IL17 Functions through the Novel REG3 STAT3 …...Molecular and Cellular Pathobiology IL17 Functions through the Novel REG3b– JAK2–STAT3 Inﬂammatory Pathway to Promote the Transition

Molecular and Cellular Pathobiology

IL17 Functions through the Novel REG3b–JAK2–STAT3 Inflammatory Pathway to Promotethe Transition from Chronic Pancreatitis toPancreatic CancerCeline Loncle1, Laia Bonjoch2, Emma Folch-Puy2, Maria Belen Lopez-Millan1, Sophie Lac1,Maria In�esMolejon1, EduardoChuluyan3, PierreCordelier4, PierreDubus5,GwenLomberk6,Raul Urrutia6, Daniel Closa2, and Juan L. Iovanna1

Abstract

Pancreatic ductal adenocarcinoma (PDAC) offers an optimalmodel for discovering "druggable" molecular pathways thatparticipate in inflammation-associated cancer development.Chronic pancreatitis, a common prolonged inflammatory dis-ease, behaves as a well-known premalignant condition thatcontributes to PDAC development. Although the mechanismsunderlying the pancreatitis-to-cancer transition remain to befully elucidated, emerging evidence supports the hypothesisthat the actions of proinflammatory mediators on cells har-boring Kras mutations promote neoplastic transformation.Recent elegant studies demonstrated that the IL17 pathwaymediates this phenomenon and can be targeted with antibo-dies, but the downstream mechanisms by which IL17 functionsduring this transition are currently unclear. In this study, we

demonstrate that IL17 induces the expression of REG3b, a well-known mediator of pancreatitis, during acinar-to-ductal meta-plasia and in early pancreatic intraepithelial neoplasia (PanIN)lesions. Furthermore, we found that REG3b promotes cellgrowth and decreases sensitivity to cell death through activa-tion of the gp130-JAK2-STAT3-dependent pathway. Geneticinactivation of REG3b in the context of oncogenic Kras-drivenPDAC resulted in reduced PanIN formation, an effect thatcould be rescued by administration of exogenous REG3b.Taken together, our findings provide mechanistic insight intothe pathways underlying inflammation-associated pancreaticcancer, revealing a dual and contextual pathophysiologic rolefor REG3b during pancreatitis and PDAC initiation. Cancer Res;75(22); 4852–62. �2015 AACR.

IntroductionPancreatic ductal adenocarcinoma (PDAC) has been recog-

nized by the scientific community, advocacy groups, and govern-ment agencies as an important national health priority because ofits physically and morally painful impact and dismal outcome.

Interestingly, in the recent past, most research efforts have pri-marily focused on how genetic and epigenetic alterations lead tothe aberrant activation of key oncogenes and inactivation oftumor suppressor pathways to consequently confer the transform-ingpancreatic cellswith growth and survival advantages. Themostcommon genetic abnormality in pancreatic cancer is oncogenicKRAS mutation, which is the initial key event for the initiationphase of pancreatic carcinogenesis. However, mutation of KRASalone is not sufficient for frank cancer progression, but ratheradditional aberrations, such as inactivation of tumor suppressorgenes and signals from the tumormicroenvironment, are requiredfor tumor promotion and progression. In this regard, pancreaticcancer that originates in the setting of inflammation (chronicpancreatitis) offers an optimal model to study these events.Chronic pancreatitis is a known premalignant disease with a53-fold lifetime cumulative risk of developing pancreatic cancer(1). Notably, oncogenic KRAS mutations are also found in thisdisease and are believed to contribute to its transformation intocancer. In fact, emerging data indicate that proinflammatorymediators can act on pancreatic cells carrying KRAS mutation tocomplete their process of neoplastic transformation throughmodulation of differentiation, growth, survival, and senescence.

In this regard, our laboratory has focused on characterizingdruggable proinflammatory pathways that function as mediatorsof the pancreatitis–cancer transition. The current study, therefore,focuses on characterizing the function of REG3b, one of the bestknown pancreatitis mediators, in the initiation of pancreatic

1Centre de Recherche en Canc�erologie de Marseille (CRCM), INSERMU1068, CNRS UMR 7258, Aix-Marseille Universit�e and Institut Paoli-Calmettes, Parc Scientifique et Technologique de Luminy, Marseille,France. 2Experimental Pathology Department, IIBB-CSIC-IDIBAPS,Barcelona, Spain. 3Laboratory of Immunomodulators, School of Med-icine, Centro de Estudios Farmacol�ogicos y Bot�anicos (CEFYBO),ConsejoNacional de InvestigacionesCientíficasyTecnol�ogicas (CON-ICET)-University of Buenos Aires, Buenos Aires, Argentina. 4INSERMUMR U1037, Centre de Recherche sur le Cancer de Toulouse, CHURangueil, Toulouse, France. 5EA2406, Histologie et pathologiemol�eculaire des tumeurs, Universit�e de Bordeaux, Bordeaux, France.6Laboratory of Epigenetics and Chromatin Dynamics, Gastroenterol-ogy Research Unit, Departments of Biochemistry and Molecular Biol-ogy, Biophysics, and Medicine, Mayo Clinic, Rochester, New York.

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

C. Loncle and L. Bonjoch contributed equally to this article.

Corresponding Author: Juan L. Iovanna, INSERM U1068, 163, Av de Luminy,Marseille 13288, France. Phone: 33-491-828803; Fax: 33-491-82886083; E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-15-0896

�2015 American Association for Cancer Research.

CancerResearch

Cancer Res; 75(22) November 15, 20154852

on February 10, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-15-0896

http://cancerres.aacrjournals.org/

cancer by KRAS. This molecule, also known as pancreatitis-asso-ciated protein (PAP), p23, or hepatocarcinoma-intestine-pancre-as protein, was originally discovered in the pancreatic juice of ratswith acute pancreatitis, butwas absent in the pancreatic juice fromhealthy rats (2). The PAP/REG3b gene and its mRNA weresubsequently cloned from several species (3–10), indicating thatREG3b is an evolutionarily conserved gene. REG3b expression isfound in a limited number of healthy tissues, such as in Panethcells of the small intestine (11), in luminal epithelial cells of theuterus in estrus (12), in alpha cells of the Langerhans islets (13),and in somatotropic cells of the pituitary gland (14). In contrast,REG3b is overexpressed in various diseased tissues, such as thepancreas with acute pancreatitis (3, 15), transformed hepatocytes(9), the brain with Alzheimer disease (16), regenerating moto-neurons (17), the brain in response to a traumatic injury (18),inflamed (19, 20) and transformed epithelial colonic cells (21),cholangiocarcinoma cells (13, 22), regenerating islet beta cells(23), the myocardium of rats with decompensated pressure-overload hypertrophy (24), pheochromocytoma cells (25), blad-der cancer cells (26), andpsoriatic keratinocytes (27). Structurally,REG3b is a 16-kDa secretory protein related to C-type lectins,although a classical lectin-related function has not been reportedyet, apart from a study suggesting that REG3b binds to bacterialproteoglycans (28). Moreover, several pro- and anti-inflamma-tory cytokines are able to induce REG3b expression, which canalso be self-induced through the canonical JAK2/STAT3-depen-dent pathway (29). Thus, based on this knowledge, we predictthat REG3b may mediate procarcinogenic effects downstream ofproinflammatorypathwayswith known transforming abilities. Totest the validity of this hypothesis,we investigatedwhether REG3bmodulates the neoplastic effects of the IL17A pathway. This ideawas recently stimulated by elegant studies, which demonstratedan essential role of IL17A in the development of PDAC afterinduction via activatedoncogenicKras (30). In fact, in vivo gain-of-function and loss-of-function studies from these investigationsboth demonstrated that IL17A promotes the initiation and pro-gression of PDAC. In vivo neutralization of IL17A led to a rapidchange in the gene expression program of PDAC cells, includingthe loss of IL6 expression and STAT3 activation, twokey regulatorsof PDAC development (30). These results suggest that signalsthrough the IL17RA can alter pancreatic cell transformation.However, the full repertoire of molecular mediators that workdownstream of IL17A to modulate the inflammation-to-cancertransition remain to be defined.

