Fatal Eosinophilic Develops in the Absence of IFN- γ and IL-17A Jobert G. Barin, G. Christian Baldeviano, Monica V. Talor, Lei Wu, SuFey Ong, DeLisa Fairweather, Djahida Bedja, This information is current as Natalie R. Stickel, Jillian A. Fontes, Ashley B. Cardamone, of September 25, 2021. Dongfeng Zheng, Kathleen L. Gabrielson, Noel R. Rose and Daniela Ciháková J Immunol 2013; 191:4038-4047; Prepublished online 18

September 2013; Downloaded from doi: 10.4049/jimmunol.1301282 http://www.jimmunol.org/content/191/8/4038

Supplementary http://www.jimmunol.org/content/suppl/2013/09/18/jimmunol.130128 http://www.jimmunol.org/ Material 2.DC1 References This article cites 70 articles, 16 of which you can access for free at: http://www.jimmunol.org/content/191/8/4038.full#ref-list-1

Why The JI? Submit online. by guest on September 25, 2021 • Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Fatal Eosinophilic Myocarditis Develops in the Absence of IFN-g and IL-17A

Jobert G. Barin,*,†,‡ G. Christian Baldeviano,‡,x Monica V. Talor,† Lei Wu,‡ SuFey Ong,‡ DeLisa Fairweather,{ Djahida Bedja,‖ Natalie R. Stickel,# Jillian A. Fontes,‡ Ashley B. Cardamone,† Dongfeng Zheng,† Kathleen L. Gabrielson,‖ Noel R. Rose,*,†,‡ and Daniela Ciha ´kova´†

CD4+ T cells play a central role in inflammatory disease, implicating a cytokine product associated with Th cell effector function as a necessary mediator of this pathophysiology. IFN-g–deficient mice developed severe experimental autoimmune myocarditis (EAM), in which mice are immunized with cardiac myosin peptide, whereas IL-17A–deficient mice were protected 2/2 2/2

from progression to dilated cardiomyopathy. We generated IFN-g IL-17A mice to assess whether IL-17 signaling was Downloaded from responsible for the severe EAM of IFN-g2/2 mice. Surprisingly, IFN-g2/2IL-17A2/2 mice developed a rapidly fatal EAM. Eosinophils constituted a third of infiltrating leukocytes, qualifying this disease as eosinophilic myocarditis. We found increased cardiac production of CCL11/eotaxin, as well as Th2 deviation, among heart-infiltrating CD4+ cells. Ablation of eosinophil development improved survival of IFN-g2/2IL-17A2/2 mice, demonstrating the necessity of eosinophils in fatal . The severe and rapidly fatal autoimmune inflammation that developed in the combined absence of IFN-g and IL-17A constitutes

a novel model of eosinophilic heart disease in humans. This is also, to our knowledge, the first demonstration that eosinophils http://www.jimmunol.org/ have the capacity to act as necessary mediators of morbidity in an autoimmune process. The Journal of Immunology, 2013, 191: 4038–4047.

yocarditis encompasses a diverse family of inflam- myocarditis, a variant associated with poor prognosis and rapidly matory heart diseases associated with a large variety of fatal outcomes (6–8). M infectious and pharmacologic causes (1, 2). We have Experimental autoimmune myocarditis (EAM) is a model of published extensively on myocarditis as a paradigm of post- human inflammatory heart disease in which animals are immunized infectious , in which autoaggressive responses repre- with cardiac Ags to recapitulate autoreactive responses following sent a critical intermediary in the development of lasting cardiologic viral clearance (9, 10). The disease is dependent on CD4+ T cells, by guest on September 25, 2021 sequelae, often culminating in inflammatory dilated cardiomyopa- implicating cytokine products of these cells as necessary signaling thy (DCM) and heart failure (3–5). Myocarditis can take several intermediaries in disease pathophysiology (11). phenotypic forms in humans, including necrotizing eosinophilic IL-17 has been described as a family of proinflammatory cyto- kines bridging innate and adaptive (12). We recently reported a critical requirement for IL-17A in mediating profibrotic *Immunology Training Program, Johns Hopkins University School of Medicine, † remodeling in the heart during EAM and subsequent progression Baltimore, MD 21205; Department of Pathology, Johns Hopkins University School 2/2 of Medicine, Baltimore, MD 21205; ‡William H. Feinstone Department of Molecular to DCM; although IL-17A mice are protected from fibrosis, Microbiology and Immunology, Johns Hopkins University Bloomberg School of they are not protected from inflammatory myocarditis (13). The Public Health, Baltimore, MD 21205; xDepartment of Parasitology, U.S. Naval Med- { archetypal member of this family, IL-17A, elicits neutrophilic ical Research Unit Six (NAMRU-6), Lima, Peru; Department of Environmental Health Sciences, Johns Hopkins University Bloomberg School of Public Health, responses in host defense to fungi and extracellular bacteria (14, Baltimore, MD 21205; ‖Department of Comparative Medicine, Johns Hopkins Uni- 15). IL-17 signaling pathways bear phylogenetic and signaling # versity School of Medicine, Baltimore, MD 21205; and Department of Hematology relationships to the TLR/IL-1 superfamily (16). Immunologically and Oncology, University of Freiburg, 79106 Freiburg, Germany relevant production of IL-17A has been identified in various cel- Received for publication May 16, 2013. Accepted for publication August 14, 2013. lular populations, including gd T cells, NK cells, and a subset of This work was supported by National Institutes of Health/National Heart, Lung, and + Blood Institute (NIH/NHLBI) Grants R01 HL67290 and R01 HL113008 and a grant- CD4 T cells, termed “Th17” cells (17, 18). Th17 differentiation in-aid from the American Heart Association (to N.R.R.), NIH/NHLBI Grant R01 is programmed by coordinate signaling from TGF-b,IL-6,and HL087033 (to D.F.), the Mary Renner Fellowship in Autoimmune Disease Research and IL-23, and is dependent on the transcription factor retinoic acid the O’Leary-Wilson Fellowship at Johns Hopkins Autoimmune Disease Research Center from the American Autoimmune-Related Disease Association (to J.G.B.), and the receptor–related orphan receptor g (19, 20). Michel Mirowski M.D. Discovery Foundation, the W.W. Smith Charitable Trust (heart The involvement of Th17 CD4+ cells in autoimmune disease research Grant H1103), and the Children’s Cardiomyopathy Foundation (to D.C.). continues to be an active area of investigation. The importance of  Address correspondence and reprint requests to Dr. Daniela Ciha´kova´, Johns Hopkins the Th17 lineage was discovered, in part, by contrasting the roles University School of Medicine, Ross 648, 720 Rutland Avenue, Baltimore, MD 21205-2196. E-mail address: [email protected] of IL-23 and IL-12 in the pathogenesis of experimental autoim- The online version of this article contains supplemental material. mune encephalomyelitis (EAE) (21). Adoptive transfer experiments have shown that Th17-conditioned cells induced qualitatively dif- Abbreviations used in this article: CVB3, coxsackievirus B3; DCM, dilated cardio- myopathy; DKO, double-knockout; EAE, experimental autoimmune encephalomy- ferent, if not more severe, autoimmune disease than did Th1 cells elitis; EAM, experimental autoimmune myocarditis; EF, ejection fraction; FS, (22–24). fractional shortening; IVSD, interventricular septal wall thickness at end diastole; LV, left ventricular; LVESD, LV end-systolic dimension; LVPWTED, LV posterior wall Th17 lineage cells are thought to be responsible for the protective thickness at end diastole; WT, wild-type. effect of IFN-g inseveralanimalmodelsofautoimmunedisease www.jimmunol.org/cgi/doi/10.4049/jimmunol.1301282 The Journal of Immunology 4039

