IL-6-Deficient Mice Are Resistant to Experimental Autoimmune Encephalomyelitis: Roles of IL-6 in the Activation and Differentiation of Autoreactive T Cells¹

Although autoreactive T cells recognizing self myelin Ags are present in normal individuals, they normally remain at the precursor stage and do not induce autoimmune diseases. However, under certain conditions, usually a combination of genetic and environmental factors, myelin-specific T cells can become activated and induce demyelinating diseases. The precise mechanisms whereby myelin-specific precursor T cells become activated and differentiate into pathogenic effector cells in vivo are not clear.

Experimental autoimmune encephalomyelitis (EAE)³ is a T cell-mediated autoimmune disease that is often used as an animal model for human multiple sclerosis. EAE can be induced in susceptible strains of animals by coadministration of specific myelin Ags and adjuvants. The myelin Ags provide the necessary peptides required for generating the Ag-specific signal, whereas the adjuvant may be crucial for generating the costimulatory signal required for T cell activation. Although cell surface molecules such as B7 and ICAM-1 have been shown to be important in costimulation, cytokines, especially proinflammatory cytokines, may also play important roles (1, 2, 3). In addition, cytokines may also be important mediators of T cell effector function ranging from direct cytotoxicity to modulation of inflammatory responses (4, 5, 6).

IL-6 is a pleiotropic cytokine produced by a variety of cells including macrophages, fibroblasts, endothelial cells, B cells, and the Th2 subset of T cells (7, 8). In vitro studies have suggested that IL-6 may regulate bone metabolism, promote growth of hemopoietic stem cells, and modulate differentiation of activated B cells into plasma cells (7, 8, 9). Although initial studies also suggested that IL-6 may costimulate T cell activation (3, 10), promote inflammation, and up-regulate immune responses to pathogens (1, 2, 3), recent reports show that IL-6 can also play an anti-inflammatory or immunosuppressive role, and may negatively regulate the acute phase responses (7, 8, 11). Little is known of the roles of IL-6 in the central nervous system (CNS). Both IL-6 and IL-6R mRNAs are expressed in normal brain tissues, predominantly by neurons. It has been suggested that IL-6 may serve as a growth factor for neurons (12) and may regulate differentiation of oligodendrocytes (13, 14). Recently, IL-6 has also been implicated in the pathogenesis of Alzheimer’s disease (15, 16), and overexpression of IL-6 in rodents, targeted to astrocytes, led to neurodegeneration and cognitive defect (17, 18). The generation of IL-6 knockout mice in 1994 lent a unique opportunity to test directly the functions of IL-6 in vivo (19, 20). IL-6-deficient mice developed normally but failed to control microbial infections and were defective in the production of T cell-dependent Abs (21, 22, 23, 24, 25, 26). IL-6-deficient mice were also compromised in their acute-phase responses to tissue injury (19, 23).

The roles of IL-6 in autoimmune diseases are not clear. In EAE, IL-6 is produced by infiltrating myelin-specific CD4⁺ T cells, macrophages, and neuroglial cells and can potentially up-regulate class II MHC expression and recruit and activate inflammatory cells (27, 28). Gijbels et al. (29) reported that EAE can be dramatically suppressed by systemic administration of anti-IL-6 Ab. However, the levels of IL-6 in the blood and the spinal fluid of animals treated with anti-IL-6 Ab were dramatically increased as compared with control animals, making it impossible to conclude whether the effect of anti-IL-6 Ab was due to its neutralization of IL-6 or its enhancement of IL-6 production (the mechanism whereby anti-IL-6 Ab increases IL-6 production in vivo is not clear) (29). In another report, anti-IL-6 Ab was shown to neither enhance nor suppress actively induced EAE in mice, whereas transgenic expression of IL-6 in CNS suppressed the disease, suggesting that IL-6 may be indeed capable of suppressing EAE (30). This latter contention was further supported by the observation that recombinant IL-6, when administered systemically, ameliorated virus-induced demyelination (31). Thus, it appears that IL-6 can both up- and down-regulate EAE. To address this IL-6 paradox and to circumvent the potential problems associated with the use of anti-IL-6-neutralizing Abs, we studied EAE in IL-6-deficient B6.129 mice. As B6.129 mice are not susceptible to myelin basic protein (MBP)- or PLP-induced EAE, we immunized the mice with an immunodominant myelin oligodendrocyte glycoprotein (MOG) peptide, i.e., MOG38–50 (32, 33). Our results strongly suggest that IL-6 plays a crucial role in the activation and differentiation of MOG-specific autoreactive T cells in vivo.

