Regulatory T Cells

Control of Autoimmune Myocarditis and Multiorgan Inflammation by Glucocorticoid-Induced TNF Receptor Family-Related Protein high, This information is current as Foxp3-Expressing CD25 + and CD25− of September 29, 2021. Regulatory T Cells Masahiro Ono, Jun Shimizu, Yoshiki Miyachi and Shimon Sakaguchi Downloaded from J Immunol 2006; 176:4748-4756; ; doi: 10.4049/jimmunol.176.8.4748 http://www.jimmunol.org/content/176/8/4748 http://www.jimmunol.org/ References This article cites 51 articles, 27 of which you can access for free at: http://www.jimmunol.org/content/176/8/4748.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

by guest on September 29, 2021 • 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 Copyright © 2006 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Control of Autoimmune Myocarditis and Multiorgan Inﬂammation by Glucocorticoid-Induced TNF Receptor Family-Related Proteinhigh, Foxp3-Expressing CD25؉ and CD25؊ Regulatory T Cells1

Masahiro Ono,*† Jun Shimizu,‡ Yoshiki Miyachi,† and Shimon Sakaguchi2*§

Anomalies of naturally occurring CD4؉ regulatory T cells (Treg) cause severe autoimmune/inflammatory diseases in humans and rodents. The transcription factor Foxp3 is currently the most specific marker for natural CD4؉ Treg, but it would be useful if other Treg markers, particularly cell surface molecules, could be elucidated. We demonstrate in this study that the vast majority of Foxp3-expressing CD4؉ T cells (whether CD25؉ or CD25؊) show constitutive high-level expression of glucocorticoid-induced Downloaded from TNFR family-related gene/protein (GITR). Transfer of T cell or thymocyte suspensions depleted of GITRhigh cells produces in BALB/c nude mice a wider spectrum and more severe forms of autoimmune diseases than does transfer of similar cell suspensions depleted of CD25؉CD4؉ T cells only. Notably, mice that receive cells depleted of GITRhigh populations develop severe multiorgan inflammation that includes fatal autoimmune myocarditis resembling giant cell myocarditis in humans, accompanying high-titer anti-myosin autoantibodies. Similar transfer of GITRhigh-depleted cells from prediabetic NOD mice to NOD-SCID mice accelerates the development of diabetes and induces skeletal muscle myositis and other autoimmune/inflammatory diseases. We conclude http://www.jimmunol.org/ that GITRhigh, Foxp3-expressing natural Treg, containing both CD25؉ and CD25؊ cell populations, contribute to preventing a variety of autoimmune/inflammatory diseases, and depletion of these cells allows the activation of even weak or rare autoreactive T cells yielding widespread severe autoimmune disease. Diseases induced in this way include many which have been suspected of an autoimmune etiology in humans without much evidence. GITRhigh, Foxp3-expressing natural Treg represent a potential target for the treatment and prevention of these diseases. The Journal of Immunology, 2006, 176: 4748–4756.

here is accumulating evidence that regulatory T cells type 1 diabetes (T1D) and thyroiditis), inflammatory bowel dis- (Treg),3 in particular naturally occurring CD25ϩCD4ϩ ease, and allergy (such as allergic dermatitis and food allergy) (4, T Treg, play crucial roles in the maintenance of immuno- 5). The disease is so severe that Ͼ80% of the patients, including by guest on September 29, 2021 logic self-tolerance and suppressive control of a variety of patho- those with a diabetes-protective HLA haplotype, develop T1D and logical and physiological immune responses (1–3). Depletion of other autoimmune diseases (5). The cause of IPEX is mutation of ϩ ϩ CD25 CD4 Treg, for example, elicits autoimmune disease in the gene FOXP3, which encodes the forkhead/winged-helix tran- multiple organs such as stomach and thyroid in otherwise normal scription factor Scurfin (5). CD25ϩCD4ϩ natural Treg in mice animals, provokes effective tumor immunity to autologous tumor specifically express Foxp3 (the murine ortholog of human cells, enhances immune responses to invading or cohabiting mi- FOXP3), and ectopic or transgenic expression of Foxp3 can con- crobes, and triggers allergic responses to innocuous environmental vert naive T cells to Treg that phenotypically and functionally substances. The best example of the role of natural Treg in humans ϩ ϩ resemble natural CD25 CD4 Treg (6–8). Foxp3-defective or is a fatal autoimmune/inflammatory disease called IPEX (immune ϩ ϩ -deficient mice fail to develop CD25 CD4 Treg, leading to the dysregulation, polyendocrinopathy, enteropathy, X-linked syn- occurrence of fatal inflammatory disease (7). Foxp3 thus appears drome), which accompanies severe autoimmune disease (such as to be a master control gene for the development and function of naturally occurring Treg, and is currently the most specific molec- *Department of Experimental Pathology, Institute for Frontier Medical Sciences, ular marker for this T cell subpopulation (6–9). A critical question, Kyoto University, Kyoto, Japan; †Department of Dermatology, Graduate School of Medicine, Kyoto University, Kyoto, Japan; ‡Tokyo Metropolitan Institute for Ger- then, in elucidating the cause and mechanism of human autoim- ontology, Itabashi-ku, Tokyo, Japan; and §Core Research for Evolutional Science and mune diseases is how FOXP3 deficiency causes such fatal auto- Technology, Japan Science and Technology Agency, Kawaguchi, Japan immune/inflammatory diseases, and whether anomaly of FOXP3- Received for publication December 12, 2005. Accepted for publication January 26, 2006. expressing Treg can be a cause of other human inflammatory The costs of publication of this article were defrayed in part by the payment of page diseases including those where etiology is unknown. charges. This article must therefore be hereby marked advertisement in accordance It has been shown that CD25, the IL-2R ␣-chain, is not only a with 18 U.S.C. Section 1734 solely to indicate this fact. useful molecular marker for natural CD4ϩ Treg but also an essen- 1 This work was supported by grants-in-aid from the Ministry of Education, Sports tial molecule for their development and function as an indispens- and Culture of Japan. able component of the high-affinity IL-2R. For example, IL-2-, 2 Address correspondence and reprint requests to Dr. Shimon Sakaguchi, Department of Experimental Pathology, Institute for Frontier Medical Sciences, Kyoto University, CD25-, or CD122 (IL-2R ␤-chain)-deficient mice develop fatal 53 Shogoin Kawahara-cho, Sakyo-ku, Kyoto 606-8507, Japan. E-mail address: autoimmune/inflammatory disease similar to Foxp3 deficiency [email protected] (10–12). Neutralization of circulating IL-2 for a limited period 3 Abbreviations used in this paper: Treg, regulatory T cell; IPEX, immune dysregu- ϩ ϩ lation, polyendocrinopathy, enteropathy, X-linked syndrome; T1D, type 1 diabetes; selectively reduces the number of CD25 CD4 Treg and elicits GITR, glucocorticoid-induced TNFR family-related protein; C, complement. autoimmune disease similar to the one produced by depletion of

