Diversity of bacteria at healthy human conjunctiva

Abstract

Purpose: Ocular surface (OS) microbiota contributes to infectious and autoimmune diseases of the eye. Comprehensive analysis of microbial diversity at the OS has been impossible because of the limitations of conventional cultivation techniques. This pilot study aimed to explore true diversity of human OS microbiota using DNA sequencing-based detection and identification of bacteria.

Methods: Composition of the bacterial community was characterized using deep sequencing of the 16S rRNA gene amplicon libraries generated from total conjunctival swab DNA. The DNA sequences were classified and the diversity parameters measured using bioinformatics software ESPRIT and MOTHUR and tools available through the Ribosomal Database Project-II (RDP-II).

Results: Deep sequencing of conjunctival rDNA from four subjects yielded a total of 115,003 quality DNA reads, corresponding to 221 species-level phylotypes per subject. The combined bacterial community classified into 5 phyla and 59 distinct genera. However, 31% of all DNA reads belonged to unclassified or novel bacteria. The intersubject variability of individual OS microbiomes was very significant. Regardless, 12 genera-Pseudomonas, Propionibacterium, Bradyrhizobium, Corynebacterium, Acinetobacter, Brevundimonas, Staphylococci, Aquabacterium, Sphingomonas, Streptococcus, Streptophyta, and Methylobacterium-were ubiquitous among the analyzed cohort and represented the putative "core" of conjunctival microbiota. The other 47 genera accounted for <4% of the classified portion of this microbiome. Unexpectedly, healthy conjunctiva contained many genera that are commonly identified as ocular surface pathogens.

Conclusions: The first DNA sequencing-based survey of bacterial population at the conjunctiva have revealed an unexpectedly diverse microbial community. All analyzed samples contained ubiquitous (core) genera that included commensal, environmental, and opportunistic pathogenic bacteria.

Figure 1.

Relative abundance of bacterial taxa in the conjunctiva. (A) Phylum-level representation of the bacteria at the OS of the four subjects calculated according to relative abundance of classified 16S rRNA gene reads. The percentage of reads that failed to classify to known bacterial phyla is indicated as Unclassified, shown in orange. The circular diagram presents average values calculated for all analyzed subjects. Color-coding legend on the right shows taxonomic identities of the classified bacteria. (B) Genus-level representation of the bacteria at the OS. Unclassified reads (31% of the total 115,003 sequences) are shown in dark blue. (C) Relative abundance of known (16S-classified) and novel (unclassified) bacterial phylotypes at the conjunctiva of the individual subjects. All percentages were calculated relative to the total number of qualified DNA reads for each individual.

Figure 2.

Individual variability of most abundant bacteria at the conjunctiva. Prevalence of the seven most abundant genera in the OS of the four analyzed subjects. Values represent percentages of DNA reads corresponding to the indicated genera relative to all classified reads generated from an individual sample (coded by colors).

Figure 3.

Putative core microbiome of the ocular surface. Genus-level representation of the 12 most prevalent bacteria at the OS of the four subjects calculated according to relative abundance of classified DNA reads. Percentage of DNA reads that fail to classify beyond bacteria is indicated as “other.” Color-coding legend on the right side of the panel designates each genus.