Human T cells from autoimmune and normal individuals can produce tumor necrosis factor

T cell clones derived from patients with autoimmune diseases were found to be capable of producing tumor necrosis factor (TNF). This was demonstrated by stimulating the clones, in the absence of accessory cells, with antibodies against the Ti/T3 complex and with recombinant interleukin 2 (IL2). Analysis of RNA extracted from these clones showed that TNF mRNA was more abundant than lymphotoxin (LT) mRNA. We also found that TNF protein in the supernatants of these clones was generally more abundant than LT as assessed by using the murine L929 cell assay. TNF production was not limited to T cells from autoimmune individuals, since the T cell tumor HUT78 and T cells purified from the peripheral blood of healthy individuals also made TNF. Unlike the T cell clones, HUT78 produced greater amounts of LT mRNA than TNF mRNA. Induction of TNF mRNA in T cells from healthy individuals displayed a two-signal requirement (phorbol myristate 13-acetate and phytohemagglutinin or OKT3 and phorbol myristate 13-acetate), similar to that described for the induction of the T cell lymphokines IL 2 and interferon-gamma (IFN-gamma). Additionally we found that IL2 alone was sufficient to induce TNF in these cells when they had been precultured with phytohemagglutinin for 7 days to express IL 2 receptors. The cloned T cells we have characterized also produce IFN-gamma which was detected in the supernatants of the clones using a radioimmunoassay. The evidence suggests that T cells can produce TNF and have the potential to deliver by themselves the dual and synergistic signals of TNF/LT and IFN-gamma to target cells, a process which may be of importance in the pathogenesis of human autoimmunity.     

In vivo lymphokine production in experimental autoimmune uveoretinitis.

Experimental autoimmune uveoretinitis (EAU) is a well-characterized model of immune-mediated intraocular inflammation. The intraocular infiltrate in EAU consists predominantly of T lymphocytes. The in vivo production of interleukin-2 (IL-2), lymphotoxin and IL-4 by these T cells was investigated by in situ hybridization using cDNA probes to lymphokine mRNA. Localization of lymphokine mRNA was found simultaneous with disease onset in areas of T-cell infiltration. Positive signal was seen over cells in the uveal tract, retina and extraocular region. Less than 10% of the population of T cells defined immunohistochemically had positive localization of mRNA for these lymphokines. The number of positive cells was similar for each of the three probes and increased as the disease progressed. The findings suggest that these lymphokines are produced in vivo in immune-mediated intraocular inflammation and may play a role in the immunopathology seen in these conditions.     

Differential effect of transforming growth factor beta on the synthesis of Th1- and Th2-like lymphokines by human T lymphocytes.

(TGF)-beta is a pluripotent cytokine exerting differential effects on distinct components of the immune response. The present report, based on lymphokine determination in culture supernatants and Northern blot analysis of lymphokine mRNA, demonstrates that TGF-beta 2 markedly inhibits interleukin (IL)-4 and IL-5 synthesis by polyclonally activated human T cells in the absence of any significant effect on (IFN)-gamma, lymphotoxin or IL-2, suggesting a modulatory effect of TGF-beta 2 on the interferon Th1/Th2) balance of immune responses. The inhibitory effect of TGF-beta on IFN-gamma production by unfractionated peripheral blood mononuclear cells is likely to reflect the blunting of natural killer cell activation by TGF-beta.     

Differential inhibition by cyclosporin A reveals two pathways for activation of lymphokine synthesis in T cells

Two pathways for the activation of lymphokine synthesis in murine T cell clones and polyclonal T cell blast populations were identified. One was induced by ligands of the T cell receptor (TCR) and led to high production of GM-CSF, IFN-gamma, and IL-3. The other was induced by IL-2 and led to production of lower levels of GM-CSF and IFN-gamma with relatively little IL-3 synthesis. Cyclosporin A (CsA) markedly inhibited TCR-independent production of lymphokine mRNA and protein at concentrations where IL-2-dependent stimulation of lymphokine production and proliferation was unaffected. Stimulation of lymphokine synthesis by phorbol myristate acetate (PMA) and the Ca2+ ionophore ionomycin, or by ionomycin alone, mimicked the TCR-dependent response. PMA on its own was a preferential stimulus for GM-CSF production, but, whereas CsA did not inhibit PMA stimulation of polyclonal T cell blasts, T cell clones displayed a biphasic response in which CsA only inhibited stimulation by high PMA concentrations. The data suggest that Ca2(+)-independent (CsA-resistant) T cell activation induces synthesis of GM-CSF and IFN-gamma but is a poor stimulus for IL-3 production. On the other hand, when Ca2(+)-dependent (CsA-sensitive) pathways are activated by TCR binding or by a Ca2+ ionophore, production of high levels of all three lymphokines can be induced.     

