Phenazine-1 carboxylic acid of Pseudomonas aeruginosa induces the expression of Staphylococcus aureus Tet38 MDR efflux pump and mediates resistance to phenazines and antibiotics

ABSTRACT

In this study, we showed that phenazine-1 carboxylic acid (PCA) of Pseudomonas aeruginosa induced the expression of Tet38 efflux pump triggering Staphylococcus aureus resistance to tetracycline and phenazines. Exposure of S. aureus RN6390 to supernatants of P. aeruginosa PA14 and its pyocyanin (PYO)-deficient mutants showed that P. aeruginosa non-PYO phenazines could induce the expression of Tet38 efflux pump. Direct exposure of RN6390 to PCA compound at 0.25× MIC led to a five-fold increase in tet38 transcripts. Expression of Tet38 protein was identified through confocal microscopy using RN6390(pRN-tet38p-yfp) that expressed YFP under control of the tet38 promoter by PCA at 0.25× MIC. The MICs of PCA of a Tet38-overexpressor and a Δtet38 mutant showed a three-fold increase and a two-fold decrease, respectively, compared with that of wild-type. Pre-exposure of RN6390 to PCA (0.25× MIC) for 1 hour prior to addition of tetracycline (1× or 10× MIC) improved bacteria viability of 1.5-fold and 2.6-fold, respectively, but addition of NaCl 7% together with tetracycline at 10× MIC reduced the number of viable PCA-exposed RN6390 of a 2.0-log₁₀ CFU/mL. The transcript levels of tetR21, a repressor of tet38, decreased and increased two-fold in the presence of PCA and NaCl, respectively, suggesting that the effects of PCA and NaCl on tet38 production occurred through TetR21 expression. These data suggest that PCA-induced Tet38 protects S. aureus against tetracycline during coinfection with P. aeruginosa; however, induced tet38-mediated S. aureus resistance to tetracycline is reversed by NaCl 7%, a nebulized treatment used to enhance sputum mobilization in CF patients.

INTRODUCTION

Staphylococcus aureus is a common commensal bacterium that regularly colonizes human skin, nares, and the gastrointestinal tract. As a result, colonization can lead to an increase in the risk of S. aureus infections that range from minor skin infections to life-threatening conditions, such as endocarditis, osteomyelitis, and sepsis (1). The emergence of multidrug-resistant (MDR) strains, such as methicillin-resistant S. aureus (MRSA), necessitates complex treatment regimens that have challenged healthcare providers attempting to eradicate S. aureus infections. Decolonization of S. aureus to reduce infections using topical antibacterial drugs such as nasal mupirocin and chlorhexidine bathing, has been utilized in intensive care units to reduce MRSA infections and bloodstream infections, however, mupirocin resistance and recolonization after treatment are challenging problems that are encountered when using this approach (2–4).

Pseudomonas aeruginosa is a ubiquitous environmental bacterium, commonly found in water, soil, and damp areas. It can, however, cause serious infections in individuals with underlying lung conditions, such as bronchiectasis and cystic fibrosis (CF), as P. aeruginosa can persist for an extended period of time following the initial infection (5). P. aeruginosa secretes many virulence factors, including the 2-heptyl-4-hydroxyquinoline-N-oxide which affects the ATP production of S. aureus (6), pyoverdine and pyochelin which are siderophores that drive S. aureus into fermentation (7), and the quorum sensing-regulated virulence factor pyocyanin (PYO), a major phenazine of P. aeruginosa (8). Many bacteria, including P. aeruginosa, use the highly conserved shikimic acid pathway to biosynthesize phenazines using chorismic acid as the branch point once the phenazine genes phzABCDEFG are expressed. Phenazine-1 carboxylic acid (PCA) is synthesized from chorismate and is also called tubermycin B due to its antimicrobial activity against M. tuberculosis (9). PCA is converted into three different phenazines under the actions of the products of three biosynthetic genes phzH, phzM, and phzS. PhzH converts PCA into phenazine 1-carboxamide (PCN), PhzM converts PCA into 5-methylphenazine-1-carboxylate (5-Me-PCA), PhzS converts PCA into 1-hydroxyphenazine (1-OH-PHZ), and PhzS also converts 5-Me-PCA into PYO. The phenazines PCA, PCN, 1-OH-PHZ, and PYO all have antimicrobial activity (8, 10).

S. aureus and P. aeruginosa are two of the most common human pathogens that are isolated from the sputum of CF patients as well as from infected wounds of diabetic foot ulcers, burns, and surgical wounds (11–13). CF patients characteristically have a thick mucus layer (sputum) coating the inside of the airways that supports the habitat of various bacteria. Furthermore, CF patients are often repetitively exposed to antibiotics, which can cause an imbalance in the natural composition of bacteria and eventually lead to the emergence of antibiotic resistance. Hypertonic saline (HTS) has been used for symptomatic therapy to improve sputum clearance and facilitate the identification of pathogenic bacteria (14). HTS works as a mucolytic agent and is part of a comprehensive, inexpensive, safe, and effective targeted therapy for CF patients (15).

In the airways of people with CF, S. aureus airway colonization often precedes colonization with P. aeruginosa. Competition between these two pathogens often results in the succession of P. aeruginosa and exclusion of S. aureus in the CF lung (7, 16, 17). Several studies illustrating the interactions between S. aureus and P. aeruginosa demonstrated that coinfection of S. aureus with P. aeruginosa enhances antibiotic resistance and virulence in both pathogens, with P. aeruginosa as the more aggressive member of the pair (18, 19). Recent studies demonstrated that while initially inhibited by P. aeruginosa-secreted products, S. aureus quickly became resistant to these toxins (20, 21). Nevertheless, how S. aureus protects itself against P. aeruginosa killing is still not well understood. S. aureus has the ability to export a large variety of unrelated compounds, including antibiotics, due to its array of efflux pumps and transporters. These native proteins confer a low-level resistance of S. aureus to the corresponding antimicrobial substrates by decreasing the intracellular antibiotic concentration via active efflux of drug compounds (22–24). Among reported S. aureus efflux pumps and transporters, we have focused on clinically important NorA and Tet38 MDR efflux pumps (25) that confer a low-level resistance to fluoroquinolones (ciprofloxacin, norfloxacin) and tetracycline, respectively.

Recently, we have reported that overexpression of the S. aureus NorA efflux pump has the capability of reducing the intracellular accumulation of PYO, resulting in a three-fold increase in the MIC of PYO as well as a four-fold increase in the MIC of ciprofloxacin (26). Furthermore, PYO at a low-level concentration induced a two-fold increase in the norA transcript levels. To identify similar phenomena involving other P. aeruginosa phenazines and other S. aureus efflux pumps, we evaluated the effects of PCA, PCN, 1-OH-PHZ, and PMS (commercial compound equivalent with P. aeruginosa 5-Me-PCA) on S. aureus RN6390 survival and resistance to phenazines and the effects of P. aeruginosa culture supernatants on efflux pump expression. We discovered that the S. aureus Tet38 efflux pump was induced by PCA at sub-MIC concentrations and caused resistance to phenazines PCA, PCN, and 1-OH-PHZ but not to PYO or PMS. We also demonstrated that Tet38 was important for S. aureus survival in the presence of PCA alone or in combination with tetracycline. Our findings indicate that coexistence of P. aeruginosa with S. aureus mediates staphylococcal resistance to antibiotics by overproducing phenazine-induced efflux pumps, and this phenomenon contributes to the survival of S. aureus when exposed to P. aeruginosa exoproducts and pump-substrate antibiotics. Interestingly, we also found that hypertonic saline, a common therapy that enhances clearance of airway mucus, may additionally affect expression of S. aureus Tet38 and reduce resistance induction by PCA.

