We have explored the phenotype and regulation of Th1 cell activation by the cytokines IL-12 and IL-18. We demonstrate that these two cytokines selectively induce IFN-γ in a differentiated Th1 cell population through the previously described p38 mitogen-activated protein (MAP) kinase pathway. Using a highly selective p38 MAP kinase inhibitor, we demonstrate that it is possible to block IFN-γ induction from activated, differentiated Th1 cells via p38 MAP kinase without disrupting the activation and differentiation of naive T cells or the proliferation of naive or differentiated T cells. In addition, IL-12 and IL-18 provide an Ag and IL-2-independent survival signal to this uniquely differentiated Th1 cell population. We hypothesize that this Ag-independent survival of Th1 cells may participate in an innate inflammatory loop with monocytes at the sites of chronic inflammation. In addition, p38 MAP kinase inhibition of this cytokine-regulated pathway may be a unique mechanism to inhibit chronic inflammation without disruption of Ag-driven activation and function of naive T cells.
Initiation of an immune response by CD4+ T cells requires presentation of MHC-bound peptide by an APC. The naive T cell specifically recognizes Ag via its clonally unique TCR and its associated CD3 complex as well as various coreceptors, including CD4 and CD28. In a productive immune response, the initial contact with Ag results in expansion of the responding CD4+ T cell population by cytokines binding to the IL-2R family. This expanded Th1 subset of CD4+ cells then produces inflammatory mediators including IFN-γ and expresses cell surface regulatory molecules such as Fas ligand when the TCR is subsequently engaged. As the source of Ag is cleared by the acute immune response, the expanded, inflammatory CD4+ T cell population is resolved largely through the death of the CD4+ effector population or via the generation of memory T cells (1, 2).
Recently, a number of laboratories have reported that IL-12 and IL-18 synergize to stimulate IFN-γ production from differentiated Th1 cells independently from Ag receptor stimulation (3, 4). The signaling pathway for IL-12 and IL-18 induction of IFN-γ is thought to be regulated by the p38 mitogen-activated protein kinase (MAPK) pathway in contrast to IFN-γ production from T cell Ag receptor stimulation of the NFAT pathway (3, 4, 5). However, in these reports, the MAPK inhibitor used has significant crossover to the c-Jun N-terminal kinase (JNK), phosphoinositol-dependent kinase, and Raf pathways as well (6). Nevertheless, Th1 induction of IFN-γ can occur by two different signaling mechanisms: 1) via Ag stimulation of the TCR complex which is inhibited by cyclosporin A and/or 2) by the cytokines IL-12 and IL-18, which are inhibited by a p38 MAPK/JNK kinase inhibitor. The difference in signaling pathways for Ag-dependent and Ag-independent IFN-γ production could be therapeutically significant since a selective p38 MAPK inhibitor could disrupt the Ag-independent, cytokine-driven inflammatory loop without impairing the ability of naive T cells to respond to new infection.
We have examined these two signaling pathways in differentiated Th1 cells and further characterized the effects of IL-12 and IL-18 on Th1 cells. We demonstrate that, in addition to their induction of IFN-γ, these innate cytokines also induce the activation markers CD25 and CD69 and prevent apoptosis while having no affect on proliferation. Furthermore, using a selective p38 MAPK inhibitor, we show that this cytokine pathway selectively stimulates IFN-γ production in a p38-dependent fashion while inhibition of p38 MAPK does not impact cell proliferation. Thus, in a chronic inflammatory environment, IL-12 and IL-18 produced by monocytes and possibly B cells can regulate IFN-γ production by a differentiated Th1 cell population, creating an Ag-independent inflammatory loop. An understanding of this unique cytokine-driven Th1 cell population and its signaling pathway involving p38 MAPK may give novel insights into the lack of efficacy in the treatment of chronic inflammatory conditions of drugs like cyclosporin and anti-TCR biologic agents that are selective for the Ag-specific pathway.
