expression and localisation of insulin receptor substrate 2 in normal intestine and colorectal...

Queries for AuthorJournal: GutPaper: gt158386Title: Expression and localisation of insulin receptor substrate 2 in normal intestine and colorectal tumours. Regulationby intestine-specific transcription factor CDX2

The proof of your manuscript appears on the following page(s).

Please note that this is a galley proof and the layout of the article may change before publication. Please read the manuscript

carefully, checking for accuracy, verifying the reference order and double-checking figures and tables. When reviewing your page

proof please keep in mind that a professional copyeditor edited your manuscript to comply with the style requirements of the journal.

This is not an opportunity to alter, amend or revise your paper; it is intended to be for correction purposes only.

During the preparation of your manuscript for publication, the questions listed below have arisen (the query number can also be

found in the gutter close to the text it refers to). Please attend to these matters and return the answers to these questions when you

return your corrections.

Please note, we will not be able to proceed with your article and publish it in print if these querieshave not been addressed.

Query

Reference

Query

1 midge station has been change to midge stage her. Is that OK?

2 Is AdhCDX2 correct in part C of figure 5

If you are happy with the proof as it stands, please email to confirm this. Changes that do not require a copy of the proof can be sent

by email (please be as specific as possible).

Email: [email protected]

If you have any changes that cannot be described easily in an email, please mark them clearly on the proof and email a scan of the

changes by replying to the eProof email or by fax: +44 (0)844 443 7468.

PLEASE RESPOND WITHIN 48 HOURS

gt158386 Module 1 Gut 19/6/09 11:35:02 Topics:

Author query sheet

0 Gut 2009;000:0–10. doi:10.1136/gut.2008.158386

Expression and localisation of insulin receptorsubstrate 2 in normal intestine and colorectaltumours. Regulation by intestine-specifictranscription factor CDX2

S Modica,1 A Morgano,2 L Salvatore,1 M Petruzzelli,1 M-T Vanier,3 R Valanzano,4

D L Esposito,2 G Palasciano,1 I Duluc,3 J-N Freund,3 R Mariani-Costantini,2

A Moschetta1

c Additional methods, figuresand a table are published onlineonly at http://gut.bmj.com/content/volXX/issueX

1 Laboratory of Lipid Metabolismand Cancer, Consorzio MarioNegri Sud, Santa Maria Imbaro,Chieti and Clinica Medica Murri,University of Bari, Italy; 2 Unit ofMolecular Pathology andGenomics, Center for Scienceson the Ageing (CeSI), ‘‘Gd’Annunzio’’ UniversityFoundation, Chieti, Italy;3 INSERM U682 and UniversiteLouis Pasteur, Faculte deMedecine, Strasbourg, France;4 Department of ClinicalPhysiopathology, University ofFlorence, Florence, Italy

Correspondence to:Dr A Moschetta, University ofBari, Consorzio Mario Negri Sud,Via Nazionale 8/A, 66030 SantaMaria Imbaro (CH), Italy;[email protected]

S Modica and A Morganocontributed equally to this study.

Revised 28 January 2009Accepted 4 February 2009Published Online First15 February 2009

ABSTRACTBackground and aims: Self-renewal and differentiationof intestinal epithelium is a tightly regulated process,whose perturbations are implicated in human colorectaltumourigenesis. The insulin/insulin-like growth factor (IGF)signalling pathway may play an important role in intestinalepithelium homeostasis. Insulin receptor substrate 2(IRS2) is a poorly characterised component in thispathway.Methods: Using complementary in vitro and in vivohuman and murine models, expression (mRNA andprotein levels), localisation (immunohistochemistry) andregulation of IRS2 were investigated in the normalintestine and colorectal tumours. In silico analysis of thehuman IRS2 promoter was performed together withreporter and chromatin immunoprecipitation assays.Results: Significant IRS2 expression was detected in theintestine, with specific protein localisation in the villusregion of the ileum and in the surface epithelium of thecolon. In human HT29 and Caco2 cells, IRS2 mRNA levelsincreased with spontaneous and induced differentiation,together with CDX2 (caudal-related homeobox protein 2),P21 and KLF4 (Kruppel-like factor 4). Adenoviral infectionwith human CDX2 induced IRS2 expression in APC-(adenomatous polyposis coli) and b-catenin-mutatedcells. On the other hand, IRS2 downregulation wasobserved in differentiated enterocytes after adenoviralinfection with short hairpin CDX2 (shCDX2), in theintestine of CDX2 heterozygous mice and in colorectaltumours of ApcMin/+ and patients with familial adenoma-tous polyposis (FAP). The human IRS2 promoter regionpresents several CDX2-binding sites where CDX2immunoprecipitated in vivo. IRS2 reporters were func-tionally activated via CDX2 and blocked via a dominant-negative CDX2 protein.Conclusions: Combining gain- and loss-of-functionapproaches, an intriguing scenario is presented wherebyIRS2 is significantly expressed in the apical intestinalcompartment and is directly controlled by CDX2 in normalintestine and tumours.

The intestinal epithelium consists of a single layerof epithelial cells mainly organised into submucosalinvaginations termed crypts of Lieberkuhn. At thecrypt bottom reside stem cells that self-renewthroughout life to generate progenitors thatoccupy the lower third of the crypt.1 2 In themidcrypt region, cells stop proliferating to differ-entiate into enterocytes. After reaching the epithelial

surface, differentiated cells undergo apoptosis andare shed into the lumen. A tight regulation of theintestinal epithelium homeostasis is required toavoid neoplastic transformation. A body of evidenceimplies that Wnt signalling is a dominant force incontrolling the process of intestinal mucosarenewal.2 One of the key Wnt signalling componentsis the scaffold protein encoded by the adenomatouspolyposis coli (APC) gene. APC is part of a multi-protein destruction complex.3 In the absence of Wntactivators, free cytoplasmatic b-catenin is seques-tered by APC into the destruction complex, phos-phorylated and targeted for ubiquitin–proteosomedegradation. In contrast, in the presence of Wntcascade activators, or when APC is mutated, b-catenin accumulates in the cytoplasm, shuttles tothe nucleus and, in association with the transcrip-tion factor T cell factor 4 (TCF-4), induces theexpression of genes required for cell proliferation.4 5

APC was identified as a germline mutated gene infamilial adenomatous polyposis (FAP),6 an autoso-mal dominantly inherited disease characterised bythe development of colorectal adenomas within thefourth decade of life, with a very high risk ofcolorectal cancer (CRC).7 Later, somatic APC inacti-vation was found to occur in the majority ofsporadic CRCs.8

