doi:10.1152/ajpheart.00928.2007 293:3356-3365, 2007. First published Oct 5, 2007;Am J Physiol Heart Circ Physiol

Bjorn Steffensen, Matthew Vincenti, Jeffrey L. Barnes and Bysani Chandrasekar Dolores M. Cortez, Marc D. Feldman, Srinivas Mummidi, Anthony J. Valente,

You might find this additional information useful...

34 articles, 13 of which you can access free at: This article cites http://ajpheart.physiology.org/cgi/content/full/293/6/H3356#BIBL

3 other HighWire hosted articles: This article has been cited by

  [PDF]  [Full Text]  [Abstract]

, May 1, 2008; 294 (5): H2078-H2087. Am J Physiol Heart Circ Physiol

K. Venkatachalam, S. Mummidi, D. M. Cortez, S. D. Prabhu, A. J. Valente and B. Chandrasekar expression in primary mouse cardiac fibroblasts

Resveratrol inhibits high glucose-induced PI3K/Akt/ERK-dependent interleukin-17 

[PDF]  [Full Text]  [Abstract], January 1, 2009; 40 (1): 4-12. Am. J. Respir. Cell Mol. Biol.

B. A. Mercer, A. M. Wallace, C. E. Brinckerhoff and J. M. D'Armiento Identification of a Cigarette Smoke-Responsive Region in the Distal MMP-1 Promoter

  [PDF]  [Full Text]  [Abstract]

, February 24, 2009; 2 (59): ra8-ra8. Sci. Signal.and S. L. Gaffen F. Shen, N. Li, P. Gade, D. V. Kalvakolanu, T. Weibley, B. Doble, J. R. Woodgett, T. D. Wood

Regulatory 2 DomainIL-17 Receptor Signaling Inhibits C/EBP{beta} by Sequential Phosphorylation of the

including high-resolution figures, can be found at: Updated information and services http://ajpheart.physiology.org/cgi/content/full/293/6/H3356

can be found at: AJP - Heart and Circulatory Physiologyabout Additional material and information http://www.the-aps.org/publications/ajpheart

This information is current as of March 11, 2009 .  

http://www.the-aps.org/.ISSN: 0363-6135, ESSN: 1522-1539. Visit our website at Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2005 by the American Physiological Society. intact animal to the cellular, subcellular, and molecular levels. It is published 12 times a year (monthly) by the Americanlymphatics, including experimental and theoretical studies of cardiovascular function at all levels of organization ranging from the

publishes original investigations on the physiology of the heart, blood vessels, andAJP - Heart and Circulatory Physiology

on March 11, 2009

ajpheart.physiology.orgD

ownloaded from

IL-17 stimulates MMP-1 expression in primary human cardiac fibroblasts viap38 MAPK- and ERK1/2-dependent C/EBP-�, NF-�B, and AP-1 activation

Dolores M. Cortez,1 Marc D. Feldman,1,2 Srinivas Mummidi,1,2 Anthony J. Valente,1 Bjorn Steffensen,3

Matthew Vincenti,4 Jeffrey L. Barnes,1,2 and Bysani Chandrasekar1,2

1Department of Veterans Affairs South Texas Veterans Health Care System, and Departments of 2Medicine and 3Periodonticsand Biochemistry, University of Texas Health Science Center, San Antonio, Texas; and 4Department of Medicine, DartmouthMedical School, Lebanon, New Hampshire

Submitted 10 August 2007; accepted in final form 4 October 2007

Cortez DM, Feldman MD, Mummidi S, Valente AJ, SteffensenB, Vincenti M, Barnes JL, Chandrasekar B. IL-17 stimulatesMMP-1 expression in primary human cardiac fibroblasts via p38MAPK- and ERK1/2-dependent C/EBP-�, NF-�B, and AP-1 activa-tion. Am J Physiol Heart Circ Physiol 293: H3356–H3365, 2007. Firstpublished October 5, 2007; doi:10.1152/ajpheart.00928.2007.—Matrix metalloproteinases (MMPs) degrade collagen and mediatetissue remodeling. The novel cytokine IL-17 is expressed duringvarious inflammatory conditions and modulates MMP expression. Weinvestigated the effect of IL-17 on MMP-1 expression in primaryhuman cardiac fibroblasts (HCF) and delineated the signaling path-ways involved. HCF were treated with recombinant human IL-17.MMP-1 expression was analyzed by Northern blotting, RT-quantitative PCR, Western blotting, and ELISA; transcriptional in-duction and transcription factor binding by EMSA, ELISA, andreporter assay; and p38 MAPK and ERK1/2 activation by proteinkinase assays and Western blotting. Signal transduction pathwayswere investigated using pharmacological inhibitors, small interferingRNA (siRNA), and adenoviral dominant-negative expression vectors.IL-17 stimulated MMP-1 gene transcription, net mRNA levels, pro-tein, and promoter-reporter activity in HCF. This response wasblocked by IL-17 receptor-Fc chimera and IL-17 receptor antibodies,but not by IL-6, TNF-�, or IL-1� antibodies. IL-17-stimulated type Icollagenase activity was inhibited by the MMP inhibitor GM-6001and by siRNA-mediated MMP-1 knockdown. IL-17 stimulated acti-vator protein-1 [AP-1 (c-Fos, c-Jun, and Fra-1)], NF-�B (p50 andp65), and CCAAT enhancer-binding protein (C/EBP)-� DNA bindingand reporter gene activities, effects attenuated by antisense oligonu-cleotides, siRNA-mediated knockdown, or expression of dominant-negative signaling proteins. Inhibition of AP-1, NF-�B, or C/EBPactivation attenuated IL-17-stimulated MMP-1 expression. IL-17 in-duced p38 MAPK and ERK1/2 activation, and inhibition by SB-203580 and PD-98059 blunted IL-17-mediated transcription factoractivation and MMP-1 expression. Our data indicate that IL-17induces MMP-1 in human cardiac fibroblasts directly via p38 MAPK-and ERK-dependent AP-1, NF-�B, and C/EBP-� activation andsuggest that IL-17 may play a critical role in myocardial remodeling.

cytokines; interleukins; matrix metalloproteinases; fibrosis

EXTRACELLULAR MATRIX (ECM) turnover in the normal heart is atightly regulated process. The alteration in the delicate balancebetween matrix metalloproteinases (MMPs) and their tissueinhibitors (TIMPs) during myocardial injury and inflammationresults in enhanced ECM degradation and remodeling (29, 30).MMPs belong to a family of related, but structurally distinct,zinc-dependent proteases that degrade various ECM proteins,

including collagens, gelatins, fibronectin, and laminins (14).MMP-1 (EC 3.4.24.7), or collagenase type I, is the firstidentified metalloproteinase that degrades interstitial collagens(collagens I, II, III, and VII) in the myocardium (14).

Sustained production of inflammatory cytokines plays acentral role in the initiation and progression of left ventricularhypertrophy to failure (11). Various cytokines have beenshown to regulate MMP-1 expression at transcriptional andposttranscriptional levels (5, 23). IL-17, a recently discoveredfamily of proinflammatory cytokines secreted mainly by asubset of T (Th17) cells, consists of six ligands (IL-17A, B, C,D, E, and F) that signal through five receptors (IL-17RA, B, C,D, and E) (4). IL-17 family members show little to no homol-ogy with other ILs and, therefore, constitute a family of theirown (4). Enhanced expression of IL-17 has been reported invarious models of inflammation, including rheumatoid arthri-tis, periodontitis, asthma, and organ rejection (4), and a causalrole for IL-17 has been demonstrated in experimental autoim-mune myocarditis (21, 28). However, a role for IL-17 inmyocardial ischemic injury, hypertrophy, and remodeling hasnot been described. Since remodeling is characterized byhypertrophy and fibrosis and since fibroblasts play a criticalrole in fibrosis through expression of MMPs, we investigatedwhether IL-17 regulates MMP-1 expression in primary humancardiac fibroblasts (HCF).

