Species-level core oral bacteriome identified by 16S rRNA pyrosequencing in a healthy young Arab population

Abstract

Background: Reports on the composition of oral bacteriome in Arabs are lacking. In addition, the majority of previous studies on other ethnic groups have been limited by low-resolution taxonomic assignment of next-generation sequencing reads. Furthermore, there has been a conflict about the existence of a 'core' bacteriome.

Objective: The objective of this study was to characterize the healthy core oral bacteriome in a young Arab population at the species level.

Methods: Oral rinse DNA samples obtained from 12 stringently selected healthy young subjects of Arab origin were pyrosequenced (454's FLX chemistry) for the bacterial 16S V1-V3 hypervariable region at an average depth of 11,500 reads. High-quality, non-chimeric reads ≥380 bp were classified to the species level using the recently described, prioritized, multistage assignment algorithm. A core bacteriome was defined as taxa present in at least 11 samples. The Chao2, abundance-based coverage estimator (ACE), and Shannon indices were computed to assess species richness and diversity.

Results: Overall, 557 species-level taxa (211±42 per subject) were identified, representing 122 genera and 13 phyla. The core bacteriome comprised 55 species-level taxa belonging to 30 genera and 7 phyla, namely Firmicutes, Proteobacteria, Actinobacteria, Bacteroidetes, Fusobacteria, Saccharibacteria, and SR1. The core species constituted between 67 and 87% of the individual bacteriomes. However, the abundances differed by up to three orders of magnitude among the study subjects. On average, Streptococcus mitis, Rothia mucilaginosa, Haemophilus parainfluenzae, Neisseria flavescence/subflava group, Prevotella melaninogenica, and Veillonella parvula group were the most abundant. Streptococcus sp. C300, a taxon never reported in the oral cavity, was identified as a core species. Species richness was estimated at 586 (Chao2) and 614 (ACE) species, whereas diversity (Shannon index) averaged at 3.99.

Conclusions: A species-level core oral bacteriome representing the majority of reads was identified, which can serve as a reference for comparison with oral bacteriomes of other populations as well as those associated with disease.

Keywords: core; high-throughput nucleotide sequencing; human; micgrobiome; mouth; pyrosequencing.

Fig. 1

Distribution of phyla and genera across samples. (a) Thirteen phyla were detected, of which 7 were present in at least 11 of the 12 subjects (core phyla). (b) Relative abundance of 30 core genera out of 122 ones detected.

Fig. 2

Distribution of phyla by subject and gender. Five core phyla (Firmicutes, Proteobacteria, Actinobacteria, Bacteroidetes and Fusobacteria) represented at least 96% of the reads. The females harbored higher proportions of Actinobacteria, TM7, Spirochaetes, and Synergistetes.

Fig. 3

Distribution of core genera by subject and gender. The thirty core genera constituted ≥89% of the microbiome in each of the study subjects, 94.7 and 97.7% in females and males, respectively. Ten of these (Streptococcus-Granulicatella in the figure key) accounted for at least 74% of the sequences.

Fig. 4

Heatmap of core species. The 55 core species (detected in at least 11 subjects) are ordered by their relative abundance across the study subjects. The map shows the relative abundance, normalized and log2-transformed, of the species within each subject. The subjects are clustered accordingly. Veillonella parvula group: V. parvula, V. dispar, V. atypica, and V. rogosae. Neisseria oralis group: N. oralis and Neisseria oral taxa 014 and 016.

Fig. 5

Species prevalence and relative abundance. The species rank is defined by its average abundance across all samples, with the most abundant species ranked first (left). (a) Species prevalence, the proportion of samples containing that species, plotted against species rank. (b) The relative abundance plotted for each sample in which the species is present – that is, for each species rank, a mark represents a sample positive for that species.

Fig. 6

Species richness and rarefaction curves. (a) Observed and estimated species richness within each subject, in the males and females and across subjects. ACE: abundance-based coverage estimator. The rarefaction curves were generated by plotting the number of species as a function of (b) number of reads from each subject, (c) number of total reads, and (d) number of sampled subjects.