Sophie Payot

Streptococcus suis is a leading cause of infection in pigs, causing extensive economic losses. In addition, it can also infect wild fauna, and can be responsible for severe infections in humans. Increasing antimicrobial resistance (AMR) has been described in S. suis worldwide and most of the AMR genes are carried by mobile genetic elements (MGEs). This contributes to their dissemination by horizontal gene transfer. A collection of 102 strains isolated from humans, pigs and wild boars in France was subjected to whole genome sequencing in order to: (i) study their genetic diversity, (ii) evaluate their content in virulence-associated genes, (iii) decipher the mechanisms responsible for their AMR and their association with MGEs, and (iv) study their ability to acquire extracellular DNA by natural transformation. Analysis by hierarchical clustering on principal components identified a few virulence-associated factors that distinguish invasive CC1 strains from the other strains. A plethora of AMR genes ( n =217) was found in the genomes. Apart from the frequently reported erm (B) and tet (O) genes, more recently described AMR genes were identified [ vga (F)/ sprA , vat (D)]. Modifications in PBPs/MraY and GyrA/ParC were detected in the penicillin- and fluoroquinolone-resistant isolates respectively. New AMR gene–MGE associations were detected. The majority of the strains have the full set of genes required for competence, i.e for the acquisition of extracellular DNA (that could carry AMR genes) by natural transformation. Hence the risk of dissemination of these AMR genes should not be neglected.

Mobile Genetic Elements (MGEs) are integrated in bacterial genomes and key elements that drive prokaryote genome evolution. Among them are Integrative and Conjugative Elements (ICEs) and Integrative Mobilizable Elements (IMEs) which are important for bacterial fitness since they frequently carry genes participating in important bacterial adaptation phenotypes such as antibiotic resistance, virulence or specialized metabolic pathways. Although ICEs and IMEs are widespread, they are as yet almost never annotated in public bacterial genomes. To address the need of dedicated strategies for the annotation of these elements, we developed ICEscreen, a tool that introduces two new features to detect ICEs and IMEs in Firmicute genomes. First, ICEscreen uses an efficient strategy to detect Signature Proteins of ICEs and IMEs based on a database dedicated to Firmicutes and composed of manually curated proteins and Hidden Markov Models (HMM) profiles. Second, ICEscreen includes a new original algorithm that detects composite structures of ICEs and IMEs that are frequent in genomes of Firmicutes but are currently not resolved by any other tool. We benchmarked ICEscreen on experimentally supported elements and on a public dataset of 246 manually annotated elements including the genomes of 40 Firmicutes and demonstrate its efficiency to detect ICEs and IMEs.