Stomach-Specific Activation of Oncogenic KRAS and STAT3-Dependent Inflammation Cooperatively Promote Gastric Tumorigenesis in a Preclinical Model

Published OnlineFirst February 2, 2016; DOI: 10.1158/0008-5472.CAN-15-3089 Cancer Molecular and Cellular Pathobiology Research

Stomach-Speciﬁc Activation of Oncogenic KRAS and STAT3-Dependent Inﬂammation Cooperatively Promote Gastric Tumorigenesis in a Preclinical Model Stefan Thiem1, Moritz F. Eissmann1, Joachim Elzer1, Anna Jonas2, Tracy L. Putoczki3, Ashleigh Poh1,3, Paul Nguyen1,3, Adele Preaudet3, Dustin Flanagan4, Elizabeth Vincan4, Paul Waring5, Michael Buchert1, Andrew Jarnicki1,6, and Matthias Ernst1

Abstract

About 5% to 10% of human gastric tumors harbor oncogenic terized by excessive STAT3 activity. To assess the contribution of mutations in the KRAS pathway, but their presence alone is often STAT3 to the spontaneously developing gastric adenomas in insufficient for inducing gastric tumorigenesis, suggesting a gp130F/F mice, which carry a knockin mutation in the Il6 requirement for additional mutagenic events or microenviron- signal transducer (Il6st), we generated Tg(Tff1-CreERT2); fl fl mental stimuli, including inflammation. Assessing the contribu- Stat3 / ;gp130F/F mice that also harbor a conditional Stat3 knock- tion of such events in preclinical mouse models requires Cre out allele and found that tamoxifen administration conferred a recombinase–mediated conditional gene expression in stem or significant reduction in their tumor burden. Conversely, excessive þ progenitor cells of normal and transformed gastric epithelium. Kras activity in Tg(Tff1-CreERT2);KrasLSL-G12D/ ;gp130F/F mice We therefore constructed a bacterial artificial chromosome con- promoted more extensive gastric inflammation, metaplastic taining transgene (Tg), comprising the regulatory elements of the transformation, and tumorigenesis than observed in Tg(Tff1- þ trefoil factor 1 (Tff1) gene and the tamoxifen-inducible Cre CreERT2);KrasLSL-G12D/ mice. Collectively, our findings demon- recombinase (CreERT2)–coding sequence. The resulting Tg strate that advanced gastric tumorigenesis requires oncogenic (Tff1-CreERT2) mice were crossed with mice harboring condi- KRAS or BRAF in concert with aberrant STAT3 activation in tional oncogenic mutations in Kras or Braf. The administration of epithelial precursor cells of the glandular stomach, providing a þ tamoxifen to the resulting adult Tg(Tff1-CreERT2);KrasLSL-G12D/ new conditional model of gastric cancer in which to investigate þ and Tg(Tff1-CreERT2);BrafLSL-V600E/ mice resulted in gastric candidate therapeutic targets and treatment strategies. Cancer Res; metaplasia, inflammation, and adenoma development, charac- 76(8); 1–11. 2016 AACR.

Introduction ities (1). The most prevalent form of gastric cancer is intestinal- type gastric adenocarcinoma, which progresses from superficial Despite a recent decline in mortality and incidence, gastric gastritis through stages of chronic gastritis, atrophic gastritis, cancer still accounts for one of the largest cancer-related mortal- metaplastic transformation, and dysplasia to invasive adenocarcinoma (2). Although chronic infection with Helicobacter pylori (H. pylori) remains the main risk factor for intestinal-type gastric 1Cancer and Inflammation, Olivia Newton-John Cancer Research Insti- tute and School of Cancer Medicine La Trobe University, Heidelberg, cancer, others include high-salt diet and viral infections (2). Australia. 2Department of Multiple Sclerosis, The Florey Institute of Tumor initiation and progression is a multistep process that Neuroscience and Mental Health, Melbourne, Australia. 3Inflammation requires the acquisition of genetic alterations and growth-pro- Division, The Walter & Eliza Hall Institute of Medical Research and Department of Medical Biology University of Melbourne, Melbourne, moting conditions in the microenvironment. The most common Australia. 4Department of Anatomy and Neuroscience, University of activating mutations in gastric cancer driver genes affect KRAS, Melbourne, Melbourne, Australia. 5Department of Pathology, Univer- PI3K, and the receptor tyrosine kinases FGFR2, ERBB2, EGFR, 6 sity of Melbourne, Melbourne, Australia. School of Biomedical MET, and related signaling molecules (3). Meanwhile, chronic Sciences and Pharmacy, University of Newcastle, Newcastle, Australia. inflammation of the gastric mucosa, triggered by bacterial infec- Note: Supplementary data for this article are available at Cancer Research tion or other environmental factors, can initiate premalignant Online (http://cancerres.aacrjournals.org/). metaplastic transformation, which often results in the loss of acid- S. Thiem and M.F. Eissmann contributed equally to this article. producing parietal cells (4). Accordingly, the excessive inflamma- Corresponding Author: Matthias Ernst, Olivia Newton-John Cancer Research tory response of the gastric epithelium of gp130F/F mice, a vali- Institute and School of Cancer Medicine at La Trobe University, 145 Studley Rd, dated preclinical model for early stage intestinal-type gastric Heidelberg, Victoria 3084, Australia. Phone: 613-9496-9775; Fax: 613-9496- cancer in humans, is associated with intestinal metaplasia and 5334; E-mail: [email protected] inevitably triggers adenoma formation in the most distal part of doi: 10.1158/0008-5472.CAN-15-3089 the stomach, referred to as antrum (5). Meanwhile, aberrant 2016 American Association for Cancer Research. activation of Kras induces hyperplasia in many epithelial tissues

www.aacrjournals.org OF1

Thiem et al.

