Interferon-gamma and IL-10 may protect from allergic polysensitization in children: preliminary evidence.

BACKGROUND: A functional defect of T regulatory cells (Treg) has been proposed as pathogenic mechanism of allergic reaction. Polysensitization is a common feature of allergic patients. AIM OF THE STUDY: It was to investigate the possible role of Treg-Th1 cytokines, in the development of new sensitizations in childhood. METHODS: Forty monosensitized (MS) children with allergic rhinitis were evaluated and followed-up for 2 years. New sensitizations were investigated. IL-10 and IFN-gamma were evaluated in in vitro experiments. RESULTS: Children remaining MS showed significant higher production of both IL-10 and IFN-gamma. CONCLUSION: This preliminary study provided evidence that IL-10 and IFN-gamma production could be defective in allergic children prone to develop polysensitization.

Levels of soluble CD27 in sera and synovial fluid and its expression on memory T cells in patients with juvenile idiopathic arthritides.

OBJECTIVE: CD27 is a member of tumour necrosis factor receptor family. Its expression is predominantly confined to mature lymphocytes and is strongly enhanced after cell activation. Shedding of the CD27 from the surface of activated cells is related to their effector phase. The aim of the present study was to evaluate the levels of soluble CD27 in sera and synovial fluids, together with its expression on circulating and synovial fluid (SF) memory T cells, in children with JIA. METHODS: Sera from 40 patients with active JIA were studied for soluble CD27. Paired SF samples were available in 20 patients. Sera from 12 age-matched patients affected with various acute infectious diseases and 12 age-matched healthy subjects were used as controls. In 8 JIA patients freshly isolated circulating and SF lymphocytes were stained for CD27 in CD4+CD45 RO+ T cell subpopulation and analyzed by cytometry. RESULTS: Soluble CD27 serum levels were significantly higher in patients with polyarticular JIA and acute systemic infectious diseases than in patients with active oligoarticular or healthy controls. Both polyarticular and oligoarticular JIA patients showed increased levels of soluble CD27 in SF when compared with paired serum samples (p = 0.01). In all the patients tested a significant enrichment of CD27- T cells was seen in the SF (median 39.5%, range 18-56%) when compared to paired CD4+CD45RO+ peripheral lymphocytes (median 19.5%, range 5-43%; p = 0.01). CONCLUSIONS: A clear enrichment of CD4+ memory SF T cells with a CD27-phenotype is observed when compared to correspondent circulating T lymphocytes. This issue is conceivably related to re-activation and recruitment of memory T cells to the site of inflammation, and to the subsequent expansion of a subpopulation of "effector" memory T cells.

Plasma levels of soluble CD30 are increased in children with chronic renal failure and with primary growth deficiency and decrease during treatment with recombination human growth hormone.

BACKGROUND: Previous studies have suggested that in vivo Th2 lymphocyte activation is related to increased soluble CD30 (sCD30) plasma levels. As various hormones (dehydroepiandrosterone, glucocorticoids, progesterone) can regulate the Th1/Th2 balance, and because growth hormone (GH) enhances lymphocyte function, we measured sCD30 plasma levels, before and after treatment with recombinant human GH (rhGH), in children with growth failure due to chronic renal failure (CRF) or to isolated GH deficiency in order to evaluate the potential effects of rhGH treatment on Th1/Th2 balance. METHODS: sCD30 plasma levels were determined by ELISA assay in 30 children with CRF (mean age 10.7+/-3.7 years), in five children with isolated GH deficiency (mean age 11.4+/-2.6 years), and in 10 normal controls (mean age 10.1+/-3.5 years). RESULTS: sCD30 levels were higher in the 30 children with CRF than in the 10 controls (179.8+/-79.4 vs 11.3+/-10.9 U/ml, P<0.001) exhibiting an inverse correlation with glomerular filtration rate (GFR) (r=-0.7860, P<0.001). In 11 children with CRF, after 19.9+/-16.7 months of rhGH treatment, a decrease of sCD30 plasma level (170+/-50 vs 134+/-49 U/ml, P<0.01) was observed. The five children with primary GH deficiency had higher sCD30 plasma level than controls (mean 147+/-105 vs 11+/-10 U/ml, P<0.004) and sCD30 plasma levels decreased to 95.2+/-109.6 U/ml after rhGH treatment. CONCLUSIONS: The finding that rhGH treatment decreased sCD30 plasma levels in children with CRF, and that children with primary GH deficiency had higher sCD30 plasma levels than controls, suggest that GH may regulate CD30 expression and possibly the balance of Th1/Th2. Whether the uraemia-induced increase in sCD30 is due to decreased renal excretion, to overproduction or both, remains to be determined.

