Porphyromonas gingivalis aggravates atherosclerotic plaque instability by promoting lipid-laden macrophage necroptosis

Abstract

At advanced phases of atherosclerosis, the rupture and thrombogenesis of vulnerable plaques emerge as primary triggers for acute cardiovascular events and fatalities. Pathogenic infection such as periodontitis-associated Porphyromonas gingivalis (Pg) has been suspected of increasing the risks of atherosclerotic cardiovascular disease, but its relationship with atherosclerotic plaque destabilization remains elusive. Here we demonstrated that the level of Pg-positive clusters positively correlated with the ratio of necrotic core area to total atherosclerotic plaque area in human clinical samples, which indicates plaque instability. In rabbits and Apoe^-/- mice, Pg promoted atherosclerotic plaque necrosis and aggravated plaque instability by triggering oxidative stress, which led to macrophage necroptosis. This process was accompanied by the decreased protein level of forkhead box O3 (FOXO3) in macrophages. The mechanistic dissection showed that Pg lipopolysaccharide (LPS) evoked macrophage oxidative stress via the TLR4 signaling pathway, which subsequently activated MAPK/ERK-mediated FOXO3 phosphorylation and following degradation. While the gingipains, a class of proteases produced by Pg, could effectively hydrolyze FOXO3 in the cytoplasm of macrophages. Both of them decreased the nuclear level of FOXO3, followed by the release of histone deacetylase 2 (HDAC2) from the macrophage scavenger receptor 1 (Msr1) promoter, thus promoting Msr1 transcription. This enhanced MSR1-mediated lipid uptake further amplified oxidative stress-induced necroptosis in lipid-laden macrophages. In summary, Pg exacerbates macrophage oxidative stress-dependent necroptosis, thus enlarges the atherosclerotic plaque necrotic core and ultimately promotes plaque destabilization.

Fig. 1

Pg contributes to atherosclerotic plaque vulnerability. a Intraplaque hemorrhage (IPH) detection with Prussian blue staining of human coronary atherosclerotic plaques. Scale bar = 100 μm. b Comparisons of the plaque numbers showing IPH in low, medium, and high groups of human coronary arteries. Group Low, n = 12; group medium, n = 10, group high, n = 11. c H&E, Masson, Oil Red O staining (scale bar = 200 μm), and CD45 and CD68 co-staining (scale bar = 100 μm) (macrophage marker) of human coronary plaques. Nuclei were labeled using DAPI. d Quantification of the ratio of necrotic area, collagen content, Oil Red O⁺ area, CD45 and CD68 positive areas of human coronary plaques, and the thickness of fibrous cap. Group low, n = 12; group medium, n = 10, group high, n = 11. Prussian blue staining of atherosclerotic plaques in aortic arches of rabbits with periodontal ligature and treated with or without Pg for 16 weeks (e) (left scale bar = 200 μm, right scale bar = 100 μm), and aortic roots of Apoe^−/− mice challenged with or without Pg for 20 weeks (i) (scale bar = 100 μm). Quantification of the plaque numbers showing IPH in aortic samples of rabbits (f), and Apoe^−/− mice (j). n = 6 per group. H&E, Masson, and Oil Red O staining, and CD45 and F4/80 co-staining (macrophage marker) of atherosclerotic plaques in aortic arches of rabbits (g) (scale bar = 200 μm), and aortic roots of Apoe^−/− mice (k) (scale bar = 200 μm in H&E, scale bar = 100 μm in the rest images). Nuclei were labeled using DAPI. Quantitative analyses of plaque size, necrotic area, collagen content, Oil Red O⁺ area, thickness of fibrous cap, CD45 and F4/80-positive areas of atherosclerotic plaques of rabbits (h), and Apoe^−/− mice (l). n = 6 per group. Results were represented as mean ± SEM. Data were analyzed by Fisher exact test (b, f, j), by Kruskal–Wallis test (necrotic area analysis in d), or by one-way ANOVA (d, h, l). ****P < 0.0001; ***P < 0.001; **P < 0.01; *P < 0.05

