Autoreactive CD4 T Cells Inducing Long-Lasting Tolerance in Central

LF 15-0195 Inhibits the Development of Rat Central Nervous System Autoimmunity by Inducing Long-Lasting Tolerance in Autoreactive CD4 T Cells This information is current as of September 27, 2021. Valérie Duplan, Patrick Dutartre, Lennart T. Mars, Roland S. Liblau, Philippe Druet and Abdelhadi Saoudi J Immunol 2003; 170:2179-2185; ; doi: 10.4049/jimmunol.170.4.2179 http://www.jimmunol.org/content/170/4/2179 Downloaded from

References This article cites 39 articles, 13 of which you can access for free at:

http://www.jimmunol.org/content/170/4/2179.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• 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 by guest on September 27, 2021

*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 © 2003 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

LF 15-0195 Inhibits the Development of Rat Central Nervous System Autoimmunity by Inducing Long-Lasting Tolerance in Autoreactive CD4 T Cells1

Vale´rie Duplan,* Patrick Dutartre,† Lennart T. Mars,* Roland S. Liblau,* Philippe Druet,* and Abdelhadi Saoudi2*

Experimental autoimmune encephalomyelitis (EAE) is a T cell-dependent autoimmune disease induced in susceptible animals by a single immunization with myelin basic protein (MBP). LF 15-0195 is a novel immunosuppressor that has been shown to have a potent immunosuppressive effect in several pathological manifestations. The purpose of this study was to investigate the effect of this drug on the induction and progression of established rat EAE and to dissect the mechanisms involved. We show that LF Downloaded from 15-0195 administration at the time of MBP immunization reduces the incidence and severity of EAE in Lewis rats. This drug also inhibits ongoing and passively induced EAE, indicating that LF 15-0195 affects already differentiated pathogenic lymphocytes. Compared with lymph node cells from untreated rats, lymphocytes from MBP-immunized rats treated with LF 15-0195 proliferated equally well in response to MBP in vitro, while their ability to produce effector cytokines and to transfer EAE into syngeneic recipients was significantly reduced. This phenomenon is stable and long-lasting. Indeed, neither IL-12 nor repeated stimulation with naive APC and MBP in vitro rendered MBP-specific CD4 T cells from protected rats encephalitogenic. In conclusion, LF http://www.jimmunol.org/ 15-0195 treatment suppresses EAE by interfering with both the differentiation and effector functions of autoantigen-specific CD4 T cells. The Journal of Immunology, 2003, 170: 2179–2185.

utoimmune diseases result from dysregulation of the im- attractive. DSG has powerful immunosuppressive properties and dif- mune system that targets its own organs, tissues, and fers from other immunosuppressive agents, both structurally and A cells for destruction. These diseases include Ͼ70 mechanistically (1–4). DSG is frequently more effective than popular chronic disorders that collectively affect millions of individuals immunosuppressants such as cyclosporin A, FK 506, or rapamycin at worldwide. In general, these diseases are associated with humoral inducing immunosuppression (1, 5, 6). DSG can prolong the survival by guest on September 27, 2021 or cell-mediated immune reactions against one or more of the of, even induce tolerance to, tissue grafts in several animal models body’s own constituents. Based on exciting results in transplanta- involving either allo- or xeno-disparities (7–9). DSG has also been tion, a number of conventional immunosuppressants, i.e., glu- effective in decreasing the severity of autoimmune diseases (10). cocorticoid, azathioprine, and cyclosporin A, are also clinically However, DSG suffers from several drawbacks, such as its low chem- used for treating autoimmune diseases. Despite the widespread use ical stability in aqueous solution and its susceptibility to oxidative of these immunosuppressive drugs, a firm conclusion about effi- metabolism in vivo, which limit its clinical use, particularly in chronic cacy has been hard to come by. Furthermore, the long term ad- autoimmune diseases that may require long term administration. LF ministration of these drugs induces several side effects. Therefore, 15-0195, the recently developed analog of DSG devoid of the above- there is a need for new, powerful, and less toxic immunosuppressive mentioned drawbacks, has improved immunosuppressive activity drugs, which could lead to a more specific immunosuppression. compared with DSG (11). LF 15-0195, which is now being investi- Among the different compounds recently developed, 15-deoxy gated as a potential therapeutic drug for human inflammatory dis- spergualin (DSG),3 a synthetic analog of spergualin that was iso- eases, has demonstrated its efficacy in allotransplantation (12) and in lated as an antitumor compound from Bacillus laterosporus, appeared Ab-mediated autoimmune diseases (13, 14). However, its effect on cell-mediated autoimmunity has not yet been tested. *Institut National de la Santeét de la Recherche Me´dicale, Unite´563, Institut Fe´d- In the present study we evaluate the effect of LF 15-0195 treat- e´ratif de Recherche 30, and Universite´ Paul Sabatier, Hoˆpital Purpan, Toulouse, France; and †Laboratoires Fournier, Daix, France ment on the development of rat experimental autoimmune encephalomyelitis (EAE). This inflammatory autoimmune disease of the Received for publication July 22, 2002. Accepted for publication December 13, 2002. CNS is considered an experimental model for multiple sclerosis The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance (15, 16). EAE arises as a consequence of the breakdown of self- with 18 U.S.C. Section 1734 solely to indicate this fact. tolerance, induced by the immunization of susceptible animals 1 This work was supported by Institut National de la Santeét de la Recherche Me´di- with myelin-derived Ags emulsified in CFA or following adoptive cale, Fournier’s Laboratories, and the Association pour la Recherche sur la Sclérose en Plaques. A.S. is supported by Center National de la Recherche Scientifique, and transfer of T cell lines or clones specific for various myelin pro- V.D. is supported by Fournier’s Laboratories. teins (15, 17, 18). Our present data show that s.c. administration of 2 Address correspondence and reprint requests to Dr. Abdelhadi Saoudi, Centre de LF 15-0195 is effective in inhibiting the induction of rat EAE and Physiopathologie Toulouse-Purpan, Institut National de la Santeét de la Recherche the progression of established disease. Analysis of cellular and Me´dicale, Unite´ 563, Hopital Purpan, 31059 Toulouse Cedex, France. E-mail: [email protected] humoral immune responses revealed that the beneficial effect of LF 3 Abbreviations used in this paper used in this paper: DSG, deoxyspergualin; EAE, ex- 15-0195 treatment is mediated by the induction of a long-lasting perimental autoimmune encephalomyelitis; LEW, Lewis rats; MBP, myelin basic protein. tolerance in MBP-specific autoaggressive CD4 T cells.

