of April 16, 2018.This information is current as

Th2 Cytokine Gene LocusIdentifies Ets-1 as a Novel Regulator of the Phylogenetic and Functional Analysis

Donata VercelliJannine M. Strempel, Roland Grenningloh, I-Cheng Ho and

http://www.jimmunol.org/content/184/3/1309doi: 10.4049/jimmunol.0804162December 2009;

2010; 184:1309-1316; Prepublished online 28J Immunol 

MaterialSupplementary

2.DC1http://www.jimmunol.org/content/suppl/2009/12/28/jimmunol.080416

Referenceshttp://www.jimmunol.org/content/184/3/1309.full#ref-list-1

, 22 of which you can access for free at: cites 55 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2010 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on April 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on April 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Phylogenetic andFunctionalAnalysis Identifies Ets-1 as aNovelRegulator of the Th2 Cytokine Gene Locus

Jannine M. Strempel,* Roland Grenningloh,†,‡ I-Cheng Ho,†,‡ and Donata Vercelli*,x,{

The Th2 cytokine gene locus has emerged as a remarkable example of coordinated gene expression, the regulation of which seems to

be rooted in an extensive array of cis-regulatory regions. Using a hypothesis-generating computational approach that integrated

multispecies (n = 11) sequence comparisons with algorithm-based transcription factor binding-site predictions, we sought to

identify evolutionarily conserved noncoding regions (ECRs) and motifs shared among them, which may underlie coregulation.

Twenty-two transcription factor families were predicted to have binding sites in at least two Th2 ECRs. The ranking of these

shared motifs according to their distribution and relative frequency pointed to a regulatory hierarchy among the transcription

factor families. GATA sites were the most prevalent and widely distributed, consistent with the known role of GATA3 as a Th2

master switch. Unexpectedly, sites for ETS-domain proteins were also predicted within several Th2 ECRs and the majority of

these sites were found to support Ets-1 binding in vitro and in vivo. Of note, the expression of all three Th2 cytokines (IL-5, -13,

and -4) was significantly and selectively decreased in Th2 cells generated from Ets-1–deficient mice. Collectively, these data suggest

that Ets-1 contributes to Th2 cytokine gene regulation by interacting with multiple cis-regulatory regions throughout the Th2

locus. The Journal of Immunology, 2010, 184: 1309–1316.

The spatio-temporal control of eukaryotic gene expressionis a complex process, orchestrated to a large extent bymechanisms that control transcription. Regulation of in-

dividual loci often relies on multiple cis-regulatory elements, lo-cated proximal (promoters) and distal (enhancers, silencers, andinsulators) to the transcription start site. Clusters of functionallyrelated genes may exhibit additional complexity and share mul-tiple cis-regulatory elements that transduce developmental and/orenvironmental cues to specify sequential expression of individualgenes, as in the b-globin gene cluster (1), or coordinated ex-pression of several genes, as is the case for the Th2 cytokine genelocus (2), which includes IL5, IL13, and IL4.The Th2 cytokine genes are grouped within a 150-kb region of

human chromosome 5q31 and the syntenic region of mouse chro-mosome 11. IL13 and IL4 are adjacent to one another, whereas IL5 isseparated from these two by a large (85 kb) DNA repair gene,RAD50, a configuration that has been preserved for $300 millionyears (3). The regulation of this locus has been studied most ex-

tensively in the Th2 lineage and was shown to provide a remarkableexample of coordinated gene expression (4, 5). Such coordinationseems to be highly adaptive, in that the distinct components of theTh2 cytokine trio cooperate, ensuring the deployment of comple-mentary antiparasite defense pathways (6). In contrast, when dys-regulated by genetic and/or environmental influences, coordinatedTh2 cytokine expression becomes critical for the pathogenesis ofallergic inflammation (7, 8).An extensive array of highly conserved cis-regulatory elements

essential for the coordinated expression of Th2 cytokine geneshas been identified by a combination of DNase I hypersensitivesite (HS) mapping and comparative sequence analysis (Fig. 2A).HSs 1 and 2 (9), located between Il13 and Il4, map to conservednoncoding sequence (CNS)-1, a potent enhancer of cytokinegene expression in T cells (10–12) (Fig. 2A). Two other regions,which map to the second intron of Il4 (HS II and III; Il4 intronicenhancer [IE]) and downstream of Il4 (HSV/Va; CNS-2), exhibitenhancer activity in Th2 cells and mast cells (13–15). SeveralHSs (RAD50 HS [RHS]5–7) clustered within the 39 end of theDNA repair gene RAD50 (16, 17) collectively function as a locuscontrol region (LCR), conferring robust Th2-specific, position-independent and copy number-dependent expression to the Il4and Il13 genes (16, 18). Finally, a single silencer located at the 39end of Il4 (HSIV) is required to suppress Il4 expression in naiveCD4+ T and Th1 cells (14, 19). Of note, the major Th2 cis-reg-ulatory elements typically map to extensive (300–600 bp) regionsthat are highly conserved between mice and humans.Several lines of evidence point to a significant degree of in-

terdependency among distant regulatory regions within the Th2

locus. Indeed, analysis of transgenic models has shown that no

single element is sufficient to fully recapitulate Th2 cytokine gene

expression at the levels driven by an intact Th2 locus (16, 18).

Conversely, deletion of any enhancer region is sufficient to

markedly inhibit the expression of more than one Th2 cytokine

gene (11, 15, 20). Moreover, characterization of long-range in-

trachromosomal interactions at the murine Th2 locus determined

that in T cells, the Il5, Il13, and Il4 promoters are physically

*Functional Genomics Laboratory, Arizona Respiratory Center, xArizona Center forthe Biology of Complex Diseases, and {Department of Cell Biology, College ofMedicine, University of Arizona, Tucson, AZ 85719; †Division of Rheumatology,Immunology, and Allergy, Department of Medicine, Brigham and Women’s Hospital;and ‡Harvard Medical School, Boston, MA 02115

Received for publication December 12, 2008. Accepted for publication November24, 2009.

This work was supported by National Institutes of Health Grant 5RO1HL66391 (toD.V.) and a National Science Foundation Integrative Graduate Education and Re-search Traineeship in Genomics (to J.M.S.).

Address correspondence and reprint requests to Dr. Donata Vercelli, 1657 East HelenStreet, 321 Thomas W. Keating Bioresearch Building, Tucson, AZ 85719. E-mailaddress: donata@arc.arizona.edu

The online version of this article contains supplemental material.

Abbreviations used in this paper: ChIP, chromatin immunoprecipitation; CNS, con-served noncoding region; ECR, evolutionarily conserved region; HS, hypersensitivesite; IE, intronic enhancer; LCR, locus control region; MULAN, multiple sequencelocal alignment and visualization tool; P, promoter; PWM, position weight matrix;rhEts-1, recombinant human Ets-1; RHS, RAD50 hypersensitive site; TF, transcrip-tion factor.

Copyright� 2010 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0804162

by guest on April 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

clustered with one another and with the distant regulatory ele-

ments, forming a distinctive chromatin hub (21).Theadvancedstructural andfunctionalcharacterizationof theTh2

locus provides a unique opportunity to decipher molecular cues thatmight dictate the function of individual regulatory regions, aswell asspecify coordinated gene expression. A simple model to orchestratethe coexpression of distinct genes predicts that their promoters willcontain a common set of transcription factor binding sites critical fortransducing relevant signals. Building on such a model, we tooka comprehensive in silico approach, and complemented it withfunctional invitroand invivostudies, todecipher the regulatory logicunderlying the coexpression of the Th2 cytokine gene cluster. Al-though computational predictions of transcription factor bindingsites have been hindered by considerable false-positive rates, suchpredictions are substantially improved by integrating motif-findingalgorithms with phylogenetic comparisons (22–24) and furtherstrengthened by functional validation. In this study, we show thatour hypothesis-generating, multipronged approach succeeded inidentifying Ets-1 as a novel, important regulator of coordinated Th2cytokine gene expression.