Fig. 2

Pg infection induces lipid-laden macrophage necroptosis enlarging necrotic core. a Non-parametric Spearman’s test between the ratio of necrotic area and the amount of Pg positive clusters within the human coronary atherosclerotic plaques. r = 0.7902, P < 0.0001. n = 33. b–g Quantification of the proportion of dying macrophages probed by TUNEL and CD45-CD68/F4/80 co-staining in human coronary plaques (b, c) (group low, n = 7; group medium, n = 7, group high, n = 6), rabbit aortic arch plaques (d, e) (n = 6 per group) and mouse aortic root plaques (f, g) (n = 6 per group). Nuclei were labeled using DAPI. Scale bar = 100 μm on top, and 20 μm on the bottom. h H&E, Masson, and Oil Red O staining of atherosclerotic plaques in aortic roots of Apoe^−/− mice untreated or infected with Pg and/or administrated with vehicle or liposomal clodronate (LC) for 8 weeks. Scale bar = 200 μm in H&E, scale bar = 100 μm in the rest images. i Quantitative analyses of plaque size, necrotic area, collagen content, Oil Red O⁺ area, thickness of fibrous cap of the plaques. n = 6 per group. j The ratio of PI⁺ cells in macrophages over a time course of stimulation with or without ox-LDL (60 μg/mL) and/or Pg (MOI = 100). n = 4 per group. k Flow cytometry analyses of the ratio of necrotic (PI⁺Annexin V⁺) and apoptotic (PI⁻Annexin V⁺) macrophages under ox-LDL (60 μg/mL) and/or Pg (MOI = 100) treatment. n = 4 per group. l The ratio of PI⁺ cells in ox-LDL loaded macrophages pre-administrated with Z-VAD-FMK (5 μM), Nec-1 (10 μM), ferrostatin-1 (FER-1, 5 μM), or VX765 (50 μM) and infected with or without Pg (MOI = 100). n = 4 per group. The co-localization of p-MLKL and CD68 or F4/80-labled macrophages in the atherosclerotic plaques of human coronary arteries (m), rabbit aortic arches (n), and mouse aortic roots (o). Nuclear DNA (blue) was counterstained with DAPI. Scale bar = 100 μm. Results were represented as mean ± SEM (c, e, g, i) or mean ± SD (j, k, l). Data were analyzed by non-parametric Spearman’s test (a), one-way ANOVA (c, i, k), unpaired Student t test (e) and two-way ANOVA (g, j, l). ****P < 0.0001; ***P < 0.001; **P < 0.01; *P < 0.05

Fig. 3

Oxidative stress serves as a principal driver of Pg–induced necroptosis. a Dihydroethidium (DHE) staining and CD45-ACTA2 co-staining of serial sections of the aortic roots isolated from Apoe^–/– mice infected with or without Pg for 8 weeks. Nuclei were labeled using DAPI. Scale bar = 100 μm. b The co-localization of NOX2 and CD45-labled macrophages in the aortic root plaques of Apoe^–/– mice infected with or without Pg for 8 weeks. Nuclei were labeled using DAPI. Scale bar = 20 μm. c Flow cytometry analyses of the reactive oxygen species (ROS) level in macrophages infected by Pg in different MOIs, and treated with or without ox-LDL (60 μg/mL). n = 4 per group. d Relative mRNA expression levels of Nox2, Cox2, Nos2, Gpx1, Sod1, and Sod2 in ox-LDL (60 μg/mL)-loaded macrophages treated with or without Pg (MOI = 100) for 24 h. n = 4 per group. Flow cytometry analyses of ROS production (e) and ratio of PI⁺ cells (f) in ox-LDL (60 μg/mL)-loaded macrophages pre-administrated with EUK134 (10 μM) and infected with Pg (MOI = 100). n = 4 per group. g Western blot analyses of RIPK3, p-MLKL, MLKL, and GAPDH expression in ox-LDL (60 μg/mL)-loaded macrophages pre-administrated with EUK134 (10 μM) and infected with Pg (MOI = 100) for 24 h. GAPDH was used as the loading control. h H&E, Masson, Oil Red O staining, and CD45 and F4/80 co-immunohistochemical staining of atherosclerotic plaques in aortic roots of Apoe^−/− mice untreated or challenged by Pg under the administration of EUK134 for 8 weeks. Scale bar = 200 μm in H&E, scale bar = 100 μm in the rest images. i Quantitative analyses of plaque size, necrotic area, Oil Red O⁺ area, and CD45 and F4/80-positive areas of the plaques in h. n = 6 per group. Results were represented as mean ± SD (c, d, e, f) or mean ± SEM (i). Data were analyzed by two-way ANOVA (c, d), or one-way ANOVA (e, f, i). ****P < 0.0001; ***P < 0.001; **P < 0.01; *P < 0.05

