SonicHedgehogPromotesDesmoplasiainPancreaticCancerTo identify SHH in the primary human tumor sections, rabbit anti-SHH primary antibody (1:200; Santa Cruz)was incubated on paraffin-embedded

Sonic Hedgehog Promotes Desmoplasia in Pancreatic CancerJenniferM. Bailey, BenjaminJ. Swanson,Tomofumi Hamada, John P. Eggers, Pankaj K. Singh,Thomas Caffery, Michel M. Ouellette, andMichael A. Hollingsworth

Abstract Purpose:We investigated the contribution of Sonic hedgehog (SHH) to pancreatic cancerprogression.Experimental Design:We expressed SHH in a transformed primary ductal-derived epithelialcell line from the human pancreas, transformed hTert-HPNE (T-HPNE), and evaluated the effectson tumor growth.We also directly inhibited the activity of SHH in vivo by administering a blockingantibody to mice challenged orthotopically with the Capan-2 pancreatic cancer cell line, which isknown to express SHHand formmoderately differentiated tumors in nude mice.Results: Our data provide evidence that expression of SHH influences tumor growth by contri-buting to the formation of desmoplasia in pancreatic cancer. We further show that SHH affectsthe differentiation and motility of human pancreatic stellate cells and fibroblasts.Conclusions:These data suggest that SHH contributes to the formation of desmoplasia in pan-creatic cancer, an important component of the tumor microenvironment.

Pancreatic cancer is the fourth leading cause of cancer-relateddeaths in the United States, resulting in 30,000 deaths eachyear, in part because it is one of the most lethal cancers with adeath-to-incidence ratio of 0.99 (1). The lethality of pancreaticcancer is largely due to late stage of diagnosis and resistance tocurrent therapies. More than 80% of patients are unresectableand most present with metastatic disease due to aggressivelocalized invasion and a high incidence of early metastasis (2).Worldwide, the incidence of death from pancreatic cancerhas increased, which, together with the lack of current thera-pies, emphasizes the need to determine the mechanisms ofpancreatic cancer progression and better targets for diagnosisand treatment.

The current model of pancreatic cancer progression includesmutations that activate K-ras, followed by inactivating muta-tions or loss of expression of tumor suppressor genes includingp53, p16INK4A , and SMAD4 (3). Sonic hedgehog (SHH) andother proteins downstream of the hedgehog pathway wererecently detected in precursor lesions and samples of primarytumors from patients with pancreatic adenocarcinoma, whichimplicates this pathway in the initiation and progression ofpancreatic cancer given its absence in normal adult pancreas(4, 5). Transgenic mouse models in which SHH was expressed

in mouse pancreatic epithelium, in concert with activated Rassignaling, enhanced the formation of pancreatic intraepithelialneoplasia and accelerated lethality, further supporting a rolefor hedgehog as an early contributor to this lethal disease (6).

We investigated the contribution of SHH to pancreatic cancerprogression by expressing SHH in a transformed primaryductal-derived epithelial cell line from the human pancreas,transformed hTert-HPNE (T-HPNE), which was describedpreviously (7, 8). This cell line has been documented torepresent an appropriate and important model system todelineate signaling mechanisms for Ras-mediated oncogenesisand growth of pancreatic ductal epithelial cells (8), and as suchrepresents a reasonable system to evaluate the superimpositionof SHH expression on pancreatic cells with a defined andlimited set of transforming insults. We also directly inhibitedthe activity of SHH in vivo using the Capan-2 pancreatic cancercell line in an orthotopic model, which is known to expressSHH and form moderately differentiated tumors in nude mice.Our data provide evidence that expression of SHH contributesto the formation of desmoplasia in pancreatic cancer. Wefurther show that SHH affects the differentiation and motilityof human pancreatic stellate cells and fibroblasts. These dataimplicate SHH as a mediator of the desmoplastic response inpancreatic cancer.

Materials andMethods

Cell lines and culture conditions. The cell line we used for theseexperiments was hTert-HPNE. The cells were originally isolated fromthe ductal structure of a human pancreas and immortalized byexpressing the catalytic subunit of telomerase (h-Tert), along withsubsequent transductions with retroviral constructs to introduce a Krasmutation, the human papillomavirus E6 and E7 peptides, and the SV40small t antigen, as described previously (7, 8). The hTert-HPNE cellswere maintained in Medium D containing 1 volume of M3 base (InCellCorp.), 3 volumes of glucose-free DMEM, 5% fetal bovine serum (FBS),5.5 mmol/L glucose, 10 ng/mL epidermal growth factor, and 50 Ag/mL

Human Cancer Biology

Authors’Affiliation: Eppley Institute, University of Nebraska Medical Center,Omaha, NebraskaReceived 2/4/08; revised 4/24/08; accepted 5/14/08.The costs of publication of this article were defrayed in part by the payment of pagecharges.This article must therefore be hereby marked advertisement in accordancewith18 U.S.C. Section1734 solely to indicate this fact.Note: Supplementary data for this article are available at Clinical Cancer ResearchOnline (http://clincancerres.aacrjournals.org/).Requests for reprints:Michael A. Hollingsworth, University of NebraskaMedicalCenter, 986805 Nebraska Medical Center, Omaha, NE 68198-6805. Phone: 402-559-8343; Fax: 402-559-3339; E-mail: [email protected].

