The role of invasion ability and proteolytic activity of Porphyromonas gingivalis in the murine model of periodontitis

Abstract

Background Periodontitis is a polymicrobial infectious disease caused by several plaque-associated bacteria, including Porphyromonas gingivalis, Treponema denticola, and Tannerella forsythia. P. gingivalis, a gram-negative obligate anaerobe, has a number of virulence factors such as cysteine proteases (gingipains), LPS, capsule, and fimbriae. Among them, gingipains have been described as a major virulence factor to evade the host immune defense by disrupting cytokine signaling network. Gingipains can degrade a number of host proteins, including cytokines/chemokines, and complement proteins. P. gingivalis has been shown to invade various host cells in vitro, including endothelial cells, gingival fibroblasts, and gingival epithelial cells. P. gingivalis has an inter-strain variability in both the proteolytic activity and invasion ability. Gingival epithelia provide the chemical, physical, and immunological barriers against the invading bacteria. The physical barrier is regulated with the expression of tight-junction (TJ) proteins such as junctional adhesion molecules (JAMs), zonula occludens (ZOs), and claudins. P. gingivalis can modulate the expression of epithelial junctional proteins. In addition, the impairment of epithelial physical barrier is involved in the pathogenesis of periodontitis. The role of invasion ability and proteolytic activity of P. gingivalis in the pathogenesis of periodontitis has not been studied at the same time. The aims of this study were to determine the effect of the invasion ability and proteolytic activity of P. gingivalis on the development of periodontitis and the direction of immune response in mice using P. gingivalis clinical isolates with different invasion ability and proteolytic activity.

Methods Three clinical isolates of P. gingivalis (P1, P4, and P8) obtained from the diseased sites of periodontitis patients were used in the present study (P1: low invasion ability and low proteolytic activity, P4: low invasion ability and high proteolytic activity, P8: high invasion ability and low proteolytic activity). To examine the effect of P. gingivalis clinical isolates with different invasion ability and proteolytic activity on the epithelial physical barrier function, the transepithelial electrical resistance (TER) of immortalized human gingival keratinocyte HOK-16B infected with P. gingivalis clinical isolates was measured. In addition, the levels of the TJ proteins (JAM-1 and ZO-1) were measured by immunofluorescence microscopy. To determine the role of P. gingivalis gingipain in the bacterial invasion and the regulation of epithelial physical barrier function, the invasion of P. gingivalis into HOK-16B cells was measured by a flow cytometric invasion assay using P. gingivalis ATCC 33277 (wild-type) and isogenic gingipain deletion mutant [KDP 129 (kgp-), KDP 133 (rgpA- rgpB-) and KDP 136 (kgp- rgpA- rgpB-)]. Additionally, the effect of wild-type and mutants on TER of HOK-16B cells was measured. To examine the role of invasion ability and proteolytic activity of P. gingivalis in the development of periodontitis in mice, the alveolar bone loss was measured four weeks after oral inoculation (six times, 2 day apart) of P. gingivalis (109 cells/mouse). To detect the bacteria within the gingival tissues after P. gingivalis inoculation, digoxigenin-labeled P. gingivalis-specific probe was prepared and then P. gingivalis was detected in paraffin-embedded tissue sections using in-situ hybridization. The presence of neutrophils or T cells in tissue sections were also determined by immunohistochemistry using specific antibody to Ly6G or CD3. To investigate the direction of adaptive immune response by P. gingivalis infection in mice, the levels of P. gingivalis-specific IgG2a, IgG1, and IgA antibodies in sera were measured by ELISA. The levels of IFN-γ, IL-4, and TGF-β1 proteins in mice cervical lymph node lysates was measured by ELISA.

Results The TER of HOK-16B cells was decreased by all P. gingivalis strain and an inter-strain difference was observed between P1 and P4 at 8 hours. The levels of TJ proteins (JAM-A and ZO-1) in HOK-16B cells was decreased by P. gingivalis clinical isolates, and P4 strain had the greatest effects. The invasion ability of gingipain mutant KDP 136 strain was significantly decreased compared with those of wild-type strain (ATCC 33277). Wild-type and KDP 133 strain remarkably decreased the TER of HOK-16B cells. When experimental periodontitis was induced using P1, P4, or P8 strain in mice, P4 strain induced higher levels of alveolar bone loss than either P1 or P8 strain. The invasion levels of P4 strain into the gingival tissues were higher than those of either P1 or P8 strain but statistical significance was not reached. Different from expectation, the number of neutrophils and T cells detected within the gingival tissues was not increased by P. gingivalis. The levels of anti-P. gingivalis IgG2a and IgG1 were not different among groups. However, the levels of anti-P. gingivalis IgA against P1 were strongly induced. Interestingly, the levels of IFN-γ, IL-4, and TGF-β1 were decreased in cervical lymph node of all P. gingivalis-infected group compared with sham control.

Conclusion P. gingivalis clinical isolates impaired epithelial barrier function through the regulation of TJ protein expression. The proteolytic activity, but not invasion ability, of P. gingivalis may play an important role in the pathogenesis of murine periodontitis model.