Materials and Methods protein in the supernatants were measured by specific ELISA (29). Tran- ␣ ␤ Rats script levels of IL-4, TNF- , TGF- , and hypoxanthine phosphoribosyltransferase were quantified by real-time quantitative PCR using the SYBR Eight- to 10-wk-old male Lewis (LEW) rats were used in this study. These Green DNA dye (ABI PRISM 5700; PerkinElmer, PE Applied Biosystems, animals were obtained from the Center d’Elevage R. Janvier (Le Genest St. Foster City, CA) (29). Isle, France) and maintained in our animal house facility under specific pathogen-free conditions. All procedures were in accordance with national Generation of T cell lines regulations on animal experiments. Preparation and maintenance of T cell lines were performed as previously Induction, treatment, and clinical evaluation of EAE described (30) with some modifications. Briefly, popliteal lymph node cells, collected 12 days after immunization with MBP, were resuspended at To induce active EAE, LEW rats were injected in the hind footpads with 4 ϫ 106 viable cells/ml in complete medium containing 1% normal rat 10 ␮g of myelin basic protein (MBP) from guinea pig emulsified in CFA serum and MBP at 20 ␮g/ml. After 3 days, viable cells were separated containing 4 mg/ml heat-killed Mycobacterium tuberculosis H37Ra using Ficoll-Hypaque and resuspended at 5 ϫ 105/ml in complete medium (Difco, Detroit, MI). MBP was prepared in our laboratory as previously containing 10% FCS and 10 U/ml rat IL-2 produced by CHO-IL-2 cells. described (19). A total of 100 ␮l of MBP-CFA was divided equally be- After 7 days of a resting phase, the viable cells were recovered and resus- tween the rear footpads. To induce passive EAE, encephalitogenic T cells pended at 5 ϫ 105 viable cells/ml in complete medium with MBP (10 from MBP-immunized donors were adoptively transferred into naive syn- ␮g/ml) and irradiated syngeneic splenocytes (1.5 ϫ 106 cells/ml) as the geneic recipients as previously described (20). Briefly, lymph node cells source of APC. The same procedure was then followed after this second (popliteal and para-aortic) from MBP-immunized rats were collected round of Ag stimulation for at least two cycles. Before the adoptive trans- 10–12 days after MBP sensitization and stimulated in vitro for 3 days with fer, the T cell lines were stimulated with MBP at 2 ␮g/ml for 3 days in the either MBP (2 ␮g/ml) or Con A (1 ␮g/ml). Viable T cells separated from presence or the absence of recombinant mouse IL-12 at 1 ng/ml. dead cells by Ficoll-Hypaque were i.v. injected into syngeneic recipients. Downloaded from LF 15-0195 (Fournier Laboratories, Daix, France) was prepared in saline ELISA for MBP-specific IgG subclasses ␮ solution and adjusted at pH 7.2, and 300 l was administered s.c. at the For detection of MBP-specific Abs, microtiters plates (Falcon 3012; BD indicated regimen and concentration. Biosciences, Oxnard, CA) were coated overnight at 4°C with 2 ␮g/ml MBP Control rats received either LF 15-0195 or saline without immunization in PBS. Bound IgG1, IgG2a, and IgG2b anti-MBP were revealed using with MBP. Animals were scored daily for clinical signs of disease on a biotinylated mouse anti-rat ␥1, ␥2a, and ␥2b mAbs, respectively. The severity scale ranging from 0 to 5: 0, normal; 1, limp tail; 2, hind limb bound biotinylated mAbs were revealed by addition of preformed strepta- weakness; 3, unilateral hind limb paralysis; 4, bilateral hind limb paralysis; vidin-peroxidase complexes (Amersham, Slough, U.K.) for 60 min at room and 5, bilateral hind limb paralysis and incontinence. Clinical signs deter- temperature. For total IgG measurement, sera were incubated with perox- http://www.jimmunol.org/ mined to be between any of these stages were given an intermediate score. idase-conjugated sheep anti-rat IgG (provided by E. Druet, Toulouse, The results are presented as the mean maximal score and/or the mean France). The plates were washed and incubated with substrate 3,3Ј-5,5Ј- cumulative score, calculated as the sum of daily disease scores of each tetramethylbenzidine (Fluka Chemie, Buchs, Switzerland). The reaction individual animal. ␮ was stopped by adding 50 l/well H2SO4 2N, and absorbance was read at Abs, flow cytometry, and immunohistology 450 nm using an automated microplate ELISA reader (Emax; Molecular Devices, Menlo Park, CA). The mAbs used for flow cytometry and immunohistology were as follows: W3/25 (anti-rat CD4) (21), OX6 (anti-rat MHC class II) (22), OX8 (anti-rat Statistical analyses CD8) (23), OX21 (anti-human C3b inactivator) (24), and OX22 (anti-rat Results are expressed as the mean Ϯ SD, and differences between groups

