Distinct functional motifs within the IL-17 receptorregulate signal transduction and targetgene expressionAmarnath Maitra†, Fang Shen†, Walter Hanel†, Karen Mossman‡, Joel Tocker§, David Swart§, and Sarah L. Gaffen†¶�

Departments of †Oral Biology and ¶Microbiology and Immunology, University at Buffalo, State University of New York, Buffalo, NY 14214; §Departmentof Inflammation Research, Amgen, Inc., Seattle, WA 98119; and ‡Department of Pathology and Molecular Medicine, McMaster University, Hamilton,ON, Canada L8N 3Z5

Edited by Laurie H. Glimcher, Harvard Medical School, Boston, MA, and approved March 19, 2007 (received for review December 27, 2006)

IL-17 is the founding member of a novel family of proinflammatorycytokines that defines a new class of CD4� effector T cells, termed‘‘Th17.’’ Mounting evidence suggests that IL-17 and Th17 cells causepathology in autoimmunity, but little is known about mechanisms ofIL-17RA signaling. IL-17 through its receptor (IL-17RA) activates genestypical of innate immune cytokines, such as TNF� and IL-1�, despiteminimal sequence similarity in their respective receptors. A previousbioinformatics study predicted a subdomain in IL-17-family receptorswith homology to a Toll/IL-1R (TIR) domain, termed the ‘‘SEFIRdomain.’’ However, the SEFIR domain lacks motifs critical for bonafide TIR domains, and its functionality was never verified. Here, weused a reconstitution system in IL-17RA-null fibroblasts to map func-tional domains within IL-17RA. We demonstrate that the SEFIR do-main mediates IL-17RA signaling independently of classic TIR adap-tors, such as MyD88 and TRIF. Moreover, we identified a previouslyundescribed‘‘TIR-like loop’’ (TILL) required for activation of NF-�B,MAPK, and up-regulation of C/EBP� and C/EBP�. Mutagenesis of theTILL domain revealed a site analogous to the LPSd mutation in TLR4,which renders mice insensitive to LPS. However, a putative salt bridgetypically found in TIR domains appears to be dispensable. We furtheridentified a C-terminal domain required for activation of C/EBP� andinduction of a subset IL-17 target genes. This structure-functionanalysis of a IL-17 superfamily receptor reveals important differencesin IL-17RA compared with IL-1/TLR receptors.

BB-loop � cytokine � inflammation � Toll/IL-1R domain �CCAAT/Enhancer binding protein

Interleukin-17 (IL-17A) is the best characterized member of anewly described cytokine family (1). Consisting of six ligands

and five receptors, the IL-17 family shares minimal homologywith other cytokines, and little is known about its mechanisms ofsignaling. IL-17 is produced primarily by T cells, particularlyCD4� but also CD8� and �� populations (2–4). In the classicview of T cell differentiation, T helper cells are divided into Th1and Th2 subsets based on cytokine profiles. A major insight intoIL-17’s place in the immune network was made with the discov-ery that IL-23 promotes IL-17 expression in a unique CD4� Tcell population (5, 6), now termed ‘‘Th17’’ (7). Development ofTh17 cells in mice is driven by TGF� and IL-6 (8–10) via theROR�t transcription factor (11). Th17 cells produce IL-17 aswell as IL-17F, TNF�, IL-6, and IL-22 (12, 13) and regulatevarious aspects of inflammation and autoimmunity.

The discovery of the Th17 subset resolved important ambiguitiesthat were not adequately explained by the Th1/Th2 paradigm. Forexample, several diseases considered to be Th1-dominated none-theless can develop in mice deficient in Th1 cytokines such as IL-12or IFN� (14, 15). Moreover, Th17 cells, IL-17, and/or IL-23 aresufficient to drive pathology in various autoimmune models, par-ticularly rheumatoid arthritis (RA) and EAE (14, 16–19). Indeed,IL-17 is elevated in human RA (20), blocking IL-17 in collagen-induced arthritis (CIA) reduces disease (21), overexpression ofIL-17 causes arthritis in rodents (22), and IL-17KO (23) or ICOSKO

mice (which cannot make IL-17) are also resistant to CIA (24). Inaddition, IL-17 negatively impacts certain infectious conditions,such as Shistosomiasis and Helicobacter pylori infections (25–28).Accordingly, IL-17 is considered an appealing target for newanticytokine therapies (29). In contrast to its pathological proper-ties, IL-17 is essential for mounting effective responses to manyextracellular microbes, largely through regulation of neutrophilmigration and granulopoiesis (30–33).

Surprisingly little is known about the nature of the IL-17Rcomplex or its mechanisms of signaling. The first IL-17 bindingprotein to be identified was IL-17RA, a single transmembranereceptor with an unusually long cytoplasmic tail and almost nohomology to known receptors (34). IL-17RA is ubiquitously ex-pressed at the mRNA level, although surface expression varieswidely (34, 35). Recent studies showed that IL-17RA is part of apreassembled complex that undergoes major conformational alter-ations after ligand binding (36). Moreover, another IL-17R super-family member, IL-17RC, has been implicated in IL-17-mediatedsignal transduction (37). Thus, the IL-17R is a multimeric complexwith dynamic subunit interactions.