ϩ ϩ CD25 CD4 Treg (13). In contrast, there is accumulating evi- purified by the MACS system (Miltenyi Biotec), using biotinylated anti- dence for the existence of regulatory activity in the CD25ϪCD4ϩ CD4, anti-CD8, or anti-CD19, respectively, with PE-labeled streptavidin T cell population. For example, CD25ϪCD45RBlowCD4ϩ T cells and anti-PE microbeads. Other T cell subpopulations were prepared as Ϫ ϩ ϩ described previously (19). For assessing GITR expression on the residual in rodents and CD25 CD45RO CD4 T cells in humans contain cells after anti-GITR (DTA-1) or anti-CD25 (7D4)ϩC (complement)-treat- cells expressing Foxp3/FOXP3 at a low level (6, 9, 14). ment or C-treatment alone, they were incubated again with DTA-1 and CD25ϪCD4ϩ T cells bear an autoimmune-suppressive activity in 7D4, washed, and stained with rabbit polyclonal biotinylated anti-murine various animal models of autoimmune disease, although the activ- GITR Ab. ϩ ϩ ity is generally weaker than that of natural CD25 CD4 Treg Quantitation of mRNA expression by real-time RT-PCR (15–17). It must be determined then to what degree CD25ϪCD4ϩ Treg contribute to the maintenance of natural self-tolerance and Quantitation of Foxp3 and HPRT mRNA by real-time PCR was described previously (6). how severe the autoimmune diseases will be, in particular what spectrum of autoimmune/inflammatory disease will develop, when Histopathology ϩ Ϫ both CD25 and CD25 Foxp3-expressing Treg are depleted from Myocarditis and other autoimmune diseases were histologically evaluated the normal immune system. and graded as described previously (21–23). CD25ϩCD4ϩ Treg express GITR (glucocorticoid-induced TNFR family related gene/protein) at higher levels than other T Immunohistochemistry of heart cells or B cells; and neonatal administration of anti-GITR mAb For immunohistochemistry of heart, myocarditis-afflicted nude mice or induces autoimmune disease similar to the one produced by normal 8-wk-old BALB/c mice were sacrificed as described, and heart was ϩ ϩ depletion of CD25 CD4 Treg (18). In addition, obtained, which was washed with cold PBS and mounted in OCT com- Downloaded from Ϫ high pound and snap frozen. Infiltrating cells in the hearts with myocarditis were CD25 GITR T cells were shown to exert suppressive ac- analyzed by staining frozen sections with rat anti-CD4 (L3T4), rat anti- tivity in organ transplantation and inflammatory bowel disease CD8a, rat anti-B220 (CD45R), and rat anti-Mac1 (CD11b), which were (19, 20). In this study, we show that specific, stable expression obtained from BD Pharmingen, followed by HRP-conjugated hamster anti- ϩ ϩ of GITR occurs in the majority of Foxp3-expressing CD25 rat IgG (H L) (Kirkegaard & Perry Laboratories) as the secondary Ab, Ϫ counterstained by hematoxylin. Controls were stained with the secondary and CD25 Treg in the thymus and periphery. With high-level ϩ Ab alone. For indirect immunohistochemistry, normal heart from BALB/c expression of GITR as a reliable marker for Foxp3 natural mice was prepared as described above and fixed with acetone or 4% para- http://www.jimmunol.org/ ϩ Ϫ Treg, we show that both CD25 and CD25 Treg in the thymus formaldehyde. Sections were first incubated with 3% FBS in PBS for 1 h, and periphery contribute to the maintenance of natural self- then with 10-times-diluted sera from myocarditis-afflicted or control mice for1hatroom temperature, and with FITC-conjugated anti-mouse IgG tolerance and hence prevention of autoimmune disease but to ϩ ϩ high (H L) with Alexa Fluor 568-conjugated phalloidin counter staining. Flu- different degrees, and that by depletion of Foxp3 GITR orescence images were taken by a Zeiss LSM510 confocal microscope cells, we can activate even weak or rare autoimmune T cell attached to an axiovert inverted microscope. clones and cause severe and wide-spectrum autoimmune/ Western blotting inflammatory diseases as in IPEX. Importantly, this thorough depletion of Foxp3ϩ GITRhigh T cells also produces a variety of Organs/tissues from female 8-wk-old BALB/c mice were homogenized novel autoimmune/inflammatory diseases that immunopatho- with a Polytron homogenizer in lysis buffer containing 150 mM NaCl, 25 by guest on September 29, 2021 logically resemble human inflammatory diseases, such as giant mM Tris-HCl (pH7.5), 5 mM EDTA (pH 8.0), 10% IGEPAL CA-630 (Sigma-Aldrich), and protease inhibitor (WAKO). SDS-PAGE was per- cell myocarditis, which have been suspected to be autoimmune formed in either 12 or 7.5% acrylamide gel, or in 10% gradient gel made in etiology. The results indicate that reduction or functional from Prosieve 50 gel solution (BioWhittaker Molecular Applications) with anomaly of natural Treg can be a cause of such rare but fatal Broad Range (Bio-Rad) as a size marker. Separated proteins were then autoimmune/inflammatory diseases, providing a clue to their blotted onto a polyvinylidene difluoride membrane, Immobilon (Millipore) using a semidry-transfer system. The transferred membrane was blocked etiology, pathogenetic mechanism, treatment, and prevention. with 5% skim milk and incubated with serially diluted sera from 1/30 to 1/3000, followed by detection with HRP-conjugated anti-mouse IgG Materials and Methods (HϩL) (Kirkegaard & Perry Laboratories) and ECL (Amersham Mice Biosciences). Eight-week-old BALB/c mice and 6-wk-old BALB/c nude mice were pur- Results chased from Japan SLC. Eight-week-old NOD mice and 6-wk-old NOD- Induction of fatal disease in nude mice by transferring T cells SCID mice were purchased from CLEA Japan. They were treated in ac- high cordance with the institutional guidelines for animal welfare. depleted of Foxp3-expressing GITR cells Reagents Cytofluorometric analysis by intracellular Foxp3 staining and cell surface staining of CD25 or GITR showed that Ͼ95% of PE-labeled anti-CD25 (PC-61), FITC-conjugated-anti-CD4 (L3T4), anti- Foxp3ϩCD4ϩ T cells were GITRhigh, and that ϳ75% of them CD8 (53–6.7), PerCP-Cy5.5-anti-CD4 (L3T4), biotinylated anti-CD25 ϩ ϩ ϩ (7D4), anti-CD4 (L3T4), anti-CD8 (53–6.7), anti-CD19 (1D3), PE-labeled were CD25 (Fig. 1a). CD25 CD4 T cells in normal naive mice rat IgG2a (isotype control for Foxp3 staining), and PE-labeled streptavidin constitutively express CTLA-4 (24); and the majority of CTLA- were all purchased from BD Pharmingen; allophycocyanin-conjugated 4ϩCD4ϩ T cells in normal naive mice were GITRhigh (data not streptavidin was obtained from eBioscience; and biotinylated polyclonal shown). Foxp3ϩ cells constituted Ͻ1% of CD8ϩ T cells (data not anti-GITR was obtained from Genzyme-Techne. Biotinylated anti-GITR shown). Thus, the majority of Foxp3ϩCD4ϩ T cells, which can be mAb (DTA-1) was reported previously (18). PE-labeled anti-mouse Foxp3 ϩ Ϫ high mAb was obtained from eBioscience and used for intracellular staining CD25 or CD25 , are GITR in normal naive mice. according to the manufacturer’s instructions. Stained cells were analyzed To assess the role of Foxp3-expressing Treg in natural self- by a FACSCalibur (BD Biosciences). FITC-conjugated anti-mouse IgG tolerance, we attempted to deplete Foxp3ϩ T cells from BALB/c ϩ (H L) was obtained from Jackson ImmunoResearch Laboratories. Alexa spleen cells by anti-GITR or anti-CD25 mAb and rabbit C, trans- Fluor 568-conjugated phalloidin was obtained from Molecular Probes. Anti-PE microbeads were obtained from Miltenyi Biotec. ferred the remaining cells to syngeneic athymic nude mice, and compared the spectrum, incidence, and severity of autoimmune/ Preparation of lymphocytes inflammatory diseases in the recipients (Table I). Anti-GITRϩC- high CD25ϩ or GITRhigh cells were depleted using mAbs and complement as treatment depleted the GITR population, whereas anti- described previously (21). CD4ϩ T cells, CD8ϩ T cells, or B cells were CD25ϩC-treatment substantially reduced GITRhigh T cells but 4750 Foxp3ϩ Treg IN AUTOIMMUNITY failed to fully deplete them (Fig. 1b). In accord with the results in became debilitated, cachectic, or cyanotic (Fig. 2, Table I). Debil- Fig. 1, a and b, Foxp3ϩ cells assessed by intracellular staining itated mice frequently exhibited marked dilatation of the heart remained as ϳ6 and ϳ2% of CD4ϩ T cells after anti-CD25ϩCor (Fig. 2a), pleural effusion, ascites, and congested liver, indicating anti-GITRϩC-treatment, respectively (Fig. 1b). The result corre- severe heart failure as the cause of death. Histological examination lated with the level of Foxp3 mRNA expression: anti-GITRϩC- revealed a marked infiltration of mononuclear inflammatory cells, treatment reduced the level to 1/30 of control CD4ϩ cells and to a appearance of multinucleate giant cells, and destruction of cardiac level equivalent to that of CD8ϩ T cells, whereas anti-CD25ϩC- myocytes (Table I, Group B and Figs. 2, b and c, and 3). Infiltrat- treatment reduced the level only to 1/10 of controls (Fig. 1c). No- ing cells were predominantly Mac1ϩ cells (macrophages) and tably, transfer of anti-GITRϩC-treated cell suspensions (GITRlow) CD4ϩ T cells, and to a lesser extent, CD8ϩ T cells and B220ϩ cells resulted in reduced survival of the recipient nude mice (50% of (B cells) (Fig. 2e). Transfer of CD25Ϫ cells did not affect the survival mice died by 40 days after transfer), whereas transfer of anti- or induce myocarditis (Table I, Group C). Ϫ CD25ϩC-treated cells (i.e., CD25 ) or C-treated control cells The sera from myocarditis-afflicted mice strongly reacted with (i.e., whole T cells) did not affect the survival (Fig. 1d). heart tissue from normal BALB/c mice by immunohistochemistry (Fig. 2f). By Western blotting, the serum from every mouse with Development of fatal autoimmune myocarditis and other histologically evident myocarditis reacted to a 200-kDa protein of autoimmune diseases in BALB/c nude mice transferred with low heart extract, whereas the sera from mice without histologically GITR T cells evident myocarditis, whether transferred with GITRlow, CD25Ϫ,or To determine the cause of the death and the extent of autoimmu- whole cells, did not (Fig. 2g). The positive sera also reacted to a nity/inflammation in nude mice transferred with GITRlow T cells, 200-kDa protein extracted from skeletal muscle, stomach, colon, Downloaded from mice were macroscopically and histologically examined when they lung, and brain, with the strongest reactivity to the heart and skeletal http://www.jimmunol.org/ by guest on September 29, 2021 FIGURE 1. Fatal disease induced in nude mice by depletion of GITRhighFoxp3ϩ T cells. a, Intracel- lular Foxp3 staining of splenic CD4ϩ T cells expressing CD25 or GITR. Representative staining in five independent experiments. b, Expression of GITR and Foxp3 after treating normal BALB/c spleen and lymph node cells with anti-GITR (DTA-1) and C, anti-CD25 (7D4) and C, or C only. Treated cells were stained by FITC anti-CD4 and biotinylated polyclonal anti-GITR and PE-conjugated streptavidin or PE anti-Foxp3. c, CD4ϩ cells purified from the cell suspensions shown in (b) were assessed for Foxp3 mRNA expression by real-time RT-PCR. CD19ϩ (B cells) and CD8ϩ T cells were freshly prepared as controls. Data are representative of three independent experiments. d, Survival of BALB/c nude mice transferred with GITRlow, CD25Ϫ, or whole lymphocytes. Spleen and lymph node cells from 6-wk-old female BALB/c mice were treated as described in a and subsequently transferred to 6-wk-old female BALB/c nude mice. The Journal of Immunology 4751

