Prostaglandin production by human gingival fibroblasts inhibited by triclosan in the presence of cetylpyridinium chloride

BACKGROUND: The effect of triclosan plus the cationic detergent cetylpyridinium chloride (CPC) was evaluated for prostaglandin inhibition in human gingival fibroblasts. Since triclosan has previously been shown to inhibit proinflammatory cytokine induced prostaglandin E2 (PGE2) production, we wanted to determine if triclosan, in the presence of CPC, could enhance these effects. METHODS: Initial studies determined that both triclosan and CPC were cytotoxic to human gingival fibroblasts in concentrations exceeding 1.0 microg/ml for either agent longer than 24 hours in a tissue culture. Therefore, subsequent studies measuring prostaglandin biosynthesis and cyclooxygenase (COX)-1 and COX-2 mRNA expression were performed in concentrations and times that did not significantly affect cell viability. RESULTS: PGE2 biosynthesis was dose dependently inhibited by both triclosan and triclosan and CPC when challenged by tumor necrosis factor (TNF)-alpha or interleukin (IL)-1beta. At pharmacologically relevant concentrations, triclosan and CPC inhibited IL-1beta-induced PGE2 production to a greater extent than triclosan alone (P = 0.02). Moreover, enhanced COX-2 mRNA repression was observed with triclosan and CPC in comparison to triclosan alone in IL-1beta and TNF-alpha stimulated cells. No effect on COX-1 gene expression was observed. Further analysis of cell signaling mechanisms of triclosan and CPC indicates that nuclear factor-kappa B (NF-kappaB) and not p38 mitogen-activated protein kinase (MAPK) signaling may be impaired in the presence of triclosan and CPC. CONCLUSION: This study indicates that triclosan and CPC are more effective at inhibiting PGE2 at the level of COX-2 gene regulation, and this combination may offer a potentially better anti-inflammatory agent in the treatment of inflammatory lesions in the oral cavity.     

Triclosan reduces microsomal prostaglandin E synthase-1 expression in human gingival fibroblasts

OBJECTIVE: The effect of triclosan (2,4,4'-trichloro-2'-hydroxydiphenyl ether) on the expression of cyclooxygenase-2 (COX-2) and microsomal prostaglandin E synthase-1 (mPGES-1) and on the translocation of the nuclear factor-kappaB (NF-kappaB) in relation to prostaglandin E2 (PGE2) production was investigated in human gingival fibroblasts challenged with tumor necrosis factor alpha (TNFalpha). METHODS: Fibroblasts were established from gingival biopsies obtained from six children. COX-2 mRNA and protein expression was quantified using mRNA quantitation and enzyme immunometric assay kits. mPGES-1 mRNA was analysed by RT-PCR, mPGES-1 protein and NF-kappaB translocation by immunoblotting. PGE2 was determined by radioimmunoassay. RESULTS: The cytokine TNFalpha enhanced the expression of mRNA as well as the protein levels of both COX-2 and mPGES-1 and subsequently the production of PGE2 in gingival fibroblasts. Treatment of gingival fibroblasts with triclosan (1 microg/ml) significantly reduced the stimulatory effect of TNFalpha (10 ng/ml) on the expression of mPGES-1 at both the mRNA and the protein level by an average of 21% and 43%, respectively, and subsequently the production of PGE2 (p<0.01). Triclosan did not, however, affect the translocation of NF-kappaB or the expression of COX-2 in TNFalpha-stimulated cells. CONCLUSION: The results show that triclosan reduces the augmented biosynthesis of PGE2 by inhibiting the mRNA and the protein expression of mPGES-1 in gingival fibroblasts. This finding may partly explain the anti-inflammatory effect of the agent previously reported in clinical studies.     