Mitogen-Induced Interleukin 2 and Gamma Interferon Production by CD4+ and CD8+ Cells of Patients with Inflammatory Arthritides. A Comparison between Cells from Synovial Fluid and Peripheral Blood

The purpose of this study was to investigate interleukin 2 (IL-2) and gamma interferon (IFN-gamma) production by purified CD4+ and CD8+ cells isolated from peripheral blood (PB) and synovial fluid (SF) of patients with rheumatoid arthritis (RA) and other inflammatory arthritides (non-RA). CD4+ and CD8+ cells were selected positively by immunomagnetic separation. Supernatants of unstimulated CD4+ and CD8+ cells from both compartments did not contain any detectable IL-2 or IFN-gamma, while supernatants of CD4+ and CD8+ cells stimulated with phytohaemagglutinin and irradiated Raji cells mostly contained both cytokines. In vitro stimulated SF CD4+ cells gave supernatants with significantly less IL-2 than supernatants from PB CD4+ cells, while in vitro stimulated SF CD4+-cell supernatants contained significantly more IFN-gamma. SF CD4+-cell supernatants contained significantly more IL-2 than the parallel CD8+ supernatants, while there was no significant difference with regard to IFN-gamma content. The pattern of differences between SF- and PB-derived T cells was the same for the two groups of patients, but the SF CD4+ cells from RA patients produced significantly less IL-2 than the corresponding cells from the non-RA group. The difference between SF and PB T cells with regard to lymphokine production is probably related to various degrees of in vivo pre-activation. The results do not indicate a major T-cell deficiency in relation to lymphokine production in RA.     

Human peripheral blood CD4+ and CD8+ T cells express Th1-like cytokine mRNA and proteins following in vitro stimulation with heat-inactivated Brucella abortus.

Defining the pattern of lymphokine production associated with Brucella abortus is critical for advancing the development of B. abortus as a vaccine carrier. In the present study we investigated the ability of heat-inactivated B. abortus or lipopolysaccharide from B. abortus to induce lymphokine production from purified human T cells in vitro. Gamma interferon (IFN-gamma), interleukin-2 (IL-2), IL-4, and IL-5 induction was assayed by mRNA-specific PCR and by enzyme-linked immunosorbent assay and bioassay for protein production. Following depletion of monocytes and B cells, B. abortus increased IFN-gamma and IL-2 mRNA expression in purified T cells compared with expression in unstimulated cells. In contrast, no IL-5 mRNA expression and only transient low-level IL-4 mRNA expression and no IL-4 protein secretion were detected. Phytohemagglutinin or phorbol myristate acetate plus ionomycin induced mRNA and protein for all these cytokines. Similar results were obtained with LPS purified from B. abortus. Removal of NK cells did not reduce lymphokine production, and enriched NK cells did not express IFN-gamma mRNA or secrete IFN-gamma protein in response to B. abortus, indicating that NK cells were not the responding population. Both CD4+ and CD8+ populations produced IFN-gamma and IL-2 in response to B. abortus. Preincubation of resting T cells with B. abortus or LPS from B. abortus for 7 days induced their differentiation into Th1-like cells as judged by their subsequent lymphokine response to phorbol myristate acetate plus ionomycin. These results suggest that B. abortus can induce differentiation of Th0 into Th1-type cells.     

Patterns of lymphokine production by primary antigen-specific/MHC restricted murine helper T cell clones.