Table I. Induction and prevention of various autoimmune diseases in BALB/c nude mice by the transfer of T cell subpopulationsa

Total Total Number of Mice with Autoimmune Disease Experimental BALB/c Nude Mice Number Group Transferred with: of Mice Mycd Gas Oop Thyr Sial Adr Ins Col Hep Gn

A Whole (3 ϫ 107) 9 0 0 0 0 0 00000 B GITRlow (6 ϫ 107) 5 3 (60) 5 (100) 5 (100) 2 (40) 5 (100) 1 (20) 0 0 1 (20) 2 (40) GITRlow (3 ϫ 107) 18 14 (78) 18 (100) 16 (89) 7 (39) 14 (78) 2 (11) 2 (11) 3 (17) 7 (39) 5 (28) GITRlow (1 ϫ 107) 5 4 (80) 5 (100) 4 (80) 0 5 (100) 0 0 1 (20) 2 (40) 2 (40) C CD25Ϫ (3 ϫ 107) 8 0 8 (100) 8 (100) 4 (50) 6 (75) 1 (13) 2 (25) 1 (13) 1 (13) 4 (50) D CD25Ϫ GITRlow (3 ϫ 107) 7 6 (86) 7 (100) 6 (86) 0 7 (100) 1 (14) 0 0 3 (43) 3 (43) E CD25ϩCD4ϩ (2 ϫ 106) ϩ 5 0 0 0 0 0 00000 GITRlow (3 ϫ 107) F CD25Ϫ (3 ϫ 107) ϩ 8 1 (13) 8 (100) 7 (88) 3 (38) 4 (50) 1 (13) 0 1 (13) 3 (38) 2 (25) GITRlow (3 ϫ 107) G CD25ϪCD4ϩ (6 ϫ 106) ϩ 6 0 6 (100) 6 (100) 0 3 (50) 0 1 (17) 0 0 3 (50) GITRlow (3 ϫ 107) H CD25ϪCD8ϩ (6 ϫ 106) ϩ 6 3 (50) 6 (100) 6 (100) 2 (33) 3 (50) 0 1 (17) 0 2 (33) 3 (50) GITRlow (3 ϫ 107)

a Indicated numbers of T cell subpopulations were prepared from BALB/c splenic and lymph node cells by the treatment with anti-GITR or anti-CD25, or a combination of Ͼ

these, and complement. Nude mice were histologically and serologically examined when they became debilitated (i.e., 20% decrease of body weight). Surviving mice were Downloaded from examined 3 mo after cell transfer. Percentages of mice with histologically evident autoimmune disease are given in parenthesis. Mycd, Myocarditis; Gas, gastritis; Oop, oophoritis; Thyr, thyroiditis; Sial, sialoadenitis; Adr, adrenalitis; Ins, insulitis; Col, colitis; Hep, hepatitis; Gn, glomerulonephritis. muscle extract (Fig. 2h), although there was no inﬂammation in the Western blotting correlated with the histological severity of myocar- skeletal muscle (data not shown). The 200-kDa protein appeared to be ditis (Fig. 2g and data not shown). cardiac myosin because the sera reacted with puriﬁed rabbit myosin as This induced myocarditis is a T cell-mediated autoimmune disease http://www.jimmunol.org/ well (Fig. 2h). The titer of this heart-reactive autoantibody assessed by because adoptive transfer of spleen cells, especially CD4ϩ T cells,

