The Burkholderia pseudomallei BpeAB-OprB efflux pump: expression and impact on quorum sensing and virulence

Abstract

BpeAB-OprB is a multidrug efflux pump of the bacterial pathogen Burkholderia pseudomallei and is responsible for conferring antimicrobial resistance to aminoglycosides and macrolides. Expression of bpeAB-oprB is inducible by its substrate erythromycin and upon entry into stationary phase. BpeR, a member of the TetR family, functions as a repressor of the bpeAB-oprB operon. bpeR expression was similarly induced at stationary phase but lagged behind the induction of bpeAB-oprB expression. The induction of bpeAB-oprB expression could be advanced to the early exponential phase by exogenous addition of the B. pseudomallei autoinducers N-octanoyl-homoserine lactone (C8HSL) and N-decanoyl-homoserine lactone (C10HSL), suggesting that the bpeAB-oprB operon may be quorum regulated. On the other hand, acyl-homoserine lactone (acyl-HSL) production was undetectable in the bpeAB-null mutant and strains which overexpress bpeR. The failure of these strains to produce acyl-HSLs seemed to be at the level of synthesis of acyl-HSLs, as growth-phase-dependent expression of the autoinducer synthase BpsI was abolished in the bpeAB-null mutant. bpsI expression remained growth phase dependent in the bpeR mutant which had functional BpeAB-OprB. BpeAB-OprB function is likewise necessary for optimal production of quorum-sensing-controlled virulence factors such as siderophore and phospholipase C and for biofilm formation. Cell invasion and cytotoxicity towards human lung epithelial (A549) and human macrophage (THP-1) cells were also significantly attenuated in both the bpeAB mutant and bpeR-overexpressing strains, thus suggesting the possibility of attenuating B. pseudomallei virulence using inhibitors of the BpeAB-OprB efflux pump.

FIG. 1.

Analyses of bpeA and bpeR expression by RT-PCR. RT-PCR results obtained using total RNA isolated from exponential-phase cultures of wild-type parental strain KHW (lane 1), KHWΔbpeAB mutant (lane 2), KHWΔbpeAB(pUCP28TbpeAB) (lane 3), KHWbpeR::Km (lane 4), KHWbpeR::Km(pUCP28TbpeR) (lane 5), and KHW(pUCP28TbpeR) (lane 6). Lane M is 1-kb Plus size markers (Invitrogen, Carlsbad, CA). The bands corresponding to bpeA, bpeR, and 16S rRNA RT-PCR products are indicated on the right. bpeA expression was absent in KHWΔbpeAB (lane 2) but was restored by complementation (lane 3). bpeR expression was absent in KHWbpeR::Km (lane 4), and complementation with pUCP28TbpeR resulted in complete repression of bpeA expression (lane 5). Overexpression of bpeR in KHW also resulted in complete repression of bpeA expression (lane 6).

FIG. 2.

Effect of erythromycin on expression of bpeAB_promoter-lacZ in wild-type B. pseudomallei and the bpeR-null mutant. (A) Effect of bpeR mutation on the basal and inducible expression of bpeAB in KHW and KHWbpeR::Km. β-Galactosidase activities expressed from bpeAB_promoter-lacZ fusion were determined in KHW (circles) and KHWbpeR::Km (triangles) harboring the reporter plasmid pCYYbpeAB. Open symbols represent basal expression of bpeAB, while closed symbols represent bpeAB expression after addition of 0.1× MIC of erythromycin (or 10 μg/ml). (B) Dose-dependent induction of bpeAB expression in KHW(pCYYbpeAB) by erythromycin. β-Galactosidase activities were assayed 4 h after the addition of erythromycin to the bacterial cultures in AB medium containing 20 mM glucose, 0.2% Casamino Acids, and 25 μg/ml tetracycline. Error bars represent standard deviations of triplicate β-galactosidase determinations for one typical experiment. Where error bars are not shown, the standard deviation was within the size of the symbol.

FIG. 3.

Growth-phase-dependent expression of bpeAB and bpeR in wild-type KHW. (A) β-Galactosidase activities were expressed from pCYYbpeAB in KHW. (B) β-Galactosidase activities were expressed from pCYYbpeR in KHW. Closed circles or squares represent β-galactosidase activities in Miller units, while open triangles represent cell densities (OD₆₀₀). Error bars represent standard deviations of triplicate determinations of cell densities and β-galactosidase activities. Where error bars are not shown, the standard deviation was within the size of the symbol. The bacteria were cultured in AB medium containing 25 μg/ml tetracycline, 20 mM glycerol, and 0.2% (wt/vol) Casamino Acids. (C) Verification of growth-phase-dependent expression of bpeA and bpeR in wild-type cells by RT-PCR. RT-PCR was performed to detect bpeA and bpeR expression from the onset of culture (0 h) to stationary phase (30 h) (upper row). RT-PCR of 16S rRNA was included as an internal control for the reaction and to normalize the amount of total RNA used. Bands corresponding to the bpeA, bpeR, and 16S rRNA transcripts are indicated on the right, while lane M is the 1-kb Plus molecular size markers.