In the current study, we report for the first time that REG3b isactivated by the IL17A in pancreatic epithelial cells with KrasG12D

mutations. In addition, we observe that REG3b expression isinduced during acinar-to-ductal metaplasia (ADM) and in earlypancreatic intraepithelial neoplasia (PanIN) lesions, which arethe histopathologic hallmarks of pancreatic cancer initiation.Moreover, in this process, REG3b primarily acts as a paracrinefactor to stimulate the development of PanIN downstream ofIL17RA. This effect of REG3b reflects its ability to promote cellgrowth and decreases sensitivity to apoptosis by coupling to thegp130-JAK2-pSTAT3 signaling pathway. Furthermore, REG3bboosts interactions between epithelial cells and immune cells,triggering the expression of some mediators, such as IL10 andTGFb, and activating the CXCL12/CXCR4 axis. Finally, wedescribe results from two antibody-based preclinical trials, whichdemonstrate the efficacy of neutralizing either IL17 or REG3b inantagonizing KRAS-induced pancreatic cancer initiation. Thus, in

addition to providing valuable mechanistic information, thisstudy offers new knowledge of significant biomedical relevance,which should be taken into consideration for the future design ofstrategies aimed at preventing inflammation-associated pancre-atic cancer development.

Materials and MethodsMice

Descriptions of the relevant strains Pdx1-cre;LSL-KrasG12D,Pdx1-cre;LSL-KrasG12D;Ink4a/Arffl/fl, and REG3b�/� can be foundelsewhere (31–33). Because animals originate from differentgenetic backgrounds, we systematically relied on littermate con-trol and experimental mice. Pancreatitis was induced through theintraperitoneal administration of cerulein (Sigma) at 250 mg/kgbody weight twice daily for one week. The antibody anti-IL17RAwas purchased from R&D Systems and anti-REG3b was obtainedfrom Dynabio S.A.; both were intraperitoneally injected twiceweekly. Recombinant REG3b was obtained from Dynabio S.A.and intraperitoneally injected daily. Mice were kept in the Exper-imental Animal House of the Centre de Canc�erologie deMarseille(CRCM), pole Luminy, following institutional guidelines.

Cell cultureAR-42J, MiaPaCa2, Panc1, and THP-1 cells obtained from the

ATCC were maintained in DMEM (Invitrogen) supplementedwith 10% FBS at 37�C with 5% CO2. Recombinant IL17A waspurchased from R&D Systems; anti-STAT3 and anti-phopho-STAT3 antibodies and neutralizing anti-gp130 were obtainedfrom Cell Signaling. The JAK2 inhibitor AG-490 was purchasedfrom Sigma. THP-1 cells were differentiated from macrophagesthrough an initial incubation with 100 nmol/L phorbol12-myristate 13-acetate (PMA; Sigma) for 48 hours at 37�C in12-well plastic Petri dishes (Nunclon; Nunc Inc.). PMA wasremoved 24 hours before the experiments. For coculture experi-ments, THP-1 cells were seeded at the lower compartment ofthe Transwell system, whereas MiaPaCa2 cells were coculturedfor 24 hours in the physically separated upper chamber (Thin-cert 12-well plate cell culture inserts, 0.4 mm pore size, GreinerBio-One BVBA).

HistologyPancreatic sections were fixed in 4% paraformaldehyde and

embedded in paraffin. Hematoxylin and eosin (H&E) was per-formed using standard procedures. Sections were probed withprimary antibodies against REG3b and IL17RA. Samples wereexamined with a Nikon Eclipse 90i microscope.

Quantification of REG3b-positive lesions per tissueThe number of peritumoral acini (low- and high-grade PanIN)

is expressed as REG3b-positive lesions perfield andwas calculatedas follows: the number of REG3b-positive lesions, which weredetermined by IHC, was counted and a percentage was calculatedaccording to the total number of lesions.

Quantification of lesions per mouseThe number of lesions per field was counted, and lesion types

were classified on H&E-stained slides. Quantification representsthe average of 15 to 20 20� fields of view for 6 mice of eachgenotype as previously reported (34).

REG3b Plays a Key Role in IL17RA Protumoral Effect

www.aacrjournals.org Cancer Res; 75(22) November 15, 2015 4853




qRT-PCRRNAs were prepared from AR-42J, THP-1, and MiaPaCa2

cells using TRIzol Reagent (Life Technologies) and reversetranscribed using Go Script (Promega) according to the man-ufacturer's instructions. Quantitative RT-PCR was performed ina Stratagene cycler using Takara reagents. Primer sequences areavailable upon request.

ImmunoblottingProtein extraction was performed on ice with a total protein

extraction buffer: 50 mmol/L HEPES pH 7.5, 150 mmol/L NaCl,20% SDS, 1 mmol/L EDTA, 1 mmol/L EGTA, 10% glycerol, 1%Triton, 25 mmol/L NaF, 10 mmol/L ZnCl2, and 50 mmol/L DTT.Before lysis, a protease inhibitor cocktail was added at 1:200(Sigma-Aldrich, NUPR1340) composed of 500 mmol/L PMSF,1 mmol/L sodium orthovanadate, and 1 mmol/L b-glyceropho-sphate. Protein concentrations were measured using a BCA Pro-tein Assay Kit (Pierce Biotechnology). Protein samples (60 mg)were denatured at 95�C and subsequently separated by SDS-PAGE. After transfer to nitrocellulose, membrane blocking wasperformed with 1% BSA; samples were probed with the primaryantibody followed by a horseradish peroxidase-coupled second-ary antibody. Images were obtained with a Fusion FX imageacquisition system (Vilber Lourmat).

Cell viabilityThe PrestoBlue reagent was added and cell viability was esti-

mated after a 3-hour incubation time, following the PrestoBluecell viability reagent protocol provided by the supplier (LifeTechnologies).

Cell proliferationCell proliferation was assessed with a commercial kit (Cell

proliferation ELISA, BrdUrd; Roche Diagnostics) according to themanufacturer's instructions. Following a 24-hour incubationwiththe recombinant REG3b, cells were BrdUrd-labeled during 3hours at 37�C. Cells were fixed and incubated with a peroxi-dase-conjugated anti-BrdUrd antibody for 90 minutes at roomtemperature. Then, the peroxidase substrate 3,30,5,50-tetramethyl-benzidine was added, and BrdUrd incorporation was quantitatedbymeasuring the differences in absorbance at awavelength of 370minus 492 nm.

Caspase-3/7 activityTo induce cell death by starvation, human PDAC cells were

cultured in nonsupplemented Earle's Balanced Salt Solution(EBSS) medium (free of glucose, amino acids, lipids, and growthfactors). To detect caspase-3/7 activation, MiaPaCa2 and Panc1cells were plated in 96-well plates, starved, and treated withrecombinant REG3b in the presence and absence of either theanti-gp130 antibody or the JAK2 inhibitor and analyzed usingApoONEHomogeneous Caspase-3/7 Assay (Promega) accordingto the manufacturer's instructions. After 1 hour of incubation, afluorescence plate reader, using respectivewavelengths of 480 and535 nm was used to measure excitation and emission in thesamples with the caspase substrate.

Statistical analysisOne-way variance analysis was used to calculate the signifi-

cances of REG3b-positive lesions per tissue (percentage values),as previously described (34). To compare REG3b mRNA expres-sions, BrdUrd incorporation, and caspase-3/7 activity after treat-

ment with different concentrations of the REG3b, we usedthe one-way variance analysis; the P value was calculated usingDCt, as described by Yuan and colleagues (35). Two-way varianceanalysis was used to compare lesion numbers per field in KrasG12D

mice carrying the indicated REG3b (REG3bþ/þ or REG3b�/�) at14 weeks of age. Values are expressed as mean � SEM. All tests ofsignificance were two-tailed and the level of significance was set at0.05. The examined cell lines represent at least three independentexperiments. All statistical tests were performed using IBM SPSSstatistics 21.

