Ormeloxifene Suppresses Desmoplasia and Enhances...

Therapeutics, Targets, and Chemical Biology

Ormeloxifene Suppresses Desmoplasia andEnhances Sensitivity of Gemcitabine inPancreatic CancerSheema Khan1, Mara C. Ebeling2, Neeraj Chauhan1, Paul A. Thompson3, Rishi K. Gara1,Aditya Ganju1, Murali M. Yallapu1, Stephen W. Behrman4, Haotian Zhao2, Nadeem Zafar5,Man Mohan Singh6, Meena Jaggi1, and Subhash C. Chauhan1

Abstract

The management of pancreatic ductal adenocarcinoma(PDAC) is extremely poor due to lack of an efficient therapyand development of chemoresistance to the current standardtherapy, gemcitabine. Recent studies implicate the intimatereciprocal interactions between epithelia and underlying stro-ma due to paracrine Sonic hedgehog (SHH) signaling in pro-ducing desmoplasia and chemoresistance in PDAC. Herein, wereport for the first time that a nonsteroidal drug, ormeloxifene,has potent anticancer properties and depletes tumor-associatedstromal tissue by inhibiting the SHH signaling pathway inPDAC. We found that ormeloxifene inhibited cell proliferationand induced death in PDAC cells, which provoked us toinvestigate the combinatorial effects of ormeloxifene withgemcitabine at the molecular level. Ormeloxifene causedpotent inhibition of the SHH signaling pathway via down-regulation of SHH and its related important downstream tar-

gets such as Gli-1, SMO, PTCH1/2, NF-kB, p-AKT, and cyclinD1. Ormeloxifene potentiated the antitumorigenic effect ofgemcitabine by 75% in PDAC xenograft mice. Furthermore,ormeloxifene depleted tumor-associated stroma in xenografttumor tissues by inhibiting the SHH cellular signaling pathwayand mouse/human collagen I expression. Xenograft tumorstreated with ormeloxifene in combination with gemcitabinerestored the tumor-suppressor miR-132 and inhibited stromalcell infiltration into the tumor tissues. In addition, invasivenessof tumor cells cocultivated with TGFb-stimulated human pan-creatic stromal cells was effectively inhibited by ormeloxifenetreatment alone or in combination with gemcitabine. Wepropose that ormeloxifene has high therapeutic index and ina combination therapy with gemcitabine, it possesses greatpromise as a treatment of choice for PDAC/pancreatic cancer.Cancer Res; 75(11); 1–13. �2015 AACR.

IntroductionPancreatic ductal adenocarcinoma (PDAC) has a poor prog-

nosis largely due to its propensity for early local invasion, distantmetastasis, and lack of effective therapies. Many chemotherapeu-tic regimens have failed and the current standard-of-care therapy,gemcitabine, extends patient survival by only a few months (1).Newer treatment options for PDACpatients are FOLFIRINOX andnab-paclitaxel/gemcitabine, which improved overall survival by4.3 and 1.8 months over gemcitabine therapy, respectively (2, 3).

However, safety profile of these drugs is less favorable thangemcitabine therapy, accounting for myelosuppression andperipheral neuropathy (2–4). Despite these advances, the overalloutcome remains miserable for this patient population. Thus,investigations on alternative approaches for PDAC therapy are ahigh research priority.

Activation of oncogenes, such as Kras, and/or inactivatingmutations or loss of expression of tumor-suppressor genes(including DPC4, p16, p53, and SMAD4) is known in PDAC(5). It has been shown that extensive desmoplasia is one of theunderlying causes of pancreatic cancer's poor prognosis andchemoresistance (6). Desmoplasia is typically characterized byexcessive production of extracellular matrix (ECM) and collagen Iand is associated with proliferation of stromal cells, myofibro-blasts, and pancreatic stellate cells. A profound role of Sonichedgehog (SHH) pathway is implicated in desmoplasia (7) andcancer progression (8), including PDAC (9). This developmentalpathway, dormant in the adult pancreas, becomes reactivatedearly in PDAC development (10). SHH is a member of theHedgehog (Hh) family of secreted signaling proteins, havingdiverse functions during vertebrate development (11). In pancre-atic tumors, intimate reciprocal interactions occur between epi-thelia and underlying stroma due to paracrine Hh signaling thatlead to desmoplasia and form a barrier to chemotherapy drug(s)penetration (12). Depletion of tumor stroma leads to the increas-ing functional vasculature that provides a feasible avenue forefficient therapeutic drug delivery (13). In addition, Hh signaling

1Department of Pharmaceutical Sciences and Center for CancerResearch, University of Tennessee Health Science Center, Memphis,Tennessee. 2Cancer Biology and Sanford Children's Health ResearchCenter, Sanford Research, Sioux Falls, South Dakota. 3Methodologyand Data Analysis Center, Sanford Research, Sioux Falls, South Dako-ta. 4Department of Surgery, University of Tennessee Health ScienceCenter, Memphis, Tennessee. 5Department of Pathology, University ofTennessee at Memphis, Memphis, Tennessee. 6Saraswati Dental Col-lege, Lucknow, Uttar Pradesh, India.

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

Corresponding Author: Subhash C. Chauhan, Department of PharmaceuticalSciences, University of Tennessee Health Science Center, 19 South Manassas,Cancer Research Building, Memphis, TN 38163. Phone: 901-448-2175; Fax: 901-448-1051; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-14-2397

�2015 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org OF1

Research. on February 5, 2020. © 2015 American Association for Cancercancerres.aacrjournals.org Downloaded from

Published OnlineFirst April 3, 2015; DOI: 10.1158/0008-5472.CAN-14-2397





http://cancerres.aacrjournals.org/



plays a key role in themaintenance of pancreatic cancer stem cells(CSC) that are involved in drug resistance, cancer recurrence, andpoor clinical outcome (10). Therefore,molecular and/or chemicalintervention to target Hh signaling and disruption in the micro-environment in tumors could be an interesting therapeuticapproach for pancreatic cancer (13). Some of the well-knownHh signaling antagonists, such as vismodegib (GDC-0449), havebeen investigated alone or as an adjuvant to the traditionalanticancer drugs but havenot yielded clinicallymeaningful results(14, 15) and have shown notable adverse effects including tera-togenic properties (16, 17). Thus, identification of novel therapieswith high therapeutic index that can target Hh and tumor pro-gression signaling pathways with no or minimal adverse effects isrequired.

Repurposing of established drugs as anticancer agents is acurrent active investigative approach. Ormeloxifene is a non-hormonal, nonsteroidal oral contraceptive molecule (18).Recent studies suggested that ormeloxifene may be effective ininhibiting breast cancer, head and neck cancer, and chronicmyeloid leukemia cells (18). Moreover, ormeloxifene isreported to have an excellent therapeutic index and is safe forchronic administration (19). This study demonstrates theinhibitory role of ormeloxifene on the SHH signaling pathway,and describes inhibitory patterns of this drug on pancreatictumor progression using bidirectional tumor stromal interac-tions. This inhibitory effect was either more pronounced orcomparable with a known smoothened (SMO) inhibitor, GDC-0449, in PDAC cells. Ormeloxifene disrupts the stroma offibrotic pancreatic tumors and inhibits the proliferating stellateand myeloid cells involved in the development of pancreaticfibrosis. Furthermore, the combinatorial effects of ormeloxi-fene with gemcitabine induce increased gemcitabine sensitivity.In addition, these studies also suggest wide use of ormeloxifenein PDAC patients due to its intended safe use in fertile women,considering the teratogenic potential of other Hh pathwayinhibitors such as cyclopamine and GDC-0449 (20, 21).

Materials and MethodsCell culture, growth conditions, and treatments

Cell lines were purchased from the ATCC and were maintainedat 37�C/5% CO2 in recommended growthmediumwith 10% FBS(RPMI, DMEM and DMEM/Ham's F12; Hyclone Laboratories).Human CSCs (CD133þ/CD44þ/CD24þ/ESAþ) were obtainedfrom Celprogen Inc. They were isolated from primary tumors andhave been described previously (22). All cell lines weremaintainedin continuous exponential growth by twice a week passaging incell type-specific media. Ormeloxifene was generously synthesizedand provided by Fathi Halaweish (South Dakota State University,Brookings, SD) as described earlier (23). Gemcitabine was pur-chased from Sigma-Aldrich (cat. no. G6423) and GDC-0449 fromSellekchem (cat. no. S1082). Cells were treated with indicateddoses of ormeloxifene, gemcitabine, and GDC-0449 after comp-letely solubilized in ethanol, PBS, and DMSO, respectively.

