Title: 2936 - Metabolomic Profiling of Antibody Response to Periodontal Pathogens


Jaakko Leskelä (Presenter)
University of Helsinki

Aino Salminen, University of Helsinki
Teemu Palviainen, Institute for Molecular Medicine Finland
Milla Pietiäinen, University of Helsinki
Anne-Mari Määttä, University of Helsinki
Susanna Paju, University of Helsinki
Juha Sinisalo, University of Helsinki and Helsinki University Hospital
Veikko Salomaa, National Institute for Health and Welfare
Jaakko Kaprio, Institute for Molecular Medicine Finland
Pirkko Pussinen, University of Helsinki


Objectives: Detection of periodontopathogen antibodies in circulation indicate exposure to periodontal pathogens. Inflammation, such as present in periodontal disease, has shown to modify lipoprotein metabolism and composition. In this study, we investigated the association between serum periodontopathogen antibody levels and various metabolite levels.

Methods: Study population consisted of 2398 individuals whose serum antibody levels against P. gingivalis and A. actinomycetemcomitans were determined by multiserotype-ELISA. In FinnTwin16 (n=551) and Parogene (n=492) both serum IgA and IgG antibody levels were available, and in FINRISK97 (n=1355) IgG antibody levels were available. We used Nuclear Magnetic Resonance (NMR)-metabolomics platform to determine in total 68 metabolic variables, including lipoprotein particles, fatty acids, amino acids, metabolic substrates, glycoproteins, and ketone bodies. With hypothesis-free approach, we estimated the associations between antibody levels and metabolic variables using linear regression model adjusted for age, sex, smoking, and kinship (FinnTwin16). Results were meta-analysed using Inverse-variance-weighted fixed-effect meta-analysis. We calculated principal components explaining >95% of the overall variation in FR97 to evaluate appropriate significance level taking into account multiple testing.

Results: The corrected significance level was set to p<0.00238. In the meta-analysis P. gingivalis IgG antibody levels associated negatively with Apolipoprotein A-I (p=0.00012, beta = -0.079), HDL-bound cholesterol (p=0.00069, beta = -0.067), and especially cholesterol bound to larger HDL particles (p=0.00017, beta = -0.078). In addition, there was negative association in P. gingivalis IgA antibody levels with Serum total cholesterol (p=0.0014, beta = -0.104) and A. actinomycetemcomitans IgG with Concentration of very large HDL particles (p=0.0011, beta = -0.060).

Conclusions: Exposure to the major periodontal pathogens, P. gingivalis and A. actinomycetemcomitans, is associated with a proatherogenic lipoprotein profile, especially low HDL cholesterol levels. This is in line with previous findings where periodontal disease is shown to associate with atherogenic diseases.

Disclosure Statement:
The submitter must disclose the names of the organizations with which any author have a relationship, the nature of the relationship, and the clinical or research area involved. The following is submitted: NONE