CD45RC) (25). The hybridomas OX6, OX8, 0X21, OX22, and W3/25 by guest on September 27, 2021 were evaluated by Mann-Whitney U test. were provided by Dr. D. Mason (Oxford, U.K.). The hybridomas ED1 (anti-rat macrophage) (26) and R73 (anti-rat TCR␣␤) (27) were provided by Dr. C. Dijkstra (Amsterdam, The Netherlands) and Dr. T. Hu¨nig (Wurz- Results burg, Germany), respectively. The phenotype of cells was determined by LF 15-0195 treatment prevents the induction of active EAE in triple staining; cells were incubated with FITC-labeled, PE-labeled, and LEW rats biotinylated mAbs. Then streptavidin-CyChrome (BD PharMingen, San Diego, CA) was added. Data were collected on 10,000 cells as determined The results obtained are summarized in Table I. All 16 LEW rats by forward and side light scatter intensity on a XL Coulter cytometer immunized with MBP and injected with saline developed typical (Coultronics, Margency, France) and analyzed using CellQuest software clinical signs of acute EAE. The first clinical signs occurred 10–11 (BD Biosciences, Mountain View, CA). days after immunization, progressing to paraparesis, with maxi- For immunohistology, rats were killed on day 15 after MBP immunization (the peak of disease), and their spinal cords were embedded in cryo- mum severity around day 15. The disease was severe and killed compound (Leica, Heidelberger, Germany). Cryostat sections were pre- 50% of the animals. The surviving rats developed a self-limiting pared from the spinal cord (5 ␮m) and were stained using the disease that lasted for ϳ1 wk (group 1). In contrast, LEW rats immunoperoxidase method. Briefly, sections were fixed in ethanol, air- immunized with MBP and treated with a daily dose of 1 mg/kg LF dried, and incubated with ED1, R73, OX6, OX8, W3/25, or isotype control 15-0195 for 7 days (group 2) or 15 days (group 3), starting from mAbs for 30 min at room temperature. After further washes the slides were incubated with peroxidase-labeled polymer conjugated to goat anti- the day of MBP immunization, developed milder EAE compared mouse-Ig (DakoEnvision, Dako, Copenhagen, Denmark). Following fur- with controls. The onset of disease was significantly delayed ( p ϭ ther washes, bound Ab was revealed using 3,3Ј-diaminobenzidine-chro- 0.0006), and only 20% of the animals died from disease (groups 2 mogen solution. We also quantified leukocyte infiltration of the spinal cord and 3). Since the animals from group 3 started to develop EAE 5 of diseased rats on day 15 (peak of disease) and of protected animals on days 15 and 50 (20 days after treatment withdrawal) after MBP immuni- days after treatment withdrawal, we tested the effect of a prolonged zation as previously described (28). treatment (30 days), starting from the day of immunization, at two different doses. A very significant reduction of incidence and se- Proliferative response and cytokines assays verity of clinical EAE was obtained with a 30-day course of LF Popliteal and para-aortic lymph node cells, collected 12 days after immu- 15-0195 treatment at 1 mg/kg (group 5). Indeed, none of the an- nization with MBP, were stimulated with different concentrations of MBP imals died from disease, and only five animals among 19 devel- in 96-well culture plates (Costar, Cambridge, MA). The culture medium Ϯ Ϯ ϭ was RPMI 1640 (Life Technologies, Cergy Pontoise, France) containing oped delayed (19.3 1.5 vs 11.5 0.8; p 0.0006) and mild 10% FCS, 1% pyruvate, 1% nonessential amino acids, 1% L-glutamine, 1% (2.8 Ϯ 1.5 vs 4.9 Ϯ 0.3; p ϭ 0.0001) clinical EAE compared with penicillin-streptomycin, and 2 ϫ 10Ϫ5 M 2-ME. Proliferation was mea- controls. The protected animals were still free of clinical signs 3 3 sured by [ H]thymidine incorporation during the last 18 h of a 72-h culture mo after treatment withdrawal. Treatment with 0.3 mg/kg for 30 period. At various times throughout the culture (24, 48, and 72 h), supernatants were removed and stored at Ϫ20°C for cytokine determination. The days (group 4) was less efficient at preventing EAE than treatment cells were harvested following stimulation, and RNA was purified for anal- with 1 mg/kg for the same period. Indeed, the severity and inci- ysis of lymphokine gene expression by RT-PCR. IFN-␥, IL-2, and IL-10 dence of EAE were not significantly different from those of the The Journal of Immunology 2181

Table I. LF 15-0195 treatment prevents actively induced EAE

MBP Day of Onsetc Severity of Diseased Survival Groups Immunizationa Treatment Dosage and Durationb Incidence (%) (mean Ϯ SD) (mean Ϯ SD) (%)

1 ϩ Saline 16/16 (100) 11.5 Ϯ 0.8e 4.9 Ϯ 0.3e 50 2 ϩ LF 15-0195 (1 mg/kg, days 0–6) 5/5 (100) 14.2 Ϯ 0.5 4.0 Ϯ 1.2 80 3 ϩ LF 15-0195 (1 mg/kg, days 0–14) 5/5 (100) 20.0 Ϯ 0.7 4.0 Ϯ 1.2 80 4 ϩ LF 15-0195 (0.3 mg/kg, days 0–29) 6/6 (100) 11.0 Ϯ 0.0 4.5 Ϯ 0.8 100 5 ϩ LF 15-0195 (1 mg/kg, days 0–29) 5/19 (26) 19.3 Ϯ 1.5 2.8 Ϯ 1.5 100 6 Ϫ Saline or LF 15-0195 0/11 (0) 100

a LEW rats in groups 1–5 were immunized on day 0 with 10 ␮g of MBP emulsified in CFA. In group 6, the rats were not immunized. b LEW rats were treated daily by s.c. injections of PBS or LF 15-0195 as indicated. The unimmunized rats (group 6) were injected with saline or LF 15-0195 (1 or 0.3 mg/kg) for 15 or 30 days. c Delay between MBP immunization and onset of EAE. d Severity of active EAE is represented by the mean of the maximal clinical score (ϮSD) of diseased rats. e Statistical analyses: differences in day of onset and severity of active EAE between group 1 and groups 2, 3, and 5 are highly significant ( p Ͻ 0.01). Group 1 vs group 4 is not significant. control animals; nevertheless, none of the six rats of this protocol MBP-stimulated immune lymph node cells were transferred, re- died from disease. The control unimmunized LEW rats (group 6) cipients developed classical EAE, with maximal clinical score of Downloaded from injected with saline or LF 15-0195 (1 or 0.3 mg/kg) for 15 days or 3. Next, we tested the effect of LF 15-0195 treatment on passive 1 mo were free of disease and appeared healthy without any signs EAE by injecting the recipients daily with 1 mg/kg of this drug for of side effects. 10 days, starting the day before the transfer of encephalitogenic T Animals with clinical EAE present inflammatory infiltrates in cells. Fig. 2 shows that recipients exposed to LF 15-0195 for 10 their CNS, but infiltration of leukocytes is not always associated days exhibited a statistically significant ( p ϭ 0.0032) decrease in

