Correspondence: Woojun Park, Department of Environmental Science and Ecological Engineering, Korea University, Seoul 136-713, Korea. Tel.: +82 23 2903067; fax: +82 29 530737; e-mail: firstname.lastname@example.org
Flavodoxin (Fld) is a bacterial electron-transfer protein that possesses flavin mononucleotide as a prosthetic group. In the genomes of the Pseudomonas species, the mioC gene is the sole gene, annotated Fld, but its function remains unclear. In this study, phenotype microarray analysis was performed using the wild-type and mioC mutant of pathogenic Pseudomonas aeruginosa PAO1. Our results showed that the mioC mutant is very resistant to oxidative stress. Different antibiotics and metals worked differently on the sensitivity of the mutant. Other pleiotropic effects of mutation in the mioC gene, such as biofilm formation, aggregation ability, motility and colony morphology, were observed under iron stress conditions. Most of the phenotypic and physiological changes could be recovered in the wild type by complementation. Mutation of the mioC gene also influenced the production of pigments. The mioC mutant and mioC over-expressed complementation cells, over-produced pyocyanin and pyoverdine, respectively. Various secreted chemicals were also changed in the mutant, which was confirmed by 1H NMR analysis. Interestingly, physiological alterations of the mutant strain were restored by the cell-free supernatant of the wild type. The present study demonstrates that the mioC gene plays an important role in the physiology of P. aeruginosa and might be considered as a suitable drug target candidate in pathogenic P. aeruginosa.
Flavodoxin (Fld) is a flavin mononucleotide-binding protein found mainly in prokaryotes (Sancho, 2006). Electrons flow from NADPH to Fld reductase and then to Fld in bacteria (Ceccarelli et al., 2004). In an effort to obtain insights into the molecular mechanism of the biological functions, several research groups have determined the solution structures of both the apo- and holo-forms of MioC (Hu et al., 2006; Sancho, 2006). Although these efforts provided insights into the mechanisms of the cofactor binding of MioC, redox partner interaction, and electron transfer mechanisms of Fld, the physiological function of MioC remains to be elucidated. Previously, we reported that Pseudomonas putida has just one Fld-encoding gene, whose homolog is annotated mioC in Escherichia coli (Yeom et al., 2009a). We also reported that the mioC gene product in P. putida interacts with ferredoxin (Fd) reductase as a preferred redox partner (Yeom et al., 2009a). The mioC gene was proven to be important for biotin synthesis in E. coli (Birch et al., 2000). However, the role of the mioC homolog in the physiology of the Pseudomonas species has never been addressed (Birch et al., 2000; Yeom et al., 2009a,b) and the PA3435 of Pseudomonas aeruginosa appears to the mioC homolog.
Pseudomonas aeruginosa is a ubiquitous environmental bacterium that is one of the top three causes of opportunistic human infections. Fds are most often involved in electron transfer roles in P. aeruginosa (Elsen et al., 2010). Functional substitution of Fd may occur with Fld (Sancho, 2006). Many sequenced bacterial genomes display a wealth of Fd genes, but fewer Fld are present. For example, the P. aeruginosa PAO1 strain has at least six genes encoding Fds, but only one Fld (PA3435) is present in its genome. It is often unclear which biological function relies on a given Fd and Fld. To elucidate the physiological function of the P. aeruginosa MioC, a phenotype microarray (PM) was performed with the wild-type and mioC mutant strains. Furthermore, we examined, for first time, the various physiologies of P. aeruginosa using the wild-type, mutant and complementation strains. Our data provide evidence that the mioC gene of P. aeruginosa is important in the response to antibiotic, metal and oxidative stresses. In addition, mutation in the mioC gene resulted in pleiotropic effects such as changes in cell aggregation, biofilm formation, pellicle production, motility and iron acquisition in P. aeruginosa.