F2008 American Association for Cancer Research.doi:10.1158/1078-0432.CCR-08-0291

www.aacrjournals.org Clin Cancer Res 2008;14(19) October1, 20085995

Cancer Research. on February 15, 2021. © 2008 American Association forclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

gentamicin. The cells were cultured at 37jC in a humidified atmospherecontaining 5% CO2.

For overexpression of SHH, the cDNA of SHH was cloned into thepWZL-blast retroviral vector [courtesy of Dr. Wade Bushman (Uni-versity of Wisconsin Medical School, Madison, WI) and Dr. MichelOuellette (Eppley Institute, Omaha, NE)]. Clones expressing SHH wereisolated by serial dilutions and selection with blastocidin. A controlvector was transduced into the control cell lines.

The Capan-2 pancreatic cancer cell line is maintained in McCoy’s 5Amedium and are cultured at 37jC in a humidified atmospherecontaining 5% CO2. Pancreatic stellate cells and fibroblasts wereisolated from an islet transplant program at the University of NebraskaMedical Center and are maintained in RPMI medium at 37jC in ahumidified atmosphere containing 5% CO2.

Western blot analysis. Whole cell lysates were prepared by adding1.5-mL cell lysis buffer [10 nmol/L Tris-HCl (pH 8.0), 150 mmol/LNaCl, 1 mmol/L phenylmethylsulfonyl fluoride and 1% Triton X-100]to cells grown to 80% confluence in T175 tissue culture flasks. Sixtymicrograms of protein were loaded onto 4% to 20% Novex Tris-glycine gradient denaturing polyacrylamide gels (Invitrogen) in a 1�SDS-PAGE running buffer, containing 1 g/L SDS, 3 g/L Tris-base, and14.4 g/L glycine. The proteins were transferred onto polyvinylpyrolidinedifluoride membranes by electrophoresis. Once transferred, themembranes were blocked in Blotto [5% dry milk resuspended in 1�TBS (0.9% NaCl, 10 mmol/L Tris pH 7.4, and 0.5% MgCl)] overnight at4jC. Anti-SHH primary antibody was incubated at 1:200 dilution atroom temperature for 2 h. The membrane was washed thrice, 10 mineach time, before the addition of the secondary antibody (goat anti-rabbit HRP) at 1:5,000 dilution for 50 min. Immune complexes wereidentified with enhanced chemiluminescence Western blotting detec-tion reagents. Membranes were stripped and subsequently blotted withanti–h-actin antibody as a loading control.

Invasion assay. The in vitro invasive potential of cells overexpressingSHH was assayed using the Biocoat Matrigel Invasion Chamber (BectonDickinson Labware). The invasion assay with pancreatic myofibroblastsincluded an initial seeding of 50,000, 250,000, or 500,000 T-HPNE orT-HPNE.SHH cells in the lower chambers. The cells were allowed toadhere for 8 h. The medium was then removed and replaced with RPMImedium containing 7% FBS. In the upper chambers, 250,000 pancreaticmyofibroblasts were seeded and then allowed to incubate for 22 h at37jC and 5% CO2. For invasion with the pancreatic fibroblasts withrhSHH, 250,000 cells were seeded in the upper chamber in RPMImedium containing 7% FBS. The lower chambers included controlRPMI with 7% FBS, 0.1 Ag/mL recombinant SHH (R&D Systems), and1.0 Ag/mL recombinant SHH. These chambers were also allowed to sitfor 22 h at 37jC and 5% CO2. For analyses of all invasion assays, cells onthe upper portion of the matrix membrane were wiped off with a cotton-tipped swab and cells on the bottom of this membrane were fixed and

stained with Diff-Quick staining kit (Allegiance) and observed by lightmicroscopy. The number of invading cells was quantified by countingthe membranes at �200 magnification.

Subcutaneous injections in athymic nude mice. Three million cellswere injected s.c. in athymic nude mice. Before injection, the cells weretrypsinized, counted, washed twice in 1� PBS, and resuspended at adensity of 3 � 106/30 AL. The cells were injected between the scapulae.Tumor growth was monitored every 2 d by measuring the diameterswith a caliper. The mice were euthanized when the tumor volumereached 1,800 mm3.

Orthotopic implantation in athymic nude mice. One million T-HPNEand T-HPNE.SHH cells were each injected into the pancreas of 12 or13 mice, respectively, to analyze tumor desmoplasia in response tooverexpression of SHH. Preparation of cells was as described above forthe s.c. injection model. Orthotopic implantation of T-HPNE cell lineswas done as described previously (9).

SHH inhibition in orthotopic models. To inhibit SHH in the Capan-2orthotopic models, we administered the 5E1 antibody by i.p. injectionsat a dosage of 500 Ag/mouse once a week for 10 wk. An IgG controlantibody, 4E11, was also given to a separate group of mice at aconcentration of 500 Ag/mouse, once per week.

Confocal and immunohistochemical analyses. To analyze cell mor-phology in the orthotopic tumors, primary tumors were resected andplaced in formalin for 24 h before being fixed for staining. Stainingwith H&E allowed for the visualization of cells within the tumorsections. Fibroblasts were visually differentiated from tumor cells on thebasis of cellular morphology and by staining with the markersdescribed below.