RESULTS

Phenazines of PYO-deficient mutants induce SA efflux pump transcript levels

P. aeruginosa PYO-deficient mutants PA14_phzM and PA14_phzS produce phenazines PCA and PCN. In addition, mutant PA14_phzM produces 1-OH-PHZ, and mutant PA14_phzS produces 5-Me-PCA (8). The compound PMS (phenazine methosulfate) is a commercial product equivalent to P. aeruginosa 5-Me-PCA and was previously demonstrated to be a substrate of the P. aeruginosa efflux pump MexGHI-OpmD (27). We included PMS in this study to determine if this phenazine could also serve as an inducer or substrate of S. aureus efflux pumps.

Exposure to culture supernatants of wild-type P. aeruginosa PA14 resulted in increased transcript levels of S. aureus RN6390 efflux pumps norA, norB, and tet38 of 2.3-, 3.1-, and 2.5-fold, respectively. Exposure of RN6390 to culture supernatants of PA14_phzM and PA14_phzS showed no increase in the norA transcript levels, as expected because of the lack of PYO synthesis in these strains. In contrast, culture supernatants of PA14_phzM caused two- and five-fold increases in norB and tet38 transcript levels, respectively, and culture supernatants of PA14_phzS caused a 2.6-fold increase in tet38 transcript levels (Fig. 1A).

Fig 1.

Induction of S. aureus efflux pump gene transcript levels by P. aeruginosa supernatants and phenazines at 0.5 x MIC. (A) RN6390 was exposed to the supernatants of PA14, PA14phzM, and PA14phzS (at 30%) for 1 hour. Then, quantitative real-time RT-PCR assays were performed to assess the level of norA, norB, and tet38 transcripts. The relative transcript level of pump gene was expressed as the fold change (FC) in pump gene transcripts of bacteria exposed versus non-exposed to P. aeruginosa supernatants. The assays were repeated three times with three different biological samples. The error bars represent the means of FC ±SEM for each assay. After 1 hour induction, the differences (*) in the FC of norA and norB of RN6390 exposed to PA14 supernatant versus RN6390 exposed to PA14phzM and PA14phzS supernatants were statistically significant as determined by a one-way ANOVA test (P < 0.05). The differences (**) in the FC of tet38 of RN6390 exposed to PA14phzM supernatant versus RN6390 exposed to PA14 and PA14phzS supernatants were statistically significant as determined by a one-way ANOVA test (P < 0.05). (B) RN6390 was exposed to 0.5x MIC of phenazines PCA, PCN, 1-OH-PHZ, and PMS for 1 hour. Then, quantitative real-time RT-PCR assays were performed to assess the level of norA, norB, and tet38 transcripts. The relative transcript level of pump gene was expressed as the fold change (FC) in pump gene transcripts of bacteria exposed versus non-exposed to phenazines. The assays were repeated three times with three different biological samples. The error bars represent the means of FC ±SEM for each assay. After 1 hour induction, the differences (**) in the FC of tet38 of RN6390 exposed to PCA or to 1-OH-PHZ versus RN6390 exposed to PCN or non-exposed were statistically significant as determined by a one-way ANOVA test (P < 0.05).

To verify that the efflux pump induction phenomenon was not limited to P. aeruginosa PA14 and S. aureus RN6390, we exposed the S. aureus strain Newman to supernatant prepared from the reference P. aeruginosa PAO1. We found that norA, norB, and tet38 transcript levels of S. aureus Newman increased 3.0-, 2.9-, and 3.2-fold, respectively. In addition, we also exposed RN6390 and Newman to supernatants prepared from previously reported clinical P. aeruginosa isolated alone in sputum samples of cystic fibrosis patients (CF-PA-1 and CF-PA-2) (26). We found that exposure of RN6390 to supernatants of the two clinical P. aeruginosa also led to similar increase in the transcript levels of norA (2.8- and 3.1-fold), norB (2.3- and 2.2-fold), and tet38 (3.8- and 3.1-fold). Exposure of Newman to clinical P. aeruginosa supernatants yielded similar pump transcript levels, norA (2.1- and 2.5-fold), norB (2.8- and 3.2-fold), and tet38 (both at 3.1-fold). Further investigations are underway to assess the interactions between secreted products of P. aeruginosa and efflux pumps of S. aureus.

Since P. aeruginosa PA14 and its mutants produce phenazines as well as various other exoproducts, we repeated the exposure assays using commercially purified PCA, PCN, 1-OH-PHZ, and PMS compounds. Table 1 shows the MIC of phenazines of RN6390. Compared with the norA, norB, and tet38 transcript levels of RN6390 without exposure to phenazines, we found a 10-fold decrease in the norA and norB transcripts when RN6390 was exposed to PCA, PCN, and 1-OH-PHZ at 0.5xMIC. In contrast, tet38 transcripts increased 5.2-fold, 1.6-fold, and 6.8-fold, respectively, when exposed to PCA, PCN, and 1-OH-PHZ at 0.5x MIC. No significant change in norA, norB, and tet38 transcript levels (FC ~1.16) occurred when exposed to PMS at 0.5x MIC compared with that of RN6390 without exposure (Fig. 1B).

TABLE 1.

MICs of phenazines and antibiotics of S. aureus strains^c

	MIC (μg/mL)
	PCA	PCN	1-OH-PHZ	PMS	PYO	TETRA	DOXY
S. aureus
RN6390	40	20	20	20	6	0.25	0.25
YW22a	20	20	20	20	6	0.125	0.125
RN6390(pL150)	40	20	20	20	6	0.25	0.25
RN6390(pL150-tet38)	120	40	40	20	6	0.5	0.5
RN6390(pRN-tet38p-yfp)	40	20	20	20	6	0.125	0.125
RN6390 + PCAb						0.5	0.5
RN6390 + NaCl 7%						0.125	0.125
RN6390 + PCAb + NaCl 7%						0.25	0.25

^a YW22: Δtet38 mutant.

^b PCA at 0.25x MIC (10 µg/mL) was added to RN6390 culture with or without NaCl.

^c PCA, phenazine-1 carboxylic acid; PCN, phenazine 1-carboxamide; 1-OH-PHZ, 1-hydroxyphenazine; PMS, phenazine methosulfate; PYO, pyocyanin; TETRA, tetracycline; DOXY, doxycycline.

PCA induced the expression of the Tet38 protein

To observe the effect of phenazines on the expression of Tet38 protein, we created the strain RN6390(pRN-tet38p-yfp), which carries a plasmid construct with the tet38 promoter fused to the yfp reporter gene. As a viability control, we used strain RN6390(pRN-sarAp-yfp) that constitutively expresses the fluorescent protein YFP. We exposed RN6390(pRN-tet38p-yfp) at OD₆₀₀ = 0.5 for 1 hour to PCA, PCN, and 1-OH-PHZ at concentrations 0.5 x MIC (10, 20; 10 µg/mL, respectively) and analyzed Tet38 reporter expression using confocal imaging.

We observed S. aureus cells (green cells) in four contiguous confocally imaged fields of view per well, each well had an average initial number of 1 × 10⁴ bacteria. The fluorescence positive control was represented by the YFP-constitutively expressed RN6390(pRN-sarAp-yfp), and the negative control of the assay was represented by RN6390(pRN-tet38p-yfp) without exposure to phenazine compound (no induction). The images shown in Fig. 2 are representative of multiple fields of view taken for each assay. Each representative field of view was imaged and presented in panels A and B of Fig. 2.

