Allergen-stimulated interleukin-4 and interferon-gamma production in primary culture: responses of subjects with allergic rhinitis and normal controls.

The balance of interleukin-4 (IL-4) to interferon-gamma (IFN-gamma) production that is induced following exposure to common environmental antigens is believed to be instrumental in determining whether hypersensitivity or clinical unresponsiveness results to that antigen. To date, evaluation of cytokine (protein) production has been based predominantly on allergen-reactive CD4 T-cell clones or activation of fresh unselected peripheral blood mononuclear cell (PBMC) populations with non-physiological stimuli such as phorbal myristate acetate (PMA) and calcium ionophore, phytohaemagglutinin (PHA), anti-CD3 or anti-CD2/anti-CD28 monoclonal antibodies (mAb). Here, ultrasensitive IL-4 and IFN-gamma assays were optimized to allow direct analysis of antigen-stimulated cytokine production by fresh human PBMC. Primary cultures of cells from grass pollen-sensitive allergic rhinitis subjects and non-atopic controls were stimulated using a range of grass pollen allergen concentrations in the absence of exogenous cytokines or polyclonal activators. The majority of subjects (45 of 52) exhibited chloroquine-sensitive, CD4-dependent cytokine production in allergen-stimulated, short-term primary culture. Median IL-4 production was substantially greater among atopics (13.0 pg/ml versus < 1 pg/ml, Mann-Whitney U test, P < 0.0000001) and IFN-gamma was lower (P = 0.008), providing direct evidence for an imbalance in both IL-4 and IFN-gamma production among circulating, pollen-reactive cells in individuals with seasonal allergic rhinitis. The distinction in the allergen-driven cytokine responses elicited from normal and atopic donors was underscored by examination of the ratios of IFN-gamma:IL-4 synthesis. Non-atopic individuals exhibited intense IFN-gamma dominance of the T-cell response, in marked contrast to that observed among grass pollen-sensitive individuals (median IFN-gamma:IL-4 ratios of 14.0 versus 0.096, P = 0.000002). The observation that essentially all individuals produced IFN-gamma (+/- IL-4) following antigen stimulation in vitro argues that the most relevant consideration in determining susceptibility to immediate hypersensitivity versus clinical tolerance to environmental allergens is not a genetically defined capacity to recognize the antigen (i.e. if allergen-reactive T cells are present in that individual) but the nature of the cytokine response.     

The effect of interferon-gamma, interleukin-4 and immunoglobulin receptor cross-linking of monocytes on allergen-specific T-cell response.

Modulation of the proliferative responses of an allergen-specific human Th2 cell line by cytokine-treated monocytes was examined. The response of this cell line to the specific allergen, Amb a V (from short ragweed pollen), increased following the addition of interleukin-1 beta (IL-1 beta). However, in the presence of exogenous interferon-gamma (IFN-gamma), there was greater than 40% reduction in the responsiveness of these T cells. The addition of IL-1 beta did not reverse the inhibitory effect of IFN-gamma. To determine the primary target cell type for IFN-gamma, autologous monocytes were pretreated with IL-4, IFN-gamma, or medium alone, and used as antigen-presenting cells (APC). We showed that the responses of T cells to Amb a V were significantly down-regulated in the presence of autologous monocytes pretreated with IFN-gamma, but not for monocytes pretreated with IL-4. Similar inhibitory effect of IFN-gamma was confirmed using a human T-cell line specific for a ragweed allergen, Amb a I, and a human T-cell clone raised against ragweed extract. Cross-linking of CD23 (Fc epsilon RII) on monocytes pretreated with IFN-gamma increased this inhibitory effect in an additive fashion, but, in the absence of IFN-gamma treatment, such cross-linking had no effect. These inhibitory effects were not due to alterations in the surface expression of HLA-DR on the monocytes, and the addition of exogenous IL-1 beta was unable to reverse these effects. In similar experiments, cross-linking of CD64 (Fc gamma R) on monocytes showed no significant effects. In conclusion, IFN-gamma is important in regulating the function of monocytes involved in Th2 cell responses to allergens. IL-4 treatment, as well as cross-linking of FcR of monocytes, have no direct effect on such response.     