Fig. 4

Gingipains play important roles in Pg-aggravated oxidative stress. Flow cytometry analyses of the ROS production (a) and the ratio of PI⁺ cells (b) in ox-LDL (60 μg/mL) loaded macrophages infected with Pg or KDP136 (MOI = 100). n = 4 per group. c Western blot analyses of RIPK3, p-MLKL, MLKL, and GAPDH expression in ox-LDL (60 μg/mL)-loaded macrophages infected with Pg or KDP136 (MOI = 100) for 24 h. GAPDH was used as the loading control. Flow cytometry analyses of the ox-LDL uptake (d), the ROS production (e), and the ratio of PI⁺ cells (f) in ox-LDL-loaded macrophages exposed to RgpA, RgpB, or Kgp (1 μg/mL). n = 5 per group in (d), n = 4 per group in (e, f). g Western blot analyses of RIPK3, p-MLKL, MLKL, and GAPDH expression in ox-LDL (60 μg/mL)-loaded macrophages treated with RgpA, RgpB, or Kgp (1 μg/mL) for 24 h. GAPDH was used as the loading control. h The co-staining of RgpA and CD68 (macrophage marker) in the coronary plaques of human. Scale bar = 20 μm. i H&E, Oil Red O staining, and CD45 and F4/80 co-staining of the aortic root plaques from Apoe^−/− mice infected with or without Pg or KDP136 for 8 weeks. Scale bar = 200 μm in H&E, scale bar = 100 μm in the rest images. j Quantitative analyses of plaque size, necrotic area, Oil Red O⁺ area, CD45, and F4/80-positive areas of the plaques in i. n = 6 per group. Results were presented as mean ± SD (a, b, d, e, f) or mean ± SEM (j). All data were analyzed by one-way ANOVA. ****P < 0.0001; ***P < 0.001; **P < 0.01; *P < 0.05

Fig. 5

Pg-enlarged oxidative stress mostly results from enhanced MSR1-mediated ox-LDL uptake. a TEM images of the aortic arch plaque of Apoe^−/− mice infected with or without Pg for 8 weeks. Scale bar = 2 μm on the left, and 500 nm on the right. b Flow cytometry analyses of the ratio of PI⁺ macrophages and the ox-LDL accumulation in PI⁺ macrophages under ox-LDL (60 μg/mL) and Pg (MOI = 100) challenge for 24 h. n = 4 per group. c qRT-PCR analyses of Cd36, Msr1, Olr1 expression in macrophages treated with or without ox-LDL (60 μg/mL) and/or Pg (MOI = 100) for 24 h. β-actin was used as control. n = 4–6 per group. Western blot analyses of MSR1, CD36, OLR1, or GAPDH expression in ox-LDL (60 μg/mL)-loaded macrophages treated with Pg (MOI = 100) (d), or RgpA, RgpB, or Kgp (1 μg/mL) (e) for 24 h. GAPDH was used as the loading control. The co-localization of MSR1 and CD45-labled macrophages in the plaques of Apoe^–/– mice aortic roots (f), rabbit aortic arches (g), and human coronary vessels (h). Scale bar = 50 μm. Flow cytometry analyses of the ox-LDL uptake (i), ROS production (j), and the ratio of PI⁺ cells (k) in Msr1-knockdown macrophages loaded with ox-LDL (60 μg/mL) and infected with Pg (MOI = 100). n = 4 per group. Flow cytometry analyses of the ox-LDL uptake (l), ROS production (m), and the ratio of PI⁺ cells (n) in fucoidan (40 μg/mL) pretreated macrophages loaded with ox-LDL (60 μg/mL) and challenged by Pg (MOI = 100). n = 4 per group. o The co-staining of p-MLKL and F4/80 (macrophage marker) in the aortic root plaques from Apoe^−/− mice treated with fucoidan and/or Pg for 8 weeks. Scale bar = 100 μm. p H&E and Oil Red O staining of aortic root plaques from Apoe^−/− mice treated with fucoidan and/or Pg for 8 weeks. Quantitation of the plaque size, as well as the ratios of necrotic area and Oil Red O⁺ area to plaque area were at right. n = 6 per group. Scale bar = 200 μm in H&E, scale bar = 100 μm in Oil Red O. Results were expressed as mean ± SD (b, c, i–n) or mean ± SEM (p). Data were analyzed by unpaired two-tailed student t-test (b), two-way ANOVA (c) or one-way ANOVA (i–n, p). ****P < 0.0001; ***P < 0.001; **P < 0.01; *P < 0.05

