Periostin is down-regulated during periodontal inflammation

Abstract

Periostin, a matricellular adapter protein highly expressed by periodontal ligament fibroblasts, is implicated in the maintenance of periodontal integrity, which is compromised during periodontal diseases. The aim of this study was to explore the influence of chronic periodontal inflammation on tissue periostin levels. Periodontal breakdown was induced in a pre-clinical ligature periodontal inflammatory disease model. Periodontal tissue specimens were harvested at baseline, 2 weeks, and 4 weeks and prepared for histologic, immunofluorescence, and micro-CT examination. Statistical analyses were conducted by Kruskal-Wallis, Mann-Whitney, and Spearman's tests. Periostin detection levels were reduced over time in response to the inflammatory process (1 ± 0.05; 0.67 ± 0.03; 0.31 ± 0.02; p < 0.001; baseline, 2, and 4 weeks, respectively). Simultaneously, alveolar bone loss increased from baseline to the 2- and 4-week time-points (0.40 ± 0.02 mm; 1.39 ± 0.08 mm; 1.33 ± 0.15 mm; p < 0.001), which was inversely correlated with the levels of periostin (ρ = -0.545; p < 0.001). In conclusion, periostin PDL tissue levels significantly decrease under chronic inflammatory response and correlate with the detrimental changes to the periodontium over time.

Trial registration: ClinicalTrials.gov NCT01180920.

Figure 1.

µCT analyses. (A) Volumetric and linear µCT images depicting ROIs (yellow cubes), used for volumetric measurements, and CEJ (green arrows) and ABC (red arrows) anatomical landmarks, used for ABL measurements. Scale bar = 1 mm. (B) ABL (mm) was significantly lower at baseline than at 2 and 4 wks (a, p < 0.001; Mann-Whitney). (C) BMD (mg/cc) was significantly lower at 2 (b, p < 0.003; Mann-Whitney) and 4 wks (c, p < 0.002; Mann-Whitney) after the ligature placement. Error bars represent SEM.

Figure 2.

Immunohistochemical analyses of the inflammatory process. (A) H&E images showing inflammatory infiltrate and periodontal fiber disorganization over time. Original magnification: 40X. Scale bar: 50 µm. (B) Myeloperoxidase- and (C) CD3-immunopositive and (D) apoptotic cell detection (black arrows). Original magnification: 60X. Scale bar: 40 µm. (E) Myeloperoxidase- and (F) CD3-immunopositive and (G) apoptotic cell quantification showing a generalized increase of all markers compared with baseline (a, p < 0.001, Mann-Whitney). Additionally, CD3-immunopositive and apoptotic cells significantly increased from 2 to 4 wks (b, p < 0.011 and c, p < 0.008, Mann-Whitney, respectively). Error bars represent SEM.

Figure 3.

Histological and immunofluorescence analyses. (A) H&E histological images showing alveolar bone reduction, inflammatory infiltrate, and periodontal fiber disorganization in the ligated interproximal region between the first and second molars over time. Original magnification: 20X. Scale bar: 100 µm. (B) Immunofluorescence images showing reduction of periostin immunolocalization (in red) over time. DAPI in blue. Original magnification: 20X. Scale bar: 100 µm. (C) Heat-map 3D representation of periostin immunofluorescence intensity from images presented in (B). (D) Significantly lower relative periostin levels in the PDL area at 2 and 4 wks compared with baseline (a, p < 0.001; Mann-Whitney). Levels at 4 wks were also significantly lower than those at 2 wks (b, p < 0.001; Mann-Whitney). Error bars represent SEM.

Figure 4.

Correlation between levels of periostin and bone destruction. (A) Correlation between mean levels of periostin and ABL (r² linear = 0.681; ρ = −0.545, p < 0.001, Spearman). (B) Correlation between periostin levels and BMD (r² linear = 0.565; ρ = 0.411, p < 0.012, Spearman). Error bars represent SEM for both the horizontal (periostin) and the vertical (ABL or BMD) axes. Orange dotted line: Confidence interval = 80%.