Low-abundance biofilm species orchestrates inflammatory periodontal disease through the commensal microbiota and complement

Abstract

Porphyromonas gingivalis is a low-abundance oral anaerobic bacterium implicated in periodontitis, a polymicrobial inflammatory disease, and the associated systemic conditions. However, the mechanism by which P. gingivalis contributes to inflammation and disease has remained elusive. Here we show that P. gingivalis, at very low colonization levels, triggers changes to the amount and composition of the oral commensal microbiota leading to inflammatory periodontal bone loss. The commensal microbiota and complement were both required for P. gingivalis-induced bone loss, as germ-free mice or conventionally raised C3a and C5a receptor-deficient mice did not develop bone loss after inoculation with P. gingivalis. These findings demonstrate that a single, low-abundance species can disrupt host-microbial homeostasis to cause inflammatory disease. The identification and targeting of similar low-abundance pathogens with community-wide impact may be important for treating inflammatory diseases of polymicrobial etiology.

Figure 1. Oral inoculation of SPF (but not GF) mice with P. gingivalis causes periodontal bone loss and elevation of the commensal bacterial load

Ten- to twelve-week-old specific-pathogen-free (SPF) or germ-free (GF) C3H mice were orally inoculated with P. gingivalis (Pg) or vehicle only (Sham). Six weeks later, the mice were assessed for periodontal bone loss (A, SPF; B, GF), levels of total cultivatable oral anaerobic bacteria (C), and changes to the qualitative composition of the microbiota detected by aerobic or anaerobic culture (D). (E) Pg and total bacteria were enumerated in the periodontal tissue of Pg-inoculated mice by real-time PCR of the ISPg1 gene (Pg) or the 16S rRNA gene (total oral bacteria). ISPg1 was selected to increase the sensitivity of Pg detection since this gene is present in 31 copies in the Pg genome (the gene copy numbers were thus divided by 31 to obtain genome equivalents). Negative ‘mm change in bone’ values indicate bone loss relative to bone levels of sham mice at the end of the experiment. Lactobacilli were detected in similar numbers following aerobic and anaerobic culture of samples from sham-treated mice, but as they represented a very low percentage of the total anaerobic counts (<0.01%), they do not appear in the anaerobic bar chart in D. In A, B, and E, data are means ± SD (n ≥ 5 mice per group). In C, CFU data are shown for each individual mouse with horizontal lines denoting mean values. In D, each organism was expressed as log₁₀ CFU and shown as a proportion of the total cultured organisms. *p <0.05; **p <0.01 versus corresponding sham control.

Figure 2. Natural periodontal bone loss in specific-pathogen-free (SPF) but not germ-free (GF) mice

(A) SPF and GF mice at the indicated ages were assessed for periodontal bone levels. Negative values indicate bone loss relative to 5-week-old SPF mice. (B) Gingiva were dissected from 10-week-old SPF or GF mice and gingival mRNA levels of the indicated molecules were determined by quantitative real-time PCR (normalized against GAPDH mRNA and expressed as fold change in SPF transcript levels relative to GF transcripts levels, which were assigned an average value of 1). Data are means ± SD (n = 5 to 6 mice per group). *p <0.05; **p <0.01 between age-matched SPF and GF mice.

Figure 3. Transmission of the commensal oral microbiota to co-caged GF mice and development of periodontal bone loss after an extended period

(A–B) Eight- to ten-week-old GF C3H mice were co-caged with SPF C3H mice. Oral swabs were taken at the indicated timepoints and the bacteria were cultured aerobically (A) and anaerobically (B). By day 14, there was complete transmission of the cultivatable oral microbiota from the SPF to the GF mice. (C) GF mice were co-caged with SPF mice and were assessed for periodontal bone levels at 6 and 16 weeks after co-caging. Negative values indicate bone loss relative to bone levels of mice maintained under GF conditions for the entire period (GF control); data are means ± SD (n = 5 to 8 mice per group). **p <0.01 versus GF control.

Figure 4. Complement-dependent elevation of the oral bacterial load and induction of bone loss

(A–B) Ten- to twelve-week-old BALB/c mice, either wild-type (WT) or deficient in C3aR (C3aR^−/−) or C5aR (C5aR^−/−), were orally inoculated with P. gingivalis (Pg) or vehicle only (Sham) and assessed for bone loss (A) and levels of oral anaerobic bacteria (B). (C) Quantitative detection, by real-time PCR of the ISPg1 gene, of wild-type Pg or a gingipain-deficient isogenic mutant (KDP128) in the periodontal tissue of inoculated WT BALB/c mice. ND, not detected. (D) WT, C3aR^−/−, or C5aR^−/− mice were orally inoculated with Pg or KDP128 and assessed for levels of oral anaerobic bacteria seven days later. (E–F) WT mice were inoculated with Pg or KDP128 and euthanized six weeks post-inoculation: Gingiva were dissected and processed for quantitative real-time PCR to determine gingival mRNA expression of the indicated molecules (normalized against GAPDH mRNA levels and presented as fold change relative to sham mice) (E), whereas defleshed maxillae were assessed for periodontal bone loss (F). (G) Effects of C5aR antagonist (C5aRA), or inactive analog control (iC5aRA ctrl), on the numbers of P. gingivalis or total bacteria in the periodontal tissue of mice with or without previous inoculation with P. gingivalis (for details see Experimental Procedures). ND, Pg not detected. (H) Periodontal bone levels in C3aR^−/− and C5aR^−/− mice relative to WT SPF or GF mice. Data are means ± SD (A, n = 9; C, E, F, and G, n = 5; H, n = 8 mice per group). In A, F, and H, negative values indicate bone loss relative to bone levels of the indicated controls (zero baseline), whereas positive values indicate increased bone levels. CFU counts are shown for each individual mouse with horizontal lines denoting mean values. *p <0.05; **p <0.01 versus corresponding control.

Figure 5. Compromised homeostasis in the periodontal tissue

(A) BALB/c mice were orally inoculated with P. gingivalis (Pg) or vehicle only (sham) and sacrificed 4 days or 6 weeks post-inoculation. Dissected gingiva were processed for real-time PCR to determine mRNA expression of the indicated molecules (normalized against GAPDH mRNA and shown as fold change relative to 4-day sham-infected mice). Data are means ± SD (n = 5 mice per group). (B) Diminished recruitment of neutrophils to the gingiva of LFA-1^−/− mice. Sagittal sections of periodontal tissue from 5-week-old wild-type (WT) or LFA-1^−/− mice were stained for the neutrophil marker LyG6. Shown are representative overlays of differential interference contrast and fluorescent confocal images. G, gingiva; S, sulcus; T, teeth. (C–D) WT and LFA-1^−/− mice were assessed for levels of oral anaerobic bacteria (C) and periodontal bone loss (D) as they aged from 5 to 16 weeks. (E–F) LFA-1^−/− mice given antibiotics-containing drinking water or plain water (control) were assessed for levels of oral anaerobic bacteria (E) and bone loss (F). (G–H) Eighteen- to twenty-week-old WT and CXCR2^−/− mice were assessed for oral anaerobic bacteria (G) and periodontal bone levels (H). (I–K) Ten-week-old WT or LFA-1^−/− mice were orally inoculated with P. gingivalis (Pg) or vehicle only (Sham) and six weeks later were assessed for levels of oral anaerobic bacteria (I) and bone loss (K). CFU are shown for each mouse and horizontal lines denote mean values. Bone loss data are means ± SD (n = 5 to 7 mice per group). *p <0.05; **p <0.01 versus corresponding control.