To identify SHH in the primary human tumor sections, rabbitanti-SHH primary antibody (1:200; Santa Cruz) was incubated onparaffin-embedded tissue sections for 2 h at 37jC. The secondaryantibody (antirabbit, 1:1,000) was incubated for 50 min at 37jC.Confirmation of desmoplasia included analysis of reactivity with a-smooth muscle actin (SMA; 1:400; Sigma), collagen I (1:50; Abcam),and fibronectin (1:50; Abcam). The secondary antibody (antimouse,1:1,000) was incubated for 60 min. Staining was developed withVectastain avidin-biotin complex reagent and 3,3¶-diaminobenzidinesubstrate.

For confocal microscopy, anti– ribonuclear protein (1:20; Chem-icon) and anti-SMA (1:500; Sigma) were incubated on paraffin-embedded tissue sections for 1 h at room temperature. Secondaryreagents, antimouse IgM-Alexa 647, and antimouse IgG-Alexa 488(1:2,000) were incubated for 1 h.

Primary cell culture. Primary human pancreatic stellate cells andmyofibroblasts were isolated as described previously (7, 10). Cells weremonitored for 3 wk and were identified by cellular morphology and byanalysis with markers described above as cells of mesenchymal lineageor pancreatic myofibroblasts.

In vitro wound healing assays. Wound healing assays were donewith the pancreatic myofibroblasts. Cells (30,000) were seeded in24-well plates and grown into a monolayer. A pipette was used to makea scratch through the monolayer and medium was replaced. RPMIcontaining 7% FBS was used as a control. RPMI containing 7% FBS and0.1 Ag/mL rhSHH and RPMI containing 7% FBS, 0.1 Ag/mL rhSHH, and0.1 Ag/mL 5E1 (Iowa hybridoma bank) served as the experimentalconditions.

Statistical analysis. The in vitro colony forming, wound healingassays, and invasion assays were analyzed with one-way ANOVA test aspart of the Prism statistical analysis package. Significance wasdetermined at the 95% confidence interval.

Results

Inhibition of SHH signaling decreases desmoplasia. Weevaluated the contribution of SHH to the growth properties ofpancreatic tumor cells by administering monoclonal antibody

Translational Relevance

The data in this article suggest that SHH contributesin a paracrine manner to the formation of desmoplasia inpancreatic cancer. The robust desmoplastic reaction seenin many pancreatic cancers, which includes fibroblasts,endothelial cells, and other stromal cell types, is believedto contribute to tumor progression by creating protumori-genic effects in the tumor microenvironment. Desmoplasiaalso creates barriers to the delivery of therapeutic com-pounds to tumor cells. Inhibition of desmoplasia wouldinhibit tumor growth and progression, and elimination ofdesmoplasia would be predicted to enhance the potentialof delivery of some therapeutics to pancreatic tumors.


www.aacrjournals.orgClin Cancer Res 2008;14(19) October1, 2008 5996



5E1 (11), a blocking antibody to SHH, to mice challengedorthotopically with Capan-2, which expresses moderately highlevels of SHH and forms moderately differentiated pancreatictumors with desmoplasia (12). One million Capan-2 cells wereimplanted orthotopically into the pancreas of athymic nudemice and 500 Ag of 5E1 were administered once per week.An isotype control IgG antibody, 4E11, was given to a controlgroup of mice. Treatment with 5E1 resulted in a decreasedtumor volume from an average of f0.45 cm3 (Capan-2controls) to f0.1 cm3 (5E1-treated mice).

Histologic and immunohistologic analyses of tumor sectionsrevealed that inhibiting SHHwith 5E1 significantly decreased thedegree of desmoplasia. Figure 1 shows staining of tumor sectionswith H&E and with the mesenchymal marker for differentiatedfibroblasts, a-SMA. The images in Fig. 1A are representative of the

desmoplastic reaction in tumors derived from the Capan-2 cellline with and without inhibition of SHH by treatment withantibody 5E1. Tumors in animals treated with the controlantibody, 4E11, showed dense desmoplastic reactions. Tumorsderived from mice treated with the 5E1 showed minimalevidence of desmoplasia.We confirmed that areas of desmoplasiacontained myofibroblasts by staining the tumor sections withSMA (Fig. 1B). Morphometric analysis revealed a significantdecrease in the number of SMA+ cells between control tumorsand tumors from mice treated with 5E1 (Fig. 1C).

SHH expression promotes a desmoplastic reaction when over-expressed in transformed HPNE cells. To further investigate therole of SHH in the production of pancreatic cancer desmo-plasia, we overexpressed SHH in a transformed primarypancreatic epithelial cell line, T-HPNE. SHH overexpression

Fig. 1. Inhibition of SHH signaling decreases desmoplasia in Capan-2 orthotopic tumors. A, histologic analysis of tumors derived following orthotopic challenge withCapan-2 and treatment with a blocking antibody for SHH (5E1), an isotype control (4E11), or no treatment (Capan-2). H&E staining, at �100 magnification, shows reactivestroma and desmoplasia in the Capan-2 and control 4EII-treated Capan-2 tumors (see black arrows).The 5E1-treated Capan-2 tumors showed little or no evidence ofdesmoplasia by histologic examination. B, immunohistochemistry with the mesenchymal marker SMA. Paraffin-embedded sections of tumors that arose from orthopicimplantation of Capan-2 cells and treatment with a blocking antibody for SHH (5E1), an isotype control (4E11), or no treatment (Capan-2). Brown coloration, positive stainingfor SMA. Counterstaining was done with hematoxylin (blue). C, morphometric quantification of the relative amount of SMA+ desmoplasia in the Capan-2 control,4E11-treated, and 5E1-treated tumors.The 5E1tumors showed a significant decrease in the amount of SMA+ desmoplasia. Relative SMA+ areas on the tumor sectionswere quantified by calculating the total cellular content of the section and expressing the results as a percentage: (SMA+ desmoplasia / all cell types in total area) � 100.Two representative results of 5E1-treated Capan-2 primary tumors showed some staining with SMA, but the areas of staining were individualized fibroblasts representing<20% of the section. Inhibition of SHH significantly decreased the amount of SMA+ desmoplasia in Capan-2 tumors (n = 5 per group).