MATERIALS AND METHODS

Materials. Recombinant human IL-6 (catalog no. 206-IL-010) andIL-17 (catalog no. 317-IL-050), neutralizing antibodies against IL-6,IL-1�, and TNF-�, and normal goat IgG (Ab 108-C) were purchasedfrom R & D Systems (Minneapolis, MN). We previously reported theeffectiveness of the anti-cytokine neutralizing antibodies (6, 19).Anti-p38, phosphorylated p38 [PhosphoPlus p38 MAP kinase(Thr180/Tyr182) antibody kit], ERK1/2 (catalog no. 9102), phosphor-ylated ERK1/2 (catalog no. 9101S), and anti-phosphorylated CCAATenhancer-binding protein (C/EBP)-� (catalog no. 3084S) antibodieswere obtained from Cell Signaling Technology (Beverly, MA).Cycloheximide [InSolution cycloheximide (CHX)], a protein synthe-sis inhibitor (10 �g/ml in DMSO); SB-203580, a p38 MAPK inhibitor(1 �M in DMSO for 30 min); PD-98059, an ERK1/2 inhibitor (10 �Min DMSO for 1 h); and DMSO were purchased from EMD Bio-sciences (San Diego, CA). GM-6001, a nonspecific hydroxamic acid-based MMP inhibitor with potent inhibitory activity against collage-nase, gelatinases, and stromelysin (15) (10 �M in DMSO for 15 min),was purchased from Upstate/Chemicon (Temecula, CA). Actinomy-

Address for reprint requests and other correspondence: B. Chandrasekar,Medicine/Cardiology, The Univ. of Texas Health Science Center, 7703 FloydCurl Dr., San Antonio, TX 78229-3900 (e-mail: chandraseka@uthscsa.edu).

The costs of publication of this article were defrayed in part by the paymentof page charges. The article must therefore be hereby marked “advertisement”in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Am J Physiol Heart Circ Physiol 293: H3356–H3365, 2007.First published October 5, 2007; doi:10.1152/ajpheart.00928.2007.

http://www.ajpheart.orgH3356

on March 11, 2009

ajpheart.physiology.orgD

ownloaded from

cin D (ActD), an RNA synthesis inhibitor (2.5 �g/ml in DMSO);�-tubulin polyclonal antibodies; and all other chemicals were pur-chased from Sigma-Aldrich (St. Louis, MO).

Cell culture. HCF (catalog no. 6300, ScienCell Research Labora-tories, San Diego, CA) were characterized by an immunofluorescentmethod using antibody to fibronectin (manufacturer’s technical datasheet). HCF were grown in fibroblast medium (FM) supplied by themanufacturer and supplemented with 2% FBS, fibroblast growthsupplement, and antibiotics (complete medium). At 70% confluency,the complete medium was replaced with FM containing 0.5% BSA.After overnight incubation (quiescent cells), IL-17 was added, and thecells were cultured for the indicated time periods. At the end of theexperimental period, culture supernatants were collected and snapfrozen. Cells were harvested, snap frozen, and stored at �80°C.

Since IL-17 is a proinflammatory cytokine and induces the expres-sion of other cytokines (27) that are known to stimulate MMP-1expression (5, 23), we determined whether IL-17-mediated MMP-1expression is dependent on IL-1�, IL-6, or TNF-�. HCF were incu-bated with IL-1�, IL-6, or TNF-� neutralizing antibodies (10 �g/mlfor 1 h; R & D Systems) before addition of IL-17. Normal goat/mouseIgG served as a control.

Adenoviral vectors and RNA interference. Recombinant, replica-tion-deficient adenoviral vectors encoding green fluorescent protein(Ad-CMV-GFP), dominant-negative (dn) IKK-�, dnp65, anddnI�B-� (S32A/S36A) have been previously described (18). Ad-CMV-dnc-Jun was purchased from Vector Biolabs. Cells were in-fected at ambient temperature with adenoviruses in PBS at a multi-plicity of infection (MOI) of 100. After 1 h, the PBS solutioncontaining adenovirus was replaced with FM containing 0.5% BSA.Assays were carried out 48 h later, and knockdown of proteins wasconfirmed by Western blotting. C/EBP-� small interfering RNA(siRNA; identification no. 114496, catalog no. 16708) was purchasedfrom Ambion (Austin, TX). Cells at 70% confluency were transfectedwith siRNA (100 nM) using the Nucleofection kit (catalog no.VPI-1002) provided by Amaxa (Gaithersburg, MD). Among the fiverecommended programs, the T-16 program gave optimal transfectionefficiency (38%) with only 9% cell death. After overnight culture inmedium containing 0.5% BSA, dead cells were removed. Controlnegative siRNA (sense, 5�-CUC GGC GUU UCA UCU GUGGdTdT) served as a control.

MMP-1 promoter-reporter assays. A 4,386-bp fragment (�4334/�52) of the 5�-flanking region of the MMP1 gene was amplified fromhuman genomic DNA (catalog no. G3041, Promega) using the fol-lowing primers: 5�-acg cgt AGA TGT AAG AGC TGG AAA GGACGG-3� (sense) and 5�-ctc gag TCA GTG CAA GGT AAG TGATGG CTT C-3� (antisense). The sense primer contained an Mlu Irestriction site at the 5� end, and the antisense primer contained an XhoI restriction site (lower case). The PCR product was cloned intopCR2.1-TOPO and subcloned into the pGL3-basic reporter vector inthe same restriction sites. The identity of the PCR product wasconfirmed by sequencing on both strands. To determine the roleof C/EBP, NF-�B, and AP-1, we generated deletion constructs lackingC/EBP, NF-�B, or AP-1 (�2685/�52, 5�-acg cgt AGA TGC TCCCAG AGG AAA C-3� for C/EBP; �1524/�52, 5�-acg cgt CAG GAATCC ATA AGG GGA GG�3� for C/EBP and NF-�B; and �62/�52,acg cgt ACC TCT GGC TTT CTG GAA GG-3� for C/EBP, NF-�B,and AP-1) and the antisense primer described above.

mRNA expression: Northern blotting and real-time quantitativePCR. DNA-free total RNA was extracted using the RNAqueous-4PCR kit (Ambion). RNA quality was assessed by capillary elec-trophoresis using the Agilent 2100 Bioanalyzer (Agilent Technolo-gies, Palo Alto, CA). All RNA samples used for quantitative PCR hadRNA integrity numbers �9.1 (on a scale of 1–10), as assigned bydefault parameters of the Expert 2100 Bioanalyzer software package(version 2.02). IL-17 receptor (IL-17R) type A (GenBank accessionno. NM_014339) expression was analyzed by RT-PCR using two setsof primers [5�-GAT GAC AGC TGG ATT CAC C-3� (sense) and

5�-CTC ATA TTC CTG GTC AGG G-3� (antisense) for set 1 and5�-GTC TGG TTA TCG TCT ATC C-3� (sense) and 5�-CAA ACTCCT GAC CTC AGA G-3� (antisense) for set 2], which resulted in a300-bp amplification product. �-Actin [GenBank accession no.NM_001101; 668-bp amplification product, 5�-CGT GCG TGA CATTAA GGA GA-3� (sense) and 5�-CAC CTT CAC CGT TCC AGTTT-3� (antisense)] served as an internal control. Northern blot analysiswas carried out as previously described (6). MMP-1 cDNA (GenBankaccession no. NM_002421.2) was amplified from reverse-transcribedHCF RNA by RT-PCR using the sense primer 5�-ATT CTA CTGATA TCG GGG CTT TGA-3� and the antisense primer 5�-ATG TCCTTG GGG TAT CCG TGT AG-3�. Expression of 28S rRNA was usedas an internal control.

MMP-1 mRNA expression was also analyzed by real-time quanti-tative PCR [5�-CAT TGA TGG CAT CCA AGC C-3� (sense) and5�-GGC TGG ACA GGA TTT TGG G-3� (antisense)] using Quanti-Tect SYBR-Green Probe RT-PCR kit (Qiagen). Each sample wasassayed in triplicate. For relative quantification, the cycle threshold(Ct) method {ratio 2 � [Ct(MMP-1) � Ct(GAPDH)]} was used,with GAPDH as a control. For copy number determination, a calibra-tion curve was obtained using serial dilutions of a linearized GAPDHcDNA with the GAPDH primer pair [5�-GAA GGT GAA GGT CGGAGT C-3� (forward) and 5�-GAA GAT GGT GAT GGG ATT TC-3�(reverse)].

MMP-1 levels. MMP-1 levels in culture supernatants were ana-lyzed using an ELISA kit according to the manufacturer’s instructions(Amersham Biosciences).

