CD39 deletion exacerbates experimental murinecolitis and human polymorphisms increasesusceptibility to inflammatory bowel diseaseDavid J. Friedmana,1, Beat M. Kunzlib,c,1, Yousif I. A-Rahimd, Jean Sevignyb, Pascal O. Berberatc, Keiichi Enjyojib,Eva Csizmadiab, Helmut Friessc, and Simon C. Robsonb,2

aRenal Division and Center for Vascular Biology Research, Beth Israel Deaconess Medical Center, Harvard University, 330 Brookline Avenue, Boston,MA02215; bTransplantation Center and Gastroenterology Division, Beth Israel Deaconess Medical Center, Harvard University, 330 Brookline Avenue, Boston,MA 02215; cDepartment of Surgery, Technische Universitat Munchen, Ismaningerstrasse 22, D-81675 Munich, Germany; and dDepartment ofGastroenterology, University of Hawaii, 651 Ilalo Street, Honolulu, HI 96822

Edited by Ralph M. Steinman, The Rockefeller University, New York, NY, and approved August 24, 2009 (received for review March 18, 2009)

CD39/ENTPD1 hydrolyzes proinflammatory nucleotides to gener-ate adenosine. As purinergic mediators have been implicated inintestinal inflammation, we hypothesized that CD39 might protectagainst inflammatory bowel disease. We studied these possibilitiesin a mouse model of colitis using mice with global CD39 deletion.We then tested whether human genetic polymorphisms in theCD39 gene might influence susceptibility to Crohn’s disease. Weinduced colitis in mice using Dextran Sodium Sulfate (DSS). Read-outs included disease activity scores, histological evidence ofinjury, and markers of inflammatory activity. We used HapMap celllines to find SNPs that tag for CD39 expression, and then comparedthe frequency of subjects with high vs. low CD39-expressiongenotypes in a case-control cohort for Crohn’s disease. Mice nullfor CD39 were highly susceptible to DSS injury, with heterozygotemice showing an intermediate phenotype compared to wild type(WT). We identified a common SNP that tags CD39 mRNA expres-sion levels in man. The SNP tagging low levels of CD39 expressionwas associated with increased susceptibility to Crohn’s disease ina case-control cohort comprised of 1,748 Crohn’s patients and 2,936controls (P � 0.005–0.0006). Our data indicate that CD39 deficiencyexacerbates murine colitis and suggest that CD39 polymorphismsare associated with inflammatory bowel disease in humans.

Crohn’s disease � ENTPD1

Inflammatory bowel disease (IBD) results when the compleximmune balance in the bowel environment is disrupted. Nor-

mally, the immune system suppresses inflammation in the pres-ence of resident gut bacteria or food antigens but can respondrapidly to potentially pathogenic bacteria or viruses with vigor-ous inflammatory responses (1). In IBD, gut immunity is epi-sodically activated in the absence of a clear ‘‘danger signal,’’leading to a constellation of symptoms including abdominal pain,diarrhea, gastrointestinal bleeding, venous thrombotic compli-cations, and often malnutrition (1). Strong genetic componentscontribute to an individual’s susceptibility to IBD, particularlyCrohn’s disease (2).

CD39, also known as ENTPD1, is a vascular and immune cellectonucleotidase that converts extracellular ATP and ADP toAMP. Since ATP is typically proinflammatory and AMP israpidly converted by CD73 into the largely anti-inflammatorymetabolite adenosine, CD39 tends to promote an anti-inflammatory and immune suppressive milieu. Within the im-mune system, CD39 is expressed on cells of both the innate andadaptive immune systems, including T cells (particularly T-regulatory cells) (3), B-cells (4), NK cells (5), NKT cells (6),dendritic cells (4), monocytes (5), and macrophages (7).

Studies in our laboratory indicate a particularly important rolefor CD39 and CD73 in the tandem generation of adenosine withsubsequent activation of adenosine A2a receptors (3). Mice nullfor CD39 have heightened expression of proinflammatory me-

diators such as IFN gamma, IL-1ß, IL-6, and TNF (8). Thesemutant mice are also prone to increased injury in a wide rangeof acute, subacute, and chronic vascular inflammatory modelsincluding transplant rejection (9), ischemia-reperfusion (10),and diabetic microvascular disease (11).

Given the prominent role of CD39 in the response to inflam-mation and adaptive immunity, we asked whether CD39 mightparticipate in regulating the complex immune balance thatgoverns susceptibility to inflammatory bowel disease (IBD). Wetested this hypothesis in mice null for CD39 by administeringDextran Sodium Sulfate (DSS), an experimental form of IBD,and characterized these mice clinically, histologically, and formarkers of immune activation.