SHH, Desmoplasia, and Pancreatic Cancer




in this cell line was confirmed by Western blot (Supplemen-tary Fig. S1). T-HPNE and T-HPNE cells expressing SHH (T-HPNE.SHH) were implanted orthotopically into the pancreasof athymic nude mice. Both T-HPNE cells and T-HPNE.SHHformed poorly differentiated tumors at the orthotopic site.Expression of SHH in the T-HPNE cells resulted in an increasein mean tumor weight from f1 g (T-HPNE) to f3 g (T-HPNE.SHH). There was dense desmoplasia in tumors producedby the T-HPNE.SHH cell line, whereas desmoplasia wasminimal in tumors derived from T-HPNE cells not expressingSHH, as evidenced by morphologic evaluation and immuno-histochemical analysis of the orthotopic tumors derived fromthe T-HPNE.SHH cell line with SMA to confirm that cells in thedesmoplastic reaction were myofibroblasts. Our analysisshowed a significant increase in cells that stained positive forthis mesenchymal marker within the tumor and at theperiphery (Fig. 2A and B). Tumors derived from the T-HPNEcell line showed minimal reactivity with a-SMA (Fig. 2A),

and the reactivity was restricted to the area around pancreaticducts, which have previously been shown to stain positive.Figure 2C shows morphometric analysis of tumors derivedfrom the T-HPNE and T-HPNE.SHH cells and indicates a signi-ficant increase in desmoplasia in the tumors derived from theT-HPNE.SHH cells.

We stained adjacent sections with collagen I and fibronectinto determine if the myofibroblasts synthesized these two well-characterized components of desmoplasia. Figure 3A showsimmunohistochemical staining of tumors derived from theT-HPNE.SHH cell line and confirms that the area containingmyofibroblasts and desmoplasia also stained positive forfibronectin and collagen I. Moreover, there were few myofi-broblasts in the tumors derived from the T-HPNE cell line(Fig. 3A), and immunohistochemical analysis of these sectionsshowed an absence of collagen I and fibronectin.

In Fig. 3B, we confirmed the presence of collagen I andfibronectin by immunohistochemistry to stain for these

Fig. 2. SHH promotes desmoplasia in an orthotopic model ofpancreatic cancer. A and B, histologic analysis of orthotopicpancreatic tumors derived from theT-HPNE.SHH cell line.Tumor sectionswere stained for themarker SMA to determine ifthere was a difference in the amount of SMA+ desmoplasia inthe SHH-expressing tumor cells. Brown, positive staining forSMA. A, peripheral sections ofT-HPNE andT-HPNE.SHHpancreatic tumors, at�40 magnification, representing the largedesmoplastic reaction in the periphery of theT-HPNE.SHHorthotopic tumors (note black arrows). Infiltrating SMA+ cellsare also indicated with a black arrow. B, immunohistochemicalanalysis with SMA to stain for myofibroblasts in theT-HPNE andT-HPNE.SHH tumor sections; �200 magnification. Brown,positive staining; black arrows, SMA+ cells. C, morphometricanalysis of desmoplasia inT-HPNE versusT-HPNE.SHHorthotopic tumors. Analysis was done by quantifying SMA+

areas on the tumor sections relative to the total cellularcontent and expressing the results as a percentage:(SMA+ desmoplasia / total area) � 100.Tumors derivedfrom theT-HPNE.SHH cell line showed significantly moredesmoplasia than the tumors derived from theT-HPNE cells(n = 13 per group).





components in tumors derived from untreated mice challengedwith the Capan-2 cell line. There is a decrease in both theintensity and quantity of both collagen I and fibronectin stainingin the Capan-2 tumors derived from the mice treated with 5E1.

Myofibroblasts express Gli-1. We sought to determine ifSHH was stimulating myofibroblasts by a paracrine mechanism.Figure 4A shows immunohistochemical analysis of humanpancreatic cancer tissue obtained from autopsy, showing that

Fig. 3. Expression of collagen I and fibronectin in orthotopic tumor sections. A, representative immunohistochemical analysis ofT-HPNE andT-HPNE.SHH tumor sections todetermine if fibronectin and collagen I are expressed in either of the tumor sections. Areas of desmoplasia were identified by staining for SMA (described in Figs.1and 2).Evidence that the SMA+ areas were expressing fibronectin and collagen I is indicated by brown staining in the areas of the tumor sections designated desmoplasia in theT-HPNE.SHH tumor sections.We did not observe positive staining in tumors derived from theT-HPNE cell line. H&E staining of the tumor sections is also shown. B,representative immunohistochemical staining for collagen I and fibronectin in the Capan-2 orthotopic tumors with no treatment, treated with 4E11, or treated with 5E1.There was a decrease in the intensity and amount of staining for both collagen I and fibronectin in the tumor sections treated with 5E1when compared with the controls.





tumor cells stain positive for SHH, whereas the surroundingfibroblasts and areas of desmoplasia were positive for Gli-1,whose expression is known to be induced following stimulationwith SHH. These data support the hypothesis that fibroblaststhat make up the desmoplastic reaction in human pancreaticcancer show evidence of paracrine signaling in response to SHHproduced by human pancreatic cancer cells.