Western blot analysis. ECM proteins (MMP-1, -2, -3, -8, -9, -10,and -13 and TIMP-1, -2, and -4) in the culture supernatants weredetermined by an antibody array (RayBio MMP antibody array 1,catalog no. AAH-MMP-1, RayBiotech, Norcross, GA) following themanufacturer’s protocol and quantified by densitometry (6). Proteinconcentrations were determined using the bicinchoninic acid method(Pierce, Rockford, IL).

MMP-1 levels were confirmed by Western blotting (6, 18, 19).Proteins were separated by 10% PAGE and electroblotted onto aHybond-P polyvinylidene difluoride membrane (Amersham Bio-sciences). The membrane was incubated with rabbit anti-humanMMP-1 (1:2,000 dilution; Chemicon International, Temecula, CA)and subsequently with horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G (New England Biolabs, Beverly, MA). Theimmunoreactive bands were detected by chemiluminescence (ECLPlus, GE Healthcare). The blots were stripped and reprobed with�-tubulin antibodies to confirm equal protein loading.

EMSA, ELISA, and reporter assays. NF-�B and AP-1 protein-DNAcomplex formation was assessed by EMSA (6, 18, 19) using HCFnuclear extracts and double-stranded consensus DNA for C/EBP(5�-TGC AGA TTG CGC AAT CTG CA-3�), NF-�B (5�-AGT TGAGGG GAC TTT CCC AGG C-3�), AP-1 (5�-CGC TTG ATG ACTCAG CCG GAA-3�), mutant C/EBP (5�-TGC AGA GAC TAG TCTCTG CA-3�), mutant NF-�B (5�-AGT TGA GGC GAC TTT CCCAGG C-3), and mutant AP-1 (5�-CGC TTG ATG ACT TGG CCGGAA-3�; Santa Cruz Biotechnology). Activation of transcription fac-tors (TF) was confirmed by ELISA (20, 21, 24) [TransAM TF ELISAkits 43296 (NF-�B), 44196 (C/EBP), and 44296 (AP-1), ActiveMotif, Carlsbad, CA].

TF activation was also analyzed by reporter assays using adenovi-ral transduction of NF-�B (Ad-NF-�B-Luc, 50 MOI; kindly providedby Dr. John F. Engelhardt) and has been previously described (19,25). Similarly, HCF were infected with Ad.AP-1-Luc (catalog no.1670, Vector Biolabs, Philadelphia, PA) reporter vector. Ad-MCS-Luc (Vector Biolabs) served as a control. Ad-�-Gal (50 MOI; VectorBiolabs) served as an internal control. �-Gal activity in cell extractswas determined using luminescent �-Gal detection kit II (BD Bio-sciences), and the results are expressed in relative light units as a ratioof firefly luciferase to �-Gal activity. C/EBP activation was confirmedusing a C/EBP reporter vector (2C/EBP-Luc) that contains two

H3357IL-17 STIMULATES CARDIAC FIBROBLAST MIGRATION

AJP-Heart Circ Physiol • VOL 293 • DECEMBER 2007 • www.ajpheart.org

on March 11, 2009

ajpheart.physiology.orgD

ownloaded from

canonical C/EBP binding sites (kindly provided by Peter Johnson,National Cancer Institute, Frederick, MD). HCF were transfected withC/EBP-Luc (3 �g) using the Nucleofection kit. Cells were cotrans-fected with Ad-�-Gal to normalize for variations in transfectionefficiency. Cells were then treated with IL-17 for 12 h. Fireflyluciferase and �-Gal activities were analyzed as described above.Neither pharmacological inhibitors, dominant-negative expressionvectors, nor siRNA affected cell viability for the indicated studyperiod.

Immune complex protein kinase assays. p38 MAPK and ERKactivities were determined as described previously (6, 18, 19). Briefly,a commercially available colorimetric assay kit [p38 MAP kinaseassay kit (nonradioactive), Cell Signaling Technology] was used todetermine p38 MAPK activity in whole cell homogenates. The assayis based on phosphorylation of activating TF (ATF)-2 by the immu-noprecipitated phosphorylated p38 MAPK. ERK activity was deter-mined in whole cell homogenates using a commercially availablecolorimetric assay kit (ERK, p44/42 MAP kinase assay kit, CellSignaling Technology).

Cell migration. HCF migration was quantified as previously de-scribed (10) using Transwell chambers with 3-�m polycarbonatemembrane (Corning) precoated with 100 �g/ml type I collagen orBSA on both sides of the membrane. HCF were trypsinized andsuspended in FM medium containing 0.5% BSA, and 1 ml containing2.0 105 cells/ml was layered on the coated insert filters. Cells werestimulated with IL-17 (10 ng/ml). The lower chamber contained IL-17at the same concentration. Plates were incubated at 37°C for 24 h.Membranes were washed with PBS, and noninvading cells on theupper surface were removed using cotton swabs. Cells migrating tothe lower surface of the membrane were determined at 540-nmabsorbance using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra-zolium bromide assay.

To determine the role of MMP-1 in IL-17-mediated cell migration,we treated HCF with GM-6001 (10 �M in DMSO for 15 min) orMMP-1 siRNA (target sequence corresponding to nucleotides 153–171 downstream from starting codon; 100 nM for 48 h) beforeaddition of IL-17. siRNA against this target sequence has been shownto knock down MMP-1 mRNA and protein expression by �90% (34).siRNA, which will not target any genes in the human genome(5�-UUC UCC GAA CGU GUC ACG UdTdT-3�; catalog no.1022076, Qiagen), and green fluorescent protein (GFP) siRNA (sense,5�-pGGCUACGUCCAGGAGCGCACC-3�) served as controls.MMP-1 knockdown was confirmed by Western blotting.

Cell death assays. Quiescent HCF were treated with IL-17 (�100ng/ml) for �48 h. Cell death was analyzed by quantitation of mono-and oligonucleosomes in the cytoplasmic fraction of cell lysates by anELISA (Cell Death Detection ELISAPLUS kit, Roche Diagnostics) (6).

Statistical analysis. Values are means � SE. For statistical analy-sis, we used ANOVA followed by an appropriate post hoc multiplecomparison test (Tukey’s method). Data were considered statisticallysignificant at P � 0.05.

RESULTS

IL-17 stimulates MMP-1 mRNA expression in human car-diac fibroblasts. Since the proinflammatory cytokine IL-17signals via IL-17RA, we first used RT-PCR to determinewhether HCF express IL-17RA. Our results demonstrated thatHCF express IL-17RA under basal conditions (data notshown). We next investigated whether IL-17 can induceMMP-1 mRNA expression in HCF. HCF were treated withIL-17 (0–100 �g/ml) for 12 h and analyzed by Northernblotting and densitometry. HCF express MMP-1 mRNA at lowlevels under basal conditions (Fig. 1A), and treatment withIL-17 for 12 h significantly increased MMP-1 expression, withpeak levels obtained with 10 ng/ml. Therefore, in all subse-

quent experiments, IL-17 was used at 10 ng/ml. IL-17 inducedMMP-1 expression in a time-dependent manner, with peaklevels of mRNA observed at 12 h (Fig. 1B). MMP-1 levelsremained at these high levels throughout the 48-h study period.IL-17-induced MMP-1 expression was also investigated byRT-quantitative PCR (qPCR). The specificity of the responseto IL-17 was verified by incubation of HCF with IL-17 orIL-17R neutralizing antibodies (10 �g/ml) for 1 h beforeaddition of IL-17. Our results indicate that the potent inductionof MMP-1 mRNA expression by IL-17 can be blocked byIL-17 or IL-17R neutralizing antibodies (Fig. 1C). These re-sults demonstrate that 1) HCF express IL-17RA, 2) IL-17 is apotent inducer of MMP-1 expression, and 3) IL-17 inducesMMP-1 expression in a time- and dose-dependent manner(Fig. 1).