Caudal-related homeobox protein 2 (CDX2) isconsidered to be a downstream mediator of APCtumour suppressor activity in intestinal cells.9 10

The human CDX2 gene is one of the threemammalian homologues of the Drosophila homeo-box-containing gene Caudal, whose encoded tran-scription factor has a primary role duringembryonic development, as assessed by the earlylethality of Cdx2-deficient mouse embryos.11 Fromthe midge stage ;onwards, the CDX2 transcriptionfactor becomes selectively expressed in the crypt–villus epithelium of the small and large intestines.12

During embryogenesis, the CDX2 gene determinesintestinal identity in the presumptive gut endo-derm13 and it later controls the transcription ofgenes linked to the differentiation of the intestinalepithelium, such as Kruppel-like factor 4 (KLF4),10

sucrase-isomaltase,14 and P21.15 The expression ofCDX2 becomes altered in human CRC in relationto the tumour grade.16–18 The relevance of thesealterations has been uncovered in mice sincereduction of Cdx2 expression facilitates tumour


Paper

Gut 2009;000:0–10. doi:10.1136/gut.2008.158386 1

progression in models of genetically induced CRC.19 20

Restoration of CDX2 in colon cancer cells has also been shownto reduce cell migration in vitro and dissemination in vivo.21 Onthe other hand, the function of CDX2 might not bestraightforward as an intestinal tumour suppressor since recentreports suggest a context- and cell type-dependent role.22–24

Insulin receptor substrate (IRS) proteins mediate the pleio-tropic effects of insulin/insulin-like growth factor (IGF) signal-ling on mitogenic and metabolic processes. IRS proteins act asdocking proteins between insulin, IGF-1 and cytokine receptors,and a complex network of intracellular molecules containing Srchomology 2 (SH2) domains that are capable of binding IRSproteins phosphorylated at tyrosine residues.25 This results inthe activation of several downstream pathways that arevariably responsible for cell survival, proliferation, differentia-tion and metabolic responses. In particular, IRS2-activatedpathways may have distinct roles in different organs or tissues,such as in female reproductive function,26 brain growth andmetabolic homeostasis.27 Among these functions, IRS2 seems tohave a primary involvement in metabolism. Indeed, IRS2-knockout mice have defects in hepatic insulin signalling andshow altered growth in only a few tissues such as certainneurons28 and pancreatic b-cells.29 It is presently unknown ifIRS2 is expressed in the intestine and how it is regulated. Incontrast, it is known that IRS1 is expressed mostly in theintestinal crypts and that its disruption inhibits growth ofintestinal epithelial cells,30 31 and confers protection againsttumour formation in the ApcMin/+ mouse background.30

Interestingly, genetic IRS1 and IRS2 polymorphisms have beenindependently associated with CRC risk.32 Recently, a somaticmutation in the IRS2 gene has been identified in human CRCspecimens.33

In the present study, we first show that IRS2 is significantlyexpressed in the villus–epithelial compartment of the adultintestine and that IRS2 is downregulated in APC-mutatedintestinal tumours. Then, combining a variety of gain- and loss-of-function approaches, we present a novel and intriguingscenario whereby intestinal IRS2 expression is directly con-trolled by CDX2 in the normal intestine and colorectal tumours.

PATIENTS AND METHODS

Patients with FAPUnrelated patients with FAP were recruited for the study afterapproval by the Ethics Committee of the University ‘‘GD’Annunzio’’ of Chieti. Written informed consent was obtainedfrom each patient before mutation analysis and tissue harvest-ing. All the included cases presented with a classical FAPphenotype and harboured pathogenetic germline APC muta-tions.34 More information on patients and gene sequencing areavailable in the Supplementary methods available online.

Mice and cell cultureHeterozygous Cdx2+/2 mice11 and wild-type littermates werehoused under pathogen-free conditions according to the guide-lines of the Ethics Committee of the University Louis Pasteur ofStrasbourg.20 ApcMin/+ mice were obtained from the JacksonLaboratory (Bar Harbor, Maine, USA). All mice were fed astandard rodent chow diet in a temperature-controlled room(23uC) on a 12 h light/dark cycle. All the experiments presentedin this study have been carried out in male mice. The EthicsCommittee of the Consorzio Mario Negri Sud approved thisexperimental set-up.

HT29 (HTB-38), CaCo2, HeLa and SW480 cells wereobtained from ATCC (ATCC-LGC Promochem, London, UK).APC- or LacZ-inducible HT29 cells were a generous gift of DrsKW Kinzler and B Vogelstein,4 and TR4 or dominant-negativeTCF-4- (DNTCF-4) inducible Ls174T cells were a generous giftof Dr H Clevers.35 Methodological details on cell culture areavailable in the Supplementary methods online.

RNA extraction and quantitative real-time PCR (qRT-PCR)Following tissue and cell RNA isolation and purity check, cDNAwas synthesised by retro-transcribing 4 mg of total RNA in atotal volume of 100 ml using the High Capacity DNA ArchiveKit (Applied Biosystems, Foster City, California, USA). qRT-PCR primers were designed using Primer Express software. qRT-PCR assays were performed as previously described using theDDCT method for relative quantification.36 Methodologicaldetails and sequences of validated primers are reported in theSupplementary methods online.

Adenoviral constructsHuman CDX2 adenovirus (AdCDX2) was generated using theViral Power Adenoviral Expression System (Invitrogen, La Jolla,California, USA). AdLacZ was used as control. Short hairpinCDX2 adenovirus (shCDX2) was generated based on apreviously validated sequence against human CDX237 usingthe BLOCK-iT U6 RNA entry vector kit (Invitrogen). shLacZwas used as a control. Details on the cloning of plasmids, and onpurification and propagation of the viruses are reported in theSupplementary methods online. The adenoviral titre wasdetermined by qRT-PCR with specific primers.38

Luciferase reporter constructsRegions between 22394/21740, 22269/21740, 22089/21740,22039/21740 and 21860/21740 of the pGL3-IRS2 (22399/+217) plasmid (a kind gift of Dr M Kasuga, Kobe University,Japan) containing the 59-flanking region of the human IRS2promoter39 were PCR amplified and subcloned into the polylinkerof the promoter-less pGL3-basic vector (Promega, Madison,Wisconsin, USA) using MluI 59 and HindIII 39 restriction sites togenerate pGL3-IRS2 (22394/21740), pGL3-IRS2 (22269/21740),pGL3-IRS2 (22089/21740), pGL3-IRS2 (22039/21740) andpGL3-IRS2 (21860/21740) reporter constructs.

