Page 102 - Read Online

Page 2 of 12 Jones et al. Microbiome Res Rep 2024;3:24 https://dx.doi.org/10.20517/mrr.2023.78

[1-4]
recent times . To date, bacterial communities from numerous body sites have been screened, both in silico
and in vitro, for antimicrobial compounds [5-11] . Recent advances in metaculturomics and metagenomic
sequencing have led to the discovery and characterisation of the urobiome [12,13] , which represents a relatively
understudied environment in terms of the diversity and novelty of bacteriocins encoded by this microbial
community.

Bacteriocins are classified as ribosomally synthesised antimicrobial peptides, which are produced by bacteria
as a defence mechanism against other bacteria present in the same environment [5,14,15] . Bacteriocins can
display both narrow- and broad-spectrum bactericidal activity but are usually most effective against bacteria
that are closely related to the producer strain [6,8,16-18] . While some bacteriocins are produced by Gram-
[19]
negative bacteria , the majority of bacteriocins characterised to date are produced by Gram-positive, lactic
acid bacteria [20-22] . Bacteriocins have been divided into three classes: class I, also known as lantibiotics, are
characterised based on the presence of the amino acid lanthionine or methyllanthionine as a result of post-
translational modifications. The primary mode of action of class I bacteriocins is targeting the cell
membrane [14,20] . Class II bacteriocins are smaller, thermostable peptides that can be further categorised into
[23]
five subclasses . They are classed as broad range antimicrobials and act by forming pores in the cell
membrane. Class III bacteriocins are larger, heat-sensitive peptides that cause bacterial cell lysis [5,14,20] . In
previous years, food preservation and other applications in the food industry were the primary focus of
[15]
bacteriocin research . More recently, this focus has shifted to antimicrobial resistance, and strategies to
[15]
improve the treatment and control of antibiotic-resistant infections , mainly centred on in vivo animal
[24]
studies . Bacteriocins have numerous desirable traits as antimicrobials, which make them particularly
attractive alternatives to antibiotics, including low toxicity, high potency, and, most importantly, the ability
to be effectively bioengineered [18,23] . Furthermore, a narrower activity spectrum than conventional antibiotics
significantly reduces undesirable collateral damage to the commensal microbiota [2,11,14] .

In silico screening of bacterial genomes for bacteriocin production has significantly reduced both the time
and cost of culture-based approaches for bacteriocin discovery , with gene mining tools such as BAGEL4
[25]
and antiSMASH7 [26,27] enabling the rapid identification of bacteriocin gene clusters. BAGEL4 scans the
bacterial genome for putative bacteriocin open reading frames (ORFs). It searches for the structural
bacteriocin gene, but also takes advantage of the common structure of bacteriocin operons and scans the
surrounding ORFs for possible accessory genes that encode immunity, transport, regulation, and
modification proteins [25,28,29] . AntiSMASH7, on the other hand, uses set “rules” that identify core biosynthetic
functions present within a genomic region to create a biosynthetic gene cluster (BGC). AntiSMASH7
combines different profile hidden Markov model “rules” to identify 81 different BGC types .
[27]
The current study is, to the best of our knowledge, the first to screen the urobiome for genes encoding
bacteriocin production. Herein, we screened 181 bacterial isolates previously isolated from the bladder ,
[30]
with the primary objective to identify novel bacteriocin clusters in the urobiome [Figure 1]. Initially, 263
putative bacteriocin gene clusters were identified, highlighting the potential of the urobiome to host a
diversity of bacteriocin producers.

METHODS
Data collection
The 181 fully sequenced genomes examined in this study were from urinary bacterial isolates, collected via
catheter, previously isolated, sequenced, and assembled by Miller-Ensminger et al. (2018) . Accession
[30]
numbers were obtained [Supplementary Data], and each individual genome was downloaded from the
European Nucleotide Archive (ENA) (https://www.ebi.ac.uk/ena/browser/home) in FASTA format.