The signaling pathways induced by IL-17 are just beginning tobe elucidated. IL-17 induces most of the same genes as IL-1� andTLR ligands (38), and a complex bioinformatics analysis pre-dicted a potential Toll-IL-1 Receptor (TIR)-like signaling do-main in IL-17R family members, termed ‘‘SEFIR’’ (39). TRAF6is necessary for IL-17 target gene expression (40), and two newreports show that the Act1 adaptor is downstream of IL-17 (41,42). Both TRAF6 and Act1 are needed to activate NF-�B, andthe majority of IL-17 target genes are NF-�B-dependent (35). Inaddition, IL-17 activates CCAAT/Enhancer binding proteins(C/EBP), particularly C/EBP� and C/EBP�, which are also vitalfor target gene expression (35, 43). Currently, very little is knownabout how IL-17 connects to C/EBP proteins or whetherC/EBP� and C/EBP� are functionally redundant.

The goal of this study was to define motifs within IL-17RA thatmediate signaling. Using bioinformatics and genetic approaches, wefound that a membrane-proximal region encompassing the SEFIRdomain is necessary for activation of NF-�B, ERK, and C/EBP andsubsequent gene expression. We also identified a critical extensionof this region with homology to a TIR BB-loop that is absolutely

Author contributions: A.M., F.S., and W.H. contributed equally to this work; A.M., F.S.,W.H., and S.L.G. designed research; A.M., F.S., W.H., and S.L.G. performed research; A.M.,F.S., and W.H. analyzed data; K.M., J.T., and D.S. contributed new reagents/analytic tools;and S.L.G. wrote the paper.

Conflict of interest statement: J.T. and D.S. own stock in Amgen.

This article is a PNAS Direct Submission.

Abbreviations: C/EBP, CCAAT/Enhancer binding protein; FL, full length; RA, rheumatoidarthritis; SEFIR, SEF/IL-17R domain; TILL, TIR-like loop; TIR, Toll/IL-1 receptor domain.

�To whom correspondence should be addressed. E-mail: sgaffen@buffalo.edu.

This article contains supporting information online at www.pnas.org/cgi/content/full/0611589104/DC1.

© 2007 by The National Academy of Sciences of the USA

7506–7511 � PNAS � May 1, 2007 � vol. 104 � no. 18 www.pnas.org�cgi�doi�10.1073�pnas.0611589104

Dow

nloa

ded

by g

uest

on

Aug

ust 1

9, 2

020

required for IL-17 function. In addition, a separable signaling motifwas located in the C-terminal region of the IL-17RA cytoplasmictail that contributes to activation of C/EBP� but not C/EBP�, aswell as a subset of IL-17 target genes. This identifies functionalsubdomains in any IL-17R family member and establishes a basis toanalyze this unique receptor family.

ResultsSystem for Evaluating Functional Domains of IL-17RA. IL-17 receptorsare notably distinct from other cytokine receptors in amino acidsequence (1). However, a domain with similarity to the well studiedTIR domain was predicted based on sophisticated biostatisticalmethods, termed a ‘‘SEFIR’’ domain, for SEF (‘‘similar expressionto FGF receptor’’) and IL-17R (39) (Fig. 1A). However, the SEFIRdomain lacks important structural elements found in bona fide TIRdomains, such as the ‘‘BB loop’’ known to be crucial for interactionsbetween TIR-containing molecules (44). To gain insight into struc-ture-function relationships within IL-17RA, we performed scan-ning and alignment of the IL-17RA cytoplasmic tail with IL-1 andTLR family members. We identified a short region downstream andslightly overlapping the SEFIR motif with homology to a BB-loop(Fig. 1B). This domain contains the glutamic acid and arginineresidues that are conserved in all IL-1 and TLR BB domains (blue)(45) as well as a number of additional residues shared with IL-1Rbut not TLR family receptors (green). Finally, there is a valine atposition 553 whose location is homologous to a valine in DrosophilaToll, where mutation destroys function (44). This site also alignswith proline-712 in murine TLR4, where a naturally occurringmutation to histidine blocks TLR4 function in LPS-resistant strainsof mice (44, 46). We term this motif the ‘‘TIR-like loop’’ (TILL),which we hypothesized may be important for mediating IL-17RA-dependent signaling.

To date, structure-function studies of IL-17RA have been tech-

nically challenging because of its ubiquitous expression (34). There-fore, we immortalized tail tip fibroblasts (MFs) from IL-17RAKO

mice. This system is relevant for evaluating IL-17RA function,because cells of mesenchymal origin such as fibroblasts, osteoblasts,and stromal cells are highly responsive to IL-17 (38, 47). To defineregions of IL-17RA important for signaling, we created HA-taggedinternal deletions and point mutations of the SEFIR and TILLdomains [Fig. 1 and supporting information (SI) Fig. 6]. Constructswere transfected into IL-17RAKO MF cells, and lines that expressedhigh, comparable levels of receptor were chosen for analysis (SI Fig.6A). The ability of each mutant to bind IL-17 also was verified (SIFig. 7). In some clones, IL-17RA expression diminished over time,so expression of IL-17RA was monitored for each clone within 2days of all experiments (data not shown).

