ABSTRACT
The resistance-nodulation-division (RND)-type efflux pump is one of the causes of the multidrug resistance of Stenotrophomonas maltophilia. The roles of the RND-type efflux pump in physiological functions and virulence, in addition to antibiotic extrusion, have attracted much attention. In this study, the contributions of the constitutively expressed SmeYZ efflux pump to drug resistance, virulence-related characteristics, and virulence were evaluated. S. maltophilia KJ is a clinical isolate of multidrug resistance. The smeYZ isogenic deletion mutant, KJΔYZ, was constructed by a gene replacement strategy. The antimicrobial susceptibility, virulence-related physiological characteristics, susceptibility to human serum and neutrophils, and in vivo virulence between KJ and KJΔYZ were comparatively assessed. The SmeYZ efflux pump contributed resistance to aminoglycosides and trimethoprim-sulfamethoxazole. Inactivation of smeYZ resulted in attenuation of oxidative stress susceptibility, swimming, flagella formation, biofilm formation, and secreted protease activity. Furthermore, loss of SmeYZ increased susceptibility to human serum and neutrophils and decreased in vivo virulence in a murine model. These findings suggest the possibility of attenuation of the resistance and virulence of S. maltophilia with inhibitors of the SmeYZ efflux pump.
INTRODUCTION
Stenotrophomonas maltophilia, a widespread environmental microorganism, is an emerging multidrug-resistant global opportunistic pathogen. S. maltophilia is generally associated with septicemia and various infections, especially respiratory infections, in immunocompromised patients and cystic fibrosis (CF) patients. The difficulty in the treatment of S. maltophilia infection has been attributed to its intrinsic multidrug resistance (1). The clinical isolates of S. maltophilia generally exhibit high resistance to β-lactams, macrolides, and aminoglycosides, which is attributed to the overexpression of β-lactamases, aminoglycoside-modifying enzymes, and multidrug efflux pumps (2).
The multidrug efflux pumps are membrane proteins that are involved in the extrusion of antibiotics and are classified into five families, including resistance-nodulation-division (RND), major facilitator superfamily (MFS), small multidrug resistance (SMR), ATP-binding cassette (ABC), and multidrug and toxic compound extrusion (MATE) (3). Of the five families, the RND-type efflux pumps form a tripartite assembly in the bacterial membrane, comprising an RND-type inner membrane protein (IMP), an outer membrane protein (OMP), and a membrane fusion protein (MFP) to link the IMP and OMP (4). RND-type efflux pumps confer resistance to a large array of drugs and make a significant contribution to drug resistance in Gram-negative bacteria such as Escherichia coli, Pseudomonas aeruginosa, and S. maltophilia.
In addition to the known roles in antibiotic extrusion, RND-type efflux pumps are increasingly recognized as components linking bacterial physiology and virulence (5). For example, inhibition of the efflux pumps impairs biofilm formation in Salmonella enterica serovar Typhimurium (6). The BpeAB-OprB pump of Burkholderia pseudomallei, the AcrAB-TolC pump of Enterobacter cloacae, and the AcrAB-TolC pump of Yersinia pestis are involved in virulence (7–9). In addition, evidence that the expression of efflux pumps is regulated by different physiological stresses further indicates the close relationship between RND-type efflux pumps and physiological functions. The MexXY efflux pump of P. aeruginosa (10, 11) and the CmeABC efflux pump of Campylobacter jejuni (12) are induced by oxidative stress. The MexEF-OprN efflux pump of P. aeruginosa (13) is activated in response to oxidative and nitrosative stresses. A variety of membrane-damaging agents can induce the MexCD-OprJ efflux pump of P. aeruginosa (14).
Eight putative RND-type efflux systems (SmeABC, SmeDEF, SmeGH, SmeIJK, SmeMN, SmeOP, SmeVWX, and SmeYZ) were annotated in the S. maltophilia K279a genome (15). Of the eight pumps, six pumps (SmeABC, SmeDEF, SmeIJK, SmeOP, SmeVWX, and SmeYZ) have been characterized, mainly focusing on their roles in antimicrobial extrusion (16–21). In particular, it was noted that the SmIJK and SmeYZ pumps are constitutively highly expressed, and both are redundant in extrusion of aminoglycosides (18, 21). It is thought that antimicrobial extrusion may not be the major function of the SmeIJK and SmeYZ pumps, and both pumps may have distinct physiological functions. Recently, we reported that the physiological role of the SmeIJK pump is linked to the maintenance of membrane integrity (18). In this article, we further elucidated the significance of the SmeYZ pump in physiological functioning. The results demonstrated that inactivation of the SmeYZ pump compromised the virulence-related physiological functions of swimming, flagella formation, oxidative stress susceptibility, biofilm formation, and protease secretion and thus decreased in vivo virulence. These findings supported the possibility that SmeYZ can develop as a new target for the treatment of S. maltophilia infection.
