Nitrate Respiration Promotes Polymyxin B Resistance in Pseudomonas aeruginosa

Abstract

Aims:

Polymyxin B (PMB) is known to require reactive oxygen species (ROS) for its bactericidal activity, but the mechanism of PMB resistance in various Pseudomonas aeruginosa strains has been poorly understood. This study examined the role of nitrate respiration (NR) of some P. aeruginosa strains in the PMB resistance.

Results:

We observed that the minimum inhibitory concentration (MIC) value of PMB against P. aeruginosa PA14 was eightfold reduced (from 2.0 to 0.25 μg/mL) by agitation, but not against P. aeruginosa PAO1 (from 2.0 to 1.0 μg/mL). Transcriptomic and phenotypic analyses using both strains and their NR mutants revealed that the higher NR in PAO1 than in PA14 accounted for the higher MIC value (i.e., PMB resistance) of PAO1, which was sufficient to compromise the antibacterial activity of PMB in Drosophila infections. We also confirmed the contribution of the NR to the PMB resistance is independent of the major catalase (KatA), suggesting that the NR might affect the ROS generation rather than the ROS disintegration. Furthermore, this PMB resistance was relatively common among clinical P. aeruginosa isolates and correlated with higher NR in those strains.

Innovation and Conclusion:

These results suggest P. aeruginosa strains could display intrinsic resistance to antibiotics in clinical settings and that NR is a crucial factor in the intrinsic antibiotic resistance, and also provide an insight into another key target for successful antibiotic treatment of P. aeruginosa infections. Antioxid. Redox Signal. 34, 442–451.

Introduction

P seudomonas aeruginosa (PA) is a well-known nosocomial pathogen that causes life-threatening infections especially in patients suffering from burn, cancer chemotherapy, chronic pulmonary diseases, and cystic fibrosis. PA infections are more and more difficult to treat due to the emergence of antibiotic resistance, thus resulting in severe outcomes worldwide and limited choices for therapeutic options (4, 6). For this reason, there has been resurgence in the use of old antibacterial regimens such as polymyxin B (PMB) (25), which can be ultimately considered as the last-resort treatment of such infections. PMB has shown relatively good clinical outcomes when it was intravenously administered to the PA-infected patients (34), but its use has been restricted to topological application due to its severe side effects (1).

PMB has in vitro bactericidal activity against almost all Gram-negative bacteria at relatively low concentrations, which involves its interaction with the lipid A of lipopolysaccharide (LPS) (17, 22, 32). PMB changes the packing order of LPS and increases the outer membrane (OM) permeability to various molecules inducing PMB itself (i.e., self-promoted uptake). Although the detailed mechanism of the OM permeability increase remains unveiled (11), it is obvious that the OM-disorganizing activity of PMB is independent of the bactericidal activity of PMB that is associated with the functions in the cytoplasmic membrane (CM) such as active transport and respiratory electron transfer (36, 38). PMB also interacts with the acidic phospholipids exposed on the CM (5) and, at high concentrations, induces ion-permeable pores in the OM and depolarizes the CM. But the bactericidal effect of PMB was evident even at lower concentrations and thus deemed independent of the CM depolarization (11).

Such complicated action of PMB on the Gram-negative bacteria is likely associated with the lower incidence of PMB-resistant bacteria (15). Nevertheless, there are PA clinical isolates that have acquired PMB resistance (3, 31, 37). The main reason for the resistance is decreased PMB binding due to the reduced anionicity of LPS, which involves modifications in the core region and/or lipid A (15). For example, the mutations in the two-component systems, PhoPQ or PmrAB result in the constitutive expression of the LPS modification (arn) operon encoding the proteins necessary for the aminoarabinosylation of the lipid A (30, 32, 37). This is the most common mechanism of the acquired PMB resistance observed in many Gram-negative bacteria. Besides the reduced LPS binding directed by LPS modification, other species-specific maneuvers of PMB resistance have been known; for example, the expression of multidrug efflux pumps or OM proteins such as OprH that can pump out PMB (42) and the presence of capsular polysaccharides that can bind to PMB by ionic interactions (7), both of which prevent PMB from reaching the bacterial target, resulting in the reduction of the effective concentration. Despite these advances in the understanding of the acquired resistance of PA to PMB, little has been known about the intrinsic resistance determinants to PMB in PA strains, considering that PA is quintessentially notorious for its intrinsic resistance to various antibacterials (33, 35).

Innovation

This is the first report to show that nitrate respiration promotes the intrinsic antibiotic resistance of the opportunistic human pathogen, Pseudomonas aeruginosa. These results also suggest that various P. aeruginosa strains could display differential nitrate respiration activity based on their own physiological adaptation in clinical settings and thus provide insight into another key factor to target for successful antibiotic treatment of P. aeruginosa infections.

In this study, we discovered the intrinsic PMB resistance of the most widely used PA strain, PAO1, which is not observed in another PA strain, PA14. The difference in intrinsic PMB resistance is directed by higher nitrate respiration (NR) even in PAO1 under aeration/agitation condition, which is sufficient to trounce the antibacterial efficacy of PMB in experimental infections. The higher NR activity in PAO1 that directs the higher expression of the major catalase, KatA, but the NR-independent contribution of KatA was suggested, in that NR could prevent the reactive oxygen species (ROS) generation by partially bypassing oxygen respiration, which might be observed at a considerable frequency in clinical PA isolates.