The SEFIR and TILL Domains Are Necessary for IL-17-DependentSignaling. We and others have identified panels of IL-17 targetgenes in mesenchymal cell types (38, 47). We used representativegenes to monitor signaling by the IL-17RA mutants describedabove, including IL-6, 24p3, and CXCL5 (LIX) and C/EBP�. Thefull length (FL) IL-17RA and the IL-17RA�665 deletion bothinduced strong secretion of IL-6 after IL-17 treatment. Similarresults were seen with IL-17 in combination with suboptimal dosesof TNF� (Fig. 2A). As shown, the dose of TNF� used in theseexperiments did not trigger appreciable IL-6 expression. Cellsexpressing the IL-17RA�527 mutant, which deletes the TILL andpart of the SEFIR, did not produce IL-6 after IL-17 and/or TNF�treatment. Consistently, the IL-17RA�426 mutant was completelydefective in IL-17 signaling to IL-6. Finally, deletion of either theSEFIR or TILL domains individually resulted in a failure torespond to IL-17 (Fig. 2A). A similar pattern was seen for IL-17induction of 24p3, CXCL5, and CXCL1; that is, IL-17RA.FL andIL-17RA�665 induced IL-17-dependent expression, whereas re-ceptors lacking the SEFIR or TILL domains failed to respond toIL-17 (Fig. 2 B and C and data not shown). Also consistent with this,activation of the 24p3 promoter was induced by IL-17RA.FL andIL-17RA�665 but not by deletions lacking SEFIR and/or TILL(data not shown).

Because IL-6, 24p3, CXCL5, and CXCL1 are regulated byNF-�B, we examined the ability of IL-17RA mutants to activateNF-�B by EMSA. As expected, IL-17RA.FL triggered a strongDNA binding activity (Fig. 2D), which was confirmed to be NF-�Bby competition and supershifting (data not shown). The IL-17RA�665 mutant induced NF-�B to a similar magnitude, but theIL-17RA�SEFIR and IL-17RA�TILL mutants failed to induceNF-�B. Thus, the SEFIR and TILL domains mediate NF-�Bactivation and expression of NF-�B-dependent genes.

IL-17 has been reported to activate various MAPK pathways, andAP1 sites are statistically overrepresented in IL-17 target promoters(35, 40, 48–50). Neither p38 nor JNK were activated by IL-17 in thiscell background (data not shown). However, activation of ERK1/2was preserved in the FL and IL-17RA�665 mutant, but not inreceptors with mutations in SEFIR or TILL (Fig. 2E). Therefore,the SEFIR/TILL motif is also upstream of ERK.

C/EBP� Is Activated by Two Distinct Domains Within IL-17RA. C/EBP�and C/EBP� are key transcription factors activated by IL-17, andboth IL-6 and 24p3 contain essential C/EBP sites in their proximalpromoters (35, 43). The mechanism by which IL-17 activates C/EBPis unclear, but expression of C/EBP� and C/EBP� mRNA andprotein are enhanced in response to IL-17 and/or TNF�. Moreover,overexpression of C/EBP� or C/EBP� can partially substitute forIL-17 signaling (38, 43). Here, we show that the TILL and SEFIRdomains are required for up-regulation of C/EBP� (Fig. 3 A and B).C/EBP� mRNA was not substantially enhanced by IL-17 and/orTNF� in IL-17RA.FL cells (data not shown). However, the LAP*isoform of C/EBP� was strongly induced in IL-17RA.FL but notIL-17RAKO, IL-17RA�SEFIR, or IL-17RA�TILL cells. Unex-

Fig. 1. IL-17RA structure. (A) IL-17RA mutants used in this study. HA,hemagglutinin. TM, transmembrane. (B) Identification of a TILL. Alignment ofhuman (hs), mouse (mm) or Drosophila (dm) TLR and IL-1R BB-loops withIL-17RA. Blue indicates amino acids conserved between the TLR/IL-1 familiesand IL-17RA. Red amino acids are conserved in IL-1/TLR but not in IL-17RA.Green residues are conserved in IL-1R and IL-17RA but not TLRs. The site of theLPSd mutation in mouse TLR4 is underlined. Arrows indicate conserved resi-dues that form a salt bridge in BB-loops.

Maitra et al. PNAS � May 1, 2007 � vol. 104 � no. 18 � 7507

IMM

UN

OLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 1

9, 2

020

pectedly, IL-17-mediated induction of C/EBP� was impaired inIL-17RA�665 cells (Fig. 3B), indicating that C/EBP� expressionalso is regulated by a region downstream of residue 665. Consistentwith this observation, IL-17-mediated induction of C/EBP DNAbinding was reduced in cells expressing IL-17RA�665 comparedwith IL-17RA.FL (Fig. 3C). In light of this finding, it was surprisingthat IL-17 induction of 24p3, IL-6, and CXCL5 were not compro-mised in IL-17RA�665-expressing cells (Fig. 2). However, expres-sion of CCL2 and CCL7 were reduced in IL-17RA�665 cells (Fig.3D), suggesting that certain genes may have a stronger dependenceon C/EBP� than others. Collectively, these data indicate that (i)expression of both C/EBP� and C/EBP� is regulated through theSEFIR/TILL domain, (ii) regulation of C/EBP� also involvessignals from a distal region in the IL-17RA tail, and (iii) signalsthrough this distal domain activate some but not all IL-17 targetgenes (SI Fig. 8).