MATERIALS AND METHODS
Bacterial strains, primers, and growth conditions.S. maltophilia KJ is a clinical isolate (22). All primers used in this study were designed based on the S. maltophilia K279a genome sequence (15). For general purposes, strains were grown aerobically in Luria-Bertani (LB) medium except as specifically noted.
Construction of deletion mutant KJΔYZ.To construct the ΔsmeYZ mutant, the 1,759-bp product containing partial smeYZ genes was amplified from the genome of S. maltophilia KJ using primers SmeYZ-F (5′-CGCAAGCTTGACCTGCGCTATGCC-3′) and SmeYZ-R (5′-TGCGAATTCCAGCAGCATCGCCTCC-3′) and cloned into pEX18Tc, generating pYZ. An internal 489-bp ClaI-ClaI DNA fragment was removed from pYZ, obtaining pΔYZ. The plasmid pΔYZ was introduced into E. coli S17-1 by transformation and mobilized into S. maltophilia KJ via conjugation (23). Transconjugants carrying deleted smeYZ in the chromosome were obtained by two-step selection on LB agar containing tetracycline (30 μg/ml)-norfloxacin (2.5 μg/ml) and then on LB agar containing 10% (wt/vol) sucrose. Mutants was confirmed by colony PCR (24).
Antimicrobial susceptibility test.The antimicrobial susceptibilities of S. maltophilia strains were tested by the serial 2-fold dilution method in Mueller-Hinton (MH) agar according to the guidelines of the Clinical and Laboratory Standards Institute (CLSI) (25). The plates were incubated at 37°C and examined visually after 24 h. The MIC was defined as the lowest concentration of antimicrobial agent that inhibited visible growth. All antibiotics were purchased from Sigma-Aldrich (USA).
Growth kinetic assay.Overnight cultures were reinoculated into fresh LB medium with an initial optical density at 450 nm (OD450) of 0.15. The OD450 readings were taken every 3 h for a total time of 24 h.
Cell surface hydrophobicity assay.The cell surface hydrophobicity values were determined using a modified two-phase partitioning system (26). The logarithmic-phase cells were washed twice with phosphate-buffered saline (PBS) buffer and resuspended in PBS before the OD450 values were recorded (A0). Bacterial partitioning was done by addition of 5 ml of bacterial suspension to 1 ml of hexadecane. The bacterial suspension in the phase system was vortexed for 2 min and allowed to stand undisturbed for phase separation for 3 min at room temperature. The lower aqueous phase was read to determine the OD450 (A). The hydrophobicity value was calculated as [1 − (A/A0)] × 100.
H2O2 susceptibility test (disk diffusion assay).The H2O2 susceptibility was tested as described previously with some modifications (27). MH plates were streaked with a cotton swab soaked in S. maltophilia cell suspension of 107 cells/ml. The sterile paper disks (6 mm in diameter) were placed on the plate, and the disks were spotted with 20 μl of 10% H2O2. The diameter of the zone of growth inhibition around each disk was measured after a 24-h incubation at 37°C.
Menadione susceptibility assay.Overnight cultures of the tested strains were diluted to an OD450 of 0.15 with LB broth. Cells were grown for 2 h, and serial dilutions were plated onto LB medium containing 0, 10, 20, and 30 μg/ml menadione. Plates were incubated for 24 h at 37°C and scored for CFU. The percentage of survival was defined as the CFU ratio of KJΔYZ to wild-type KJ.
Swimming assay.To assess swimming motility, 5 μl of overnight culture cultivated at 37°C in LB broth was inoculated at the swimming agar surface (1% tryptone, 0.5% NaCl, and 0.15% agar) (28). The plates were incubated at 37°C for 48 h. Results are expressed as diameters (millimeters) of swimming zones.