Materials and Methods
Recombinant mouse IL-12, recombinant mouse IL-18, and SC-409 were gifts from Pharmacia (Chesterfield, MO). The IL-12 used for Th1 differentiation was a gift from Dr. K. Murphy (Washington University, St. Louis, MO). Cyclosporin A and SB203580 were purchased from Calbiochem (La Jolla, CA).
Cell lines were derived from myelin basic protein (MBP)-specific (Ac1–11) TCR-transgenic mice (Vα4 and Vβ8.2) on the B10.PL background (7) as previously described (8). RPMI 1640 culture medium supplemented with 10% (v/v) heat-inactivated FCS, 15 mM HEPES, 1% nonessential amino acids, 1% sodium pyruvate, 1% l-glutamine, 0.5% penicillin/streptomycin was used in all experiments. Fresh CD4+ T cells were prepared as described previously (9) by harvesting splenocytes from wild-type MBP TCR-transgenic mice, purified with Histopaque 1.119, and incubated on a nylon wool column for 1 h at 37°C. Nonadherent cells were washed and resuspended to a concentration of 20–50 × 106 cells/ml and incubated with an additional equal volume of culture supernatant containing anti-CD8 (3.155) and anti-heat-stable Ag (J11d) on ice for 1 h. Cells were washed and resuspended in reconstituted Low-Tox rabbit (Accurate Chemical and Scientific, Westbury, NY) complement (1/10 dilution) in a volume equal to the volume of hybridoma supernatant used and incubated at 37°C for 1 h. Live cells were purified by flotation on a Histopaque 1.119 step gradient and cultured at a concentration of 1 × 106 cells/ml, 5 × 106 irradiated splenocytes with 10 μg/ml MBP, 10 U/ml mouse IL-2, and 10 U/ml recombinant mouse IL-12 for production of Th1 lines. To prepare Th2 cell lines, 100 U/ml IL-4 replaced IL-12 in the culture.
RNase protection assay
Total RNA was isolated from samples using TriReagent (Molecular Research Center; Cincinnati, OH) as per the manufacturer’s instructions. Multiprobe DNA templates mCK-1, mCK-3b, mAPO-2, and mAPO-3 were purchased from BD PharMingen (San Diego, CA). 32P-Labeled UTP (PerkinElmer Life Sciences; Boston, MA) was used in the generation of radioactive RNA probes. Transcription from the DNA templates was accomplished using Ambion’s (Austin, TX) MAXIscript T7/T3 kit as per the manufacturer’s instructions. The subsequent hybridization, RNA digestion, and the resolution of bands on an acrylamide gel were performed using Ambion’s RNase protection assay III kit following the manufacturer’s recommendations. Bands were visualized after exposure of the gel to x-ray film with an enhancer screen at −80°C.
Biotinylated rat anti-mouse CD25 (7D4), biotinylated rat IgM, biotinylated hamster anti-mouse CD69 (H1.2F3), biotinylated hamster IgG, and streptavidin-PE were purchased from BD PharMingen. After washing once in FACS medium (HBSS supplemented with 0.2% BSA, 0.1% sodium azide, and 15 mM HEPES), cells were incubated with 10 μg/ml biotinylated Ab or the isotype control Ab for 20 min on ice. Cells were washed three times in FACS medium and then resuspended in 5 μg/ml streptavidin-PE for 20 min on ice. After washing three times, the stained cells were identified using a FACSCalibur flow cytometer (BD Immunocytometry Systems, San Jose, CA) and then analyzed by either CellQuest (BD Immunocytometry Systems) or WinMDI (J. Trotter, The Scripps Research Institute; La Jolla, CA).