FIGURE 2. Autoimmune myocarditis induced in low

nude mice by the transfer of GITR T cells. a, by guest on September 29, 2021 Macroscopic view of hearts. GITRlow cell-transferred mice showed enlargement of the heart (left) compared with controls (right). H&E staining of the heart of a GITRlow cell-transferred nude mouse 4 wk after cell transfer (b and c) or a CD25Ϫ cell-transferred mouse 3 mo after cell transfer (d). Infiltrating cells are chiefly lymphocytes, macrophages, and oc- casional giant cells (arrows in inset) (H&E staining; original magnification, ϫ10 or ϫ40). e, Immunohis- tochemistry shows CD4ϩ, CD8ϩ, B220ϩ, or Mac- 1ϩ-infiltrating cells in the heart. Nuclei were counterstained by hematoxylin. f, Immunohistochemistry showing reactivity of the sera of myocarditis-afflicted mice to normal heart tissue. Sera from mice with or without myocarditis were incubated with heart tissue from normal BALB/c mice. Subse- quently, heart tissue was stained by FITC-conjugated anti-mouse IgG and counterstained with Alexa Fluor 568-conjugated phalloidin and analyzed by a confocal microscopy. g, Western blotting of heart extracts with sera from myocarditis-afflicted mice (lanes 1–5), or from mice without myocarditis fol- lowing transfer of GITRlow cells (lane 6), CD25Ϫ cells (lanes 7 and 8), or whole lymphocytes (lanes 9 and 10). h, Reactivity of sera from myocarditis-afflicted mice to extracts from various tissues or purified myosin H chains, assessed by Western blotting. i, Adoptive transfer of myocarditis to naive nude mice by CD4ϩ, CD8ϩ, or unfractionated spleen cells from myocarditis-bearing mice. 4752 Foxp3ϩ Treg IN AUTOIMMUNITY

FIGURE 3. Prevention of myocarditis in nude mice by the cotransfer of T cell subpopulations. Nude mice transferred with various cell populations, as shown in Table I, were histologically examined for the development of myocarditis. Mice were sacriﬁced and hearts analyzed at 3 mo posttransfer, or earlier if animals became debilitated and had to be euthanized. Numbers of transferred cells are shown in parentheses.

was able to secondarily transfer the disease to a fresh cohort of hepatitis) at high incidences even when early death due to myo- syngeneic athymic nude mice (Fig. 2i). When the donor mice were carditis made it difficult to accurately assess the incidences of other myocarditis-free but developed other organ-specific autoimmune dis- slowly developing autoimmune/inflammatory diseases in BALB/c eases (such as gastritis, thyroiditis, and oophoritis; Table I), the cell nude recipients (Table I). transfer failed to elicit myocarditis in the recipients but successfully We showed in a previous report that transfer of GITRlow T cells produced the latter (Ref. 25 and data not shown). Transfer of the sera produced autoimmune diseases in syngeneic BALB/c nude mice Downloaded from alone from myocarditis-afflicted mice failed to produce myocarditis or without reduction of survival (18). The difference between the pre- other diseases in athymic or euthymic mice (data not shown). vious and the present results could be attributed to the strains of Thus, transfer of GITRlow cells, in contrast with that of CD25Ϫ nude mice used. The BALB/c nude substrain previously used (i.e., cells, induced in BALB/c nude mice T cell-mediated autoimmune BALB/cClea-nu) developed nonfatal and histologically mild myo- myocarditis that immunopathologically resembled giant cell myocarditis at 20% incidence after transfer of GITRlow T cells (18).

carditis in humans (26, 27). The transfer also produced other au- There are subtle differences among substrains of BALB/c nude http://www.jimmunol.org/ toimmune diseases (such as gastritis, thyroiditis, sialadenitis, and mice in the incidence, severity, and spectrum of autoimmune disease by guest on September 29, 2021 FIGURE 4. Induction of severe autoimmunity in NOD-SCID mice by the transfer of GITRlow T cells from NOD mice. a, Intra- cellular staining of splenocytes from prediabetic NOD mice for Foxp3 expression. Total splenocyteswerestainedbyPerCP-Cy5.5-anti- CD4, FITC-anti-CD8, and biotinylated anti- CD25 or anti-GITR and allophycocyanin- conjugated streptavidin, and subsequently fixed and stained by PE-anti-Foxp3. Body weight (b) and incidence of hyperglycemia (Ͼ350 mg/dl) (c) in NOD-SCID mice transferred with anti-GITRϩC (GITRlow), anti- CD25ϩC (CD25Ϫ), or C-treated (whole) spleen and lymph node cells from nondiabetic NOD mice. Recipient NOD-SCID mice were sacrificed for histological examination when they became debilitated. Asterisks mean significant differences (p Ͻ 0.05). Cross (ϩ) indicates that all mice in this group (GITRlow) had to be sacrificed because of debilitation and hyperglycemia (at 7 wk posttransfer). d, Representative histology of various tissues of GITRlow cell-transferred NOD-SCID mice or whole-cell-transferred mice as control (H&E staining; original magnification, ϫ10). The Journal of Immunology 4753

Table II. Induction of various autoimmune diseases/inﬂammations in NOD-SCID mice by the transfer of T cell subpopulationsa

Total Total Number of Mice with Autoimmune Disease Experimental NOD-SCID Mice Number Group Transferred with: of Mice Mycd Mys Col Gas Thyr Oop Sial Adr Ins Hep Gn

A Whole (1 ϫ 107) 5 0 (0) 0 (0) 0 (0) 1 (20) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) B GITRlow (1 ϫ 107) 5 0 (0) 4 (80) 5 (100) 5 (100) 2 (40) 1 (20) 1 (20) 1 (20) 5 (100) 5 (100) 1 (20) C CD25Ϫ (1 ϫ 107) 5 0 (0) 1 (20) 0 (0) 5 (100) 4 (80) 3 (60) 5 (100) 3 (60) 4 (80) 1 (20) 1 (20)

a Indicated numbers of T cell subpopulations from 6-wk-old nondiabetic NOD mice were prepared as described in Table I. NOD-SCID mice were histologically and serologically examined when they became debilitated (i.e., Ͼ20% decrease of body weight) or their blood glucose levels became higher than 350 mg/dl in two successive measurements. Mys, Myositis. See Table I for other abbreviations.