T cell clones raised from chronically infected healthy humans by stimulation with Toxoplasma gondii excretory-secretory antigens cross-react with live tachyzoites: characterization of the fine antigenic specificity of the clones and implications for vaccine development.

Excreted-secreted Ags (ESA) of Toxoplasma gondii (Tg) play an important role in the stimulation of the host immune system in both acute and chronic infections. To identify the parasite Ag(s) involved in the maintenance of T cell-mediated long term immunity, 40 ESA-specific T cell clones were derived from three chronically infected healthy subjects. All the clones were CD4+ and recognized both ESA and live tachyzoites in a HLA-DR-restricted manner. Conversely, CD4+ tachyzoite-specific T cell clones from the same subjects proliferated in response to ESA, pointing to shared immunodominant Ags between ESA and Tg tachyzoites. By T cell blot analysis using SDS-PAGE-fractionated parasite extracts, the following patterns of reactivity were detected. Of 25 clones, 6 recognized Tg fractions in the 24- to 28-kDa range and proliferated to purified GRA2, 5 reacted with Tg fractions in the 30- to 33-kDa range; and 4 of them proved to be specific for rSAg1. Although surface Ag (SAg1) is not a member of ESA, small amounts of this protein were present in ESA preparation by Western blot. Of 25 clones, 8 responded to Tg fractions in the 50- to 60-kDa range but not to the 55-kDa recombinant rhoptries-2 parasite Ag, and 6 did not react with any Tg fraction but proliferated in response to either ESA or total parasite extracts. In conclusion, CD4+ T cells specific for either ESA (GRA2) or SAg1 may be involved in the maintenance of long term immunity to Tg in healthy chronically infected individuals.

Ultrastructural and functional studies of the interaction between IL-12 and IL-2 for the generation of lymphokine-activated killer cells.

IL-12 promotes generation of LAK activity in short-term-cultured NK cells, but information on the structure and function of IL-12-induced LAK cells is not yet available. The latter issues have been here investigated with emphasis on interactions between IL-12 and IL-2. Peripheral blood mononuclear cells (MNC) exposed to IL-12 for 5-7 days displayed a decrease in the amount and density of the matrix of large granular lymphocyte (LGL)-associated granules. In cells cultured with IL-12 and IL-2 for 5-7 days, empty vacuoles were predominant and the electron-dense matrix was scanty. In MNC incubated with IL-2 for 5-7 days, most granules were loaded with electron-dense matrix. IL-12 and IL-2 displayed an additive effect on LAK cell cytotoxicity until approximately 48 h in culture which was followed by a sharp decline. Immunocytochemical and biochemical studies demonstrated that MNC cultured for 5-7 days with IL-12 and IL-2 displayed downregulated perforin expression and upregulated granzyme B expression. Fas ligand expression was virtually undetectable in MNC cultured for 5-7 days with or without cytokines. It appears that perforin downregulation plays a major role in the reduced cytotoxicity of MNC cultured with IL-12 and IL-2 for 5-7 days.

Peritoneal mesothelial cells produce complement factors and express CD59 that inhibits C5b-9-mediated cell lysis.

The CD59 membrane protein confers protection from C5b-9-mediated cell lysis. Because evidence exists for complement (C) activation and generation of C5b-9 in the peritoneal cavity during chronic peritoneal dialysis (CPD), we investigated, on mesothelial cell (MC) lines, the expression of CD59 and the production of C components. Four MC lines were obtained from children on CPD, and two from non uremic children. CD59 expression on MCs was investigated with anti-CD59 monoclonal antibody (mAb) and polyclonal goat immunoglobulin G (IgG). MC lines were positive for staining with anti-CD59 mAb. Western blotting analysis of MC membrane demonstrated a band with the same molecular weight as CD59. Incubation of MC with anti-CD59 mAb abrogated the protective effect of CD59 (100% cytotoxicity). C3, C4, and C6 were detected in the supernatants of MC; in non uremic MC supernatants, C5, C7, C8, and C9 were also detectable, and C4 concentration was tenfold higher. CD59 expression confers to MCs protection from C5b-9-mediated lysis. MCs produce C factors. These findings suggest that production of complement components and expression of CD59 on MCs could play a role both in peritoneal cavity infection (decreased complement production) and in peritoneal membrane damage (decreased CD59 expression and reduced remesothelialization owing to MC lysis).