FIG. 4.

Advancement of growth-phase-dependent induction of bpeAB-lacZ expression to early exponential phase by exogenous C8HSL and C10HSL. (A) Growth curves (OD₆₀₀) of KHW in the presence (○) or absence (□) of 100 nM C8HSL. Also shown are the β-galactosidase activities (Miller units) representing bpeAB expression from pCYYbpeAB from onset of culture (0 h) to early exponential phase (12 h) in the absence (grey bars) or presence (black bars) of C8HSL, respectively. (B) Growth curves (OD₆₀₀) of KHW in the presence (○) or absence (□) of 100 nM C10HSL. Also shown are β-galactosidase activities (Miller units) representing bpeAB expression in the absence (grey bars) or presence (black bars) of C10HSL, respectively. (C) Dose-dependent induction of bpeAB-lacZ expression by exogenous C8HSL. The horizontal axis represents the concentrations of exogenous C8HSL added to the culture medium. Black bars represent cultures to which different amounts of C8HSL were added, while grey bars represent control cultures to which no C8HSL was added. C8HSL was added to the bacterial cultures in AB medium containing 20 mM glucose, 0.2% Casamino Acids, and 25 μg/ml tetracycline about 1 h after inoculation from an overnight culture (OD₆₀₀ of ∼0.1 to 0.2), and β-galactosidase assays were performed on aliquots of the cultures 4 h after the addition of exogenous C8HSL. Error bars represent standard deviations of triplicate β-galactosidase determinations for one typical experiment.

FIG. 5.

(A) Detection of autoinducer production by B. pseudomallei KHW and its derivatives. Vertical streaks represent the E. coli JB525 reporter strain harboring the luxR-luxI_promoter-GFP plasmid (pJBA132), while the horizontal streaks represent (I) KHW, (II) KHWΔbpeAB mutant, (III) KHWΔbpeAB(pUCP28T-bpeAB)-complemented mutant, (IV) KHWbpeR::Km mutant, (V) KHWbpeR::Km(pUCP28T-bpeR)-complemented mutant, and (VI) KHW(pUCP28TbpeR), respectively. Fluorescence on the vertical streaks indicates the activity of autoinducers produced by the horizontal streaks of tester strains. (B) Effect of BpeAB-OprB on growth-phase-dependent expression of bpsI_{promoter-lacZ}. The plasmid pSYI carrying bpsI_promoter-lacZ was introduced into KHW, KHWΔbpeAB, and KHWbpeR::Km, respectively, to study the effect of BpeAB-OprB on the expression of the autoinducer synthase BpsI. Aliquots of the bacteria cultured in AB medium containing 20 mM glycerol, 0.2% Casamino Acids, and 25 μg/ml tetracycline were assayed at different time intervals from the onset of culture (0 h) to stationary phase (30 h). Dotted lines represent the cell densities of KHW(pSYI) (□), KHWΔbpeAB(pSYI) (○), and KHWbpeR::Km(pSYI) (⧫), respectively, while β-galactosidase activities are represented as bars [KHW(pSYI), black bars; KHWΔbpeAB(pSYI), gray bars; and KHWbpeR::Km(pSYI), striped bars]. All measurements were done in triplicate; the means and standard errors are shown.

FIG. 6.

Effect of BpeAB-OprB on siderophore and phospholipase C production and biofilm formation in B. pseudomallei. Siderophore activities, phospholipase C activities, and biofilm formation were measured in KHW, KHWΔbpeAB, KHWΔbpeAB complemented with pUCP28TbpeAB, KHWbpeR::Km, KHWbpeR::Km complemented with pUCP28TbpeR, and KHW harboring pUCP28TbpeR, respectively. (A) Optimal siderophore production in B. pseudomallei is dependent on the BpeAB-OprB function, and overexpression of bpeR in KHWbpeR::Km(pUCP28TbpeR) and KHW(pUCP28TbpeR) reduced siderophore production. Siderophore activities were assayed in the supernatants of 24-h-old cultures and were determined by measuring the differential in OD₆₃₀ readings between the test and the sample blank. The values shown have been normalized for cell density by being expressed as a ratio of ΔOD₆₃₀/OD₆₀₀. (B) Optimal PLC secretion by B. pseudomallei KHW is dependent on BpeAB-OprB, and overexpression of bpeR in KHWbpeR::Km(pUCP28TbpeR) and KHW(pUCP28TbpeR) inhibited PLC secretion. PLC activities were determined in the supernatants of 24-h cultures, and the values shown have been normalized for cell density by being expressed as a ratio of OD₄₁₀/OD₆₀₀. (C) Optimal biofilm formation is also dependent on BpeAB-OprB. Biofilm formation was significantly reduced in KHWΔbpeAB and KHW overexpressing bpeR. Each bar is the average reading (± standard deviation) from three independent experiments.