induced by removal of CD25ϩCD4ϩ T cells (25). BALB/cSlc nude Acceleration of the onset of T1D and induction of other mice and BALB/cSlc mice, which were used in the present experi- autoimmune diseases in NOD-SCID mice by the transfer of ments, are more susceptible than other BALB/c substrains to various GITRlow T cells organ-specific autoimmune diseases. The BALB/cSlc athymic and ϩ Ϫ Diabetes-prone NOD mice also harbor CD25 and CD25 Foxp3- euthymic strains in our facility were ensured to be specific pathogen- ϩ expressing CD4 T cells, the majority (Ͼ97%) of which were Downloaded from free, in particular, free of Coxsackie virus and other viruses that are GITRhigh, comparable to what was shown in BALB/c mice (Fig. reported to cause myocarditis (28). 4a). To determine then whether the development of severe auto- high high Ϫ ϩ ϩ ϩ immune diseases after removing GITR T cells is unique to GITR CD25 CD4 T cells and CD25 CD4 T cells in low Ϫ normal naive mice bear autoimmune-suppressive activity BALB/c mice, we transferred GITR or CD25 T cells from NOD mice to NOD-SCID mice (Table II). Mice receiving The development of autoimmune myocarditis in nude mice after GITRlow cell transfers began to lose body weight ϳ2 wk after cell http://www.jimmunol.org/ low Ϫ transfer of GITR cells, but not CD25 cells, cannot be due to transfer (Fig. 4b). The onset of diabetes was accelerated in these different numbers of autoimmune effector T cells transferred, be- Ͼ low mice, which started to develop hyperglycemia ( 350 mg/dl) cause transfer of GITR cells at a dose much smaller than Ϫ Ϫ within 1 mo after cell transfer, compared with Ͼ2 mo for CD25 CD25 T cells could still induce myocarditis (Table I, Group B). low ϩ low low cell-transferred mice (Fig. 4c). GITR -cell-transferred mice also Furthermore, CD25 GITR cells, which exist in GITR but Ϫ Ϫ developed severe diarrhea with debilitation, whereas CD25 cell- not in CD25 cells, are not “heart-specific effector cells” because transferred mice did not (Fig. 4b). Histologically, colitis, myositis their depletion from GITRlow cells by mixed mAbs of anti-CD25 and anti-GITRϩC did not reduce the incidence of autoimmune of skeletal muscle (but not the myocardium), and hepatitis with myocarditis (Table I, Group D; and Fig. 3). It is more likely, there- inflammatory destruction of the portal area frequently developed in by guest on September 29, 2021 low fore, that CD25ϪGITRhigh cells bear an autoimmune regulatory GITR cell-transferred NOD-SCID mice despite their early de- activity, and anti-GITRϩC-treatment depletes not only CD25ϩ mise after cell transfer due to severe diarrhea and debilitation (Fig. Treg but also CD25Ϫ Treg. To examine this possibility further, we 4d and Table II). Both groups developed multiple organ-specific cotransferred to nude mice the same number of GITRlow cells and autoimmune diseases including gastritis and thyroiditis (Table II). ϩ CD25Ϫ cells or an equivalent number of CD25ϪCD4ϩ cells, and Thus, GITRhigh cell populations would include both CD25 and Ϫ assessed the occurrence of myocarditis and other autoimmune dis- CD25 Treg in NOD mice as well, as shown in BALB/c mice. eases (Table I, Groups F and G; and Fig. 3). The cotransfer effec- These data, specifically with respect to myositis, colitis, and hep- tively suppressed the development of myocarditis, but not other atitis, should suggest a similar phenomenon to that which was Ϫ ϩ autoimmune diseases, indicating that GITRhighCD25 CD4 T observed in BALB/c mice. cells in the CD25ϪCD4ϩ population could suppress myocarditis but not the occurrence of other diseases. Apparently, this myocarditis-specific suppression by CD25ϪCD4ϩ T cells also makes it Thymic production of GITRhigh Treg low unlikely that severe disease induction by GITR cells is due to Single-cell staining revealed that Foxp3ϩ thymocytes comprised low possible activation of GITR cells by agonistic DTA-1 attached Ͼ high 5–6% of CD4-SP thymocytes (Fig. 5a). The majority ( 98%) of to them during in vitro Ab treatment for depletion of GITR Foxp3ϩCD4-SP thymocytes were GITRhigh, compared with cells (18). ϩ ϩ ϳ70% of Foxp3 thymocytes as being CD25 (Fig. 5a). Compared with CD25ϪCD4ϩ T cells, CD25ϪCD8ϩ T cells Transfer of anti-GITRϩC-treated BALB/c thymocytes, at a were much less effective in this autoimmune suppression, whereas ϩ ϩ dose allowing the majority of the recipient mice to survive more a small number of CD25 CD4 T cells completely suppressed the than 3 mo, produced various autoimmune diseases in BALB/c development of every autoimmune disease (Table I, Groups E, G, and H; and Fig. 3). nude mice, including histologically evident autoimmune myocar- Taken together, these results indicate that not only ditis (Fig. 5b and Table III). In contrast, transfer of anti-CD25 and CD25ϩCD4ϩ T cells but also GITRhighCD25Ϫ cells in the C-treated thymocytes failed to produce myocarditis, although it CD25ϪCD4ϩ population hold an autoimmune-suppressive activ- induced other autoimmune diseases in a similar spectrum of or- high ity. Compared with the potent autoimmune-suppressive activity of gans. Thus, the normal thymus generates GITR CD4-SP thy- ϩ the former, the latter is sufficient in number or suppressive activity mocytes, which include the majority of Foxp3 Treg with auto- ϩ to inhibit the development of myocarditis but insufficient to sup- immune-preventive activity. Some of such Foxp3 Treg are ϩ Ϫ press other autoimmune diseases such as gastritis, thyroiditis, and CD25 , others are CD25 . In addition, mature CD4-SP thymo- oophoritis, which can be easily induced in BALB/c mice by de- cytes contain T cells with various self-reactive specificities includ- pletion of CD25ϩCD4ϩ Treg alone (3). ing those against cardiac Ags. 4754 Foxp3ϩ Treg IN AUTOIMMUNITY