Identification of Essential Amino Acid Residues in the TILL Motif.Some of the conserved residues within the TILL motif correspondto residues in the BB-loop that form a key salt bridge; namely,Glu-533 or Glu-534 with Arg-549. In addition, the TILL containsa conserved valine at position 553 that corresponds to an essentialresidue in Drosophila Toll and murine TLR4 (44, 46). To test the

hypothesis that the TILL domain forms a structure similar to aBB-loop, we mutated the charged residues to Ala or Val-553 to Hisand evaluated the ability to activate IL-17-dependent signaling.IL-17RAKO cells were transiently transfected with IL-17RA.FL orthe mutant receptors and a 24p3 promoter-luciferase reporter (38).Surprisingly, all of the putative salt bridge mutants were fullycompetent to activate the reporter in response to IL-17 and/orTNF� (Fig. 4A). However, the IL-17RA.V553H mutant showedconsiderably impaired signaling activity. Therefore, stable IL-17RAKO cells expressing IL-17RA.V553H were created to deter-mine whether this mutant was defective in other TILL-dependentsignaling events. In three independent lines, IL-17-mediated IL-6expression was blocked (Fig. 4B). Similarly, IL-17-mediated induc-tion of 24p3, CXCL5, C/EBP�, C/EBP�, and NF-�B were impaired(Figs. 2D and 4 C–E). Thus, the TILL domain appears to containsome structural similarities to a genuine TIR domain but also hasimportant disparities that are likely to be the basis for differencesbetween IL-17RA and classic TIR domain-containing receptors.

IL-17RA Does Not Use Canonical TIR Signaling Intermediates. In theIL-1/TLR family, the TIR domain is critical for engaging adaptorsthat lead to activation of TRAF6. There is convincing evidence thatTRAF6 is a key downstream mediator of IL-17-dependent signal

Fig. 2. The SEFIR and TILL domains are required for IL-17 target gene expression and NF-�B activation. IL-17RAKO cells expressing the indicated receptors werestimulated in triplicate with IL-17 (100 ng/ml) and/or TNF� (2 ng/ml) for 4 h. (A) Expression of IL-6 was assessed by ELISA. (B and C) 24p3 and CXCL5 were assessed byreal-timeRT-PCR.Dataarepresentedas fold increaseoveruntreated (normalizedtoGAPDH)foreachcell line. (D) IL-17RAKO cells stablyexpressingthe indicated IL-17RAconstructs were stimulated for 30 min with nothing (‘‘U’’), IL-17, and/or TNF�. Nuclear extracts were subjected to EMSA with an NF-�B probe. (E) The indicated cell lineswere stimulated with IL-17 (100 ng/ml), IL-1� (20 ng/ml), or TNF� (20 ng/ml) and lysates were immunoblotted with Abs to phospho-ERK1/2 (Upper) or total ERK (Lower).

7508 � www.pnas.org�cgi�doi�10.1073�pnas.0611589104 Maitra et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 1

9, 2

020

transduction, and IL-17RA can bind TRAF6 in overexpressionsystems (40). However, IL-17RA does not contain a canonicalTRAF6-binding motif (51) and, thus, is unlikely to bind TRAF6directly. Because TILL resembles the TIR BB-loop, we speculatedthat IL-17 might employ the canonical TLR and IL-1R adaptor,MyD88. However, MyD88-deficient fibroblasts induced potentIL-17-dependent IL-6 expression (Fig. 5). Similarly, TRIF-deficientcells also induced strong IL-17-dependent signaling (Fig. 5). Cou-pled with the recent finding that IL-17 activates Act1 (41, 42), thesedata suggest that immediate proximal mediators of IL-17R signal-ing are distinct from IL-1/TLR family adaptors.

DiscussionCytokines receptors are classified into limited groups based onsequence homology, which generally share common signaling path-ways. For example, Type I and II hematopoietin receptors use theJAK-STAT pathway, whereas TIR-domain receptors typically ac-tivate TRAF6 and NF-�B. Because the IL-17R superfamily sharesonly minimal homology with other receptors, we undertook todelineate regions within IL-17RA critical for signal transduction.We found that the ‘‘SEFIR’’ domain, predicted to be a potentialsignaling motif with distant similarity to TIR domains (39), isindeed required for IL-17-induced signaling. Consistent with this,Qian et al. (42) recently showed that a SEFIR deletion in humanIL-17RA fails to bind Act1, a newly identified IL-17RA signalingmolecule. In addition, we identified a region downstream of the

SEFIR domain termed a TILL that is critically important foractivation of NF-�B and C/EBP transcription factors and subse-quent gene expression. Detailed mutagenesis of TILL indicates thatthis region contains features similar to TIR domains such as a keyresidue at position V553, analogous to Toll and TLR4, but alsoimportant differences, such as the lack of an apparent salt bridge(44). Furthermore, the C-terminal location of the TILL with respectto the SEFIR is different from TIR BB-loops, which link the second�-strand to the second �-helix (45). Therefore, it will be interestingto determine whether the TILL adopts a conformation similar to aBB-loop, or links to an as-yet-unidentified functional domain inIL-17RA. Surprisingly, we also discovered a distal domain in theIL-17RA tail that contributes to activation of C/EBP� and isnecessary for optimal induction of a subset of IL-17 target genes. Todate, we have found no homology of this subdomain to othermolecules.

