Furanones as potential anti-bacterial coatings on biomaterials

ARTICLE IN PRESS

Biomaterials 25 (2004) 5003–5012

Furanones as potential anti-bacterial coatings on biomaterials J.K. Bavejaa,b, M.D.P. Willcoxa, E.B.H. Humea, N. Kumarc, R. Odellb, L.A. Poole-Warrenb,* a

Cooperative Research Centre for Eye Research and Technology, University of New South Wales, Sydney, NSW 2052, Australia b Graduate School of Biomedical Engineering, University of New South Wales, Sydney, NSW 2052, Australia c Centre for Marine Biofouling and Bio-innovation, University of New South Wales, Sydney, NSW 2052, Australia Received 20 December 2002; accepted 13 May 2003

Abstract A major barrier to the long-term use of medical devices is development of infection. Staphylococcus epidermidis is one of the most common bacterial isolates from these infections with biofilm formation being their main virulence factor. Currently, antibiotics are used as the main form of therapy. However with the emergence of staphylococcal resistance, this form of therapy is fast becoming ineffective. In this study, the ability of a novel furanone antimicrobial compound to inhibit S. epidermidis adhesion and slime production on biomaterials was assessed. Furanones were physically adsorbed to various biomaterials and bacterial load determined using radioactivity. Slime production was assessed using a colorimetric method. Additionally, the effect of the furanone coating on material surface characteristics such as hydrophobicity and surface roughness was also investigated. The results of this study indicated that there was no significant change in the material characteristics after furanone coating. Bacterial load on all furanonecoated materials was significantly reduced (po0.001) as was slime production (po0.001). There is a potential for furanone-coated biomaterials to be used to reduce medical device-associated infections. r 2004 Elsevier Ltd. All rights reserved. Keywords: Biomaterials; Anti-bacterial; Biofilm; Bacterial adhesion; Surface modification

1. Introduction The use of medical implants has increased immensely over the last decade. This ranges from the simple use of catheters to draw blood to life saving devices such as the total artificial heart. This has resulted in not only a better quality of life but also longer survival of patients. However, a major barrier to long-term use of these implants is development of infection. Implant-related infection can result in high morbidity and mortality for the patient and can also increase hospital costs. It is now well accepted that the coagulase negative staphylococci (CNS) are a major cause of implant related infections [1,2]. CNS are non-pathogenic microorganisms that reside in harmony on the human skin and various mucous membranes without causing harm. Staphylococcus epidermidis is one of the most common CNS isolated from medical device infections. *Corresponding author. Tel.: +61-2-9385-3905; fax: +61-2-663 2108. E-mail address: [email protected] (L.A. Poole-Warren). 0142-9612/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2004.02.051

Until recently the appearance of S. epidermidis on a medical device was dismissed as contamination due to its large numbers and ubiquitous distribution. However, despite having low pathogenic potential under normal circumstances, this microorganism has now evolved into being the leading cause of infection in the immunocompromised host or in the presence of a medical device. There are two main characteristics of S. epidermidis that allow persistence of infection. These are the ability of the bacteria to adhere onto surfaces in multilayered cell clusters, followed by the production of a mucoid substance more commonly known as slime or glycocalyx [3,4]. The adherent bacteria and slime are collectively known as biofilm. Once embedded in this biofilm layer the microorganisms are protected from the host’s immune cells and from the action of anti-microbial agents [5–7]. In many cases, the only effective therapy for these infections is removal and replacement of the device. An alternative approach to overcome this problem is based on the prevention of biofilm formation.

ARTICLE IN PRESS 5004

J.K. Baveja et al. / Biomaterials 25 (2004) 5003–5012

Researchers in the past have used various methods to try and prevent bacterial adhesion and slime production to biomaterials. These include physical and chemical modification of the biomaterial surface. Bridgett et al. [8] tested bacterial adhesion to cerebrospinal fluid shunts coated with a hydrogel material that created a more hydrophilic surface. This coating, although effective in reducing bacterial adhesion, was difficult to apply uniformly. Silver-impregnated cuffs on catheters have been another approach based on the anti-microbial activity of silver ions [9]. However, this approach is limited by the degradation of the cuff resulting in the loss of the coated silver ions and thereby the antimicrobial activity [10]. Antibiotics such ciprofloxacin, gentamicin, minocycline, and rifampin have also been coated onto and incorporated into biomaterials with some success [11,12]. However, with the emergence of staphylococcal resistance, this form of therapy may be short-lived and may soon become ineffective [13]. Given the serious implications associated with medical device infections and the paucity of current successful preventative strategies, there is a need for development of methods designed to prevent bacterial colonisation and slime production on medical devices. In the marine environment, many organisms defend themselves against biofilm formation through production of specific defense chemical compounds. For instance, the subtidal red alga Delisea pulchra produces a unique class of halogenated furanones called fimbrolides that reside in vesicles on the surface of the algae and inhibit fouling of their surface by marine organisms. Fimbrolides are five-membered ring lactone natural products called furanones in this study. A variety of analogues of the natural furanones have been synthesised and evaluated for their efficacy. These compounds have been shown to possess potent anti-microbial activity against a number of Gram-positive and Gramnegative bacteria [14,15]. Other researchers have also demonstrated antimicrobial properties of other furanone compounds. Most of these studies have tested the efficacy of the furanone compounds against Staphylococcus aureus and Esherichia coli [16–19]. Kozminykh et al. [16] tested various analogues of 3(2 H) furanones and found varying antimicrobial activity against the S. aureus P-209 strain. Gein et al. [17] used 4-aroyl-3-hydroxy-2,5-dihydrofuran-2-ones and found activity against the same S. aureus strain and an E. coli M-17 strain. Khan and Husain, [18] published data that another derivative of furanones, the 3-arylidene-5-(biphenyl-4-yl)-2(3 H)-furanones were also effective as antimicrobial agents against S. aureus and E. coli ranging from an MIC of 10–100 mg/ml. More recently, 5H-furan-2-ones from fungal cultures of Aporpium caryae also were able to inhibit S. aureus ATCC 25923 and E. coli 25922 [19]. Additionally, there is also evidence of inhibition of growth of Bacillus