Four- to 6-wk-old (B6 × 129)F₂ (B6.129) mice homozygous for IL-6 mutation and their littermate controls were purchased from The Jackson Laboratory (Bar Harbor, ME) and were housed in the University of Pennsylvania Animal Care Facilities. The IL-6 gene mutation was created by inserting the neo^r cassette into the second exon of the IL-6 gene (19). Mice were screened for IL-6 gene mutation by RT-PCR and Southern blot analysis (19).

All mice received a s.c. injection on flanks of 200 μg of MOG38–50 peptide in 0.1 ml of PBS emulsified in an equal volume of CFA containing 4 mg/ml of Mycobacterium tuberculosis H37RA (Difco, St. Louis, MO) and an i.v. or i.p. injection of 200 ng of pertussis toxin in 0.1 ml of PBS. A second injection of pertussis toxin (200 ng per mouse) was given 24 or 48 h later. Mice were examined every day for signs of EAE and scored as follows (34): 0, no disease; 1, tail paralysis; 2, hind limb weakness; 3, hind limb paralysis; 4, hind limb plus forelimb paralysis; 5, moribund or dead.

MOG38–50 peptide was synthesized using Fmoc solid phase methods and purified through HPLC by Research Genetics (Huntsville, AL). Pertussis toxin was purchased from List Biological Laboratories (Campbell, CA). The following reagents were purchased from PharMingen (San Diego, CA): purified rat anti-mouse IL-2 (clone JES-1A12), IL-4 (clone BVD4-1D11), IFN-γ (clone XMG1.2) mAb; recombinant mouse IL-2, IL-4, IFN-γ. Quantitative ELISA for IL-2, IL-4 and IFN-γ was performed using paired mAbs specific for corresponding cytokines per manufacturer’s recommendations (35).

Splenocytes (1.5 × 10⁶ cells/well) were cultured in 0.2 ml of serum-free medium X-Vivo 20 (BioWhittaker, Walkersville, MD), containing various concentrations of MOG38–50 peptide, OVA, or Con A (Sigma, St. Louis, MO) (34). Culture supernatants were collected 40 h later for cytokine assays. For the proliferation assay, 1 μCi of [³H]thymidine was added to each culture at 72 h, and cells were harvested 16 h later. Radioactivity was counted using a flatbed beta counter (Wallac, Gaithersburg, MD).

Brains and spinal cords were harvested at the end of each experiment, fixed in 10% formalin, and embedded in paraffin. Five-micrometer-thick paraffin sections were stained with hematoxylin and eosin or with Luxol Fast Blue as described (36).

Single-cell suspensions of splenocytes were first incubated for 45 min with one of the following Abs: anti-mouse CD3-FITC (clone 500-A2), anti-mouse CD4-PE (clone CT-CD4), anti-mouse CD8a-PE (clone CT-CD8a), or anti-mouse B220-FITC (clone RA3-6B2) (Caltag Laboratories, Burlingame, CA). Cells were then washed three times and analyzed directly using a FACScan flow cytometer (Becton Dickinson, Mountain View, CA).

Disease severity, day of onset, and cytokine concentrations were analyzed by analysis of variance (ANOVA).

To study the roles of IL-6 in the development of EAE, we immunized IL-6-deficient B6.129 mice (IL-6^−/−) and their control littermates (IL-6^+/+) with MOG38–50 peptide, and monitored the disease by both physical examination and histochemistry. Fig. 1 illustrates the typical disease course in B6.129 mice. EAE developed virtually in all control B6.129 mice, starting approximately 7 days after immunization. The early onset of EAE in these mice contrasts with disease in other strains of mice in which the symptoms usually develop much later (32, 33). The maximal clinical scores were 2.25 ± 0.25 and 3.0 ± 0.76 for male and female mice, respectively (Table I). Disease in female mice appeared to be more severe than in the male because 50% of female mice died by day 9 (as compared with 0% in the males) (Table I). Remarkably, neither male nor female mice that were deficient in IL-6 developed any symptoms of EAE during the entire period of observation, suggesting that IL-6 is crucial for the development of EAE.

Consistent with these results, histologic examination of the CNS tissue 10 and 28 days after immunization revealed dramatic differences in the two groups. In the control B6.129 group, multiple inflammatory foci were observed in the white matter of the brain and spinal cord, with the latter being more severely inflicted. Fig. 2,A is a representative section of the spinal cord of B6.129 mice 10 days after immunization. By contrast, in the CNS tissues of IL-6-deficient mice, no inflammatory lesion was ever detected (Fig. 2 C). Luxol Fast Blue staining of CNS tissues demonstrated severe demyelination in control B6.129 mice, but not in B6.129 mice deficient in IL-6. Taken together, these results strongly suggest that EAE may not be induced in the absence of IL-6.

