Sequence specificity defines the effectiveness of PPMOs targeting Pseudomonas aeruginosa

ABSTRACT

Development of new therapeutics against antibiotic resistant pathogenic bacteria is recognized as a priority across the globe. We have reported using peptide-conjugated phosphorodiamidate morpholino oligomers (PPMOs) as species-specific antibiotics. The oligo sequences, 11 bases are designed to be complementary to specific essential genes near the Shine-Dalgarno site and inhibit translation. Here, we analyzed target specificity and the impact of genetic mutations on lead PPMOs targeting the rpsJ or acpP gene of Pseudomonas aeruginosa. Mutants in P. aeruginosa PAO1 were generated with four, two, or one base-pair mutations within the 11-base target sequence of the rpsJ gene. All mutants exhibited increased MICs compared to wild-type PAO1 when treated with the RpsJ PPMO, and the increase in the MICs was proportional to the number of base-pair mutations. Among single base-pair mutants, mutations in the middle of the sequence were more impactful than mutations in 5′ or 3′ end of the sequence. The increased MICs shown by the rpsJ mutants could be reversed by PPMOs designed to target the mutated rpsJ sequence. BALB/c mice infected intratracheally with mutants demonstrated increased lung burden when treated with RpsJ PPMO compared to wild-type PAO1-infected mice treated with RpsJ PPMO. Treating mice with a PPMOs designed to specifically target the mutant sequence was more effective against these mutant strains. These experiments confirm target specificity of two lead P. aeruginosa PPMOs and illustrate one potential mechanism of resistance that could emerge from an antisense approach.

INTRODUCTION

Development of antibiotic resistance in pathogenic bacteria is recognized as a major public health threat (1, 2). Loss of effectiveness of antibiotics threatens the success of many medical interventions such as general surgery, cancer therapy, and organ transplantation (3 - 5). Use of broad-spectrum antibiotics is considered as a key reason for development of antibiotic resistance in both targeted and non-targeted bacteria (6 - 8). Advances in the rapid identification of pathogens will make the prospect of pathogen-specific antibiotics a viable strategy (9). Previously, we and others reported using peptide-conjugated phosphorodiamidate morpholino oligomers (PPMOs) as species-specific antibiotics both in vitro and in vivo against a variety of pathogens, including Pseudomonas aeruginosa (10), Escherichia coli (11), Salmonella enterica (12), Acinetobacter baumannii (13), and members of the Burkholderia cepacia complex (14). Phosphorodiamidate morpholino oligomers (PMOs) used in our studies are 11-bases long and retain their natural nucleobase with a synthetic morpholino and phosphorodiamidate backbone. PMOs are designed to be complementary to mRNAs of essential genes and are most active near the Shine-Dalgarno sequence or the translation start site of the target gene. Complementary binding of PMOs with targeted mRNA is thought to exert their effect through inhibition of protein translation (15 - 17). PMOs are conjugated with arginine-rich cell-penetrating peptides to facilitate the intracellular entry of PMOs (18, 19).

As PMOs are thought to exert their effects through translation inhibition as a result of steric hindrance, we hypothesized that mutations in the target mRNA sequence would lead to reduced efficacy of the PPMO. Understanding specificity and the impact of target mutations would be important for ongoing clinical development of PPMOs as new antibiotics (20, 21). Here, we tested this hypothesis by mutating gene sequences corresponding to those targeted by lead P. aeruginosa PPMOs.

MATERIALS AND METHODS

All reagents were of analytical grade and were obtained from Sigma-Aldrich (St. Louis, MO), Thermo Fisher Scientific (Waltham, MA), or Acros Organics (Geel, Belgium) unless otherwise specified. Morpholinopropanesulfonic acid (MOPS) minimal medium was made according to the method of Neidhardt et al. with 1 g/L glucose and 100 µg/L thiamine (22).

Bacterial strains and plasmids

The P. aeruginosa strain PAO1 (ATCC 15692) and E. coli DH5α (ATCC 53868) used in this study were obtained from the American Type Culture Collection. Plasmid pRK2013 (23) was maintained in DH5α in LB media with 50 µg/mL kanamycin. pEX18Tc (24) was a kind gift of Herbert Schweizer and maintained in DH5α in LB media with 6.25 µg/mL tetracycline. pwFRT-TelR (25) was a kind gift from Tung Hoang. pFRT-Term-TelR is a derivative plasmid generated in the lab using pwFRT-TelR. Two synthetic terminator sequences were added upstream and downstream of the tellurite resistance gene of pwFRT-TelR using PCR cloning. Sequence of pFRT-Term-TelR can be found in Sequence S1 in the supplemental material. All the plasmids with tellurite resistance were maintained in DH5α in LB media with 30 µg/mL potassium tellurite.