Statistical analysisAll results are expressed as the mean (SEM). Data distribution andgene expression statistical analysis was performed using NCSSstatisticaland power analysis software 2007 (Kaysville,Utah, USA).Multiple groups were tested by one-way analysis of variance(ANOVA) followed by the Fisher least significant difference test forunpaired data. Comparisons of two groups were performed usingthe Student t test followed by a Mann–Whitney U test whereappropriate. A p value ,0.05 was considered significant.

Methods used for western blot, immunohistochemistry andimmunocytochemistry, alkaline phosphatase measurements,transient transfection and luciferase activity, chromatin immu-noprecipitation (ChIP) and gel-shift electromobility shift assays(EMSAs) are available as Supplementary methods online.

RESULTSIRS2 expression and localisation in the intestineThe first step of our study consisted of the assessment of IRS2gene expression in the murine gastrointestinal tract and theexact localisation of IRS2 protein in the crypt–villus axis of the


Paper

2 Gut 2009;000:0–10. doi:10.1136/gut.2008.158386

intestinal mucosa. Significant IRS2 mRNA levels were found inthe entire enterohepatic system (liver, stomach, duodenum,jejunum, ileum and colon), as compared with white and brownadipose tissue that showed the highest IRS2 expression levels(fig 1A). Immunohistochemical analysis of the terminal ileumand proximal colon revealed clear IRS2 cytoplasmic localisationwithin the differentiated absorptive enterocytes of the villuscompartment (ileum) and surface epithelium (colon), togetherwith L-FABP (L-type fatty acid-binding protein) (fig 1B). NoIRS2 labelling was observed in the Ki67-stained proliferatingprogenitor cells of the crypt compartment (fig 1B).

Changes in IRS2 expression during intestinal cell differentiationSince IRS2 is solely expressed in the differentiated apicalcompartment of the intestinal epithelium, we analysed the

behaviour of IRS2 expression in a human CRC cell line duringspontaneous and induced in vitro differentiation. As a first step,IRS2 mRNA and protein expression was evaluated in HT29 cellsin preconfluent confluent and postconfluent states.40 Underthese conditions, a clear induction of IRS2 protein expressionwas observed together with the upregulation of CDX2 protein(fig 2A).40 Alkaline phosphatase (ALP) activity was used as asurrogate marker of enterocyte differentiation (fig 2B,C). A netincrease in IRS2 mRNA levels was observed during confluence-induced growth arrest and enterocyte differentiation (fig 2B),which was accompanied by increased expression of genes thatare known to be associated with the degree of intestinaldifferentiation, such as CDX2, KLF4 and P21. Since in vitropostconfluence growth also increases cellular metabolic require-ments, in order to know if the induction of CDX2 and IRS2 is


Figure 1 Insulin receptor substrate 2(IRS2) expression levels and localisationin mice tissues. (A) IRS2 mRNAexpression was measured by quantitativereal-time PCR using cyclophilin as areference gene. Results are expressed asthe mean (SEM). The mean Ct of ileumsamples is reported in number. WAT,white adipose tissue; BAT, brownadipose tissue. (B) Paraffin-embeddedterminal ileum (upper panel) and proximalcolon (lower panel) specimens wereimmunostained with IRS2 antibody todetermine the expression and localisationof IRS2 protein. Immunostaining of Ki67and L-type fatty acid-binding protein (L-FABP) was used as a marker of theproliferative crypt and differentiated villuscompartments, respectively (6400magnification).

Paper

Gut 2009;000:0–10. doi:10.1136/gut.2008.158386 3

related to differentiation or increased metabolic requirements,we studied SW480 colon cancer cells that are not able todifferentiate postconfluence.41 Interestingly, there was noincrease in ALP activity in postconfluent SW480 cells with noinduction of CDX2, IRS2 and P21 mRNA levels (Supplementaryfig S1). Thus, in the postconfluent state, SW480 cells increasetheir metabolic requirements without overexpressing CDX2 andIRS2.

Treatment with sodium butyrate (NaB) at 1 mM concentra-tion promotes differentiation of intestinal cell lines. Also, CDX2expression is stimulated in HT29 cells by NaB treatment at adose that promotes cell differentiation.42 Here we show thatIRS2 mRNA (fig 2C) and protein (fig 2A) levels significantlyincreased in HT29 cells after 24 h treatment with 1 mM NaB.Upregulation of IRS2 was accompanied by increased CDX2,KLF4 and P21 mRNA levels and ALP activity (fig 2C). As afurther confirmation, an increase in CDX2, IRS2, KLF4 and P21mRNA levels was also observed in Caco2 cells duringdifferentiation (fig 2D).

The CDX2–IRS2 axis regulates IGF responses during intestinalcell differentiationSince IRS2 is presumably part of the IGF signalling pathways inthe intestine, we decided to study IGF-1-induced activation ofextracellular signal-regulated kinase (Erk) in preconfluent (lowCDX2–IRS2 expressing) and postconfluent (high CDX2–IRS2expressing) HT29 colon cancer cells. We therefore treatedpreconfluent and postconfluent HT29 cells with 100 nmol/lIGF-1 for 0–30 min and we measured total and phosphorylatedforms of Erk1 and Erk2.43 Representative immunoblots arereported in fig 3A. Also, we performed a fluorescence-activatedcell sorting (FACS) analysis of the same cells with and withouttreatment with 100 nmol/l IGF-1. Interestingly, when the levelsof endogenous IRS2–CDX2 were low (ie, preconfluent HT29cells), IGF-1 was able to phosphorylate both Erk1 and Erk2 to acertain extent and to induce a strong increase in cell prolifera-tion, as shown by the significant amount of cells in S-phase(from 13.49% to 57.52%, vehicle vs IGF-1, p ,0.01, fig 3B). Onthe other hand, in postconfluent HT29 cells that express high


Figure 2 Insulin receptor substrate 2 (IRS2) expression levels during confluence-induced growth arrest and sodium butyrate (NaB)-induceddifferentiation in HT29 cells and after confluence in Caco2 cells. (A) Expression levels of IRS2 and CDX2 (caudal-related homeobox protein 2) in HT29cells (preconfluence, confluence and postconfluence; vehicle and NaB treated) were observed by western blot. Alkaline phosphatase (ALP) activity wasanalysed as a marker of differentiation in HT29 cells (preconfluence, confluence and postconfluence (B); vehicle and NaB treated (C)). Quantitative real-time PCR (qRT-PCR) was performed to measure IRS2, CDX2, KLF4 (Kruppel-like factor) and P21 mRNA levels in HT29 cells (preconfluence, confluenceand postconfluence (B); vehicle and NaB treated (C)). (D) qRT-PCR was performed to measure IRS2, CDX2, KLF4 and P21 mRNA levels in Caco2 cells atdifferent time points after confluence. Values shown represent the mean (SEM) for four independent determinations performed in duplicate. Cyclophilinwas used as a reference gene and values were normalised to data obtained from preconfluent or vehicle-treated cells. Different lower case lettersindicate significant statistical difference (b different from a, p .0.05; c different from a and b, p ,0.05, analysis of variance).

