The response of Th cells to cytokines is normally strictly regulated, such that following antigenic stimulation, Th cells respond for only a short period of time, after which they become refractory to cytokine-mediated effects. IL-12, a costimulator of Th1 having no proliferation-inducing capacity of its own, allows Th1 clones and lines to respond to IL-4 when they would otherwise be unable to respond to this cytokine. Cells that have proliferated in response to IL-4 plus IL-12 are fully able to be subsequently activated by specific Ag and APC. Additionally, the response to IL-4 of Th1 effector cells derived from normal murine spleen is enhanced significantly by IL-12. Furthermore, in the presence of IL-12, stimulated Th2 can induce proliferation of Th1 via IL-4 production, in a dual chamber culture system. We hypothesize that the effects of IL-4 and IL-12 represent a novel, positive cross-regulatory pathway that acts on Th1, and is mediated by Th2 (the IL-4 source) and APC (the IL-12 source). We propose this as a way for a Th2 immune response to positively influence an ongoing or waning Th1 response.
The cellular arm of the immune system is generally controlled by Th1 cells, which selectively secrete IL-2, IFN-γ, and lymphotoxin. The second type of Th cells, Th2, generally mediates humoral immunity and also displays a unique cytokine secretion profile, including IL-4, IL-5, and IL-10 1, 2 . Several of these cytokines, along with certain APC-derived cytokines such as IL-1 and IL-12, are involved in complex differentiation and cross-regulation pathways between Th1 and Th2 cells. These pathways of differentiation and cross-regulation between Th1/Th2, which have both been recently reviewed 3, 4, 5, 6, 7, 8 , have positive or stimulatory, and negative or inhibitory components.
Two of the cytokines involved in Th1/Th2 interactions, IL-4 and IL-12, have been shown to be essential for differentiation of Ag-activated precursor T cells, termed Th0, into the two Th subtypes. IL-12 acts with IFN-γ to strongly promote differentiation into Th1, and IL-4 acts similarly for the generation of Th2 3, 5, 7 . Thus, with regard to these differentiation-promoting functions, IL-4 and IL-12 are mutually antagonistic. In fact, when Th0 cells differentiate into Th2 in the presence of IL-4, their ability to respond to IL-12 is rapidly extinguished (within 16–48 h) 9, 10 . This has been shown to be due to down-regulation of the signaling chain (the β2-chain) of the heterodimeric IL-12R 11, 12 .
Following differentiation, Th1 cells have been shown to be able to respond to both IL-4 and IL-12. IL-4 acts as a direct stimulus to promote proliferation in Th1 cells that have been activated by Ag stimulation through the TCR 13 . Th1 cells typically do not respond to IL-4 in the absence of TCR ligation. Furthermore, IL-4 can synergize with the autocrine Th1 growth factor IL-2 to induce high levels of proliferation 13, 14, 15 . IL-12, on the other hand, has no ability to directly stimulate Th1 proliferation, but rather acts as a costimulatory molecule, allowing enhanced Th1 responsiveness to cytokines produced after TCR ligation 16 . IL-12 also is a potent inducer of IFN-γ production by Th1 17 .
The effects of IL-4 and IL-12 on Th cell differentiation are essentially mutually exclusive. However, there are examples in which IL-4 and IL-12 interact simultaneously in the same cell, and this can occur in various types of cells. In macrophages and dendritic cells, IL-12 production was inhibited by IL-4 when the cells were stimulated with LPS and IFN-γ, whereas IL-12 production induced by CD40-CD40 ligand interaction was enhanced by IL-4 18 . IL-12 enhances differentiation of murine erythroid progenitor cells induced by erythropoietin plus IL-4 19 . The two cytokines have also been described to interact in T cells. They have been shown to synergistically promote CTL responses 20 . An IL-4-dependent cell line CT4S responds to IL-4 and IL-12 in a synergistic fashion 21 , and Kennedy et al. demonstrated that murine Th1 clones responded to a combination of IL-4 and IL-12 16 . Interestingly, their Th1 cells did not have to be activated by stimulation through the TCR to be able to respond to the two cytokines.