Results

PAO1, but not PA14, displays intrinsic PMB resistance

In our previous study, we observed that a widely used P. aeruginosa (PA) strain, PAO1, displays higher resistance to H₂O₂ and PMB than another PA strain, PA14 (19). Despite this observation, we have confirmed that both strains have the same minimum inhibitory concentration (MIC) value of 2.0 μg/mL as assessed by the conventional microdilution assay according to the Clinical & Laboratory Standard Institute (CLSI) guideline (10) (Table 1). In the previous study, we used different media (LBN; LB media containing 15 mM nitrate) for either surface cultures on the plates or submerged cultures under shaking (i.e., vigorous aeration/agitation). Therefore, we exploited the same culture conditions for MIC determination to verify the differential PMB resistance of the two strains. As shown in Table 1, the less MIC values were observed in the shaking condition than in the conventional assay condition for the tested PA strains, including PA14 and PAO1 as well as the PMB-resistant clinical isolates (P26 and P80) (24). This confirmed the quantitative difference in antibacterial activity of PMB between both culture conditions. Duetz et al. (14) reported that agitation conditions affect oxygen transfer rate (OTR) of the culture media: the OTR of nutrient broth was increased from 12.5 to 20.7 mmol O₂ L⁻¹·h⁻¹ just by 75° angled agitation. Moreover, surface plate cultures are most likely more oxic than submerged broth cultures. Thus, our previous culture conditions might have ensured higher OTR than the conventional microdilution assay condition. Since PMB involves ROS generation for its bactericidal activity (12), it is most likely that the alteration in the oxygen utilization of PA strains might affect their resistance to PMB.

Table 1.

Minimum Inhibitory Concentrations (μg/ml) of Polymyxin B Against Pseudomonas aeruginosa Strains Under Different Conditions^a

Strain	5-mL test tube (LBN/MH)	96-Well microtiter plate (LBN/MH)
PA14	0.25/0.25	2.0/2.0
PAO1	1.0/0.25	2.0/2.0
P26^b	4.0/4.0	8.0/8.0
P80^b	2.0/2.0	8.0/8.0

MIC values were obtained using both LBN broth and MH either under shaking (5-mL test tube) or conventional (96-well microtiter plate) conditions.

Clinical isolates of colistin-resistant PA strains.

LBN, LB supplemented with 15 mM KNO₃; MH, Müller–Hinton II; MIC, minimum inhibitory concentration; PA, Pseudomonas aeruginosa.

It is noted that, however, even under the shaking condition, PAO1 displayed higher MIC values than PA14 did in LBN but not in Müller-Hinton II broth (Table 1), suggesting that PAO1, but not PA14, showed some resistance to PMB in LBN only under agitation condition as in the surface culture conditions as our previous study (19). This “operational” or “intrinsic” PMB resistance that PAO1 displays clearly differs from the resistome-mediated acquired resistance of P26 and P80, because the resistance was conditionally observed (i.e., in LBN only under shaking condition in vitro).

The intrinsic PMB resistance of PAO1 involves NR

In an attempt to elucidate the physiological clues associated with the intrinsic PMB resistance of PAO1, we compared the transcriptomes of PAO1 and PA14 in LBN broth under shaking condition (Supplementary Table S1), which identified the genes differentially regulated in both strains. Most of the upregulated genes by more than twofolds in PAO1 belong to the genes involved in NR even in the normal aerobic culture condition. This observation is in a good agreement with our previous study that the aerobic NR triggers the expression of katAp2, which is dependent on Anr, the master regulator of the NR in PA (9, 39, 44). To verify the transcriptome results, the promoter activities of the selected genes (nirS for nitrite reductase and fhp for flavohemoprotein) were examined in both PAO1 and PA14 by using lacZ transcriptional fusion (Fig. 1A). The higher transcription activities were observed in PAO1 than in PA14, indicating that PAO1 has higher NR activity than PA14. We also measured the mRNA levels of the nirS gene in the PMB-resistant strains (P26 and P80) (Fig. 1B). P26 and P80 have a mutation in pmrA and phoQ, respectively, which induce the overexpression of the pmrH gene required for the PMB resistance (24). The similar nirS mRNA levels of P26 and P80 as that of PA14 indicate that the acquired PMB resistance in P26 and P80 is not associated with higher NR and that the mechanism of the intrinsic PMB resistance of PAO1 might differ from those of the acquired PMB resistance. Besides the gene expression results, the higher NR activity in PAO1 than in PA14 was also confirmed by the higher steady-state level of nitrite (NO²⁻) (Fig. 1C), in that nitrite, a stable intermediate in NR is one of the tractable biomarkers of NR and/or nitric oxide production (27).

FIG. 1.

Transcriptional levels and enzyme activity of nitrate respiration genes. (A). Promoter activities of fhp and nirS in PA14 and PAO1. The transcriptional fusions for fhp and nirS were used to measure the corresponding promoter-driven LacZ activity (Miller unit) in both PA14 (empty) and PAO1 (filled) cells harboring the vector alone (pQF50) or one of the fusions (either pQF50-fhp or pQF50-nirS). The values are the averages from the three independent experiments with the error bars representing the standard deviations. The statistical significance based on unpaired t-tests: **p < 0.01. (B) Transcriptional levels of nirS in PA14, PAO1, and the colistin-resistant PA strains (P26 and P80). Data were normalized to the level in PA14. Error bars represent the standard deviations of three independent measurements. The statistical significance based on unpaired t-tests with that in PA14: **p < 0.01. (C) NO₂ ⁻ accumulation during growth. NO₂ ⁻ accumulation was measured by Griess reagents using the culture supernatants of PA14 (empty) and PAO1 (filled) at the indicated time points. The NO₂ ⁻ concentrations were calculated from the OD₅₄₀ values using a standard curve. The values are the averages from the three independent experiments with the error bars representing the standard deviations. PA, Pseudomonas aeruginosa.

To verify whether NR is indeed necessary for the intrinsic PMB resistance of PAO1, we exploited the anr mutants of PAO1 and PA14, which are completely defective in NR (43) and measured their MIC values. As shown in Figure 2, unlike the PA14 anr mutant, whose susceptibility to PMB was no less than that of the isogenic wild type, the PAO1 anr mutant showed lower MIC value than the wild type (i.e., 0.25 μg/mL vs. 2.0 μg/mL) under LBN-agitation condition, similar to the MIC values of PA14 and the PA14 anr mutant (i.e., 0.25 μg/mL). All these results led us to the conclusion that the intrinsic PMB resistance of PAO1 might involve the Anr-mediated upregulation of the NR genes and the subsequently increased NR activity under LBN-agitation condition.

FIG. 2.