Fig 2.

Confocal imaging of S. aureus RN6390 exposed to phenazines at 0.5 x MIC and PCA at different sub-MIC concentrations. S. aureus RN6390 (10⁴ bacteria/well) carrying plasmid construct (pRN-tet38p-yfp) (A, expressing YFP by induction of tet38 promoter or spontaneous expression in the absence of induction) were grown in TSB media supplemented with chloramphenicol at 10  µg/mL. Bacteria were exposed to PCA, PCN, and 1-OH-PHZ at 0.5 x MIC (20, 10, and 10  µg/mL, respectively) and then were imaged after 4 hours. Four fields of view were selected per well and all fields of view of each condition showed similar Differential Interference Contrast (DIC) (B). Shown are representative fields of view for each assay at 4 hours of exposure. The experiments were done in triplicate with three biological samples. Induction by PCA at 0.5xMIC and PCA at 0.25xMIC showed similar results. No significant induction was found after 4 hours of exposure to PCA at 0.125x MIC. No induction with PCN or with 1-OH-PHZ at 0.5x MIC. RN6390 carrying plasmid (pRN-sarAp-yfp) expressed YFP constitutively and was used as a positive control. Scale bar = 5 µm.

Figure 2A showed the induction of YFP/Tet38 by phenazines (fluorescent images, YFP) and Fig. 2B showed the total number of S. aureus RN6390(pRN-tet38p-yfp) exposed to phenazines (using differential interference contrast, DIC). In comparing panel A to panel B, we estimated the ratio of green S. aureus (A, cells expressing YFP) from the total bacteria (B, DIC) in one field of view. The small number of green cells ( ~10 cells) in the well with no compound represented the spontaneous expression of the YFP/Tet38 bacteria (ratio ~10⁻³, with a total bacteria per well ~10⁴ cells). We found that PCA, but not PCN or 1-OH-PHZ, at 0.5x MIC induced the expression of YFP/Tet38 (Fig. 2A and B, ratio A/B ~1). RN6390(pRN-tet38p-yfp) was further exposed to PCA at 0.25x MIC (10 µg/mL), and 0.125x MIC (5 µg/mL), and the fluorescent signals showed that PCA at 0.25x MIC induced the expression of YFP/Tet38 at a similar level as that of PCA at 0.5x MIC (Fig. 2A and B, ratio A/B ~1). The small number of green cells (average ~13 green cells from observation of 4 field of views) that appeared following exposure to PCA at 0.125 x MIC (ratio A/B ~1.3×10⁻³) was comparable to that of bacteria without induction (Average ~12–14 green cells from observation of 4 field of views) (Fig. 2A).

These findings differed from the real-time qPCR data, which showed a significant increase in tet38 transcripts (6.8-fold) when the bacteria were exposed to 1-OH-PHZ at 0.5xMIC (Fig. 1B). The reason for these differences is not known and further investigations are underway to determine if the problems occurred at the transcription or translation level.

PCA-induced Tet38 protects S. aureus RN6390 against tetracycline

To evaluate the tet38 induction effect of PCA on the survival of S. aureus RN6390 when exposed to tetracycline, we carried out two series of assays in parallel using RN6390(pRN-tet38p-yfp) exposed to PCA at 0.25x MIC ±tetracycline at 1x MIC or 10 x MIC (Table 1). The plasmid construct pRN-tet38p-yfp did not modify the MICs of phenazines or tetracycline and doxycycline in RN6390. In the first series, PCA was added to the bacterial culture simultaneously with tetracycline, and in the second series, PCA was added for 1 hour prior to addition of tetracycline, allowing induction prior to tetracycline exposure. All assays were carried out for a period of 0–4 hours followed by confocal imaging. In parallel, as a viability control, RN6390(pRN-sarAp-yfp) with YFP constitutively expressed, was treated with PCA and tetracycline in the same manner and was analyzed by confocal imaging. The images in Fig. 3 are representative of multiple fields of view taken for each assay.

Fig 3.

Confocal imaging of S. aureus RN6390 exposed to phenazines at 0.25x MIC with and without tetracycline at 1x MIC and 10x MIC. S. aureus RN6390 (10⁴ bacteria/well) carrying plasmid construct (pRN-tet38p-yfp) (A, expressing YFP by induction of tet38) or plasmid construct (pRN-sarAp-yfp) (B, viability control, expressing YFP constitutively) were grown in TSB media supplemented with chloramphenicol at 10  µg/mL. Bacteria with either plasmid constructs were exposed to PCA at 0.25x MIC with and without tetracycline at 1x MIC or 10x MIC for a total of 4 hours. Four fields of view were selected per well and all fields of view of each condition showed similar observations. Shown are representative fields of view for each assay at 4 hours of exposure. The experiments were done in triplicate with three biological samples. RN6390 carrying plasmid (pRN-sarAp-yfp) was used as a positive control of YFP expression (green cells) and RN6390(pRN-tet38p-yfp) without induction (no induction) was used as a negative control. Scale bar = 5 µm.YFP, Fluorescent signal; DIC, Differential Interference Contrast*PCA +Tetra1x MIC, PCA at 0.25x MIC was added together with tetracycline at 1x MIC to the bacterial culture. *PCA +Tetra10x MIC, PCA at 0.25x MIC was added together with tetracycline at 10x MIC to the bacterial culture. **PCA/Tetra1x MIC, PCA at 0.25x MIC added for 1 hour then tetracycline at 1x MIC was added to the bacterial culture. **PCA/Tetra10x MIC, PCA at 0.25x MIC added for 1 hour then tetracycline at 10x MIC was added to the bacterial culture.

Starting from 10⁴ bacteria per well of RN6390(pRN-tet38p-yfp) and RN6390(pRN-sarAp-yfp) and 4 hours of drug exposure, we found that exposure to PCA at 0.25x MIC did not significantly alter the viability of both bacteria when compared with the no induction condition (Fig. 3A and B). When PCA was added to the cultures for 4 hours, we counted an average of 160 green cells RN6390(pRN-tet38p-yfp) (Fig. 3A, YFP) and an average of 162 green cells of RN6390(pRN-sarAp-yfp) (Fig. 3B) from 4 field of views (ratio induction/viability ~0.99).

Compared with bacteria grown without exposure, addition of tetracycline at 1x MIC (0.25 µg/mL) decreased the number of RN6390(pRN-tet38p-yfp) to ~30 cells and RN6390(pRN-sarAp-yfp) to ~33 cells (average of 4 field of views) (ratio ~0.91). This finding suggested that tetracycline at 1x MIC had a limited induction effect on the expression of YFP/Tet38 (Fig. 3A, YFP), a phenomenon that was previously reported in our study of Tet38 efflux pump (28). Addition of tetracycline at 10x MIC (2.5 µg/mL) decreased the number of RN6390(pRN-tet38p-yfp) to ~14 cells and RN6390(pRN-sarAp-yfp) to ~13 cells (average of 4 field of views) (ratio ~1.1).

Exposure of bacteria to a mix of PCA and tetracycline (1x MIC or 10x MIC) reduced the number of RN6390(pRN-tet38p-yfp) to ~35 and 16 cells, and RN6390(pRN-sarAp-yfp) to ~40 and 20 cells, respectively (average of 4 field of views). The ratios were 0.89 and 0.8, respectively for tetracycline 1x MIC and 10x MIC (Fig. 3A and B, YFP).

