Critical Role for IL-4 and IL-10 Th1

Generation and Regulation of Human Th1-Biased Immune Responses In Vivo: A Critical Role for IL-4 and IL-10

This information is current as Alla Skapenko, Gerald U. Niedobitek, Joachim R. Kalden, of September 28, 2021. Peter E. Lipsky and Hendrik Schulze-Koops J Immunol 2004; 172:6427-6434; ; doi: 10.4049/jimmunol.172.10.6427 http://www.jimmunol.org/content/172/10/6427 Downloaded from

References This article cites 51 articles, 25 of which you can access for free at: http://www.jimmunol.org/content/172/10/6427.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 28, 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 © 2004 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Generation and Regulation of Human Th1-Biased Immune Responses In Vivo: A Critical Role for IL-4 and IL-101

Alla Skapenko,*† Gerald U. Niedobitek,‡ Joachim R. Kalden,† Peter E. Lipsky,§ and Hendrik Schulze-Koops2*†

Tissue damage in many human autoimmune diseases is mediated by activated autoantigen-specific Th1 cells. Delineation of the regulatory mechanisms controlling a Th1-biased human immune reaction and its pathologic potential is, therefore, a critical step in the understanding of autoimmune diseases. In this study, we introduce a novel means to investigate human Th1-biased immune responses in vivo. Intraperitoneal injection of human mononuclear cells into immunodeficient mice generates a xenogeneic Th1- biased human immune response characterized by systemic inflammation and leukocytic infiltrates with a granuloma-like architecture in the liver, and the perigastrointestinal and perirenal fatty tissue. Th1 cell activation was dependent on the presence of

APCs and could be blocked by cyclosporine. Importantly, neutralization of endogenously produced IL-4 and IL-10 markedly Downloaded from exaggerated the immune response, whereas exogenous IL-4 and IL-10 inhibited systemic Th1 immunity. Thus, the model described in this paper presents a useful means to analyze the regulation of human immune reactions in an in vivo situation. The results suggest that both IL-4 and IL-10 contribute to controlling the development of a human Th1-biased immune reaction. The Journal of Immunology, 2004, 172: 6427–6434.

3 utoimmune diseases are caused by failure of self-toler- and of experimental autoimmune encephalomyelitis (EAE), a http://www.jimmunol.org/ ance and subsequent immune responses against autolo- model of human multiple sclerosis (8). Furthermore, treatment A gous Ags (1). Convincing evidence exists that self-tol- with rIL-4 induces a switch from a Th1-type to a Th2-type re- erance is an active dynamic state in which potentially pathogenic sponse and prevents proteoglycan-induced arthritis, a different autoreactive cells are prevented from causing disease by regulatory model of human inflammatory arthritis (9). mechanisms (2). The breakdown of such mechanisms might, there- The anti-inflammatory role of another immunomodulatory cy- fore, result in the development of pathologic autoimmune reac- tokine, IL-10, has also been shown in animal models. IL-10-defi- tions. It has become apparent that the destructive effector mecha- cient mice are more susceptible to EAE when compared with wild- nisms of many systemic autoimmmune diseases are mediated by type mice (10). Diabetes induced by adoptively transferred activated autoantigen-specific Th1 cells (2, 3). Therefore, the lymphocytes into NOD mice can be prevented by IL-10-trans- by guest on September 28, 2021 mechanisms controlling the evolution of Th1-biased immune reduced islet-specific Th1 lymphocytes (11). Moreover, the effect of sponses play a critical role in the development of pathogenic au- regulatory T cells in a transfer model of colitis can be abrogated by toimmune reactions. neutralizing Abs to TGF-␤ and IL-10, resulting in the emergence Delineation of the mechanisms controlling Th1-mediated im- of tissue pathology (12, 13). Although the pathways are complex, munity has largely been derived from animal models. For example, there is convincing evidence that anti-inflammatory cytokines play amelioration of autoimmune diabetes in nonobese diabetic (NOD) essential roles in regulating the development and perpetuation of mice, a murine model of human insulin-dependent diabetes mel- chronic Th1-mediated autoimmune responses in animals. litus, is associated with increased expression of the Th2-derived Analysis of the role of cytokines in regulating T cell immunity cytokines IL-4 and IL-5 (4, 5). Moreover, pancreatic expression of in humans is hampered by the inability to address these questions IL-4 completely prevents diabetes in NOD mice (6). Injection of directly in vivo. In vitro, IL-4 and IL-10 clearly exhibit an anti- IL-4-transduced cells reduces the incidence and severity of colla- inflammatory effect, because they induce expression of the IL-1R gen-induced arthritis, a model of human inflammatory arthritis (7), antagonist (14, 15) and down-regulate the production of proinflammatory cytokines, such as IL-1 and TNF from human monocytes (16, 17). Moreover, IL-4 has a direct inhibitory effect on the de- *Nikolaus Fiebiger Center for Molecular Medicine, Clinical Research Group III, †De- velopment of human Th1 cells (18), and IL-10 is able to prevent partment of Internal Medicine III and Institute for Clinical Immunology, and ‡De- partment of Pathology, University of Erlangen-Nuremberg, Erlangen, Germany; and Th1 effector functions by induction of long-lasting T cell unre- ¤National Institute of Arthritis and Musculoskeletal and Skin Diseases, Bethesda, MD sponsiveness (19). Recent studies aiming to dissect the impact of 20892 new treatment approaches in human autoimmune diseases have Received for publication August 29, 2003. Accepted for publication March 16, 2004. revealed that clinical benefit may be associated with enhanced Th2 The costs of publication of this article were defrayed in part by the payment of page cell differentiation in vivo (20Ð22). Moreover, administration of charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. IL-4 or IL-10 to patients with psoriasis resulted in improvement of skin disease (23Ð25). In contrast, treatment of rheumatoid arthritis 1 This study was supported by the Deutsche Forschungsgemeinschaft, Grants DFG Schu 786/2-3 and 2-4, the German Society for Immunology, and the Interdisciplinary (RA), a prototype human Th1-biased autoimmune disease, with Center for Clinical Research at the University of Erlangen-Nuremberg (Project B27). IL-4 or IL-10 has largely failed to down-modulate inflammation 2 Address correspondence and reprint requests to Dr. Hendrik Schulze-Koops, Niko- laus Fiebiger Center for Molecular Medicine, Clinical Research Group III, University of Erlangen-Nuremberg, Department of Internal Medicine III and Institute for Clinical Immunology, Glueckstrasse 6, 91054 Erlangen, Germany. E-mail address: Schulze- 3 Abbreviations used in this paper: EAE, experimental autoimmune encephalomyeli- [email protected] tis; RA, rheumatoid arthritis; GvH, graft vs host.