Fig. 6

Proteolysis of FOXO3 by gingipains upregulates the transcription of MSR1. a GO enrichment analysis of differential proteins between ox-LDL (60 μg/mL)-loaded macrophages treated with Pg (MOI = 100) versus PBS for 24 h (n = 4 per genotype). P < 0.05 represented statistically significant difference, and upregulated or downregulated proteins were identified with a fold-change (Pg/PBS) > 1.15 or <0.87. qRT-PCR (b) and Western blot (c) analyses of TPT1, BNIP3, FOXO3, CD74, HDAC4, PDCD4, and TAX1BP1 expression in ox-LDL (60 μg/mL)-loaded macrophages infected with Pg or KDP136 (MOI = 100) for 24 h. β-actin was used as control, and n = 4 per group (b). GAPDH was employed as the loading control in (c). Western blot analyses (d) and immunohistochemical staining (e) of FOXO3 in ox-LDL (60 μg/mL)-loaded macrophages infected with Pg, KDP136 (MOI = 100), RgpA, RgpB, Kgp, or Pg-LPS (1 μg/mL) for 24 h. GAPDH was employed as the loading control in (d). Scale bar = 10 μm in (e). The co-staining of FOXO3 and F4/80 (macrophage marker) in the plaques of rabbit aortic arches (f), and mouse aortic roots (g). Scale bar = 20 μm. h Silver staining of recombinant human FOXO3 (rFOXO3, 207.9 μM) incubated with RgpA, RgpB or Kgp (2.1 μM) for 60 min at 37 °C. Red arrowhead points to original rFOXO3. Blue arrowhead points to RgpA. Orange arrowhead points to RgpB. Green arrowhead points to Kgp. Dashed lines encircle rFOXO3 fragments. i Western blot analyses of MSR1 expression in Foxo3 siRNA (si-Foxo3) transfected macrophages loaded with ox-LDL (60 μg/mL) and stimulated with RgpA, RgpB, or Kgp (1 μg/mL) for 24 h. GAPDH was used as control. Flow cytometry analyses of the ox-LDL uptake (j), ROS production (k), and the ratio of PI⁺ cells (l) in si-Foxo3 transfected macrophages treated with RgpA, RgpB, or Kgp (1 μg/mL) in the presence of ox-LDL (60 μg/mL). n = 4 per group. m Flow cytometry analyses of the ratio of PI⁺ cells in si-Foxo3 transfected macrophages pre-administrated with or without fucoidan (40 μg/mL) and exposed to ox-LDL (60 μg/mL) for 24 h. n = 4 per group. n Flow cytometry analyses of the proportion of PI⁺ cells in Msr1 and/or Foxo3 knockdown macrophages treated with ox-LDL (60 μg/mL) for 24 h. n = 4 per group. Results were expressed as mean ± SD. Data were analyzed by two-way ANOVA (b) or one-way ANOVA (j–n). ****P < 0.0001; ***P < 0.001; **P < 0.01; *P < 0.05

Fig. 7

HDAC2 is recruited to the promoter by FOXO3 and inhibits MSR1 transcription. a Potential FOXO3 binding sites on mouse Msr1 promoter predicted by JASPAR database. b CHIP was conducted using FOXO3 antibody and IgG (negative control). c Vector information of the wild-type Msr1 promoter (−496 /−490) and its mutation/depletion mutants. d Dual luciferase reporter assay of the activities of the wild-type Msr1 promoter, and its mutation and depletion mutants in macrophages loaded with ox-LDL (60 μg/mL). n = 4 per group. e Dual luciferase reporter assay of the activities of the wild-type Msr1 promoter in macrophages loaded with ox-LDL (60 μg/mL) and infected with Pg or KDP136 (MOI = 100). n = 4 per group. f, g CHIP-PCR analyses of the histone H3 and histone H4 acetylation level in ox-LDL (60 μg/mL)-loaded macrophages treated with Pg or KDP136 (MOI = 100). n = 4 per group. qRT-PCR analyses of Msr1 expression (h), flow cytometry analyses of the ox-LDL uptake (i) and the ratio of PI⁺ cells (j) in ox-LDL (60 μg/mL)-loaded macrophages pre-administrated with TSA (1 μM) and challenged with Pg (MOI = 100) for 24 h. n = 4 per group. k FOXO3 and HDAC2 binding detection by Co-IP in ox-LDL (60 μg/mL) loaded macrophages infected with Pg or KDP136 (MOI = 100) for 24 h. l FOXO3 and HDAC2 binding detection by Co-IP in si-Foxo3 transfected macrophages upon ox-LDL (60 μg/mL) stimulation for 24 h. m Schematic diagram showing the mechanism of Pg-promoted macrophage necroptosis. In macrophages, gingipains secreted by Pg directly hydrolyze FOXO3 in the cytoplasm, thereby inhibiting its nuclear translocation. Additionally, Pg-LPS activates MAPK/ERK-mediated phosphorylation of FOXO3, promoting its nuclear export, and subsequent ubiquitination (Ub) and degradation. The reduction in FOXO3 levels diminishes HDAC2 binding to the MSR1 promoter and increases promoter acetylation, leading to upregulation of MSR1 expression. This enhanced MSR1 expression promotes lipid uptake and the following oxidative stress, ultimately triggering macrophage necroptosis. Diagram is generated by figdraw.com. Results were represented as mean ± SD. Data were analyzed by one-way ANOVA (d, e, h, i, j), or Kruskal–Wallis test (f, g). ****P < 0.0001; ***P < 0.001; **P < 0.01; *P < 0.05