼ 0.05). Representative excised tumors fromall animals continued to demonstrate that SRT cells conferreda significant growth advantage over SC or SCyc (nonradiother-apy) stroma. Furthermore, tumors pretreated with cyclopa-mine were able to inhibit tumor regrowth despite beingirradiated (Fig. 7A and C).

DiscussionRadiotherapy remains the cornerstone for local manage-

ment of HNSCC either definitively or in the postoperativesetting. Whether radiotherapy can induce GLI1/HhP upregu-lation through an EMT phenotype as a mechanism for tumorrepopulation and the respective relevance of irradiated cancercells versus intratumor stromal cells for tumor repopulationremain unknown. Accelerated tumor repopulation followingradiotherapy is associated with locoregional recurrence inpatients and a better understanding of this mechanism iscritical to improve LC and OS.

The interface between tumor margin/edge and tumor stro-ma is an important distinction for cancer biology. First, wedemonstrated that GLI1 is significantly upregulated at thetumor edge compared with the core in nonirradiated samples.Second, following radiotherapy, GLI1 gene expression wasupregulated at the peripheral tumor edge compared with thetumor core. Of note, this was taken at a solitary time point (48hours after radiotherapy), and even though overall Gli1 contenthas increased in both cells, there is no observed preference forstaining at the tumor edge. On the basis of this experiment,accumulation of nGli1 preferentially at the tumor edge mayoccur more rapidly as our immunoblot results suggest andtherefore tumor harvested at an earlier time point mightillustrate this.

First, the implications of our findings are important becausetumors with upregulated Gli1 and/or the EMT pathway havebeen shown to be more chemo- and radioresistant, morelocally invasive, and metastatic (7, 21–23). Second, the periph-eral tumor cells model those cell islands left behind aftersurgical resection ("positive margin"). This is clinically signif-icant as a positive margin indicates a high risk for recurrenceand patients with positive margins often experience greaterlocal and distant failure rates when they fail to receive ade-quate doses of radiotherapy/chemoradiotherapy (24, 25). Thechange in tumor milieu, increased inflammation, and localhypoxia following surgery may provide additional factors thatallow a high Gli1-expressing tumor more opportunity to pro-liferate/flourish compared with a low Gli1-expressing tumor.Our results demonstrated that GLI1 expression is inherentlyhigher at the tumor margin compared with the core and thatradiotherapy can further induce GLI1 overexpression at thetumor margin. However, these findings need to be cautiouslyinterpreted andwe caveat that tumor heterogeneity may causevariations in gene expression between tumor margin andcenter and this may explain the more uniform Gli1 proteinexpression following irradiation observed by IHC as comparedwith the quantitative real-time PCR (qRT)-PCR results, whichdemonstrated greater GLI1 expression at the tumor margin.Another explanation is that radiotherapy leads to GLI1 upre-gulation, which in turn upregulates certain EMT-associatedgenes. A limited number of these cells that are more EMT-likecan transform to amoremesenchymal phenotype and settle into the existing environment or translocate away from thesource of insult (i.e., local radiotherapy). As they are awayfrom the area of insult, these cells can downregulate variousmesenchymal genes and reestablish their epithelial phenotype(mesenchymal-to-epithelial phenotype) allowing them to grow

Gan et al.





in a new microenvironment. This has been illustrated in asquamous cell epithelial cancer model using a conditionaldoxycycline regulatory system controlling TWIST1 expression(26). The authors demonstrated that TWIST1 expression wasnecessary for EMT but shutting down TWIST1 was necessaryfor establishing distant metastasis. This explanation takes intoaccount both the tumor margin and heterogeneity issuesproviding a possible mechanism for Gli1-mediated treatmentfailure and distant metastasis.In the chronically irradiated cells, we observed that HhP

genes, GLI1 andGLI2, and EMT genes, ZEB2, TWIST1, and VIM,were upregulated though not as robustly as when cells were

acutely irradiated. These genes remained elevated 2 to 8 weeksafter radiotherapy, suggesting that a surviving population ofirradiated cancer cells adapt gene expression, allowing forsurvival and repopulation. Other groups, including our own,have demonstrated that inhibition of HhP or EMT pathwaygenes, TWIST, SNAI1, or SLUG, can reverse (chemo)radiother-apy resistance and decrease tumor invasiveness, progression,and metastasis (8, 9, 11, 22, 27). Therefore, we decided toevaluate whether HhP inhibition could enhance tumor controlby either increasing radiosensitivity or preventing tumorrepopulation in vivo. A recent article indicated that inadequateradiotherapy doses in other tumor cell lines can lead to