The observation by Kennedy et al. 16 that IL-4 plus IL-12 induced proliferation in resting Th1 in an Ag-independent manner prompted us to investigate the mechanism by which proliferation is stimulated, and also to determine whether this in vitro phenomenon could represent a novel Th1/Th2 cross-regulatory pathway. We hypothesized that Th2 cells could positively regulate resting Th1 cells through the direct action of IL-4 and the indirect action of IL-12 derived from APC. We present in this work the results of our attempts to better understand the action of IL-4 plus IL-12 on Th1, and to assign biological significance to the phenomenon.
Materials and Methods
Female DBA/2J mice used for these studies were purchased from The Jackson Laboratory (Bar Harbor, ME). They were used between 2 and 6 mo of age.
Cells and lines
Th lines and clones were generated as previously described 22 . Briefly, female DBA/2J mice were immunized at the base of the tail with 100 μg of synthetic peptide in CFA. The two peptides used were derived from sperm whale myoglobin (SpW Mb) protein (amino acids 110–121 and 132–147) and were synthesized by the University of Pittsburgh Peptide Synthesis Facility (Pittsburgh, PA). Clones were generated from inguinal lymph node cell lines by standard techniques. Clones and lines were maintained in RPMI 1640 (Life Technologies, Grand Island, NY) supplemented with HEPES (10 mM), penicillin (100 U/ml), streptomycin (10 μg/ml), 2-ME (0.16 mM), l-glutamine (2 mM), and 10% FCS (Life Technologies), and were restimulated every 10–14 days with irradiated (2500 rad) syngeneic spleen APCs and specific antigenic peptide. Cytokine expression was used to delineate Th1 (expressing IL-2 and IFN-γ) and Th2 (expressing IL-4 or IL-10) phenotypes. The lines AII (110–121 specific) and 132 (132–147 specific) were polarized for the production of Th1 and Th2 cytokines, respectively. Clones A.2.5S and A.4.3 (both 110–121 specific) were shown to be Th1, and clone 31.F.6 was shown to be Th2. The Th2 clone 13.26 has been described previously 23 .
Spleen T cells were purified by a multistep depletion process. Culture medium used for T cell preparation was DMEM (Life Technologies) supplemented as described above. Fresh spleen cells were first incubated in plastic cell culture plates (#3003 plates; Falcon) for 45–60 min at 37°C, with 5 × 107 cells in 5 ml medium per plate. After gentle rinsing, nonadherent cells were incubated with anti-B cell mAb-containing hybridoma culture supernatant (anti-B220) and anti-MHC class II mAb supernatant (MKD6; anti-I-Ad) for 30 min at 4°C. This was followed by incubation with rabbit complement (diluted 1/10 in culture medium, Cedarlane Low-Tox-M rabbit complement; Accurate Chemical, Westbury, NY) for 40 min at 37°C. Cells prepared in this fashion generally were greater than 95% T cells, as determined by staining for specific T cell markers and flow cytometry, as described previously 22 .
Spleen-derived APCs were prepared by complement depletion, as described above, using anti-Thy-1.2 24 mAb-containing hybridoma culture supernatant to remove T cells. APCs were irradiated with 2500 rad before use.
T cells were plated in 96-well microtiter culture plates (Corning Glass, Corning, NY) with 5 × 104 cells/well. Cytokines and or anti-cytokine mAbs were added, and the total culture volume was adjusted to 0.2 ml. Experiments using Th clones were performed using RPMI-based culture medium, and those using splenic T cells with DMEM-based culture medium. Cultures were incubated at 37°C for 24 h, at which time the wells were pulsed with [3H]thymidine ([3H]TdR; DuPont NEN, Boston, MA) at 0.5 μCi/well. [3H]TdR incorporation was measured by harvesting cells onto filter plates using a 96-well cell harvester (Packard Instrument, Downers Grove, IL) and counting on a TopCount liquid scintillation counter (Packard).