and to provide clinical benefit (26). Therefore, the precise mech- Cytokine determination anisms of regulation of Th1-mediated immune reactions in humans Human cells were recovered from the peritoneal cavity of the mice. To remain obscure, and as a result, the impact of targeted interven- assess the acquired capacity of T cells for cytokine production, 2 ϫ 106 tions remains unpredictable. recovered cells were restimulated with ionomycin (1 mM; Calbiochem, To address these issues, we sought to develop a model in which Schwalbach, Germany) and PMA (20 ng/ml; Sigma-Aldrich) for 5 h in the ␮ human Th1-biased immune responses could be induced and ana- presence of 2 M monensin (Sigma-Aldrich). Cells were fixed with 4% paraformaldehyde (Sigma-Aldrich), and cytoplasmic human IFN-␥ and lyzed in vivo. Inducing T cell activation by transfer of human T IL-4 were detected by flow cytometry after intracellular staining with cells and APC into SCID mice provided an appropriate model FITC-labeled anti-IFN-␥ and PE-labeled anti-IL-4. The numbers of cyto- system. In this model, a Th1-biased immune response developed kine-producing T cells were determined from the total population of gated that was associated with lymphocytic infiltrations into several or- lymphocytes. Analysis of extracellular markers revealed that recovered lymphocytes contained Ͻ1% CD19-positive B cells. gans. The in vivo evolution of the human Th1 immune response To analyze the serum levels of the human inflammatory cytokines IFN-␥ was tightly regulated by endogenously produced IL-4 and IL-10. and TNF and of human IL-4, blood was taken from the tail vein of the Further analysis of the reaction showed that both cytokines admin- mice, sera were collected, and the cytokine levels were measured using istered exogenously lead to a down-modulation of human Th1 im- commercially available high-sensitivity ELISA kits that were highly spe- munity. Using this novel in vivo model to understand the regula- cific for the human cytokines (sensitivity thresholds, 0.12, 8, and 0.13 pg/ml for TNF, IFN-␥, and IL-4, respectively; R&D Systems). tion of a Th1-biased immune response should provide important insights into the regulation of human immune reactions and their pathologic consequences. Histopathologic analysis

Mouse tissues were sampled immediately after sacrifice, fixed in 5% neu- Downloaded from tral buffered formalin, and embedded in paraffin wax using standard his- Materials and Methods tological procedures. Three-micrometer paraffin sections were stained with Reagents and Abs H&E for morphological assessment by light microscopy. For immunohistochemistry, a polyclonal CD3 antiserum, monoclonal The following mAbs were used for purification and staining of human CD4, CD8, CD20, and CD68 Abs (all specific for human Ags), and a cells: anti-CD16, anti-CD19; FITC-conjugated anti-CD3, PE-labeled anti- monoclonal eosinophilic peroxidase Ab (all from DAKO) were used. Par- CD4, FITC-labeled anti-CD4 (Sigma-Aldrich, Taufkirchen, Germany);

afﬁn sections were dewaxed and subjected to Ag retrieval in 0.1 M citrate http://www.jimmunol.org/ FITC-labeled anti-CD14, PE-labeled anti-CD25 (Cymbus Biotechnology, buffer (pH 6.0) using a pressure cooker. Following incubation with appro- Hants, U.K.); FITC-labeled anti-HLA-DR (DAKO, Hamburg, Germany); priately diluted primary Abs, sections were incubated with a biotin-labeled PE-labeled anti-CD8, PE-labeled anti-IL-4 (MP4-25D2), and FITC-labeled goat anti-rabbit serum (for CD3; DAKO) or with biotinylated rabbit anti- ␥ anti-IFN- (4S.B3) (BD PharMingen, Heidelberg, Germany). CFSE was mouse Igs (for all others; DAKO). Bound Abs were detected using a obtained from Molecular Probes (Leiden, The Netherlands). Cyclosporine streptavidin-biotinylated alkaline phosphatase complex (DAKO) and Fast was purchased from Sigma-Aldrich. Human rIL-4 and the neutralizing Red as a chromogen (Sigma-Aldrich). Stained sections were counterstained mAb to IL-4 were from Perbio Science (Bonn, Germany). Human rIL-10 with hematoxylin and examined by light microscopy. and the neutralizing mAb to IL-10 were from R&D Systems (Wiesbaden, Germany). The neutralizing mAbs were highly speciﬁc for human cytokines with no measurable cross-reactivity to murine cytokines. Isotype Abs Statistical and mathematical analysis