Figure 7. Effect of radiotherapy (RT)on in vivo tumor repopulation inTU167. A, all potential flank tumorswere measured at 3 (top) and 6(bottom) weeks postimplantation.Stromal:tumor (S:T) and tumor:tumor (T:T) combinations wereimplanted on the left and rightflanks, respectively. B, red, orange,green, blue labels correspondto pretreatment control,cyclopamine, radiotherapy anddual therapy groups. At 6 weekspostimplantation, finalmeasurements were taken andtumors harvested for analysis.C, representative excised tumorsare shown for each of the fourpretreatment arms, withcorresponding left and right flanktumors for the twodifferent groups.






increased tumor regrowth mediated by the HhP (21). There-fore, direct inhibition of Gli1 activation in vivo may lead toenhanced tumoricidal effects when combined with radiother-apy. When we treated our chronically irradiated HN11 cellswith cyclopamine, downregulation of GLI2 and VIM occurred,indicating that HhP inhibition in HNC may be an appropriateanticancer target when combined with radiotherapy.

Our IHC results demonstrated thatGli1 is upregulated in bothtumors and in the tumor stroma following radiotherapy and thisupregulationwas mitigated with cyclopamine. Interestingly, thedownstream stem cell marker and DNA damage responseprotein ALDH1 and BMI1, respectively, were both downregu-lated following cyclopamine therapy, suggesting that thesegenes are regulated by Gli1 and thatHhP inhibitionmay preventor slow the transition to a more "stem cell–like" state. In onestudy evaluating patients with high versus low ALDH1 expres-sion pre- and postsurgery, patients with colorectal cancerundergoing neoadjuvant chemoradiotherapy who expressedhigh levels of ALDH1 had greater rates of local failure andrecurrence (28). Expression of ALDH1 in HNSCC and colorectalcancer is associated with increased nodal and distant metasta-sis, worse OS, and is mediated through TWIST1 and SNAI1expression (29, 30). These findings highlight the concern thatchemoradiotherapy can induce ALDH1 (and BMI1) expressionwhich is a poor prognostic factor. Our results are the first todemonstrate that adding cyclopamine to radiotherapy cansuppress Gli1 expression in HPV-negative HNC and therebydownregulate ALDH1 and BMI1 expression in vivo.

Accelerated tumor repopulation following radiotherapy is awell-known phenomenon but its mechanism remainsunknown. Our in vitro results suggest that both HN11 andTU167 upregulate the HhP acutely and chronically followingradiotherapy. Furthermore, cyclopamine can suppress radio-therapy-induced GLI1 in HN11 but not in TU167 despite our invivo results, and this is likely due to lower baseline levels of HhPexpression in HN11 compared with TU167. Given the differ-ence in our in vitro versus our in vivo data, this led us toquestion whether an alternative target may be the stroma. Ourown LCM qRT-PCR and IHC results (Figs. 1C and 4E) show themouse stroma also upregulated GLI1 following radiotherapybut that GLI1 is suppressed following cyclopamine treatment.One possible explanation is that irradiated stromamay providea nurturing environment to allow circulating mesenchymaltumor cells a nidus to grow. Researchers have demonstratedthat irradiated fibroblasts undergo changes, including matrixremodeling (i.e., release of MMPs) and cytokine release (i.e.,TGFb, basic FGF). This may allow for an altered albeit richgrowth environment allowing tumor cells to flourish. A grow-ing body of literature in prostate, breast, pancreas, and othercancers suggests that HhP upregulation in the tumor micro-environment through paracrine signaling plays a substantialrole in tumor growth, survival, angiogenesis, and metastasis(27, 31–34). More specifically, cancer-associated fibroblasts(CAF) have been shown to upregulate both the Smoothenedreceptor and in turnGLI1 (35) and SNAI1 (36), whereas anothergroup demonstrated that inhibition of Smoothened onCAFs inturn translated to decreased GLI1 expression and the down-stream EMT gene SNAI1 in pancreatic cancer cells (37), sup-

porting a paracrine signaling model for EMT. These findingsindicate that both stromal and tumor Gli1 may play animportant role in tumor treatment resistance and repopula-tion and targeting them may not only enhance CAF sensitivityto genotoxic agents but also inhibit the EMTprocess in tumors.

