PanIN Neuroendocrine Cells Promote Tumorigenesis via Neuronal … · slides from 8-, 12-, and...

Microenvironment and Immunology

PanIN Neuroendocrine Cells PromoteTumorigenesis via Neuronal Cross-talkSmrita Sinha1,2,3, Ya-Yuan Fu3, Adrien Grimont1, Maren Ketcham4, Kelly Lafaro1,Joseph A. Saglimbeni1, Gokce Askan1,5, Jennifer M. Bailey6, Jerry P. Melchor1,Yi Zhong1, Min Geol Joo7, Olivera Grbovic-Huezo1, In-Hong Yang7, Olca Basturk5,Lindsey Baker8, Young Park8, Robert C. Kurtz2, David Tuveson8, Steven D. Leach1,and Pankaj J. Pasricha3

Abstract

Nerves are a notable feature of the tumor microenvironmentin some epithelial tumors, but their role in the malignantprogression of pancreatic ductal adenocarcinoma (PDAC) isuncertain. Here, we identify dense innervation in the micro-environment of precancerous pancreatic lesions, known aspancreatic intraepithelial neoplasms (PanIN), and describe aunique subpopulation of neuroendocrine PanIN cells thatexpress the neuropeptide substance P (SP) receptor neurokinin1-R (NK1-R). Using organoid culture, we demonstrated thatsensory neurons promoted the proliferation of PanIN orga-noids via SP-NK1-R signaling and STAT3 activation. Nerve-

responsive neuroendocrine cells exerted trophic influences andpotentiated global PanIN organoid growth. Sensory denerva-tion of a genetically engineered mouse model of PDAC ledto loss of STAT3 activation, a decrease in the neoplastic neu-roendocrine cell population, and impaired PanIN progressionto tumor. Overall, our data provide evidence that nervesof the PanIN microenvironment promote oncogenesis, likelyvia direct signaling to neoplastic neuroendocrine cells capableof trophic influences. These findings identify neuroepithelialcross-talk as a potential novel target in PDAC treatment.Cancer Res; 77(8); 1868–79. �2017 AACR.

IntroductionNeuronal influences on tumorigenesis have been described

in prostate (1) and gastric cancers (2) but their role in pancre-atic ductal adenocarcinoma (PDAC) remains unclear. PDAChas a tropism for nerves and is associated with high rates ofperineural invasion that portends a worse prognosis (3). Recip-rocal molecular signaling between PDAC and nerves appears todrive perineural invasion (4, 5). In turn, human PDAC appears

to cause global neural remodeling as tumors show increasedneural hypertrophy (3). Whether nerves are selectively presentin the neoplastic microenvironment and can drive tumorigen-esis of PDAC or its precursor lesion, pancreatic intraepithelialneoplasia (PanIN), is unknown.

The pancreas is richly innervated by intrinsic, autonomic, andsensory nerves. Visceral sensory afferent nerves project fromdorsal root ganglia (DRG) and vagal nodose ganglia (NG)neurons (6), and have both afferent (sensory) and efferent (tissuetargeting) functions (for review see ref. 7). Unmyelinated affer-ents, C and Ad fibers, express the transient receptor potentialcation channel subfamily V member 1 (TRPV1) channel thatmediates the release of inflammatory neuropeptides such assubstance P (SP; 8). SP binding to its target G protein-coupledreceptor (GPCR), neurokinin 1 receptor (NK1-R), activates severaloncogenic pathways (for review see ref. 9) in non-neuronalhuman cancers (10, 11). The SP–NK1-R axis, for example, canstimulate the Janus kinase (JAK)-signal transducer and activator oftranscription (STAT; ref. 12) signaling pathway, which is also animportant driver of Kras-mediated oncogenesis in PDAC (13).

We hypothesized that sensory nerves in the PanIN microenvi-ronment communicated directly with the PanIN epithelium todrive tumorigenesis. We identified a small population of uniqueneoplastic neuroendocrine cells expressing the neuropeptidereceptor NK1-R in both murine and human PanINs. Using orga-noid culture techniques, we demonstrated that these neuron-responsive NK1-Rþ cells exerted trophic influences and potenti-ated global PanIN organoid growth. Accordingly, sensory dener-vation in a PDAC mouse model decreased the NK1-Rþ cellpopulation and protected against PanIN progression, confirminga protumorigenic role of sensory nerves.

1David M. Rubenstein Center for Pancreatic Cancer Research, Memorial SloanKettering Cancer Center, New York, New York. 2Gastroenterology andNutrition Service, Memorial Sloan Kettering Cancer Center, New York, NewYork. 3Division of Gastroenterology and Hepatology, Johns Hopkins Hospital,Baltimore, Maryland. 4Weill Cornell Medical College, New York, New York.5Gastrointestinal Pathology, Memorial Sloan Kettering Cancer Center, NewYork, New York. 6Division of Surgical Oncology, Johns Hopkins Hospital,Baltimore, Maryland. 7Department of Biomedical Engineering, Johns HopkinsUniversity, Baltimore, Maryland. 8Cold Spring Harbor Laboratory, Cold SpringHarbor, New York.

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

S.D. Leach and P.J. Pasricha share co-senior authorship of and contributedequally to this article.

Corresponding Authors: Steven D. Leach, Memorial Sloan Kettering CancerCenter, 1275 York Avenue, Box 20, New York, NY 10065. Phone: 646-888-3662;Fax: 646-888-3235; E-mail: [email protected]; and Pankaj J. Pasricha, Divisionof Gastroenterology and Hepatology, The Johns Hopkins Hospital, Baltimore,MD 21202. E-mail: [email protected].

doi: 10.1158/0008-5472.CAN-16-0899

�2017 American Association for Cancer Research.

CancerResearch

Cancer Res; 77(8) April 15, 20171868

on October 9, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst January 19, 2017; DOI: 10.1158/0008-5472.CAN-16-0899

http://cancerres.aacrjournals.org/

Materials and MethodsMouse strains

Pdx1:Cre, LSL-KrasG12D and LSL-Trp53R172H mice (The JacksonLaboratory)were bred to generate KCPdx1 and KPCPdx1mice.Mist1:CreERT2; LSL-KrasG12D and LSL-tdTomato mice were bred to gen-erate KCMist1tdT mice. KCMist1tdT mice were injected with Tamox-ifen and treated with cerulein as described previously (14). Animalstudies were approved by the Animal Care and Use Committees ofJohns Hopkins University School of Medicine, Memorial SloanKettering Cancer Center and Cold Spring Harbor Laboratory.

Cell and organoid linesA6L, MIA PaCa-2, and Capan-2 cells (Iacobuzio-Donahue

laboratory) and 3T3 and 293T (ATCC) cells were maintained inRPMI-1640 or DMEM (Gibco) supplemented with 10% FBS(Sigma-Aldrich), GlutaMAX (Gibco), and 1� penicillin-strepto-mycin (Gibco). Human pancreatic ductal epithelial (HPDE) cellswere cultured according to established protocols (15). MurinePanINorganoids (derived fromKCPdx1 andKCMist1tdTmice)werecultured according to published protocols (16). A6L cells (2008)were authenticated with whole exome sequencing and confirmedto have mutant Kras (17). HPDE (2013), MIA PaCa-2 (2016),Capan-2 (2016), 3T3 (2015) and 293T (2016) cells were authen-ticated using short tandem repeat profiling.Organoids underwentlox-KrasG12D allele genotyping (2016).