Materials and methods
Bacterial strains, plasmids, and growth conditions
The wild-type and mioC mutant strains of P. aeruginosa PAO1 were purchased from Washington University Genome Center. Antibiotics (tetracycline, 20 μg mL−1; kanamycin, 100 μg mL−1) were added where necessary. The open reading frames of the mioC genes for the mioC over-expressed complementation stain were PCR-amplified using PAmioC-OE F (CGCAAGCTTAATGCCCGGCTTACCCCTGTTG)/PAmioC-OE R (CGCGGATCCCGTTATTCGCCCTACCGCTTGTCC) primer pairs. The amplified mioC gene fragments were cloned into the HindIII/BamHI sites of pBBR1MCS-2 to yield pBBR1-mioC. The pBBR1-mioC was transformed into E. coli Top10 via electroporation. Escherichia coli Top10 cells were grown with aeration at 37 °C in lysogeny broth (LB) medium supplemented with kanamycin (100 μg mL−1). The cells were harvested and pBBR1-mioC isolated using the MiniPrep plasmid purification kit (Takara). pBBR1-mioC was transformed into P. aeruginosa mioC mutant cell via electroporation. In the cases of the growth and lysis curves, cells were cultured with LB medium at 37 °C with aeration. The cell-free supernatants (CFS) were prepared by filtering a culture of each tested strain through a 0.22-mm pore-size filter (Sartorius). All chemicals were acquired from Sigma (Sigma Chemical, St. Louis, MO) unless otherwise stated.
Twenty 96-well PM plates (Biolog Co.) were used with the following nomenclature: metabolic panels, PM1–PM10; sensitivity panels, PM11–PM20. PM experiments were conducted according to the manufacturer's instruction. The cells were inoculated into the PM plates and incubated at 37 °C for 48 h. Cell growth was reflected by the development of purple coloration as monitored and recorded by OmniLog PM Station and PM Kinetic (Biolog). Further information on the compounds tested with the PM kit can be found at the Biolog website.
The wild-type, mioC mutant and the mioC over-expressed complementation cells were grown overnight in LB medium and diluted 100-fold with fresh LB medium with vigorous aeration. After the cells reached exponential phase (OD600 nm ~ 0.5), serially diluted cells were spotted on LB agar with paraquat, hydrogen peroxide (H2O2), cumen hydroperoxide (CHP), ampicillin (Amp), gentamicin (Gm), norfloxacin (Nor), 2, 2′-dipyridyl, arsenic (As), zinc (Zn), and copper (Cu). Spotted LB agar plates were grown at 37 °C for 1 day.
Biofilm formation assay
Polystyrene 96-well microtiter plates (BD Biosciences, San Jose, CA) were utilized as abiotic surfaces for biofilm formation study. Bacterial cultures were grown overnight, washed twice in phosphate-buffered saline and inoculated at 106 CFU mL−1 in LB broth with a variety of substrates. After 48 h of incubation at 30 °C, biofilm formation was determined via crystal violet staining and quantified by measuring the absorbance at 595 nm, normalized by the absorbance at 600 nm (Lee et al., 2010).
Colony morphology assay
Pseudomonas aeruginosa strains were grown in LB medium to exponential phase, then 10 μL were spotted onto 1% agar plates containing 10 g L−1 tryptone, supplemented with 40 μg L−1 Congo Red and 20 μg L−1 Coomassie Blue (Dietrich et al., 2008). Plates were incubated either with or without ferric chloride (1 mM) and 2, 2′-dipyridyl (0.5 mM) at 30 °C.
Aggregation and motility assay
Pseudomonas aeruginosa cells were grown in LB to exponential phase. The cells were incubated at 30 °C for 24 h without agitation. Subsequently, the aggregation percentage was obtained according to a previous report (Liu et al., 2008). For the swimming and swarming assay, 2 μL of cells grown overnight were inoculated in plates with modified M9 medium, [20 mM NH4Cl; 12 mM Na2HPO4; 22 mM KH2PO4; 8.6 mM NaCl; 1 mM MgSO4; 1 mM CaCl2 2H2O; 10 mM glucose; 0.5% casamino acids (Difco)] solidified with Bacto-agar (Difco; swimming 0.2%; swarming 0.5%) for 24 h at 30 °C.
Measurements of pyocyanin and pyoverdine
The pyocyanin assay is based on the absorbance of pyocyanin at 520 nm in acidic solution (Essar et al., 1990). A 5-mL sample of culture grown in LB was extracted with 3 mL of chloroform and then re-extracted into 1 mL of 0.2 N HCl to give a pink to deep red solution. The absorbance of this solution was measured at 520 nm. To measure pyoverdine production, bacteria were grown in LB to stationary phase and the absorbance of the culture supernatants was measured. Pyoverdine has a characteristic absorbance spectrum with a peak at 403 nm (Hohnadel et al., 1986). Spectral analysis of CFS was performed using an Optizen 2120 UV/VIS spectrophotometer (Mecasys, Korea).