Intracellular cytokine staining
FITC-conjugated rat anti-mouse IFN-γ (XMG1.2), FITC-conjugated rat IgG1, PE-conjugated rat anti-mouse IL-4 (BVD4-1D11), and PE-conjugated rat IgG2b were purchased from BD PharMingen. After 5–7 days poststimulation, cells were purified from culture over a Histopaque 1.077 gradient. Briefly, 1 × 106 cells were restimulated on plates precoated with anti-CD3 (10 μg/ml PBS), with concurrent blocking of secretion by treatment with 1 μg/ml brefeldin A (Sigma-Aldrich, St. Louis, MO) for 2–4 h at 37°C. Cells were blocked with 10% buffered Formalin phosphate (10% paraformaldehyde in PBS) for 20 min at room temperature, washed with PBS, and permeabilized with PBS supplemented with 0.5% BSA, 0.1% sodium azide, and 0.1% saponin for 10 min at room temperature. After washing, the cells were stained with 10 μg/ml IL-4-PE and IFN-γ-FITC or isotype controls for 5 min at room temperature. After washing four times, cells were collected using a FACSCalibur (BD Immunocytometry Systems) and were analyzed using CellQuest (BD Immunocytometry Systems).
The hamster anti-mouse IFN-γ (H22) capture Ab and the polyvalent goat anti-mouse IFN-γ Abs were gifts from Dr. B. Schreiber (Washington University). Bovine anti-goat IgG HRP-conjugated Ab was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Recombinant mouse IFN-γ for the standard curve was purchased from Calbiochem (San Diego, CA). ABTS substrate (Roche, Indianapolis, IN) was dissolved in 0.1 M sodium citrate buffer at a concentration of 0.55 mg/ml. Ninety-six-well flat-bottom enzyme immunoassay/RIA plates (Corning, Corning, NY) were coated with 100 μl of H22 Ab diluted to 1 μg/ml in sodium carbonate buffer. After incubating overnight at 4°C, the wells were washed three times with PBST (PBS plus 0.05% Tween 20; Sigma-Aldrich). The plate was blocked using 150 μl of 1% BSA in PBS for 1 h at room temperature. After washing three times in PBST, 100 μl of diluted experimental samples was added to the wells in triplicate and incubated at 37°C for 90 min. The wells were washed three times with PBST and then 100 μl of polyvalent goat anti-mouse IFN-γ (diluted 1/4000 in PBST) was added to each well, incubating for 1 h at room temperature. The wells were washed three times with PBST and then 100 μl of bovine anti-goat IgG-HRP (diluted to 1 μg/ml PBST) was added to each well, incubating for 1 h at room temperature. The wells were washed four times with PBST and then 100 μl of ABTS substrate supplemented with 30% H2O2 (1 μl/ml ABTS) was added to each well and analyzed at A414 using a spectrophotometer.
T cells (105) were treated with various stimuli and inhibitors (to a total volume of 200 μl) in triplicate in a flat-bottom 96-well plate for 24 h at 37°C. Afterward, 1 μCi [3H]thymidine in 50 μl of medium (replacing 50 μl removed for cytokine analysis) was added to each sample. After incubating for an additional 16–18 h at 37°C, the cells were harvested onto glass fiber filters, lysed with water, and the levels of [3H]thymidine incorporated in the DNA was measured by a scintillation counter.
5-(and 6-)Carboxyfluorescein diacetate succinimidyl ester (CFDA SE) staining, and analysis
CFDA SE was purchased from Molecular Probes (Eugene, OR). Stock CFDA SE (5 mM in DMSO) was diluted 1/100 in PBS. Then 110 μl of diluted CFDA SE per ml of culture was added directly to the cell cultures and mixed rapidly. After 5 min at room temperature, the labeled culture was pelleted, the supernatant aspirated, and then washed once in PBS. Pelleted cells were resuspended in 3 mM EDTA and incubated at 37°C for 5 min. The cells were washed once in PBS, resuspended in culture medium, and then stimulated with various cytokines. After incubating for different lengths of time, the cells were washed once in PBS, resuspended in EDTA, and incubated for 5 min at 37°C. Pelleted cells were washed once in supplemented RPMI 1640, pelleted, and resuspended in FACS medium. The cells were identified using a FACSCalibur (BD Immunocytometry Systems) flow cytometer under the FL-1 channel and analyzed using CellQuest (BD Immunocytometry Systems).