2 2 (25, 26). IFN-g / mice develop more severe EAM than do wild- mation of the percentage of area of myocardium infiltrated with inflam- type (WT) controls, and progress to fibrosis and DCM at later matory cells, fibrosis, and cardiomyocyte necrosis, determined from five timepoints (27–30). More severe disease may be elicited in IFN-g2/2 sections per heart according to the following scoring system: grade 0, no inflammation; grade 1, , 10% of the heart section is involved; grade 2, mice through unconstrained outgrowth of autoaggressive Th17 10–30%; grade 3, 30–50%; grade 4, 50–90%; grade 5, . 90% (9). Grading clones in vivo. Several investigators have reported increased Th17 was performed by a minimum of two independent, blinded investigators responses in association with severe disease in IFN-g2/2 mice, and averaged. consistent with this hypothesis (31–35). These findings led us to Flow cytometry investigate whether the protective effect of IFN-g in EAM is mediated through suppressing Th17 differentiation in autoag- Heart-infiltrating leukocytes were isolated from animals perfused for 3 gressive CD4+ T cells. 3 min with 1 PBS + 0.5% BSA and digested in gentleMACS C Tubes according to the manufacturer’s instructions (Miltenyi Biotec). Spleno- Unexpectedly, we found that mice deficient in both IFN-g and cytes were also extracted into single-cell suspension in 13 PBS + 0.5% IL-17A developed a severe, rapidly fatal form of EAM, charac- BSA, and RBCs were lysed by , 5 min incubation in ACK lysis buffer terized by extensive eosinophilic infiltration, cardiomyocyte ne- (Biofluids). Prior to surface staining, viability was determined by crosis, and thrombogenesis, with evidence of Th2 deviation and LIVE/DEAD staining according to the manufacturer’s instructions 2 2 (Molecular Probes). Cells were washed, and FcgRII/III was blocked rapid decline of cardiac function. Comparisons to both IL-17A / 2/2 with aCD16/32 (eBioscience). Surface markers were stained with and IFN-g mice suggest that both cytokines collaborate in the fluorochrome-conjugated mAbs (eBioscience, BD Pharmingen, Bio- suppression of autoaggressive Th2 differentiation and eosinophilic Legend). Samples were acquired on the LSR II quad-laser cytometer cardiac infiltration. In this article, we demonstrate a requisite role running FACSDiva 6 (BD Immunocytometry). Data were analyzed with for eosinophils in effecting the rapid mortality of eosinophilic FlowJo 7.6 (TreeStar Software). Downloaded from myocarditis. We further provide evidence of specific features of ELISA autoimmune pathophysiology independently mediated by eosin- ophils, and the Th2 effector program, respectively. Supernatants from stimulated cells were stored at –80˚C. For cardiac homogenates, tissues were snap frozen, stored at –80˚C, homogenized in MEM + 2% FBS, and stored at –80˚C until used in ELISA or Linco assays. Materials and Methods Homogenate cytokine levels were normalized to wet heart weights prior to Mice homogenization. Linco multiplex cytokine assays (Millipore) were used http://www.jimmunol.org/ according to the manufacturer’s instructions, and acquired on a Luminex IL-17A2/2 founder mice were kindly provided by Yoichiro Iwakura xMAP reader. Alternatively, quantitative sandwich ELISA for supernatants (University of Tokyo, Tokyo, Japan). IFN-g2/2 and WT BALB/cJ mice was determined by colorimetric ELISA kits according to the manu- were obtained from The Jackson Laboratory (Bar Harbor, ME). IL-17A2/2 facturers’ recommended protocols (R&D Systems, IBL). Total serum IgE mice were crossed to IFN-g2/2 and bred to homozygosity at both loci was determined by sandwich ELISA (BD Biosciences). employing established PCR genotyping (36, 37). Established IFN-g2/2IL- 17A2/2 BALB/c mice were outcrossed to the DdblGATA1 BALB/c strain Real-time PCR (JAX 5653), and resulting progeny were intercrossed for a minimum of Tissue mRNA was extracted in TRIzol (Invitrogen), quantitated by two generations to generate homozygous triple-mutant founders. As a re- Nanodrop (Thermo Fisher), DNase digested (Ambion), reverse transcribed sult, cross-intermediates did not include DdblGATA1 3 single-knockout

(Fermentas), and amplified with iQ SYBR Green Mastermix (Bio-Rad) by guest on September 25, 2021 controls, which may be the focus of future studies. All mice were main- acquired on the MyiQ2 thermocycler (Bio-Rad) running iQ5 software tained in the Johns Hopkins University School of Medicine specific (Bio-Rad). Synthetic oligonucleotide primers were commercially synthe- pathogen–free vivarium. Experiments were conducted on 6- to 12-wk-old sized, purified, and assayed by mass spectroscopy (Integrated DNA Tech- male mice, in compliance with the Animal Welfare Act and the principles nologies). A total of 250 pmol of each primer template (Supplemental Table I) set forth in the Guide for the Care and Use of Laboratory Animals (38). All was used for 2 mg cDNA template for each reaction. Reactions were run as methods and protocols were approved by the Animal Care and Use individual primer pairs for individual animals. Data were analyzed by the Committee of the Johns Hopkins University. 2–DDCt method of Livak et al. (42), comparing threshold cycles first to Hprt Induction of EAM expression in individual animals, then DCt of target genes in naive WT BALB/c controls. Certain data are pseudolinearized to depict downregu- DD DD For the induction of EAM, we employed the myocarditogenic peptide lation (2– Ct , 1) as –2 Ct. MyHCa614–629 (Ac-SLKLMATLFSTYASAD) (39), commercially syn- thesized by 9-fluorenylmethoxycarbonyl chemistry and purified to a min- Hydroxyproline determination imum of 90% by HPLC (GenScript). On days 0 and 7, mice received s.c. Cardiac samples are weighed, then hydrolyzed in 12 N HCl overnight at axillary instillations of 50–100 mg MyHCa peptide emulsified in 614–629 120˚C. Samples are dried in 96-well format plates, then incubated with CFA (Sigma-Aldrich) supplemented to 5 mg/ml heat-killed Mycobacte- 50 mm Chloramine-T (Sigma-Aldrich), followed by 1 M dimethylamino- rium tuberculosis strain H37Ra (Difco). On day 0, mice also received 500 benzaldehyde (Sigma-Aldrich) against a 1–100 mg/ml standard curve of ng pertussis toxin i.p. (List Biologicals). hydroxyproline (Sigma-Aldrich). Samples are read at 570 nm, and values Coxsackie viral infection normalized to starting sample mass (43). Male 6- to 8-wk-old mice were maintained under pathogen-free con- Echocardiography ditions in the animal facility at the Johns Hopkins School of Medicine. Transthoracic echocardiography was performed using the Acuson Sequoia Mice were inoculated i.p. with 103 PFU heart-passaged coxsackievirus C256 ultrasonic imaging system (Siemens) with a 13 MHz transducer. B3 (CVB3) (containing CVB3 and cardiac myosin) diluted in sterile Conscious, depilated, previously trained mice were gently held in PBS on day 0. This autoimmune model of CVB3 myocarditis closely a supine position. The heart was imaged in the two-dimensional mode in resembles EAM (40). CVB3 (Nancy strain) was originally obtained from the parasternal short-axis view, and an M-mode cursor was positioned the American Type Culture Collection (Manassas, VA) and grown in Vero perpendicular to the interventricular septum and the left ventricular cells (American Type Culture Collection), as previously described (41). (LV) posterior wall at the level of the papillary muscles. From M-mode, Myocarditis was assessed at day 10 post infection, during the peak of acute the wall thicknesses and chamber dimensions were measured. For each inflammation. mouse, three to five values for each measurement were obtained and Assessment of EAM averaged for evaluation. The LV end-diastolic dimension, LV end- systolic dimension (LVESD), interventricular septal wall thickness Mice were most commonly evaluated for the development of EAM at the at end diastole (IVSD), and LV posterior wall thickness at end diastole peak of disease on day 21. Heart tissues were fixed in SafeFix (Thermo (LVPWTED) were measured from a frozen M-mode tracing. Fractional Fisher Scientific). Tissues were embedded longitudinally, and 5-mm step shortening [(FS), %], ejection fraction [(EF), %], LV mass, and relative sections were cut and stained with H&E, or Masson’s Trichrome Blue wall thickness were calculated from these parameters, as previously de- (Histoserv). Myocarditis severity was evaluated by microscopic approxi- scribed (44). 4040 IFN-g AND IL-17A IN THE REGULATION OF MYOCARDITIS

Statistics Multiple group comparisons were performed by ANOVA, followed by the Tukey–Kramer and Bonferroni posttest (StatPlus). Group survival was determined by Mantel–Cox log rank (GraphPad Prism). Asterisks denote statistically significant comparisons to WT.