subtilis by furanone compounds [19–21]. To date, there have been few reports of furanone activity against S. epidermidis, which are the most commonly isolated pathogens from medical device infections. The aim of this study was to assess the effect of a furanone compound, 3-(10 -bromohexyl)-5-dibromomethylene-2(5H)-furanone on adhesion and slime production of S. epidermidis on polymeric materials commonly used in biomaterials for medical devices.

2. Materials and methods 2.1. Microorganism The S. epidermidis strain used in this study was ATCC 35984. This strain was isolated from a catheter sepsis case and is known to be a high slime producer [22]. The isolate was stored at –85 C until required. Frozen cultures were revived by plating on chocolate agar and cultures were maintained on chocolate agar plates and replated from frozen cultures every 1–2 weeks. 2.2. Biomaterials Six commercially available polymer materials commonly used for medical devices were used in this study. These included silicone (Resmed., North Ryde, Australia), poly(vinyl chloride), (PVC; Tuta Healthcare, Lane Cove, Australia), polyether polyurethane (PU; Dow Chemical Co., MI, USA), polyethylene (PE), polypropylene (PP), and polytetrafluorethylene (PTFE; Dotmar, North Ryde, Australia). Silicone was supplied as a sheet cast from a two-part (A and B) medical grade product. PTFE, PE, and PU were supplied as compression molded sheets whereas PVC and PP were supplied as extruded sheets. Squares of 1 cm  1 cm dimension were cut from the flat sheets of the materials for use in experiments. The biomaterials were prepared to remove contamination by soaking with agitation overnight in a solution of 2% Decon 90 prepared with Milli-Q water. The detergent solution was then drained and the samples rinsed 4–6 times in Milli-Q water followed by soaking in Milli-Q water overnight. Finally, the samples were dried in a laminar airflow hood, packaged, and sterilised using 100% ethylene oxide. Samples were degassed at room temperature for 7 days prior to use. 2.3. Furanone coating Polymer materials were coated with the furanone compound, 3-(10 -bromohexyl)-5-dibromomethylene2(5H)-furanone using physical adsorption. The chemical structure of this compound is shown in Fig. 1. A stock solution of the furanone compound (1 mg/ml) was prepared in ethanol and 40 ml of this solution was

ARTICLE IN PRESS J.K. Baveja et al. / Biomaterials 25 (2004) 5003–5012

Fig. 1. Structure of the furanone compound used in this study.

placed on each side of the 1 cm2 test samples. The ethanol was allowed to evaporate resulting in 40 mg/cm2 of furanone. Control samples were coated on both sides with 40 ml of ethanol without furanone added. The amount of furanone associated with each biomaterial was assessed using X-ray photoelectron spectroscopy (XPS). 2.3.1. X-ray photoelectron spectroscopy The amount of furanone on the surface of the biomaterials was analysed using XPS (ESCALAB220iXL, VG Scientific, West Sussex, England) at the start (0 h) and completion of the experiment (24 h). The X-ray source was monochromated Al K alpha and the photoenergy was 1486.6 eV with a source power of 120 W. Five points of 0.5 mm2 were examined for each material and the average taken. The results are presented as the atom percent of bromine (%Br)7standard deviation as an indicator of the furanone compound present on the surface. XPS was also carried out on the control squares at both time points.

5005

Scanning electron microscopy (SEM) was performed on the samples following 1 and 24 h incubations. Following the washing step, the bacteria on the biomaterials were fixed by placing in 2.5% gluteraldehyde (v/v) in 0.1 m cacodylate buffer for 2 h. The samples were then dried with ethanol steps—50% aqueous ethanol (v/v) for 15 min, 75% aqueous ethanol (v/v) for 15 min, 95% aqueous ethanol (v/v) for 15 min, 100% aqueous ethanol (v/v) for 15 min, and 100% aqueous ethanol (v/v) for 15 min. The biomaterials were then sputter coated with gold using a Polaroan SC150 sputter coater and examined using a Hitachi S360 scanning electron microscope. 2.5. Slime assay Three FC and 3 control squares of each biomaterial were used for each of the 3 experiments carried out. Bacteria were prepared and exposed to biomaterials as in the adhesion assay with the exception of the addition of the tritiated thymidine. At the completion of the 24 h time period, a slime detection assay was performed according to Tsai et al. [23]. Briefly, slime on biomaterials was fixed using Carnoy’s solution consisting of glacial acetic acid, chloroform and absolute alcohol (1:3:6, v/v) and stained using 0.1% toluidine solution. The slime was then hydrolysed by suspending the biomaterials in a 0.2 m NaOH solution and heating in a water bath at 85 C for 1 h. The colour was quantified with a spectrophotometer at 590 nm. Slime production was expressed as optical density (OD)/cm271 standard error of the mean.