Cell proliferation by MTS assayThe antiproliferative effect of ormeloxifenewasdeterminedafter

48 hours using the CellTiter 96 AQueous One solution assay (cat.no.G5421; Promega) using amicroplate reader (BioMate 3UV-Visspectrophotometer; Thermo Electron Corporation). Ethanol- orPBS-containingmedium served as the vehicle control. In addition,

the antiproliferative effect of ormeloxifene was determined at 24and 48 hours using Cell Counting Kit-8 (Mayflower Bioscience)and the percentage viability of Panc-1 and BxPC-3 cells was deter-mined after treatment with GDC-0449 and ormeloxifene. Theantiproliferative effect of each treatment was calculated as a per-centage of cell growth with respect to the vehicle control.

Cell proliferation by xCELLigence assayPDAC cells (10,000 cells/well) were seeded in E-plate (Roche)

following the xCELLigence Real Time Cell Analyzer (RTCA) DPinstrument manual as provided by the manufacturer (Roche;ref. 24). After 24 hours, ormeloxifene or the vehicle control wasadded and the experiment was allowed to run for 100 hours.Average baseline cell index for ormeloxifene-treated cells com-pared with control cells was calculated for at least two measure-ments from three replicated experiments.

Flow cytometric analysis of apoptosis and necrosisBxPC-3 and Panc-1 cells (1 � 106) were treated for 24 hours

with ormeloxifene (15 mmol/L) and gemcitabine (100 nmol/L)alone and in combination. Cells were stained with Annexin V–FITC and propidium iodide (PI). The apoptotic and necroticpopulations were detected as described earlier (25). Cells werescanned in FL-1 (FITC) versus FL-2 (PI) channels and analyzedusing an Accuri C6 flow cytometer (Accuri Cytometers, Inc.).

Cell-cycle analysisCells were exposed to ormeloxifene (15 mmol/L) and gemci-

tabine (100 nmol/L) alone or in combination for 24 hoursand stained with Telford Reagent containing PI (cat. no. P-4170;Sigma-Aldrich). Cells were analyzed with an Accuri C6 flow cyto-meter. Cellswith hypodiploidDNA(content less thanG0–G1)weredeemed apoptotic (sub-G0–G1).

Dual-Luciferase reporter assayDual-Luciferase reporter assay was carried out to investigate the

effect of treatments on Gli-1 and NF-kB transcriptional activityusing a luciferase assay kit (cat. no. E2940; Promega) according tothe manufacturer's protocol. BxPC-3 and Panc-1 cells were trans-fected with luciferase reporter constructs [NF-kB, gift from Dr.Ajay Singh, Mitchell Cancer Institute, Fairhope, AL, Cignal GLIReporter (luc) Kit, cat. no. CCS-6030L; Qiagen] and treated withormeloxifene and gemcitabine alone or in combination for 24hours. The normalized luciferase activity was expressed as a ratioof firefly luciferase to Renilla luciferase units.

Indirect coculture of PDAC cells and pancreatic stromal cellsHuman pancreatic stromal cell (PSC) fibroblasts and stellate

cells were attained from an islet transplant program and main-tained in CMRL-1066 medium (cat. no. 15110; Corning) supple-mented with 10% FBS, penicillin sodium, and streptomycinsulfate at 37�C in humidified atmosphere containing 5% CO2.Human PSCs (3 � 106 cells/culture insert) were seeded into theculture inserts of 1.0-mm pore size (BD Biosciences) in CMRL-1066 media. On day 2, the culture inserts were placed into 6-wellplates containing Panc-1 cells (0.8 � 106 cells/well), followed bytreatment with ormeloxifene (10 mmol/L) and gemcitabine (100nmol/L) and incubated up to 2 days in DMEM medium. Asprevious studies have shown TGFb to be a potent inducer ofepithelial–mesenchymal transition (EMT) in several cancer cellsincluding pancreatic cancer cells (26, 27), we used recombinant

Khan et al.

Cancer Res; 75(11) June 1, 2015 Cancer ResearchOF2




TGFb (2 ng/mL) to stimulate the stromal cells as a mediator ofPSC-induced EMT in cells.

Clonogenic assayFor the clonogenic assay, 500 cells were treated with indicated

concentrations of ormeloxifene for 12 days. The visible colonies(� 50 cells) were counted following hematoxylin staining (FisherScientific) and the percentage of colonies was calculated as com-pared with control, as described earlier (28).

Cell motility, migration, and invasion assaysCellmotility was analyzedwith a Boyden's chamber assay (28).

For cell invasion assays, BD Biocoat Matrigel Invasion Chambers(BDBiosciences)were used as per themanufacturer's instructions.After 48 hours of incubation, the invading cells were stained andcounted in 10 fields of view. In addition, a wound-healingmigration assay was also used to evaluate the effect of ormelox-ifene on the migratory ability of cancer cells. The cell monolayerwas scraped using a micropipette tip and 48 to 72 hours aftertreatment, the residual gap length was calculated from photo-micrographs. To further confirm these findings, real-time migra-tion andproliferationwereperformedby the xCELLigence system,which is an electrical impedance-basedmethod that allows for themeasurement of cell migration and proliferation in real-time(24). Briefly, 4 � 104 cells were seeded per chamber of cellinvasion and migration (CIM) plate and the cells were analyzedin xCELLigence instrument at 37�C, 5% CO2 for migration andinvasion assays.

Immunoblot analysisHuman PDAC cells (1 � 106) were treated with ormeloxifene

(15 mmol/L) and gemcitabine (100 nmol/L) alone and in com-bination for 24 hours. Total cell lysates were prepared followedby immunoblotting for various indicated proteins as describedearlier (25).

Reverse transcription–quantitative real-time PCRTotal RNA was extracted using TRIzol reagent (cat. no. AM

9738; Invitrogen) and integrity was checked with an RNA 6000Nano Assay kit and 2100 Bioanalyzer (Agilent Technologies). ThemRNA expression levels were determined by quantitative real-time PCR (qRT-PCR) using TaqMan PCR master mixture andTaqMan-specific probes (Applied Biosystems). The expression ofgenes was normalized to the 18S rRNA gene.

Tumorsphere assayPancreatic CSCs were plated on ultra-low attachment plates

(Corning) at a density of 1 � 103/100 mL well/96-well plate andtreated with ormeloxifene or GDC-0449 (2.5–10 mmol/L). Theplates were allowed to grow for 7 days in 0.5% serum-freemedium (Cellprogen) to form primary spheres. Following theincubation, the primary spheres were dissociated into single-cellsuspension and plated at a density of 1 � 104/2 mL/6-well ultra-low attachment plate. Secondary spheres were counted after 7 to10 days in culture.

In vivo tumor xenograft modelSix-week-old female athymic nude (nu/nu) mice (22–25 mg)

were purchased from Charles River Laboratories International,Inc., and maintained in a pathogen-free environment. The mice

were injected with BxPC-3 cells intraperitoneally (i.p.; 3 � 106)and (5 � 106) cells in 200 mL PBS/Matrigel subcutaneously. Onday15, themicewere treatedwith vehicle (ethanol), ormeloxifene(200 mg), gemcitabine (500 mg), or their combination via i.p.injections, three times a week, for 6 subsequent weeks. Mice wereweighed twice a week to monitor their health and tumor growth.Tumor volume (V) was estimated from the length (l), width (w),andheight (h) of the tumor using the formulaV¼ 1/4 0.52 (l�w�h), as described previously (Supplementary Fig. S4; ref. 28). Forty-five days after the first drug injection, mice were euthanized andtumor burden (wet weight) and metastases were noted. Theorgans, including pancreas, were harvested and checked formetastases. The data were modeled with time (discrete), group(control, ormeloxifene, gemcitabine, and ORMþGEM), and theinteraction between them. Primary analyses involved plannedcomparisons (separately for each time point) between controland ormeloxifene/gemcitabine versus ORMþGEM. Animal carewas performed in accordance with institutional guidelines andall animal experiments were carried out using protocols approv-ed by the Sanford Research Institutional Animal Care and UseCommittee.

