The purpose of this study was to evaluate a commercial antimicrobial formulation, Byotrol™ G32, as a potential coating for impeding biofilm formation on medical devices such as urinary catheters.
The purpose of this study was to evaluate a commercial antimicrobial formulation, Byotrol™ G32, as a potential coating for impeding biofilm formation on medical devices such as urinary catheters.
The antimicrobial activity of Byotrol™ G32 and its individual constituents has been tested on planktonic and biofilm cultures of uropathogenic bacteria. The Byotrol™ G32 formulation was superior with MICs ranging from 3 μg ml−1 to 15 μg ml−1 for planktonic cultures and 3–20 μg ml−1 for biofilms. Furthermore, Byotrol™ G32 was able to remove established biofilms and act as an antibiofilm surface coating.
Byotrol™ G32 displays impressive antimicrobial activity both in suspension and as a coating. Pretreating medical devices with Byotrol™ G32 may significantly impede biofilm formation and prolong the lifetime of the device.
Medical devices are indispensable in health care. They are, however, a predisposing factor in infection. This research has demonstrated that Byotrol™ G32 reduces bacterial growth and subsequent biofilm formation. Application of Byotrol™ G32 as a medical device coating could have a significant impact on the costs associated with device replacement and patient morbidity and mortality.
It is estimated that over half of all nosocomial infections are associated with medical devices, such as artificial heart valves, prosthetic devices, surgically implanted devices, contact lenses, wound drainage tubes, dressings, intrauterine contraception devices, sutures, intravenous catheters and urinary catheters (Smith et al. 1991; Costerton et al. 1999; Richards et al. 1999). Systemic and chronic infections are a result of biofilm formation on the surface of medical devices (Costerton et al. 1999). Biofilms are organized multicellular communities of bacteria attached to a surface and embedded in a protective polymer matrix. The biofilm phenotype is a ubiquitous characteristic of bacteria that constitutes a protected growth mode to facilitate survival in hostile environments (Costerton et al. 1987, 1995, 1999). It offers increased resistance to host defences and antimicrobials, and consequently, biofilms are notoriously difficult to treat and commonly manifest as chronic or recurrent infections (Patel 2005; Vuong et al. 2005; Anderson and O'Toole 2008). The most effective method to impede biofilm development is to avoid or reduce the initial adhesion of bacteria to the surface. The nonspecific attachment of bacteria to any surface is a key determinant in subsequent biofilm formation; therefore, many approaches have been adopted to prevent bacterial attachment to surfaces of medical devices (Banerjee et al. 2011; Knetsch and Koole 2011). Currently, the most widely used antimicrobial biomaterials are those that have silver-modified coatings (Boswald et al. 1999; Davenport and Keeley 2005). These are now routinely used in wound management (Davenport and Keeley 2005; Silver et al. 2006; Atiyeh et al. 2007), and while they can reduce the risk of infection, they introduce a host of problems. For example, the uptake of silver ions by bacterial cells has resulted in the emergence of silver-resistant strains (Silver 2003; Percival et al. 2005; Silver et al. 2006), indeed silver-resistant Pseudomonas aeruginosa has been isolated from burn patients who have been treated with silver-coated wound dressings (Modak and Fox 1981). Cationic compounds represent a suitable alternative as they define a structurally diverse class of antimicrobials (Banerjee et al. 2011). Such chemical diversity leads to a broad spectrum of activity and different modes of action. Furthermore, despite over a century of use, only trace levels of cross-resistance have been observed in the clinical environment (Gilbert and Moore 2005; Jaglic and Cervinkova 2012). Antibiofilm coatings containing novel cationic compounds could, therefore, have a significant impact on the clinical setting.
Byotrol™ G32 is a successful commercial antimicrobial hygiene product that is based on a novel proprietary mixture of poly(hexamethylene biguanide) chloride (PHMB); didecyldimethylammonium chloride (DDQ) and alkyl (C12, 70%; C14, 30%) dimethyl benzyl ammonium chloride (BAC). In this study, we evaluated the in vitro efficacy of Byotrol™ G32 for impeding biofilm formation.
