An in vitro investigation of bacteria-osteoblast competition on oxygen plasma-modified PEEK Edward T. J. Rochford,1,2 Guruprakash Subbiahdoss,3 T. Fintan Moriarty,1 Alexandra H. C. Poulsson,1 Henny C. van der Mei,3 Henk J. Busscher,3 R. Geoff Richards1,2 1

AO Research Institute Davos, 7270 Davos, Switzerland Institute of Biological, Environmental and Rural sciences, Aberystwyth University, Wales, United Kingdom 3 Department of BioMedical Engineering, University of Groningen and University Medical Center Groningen, Groningen, The Netherlands 2

Received 7 November 2013; revised 5 February 2014; accepted 11 February 2014 Published online 25 February 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.a.35130 Abstract: Polyetheretherketone (PEEK) films were oxygen plasma treated to increase surface free energy and characterized by X-ray photoelectron microscopy, atomic force microscopy, and water contact angles. A parallel plate flow chamber was used to measure Staphylococcus epidermidis, Staphylococcus aureus, and U-2 OS osteosarcomal cell-line adhesion to the PEEK films in separate monocultures. In addition, bacteria and U-2 OS cells were cocultured to model competition between osteoblasts and contaminating bacteria for the test surfaces. Plasma treatment of the surfaces increased surface oxygen content and decreased the hydrophobicity of the materials, but did not lead to a significant difference in bacterial or U-2 OS cell adhesion in the monocultures. In the S. epidermidis coculture experiments, the U-2 OS cells adhered in greater numbers on the treated surfaces

compared to the untreated PEEK and spread to a similar extent. However, in the presence of S. aureus, cell death of the U-2 OS occurred within 10 h on all surfaces. The results of this study suggest that oxygen plasma treatment of PEEK may maintain the ability of osteoblast-like cells to adhere and spread, even in the presence of S. epidermidis contamination, without increasing the risk of preoperative bacterial adhesion. Therefore, oxygen plasma-treated PEEK remains a promising method to improve implant surface free energy for C 2014 Wiley Periodicals, Inc. J Biomed Mater osseointegration. V Res Part A: 102A: 4427–4434, 2014.

Key Words: Staphylococcus, biomaterial-associated infections, bacterial adhesion, coculture, PEEK, plasma treatment, osteoblasts, surface modification

How to cite this article: Rochford ETJ, Subbiahdoss G, Moriarty TF, Poulsson AHC, van der Mei HC, Busscher HJ, Richards RG. 2014. An in vitro investigation of bacteria-osteoblast competition on oxygen plasma-modified PEEK. J Biomed Mater Res Part A 2014:102A:4427–4434.

INTRODUCTION

Polyetheretherketone (PEEK) is a radiolucent polymer which, when compared with other orthopaedic polymers exhibits relatively high strength and also a high resistance to both chemical and physical degradation.1 In addition, PEEK is commonly implanted and used as a major component in spinal fusion cages and craniomaxillofacial implants2; however, it has yet to be as extensively investigated as other materials.3 Like many polymers, the relatively low surface free energy of this material limits osseointegration,4–6 potentially leading to fibrous encapsulation, implant loosening, and reduced wound healing. In addition, fibrous encapsulation can provide a niche for infection to develop isolated from the full extent of the immune system.7,8 Recent research has shown that a novel, stable, oxygen plasma treatment of PEEK enhances primary human osteoblast attachment, differentiation, and

mineralization in vitro.9,10 This is believed to be, in part, due to an increase in surface free energy through the incorporation of specific functional groups on the treated material. Previously we have shown by using a bacterial adhesion chamber based on the CDC Biofilm Reactor, that oxygen plasma treatment of orthopaedic grade PEEK OPTIMAV does not lead to a significant increase in bacterial adhesion in a protein-free, preoperative contamination model.11 In the current investigation, to further explore the effect of this surface modification on biomaterial-associated infection risk, bacterial adhesion and U-2 OS osteosarcomal cell-line adhesion and spreading on oxygen plasma-treated PEEK films were investigated using a parallel plate flow chamber (PPFC). In addition, both bacterial and eukaryotic cell adhesion were evaluated simultaneously using the coculture method developed by Subbiahdoss et al.12 R