ResultsThe inducible REG3b expression characterizes the transitionperiod between inflammation and cancer

Our laboratory has been critically involved in characterizing therole of REG3b in acute pancreatitis (32). Chronic pancreatitisresults from repeated episodes of acute pancreatitis, and thisprolonged inflammatory condition is accompanied by the con-tinued synthesis of REG3b. Thus, as chronic pancreatitis precedesPDAC, we hypothesized that REG3b may fulfill a previouslyunsuspected role in mechanistically linking both disease process-es. We were also guided by the recent discovery that the proin-flammatory cytokine IL17A promotes PDAC development andsought to investigate if REG3b is a mediator of this phenomenon.Consequently, we first performed IHC to examine the expressionof REG3b and IL17RA proteins in the pancreas of Pdx1-Cre;KrasG12D mice. Figure 1A shows that REG3b and IL17RA proteinsare barely detectable in normal pancreatic exocrine cells andwithin pancreatic islets, REG3b, but not IL17RA, is only evidentin the glucagon-producing cells. In the pancreas of Pdx1-Cre;KrasG12D animals, IL17RA is expressed in almost all acinar cells,reaches its highest levels in both ADM and early PanIN lesions,and is lost in more advanced PanIN lesions. Notably, we alsofound that the expression of REG3b in the exocrine pancreasfollows the same pattern as IL17RA.

Treatment of the AR-42J cultured acinar cell line with IL17Aincreases the levels of both REG3b mRNA and protein in a dose-dependent manner (Fig. 1B and C), revealing that they are part ofthe same pathway. Thus, both IL17RA and REG3b appear to beinduced by the oncogenic activation of KRAS to act early duringthe process of neoplastic transformation. Next, we sought to gaininsight into the causal relationship between IL17RA activation, inKRAS mutant carrying cells, and the induction of REG3b. For thispurpose, we neutralized IL17RA receptor activity in Pdx1-Cre;KrasG12D and Pdx1-Cre;Kraswt mice through the systemic admin-istration of a specific anti-IL17RA antibody and measured theexpression of the REG3b protein in the pancreas by Western blot.The result of this experiment, shown in Fig. 1D, demonstrates thatthe levels of REG3b protein is higher in the KrasG12D pancreascompared with the Kraswt pancreas. Moreover, Fig. 1D shows thatthe neutralization of IL17RA receptor in the pancreas of KrasG12D

mice almost completely inhibited the upregulation of REG3bprotein levels. Combined, these results suggest that, in the contextof oncogenic KRAS mutation, REG3b functions downstream ofthe IL17A.

Then, we examined whether the expression of REG3b is depen-dent on IL17RA activation in response to pancreatitis. Therefore,we repeatedly injectedmice with supramaximal doses of cerulein,which induces a strongpancreatic inflammation (36). IL17RAwasagain neutralized through the systemic injection of anti-IL17RA

Loncle et al.

Cancer Res; 75(22) November 15, 2015 Cancer Research4854




antibodies. Interestingly, the expression of REG3b in the KrasG12D

positive pancreas seemed almost completely dependent onIL17RA activation (Fig. 1D). However, the expression of REG3bis not dependent on IL17RA in the pancreas with pancreatitis, asIL17RA immunoneutralization does not change the expression ofREG3b. However, IL17RA expression is induced by pancreatitis ina REG3b-independent manner (Fig. 1E). Thus, we hypothesizethat increased levels of REG3b, likely caused by various inflam-matory mediators, accompany the development and progressionof pancreatitis until initiating oncogenic KRAS mutation triggersthis event to be exclusively dependent on IL17A signaling.

REG3bpromotes KRAS-mediated PDAC initiation in bothmiceand humans

To investigate the role of REG3b in the formation of KRAS-induced PanIN lesions, we crossed REG3b�/� mice with Pdx1-Cre;KrasG12D animals. Control Pdx1-Cre;KrasG12D;REG3bþ/þ

mice exhibited ADM, as well as low- and high-grade PanIN, at14 weeks (Fig. 2; average lesions per type and power field wereADM¼ 8.3 and PanIN 1a¼ 6.4, PanIN 1b ¼ 5.9, PanIN 2 ¼ 2.3,and PanIN 3 ¼ 0.6, n ¼ 6). In contrast, Pdx1-Cre;KrasG12D;REG3b�/� animals displayed a reduced number of both ADMand PanIN lesions (Fig. 2; average lesions per type and field wereADM¼ 2.2 and PanIN 1a¼ 0.6, PanIN 1b ¼ 0.3, PanIN 2 ¼ 0.3,and PanIN 3 ¼ 0.1, n ¼ 6). These differences were statisticallysignificant for various genotypes and lesion types (P < 0.001).Thus, we next investigated whether exogenous administration ofrecombinant REG3b protein is able to rescue PanIN developmentin Pdx1-Cre;KrasG12D;REG3b�/� animals. We performed dailyintraperitoneal injections of 1 mg recombinant REG3b proteinbeginning at 5 weeks of age. The results of this experiment, shownin Fig. 3, demonstrate that the recombinant REG3b increases the

number of PanIN lesions in Pdx1-Cre;KrasG12D;REG3b�/� micecloser to the number found in Pdx1-Cre;KrasG12D;REG3bþ/þ

animals. The average lesions per type and field were ADM ¼

PanINs1a 1b 32

Lesio

ns p

er fi

eld

0

2

4

10

8

6

PanINs1a1b 32

Lesio

ns p

er fi

eld

0

2

4

10

8

6

Pdx1-Cre;KrasG12D

REG3β+/+ REG3β–/–

Pdx1-Cre;KrasG12D;REG3βββ+/+ Pdx1-Cre;KrasG12D;REG3ββ–/–A

B

Figure 2.REG3b expression is required for the KrasG12D oncogene to induce PanIN. A,images representative of the pancreas in Pdx1-cre;LSL-KrasG12D;REG3b�/� andPdx1-cre;LSL-KrasG12D;REG3bþ/þ mice. H&E staining. B, number of ADM andPanIN lesions per �20 tissue field in Pdx1-cre;LSL-KrasG12D;REG3b�/� andPdx1-cre;LSL-KrasG12D;REG3bþ/þ mice at 14 weeks of age. Means are denotedby horizontal bars. Differences between genotypes and lesion types weresignificant (P < 0.001).

1,000500100500

REG3

β m

RNA

expr

essio

n (fo

ld c

hang

e)

0

10

20

50

40

30

*

**

******

IL17A (ng/mL)

Vehicl

e

Vehicl

e

Vehicl

eIL17A

An�-IL 17RA

An�-IL17RA

B

REG3β

β-Tubulin

C

REG3β

β-Tubulin

Pdx1-Cre;KrasG12D

Pdx1-Cre;Kraswt

D

β-Tubulin

REG3β

IL17RA

REG3β–/– REG3β+/+ + an�-IL17RAREG3β+/+

Pancrea��s − + − + − +

E

LL

DD

AciniAcini

L

PanIN

ADM L

PanINADM

Cont

rol

IL17RAREG3βββ

Pdx1

-Cre

;Kra

sG12

D

A

7±3 2±1 3±1 2±2

Figure 1.Expression of REG3b and IL17RA in the pancreas. A, IHC detection of REG3b and IL17RA proteins in control and for 14-week-old Pdx1-Cre;KrasG12D mice. D, duct;L, Langerhans Islet. B, AR-42J cells were treated with increasing concentrations of recombinant IL17A and the expression of REG3b was measuredby qRT-PCR. C, AR-42J cells were treated with 500 ng/mL of recombinant IL17A and the expression of REG3b was measured by Western blotting. D, Pdx1-Cre;KrasG12D and Pdx1-Cre;Kraswt mice were treated with a systemic injection of anti-IL17RA antibody and the pancreatic expression of the REG3b protein wasmeasured byWestern blotting. E, pancreatitis was induced in REG3b�/�, REG3bþ/þ, and REG3bþ/þ treated with anti-IL17RA antibody through systematic injectionand the pancreatic expression of the REG3b protein was measured by Western blotting. � , P < 0.05; �� , P < 0.01; �� , P < 0.001.






2.8 and PanIN 1a ¼ 0.9, PanIN 1b ¼ 0.5, PanIN 2 ¼ 0.4, andPanIN 3 ¼ 0.2 (n ¼ 6) for Pdx1-Cre;KrasG12D;REG3b�/� micetreated with the vehicle; whereas average values were ADM ¼ 6.1and PanIN 1a¼ 3.8, PanIN 1b¼ 3.5, PanIN 2¼ 2.0, and PanIN 3¼ 1.6 (n ¼ 6) for Pdx1-Cre;KrasG12D;REG3b�/� mice treatedcontinually with recombinant REG3b. These differences werestatistically significant for various genotypes and lesion types(P < 0.01). Consequently, we conclude that the expression ofREG3b is necessary for PanIN development induced by theKrasG12D oncogene during the initiation period of pancreaticcarcinogenesis.