Results Concurrent ablation of IFN-g and IL-17A reveals a protective effect of IL-17A in inflammatory autoimmune heart disease To test our original hypothesis that IL-17 signaling mediates the severe disease observed in IFN-g–deficient mice, we crossed an IL-17A-null allele to the IFN-g2/2 background, to generate IFN- g2/2IL-17A2/2 double-knockout (DKO) mice. EAM was induced in DKO mice, IFN-g2/2 single-knockout, IL-17A2/2 single- knockout, and WT controls by immunization with myocarditogenic peptide (MyHCa614–629) in CFA. As we previously reported, IFN- g2/2 animals developed severe EAM, and IL-17A2/2 animals developed EAM of similar severity to WT animals (13, 29). Unexpectedly, accelerated mortality was observed in IFN-g2/2 Downloaded from IL-17A2/2 animals, beginning at day 14 post immunization (Fig. 1A). In contrast, IFN-g2/2 mice died only at later timepoints, con- sistent with our previously published reports (28). Across experi- ments, between 50 and 100% of all immunized IFN-g2/2IL- 17A2/2 mice consistently died during this timeframe, in both male 2/2 2/2 and female IFN-g IL-17A animals (data not shown). The http://www.jimmunol.org/ survival of IFN-g2/2IL-17A2/2 mice was significantly poorer than IFN-g2/2 control animals by Kaplan–Meier log-rank esti- mates (p , 0.001). Moreover, IFN-g2/2IL-17A2/2 mice had more severe cardiac inflammation at time of death than did IFN-g2/2 or other control mice (Fig. 1B). The degree of disease severity in IFN-g2/2IL-17A2/2 mice was unprecedented for the EAM model. The majority of IFN-g2/2IL- 17A2/2 animals had such severe myocarditis that the tissue re- sembled secondary lymphoid tissue. Infiltrating cells included by guest on September 25, 2021 numerous mononuclear cells, as well as frequent polymorphonu- clear cells (Fig. 1C). We observed extensive pericardial inflam- mation and fibrosis, as well as extensive cardiomyocyte necrosis, , intra-atrial thrombosis, and the formation of ectopic lymphoid follicle-like structures throughout the myocardium (Supplemental Fig. 1). In parallel experiments, we investigated the response of IFN-g2/2 IL-17A2/2 mice (and appropriate single-knockout controls) to acute infection with CVB3, a cardiotropic enterovirus that has been previously described as an important agent in the elicitation of myocarditis in humans, as well as mouse models (45–47). Similar to the EAM model, IFN-g–deficient mice developed more severe cardiac inflammation following CVB3 infection (48). At day 10 following infection, 2 of 13 IFN-g2/2IL-17A2/2 mice had expired, and the remaining animals were moribund (Supplemental Fig. 2A), whereas all mice of the control geno- types survived to this timepoint. The severity of histopathologic cardiac inflammation was largely comparable across strains (Supplemental Fig. 2B, 2C). IFN-g2/2IL-17A2/2 mice rapidly develop rapidly compromised cardiac function, independent of cardiac fibrosis FIGURE 1. IL-17 accelerates lethal EAM on an IFN-g–deficient back- 2/2 2/2 2/2 g2/2 2/2 ground. (A)SurvivalofmaleIFN-g IL-17A (violet, n =15),IFN-g Echocardiographic imaging of IFN- IL-17A mice at day 2/2 14 demonstrated greatly compromised cardiac function and dra- (red, n = 14), IL-17A (blue, n = 5), and WT (gray, n =5)miceduring the first 28 d of EAM. (B) Cardiac histopathology of IFN-g2/2IL-17A2/2 matic loss of cardiac contractility, as evidenced by diminished FS 2/2 2/2 2/2 2/2 2/2 (violet), IFN-g (red), IL-17A (blue), and WT (open) mice at time of andEFinIFN-g IL-17A , but not IFN-g ,miceatthis death from (A). Prematurely expired mice from (A) are represented as early timepoint (Fig. 2A). We did not observe overt signs of LV crosses, whereas animals that survived to endpoint are depicted as diamonds. dilation (Supplemental Fig. 3); however, LV end-systolic dimen- (C) Representative histopathology from median animals of each group; 2/2 2/2 sions were increased in IFN-g IL-17A mice (Fig. 2B), as staining by H&E, magnification at 32.5 (left) and 3100 (right). Individual was the resulting calculated LV mass (Fig. 2C). data points represent individual animals; bars indicate mean of each group. The Journal of Immunology 4041

Cardiac fibrosis was assessed at day 21 of EAM by staining with Masson’s Trichrome Blue (Fig. 3A). Total collagen content of was assayed by quantitative determination of hydroxypro- line content, demonstrating diminished fibrosis in IFN-g2/2IL- 17A2/2 mice, compared with IFN-g2/2 mice (Fig. 3B). Transcrip- tional regulation of fibrotic collagens in IFN-g2/2IL-17A2/2 hearts was interrogated by real-time PCR. Similar patterns of expression were observed for collagens I and III (Fig. 3B, 3C), as well as matrix metalloproteinase 2 and matrix metalloproteinase 3 (Supplemental Fig. 3). Together, these data confirm our previous findings that IL- 17A drives cardiac fibrosis, whereas IFN-g counterregulates this remodeling process (13). However, it makes such fibrotic dilative remodeling unlikely to mediate the severe, morbid heart disease of IFN-g2/2IL-17A2/2 mice.

Cellular mediators of severe EAM in IFN-g2/2IL-17A2/2 mice To address mechanisms by which IL-17A could protect mice from eosinophilic myocarditis, we performed comprehensive cytometric profiling of the cardiac infiltrate at the peak of EAM in surviving Downloaded from mice (Fig. 4A). Among the IFN-g2/2IL-17A2/2 mice that sur- vived to day 21, the total number of CD45+ leukocytes was similar to that seen in IFN-g2/2 animals (Fig. 4B, left). Notably, in- creased infiltration was observed in IFN-g2/2 hearts at day 21 of EAM; however, this recruitment was abolished in 2/2 2/2 IFN-g IL-17A mice (Fig. 4B–D). http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 3. Cardiac remodeling in the EAM of IFN-g2/2IL-17A2/2 mice. (A) Representative histopathology of IFN-g2/2 (left) and IFN-g2/2 IL-17A2/2 (right) mice at day 21 of EAM, staining by Masson’s Tri- chrome Blue, 32.5 (left) and 3100 (right) magnification. (B) Total cardiac hydroxyproline content on days 12 (left) and 21 (right) of EAM, nor- malized to heart weight, expressed as parts per million (ppm). (C) Tran- scription of collagen I (Col1a2, left) and collagen III (Col3a1, right) at day 21 of EAM by real-time PCR. Individual data points represent individual animals; bars indicate mean of each group.

