The human urine virome in association with urinary tract infections

Abstract

While once believed to represent a sterile environment, the human urinary tract harbors a unique cellular microbiota. We sought to determine whether the human urinary tract also is home to viral communities whose membership might reflect urinary tract health status. We recruited and sampled urine from 20 subjects, 10 subjects with urinary tract infections (UTIs) and 10 without UTIs, and found viral communities in the urine of each subject group. Most of the identifiable viruses were bacteriophage, but eukaryotic viruses also were identified in all subjects. We found reads from human papillomaviruses (HPVs) in 95% of the subjects studied, but none were found to be high-risk genotypes that are associated with cervical and rectal cancers. We verified the presence of some HPV genotypes by quantitative PCR. Some of the HPV genotypes identified were homologous to relatively novel and uncharacterized viruses that previously have been detected on skin in association with cancerous lesions, while others may be associated with anal and genital warts. On a community level, there was no association between the membership or diversity of viral communities based on urinary tract health status. While more data are still needed, detection of HPVs as members of the human urinary virome using viral metagenomics represents a non-invasive technique that could augment current screening techniques to detect low-risk HPVs in the genitourinary tracts of humans.

Keywords: HPV; human microbiome; papillomavirus; urinary tract infections; virobiota; virome.

Figure 1

Read mappings of select viromes to Human papillomaviruses (HPVs). (A) Represents URN2 mapping to HPV Type 96, (B) represents URN6 mapping to HPV Type 49, and (C) represents URP12 mapping to HPV Type 178. The relative coverage of each contig is represented, along with annotated open reading frames (ORFs) above each box. The portions of the contigs identified are represented by the colored boxes for each subject. The length of each contig is denoted at the top of each panel. The relative proportions of reads mapping to each virus are represented in green.

Figure 2

Bar graphs of the percentage of subjects with detectable HPV Types 49 and 178 by quantitative PCR. White bars indicate those subjects with negative urine cultures, and black bars represent subjects with urinary tract infections.

Figure 3

Bar graphs of the mean percentages of contigs (± standard error) with viral homologs in the NR database from all of the subjects. White bars indicate those subjects with negative urine cultures, and black bars represent subjects with urinary tract infections.

Figure 4

Rarefaction analysis based on the Homologous Virus Diversity Index using the Shannon Index (A) and the chao1 Index (B) for urine viral communities. The index values are shown on the y-axis and the number or reads sampled are shown on the x-axis. White boxes indicate those subjects with negative urine cultures, and black triangles represent subjects with urinary tract infections. Dashed lines indicate those subjects with negative urine cultures, and solid lines represent subjects with urinary tract infections.

Figure 5

Plots of alpha diversity for urine viromes (A) and bacteria biota using 16S rRNA (B). Viral diversity was determined using the Homologous Virus Diversity Index based on the Shannon Index, and bacterial diversity determined using the Shannon Index. White diamonds indicate those subjects with negative urine cultures, and black diamonds represent subjects with urinary tract infections.

Figure 6

Principal coordinates analysis of beta diversity of the urine viromes (A), and bacterial biota using 16S rRNA (B). White circles indicate those subjects with negative urine cultures, and black circles represent subjects with urinary tract infections.

Figure 7

Pie charts of phage BLASTX hits (A,B) and bacterial taxonomy using 16S rRNA (C,D). (A) and (C) represent subjects with negative urine cultures, and (B) and (D) represent subjects with urinary tract infections. The “^*” represents significant differences (p = 0.05) between the proportions of different phyla represented between UTI+ and UTI− subjects.