Resistance to EAE in IL-6-deficient mice can be either due to the inability of myelin-specific T cells to differentiate into effector T cells in the periphery or due to the inability of differentiated effector T cells to induce demyelinating inflammation in the CNS, or both. To address this issue, we first examined whether activation and differentiation of myelin-specific T cells were normal in IL-6-deficient animals. Splenocytes were, therefore, collected from both control and IL-6-deficient mice 10 days after immunization and tested in vitro for their cytokine production and proliferation in response to MOG38–50 peptide. As shown in Fig. 3, splenocytes of control animals proliferated vigorously in response to MOG peptide and produced significant amount of both Th1 (IL-2 and IFN-γ) and Th2 (IL-4) type cytokines. By contrast, splenocytes from IL-6-deficient animals produced significantly less amount of these cytokines. Specifically, at a MOG concentration of 5 μg/ml, the amount of IL-2 produced in the control culture was 184 ± 39 pg/ml, and this was reduced by 13-fold in IL-6-deficient cell culture (14 ± 7.5 pg/ml). Similarly, the amount of IL-4 produced was decreased by 11-fold in IL-6-deficient culture (from 30.5 ± 13 pg/ml in the control to 2.7 ± 0.6 pg/ml), whereas IFN-γ was decreased by 9-fold (from 6071 ± 81 pg/ml in the control to 706 ± 110 pg/ml). Proliferation of splenocytes in response to MOG was also decreased in IL-6-deficient culture, although the decrease was not as dramatic as the reduction in cytokine production (Fig. 3).

Taken together, these results indicate that both activation and differentiation of MOG-specific T cells are hindered in IL-6-deficient mice.

Despite the potential roles of IL-6 in hematopoiesis, IL-6-deficient mice develop normally and acquire a normal immune system. No structural abnormality was ever observed in the lymphoid or nonlymphoid organs of these mice (19, 20, 21, 22, 23, 24, 25). Although a slight decrease in T lymphocyte numbers was noted in some colonies of IL-6-deficient mice (19), we have found that the numbers of cells in the spleen and lymph node of IL-6-deficient mice were essentially the same as those of the control littermates. Table II summarizes the splenocyte and lymph node cell counts in normal and IL-6-deficient mice from two independent experiments. No significant differences were observed between the two groups. To further characterize the individual lymphocyte subsets in IL-6 deficient mice, we determined the frequencies of CD3⁺, CD4⁺, CD8⁺, and B220⁺ cells in the spleen by flow cytometry. As shown in Fig. 4, the percentages of CD3⁺, CD4⁺, CD8⁺, and B220⁺ cells in IL-6-deficient mice are comparable to those in normal mice. These results suggest that the reduced anti-MOG immune responses in IL-6 deficient mice may not be due to the decrease in MOG-specific T cells.

To further determine the immune function of IL-6-deficient mice, we tested the polyclonal immune responses of splenocytes to mitogen Con A. Splenocytes from normal and IL-6-deficient mice were stimulated with Con A, and their proliferation and cytokine production determined. As shown in Table III, splenocytes from both normal and IL-6-deficient mice proliferated vigorously in response to Con A and produced large amounts of IL-2 and IFN-γ. Interestingly, IL-6-deficient splenocytes produced significantly higher amount of IL-4 than normal control cells. Experiments are under way to determine the potential roles of IL-6 in polyclonal T cell activation and IL-4 secretion.

To establish whether the blockade of T cell activation in IL-6-deficient mice also applies to other soluble Ags, we tested immune responses of IL-6-deficient mice to a model foreign Ag ovalbumine (OVA). Mice were immunized with OVA in a similar manner as for MOG and tested in vitro for their splenocyte proliferation and cytokine production in response to OVA. As shown in Fig. 5, similar blockade in T cell activation and differentiation was present in IL-6-deficient mice for OVA-specific cells. Proliferation and IL-2/IFN-γ production were dramatically reduced in IL-6-deficient mice as compared with normal control littermates. Little IL-4 or IL-10 was detected in the cultures. These results are consistent with reports that systemic T cell responses to particulate Ags or mucosal immune responses to soluble Ags are hindered in mice deficient in IL-6 (20, 21, 22, 23, 25, 26).

Although cytokines have long been implicated in the development of autoimmune diseases, their precise roles in the activation and effector function of autoreactive lymphocytes remain to be established. Recent work suggests that proinflammatory cytokines such as IL-1, IL-6, and TNF may be important mediators of inflammation (4, 5, 6). These cytokines may mediate inflammation directly through their cytotoxicity, or indirectly through modulating the functions of inflammatory cells such as macrophages and granulocytes. However, whether any of the proinflammatory cytokines effects autoimmune disease through regulating the activation stage of autoreactive T cells is unknown. To our knowledge, this is the first report examining the roles of IL-6 in the activation of autoreactive T cells in vivo using IL-6-deficient mice. Results presented here strongly suggest that cytokine IL-6 may play important roles in the activation and differentiation of autoreactive T cells in vivo.