1H NMR analysis of CFS
Cultures grown in LB underwent NMR analysis. After allowing the cultures to grow in 50 mL of LB medium (37 °C, 16 h, with agitation), the cells were harvested with centrifugation (4 °C, 1 h, 2800 g). Supernatants were lyophilized until they could be analyzed. All 1H NMR spectra were acquired on a Varian Inova 600-MHz NMR spectrometer (Varian) at ambient temperature. The NMR spectral data were reduced to 0.001 p.p.m. spectral buckets and the region corresponding to water (4.6–4.8 p.p.m.) was removed (Jung et al., 2012).
The mioC gene influenced various phenotypes and physiologies of P. aeruginosa
The PM assay of the mioC mutant was performed with chemicals using the Biolog system (Fig. 1a). Sensitivity to the antibiotics, metals and chelator was detected. Although some antibiotics and metals did not significantly affect the PM results, a mutant strain has resistance or sensitivity under various antibiotics and metals. Therefore, our data suggested that the mioC gene might be involved in antibiotic resistance and metabolism of metals in P. aeruginosa. Sensitivity tests were performed with the wild-type, mioC mutant and mioC over-expressed complementation strains using antibiotics, metals and chelator (Fig. 1b and c). Our laboratory experimental data were consistent with those of the PM assay. The mutant strain was resistant against oxidative stresses, including superoxide [paraquat (PQ)] and peroxide (H2O2 and CHP) stresses (Fig. 1b). The mutant strain was also resist to Amp and Gm antibiotics (Fig. 1b). However, the mioC mutant strain was sensitive to Nor, which is a synthetic fluoroquinolone-class antibiotic. In addition, the iron chelator 2, 2′-dipyridyl was able to kill the mioC mutant strain (Fig. 1b).
Subsequently, bacterial sensitivities were tested with three different metals: As, Zn and Cu (Fig. 1c). Consistent with the PM assay, the mutant was notably sensitive to As and Zn. Although Cu was not used in the PM assay, we performed the sensitivity test using Cu because it is known to promote cell death. However, the sensitivity of the mutant to Cu was not different from that of the other two strains. To summarize, we confirmed the results observed with the Biolog PM system using sensitivity tests.
Pleiotropic effect of the mioC mutant strain under iron stress
The mioC mutant strain displayed significant reductions in biofilm formation during static aerobic growth (Fig. 2a). Therefore, we thought that the mutant might be able to reduce cell aggregation of P. aeruginosa under biofilm conditions. Interestingly, aggregation of the mutant cell was reduced during static aerobic growth (Supporting Information, Fig. S1). Under iron excess condition, biofilm formations of the mutant and over-expressed complementation strains were reduced compared with that of the wild type (Fig. 2a and Fig. S2). Thus, the balance of the mioC gene product may be important for maintaining biofilm formation ability under iron excess condition. Interestingly, biofilm formation of the mutant was significantly induced by the iron chelator 2,2′-dipyridyl compared with the other two strains (Fig. 2a and Fig. S2). The growth mutant appeared to be slower under the iron chelator than was the wild type (Fig. 1b), whereas biofilm formation ability was enhanced by 0.5 mM dipyridyl (Fig. 2a and Fig. S2). No biofilm formation occurred in the absence of dipyridyl, but robust biofilm formation occurred in the presence of dipyridyl, which clearly demonstrated that dipyridyl treatment increased biofilm formation of the mutant (Fig. S2). In addition, biofilm formation was increased in the mioC mutant cell under Zn and As stresses (Fig. 2b). Consistent with sensitivity data, biofilm formation under Cu stress was similar to that under normal conditions (Fig. 2b). Subsequently, the colony morphology test was performed using Congo red and Brilliant blue (Fig. 2c–e). Congo red and Brilliant blue, a constituent of the agar used in the experiments, are known to bind the glucose-rich exopolysaccharide pellicle and proteins, respectively (Dietrich et al., 2008). Interestingly, red color formation was not observed in the mioC mutant strain, compared with the wild type under iron-rich conditions (Fig. 2d). Red color was recovered in the mioC over-expressed complementation strain under iron excess (Fig. 2d). However, this pellicle appeared in the mutant but disappeared in the other two strains under iron depletion (Fig. 2e). We also performed motility tests (Fig. S3). Interestingly, the swarming motility of the mioC mutant strain had a branch form. The mioC mutant cells may be more affected by iron depletion compared with other strains and they will search for iron in order to survive, thereby giving their motility form a unique structure. Taken together, we concluded that the mioC gene plays key roles in establishing biofilms, pellicle formation and motility under iron excess and depletion conditions.
