Massive interleukin-12-induced interferon-γ production by interleukin-15-stimulated lamina propria lymphocytes followed by down-regulation of the interleukin-12 receptor

Abstract

The intestinal mucosal immune response must differentiate between harmless foreign antigens and pathogens, a distinction that may depend upon changes in the cytokine milieu. A key cytokine in the adaptive immune response is interleukin-12 (IL-12), secreted by antigen-presenting cells (APC) immediately after encounter with a pathogen. IL-12 is important in the priming and polarization of naïve T cells. Here, we show that IL-12 and IL-15 direct human intestinal lamina propria lymphocytes (LPL) in the absence of T-cell receptor engagement to secrete extremely high amounts of interferon-γ (IFN-γ), greater than with any other stimulus. The functional synergy of IL-12 with IL-15 surprisingly operates independently of signal transducer and activator of transcription 1 (STAT1), STAT3, STAT4, or STAT5 phosphorylation and occurs during transcription. Four-colour immunofluorescence showed that IL-12 receptor β1 is found on the CD4⁺ T cells expressing intracytoplasmic IFN-γ. Importantly, IL-12 receptors β1 and β2 are not up-regulated by IL-12, unlike findings using antigen-specific T cells, and are lost over time. This study demonstrates the early and massive IFN-γ response of LPL to IL-12 and IL-15, providing the tools to deal with a pathogen. The down-regulation of IL-12 receptors may curtail any excess damaging inflammation.

Introduction

The intestinal immune system must differentiate between harmless foreign antigens and pathogens. This critical distinction depends upon changes in the cytokine milieu and is essential for the balance between a targeted appropriate immune response and destructive inflammation. A key cytokine is interleukin-12 (IL-12), which is secreted by antigen-presenting cells (APC) immediately after encounter with a pathogen.¹ IL-12 is required for optimal initiation and probably for maintenance of cell-mediated immunity to infection.² This has been shown, for example, by the persistence of Salmonella in knock-out animal models and in patients with defects in IL-12 secretion.³ IL-12 activates two members of the Janus kinase family, Jak 2 and Tyk 3,⁴ which then phosphorylate the IL-12 receptor (IL-12R), providing docking sites for the transcription factor, signal transducer and activator of transcription 4 (STAT4). In some systems, STAT1, STAT3, and STAT5 are also activated.^5–7

Critical to IL-12 production and responsiveness is IL-15.⁸ IL-15 is constitutively synthesized by many cell types including APC, stromal cells, endothelial cells, and epithelial cells, all found in the intestinal mucosa. Although only a few IL-15-containing APC are present in normal mucosa,⁹ they may support local T-cell activities. IL-15 can serve as a survival factor and growth promoter for antigen-experienced CD4⁺ T cells.¹⁰ IL-2, in contrast, is present transiently with T-cell activation and promotes antigen-induced cell death. Compared with IL-2, IL-15 is more resistant to inhibition by down-regulatory cytokines, permitting its action to be more constant in a mixed cytokine environment.¹¹ IL-2 and IL-15 phosphorylate JAK1 and JAK 3, both functionally coupled to receptors that use the common γ chain (γc).¹² There is a rapid induction of DNA-binding complexes that contain STAT3 and STAT5, both of which are tyrosine phosphorylated.¹³ The functional synergy between IL-12 and IL-2 is associated with a prominent increase in STAT1 and STAT3 serine phosphorylation over that observed with IL-12 or IL-2 alone.¹⁴

The cytokines raised during infection, such as Mycobacterium bovis, set up a positive T helper 1 (Th1) feedback cycle. To begin, activated APC secrete IL-12, an action that may require interferon-γ (IFN-γ) depending upon the pathogen ¹⁵. IL-12 markedly increases IFN-γ production by T cells and natural killer (NK) cells in the context of constitutive IL-15 release.¹⁶ IFN-γ, in turn, up-regulates IL-12 and IL-15 synthesis by the APC.¹⁷ IL-12, IL-15, or IFN-γ can each up-regulate IL-12 receptor (R) expression, furthering the Th1 response.¹⁸ This positive feedback loop participates in an optimal adaptive immune response against pathogens.