In contrast, when the bacteria were first exposed to PCA for 1 hour then exposed to tetracycline at 1x or 10x MIC for an additional 3 hours, the number of YFP-induced green cells was significantly higher than that of concurrent tetracycline +PCA for both concentrations of tetracycline (average ~162 cells for 1x MIC and 125 cells for 10x MIC) (Fig. 3A, YFP). Similar observations were found with RN6390(pRN-sarAp-yfp) with ~180 cells for 1x MIC and 120 cells for 10x MIC (Fig. 3B, YFP). Ratios were 0.9 and 1.04, respectively for tetracycline 1x and 10x MIC. The DIC images were added to Fig. 3 to show the total bacteria (alive and dead) for each field of view. We performed colony count for each condition to evaluate the CFU/mL of survivors (described in the next section).

With a ratio of RN6390(pRN-tet38p-yfp) for induction vs RN6390(pRN-sarAp-yfp) for viability varying between 0.8–1.0, these data suggested that survivor RN6390 cells were those that overexpressed Tet38. These data suggested that PCA-induced Tet38 improved S. aureus survival in the presence of tetracycline especially when induction occurred prior to addition of tetracycline.

To confirm the findings of confocal imaging, we also studied S. aureus survival on exposure to PCA and tetracycline. RN6390 with or without exposure to PCA at 0.25x MIC showed similar growth curves over a period of 24 hours (Fig. 4A). Compared with RN6390 grown with or without exposure to PCA at 0.25x MIC, exposure to tetracycline at 1x MIC or to the combination (PCA +tetracycline) led to a reduction of 3.0-log₁₀ CFU/mL after 24 hours of RN6390. In contrast, pre-exposure to PCA for 1 hour led to a reduction of 2.0-log₁₀ CFU/mL of viable cells after exposure to tetracycline for the same period (Fig. 4A). Thus, pre-exposure to PCA increased the number of RN6390 survivors of tetracycline (1x MIC) by 1.5-fold.

Fig 4.

S. aureus time-kill curves following exposure to PCA at 0.25x MIC with and without tetracycline at 1x MIC and 10x MIC. S. aureus RN6390 and YW22 (OD₆₀₀ ~0.5, 10⁷ CFU/mL) were exposed to PCA at 0.25x MIC with and without tetracycline at 1x and 10x MIC for a period of 24 hours. The experiments were repeated three times with three biological samples for each antibiotic. The figure represents the log₁₀ of S. aureus CFU/mL with or without antibiotic exposure (error bars = log₁₀CFU/mL ±SD for each condition). YW22, Δtet38 mutant. PCA/+Tetra 1xMIC: PCA at 0.25 x MIC added for 1 hour then tetracycline at 1x MIC was added to the bacterial culture. PCA +Tetra 1xMIC: PCA at 0.25x MIC was added together with tetracycline at 1x MIC to the bacterial culture. PCA/+Tetra 10x MIC: PCA at 0.25x MIC added for 1 hour then tetracycline at 10x MIC was added to the bacterial culture. PCA +Tetra 10xMIC: PCA at 0.25x MIC was added together with tetracycline at 10 x MIC to the bacterial culture. (A) S. aureus RN6390 kill curves following exposure to PCA at 0.25x MIC ±tetracycline at 1x for a period of 24 hours. (B) S. aureus RN6390 kill curves following exposure to PCA at 0.25x MIC ±tetracycline at 10x MIC for a period of 24 hours. (C) Δtet38 mutant kill curves following exposure to PCA at 0.25x MIC with and without tetracycline at 1x MIC for a period of 24 hours. (D) Δtet38 mutant kill curves following exposure to PCA at 0.25x MIC with and without tetracycline at 10x MIC for a period of 24 hours.

Exposure to tetracycline at 10x MIC or to concurrent PCA and tetracycline led to a reduction of 4.6-log₁₀ CFU/mL of RN6390, while a pre-exposure to PCA for 1 hour led to a reduction of 2.0-log₁₀ CFU/mL of viable cells after exposure to tetracycline (Fig. 4B). Thus, pre-exposure to PCA increased the number of RN6390 tetracycline (10x MIC) survivors by 2.3-fold.

Absence of Tet38 increases S. aureus susceptibility to phenazines and tetracycline

To confirm the role of Tet38 in S. aureus resistance to phenazines and tetracycline, we determined the MICs of PCA, PCN, 1-OH-PHZ, PYO, tetracycline, and doxycycline of RN6390 and its Δtet38 mutant (YW22) and a Tet38-overexpressor RN6390(pLI50-tet38) (Table 1). We found that the Tet38-overexpressor exhibited an increase of three-fold in the MICs of PCA, an increase of two-fold in the MIC of PCN and 1-OH-PHZ, and no change in the MICs of PMS and PYO when compared with the MICs of RN6390. In contrast, the tet38 mutant YW22 showed a decrease of two-fold in the MICs of PCA and no change in the MICs of PCN, 1-OH-PHZ, PMS, and PYO when compared with the MICs of RN6390. As a control of the Tet38 activity, we determined the MICs of tetracycline and doxycycline. As expected, Tet38-overexpressor showed an increase of four-fold in the MICs of tetracycline and doxycycline, while YW22 showed a decrease of two-fold in the MICs of both antibiotics (Table 1) when compared with the MICs of RN6390 transformed with the empty vector pLI50.

We next performed growth curves of Δtet38 mutant YW22 in the same manner, exposed or non-exposed to PCA at 0.25x MIC, exposed or non-exposed to PCA in the presence of tetracycline at 1x MIC and 10 x MIC for a period of 0–24 hours, and carried out colony counts to determine the S. aureus CFU/mL. Compared with assays done with RN6390, exposure of YW22 to PCA (MIC_PCA of YW22 = 20 µg/mL; 0.25x MIC = 5 µg/mL) led to a reduction in viable cells of 0.5-log₁₀. A 24 hour exposure of YW22 to tetracycline at 1x MIC (0.125 µg/mL) yielded a reduction in viable cells of 3.0-log₁₀ CFU/mL.

Exposure of YW22 to the combination of PCA at 0.25x MIC with tetracycline at 1x MIC, added immediately or successively, yielded similar reduction in viable bacterial cells (~ 3.0-log₁₀ CFU/mL for 24 hours) (Fig. 4C). The same assays carried out with PCA and tetracycline at 10x MIC (0.6 µg/mL) showed similar results with a reduction of viable cells approximately 0.5-log₁₀ CFU/mL for exposure to PCA alone and 5.5-log₁₀ CFU/mL for exposure to PCA in combination with tetracycline for a period of 24 hours (Fig. 4D). These data showed that S. aureus protection against PCA and tetracycline depended on the presence of an intact and functional Tet38.

Hypertonic NaCl at 7% decreased the number of PCA-induced Tet38-overexpressor S. aureus in the presence of tetracycline

We have shown previously that high concentrations of NaCl ( ≥1.2M, 7%) resulted in reduced expression of Tet38, which also appears to be able to import sodium in exchange for tetracycline export, resulting in protection of S. aureus from salt stress when Tet38 expression is decreased (29). In this study, we assessed the effect of NaCl 7% on the susceptibility to tetracycline of PCA-pre-exposed S. aureus. We selected tetracycline at 10x MIC based on our kill curve data (Fig. 4B). We first exposed RN6390 to PCA (0.25x MIC) for one hour and then added tetracycline (10x MIC) with and without NaCl for 24 hours. RN6390 grew in similar manner in TSB media with and without PCA and supplemented or not with NaCl (Fig. 5A). When compared with RN6390 grown without antibiotic, RN6390 with and without concurrent PCA and tetracycline showed a reduction of 6.0-log₁₀ CFU/mL (from 10.5-log₁₀ to 4.5-log₁₀ CFU/mL). When RN6390 was pre-exposed to PCA for 1 hour then tetracycline was added, the reduction in viable cells was 3.0-log₁₀ CFU/mL (from 10.5-log₁₀ to 7.5-log₁₀ CFU/mL). These data confirmed our kill curve data reported in Fig. 4, with survivors RN6390 pre-exposed to PCA increased two-fold. In contrast, addition of (NaCl 7% + tetracycline) to PCA-pre-exposed RN6390 and compared that with (tetracycline +PCA-pre-exposed RN6390), we found the number of survivors RN6390 decreased 2.2-log₁₀ CFU/mL (from 7.5-log₁₀ to 5.3-log₁₀ CFU/mL) (Fig. 5A). These data suggested that hypertonic saline enhanced tetracycline activity in killing RN6390 pre-exposed to PCA.