Effect of mioC mutation on pigment production in P. aeruginosa
The mioC depletion and over-expression cells produced more pigments in LB medium (Fig. 3). In general, P. aeruginosa produce two types of pigment: the fluorescent pigment pyoverdine and the blue pigment pyocyanin (Youard et al., 2011). The latter is produced abundantly in low-iron content media and functions in iron metabolism and infection (Price-Whelan et al., 2007). To investigate pigment production, we performed pyocyanin and pyoverdine production analysis using the wild-type, mioC mutant and mioC over-expressed strains (Fig. 3a and b). Interestingly, mutant and over-expressed cells abundantly produced pyocyanin and pyoverdine, respectively, compared with the wild-type strain (Fig. 3a and b). Subsequently, absorbance scanning of CFS using a spectrophotometer was conducted (Fig. 3c). The absorbance spectra of mutant CFS indicated that the mioC mutant strain could produce plentiful pyocyanin (about 310 nm) compared with the wild-type strain (Fig. 3c; green arrow). Data of the mioC over-expressed strain suggested that cells could produce abundant pyoverdine (about 375 nm) compared with the wild-type strain (Fig. 3c; blue arrow). To determine the secreted chemicals of the mioC mutant, 1H NMR analysis was performed to compare the fresh LB growth medium with CFS from the wild type and mioC mutant (Fig. 3d). Some peaks appeared in the analysis of the wild-type CFS in the 2 p.p.m. region (Fig. 3d), whereas the mioC mutant CFS showed other patterns (Fig. 3d). Unfortunately, the actual compounds could not be identified in the NMR analysis. Our data showed that fine modulation of MioC amounts is important for pigment production, that the mioC gene might influence the production of various secondary metabolites, and that these changes might change the physiology in P. aeruginosa.
CFS changes colony morphology, biofilm and motility in P. aeruginosa
To investigation the secreted materials, we tested the physiological alteration using CFS of the wild-type and mioC mutant cells. Ten percent CFS of the wild-type and mioC mutant cells were used as a constituent of the medium. In the studies using the wild-type CFS, the colony morphology and pellicle formation of the mioC mutant cells were restored to wild type with the wild-type CFS (Fig. 4a). In particular, the mutant cells showed red pigment, which is pellicle extracellular polymeric substances (EPS), under iron excess. Therefore, secreted chemicals in the wild-type CFS may have stimulated production of pellicle in the mutant cells. We also performed the cell morphology test using CFS of the mioC mutant cells (Fig. 4b). The white region of colony of the wild-type and over-expressed cells slightly increased with the mioC mutant CFS. Interestingly, under iron depletion with 2,2′-dipyridyl (0.5 mM), the growth defect of the mioC mutant was not restored with the wild-type CFS, but the mioC mutant cells were very sensitive to iron depletion (Fig. 4b). Therefore, the mioC mutant cells may strongly inhibit iron acquisition with mutant CFS. We speculated that some chemicals of the wild-type CFS may have stimulated production of pellicle and that mutant CFS may have inhibited production of pellicle and iron utilization in P. aeruginosa. Subsequently, we performed biofilm assay using CFS of the wild-type and mutant cells (Fig. 4c). Interestingly, biofilm formation of the mioC mutant cells was induced by 10% wild-type CFS, a result that coincided with data of colony morphology (Fig. 4a and c). Therefore, the wild-type CFS may contain chemicals that can stimulate production of pellicle EPS and biofilm formation. The swarming motility test was conducted with CFS. Interestingly, the swarming motility using the mioC mutant CFS had a branch form in the wild-type and mioC over-expressed cells (Fig. S4). However, the swarming motility using the wild-type CFS was not changed (data not shown). Therefore, the wild-type and mioC over-expressed cells may have sensed the strong iron depletion and interfered with the iron utilization by mutant CFS.
Table 1. PM results of the mioC mutant and its wild-type strains in Pseudomonas aeruginosa PAO1
The OmniLog-PM software generates time course curves for respiration or growth and calculates differences in the areas under the curve for mutant vs. wild-type cells. The units are arbitrary. Positive values indicate that the mutant showed greater rates of respiration or growth than the parent strain. Negative values indicate that the mutant showed lower rates of respiration or growth than the parent strain.
Chemicals were tested in 96-well phenotypic microarray plates.