In situ hybridization for microRNA detection and expressionWe detected the expression of miR-132 in formalin-fixed par-

affin-embedded (FFPE) tissues of control and treated xenograftmice. We used an in situ hybridization technique and used aBiochain kit (cat. no. K2191050; Biochain IsHyb In Situ hybrid-ization kit) as previously described (29). Briefly, tissues werehybridized with hybridization buffer and digoxigenin-labeledprobe (EXIQON) at 45�C overnight followed by incubation withthe AP-conjugated anti-digoxingenin antibody and NBT/BCIP(Pierce) and nuclear fast red counterstaining.

Immunofluorescence and immunohistochemical analysesImmunofluorescence and IHC analyses were used to analyze

the untreated and treated xenograft tumor tissues to detectchanges associated with the expression of important proteinsinvolved in tumor–stromal interactions as described previously(30). The slides were stained with specific antibodies followingheat-induced antigen retrieval techniques and imaged using alaser scanning confocal microscope (Nikon TIRF) with a �200Apochromat objective for immunofluorescence. For immunohis-tochemistry, the slides were stained using Biocare's MACH4Universal HRP-Polymer kit (Biocare Medical) and analyzed aspreviously described (29, 30).

Statistical analysesStatistical significance of the studies was analyzed by the

Student t test. Differences with P values of < 0.05 are consideredsignificant. Tumor size values were examined at the day 50 point,using an ANOVA approach. Tests of main effects (differencesbetween treatments) and contrasts were performed.

ResultsOrmeloxifene treatment suppresses tumorigenic features ofPDAC cells

Ormeloxifene was found to have an anticancer effect on alltested PDAC cells (Fig. 1A and Supplementary Fig. S1A). Toconfirm these results, we measured the growth in real time for

Ormeloxifene Suppresses Desmoplasia in Pancreatic Cancer

www.aacrjournals.org Cancer Res; 75(11) June 1, 2015 OF3




duration of 100 hours using the xCELLigence system (Fig. 1B).This assaymonitors cell growth in real time bymeasuring changesin electric impedance between two golden electrodes embeddedin the bottom of the cell culture wells. The impedance, which isconverted to a cell index value, is directly proportional to thenumber of cells and also reflects the cells' viability, morphology,and adhesion strength (31). The growth curve, which is presentedas a baseline cell index, showed that ormeloxifene significantlyreduced the baseline cell index compared with the control cells(Fig. 1B). Furthermore, ormeloxifene treatment inhibited theclonogenic potential of PDAC cells (BxPC-3, Panc-1, AsPC-1,MiaPaca, and HPAF-II) as evident by the decreased number ofcolonies after ormeloxifene treatment (Fig. 1C). Moreover, orme-loxifene was also found to inhibit cellular motility (Fig. 2A) andinvasion (Fig. 2B) of PDAC cells. The inhibition of the migratoryability of cells is demonstrated by wound-healing assay (Supple-mentary Fig. S1B) and cellular invasion byMatrigel invasion assay(Supplementary Fig. S1C), whichwas further confirmed using thexCELLigence method (Fig. 2C).

In addition, we sought to compare the anticancer potential oformeloxifene with a known SMO inhibitor (GDC-0449) inhuman PDAC cells. Ormeloxifene showed more pronounced orcomparable inhibitory effect on cell proliferation, clonogenicity,

and invasion than GDC-0449 at equal indicated concentrations(Supplementary Fig. S2A–S2C). Inhibition of cell viability andinvasion was observed within 48 hours following exposure tothese drugs.

Ormeloxifene treatment inhibits tumorsphere formation inpancreatic stem cells

We observed a significant effect of ormeloxifene on tumor-sphere formation in CSCs as reflected by a reduction in size andnumber of tumorspheres in cells upon treatment, suggesting theclonogenic depletion of the CSCs. In contrast, GDC-0449 did notshow a significant effect on secondary tumorsphere formation(Supplementary Fig. S2D).

Ormeloxifene inhibits SHH signaling in PDAC cellsThe SHH signaling pathway has been implicated in the devel-

opment of pancreatic cancer (9). Therefore, PDAC cells weretreated with ormeloxifene and changes in the SHH signalingpathway were evaluated by Western blot and qRT-PCR analyses.Ormeloxifene treatment effectively inhibited SHH expression atprotein andmRNA levels at indicated concentrations (Fig. 2D andSupplementary Fig. S3A and S3B). Ormeloxifene treatment alsoinhibited the expression of Gli-1, SMO, cyclin D1, and p-AKT, the

120

100

80

60

40

20

00 5 10 15 20 25

Concentration (μmol/L)

% C

ell p

rolif

erat

ion

% C

lono

geni

c ef

ficie

ncy

AsPC-1

AsPC-1

Colo-357

Panc-1

Panc-1

CFPAC

BxPC-3

BxPC-3

HPAF-II

HPAF-II

SW-1990

Panc 02.03

Capan-1

MiaPaca

MiaPac

15

10

5

00 20 40 60 80 100 0 20 40 60 80 100

2

1.5

1

0.5

0

6

4

2

0

4

3

2

1

00 20 40 60 80 100 0 20 40 60 80 100

Time (h)

Nor

mal

ized

cel

l ind

ex

AsPC-1 Panc-1

HPAF-IIEtOH10 μmol/L20 μmol/L30 μmol/L

BxPC-3

ON

OH3COCH3

CH3

1209060300

1209060300

1209060300

1209060300

1209060300

CT 2.5 μmol/L 5 μmol/L 10 μmol/L





A

B

C

a

Figure 1.Determination of proliferation,clonogenicity, and cytotoxicityprofiles of ormeloxifene in PDAC cells.A, structure of ormeloxifene (IUPACname: 1-[2-[4-[(3S,4R)-7-methoxy-2,2-dimethyl-3-phenyl-chroman-4-yl] phenoxy] ethyl] pyrrolidine) andits effect on cell growth wasmonitored by the MTS assay for 48hours and is shown as percentage. B,effect of ormeloxifene on cellproliferation with respect to time wasalso confirmed by xCELLigence RTCA.C, clonogenicity assay was performedtodetermine the ability of cells to formcolonies (percentage inhibition)following treatment. Cells werephotographed and counted usingAlphaEaseFC (Alpha ImagerHP AIC)software analysis tool. Bars representmean � SD; n ¼ 3; � , P < 0.05;��, P < 0.001.

Khan et al.





key downstream proteins that drive the oncogenic signaling ofSHH pathway in BxPC-3 and MiaPaca cells (Fig. 2D and Supple-mentary Fig. S3A; ref. 12). Ormeloxifene treatment also increasedthe expressionof tumor-suppressor SUFU,which interacts directlywith Gli-1 proteins to repress SHH signaling (Fig. 2D and Sup-plementary Fig. S3A; ref. 32).

Importantly, ormeloxifene treatment caused amarked (�70%)decrease in the expression of the SHH transcription factor, NF-kB-65 (33), and its downstream target, cyclin D1 (34), within 24hours (Fig. 2D and Supplementary Fig. S3C). Cyclin D1 is animportant mediator of SHH-induced cell proliferation and car-

cinogenesis. These data suggest that ormeloxifene treatmenteffectively inhibits tumorigenic phenotypes via modulation ofSHH and its downstream signaling molecules.

Ormeloxifene and gemcitabine in combination induceapoptosis in PDAC cells

We investigated whether ormeloxifene treatment enhanced theapoptotic index in gemcitabine-resistant PDAC cells (Panc-1 andBxPC-3). Our data show that when combined, ormeloxifene (15mmol/L), and gemcitabine (100 nmol/L) induced a significantlyhigher (21%) apoptotic population in 24hours as comparedwith

CT 10 μmol/L 20 μmol/L






AsP

C-1

BxP

C-3

AsP

C-1

Cel

ls/F

.O.V

.C

ells

/F.O

.V.

