Role of tumor necrosis factor-α in the regulation of activated synovial T cell growth: Downregulation of synovial T cells in rheumatoid arthritis patients

To characterize the role of tumor necrosis factor (TNF)-α in regulating synovial T cell growth, cell cycle progression associated with TNF-α in mitogen-activated synovial T cells of patients with rheumatoid arthritis (RA) were analyzed. After mitogen stimulation, the majority of synovial T cells in RA patients accumulated in S-phase. Anti-human TNF-α monoclonal antibody and soluble recombinant human TNF receptor (rhTNFR) can block S-phase accumulation. Furthermore, synovial fluid (SF) from RA patients was able to inhibit the proliferation of these S-phase-accumulated T cells. These data indicate that TNF-α could regulate activated synovial T cell growth by driving them into S-phase. Combined with the activities of other components of SF, TNF-α seems to play an important role in down-regulating activated synovial T cells in RA patients. In addition, the elevated level of soluble TNFR in the SF of disease-active RA patients is believed to be associated with the promotion of synovial T cell responses.     

T cells which do not express membrane tumor necrosis factor-α activate macrophage effector function by cell contact-dependent signaling of macrophage tumor necrosis factor-α production

Previous studies have suggested that T cell contact-dependent signaling of macrophages (MΦ) is mediated by membrane tumor necrosis factor-α (memTNF-α), based on the observation that anti-TNF-α could inhibit T cell-mediated MΦ activation. The current report confirms that anti-TNF-α does inhibit activation of interferon-γ (IFN-γ)-primed MΦ by paraformaldehyde-fixed activated T cells. However, the involvement of membrane molecules other than memTNF-α in the contact-dependent signaling is suggested by two lines of evidence. First, the T_H2 clone, AK8, displayed neither secreted TNF-α/β nor memTNF-α/β detectable by bioassay or immunofluorescence. Nonetheless, AK8 cells were equally effective, on a per cell basis, in contact-dependent signaling of MΦ activation as T_H2 and T_H1 cells which do express memTNF-α. Second, the expression of memTNF-α by the T_H clone, D10.G4, is maximal 24 h after activation, whereas the ability of this clone to activate MΦ is maximal at 6–8 h of activation and declines thereafter. Since TNF-α is known to play a critical role in activation of MΦ effector function, it was hypothesized that T cell membrane components other than memTNF-α might signal MΦ production of TNF-α, thus allowing autocrine TNF-α stimulation of MΦ effector function. In support of this, it is demonstrated that paraformaldehyde-fixed activated T_H2 cells can induce de novo production and release of TNF-α by MΦ. This effect was not an artifactual result of paraformaldehyde fixation since paraformaldehyde-fixed resting T cells did not induce TNF-α gene expression. Previous studies have demonstrated a role for autocrine TNF-α stimulation in LPS induction of effector function in recombinant IFN-γ-primed MΦ. The current study confirms that TNF-α plays a critical role in T cell contact-dependent signaling of MΦ but indicates that memTNF on the T cells may not be a sine qua non factor for contact-dependent signaling. The data suggest that other T cell membrane molecules contribute to activation of MΦ effector function by stimulation of MΦ TNF-α production.     

Mice heterozygous for a deletion of the tumor necrosis factor-α and lymphotoxin-α genes: biological importance of a nonlinear response of tumor necrosis factor-α to gene dosage

The tumor necrosis factors (TNF-α and lymphotoxin, or LT-α) are important mediators of the immune and inflammatory responses, and it has been proposed that a positive feedback loop could boost the expression of the TNF to sufficiently high levels to fend off infections. To investigate this phenomenon and its biological consequences, we have generated LT-α/TNF-α knockout mice and compared mice having one or two functional LT-α/TNF-α alleles. In response to lipopolysaccharide (LPS) stimulation, TNF-α levels in the circulation or in the supernatant of macrophage cultures were 20- to 100-fold lower in heterozygous samples than in their wild-type counterparts. This differential increased with the intensity of stimulation and throughout the response, supporting the involvement of a positive feedback loop. Moreover, the heterozygous mice had an increased bacterial load following Listeria monocytogenes infection and exhibited a bimodal response to the association of D -galactosamine and LPS which was similar to that of wild-type mice at low doses of LPS and more like that of homozygous mutants at high doses. These results therefore establish the biological importance of the nonlinear response of TNF-α levels to gene dosage, and these mice provide a unique tool to study how the propensity to produce TNF can determine the immunological fitness of individuals.     