Eosinophils were most abundant in IFN-g2/2IL-17A2/2 hearts at day 21 (Fig. 4B–E). By examining the loss of light scattering of heart-infiltrating eosinophils as an indicator of degranulation (49), we observed more numerous degranulated eosinophils at day 12 of EAM in IFN-g2/2IL-17A2/2 mice (data not shown). Together, 2/2 2/2 these data led us to conclude that concurrent ablation of IL-17A FIGURE 2. IFN-g IL-17A mice rapidly develop inflammatory 2/2 cardiomyopathy, as determined by echocardiographic imaging at day 14 replaced the predominantly neutrophilic infiltrate of IFN-g of EAM. (A)%FS(left)and%EF(right); (B)LVESD;(C)LVmassin mice with eosinophils. IFN-g2/2IL-17A2/2, IFN-g2/2, IL-17A2/2, and WT mice. Individual data Quantitatively, the most substantial differences in cellular in- points represent individual animals; bars indicate mean of each group. filtration were observed among granuloid cell types, although 4042 IFN-g AND IL-17A IN THE REGULATION OF MYOCARDITIS Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 4. Composition of inflammatory leukocytic cardiac infiltrates in IFN-g2/2IL-17A2/2 mice at day 21 of EAM. (A) Distribution of infiltrating inflammatory leukocytic populations as normalized means of groups; areas of pie graphs approximately correspond to absolute numbers. Notable pop- ulations are highlighted. (B) Total intracardiac CD45+ leukocytes, Ly6GhiCD11b+ , Siglec F+Ly6Gint eosinophils, and FcεRIahi mast cells expressed as absolute numbers; (C) neutrophils and eosinophils expressed as a proportion of total CD45+ leukocytes in WT, IFN-g2/2, and IFN-g2/2IL- 17A2/2 mice. (D) Representative bivariate gating of cardiac granulocytes, depicting median individual of each group. (E) Representative cardiac eosinophil staining in an IFN-g2/2IL-17A2/2 animal at day 21 of EAM, by Verfuerth-Luna (74), 3400 magnification. All cells are gated on viable CD45+ events. Individual data points represent individual animals; bars indicate mean of each group. we also found differences between IFN-g2/2IL-17A2/2 mice and mice, we did not observe differences in the number of eosino- IFN-g2/2 animals in infiltration with other cell types. IFN-g2/2 phils, regulatory T cells, or dendritic cells (data not shown). We IL-17A2/2 hearts tended to be more infiltrated with F4/80+ also did not find differences in numbers of effector T cells, NK CD11b+ macrophages, as well as FcεRIahi mast cells (Fig. 4B, cells, gd T cells, monocytes, or B cells (data not shown). In the data not shown). peripheral blood of naive IFN-g2/2IL-17A2/2 mice, we observed We examined the spleens and blood of naive IFN-g2/2IL-17A2/2 a modest increase in the proportion of eosinophils. Upon mock mice to determine whether these differences in infiltrating leukocyte immunization with PBS/CFA, IFN-g2/2IL-17A2/2 mice did not populations represent pre-existing disparities in the frequencies develop peripheral eosinophilia or eosinophilic cardiac infiltrates of these cell subsets prior to the induction of EAM. Comparing (data not shown). These data indicate that IFN-g2/2IL-17A2/2 the spleens of unimmunized IFN-g2/2IL-17A2/2 and IFN-g2/2 mice do not exhibit signs of general systemic eosinophilia. From The Journal of Immunology 4043 this, we conclude that eosinophilic infiltration of the hearts of IFN-g2/2IL-17A2/2 mice during EAM is specifically autoimmune, and not a generalized inflammatory disease of the IFN-g2/2IL- 17A2/2 strain. Eosinophilic mediators of severe EAM in IFN-g2/2IL-17A2/2 mice To understand the differences in recruitment of granulocytic effectors under coordinate control by IFN-g and IL-17A, we ex- amined the expression of a variety of cytokines, chemokines, and growth factors in the hearts and autoreactive T cells of IFN-g2/2IL- 17A2/2 and IFN-g2/2 mice with EAM. As expected, chemokines, cytokines, and growth factors associated with Th17-dependent recruitment of neutrophils were largely downregulated in the ab- sence of IL-17, in a manner that did not track with cardiac disease (data not shown). Eosinophil recruitment to sites of inflammation is largely me- diated by chemokine utilization of the receptor CCR3 by the ligand

CCL11/eotaxin-1 (50). Interrogation of cardiac homogenates of Downloaded from IFN-g2/2IL-17A2/2 hearts during EAM demonstrated substan- tially greater production of CCL11/eotaxin-1 at day 12, and CCL24/eotaxin-2, compared with WT controls (p , 0.001), at day 21 (Fig. 5A). No differences were detected in CCL26/eotaxin-3 expression at day 12; no expression was detectable at day 21 (data

not shown). Altogether, these data indicate that IFN-g and IL-17A http://www.jimmunol.org/ collaborated in suppressing eosinophilic recruitment to the heart, via suppression of CCL11/eotaxin-1 and CCL24/eotaxin-2 pro- duction. We did not detect production of CCL11/eotaxin-1 from restimulated splenocyte cultures (data not shown), indicating its likely production by a nonhematopoietic, heart-resident cell type. Autoaggressive Th2 deviation in severe EAM in IFN-g2/2 IL-17A2/2 mice