PPMOs

PMO sequences were designed using a custom webtool that has inputs for a taxon ID, gene, alignment region, and oligomer length. As previously described, our lead PPMOs target acpP, rpsJ, and lpxC (10). PPMOs were synthesized by Sarepta Therapeutics (Cambridge, MA). PPMO gene targets, sequences, peptides, and peptide conjugation sites (5′ versus 3′) are indicated in Table 1. Scrambled (Scr) PPMO controls are 11-mer random sequences conjugated to the same peptides and in the same orientation as active PPMOs; however, they are not complementary to the essential gene target (13, 19, 26).

TABLE 1.

PPMOs^a

Target	NCBI gene ID	Designation	Base sequence (5′ to 3′)	Location of gene relative to start site	5′ end	3′ end
RpsJ	881717	RpsJ-RXR(3′)	CCT CAG ACT CC	−15 to −5	TEG	(RXR)4XB
RpsJ	881717	RpsJ-RXR(5′)	CCT CAG ACT CC	−15 to −5	(RXR)4XB	H
Mutant RpsJ	N/A	RpsJ_1bp5′-RXR(3′)	CCT CAG ACT CG	−15 to −5	TEG	(RXR)4XB
Mutant RpsJ	N/A	RpsJ_1bpM -RXR(3′)	CCT CAG ACA CC	−15 to −5	TEG	(RXR)4XB
Mutant RpsJ	N/A	RpsJ_1bp3′-RXR(3′)	GCT CAG ACT CC	−15 to −5	TEG	(RXR)4XB
Mutant RpsJ	N/A	RpsJ_2bp-RXR(3′)	CCT CCT ACT CC	−15 to −5	TEG	(RXR)4XB
Mutant RpsJ	N/A	RpsJ_4bp-RXR(3)	CCT CCT AAG CC	−15 to −5	TEG	(RXR)4XB
AcpP	879895	AcpP-RXR(3′)	CTC ATA CCT TG	−6 to +5	TEG	(RXR)4XB
AcpP	879895	AcpP-R6G(3′)	CTC ATA CCT TG	−6 to +5	TEG	R6G
Scr	N/A	SCR-RXR(3′)	CTG AGC ACG AC	N/A	TEG	(RXR)4XB

^a For the location relative to the start site, we defined ‘A’ of ATG as +1. Non-common. Abbreviations are as follows: R, Arginine; G, Glycine; B, β Alanine; X, 6-aminohexanoic acid (aminocaproic acid); TEG, Triethylene glycol; Scr, Scramble.

Allelic exchange approach to generate mutations in PPMO sites in P. aeruginosa PAO1

Genome modifications were performed by allelic exchange according to the procedure described in Hmelo et al. (27), with modifications described herein. Q5 High-Fidelity 2X Master Mix (NEB) was used for PCR amplifications. PCR products were combined using HiFi DNA Assembly Master Mix (NEB) to generate all derivative plasmids. All primers used are described in Table S1. The backbone of the allelic exchange vector contained sacB and tet genes from pEX18Tc, amplified using primers ON1- ON4. The tellurite resistance gene (telR) of pFRT-Term-TelR was PCR amplified using primers ON21 and ON22 (Fig. S1b; Table S1). For rpsJ allelic exchange vectors, the two arms of homology were amplified using primers ON9–ON12 (Fig. S1a). For acpP, the two arms of homologies were amplified using primers ON15–ON18 (Table S1). HiFi DNA Assembly Kit was used to create the allelic exchange vector as mapped in Fig. S1d. Plasmids with mutant alleles of rpsJ or acpP PPMO binding sites were amplified by site directed mutagenesis of plasmid containing wild-type PPMO binding site as the template, shown in Fig. S1d. ON23–ON27 primers were used to generate rpsJ mutants with 4 bp, 2 bp, 1 bp M′, 1 bp 5′, and 1 bp 3′ mutations. Sequences of these rpsJ mutants are shown in Fig. 1a. ON28 and ON29 primers were used to generate acpP mutants with 2 bp and 1 bp 5′ mutations (Table S1). Sequences of acpP mutants are shown in Fig. 2a Allelic exchange vectors were transformed into DH5α and clones were selected based on tellurite resistance (30 µg/mL), sacB sensitivity (10% sucrose wt/vol in LB with no salt), and sequencing the PPMO binding site. Confirmed vectors were isolated and transformed into DH5α already containing pRK2013 plasmid (Donor vector). Mating reactions were arranged between P. aeruginosa PAO1 with E. coli DH5α carrying both the allelic exchange vector and pRK2013 plasmid (Donor vector) by mixing freshly grown cells (1:5 ratio) and plating on LB agar plates. P. aeruginosa mutants were selected based on tellurite resistance (120 µg/mL) on Difco Pseudomonas isolation agar (PIA) and sacB sensitivity (10% sucrose wt/vol in LB with no salt) followed by confirmation of mutations with PCR sequencing. rpsJ mutants were confirmed with ON13 and ON14 sequencing primers, and acpP mutants were confirmed with ON19 and ON20 primers.