IL-17 stimulates MMP-1 protein expression. We next inves-tigated whether IL-17 also stimulates MMP-1 protein expres-sion. Western blot analysis of whole cell homogenates usingantibodies that recognize latent and active forms of MMP-1revealed that, under basal conditions, HCF expressed latent andactive forms of MMP-1 (Fig. 2A), whereas treatment withIL-17 resulted in a modest increase in the latent form but a

Fig. 1. IL-17 stimulates matrix metalloproteinase (MMP)-1 expression.A: Northern blot of MMP-1 mRNA expression in DNA-free total RNA isolatedfrom quiescent human cardiac fibroblasts (HCF) treated with IL-17 for 12 h.28S rRNA served as loading control. B: kinetics of IL-17-mediated MMP-1expression shown in Northern blot of MMP-1 mRNA expression in quiescentHCF treated with IL-17 (10 ng/ml) for �48 h. C, control. C: RT-quantitativePCR (qPCR) analysis of MMP-1 mRNA expression in quiescent HCF treatedwith IL-17 or IL-17 receptor (IL-17R) neutralizing antibodies for 1 h before12 h of treatment with IL-17 (10 ng/ml). Normal goat/mouse IgG and GAPDHserved as internal controls. IL-17-mediated MMP-1 expression is blocked byIL-17 or IL-17 receptor (IL-17R) neutralizing antibodies. *P � 0.01 vs.untreated (ANOVA). †P � 0.01 vs. IL-17 (ANOVA).

H3358 IL-17 STIMULATES CARDIAC FIBROBLAST MIGRATION

AJP-Heart Circ Physiol • VOL 293 • DECEMBER 2007 • www.ajpheart.org

on March 11, 2009

ajpheart.physiology.orgD

ownloaded from

significant increase in the active form (Fig. 2A). Furthermore,IL-17 increased MMP-1 secretion from the HCF (Fig. 2B).These results demonstrate that IL-17 1) stimulates MMP-1protein expression, with a modest increase in the latent formand a significant increase in the active form, and 2) stimulatesMMP-1 secretion (Fig. 2).

IL-17 regulates MMP-1 expression at the transcriptionallevel. We next determined whether IL-17-mediated MMP-1expression is regulated at the transcriptional and/or transla-tional level. HCF were treated with CHX or ActD for 1 h,IL-17 (10 ng/ml for 12 h) was added, and RNA was isolatedand analyzed for MMP-1 expression by RT-qPCR. IL-17-mediated MMP-1 expression was significantly inhibited by theRNA synthesis inhibitor ActD, but not by the protein synthesisinhibitor CHX (Fig. 3A). Further studies indicated that the rateof degradation of MMP-1 mRNA was similar in ActD-treatedand untreated controls (data not shown), indicating that mRNAstabilization did not contribute significantly to the induction ofMMP-1 by IL-17. We next determined whether IL-17 inducesMMP-1 promoter-reporter gene activity. Consistent with ourearlier observations, IL-17 stimulated the MMP-1 promoter-

reporter (pGL3-4334) activity, and incubation with IL-17 orIL-17R neutralizing antibodies blocked this effect (Fig. 3B). Ithas been reported previously that several IL-17 responsivegenes contain NF-�B, AP-1, and C/EBP binding elements intheir cis-regulatory regions (27). Since these elements also playa role in cytokine- and growth factor-induced MMP-1 expres-sion, we determined whether these elements mediate IL-17-stimulated MMP-1 transcription. Transfection of the deletionconstructs lacking C/EBP (pGL3-2685), C/EBP and NF-�B(pGL3-1524), or all three binding sites (pGL3-62) and treat-ment with IL-17 showed that IL-17-induced MMP-1 promoter-reporter activity was significantly attenuated by the deletion ofC/EBP or NF-�B, but this inhibition was significantly morepronounced when all three sites were deleted (Fig. 3C). Theseresults demonstrate that 1) IL-17 does not affect MMP-1mRNA stability, 2) IL-17 regulates MMP-1 expression at thetranscriptional level, and 3) IL-17-mediated MMP-1 transcrip-tion is dependent on C/EBP, NF-�B, and AP-1 (Fig. 3).

IL-17 activates NF-�B, AP-1, and c/EBP. Since we demon-strated that IL-17-mediated MMP-1 transcription is dependenton NF-�B, AP-1, and C/EBP (Fig. 3), we next investigatedwhether IL-17 induces activation of these TFs. TF activationswere analyzed by EMSA and reporter assays, and their subunitcomposition was determined by ELISA. IL-17 induced NF-�BDNA binding activity, which was blunted by pretreatment withIL-17 or IL-17R neutralizing antibodies (Fig. 4A). Comple-menting these EMSA results, IL-17 also induced NF-�B-dependent reporter gene activity (Fig. 4B). ELISA of nuclearprotein extracts revealed that p65 and p50 contribute to IL-17-mediated NF-�B activation (Fig. 4C). Similarly, IL-17 inducedAP-1 DNA binding (Fig. 4D) and reporter gene activities (Fig.4E), and c-Fos, FosB, c-Jun, Fra-1, and JunD contributed to itsactivation (Fig. 4F). IL-17 induced C/EBP DNA binding (Fig.4G) and reporter gene activities (Fig. 4H), and C/EBP-�, butnot C/EBP-�, contributed to its activation (Fig. 4I). Theseresults demonstrate that IL-17 potently induces NF-�B (p65

Fig. 2. IL-17 induces MMP-1 protein expression. A: Western blot analysis ofMMP-1 protein levels in whole cell homogenates from quiescent HCF treatedwith IL-17 (10 ng/ml) for 24 h. �-Tubulin served as internal control. B: ELISAquantification of MMP-1 levels in culture supernatants from quiescent HCFtreated as in described A. IL-17 stimulates MMP-1 secretion. *P � 0.001 vs.untreated (ANOVA).

Fig. 3. IL-17 stimulates MMP-1 expression at transcriptional and posttranscriptional levels. A: quantification by RT-qPCR of MMP-1 expression in quiescentHCF treated with IL-17 and/or cyclohexamide (CHX, 10 �g/ml in DMSO) or actinomycin D (ActD, 10 �g/ml in DMSO) for 12 h. IL-17-mediated MMP-1expression is regulated at transcriptional and posttranscriptional levels. B: IL-17 stimulation of MMP-1 promoter-reporter activity in quiescent HCF transfectedwith the full-length MMP-1 promoter-reporter vector (pGL3-4334; 3 �g) and 24 h later treated with IL-17 (10 ng/ml) for 12 h. Specificity of IL-17 was verifiedby 1 h of preincubation of cells with IL-17 or IL-17R neutralizing antibodies (10 �g/ml). pGL3-basic vector (3 �g) served as control. Cells were infected withadenovirus expressing �-galactosidase [Ad-�-Gal, 50 multiplicity of infection (MOI)]. Firefly luciferase and �-Gal activities were analyzed, and results arepresented as fold increase relative to untreated and represent ratio of firefly luciferase to �-Gal activity. *P � 0.001 vs. untreated. †P � 0.01 vs. IL-17. C: deletionof CCAAT enhancer-binding protein (C/EBP), NF-�B, or activator protein (AP)-1 binding sites blunts IL-17-mediated MMP-1 promoter-reporter activity inquiescent HCF transfected with the full-length vector (pGL3-4334) or deletion constructs lacking C/EBP (pGL3-2685), NF-�B (pMMP1–1524), or all 3 bindingsites (pGL3-62). Ad-�-Gal served as control. *P � 0.01 vs. untreated. †P � 0.05 vs. pGL3-4334. §P � 0.01 vs. pGL3-4334.

H3359IL-17 STIMULATES CARDIAC FIBROBLAST MIGRATION

AJP-Heart Circ Physiol • VOL 293 • DECEMBER 2007 • www.ajpheart.org

on March 11, 2009

ajpheart.physiology.orgD

ownloaded from

and p50), AP-1 (c-Fos, FosB, c-Jun, Fra-1, and JunD), andC/EBP-� activation in HCF (Fig. 4).

Targeting TF activation blunts IL-17-mediated MMP-1 ex-pression. We have demonstrated that IL-17 induces NF-�B,AP-1, and C/EBP activation (Fig. 4). We have also shown thatdeletion of the NF-�B, AP-1, or C/EBP binding site bluntsIL-17-mediated MMP-1 transcription (Fig. 3). We next inves-tigated whether targeting NF-�B, AP-1, or C/EBP-� activationwill inhibit IL-17-mediated MMP-1 expression. Activation ofNF-�B, AP-1, and C/EBP-� was targeted by adenoviral trans-duction with dominant-negative expression vectors, phospho-rothioated antisense oligonucleotides (ODN), or RNA interfer-ence. IL-17-induced MMP-1 promoter-dependent reportergene activity was significantly attenuated by adenoviral trans-duction with dnIKK-�, dnp65, or dnI�B-� (S32A/S36A)(Fig. 5A). Similarly, c-Fos, c-Jun antisense ODN, or Ad.dnc-Jun significantly attenuated IL-17-dependent MMP-1 promoterreporter activity (Fig. 5B), as did siRNA-mediated C/EBP-�

knockdown (Fig. 5C). Furthermore, targeting NF-�B, AP-1, orC/EBP-� attenuated IL-17-mediated MMP-1 mRNA expres-sion (Fig. 5D). These results demonstrate that activation ofNF-�B, AP-1, or C/EBP-� is a significant mechanism inIL-17-mediated MMP-1 transcription and mRNA expression(Fig. 5).