Recent advances in genetics and genomics, such as the Hap-Map and the advent of genome wide association studies, havegenerated a wealth of human genetic data. Using data from geneexpression profiling studies and Crohn’s case-control studies, weare able to identify human SNPs strongly associated with CD39mRNA expression and have determined the effect of these SNPson susceptibility to Crohn’s disease.

ResultsInitial Conditions. We studied adult wild-type (WT), heterozygous(hz) for CD39, and CD39-null (KO) C57BL6 mice at 16–22weeks of age. Mice in the treatment groups for DSS (WT 37.9 �1.0 g vs. hz 34.9 � 3.0 g vs. CD39-null 35.4 � 1.2 g; P � 0.05, n �5 for each group) had comparable body weights at initiation ofthe study.

Clinical Course of DSS Colitis. DSS treatment of C57BL6 miceresulted in acute colitis in all mice. We evaluated the clinicalseverity of the colitis using a disease activity index (DAI) (Fig.1A), which incorporates weight loss, stool consistency, and GIbleeding (Fig. 1B). CD39-null mice had significantly worse colitisthan WT mice starting on day 2 and continuing through day 7.Heterozygote mice were intermediate between WT and CD39-null at all time points, but did not differ significantly from eithergroup at any time point (Fig. 1 A).

We measured the hematocrit in each group as another indi-cator of disease severity. There is no difference at baseline in

Author contributions: D.J.F., B.M.K., and S.C.R. designed research; D.J.F., B.M.K., Y.I.A.-R.,J.S., and E.C. performed research; J.S. contributed new reagents/analytic tools; D.J.F.,B.M.K., Y.I.A.-R., P.O.B., K.E., H.F., and S.C.R. analyzed data; and D.J.F., B.M.K., and S.C.R.wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

1D.J.F. and B.M.K. contributed equally.

2To whom correspondence should be addressed. E-mail: srobson@bidmc.harvard.edu.

This article contains supporting information online at www.pnas.org/cgi/content/full/0902869106/DCSupplemental.

16788–16793 � PNAS � September 29, 2009 � vol. 106 � no. 39 www.pnas.org�cgi�doi�10.1073�pnas.0902869106

Dow

nloa

ded

by g

uest

on

Aug

ust 7

, 202

1

hematocrit (approximately 50) between WT and CD39-null mice(12). However, colitis resulted in decreased hematocrit for bothCD39-null (27.8 � 0.8%) and heterozygous (28.0 � 0.4%) micecompared to WT mice (33.8 � 1.0%; P � 0.012 for bothcomparisons (Fig. 1C)).

Pathology. Next we examined bowel from mice in each group.Histologically, CD39-null mice displayed severe transmural in-f lammation with dense wall thickening, profound leukocyticinfiltration and ulceration of colonic mucosa (Fig. 2C), whencompared to heterozygous (Fig. 2B) and WT mice (Fig. 2 A). Wealso examined myeloperoxidase activity in the colons of the miceas a quantitative index of inflammation. CD39 null mice showedsignificantly increased MPO activity (P � 0.01) when comparedto WT mice, with heterozygous mice demonstrating an inter-mediate level of inflammation (Fig. 2D).

Apyrase Rescue. We pretreated CD39-null mice with parenteralapyrase, a soluble factor with enzymatic activity essentiallyidentical to CD39, before chronic exposure with DSS and usedweight as a proxy for colitis severity. Mice receiving apyrasebefore DSS lost no weight compared to mice receiving onlyplacebo (38.0 � 1.0 g vs. 37.7 � 2.3 g), while mice receiving onlyDSS experienced significant weight loss (32.8 � 0.8g; P � 0.05compared to both groups) (Fig. S1, n � 6–10 per group).

Real-Time PCR of CD39 mRNA by Genotype: High Vs. Low ENTPD1Expressors. We analyzed available whole-genome expression pro-filing data in public repositories to generate hypotheses about SNPsin the ENTPD1 gene associated with CD39 mRNA expression

levels in humans. Several SNPs in strong linkage dysequilibriumappeared to correlate with CD39 expression, and we chosers10748643 as the tag SNP because of its proximity to the CD39promoter (�850 base pairs into intron 1, a region often associatedwith transcriptional regulation). We directed tested cell lines fromHapMap subjects of European ancestry who were either AA (n �6) or GG (n � 6) at rs10748643 for CD39 mRNA expression. GGhomozygotes had mRNA levels 43% higher (P � 0.0004), with nooverlap between groups (Fig. 3).