Flagella staining.The presence of flagella was determined by negative staining and transmission electron microscopy (TEM) (Hitachi H-7650 microscope). Bacteria were negatively stained using the drop method for 10 s with 1% phosphotungstic acid (pH 7.4) on Formvar-coated copper grids (29).
Assays of biofilm formation.The biofilm assay was carried out as described previously with some modifications (30). The overnight-cultured bacteria were adjusted to an OD450 of 0.1, and 200 μl of bacterial suspension was inoculated into each microtiter well, with a minimum of three wells per bacterial strain for each assay. After a 48-h incubation at 37°C, the total cell biomass was estimated by enumerating the CFU. The culture supernatants were decanted, and the wells were washed with distilled water. The biofilms in the well were stained by 200 μl of 1% crystal violet for 15 min. The stained biofilms were rinsed with distilled water, extracted with 200 μl of 70% ethanol, and quantified by measuring the A570 of dissolved crystal violet. Uninoculated medium controls were included. The levels of biofilm formation were expressed as the crystal violet staining relative to the cell biomass (A570/CFU).
Secreted protease activity assay.The secreted protease activity of the bacteria was assayed using LB agar containing 1% skim milk. For the convenience of bacterial loading, the skin milk agar was prepared with a 6-mm-diameter hole in the center. Strains tested were cultured in LB agar at 37°C for 20 h and adjusted with LB broth to an OD450 of 1.0, and 40 μl of these cultures was dripped onto the hole of the skim milk agar plates. After incubation at 37°C for 72 h, the proteolytic activity of bacteria was assessed by measuring the transparent zones around the bacteria.
Serum killing assays.Serum killing assays were performed according to the published method with some modifications (31). The survival of exponential-phase bacteria (wild-type KJ and mutant KJΔYZ) in nonimmune human serum was measured. A log-phase inoculum of 5 ×104 CFU was mixed at a 1:1 (vol/vol) ratio with mixed nonimmune human serum or heat-inactivated serum (56°C for 30 min) donated by 2 healthy volunteers. The final mixture, comprising 50% nonimmune serum or heat-inactivated serum by volume, was incubated at 37°C for 1 to 3 h. To determine the number of viable bacteria after exposure to serum, an aliquot of each bacterial suspension was removed at the designated time point, diluted 10-fold by addition of Mueller-Hinton broth, and plated on Mueller-Hinton agar. The mean survival ratio was plotted.
Killing assays by human neutrophils.Human neutrophils were freshly isolated from peripheral blood donated by 2 healthy volunteers. An inoculum containing 105 CFU of bacteria was opsonized with 25% normal human serum for 15 min on ice and incubated with or without 105 human neutrophils in 1× phosphate-buffered saline at 37°C for 45 min. The percentages of survival of wild-type KJ and mutant KJΔYZ were calculated on the basis of the viable counts relative to those for the no-neutrophil controls (32).
Mouse inoculation experiments.Virulence was evaluated by mortality in a murine model of septicemia generated by intraperitoneal injection (33). Groups of 8-week-old male C57BL/6 mice were infected intraperitoneally with the wild-type KJ and mutant KJΔYZ (109 CFU; 5 mice for each dose). The exact inoculation dose was confirmed by serial dilution and plating on Mueller-Hinton agar. Mice were monitored for 14 days. A control group without bacterial injection was also included. All animal care procedures and protocols were approved by the Institutional Animal Care and Use Committee of National Yang-Ming University.
Statistical analyses.Data are presented as means ± standard deviations. The statistical significance was assessed by a 2-tailed Student's t test using Prism 5 (GraphPad) software. Survival was assessed by Kaplan-Meier analysis with a log rank test. P values of <0.05 were considered significant.
RESULTS
Loss of SmeYZ affects susceptibility to aminoglycosides and SXT.Because the SmeYZ is an RND-type efflux pump, the contribution of SmeYZ to antimicrobial susceptibility was first evaluated. The smeYZ mutant (KJΔYZ) was more susceptible to the aminoglycosides (amikacin, gentamicin, and kanamycin) (Table 1), which is consistent with the previous findings (21). In addition, deletion of SmeYZ compromised resistance to trimethoprim-sulfamethoxazole (SXT) (Table 1).