with clinical signs of EAE (31, 32). Therefore, we analyzed EAE compared with control animals injected with the same effec- http://www.jimmunol.org/ whether the protection from EAE by LF 15-0195 is associated with tor cells without LF 15-0195 treatment. Moreover, lymph node inhibition of the development of inflammatory infiltrates in the cells from MBP-immunized LEW rats when restimulated in vitro CNS. Immunohistologic examination of the spinal cords of MBP- in the presence of LF 15-0195 had a reduced capacity to transfer immunized LEW rats, treated or not with LF 15-0195 at 1 mg/kg EAE (data not shown). Altogether these data indicate that LF 15- for 15 days, shows a marked reduction in the magnitude of leu- 0195 can block the pathogenic effect of differentiated autoantigen- kocyte infiltration in the protected animals. This reduction was specific T cells and suggest that this drug may be used to inhibit observed for CD4 and CD8T cells and macrophages, and there was the progression of established disease. a parallel reduction of MHC class II Ag expression on the micro- glia. Fig. 1A shows representative spinal cord sections in which Analysis of Ag-specific T cell responses in LF 15-0195-treated by guest on September 27, 2021 leukocyte infiltration and MHC Ag class II expression are ob- or untreated MBP-immunized LEW rats served in diseased rats, but not in protected rats. These results were To investigate the mechanisms involved in prevention of EAE, we confirmed by the quantification of leukocyte infiltration in the spi- compared the proliferative capacity and the cytokine profiles of nal cord after purification using a Percoll gradient (Fig. 1B). The immune lymph node cells from MBP-immunized LEW rats, CNS of LF 15-0195-treated rats were still free of leukocyte infil- treated daily, or not, with LF 15-0195 for a total of 12 days from trates 20 days after treatment withdrawal (Fig. 1B), indicating that, the day of MBP immunization. The proliferative response after unlike rats treated with other immunosuppressive drugs (33), LF MBP stimulation in vitro was similar in the two groups (Fig. 3A). 15-0195-protected rats did not develop relapsing disease after The analysis of Ag-specific type 1 (IFN-␥ and TNF-␣) and type 2 treatment withdrawal. In combination, our results show that LF (IL-4 and IL-10) cytokine production revealed that lymph node 15-0195 treatment prevents active EAE, and that the protection cells from LF 15-0195-treated animals produced significantly less afforded is associated with a markedly reduced encephalomyelitis, IFN-␥ (Fig. 3B) and IL-10 (Fig. 3C) and expressed low amounts of as evaluated histologically. TNF-␣ mRNA (Fig. 3D) compared with lymph node cells from untreated MBP-immunized LEW rats. The expression of TGF-␤ LF 15-0195 treatment also reduces the severity of ongoing and (Fig. 3E) and IL-4 (Fig. 3F) mRNA by immune lymph node cells passive EAE from protected and diseased rats is not significantly different. To study the effect of LF 15-0195 on the progression of ongoing Taken together, these data demonstrate that although MBP-specific EAE, daily treatment of MBP-immunized LEW rats with LF 15- T cells proliferated equally well in vitro, the production of effector 0195 was initiated on day 7 (before appearance of EAE clinical (IFN-␥ and TNF-␣) and regulatory (IL-10) cytokines was lower in signs) or day 11 (animals have at least a limp tail) after MBP LF 15-0195-treated compared with untreated rats. This difference immunization and continued for a period of 15 days. Table II was not observed when lymph node cells were stimulated with shows that during this treatment, MBP-immunized LEW rats de- Con A (data not shown), suggesting that this treatment did not veloped less severe clinical EAE (the maximal clinical score and have a general immunosuppressive effect. mortality were significantly reduced) compared with control MBP- immunized LEW rats. Treatment with LF 15-0195 at 2 mg/kg Analysis of Ag-specific Ab responses in LF 15-0195-treated or proved more efficient than 1 mg/kg, and the effect of LF 15-0195 untreated MBP-immunized LEW rats was more pronounced if started early, before the appearance of The sera obtained from MBP-immunized LEW rats, treated, or not, clinical EAE (day 7). These data suggest that LF 15-0195 may with LF 15-0195 for 30 days at 1 mg/kg, starting from the day of affect the pathogenicity of already differentiated effector lympho- immunization, were analyzed for the presence of MBP-specific cytes. This is further supported by the effect of LF 15-0195 on IgG. The anti-MBP IgG response was drastically reduced in the adoptively transferred EAE. We determined that when 25 ϫ 106 sera from LF 15-0195-treated animals (Fig. 4A), and this reduction 2182 BENEFICIAL EFFECT OF LF 15-0195 ON RAT EAE

FIGURE 2. LF 15-0195 treatment reduces the severity of passively induced EAE. Lymph nodes cells were removed from MBP-immunized rats on day 12 after MBP challenge and were stimulated in vitro with MBP for 3 days as indicated in Materials and Methods. Viable leukocytes (25 ϫ 106) were injected i.v. into each naive syngeneic recipient that was injected daily with either PBS (f; n ϭ 7) or LF 15-0195 at 1 mg/kg for 10 days (Ⅺ; n ϭ 7), starting 1 day before the transfer. The recipients were examined daily for signs of EAE. The results are expressed in A as the mean daily clinical score of each experimental group and in B as the mean cumulative Downloaded from disease score of each experimental group, calculated as the sum of the daily disease score for each individual animal. Results represent the pooled data -p Ͻ 0.01 (by Mann ,ءء ;p Ͻ 0.05 ,ء .of two independent experiments Whitney U test). http://www.jimmunol.org/ Taken together, the results obtained demonstrated that LF 15-0195 treatment affected both type 1- and type 2-associated Ag-speciﬁc IgG responses.