Paper

4 Gut 2009;000:0–10. doi:10.1136/gut.2008.158386

levels of CDX2 and IRS2, there is a basal increase in total andphosphorylated Erk1 and 2 as compared with preconfluentHT29 cells. Furthermore, under these conditions, IGF-1 wasunable to induce any changes in total and phosphorylated Erk orin cell proliferation, as shown by no changes in the amounts ofcells in phases G0G1, G2M and S (fig 3B). Taking these datatogether, we conclude that when CDX2 induces IRS2 expres-sion in postconfluent HT29 colon cancer cells, IGF-1 fails tostimulate phosphorylation of Erk1 and 2 as well as to induce cellproliferation. These findings are consistent with the idea that inthe intestine IRS2 is expressed in differentiated low proliferatingenterocytes, where it may counteract IGF-1-induced proliferation.

Downregulation of IRS2 expression in colorectal tumours frompatients with FAP and ApcMin/+ miceEnterocyte differentiation from proliferative crypt precursors isa homeostatic process where Wnt signalling and APC statusrepresent the dominant factors in controlling cell fate. FAP is aclassic model of CRC, where the earliest phase is exclusivelyinitiated by and dependent on APC gene mutations. FAPadenomas grow within the top crypt–villus compartmentretaining a ‘‘crypt progenitor-like’’ phenotype generated by

the loss of APC and the subsequent accumulation of the b-catenin protein at high levels.2 We therefore studied IRS2 andCDX2 expression levels in colorectal adenomas obtained byendoscopic resection from patients with FAP with molecularlydefined germline and somatic APC mutations.6 IRS2 and CDX2mRNA levels were reduced in adenomas as compared withmatched normal mucosa (fig 4A). Also, IRS2 protein wasdecreased in the aberrant crypt foci and colon tumours ofApcMin/+ mice as revealed by immunohystochemical analysis,where Ki67 was used as control for the proliferative compart-ments (fig 4B).

CDX2 is expressed in a heterogeneous pattern in some CRCsin relation to their degree of differentiation,18 22 and sometimesamplified.23 Thus, we analysed the expression of IRS2 in humancolon cancer samples that presented CDX2 expression hetero-geneously. Intriguingly, there was also a high correspondencebetween CDX2 and IRS2 levels in human colon tumours withcertain CDX2 expression (fig 4C), thus supporting our hypoth-esis of IRS2 being a target of CDX2.

Downregulation of IRS2 mRNA levels in CDX2+/2 miceTo confirm further that IRS2 is a direct target of CDX2, weanalysed the IRS2 and CDX2 mRNA levels in the intestine ofCDX2+/2 mice. While homozygous CDX22/2 mice are embryo-nic lethal, CDX2 heterozygous mice have a compromised CDX2function, develop gastric heteroplasia in the intestine, and arehypersensitive to CRC upon treatment with a DNA mutagen20

and under genetic predisposition in the ApcMin/+ background.19

Here we show that IRS2 expression levels are lower in theterminal ileum and proximal colon of CDX2+/2 mice ascompared with corresponding wild-type mice (fig 4D).Interestingly, IRS2 levels were positively correlated withCDX2 levels in all the examined samples.

CDX2 and IRS2 mRNA levels are upregulated in HT29 cells withAPC-inducible expression and unchanged in Ls174T cells withDNTCF4-inducible expressionSince IRS2 expression is significantly decreased in mouse andhuman tumours carrying an APC mutation, we decided toinvestigate if re-expression of a wild-type APC in an APC-mutated cell line would restore IRS2 expression. Like most CRCcell lines,44 HT29 cells have no functional APC protein, whiletwo C-terminally truncated isoforms of approximately 100 and200 kDa, expressed by mutated APC alleles, are present.Inducible expression of full-length APC in HT29 cells resultsin a significant decrease of cell proliferation via induction of celldeath through apoptotic pathways.45 In line with previousreports,9 10 when APC expression is re-established in HT29-APCcells via ZnCl2, CDX2 (fig 5A) and KLF4 and P21 (data notshown), mRNA levels are significantly increased. Under theseconditions, there is a net upregulation of IRS2 expression levels(fig 5A), accompanied by an increase in P21 mRNA levels (datanot shown). HT29-b-Gal cells treated with ZnCl2 exhibited nosignificant difference in CDX2 and IRS2 mRNA levels (fig 5A). Anet increase in IRS2 protein levels in HT29APC cells after ZnCl2was observed via immunocytochemical analysis (fig 5B). Incontrast, in Ls174T cells that present a mutated b-catenin withtranscriptionally activated b-catenin–TCF-4 complexes, theinduction of DNTCF4 decreased b-catenin–TCF-4 target genesand increased P21 mRNA levels (data not shown),35 but did notalter the mRNA levels of CDX2 and IRS2 (fig 5B). These dataunderscore the necessity for a functional cytoplasmic APCprogramme for the expression of IRS2.


Figure 3 Activation of extracellular signal-regulated kinase (Erk) byinsulin-like growth factor-1 (IGF-1) in HT29 cells after confluence. (A)Stimulation of Erk in preconfluent and postconfluent HT29 cells wasstudied in the absence or presence of 100 nmol/l IGF-1 for the indicatedtimes. Erk1/2 phosphorylation and protein were evaluated byimmunoblotting with antibodies specific to total or phoshorylated Erk1/2.(B) Fluorescence-activated cell sorting (FACS) analysis in preconfluentand postconfluent HT29 cells with or without the 100 nmol/l IGF-1.Preconfluent cells treated with IGF-1 showed an accumulation in S-phase; conversely, postconfluentcells showed an accumulation in G0G1.

