Distinct and complex bacterial profiles in human periodontitis and health revealed by 16S pyrosequencing

Abstract

Periodontitis has a polymicrobial etiology within the framework of a complex microbial ecosystem. With advances in sequencing technologies, comprehensive studies to elucidate bacterial community differences have recently become possible. We used 454 sequencing of 16S rRNA genes to compare subgingival bacterial communities from 29 periodontally healthy controls and 29 subjects with chronic periodontitis. Amplicons from both the V1-2 and V4 regions of the 16S gene were sequenced, yielding 1 393 579 sequences. They were identified by BLAST against a curated oral 16S database, and mapped to 16 phyla, 106 genera, and 596 species. 81% of sequences could be mapped to cultivated species. Differences between health-and periodontitis-associated bacterial communities were observed at all phylogenetic levels, and UniFrac and principal coordinates analysis showed distinct community profiles in health and disease. Community diversity was higher in disease, and 123 species were identified that were significantly more abundant in disease, and 53 in health. Spirochaetes, Synergistetes and Bacteroidetes were more abundant in disease, whereas the Proteobacteria were found at higher levels in healthy controls. Within the phylum Firmicutes, the class Bacilli was health-associated, whereas the Clostridia, Negativicutes and Erysipelotrichia were associated with disease. These results implicate a number of taxa that will be targets for future research. Some, such as Filifactor alocis and many Spirochetes were represented by a large fraction of sequences as compared with previously identified targets. Elucidation of these differences in community composition provides a basis for further understanding the pathogenesis of periodontitis.

Dysbiosis and Alterations in Predicted Functions of the Subgingival Microbiome in Chronic Periodontitis

“…2B). This separation according to clinical phenotype was also observed when data sets reported by the HMP (healthy subjects only), Abusleme et al (31), and Griffen et al (30) (both healthy and periodontitis groups) were included in the analysis (Fig. 2C to G).…”

“…Raw sequencing reads were obtained from three other studies: the HMP (42), Griffen et al (30), and Abusleme et al (31). The reads were downloaded from the HMP DACC website and the Sequence Read Archive (SRA) (SRA accession number SRP009299 for data reported by Griffen et al and SRA accession number SRP012422 for data reported by Abusleme et al).…”

“…Recently, culture-independent molecular methods targeting small-subunit rRNA sequences (16S rRNA) in combination with high-throughput DNA sequencing have emerged as a popular approach to profile microbial flora from diverse sites of the human body. By amplification and deep sequencing of 16S rRNA gene segments, this approach has provided an unprecedented view of human-associated microbial communities and has uncovered novel or previously unrecognized microbes associated with health and disease (16,(28)(29)(30)(31). The Human Microbiome Project (HMP) data set is currently the largest reference data set (143 individuals) for the subgingival microbiome (32).…”

“…However, because this data set was generated from individuals with minimal or no periodontal disease, the entire spectrum of the subgingival microbiota was not captured. Several groups have since used the 16S rRNA and metagenomic approaches to survey subgingival microbiota from subjects recruited in the community (30,33). These "community" data sets have demonstrated a difference between periodontitis-associated and health-associated microbiota in both the overall microbial community structure (30,31) and the predicted functional capabilities (34,35).…”

“…We used 16S rRNA deep sequencing to analyze subgingival microbiota from 25 subjects with chronic periodontal disease and 25 periodontally healthy controls and compared our data set with previously reported data sets from the HMP, Griffen et al (30), and Abusleme et al (31). In addition, we used PICRUSt (phylogenetic investigation of communities by reconstruction of unobserved states) (36), a bioinformatics tool designed to predict metagenome functional content from marker gene surveys and full genomes, to predict functional compositions of subgingival metagenomes in healthy and diseased individuals.…”