ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Aug. 2005, p. 3281–3288 0066-4804/05/$08.00⫹0 doi:10.1128/AAC.49.8.3281–3288.2005 Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Vol. 49, No. 8
Secretion of Proteases by Pseudomonas aeruginosa Biofilms Exposed to Ciprofloxacin Ewa Ołdak and Elz˙bieta A. Trafny* Department of Microbiology and Epidemiology, Military Institute of Hygiene and Epidemiology, Kozielska 4, 01-163 Warsaw, Poland Received 26 July 2004/Returned for modification 28 December 2004/Accepted 20 May 2005
Pseudomonas aeruginosa proteases are considered important virulence factors which damage host tissues and interfere with host antibacterial defense mechanisms. P. aeruginosa biofilm cells are not completely killed by antibacterials, and therefore this study addresses the question whether ciprofloxacin attenuates the virulence of biofilm communities by abolishing their secretion of proteases. The surviving cells of the colony biofilms studied, despite their cyclical exposure to four doses of ciprofloxacin at bactericidal concentrations (one dose a day), still secreted active proteases to the environment surrounding the biofilms. The biofilm cells secreted elastase B (LasB) over the duration of the experiments as confirmed by Western immunoblot analysis. The colony biofilms did not secrete LasA—a protease with staphylolytic activity. The same profiles on zymogram gels with gelatin were observed for the proteases secreted by both ciprofloxacin-exposed and unexposed (control) biofilms. Total proteolytic activities of the colony biofilms studied were significantly reduced after exposure to ciprofloxacin at bactericidal concentrations—after 96 h of exposure they dropped to 38% for the strain intermediate resistant to ciprofloxacin and to 65% for the strain highly resistant to the antibiotic, relative to the control biofilms. The surviving cells of the colony biofilms after their release into a fresh medium displayed transient increased resistance to ciprofloxacin compared to their planktonic counterparts. Pseudomonas aeruginosa is an opportunistic pathogen. However, serious pneumonias and sepsis due to an infection with this organism may be life threatening for patients with burns or cystic fibrosis and also for immunocompromised individuals. In patients with persistent lung infections, P. aeruginosa has been shown to grow as a biofilm (19). The reduced susceptibility to antibiotics displayed by P. aeruginosa biofilms, compared with that exhibited by the bacterial cells in liquid cultures, is a well-known phenomenon (9, 11, 12). A complete characterization of the cells from many bacterial species that survived treatment with an antibiotic has not yet been accomplished. In particular, it has not been answered, until now, whether the surviving cells are in a dormant state (31) or, quite the reverse, whether the metabolic processes of survivors have not been arrested upon their exposure to antibiotic (14), and these organisms display other physiological features that enable them to persist in the presence of antibiotic. Some antibiotics in concentrations below the MIC for the organisms also affect the synthesis of proteins (4). The production of some of the proteins may even be upregulated in the presence of ciprofloxacin at sublethal concentrations (3). Assuming that the physiology of the surviving cells is not somehow changed by ciprofloxacin to that of a completely nonvirulent phenotype, the proteases secreted by P. aeruginosa biofilms, despite the antimicrobial treatment, may still damage mammalian matrix proteins. Hence, we intended to characterize the production of proteases by the cells within P. aeruginosa
biofilms that survive treatment with ciprofloxacin at bactericidal concentrations. In this study, the secretion of proteases by the surviving cells in the presence of the antibiotic is demonstrated. Furthermore, an increased resistance of these cells to ciprofloxacin is shown for 10 passages in the antibiotic-free medium. MATERIALS AND METHODS Bacterial cells and culture. Pure cultures of two P. aeruginosa strains, 1159 (PA1159) and 1230 (PA1230), were used. Both strains were nonmucoid burn wound isolates from the bacterial collection of our department. Mucoidity of the strains was assessed on Pseudomonas Isolation Agar (Difco). Stock cultures were maintained at ⫺70°C in 50% skim milk supplemented with 0.7% NaCl. Selection of medium for colony biofilm culture. For each experiment, the strains were grown overnight at 37°C in Luria-Bertani broth (LB broth; Sigma). Then, the 0.22-m-pore-size filters (MF-Millipore membrane filters, consisting of mixed cellulose esters, 47-mm diameter) were seeded with 200 l of an overnight culture. The mean cell numbers in the inocula were 8.89 ⫾ 0.33 log10 CFU for PA1159 and 8.4 ⫾ 0.15 log10 CFU for PA1230. The number of bacterial cells was estimated by plate count. The membrane filters were placed on the nutrient agar (Difco) and incubated at 37°C for 24 h. The membrane filters, with the 24-h-old colony biofilms oriented upwards, were transferred to the following media: (i) chemically defined (CD) liquid medium poured (4 ml) into petri dishes (dimensions, 60 ⫻ 15 mm; Sigma), (ii) LB broth poured (4 ml) into petri dishes, (iii) CD medium solidified with 1.5% agar, and (iv) LB agar. The CD medium (pH 7.4) consisted of glucose (30 mM), NaCl (8 mM), K2HPO4 (60 mM), KH2PO4 (35 mM), ZnCl2 (0.025 mM), (NH4)2SO2 (15 mM), L-glutamine (7 mM), CaCl2 (0.05 mM), FeCl3 (0.017 mM), C6H5Na3O7 (35 mM), MgCl2 (1.4 mM), thiamine (0.15 mM), DL-arginine (0.22 mM), uracil (0.2 mM), and nicotinic acid (0.1 mM). The medium was sterilized using filters (pore size, 0.22 m; Millipore). The bacterial colonies on the membrane filters were incubated at 37°C for 48 h. The media were exchanged for fresh media every 24 h. Enumeration of viable cells in colony biofilms. The individual membrane filter, with colonies that had been accumulated, was put into the flask containing 20 ml of NaCl. The colony biofilm was taken off the membrane by pipette suction. The membrane filter was flushed several times with the saline in the flask. Then, the filter and the bacterial suspension were vigorously vortexed. The organisms were serially diluted, and the number of viable bacterial cells was estimated by plate counts.