We also analyzed adjacent sections of T-HPNE.SHH tumorswith antibodies to SMA and Gli-1, which showed evidence thatSMA+ myofibroblasts expressed Gli-1, supporting the hypothesisthat these cells were responding to SHH produced by theT-HPNE-SHH cells in a paracrine manner (Fig. 4B).

Myofibroblasts and SMA+ cells are of murine origin. We useda human-specific antibody to a ribonuclear protein and anantibody that binds to both murine and human SMA todetermine if the SMA+ cells were of human or murine origin.Figure 5A presents confocal analysis of expression of thesemarkers in orthotopic tumors derived from human T-HPNE,T-HPNE.SHH, Capan-2, and 5E1-treated Capan-2 tumors. Themajority of these cells were clearly of murine origin and showedmyofibroblastic morphology.

SHH promotes the differentiation and proliferation of pan-creatic stellate cells. Confocal analysis of the 5E1-treatedCapan-2 tumors (Fig. 5A) revealed single SMA+ cells embedded

in the human tumors, raising the possibility that these werestellate cells that were progenitors to the myofibroblastsobserved in Capan-2 tumors when SHH signaling was notinhibited (Fig. 5A). Capan-2 tumors were stained for cyclin D1,which was detected in the myofibroblasts, supporting the hypo-thesis that these represented stellate cells within the tumor thatwere proliferating in response to SHH stimulation (Fig. 5B).

We isolated pancreatic stellate cells from the human pancreasusing previously described techniques (7, 10) and confirmedtheir lineage by PCR to identify markers of both epithelial andmesenchymal lineage. Consistent with the anticipated pheno-type for pancreatic stellate cells, the isolated cells expressedhigher levels of SMA and vimentin and displayed a spindle-like morphology (Supplementary Fig. S2). We stimulated thesecells with rhSHH at 1 and 10 Ag/mL for 24 hours and observedan increase in the mesenchymal markers SMA, vimentin, anddesmin and a decrease in the epithelial marker cytokeratin 19(Fig. 5C). These data are the first to suggest that SHH induces thedifferentiation of pancreatic stellate cells into myofibroblasts.

SHH in vitro promotes the invasion of human pancreaticmyofibroblasts. We analyzed the ability of human myofibro-blasts from the pancreas to respond to SHH in Matrigelinvasion assays to determine if SHH contributes to increaseddesmoplasia by enhancing the motility and invasive capacity of

Fig. 4. SHH paracrine signaling betweentumor and stroma. A, representativeimmunohistochemical analysis of primarypancreatic ductal adenocarcinoma tissuesections from patients with pancreatic cancer.Brown color, positive staining for Gli-1orSHH.We observed positive staining forSHH in the tumor cells, whereas there wasminimal staining in the tumor cells for Gli-1(black arrows, positive staining;white arrows,negative staining).We observed minimalstaining for Gli-1in the tumor cells andobserved reactivity with the Gli-1antibody inareas of desmoplasia. B, immunohistochemicalanalysis ofT-HPNE.SHH tumors, showing bothSMA and Gli-1staining; �200 magnification.The brown staining in adjacent sectionsindicates that the SMA+ myofibroblastsinfiltrating theT-HPNE.SHH tumor sectionare expressing Gli-1 (n = 3).





these cells. Human pancreatic myofibroblasts, which expressthe markers vimentin and SMA and lack expression ofcytokeratin 19 (Supplementary Fig. S3A), were isolated andshown to express Gli-1 when stimulated with SHH (Supple-mentary Fig. S3B). Figure 6A shows the results of a Matrigelinvasion assay in which increasing numbers of T-HPNE andT-HPNE.SHH cells were seeded in the lower chamber andmyofibroblasts were seeded in the upper chamber. For culturescontaining 250,000 and 500,000 T-HPNE cells expressingSHH, there was a significant increase in the invasive capacityof myofibroblasts as compared with T-HPNE controls. Werepeated this experiment with recombinant SHH in the lowerchamber (rhSHH, R&D Systems). Figure 6B shows that therewas a concentration-dependent and significant increase in theability of the myofibroblasts to invade through a Matrigelmembrane in response to exposure to rhSHH.

We also conducted an in vitro wound healing assay todetermine if SHH increased the ability of the myofibroblasts ina monolayer to migrate into a wound. Figure 6C and D showsthe visual and quantitative results from this experiment. After48 hours, SHH significantly increased the ability of themyofibroblasts to recover from the wound (90% recovery withSHH versus 45% recovery in the control cells and cells withSHH and the 5E1 SHH inhibitor). These experiments supportthe hypothesis that SHH increases the migration and invasionof human pancreatic myofibroblasts in vitro.