Synovial fluid T cell clones from oligoarticular juvenile arthritis patients display a prevalent Th1/Th0-type pattern of cytokine secretion irrespective of immunophenotype.

The aim of the present study was to investigate the patterns of cytokine production by T cell clones raised from in vivo activated synovial fluid (SF) mononuclear cells (MNC) of five patients with oligoarticular juvenile arthritis (JA). Freshly isolated SF T cells were cultured in vitro with low dose recombinant IL-2 and subsequently cloned by limiting dilution. Sixty-four clones were obtained from the five patients studied. Fifty-nine clones were TCR alpha/beta+, either CD4+ (n = 43) or CD8+ (n = 15). The remaining five clones were TCR gamma/delta+, CD4-, CD8-. Clone immunophenotypes differed in the individual patients. Forty-four T cell clones were stimulated with phytohaemagglutinin (PHA) and phorbol myristate acetate (PMA) and supernatants tested for the presence of IL-2, IL-4, IL-5 and interferon-gamma (IFN-gamma) by ELISA or bioassays. Cytokine mRNA accumulation was tested by reverse transcriptase-polymerase chain reaction (RT-PCR). Most of 44 clones tested released large amounts of IFN-gamma irrespective of the immunophenotype. Of these, 27 were classified as Th1-type and 17 as Th0-type based upon the IFN-gamma/IL-4 ratio in culture supernatants. Finally, when 10 representative T cell clones were tested for pro- and anti-inflammatory cytokines, gene expression by RT-PCR, all of them were found to express the granulocyte-macrophage colony-stimulating factor (GM-CSF), tumour necrosis factor-alpha (TNF-alpha), IL-10 and transforming growth factor-beta 1 (TGF-beta1) genes, and half of them IL-6 and IL-8 mRNA. In conclusion, T cell clones, that represent the progeny of in vivo activated SF T cells from oligoarticular JA patients, display heterogeneous immunophenotypes, but all share the ability to produce large amounts of IFN-gamma, with a predominant Th1/Th0 pattern. The expression of pro- and anti-inflammatory cytokine genes in these clones suggests that in vivo activated SF T cells modulate joint inflammation in a complex fashion.

Functional and molecular characterization of tumour-infiltrating lymphocytes and clones thereof from a major-histocompatibility-complex-negative human tumour: neuroblastoma.

Neuroblastoma (NB) is a major-histocompatibility-complex(MHC)-negative neuroectodermal tumour that is often infiltrated with lymphocytes. A detailed characterization of NB-associated tumour-infiltrating lymphocytes (TIL) has never been carried out. Here we have investigated the immunophenotype and the cytotoxic activities of TIL from nine and seven NB patients respectively. Furthermore, the T cell receptor (TcR) variability and the patterns of cytokine gene expression of fresh versus recombinant (r) interleukin (IL)-2-cultured TIL were studied in four NB cases. The results obtained showed the following: (1) freshly isolated TIL were comprised of a mixture of CD4+ and CD8+ T cells partially expressing HLA-DR and/or CD25. The CD4/CD8 ratio ranged from 0.5 to 5 in the different cases. Upon culture of TIL with rIL-2, an increased proportion of CD56+ and CD8+ lymphocytes was consistently observed; (2) IL-2-expanded TIL lysed natural-killer(NK)sensitive and lymphokine-activated-killer(LAK)-sensitive target cell lines; (3) reverse-transcriptase/polymerase-chain-reaction (RT-PCR) experiments showed that most TcR V beta genes were expressed both in fresh and in cultured TIL, suggesting that such cell populations were polyclonal; (4) interferon gamma, IL-4, IL-5, tumour necrosis factor (TNF) alpha, IL-8, IL-10 mRNA and, to a lesser extent, IL-2 mRNA were expressed by cultured TIL, as assessed by RT-PCR; the corresponding tumour samples consistently contained TNF alpha, IL-8 and IL-10 mRNA, whereas IL-2 and IFN gamma mRNA were faintly expressed in some NB tumours and IL-4 and IL-5 mRNA were never detected. A total of 90 clones were subsequently raised from IL-2-expanded TIL from six NB patients; 87/90 clones were of T cell lineage with a CD4+ or CD8+ immunophenotype, whereas the 3 remaining clones were of NK cell origin. Upon triggering of the CD3-TcR complex, 64% CD4+ and 77% CD8+ T cell clones killed the murine P815 mastocytoma cell line. Virtually no T cell clone lysed a LAK-sensitive NB cell line whereas 15% CD4+ and 17% CD8+ clones mediated NK-like activity against the K562 cell line. Finally, the patterns of cytokine production by CD4+ clones were roughly consistent with those of a T helper (TH) 1 profile and similar to those observed in CD8+ clones.