Cytokines used in these assays included IL-1 (Genzyme, Cambridge, MA), IL-2 and IL-4 (R&D Systems, Minneapolis, MN), and IL-12 (3.5 × 105 U/per μg; a gift from Dr. M. Lotze, University of Pittsburgh, PA). These cytokines were neutralized by mAbs including anti-IL-2 (PharMingen, San Diego, CA), anti-IL-4 25 (10% hybridoma 11B11 supernatant), and anti-IL-12 (G28-A; a gift from Dr. J. Flynn, University of Pittsburgh, PA). IL-1 was neutralized using IL-1R antagonist (R&D Systems). Assays of IL-1 and IL-4 responsiveness in Th2 were performed on plates that had been coated overnight with 10 μg/ml anti-TCR mAb (H57-597) 22, 26 .
Treatment of splenic T cells for generation of activated Th1 and Th2 effector cells
Activated Th effector cells were generated essentially as previously described 27 . Fresh splenic T cells were cultured in 12-well plates (5 × 106 cells/well), with irradiated T-depleted splenic APCs (1 × 107 cells/well), Con A (2.5 μg/ml), and IL-2 (10 U/ml). Cells that had undergone such treatment were referred to as having been generated under neutral conditions. In addition, Th1 effector cells were generated by adding IL-12 (50 U/ml) and anti-IL-4 (10% hybridoma 11B11 supernatant) to the reagents used to generate cells under neutral conditions. Th2 effector cells were similarly generated using IL-4 (300 pM) and anti-IFN-γ (10 μg/ml; PharMingen). Cells were incubated under these three conditions for various lengths of time, after which they were washed and purified on Ficoll-Hypaque (Sigma, St. Louis, MO) before further use. The development of Th1 and Th2 effectors was monitored by assaying culture supernatants for the Th1-specific cytokine IFN-γ, and the Th2-specific cytokine IL-5 by ELISA assay. ELISAs were performed using cytokine-specific paired primary and secondary reagents (PharMingen), as described 28, 29 .
Assays of Th1/Th2 cytokine cross-regulation were performed using a dual chamber transwell culture system (Costar). Th2 clones or Con A-generated Th2 effector cells were placed in the bottom chamber of the transwell either alone, or following coating of the chamber with anti-TCR mAb. Variable numbers of Th2 cells were used depending on the particular experiment. Th1 clones or effectors (5 × 105 cells/well) were cultured in the upper chamber. Cells were incubated in this manner either alone or in the presence of IL-4, IL-12, or anti-IL-4 mAb. Total culture volume was 1 ml, which allowed for an approximate volume of 0.25 ml in the upper chamber. Assays with cytokine- or anti-TCR-stimulated cells were incubated for 24 or 48 h, respectively, before an 18-h pulse with [3H]TdR (2.5 μCi/well). Cells were transferred from the transwells into 96-well plates before harvesting and counting, as described above.
The combination of IL-4 and IL-12 induces proliferation in resting Th1 cells
Upon antigenic stimulation through the TCR, Th cells acquire the ability to proliferate in response to cytokines. Typically, activated Th cells respond to stimulation by cytokines for only a limited period of time, after which they become refractory to further stimulation 15 . This principle is demonstrated in Fig. 1 for the response of a Th1 and a Th2 clone to IL-4 in the days following stimulation with specific Ag presented by APC. Our clones, of which those in Fig. 1 are representative, respond to IL-4 maximally at 2 days poststimulation. On subsequent days, the response to IL-4 diminishes, until at some point stimulation is entirely absent. The representative Th1 clone lost all IL-4 responsiveness by 4 days poststimulation (Fig. 1,A), whereas the Th2 clone retained some ability to proliferate in response to IL-4, albeit at a greatly reduced level, 2 wk poststimulation (Fig. 1 B). Normally, our clones do not exhibit IL-4 responsiveness beyond 5–7 days poststimulation, and are termed resting Th cells.