Discussion

Hedgehog signaling was first identified and characterized inDrosophila and has been shown to play a role in organogenesisand differentiation. During vertebrate patterning, hedgehog

Fig. 5. SHH enhances pancreatic stellate cell proliferation and differentiation and enhances pancreatic myofibroblast invasion. A, confocal microscopy ofT-HPNE andT-HPNE.SHH tumors (as indicated) using antihuman ribonuclear protein (blue) and anti-SMA (green). Differential interference contrast microscopy (DIC) plus overlayshows that in theT-HPNE.SHH tumors, there are cells positive for SMA but not human ribonuclear protein, which indicates that the majority of the myofibroblasts are ofmurine origin. Confocal analysis of anti ^ ribonuclear protein (RNP) or SMA expression on Capan-2 control tumors and tumors treated with the 5E1inhibitor (as indicated).Blue, antihuman ribonuclear protein; green, SMA. In the 5E1-teated Capan-2 tumors, isolated SMA+ cells can be identified as green in the overlay. B, serial sectionsshowing SMA and cyclin D1staining of SMA+ myofibroblasts in Capan-2 tumors (brown stain). C, real-time PCR analysis on markers of mesenchymal and epithelialdifferentiation done on pancreatic stellate cells after 24-h stimulation with recombinant human SHHat either1or10 Ag/mL. SHH stimulation increased the expressionof desmin, vimentin, and SMA and decreased the expression of cytokeratin19.





signaling induces differentiation of the skeleton, centralnervous system, early gut endoderm, limb axes and ventralsomite buds (13, 14). Interestingly, SHH has not been shownto contribute to normal differentiation of the pancreas, andectopic expression in the epithelia of the embryonic pancreaticmesoendoderm leads to malformation of the pancreaticmesoderm and an increase in the number of mesenchymalcells in the developing pancreatic compartment (4).

There are three known human hedgehog family members:Sonic (SHH), Indian (Ihh), and Desert (Dhh). After translationof the 45-kDa protein, the hedgehog proteins undergo anautocatalytic cleavage facilitated by a cholesterol modification,an event that yields a 19-kDa NH2-terminal protein that issecreted and a 25-kDa COOH-terminal protein that remainsassociated with the cell (15, 16). The secreted ligand binds to a12-pass transmembrane protein, Patched (Ptch), which relievesthe inhibitory effect of Patched on another transmembraneprotein, Smoothened (SMO; reefs. 15–17), which in turnenables hedgehog signaling. The complex signaling cascadedownstream of SMO activation is not well characterized, yet

ultimately SHH transduces a signal that induces nucleartranslocation of Gli, a target transcription factor, and thetranscriptional activation of hedgehog target genes, includingPtch, Gli , and hedgehog interaction protein, Hhip (18, 19).

Several reports have implicated the misregulation of thehedgehog signaling pathway in the initiation and progressionof pancreatic cancer (5, 20). Mutations in this pathway areassociated with cancer initiation and progression, as mutationsin the tumor suppressor Patched are characterized in basalcell carcinoma, medulloblastoma, and rhabdomyosarcoma,whereas oncogenic mutations in the downstream transcrip-tion factor Gli have been identified in glioblastoma and basalcell carcinoma (21, 22). Pancreatic cancer presents a differentoncogenic insult within the hedgehog pathway, as SHH is up-regulated in the tumor cells but absent in the normal pancreas.

We observed that expression of SHH was correlated withdesmoplasia in two cell lines grown orthotopically in nudemice. Desmoplasia, a common and well-noted histologicfeature of pancreatic cancer, has been proposed to contributeto progression of the disease (23); however, the molecular

Fig. 6. SHH increases the in vitro invasive capacity of primary cells from the human pancreas. A, invasion assay in Matrigel invasion chambers.T-HPNE cells orT-HPNE.SHHcells were seeded in the lower chambers at densities of 50,000, 250,000, and 500,000 per well.With 250,000 and 500,000 cells, SHH secreted from theT-HPNE.SHH cellline significantly increased the invasive potential of an immortalized humanpancreatic myofibroblast cell line (*,P < 0.05). Columns, meannumber of invasive cells for each cellline (n = 3); bars, SE. B, invasion assay in Matrigel invasion chambers. Recombinant SHH in the lower chamber of the invasion assay significantly increased the invasivepotential of an immortalized humanpancreatic myofibroblast cell line (**, P < 0.01). Columns, meannumber of invasive cells for each cell line (n = 3); bars, SE.C andD, in vitrowound healing assay. C, pictures of initial time point controls and 48-h time points showing the effects of SHH stimulation on pancreatic myofibroblasts.D, quantitativeanalysis of the in vitro wound healing assay (percent closure). SHH-stimulated immortalized human pancreatic myofibroblasts showed a significant increase in their abilityto recover from the wound (*, P = 0.05).The addition of the 5E1blocking antibody significantly inhibited the SHH-mediated wound healing ability back to control levels(*, P = 0.05).Values are shown for each cell line; bars, SD.





processes that cause the desmoplastic reaction are not wellcharacterized. Previous experiments in prostate cancer havedetermined that SHH contributes to desmoplasia and tumorgrowth (24); thus, we propose that SHH has a similarmechanistic role in pancreatic cancer.

We explored the requirement for SHH in the induction ofdesmoplasia by administering a SHH-neutralizing antibody,5E1, to mice with Capan-2 cells injected orthotopically intotheir pancreas. Capan-2 tumors express SHH and produce adesmoplastic reaction when grown orthotopically in mice.Capan-2 tumors treated with the SHH-inhibiting antibody weresmaller and did not elicit a desmoplastic response whencompared with the control tumors derived from the Capan-2cell line that received no treatment or an isotype controlantibody. Similarly, SHH expression in T-HPNE cells (T-HPNE.SHH) produced larger tumors that showed significantdifferences in cellular morphology. Within the tumor sectionsof the T-HPNE.SHH cell line, we noted the presence ofinfiltrating fibroblasts as long, extended cells with elongatednuclei. This fibrotic infiltration was accompanied by expressionof collagen I and fibronectin, and represented further evidenceof a desmoplastic reaction within tumors from the SHHoverexpressing transformed cell line (25, 26). This reactionwas not observed in the tumor sections from the T-HPNEcell line.