Molecular studies of IL-17RA have been hindered by its wide-spread expression. In the present study, we developed an unam-biguous system to analyze IL-17RA by reconstituting IL-17RAmutants in fibroblasts from IL-17RAKO mice (30). Fibroblasts andother mesenchymal cells contribute to IL-17-mediated pathology inrheumatoid synovium, probably by expression of inflammatoryeffectors with joint-destructive potential (20, 29). Importantly,IL-17RAKO cells reconstituted with IL-17RA activate the samegenes that are induced by IL-17 in cultured fibroblasts (38, 47).Although the in vivo relevance of many IL-17 target genes has not

Fig. 3. Activation of C/EBP transcription factors requires the SEFIR/TILL domain and a distal domain of the IL-17RA tail. (A) Up-regulation of C/EBP� requiresthe SEFIR/TILL domain. C/EBP� mRNA was assessed by real-time RT-PCR after 4 h of stimulation. Data are presented as fold increase over untreated (normalizedto GAPDH) for each line. (B) Compromised expression of C/EBP� in IL-17RA�665 cells. Expression of C/EBP� and C/EBP� was assessed by Western blotting lysatesfrom the indicated cell lines. (C) Compromised C/EBP� DNA binding in IL-17RA�665 cells. Indicated cells were stimulated for 4 h with IL-17 and/or TNF� and nuclearextracts subjected to EMSA with a probe corresponding to the 24p3 C/EBP site (35). Competition with 50� unlabeled probe confirms specificity (lane 12). (D)Compromised expression of CCL2 and CCL7 in IL-17RA�665 cells. Indicated lines were stimulated with IL-17 and/or TNF� for 4 h, and gene expression was assessedby real-time RT-PCR. Data are presented as fold increase over untreated (normalized to GAPDH) for each line.

Maitra et al. PNAS � May 1, 2007 � vol. 104 � no. 18 � 7509

IMM

UN

OLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 1

9, 2

020

been determined, IL-17-mediated regulation of neutrophil migra-tion via CXC chemokines is well established in infection models (30,52). For example, we found that IL-17RAKO mice are highlysusceptible to periodontal infections; this phenotype is due largelyto a severe reduction in expression of CXCL1 and CXCL5, leadingto a failure of neutrophils to reach the infected gingiva (33).

Although IL-17RA was initially considered to be structurallyunique (34), a bioinformatics study predicted a ‘‘SEFIR’’ regionwith homology to TIR domains (39). However, important elementswere absent in this alignment, such as the BB-loop. We found aregion with homology to a BB-loop within IL-17RA locateddownstream and slightly overlapping the SEFIR domain (Fig. 1),which we termed a TILL. Deletion of either SEFIR or TILLeliminated NF-�B- and C/EBP-dependent signaling as well as ERKactivation (Figs. 2–4). Thus, the SEFIR and TILL domains com-prise an extended functional region in IL-17RA required for

activating downstream events. Although the three-dimensionalstructure of this motif remains unknown, we identified an essentialamino residue, V553. This residue lies in an analogous position toPro-712 in TLR4, which is altered to histidine in LPS-resistantC3H/HeJ mice (46). In TLRs, this residue is not required forstructural integrity (44), but whether this is also true for IL-17receptors remains to be determined. Unlike true TIR domains,there is no evidence for a stabilizing salt bridge in the TILL, becausethe E533A, E534A, and R549A mutants all reconstituted IL-17signaling indistinguishably from wild type (Fig. 4A). Accordingly, itwas not surprising that IL-17 signals independently of MyD88 andTRIF, adaptors that interact with TIR domains (Fig. 5). It wasdiscovered recently that Act1 may be a key link to IL-17R signaling,and Act1 could be coimmunoprecipitated with IL-17RA throughthe Act1 SEFIR-like domain (41). Thus, homotypic SEFIR inter-actions may be analogous to TIR–TIR associations.

Whereas NF-�B is an important IL-17 signaling pathway, C/EBPtranscription factors are also vital for induction of many IL-17target genes (43). Our microarray studies showed that C/EBP� andC/EBP� themselves are induced by IL-17, suggesting a positivefeedback loop whereby up-regulation of C/EBP amplifies expres-sion of IL-6 and other target genes (38, 43). Although induced onlymildly at the mRNA level, protein expression of both C/EBP� andC/EBP� is strongly enhanced by IL-17 (Fig. 3). Reconstitution ofC/EBP��KO cells with C/EBP� or C/EBP� can rescue IL-17-dependent IL-6 secretion and 24p3 promoter activity (53), suggest-ing that there is considerable redundancy in C/EBP function.However, the present study shows that regulation of C/EBP� andC/EBP� by IL-17 occurs by separable pathways, because theIL-17RA�665 mutant induces normal expression of C/EBP� butnot C/EBP� (Fig. 3). Moreover, IL-17RA�665 activates someIL-17-dependent genes normally (IL-6, CXCL5, and 24p3), but notothers (CCL2 and CCL7). Therefore, IL-17 uses C/EBP� andC/EBP� differentially to control gene expression, and these factorsare clearly not entirely redundant.