* Corresponding author. Mailing address: Department of Microbiology and Epidemiology, Military Institute of Hygiene and Epidemiology, Kozielska 4, 01-163, Warsaw, Poland. Phone: (48) 22 685-32-06. Fax: (48) 22 838-10-69. E-mail:
[email protected]. 3281
3282
OŁDAK AND TRAFNY
Determination of biofilm wet weight. The wet weight of the biofilms was determined gravimetrically in each experiment. Three membrane filters with biofilms formed by the strains analyzed were weighed, and the average weight was calculated (after subtraction of the mean filter weight). Culture of colony biofilm on liquid media. Taking into consideration the appearance, higher wet weight, and lower number of viable cells of the colony biofilms grown on liquid media (dome-like smooth colonies) compared to the biofilms grown on solid media (flat and rough colonies), we chose to grow the colony biofilms on liquid media in our experiments, as follows: the 24-h-old colony biofilms (prepared as described above) on their membrane filters were placed onto a surface of CD liquid medium and then incubated at 37°C for 96 h. Every 24 h, the CD media underneath the membrane filters and the membranes with biofilm bacteria were collected. Exposure of colony biofilms to ciprofloxacin. The 24-h-old colony biofilms were exposed to ciprofloxacin at a concentration of 4 g per ml (10). Ciprofloxacin (Ciprobay 100) was purchased from Bayer as an intravenous solution (2 mg/ml). Strain PA1159 was determined as intermediate resistant to ciprofloxacin (MIC, 2 g/ml) according to the CLSI (formerly NCCLS) breakpoints (24), and the ciprofloxacin concentration of 4 g per ml corresponded to two times the MIC and was equal to the minimal bactericidal concentration. Strain PA1230 displayed a much higher level of resistance to ciprofloxacin (MIC, 32 g/ml), and the investigated concentration of the drug was eight times lower than the MIC. Thus, PA1230 biofilms were exposed to ciprofloxacin at a concentration of 64 g/ml (corresponding to two times the MIC and equal to the minimal bactericidal concentration) to compare the influence of suprainhibitory concentration of the antibiotic on the biofilms formed by the highly resistant strain. The maximum exposure time for the biofilms was 96 h. Preparation of media collected from underneath the colony biofilms. The sterility of collected media was assessed each time by plating (on nutrient agar). Then, the CD medium was desalted by extensive dialysis versus phosphatebuffered saline, pH 7.4, and concentrated four times with Aquacide II (Calbiochem) at a temperature of 4°C. The CD medium after concentration (CDMC) was assayed for protein concentration and protease activity as described below. Estimation of protein concentration and total protease activity. Protein concentration was measured in CDMC every 24 h with the Bio-Rad protein assay (Bio-Rad). Total protease activity was estimated using azocasein as substrate (28). The absorbance of the samples was read at A440 in the wells of the microtiter plates, using a microplate reader (Bio-Rad). Zymography. Proteases secreted by biofilm-grown P. aeruginosa were resolved in 7.5% sodium dodecyl sulfate (SDS)-polyacrylamide gels containing the following substrates for P. aeruginosa proteases: 0.1% gelatin (Sigma), 0.1% ␣-casein (Sigma), and 1% proteoglycans (isolated from chicken cartilage [13]). Samples (20 l) of CDMC were mixed with a sample buffer without reducing agents. Samples were not boiled. SDS-polyacrylamide gel electrophoresis was carried out in a Mini-Protean II apparatus (Bio-Rad). The proteins were electrophoresed under standard conditions. After electrophoresis, SDS was removed from the gels by immersion in a solution containing 2.5% Triton X-100 (Sigma), and then the gels were incubated overnight at room temperature in 50 mM Tris-HCl (pH 8.0), supplemented with CaCl2 (1 mM), ZnCl2 (0.001 mM), and NaCl (150 mM) (33). The protease activity was visualized by staining the gels with Coomassie brilliant blue (Bio-Rad). The characteristic protease profile (number of bright bands and their location in the gel) for each strain was recorded using a Gel-DOC 1000 apparatus (Bio-Rad). The densitometric analysis of the bright bands in the stained gels was performed with MultiAnalyst software, version 1.1 (Bio-Rad). To determine the protease profiles, at least three polyacrylamide gels were run in each experiment. Immunodetection of elastase. Samples (20 l) of CDMC were electrophoresed, either in 10% SDS-polyacrylamide gels under reducing conditions or in 7.5% SDS-polyacrylamide gels under nonreducing conditions, in a Mini-Protean II apparatus (Bio-Rad) under standard conditions. After electrophoresis, the proteins were electroblotted onto nitrocellulose membranes (Millipore) for 1 h at 120 V using a Mini Trans Blot apparatus (Bio-Rad). Immunodetection of elastase was performed with rabbit polyclonal antibodies against P. aeruginosa elastase followed by donkey anti-rabbit immunoglobulin G-horseradish peroxidase conjugate (Jackson ImmunoResearch). The polyclonal antibodies against elastase were a gift, generously donated by J. Fukushima. The immunoblots were developed using SuperSignal West Pico chemiluminescent substrate (Pierce) and imaged with X-ray film (Roche), followed by quantification by densitometry. Detection of proteases in a biofilm matrix. The colony biofilms were harvested, and the bacterial suspensions were stirred on a magnetic stirrer for 2 h at 4°C to separate exopolysaccharides from the biofilm cells (1). Of this suspension 1.5 ml was centrifuged for 20 min at 10,000 rpm. The supernatants were collected and