and, in the gastric epithelium, is associated with some hallmarks ATG initiation codon in the Tff1 locus. We injected the resulting of spasmolytic peptide–expressing metaplasia (SPEM) and pseu- BAC transgene in the pronucleus of CBB6F1-derived one-cell dointestinal metaplasia (6). However, oncogenic Kras activation embryos and genotyped transgenic founder mice and their off- in isolation is usually insufficient to trigger the formation of spring by PCR (refer to Supplementary Methods). gastric adenomas (7, 8). Thus, akin to the paradigms for the intestinal epithelium, gastric tumorigenesis may also depend on Tamoxifen administration the sequential acquisition of mutations within the precancerous Tamoxifen (Sigma T5648) was dissolved in 100% ethanol metaplastic epithelium or require a permissive tumor microen- (Sigma) and diluted 1:10 in sunflower oil (Sigma). Mice were fl vironment characterized by subclinical in ammation. injected intraperitoneally (100 mL) for 5 consecutive days with The formation of solid tumors is widely believed to depend on tamoxifen (1 mg/20 g body weight; twice daily) or vehicle, unless cancer stem cells, which, at least in the case of the intestinal indicated otherwise. The tamoxifen-containing emulsions were epithelium, express similar markers as their nontransformed prepared freshly every 3 days. tissue stem cells. Accordingly, the gastrointestinal stem cell marker locus Lgr5 can confer Cre recombinase–mediated cellular trans- b formation at the base of the glands of the pylorus and corpus -Gal staining region of the stomach (9). However, the higher abundance of Stomachs and intestines were opened longitudinally, washed Lgr5-positive stem cells in the intestinal epithelium results in three times in PBS by vigorous shaking, and pinned out on fi lethal intestinal tumorigenesis, preventing the analysis of gastric silicone-coated plates. Tissues were pre xed in 4% paraformal- fl fl tumor formation in corresponding Lgr5CreERT2;Apc / mice, which dehyde (PFA; Sigma) for one hour at 4 C, washed three times with b harbor conditional null alleles of the tumor suppressor gene Apc -Gal Wash Buffer (20 minutes/room temperature), and incu- fi bated in b-Gal Staining Solution (refer to Supplementary Meth- (9). Meanwhile, Cre expression that is largely con ned to com- mitted gastric epithelial progenitors in Atp4b-Cre or Tg(Tff2- ods) at 37 C overnight. Photographs were taken and the organs fi CreERT2) mice is insufficient to reproducibly and reliably trans- were xed in 2% PFA (Sigma) at 4 C overnight, transferred to 70% fi duce tumorigenesis, and neither of these Cre-drivers have been ethanol, paraf n sectioned, and counterstained with Nuclear Fast tested for their capacity to confer recombinase activity within Red. established tumors (10, 11). We therefore exploited the regulatory sequences of the stomach-specific trefoil factor 1 (Tff1) gene, YFP detection which is primarily expressed in the epithelium of the glandular Following stomach dissection, specimens were washed in PBS, stomach within the corpus and antrum, and to a lesser extent also fixed in 4% PFA solution (Electron Microscopy Sciences) over- in the neoplastic counterpart in gastric cancer (12, 13). Impor- night, incubated in 30% and 50% sucrose/PBS (w/v) solution, tantly, Tff1 deficiency can give rise to gastric tumorigenesis in mice embedded in OCT (Tissue-Tek Sakura), and snap frozen on dry through a cell-intrinsic mechanism (14), arguing that Tff1-expres- ice. Frozen stomach sections (8 mm) were counterstained with sing gastric epithelium can provide the cell of origin for gastric DAPI before mounting. YFP (shown as green) and DAPI (shown cancer. We therefore exploited bacterial artificial chromosome as blue) were imaged using a confocal microscope (Zeiss LSM (BAC) transgenesis to confer tamoxifen-inducible CreERT2 activ- 780). ity in the glandular stomach of corresponding transgenic Tg(Tff1- CreERT2) mice. This confers long-term retention of conditional Histologic and immunohistochemical analysis b b fl -galactosidase ( -Gal) or yellow uorescent protein (YFP) Following dissection, tissues were washed in PBS, fixed in 10% reporter activity, or of somatically mutated genes, in the gastric neutral-buffered formalin solution (pH 7.4) overnight, and pro- epithelium and in corresponding tumor cells. cessed. Sections (4 mm) were stained with hematoxylin and eosin, Alcian blue or periodic acid-Schiff. Immunohistochemical anal- Materials and Methods ysis was performed as described previously (5).Histopathology Mice and study approval scoring of gastric mucosa and tumors was performed as described Homozygous mice for a mutated version of the IL6 signal previously (21, 22). transducer gp130, encoded by the Il6st gene, and referred to as Y757F F/F fl/fl gp130 (gp130 ), or for a floxed Stat3 allele (Stat3 ), as well Protein extraction and immunoblot analysis LSL-G12D/þ LSL-V600E/þ as heterozygous Kras , Braf ; LSL-LacZ13 Protein lysates from snap-frozen tissues were prepared using (LacZ), and Rosa26-LSL-YFP (YFP) mice were propagated on a the TissueLyser II (Qiagen) and RIPA lysis buffer (Sigma), sepa- mixed C57BL/6 129/Sv background (15–20). Mice were rated by SDS-PAGE, and transferred to nitrocellulose membranes cohoused under specific pathogen-free conditions, and age- and by iBlot (Invitrogen). Protein bands were visualized and quanti- gender-matched littermates were used for experiments. Animal fied using the Odyssey Infrared Imaging System (LI-COR studies were approved by the Animal Ethics Committees of the Biosciences). Ludwig Institute for Cancer Research Institute, the Walter and Eliza Hall Institute (Melbourne, Australia), and the Olivia New- Statistical analysis ton-John Cancer Research Institute (Heidelberg, Australia). Unless otherwise stated, comparisons among mean values were performed by ANOVA or a two-tailed Student t test as appro- Generation of the transgenic Tff1-CreERT2 BAC priate using Prism 5 software (GraphPad). P < 0.05 was consi- To generate the Tff1-CreERT2 BAC, we used "DNA recombi- dered statistically significant. neering" to insert the CreERT2 open reading frame with a poly- Further information is provided in the Supplementary adenylation signal immediately downstream of the endogenous Methods.

OF2 Cancer Res; 76(8) April 15, 2016 Cancer Research

Kras and Stat3 Drive Gastric Tumorigenesis

Figure 1. Tff1-CreERT2-mediated recombination pattern in the stomach. A, representative sections (left) and whole mount (right) of stomachs from Tg(Tff1-CreERT2);LacZ mice stained for b-Gal activity 6 hours and 120 days after a single injection of tamoxifen (TAM; 1 mg/20 g bodyweight). Vehicle control images are shown in Supplementary Fig S1E. Sections were counterstained with Nuclear Fast Red. Scale bars, 50 mm. B, quantiﬁcation of the staining pattern of well-oriented glands for b-Gal activity presented in A (absent, no blue cells within gland; partial, at least one blue cell within gland; and full, all blue cells in gland). Graphs show mean SEM, n 3. C, representative immunoﬂuorescence image of antrum sections of Tg(Tff1-CreERT2);YFP mice 6 hours after a single administration of tamoxifen (1 mg/20 g bodyweight) or oil as vehicle control (no TAM). Merged images show DAPI in blue and YFP in green. Scale bars, 50 mm.

