Glycoside hydrolases catalyze the disruption of biofilms

We hypothesized that the exogenous application of the glycoside hydrolases PelA_h and PslG_h to Pel- and Psl-dependent biofilms, respectively, would result in hydrolysis of the exopolysaccharides, thereby disrupting the established biofilms. To assay for biofilm disruption, we produced biofilms using the following strains: PA14 (Pel-dependent matrix); PAO1 (Psl-dependent matrix); and l-arabinose–inducible P. aeruginosa PAO1 ΔwspF Δpsl P_BADpel and PAO1 ΔpelF P_BADpsl, which exclusively produce Pel and Psl, respectively. PelA_h and a putative catalytically inactive E218A variant (PelA_h E218A) were applied to Pel-dependent biofilms, whereas PslG_h and an inactive E165Q/E276Q variant (PslG_h E165Q/E276Q) were applied to Psl-dependent biofilms. Images captured with confocal microscopy coupled with fluorescence-labeled lectins from Hippeastrum hybrid Amaryllis (HHA; specific for Psl) and Wisteria floribunda (WFL; specific for Pel) demonstrated that catalytically active hydrolases, but not the inactive variants, were capable of degrading the Pel- and Psl-dependent biofilm biomass on the basis of the elimination of fluorescence signal following treatment (Fig. 1).

Open in a separate window

Fig. 1

The glycoside hydrolases PslG_h and PelA_h hydrolyze the exopolysaccharides Pel and Psl in a biofilm.

Representative confocal images of Psl biofilms grown statically for 24 hours (top) and Pel biofilms cultivated for 48 hours (bottom) under flow conditions and treated with wild-type hydrolases or hydrolases that have point mutations to catalytic residues. Biofilms were stained with the HHA Psl-specific lectin (green) and WFL Pel-specific lectin (red). Scale bars, 30 μm.

Crystal violet staining was subsequently utilized to quantify the effect of hydrolase treatment on the total biofilm biomass. A 2-hour treatment of a Pel-dependent biofilm with PelA_h resulted in disruption of 99% of the biomass, whereas both the PelA_h E218A variant and PslG_h, added in 100-fold excess relative to PelA_h, exhibited no significant difference compared to that of the untreated biofilm (Fig. 2A). Similar results were obtained for Psl-dependent biofilms wherein only the treatment with a catalytically active PslG_h resulted in a 98.5% reduction in biofilm biomass. The activity of both enzymes was observed to be dose-dependent. When incubated with established biofilms for 1 hour, PelA_h exhibited a half maximal effective concentration (EC₅₀) of 35.7 ± 1.1 nM, whereas PslG_h had an EC₅₀ of 12.9 ± 1.1 nM (Fig. 2B). Time course experiments using fixed concentrations of PelA_h and PslG_h revealed a continuous decrease in biofilm biomass over time (fig. S1). The activity of both enzymes was minimally affected by the presence of serum with observed EC₅₀ of 66.1 ± 1.2 nM for PelA_h and 6.7 ± 1.1 nM for PslG_h (fig. S2). Combined, these results indicate that biofilm disruption is catalytic, rapid, and exopolysaccharide-specific.

Open in a separate window

Fig. 2

The glycoside hydrolases PelA_h and PslG_h catalyze the inhibition and disruption of P. aeruginosa biofilms.

(A) Crystal violet staining of biofilms following the exogenous addition of glycoside hydrolases or catalytic variants. (B) Dose-response curves to examine the disruption of biofilm biomass by the exogenous treatment of each glycoside hydrolase and variant. (C) Dose-response curves to examine the prevention of biofilm biomass in the presence of various glycoside hydrolases. Each data point represents the mean from three independent experiments of n = 3 crystal violet microtiter plate wells. EC₅₀ values were calculated using nonlinear least-squares fitting to a dose-response model. Error bars indicate SEM. ***P ≤ 0.001. NS, no significant difference.