The well-characterized laboratory strain Escherichia coli K12 (XL1 blue phage) and clinical isolates of E. coli, Klebsiella pneumoniae and Ps. aeruginosa obtained from urinary tract infections and characterized by Vitek® 2 (BioMérieux, Inc., Basingstoke, UK) were used in this study. The staff at the Central Manchester Foundation Trust, Clinical Sciences Building 2, Manchester, UK, kindly provided the clinical isolates used in this study. Stocks were stored at −80°C in 80% glycerol. Stock bacteria were cultured for 18 h on Luria–Bertani (LB) agar plates every two weeks. Plates were stored at 4°C. Overnight cultures were prepared by inoculating LB broth with several colonies from the working culture plates and incubating for 18 h with shaking at 200 rev min−1. The inoculum was standardized to 1 × 108 CFU ml−1 using the Miles and Misra method for the determination of viable cell counts. All cultures were incubated in an aerobic atmosphere at 37°C.
The composition of Byotrol™ G32 (Byotrol™ Technology Ltd., Daresbury, UK) is PHMB (poly(hexamethylene biguanide) chloride); DDQ (didecyldimethylammonium chloride) and BAC (alkyl (C12, 70%; C14, 30%) dimethyl benzyl ammonium chloride).
Poly(hexamethylene biguanide) was obtained from Arch Chemicals (Tradename-Vantocil TG) as a 20 wt% solution in water. DDQ was obtained from Lonza (Tradename - Lonza Bardac 2240) and received as a 40 wt% solution in water. BAC was obtained from Thor (Tradename – Acticide BAC 50 mol l−1) and received as a 50 wt% solution in water.
The minimum inhibitory concentration (MIC) of Byotrol™ G32 and its constituents was determined by incubating increasing concentrations of each with 100 μl of a 1 : 50 dilution of overnight cultures prepared in LB in 96-well flat-bottomed nontissue culture–treated polystyrene microtitre plates (Greiner Bio-one Ltd., Gloucestershire, UK). Eight technical replicates were prepared on each plate for each concentration of antimicrobial tested. The microtitre plates were incubated under aerobic, static conditions for 18 h at 37°C.
Positive growth controls were prepared by inoculating eight wells with 100 μl of 1 : 50 dilutions of bacteria and 100 μl sterile distilled water in which the antimicrobial was prepared. Negative controls were prepared by dispensing 100 μl LB broth and 100 μl sterile distilled water into a further 8 wells. The microtitre plates were incubated under aerobic, static conditions for 18 h at 37°C.
After the incubation period, the optical density at 595 nm (OD595) of planktonic growth was measured to quantify the MIC using a spectrophotometer (BMG Labtech FLUOstar OPTIMA). The average optical density (OD) from the eight negative control wells was subtracted from the average OD from the eight technical replicates of each concentration in the test wells. The MIC is determined as the lowest concentration of antimicrobial that completely inhibits visual bacterial growth, or an OD595 < 0·05.
After determining the MIC, the microtitre plate biofilm formation assay was used according to the method by Christensen et al. (1985). Briefly, excess media and any planktonic cells were decanted from the microtitre plate, and each well was washed with 200 μl sterile phosphate-buffered saline (PBS) (Sigma-Aldrich Company Ltd.). The plate was left in an inverted position to air dry overnight at room temperature. Each well was stained with 150 μl 0·4% (w/v) crystal violet (Sigma-Aldrich Company Ltd.) at room temperature for 10 min and washed with running tap water until the excess stain was removed and the running water appeared colourless. The plate was inverted and left to dry overnight at room temperature. The biofilm density was quantified by solubilizing the ammonium crystal violet stain with 200 μl 99·5% ethanol (Fisher Scientific UK Ltd., Loughborough, UK) and measuring the OD595 of solubilized crystal violet in each well using a spectrophotometer plate reader (BMG Labtech FLUOstar OPTIMA). The average OD from the eight negative control wells was subtracted from the average OD from the eight technical replicates of each concentration in the test wells.
Eight-well glass chamber slides (Lab-Tek™ Chamber slide™; Nunc, Fisher Scientific, Loughborough, UK) were used to analyse biofilms by bright field microscopy. Each slide had six increasing concentrations of antimicrobial agent, one in each well. The concentration of antimicrobial agent in each well was prepared by dispensing 150 μl of the antimicrobial agent and 150 μl of a 1 : 50 dilution of inoculum prepared from an overnight culture as described previously. The remaining two wells contained a positive growth control prepared with 150 μl inoculum and 150 μl water, and a negative control with 150 μl media and 150 μl water. Slides were prepared in triplicate. After incubation, the slides were washed with 1 ml PBS and immediately stained with 0·4% crystal violet for 10 min. Excess stain was washed with running water and left to dry in an inverted position overnight. The chambers were removed and microscopic images were collected on an Olympus BX51 upright microscope using a 100×/1·30 UPlanFLN objective. Images were captured using a CoolSnap HQ camera (Photometrics) through MetaVue Software (Molecular Devices). Images were then processed and analysed using ImageJ (http://rsb.info.nih.gov/ij).