ANTIMICROB. AGENTS CHEMOTHER. sterilized using filters. The presence of active proteases in supernatants was determined by zymography. Staphylolytic activity assay. To estimate the activity of LasA protease, the following assay was performed: an overnight culture of Staphylococcus aureus ATCC 25923 was boiled for 10 min and centrifuged for a further 10 min at 14,000 rpm. The pellet was resuspended in 10 mM phosphate buffer, pH 7.5, to an optical density at 595 nm of 0.8. A 100-l volume of CDMC was added to 900 l of the suspension of the bacterial cells. Staphylolytic activity of the samples was determined spectrophotometrically by monitoring the decrease in the absorbance at 595 nm. The absorbance was recorded every 5 min for 1 h (7). Determination of MICs for surviving cells. Colony biofilms of PA1230 and PA1159 were exposed to ciprofloxacin for 96 h. The bacteria were removed from membranes every 24 h during this period, resuspended 200-fold in 0.9% NaCl, and adjusted to a concentration of 1 ⫻ 107 CFU/ml. Then, 5 l of bacterial suspension was transferred into the wells of microtiter plates and treated with ciprofloxacin, which was serially diluted by a factor of two in Mueller-Hinton broth (the total volume in the well was 100 l). The plates were incubated at 37°C for 20 h. The amount of bacterial growth inside the wells was determined at 595 nm using a microplate reader (Bio-Rad). The optical density below 0.1 at 595 nm in the wells with the antibiotic was determined as the end point of the bacterial growth, confirmed by visual inspections. The organisms taken from PA1230 colony biofilms exposed for 48 h to sub- or suprainhibitory concentrations of antibiotic were also regrown in fresh LB medium. Fifteen passages were performed for each strain, and a determination of the MICs was repeated every passage. Each experiment was performed at least three times. Statistical analysis. Data were analyzed using unpaired Student’s t test to determine whether the mean values of protein concentration, protease activity, and elastase activity of the biofilm bacteria were different from the corresponding values of the control biofilm bacteria, not exposed to antibiotics. Statistical significance was determined when P was ⱕ0.05.
RESULTS Effects of ciprofloxacin on wet weight and viability of bacteria in P. aeruginosa colony biofilms. The average number of viable bacteria within 24-h colony biofilms of PA1230 was 9.94 ⫾ 0.22 log10 CFU per membrane and did not significantly change during culture on CD medium—an average number of 9.9 ⫾ 0.21 log10 CFU per membrane was observed after 96 h (Fig. 1A). Ciprofloxacin at a concentration of 4 g/ml (equal to one-eighth the MIC) had a negligible effect on the viability of cells within biofilms formed by this strain. However, when the colony biofilms of PA1230 were exposed to ciprofloxacin at a concentration of 64 g/ml for 96 h (two times the MIC), the number of viable cells decreased to 8.9 ⫾ 0.22 log10 CFU, as observed on the membrane filters (a reduction of 90.4%). After 96 h of development on nutrient agar, the average number of bacteria within the colony biofilms of PA1159 was 10.89 ⫾ 0.65 log10 CFU per membrane (Fig. 1B). After 96 h of growth in the presence of the antibiotic the viable biofilm cells experienced a log reduction of 3.0 compared to the control biofilms. These results indicate that the substantial majority of the cells (99.98%) in the biofilm communities were killed by ciprofloxacin. To observe whether the population of the surviving cells was able to regrow when the antibiotic had lost its activity, another set of experiments was performed. The colony biofilms of PA1159 were exposed only once to ciprofloxacin (at a concentration of 4 g per ml of CD medium) and cultured for 96 h. The control biofilms were grown on CD medium, which had not been exchanged. After 24 h of exposure, the biofilms of PA1159 showed a 3.12-log (an average reduction of 99.25%) decline in viable cells compared with the control biofilms (Fig. 1C). During a continuous period of culture for an additional 72 h, the number of viable cells increased and by day 3 was
VOL. 49, 2005
P. AERUGINOSA BIOFILMS AND PROTEASES
3283
FIG. 2. An average wet weight of the colony biofilms formed on filter membranes by PA1159 (A) and PA1230 (B) strains upon exposure to ciprofloxacin. The colony biofilms were exposed to subinhibitory (ƒ) or suprainhibitory (E) concentrations of ciprofloxacin in CD medium. F, an average wet weight of the control biofilms. The results are expressed as means with standard deviations of three independent experiments. *, P ⱕ 0.005.