Susceptibility tests of S. maltophilia KJ and KJΔYZ
Loss of SmeYZ affects virulence-related physiological characteristics of S. maltophilia KJ.The bacterial growth was assessed by monitoring the OD450 and viable counts at the different growth time points. The logarithmic-phase growth of KJΔYZ was comparable to that of wild-type KJ, but the OD450 of KJΔYZ cells gradually declined after 9 h of culture (Fig. 1A). We also noticed that the KJΔYZ cells were gradually aggregated onto the surface of the culture bottle after 9 h. However, Fig. 1B demonstrates that the viable cells of KJ and KJΔYZ were comparable during the growth course monitored. In addition, we also noticed that the bacterial colony of KJΔYZ was homogeneously smaller than that of KJ when a single colony was isolated by the three-phase streaking method on LB agar (Fig. 1C). To further address whether the bacterial surface characteristics affect cell stacking, the cell surface hydrophobicity values of KJ and KJΔYZ were determined. Figure 1D shows that the hydrophobicity value of KJΔYZ was higher than that of KJ.
Bacterial growth curves, viability, and colony sizes. (A) Growth curves. The overnight-cultured bacteria were inoculated into fresh LB broth at the initial OD450 of 0.15. The bacterial growth was monitored by recording the OD450 every 3 h. Data are the means from three independent experiments. Error bars indicate the standard deviations for three triplicate samples. (B) Bacterial viability was monitored by counting the numbers of CFU. Data are the means from three independent experiments. Error bars indicate the standard deviations for three triplicate samples. (C) Single colony isolation. The single colonies of KJ and KJΔYZ cells were isolated by the three-phase streaking method on LB agar. (D) Bacterial surface hydrophobicity values. The bacterial surface hydrophobicity values were determined by partitioning of cells in two-phase systems (water-hexadecane). Data are the means from three independent experiments. Error bars indicate the standard deviations for three triplicate samples. *, P < 0.05, significance calculated by Student's t test.
The role of SmeYZ in the response to oxidative stresses was tested by H2O2 and menadione susceptibility assays. The mean zone of H2O2 inhibition exhibited by KJΔYZ (55 ± 3.0 mm) was larger than that produced by the parental strain KJ (45 ± 3.2 mm) (Fig. 2A). Furthermore, the survival rates of strains KJ and KJΔYZ were determined by the plate count method after the strains tested were exposed to different concentrations of menadione. The viability of KJΔYZ relative to that of KJ decreased when the menadione concentration was increased (Fig. 2B), supporting the observation that KJΔYZ is more susceptible to oxidative stress than wild-type KJ.
Oxidative stress susceptibility assay. Data are the means from three independent experiments. Error bars indicate the standard deviations for three triplicate samples. *, P < 0.05, significance calculated by Student's t test. (A) H2O2 susceptibility. The MH agar was uniformly spread with bacterial cell suspension. Sterile filter paper with 20 μl of 10% H2O2 was placed onto MH agar, and the diameter of a zone of growth inhibition was measured after 24 h of incubation at 37°C. (B) Menadione tolerance assay. Serial dilutions of the logarithmic-phase bacterial cells were plated onto LB containing 0, 10, 20, and 30 μg/ml menadione. The CFU were scored after 24 h of incubation at 37°C. The percentage of relative survival was defined as the CFU ratio of mutant to wild type.
The swimming abilities of KJ and KJΔYZ cells for migration through semisolid agar (0.15% agar) were evaluated. After 48 h of culture, KJ cells exhibited a swimming zone of 32 ± 2 mm. In contrast, no swimming zone was observed for KJΔYZ cells except the KJΔYZ grew a lawn of approximately 6 mm (Fig. 3A). Since the flagella are the critical organelles determining bacterial motility, the impact of smeYZ on flagella formation was examined. Electron microscopy analyses showed that wild-type KJ cells had several flagellar structures in the pole region. In contrast, no flagella were observed in KJΔYZ cells (Fig. 3B).
Motility ability and flagella formation. (A) Motility ability. Five microliters of bacterial cell suspension was inoculated into the swimming agar (1% tryptone, 0.5% NaCl, and 0.15% agar). Results were expressed as diameters (millimeters) of swimming zones after 48 h of incubation at 37°C. Data are the means from three independent experiments. Error bars indicate the standard deviations for three triplicate samples. *, P < 0.05, significance calculated by Student's t test. (B) Flagella formation. The flagella were negatively stained with 1% phosphotungstic acid (pH 7.4) and observed by TEM.