Fig 5.

S. aureus RN6390 and Δtet38 mutant YW22 exposed to PCA at 0.25x MIC, NaCl at 7%, and tetracycline at 10x MIC for 24 hours. tet38 and tetR21 transcript levels after exposure to PCA, NaCl, and tetracycline.S. aureus RN6390 and YW22 (OD₆₀₀ ~0.5, 10⁷ CFU/mL) were exposed to PCA at 0.25x MIC, NaCl 7%, and tetracycline 10x MIC, alone or in combination, for a period of 24 hours. The experiments were repeated three times with three biological samples for each antibiotic. The figure (A and B) represents the log₁₀ of S. aureus CFU/mL with or without exposure to PCA, NaCl, or tetracycline (error bars = log₁₀CFU/mL ±SD for each condition). (C) RN6390 exposed to PCA, NaCl, or both for 1 hour then quantitative real-time RT-PCR assays were performed to assess the levels of tetR21 and tet38 transcripts. The relative gene transcript level was expressed as the fold change (FC) in gene transcripts of RN6390 exposed versus non-exposed to compounds. The assays were repeated three times with three different biological samples. The error bars represent the means of FC ± SEM for each assay. The differences in the FC of tetR21 and tet38 of RN6390 exposed to PCA or to NaCl, or to PCA + NaCl versus RN6390 non-exposed were statistically significant as determined by a one-way ANOVA with t test (P < 0.05). RN, RN6390; YW22, Δtet38 mutant. PCA, PCA at 0.25x MIC; NaCl, NaCl at 7% (1.2 M); Tetra, tetracycline at 10 x MIC. +PCA; +NaCl; +Tetra: S. aureus exposed to compounds added alone. +PCA + NaCl; +PCA + Tetra: S. aureus exposed to compounds added together. +PCA/+Tetra: S. aureus exposed to PCA for 1 hour then tetracycline was added to the culture. +PCA/+NaCl + Tetra: S. aureus exposed to PCA for 1 hour then NaCl and tetracycline were added together to the culture. (A) RN (RN6390) exposed to PCA, NaCl, and tetracycline for 24 hours. (B) YW22 (Δtet38 mutant) exposed to PCA, NaCl, and tetracycline for 24 hours. (C) Relative tetR21 and tet38 transcripts under induction by PCA and NaCl.

To determine if Tet38 played a central role in the PCA-NaCl-Tetracycline interactions that affected the susceptibility of S. aureus to tetracycline, we repeated the assays using the Δtet38 mutant YW22 (Fig. 5B). We found that YW22 did not demonstrate an effect of saline on survival in the presence of tetracycline after PCA induction, with similar reductions in viable cells of 6.0-log₁₀ CFU/mL seen with exposure to PCA ±NaCl ± tetracycline. Thus, the effects of saline on tetracycline susceptibility after PCA induction are dependent on intact tet38.

PCA at 0.25x MIC and NaCl at 7% exercised opposite effects on tet38 transcription by affecting the expression of the tetR21 regulator

We exposed RN6390 to PCA, NaCl, and PCA + NaCl then observed the changes in tetR21 and tet38 transcript levels in comparison with RN6390 without exposure. TetR21 is a repressor of tet38 expression in S. aureus (28). We found that PCA and NaCl inversely affected the transcription of tetR21, leading to opposite effects on tet38 expression. When RN6390 was exposed to PCA, the relative tetR21 and tet38 transcript levels decreased two-fold and increased 2.19-fold, respectively (Fig. 5C). In contrast, exposure of RN6390 to NaCl led to an increase of 2.18-fold for tetR21 and a decrease of two-fold for tet38. Exposure of RN6390 to PCA and NaCl led to a slight increase of 1.5-fold for tetR21 and tet38 transcripts (Fig. 5C). We carried out MICs to assess the susceptibility to tetracycline and doxycycline of RN6390 after exposure to PCA and NaCl. We found that RN6390 +PCA showed a two-fold increase in tetracycline and doxycycline MICs (from 0.25 to 0.5 µg/mL) while RN6390 in presence of NaCl showed a two-fold decrease in tetracycline and doxycycline MICs (from 0.25 to 0.125 µg/mL). Exposure of RN6390 to (PCA + NaCl) did not change the MICs of either antibiotics (Table 1). Thus, exposure to NaCl 7% counteracts PCA effects on tet38 expression and tetracycline resistance in keeping with modulation of tetR21 expression.

DISCUSSION

Phenazines are a large group of heterocyclic compounds essential for many bacterial metabolic processes, such as iron acquisition, biofilm formation, and signaling events (30). Many microbes including Pseudomonas, Burkholderia, and Streptomyces produce phenazines, and the biosynthesis of these microbial metabolites as well as their respective functions vary among the producing bacteria (30–32). Phenazines are also virulence factors that can cause damage to human host cells. As examples, P. aeruginosa pyocyanin (PYO) causes lung damage due to the activation of reactive oxygen species (ROS), and 1-hydroxyphenazine (1-OH-PHZ) inhibits the beating of the human respiratory cilia (33). Phenazines produced by P. aeruginosa also exhibit high levels of anti-staphylococcal activity, as shown in a study of adolescent CF patients with persistent S. aureus lung infections, wherein later acquisition of P. aeruginosa resulted in the production of phenazines and the eradication of S. aureus (34).

Phenazine-1 carboxylic acid (PCA) is the precursor of other natural phenazines such as phenazine-1 carboxamide (PCN), 1-OH-PHZ, and PYO. PCA also exhibits antifungal activity against human fungal pathogens, such as Trichophyton rubrum (8, 9).

To compete with P. aeruginosa in hostile environments, S. aureus uses transmembrane efflux pumps to defend against anti-staphylococcal exoproducts, including natural phenazines. S. aureus efflux pumps and transporters have varying and sometimes broad substrate profiles to transport metabolites such as salts, vitamins, fatty acids, sugars, amino acids, as well as antibiotics and toxic compounds. Among S. aureus transporters, Tet38 and NorA are known MDR efflux pumps that belong to the major facilitator superfamily (MFS) of transporters with Tet38 having 14 transmembrane segments (TMS), and NorA having 12 TMS (28, 35, 36). Tet38 is a ubiquitous and multi-functional S. aureus MDR efflux pump that confers resistance to tetracycline, chloramphenicol, and antibacterial unsaturated free fatty acids. The Tet38 protein also participates in host cell colonization and internalization by S. aureus (36). We previously reported that sodium selectively reduces tetracycline efflux by Tet38, and under exposure to high concentration of NaCl, S. aureus adjusted its survival in part by decreasing the production of Tet38 protein (29).