We and others have shown previously that IL-1, which is a costimulatory factor for Th2, can prolong the ability of some Th2 clones to proliferate in response to IL-4 22, 30, 31, 32 . Similarly, IL-12 has been described as a costimulatory cytokine for activated Th1, and surprisingly, it also appears to modulate the response of resting Th1 to IL-4 16 . We thus tested responsiveness of our Th1 clones and lines to the combination of IL-4 and IL-12. We found that not only could IL-12 enhance proliferation of Th1 clones in response to antigenic stimulation as expected (16 and data not shown), but that this cytokine enabled resting Th1 cells to respond to IL-4. As shown in Fig. 2, two Th1 clones (A.2.5S and A.4.3, and a Th1 polarized line, AII; Fig. 2, A–C) were completely unable to respond to high levels of IL-4 (up to 300 pM). Yet when IL-12 was added, the cells were able to proliferate in response to IL-4. The level of proliferation observed was directly related to the amount of IL-12 added at a given concentration of IL-4 (Fig. 2, E and F). The amount of IL-12 added was lower than that used in the previously published report (0.1–100 U/ml or 0.286–286 pg/ml, versus 1–20 ng/ml) 16 . The cells did not proliferate in response to IL-12 alone, but only when the two cytokines were present together. The effect was observed in all of the Th1 lines and clones that have been tested (Fig. 2 and data not shown). In contrast, Th2 cells, which normally make IL-4 and proliferate in response to it in an autocrine fashion after activation through the TCR (13 and data not shown), did not respond to the combination of IL-4 and IL-12 when in a resting state (Fig. 2 D). This is not at all unexpected because Th2 cells lose the ability to respond to IL-12 early in their differentiation from a naive phenotype 9, 10 .
Analysis of possible mechanisms of the IL-4 + IL-12 effect
We attempted to define the mechanism by which IL-12 renders Th1 cells susceptible to stimulation by IL-4. First, we wanted to determine whether the two cytokines had to be present together to induce Th1 proliferation, or if cells could be treated with IL-12 and then respond to IL-4 subsequently in a manner analogous to antigenic stimulation through the TCR, such as in Fig. 1. Our previous work showed that in some Th2 clones, IL-1 pretreatment mediates a subsequent, prolonged responsiveness to IL-4 22 . In contrast, we and others found that pretreatment with IL-12 for up to 24 h did not render Th1 cells able to subsequently respond to IL-4 (16 and data not shown). In fact, both IL-4 and IL-12 had to be present essentially throughout the duration of a proliferation assay to stimulate a proliferative response. A Th1 clone was incubated with IL-4 and IL-12, and neutralizing Abs to either IL-4 or IL-12 were added at various times poststimulation. When neutralizing mAbs to either IL-4 or IL-12 were added at time points before 24 h after the beginning of the assay, little or no proliferation was observed (Fig. 3, A and B). Adding either mAb beyond 24 h had no effect during the remaining 18 h of the assay (i.e., during the [3H]TdR pulse). These results are in contrast to our results from experiments examining the costimulatory effects of IL-1 on Th2 cells. In these experiments, the presence of IL-1 was only required early, and maximal IL-4 responsiveness was observed after IL-1 had been present for 8 h (Fig. 3,C). Similar to Th1, IL-4 had to be present throughout the duration of the assay in order for IL-1 to elicit a response in Th2 (Fig. 3 D). In addition, the costimulatory effect of IL-1 required prior antigenic stimulation in the form of anti-TCR mAb coated on the plate, which was not required in the case of IL-12 costimulating the IL-4 response in Th1.
A likely possibility for the proliferative effect of IL-4 + IL-12 was that IL-12 up-regulated expression of the IL-4R on the cell surface. We performed flow-cytometric analysis on Th1 cells before and after IL-12 treatment and found that IL-4R expression was comparable in both cases. In a representative experiment, IL-4R was detected on essentially all resting Th1 clone cells (mean fluorescence intensity of 16.81 versus the 6.35 for staining with an isotype control Ab). After overnight stimulation with IL-12, IL-4R expression was essentially unchanged, with a similar number of cells staining positively (mean fluorescence intensity of 16.37 versus 6.26 for the control). Furthermore, this result was confirmed by RT-PCR analysis on RNA obtained from unstimulated and IL-4 + IL-12-stimulated cells. RNA was extracted at several time points ranging from 2–24 h poststimulation. Image analysis of RT-PCR products revealed no difference in levels of IL-4R mRNA expression relative to the level of β-actin mRNA (data not shown). Thus, it seems unlikely that up-regulation of the IL-4R by IL-12 mediates the induction of IL-4 responsiveness.