riety of approaches. For example, it can be produced in genetically susceptible strains of mice by immunization with cardiac myosin along with potent adjuvant, repeated inoculation of dendritic cells pulsed with myosin peptides, or infection with Coxsackie virus and other cardiotropic viruses (23, 28–31). BALB/c mice deficient of PD-1, a coinhibitory molecule expressed by activated T and B cells, spontaneously develop cardiomyopathy accompanying autoantibody specific for cardiac troponin (32). CTLA-4 or TGF-␤- deficient mice also succumb to severe myocarditis (33–36). Fur- thermore, NOD mice expressing transgenic human HLA-DQ8 in the absence of endogenous class II MHC spontaneously develop severe myocarditis accompanying autoantibody specific for cardiac myosin H chain (37). Thus, autoimmune myocarditis devel- ops as the result of altered antigenicity or excessive presentation of cardiac self-Ags, and also as the consequence of alteration in the generation or peripheral control of self-reactive lymphocytes specific for cardiac self-Ags. Assuming that the various forms of myocarditis in humans are heterogeneous in etiology, it remains to be Downloaded from FIGURE 5. Induction of myocarditis in nude mice by the transfer of determined which of the mechanisms revealed in animals are rel- thymocytes depleted of GITRhighFoxp3ϩthymocytes. a, Intracellular stain- evant to particular types of myocarditis in humans. It is of note in ing of thymocytes for Foxp3 expression. Total thymocytes were stained by this regard that a type of giant cell myocarditis and dilated car- PerCP-Cy5.5-anti-CD4, FITC-anti-CD8, and biotinylated anti-CD25 or anti- diomyopathy in humans is frequently associated with other organ- GITR and allophycocyanin-conjugated streptavidin, and subsequently specific autoimmune diseases, such as T1D and thyroiditis (26, fixed and stained by PE-anti-Foxp3. b, Thymocytes from normal female high ϩ 27). Given the fact that developmental or functional anomaly of 6-wk-old BALB/c mice were depleted of GITR or CD25 thymocytes natural Treg can be a cause of autoimmune disease as demon- http://www.jimmunol.org/ by the treatment with anti-GITR or anti-CD25, respectively, and C (see Fig. 2), and subsequently transferred to BALB/c nude recipients, which strated in IPEX and possibly in other autoimmune polyendocri- were histologically examined 3 mo later. Myocarditis was histologically nopathy syndromes (4, 38), our present results indicate that one of graded as in Fig. 2i or Fig. 3. the causes of giant cell myocarditis in humans can be genetically or environmentally induced abnormality in natural Treg. In this way, the disease would share a common pathogenetic basis with Discussion other more common organ-specific autoimmune diseases such as The transcription factor Foxp3 is currently the most specific T1D, chronic thyroiditis, and autoimmune gastritis (26, 27). Myo- ϩ carditis in CTLA-4 or TGF-␤ deficiency could be attributed to marker for naturally arising CD4 Treg (6–9). We have shown in by guest on September 29, 2021 this study that high expression of GITR occurs in the majority of defective development or function of natural Treg because both Foxp3-expressing CD4ϩ T cells in normal mice, whether they are molecules are engaged in the maintenance and activation of natural CD25ϩ or CD25Ϫ, and in the thymus or the periphery. Transfer of Treg (24, 39–43). T cell/thymocyte suspensions depleted of GITRhigh cells indeed A key finding in this study is that when the reduction of natural produced more severe autoimmune/inflammatory diseases in a Treg is very severe, novel autoimmune diseases develop, which wider spectrum of organs in T cell-deficient recipient mice than the fail to occur with mild Treg reduction; e.g., development of auto- transfer of similar populations depleted of CD25ϩ T cells/thymo- immune myocarditis in BALB/c mice after depletion of GITRhigh ϩ cytes alone. The autoimmune/inflammatory diseases thus induced T cells but not CD25 T cells alone. We have shown that whatever are immunopathologically similar to their human counterparts. the procedures of reducing the number or function of natural Treg The GITRhigh cell depletion induced autoimmune myocarditis at are, the reduction produces a similar spectrum of organ-specific a high incidence in BALB/c mice in addition to other organ-spe- autoimmune diseases in a particular strain of mice; for example, cific autoimmune diseases, such as gastritis and thyroiditis, which gastritis, oophoritis, thyroiditis, and other autoimmune diseases in are commonly produced by depletion of CD25ϩ Treg. The myo- that order of incidences in BALB/c mice (13, 44). Similarly, NOD carditis accompanied the development of autoantibodies specific mice are genetically prone to develop T1D and thyroiditis; and for cardiac myosin (23, 29, 30). Like gastritis and thyroiditis in- Treg reduction accelerates the occurrence of T1D and thyroiditis in duced by Treg depletion, it is a bona fide autoimmune disease NOD mice (13, 45). These results, when taken together, indicate because T cells, especially CD4ϩ T cells, were able to transfer the that the phenotype and severity of autoimmune/inflammatory dis- disease adoptively to naive nude mice (25, 29). It has been shown eases elicited by Treg reduction depends, at least in part, on the that autoimmune myocarditis can be induced in animals by a va- degree of Treg reduction, but also on the genetic background of the

Table III. Induction of various autoimmune diseases in BALB/c nude mice by the transfer of thymocyte subpopulationsa

Total Total Number of Mice with Autoimmune Disease Experimental BALB/c Nude Mice Number Group Transferred with: of Mice Mycd Gas Oop Thyr Sial Adr Ins Col Hep Gn

A Whole (3 ϫ 107) 7 0 6 (86) 0 0000000 B GITRlow (3 ϫ 107) 12 9 (75) 12 (100) 4 (33) 1 (8.3) 7 (58) 1 (8.3) 4 (33) 2 (17) 4 (33) 1 (8.3) C CD25Ϫ (3 ϫ 107) 8 0 8 (100) 8 (100) 2 (25) 5 (63) 0 0 0 0 3 (38)

a Indicated numbers of thymocyte subpopulations from BALB/c mice were prepared as described in Table I. Nude mice were histologically and serologically examined 3 mo after cell transfer. The Journal of Immunology 4755 hosts. The more severe the Treg reduction, the wider the spectrum 7. Fontenot, J. D., M. A. Gavin, and A. Y. Rudensky. 2003. Foxp3 programs the ϩ ϩ of autoimmune diseases in a genetically determined hierarchical development and function of CD4 CD25 regulatory T cells. Nat. Immunol. 4: 330–336. pattern. Furthermore, it is likely that normal mice harbor those 8. Khattri, R., T. Cox, S. A. Yasayko, and F. Ramsdell. 2003. An essential role for self-reactive T cells (e.g., heart-reactive T cells in BALB/c mice), Scurfin in CD4ϩCD25ϩ T regulatory cells. Nat. Immunol. 4: 337–342. 9. Fontenot, J. D., J. P. Rasmussen, L. M. Williams, J. L. Dooley, A. G. Farr, and which may be low in number or pathogenicity and therefore easily A. Y. Rudensky. 2005. Regulatory T cell lineage specification by the forkhead subjected to suppressive control in the physiological state even by transcription factor foxp3. Immunity 22: 329–341. CD25Ϫ Treg. 10. Schorle, H., T. Holtschke, T. Hunig, A. Schimpl, and I. Horak. 1991. Develop- ϩ ment and function of T cells in mice rendered interleukin-2 deficient by gene The present experiments demonstrated that both CD25 and targeting. Nature 352: 621–624. Ϫ ϩ CD25 CD4 Treg actively engage in the maintenance of natural 11. Willerford, D. M., J. Chen, J. A. Ferry, L. Davidson, A. Ma, and F. W. Alt. 1995. self-tolerance, although the former are more abundant and more Interleukin-2 receptor ␣ chain regulates the size and content of the peripheral lymphoid compartment. Immunity 3: 521–530. potent in suppressing autoimmunity. We previously showed that 12. Malek, T. R., A. Yu, V. Vincek, P. Scibelli, and L. Kong. 2002. CD4 regulatory Ϫ ϩ the GITRhighCD25 CD4 population in normal naive mice gave T cells prevent lethal autoimmunity in IL-2R␤-deficient mice: implications for rise to Foxp3highCD25ϩCD4ϩ Treg in vivo; and the latter were the nonredundant function of IL-2. Immunity 17: 167–178. 13. Setoguchi, R., S. Hori, T. Takahashi, and S. Sakaguchi. 2005. Homeostatic main- suppressive upon in vitro anti-CD3 stimulation (19). It is also tenance of natural Foxp3ϩCD25ϩCD4ϩ regulatory T cells by interleukin (IL)-2 known that CD25ϩCD4ϩ Treg may lose the expression of CD25, and induction of autoimmune disease by IL-2 neutralization. J. Exp. Med. 201: 723–735. especially when they are subjected to homeostatic proliferation in 14. Yagi, H., T. Nomura, K. Nakamura, S. Yamazaki, T. Kitawaki, S. Hori, a T cell-deficient environment (17, 19, 46). It is thus likely that M. Maeda, M. Onodera, T. Uchiyama, S. Fujii, and S. Sakaguchi. 2004. Crucial ϩ ϩ Foxp3-expressing CD25ϩ and CD25ϪCD4ϩ Treg present in nor- role of FOXP3 in the development and function of human CD25 CD4 regulatory T cells. Int. Immunol. 16: 1643–1656. Downloaded from mal naive mice are functionally similar and developmentally con- 15. Stephens, L. A., and D. Mason. 2000. CD25 is a marker for CD4ϩ thymocytes tiguous, but at different stages of their activation and/or differen- that prevent autoimmune diabetes in rats, but peripheral T cells with this function ϩ Ϫ tiation (9, 47), and that their CD25 phenotype can be convertible are found in both CD25 and CD25 subpopulations. J. Immunol. 165: ϩ 3105–3110. in the periphery. The normal thymus also contains both CD25 16. Lehmann, J., J. Huehn, M. de la Rosa, F. Maszyna, U. Kretschmer, V. Krenn, Ϫ ␣ ␤ and CD25 Treg. Yet, it remains to be determined whether M. Brunner, A. Scheffold, and A. Hamann. 2002. Expression of the integrin E 7 ϩ Ϫ CD25ϩ and CD25Ϫ Treg in the periphery are all produced by the identifies unique subsets of CD25 as well as CD25 regulatory T cells. Proc. Ϫ Natl. Acad. Sci. USA 99: 13031–13036. thymus as functionally mature Treg, or peripheral CD25 naive T 17. Annacker, O., R. Pimenta-Araujo, O. Burlen-Defranoux, T. C. Barbosa, http://www.jimmunol.org/ cells differentiate to CD25ϩ/CD25Ϫ Treg under certain in vivo A. Cumano, and A. Bandeira. 2001. CD25ϩCD4ϩ T cells regulate the expansion of peripheral CD4 T cells through the production of IL-10. J. Immunol. 166: conditions (48–51). 3008–3018. In conclusion, the degree of reduction of Foxp3-expressing nat- 18. Shimizu, J., S. Yamazaki, T. Takahashi, Y. Ishida, and S. Sakaguchi. 2002. Stim- ϩ ϩ ural Treg, whether they are CD25ϩ or CD25Ϫ, can determine the ulation of CD25 CD4 regulatory T cells through GITR breaks immunological self-tolerance. Nat. Immunol. 3: 135–142. phenotype and severity of autoimmune/inflammatory disease ac- 19. Nishimura, E., T. Sakihama, R. Setoguchi, K. Tanaka, and S. Sakaguchi. 2004. cording to a hierarchical genetic susceptibility of each autoimmune Induction of antigen-specific immunologic tolerance by in vivo and in vitro an- ϩ ϩ ϩ disease (3). High incidences of severe autoimmune diseases, es- tigen-specific expansion of naturally arising Foxp3 CD25 CD4 regulatory T cells. Int. Immunol. 16: 1189–1201. pecially T1D and thyroiditis, in IPEX could be attributed to severe 20. Uraushihara, K., T. Kanai, K. Ko, T. Totsuka, S. Makita, R. Iiyama,