Fig 1.

Mutations in rpsJ gene sequence result in varied levels of resistance to RpsJ PPMOs in P. aeruginosa. (a) Heat map comparison of MICs (μM) for PPMOs against P. aeruginosa wild type and mutant strains grown in MOPS media. RpsJ-RXR(3′): rpsJ sequence targeted PMO with 3′ peptide; RpsJ-RXR(5′): rpsJ sequence targeted PMO with 5′ peptide; AcpP-RXR(3′): acpP sequence targeted with 3′ peptide; and Scr-RXR(3′): scrambled sequence with 3′ peptide. Mutation within the rpsJ sequence targeted by PPMOs are indicated in black letters. PAO1t_r contains the wild-type rpsJ gene along with the selection gene (telR) in mutants used to identify mutant strains. MICs were obtained from at least three independent biological replicates. CFUs were enumerated for P. aeruginosa wild type and mutant strains by serial dilution and plating for (b)16 µM (c) 8 µM and (d) 0 of RpsJ-RXR(3′). Significance was calculated using ANOVA/Tukey’s test. Each histogram represents the arithmetic mean  ±  S.D. (n  =  3, ****P  <  0.0001).

Fig 2.

Mutations in acpP gene sequence result in varied levels of resistance to AcpP PPMOs, but retain sensitivity to other gene-targeted PPMOs. Heat map comparison of MICs (µM) for PPMOs against P. aeruginosa wild type and mutant strains grown in MOPS media. AcpP-RXR(3′): acpP sequence targeted PMO with 3′ peptide; AcpP-R6G(3′): acpP sequence targeted PMO with 3′ peptide; RpsJ-RXR(3′): rpsJ sequence targeted with 3′ peptide; and Scr-RXR(3′): scrambled sequence with 5′ peptide. Mutation within the acpP sequence targeted by PPMOs are indicated in black letters. MICs were obtained from at least three independent biological replicates. (b) CFUs were enumerated for P. aeruginosa wild type and mutant strains by serial dilution and plating for three different concentrations, 8 µM, 1 µM, and 0 of AcpP-RXR(3′). Significance was calculated using ANOVA/Tukey’s test. Each histogram represents the arithmetic mean  ±  S.D. (n  =  3,****P  <  0.0001).

Bacterial susceptibility testing

All minimum inhibitory concentration (MIC) determinations were performed in 96-well non-tissue culture treated plates. MICs of the PPMOs were determined according to the Clinical and Laboratory Standards Institute (CLSI) broth microdilution method, with minor modifications (28). Briefly, each bacterial strain was diluted to a final concentration of approximately 5 × 10⁵ CFU/mL in MOPS medium, and the PPMO was serially diluted twofold between 16 and 0.125 µM in a 96-well tissue culture plate. Similarly, the plates were then covered with gas permeable membrane strips (MIDSCI, St. Louis, MO) and incubated at 37°C and 225 rpm in a shaking incubator for 18 to 20 hours. The optical densities at 600 nm (OD₆₀₀) were measured using a microtiter plate reader and the lowest dose of PPMO at which the average OD₆₀₀ that measured ≤0.06 was recorded as the MIC. At least three independent experiments were performed for each PPMO. The MICs for meropenem, colistin sulfate, and tobramycin were also obtained, as mentioned above. After OD₆₀₀ measurement, wells containing particular concentrations of PPMO were serially diluted in phosphate buffered saline and plated for CFU enumeration.

RpsJ and AcpP protein analysis utilizing label-free quantitation by LC-MS/MS

Total protein extracts were prepared from bacterial pallets grown overnight in MOPS using B-PER reagent (Thermo Fisher Scientific, Waltham, MA). Label-free quantitation by LC-MS/MS was performed as described in Behrmann et al. (29). Sum of the peak intensities for all the peptides identified for either RpsJ or AcpP was used for quantification after normalizing with total protein concentration of each sample.

