Pseudomonas aeruginosa is a significant nosocomial pathogen causing severe diseases, including sepsis. Carbapenem-resistant P. aeruginosa is a priority 1 pathogen according to the WHO, necessitating novel therapeutics. Antibiotic resistance is a global health crisis, driving renewed interest in bacteriophage therapy. Phage therapy offers advantages like specificity, replication at the infection site, low cost, and low toxicity. However, the in vivo dynamics of treating pan-resistant P. aeruginosa with phage are poorly understood. This study aimed to investigate the efficacy of phage therapy against pan-resistant P. aeruginosa in vivo, focusing on the interplay between bacterial adaptation to the lung environment and the development of phage resistance, and the potential for phage steering to overcome these challenges. Hospital-acquired pneumonia (HAP) is a leading cause of death from nosocomial infections, with P. aeruginosa being a frequent isolate, often exhibiting carbapenem resistance. The high mortality associated with P. aeruginosa bacteremia emphasizes the urgent need for effective new therapies. Previous clinical case studies and in vitro studies demonstrate the potential of phage therapy, including the concept of 'phage steering'—where phage resistance leads to re-sensitization to antibiotics. However, in vivo demonstration of phage steering to clear an infection remained lacking.

Existing literature highlights the effectiveness of phage therapy in treating various bacterial infections. Studies using murine models of chronic lung infection with P. aeruginosa demonstrated the efficacy of phage PEIL20 in clearing bacterial loads. However, complexities in the in vivo phage-bacteria dynamics remained unclear. Environmental factors such as limited oxygen availability, mucins, and polysaccharides, are known to influence phage resistance, but their effects hadn't been fully investigated within the context of bacterial adaptation and phage therapy. In vitro studies have indicated the possibility of 'phage steering,' where phage-resistant bacteria become re-sensitized to antibiotics; yet, this phenomenon lacked confirmation in vivo. Previous studies have used phage cocktails and single phage therapy, yielding variable success rates. The effectiveness of phage therapy is often dependent on the choice of phage and the specific bacterial strain.

This study employed an in vivo model of systemic P. aeruginosa infection using a pan-resistant strain carrying a megaplasmid with antimicrobial resistance genes. A four-phage cocktail (PELP20, PT6, PMN, 14/1) was tested for its efficacy. The susceptibility of 151 clinical isolates to the phages was determined using a direct spot test. An invasive respiratory model was established in mice using strain B9 (T2436), demonstrating systemic spread following intranasal infection. Phage treatment was initiated both early (simultaneously with infection) and late (after systemic spread) via intranasal and intravenous routes. Bacterial loads in lungs, liver, blood, kidneys, and spleen were assessed at various time points. Phage resistance was evaluated using efficiency of plating (EOP) on in vivo recovered isolates. The influence of the lung environment (oxygen availability, polyamines, mucins) on phage resistance was investigated by culturing P. aeruginosa under different conditions. Antibiotic susceptibility was assessed via disk diffusion assays and MIC determination for tobramycin and meropenem. Whole-genome sequencing was performed to identify genetic alterations associated with phage resistance and antibiotic re-sensitization. Membrane permeability assays were conducted to examine outer membrane integrity and cytoplasmic membrane polarization in phage-resistant isolates. Finally, phage steering was evaluated by pre-treating mice with phage and subsequently administering meropenem or tobramycin.

The phage cocktail exhibited strong therapeutic potential in the in vivo model, significantly reducing or clearing P. aeruginosa from various organs, particularly when administered early and intravenously. Surprisingly, even untreated mice displayed increased phage resistance in lung isolates, suggesting adaptation to the in vivo environment. This resistance was linked to reduced phage adsorption, potentially due to modifications in the bacterial outer membrane. Phage-treated isolates showed increased resistance to the corresponding phages but were re-sensitized to a broad range of antibiotics. This re-sensitization was associated with increased outer membrane permeability. Phage steering was demonstrated as a successful therapeutic strategy: pre-exposure to the phage cocktail re-sensitized the infection to antibiotics, leading to bacterial clearance. Genetic analysis revealed alterations in LPS biosynthesis genes in non-phage-treated lung isolates, potentially explaining phage resistance. In phage-treated isolates, increased outer membrane permeability, rather than genetic changes in AMR genes, accounted for the observed re-sensitization to antibiotics. The study also examined the role of the lung environment (oxygen, polyamines, mucins) in contributing to phage resistance independently of phage treatment.

This study demonstrates the potential of phage therapy, particularly phage steering, in treating pan-resistant P. aeruginosa infections. The findings highlight the importance of considering bacterial adaptation to the in vivo environment and its impact on phage efficacy. The re-sensitization of phage-resistant bacteria to antibiotics opens new therapeutic avenues. The differences in phage efficacy between different organs suggest the impact of phage pharmacokinetics and tissue-specific factors on treatment outcomes. The observed re-sensitization to antibiotics through altered outer membrane permeability points to a post-transcriptional mechanism. This study successfully translates the in vitro concept of phage steering into an effective in vivo strategy.

This research conclusively demonstrates the efficacy of phage therapy, especially the novel application of phage steering, in combating pan-resistant P. aeruginosa infections. The study successfully bridges the gap between in vitro findings and in vivo applications, paving the way for potential clinical translation. Future research should focus on optimizing phage cocktails, refining administration routes, and investigating the long-term effects of phage steering. Exploring the detailed mechanisms underlying the observed re-sensitization to antibiotics could lead to further enhancements in treatment strategies.

The study used a specific strain of P. aeruginosa and a specific phage cocktail. The generalizability of these findings to other strains and phage cocktails needs further investigation. The direct spot test might have underestimated the phage’s true efficacy. While a broad range of isolates were screened, the majority were keratitis isolates; testing a wider variety of infection types would strengthen the results. Further studies are needed to optimize the phage cocktail composition and delivery method for improved therapeutic efficacy.