There are several possible down-regulatory mechanisms that limit the pro-inflammatory response. For one, IL-12 production is short lived.¹ T cells compete for access to APC and for growth and viability signals. Specialized regulatory T cells control excess expansion. In addition, IFN-γ, which plays a role in the destruction of the pathogen, subsequently regulates the pool size of Th1 cells.¹⁹

Lamina propria lymphocytes (LPL) in the intestinal mucosa contain chronically-activated memory T cells. They respond incompletely to ligation of the CD3/T-cell receptor (TCR) complex,²⁰ but, as shown here, are markedly responsive to IL-12 and IL-15. Cytokine-stimulated, TCR-independent proliferation and IFN-γ production have been described using LPL from Crohn’s disease, while very low values were reported using normal LPL.^9,21–23 With excess IL-12 and IL-15 in Crohn’s disease and up-regulated IL-12R expression, there is heightened IFN-γ release.^9,24,25 IFN-γ, in turn, stimulates LPL to produce IL-12 and IL-15.²⁴ The resulting positive feedback loop is thought to perpetuate the inflammation, both in Crohn’s disease and in animal models.^9,26 It is unclear how this system in the normal host provides an adaptive immune response yet avoids destructive inflammation. A detailed analysis of the normal state is needed in order to know what is abnormal.

Methods

Isolation of LPL and peripheral blood lymphocytes (PBL)

Human jejunal mucosa was obtained after informed consent from individuals undergoing gastric bypass operations for morbid obesity. This study was approved by the Institutional Review Board at UMDNJ-Robert Wood Johnson Medical School. The minced mucosa was treated in a shaking water bath (37°) with 1 mm dithiothreitol, then with 0·75 mm ethylenediamine tetra-acetic acid (both from Sigma-Aldrich, St Louis, MO), and finally collagenase (Worthington Scientific, Malvern, PA) as described previously.²⁷ The cells from the collagenase digestion were isolated by a 60–40% Percoll density gradient (Pharmacia Fine Chemicals, Piscataway, NJ). The epithelial cells were then removed by negative selection using anti-epithelial cell antigen (BER-EP4; Dako, Glostrup, Denmark) followed by magnetic beads coated with goat anti-mouse immunoglobulin G (IgG; Polysciences, Warrington, PA). The CD4 : CD8 ratio of the LPL averaged 2 : 1 to 3 : 1. Any preparation containing <90% CD45⁺ lymphocytes was discarded as epithelial cell contamination reduces lymphocyte viability. For example, contamination with <5% epithelial cells results in about 10% non-viable cells, while contamination with 25% epithelial cells results in 35–40% non-viable cells. PBL were isolated by Ficoll density gradient centrifugation.

Cell culture

Cells (2 × 10⁵/0·2 ml) were cultured in RPMI medium containing 10% fetal calf serum, 1% glutamine and 1% antibiotic–antimycotic solution (complete medium). Some cultures were supplemented with combinations of IL-12 (p35 and p40), IL-15, and IL-18 (10 ng/ml, R&D Systems, Minneapolis, MN). Other cultures had phytohaemagglutinin (PHA, 1 μg/ml; Burroughs-Wellcome, Greenville, NC), staphylococcal enterotoxin B (SEB, 100 ng/ml), anti-CD3 antibody coating the plate (100 pg/ml; R&D Systems), antibodies recognizing CD2 epitopes, T11₂ and T11₃ (1:500 dilution, generous gift of Dr Stuart Schlossman, Dana-Farber Institute, Boston, MA), and CD2 blocking antibodies (R&D Systems).²⁸

Four-colour immunofluorescence

All immunofluorescence staining was performed by incubating cells with fluorochrome- or biotin-conjugated antibodies according to standard protocols. Antibodies recognizing CD4, CD8, IL12 receptor(R) β1, CD28, and ic IFN-γ were used for four-colour immunofluorescence. Single-colour staining was performed with IL-12 receptor (R) β1, IL-12Rβ2, IL-18Rα, and CD25 (all antibodies from R&D Systems). Immunofluorescence was analysed by flow cytometry (Beckman Coulter FC500; Beckman Coulter Inc., Fullerton, CA). Relative fluorescence intensity (RFI), determined by flow cytometry, is the fold-increase in fluorescence compared to an IgG isotype control. For intracytoplasmic markers, the cells were cultured with PHA (1 μg/ml), phorbol 12-myristate 13-acetate (PMA, 50 ng/ml), and brefeldin A (5 μg/ml) (all from Sigma-Aldrich) for 18 hr, then permeabilized with Cytofix/Cytoperm (BD Biosciences, San Jose, CA), and finally stained for IFN-γ expression.