MIC measurements for PA14, PAO1, and their anr mutants. MIC values of PMB were measured against PA14 and PAO1 WT and their anr mutants in LBN under two different conditions. The numbers indicate the concentrations of PMB (μg/mL) and the vertical red lines denote the boundaries between MICs and the maximal growth-permitting concentrations. MIC, minimum inhibitory concentration; PMB, polymyxin B; WT, wild type. Color images are available online.

NR-associated PMB resistance is independent of KatA

KatA, the major catalase of PA, is preponderantly responsible for detoxifying H₂O₂ and its expression is upregulated under NR conditions (9). More importantly, our recent observation that the KatA expression is higher in PAO1 than in PA14 (19) has prompted us to investigate whether the NR-mediated upregulation of KatA might be necessary for the intrinsic PMB resistance of PAO1. We first generated the nirSkatA double mutants of PAO1 and PA14 to address the genetic interaction between NR and KatA on the PMB resistance. The created mutants were tested for their ability to grow in the prolonged exposure to PMB (0.125 μg/mL) in LBN broth as described previously (19). As shown in Figure 3, no difference was observed between the wild type and the nirS mutant as well as between the katA and the nirSkatA mutants in the PA14 background. However, as expected from Figure 2, the PAO1 nirS mutant showed similar susceptibility as the wild type PA14 or the PA14 nirS mutant (Fig. 3C). Interestingly, the PAO1 nirS mutant showed similar PMB susceptibility as the PAO1 katA mutant, whereas the PAO1 nirSkatA mutant was even more susceptible to PMB than the PAO1 nirS mutant (Fig. 3D). This result indicates that the NR-mediated intrinsic PMB resistance of PAO1 is independent of (i.e., not hypostatic to) the KatA function, although it is also probable that reactive nitrogen intermediates such as nitric oxide generated during NR might induce KatA expression, the KatA-independent contribution of NR is evident, most likely affecting different step(s) of the PMB toxicity.

FIG. 3.

Growth of nirS mutants upon PMB treatment. Growth curves were measured for PA14 (A, C) and PAO1 (B, D) WT (filled circle) and their nirS (filled triangle), katA (empty square), and nirSkatA (empty diamond) mutants. Each cell was grown in LBN containing either nothing (−PMB) (A, B) or 125 ng/mL of PMB (+PMB) (C, D). The growth was measured by OD₆₀₀ at the indicated time points. The values are the averages from the three independent experiments with the error bars representing the standard deviations. LBN, LB media containing 15 mM nitrate.

To gain a mechanistic insight into the NR-mediated PMB resistance, we investigated whether or not the PMB susceptibility of the nirS mutant could be recovered by an ROS quenching chemical, thiourea (TU) (18). The nirS, katA, and nirSkatA mutants were pulse-treated with PMB (2.0 μg/mL for 1 h) followed by TU treatment (5 mM), then the surviving cell proportions were counted after fluorescent staining of the live and dead cells (Fig. 4A, B). Clearly, based on the proportions of dead cells, it is notable that all the nirS and/or katA mutants survived the PMB treatment in the presence of TU. This result suggests that NR could reduce the accumulation of ROS presumably in part by bypassing oxygen respiration, the main source of ROS generation. Considering that PMB treatment can induce ROS generation (12, 26, 36), we were prompted to address whether the PMB treatment could generate ROS or just require ROS without affecting the amount of ROS. To this end, the steady-state level of H₂O₂ was measured in those mutants upon PMB treatment by using H₂O₂-specific dye, Amplex Red. As shown in Figure 4C, more ROS accumulation (31.7 nM of H₂O₂) was observed in PA14 than in PAO1 (19.76 nM H₂O₂). Most importantly, the H₂O₂ level in the PAO1 nirS mutant was increased to 27.68 nM, indicating the possibility that higher NR activity in PAO1 effectively mitigates the ROS accumulation. Taken together, higher KatA and NR activities are deemed required for the intrinsic PMB resistance in PAO1. This dismantling of the contributions of KatA and NR to the PMB resistance suggests that NR could prevent ROS generation, whereas KatA detoxifies generated ROS.

FIG. 4.

Antioxidant effect on PMB treatment. (A, B) The survival of PA14 and PAO1 WT and their mutants (nirS, katA, and nirSkatA) was measured by Live/Dead BacLight staining upon PMB (2.0 μg/mL) treatment in the presence or absence of 5 mM of TU. The microscopic observation (A) and its quantification (B) are presented as described in Materials and Methods section. The cell counts are the averages from the three independent experiments with the error bars representing the standard deviations. (C) H₂O₂ accumulation in PA14, PAO1, and their mutants (nirS, katA, and nirSkatA) upon PMB (125 ng/mL) treatment. The H₂O₂ concentrations were calculated from the fluorescence values as described in Materials and Methods section. The H₂O₂ concentrations are the averages from the three independent experiments with the error bars representing the standard deviations. The statistical significance based on unpaired t-test of all the mutants compared with the corresponding wild type: *p < 0.05. TU, thiourea. Color images are available online.

Intrinsic PMB resistance compromises the PMB treatment in infection conditions

To observe the significance of this NR-mediated intrinsic PMB resistance of PAO1 in infection conditions, we have investigated the antibacterial efficacy of PMB using the Drosophila systemic infection model. Flies were infected with both strains and fed with PMB after infection to see if the intrinsic PMB resistance can compromise the antibacterial activity of PMB in the infection condition. As shown in Figure 5A, PMB treatment (2 mg/mL per vial) was able to rescue the flies from the mortal infection caused by PA14. In contrast, however, the PAO1-infected flies were not rescued by PMB treatment (Fig. 5B). More importantly, the involvement of NR in the PMB resistance of PAO1 was also confirmed, in that the nirS mutation compromised the PMB resistance in PAO1. It is also noted that the mortality of the PA14 nirS mutant was no less than that of the isogenic wild type, irrespective of PMB treatment. This result underlines that the NR-mediated intrinsic PMB resistance of PAO1 is sufficient to overcome the antibacterial efficacy of PMB in infection conditions.

FIG. 5.