We also tested whether IL-4 + IL-12 treatment of Th1 induced the production of IL-2, which Th1 cells normally utilize for proliferation following antigenic stimulation. However, proliferation of clone A.2.5S, which responds to IL-2 constitutively (unpublished observation), was unaffected by neutralizing Ab to IL-2 after treatment with IL-4 + IL-12 (Fig. 4,B). The amount of anti-IL-2 mAb used was able to inhibit comparable levels of proliferation in the same clone in response to exogenously added murine IL-2 (Fig. 4 A). Furthermore, RT-PCR analysis showed no detectable IL-2 mRNA in cells stimulated with IL-4 + IL-12 (data not shown). Therefore, it appears that IL-12 and IL-4 could interact within cells to promote proliferation, possibly through common intracellular signaling pathways.
Possible functional significance of Th1 responsiveness to IL-4 + IL-12
We have shown that IL-4 and IL-12, when present together, provide a modest stimulatory signal to Th1 cells, and we know that the effect occurs at low levels of IL-12 (as low as 0.1 U/ml, or 0.286 pg/ml, in Fig. 2). We wanted to determine whether IL-12 would allow Th1 to proliferate in response to IL-4 at levels produced by Th2. To do this, cultures were set up in transwell culture plates, which consists of upper and lower chambers separated by a membrane filter that is permeable to soluble factors such as cytokines, but not to cells. The Th1 clone was cultured in the upper chamber, with a Th2 clone in the lower chamber. A Th1 clone placed in the upper chamber proliferated in response to IL-4 plus IL-12, and the response could be blocked by neutralizing anti-IL-4 Ab (Fig. 5,A). Stimulation of the Th2 clone with anti-TCR mAb, which induced IL-4 production but not proliferation (data not shown), had no effect on the Th1 unless IL-12 was added (Fig. 5,B). At the concentration of IL-12 used in this experiment (50 U/ml or 143 pg/ml), there was some proliferative response of the Th1 to IL-12 alone in some, but not all experiments. We assume this IL-12-induced proliferation is the result of a failure of the Th1 cells to reach a fully resting state in these cases, since previously activated cells do appear to be able to respond to IL-12 alone for a period of time (for example, see Fig. 6). However, stimulation of the Th2 clearly enhanced Th1 proliferation in the presence of IL-12, which could be blocked by anti-IL-4 mAb (Fig. 5,B). The anti-TCR mAb, which was coated onto the bottom of the lower chamber, had no effect on the Th1 in the upper chamber in the absence of Th2 or IL-12 (compare the second bars from the left in Fig. 5, A and B, which differ only by the presence or absence of anti-TCR mAb). These results suggest that this form of cross-regulation could occur in a natural setting in which a Th2 effector response is underway in the simultaneous presence of an IL-12 source, such as macrophages or dendritic cells 33, 34, 35, 36 .
The proliferation in response to IL-4 and IL-12 indicates that there could be an increased pool of cells capable of responding to specific Ag. Therefore, we wanted to be sure that cells that had proliferated in response to IL-4 plus IL-12 could still respond to Ag. Th1 clones were treated with the combination of cytokines, were tested for Ag responsiveness, and were found to be fully capable of proliferating in response to Ag presented by APC (data not shown).
Effect of IL-4 + IL-12 can be observed in freshly derived T cell effectors
We assessed responses of normal T cells to IL-4 and IL-12. T cells were purified from murine spleen and were assayed for responsiveness to IL-4 and IL-12. Unlike the Th1 lines and clones (Fig. 2), freshly isolated spleen T cells exhibited no proliferation in response to IL-4 and IL-12 (Fig. 6,A). However, upon activation with Con A and IL-2, a response was seen in response to both cytokines. Two days after Con A stimulation, there was a vigorous response to either IL-4 or IL-12 alone, and IL-12 slightly enhanced the IL-4 response, particularly at lower concentrations of IL-4 (Fig. 6,B). By 6 days poststimulation, the response to IL-12 alone was absent, the response to IL-4 alone was minimal, and IL-12 had a more significant enhancing effect on the IL-4 response (Fig. 6 C).