Results Likewise, the absence of reporter activity in the forestomach, and the gradually increased expression of the Tff1-CreERT2 transgene Generation of Tg(Tff1-CreERT2) mice from the squamous forestomach to the antrum, with a sharp To confer Cre activity to the gastric epithelium without affecting decrease towards the proximal small intestine, coincided with a endogenous gene expression, we took advantage of BAC transgen- similar expression pattern of endogenous Tff1 (Supplementary esis and DNA recombineering. We generated a BAC vector encod- Fig. S2B; refs.12, 13). We also observed b-Gal–positive cells ing the CreERT2 recombinase inserted at the translational initi- towards the base of antral glands, where Lgr5-positive stem cells ation sites of the Tff1 locus (Supplementary Fig. S1A–S1C). We reside (9). Indeed, FACS-sorted GFP-positive cells from stomachs derived four independent Tg(Tff1-CreERT2) founder lines and of Lgr5CreERT2-IRES-GFP mice also expressed Tff1, although we found confirmed retention of the transgenic BAC after germline trans- stronger Tff1 expression in GFP-negative stomach epithelial cells mission (Supplementary Fig. S1D). (Supplementary Fig. S2C; ref. 2). Consistent with expression of the Tff1-CreERT2 transgene in long-lived stem cells, approximate- Tff1-CreERT2 mediates long-term somatic mutagenesis in the ly 20% of antral glands stained entirely b-Gal–positive 120 days glandular stomach after tamoxifen administration, with another 40% staining par- We first determined CreERT2 transgene activity through the tially positive for the reporter (Fig. 1B). However, we did not assessment of b-Gal or YFP activity conferred by activation of the detect completely b-Gal–traced glands in the corpus, suggesting conditional LacZ or YFP reporter allele in the corresponding Tg that the Troy-positive stem cells at the base of the corpus, which (Tff1-CreERT2) compound mutant strain. Six hours after a single can reconstitute entire corpus glands (23), do not express the Tff1- tamoxifen administration (1 mg/20 g body weight), we detected CreERT2 transgene. b-Gal or YFP-positive cells in the pit regions of gastric glands in the To induce more widespread and uniform CreERT2 activity, we corpus and antrum and also often along entire antral glands (Fig. administered tamoxifen (1 mg/20 g mouse, twice daily) over 5 1A–C and Supplementary Fig. S1E). These staining patterns consecutive days. Three days later, we again observed >20% of suggest CreERT2 expression within all cells of the antral glands entirely b-Gal–traced glands in the antrum, but not in the corpus but only some cells in the corpus, thereby reflecting the distribu- (Fig. 2A and B). Over a 360-day follow-up period, the frequency of tion of endogenous Tff1 protein (Supplementary Fig. S2A). entirely stained glands increased slightly in the antrum, but was

www.aacrjournals.org Cancer Res; 76(8) April 15, 2016 OF3

Thiem et al.

Figure 2. Long-term retention of b-Gal activity in the stomach of Tg(Tff1-CreERT2) mice. A, representative whole-mount stomachs from Tg(Tff1-CreERT2);LacZ mice stained for b-Gal activity following 5-day tamoxifen (TAM) treatment protocol (1 mg/20 g body weight; twice daily) and follow-up period as indicated. Corresponding sections were counterstained with Nuclear Fast Red. Boxed regions are shown at higher magniﬁcations. Scale bars, 100 mm. B, quantiﬁcation of the b-Gal staining pattern of well-oriented glands of the antrum and corpus of tissue presented in A (absent, no blue cells within gland; partial, at least one blue cell within gland; and full, all blue cells in gland). Graphs show mean SEM, n 3.

never observed in the corpus, where only 20% of all oxyntic glands goblet-like cells (Fig. 3C) that stained positive for acidic mucus showed some b-Gal staining. A transient increase of entirely and neutral glyco and mucoproteins (Supplementary Fig. S4B). stained glands 10 days after tamoxifen administration is consis- The tumor-adjacent antral mucosa showed similar signs of tent with the mucosal repair elicited by the transient loss of mucus metaplasia, coinciding with epithelial hyperplasia and parietal cell associated with high doses of tamoxifen administra- mononuclear cell infiltrates in the lamina propria (Fig. 3C and tion (i.e., >3 mg/20 g body weight), and which spontaneously Supplementary Fig. 4C). In contrast, the corpus revealed only resolves within three weeks (24). In contrast, stomachs of vehicle- mild inflammation and occasionally dilated and disorganized treated Tff1-CreERT2–positive or tamoxifen-treated Tff1- glands, which therefore clearly demarcated the corpus–antrum CreERT2–negative littermates showed no b-Gal staining (Supple- border in KrasG12D-expressing mice (Fig. 3C). mentary Fig. S3A). We also observed b-Gal and YFP (data not As expected from the widespread recombination of the latent shown) staining in the Brunner glands and in some airway KrasLSL-G12D allele (Supplementary Fig. S4A), we observed stain- epithelial cells (Supplementary Fig. S3B), but not in the pancreas ing of the phosphorylated (activated) Erk1/2 isoforms at the (Supplementary Fig. S3C and S3D). As all four transgenic founder luminal edges of KrasG12D-induced tumors and more prominent- lines yielded similar reporter staining patterns, we concluded that ly in the adjacent hyperplastic antrum, irrespective of the presence the Tff1-CreERT2 transgene replicates the expression of endoge- of tumors (Fig. 4A and B). Surprisingly, all tumors in Tg(Tff1- þ nous Tff1 (25, 26), and confers recombinase activity to long-lived CreERT2);KrasLSL-G12D/ mice displayed extensive staining of the stem/progenitor cells, particularly within the antrum. activated, tyrosine-phosphorylated form of Stat3 (Fig. 4A and B), which mediates the tumor-promoting activity of the inflamed Activation of KrasG12D or BrafV600E in the gastric epithelium microenvironment (27, 28). Collectively, these results suggest induces metaplasia, gastritis, and tumor formation that oncogenic KrasG12D triggers excessive Mek/Erk pathway acti- To explore whether the oncogenic KrasG12D mutation in epi- vation, which, in a tumor-specific manner, coincides with aber- thelial stem/progenitor cells is sufficient to initiate tumorigenesis rant Stat3 activation, as indicated by excessive staining for phos- in the gastric antrum, we treated adult Tg(Tff1-CreERT2); photyrosine (pY)-Stat3. þ KrasLSL-G12D/ mice with tamoxifen. Nine months later, we We detected widespread expression of the SPEM marker Tff2 detected large adenomas in one third of the mice analyzed (27, 29) in all KrasG12D-induced adenomas (Fig. 4C) and at the (Fig. 3A; Supplementary Table S1), whereas stomachs of tamox- base of the adjacent antral mucosa (Supplementary Fig. S4D). ifen-treated Tg(Tff1-CreERT2)-negative littermates remained This coincided with Alcian blue staining in cells of the deep histologically normal and lacked evidence for the recombined antral glands (Supplementary Fig. S4B). We also detected signs KrasG12D allele (Fig. 3A and B and Supplementary Fig. 4A). of pseudointestinal metaplasia within the tumor-adjacent glan- KrasG12D-induced tumors were characterized by glandular dular epithelium and in the adenomas, comprising prominent structures at the luminal edges and regions with intestinal appearance of goblet-like cells alongside the induction of the