Antibacterial efficacy of PMB in Drosophila systemic infection. The flies that had been infected with one of PA14 (A) and PAO1 (B) WT (circles) and their nirS mutants (diamonds) were transferred to a new cornmeal medium overlaid with either water (empty) or 2 μg/mL of PMB (filled). The dashed lines represent the time required to reach 50% mortality in each infection. The values are the averages from three independent experiments using a group of 30 flies per vial. The statistical significance based on a log-rank test: **p < 0.01; ***p < 0.001. Color images are available online.

NR correlates with the intrinsic PMB resistance in PA clinical strains

The aforementioned results are supporting that the higher NR activity in PAO1 is an important determinant for the intrinsic PMB resistance, which is not observed in PA14. We have wondered how common this strain-specific intrinsic resistance is in clinical PA isolates. To this end, the MIC values were measured for the PA isolates in LBN under both shaking and conventional assay conditions. Based on the differences between MIC values measured in both conditions, the tested 29 strains were divided into 3 groups (I to III) (Supplementary Table S2): 19 group I strains, including PA14, showed no PMB resistance, whereas 10 group II strains, including PAO1, showed intrinsic PMB resistance. The group III strains (P26 and P80) exhibit acquired PMB resistance.

To test whether the group I and II strains might differ in NR activity, the nirS-lacZ fusion construct was introduced into the clinical PA strains and β-galactosidase assay was performed as in Figure 1A (Fig. 6). It was conspicuous that most (15 out of 19) of the group I strains showed relatively lower nirS promoter activities, whereas the 10 group II strains showed higher nirS promoter activities. A few outliers existed especially among the group I strains, presumably due to other contributing factors, including differential expression of KatA (19) and/or the resistance determinants such as efflux pumps and membrane modification (13, 29). Nevertheless, the statistically significant correlation between the nirS promoter activity (i.e., NR) and the intrinsic PMB resistance in clinical PA strains suggests that the NR-mediated PMB resistance might be a relatively common trait of PA strains in the clinical settings.

FIG. 6.

nirS promoter activities in PA strains. The nirS promoter activities of 19 group I (MIC of 0.25 μg/mL; empty box) and 10 group II (MIC of 1.0 μg/mL; gray box) strains (Supplementary Table S2) are presented by box-whisker plot: the lower boundary of the box indicates the 25th percentile, a black line within the box indicates the median, and the upper boundary of the box indicates the 75th percentile; the whiskers below and above the box represents the 10th and 90th percentiles, respectively, with the four outliers above the whisker of group I. Total numbers of tested strains of each group are designated above the whiskers. The p-value from unpaired t-test is indicated.

Discussion

This is the first study to highlight the function of NR in PMB resistance, initially based on the comparison using two widely used P. aeruginosa (PA) laboratory strains, PAO1 and PA14. It becomes more and more important to reincarnate the old drugs such as PMB into the modern antibacterial regimen in the era of antimicrobial resistance, which has been facilitated by the recent advancement in the knowledge about the antibacterial and resistance mechanisms of the old drugs. This study provides an important insight into another layer of the molecular mechanisms of antibiotic resistance to PMB in PA strains and a key consideration for treatment options against PA infections, given that PMB is currently one of the last-resort antibiotics for the treatment of infections caused by the multidrug-resistant Gram-negative bacteria.

The bactericidal mechanism of PMB is generally attributed to its ability to bind to the OM lipid A. The classical explanation is that PMB-lipid A binding results in membrane dysfunction due to membrane lysis and/or component exchange at the OM and the CM through its hydrophobic parts such as the N-terminal fatty acyl chain and the amino acid residues such as Phe and Leu. An alternative mechanism that has been recently suggested is the involvement of highly toxic ROS that can be generated through aberrant electron transfers in the PMB-perturbed membrane. Superoxide anion (O₂ ⁻) can be formed upon entry of PMB into and cross OM and CM (36). O₂ ⁻ is subjected to enzymatic conversion to hydrogen peroxide (H₂O₂), which leads to the formation of the most toxic ROS, hydroxyl radical (^•OH) through Fe²⁺ oxidation to Fe³⁺, which reverts back to Fe²⁺ by absorbing an electron from another molecule of H₂O₂ (Fenton reaction) or from O₂ ⁻ (Harber–Weiss reaction).

Despite some disputes still going on (21, 28), ROS generation is regarded as a common killing mechanism for the bactericidal antibiotics (21, 26). It is yet to be ascertained that ROS generation is the sole or at least the primary contributor to the antibacterial activity of PMB. However, the data in this study validated that PAO1, but not PA14, has intrinsic resistance to PMB directed by higher NR activity. Therefore, it is certain that ROS generation can be regarded as an important contributor to the antibacterial mechanism of PMB in this context. More importantly, this intrinsic resistance of PAO1 is sufficient to abrogate the antibacterial activity of PMB in Drosophila systemic infections and some clinical PA isolates have higher NR activity, displaying higher MIC in vitro. Thus, it needs to pay more attention to use PMB for the treatment of multidrug-resistant PA infections.

Yoon et al. (41) previously showed that PA respires on nitrate during chronic lung infections and tends to form robust biofilms during anaerobic growth. This implies that PMB resistance might be triggered for some PA strains under the NR and/or biofilm growth conditions, which needs to be further investigated in the relevant infection conditions. Recently, the PMB resistome of PA14 was characterized using its nonredundant transposon mutant library, which identified the genes involved in the intrinsic layer of resistance in PA14 (16). Although they might have provided a comprehensive snapshot of the PMB resistance mechanisms in PA strains, we were able to identify another layer of the PMB resistance that was more evidently observed in the PA strains with higher NR activity as PAO1. Since the higher NR activity is deemed required for successful chronic infections by PA, targeting the physiological determinants to trigger NR would be combined with PMB treatment to ensure the antibacterial efficacy of PMB to treat chronic infections. Because the intrinsic resistance and the associated mechanisms involving NR diminishes the effectiveness of PMB and potentially other related antibiotics, the successful inhibition of NR can be considered as an augmentation therapy. More importantly, the fact that the enzymes in NR are not found in human and livestock prompts us to design specific inhibitors targeting these enzymes for the successful treatment of the chronic PA infections.