Effect of Emdogain on human periodontal fibroblasts in an in vitro wound-healing model

OBJECTIVE: The aim of this study was to evaluate the influence of Emdogain (EMD) on cultured gingival fibroblasts, periodontal ligament fibroblasts and dermal fibroblasts, using an in vitro model of wound healing. BACKGROUND: Enamel matrix derivative has been demonstrated to promote periodontal regeneration. However, the precise mechanisms by which this agent acts are still unclear. METHODS: The effect of EMD on proliferation of the cells was studied using subconfluent cultures of gingival fibroblasts and periodontal ligament fibroblasts. The cells were made quiescent overnight and then stimulated with various concentrations of EMD (10, 50, 100 and 150 microg/ml) for 24 h. Negative and positive controls were cells cultured in media containing 0.2% and 10% fetal calf serum (FCS). The DNA synthesis was measured by the cellular uptake of [3H]thymidine. For in vitro wounding the cells were cultured, wounded and stimulated with 0.2% FCS, 10% FCS and EMD at a concentration of 20 microg/ml. The percentage of wound fill after treatment was measured after d 1, 4, 6, 12 and 16. The proliferation of cells was also calculated by the extent of incorporation of crystal violet. RESULTS: The results demonstrated that cells in vitro fill an empty space by a combination of proliferation and cell migration. The most rapid closure of a wound area occurred where both proliferation and migration can occur as was seen when wounded cultures were maintained in 10% FCS or at a concentration of 20 microg/ml EMD which promoted proliferation. CONCLUSIONS: Therefore, EMD appears to exert an influence on cells that is compatible with improved wound healing.     

Nitric oxide synthase type-II is synthesized by human gingival tissue and cultured human gingival fibroblasts

Nitric oxide is known to be an important inflammatory mediator, and is implicated in the pathophysiology of a range of inflammatory disorders. The aim of this study was to determine the localization and distribution of endothelial NOS (NOS-II) in human gingival tissue, and to ascertain if human gingival fibroblasts express NOS-II when stimulated with interferon gamma (IFN-gamma) and bacterial lipopolysaccharide (LPS). The distribution of NOS-II in inflamed and non-inflamed specimens of human gingivae was studied using a monoclonal antibody against nitric oxide synthase II. Cultures of fibroblasts derived from healthy human gingivae were used for the cell culture experiments. The results from immunohistochemical staining of the tissues indicated an upregulation of NOS-II expression in inflamed compared to non-inflamed gingival tissue. Fibroblasts and inflammatory cells within the inflamed connective tissue were positively stained for NOS-II. In addition, basal keratinocytes also stained strongly for NOS-II, in both healthy and inflamed tissue sections. When cultured human gingival fibroblasts were stimulated by INF-gamma and Porphyromonas gingivalis LPS, NOS-II was more strongly expressed than when the cells were exposed to LPS or IFN-gamma alone. These data suggest that, as for other inflammatory diseases, NO plays a role in the pathophysiology of periodontitis.     

Focal adhesion kinase mediates human leukocyte histocompatibility antigen class II-induced signaling in gingival fibroblasts

BACKGROUND AND OBJECTIVE: The role of human leukocyte antigen class II molecules on nonantigen-presenting cells has been a matter of controversy. We previously reported that human leukocyte antigen class II molecules on human gingival fibroblasts do not present antigens, but transduce signals into the cells by making a complex with antigenic peptide T-cell receptor or by stimulating cell surface human leukocyte antigen-DR molecules with human leukocyte antigen-DR antibody (L243), which mimics the formation of the human leukocyte antigen class II-antigenic peptide T-cell receptor complex, resulting in the expression of several cytokines. The aim of this study was to detect human leukocyte antigen class II-associated molecules mediating human leukocyte antigen class II-induced signals into the cells. MATERIAL AND METHODS: Antibody-based protein-microarray analysis was performed to detect activated signaling molecules in gingival fibroblasts stimulated via human leukocyte antigen class II molecules. Then, we examined if these molecules structurally associate with human leukocyte antigen class II and actually transduce signals into the cells. RESULTS: Stimulation of human leukocyte antigen class II on gingival fibroblasts by L243 resulted in enhanced phosphorylation of focal adhesion kinase. Focal adhesion kinase was co-immunoprecipitated with human leukocyte antigen-DR by L243. Stimulation of gingival fibroblasts with L243 induced phosphorylation of focal adhesion kinase. Luteolin, a putative focal adhesion kinase inhibitor, suppressed phosphorylation of focal adhesion kinase and dose dependently inhibited human leukocyte antigen class II-induced cytokine production. CONCLUSION: Focal adhesion kinase is structurally associated with human leukocyte antigen-DR and mediates human leukocyte antigen class II-induced signals in gingival fibroblasts.     