OF4 Cancer Res; 76(8) April 15, 2016 Cancer Research

Kras and Stat3 Drive Gastric Tumorigenesis

Figure 3. Activation of oncogenic Kras in the gastric epithelium triggers tumorigenesis, gastritis, and metaplasia. A, representative whole-mount stomachs from KrasLSL-G12D/þ mice of the indicated Tff1-CreERT2 genotypes, collected nine months after tamoxifen administration (5 consecutive days; 1 mg/20 g body weight; twice daily). H&E- stained cross-sections, cut along the dotted line, extend from the fundus to the proximal end of the small intestine. Tumors are outlined with red circles (arrows), and boxed regions are shown at higher magniﬁcations in B and C. B and C, the corpus and antrum of tamoxifen-treated CreERT2-negative mice (B) remain histologically normal, whereas the mucosa of tamoxifen-treated CreERT2-positive mice (C) shows a sharp morphologic demarcation at the corpus/antrum border (C, ii, arrowheads). The predominantly normal corpus exhibits some disorganized and dilated glands (C, i), whereas the frequently distorted glands in the antrum are associated with severe chronic gastritis (C, iv) and extensive mucus metaplasia with goblet-like cells (C, iii). The tumors display glandular structures at the luminal edges (C, v) and mucus-containing, goblet-like cell clusters (arrows) in the core (C-vi). Scale bars, 50 mm.

intestinal marker gpA33 (30, 31) and to a lesser extent of Cdx2 response to Tff1-CreERT2 activation or transgenic expression (Fig. 4C and D and Supplementary Fig. S4E and S4F). Consis- from the K19 promoter (6). tent with the discordant expression between gpA33 and the To establish whether the observations associated with Tg(Tff1- þ intestinal master regulator Cdx2, with the latter being expressed CreERT2);KrasLSL-G12D/ mice also occurred with other mutations at the base of the glands (31, 32), ectopic expression of the within the Ras–Erk pathway, we generated Tg(Tff1-CreERT2); þ intestinal marker Cdx2 was limited to some epithelial cells at BrafLSL-V600E/ compound mice to mimic a rare mutation in the base of antral glands in tumor-bearing mice (Supplemen- human gastric cancer (34). We observed large adenomas in the þ tary Fig. S4E). Meanwhile, we detected the intestinal mucin antrum of 10/15 Tg(Tff1-CreERT2);BrafLSL-V600E/ mice 8 months Muc2 only in a few tumor cells (Fig. 4C), albeit reminiscent of after tamoxifen administration (Supplementary Fig. S5A; Supple- MUC2 expression in humans being most prominent in gastric mentary Table S1). Strikingly, these lesions replicated the histo- þ cancer tissues (33). Thus, SPEM and pseudointestinal metapla- pathologic features observed in Tg(Tff1-CreERT2);KrasLSL-G12D/ sia appear in transformed hyperplastic epithelium and tumors mice, including disorganized glandular structures, SPEM, and irrespective of the mucosal cell type(s) that express KrasG12D in regions of pseudointestinal metaplasia with goblet-like cells that

www.aacrjournals.org Cancer Res; 76(8) April 15, 2016 OF5

Thiem et al.

Figure 4. KrasG12D-induced tumorigenesis is associated with expression of markers of metaplasia and Stat3 activation. A and C, representative immunohistochemical staining for pErk1/2 and pY-Stat3 (A) and Tff2, Muc2, and gpA33 (C) on antrum and tumor sections from tamoxifen-treated KrasLSL-G12D/þ mice of the indicated Tff1-CreERT2 genotypes. Scale bars, 500 mm (A, inset, 100 mm), 100 mm (C, inset, 20 mm). B and D, immunoblot analysis of antral and pooled gastric tumor tissue from individual tamoxifen-treated KrasLSL-G12D/þ mice of the indicated Tff1-CreERT2 genotypes (B). Intestinal tissue was used as positive control (D).

stained positive for acidic mucus (Supplementary Fig. S5A–S5C). gene" inactivation in the gastric mucosa and established Predictably, the antrum of all BrafV600E-expressing mice showed tumors. elevated Mek1/2 and Erk1/2 phosphorylation, irrespective of the presence of tumors, and all tumors also displayed extensive KrasG12D activation in gp130F/F mice exacerbates metaplasia and staining of pY-Stat3 (Supplementary Fig. S5D and S5E). increases tumor burden and progression Although tumor latency and incidence differs between þ Tff1-CreERT2 confers recombinase activity to the metaplastic gp130F/F and Tg(Tff1-CreERT2);KrasLSL-G12D/ mice (Supple- epithelium in established gastric tumors mentary Table S1), both models are associated with hallmarks Homozygous gp130F/F mice, which carry a constitutive tyrosine of inflammation and metaplastic transformation. Furthermore, to phenylalanine substitution mutation at amino acid 757 to as Stat3 activity is moderately elevated in KrasG12D-induced prevent binding of the negative regulator Socs3, as well as acti- gastric tumors (Fig. 4A and B), we hypothesized that further vation of the Ras signalling cascade, spontaneously develop augmentation of Stat3 and other signaling molecules down- gastric adenomas in response to excessive IL11-dependent Stat3 stream of Kras in compound gp130F/F;Tg(Tff1-CreERT2); þ activation with molecular hallmarks of intestinal-type gastric KrasLSL-G12D/ mice may increase tumor frequency and progres- cancer (5, 35). Although Tff1 expression is reduced in tumors of sion of the tubular adenomas that spontaneously arise in gp130F/F mice (Supplementary Fig. S6A; ref.15), we observed gp130F/F mice. Indeed, 3 months after tamoxifen administra- extensive b-Gal staining throughout the tumor epithelium of tion to 8-week-old tumor-bearing gp130F/F;Tg(Tff1-CreERT2); þ gp130F/F;Tg(Tff1-CreERT2);LacZ mice and detected robust, long- KrasLSL-G12D/ mice, we detected increased tumor numbers term retention of b-Gal (Supplementary Fig. S6B). Although the when compared with those from the gp130F/F;Kras wild-type latter was concentrated towards the tumor edges, immunohisto- (Kras-wt) cohort, which comprised gp130F/F;Tg(Tff1-CreERT2) þ chemical analysis for b-Gal revealed staining of most tumor and gp130F/F;KrasLSL-G12D/ mice (Fig. 6A). glands (Supplementary Fig. S6C). We therefore hypothesized that Surprisingly, the extent of epithelial hyperplasia and the extent the Tff1-CreERT2 transgene–mediated genetic ablation of tumor of inflammatory cell infiltration were similar in tumors of all promoters could counteract their effect, and treated tumor-bear- genotypes (Fig. 6B and C). Strikingly however, goblet-like cells fl fl fl fl ing gp130F/F;Tg(Tff1-CreERT2);Stat3 / and gp130F/F;Stat3 / con- and associated mucus metaplasia as well as dysplastic regions trol mice, harboring a "floxed" Stat3 allele, with tamoxifen. We were consistently more prominent in tumors of gp130F/F;Tg(Tff1- þ observed a significant reduction in tumor burden and numbers CreERT2);KrasLSL-G12D/ mice (Fig. 6B–E). Importantly, approx- in mice of the CreERT2-positive cohort four weeks later (Fig. imately half of all tumors analyzed from 3-month-old gp130F/F;Tg þ 5A–C). This coincided with reduced levels of total Stat3 expres- (Tff1-CreERT2);KrasLSL-G12D/ mice contained clusters of submu- sion in the tumors and adjacent mucosa and a reduction of cosal glands, likely to reflect an increased propensity of the nuclear accumulation of the pY-Stat3 (Fig. 5D–F). Thus, the infrequent epithelial invasion through the muscularis mucosae Tff1-CreERT2 transgene is sufficient to effectively confer "onco- seen in 8- to 10-month-old gp130F/F mice (Fig. 6C and E; ref. 36).