Outgrowth of interleukin-2-dependent T cell lines with cytotoxic activity against mesothelial cells from cultures of peritoneal effluent cells of children on CPD.

Previous studies on the peritoneal immune system described the presence of activated T lymphocytes in peritoneal effluents (PE) from patients on chronic peritoneal dialysis (CPD), and showed that mesothelial cells (MC) can present antigens to T cells. In order to better define phenotypic and functional characteristics of T cells and their interactions with MC, we isolated PE cells from 15 children. At the immunophenotypic analysis, high percentages of activated T cells were identified (mean value: 15% double staining for CD3/DR; 12% CD25+). T cells with gamma/delta T cell receptor (mean 5%) and natural killer cells (mean 17%) were also present in elevated numbers. MC lines (n = 7) and interleukin-2-dependent T cell lines (9 CD4+; 1 CD8+) were also obtained by incubating PE cells under different conditions. Two cell lines showed a major histocompatibility complex (MHC) restricted cytotoxic activity against autologous MC; two lines killed allogeneic MC; one line killed both autologous and allogeneic MC. Although the hypothesis that activated T cells could kill MC after recognition of surface structures modified by dialysis fluid, or during antigen presentation, needs to be further investigated, our data suggest that the subsets of lymphocytes we identified could play an important role in the mechanisms of peritoneal membrane defense.

Toxoplasma gondii-specific CD4+ T cell clones from healthy, latently infected humans display a Th0 profile of cytokine secretion.

Human Toxoplasma gondii (Tg)-specific T cell clones were raised by infecting peripheral blood mononuclear cells (MNC) from two healthy, latently infected individuals with Tg trophozoites. All of the clones had a CD4+ immunophenotype and produced simultaneously interleukin (IL)-2, interferon (IFN)-gamma, IL-4 and IL-5 upon mitogen or antigen stimulation. Tg-specific T cell clones were classified as T helper of type 0 (Th0) since most of them released roughly comparable amounts of IFN-gamma and IL-4. In some clones, a trend to an increased production of IFN-gamma following antigen-specific as compared to non-specific stimulation was observed. The Th0 phenotype was also expressed by T cell clones that had been raised from bulk cultures performed in the presence of IL-4 or IFN-gamma. All of the Tg-specific T cell clones were cytolytic in a non-specific assay which involves the triggering of the CD3-T cell receptor (TcR) complex. Some clones specifically lysed an autologous lymphoblastoid cell line (LCL) that had been infected with Tg trophozoites. Finally, most of the Tg-specific T cell clones produced IL-10, irrespective of whether they had been raised from bulk cultures incubated in the presence or absence of IL-4 or IFN-gamma. Taken together, these findings suggest that Tg-specific Th0 helper cell clones from healthy, latently infected individuals, beside activating toxoplasmacidal mechanisms through IFN-gamma release, might limit the magnitude of the immune response of the parasite by killing Tg-infected antigen-presenting cells and by releasing IL-10.

[Rational bases for new approaches to the therapy of pediatric solid tumors: immunotherapy and gene therapy]. / Le basi razionali per nuovi approcci alla terapia dei tumori solidi pediatrici: immunoterapia e terapia genica.

Neuroblastoma is one of the commonest solid tumors in children. Conventional therapeutic approaches, such as surgery, chemotherapy and radiotherapy, fail to control tumor progression in stage III and IV patients. The search for novel therapeutic strategies should necessarily take into account immunotherapy and gene therapy. Here the theoretical bases for the development of such approaches are discussed. Studies carried out with neuroblastoma (NB) cell lines have shown that neoplastic cells express a wide array of potential tumor associated antigens (TAA) but are devoid of HLA molecules which are necessary for TAA presentation to the host immune system. Transfection of NB cells with the interferon gamma gene appears a promising approach, since this cytokine up-regulates the expression of class I HLA molecules in NB cells. Other cytokines of potential interest for gene transfer studies are interleukin 2 (IL2) and interleukin 12 (IL12).