Glycoside hydrolases inhibit biofilm formation but not bacterial growth

Because the glycoside hydrolases were effective at disrupting established biofilms, we next sought to determine whether the application of enzyme to bacterial culture could be utilized as a prophylactic strategy to prevent biofilm formation. The addition of PelA_h, but not PslG_h, to Pel-producing P. aeruginosa abrogated biofilm formation. Dose titration indicated that Pel biofilms could be prevented over 24 hours by the addition of PelA_h with an EC₅₀ of 69.3 ± 1.2 nM (Fig. 2C). As visualized in borosilicate tubes, PelA_h prevented pellicle biofilm at the air-liquid interface, and bacterial cells grew exclusively in the planktonic state (fig. S3). Addition of PelA_h E218A, which cannot catalyze the disruption of biofilms, resulted in a statistically significant reduction in biofilm biomass at concentrations of ≥500 nM. Although an accurate EC₅₀ value could not be readily determined for this catalytic variant, 5 μM of the enzyme variant (>70 times greater than the EC₅₀ of the wild type) resulted in <50% reduction of the biomass relative to untreated cells. The presence of ≥10 μM PslG_h did not affect the ability of P. aeruginosa to form Pel-dependent biofilms. Consistent with the results for PelA_h on Pel-dependent biofilms, the addition of 1 μM PslG_h to Psl-producing cultures under biofilm-forming conditions resulted in a complete inhibition of biofilm formation, and dose titration indicated that PslG_h had an EC₅₀ of 4.1 ± 1.1 nM over 24 hours. To examine whether the effect was the direct result of PslG_h activity, we tested the catalytically inactive variant PslG_h E165Q/E276Q. This variant was >100-fold less effective in biofilm prevention (EC₅₀ of 466.5 ± 1.1 nM) relative to the catalytically active enzyme. Addition of ≥10 μM PelA_h had no effect on Psl-dependent biofilm production.

We next examined the length of time in which the glycoside hydrolases could prevent biofilm formation. A single dose of PelA_h or PslG_h prevented biofilm formation for 48 and 72 hours, respectively. The formation of a Pel-dependent biofilm at 72 hours was associated with proteolytic degradation of PelA_h, whereas no degradation of PslG_h was observed over the entirety of the experiment (fig. S4). The growth rate of P. aeruginosa PAO1 exposed to ≥20 μM of either glycoside hydrolase was unaffected over 6 hours of static growth when compared to a no-treatment control (fig. S5). The absence of enzyme cross-reactivity between exopolysaccharides indicates that biofilm inhibition is highly specific. This result, combined with bacterial growth curves, demonstrates that exogenous glycoside hydrolases do not impede biofilm formation by altering cell viability and growth.

Biofilm-disrupting enzymes are noncytotoxic

Because exogenous PelA_h and PslG_h did not affect P. aeruginosa cell viability and growth, we next sought to examine whether enzyme treatment affected mammalian cells. IMR-90 human lung fibroblast cells treated for 5 hours with concentrations of up to 1 mg/ml of either enzyme, which are ~100-fold above the concentration required for effective biofilm disruption, resulted in no significant difference in cell area or length-to-width ratio (Fig. 3A). Following a 48-hour incubation, no significant difference in cellular viability was observed in PelA_h- and PslG_h-treated cells regardless of the media used (Fig. 3B). The Clostridium difficile toxin TcdB (62) was used as a control for cell morphology because it results in cell rounding, whereas digitonin, which permeabilizes the cells, was used to monitor cellular viability (Fig. 3B). Western blot analysis of the media confirmed that PelA_h and PslG_h remained intact for the 48-hour duration of the experiment (fig. S6). Together, these results suggest that the enzymes do not interfere with mammalian cell morphology and viability.

Open in a separate window

Fig. 3

The glycoside hydrolases PelA_h and PslG_h are noncytotoxic.

(A) IMR-90 cellomics assay to measure the length-to-width ratio (LWR) of the cells using CellTracker Orange CMRA. (B) IMR-90 fibroblast cell viability assay using PrestoBlue reagent. All data were normalized to a no-treatment control (100%). The C. difficile toxin TcdB was used as a positive control in cell morphology assays, and the detergent digitonin was utilized as a negative control in cell viability assays. Each data point represents the mean from three independent experiments of n = 3 cellomic and PrestoBlue measurements in microtiter plate wells. Error bars indicate SEM. ***P ≤ 0.001. NS, no significant difference.