We recently reported that S. aureus NorA MDR efflux pump included PYO as a substrate, and when overexpressed, NorA caused a decrease in the cellular accumulation of PYO in S. aureus (26). P. aeruginosa produces an array of phenazines in addition to PYO, suggesting that the ability of some strains of S. aureus to co-colonize with P. aeruginosa in the airways of CF patients or in chronic wounds, may be influenced by other efflux mechanisms for other phenazine compounds.

In this study, we investigated other S. aureus candidate efflux pumps by exposing a wild-type S. aureus strain to supernatants prepared from PYO-deficient mutants of the reference P. aeruginosa PA14. We found that exoproducts other than PYO were able to induce the expression of S. aureus tet38 efflux pump gene. Further exposure of S. aureus wild-type to phenazine compounds showed that PCA, PCN, and 1-OH-PHZ were able to induce the transcription of tet38. Surprisingly, only PCA was able to induce the expression of a reporter of the Tet38 protein at a concentration as low as 0.25x MIC.

To investigate the effects of a PCA-induced Tet38 efflux pump in S. aureus resistance to phenazines and to tetracycline antibiotics, we performed exposure assays using a combination of PCA with tetracycline at 1x MIC and 10x MIC. We found that S. aureus benefited from an exposure to PCA only if S. aureus was first exposed to PCA followed by exposure to tetracycline. The difference in viable cells was most pronounced when S. aureus was successively exposed to PCA at 0.25x MIC and then to tetracycline at 10x MIC in comparison to S. aureus exposure to tetracycline concurrently with PCA (difference ~2.6-log₁₀ CFU/mL). To verify the role of Tet38 in S. aureus survival against PCA and tetracycline, we repeated the exposure assays using a Δtet38 mutant. We found that PCA conferred a beneficial effect on S. aureus viability in the presence of tetracycline and required a functional and intact Tet38 protein. In a recent study by Fu et al., PCA and PYO were reported to act as ligands that bound to the Tet38 regulator TetR21 and controlled the expression of the PYO transporter HprS causing resistance to PYO in S. aureus (37). Induction of Tet38 by PCA could result from either a PCA-mediated decrease in self-regulated TetR21 production or from the loss of binding of TetR21-PCA complex to the tet38 promoter. Utilizing real-time qPCR, we found that exposure to PCA at 0.25x MIC led to a two-fold reduction in tetR21 transcript level in parallel with an increase of two-fold in tet38 transcription. Overexpression of Tet38 via PCA exposure contributed to the increase in the MICs of PCA, PCN, and 1-OH-PHZ, but not to PYO or PMS (a synthetic compound analog to 5-Me-PCA). Thus, S. aureus efflux pumps differ in phenazine substrate specificity, with Tet38 exporting PCA and NorA and HprS exporting PYO (26, 37).

We previously reported that S. aureus overexpressing Tet38 grew poorly under elevated levels of NaCl (≥ 2M; 10%) but was able to grow at lower concentrations of salt (≤ 1.2M; 7%) (29). Furthermore, hypertonic saline at 7% has been used as a mucolytic agent to stabilize and improve lung function in adolescent CF patients (15). We attempted to evaluate the potency of a combination of hypertonic saline and tetracycline at 10x MIC in the killing of S. aureus Tet38-overexpressor through PCA induction. Interestingly, the addition of saline to tetracycline following a pre-incubation of S. aureus with PCA, led to a significant decrease (2.2-log₁₀ CFU/mL) in the number of S. aureus survivors. Furthermore, exposure of RN6390 to NaCl at 7% (1.2 M) increased tetR21 transcript levels 2.2-fold and decreased tet38 transcript levels two-fold. These changes resulted in a two-fold decrease in the MICs of tetracycline and doxycycline of RN6390 (Fig. 5C; Table 1). We previously demonstrated that NaCl at 7% decreased tet38 transcript levels via the KdpD/E two-component regulators (29). In this study, we showed that NaCl at 7% induced the transcription of tetR21 and repressed tet38 in addition to the KdpD/E effect. When PCA and NaCl were added together to a culture of RN6390, the transcripts of tetR21 and tet38 showed a slight increase (~1.5 fold) suggesting a possible interaction between TetR21 and other S. aureus components. Thus, therapeutic benefits of nebulized hypertonic saline treatments may go beyond simple mucolysis and induced expectoration.

We found a two-fold decrease in tetR21 transcript under induction by PCA at 0.25x MIC (10 µg/mL), which was the opposite of study reported by Fu et al. (37). This difference could be due to different strain backgrounds between RN6390 used in this study versus MW2 in study of Fu et al. In this study, we reported an in vitro system using commercial PCA compound and other phenazines to test the corresponding variations in tet38 efflux pump of the reference wild-type strain RN6390. To determine that the phenomenon was not limited to one specific S. aureus strain (RN6390), we repeated this assay using the MSSA Newman strain exposed to PCA at 0.25xMIC (10 µg/mL) and observed similar results with an increase in tet38 transcript level as well as a subsequent protection of Newman against tetracycline at 10xMIC (data not shown). Further investigations will be carried out to assess the interactions between PCA, NaCl, TetR21, KdpD/E, and tet38 promoter to determine how PCA and salt affect the expression of tet38. We will include S. aureus such as USA300, Cowan I, 6850, and JE2 (20) to assess the variation in tet38 expression of different S. aureus in response to PCA and NaCl. Furthermore, in a co-culture system, P. aeruginosa production of phenazines such as PCA, PCN, PYO, and 1-OH-PHZ varies among different strains and is dependent on the growth media. We will include PAO1, PA14, and selected clinical P. aeruginosa in our future study to address the impact of strain-specific interactions.

It is noteworthy to mention that some clinical S. aureus strains, isolated alone or co-isolated with P. aeruginosa or other bacteria, naturally overexpressed efflux pump Tet38, as well as several other efflux pumps, with a level of expression varied between 2.5–4.01-fold when compared with the MSSA reference strain Newman (data not shown). The magnitude of Tet38 overexpression in these clinical S. aureus isolates was similar to that of Tet38 engineered strain RN6390(pLI50-tet38) together with an increase of two to four-fold in the MIC of tetracycline, doxycycline, and phenazines. These data suggested that Tet38 and other S. aureus efflux pumps carried various natural functions to maintain the well-being of the bacterial cell, including exporting a host of toxic compounds including antibiotics.

Taken together, this study showed that phenazine PCA was a substrate and an inducer of the Tet38 efflux pump, and the PCA-induced Tet38 protected S. aureus against phenazines and tetracycline. Tet38 is the third MFS efflux pumps after NorA and HprS that have been shown to include a P. aeruginosa exo-product phenazine as a substrate. While overexpression of Tet38 was beneficial to S. aureus survival against tetracycline and phenazines, this effect is reversed by hypertonic saline, suggesting that in addition to its mucolytic activity, it may enhance susceptibility to some phenazine-induced efflux pump substrates. Further investigations are underway to evaluate the effect of salt in S. aureus susceptibility to other Tet38 substrates.

Furthermore, several studies have shown that coinfection by S. aureus and P. aeruginosa caused more severe and complicated lung infections in CF patients. However, the reason behind this phenomenon was still not well understood (11, 38). Understanding how the interactions "Tet38-Phenazines" manifest in vivo would be important and require survival study of CF S. aureus in co-culture with CF P. aeruginosa (39), to determine if the coexistence would be dependent on the efflux pump profile of S. aureus and the phenazines of the P. aeruginosa.

In conclusion, S. aureus and other bacteria use efflux pumps and transporters as important tools to adapt to diverse environments including common polymicrobial environments with potential exposure to antibacterial exoproducts. Long-term adaptations that enable co-infection and contribute to antibiotic resistance are important topics for additional study and demonstrate a need for more targeted therapies that might benefit patients with CF who are chronically colonized with these pathogens.