IL-17 induces p38 MAPK and ERK1/2 activation. MAPKsare important mediators of a variety of physiological andpathological cellular processes, including cell death, cell sur-vival, proliferation, and migration (20). Since IL-17 inducedMMP-1 transcription and mRNA expression via NF-�B, AP-1,and C/EBP-� activation (Fig. 5) and since these TFs serve asnuclear effectors of MAPKs, we investigated whether IL-17induces MAPK activation in HCF. Quiescent HCF weretreated with IL-17 for 30 min, and cleared cell lysates werethen analyzed for MAPK activation by Western blotting usingactivation-specific antibodies. Kinase activity was determinedby immune complex kinase assays. IL-17 induced p38 MAPK

Fig. 4. IL-17 stimulates AP-1, NF-�B, and C/EBP activation. A: IL-17 stimulation of NF-�B DNA binding activity in quiescent HCF treated with IL-17 (10ng/ml) for 2 h. Nuclear protein was extracted and analyzed by EMSA using labeled double-stranded consensus NF-�B oligodeoxynucleotides (ODNs). Specificityof IL-17 was verified by preincubation of HCF with IL-17 or IL-17R antibodies. B: IL-17 stimulation of NF-�B-dependent reporter gene activity in quiescentHCF transduced with Ad-NF-�B-Luc (50 MOI). Ad-MCS-Luc (50 MOI) served as control; Ad-�-Gal (50 MOI) served as internal control. After 24 h, cells weretreated with IL-17 (10 ng/ml), and firefly luciferase and �-Gal activities were determined. C: contribution of p50 and p65 to IL-17-mediated NF-�B activation.Nuclear extracts from quiescent HCF were treated with IL-17 as described in A and analyzed by ELISA for p50 and p65. OD, optical density.D–I : IL-17-mediated AP-1 and C/EBP activation determined by EMSA [AP-1 (D) and C/EBP (G)], reporter [AP-1 (E) and C/EBP (H)] assays, and ELISA [AP-1(F) and C/EBP (I)]. Arrows in A, D, and G indicate transcription factor-specific DNA-protein complexes. *P � 0.001 vs. respective untreated (B, E, and H);P � 0.01 vs. untreated (C, F, and I).

H3360 IL-17 STIMULATES CARDIAC FIBROBLAST MIGRATION

AJP-Heart Circ Physiol • VOL 293 • DECEMBER 2007 • www.ajpheart.org

on March 11, 2009

ajpheart.physiology.orgD

ownloaded from

phosphorylation (Fig. 6A) and activity (Fig. 6B), which wereblocked by the inhibitor SB-203580. Similarly, IL-17 inducedERK1/2 phosphorylation (Fig. 6C) and activity (Fig. 6D),which were blunted by PD-98059. These results demonstratethat IL-17 potently induces p38 MAPK and ERK1/2 activationin HCF (Fig. 6).

IL-17 induces TF activation and MMP-1 mRNA expressionvia p38 MAPK and ERK1/2. Our results show that IL-17potently induces NF-�B, AP-1, and C/EBP-� activation inHCF (Fig. 4). IL-17 also induced p38 MAPK and ERK1/2activation (Fig. 6). Therefore, we next investigated whetherIL-17 induces TF activation via p38 MAPK and ERK1/2.Quiescent HCF were treated with SB-203580 or PD-98059 andthen with IL-17. TF activation was analyzed after 2 h byELISA of nuclear protein extracts. MMP-1 mRNA expressionwas analyzed after 12 h by RT-qPCR using total RNA. IL-17stimulated nuclear translocation of NF-�B p65, an effect sig-nificantly attenuated by SB-203580 and PD-98059 (Fig. 7A).Similarly, SB-203580 and PD-98059 attenuated IL-17-medi-ated AP-1 (c-Fos; Fig. 7B) and C/EBP (C/EBP-�; Fig. 7C)

activation. Furthermore, SB-203580 and PD-98059 attenuatedIL-17-mediated MMP-1 mRNA expression (Fig. 7D). Theseresults demonstrate that IL-17 induces TF activation andMMP-1 mRNA expression via p38 MAPK and ERK1/2 acti-vation (Fig. 7).

Inhibition of MMP-1 expression blocks IL-17-mediated HCFmigration. Fibroblast migration and proliferation are criticalprocesses in wound healing and scar formation followingischemia, infarction, and inflammation (12). Since MMPs de-grade ECM and facilitate cell migration, we investigatedwhether IL-17 induces HCF migration in an MMP-1-depen-dent manner. IL-17 does indeed stimulate HCF migration, andpretreatment with the broad-spectrum MMP inhibitor GM-6001 or siRNA-mediated MMP-1 knockdown attenuated IL-17-dependent HCF migration (Fig. 8A); knockdown of MMP-1was confirmed by Western blotting (Fig. 8B). The effect ofGM-6001 appeared to be more pronounced than was MMP-1knockdown. However, neither control siRNA (Fig. 8A) norGFP siRNA (data not shown) modulated IL-17-mediated HCFmigration. These results demonstrate that IL-17 induces HCF

Fig. 5. IL-17 stimulates MMP-1 promoter-reporter activity in an NF-�B-, AP-1-, and C/EBP-�-dependent manner. A: attenuation of IL-17-mediated MMP-1promoter-reporter activity by dominant-negative (dn) IKK-�, dnp65, and dnI�B-� in quiescent HCF transfected with pGL3-4334 and transduced with adenoviralvector expressing dnIKK-�, dnp65, or dnI�B-� for 24 h and then treated with IL-17 for 12 h. Ad-�-Gal served as control. B: c-Fos and c-Jun antisense (AS)ODNs or adenoviral transduction of dnc-Jun block IL-17-mediated MMP-1 promoter-reporter activity in quiescent HCF transfected with pGL3-4334 and treatedwith phosphorothioated c-Fos or c-Jun AS ODN or transduced with Ad.dnc-Jun. Scrambled ODN or Ad-�-Gal served as control. C: attenuation ofIL-17-mediated MMP-1 promoter-reporter activity by knockdown of C/EBP-� in quiescent HCF transfected with pGL3-4334 and treated with C/EBP-� smallinterfering RNA (siRNA; 100 nM) for 48 h and then with IL-17 (10 ng/ml). Scrambled siRNA served as control. Firefly luciferase and �-Gal activities wereanalyzed after 12 h. D: attenuation of IL-17-mediated MMP-1 mRNA expression by inhibition of AP-1, NF-�B, or C/EBP-� in quiescent HCF treated asdescribed in A–C with AS ODN, dominant-negative expression vectors, or siRNA before 24 h of IL-17 (10 ng/ml) treatment. MMP-1 expression was quantifiedby RT-qPCR. GAPDH served as internal control. *P � 0.001 vs. untreated. †P � 0.01 (A); P � 0.05 (B, C, D) vs. IL-17. ††P � 0.01 vs. IL-17.

H3361IL-17 STIMULATES CARDIAC FIBROBLAST MIGRATION

AJP-Heart Circ Physiol • VOL 293 • DECEMBER 2007 • www.ajpheart.org

on March 11, 2009

ajpheart.physiology.orgD

ownloaded from

migration, at least in part, in an MMP-1-dependent manner(Fig. 8).