Preparation of T cell-depleted splenocytes, adherent APCs
Irradiated B10.PL splenocytes were suspended to a concentration of 50 × 106 cells/ml in warm medium. T cells were removed by adding an equal total volume of anti-CD4 hybridoma supernatant (RL.172) plus anti-CD8 supernatant (3.155) and incubating on ice for 1 h. Cells were washed with medium and resuspended in reconstituted (1/10) Low-Tox rabbit complement (Accurate Chemical and Scientific) in a volume equal to the volume of Ab used, and incubated at 37°C for 1 h. The cells were washed once in supplemented RPMI 1640 medium, purified over a Histopaque 1.119 gradient and were then ready to be used for experiments as T cell-depleted splenocytes. To prepare wells of adherent APCs, 5 × 105 cells/well were added to each well for a 96-well plate or 5 × 106 cells/well for a 24-well plate. The plate was incubated overnight at 37°C and then washed three times with fresh, warm medium. After washing, the cells were kept moist with either 50 μl (96-well plate) or 500 μl (24-well plate) of medium until ready for stimulation.
Our initial experiment confirms the observations reported by Yang et al. (5) using a different Th1 cell line specific for MBP, namely, that IL-12 and IL-18 synergize to induce IFN-γ production by a Th1 cell line (Fig. 1,A). These same innate inflammatory cytokines also synergize to increase proliferation as measured by [3H]thymidine incorporation in Th1 but not Th2 cells of the same Ag specificity (Fig. 1 B). Although both Th1 and Th2 cells predominantly express the p38α isoform (our unpublished data), it is not surprising that there is no effect on Th2 cells since IL-4 negatively modulates the receptors for both IL-12 and IL-18 (10, 11).
Historic data demonstrate a role for p38 MAPK in various T cell functions using the compound SB203580. This compound has been reported not to be specific for the p38 MAPK pathway, also inhibiting the JNK and phosphoinositol-3-dependent protein kinase 1 pathways as well (6). Thus, we compared the SB203580 compound with a more selective p38 MAPK inhibitor, SC-409. SC-409 has an IC50 of 57 nM on p38α vs an IC50 of 2.48 μM on p38β, whereas its IC50 value on JNK2/3 >30 μM; likewise, this compound inhibits p38 MAPK in cellular and in vivo assays of inflammation. For example, the IC50 for SC-409 in LPS-induced TNF-α production from monocytes is 0.04 μM (data not shown).
Fig. 2 confirms the observations by others that Ag receptor (TCR) stimulation of IFN-γ production is selectively blocked by cyclosporin (Fig. 2,A) while IL-12- and IL-18-induced IFN-γ production is selectively sensitive to the dual p38/JNK inhibitor SB203580 (Fig. 2,B) and extends these observations with the more selective p38 MAPK inhibitor SC-409 (Fig. 2,C). In contrast, while cyclosporin selectively blocks APC- plus peptide-induced proliferation, (Fig. 2,A), the SB203580 compound blocks both Ag- and cytokine-driven [3H]thymidine incorporation (Fig. 2,B), and the higher IC50 suggests it may not be acting through inhibition of p38 MAPK. In separate experiments, we have observed that similar concentrations of SB203580 block IL-2-stimulated [3H]thymidine incorporation by inhibiting G1 to S phase transition (our unpublished data), which is consistent with other reports that this compound affects Rb phosphorylation in a p38 MAPK-independent fashion (12). The lack of effect of the more selective p38 MAPK inhibitor SC-409 on [3H]thymidine incorporation (Fig. 2 C) further indicates that cytokine-driven proliferation is not dependent on activation of p38 MAPK.
