Functional analysis of fibroblast-activating factor from porphyromonas gingivalis W50

Periodontal diseases are complex, multifactorial, polymicrobial infections characterized by the destruction of tooth-supporting tissues. The disease begins as acute inflammation of the gingival tissue and untreated infections can progress to formation of teeth pockets, and eventually loss of teeth. According to the World Health Organization, periodontal disease affects 10 - 15% of adult populations worldwide and it was shown that periodontitis also enhances the risk for cardiovascular diseases and diabetes. While human subgingival plaque harbors more than 500 bacterial species, considerable evidence points to Porphyromonas gingivalis, a Gram-negative anaerobic bacterium, as the major etiologic agent of chronic periodontitis. This black-pigmented bacteria produces a myriad of virulence factors that apparently contribute to periodontal tissue destruction either directly or indirectly by modulating the host inflammatory response. Among them is fibroblast-activating factor (FAF) encoded by the gene, faf. This 24 kDa outer-membrane protein affects the supporting tissue of the periodontium, among which is the proliferative stimulating effect on human gingival fibroblasts (HGFs). It has been shown that FAF undergoes post-translational processing in which the N-terminal signal peptide is cleaved before the protein is secreted. Both full length faf and faf without the signal peptide sequence (designated δfaf) was amplified and cloned into pET28a for overexpression in E. coli BL21. SDS-PAGE analysis illustrated a higher expression level of δFAF than its full-length counterparts. An optimum expression level was obtained at 0.5 mM IPTG for 16 h at 25ºC. Both variants of purified recombinant FAF proteins at concentrations between 0.5 - 1.5 μg/ml significantly enhanced cell proliferation and DNA synthesis in mouse embryonic fibroblasts (NIH3T3), human dermal fibroblasts (HDFn) and HGF in a time- and dose-dependent manner. The biological activity of the proteins is suggestive of a specific target cell effect as several cancerous cell lines (MCF7 and HeLa) and T-lymphocytes (Jurkat) were not responsive to both recombinant proteins. The recombinant proteins were also not able to revive fibroblasts under 24 h starvation. On the other hand, both FAF variants and growth factors (EGF and FGF) showed a synergistic effect on the proliferation of fibroblasts. However, the biological activities of the proteins were abolished when heated to 60°C or in the presence of selective protease inhibitors. In addition, quantitative real-time PCR analysis demonstrated that both recombinant proteins induced a down-regulation of collagen synthesis in the fibroblasts while bone resorption assays showed an increase in calcium release in mouse calvarial culture. This ultimately causes destruction of the connective tissue attachment to the root surface and resorption of the bone. Meanwhile, treating the human fibroblasts HGF with both recombinant proteins triggered the secretion of proinflammatory cytokine, IL-8, and the synthesis was sustained for almost a week. However, the secondary messenger, cAMP, appeared not to be involved in the signal transduction and hence, the activity of both FAFC and δFAFN in the host cells. Immunohistochemical studies successfuly showed the binding of both recombinant proteins on the cell surface receptor of the fibroblasts. However, the isolation of the binding partner of δFAF protein from the host membrane fractions using pull-down assay was not successsful. To determine the localization of the functional domains of FAF, several truncation mutants were constructed. A mutant construct, FAF∆(121-139), showed that a deletion of a peptide fragment spanning residues 121 – 139 rendered the FAF inactive, suggesting that these residues are likely involved in modulating the protein’s biological activity