Paper

Gut 2009;000:0–10. doi:10.1136/gut.2008.158386 5

Increased IRS2 expression in HT29 and Ls174T cells viaadenoviral CDX2 infectionTo prove if CDX2 was also able to induce IRS2 expression in theabsence of a functioning APC programme, we evaluatedexpression of IRS2 in APC-mutated HT29 cells after infectionwith CDX2 or control LacZ adenoviruses. Induction of CDX2expression in HT29 cells was correlated with a significantupregulation of IRS2 (fig 5C), and KLF4 and P21 mRNA levels(data not shown). Also, the forced CDX2 expression viaAdCDX2 in preconfluent HT29 cells is linked to increasedIRS2 protein levels (Supplementary fig S2). Interestingly, thesame results are obtained in b-catenin-mutated Ls174T cellsunder the same conditions (fig 5C), thus underlining theindependence of CDX2 transcriptional activity from thecytoplamic APC and nuclear b-catenin–TCF-4 status of the cell.

CDX2 is needed for IRS2 expression during intestinal celldifferentiationTo verify if the induction of IRS2 expression during enterocytedifferentiation depends on CDX2 levels, we infected HT29 cells

with shCDX2 or shLacZ adenoviruses. The shCDX2 adenovirusthat we generated was able to induce an 85% decrease inendogenous CDX2 mRNA levels (fig 5D). Significant down-regulation of both KLF4 and IRS2 mRNA levels was observedafter NaB- or confluence-induced differentiation (fig 5D), thusproving the necessity of CDX2 for IRS2 expression duringintestinal cell differentiation.

Direct activation of the IRS2 reporter via CDX2 and binding ofCDX2 to the IRS2 promoterCDX2 is a DNA-binding transcription factor, which recognisesTTTAT or ATAAA motifs in the promoter region of its targetgenes.46 We therefore performed an in silico analysis of thehuman IRS2 gene promoter that identified several putativeCDX2 binding sites (fig 6A). We then generated a series ofreporters of the IRS2 gene promoter including a progressivelydecreasing number of the putative binding sites. pGL3-basic wasused as a control. Details of the reporter constructs are given inthe Patients and methods section and in fig 6A. Potential CDX2binding sites were proved to be functionally active in HeLa cells


Figure 4 (A) Insulin receptor substrate2 (IRS2) and caudal-related homeoboxprotein 2 (CDX2) expression in colonadenomas from patients with familialadenomatous polyposis (FAP).Quantitative real-time PCR (qRT-PCR) wasperformed on total RNA from colorectaladenomas and adjacent normal mucosasamples obtained from patients with FAP.Values shown represent the mean (SEM)of four independent determinationsperformed in duplicate. Cyclophilin wasused as a reference gene and values werenormalised to data obtained in the normalmucosa (plotted as 0 on the x-axis). (B)Paraffin-embedded specimens fromnormal colonic mucosa, aberrant cryptfoci (ACF) and colon tumours of ApcMin/+

mice. Sections were immunostained withthe IRS2 antibody to determineexpression and localisation of IRS2protein in normal tissue and tumours.Immunostaining of Ki67 protein was usedas a marker of the proliferative cellcompartment (6400 magnification). (C)Immunohistochemical analysis ofsections of human colorectal tumourswith certain CDX2 expression.Representative fields of sections ofhuman colon tumours stained for CDX2and IRS2 (6160 magnification). (D) IRS2expression was measured in the ileumand colon of CDX2+/2 mice tissues andcorresponding wild-type mice. qRT-PCRwas performed to measure IRS2 andCDX2 mRNA levels. Values shownrepresent the mean (SEM) for twoindependent determinations performed induplicate. Cyclophilin was used as areference gene and values werenormalised to data obtained from wild-type mice. Different lower case lettersindicate significant statistical difference(b different from a, p ,0.05, t test).

Paper

6 Gut 2009;000:0–10. doi:10.1136/gut.2008.158386

(fig 6B) after co-transfection with CDX2 plasmids. The full-length pGL3-IRS2 (22399/+217)39 presents a 2.5-fold inductionby CDX2, yet with a very high basal activity (data not shown).Importantly, a dominant-negative CDX2 mutant expressionplasmid was able to block activation of IRS2 reporters by CDX2(fig 6C).

To prove that CDX2 binds in vivo the IRS2 promoter motifsthat are functionally active in the reporter assay, we performeda ChIP assay in postconfluent (endogenous CDX2 expressing)HT29 cells. As shown in fig 6D, significant precipitation ofCDX2 was observed in the specific regions of the IRS2 genepromoter. Specific validated qRT-PCR primers were usedtogether with a series of controlled settings (no antibody,aspecific antibody, primers amplifying the region of the IRS2gene outside the promoter or far inside the promoter). We

recognise that the promoter motifs are located within 200–400 bp, thus decreasing the specificity of the pull-down for thesingle motifs, for which DNA segment length should be at least600–700 bp. Thus, we decided to present the data as the meanof the copies obtained by the different primer couples. We alsoperformed a ChIP assay in preconfluent (virtually no CDX2expressing) HT29 cells after infection with AdCDX2 andconfirmed the data obtained on endogenous CDX2 withpostconfluent cells. Finally, we performed a gel-shift EMSAusing the four oligonucleotide probes and we proved that atleast two of the four present a strong binding with CDX2 withan elegant competition with the same unlabelled probe (fig 6E).Taken together, the data obtained in the reporter assay, ChIPand gel-shift EMSAs support the compelling hypothesis thatIRS2 is a direct target of CDX2.