This assay determines the effect of Byotrol™ G32 on biofilm cells. 200 μl of a 1 : 100 dilution of an overnight culture was dispensed into wells of a microtitre plate. The Minimum Biofilm Eradication Concentration (MBEC) assay is a modification of the method described by Ceri et al. (1999). A lid with protruding pegs (transferable solid-phase screening system, Nunc) was placed into the inoculated wells and incubated under static conditions for 18 h. The lid carrying pegs were transferred to a microtitre plate containing 200 μl sterile PBS, shaken to remove any nonadhered bacterial cells and then placed in a microtitre plate containing a concentration range of each antimicrobial, which was prepared in the same manner as described for the MIC assay and microtitre plate biofilm production assay. The plates were incubated under static conditions for 18 h. The pegs were then placed in 200 μl PBS, shaken briefly and immediately placed in 200 μl fresh sterile media and incubated for a further 18 h to allow for regrowth of viable bacterial cells on the pegs. After incubation, the OD595 of the plates containing any planktonic growth was read. The pegs were washed in 200 μl PBS and allowed to dry overnight at room temperature before placing the pegs in 200 μl 0·4% crystal violet for 15 min. The pegs were washed in running water and left to air dry overnight at room temperature. The biofilm density was quantified by solubilizing the ammonium crystal violet stain with 200 μl 99·5% ethanol and measuring the OD595 of solubilized crystal violet using a spectrophotometer plate reader. The average OD from the eight negative control wells was subtracted from the average OD from the eight technical replicates of each concentration in the test wells.
The wells of a glass chamber slide were pretreated by incubating overnight at 37°C with a 5 μg ml−1 and 1 mg ml−1 solution of Byotrol™ G32. After incubation, excess Byotrol™ G32 was decanted and slides were re-incubated for a further 18 h at 37°C to allow for evaporation of water. A 1 : 100 dilution of overnight cultures of bacteria was prepared in LB broth, 100 μl of which was added to pretreated slides and incubated for 8 h at 37°C. A minimum of four biological replicates were prepared for each concentration. After the incubation period, the slides were washed with 1 ml PBS, stained with 250 μl crystal violet and solubilized with 300 μl ethanol. The OD595 measurements were determined as for the microtitre plate biofilm formation assay.
Wells of a glass chamber slide were precoated with 300 μl of a 1 mg ml−1 solution of Byotrol™ G32 overnight at 37°C. After incubation, excess Byotrol™ was decanted and the slides were re-incubated for a further 18 h at 37°C to allow for evaporation of water.
Topographies were recorded using atomic force microscopy (AFM) (PSIA Inc, XE100, Surrey, UK) in noncontact mode. A commercial silicon cantilever (PSIA Inc, 910M-NSC15) with a nominal spring constant of about 40 N/m was used.
The glass slides with precoated Byotrol™ G32 were secured to a metal disc using double sided tape and installed on the AFM scanner. An area of 20 μm2 was scanned. A scratch was made across the bottom of the glass chamber using a 0·8 × 40 mm needle. The resulting topography image of this scratch gives an indication of the film thickness on the precoated glass slides.
The antimicrobial activity of the Byotrol™ G32 formulation and its individual constituents was tested on a planktonic laboratory E. coli K12 strain and clinical strains of E. coli, Kl. pneumoniae and Ps. aeruginosa, all isolated from patients with urinary tract infections. For the Byotrol™ G32 formulation, antimicrobial activity was observed against all isolates (Fig. 1) with MICs ranging from 3 μg ml−1 for the two E. coli isolates to 15 μg ml−1 for Kl. pneumoniae and Ps. aeruginosa. A comparison of the MICs for the individual constituents of Byotrol™ G32, displayed in Table 1, reveals that the G32 formulation outperforms the individual constituents.