We then treated spleen T cells with Con A + IL-2 alone and under conditions that favor the development of Th1 or Th2 effector cells (Con A + IL-2 and IL-12 + anti-IL-4, or IL-4 + anti-IFN-γ, respectively) 27 . Appropriate cytokines were generated under each of the conditions (IFN-γ for Th1 and IL-5 for Th2), as determined by ELISA assay of culture supernatants (Fig. 7, top panels). At 6 days poststimulation, IL-12 slightly enhanced the IL-4 response in cells stimulated under neutral conditions, as shown previously in Fig. 6,C (Fig. 7,A), and cells generated under Th2-promoting conditions exhibited little or no IL-12 responsiveness (Fig. 7,C). In cells generated under Th1-promoting conditions, IL-12 induced a significant enhancement of proliferation even at very low levels of IL-4 (Fig. 7 B). Therefore, normal Th effector cells with a Th1 phenotype do respond to the combination of IL-4 and IL-12 in an Ag-independent fashion in a manner analogous to that previously observed in Th1 clones.
Finally, we wanted to demonstrate that this form of Th1/Th2 cross-regulation could occur between normal spleen T cells using the transwell system. In a control experiment, 6-day-old Con A + IL-2-stimulated Th1 effector cells were placed in the top well, and Th2 effector cells were placed in the bottom well. At the concentrations used, IL-4 and IL-12 individually did not induce detectable Th1 proliferation above the background level, but as expected, induced Th1 proliferation when present together (Fig. 8,A). All of the proliferation observed in response to IL-4 + IL-12 was blocked by anti-IL-4 mAb (Fig. 8,A). Th2 cells in the bottom chamber did not respond to the level of IL-4 used alone or in conjunction with IL-12 (data not shown). Th1 and Th2 effector cells were similarly cultured with the Th2 cells stimulated by anti-TCR mAb coated onto the bottom of the wells. Anti-TCR mAb stimulation of Th2 led to IL-4 production, which induced proliferation of the Th2 cells themselves (Fig. 8,B). The Th1 cells in the top chamber of the same wells also proliferated vigorously in response to the IL-4 produced by the Th2 cells in the bottom chamber (Fig. 8,B). The addition of IL-12 in this experiment had no additional proliferative effect (data not shown), and we hypothesized that high levels of IL-4 may have rendered any IL-12-specific effect unobservable, as was the case with high levels of IL-4 shown in Fig. 7,B. In a separate experiment, the concentration of IL-4 was reduced by lowering the number of Th2 effector cells in the bottom chamber (Fig. 8,C). In this experiment, only a small amount of IL-4-specific proliferation was induced in the Th1 cells in the upper chamber. As was seen on occasion with the Th1 clones, IL-12 alone had some stimulating capacity, but greatly enhanced Th1 proliferation was induced by adding IL-12 to wells containing anti-TCR mAb-stimulated Th2 cells in the bottom chamber (Fig. 8 C). This proliferation was inhibited by the anti-IL-4 mAb. These experiments show that activated Th2 cells can provide bystander help for Th1 in the presence of an IL-12 source.
The roles of IL-4 and IL-12 in Th cell differentiation have been well described. IL-12, along with IFN-γ, mediates the generation of the Th1 subtype, and IL-4 promotes differentiation of Th0 to the Th2 subtype 3, 5, 7 . Therefore, it is generally thought that the roles of IL-4 and IL-12 are mutually antagonistic. However, it has been shown that Th1 clones that have returned to a resting, nonproliferating state following antigenic stimulation could be induced to proliferate in an Ag-independent manner by a combination of IL-4 and IL-12 16 . We have confirmed these observations using our independently derived Th1 lines and clones from a different strain of mice, suggesting that this may be a general phenomenon. Furthermore, and most interestingly, we found that IL-12 greatly enhanced the IL-4 response of activated normal splenic Th1 cells, and that the combined effect of the two cytokines became more prominent with time after activation, when the IL-4 response was waning. We hypothesized that the effect of IL-4 + IL-12 could represent a novel Th1/Th2 cross-regulatory pathway that benefits Th1.