To determine whether the effect of REG3b on stimulatingKRAS-induced PanIN formation is dependent upon IL17RA sig-naling, we treated a cohort of Pdx1-Cre;KrasG12D;REG3b�/� micewith recombinant REG3b protein, although simultaneously inhi-biting IL17RA with a specific neutralizing antibody. The neutral-izing anti-IL17RA antibody had no significant influence on eitherthe number of lesions or the grade of PanIN promoted byrecombinant REG3b (Fig. 3; average lesions per type and fieldwere ADM¼ 7.6 and PanIN 1a¼ 4.9, PanIN 1b¼ 3.3, PanIN 2¼1.9, and PanIN 3¼ 1.7, n¼ 6). These differences were statisticallysignificant for Pdx1-Cre; KrasG12D;REG3b�/� mice treated withvehicle (P < 0.01), but no significant differences were observed forPdx1-Cre;KrasG12D;REG3b�/� mice that received the combinedtreatment of recombinant REG3b protein and neutralizing anti-IL17RA antibodies (P > 0.05). Therefore, recombinant REG3b isable to rescue the development of PanIN in Pdx1-Cre; KrasG12D;

REG3b�/�mice, suggesting that REG3b plays a paracrine effect onstimulating the formation of PanIN lesion by KRAS containingpancreatic cells.

Because the systemic administration of recombinant REG3bprotein rescues PanIN formation inREG3bdeficientmice,wenextsought to determine whether, conversely, the systemic neutrali-zation of either REG3b or IL17RA potentially inhibits PanINdevelopment in Pdx1-Cre; KrasG12D;REG3bþ/þmice. As shownin Fig. 4, the treatment of these mice with either anti-REG3b oranti-IL17RA antibodies (beginning at 5 weeks old and for 9consecutive weeks) had a significant effect on the developmentof PanIN lesions. Average lesions per type and field decreased asfollows: ADM¼ 7.6 to 2.1 and PanIN 1a¼ 5.8 to 1.7, PanIN 1b¼5.3 to 1.2, PanIN 2 ¼ 2.3 to 0.8, and PanIN 3 ¼ 1.6 to 0.3 foranimals treated with anti-IL17RA (n ¼ 6); and ADM ¼ 7.6 to 2.6andPanIN1a¼5.8 to 1.5, PanIN1b¼5.3 to 0.8, PanIN2¼2.3 to0.6, and PanIN 3¼ 1.6 to 0.2 for animals treated with anti-REG3bantibodies (n ¼ 6). These differences were statistically significantfor Pdx1-Cre;KrasG12D;REG3bþ/þ mice treated with the vehicleand those treated with anti-IL17RA and anti-REG3b antibodies(P < 0.01). We found no significant differences for mice treatedwith anti-IL17RA and anti-REG3b antibodies (P > 0.05). Thesefindings indicate that neutralizing either IL17RA or REG3b withspecific antibodies is a useful strategy to inhibit PanIN develop-ment in Pdx1-Cre;KrasG12D;REG3bþ/þ mice. Because therapeuticmodalities to block IL17RA are actively being tested in clinicaltrials formany human diseases, our findings support the idea that

PanINs1a 1b 32

Lesio

ns p

er fi

eld

0

2

4

10

8

6

PanINs1a1b 32

Lesio

ns p

er fi

eld

0

2

4

10

8

6

Pdx1-Cre;KrasG12D;REG3β–/–

Vehicle +REG3β

PanINs1a 1b 32

0

2

4

10

8

6

+REG3β+ an�-IL17RA

Pdx1-Cre;KrasG12D;REG3βββ–/– Pdx1-Cre;KrasG12D;REG3ββ–/– Pdx1-Cre;KrasG12D;REG3ββ–/–

Vehicle +REG3β +REG3β + an�-IL17RA

A

B

Figure 3.Systemic REG3b injection rescues PanIN development in Pdx1-cre;LSL-KrasG12D;REG3b�/�mice. A, images representative of the pancreas in Pdx1-cre;LSL-KrasG12D;REG3b�/� mice treated with vehicle, with recombinant REG3b, and with recombinant REG3b in combination with neutralizing anti-IL17RA antibody.B, number of ADM and PanIN lesions per �20 tissue field in Pdx1-cre;LSL-KrasG12D;REG3b�/� mice treated with vehicle, with recombinant REG3b, and withrecombinant REG3b in combination with the anti-IL17RA antibody at 14 weeks of age. Means are denoted by horizontal bars. Differences between genotypesand lesion types were significant (P < 0.01) for Pdx1-Cre;KrasG12D;REG3b�/� mice treated with vehicle and Pdx1-cre;LSL-KrasG12D;REG3b�/� mice treated withrecombinant REG3b and with a combination of recombinant REG3b and the anti-IL17RA antibody; no difference was found for Pdx1-Cre;KrasG12D;REG3b�/� micetreated with recombinant REGb protein and with a combination of recombinant REGb and the neutralizing anti-IL17RA antibody (P > 0.05).

Loncle et al.





that this strategy may have beneficial effects in preventing ortreating preneoplastic pancreatic cancers,which arise in the settingof chronic pancreatitis.

We next evaluated the role of REG3b at the postinitiation step,namely, on frank PDAC. Toward this end, we first used the Pdx1-Cre;KrasG12D;Ink4a/Arffl/fl mouse model, which develops PDACat a few weeks of age and evolves according to a natural history

that is similar to that seen in humans (37). Using pancreaticsamples from 8-week-old mice in which PDAC has alreadydeveloped, we observed a strong reactivity for REG3b with thefollowing percentages of positive lesion staining: 82 � 13% forADM, 46� 6% for PanIN 1a, 41� 3% for PanIN 1b, 11� 2% forPanIN 2, 3� 2% for PanIN 3, and 0% in well-established PDACs(Fig. 5). Thus, these results led us to characterize the expression of

An�-IL17RAVehicle An�-REG3β

PanINs1a 1b 32

Lesio

ns p

er fi

eld

0

2

4

10

8

6

Lesio

ns p

er fi

eld

Pdx1-Cre;KrasG12D;REG3β+/+

Vehicle

PanINs1a1b 32

0

2

4

10

8

6

An�-IL17RA

PanINs1a1b 32

0

2

4

10

8

6

An�-REG3β

Pdx1-Cre;KrasG12D;REG3βββ+/+ Pdx1-Cre;KrasG12D;REG3ββ+/+ Pdx1-Cre;KrasG12D;REG3ββ+/+A

B

Figure 4.Systemic injection of anti-REG3band anti-IL17RA antibodies inhibitsPanIN development in Pdx1-cre;LSL-KrasG12D;REG3bþ/þ mice. A, imagesrepresentative of the pancreas inPdx1-cre;LSL-KrasG12D;REG3bþ/þ

mice treated with vehicle, with anti-REG3b, and with the neutralizing anti-IL17RA antibodies. B, number of ADMand PanIN lesions per�20 tissue fieldin Pdx1-cre;LSL-KrasG12D;REG3bþ/þ

mice treated with vehicle, with anti-REG3b, and with the neutralizing anti-IL17RA antibodies at 14 weeks of age.Means are denoted by horizontal bars.Differences between genotypes andlesion typeswere significant (P <0.01)for Pdx1-Cre;KrasG12D;REG3bþ/þ micetreatedwith vehicle andPdx1-cre;LSL-KrasG12D;REG3bþ/þ mice treated withanti-REG3b and with the neutralizinganti-IL17RA antibodies; no differencewas found for Pdx1-Cre;KrasG12D;REG3bþ/þ mice treated with anti-REG3b and with the neutralizinganti-IL17RA antibodies (P > 0.05).

1a 2PanINs1b 3

% o

f Pos

i�ve

stai

ning

cel

ls

****

**

% o

f pos

i�ve

stai

ning

cel

ls

0

15

30

75

60

45

Pdx1-Cre;KrasG12D;Ink4a/Arffl/fl Peritumoral region

Perit

umor

al

Well-differen�ated (REG3b+)

Well-differen�ated (REG3b-)

Well

-diff

eren

�ated

Poor

ly diff

eren

�ated

Poorly differen�ated

A B

*

**

0

20

40

100

80

60

Figure 5.REG3b is expressed in early PDAC lesions in mice and humans. A, IHC image representative of REG3b detection in the pancreas of Pdx1-Cre;KrasG12D;Ink4a/Arffl/fl

mice; quantification of positive REG3b lesions expressed as the percentage of positive staining cells. B, IHC detection of REG3b in human PDAC and quantificationof positive REG3b lesions expressed as the percentage of positive staining cells. � , P < 0.05; �� , P < 0.01.