Materials and Methods

Bacterial strains and culture conditions

All bacterial strains and plasmids used in this study are listed in Table 2. Strains of P. aeruginosa (PA) were grown at 37°C using Luria–Bertani (LB) broth, LB containing 15 mM KNO₃ (LBN) or on 2% Bacto-agar (Difco) LB plates for 18 h. Bacterial cultures that had been grown to the late stationary growth phase (OD₆₀₀ of 5.0) were inoculated into LB broth with an inoculum of 1.6 × 10⁷ colony forming units (cfu)/mL and then grown to the late-logarithmic growth phase (OD₆₀₀ of 0.7) before being used for experiments.

Table 2.

Strains Used in This Study

Strain or plasmid	Relevant characteristics or purpose	Reference or source
Pseudomonas aeruginosa
PA14	Wild type laboratory strain; Rif^R	Laboratory collection
PA14 anr	PA14 with in-frame deletion of anr; Rif^R	(9)
PA14 nirS	PA14 with in-frame deletion of nirS; Rif^R	This study
PA14 katA	PA14 with in-frame deletion of katA; Rif^R	(23)
PA14 katAnirS	PA14 with in-frame deletion of katA and nirS; Rif^R	This study
PAO1	Wild type laboratory strain	Laboratory collection
PAO1 anr	PAO1 with in-frame deletion of anr	This study
PAO1 nirS	PAO1 with in-frame deletion of nirS	This study
PAO1 katA	PAO1 with in-frame deletion of katA	(19)
PAO1 katAnirS	PAO1 with in-frame deletion of katA and nirS	This study
PAK	Wild type laboratory strain	Laboratory collection
PMM3 ∼ 52	Clinical isolates from various Asian countries	Laboratory collection
Escherichia Coli
DH5α	Multipurpose cloning	Laboratory collection
S17–1	Conjugal transfer of mobilizable plasmid; Tp^R; Sm^R	Laboratory collection
Plasmids
pEX18T	Positive selection suicide vector for allelic exchange; Cb^R	Laboratory collection
pEX18T-anr^a	pEX18T with the in-frame deletion in the anr gene	(9)
pEX18T-katA^a	pEX18T with the in-frame deletion in the katA gene	(9)
pEX18T-nirS^a	pEX18T with the in-frame deletion in the nirS gene	This study
pQF50	lacZ transcriptional fusion; Cb^R	Laboratory collection
pQF50-nirS^b	pQF50 with the nirS promoter	This study
pQF50-fhp^b	pQF50 with the fhp promoter	(2)

The deletion constructs were generated for both PA14 and PAO1.

The promoter fusion constructs were generated using either PA14 or PAO1 chromosome.

Cb^R, carbenicillin- and ampicillin-resistant; Rif^R, rifampicin-resistant; Sm^R, streptomycin-resistant; Tp^R, trimethoprim-resistant.

Oligonucleotide primers

The DNA oligonucleotide primers used for gene deletion, expression, and detection in this study are listed in Supplementary Table S3.

Generation of deletion mutants

All the in-frame deletion mutants were created using pEX18T. Oligonucleotide primers were designed using the PAO1 genome sequence. anr, nirS, and nirSkatA mutants were created by SOEing (splicing by overlap extension) polymerase chain reaction (PCR) as described previously (23), using four oligonucleotide primers listed in Supplementary Table S3. The resulting PCR products were cloned into pEX18T and then introduced into both PAO1 and PA14 chromosomes. In general, >30% of the coding regions were deleted for all the deletions, which were verified by PCR.

Measurement of MICs

The MICs of PMB were determined against 31 PA strains using the standard microbiological procedures according to the CLSI guideline, as per document no. (M7-A10) and the 5-mL test tube assay for vigorous agitation. In brief, bacterial cells that had been grown to OD₆₀₀ of 0.7 were diluted to 5.0 × 10⁵ cfu/mL and mixed with the equal volume of 30 mM KNO₃-amended LB containing serially diluted PMB (from 16 μg/mL to 125 ng/mL) or Müller–Hinton II broth with PMB (from 16 μg/mL to 125 ng/mL). Then, the mixtures were grown at 37°C for 18 h either with 96-well microtiter plates in a shaking incubator (HANBEAK, Korea) (200 rpm) or with 5-mL test tubes in a rotator (Glas-col^®) (80 rpm). The MIC values were recorded as the lowest concentration at which no signs of growth were observed based on the OD₆₀₀ value of <0.05 after 18-h incubation.

Transcriptome analysis

Affymetrix PAO1 GeneChip was used for microarray analysis. RNA was isolated from the late logarithmic-growth phase (OD₆₀₀ of 0.7) cultures in LBN broth of either PAO1 or PA14. Trizol (Invitrogen) was used and the samples were treated according to the manufacturers' recommendation. The eluted RNA was treated with DNase I for 1 h at 37°C. After enzyme inactivation at 65°C for 1 h, further treatment of RNA samples was done as described by the manufacturer (Affymetrix) with minor modifications as suggested elsewhere (40). cDNA synthesis, fragmentation, and labeling and hybridization, and washing were performed according to the manufacturer's instruction. Three independent experiments were performed using independent RNA preparations. The data were analyzed using Affymetrix microarray suite software suite (MAS) (version 5.0) with the default parameters. Gene expression changes were calculated by comparisons of the signal averages from PAO1 and PA14 cultures. The statistical significance of the gene expression changes was analyzed according to the Mann–Whitney U test with a cutoff p-value of 0.05 (8).

Construction of lacZ fusions and β-galactosidase assay

The transcriptional fusion of nirS were created by SOEing PCR using four oligonucleotide primers (Supplementary Table S3). In brief, the nirS promoter regions from both PAO1 and PA14 were prepared using the primer pair (nirS-lacZ-F and nirS-lacZ-UR) and the cognate chromosomes. The partial lacZ coding region was amplified using the primers pair (nirS-lacZ-DF and pQF50-R1) and pQF50. Then, the PCR products were fused by SOEing PCR using nirS-lacZ-F and pQF50-R1 and the amplified products were cloned into pQF50. The fhp transcriptional fusions were created as described elsewhere (2) with the primers listed in Supplementary Table S3. The lacZ fusion constructs were then used to transform the cognate PA strain by electroporation and the pQF50-nirS for PAO1 was used to transform the clinical PA strains except for PA14. LacZ (β-galactosidase) activity was determined as previously described (20). The results are presented as the means with standard deviations and were analyzed by unpaired t-tests using GraphPad Prism 5.0 statistical software. For box and whisker plotting, unpaired t-tests were performed to analyze one-tailed p-value with confidence intervals of 95%. A p-value of <0.05 was considered statistically significant.