Mouse experiments

All animal procedures were conducted according to the guidelines of the Emory University Institutional Animal Care and Use Committee (IACUC) under approved protocol number PROTO 201700441. The study was carried out in strict accordance with established guidelines and policies at Emory University School of Medicine and recommendations in the Guide for Care and Use of Laboratory Animals, as well as local, state, and federal laws. Six- to eight-week-old female BALB/c mice (Jackson Laboratories, Bar Harbor, ME) were anesthetized by intraperitoneal injection of 0.2 mL of a cocktail of ketamine (100 mg/mL) and xylazine (5 mg/mL) and infected by noninvasive intratracheal instillation with 5 × 10⁶ CFU (25 µL) of P. aeruginosa PAO1 or isogenic mutants, as previously described (30). At 15 minutes post-infection, mice were treated via the same route with 300 µg (15 mg/kg) (31) of the indicated PPMOs or PBS in a 25 µL volume. Mice were euthanized at 24 hours post-infection, and whole lungs were collected aseptically, weighed, and homogenized for 20 seconds in 1 mL of PBS, followed by serial dilution onto PIA, and plated for CFU enumeration.

RESULTS

Mutations in the target gene sequence impact the MIC of PPMOs to various degrees in P. aeruginosa

Targeted rpsJ mutant strains were generated utilizing homologous recombination techniques (Fig. S1). One, two, and four base-pair mutations were made in the target sequences in order to assess the impact of position and number of mismatches on PPMO MICs. A similar approach was used to generate acpP mutant strains with 1 and 2 bp mutations. All mutants and wild-type control achieved a similar OD₆₀₀ value and CFUs compared to wild-type PAO1 after 18 hours in the growth conditions used for MIC assays (Fig. S2a and b).

The impact on PPMO MIC was dependent on the number of base mismatches in the specific gene target sequence. rpsJ mutants with a 1 bp change either had no impact on the RpsJ PPMOs MIC or increased the MIC by two to fourfold depending on the location of 1 bp mutation within the PPMO site (Fig. 1a). However, 2 and 4 bp mutations had significant impacts on MICs (>16 µM). To confirm that this effect was specific to the change in the rpsJ sequence, we tested the PPMO targeting acpP. PPMOs targeting acpP were equally effective in the rpsJ mutants as in wild-type PAO1 (Fig. 1a). In addition, mutants and wild-type control were equally sensitive to traditional antibiotics as wild-type PAO1 (Fig. S3a through f) indicating that rpsJ mutants were resistant to RpsJ PPMOs due to the mutations in the PPMO target site and not by other antibiotic resistant mechanisms. The impact on MIC of various mutations was confirmed when quantitating bacterial burden in vitro and reduction in CFU/mL was sequence mismatch dependent (Fig. 1b). While 2 and 4 bp mutants showed no reduction in CFU/mL, 1 bp mutants showed a 1–5 log reduction in CFU/mL, depending on position of the mutated base and concentration of PPMO used.

Similar patterns were seen when mutating a second gene target, acpP (Fig. 2). A 1 bp mutant at the 5′ side of the target sequence (PAO1t_a1bp 5′) had no impact on MIC while a 2 bp mutant resulted in a multifold reduction with MICs of >8 µM (Fig. 2a). The RpsJ PPMO and traditional antibiotics remained equally active in both wild-type PAO1 and all acpP mutants that were constructed. Similar to what was seen with rpsJ, quantitative cultures demonstrated both dose-dependent and sequence mismatch dependent effects (Fig. 2b).

It is a possibility that introduced mutations could change mRNA stability and protein levels compared to wild-type PAO1. Label-free quantitation by LC-MS/MS was used to compare the protein levels of RpsJ or AcpP among mutants. There were no significant differences in the level of RpsJ or AcpP proteins compared to wild-type PAO1 (Fig. S4).

RpsJ mutants are sensitive to PPMOs targeting the altered rpsJ gene sequences

We generated new PPMOs for each specific mutated rpsJ sequence and compared their efficacy with the compatible mutant in MIC assays. As seen in Fig. 3, RpsJ PPMOs that matched the corresponding mutant sequence were highly active with MICs that were similar to what was achieved in wild-type PAO1. Quantitative cultures confirmed what was found in MIC assays (Fig. 3b and c).

Fig 3.