In this study we examined a panel of CD4+ antigen specific/MHC restricted T cell clones for their ability to secrete IL-2, IL-4, and IFN-gamma upon stimulation with con A, three lymphokines which are diagnostic for the TH1 and TH2 subtypes of helper T cells. Eight of the twelve clones we analyzed did not fit the classical TH1/TH2 patterns of lymphokine secretion. Seven of these clones secreted both IL-2 and IL-4 and two of these also produced IFN-gamma. The remaining non-classical clone secreted IL-4 and IFN-gamma but not IL-2. Data from the subcloning of the IL-2/IL-4/IFN-gamma triple producers were not consistent with the parental lines being a mixture of TH1 and TH2 cells. The IL-2/IL-4 double producers (IFN-gamma negative) cannot be explained by the parental lines being a mixture of the TH1 and TH2 subtypes. Nevertheless, these double producers were subcloned and the results provided convincing evidence that clones which secrete both IL-2 and IL-4 do exist. Lymphokine loss variants involving IL-2, IL-4 or IFN-gamma were observed among subclones derived from the double and triple producers as well as in several parental lines maintained in continuous culture. We also observed the appearance of inducible IFN-gamma production in some subclones derived from parental clones where production of IFN-gamma was not detectable. The phenotypes of these variants failed to indicate an obvious trend toward the TH1 and TH2 subtypes. Thus, our results suggest that more heterogeneity in the population of CD4+ helper T cells exists than can be explained by the TH1 and TH2 subtypes of these cells.     

There is No Correlation Between Function and Lymphokine Production of HBs-Antigen-Specific Human CD4+-Cloned T Cells

The question whether antigen-specific human CD4+ T cells can be classified on the basis of appropriate and fixed lymphokine production patterns and their corresponding functions still remains to be elucidated. We generated ten CD4+ T-cell clones specific for HBsAg from HBsAb-positive but HBsAg-negative individuals. Seven of these clones exhibited helper activity for HBsAb response, while the three other clones did not. Both helper- and non-helper-type T-cell clones produced interleukin 4 (IL-4) after antigenic stimulation. By stimulation with phytohaemagglutinin (PHA) plus phorbol myristate acetate (PMA), three of the seven helper-type clones produced interleukin 2 (IL-2) in addition to IL-4. However, the other four helper-type clones did not produce IL-2 by such stimulation, although they continued the production of IL-4. All non-helper-type T-cell clones produced a large amount of IL-2, and some of them completely became an IL-2 producer after certain stimulation. These results suggested that both helper- and non-helper-type CD4+ T-cell clones specific for HBsAg might have no strict pattern of lymphokine production as in the TH1/TH2 dichotomy of murine CD4+ T cells. The data also revealed that lymphokine-producing capacity of individual cloned T cells is changeable depending upon the sort of activation.     

Induction of global anergy rather than inhibitory Th2 lymphokines mediates posttrauma T cell immunodepression

Depressed mitogen-induced IL-2 and IFN-gamma responses after severe mechanical or thermal injury are postulated to result from an expansion of Th2 lymphocytes with concomitant excessive production of IL-4 and/or IL-10. Here, we simultaneously assessed proliferation and Th1 (IFN-gamma) versus Th2 (IL-10, IL-4) lymphokine production in trauma patients' isolated T cells stimulated in a costimulation sufficient, antigen presenting cell independent system (anti CD3 + anti-CD4). T cells with depressed proliferation and IL-2 production simultaneously lost IL-4, IL-10, and IFN-gamma protein and mRNA responses. Exogenous IL-12 addition did not restore IFNgamma responses, but exogenous IL-2 partially restored IL-4, IFN-gamma, and IL-10 production. Although initially partially restored by exogenous IL-2 or stimulation with PMA + ionomycin, patient T cells with persisting anergy progressively lost even these lymphokine and proliferative responses. Development of global T cell anergy was not a result of lost T cell viability or protein synthesis, since it corresponded to predominance of anergic T cells with upregulated expression of CD11b, but downregulated CD28 and CD3 expression. Thus, the subset of posttrauma patients whose isolated T cells become unresponsive experienced progressively worsening global anergy, mediated not by an increased production of Th2 lymphokines, but possibly by T cell incapacity to be activated through TCR triggering or Ca(2+) mobilization.     

P cell stimulating factor release: a useful assay of T cell activation in vitro

The coordinate production of multiple lymphokines by activated T cells allows the investigator a choice of assays for antigen recognition in vitro. In this paper, we examine the production of P cell stimulating factor (PSF, a lymphokine identical to interleukin 3) by L3T4+ Lyt-2-T cell clones. PSF production is antigen-specific, Ia-restricted and inhibited by a monoclonal antibody to L3T4, PSF, interferon-gamma (IFN-gamma) and interleukin 2 (IL 2) are all rapidly secreted after antigen or mitogen stimulation. The measurement of PSF avoids problems inherent in the IL 2 and IFN-gamma assays, namely absorption of IL 2 by the producer T cells and failure of the interferon assay to distinguish between interferons of the alpha, beta and gamma classes.     