Real-time PCR

RNA was extracted from the late-logarithmic growth phase (OD₆₀₀ of 0.7) cultures with RNeasy Mini Kit (QIAGEN, Germany) according to the manufacturer's instruction. The cDNA samples that had been synthesized from the RNA samples (1.0 μg/mL) using ReverTra Ace qPCR RT kit (Toyobo, Japan) were subjected to qPCR with Thunderbird SYBR qPCR Mix (Toyobo) for StepOnePlus qPCR system (Applied Biosystems). The cDNA synthesis and qPCR primers are listed in the Supplementary Table S3.

Measurement of NO₂ ⁻ accumulation

The steady-state level of NO₂ ⁻ during the growth in LBN was measured using the modified Griess reagent (Sigma). In brief, the cells that had been grown up to the OD₆₀₀ of 3.0 were centrifuged at 8000 g for 5 min and the supernatants were collected. The aliquot (100 μL) from the supernatants at the indicated time points were mixed with the equal volume of modified Griess reagent and incubated at 25°C for 15 min. The NO₂ ⁻ concentrations were calculated from the absorbance of the supernatants that had been measured in 96-well microtiter plates at 540 nm. The measurements were repeated three times and the results were presented as the means with standard deviations.

PMB-mediated growth inhibition

Bacterial cells that had been grown to OD₆₀₀ of 0.7 were diluted to 5.0 × 10⁵ cfu/mL in LBN broth with 125 ng/mL of PMB, then incubated in a rotator (Glas-col) under shaking (80 rpm) for 15 h at 37°C. Aliquots (100 μL) from the cultures were collected at the indicated time points for OD₆₀₀ measurement.

Microscopic observation of cell death

Live/Dead-BacLight^® (Thermo-Fischer) was used to observe the cell death. In brief, cells that had been grown to OD₆₀₀ of 0.5 were pulse-treated with 2.0 μg/mL of PMB in LBN broth with or without 5 mM TU (Sigma) in the rotator under shaking (80 rpm) for 1 h at 37°C. After incubation, 1 mL of each culture was washed with PBS three times and resuspended in the equal volume of PBS and 3 μL of the dye was added to the resuspension, followed by incubation for 15 min in the dark. The stained cells were mounted on a 2% agarose pad for fluorescence observation using a fluorescent microscopy (Zeiss, Germany). Two single channel (green: viable cells, red: dead cells) pictures of three different random locations were selected and saved for further analysis. The selected pictures were then converted into 8-bit images using ImageJ (version 1.52k) for cell counting. Particle analyzer (default) and cell-counter (add-on) were used to count the numbers of viable and dead cells. The counting results were presented as the means with standard deviations of the three independent experiments.

Measurement of H₂O₂ accumulation

H₂O₂ accumulation was measured using the Amplex^® Red Hydrogen Peroxide assay kit (Thermo-Fischer). In brief, aliquots (50 μL) of the cells that had been grown for the indicated time points in the rotator (Glas-col) under shaking (80 rpm) with LBN containing 125 ng/mL of PMB were mixed with the equal volume of Amplex Red reagent, followed by incubation for 30 min. The fluorescence was measured using 96-well microtiter plates at 580 nm by excitation at 550 nm. H₂O₂ concentrations were calculated by using a standard curve. The measurement was repeated three times and the results were presented as the means with standard deviations. The results were analyzed by unpaired t-test with GraphPad Prism 5.0 statistical software. A p-value of <0.05 was considered statistically significant.

Determination of antibacterial efficacy in vivo

To determine antibacterial efficacy of PMB in vivo, Drosophila systemic infection was performed as previously described (20). PA cells that had been grown to OD₆₀₀ of 3.0 were diluted into 10 mM MgSO₄ (up to 10⁷ cfu/mL). Drosophila melanogaster strain Oregon R was grown at 25°C with 5 mL of corn meal-dextrose medium (0.93% agar, 6.24% dry yeast, 4.08% corn meal, 8.62% dextrose, 0.1% methyl paraben, and 0.45% [v/v] propionic acid). For infection, 4- to 5-day-old adult female flies were infected by pricking at the dorsal thorax with a 10 μm needle (Ted Pella) that had been dipped in the bacterial suspension in 10 mM MgSO₄. Infected flies were transferred to a new medium overlaid with PMB at the final concentration of 2 mg/mL. Survival rate of infected flies were monitored for up to 54 h postinfection. Flies that died within 12 h were excluded in mortality determination. Mortality assay was repeated at least three times. Log-rank tests were used to compare the survival rate difference between the groups. A p-value of <0.05 was considered to be significant.

Authors' Contributions

Y.-H.C. conceived and designed the research. B.-o.K, H.-J.J., and I.-Y.C. designed and performed the experiments, and collected and analyzed the experimental data. H.-W.B. and E.S.K. provided reagents. B.-o.K., H.-J.J., and Y.-H.C. wrote the article. All authors reviewed the article.

Footnotes

Author Disclosure Statement

No competing interests exist.

Funding Information

This study was supported by the National Research Foundation of KOREA (NRF) Grant (NRF-2017R1A2B3005239).

Supplementary Material

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

Abbreviations Used

References

Abdelraouf

, Braggs

, Yin

, Truong

, Hu

, and Tam

. Characterization of polymyxin B-induced nephrotoxicity: implications for dosing regimen design. Antimicrob Agents Chemother, 56: 4625–4629, 2012.

Arai

, Hayashi

, Kuroi

, Ishii

, and Igarashi

. Transcriptional regulation of the flavohemoglobin gene for aerobic nitric oxide detoxification by the second nitric oxide-responsive regulator of Pseudomonas aeruginosa . J Bacteriol, 187: 3960–3968, 2005.