|Minimum inhibitory concentration (μg ml−1)|
|E. coli K12||3||15||20||60|
|E. coli clinical isolate||3||60||30||100|
|Kl. pneumoniae clinical isolate||15||100||100||200|
|Ps. aeruginosa clinical isolate||15||200||100||400|
Next, the amount of biofilm produced by the remaining viable planktonic cells after 18 h was determined by crystal violet staining of the adhered bacterial cells. Figure 2 illustrates that low concentrations of Byotrol™ G32 (3–20 μg ml−1) reduce the biofilm development of all isolates. To confirm these results, biofilms stained with crystal violet were visualized by bright field microscopy (Fig. 3). In this instance, bright field microscopy of bacterial cells stained with crystal violet was preferable to fluorescence microscopy, as it avoided nonspecific binding of dyes to Byotrol™ G32, which we observed for the standard fluorescent biofilm labelling dye FilmTracer™ SYPRO® Ruby biofilm matrix stain (Invitrogen™, Life Technologies Ltd., Paisley, UK), and for the Live/Dead® stain (Invitrogen™). Figure 3 demonstrates that the number of adherent bacteria is substantially reduced in relation to the growth control, at concentrations of Byotrol™ G32 as low as 3 μg ml−1, confirming the ability of Byotrol™ G32 to impede biofilm development. Consistent with previous results, reducing Ps. aeruginosa biofilm development required higher concentrations of Byotrol™ G32. Furthermore, comparison to the minimum biofilm inhibitory concentrations for the individual constituents displayed in Table 2 reveals that again the G32 formulation outperforms any of the individual constituents in its ability to reduce the degree of bacterial growth.
|Minimum biofilm inhibitory concentration (μg ml−1)|
|E. coli K12||3||20||20||60|
|E. coli clinical isolate||5||60||30||100|
|Kl. pneumoniae clinical isolate||15||100||30||60|
|Ps. aeruginosa clinical isolate||20||400||100||400|
The impact that Byotrol™ G32 had on established biofilms was also determined by performing the MBEC assay. The results, which are presented in Fig. 4, show that Byotrol™ G32 was effective against biofilms of E. coli and Kl. pneumoniae; >50% of the biofilm was eradicated at concentrations of 60 μg ml−1. Ps. aeruginosa was more difficult to eradicate requiring 400 μg ml−1 to reduce the biofilm mass by 50%.
Byotrol™ G32 was used to coat glass surfaces by incubating a solution at 37°C and then drying. Atomic force microscopy was used to verify the extent of coating and access surface morphology. Figure 5(a) shows the atomic force micrographs of a glass surface coated with 5 μg ml−1 and 1 mg ml−1 Byotrol™ G32. A scratch drawn along the surface allowed the depth of the coatings to be measured (Fig. 5b). At a concentration of 5 μg ml−1, the coating depth was measured as 20 nm. At a concentration of 1 mg ml−1, the surface is evenly coated and the depth of the coating was measured as 60 nm (Fig. 5b). Biofilm development on the coated surfaces was determined by crystal violet staining of a biofilm grown for 8 h. On the surface coated with 1 mg ml−1 Byotrol™ G32, no biofilm development was observed for the two E. coli strains and Kl. pneumoniae, and Ps. aeruginosa biofilm development was impeded by ~60%.
Healthcare professionals, to support the care and treatment of patients, increasingly use medical devices such as catheters, shunts, orthopaedic implants and wound dressings. While medical devices offer regained structure and function to the body, they are a persistent source of infection (Darouiche 2001; Van and Michiels 2005). Such infections not only present profound economic burdens for society but infections associated with the insertion of a medical device lead to significant levels of morbidity and are sometimes fatal (Polonio et al. 2001). Bacterial communities known as biofilms play a central role in device-associated infections (Costerton 2007; Hatt and Rather 2008). Preventing biofilm formation is key to prolonging the lifetime of any medical device and reducing infection-related complications.
The ability of many conventional antimicrobials to inhibit biofilm formation has been assessed and some success has been found with the use of heavy metals, particularly silver. Hydrogel-coated latex catheters impregnated with silver on both the outer surface and lumen of the catheter have shown some effectiveness, however, only in the short-term (Bologna et al. 1999; Verleyen et al. 1999), possibly because the main challenge of the urinary catheter is the hugely mixed population of resistant micro-organisms that form a biofilm and cause infection, including bacteria, which display heavy metal resistance (Woods et al. 2009). Therefore, silver may not be an ideal inhibitor in the long term, as it may select for organisms with silver resistance. Furthermore, there is conflicting evidence in the literature as to the efficacy of silver in the clinical environment (Johnson et al. 2006).