BxP

C-3

300

200

100

0

300

200

100

0

60

45

30

15

0

60

45

30

15

0

BxPC-3

BxPC-3

AsPC-1

AsPC-1

Cell migration

Cell invasion

EtOH

10 μmol/L

20 μmol/L

0.6

0.5

0.4

0.3

0.2

0.1

0

0 4 8 12 16 20 24

2

1.6

1.2

0.8

0.4

00 5 10 15 20 25

Time (h)

ORM BxPC-3

μmol/L CT 5 10 15 20

SHH

Cyclin D1

P-AKT

Caspase-3

IKB-α

NFκB-65

SUFU

Gli-1

β-Actin

Cel

l ind

ex

Cells: P

anc-1A

B D

C

Figure 2.Ormeloxifene targets SHH signalingpathway and inhibits PDACCIM. A andB, effect of ormeloxifene on cellinvasion through Matrigel invasionassay (A) and cell migration abilitythrough migration assay (B). Cellswerephotographedat amagnificationof �200 and counted using animaging system. Bars represent mean� SD; n ¼ 3; �, P < 0.0001. C, effect oformeloxifene on CIM ability wasconfirmed using xCELLigence system.D, effect of ormeloxifene on SHH andits associated protein targets wasdetermined by Western blotting.b-Actin was used as an internalcontrol. Data are representative ofthree individual experiments. F.O.V.,field of view.






ormeloxifene and gemcitabine treatment alone (Fig. 3A). How-ever, PI-positive postapoptotic/necrotic cell population was rel-atively small, suggesting that the induced cytotoxicity was pre-

dominantly through activation of apoptotic pathways. These datasuggest that ormeloxifene-alone induced cell death does notinvolve the release of phosphatidylserine onto the outer leaflet,

5.00

4.00

3.00

2.00

1.00

0.00

Panc-1

BxPC-3

Fol

d ch

ange

Fol

d ch

ange

Rel

ativ

e fo

ld c

hang

e

10.00

8.00

6.00

4.00

2.00

0.00

CT

CT

OR

MO

RM

GE

MG

EM

OR

M +

GE

MO

RM

+ G

EM

SHH

Panc-1

Panc-1

Panc-1

Panc-1

BxPC-3

BxPC-3

BxPC-3

BxPC-3

BxPC-3

SMO

PTCH1

PTCH2

2.00

0.00

-2.00

-4.00

-6.00

5.00

0.00

-5.00

-10.00

-15.00

10.00

0.00

-10.00

-20.00

-30.00

43210

-1-2

CT

OR

M

OR

M

GE

M

+G

EM

CT

CT

ORM

ORM

ORMGEM+ GEM

SHH

Gli-1

Bcl-xL

SMO

b-Actin

G0–G1 G2–MS

CT ORM GEM ORM+GEM

GEM

GEM

ORM+GEM

ORM+GEM

CT ORM

FL2-A

A

B

D

CF

I3-A

Cells:

Cells:

Panc-1Percent population in

Percent population in

each phase

each phase

Treatment

Treatment

CTORMGEMORM+GEM

BxPC-3

CTORMGEMORM+GEM

G0–G1

G0–G1

S

S

G2–M

G2–M

54.59

36.29 20.98 27.66

74.38

55.14

73.76

19.05

5.87

18.86

5.59

21.68

15.17

21.75

15.79

51.10 15.19 15.02

12.97 32.25 41.55

38.72 29.76 8.92

Figure 3.Ormeloxifene in combination with gemcitabine induces apoptosis and modulates key targets of SHH pathway. A and B, flow cytometric analysis of AnnexinV–positive cells and cells in G0–G1 stage after treatment. Bars represent mean � SD; n ¼ 3; � , P < 0.01; �� , P < 0.001; �� , P < 0.0001 as compared withCT. C, relative fold change in the mRNA levels of key molecules involved in SHH pathway by qRT-PCR. Bars represent mean � SD; n ¼ 3; �, P < 0.01;�� , P < 0.001; ��, P < 0.0001. D, Western blot analysis indicating the effect of ormeloxifene (15 mmol/L), gemcitabine (100 nmol/L), and their combinationon the important proteins in SHH pathway. Data are representative of one of three similar experiments. ORM, ormeloxifene; GEM, gemcitabine.

Khan et al.





indicative of Annexin V–positive apoptotic cells and mitochon-drial apoptotic signaling. Instead, it may involve death receptor–mediated extrinsic apoptotic signaling. Therefore, we sought toinvestigate the effect of ormeloxifene on cell-cycle phase distri-bution. Typically, D-type cyclins are required for the progressionof cells from the G1-phase of the cell cycle to S-phase (35).Ormeloxifene treatment decreased the expression of cyclin D inBxPC-3 and MiaPaca cells (Fig. 2D and Supplementary Fig. S3A).Ormeloxifene treatment led to cell-cycle arrest at sub-G0–G1

phase in Panc-1 and BxPC-3 cells. Cells in sub-G1-phase increasedup to 74% after ormeloxifene treatment (15 mmol/L), while cellsin the S-phase decreased from 19% to 5%. However, gemcitabinetreatment did not show an additional effect on cell-cycle phases(Fig. 3B). Similar effects in cell-cycle phase distribution wereobserved in BxPC-3 cells, which showed significant inhibition ofthe G2–M phase upon treatment with ormeloxifene and gemci-tabine in combination.

Ormeloxifene and gemcitabine combination targets SHHsignaling pathway and inhibits CIM in PDAC cells

In addition, we investigated combinatorial effects of ormelox-ifene and gemcitabine on SHH and downstream signaling mole-cules. Treatment with ormeloxifene and gemcitabine has relative-ly more pronounced inhibitory effects on the expression of SHH,Gli-1, and SMO as compared with ormeloxifene or gemcitabinealone (Fig. 3D). This reveals the potentiated effects of ormelox-ifene in combination with gemcitabine. We also confirmed theseresults by qRT-PCR analysis and observed an apparent decrease inthe mRNA levels of main effectors of the SHH signaling pathwayin response to ormeloxifene alone or in combination with gem-citabine. This included decreased expression of SHH (4-fold),SMO (5-fold), and patched 1/2 (PTCH1/2) compared with thecontrol (Fig. 3C). Ormeloxifene alone or in combination withgemcitabine also showed a marked (�40%) decrease in the levelof antiapoptotic, Bcl-xL protein (Fig. 3D). The Bcl-xL protein isalso an important mediator of SHH and is transcriptionallyregulated by SHH through the Gli-1 transcription factor (34). Inaddition, ormeloxifene alone or in combination with gemcita-bine inhibited the Gli-1 and NF-kB-65 transcriptional activity inPDAC cells (Fig. 4A and Supplementary Fig. S3C). These resultspresent first evidence that ormeloxifene inhibits the SHH–Gli-1signaling pathway in PDAC.

Moreover, we evaluated the ability of ormeloxifene and gem-citabine to inhibit tumor progression and found that ormelox-ifene inhibited motility (Fig. 4B) and the migratory ability ofPDAC cells as demonstrated by wound healing (Fig. 4C).

Ormeloxifene and gemcitabine combination efficientlyabrogates TGFb-induced SHH signaling

The interactions among the stromal and tumor cells and thevarious cytokines embedded in the ECM contribute to the neo-plastic phenotype (36). In addition to the activated tumor–stromal myofibroblasts (characterized by the expression of con-tractile genes such as smooth-muscle actin, aSMA; ref. 37), theactivated PSCs that are characterized by expression of the stellatecell activation-associated protein (cygb/STAP) are identified as themajor source of the excessive stromal ECM production in pan-creatic tumors (6). Here, we show for the first time that indirectcoculture of PDAC cells with PSCs (stimulated with TGFb)induced enhanced secretion of SHH and chemokine CXCL12(stromal cell–derived factor-1, SDF1) that was abrogated by

ormeloxifene alone or in combination with gemcitabine. Gemci-tabine treatment alone did not show any effect as observedthrough ELISA of the conditioned media in which the PDACcells were cultured (Fig. 4D). CXCL12 is abundantly produced bythe stromal cells that induce SHH expression, which promotesprogression, metastasis, and chemoresistance of PDAC cells (38).In addition, treatment with ormeloxifene alone or in combina-tion inhibited the proliferation of PSCs as depicted by thedecreased expression of aSMA and cygb/STAP in the immuno-fluorescence of PSCs (Fig. 4D, bottom). These results suggest thatormeloxifene not only reduces the number of stromal cellsinvolved in the development of pancreatic fibrosis but alsoinhibits the paracrine SHH signaling between cancer and stromalcells that leads to desmoplasia and causes chemoresistance.