Materials and MethodsMultispecies comparative analysis

Genomic sequences corresponding to the human Th2 cytokine gene cluster(20 kb downstream of IL5 through the 39 end of KIF3A) and the syntenicregions in Pan troglodytes (chimpanzee), Papio anubis (baboon), Callithrixjacchus (marmoset),Otolemurgarnetti (bush baby),Bos taurus (cow),Canisfamiliaris (dog), Rattus norvegicus (rat), Mus musculus (mouse), Mono-delphis domestica (opossum), and Gallus gallus (chicken) were obtainedfrom NCBI. Some of the sequence used was generated by the National In-stitutes of Health Intramural Sequencing Center (www.nisc.nih.gov).

Generation of a multispecies alignment and identification of evolu-tionarily conserved regions (ECRs) were performed with the MUltiplesequence Local AligNment and visualization tool (MULAN) program(http://mulan.dcode.org/) (25). MULAN uses a local alignment strategyusing the threaded blockset aligner program and uses the phylogeneticrelationships of the sequences provided to build the multispecies alignment(25). Repeat masking was performed on all sequences with the species-appropriate filters prior to alignment. A large gap due to incomplete se-quence data was detected at the 39 end of RAD50 in C. jacchus. Therefore,our analysis of this region (which spans RHS6.1, 6.2, and 7 within the Th2LCR) did not include this species. Because of the very close relationshipbetween human and marmoset, it is unlikely that exclusion of this speciessignificantly altered the multispecies alignment or the transcription factorbinding-site profiles generated for this region.

Identification of conserved transcription factor binding sites

Genomic regions (excluding exons) identified by the MULAN program asmultispecies ECRs ($70% identity over $100 bp in all species examined)were examined for putative transcription factor binding sites. Ungappedsequences from each species were analyzed individually with the Mat-Inspector program, using the matrix library version 6.0, which contains 431position weight matrices (PWMs) for vertebrate transcription factors, rep-resenting 148 families, (Genomatix, Munich, Germany, http://www.ge-nomatix.de/). PWMs describe transcription factor binding sites in terms ofthe complete nucleotide distribution for each single position, allowing forquantification of similarity between the weight matrix and the putativebinding motif (26). Detection criteria were set to only report matches thatdemonstrated a core similarity score$0.85 and amatrix similarity score thatmet or exceeded the optimization threshold defined for each PWM. Theoptimization threshold is a parameter designed to significantly reduce therate of false positives by limiting detection to no more than three hits within10 kb of nonregulatory sequence (26). We identified those transcriptionfactor binding sites predicted independently in 100% of the species exam-ined, includingmatches to differentmatriceswithin a family.Matrix familiesrepresent functionally related transcription factors that also have PWMs thatare essentially indistinguishable (26). Of these sites, those thatmapped to thesame locationwithin themultispecies alignmentwere included in the profile.For those cases in which multiple matrix families met the conservation cri-teria and mapped to the same location in the alignment, all families wereincluded in the profile.

EMSA

Single-stranded complementary oligonucleotides were annealed and PAGEpurified. Annealed oligonucleotides were end-labeled with g-[32P] ATP usingT4 polynucleotide kinase. Binding reactions consisted of 50 ng human re-combinant Ets-1 (Axxora, SanDiego,CA) in binding buffer (25mMTris-Cl, 1mM EDTA, 60 mMKCl, 6 mMMgCl, 10 mMDTT, 1 mg Poly (dI·dC), 10%glycerol). Competition assays were performed by preincubation for 30 min atroomtemperaturewithunlabeledoligonucleotide competitors (30- and 90-foldmolar excess) prior to addition of the probe. Samples were incubated withradiolabeled probes for 30 min at room temperature and run on a 5% (w/v)polyacrylamide gel (20 mAmp, 4 h, 4˚C), which was dried and subjected toautoradiography. Oligonucleotide sequences used as probes and/or com-petitors included a previously defined high-affinity Ets-1 motif (SC1_ETS1CGGCCAAGCCGGAAGTGAGTGCC; the core consensus is underlined)(27), a mutant of this motif (SC1_ETS1mut CGGCCAAGCCttAAGT-GAGTGCC), and the conservedETS sites predictedwithin the humanTh2 cis-regulatory regions (IL5P: TGTCTTTGAGGAAATGAATAA, IL13P: GTT-CGGGGAGGAAGTGGGTAG, IL4P.1: GATTTCACAGGAACATTTTAC,IL4P.2: TTTTCTCCTGGAAGAGAGGTG, IL13P: GTTCGGGGAGGAA-GTGGGTAG, RHS5: GGTAACACAGGAAGTCAGCAG, IL4IE: AT-GAAAACAGGAACTGAAATG, HSIV: TCTGCCACAGGATATGGGTAG,CNS2: TGGGTCACAGGAAGCCCAAGA). The intensity of the Ets-1complexes was determined by densitometry.

Mice

Female 3–6-wk-old C57BL/6 mice were obtained from The JacksonLaboratory (Bar Harbor, ME). Ets-1–deficient mice were described pre-viously (28). The animals were housed under specific pathogen-free con-ditions and experiments were performed in accordance with theinstitutional guidelines for animal care at the University of Arizona and theDana-Farber Cancer Institute (Boston, MA) under approved protocols.

T cell isolation and in vitro differentiation of Th2 cells

Murine Th2 cells were generated essentially as described (29). Naive CD4+

T cells were isolated from spleens by negative selection, followed byenrichment using anti-CD62L–coated magnetic beads (Miltenyi Biotec,Auburn, CA). Cells (2–5 3 106) were cultured in the presence of anti-CD3ε Ab (145.2C11, 1 mg/ml) and anti-CD28 (37.51, 1 mg/ml) (BDPharmingen, San Diego, CA) in a flask coated with goat anti-hamster IgG(0.2 mg/ml), under Th2 skewing conditions (1000 U/ml IL-4, 3 mg/ml anti-IL-12, and 5 mg/ml anti–IFN-g; complete medium) for 3 d. Then the cellswere expanded in complete medium containing IL-2 (20 U/ml) for 7 d.Cytokine expression at the single-cell level was examined by intracellularstaining using the following Abs: anti–IL-4 PE (11B11), anti–IL-13 Alexa(eBio13A), and anti–IFN-g FITC (XMG1.2). For restimulation, cells wereincubated with increasing concentrations (0.3–3 mg/ml) of anti-CD3 mAb,in the presence of constant amounts of anti-CD28 Ab (2 mg/ml), for 24 h.Cytokine secretion was quantified by ELISA (Quantikine Immunoassays,R&D Biosystems, Minneapolis, MN).