Partial purification and characterization of a tumor necrosis factor-α converting activity

Tumor necrosis factor (TNF)-α is initially synthesized as an extracellular membrane-associated 26-kDa protein that is further cleaved at Ala⁷⁶-Val⁷⁷ to yield the soluble 17-kDa form. Recently, peptide-hydroxamate metallopro-teinase inhibitors have been reported to block the proteolytic processing of TNF-α, thus suggesting that the putative TNF-α converting enzyme (TACE) is a zinc-dependent metalloendopeptidase. In this report, we characterize a TNF-α converting activity (TACA) that cleaves in vitro the human 26-kDa TNF-α at the physiological processing site. The chromatography steps followed for purification and the use of a panel of proteinase inhibitors indicate that the enzyme responsible for TACA is a membrane glycosylated metalloendopeptidase which is most likely different from the matrix-degrading metalloproteinases. The failure of TACA to process a Val⁷⁷→Gly⁷⁷ precursor mutant emphasizes the importance of hydrophobic residue at P1' position. In addition, TACA is not able to cleave the mouse pro-TNF-α and does not catalyze in vitro the processing of other transmembrane proteins susceptible to metalloproteinase-mediated shedding, such as interleukin-6 or TNF receptors. These studies suggest the existence of an enzyme specific for TNF-α within the metalloproteinases involved in the processing/shedding of a number of cytokines and cytokine receptors.     

Apoptosis of cultured mouse luteal cells induced by tumor necrosis factor-α and interferon-γ

Background: Macrophages and T lymphocytes have been identified in the regressing corpus luteum, and they are thought to participate in structural luteolysis (destruction and removal of luteal cells). Since these cells produce cytokines such as tumor necrosis factor-α (TNF-α) and interferon-γ (IFN-γ), we investigated the effects of these two cytokines on death of luteal cells in vitro. Methods: Mouse luteal cells were cultured in serum-free medium with TNF-α at 0,500,1,000,3,000, or 5,000 U/ml in the presence or absence of IFN-γ at 1,000 U/ml for 3 or 6 days. Then, for estimation of the actions of these cytokines on induction of luteal cell death, we determined the number of viable cells, the percentage of fragmented DNA in total DNA extracted from cultured cells, and the percentage of cells with fragmented DNA in their nuclei by the trypan blue exclusion test, the sensitive micromethod for DNA assay, and the in situ DNA 3′ end labeling method, respectively. DNA fragmentation was also analysed by agarose gel electrophoresis, and cultured cells were examined by electron microscopy. Results:On day 3 of culture, IFN-γ alone at 1,000 U/ml or TNF-α alone at 500–5,000 U/ml did not decrease the number of viable cells, but a combination of IFN-γ (1,000 U/ml) and TNF-α (5,000 U/ml) did. On day 6, IFN-γ alone at 1,000 U/ml or TNF-α alone at 500, 1,000 and 3,000 U/ml did not decrease the number of viable cells, whereas TNF-α alone at 5,000 U/ml did, and combinations of IFN-γ and TNF-α at 1,000, 3,000, and 5,000 U/ml decreased the number of viable cells in proportion to the concentration of TNF-α. On days 3–6 of culture, combinations of IFN-γ and TNF-α that decreased the number of viable cells also increased the percentages of fragmented DNA in total DNA of cultured luteal cells and the percentages of luteal cells with fragmented DNA in their nuclei. Agarose gel electrophoresis of fragmented DNA showed a ladder-like pattern, and electron microscopic examination showed luteal cells with the characteristics of apoptosis. Conclusions: The presence of IFN-γ modulates the ability of TNF-α to induce a reduction in the number of viable cells, although TNF-α alone at high concentrations can induce a reduction in the number of viable cells. © 1995 Wiley-Liss, Inc.     