FIG. 1. Killing of bacteria within the colony biofilms of PA1230 (A) and PA1159 (B) strains by ciprofloxacin at a concentration of 4 g (ƒ) or 64 g (■) per ml of CD medium. The CD medium was exchanged for a fresh medium containing antibiotic every 24 h. In the same periods of time growth of the control biofilms (F), not exposed to antibiotic, was monitored. (C) The colony biofilms of strain PA1159 were exposed to ciprofloxacin at a concentration of 4 g per ml (}) of CD medium only once and the medium was not exchanged. E, growth of the control biofilms on CD medium that was not exchanged for 120 h. The arrows indicate the point in time of exposure to ciprofloxacin. The values are averages of the results from three independent experiments each performed in three replications. *, P ⱕ 0.005.
comparable to that of the control biofilms despite the lack of a supply of fresh nutrients. After 96 h of growth on CD medium, the wet weight of the colony biofilms of PA1159 and PA1230 reached 0.537 ⫾ 0.1 g and 0.490 ⫾ 0.14 g per membrane, respectively (Fig. 2A and B). Ciprofloxacin exposure had a significant influence on the wet weight of the biofilms. Treatment of the PA1159 and PA1230 biofilms with ciprofloxacin at a concentration equal to two times the MIC reduced the wet weight of the biofilms by 46% and 34%, respectively. However, exposure of these biofilms to ciprofloxacin at a subinhibitory concentration (4 g/ ml) had no effect on their wet weight. The surviving cells secrete active proteases. P. aeruginosa grown in liquid cultures secretes several proteases. Zymogra-
3284
OŁDAK AND TRAFNY
FIG. 3. Analysis of proteases secreted by the colony biofilms of PA1159 and PA1230 strains. The biofilms were grown for 72 h, and the media underneath the biofilms (CDMC) were prepared for analysis as described in Materials and Methods. (A) CDMC samples from the cultures of PA1159 (lane 1) and PA1230 (lane 2) biofilms were analyzed by gelatin zymography. (B) Western blot analysis of nondenatured CDMC samples from cultures of PA1159 (lane 1) and PA1230 (lane 2) biofilms was performed with anti-P. aeruginosa elastase antibodies. The zymograms and the blots are typical representatives of at least three independent experiments.
phy using gelatin as the substrate enables the detection of the activity of protease IV, elastase B (LasB), elastase A (LasA), and alkaline protease of this organism (6). In this study, active proteases were observed in zymograms of the nondenatured samples of CDMC from the cultures of the colony biofilms of PA1159 and PA1230 strains. Two different profiles of the bright bands were observed in the gels. Profile 1 was specific for PA1159 biofilm and consisted of two bands of 55 and 158 kDa in molecular mass. Profile 2 was specific for PA1230 biofilm and consisted of three bands corresponding to molecular masses of 55, 116, and ⬎200 kDa (Fig. 3A). The 55-kDa band was detected only in the samples of CDMC after 24 h of culture on this medium. Blots of nondenatured proteins of the CDMC samples demonstrated that after the reaction with rabbit antielastase antibodies the bands appeared at the same positions as the bands migrating at 116 and 158 kDa in zymograms (Fig. 3B). The immunoblots of denatured proteins showed immunological reaction for the bands migrating at 33 kDa (data not shown), exactly at the same position as the LasB bands observed by others (29). In zymograms containing casein and proteoglycans, proteases migrated at the same positions as in zymograms containing gelatin. Caballero and coworkers (6) showed that the bands at 158 kDa in the gels with casein represented the proteolytic activity of LasB, although in the gels with gelatin the bright bands at 158 kDa represented the activity of LasA and LasB, simultaneously. The protein bands of 116 and 158 kDa were quantified in the polyacrylamide gels containing gelatin and casein by densitometric analysis, and the results were compared. There were no significant differences in the relative activities of these proteins in the gels (data not shown). To verify whether P. aeruginosa biofilms secrete LasA, the staphylolytic activity of the CDMC samples was assayed spectrophotometrically using a heat-killed Staphylococcus aureus suspension. No staphylolytic activity was de-
ANTIMICROB. AGENTS CHEMOTHER.
tected in these samples. Thus, we concluded that LasA is not secreted by PA1159 and PA1230 biofilms, and thus, the bright bands at 116 and 158 kDa in Coomassie blue-stained gels represented only the activity of LasB of P. aeruginosa. We quantified the amounts of LasB by densitometric analysis of gelatin zymograms, because these results were much more reproducible in the repetitious experiments than the results of the elastin-Congo Red assay, generally used to quantify the elastase activity (26). It has been postulated that the cells that survive antibiotic treatment are metabolically quiescent (31). Therefore, we investigated whether the cells that survived exposure to ciprofloxacin still retain their capacity to release proteases. The biofilms of PA1159 and PA1230 strains, exposed to ciprofloxacin, persistently secreted proteases, which in gelatin zymograms demonstrated the same profiles as those observed for proteases released by the corresponding control biofilms (data not shown). The only differences between CDMC samples from the control and the antibiotic-treated biofilms were in width and intensity of the clearance zones in the Coomassie blue-staining gels. We found that the bands corresponding to elastase activity were present in each sample of both the control and exposed biofilms, and the width and intensity of these bands were most variable over the course of the experiments. The bands corresponding to molecular masses of ⬎200 kDa were found to display the same width and intensity in all CDMC samples from PA1230 biofilms. The width and intensity of the bands corresponding to LasB (band at 158 kDa for PA1159 and band at 116 kDa for PA1230) were quantified with Multi-Analyst software and plotted versus time in Fig. 4. Exposure of PA1159 colony biofilms to ciprofloxacin at suprainhibitory concentrations resulted in suppression of the gelatinase activity of LasB, compared to the control biofilms. A similar drop in gelatinase activity of LasB was observed for PA1230 biofilms at sub- and suprainhibitory concentrations of ciprofloxacin used for treatment. Surprisingly, PA1230 biofilms exposed to ciprofloxacin at a concentration corresponding to two times the MIC secreted significantly more protease on the third and fourth day of the experiment than in the first 2 days. Gelatin zymography of the CDMC samples from the control biofilms of PA1230 showed a progressive increase in LasB activity over the duration of the experiment as opposed to the constant activity of LasB in the CDMC samples observed for PA1159 control biofilms. Apart from the assays of proteolytic activity performed with the zymography technique, we also quantified total proteolytic activity of the CDMC samples using azocasein as a substrate (Fig. 5). Total protease activity in the CDMC samples from PA1159 and PA1230 control biofilms showed an increase of the biofilms’ growth over time; however, PA1230 biofilms produced more proteolytic activity than PA1159 biofilms. Ciprofloxacin at subinhibitory concentrations inhibited proteolytic activity of PA1230 biofilms only after a long exposure to the antibiotic (96 h). Total proteolytic activity of PA1159 and PA1230 biofilms was significantly reduced after 48 h of exposure to ciprofloxacin at suprainhibitory concentrations relative to the control biofilms. After 96 h of exposure of PA1159 and PA1230 biofilms to ciprofloxacin, proteolytic activity in the CDMC samples dropped to 38% and 65%, respectively, compared to the control biofilms.