RpsJ mutants can be rescued by PPMOs targeting the mutant sequence. Heat map comparison of MICs (μM) for PPMOs against P. aeruginosa wild type and mutant strains grown in MOPS media. RpsJ-RXR(3′): wild-type rpsJ sequence targeted PMO with 3′ peptide; RpsJ_1bp5′-RXR(3′): PAO1t_r1bp5′ targeted PMO with 3′ peptide; RpsJ_1bpM -RXR(3′): PAO1t_r1bpM targeted PMO with 3′ peptide; RpsJ_1bp3′-RXR(3′): PAO1t_r1bp3′ targeted PMO with 3′ peptide; RpsJ_2bp-RXR(3′): PAO1t_r2bp targeted PMO with 3′ peptide; and RpsJ_4bp-RXR(3′): PAO1t_r4bp targeted PMO with 3′ peptide. MICs were obtained from at least three independent biological replicates. CFUs were enumerated for P. aeruginosa wild type and mutant strains by serial dilution and plating for (b) PAO1t_r4bp and (c) PAO1t_r2bp for three different concentrations, 16 µM, 8 µM, and 4 µM of RpsJ-RXR(3′), and PPMO designed to target the mutant PPMO sites. Significance was calculated using ANOVA/Tukey’s test. Each histogram represents the average  ± S.D. (n  =  3, **P  <  0.01,****P  <  0.0001).

In vivo activity of PPMOs is dependent on the number of target base-mismatches and can be rescued with new PPMOs

The therapeutic impact of mutations in the PPMO target site was further assessed by utilizing an in vivo model of acute pneumonia (Fig. 4). BALB/c mice were infected intratracheally with P. aeruginosa PAO1 or mutants (PAO1t_r1bp3′or PAO1t_r2bp) followed by a single treatment 15 minutes post-infection with PPMOs or PBS control. Lungs were harvested at 24 hours post-infection and CFU/g were enumerated. In vivo studies confirmed what had been seen in vitro. Mice infected with mutants had a higher lung burden of bacteria compared to wild-type PAO1 infected after treating with PPMO RpsJ-RXR(3′) that was designed to wild-type rpsJ sequence. PPMOs that had 1–2 bp differences from their target sequences resulted in a lung burden that was in between that seen in perfect matches and the lung burden in mice treated with either PBS or the scrambled PPMO (Fig. 4). Importantly, treatment of PPMOs [RpsJ_1bp3′-RXR(3′) and RpsJ_2bp-RXR(3′)] that were designed for mutants resulted a lower lung burden of mutant bacteria (Fig. 4).

Fig 4.

In vivo activity of PPMOs is dependent on base-pair specificity. Bacterial loads in the lungs of BALB/c mice infected intratracheally with P. aeruginosa PAO1 or mutants, PAO1t_r1bp3′ or PAO1t_r2bp, and treated with indicated PPMOs, Scr, or PBS control, at 15 minutes post-infection. The PPMOs, RpsJ-RXR(3′), RpsJ_1bpM -RXR(3′), and RpsJ_2bp-RXR(3′) designed for PAO1 wild type, PAO1t_r1bp3′, and PAO1t_r2bp, respectively. Mice were euthanized 24 hours post-infection and lungs were collected aseptically, weighed, and homogenized in 1 mL of PBS. Tissue homogenates were diluted and plated on PIA for CFU enumeration. The different colors reflects the different strains used to infect the mice. Closed symbols reflect treatments using compounds with 100% homology to the target binding sequence. Open symbols reflect treatments where at least 1 bp mismatch was present. Each point represents a single mouse. Error bars represent standard errors of the means (SEM). Data were analyzed using Graphpad Prism 9.5. Statistical differences were determined by one-way ANOVA and Dunn multiple comparison test and indicated as follows: **P < 0.01, ****P < 0.0001. These data are combined from two independent experiments.

DISCUSSION

P. aeruginosa displays resistance to a variety of existing antibiotics by intrinsic, acquired, and adaptive mechanisms (32). The discovery and development of novel antibiotics against P. aeruginosa infections are increasingly needed (33). In previous studies, PPMOs were designed to target essential genes such as lpxC, acpP, and rpsJ in P. aeruginosa (10, 31). Lead PPMOs displayed in vitro bactericidal activity against a large panel of clinical strains including multidrug-resistant ones and a survival benefit in vivo (10, 31). In this current study, we further validated the targets of two lead PPMOs, acpP and rpsJ, by demonstrating the impact of mutations in the PPMO target sequences. The results indicate a dose-dependent impact of the number of base mismatches on both in vitro and in vivo activity. Importantly for this analysis, the mutations generated in the upstream regions of the essential genes did not cause any major growth defects or impact on resistance to other small molecule antibiotics.