(mouse IgG2b and rat IgG1) were purchased from BD PharMingen. by guest on September 28, 2021 Linear regression was calculated using the InStat computer program (GraphPad, San Diego, CA). Based on CFSE fluorescence intensity, the Mice frequencies (F) of CD4 and CD8 cells that had undergone divisions (from ϭ Mice congenic for the scid mutation on the NOD genetic background were d 1 for the first cycle to n cell divisions) was calculated as follows: purchased from M&B (Ry, Denmark). The animals were maintained under pathogen-free conditions in the animal facility of Nikolaus Fiebiger Center n (Erlangen, Germany). Mice were used at 6Ð12 wk of age. ϭ ͸͑ Ϫd͒ F 2 Fd dϭ1 Human cell preparation PBMC were obtained by Ficoll Hypaque (Sigma-Aldrich) gradient cen- Results trifugation of heparinized venous blood from young healthy volunteers not Development of a human Th1 immune reaction in SCID mice taking any medications. For further T cell or monocyte preparation, PBMC were incubated with sheep erythrocytes, and T cells were isolated from the Previous studies have shown that human PBMC remain function- rosette-positive cells by negative-selection panning using anti-CD16 and ally active after injection into SCID mice (27). Moreover, transfer anti-CD19 as previously described (20). Monocytes were purified from the of human cells into SCID mice results in the development of a fraction of rosette-negative cells using the monocyte isolation kit from xenogeneic graft-vs-host (GvH) disease (28). For characterization Miltenyi Biotec (Bergisch Gladbach, Germany), according to the manu- facturer’s instructions. The frequencies of cell populations within PBMC, of the developing immune reaction of human cells against mouse and homogeneity and purity of the isolated T cells and monocytes were tissue, human PBMC were labeled with CFSE and injected into routinely assessed by flow cytometry. Typically, Ն95% of the T cells were SCID mice (Fig. 1A). Human CD4 and CD8 T cells spontaneously Ն positive for CD3 and CD4, 90% of the monocytes stained brightly with proliferated after introduction into SCID mice. The degree of CD8 a mAb to CD14, and Ͼ98% of the cells were viable after the purification procedure. T cells were negative for the activation markers CD25, CD30, T cell expansion exceeded that of CD4 cells. Because CFSE pat- CD69, and HLA-DR. tern analysis indicated a similar proliferative rate of CD4 and CD8 cells (data not shown), this was likely to be a consequence of Injection with human cells and treatment of mice increased precursor frequencies of CD8 cells capable of undergoing proliferation in response to murine Ags (Table I). Of note, the Isolated cell populations were resuspended in PBS and injected i.p. into mice in a total volume of 0.2 ml. For CFSE labeling, 10 ϫ 106 PBMC were percentage of T cells initially reacting to mouse tissue was differ- resuspended in 1 ml of PBS and labeled for 8 min with 10 ␮M CFSE at ent among donors. Additional experiments using nonlabeled room temperature. Cyclosporine treatment was performed daily with 0.02Ð PBMC indicated that greater expansion of CD8 compared with 0.04 mg from days 0 through 14 by i.p. injection. Treatment with IL-4 CD4 cells led to an inverted CD4/CD8 ratio (Fig. 1B). (0.008Ð0.2 mg) or IL-10 (0.0016Ð0.04 mg) was performed daily by i.p. injection of recombinant cytokines from days 14 to 19. Treatment with Analysis of activation markers on the T cell surface revealed blocking Abs for human IL-4 or IL-10 (100 ␮g) was performed by three that the frequency of CD4 T cells expressing HLA-DR and CD25 i.p. injections during the 14-day experiments, at days 0, 5, and 10. increased with time (Fig. 1C). The percentage of T cells capable of The Journal of Immunology 6429 Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 1. Human PBMC develop a Th1-biased immune reaction after i.p. injection into SCID mice. A total of 50 ϫ 106 human PBMC was injected i.p. into SCID mice. At the indicated time points, human cells were recovered from the peritoneal cavity, and analyzed by flow cytometry. A, Before injection, PBMC were labeled with CFSE. Frequencies of cells with reduced CFSE fluorescence indicative of proliferative cycles were assessed after counterstaining with mAbs to CD4 or CD8. B, The CD4/CD8 ratio in recovered cell populations was calculated after surface staining with mAbs to CD4 and CD8. C, Percentage of CD4 cells expressing the activation marker HLA-DR or CD25. D, The frequency of cells capable of IFN-␥ production was assessed by intracellular flow cytometry after a 5-h in vitro stimulation with PMA and ionomycin. One representative of eight independent experiments with different donors is shown. E, Liver sections from day 14 are illustrated after staining with H&E (i–iii) or with a mAb to human CD3 (iv). Polymorphic lymphoid cell infiltrates are observed in a large (i) and in small portal tracts (ii and iii) around small bile ducts. Note that bile duct epithelial cells occasionally show degenerative changes (ii). Immunostaining of a serial section to iii identifies the majority of infiltrating cells as human CD3-positive T cells (iv). Original magnification: i, iii, and iv, ϫ200; ii, ϫ400. F, A typical epitheloid cell granuloma from perigastric fatty tissue is illustrated (i, H&E staining). Immunostaining reveals a row of human CD68-positive macrophages surrounding an area of central necrosis (ii). The polymorphic lymphoid cell infiltrate adjacent to the macrophages consists of human CD3-positive T cells (iii) with a large proportion of CD4-positive cells (iv) and a smaller number of CD8-positive cells (v). Adjacent to the granuloma are aggregates of human CD20-positive B cells (vi). Original magnification: i, ii, iii, and vi, ϫ200; iv and v, ϫ400.

producing IFN-␥ and IL-4 as indicators of a Th1 or a Th2 re- 31). By contrast, eosinophils, indicative of Th2 activation, could sponse, respectively, was determined by intracellular flow cytom- not be detected in the granulomas (data not shown). etry. As demonstrated in Fig. 1D, the frequency of IFN-␥-producing T cells increased with time of the xenogeneic immune reaction. In contrast, IL-4-producing T cells could not be detected by intra- Cylosporine prevents the human Th1 immune reaction in SCID cytoplasmic staining after ex vivo stimulation in the recovered T mice cells (data not shown). Importantly, histopathological analysis re- To define the role of T cells in the initiation of the xenogeneic GvH vealed infiltration of activated human lymphocytes into the portal reaction in the SCID mouse more precisely, experiments were con- tracts of the liver, and into the perigastrointestinal and perirenal ducted in which PBMC-injected mice were treated with cyclospor- fatty tissues (Fig. 1, E and F). The lymphocytic infiltrates were ine, an inhibitor of T cell activation (Fig. 2). Cyclosporine treat- frequently organized in granuloma-like structures with pallisading ment had a marked inhibitory effect on the activation of CD4 T human macrophages and central necrosis (Fig. 1F), further indi- cells (Fig. 2A). The effect of cyclosporine was even more pro- cating the Th1-biased nature of the inflammatory immunity (29Ð nounced with regard to the expansion of IFN-␥ producers, which 6430 REGULATION OF HUMAN Th1 IMMUNITY IN VIVO

Table I. Frequencies of T cells initiating spontaneous proliferation cytes to T cells regulates the strength of the Th1-mediated xeno- after introduction into SCID mice geneic reaction.