Apoptosis and necrosis were detected by Annexin conjugated to fluorescein isothiocyanate (FITC) and propidium iodide (Molecular Probes, Eugene, OR), respectively, and read by flow cytometry.

Cell functions

Proliferation was measured by tritiated [³H]thymidine incorporation.²⁷ Cytokine levels (IFN-γ, tumour necrosis factor-α (TNF-α), IL-2, IL-4, IL-5, IL-10, and IL-12) were measured by enzyme-linked immunosorbent assay (ELISA; R&D Systems). Dose–response curves showed that IFN-γ was maximally produced starting at 5 ng/ml of IL-12 and IL-15 in most experiments, so 10 ng/ml was used for this study.

IFN-γ mRNA levels were determined by a quantitation kit (R&D Systems). The RNA preparation was obtained using Trizol Reagent (Invitrogen, Carlsbad, CA). Hybridization of cellular RNA was performed with gene-specific biotin-labelled capture oligonucleotide probes and digoxigenin-labelled detection probes in a microplate. The hybridization products were then transferred to a streptavidin-coated microplate, and the RNA/probe hybrid was captured. Following a wash to remove unbound conjugate, colour was developed, and its quantitation was proportional to the amount of gene-specific mRNA in the original samples.

Detection of STAT phosphorylation

LPL were stimulated for 2, 4, 8, or 18 hr with complete medium supplemented with IL-12, IL-15, or both combined (10 ng/ml) (R&D Systems). Nuclear extracts were isolated using a Nuclear Extract Kit (Active Motif, Carlsbad, CA) and samples were analysed by 10% sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE). They were then transferred to nitrocellulose paper, and probed with antibodies against phosphorylated STAT proteins (1 : 200 dilution; R&D Systems).

Statistics

Statistical analysis was performed using the Mann–Whitney Wilcoxon rank sum test or the Student’s t-test depending upon the distribution of the data. For more than two data sets, analysis of variance (anova) was used with the Tukey test to evaluate pairs within the group.

Results

IL-12Rβ1 expression found preferentially on IFN-γ-producing CD4⁺ T cells

LPL had low-level constitutive expression of IL-12Rβ1 (RFI averaging 1·3), suggesting that they were armed for an immediate response to IL-12. Two-colour immunofluorescence revealed that IFN-γ was found in a larger number of CD4⁺ than CD8⁺ T cells (32 ± 8% versus 15 ± 7%, respectively) and at a higher density (mean RFI 1·3-fold greater for CD4⁺ than CD8⁺ T cells) (Fig. 1). Three- and four-colour immunofluorescence revealed that, within the CD4 subset, IFN-γ was found equally in CD28⁺ and CD28⁻ T cells in one experiment and in greater amounts by the CD28⁺ T cells in two others. Importantly, when expression of IL-12Rβ1 on CD4⁺ T cells was compared to intracytoplasmic IFN-γ, the IL-12Rβ1⁺ subset contained more IFN-γ than did the IL-12Rβ1⁻ cells (RFI of 3·8 and 2·1, respectively). IFN-γ, then, was concentrated in the LPL CD4⁺ T cells expressing IL-12Rβ1. No IL-12Rβ2 could be detected at 0, 2, 4, or 18 hr after stimulation with IL-15 and IL-12 (Fig. 2a).

Figure 1.

LPL were cultured for 18 hr with PHA (1 μg/ml), PMA (50 ng/ml), and brefeldin A (5 μg/ml), and four-colour immunofluorescence performed using CD4, CD8, IL-12Rβ1, and intracytoplasmic IFN-γ. The numbers represent the percentage of events in that quadrant. The numbers of experiments are reported in the text.

Figure 2.

(a) Expression of IL-12Rβ1 and IL-12Rβ2 is shown on fresh LPL or LPL after culture for 3 days with medium, IL-12, IL-15, or IL-12 with IL-15 (all at 10 ng/ml) (n= 5). (b) Expression of IL-12Rβ1 and IL-12Rβ2 is shown on LPL after culture for 7 days with IL15 in the presence or absence of IL-12 (both at 10 ng/ml) (n= 4). (c) Expression of the surface receptors listed on the x-axis was measured after culture for 3 days with combinations of the cytokine stimuli (all at 10 ng/ml). Relative fold increase in fluorescence intensity (RFI) was determined by flow cytometry. Values were compared using anova with the Tukey test to analyse pairs within the groups (n= 5).