ϩ Ϫ ϩ by guest on September 29, 2021 deficiency of both CD25 and CD25 FOXP3 Treg from the T. Nakamura, and M. Watanabe. 2003. Regulation of murine inflammatory bowel disease by CD25ϩ and CD25ϪCD4ϩ glucocorticoid-induced TNF receptor fam- beginning of their ontogeny (5). It is possible that reduction or ϩ ily-related gene regulatory T cells. J. Immunol. 171: 708–716. dysfunction of natural Treg, along with host genetic susceptibility, 21. Shimizu, J., S. Yamazaki, and S. Sakaguchi. 1999. Induction of tumor immunity could be a cause of certain other autoimmune/inflammatory dis- by removing CD25ϩCD4ϩ T cells: a common basis between tumor immunity and autoimmunity. J. Immunol. 163: 5211–5218. eases where the etiology is currently unknown. Furthermore, the 22. Itoh, M., T. Takahashi, N. Sakaguchi, Y. Kuniyasu, J. Shimizu, F. Otsuka, and present results indicate that natural Treg can be exploited for the S. Sakaguchi. 1999. Thymus and autoimmunity: production of CD25ϩCD4ϩ prevention and treatment of various autoimmune/inflammatory naturally anergic and suppressive T cells as a key function of the thymus in maintaining immunologic self-tolerance. J. Immunol. 162: 5317–5326. diseases, including autoimmune giant cell myocarditis and result- 23. Neu, N., N. R. Rose, K. W. Beisel, A. Herskowitz, G. Gurri-Glass, and ing dilated cardiomyopathy. S. W. Craig. 1987. Cardiac myosin induces myocarditis in genetically predis- posed mice. J. Immunol. 139: 3630–3636. 24. Takahashi, T., T. Tagami, S. Yamazaki, T. Uede, J. Shimizu, N. Sakaguchi, Acknowledgments T. W. Mak, and S. Sakaguchi. 2000. Immunologic self-tolerance maintained by CD25ϩCD4ϩ regulatory T cells constitutively expressing cytotoxic T lympho- We thank Z. Fehervari for critical reading of the manuscript and cyte-associated antigen 4. J. Exp. Med. 192: 303–310. T. Matsushita and K. Kogishi for histology. 25. Sakaguchi, S., N. Sakaguchi, M. Asano, M. Itoh, and M. Toda. 1995. Immuno- logic self-tolerance maintained by activated T cells expressing IL-2 receptor ␣-chains (CD25): breakdown of a single mechanism of self-tolerance causes Disclosures various autoimmune diseases. J. Immunol. 155: 1151–1164. The authors have no financial conflict of interest. 26. Cooper, L. T., Jr. 2000. Giant cell myocarditis: diagnosis and treatment. Herz. 25: 291–298. 27. Limas, C. J., P. Iakovis, A. Anyfantakis, C. Kroupis, and D. V. Cokkinos. 2004. Familial clustering of autoimmune diseases in patients with dilated cardiomyop- References athy. Am. J. Cardiol. 93: 1189–1191. 1. Maloy, K. J., and F. Powrie. 2001. Regulatory T cells in the control of immune 28. Fairweather, D., S. Frisancho-Kiss, and N. R. Rose. 2005. Viruses as adjuvants pathology. Nat. Immunol. 2: 816–822. for autoimmunity: evidence from Coxsackievirus-induced myocarditis. Rev. Med. ϩ ϩ 2. Shevach, E. M. 2002. CD4 CD25 suppressor T cells: more questions than Virol. 15: 17–27. answers. Nat. Rev. Immunol. 2: 389–400. 29. Smith, S. C., and P. M. Allen. 1991. Myosin-induced acute myocarditis is a T ϩ 3. Sakaguchi, S. 2004. Naturally arising CD4 regulatory T cells for immunologic cell-mediated disease. J. Immunol. 147: 2141–2147. self-tolerance and negative control of immune responses. Annu. Rev. Immunol. 30. Smith, S. C., and P. M. Allen. 1992. Expression of myosin-class II major histo- 22: 531–562. compatibility complexes in the normal myocardium occurs before induction of 4. Wildin, R. S., S. Smyk-Pearson, and A. H. Filipovich. 2002. Clinical and mo- autoimmune myocarditis. Proc. Natl. Acad. Sci. USA 89: 9131–9135. lecular features of the immunodysregulation, polyendocrinopathy, enteropathy, X 31. Eriksson, U., R. Ricci, L. Hunziker, M. O. Kurrer, G. Y. Oudit, T. H. Watts, linked (IPEX) syndrome. J. Med. Genet. 39: 537–545. I. Sonderegger, K. Bachmaier, M. Kopf, and J. M. Penninger. 2003. Dendritic 5. Gambineri, E., T. R. Torgerson, and H. D. Ochs. 2003. Immune dysregulation, cell-induced autoimmune heart failure requires cooperation between adaptive and polyendocrinopathy, enteropathy, and X-linked inheritance (IPEX), a syndrome innate immunity. Nat. Med. 9: 1484–1490. of systemic autoimmunity caused by mutations of FOXP3, a critical regulator of 32. Okazaki, T., Y. Tanaka, R. Nishio, T. Mitsuiye, A. Mizoguchi, J. Wang, T-cell homeostasis. Curr. Opin. Rheumatol. 15: 430–435. M. Ishida, H. Hiai, A. Matsumori, N. Minato, and T. Honjo. 2003. Autoantibod- 6. Hori, S., T. Nomura, and S. Sakaguchi. 2003. Control of regulatory T cell de- ies against cardiac troponin I are responsible for dilated cardiomyopathy in PD- velopment by the transcription factor Foxp3. Science 299: 1057–1061. 1-deficient mice. Nat. Med. 9: 1477–1483. 4756 Foxp3ϩ Treg IN AUTOIMMUNITY