Regulation of C/EBP proteins is highly complex. C/EBP� iscontrolled primarily by expression rather than by posttranslationalmodifications. At the transcriptional level, both 5� and 3� promoter/enhancer regions have been described (54, 55), although themechanism by which IL-17 regulates C/EBP� is undefined. Ourdata indicate that C/EBP� expression is downstream of SEFIR/TILL, perhaps mediated by NF-�B or ERK (Fig. 3). In contrast,C/EBP� is largely controlled posttranslationally. C/EBP� exists inmultiple isoforms (56), and IL-17 seems to preferentially induce thelargest of these, LAP* (Fig. 3). Moreover, C/EBP� is phosphory-lated on several sites, and numerous serine-threonine kinases have

Fig. 4. A point mutation in TILL renders IL-17RA nonfunctional. (A) Valine-553 is an essential residue in the TILL domain. IL-17RAKO cells were transientlytransfected in triplicate with a vector control, IL-17RA.FL, or the indicatedmutations together with a 24p3-promoter-luciferase reporter. Cells werestimulated with IL-17 and/or TNF� for 6h and normalized to Renilla-Luciferase.(B–E) Valine-553 in the TILL domain is essential for IL-17 target gene expres-sion. The indicated cell lines were incubated with IL-17 and/or TNF� for 24 h,and IL-6 was assessed in triplicate by ELISA (A). 24p3, CXCL5, and C/EBP� wereassessed by real-time RT-PCR (C–E). Data are presented as fold increase overuntreated (normalized to GAPDH) for each line.

Fig. 5. IL-17 signals independently of MyD88 and TRIF. Murine embryonicfibroblast cells from MyD88KO or TRIFKO mice (61, 62) were treated with IL-17and/or TNF�, poly I:C (12.5 �g/ml), or LPS (10 ng/ml) for 6 or 24 h, and IL-6 wasassessed in triplicate.

7510 � www.pnas.org�cgi�doi�10.1073�pnas.0611589104 Maitra et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 1

9, 2

020

been implicated in its regulation (55–58). Determining the targetsites on C/EBP� required for IL-17-mediated activation and theupstream signals responsible will be essential for fully understand-ing the IL-17 signaling cascade.

Mapping functional domains in cytokine receptors has beenenlightening in terms of understanding signal transduction. In theIL-2R, different tyrosine residues preferentially mediate STAT5versus phosphatidylinositol 3-kinase (59). Similarly, separable re-gions within gp130 are required for STAT activation versus MAPKactivation and mediate different activities in vivo (60). The presentreport reveals two separable motifs involved in IL-17RA signaltransduction (SI Fig. 8). The biological significance of these sig-naling pathways is still not fully known, but this work sets the stagefor probing IL-17R superfamily members functions in vivo.

Materials and MethodsCell Culture and Luciferase Assays. IL-17RAKO fibroblasts fromIL-17RAKO mice (30, 36) were immortalized with the SV40 Tantigen. Murine embryonic fibroblasts were obtained fromTRIF�/� (61) and MyD88�/� mice (62). All cells were cultured in�-MEM (Sigma, St. Louis, MO), with 10% FBS and antibiotics(Invitrogen, Carlsbad, CA). Cells were transfected with FuGENE6 (Roche, Indianapolis, IN) and stable lines were selected in Zeocin(Invitrogen). Cytokines were from R&D Systems (Minneapolis,MN). LPS and Poly I:C were from Sigma. Luciferase assays wereperformed as described (38).

Plasmids. IL-17RA was generated by RT-PCR from HT-2 cells.Mutations were made by PCR and subcloned in pcDNA3.1-Zeocin(Invitrogen). Point mutations were made with the QuikChangemutagenesis kit (Stratagene, La Jolla, CA) and confirmed bysequencing.

ELISA and Flow Cytometry. IL-6 ELISAs were performed withcommercial kits (eBioscience, San Diego, CA). For FACS, 106 cellswere stained with mAbs M177, M750, or M751, followed byanti-rat-PE (BD PharMingen, San Jose, CA). For binding studies,a cDNA encoding huIL17 fused to huFc (IgG1) was expressed inCOS cells and purified over Protein A columns. Cells were stainedwith huIL17.Fc, followed by anti-huIgG-APC (Jackson Immunore-search, West Grove, PA). Data were analyzed on a FACSCaliburwith Cell Quest software (BD Biosciences, San Jose, CA).

Western Blotting, Immunoprecipitation, and EMSA. EMSAs wereperformed as described (35). Antibodies to C/EBP and histonewere from Santa Cruz Biotechnology (Santa Cruz, CA), HA Abswere from Roche, and ERK Abs were from Cell Signaling Tech-nology (Beverly, MA).