CD4 T cells from protected animals have a diminished capacity to transfer EAE As shown in Fig. 5, classical EAE developed in all seven naive LEW rats given 25 ϫ 106 MBP-stimulated lymph node cells from

FIGURE 1. LF 15-0195 treatment suppresses histological encephalo- rats immunized with MBP and injected with PBS (untreated by guest on September 27, 2021 myelitis in MBP-immunized LEW rats. Spinal cords were removed from group). Paralysis developed, as expected, 4 or 5 days after adoptive LEW rats that were immunized with MBP in CFA and treated, or not, with transfer. In contrast, naive LEW rats receiving 25 ϫ 106 MBP- LF 15-0195, starting from the day of MBP immunization. A, The photo- stimulated lymph node cells from protected rats (treated group) graphs show immunohistologic staining of spinal cord cryostat sections from a diseased rat (left panels) and a protected rat (right panels) after developed a very mild form of the disease, which appeared later treatment with LF 15-0195 at 1 mg/kg for 15 days. The sections are stained than in the controls (Fig. 5, A and B). Similar results were obtained for MHC class II Ag using OX6 mAb (upper panels) or for TCR␣␤ T cells when immune lymph node cells were stimulated with Con A in- using R73 mAb (lower panels). Magnification, ϫ400. B, The histograms stead of MBP (Fig. 5, C and D), indicating that polyclonal T cell show the absolute number of leukocytes that infiltrate the spinal cords of activation cannot overcome the MBP-specific CD4 T cell toler- naive (n ϭ 3), protected (n ϭ 3), and diseased (n ϭ 9) rats on the indicated ance. This reduced ability of lymph node cells from protected rats days after MBP immunization. Results are expressed as the mean value to transfer EAE is a stable phenomenon that is persistent despite obtained from three to seven individual rats per group. repeated in vitro stimulation with the autoantigen and APC from syngeneic naive rats. Indeed, MBP-specific T cell lines derived from LF 15-0195-protected rats, after five rounds of stimulation, concerns all the Ag-specific IgG subclasses tested (Fig. 4, BÐD). have a reduced capacity to transfer EAE compared with those ob- The control nonimmunized LEW rats, treated, or not, with LF tained from diseased rats (Fig. 6, A and B, six recipients in each 15-0195, did not show any Ab response to MBP (data not shown). group). Although these MBP-specific T cells lines proliferated

Table II. Effect of LF 15-0195 treatment on the progression of actively induced EAE

Day of Onsetb Severity of Diseasec Survival Groupsa Treatment Dosage and Duration Incidence (%) (mean Ϯ SD) (mean Ϯ SD) (%)

1 Saline 13/13 (100) 11.3 Ϯ 0.8 5.0 Ϯ 0.0d 33 2 LF 15-0195 (1 mg/kg, days 7–21) 5/5 (100) 11.4 Ϯ 0.6 2.4 Ϯ 1.5 100 3 LF 15-0195 (1 mg/kg, days 11–25) 5/5 (100) 12.2 Ϯ 1.1 4.0 Ϯ 0.7 80 4 LF 15-0195 (2 mg/kg, days 7–21) 1/4 (25) 11 1 100 5 LF 15-0195 (2 mg/kg, days 11–25) 4/4 (100) 11.0 Ϯ 0.0 2.0 Ϯ 2.0 75

a LEW rats in groups 1–5 were immunized on day 0 with 10 ␮g of MBP emulsiﬁed in CFA. b Delay between MBP immunization and onset of EAE. c Severity of active EAE is represented by the mean of the maximal clinical score (ϮSD) of diseased rats. d Statistical analyses: differences in severity of EAE between group 1 and groups 2–5 are highly signiﬁcant ( p Ͻ 0.01). The Journal of Immunology 2183

FIGURE 3. LF 15-0195 treatment reduces the MBP-specific cellular immune response. Lymph node cells from MBP-immunized LEW rats treated (Ⅺ) f or not treated ( ) with LF 15-0195 (1 mg/kg) for 12 days starting from the day of MBP sensitization were collected and stimulated in vitro with the indicated Downloaded from concentrations of MBP. A, Proliferation was assessed with an 18-h [3H]thymidine pulse added after 48 h of culture, and results are expressed as the mean [3H]thymidine incorporation adjusted for background proliferation (⌬cpm) Ϯ SD. Tissue culture supernatants were assayed at 48 h for IFN-␥ (B) and IL-10 (C) proteins using capture ELISA. The results of one representative experiment of three are shown and are expressed as the mean (ϮSD) of values obtained from four individual rats per group. TNF-␣ (D), TGF-␤ (E), and IL-4 (F) mRNA expression were assayed by quantitative RT-PCR in lymph node cells following MBP stimulation (20 ␮g/ml) for 24 h. Results are expressed as the mean cytokine/hypoxanthine phosphoribosyltransferase ratio Ϯ SD of values .(p Ͻ 0.05 (by Mann-Whitney U test ,ء .obtained from four individual rats in each group and represent two independent experiments http://www.jimmunol.org/ equally well in vitro, the production of IL-10 and IFN-␥ was lower significantly the encephalogenicity of MBP-specific T cell lines in T cell lines obtained from protected rats compared with those obtained from diseased rats (Fig. 6C). Taken together, these data obtained from diseased animals (data not shown). We also tested show that LF 15-0195 induces tolerance in MBP-specific CD4 T whether the in vitro stimulation of these lines in the presence of the Th1-promoting cytokine, IL-12, would increase their ability to transfer EAE. We observed that although IL-12 increases the ability of the T cell lines from protected rats to produce type 1 effector

cytokines (data not shown), it does not unmask their reduced abil- by guest on September 27, 2021 ity to transfer EAE (Fig. 6C). In contrast, this cytokine increases

