CD40 ligation triggers COX-2 expression in endothelial cells: evidence that CD40-mediated IL-6 synthesis is COX-2-dependent.

OBJECTIVE: To investigate whether CD40-CD154 interactions on HUVEC can trigger COX-2 synthesis as well as PGE2 and PGI2 secretion in vitro and explore whether the CD40-triggered prostanoids provide costimulatory signals for IL-6 secretion in this cell type. MATERIALS AND METHODS: COX-2 protein expression was examined in HUVEC using Western blot analysis. ELISAs were employed to assess PGE2, PGI2 and IL-6 synthesis. RESULTS: We found that COX-2 expression is upregulated when HUVEC are cultured with CD154+ D1.1 cells but not CD154- B2.7 cells. This effect was specifically inhibited by anti-CD154 mAb, and was amplified by the presence of IFNgamma. Analysis of cell supernatants showed a concomitant rise in PGE2 and PGI2 secretion triggered by CD154+ D1.1 cells, or rsCD154. Use of selective (NS-398) and non-selective (ibuprofen) COX-2 inhibitors effectively inhibited prostanoid synthesis triggered by CD40 ligation. Reduction in prostanoid levels by NS-398 was accompanied by a reduction in IL-6 secretion levels triggered by CD40 ligation. Furthermore, exogenously added PGE2 triggered a dose-dependent IL-6 secretion, which was unaffected by NS-398. CONCLUSIONS: These studies demonstrate that CD40 ligation upregulates HUVEC COX-2 expression and function. Moreover, the data strongly suggest that CD154-induced IL-6 secretion in HUVEC is dependent on COX-2 activity.

Cyclooxygenase-2 is upregulated in inflamed gingival tissues.

BACKGROUND: Increased release of prostaglandins (PG) within periodontal tissues is considered to play a pathogenetic role during periodontal disease progression. The rate-limiting step in the formation of PG from arachidonic acid is catalyzed by cyclooxygenase (COX). Currently there are 2 known isoforms of the enzyme. COX-1 is constitutively expressed in various tissues whereas COX-2 is an inducible enzyme believed to be responsible for PG synthesis at sites of inflammation. The purpose of this study was to compare COX-2 expression in inflamed and healthy human gingiva and further explore some of the pathogenetic mechanisms which may lead to elevated COX-2 expression in vivo. METHODS: Thirty-two gingival biopsies were obtained during routine oral surgical procedures and were processed histologically using hematoxylin and eosin to determine the degree of inflammation. Of these biopsies, 7 with low and 7 with high histological levels of inflammation were further processed immunohistochemically in order to assess the levels of COX-2 expression in situ. To explore some potential mechanisms of COX-2 upregulation, gingival connective tissue primary cell cultures were established and challenged with periodontal bacteria or proinflammatory cytokines in vitro. The levels of COX-2 expression were analyzed by Western blot of cell lysates. COX-2 activity was assessed by quantifying prostaglandin E2 (PGE2) levels in culture supernatants by competitive EIA. RESULTS: We have shown by immunohistochemistry that COX-2 expression was significantly higher (P < 0.01) in tissues with higher levels of inflammatory infiltrates. Expression of COX-2 was detected in gingival epithelium, endothelial cells as well as cells with fibroblast morphology. In vitro studies indicated that gingival fibroblasts (GF) did not express COX-2 constitutively. However, when these cells were challenged with interleukin (IL)-1 beta or bacterial cells (A. actinomycetemcomitans JP2 or B. forsythus ATCC 43037), COX-2 expression as well as COX-2 activity were upregulated. COX-2 expression was upregulated as early as 2 hours post IL-1 beta challenge and was accompanied by a sustained PGE2 release in the culture supernatants. Cyclosporin A (CsA) did not inhibit COX-2 expression induced by bacterial challenge. In contrast, NS-398, a selective inhibitor of COX-2 activity, almost completely abolished PGE2 synthesis by these cells in response to bacterial or cytokine challenge. CONCLUSIONS: We conclude that COX-2 expression is significantly upregulated in inflamed periodontal tissues. Both inflammatory cytokines such as IL-1 beta and bacterial constituents may be responsible for the enhanced COX-2 expression and PGE2 synthesis in vivo.

Regulation of gingival fibroblast interleukin-6 secretion by cyclosporine A.

