Quantitation of Bacteroides forsythus in subgingival plaque - Comparison of immunoassay and quantitative polymerase chain reaction

Our objective was to compare three methods (enzyme-linked immunosorbent ass ay [ELISA], endpoint and quantitative polymerase chain reaction [E-PCR and Q-PCR]) for detection and quantitation of Bacteroides forsythus in 56 plaqu e samples from seven subjects with progressive periodontal disease. Samples collected in buffer were pelleted and resuspended in 500 mu l of water. Fi fty mu l aliquots were removed for an ELISA performed on bacteria or plaque immobilized on 96-well plates and probed with B. forsythus specific antibo dy. An occurrence of 3.7+/-0.6 . 10(4) or more bacteria were detected by EL ISA in pure culture; 26 of 54 plaque samples were positive, two samples cou ld not be analyzed. Samples for PCR were autoclaved for 10 min prior to use . The detection level of E-PCR using primers specific for B.forsythus 16S r RNA was 200 cells and 42 out of 56 samples were positive based on ethidium bromide stained agarose gels. Q-PCR using the same primers combined with a nested fluorescent oligonucleotide probe detected 10+/-0.32 bacteria in pur e culture; 43 of 56 plaque samples were positive. The ELISA and Q-PCR obtai ned identical results with 36 of the 54 samples assayed; there were one fal se positive and 17 false negative ELISA results using Q-PCR as standard. Th e positive proportions of plaque samples were almost the same for E-PCR and Q-PCR. We conclude that the PCR methods are more appropriate for a multice nter study because of greater sensitivity and convenience of sample transpo rtation from clinics to a central laboratory. (C) 2000 Published by Elsevie r Science B.V. All rights reserved.