33. Shull, M. M., I. Ormsby, A. B. Kier, S. Pawlowski, R. J. Diebold, M. Yin, 41. Fahlen, L., S. Read, L. Gorelik, S. D. Hurst, R. L. Coffman, R. A. Flavell, and R. Allen, C. Sidman, G. Proetzel, D. Calvin, et al. 1992. Targeted disruption of F. Powrie. 2005. T cells that cannot respond to TGF-␤ escape control by the mouse transforming growth factor-␤ 1 gene results in multifocal inflamma- CD4ϩCD25ϩ regulatory T cells. J. Exp. Med. 201: 737–746. tory disease. Nature 359: 693–699. 42. Salomon, B., D. J. Lenschow, L. Rhee, N. Ashourian, B. Singh, A. Sharpe, and J. A. Bluestone. 2000. B7/CD28 costimulation is essential for the homeostasis of 34. Kulkarni, A. B., C. G. Huh, D. Becker, A. Geiser, M. Lyght, K. C. Flanders, ϩ ϩ A. B. Roberts, M. B. Sporn, J. M. Ward, and S. Karlsson. 1993. Transforming the CD4 CD25 immunoregulatory T cells that control autoimmune diabetes. growth factor-␤1 null mutation in mice causes excessive inflammatory response Immunity 12: 431–440. and early death. Proc. Natl. Acad. Sci. USA 90: 770–774. 43. Read, S., V. Malmstrom, and F. Powrie. 2000. Cytotoxic T lymphocyte-associated antigen 4 plays an essential role in the function of CD25ϩCD4ϩ regulatory 35. Waterhouse, P., J. M. Penninger, E. Timms, A. Wakeham, A. Shahinian, cells that control intestinal inflammation. J. Exp. Med. 192: 295–302. K. P. Lee, C. B. Thompson, H. Griesser, and T. W. Mak. 1995. Lymphoprolif- 44. Sakaguchi, S., and N. Sakaguchi. 2000. Role of genetic factors in organ-specific erative disorders with early lethality in mice deficient in CTLA-4. Science 270: autoimmune diseases induced by manipulating the thymus or T cells, and not 985–988. self-antigens. Rev. Immunogenet. 2: 147–153. 36. Tivol, E. A., F. Borriello, A. N. Schweitzer, W. P. Lynch, J. A. Bluestone, and 45. Alyanakian, M. A., S. You, D. Damotte, C. Gouarin, A. Esling, C. Garcia, A. H. Sharpe. 1995. Loss of CTLA-4 leads to massive lymphoproliferation and S. Havouis, L. Chatenoud, and J. F. Bach. 2003. Diversity of regulatory CD4ϩ T fatal multiorgan tissue destruction, revealing a critical negative regulatory role of cells controlling distinct organ-specific autoimmune diseases. Proc. Natl. Acad. CTLA-4. Immunity 3: 541–547. Sci. USA 100: 15806–15811. 37. Elliott, J. F., J. Liu, Z. N. Yuan, N. Bautista-Lopez, S. L. Wallbank, K. Suzuki, 46. Gavin, M. A., S. R. Clarke, E. Negrou, A. Gallegos, and A. Rudensky. 2002. ϩ ϩ D. Rayner, P. Nation, M. A. Robertson, G. Liu, and K. M. Kavanagh. 2003. Homeostasis and anergy of CD4 CD25 suppressor T cells in vivo. Nat. Im- Autoimmune cardiomyopathy and heart block develop spontaneously in HLA- munol. 3: 33–41. DQ8 transgenic IA␤ knockout NOD mice. Proc. Natl. Acad. Sci. USA 100: 47. Huehn, J., K. Siegmund, J. C. Lehmann, C. Siewert, U. Haubold, M. Feuerer, 13447–13452. G. F. Debes, J. Lauber, O. Frey, G. K. Przybylski, et al. 2004. Developmental stage, phenotype, and migration distinguish naive- and effector/memory-like 38. Kriegel, M. A., T. Lohmann, C. Gabler, N. Blank, J. R. Kalden, and ϩ ϩ ϩ J. Exp. Med. H. M. Lorenz. 2004. Defective suppressor function of human CD4 CD25 reg- CD4 regulatory T cells. 199: 303–313. 48. Apostolou, I., A. Sarukhan, L. Klein, and H. von Boehmer. 2002. Origin of

ulatory T cells in autoimmune polyglandular syndrome type II. J. Exp. Med. 199: Downloaded from regulatory T cells with known speciﬁcity for antigen. Nat. Immunol. 3: 756–763. 1285–1291. 49. Apostolou, I., and H. von Boehmer. 2004. In vivo instruction of suppressor com- 39. Huber, S., C. Schramm, H. A. Lehr, A. Mann, S. Schmitt, C. Becker, mitment in naive T cells. J. Exp. Med. 199: 1401–1408. M. Protschka, P. R. Galle, M. F. Neurath, and M. Blessing. 2004. Cutting edge: 50. Thorstenson, K. M., and A. Khoruts. 2001. Generation of anergic and potentially TGF-␤ signaling is required for the in vivo expansion and immunosuppressive ϩ ϩ ϩ immunoregulatory CD25 CD4 T cells in vivo after induction of peripheral tol- capacity of regulatory CD4 CD25 T cells. J. Immunol. 173: 6526–6531. erance with intravenous or oral antigen. J. Immunol. 167: 188–195. 40. Marie, J. C., J. J. Letterio, M. Gavin, and A. Y. Rudensky. 2005. TGF-␤1 main- 51. Curotto de Lafaille, M. A., A. C. Lino, N. Kutchukhidze, and J. J. Lafaille. 2004. tains suppressor function and Foxp3 expression in CD4ϩCD25ϩ regulatory T CD25Ϫ T cells generate CD25ϩFoxp3ϩ regulatory T cells by peripheral expan-

cells. J. Exp. Med. 201: 1061–1067. sion. J. Immunol. 173: 7259–7268. http://www.jimmunol.org/ by guest on September 29, 2021