Production of Rantes/CCL5 in human gingival fibroblasts challenged with tumor necrosis factor alpha

Chemokines are small-secreted proteins that stimulate the directional migration of leukocytes and thereby mediate the inflammatory process. The present study investigates the capacity of human gingival fibroblasts to produce the beta chemokine Rantes/CCL5. In situ hybridization, immunohistochemistry and ELISA were used to measure the induction of Rantes/CCL5 at the mRNA and protein levels, both in unstimulated gingival fibroblasts as well as in fibroblasts treated with the proinflammatory cytokines tumor necrosis factor (TNF)alpha or interleukin (IL)-1beta. TNFalpha in different concentrations (0.1-10 ng/ml) induced Rantes/CCL5 mRNA expression and protein production in 24-h cultures of human gingival fibroblasts. The expression of Rantes/CCL5-mRNA and protein production, induced by TNFalpha, was evident at 6 h and thereafter increased continuously during the study period (24 h). IL-1beta (3-300 pg/ml) also enhanced the production of Rantes/CCL5 in gingival fibroblasts. The amount of Rantes/CCL5 induced by IL-1beta (300 pg/ml), however, was less than that induced by TNFalpha (10 ng/ml). The study suggests that human gingival fibroblasts, by producing the chemokine Rantes/CCL5, participate in the regulation of the host response during the inflammatory process in the periodontal tissue.     

Cyclosporin A and hydroxycyclosporine (M-17) affect the secretory phenotype of human gingival fibroblasts

The responsiveness of human gingival fibroblast populations to cyclosporin A (CsA) and its principal metabolite, hydroxycyclosporine (M17), was evaluated in cell culture. Gingival fibroblasts exhibited a dose-dependent accumulation and bell-shaped distribution of dansylated CsA. A 100-fold excess of non-labeled CsA prevented the accumulation of the fluorescent probe in the fibroblasts. Both CsA (400 ng/ml) and M17 (100 ng/ml) stimulated mean gingival fibroblast cell number to 23.2% and 36.7% above controls, and reduced mean collagen production by 37.7% and 37.4% below controls, respectively; however, neither CsA nor M17 affected mean protean production in comparison to control cultures. Analyses of responses to CsA and M17 by ligand-accumulating and non-accumulating fibroblasts sorted out from the parent cultures did not provide consistent interstrain responses either by cells representing the upper quartile of fluorescence or cells representing the bottom quartiles of fluorescence. These data demonstrate that CsA is accumulated by gingival fibroblasts and that CsA and M17 are potent modulators of gingival fibroblast phenotype.     

Cyclosporin A upregulates prostaglandin E2 production in human gingival fibroblasts challenged with tumor necrosis factor alpha in vitro