Neuronal cocultureDorsal root ganglia (DRG) from 4- to 8-week-old C57BL/6J

mice were harvested and cultured in supplemented NeurobasalMedium (NBM; Gibco) as previously described (18). Growthfactor–free NBM was used for all coculture experiments. DRGneurons were plated with PDAC cell lines (5 � 103 cells/well) inthe microfluidic device for 48 hours (Supplementary Fig. S1A) orwith organoids (1�103 cells/well) in Transwells (Corning) for 96hours (Supplementary Fig. S1B). Organoid proliferation wasmeasured with CellTiter-Glo assay for ATP (Promega) or MTT(Promega). Inhibitors L-733,060, RP 67580 and Stattic were usedin some experiments.

Sensory denervation of KPCPdx1 mice7 day old KPCPdx1 mice underwent a single subcutaneous

injection of resiniferatoxin (RTX) or vehicle solutions. Earlyprophylactic denervation before the development of PanINsobviated the concern of possible interactions of RTX with PanINcells. Physiologic testing for somatic denervation was performedusing the capsaicin-induced eye wipe response.

IHC and immunofluorescenceFixed cells and deparrafinized murine and human sections were

subject to immunofluorescence and hematoxylin and eosin (H&E)analyses as per standard protocols. Organoids were stained inwholemount in chamber slides (Lab-Tek). Pancreas tissues were opticallycleared and stained as per established protocols for 3D analysis (19)and each view was scanned to a depth of 150 mm. Axon lengthdensities (mm/mm3) were calculated using Avizo 7.1 software.

Histopathologic analysisHistopathologic analyses were performed on de-identified

slides from 8-, 12-, and 16-week-old KPCPdx1 pancreas. For eachpancreas, 3 to 5 sections were sampled 150 to 200 mm apart and5 to 15 random views were taken for each section. Lesions

were classified as acinar to ductal metaplasia (ADM), PanIN1(1A and 1B; "early"), PanIN2/3 ("late") and PDAC (tumor) basedon the classification consensus (20). PanIN and stroma burdenwere expressed as the percentage of total analyzed surface areaoccupied per lesion as previously described (SupplementaryFig. S2A; ref. 21). Proliferative activity, which correlates with degreeof PanIN dysplasia (22), was assessed with Ki67 labeling. Cytoker-atin 19 (CK19) staining of tumors further confirmed the presenceof invasive cells in the stroma (Supplementary Fig. S2B). Thefraction of mice with tumors at all ages was calculated.

Flow cytometry and clonogenic assaysFlow cytometry on live PanIN organoid and PDAC cells

were performed using a FITC-conjugated anti-NK1-R antibody(Alomone) in a LSRFortessa 3 cell analyzer (BD Biosciences) basedon established protocols (14). tdTomatoþ or NK1-Rþ and NK1-R�

PanIN cells (using the conjugated anti-NK1-R antibody) fromKCMist1tdTmice were FACS sorted on a FACSAria III cell sorter (BDBiosciences) based on established protocols (14). In clonogenicassays, sorted cells were cultured in 50% Matrigel (MG; Corning)with neurons (5,000 sorted cells/well) or in organoid media(500 sorted cells/well) for 7 days (Supplementary Fig. S3A–S3D).

Reassociation assaysFACS sorted NK1-R� and NK1-Rþ cells were used to derive

NK1-Rlow and NK1-Rhi organoids, respectively (SupplementaryFig. S3A–S3D).NK1-Rlow andNK1-Rhi organoidsweredissociatedinto single cells and transduced with lentiviral constructs contain-ing pLKO.1-puro-CMV empty vector and pLKO.1-puro-CMV-TurboGFP (Sigma), respectively, as per the manufacturer's pro-tocol. NK1-Rlow and NK1-Rhi/GFP organoids were plated as singlecells at a 20:1 ratio (5 � 103 total cells/well) and imaged every 4hours for 4 days in a microscope incubation chamber. NK1-Rlow

andNK1-Rhi/GFPwere plated at a 1:1 ratio (2�104 total cells/well)and subject to flow cytometry analysis at several time points.

Quantitative real-time PCRRNA extraction and cDNA synthesis was done as described

previously (21). qPCR was performed on a QuantStudio 6 FlexSystem (Applied Biosystems). TaqMan Universal PCR MasterMix (Applied Biosystems) was used with chromogranin A (CgA)and Gapdh primers, Mm99999915_g1 and Mm00514341_m1(Thermo Fisher), respectively. All other cDNAs were amplifiedwith PowerUp SYBRGreenMasterMix (Applied Biosystems)withthe primers listed in Supplementary Table S1. Relative amountsof mRNA were calculated by the comparative DCt method withGapdh and/or Hprt as house-keeping genes.

Western blottingOrganoids were starved for 24 hours, dissociated and incu-

bated with media with or without SP. Cell extracts were preparedaccording to standard protocols using cell lysis buffer (RIPA;Pierce) with protease and phosphatase inhibitors (Roche).Membranes were incubated with antibodies and developed withTrident ECL (Genetex). Densitometric analysis was done inImageJ.

Substance P ELISASubstance P (SP) concentration was quantified using a Fluo-

rescent Immunoassay Kit as per the manufacturer's protocol(Phoenix Pharmaceuticals, Inc.).

PanIN Neuroendocrine Cells Promote Tumorigenesis

www.aacrjournals.org Cancer Res; 77(8) April 15, 2017 1869




Statistical analysisData were analyzed using GraphPad Prism (GraphPad Soft-

ware, Inc.) and were expressed as mean � SEM. Comparisonsbetween groups where data were normally distributed weremadewith the Student t test, and comparisons with categorical inde-pendent variables were performed using Fisher exact test. Detailedprotocols and standard procedures are described in Supplemen-tary Materials and Methods.

ResultsPanINs and PDAC cells demonstrate neurotropism

Increased global nerve density has been suggested in pan-creas from human PDAC (3) and murine PanINs (23) com-

pared with densities in pancreas from disease-free subjects.We aimed to characterize the nerve density in the immediatePanIN microenvironment compared with respective PanIN-freepancreas in isogenic mice to control for neural remodelingchanges that may occur with genotype and disease state vari-ables. We used 3D whole mount staining to reconstruct thefine intra-visceral axons (<5 mm) that are not easily visualizedon 2D histology (Supplementary Fig. S4A and S4B; ref 19). In12-week-old KPCPdx1 mice, axons were observed in close prox-imity to PanIN epithelial cells in the immediate microenviron-ment (Fig. 1A and B; Supplementary Video SV1). We quantifiedthe nerve density surrounding PanINs versus their respectivePanIN-free pancreas tissue in 12-week KCPdx1 mice, which have

Figure 1.

Innervation of PanINs. Three-dimensional immunofluorescenceprojections (scan depths, 150 mm) ofPanIN epithelium stained with E-cadherin (red), axons with the pan-neuronal marker PGP9.5 (green), andnuclei with DAPI (blue) in mousepancreas. A, Late PanIN in a 12-weekKPCPdx1mouse is amultilobar structure(arrows)with infiltrating axons (inset);scale bars, 25 mm. B, Increasedinnervation of an early PanIN lesion(yellow dashes) in a 12-week KCPdx1

mouse pancreas compared withadjacent pancreas tissue (asterisks);scale bars, 100 mm. C, Sampleprojections of axons in PanIN-freetissue and PanIN lesions used forquantification of nerve densities(PanIN epithelial staining not shown tohighlight axonal network); scale bars,100 mm. D, Axon densityquantifications (n ¼ 6–13 scans foreach group; n ¼ 3–4 mice per group).The data represent mean � SEM(two-tailed unpaired t test;��, P � 0.01; �� , P � 0.001).