OF6 Cancer Res; 76(8) April 15, 2016 Cancer Research

Kras and Stat3 Drive Gastric Tumorigenesis

Figure 5. Epithelial Stat3 is required for gastric tumor growth in gp130F/F mice. A, schematic representation of spontaneous gastric tumor growth in gp130F/F mice and of fl fl the tamoxifen (TAM) treatment protocol used (1 mg/20 g body weight; once daily). B, representative whole-mount stomachs of 12-week-old gp130F/F;Stat3 / littermates of the indicated Tff1-CreERT2 genotypes, collected 4 weeks after tamoxifen administration. Tumors are outlined with black circles. C, total weight (left) fl fl and number (right) of tumors resected from individual gp130F/F;Stat3 / mice of the indicated Tff1-CreERT2 genotypes (n ¼ 9/cohort), collected 4 weeks after tamoxifen administration. Horizontal lines refer to mean values ( , P < 0.01 and , P < 0.001). D, immunoblot analysis of antral and pooled gastric tumor tissue from individual gp130F/F;Stat3fl/fl mice of the indicated Tff1-CreERT2 genotypes showing reduced Stat3 levels in tamoxifen-treated Tff1-CreERT2–positive mice. Numbers refer to relative intensities of the pY-Stat3 or Stat3 bands normalized for the amount of b-actin, with the Cre samples arbitrarily being set as 1. E and F, representative immunohistochemical staining for pY-Stat3 (E) and total Stat3 (F) on tumor and antrum sections from 12-week-old gp130F/F;Stat3fl/fl littermates of the indicated Tff1-CreERT2 genotypes, collected 4 weeks after tamoxifen administration. In Tff1-CreERT2–positive mice, many epithelial cells had lost Stat3 expression (inset), whereas others still showed Stat3 staining (arrowheads). Scale bars, 100 mm (inset, 50 mm).

www.aacrjournals.org Cancer Res; 76(8) April 15, 2016 OF7

Thiem et al.

Figure 6. KrasG12D activation in established gastric tumors increases tumor burden and formation of submucosal glands. A, total tumor number for individual Tff1-CreERT2–positive mice of the indicated genotypes, collected at the indicated time after tamoxifen (TAM; 5 consecutive days; 1 mg/20 g body weight; twice daily) administration (n 4/cohort; , P < 0.05; , P < 0.01; and , P < 0.001). B, differential pathology scores demonstrating the severity of the indicated conditions in gastric tumors of Tff1-CreERT2–positive mice of the indicated genotypes (n 4/cohort; , P < 0.05). A score of 3 denotes the highest pathologic severity, and a score of 0 denotes the normal (wild type) condition (21, 22). C, representative H&E-stained tumor sections from Tff1-CreERT2–positive mice of the indicated genotypes, collected three months after tamoxifen administration (5 consecutive days; 1 mg/20 g body weight; twice daily). Boxed regions are shown at higher magniﬁcation in D and E, and the dotted gray lines indicate the border between the hyperplastic/dysplastic glandular epithelium and the underlying muscularis mucosae. Scale bars, 1 mm. D and E, sections were stained with H&E, Alcian blue (AB), or periodic acid-Schiff (PAS). Tumors from gp130F/F;Kras-wt control mice (D) have an organized glandular appearance (D, i), with only subtle þ metaplastic transformation and mucus production (D, ii). Tumors from gp130F/F;KrasG12D/ mice (E) reveal disorganized and dysplastic structures (E, i), extensive goblet cell-like/mucus metaplasia (E, ii), and submucosal glands (E, iii). Scale bars, 100 mm.

OF8 Cancer Res; 76(8) April 15, 2016 Cancer Research

Kras and Stat3 Drive Gastric Tumorigenesis

Figure 7. Coactivation of Kras and gp130 signaling induces intestinalization of the gastric mucosa. A, H&E, Alcian blue (AB), and periodic acid-Schiff (PAS)–stained sections of the antral gastric mucosa of Tff1-CreERT2–positive mice of the indicated genotypes, collected at the indicated time after tamoxifen (TAM) administration (5 consecutive days; 1 mg/20 g body weight; twice daily). Boxed regions are shown at higher magniﬁcations. Scale bars, 100 mm. B, differential pathology scores of the gastric antrum (top) and corpus (bottom) of Tff1-CreERT2–positive mice of the indicated genotypes (n 4/cohort; , P < 0.05). A score of 3 denotes the highest pathologic severity, and a score of 0 denotes the normal (wild type) condition (21, 22).