Building on our findings, we hypothesized that the HhPmayplay a role in the HPV negative tumor microenvironment andtumor repopulation following radiotherapy. When we irradi-ated stroma and recombined them with fresh tumor cells, itresulted in significant rapid tumor regrowth after reimplan-tation compared with control or cyclopamine alone. Com-pared with the irradiated tumor recombined with fresh tumorcells, we observed a nonsignificant trend at 3 weeks and amorepronounced regrowth effect at 6 weeks. Interestingly, we didnot see substantial tumor regrowth in the irradiated stroma (ortumors) treated with cyclopamine when combined with freshtumor at either time point. Our findings are novel as theyimplicate Gli1 as a potential target in the stromal microenvi-ronment important for radiotherapy-induced HNSCC tumor(re)growth. However, additional work is needed to understandwhat may be driving this regrowth but our findings suggestthat HhP may be involved. As the tumor microenvironmentdecisively contributes to therapy resistance (38), modulatingits signaling through targeted therapy may enhance LC rates.Our findings suggest that irradiated HPV-negative HNSCCstroma (and to a lesser degree, irradiated HNSCC tumor)utilize the HhP to mediate tumor repopulation. Identifyingthe paracrine signaling molecule(s) interacting with the tumorstroma will provide valuable information regarding tumorbiology that could be applied towards novel drug discovery.There is a distinct lack of information regarding HPV-positiveHNSCC and the HhP in the literature. Given their excellentresponse to traditional chemoradiotherapy, we would hypoth-esize that theHhPmay not play as important a role but this willneed to be addressed with formal experimentation.

Our observation that GLI1- and EMT-associated genes andproteins can be similarly upregulated following acute orchronic irradiation implicates their importance as radio-responsive genes. Our transcriptome analysis of TU167 showedthat it was less epithelial compared with HN11 and had greatergene enrichment for the Hedgehog and mTOR pathways.Higher basal expression of HhP genes in TU167 may explainits relative insensitivity to cyclopamine in vitro but this wasovercome with higher doses of drug. In esophageal cancer,TNFa has been shown to activate/stabilize Gli1 via mTOR/S6K1 and not through the canonical Smoothened pathway (6).Furthermore, radiotherapy, either through EGFR phosphory-lation (39) or directly through DNA damage response proteinssuch as DNA-PKcs (40), has been shown to activate the PI3K/AKT/mTORpathway. Our pharmacologicfindings suggest thatradiotherapy-induced nGli1 translocation may signal throughthemTOR/S6K1 pathway inmoremesenchymal appearing celllines such as TU167 and may be the primary driver forradiotherapy-induced nGli1 accumulation. However, in moreepithelial cell lines such as HN11, rapamycin was unable tosuppress nGli1 accumulation yet GLI1 mRNA remained sup-pressed following pharmacologic inhibition. One possibleexplanation is that rapamycin may affect GLI1 expression

Gan et al.





irrespective of nuclear accumulation. One hypothesis is thatmesenchymal versus epithelial cells likely signal differentlyupstream but converge through mTOR/S6K1. Others havesimilarly demonstrated that the mTOR pathway is upregu-lated in head and neck cancers and may be an attractiveclinical target. One group has shown that rapamycin com-bined with an AKT inhibitor, enzastaurin, can decreasesurvival, angiogenesis, and proliferation in preclinical HNCmodels (41). Furthermore, two phase I clinical studies havebeen completed: the first in patients with resected, locallyadvanced head and neck cancer combining everolimus,weekly cisplatin, and radiation and the second, combiningeverolimus with docetaxel and cisplatin as part of an induc-tion chemotherapy regimen. Both clinical studies have dem-onstrated tolerability and safety and are presumed to beprogressing to phase II (42, 43). It is enticing to consider thatthe combination of an mTOR inhibitor, cyclopamine, andradiotherapy in patients with recurrent or locally advancedhead and neck cancer may be more efficacious as it targetsboth the tumor and stroma-mediated components involvedwith tumor repopulation.Our results show that the HhP and Gli1 play an important

role in HNSCC tumor biology. We have demonstrated that themTOR/S6K1 pathwaymay contribute to radiotherapy-inducedGli1 signaling in vitro. Radiotherapy combined with targetingHhP led to increased HNSCC cytotoxicity in vitro and in vivoand limited tumor repopulation in vivo. In locally recurrent ormetastatic head and neck tumors that may be mTOR driven,combining mTOR and HhP inhibitors with radiotherapy may

allow an opportunity to target primary and alternativemechanisms, including stroma signaling, which may lead toimproved therapeutic efficacy.