FIGURE 5. MBP-speciﬁc CD4 T cells from LF 15-0195-treated rats have a diminished capacity to transfer EAE. Lymph node cells from animals in which EAE was completely suppressed by LF 15-0195 treatment at 1 mg/kg, starting from the day of MBP challenge (Ⅺ; n ϭ 11 for A and B and n ϭ 4 for C and D) or from control PBS-treated diseased rats (f; n ϭ 7 for A and B and n ϭ 4 for C and D) were adoptively transferred. Lymph nodes were removed on day 10 or 12 following MBP immunization, and cells were either cultured in vitro with MBP (2 ␮g/ml) or Con A (1 ␮g/ml) FIGURE 4. LF 15-0195 treatment reduces MBP-speciﬁc IgG responses. for 3 days. Viable MBP-stimulated leukocytes (25 ϫ 106; A and B)or108 LEW rats were immunized with MBP in CFA and treated (Ⅺ; six rats) or not viable Con A-stimulated lymphocytes (C and D) were injected i.v. into treated (f; four rats) with LF 15-0195 at 1 mg/kg for 30 days, starting from the each naive syngeneic recipient. The results shown represent in A and C the day of immunization. The IgG (A), IgG1 (B), IgG2a (C), and IgG2b (D) anti- mean clinical scores for paralysis of each experimental group and in B and MBP Ab titers were measured by ELISA on days 5, 11, 17, 24, 31, and 39 after D the mean total disease score of each experimental group. Results in A and immunization. Results are expressed as absorbance at 450 nm of a pool of sera B are from two independent experiments, and those in C and D are from .(p Ͻ 0.01 (by Mann-Whitney U test ,ءء ;p Ͻ 0.05 ,ء .diluted 1/800 for total IgG and 1/1600 for IgG subclasses. one experiment 2184 BENEFICIAL EFFECT OF LF 15-0195 ON RAT EAE

ing the activation of autoreactive lymphocytes, the beneficial effect of LF 15-0195 may be due to a direct effect of this drug on T cell activation or to an indirect effect by interfering with APC functions. It has been shown that DSG inhibits the T cell proliferation induced by T cell mitogens or alloantigens (34, 35) and the generation of secondary CTLs (36). Furthermore, DSG inhibits Ag- specific T cell proliferation at the level of the Ag-presenting monocytes by interfering with the capacity of these cells to process and present Ags (37). Herein we have excluded that the protective effect of LF 15-0195 is attributed to an inhibition of the proliferation of autoreactive T lymphocytes. The immune lymph node cells from protected or diseased animals proliferate equally after stimulation with MBP or a dominant peptide of this Ag. Further- more, we observed that dendritic cells from LF 15-0195-treated animals present Ag as efficiently as those from untreated animals to MBP-specific T cell lines (unpublished observations). These results suggest that the beneficial effect of LF 15-0195 on EAE may not be mediated by a defect in Ag presentation by profes-

sional APC or by induction of anergy in Ag-specific CD4 T cells. Downloaded from In EAE the immunization with MBP elicits an autoimmune response toward myelin Ag, whereafter autoreactive T lymphocytes FIGURE 6. LF 15-0195 treatment induces a long-lasting tolerance in and macrophages migrate into the CNS and produce inflammatory MBP-specific CD4 T cells. A and B, Adoptive transfer of EAE with MBP- cytokines such as IFN-␥, TNF-␣, and lymphotoxin. These inflam- specific T cell lines generated after five rounds of stimulation with MBP, matory cytokines have crucial roles in initiating and perpetuating

