A community analysis of dental plaque in pre-pubertal children

Healthy children, aged between 5 and 9 years who had not taken antibiotics in the preceding three months were recruited. Plaque was sampled from the gingival crevice of either the lower left or lower right first permanent molar to estimate the prevalence of Porphyromonas gingivalis, Actinobacillus actinomycetemcomitans and Tannerella forsythensis which have been implicated as main etiological agents of periodontal disease. Analysis of plaque from subjects without gingivitis (n = 65) and those with gingivitis (n = 53) by PCR targeting of the 16S rRNA gene demonstrated negligible differences in the prevalence of these pathogens with the exception of T. forsythensis. This pathogen was detected more frequently in children with no gingivitis (P = 0.03). Community analysis of plaque was carried out to determine (i) any significant differences in the microbiota of both cohorts and (ii) if specific taxa influenced the prevalence of the three periodontal pathogens. Community analysis was attempted by using both culture dependent and culture independent techniques. The culture dependent technique involved community level physiological profiling (CLPP) and attempted to measure metabolic differences between cohorts. Preliminary studies demonstrated that this technique was not suitable for small sample volumes, such as plaque from single sites and was abandoned. The culture independent technique, denaturing gradient gel electrophoresis (DGGE), is a rapid and cost effective method for analysing bacterial communities. Statistical analysis of the DGGE data for both cohorts suggested that bacterial diversity was lower in subjects with gingivitis (P = 0.009). Logistic regression analysis of the DGGE banding patterns demonstrated that specific bands were significantly associated with gingival health and with gingivitis. Similarly, other bands were significantly associated with the prevalence of the three periodontal pathogens. These bands were excised and PCR-cloned. The 16S rRNA sequence data for these clones demonstrated that these DGGE bands were mixed with DNA of multiple taxa. Further attempts to separate individual excised bands over shorter denaturing gradients proved unsuccessful.