Cytokine stimulation of T lymphocytes regulates their capacity to induce monocyte production of tumor necrosis factor-α, but not interleukin-10: Possible relevance to pathophysiology of rheumatoid arthritis

Previous studies in the laboratory have shown that the pro-inflammatory cytokine tumor necrosis factor (TNF)-α plays a pivotal role in the pathogenesis of rheumatoid arthritis (RA). The mechanisms involved in regulating monocyte/macrophage cytokine production are not yet fully understood, but are thought to involve both soluble factors and cell/cell contact with other cell types. We and others have previously demonstrated that T cells activated through the T cell receptor/CD3 complex induce monocyte TNF-α production by contact-mediated signals. In this report, we investigated further whether T cells activated by cytokines in the absence of T cell receptor stimulation also regulate monocyte cytokine production. T cells were activated in an antigen-independent manner using the cytokines interleukin (IL)-15 or IL-2 alone, or in combination with IL-6 and TNF-α. Subsequently, T cells were fixed and incubated with monocytes. Fixed, cytokine-stimulated T cells induced monocytes to secrete TNF-α in a dose-dependent manner, but did not induce secretion of IL-10, a potent endogenous down-regulator of TNF-α and other pro-inflammatory cytokines. Stimulation of monocyte TNF-α was markedly inhibited when T cells were physically separated from monocytes within the tissue culture well, confirming that T cell contact is necessary. T cell acquisition of monocyte-activating capacity was shown to be dependent on the period of cytokine stimulation, with T cells activated for 8 days more effective than T cells activated for shorter periods. Addition of interferon-γ or granulocyte/macrophage colony-stimulating factor to the T cell/monocyte cultures enhanced T cell induction of monocyte TNF-α by threefold and ninefold, respectively. The results from this model of cognate interaction suggest that cytokine-stimulated T cells, interacting with macrophages in the rheumatoid synovial membrane, may contribute to the continuous excessive production of TNF-α observed in the RA joint, and to the imbalance of pro-inflammatory cytokines over anti-inflammatory cytokines.     

Involvement of a serine protease in the synthesis of platelet-activating factor by endothelial cells stimulated by tumor necrosis factor-α or interleukin-1α

It has been shown that production of platelet-activating factor PAF, 1-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine) by endothelial cells (EC) stimulated with tumor necrosis factor (TNF)-α and interleukin (IL)-1α requires the synthesis of new proteins and is regulated by anti-proteinases. Here, we demonstrate that TNF-α and IL-1α induce the expression by EC of a 34-kDa diisopropyl fluorophosphate-binding protein immunoprecipitated by an antihuman elastase antibody. This protein is released in the medium and cleaves the chromogenic substrate N-methoxysuccinyl- Ala-Ala-Pro-Val p-anilide, which is specific for elastase. The generation of this elastase-like protein seems to be important for the synthesis of PAF induced by TNF-α and IL-lα, as suggested by the following observations: (a) it precedes the synthesis of PAF; (b) the inhibitors of serine protease and anti-human elastase antibody prevent the synthesis of PAF and the activation of 1-O-alkyl-2-lyso-glycerophosphocholine acetyl-CoA: acetyltransferase, which is a key enzyme of the PAF remodelling pathway; (c) elastase, at concentrations similar to that detectable in the medium of cytokine-activated EC, elicits a rapid synthesis of PAF by EC. High-performance liquid chromatography-tandem mass spectrometric analysis of bioactive PAF demonstrates that the molecular species produced after stimulation of EC with TNF-α, IL-1α or elastase are similar, with a predominant synthesis of the alkyl species. These results indicate that TNF-α and IL-1α stimulate the production of a serine protease which is critical in the activation of enzymes involved in PAF synthesis, suggesting the potential involvement of this mechanism in the regulation of EC functions.     

Interleukin-12 activates human γδ T cells: synergistic effect of tumor necrosis factor-α

γδ T cell populations are known to expand in response to intracellular bacterial infectious agents regardless of previous priming. We have shown previously that soluble factor(s) produced by Mycobacterium-stimulated monocytes activate cord blood γδ T cells to proliferate. In this study, we investigated whether cytokines produced by monocytes are responsible for γδ T cell activation in vitro: interleukin (IL)-1β, IL-6, IL-8, IL-12, tumor necrosis factor (TNF)-α and granulocyte/macrophage colony-stimulating factor were examined. Recombinant human IL-12 stimulated γδ T cells, but not αβ T cells in peripheral blood mononuclear cells, to express CD25 on their surfaces, and to expand in number in vitro. IL-12-primed γδ T cell numbers increased to a greater extent in the culture to which exogenous IL-2 (5 U/ml) was added. Anti-TNF-α monoclonal antibody inhibited IL-12-induced up-regulation of CD25 on γδ T cells, suggesting that endogenous TNF-α may play a role in IL-12-induced activation of γδ T cells. Recombinant TNF-α synergistically augmented IL-12-induced activation of γδ T cells. Furthermore, IL-12 up-regulated TNF receptors on γδ T cells in vitro: TNF-α binding to its receptor induced CD25 expression on the γδ T cells in an autocrine or paracrine fashion, or perhaps both. It also became evident that both IL-12 and TNF-α were produced by mycobacterial lysate-stimulated monocytes. Taken together, these results suggest that upon confrontation with mycobacterial organisms, γδ T cells can be quickly and antigen-nonspecifically activated by soluble factors including IL-12 and TNF-α, both of which are produced by mononuclear phagocytes in response to mycobacterial organisms.     