2.4. Adhesion assay 2.6. Biomaterial surface characteristics Three furanone-coated (FC) and 3 control (C) squares of each biomaterial were used for each experiment and experiments were repeated 3 times. Bacteria were grown for 16 h in 10 ml tryptone soy broth (TSB) supplemented with 0.25% glucose and 8 mCi/ml of tritiated thymidine, (specific activity of 40–60 Ci/mmol Amersham, Buckinghamshire, England). The bacterial culture was adjusted using a spectrophotometer to give an optical density at 660 nm of 0.1 (OD660=0.1). The viable count of this preparation was approximately 1  107 colony forming unit (cfu)/ml. Bacterial suspension (1 ml) and 1 mCi of tritiated thymidine was placed with each biomaterial square and incubated at 37 C for 1 h. At the completion of the time period, the squares were washed three times in phosphate-buffered saline (PBS, pH 7.4). To determine the radioactive counts at 24 h, one ml of fresh TSB and 1 mCi of tritiated thymidine was added and incubated at 37 C for 24 h. The squares were subsequently washed twice in PBS and placed in 5 ml scintillation fluid. The radioactive counts were determined using a beta counter and expressed as counts per minute (cpm) per 2 cm271 standard error of the mean.

Surface characteristics of both furanone coated and control materials were analysed using atomic force microscopy (AFM) and contact angle measurements. These are described in detail below. 2.6.1. Atomic force microscopy (AFM) The roughness of the material surfaces was determined using the Dimension 3000 Scanning Probe Microscope. In particular, tapping mode atomic force microscopy (AFM) was used to analyse an area of 10  10 mm. The average roughness was then generated by the NanoScope software and represented as the arithmetic average of the deviations from the center plane in the chosen area. Three to five points were analysed on each square and the average taken. Results are expressed as mean7standard deviation of all measurements. 2.6.2. Contact angle measurements Biomaterial surface hydrophobicity was determined by contact angle measurements made using

ARTICLE IN PRESS 5006

J.K. Baveja et al. / Biomaterials 25 (2004) 5003–5012

a Rame! –Hart contact angle goniometer (Mountain Lakes, NJ, USA; model number 100-00). Distilled water was used as the contact angle liquid. A drop of distilled water was placed on the surface of the biomaterial and the image was captured by a camera equipped with the Rame! –Hart 2001 imaging software. The angle made by this drop of water with respect to the biomaterial was calculated by the software. Contact angles of 5 drops were analysed on each sample and the experiment was repeated twice. Results are expressed as mean 7 standard deviation of the measurements.

day, respectively [24]. The analysis was carried out using Minitab 13 (GLM). The Bonferroni method was used for the pairwise comparisons among materials [25]. The roughness and contact angle data were analyzed using two-way analysis of variance with factors T and M. Spearman’s rank correlation coefficient was used to test for association between variables [25].

3. Results 3.1. Furanone coating efficacy

2.7. Cytotoxicity of the furanone compound A cell growth inhibition assay was conducted to determine the effect of FC materials on mammalian cells. Earle’s L, NCTC clone 929 (Murine) cells were used in this study. The cells were grown in minimum essential media with non-essential amino acids (MEM/ NA) supplemented with 10% fetal bovine serum (FBS) and grown to confluency in plastic petri dishes. After a 24 h inoculation period, the growth medium was aspirated and replaced with extract supplemented medium prepared from the test materials as described below. The test materials were silicone with and without the furanone coating. The materials were extracted by incubating 3 cm2 in 2.5 ml saline for a period of 72 h at 50 C. This extract was diluted 1:3 with 1X Eagle’s minimum essential medium (EMEM) to give a final concentration of 25%. Negative (saline) and positive (5% ethanol) controls were also tested, prepared with 1X EMEM to give a final concentration of 25%. The cell monolayer was then cultured with the different media for a further 48 h. At the completion of the test period, the cells were harvested and counted using a Coulter counter. Results were expressed as mean cell number 71 standard deviation. Cell growth inhibition was calculated as ð1  ðcell numbers in test=cell numbers in negative controlÞÞ  100

CGI greater than 30% were considered to represent significant inhibition of cell growth. 2.8. Statistical analyses A multiplicative model was assumed for the number of radioactive counts or the amount of slime measured on a specimen and statistical analyses were performed on the logarithms of the data. The count and slime data were analyzed separately by analysis of variance, with two fixed factors, treatment (furanone, control) and material (PVC, PE, PU, silicone, PTFE, PP), and one random factor, the day on which the experiment was conducted (1, 2, or 3). The fixed effects T, M, and T  M were tested over their respective interactions with D, where T, M, and D represent treatment, material and

The amount of furanone on the surface as determined by XPS at 0 and 24 h is given in Table 1. The amount of furanone on the surface of silicone and PP increased during the 24 h time period while it remained constant on PU. In the case of PVC and PE, the amount of furanone decreased over the time period. It was harder to coat the PTFE squares as the furanone compound tended to accumulate at one spot resulting in significantly varying XPS measurements from one spot to another. It was therefore difficult to determine the exact amount of furanone on PTFE. 3.2. Bacterial adhesion on furanone-coated and control biomaterials At the 1 h time point, the results showed that radioactive counts were low and similar on the FC and control biomaterials. Mean7standard deviation counts on the control biomaterials were 6467162 compared with 7687167 on the FC biomaterials. Bacterial distribution and surface morphology of the biomaterials appeared similar on both the control and the FC biomaterials at the 1 h time point. This is illustrated by the SEM images shown in Fig. 2. Another observation from the SEM images in Fig. 2 was the low bacterial numbers on the biomaterials supporting the results of the radioactive counts. Fig. 3 shows the radioactive counts per minute (cpm) per 2 cm2 of

