Genomic Characterization of Intestinal Colonizing Pseudomonas juntendi Strains Harboring bla_VIM-2

Introduction

Pseudomonas juntendi, first described in 2019 from sputum samples of patients in Japan and Myanmar,¹ is an emerging member of the Pseudomonas putida group with increasing clinical relevance. Once considered an under-studied environmental species, it is now increasingly recognized as a potential nosocomial pathogen. Emerging clinical evidence highlights its role in opportunistic infections, particularly among immunocompromised individuals.² This Gram-negative bacterium is associated with multidrug-resistant (MDR) healthcare-associated infections, including ventilator-associated pneumonia and catheter-related bacteremia, with mortality rates exceeding 28% in ICU settings.³

The clinical significance of this species is compounded by its genomic plasticity, which facilitates the acquisition of horizontally transferred resistance determinants. Of particular concern is bla_VIM-2, a metallo-β-lactamase gene conferring carbapenem resistance, which has been detected in P. juntendi isolates from clinical and environmental sources,^4,5 However, the genetic context and transmission dynamics of this resistance gene remain poorly characterized.

A critical knowledge gap persists regarding the clonal spread and genomic stability of bla_VIM-2 harboring P. juntendi in human populations, particularly within the intestinal microbiota—a potential reservoir for horizontal gene transfer. This study investigates two MDR P. juntendi strains isolated from human fecal specimens, aiming to characterize their resistomes, plasmid profiles, and evolutionary links to aquatic reservoirs using hybrid whole-genome sequencing.

Materials and Methods

Bacterial Isolation and Phenotyping

P. juntendi strain L2353hy was isolated from a 56-year-old male with diarrhea in July 2020, whereas strain L2891hy was isolated from a 71-year-old male with diarrhea who was treated at the First Affiliated Hospital, College of Medicine, Zhejiang University, October 2020. Species identification was confirmed by ANI analysis using JSpeciesWS (http://jspecies.ribohost.com/jspeciesws/) against 58 global P. juntendi genomes (NCBI Pathogen Detection Database, Table S1). Antimicrobial susceptibility profiles were determined using CLSI M100-Ed34 (2024)-recommended broth microdilution (MIC) and disk diffusion for carbapenems (meropenem and imipenem), cephalosporins (ceftazidime and cefepime), aminoglycosides (amikacin), and fluoroquinolones (ciprofloxacin).

Genomic Characterization

Hybrid Nanopore Minion (Oxford Nanopore Technologies, Oxford, UK) and Illumina NovaSeq (San Diego, CA, USA, 150bp paired-end) sequencing were performed on the extracted genomic DNA (QIAamp DNA Mini Kit). The hybrid assembly used Unicycler v0.5.0⁶ with SPAdes-polished⁷ Illumina reads, achieving >100× coverage (Q30 ≥95%). Genome annotation incorporated Prokka⁸ v1.14.6, CheckM v1.2.0⁹ and RASTtk (https://www.bv-brc.org/app/Annotation) v3.47.11. Resistance determinants were identified using ResFinder v4.1, and plasmid types were determined using PlasmidFinder 2.1 (https://cge.food.dtu.dk/services/PlasmidFinder). CARD v3.2.4 (https://card.mcmaster.ca/), with ≥90% identity threshold. Virulence-associated genes were identified using the Virulence Factor Database, (VFDB) (https://www.vfdb.org/). Plasmid replicon typing combined PlasmidFinder 2.1 (https://cge.food.dtu.dk/services/PlasmidFinder-2.0/) and further verified using S1 nuclease pulsed-field gel electrophoresis (S1-PFGE, CHEF Mapper XA System).

Phylogenetic Analysis

Core genome SNP phylogeny was reconstructed using Parsnp v1.7.4 (Harvest Suite, https://github.com/marbl/parsnp). The pairwise SNP distances were calculated via SNP-dists v0.7.0 (https://github.com/tseemann/snp-dists). Phylogenetic trees were visualized using iTOL v6.7 with bootstrap support (1000 replicates). Comparative analysis included 58 global isolates (Table S1) from 10 countries.

Results

The genomes of strains L2353hy (5,820,718 bp) and L2891hy (5,699,712 bp) were assembled with a GC content of 62% and contained 5287 and 5116 coding DNA sequences (CDSs), respectively. Completeness was assessed using CheckM, which confirmed 100% completeness and no contamination by either strain. These high-quality assemblies provided a robust basis for further functional and comparative genomic studies. Average nucleotide identity analysis of P. juntendi strains L2353hy and L2891hy, and 58 global isolates revealed high nucleotide conservation (Figure 1A). A heat map was constructed using ANI values, with a color gradient representing identity percentages ranging from 75% (blue) to 100% (red). The established species boundary threshold (95% ANI) confirmed that both the clinical strains were P. juntendi. Notably, L2353hy and L2891hy exhibited 99.9% ANI values, which indicated clonal relatedness. Furthermore, both strains shared 99.2% ANI with P. juntendi yb_3(GCA_043103875.1; fish intestinal isolate; Wenzhou, China) (Figure 1B).

Figure 1 Continued.