Elevations in the expression of eotaxin CCR3 ligands implied by guest on September 25, 2021 + deviation of autoreactive CD4 T cells to a Th2 phenotype. To FIGURE 5. Cardiac and spleen production of eosinotropic cytokines and 2/2 2/2 + further examine whether autoreactive IFN-g IL-17A CD4 chemokines. (A) Cardiac transcription of CCL11/eotaxin-1 at day 12 and cells were preferentially differentiating into Th2 cells at the CCL24/eotaxin-2 at day 21 of EAM by real-time PCR. (B) Absolute single-cell level, we performed intracellular staining for cytokines enumeration of heart-infiltrating IL-4+IL-13+ double-positive CD4+CD3ε+ of heart-infiltrating CD4+ T cells at day 12 of EAM. CD4+ T cells Th2 cells at day 12 of EAM. (C) Representative bivariate plots of IL-4 and + + + staining for both IL-4 and IL-13 were increased in the hearts of IL-13 staining of viable intracardiac CD4 CD3ε CD45 -gated lympho- IFN-g2/2IL-17A2/2 mice, compared with the single-knockout cytes, depicting median individual of each group. Individual data points represent individual animals; bars indicate mean of each group. controls, further confirming the expansion of autoaggressive CD4+ cells with a Th2 phenotype in the combined absence of IFN-g and IL-17A (Fig. 5B, 5C). Together, these data indicate that autore- active CD4+ cells are constrained from Th2 differentiation by a binding site in the promoter of GATA1. This mutation selectively synergistic action of IFN-g and IL-17. Intriguingly, this suppres- ablates differentiation of eosinophils, while leaving erythroid line- sive function appears to be preferentially targeted toward eosin- ages unaffected (51). The resulting triple-mutant DdblGATA1 3 IFN- ophil-tropic functions of Th2 cells. It further implicates this g2/2IL-17A2/2 mice showed fewer Siglec F+Ly6Gint eosinophils in eosinophilic Th2 deviation in the early mortality of IFN-g2/2IL- peripheral blood (Fig. 6A). 17A2/2 EAM. In contrast, IL-5 production in cardiac homoge- Upon induction of EAM, eosinophil-deficient triple-mutant mice nates, cardiac transcripts, and restimulated splenocytes did not were protected from the fatal heart failure observed in IFN-g2/2IL- track with eosinophilia or disease in a cogent manner (data not 17A2/2 mice (Fig. 6B). This protection was incomplete, as 5 of 27 shown). From these findings, we conclude that the collaborative triple-mutant animals died before the termination of the experi- suppression of cardiac eosinophilia by IFN-g and IL-17A is medi- ment on day 21; this survival rate approximately matches that of ated through eosinotropic chemokines that may be Th2 dependent, IFN-g2/2 animals (Fig. 1A, data not shown). Statistically com- rather than through local or systemic expansion of eosinophils by paring DdblGATA1 3 IFN-g2/2IL-17A2/2 and IFN-g2/2IL-17A2/2 IL-5. mice, Mantel–Cox log-rank survival estimates were significantly different (p = 0.0039). We observed similar low rates of mortality Genetic ablation of eosinophils reverses the mortality of 3 2/2 2/2 2/2 2/2 in either male or female DdblGATA1 IFN-g IL-17A mice IFN-g IL-17A EAM (data not shown). Surprisingly, the severity of cardiac enlargement To demonstrate conclusively that eosinophils acted as requisite or inflammation did not differ between IFN-g2/2IL-17A2/2 and effectors of the severe, fatal cardiac disease in the EAM of DdblGATA1 3 IFN-g2/2IL-17A2/2 mice (Fig. 6C, 6D). IFN-g2/2IL-17A2/2 mice, we crossed onto this background Cardiac function was assessed at day 15 by echocardiographic a mutated allele bearing a deletion of the high-affinity double-GATA imaging. Consistent with their greater survival, DdblGATA1 3 4044 IFN-g AND IL-17A IN THE REGULATION OF MYOCARDITIS Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 6. Genetic ablation of eosinophil development reverses the mortality of IFN-g2/2IL-17A2/2 mice. (A) Established IFN-g2/2IL-17A2/2 BALB/c mice were outcrossed to the DdblGATA1 BALB/c strain (JAX 5653), and resulting progeny intercrossed for a minimum of two generations, to generate the homozygous triple-mutant founders depicted in representative bivariate plots of CD45+-gated peripheral blood granulocytes from naive DdblGATA1 3 IFN-g2/2IL-17A2/2 mice (orange) and control strains; as well as quantitative analyses of Siglec F+Ly6Gint eosinophils. (B) Survival of male IFN-g2/2IL-17A2/2 (violet), WT BALB/c (gray, dashed), and DdblGATA1 3 IFN-g2/2IL-17A2/2 mice (orange, dashed) mice during the first 21 d of EAM. Data represent a composite of two independent experiments. (C) Relative heart weight at day 21 (left) and cardiac histopathology (right)of DdblGATA1 3 IFN-g2/2IL-17A2/2 and control mice at time of death. Prematurely expired mice from (A) are represented as crosses, whereas animals that survived to endpoint are depicted as diamonds. (D) Representative histopathology of median individual animals; staining by H&E, magnification at 32.5 (left) and 3100 (right). (E) Echocardiographic imaging of DdblGATA1 3 IFN-g2/2IL-17A2/2 and control mice on day 15 of EAM. (F) Linear regressive correlation of LVPWTED and survival. Individual data points represent individual animals; bars indicate mean of each group.

IFN-g2/2IL-17A2/2 mice had substantially improved cardiac 6F). Together, these data demonstrate that eosinophils are requisite function at this timepoint (Fig. 6E). Functional improvement of FS mediators of the fatally severe myocarditis of IFN-g2/2IL-17A2/2 and EF were largely due to diminished wall thickness parameters mice, by driving inflammatory hypertrophic cardiomyopathy. (IVSD, LVPWTED, LV mass, and relative wall thickness), rather than signs of ventricular dilation. Consistent with this observation, Discussion among the primary measurements in M-mode echocardiography We report in this article that in concert with IFN-g, IL-17A exerts (LVend-diastolic dimension, LVESD, IVSD, LVPWTED, and heart a protective effect in eosinophilic heart disease. IL-17 has been rate), the thickness of the LV posterior wall correlated best with the ascribed protective functions in various other inflammatory dis- improved survival of DdblGATA1 3 IFN-g2/2IL-17A2/2 mice (Fig. eases, including CD45RBhi transfer colitis (52), dextran The Journal of Immunology 4045 sulfate sodium colitis (53), and uveitis secondary to spondy- largely restored cardiac function, underscoring the important role loarthropathy (54). These results indicate that, in addition to an- of the eosinophil in mediating the death of IFN-g2/2IL-17A2/2 atomic considerations, IL-17A has divergent pathophysiologic mice. We are further undertaking studies using triple-mutant mice functions, depending on the context of other cytokines. to examine specific eosinophil chemotactic and effector pathways The paradoxical protective effect of IFN-g in several models of in mediating cardiotoxic mortality. autoimmune disease provides evidence that IFN-g can suppress We interrogated a number of biomarkers to assess alternative disease, contrary to the expected proinflammatory role of Th1 deviation pathways that may contribute to the severe EAM of cells. This paradox had seemingly been resolved by the discovery IFN-g2/2IL-17A2/2 mice. We did not find compelling evidence of the Th17 lineage. IL-23p19–deficient animals were protected for Tregs, Th9, nonclassical Th17, or Th22 cells contributing to from EAE, whereas IL-12p35–deficient animals developed disease the disease phenotype of DKO mice (data not shown). We also did comparable to that of controls (21). not find persuasive evidence of mast cell involvement in the dis- We conclude from our prior experiments that the Th17 lineage ease of IFN-g2/2IL-17A2/2 mice. Despite the fact that more mast prompts the pathogenesis of EAM. IL-17A is the prototypic cells were detected in IFN-g2/2IL-17A2/2 hearts, levels of in- product of Th17 cells, which differentiate through a process re- tracardiac histamine and serum IgE did not track with disease quiring TGF-b, IL-6, and IL-23 (17, 18, 55). Although IL-17A is (data not shown). Although these data do not definitively rule out not essential for the inflammatory process of EAM, it instead the participation of these pathways, they emphasize that eosino- drives cardiac remodeling, fibrosis, and DCM development (13). philic Th2 deviation is primarily responsible for the severe, fatal In WT mice, both Th1 and Th17 cells infiltrate the heart during disease of IFN-g2/2IL-17A2/2 mice.