CD4 Cells (%) CD8 Cells (%) IL-4 and IL-10 are regulators of human Th1 immunity

Donor 1 0.46 1.07 A large body of in vitro data suggests that the development of a Donor 2 1.09 2.43 human Th1-mediated immune reaction is controlled by cytokines, such as IL-4 and IL-10 (26). However, conclusive evidence from in vivo human immune responses has not been provided to date. To address this question, human endogenously produced IL-4 and was completely prevented by cyclosporine (Fig. 2B). As a reflec- IL-10 were neutralized during the development of the xenogeneic tion of the activity of effector T cells in vivo, the inflammatory Th1-biased reaction resulting from the injection of PBMC into human cytokines IFN-␥ and TNF and human IL-4 were measured SCID mice by mAb that were specific for the human cytokines in the serum of the animals. The serum level of human IFN-␥ was with no cross-reactivity to murine IL-4 and IL-10 (Fig. 4). The markedly diminished in the cyclosporine-treated mice compared blockade of endogenous human IL-4 led to a significant enhance- with PBS-treated mice, whereas the serum level of human TNF ment of the Th1-biased immune response (Fig. 4, AÐC). Whereas was only slightly decreased (Fig. 2C). Human IL-4 was not de- the frequency of activated CD4 T cells within the cells recovered tectable in any of the mice (data not shown). Together, the data from anti-IL-4-treated mice was comparable with that of control strongly indicate that the immune reaction of human PBMC to mice (Fig. 4A), neutralization of endogenous IL-4 resulted in a mouse tissue generates a T cell-mediated, Th1-biased immune significantly increased frequency of T cells capable of IFN-␥ pro- Downloaded from response. duction (1.2 Ϯ 0.2-fold increase, p Ͻ 0.004; B). When the in vivo activity of effector T cells was assessed, the effect of IL-4 neutral- The xenogeneic Th1-biased immune response requires APC ization became even more pronounced, because neutralization of The development of a specific immune response requires APC IL-4 resulted in a Ͼ2-fold increase in the serum concentrations of (32). Monocytes represent potential APC in the pool of PBMC. To human IFN-␥ (3.0 Ϯ 2.4-fold increase, p Ͻ 0.05) and human TNF

evaluate the impact of monocytes on the development of the hu- (2.2 Ϯ 1.5-fold increase, p ϭ 0.06) in the serum of the mice (Fig. http://www.jimmunol.org/ man Th1-biased xenogeneic GvH reaction, T cells and monocytes 4C). Because IFN-␥ production is not affected directly by IL-4 were purified from the peripheral blood of the same donor and (33), this is likely to be a consequence of markedly increased num- injected into SCID mice separately or at a purposeful ratio of five bers of activated cytokine-producing T cells rather than of en- T cells to one monocyte (Fig. 3, A–C). Activation of CD4 T cells, hanced secretion of human IFN-␥ and TNF by the activated T their differentiation into effectors capable of IFN-␥ production, and cells. in vivo secretion of IFN-␥ and TNF were all observed only in the Neutralization of endogenously produced IL-10 during the xe- mice that had received T cells together with monocytes. nogeneic human Th1-biased immune response did not interfere To characterize the role of monocytes in the development of the with the activation of CD4 T cells, because the frequencies of human xenogeneic Th1-biased reaction in greater detail, we con- HLA-DR-positive CD4 T cells recovered from control mice and by guest on September 28, 2021 ducted experiments with different ratios of purified T cells to from anti-IL-10-treated mice were comparable (Fig. 4D). Inhibi- monocytes, keeping the absolute number of injected T cells con- tion of IL-10, however, resulted in a significant increase of the stant (Fig. 3D). Levels of human cytokines indicative of an in vivo frequencies of effector T cells capable of IFN-␥ production (1.2 Ϯ Th1-biased effector cell activation (IFN-␥ and TNF) were mea- 0.3-fold increase, p Ͻ 0.02; Fig. 4E). Consequently, blockade of sured in the serum of the animals. A marked increase in the con- IL-10 resulted in a tendency for increased Th1-mediated effector centration of IFN-␥ and TNF was detected in the serum of the functions in vivo as documented by increased serum levels of animals that had been injected with mixtures that contained higher IFN-␥ (1.9 Ϯ 1.0-fold increase, p ϭ 0.08) and TNF (1.8 Ϯ 0.8-fold ratios of monocytes to T cells. Notably, when the in vivo produc- increase, p Ͻ 0.05; Fig. 4F). Thus, the development of a Th1- tion of IFN-␥ from PBMC of various donors was analyzed as a biased human immune reaction is tightly controlled by the endo- function of the monocyte-to-T cell ratio within the injected PBMC, genously produced anti-inflammatory cytokines IL-4 and IL-10. It the IFN-␥ concentration in the serum of the animals correlated should be noted that no IL-10 mRNA could be detected in recov- directly with this ratio (Fig. 3E). Thus, the ratio of human mono- ered cells even when using highly sensitive real-time PCR, and

FIGURE 2. Cyclosporine prevents the Th1-biased human immune response in SCID mice. A total of 50 ϫ 106 freshly isolated human PBMC was injected into SCID mice. Mice were treated daily by i.p. injection with cyclosporine or PBS as control. Analysis was performed on day 14. Human cells were recovered from the peritoneal cavity and analyzed by ﬂow cytometry for extracellular expression of CD4 and HLA-DR (A), and for cytoplasmic IFN-␥ after in vitro stimulation with PMA and ionomycin (B). For comparison, values from freshly isolated PBMC are shown (Before). C, Serum levels of the human inﬂammatory cytokines IFN-␥ and TNF were determined by ELISA. One representative of six independent experiments with different donors is shown. The Journal of Immunology 6431 Downloaded from

FIGURE 3. Monocytes are required for the development of the Th1-biased immune reaction of human cells in SCID mice. Human T cells and monocytes were purified by negative selection from PBMC. A–C, T cells (50 ϫ 106) and monocytes (10 ϫ 106) from the same donor were injected either separately or together into mice. As a control, mice were injected with PBMC that contained 50 ϫ 106 T cells. Analysis was performed on day 14. Human cells were recovered from the peritoneal cavity and analyzed by flow cytometry for CD4 and HLA-DR (A), and cytoplasmic IFN-␥ (B). C, Serum levels of the human inflammatory cytokines IFN-␥ and TNF were determined by ELISA. One representative of three independent experiments with different donors is shown. ϫ 6 ϫ 6 ϫ 6 ϫ 6 n.d., Nondetectable. D, A total of 50 10 T cells was injected together with different numbers of monocytes (6.25 10 , 12.5 10 ,or25 10 , http://www.jimmunol.org/ respectively) from the same donor into mice. Serum levels of the human inflammatory cytokines IFN-␥ and TNF were determined at day 14 by ELISA. E, Serum concentrations of human IFN-␥ (day 14) were plotted as a function of the monocyte-to-T cell ratio within the injected PBMC from 19 independent experiments, and the linear regression was calculated. The serum IFN-␥ concentration significantly correlated with the monocyte-to-T cell ratio of the inoculum.

