Fig 1.
P. gingivalis gingipains degrade JAM1 in IHGE cells.
(A) IHGE cells were infected for 1 h with P. gingivalis WT or the Δkgp ΔrgpA ΔrgpB mutant at an MOI of 100. The cells were then analyzed by immunoblotting with the indicated antibodies. (B–E) IHGE cells were transiently transfected with plasmid encoding Myc-mCherry-CLDN1 (B), Myc-mCherry-CLDN4 (C), Myc-mCherry-OCLN (D), or HA-TJP1 (E). After 48 h of incubation, cells were infected with P. gingivalis at an MOI of 100 for 1 or 3 h. The cells were then analyzed by immunoblotting with the indicated antibodies. β-actin was used as a loading control. IB, immunoblot.
Fig 2.
Confocal microscopic images of JAM1 in IHGE cells infected with P. gingivalis WT or the Δkgp ΔrgpA ΔrgpB mutant.
(A) IHGE cells were infected with P. gingivalis WT or the Δkgp ΔrgpA ΔrgpB mutant at an MOI of 100 for 1.5 h. The cells were then fixed, stained with DAPI (cyan) and anti-JAM1 (yellow), and analyzed by confocal microscopy. Scale bars, 10 μm. (B, C) Schematic illustration (B) and confocal microscopic cross-sectional images (C) of the 3D-tissue model of IHGE cells. Gingival epithelial tissues on coverslips were infected with P. gingivalis WT or the Δkgp ΔrgpA ΔrgpB mutant for 2 h. The tissues were then fixed, stained with anti-JAM1 (white) and Alexa Fluor 568–conjugated phalloidin (magenta), and analyzed by confocal microscopy. Scale bars, 30 μm.
Fig 3.
Confocal microscopic images of JAM1 in IHGE cells.
(A) Schematic view of the JAM1 construct. The Myc-mCherry tag is indicated by the magenta box, and the HA-tag is indicated by the green box. The signal peptide is indicated by the blue box, and IG-LIKE and TMD domains are indicated by the gray boxes and orange box, respectively. (B) Confocal microscopic images of an IHGE cell expressing Myc-mCherry-JAM1, stained with anti-JAM1 (green) and anti-HA (cyan). mCherry is shown in magenta. Higher magnifications of the areas indicated by the white boxes in the left panels are shown at right. Scale bar, 5 μm. (C–F) IHGE cells were transiently transfected with plasmids encoding Myc-mCherry–tagged HA-inserted JAM1; in (C) and (D), the cells were also transfected with EGFP-SEC61β (green). Following 48 h of incubation, cells were fixed, stained with anti-HA (D, magenta; F, green), and also stained with Alexa Fluor 488–conjugated phalloidin (E, green) or Alexa Fluor 633–conjugated phalloidin (F, magenta). The cells were then analyzed by immunofluorescence microscopy. Higher magnifications of the areas indicated by white boxes in the upper panels are shown on the lower side. Scale bars, 5 μm. See also S9A–S9D Fig.
Fig 4.
P. gingivalis, but not S. gordonii or F. nucleatum, degrades JAM1 in IHGE cells.
(A) Schematic view of the structure of Myc-mCherry–tagged HA-inserted JAM1. The immature form of Myc-mCherry–tagged HA-inserted JAM1 was used as an internal control to monitor degradation flux of the mature form of HA-inserted JAM1. IHGE cells were transiently transfected with Myc-mCherry–tagged HA-inserted JAM1 plasmid. Following 48 h of incubation, cells were infected for 1 or 2 h with the indicated bacteria at an MOI of 100. The cells were then analyzed by immunoblotting using the indicated antibodies. (B–E) IHGE cells were transiently transfected with the HA-inserted JAM1 plasmid. Following 48 h of incubation, the cells were infected with P. gingivalis ATCC 33277 (B), P. gingivalis TDC60 (C), S. gordonii DL-1 (D), or F. nucleatum ATCC 25586 (E) at an MOI of 100 for the indicated times. The cells were then analyzed by immunoblotting with the indicated antibodies.
Fig 5.
The K134 and R234 residues are involved in degradation of JAM1 by P. gingivalis in IHGE cells.
(A) Schematic view of the JAM1 structure and derivatives. SP, IG-LIKE, or TMD domains are indicated by gray boxes. HA-tag is shown in green. The point mutations K134H and R234H are shown in magenta. (B) IHGE cells were transiently transfected with plasmid encoding HA-inserted JAM1 or the indicated JAM1 mutants. Following 48 h of incubation, the cells were infected with P. gingivalis at an MOI of 100 for 1 h, and then analyzed by immunoblotting using the indicated antibodies. (C) IHGE cells stably expressing HA-tagged JAM1 Δ (1–133) or HA-tagged JAM1 Δ (1–133) K134H R234H were infected for 1 h with P. gingivalis WT or the Δkgp ΔrgpA ΔrgpB mutant at an MOI of 100. The cells were then fixed, stained with DAPI (cyan) and anti-HA (yellow), and analyzed by immunofluorescence microscopy. Scale bars, 10 μm.
Fig 6.
JAM1 is required for epithelial barrier function of IHGE cells.