Table 1 Percentage of bromine (% Br) on the biomaterial surface7standard deviation at 0 and 24 h Material

Percent of bromine (% Br) on the biomaterial surface (0 h)

Percent of bromine (% Br) on the biomaterial surface (24 h)

Silicone PTFE PVC PP PE PU

0.1170.07 5.274.03 0.4170.14 0.9070.68 3.3471.40 0.1370.05

0.2270.05 8.10715.11 0.0770.02 0.8470.54 0.5670.89 0.2570.03

ARTICLE IN PRESS J.K. Baveja et al. / Biomaterials 25 (2004) 5003–5012

5007

Control Furanone

cpm / 2 cm2

105

104

103 PVC

PE

PU

Sil

PTFE

PP

Fig. 3. Radioactive counts per minutes (cpm)71 standard error of the mean on the control and furanone-coated biomaterial squares at the 24 h time point.

Table 2 Ratio of the bacterial load on furanone-coated materials to the bacterial load on control materials (geometric mean, minimum and maximum) Material

PVC PE PU Sil PTFE PP

Fig. 2. SEM images of control and furanone-coated biomaterials at 0 h (1K X magnification)—PVC C (a), PVC F (b), PE C (c), PE F (d), PU C (e), PU F (f), Silicone C (g), Silicone F (h), PTFE C (i), PTFE F (j), PP C (k), PP F (l).

biomaterial at 24 h and indicates the total bacterial load associated with the surface. These results clearly show that the furanone compound was able to reduce bacterial load across all biomaterials (po0.001).

Ratio furanone/control Mean

Minimum

Maximum

0.33 0.11 0.14 0.03 0.06 0.10

0.15 0.05 0.04 0.02 0.02 0.06

0.85 0.20 0.31 0.08 0.15 0.18

Although the reduction in bacterial load by furanone was somewhat less on PVC than on other materials, this was possibly due to chance since the analysis of variance indicated that the T  M interaction was not significant (p=0.17). Table 2 shows the ratio of the bacterial load on FC materials to the bacterial load on the control materials. The estimated ratio of bacterial load on coated to control materials was 0.10 with a 95% confidence interval (0.06, 0.17). The differences among material were also highly significant (p=0.001). Pairwise comparisons of materials at a 5% significance level suggested that PE had significantly lower load than PTFE, silicone, or PU and that PP had a lower load than PU. There were significant differences among days but the interactions of T, M, and T  M with D were not significant. Thus the effects noted above were consistent across days. The results of the radioactive counts were further confirmed by the SEM images shown in Fig. 4. The SEM images showed most bacteria on the control silicone, PTFE, and PU materials as compared to PP, PE, and PVC. The bacteria had a tendency towards adhering to the edges of the PE squares and to adhere in small circular areas on the PP.

ARTICLE IN PRESS 5008

J.K. Baveja et al. / Biomaterials 25 (2004) 5003–5012

2.0 Control Furanone

OD570

1.5

1.0

0.5

0.0 PVC

PE

PU

Sil

PTFE

PP

Fig. 5. Bacterial slime production 71 standard error of the mean on control and furanone-coated biomaterial squares at the 24 h time point.

Table 3 Ratio of slime levels on furanone-coated materials to slime levels on control materials (geometric mean, minimum and maximum) Material

PVC PE PU Sil PTFE PP

Ratio furanone/control Mean

Minimum

Maximum

0.67 0.11 0.70 0.03 0.25 0.11

0.45 0.08 0.57 0.01 0.13 0.10

1.03 0.15 0.93 0.11 0.38 0.12

3.3. Slime quantitation on furanone-coated and control biomaterials Slime production was also significantly reduced on FC biomaterials (po0.001), as shown in Fig. 5, but the degree of reduction, which is tabulated in Table 3, varied widely across materials. Furanone reduced slime by an average of about 90% on PE, silicone, and PP, 70% on PTFE, and 30% on PVC and PU. The analysis of variance indicated that the T  M interaction was highly significant (po0.001). While the effect of furanone overall was also significant (po0.001), this must be interpreted in light of the dependence of the effect on the material. As with adhesion, there were significant differences among days but the interactions of T, M, and T  M with D were not significant. 3.4. Analysis of bacterial adhesion and slime production

Fig. 4. SEM images of control and furanoen-coated biomaterials at 24 h (150 X magnification)—PVC C (a), PVC F (b), PE C (c), PE F (d), PU C (e), PU F (f), Silicone C (g), Silicone F (h), PTFE C (i), PTFE F (j), PP C (k), PP F (l).

