Deciphering the role of insertion sequences in the evolution of bacterial epidemic pathogens with panISa software.

Xavier Bertrand,Didier Hocquet,Charlotte Couchoud,Benoit Valot

doi:10.1099/mgen.0.000356

Abstract

Next-generation sequencing (NGS) is now widely used in microbiology to explore genome evolution and the structure of pathogen outbreaks. Bioinformatics pipelines readily detect single-nucleotide polymorphisms or short indels. However, bacterial genomes also evolve through the action of small transposable elements called insertion sequences (ISs), which are difficult to detect due to their short length and multiple repetitions throughout the genome. We designed panISa software for the ab initio detection of IS insertions in the genomes of prokaryotes. PanISa has been released as open source software (GPL3) available from https://github.com/bvalot/panISa. In this study, we assessed the utility of this software for evolutionary studies, by reanalysing five published datasets for outbreaks of human major pathogens in which ISs had not been specifically investigated. We reanalysed the raw data from each study, by aligning the reads against reference genomes and running panISa on the alignments. Each hit was automatically curated and IS-related events were validated on the basis of nucleotide sequence similarity, by comparison with the ISFinder database. In Acinetobacter baumannii , the panISa pipeline identified ISAba1 or ISAba125 upstream from the ampC gene, which encodes a cephalosporinase in all third-generation cephalosporin-resistant isolates. In the genomes of Vibrio cholerae isolates, we found that early Haitian isolates had the same ISs as Nepalese isolates, confirming the inferred history of the contamination of this island. In Enterococcus faecalis , panISa identified regions of high plasticity, including a pathogenicity island enriched in IS-related events. The overall distribution of ISs deduced with panISa was consistent with SNP-based phylogenic trees, for all species considered. The role of ISs in pathogen evolution has probably been underestimated due to difficulties detecting these transposable elements. We show here that panISa is a useful addition to the bioinformatics toolbox for analyses of the evolution of bacterial genomes. PanISa will facilitate explorations of the functional impact of ISs and improve our understanding of prokaryote evolution.

Highlights

Whole-g enome sequencing (WGS) is becoming the gold- standard technique for investigating the evolution of bacterial pathogen genomes during their spread
Holt et al [13] reported the detection of ISAba1 and ISAba125 upstream from the cephalosporinase-encoding ampC gene by a PCR approach for the identification of genetic events leading to antibiotic resistance [13]
insertion sequences (ISs) can contribute to antibiotic resistance and virulence, depending on the nature of the genes that are disrupted or imported

Summary

Introduction

Whole-g enome sequencing (WGS) is becoming the gold- standard technique for investigating the evolution of bacterial pathogen genomes during their spread. Application of the appropriate pipelines to sequencing data results in the detection of single-nucleotide polymorphisms (SNPs) or small insertion/deletion (indels) after the alignment of reads with a reference genome sequence. Bacterial genomes evolve through the insertion of insertion sequences (ISs), which are widespread and occur in all domains of life [1]. ISs are mobile autonomous elements formed by (i) one or two transposase-encoding genes, (ii) two terminal inverted repeats (IRs), and (iii) two direct repeated sequences (DRs) [2]. ISs are sorted into families using the amino acid similarity of their transposase [3]. In 2019, the ISFinder database reported more than 4000 ISs belonging to 29 families [2, 3]

Methods

Results

Conclusion