Sinha et al.

Cancer Res; 77(8) April 15, 2017 Cancer Research1870




only early PanINs, and age-matched KPCPdx1 mice, which havemostly late PanINs (Fig. 1C and D; Supplementary Fig. S4C).KCPdx1 and KPCPdx1 mice had a 3.2 � 0.84 and 10.1 � 2.4-foldincrease in nerves densities compared with adjacent tissues,respectively. PanIN-associated fibers had a grossly aberrantbranching pattern compared with acinar axons. Similar to ourfindings, neoplasia-associated nerves in prostate cancer showevidence of new sprouting and axonogenesis (1).

Having observed markedly increased axon density aroundPanIN lesions in vivo, we investigated whether neoplastic pan-creatic cells actively recruited more axons. We cocultured DRGneurons with human PDAC or non-neoplastic HPDE cells in amicrofluidic system that allows axonal interactions and modelsintrapancreatic sensory innervation (Supplementary Fig. S1A).We used PDAC cell lines because of technical limitations of 3Dculture of organoids in the device. The cocultured PDAC celllines, A6L and MIA PaCa-2, recruited 3.30 � 0.31 and 3.43 �0.46 times more sensory axons, respectively, than the HPDEcells (Supplementary Fig. S4D and S4E). Cocultured axons alsoexpressed synapsin proteins, indicating active vesicular trans-port and neurotransmitter release (Supplementary Fig. S4F andS4G). Reflective of these in vitro findings, PanIN cells likelyactively recruit axons to their microenvironments. Interestingly,

mutations in axon guidance genes, such as semaphorins, inhuman PDAC are significant, associated with a worse prognosisand may explain increased axon recruitment (24).

PanIN neuroendocrine cells express the substance Pneuropeptide receptor

Because we observed axons in close proximity to PanIN epi-thelial cells and showed that PDAC cells robustly recruitedsensory axons, we hypothesized that nerves, like other stromalcomponents (21; for review see ref. 25), may communicate withand regulate the PanIN epithelium. The KPCPdx1 PanIN epitheliawere negative for expression of several nerve-responsive elementssuch as the neurotrophic receptors p75, TRKA, TRKB, and TRKCthat can be expressed in human PDAC (data not shown; ref. 4).The well-characterized sensory neuropeptide SP and its receptor,NK1-R, have been shown to mediate neurogenic inflammationin the pancreas (26) and promote mitogenesis in several cancers(9–11). We discovered that a subpopulation of PanIN epithelialcells in the KPCPdx1 mouse expressed NK1-R. These cellswere present as a low-abundance population in the PanIN epi-thelium and expressed NK1-R in a cytoplasmic and membranouspattern (Fig. 2A and Supplementary Fig. S5A–S5C). In areas ofPanIN-free pancreas, NK1-R was expressed only in islets (Fig. 2A;

Figure 2.

PanIN neuroendocrine cells express theneuropeptide receptor NK1-R. Confocalimmunofluorescence analyses of murineand human PanINs stained with E-cadherin (gray) and nuclei with DAPI(blue). A, PanIN in a KPCPdx1 pancreas(white dashes) next to an islet (yellowdashes). Note PanIN NK1-Rþ/CgAþ cells(arrowheads) and NK1-R�/CgAþ cells(arrows). B, Human PanIN lesion withNK1-Rþ /CgAþ cells (arrowheads) andNK1-R�/CgAþ cells (arrows). C,tdTomatoþ PanINs in a KCMisttdTpancreas. All PanIN NK1-Rþ cells aretdTomatoþ (arrowheads). All scale bars,25 mm. D, Quantification of NK1-Rþ

cells/PanIN. Seventy sections from miceof different ages (n ¼ 10) and 33 sectionsfrom patients (n ¼ 16) were quantified.The data represent mean � SEM(two-tailed unpaired t test; � , P � 0.05).






Supplementary Fig. S5A). NK1-R expression in islets raised thequestion of whether PanIN NK1-Rþ cells also displayed featuresof endocrine differentiation. PanIN NK1-Rþ cells expressed thepan-endocrine marker chromogranin A (CgA) but not insulin,confirming their identity as neuroendocrine cells.

We surveyed PDAC tumor sections from16 patients containingsynchronous PanIN lesions. NK1-Rþ and CgAþ neuroendocrinecells were identified in one ormore PanINs of all patients (Fig. 2B;Supplementary Fig. S5D and S5E). As in the mouse, NK1-Rexpression was limited to islets in disease-free areas. The frequen-cy of NK1-Rþ cells per lesion increased with PanIN grade inKPCPdx1 mouse but not in human PanINs (Fig. 2D). Even earlyKPCPdx1 PanINs composed of a few neoplastic cells also expressedthe NK1-R, suggesting a potential early and persistent role inPanIN evolution (Supplementary Fig. S5A).

BecauseNK1-RandCgAare generally expressedby islet cells,weinvestigated whether the NK1-Rþ cells were intrinsic neoplasticcells, or alternatively, of an endocrine lineage that had beenincorporated into the PanIN epithelium. We stained for NK1-Rin the lineage traced KCMist1tdTmouse that expresses tdTomatoonly in cells that have undergone Cre-mediated recombination asdirected by the acinar-specific Mist1 promoter (27). All NK1-Rþ

cells in the PanIN epithelium coexpressed tdTomato, confirming

that these were acinar cell-derived neoplastic PanIN epithelialcells (Fig. 2C; Supplementary Fig. S5F).

We incidentally saw expression of NK1-R on a subpopulationof CgAþ enteroendocrine cells (EEC) in human duodenal villi(Supplementary Fig. S6A), suggesting similarities between PanINand intestinal neuroendocrine cells. Even though PanIN neuro-endocrine cells were a subpopulation in the epithelium, wehypothesized they may have the potential to affect global PanINbiology through paracrine niche influences, similar to the wayEECs can indirectly regulate crypt stem cells (28).

Sensory neurons promote PanIN organoid proliferationTo investigate whether PanIN NK1-R expression could medi-

ate neuroepithelial cross-talk, we used KCPdx1 PanIN organoidsin coculture experiments. The organoid culture is a powerfultool to study noncancerous pancreatic epithelial cells, such asPanIN cells, as it allows for propagation of these cells withoutthe need of a mesenchymal niche while preserving the char-acteristics of the source epithelium (16, 29). All PanIN orga-noid lines studied expressed cytoplasmic and membranousNK1-R as detected on immunofluorescence staining and immu-noblotting (Fig. 3A and B). In contrast with the low abundanceNK1-Rþ cells in vivo, PanIN organoid cells uniformly expressed

Figure 3.