Figure 5 (A) HT29 adenomatous polyposis coli (APC)-inducible and LacZ-inducible cells (upper panel) were treated with 120 mM ZnCl2 or vehicle for24 h. Ls174T TR4 and dominant-negative T cell factor 4 (DNTCF-4)-inducible cells (lower panel) were treated with 1 mg/ml doxycyclin (DOX) or vehiclefor 24 h. Quantitative real-time PCR (qRT-PCR) was performed to measure IRS2 and CDX2 mRNA levels. Values shown represent the mean (SEM) forfour independent determinations performed in duplicate. Cyclophilin was used as a reference gene and values were normalised to data obtained invehicle-treated cells. (B) Immunostaining was performed to visualise IRS2 protein in HT29-APC-inducible cells treated with ZnCl2 or vehicle (61260magnification). (C) <Expression of insulin receptor substrate 2 (IRS2) in HT29 (upper panel) and Ls174T (lower panel) cells after adenoviral caudal-related homeobox protein 2 (CDX2) infection. qRT-PCR was performed to measure IRS2 and CDX2 mRNA levels 48 h after infection with 100multiplicity of infection (MOI) of CDX2 or control LacZ adenoviruses. Values shown represent the mean (SEM) for four independent determinationsperformed in duplicate. Cyclophilin was used as a reference gene and values were normalised to data obtained from control LacZ adenovirus-infectedcells. (D) IRS2, Kruppel-like factor 4 (KLF4) and CDX2 expression levels were measured in HT29 cells treated with sodium butyrate (NaB) and underconfluence after infection with short hairpin (sh) CDX2 or shLacZ control adenoviruses. qRT-PCR was performed to measure IRS2 and CDX2 mRNAlevels. Values shown represent the mean (SEM) of four independent determinations performed in duplicate. Cyclophilin was used as a reference geneand values were normalised to data obtained from adenovirus shLacZ-treated cells. Different lower case letters indicate significant statistical difference(b different from a, p ,0.05, t test).

Paper

Gut 2009;000:0–10. doi:10.1136/gut.2008.158386 7

DISCUSSIONIn the present study we show that IRS2 is significantlyexpressed in the intestine where it is localised in thedifferentiated absorptive enterocytes of the apical villus–epithelial compartment. IRS2 expression is highly regulatedduring the differentiation of intestinal epithelial cells via a directmechanism involving the transcription factor CDX2.

The process of intestinal mucosa self-renewal and differentia-tion is tightly controlled by a series of events that involveseveral regulatory transcriptional pathways.2 The intestinalepithelium is organised in crypt–villus functional units; theseunits are organised in a dynamic scenario in which progenitorcells are generated from the stem cell population at the cryptbottom and migrate along the crypt axis until exfoliation as


Figure 6 Transactivation of insulin receptor substrate 2 (IRS2) reporter assay by caudal-related homeobox protein 2 (CDX2). (A) Shown is aschematic representation of the IRS2 gene promoter. Putative CDX-binding elements (CBS1, CBS2, CBS3, CBS4) are depicted. Transcriptionalactivation of the different reporter constructs was measured in HeLa cells in the presence of wild-type full-length human CDX2 (B) or dominant-negativemutant CDX2 (C) expression plasmids or control (pCDNA4b). pGL3-basic was used as a negative control. Different lower case letters indicatesignificant statistical difference (b different from a, p ,0.05; c different from b and a, p ,0.05; d different from c, b and a, p ,0.05, analysis ofvariance). (D) A chromatin immunoprecipitation assay was performed in preconfluent (low CDX2 expressing) and postconfluent (high CDX2 expressing)HT29 cells and in preconfluent HT29 cells after transfection of CDX2 expression plasmid or control (pCDNA4B). Experiments were performed with andwithout the CDX2 antibody. Quantitative real-time PCR (qRT-PCR) was used to measure the amount of DNA that was pulled down during theexperiments. Different lower case letters indicate significant statistical difference. Experiments were performed with and without the CDX2 antibody.qRT-PCR was used to measure the amount of DNA that was pulled down during the experiments. Different lower case letters indicate significantstatistical difference (b different from a, p ,0.05, t test). (E) Gel-shift electrophoretic mobility shift assays were performed to show the physicalinteraction between CDX2 protein and IRS2 promoter DNA. Synthetic oligonucleotides representing CDX2 putative binding sites on the human IRS2gene promoter are shown. Note in bold the CDX core-binding sequence (ATAAA). Nuclear extract (5 mg) from HeLa cells infected with adenovirus (Ad)CDX2 was incubated with the 32P-labelled IRS2 oligonucleotide probe alone (lane 2) and in the presence of 50, 100 or 500 unlabelled IRS2 competitoroligonucleotides (lanes 3–5). In lane 1, only 32P-labelled IRS2 oligonucleotide was added. Samples were loaded on a 5% acrylamide gel.

Paper

8 Gut 2009;000:0–10. doi:10.1136/gut.2008.158386

terminally differentiated cells at the villus tip. The mechanismscontrolling cell transitions from one compartment to the nextare complex and involve transcription factors that switchcompartment-specific genes on and off. CDX2 has been shownto confer positional information along the crypt–villus axis,inducing the transcription of genes associated with intestinalepithelium differentiation. Multiple pieces of evidence indicatethat the Wnt cascade and the APC programme act as dominantforces controlling cell migration and differentiation along thecrypt–villus axis.2 Mutations in this cascade inevitably result indevelopment of CRC. Expression of CDX2 has been found todecrease with tumour grade in human CRC16 and in mouse andrat models of intestinal tumourigenesis.11 16 Nevertheless, it hasbeen reported that Cdx2 expression is maintained but hetero-geneous in most human colon cancers,18 22 with levels ofexpression related to the degree of differentiation of thetumours.17 Knowledge of the factors that control cell fate inthe intestine will contribute to the discovery of the nutritionaland metabolic hits that drive the downstream elements ofactivation of the constitutive Wnt pathway during CRCprogression.7

A growing body of evidence implies IGF-1, IGF-2, IGF-bindingproteins (IGFBPs) and insulin in CRC progression.47–50 The IRSproteins are central mediators of insulin/IGF signalling and areimplicated in basic cellular functions such as growth, survivaland metabolism.25 IRS1 is considered a marker of active IGF-1signalling pathways in tumours.50 IGF-1 receptor-inducedactivation of IRS1 regulates subcellular localisation and activityof b-catenin, activating b-catenin/TCF-4 target genes, such ascyclin D1.51 Reductions in crypt IRS1 levels increase apoptosis ofcrypt stem or progenitor cells, protect against b-catenin-drivenintestinal tumours and reduce Sox9, a b-catenin/TCF-4 targetgene. Indeed, disruption of IRS1 inhibits growth of colon cancercells31 52 and leads to a protection against tumour formation inthe background of ApcMin/+ mice.30 While IRS1 seems to beprimarily involved in the antiapoptotic and trophic actions ofinsulin and IGF, IRS2 might have a preferential role inmetabolism.47 Nothing is currently known about its expressionand regulation in intestinal epithelia.