EAM, but Th17 cells produce higher levels of inflammatory Among the types of myocarditis seen in patients, eosinophilic Downloaded from cytokines, such as TNF-a, that have previously been shown to necrotizing myocarditis has one of the worst prognoses, often drive cardiac inflammation (13). In addition, we have found that leading to rapid cardiac failure and death (6, 67, 68). Most often, Th17-polarized cells are sufficient to transfer myocarditis (G.C. eosinophilic myocarditis in patients is seen as a sequela of Baldeviano and D. Ciha ´kova´, unpublished observations). Our data eosinophilia-associated systemic disease, in which infiltration remain consistent with the idea that Th17 cells mediate the severe is typically perivascular, with less profound damage to car- 2/2 disease of IFN-g animals, as evidenced by enhanced neutro- diomyocytes and better prognosis (69). However, necrotizing http://www.jimmunol.org/ phil infiltration, as well as increased cardiac production of G-CSF eosinophilic myocarditis may not be associated with increased and IL-6 in IFN-g2/2 mice (data not shown). numbers of peripheral eosinophils in the periphery. It progresses The myocarditic phenotype observed in IFN-g2/2IL-17A2/2 quickly, with severe eosinophilic infiltration in the myocardium, mice is the most severe yet reported, even more so than the EAM severe cardiomyocyte necrosis, intraventricular thrombi, and poor of IFN-g2/2, IFN-gRI2/2, or IL-132/2 animals (28, 44, 56). The prognosis (69–71). All of these features were observed as part of rapid mortality begins at approximately day 14, shortly after the the necrotizing eosinophilic myocarditis in IFN-g2/2IL-17A2/2 emergence of inflammatory cardiac lesions in WT mice, consis- mice. tent with the myocarditic process being the cause of death in In a broader context, we can further envision that autoimmune IFN-g2/2IL-17A2/2 animals. At this timepoint, we observed a diseases, as a whole, may be more heterogeneous than the Th17- by guest on September 25, 2021 greater proportion of degranulating eosinophils, continuing to ex- centered view popularized since the discovery of the cell lineage. tensive eosinophilic infiltrates by day 21, at which point eosi- Th1 cells have been described as potentiating autoaggressive ef- nophils constitute roughly a third of infiltrating cells. Most impor- fector functions in several models, albeit to a generally lesser tantly, concurrent developmental ablation of eosinophils improved degree than Th17 cells (22, 24, 72). Clearly, the pathogenicity of survival of IFN-g2/2IL-17A2/2 mice. various cell lineages is linked to the location and inflammatory Upstream of this fatal eosinophilic recruitment, we observed context of the model system in question. Our data argue for sub- enhanced Th2 deviation in IFN-g2/2IL-17A2/2 mice. Th2 devi- stantially greater redundancy among autoaggressive effector T cell ation in IFN-g2/2 mice corresponds to the classical Th1/Th2 subsets than has been previously appreciated. model (57). We have found novel evidence that IL-17 signaling It remains to be seen whether this redundancy is true for other collaborates with IFN-g in the suppression of production of IL-4 autoimmune disease models. IFN-g2/2IL-17A2/2 mice were not and IL-13 by autoaggressive Th2 cells. Although other Th2 and protected from EAE, nor did they develop accelerated disease eosinophilic autoimmune disease models have been described (73). This fundamental disjoint between EAE and EAM may re- (58–65), to our knowledge this is the first that has been elicited by side in the fact that the CNS lies behind the blood–brain barrier an intrinsically deviated response. Moreover, these data conclusively and is historically regarded as a site of immune privilege. It is demonstrate a primary role for eosinophils in eliciting the rapid evident that different anatomic compartments possess different mortality of this deviated autoimmune response. The transcrip- mechanisms to suppress inflammatory responses. Our data suggest tion factor eomesodermin has recently been reported to be that what we view as a predominantly Th1/Th17 response may be a critical regulator of this regulatory hierarchy among the Th2 beneficial, insofar as it suppresses a potentially more devastating cytokines, by selectively suppressing targeting of the Th2 tran- outcome. Because eosinophilic myocarditis is recognized as a scription GATA3 to the Il5 locus in memory Th2 cells (66). These distinct clinical entity with a particularly poor prognosis, our data provide novel evidence that IFN-g and IL-17A are selec- findings indicate that shared etiologic mechanisms may result in tively collaborative in controlling eosinophilic effector functions a qualitatively diverse spectrum of disease outcomes, depending of the Th2 response. on T cell polarization and the downstream effector mechanisms Importantly, we found evidence for pathophysiologic con- involved. tributions of eosinophils that directly affect survival, which seg- These findings may have further implications for the deviation regate away from effects on net cardiac inflammation. Eosinophil- strategies entering clinical use as the first generation of IL-12/ deficient DdblGATA1 3 IFN-g2/2IL-17A2/2 mice developed 23p40 antagonists are deployed for use in treating autoimmune severe cardiac infiltrates and inflammation, comparable to that in disease. Caution may be warranted in the investigational use IFN-g2/2IL-17A2/2 mice, but did not die. Importantly, this pro- of these drugs in diseases with the potential for eosinophilic tection observed in the triple-mutant genotype was associated with deviation. 4046 IFN-g AND IL-17A IN THE REGULATION OF MYOCARDITIS

Acknowledgments 26. Hu, X., and L. B. Ivashkiv. 2009. Cross-regulation of signaling pathways by interferon-gamma: implications for immune responses and autoimmune dis- We thank R. Lee Blosser and Ada Tam for expert assistance with flow cyto- eases. Immunity 31: 539–550. metric analysis methods; Nicole Barat and the laboratories of J. Steven 27. Eriksson, U., M. O. Kurrer, W. Sebald, F. Brombacher, and M. Kopf. 2001. Dual Dumler for assistance with multiplex cytokine analyses; Norman J. Barker role of the IL-12/IFN-gamma axis in the development of autoimmune myocar- for expert assistance with photomicrography; Yoichiro Iwakura for the gen- ditis: induction by IL-12 and protection by IFN-gamma. J. Immunol. 167: 5464– 2/2 5469. erous provision of IL-17A founder animals; and Patrizio Caturegli, 28. Afanasyeva, M., D. Georgakopoulos, D. F. Belardi, D. Bedja, D. Fairweather, Mark Lindsay, and Abdel Hamad for critical discussion and reading. Y. Wang, Z. Kaya, K. L. Gabrielson, E. R. Rodriguez, P. Caturegli, et al. 2005. Impaired up-regulation of CD25 on CD4+ T cells in IFN-gamma knockout mice is associated with progression of myocarditis to heart failure. Proc. Natl. Acad. Disclosures Sci. USA 102: 180–185. The authors have no financial conflicts of interest. 29. Afanasyeva, M., Y. Wang, Z. Kaya, E. A. Stafford, K. M. Dohmen, A. A. Sadighi Akha, and N. R. Rose. 2001. Interleukin-12 receptor/STAT4 signaling is required for the development of autoimmune myocarditis in mice by an interferon- References gamma-independent pathway. Circulation 104: 3145–3151.  30. Barin, J. G., M. V. Talor, G. C. Baldeviano, M. Kimura, N. R. Rose, and 1. Ciha´kova´, D., and N. R. Rose. 2008. Pathogenesis of myocarditis and dilated D. Ciha ´kova´. 2010. Mechanisms of IFNg regulation of autoimmune myocarditis. cardiomyopathy. Adv. Immunol. 99: 95–114. Exp. Mol. Pathol. 89: 83–91. 2. Andre´oletti, L., N. Le´veˆque, C. Boulagnon, C. Brasselet, and P. Fornes. 2009. 31. Berghmans, N., A. Nuyts, C. Uyttenhove, J. Van Snick, G. Opdenakker, and Viral causes of human myocarditis. Arch. Cardiovasc. Dis. 102: 559–568. H. Heremans. 2011. Interferon-g orchestrates the number and function of Th17 3. Afanasyeva, M., and N. R. Rose. 2004. Viral infection and heart disease: auto- cells in experimental autoimmune encephalomyelitis. J. Interferon Cytokine Res. immune mechanisms. In Infection and Autoimmunity. Y. Shoenfeld, and 31: 575–587. N. R. Rose, eds. Elsevier, Amsterdam, p. 299–318. 32. Qian, X., H. Ning, J. Zhang, D. F. Hoft, D. J. Stumpo, P. J. Blackshear, and 4. Rose, N. R. 2005. The Significance of Autoimmunity in Myocarditis. In Chronic