Defects in the regulation of anti-DNA antibody production in aged lupus-prone (NZB x NZW)F1 mice: analysis of T-cell lymphokine synthesis.

(NZB x NZW)F1 (B/W) mice spontaneously develop a lupus-like syndrome characterized by an increased level of autoantibodies in old mice. We analysed the role of T cells in the regulation of anti-DNA antibody production by B cells in vitro as a function of age. In cultures of old mouse T and B cells, IgG and IgM anti-DNA antibodies were synthesized at high levels, in contrast to consistently lower amounts, particularly of IgG, measured in cultures of young mouse cells. Addition of young mouse T cells to old B cells inhibited IgG, but not IgM, anti-DNA production, whereas T cells from old mice stimulated IgG synthesis by young mouse B cells. Addition of supernatants harvested from concanavalin A (Con A)-stimulated T cells to B-cell cultures induced similar effects. Therefore, we evaluated possible modifications of lymphokine synthesis compared to that of the healthy NZW parent. T cells from old mice were able to secrete normal levels of interferon-gamma (IFN-gamma) and interleukin (IL)-10; however, secretion of IL-2 and IL-4 was dramatically decreased. Semi-quantitative polymerase chain reaction analysis of constitutive RNA messengers showed increased IFN-gamma levels in young and old B/W mice, and normal IL-10 mRNA levels in young and higher levels in old mice. Constitutive IL-2 and IL-4 mRNA were detected only after Con A stimulation and their levels decreased in old compared to young B/W mice; in particular IL-2 mRNA was considerably lower in old B/W than in control NZW mice. Taken together, these results suggest that, despite constitutive T-cell abnormalities, young B/W mice are able partially to control their lymphokine production, whereas aged mice exhibit a deficient synthesis, associated with an increased capacity to produce IFN-gamma.     

Peripheral blood and intrathyroidal T cell clones from patients with thyroid autoimmune diseases

For a better understanding of the pathogenesis of thyroid autoimmune diseases, we have studied morphological and functional properties of T clones from peripheral blood lymphocytes (PBL) and from intrathyroidal lymphocytes (ITL) obtained from 3 patients with Graves' disease or 1 Hashimoto's thyroiditis. Investigations were carried out on clones cultured alone or cocultured with autologous thyrocytes. Clonage efficiency ranged from 30% to 33% for PBL and 10% to 36% for ITL. A predominance of CD4-positive clones was observed whatever the origin of the lymphocytes or the autoimmune pathology. Gamma interferon (IFN-gamma) was detected in the majority (17/19) of the clones tested. Intracytoplasmic interleukin (IL-4) was secreted in 7/19 clones and both cytokines were produced in 5/19 clones. In coculture a proliferative response and tumour necrosis factor (TNF-alpha) production were observed with 6 clones (4 from Graves thyrocytes and 2 from thyroiditis). No cytotoxic clone was derived from Graves or thyroiditis tissues. These data demonstrate that the large majority of T clones are principally CD4-T cells; all the clones secreted TNF-alpha and a large majority produced IFN-gamma. Only a few clones produced IL-4 alone or associated with IFN-gamma. Six T clones induced proliferative response and of TNF-alpha secretion in coculture. Further investigations must be performed on these antigen-reactive T clones to analyse their role in the pathogenesis of the human thyroid autoimmune diseases.     

Cell-mediated immunity to chemically xenogenized tumors--IV. Production of lymphokine activity by, and in response to, highly immunogenic cells

To determine whether a novel pattern of lymphokine production might be involved in the superior immunogenicity of chemically xenogenized tumors over that of parental cells, we tested a panel of murine tumors xenogenized by DTIC for production of soluble factors with lymphokine-like activity and induction of lymphokine release from na?ve or specifically sensitized lymphocytes. In the L5178Y tumor system, a majority of xenogenized but not parental clones produced an IL-1-like factor, and this was associated, as a rule, with class II antigen expression and antigen-presenting ability. However, no such properties were exhibited by the xenogenized variants of P815 and L1210Ha cells, which nevertheless occasionally expressed other lymphokine (GM-CSF, IL-3) activities. On examining the ability of xenogenized and parental tumors to cause release of IL-1, IL-2, IL-3, IFN-gamma, TNF/LT and GM-CSF from T-cells, we found, as a rule, an increased lymphokine production when lymphocytes primed in vivo to a xenogenized tumor were restimulated in vitro with the same or parental cells.     