Sensory neurons increase PanIN organoid proliferation. A, Immunofluorescence confocal projection of a PanIN organoid stained with NK1-R (red), EpCAM(gray), and nuclei (blue); scale bars, 25 mm. B, Western blot analysis for NK1-R expression in mouse brain (positive control) and PanIN organoids(b-actin, loading control). C, Proliferation of cocultured PanIN organoids (RLU, random luminescence units; n ¼ 3–4). Note break in y-axis. Representativebright field (BF) images of viable MTT-stained organoids (arrowheads) in this experiment; scale bars, 200 mm. D, Effects of NK-1R inhibitors onorganoids cocultured with neurons (n ¼ 4–6). Monoculture controls not shown. RLU values are normalized to monoculture controls in each experiment.E, Live confocal images of organoid colonies from FACS sorted and cocultured tdTomatoþ single PanIN cells from a KCMisttdT mouse pancreas; scalebars, 500 mm. All cells are tdTomatoþ (inserts); scale bars, 50 mm. Organoid colony (>20-mm diameter) counts (n ¼ 2). The data represent mean � SEM(two-tailed unpaired t test; �� , P � 0.001; �� , P � 0.0001).

Sinha et al.





NK1-R, possibly indicating enrichment of receptor expressionin organoid culture. The universal expression of NK1-R inorganoid cells and minimal expression in PDAC cell lines wasconfirmed by flow-cytometry studies (Supplementary Fig. S7A).PanIN organoids also diffusely expressed CgA (SupplementaryFig. S7B).

To recapitulate in vivo conditions with the potential forreciprocal signaling between neurons and cancer cells, wecocultured DRG neurons with PanIN organoids (Supplemen-tary Fig. S1B). Neurons increased organoid proliferation to agreater degree than 3T3 murine fibroblast cells (38.7� � 3.5-fold vs. 9.3� � 0.80-fold), an established tumorigenic stromalcell type in the microenvironment (Fig. 3C; ref. 30). Neuron-mediated PanIN organoid proliferation could be partially butsignificantly blocked by two NK1-R antagonists L-733,060 andRP 67580, by 43.2 and 32.3%, respectively (Fig. 3D; Supple-mentary Fig. S7C). To confirm that the neuronal effect onorganoid proliferation was mediated by a soluble factor, weincubated organoids with neuronal conditioned media andfound 3.1� � 1.4-fold increased proliferation that was alsoblocked by L-733,060 by 31.5% (Supplementary Fig. S7D).Neuronal conditioned media contained SP (20–40 pg/mL)as measured by ELISA. The decreased degree of proliferationin conditioned media compared with coculture suggestedthat reciprocal cross-talk between organoids and neuronsmay help mediate neoplastic proliferation. Unlike neuronalcoculture, SP incubation alone did not cause organoid prolif-eration likely due to known peptide instability in solution(data not shown; ref. 31). Very limited studies have shownthat SP alone mediates proliferation of PDAC cell lines andthe data are inconsistent (32, 33). Because NK1-R antagonismonly partly blocked PanIN organoid proliferation, additionalneuronal factors such as nerve growth factor (4) and glial cell-derived neurotrophic factor (5) could drive PanIN oncogenesisbut were not explored in this study.

Having observed that DRG sensory neurons exerted a robustproliferative influence on PanIN organoids, we hypothesized thatneurons could support the survival of primary PanIN cells. FACS-sorted PanIN cells from the KCMist1tdT pancreas formed 6.1- �1.2-fold more organoid colonies in the presence of neurons(Fig. 3E) as additionally confirmed by the proliferation assay(Supplementary Fig. S7E). Thus, sensory neurons supported thegrowth of organoid colonies from single primary cells in a 3Dmatrix, a process that additionally requires multiple growthfactors (16).

PanIN neuroendocrine cells exert trophic influencesGiven their similarity to intestinal EECs that are known to

function as nerve-responsive niche cells (28), we investigatedwhether the subpopulation of NK1-Rþ PanIN cells could influ-ence global PanIN growth. In clonogenic assays, FACS sortedNK1-R� and NK1-Rþ cells from KCMist1Tdt PanINs were platedat low density with neurons or in organoid media (Fig. 4A;Supplementary Fig. S3A). Strengthening our studies of SP-NK1-R–mediated neuroepithelial cross-talk, only NK1-Rþ cellsresponded to DRG coculture and formed 1.9- � 0.09-fold moreorganoids than monoculture controls. Even though NK1-Rinhibitors partially inhibited organoid proliferation in ourprior studies, NK1-Rþ cells exclusively responded to neuronalcoculture. Thus, NK1-Rþ cells may be the sole sensory nerve-responsive PanIN cells in vivo. Surprisingly, the combination of

NK1-R� and NK1-Rþ cells formed the most colonies (3.5- �0.21-fold over control) compared with the individual cultures,suggesting that the neuron-responsive NK1-Rþ cells potentiatedglobal organoid growth. This effect was accentuated in organoidmedia where combined NK1-R� and NK1-Rþ cells formedapproximately 20-fold more colonies than the individual cul-tures. These data resonated studies of Paneth cells' strikingpotentiation of intestinal organoid growth (34).

Interestingly, conditioned media from NK1-R� and NK1-Rþ

cells did not influence the cells' clonogenic potential in recip-rocal experiments (data not shown). This indicated likely con-tact-dependent signaling. To better study NK1-Rþ contact-medi-ated influences, we derived and validated new NK1-Rlow andNK1-Rhi organoids (Supplementary Fig. S3A). Freshly sortedNK1-R� cells, although initially depleted for tachykinin receptor1 (Tacr1) relative to NK1-Rþ cells, expressed the receptor afterone week in organoid culture (Supplementary Fig. S3B). Becausethe NK1-R may support Wnt signaling (35), receptor upregula-tion and universal expression in our organoid progenitor culture(Fig. 3A) was not surprising. NK1-Rhi organoids were enrichedfor NK1-R protein on immunoblotting (Supplementary Fig.S3C) and did not contain any non-neoplastic cells as deter-mined by lox-KrasG12D genotyping (Supplementary Fig. S3D). Inthe reassociation assay, NK1-Rlow and NK1-Rhi/GFP organoidsformed composite organoids when plated as single cells at aroughly in vivo ratio of 20:1 (Fig. 4B and C; Supplementary Fig.S8A). This plating ratio was important so as to not drive thereaction in a biased direction. Surprisingly, the NK1-Rhi/GFP cellsmaintained restrictive expansion within isolated areas of thecomposite organoids andnohomogenousNK1-Rhi/GFP organoidswere seen. This finding was reminiscent of Paneth cell reassocia-tion with intestinal cells into distinct regions within heteroge-neous organoids (34). Furthermore, NK1-Rlow and NK1-Rhi/GFP

organoid reassociationwas a dynamic process with highlymobileorganoids that appeared to preferably merge heterogeneously(Supplementary Video SV2). NK1-Rhi/GFP cells also maintaineda constant low cell population in composite organoids over timewhen plated at equal starting ratios (Fig. 4D; SupplementaryFig. S8B). Strikingly, the NK1-Rhi/GFP cell frequencies (rangingfrom 6.7% � 0.11% to 9.9% � 2.9%) closely reflected theirobserved ratios in PanIN epithelium in vivo (Fig. 2D). Eventhough NK1-Rhi/GFP organoids were observed to grow compara-tively to NK1-Rlow organoids in individual cultures, they showedrestricted expansion in the composite cultures.