We morphologically investigated the nontumor mucosa for and conclude that the employed BAC-based transgene confers and quantifiable evidence of metaplasia and compared stomachs restricts CreERT2 activity to Tff1 expressing epithelial cells along þ from 3-month-old CreERT2-positive gp130F/F;KrasLSL-G12D/ and the entire length of the antral glands and some cells in the pit gp130F/F;Kras-wt mice with those obtained from 9-month-old region of the corpus (12). As the gastric mucosa continuously and þ þ þ CreERT2-positive gp130 / ;KrasLSL-G12D/ mice. Despite the dif- rapidly undergoes complete renewal, the abundant long-term ferent ages, we found a similar prevalence of glandular disorga- retention of LacZ or YFP reporter activity in the antrum, tumors, nization, hyperplasia, goblet-like cell metaplasia, and inflamma- and to a lesser extent, in the corpus, indicates Tff1-CreERT2 activity þ þ þ tory cell infiltrates in the antral mucosa of gp130 / ;KrasLSL-G12D/ within the long-lived stem and progenitor cell compartment. This and gp130F/F;Kras-wt mice, which was less prominent in the provides an important advantage over many existing Cre-driver corpus (Fig. 7A and B and Supplementary Fig. S7A). Most likely strains when modeling diseases with a long latency (i.e., cancer) or attributable to the gp130F/F mutation (5, 15), the gastric mucosa of in situations where stem cell loci (i.e., Lgr5) also confer recombi- þ the antrum and corpus of gp130F/F;KrasLSL-G12D/ mice showed a nase activity to other compartments outside of the stomach. striking transformation into an intestinal phenotype associated High concentrations of tamoxifen (>3 mg/20 g mouse) can with extensive hyperplasia, signs of inflammation, and large induce transient mucosal damage and parietal cell atrophy 3 days numbers of goblet-like cells (Fig. 7A and B and Supplementary after the first tamoxifen administration; however, these cytotoxic Fig. S7A). Indeed, these mucus-producing cells had replaced most side effects fully resolve within 1 to 3 weeks through increased of the gastric cell types in the corpus, which showed a severe loss of gastric stem cell proliferation (24).Our data suggest an involve- parietal and chief cells (Fig. 7B and Supplementary Fig. S7B). ment of Tff1-expressing, possibly Lgr5-positive, stem cells in this Collectively, our observations suggest that simultaneous exces- process, because we observe a maximal number of traced antral sive Stat3 activation and oncogenic KrasG12D expression cooper- glands 10 days after tamoxifen administration. Although high ate to increase tumor progression and to trigger the formation of tamoxifen concentrations could also induce transient metaplastic submucosal glands and invasion, with the latter coinciding with changes (23, 24), this is less likely to interfere in situations of extensive metaplastic transformation and inflammatory cell accu- "long latency phenotypes" (i.e., KrasG12D and BrafV600E–induced mulation in the tumor-adjacent precancerous gastric mucosa. tumorigenesis) than with short-term lineage tracing experiments reported by Huh and colleagues (24). Discussion The frequency and extended latency period of tumor develop- þ ment in Tg(Tff1-CreERT2);KrasLSL-G12D/ and Tg(Tff1-CreERT2); þ Here, we describe the generation and validation of Tg(Tff1- BrafLSL-V600E/ mice is reminiscent of similar findings in mice with CreERT2) mice as a novel gastric tissue–specific Cre-driver strain systemic Kras activation (7, 8) and suggest the need to overcome

www.aacrjournals.org Cancer Res; 76(8) April 15, 2016 OF9

Thiem et al.

tumor-suppressing mechanisms, including Ink4a/Arf–mediated Although various animal models have been employed to cell-cycle arrest (37). As inflammation has been shown to over- identify molecular mechanisms involved in gastric tumorigenesis, come oncogene-induced senescence (38) and to enable epithe- there is a need for mouse models that allow for the consecutive lial-to-mesenchymal transition (39), we speculated that inflam- induction of mutations to mimic advanced human disease. Here, þ mation-associated Stat3 activation may also facilitate these pro- we replicate in gp130F/F;Tg(Tff1-CreERT2);KrasLSL-G12D/ mice, as cesses. This hypothesis is supported by our observation of exten- a proof of concept, the sequential genetic changes that collectively sive Stat3 activation within the tumors of KrasG12D mice, but not facilitate progression from tubular adenomas to invasive adeno- in the adjacent preneoplastic epithelium. We functionally validat- carcinomas. To our knowledge, gp130F/F;Tg(Tff1-CreERT2); þ þ ed this correlation in gp130F/F;Tg(Tff1-CreERT2); KrasLSL-G12D/ KrasLSL-G12D/ mice are the first conditional gastric tumor model mutant mice by demonstrating tumor-promoting synergies in which two oncogenic pathways can be activated independently between excessive, inflammation-driven Stat3 and oncogenic and alongside the CEA/SV40-T, the Tff1 knockout and the double- KrasG12D signaling. Our findings complement similar observations transgenic K19-C2mE (GAN) strains, one of a few models that in Kras-driven human xenografts, including pancreas cancer mod- reproducibly show submucosal invasion (41–43). Moreover, as els, where neoplastic growth required persistent Stat3 activity (38). we show here functionally for Stat3 gene expression, the extensive Thus, inflammation may not only promote metaplastic transfor- Tff1-CreERT2 transgene expression throughout the proliferative mation of the (Kras-) mutant glandular epithelium, but indeed tumor epithelium provides a novel Cre-driver strain to assess the may overcome oncogene-induced intrinsic barriers that safeguard contribution of putative oncogenes to gastric tumor progression. against aberrant proliferation. Furthermore, the widespread Cre expression throughout the It still remains unclear whether chronic inflammation and glandular stem cell compartment of Tg(Tff1-CreERT2) mice metaplastic transformation, which are commonly associated enables functional testing of the contribution of candidate genes with human gastric cancer, are required for tumorigenesis or from human genome-wide association studies to gastric cancer whether they account for epiphenomena. Likewise, whether initiation and progression. persistent inflammation and metaplasia develop in parallel or whether one condition causes the other has not been resolved Disclosure of Potential Conflicts of Interest yet. Here, we find Tff2 expression in Alcian blue–positive antral No potential conflicts of interest were disclosed. glands of the KrasG12D-mutant epithelium, at least in tumor- bearing mice, with metaplasia characterized by some intestinal Authors' Contributions hallmarks, including the presence of goblet-like cells and the Conception and design: S. Thiem, M.F. Eissmann, T.L. Putoczki, M. Ernst expression of intestinal marker gpA33 (30, 31). Our observa- Development of methodology: S. Thiem, M.F. Eissmann, J. Elzer, T.L. Putoczki Acquisition of data (provided animals, acquired and managed patients, tions also clarify the contribution of gp130-dependent exces- provided facilities, etc.): S. Thiem, M.F. Eissmann, J. Elzer, A. Jonas, T.L. sive Stat3 activity to Kras-induced tumorigenesis. First, in the Putoczki, A. Poh, A. Preaudet, D. Flanagan, P. Waring, M. Buchert G12D Kras -mutant epithelium, elevated Stat3 activity correlates Analysis and interpretation of data (e.g., statistical analysis, biostatistics, with enhanced metaplastic transformation and associated loss computational analysis): S. Thiem, M.F. Eissmann, T.L. Putoczki, M. Buchert, of parietal and chief cells and promotes tumor development. As A. Jarnicki, M. Ernst the gastric epithelium of tumor-bearing and tumor-free Writing, review, and/or revision of the manuscript: S. Thiem, M.F. Eissmann, M. Buchert, M. Ernst KrasG12D or BrafV600E mice showed goblet-like cell metaplasia, Administrative, technical, or material support (i.e., reporting or organizing we conclude that the tumor-promoting secondary events, data, constructing databases): S. Thiem, A. Jonas, P. Nguyen, D. Flanagan, which are implicated by the long latency period required for E. Vincan tumors to emerge, are likely to occur in metaplastically trans- Study supervision: T.L. Putoczki, M. Ernst formed cells. Second, excessive Stat3 activation in KrasG12D or Other (obtained funding): M. Ernst BrafV600E cells confers the potential for hyperplastic/dysplastic transformation and, when further stimulated through excessive Acknowledgments gp130 signaling, the appearance of submucosal glands. It is there- The authors thank the LICR and WEHI Histology and Animal Facility Staff for – expert technical assistance. The authors also thank Sue Bath for the pronuclear fore tempting to speculate that interference with gp130/Stat3 microinjections and Joan Heath for helpful discussions. activating cytokines might provide a novel therapeutic target for G12D Kras -dependent gastric tumorigenesis. Indeed, we have ob- Grant Support served that systemic limitation of Stat3 expression in tamoxifen- This work was supported in part by a Project Grant (APP1007523) and a LSL-V600E/þ þ/ induced compound Tg(Tff1-CreERT2);Braf ;Stat3 Program Grant (APP487922) from the NHMRC and by funds from the Oper- markedly reduced tumor formation (Eissmann and Ernst, ational Infrastructure Support Program provided by the Victorian Government, unpublished data). Furthermore, the notion of inflammatory Australia (M. Ernst). This work was also initiated through the generous financial cytokines promoting Kras-initiated gastric tumors (39) is con- support of Ludwig Cancer Research. The costs of publication of this article were defrayed in part by the payment of sistent with observations by Okumura and colleagues (6) and fi fi fl page charges. This article must therefore be hereby marked advertisement in ts with emerging ndings that in ammation downregulates accordance with 18 U.S.C. Section 1734 solely to indicate this fact. mismatch repair proteins and the clinical observations that KRAS mutations occur more frequently in gastric cancers with Received November 10, 2015; revised December 20, 2015; accepted January DNA mismatch repair deficiency (40). 11, 2016; published OnlineFirst February 2, 2016.