IL-12 stimulates IFN-γ production without STAT phosphorylation

To determine whether the constitutive expression of IL-12Rβ1 translated to an early response to IL-12, LPL were cultured with IL-12 alone or combined with IL-18 or IL-15, and IFN-γ mRNA was quantitated (Fig. 3). Transcripts were present in freshly-isolated LPL (150 amol/ml for 1 × 10⁶ cells), indicating low-level constitutive production of IFN-γ. IL-12 alone or IL-12 with IL-18 did not affect IFN-γ mRNA levels at either 8 or 18 hr. By 8 hr, IL-15 increased IFN-γ mRNA to levels above that with medium alone although secretion of protein was minimal (<10 pg/ml). At 18 hr, IL-12 significantly increased IL-15-induced transcription of IFN-γ. Together, these data illustrated an early response of LPL to IL-15 with detectable changes in IFN-γ mRNA by 8 hr. The up-regulatory response with IL-12 lagged behind with transcripts rising at 18 hr.

Figure 3.

LPL were cultured for 8 or 18 hr with the stimuli listed (all at 10 ng/ml), then tested for IFN-γ mRNA content in 1 × 10⁶ cells. Values were compared using anova with the Tukey test to analyse pairs within the groups (n= 6).

LPL were again stimulated with combinations of IL-12, IL-18, and IL-15. IFN-γ protein production was measured on day 3 (Fig. 4a). Unlike the mRNA data, there was a modest rise in IFN-γ protein with either IL-12 alone or IL-12 combined with IL-18. A similar modest increase occurred with IL-15 alone. In striking contrast, IFN-γ generation was markedly up-regulated when IL-12 was combined with IL-15. This tremendous amount of IFN-γ, averaging 625 ng/ml, exceeded that induced by PHA, SEB, antibody ligation of the CD3/TCR complex, or activating antibodies of the CD2 receptor (called anti-T11₂ and T11₃) (Fig. 4b). Antibody triggering of CD2, the most potent known stimulus of LPL, induced eightfold less IFN-γ than did the combination of IL-12 with IL-15.

Figure 4.

LPL were cultured for 3 days and the resulting IFN-γ or IL-10 production determined by ELISA. (a) IFN-γ production by LPL activated with cytokines (all at 10 ng/ml) (n= 6); (b) IFN-γ production by LPL activated with PHA (1 μg/ml), SEB (100 ng/ml), anti-CD3 antibody (50 ng/ml), and anti-T11₂ and T11₃ antibodies (1 : 500 dilution) (n= 6); and (c) IL-10 production by LPL activated with cytokines (all at 10 ng/ml) (n= 6). Values were compared using anova with the Tukey test to analyse pairs within the groups.

To evaluate STAT phosphorylation, LPL were cultured for 2, 4, 8, or 18 hr with medium alone or with IL-15, IL-12, or both cytokines combined. STAT phosphorylation was measured by Western blot assay. Surprisingly, there was no signal whether or not IL-12 supplemented IL-15 (Fig. 5). This differed from the measurable STAT phosphorylation by CD3-activated PBL stimulated with IL-12 and IL-15.

Figure 5.

(a) LPL were cultured for 18 hr with medium alone, IL-12, IL-15, or both combined (all at 10 ng/ml) and phosphorylation of STAT1, STAT3, STAT4, and STAT5 determined by Western blot assay (n= 3). All results were negative including the experiment with IL-15 and IL-12 shown below. (b) For a positive control, PBL were cultured for 18 hr with anti-CD3 antibody, IL-15, and IL-12 and phosphorylation demonstrated.

IL-12 has a lower up-regulatory effect on proliferation than on IFN-γ production

Proliferation was measured because it contributes to the inflammatory cell infiltrate that occurs with infection (Fig. 6). IL-15 induced more proliferation than medium alone. IL-12 increased the IL-15-induced response but not to the same degree as seen with IFN-γ production.

Figure 6.