Expression and cleavage of tumor necrosis factor-α and tumor necrosis factor receptors by human monocytic cell lines upon direct contact with stimulated T cells

Tumor necrosis factor-α (TNF-α) is a potent cytokine in inflammatory processes. A variety of mechanisms that modulate its activity have been described, one being its binding to soluble receptors (sTNFR). In this study, we demonstrate that human monocytic cells such as THP-1 respond to direct contact with a membrane preparation of stimulated HUT-78 cells by producing TNF-α and by releasing sTNFR-p75, but not sTNFR-p55, with different kinetics. TNF-α concentration peaked after 12 h of contact and then decreased, whereas sTNFR-p75 production increased progressively upon cell/cell contact. The decrease in TNF-α concentration is not due to trapping of TNF-α by its soluble receptors or other soluble or cell-associated molecules, but rather to a proteolytic activity associated to THP-1 cells. On the other hand, the increase in sTNFR-p75 release does not result from an increase in the cleavage of pre-existing cell-associated sTNFR-p75 but from an increase in TNFR-p75 expression, immediately followed by the cleavage of its extracellular domain. Phenylmethylsulfonylfluoride, a serine protease inhibitor, has a negative effect on both TNF-α degradation and sTNFR-p75 release by THP-1 cells. Thus, there may be an enzymatic activity associated to THP-1 cells that plays an important role in the neutralization of TNF-α activity both by degrading the molecule and by cleaving its receptors at the cell surface.     

Induction of macrophage nitric oxide production by interferon-γ and tumor necrosis factor-α is enhanced by interleukin-10

Interleukin-10 (IL-10) has been reported to inhibit nitric oxide (NO) synthesis and microbicidal activity of interferon-γ (IFN-γ)-stimulated macrophages (MΦ) by preventing the secretion of tumor necrosis factor-α (TNF-α) which serves as an autocrine activating signal. We have examined the effects of recombinant IL-10 on the capacity of IFN-γ together with exogenous TNF-α to induce NO synthesis by bone marrow-derived MΦ. Under these conditions and in contrast to its reported deactivating potential, IL-10 strongly enhanced NO synthesis measured as nitrite (NO) release (half maximal stimulation at approximately 10 U/ml). IL-10 further increased NO production by MΦ stimulated in the presence of optimal concentrations of prostaglandin E₂, a positive modulator of MΦ activation by IFN-γ/TNF-α. Increased steady state levels of NO synthase mRNA were observed in 4-h IFN-γ/TNF-α cultures and enhanced NO release was evident 24 h but not 48 h after stimulation. These results suggest that the effects of IL-10 on MΦ function are more complex than previously recognized.     

Biphasic control of nuclear factor-χB activation by the T cell receptor complex: role of tumor necrosis factor α

The regulation of nuclear factor (NF)-χB activation by the T cell receptor (TcR)/CD3 complex in primary human T cells has been studied at various times after activation. Only p50 NF-χB protein bound the χB element of interleukin-2 receptor (IL-2R) α chain promoter on resting T cells. However, immediately after TcR/CD3 cross-linking (after approximately 1 h; immediate) binding of p50.p65 heterodimers was observed. p50.c-rel heterodimers were also detected bound to this sequence at early time points (7–16 h; early), and both remained active at later time points (40 h; late) after activation. This regulation takes place mainly at the level of nuclear translocation of p65 and c-rel, at immediate and early time points. Activation also induced c-rel and p105/p50 mRNA synthesis, but not p65 mRNA whose expression was constitutive. Interestingly, all those early and late events, but not the immediate ones, were inhibited by a neutralizing anti-tumor necrosis factor α (TNF-α) monoclonal antibody. Similarly, cycloheximide prevented the p65 and c-rel translocation and consequent formation of active binding heterodimers, at early and late times. Cyclosporin A impaired not only early and late, but also immediate events; however, addition of TNF-α prevented all inhibition. These results indicate that the regulation of NF-χB activation during T cell activation by TcR/CD3 signals is biphasic: TcR/CD3 triggers its immediate translocation, which is transient if no TNF-α is present. TNF-α, therefore, emerges as the main factor responsible for a second phase of NF-χB regulation, controlling both translocation of p65 and c-rel, and new mRNA synthesis for c-rel and p105/p50.     