IL-17 induces MMP-2, -3, -9, and -13 and TIMP-1 expres-sion in HCF. In addition to MMP-1, MMP-2, -9, and -13 alsoplay critical roles in ECM degradation and myocardial remod-eling. These MMPs also contain similar TF binding sites intheir promoters and, therefore, may be transcriptionally regu-lated by IL-17. The proteolytic activity of MMPs is tightlyregulated by various physiological inhibitors termed TIMPs.Since TIMPs such as TIMP-1 contain similar cis elements,including AP-1 and NF-�B, in their promoter regions, we

hypothesized that IL-17 may regulate their expression as well.Therefore, we used an antibody array to investigate the expres-sion of various MMPs and TIMPs (Fig. 9A). Confirming ourearlier results obtained using ELISA (Fig. 2B), results in Fig.9B show that IL-17 potently induces MMP-1 expression. Inaddition, IL-17 also induced the expression of MMP-2, -3, -9,and -13 (Fig. 9B). However, IL-17 failed to significantly affectMMP-8 and -10 expression. Similarly, IL-17 induced TIMP-1,but failed to modulate TIMP-2 and -4, expression (Fig. 9B).These experiments were performed three times, and the resultsare summarized in Fig. 9C. Together, these results indicate that

Fig. 6. IL-17 induces p38 MAPK and ERK1/2 activation.A: inhibition of IL-17-mediated p38 MAPK activation bySB-203580 (SB). Quiescent HCF were treated with SB-203580 (1 �M in DMSO) for 30 min and then with IL-17(10 ng/ml) for 30 min, and total and phosphorylated p38(P-p38) MAPK levels were analyzed by Western blotting.B: IL-17 induction of p38 MAPK activity in quiescent HCFtreated with SB-203580 for 30 min and then with IL-17.p38 MAPK activity was determined by immune complexkinase assay, with activating transcription factor (ATF)-2used as a substrate. �-Tubulin served as internal control.C: inhibition of IL-17-induced ERK1/2 activation by PD-98059 (PD) in quiescent HCF treated with PD-98059 (10�M in DMSO) for 1 h and then with IL-17. Total andphosphorylated ERK1/2 (pERK1/2) levels were analyzedby Western blotting. D: IL-17 induction of ERK1/2 activityin quiescent HCF treated with PD-98059 and then withIL-17. ERK1/2 activity was determined by immune com-plex kinase assay, with Elk used as substrate. �-Tubulinserved as internal control. Experiments were performed �3times.

Fig. 7. IL-17-mediated NF-�B, AP-1, and C/EBP-�activation and MMP-1 expression are dependent onp38 MAPK and ERK1/2 activation. A: attenuationof IL-17-mediated NF-�B activation by inhibitionof p38 MAPK and ERK1/2 in quiescent HCFtreated with SB-203580 (1 �M in DMSO for 30min), PD-98059 (10 �M in DMSO for 1 h), orDMSO and then with IL-17 (10 ng/ml for 2 h).Nuclear p65 levels were quantified by ELISA.B: attenuation of IL-17-mediated AP-1 activation byinhibition of p38 MAPK and ERK1/2. Nuclear pro-tein extracts isolated as in described A were ana-lyzed for c-Fos levels by ELISA. C: attenuation ofIL-17-mediated C/EBP-� activation by inhibition ofp38 MAPK and ERK1/2. Nuclear protein extractsisolated as described in A were analyzed forC/EBP-� levels by ELISA. D: attenuation of IL-17-mediated MMP-1 mRNA expression by inhibitionof p38 MAPK and ERK1/2 in quiescent HCFtreated with SB-203580 or PD-98059 and then withIL-17. After 24 h, total RNA was isolated andMMP-1 expression was analyzed by RT-qPCR.*P � 0.001 (A, C, D); P � 0.01 (B) vs. untreated.†P � 0.01 (A, C); P � 0.05 (B); P � 0.001 (D) vs.IL-17.

H3362 IL-17 STIMULATES CARDIAC FIBROBLAST MIGRATION

AJP-Heart Circ Physiol • VOL 293 • DECEMBER 2007 • www.ajpheart.org

on March 11, 2009

ajpheart.physiology.orgD

ownloaded from

IL-17 induces the expression of various MMPs and TIMP-1,which play a role in myocardial remodeling (Fig. 9).

DISCUSSION

The results from this study show that the novel cytokineIL-17 is a potent inducer of MMP-1 expression in primaryHCF and stimulates MMP-1 expression independently of IL-1�, IL-6, and TNF-�. IL-17 regulates MMP-1 expression at thetranscriptional level and is dependent on AP-1, NF-�B, andC/EBP-� activation. More importantly, IL-17 induces HCFmigration in an MMP-1-dependent manner. Since MMPs de-grade ECM and facilitate migration, our results suggest thatIL-17 may be potentially important in myocardial injury,remodeling, and failure.

IL-17s constitute a newly discovered and unique family ofcytokines that show no structural homology to other ILs (4).IL-17 is expressed mainly by a subset of CD4� T cells, i.e.,Th17 cells. Many cell types express IL-17Rs and are, therefore,targets of IL-17. Increased IL-17 levels have been detected invarious models of inflammation, including rheumatoid arthritisand periodontitis (1, 4). However, few studies have reportedthe role of IL-17 in myocardial inflammation and injury.Sonderegger et al. (28) demonstrated that administration ofIL-23 neutralizing antibodies attenuated experimental autoim-mune myocarditis (EAM). Since IL-23 stimulates a pathogenicIL-17-producing T cell population, these authors hypothesizedthat targeting IL-17 would reduce EAM. They targeted IL-17expression by an active vaccination approach that breaks B celltolerance and found that neutralization of IL-17 effectivelyreduced heart autoantibody responses and myocardial inflam-mation (28). Recently, Rangachari et al. (21) showed that micelacking T-bet, a T-box TF essential for Th1 lineage differen-tiation, develop severe EAM. Using T-bet�/� IL-12R�1�/�

and T-bet�/� IL-12p35�/� mice and antibody depletion exper-

iments, these authors reported that IL-23 and IL-17 are criticalfor EAM pathogenesis.

EAM serves as an animal model for postinfectious myocar-ditis and cardiomyopathy. Dilated cardiomyopathy is charac-terized by increased MMP-1 expression (13). It is thereforeplausible that IL-17 may play a causal role in dilated cardio-myopathy via enhanced expression of various MMPs, includ-ing MMP-1. Although the above-mentioned studies demon-strate a role for IL-17 in myocarditis and cardiomyopathy, Liet al. (16) showed that local expression of soluble IL-17R-immunoglobulin chimera (sIL-17R-Ig) prolongs graft survivalin rat cardiac allografts by suppressing cytokine responses andleukocyte infiltration. Together, these studies suggest thatIL-17 may be a therapeutic target to reduce cardiac inflamma-tion and injury.

Although the above-mentioned studies show that neutraliza-tion of IL-17 blunts myocardial inflammation mainly by reg-ulating Th1 cell responses and attenuating inflammatory andimmune cell infiltration (16, 21, 28), it is not known whetherIL-17 affects myocardial biology directly. It is also not knownwhether myocardial constituent cells express IL-17 andwhether IL-17 affects myocardial cells differentially. Using

Fig. 9. IL-18 stimulates secretion of MMPs and tissue inhibitors of MMPs(TIMPs). A: antibody array that detects various MMPs and TIMPs simulta-neously was used to assess whether IL-17 induces other MMPs and TIMPs.POS, positive control; NEG, negative control. B: IL-17 stimulates secretion ofMMPs and TIMPs. Quiescent HCF were stimulated with saline or IL-17 (10ng/ml) for 24 h, and culture supernatants were collected and analyzed forextracellular matrix proteins. C: quantitation of signals in B by image analysis.Intensity of signals was normalized to saline-treated control samples, andresults are expressed as fold increases. *P � 0.05; **P � 0.001 vs. respectivesaline (by ANOVA).

Fig. 8. IL-17 stimulates HCF migration in an MMP-1-dependent manner.A: quiescent HCF were layered on collagen type I-coated polycarbonatemembrane and treated with the MMP-1 inhibitor GM-6001 (10 �M in DMSOor DMSO alone for 15 min) and then with IL-17. MMP-1 expression was alsotargeted by siRNA: quiescent HCF were treated with MMP-1 or control siRNA(100 nM for 48 h), layered on collagen-coated filters, and treated with IL-17.Cell migration was determined after 24 h. B: knockdown of MMP-1 proteinwas confirmed by Western blotting. �-Tubulin served as internal control. *P �0.01 vs. untreated. †P � 0.05 vs. IL-17.