Chromatin immunoprecipitation

Murine Th2 cells were treated with 1% formaldehyde for 10 min at roomtemperature followedby theadditionofglycine (125mMfinalconcentration)to halt cross-linking. Cells were harvested, washed twice with 13 PBS, andresuspended in cell lysis buffer (5 mM PIPES [pH 8], 85 mM KCl, 0.5%Nonidet P-40) supplemented with protease inhibitors (10 mg/ml Aprotinin,13 EDTA-free complete protease inhibitor mixture [Roche, Basel Swit-zerland], and 1 mM PMSF). Nuclei were collected by centrifugation andlysed in nuclear lysis buffer (50 mM Tris-HCl [pH 8.1], 10 mM EDTA, 1%SDS) supplemented with protease inhibitors as above. Chromatin wassheared by sonication to yield the majority of fragments in the 200–600-bpsize range. An aliquot of chromatin (1 3 107 whole cell equivalents) waspreclearedwith a 50%ProteinA Sepharose slurry (150ml) for 20min at 4˚C.Precleared chromatin was then diluted 10-fold in IP dilution buffer (0.01%SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl [pH 8.1], 167mM NaCl) containing protease inhibitors. Immunoprecipitation reactionswere performedwith 53 106whole-cell equivalents overnight at 4˚Cwith ananti–Ets-1 Ab (C-20, Santa Cruz Biotechnology, Santa Cruz, CA) or normalrabbit IgG (10 mg). Immunocomplexes were collected with Protein A aga-rose beads (Millipore, Bedford, MA), washed once each with low-salt wash,high-salt wash, and LiCl wash, and washed twice with TE. Im-munocomplexes were eluted with IP elution buffer (1% SDS, 50 mMNaHCO3), and cross-links were reversed with an overnight incubation at 65˚C. DNA was purified by phenol/chloroform extraction and ethanol pre-cipitation using glycogen as a carrier. Real-timePCRwas performedwith theQuantitect SYBR Green PCR kit (Qiagen, Valencia, CA) on an ABI Prism

1310 Ets-1 CONTRIBUTES TO Th2 CYTOKINE GENE COREGULATION

by guest on April 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

7900 Sequence Detection System (Applied Biosystems, Foster City, CA).The following primer pairs were used: Il5 promoter: 59-ACCCTGAGTTT-CAGGACTCG-39, 59-TCAGAACACTCAAGTGCAGAAG-39; Rad50H-S5: 59-GTTTGTCGTCTCAGAGTAAAGC-39, 59-GTCACATCCTGAGCATGTTCC-39; Il13 promoter: 59-GTGAGAAG-ATGGAATCTGGC-39, 59-CCCTACTCTTTCCTCATACTG-39; Il4 promoter: 59-CCT-ACGCTGA-TAAGATTAGTCTG-39, 59-CACCAGATTGTCAGTTATTCTG-39; Il4 IE:59-GATGTGACAGGCTGATAGTG-39, 59 GAGACCACTGGTAAG-TC-AAGAC-39; HSIV: 59-GATTGAAACTCATTTGCATGTC-39, 59-GTGT-TCTGACC-ATCCAAGTGTAG-39; CNS-2: 59-GAAGCAGTGTGGGAGAGCGT-39, 59-GGGAAA-GTGATCGTGAGGAG-39; and Cd14 promoter:59-GAATGCCCTAATGCCACTCTG-39 and 59-GCTTGTTCGACCAATTTGGC-39. PCR reactions were performed under the following cyclingconditions: 15 min at 95˚C, followed by 40 cycles of 15 s at 95˚C, 30 s at 57˚C, and 30 s at 72˚C. Dissociation curve analysis and agarose gel electro-phoresis confirmed amplification of a single product for each primer set. Inaddition, PCR efficiency was optimized by adjusting the primer concentra-tion and running serial dilutions of input DNA to generate a standard curve.Chromatin immunoprecipitation (ChIP) data are represented as the PCRsignal generated for the sample immunoprecipitated with anti–Ets-1 Abrelative to normal rabbit IgG (specific enrichment).

Real-time PCR

Total RNA was extracted using TRIzol reagent and reverse-transcribedusing oligo(dT) priming. Negative control reactions without reverse tran-scriptasewere always included. Semiquantitative RT-PCRwas performed induplicate on an ABI Prism 7900 Sequence Detection System using SYBRgreen dye detection. mRNA levels were measured using predesigned primersets for murine Il4, Il5, Il13, Il10, Gata3, and Gapdh (Quantitect PrimerAssays, Qiagen). Products were verified by melting curve analysis. Th2cytokine and Gata3 mRNA levels are expressed relative to Gapdh mRNAabundance within the same sample.

ResultsMost Th2 cis-regulatory elements predate the divergencebetween placental and nonplacental mammals

Toinvestigatetheevolutionaryhistoryof theTh2regulatoryelementsandestablishanappropriate scopeof species for subsequent binding-site analyses, we first assessed to what extent the extensive array ofTh2 regulatory regions characterized in humans and mice is con-served in additional placental mammals and evolutionarily moredistant species. Beyond placentalmammals, orthologs for IL5, IL13,and IL4 were recently identified for a marsupial, the laboratoryopossum (Monodelphis domestica) (30), and the chicken (G. gallus)(3). Although a draft sequence of the platypus (Ornithorhynchusanatinus) genome was recently released (31), the genomic regionexpected to contain orthologs of the Th2 cytokine genes is in-completely covered and could not be included in our analysis.Furthermore, because comparative analysis of syntenic regions andmining of expressed sequence tag libraries failed to identify thesegenes in earlier vertebrate species (teleosts) (32), we did not extendour analysis beyond the avian order.Fig. 1 shows the phylogenetic relationships and estimated di-

vergence times (33–35) for the 11 species included in our study. Amultiple DNA sequence alignment for the entire Th2 cytokinegene cluster (Fig. 2A) was generated with the MULAN program(25), which relies on a local alignment strategy and integrates thephylogenetic relationships among the sequences. Fig. 2B showsthe MULAN-generated alignment for selected regions within theTh2 locus in a subset of species representative of the clades weanalyzed. The ability to detect the exons for IL5, IL13, and IL4 bymultiple species alignment decreased with increasing phyloge-netic distance, consistent with previous reports that the opossum(30) and chicken (3, 32) Th2 cytokines exhibit relatively lowhomology to the human proteins. In contrast, all exons for theDNA repair gene RAD50 were highly conserved and readilyidentified in all species (Fig. 2B and data not shown). As expected,all of the Th2 regulatory regions previously identified in pairwise

human–mouse alignments (Fig. 2A) and validated in functional

studies (10, 11, 15, 17–20) were highlighted as ECRs in the seven

placental mammalian species examined (Fig. 2B). In addition, this

analysis identified a novel noncoding ECR, located ∼6 kb up-

stream of the IL5 gene (IL-5/-6). Of note, only a single noncoding

element (110 bp within RHS5) was conserved in the syntenic

region in chicken. In contrast, the majority of the noncoding ECRs

were highly conserved in the orthologous opossum locus (Fig.

2B), with the exception of the functionally undefined IL5/-6, the

distal region of the IL13 promoter (which encompasses a con-

served GATA-3–responsive element) (36), and HSIV, which is

critical for Il4 silencing in murine Th1 cells (19).Collectively, these data indicate that the majority of conserved

regulatory regions within the Th2 cytokine locus predate the di-vergence of marsupials and placental mammals ∼180 million yearsago (35, 37). The phylogenetic distance between chicken andmammals might preclude alignment-based detection of additionalregulatory regions.