Fld has been found in prokaryotes of all major taxa (Zurbriggen et al., 2007). Fld is typically induced as an adaptive resource under environmental or nutritional hardships such as iron limitation (Zurbriggen et al., 2007). Interestingly, Fld expression confers tolerance to iron deficit and abiotic stress when introduced in plants (Zurbriggen et al., 2007). Therefore, Fld may be important to the resistance of various stresses in bacteria. We performed PM analysis to investigate the Fld function and our result suggested that the mioC gene mutation changed the physiology of P. aeruginosa in response to oxidative, metal and antibiotic stresses. Interestingly, the mioC mutant was significantly sensitive to norfloxacin and colistin, whereas the mutant was resistant to ampicillin, polymyxin B and gentamicin. Norfloxacin is a fluoroquinolone antibiotic and functions by inhibiting DNA gyrase (Leigh & Emmanuel, 1984). The mioC gene of P. aeruginosa was induced 1.5-fold under norfloxacin (GDS2317, GEO database), which suggested that the mioC gene might be important for defense against norfloxacin. Each antibiotic has a different mode of action. It remains unclear why different antibiotics work differently to the mutant sensitivity. Because the function of MioC has been characterized for first time, we believed that our global phenotypic analysis will be useful resource to the scientific field.
Our data demonstrated that Fld may be linked to biofilm, aggregation and motility under various stresses. It has been reported that flavodoxin gene was induced under biofilm condition of P. aeruginosa (Anderson et al., 2008). The flavodoxin A gene was also induced 5.3-fold in the biofilms of E. coli (Hancock & Klemm, 2007). It has been shown that biofilm formation and cell aggregation are connected (O'Toole & Kolter, 1998). Our data showed that the mioC mutant is defective in both biofilm formation and aggregation, which suggested that the mioC gene may be important for biofilm formation in P. aeruginosa, which is consistent with other reports. Interestingly, biofilm formation of the mioC mutant was boosted under iron depletion and some metal stresses. Fld has been shown to replace bacterial ferredoxin and this protein can enhance bacterial tolerance to iron starvation (Sancho, 2006). Therefore, the mioC gene mutant may feel stressed under iron depletion so that more biofilms are produced for their survival under this condition. Also, metals are known to induce oxidative stress in bacterial cell and bacterial Fld influences in the defense against oxidative stress (Imlay, 2006; Sancho, 2006). Thus, the mioC mutant is in danger under excess metal conditions and induces biofilm formation as a defense. It has been shown that motility is important for E. coli and P. aeruginosa biofilm formation (O'Toole & Kolter, 1998; Pratt & Kolter, 1999). Consistent with those data, we demonstrated that motility and biofilm formation were enhanced in the mioC mutant under iron-depleted conditions.
Pyocyanin has been reported to function as an electron shuttle for iron acquisition (Hernandez et al., 2004). Natural products such as pyocyanin may promote microbial metal reduction in the environment (Hernandez et al., 2004). In addition, pyocyanin alters the carbon flux of carbon metabolism (Price-Whelan et al., 2007). In this study, we suggested that the mioC mutant strain may be very sensitive to iron limitation, over-producing pyocyanin in response. The mutant cells were also sensitive to metal stresses. Therefore, the mioC mutant cell may recognize the deficiency of the reduced metal due to depletion of Fld, which functions as an electron donor in bacteria, and therefore produces pyocyanin to acquire metals from the environment. Interestingly, cell death after the stationary phase was accelerated in the mioC mutant cell, whereas there was no difference in exponential growth rate between the cells (wild type, 0.43 ± 0.04; ∆mioC, 0.41 ± 0.03; mioC OE, 0.41 ± 0.05) (Fig. S5). This means that pyocyanin-induced over-production of mutant may be able to promote cell death with redox imbalance, because pyocyanin generates reactive oxygen species that induce oxidative stress in bacteria (Hassan & Fridovich, 1980).
It has been proposed that the long-chain Flds may have preceded the shorter ones, such as MioC (Sancho, 2006). Interestingly, Fld is not present in higher eukaryotes and appears fused in multi-domain proteins of eukaryotes. Escherichia coli has some Fld in its genome; however, one Fld (MioC) is annotated in the Pseudomonas species chromosomes (Yeom et al., 2009a). Therefore, Pseudomonas species may be closer from an evolutionary perspective to eukaryotes than E. coli is, and MioC of the Pseudomonas species may play more important roles than do those of E. coli. In this work, we demonstrated that the mioC gene has functions related to biofilms, cell aggregation, motility, cell lysis and EPS production. As these physiologies may be important for P. aeruginosa virulence (Vasil & Ochsner, 1999; Shapiro et al., 2002; Rybtke et al., 2011), the mioC gene might be a useful therapeutic target for pathogenic bacteria.
This work was supported by the MEST/NRF program (grant # 2009-0076488) to W.P.