High Vs. Low Expressors Across Four Population Groups. After pub-lication of complete HapMap gene expression profiling sets, we

Fig. 1. Experimental assessment of acute DSS colitis. (A) Disease activityindex (DAI) correlated with severity of disease and CD39-null mice displayedworse DAI compared to WT DSS colitis. (B) Definition of disease activity index.(C) Hematocrit (Hct) values correlated with the pathological findings. A dropin the Hct occurred in all mice, but WT mice maintained a significant higher Hctthan heterozygous or CD39-null mice.

Fig. 2. Acute DSS colitis pathology and myeloperoxidase (MPO) activity. (A)WT mice showed thickening of the colon wall with discrete inflammatoryinfiltrates, but the architecture and polarity of tissue layers were maintained.(B) DSS caused moderate inflammation in heterozygous mice and (C) severeinflammation in CD39-null mice, affecting the epithelial and subserosal layerin heterozygous and severe inflammation in CD39-null mice, with mononu-clear cell infiltrates into the muscularis. (D) Leukocyte recruitment into thecolonic wall during DSS colitis was determined by a standard MPO activityassay. DSS treatment was associated with a significant (P � 0.01) increase inMPO activity in CD39-null mice.

Fig. 3. CD39 mRNA expression by genotype at rs10748643. CD39 mRNAexpression levels (by real time PCR) were normalized for ribosomal 18S subunitexpression level in HapMap lymphocytes. CD39 expression from cell lines fromsubjects with GG alleles at rs10748643 was higher than expression fromsubjects with AA alleles. There was no overlap between groups.

Friedman et al. PNAS � September 29, 2009 � vol. 106 � no. 39 � 16789

MED

ICA

LSC

IEN

CES

Dow

nloa

ded

by g

uest

on

Aug

ust 7

, 202

1

examined CD39 mRNA expression in 210 HapMap subjects fromfour diverse populations to confirm our in vitro findings in a largerset of European subjects and to determine if this expressionphenotype could be generalized to people of other ancestries. InEuropeans, Africans, Chinese, and Japanese subjects, rs10748643genotype correlates strongly with CD39 expression. GG is associ-ated with much higher levels than AA (46–77% increase), withcarriers of the AG genotype displaying intermediate expressionlevels (Fig. 4 and Tables 1 and 2). In Chinese subjects, the GGgenotype was uncommon and AG and GG genotypes were pooledfor analysis. For all populations combined, the G allele was asso-ciated with higher expression levels with a P value on the order of10�22 using an additive, linear model (13). These results wereconfirmed in a second genome wide study, with similarly strongassociation (10�28) (14).

Association of a SNP Tagging ENTPD1 Expression with Crohn’s Disease.Next, we asked whether this SNP strongly correlated to CD39expression might influence human susceptibility to colitis, as wesaw in our mouse models. We obtained data from the WellcomeTrust Case Control Consortium study in Crohn’s disease to testthis hypothesis. At rs10748643, there was significant enrichmentof the low CD39-expressing AA genotype in Crohn’s diseasecases, while controls were enriched for the higher CD39-expressing GG allele (Table 3). Using the Wellcome Trust CaseControl Consortium (WTCCC) expanded control panel (allwithout IBD), we noted an increase in statistical significance inadditive, genotypic, allelic, and homozygote vs. homozygotemodels (Table 3). The odds ratio for the A allele was 1.13(1.04–1.23) or 1.14 (1.06–1.22) using the standard control orexpanded control panel, respectively.

DiscussionIBD may represent a tipping of the immune balance towardexcessive inflammation in genetically susceptible individuals.Crohn’s disease in particular has been associated with a skewingtoward a Th1/IFN gamma-driven inflammatory response (1).Interestingly, CD39-null mice also exhibit a general tendencytoward exaggerated Th1/IFN gamma responses (15).

In the selected mouse model (acute DSS exposure) thatrecapitulates important cardinal features of colitis in IBD suchas diarrhea, weight loss, and gastrointestinal bleeding, we ob-served that CD39-null mice suffer more severe injury than WTmice. Although heterozygous mice did not differ from eithergroup in our overall index (DAI) of IBD-activity, these mice didshow a potential CD39 dose effect in limiting inflammation.DSS-colitis has traditionally been associated with features ofboth Crohn’s disease and ulcerative colitis in rodents, with a biastoward the former.