REG3b in human PDAC samples carryingmutations in KRAS. Wefound that the expression of REG3b in human tumors appears as astrong reaction in the nontransformed peri-tumor region as wellas in some, but not all, well-differentiated PDACs. In contrast,REG3b staining was negative in poorly differentiated PDACs (Fig.5). Thus, altogether, these results further confirm that REG3b isexpressed in the early phases of PDAC development.

REG3b promotes pancreatic cell growth by inducing cellproliferation and inhibiting apoptosis

To gain insight into both the cellular and molecular mechan-isms underlying the function of REG3b during pancreatic cancerinitiation, we treated two PDAC-derived cell lines (MiaPaCa2and Panc1) with increasing concentrations of recombinantREG3b at 24-hour intervals for a period of 72 hours and mea-sured BrdUrd incorporation. When compared with control, cellsthat were treated with REG3b at a concentration of 50 ng/mLhad an increase in BrdUrd incorporation of 2.3-fold and 1.7-fold(P < 0.05) for MiaPaCa2 and Panc1 cells, respectively (Fig. 1B).Moreover, at a concentration of 100 ng/mL for REG3b, thevalues for BrdUrd incorporation increased to 2.3-fold and 3.9-fold (P < 0.01). These results demonstrate that an importanteffect of REG3b, as it relates to cancer cells, is its ability tostimulate their proliferation. Thus, we complemented thesestudies by measuring the effects of REG3b on pancreatic celldeath. For this purpose, we activated the apoptotic program ofpancreatic cancer cells by serum starvation for 24, 48, and 72hours. At the same time, cells were also incubated with orwithout recombinant REG3b protein at a concentration of100 ng/mL. Apoptosis was determined by measuring cell viabil-ity and caspase-3/7 activities at the end of the experiment. Figures6B and 7C show that the treatment of both MiaPaCa2 and Panc1cells with REG3b for 48 and 72 hours increased their resistance toapoptosis as evidenced by increased cell viability and decreasedcaspase-3/7 activity. Thus, we conclude that REG3bpromotes cellgrowth and protects against cell death in PDAC cells. These dataare relevant to better understanding the role of this proteinduring the process of initiation, because it is well-establishedthat both of these processes are crucial for normal cells to acquirethe ability to undergo ADM and form PanIN lesions in responseto oncogenic KRAS.

A novel IL17-regulated paracrine pathway involving REG3b-JAK2-STAT3 signaling promotes KRAS-mediated PDACinitiation

Our results demonstrate that REG3b is a mediator of bothinflammation and cancer. This finding is reminiscent of the effectsrecently described by several laboratories for STAT3, which,similar to REG3b, also works downstream of proinflammatorypathways. Thus, we hypothesized that REG3bmay signal via thisintracellular signaling pathway. To evaluate the influence of thisprotein on STAT3 phosphorylation, which is a marker for itsactivation, we treated pancreatic cells with 100 ng/mL of recom-binant REG3b for 1 hour, which demonstrated that REG3bsignificantly increased the phosphorylation-dependent activationof STAT3 (Fig. 6D). Next, we extended our mechanistic investiga-tions to determine whether this effect occurs via gp130, a proteinthat functions as a receptor for REG3b. Interestingly, STAT3activation in response to REG3bwas almost completely inhibitedby pretreating the cells with an anti-gp130 neutralizing antibody.Similar results were obtained using the JAK2 inhibitor AG-490,

indicating that this kinase links the REG3b-gp130 interaction toSTAT3 activation. Thus, we formally assessed whether the acti-vation of this pathway was linked to the effects of REG3b onBrdUrd incorporation and cell death. Cells were treated with 100ng/mL of recombinant REG3b for 24 hours either in the presenceor absence of the anti-gp130 antibody and proliferation wasmeasured by BrdUrd incorporation. Similar experiments werealso performed using the JAK2 inhibitor AG-490. The results ofthese experiments shown in Fig. 6E demonstrate that both theanti-gp130 antibody and AG-490 readily inhibit the growth-promoting effect of REG3b on MiaPaCa2 and Panc1 cells.Finally, we also evaluated whether gp130 and JAK2 wereinvolved in REG3b-mediated apoptosis resistance. Accordingly,PDAC-derived MiaPaCa2 and Panc1 cells were treated withrecombinant REG3b in the presence and absence of either theanti-gp130 antibody or AG-490. Figure 6F shows that the anti-apoptotic effect of recombinant REG3b was almost completelyinhibited by the anti-gp130 antibody, as well as the AG-490compound. Based on these results, we conclude that REG3b is, atleast partly, responsible for the activation of the intracellularpathway led by the gp130 receptor, JAK2, and phosho-STAT3,which therefore promotes pancreatic cell growth and increasesresistance to cell death.

REG3b regulates the phenotype of bothPDAC-derived cells andmacrophages in a cell-to-cell-dependent interaction

The experiments described above have demonstrated thatREG3b is a paracrine factor released by the Kras-transformed aswell as inflamed pancreatic epithelial cells, similar to many othercancer promoting proteins that are synthesized and released bythe epithelial compartment of the initiating PanIN lesions. Recentstudies, however, indicate that inflammatory mediators also playa role in stimulating protumoral or inhibiting antitumoralresponses in the tumor microenvironment, which led us toestablish a coculture model to evaluate whether REG3b has theability to function in this manner. This coculture model involvedthe use of human pancreatic cancer cells (MiaPaCa2) and mac-rophage-differentiated THP-1 monocytes, both grown in asingle chamber that contains a separation to avoid cell-to-cellcontact. Because we along with others have shown that theCXCL12-CXCR4 chemokine signaling pathway is key in promot-ing pancreatic cancer, we first analyzed whether REG3b affects theactivation of this signaling pathway. As shown in Fig. 7, REG3binduced an increase in the expression of CXCR4mRNA inmacro-phages (1.0 � 0.1 vs. 1.5 � 0.1 folds, P < 0.01). This increase inCXCR4mRNAwas unalteredwhethermacrophageswere culturedalone or in the presence of pancreatic cancer cells. Parallel experi-ments showed that REG3b did not modify the expression ofCXCL12 in pancreatic cancer cells when cultured alone. In con-trast, REG3b did, however, induce a marked increase in theexpression of CXCL12 when pancreatic cancer cells were cocul-tured with macrophages (0.6 � 0.1 vs. 9.7 � 1.2 folds, P < 0.01).Similarly, pancreatic cancer cells exhibited a comparable effect inthe expression of TGFb (2.1� 0.2 vs. 3.8� 0.4 folds, P < 0.05) andIL10 (0.6 � 0.1 vs. 4.3 � 0.3 folds, P < 0.01), other cytokines,which like CXCL12 modulate pancreatic cancer development.Control experiments demonstrated that, in the absence of REG3b,cocultured cells did not change the expression of thesemediators.Notably, REG3b also induced THP-1 macrophages from thecoculture to increase the synthesis of IL10 (2.4 � 0.4 vs. 7.2 �0.8 folds, P < 0.01) and MRC-1 (mannose receptor; 1.1 � 0.2 vs.

Loncle et al.





2.4 � 0.2 folds, P < 0.01), two typical markers of the M2phenotype. Taken together, these results demonstrate that REG3bcan induce robust communication between pancreatic cancercells and cells, which, like macrophages, are commonly foundin the desmoplastic microenvironment where they regulate manycytokine-mediated immunotumoral responses. These effects ofREG3b appear to act independently yet complementary to its rolein promoting PanIN development by acting on the gp130 recep-tor present in epithelial cells.

