Cytoplasmic truncation of the p55 tumour necrosis factor (TNF) receptor abolishes signalling, but not induced shedding of the receptor.

The mechanistic relationship between the signalling for the TNF effects by the human p55 TNF receptor (hu-p55-TNF-R) and the formation of a soluble form of the receptor, which is inhibitory to these effects, was explored by examining the function of C-terminally truncated mutants of the receptor, expressed in rodent cells. The 'wild-type' receptor signalled for a cytocidal effect when cross-linked with specific antibodies and exhibited spontaneous shedding. Shedding of the receptor was not affected by TNF but was markedly enhanced by 4 beta-phorbol-12-myristate-13-acetate (PMA). Receptor mutants with 53%, 83% and 96% C-terminal deletions could not signal for the cytocidal effect. Furthermore, they were found to associate with the endogenous rodent receptors, interfering with their signalling. Yet even the deletion of 96% of the intracellular domain did not abolish shedding of the receptor in response to PMA. These findings suggest that signalling and shedding of the p55 TNF-R are mechanistically distinct.     

IL-1 and TNF transmodulate epidermal growth factor receptors by a protein kinase C-independent mechanism

The effect of the human rIL-1 alpha and rTNF-alpha on the binding of 125I-labeled epidermal growth factor ([125I]EGF) to its receptor (EGF-R) has been studied in human gingival fibroblasts (HuGi). Incubation of these cells with recombinant cytokines at 37 degrees C caused a rapid, dose-dependent decrease in their ability to subsequently bind subsaturating levels of [125I]EGF at 4 degrees C. Inhibition was evident at 5 min after addition of cytokines, reached a maximal level (60-70% reduction) after 15 to 30 min, and declined thereafter. Normal EGF binding was attained by 2 h. Half-maximal inhibition of EGF binding occurred at 10 pM IL-1 and 50 pM TNF. The two cytokines were not additive in their effect. Competition experiments at 4 degrees C showed that the cytokines did not interact directly with EGF-R; Scatchard analysis of binding of [125I]EGF to HuGi after treatment with IL-1 and TNF revealed an increase in EGF-R Kd from 0.75 nM to 2.9 nM with no change in receptor number. The effect of IL-1 and TNF on EGF-R was compared with that of the tumor-promotor PMA which is known to "transmodulate" EGF-R affinity by activating protein kinase C which then phosphorylates EGF-R. PMA caused a greater inhibition of EGF binding to HuGi (80 to 85% inhibition; ED50 = 500 pM), and recovery of binding was much slower. Importantly, in HuGi made deficient in protein kinase C by prolonged incubation with PMA, addition of fresh PMA no longer affected EGF binding, while the response to IL-1 and TNF was intact. Cytokine- but not PMA-mediated EGF-R transmodulation was partially reversed by treatment of the cells with millimolar concentrations of the kinase inhibitor amiloride. HuGi were incubated with H3 32PO4, stimulated with PMA or cytokines, and EGF-R were immunoprecipitated; IL-1 and TNF, like PMA, caused a 2- to 5-fold increase in receptor phosphorylation. We conclude that occupation of IL-1 and TNF-R activates a protein kinase, distinct from kinase C, for which EGF-R is a substrate.     

Tumor necrosis factor receptors and cytocidal activity are down-regulated by activators of protein kinase C

Human HeLa cells and murine L(S) cells are highly sensitive to the cytocidal activity of tumor necrosis factor (TNF) when simultaneously treated with the inhibitor of protein synthesis cycloheximide. This cytocidal activity of TNF was inhibited up to 90% in both cell lines after a 15-60-min pretreatment with 3-10 ng/ml of phorbol 12-myristate 13-acetate (PMA). This inhibition was long lasting for HeLa cells but transient for L(S) cells. The protection afforded by PMA was most effective when the cells were pretreated with this phorbol ester, but it decreased when PMA was added together with TNF or after TNF addition. This finding suggested that PMA interfered with one of the early steps in the mechanism of action of TNF. A pretreatment with the calcium ionophore A23187 also reduced the cytocidal activity of TNF in both HeLa and L(S) cells to about the same extent. Treatment of these cells with either PMA or A23187 significantly decreased the binding of 125I-TNF to cell surface receptors. This decrease paralleled the time course and dose-response of the inhibition of cytocidal activity. In addition, treatment of HeLa cells with 1-oleyl-2-acetyl-glycerol (OAG) also induced a rapid loss of TNF binding capacity. Since OAG, PMA, and A23187 are all activators of protein kinase C (Ca2+/phospholipid-dependent enzyme), these results suggest that this kinase is involved in modulation of TNF sensitivity. Furthermore, depletion or inhibition of protein kinase C antagonized PMA-induced effects on TNF cytotoxicity and binding to receptors. Internalization of bound TNF was not significantly affected by PMA treatment, and Scatchard analysis of binding data indicated that PMA decreased TNF receptor binding affinity rather than the number of TNF-binding sites. These findings suggest that protein kinase C may have a physiological role in mediating TNF sensitivity.     