Using gain- and loss-of-function models, we provide evidencefor a direct regulation of IRS2 expression via CDX2.Interestingly, induction of IRS2 expression by CDX2 appearsto be independent from APC. On one hand, re-expression of awild-type APC protein in APC-mutated HT29 cells45 is able toinduce the expression of CDX2 and IRS2. On the other hand,there is no restoration of CDX2 and IRS2 expression afterinhibition of b-catenin/TFC-4 transcriptional activity viaDNTCF-4 in b-catenin-mutated Ls174T cells.35 These findingspoint to an APC-dependent mechanism as the basis of theregulation of CDX2 expression in intestinal cells. Nevertheless,expression of CDX2 via adenoviral infection is able to activatethe transcription of IRS2 as well as KLF4 in both cellularmodels, thus proving the independence of CDX2 transcriptionalactivity from the APC or b-catenin/TCF-4 status of the cells.

Here we present a novel scenario for the regulation of IRS2 inthe intestinal architecture. In the crypt–villus axis, the IRS2protein is specifically localised in the terminally differentiatedsurface compartment. IRS2 expression is almost lost in color-ectal adenomas initiated by APC mutation. Furthermore, IRS2expression directly correlates with the degree of spontaneous orinduced enterocytic differentiation in CRC cells and withgrowth arrest, IRS2 being a direct target of the intestine-specifictranscription factor CDX2. One could speculate that asproliferative progenitors migrate along the crypt axis to

differentiate into mature enterocytes, CDX2, by inducingIRS2 expression, could contribute to switching insulin/IGF-related signalling pathways towards differentiation and meta-bolism. Future studies aimed at disclosing cell type-specificactivities of IRS proteins via the generation of intestinal-specifictransgenic mice are needed to verify this intriguing possibility.

Acknowledgements: We are indebted to Drs J Auwerx, F Beck, H Clevers, MGerman, F Giorgino, M Kasuga, K W Kinzler, A V Lee, S Perrini, K Schoonjans and BVogelstein for their valuable tools. We thank A Di Santo, V Evangelista, G Lo Sasso, NMartelli and S Manarini for their valuable help during the study.

Funding: A Moschetta is funded by a Start Up Grant 2005 of the Italian Associationfor Cancer Research (AIRC, Milan, Italy), by University of Bari, Italy, and by ERC-STG-IDEAS. A Morgano is a fellow of the G d’Annunzio University Oncology Program. MP issupported by the Rosario Samanin Fund. M-TV is supported by the Association pour laRecherche sur le Cancer, France.

Competing interests: None.

Ethics approval: The Ethics Committees of Consorzio Mario Negri Sud (Santa MariaImbaro, Chieti, Italy), University ‘‘G. D’Annunzio’’ (Chieti, Italy) and University LouisPasteur (Strasbourg, France) approved the study protocols for ApcMin/+ mice, patientswith FAP and CDX2+/2 mice, respectively.

REFERENCES1. Booth C, Potten CS. Gut instincts: thoughts on intestinal epithelial stem cells. J Clin

Invest 2000;105:1493–9.2. Clevers H. Wnt/beta-catenin signaling in development and disease. Cell

2006;127:469–80.3. Behrens J, Jerchow BA, Wurtele M, et al. Functional interaction of an axin homolog,

conductin, with beta-catenin, APC, and GSK3beta. Science 1998;280:596–9.4. He TC, Sparks AB, Rago C, et al. Identification of c-MYC as a target of the APC

pathway. Science 1998;281:1509–12.5. Sansom OJ, Reed KR, Hayes AJ, et al. Loss of Apc in vivo immediately perturbs Wnt

signaling, differentiation, and migration. Genes Dev 2004;18:1385–90.6. Kinzler KW, Nilbert MC, Su LK, et al. Identification of FAP locus genes from

chromosome 5q21. Science 1991;253:661–5.7. Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell

1990;61:759–67.8. Kinzler KW, Vogelstein B. Lessons from hereditary colorectal cancer. Cell

1996;87:159–70.9. da Costa LT, He TC, Yu J, et al. CDX2 is mutated in a colorectal cancer with normal

APC/beta-catenin signaling. Oncogene 1999;18:5010–4.10. Dang DT, Mahatan CS, Dang LH, et al. Expression of the gut-enriched Kruppel-like

factor (Kruppel-like factor 4) gene in the human colon cancer cell line RKO isdependent on CDX2. Oncogene 2001;20:4884–90.

11. Chawengsaksophak K, James R, Hammond VE, et al. Homeosis and intestinaltumours in Cdx2 mutant mice. Nature 1997;386:84–7.

12. Silberg DG, Swain GP, Suh ER, et al. Cdx1 and cdx2 expression during intestinaldevelopment. Gastroenterology 2000;119:961–71.

13. Beck F, Chawengsaksophak K, Waring P, et al. Reprogramming of intestinaldifferentiation and intercalary regeneration in Cdx2 mutant mice. Proc Natl Acad SciUSA 1999;96:7318–23.

14. Suh E, Chen L, Taylor J, et al. A homeodomain protein related to caudal regulatesintestine-specific gene transcription. Mol Cell Biol 1994;14:7340–51.

15. Bai YQ, Miyake S, Iwai T, et al. CDX2, a homeobox transcription factor, upregulatestranscription of the p21/WAF1/CIP1 gene. Oncogene 2003;22:7942–9.

16. Ee HC, Erler T, Bhathal PS, et al. Cdx-2 homeodomain protein expression in humanand rat colorectal adenoma and carcinoma. Am J Pathol 1995;147:586–92.

17. Subtil C, Guerin E, Schneider A, et al. Frequent rearrangements and amplification ofthe CDX2 homeobox gene in human sporadic colorectal cancers with chromosomalinstability. Cancer Lett 2007;247:197–203.

18. Brabletz T, Spaderna S, Kolb J, et al. Down-regulation of the homeodomain factorCdx2 in colorectal cancer by collagen type I: an active role for the tumor environmentin malignant tumor progression. Cancer Res 2004;64:6973–7.

19. Aoki K, Tamai Y, Horiike S, et al. Colonic polyposis caused by mTOR-mediatedchromosomal instability in Apc+/Delta716 Cdx2+/2 compound mutant mice. NatGenet 2003;35:323–30.

20. Bonhomme C, Duluc I, Martin E, et al. The Cdx2 homeobox gene has a tumoursuppressor function in the distal colon in addition to a homeotic role during gutdevelopment. Gut 2003;52:1465–71.

21. Gross I, Duluc I, Benameur T, et al. The intestine-specific homeobox gene Cdx2decreases mobility and antagonizes dissemination of colon cancer cells. Oncogene2008;27:107–15.