DiscussionElegant studies, primarily performed during the last decade,

demonstrate that inflammatory cells and other stromal elementsexert a potent protumoral effect on oncogenic Kras-activatedepithelial cells (38). Noteworthy, many of the stimuli found topromote tumor growth are also activated during pancreatitis. Infact, chronic pancreatitis is known to accelerate the process ofpancreatic tumorigenesis in both humans and mice (1, 39).Recently, using a sophisticated combination of transgenic mice

BrdU

rd in

corp

ora�

on(fo

ld c

hang

e)

0

1

2

5

4

3 *

**

REG3β (ng/mL)

0

1

2

5

4

3*

**

REG3β (ng/mL)

Livi

ng c

ells

(% o

f 24

h st

arve

d)

0

25

50

125

100

75

24 h48 720

25

50

125

100

75

24 h48 72

*

ns*

*ns

*

0

2

4

10

8

6

*

24 h48 72

nsns

BrdU

rd in

corp

ora�

on(fo

ldch

ange

)

0

1

2

5

4

3

**

nsns

*

0

1

2

5

4

3

**

ns

ns

*

A

B

0

5

10

25

20

15

*

24 h48 72

Casp

ase

3/7

ac�v

ity(a

rbita

ryun

itsx1

03 )

nsns

C

D

E

REG3β REG3β

p-STAT3

STAT3

0

5

10

25

20

15

*

Casp

ase

3/7

ac�v

ity(a

rbita

ryun

itsx1

03 )

nsF

*

0

2

4

10

8

6

*ns *

MiaPaCa2 Panc1

MiaPaCa2 Panc1

MiaPaCa2 Panc1 MiaPaCa2 Panc1

MiaPaCa2 Panc1MiaPaCa2 Panc1

Figure 6.REG3b promotes cellular growth and increases cell resistance to cell death through a gp130-, JAK2-, and STAT3-dependent mechanism. A, MiaPaCa2 andPanc1 cells were treated with 50 and 100 ng/mL of recombinant REG3b protein and BrdUrd incorporation was measured after 24 hours. Data are expressed asfold changes relative to untreated cells. B and C, MiaPaCa2 and Panc1 cells were cultivated in EBSS media for 24, 48, and 72 hours in either the absence (blue) orpresence (red) of recombinant REG3b protein. Cell viability (B) and caspase 3/7 activity (C) were measured. D, MiaPaCa2 and Panc1 cells were cultivatedin the presence of recombinant REG3b protein together with either the anti-gp130 or the AG-490 compound and the expression of phosphoSTAT3 or STAT3was measured by Western blotting. E and F, MiaPaCa2 and Panc1 cells were cultivated in EBSS media for 72 hours in the absence (blue) and presence (red)of recombinant REG3b protein together with either the anti-gp130 or the AG-490 compound and BrdUrd incorporation (E) and caspase 3/7 activity (F) weremeasured. � , P < 0.05; �� , P < 0.01. ns, nonsignificant.






with specific neutralizing antibodies, McAllister and colleaguesdemonstrated that TH17 and IL17þ/gdT cells, infiltrating thepancreatic stroma, produce IL17A so as to promote PanIN devel-opment through the activation of the STAT3 signaling pathway(30). This study also determined that the IL17RA receptor isexpressed in pancreatic cells upon oncogenic Kras-activation.Combined, therefore, these observations reveal that inflamma-tion induces immune cells to migrate into the pancreatic stromawhere they produce and release the master cytokine IL17A.Simultaneously, the expression of the IL17RA is induced inepithelial pancreatic cells after oncogenic Kras activation, therebysetting the conditions for this pathway to be activated in aparacrine fashion. However, the mechanisms that work down-stream of this pathway to promote KRAS-induced pancreaticcarcinogenesis remain to be fully understood. In this study, weused several complementary approaches to demonstrate that the

expression of REG3b is induced downstream of the IL17RAreceptor and that REG3b is necessary for PanIN development inmice. In addition, we were able to show that REG3b activates agp130-, JAK2-, and STAT3-dependent signaling pathway to pro-mote cell growth and increase cell resistance in PDAC-derivedcells. Unexpectedly, we also found that REG3b promotes theexpression of mediators in both PDAC-derived cells and macro-phages in a cell-to-cell-dependent manner. The Pdx1-Cre;KrasG12D;REG3bþ/þ mice, in which the REG3b protein was neu-tralized with a specific antibody against REG3b (Fig. 4), wereunable to develop PanIN, demonstrating that REG3b is an essen-tial factor for PanIN development. Therefore, REG3b should beconsidered a potential target for the treatment of patients withinflammation-associatedPDAC. In summary,wehave establishedthat REG3b is a key effector of the IL17A/IL17RA receptor pathwayacting during both the inflammation and transformation steps ofinflammation-induced pancreatic cancer,which functions accord-ing to the model shown in Supplementary Fig. S1.

Our study also demonstrates that in addition to working onreceptors of KRAS-transformed pancreatic epithelial cells, REG3bmodulates the function of inflammatory and PDAC-derived cellswhen both cells were cocultivated (Fig. 7). This immunomodu-latory activity of REG3b on THP-1macrophages occurs complete-ly independent of the cell-to-cell contact, because, in our exper-imental set-up, these cells were cocultured in chambers carrying abarrier that prevents their direct interactions. We find that REG3bactivates the CXCL12/CXCR4 signaling cascade, which sustainstumor cell growth, induces angiogenesis, facilitates tumor escapethrough evasion of immune surveillance, and promotes metas-tasis with neural invasion (40). Moreover, we show that REG3binduces the expression of IL10, TGFb andMRC-1, which togetherbehave as immunomodulators and increase the synthesis of thepancreatic desmoplastic reaction (41). Thus, these results under-score the fact that, in addition to its effects on pancreatic cancercells, REG3b can regulate the timely release of chemokines thatsupport the dialogue between tumor cells and cells from itsmicroenvironment (42). Thus, it is likely that in the complexchemokine-rich milieu of chronic pancreatitis, REG3b facilitatesthe transforming function of KRAS by acting on both pancreaticcells and its microenviroment. We are optimistic that futurestudies using state-of-the-art conditional knockout models fordeleting REG3b in different cell types will extend these observa-tions and further define the relative importance of each of thesemechanisms to the development of inflammation-associatedpancreatic cancer.

Notably, studies seeking to better understand the mechanismsthat underlie the transition from pancreatitis into cancer haverecently gained popularity since a model for studying this processwas established by Guerra and colleagues (39). We have recentlyused this model to demonstrate that an EGF-dependent proin-flammatory NFATc1-STAT3 pathway promotes the transitionof chronic pancreatitis into cancer (43). Similarly, Liou andcolleagues found that RANTES and TNFa produced by activatedstromal macrophages have a significant effect on the transforma-tion of ADM following activation of NF-kB signaling (44). More-over, McAllister and colleagues discovered that IL17-secretingcells play a similar role (30). Thus, from considering these studiesand others in aggregate, it appears that more than one proin-flammatory mediator can promote PanIN development by KRAS.It is even likely that epithelial cells from the pancreatic parenchy-ma or mesenchymal cells from the stroma synthesize mediators

CXCL

12m

RNA

(rel

a�ve

exp

ress

ion)

0

2.5

5.0

10.0

7.5

+

+ THP-1

TGFβ

mRN

A(r

ela�

veex

pres

sion)

0

1

2

4

3

+ THP-1

IL10

mRN

A(r

ela�

veex

pres

sion)

0

1

2

4

3

+

+ THP-1

*5

+*

+*MiaPaCa2

CXCR

4m

RNA

(rel

a�ve

expr

essio

n)

0

0.5

1.0

2.0

1.5 *

+ MiaPaCa2

*

MRC

-1m

RNA

(rel

a�ve

expr

essio

n)

0

2

4

8

6

+ MiaPaCa2

+

IL10

mRN

A(r

ela�

veex

pres

sion)

0

0.5

1.0

2.0

1.5

+

+ MiaPaCa2

*2.5

+*

THP-1

Control

+ REG3β

Figure 7.REG3b induces changes in gene expression of macrophages and PDAC cellswhen they are cocultured. Macrophage-differentiated THP-1 cells (left) andMiaPaCa2 cells (right) were cultured alone or in combination using transwellinserts. Cellswere treatedwith 500ng/mLof recombinantREG3bprotein andRNA was obtained after 24 hours. THP-1 cells increased the expression ofCXCR4 in response to REG3b; a similar increasewas observed in the presenceof MiaPaCa2 cells. By contrast, REG3b induced the expression of MRC-1 andIL10 only when THP-1 cells were cocultured with MiaPaCa2 cells. Coculturealso promoted the expression of MRC-1 even in the absence of REG3b. On theother hand, REG3b had no effect on MiaPaCa2 cells alone, but stronglyinduced the expression of CXCL12, TGFb, and IL10 when cocultured withTHP-1 cells.� , P < 0.05 versus its corresponding control;þ, P < 0.05 versus nococulture.