The effect of Cyclosporin A (CsA) on prostaglandin E₂ (PGE₂) production in human gingival fibroblasts challenged with tumor necrosis factor alpha (TNF-α) was studied. TNF-α (1-100 ng/ml) dose-dependently stimulated PGE₂; formation in 24 h cultures. CsA (1-100 ng/ml) did not induce PGE₂; formation itself but potentiated TNF-α induced PGE; formation in gingival fibroblasts in a manner dependent on the concentrations of both CsA and TNF-α. TNF-α (10 ng/ml) stimulated the release of [³H]-arachidonic acid (A.A) from prelabelled fibroblasts that was potentiated by CsA (100 ng/ml). Addition of exogenous unlabelled AA (5-20 μM/ml) to the cells resulted in enhanced PGE₂: formation that was not potentiated by CsA (100 ng/mi). Furthermore. CsA (100 ng/ml) did not further increase the level of cyclooxygenase-2 mRNA induced by TNF-α (10 ng/ml). although PGE₂ formation was enhanced. The results indicate that CsA and TNF-α act in concert on PGE₂ formation in gingival fibroblasts. which may be of importance in the pathogenesis of gingival overgrowth induced by the drug.     

The effects of folic acid application on IL-1beta levels of human gingival fibroblasts stimulated by phenytoin and TNFalpha in vitro: a preliminary study

Folic acid (FA), that is required for the integrity of gingival tissues, was found to decrease in patients using phenytoin (PHT). Interleukin-1 beta (IL-1beta) was reported to enhance the extracellular matrix synthesis in fibroblasts. The purpose of this study was to assess the role of FA supplementation on PHT-induced overgrowth by investigating its effect on IL-1beta production of human gingival fibroblasts induced by tumor necrosis factor alfa (TNFalpha) in cell culture. PHT (20 microg/ml), FA (20 or 40 ng/ml) + PHT, PHT + TNF (10 ng/ml), FA (20 or 40 ng/ml) + PHT + TNF, or only culture medium (control) was added to 24-well plates containing fibroblasts. After an incubation period of 72 h, culture medium and cells were harvested separately. Then, IL-1beta levels in cell lysate were measured using enzyme-linked immunosorbent assay. The cellular IL-1beta level in the PHT group was 1 pg/ml. In PHT + 20 or 40 ng/ml FA-added cultures, the results obtained respectively were 0.8 and 0.7 pg/ml, whereas the control group value was 0.7 pg/ml. IL-1beta level was 4 pg/ml in the cultures that PHT and TNFalpha were applied simultaneously (P < 0.05). When PHT and either 20 or 40 ng/ml FA were simultaneously added into TNFalpha-induced cultures, the IL-1beta levels were 1.8 and 1.3 pg/ml, respectively. IL-1beta level in gingival tissues might play a role in PHT-induced overgrowth by increasing in the gingival tissues, and FA application might play a role in decreasing gingival tissues. However, further studies are needed for a more complete understanding of PHT-induced gingival overgrowth at the cellular level.     

Effect of IFN–γon the expression of MHC class I and class II antigens in a human malignant oral epithelial cell line

This study examined the expression of major histocompatibility complex (MHC) antigens in the KB human carcinoma cell line. MHC class I antigens (HLA-ABC and beta 2microglobulin) were constitutively expressed but not significantly enhanced by interferon-gamma (IFN-gamma). By contrast, MHC class II antigens were absent or expressed minimally on unstimulated KB cells, but DR and DP were significantly stimulated in a non-coordinated pattern (HLA-DR greater than -DP greater than -DQ) by IFN-gamma.     

Gamma-interferon enhances expression of CD14/MyD88 and subsequent responsiveness to lipopolysaccharide from Actinobacillus actinomycetemcomitans in human gingival fibroblasts