APC, and IL-2 as described in Materials and Methods. Naive syngeneic CNS inflammation. In the present study we show that although http://www.jimmunol.org/ 6 recipients were injected i.v. with 10 viable T cell lines generated from immune lymph node cells from LF 15-0195-protected animals pro- Ⅺ lymph node cells from MBP-immunized and LF 15-0195-treated rats ( ; liferate equally well in response to MBP in vitro, they produce ϭ f ϭ n 6) or from control, MBP-immunized, untreated rats ( ; n 6). The significantly lower amounts of effector cytokines (IFN-␥ and results represent in A the mean clinical score of each experimental group TNF-␣) compared with those from untreated MBP-immunized and in B the mean cumulative disease score of each experimental group. Results are derived from two independent experiments. C, T cell lines LEW rats. This may explain their inability to induce CNS inflam- generated from protected (Ⅺ; four recipients in each group) and diseased mation and to recruit other inflammatory leukocytes in the CNS. animals (f; four recipients in each group) were stimulated with MBP in the LF 15-0195 may therefore mediate some of its immunosuppressive presence or the absence of IL-12 for 3 days before the transfer. Viable T properties by interfering with the ability of autoantigen-specificT cells (106) were injected i.v. into each naive syngeneic recipient. The re- cells to differentiate into effector cytokine-producing cells. How- by guest on September 27, 2021 sults show the mean cumulative disease score of each experimental group. ever, LF 15-0195 reduces the severity of both ongoing and pas- p Ͻ 0.01 (by Mann-Whitney U test). sively induced EAE, suggesting that this compound also affects ,ءء ;p Ͻ 0.05 ,ء already differentiated pathogenic CD4 T cells. Whether this effect is direct or indirect remains to be elucidated. cells, and this phenomenon is stable and irreversible by exogenous Interestingly, our data show that LF 15-0195 treatment induces IL-12. a stable and persistent tolerance in Ag-specific CD4 T cells. In- deed, the T cell lines obtained from LF 15-0195-treated animals, Discussion despite having been repeatedly restimulated in vitro in the absence In the present study we show that LF 15-0195 prevents clinical of LF 15-0195, were unable to transfer severe EAE. Furthermore, signs of actively induced EAE when administrated daily at 1 mg/ the stimulation of these lines with IL-12, a key cytokine for the kg/day s.c. from the day of MBP immunization of LEW rats. The differentiation of T cells into Th1, did not render these lines patho- dose and the duration of treatment are critical parameters. A 15- genic. Recently, it has been demonstrated in fully MHC-mis- day treatment is less efficient than the 30-day treatment, and a matched heart allograft transplantation that LF 15-0195 treatment dosage of 1 mg/kg is more potent than 0.3 mg/kg in preventing induces graft-specific tolerance (38). This tolerance requires the EAE. Continuous injections of this compound are not necessary, presence of donor APCs during LF 15-0195 injections, indicating since rats injected with LF 15-0195 for 30 days were still free of that Ag stimulation is required during LF 15-0195 treatment for clinical symptoms of EAE 3 mo after treatment withdrawal. How- promoting Ag-specific tolerance. Therefore, LF 15-0195 may sup- ever, a 15-day treatment is not sufficient to induce this long-lasting press any B and T immune response in a specific manner provided protection, since the animals develop EAE once the treatment the Ag is presented to the immune system during the treatment stopped. LF 15-0195 treatment reduces also the severity of ongo- period. Our present results support and extend these conclusions to ing and passively induced EAE, indicating that this drug may op- self-reactive lymphocytes. Indeed, the beneficial effect of LF 15- erate after pathogenic autoreactive T cells have differentiated. In 0195 in our study is not mediated by nonspecific immunosuppres- agreement with this observation, we showed recently that in vitro sion, since the inhibition of effector cytokine production is ob- stimulation of immune lymph node cells from MBP-immunized served only when immune lymph node CD4 T cells are stimulated rats with the autoantigen in the presence of LF 15-0195 reduced with MBP, but not after stimulation with the T cell mitogen Con their ability to transfer EAE (data not shown). A. Similar results are obtained in another animal model of auto- LF 15-0195 may affect a number of events involved in the immunity, experimental autoimmune myasthenia gravis (14). In pathogenesis of EAE, such as the activation of autoreactive T cells, the model of allo-transplantation, the tolerance induced by LF 15- the homing of these T cells into the CNS, and the interactions 0195 is mediated by Ag-specific regulatory T cells that express between infiltrating T cells and local parenchymal cells. Concern- CD25, proliferated weakly in response to alloantigen and transfer The Journal of Immunology 2185 tolerance to second syngeneic recipients (38). In contrast, there are 14. Duplan, V., P. Dutartre, P. Druet, and A. Saoudi. 2002. LF 15-0195 prevents from several lines of evidence indicating that, at least during the first 2 the development and inhibits the progression of rat experimental autoimmune myasthenia gravis. J. Neuroimmunol. 129:115. wk of LF 15-0195 treatment, regulatory T cells are not involved in 15. Raine, C. S. 1984. Analysis of autoimmune demyelination: its impact upon mul- the protection from EAE: 1) CD4 T cells from protected and dis- tiple sclerosis. Lab. Invest. 50:608. eased rats express similar levels of CD25 and proliferate in a sim- 16. Steinman, L. 1996. Multiple sclerosis: a coordinated immunological attack against myelin in the central nervous system. Cell 95:299. ilar way; 2) in a cotransfer experiment, T cell lines from protected 17. Mason, D. 1991. Genetic variation in the stress response: susceptibilty to exper- animals are unable to prevent the encephalitogenic anti-MBP T imental allergic encephalomyelitis and implications for human inflammatory dis- cells to transfer EAE (our unpublished observations); 3) the neu- ease. Immunol. Today 12:57. ␤ 18. Swanborg, R. 1995. Experimental autoimmune encephalomyelitis in rodents as a tralization of TGF- , a key regulatory cytokine, during LF 15- model for human demyelinating disease. Clin. Immunol. Immunopathol. 77:4. 0195 treatment did not abrogate the protection (our unpublished 19. Brostoff, S. W., and D. W. Mason. 1984. Experimental allergic encephalomyeli- observations); 4) MBP-specific CD4 T cells from LF 15-0195 tis: successful treatment in vivo with a monoclonal antibody that recognizes T helper cells. J. Immunol. 133:1938. treatment produced lower amounts of IL-10, a cytokine produced 20. Sedgwick, J. D., and D. W. Mason. 1986. The mechanism of inhibition of ex- and involved in the suppressive effect mediated by regulatory T perimental allergic encephalomyelitis in the rat by monoclonal antibody against cells (39); and 5) LF 15-0195 reduces the severity of ongoing and CD4. J. Neuroimmunol. 13:217. 21. Williams, A. F., G. Galfre´, and C. Milstein. 1977. Analysis of cell surface by passively induced EAE; this effect is observed after a short period xenogeneic myeloma-hybrid antibodies: differentiation antigens of rat lympho- of treatment not consistent with generation of regulatory T cells. cytes. Cell 12:663. In conclusion, we demonstrated that the immunosuppressant LF 22. McMaster, W. R., and A. F. Williams. 1979. Identification of Ia glycoproteins in rat thymus and purification from rat spleen. Eur. J. Immunol. 9:426. 15-0195 can be used as a preventive and curative treatment in 23. Brideau, R. J., P. B. Carter, W. R. McMaster, and M. Webb. 1980. Two subsets experimental autoimmunity. The beneficial effect of this drug is of rat T lymphocytes defined with monoclonal antibodies. Eur. J. Immunol. 10: Downloaded from accompanied by decreased Ag-specific T cell responses. Although 609. 24. Hsiung, L. M., A. N. Barclay, M. R. Brandon, E. Sim, and R. R. Porter. 1982. the molecular mechanisms underlying the effect of LF 15-0195 Purification of human C3b inactivator by monoclonal-antibody affinity chroma- have yet to be elucidated, our data suggest that LF 15-0195 is a tography. Biochem. J. 203:293. promising therapeutic reagent for the treatment of human autoim- 25. Spickett, G. P., M. R. Brandon, D. W. Mason, A. F. Williams, and G. R. Woollett. 1983. MRC OX-22, a monoclonal antibody that labels a new subset of T lym- mune diseases. phocytes and reacts with the high molecular weight form of the leukocyte-com-

mon antigen. J. Exp. Med. 158:795. http://www.jimmunol.org/ Acknowledgments 26. Dijkstra, C. D., E. A. Dopp, P. Joling, and G. Kraal. 1985. The heterogeneity of We thank Isabelle Bernard for excellent technical assistance, and Maryline mononuclear phagocytes in lymphoid organs: distinct macrophage subpopula- tions in the rat recognized by monoclonal antibodies ED1, ED2 and ED3. Im- Calise and Patrick Aregui (IFR 30, Toulouse, France) for taking care of the munology 54:589. animal house. 27. Hunig, T., H.-J. Wallny, J. K. Hartley, A. Lawetzky, and G. Tiefenthaler. 1989. A monoclonal antibody to a constant determinant of the rat T cell antigen receptor References that induces T cell activation: differential reactivity with subsets of immature and 1. Kasai, M., T. Higa, T. Naohara, N. Masauzi, M. Watanabe, A. Matsuura, mature T lymphocytes. J. Exp. Med. 169:73. H. Ohizumi, J. Tanaka, K. Morii, Y. Kiyama, et al. 1994. 15-Deoxyspergualin 28. Sedgwick, J. D., S. Schwender, H. Imrich, R. Dorries, G. W. Butcher, and controls cyclosporin- and steroid-resistant intestinal acute graft-versus-host dis- V. ter Meulen. 1991. Isolation and direct characterization of resident microglial cells from the normal and inﬂamed central nervous system. Proc. Natl. Acad. Sci.