Figure 1 Comparative genomic analysis and plasmid distribution. (A) The heatmap was constructed using ANI values, with a color gradient from 75% (blue) to 100% (red), showing genetic similarity among strains. (B) Prevalence of key carbapenemase genes (IMP-16, IMP-1, KPC-2, etc) across different countries. Strains harboring blaVIM-2 are highlighted in red. The tree scale indicates evolutionary relationships at 0.01 substitutions per site. (C) Genetic environment of blaVIM-2 created by Easyfig. Arrows indicate: antibiotic resistance (red), heavy metal resistance (Orange), membrane modification (purple), mobile genetic elements (green), regulatory proteins (blue), and metabolic or hypothetical proteins (gray). Blue shading indicates homologous regions.

Identification of virulence factors showed that both strains harbored chromosomally encoded virulence determinants, including lipopolysaccharide biosynthesis genes (waaF), transcriptional regulators (vfr and rpoNS), flagellar assembly and motility genes (flgCGHI, fleQ, flhA, fleN, cheY, pa1464, fliAGIMNPQ, motC), iron acquisition systems (pvdS and mbtH-like), alginate biosynthesis machinery (algU, alg8, algIA), twitching motility components (pilH), and catabolite-repression regulators (crc). These findings suggest that P. juntendi possesses a robust genetic arsenal for host colonization, biofilm formation, and immune evasion, potentially contributing to its persistence in clinical settings.

ResFinder analysis identified 13 resistance genes in L2353hy and L2891hy, including aminoglycoside resistance genes (aadA13, aph(3′)-VIb, aac(6′)-IIa), β-lactam resistance genes (bla_VIM-2 and bla_OXA-246), quinolone resistance gene qnrVC6, phenicol resistance gene cmlA1, rifampin resistance gene arr-3, sulfonamide resistance gene sul1, tetracycline resistance gene tet(A), and trimethoprim resistance gene dfrB4. The presence of bla_VIM-2 within the conserved integron structure highlighted the role of this species as a reservoir for carbapenem resistance.

PlasmidFinder analysis confirmed the absence of plasmid replicons in both L2353hy and L2891hy isolates. S1 nuclease pulsed-field gel electrophoresis (S1-PFGE) showed no plasmid bands for either strain (Figure S1). These results indicate that all identified resistance genes, including bla_VIM-2, are chromosomally encoded, highlighting the genomic integration of resistance determinants in these clinical isolates.

Gene environment analysis revealed a diverse array of sequences surrounding bla_VIM-2, with the conserved structure sequence eptA-czcD-Tn3-hin-xerC- bla_VIM-2-hin-hypothetical protein-folP-hypothetical protein--aphA-IS91 present in two strains. The predominant MGEs surrounding bla_VIM-2 were Tn3 and IS 91 (Figure 1C).

Discussion

Core genome phylogeny revealed that L2353hy and L2891hy clustered within a distinct clade with three closely related P. juntendi strains: P. juntendi yb_3, P. juntendi 12815 (GCA_014062135.1; human urine isolate; Brazil), P. juntendi BML-PP047 (GCA_021602565.1; human urine isolate; Japan). This close phylogenetic relationship suggests a potential zoonotic transmission pathway, with aquatic reservoirs serving as sources for clinically relevant P. juntendi strains. The clustering of human isolates from geographically distinct regions (China, Brazil, and Japan) further underscores the global dissemination of this lineage.

Although bla_VIM-2 is typically embedded within integrons that are not autonomously transferable, their association with mobile genetic elements (MGEs) facilitates horizontal dissemination¹⁰ The increasing global prevalence of bla_VIM-2-mediated carbapenem resistance poses a significant threat to public health, particularly in healthcare settings where MDR pathogens thrive.¹¹ This study identified P. juntendi as a potential reservoir for carbapenem-resistance genes. The close relationship between human and animal strains suggests bidirectional resistance gene flow at the human-animal interface. By analyzing the genetic environment surrounding bla_VIM in two strains (L2353hy and L2891hy), we identified a highly conserved gene cluster, including lipopolysaccharide biosynthesis genes (waaF), transcriptional regulators (vfr, rpoNS), flagellar assembly and motility genes (flgCGHI, fleQ, flhA, fleN, cheY, pa1464, fliAGIMNPQ, motC), iron acquisition systems (pvdS, mbtH-like), alginate biosynthesis machinery (algU, alg8, algIA), twitching motility components (pilH), and catabolite repression regulators (crc). These findings suggest that P. juntendi possesses a robust genetic arsenal for host colonization, biofilm formation, and immune evasion, potentially contributing to its persistence in clinical settings. This region not only contains multiple antibiotic resistance genes (bla_VIM-2, aphA, merA, merR and merT) but also includes several recombinases (Tn3 and IS91) and other accessory proteins. This structure suggests that this area may constitute a multidrug resistance island (MDRI) with the potential for rapid dissemination through horizontal gene transfer. The high-quality genomes generated here enhance genomic resources for P. juntendi and inform future surveillance of carbapenemase-carrying pathogens across ecological settings.

Conclusion

The detection of bla_VIM-2 harboring P. juntendi in human fecal microbiota adds to the growing evidence of its role as a reservoir for carbapenem resistance. The close genomic relationship between human and animal isolates highlights potential bidirectional transmission at the human-animal interface. These findings support the adoption of a One Health approach to surveillance for carbapenemase producing Pseudomonas species.