The crystal violet biofilm assay was used to determine the ability of KJ and KJΔYZ strains to form biofilms. The statistical analysis revealed that KJΔYZ had a reduced ability to form biofilm in comparison to the wild-type KJ (Fig. 4A). The secreted protease activity was determined by measuring the transparent zones around the bacteria. The transparent zone of KJΔYZ was smaller than that of wild-type KJ (Fig. 4B), indicating that KJΔYZ exhibited a lower secreted protease activity.
Biofilm formation and secreted protease activity assay results. Data are the means from three independent experiments. Error bars indicate the standard deviations for three triplicate samples. *, P < 0.05, significance calculated by Student's t test. (A) Biofilm formation. Each microtiter well was inoculated with 200 μl of the OD450 0.1 bacterial culture and incubated at 37°C for 48 h. The total cell biomass was estimated by enumerating the CFU. The amount of biofilm was determined by crystal violet staining. The stained biofilms were quantified by measuring the A570 of dissolved crystal violet. (B) Secreted protease activity assay. Forty microliters of bacterial cell suspension was dipped onto LB agar containing 1% skim milk. After incubation at 37°C for 72 h, the proteolytic activity of the bacteria was assessed by measuring the transparent zones around the bacteria.
Loss of smeYZ affects susceptibility to human serum and neutrophils.To compare the sensitivity to the serum's bactericidal effect between wild-type KJ and mutant KJΔYZ cells, serum killing assays were performed. Heat-inactivated serum was used as the control, and no killing effects were shown (data not shown). Killing of the KJΔYZ cells by nonimmunized healthy human serum was significantly more efficient than that of KJ cells at the designated time point (Fig. 5A).
Susceptibility to human serum and neutrophils. Data represent the means from 3 independent trials. Error bars represent the standard deviations. (A) Nonimmune healthy human serum sensitivity assays of the wild-type KJ and KJΔYZ strain. **, P < 0.0001; *, P < 0.001. (B) Bacterial susceptibilities to killing by human neutrophils of the wild-type KJ and KJΔYZ strains. Survival rates indicate the percentages of survival of wild-type or mutant strains calculated on the basis of viable counts relative to those for the no-neutrophil controls.
We then analyzed neutrophil-mediated killing of wild-type KJ and KJΔYZ cells. Compared to those for KJ cells, the survival numbers for KJΔYZ cells were markedly reduced after incubation with human neutrophils for 45 min (Fig. 5B). Thus, the SmeYZ pump of S. maltophilia confers protection again, not only for serum killing but also for neutrophil-mediated killing.
Loss of smeYZ affects virulence to mice.Finally, the virulence in vivo of KJ and KJΔYZ cells was compared using the murine model of septicemia generated by intraperitoneal injection (Fig. 6). Upon intraperitoneal infection of mice, KJΔYZ cells were less virulent than KJ cells. The survival rates of mice following infection with KJ and KJΔYZ strains were 22.2% and 77.8%, respectively (P = 0.0659).
Virulence to mice. Mouse lethality following challenges with wild-type KJ and KJΔYZ strains is presented. Male C57BL/6 mice (n = 9 from the summary of two independent experiments) were inoculated by intraperitoneal injection with 1 × 109 CFU of the wild-type KJ and KJΔYZ strains. Survival was assessed for 14 days following infection.
DISCUSSION
Crossman et al. have reported that SmeYZ contributes to the extrusion of aminoglycosides (15). In this study, we further demonstrated that SXT is also a substrate for the SmeYZ pump in addition to aminoglycosides. Historically, SXT is considered the first line of defense in S. maltophilia infection. The known underlying mechanism responsible for SXT resistance in S. maltophilia is the gain in the sul1 resistance gene of class 1 integrons and/or the sul2 resistance gene of the insertion element (34, 35). The finding in this article is the first report linking the RND-type efflux pump to SXT resistance in S. maltophilia.
A smaller single colony in LB agar was observed for KJΔYZ cells than for wild-type KJ cells (Fig. 1C), although the size of an individual KJΔYZ cell was similar to that of an individual KJ cell, as shown by the TEM observation (Fig. 3B). Furthermore, the hydrophobicity value of KJΔYZ cells was higher than that of KJ cells (Fig. 1D). These observations support the speculation that an unidentified substance, which is a physiological substrate of the SmeYZ pump, may act as the extracellular matrix affecting the bacterial surface hydrophobicity and/or the stacking of bacterial cells but not affecting the morphology of each cell. These results strongly suggest that the OD450 decline in KJΔYZ cells in the stationary phase (Fig. 1A) largely results from cell aggregation.