22. Witek ME, Nielsen K, Walters R, et al. The putative tumor suppressor Cdx2 isoverexpressed by human colorectal adenocarcinomas. Clin Cancer Res2005;11:8549–56.

23. Martin ES, Tonon G, Sinha R, et al. Common and distinct genomic events in sporadiccolorectal cancer and diverse cancer types. Cancer Res 2007;67:10736–43.


Paper

Gut 2009;000:0–10. doi:10.1136/gut.2008.158386 9

24. Dang LH, Chen F, Ying C, et al. CDX2 has tumorigenic potential in the human coloncancer cell lines LOVO and SW48. Oncogene 2006;25:2264–72.

25. Taniguchi CM, Emanuelli B, Kahn CR. Critical nodes in signalling pathways: insightsinto insulin action. Nat Rev Mol Cell Biol 2006;7:85–96.

26. Burks DJ, Font dM, Schubert M, et al. IRS-2 pathways integrate femalereproduction and energy homeostasis. Nature 2000;407:377–82.

27. Lin X, Taguchi A, Park S, et al. Dysregulation of insulin receptor substrate 2 in betacells and brain causes obesity and diabetes. J Clin Invest 2004;114:908–16.

28. Kubota N, Terauchi Y, Tobe K, et al. Insulin receptor substrate 2 plays a crucial rolein beta cells and the hypothalamus. J Clin Invest 2004;114:917–27.

29. Withers DJ, Gutierrez JS, Towery H, et al. Disruption of IRS-2 causes type 2diabetes in mice. Nature 1998;391:900–4.

30. Ramocki NM, Wilkins HR, Magness ST, et al. IRS-1 deficiency promotes apoptosisin the putative intestinal crypt stem cell region, limits Apcmin/+ tumors, andregulates Sox9. Endocrinology 2008;149:261–7.

31. Shi B, Sepp-Lorenzino L, Prisco M, et al. Micro RNA 145 targets the insulin receptorsubstrate-1 and inhibits the growth of colon cancer cells. J Biol Chem2007;282:32582–90.

32. Slattery ML, Samowitz W, Curtin K, et al. Associations among IRS1, IRS2, IGF1, andIGFBP3 genetic polymorphisms and colorectal cancer. Cancer Epidemiol BiomarkersPrev 2004;13:1206–14.

33. Wood LD, Parsons DW, Jones S, et al. The genomic landscapes of human breastand colorectal cancers. Science 2007;318:1108–13.

34. Cama A, Palmirotta R, Curia MC, et al. Multiplex PCR analysis and genotype–phenotype correlations of frequent APC mutations. Hum Mutat 1995;5:144–52.

35. Van de Wetering M, Sancho E, Verweij C, et al. The beta-catenin/TCF-4 compleximposes a crypt progenitor phenotype on colorectal cancer cells. Cell2002;111:241–50.

36. Moschetta A, Bookout AL, Mangelsdorf DJ. Prevention of cholesterol gallstonedisease by FXR agonists in a mouse model. Nat Med 2004;10:1352–8.

37. Scholl C, Bansal D, Dohner K, et al. The homeobox gene CDX2 is aberrantlyexpressed in most cases of acute myeloid leukemia and promotes leukemogenesis.J Clin Invest 2007;117:1037–48.

38. Thomas MA, Lichtenstein DL, Krajcsi P, et al. A real-time PCR method to rapidly titeradenovirus stocks. Methods Mol Med 2007;130:185–92.

39. Iwamoto K, Mori H, Okazawa H, et al. Identification of a single nucleotidepolymorphism showing no insulin-mediated suppression of the promoter activity inthe human insulin receptor substrate 2 gene. Diabetologia 2002;45:1182–95.

40. Qualtrough D, Hinoi T, Fearon E, et al. Expression of CDX2 in normal and neoplastichuman colon tissue and during differentiation of an in vitro model system. Gut2002;51:184–90.

41. de Gottardi A, Touri F, Maurer CA, et al. The bile acid nuclear receptor FXR and thebile acid binding protein IBABP are differently expressed in colon cancer. Dig Dis Sci2004;49:982–9.

42. Domon-Dell C, Wang Q, Kim S, et al. Stimulation of the intestinal Cdx2 homeoboxgene by butyrate in colon cancer cells. Gut 2002;50:525–9.

43. Perrini S, Natalicchio A, Laviola L, et al. Abnormalities of insulin-like growth factor-Isignaling and impaired cell proliferation in osteoblasts from subjects withosteoporosis. Endocrinology 2008;149:1302–13.

44. Su LK, Johnson KA, Smith KJ, et al. Association between wild type and mutant APCgene products. Cancer Res 1993;53:2728–31.

45. Morin PJ, Vogelstein B, Kinzler KW. Apoptosis and APC in colorectal tumorigenesis.Proc Natl Acad Sci USA 1996;93:7950–4.

46. Traber PG, Wu GD, Wang W. Novel DNA-binding proteins regulate intestine-specifictranscription of the sucrase-isomaltase gene. Mol Cell Biol 1992;12:3614–27.

47. White MF. Insulin signaling in health and disease. Science 2003;302:1710–1.48. Kaaks R, Toniolo P, Akhmedkhanov A, et al. Serum C-peptide, insulin-like growth

factor (IGF)-I, IGF-binding proteins, and colorectal cancer risk in women. J NatlCancer Inst 2000;92:1592–600.

49. Singh P, Rubin N. Insulinlike growth factors and binding proteins in colon cancer.Gastroenterology 1993;105:1218–37.

50. Werner H, Le Roith D. The insulin-like growth factor-I receptor signaling pathwaysare important for tumorigenesis and inhibition of apoptosis. Crit Rev Oncog1997;8:71–92.

51. Chen J, Wu A, Sun H, et al. Functional significance of type 1 insulin-like growthfactor-mediated nuclear translocation of the insulin receptor substrate-1 and beta-catenin. J Biol Chem 2005;280:29912–20.

52. Simmons JG, Ling Y, Wilkins H, et al. Cell-specific effects of insulin receptorsubstrate-1 deficiency on normal and IGF-I-mediated colon growth. Am J PhysiolGastrointest Liver Physiol 2007;293:G995–1003.


Paper

10 Gut 2009;000:0–10. doi:10.1136/gut.2008.158386

expression and localisation of insulin receptor substrate 2 in normal intestine and colorectal...

Documents