Functionally distinct human T cell clones that produce lymphokines with IL-2-like activity

Various functionally distinct human T cell clones derived from in vitro mixed leukocyte cultures are found to secrete lymphokines with detectable Interleukin-2 (IL-2)-like activity upon antigenic stimulation. These lymphokine producing clones are not only dependent on exogenous growth factors provided in the form of crude phytohemagglutinin (PHA)-induced spleen conditioned medium for growth but can themselves be driven to proliferate by their own lymphokines. The induction of lymphokine production appears to be antigen-specific and the lymphokines secreted are believed to contain nonantigen and nonspecies specific IL-2-like activity. We show here that IL-2-like lymphokines are produced by a subset of T lymphocyte clones, i.e., the help-independent cytotoxic T cells clones that have the capacity to proliferate to, as well as to lyse, the original sensitizing cells. In addition, other T cell clones capable of producing active lymphocytes include those clones that have the T helper cell characteristics, i.e., can undergo antigen-induced proliferation but are not cytotoxic; in the presence of lectin (PHA), however, some helper T cells (Th) but not others, can express lectin-dependent cell-mediated lysis. Finally, yet another subset of T lymphocytes, the help-dependent cytotoxic T cell clones that cannot proliferate to antigenic stimulation, was found to produce no detectable IL-2-like activity.     

Transforming growth factor-beta1 suppresses interleukin-15-mediated interferon-gamma production in human T lymphocytes

One of the most remarkable means by which tumour cells manage to evade recognition and elimination by the immune system is the release of immunosuppressive mediators, such as interleukin (IL)-10 or transforming growth factor-beta (TGF-beta). For antitumour immunotherapies to reach their full potential, cytokine cocktails will have to be custom-tailored to the tumour's individual cytokine microenvironment. One of the components of such a cytokine cocktail may be interleukin (IL)-15, which has demonstrated an excellent stimulatory potential of antitumour immunity. In an in vitro model, we have previously been able to show that the negative effects of IL-10 on IL-15-mediated cytotoxic T-cell activation can be outweighed by the addition of interleukin (IL)-12. The mechanism by which TGF-beta may influence the effect of IL-15 remains poorly understood, however. We have therefore taken our T-cell model further and have studied the effect of TGF-beta on IL-15-mediated interferon-gamma (IFN-gamma) production. In activated, IL-15-stimulated peripheral blood T lymphocytes, TGF-beta suppressed IFN-gamma mRNA and protein levels by approximately 75%. This effect was likewise observed on both CD4+ and CD8+ T cells and, in contrast to the effect of IL-10 in this system, could not be neutralized by the addition of IL-12. Thus, immunotherapy for TGF-beta-producing tumours may benefit from the addition of TGF-neutralizing activity rather than IL-12.     

Differential regulation of lymphokine production in mitogen-stimulated murine spleen cells.

Activation of murine spleen cells in vitro with soluble anti-CD3 monoclonal antibody and phorbol 12-myristate 13-acetate (PMA) induced an initial production of interleukin 2 (IL2), interferon-gamma (IFN-gamma), IL4 and IL5, followed by a refractory state during which the T cells did not produce lymphokines when stimulated with some common mitogens. The refractory state was long-lasting, depended on the presence of anti-CD3 and PMA but could be reverted by incubation in fresh medium. Pokeweed mitogen (PWM) differed from other mitogens tested, since stimulation by PWM and PMA induced lymphokine production and proliferation also in the refractory cells. Furthermore, PWM stimulation selectively induced IL4 and IFN-gamma production but not IL2 and IL5, as detected by intracellular cytokine-specific immunofluorescence and in situ hybridization for mRNA. The results indicate differential regulation of lymphokine production in primary lymphocytes.     

Production of interleukin 10 and transforming growth factor beta in concomitant allergy and autoimmunity.

