Infectious diseases caused by antimicrobial-resistant bacteria pose significant threats to public health. According to O’Neill (2015), without a global program addressing antimicrobial resistance (AMR), an estimated 10 million people annually could be died to bacterial infections resistant to antibiotics by 2050. Among various types of Pseudomonas that commonly inhabit the environment, Pseudomonas aeruginosa is the species most frequently associated with infections in humans. This pathogen has the potential to cause various infections including pneumonia (Rossi et al., 2021; Qin et al., 2022). Along with other pathogens like Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, and Enterobacter spp., P. aeruginosa is classified as an ESKAPE pathogen, which are major contributors to multidrug resistance found in clinical human patients frequently (Mulani et al., 2019). In this context, phage therapy has received significant attention as a promising antimicrobial strategy because phages possess a bactericidal mechanism entirely distinct from antibiotics, enabling them to effectively combat antimicrobial-resistant bacteria (Kortright et al., 2019; Strathdee et al., 2023).

Despite high expectations for phage therapy, previous clinical trials on various phage therapies have reported the emergence of phage-resistant variants (Fujiki et al., 2023; Oromí-Bosch et al., 2023). This represents a significant concern for the successful development of phage-based therapies, similar to the case of resistance seen with classical antibiotics. In the case of P. aeruginosa, it has been reported that the polysaccharide O-antigen plays a role in LPS-targeting phages infection. Mutations in the polymerase encoded by wzy, responsible for lipopolysaccharide repeat unit polymerization, have been linked to the ability to evade LPS-targeting phages infections. These mutations have been observed in P. aeruginosa variants resistant to Pbunavirus strains KPP22 (Uchiyama et al., 2016), E215, E217, Litunavirus PYO2 and DEV (Forti et al., 2023) and Pakpunavirus PaP1 (Li et al., 2018). Moreover, galU, for synthesizing the complete core oligosaccharide, plays a crucial role in Pseudomonas phage infections. Removing galU from the Pseudomonas chromosome leads to a deficiency in O-antigen polysaccharide for phage adsorption (Le et al., 2014; Nakamura et al., 2021). Notably, deleting galU during phage infection coupled with a substantial chromosomal deletion by Mut associated with the DNA mismatch repair system, results in phage resistance due to the absence of a phage receptor (Shen et al., 2018). Since hmgA, responsible for converting red-pigment homogentisic acid to 4-maleylacetoacetate, is located near galU on the chromosome, P. aeruginosa phage-resistant variants exhibit a brown color phenotype if the extensive chromosomal deletion includes both galU and hmgA. These bacterial cells are specifically termed as brown mutants (brmts) (Sanz-García et al., 2018; Shen et al., 2018).

Fluoroquinolones, such as ciprofloxacin (CPFX) and levofloxacin (LVFX), are among the most commonly prescribed and effective antimicrobials for treating infections caused by P. aeruginosa. Currently, they are the only oral treatment options available for Pseudomonas infections (Zakhour et al., 2022). However, P. aeruginosa isolated from patients in clinics often exhibits resistance to fluoroquinolones (Rehman et al., 2019; Ruiz, 2019), suggesting that antibiotics commonly used in clinics may no longer be appropriate for this purpose. The main molecular mechanisms through which P. aeruginosa can decrease fluoroquinolone sensitivity include mutations in the quinolone resistance determining region (QRDR) in both gyrA and parC, known to play an integral role in quinolone resistance in P. aeruginosa (Rehman et al., 2019). Additionally, it is well-known that P. aeruginosa develops resistance through several multi-drug efflux (Mex) systems, such as MexAB-OprM, MexEF-OprN, and MexXY-OprM (Rehman et al., 2019).

To address phage resistance, exploring the fitness trade-offs induced by phage resistance has uncovered new avenues in the field of phage therapy. One strategy aims to restore the antibiotic susceptibility of antimicrobial-resistant bacteria as compensation for the development of phage resistance (Mangalea and Duerkop, 2020; Fujiki et al., 2023; Oromí-Bosch et al., 2023). In a previous study, we found that chromosomal deletions cause simultaneous deletion of the phage receptor encoding galU and the drug efflux transporter encoding mexX and mexY contributing to a trade-off between phage and antibiotics resistance in a P. aeruginosa veterinary isolate (Nakamura et al., 2021). However, the applicability of this concept are not completely understood. In this context, we summarize the key achievements related to fitness trade-offs between phage and antibiotics resistance via function of drug efflux transporters. Furthermore, we investigated whether this strategy can be applied to diverse P. aeruginosa strains from the aspects of their genome structure, and to quinolone-resistant P. aeruginosa strains that are not dependent on drug efflux transporter.

It has been observed that the function of drug efflux pumps is compromised in phage-resistant variants of P. aeruginosa either directly or indirectly due to infection by Pseudomonas phages (Chan et al., 2016; Koderi Valappil et al., 2021; Nakamura et al., 2021). Consequently, the restoration of the antibiotics sensitivity has been observed in several studies; however, previous analyses have utilized strains in which the absence of mutations in QRDR was not explicitly confirmed (Chan et al., 2016; Koderi Valappil et al., 2021), or a strain without known QRDR mutations (e.g., Pa12 isolate) (Nakamura et al., 2021). Therefore, it has been suggested for the first time that the loss of MexXY function alone may have limitations in inducing antibiotics sensitization in QRDR mutant strains through the current analysis. In strains harboring mutations in QRDR, it may be possible to induce mutations in DNA gyrase by infecting them with phages that target DNA gyrase within their infection cycle (De Smet et al., 2021), thereby applying another selective pressure. This could potentially impose a fitness cost for antibiotic sensitization through alteration of DNA gyrase function associated with phage resistance.