ease after allogeneic bone marrow transplantation. Bone Marrow Transplant. by guest on September 27, 2021 14:315. USA 88:7438. 2. Nadler, S. G., M. A. Tepper, B. Schacter, and C. E. Mazzucco. 1992. Interaction 29. Cautain, B., J. Damoiseaux, I. Bernard, E. Xystrakis, E. Fournie, P. van Breda of the immunosuppressant deoxyspergualin with a member of the Hsp70 family Vriesman, P. Druet, and A. Saoudi. 2002. The CD8 T cell compartment plays a of heat shock proteins. Science 258:484. dominant role in the deficiency of Brown-Norway rats to mount a proper type 1 3. Schreiber, S. L. 1991. Chemistry and biology of the immunophilins and their immune response. J. Immunol. 168:162. immunosuppressive ligands. Science 251:283. 30. Ben-Nun, A., H. Wekerle, and I. R. Cohen. 1981. Vaccination against autoim- 4. Tepper, M. A., S. Nadler, C. Mazzucco, C. Singh, and S. L. Kelley. 1993. 15- mune encephalomyelitis with T-lymphocyte line cells reactive against myelin Deoxyspergualin, a novel immunosuppressive drug: studies of the mechanism of basic protein. Nature 292:60. action. Ann. NY Acad. Sci. 685:136. 31. Happ, M. P., P. Wettstein, B. Dietzschold, and E. Heber-Katz. 1988. Genetic 5. Yuh, D. D., and R. E. Morris. 1993. The immunopharmacology of immunosup- control of the development of experimental allergic encephalomyelitis in rats. pression by 15-deoxyspergualin. Transplantation 55:578. Separation of MHC and non-MHC effects. J. Immunol. 141:1489. 6. Ohlman, S., G. Gannedahl, G. Tyden, G. Tufveson, and C. G. Groth. 1992. 32. Roth, M. P., C. Viratelle, L. Dolbois, M. Delverdier, N. Borot, L. Pelletier, Treatment of renal transplant rejection with 15-deoxyspergualin: a dose-finding P. Druet, M. Clanet, and H. Coppin. 1999. Susceptibility to experimental auto- study in man. Transplant. Proc. 24:318. immune encephalomyelitis in (LEXxBN)F2 rats is controlled by loci in the IL-4 7. Reichenspurner, H., A. Hildebrandt, P. A. Human, D. H. Boehm, A. G. Rose, and IL-6 regions. J. Immunol. 162:1917. J. A. Odell, B. Reichart, and H. U. Schorlemmer. 1990. 15-Deoxyspergualin for 33. McCombe, P. A., J. Harness, and M. P. Pender. 1999. Effects of cyclosporin A induction of graft nonreactivity after cardiac and renal allotransplantation in pri- treatment on clinical course and inflammatory cell apoptosis in experimental mates. Transplantation 50:181. autoimmune encephalomyelitis induced in Lewis rats by inoculation with myelin 8. Nemoto, K., M. Hayashi, J. Ito, Y. Sugawara, T. Mae, H. Fujii, F. Abe, A. Fujii, basic protein. J. Neuroimmunol. 97:60. and T. Takeuchi. 1991. Deoxyspergualin in lethal murine graft-versus-host dis- 34. Fujii, H., T. Takada, K. Nemoto, F. Abe, A. Fujii, and T. Takeuchi. 1989. In vitro ease. Transplantation 51:712. immunosuppressive properties of spergualins to murine T cell response. 9. Thomas, F., K. Pittman, T. Ljung, and E. Cekada. 1995. Deoxyspergualin is a J. Antibiot. 42:788. unique immunosuppressive agent with selective utility in inducing tolerance to 35. Kerr, P. G., and R. C. Atkins. 1989. The effects of deoxyspergualin on lympho- pancreas islet xenografts. Transplant. Proc. 27:417. cytes and monocytes in vivo and in vitro. Transplantation 48:1048. 10. Schorlemmer, H. U., and G. Dickneite. 1993. Preclinical studies with 15-deoxyspergualin in various animal models for autoimmune diseases. Ann. NY Acad. 36. Nishimura, K., and T. Tokunaga. 1989. Mechanism of action of 15-deoxysper- Sci. 685:155. gualin. I. Suppressive effect on the induction of alloreactive secondary cytotoxic 11. Lebreton, L., E. Jost, B. Carboni, J. Annat, M. Vaultier, P. Dutartre, and T lymphocytes in vivo and in vitro. Immunology 68:66. P. Renaut. 1999. Structure-immunosuppressive activity relationships of new an- 37. Hoeger, P. H., M. A. Tepper, A. Faith, J. A. Higgins, J. R. Lamb, and R. S. Geha. alogues of 15-deoxyspergualin. II. Structural modifications of the spermidine 1994. Immunosuppressant deoxyspergualin inhibits antigen processing in mono- moiety. J. Med. Chem. 42:4749. cytes. J. Immunol. 153:3908. 12. Chiffoleau, E., G. Beriou, P. Dutartre, C. Usal, J. P. Soulillou, and M. C. Cuturi. 38. Chiffoleau, E., G. Beriou, P. Dutartre, C. Usal, J. P. Soulillou, and M. C. Cuturi. ϩ 2002. Induction of donor-specific allograft tolerance by short-term treatment with 2002. Role for thymic and splenic regulatory CD4 T cells induced by donor LF15-0195 after transplantation: evidence for a direct effect on T-cell differen- dendritic cells in allograft tolerance by LF15-0195 treatment. J. Immunol. 168: tiation. Am. J. Transplant. 2:745. 5058. 13. Tesch, G. H., P. A. Hill, M. Wei, D. J. Nikolic-Paterson, P. Dutartre, and 39. Asseman, C., S. Mauze, M. W. Leach, R. L. Coffman, and F. Powrie. 1999. An R. C. Atkins. 2001. LF15-0195 prevents the induction and inhibits the progres- essential role for interleukin 10 in the function of regulatory T cells that inhibit sion of rat anti-GBM disease. Kidney Int. 60:1354. intestinal inflammation. J. Exp. Med. 190:995.