VOL. 49, 2005
P. AERUGINOSA BIOFILMS AND PROTEASES
3285
FIG. 4. Time course of the elastase in zymograms of the CDMC samples from the cultures of PA1159 (A) and PA1230 (B) colony biofilms. F, control biofilms, not exposed to ciprofloxacin; ƒ, biofilms exposed to subinhibitory concentration of the antibiotic; E, biofilms exposed to suprainhibitory concentrations of the antibiotic. Experiments were repeated at least three times, and the results are expressed as means ⫾ standard deviations. , P ⱕ 0.05; *, P ⱕ 0.001 compared to the control biofilms.
FIG. 5. Time plot of total proteolytic activity of the CDMC samples from the cultures of PA1150 (A) and PA1230 (B) biofilms grown in the presence of ciprofloxacin at suprainhibitory (E) or subinhibitory (ƒ) concentrations. Black dots (F) indicate total proteolytic activity of the control biofilms not exposed to ciprofloxacin. Shown are the averages ⫾ standard deviations of at least three experiments, each performed in , P ⱕ 0.05; *, P ⱕ 0.001 compared to CDMC samples triplicate. from the cultures of control biofilms.
To address the question whether the contribution of proteases in total proteins secreted by the colony biofilms exposed to ciprofloxacin is altered relative to the control biofilms, total activity of proteases in the azocasein assay and the activities of elastase in zymograms were both normalized by dividing values for optical density at 440 nm and the areas of bands (mm), respectively, by protein concentrations (mg) in corresponding samples of CDMCs. The results were expressed as a percentage of their counterparts calculated for the control biofilms. The results obtained for the control biofilms were considered 100%. Surprisingly, as shown in Fig. 6A an enormous increase in relative elastase activity was observed after 48 h and 72 h of exposure of PA1159 colony biofilms to ciprofloxacin at a concentration equal to two times the MIC. This phenomenon was observed only for PA1159 biofilms. Relative elastase activity in CDMC from PA1230 biofilms exposed to supra- and subin-
hibitory concentrations of antibiotic was diminished compared to the control biofilms (Fig. 6B and C). Relative total proteolytic activity for the CDMC samples from the antibiotic-exposed PA1230 biofilms was lower than in corresponding samples of CDMC from the control biofilms when ciprofloxacin was used at a suprainhibitory concentration. Relative total proteolytic activity for the CDMC samples from the antibioticexposed PA1159 biofilms was at the same level as in the control biofilms for 3 days of the consecutive treatment with ciprofloxacin and dropped below this level by day 4. The wet weight of the biofilms cultivated using our model significantly increased over the duration of the experiments. Since the number of viable cells in the biofilms did not increase significantly, we concluded that an extracellular matrix was accumulated in the biofilms. The part of the proteases produced by biofilm cells might be trapped by the extracellular
3286
OŁDAK AND TRAFNY
ANTIMICROB. AGENTS CHEMOTHER. TABLE 1. Susceptibility of the surviving cells to ciprofloxacin shown as multiples of the MICs for planktonic P. aeruginosa cells Multiples of the MICa Duration of exposure to ciprofloxacin (h)
24 48 72 96
PA1230 biofilmsb 1/8 the MIC
2 times the MIC
1 4 4 4
4 4 4 8
PA1159 biofilms,c 2 times the MIC
2 4 4 8
a
Values estimated for planktonic cells. The colony biofilms exposed to sub- and suprainhibitory concentration of ciprofloxacin. c The colony biofilms exposed to suprainhibitory concentrations of ciprofloxacin. b
matrix. We assessed the presence of protease activity inside the matrix by running the zymogram gels with extracellular matrix samples, devoid of the bacterial cells after centrifugation and filtration. The zone of clearing in the zymogram is proportional to the amount of the enzyme loaded on the gel (21). Densitometric analysis of these zymograms reveals semiquantitatively that the proteolytic activity of samples of extracellular matrix has been at a level comparable to that in the CDMC samples (data not shown). It indicates that extracellular proteins diffuse off the extracellular matrix. Similar diffusion has been shown previously for beta-lactamase (5). Thus, we believe that the proteolytic activities in CDMC reflected the proteolytic activities of P. aeruginosa colony biofilms. MICs of ciprofloxacin for surviving cells. The cells recovered from the biofilms of PA1230 and PA1159 strains, not exposed to ciprofloxacin, experienced complete growth inhibition at the same concentration of antibiotic as planktonic cells, independently of the day of sampling (1 to 4 days). As shown in Table 1, the organisms recovered from the PA1159 and PA1230 biofilms exposed to ciprofloxacin required a severaltimes-higher concentration of ciprofloxacin for the inhibition of growth than did planktonic cells. To examine whether the surviving cells from the biofilms treated with antibiotic would be able to change to planktonic phenotype and become sensitive to ciprofloxacin again, the surviving cells from the PA1230 biofilms exposed to sub- and suprainhibitory concentrations for 48 h were passaged in antibiotic-free broth for 15 passages. The cells that survived treatment with subinhibitory concentrations of ciprofloxacin recovered their susceptibility to the antibiotic, similarly to planktonic cells after three passages, while the cells that survived treatment of the biofilms with suprainhibitory concentrations of ciprofloxacin did not recover their initial susceptibility to the antibiotic until the 10th passage in fresh medium without ciprofloxacin (data not shown). FIG. 6. Total protease (F) and elastase (ƒ) activities normalized to the protein concentration in the corresponding CDMC samples, relative to the control biofilm value, which was taken as 100%. (A) PA1159 biofilm exposed to ciprofloxacin at two times the MIC; (B) PA1230 biofilms exposed to ciprofloxacin at two times the MIC; (C) PA1230 biofilms exposed to ciprofloxacin at one-eighth the MIC. Experiments were performed in three repetitions, and the means ⫾ standard deviations are shown.