Importantly, one-base mismatches in the PPMO binding site did not automatically lead to an increase in the MIC. Among single base mutations in the rpsJ gene, our results indicate that a mutation in the middle of the target sequence impacts the MIC significantly more compared to one base mutation in either the 5′ or 3′ end. In addition, the MIC achieved by the original PPMO in the wild-type strain and MICs achieved by newly designed PPMO for mutants were not equivalent in some cases. It is unclear whether this finding is related to other factors such as secondary structure of the mRNA itself. Prokaryotic mRNA secondary structure is known to affect translation efficiency (34), similarly we believe it can have an effect on affinity towards antisense molecules as well (35). Loss of activity due to base mismatches has been demonstrated by Liam et al. using peptide nucleic acids (PNAs) (36, 37). They reported a substantial loss of inhibition in PNAs targeting beta-lactamase in E. coli due to mutations of two or six bases in the target site (36). Similar to our study, a PNA targeting the mutation could re-establish inhibition. The effect of the location of two consecutive mismatches in PNA bases to the target site had been studied by Jung et al. targeting acpP gene in Salmonella. They observed mismatches in the PNA resulted in positional related growth inhibition of Salmonella. PNAs that carry mismatches at either end of the sequence retained the activity while central mismatches resulted in substantial loss of activity (38). The impact of location on the mutations we observed aligned with the studies of Holen et al. (39) and Amarzguioui et al. (40), utilizing siRNA approaches in mammalian cell lines as well. Both these previous studies highlighted that those mutations in the central part of the antisense strand caused a pronounced decrease in activity, while mutations in the 5′ and 3′ ends were relatively tolerated. Additionally, Amarzguioui et al. (40) reported that siRNA generally tolerated mutations in the 5′ end, while the 3′ end exhibited less tolerance, which was similar to our observations in mutations in 5′ and 3′ ends of the PPMO. However, when we generated a 2 bp mismatch or greater in the binding sequence, we had complete loss of activity. Comparable results were observed by Bautla et al. (41) and Wilda et al. (42) in 21–23 nucleotide long siRNA when a central double mismatch with the target RNA was present. Our in vivo results were similar to what was seen in the in vitro setting.

PPMOs act as translation inhibitors and this is thought to occur through steric blockade of the ribosome complex binding to mRNA. It has been demonstrated that PMO position is directly related to in vitro activity, with the most active PPMOs usually targeting a region between the start ATG and the Shine-Delgarno sequence. For any given essential gene, this region is usually 100% identical in sequence for any given strain, species, and genus. Therefore, when designing PPMOs, we are usually successful in finding an 11-mer that covers all available genomes of interest. Although this region of a given essential gene seems to be evolutionarily conserved, this does not suggest that target sequence-based resistance mutations will not occur. Indeed, target mutation-induced drug resistance is a major mechanism of antibiotic resistance in general (43). In addition to the target validation we have performed in this study, we illustrate one potential mechanism of resistance that could emerge from an antisense approach. The fact that some PPMOs with single-base mismatches could still retained activity is encouraging; however, more work is needed to determine the actual frequency of resistance of PPMOs and the specific molecular drivers of resistance. If sequence-based mutations become a major mechanism of PPMO resistance, then one approach to circumventing resistance would be to combine PPMOs targeting more than one gene into a single formulation.

One limitation of this study is that we were unable to generate mutations in every base position of the target sequence. There were examples of mutants we attempted to generate downstream of the ATG sequence of lpxC gene which failed due to either non-viability or potential DNA repair mechanisms (data not shown). Although this finding could imply that there are certain areas of a target essential gene that could be even more potent regions for PPMO inhibition, there are a number of variables that go into the design of PPMOs, including positioning of the PMO, trying to avoid off-target hits in commensal bacteria, and ensuring that the 11-mer of interest is homologous across all known sequenced strains of a particular genus and species. Although we were unable to make all mutant combinations, we have demonstrated both target specificity of two of our lead Pseudomonas PPMOs as well as begun to understand the positional impact of mutations on antibacterial activity. Future and ongoing studies aim to elucidate the frequency and mechanisms of resistance of PPMOs in P. aeruginosa.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/aac.00245-23.

Figures S1 to S4, Sequence S1, Table S1. aac.00245-23-s0001.pdf.

Supplemental material.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.