Further characterization of NK1-Rhi organoids with qPCRshowed significantly decreased expression of proposed pancre-atic, and potentially neoplastic, progenitor markers relative toNK1-Rlow cells, including Lgr5 (29) and doublecortin-likekinase-1 (Dclk1; Fig. 4E; refs. 14, 36). Hedgehog ligands, Shhand Ihh, which cooperate with mutant Kras in early PanINdevelopment and are potentially activated in neoplastic "pro-genitor" cells (37), were also decreased. The Wnt pathway is acritical mediator of progenitor biology (for review see ref. 38)and is notably activated in pancreatic injury and in organoidcultures (29). Like Hedgehog signaling, the canonical Wntpathway is also critical for early PanIN development (39).Despite expressing many Wnt ligands similar to NK1-Rlow

organoids, NK1-Rhi organoids had decreased activation of thedownstream canonical versus noncanonical pathway (Fig. 4E;Supplementary Fig. S8C) as well as decreased expression ofseveral members of the Wnt receptor, frizzled (Fzd), family.






(Supplementary Fig. S8D). Thus, the expression profile of NK1-Rhi cells was less "progenitor"-like than of NK1-Rlow cells.

Considering together their ability to greatly potentiate com-posite organoid growth despite being a restricted subpopula-tion and their lower expression of known pancreatic progenitormarkers and pathways, NK1-Rþ cells appeared to function astrophic "niche" cells with the potential to provide growthsignals to NK-1R� cells. The NK1-Rþ PanIN cells were alsoKi67� on immunofluorescence analysis, further suggesting thatthey were not proliferating progenitor cells in vivo (Supplemen-tary Fig. S8E).

SP stimulates STAT3 phosphorylation in PanIN organoidsBecause sensory neurons robustly increased PanIN organoid

proliferation and selectively signaled to NK1-Rþ PanIN cells, wehypothesized that SP should activate oncogenic pathways.Organoids incubated with SP were screened for known SP-mediated signaling pathways (9, 12, 40). SP rapidly phosphor-ylated JAK2 and STAT3 in the PanIN organoids but did notaffect MAPK or AKT phosphorylation (Fig. 5A and B). Stattic, asmall-molecule inhibitor of STAT3, decreased the proliferationof cocultured organoids by 50.8%, suggesting that the neuron-mediated organoid proliferation is significantly mediatedthrough STAT3 activation (Fig. 5C). Indeed, the NK1-Rþ PanINcells in vivo had a significantly increased p-STAT3 expression (by

35.4% � 7.9%) compared with NK1-R� cells within the samePanIN lesion, suggesting that NK1-Rþ PanIN cells may be moreresponsive to neuropeptide signaling (Fig. 5D).

Sensory denervation of KPCPdx1 mice decreases PanINprogression

Beacuse nerves were abundant in the PanIN microenviron-ment and promoted PanIN organoid proliferation, we hypoth-esized that sensory axons may play an important role in PanINgrowth in vivo. A possible role of sensory nerves on PanINtumorigenesis has been suggested by studies that show capsa-icin, a TRPV1 antagonist at high doses, decreases tumor growthin PDAC mouse models (41, 42). However, the mechanism forcapsaicin's chemoprotective effects and the specific role ofdecreased intrapancreatic innervation is unknown. We system-ically denervated KPCPdx1 mice with resiniferatoxin (RTX), asuper agonist of the TRPV1 receptor that causes highly selectiveand rapid degeneration of the TRPVþ sensory afferent nervefibers (43).

In these studies, effective sensory denervation was confirmedby both the capsaicin-induced eye wipe response as well assensory axon density counts in pancreas tissue. DenervatedKPCPdx1 mice had significantly decreased corneal sensitivity tocapsaicin for up to 12 weeks compared with control mice,indicating significant loss of TRPV1þ somatic innervation

Figure 4.

PanIN neuroendocrine cells exerttrophic influences. A, Clonogenicassays of FACS sorted NK1-R� andNK1-Rþ KCMisttdT single PanIN cells(n ¼ 3–4 for each experiment; RLU,random luminescence units). Data arenormalized to the monoculturecontrols and the NK1-R� group in thecoculture and organoid mediaexperiments, respectively. B and C,Z-stack live immunofluorescence andbright field (BF) images of thereassociation of single NK1-Rhi/GFP

(yellow and green; arrows) andNK1-Rlow (red) organoid cells intocomposite organoids; scale bars,100 mm. D, Population fractions ofNK1-Rlow (GFP�) and NK1-Rhi/GFP

(GFPþ) organoid cells culturedtogether as single cells at a 1:1 ratio (onday 0) as determined by FACSanalysis over time. E, qPCR analysis.The data represent mean � SEM(two-tailed unpaired t test; � ,P�0.05;��, P � 0.01; �� , P � 0.001; and�� , P � 0.0001).

Sinha et al.





(Supplementary Fig. S9A). The 16-week control mice weremore withdrawn and had a significantly decreased eye wiperesponse compared with earlier time points, likely reflectingtumor-associated morbidity. Although we had evidence forrobust somatic sensory denervation, it was critical to confirmintrapancreatic loss of TRPV1þ fibers as we hypothesized

local effects of nerves on PanINs. Pancreatic sensory nervedensities were quantified with SP staining, which overlapsexclusively with TRPV1 (Fig. 6E; Supplementary Fig. S9B;ref. 44). RTX-treated mice had a significant decrease in pancreassensory nerve density compared with control mice at 8 and12 weeks but not at 16 weeks, suggesting likely axon

Figure 5.

SP induces JAK2 and STAT3 phosphorylation in PanIN organoids. A, Western blot analysis for the proteins shown from PanIN organoids incubated withmedia � SP. B, Quantification of relative band densities (a.u., arbitrary units) of phospho-protein/total protein (b-actin, loading control; n ¼ 3). C,Proliferation of cocultured PanIN organoids with STAT3 inhibitor, Stattic (n ¼ 3). RLU (random luminescence units) values are normalized tomonoculture controls. D, Representative immunofluorescence confocal projection of a PanIN lesion (dashes) in a 12-week KPCPdx1 pancreas. p-STAT3(red) expression in NK1-Rþ cells (green; short arrows) versus NK1-R� cells (long arrows) within the same lesion. Nonspecific staining in PanIN lumen(asterisks); scale bar, 25 mm. Quantification of p-STAT3þ nuclei within the same PanIN lesions of 12-week control KPCPdx1 mice (n ¼ 5; n ¼ 17 scans). Thedata represent mean � SEM (two-tailed unpaired t test; � , P � 0.05; �� , P � 0.01; �� , P � 0.0001).






regeneration over time. Interestingly, at this later time pointthe regrowth of pancreatic visceral sensory fibers was notassociated with an increase in the eye wipe response (Supple-mentary Fig. S9A), suggesting maintenance of peripheralsomatic denervation.