Barrow

and Kwon

. Alterations in two-component regulatory systems of phoPQ and pmrAB are associated with polymyxin B resistance in clinical isolates of Pseudomonas aeruginosa . Antimicrob Agents Chemother, 53: 5150–5154, 2009.

Blair

, Webber

, Baylay

, Ogbolu

, and Piddock

. Molecular mechanisms of antibiotic resistance. Nat Rev Microbiol, 13: 42–51, 2015.

Brandenburg

, Rubinstein

, Sadikot

, and Onyuksel

. Polymyxin B self-associated with phospholipid nanomicelles. Pharm Dev Technol, 17: 654–660, 2012.

Cabot

, Zamorano

, Moya

, Juan

, Navas

, Blazquez

, and Oliver

. Evolution of Pseudomonas aeruginosa antimicrobial resistance and fitness under low and high mutation rates. Antimicrob Agents Chemother, 60: 1767–1778, 2016.

Campos

, Vargas

, Regueiro

, Llompart

, Alberti

, and Bengoechea

. Capsule polysaccharide mediates bacterial resistance to antimicrobial peptides. Infect Immun, 72: 7107–7114, 2004.

Chang

, Small

, Toghrol

, and Bentley

. Microarray analysis of Pseudomonas aeruginosa reveals induction of pyocin genes in response to hydrogen peroxide. BMC Genomics, 6: 115, 2005.

Chung

, Kim

, Jang

, and Cho

. Dual promoters of the major catalase (KatA) govern distinct survival strategies of Pseudomonas aeruginosa . Sci Rep, 6: 31185, 2016.

10.

CLSI. Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobicaly: Approved Standard. M07-A10. Wayne, PA: Clinical and Laboratory Standards Institute, 2015.

11.

Daugelavicius

, Bakiene

, and Bamford

. Stages of polymyxin B interaction with the Escherichia coli cell envelope. Antimicrob Agents Chemother, 44: 2969–2978, 2000.

12.

Dong

, Dong

, Catalano

, Moore

, Liang

, and Mekalanos

. Generation of reactive oxygen species by lethal attacks from competing microbes. Proc Natl Acad Sci U S A, 112: 2181–2186, 2015.

13.

Dreier

and Ruggerone

. Interaction of antibacterial compounds with RND efflux pumps in Pseudomonas aeruginosa . Front Microbiol, 6: 660, 2015.

14.

Duetz

, Ruedi

, Hermann

, O'Connor

, Buchs

, and Witholt

. Methods for intense aeration, growth, storage, and replication of bacterial strains in microtiter plates. Appl Environ Microbiol, 66: 2641–2646, 2000.

15.

Falagas

, Rafailidis

, and Matthaiou

. Resistance to polymyxins: mechanisms, frequency and treatment options. Drug Resist Updat, 13: 132–138, 2010.

16.

Fernandez

, Gooderham

, Bains

, McPhee

, Wiegand

, and Hancock

. Adaptive resistance to the “last hope” antibiotics polymyxin B and colistin in Pseudomonas aeruginosa is mediated by the novel two-component regulatory system ParR-ParS. Antimicrob Agents Chemother, 54: 3372–3382, 2010.

17.

Helander

, Kilpelainen

, and Vaara

. Increased substitution of phosphate groups in lipopolysaccharides and lipid A of the polymyxin-resistant pmrA mutants of Salmonella typhimurium: a 31P-NMR study. Mol Microbiol, 11: 481–487, 1994.

18.

Kelner

, Bagnell

, and Welch

. Thioureas react with superoxide radicals to yield a sulfhydryl compound. Explanation for protective effect against paraquat. J Biol Chem, 265: 1306–1311, 1990.

19.

Kim

, Chung

, and Cho

. Differential expression of the major catalase, KatA in the two wild type Pseudomonas aeruginosa strains, PAO1 and PA14. J Microbiol, 57: 704–710, 2019.

20.

Kim

, Park

, Heo

, and Cho

. Drosophila melanogaster-based screening for multihost virulence factors of Pseudomonas aeruginosa PA14 and identification of a virulence-attenuating factor, HudA. Infect Immun, 76: 4152–4162, 2008.

21.

Kohanski

, Dwyer

, Hayete

, Lawrence

, and Collins

. A common mechanism of cellular death induced by bactericidal antibiotics. Cell, 130: 797–810, 2007.

22.

Lee

, Hsu

, Turk

, and Groisman

. The PmrA-regulated pmrC gene mediates phosphoethanolamine modification of lipid A and polymyxin resistance in Salmonella enterica . J Bacteriol, 186: 4124–4133, 2004.

23.

Lee

, Heo

, Lee

, and Cho

. KatA, the major catalase, is critical for osmoprotection and virulence in Pseudomonas aeruginosa PA14. Infect Immun, 73: 4399–4403, 2005.

24.

Lee

and Ko

. Mutations and expression of PmrAB and PhoPQ related with colistin resistance in Pseudomonas aeruginosa clinical isolates. Diagn Microbiol Infect Dis, 78: 271–276, 2014.

25.

, Nation

, Turnidge

, Milne

, Coulthard

, Rayner

, and Paterson

. Colistin: the re-emerging antibiotic for multidrug-resistant Gram-negative bacterial infections. Lancet Infect Dis, 6: 589–601, 2006.

26.

Lima

, Ferreira

, Nunes Neto

, Monteiro

, Santos

ARC

, Tavares

, Denadai

AML

, Bomfim

MRQ

, Dos Santos

, Marques

, and de Souza Monteiro

. Evaluation of the interaction between polymyxin B and Pseudomonas aeruginosa biofilm and planktonic cells: reactive oxygen species induction and zeta potential. BMC Microbiol, 19: 115, 2019.

27.

Line

, Alhede

, Kolpen

, Kuhl

, Ciofu

, Bjarnsholt

, Moser

, Toyofuku

, Nomura

, Hoiby

, and Jensen

. Physiological levels of nitrate support anoxic growth by denitrification of Pseudomonas aeruginosa at growth rates reported in cystic fibrosis lungs and sputum. Front Microbiol, 5: 554, 2014.

