Establishing a Pseudomonas aeruginosa-Caenorhabditis elegans embryogenesis infection model for assessment of the Pseudomonas aeruginosa virulence and identification of novel virulence factors

Abstract

Pseudomonas aeruginosa is a prevalent Gram-negative opportunistic pathogen that poses a significant threat to human health due to its production of a wide array of virulence factors and exceptional capacity for developing antimicrobial resistance. Despite several existing infection models, our understanding of the factors and mechanisms underlying the virulence and pathogenesis of P. aeruginosa remains incomplete.

In this study, a novel and robust Pseudomonas aeruginosa - Caenorhabditis elegans embryogenesis infection model is established and key procedures of the assay, including C. elegans embryo preparation, bacterial culture medium, and conditions for virulence factors production were optimized. The robustness of this model was validated through assessing the virulence of both P. aeruginosa model strains and clinical isolates. It was revealed that canonical virulence factors, such as pyocyanin and proteolytic enzymes, are not responsible for the inhibition of C. elegans embryogenesis, suggesting novel determinants at play. Since the infection assay hinted that heat-stable and diffusible small molecules might be responsible for inhibiting the C. elegans embryogenesis, I focused on exploring the secondary metabolites of P. aeruginosa by constructing and screening a CRISPRi library targeting to secondary metabolites biosynthesis in PAO1.

Bioinformatic search combined with 2 common secondary metabolite database, clusters of orthologous groups, and pathway annotations identified 18 secondary metabolites BGCs and 9 potential secondary metabolites biosynthetic enzymes outside BGCs in the PAO1 genome. Leveraging our previously devised integrative type I-F CRISPRi system, an arrayed CRISPRi library containing 71 constructs was successfully constructed. These constructs specifically targeted 34 transcription units, with 2 or 3 designed crRNAs for each unit. After verifying target gene repression by RT-qPCR and re-designing of crRNA for those constructs displaying moderate to low repression effect, a well-defined, arrayed CRISPRi library containing 34 constructs was obtained. This library effectively repressed all target transcription units by more than 60%.

The arrayed CRISPRi library was subjected to P. aeruginosa - C. elegans embryogenesis infection assay. Among the 34 CRISPRi strains tested, inhibition of C. elegans embryogenesis was attenuated in 21 CRISPRi strains compared to the parent strain PAO1. Subsequent investigations, involving both CRISPRi strains and gene deletion mutants, verified that disrupting the pseudopaline and pyrroloquinoline quinone BGCs attenuated the C. elegans embryogenesis-inhibiting capacity of PAO1, suggesting that the corresponding secondary metabolites contribute to the pathogenesis of PAO1 to C. elegans embryos.

In conclusion, this study has revealed the vast potential of the embryogenesis infection model as a robust tool for assessing the virulence of P. aeruginosa and identifying new virulence factors. Furthermore, the CRISPRi library I developed has proven to be valuable in elucidating the novel functions of secondary metabolites. Together, these tools have markedly advanced our understanding of P. aeruginosa, offering significant insights into its pathogenesis.