BACKGROUND: Cyclosporine A (CsA) is a widely used immunosuppressant, with clinical applications ranging from organ transplants to chronic inflammatory diseases. One of the side effects associated with CsA treatment is the development of gingival overgrowth. Exuberant growth of connective tissue within the periodontium can result from hyperactivity of resident fibroblasts. Fibroblasts are capable of secreting interleukin-6 (IL-6), which has been shown to enhance proliferation as well as collagen and glycosaminoglycan synthesis by these cells. We tested the hypothesis that one of the pathogenetic mechanisms underlying CsA-induced fibrosis is an enhanced IL-6 secretion by gingival fibroblasts (GF) in response to this drug. METHODS: The ability of CsA to upregulate GF IL-6 secretion alone or in combination with bacterial challenge or other inflammatory cytokines was tested in an in vitro system. Fibroblast cultures were established from systemically healthy gingival tissue donors and were challenged with CsA in the absence or presence of bacteria, IL-1beta, or tumor necrosis factor (TNF) alpha as co-stimulants. Nifedipine and phenytoin were also tested to further support findings with CsA. After 72 hours of incubation, culture supernatants were collected and analyzed for IL-6 content by ELISA. RESULTS: We have shown that GF respond to CsA with an increase in IL-6 secretion. The magnitude of this response varies among cultures derived from different tissue donors. We have also demonstrated that GF IL-6 responses to bacterial challenge or TNFalpha are downregulated by CsA. However, CsA synergizes with IL-1beta to further upregulate IL-6 secretion, and this effect is shared by phenytoin and nifedipine. CONCLUSIONS: We conclude that one of the pathogenetic mechanisms underlying drug-induced gingival overgrowth may be enhanced secretion of IL-6 by GF in response to these medications. This is the first report on direct and indirect effects of gingival overgrowth-related medications on GF IL-6 metabolism. This work will lay the foundation for future studies directed towards the development of prevention or treatment modalities for gingival overgrowth based on blocking the fibrogenic activities of IL-6 at the cellular level.

Increased presence of interleukin-6 (IL-6) and IL-8 secreting fibroblast subpopulations in adult periodontitis.

Periodontitis is a chronic inflammation of the supporting structures of the dentition which constitutes one of the most common causes of adult tooth loss. While certain microorganisms have been associated with the onset of the disease process, the exact pathogenetic mechanisms underlying periodontal destruction are still poorly understood. We have tested the hypothesis that gingival fibroblasts from diseased sites contribute to pathogenesis by possessing a secretory phenotype characterized by an exuberant secretion of inflammatory mediators and cytokines. Of the cytokines and mediators tested, fibroblast IL-1beta and prostaglandin E2 (PGE2) secretion was not different between health and disease. However, we have shown that fibroblasts from periodontal lesions produce in vitro greater amounts of IL-6 and IL-8 constitutively than healthy controls. When fibroblasts were stimulated with a panel of endogenous or exogenous response modifiers, the magnitude of cytokine and mediator stimulation above constitutive levels did not differ between health and disease. A strong positive correlation was identified between IL-6 or IL-8 constitutive secretion levels in vitro and the in situ expression of these cytokines within the connective tissues from where these cells originated, indicating that the in vitro phenotype mirrors their in vivo function. Furthermore, we present evidence which indicates that increased cytokine secretion by fibroblasts in disease is due to an elevated proportion of subpopulations with higher cytokine secretory capacity. Finally, we demonstrated that cultures from diseased sites are composed of cells with higher levels of constitutive CD40 expression, which may contribute to the increased IL-6 and IL-8 secretory phenotype.

CD40 expression by gingival fibroblasts: correlation of phenotype with function.

CD40, a member of the tumor necrosis factor-alpha receptor family, is constitutively expressed by cells of hematopoietic and non-hematopoietic origin, including fibroblasts. Signaling through this receptor molecule regulates inflammatory cytokine secretion by many cell types. Based on the recently described cytokine secretory heterogeneity of fibroblast cell subsets, we hypothesized that secretion of inflammatory cytokines by gingival fibroblast cultures may be dictated by the existence of differential proportions of cytokine-secreting subpopulations which express high levels of CD40. After examining a large number of gingival fibroblast (GF) cultures we find that the frequency of IL-6- and IL-8-secreting cells mirrors the frequency of cells expressing high levels of CD40 in these cultures. In addition, we demonstrate a direct functional relationship between CD40 expression and IL-6 or IL-8 secretion by showing that ligation of this molecule on GF, and CD40+ fibroblast subsets in particular, up-regulates secretion of these cytokines in vitro.