The colorimetric method used in this study for quantitation of slime is not specific for glycocalyx and may also stain biomass, thus the slime result reflects the total biofilm formation rather than slime alone. However, the fact that the T  M interaction was significant in the analysis of the slime data but not significant in the

ARTICLE IN PRESS J.K. Baveja et al. / Biomaterials 25 (2004) 5003–5012

analysis of the adhesion data suggests that the assay is detecting more than simply a change in biomass. Given that the adhesion assay detects bacterial biomass alone (via uptake of radiolabel) and the slime assay detects glycocalyx plus biomass, there should be an association between the two assays. For example, if the assay was detecting equivalent changes in slime production to the changes in biomass, this correlation coefficient would be expected to be close to 1. Calculation of Spearman’s rank correlation coefficient showed that it was 0.55 (p=0.017), which suggests some association between the assays, but highlights that there are more complex differential changes in slime production and biomass on the different materials tested. 3.5. Biomaterial characteristics

5009

Table 5 Average contact angles of biomaterials 7 standard deviation expressed as degrees Material

Control7SD ( )

Furanone-coated7SD ( )

PVC PE PU Silicone PTFE PP

7970.5 8172.7 6971.2 10871.3 11671.6 10372.5

80.670.6 8271.6 6971.9 11673.0 11675.0 9773.0

materials. This did not change significantly when the biomaterials were coated with the furanone compound. 3.6. Cytotoxicity of the furanone compound

In order to determine whether alterations in surface characteristics may account for a lower bacterial load following the furanone coating, the contact angles and surface roughness were investigated. 3.5.1. Roughness of the biomaterials by atomic force microscopy AFM provides the mean roughness of the sample in nanometers. The roughness measurements for the FC and control materials are shown in Table 4. PU was the smoothest material for both the FC and control materials. PVC was the roughest control material, however, when coated with the furanone compound, PE became the roughest material with an average roughness measurement of 130 nm. Although there were substantial differences among the biomaterials, the furanone coating had no significant effect on roughness. 3.5.2. Hydrophobicity/hydrophilicity by contact angle measurements Contact angle measures the hydrophobicity or hydrophilicity of materials. A higher surface/water contact angle is an indication of a more hydrophobic material. Table 5 shows the contact angles for the FC and control biomaterials. The highest contact angle was 116 for PTFE and the lowest was 69 for PU for the uncoated

Table 4 Average roughness of biomaterials 7 standard deviation expressed in nm calculated from atomic force measurements Material

Control7SD (nm)

Furanone-coated (nm)

PVC PE PU Silicone PTFE PP

15876.5 128718.0 1277.4 2777.2 6476.2 4172.1

109719.0 130719.0 1773.6 36712.6 5677.4 5076.6

The CGI assay showed minimal inhibition of L929 cell growth when exposed to extracts of either silicone alone or FC silicone. Cell numbers after 48 h exposure to the material extracts and controls are shown in Fig. 6. The % inhibition was 15% for silicone material alone and 6% for the FC material. CGI for the positive control was 61% which was within the expected range.

4. Discussion In this study, various polymeric biomaterials were coated with a furanone compound by physical adsorption and the antimicrobial activity of the coating determined by assessing bacterial adhesion and slime production. The results showed that the bacterial load on control and FC biomaterials were similar at 1 h post inoculation indicating that initial irreversible adhesion was not altered by furanone coating. There was also not much difference in the bacterial load on the control materials. This has previously been reported by Gottenbos et al. [26], who found that at 1 h, bacterial attachment (S. epidermidis and Pseudomonas aeruginosa) was similar on poly(dimethylsiloxane), Teflon, PE, PP, PU, poly(ethylene terephthalate), poly(methyl methacrylate), and glass. However, at 24 h, the furanone coating significantly inhibited bacterial load and slime production on all the biomaterials tested. This indicated that the furanones are able to have an effect on bacterial growth and biofilm formation once the bacteria are exposed to the surface adsorbed furanone. Significant variability in microbial adhesion and biofilm production on the different biomaterials without furanone coatings was observed. Similar results have been reported previously in the literature. Arciola et al. [27] found that there was a difference in bacterial adhesion between seven different silicone-based materials. Similarly, Ludwika et al. [28] found a varying degree

ARTICLE IN PRESS J.K. Baveja et al. / Biomaterials 25 (2004) 5003–5012

5010

8.00E+05 7.00E+05 Cells per plate

6.00E+05 5.00E+05 4.00E+05 3.00E+05 2.00E+05 1.00E+05 0.00E+00 Negative (Saline)

Positive (5% ethanol)

Silicone with furanone

Silicone without furanone

Extract Type Fig. 6. Cell growth inhibition assay to determine the cytotoxicity of the furanone coating when placed on the silicone material. Comparison of the cells per plate for the negative (saline), positive (5% ethanol), and silicone with the furanone compound and without the furanone compound. The furanone-coated silicone material was non-cytotoxic as there was no significant inhibition of the L929 cells.

of bacterial attachment to different synthetic polymers including silicone rubber, PP, polyetherurethane, and cellulose acetate. In another study, Sheth et al. [29] found more CNS on commercially available PVC than Teflon catheters in an in vitro study using the rolling technique on blood agar plates. However, in the present study, there was greater adhesion to the PTFE material than PVC. This difference may possibly be attributed to different sources of biomaterials resulting in different surface properties, variations in methodology used and bacterial strains. Bacterial adhesion depends on various factors including the nature of the polymer material, cell surface characteristics of the bacteria and environmental factors such as temperature, surrounding medium, and bacterial concentration [30]. The bacterial characteristics and environmental factors were kept constant between the FC and control materials; however, material surface characteristics may be altered by the coating process or by the presence of the coating itself. Critical material characteristics that can influence bacterial adhesion to biomaterials are surface chemistry [27], the related factor of surface energy/hydrophobicity [28,31], and surface roughness [32]. The results of this study indicate that there are no significant differences in surface roughness and hydrophobicity (Tables 4 and 5) between the FC and control materials. This suggests that the reduction in bacterial load and slime production can be attributed to the presence of the furanone compound on the biomaterial surface. It does not however account for the variability of adhesion on different uncoated biomaterials. The influence of surface hydrophobicity on bacterial adhesion still remains controversial. Some authors have reported that increasing hydrophobicity of polymeric biomaterials increased staphylococcal adhesion [28,33].