Enzymes potentiate antibiotics and ameliorate human neutrophil killing

Previous studies have demonstrated that both Pel and Psl enhance antibiotic resistance (7, 8). We therefore theorized that glycoside hydrolase degradation of these polymers could potentiate the activity of antimicrobial agents. Because the antibiotic colistin targets the cell membrane of P. aeruginosa, it is active against both metabolically active and dormant cells found within biofilms (63). The effect of combining glycoside hydrolase treatment with subinhibitory concentrations of colistin was therefore examined. As predicted, the treatment of P. aeruginosa Pel- and Psl-dependent 24-hour biofilm cultures with colistin (50 μg/ml) or those grown in the presence of PelA_h or PslG_h (2 μM) alone had no effect on the viability of the bacteria (Fig. 4A). However, prophylactic treatment with PelA_h or PslG_h before treatment with colistin resulted in an approximately 2.5-log reduction in bacterial colony-forming units (CFUs). A 5-hour cotreatment of 24-hour biofilms with either PelA_h or PslG_h and colistin resulted in a similar reduction in CFUs (Fig. 4B). These results indicate that these glycoside hydrolases are compatible with antibiotics and can potentiate the antimicrobial activity of colistin.

Open in a separate window

Fig. 4

Glycoside hydrolases potentiate antibiotics and increase human neutrophil killing.

(A) CFUs for PAO1 ΔwspF Δpsl P_BADpel (left) and PAO1 ΔpelF P_BADpsl (right) following growth in the presence or absence of glycoside hydrolases before treatment with colistin. The mean was calculated from LB agar plate counts from three independent experiments. (B) CFUs for biofilm cultures in (A) after treatment with glycoside hydrolases and colistin for 5 hours. The mean was calculated from LB agar plate counts from three independent experiments. (C) HL-60 neutrophil killing of strain PAO1 ΔwspF Δpsl P_BADpel and PAO1 ΔpelF P_BADpsl following biofilm formation and treatment with PelA_h and PslG_h and their catalytic variants, respectively. Percent killing was normalized to a no-treatment control. Error bars indicate SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. NS, no significant difference.

Because P. aeruginosa exopolysaccharides enhance resistance to human neutrophil killing, we next investigated whether glycoside hydrolase treatment could enhance susceptibility to immunomediated killing. To determine whether hydrolase treatment could affect neutrophil killing, we examined the ability of PelA_h and PslG_h to enhance the susceptibility of P. aeruginosa to the human HL-60–derived neutrophils. Treatment of Pel-dependent P. aeruginosa biofilms with PelA_h increased the degree of HL-60–mediated microbial killing from approximately 22 to 42% (Fig. 4C). This was not observed in a PelA_h E218A variant, indicating that the enhanced susceptibility to neutrophils is due to the catalytic activity of the enzyme that disrupts the biofilm. This effect was specific to PelA_h because neither PslG_h nor the E165Q/E276Q variant was significant to ameliorate neutrophil killing or affect neutrophil activity (Fig. 4C). Combined, these data provide further evidence that PelA_h and PslG_h do not affect mammalian cell function, and that PelA_h can function to enhance neutrophil killing of P. aeruginosa.

Cloning, expression, and purification of PelA and PslG constructs

PslG_h was purified as previously described (60). The DNA sequence of pelA from P. aeruginosa PAO1 was obtained from GenBank under accession no. AAG06452.1 (72). The PRED-TAT server (73) indicated that PelA has a TAT (twin-arginine translocation) signal sequence from residues 1 to 45. To obtain soluble protein constructs, we amplified pelA from genomic DNA by polymerase chain reaction using the primers CTGCATATGGGCGGGCCGTCCAGCGTGGCG and TTTCTCGAGTCACGGTTGCAC CTCGACGTC, respectively. Introduced NdeI and XhoI restriction sites were underlined, and each gene was ligated into the pET28a (Novagen) expression vector encoding an N-terminal polyhistidine tag. This generated PelA_47–303. Site-directed mutagenesis was performed to generate protein variants, using the QuikChange Lightning Kit according to the prescribed protocol (Agilent Technologies). Generated constructs were verified by sequencing performed by ACGT DNA Technologies Corporation.

