IADR Abstract Archives
Identification of Oral Pathogenic Bacteria by Signature Volatile Organic Compounds
Objectives: Breath testing for the identification of signature volatile organic compounds (VOCs) emitted by cariogenic and periopathogenic bacteria is a promising modality for early detection of susceptibility to caries and periodontal disease. Our study aims at investigating signature molecular patterns of common oral bacteria, as a first step toward the development of a computerized breath test.
Methods: Headspaces above colonies of S.mutans, S.sanguis, P.gingivalis and F.nucleatum were trapped after 24 and 48 hours of cultivation, together with aerobic and anaerobic controls, and analyzed using gas chromatography and mass spectrometry. Statistical analysis of the peak and area under the curve for each material was performed.
Results: Out of 228 materials detected, 181 were present in both control and trial headspaces, while 47 (20.6%) were present in trial headspaces and absent from the controls. A unique VOC signature, consisting of 20-35 molecules, was detected for each of the bacterial strains, differing by molecule types and spectrometer peak sizes (measured by AUC). Among these, two molecules were detected in the signature of P.gingivalis alone; similarly, two unique molecules were detected for F.nucleatum and three for S.mutans.
Conclusions: A unique signature for each bacterial strain allows the development of an algorithm for statistical detection of bacteria by their VOC profile alone. Some bacteria emit unique VOCs, allowing a dichotomous (rather than statistical) detection, thus further improving the diagnostic capability of the proposed algorithm. Additional research is required with other types and strains of bacteria – including samples from active periodontal and carious lesions.
Methods: Headspaces above colonies of S.mutans, S.sanguis, P.gingivalis and F.nucleatum were trapped after 24 and 48 hours of cultivation, together with aerobic and anaerobic controls, and analyzed using gas chromatography and mass spectrometry. Statistical analysis of the peak and area under the curve for each material was performed.
Results: Out of 228 materials detected, 181 were present in both control and trial headspaces, while 47 (20.6%) were present in trial headspaces and absent from the controls. A unique VOC signature, consisting of 20-35 molecules, was detected for each of the bacterial strains, differing by molecule types and spectrometer peak sizes (measured by AUC). Among these, two molecules were detected in the signature of P.gingivalis alone; similarly, two unique molecules were detected for F.nucleatum and three for S.mutans.
Conclusions: A unique signature for each bacterial strain allows the development of an algorithm for statistical detection of bacteria by their VOC profile alone. Some bacteria emit unique VOCs, allowing a dichotomous (rather than statistical) detection, thus further improving the diagnostic capability of the proposed algorithm. Additional research is required with other types and strains of bacteria – including samples from active periodontal and carious lesions.