OBJECTIVES: CD14, toll-like receptor 4 (TLR4) and MyD88 have been shown to mediate responsiveness in host cells to lipopolysaccharide. We investigated here the regulatory effects of inflammatory cytokines on the expression of membrane CD14 (mCD14), TLR4 and MyD88, and on subsequent responsiveness to lipopolysaccharide from Actinobacillus actinomycetemcomitans in human gingival fibroblasts. MATERIALS AND METHODS: Following treatment with either interleukin-1beta, tumor necrosis factor-alpha (TNF-alpha) or gamma-interferon (IFN-gamma), expression of mCD14/TLR4 and MyD88 was determined by flow cytometry and western blotting, respectively. After pretreatment with IFN-gamma, cells were pre-incubated with either anti-CD14 antibody MY4 or anti-TLR4 antibody HTA125 and subsequently treated with A. actinomycetemcomitans lipopolysaccharide. Then, phosphorylation of mitogen-activated protein (MAP) kinases and IkappaBalpha was examined by western blotting, and production of interleukin-6 and interleukin-8 was measured by their respective enzyme-linked immunosorbent assay (ELISA) kits. RESULTS: IFN-gamma stimulated expression of mCD14, whereas -1beta and TNF-alpha did not. Expression of MyD88 but not TLR4 was also enhanced by IFN-gamma. The lipopolysaccharide activated MAP kinases, such as extracellular signal-regulated kinase (ERK), c-Jun N-terminal kinase (JNK) and p38, and IkappaBalpha and stimulated production of interleukin-6 and interleukin-8. The lipopolysaccharide-stimulated interleukin-6 and interleukin-8 production was markedly inhibited by MY4 or HTA125. Pretreatment with IFN-gamma augmented the following activation of MAP kinases and IkappaBalpha and production of interleukin-6 and interleukin-8 in response to the lipopolysaccharide. CONCLUSIONS: These results suggest that the augmentation by IFN-gamma of the responsiveness to A. actinomycetemcomitans lipopolysaccharide, such as activation of MAP kinases and IkappaBalpha and terminal cytokine production in human gingival fibroblasts, may be partially mediated by up-regulation of CD14 and MyD88 expression.     

Androgen metabolic response to indomethacin and the alkaline phosphatase inhibitor levamisole in fibroblasts

OBJECTIVES: The aim of this investigation is to study the effects of indomethacin (I) and the alkaline phosphatase (ALP) inhibitor levamisole (L) on androgen 5alpha-reductase expression in gingival and periosteal fibroblasts, in the context of repair in the periodontium. Chronically inflamed human gingival fibroblasts (HGF) were used to demonstrate the comparative effects of L on HGF and human oral periosteal fibroblasts (HPF). MATERIAL AND METHODS: Monolayer cultures of six cell lines of HPF of the fifth to ninth passage were incubated in duplicate with 14C-testosterone/14C-4-androstenedione as substrates in Eagle's MEM; I was added at concentrations of 1 and 3 microg/ml in the presence or absence of the established inhibitory concentration of 30 microg/ml L and incubated for 24 h. The medium was solvent extracted for radioactive metabolites, separated by thin layer chromatography and quantified. RESULTS: L caused 50% inhibition of 5alpha-reductase and 17beta-hydroxysteroid dehydrogenase activity in HGF. In HPF, 5alpha-reductase expression was enhanced by I with both substrates, by 65-76% (n = 6; p<0.01), inhibited by 30-50% (n = 6; p<0.01) with L and restored to control values in combination. CONCLUSION: Yields of androgen metabolites may be linked to ALP activity, with implications on healing, during adjunctive treatment of inflammatory periodontal disease with I.     

Nicotine increases the collagen-degrading ability of human gingival fibroblasts

BACKGROUND AND OBJECTIVE: The objective of this study was to determine the effects that nicotine and the combination of nicotine and Porphyromonas gingivalis supernatant have on human gingival fibroblast-mediated collagen degradation. MATERIAL AND METHODS: Human gingival fibroblasts were cultured with 25-500 microg/ml of nicotine in collagen-coated six-well plates. On days 1-5, the conditioned media was collected for zymography and western blot analyses of matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs). The cells were then removed and the collagen cleavage visualized by Coomassie blue staining. To examine the combined effect, 250 microg/ml of nicotine and 10% v/v culture supernatant of P. gingivalis ATCC 33277 were added to the human gingival fibroblasts. The mRNA levels of multiple MMPs and TIMPs were monitored. RESULTS: Nicotine increased the human gingival fibroblast-mediated collagen cleavage. The MMP-14 and MMP-2 produced by the nicotine-treated human gingival fibroblasts more readily underwent zymogen activation. Nicotine treatment resulted in TIMP-2 redistribution to the cell surface. The mRNAs of multiple MMPs and TIMPs were unaltered by nicotine. An additive collagen cleavage effect was observed when the human gingival fibroblasts were treated with both nicotine and P. gingivalis. CONCLUSION: Nicotine increased human gingival fibroblast-mediated collagen degradation, in part through the activation of membrane-associated MMPs. Nicotine and P. gingivalis had an additive effect on human gingival fibroblast-mediated collagen degradation.     