E. coli BL21 (DE3) CodonPlus cells (Stratagene) were transformed with the expression plasmid and grown in 2 liters of Lauria-Bertani (LB) broth containing kanamycin (50 μg/ml) at 37°C. When the optical density at 600 nm (OD₆₀₀) of the cell culture reached 0.5 to 0.6, protein expression was induced by the addition of isopropyl β-d-1-thiogalactopyranoside to a final concentration of 0.5 mM. After induction, the cells were incubated overnight at 18°C with shaking at 200 rpm, before being harvested by centrifugation at 5000g for 30 min at 4°C. Cell pellets containing PelA_47–303 were resuspended in 40 ml of buffer A [20 mM imidazole, 50 mM tris-HCl (pH 7.5), 300 mM NaCl, and 2% (v/v) glycerol] with one SIGMAFAST Protease Inhibitor Tablet. The cells were lysed by at least three passes through an EmulsiFlex C3 homogenizer at 100 MPa (Avestin Inc.), and the resulting cell debris was separated from soluble protein by centrifugation at 35,000g for 30 min. The supernatant was applied to 5 ml of Ni-NTA Superflow resin packed into a gravity column (Qiagen) preequilibrated with buffer A. The column was washed with 3 CV (column volume) of buffer A, and the expressed protein was eluted with buffer A supplemented with 250 mM imidazole. The eluted fractions were concentrated to a 1- to 2-ml volume using an Amicon Ultra centrifugation filter device (Millipore) with a 10-kDa cutoff, and the protein was further purified by size exclusion chromatography using a HiLoad 16/60 Superdex 200 gel filtration column (GE Healthcare). The protein was judged to be >95% pure by SDS–polyacrylamide gel electrophoresis, and the protein could be concentrated to 8 to 10 mg/ml and stored at 4°C for at least 1 month without precipitation or degradation.

Confocal microscopy

Psl biofilms were grown overnight at room temperature in uncoated 15 μ-Slide VI^0.4 flow cell chambers (ibidi GmbH). The channels were inoculated with 200 μl of a culture with an OD₆₀₀ of 0.5 grown in LB no salt (LBNS) supplemented with 0.5% arabinose. Biofilms were washed three times with sterile phosphate-buffered saline and then treated with PslG_h, PslG_h E165Q/E276Q, and buffer-only control [50 mM tris (pH 7.5), 150 mM NaCl, and 10% (v/v) glycerol] statically for 1 hour at room temperature. The final enzyme concentration was 86 nM. After digestion, the biofilms were stained with fluorescein isothiocyanate–conjugated HHA lectin (100 μg/ml; EY Laboratories) for 2 hours at 4°C, as previously described by Ma et al. (74). The biofilms were then washed and fixed with 4% paraformaldehyde. Fluorescent images were acquired with an Olympus FV1000 filter confocal system using a 20× LUCPLFLN objective lens with a numerical aperture of 4.5 (Olympus America Inc.). Images were analyzed and constructed using the Olympus FluoView version 03.01 software.

Pel-dependent biofilms were cultivated as previously described (46), with minor modifications. Flow cell chambers were inoculated with a mid-log LB culture of P. aeruginosa PA14 that was diluted with glucose minimal medium (final glucose concentration of 0.3 mM) to an OD₆₀₀ of 0.01. Cells were allowed to attach for 1 hour before induction of flow. Biofilms were grown on glucose minimal medium for 2 days at room temperature at a constant flow rate (10 ml/hour) before treatment with PelA_h, PelA_h E218A, and buffer-only control [20 mM Tris (pH 8.0), 150 mM NaCl, and 10% (v/v) glycerol] statically for 1 hour at room temperature. The final enzyme concentration was 85 nM. After digestion, biofilms were washed, and Pel was then stained with fluorescein-labeled WFL lectin (100 μg/ml; Vector Laboratories) for 15 min. Stained biofilms were washed before visualization on a Zeiss LSM 510 confocal laser scanning microscope. Image analysis was conducted using the Velocity software (Improvision). Experiments were performed in biological duplicate.

