Genetic and Virulence Factors Behind the Success of High-Risk Pseudomonas aeruginosa Clones: Insights from Comparative Genomics and an Experimental Infection Model

Romário Oliveira De Sales¹Laura Leaden¹Dayble Martins¹Paula Koga²Alexandra Toniolo²Fernando Gatti De Menezes²Marcelo Alves Mori³Marines D V Martino²Patricia Severino^1*

• ¹Albert Einstein Research and Education Institute, Hospital Israelita Albert Einstein, SAO PAULO, Brazil
• ²Hospital Israelita Albert Einstein, São Paulo, Brazil
• ³Universidade Estadual de Campinas, Campinas, Brazil

Introduction: Pseudomonas aeruginosa causes severe healthcare-associated infections. High-risk clones are defined by global dissemination and multidrug resistance, yet virulence is heterogeneous. We sought to map accessory-genome determinants associated with high-risk clones by integrating whole-genome sequencing (WGS) with a Caenorhabditis elegans infection model. Methods: We analyzed 84 clinical isolates plus publicly available genomes using WGS, phylogenomics, and resistome/virulome profiling. Virulence was measured by C. elegans slow-killing (SK). A GWAS of accessory-genome sub elements (AGEs) identified loci with high- (HVA) or low-virulence association (LVA). Coding sequences were annotated with Prokka and InterPro. Results: Although HR and sporadic clones carried a similar total number of antimicrobial-resistance genes, 15/67 (22.38%) genes/variants were enriched in high-risk clones, producing class-level enrichment (p < 0.002) for aminoglycosides, phenicols, trimethoprim, sulfonamides, and tetracyclines, but not β-lactams or fosfomycin. Many determinants are recognized mobile-element cargo such as integron cassettes or plasmid/ICE-borne genes (e.g., aadA, dfrB, blaVIM-2, crpP, cmlA, floR), indicating a mobility-linked resistome in high-risk clones. GWAS identified 113 AGEs linked to SK virulence (42 HVA, 71 LVA). HVA regions were enriched for pyoverdine (fpvA, pvdE, pvdD) and LPS O-antigen (wbpA/B/D) loci, whereas LVA regions were enriched for ICE/conjugation/integrase motifs. cdsA and clpP were newly associated with P. aeruginosa virulence. Phenotypically, high-risk clones were more often strong biofilm producers and none were non-producers. High-risk clones were not consistently more virulent in SK, suggesting success reflects persistence traits (mobile DNA and biofilm under antibiotic pressure). Conclusions: Accessory-genome GWAS revealed two risk dimensions: acute-virulence programs (HVA) versus mobility functions (LVA) favoring persistence and spread. Because SK measures acute virulence, readouts did not align with high-risk designations. Genomic reports should combine HR assignment with accessory-genome and effector profiling to support earlier containment and mechanism-aware, biofilm-focused care.