DISCUSSION A colony biofilm model of growth has been chosen for carrying out our experiments. The colony biofilms have been well characterized by others (15, 35, 36). In this model of biofilm growth, the nutrients diffuse into the colony from the bottom while oxygen diffuses from the top. Colony biofilms develop increased antibiotic resistance similarly to highly structured
VOL. 49, 2005
biofilms (23). In the present experiments, we were interested in the response of the entire biofilm to the treatment with bactericidal agent. Studies by others have shown that ciprofloxacin was penetrating through the full extent of the colony biofilm and was acting on the bacteria at the distal edge (34). Thus, the colony biofilm model perfectly fits the goals to be achieved in this study. Ciprofloxacin has been selected for the experiments as it is a drug equally active against P. aeruginosa in the logarithmic and stationary phases of growth, i.e., the bacteria in a low metabolically active state (2). Our model of biofilm growth enables the development of wet and smooth colony biofilms that resemble the biofilms in natural settings. The production of the extracellular polysaccharide matrix is a hallmark of P. aeruginosa biofilms, and it has been observed in our study. Treatment of such biofilms with ciprofloxacin at concentrations equal to two times the MIC for 24 h produced a reduction of 98.5% and 94% in the number of viable cells for intermediate (PA1159) and highly resistant (PA1230) P. aeruginosa strains, respectively. Most of the biofilm cells were killed by antibiotic, but at the same time millions of them were still alive. These findings support most of the results reviewed recently by Lewis (22) on the production of persister cells by a biofilm population as a response to the treatment with antibiotics. However, it is still poorly understood whether physiology of survivors is different from that of isogenic planktonic cultures and intact biofilms, though it is obvious that survivors are able to grow in fresh medium without the antibiotic. We believe that the biofilm population of cells secretes the active proteases at least to gain a fresh supply of nutrients (32). This ability would likely ensure survival of the bacterium in human tissues during chronic infections. P. aeruginosa in liquid cultures produces and secretes a number of proteases: elastase A (LasA), elastase B (LasB), protease IV, and alkaline protease (6). The expression of protease activity is not restricted by the mucoid phenotype of planktonic P. aeruginosa strains from cystic fibrosis patients, as 56% of mucoid strains secreted LasB and 59% of these strains secreted alkaline protease (17). Host response to P. aeruginosa proteases increases in chronically infected patients year by year, indicating that these proteins are secreted most of the time (8). LasB and alkaline protease have been found in the sputa of cystic fibrosis patients, and the elevated levels of these proteases were correlated with the poor clinical conditions of the patients (16). An enhanced release of LasB in microaerobic conditions of culture, which are specific conditions of growth inside P. aeruginosa biofilms, has been observed by Sabra and coworkers (27). In our experiments, both colony biofilms of PA1159 and PA1230 strains secreted proteases over 4 days of experiment. The CDMC samples recovered from under the membranes with biofilms of two P. aeruginosa strains exhibited two different patterns of exoprotease activity, as has been observed in zymogram gels. Similar patterns have also been observed for the ocular P. aeruginosa isolates, which are classified into three groups based on their protease profiles in gelatin gels. The protease profiles observed for PA1159 and PA1230 biofilms closely resemble the profiles of the ocular isolates (37). Sub- and suprainhibitory concentrations of various antibiotics may markedly suppress or augment the secretion of virulence factors by bacteria. Ciprofloxacin inhibits DNA gyrase
P. AERUGINOSA BIOFILMS AND PROTEASES
3287
and therefore disrupts the process of maintaining the proper DNA topology. By this process, it may interfere in the regulatory mechanisms that control extracellular enzyme production (30). Fluoroquinolones at a level even below the MIC are still capable of affecting these topological characteristics of DNA (4). To our knowledge, the secretion of proteases by P. aeruginosa biofilms upon treatment with ciprofloxacin has not previously been evaluated in vitro or in vivo. In this study, the exposure of the colony biofilms to ciprofloxacin has not changed their protease profiles in zymograms. The surviving cells within PA1159 and PA1230 biofilms secrete two or three proteases, respectively. Both strains grown as biofilms secrete LasB, which is considered to be an important P. aeruginosa virulence factor. LasB possesses the ability to degrade extracellular matrix components, such as collagen III and IV, elastin, and proteoglycans as well as host matrix protease inhibitors (20). In this study, the relative amount of LasB increased starting from the second day of ciprofloxacin exposure of the biofilm formed by the PA1159 strain, which is intermediate resistant to the antibiotic. LasB and another protease with a molecular mass above 200 kDa are secreted by the colony biofilms even 96 h following their first exposure to ciprofloxacin. However, the relative amount of total proteases secreted by PA1230 and PA1159 biofilms exposed to ciprofloxacin was lower than or remained at the same level as, respectively, that in the control biofilms not exposed to the drug. Thus, the secretion of total proteases by P. aeruginosa biofilms exposed