Recently, Drake et al. [34] also observed this trend with a titanium surface. They found that a strain of Streptococcus sanguis adhered less to a more hydrophilic titanium sample. There have also been reports of decreased bacterial adhesion when the biomaterial has been modified to make it more hydrophilic [35]. However, others have found no correlation between hydrophobicity of a material and bacterial adhesion [36,37]. The results from this study also show no significant correlation between the hydrophobicity of the material and bacterial load. Similarly, the correlation between the roughness of a biomaterial and bacterial adhesion remains to be fully understood. Tebbs et al. [38] tested the effect of surface roughness on the adhesion of the S. epidermidis ATCC 35984 strain. They reported a reduced amount of bacterial adhesion to a smooth central venous catheter (CVC) as compared to one with a rougher surface. Other researchers have found similar effects of increased bacterial colonisation on rougher biomaterials [32,35,39]. However, An [30] found no difference in bacterial adhesion with roughened titanium implants. In this study, materials were coated by physical adsorption. While most studies use covalently bound antimicrobials [40–42], physical adsorption of antibiotics has also been previously reported in the literature [43]. In this study, XPS data suggested that there was no bromine present on the control samples (data not shown). This indicated that the bromine detected on the coated samples must be due to the presence of the furanone compound. The results also suggested that furanones persisted on the material surfaces following incubation for 24 h. The limitations of physically coating materials include loss of furanones into solution during incubation as seen with the decreased bromine (%Br) on PVC and PE and variation in uniformity of the coating

ARTICLE IN PRESS J.K. Baveja et al. / Biomaterials 25 (2004) 5003–5012

on the biomaterial surface. The variation in uniformity could account for the slight increase in the bromine (%Br) with different spots being sampled at 0 and 24 h. Slime is a feature of biofilm formation. It is well established now that slime is the main virulence factor of S. epidermidis. Antibiotics do not provide effective therapy once a biofilm has been established by the bacteria [44]. This is thought to be due to the slower penetration and uptake of the antibiotics but more recently is believed to be due to the change in the bacterial population in the slime matrix [45]. This group was able to show thorough proteomic analysis that there was a change in over 800 proteins of Pseudomonas aeruginosa when the bacteria were in a biofilm as compared to the planktonic culture cells. This study shows that furanones are able to reduce biofilm production on all biomaterials studied. Although this reduction may be due to the reduced amount of bacteria present on the FC biomaterials, slime production itself is possibly also reduced by a direct effect on the gene encoding adhesion and slime production. This mechanism is currently being investigated in our laboratory. There are various other mechanisms that could also account for the action of furanones on reduced slime production. Furanone compounds have been shown to be able to inhibit quorum sensing in various gram positive and gram negative bacteria [20,46–48]. Additionally, furanones have also been shown to inhibit the 2 component system found in bacteria [49,50]. For an antimicrobial coating to be used for the prevention of infection on medical devices, it should not cause significant adverse host responses. Cytotoxicity testing of the furanone coating on silicone material at the concentration that was antimicrobial showed no significant reduction in L929 numbers. Unpublished data from our laboratory also suggest that once this compound has been bound to a material, it passes bacterial reverse mutation assay mutagenicity tests (AMES). Other furanone compounds have been shown to be both cytotoxic and mutagenic [51–53]. However, for the purpose of being used as potential coatings, the furanones in this study are both non-cytotoxic and nonmutagenic when bound to a material, which is relevant to medical device applications. In conclusion, this study has indicated that physically adsorbed furanones could be used as coatings to prevent implant-related infections in a clinical environment. This is especially applicable to commonly used biomaterials for implantable devices such as silicone, ePTFE and PP. There are however still some limitations to this study due to the nature of physically adsorbing the compound onto the biomaterial. This issue is currently being examined with covalently bound furanone polymers to assess if the activity is retained and will be supplemented by clarification of the mechanisms of action of furanones bound to surfaces.

5011

Acknowledgements The authors would like to thank Ms. Kate Noble for help with SEM work, Ms. Katie Levick, and Ms. Margaret Budanovic for help with AFM work, Dr. Ashley Jones for help with contact angle measurement and Ms. Lynn Ferris for conducting the cytotoxicity assay. This research was supported by the Cooperative Research Centre for Eye Research and Technology (CRCERT) and the Faculty of Engineering, UNSW.