Repression of farnesoid X receptor during the acute phase response

The acute phase response is associated with changes in the hepatic expression of genes involved in lipid metabolism. Nuclear hormone receptors that heterodimerize with retinoid X receptor (RXR), such as thyroid receptors, peroxisome proliferator-activated receptors, and liver X receptors, modulate lipid metabolism. We recently demonstrated that these nuclear hormone receptors are repressed during the acute phase response induced by lipopolysaccharide (LPS), consistent with the known decreases in genes that they regulate. In the present study, we show that LPS significantly decreases farnesoid X receptor (FXR) mRNA in mouse liver as early as 8 h after LPS administration, and this decrease was dose-dependent with the half-maximal effect observed at 0.5 microg/100 g of body weight. Gel-shift experiments demonstrated that DNA binding activity to an FXR response element (IR1) is significantly reduced by LPS treatment. Supershift experiments demonstrated that the shifted protein-DNA complex contains FXR and RXR. Furthermore, the expression of FXR target genes, SHP and apoCII, were significantly reduced by LPS (70 and 60%, respectively). Also, LPS decreases hepatic LRH expression in mouse, which may explain the reduced expression of CYP7A1 in the face of SHP repression. In Hep3B human hepatoma cells, both tumor necrosis factor (TNF) and interleukin-1 (IL-1) significantly decreased FXR mRNA, whereas IL-6 did not have any effect. TNF and IL-1 also decreased the DNA binding activity to an IR1 response element and the expression of SHP and apoCII. Importantly, TNF and IL-1 almost completely blocked the expression of luciferase activity linked to a FXR response element promoter construct transfected into Hep3B cells. Together with our earlier studies on the repression of RXRs, peroxisome proliferator-activated receptors, LXRs, thyroid receptors, constitutive androstane receptor, and pregnane X receptor, these results suggest that decreases in nuclear hormone receptors are major contributors to the decreased gene expression that occurs in the negative acute phase response.     

Differential induction of interleukin-6 production in monocytes, endothelial cells and smooth muscle cells.

Studying the production of IL-6 (interleukin-6) by monocytes, endothelial cells and smooth muscle cells we observed that cytokine inducers like IL-1, TNF alpha (tumor necrosis factor alpha), LPS (lipopolysaccharide), SAC (Staphylococcus Aureus Cowan 1) and PMA could be divided roughly into two categories. Bacterial products such as LPS or SAC have a potent IL-6 inducing effect on monocytes and minor or no effect on endothelial- and smooth muscle cells. The other category comprising IL-1, TNF alpha and PMA induces IL-6 production in endothelial- and smooth muscle cells. Only IL-1 induces IL-6 production in monocytes as well as in endothelial cells and smooth muscle cells. In addition to IL-6, also IL-1 and TNF alpha are produced by monocytes however with different kinetics. None of the stimuli had any inhibitory effect on IL-6 production with the exception of PMA. Whereas PMA induced IL-6 production in endothelial cells and it potentiated the induction of IL-6 by IL-1 in these cells, it inhibited LPS-stimulated IL-6 production in monocytes. In line with the effects of PMA, staurosporin induced IL-6 production in monocytes and it inhibited IL-1 driven IL-6 production by endothelial cells.     