J. Liu. 2011. Posttranscriptional regulation of IL-23 expression by IFN-gamma Downloaded from Viral and Inflammatory Cardiomyopathy. H.-P. Schultheiss, J.-F. Kapp, and through tristetraprolin. J. Immunol. 186: 6454–6464. G. Gro¨tzbach, eds. Springer Verlag, Du¨sseldorf, Germany, p. 141–154. 33. Nakae, S., Y. Iwakura, H. Suto, and S. J. Galli. 2007. Phenotypic differences 5.Eckart,R.E.,S.L.Scoville,C.L.Campbell,E.A.Shry,K.C.Stajduhar, between Th1 and Th17 cells and negative regulation of Th1 cell differentiation R. N. Potter, L. A. Pearse, and R. Virmani. 2004. Sudden death in young adults: a by IL-17. J. Leukoc. Biol. 81: 1258–1268. 25-year review of autopsies in military recruits. Ann. Intern. Med. 141: 829–834. 34. Koike, T. 2012. Interferon-g-independent suppression of Th17 cell differentia- 6. Ginsberg, F., and J. E. Parrillo. 2005. Eosinophilic myocarditis. Heart Fail. Clin. tion by T-bet expression in mice with autoimmune arthritis. Arthritis Rheum. 64: 1: 419–429. 40–41. 7. Cooper, L. T., Jr. 2009. Myocarditis. N. Engl. J. Med. 360: 1526–1538. 35. Kelchtermans, H., E. Schurgers, L. Geboes, T. Mitera, J. Van Damme, J. Van 8. Magnani, J. W., and G. W. Dec. 2006. Myocarditis: current trends in diagnosis http://www.jimmunol.org/ Snick, C. Uyttenhove, and P. Matthys. 2009. Effector mechanisms of interleukin-17 and treatment. Circulation 113: 876–890. in collagen-induced arthritis in the absence of interferon-gamma and counter- 9. Ciha´kova´, D., R. B. Sharma, D. Fairweather, M. Afanasyeva, and N. R. Rose. action by interferon-gamma. Arthritis Res. Ther. 11: R122. 2004. Animal models for autoimmune myocarditis and autoimmune thyroiditis. 36. Nakae, S., Y. Komiyama, A. Nambu, K. Sudo, M. Iwase, I. Homma, Methods Mol. Med. 102: 175–193. K. Sekikawa, M. Asano, and Y. Iwakura. 2002. -specific T cell sensiti- 10. Neu, N., N. R. Rose, K. W. Beisel, A. Herskowitz, G. Gurri-Glass, and S. W. Craig. zation is impaired in IL-17-deficient mice, causing suppression of allergic cel- 1987. Cardiac myosin induces myocarditis in genetically predisposed mice. J. lular and humoral responses. Immunity 17: 375–387. Immunol. 139: 3630–3636. 37. Dalton, D. K., S. Pitts-Meek, S. Keshav, I. S. Figari, A. Bradley, and T. A. Stewart. 11. Smith, S. C., and P. M. Allen. 1991. Myosin-induced acute myocarditis is a T cell-mediated disease. J. Immunol. 147: 2141–2147. 1993. Multiple defects of immune cell function in mice with disrupted interferon- 12. Weaver, C. T., R. D. Hatton, P. R. Mangan, and L. E. Harrington. 2007. IL-17 gamma genes. Science 259: 1739–1742. family cytokines and the expanding diversity of effector T cell lineages. Annu. 38. Care, C. U. G. t., U. L. Animals, and N. R. Council. 2011. Guide for the Care Rev. Immunol. 25: 821–852. and Use of Laboratory Animals, 8th Ed. The National Academies Press, by guest on September 25, 2021 13. Baldeviano, G. C., J. G. Barin, M. V. Talor, S. Srinivasan, D. Bedja, D. Zheng, Washington, D.C. K. Gabrielson, Y. Iwakura, N. R. Rose, and D. Ciha ´kova´. 2010. Interleukin-17A 39. Pummerer, C. L., K. Luze, G. Gra¨ssl, K. Bachmaier, F. Offner, S. K. Burrell, is dispensable for myocarditis but essential for the progression to dilated car- D. M. Lenz, T. J. Zamborelli, J. M. Penninger, and N. Neu. 1996. Identification diomyopathy. Circ. Res. 106: 1646–1655. of cardiac myosin peptides capable of inducing autoimmune myocarditis in 14. Kolls, J. K., and A. Linde´n. 2004. Interleukin-17 family members and inflam- BALB/c mice. J. Clin. Invest. 97: 2057–2062. mation. Immunity 21: 467–476. 40. Fairweather, D., K. A. Stafford, and Y. K. Sung. 2012. Update on coxsackievirus 15. Iwakura, Y., H. Ishigame, S. Saijo, and S. Nakae. 2011. Functional specialization B3 myocarditis. Curr. Opin. Rheumatol. 24: 401–407. of interleukin-17 family members. Immunity 34: 149–162. 41. Fairweather, D., and N. R. Rose. 2007. Coxsackievirus-induced myocarditis in 16. Gaffen, S. L. 2009. Structure and signalling in the IL-17 receptor family. Nat. mice: a model of autoimmune disease for studying immunotoxicity. Methods 41: Rev. Immunol. 9: 556–567. 118–122. 17. Harrington, L. E., R. D. Hatton, P. R. Mangan, H. Turner, T. L. Murphy, 42. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression K. M. Murphy, and C. T. Weaver. 2005. Interleukin 17-producing CD4+ effector data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. T cells develop via a lineage distinct from the T helper type 1 and 2 lineages. Methods 25: 402–408. Nat. Immunol. 6: 1123–1132. 43. Edwards, C. A., and W. D. O’Brien, Jr. 1980. Modified assay for determination of hydroxyproline in a tissue hydrolyzate. Clin. Chim. Acta 104: 161–167. 18. Park, H., Z. Li, X. O. Yang, S. H. Chang, R. Nurieva, Y. H. Wang, Y. Wang,  L. Hood, Z. Zhu, Q. Tian, and C. Dong. 2005. A distinct lineage of CD4 T cells 44. Ciha´kova´, D., J. G. Barin, M. Afanasyeva, M. Kimura, D. Fairweather, M. Berg, regulates tissue inflammation by producing interleukin 17. Nat. Immunol. 6: M. V. Talor, G. C. Baldeviano, S. Frisancho, K. Gabrielson, et al. 2008. 1133–1141. Interleukin-13 protects against experimental autoimmune myocarditis by regu- 19. Veldhoen, M., R. J. Hocking, C. J. Atkins, R. M. Locksley, and B. Stockinger. lating macrophage differentiation. Am. J. Pathol. 172: 1195–1208. 2006. TGFbeta in the context of an inflammatory cytokine milieu supports de 45. Fairweather, D., S. Frisancho-Kiss, and N. R. Rose. 2005. Viruses as adjuvants novo differentiation of IL-17-producing T cells. Immunity 24: 179–189. for autoimmunity: evidence from Coxsackievirus-induced myocarditis. Rev. 20. Ivanov, I. I., B. S. McKenzie, L. Zhou, C. E. Tadokoro, A. Lepelley, J. J. Lafaille, Med. Virol. 15: 17–27. D. J. Cua, and D. R. Littman. 2006. The orphan nuclear receptor RORgammat 46. Fairweather, D., Z. Kaya, G. R. Shellam, C. M. Lawson, and N. R. Rose. 2001. directs the differentiation program of proinflammatory IL-17+ T helper cells. From infection to autoimmunity. J. Autoimmun. 16: 175–186. Cell 126: 1121–1133. 47. Fairweather, D., S. Frisancho-Kiss, S. A. Yusung, M. A. Barrett, S. E. Davis, 21. Bettelli, E., Y. Carrier, W. Gao, T. Korn, T. B. Strom, M. Oukka, H. L. Weiner, R. A. Steele, S. J. Gatewood, and N. R. Rose. 2005. IL-12 protects against and V. K. Kuchroo. 2006. Reciprocal developmental pathways for the generation coxsackievirus B3-induced myocarditis by increasing IFN-gamma and macro- of pathogenic effector TH17 and regulatory T cells. Nature 441: 235–238. phage and neutrophil populations in the heart. J. Immunol. 174: 261–269. 22. Luger, D., P. B. Silver, J. Tang, D. Cua, Z. Chen, Y. Iwakura, E. P. Bowman, 48. Fairweather, D., S. Frisancho-Kiss, S. A. Yusung, M. A. Barrett, S. E. Davis, N. M. Sgambellone, C. C. Chan, and R. R. Caspi. 2008. Either a Th17 or a Th1 S. J. Gatewood, D. B. Njoku, and N. R. Rose. 2004. Interferon-gamma protects effector response can drive autoimmunity: conditions of disease induction affect against chronic viral myocarditis by reducing mast cell degranulation, fibrosis, dominant effector category. J. Exp. Med. 205: 799–810. and the profibrotic cytokines transforming growth factor-beta 1, interleukin-1 23. Stummvoll, G. H., R. J. DiPaolo, E. N. Huter, T. S. Davidson, D. Glass, beta, and interleukin-4 in the heart. Am. J. Pathol. 165: 1883–1894. J. M. Ward, and E. M. Shevach. 2008. Th1, Th2, and Th17 effector T cell- 49. Shinkai, K., M. Mohrs, and R. M. Locksley. 2002. Helper T cells regulate type-2 induced autoimmune gastritis differs in pathological pattern and in susceptibil- innate immunity in vivo. Nature 420: 825–829. ity to suppression by regulatory T cells. J. Immunol. 181: 1908–1916. 50. Rothenberg, M. E., and S. P. Hogan. 2006. The eosinophil. Annu. Rev. Immunol. 24. Ja¨ger, A., V. Dardalhon, R. A. Sobel, E. Bettelli, and V. K. Kuchroo. 2009. Th1, 24: 147–174. Th17, and Th9 effector cells induce experimental autoimmune encephalomy- 51. Yu, C., A. B. Cantor, H. Yang, C. Browne, R. A. Wells, Y. Fujiwara, and elitis with different pathological phenotypes. J. Immunol. 183: 7169–7177. S. H. Orkin. 2002. Targeted deletion of a high-affinity GATA-binding site in the 25. Damsker, J. M., A. M. Hansen, and R. R. Caspi. 2010. Th1 and Th17 cells: GATA-1 promoter leads to selective loss of the eosinophil lineage in vivo. J. Exp. adversaries and collaborators. Ann. N. Y. Acad. Sci. 1183: 211–221. Med. 195: 1387–1395. The Journal of Immunology 4047