Combined ormeloxifene and gemcitabine treatment effectivelyinhibits tumor burden in mice model

To investigate the anticancer effects of ormeloxifene, we used asubcutaneous (for solid tumor) and i.p. (metastatic) preclinicalmurine xenograft model generated with gemcitabine-resistantBxPC-3 cells. Both ormeloxifene and gemcitabine, administeredalone, inhibited overall tumor burden, but combination treat-ment of the two was more efficacious than either of them alone(Fig. 5A and B). When compared with the control mice, micetreated with ormeloxifene (P ¼ 0.0301) or gemcitabine (P ¼0.0009) or ORMþGEM in combination (P < 0.0001) showed amarked reduction in tumor weight (Fig. 5B). Moreover, in the i.p.model, tumors barely developed in the ORMþGEM–treatedgroup. Upon further examination, we also found there were feweror nometastases in themice treatedwith ormeloxifene alone or incombination with gemcitabine (Fig. 5B, inset table). miR-132 isdownregulated in pancreatic cancer, which contributes to pan-creatic cancer development (39). Treatment of ormeloxifenealone or in combination with gemcitabine leads to increasedlevels ofmiR-132 in xenograft tumors (Fig. 5C). These data furtherconfirm that ormeloxifene treatment along with gemcitabinecould be an effective therapeutic modality for pancreatic cancer.

Ormeloxifene inhibits tumor desmoplasia and the host cellsinvading the tumor

To elucidate the basis of the potentiated antitumorigenic effectsof ormeloxifene in combination with gemcitabine in mice, weanalyzed the FFPE tumor tissues through tumor histopathology,immunofluorescence, and IHC analyses. We observed a clear in-hibition of Gli-1 expression in tumor tissues from mice treatedwith either ormeloxifene alone or in combination with gemcita-bine (Fig. 5D). In contrast to vehicle- or gemcitabine-treatedmice,which exhibited profuse desmoplastic tumor stroma,mice treatedwith ormeloxifene showed markedly depleted desmoplastic stro-ma. This was evidenced by a decrease in collagen I content intumor xenografts and the invading host mice cells migrating intothe tumors (Fig. 6A). It was found that only ormeloxifene, but notgemcitabine, reduced the amount of collagen I deposition. Inter-estingly, these differences were apparent in mice treated withormeloxifene or ORMþGEM. In addition, ormeloxifene aloneor in combination with gemcitabine treatment showed decreasednumber of activated stromal cell populations as identified by thereduced expression ofaSMAandfibroblast surface protein (FSP)–positive stromal myofibroblasts (Fig. 6A) and cygb/STAP positiveactivated stellate cells (Fig. 6B; ref. 37). Gemcitabine alone treat-ment did not show any effects on these parameters. This decrease






in proliferation was accompanied by a decrease in SHH expres-sion in ormeloxifene andORMþGEM–treated tumor tissues (Fig.6B). Furthermore, we analyzed these tumor tissues for the pres-ence of tumor-infiltratingmacrophages and found a large increas-ing population of macrophages in ormeloxifene-treated micetumor tissues (Fig. 6B) that might become tumoricidal andfacilitate the depletion of the tumor stroma (40, 41). This signifiesthat ormeloxifene alone or in combination with gemcitabineinhibits the host cells invading the tumor tissue and disrupts thedesmoplastic stroma that can facilitate the delivery and enhancethe efficacy of gemcitabine.

DiscussionPancreatic tumors are typically characterized by a high desmo-

plastic reaction (42). Desmoplasia plays an important role toinitiate cross-talk between stromal-cancer cells, limit the delivery

and effectiveness of chemotherapy, and induce chemoresistance.The SHH pathway is a major player for desmoplasia and isactivated in both stromal and cancer cells in PDAC (7, 43).Therefore, suppression of the Hh pathway and desmoplasia maylimit the molecular/clinical course of PDAC and improve drug(s)access in tumors (13). Currently available Hh pathway antago-nists, including GDC-0449, have been investigated as a singleagent or in combination with conventional chemotherapies forcancer treatment (14, 15). GDC-0449 is an SMO (Hh) inhibitorapproved by the FDA for the treatment of locally advanced andmetastatic basal cell carcinomas. But severe toxicity issues andadverse effects (fatigue, nausea, asthenia, mucositis, peripheralsensory neuropathy, dysgeusia, muscle spasms, and dehydration)and the lack of strong efficacy, limits its use in cancer therapy (44).In addition, no significant improvement in survival of pancreatic,colon, and ovarian cancer patients is noticed in recent clinicaltrials of Hh signaling inhibitors. As other signaling pathways

Cells: BxPC-3

A

B

D

C

CNT ORM

Panc-1

Con

trol

OR

M20

μm

ol/L

OR

M +

GE

MG

EM

100

nmol

/L

Cel

l-mig

ratio

n as

say

BxPC-3

ORM +GEM

GEM

CTHours

0

24

0

24

SDF1

CT+No

TGFβ

CT+TGFβ

GEM+TGFβ

ORM+TGFβ

ORM+GEM+TGFβ

NoPSC

CT

+T

GF

βG

EM

+T

GF

βO

RM

+T

GF

βG

EM

+O

RM

+T

GF

βC

T

SHH

DAPICygb/STAP

DAPIα SMA

Pan

c-1

BxP

C-3

ORMORM +GEM

GEM

Gli-

1 lu

cife

rase

act

ivity 1,500

1,200

900

600

300

0

Pg/

mL

20

15

10

5

0

Figure 4.Ormeloxifene and gemcitabinecombination inhibits Gli-1transcriptional activity and inhibitspancreatic cancer invasiveness. A andB, treatment of ormeloxifene alone orwith gemcitabine inhibited Gli-1transcriptional activity (A) andinhibited cell migration (B). C, wound-healing assay. The initial (0 hours) andthe residual gap length, 48 hours afterwounding, were analyzed fromphotomicrographs taken at amagnification of �200. D, indirectcoculture of PDAC and stromal cellsand treatmentwith ormeloxifene aloneand in combination with gemcitabine.ELISA was performed to observe theeffect on the secretion of key proteins(SHH and SDF1) involved in tumorstromal interactions. Bars representmean � SD; n ¼ 3; � , P < 0.01; �� , P <0.001. Immunofluorescence indicatesthat treatment with ormeloxifene andgemcitabine in the presence of PSCsreduces the number of myofibroblastsexpressing Cygb/STAP and aSMA.Images were captured using a laserscanning confocal microscope (NikonTIRF) with a �200 Apochromatobjective for immunofluorescence.The above-mentioned experimentswere performed after treatment withormeloxifene (15mmol/L), gemcitabine(100 nmol/L), and their combination(15 mmol/Lþ100 nmol/L) unlessotherwise indicated. ORM,ormeloxifene; GEM, gemcitabine.

Khan et al.





(such as PI3K or TGFb signaling; refs. 45, 46) are also known toactivate transcriptional activity of Gli in addition to SMO, thetherapeutic efficacy of SMO inhibitors is compromised in cancer.This is a probable rationale for shifting interest from SMOinhibitors toward more specific Gli inhibitors in order to effec-tively suppress the Hh signaling pathway. In this endeavor, wehave identified ormeloxifene, a nonsteroidal triphenylethylenecompound that effectively blocks the Hh signaling pathway byinhibiting the important effectors of this pathway, such as SHH,SMO, Gli-1, and SDF-1 (CXCL12). Ormeloxifene disrupts mul-tiple paracrine factors that are important for the maintenance of

Hh signaling, and thus inhibits stromal and tumor cell cross-talkwithin the tumor.