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

Authors' ContributionsConception and design: G.N. Gan, J. Eagles, X.-J. Wang, A. JimenoDevelopment of methodology: G.N. Gan, J. Eagles, S.B. Keysar, X.-J. Wang,A. JimenoAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): G.N. Gan, J. Eagles, M.J. Glogowska, C. Altunbas,R.T. Anderson, P.N. Le, J.J. Morton, B. Frederick, A. JimenoAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis):G.N. Gan, J. Eagles, S.B. Keysar, G.Wang, C. Altunbas,P.N. Le, J.J. Morton, A. JimenoWriting, review, and/or revision of the manuscript: G.N. Gan, J. Eagles,S.B. Keysar, R.T. Anderson, P.N. Le, D. Raben, X.-J. Wang, A. JimenoAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): G.N. Gan, J. Eagles, R.T. Anderson,A. JimenoStudy supervision: A. Jimeno

Grant SupportThis work was supported by 1R01CA149456 (A. Jimeno), R21DE019712 (A.

Jimeno), 2012 Cancer League of Colorado Grant (G.N. Gan and A. Jimeno), 2013ASTRO Resident/Fellow Research Seed Grant RA2013-1 (G.N. Gan and A.Jimeno), T32CA174648 (P.N. Le), NIH Cancer Center Support Grant P30CA046934, and the Paul Wehling Fund.

The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received May 7, 2014; revised September 11, 2014; accepted September 19,2014; published OnlineFirst October 8, 2014.

References1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA Cancer

J Clin 2013;63:11–30.2. Hall E, Giaccia A. Radiobiology for the Radiologist. 7th ed. Philadel-

phia, PA: Lippincott Williams and Wilkins; 2012.3. Polyak K, Weinberg RA. Transitions between epithelial and mesen-

chymal states: acquisition of malignant and stem cell traits. Nat RevCancer 2009;9:265–73.

4. Ohta H, Aoyagi K, Fukaya M, Danjoh I, Ohta A, Isohata N, et al. Crosstalk between hedgehog and epithelial-mesenchymal transition path-ways in gastric pit cells and in diffuse-type gastric cancers. Br JCancer2009;100:389–98.

5. Isohata N, Aoyagi K, Mabuchi T, Daiko H, Fukaya M, Ohta H, et al.Hedgehog and epithelial-mesenchymal transition signaling in normaland malignant epithelial cells of the esophagus. Int J Cancer2009;125:1212–21.

6. WangY,DingQ, YenCJ, XiaW, Izzo JG, Lang JY, et al. The crosstalk ofmTOR/S6K1 and Hedgehog pathways. Cancer Cell 2012;21:374–87.

7. Kawamoto A, Yokoe T, Tanaka K, Saigusa S, Toiyama Y, Yasuda H,et al. Radiation induces epithelial-mesenchymal transition in colorectalcancer cells. Oncol Rep 2012;27:51–7.

8. Kurrey NK, Jalgaonkar SP, Joglekar AV, Ghanate AD, Chaskar PD,Doiphode RY, et al. Snail and slug mediate radioresistance andchemoresistance by antagonizing p53-mediated apoptosis andacquiring a stem-like phenotype in ovarian cancer cells. Stem Cells2009;27:2059–68.

9. Tsukamoto H, Shibata K, KajiyamaH, Terauchi M, NawaA, Kikkawa F.Irradiation-induced epithelial-mesenchymal transition (EMT) related toinvasive potential in endometrial carcinoma cells. Gynecol Oncol2007;107:500–4.

10. Chung CH, Dignam JJ, Hammond ME, Klimowicz AC, Petrillo SK,Magliocco A, et al. Glioma-associated oncogene family zinc finger 1

expression and metastasis in patients with head and neck squamouscell carcinoma treated with radiation therapy (RTOG 9003). J ClinOncol 2011;29:1326–34.

11. Keysar SB, Le PN, Anderson RT, Morton JJ, Bowles DW, Paylor JJ,et al. Hedgehog signaling alters reliance onEGF receptor signaling andmediates anti-EGFR therapeutic resistance in head and neck cancer.Cancer Res 2013;73:3381–92.

12. Jimeno A, Kulesza P, Kincaid E, Bouaroud N, Chan A, Forastiere A,et al. C-fos assessment as a marker of anti-epidermal growth factorreceptor effect. Cancer Res 2006;66:2385–90.

13. Yonesaka K, Zejnullahu K, Lindeman N, Homes AJ, Jackman DM,Zhao F, et al. Autocrine production of amphiregulin predicts sensitivitytobothgefitinib andcetuximab inEGFRwild-typecancers.ClinCancerRes 2008;14:6963–73.