Because the SC-409 p38 MAPK inhibitor did not block proliferation, it was possible to test whether p38 MAPK was involved in the regulation of the activation and differentiation of naive cells. We prepared naive CD4+ T cells from anti-MBP TCR-transgenic mice and stimulated them in culture with irradiated APCs, MBP, and IL-2 with or without a Th1-inducing agent, IL-12, or heat-killed Listeria monocytogenes (hkLM) and in the absence or presence of the SC-409 compound or cyclosporin A. Five days after the primary or 5 days after a second weekly stimulation with Ag and IL-2, the cells were harvested, stimulated with anti-CD3 in the presence of brefeldin A, and stained for intracellular IL-4 and IFN-γ (Table I). The results demonstrate that SC-409 has little affect on basal or exogenous IL-12-induced Th1 differentiation and appears to enhance the response to Listeria. As expected, cyclosporin A blocked the expansion of the primary T cells and thus could not be tested in a second stimulation.
|Inducing Agent .||Inhibitor .||Primary Stimulationa .||Secondary Stimulationa .|
|Inducing Agent .||Inhibitor .||Primary Stimulationa .||Secondary Stimulationa .|
Percent Th1 cells assessed by intracellular cytokine staining (IL-4 negative, IFN-γ positive).
ND, not done.
The effect of SC-409 on Th1 differentiation was tested in a separate experiment comparing the effect of different doses of SC-409 with exogenous IL-12 or LPS. Five days after the secondary stimulation, the percentage of Th1 cells was determined by intracellular cytokine staining. In this experiment, there was a slight, insignificant enhancement of differentiation in response to exogenous IL-12, but again significant enhancement with the more physiological stimulus LPS (Fig. 3). In three independent experiments, the mean Th1 differentiation in cultures with Ag plus IL-2 alone was 8.4 ± 3.4%, plus IL-2 + IL-12 was 33 ± 6.6%, and plus IL-2 + IL-12 + 1 μM SC-409 was 27 ± 2.6%. The enhanced Th1 differentiation in the presence of LPS or hkLM may be through the p38 MAPK suppression of monocyte-derived IL-10 production (13). Thus, inhibition of p38 MAPK does not inhibit the differentiation of naive T cells into Th1 cells.
We next investigated the mechanism by which T cell proliferation was regulated by comparing stimulation of [3H]thymidine incorporation by IL-12 and IL-18 with the stimulation seen by IL-2. Fig. 4 demonstrates that IL-12- and IL-18-stimulated proliferation is less vigorous than that stimulated by IL-2. Likewise, IL-12- and IL-18-stimulated [3H]thymidine incorporation is not blocked by an amount of antimurine IL-2 capable of specifically neutralizing 100 U/ml murine recombinant IL-2. This demonstrates that IL-12 and IL-18 do not stimulate T cell proliferation via the release of IL-2.
Since the [3H]thymidine incorporation was independent of IL-2, it was important to determine whether IL-12 and IL-18 actually stimulated cell division in the Th1 population. Cell division was examined by two alternative methods, namely, labeling the cells with the vital dye CFSE and assessing the expansion of the live cell population (trypan blue exclusion) without exogenous cytokines, with IL-12 and IL-18 or with IL-2. The number of live cells and percent dead cells was examined with time (Fig. 5,A). The number of live cells decreased in the absence of exogenous cytokine, remained relatively constant in the presence of IL-12 and IL-18, and increased significantly with IL-2. After 3 days, the number of viable cells recovered (assessed by trypan blue exclusion) was compared with the starting population and cell division was assessed by FACS analysis of dye dilution. Fewer than 60% of the cells survive in the absence of exogenous cytokines. The IL-2-treated population underwent 2.5 doublings as indicated by the 6-fold increase in cell number and the dilution of CFSE. The IL-12- and IL-18-treated population did not complete one division as demonstrated by the <2-fold increase in cell number and the limited dilution of CFSE. In a separate experiment (Fig. 5,C), the ability of IL-12 and IL-18 to suppress apoptosis was evident by the ability to suppress the development of annexin V binding, an early indicator of the apoptotic process. Thus, IL-12 and IL-18 do not actually stimulate cell division, but rather provide a survival signal allowing cells already in cycle to complete it and hence incorporate [3H]thymidine; they do not stimulate another round of division. In contrast, IL-2 provides both survival and proliferation signals. These data explain the limited [3H]thymidine incorporation stimulated by IL-12 and IL-18 compared with IL-2 (Fig. 4).