Nocardia fractions,NLD and NWSM,induce tumor necrosis factor-α secretion in human monocytes: Role of protein kinase C

Nocardia lysozyme digest (NLD) and Nocardia water-soluble mitogen (NWSM) are two fractions derived from Nocardia opaca. In this report, we demonstrated that both fractions elicited significant secretion of tumor necrosis factor-α (TNF-α) in human monocytes. Supernatants from monocytes stimulated with NWSM and low concentrations of NLD displayed a cytotoxic activity against TNF-α-sensitive L929 cells, but supernatants from monocytes stimulated with high concentrations of NLD failed to lyse L929 cells. This latter phenomenon might be related to the secretion of an inactive form of TNF-α or the release of an inhibitor of TNF-α cytotoxic activity. Since it is well established that protein kinase C (PKC) plays a major role in the signaling of several monocyte activators, we investigated the putative role of PKC in cytokine synthesis induced by NLD and NWSM fractions. TNF-α secretion in response to both Nocardia fractions was inhibited by sphingosine, staurosporine and calphostin C, known PKC inhibitors, as well as by a PKC depletion procedure. In addition, NLD and NWSM induced a transient increase in [³H] phorbol dibutyrate binding, which assessed the activation of PKC. The data suggest the involvement of PKC in the signaling of NLD and NWSM fractions leading to the synthesis and the secretion of TNF-α by human monocytes.     

HLA class II molecule-mediated signal transduction mechanism responsible for the expression of interleukin-1β and tumor necrosis factor-α genes induced by a staphylococcal superantigen

Superantigens including staphylococcal enterotoxins (SE) bind to major histocompatibility complex class II molecules and interact with T cells bearing particular vβ chains. SEB was shown to induce the expression of interleukin (IL)-1β and tumor necrosis factor (TNF)-α genes in human peripheral blood monocytes bearing HLA class II molecules. Monoclonal antibodies directed against HLA-DR and -DQ abolished the SEB-induced expression of both the EL-1β and TNF-α genes, suggesting that the HLA class II molecules mediated the gene expression. Therefore, we investigated the signal transduction mechanism responsible for the expression of IL-1β and TNF-α genes induced by binding of SEB to the HLA class II molecules. Three protein tyrosine kinase (PTK) inhibitors, genistein, herbimycin A., and tyrphostin, each of which has a different mechanism of action, strongly inhibited the expression of the monokine mRNA induced by SEB. Analyses of PTK activity revealed that SEB induced a rapid increase of membrane-associated PTK activity and this was blocked by tyrphostin. Furthermore, H-7 inhibited the expression of the monokine mRNA induced by SEB, suggesting the involvement of protein kinase C (PKC) in the signaling pathway. The involvement of PKC was confirmed by the observations that phorbol 12-myristate 13-acetate (PMA), a direct activator of PKC, induced the expression of the monokine mRNA and that SEB evoked the activation of membrane-associated PKC. Both activation of PKC and expression of the monokine mRNA induced by SEB appeared to be inhibited by tyrphostin, but those induced by PMA were not. Taken together, these findings indicate that both PTK and PKC play essential roles in HLA class II molecule-mediated signal transduction elicited by SEB and that PTK activation may precede PKC activation in the signaling pathway.     

Staphylococcal enterotoxin B and tumor-necrosis factor-α-induced relapses of experimental allergic encephalomyelitis: Protection by transforming growth factor-β and interleukin-10