BACKGROUND: The immunologic relationship between T(H)1-type autoimmune disorders and T(H)2-type allergic disorders and the role of T-cell regulation in humans is as yet unclear. The regulatory cytokine production capacity of individuals with concomitant allergy and T(H)1-type autoimmunity may provide insight into the role of T-cell regulation in both disorders. OBJECTIVES: To examine the production capacity of interleukin 10 (IL-10) and transforming growth factor beta (TGF-beta), 2 regulatory cytokines, in individuals with concomitant allergic rhinitis and T(H)1-type autoimmune diagnoses and to compare that capacity with that in individuals with allergic rhinitis only and individuals with neither diagnosis. METHODS: Seventeen case subjects and 17 age-, sex-, and ethnicity-matched controls with allergic rhinitis only were recruited from an allergy clinic. Fourteen matched controls with neither diagnosis were recruited from the general population. Peripheral blood mononuclear cells were obtained and cultured with and without mitogen stimulation (lipopolysaccharide and phytohemagglutinin). Cytokine levels from culture supernatants were measured by enzyme-linked immunosorbent assay. RESULTS: Cases with allergic rhinitis and autoimmune diseases had significantly lower unstimulated day 3 IL-10 levels compared with controls with allergic rhinitis only (P = .05) and significantly lower stimulated day 5 TGF-beta levels compared with controls with neither diagnosis (P = .02). Cases had consistently lower regulatory capacity compared with both control groups, as measured by an additive index using IL-10 and TGF-beta levels. CONCLUSION: Individuals with concomitant allergic rhinitis and T(H)1-type autoimmune disorders have a lower regulatory cytokine production capacity than individuals with allergic rhinitis only and those with neither diagnosis.     

In Situ Hybridization of Interferon-Gamma Producing Peripheral Blood Mononuclear Cells

Individual interferon-gamma (IFN-gamma) producing cells in activated human peripheral blood mononuclear cells (PBMC) were characterized by in situ hybridization using [35S]-labelled antisense RNA probes. The proportion of positive cells expressing IFN-gamma mRNA varied according to the substances used for stimulation. IFN-gamma mRNA expressed a relatively low percentage of 1-8% PBMC after a single stimulus with mitogens or OKT-3 antibody and 20-30% of the cells were identified to synthesize IFN-gamma mRNA after stimulation with PHA + P-MA + OKT-3 antibody. The expression of IFN-gamma mRNA and production of the lymphokine was dependent on accessory cells. If accessory cells were replaced by recombinant interleukin-1 (IL-1) plus interleukin-6 (IL-6), then T-cell proliferation to phytohaemagglutinin (PHA) could be partially restored and measurable amounts of IFN-gamma were detected. The addition of interleukin-2 (IL-2) or phorbol-12-myristate-13-acetate to T cells stimulated with PHA, IL-1 and IL-6 did not restore the production of IFN-gamma to an extent comparable to that produced by T cells stimulated in the presence of accessory cells. In further studies, depletion of T-cell subsets showed that CD3+, CD4+, CD8+, CD29+ and CD45RA+ cells were involved in IFN-gamma production after mitogenic stimulation. In conclusion, our data demonstrate that IFN-gamma production is dependent on signals from accessory cells and IFN-gamma is synthesized by only a small proportion of T cells, that did not belong to a unique population, characterized by conventional cellular surface antigens.     

Cytokine profiles of BAL T cells and T-cell clones obtained from human asthmatic airways after local allergen challenge.

BACKGROUND: This study assessed the heterogeneity of cytokine expression in asthma before and after local allergen challenge. METHODS: BAL T cells were obtained 10 min or 24 h after local endobronchial allergen challenge in atopic asthmatic subjects. T cells were cloned by direct limiting dilution. mRNA expression was assessed by RT-PCR, and cytokine protein production by ELISA. RESULTS: Unstimulated baseline BAL T cells expressed mRNA for IFN-gamma, IL-13, and TNF-alpha. A minority of samples expressed IL-4 and IL-5, but no IL-3 mRNA was detected. PHA stimulation increased expression of IL-3, IL-4, and IL-5 mRNA in 4/6 samples. IL-13 and GM-CSF mRNA were found in BAL cells after allergen challenge, but expression of IFN-gamma was reduced. Both IL-4 and IL-3 were strongly upregulated after PHA stimulation, while the expression of TNF-alpha and IFN-gamma was reduced, compared to equivalent baseline samples. Seventeen panels of BAL T-cell clones were derived (average cloning efficiency 1/40 T cells). Seven panels survived to 8 weeks for analysis. Clones derived 4 h after saline challenge showed strong mRNA signals for IL-13, IL-4, and IFN-gamma, whereas clones derived 24 h after allergen challenge expressed IL-13, GM-CSF, IL-3, IL-4, and often IL-5 (i.e., closer to the Th2 profile). There was considerable heterogeneity in the patterns of cytokine mRNA and protein production by different clones. CONCLUSIONS: T cells from asthmatic airways produce IL-13, IFN-gamma, and TNF-alpha, but after allergen challenge, type 2 cytokines are upregulated. mRNA and protein analysis provide complementary information on airways T-cell cytokine profiles.