Subversion of Lipopolysaccharide Signaling in Gingival Keratinocytes via MCPIP-1 Degradation as a Novel Pathogenic Strategy of Inflammophilic Pathobionts

Abstract

Periodontal disease (PD) is an inflammatory disease of the supporting tissues of the teeth that develops in response to formation of a dysbiotic biofilm on the subgingival tooth surface. Although exacerbated inflammation leads to alveolar bone destruction and may cause tooth loss, the molecular basis of PD initiation and progression remains elusive. Control over the inflammatory reaction and return to homeostasis can be efficiently restored by negative regulators of Toll-like receptor (TLR) signaling pathways such as monocyte chemoattractant protein-induced protein 1 (MCPIP-1), which is constitutively expressed in gingival keratinocytes and prevents hyperresponsiveness in the gingiva. Here, we found that inflammophilic periodontal species influence the stability of MCPIP-1, leading to an aggravated response of the epithelium to proinflammatory stimulation. Among enzymes secreted by periodontal species, gingipains-cysteine proteases from Porphyromonas gingivalis-are considered major contributors to the pathogenic potential of bacteria, strongly influencing the components of the innate and adaptive immune system. Gingipain proteolytic activity leads to a rapid degradation of MCPIP-1, exacerbating the inflammatory response induced by endotoxin. Collectively, these results establish a novel mechanism of corruption of inflammatory signaling by periodontal pathogens, indicating new possibilities for treatment of this chronic disease. IMPORTANCE Periodontitis is a highly prevalent disease caused by accumulation of a bacterial biofilm. Periodontal pathogens use a number of virulence strategies that are under intensive study to find optimal therapeutic approaches against bone loss. In our work, we present a novel mechanism utilized by the key periodontal pathogen Porphyromonas gingivalis, based on the selective degradation of the negative regulator of inflammation, MCPIP-1. We found that the diminished levels of MCPIP-1 in gingival keratinocytes-cells at the forefront of the fight against bacteria-cause sensitization to endotoxins produced by other oral species. This results in an enhanced inflammatory response, which promotes the growth of inflammophilic pathobionts and damage of tooth-supporting tissues. Our observation is relevant to understanding the molecular basis of periodontitis and the development of new methods for treatment.

Keywords: MCPIP-1; Porphyromonas gingivalis; gingipains; lipopolysaccharide; periodontitis.

FIG 1

Repeated infection with P. gingivalis W83 induces bone loss in a murine model, contributing to depletion of MCPIP-1 protein in gingiva. (A) Graphical representation of P. gingivalis-induced periodontitis in mice. (B) Visualization of P. gingivalis W83-induced bone loss in wild-type (WT) mice. Representative images of methylene blue (upper) and sectional micro-computed tomography (micro-CT) analysis of gingival tissue (bottom) with indicated cementoenamel junction (CEJ) to alveolar bone crest (ABC) distances on the second molar (yellow lines). (C) Quantitative analysis of the CEJ-ABC distances of the second molar. n = 6 in control group; n = 8 in infected group (+W83). (D) Levels of anti-W83 and anti-RgpA IgG antibodies in murine sera. Results for the control mice are shown as 1. (E) Visualization of MCPIP-1 protein level (green) by confocal laser scanning microscopy in murine gingiva, along with protein quantification presented as a percentage of area of the fluorescence signal. Analysis performed based on 15 images. (F) Relative expression of Mcpip-1 transcript in gingival tissue. All data represent mean values ± standard error of the mean (SEM). *, P < 0.05; **, P < 0.01; ****, P < 0.0001; ns, not significant.

FIG 2

Proteolysis of MCPIP-1 depends on gingipain activity. (A) Recombinant MCPIP-1 protein was incubated with the P. gingivalis wild-type W83 strain, the gingipain-deficient ΔKΔRAB strain, or outer membrane vesicles (OMVs) isolated from both species for 1 h at 37°C. Wild-type bacteria and OMVs were heat-inactivated (HI) at 90°C for 20 min or inactivated with the gingipain-specific inhibitors KYT-1 and KYT-36 (KYTs) for 20 min. (B) Western blot analysis of protein lysates of gingival keratinocytes (TIGKs) infected with P. gingivalis W83 or ΔKΔRAB strains (multiplicity of infection [MOI], 1:100). Gingipain activity was inhibited using a mixture of KYT-1 and KYT-36 inhibitors. (C) Time-dependent MCPIP-1 protein degradation in TIGKs estimated by Western blotting, quantified by densitometry, and presented as a fold change normalized to β-actin levels. Cells were incubated with OMVs with or without KYTs. (D) Recombinant MCPIP-1 protein was incubated for 60 min at 37°C with purified gingipains RgpA, RgpB, or Kgp at indicated molar ratios (gingipain:MCPIP-1) and (E) with RgpA at a 1:10,000 molar ratio for the indicated times. (F) Determination of MCPIP-1 level after exposure of recombinant protein to active RgpA or RgpA inactivated by KYT-1 at a molar ratio of 1:10,000. (A to F) Representative Western blot results are shown.