In CD39-null mice, injury spanned the bowel wall, a typicaland critically important feature of Crohn’s disease that is notseen in classic ulcerative colitis (16). Examination of large bowelin the treated mouse groups revealed substantially more severehistological lesions in the null mice, with loss of crypts and a farmore vigorous inflammatory response. Increased MPO activityin CD39-null mice provided further evidence of this heightenedinflammatory response, emphasizing the role of infiltratingmonocyte-macrophages in this disease process. Treatment ofCD39-null mice with soluble apyrase could reverse weight lossfrom chronic DSS treatment, implicating the nucleotide hydro-lyzing capacity of CD39 as the functional protective mechanism.

Immune system dysregulation is generally considered themajor cause of IBD, but there is a growing appreciation for therole of the vasculature in modifying this response (17–19). CD39is both the dominant immune cell and vascular ectonucleotidase,so its protective effect against DSS colitis in mice may be due toeither of these two roles, or perhaps both. CD39 is important forTreg cell function through generation of adenosine (3), and wewould propose this function as the most likely factor in ourmodel. However, recently others have shown that CD39 mayspecifically modify vascular regulation in murine colitis (20). Weexpect tissue specific deletion of CD39 in immune cells orvascular cells will answer this question definitively in the future.

In mice, several experimental models have been publishedshowing involvement of purinergic signaling in IBD and toxin-induced colitis. The ultimate products of ectonucleotidases arenucleosides/adenosine that activate type 1 A2A adenosine re-ceptors on immune and vascular cells, markedly attenuatingintestinal inflammation in some animal models of IBD (21, 22).CD39 is not expressed on epithelial surfaces but rather hassimilar tissue distribution to the A2A receptor, and the similarresponse to colitis in CD39- and A2A-null mice suggests afunctional interaction. A2A activation protects against colitis inthese studies, associated with decreased production of inflam-matory cytokines such as TNF-� and IFN gamma. Notably, thesecytokines are upregulated in CD39-null mice (23). In contrast,conflicting recent reports show paradoxically that activation ofanother adenosine receptor, A2B, may be either pathogenic or

Table 1. HapMap cell lines from 4 ancestral populations

Genotype CD39 CEU YRI JPT CHB

AA 9 10 20 26AG 30 35 19 17GG 21 15 6 2

Number of cell lines per genotype in each ancestral population.

Table 2. CD39 expression in HapMap cell lines

Relative expression

CEU YRI JPT CHB

AA 1.00 1.00 1.00 1.00AG 1.34 1.13 1.21 1.20GG 1.77 1.46 1.61 1.72

Relative CD39 expression for each population based on genotype atrs10748643.

Fig. 4. CD39 expression in 210 subjects across four populations. (A) CD39mRNA expression in cell lines from subjects with Caucasian/European (CEU),Yoruba (YRI), Chinese (CHB), or Japanese (JPT) ancestry with each rs10748643genotype. For Chinese subjects, AG and GG were combined because therewere only two GG subjects.

16790 � www.pnas.org�cgi�doi�10.1073�pnas.0902869106 Friedman et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 7

, 202

1

of benefit in mouse colitis models (24–26). A2B is a loweraffinity adenosine receptor that is also expressed on the epithe-lial surface and has differing effects to the A2A receptor.

Quite recently, investigators have shown that luminal ATPproduced by gut bacteria plays a critical role in Th17 T-cellimmune deviation in the lamina propria (27). Administration ofATP worsened and luminal administration of apyrase amelio-rated colitis in mouse models (27). These findings suggest thatmore efficient ATP scavenging in the gut environment by thehost organism might protect against colitis. The integral role ofthe ATP receptor P2�7 in the inflammasome complex (28), acritical coordinator of innate immunity, also suggests a potentialcandidate pathway that may mediate this effect. Other studieshave suggested roles for various P2 receptors in colitis, partic-ularly P2�3, but little consensus has emerged (29–32).

We sought further evidence for relevance of the purinergicsystem to human IBD by identifying SNPs associated with highor low CD39 expression. We took advantage of the fact thatCD39 was initially described as a B-cell activation marker (4),and that B-lymphoblast cell lines are available for all HapMapsubjects, along with genotyping at several million SNPs. Initiallywe identified rs10748643 due to its proximity to the promoter,but because of high LD between this SNP and several dozenothers, we are not able to determine if rs10748643 is thefunctional SNP or a SNP that tags the functional SNP. In eithercase, it does appear to distinguish high versus low CD39 expres-sors. We suspect the GG genotype leads to an increase inexpression in Caucasians of closer to the 77% calculated fromour bioinformatic studies than the 43% observed in our in vitrowork, due to chance inclusion of the highest CD39-expressingAA cell line in the CEU HapMap collection in our abbreviatedpanel of six AA cell lines. Nonetheless, all six AA cell lines hadlower CD39 mRNA expression than all six GG cell lines tested.The correlation with this set of SNPs tagged by rs10748643 andCD39 expression was consistent in cell lines across four popu-lation groups, although the frequency of the A vs. G allelesvaried widely between ancestral groups. Previously, we haveshown that CD39 expression levels in B-lymphoblasts correlatewell with ATPase and ADPase functional activity (33).