52. O’Connor, W., Jr., M. Kamanaka, C. J. Booth, T. Town, S. Nakae, Y. Iwakura, 63. Cookston, M., M. Stober, and S. G. Kayes. 1990. Eosinophilic myocarditis in J. K. Kolls, and R. A. Flavell. 2009. A protective function for interleukin 17A in CBA/J mice infected with Toxocara canis. Am. J. Pathol. 136: 1137–1145. T cell-mediated intestinal inflammation. Nat. Immunol. 10: 603–609. 64. Schaffer, S. W., E. R. Dimayuga, and S. G. Kayes. 1992. Development and 53. Ogawa, A., A. Andoh, Y. Araki, T. Bamba, and Y. Fujiyama. 2004. Neutrali- characterization of a model of eosinophil-mediated cardiomyopathy in rats zation of interleukin-17 aggravates dextran sulfate sodium-induced colitis in infected with Toxocara canis. Am. J. Physiol. 262: H1428–H1434. mice. Clin. Immunol. 110: 55–62. 65. Hirasawa, M., Y. Ito, M. A. Shibata, and Y. Otsuki. 2007. Mechanism of in- 54. Kezic, J. M., T. T. Glant, J. T. Rosenbaum, and H. L. Rosenzweig. 2012. Neu- flammation in murine eosinophilic myocarditis produced by adoptive transfer tralization of IL-17 ameliorates uveitis but damages photoreceptors in a murine with ovalbumin challenge. Int. Arch. Allergy Immunol. 142: 28–39. model of spondyloarthritis. Arthritis Res. Ther. 14: R18. 66. Endo, Y., C. Iwamura, M. Kuwahara, A. Suzuki, K. Sugaya, D. J. Tumes, 55. Veldhoen, M., R. J. Hocking, R. A. Flavell, and B. Stockinger. 2006. Signals K. Tokoyoda, H. Hosokawa, M. Yamashita, and T. Nakayama. 2011. Eomeso- mediated by transforming growth factor-beta initiate autoimmune encephalo- dermin controls interleukin-5 production in memory T helper 2 cells through myelitis, but chronic inflammation is needed to sustain disease. Nat. Immunol. 7: inhibition of activity of the transcription factor GATA3. Immunity 35: 733–745. 1151–1156. 67. Rezaizadeh, H., M. Sanchez-Ross, E. Kaluski, M. Klapholz, B. Haider, and 56. Eriksson, U., M. O. Kurrer, R. Bingisser, H. P. Eugster, P. Saremaslani, C. Gerula. 2010. Acute eosinophilic myocarditis: diagnosis and treatment. Acute F. Follath, S. Marsch, and U. Widmer. 2001. Lethal autoimmune myocarditis in Card. Care 12: 31–36. interferon-gamma receptor-deficient mice: enhanced disease severity by im- 68. Al Ali, A. M., L. P. Straatman, M. F. Allard, and A. P. Ignaszewski. 2006. Eo- paired inducible nitric oxide synthase induction. Circulation 103: 18–21. sinophilic myocarditis: case series and review of literature. Can. J. Cardiol. 22: 57. Mosmann, T. R., and R. L. Coffman. 1989. TH1 and TH2 cells: different patterns 1233–1237. of lymphokine secretion lead to different functional properties. Annu. Rev. 69. Cooper, L. T., and K. J. Zehr. 2005. Biventricular assist device placement and Immunol. 7: 145–173. immunosuppression as therapy for necrotizing eosinophilic myocarditis. Nat. 58. Khan, S., and S. R. Orenstein. 2008. Eosinophilic gastroenteritis. Gastroenterol. Clin. Pract. Cardiovasc. Med. 2: 544–548. Clin. North Am. 37: 333–348, v (v.). 70. Roehrl, M. H., M. P. Alexander, S. B. Hammond, M. Ruzinova, J. Y. Wang, and 59. Nelson, A. M. 1996. Localized forms of scleroderma, including morphea, C. J. O’Hara. 2011. Eosinophilic myocarditis in hypereosinophilic syndrome. linear scleroderma, and eosinophilic fasciitis. Curr. Opin. Rheumatol. 8: Am. J. Hematol. 86: 607–608. 473–476. 71. Gupta, S., D. W. Markham, M. H. Drazner, and P. P. Mammen. 2008. Fulminant 60. Alfadda, A. A., M. A. Storr, and E. A. Shaffer. 2011. Eosinophilic colitis: an myocarditis. Nat. Clin. Pract. Cardiovasc. Med. 5: 693–706. Downloaded from update on pathophysiology and treatment. Br. Med. Bull. 100: 59–72. 72. Domingues, H. S., M. Mues, H. Lassmann, H. Wekerle, and G. Krishnamoorthy. 61. Afanasyeva, M., Y. Wang, Z. Kaya, S. Park, M. J. Zilliox, B. H. Schofield, 2010. Functional and pathogenic differences of Th1 and Th17 cells in experi- S. L. Hill, and N. R. Rose. 2001. Experimental autoimmune myocarditis in A/J mental autoimmune encephalomyelitis. PLoS ONE 5: e15531. mice is an interleukin-4-dependent disease with a Th2 phenotype. Am. J. Pathol. 73. Codarri, L., G. Gyu¨lve´szi, V. Tosevski, L. Hesske, A. Fontana, L. Magnenat, 159: 193–203. T. Suter, and B. Becher. 2011. RORgt drives production of the cytokine GM-CSF 62. Abston, E. D., J. G. Barin, D. Ciha ´kova´, A. Bucek, M. J. Coronado, J. E. Brandt, in helper T cells, which is essential for the effector phase of autoimmune neu- D. Bedja, J. B. Kim, D. Georgakopoulos, K. L. Gabrielson, et al. 2012. IL-33 roinflammation. Nat. Immunol. 12: 560–567.

independently induces eosinophilic and cardiac dilation: ST2 improves 74. Luna, L. 1968. Manual of Histologic Staining Methods of the Armed Forces http://www.jimmunol.org/ cardiac function. Circ Heart Fail 5: 366–375. Institute of Pathology. McGraw-Hill, New York. by guest on September 25, 2021