We furthered evaluated the effects of SHH signaling in vitroon primary pancreatic myofibroblasts and found that SHHacted as a chemoattractant. This result complements previousstudies, which showed that cancer cells recruit fibroblasts, andprovides the first reported evidence that SHH, secreted fromtransformed pancreatic epithelia, contributes to the recruitmentand organization of fibroblasts to enhance tumor progression(27). We also show that SHH can induce the differentiation and

proliferation of human pancreatic stellate cells, which can alsobe a mechanism whereby SHH induces desmoplasia.

In summary, our data are significant in that they identifySHH as a regulator of one critical pathologic outcome ofpancreatic ductal adenocarcinoma: the production of desmo-plasia, or the infiltration of fibrous tissue into neoplasms,which results from the proliferation of fibroblasts and theincrease in fibrosis within the stroma (28–30). Desmoplasiais a common histologic feature of many adenocarcinomas(31, 32). The stroma, or the complex microenvironment sur-rounding tumor cells, includes inflammatory cells, fibroblasts,extracellular matrix, and small blood and lymphatic vessels(33, 34). Previous experiments have shown that TGFh is amediator of fibrosis (12) in pancreatic cancer. In data notshown, we conducted a microarray on T-HPNE versus T-HPNE.SHH cells and determined that the expression of SHHdid not up-regulate the expression of TGFh. Thus, it is possiblethat SHH directly influences the desmoplastic reaction, but itremains possible that SHH acts by inducing signals that interactwith TGFh signaling pathways (35). We have also shown thatSHH can act as a chemoattractant for myofibroblasts ininvasion assays and in wound healing assays, where SHHincreased the ability of pancreatic myofibroblasts to recoverfrom a wound. These data suggest that SHH is a critical playerin the tumor microenvironment. Further elucidation of itsfunctions will lead to a better understanding of the patho-genesis of this lethal disease and the potential for therapeutictargeting of SHH to disrupt the microenvironment in pan-creatic cancer.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.



References1. Jemal A,Tiwari RC, MurrayT, et al. Cancer statistics,2004. CACancerJClin 2004;54:8^29.

2.Yeo CJ, CameronJL, Lillemoe KD, et al. Pancreatico-duodenectomy with or without distal gastrectomyand extended retroperitoneal lymphadenectomy forperiampullary adenocarcinoma. Part 2: randomizedcontrolled trial evaluating survival, morbidity, andmortality. Ann Surg 2002;236:355^66; discussion366^8.

3.Hruban RH, GogginsM, ParsonsJ, Kern SE.Progres-sion model for pancreatic cancer. Clin Cancer Res2000;6:2969^72.

4. Kayed H, Kleeff J, OsmanT, Keleg S, Buchler MW,Friess H. Hedgehog signaling in the normal and dis-eased pancreas. Pancreas 2006;32:119^29.

5.Thayer SP, di Magliano MP, Heiser PW, et al. Hedge-hog is an early and late mediator of pancreatic cancertumorigenesis. Nature 2003;425:851^6.

6. Pasca di Magliano M, Sekine S, ErmilovA, Ferris J,Dlugosz AA, Hebrok M. Hedgehog/Ras interactionsregulate early stages of pancreatic cancer. Genes Dev2006;20:3161^73.

7. Lee KM, Nguyen C, Ulrich AB, Pour PM, OuelletteMM. Immortalization with telomerase of the Nestin-positive cells of the human pancreas. Biochem Bio-phys Res Commun 2003;301:1038^44.

8. Campbell PM, Groehler AL, Lee KM, Ouellette MM,Khazak V, Der CJ. K-Ras promotes growth trans-formation and invasion of immortalized human pan-creatic cells by Raf and phosphatidylinositol 3-kinasesignaling. Cancer Res 2007;67:2098^106.

9. Tsutsumida H, Swanson BJ, Singh PK, et al. RNAinterference suppression of MUC1 reduces thegrowth rate and metastatic phenotype of humanpancreatic cancer cells. Clin Cancer Res 2006;12:2976^87.

10. LawsonT, Ouellette M, Kolar C, Hollingsworth M.Culture and immortalization of pancreatic ductal epi-thelial cells. Methods Mol Med 2005;103:113^22.

11. EricsonJ,Morton S, Kawakami A, Roelink H, JessellTM.Two critical periods of Sonic Hedgehog signalingrequired for the specification of motor neuron identity.Cell 1996;87:661^73.

12. Lohr M, Schmidt C, Ringel J, et al. Transforminggrowth factor-h1 induces desmoplasia in an experi-mental model of human pancreatic carcinoma. CancerRes 2001;61:550^5.

13. Echelard Y, Epstein DJ, St-Jacques B, et al. Sonichedgehog, a member of a family of putative signalingmolecules, is implicated in the regulation of CNSpolarity. Cell 1993;75:1417^30.