MATERIALS AND METHODS

Bacterial strains and growth conditions

The bacterial strains, plasmids, and primers used in this study are listed in Table 2.

TABLE 2.

Bacterial strains, plasmids, and primers used in this study

Strain, plasmid, or primer	Genotype or relevant characteristic(s)	Reference or source
S. aureus strains
RN6390	Wild-type	(29)
Newman	Laboratory strain, high level of clumping factor	(40)
YW22	Δtet38 mutant from RN6390	(29)
RN6390 (pLI50-tet38)	tet8 overexpressor, CmR
RN6390 (pRN-sarAp-yfp)	sarA promoter-dependent YFP with sod RBS, CmR	This study
RN6390 (pRN-tet38p-yfp)	tet38 promoter-dependent YFP with sod RBS, CmR	This study
P. aeruginosa strains
PA14	Reference strain, virulent burn wound isolate	(41)
PA14phzM	Transposon insertion phzM mutant	(42)
PA14phzS	Transposon insertion phzS mutant	(42)
PAO1	Reference strain	(31)
CF-PA-1 and CF-PA-2	CF clinical strain	(26)
Escherichia coli strain
DH5α	F- Φ80lacZΔM15 Δ(lacZYA-argF) U169 recA1 endA1 hsdR17(rk-, mk+) phoA supE44 thi-1 gyrA96 relA1 λ-	Life Technologies
Plasmids
pLI50	Shuttle plasmid E. coli–S. aureus, CmR	(29)
pRN12	Shuttle vector E.coli–S.aureus, CmR Expression of sGFP from the sarA-P1 promoter	(43)
Primers for real-time RT-PCR assays
gmk	Forward 5’TCAGGACCATCTGGAGTAGGTAAAG 3’ Reverse 3’CAAATGCGTGAAGGTGAAGTTGATG 5’
tet38	Forward 5’ATCGTAGTATTTACGTTGCC 3’ Reverse 3’GGCTTAATTCTAGTGGCAAC 5’
norA	Forward 5’TGGCCACAATTTTTCGGTAT 3’ Reverse 3’CTTTGGCTACATGTCAGCGA 5’
norB	Forward 5’CTCGGATGCAAGAAACCAAT 3’ Reverse 3’GCTTCTGCATTAGGTGGAGC 5’
Primers for plasmid constructs pRN-sarAp-yfp and pRN-tet38p-yfp
yfp (KpnI)-F	Forward 5’ ATATGGTACCATGAGTAAAGGAGA 3’
yfp (EcoRI)-R	Reverse 3’ GTTGTTGAATTCGCATGGATGACTATACAATAA 5’
tet38p-F (NheI)	Forward 5’ ATGCGCTAGC CAATATCAATTACC 3’
tet38p-R	Reverse 3’ AAATTGCACTGTCAAGGAGGATGATTATTT
yfp-F	Forward 5’ AGGAGGATGATTATTTATGAGTAAAGGAGAAGAACTTTTC
yfp (EcoRI)-R	Reverse 3’GTTGTTGAATTCGCATGGATGACTATACAATAA

Tetracycline, doxycycline, chloramphenicol, phenazine-1 carboxylic acid (PCA), phenazine-1 carboxamide (PCN), 1-hydroxyphenazine (1-OH-PHZ), phenazine methosulfate (PMS), and pyocyanin (PYO) were purchased from Sigma-Aldrich (St. Louis, MO).

S. aureus containing plasmids pLI50, pRN12 and their various derived constructs were grown at 37°C in trypticase soy broth media (TSB) supplemented with chloramphenicol at 10 µg/mL. All other bacteria were grown in TSB media unless otherwise stated.

Antibiotic susceptibility assay

The MIC was determined by broth microdilution at 37°C for 24 hours, as previously described (28). A log-phase culture of S. aureus (OD₆₀₀ = 0.5) grown in TSB media was diluted 100-fold and inoculated into microtiter plates (Fisher Scientific, Pittsburgh, PA) containing two-fold serial dilutions of phenazines or antibiotics. MIC was the lowest drug concentration that produced no visible turbidity after incubation at 37°C for 24 hours.

To determine the MICs of RN6390 pre-exposed to PCA at 0.25 x MIC, NaCl 7%, or a combination of PCA +NaCl, RN6390 was cultivated in the presence of compounds for 1 hour then diluted 100-fold and inoculated into microtiter plates (Fisher Scientific, Pittsburgh, PA) containing the appropriate inducer (PCA, NaCl, or both) and two-fold serial dilutions of tetracycline or doxycycline. MIC was the lowest drug concentration that produced no visible turbidity after incubation at 37°C for 24 hours.

Construction of plasmids pRN-sarAp-yfp and pRN-tet38p-yfp

We constructed the plasmids pRN-sarAp-yfp and pRN-tet38p-yfp using a technique that was previously described with some modifications (44). All primers used in this study were synthesized by Eton Bioscience Inc. (Eton Bioscience, Boston, MA) and are listed in Table 2. The sod RBS-yfp gene was amplified from plasmid pAH16 (45) using primers yfp (KpnI)-F and yfp (EcoRI)-R (Table 2). Plasmid pRN12 that carried the ametrine reporter gene fused to the sarA promoter was purchased from Addgene (43) (Addgene, Watertown, Massachusetts). pRN12 was digested with enzymes KpnI and EcoRI to remove the mAmetrine reporter gene of pRN12 and replaced with the amplified and KpnI/EcoRI digested PCR product (sod RBS—yfp) to yield plasmid construct pRN-sarAp-yfp that was used as a bacterial cell viability control in confocal experiments.

In parallel, we created the plasmid construct pRN-tet38p-yfp by performing a series of three PCR reactions using primers specific for tet38 promoter (tet38p) and yfp gene (Table 2) (44). We amplified a 354 bp tet38 promoter from the chromosome of S. aureus RN6390 using primers tet38p-F (NheI) and tet38p-R. The tet38p-R primer carried an overhang sequence of 16-nucleotide (AGGAGGATGATTATTT) that was complementary to the 16-nucleotide sequence of the PCR product yfp amplified using primers yfp-F and yfp (EcoRI)-R. A third PCR reaction was carried out using the tet38p and yfp PCR products as templates and the primers tet38p-F (NheI I) and yfp (EcoRI)-R. The final PCR product was digested with NheI and EcoRI and cloned into plasmid pRN12 previously digested with the same restriction enzymes. The construct pRN-tet38p-yfp was transformed into S. aureus RN6390 and grown in TSB supplemented with chloramphenicol at 10 µg/mL. Confocal imaging was carried out to visualize and evaluate the expression of YFP when the tet38 promoter was induced.

Quantitative real-time RT-PCR assay

The real-time RT-PCR assays were done as previously described (26). Total S. aureus RNA was extracted from lysostaphin-treated cells using the RNeasy midi kit (Qiagen, Valencia, CA). cDNAs were synthesized using the Verso cDNA synthesis kit (Thermo Scientific, ABgene, Epsom, Surrey, United Kingdom), followed by real-time qRT-PCR assays using EvaGreen dye and the CFX96 real-time system (Bio-Rad, Hercules, CA). Primers designed for the qRT-PCR assays were synthesized at Eton Bioscience Inc. (Eton Bioscience, Boston, MA), and are listed in Table 2. The housekeeping gene gmk was used as an internal control. All samples were analyzed in triplicate, and expression levels normalized against gmk gene expression, which remained unchanged following exposure to 0.5 x MIC concentration of phenazines or supernatants of P. aeruginosa. The assays were repeated with three independent biological samples. Statistical analyses were performed using a one-way ANOVA with a t-test as post-hoc test and a Bonferroni adjustment to determine the significance of differences in gene expression values.

