Method: 20 PER+ACS patients, 14 PER-ACS patients and 15 HS subjects were recruited from Gazi University Hospital and Department of Periodontology, Ankara, Turkey. Paper points were used for sampling subgingival microbiota from 6 most inflamed periodontal sites. The samples were pooled per subject. After DNA extraction, 16S rRNA genes were PCR amplified and DNA hybridization performed using a microarray (HOMIM) for over 300 bacterial phylotypes. Partial least squares projections to latent structures-discriminant analysis (PLS-DA) was used for multivariate modeling and analysis of the data.
Result: Altogether, 130 phylotypes/bacterial groups were identified, 112 in PER+ACS, 97 in PER-ACS and 102 in HS. PLS-DA showed separation of the 3 study groups. Regression coefficient plots for a PLS model where the phylotypes referred to x data and the 3 subject groups y data (y1, y2, y3) revealed phylotypes significantly positively or negatively correlating to the 3 groups. Phylotypes with the highest positive correlations differed among the groups and included Granulicatella, Veillonella, Leptotrichia, Streptococcus, Synergistes and Eubacterium spp. for PER+ACS, Campylobacter, Desulfobulbus, Eubacterium, Treponema, Dialister, Prevotella, Parvimonas, Streptococcus, Tannerella, and Filifactor spp. for PER-ACS and Capnocytophaga, Kingella, Propionibacterium, Streptococcus, Lautropia, Sphaerocytophaga and Slackia spp. for HS.
Conclusion: Present bacterial findings, including the significance of Granulicatella elegans and several Veillonella species for distinguishing the PER+ACS group, encourage further studies on the structure, function and systemic importance of subgingival bacterial communities in periodontitis.