Down-regulation of mannose receptor activity in macrophages after treatment with lipopolysaccharide and phorbol esters

We have investigated the effects of LPS and PMA on the expression of functional mannose receptors in rat bone marrow-derived macrophages. After 48 h of treatment with LPS (10 ng/ml) and PMA (100 nM), mannose receptor activity was reduced by 70 to 80%. The effect of these agents on receptor activity was not reversible, and activity continued to decline after the agents were removed. Pretreatment of cells with dexamethasone was effective in blocking the LPS/PMA-induced down-regulation. Serine protease inhibitors did not block the reduction in receptor activity, suggesting that proteolysis is not involved in receptor down-regulation. LPS/PMA treatment did not increase turnover of the receptor. Ligand uptake studies showed that the total capacity of the uptake system was reduced by 80%, although the Kuptake was unaffected. Binding of 125I-mannose-BSA to intact macrophages showed a 70% decrease in surface receptor activity after treatment with LPS/PMA. LPS/PMA treatment had no effect on total receptor synthesis as quantitated by immunoprecipitation of metabolically labeled receptor. However, binding of metabolically labeled receptor to mannose-Sepharose, and binding of 125I-mannose-BSA to immunoprecipitated receptor revealed that intracellular plus surface binding sites were reduced to approximately 30% after LPS/PMA treatment. These results suggest that LPS/PMA treatment of macrophages results in an inactivation of mannose receptors with no effect on receptor turnover or biosynthesis.     

Modulation of two forms of tumor necrosis factor receptors and their cellular response by soluble receptors and their monoclonal antibodies.

Recently, two different receptors for human tumor necrosis factor (TNF) with molecular masses of 60 kDa (p60) and 80 kDa (p80) have been identified. In this report, we investigated the effect of the soluble forms of these receptors and monoclonal antibodies against them on ligand interaction, receptor down-regulation, and mediation of cellular response in U-937 cells. Our results indicate that p60 and p80 constitute 20-30 and 60-80% of the total TNF-binding sites on U-937 cells, respectively. However, by cross-linking, only the p80 form of the receptor could be detected. In contrast to unlabeled TNF, the anti-p60 and anti-p80 antibodies together only partially inhibited ligand binding, and this inhibition was not additive. Lack of additive inhibition of binding was found to be not due to stereo-chemical hindrance. TNF binding to cells can be completely displaced by soluble forms of either the p60 or p80 receptor. However, 100-fold more of the p80 than the p60 form of the soluble receptor is needed for equivalent displacement. Under optimum conditions, TNF and the anti-p80 and anti-p60 antibodies down-regulated 30, 80, and 20% of the TNF receptors, respectively. The anti-p60 and anti-p80 antibodies down-regulated not only their own receptors, but also reciprocal receptors, suggesting a cross-communication between the p60 and p80 forms of the TNF receptor. In spite of inhibiting as much as 80% of TNF binding, none of the receptor antibodies significantly inhibited the cytotoxic response to TNF in U-937 cells. Soluble forms of both receptors, however, completely abrogated the cellular response to TNF. Thus, overall, our results indicate that the antibodies against both receptors together inhibit the majority of the receptor-ligand interaction without any significant effect on the biological response to TNF.     

Downregulation of tumor necrosis factor receptors of macrophages by interferons and interleukin-1. Role of protein kinase C activation

Freshly harvested murine peritoneal macrophages and a line of transformed murine macrophages (RAW) were used in experiments designed to investigate the effect of different interferons (IFN) and interleukin-1 (IL-1) on tumor necrosis factor (TNF) receptors. Low concentrations of IFN-gamma or somewhat higher concentrations of IFN-alpha drastically downregulated the TNF receptors of RAW cells. A similar, but less pronounced, downregulation of TNF receptors was observed in peritoneal macrophages treated with these IFNs. This downregulation could not be accounted for by an induction of TNF secretion. Furthermore, IFN-alpha and gamma interacted synergistically in downregulating TNF receptors of RAW cells. IL-1 also downregulated TNF receptors. When RAW cells were treated with inhibitors of protein kinase C, the downregulation of TNF receptors by IFNs or IL-1 was reversed, and TNF binding increased up to 2-fold over that of untreated cells. Such increase was also observed in RAW cells treated only with the inhibitor of protein kinase C, staurosporine. However, TNF receptors decreased in peritoneal macrophages treated with staurosporine. This finding was explained by activation of macrophages by staurosporine, which induced secretion of TNF. These findings indicate that protein kinase C activity regulates TNF receptors in macrophages.     