The benefits accruing to the author or authors from a commercial or industrial party will be applied to a research fund, nonprofit institution or other organization with which the author(s) are associated. Additional Supporting Information may be found in the online version of this article. Correspondence to: T. F. Moriarty; e-mail: [email protected] Contract grant sponsor: Invibio Biomaterial Solutions

C 2014 WILEY PERIODICALS, INC. V

4427

MATERIALS AND METHODS

Sample preparation PEEK film, 100 mm in thickness, was supplied by Invibio Biomaterial Solutions (UK) and cut to a size of 75 3 25 mm2. The films were cleaned by ultrasonication in a series of propan-2-ol (Sigma Aldrich, Switzerland), 70% ethanol (Sigma Aldrich, Switzerland) and ultrapure water (MilliQ, 18.2 MX), twice for 15 min in each solution. Sample films were exposed to oxygen gas plasma in an EMITECH K1050X plasma cleaner (Quorum Technologies, UK) for 900 or 1800 s. Following plasma treatment the samples were washed twice for 30 min in ultrapure water.9,10 Untreated PEEK films were used as control materials. All of the films were steam sterilised in an autoclave (121 C, 20 min) before use. Surface characterization Surface chemistry was analyzed by X-ray photoelectron spectroscopy (XPS) as described previously.10,11 Additionally, contact angle measurements of 10 lL ultrapure water droplets (21 C) were attained using a Kruess Drop Shape Analysis system (Kruess, Germany). An atomic force microscope (AFM) fitted with a silicon nitride tip in tapping mode was used to assess topography in 60 lm2 areas, as previously described.10,11 Nanoscope 6.13R1(R) software (Digital instruments Veeco, 2002) was used for AFM data analysis. A range of roughness values were calculated to effectively compare the surfaces including: root mean squared roughness (Rq), mean roughness (Ra), topographical skew (Rsk), and the surface area within the scan. The roughness values were calculated as a mean from 3 or more scans. Bacterial growth conditions and harvesting Staphylococcus aureus V8189-94 (clinical isolate) and Staphylococcus epidermidis RP12 (ATCC 35983) were cultured from a frozen stock on tryptone soy agar (Oxoid, UK) plates and grown overnight at 37 C. The plates were then kept at 4 C. For each experiment a single colony was inoculated into 10 mL of tryptone soy broth (TSB: Oxoid, UK) and grown for 24 h at 37 C. This culture was then used to inoculate 200 mL of fresh TSB which was grown at 37 C for a further 18 h. The bacteria were harvested by centrifugation at 5000 3 g for 5 min at 10 C and washed twice with sterile phosphate buffered saline (PBS, 10 mM potassium phosphate, 0.15M sodium chloride, pH 7.0). The harvested bacteria were sonicated on ice (3 3 10 s) to break up any bacterial aggregates and the suspension was adjusted to contain 3 3 105 bacteria mL21 using a B€ urker-T€ urk counting chamber. Bacterial adhesion Bacterial adhesion was assessed using a PPFC and a phase contrast microscope.12 In preparation, the flow chamber was cleaned with 2% Extran MA02 soap (Merck, Germany) before being rinsed with tap water. Glass slides and sampleholding polymethyl methacrylate slides were sonicated for 3 min in 2% RBS35 soap solution (Omniclean, The Netherlands), rinsed with ultrapure water followed by methanol

4428

ROCHFORD ET AL.

