Defining the Infant Male Urobiome and Moving Towards Mechanisms in Urobiome Research

Abstract

The urinary bladder harbors a community of microbes termed the urobiome, which remains understudied. In this study, we present the urobiome of healthy infant males from samples collected by transurethral catheterization. Using a combination of extended culture and amplicon sequencing, we identify several common bacterial genera that can be further investigated for their effects on urinary health across the lifespan. Many genera were shared between all samples suggesting a consistent urobiome composition among this cohort. We note that, for this cohort, early life exposures including mode of birth (vaginal vs. Caesarean section), or prior antibiotic exposure did not influence urobiome composition. In addition, we report the isolation of culturable bacteria from the bladders of these infant males, including Actinotignum schaalii, a bacterial species that has been associated with urinary tract infection in older male adults. Herein, we isolate and sequence 9 distinct strains of A. schaalii enhancing the genomic knowledge surrounding this species and opening avenues for delineating the microbiology of this urobiome constituent. Furthermore, we present a framework for using the combination of culture-dependent and sequencing methodologies for uncovering mechanisms in the urobiome.

Keywords: Urobiome; pangenome; pediatric urology; urinary microbiome; urinary tract infection.

Figure 1. Study Schematic and Analysis Workflow.

A) Illustration of study design. Fifty male infants were sterilely catheterized in the operating theatre prior to undergoing circumcision. Urine was immediately plated for extended quantitative urine culture (EQUC). DNA was extracted from urine samples, amplified with V4 16S rRNA primers, and sequenced using Illumina paired-end chemistry. The combination of urine culture and sequencing results was used to describe the urobiome composition. B) Illustration of analysis workflow and evaluation of potential contaminant sources. Sampling controls were collected contemporaneously with urine samples in the operating theatre. Extraction blanks and a mock microbial community dilution series were used to benchmark DNA extraction. No template blanks were subjected to 16S rRNA PCR amplification to benchmark PCR amplification. All controls mentioned were subjected to Illumina paired-end sequencing. The Decontam package in R was used to filter potential contaminant sequences.

Figure 2. 16S rRNA Amplicon Sequencing Reveals a Consistent Urobiome Composition.

A) Beta diversity between infant urine samples and sampling controls. Beta diversity was calculated by the phyloseq “ordinate” function using Bray-Curtis distances. Urine samples were significantly different than sampling controls by permutational multivariate analysis of variance (PERMANOVA, p=0.001). PERMANOVA was calculated using the vegan function “adonis2”. B) Phyla-level taxonomic profiles of urine samples from 50 infants. Urine samples are depicted along the vertical axis and taxonomic relative abundance on the x-axis. Plot created with the microViz function “comp_barplot”. C) Alpha diversity metrics (Shannon index and Chao1) between urine from infants born by vaginal delivery vs. Caesarean section (left); and between urine from infants previously exposed to antibiotics vs. antibiotic naïve (right). Alpha diversity was calculated within the phyloseq package using the “plot_richness” function. Alpha diversity was not significant different between groups by Wilcoxin rank sum test (p>0.05). D) Relative abundance of Lactobacillaceae in urine samples between infants born by vaginal delivery vs. Caesarean section. The Lactobacillaceae family was agglomerated with the phyloseq command “tax_glom”. There was no significant difference in Lactobacillaceae abundance between groups by Wilcoxin rank sum test (p>0.05).

Figure 3. Concordance between EQUC and Amplicon Sequencing Results.

Co-occurrence detection patterns of taxonomic families between EQUC and amplicon sequencing methodologies. Taxonomic families are arranged vertically and patient samples horizontally. The rectangles indicate the detection of the family by EQUC (maroon), 16S rRNA amplicon sequencing (light blue), both methodologies (dark blue), or neither methodology (gray).

Figure 4. Genomic Characterization of Actinotignum schaalii Isolates.

A) Nine A. schaalii genomes isolated by EQUC from distinct subjects were subjected to whole genome sequencing. A. schaalii pangenome of the nine isolates visualized using anvi’o. Core genes were present in 100% of isolates (9/9) while the accessory genome consists of gene present in <9 of the genomes. Clustering of the genomes is based on average nucleotide identity (ANI), shown in the upper right matrix. B) Relative abundance of COG categories represented in the core and accessory genomes. C) Presence-absence matrix of fitness factors and antimicrobial resistance genes. ABRicate was used to screen contigs using the MegaRes, ResFinder, and Virulence Factor databases.