Modulation of lipopolysaccharide-induced production of tumor necrosis factor, interleukin 1, and interleukin 6 by synthetic precursor Ia of lipid A

Abstract Endotoxin (lipopolysaccharide, LPS) induces the production of mediators of inflammation, which exerts pathophysiological effects such as fever or shock in mammals. In the present study we have investigated the modulation of LPS by the synthetic non-active tetraacylated precursor Ia of lipid A (compound 406) in the induction of tumor necrosis factor (TNF), interleukin 1 (IL-1) and interleukin 6 (IL-6) in human peripheral blood mononuclear cells (PBMC) and in human peripheral blood monocytes (PBMo). PBMC stimulated with LPS released TNF in a concentration dependent manner. Release of biologically active TNF, IL-1 and IL-6 was first detectable 4 h after LPS stimulation. Compound 406 alone in all concentrations tested did not induce TNF, IL-1 or IL-6 release, intracellular TNF or IL-1β, or mRNA for TNF or IL-1. Added to PBMC 1 h before LPS compound 406 enhanced or suppressed TNF release and suppressed IL-1 and IL-6 release depending on the ratio of concentrations between stimulator (LPS) and modulator (compound 406). In contrast to LPS stimulation alone TNF, IL-1 and IL-6 release in presence of compound 406 was delayed and first detectable after 6 to 8 h. Compound 406 was able to suppress LPS-induced intracellular TNF and IL-1β in PBMC. Added to PBMo 1 h before LPS it totally inhibited the production of mRNA for TNF and IL-1. When added to PBMC 1 h after LPS, TNF release was suppressed in a concentration-dependent way and release of biologically active TNF, IL-1 and IL-6 could again be detected for the first time after 4 h. Compound 406 was not able to inhibit phorbol 12-myristate 13-acetate (PMA)-induced TNF and IL-1 release in PBMo which suggests that its modulating effect is LPS-specific. This study provides evidence that the modulating effect of compound 406 on the LPS induction of TNF, IL-, 1 and IL-6 could be due to competitive binding.     

Tumor necrosis factors alpha and beta differ in their capacities to generate interleukin 1 release from human endothelial cells

The capacity of the tumor necrosis factors, TNF-alpha and TNF-beta, products of activated macrophages and lymphocytes, respectively, to stimulate interleukin 1 (IL-1) release from endothelial cells derived from human umbilical veins was examined in vitro. Recombinant TNF-alpha caused IL-1 release by 4 hr with maximal levels of 17 U/ml by 24 hr; half-maximal stimulation occurred at approximately 80 pM. In contrast, recombinant TNF-beta was a relatively poor stimulus for IL-1 release. Even at concentrations as high as 600 pM, only 3 U of IL-1/ml were recovered; maximal IL-1 release (10 to 12 U/ml) required up to 5 nM TNF-beta. Natural, glycosated human TNF-beta was comparable in activity to recombinant TNF-beta. TNF-beta did not directly inhibit the IL-1 comitogenesis assay, nor was there evidence that TNF-beta induced the release of an IL-1 inhibitor, in that supernatants generated in the presence of TNF-beta did not inhibit thymocyte proliferation to a recombinant IL-1 standard. Binding of the recombinant TNF to endothelial monolayers was assessed by using [125I]TNF-alpha in competition studies with cold TNF-alpha and TNF-beta. Binding of TNF-alpha was half-maximal at 80 pM with an average of 664 receptors/cell and Kd = 0.043 nM. Although TNF-beta was capable of fully competing for [125I]TNF-alpha binding, half-maximal binding occurred at 800 pM TNF-beta. These data suggest that the TNF receptors on human endothelial cells may reflect the structural differences between these two homologous cytokines.     

Selective decrease in cell surface expression and mRNA level of the 55-kDa tumor necrosis factor receptor during differentiation of HL-60 cells into macrophage-like but not granulocyte-like cells.

Expression of the two known receptors for TNF was studied in the promyelocytic leukemia cell line HL-60 before and after differentiation of the cells along the granulocyte lineage (induced by incubation with retinoic acid), or along the macrophage lineage (induced by incubation with the phorbol diester, PMA). The extent of inhibition of TNF binding by receptor-specific antisera, as well as the size of the complexes formed after cross-linking TNF to its receptors on intact cells, indicated that both receptor species were expressed on the surface of the undifferentiated HL60 cells. Differentiation into granulocyte-like cells resulted in some increase in TNF binding. The increase was apparently due to enhanced expression of the 75-kDa TNF-R, whereas the amounts of the 55-kDa TNF-R did not change significantly. In contrast, in HL-60 cells induced to differentiate into macrophage-like cells, expression of the 55-kDa TNF-R species was completely abolished. The pattern of TNF-R expression in the differentiated HL-60 cells was similar to that observed in leukocytes isolated from peripheral blood: on granulocytes, there were about equal amounts of both receptor species, whereas on monocytes the 75-kDa receptor was predominant. The loss of 55-kDa receptors during differentiation of HL-60 cells into macrophage-like cells was accompanied by a pronounced decrease in the level of the mRNA for that receptor, suggesting that at least part of the change in TNF-R expression is due to mechanisms that control the amounts of receptor mRNA. Although little is yet known regarding the functional differences between the two receptor species, marked changes in the pattern of their expression, as observed during HL-60 cell differentiation, are likely to alter the kind of response of the cells to TNF and may therefore play an important role in the coordination of TNF effects in the organism.     