We thank J. Hummel for preparation of murine embryonic fibroblasts andS. Akira (Osaka University, Osaka, Japan) for MyD88�/� and TRIF�/�

mice. S.L.G. was supported by the Arthritis Foundation and NationalInstitutes of Health Grants AR050458 and AI43929. K.M. was supported bythe Canadian Institutes of Health Research.

1. Aggarwal S, Gurney AL (2002) J Leukoc Biol 71:1–8.2. Fossiez F, Djossou O, Chomarat P, Flores-Romo L, Ait-Yahia S, Maat C, Pin, J-J,

Garrone P, Garcia E, Saeland S, et al. (1996) J Exp Med 183:2593–2603.3. Liu XK, Clements JL, Gaffen SL (2005) Mol Cells 20:329–337.4. Lockhart E, Green AM, Flynn J (2006) J Immunol 177:4662–4669.5. Aggarwal S, Ghilardi N, Xie MH, De Sauvage FJ, Gurney AL (2002) J Biol Chem

3:1910–1914.6. Happel KI, Zheng M, Young E, Quinton LJ, Lockhart E, Ramsay AJ, Shellito

JE, Schurr JR, Bagby GJ, Nelson S, et al. (2003) J Immunol 170:4432–4436.7. Weaver CT, Harrington LE, Mangan PR, Gavrieli M, Murphy KM (2006)

Immunity 24:677–688.8. Veldhoen M, Hocking RJ, Atkins CJ, Locksley RM, Stockinger B (2006)

Immunity 24:179–189.9. Mangan PR, Harrington LE, O’Quinn DB, Helms WS, Bullard DC, Elson CO,

Hatton RD, Wahl SM, Schoeb TR, Weaver CT (2006) Nature 441:231–234.10. Bettelli E, Carrier Y, Gao W, Korn T, Strom TB, Oukka M, Weiner HL, Kuchroo

VK (2006) Nature 441:235–238.11. Ivanov I, McKenzie B, Zhou L, Tadokoro C, Lepelley A, Lafaille J, Cua D,

Littman D (2006) Cell 126:1121–1133.12. Tato CM, Laurence A, O’Shea JJ (2006) J Exp Med 203:809–812.13. Liang SC, Tan XY, Luxenberg DP, Karim R, Dunussi-Joannopoulos K, Collins

M, Fouser LA (2006) J Exp Med 203:2271–2279.14. Cua DJ, Kastelein RA (2006) Nat Immunol 7:557–559.15. Gor DO, Rose NR, Greenspan NS (2003) Nat Immunol 4:503–505.16. Lubberts E, Koenders MI, van den Berg WB (2005) Arthritis Res Ther 7:29–37.17. Langrish CL, Chen Y, Blumenschein WM, Mattson J, Basham B, Sedgwick JD,

McClanahan T, Kastelein RA, Cua DJ (2005) J Exp Med 201:233–240.18. Sato K, Suematsu A, Okamoto K, Yamaguchi A, Morishita Y, Kadono Y, Tanaka

S, Kodama T, Akira S, Iwakura Y, et al. (2006) J Exp Med 203:2673–2682.19. Lubberts E, van den Bersselaar L, Oppers-Walgreen B, Schwarzenberger P,

Coenen-de Roo CJ, Kolls JK, Joosten LA, van den Berg WB (2003) J Immunol170:2655–2662.

20. Kotake S, Udagawa N, Takahashi N, Matsuzaki K, Itoh K, Ishiyama S, Saito S,Inoue K, Kamatani N, Gillespie MT, et al. (1999) J Clin Invest 103:1345–1352.

21. Lubberts E, Koenders MI, Oppers-Walgreen B, van den Bersselaar L, Coenen-deRoo CJ, Joosten LA, van den Berg WB (2004) Arthritis Rheum 50:650–659.

22. Lubberts E, Joosten LA, van de Loo FA, Schwarzenberger P, Kolls J, van denBerg WB (2002) Inflamm Res 51:102–104.

23. Nakae S, Nambu A, Sudo K, Iwakura Y (2003) J Immunol 171:6173–6177.24. Nurieva RI, Treuting P, Duong J, Flavell RA, Dong C (2003) J Clin Invest

111:701–706.25. Luzza F, Parrello T, Monteleone G, Sebkova L, Romano M, Zarrilli R, Imeneo

M, Pallone F (2000) J Immunol 165:5332–5337.26. Kullberg MC, Jankovic D, Feng CG, Hue S, Gorelick PL, McKenzie BS, Cua DJ,

Powrie F, Cheever AW, Maloy KJ, et al. (2006) J Exp Med 203:2485–2494.27. Hue S, Ahern P, Buonocore S, Kullberg M, Cua D, McKenzie B, Powrie F, Malloy

K (2006) J Exp Med 203:2473–2483.28. Rutitzky LI, Lopes da Rosa JR, Stadecker MJ (2005) J Immunol 175:3920–3926.29. Lubberts E (2003) Curr Opin Investig Drugs 4:572–577.30. Ye P, Rodriguez FH, Kanaly S, Stocking KL, Schurr J, Schwarzenberger P, Oliver

P, Huang W, Zhang P, Zhang J, et al. (2001) J Exp Med 194:519–527.