H3363IL-17 STIMULATES CARDIAC FIBROBLAST MIGRATION

AJP-Heart Circ Physiol • VOL 293 • DECEMBER 2007 • www.ajpheart.org

on March 11, 2009

ajpheart.physiology.orgD

ownloaded from

multiplexed immunoassays, we observed for the first time thathuman cardiac fibroblasts secrete various proinflammatorycytokines, including low levels (4.1 pg/ml) of IL-17 at basalconditions (unpublished observations). Since no cells of anon-HCF phenotype contaminate these cultures, our resultssuggest that cells other than Th17 lineage may secrete IL-17.Thus fibroblast-secreted IL-17 may affect fibroblasts and othermyocardial cells via autocrine and paracrine mechanisms.

In the present study, we show that IL-17 induces MMP-1expression in cardiac fibroblasts, and these stimulatory effectsare independent of other proinflammatory cytokines. IL-17induced MMP-1 expression via enhanced transcription, ratherthan MMP-1 mRNA stability. Furthermore, IL-17 stimulatedMMP-1 transcription via NF-�B, AP-1, and C/EBP-� activa-tion. It has been previously demonstrated that cytokines induceMMP-1 expression via c-Jun induction (22). Although we havenot investigated whether IL-17 induces c-Jun expression, ourstudies show that IL-17-mediated AP-1 DNA binding involvesvarious subunits including c-Jun, and treatment with c-Junantisense ODN and adenoviral transduction of dnc-Jun atten-uate IL-17-mediated MMP-1 transcription. Our results alsoshow that IL-17-mediated AP-1 activation includes Fra-1.Since Fra-1 confers invasiveness and motility in various cancercell lines (2), it is plausible that Fra-1 may function in a similarfashion in mediating IL-17-dependent fibroblast migration.

In addition to AP-1, our results indicate that IL-17-mediatedMMP-1 expression is dependent on NF-�B and C/EBP-�activation. In support of our studies, Raymond et al. (23)recently demonstrated that NF-�B and C/EBP-� cooperativelyinduce IL-1�-mediated MMP-1 transcription in chondrocytes.They further demonstrated that IL-1� induces phosphorylationof C/EBP-� on threonine 235, enhancing C/EBP-� transacti-vation potential. In that study, IL-1� induced C/EBP-� phos-phorylation via ERK activation (23). Our results show thatIL-17 induces MMP-1 expression via p38 MAPK and ERKactivation. Together, these studies demonstrate that p38MAPK- and ERK-dependent coordinated activation of NF-�B,AP-1, and C/EBP-� plays a role in IL-17-mediated MMP-1induction in cardiac fibroblasts. Since MMPs such as MMP-3and -9 are also responsive to AP-1 and NF-�B activation (6,26, 33), IL-17 may induce their expression and, thus, play acritical role ECM regulation and myocardial remodeling.

In fact, our results demonstrate that, in addition to MMP-1,IL-17 induces the expression of other members of the MMPfamily, such as MMP-2, -3, -9, and -13, which play criticalroles in myocardial remodeling. Similar to MMP-1, theseMMPs are also regulated predominantly at the transcriptionallevel. On the basis of the composition of cis-regulatory regionsin their promoters, MMPs are arbitrarily grouped into threecategories (32). Group 1 consists of MMP-1, -3, -7, -9, -10,-12, and -13, which contain a TATA box and an AP-1 bindingsite proximal to the transcriptional start site. MMP-9 alsocontains an NF-�B site at the distal region, and its expressionis regulated at the transcriptional levels via AP-1 and NF-�B.Group 2 consists of MMP-8 and -11, which contain a TATAbox but lack a proximal AP-1 site. Group 3 consists of MMP-2and -14, which lack the TATA box and the proximal AP-1 site.Because of these variations, it is possible that these MMPs areregulated differently. Our results show that IL-17 potentlyinduces the expression of various group 1 MMPs. However,lack of significant induction of MMP-10 expression suggests

that its induction is not solely dependent on AP-1 activation.The MMP-8 promoter, which contains a TATA box but lacksa proximal AP-1 site, failed to respond to IL-17. In contrast toits effects on MMP-1 induction, IL-17 failed to induce MMP-8expression, despite three potential C/EBP binding sites at �70,�112, and �164 in its cis-regulatory region. IL-17 also in-duces the expression of MMP-2 in HCF. MMP-2 containsneither a TATA box nor the proximal AP-1 site. However,MMP-2 contains two AP-2 binding sites at �61 and �1649. Itis possible that IL-17 may induce MMP-2 via AP-2 activation.However, Bergman et al. (3) demonstrated that the MMP-2promoter contains an AP-1 binding site at �1670 that bindsFosB and JunB, suggesting that further critical analysis of TFbinding sites is necessary.

Transcription is a complex process and is regulated atmultiple stages. It is possible that IL-17 may regulate MMPexpression at transcriptional and posttranscriptional levels.Despite few CpG islands in the promoter regions of MMPs,epigenetic mechanisms have recently been shown to signifi-cantly affect MMP expression (31). For example, increasedpromoter methylation was shown to suppress MMP-9 tran-scription (9). Therefore, it is possible that IL-17 may regulateexpression of MMPs via genetic and epigenetic mechanisms. Itis also possible that IL-17-induced MMP induction is mediatedby an intermediary, inasmuch as IL-17 is a potent inducer ofvarious proinflammatory cytokines that are known to induceMMP expression. However, in the present study, we haveshown that neutralization of IL-�, IL-6, and TNF-� fails toinhibit IL-17-mediated MMP-1 induction.

MMPs are synthesized as proenzymes and, upon secretion,bind to various ECM components (31, 32). These storedproforms are immediately available and become activatedduring inflammation and injury. In addition to ECM degrada-tion, MMPs also release ECM-bound growth factors and otherbiological molecules. For example, TGF-�, a growth factorreadily expressed after myocardial ischemic injury (7), issecreted as a latent and inactive form due to an intact prodo-main, the latency-associated peptide (17). MMP-1 degradeslatency-associated peptide and releases mature TGF-�, andmature TGF-� downregulates MMP-1 expression. This dy-namic coregulation and downregulation of MMP-1 expressionmay result in reduced tissue injury. In fact, active MMP-1 hasbeen shown to induce cardiomyocyte death (8), and thesecytotoxic effects are blunted by exogenous addition of matureTGF-�. However, studies are in progress to determine whetherIL-17 coregulates MMP-1 and TGF-� expression.

Our studies have important clinical implications. 1) We haveestablished for the first time that IL-17, a novel proinflamma-tory cytokine, induces primary cardiac fibroblast migration inan MMP-1-dependent manner. Since fibroblast migration andproliferation are two critical steps in cardiac fibrosis, ourresults indicate that IL-17 may play a role in myocardialremodeling. 2) IL-17 induced NF-�B, AP-1, and C/EBP-�activation. Therefore, IL-17 may upregulate NF-�B-, AP-1-,and C/EBP-�-responsive proinflammatory cytokines, chemo-kines, adhesion molecules, and MMPs in fibroblasts and othermyocardial constituent cells. IL-17 may synergize with thesemediators and induce myocardial inflammation and injury. 3)Neutrophils, at least initially, play a role in myocardial isch-emic injury. IL-17, a potent inducer of neutrophil chemoattrac-tants (24), may amplify the inflammatory cascade during

H3364 IL-17 STIMULATES CARDIAC FIBROBLAST MIGRATION

AJP-Heart Circ Physiol • VOL 293 • DECEMBER 2007 • www.ajpheart.org

on March 11, 2009

ajpheart.physiology.orgD

ownloaded from

ischemic injury via recruitment of neutrophils to the site ofinjury/inflammation. Therefore, targeting IL-17 expressionmay reduce fibrosis and remodeling following myocardialinflammation and injury.

GRANTS

This work was supported in part by the Research Service of the Departmentof Veterans Affairs (B. Chandrasekar) and National Institutes of Health GrantsDE-017139 and K02 DE-016312 (B. Steffensen).

REFERENCES

1. Beklen A, Ainola M, Hukkanen M, Gurgan C, Sorsa T, Konttinen YT.MMPs, IL-1, and TNF are regulated by IL-17 in periodontitis. J Dent Res86: 347–351, 2007.

2. Belguise K, Kersual N, Galtier F, Chalbos D. FRA-1 expression levelregulates proliferation and invasiveness of breast cancer cells. Oncogene24: 1434–1444, 2005.