(Biosolve, The Netherlands), ultrapure water again and sterilised in 70% ethanol. The sealed flow chamber and glass slides were then steam sterilised in an autoclave at 121 C for 20 min. A PEEK sample was fixed into the sample slot and the sterile PPFC was assembled. The flow chamber was filled with sterile PBS, taking care to remove any air bubbles and maintained at 37 C using a heating plate. The bacterial suspension (3 3 105 bacteria mL21) was then added to the chamber at a constant shear rate of 11 s21, by hydrostatic pressure as maintained by a Watson Marlow 505S peristaltic pump. Bacterial adhesion was measured using a CCD camera (Basler AG, Germany) mounted on a Leica DM2000 (Leica Microsystems, Germany) phasecontrast microscope. A modified version of MATLAB (vs. 6.5.1.) was used to capture images from the centre of the surface after the first adhesion event at predetermined intervals for 2.5 h. All of the images captured were a summation of 15 consecutive micrographs with 1 s time intervals in between so as to eliminate non-adherent moving bacteria from the final image. The density of adherent bacteria after 2.5 h was ascertained by manually counting the adherent bacteria in the micrographs. Osteoblast-like cell culturing and harvesting U-2 OS osteosarcomal cells (ATCC HTB-96) were cultured in low glucose (1 g L21) Dulbecco’s modified Eagle’s medium 1 Glutamax 1 pyruvate (DMEM: Invitrogen, The Netherlands) supplemented with 10% foetal bovine serum (FBS) and 0.2 mM of ascorbic acid-2-phosphate (AA2P: Sigma, The Netherlands), denoted as "DMEM1FBS". U-2 OS cells were expanded to 70–90% confluency in a humidified, 5% CO2 incubator at 37 C, and extracted using trypsin/ethylenediaminetetraacetic acid (Invitrogen, The Netherlands). At each passage, 6 3 106 cells were harvested and resuspended in 10 mL of modified culture media containing 98% DMEM1FBS and 2% TSB.12 U-2 OS cell adhesion The PPFC was cleaned, sterilized, and filled with PBS as described in the “Bacterial adhesion” section; however, a 4 sample-holding slide allowing the surfaces to be evaluated simultaneously was used. This method was first validated with untreated samples in each position (data not shown). The 10 mL U-2 OS suspension was added to the PPFC apparatus and allowed to fill the chamber. The cell suspension was kept static at 37 C for a further 1.5 h to permit adhesion. Following this incubation period, fresh, prewarmed modified culture medium was flowed through the PPFC at a shear rate of 0.14 s21 for 48 h. After 48 h flow was stopped and the sample slide was removed and fixed with 3.7% formaldehyde in cytoskeleton stabilization buffer (0.1M piperazine-N,N0 -bis(2-ethanesulfonic acid), 1 mM ethylene glycol tetraacetic acid, 4% w/v polyethylene glycol 8000, pH 6.9). Following fixation the cells were incubated in 0.5% Triton X-100 for 3 min, rinsed with PBS and stained for 30 min in 5 mL PBS containing 49 lL DAPI (Sigma, The Netherlands) and 2 lg mL21 TRITC-phalloidin (Sigma, The Netherlands). Any unbound stain was removed by rinsing with

AN IN VITRO INVESTIGATION OF BACTERIA-OSTEOBLAST COMPETITION

ORIGINAL ARTICLE

TABLE I. Surface Characteristics of the Plasma-Treated PEEK Films (mean 6 s.d.) Treatment Time (s) 0 900 1800

WCA (degrees)

Atom% Oxygen Content

82 6 3 53 6 2 51 6 4

12.0 6 0.9 13.1 6 0.2 15.2 6 0.3

PBS. The cell density and surface coverage (confluency) of each sample was subsequently assessed quantitatively using Leica Application Suite software connected to a Leica DM400B (Switzerland) fluorescent microscope and ScionImage software (Scion Corporation). U-2 OS and bacteria coculture adhesion The PPFC was cleaned, sterilized, and filled with PBS as described in the “Bacterial adhesion” section. Suspensions of bacteria and U-2 OS cells were also prepared as described in the "Bacterial growth conditions and harvesting" and "Osteoblast-like cell culturing and harvesting" sections respectively. Again, the 4 sample holding slide was used. First, bacteria, either S. epidermidis RP12 or S. aureus V8189-94, were flowed through the PPFC at a shear rate of 11 s21 for 15 min to give 1 3 104 adherent bacterial cm22 on the untreated PEEK films. Following initial bacterial adhesion, flow of the bacterial suspension was stopped and the nonadherent bacteria were removed by flowing PBS through the PPFC. Following the removal of nonadhered bacteria the PPFC was filled with U-2 OS cells which were permitted to adhere in static conditions for 1.5 h. Flow of modified DMEM for 48 h and subsequent fixing, staining, and visualization was performed as described in the “U-2 OS cell adhesion” section.12 Statistics Experiments were performed in three separate replicates (n 5 3). The mean data is shown in the results section with standard errors. PASW statistics (v.18, IBM SPSS) was used to analyze the data and a p-value < 0.05 considered significant. The difference between U-2 OS and bacterial adhesion to the treatments in mono-culture was analyzed by ANOVA with post-hoc LSD tests. Additionally, pairwise comparisons were made between the cell culture time-points using Ttests. For the coculture U-2 OS results Kruskal-Wallis tests were performed. RESULTS