14. AmadorML, Oppenheimer D, Perea S, Maitra A, Cusatis G, Iacobuzio-Donahue C, et al. An epidermal growth factor receptor intron 1polymorphismmediates response to epidermal growth factor receptorinhibitors. Cancer Res 2004;64:9139–43.

15. Frederick MJ, Holton PR, Hudson M, Wang M, Clayman GL. Expres-sion of apoptosis-related genes in human head and neck squamouscell carcinomas undergoing p53-mediated programmed cell death.Clin Cancer Res 1999;5:361–9.

16. Frederick BA, Helfrich BA, Coldren CD, Zheng D, Chan D, Bunn PAJr, et al. Epithelial to mesenchymal transition predicts gefitinibresistance in cell lines of head and neck squamous cell carcinomaand non-small cell lung carcinoma. Mol Cancer Ther 2007;6:1683–91.

17. Rangan SR. A new human cell line (FaDu) from a hypopharyngealcarcinoma. Cancer 1972;29:117–21.

18. Keysar SB, Astling DP, Anderson RT, Vogler BW, Bowles DW, MortonJJ, et al. A patient tumor transplant model of squamous cell cancer






identifies PI3K inhibitors as candidate therapeutics in defined molec-ular bins. Mol Oncol 2013;7:776–90.

19. Zhao M, Sano D, Pickering CR, Jasser SA, Henderson YC, ClaymanGL, et al. Assembly and initial characterization of a panel of 85genomically validated cell lines from diverse head and neck tumorsites. Clin Cancer Res 2011;17:7248–64.

20. Huntzicker EG, Estay IS, Zhen H, Lokteva LA, Jackson PK, Oro AE.Dual degradation signals control Gli protein stability and tumor for-mation. Genes Dev 2006;20:276–81.

21. Ma J, Tian L, Cheng J, Chen Z, Xu B, Wang L, et al. Sonic hedgehogsignaling pathway supports cancer cell growth during cancer radio-therapy. PLoS ONE 2013;8:e65032.

22. Feldmann G, Dhara S, Fendrich V, Bedja D, Beaty R, Mullendore M,et al. Blockade of hedgehog signaling inhibits pancreatic cancerinvasion and metastases: a new paradigm for combination therapy insolid cancers. Cancer Res 2007;67:2187–96.

23. Sims-Mourtada J, Izzo JG, Apisarnthanarax S, Wu TT, Malhotra U,Luthra R, et al. Hedgehog: an attribute to tumor regrowth after che-moradiotherapy and a target to improve radiation response. ClinCancer Res 2006;12:6565–72.

24. Bernier J, Domenge C, Ozsahin M, Matuszewska K, Lefebvre JL,Greiner RH, et al. Postoperative irradiationwith orwithout concomitantchemotherapy for locally advanced head and neck cancer. N Engl JMed 2004;350:1945–52.

25. Cooper JS, Pajak TF, Forastiere AA, Jacobs J, Campbell BH, SaxmanSB, et al. Postoperative concurrent radiotherapy and chemotherapyfor high-risk squamous-cell carcinoma of the head and neck. N Engl JMed 2004;350:1937–44.

26. Tsai JH, Donaher JL, Murphy DA, Chau S, Yang J. Spatiotemporalregulation of epithelial-mesenchymal transition is essential for squa-mous cell carcinoma metastasis. Cancer Cell 2012;22:725–36.

27. Zeng J, Aziz K, Chettiar ST, Aftab BT, Armour M, Gajula R, et al.Hedgehog pathway inhibition radiosensitizes non-small cell lung can-cers. Int J Radiat Oncol Biol Phys 2013;86:143–9.

28. Deng Y, Zhou J, Fang L, Cai Y, Ke J, Xie X, et al. ALDH1 is anindependent prognostic factor for patientswith stages II-III rectal cancerafter receiving radiochemotherapy. Br J Cancer 2014;110:430–4.

29. Huang CF, Xu XR, Wu TF, Sun ZJ, Zhang WF. Correlation of ALDH1,CD44, OCT4 and SOX2 in tongue squamous cell carcinoma and theirassociation with disease progression and prognosis. J Oral PatholMed 2014;43:492–8.

30. Kim YH, Kim G, Kwon CI, Kim JW, Park PW, Hahm KB. TWIST1 andSNAI1 as markers of poor prognosis in human colorectal cancer areassociated with the expression of ALDH1 and TGF-beta1. Oncol Rep2014;31:1380–8.