Thedenervatedmice had reduced late-stage, but not early-stage,PanINburden at all time points comparedwith the control group,suggesting an influence on PanIN progression but not PanINinitiation (Fig. 6A and B). Supporting this interpretation, therewas no difference in ADM lesions between the two groups(Supplementary Fig. S9C). In another study, STAT3 inactivationin the neoplastic epithelium of a different PDAC murine modelalso prevented PanIN progression but not initiation (13). There-fore, STAT3 may have minimal or no effects on PanIN initiation,which most likely depends on cell-autonomous mutation-drivenmechanisms. Denervated mice also had a significant decrease inthe PanIN-associatedfibro-inflammatory stroma.Consistentwiththeir impaired PanIN progression, overall fewer denervated micedeveloped PDAC (7.7% of denervated vs. 50% of controlmice; Fig. 6C). The decreased dysplasia of denervated PanINswas further confirmed by a significantly lower Ki67 expression(Fig. 6D; Supplementary Fig. S9D). Furthermore, consistent with

the observation that SP activated STAT3 in PanIN organoids,PanINs in axotomized pancreata had profoundly decreasedSTAT3 phosphorylation in PanIN epithelial cells (Fig. 6D; Sup-plementary Fig. S10A and S10B) as well as in NK1-Rþ cells(Supplementary Fig. S11A). Interestingly, denervation also ledto a decrease in NK1-Rþ PanIN cell abundance (Fig. 6D). Takentogether, these findings demonstrated that decreased sensoryintrapancreatic innervation reduced PanIN progression to PDACin vivo, potentially through both impaired epithelial STAT3 acti-vation and neuroendocrine cell maintenance.

DiscussionPanIN progression is not only accompanied by accumulating

genetic mutations and cellular atypia but also the developmentof a complex tumor-associated stroma that promotes tumorprogression through direct and indirect influences on the neo-plastic epithelium (for review see refs. 25, 45). Here, wedescribe nerves as bona fide members of the PanIN microenvi-ronment and demonstrate that they support PanIN tumorigen-esis likely through cross-talk with a novel population of neo-plastic neuroendocrine cells (Fig. 7).

Figure 6.

Sensory denervation decreasesPanIN progression to PDAC. A,Representative H&E pancreassections. Normal pancreas (asterisks),PanINs (arrowheads), and highlydysplastic ducts of tumors (insets;arrows) are shown; scalebars, 200mm.B, Histopathologic quantifications ofPanIN and stroma burden (n ¼ 4–5mice per group per time point). C,The percentage of 8-, 12-, and 16-week-old mice with tumors (n ¼ 12control and n ¼ 13 denervated). D,Quantifications of Ki-67þ, p-STAT3þ

and NK1-Rþ cells per PanIN (n¼ 13–44PanINs per analysis and 3–5 miceper group). E, Representative 3Dimmunofluorescence projections ofsubstance P (SP) axons (green).Pancreatic axon densityquantifications (n ¼ 3 views permouse; n ¼ 3 mice per group per timepoint); scan depth, 150 mm; scale bars,50 mm. The data represent mean �SEM (two-tailed unpaired test orFisher exact test; � , P � 0.05;��, P � 0.01; �� , P � 0.001; and�� , P � 0.0001).

Sinha et al.





Other examples of neuronal influences on non-neuronaltumors via cross-talk with the neoplastic stem cell niche areemerging. For example, autonomic nerves signal directly togastric stem cells to upregulate the oncogenic Wnt pathway anddrive gastric tumorigenesis (2). Here, we propose neuromodu-lation of PanIN neuroendocrine cells that influences globalPanIN progression. Even though human PDACs commonlyexpress endocrine markers (46, 47), a functional role of endo-crine differentiation in PanIN and PDAC has not been estab-lished. In our study, NK1-Rþ neuroendocrine PanIN cellsappeared in the earliest murine neoplastic lesions, exclusivelyresponded to neuronal signaling and functioned as trophic"niche" cells in organoid culture. NK1-Rþ cells' dramatic poten-tiation of composite organoid proliferation depended on con-tact mediated signaling, the exact mechanism of which needs tobe elucidated. NK1-Rþ cells may serve as a source of Wnt ligandsthat are critical for PanIN tumorigenesis (39) and that candepend on close cell contact for signal transduction (48).Interestingly, the NK1-Rþ organoids highly expressed Wnt7b(Supplementary Fig. S8C), which mediates pancreatic progen-itor cell growth during development (49). To formally establishwhether NK1-Rþ cells are required for PanIN progression in vivo,future studies should employ such strategies as toxin-dependentelimination of NK1-Rþ cells in PanIN epithelium, as has beendone for DCLK1þ PanIN progenitor cells (36).

The NK1-Rþ cells were Ki67�in vivo but expectedly proliferatedunder permissive organoid culture conditions (29). Intriguingly,the NK1-Rþ cells recapitulated their in vivo role as a restrictedsubpopulation when associated with NK1-R� cells. Thus, neuro-nal activation of STAT3 in NK1-Rþ cells in vivo may promoteneuroendocrine cell maintenance rather than proliferation,reflecting the complex role of STAT3 in cancer epithelium (forreview see ref. 50). Furthermore, NK1-Rþ cells may exclusively

influence PanIN versus tumor biology as they were not expressedin tumors or PDAC cells lines. The long-term effects of intrapan-creatic sensory denervation on tumor progression and survivalrequire further investigation. Also, the SP–NK1-R axis is likely notinvolved in axon recruitment by PDAC cell lines that do notnotably express the NK1-R. Such time- and context-specific rolesof oncogenic factors are well described in PanIN and PDACtumorigenesis (25).

SP-induced JAK2 and STAT3 phosphorylation in the PanINorganoids provided evidence for neuropeptide activation ofa key transcription factor in PanIN cells. Epithelial STAT3activation by stromal cell–derived IL-6 is critical to Kras-medi-ated PanIN tumorigenesis (13). Here, we show neuronal SP inthe microenvironment can serve as an additional activator ofSTAT3, as axotomized mice had markedly decreased PanINp-STAT3 expression. Notably, p-STAT3 was also reduced inNK1-R� cells, suggesting the potential for a more complex roleof nerves in mediating stromal-epithelial signaling. Althoughwe did not explore the effects of sensory denervation in alter-ing the inflammatory compartment, our studies establishedan avenue of direct cross-talk between sensory nerves and thePanIN epithelium.

In summary, sensory nerves promote PanIN tumorigenesispotentially through direct cross-talk to unique neoplastic neu-roendocrine cells. Our studies define a unique subpopulationof neural-responsive PanIN neuroendocrine cells capable ofexerting trophic influences to stimulate global PanIN growth.These results highlight the heterogeneous nature of the neo-plastic epithelium and its associated microenvironment, andillustrate the potential for their dynamic interactions. Disrup-tion of neuronal influences, perhaps even at the tumor stage,may have therapeutic implications in the chemopreventionand/or treatment of human pancreatic cancer.

Figure 7.

Neuroepithelial cross-talk. Sensoryinnervation of a pancreatic PanINlesion. The sensory neuron residing inthe dorsal root ganglion (DRG)projects a visceral sensory afferentnerve into the pancreas. Vagal nodoseganglia (NG) projections are omitted.The terminal axon in themicroenvironment is in closeproximity to the PanIN epithelium. SPbinding its membrane receptor, NK-1Rexpressed on NK1-Rþ neuroendocrinecells (red), leads to STAT3phosphorylation and maintenance ofthe NK1-Rþ cells, which signal topotentiate PanIN growth. Treatmentwith resiniferatoxin (RTX) results inloss of the intrapancreatic sensoryfibers regardless of the origin of theprojections.