Cell aggregation generally correlates positively with biofilm formation. However, a contradiction occurs in KJΔYZ cells. An elevated hydrophobicity value (Fig. 1D) should confer KJΔYZ cells with a high tendency for aggregation; however, the biofilm formation of KJΔYZ cells was decreased (Fig. 4A). A possible explanation is that SmeYZ extrudes a substrate that compromises directly or indirectly the formation of the biofilm-associated extracellular matrix. This compound can be a component of such a matrix but can also be other substrates involved in biofilm formation such as quorum-sensing signals, among others.
For any infection, the prerequisite event is the encounter of the pathogenic bacteria with host cell surfaces. For this aspect, flagella can enhance the bacterial pathogenicity by providing bacterial motility (36, 37). In addition, a linkage between motility and biofilm formation has been reported (38–40). Given the fact that the smeYZ mutant totally abolishes flagella formation (Fig. 3B), the pleiotropic defects of mutant KJΔYZ in biofilm formation and virulence are rational.
Reactive oxygen species (ROS) are produced naturally during aerobic growth or by host immune cells during infection. Oxidative stress susceptibility is a critical determinant for bacterial survival either in physiological growth or during infection. In addition, S. maltophilia is frequently isolated from patients with cystic fibrosis (CF). CF lungs are known to be enriched in ROS (41, 42). In this study, we demonstrated that the smeYZ mutant is more susceptible to redox compounds (H2O2 and menadione) (Fig. 2) and human serum and neutrophils (Fig. 5). Therefore, the SmeYZ pump might play a critical role in the oxidative balance for normal physiology. Furthermore, during colonization or infection, SmeYZ can protect S. maltophilia from the attack of antimicrobial agents and ROS existing in human serum or neutrophils. The involvement of efflux pumps in the alleviation of oxidative stress has been reported in MacABC of Salmonella enteric serovar Typhimurium (43) and MacABCsm of S. maltophilia (30). Although the exact mechanism as to how SmeYZ alleviates oxidative stresses is not clear right now, the SmeYZ pump may export cellular components damaged by ROS or antioxidants to lower the amount of extracellular H2O2.
The involvement of efflux pumps in virulence has been reported in several microorganisms. The proposed mechanisms include the following: (i) efflux pumps contribute to the production of virulence factors (44, 45); (ii) efflux pumps can extrude host-derived antimicrobial molecules and alleviate the host-mediated attack (46, 47); and (iii) efflux pumps expel certain molecules, which extensively modulate bacterial physiology and thus indirectly affect virulence (48). In view of the pleiotropic phenotype change in KJΔYZ cells, aminoglycosides should not be the sole substrate for the SmeYZ pump. The contribution of SmeYZ to virulence should be attributed to extrusion of some unidentified substrates that are directly or indirectly involved in virulence. Further elucidation of the exact substrates of the SmeYZ pump and the underlying mechanisms responsible for virulence is worthwhile.
Clinical S. maltophilia isolates with different SmeYZ expression levels, from low to high, have been reported (21). Furthermore, the strain KJ used in this article is a clinical isolate from a patient without aminoglycoside treatment (22). These facts support the observation that SmeYZ overexpression has clinical relevance and imply that aminoglycoside extrusion seems not to be the significant outcome of SmeYZ overexpression. The impact of efflux pump overexpression on bacterial fitness has been reported, but no consistent conclusion has been made yet. SmeDEF overexpression in S. maltophilia has a fitness cost (49); nevertheless, MtrCDE overexpression in Neisseria gonorrhoeae confers increased fitness in vivo (50). The actual impact of SmeYZ overexpression on fitness has not been assessed. However, we demonstrated in this article that inactivation of an overexpressed SmeYZ pump reduces motility, biofilm formation, oxidative stress susceptibility, and virulence. Accordingly, SmeYZ is an attractive target for the development of new therapeutic strategies for S. maltophilia infection, such as an effective inhibitor(s) against the SmeYZ pump.
ACKNOWLEDGMENT
This work was supported by grant NSC 101-2320-B-010-053-MY3 from the Ministry of Science and Technology, Taiwan.
FOOTNOTES
- Received 13 February 2015.
- Returned for modification 6 March 2015.
- Accepted 20 April 2015.
- Accepted manuscript posted online 27 April 2015.
- Copyright © 2015, American Society for Microbiology. All Rights Reserved.