Loncle et al.





that can act at different times during the initiation process (earlyvs. late PanINs), though this idea has remained experimentallyunexplored. In addition, similar to what we observe with REG3b,which expression disappears after the cancer initiation step,different factors secreted in either an autocrine manner by KRASmutant pancreatic cells or a paracrine manner by cells from aninflammatory-protumoral microenvironment may be required atsubsequent steps during carcinogenesis. In fact, decades ofgenome-wide expression profiling studies have clearly demon-strated thatwith each additionalmutation or epigenetic alterationacquired during the transformation process, pancreatic cancercells switch on and off the synthesis of different cytokines andgrowth factors.However, studies that seek to definewhich of thesefactors are operational during the pancreatitis–cancer transition,like the one reported here, have recently begun. These types ofstudies are of paramount importance because patients withchronic pancreatitis are the largest group of individuals at riskfor dying of pancreatic cancer. More importantly, these patientsalso display the highest score among all high-risk populations foracquiring this disease (1).

Thus, the current study was designed to define whether one ofthe best-characterized mediators of pancreatitis, REG3b, alsopromotes the development of pancreatic cancer. Indeed, ourresults not only support this hypothesis, but also discoveredmechanisms by which REG3b promotes cancer development atthe molecular level through its involvement in a IL17A-REG3b-gp130-JAK2-STAT3 pathway and at the cellular level by increasingcell proliferation and survival. Moreover, at the whole organismlevel, our antibody-based preclinical trials demonstrate that inhi-bition of the IL17A-REG3b pathway antagonizes KRAS-inducedpancreatic cancer initiation. Thus, this new knowledge changesthe previously established paradigm, which considered REG3b asan exclusive pancreatitis-associated protein, into a new paradigmthat describes this protein as a druggable link between pancreatitisand pancreatic cancer. Furthermore, we also have studied severalmolecules within this pathway, namely, IL17, REG3b, JAK2, andSTAT3, for which inhibitors are being tested in clinical trials,thereby contributing to build the foundation of future combina-torial therapies. In conclusion, the experiments reported here

support the idea that premalignant diseases of the pancreasactively establish mechanisms that can self-perpetuate inflamma-tion, and, more importantly, also lead to neoplastic transforma-tion. Fortunately, we demonstrate that antagonizing thesemechanisms have a beneficial effect in preventing this malignantswitch.

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

Authors' ContributionsConception and design: E. Folch-Puy, E. Chuluyan, R. Urrutia, D. Closa,J.L. IovannaDevelopment of methodology: L. Bonjoch, M.B. Lopez-MillanAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): C. Loncle, M.B. Lopez-MillanAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): C. Loncle, M.B. Lopez-Millan, S. Lac, P. Cordelier,P. Dubus, R. Urrutia, D. Closa, J.L. IovannaWriting, review, and/or revision of the manuscript: C. Loncle, L. Bonjoch,E. Folch-Puy, M.B. Lopez-Millan, M.I. Molejon, E. Chuluyan, P. Cordelier,P. Dubus, G. Lomberk, R. Urrutia, D. Closa, J.L. IovannaAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): M.B. Lopez-Millan, S. Lac, G. Lomberk,R. UrrutiaStudy supervision: G. Lomberk, R. Urrutia, J.L. Iovanna

Grant SupportThis work was supported by La Ligue Contre le Cancer, INCa, Canceropole

PACA, DGOS (labellisation SIRIC), and INSERM to J.L. Iovanna; NIH grantsDK52913, the Mayo Clinic Center for Cell Signaling in Gastroenterology(P30DK084567), and the Mayo Foundation to R. Urrutia; by Fraternal Orderof Eagles Cancer Award toG. Lomberk; and the FIS grant from Instituto de SaludCarlos III (PI13/01224) to E. Folch-Puy.

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 April 9, 2015; revised August 12, 2015; accepted August 14, 2015;published OnlineFirst September 24, 2015.

References1. Lowenfels AB,Maisonneuve P, DiMagno EP, Elitsur Y, Gates LK Jr., Perrault

J, et al. Hereditary pancreatitis and the risk of pancreatic cancer. Interna-tional Hereditary Pancreatitis Study Group. J Natl Cancer Inst 1997;89:442–6.

2. Keim V, Iovanna JL, Rohr G, Usadel KH, Dagorn JC. Characterization of arat pancreatic secretory protein associated with pancreatitis. Gastroenter-ology 1991;100:775–82.

3. Iovanna J, Orelle B, Keim V, Dagorn JC. Messenger RNA sequence andexpression of rat pancreatitis-associated protein, a lectin-related proteinoverexpressed during acute experimental pancreatitis. J Biol Chem 1991;266:24664–9.

4. Orelle B, Keim V, Masciotra L, Dagorn JC, Iovanna JL. Human pancreatitis-associated protein. Messenger RNA cloning and expression in pancreaticdiseases. J Clin Invest 1992;90:2284–91.

5. Dusetti NJ, Frigerio JM, Keim V, Dagorn JC, Iovanna JL. Structural orga-nization of the gene encoding the rat pancreatitis-associated protein.Analysis of its evolutionary history reveals an ancient divergence from theother carbohydrate-recognition domain-containing genes. J Biol Chem1993;268:14470–5.

6. Dusetti NJ, Frigerio JM, Fox MF, Swallow DM, Dagorn JC, Iovanna JL.Molecular cloning, genomic organization, and chromosomal localization

of the human pancreatitis-associated protein (PAP) gene. Genomics1994;19:108–14.

7. Itoh T, Teraoka H. Cloning and tissue-specific expression of cDNAs for thehuman and mouse homologues of rat pancreatitis-associated protein(PAP). Biochimica et biophysica acta 1993;1172:184–6.

8. Lasserre C, Simon MT, Ishikawa H, Diriong S, Nguyen VC, Christa L, et al.Structural organization and chromosomal localization of a human gene(HIP/PAP) encoding a C-type lectin overexpressed in primary liver cancer.Eur J Biochem/FEBS 1994;224:29–38.

9. Lasserre C, Christa L, Simon MT, Vernier P, Brechot C. A novel gene (HIP)activated in human primary liver cancer. Cancer Res 1992;52:5089–95.

10. Katsumata N, Chakraborty C, Myal Y, Schroedter IC, Murphy LJ, Shiu RP,et al. Molecular cloning and expression of peptide 23, a growth hormone-releasing hormone-inducible pituitary protein. Endocrinology 1995;136:1332–9.

11. Masciotra L, Lechene de la Porte P, Frigerio JM, Dusetti NJ, Dagorn JC,Iovanna JL. Immunocytochemical localization of pancreatitis-associatedprotein in human small intestine. Dig Dis Sci 1995;40:519–24.

12. Chakraborty C, Vrontakis M, Molnar P, Schroedter IC, Katsumata N,Murphy LJ, et al. Expression of pituitary peptide 23 in the rat uterus:regulation by estradiol. Mol Cell Endocrinol 1995;108:149–54.






13. Christa L, Simon MT, Brezault-Bonnet C, Bonte E, Carnot F, Zylberberg H,et al. Hepatocarcinoma-intestine-pancreas/pancreatic associated protein(HIP/PAP) is expressed and secreted by proliferating ductules as well as byhepatocarcinoma and cholangiocarcinoma cells. Amer J Pathol 1999;155:1525–33.

14. Yamamoto T, KatsumataN, Tachibana K, FriesenHG,Nagy JI. Distributionof a novel peptide in the anterior pituitary, gastric pyloric gland, andpancreatic islets of rat. J Histochem Cytochem 1992;40:221–9.

15. Iovanna JL, Keim V, Michel R, Dagorn JC. Pancreatic gene expression isaltered during acute experimental pancreatitis in the rat. Amer J Physiol1991;261:G485–9.

16. Duplan L, Michel B, Boucraut J, Barthellemy S, Desplat-Jego S, Marin V,et al. Lithostathine and pancreatitis-associated protein are involved in thevery early stages of Alzheimer's disease. Neurobiol Aging 2001;22:79–88.