Modulation of the epidermal growth factor receptor by brain-derived growth factor in Swiss mouse 3T3 cells

Incubation of Swiss mouse 3T3 cells at 37 degrees C with bovine brain-derived growth factor (BDGF) decrease the cell surface 125I-EGF binding activity of these cells by 70-80%. This down-modulation of the EGF receptor by BDGF was time, temperature, and dose dependent. Scatchard plot analysis indicated that BDGF binding led to a selective decrease in the number of high-affinity EGF receptors. The BDGF-induced down-modulation of the EGF receptor was completely blocked by protamine, a potent inhibitor of receptor binding and mitogenic activities of BDGF. BDGF down-modulated the EGF receptor in phorbol myristic acetate (PMA)-pretreated cells, as well as in control cells. Furthermore, PMA-pretreated cells responded mitogenically to BDGF, whereas PMA itself failed to stimulate the mitogenic response of PMA-pretreated cells. This BDGF-induced down-modulation of the EGF receptor in PMA-desensitized cells suggests that BDGF down-regulates the EGF receptor by a mechanism distinct from that of PMA. Incubation of cells with compounds which are known to inhibit pinocytosis blocked the down-modulation induced either by BDGF or by platelet-derived growth factor (PDGF) but had no effect on the PMA-induced down-modulation. Incubation of cells with inhibitors of receptor recycling enhanced the BDGF-induced down-modulation of the EGF receptor. These results suggest that BDGF and PDGF induce down-modulation of the EGF receptor by increasing the internalization of cell surface high-affinity receptors and that the internalization process may not be required for down-modulation induced by PMA.     

Cell density-dependent regulation of cell surface expression of two types of human tumor necrosis factor receptors and its effect on cellular response

Tumor necrosis factor (TNF) is a multipotential cytokine known to regulate the growth of a wide variety of normal and tumor cells. It has been shown that the density of cells in culture can modulate the growth regulatory activities of TNF, the mechanism of which, however, is not understood. In this report, we investigated the effect of cell density on the expression of TNF receptors. The receptors were examined on epithelial cells (e.g., HeLa), which primarily express the p60 form, and on myeloid cells (e.g., HL-60) known to express mainly the p80 form. We observed that binding of TNF to both cell lines decreased with increase in cell density. Scatchard analysis of binding on HeLa and HL-60 cells revealed a 4- to 5-fold reduction in the number of TNF receptors without any significant change in receptor affinity in both cell types at high density. The decrease in TNF receptor numbers at high cell density was also observed in several other epithelial and myeloid cell lines. The downmodulation at high cell density was unique to TNF receptors, since minimum change in other cell surface proteins was observed as revealed by fluorescent activated cell sorter analysis. Neutralization of binding with antibodies specific to each type of the receptors revealed that both the p60 and p80 forms of the TNF receptor were equally downmodulated. A decrease in leucine incorporation into proteins was observed with increase in cell density, suggesting a reduction in protein synthesis. Since inhibition of protein synthesis by cycloheximide also leads to a decrease in TNF receptors, it is possible that the density-dependent reduction in TNF receptor number is due to an overall decrease in protein synthesis. The density-dependent decrease in TNF receptors was accompanied by a decrease in intracellular reduced glutathione levels. A reduction in the number of receptors on TNF sensitive tumor cells induced by cell-density correlated with increase in resistance to the cytokine.     