Disclosure of Potential Conflicts of InterestS.D. Leach reports receiving a commercial research grant from Bristol Myers

Squibb. No potential conflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: S. Sinha, Y.-Y. Fu, S.D. Leach, P.J. PasrichaDevelopment of methodology: S. Sinha, Y.-Y. Fu, Y. Zhong, M.G. Joo,O. Grbovic-Huezo, I.-H. Yang, S.D. Leach, P.J. PasrichaAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): S. Sinha, Y.-Y. Fu, M. Ketcham, K. Lafaro, J.A.Saglimbeni, J.M. Bailey, M.G. Joo, O. Basturk, Y. Park, D. Tuveson, P.J. PasrichaAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): S. Sinha, Y.-Y. Fu, P.J. PasrichaWriting, review, and/or revision of the manuscript: S. Sinha, A. Grimont,K. Lafaro, J.P. Melchor, L. Baker, S.D. Leach, P.J. PasrichaAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): S. Sinha, G. Askan, Y. Zhong,O.Grbovic-Huezo,O. Basturk, R.C. Kurtz, P.J. PasrichaStudy supervision: S. Sinha, Y. Zhong, S.D. Leach, P.J. PasrichaOther (experimental design and discussion input): A. Grimont, J.P. Melchor

AcknowledgmentsWe thank Dr. Mark Donowitz and Mr. John Gibas at JHH for microscopy

support. We thank Dr. Christine Iacabuzio-Donahue for histology support andproviding the A6L cells and Dr. Ming S. Tao for providing the HPDE cells. Weacknowledge the utilizedMSKCC core facilities supported by the Cancer CenterSupport Grant (CCSG) P30 CA008748.

Grant SupportThis work was supported by NIHR01DK073558 (P.J. Pasricha,

R01DK097087 (S.D. Leach), grants P30DK089502 and T32-DK007632(M. Donowitz), and P30 CA008748 Cancer Center Support Grant (CCSG)to MSKCC.

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 3, 2016; revised December 21, 2016; accepted December 26,2016; published OnlineFirst April 6, 2017.

References1. Magnon C, Hall SJ, Lin J, Xue X, Gerber L, Freedland SJ, et al. Autonomic

nerve development contributes to prostate cancer progression. Science2013;341:1236361.

2. Zhao CM, Hayakawa Y, Kodama Y, Muthupalani S, Westphalen CB,AndersenGT, et al.Denervation suppresses gastric tumorigenesis. Sci TranslMed 2014;6:250ra115.

3. Ceyhan GO, Bergmann F, Kadihasanoglu M, Altintas B, Demir IE, Hinz U,et al. Pancreatic neuropathy and neuropathic pain–a comprehensivepathomorphological study of 546 cases. Gastroenterology 2009;136:177–86.

4. Miknyoczki SJ, Lang D, Huang L, Klein-Szanto AJ, Dionne CA, Ruggeri BA.Neurotrophins and Trk receptors in human pancreatic ductal adenocarci-noma: expression patterns and effects on in vitro invasive behavior. Int JCancer 1999;81:417–27.

5. CeyhanGO,GieseNA, ErkanM, Kerscher AG,WenteMN,Giese T, et al. Theneurotrophic factor artemin promotes pancreatic cancer invasion. AnnSurg 2006;244:274–81.

6. Fasanella KE, Christianson JA, Chanthaphavong RS, Davis BM. Distribu-tion and neurochemical identification of pancreatic afferents in themouse.J Comp Neurol 2008;509:42–52.

7. Li Q, Peng J. Sensory nerves and pancreatitis. Gland Surg 2014;3:284–92.

8. Caterina MJ, Schumacher MA, Tominaga M, Rosen TA, Levine JD, Julius D.The capsaicin receptor: a heat-activated ion channel in the pain pathway.Nature 1997;389:816–24.

9. Steinhoff MS, von Mentzer B, Geppetti P, Pothoulakis C, BunnettNW. Tachykinins and their receptors: contributions to physiologi-cal control and the mechanisms of disease. Physiol Rev 2014;94:265–301.

10. Garcia-Recio S, Fuster G, Fernandez-Nogueira P, Pastor-Arroyo EM,Park SY, Mayordomo C, et al. Substance P autocrine signalingcontributes to persistent HER2 activation that drives malignantprogression and drug resistance in breast cancer. Cancer Res 2013;73:6424–34.

11. Gillespie E, Leeman SE, Watts LA, Coukos JA, O'Brien MJ, Cerda SR, et al.Truncated neurokinin-1 receptor is increased in colonic epithelial cellsfrom patients with colitis-associated cancer. Proc Natl Acad Sci U S A2011;108:17420–5.

12. KoonHW, ZhaoD, Zhan Y, Rhee SH,Moyer MP, Pothoulakis C. SubstanceP stimulates cyclooxygenase-2 and prostaglandin E2 expression throughJAK-STAT activation in human colonic epithelial cells. J Immunol2006;176:5050–9.

13. LesinaM, KurkowskiMU, Ludes K, Rose-John S, TreiberM, Kloppel G, et al.Stat3/Socs3 activation by IL-6 transsignaling promotes progression ofpancreatic intraepithelial neoplasia and development of pancreatic cancer.Cancer Cell 2011;19:456–69.

14. Bailey JM,Alsina J, RasheedZA,McAllister FM, FuYY, PlentzR, et al.DCLK1marks a morphologically distinct subpopulation of cells with stem cellproperties in preinvasive pancreatic cancer. Gastroenterology 2014;146:245–56.

15. Ouyang H, Mou L, Luk C, Liu N, Karaskova J, Squire J, et al. Immortalhuman pancreatic duct epithelial cell lines with near normal genotype andphenotype. Am J Pathol 2000;157:1623–31.

16. Boj SF, Hwang CI, Baker LA, Chio II, Engle DD, Corbo V, et al. Organoidmodels of human and mouse ductal pancreatic cancer. Cell 2015;160:324–38.

17. Jones S, Zhang X, Parsons DW, Lin JC, Leary RJ, Angenendt P, et al. Coresignaling pathways in human pancreatic cancers revealed by global geno-mic analyses. Science 2008;321:1801–6.

18. Winston J, Toma H, Shenoy M, Pasricha PJ. Nerve growth factor regulatesVR-1 mRNA levels in cultures of adult dorsal root ganglion neurons. Pain2001;89:181–6.

19. Fu YY, Tang SC. At the movies: 3-dimensional technology and gastroin-testinal histology. Gastroenterology 2010;139:1100–5.

20. Hruban RH, Adsay NV, Albores-Saavedra J, Anver MR, Biankin AV, BoivinGP, et al. Pathology of genetically engineered mouse models of pancreaticexocrine cancer: consensus report and recommendations. Cancer Res2006;66:95–106.

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

22. Klein WM, Hruban RH, Klein-Szanto AJ, Wilentz RE. Direct correlationbetween proliferative activity and dysplasia in pancreatic intraepithelialneoplasia (PanIN): additional evidence for a recently proposed model ofprogression. Mod Pathol 2002;15:441–7.

23. Stopczynski RE, Normolle DP, Hartman DJ, Ying H, DeBerry JJ, BielefeldtK, et al. Neuroplastic changes occur early in the development of pancreaticductal adenocarcinoma. Cancer Res 2014;74:1718–27.

24. Biankin AV, Waddell N, Kassahn KS, Gingras MC, Muthuswamy LB, JohnsAL, et al. Pancreatic cancer genomes reveal aberrations in axon guidancepathway genes. Nature 2012;491:399–405.