to ciprofloxacin at pharmacologically allowable concentrations is not elevated in the resistant population of cells within the biofilms. However, ciprofloxacin at this concentration augments the secretion of elastase by the biofilms formed by the intermediate resistant P. aeruginosa strain. In our studies, the bacterial cells taken from 72-h-old biofilms after their exposure to ciprofloxacin for 48 h did not display fully restored sensitivity to the antibiotic. The level of resistance observed for the surviving cells has been found to increase over the duration of exposure to ciprofloxacin. Given these results, we were interested to know whether the resistance of the surviving cells to ciprofloxacin is a stable phenomenon. The initial susceptibility of P. aeruginosa strains has been recovered after 10 passages of the surviving cells, which were derived from the biofilms exposed to ciprofloxacin for 48 h. The findings of our study are not consistent with the observations by Lewis (22) that the progeny of persisters after treatment with bactericidal antibiotic, immediately after the cessation of antimicrobial treatment, would not be more resistant to the antibiotic than the original population. An increase in the expression of efflux pumps may account for this phenomenon, as has been recently shown by others (18, 25). In conclusion, the colony biofilms formed by strains of P. aeruginosa intermediate and highly resistant to ciprofloxacin secrete proteases. The secretion of these virulence factors by the colony biofilms is influenced by the bactericidal concentrations of the drug. However, the secretion of proteases is not attenuated even after exposure to ciprofloxacin for 4 days. REFERENCES 1. Bagge, N., M. Schuster, M. Hentzer, O. Ciofu, M. Givskov, E. P. Greenberg, and N. Høiby. 2004. Pseudomonas aeruginosa biofilms exposed to imipenem
3288
2.
3.
4. 5.
6.
7.
8. 9. 10. 11. 12. 13. 14.
15.
16.
17. 18.
OŁDAK AND TRAFNY
exhibit changes in global gene expression and -lactamase and alginate production. Antimicrob. Agents Chemother. 48:1175–1187. Bakker-Woudenberg, I. A., M. T. ten Kate, L. Guo, P. Working, and J. W. Mouton. 2002. Ciprofloxacin in polyethylene glycol-coated liposomes: efficacy models of acute or chronic Pseudomonas aeruginosa infection. Antimicrob. Agents Chemother. 46:2575–2581. Bisognano, C., P. Vaudaux, P. Rohner, D. P. Lew, and D. C. Hooper. 2000. Induction of fibronectin-binding proteins and increased adhesion of quinolone-resistant Staphylococcus aureus by subinhibitory levels of ciprofloxacin. Antimicrob. Agents Chemother. 44:1428–1437. Braga, P. C., M. T. Sala, and M. Dal Sasso. 1999. Pharmacodynamic effects of subinhibitory concentrations of rufloxacin on bacterial virulence factors. Antimicrob. Agents Chemother. 43:1013–1019. Budhani, R. K., and J. K. Struthers. 1998. Interaction of Streptococcus pneumoniae and Moraxella catarrhalis: investigation of the indirect pathogenic role of -lactamase-producing moraxellae by use of a continuousculture biofilm system. Antimicrob. Agents Chemother. 42:2521–2526. Caballero, A. R., J. M. Moreau, L. E. Engel, M. E. Marquart, J. M. Hill, and R. J. O’Callaghan. 2001. Pseudomonas aeruginosa protease IV enzyme assays and comparison to other Pseudomonas proteases. Anal. Biochem. 290:330– 337. Diggle, S. P., K. Winzer, A. Lazdunski, P. Williams, and M. Camara. 2002. Advancing the quorum in Pseudomonas aeruginosa: MvaT and the regulation of N-acylhomoserine production and virulence gene expression. J. Bacteriol. 184:2576–2586. Doring, G., H. J. Obernesser, K. Botzenhart, B. Flehmig, N. Høiby, and A. Hofmann. 1983. Proteases of Pseudomonas aeruginosa in patients with cystic fibrosis. J. Infect. Dis. 147:744–750. Drenkard, E. 2003. Antimicrobial resistance of Pseudomonas aeruginosa biofilms. Microbes Infect. 5:1213–1219. Garrelts, J. C., G. Jost, S. F. Kowalsky, G. J. Krol, and J. T. Letteri. 1996. Ciprofloxacin pharmacokinetics in burn patients. Antimicrob. Agents Chemother. 40:1153–1156. Gilbert, P., D. G. Allison, and A. J. McBain. 2002. Biofilms in vitro and in vivo: do singular mechanisms imply cross-resistance? J. Appl. Microbiol. Symp. Suppl. 92:98S–110S. Greenberg, E. P. 2003. Bacterial communication and group behavior. J. Clin. Investig. 112:1288–1290. Haskall, V. C., and J. H. Kimura. 1982. Proteoglycans: isolation and characterization. Methods Enzymol. 82:679–800. Hu, Y., J. A. Mangan, J. Dhillon, K. M. Sole, D. A. Mitchison, P. D. Butcher, and A. R. M. Coates. 2000. Detection of mRNA transcripts and active transcription in persistent Mycobacterium tuberculosis induced by exposure to rifampin or pyrazinamide. J. Bacteriol. 182:6358–6365. Huang, C. T., K. D. Xu, G. A. McFeters, and P. S. Steward. 1998. Spatial patterns of alkaline phosphatase expression within bacterial colonies and biofilms in response to phosphate starvation. Appl. Environ. Microbiol. 64:1526–1531. Jaffar-Bandjee, M. C., A. Lazdunski, M. Bally, J. Carrere, J. P. Chazalette, and C. Galabert. 1995. Production of elastase, exotoxin A, and alkaline protease in sputa during pulmonary exacerbation of cystic fibrosis in patients chronically infected by Pseudomonas aeruginosa. J. Clin. Microbiol. 33:924– 929. Jagger, K. S., D. R. Bahner, and R. L. Warren. 1983. Protease phenotype of Pseudomonas aeruginosa patients with cystic fibrosis. J. Clin. Microbiol. 17: 55–59. Jumbe, N., A. Louie, R. Leary, W. Liu, M. R. Deziel, V. H. Tam, R. Bachhwat, C. Freeman, J. B. Kahn, K. Bush, M. N. Dudley, M. M. Miller, and G. L. Drusano. 2003. Application of mathematical model to prevent in vitro amplification of antibiotic-resistant bacterial populations during therapy. J. Clin. Investig. 112:275–285.