References [1] Christensen GD, Baldassarri L, Simpson WA. Colonisation of medical devices by coagulase-negative staphylococci. In: Bisno AL, Waldovogel FA, editors. Infections associated with indwelling medical devices, 2nd ed. Washington DC 20005: American Society for Microbiology; 1994. [2] Rupp ME, Archer GD. Coagulase-negative staphylococci: pathogens associated with medical progress. Clin Infect Dis 1994;19:231–45. [3] Peters G, Locci R, Pulverer G. Adherence and growth of coagulase-negative staphylococci on surfaces of intravenous catheters. J Infect Dis 1982;146:479–82. [4] Franson TR, Sheth NK, Rose HD, Sohnle PG. Scanning electron microscopy of bacteria adherent to intravascular catheters. J Clin Microbiol 1984;20:500–5. [5] Gray ED, Peters G, Verstegen M, Regelmann WE. Effect of extracellular slime substance from Staphylococcus epidermidis on human cellular immune response. Lancet 1984;18:365–7. [6] Johnson GM, Lee DA, Regelmann WE, Gray ED, Peters G, Quie PG. Interference with granulocyte function by Staphylococcus epidermidis slime. Infect Immun 1986;54:13–20. [7] Duguid IG, Evans E, Brown MR, Gilbert P. Effect of biofilm culture upon the susceptibility of Staphylococcus epidermidis to tobramycin. J Antimicrob Chemother 1992;30:803–10. [8] Bridgett MJ, Davies MC, Denyer SP, Eldridge PR. In vitro assessment of bacterial adhesion to hydromer-coated cerebrospinal fluid shunts. Biomaterials 1993;14:184–8. [9] Maki DG, Cobb L, Garman JK, Shapiro JM, Ringer M, Helgerson RB. An attachable silver-impregnated cuff for prevention of infection with central venous catheters: a prospective randomized multicenter trial. Am J Med 1988;85:307–14. [10] Raad I. Intravascular-catheter-related infections. Lancet 1998;351:893–8. [11] Raad I, Darouiche RO, Hachem R, Sacilowski M, Bodey GP. Antibiotics and prevention of microbial colonization of catheters. Antimicrob Agents Chemother 1995;39:2397–400. [12] Schierholz JM, Steinhauser H, Rump AF, Berkels R, Pulverer G. Controlled release of antibiotics from biomedical polyurethanes: morphological and structural features. Biomaterials 1997;18: 839–44. [13] Lelievre H, Lina G, Jones ME, Olive C, Forey F, RousselDelvallez M, Nicolas-Chanoine MH, Bebear CM, Jarlier V, et al. Emergence and spread in French hospitals of methicillin-resistant Staphylococcus aureus with increasing susceptibility to gentamicin and other antibiotics. J Clin Microbiol 1999;37:3452–7. [14] Kjelleberg S, Steinberg PD, Holmstrom C, Back A. Inhibition of gram positive bacteria. Patent No. PCT/AU99/00284, 1999. [15] Holmstrom C, Kjelleberg S. Microbial inhibitory compositions. Patent No. PCT/AU01/00295, 2001.

ARTICLE IN PRESS 5012

J.K. Baveja et al. / Biomaterials 25 (2004) 5003–5012

[16] Kozminykh VO, Igidov NM, Kozminykh EN, Aliev ZG. Reactions of 5-aryl-furan-2,3-diones with acylmethylenetriphenylphosphoranes: synthesis and biological activity of 3(2 H)furanone derivatives. Pharmazie 1993;48:99–106. [17] Gein VL, Gein LF, Bezmaternykh EN, Voronina EV. Synthesis and antimicrobial activity of 4-aroyl-3-hydroxy-2,5-dihyrofuran2-ones and their derivatives. Pharm Chem J 2000;34:254–6. [18] Khan MS, Husain A. Syntheses and reactions of some new 2arylidene-4-(biphenyl-4-yl)-but-3-en-4-olides with a study of their biological activity. Pharmazie 2002;57:448–52. [19] Levy LM, Cabrera GM, Wright JE, Seldes AM. 5H-furan-2-ones from fungal cultures of Aporpium caryae. Phytochemistry 2003;62:239–43. [20] Ren D, Sims JJ, Wood TK. Inhibition of biofilm formation and swarming of Bacillus subtilis by (5Z)-4-bromo-5-(bromomethylene)-3-butyl-2(5H)-furanone. Lett Appl Microbiol 2002;34:293–9. [21] Darouiche RO. Device-associated infections: a macroproblem that starts with microadherence. Clin Infect Dis 2001;33:1567–72. [22] Christensen GD, Simpson WA, Younger JJ, Baddour LM, Barrett FF, Melton DM, Beachey EH. Adherence of coagulasenegative staphylococci to plastic tissue culture plates: a quantitative model for the adherence of staphylococci to medical devices. J Clin Microbiol 1985;22:996–1006. [23] Tsai CL, Schurman DJ, Smith RL. Quantitation of glycocalyx production in coagulase-negative Staphylococcus. J Orthop Res 1988;6:666–70. [24] Zar JH. Biostatistical analysis, 4th ed. Upper Saddle River, NJ: Prentice-Hall; 1999. [25] Quinn GP, Keough MJ. Experimental design and data analysis for biologists. Cambridge: Cambridge University Press; 2002. [26] Gottenbos B, van der Mei HC, Busscher HJ. Initial adhesion and surface growth of Staphylococcus epidermidis and Pseudomonas aeruginosa on biomedical polymers. J Biomed Mater Res 2000;50:208–14. [27] Arciola CR, Cenni E, Caramazza R, Pizzoferrato A. Seven surgical silicones retain Staphylococcus aureus differently in vitro. Biomaterials 1995;16:681–4. [28] Ludwicka A, Jansen B, Wadstrom T, Pulverer G. Attachment of staphylococci to various synthetic polymers. Zentralbl Bakteriol Mikrobiol Hyg [A] 1984;256:479–89. [29] Sheth NK, Rose HD, Franson TR, Buckmire FL, Sohnle PG. In vitro quantitative adherence of bacteria to intravascular catheters. J Surg Res 1983;34:213–8. [30] An YH, Friedman RJ. Concise review of mechanisms of bacterial adhesion to biomaterial surfaces. J Biomed Mater Res 1998;43:338–48. [31] John SF, Derrick MR, Jacob AE, Handley PS. The combined effects of plasma and hydrogel coating on adhesion of Staphylococcus epidermidis and Staphylococcus aureus to polyurethane catheters. FEMS Microbiol Lett 1996;144:241–7. [32] Verheyen CC, Dhert WJ, de Blieck-Hogervorst JM, van der Reijden TJ, Petit PL, de Groot K. Adherence to a metal, polymer and composite by Staphylococcus aureus and Staphylococcus epidermidis. Biomaterials 1993;14:383–91. [33] Galliani S, Cremieux A, van der Auwera P, Viot M. Influence of strain, biomaterial, proteins, and oncostatic chemotherapy on Staphylococcus epidermidis adhesion to intravascular catheters in vitro. J Lab Clin Med 1996;127:71–80. [34] Drake DR, Paul J, Keller JC. Primary bacterial colonization of implant surfaces. Int J Oral Maxillofac Implants 1999;14:226–32. [35] Kristinsson KG. Adherence of staphylococci to intravascular catheters. J Med Microbiol 1989;28:249–57.