IL-4 mRNA was found only at a very low level (data not shown), we sought to delineate the effect of IL-4 as a means to down- suggesting that the mechanisms of action of IL-4 and IL-10 on the modulate an established Th1-mediated immune response. SCID development of a human Th1-biased immune response, although mice were injected with PBMC, and starting at day 14, they were by guest on September 28, 2021 different, are very precise and efficient processes. treated daily for 5 days with recombinant human IL-4 (Fig. 5). A 5-day treatment with IL-4 did not alter the frequency of HLA-DR- IL-4 down-modulates established Th1-biased immunity expressing T cells (Fig. 5A). However, the Th1-biased differenti- Neutralization of endogenous IL-4 led to an unbalanced and ex- ation of T cells into IFN-␥-producing effectors was significantly aggerated development of the xenogeneic human Th1-biased im- inhibited in response to IL-4 treatment (reduction to 92 Ϯ 7% of mune response (Fig. 4, A–C), suggesting the potential of this cy- control, p Ͻ 0.03; Fig. 5B). Moreover, IL-4 caused a significant tokine to function as an immune modulator in humans. Therefore, suppression of the Th1-biased effector functions in vivo, as

FIGURE 4. Endogenously produced IL-4 and IL-10 control the development of the human Th1-biased immune reaction. A total of 50 ϫ 106 freshly isolated human PBMC was injected into SCID mice. At days 0, 5, and 10, mice were treated with mAbs to human IL-4 (␣IL-4) (A–C) or human IL-10 (␣IL-10) (D–F). As a control, animals were treated with isotype-matched control mAbs (IgG). Analysis was performed on day 14. Human cells were recovered from the peritoneal cavity and analyzed by ﬂow cytometry for CD4 and HLA-DR (A and D), and cytoplasmic IFN-␥ (B and E). C and F, Serum levels of the human inﬂammatory cytokines IFN-␥ and TNF were determined by ELISA. One representative of 13 independent experiments for ␣IL-4 and one of 13 independent experiments for ␣IL-10 with different donors are shown. 6432 REGULATION OF HUMAN Th1 IMMUNITY IN VIVO

FIGURE 5. Exogenous IL-4 diminishes the human Th1-biased immune reaction. A total of 50 ϫ 106 freshly isolated human PBMC was injected into SCID mice. Starting at day 14, mice were treated daily with human IL-4 for 5 days. As a control, animals were treated with PBS. Analysis was performed on day 19. Human cells were recovered from the peritoneal cavity and analyzed by ﬂow cytometry for CD4 and HLA-DR (A), and cytoplasmic IFN-␥ (B) after in vitro stimulation with PMA and ionomycin. C, Serum levels of the human inﬂammatory cytokines IFN-␥ and TNF were determined by ELISA. One representative of six independent experiments with PBMC from different donors is shown. indicated by substantially decreased concentrations of human clinical applications. In this study, we introduce a convenient and IFN-␥ and TNF in the serum of the treated animals (reduction to reliable model to analyze human immune responses in vivo. The Downloaded from 50 Ϯ 43 and 63 Ϯ 32% of control, respectively, p Ͻ 0.04; Fig. system provides the opportunity to investigate in detail the mech- 5C). Together, the data indicate that administration of exogenous anisms involved in regulating human immunity and, in particular, IL-4 is able to down-modulate an established Th1-biased immune the analysis of the development of human Th1 responses at the response. molecular level. Thus, the model represents a novel and useful tool for the study of human immunity.

IL-10 targets Th1 effector functions Human tissue engrafted into SCID mice has served as a model http://www.jimmunol.org/ IL-10 has been shown to be a very powerful inhibitor of IFN-␥- for in vivo human research in HIV, transplantation, and tumor mediated effector functions in the mouse by preventing the pro- biology (27, 28, 35, 36). The goal of most such investigations was duction of IFN-␥ (34). The data from the anti-IL-10 experiments to establish functional human tissue in nonhuman living organ- (Fig. 4, DÐF) suggest a similar mechanism of IL-10 activity in isms. However, little attention was paid to the xenogeneic reaction humans. To test this hypothesis, we investigated the effect of ex- that occurs after engraftment of immunocompetent human cells ogenous human IL-10 on an established Th1-mediated immune into mice, although it has been shown that immunocompromised reaction of human PBMC (Fig. 6). IL-10 treatment reduced the recipients of xenogeneic cells develop severe T cell-dependent frequency of HLA-DR-positive CD4 T cells but, in contrast to GvH disease (28, 37, 38). To induce human immune reactions in by guest on September 28, 2021 IL-4, had no effect on the differentiation of cells capable of pro- vivo, therefore, we transferred human PBMC into SCID mice and ducing IFN-␥ (Fig. 6, A and B). However, IL-10 treatment signif- could, for the first time, conclusively show that a human APC- icantly inhibited the in vivo production of IFN-␥ and TNF as as- dependent Th1-biased cellular immune response develops. The sessed by measurement of serum levels of these cytokines magnitude of this Th1-biased immune reaction is comparable to (reduction to 51 Ϯ 19 and 60 Ϯ 32%, respectively; p Ͻ 0.002 and that of physiological immune reactions in the human body (39, p Ͻ 0.03; Fig. 6C). Thus, IL-10 might inhibit effector functions of 40). For example, the serum IFN-␥ level in RA and osteoarthritis the human Th1-mediated immune response by preventing the pro- patients varies in the range of 7Ð16 pg/ml (40). Even the strongest duction of these inflammatory cytokines. inducer of IFN-␥ production, IL-12, induces IFN-␥ serum levels in a comparable range (40Ð140 pg/ml) (39). Moreover, the develop- Discussion ment of the human immune response was characterized by lym- Modern immunological research is dependent on the use of animal phocytic infiltrations into different murine organs. The frequent models to dissect the complex, but tightly regulated cellular and granuloma-like architecture of the infiltrates with a central necrotic molecular interactions that occur during an immune response. area surrounded by dense accumulation of T cells and macro- However, animal models cannot provide all of the answers that are phages is reminiscent of granulomatous lesions in tuberculosis, required to understand human immunity completely, or to translate sarcoidosis, and Wegener’s granulomatosis, all of which are char- basic research to useful clinical applications efficiently. Therefore, acterized by Th1-mediated granuloma formation (29Ð31, 41, 42). it is essential to develop reliable models of human immunity to Systemic Th1 inflammation evolved reliably and consistently foster the translation of fundamental principles into appropriate after i.p. injection of PBMC from healthy individuals into SCID