We extended our analysis of the effects of IL-12 and IL-18 on differentiated Th1 cells by further phenotyping these cells in terms of induction of cell surface activation markers and the profile of secreted lymphokines typically associated with Ag activation. Fig. 6 demonstrates that IL-12 and IL-18 induce both CD25 and CD69, although their relative contributions are different for each activation marker. IL-12 partially induces CD25, but requires IL-18 to achieve maximal expression of CD25. On the other hand, IL-18 alone is capable of maximal induction of CD69. The differential effects of the cytokines on these activation markers may be useful in determining the cytokine environment in inflammatory lesions in vivo. Inhibition of p38 MAPK with SC-409 had little or no effect on the induction of either CD25 or CD69 (data not shown), indicating that induction of these markers is p38 MAPK independent.
Finally, using RNase protection analysis we compared the spectrum of lymphokines normally produced with Ag/TCR stimulation with those stimulated by IL-12 and IL-18 (Fig. 7). Strikingly, IL-12 and IL-18 are very selective in their capacity to induce IFN-γ. Note that IL-2 is not produced in response to IL-12/IL-18 stimulation as it is in response to anti-CD3, confirming the IL-2-independent [3H]thymidine incorporation demonstrated earlier (Fig. 4). Likewise, there is no induction of other inflammatory mediators, including TNF-α or lymphotoxin, by IL-12 and IL-18. Thus, IL-12 and IL-18 provide a survival signal and selectively stimulate IFN-γ in this differentiated Th1 cell population, and the IFN-γ production is dependent on p38 MAPK activation. Since the role of IL-12/IL-18 in an Ag-independent cellular cytokine loop likely depends on IFN-γ stimulation of other cells such as monocytes, these Th1 cells may have a pivotal role in sustaining a chronic inflammatory lesion.
It is becoming increasingly clear that various chronic inflammatory diseases are dependent on Th1 cytokines and monokines (14, 15, 16, 17, 18). The experiments described herein characterize a potential novel, Ag-independent inflammatory loop between activated Th1 CD4+ cells and monocytes that is cytokine driven. The experiments confirm earlier work demonstrating that Ag-stimulated IFN-γ production uses the cyclosporin A-dependent NFAT signaling pathway while cytokine-stimulated IFN-γ production is dependent on the p38 MAPK pathway (5, 19). In addition, we determined that a selective p38 MAPK inhibitor can disrupt the cytokine pathway without interference with the activation and differentiation of new Th1 effectors. This provides a further rationale for possible therapeutic intervention for p38 MAPK inhibitors in chronic inflammation without compromising the immune system’s capacity to respond to new infectious challenges.
Our data agree with the earlier work demonstrating p38 MAPK-dependent induction of IFN-γ by IL-12 and IL-18, but p38 MAPK-independent induction of IFN-γ by TCR stimulation (5). However, both our data and that of Yang et al. (5) are in conflict with other work by Lu et al. (20) indicating a GADD45γ-dependent activation of p38 MAPK by TCR contributing to TCR-induced induction of IFN-γ. A possible difference is that we and Yang et al. (5) used TCR-transgenic models allowing us to use physiological amounts of defined Ag and APCs as the TCR stimulus. In contrast, Lu et al. (20) examined TCR-dependent IFN-γ production in wild-type and GADD45γ null cells by stimulating with immobilized anti-CD3 under conditions resulting in the death of most of the responding cells. It is possible under supraphysiological signaling conditions TCR stimulation of p38 MAPK becomes more important, but the experiments herein demonstrate that TCR-induced p38 MAPK activation is less important under more physiological conditions. This issue could best be resolved with p38 MAPK null cells.