The combined effect of IL-4 and IL-12 may be a novel means of maintaining a Th1 effector response that is waning due to, for example, suboptimal concentrations of Ag. This could represent a way of expanding the pool of Th1 effector cells, to generate a larger number of cells capable of subsequently responding to specific Ag. Several pieces of evidence support this hypothesis. The first is that IL-4 + IL-12 clearly induces proliferation in Th1 clones and in Th1 effector cells derived from normal spleen. The effect is observed at low (pg range) levels of IL-12. In the presence of IL-12, Th1 effector cells generated from normal spleen by Con A stimulation were particularly responsive to low concentrations of IL-4 (Figs. 7,B and 8C). In addition, we showed that the amount of IL-4 produced by in vitro stimulated Th2 effector cells was sufficient to promote this effect (Fig. 8). Con A-activated cells never completely lost their ability to respond to IL-4 in the manner the Ag-stimulated Th1 clones did, but the level of proliferation in response to IL-4 alone was minimal by 6 days poststimulation at all but very high concentrations of IL-4. As expected, since Th2 cells lack functional IL-12R, Th2 clones, as well as newly generated Th2 effector cells, were completely unable to respond to IL-4 and IL-12, demonstrating that this is a purely Th1 phenomenon. A similar unresponsiveness to IL-12 has been observed in human T cells activated with PHA under conditions that would result in the generation of Th2 37 . Therefore, all of our evidence suggests that previously activated Th1 cells in vivo could gain a survival advantage from a bystander effect mediated by Th2-derived IL-4 and APC-derived IL-12.
IL-12 is a known costimulatory molecule for Th1 16 . The proliferative effect of IL-4 + IL-12 on Th1 is similar to that seen in Th2 stimulated by a combination of IL-1 and IL-4. IL-1 is able to prolong the response of previously activated Th2 to IL-4 beyond the normally short period (2–3 days) of poststimulatory IL-4 responsiveness 15 . A distinct difference between the two effects is that while pretreatment of Th2 cells with IL-1 renders them able to respond to IL-4 for some time afterward, IL-12 had to be present together with IL-4 to stimulate Th1 proliferation (Fig. 3). It was not possible to pretreat Th1 cells with IL-12 such that they would subsequently respond to IL-4 (data not shown). The effects of IL-1 on Th2 and IL-12 on Th1 also differ in that IL-1 is not a differentiation factor for Th2 like IL-12 is for Th1 5, 7 . Another possible costimulator of the IL-4 response in Th1 may be IL-18, which is more like IL-1 in that it has similar functions as IL-12, but does not appear to be a Th1 differentiation factor 38, 39 . Cooperativity between IL-18 and IL-4 would likely occur via a pathway distinct from IL-4 and IL-12 since IL-18 can induce IL-2 production from Th1 38 , which would act with the IL-4 to synergistically promote Th1 proliferation 13, 14, 15 . In contrast, we found no evidence that IL-2 was involved in Th1 proliferation in response to IL-4 + IL-12 (Fig. 4). A common thread between the autocrine IL-4 effect on Th2 involving IL-1, and the cross-regulatory IL-4 effect on Th1 involving IL-12 is that both cytokines can be produced by the same cell type, activated macrophages 33, 34, 36 . So, although the B cell is the preferential APC for Th2 8 , these similar pathways are likely to occur, at least to varying degrees, in the same environment.