The effect of triclosan on mediators of gingival inflammation

Abstract Triclosan (2,4,4′,-trichloro-2′-hydroxydiphenylether) is a well-known and widely used nonionic antibacterial agent which has recently been introduced in toothpastes and mouthrinses. The efficacy of triclosan-containing toothpaste and mouthrinse to reduce both plaque and gingivitis in long-term clinical trials has been well documented. Until recently, it was generally assumed that triclosan's effect on gingival inflammation was due to its antimicrobial and anti-plaque effect. It has now become apparent that triclosan may have a direct anti-inflammatory effect on the gingival tissues. Several in vitro studies were conducted to evaluate the effect of triclosan on 4 primary enzymes of the pathways of arachidonic acid metabolism, cyclo-oxygenase 1, cyclo–oxygenase 2, 5-lipoxygenase and 15-lipoxygenase. These pathways lead to the production of known mediators of inflammation such as the prostaglandins, leukotrienes and lipoxins. Triclosan inhibited both cyclooxygenase 1 and cyclo-oxygenase 2 with IC–50 values of 43 μM and 227 μM, respectively. Triclosan also inhibited 5–lipoxygenase with an IC-50 of 43 μM. The 15-lipoxygenase was similarly inhibited by triclosan with an IC-50 of 61 μM. Hence, triclosan has the ability to inhibit both the cyclo-oxygenase and lipoxygenase pathways of arachidonic acid metabolism with similar efficacy. In cell culture experiments, it was found that triclosan inhibited IL-1β induced prostaglandin E₂ production by human gingival fibroblasts in a concentration dependent manner, and at relatively low concentrations. These data, taken together, indicate that triclosan can inhibit formation of several important mediators of gingival inflammation. The data further suggest that the necessary triclosan concentrations could probably be achieved in local tissues, such as the gingiva, from topical applications of triclosan-containing toothpaste and mouthrinse.     

Cyclosporin A regulates interleukin-1beta and interleukin-6 expression in gingiva: implications for gingival overgrowth

BACKGROUND: Gingival overgrowth is a common side effect following the administration of cyclosporin A (CsA); however, the cellular mechanisms remain poorly understood. CsA's immunosuppressant properties involve the regulation of synthesis and cellular response to cytokines. A CsA-induced alteration in the cytokine profile within gingival tissue could provide a mechanism for gingival hyperplasia. The aim of this study was to investigate the effects of CsA on the production of 2 cytokines - interleukin-1beta (IL-1beta) and interleukin-6 (IL-6) - by both gingival fibroblasts and peripheral blood mononuclear cells (PBMC). METHODS: Cells were stimulated for 24 hours in the presence of CsA over a concentration range of 100 to 2,000 ng/ml and the resultant cytokine production determined by ELISA. In addition, levels of both cytokines within normal, inflamed, and overgrown gingival tissue were determined. RESULTS: CsA inhibited IL-6 production by gingival fibroblasts in a dose-dependent manner. In contrast, at a concentration of 2,000 ng/ml, CsA stimulated IL-6 production by PBMC (P <0.05). Fibroblasts derived from overgrown gingiva produced significantly higher levels of IL-6 than their normal counterparts (P <0.05). CsA inhibited IL-1beta production by PBMC over the whole concentration range (P <0.05). IL-1beta was not found in measurable quantities in any of the fibroblast cultures. Levels of IL-6 extracted from overgrown gingival tissue were significantly higher than in inflamed or normal tissue. In contrast IL-1beta levels in overgrown tissue were not statistically significantly greater than those in inflamed tissue. CONCLUSIONS: These results show that CsA does regulate cytokine expression in gingival tissue. This effect may play an important role in the pathogenesis of CsA-induced gingival overgrowth.     