17. Nishimune H, Vasseur S, Wiese S, Birling MC, Holtmann B, Sendtner M,et al. Reg-2 is a motoneuron neurotrophic factor and a signalling inter-mediate in the CNTF survival pathway. Nat Cell Biol 2000;2:906–14.

18. Ampo K, Suzuki A, Konishi H, Kiyama H. Induction of pancreatitis-associated protein (PAP) family members in neurons after traumatic braininjury. J Neurotrauma 2009;26:1683–93.

19. OgawaH, Fukushima K,NaitoH, Funayama Y, UnnoM, Takahashi K, et al.Increased expression of HIP/PAP and regenerating gene III in humaninflammatory bowel disease and a murine bacterial reconstitution model.Inflamm Bowel Dis 2003;9:162–70.

20. Rodenburg W, Keijer J, Kramer E, Roosing S, Vink C, Katan MB, et al.Salmonella induces prominent gene expression in the rat colon. BMCMicrobiol 2007;7:84.

21. Rechreche H, Montalto G, Mallo GV, Vasseur S, Marasa L, Soubeyran P,et al. pap, reg Ialpha and reg Ibeta mRNAs are concomitantly up-regulatedduring human colorectal carcinogenesis. Int J Cancer J Int du Cancer1999;81:688–94.

22. Chen CY, Lin XZ, Wu HC, Shiesh SC. The value of biliary amylase andhepatocarcinoma-intestine-pancreas/pancreatitis-associated protein I (HIP/PAP-I) in diagnosing biliary malignancies. Clin Biochem 2005;38:520–5.

23. NarushimaY,UnnoM,NakagawaraK,MoriM,MiyashitaH, Suzuki Y, et al.Structure, chromosomal localization and expression of mouse genesencoding type III Reg, RegIII alpha, RegIII beta, RegIII gamma. Gene1997;185:159–68.

24. Fitzgibbons TP, Paolino J, Dagorn JC,Meyer TE. Usefulness of pancreatitis-associated protein, a novel biomarker, to predict severity of disease inambulatory patients with heart failure. Amer J Cardiol 2014;113:123–6.

25. WaelputW, Verhee A, Broekaert D, Eyckerman S, Vandekerckhove J, BeattieJH, et al. Identification and expression analysis of leptin-regulated imme-diate early response and late target genes. Biochem J 2000;348(Pt 1):55–61.

26. Nitta Y, Konishi H, Makino T, Tanaka T, Kawashima H, Iovanna JL, et al.Urinary levels of hepatocarcinoma-intestine-pancreas/pancreatitis-associ-ated protein as a diagnostic biomarker in patients with bladder cancer.BMC Urol 2012;12:24.

27. Lai Y, Li D, Li C, Muehleisen B, Radek KA, Park HJ, et al. The antimicrobialprotein REG3A regulates keratinocyte proliferation and differentiationafter skin injury. Immunity 2012;37:74–84.

28. Lehotzky RE, PartchCL,Mukherjee S, CashHL, GoldmanWE,Gardner KH,et al. Molecular basis for peptidoglycan recognition by a bactericidal lectin.Proc Natl Acad Sci USA 2010;107:7722–7.

29. Folch-Puy E, Granell S, Dagorn JC, Iovanna JL, Closa D. Pancreatitis-associated protein I suppresses NF-kappa B activation through a JAK/STAT-mediated mechanism in epithelial cells. J Immunol 2006;176:3774–9.

30. McAllister F, Bailey JM, Alsina J, Nirschl CJ, Sharma R, Fan H, et al.Oncogenic Kras activates a hematopoietic-to-epithelial IL-17 signaling axisin preinvasive pancreatic neoplasia. Cancer Cell 2014;25:621–37.

31. Aguirre AJ, Bardeesy N, Sinha M, Lopez L, Tuveson DA, Horner J, et al.Activated Kras and Ink4a/Arf deficiency cooperate to produce metastaticpancreatic ductal adenocarcinoma. Genes Dev 2003;17:3112–26.

32. Gironella M, Folch-Puy E, LeGoffic A, Garcia S, Christa L, Smith A, et al.Experimental acute pancreatitis in PAP/HIP knock-out mice. Gut 2007;56:1091–7.

33. Tuveson DA, Shaw AT, Willis NA, Silver DP, Jackson EL, Chang S, et al.Endogenous oncogenic K-ras(G12D) stimulates proliferation and wide-spread neoplastic and developmental defects. Cancer Cell 2004;5:375–87.

34. Garcia MN, Grasso D, Lopez-Millan MB, Hamidi T, Loncle C, Tomasini R,et al. IER3 supports KRASG12D-dependent pancreatic cancer developmentby sustaining ERK1/2 phosphorylation. J Clin Invest 2014;124:4709–22.

35. Yuan JS, Reed A, Chen F, Stewart CN Jr. Statistical analysis of real-time PCRdata. BMC Bioinformat 2006;7:85.

36. Adler G, Hupp T, Kern HF. Course and spontaneous regression of acutepancreatitis in the rat. Virchows Arc A, Pathol AnatHistol 1979;382:31–47.

37. CanoCE, Hamidi T, GarciaMN, GrassoD, Loncle C, Garcia S, et al. Geneticinactivation of Nupr1 acts as a dominant suppressor event in a two-hitmodel of pancreatic carcinogenesis. Gut 2014;63:984–95.

38. Neesse A, Krug S, Gress TM, Tuveson DA, Michl P. Emerging concepts inpancreatic cancer medicine: targeting the tumor stroma. OncoTargets Ther2013;7:33–43.

39. Guerra C, Schuhmacher AJ, Canamero M, Grippo PJ, Verdaguer L, Perez-Gallego L, et al. Chronic pancreatitis is essential for induction of pancreaticductal adenocarcinoma by K-Ras oncogenes in adult mice. Cancer Cell2007;11:291–302.

40. ZhongW,ChenW, ZhangD, Sun J, Li Y, Zhang J, et al. CXCL12/CXCR4 axisplays pivotal roles in the organ-specific metastasis of pancreatic adeno-carcinoma: a clinical study. Exp Therapeutic Med 2012;4:363–9.

41. DjaldettiM, BesslerH.Modulators affecting the immune dialogue betweenhuman immune and colon cancer cells. World J Gastrointest Oncol 2014;6:129–38.

42. Iovanna JL, Marks DL, Fernandez-Zapico ME, Urrutia R. Mechanisticinsights into self-reinforcing processes driving abnormal histogenesisduring the development of pancreatic cancer. Amer J Pathol 2013;182:1078–86.

43. Baumgart S, Chen NM, Siveke JT, Konig A, Zhang JS, Singh SK, et al.Inflammation-induced NFATc1-STAT3 transcription complex promotespancreatic cancer initiation by KrasG12D. Cancer Dis 2014;4:688–701.

44. Liou GY, Doppler H, Necela B, Krishna M, Crawford HC, Raimondo M,et al. Macrophage-secreted cytokines drive pancreatic acinar-to-ductalmetaplasia through NF-kappaB and MMPs. J Cell Biol 2013;202:563–77.


Loncle et al.




2015;75:4852-4862. Published OnlineFirst September 24, 2015.Cancer Res Celine Loncle, Laia Bonjoch, Emma Folch-Puy, et al. Pancreatitis to Pancreatic CancerInflammatory Pathway to Promote the Transition from Chronic

STAT3−JAK2−βIL17 Functions through the Novel REG3

Updated version

10.1158/0008-5472.CAN-15-0896doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2015/10/08/0008-5472.CAN-15-0896.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/75/22/4852.full#ref-list-1

This article cites 44 articles, 10 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/75/22/4852.full#related-urls

This article has been cited by 6 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://cancerres.aacrjournals.org/content/75/22/4852To 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-15-0896

http://cancerres.aacrjournals.org/content/suppl/2015/10/08/0008-5472.CAN-15-0896.DC1

http://cancerres.aacrjournals.org/content/75/22/4852.full#ref-list-1

http://cancerres.aacrjournals.org/content/75/22/4852.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/75/22/4852


IL17 Functions through the Novel REG3 STAT3 …...Molecular and Cellular Pathobiology IL17 Functions through the Novel REG3b– JAK2–STAT3 Inﬂammatory Pathway to Promote the Transition

Documents