References 1. Hartgrink HH, Jansen EP, van Grieken NC, van de Velde CJ. Gastric cancer. 2. Fox JG, Wang TC. Inﬂammation, atrophy, and gastric cancer. J Clin Invest Lancet 2009;374:477–490. 2007;117:60–69.

OF10 Cancer Res; 76(8) April 15, 2016 Cancer Research

Kras and Stat3 Drive Gastric Tumorigenesis

3. Deng N, Goh LK, Wang H, Das K, Tao J, Tan IB, et al. A comprehensive 25. Hertel SC, Chwieralski CE, Hinz M, Rio MC, Tomasetto C, Hoffmann W. survey of genomic alterations in gastric cancer reveals systematic patterns of Profiling trefoil factor family (TFF) expression in the mouse: identification molecular exclusivity and co-occurrence among distinct therapeutic targets. of an antisense TFF1-related transcript in the kidney and liver. Peptides Gut 2012;61:673–84. 2004;25:755–62. 4. Goldenring JR, Nomura S. Differentiation of the gastric mucosa III. Animal 26. dos Santos Silva E, Ulrich M, Doring€ G, Botzenhart K, Gott€ P. Trefoil factor models of oxyntic atrophy and metaplasia. Am J Physiol Gastrointest Liver family domain peptides in the human respiratory tract. J Pathol Physiol 2006;291:G999–1004. 2000;190:133–42. 5. Thiem S, Pierce TP, Palmieri M, Putoczki TL, Buchert M, Preaudet A, et al. 27. Putoczki TL, Thiem S, Loving A, Busuttil RA, Wilson NJ, Ziegler PK, et al. mTORC1 inhibition restricts inflammation-associated gastrointestinal Interleukin-11 Is the Dominant IL-6 Family Cytokine during Gastrointes- tumorigenesis in mice. J Clin Invest 2013;123:767–81. tinal Tumorigenesis and Can Be Targeted Therapeutically. Cancer Cell 6. Okumura T, Ericksen RE, Takaishi S, Wang SS, Dubeykovskiy Z, Shibata W, 2013;24:257–71. et al. K-ras mutation targeted to gastric tissue progenitor cells results in 28. Bollrath J, Phesse T, Von Burstin VA, Putoczki T, Bennecke M, Bateman T, chronic inflammation, an altered microenvironment, and progression to et al. gp130-mediated Stat3 activation in enterocytes regulates cell survival intraepithelial neoplasia. Cancer Res 2010;70:8435–45. and cell-cycle progression during colitis-associated tumorigenesis. Cancer 7. Ray KC, Bell KM, Yan J, Gu G, Chung CH, Washington MK, et al. Epithelial Cell 2009;15:91–102. tissues have varying degrees of susceptibility to Kras(G12D)-initiated 29. Giraud AS. Metaplasia as a premalignant pathology in the stomach. tumorigenesis in a mouse model. PLoS One 2011;6:e16786. Gastroenterology 2007;132:2053–6. 8. Matkar SS, Durham A, Brice A, Wang TC, Rustgi AK, Hua X. Systemic 30. Johnstone CN, Tebbutt NC, Abud HE, White SJ, Stenvers KL, Hall NE, et al. activation of K-ras rapidly induces gastric hyperplasia and metaplasia in Characterization of mouse A33 antigen, a definitive marker for basolateral mice. Am J Cancer Res 2011;1:432–445. surfaces of intestinal epithelial cells. Am J Physiol Gastrointest Liver Physiol 9. Barker N, Huch M, Kujala P, van de Wetering M, Snippert HJ, van Es JH, et al. 2000;279:G500–510. Lgr5(þve) stem cells drive self-renewal in the stomach and build long-lived 31. Johnstone CN, White SJ, Tebbutt NC, Clay FJ, Ernst M, Biggs WH, et al. gastric units in vitro. Cell Stem Cell 2010;6:25–36. Analysis of the regulation of the A33 antigen gene reveals intestine-specific 10. Syder AJ, Karam SM, Mills JC, Ippolito JE, Ansari HR, Farook V, et al. A mechanisms of gene expression. J Biol Chem 2002;277:34531–9. transgenic mouse model of metastatic carcinoma involving transdifferen- 32. Mutoh H, Hakamata Y, Sato K, Eda A, Yanaka I, Honda S, et al. Conversion tiation of a gastric epithelial lineage progenitor to a neuroendocrine of gastric mucosa to intestinal metaplasia in Cdx2-expressing transgenic phenotype. Proc Natl Acad Sci U S A 2004;101:4471–6. mice. Biochem Biophys Res Commun 2002;294:470–9. 11. Quante M, Marrache F, Goldenring JR, Wang TC. TFF2 mRNA transcript 33. Ilhan_ O,€ Han U,€ Onal€ B, Celik SY, et al. Prognostic significance of MUC1, expression marks a gland progenitor cell of the gastric oxyntic mucosa. MUC2 and MUC5AC expressions in gastric carcinoma. Turk J Gastroen- Gastroenterology 2010;139:2018–27. terol 2010;21:345–52. 12. Taupin D, Podolsky DK. Trefoil factors: initiators of mucosal healing. Nat 34. van Grieken NC, Aoyama T, Chambers PA, Bottomley D, Ward LC, Inam Rev Mol Cell Biol 2003;4:721–32. I, et al. KRAS and BRAF mutations are rare and related to DNA mismatch 13. Karam SM, Tomasetto C, Rio MC. Trefoil factor 1 is required for the repair deficiency in gastric cancer from the East and the West: results commitment programme of mouse oxyntic epithelial progenitors. Gut from a large international multicentre study. Br J Cancer 2013;108: 2004;53:1408–15. 1495–501. 