We examined the effect of rs10748643 in data obtained froma carefully collected and phenotyped case-control study ofCrohn’s disease in people of European ancestry. We found thatcontrol subjects were more likely to have the high CD39-expressing GG genotype, and Crohn’s patients were more likelyto have the low-expressing AA genotype. Although the effectsize was modest in terms of odds ratio (about 1.27 whencomparing homozygote groups), the high frequency of the risk(A) allele, about 40% in Caucasians, means that the effect of thisvariant on populations as a whole (population attributable risk)may be more important than variants in other genes with higherrelative risks but very low frequencies. Conversely, there may berare CD39 variants with greater effects on expression level andmore impact on relative risk.

Although the P value of our variant is also modest, we thinkthis result is best appreciated from the perspective of Bayesianprobability. Rather than ask 500,000 simultaneous tests with

equal prior probability of significance, we first identified a genewith functional importance in the specific disease process in ananimal model. Secondly, we found a SNP that strongly correlateswith expression of that functional gene. Given these first twopieces of evidence, the prior probability of this SNP’s effect onCrohn’s disease is markedly increased compared to a randomlychosen SNP. In this Bayesian context, a P value of 0.005 (or0.0007 using the expanded controls) provides strong evidence ofan effect of this SNP in human Crohn’s disease.

While many studies identify genetically modified mice sus-ceptible to an injury model, and many others find SNPs associ-ated with human disease, it is uncommon when both functionaldata in mice and studies demonstrating relevance in humans arein alignment. We believe that evidence supporting both of thesecriteria makes our study particularly interesting, especially sincethe SNP we identified definitively alters gene expression levels.Our data are particularly intriguing given the links betweeninflammation and thrombosis in IBD (34) and the observationthat CD39 was first characterized as a vascular thromboregula-tory factor (35). Our future aim is to further define the patho-genetic mechanisms underlying the role of CD39 in IBD, con-clusively identify the specific functional SNP and molecular generegulation mechanisms, and analyze more patients to rigorouslyestablish the genetic association of this CD39 polymorphismwith Crohn’s disease in humans.

Materials and MethodsAnimals. Pathogen-free C57BL6 CD39-null and matched wild-type (WT) micewere studied in accordance with standard institutional animal welfare guide-lines. The derivation and characterization of CD39-null mice have been de-scribed elsewhere (12). Genotyping of the CD39 alleles was performed aspreviously described (12). Animal care and experiments were carried outunder the guidelines and protocols approved by the Animal Care and UseCommittee, Beth Israel Deaconess Medical Center, Harvard University, Boston.

Induction and Clinical Assessment of Experimental (Acute) Colitis. DextranSodium Sulfate (DSS) colitis. CD39-null (n � 8), heterozygous for CD39 (n � 11),and WT (n � 12) mice were equally treated with 3% DSS in standard drinkingwater. There was no restriction regarding the dose of DSS solution that wasprovided ad libitum for 7 consecutive days. Control mice received standarddrinking water. Body weights, hemoccult, gross blood, and stool consistencywere analyzed on a daily basis. Disease activity index (DAI) was calculated byscoring percent weight loss, intestinal bleeding [no blood, occult blood(hemoccult �), or gross blood], and stool consistency (normal stool, loosestool, or diarrhea), as previously described (36). The clinical features werescored separately and then correlated with a histological score: DAI � (bodyweight loss) � (diarrhea score) � (rectal bleeding score) (Fig. 1B).

Chronic DSS Colitis. Sixteen- to eighteen-week-old CD39 null mice werefurther studied with a longer course of DSS, with or without apyrase toreconstitute CD39 function. CD39-null mice were treated with four 1-weekintervals of DSS in standard drinking water followed by 1 week off DSStreatment. The total duration of the treatment was 7 weeks. We treated withparenteral apyrase and compared three different conditions: CD39-null micewith DSS (n � 10), CD39-null mice with DSS and apyrase (n � 10), and CD39-nullmice receiving only apyrase (n � 6). We used grade-VII apyrase (derived fromS tuberosum) purchased from Sigma at a dose of 2.5 U/g twice a day by i.p.injection. This apyrase has comparable (approximately1:1) ATPase andADPase activity, paralleling the nucleotide preference of CD39.