[36] Espersen F, Wurr M, Corneliussen L, Hog AL, Rosdahl VT, Frimodt-Moller N, Skinhoj P. Attachment of staphylococci to different plastic tubes in vitro. J Med Microbiol 1994;40:37–42. [37] Karakecili AG, Gumusderelioglu M. Comparison of bacterial and tissue cell initial adhesion on hydrophilic/hydrophobic biomaterials. J Biomater Sci Polym Ed 2002;13:185–96. [38] Tebbs SE, Sawyer A, Elliott TS. Influence of surface morphology on in vitro bacterial adherence to central venous catheters. Br J Anaesth 1994;72:587–91. [39] Bellon JM, N GH, Jurado F, Carranza A, Bujan J. In vitro interaction of bacteria with polypropylene/ePTFE prostheses. Biomaterials 2001;22:2021–4. [40] Yorganci K, Krepel C, Weigelt JA, Edmiston CE. In vitro evaluation of the antibacterial activity of three different central venous catheters against gram-positive bacteria. Eur J Clin Microbiol Infect Dis 2002;21:379–84. [41] Gottenbos B, van der Mei HC, Klatter F, Nieuwenhuis P, Busscher HJ. In vitro and in vivo antimicrobial activity of covalently coupled quaternary ammonium silane coatings on silicone rubber. Biomaterials 2002;23:1417–23. [42] Maki DG, Stolz SM, Wheeler S, Mermel LA. Prevention of central venous catheter-related bloodstream infection by use of an antiseptic-impregnated catheter. A randomized, controlled trial. Ann Intern Med 1997;127:257–66. [43] Kockro RA, Hampl JA, Jansen B, Peters G, Scheihing M, Giacomelli R, Kunze S, Aschoff A. Use of scanning electron microscopy to investigate the prophylactic efficacy of rifampinimpregnated CSF shunt catheters. J Med Microbiol 2000;49: 441–50. [44] Anwar H, Strap JL, Chen K, Costerton JW. Dynamic interactions of biofilms of mucoid Pseudomonas aeruginosa with tobramycin and piperacillin. Antimicrob Agents Chemother 1992;36:1208–14. [45] Sauer K, Camper AK, Ehrlich GD, Costerton JW, Davies DG. Pseudomonas aeruginosa displays multiple phenotypes during development as a biofilm. J Bacteriol 2002;184:1140–54. [46] Givskov M, de Nys R, Manefield M, Gram L, Maximilien R, Eberl L, Molin S, Steinberg PD, Kjelleberg S. Eukaryotic interference with homoserine lactone-mediated prokaryotic signalling. J Bacteriol 1996;178:6618–22. [47] Gram L, de Nys R, Maximilien R, Givskov M, Steinberg PD, Kjelleberg S. Inhibitory effects of secondary metabolites from the red alga Delisea pulchra on swarming motility of Proteus mirabilis. Appl Environ Microbiol 1996;62:4284–7. [48] Ren D, Sims JJ, Wood TK. Inhibition of biofilm formation and swarming of Escherichia coli by (5Z)-4-bromo-5-(bromomethylene)-3-butyl-2(5H)-furanone. Environ Microbiol 2001;3:731–6. [49] England D, Kjelleberg S. Inhibition of two-component signal transduction systems. Patent No. PCT/AU00/01553, 2001. [50] Kjelleberg S, Rice S, McDougald D. Regulation of bacterial virulence. Patent No. PCT/AU01/01621, 2002. [51] LaLonde RT, Bu L, Henwood A, Fiumano J, Zhang L. Bromine-, chlorine-, and mixed halogen-substituted 4-methyl-2(5H)-furanones: synthesis and mutagenic effects of halogen and hydroxyl group replacements. Chem Res Toxicol 1997;10:1427–36. [52] Janecki T, Blaszczyk E, Studzian K, Rozalski M, Krajewska U, Janecka A. New stereocontrolled synthesis and biological evaluation of 5-(10 -hydroxyalkyl)-3-methylidenetetrahydro-2-furanones as potential cytotoxic agents. J Med Chem 2002;45: 1142–5. [53] Kim Y, Nam NH, You YJ, Ahn BZ. Synthesis and cytotoxicity of 3,4-diaryl-2(5H)-furanones. Bioorg Med Chem Lett 2002;12: 719–22.