LPL were cultured for 3 days with the combinations of cytokines listed (all at 10 ng/ml). Proliferation was measured by [³H]thymidine incorporation. Values were compared using anova with the Tukey test to analyse pairs within the groups (n= 6).

Because LPL are prone to apoptosis, IL-12 may improve survival and thereby increase IFN-γ production and proliferation. However, after a 4- or 18-hr incubation with IL-12 and/or IL-15, the numbers of apoptotic and necrotic cells were the same as without stimulation (Fig. 7).

Figure 7.

LPL were cultured for 18 hr with the cytokines listed (all at 10 ng/ml). Apoptosis and necrosis were determined by Annexin and PI staining of ungated samples. The numbers of cells in N2 and N4 combined were consistently <10% of the whole (n= 3).

IL-12 induces a Th0 cytokine profile

To determine whether IL-15 with or without IL-12 induces a Th1- or Th2-pattern of cytokine release, the production of TNF-α, IL-2, IL-4, and IL-5 was measured by ELISA in medium after culturing LPL for 3 days. Neither the Th1 cytokines (TNF-α and IL-2) nor the Th2 cytokines (IL-4 and IL-5) could be detected (all <25 pg/ml), arguing against a strict Th1 or Th2 profile. IL-12 production was also undetectable in IL-15-stimulated cultures. IL-12 caused a rise in IL-10 production, but the amounts were consistently over 100-fold less than those of IFN-γ (Fig. 4c). These experiments show that IL-15 and IL-12 selectively induce IFN-γ and, to a much lesser extent, IL-10, forming a Th0 cytokine profile.

Down-regulation of IL-12Rβ1 with time

IL-12 and IL-15 have been shown to increase expression of IL-12 receptors on activated T cells.^18,24 To determine whether this is true for LPL, they were grown for 3 days with combinations of IL12, IL-15, and IL18, then stained for IL-12Rβ1, IL-18Rα, or CD25 (Fig. 2). To assure that free IL-12 or IL-18 was not masking receptor expression, the cells were exposed to a 100 nm glycine HCl solution (pH 3) to free any receptor-bound cytokines just before staining. Despite the increase in IL-18Rα with IL-15 (and especially with IL-12).²⁹ and the up-regulation of CD25 with IL-15, expression of IL-12Rβ1 remained the same regardless of the cytokine stimulus.

LPL were then cultured with IL-15 with or without IL-12 for 7 days. Only cultures containing IL-15 survived. There was no detectable IL-12Rβ1 or IL12Rβ2 expression whether or not IL-12 was used with IL-15 (Fig. 2) (RFI < 1·1).

Discussion

The axis formed between IL-12 and IFN-γ is essential for protective immunity against pathogens. IL-12 is released upon first encounter with pathogen.¹⁵ Along with IL-15, it up-regulates IFN-γ production¹⁶ which, in turn, stimulates IL-12 and IL-15 production by the APC.¹⁷ This positive feedback loop has been described with the host response to pathogens as well as with Crohn’s disease,^9,26 indicating that it can be a normal or abnormal response. There may be some overlap between the two as demonstrated by a destructive inflammatory response that occurs as the host reacts to the pathogen, as in infectious colitis. However, the response is down-regulated once the pathogen is eliminated. The mechanisms by which this positive feedback loop is down-regulated is unclear. The present study shows that normal LPL constitutively express IL-12Rβ1, permitting an early response to IL-12. This action is down-regulated by a loss of IL-12 receptor expression.

In the present study, the response of LPL to IL-15 precedes that to IL-12 in the cytokine cascade, as shown by mRNA quantitation. This agrees with previous studies showing that IL-15 modulates the production of and responsiveness to IL-12 and that both contribute to an optimal response to infection.⁸ Kinetics in the present study indicated that IFN-γ mRNA rose by 8 hr with IL-15 and 18 hr with superimposed IL-12. This early response to IL-12 was aided by constitutive expression of IL-12Rβ1. Although not measurable by flow cytometry here, IL-12Rβ2 must be expressed transiently or at low levels on the LPL in order to get an IL-12 response. IFN-γ accumulated in the medium over time so that by 72 hr, the levels exceeded that with mitogen, superantigen, or ligation of CD2 or CD3/TCR, stressing the tremendous effects of IL-12 and IL-15 on LPL. In agreement with previous studies, triggering the CD2 receptor resulted in greater IFN-γ production than triggering the CD3/TCR complex.³⁰

In order to get such vigorous responses, epithelial cells in the LPL preparation had to be removed after the density gradient centrifugation, a step not commonly used in other studies.^9,21–23 Because LPL viability declines with epithelial cell contamination, this may explain the low responses of normal LPL found previously.