25. Sinha S, Leach SD. New insights in the development of pancreatic cancer.Curr Opin Gastroenterol. 2016 Jul 21. [Epub ahead of print].

26. Bhatia M, Saluja AK, Hofbauer B, Frossard JL, Lee HS, Castagliuolo I, et al.Role of substance P and the neurokinin 1 receptor in acute pancreatitis andpancreatitis-associated lung injury. Proc Natl Acad Sci U S A 1998;95:4760–5.

27. Shi G, Zhu L, Sun Y, Bettencourt R, Damsz B, Hruban RH, et al. Loss ofthe acinar-restricted transcription factor Mist1 accelerates Kras-inducedpancreatic intraepithelial neoplasia. Gastroenterology 2009;136:1368–78.

Sinha et al.





28. Amcheslavsky A, Song W, Li Q, Nie Y, Bragatto I, Ferrandon D, et al.Enteroendocrine cells support intestinal stem-cell-mediated homeostasisin Drosophila. Cell Rep 2014;9:32–9.

29. Huch M, Bonfanti P, Boj SF, Sato T, Loomans CJ, van de Wetering M,et al. Unlimited in vitro expansion of adult bi-potent pancreasprogenitors through the Lgr5/R-spondin axis. EMBO J 2013;32:2708–21.

30. Hwang RF, Moore T, Arumugam T, Ramachandran V, Amos KD, Rivera A,et al. Cancer-associated stromal fibroblasts promote pancreatic tumorprogression. Cancer Res 2008;68:918–26.

31. Watson SP. Rapid degradation of [3H]-substance p in guinea-pig ileumandrat vas deferens in vitro. Br J Pharmacol 1983;79:543–52.

32. Friess H, Zhu Z, Liard V, Shi X, Shrikhande SV, Wang L, et al. Neurokinin-1receptor expression and its potential effects on tumor growth in humanpancreatic cancer. Lab Invest 2003;83:731–42.

33. Li X, Ma G, Ma Q, Li W, Liu J, Han L, et al. Neurotransmitter substance Pmediates pancreatic cancer perineural invasion via NK-1R in cancer cells.Mol Cancer Res 2013;11:294–302.

34. Sato T, van Es JH, Snippert HJ, Stange DE, Vries RG, van den Born M, et al.Paneth cells constitute the niche for Lgr5 stem cells in intestinal crypts.Nature 2011;469:415–8.

35. Ilmer M, Garnier A, Vykoukal J, Alt E, von Schweinitz D, Kappler R, et al.Targeting the neurokinin-1 receptor compromises canonicalWnt signalingin hepatoblastoma. Mol Cancer Ther 2015;14:2712–21.

36. Westphalen CB, Takemoto Y, Tanaka T, Macchini M, Jiang Z, Renz BW,et al. Dclk1 defines quiescent pancreatic progenitors that promoteinjury-induced regeneration and tumorigenesis. Cell Stem Cell 2016;18:441–55.

37. Pasca di Magliano M, Sekine S, Ermilov A, Ferris J, Dlugosz AA, Hebrok M.Hedgehog/Ras interactions regulate early stages of pancreatic cancer. GenesDev 2006;20:3161–73.

38. Klaus A, Birchmeier W. Wnt signalling and its impact on development andcancer. Nat Rev Cancer 2008;8:387–98.

39. Zhang Y, Morris JPt, Yan W, Schofield HK, Gurney A, Simeone DM, et al.Canonical wnt signaling is required for pancreatic carcinogenesis. CancerRes 2013;73:4909–22.

40. Koon HW, Zhao D, Zhan Y, Moyer MP, Pothoulakis C. Substance Pmediates antiapoptotic responses in human colonocytes by Akt activation.Proc Natl Acad Sci U S A 2007;104:2013–8.

41. BaiH, LiH, ZhangW,Matkowskyj KA, Liao J, Srivastava SK, et al. Inhibitionof chronic pancreatitis and pancreatic intraepithelial neoplasia (PanIN) bycapsaicin in LSL-KrasG12D/Pdx1-Cre mice. Carcinogenesis 2011;32:1689–96.

42. Saloman JL, Albers KM, Li D, Hartman DJ, Crawford HC, Muha EA, et al.Ablation of sensory neurons in a genetic model of pancreatic ductaladenocarcinoma slows initiation and progression of cancer. Proc NatlAcad Sci U S A 2016;113:3078–83.

43. Karai L, Brown DC, Mannes AJ, Connelly ST, Brown J, Gandal M, et al.Deletion of vanilloid receptor 1-expressing primary afferent neurons forpain control. J Clin Invest 2004;113:1344–52.

44. Price TJ, Flores CM. Critical evaluation of the colocalization betweencalcitonin gene-related peptide, substance P, transient receptor potentialvanilloid subfamily type 1 immunoreactivities, and isolectin B4 binding inprimary afferent neurons of the rat and mouse. J Pain 2007;8:263–72.

45. Neesse A, Algul H, Tuveson DA, Gress TM. Stromal biology and therapy inpancreatic cancer: a changing paradigm. Gut 2015;64:1476–84.

46. Chen J, Baithun SI, Pollock DJ, Berry CL. Argyrophilic and hormoneimmunoreactive cells in normal and hyperplastic pancreatic ducts andexocrine pancreatic carcinoma. Virchows Arch A Pathol Anat Histopathol1988;413:399–405.

47. Eusebi V, Capella C, Bondi A, Sessa F, Vezzadini P,Mancini AM. Endocrine-paracrine cells in pancreatic exocrine carcinomas. Histopathology 1981;5:599–613.

48. Farin HF, Jordens I, Mosa MH, Basak O, Korving J, Tauriello DV, et al.Visualizationof a short-rangeWnt gradient in the intestinal stem-cell niche.Nature 2016;530:340–3.

49. Afelik S, Pool B, Schmerr M, Penton C, Jensen J. Wnt7b is required forepithelial progenitor growth and operates during epithelial-to-mesenchy-mal signaling in pancreatic development. Dev Biol 2015;399:204–17.

50. YuH, LeeH,HerrmannA, Buettner R, Jove R. Revisiting STAT3 signalling incancer: new and unexpected biological functions. Nat Rev Cancer 2014;14:736–46.






2017;77:1868-1879. Published OnlineFirst January 19, 2017.Cancer Res Smrita Sinha, Ya-Yuan Fu, Adrien Grimont, et al. Cross-talkPanIN Neuroendocrine Cells Promote Tumorigenesis via Neuronal

Updated version

10.1158/0008-5472.CAN-16-0899doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2017/01/19/0008-5472.CAN-16-0899.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/77/8/1868.full#ref-list-1

This article cites 48 articles, 17 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/77/8/1868.full#related-urls

This article has been cited by 5 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/77/8/1868To 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-16-0899

http://cancerres.aacrjournals.org/content/suppl/2017/01/19/0008-5472.CAN-16-0899.DC1

http://cancerres.aacrjournals.org/content/77/8/1868.full#ref-list-1

http://cancerres.aacrjournals.org/content/77/8/1868.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/77/8/1868


PanIN Neuroendocrine Cells Promote Tumorigenesis via Neuronal … · slides from 8-, 12-, and...

Documents

Transcript of PanIN Neuroendocrine Cells Promote Tumorigenesis via Neuronal … · slides from 8-, 12-, and...