Application of immunoprinting for assessment of fibroblast secretory heterogeneity.

Fibroblasts are capable of secreting a number of inflammatory mediators and cytokines and may exhibit a marked heterogeneity in this capacity. The relative frequency of cytokine-secreting fibroblasts in chronic inflammatory connective tissue disorders may affect the amount of these molecules secreted locally and dictate the intensity or chronicity of the disease process. We have devised a simple, in situ immunodetection method for assessment of the frequency of actively secreting cells in adherent cell cultures. Our technique is based on the principle of immunoprinting and is coupled with an enzyme-linked immunodetection system. The methodology, sensitivity, specificity and accuracy of this technique are described.

Gingival fibroblast cytokine profiles in Actinobacillus actinomycetemcomitans-associated periodontitis.

Within the in vivo environment human gingival fibroblasts (HGF) may be challenged with bacteria or bacterial products. This interaction may result in the release of cytokines which are directly or indirectly involved in connective tissue and bone catabolism, such as interleukin (IL)-1 beta, IL-6, and IL-8. Our investigation has tested the hypothesis that HGF from Actinobacillus actinomycetemcomitans (Aa)-infected patients with rapidly destructive forms of periodontitis, such as localized juvenile periodontitis (LJP), respond to Aa challenge with an exaggerated secretion of IL-1 beta, IL-6, and IL-8. We have compared the in vitro profiles of these cytokines by Aa-challenged HGF obtained from 2 healthy subjects, 2 Aa-infected, slowly progressing adult periodontitis (AP) patients and 2 LJP patients. HGF were challenged throughout a 48-hour period with formalinized whole bacterial cells, and culture supernatants were analyzed for cytokine content using RIA. No differences were noted in the IL-1 beta secretion levels among the different HGF cultures. Although basal (unchallenged) IL-6 and IL-8 production was similar in all HGF cultures, HGF from the two LJP patients responded to Aa challenge with a more rapid IL-6 and a more pronounced IL-8 secretion than healthy or AP HGF. We also tested the ability of human serum antibodies against Aa to moderate the Aa-elicited HGF cytokine secretion by adding human serum, with normal or elevated antibody content. Both sera appeared to have an upregulating effect on IL-6 and IL-8 secretion. Depletion of 95% of the anti-Aa antibody from serum by absorption did not affect its activity. Based on the response of HGF from two LJP patients, we propose that Aa-induced pathology in LJP may be modulated by stimulation of rapid and/or exaggerated secretion of cytokines with potential catabolic effects, although studies with a larger group of LJP patients are needed to further test this hypothesis. Furthermore, serum antibodies against this microorganism do not appear to have a neutralizing effect in cytokine-eliciting HGF-Aa interactions.

Production of inflammatory mediators and cytokines by human gingival fibroblasts following bacterial challenge.

Bacteria can indirectly affect the course of periodontal diseases by activating host cells to produce and release inflammatory mediators and cytokines. These mediators and cytokines manifest potent proinflammatory and catabolic activity and may play key roles in local amplification of the immune response as well as in periodontal tissue breakdown. This study tested the effect of Actinobacillus actinomycetemcomitans (Aa) and Campylobacter rectus (Cr) challenge on PGE2, IL-1 beta, IL-6 and IL-8 production by human gingival fibroblasts (HGF). Contact-inhibited HGF were prepared and formalin-killed bacterial cells (Aa JP2, ATCC 29523 & 33384 and Cr ATCC 33238) at 10(6)-10(9) were added to the HGF. Culture supernatants were collected at varying time intervals and analyzed for cytokine and mediator content. All concentrations of Aa JP2 and Cr ATCC 33238 suppressed IL-1 beta production up to approximately 50% during the initial 3-12-h period. No bacterial concentration tested was able to increase IL-1 beta production above the maximum basal levels. Both bacterial species stimulated production of IL-6 and IL-8. Aa JP2 did not affect PGE2 levels significantly, whereas Cr ATCC 33238 was stimulatory only at the highest concentration tested (10(9)). There were no significant differences among the three Aa strains with respect to IL-1 beta production. However, Aa ATCC 29523 and ATCC 33384 were less capable of stimulating IL-6 secretion and more efficient in stimulating IL-8 production than Aa JP2. In general, Cr was the most potent enhancer of cytokine and mediator production by HGF. In conclusion, Aa and Cr are capable of amplifying the local immune response and promoting periodontal tissue inflammation by stimulating HGF to secrete mainly IL-6 and IL-8.