Here, we have studied the antimicrobial and biofilm inhibitory properties of a proprietary hygiene product, Byotrol™ G32, which is a commercial formulation based upon a mixture of poly(hexylmethylbiguanide) chloride (PHMB), didecyldimethylammonium chloride (DDQ) and dimethyl benzyl ammonium chloride (BAC). PHMB is a polymeric cationic antimicrobial agent, which for many years has been used in the domestic, food and medical industries (Gilbert and Moore 2005; Kim et al. 2011). BAC and DDQ are quaternary ammonium compounds that have also been well studied as antimicrobial agents and used in a wide variety of settings (Ioannou et al. 2007). Although PHMB, DDQ and BAC all display impressive antimicrobial activity, we demonstrate that these constituents are synergistic in the Byotrol™ G32 formulation and present a much greater degree of antimicrobial activity.
The microtitre plate biofilm formation assay demonstrates a correlation between the MIC results for the inhibition of planktonic cells and the inhibition of biofilm forming cells, that is, the fewer the planktonic cells, the fewer the number of viable cells that are able to form a biofilm. However, the results from the microtitre plate biofilm formation assay validates the activity of Byotrol™ G32 against planktonic cells and therefore as an inhibitor of early biofilm formation. Ps. aeruginosa is inherently resistant to many antimicrobial agents due to surface factors such as its outer membrane impermeability and active drug efflux mechanisms (Drenkard 2003; Trott et al. 2007). The MIC for the planktonic cells of Ps. aeruginosa is lower than the biofilm inhibition concentration, which may be a result of fewer planktonic cells being killed; therefore, perhaps a greater number of cells were able to proliferate as a biofilm.
By comparison, Byotrol™ is least effective against Ps. aeruginosa. This may be explained by the suggestion that biocides, which have a broad spectrum of activity and nonspecific targets, may cause nonspecific resistance mechanisms, for example in Ps. aeruginosa, the hyperexpression of multidrug efflux pumps (Gilbert and McBain 2003). There may also be changes to the outer membrane, which reduces the permeability of the membrane to the biocide (McDonnell and Russell 1999; Gilbert 2005). However, the higher MIC of Byotrol™ for Ps. aeruginosa compared to that of the other organisms tested is likely to be due to increased tolerance rather than resistance (Gilbert and McBain 2003).
Intervention at the early stages of biofilm development and inhibiting planktonic cells before they are able to attach to a surface is most desirable. For this reason, the inhibition of viable planktonic bacteria to attach to the surface was evaluated at 18 h with the microtitre plate biofilm formation assay and 8 h for the pretreatment of a glass surface, as these time points reflect the early stages of biofilm formation.
Pretreatment of a surface with Byotrol™ G32 in this study provided the preliminary data as to the efficacy of Byotrol™ as an antimicrobial coating. AFM clearly shows the deposition of a nonstructured thin film over the concentration range for Byotrol™ G32. Glass has a negative zeta potential at pH 7 (Gu and Li 2000); therefore, we propose that as PHMB is a positively charged polyelectrolyte, it strongly adheres to the surface through opposite charge interaction (Borkovec and Papastavrou 2008) and serves to enhance the weaker adhesion of both BAQ and DDQ through the formation of a composite film.
To place these findings in a clinical context, the stability and lifetime of this coating under the flow of urine would also be an important factor in the success of Byotrol™ as a catheter coating. The more stable the coating, the longer the lifetime of the catheter. That being the case, any long-term toxicity associated with Byotrol™ against uroepithelial cells would also need to be assessed. It should be noted that PHMB and BAC, the compounds demonstrating the highest antimicrobial activity in the Byotrol™ formulation, are already widely used in the environmental and clinical setting, including wound care dressings and ophthalmic solutions. In a clinical review, PHMB was stated to have good clinical safety with no known toxic risks (Gray et al. 2010), and even with long-term use, BAC was stated to pose no clinical risk (Marple et al. 2004). Therefore, although cytotoxicity testing would need to be performed prior to a Byotrol™-coated catheter entering the clinical setting, it is probable that Byotrol™ should be safe to use.
In conclusion, Byotrol™ G32 has the ability to reduce bacterial growth as an antimicrobial in suspension, as a coating, and is also able to disrupt an existing biofilm. It may therefore hold great promise not only as an antimicrobial coating for medical devices but also as a sanitizing agent for the removal of established biofilms.
We would like to thank Age UK for funding this research and Byotrol Technology Ltd. for sample provision. The microscopes used in this study were part of The Bioimaging Facility at the University of Manchester, purchased with grants from BBSRC, the Wellcome Trust and the University of Manchester Strategic Fund.