Quantitative T-cell interferon-gamma responses to Mycobacterium tuberculosis-specific antigens in active and latent tuberculosis

The objective was to compare the quantitative T-cell responses measured by the commercial interferon-gamma (IFNγ) release assays (IGRAs) in active and latent tuberculosis (TB) states. T-cell responses of culture-proven TB cases were compared with those of contacts with positive IGRA results and tuberculin skin tests ≥ 15 mm. T-SPOT.TB results in 270 active TB cases and 183 community contacts showed the median spot-forming cells (SFCs) above negative control/2.5 × 10⁵ peripheral blood mononuclear cells to be 27 (−1 to 203) vs 10 (−2 to 174) in response to ESAT-6 (p < 0.001); and 37 (0 to 293) vs 13 (0 to 225) to CFP-10 (p < 0.001). The median IFNγ levels (antigen minus nil control) as measured by QuantiFERON-TB Gold In-tube in 270 cases and 142 contacts in congregate settings was 2.3 IU/ml (−0.58 to 31.44) vs 1.7 IU/ml (0.35 to 26.51, p = 0.98). Quantitative T-cell responses as measured by the T-SPOT.TB may indicate mycobacterial burden and disease activity, but cannot be used to discriminate active from latent TB.     

Role of interferon-gamma in T-cell responses to Semliki Forest virus-infected murine brain cells

Primary brain cell cultures prepared from newborn C3H mice were infected with Semliki Forest virus (SFV) or treated with a beta-propiolactone-inactivated preparation of SFV (BPL-SFV). The effects of recombinant interferon-gamma (IFN-gamma) treatment on SFV replication, SFV antigen display, major histocompatibility complex (MHC) class I and class II antigen expression, susceptibility to lysis by SFV-specific cytotoxic T lymphocytes (CTL) and the ability to stimulate SFV-specific T lymphocytes to release IFN-gamma were determined. The IFN-gamma treatment prevented replication of SFV, as determined by incorporation of [3H]uridine into SFV-RNA, and reduced expression of SFV antigens on the cell surface, as determined by lysis with antibody and complement or indirect immunofluorescence. BPL-SFV-treated brain cells expressed no SFV antigen detectable by lysis with antibody and complement or indirect immunofluorescence. IFN-gamma increased expression of MHC class I and class II antigens, measured by indirect immunofluorescence, susceptibility to killing by alloreactive T-cell lines and ability to stimulate an allogeneic mixed lymphocyte reaction (MLR). Brain cells infected with SFV or treated with BPL-SFV were susceptible to killing by the CTL. The killing was MHC restricted and neither uninfected nor untreated cells were killed. IFN-gamma treatment prior to SFV infection or BPL-SFV treatment resulted in an augmentation of lysis by the CTL, indicating that even where SFV antigen expression is reduced or present at very low levels, in the context of enhanced MHC class I expression cells remain susceptible to CTL killing. Brain cells treated with BPL-SFV stimulated SFV-specific T cells to release IFN-gamma. Pretreatment of brain cells with IFN-alpha beta or IFN-gamma prior to BPL-SFV treatment markedly increased the ability of the cells to stimulate the SFV-specific T cells to release IFN-gamma. Release of IFN-gamma was MHC restricted and brain cells untreated with BPL-SFV did not stimulate IFN-gamma release. IFN-gamma released by T cells stimulated with BPL-SFV-treated brain cells increased class II MHC expression by brain cells as assessed by indirect immunofluorescence.     

Myocytes respond to both interleukin-4 and interferon-gamma: cytokine responsiveness with the potential to influence the severity and course of experimental myasthenia gravis

Messenger RNA that encodes for interleukin-15 (IL-15) has been reported to be constitutively expressed in skeletal muscle, although the protein product is not generally observed. Furthermore, interferon-gamma (IFN-gamma) has been reported to exacerbate symptoms of experimental myasthenia gravis (EAMG). Therefore, since IL-15 is an activator of IFN-gamma-producing cells, the hypothesis that drove the study reported below proposes that muscle is not a passive participant in the development of disease symptoms in EAMG and, in fact, plays a very important active role by producing immunomodulating factors that can influence the eventual immunopathological impact of the immune system on muscle. Tests of this hypothesis, made using a monoclonal skeletal myocyte line from the Lewis rat, have indicated that myocytes produce IL-15 protein following exposure to interleukin-4 (IL-4), an interesting paradox in light of the usual anti-inflammatory role played by IL-4. Furthermore, the level of IL-15 also can be regulated by IFN-gamma itself. Although yet to be confirmed in vivo, IFN-gamma has been shown to be capable of activating cultured myocytes in a variety of ways that could influence the ongoing autoimmune response associated with EAMG.     