Microtiter dish biofilm assay

For examination of biofilm prevention, Psl and Pel arabinose–inducible P. aeruginosa PAO1 (PAO1 ΔpelF P_BADpsl and PAO1 ΔwspF Δpsl P_BADpel) and clinical isolates were grown at 37°C overnight with shaking at 200 rpm. The cultures were normalized to an OD₆₀₀ of 0.5 and then diluted 1:100 in LBNS. l-Arabinose was added to a final concentration of 0.5% (w/v) to induce exopolysaccharide biosynthesis and biofilm formation. Diluted culture (95 μl) was added to sterile 96-well polystyrene microtiter plates (Thermo Scientific cat. no. 243656), and varying concentrations of PelA_h or PslG_h (0.1 to 5 μM) were added in 5-μl aliquots to give a final volume of 100 μl. The cultures were incubated statically for 24 hours at 25°C to allow for biofilm formation. For elimination of edge effects, ~200 μl of sterile water was placed in all outside wells, and the plate was sealed with parafilm. After incubation, nonadherent cells and media were removed by thoroughly washing the plate with deionized water. The wells were stained with 150 μl of 0.1% (w/v) crystal violet for 10 min and then rinsed with water. The remaining dye was solubilized by addition of 150 μl of 95% (v/v) ethanol and left for 10 min, after which the absorbance was measured at 595 nm using a SpectraMax M2 spectrophotometer (Molecular Devices). The amount of biofilm was proportional to the absorbance from staining with crystal violet (75).

For biofilm disruption assays, biofilm cultures were grown statically for 24 hours. Following incubation, nonadherent cells and media were removed by washing the plate with distilled water. The wells were filled with 95 μl of 100 mM HEPES sodium buffer (pH 7.0) followed by 5 μl of varying concentrations of each hydrolytic enzyme (2 to 5 μM). Reactions were allowed to proceed for up to 60 min at 25°C on a rotating nutator, at which time the reaction was quenched by washing the plates with distilled water. The wells were stained with 150 μl of 0.1% (w/v) crystal violet for 10 min and then washed and solubilized with ethanol before quantification. All reactions were completed in triplicate. The addition of kanamycin (2.5 mg/ml) to the culture before biofilm formation was used as a positive control.

P. aeruginosa growth assay

To assay for glycoside hydrolase cytotoxicity to P. aeruginosa PAO1, we set up a bacterial growth assay as described for the biofilm inhibition assay, with 25 μM PelA_h or PslG_h added at the time of inoculation. The bacteria were grown statically at 37°C in a thermocontrolled SpectraMax M2 spectrophotometer. At 30-min intervals, the OD₆₀₀ of each culture was measured for a duration of 6 hours. An untreated culture was used as a control.

Antibiotic susceptibility assay

Overnight cultures of P. aeruginosa PAO1 ΔwspF Δpsl PBADpel (Pel-dependent) and P. aeruginosa PAO1 ΔpelF P_BADpsl (Psl-dependent) were diluted to an OD₆₀₀ of 0.05 in LBNS with 0.5% (w/v) l-arabinose and were grown statically in polystyrene tubes at a final volume of 1 ml per tube. For prophylactic treatment, 2 μM PelA_h or PslG_h was added at the time of inoculation and incubated with the bacteria for 24 hours at 25°C. Following incubation, colistin was added at a final concentration of 50 μg/ml and was incubated for 24 hours. For treatment of 24-hour biofilms, 2 μM PelA_h or PslG_h was added concurrently with a final colistin concentration of 50 μg/ml and was incubated with the biofilm at 25°C for 5 hours. Cultures that were not subjected to treatment or were only treated with enzymes served as controls. Adherent biofilm and embedded cells were resuspended by scraping the tubes and vigorous pipetting to ensure removal of all cellular materials and to prevent experimental bias. Viability was quantified by serial dilutions and CFU counts on LB agar plates of the surviving population. Experiments were performed three times to obtain both mean and SE.