Phylogenetic profiling of Th2-responsive elements

The Th2 cytokine locus ECRs identified by comparative analysiswere then examined to search for shared transcription factorbinding sites that might contribute to coordinated regulation of theTh2 locus. Because some of the Th2 ECRs were undetectable in theopossum genome, we profiled transcription factor binding sitesacross nine placental mammalian species, representing four majorclades (primates, carnivores, artiodactyls, and rodents). Multi-species comparisons of ECRs increase the overall sequence di-vergence, which serves to refine alignments and reveal invariantpositions more likely to reflect true functional constraints asso-ciated with transcription factor binding sites (22–24, 38). Indeed,the sequence diversity within the noncoding Th2 ECRs capturedby the multiple species alignment was ∼20% greater than thatobtained by pairwise comparison between human and mouse se-quences (data not shown). Th2 ECRs from placental mammalswere examined for putative transcription factor binding sites usingthe MatInspector program (26). Searches were conducted on un-gapped sequence from each species using a vertebrate PWM li-brary and stringent match criteria. Sites predicted independently inall species and residing in the same position within the multiplespecies alignment were included in the profile.Within the Th2 ECRs, which collectively span ∼4.8 kb, we

identified 103 conserved motifs representing targets for 52 distincttranscription factor families (data not shown). Fig. 3 shows that 22transcription factor families were predicted to have binding siteswithin at least two Th2 regulatory regions. When multiple sites fora particular factor were found, they typically resided in different

Bos taurus

Canis familiaris

Mus musculus

Rattus norvegicus

Otolemur garnettii

Callithrix jacchusPapio anubis

Pan troglodytes

Homo sapiens

100

Million Years

0

Monodelphis domestica

Gallus gallus

300 200

FIGURE 1. Phylogenetic relationships and estimated divergence times

for species used in comparative analysis. Divergence time estimates were

provided by the literature (33–35).

The Journal of Immunology 1311

by guest on April 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

regions, rather than clustering within one or few regulatory ele-ments. In fact, the number of conserved sites and the number ofdistinct regulatory regions they are predicted to target were pos-itively correlated (R2 = +0.81). The number of motifs common todistinct regions did not correlate with GC content nor linearproximity (data not shown). Therefore, these data are unlikely tomerely reflect genomic bias. Rather, they may highlight an un-derlying regulatory network within the Th2 locus.

Identification and functional validation of Ets-1 binding sites inthe Th2 locus

The ranking of transcription factor binding sites shared across theTh2 locus, according to their distribution and relative frequency(Fig. 3, Supplemental Fig. 1), pointed to a regulatory hierarchyamong transcription factor families. Interestingly, the most prev-alent and widely distributed motifs were those for GATA proteins,consistent with the notion that GATA-3 acts as a master switch forTh2 cell differentiation and Th2 cytokine expression (36, 39–42).This finding validated our approach and by extension raised thepossibility that members of the ETS and SORY/HMG transcrip-tion factor families, which like GATA proteins were predicted totarget six or more Th2 cis-elements, might also play a role in thecoordinated regulation of the Th2 cytokine locus.Our subsequent analysis focused on the ETS family, which had

not been previously implicated in orchestrating the coordinatedexpression of the Th2 cytokines, and particularly on Ets-1, the

family member that is predominant in T cells (43). To functionallyvalidate the ETS motifs predicted within the Th2 ECRs, we usedEMSA and compared and contrasted the ability of oligonucleo-tides corresponding to individual putative ETS sites to bind re-combinant human Ets-1 (rhEts-1). Fig. 4A shows the results of anexperiment representative of this approach. Incubation of rhEts-1with a high-affinity Ets-1 consensus probe (27) led to the forma-tion of a complex (lane 1) that was competed almost completelyupon incubation with a 30-fold molar excess of unlabeled Ets-1consensus oligonucleotide (lane 2) and was specifically super-shifted by an anti–Ets-1 Ab (data not shown). Oligonucleotidesspanning the putative ETS sites in HSIV (lanes 4 and 5) and theIL13 promoter (lanes 6 and 7) also competed Ets-1 binding, albeitless effectively. In contrast, an oligonucleotide encompassing oneof the two putative ETS sites within the IL4 promoter (IL4P.1,lanes 12 and 13) failed to compete Ets-1 binding. Similar ex-periments were performed to test each predicted ETS motif, andthe intensity of the relevant bands was measured by densitometryand compared with the intensity of the uncompeted Ets-1 com-plex. The Ets-1 consensus and an Ets-1 oligonucleotide containingtwo mutations in the ETS core were used as positive and negativecontrols throughout this study.Fig. 4B shows the competition curves for the eight putative ETS

sites identified within the Th2 locus. The motifs within RHS5 andthe IL13 promoter competed Ets-1 binding almost as effectively asthe Ets-1 consensus oligonucleotide. The sites within HSIV, CNS2,

HS

IV

HS

V (C

NS-

2)

RAD50

HS

II (IL

4 IE

)

HS

I (IL

4P)

HS

1,2

(IL13

P)

RHS

5RH

S 6

RHS

7

RHS

1 (IL

5P)

HSS

1-2

(CN

S-1)

31LI5LI IL4 KIF3A

LCR

HS

Va

HSS

3

Mus

Canis

Oto

Papio

IL5

Mono

Gallus

Th2 LCR (Rad50 3’ End)RHS5 RHS6.1/6.2 RHS7P

IL4IEP HSIV CNS-2

Mus

Canis

Oto

Papio

Mono

Gallus

P HSS3CNS-1

IL13

IL5/-6*

A

B

Perc

ent I

dent

ity

50

100

50

100

50

100

50

100

50

100

50

100

Perc

ent I

dent

ity

50

100

50

100

50

100

50

100

50

100

50

100

50

100

50

100

50

100

50

100

50

100

50

100

50

100

50

100

50

100

50

100

50

100

50

100

FIGURE 2. Th2 cis-regulatory ele-

ments predate the divergence of pla-

cental and nonplacental mammals. A,

Map of the Th2 cytokine gene cluster

with the locations of murine DNase I

hypersensitive sites (vertical arrows). B,

MULAN-generated graphical represen-

tation of DNA sequence alignments be-

tween the indicated species and the

reference human sequence. Selected re-

gions of the Th2 cytokine locus are

shown, with gene annotations and fea-

tures noted above the plots. Shaded re-

gions indicate multispecies ECRs with

gene features specified by color (exon:

blue; untranslated region: yellow; intron:

pink; intergenic: red). Dashed boxes

denote regions that are conserved among

the placental mammals but not in opos-

sum or chicken. P, promoter.

1312 Ets-1 CONTRIBUTES TO Th2 CYTOKINE GENE COREGULATION

by guest on April 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

and IL4IE exhibited intermediate binding ability. Competition bythe IL5 promoter site was more modest but still appreciable. Incontrast, individual ETSmotifs within the IL4 promoter (IL4P.1 andIL4P.2) failed to compete Ets-1 binding.Fig. 4C shows the core motif defined specifically for Ets-1 (27)

and the sequences of the Th2 Ets-1 binding sites, ranked by rel-ative Ets-1 binding ability. Relative to the Ets-1 core, all se-quences containing no or one mismatch supported Ets-1 binding,albeit to different extents. In contrast, two mismatches seemed tocompromise binding, as shown by the poor interactions detectedfor IL5 promoter, IL4P.1, and IL4P.2.Collectively, these data show that the majority of the ETS sites

predicted within the Th2 ECR behave as bona fide Ets-1–bindingmotifs in vitro, suggesting that this protein (and/or possibly othermembers of the ETS family) might bind at multiple locationsthroughout the locus, thereby participating in the concerted reg-ulation of Th2 cytokine gene expression.