Experimental investigations indicate that ormeloxifene inhibitsproliferation, invasion, and clonogenicity of PDAC cells (Fig. 1and Supplementary Fig. S2), comparable with cells treated withGDC-0449. In addition, reduced tumorsphere formation of CSCsthat were treated with ormeloxifene indicates that ormeloxifenealso inhibits pancreatic CSC proliferation and self-renewal. Thissuggests that the anticancer effects of ormeloxifene are greater orcomparable with GDC-0449. Investigations of the mechanism oformeloxifene-induced cell death showed the induction of cell-

A

B

D

C

ORMEtOH ORM + GEMGEM

ORMEtOH

ISH

IHC

miR

-132

Gli-

1

ORM + GEMGEM

ORM

0 20 40Days

Effect of ORM on tumor development and dissemination

Treatment Mice(n)

No. of mice with dissemination to

intestine pancreas liver

Av.metastases

Meanvolume

(mm3) ofsecondary

tumors

Meantumorweight(gm)

(pr+sec)

60

EtOH

ORM + GEMGEM

EtOHORMGEMORM + GEM

6666

*Significantly different from EtOH group (P < 0.0001),

Significantly different from EtOH group (P < 0.001),

Significantly different from EtOH group (P < 0.01)

Significantly different from GEM group (P < 0.01) Significantly different from ORM group (P < 0.01)

60*0*0*

22.57.5 0*0*

10*0*0*

32 0*0*

21.20.6 0.2*

210.19.2

55.919.6*

ORMEtOH ORM+

GEM

GEM

Avg

. tum

or v

ol.

(mm

3 )

1,000

800

600

400

200

0 Tum

or w

eigh

t (g) 2.5

21.5

10.5

0

Figure 5.Ormeloxifene and gemcitabine incombination inhibit tumor growth inpancreatic xenograft mice. A,photographs of xenograft mice fromeach treatment group. B, averagetumor volume and average tumorweight was determined. Barsrepresent mean � SD; � , P < 0.01;�� , P < 0.0001. The correspondingtable shows the effect of ormeloxifeneon tumor development anddissemination. C, in situ hybridizationfor tumor-suppressor miR-132 wasperformed on the excised tumortissues from treated mice. D, IHCstaining showing the inhibition of Gli-1expression in tumor tissues from micetreated with ormeloxifene alone orwith gemcitabine. Images werecaptured using Nikon phase contrastmicroscope with �200 objective.Bars represent mean� SD; � , P < 0.01;�� , P < 0.0001. ORM, ormeloxifene;GEM, gemcitabine.






cycle arrest at G0–G1 phase, suggesting that ormeloxifene mayinduce apoptosis. It was also an intriguing observation thattreatment of ormeloxifene in combination with gemcitabineshowed an increasing population of Annexin V–positive cells ascompared with when both were used alone. These results indicatethat in the presence of ormeloxifene, gemcitabine induces higherapoptotic cell death that might be triggered through the mito-chondrial pathway. Alternatively, the other possibility is thatormeloxifene might involve death receptor–mediated cell death.

Altogether, the results indicated that ormeloxifene potentiates theanticancer effect of gemcitabine when used in combination.

It has been reported that NF-kB (33) and SHH (7, 43) signalingpathways play crucial roles in PDAC progression and drug resis-tance, including gemcitabine. Ormeloxifene treatment stabilizesIkBa, which inhibits protein and transcriptional activity of NF-kB-65, preventing it from binding to the SHH promoter andleading to its transcription (33). This was further confirmed onfinding that ormeloxifene alone or in combination with gemci-tabine inhibits the main downstream targets of SHH, SMO, andPTCH1/2 and downregulates the expression and transcriptionalactivity of Gli-1 in PDAC cells. No such effects were found in cellswhen treatedwith gemcitabine alone. In the absence of SHH, cellshave small amounts of PTCH1/2 and Gli and therefore, the highconcentrations of these transcripts generally indicate involvementof the SHH pathway in PDAC (47). The aberrantly activated SHHbinds to its receptor PTCH1/2 and inhibits the suppressive effectof PTCH1/2 on SMO, which activates Gli-1 to transcribe Hhoncogenic target genes (43). Ormeloxifene inhibits AKT phos-phorylation, which is known to activate Gli-1 (48). The observa-tions collected from coculturing the PDAC cells with stromal cellsindicate the inhibition of paracrine stromal cell signaling throughthe inhibition of their proliferation and secretion of SHH andSDF1. All these results confirm that ormeloxifene inhibits Hhsignaling in PDAC cells; thus, we hypothesized that it might alsodisrupt the stroma of pancreatic tumors and alter the desmoplas-tic reaction.

Recent studies implicate the profound role of stroma in drugresistance in numerous tumor types (49). Thus, treatment para-digms targeting both neoplastic cells and stromal components areemerging for PDAC (50). The enhanced antitumor effect oformeloxifene and gemcitabine combination treatment wasobserved in xenograft mouse models when compared with theirtreatment alone. An abundant stromal component was observedin the control and gemcitabine-treated tumor tissues while micetreated with ormeloxifene alone or ORMþGEM combinationshowed markedly less stromal component and invaded stromaltissue. This was indicated by the presence of reduced numbers ofstroma myofibroblasts infiltrating the tumor tissue, as indicatedby reduced PSCs, aSMA, FSP, and cygb/STAP expression in thetumor tissue. This might be well supported by the observationsshowing the inhibition of both mouse and human collagen I intissues, as activated PSCs are the predominant source of collagenin the desmoplastic reaction in pancreatic cancers (6). Interest-ingly, we also observed an increased number of macrophages inthe tumors obtained fromormeloxifene-treatedmice. The increas-ingmacrophage recruitment to tumor site can be explained as theemergence of tumor immunity that serves as a part of immunesurveillance for targeting tumor stroma in the treatment of cancer.The activated macrophages may rapidly infiltrate the tumors,become tumoricidal and facilitate the depletion of tumor stroma.

These findings suggest that ormeloxifene inhibits desmoplasticreaction in PDAC. Due to the toxicity and morbidity of availabledrugs, there is an urgent need for effective therapies that couldtarget both tumor and stromal compartments to regulate pancre-atic tumor growth. Therefore, ormeloxifene might be a drug ofchoice as it is very safe for chronic use in humans. Our resultssignify that ormeloxifene is effective in targeting Hh and tumorprogression signaling pathways.Our results emphasize that orme-loxifene interrupts the tumor–stromal interactions to inhibit thereciprocal relationship between these two components, leading to

DAPI/mouse collagen-1/Human collagen-1

A

B

DAPI/α SMADAPI/fibroblast SP

DAPI/SHH DAPI/Cygb/STAP

Eth

anol

OR

MG

EM

OR

M+

GE

ME

than

olO

RM

GE

MO

RM

+G

EM

Hematoxylin/F4/80

Figure 6.Representative photomicrographs of immunofluorescence studies onexcised xenograft tumor tissues using confocal microscopy. A and B,treatment of ormeloxifene aloneor in combinationwith gemcitabine inhibitedboth human andmouse collagen I, FSP, and therefore reduced the number oftotal stroma cells within the tumor, which is indicated by reducedmyofibroblasts expressing aSMA and cygb/STAP. This was observed using alaser scanning confocal microscope (Nikon TIRF), original magnifications�200. In addition, tissues were stained for F4/80 that indicated increasednumber of macrophages infiltrating into the tumor. ORM, ormeloxifene; GEM,gemcitabine.

Khan et al.





reduction of tumor progression, invasion, metastasis, and che-moresistance (Fig. 7). This facilitates the anticancer effects oformeloxifene and potentiates the chemotherapeutic effects ofgemcitabine for pancreatic cancer treatment. Our results haveimportant implications toward the development of effectivetherapy for pancreatic cancer treatment.