3. Bergman MR, Cheng S, Honbo N, Piacentini L, Karliner JS, LovettDH. A functional activating protein 1 (AP-1) site regulates matrix metal-loproteinase 2 (MMP-2) transcription by cardiac cells through interactionswith JunB-Fra1 and JunB-FosB heterodimers. Biochem J 369: 485–496,2003.

4. Bettelli E, Oukka M, Kuchroo VK. T(H)-17 cells in the circle ofimmunity and autoimmunity. Nat Immunol 8: 345–350, 2007.

5. Borden P, Solymar D, Sucharczuk A, Lindman B, Cannon P, HellerRA. Cytokine control of interstitial collagenase and collagenase-3 geneexpression in human chondrocytes. J Biol Chem 271: 23577–23581, 1996.

6. Chandrasekar B, Mummidi S, Mahimainathan L, Patel DN, BaileySR, Imam SZ, Greene WC, Valente AJ. Interleukin-18-induced humancoronary artery smooth muscle cell migration is dependent on NF-�B- andAP-1-mediated matrix metalloproteinase-9 expression and is inhibited byatorvastatin. J Biol Chem 281: 15099–15109, 2006.

7. Chandrasekar B, Streitman JE, Colston JT, Freeman GL. Inhibition ofnuclear factor-�B attenuates proinflammatory cytokine and induciblenitric-oxide synthase expression in postischemic myocardium. BiochimBiophys Acta 1406: 91–106, 1998.

8. Chen H, Li D, Saldeen T, Mehta JL. TGF-�1 attenuates myocardialischemia-reperfusion injury via inhibition of upregulation of MMP-1.Am J Physiol Heart Circ Physiol 284: H1612–H1617, 2003.

9. Chicoine E, Esteve PO, Robledo O, Van Themsche C, PotworowskiEF, St-Pierre Y. Evidence for the role of promoter methylation in theregulation of MMP-9 gene expression. Biochem Biophys Res Commun297: 765–772, 2002.

10. Colston JT, de la Rosa SD, Freeman GL. Impact of brief oxidant stresson primary adult cardiac fibroblasts. Biochem Biophys Res Commun 316:256–262, 2004.

11. Diwan A, Tran T, Misra A, Mann DL. Inflammatory mediators and thefailing heart: a translational approach. Curr Mol Med 3: 161–182, 2003.

12. Eckes B, Zigrino P, Kessler D, Holtkotter O, Shephard P, Mauch C,Krieg T. Fibroblast-matrix interactions in wound healing and fibrosis.Matrix Biol 19: 325–332, 2000.

13. Felkin LE, Birks EJ, George R, Wong S, Khaghani A, Yacoub MH,Barton PJ. A quantitative gene expression profile of matrix metallopro-teinases (MMPS) and their inhibitors (TIMPS) in the myocardium ofpatients with deteriorating heart failure requiring left ventricular assistdevice support. J Heart Lung Transplant 25: 1413–1419, 2006.

14. Ganea E, Trifan M, Laslo AC, Putina G, Cristescu C. Matrix metal-loproteinases: useful and deleterious. Biochem Soc Trans 35: 689–691,2007.

15. Levy DE, Lapierre F, Liang W, Ye W, Lange CW, Li X, Grobelny D,Casabonne M, Tyrrell D, Holme K, Nadzan A, Galardy RE. Matrixmetalloproteinase inhibitors: a structure-activity study. J Med Chem 41:199–223, 1998.

16. Li J, Simeoni E, Fleury S, Dudler J, Fiorini E, Kappenberger L, vonSegesser LK, Vassalli G. Gene transfer of soluble interleukin-17 receptorprolongs cardiac allograft survival in a rat model. Eur J Cardiothorac Surg29: 779–783, 2006.

17. Massague J. The transforming growth factor-� family. Annu Rev CellBiol 6: 597–641, 1990.

18. Patel DN, Bailey SR, Gresham JK, Schuchman DB, Shelhamer JH,Goldstein BJ, Foxwell BM, Stemerman MB, Maranchie JK, ValenteAJ, Mummidi S, Chandrasekar B. TLR4-NOX4-AP-1 signaling medi-ates lipopolysaccharide-induced CXCR6 expression in human aorticsmooth muscle cells. Biochem Biophys Res Commun 347: 1113–1120,2006.

19. Patel DN, King CA, Bailey SR, Holt JW, Venkatachalam K, AgrawalA, Valente AJ, Chandrasekar B. Interleukin-17 stimulates C-reactiveprotein expression in hepatocytes and smooth muscle cells via p38 MAPKand ERK1/2-dependent NF-�B and C/EBP� activation. J Biol Chem 282:27229–27238, 2007.

20. Petrich BG, Wang Y. Stress-activated MAP kinases in cardiac remodel-ing and heart failure: new insights from transgenic studies. TrendsCardiovasc Med 14: 50–55, 2004.

21. Rangachari M, Mauermann N, Marty RR, Dirnhofer S, Kurrer MO,Komnenovic V, Penninger JM, Eriksson U. T-bet negatively regulatesautoimmune myocarditis by suppressing local production of interleukin17. J Exp Med 203: 2009–2019, 2006.

22. Ray A, Shakya A, Ray BK. Inflammation-responsive transcription fac-tors SAF-1 and c-Jun/c-Fos promote canine MMP-1 gene expression.Biochim Biophys Acta 1732: 53–61, 2005.

23. Raymond L, Eck S, Mollmark J, Hays E, Tomek I, Kantor S, ElliottS, Vincenti M. Interleukin-1� induction of matrix metalloproteinase-1transcription in chondrocytes requires ERK-dependent activation ofCCAAT enhancer-binding protein-�. J Cell Physiol 207: 683–688, 2006.

24. Ruddy MJ, Shen F, Smith JB, Sharma A, Gaffen SL. Interleukin-17regulates expression of the CXC chemokine LIX/CXCL5 in osteoblasts:implications for inflammation and neutrophil recruitment. J Leukoc Biol76: 135–144, 2004.

25. Sanlioglu S, Williams CM, Samavati L, Butler NS, Wang G, McCrayPB Jr, Ritchie TC, Hunninghake GW, Zandi E, Engelhardt JF.Lipopolysaccharide induces Rac1-dependent reactive oxygen species for-mation and coordinates tumor necrosis factor-� secretion through IKKregulation of NF-�B. J Biol Chem 276: 30188–30198, 2001.

26. Sato H, Seiki M. Regulatory mechanism of 92-kDa type IV collagenasegene expression which is associated with invasiveness of tumor cells.Oncogene 8: 395–405, 1993.

27. Shen F, Hu Z, Goswami J, Gaffen SL. Identification of commontranscriptional regulatory elements in interleukin-17 target genes. J BiolChem 281: 24138–24148, 2006.

28. Sonderegger I, Rohn TA, Kurrer MO, Iezzi G, Zou Y, Kastelein RA,Bachmann MF, Kopf M. Neutralization of IL-17 by active vaccinationinhibits IL-23-dependent autoimmune myocarditis. Eur J Immunol 36:2849–2856, 2006.

29. Spinale FG. Matrix metalloproteinases: regulation and dysregulation inthe failing heart. Circ Res 90: 520–530, 2002.

30. Spinale FG, Coker ML, Bond BR, Zellner JL. Myocardial matrixdegradation and metalloproteinase activation in the failing heart: a poten-tial therapeutic target. Cardiovasc Res 46: 225–238, 2000.

31. Vincenti MP, Brinckerhoff CE. Signal transduction and cell-type spe-cific regulation of matrix metalloproteinase gene expression: Can MMPsbe good for you? J Cell Physiol 213: 355–364, 2007.

32. Yan C, Boyd DD. Regulation of matrix metalloproteinase gene expres-sion. J Cell Physiol 211: 19–26, 2007.

33. Yokoo T, Kitamura M. Dual regulation of IL-1�-mediated matrixmetalloproteinase-9 expression in mesangial cells by NF-�B and AP-1.Am J Physiol Renal Fluid Electrolyte Physiol 270: F123–F130, 1996.

34. Yuan J, Dutton CM, Scully SP. RNAi-mediated MMP-1 silencinginhibits human chondrosarcoma invasion. J Orthop Res 23: 1467–1474,2005.

H3365IL-17 STIMULATES CARDIAC FIBROBLAST MIGRATION

AJP-Heart Circ Physiol • VOL 293 • DECEMBER 2007 • www.ajpheart.org

on March 11, 2009

ajpheart.physiology.orgD

ownloaded from