Ets-1 binds several regulatory regions in the murine Th2 locusin vivo

To examine whether Ets-1 binds to the endogenous Th2 ECRs,ChIP experiments were performed on chromatin isolated fromnaive CD4+ T cells cultured under Th2-skewing conditions for 7 d.Immunoprecipitation relied upon an Ab specific for Ets-1 ornormal rabbit IgG as a negative control. Real-time PCR wasperformed using primers that target the individual Th2 ECRspreviously tested for Ets-1 binding in vitro. The promoter of Cd14,a gene that is not expressed in murine Th2 cells (data not shown),was amplified for comparison.Fig. 5 shows that, relative to the Cd14 promoter, Ets-1–con-

taining complexes were specifically enriched at all three Th2 cy-tokine gene promoters. The strongest occupancy was detected atthe Il4 promoter, a region that contains two ETS-binding motifs.Notably, the two ETS motifs in this region failed to bind Ets-1when tested individually in vitro, suggesting that the strong signaldetected in vivo reflects cooperative binding. Ets-1 was alsoconsistently found to bind distal Th2 regulatory regions, includingRHS5 of the Th2 LCR, the Il4 silencer region (HSIV), and a Th2cytokine enhancer (CNS2). An Ets-1–specific signal was detected

at IL4IE, but it barely exceeded the one generated for the Cd14promoter. Of note, our ChIP experiments detected Ets-1 and noother members of the large ETS family, because specific enrich-ment was completely abrogated at all of the Th2 ECRs in Ets-1–deficient Th2 cells (data not shown). Overall, these data show thatEts-1 binds to multiple Th2 regulatory regions in vivo, supportingthe possibility that Ets-1 participates in the coordinated regulationof the Th2 cytokine gene cluster.

Ets-1 is required for optimal expression of all three Th2cytokine genes

To test the role of Ets-1 in Th2 cytokine gene expression, we com-pared the levels of Il4, Il13, and Il5mRNA and protein produced byTh2 cells from Ets-1+/2 or Ets-12/2 mice. Naive CD4+ T cellsisolated from these mice were cultured under Th2-skewing con-ditions for 7 d and stimulated for 24 hwith increasing concentrationsof plate-bound anti-CD3mAb, in the presence of a constant amountof anti-CD28Ab.mRNA levelswere assessed by real-timeRT-PCR.Fig. 6A shows that Ets-12/2 Th2 cells exhibited a marked decreasein Il4, Il5, and Il13mRNA relative to Ets-1+/2Th2 cells. In contrast,transcript levels for Il10, a cytokine gene expressed by Th2 cells andlocated on mouse chromosome 1, were comparable in Ets-1+/2 andEts-12/2 Th2 cells. Impaired Th2 cytokine transcription in cellslacking Ets-1 was unlikely to reflect a defect in Th2 cell differen-tiation per se, because Gata3 mRNA levels in these cells were un-affected (Fig. 6B).In parallel, we examined the impact of Ets-1 deficiency on Th2

cytokine secretion. Fig. 6C shows that, consistent with the de-crease observed in Th2 cytokine transcripts, Ets-12/2 Th2 cellswere also impaired in their ability to release IL-4, -13, and -5, butnot IL-10. Collectively, these data demonstrate that Ets-1 acts asa global positive regulator of the expression of the Th2 cytokinegene cluster, thereby providing functional validation for ourphylogenetic analysis.

DiscussionThe Th2 cytokine gene locus typically acts as a unit, the distinctcomponents of which (IL5, IL13, and IL4) are coregulated andcoexpressed in response to T cell activation and differentiation.

LCR

Description IL5P

IL5 (

-6)

RHS5

RHS6.1

RHS6.2

RHS7IL

13P

HSS3CNS-1

IL4P

IL4I

EHSIV

CNS-2

GATA binding factors GATA 1113111 3Human and murine ETS1 factors ETSF 1112111

SOX/SRY and related HMG box factors SORY/HMG 111111

Factors with moderate activity to homeo domain consensus sequence HOXF 1111

Signal transducer and activator of transcription STAT 1221

MAF and AP1 related factors AP1R 111

Nuclear factor of activated T-cells NFAT 111

F1PSCG/1PS srotcaf xob CG 1 1 1

FPBT nietorp gnidnib ATAT 111

cAMP responsive element binding proteins CREB 1 2

EGR /nerve growth factor induced protein C & related factors EGRF 1 1

Ectopic viral integration site EVI 11

DHKF rotcaf niamod daehkroF 11

Growth factor independence-transcriptional repressor GFI1 2 1

CXOHsexelpmoc XBP - XOH 11

Interferon regulatory factors IRFF 11

Myc associated zinc fingers MAZF 1 1

Nuclear factor kappa B/c-rel NFKB 11

RBP-Jkappa/CBF1; Notch pathway RBPF 11

Human acute myelogenous leukemia factors (RUNX) RUNX 11

TALE homeodomain class recognizing TG motifs TALE 1 2Zinc binding protein factors ZBPF 1 1

TF Family

6+

4

3

2

*

FIGURE 3. Common putative transcription factor binding motifs among the Th2 cis-regulatory regions. Shown are those transcription factor (TF)

families that exhibit conserved target sites in more than one Th2 element. Noted within each box is the number of individual conserved sites predicted for

that region. Transcription factors are ranked according to the number of Th2 cis-elements that contain conserved putative binding motifs. The DNA se-

quence of each putative GATA and ETS motif for each species analyzed is provided as supplemental material (Supplemental Fig. 1).

The Journal of Immunology 1313

by guest on April 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

This expression pattern is likely to be critical for effective immuneresponses, but its molecular underpinnings remain only partiallyunderstood. To investigate the molecular basis for coordinatedregulation of Th2 cytokine genes, we combined comparative andfunctional analyses, and we focused on transcription factor bind-ing sites that are highly conserved across placental mammals andcommon to more than one Th2 regulatory region. Our comparisonof the Th2 cytokine gene locus in evolutionarily distant vertebratespecies (human to chicken) revealed that although gene content,order, and orientation have been preserved for.300 million years,nearly all of the distal cis-regulatory elements known to underlieTh2 coregulation arose after the split between birds and mammals.

The main exception seems to be a single region conserved inchicken that corresponds to one of several murine DNase I HSswithin the Th2 LCR (RHS5). It is tempting to speculate that thisregion might participate in transcriptional regulation of thechicken orthologs of mammalian type-2 cytokines.Despite remarkable conservation of the majority of Th2 regu-

latory regions, a few elements were not identifiable in the opossumlocus. These included a GATA-3–responsive element 59 of IL13and the Th1-specific IL4 silencer, HSIV. The emergence of a Th1-selective IL4 silencer suggests that a fine-tuning of the Th1/Th2dichotomy may have occurred during mammalian evolution. Re-sistance to certain mammalian pathogens (e.g., Leishmania spp.)requires extremely polarized Th cell responses (44, 45). Thus,regulatory mechanisms that place increasingly tighter restrictionson alternative cytokine responses might be adaptive to the host.Perhaps most importantly, our identification of highly conserved

Ets-1 binding sites in multiple cis-regulatory regions in the Th2locus, and the finding that the expression of Il5, Il13, and Il4transcripts and protein were substantially and selectively impairedin Ets-1–deficient murine Th2 cells, for the first time define Ets-1as a global regulator of coordinated Th2 cytokine gene expression.Previously, GATA-3, the Th2 master regulator, was the onlytranscription factor known to have a broad influence on Th2 locusregulation, directly controlling chromatin structure and transcrip-tional activity (46). The phenotype of Ets-1 deficiency (globalreduction of Th2 cytokine expression) is less severe than thephenotype of GATA-3 deficiency (complete inhibition of Th2 celldevelopment) (47) but is similar in scope. Of note, GATA-3 andEts-1 were shown to regulate the Il5 promoter cooperatively bybinding to adjacent sites (48–50). This finding raises the possi-bility that the Th2 cytokine defect detected in Ets-1 knockout miceresults from an impairment of the functional synergism betweenGATA-3 and Ets-1. However, although GATA and ETS sites

0

5

10

15

20

25

30

35

40

45

IL5PRHS5 IL4P

IL4IEHSIV

CNS2CD14P

IL13P

Spec

ific

Enr

ichm

ent

FIGURE 5. Ets-1 binds several regulatory regions in the endogenous

murine Th2 locus. Spleens from C57BL/6 mice were harvested, and naive

CD4+ T lymphocytes were isolated and cultured under Th2-skewing

conditions for 7 d. Chromatin was immunoprecipitated with an anti–Ets-1

Ab or normal rabbit IgG as a negative control. The precipitated DNA was

amplified by quantitative real-time PCR using primers specific for the Th2

ECRs or the Cd14 promoter as a comparison. All PCR reactions were

carried out in duplicate. Results are expressed as specific enrichment for

Ets-1 relative to the IgG control. Shown are mean 6 SE values for two

independent experiments, one of which was conducted on chromatin iso-

lated from cells pooled from four mice.