28.

Liu

and Imlay

. Cell death from antibiotics without the involvement of reactive oxygen species. Science, 339: 1210–1213, 2013.

29.

Masuda

, Sakagawa

, and Ohya

. Outer membrane proteins responsible for multiple drug resistance in Pseudomonas aeruginosa . Antimicrob Agents Chemother, 39: 645–649, 1995.

30.

McPhee

, Lewenza

, and Hancock

. Cationic antimicrobial peptides activate a two-component regulatory system, PmrA-PmrB, that regulates resistance to polymyxin B and cationic antimicrobial peptides in Pseudomonas aeruginosa . Mol Microbiol, 50: 205–217, 2003.

31.

Moskowitz

, Brannon

, Dasgupta

, Pier

, Sgambati

, Miller

, Selgrade

, Miller

, Denton

, Conway

, Johansen

, and Hoiby

. PmrB mutations promote polymyxin resistance of Pseudomonas aeruginosa isolated from colistin-treated cystic fibrosis patients. Antimicrob Agents Chemother, 56: 1019–1030, 2012.

32.

Moskowitz

, Ernst

, and Miller

. PmrAB, a two-component regulatory system of Pseudomonas aeruginosa that modulates resistance to cationic antimicrobial peptides and addition of aminoarabinose to lipid A. J Bacteriol, 186: 575–579, 2004.

33.

Murray

, Kwon

, Marcotte

, and Whiteley

. Intrinsic antimicrobial resistance determinants in the Superbug Pseudomonas aeruginosa . mBio, 6: e01603-15, 2015.

34.

Nelson

, Eiras

, Gomez-Simmonds

, Loo

, Satlin

, Jenkins

, Whittier

, Calfee

, Furuya

, and Kubin

. Clinical outcomes associated with polymyxin B dose in patients with bloodstream infections due to carbapenem-resistant Gram-negative rods. Antimicrob Agents Chemother, 59: 7000–7006, 2015.

35.

Okamoto

, Gotoh

, and Nishino

. Pseudomonas aeruginosa reveals high intrinsic resistance to penem antibiotics: penem resistance mechanisms and their interplay. Antimicrob Agents Chemother, 45: 1964–1971, 2001.

36.

Sampson

, Liu

, Schroeder

, Kraft

, Burd

, and Weiss

. Rapid killing of Acinetobacter baumannii by polymyxins is mediated by a hydroxyl radical death pathway. Antimicrob Agents Chemother, 56: 5642–5649, 2012.

37.

Schurek

, Sampaio

, Kiffer

, Sinto

, Mendes

, and Hancock

. Involvement of pmrAB and phoPQ in polymyxin B adaptation and inducible resistance in non-cystic fibrosis clinical isolates of Pseudomonas aeruginosa . Antimicrob Agents Chemother, 53: 4345–4351, 2009.

38.

Teuber

. Phosphorylation of methyl-alpha-D-glucopyranoside in polymyxin B-treated. Salmonella typhimurium J Bacteriol, 100: 1417–1419, 1969.

39.

Trunk

, Benkert

, Quack

, Munch

, Scheer

, Garbe

, Jansch

, Trost

, Wehland

, Buer

, Jahn

, Schobert

, and Jahn

. Anaerobic adaptation in Pseudomonas aeruginosa: definition of the Anr and Dnr regulons. Environ Microbiol, 12: 1719–1733, 2010.

40.

Wolfgang

, Kulasekara

, Liang

, Boyd

, Wu

, Yang

, Miyada

, and Lory

. Conservation of genome content and virulence determinants among clinical and environmental isolates of Pseudomonas aeruginosa . Proc Natl Acad Sci U S A, 100: 8484–8489, 2003.

41.

Yoon

, Hennigan

, Hilliard

, Ochsner

, Parvatiyar

, Kamani

, Allen

, DeKievit

, Gardner

, Schwab

, Rowe

, Iglewski

, McDermott

, Mason

, Wozniak

, Hancock

, Parsek

, Noah

, Boucher

, and Hassett

. Pseudomonas aeruginosa anaerobic respiration in biofilms: relationships to cystic fibrosis pathogenesis. Dev Cell, 3: 593–603, 2002.

42.

Young

, Bains

, Bell

, and Hancock

. Role of Pseudomonas aeruginosa outer membrane protein OprH in polymyxin and gentamicin resistance: isolation of an OprH-deficient mutant by gene replacement techniques. Antimicrob Agents Chemother, 36: 2566–2568, 1992.

43.

Zimmermann

, Reimmann

, Galimand

, and Haas

. Anaerobic growth and cyanide synthesis of Pseudomonas aeruginosa depend on anr, a regulatory gene homologous with fnr of Escherichia coli . Mol Microbiol, 5: 1483–1490, 1991.

44.

Zumft

. Cell biology and molecular basis of denitrification. Microbiol Mol Biol Rev, 61: 533–616, 1997.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.09 MB

0.07 MB

0.08 MB

Nitrate Respiration Promotes Polymyxin B Resistance in Pseudomonas aeruginosa

Abstract

Aims:

Results:

Innovation and Conclusion:

Introduction

Innovation

Results

PAO1, but not PA14, displays intrinsic PMB resistance

The intrinsic PMB resistance of PAO1 involves NR

NR-associated PMB resistance is independent of KatA

Intrinsic PMB resistance compromises the PMB treatment in infection conditions

NR correlates with the intrinsic PMB resistance in PA clinical strains

Discussion

Materials and Methods

Bacterial strains and culture conditions

Oligonucleotide primers

Generation of deletion mutants

Measurement of MICs

Transcriptome analysis

Construction of lacZ fusions and β-galactosidase assay

Real-time PCR

Measurement of NO2 − accumulation

PMB-mediated growth inhibition

Microscopic observation of cell death

Measurement of H2O2 accumulation

Determination of antibacterial efficacy in vivo

Authors' Contributions

Footnotes

Author Disclosure Statement

Funding Information

Supplementary Material

Abbreviations Used

References

Supplementary Material

Measurement of NO₂ ⁻ accumulation

Measurement of H₂O₂ accumulation