Gangliosides interact with interleukin-4 and inhibit interleukin-4-stimulated helper T-cell proliferation.

Gangliosides are potent immunosuppressive agents in vitro, and gangliosides shed from tumours in vivo may play an important role in the escape of tumours from immune destruction. We have investigated the effect of gangliosides on interleukin-4 (IL-4)-mediated processes in the murine helper T-cell line HT-2. Various gangliosides inhibited IL-4-stimulated DNA synthesis in HT-2 with IC50 values in the range 26-60 micrograms/ml. However, the proliferation of four lymphokine-independent cell lines was unaffected by 500 micrograms/ml gangliosides, as was the IL-1-stimulated secretion of IL-2 by EL-4 NOB-1 cells. Gangliosides were highly effective inhibitors when added to G0-G1-synchronized HT-2 cells during the first 6 hr after IL-4 stimulation, indicating that they act early in the IL-4 signalling pathway. High levels of exogenous IL-4 completely reversed inhibition of proliferation by gangliosides, which suggests that gangliosides compete with cellular IL-4 receptors for available lymphokine. Receptor-binding experiments confirmed that gangliosides blocked binding of [125I]IL-4 to receptors on intact HT-2 cells in a dose-dependent fashion. Gel-filtration fast protein liquid chromatography (FPLC) demonstrated that [125I]IL-4 co-eluted with ganglioside micelles after co-incubation before chromatography, and an overlay technique showed that IL-4 bound efficiently to gangliosides on thin-layer chromatography plates. Taken together, these results indicate that gangliosides act as potent suppressors of IL-4-dependent processes in lymphocytes, and that their mechanism of action involves direct interaction with IL-4, thus preventing IL-4 binding to high-affinity IL-4 receptors. This information helps to explain the diverse immunosuppressive actions reported for gangliosides, both in vitro and in vivo.     

CD4(+) T cell clones producing both interferon-gamma and interleukin-10 predominate in bronchoalveolar lavages of active pulmonary tuberculosis patients.

The pattern of cytokine production in T cell clones derived from bronchoalveolar lavages (BAL) of active pulmonary tuberculosis (TB) patients was analyzed in clones obtained by limiting dilution procedures which expand with high efficiency either total T lymphocytes, independently of their antigen-recognition specificity, or Mycobacterium tuberculosis-specific T cells. BAL-derived clones, representative of CD4(+) cells from five patients with active TB, produced significantly higher amounts of IFN-gamma than BAL-derived CD4(+) clones from three inactive TB donors or four controls (with unrelated, noninfectious pathology). Average IL-4 and IL-10 production did not differ significantly in the three groups. Although these data suggest a predominant Th1 response to M. tuberculosis infection in the lungs, the majority of BAL-derived CD4(+) clones produced both IFN-gamma and IL-10 and the percentage of clones with this pattern of cytokine production was significantly higher in clones derived from BAL of active TB patients than from controls. Only rare clones derived from peripheral blood (PB)-derived CD45RO(+) CD4(+) T cells of both patients (nine cases) and controls (four cases) produced both IFN-gamma and IL-10; instead, the IL-10-producing clones derived from PB T cells most often also produced IL-4, displaying a typical Th2 phenotype. Higher average amounts of IFN-gamma and IL-10 were produced by BAL-derived CD8(+) clones of four active TB patients than of four controls, although the frequency of CD8(+) clones producing both IFN-gamma and IL-10 was lower than that of CD4(+) clones. The M. tuberculosis-specific BAL-derived T cell clones from three active TB patients were almost exclusively CD4(+) and produced consistently high levels of IFN-gamma often in association with IL-10, but very rarely with IL-4. Unlike the BAL-derived clones, the M. tuberculosis-specific clones derived from PB CD45RO(+) CD4(+) T cells of three different active TB patients and two healthy donors showed large individual variability in cytokine production as well as in the proportion of CD4(+), CD8(+), or TCR gamma/delta(+) clones. These results indicate the predominance of CD4(+) T cells producing both the proinflammatory cytokine IFN-gamma and the anti-inflammatory cytokine IL-10 in BAL of patients with active TB.     