Confocal imaging assay

S. aureus RN6390 transformed with pRN-sarAp-yfp or pRN-tet38p-yfp were grown in TSB supplemented with 10 µg/mL of chloramphenicol at 37°C. At OD₆₀₀ ~0.5 (time = 0), phenazines alone or with tetracycline at 1x MIC or 10x MIC were added to the cultures, and the bacteria were exposed to phenazines ± antibiotics for 4 hours, except when PCA at 0.25x MIC was added first for 1 hour, then tetracycline was added for an additional 3 hours. The bacterial samples were subjected to confocal imaging as previously described (44). The S. aureus cultures were adjusted to have 10⁴ bacterial cells per well in a Nunc Lab-Tek II 8-Well Borosilicate Glass Chambered Coverglass (Thermo Scientific 155409PK). The chambered cover glass was mounted onto a Nikon Ti-E inverted microscope fitted with a spinning disc confocal C54-X1 head (Yokogawa, Sugar Land, TX). Solid-state lasers were used to produce excitation wavelengths of 488 nm (relevant to this study). Bacteria were imaged using a Nikon 100 x objective (1.49 NA, oil immersion objective, Nikon). Images were captured using an electron-multiplying charge-coupled device (EM-CCD) camera (C9100-13; Hamamatsu, Bridgewater, NJ) and analyzed using MetaMorph software version 7.10.5.476 (Molecular Devices, Downington, PA) (46). The confocal assays were repeated using three independent biological samples. The visualization with confocal imaging was performed in parallel with colony counting for accuracy of the observations and to evaluate the statistical significance of the difference in bacterial survival following exposure to PCA and tetracycline. All confocal images were processed and set to the same brightness/contrast as the positive control in each assay using FIJI (ImageJ2) version 2.3.0/1.53q.

Supernatant exposure assays

We grew P. aeruginosa PA14, PA14_phzM, and PA14_phzS in TSB media overnight then followed the protocol described by Niggli et al. (20) to prepare the assay media with a proportion of 30% P. aeruginosa supernatant in TSB media. In brief, we centrifuged 10 mL of overnight bacterial cultures then discarded the pellets. The supernatants were filtered using 0.2 µm filter cups and the sterile supernatants were added to TSB in the proportion of 30% P. aeruginosa supernatant +70% TSB. The mixtures were used as supernatant assay media for the S. aureus exposure assays. In parallel, we grew S. aureus RN6390 in TSB media from an overnight culture until OD₆₀₀ reached 0.5. We centrifuged a series of 2 mL of RN6390 cultures. Each pellet was resuspended in 2 mL of the assay media prepared with the appropriate P. aeruginosa PA14 or mutants. The supernatant exposure assay was carried out for 1 hour then total RNAs of RN6390 were extracted and submitted to real-time qPCR to determine the relative pump gene norA, norB, and tet38 transcripts. The relative gene transcripts were expressed as the fold change (FC) of RN6390 pump gene exposed versus non-exposed to PA supernatants. RN6390 grew in 100% TSB served as a positive control of S. aureus growth, and S. aureus grew in saline TSB (0.9% NaCl solution +70% TSB) served as a growth in reduced nutrient control (negative control). We used the gene gmk as an internal control of the assay.

The supernatants of P. aeruginosa PAO1 and two CF clinical isolates CF-PA-1 and CF-PA-2 were prepared as described above. Supernatant exposure assay using S. aureus Newman was carried out as was done with RN6390.

Growth curves and kill curves of S. aureus strains RN6390 exposed to PCA with and without tetracycline

S. aureus RN6390 and Δtet38 mutant YW22 were cultured overnight at 37°C in TSB media. Then, 0.1 mL of each culture was transferred into 10 mL of fresh TSB liquid media and allowed to grow at 37°C under shaking for a period of 24 hours. Bacterial samples were collected at 0, 2, 4, and 24 hours, diluted, and plated on TSB agar for colony counts (CFU/mL) (29).

PCA at 0.25x MIC and tetracycline at 1x MIC and 10x MIC were added to the bacterial cultures at OD₆₀₀ ~0.5 to perform S. aureus growth/kill curves. The concentrations of antibiotics at 1x and 10x were adjusted depending on the MICs of the RN6390 and YW22. S. aureus RN6390 and YW22 were cultivated overnight, and 0.5 mL of the overnight cultures were added to 50 mL of fresh TSB media. At OD₆₀₀ ~0.5 (10⁷ CFU/mL), we carried out a series of 4 conditions and added to a series of 5 mL S. aureus cultures (a) PCA at 0.25x MIC; (b) Tetracycline at 1x MIC; (c) PCA at 0.25x MIC and tetracycline at 1x MIC; (d) PCA at 0.25x MIC for 1 hour then added tetracycline at 1x MIC. We repeated the series of assays with PCA at 0.25x MIC and tetracycline at 10x MIC.

All S. aureus continued to grow for 24 hours with samples collected at 0, 2, 4, and 24 hours for colony counts (CFU/mL). All experiments were repeated using three independent biological samples. Statistical analyses were performed using a one-way ANOVA with a t-test as post-hoc test and a Bonferroni adjustment to determine the significance of differences in the growth of S. aureus strains RN6390 and tet38 mutant YW22 in the presence of PCA ± tetracycline at 1x MIC and 10x MIC.

S. aureus RN6390 and Δtet38 mutant exposed to PCA with or without tetracycline and NaCl at 7%

S. aureus were cultivated overnight, and 0.5 mL of the overnight cultures were added to 50 mL of fresh TSB media and allowed to grow at 37°C under shaking. The concentrations of tetracycline at 10x were adjusted depending on the MICs of RN6390 and YW22. PCA at 0.25x MIC, tetracycline at 10x MIC, and NaCl at 7% final concentration were added to the bacterial cultures as described below. At OD₆₀₀ ~0.5 (10⁷ CFU/mL), we carried out a series of 8 conditions and added to a series of 5 mL S. aureus cultures (1) No compounds; (2)+ PCA at 0.25x MIC; (3)+ NaCl 7%; (4)+ PCA + NaCl; (5)+Tetracycline at 10x MIC; (6)+PCA + tetracycline at 10x MIC; (7)+PCA for 1 hour then added tetracycline at 10x MIC; (8)+ PCA for 1 hour then added tetracycline +NaCl 7%. S. aureus was cultivated at 37°C for 24 hours then colony counts were performed.

Statistical analysis

All experiments were repeated using three independent biological samples and data were expressed as a mean ± SD. Statistical analyses were performed using a one-way analysis of variance (ANOVA). The pairwise comparison was done with a t-test to compare sample groups combined with a Bonferroni adjustment. The threshold for significance was set at a P-value < 0.05.

The differences between the number of viable cells RN6390 or YW22 ± PCA (0.25x MIC) exposed to tetracycline at 1x MIC or 10x MIC versus non-exposed were statistically significant as determined by a one-way ANOVA with a pairwise t-test and Bonferroni adjustment (P < 0.05) (Fig. 4).

The differences between the number of viable cells RN6390 or YW22 ±PCA (0.25x MIC) ±NaCl at 7% exposed to tetracycline at 10x MIC versus non-exposed were statistically significant as determined by a one-way ANOVA with a pairwise t-test and Bonferroni adjustment (P < 0.05) (Fig. 5A and B).