Human neutrophil killing assay

Overnight cultures of P. aeruginosa PAO1 ΔwspF Δpsl P_BADpel and P. aeruginosa PAO1 ΔpelF P_BADpsl were diluted to an OD₆₀₀ of 0.05 in LBNS with 0.5% l-arabinose and inoculated in a 96-well tissue culture–treated plate at a final volume of 100 μl per well. The plate was incubated statically at 28°C for 20 hours. Supernatants were aspirated, and 100 μl of phenol red–free RPMI medium + 10% fetal bovine serum (FBS) containing 0.5 μM PelA_h or PslG_h was added. The plate was incubated at room temperature on the nutator for 1 hour. Following pretreatment with hydrolase, 100 μl of RPMI + 10% FBS containing 6 × 10⁶ differentiated HL-60 cells was added to the wells, and the plate was incubated for 90 min at 37°C with 5% CO₂. Wells were aspirated, and the supernatant was diluted between 1:200,000 and 1:400,000 and plated (50 μl) onto LB agar. Two hundred microliters of 2 μM PelA_h and 2 μM PslG_h was added to aspirated wells, and the plate was incubated at room temperature on the nutator for 1 to 1.5 hours. The wells were aspirated, diluted, and plated onto LB agar as described above.

Cell morphology and viability assays

IMR-90 human lung fibroblast cells were seeded in 96-well CellBIND plates (Corning) at a density of 8000 to 10,000 cells per well. The next day, the medium was exchanged with a medium containing 1 μM CellTracker Orange CMRA (Molecular Probes) in serum-free Eagle’s minimum essential medium (Wisent) or supplemented with 10% FBS. After 60 min, excess dye was removed by media exchange. PslG_h and PelA_h were added to a final concentration of 1 mg/ml. The detergent digitonin and the C. difficile toxin TcdB were utilized as controls and added to a final concentration of 0.03 mg/ml and 0.5 pM, respectively (62). The cell plates were returned to the incubator for 5 hours before imaging. CellTracker-labeled cells were evaluated on a Cellomics ArrayScan VTI HCS reader (Thermo Scientific) using the target acquisition mode, a 10× objective, and a sample rate of 100 objects per well. Following a 24-hour incubation, cells in serum-free medium were supplemented with 10% FBS. Forty-eight hours after incubation, PrestoBlue reagent was added to the cells at a 1:10 reagent/media ratio and allowed to incubate for 5 hours. The microtiter plates were read in a SpectroMax M2 plate reader with λ_ex (excitation wavelength) of 555 nm and λ_em (emission wavelength) of 585 nm. To probe the presence of PslGh and PelAh, we conducted Western blot analysis using polyclonal antibodies that were previously generated against both enzymes (47, 60).

Funding: This work was supported by operating grants from the Canadian Institutes of Health Research (CIHR) (grant 43998 to P.L.H., grant 123306 to D.C.S., grant 286650 to R.A.M., and grant 81361 to P.L.H. and D.C.S.), Cystic Fibrosis Canada (CFC) (to D.C.S. and P.L.H.), the NIH (grant R01AI097511 to D.J.W. and grant 2R01AI077628 to M.R.P.), and the Natural Sciences and Engineering Research Council of Canada (grant RGPIN 418405 to R.A.M.). P.B. has been supported in part by a CFC postdoctoral fellowship and a Banting Fellowship from CIHR. B.D.S. has been supported by graduate scholarships from CFC and CIHR. L.K.J. is the recipient of an American Heart Association Postdoctoral Fellowship (14POST20130017). P.L.H. is the recipient of a Canada Research Chair. Author contributions: P.B., M.R.P., D.C.S., D.J.W., and P.L.H. designed the study and wrote the paper. P.B. completed the biofilm inhibition, biofilm disruption, antibiotic, and cytotoxicity assays. P.J.H. and M.J.P. completed the biofilm disruption assays and confocal microscopy. B.D.S. and M.J.L. completed the neutrophil assays. N.A. designed the vectors for the expression and purification of the PelA hydrolase and performed the initial Pel biofilm inhibition studies. L.K.J. completed the confocal microscopy experiments on the Pel-dependent biofilms. P.B., J.T., and R.A.M. developed and completed the cytotoxicity experiments. All authors approved the final version of the manuscript. Competing interests: The authors declare that they have no competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Additional data related to this paper may be requested from the authors.