Table 3. Association of rs10748643 with Crohn’s disease in a case-control cohort

AA AG GG Additive: P Allelic: P, O.R. Genotypic: P AA vs GG: P, O.R.

Crohn’s cases 391 (22.4%) 885 (50.6%) 472 (27%)Controls 577 (19.7%) 1472 (50.1%) 887 (30.2%) 0.005 0.005, 1.13 (1.04–1.23) 0.0197 0.005, 1.27 (1.07–1.51)Expanded controls 2108 (19.9%) 5236 (49.4%) 3,256 (30.7%) 0.0006 0.0007, 1.13 (1.05–1.22) 0.0026 0.0008, 1.28 (1.11–1.48)

Frequency of AA, AG, and GG genotypes at rs10748643 in control subjects and patients with Crohn’s disease from the Wellcome Trust Case Control Consortium.Expanded controls include additional subjects without Crohn’s disease from the WTCCC (see Methods). O.R.�Odds Ratio, with confidence intervals. The additivemodel is a 1-degree of freedom �2 test for trend. The allelic and homozygote models are 1-degree of freedom �2 tests. The genotypic model is a 2-degree offreedom �2 test.

Friedman et al. PNAS � September 29, 2009 � vol. 106 � no. 39 � 16791

MED

ICA

LSC

IEN

CES

Dow

nloa

ded

by g

uest

on

Aug

ust 7

, 202

1

Histological Assessment of Colitis. Specimens were embedded and snap-frozenas previously described (37) and stored at �80 °C. Four- to five-micrometertissue sections were stained with hematoxylin and eosin. Stained sectionswere examined for evidence of colitis as previously published, (38) using ascriteria the presence of lymphocytic infiltration, macrophages or polymor-phonuclear cells, elongation and/or distortion of crypts, crypt abscesses, re-duction in goblet cell number, ulceration, and edema formation of the colonwall.

Colonic Myeloperoxidase (MPO) Activity. Colon samples were obtained fromcontrol and DSS treated animals, and prepared as previously decribed (39). Thesamples were thawed for MPO activity assay determination according to theo-dianisidine method as previously described (39). In brief, tissue samples werethawed, weighed, suspended in 50 mM potassium phosphate buffer (Kpi), pH6.0, containing 0.5% hexadecyltrimethylammonium bromide buffer (0.1 g/20mL Kpi), and homogenized. After sonication the sample was microcentri-fuged. The reaction was started by mixing and incubating the supernatant in50 mM Kpi, 20 mg/mL o-dianisidine dihydrochloride, and 20 mM hydrogenperoxide. The reaction was stopped by adding 2% sodium azide. The changeof the absorbance was monitored at 460 nm for 10 min. MPO activity wasexpressed as the amount of enzyme necessary to produce a change in absor-bance of 1.0 per min/g wet weight of colonic tissues.

Cell Culture. Lymphoblasts from HapMap participants of European (CEU)ancestry were purchased from Coriell Institute (www.coriell.org). Genotypesfor each cell line were obtained from www.hapmap.org. Cells were grown inRPMI 1640 media with 2 mM L-glutamine, 15% FBS, penicillin-streptomycinmix, and 10 mM HEPES buffer. Cells were seeded at 200,000 per mL andharvested 24–48 h later in log-phase growth.

Real-Time PCR. mRNA was isolated using RNeasy (Qiagen) according to themanufacturer’s instructions. mRNA (0.5 �g) was reversed transcribed to cDNAusing the Taqman Reverse Transcription Kit (Applied Biosystems). Probe-primer sets for ENTPD1 and the 18S ribosomal subunit (loading control) wereobtained from Applied Biosystems. Real-time PCR was performed with anApplied Biosystems 7900 system. ENTPD1 mRNA expression was normalizedusing 18S expression. mRNA was quantified twice from each cell line andaveraged.

Bio-Informatic Assessment of ENTPD1 Expression. Expression data for 210HapMap cell lines was downloaded from the Wellcome Trust Sanger Institute(http://www.sanger.ac.uk/humgen/genevar/). The methods for both genotyp-ing and expression profiling of these cell lines have been described in detail(13). ENTPD1 expression level for each genotype was calculated. We usedlog2-transformed data for tests of statistical significance.