ConclusionIn summary, our study provides new evidence regarding the

anticancer effects of ormeloxifene in PDAC. This study demon-strates novel role of an existing drug, ormeloxifene to inhibitthe SHH pathway and desmoplasia, resulting in tumor growthinhibition and potentiation of the antitumor effect of gemcita-bine. This suggests that a combination of ormeloxifene andgemcitabine may have the capacity to inhibit the SHH signalingcascade in PDAC cells and alter the behavior of surroundingstromal cells so that cancer progression is repressed. Therefore,this study provides evidence that ormeloxifene in combinationwith gemcitabine could serve as a novel therapeutic interventionfor pancreatic cancer.

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

Authors' ContributionsConception and design: S. Khan, M.M. Yallapu, S.W. Behrman, M.M. Singh,M. Jaggi, S.C. ChauhanDevelopment of methodology: S. Khan, M.M. Yallapu, S.C. Chauhan

Acquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): S. Khan, M.C. Ebeling, N. Chauhan, R.K. Gara,M.M. Yallapu, S.C. ChauhanAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): S. Khan, N. Chauhan, P.A. Thompson, R.K. Gara,A. Ganju, S.W. Behrman, M. Jaggi, S.C. ChauhanWriting, review, and/or revision of the manuscript: S. Khan, P.A. Thompson,M.M. Yallapu, S.W. Behrman, M.M. Singh, S.C. ChauhanAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): S. Khan, H. Zhao, N. Zafar, M. Jaggi,S.C. ChauhanStudy supervision: M. Jaggi, S.C. Chauhan

AcknowledgmentsThe authors thank the Kosten Foundation for pancreatic cancer research

support. The authors are also thankful to Cathy Christopherson (SanfordResearch) for editorial assistance.

Grant SupportThis work was partially supported by grants from Department of Defense

(PC073887 to S.C. Chauhan and PC073643 to M. Jaggi), the NIH (RO1CA142736 to S.C. Chauhan), Pilot grant to S. Khan (8P20GM103548-02) andUO1 CA162106A (S.C. Chauhan and M. Jaggi) and the College of Pharmacy2013 Dean's Seed Grant of the University of Tennessee Health Science Center(M. Jaggi and M.M. Yallapu).

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

Received August 20, 2014; revised January 29, 2015; accepted February 21,2015; published OnlineFirst April 3, 2015.

Stellate cells

Endothelial cell

Normal fibroblast

CAFs

Proangiogenic TAM

Antitumorigenicmacrophage

PDAC cell Stromal cell

PTCH 1/2 SMO

Gli

CXCR4

Secretorypathway

catalyticsignalingSHH

Target genes

SUFU

CXCL 12

MMPsCollagen-1

VEGFbFGF

Paracrine

signalingORM

ORM

ORM

Tumor progression

VEGF

TGFβ

NF-κβ

Release of growthfactors

Figure 7.Diagrammatic representation of theormeloxifene-mediatedmodulation ofthe SHH signaling pathway.






References1. Burris HA III, Moore MJ, Andersen J, Green MR, Rothenberg ML, Modiano

MR, et al. Improvements in survival and clinical benefitwith gemcitabine asfirst-line therapy for patients with advanced pancreas cancer: a randomizedtrial. J Clin Oncol 1997;15:2403–13.

2. Moorcraft SY, Khan K, Peckitt C, Watkins D, Rao S, Cunningham D, et al.FOLFIRINOX for locally advanced or metastatic pancreatic ductal adeno-carcinoma: the royal marsden experience. Clin Colorectal Cancer 2014;13:232–8.

3. Conroy T,Desseigne F, YchouM,BoucheO,GuimbaudR, Becouarn Y, et al.FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl JMed 2011;364:1817–25.

4. Von Hoff DD, Ervin T, Arena FP, Chiorean EG, Infante J, Moore M, et al.Increased survival in pancreatic cancer with nab-paclitaxel plus gemcita-bine. N Engl J Med 2013;369:1691–703.

5. Han H, Bearss DJ, Browne LW, Calaluce R, Nagle RB, Von Hoff DD.Identification of differentially expressed genes in pancreatic cancer cellsusing cDNA microarray. Cancer Res 2002;62:2890–6.

6. Apte MV, Park S, Phillips PA, Santucci N, Goldstein D, Kumar RK, et al.Desmoplastic reaction in pancreatic cancer: role of pancreatic stellate cells.Pancreas 2004;29:179–87.

7. Bailey JM, SwansonBJ,Hamada T, Eggers JP, SinghPK,Caffery T, et al. Sonichedgehog promotes desmoplasia in pancreatic cancer. Clin Cancer Res2008;14:5995–6004.

8. Pasca di Magliano M, HebrokM. Hedgehog signalling in cancer formationand maintenance. Nat Rev Cancer 2003;3:903–11.

9. Thayer SP, di Magliano MP, Heiser PW, Nielsen CM, Roberts DJ, LauwersGY, et al. Hedgehog is an early and late mediator of pancreatic cancertumorigenesis. Nature 2003;425:851–6.

10. Merchant AA, Matsui W. Targeting hedgehog—a cancer stem cell pathway.Clin Cancer Res 2010;16:3130–40.

11. Varjosalo M, Taipale J. Hedgehog: functions and mechanisms. Gene Dev2008;22:2454–72.

12. Kelleher FC. Hedgehog signaling and therapeutics in pancreatic cancer.Carcinogenesis 2011;32:445–51.

13. Olive KP, Jacobetz MA, Davidson CJ, Gopinathan A, McIntyre D,Honess D, et al. Inhibition of Hedgehog signaling enhances deliveryof chemotherapy in a mouse model of pancreatic cancer. Science 2009;324:1457–61.

14. Kim EJ, Sahai V, Abel EV, Griffith KA, Greenson JK, Takebe N, et al. Pilotclinical trial of hedgehog pathway inhibitor GDC-0449 (vismodegib) incombination with gemcitabine in patients with metastatic pancreaticadenocarcinoma. Clin Cancer Res 2014;20:5937–45.

15. LoRusso PM, Rudin CM, Reddy JC, Tibes R,Weiss GJ, BoradMJ, et al. PhaseI trial of hedgehog pathway inhibitor vismodegib (GDC-0449) in patientswith refractory, locally advanced or metastatic solid tumors. Clin CancerRes 2011;17:2502–11.

16. Juhasz ML, Marmur ES. Systematic review of vismodegib toxicity profilein the treatment of advanced Basal cell carcinomas compared to othersystemic therapies in dermatology. J Drugs Dermatol 2014;13:729–33.

17. Sheikh A, Alvi AA, Aslam HM, Haseeb A. Hedgehog pathway inhibitors—current status and future prospects. Infect Agent Cancer 2012;7:29.

18. Gara RK, SundramV, Chauhan SC, JaggiM. Anti-cancer potential of a novelSERM ormeloxifene. Curr Med Chem 2013;20:4177–84.

19. Singh MM. Centchroman, a selective estrogen receptor modulator, as acontraceptive and for the management of hormone-related clinical dis-orders. Med Res Rev 2001;21:302–47.

20. Lipinski RJ, Hutson PR, Hannam PW, Nydza RJ, Washington IM, MooreRW, et al. Dose- and route-dependent teratogenicity, toxicity, and phar-macokinetic profiles of the hedgehog signaling antagonist cyclopamine inthe mouse. Toxicol Sci 2008;104:189–97.

21. LoRusso PM, Piha-Paul SA, Mita M, Colevas AD, Malhi V, Colburn D, et al.Co-administration of vismodegib with rosiglitazone or combined oralcontraceptive in patients with locally advanced or metastatic solid tumors:a pharmacokinetic assessment of drug–drug interaction potential. CancerChemother Pharm 2013;71:193–202.

22. Shankar S, Nall D, Tang SN, Meeker D, Passarini J, Sharma J, et al.Resveratrol inhibits pancreatic cancer stem cell characteristics in humanand KrasG12D transgenic mice by inhibiting pluripotency maintainingfactors and epithelial-mesenchymal transition. PLoSONE 2011;6:e16530.

23. Ji X MY, Liu Y, Jin T, Song P. J Chin Pharm Sci 1998;7:69–71.

24. Limame R, Wouters A, Pauwels B, Fransen E, Peeters M, Lardon F, et al.Comparative analysis of dynamic cell viability, migration and invasionassessments by novel real-time technology and classic endpoint assays.PLoS ONE 2012;7:e46536.