Nil

30X

30X

30X

90X

90X

90X

HSIVEts-1cons. IL13P

1 2 3 4 5 6 7

GGCCAAGCCGGAAGTGAGTGC

GGTAACACAGGAAGTCAGCAG

GTTCGGGgAGGAAGTGGGTAG

TCTGCCACAGGATaTGGGTAG

ATGAAAACAGGAAcTGAAATG

TGGGTCACAGGAAGCCCAAGA

TGTCTTTgAGGAAaTGAATAA

GATTTCACAGGAAcaTTTTAC

TTTTCTCCtGGAAGaGAGGTG

0

20

40

60

80

100

0 30 90Molar Excess Competitor

Perc

ent I

nhib

ition

Ets -1 consensus

Rad HS5

IL13P

HSIV

CNS2

IL4IE

IL5P

IL4P.1

IL4P.2

CCAGGAATG

TC

Ets-1 Core

Ets-1

A

B C

GGCCAAGCCttAAGTGAGTGCEts-1 Mut

Nil

30X

90X

Ets-1cons.

Nil

30X

90X

IL4P.1

8 9 10 11 12 13

FIGURE 4. Ets-1 binds to the majority of ETS

motifs predicted in the Th2 ECR. A, Ets-1 pro-

tein/DNA interactions were characterized using

a g2[32P]-labeled Ets-1 consensus probe. Puri-

fied rhEts-1 (50 ng) was preincubated with an

increasing molar excess of the indicated com-

petitors before addition of a high-affinity Ets-1

consensus probe. B, EMSA blots were analyzed

by densitometry, and the results are expressed as

the percentage inhibition of the Ets-1 band rela-

tive to the uncompeted Ets-1 complex. C, Se-

quences of the Ets-1 binding site core, the high-

affinity Ets-1 consensus probe, and the eight

conserved ETS motifs predicted within the Th2

ECRs. Bold type indicates nucleotide sites within

the Ets-1 core; nucleotides that differ from the

defined core sequence are marked by lower case

letters.

1314 Ets-1 CONTRIBUTES TO Th2 CYTOKINE GENE COREGULATION

by guest on April 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

coexist in all but two of the Ets-1–binding ECRs, these sites are inclose proximity only in the Il5 promoter and in CNS2, the en-hancer located 39 of Il4. This arrangement suggests that alteredcooperativity between Ets-1 and GATA-3 may contribute to, butnot be the only mechanism for, the locus-wide defect in Th2 cy-tokine expression observed in Ets-12/2 mice.The consequences of Ets-1 deficiency are not limited to Th2

cells. Ets-12/2 Th1 cells exhibit a significant defect in the es-tablishment of polarized cytokine expression programs, with de-fective IFN-g production and inappropriate expression of Il4 (51).The latter mirrors the phenotype of CD4+ T cells lacking the Il4silencer, HSIV (19, 51). Interestingly, our bioinformatic analysisidentified a highly conserved ETS binding site within HSIV thatdid support Ets-1 binding in vitro and in vivo. Loss of Ets-1

binding at HSIV may play a role in the abnormal regulation of Il4expression detected in the absence of Ets-1. Overall, our studyhighlighted specific functional elements that Ets proteins maytarget to orchestrate Th2 cytokine expression. Interestingly, ouranalysis also revealed a remarkable heterogeneity in the sequenceof Ets-1–binding motifs, particularly their flanking regions. Thisheterogeneity might result in a functional hierarchy among Ets-1–binding sites, particularly under conditions of limited Ets avail-ability, as well as in differential interaction between Ets-1 and itsprotein partners.Synergism between homotypic factors that occupy distant DNA

elements is a regulatory arrangement that has been appreciated sinceearly studies conducted with phage l (52). The highly conserved,common motifs we identified across the Th2 locus may providea molecular basis for functional cooperation between distant Th2regulatory regions. Indeed, data describing the three-dimensionalarchitecture of the murine Th2 locus highlight the potential for di-rect communication among distant Th2 cis-elements. In naive CD4+

T cells, the Il5, Il13, and Il4 promoters are physically clustered withone another and with distant Th2 cis-regulatory regions (21).Analogous to a single regulatory element containing tandem arraysof identical binding sites, three-dimensional clustering of homo-typic sites may serve as a strategy for the Th2 cytokine genes toeffectively vie for limited amounts of trans-acting factors and, inturn, foster expression of the locus as a whole.Very little is known about how long-range chromatin contacts are

initiated and/or maintained, yet the earliest evidence supportinga DNA looping model pointed toward the homo-oligomerization oftranscriptional activators and/or repressors bound to distant targetsites (52–54). These early studies were conducted on artificialDNA templates, but more recent experiments in a native chro-mosomal context also link sequence-specific DNA-binding pro-teins to chromosomal looping and implicate distant, matchingbinding sites in this process. For example, the erythroid-specificKruppel-like factor directly contributes to the formation of anactive chromatin hub in the b-globin locus by interacting withmultiple cis-regulatory elements distributed across the locus (55).Similarly, androgen receptor-mediated regulation of the prostate-specific Ag gene proceeds via independent recruitment of andro-gen receptors to a distant upstream enhancer and the proximalpromoter, which subsequently form a stable chromosomal loop(56). The distribution of the highly conserved GATA motifs acrossthe Th2 locus seems to reflect the role that GATA-3 plays inhigher-order chromatin architecture, because ectopic expression ofGATA-3 in fibroblasts induces chromatin contacts between theTh2 LCR and the Il4 and Il13 promoters (21). In contrast, theseassociations were relatively weak, suggesting that additional Tcell lineage-restricted factors may be required to reinforce and/ormaintain these long-range interactions (21). Our finding thatfunctional binding sites for Ets-1 are distributed across the Th2locus raises the intriguing possibility that the contribution of Ets-1to the coordinated regulation of Th2 cytokine genes may extend tothe higher-order chromatin architecture of the locus.

DisclosuresThe authors have no financial conflicts of interest.

References1. Stamatoyannopoulos, G. 2005. Control of globin gene expression during de-

velopment and erythroid differentiation. Exp. Hematol. 33: 259–271.2. Lee, G. R., S. T. Kim, C. G. Spilianakis, P. E. Fields, and R. A. Flavell. 2006. T

helper cell differentiation: regulation by cis elements and epigenetics. Immunity24: 369–379.

