Genomic Potential vs. Pathogenic Reality: Functional Virulence Variation in Pseudomonas aeruginosa

Abstract

Despite the rapid expansion of bacterial genome sequencing, predicting pathogen virulence from genomic information remains a major challenge. To study this association, we integrate comparative genomics, functional phenotyping, host-pathogen interaction assays, and infection outcomes across a diverse collection of 28 Pseudomonas aeruginosa isolates, representing clinically relevant diversity in serotype, type IV pilus group, and flagellar type. While virulence-associated systems were broadly retained across the strain collection, isolates displayed substantial variation in phenotypic behavior and pathogenicity. Genotypic classifications based on pilus, flagellar, siderophore, or secretion system characteristics were poor predictors of virulence-associated phenotypes. Pathogenicity in a Galleria mellonella model varied markedly among strains and was most strongly associated with surface-associated motility, whereas other virulence traits and quorum-sensing genotypes showed limited predictive value. In a three-dimensional lung epithelial cell model, inflammatory responses were negatively correlated with activity of the type III secretion system effector ExoS, suggesting a role in modulation of host immune signaling. Together, these findings demonstrate that virulence emerges from functional deployment of conserved virulence repertoires, rather than their mere presence. By linking genomic diversity to phenotypic expression, pathogenicity, and host responses within a single strain collection, this study highlights fundamental limitations of gene-content-based virulence prediction and underscores the importance of integrating genomic and functional data to better understand P. aeruginosa virulence.