Crohn’s Disease Genetics. Genotypes at rs10748643 for both control andCrohn’s patients in the Wellcome Trust Case Control Consortium (WTCCC)were downloaded from http://www.wtccc.org.uk/. Data were available for1,748 Crohn’s cases and 2,936 apparently healthy controls. We conducted asecond analysis of the data using an expanded control panel (10,600 subjects)that included the healthy controls plus additional subjects from the WellcomeTrust case cohorts. We included hypertension, coronary artery disease, bipolardisorder, and type 2 diabetes cohorts that are not expected to have diseaseSNPs in common with Crohn’s disease, but excluded subjects from autoim-mune disease cohorts (type 1 diabetes and rheumatoid arthritis) that are likelyto have disease SNPs that overlap with Crohn’s disease (40).

Statistical Analysis. Statistical analysis was performed with Microsoft Excel orGraphpad Prism 4 software. For mouse data, we used ANOVA with Newman-Keuls posthoc test for parametric analyses between three or more groups andthe Kruskal-Wallis test with Dunn’s posthoc test for nonparametric analyses.For cell line data, we used unpaired Student t-test for comparison betweengroups. For Crohn’s genetic data, we analyzed the data under a genotypicmodel (two-degree of freedom �2 test), an allelic model (one-degree offreedom �2 test), a homozygote vs. homozygote model (1- degree of freedom�2 test), and an additive model (one-degree of freedom �2 test for lineartrend).

ACKNOWLEDGMENTS. We thank Mark Daly for advice and suggestions; theWellcome Trust and the authors of the Wellcome Trust Case Control Consor-tium study (40) for making their data publicly available; Stranger et al. (13) formaking their gene expression data publicly available; and the National Insti-tute of General Medical Sciences and the Coriell Cell Repository for theirservice in providing human cell lines for use by the research community. Thiswork was supported by National Institutes of Health Grants K08 DK076868,HL57307, HL63972, and HL076540 (to D.J.F. and S.C.R.) and German ResearchFoundation Grants DFG KU 1957/1-1 and DFG KU 1957/3-1 (to B.M.K.).

1. Cho JH (2008) The genetics and immunopathogenesis of inflammatory bowel disease.Nat Rev Immunol 8:458–466.

2. Tysk C, Lindberg E, Jarnerot G, Floderus-Myrhed B (1988) Ulcerative colitis and Crohn’sdisease in an unselected population of monozygotic and dizygotic twins. A study ofheritability and the influence of smoking. Gut 29:990–996.

3. Deaglio S, et al. (2007) Adenosine generation catalyzed by CD39 and CD73 expressedon regulatory T cells mediates immune suppression. J Exp Med 204:1257–1265.

4. Maliszewski CR, et al. (1994) The CD39 lymphoid cell activation antigen. Molecularcloning and structural characterization. J Immunol 153:3574–3583.

5. Pulte ED, et al. (2007) CD39/NTPDase-1 activity and expression in normal leukocytes.Thromb Res 121:309–317.

6. Beldi G, et al. (2008) The role of purinergic signaling in the liver and in transplantation:Effects of extracellular nucleotides on hepatic graft vascular injury, rejection andmetabolism. Front Biosci 13:2588–2603.

7. Goepfert C, et al. (2001) Disordered cellular migration and angiogenesis in cd39-nullmice. Circulation 104:3109–3115.

8. Enjyoji K, et al. (2008) Deletion of Cd39/Entpd1 results in hepatic insulin resistance.Diabetes 57:2311–2320.

9. Robson SC, et al. (2005) Ectonucleotidases of CD39 family modulate vascular inflam-mation and thrombosis in transplantation. Semin Thromb Hemost 31:217–233.

10. Grenz A, et al. (2007) Contribution of E-NTPDase1 (CD39) to renal protection fromischemia-reperfusion injury. FASEB J 21:2863–2873.

11. Friedman DJ, Rennke HG, Csizmadia E, Enjyoji K, Robson SC (2007) The vascularectonucleotidase ENTPD1 is a novel renoprotective factor in diabetic nephropathy.Diabetes 56:2371–2379.

12. Enjyoji K, et al. (1999) Targeted disruption of cd39/ATP diphosphohydrolase results indisordered hemostasis and thromboregulation. Nat Med 5:1010–1017.

13. Stranger BE, et al. (2007) Population genomics of human gene expression. Nat Genet39:1217–1224.

14. Dixon AL, et al. (2007) A genome-wide association study of global gene expression. NatGenet 39:1202–1207.

15. Dwyer KM, et al. (2007) CD39 and control of cellular immune responses. PurinergicSignal 3:171–180.

16. Podolsky DK (2002) Inflammatory bowel disease. N Engl J Med 347:417–429.17. Lomax AE, O’Reilly M, Neshat S, Vanner SJ (2007) Sympathetic vasoconstrictor regu-

lation of mouse colonic submucosal arterioles is altered in experimental colitis.J Physiol 583:719–730.