3. Avery, S., L. Rothwell, W. D. Degen, V. E. Schijns, J. Young, J. Kaufman, andP. Kaiser. 2004. Characterization of the first nonmammalian T2 cytokine gene

0

20

40

60

80

100

0 0.03 0.3 30

50

100

150

200

250

300

0

20

40

60

80

100

0 0.03 0.3 3

ng/m

l

0

20

40

60

80

100

0 0.03 0.3 3 0 0.03 0.3 3

IL-4

IL-5

IL-13

IL-10

anti-CD3 [µg/ml]

+/--/-

Il4

0.00

0.01

0.02

0.03

0.04

0.05

0 0.03 0.3 3

0.00

0.05

0.10

0.15

Il13

0.0

0.1

0.2

0.3

0.4

Il10

0.0

0.1

0.2

Rel

ativ

e E

xpre

ssio

n

0 0.03 0.3 3

Il5

0 0.03 0.3 3

0 0.03 0.3 3

anti-CD3 [µg/ml]

Gata3

Rel

ativ

e E

xpre

ssio

n

anti-CD3 [µg/ml]0 0.03 0.3 3

B

A

+/--/-

+/--/-

C

0.00

0.02

0.04

0.06

FIGURE 6. Ets-1 is required for optimal expression of all three Th2 cy-

tokine genes. Naive CD4+ T cells were isolated from Ets-1+/2 or Ets-12/2

mice and cultured under Th2-skewing conditions in vitro. After 7 d, cells

were stimulated with the indicated concentrations of plate-bound anti-CD3

mAb, in the presence of a constant amount of anti-CD28 mAb (2mg/ml), for

24 h.A andB, mRNA levels were assessed by quantitative real-time RT-PCR

using SYBR green dye detection. All PCR reactions were carried out in

duplicate. mRNA levels are expressed relative toGapdhmRNA abundance.

C, Th2 cell supernatantswere collected and cytokine secretionwasmeasured

by ELISA. All samples were measured in duplicate. The figure shows one

representative experiment of three.

The Journal of Immunology 1315

by guest on April 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

cluster: the cluster contains functional single-copy genes for IL-3, IL-4, IL-13,and GM-CSF, a gene for IL-5 that appears to be a pseudogene, and a gene en-coding another cytokinelike transcript, KK34. J. Interferon Cytokine Res. 24:600–610.

4. Kelly, B. L., and R. M. Locksley. 2000. Coordinate regulation of the IL-4, IL-13,and IL-5 cytokine cluster in Th2 clones revealed by allelic expression patterns. J.Immunol. 165: 2982–2986.

5. Cousins, D. J., T. H. Lee, and D. Z. Staynov. 2002. Cytokine coexpression duringhuman Th1/Th2 cell differentiation: direct evidence for coordinated expressionof Th2 cytokines. J. Immunol. 169: 2498–2506.

6. Anthony, R. M., L. I. Rutitzky, J. F. Urban, Jr., M. J. Stadecker, and W. C. Gause.2007. Protective immune mechanisms in helminth infection. Nat. Rev. Immunol.7: 975–987.

7. Wills-Karp, M., J. Luyimbazi, X. Xu, B. Schofield, T. Y. Neben, C. L. Karp, andD. D. Donaldson. 1998. Interleukin-13: Central mediator of allergic asthma.Science 282: 2258–2261.

8. Shi, H., S. Qin, G. Huang, Y. Chen, C. Xiao, H. Xu, G. Liang, Z. Xie, X. Qin,J. Wu, et al. 1997. Infiltration of eosinophils into the asthmatic airways caused byinterleukin 5. Am. J. Respir. Cell Mol. Biol. 16: 220–224.

9. Takemoto, N., N. Koyano-Nakagawa, T. Yokota, N. Arai, S. Miyatake, and K.-I. Arai. 1998. Th2-specific DNase I-hypersensitive sites in the murine IL-13 andIL-4 intergenic region. Int. Immunol. 10: 1981–1985.

10. Loots, G. G., R. M. Locksley, C. M. Blankenspoor, Z. E. Wang, W. Miller,E. M. Rubin, and K. A. Frazer. 2000. Identification of a coordinate regulator ofinterleukins 4, 13, and 5 by cross-species sequence comparison. Science 288:136–140.

11. Mohrs, M., C. M. Blankespoor, Z. E. Wang, G. G. Loots, V. Afzal, H. Hadeiba,K. Shinkai, E. M. Rubin, and R. M. Locksley. 2001. Deletion of a coordinateregulator of type 2 cytokine expression in mice. Nat. Immunol. 2: 842–847.

12. Lee, G. R., P. E. Fields, and R. A. Flavell. 2001. Regulation of IL-4 gene ex-pression by distal regulatory elements and GATA-3 at the chromatin level. Im-munity 14: 447–459.

13. Henkel, G., D. L. Weiss, R. McCoy, T. Deloughery, D. Tara, and M. A. Brown.1992. A DNase I-hypersensitive site in the second intron of the murine IL-4 genedefines a mast cell-specific enhancer. J. Immunol. 149: 3239–3246.

14. Agarwal, S., and A. Rao. 1998. Modulation of chromatin structure regulatescytokine gene expression during T cell differentiation. Immunity 9: 765–775.

15. Solymar, D. C., S. Agarwal, C. H. Bassing, F. W. Alt, and A. Rao. 2002. A 39enhancer in the IL-4 gene regulates cytokine production by Th2 cells and mastcells. Immunity 17: 41–50.

16. Fields, P. E., G. R. Lee, S. T. Kim, V. V. Bartsevich, and R. A. Flavell. 2004.Th2-specific chromatin remodeling and enhancer activity in the Th2 cytokinelocus control region. Immunity 21: 865–876.

17. Lee, D. U., and A. Rao. 2004. Molecular analysis of a locus control region in theT helper 2 cytokine gene cluster: a target for STAT6 but not GATA3. Proc. Natl.Acad. Sci. USA 101: 16010–16015.

18. Lee, G. R., P. E. Fields, T. J. Griffin, and R. A. Flavell. 2003. Regulation of theTh2 cytokine locus by a locus control region. Immunity 19: 145–153.

19. Ansel, K. M., R. J. Greenwald, S. Agarwal, C. H. Bassing, S. Monticelli,J. Interlandi, I. M. Djuretic, D. U. Lee, A. H. Sharpe, F. W. Alt, and A. Rao.2004. Deletion of a conserved Il4 silencer impairs T helper type 1-mediatedimmunity. Nat. Immunol. 5: 1251–1259.

20. Lee, G. R., C. G. Spilianakis, and R. A. Flavell. 2005. Hypersensitive site 7 ofthe TH2 locus control region is essential for expressing TH2 cytokine genes andfor long-range intrachromosomal interactions. Nat. Immunol. 6: 42–48.

21. Spilianakis, C. G., and R. A. Flavell. 2004. Long-range intrachromosomalinteractions in the T helper type 2 cytokine locus. Nat. Immunol. 5: 1017–1027.

22. Loots, G. G., and I. Ovcharenko. 2004. rVISTA 2.0: evolutionary analysis oftranscription factor binding sites. Nucleic Acids Res. 32: W217–221.

23. Lenhard, B., A. Sandelin, L. Mendoza, P. Engstrom, N. Jareborg, andW. W. Wasserman. 2003. Identification of conserved regulatory elements bycomparative genome analysis. J. Biol. 2: 13.

24. Sumiyama, K., C. B. Kim, and F. H. Ruddle. 2001. An efficient cis-elementdiscovery method using multiple sequence comparisons based on evolutionaryrelationships. Genomics 71: 260–262.

25. Ovcharenko, I., G. G. Loots, B. M. Giardine, M. Hou, J. Ma, R. C. Hardison,L. Stubbs, and W. Miller. 2005. Mulan: multiple-sequence local alignment andvisualization for studying function and evolution. Genome Res. 15: 184–194.

26. Cartharius, K., K. Frech, K. Grote, B. Klocke, M. Haltmeier, A. Klingenhoff,M. Frisch, M. Bayerlein, and T. Werner. 2005. MatInspector and beyond: pro-moter analysis based on transcription factor binding sites. Bioinformatics 21:2933–2942.