Effect of nicotine-treated epithelial cells on the proliferation and collagen production of gingival fibroblasts

BACKGROUND, AIMS: Several in vitro and in vivo studies have indicated that tobacco smoking may be an important risk factor for the development and severity of inflammatory periodontal disease. METHOD: In the present study, we developed an in vitro model to study the interactions between nicotine-treated epithelial cells (EC) and gingival fibroblasts (GF) derived from the same patient. EC were treated with nicotine concentrations varying from 1 microg/ml to 500 microg/ml and their effect on different functions of GF was studied. The proliferation of GF was evaluated by the incorporation of 3H-thymidine. A dose-dependent inhibition was observed with nicotine concentrations > or =100 microg/ml. Similar results were observed when studying the total protein synthesis of GF by incorporation of 3H-proline into non-dialyzable material. RESULTS: When collagen production was evaluated by 3H-proline incorporation into collagenase-sensitive protein, a dose-dependent reduction was observed: the degree of inhibition varied from 25% with 50 microg/ml nicotine, to almost 60% with 500 microg/ml. Interestingly, the production of non-collagenous proteins decreased by almost 50% only when EC were treated with the highest concentration of nicotine. CONCLUSIONS: The results suggest that epithelial cells, acting as mechanical barrier, can reduce but not completely eliminate the deleterious effect of nicotine on gingival fibroblasts.     

Epidermal growth factor stimulates urokinase-type plasminogen activator expression in human gingival fibroblasts. Possible modulation by genistein and curcumin

BACKGROUND: Regulation of the extracellular matrix turnover is a crucial process in wound healing and the progress of periodontal disease. It has been proposed that urokinase-type plasminogen activator (uPA), under the control of growth factors or cytokines, provides the proteolytic potential to the accomplishment of these cellular events. Epidermal growth factor (EGF) is one of the growth factors that has been shown to be active in uPA regulation. METHODS: In this study, we have assessed the effect of EGF on uPA expression in primary cultures of human gingival fibroblasts. We also studied the signaling pathways involved in this process and the role of the dietary phytoestrogens curcumin and genistein as potential modulators of this response. RESULTS: Human gingival fibroblasts expressed a basal uPA activity, which was inhibited by genistein, but not by curcumin. After treatment with 10 ng/ml EGF, uPA production was strongly stimulated. Exposure to genistein and curcumin inhibited EGF-stimulated urokinase production, although only genistein showed a statistically significant inhibitory response. Using more specific inhibitors, we found that the mitogen-activated extracellular kinase and c-Jun N-terminal kinase (JNK) inhibitors PD98059 and SP600125 also blocked the EGF-dependent stimulatory effect. On the other hand, SB203580, inhibitor of the p38 member of mitogen-activated protein kinase family, did not alter this response. In accordance to these findings, EGF stimulated a potent activation of JNK and a mild activation of extracellular signal-regulated kinases 1/2. Finally, EGF stimulated the phosphorylation of its receptor and tyrphostin (AG1478), curcumin and genistein were able to inhibit this stimulatory effect. CONCLUSIONS: These results indicate that EGF constitutes a strong stimuli on uPA expression in human gingival fibroblasts. Our data also shows that EGF-stimulated uPA production involves the activation of the extracellular signal-regulated kinases 1/2 and JNK signaling pathways and might be modulated by the natural phytoestrogens curcumin and genistein.