Effect of cell cycle on the regulation of the cell surface and secreted forms of type I and type II human tumor necrosis factor receptors

The cell cycle has been shown to regulate the biological effects of human tumor necrosis factor (TNF), but to what extent that regulation is due to the modulation of TNF receptors is not clear. In the present report we investigated the effect of the cell cycle on the expression of surface and soluble TNF receptors in human histiocytic lymphoma U-937. Exposure to hydroxyurea, thymidine, etoposide, bisbensimide, and democolcine lead to accumulation of cells primarily in G₁/S, S, S/G₂/M, G₂/M, and M stages of the cell cycle, respectively. Whilie no significant change in TNF receptors occurred in cells arrested in G₁/S or S/G₂ stages, about a 50% decrease was observed in cells at M phase of the cycle. Scatchard analysis showed a reduction in receptor number rather than affinity. In contrast, cells arrested at S phase (thymidine) showed an 80% increase in receptor number. The decrease in the TNF receptors was not due to changes in cell size or protein synthesis. The increase in receptors, however, correlated with an increase in total protein synthesis (to 3.8-fold of the control levels). A proportional change was observed in the p60 and p80 forms of the TNF receptors. A decrease in the surface receptors in cells arrested in M phase correlated with an increase in the amount of soluble receptors. The cellular response to TNF increased to 8- and 2-fold in cells arrested in G₁ and S phase, respectively; but cells at G2/M phase showed about 6-fold decrease in response. In conclusion, our results demonstrate that the cell cycle plays an important role in regulation of cell-surface and soluble TNF receptors and also in the modulation of cellular response. © 1995 Wiley-Liss, Inc.     

Synthesis and release of platelet-activating factor by human vascular endothelial cells treated with tumor necrosis factor or interleukin 1 alpha

Human endothelial cells synthesize large amounts of platelet-activating factor (PAF) after 30-min treatment with recombinant tumor necrosis factor (TNF). Synthesis of PAF peaks at 4-6 h, whereas in endothelial cells treated with interleukin 1 alpha (IL-1) it peaks at 8-12 h. More than twice as much PAF is synthesized in response to optimal concentrations of TNF than in response to IL-1. However, PAF synthesis is stimulated by lower molar concentrations of IL-1 than TNF. About 30% of PAF produced in response to either TNF or IL-1 is released into the medium, whereas approximately 70% remains cell-associated. Experiments with labeled precursors show that PAF is synthesized de novo in response to TNF. This activity of TNF is inhibited by treating endothelial cells with the inhibitors of protein or RNA synthesis cycloheximide or actinomycin D. This finding may be explained by the observation that TNF induces in endothelial cells an acetyltransferase required for PAF synthesis. The induction of this enzymatic activity precedes the peak of PAF synthesis in TNF-treated cells. After prolonged incubation with either TNF or IL-1, endothelial cells no longer respond to the same monokine, but are still capable of producing PAF when treated with the other monokine. The finding that these monokines do not show reciprocal tachyphylaxis in endothelial cells may be explained by their binding to different receptors. In cells treated simultaneously with different concentrations of TNF and IL-1, PAF synthesis is stimulated in an additive rather than synergistic way. This suggests that PAF is synthesized by the same pathway in response to TNF or IL-1.     

Discoordinate expression of stromelysin, collagenase, and tissue inhibitor of metalloproteinases-1 in rheumatoid human synovial fibroblasts. Synergistic effects of interleukin-1 and tumor necrosis factor-alpha on stromelysin expression.

Primary and passaged human synovial fibroblasts isolated from rheumatoid pannus were treated with recombinant interleukin-1 (IL-1) alpha or beta, tumor necrosis factor-alpha (TNF), or phorbol myristate acetate (PMA) to determine the effects of these stimuli on the relative expression of stromelysin, collagenase, and tissue inhibitor of metalloproteinases (TIMP). The steady-state mRNA levels for these genes and glyceraldehyde-3-phosphate dehydrogenase were determined on Northern blots. Immunoblot analyses of the conditioned media using monoclonal antibodies generated against recombinant human stromelysin, collagenase, or TIMP showed that protein levels reflected the corresponding steady-state mRNA levels. The results revealed that 1) stromelysin and collagenase were not always coordinately expressed; 2) IL-1 was more potent than TNF or PMA in the induction of stromelysin expression; 3) neither IL-1 nor TNF significantly affected TIMP expression; 4) PMA induced both metalloproteinase and TIMP expression; and 5) the combination of IL-1 plus TNF had a synergistic effect on stromelysin expression. Dose response and time course experiments demonstrated that the synergistic effect of IL-1 plus TNF occurred at saturating concentrations of each cytokine and lasted for 7 days. In summary, the ability of IL-1 and TNF to preferentially induce stromelysin and collagenase expression, versus TIMP, may define a pivotal role for these cytokines in the pathogenesis of rheumatoid arthritis.