The experiments presented here also demonstrate that IL-12 and IL-18 not only selectively stimulate IFN-γ from differentiated Th1 cells, but these cytokines also provide an IL-2-independent survival signal to the same cells. The cytokine-stimulated survival and IFN-γ secretion set the stage for the maintenance of an Ag-independent inflammatory loop between differentiated Th1 cells and monocytes at a site of chronic inflammation. We have not explored the mechanism of this IL-12 and IL-18 protection from apoptosis, except that it does not appear to involve the induction of anti-apoptotic bcl-2 family members (our unpublished observations). However, Yang et al. (5) demonstrated that the IL-12 and IL-18 induction of p38 MAPK activation was dependent on the NFκB-dependent induction of GADD45β and De Smaele et al. (21) have demonstrated that NFκB induction of GADD45β is responsible for the anti-apoptotic signaling of TNF-α which was also independent of bcl-2 induction (21). Induction of GADD45β is not sufficient for p38 MAPK activation (5) so this anti-apoptotic pathway is not affected by SC-409. However, TNF-α and other inflammatory cytokines like IL-12 and IL-18 may converge in this mechanism to prolong the life of inflammatory cells, further promoting chronic inflammation.
Fig. 8 presents a model whereby Th1 cells that normally die or become quiescent after an acute infection may be co-opted by monocytes producing IL-12 and IL-18 to participate in a chronic inflammatory loop. Would a loop of IFN-γ, IL-12, and IL-18 be sustained? Recent experiments demonstrating an NK cell, IFN-γ-dependent inflammatory response to parenterally administered IL-12 and IL-18 suggests that it may (22). NK cells may also participate in such a loop, but the data here and elsewhere suggest that activated CD4+ Th1 cells can also be important at a local site of inflammation.
Disruption of the IFN-γ gene has been reported to increase the severity and/or decrease resistance to induction of experimental autoimmune encephalomyelitis, a rodent model for multiple sclerosis (23, 24). This may be due to the role of IFN-γ on the primary expansion of effector T cell populations (25). Paradoxically, transgenic expression of IFN-γ in the CNS converts an acute model of experimental autoimmune encephalomyelitis to a chronic form of disease (26). Thus, cytokines like IFN-γ may have opposing activities in acute vs chronic stages of disease. The ability of IFN-γ to promote and sustain chronic disease is in accord with our model.
The model implies that there may be a necessary “transforming” event to place the Th1 and the monocytes in a state of activation sufficient to maintain the chronic cytokine-driven loop. Additional experiments will be required with both Th1 cells and monocytes to understand whether such transforming events occur and what their nature may be. Gene expression in the T cell resulting in the survival and sustained activation of cytokine expression may be continually evolving within the changing cytokine milieu creating a stable Th1 cell population that is unresponsive to classical death signals.
The data presented here provide a model to explain why patients with diseases that appear to have a Th1 component (e.g., the presence of T cells at the site of inflammation and/or by the profile of cytokines produced) would be relatively unresponsive to classical immunosuppressive drugs like cyclosporin A or antiproliferative agents designed to inhibit the Ag-driven T cell activation pathway. Differences in the relative contributions of the Ag vs the cytokine pathways during the clinical course of a T cell-driven chronic inflammatory disease could explain why cyclosporine is less effective with more toxicity in the treatment of rheumatoid arthritis (27) while being very effective in graft-vs-host disease (28). Likewise this model may also explain why some chronic inflammatory diseases are unresponsive to anti-TNF agents, since this simple paradigm does not require TNF for its propagation. Further work is needed to dissect the generation and role of the “innate Th1 cell” in sustaining chronic inflammation.
This work was supported by National Institutes of Health Grant AI45861 and a grant from the Pharmacia Corporation.
Abbreviations used in this paper: MAPK, mitogen-activated protein kinase; JNK, c-Jun N-terminal kinase; MBP, myelin basic protein; CFDA SE, 5-(and 6-)carboxyfluorescence succinimidyl ester; hkLM, heat-killed L. monocytogenes.