14. Soutto M, Peng D, Katsha A, Chen Z, Piazuelo MB, Washington MK, et al. 35. Ernst M, Najdovska M, Grail D, Lundgren-May T, Buchert M, Tye H, et al. Activation of b-catenin signalling by TFF1 loss promotes cell proliferation STAT3 and STAT1 mediate IL-11-dependent and inflammation-associated and gastric tumorigenesis. Gut 2015;64:1028–39. gastric tumorigenesis in gp130 receptor mutant mice. J Clin Invest 15. Tebbutt NC, Giraud AS, Inglese M, Jenkins B, Waring P, Clay FJ, et al. 2008;118:1727–38. Reciprocal regulation of gastrointestinal homeostasis by SHP2 and STAT- 36. Judd LM, Alderman BM, Howlett M, Shulkes A, Dow C, Moverley J, Grail mediated trefoil gene activation in gp130 mutant mice. Nat Med D, et al. Gastric cancer development in mice lacking the SHP2 binding 2002;8:1089–97. site on the IL-6 family co-receptor gp130. Gastroenterology 2004;126: 16. Alonzi T, Middleton G, Wyatt S, Rizzuto G, Libert C, Poli V, et al. Essential 196–207. role of STAT3 in the control of the acute-phase response as revealed by 37. Bennecke M, Kriegl L, Bajbouj M, Retzlaff K, Robine S, Jung A, Arkan MC, inducible gene inactivation in the liver. Mol Cell Biol 2001;21:1621–32. Kirchner T, Greten FR. Ink4a/Arf and oncogene-induced senescence pre- 17. Jackson EL, Willis N, Mercer K, Bronson RT, Crowley D, Montoya R, et al. vent tumor progression during alternative colorectal tumorigenesis. Cancer Analysis of lung tumor initiation and progression using conditional Cell 2010;18:135–46. expression of oncogenic K-ras. Genes Dev 2001;15:3243–8. 38. Guerra C, Collado M, Navas C, Schuhmacher AJ, Hernandez-Porras I, 18. Thorey IS, Muth K, Russ AP, Otte J, Reffelmann A, von Melchner H. Selective Canamero M, et al. Pancreatitis-induced inflammation contributes to disruption of genes transiently induced in differentiating mouse embry- pancreatic cancer by inhibiting oncogene-induced senescence. Cancer Cell onic stem cells by using gene trap mutagenesis and site-specific recombi- 2011;19:728–39. nation. Mol Cell Biol 1998;18:3081–8. 39. Gough DJ, Corlett A, Schlessinger K, Wegrzyn J, Larner AC, Levy DE. 19. Dankort D, Filenova E, Collado M, Serrano M, Jones K, McMahon M. A new Mitochondrial STAT3 supports Ras-dependent oncogenic transformation. mouse model to explore the initiation, progression, and therapy of Science 2009;324:1713–36. BRAFV600E-induced lung tumors. Genes Dev 2007;21:379–84. 40. Colotta F, Allavena P, Sica A, Garlanda C, Mantovani A. Cancer-related 20. Srinivas S, Watanabe T, Lin CS, William CM, Tanabe Y, Jessell TM, et al. Cre inflammation, the seventh hallmark of cancer: links to genetic instability. reporter strains produced by targeted insertion of EYFP and ECFP into the Carcinogenesis 2009;30:1073–81. ROSA26 locus. BMC Dev Biol 2001;1:4–11. 41. Thompson J, Epting T, Schwarzkopf G, Singhofen A, Eades-Perner AM, van 21. Rogers AB, Taylor NS, Whary MT, Stefanich ED, Wang TC, Fox JG, et al. der Putten, et al. A transgenic mouse line that develops early-onset invasive Helicobacter pylori but not high salt induces gastric intraepithelial neo- gastric carcinoma provides a model for carcinoembryonic antigen-targeted plasia in B6129 mice. Cancer Res 2005;65:10709–15. tumor therapy. Int J Cancer 2000;86:863–9. 22. Rogers AB, Houghton J. Helicobacter-based mouse models of digestive 42. Soutto M, Belkhiri A, Piazuelo MB, Schneider BG, Peng D, Jiang A, et al. Loss system carcinogenesis. Methods Mol Biol 2009;511:267–95. of TFF1 is associated with activation of NF-kB-mediated inflammation 23. Stange DE, Koo B-K, Huch M, Sibbel G, Basak O, Lyubimova A, et al. and gastric neoplasia in mice and humans. J Clin Invest 2011;121: Differentiated Troy þ Chief cells act as reserve stem cells to generate all 1753–67. lineages of the stomach epithelium. Cell 2013;155:357–368. 43. Oshima H, Matsunaga A, Fujimura T, Tsukamoto T, Taketo MM, Oshima 24. Huh WJ, Khurana SS, Geahlen JH, Kohli K, Waller RA, Mills JC. Tamoxifen M. Carcinogenesis in mouse stomach by simultaneous activation of the induces rapid, reversible atrophy, and metaplasia in mouse stomach. Wnt signaling and prostaglandin E2 pathway. Gastroenterology 2006;131: Gastroenterology 2012;142:21–4. 1086–95.

www.aacrjournals.org Cancer Res; 76(8) April 15, 2016 OF11

Stomach-Specific Activation of Oncogenic KRAS and STAT3-Dependent Inflammation Cooperatively Promote Gastric Tumorigenesis in a Preclinical Model

Stefan Thiem, Moritz F. Eissmann, Joachim Elzer, et al.

Cancer Res Published OnlineFirst February 2, 2016.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-15-3089

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2016/02/04/0008-5472.CAN-15-3089.DC1

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/early/2016/04/05/0008-5472.CAN-15-3089. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.