FIG 3

RgpA degrades MCPIP-1 protein in gingival keratinocytes. (A) Representative Western blot analysis, using specific anti-MCPIP-1 antibodies, of protein lysates from TIGK cells treated with 2 nM gingipains (RgpA, RgpB, or Kgp) for the indicated times. (B) Densitometry analysis demonstrating MCPIP-1 protein fold change normalized to β-actin level in TIGK lysates. (C) Visualization of MCPIP-1 protein (red) and RgpA (green) level by confocal laser scanning microscopy in TIGKs. (D) Quantitative analysis of MCPIP-1 and RgpA fluorescence signal. Data show mean fluorescence values (± standard deviation [SD]) in 8 tested fields of view. **, P < 0.01; ****, P < 0.0001; ####, P < 0.0001.

FIG 4

RgpA induces hyperresponsiveness of gingival keratinocytes to endotoxin and Gram-negative bacteria. (A) TIGK cells were prestimulated with 2 nM RgpA for 1 h and treated with lipopolysaccharide (LPS; 20 μg/ml) or infected with bacteria for an additional 3 h. Heat map represents the changes in the expression of genes regulated by LPS signaling in the presence of RgpA. Relative mRNA levels were determined by quantitative reverse transcription-PCR (qRT-PCR). Expression of each gene in cells stimulated only with LPS was established as 1. (B) Quantitative RT-PCR analysis of the proinflammatory cytokines and MCPIP-1 mRNA level after treatment of cells with RgpA, LPS, and/or RgpA 1 h prior to LPS in comparison to untreated cells. (C, D) Relative expression of IL-1β, IL-6, IL-8 and MCPIP-1 at the mRNA and protein level in TIGK cells. Gingival keratinocytes were pretreated for 1 h with RgpA, and then the specific-RgpA inhibitor KYT-1 was added for 30 min. Subsequently, cells were stimulated with LPS for 3 h. ND, not detected. (E) Relative expression of IL-1β after treatment of TIGK cells with cytochalasin D (+cytD) for 30 min prior to their stimulation with LPS and RgpA/LPS in comparison to cytochalasin D-untreated cells. LPS-stimulated cells are presented as 100%. (F) Comparison of IL-1β mRNA expression determined by qRT-PCR after a 3-h reaction of TIGKs to different Toll-like receptor (TLR) agonists in comparison to cells additionally pretreated with RgpA for 1 h. TIGKs stimulated with corresponding agonists are presented as 100%. (G, H) Gingiva keratinocytes were preincubated for 1 h with RgpA (2 nM) and then exposed for 3 h to (G) the P. gingivalis gingipain-deficient ΔKΔRAB strain at an MOI of 1:25, or (H) F. nucleatum at an MOI of 1:10. The level of proinflammatory cytokines was determined by qRT-PCR. TIGKs stimulated with bacteria are presented as 100%. Data represent mean values from three independent experiments ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, not significant.

FIG 5

The LPS sensitivity of keratinocytes depends on MCPIP-1 expression. (A) Depletion of Mcpip-1 in keratinocytes determined by qRT-PCR. (B, C) Wild-type (WT) and Mcpip-1-deficient (Mcpip-1^eKO) murine keratinocytes were exposed to RgpA (2 nM) for 1 h, followed by stimulation with LPS (20 μg/ml) for an additional 3 h. Relative expression of (B) IL-1α and (C) IL-1β mRNA was determined by qRT-PCR. Cells stimulated with LPS are presented as 100%. (D) Relative expression of Mcpip-1 transcript in murine gingival tissue (n = 4). (E) Quantitative analysis of the CEJ-ABC distances of the second molar of the P. gingivalis-induced bone loss in WT and Mcpip-1^eKO mice after repetitive infections with wild-type strain W83 or with the gingipain-deficient ΔKΔRAB mutant (n = 4 in each group). (F) Levels of anti-P. gingivalis IgG antibodies in murine sera. The values in WT individuals were defined as 100%. All data represent mean values ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, not significant.

FIG 6

Highly proteolytically active periodontal bacteria degrade the MCPIP-1 protein. Representative Western blot of recombinant human MCPIP-1 protein incubated for 1 h with bacterial supernatants (upper) or bacteria (bottom) classified as pathogens (Porphyromonas gingivalis, Tannerella forsythia, Prevotella intermedia, or Fusobacterium nucleatum) or commensals (Streptococcus oralis, Actinomyces naeslundii, or Streptococcus salivarius).

FIG 7

Schematic representation of the putative role of RgpA in the regulation of LPS signaling in gingival keratinocytes. RgpA secreted by P. gingivalis penetrates into the cytoplasm, degrades MCPIP-1, and sensitizes epithelial cells to endotoxins and to both commensal and pathogenic species. Additionally, the released cytokines activate a proinflammatory signaling cascade via the autocrine pathway, leading to exacerbation of the local inflammatory reaction.