Effects of interleukin-12 and interleukin-15 on measles-specific T-cell responses in vaccinated infants

Understanding the infant host response to measles vaccination is important because of their increased mortality from measles and the need to provide effective protection during the first year of life. Measles-specific T and B-cell responses are lower in infants after measles vaccination than in adults. To define potential mechanisms, we investigated age-related differences in measles-specific T-cell proliferation, CD40-L expression, and IFN-gamma production after measles immunization, and the effects of rhIL-12 and rhIL-15 on these responses. Measles-specific T-cell proliferation and mean IFN-gamma release from infant PBMCs were significantly lower when compared with responses of vaccinated children and adults. Infant responses increased to ranges observed in children and adults when both rhIL-12 and rhIL-15 were added to PBMC cultures. Furthermore, a significant rise in T-cell proliferation and IFN-gamma release was observed when infant PBMCs were stimulated with measles antigen in the presence of rhIL-12 and rhIL-15 compared to measles antigen alone. CD40-L expression by infant and adult T cells stimulated with measles antigen was comparable, but fewer infant CD40-L(+) T cells expressed IFN-gamma. These observations suggest that lower measles-specific T-cell immune responses elicited by measles vaccine in infants may be due to diminished levels of key cytokines.     

Contrasting effects of interferon-gamma and interleukin-4 on the interleukin-6 activity of stimulated human monocytes.

Stimulated human monocytes/macrophages are a source of interleukin-6 (IL-6), which is a likely mediator involved in immune and inflammatory reactions. The means to control production of IL-6 by these cells could therefore have therapeutic applications. We report here, for lipopolysaccharide (LPS)-stimulated human monocytes in vitro, that the lymphokine, interferon-gamma (IFN-gamma) (100 U/ml), enhanced the level of IL-6 activity, whereas another lymphokine, interleukin-4 (IL-4) (greater than or equal to 0.1 U/ml; greater than or equal to 1.2 x 10(-11) M), suppressed it. The effects of the two lymphokines were manifested at the level of mRNA. The action of the IL-4 was similar to that of the glucocorticoid, dexamethasone, but observed at a lower molar concentration. Such regulation of monocyte IL-6 activity is similar to that found previously for interleukin-1 (IL-1) and tumour necrosis factor-alpha (TNF-alpha) synthesis.     

Immune responses incongenitally thymus-less mice. I. Absence of response to oxazolone

Cell proliferation in regional lymph nodes of mice sensitized with oxazolone was measured by uptake of ¹²⁵IUdR. Experimental factors affecting incorporation of IUdR into lymph nodes are described. The delayed hypersensitivity response was assessed by measuring the increase in ear thickness after subsequent challenge with oxazolone. Unlike normal mice, thymus-less `nude' mice showed no detectable response to oxazolone. Implantation of an allogeneic neonatal thymus, either subcutaneously or under the kidney capsule, adoptively conferred on `nude' mice the ability to mount a lymphoproliferative response.     

Contact sensitivity to oxazolone in the chicken: evidence for Arthus type hypersensitivity of the cutaneous reaction.

Cutaneous hypersensitivity reaction can be induced in chickens by skin painting with oxazolone, 33 mg/Kg of body weight (KBW). The B cell contribution to the generation of the cutaneous reaction has been a matter of controversy. In an attempt to characterize this reaction we placed special interest on the possibility that the nature of this reaction could be Arthus type hypersensitivity. From the kinetics study on the cutaneous hypersensitivity after challenge with oxazolonated egg-albumin (EA-OX) it was excluded that the nature of this reaction could be delayed type hypersensitivity. Immune sera transfer experiments demonstrated that the cutaneous reaction was antibody dependent. Serum anti-oxazolone antibody titers in sensitized chickens were assayed by antiglobulin haemagglutination, using oxazolone coupled sheep erythrocytes (OX-SRBC). High titres of IgG were found in contact sensitized chickens. Furthermore this cutaneous reaction was characterized by neutrophils, inflammatory edema, rare thrombotic occlusion of small venules and on absence of monocytes. The utilization of complete Freunds' adjuvant (CFA) given at sensitization demonstrated that CFA enhanced oxazolone antibodies in the sera of immunized chickens without a correlated increase in the intensity of the cutaneous reaction to EA-OX. Animals sensitized to oxazolone (33 mg/KBW) without CFA and challenged intravenously seven days later with oxazolone coupled to autologous chicken red blood cells (OX-CRBC) died from anaphylactic shock; instead animals with the same treatment but with CFA given at sensitization did not die from anaphylactic shock. Taken collectively it was concluded that the cutaneous reaction to oxazolone in the chicken can be categorized as Arthus hypersensitivity. The relationship between cutaneous Arthus reaction and anaphylactic shock in chickens sensitized to oxazolone is discussed.     