14. Roelink H, Augsburger A, Heemskerk J, et al. Floorplate and motor neuron induction by vhh-1, a verte-brate homolog of hedgehog expressed by the noto-chord. Cell 1994;76:761^75.

15. PorterJA,Young KE, Beachy PA. Cholesterol modi-fication of hedgehog signaling proteins in animal de-velopment. Science1996;274:255^9.

16. Porter JA, Ekker SC, ParkWJ, et al. Hedgehog pat-terning activity: role of a lipophilic modification medi-ated by the carboxy-terminal autoprocessing domain.Cell 1996;86:21^34.

17. MarigoV, Davey RA, ZuoY, Cunningham JM,TabinCJ. Biochemical evidence that patched is the Hedge-hog receptor. Nature1996;384:176^9.

18. Lipinski RJ, Gipp JJ, Zhang J, Doles JD, BushmanW. Unique and complimentary activities of the Glitranscription factors in Hedgehog signaling. Exp CellRes 2006;312:1925^38.

19. Ingham PW, McMahon AP. Hedgehog signaling inanimal development: paradigms and principles. GenesDev 2001;15:3059^87.

20. Kayed H, Kleeff J, Keleg S, et al. Indian hedgehogsignaling pathway: expression and regulation in pan-creatic cancer. Int JCancer 2004;110:668^76.

21. ZedanW, Robinson PA, Markham AF, High AS. Ex-pression of the Sonic Hedgehog receptor ‘‘PATCHED’’in basal cell carcinomas and odontogenic keratocysts.JPathol 2001;194:473^7.

22. Pasca di MaglianoM, HebrokM. Hedgehog signal-ling in cancer formation and maintenance. Nat RevCancer 2003;3:903^11.

23. Korc M. Pancreatic cancer-associated stroma pro-duction. AmJSurg 2007;194:S84^6.

24. Fan L, Pepicelli CV, Dibble CC, et al. Hedgehogsignaling promotes prostate xenograft tumor growth.Endocrinology 2004;145:3961^70.

25.Hauptmann S, Zardi L, Siri A, et al. Extracellular ma-trix proteins in colorectal carcinomas. Expression oftenascin and fibronectin isoforms. Lab Invest 1995;73:172^82.

26. Hanamura N,YoshidaT, Matsumoto E, KawaradaY,SakakuraT. Expression of fibronectin and tenascin-C





mRNA by myofibroblasts, vascular cells and epithelialcells in human colon adenomas and carcinomas. Int JCancer1997;73:10^5.

27. Ishii G, SangaiT, ItoT, et al. In vivo and in vitro char-acterization of human fibroblasts recruited selectivelyinto human cancer stroma. Int J Cancer 2005;117:212^20.

28. Mollenhauer J, Roether I, Kern HF. Distribution ofextracellular matrix proteins in pancreatic ductal ade-nocarcinoma and its influence on tumor cell prolifera-tion in vitro. Pancreas1987;2:14^24.

29. Kuniyasu H, Abbruzzese JL, Cleary KR, Fidler IJ.Induction of ductal and stromal hyperplasia by basic

fibroblast growth factor produced by human pancre-atic carcinoma. Int J Oncol 2001;19:681^5.

30. Kloppel G, Lingenthal G, von Bulow M, Kern HF.Histological and fine structural features of pancreaticductal adenocarcinomas in relation to growth andprognosis: studies in xenografted tumours and clin-ico-histopathological correlation in a series of 75cases. Histopathology1985;9:841^56.

31. Hartel M, Di Mola FF, Gardini A, et al. Desmoplasticreaction influences pancreatic cancer growth beha-vior.WorldJ Surg 2004;28:818^25.

32. Desmouliere A, Guyot C, Gabbiani G. The stromareaction myofibroblast: a key player in the control

of tumor cell behavior. Int J Dev Biol 2004;48:509^17.

33. Bardeesy N, DePinho RA. Pancreatic cancer biolo-gy and genetics. Nat Rev Cancer 2002;2:897^909.

34. Bachem MG, Schunemann M, Ramadani M, et al.Pancreatic carcinoma cells induce fibrosis by stimulat-ing proliferation and matrix synthesis of stellate cells.Gastroenterology 2005;128:907^21.

35.YooYA, Kang MH, KimJS, Oh SC. Sonic hedgehogsignalingpromotes motility and invasiveness of gastriccancer cells through TGF-h-mediated activation ofthe ALK5-Smad 3 pathway. Carcinogenesis 2008;29:480^90.



2008;14:5995-6004. Clin Cancer Res Jennifer M. Bailey, Benjamin J. Swanson, Tomofumi Hamada, et al. CancerSonic Hedgehog Promotes Desmoplasia in Pancreatic

Updated version

http://clincancerres.aacrjournals.org/content/14/19/5995

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2012/03/07/14.19.5995.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/14/19/5995.full#ref-list-1

This article cites 35 articles, 7 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/14/19/5995.full#related-urls

This article has been cited by 50 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] at

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/14/19/5995To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/content/suppl/2012/03/07/14.19.5995.DC1

http://clincancerres.aacrjournals.org/content/14/19/5995.full#ref-list-1

http://clincancerres.aacrjournals.org/content/14/19/5995.full#related-urls

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

mailto:[email protected]



Documents

SonicHedgehogPromotesDesmoplasiainPancreaticCancerTo identify SHH in the primary human tumor sections, rabbit anti-SHH primary antibody (1:200; Santa Cruz)was incubated on paraffin-embedded