Clonal analysis of peripheral blood lymphocytes from three patients with advanced neuroblastoma receiving recombinant interleukin-2 and interferon alpha.

In this study we have investigated, at the population and the clonal levels, the immunophenotypes and the non-specific cytotoxic functions of peripheral blood lymphocytes from three stage IV neuroblastoma patients receiving treatment with recombinant interleukin-2 (IL-2) and interferon alpha (IFN alpha). Both IL-2 alone and the combination of IL-2 and IFN alpha caused an in vivo expansion of CD56+, CD3- NK cells most of which expressed the p75 molecule, i.e. the beta chain of the IL-2 receptor. Peripheral blood mononuclear cells (PBMC), drawn after treatment, displayed an increased NK activity, but no lymphokine-activated killer (LAK) activity. However, the subsequent in vitro culture of PBMC with high-dose IL-2 induced the generation of a potent LAK activity, which was mediated by an expanded population of CD3+, CD8+ T cells. Finally lymphocytes that had been isolated after cytokine therapy were cloned, in the presence of low-dose phytohemagglutin, immediately or following culture with IL-2. Clones derived from LAK cells expanded in vitro had predominantly a CD3+, CD8+ immunophenotype, whereas those raised from freshly separated lymphocytes were either CD3+, CD4+ or CD3+, CD8+ in equal proportions. Most of the above clones were poorly or not at all cytolytic against NK-sensitive or NK-resistant targets. In contrast, the few NK clones obtained (CD3-, CD56+) lysed all targets with high efficiency.

Normal levels of soluble CD4 in sera from patients with juvenile chronic arthritis.

Sera from a group of patients with juvenile chronic arthritis (JCA) were tested for soluble CD4 (sCD4). In most cases normal levels of the molecule were detected independent of disease activity. Similar results were obtained when sera from a population of adult rheumatoid arthritis (RA) patients were analyzed. Immunophenotypic studies of circulating mononuclear cells from seven JCA patients with active disease showed that T cells did not express activation markers. Finally, preliminary experiments showed that sCD4 levels were high in the synovial fluids from 3 RA patients as compared with paired serum determinations.

Targeting of "T" lymphocytes against human hepatoma cells by a bispecific monoclonal antibody: role of different lymphocyte subsets.

In an attempt to construct bispecific monoclonal antibodies (bimAbs) able to target cytotoxic T lymphocytes against human hepatoma cells, an HGPRT-deficient mutant of the Hepama-6 hybridoma, which produces an antihuman-hepatoma mAb, was directly fused with splenocytes from Balb/C mice immunized by a polyclonal cytotoxic T-cell line. Hybrid hybridomas were selected in HAT medium, and their supernatants were directly screened for the ability to induce IL-2-cultured cytotoxic T lymphocytes to kill hepatoma cells in a 51Cr-release assay. The selected hybrid hybridoma, termed DQ-33, secretes a bimAb, which reacts with a CD3-associated determinant. When resting peripheral-blood lymphocytes were used as effector cells, virtually no cytolytic activity could be induced by DQ-33, whereas phytohemagglutinin-activated lymphocytes that had been expanded in vitro in IL-2-containing medium could be efficiently targeted against hepatoma cells. Targeting by DQ-33 bimAb was analyzed on different subsets of IL-2-cultured lymphocytes. It was evident that CD+4-8+ TCR alpha/beta+ and CD3+4-8-TCR gamma/delta+ lymphocytes were efficiently induced by bimAb to lyse human hepatoma cells, whereas no induction of cytolysis could be observed when CD3 + 4 + 8-TCR alpha/beta+ cells were used as effectors. DQ-33 bimAb was also able to induce lymphokine secretion (IL-2, GM-CSF and TNF-alpha) by all the different subsets of lymphocytes analyzed in the presence of target cells expressing the relevant antigen, independent of the expression of cytolytic activity.

A novel surface molecule expressed by long-term cultured T and natural killer cells is involved in cell activation.