Inducible interleukin-4-secreting cells provoked in mice during chemical sensitization.

It has been demonstrated previously that chemical contact and respiratory allergens differ with respect to the quality of immune responses they will provoke in mice. Trimellitic anhydride (TMA), a human respiratory allergen, induces in mice responses consistent with the preferential activation of Th2-type cells, resulting in the production of IgE anti-hapten antibody and an increase in the serum concentration of IgE. In contrast, oxazolone (OX), a potent contact allergen considered not to cause respiratory hypersensitivity, induces instead Th1-type responses in mice characterized by vigorous IgG2a antibody production and a failure to elicit IgE. In the present study we have extended these investigations and have examined the capacity of these chemicals to stimulate inducible interleukin-4 (IL-4) production by draining lymph node cells (LNC). IL-4 was measured in the supernatants of draining LNC cultured for various periods in the presence or absence of concanavalin A (Con A). Following primary topical exposure to the chemical allergens, Con A-stimulated LNC from OX-treated mice secreted significantly more IL-4 than did LNC from mice exposed to trimellitic anhydride (TMA). A different pattern of IL-4 secretion was observed following culture with Con A of LNC prepared from lymph nodes draining the sites of secondary exposure to these chemicals. In this case significantly higher concentrations of IL-4 were produced by TMA-treated mice. Detectable levels of IL-4 (> 300 pg/ml) were not found following culture of draining LNC from sensitized mice in the absence of Con A or following culture of LNC from naive mice with or without Con A. These data demonstrate that chemical allergens of different types stimulate discrete and changing patterns of inducible IL-4 synthesis consistent with the selective activation of Th-cell subpopulations.     

Venom immunotherapy modulates interleukin-4 and interferon-gamma messenger RNA expression of peripheral T lymphocytes.

The mechanism by which specific immunotherapy exerts its beneficial effect remains unclear. In order to evaluate the influence of venom immunotherapy on the T-cell cytokine pattern of allergic reactions, we studied interleukin-4 (IL-4) and interferon-gamma (IFN-gamma) mRNA expression of peripheral T lymphocytes from 12 patients undergoing rush venom desensitization, before treatment at Day 0 (D0), at Day 15 (D15) and Day 90 (D90) after treatment, and from seven controls. Antigen-specific T-cell proliferation was also determined. Cytokine mRNA expression was evaluated using in situ hybridization, 24 hr after culture of peripheral T cells with medium, venom, or an unrelated allergen. Allergen-induced T-cell proliferation decreased at D15 and D90 of rush immunotherapy (P < or = 0.02). In venom-stimulated cultures of the patient group, there was a decrease in IL-4 mRNA-positive cells at D15 and D90 (P < or = 0.001). Before desensitization, IFN-gamma mRNA expression was lower in patients than in controls and did not increase after in vitro allergen stimulation. In contrast, after immunotherapy, spontaneous IFN-gamma mRNA expression increased, but only at D90 (P < or = 0.001). The cytokine pattern observed at D90 after immunotherapy was similar to that observed in control subjects. In conclusion, venom immunotherapy induced an altered cytokine mRNA pattern in allergen-stimulated T cells which was dissociated from the early changes of allergen-induced T-cell responsiveness.     

Spontaneous secretion of interleukin-4, interleukin-10 and interferon-gamma by first trimester decidual mononuclear cells

PROBLEM: A T-helper cell type 2 (Th2) cytokine dominated microenvironment has been predicted to be crucial for successful pregnancy. However, little information is available about local cytokine secretion in the human decidua. We determined the spontaneous secretion of interleukin-4 (IL-4), interferon-gamma (IFN-gamma) and IL-10 by decidual mononuclear cells at the single cell level and compared it with their secretion by peripheral blood mononuclear cells (PBMC) in the first trimester of pregnancy. METHODS OF STUDY: The cytokine secretion from decidual and blood cells was detected by a sensitive enzyme-linked immunosorbent spot-forming cell (ELISPOT)-assay. RESULTS: Cells secreting IL-4 (median 153, range 8-530), IL-10 (median 188, range 32-1600) and IFN-gamma (median 123, range 15-1140) were detected in all decidual and blood samples. The cytokine secretion showed a co-linear pattern in both the blood and decidua, i.e. when one cytokine was secreted at high levels, the others followed the trend. No correlation was found between the number of cytokine secreting cells in blood and decidua for any of the cytokines. CONCLUSIONS: Interleukin-4 and IL-10 are locally secreted in the decidua early during normal pregnancy, probably counteracting the fetal rejecting effects of co-expressed IFN-gamma. The cytokine secretion by blood cells does not generally reflect the local secretion pattern during first trimester pregnancy.     