FIGURE 6. Exogenous IL-10 diminishes the human Th1-biased immune reaction. A total of 50 ϫ 106 freshly isolated human PBMC was injected into SCID mice. Starting at day 14, mice were treated daily with human IL-10 for 5 days. As a control, animals were treated with PBS. Analysis was performed on day 19. Human cells were recovered from the peritoneal cavity and analyzed by flow cytometry for CD4 and HLA-DR (A), and cytoplasmic IFN-␥ (B). C, Serum levels of the human inflammatory cytokines IFN-␥ and TNF were determined by ELISA. One representative of six independent experiments with PBMC from different donors is shown. The Journal of Immunology 6433 mice. However, the degree of Th1 inflammation varied, largely APC function (34). In accordance with these findings, the IL-10- depending on the ratio of injected APC to T cells (Fig. 3), which mediated immune suppression observed in the current study is emphasizes the specific nature of the immune response. Of note, likely the result of the inhibition of unique functions of APC that PBMC from a small minority of donors repeatedly failed to de- are required for IFN-␥ secretion in vivo. However, T cell activa- velop systemic inflammation in the SCID mice, despite normal tion and differentiation into Th1 cells appear to be intact. In con- frequencies and numbers of monocytes and T cells. Whether the trast, the immunosuppressive effect of IL-4 imparts suppression of lack of response of the PBMC from this subset of donors reflects T cell activation and differentiation. specific anergy to murine Ags or the presence of regulatory cells of Together, the system described in this study presents a useful some kind remains to be delineated. With the exception noted, the tool to investigate in detail the mechanisms involved in regulating model is robust in that injection of PBMC from the vast majority human immunity. The data demonstrate that both endogenous IL-4 of donors results in spontaneous systemic Th1 inflammation. and IL-10 control human Th1-mediated immune reactions. As a A striking observation in this study was the finding that the in logical consequence, rIL-4 and rIL-10 administered during an es- vivo development of human Th1 immunity is tightly controlled by tablished immune response repress Th1-mediated effector func- endogenously produced IL-4 and IL-10 to a similar extent. Our tions by inhibiting proinflammatory cytokine production, although data suggest that the role of IL-4 in the regulation of Th1 immunity by different ways. However, whereas IL-10-based treatment, al- in humans is much more pronounced than in the mouse. In the though effective, might carry the risks associated with general im- mouse, IL-10 neutralization results in an accelerated onset of col- munosuppression, such as enhanced susceptibility to infectious lagen-induced arthritis and increased severity, whereas IL-4 block- agents (34), the immunosuppressive action of IL-4 appears to be ade is without effect (43). Moreover, IL-10-deficient mice are more more selective, because IL-4 targets primarily T cells undergoing Downloaded from susceptible and develop more severe EAE when compared with activation through their Ag receptor (33, 51). IL-4-deficient mice (10). In contrast, our findings using neutralizing Abs to IL-4 and IL-10 suggest that, in humans, the regulatory Acknowledgments role of these two cytokines for Th1-mediated immunity is We thank Daniela Thein for expert technical assistance, Jan Leipe and weighted differently. Stefan Mattyasovszky for their skillful preparations of murine tissue, Ruediger Mueller for purification of monocytes, and Andreas Pahl for the Supportive indications for this conclusion derive from the anal- http://www.jimmunol.org/ real-time PCR. Thomas Winkler’s advice on handling of the animals is ysis of patients with systemic autoimmune diseases, which are kindly acknowledged. characterized by unresolved inflammation mediated by Th1 autoimmune activity. A typical feature of human autoimmune diseases References appears to be the absence of the Th2 cytokine IL-4 together with 1. Ridgway, W. M., H. L. Weiner, and C. G. Fathman. 1994. Regulation of auto- simultaneous up-regulation of IL-10 (21, 44Ð48), which, in con- immune response. Curr. Opin. Immunol. 6:946. 2. O’Garra, A., L. Steinman, and K. Gijbels. 1997. CD4ϩ T-cell subsets in auto- trast to the situation in the mouse, is not a typical Th2 cytokine in immunity. Curr. Opin. Immunol. 9:872. humans (49). When IL-4 is measured directly in the synovial fluid 3. O’Shea, J. J., A. Ma, and P. Lipsky. 2002. Cytokines and autoimmunity. Nat. Rev. or in PBMC of RA patients, it is virtually absent (21, 44). More- Immunol. 2:37.