Two monoclonal antibodies (mAb), termed ED6 and LD6, were obtained by immunizing mice with cytotoxic T cell lines expressing the T cell receptor (TcR) gamma/delta. These mAb were selected according to their ability to trigger the cytolytic program of the immunizing cell lines in a redirected killing assay. Both mAb recognized molecule(s) expressed on the surface of most long-term cultured TcR gamma/delta +, TcR alpha/beta + and CD3-CD16+ lymphocytes, while it was absent on resting peripheral blood lymphocytes. In addition both mAb reacted with neoplastic B cell lines, Epstein-Barr virus-transformed B cell lines, small cell lung cancer and glioma cell lines, while no surface reactivity was detected on ovarian, breast, colon and non-small cell lung cancer lines. The functional activity of these mAb was studied by two cytolytic assays. Both mAb were able to trigger the cytolytic program of CD3+TcR gamma/delta + polyclonal cell lines and of a CD3-CD16+ NK cell clone against the murine mastocytoma target cell line P815 (Fc receptor+) in a 4-h 51Cr-release assay. In addition, ED6 and LD6 hybridomas were lysed by TcR gamma/delta + effector cells while other hybridomas (obtained from the same fusion) were not lysed. ED6 and LD6 mAb (in the presence of submitogenic doses of the phorbol 12-myristate 13-acetate) also induced the secretion of interleukin 2 by ED6/LD6+ T cell clones expressing TcR gamma/delta or alpha/beta. mAb-induced surface antigen modulation experiments showed that the antigenic determinant recognized by ED6 and LD6 co-modulated, thus indicating that the two mAb probably recognize the same or closely associated molecules. The molecular characteristics of the antigen recognized by the mAb were investigated by Western blot analysis. The LD6 mAb recognized a major band of approximately 65 kDa, both under nonreducing and reducing conditions. These data indicate that ED6 and LD6 mAb recognize a novel non-lineage-specific activation antigen which is involved in the induction of the functional program of long-term cultured T or natural killer cells.

Bispecific monoclonal antibodies directed to CD16 and to a tumor-associated antigen induce target-cell lysis by resting NK cells and by a subset of NK clones.

CD16 surface antigens represent activatory molecules in CD3-16+ NK cells. In order to target NK cells against relatively NK-resistant ovarian carcinomas, we used an anti-CD16 monoclonal antibody (MAb) (VD4), together with an anti-ovarian carcinoma-associated antigen (MOV19), to construct biMAbs. To this end, hybrid hybridomas were generated by fusing a TK-deficient VD4 hybridoma mutant with a HGPRT-deficient MOV19 hybrid. Supernatants from hybrid hybridomas that had been selected in HAT medium were screened for their ability to induce a CD3-16+ NK clone to lyse an MOV19+ ovarian carcinoma cell line in a 4-hr 51Cr-release assay. The NMB.45 hybrid hybridoma secreted a biMAb which triggered lysis of MOV19+ but not of MOV19- target cells. Some degree of target cell lysis was also observed with MOV19 MAb (due to ADCC mechanisms), while the VD4 MAb was ineffective. HPLC fractionation of MAbs secreted by the hybrid hybridoma made it possible to identify 4 different peaks, one of which appeared to contain functional biMAb molecules. HPLC-purified biMAb (100 ng/ml) induced resting PBL to lyse the "NK-resistant" IGROVI ovarian carcinoma cell line. Fresh MOV19+ tumor cells were also lysed, although with lower efficiency. When IL-2-activated lymphocytes were used as a source of effectors, biMAb caused only minor increases in the IL-2-induced cytolytic activity. Further analyses of the effect of biMAb were performed at the clonal level. Among CD3-16+ NK cell clones, a clear enhancing effect could be observed only in GL183+ but not in GL183- clones. In CD3+ cytotoxic clones a triggering effect could be detected in one out of 4 TCR gamma/delta+ clones but not in TCR alpha/beta+ clones.

Activated CD3- CD16+ natural killer cells express a subset of the lymphokine genes induced in activated alpha beta + and gamma delta + T cells.

In this study we analysed the potential of highly purified polyclonal TcR alpha beta+, TcR gamma delta + and CD3- NK cells, to produce lymphokines in response to mitogenic stimulation. RNA hybridizations were performed to detect with high sensitivity the induction of multiple lymphokine genes. Upon stimulation with lectin and phorbol ester TcR gamma delta + lymphocytes expressed the same set of lymphokine genes as the TcR alpha beta + lymphocytes expressed the same set of lymphokine genes as the TcR alpha beta + lymphocytes, which included IL-2, -3, -4, -5, GM-CSF, TNF alpha and beta, IFN gamma. In contrast, a more limited set of lymphokine genes (GM-CSF, TNF alpha and beta, IFN gamma) was induced in activated CD3- NK cells, thus indicating that this subpopulation of cells may display different regulatory functions, with respect to CD3+ T lymphocytes.