(A) Schematic image of the culture insert system. Monolayer of IHGE cells stably expressing shLuc, shJAM1 #110, or shJAM1 #508 were cultured in culture inserts. FITC-labeled tracer was added to culture media in the upper compartment. Following 30 min of incubation, the transmission of tracer from the upper compartment to the lower compartment was analyzed by spectrometry. (B) JAM1 expression in IHGE cells stably expressing shLuc, shJAM1 #110, or shJAM1 #508 was analyzed by immunoblotting with the indicated antibodies. (C–I) Permeability to 40 kDa FITC-dextran (C), 3–5 kDa FITC-dextran (D), 0.5 kDa Lucifer Yellow (E), FITC–P. gingivalis LPS (F), FITC–P. gingivalis PGN (G), FITC–E. coli LPS (H), or FITC–S. aureus PGN (I) in IHGE cells expressing shLuc and shJAM1. Results are expressed as fold change relative to cells expressing shLuc and are the means ± SD of eight technical replicates. *, p<0.05, two-tailed Dunnett’s test (C-G) or two-tailed t test (H, I).
Fig 7.
JAM1 is required for epithelial barrier function of gingival epithelial tissues.
(A, B) Schematic illustration (A) and confocal microscopic cross-sectional images (B) of 3D-tissue model expressing shLuc or shJAM1. Gingival epithelial tissues were fixed, stained with anti-JAM1 (white) and Alexa Fluor 568–conjugated phalloidin (magenta), and analyzed by confocal microscopy. Scale bars, 30 μm. (C–G) Permeability to 40 kDa FITC–dextran (C), FITC–P. gingivalis LPS (D), FITC–P. gingivalis PGN (E), FITC–E. coli LPS (F), and FITC–S. aureus PGN (G) in gingival epithelial tissues expressing shLuc and shJAM1. Results are expressed as fold change relative to epithelium expressing shLuc and are the means ± SD of seven technical replicates. *, p<0.05, one-tailed t test (C–E).
Fig 8.
P. gingivalis gingipains penetrate the epithelial barrier of IHGE cells.
(A, B) Schematic image of the culture insert system (A). A monolayer of IHGE cells was cultured in the upper compartment and on a coverslip in the lower compartment. Cells in the upper compartment were infected with P. gingivalis WT or the Δkgp ΔrgpA ΔrgpB mutant at an MOI of 100. Following 1 h of incubation, cells in the lower compartment were fixed, stained with DAPI (cyan) and anti-JAM1 (yellow), and analyzed by confocal microscopy (B). Scale bars, 10 μm. (C, D) Schematic image of the culture insert system (C). A monolayer of IHGE cells (WT) was cultured in the upper compartment and IHGE cells stably expressing Myc-mCherry–tagged HA-inserted JAM1 were cultured on a coverslip in the lower compartment. Cells in the upper compartment were infected with P. gingivalis WT or the Δkgp ΔrgpA ΔrgpB mutant at an MOI of 100. Following 1 h of incubation, cells in the lower compartment were fixed, stained with anti-HA (green), and analyzed by confocal microscopy (D). Scale bars, 10 μm. (E, F) Schematic image of the culture insert system (E). A monolayer of IHGE cells (WT) was cultured in the upper compartment and IHGE cells stably expressing HA-JAM1 Δ (1–133) or Δ (1–133) K134H R234H were cultured on a coverslip in the lower compartment. Cells in the upper compartment were infected with P. gingivalis WT or the Δkgp ΔrgpA ΔrgpB mutant at an MOI of 100. Following 1 h of incubation, cells in the lower compartment were fixed, stained with DAPI (cyan) and anti-HA (yellow), and analyzed by confocal microscopy (F). Scale bars, 10 μm. (G-I) Schematic image of the culture insert system (G). JAM1-expressing IHGE cells (WT or overexpressing JAM) were infected with P. gingivalis for 1 h and analyzed by immunoblotting with the indicated antibodies (H). A monolayer of IHGE cells (WT or overexpressing JAM1) was cultured in the culture insert. Cells were infected with P. gingivalis at an MOI of 100 in the upper compartment. Following 1 h of incubation, cells in the lower compartment were fixed, stained with DAPI (cyan) and anti-JAM1 (yellow), and analyzed by confocal microscopy (I). Scale bars, 10 μm.
Fig 9.
P. gingivalis degrades JAM1 of gingival epithelium, causing penetration of LPS and PGN.
(A, B) Schematic illustration of the three-dimensional culture (A) and confocal microscopic cross-sectional images (B) of the three-dimensional culture of IHGE cells. Gingival epithelial tissues (WT or overexpressing JAM1) were infected with P. gingivalis for 30 min. Tissues were then fixed, stained with anti-JAM1 (white) and Alexa Fluor 568–conjugated phalloidin (magenta), and analyzed by confocal microscopy. Scale bars, 30 μm. (C–G) Permeability to 40 kDa FITC-dextran (C), FITC–P. gingivalis LPS (D), FITC–P. gingivalis PGN (E), FITC–E. coli LPS (F), and FITC–S. aureus PGN (G) of gingival epithelial tissues (WT or overexpressing JAM1) infected with P. gingivalis. Three-dimensional tissues on culture inserts were infected with P. gingivalis and FITC-labeled tracer in the upper compartment. Following 30 min of incubation, the transmission of tracer from the upper compartment to the lower compartment was analyzed by spectrometry. Results are expressed as fold change relative to uninfected WT cells and are the means ± SD of seven technical replicates. *, p<0.05, one-tailed t test (closed testing procedure).
Fig 10.
Proposed model of how P. gingivalis gingipains send bacterial virulence factors through the gingival epithelium.
P. gingivalis gingipains degrade JAM1, which increases the permeability of gingival epithelium to gingipains and other factors. Subsequently, gingipains are transferred to the deeper epithelium to further degrade JAM1, which allows LPS and PGN to penetrate the gingival epithelium and reach subepithelial tissues. Finally, gingipains, LPS, and PGN induce inflammation in gingival tissues.