4. Fox, C. J., and J. S. Danska. 1997. IL-4 expression at the onset of islet inflam- by guest on September 28, 2021 over, we have demonstrated that memory CD4 T cells from pa- mation predicts nondestructive insulitis in nonobese diabetic mice. J. Immunol. tients with RA show an impaired differentiation into IL-4-produc- 158:2414. ing cells (45). In marked contrast, high levels of functionally active 5. Tian, J., M. A. Atkinson, M. Clare-Salzler, A. Herschenfeld, T. Forsthuber, P. V. Lehmann, and D. L. Kaufman. 1996. Nasal administration of glutamate IL-10 are detected in RA synovial fluid and in the peripheral cir- decarboxylase (GAD65) peptides induces Th2 responses and prevents murine culation of RA patients (46, 47). In vitro analysis of mononuclear insulin-dependent diabetes. J. Exp. Med. 183:1561. 6. Mueller, R., T. Krahl, and N. Sarvetnick. 1996. Pancreatic expression of inter- cells from multiple sclerosis patients shows elevated numbers of leukin-4 abrogates insulitis and autoimmune diabetes in nonobese diabetic cells producing IL-10 (48). Thus, although IL-10 is produced in (NOD) mice. J. Exp. Med. 184:1093. substantial amounts in human autoimmune diseases, it appears to 7. Morita, Y., J. Yang, R. Gupta, K. Shimizu, E. A. Shelden, J. Endres, J. J. Mule, K. T. McDonagh, and D. A. Fox. 2001. Dendritic cells genetically engineered to be insufficient to control an autoimmune reaction in the absence express IL-4 inhibit murine collagen-induced arthritis. J. Clin. Invest. 107:1275. of IL-4. 8. Shaw, M. K., J. B. Lorens, A. Dhawan, R. DalCanto, H. Y. Tse, A. B. Tran, Analysis of IL-4- and IL-10-induced down-modulation of the C. Bonpane, S. L. Eswaran, S. Brocke, N. Sarvetnick, et al. 1997. Local delivery of interleukin-4 by retrovirus-transduced T lymphocytes ameliorates experimen- established human Th1-biased immune response described in this tal autoimmune encephalomyelitis. J. Exp. Med. 185:1711. study suggests that the mechanism of action of these cytokines is 9. Finnegan, A., K. Mikecz, P. Tao, and T. T. Glant. 1999. Proteoglycan (aggrecan)- induced arthritis in BALB/c mice is a Th1-type disease regulated by Th2 cyto- different. Whereas the IL-4 effect was evident at the steps of T cell kines. J. Immunol. 163:5383. differentiation into IFN-␥-producing effectors, the immunomodu- 10. Bettelli, E., M. P. Das, E. D. Howard, H. L. Weiner, R. A. Sobel, and latory effect of IL-10 was detectable only at the level of Th1- V. K. Kuchroo. 1998. IL-10 is critical in the regulation of autoimmune encephalomyelitis as demonstrated by studies of IL-10- and IL-4-deficient and trans- mediated effector functions, e.g., in vivo production of inflamma- genic mice. J. Immunol. 161:3299. tory cytokines. Interestingly, the central necrotic areas of the 11. Moritani, M., K. Yoshimoto, S. Ii, M. Kondo, H. Iwahana, T. Yamaoka, T. Sano, granulomas were reduced by IL-10, but not by IL-4 treatment (data N. Nakano, H. Kikutani, and M. Itakura. 1996. Prevention of adoptively transferred diabetes in nonobese diabetic mice with IL-10-transduced islet-specific not shown), further suggesting a direct anti-inflammatory role of Th1 lymphocytes: a gene therapy model for autoimmune diabetes. J. Clin. Invest. IL-10. IL-4 and IL-10 exert a broad spectrum of different rather 98:1851. 12. Powrie, F., J. Carlino, M. W. Leach, S. Mauze, and R. L. Coffman. 1996. A than overlapping immunomodulatory activities (33, 34). Whereas critical role for transforming growth factor-␤ but not interleukin-4 in the sup- IL-4 predominantly targets T cells as effectors of an immune re- pression of T helper type 1-mediated colitis by CD45RBlowCD4ϩ T cells. J. Exp. action, the action of IL-10, in contrast, rather focuses on APC, Med. 183:2669. 13. Asseman, C., S. Mauze, M. W. Leach, R. L. Coffman, and F. Powrie. 1999. An such as macrophages. The most pronounced regulatory capacity of essential role for interleukin-10 in the function of regulatory T cells that inhibit IL-4 is the direct negative regulation of Th1 differentiation from intestinal inflammation. J. Exp. Med. 190:995. CD4 T cells (50). In contrast, IL-10 modulates expression of cy- 14. Cassatella, M. A., L. Meda, S. Gasperini, F. Calzetti, and S. Bonora. 1994. In- terleukin-10 (IL-10) upregulates IL-1 receptor antagonist production from lipopo- tokines, such as IL-12, and surface molecules by myeloid cells, lysaccharide-stimulated human polymorphonuclear leukocytes by delaying with important consequences for their ability to activate and sus- mRNA degradation. J. Exp. Med. 179:1695. 15. Wong, H. L., G. L. Costa, M. T. Lotze, and S. M. Wahl. 1993. Interleukin (IL)-4 tain immune responses. Most effects of IL-10 on T cells are, there- differentially regulates monocyte IL-1 family gene expression and synthesis in fore, indirect and are mediated via its down-modulatory activity on vitro and in vivo. J. Exp. Med. 177:775. 6434 REGULATION OF HUMAN Th1 IMMUNITY IN VIVO

16. de Waal Malefyt, R., J. Abrams, B. Bennett, C. G. Figdor, and J. E. de Vries. 34. Moore, K. W., R. de Waal Malefyt, R. L. Coffman, and A. O’Garra. 2001. In- 1991. Interleukin-10 (IL-10) inhibits cytokine synthesis by human monocytes: an terleukin-10 and the interleukin-10 receptor. Annu. Rev. Immunol. 19:683. autoregulatory role of IL-10 produced by monocytes. J. Exp. Med. 174:1209. 35. Feuerer, M., P. Beckhove, L. Bai, E. F. Solomayer, G. Bastert, I. J. Diel, 17. Essner, R., K. Rhoades, W. H. McBride, D. L. Morton, and J. S. Economou. C. Pedain, M. Oberniedermayr, V. Schirrmacher, and V. Umansky. 2001. Ther- 1989. IL-4 down-regulates IL-1 and TNF gene expression in human monocytes. apy of human tumors in NOD/SCID mice with patient-derived reactivated mem- J. Immunol. 142:3857. ory T cells from bone marrow. Nat. Med. 7:452. 18. Maggi, E., P. Parronchi, R. Manetti, C. Simonelli, M. P. Piccinni, F. S. Rugiu, 36. Tary-Lehmann, M., P. V. Lehmann, D. Schols, M. G. Roncarolo, and A. Saxon. M. De Carli, M. Ricci, and S. Romagnani. 1992. Reciprocal regulatory effects of 1994. Anti-SCID mouse reactivity shapes the human CD4ϩ T cell repertoire in IFN-␥ and IL-4 on the in vitro development of human Th1 and Th2 clones. hu-PBL-SCID chimeras. J. Exp. Med. 180:1817. J. Immunol. 148:2142. 37. Surh, C. D., H. Kosaka, and J. Sprent. 1997. Rat stem cells developing in irra- 19. Groux, H., M. Bigler, J. E. de Vries, and M. G. Roncarolo. 1996. Interleukin-10 ϩ diated SCID mice fail to become tolerized and cause lethal graft-versus-host induces a long-term antigen-speciﬁc anergic state in human CD4 T cells. J. Exp. disease. Am. J. Pathol. 151:591. Med. 184:19. 38. Miki, T., Y. H. Lee, A. Tandin, V. Subbotin, A. Goller, A. Kovscek, J. J. Fung, 20. Schulze-Koops, H., L. S. Davis, T. P. Haverty, M. C. Wacholtz, and P. E. Lipsky. and L. A. Valdivia. 2001. Hamster-to-rat bone marrow xenotransplantation and 1998. Reduction of Th1 cell activity in the peripheral circulation of patients with humoral graft vs. host disease. Xenotransplantation 8:213. rheumatoid arthritis after treatment with a non-depleting humanized monoclonal antibody to CD4. J. Rheumatol. 25:2065. 39. Portielje, J. E., C. H. Lamers, W. H. Kruit, A. Sparreboom, R. L. Bolhuis, 21. Schulze-Koops, H., P. E. Lipsky, A. F. Kavanaugh, and L. S. Davis. 1995. El- G. Stoter, C. Huber, and J. W. Gratama. 2003. Repeated administrations of interleukin (IL)-12 are associated with persistently elevated plasma levels of IL-10 evated Th1- or Th0-like cytokine mRNA in peripheral circulation of patients with ␥ ␣ rheumatoid arthritis: modulation by treatment with anti-ICAM-1 correlates with and declining IFN- , tumor necrosis factor- , IL-6, and IL-8 responses. Clin. clinical beneﬁt. J. Immunol. 155:5029. Cancer Res. 9:76. 22. Schwitalle, Y., R. Laub, K. Wenk, R. Voll, and F. Emmrich. 1996. The anti-CD4 40. Klimiuk, P. A., S. Sierakowski, R. Latosiewicz, B. Cylwik, J. Skowronski, and monoclonal antibody Max. 16H5 induces an increase of intracellular IL-4 in J. Chwiecko. 2001. Serum cytokines in different histological variants of rheuma- human T cell clones. Immunobiology 196:21. toid arthritis. J. Rheumatol. 28:1211. 23. Ghoreschi, K., P. Thomas, S. Breit, M. Dugas, R. Mailhammer, W. Van Eden, 41. Orme, I. M. 1998. The immunopathogenesis of tuberculosis: a new working R. Van Der Zee, T. Biedermann, J. Prinz, M. Mack, et al. 2003. Interleukin-4 hypothesis. Trends Microbiol. 6:94. Downloaded from therapy of psoriasis induces Th2 responses and improves human autoimmune 42. Nanki, T., and P. E. Lipsky. 2000. Cytokine, activation marker, and chemokine ϩ disease. Nat. Med. 9:40. receptor expression by individual CD4 memory T cells in rheumatoid arthritis 24. Asadullah, K., W. Sterry, K. Stephanek, D. Jasulaitis, M. Leupold, H. Audring, synovium. Arthritis Res. 2:415. H. D. Volk, and W. D. Docke. 1998. IL-10 is a key cytokine in psoriasis: proof 43. Joosten, L. A., E. Lubberts, P. Durez, M. M. Helsen, M. J. Jacobs, M. Goldman, of principle by IL-10 therapy: a new therapeutic approach. J. Clin. Invest. and W. B. van den Berg. 1997. Role of interleukin-4 and interleukin-10 in murine 101:783. collagen-induced arthritis: protective effect of interleukin-4 and interleukin-10 25. McInnes, I. B., G. G. Illei, C. L. Danning, C. H. Yarboro, M. Crane, T. Kuroiwa, treatment on cartilage destruction. Arthritis Rheum. 40:249.