ANTIMICROB. AGENTS CHEMOTHER. 19. Kobayashi, H. 2001. Airway biofilm disease. Int. J. Antimicrob. Agents 17:351–356. 20. Komori, Y., T. Nonogaki, and T. Nikai. 2001. Hemorrhagic activity and muscle damaging effect of Pseudomonas aeruginosa metalloproteinase (elastase). Toxicon 39:1327–1332. 21. Lantz, M. S., and P. Ciborowski. 1994. Zymographic techniques for detection and characterization of microbial proteases. Methods Enzymol. 235: 563–594. 22. Lewis, K. 2001. Riddle of biofilm resistance. Antimicrob. Agents Chemother. 45:999–1007. 23. Mah, T. F., B. Pitts, B. Pellock, G. C. Walker, P. S. Steward, and G. A. O’Toole. 2003. A genetic basis for Pseudomonas aeruginosa biofilm antibiotic resistance. Nature 426:306–310. 24. National Committee for Clinical Laboratory Standards. 1997. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, 4th ed. Approved standard M7-A4. National Committee for Clinical Laboratory Standards, Wayne, Pa. 25. Oliver, A., B. R. Levin, C. Juan, F. Baquero, and J. Blazquez. 2004. Hypermutation and the preexistence of antibiotic-resistant Pseudomonas aeruginosa mutants: implications for susceptibility testing and treatment of chronic infections. Antimicrob. Agents Chemother. 48:4226–4233. 26. Rust, L., C. R. Messing, and B. H. Iglewski. 1994. Elastase assays. Methods Enzymol. 235:554–562. 27. Sabra, W., E. J. Kim, and A. P. Zheng. 2002. Physiological responses of Pseudomonas aeruginosa PAO1 to oxidative stress in controlled microaerobic and aerobic cultures. Microbiology 148:3195–3202. 28. Schmidtchen, A., H. Wolff, and C. Hansson. 2001. Differential proteinase expression by Pseudomonas aeruginosa derived from chronic leg ulcers. Acta Derm. Venereol. 81:406–409. 29. Schmidtchen, A., E. Holst, H. Tapper, and L. Bjo ¨rck. 2003. Elastase-producing Pseudomonas aeruginosa degrade plasma proteins and extracellular products of human skin and fibroblasts, and inhibit fibroblast growth. Microb. Pathog. 34:47–55. 30. Sonstein, S. A., and J. C. Burnham. 1993. Effect of low concentrations of quinolone antibiotics on bacterial virulence mechanisms. Diagn. Microbiol. Infect. Dis. 16:277–289. 31. Sufya, N., D. G. Allison, and P. Gilbert. 2003. Clonal variation in maximum specific growth rate and susceptibility towards antimicrobials. J. Appl. Microbiol. 95:1261–1267. 32. Travis, J., and J. Potempa. 2000. Bacterial proteinases as targets for the development of second-generation antibiotics. Biochim. Biophys. Acta 1477: 35–50. 33. Twining, S. S., S. E. Kirschner, L. A. Mahnke, and D. F. Frank. 1993. Effect of Pseudomonas aeruginosa elastase, alkaline protease, and exotoxin A on corneal proteinases and proteins. Investig. Ophthalmol. Vis. Sci. 34:2699– 2712. 34. Walters, M. C., III, F. Roe, A. Bugnicourt, M. J. Franklin, and P. S. Stewart. 2003. Contributions of antibiotic penetration, oxygen limitation, and low metabolic activity to tolerance of Pseudomonas aeruginosa biofilms to ciprofloxacin and tobramycin. Antimicrob. Agents Chemother. 47:317–323. 35. Zahller, J., and P. S. Steward. 2002. Transmission electron microscopic study on antibiotic action on Klebsiella pneumoniae biofilm. Antimicrob. Agents Chemother. 46:2679–2683. 36. Zheng, Z., and P. S. Steward. 2002. Penetration of rifampin through Staphylococcus epidermidis biofilms. Antimicrob. Agents Chemother. 46:900–903. 37. Zhu, H., S. J. Thuruthyil, and M. D. P. Willcox. 2002. Determination of quorum-sensing signal molecules and virulence factors of Pseudomonas aeruginosa isolates from contact lens-induced microbial keratitis. J. Med. Microbiol. 51:1063–1070.