31. Fingas CD, Bronk SF, Werneburg NW, Mott JL, Guicciardi ME, Caza-nave SC, et al. Myofibroblast-derived PDGF-BB promotes Hedgehog

survival signaling in cholangiocarcinoma cells. Hepatology 2011;54:2076–88.

32. Harris LG, Pannell LK, Singh S, Samant RS, Shevde LA. Increasedvascularity and spontaneous metastasis of breast cancer by hedge-hog signaling mediated upregulation of cyr61. Oncogene 2012;31:3370–80.

33. Mills LD, Zhang Y, Marler RJ, Herreros-VillanuevaM, Zhang L, AlmadaLL, et al. Loss of the transcription factor GLI1 identifies a signalingnetwork in the tumor microenvironment mediating KRAS-inducedtransformation. J Biol Chem 2013;288:11786–94.

34. Levina E, Chen M, Carkner R, Shtutman M, Buttyan R. ParacrineHedgehog increases the steroidogenic potential of prostate stromalcells in a Gli-dependent manner. Prostate 2012;72:817–24.

35. Walter K, Omura N, Hong SM, Griffith M, Vincent A, Borges M, et al.Overexpression of smoothened activates the sonic hedgehog signal-ing pathway in pancreatic cancer-associated fibroblasts. Clin CancerRes 2010;16:1781–9.

36. Choe C, Shin YS, Kim SH, Jeon MJ, Choi SJ, Lee J, et al. Tumor-stromal interactions with direct cell contacts enhance motility of non-small cell lung cancer cells through the hedgehog signaling pathway.Anticancer Res 2013;33:3715–23.

37. Hwang RF, Moore TT, Hattersley MM, Scarpitti M, Yang B, Dever-eaux E, et al. Inhibition of the hedgehog pathway targets the tumor-associated stroma in pancreatic cancer. Mol Cancer Res 2012;10:1147–57.

38. Sun Y, Campisi J, Higano C, Beer TM, Porter P, Coleman I, et al.Treatment-induced damage to the tumor microenvironment promotesprostate cancer therapy resistance through WNT16B. Nat Med2012;18:1359–68.

39. Pickhard AC, Margraf J, Knopf A, Stark T, Piontek G, Beck C, et al.Inhibition of radiation induced migration of human head and necksquamous cell carcinoma cells by blocking of EGF receptor pathways.BMC Cancer 2011;11:388.

40. Boehme KA, Kulikov R, Blattner C. p53 stabilization in response toDNAdamage requiresAkt/PKBandDNA-PK. ProcNatl Acad Sci USA2008;105:7785–90.

41. Liu J, Kuo WL, Seiwert TY, Lingen M, Ciaccio MF, Jones RB, et al.Effect of complementary pathwayblockade onefficacyof combinationenzastaurin and rapamycin. Head Neck 2011;33:1774–82.

42. Fury MG, Lee NY, Sherman E, Ho AL, Rao S, Heguy A, et al. A phase 1study of everolimusþweekly cisplatinþ intensity modulated radiationtherapy in head-and-neck cancer. Int J Radiat Oncol Biol Phys2013;87:479–86.

43. FuryMG, Sherman E, HoAL, XiaoH, Tsai F, NwankwoO, et al. A phase1 study of everolimus plus docetaxel plus cisplatin as inductionchemotherapy for patients with locally and/or regionally advancedhead and neck cancer. Cancer 2013;119:1823–31.


Gan et al.




2014;74:7024-7036. Published OnlineFirst October 8, 2014.Cancer Res Gregory N. Gan, Justin Eagles, Stephen B. Keysar, et al. Tumor Repopulation in Head and Neck Squamous CancersHedgehog Signaling Drives Radioresistance and Stroma-Driven

Updated version

10.1158/0008-5472.CAN-14-1346doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2014/10/09/0008-5472.CAN-14-1346.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/74/23/7024.full#ref-list-1

This article cites 42 articles, 16 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/74/23/7024.full#related-urls

This article has been cited by 1 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://cancerres.aacrjournals.org/content/74/23/7024To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-14-1346

http://cancerres.aacrjournals.org/content/suppl/2014/10/09/0008-5472.CAN-14-1346.DC1

http://cancerres.aacrjournals.org/content/74/23/7024.full#ref-list-1

http://cancerres.aacrjournals.org/content/74/23/7024.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/74/23/7024


hedgehog signaling drives radioresistance and stroma...

Documents