31. Huang W, Na L, Fidel PL, Schwarzenberger P (2004) J Infect Dis 190:624–631.32. Kelly MN, Kolls JK, Happel K, Schwartzman JD, Schwarzenberger P, Combe C,

Moretto M, Khan IA (2005) Infect Immun 73:617–621.33. Yu J, Ruddy M, Wong G, Sfintescu C, Baker P, Smith J, Evans R, Gaffen S

(January 3, 2007) Blood, 10.1182/blood-2005–09-010116.34. Yao Z, Fanslow WC, Seldin MF, Rousseau, A-M, Painter SL, Comeau MR,

Cohen JI, Spriggs MK (1995) Immunity 3:811–821.35. Shen F, Hu Z, Goswami J, Gaffen SL (2006) J Biol Chem 281:24138–24148.36. Kramer J, Yi L, Shen F, Maitra A, Jiao X, Jin T, Gaffen S (2006) J Immunol

176:711–715.37. Toy D, Kugler D, Wolfson M, Vanden Bos T, Gurgel J, Derry J, Tocker J,

Peschon JJ (2006) J Immunol 177:36–39.38. Shen F, Ruddy MJ, Plamondon P, Gaffen SL (2005) J Leukoc Biol 77:388–399.39. Novatchkova M, Leibbrandt A, Werzowa J, Neubuser A, Eisenhaber F (2003)

Trends Biochem Sci 28:226–229.40. Schwandner R, Yamaguchi K, Cao Z (2000) J Exp Med 191:1233–1239.41. Chang SH, Park H, Dong C (2006) J Biol Chem 281:35603–35607.42. Qian Y, Liu C, Hartupee J, Altuntas CZ, Gulen MF, Jane-Wit D, Xiao J, Lu Y,

Giltiay N, Liu J, et al. (2007) Nat Immunol 8:247–256.43. Ruddy MJ, Wong GC, Liu XK, Yamamoto H, Kasayama S, Kirkwood KL, Gaffen

SL (2004) J Biol Chem 279:2559–2567.44. Xu Y, Tao X, Shen B, Horng T, Medzhitov R, Manley J, Tong L (2000) Nature

408:111–115.45. Rock F, Hardiman G, Timans J, Kastelein R, Bazan J (1998) Proc Natl Acad Sci

USA 95:588–593.46. Poltorak A, He X, Smirnova I, Liu MY, Van Huffel C, Du X, Birdwell D, Alejos

E, Silva M, Galanos C, et al. (1998) Science 282:2085–2088.47. Park H, Li Z, Yang XO, Chang SH, Nurieva R, Wang YH, Wang Y, Hood L, Zhu

Z, Tian Q, et al. (2005) Nat Immunol 6:1133–1141.48. Jovanovic D, di Battista JA, Martel-Pelletier J, Jolicoeur FC, He Y, Zhang M,

Mineau F, Pelletier, J-P (1998) J Immunol 160:3513–3521.49. Faour WH, Mancini A, He QW, Di Battista JA (2003) J Biol Chem 278:26897–26907.50. Granet C, Miossec P (2004) Cytokine 26:169–177.51. Pullen SS, Dang TT, Crute JJ, Kehry MR (1999) J Biol Chem 274:14246–14254.52. Linden A, Laan M, Anderson G (2005) Eur Respir J 25:159–172.53. Ruddy MJ, Shen F, Smith J, Sharma A, Gaffen SL (2004) J Leukoc Biol

76:135–144.54. Cantwell CA, Sterneck E, Johnson PF (1998) Mol Cell Biol 18:2108–2117.55. Yamada T, Tsuchiya T, Osada S, Nishihara T, Imagawa M (1998) Biochem

Biophys Res Commun 242:88–92.56. Ramji DP, Foka P (2002) Biochem J 365:561–575.57. Tang QQ, Gronborg M, Huang H, Kim JW, Otto TC, Pandey A, Lane MD (2005)

Proc Natl Acad Sci USA 102:9766–9771.58. Kalvakonalu D, Roy S (2005) J Interferon Cytokine Res 25:757–769.59. Gaffen SL (2001) Cytokine 14:63–77.60. Sims N, Jenkins B, Quinn J, Nakamura A, Glatt M, Gillespie M, Ernst M, Martin

T (2004) J Clin Invest 113:379–389.61. Yamamoto M, Sato S, Hemmi H, Hoshino K, Kaisho T, Sanjo H, Takeuchi O,

Sugiyama M, Okabe M, Takeda K, et al. (2003) Science 301:640–643.62. Adachi O, Kawai T, Takeda K, Matsumoto M, Tsutsui H, Sakagami M, Nakanishi

K, Akira S (1998) Immunity 9:143–150.

Maitra et al. PNAS � May 1, 2007 � vol. 104 � no. 18 � 7511

IMM

UN

OLO

GY

Dow

nloa

ded

by g

uest

on

Aug

ust 1

9, 2

020