As mentioned above, proliferation of Th1 in response to IL-4 + IL-12 did not appear to involve IL-2 production. Further attempts were made to define the mechanism by which IL-12 appears to promote IL-4 responsiveness in Th1. Since IL-4 does not appear to alter IL-12R expression 37 , and because IL-12 is known to act on Th1 only as a costimulator, and not as a direct inducer of proliferation, we examined IL-4R expression. However, expression of the IL-4R did not appear to be altered either at the cell surface protein or mRNA levels. The only plausible remaining explanation is that modulation of intracellular signaling pathways may be involved. IL-12 and IL-4 both signal through the Jak-STAT pathway of signaling molecules 40, 41 , and although the signals for IL-4-induced proliferation are not thought to be transmitted through this pathway 42 , there is at least the potential for overlap of signal transduction between the two cytokines. We are currently investigating interactions between intracellular signaling pathways as a possible mechanism for the observed effect of IL-4 + IL-12 on Th1. In any case, it appears that the proliferation-inducing activity of IL-4 plus IL-12 on Th1 occurs through a direct effect. This idea is supported by the fact that IL-4 and IL-12 had to be present together to have a proliferative effect on Th1 (Fig. 3).
IL-4 and IL-12 clearly play vastly different roles in Th cell differentiation, leading to the generation of Th2 and Th1, respectively. Therefore, it may initially seem counterintuitive that the two should have any kind of combined positive effects. For example, probably the most well-studied model in the Th1/Th2 paradigm is infection of mice with Leishmania major, in which mouse strains that develop an IL-12-driven Th1 response are disease resistant, and those that develop an IL-4-driven Th2 response succumb to the infection 43 . However, this type of immune response that is highly polarized in a single direction may be the exception rather than the rule. Successful control of infections may normally require a cooperative response comprised of both Th1 and Th2 components. Therefore, a more prototypical model may be the infection of mice with Plasmodium chabaudi chabaudi, in which a Th1 response is required for control of the initial stages of disease, and a Th2 response is required for final disease resolution 44, 45 . There are similar examples in humans, in which a combined Th1/Th2 response is necessary, such as infection with measles virus 46 . These situations demonstrate the potential benefit for cooperativity between Th1 and Th2, such as the positive effect of IL-2 on Th2 15 , and the effect of IL-4 + IL-12 on Th1 that we have investigated.
In conclusion, the proliferative effect of IL-4 and IL-12 on Th1 appears to represent a novel positive cross-regulatory effect of Th2 and APC. This stimulatory effect could increase numbers of potentially relevant Th1 effector cells that have homed to an area of inflammation. Such an area would logically contain a wide variety of cells and cytokines, both inhibitory and stimulatory, including Th2 cells as an IL-4 source, and activated macrophages or dendritic cells that could provide IL-12 34, 35 . Whether or not the IL-4 + IL-12 effect actually occurs, and the magnitude to which it occurs, would depend on the dynamics of the balance between Th1 and Th2 in each case. Th1/Th2 cross-regulation has many inhibitory and stimulatory facets that operate simultaneously during an immune response 8 . So, although Th2-derived IL-10 can inhibit IL-12 production, its effect can be counteracted by the IL-12-inducing effect of IFN-γ derived from Th1. Therefore, it is plausible to think that adequate concentrations of IL-4 and IL-12 could be present to mediate the combined effect on Th1 during at least some immune responses.
Although Th1 clones that were seemingly resting responded to IL-4 + IL-12, they do probably represent chronically activated effector cells 47 , and spleen-derived Th effector cells also required prior activation to be able to respond to the two cytokines. We therefore favor the notion that IL-4 + IL-12 is a means to enhance or prolong an ongoing Th1 response, rather than a way to activate truly resting cells. This makes the scenario more plausible since it would probably be undesirable to promote indiscriminate proliferation of Th1 cells even if it was occurring within a confined area of inflammation. Naturally, many other positive and negative cross-regulatory pathways would also be operative at the same site 8 . Thus, the combined effect of IL-4 and IL-12 on Th1 appears to add an additional level of complexity to understanding or predicting the dynamics and eventual outcome of an immune response.
We thank Dr. Michael Lotze for the gift of IL-12 and Dr. Joanne Flynn for the gift of anti-IL-12 mAbs.
This work was supported by National Institutes of Health Grant AI31427.