27. Nye, J. A., J. M. Petersen, C. V. Gunther, M. D. Jonsen, and B. J. Graves. 1992.Interaction of murine ets-1 with GGA-binding sites establishes the ETS domainas a new DNA-binding motif. Genes Dev. 6: 975–990.

28. Barton, K., N. Muthusamy, C. Fischer, C. N. Ting, T. L. Walunas, L. L. Lanier,and J. M. Leiden. 1998. The Ets-1 transcription factor is required for the de-velopment of natural killer cells in mice. Immunity 9: 555–563.

29. Avni, O., D. Lee, F. Macian, S. J. Szabo, L. H. Glimcher, and A. Rao. 2002. T(H)cell differentiation is accompanied by dynamic changes in histone acetylation ofcytokine genes. Nat. Immunol. 3: 643–651.

30. Wong, E. S., L. J. Young, A. T. Papenfuss, and K. Belov. 2006. In silico iden-tification of opossum cytokine genes suggests the complexity of the marsupialimmune system rivals that of eutherian mammals. Immunome Res. 2: 4.

31. Warren, W. C., L. W. Hillier, J. A. Marshall Graves, E. Birney, C. P. Ponting,F. Grutzner, K. Belov, W. Miller, L. Clarke, A. T. Chinwalla, et al. 2008. Genomeanalysis of the platypus reveals unique signatures of evolution. Nature 453: 175–183.

32. Huising, M. O., C. P. Kruiswijk, and G. Flik. 2006. Phylogeny and evolution ofclass-I helical cytokines. J. Endocrinol. 189: 1–25.

33. Goodman, M. 1999. The genomic record of Humankind’s evolutionary roots.Am. J. Hum. Genet. 64: 31–39.

34. Murphy, W. J., E. Eizirik, W. E. Johnson, Y. P. Zhang, O. A. Ryder, andS. J. O’Brien. 2001. Molecular phylogenetics and the origins of placentalmammals. Nature 409: 614–618.

35. Blair Hedges, S., and S. Kumar. 2003. Genomic clocks and evolutionary time-scales. Trends Genet. 19: 200–206.

36. Yamashita, M., M. Ukai-Tadenuma, M. Kimura, M. Omori, M. Inami,M. Taniguchi, and T. Nakayama. 2002. Identification of a conserved GATA3response element upstream proximal from the interleukin-13 gene locus. J. Biol.Chem. 277: 42399–42408.

37. Mikkelsen, T. S., M. J. Wakefield, B. Aken, C. T. Amemiya, J. L. Chang,S. Duke, M. Garber, A. J. Gentles, L. Goodstadt, A. Heger, et al; Broad InstituteGenome Sequencing Platform Broad Institute Whole Genome Assembly Team.2007. Genome of the marsupial Monodelphis domestica reveals innovation innon-coding sequences. Nature 447: 167–177.

38. Cooper, G. M., M. Brudno, E. D. Green, S. Batzoglou, and A. Sidow; NISCComparative Sequencing Program. 2003. Quantitative estimates of sequencedivergence for comparative analyses of mammalian genomes. Genome Res. 13:813–820.

39. Zheng, W.-P., and R. A. Flavell. 1997. The transcription factor GATA-3 isnecessary and sufficient for Th2 cytokine gene expression in CD4 T cells. Cell89: 587–596.

40. Zhang, D.-H., L. Cohn, P. Ray, K. Bottomly, and A. Ray. 1997. Transcriptionfactor GATA-3 is differentially expressed in murine Th1 and Th2 cells andcontrols Th2-specific expression of the interleukin-5 gene. J. Biol. Chem. 272:21597–21603.

41. Kishikawa, H., J. Sun, A. Choi, S. C. Miaw, and I. C. Ho. 2001. The cell type-specific expression of the murine IL-13 gene is regulated by GATA-3. J. Im-munol. 167: 4414–4420.

42. Lavenu-Bombled, C., C. D. Trainor, I. Makeh, P.-H. Romeo, and I. Max-Audit.2002. Interleukin-13 gene expression is regulated by GATA-3 in T cells: role ofa critical association of a GATA and two GATG motifs. J. Biol. Chem. 277:18313–18321.

43. Hollenhorst, P. C., D. A. Jones, and B. J. Graves. 2004. Expression profiles framethe promoter specificity dilemma of the ETS family of transcription factors.Nucleic Acids Res. 32: 5693–5702.

44. Heinzel, F. P., M. D. Sadick, B. J. Holaday, R. L. Coffman, and R. M. Locksley.1989. Reciprocal expression of interferon g or interleukin 4 during the resolutionor progression of murine leishmaniasis. Evidence for expansion of distinct helperT cell subsets. J. Exp. Med. 169: 59–72.

45. Noben-Trauth, N., P. Kropf, and I. Muller. 1996. Susceptibility to Leishmaniamajor infection in interleukin-4-deficient mice. Science 271: 987–990.

46. Wilson, C. B., E. Rowell, and M. Sekimata. 2009. Epigenetic control of T-helper-cell differentiation. Nat. Rev. Immunol. 9: 91–105.

47. Pai, S. Y., M. L. Truitt, and I. C. Ho. 2004. GATA-3 deficiency abrogates thedevelopment and maintenance of T helper type 2 cells. Proc. Natl. Acad. Sci.USA 101: 1993–1998.

48. Blumenthal, S. G., G. Aichele, T. Wirth, A. P. Czernilofsky, A. Nordheim, andJ. Dittmer. 1999. Regulation of the human interleukin-5 promoter by Ets tran-scription factors. Ets1 and Ets2, but not Elf-1, cooperate with GATA3 and HTLV-I Tax1. J. Biol. Chem. 274: 12910–12916.

49. Wang, J., M. F. Shannon, and I. G. Young. 2006. A role for Ets1, synergizingwith AP-1 and GATA-3 in the regulation of IL-5 transcription in mouse Th2lymphocytes. Int. Immunol. 18: 313–323.

50. Wang, J., and I. G. Young. 2007. Eosinophilic inflammation: mechanisms reg-ulating IL-5 transcription in human T lymphocytes. Allergy 62: 1131–1138.

51. Grenningloh, R., B. Y. Kang, and I.-C. Ho. 2005. Ets-1, a functional cofactor ofT-bet, is essential for Th1 inflammatory responses. J. Exp. Med. 201: 615–626.

52. Ptashne, M. 1992. A Genetic Switch. Blackwell Scientific Publications and CellPress, Cambridge, MA.

53. Su, W., S. Jackson, R. Tjian, and H. Echols. 1991. DNA looping between sitesfor transcriptional activation: self-association of DNA-bound Sp1. Genes Dev. 5:820–826.

54. Knight, J. D., R. Li, and M. Botchan. 1991. The activation domain of the bovinepapillomavirus E2 protein mediates association of DNA-bound dimers to formDNA loops. Proc. Natl. Acad. Sci. USA 88: 3204–3208.

55. Drissen, R., R. J. Palstra, N. Gillemans, E. Splinter, F. Grosveld, S. Philipsen,and W. de Laat. 2004. The active spatial organization of the b-globin locusrequires the transcription factor EKLF. Genes Dev. 18: 2485–2490.

56. Wang, Q., J. S. Carroll, and M. Brown. 2005. Spatial and temporal recruitment ofandrogen receptor and its coactivators involves chromosomal looping andpolymerase tracking. Mol. Cell 19: 631–642.

1316 Ets-1 CONTRIBUTES TO Th2 CYTOKINE GENE COREGULATION

by guest on April 16, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from