18. Hatoum OA, et al. (2005) Novel mechanism of vasodilation in inflammatory boweldisease. Arterioscler Thromb Vasc Biol 25:2355–2361.

19. Harris NR, Whatley JR, Carter PR, Morgan GA, Grisham MB (2009) Altered microvascularhemodynamics during the induction and perpetuation of chronic gut inflammation.Am J Physiol Gastrointest Liver Physiol 296:G750–754.

20. Neshat S, et al. (2009) Loss of purinergic vascular regulation in the colon during colitisis associated with upregulation of CD39. Am J Physiol Gastrointest Liver Physiol296:G399–405.

21. Cavalcante IC, et al. (2006) Effect of novel A2A adenosine receptor agonist ATL 313 onClostridium difficile toxin A-induced murine ileal enteritis. Infect Immun 74:2606–2612.

22. Odashima M, et al. (2005) Activation of A2A adenosine receptor attenuates intestinalinflammation in animal models of inflammatory bowel disease. Gastroenterology129:26–33.

23. Enjyoji K, et al. (2008) Deletion of cd39/entpd1 results in hepatic insulin resistance.Diabetes 57:2311–2320.

24. Kolachala VL, et al. (2008) A2B adenosine receptor gene deletion attenuates murinecolitis. Gastroenterology 135:861–870.

25. Kolachala V, et al. (2008) Blockade of adenosine A2B receptors ameliorates murinecolitis. Br J Pharmacol 155:127–137.

26. Frick JS, et al. (2009) Contribution of adenosine A2B receptors to inflammatoryparameters of experimental colitis. J Immunol 182:4957–4964.

27. Atarashi K, et al. (2008) ATP drives lamina propria T(H)17 cell differentiation. Nature455:808–812.

28. Mariathasan S, et al. (2006) Cryopyrin activates the inflammasome in response to toxinsand ATP. Nature 440:228–232.

29. Rybaczyk L, et al. (2009) New bioinformatics approach to analyze gene expressions andsignaling pathways reveals unique purine gene dysregulation profiles that distinguishbetween CD and UC. Inflamm Bowel Dis 15:971–984.

30. Yiangou Y, et al. (2001) ATP-gated ion channel P2X(3) is increased in human inflam-matory bowel disease. Neurogastroenterol Motil 13:365–369.

31. Grbic DM, Degagne E, Langlois C, Dupuis AA, Gendron FP (2008) Intestinal inflamma-tion increases the expression of the P2Y6 receptor on epithelial cells and the release ofCXC chemokine ligand 8 by UDP. J Immunol 180:2659–2668.

32. Wynn G, Ma B, Ruan HZ, Burnstock G (2004) Purinergic component of mechanosensorytransduction is increased in a rat model of colitis. Am J Physiol Gastrointest Liver Physiol287:G647–657.

33. Friedman DJ, et al. (2009) Functional ENTPD1 polymorphisms in African Americans withdiabetes and end-stage renal disease. Diabetes 58:999–1006.

34. Danese S, et al. (2007) Inflammation and coagulation in inflammatory bowel disease:The clot thickens. Am J Gastroenterol 102:174–186.

16792 � www.pnas.org�cgi�doi�10.1073�pnas.0902869106 Friedman et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 7

, 202

1

35. Robson SC, Sevigny J, Zimmermann H (2006) The E-NTPDase family of ectonucleoti-dases: Structure function relationships and pathophysiological significance. PurinergicSignal 2:409–430.

36. Berberat PO, et al. (2005) Heme oxygenase-1-generated biliverdin ameliorates exper-imental murine colitis. Inflamm Bowel Dis 11:350–359.

37. Kunzli BM, et al. (2007) Upregulation of CD39/NTPDases and P2 receptors in humanpancreatic disease. Am J Physiol Gastrointest Liver Physiol 292:G223–230.

38. Boirivant M, Fuss IJ, Chu A, Strober W (1998) Oxazolone colitis: A murine model of Thelper cell type 2 colitis treatable with antibodies to interleukin 4. J Exp Med 188:1929–1939.

39. Bauer P, Russell JM, Granger DN (1999) Role of endotoxin in intestinal reperfusion-induced expression of E-selectin. Am J Physiol 276:G479–484.

40. Anonymous (2007) Genome-wide association study of 14,000 cases of seven commondiseases and 3,000 shared controls. Nature 447:661–678.

Friedman et al. PNAS � September 29, 2009 � vol. 106 � no. 39 � 16793

MED

ICA

LSC

IEN

CES

Dow

nloa

ded

by g

uest

on

Aug

ust 7

, 202

1