Genome-wide co-occurrence patterns link mobile genetic elements, antimicrobial resistance and defense systems in Pseudomonas aeruginosa

Abstract Background Pseudomonas aeruginosa remains an urgent public health threat due to the many life-threatening diseases it causes, wide dissemination of multidrug resistant strains, and ability to survive and persist on hospital surfaces. Its accessory genome, composed of genes present in a subset of strains, may provide a clue to its success as an opportunistic pathogen. Here, we aim to characterize the pool of mobile genetic elements (MGE), anti-phage defenses and antimicrobial resistance (AMR) genes in P. aeruginosa. Results Using a dataset of 1,045 complete genomes, we identified 173 plasmids, 745 ICE/IME, 4,351 phage sequences, and 40,110 insertion sequence (IS) elements. Most pairs of genomes carrying specific combinations of defenses and MGEs exhibit low Jaccard similarity coefficient (< 20%) indicating that most unique MGE-defense combinations can be found in dissimilar genomes. Nonetheless, despite their considerable diversity and differential distribution among genomes, we found a highly connected network of co-occurring MGEs and AMR genetic determinants. IncP was connected to the largest and most diverse set of AMR determinants (n = 93) from nine antimicrobial classes (aminoglycoside, beta-lactam, phenicol, polymyxin, quinolone, sulfonamide, tetracycline, trimethoprim, multiple drugs) and heavy metals. Genetic elements conferring resistance to aminoglycoside, beta-lactam and quinolone were enriched in genomes carrying IS3. Genes conferring resistance to aminoglycoside, beta-lactam and trimethoprim were more frequently found in genomes carrying Autographiviridae sequences. Conclusions Our study highlights the need for public health strategies that disrupt MGE-mediated transmission of AMR genes, integrate genomic surveillance of MGEs and other accessory genes into infection control programs, and guide the design of combination therapies that consider bacterial defense systems. Our findings have implications to understanding the contributions of strain diversity in health and disease, the evolution of multidrug resistance, and the nature of the bacterial pan-genome.