Activation of T cells may require a minimum of two signals: the first signal delivered by MHC-peptide complex and a second signal delivered either through cell surface molecules or soluble cytokines. Although recent studies on costimulation have focused primarily on cell surface molecules such as B7, cytokines may also play important roles in promoting T cell activation. Our data suggest that IL-6 may be one of such molecules that is required for the activation of autoreactive T cells in vivo. In the absence of IL-6, autoreactive T cells may not be activated and may not differentiate into Th1 or Th2 type effector cells; therefore, autoimmune encephalomyelitis may not occur even with active immunization with myelin Ags and adjuvants.

The precise mechanisms of IL-6 action in the generation of autoreactive effector T cells need now to be investigated. IL-6 may deliver direct costimulatory signals to T cells (3, 10). In this regard, IL-6 receptor has been shown to be expressed on both resting and activated T cells and IL-6 receptor ligation activated CCAAT/enhancer-binding protein and AP-1 family of transcription factors (37, 38, 39, 40). Therefore, IL-6 may directly act as a costimulatory molecule for T cell activation. On the other hand, IL-6 may indirectly regulate T cell activation through modulating the expression and function of other molecules such as class II MHC that is required for T cell activation. In the absence of IL-6, these pathways may be blocked. Experiments are under way to prove or disprove these hypotheses.

Our observation that polyclonal T cell activation is normal in IL-6-deficient mice confirms earlier reports that immune response to mitogens in IL-6-deficient mice is comparable to that of normal mice (19, 20). However, these results also question why IL-6 is required for activation of MOG-specific or OVA-specific T cells but not polyclonal T cell activation induced by mitogens. One possibility is that polyclonal activation induced by Con A may not require the presence of all costimulatory molecules, since Con A may directly interact with certain glycosylated receptor molecules on the surface of T cells. Indeed, Perrin et al. (41) recently showed that B7-1 and B7-2 may be the only costimulatory molecules involved during Con A-stimulated T cell activation. To determine whether B7 expression is normal in IL-6-deficient mice, we examined B7-1 and B7-2 expression by flow cytometry. Briefly, splenocytes and lymph node cells were stained with flurochrome-labeled Abs to B7-1, B7-2 and various cell surface markers for B cells, T cells, and macrophages. Multicolor flow cytometry was performed to determine the levels of B7 expression in different cell types. B7-1 was not detected in either normal or IL-6-deficient mice. By contrast, B7-2 was expressed in both normal and IL-6-deficient mice by macrophages. Interestingly, the levels of B7-2 in IL-6 deficient mice are either comparable to or slightly higher than those in normal mice (B.H. and Y.C., unpublished observations).

It is to be noted that our experiments do not directly address the question whether IL-6 plays a role in the effector stage of EAE. Activation of autoreactive T cells is severely compromised in IL-6-deficient mice, making it difficult to determine the functions of IL-6-deficient effector cells in EAE. Nonetheless, the complete resistance of IL-6-deficient mice to EAE suggests that disease could not develop despite the presence of low degree of activation of MOG-specific T cells in IL-6 deficient mice (as shown in Fig. 3). Studies are under way to further address these issues.

Although MBP and PLP have been most widely used in previous EAE studies, MOG has been recently shown to be an important autoantigen involved in autoimmune demyelination. Thus, in multiple sclerosis patients, anti-MOG immune responses were shown to be significantly elevated, suggesting that MOG may be a target Ag involved in the pathogenesis of the disease (42). In mice, EAE induced by MOG is characterized by focal CNS inflammation with severe demyelination, similar to that induced by MBP or PLP (32, 33, 43). Although anti-MOG Abs have been implicated in the pathogenesis of MOG-induced EAE, MOG-specific T cells play a crucial role in the initiation of the disease (32, 33, 43). In the course of our current study, we have also tested anti-MOG38–50 Abs in the blood of mice immunized with MOG38–50 peptide by ELISA (E.B.S. and Y.C., unpublished observation). Little or no anti-MOG38–50 Abs were detected in immunized normal or IL-6 deficient B6.129 mice, suggesting that MOG38–50 may not contain B cell epitopes recognized in these mice. Therefore, EAE induced in B6.129 mice by MOG38–50 peptide may be primarily mediated by T cells.

In summary, we have discovered a critical role for IL-6 in the development of EAE. This finding may not only be important for our understanding the basic mechanisms of autoimmunity but also aid in designing novel therapeutic strategies for the treatment of autoimmune diseases such as multiple sclerosis.