The acquisition of phage resistance through chromosomal deletions may not be a predominant mechanism for phage resistance. In the selection of brmts as shown in Figure 1F, only one strain out of 30 yielded brmt against four types of phages, suggesting that genetic mutations in phage receptor genes play a significant role in the acquisition of phage resistance. For instance, it has been shown that phage-resistant variants against Pseudomonas phages accumulate mutations in the genes such as wzy (Uchiyama et al., 2016; Li et al., 2018; Forti et al., 2023), leading to shorter lengths of LPS. However, such mutant strains are characterized solely by point mutations in a single gene and do not involve chromosomal deletions. It has also been reported that no effective antibiotic sensitization was observed in phage-resistant variants without chromosomal deletions in P. aeruginosa (Nakamura et al., 2021). Hence, increased selection pressure, leading to higher frequencies of chromosomal deletions involving MexXY, is speculated to be necessary to target the trade-off between phage and antibiotic sensitivity. Shen et al. (2018) reported a positive correlation between the expression level of MutL and the frequency of brmts occurrence in P. aeruginosa. Therefore, strategies such as inducing overexpression of MutL or promoting excessive activity of the Mut system in targeted P. aeruginosa strains during phage infection might serve as effective selection pressures to enhance the frequency of MexXY loss associated with chromosomal deletions, thereby inducing antibiotic resensitization. Additionally, studies aiming to achieve control of antibiotic sensitivity through the antibiotics themselves has been attempted, and it has been shown that substantial chromosomal deletion strains can also be selected by antibiotic-induced selection pressure (Cabot et al., 2016; Yen and Papin, 2017; Sanz-García et al., 2018). Therefore, there is a potential to impose higher selection pressure effectively by combining phages with other antibiotics. In addition, as shown in Figure 1E, there is also a considerable possibility that phages may not cover P. aeruginosa strains targeted. Hence, it is desirable to expand the diverse Pseudomonas phages capable of targeting a wide range of P. aeruginosa strains prior to actual applications.

To ensure effective management of trade-offs between phage and antibiotic sensitivity in clinical settings, it will be crucial to organize the phages and antibiotics that are involved in such trade-off relationships and understand their molecular basis. In fact, there are many other mechanisms underlying the trade-off between phage and antibiotic susceptibility (Mangalea and Duerkop, 2020; Fujiki et al., 2023; Oromí-Bosch et al., 2023). Interestingly, previous reports (Oechslin et al., 2017; Li et al., 2022; Menon et al., 2022; Wannasrichan et al., 2022) have highlighted other factors contributing to the restoration of antibiotic sensitivity in P. aeruginosa strains with large chromosomal deletions associated with phage resistance. These factors include changes in cell membrane potential and alterations in membrane permeability to antibiotics, such as cefepime (Li et al., 2022), fluoroquinolones such as CPFX (Oechslin et al., 2017) and LVFX (Li et al., 2022), and colistin (Menon et al., 2022; Wannasrichan et al., 2022). However, the deleted region in these studies (Oechslin et al., 2017; Li et al., 2022; Wannasrichan et al., 2022) contained galU and hmgA likely includes mexX and mexY as demonstrated in this study (Figure 1D). Additionally, the study by Menon et al. (2022) confirmed the presence of mexX and mexY (as well as galU and hmgA) within the deleted region. As it has been reported that MexXY contributes to the efflux of cephalosporins (Hocquet et al., 2006) and colistin (Puja et al., 2020), in addition to quinolones, it is important to carefully consider the influence of loss of MexXY as a factor in those antibiotic resensitization associated with phage resistance. Furthermore, another study indicates that antibiotic resensitization is limited when the mutant strain harbors extensive chromosomal deletions but MexXY remains intact (Barceló et al., 2021). Therefore, the contribution of MexXY loss would be considered significant, particularly in strains lacking other resistance mechanisms such as QRDR mutations. On the other hand, our previous analysis of the Pa12 isolate focused on the loss of MexXY associated with chromosomal deletions (Nakamura et al., 2021), but we did not examine changes in membrane potential or other related factors at that time. Therefore, it will be important to carefully examine both aspects and comprehensively discuss the molecular basis of changes in antibiotic sensitivity associated with chromosomal deletions, taking into account factors such as MexXY loss and changes in membrane potential and permeability.

Taken together, this study reveals that galU is closely located to mexX and mexY in diverse P. aeruginosa strains (255 strains), implying that a chromosomal deletion leading to phage resistance is likely to result in the deletion of mexX and mexY in general. Based on the previous report (Nakamura et al., 2021), while it could contribute to restoring sensitivity to antibiotics such as fluoroquinolones, the counterexample using the BEI16345 strain has shown that in P. aeruginosa strains with mutations in the QRDR, the loss of MexXY due to chromosomal deletions is insufficient to lead to significant changes in sensitivity to fluoroquinolones. This suggests that further studies will be required to apply this kind of trade-off to phage therapy, considering the presence or absence of QRDR mutations.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

JF: Conceptualization, Data curation, Funding acquisition, Project administration, Supervision, Validation, Visualization, Writing – original draft. KN: Data curation, Validation, Writing – review & editing. YI: Investigation, Writing – review & editing. HI: Funding acquisition, Supervision, Writing – review & editing.