High-resolution differentiation of enteric bacteria in premature infant fecal microbiomes using a novel rRNA amplicon

Abstract

Identifying and tracking microbial strains as microbiomes evolve are major challenges in the field of microbiome research. We utilized a new sequencing kit that combines DNA extraction with PCR amplification of a large region of the rRNA operon and downstream bioinformatic data analysis. Longitudinal microbiome samples of coadmitted twins from two different neonatal intensive care units (NICUs) were analyzed using an;2,500-base amplicon that spans the 16S and 23S rRNA genes and mapped to a new, custom 16S-23S rRNA database. Amplicon sequence variants (ASVs) inferred using DADA2 provided sufficient resolution for the differentiation of rRNA variants from closely related but not previously sequenced Klebsiella, Escherichia coli, and Enterobacter strains, among the first bacteria colonizing the gut of these infants after admission to the NICU. Distinct ASV groups (finger-prints) were monitored between coadmitted twins over time, demonstrating the potential to track the source and spread of both commensals and pathogens. The high-resolution taxonomy obtained from long amplicon sequencing enables the tracking of strains temporally and spatially as microbiomes are established in infants in the hospital environment. IMPORTANCE Achieving strain-level resolution is a major obstacle for source tracking and temporal studies of microbiomes. In this study, we describe a novel deep-sequencing approach that provides species-and strain-level resolution of the neonatal microbiome. Using Klebsiella, E. coli, and Enterobacter as examples, we could monitor their temporal dynamics after antibiotic treatment and in pairs of twins. The strain-level resolution, combined with the greater sequencing depth and decreased cost per read of PacBio Sequel 2, enables this advantageous source-and strain-track-ing analysis method to be implemented widely across more complex microbiomes.