For naïve T cells to respond to IL-12, they must first be activated by APC through TCR ligation resulting in their expression of both the β1 and β2 chains of IL-12R. Here, LPL constitutively display IL-12Rβ1 and demonstrate marked IL-12 reactivity, suggesting that they have been previously-activated. IL-12 coupled with IL-15 caused a spike in IFN-γ and IL-10 production and a lesser increase in proliferation. This difference was partly because of the placement of IL-12Rβ1 on the IFN-γ-producing CD4⁺ T cells. Cytokine-induced proliferation, in contrast, has been shown previously to be a property of both the CD4⁺ and CD8⁺ T-cell subsets.³¹

In the present study, the marked IL-12 up-regulation of IFN-γ production occurred at the protein and mRNA levels. IL-12 surprisingly did not phosphorylate STAT proteins, an event that occurs before transcription. This differs from most other described IL-12-induced processes^5–7 which involve phosphorylation of STAT proteins and alterations in the transcriptional rate of the target proteins.

The striking effect of IL-12 could be the result of synergy with CD2 triggering.³² A previous study showed that LPL, unlike naïve T cells, constitutively display epitopes of CD2 that are associated with activation, particularly T11₃, and this marker is functional as demonstrated by marked increases in proliferation and IFN-γ production with anti-T11₃ antibody.²⁷ However, blocking CD2 with specific antibodies here did not reduce the IFN-γ release with IL-2 or IL-15, whether or not IL-12 was present (data not shown).

IL-12 can also have an enhanced response in conjunction with IL-18. These two cytokines synergistically increase IFN-γ synthesis by activating different portions of the IFN-γ promoter and by reciprocally up-regulating the other’s receptor.^29,33,34 Using antigen-specific T cells, IL-12 and IL-18 increase IL-12Rβ1 and β2, while IL-12 increases IL-18Rα and β.^29,34 Although IL-18 co-operated here with IL-12 to increase IFN-γ and IL-10 release by LPL and IL-12 promoted IL-18R expression, IL-18 did not affect IL-15-triggered responses, nor did it alter IL-12Rβ1.

Surprisingly, IL-12 did not induce a classic Th1 cytokine profile as IL-2 and TNF-α were not raised. In addition, the Th2 cytokines, IL-4 and IL-5, were not affected. Rather, a Th1 cytokine, IFN-γ, and a Th2 cytokine, IL-10, were produced, comprising a Th0 profile. IL-10 synthesis was measured as LPL have been shown to be high producers of this cytokine³⁵ and as IL-10 often reverses some of the effects of IFN-γ and down-regulates the IL-12 response. For example, IFN-γ can disrupt the epithelial cell barrier, while IL-10 reverses this effect.³⁶ Although IL-10 was 100-fold less concentrated than IFN-γ, it is unknown whether this amount will significantly alter the actions of IFN-γ. Ongoing experiments are addressing the question as to whether IL-10 could be regulating IL-12-induced IFN-γ production.

In the present study, LPL did not display more IL-12Rβ1 with IL-12, although IL-12 up-regulation of its own receptor has been shown in many other systems.¹⁸ In fact, by day 7, no detectable IL-12Rβ1 was found. This may be a regulatory mechanism to arrest the IL-12-triggered events in the mucosa. In contrast, up-regulated IL-12R in Crohn’s disease may explain the persistent destructive inflammation in this disease.

In summary, IL-12 causes marked IL-15-mediated up-regulation of IFNγ production by LPL, greater than that induced by other known stimuli, caused by placement of IL-12R on the CD4⁺ IFN-γ-producing cells. This response, in the absence of TCR ligation, suggests that LPL respond quickly and markedly to their cytokine milieu rather than to cognate antigen. This IL-12-induced up-regulation occurs at the transcriptional level surprisingly without phosphorylation of the STAT proteins. Unlike other systems, IL-12R is not up-regulated by IL-12 or IL-18, and the IL-12 effect is lost over time. This is a novel mechanism by which the IL-12 response is controlled, thereby appropriately limiting the inflammatory response.