A study was made of the ability of the superantigen staphylococcal enterotoxin B (SEB) to induce relapses of experimental allergic encephalomyelitis (EAE) in SJL mice that had partially or completely recovered from acute EAE. We find that a single injection of 0.05 mg SEB i.v. induces mild relapses in 50% of such mice. In addition, tumor necrosis factor (TNF)-α (0.2 μg, i.p.) also induces EAE relapses in 43% of SJL mice when injected 1–2 months after recovery. SEB does not induce a second relapse if reinjected when Vβ17a⁺ T cells are still partially deleted. In these mice, however, TNF-α is equally effective in inducing relapses as in mice that did not receive SEB previously. We showed earlier that transforming growth factor (TGF)-β and TNF-α have antagonistic effects on experimental autoimmune diseases; e.g., in spontaneously relapsing EAE, TGF-β and anti-TNF were protective, while anti-TGF-β caused disease exacerbation. Interleukin (IL)-10 is also known to counteract certain TNF effects. We now find that both human IL-10 and TGF-β2 lower the incidence of EAE relapses when given simultaneously with SEB or TNF-α. The protective effect of TGF-β is significant only against relapses induced by SEB (reduced to 9%), and that of IL-10 only against relapses induced by TNF (reduced to 0%) with the treatment regimens employed. Neutralizing anti-TGF-β does not increase the incidence of SEB-induced EAE relapses. In contrast, anti-IL-10 increases both the incidence and the severity of such relapses. We conclude that TNF production is probably important in causing EAE relapses, but that other aspects of the SEB-induced reactivation of myelin-specific T cells also contribute. Furthermore, endogenous IL-10 rather than TGF-β production appears to limit the susceptibility to induction of EAE relapses in this model.     

Relevance of IL-1 and TNF gene polymorphisms on interleukin-1β and tumor necrosis factor-α gastric mucosal production

We aimed to evaluate the influence of Helicobacter pylori infection and IL-1/TNF gene polymorphisms on interleukin (IL)-1β and tumor necrosis factor (TNF)-α gastric mucosal production. IL-1β and TNF-α levels in homogenized biopsy specimens taken from the antrum and corpus of 81 patients were measured by enzyme-linked immunosorbent assay. Genomic DNA was typed for the IL1B-511, IL1B+3954, variable number of tandem repeat (VNTR) IL1RN, TNFA-308, TNFA-238, LTA NcoI, and LTA Bsi gene polymorphisms by polymerase chain reaction, restriction fragment length polymorphism, and TaqMan assays. H. pylori infection and CagA/VacA antibody status were determined by Western blot. IL-1β and TNF-α protein levels were significantly higher in the gastric antrum of patients infected with H. pylori compared with uninfected patients [9.54 (5.07–16.28) vs. 4.55 (3.69–8.28) pg IL-1β/mg protein, p = 0.004, and 1.5 (0.7–2.71) vs. 0.63 (0.3–1.26) pg TNF-α/mg protein, p = 0.001]. Among H. pylori-infected individuals, carriers of the IL1RN*2 allele had significantly higher antrum mucosal IL-1β levels than noncarriers [15.97 (9.59–26.6) vs. 10.08 (7.72–13.33), p = 0.008]. No association between gastric mucosal TNF-α levels and genotypes of the TNFA and LTA gene polymorphisms was reported. Our results indicate that the VNTR polymorphism of the IL1RN gene influences IL-1β gastric mucosal production in patients infected with H. pylori.     

Essential role of tumor necrosis factor-α in the differentiation of human tonsil in vivo induced B cells capable of spontaneous and high-rate immunoglobulin secretion

Human tonsils contain B cells capable of spontaneous and high-rate immunoglobulin (Ig) secretion in vitro. These cells are in vivo induced mature B cells, and, as such, they provide an adequate model for studying tonsil B cell differentiation. The present report analyzes the effect of a variety of factors on purified tonsil B cells capable of spontaneous IgG secretion in fetal calf serum (FCS)-containing and serum-free supplemented cultures. Tumor necrosis factor-(TNF)α was found to be important for these B cells to reach the high-rate IgG-secreting stage, as is indicated by the following findings: (a) none of the factors used modified tonsil B cell IgG secretion in FCS-containing cultures; (b) TNF-α (5-20 ng/ml), but not other cytokines or factors including interleukin (IL)-6, was capable of restoring missing IgG production in serum-free supplemented cultures of tonsil B cells; and (c) IgG secretion in FCS-containing cultures was inhibited by the addition of blocking anti-TNF-α antibodies, but not anti-IL-6 antibodies, and this inhibition could be specifically reversed by exogenous TNF-α. TNF-α was actively produced by tonsil B cells (range 120-750 pg/ml) in the presence, but not in the absence, of FCS. The TNF-α inductive effect occurred during the first 12 h of culture and did not require DNA synthesis. These results indicate that the early and endogenous generation of TNF-α seems to be essential for tonsil in vivo induced B cells to differentiate into the high-rate Ig-secreting stage.