Intracellular expression of interleukin-4 and interferon-gamma by a Mycobacterium tuberculosis antigen-stimulated CD4+ CD57+ T-cell subpopulation with memory phenotype in tuberculosis patients

In some chronic pathological conditions, antigen persistence activates and expands the CD4+ CD57+ T-cell subset. The host immune response against tuberculosis infection is maintained through the continuous presence of antigen-stimulated effector/memory helper T cells. To determine whether CD4+ CD57+ T cells were also expanded in human tuberculosis, we analysed (by flow cytometry) the phenotype of peripheral blood CD4+ T cells from 30 tuberculosis patients and 30 healthy controls. We observed a significant increase in the CD4+ CD57+ T-cell subset in tuberculosis patients in comparison to healthy controls (P < 0.001). Most CD4+ CD57+ T cells exhibited a CD28- CD45RO+ CD62L- phenotype, which is associated with memory cells. In vitro, a higher number of antigen-stimulated CD4+ CD57+ T cells produced intracellular interferon-gamma and interleukin-4 compared with antigen-stimulated CD4+ CD57- T cells (P < 0.001). These findings suggest that the majority of CD4+ CD57+ T cells correspond to a phenotype of activated memory T cells.     

Immunodominance in CD4 T-cell responses: implications for immune responses to influenza virus and for vaccine design

CD4 T cells play a primary role in regulating immune responses to pathogenic organisms and to vaccines. Antigen-specific CD4 T cells provide cognate help to B cells, a requisite event for immunoglobulin switch and affinity maturation of B cells that produce neutralizing antibodies and also provide help to cytotoxic CD8 T cells, critical for their expansion and persistence as memory cells. Finally, CD4 T cells may participate directly in pathogen clearance via cell-mediated cytotoxicity or through production of cytokines. Understanding the role of CD4 T-cell immunity to viruses and other pathogens, as well as evaluation of the efficacy of vaccines, requires insight into the specificity of CD4 T cells. This review focuses on the events within antigen-presenting cells that focus CD4 T cells toward a limited number of peptide antigens within the pathogen or vaccine. The molecular events are discussed in light of the special challenges that the influenza virus poses, owing to the high degree of genetic variability, unpredictable pathogenicity and the repeated encounters that human populations face with this highly infectious pathogenic organism.     

Reciprocal action of interferon-gamma and interleukin-4 promotes granulomatous inflammation induced by Rhodococcus aurantiacus in mice.

An intravenous injection of Rhodococcus aurantiacus to mice causes granulomatous inflammation dependent on endogenous interferon-gamma (IFN-gamma). The present study examined the role of endogenous interleukin-4 (IL-4) on granulomatous inflammation. Endogenous IL-4 in the spleen extracts was not detected during the phase of granuloma formation by enzyme-linked immunosorbent assay (ELISA). However, IL-4 protein level was elevated during the phase of granuloma regression. IL-4 mRNA expression in the livers and spleens was also elevated during the phase of granuloma regression. In addition, IL-4 levels during the phase of granuloma formation were increased by treatment with anti-IFN-gamma monoclonal antibody (mAb), suggesting that endogenous IFN-gamma might inhibit IL-4 production during the phase of granuloma formation. Administration of anti-IL-4 mAb on weeks 3 and 4 after the inoculation inhibited the regression of granulomas and augumented IFN-gamma level at 5 weeks. Endogenous IFN-gamma was produced by CD4+ T cells during the phase of granuloma regression and endogenous IL-4 was produced by both CD4+ and CD8+ T cells. These findings suggest that during the phase of granuloma formation endogenous IL-4 might be inhibited by IFN-gamma, while during the phase of granuloma regression endogenous IL-4 might play a crucial role in the reduction of granulomas and IFN-gamma production.