25. Khan S, Kaur R, Shah BA, Malik F, Kumar A, Bhushan S, et al. A Novelcyano derivative of 11-Keto-beta-Boswellic acid causes apoptotic deathby disrupting PI3K/AKT/Hsp-90 cascade, mitochondrial integrity, andother cell survival signaling events in HL-60 cells. Mol Carcinogen2012;51:679–95.

26. Kalluri R, Weinberg RA. The basics of epithelial–mesenchymal transition.J Clin Invest 2009;119:1420–8.

27. BernaMJ, SeizO,Nast JF, BentenD, BlakerM,Koch J, et al. CCK1 andCCK2receptors are expressed on pancreatic stellate cells and induce collagenproduction. J Biol Chem 2010;285:38905–14.

28. ChauhanSC, EbelingMC,MaherDM,KochMD,WatanabeA,AburataniH,et al. MUC13mucin augments pancreatic tumorigenesis. Mol Cancer Ther2012;11:24–33.

29. Khan S, Ebeling MC, Zaman MS, Sikander M, Yallapu MM, Chauhan N,et al. MicroRNA-145 targets MUC13 and suppresses growth and invasionof pancreatic cancer. Oncotarget 2014;5:7599–609.

30. YallapuMM, EbelingMC, Khan S, SundramV, ChauhanN,Gupta BK, et al.Novel curcumin-loaded magnetic nanoparticles for pancreatic cancertreatment. Mol Cancer Ther 2013;12:1471–80.

31. Abassi YA, Xi B, Zhang W, Ye P, Kirstein SL, Gaylord MR, et al.Kinetic cell-based morphological screening: prediction of mechanismof compound action and off-target effects. Chem Biol 2009;16:712–23.

32. DunaevaM, Michelson P, Kogerman P, Toftgard R. Characterization of thephysical interaction of Gli proteins with SUFU proteins. J Biol Chem2003;278:5116–22.

33. Kasperczyk H, Baumann B, Debatin KM, Fulda S. Characterization of sonichedgehog as a novel NF-kappaB target gene that promotes NF-kappaB–mediated apoptosis resistance and tumor growth in vivo. FASEB J 2009;23:21–33.

34. Morton JP, Mongeau ME, Klimstra DS, Morris JP, Lee YC, Kawaguchi Y,et al. Sonic hedgehog acts at multiple stages during pancreatic tumorigen-esis. Proc Natl Acad Sci U S A 2007;104:5103–8.

35. Matsushime H, Roussel MF, Ashmun RA, Sherr CJ. Colony-stimulatingfactor 1 regulates novel cyclins during the G1 phase of the cell cycle. Cell1991;65:701–13.

36. Shekhar MP, Pauley R, Heppner G. Host microenvironment in breastcancer development: extracellular matrix–stromal cell contribution toneoplastic phenotype of epithelial cells in the breast. Breast Cancer Res2003;5:130–5.

37. LiuM,Xu J,DengH. Tangledfibroblasts in tumor–stroma interactions. Int JCancer 2011;129:1795–805.

38. Singh AP, Arora S, Bhardwaj A, Srivastava SK, Kadakia MP, Wang B, et al.CXCL12/CXCR4 protein signaling axis induces sonic hedgehog expres-sion in pancreatic cancer cells via extracellular regulated kinase- and Aktkinase-mediated activation of nuclear factor kappaB: implicationsfor bidirectional tumor–stromal interactions. J Biol Chem 2012;287:39115–24.

39. Zhang S,Hao J, Xie F,HuX, LiuC, Tong J, et al. Downregulation ofmiR-132by promoter methylation contributes to pancreatic cancer development.Carcinogenesis 2011;32:1183–9.

40. Beatty GL, Chiorean EG, Fishman MP, Saboury B, Teitelbaum UR,Sun W, et al. CD40 agonists alter tumor stroma and show efficacyagainst pancreatic carcinoma in mice and humans. Science 2011;331:1612–6.

41. Mantovani A, Sica A. Macrophages, innate immunity and cancer: balance,tolerance, and diversity. Curr Opin Immunol 2010;22:231–7.

42. Merika EE, Syrigos KN, SaifMW.Desmoplasia in pancreatic cancer. Canwefight it? Gastroenterol Res Pract 2012;2012:781765.

43. Bailey JM, Mohr AM, Hollingsworth MA. Sonic hedgehog paracrine sig-naling regulates metastasis and lymphangiogenesis in pancreatic cancer.Oncogene 2009;28:3513–25.

44. Berlin J, Bendell JC, Hart LL, Firdaus I, Gore I, Hermann RC, et al. Arandomized phase II trial of vismodegib versus placebo with FOLFOX orFOLFIRI and bevacizumab in patients with previously untreatedmetastaticcolorectal cancer. Clin Cancer Res 2013;19:258–67.

Khan et al.





45. Javelaud D, Pierrat MJ, Mauviel A. Crosstalk between TGF-beta andhedgehog signaling in cancer. FEBS Lett 2012;586:2016–25.

46. Ramaswamy B, Lu Y, Teng KY, Nuovo G, Li X, Shapiro CL, et al.Hedgehog signaling is a novel therapeutic target in tamoxifen-resistantbreast cancer aberrantly activated by PI3K/AKT pathway. Cancer Res2012;72:5048–59.

47. ElaminMH, Shinwari Z, Hendrayani SF, Al-Hindi H, Al-Shail E, Khafaga Y,et al. Curcumin inhibits the sonic hedgehog signaling pathway and triggersapoptosis in medulloblastoma cells. Mol Carcinogen 2010;49:302–14.

48. Stecca B, Mas C, Clement V, Zbinden M, Correa R, Piguet V, et al.Melanomas require HEDGEHOG-GLI signaling regulated by interactionsbetween GLI1 and the RAS-MEK/AKT pathways. Proc Natl Acad Sci U S A2007;104:5895–900.

49. Straussman R, Morikawa T, Shee K, Barzily-Rokni M, Qian ZR, Du J, et al.Tumour micro-environment elicits innate resistance to RAF inhibitorsthrough HGF secretion. Nature 2012;487:500–4.

50. HeinemannV,HaasM, Boeck S. Systemic treatment of advanced pancreaticcancer. Cancer Treat Rev 2012;38:843–53.






Correction

Correction: Ormeloxifene SuppressesDesmoplasia and Enhances Sensitivity ofGemcitabine in Pancreatic Cancer

In the original version of this article (1), the loading controls for theWestern blots inFigs. 2D and 3D were missing for all different percentages of gels. These have beenadded in the recent online HTML and PDF versions of the article. The authors regretthis omission.

Reference1. Khan S, Ebeling MC, Chauhan N, Thompson PA, Gara RK, Ganju A, et al. Ormeloxifene sup-

presses desmoplasia and enhances sensitivity of gemcitabine in pancreatic cancer. Cancer Res2015;75:2292–304.

Published online May 1, 2018.doi: 10.1158/0008-5472.CAN-18-0427�2018 American Association for Cancer Research.

CancerResearch

Cancer Res; 78(9) May 1, 20182444

http://crossmark.crossref.org/dialog/?doi=10.1158/0008-5472.CAN-18-0427&domain=pdf&date_stamp=2018-4-27

Published OnlineFirst April 3, 2015.Cancer Res Sheema Khan, Mara C. Ebeling, Neeraj Chauhan, et al. Sensitivity of Gemcitabine in Pancreatic CancerOrmeloxifene Suppresses Desmoplasia and Enhances

Updated version

10.1158/0008-5472.CAN-14-2397doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2015/04/04/0008-5472.CAN-14-2397.DC1

Access the most recent supplemental material at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://cancerres.aacrjournals.org/content/early/2015/05/15/0008-5472.CAN-14-2397To 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-14-2397

http://cancerres.aacrjournals.org/content/suppl/2015/04/04/0008-5472.CAN-14-2397.DC1

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/early/2015/05/15/0008-5472.CAN-14-2397


Ormeloxifene Suppresses Desmoplasia and Enhances...

Documents

Transcript of Ormeloxifene Suppresses Desmoplasia and Enhances...