R. Schlimgen, E. Lee, B. Foster, D. Flemming, et al. 2001. IL-10 improves skin 44. Miossec, P., M. Naviliat, A. Dupuy d’Angeac, J. Sany, and J. Banchereau. 1990. http://www.jimmunol.org/ disease and modulates endothelial activation and leukocyte effector function in Low levels of interleukin-4 and high levels of transforming growth factor-␤ in patients with psoriatic arthritis. J. Immunol. 167:4075. rheumatoid synovitis. Arthritis Rheum. 33:1180. 26. van Roon, J. A., F. P. Lafeber, and J. W. Bijlsma. 2001. Synergistic activity of 45. Skapenko, A., J. Wendler, P. E. Lipsky, J. R. Kalden, and H. Schulze-Koops. interleukin-4 and interleukin-10 in suppression of inflammation and joint destruc- 1999. Altered memory T cell differentiation in patients with early rheumatoid tion in rheumatoid arthritis. Arthritis Rheum. 44:3. arthritis. J. Immunol. 163:491. 27. Mosier, D. E., R. J. Gulizia, S. M. Baird, and D. B. Wilson. 1988. Transfer of a 46. Katsikis, P. D., C. Q. Chu, F. M. Brennan, R. N. Maini, and M. Feldmann. 1994. functional human immune system to mice with severe combined immunodefi- Immunoregulatory role of interleukin-10 in rheumatoid arthritis. J. Exp. Med. ciency. Nature 335:256. 179:1517. 28. Duchosal, M. A., S. A. Eming, P. J. McConahey, and F. J. Dixon. 1992. Char- 47. Cush, J. J., J. B. Splawski, R. Thomas, J. E. McFarlin, H. Schulze-Koops, acterization of hu-PBL-SCID mice with high human immunoglobulin serum lev- L. S. Davis, K. Fujita, and P. E. Lipsky. 1995. Elevated interleukin-10 levels in els and graft-versus-host disease. Am. J. Pathol. 141:1097. patients with rheumatoid arthritis. Arthritis Rheum. 38:96. 29. Cooper, A. M., D. K. Dalton, T. A. Stewart, J. P. Griffin, D. G. Russell, and by guest on September 28, 2021 I. M. Orme. 1993. Disseminated tuberculosis in interferon-␥ gene-disrupted mice. 48. Navikas, V., J. Link, W. Palasik, M. Soderstrom, S. Fredrikson, T. Olsson, and J. Exp. Med. 178:2243. H. Link. 1995. Increased mRNA expression of IL-10 in mononuclear cells in 30. Csernok, E., A. Trabandt, A. Muller, G. C. Wang, F. Moosig, J. Paulsen, multiple sclerosis and optic neuritis. Scand. J. Immunol. 41:171. 49. Yssel, H., R. de Waal Malefyt, M. G. Roncarolo, J. S. Abrams, R. Lahesmaa, A. Schnabel, and W. L. Gross. 1999. Cytokine profiles in Wegener’s granulo- ϩ matosis: predominance of type 1 (Th1) in the granulomatous inflammation. Ar- H. Spits, and J. E. de Vries. 1992. IL-10 is produced by subsets of human CD4 thritis Rheum. 42:742. T cell clones and peripheral blood T cells. J. Immunol. 149:2378. 31. Agostini, C., U. Basso, and G. Semenzato. 1998. Cells and molecules involved in 50. Seder, R. A., W. E. Paul, M. M. Davis, and B. Fazekas de St. Groth. 1992. The the development of sarcoid granuloma. J. Clin. Immunol. 18:184. presence of interleukin-4 during in vitro priming determines the lymphokine- ϩ 32. Joyce, S. 2001. Immune recognition, response, and regulation: how T lympho- producing potential of CD4 T cells from T cell receptor transgenic mice. J. Exp. cytes do it. Immunol. Res. 23:215. Med. 176:1091. 33. Nelms, K., A. D. Keegan, J. Zamorano, J. J. Ryan, and W. E. Paul. 1999. The 51. Rocken, M., J. Urban, and E. M. Shevach. 1994. Antigen-specific activation, IL-4 receptor: signaling mechanisms and biologic functions. Annu. Rev. Immunol. tolerization, and reactivation of the interleukin-4 pathway in vivo. J. Exp. Med. 17:701. 179:1885.