Surface characterization Oxygen plasma treatment led to an increase in wettability as shown by a decrease in water contact angle by 30 and 33 , after 900 s and 1800 s exposure, respectively (Table I). The change in wettability was shown to be due to an increase in surface oxygen content by XPS. The surface oxygen increased by 1.1 atom% after 900 s and 3.2 atom% after 1800 s of oxygen plasma treatment (Table I). The results of the AFM analysis showed that 1800 s of exposure

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | DEC 2014 VOL 102A, ISSUE 12

to oxygen plasma decreased the Rq and Ra of the films (Table II, Supporting Information Fig. 1), though this difference was not significant (p 5 0.055). Bacterial adhesion There was no significant difference in bacterial adhesion between the different treated and untreated surfaces for both of the organisms included [Fig. 1(A), S. epidermidis RP12 p 5 0.300 and Fig. 1(B), S. aureus V8189-94 p 5 0.456]. However, the S. epidermidis strain adhered at a higher surface density than the S. aureus strain on all the surfaces. Over the 2.5 h experiment, S. epidermidis RP12 adhered at a steady rate of 19.5, 17.0, and 23.7 bacteriacm22 s21 for the untreated, 900 s and 1800 s treated surfaces, respectively. For S. aureus V8189-94 these values were lower at 12.5, 10.9, and 13.0 bacteriacm22 s21. U-2 OS osteoblast adhesion The initial adhesion density of the U-2 OS cells was similar on all of the surfaces, 9 3 104 cells cm22 after 1.5 h (p 5 0.250) [Fig. 2(A)]. After the following 48 h flow period, no difference was observed between the cell densities as a result of the different treatment times (p 5 0.735). Additionally, there was no significant net increase or decrease in cells over the 48 h flow period (p > 0.05). The confluency of the U-2 OS cells on the surfaces was also assessed at 48 h. There was no significant difference in surface coverage by the cells on all of the surfaces, each with confluency values of approximately 40% after 48 h (p 5 0.803) [Fig. 2(B)]. When confluency was correlated to cell density for all of the data collected, a linear relationship with an R2 value of 0.81 was calculated. This correlation suggests that individual cell spreading was similar on all of the surfaces and confluency was a function of cell density and not of oxygen plasma treatment. U-2 OS adhesion in the coculture model The 4 sample holding PMMA slide was validated and shown to produce similar results independently of sample position (data not shown). Additionally, because the adhesion of bacteria was shown to be independent of treatment, the number of adherent bacteria was presumed to be equal on the different surfaces prior to U-2 OS cell adhesion, at 1 3 104 bacteria cm22. The number of adherent U-2 OS cells was quantified after the initial 1.5 h static adhesion period and after the subsequent 48-h period with low shear flow. In the presence of S. aureus V8189-94, U-2 OS cell death and cytosol leakage was observed after 10 h on all surfaces (Supporting Information Fig. 2). However, in the TABLE II. AFM Roughness Values of the Plasma-Treated PEEK Films (mean 6 s.d.) Treatment Time (s)

Rq (nm)

Ra (nm)

Rsk

s.a. (mm)

0 900 1800

37 6 15 28 6 8 20 6 4

28 6 12 21 6 8 15 6 3

1.2 6 1.8 2.7 6 3.5 0.1 6 0.4

3602 6 3 3609 6 6 3606 6 3

4429

FIGURE 1. Mean bacterial adhesion to the PEEK films with and without an oxygen plasma treatment after 2.5 h in the PPFC for A: S. epidermidis RP12 and B: S. aureus V8189-94. 6 s.e. n 5 3. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

S. epidermidis RP12 coculture experiments, U-2 OS cells survived until after the final 48-h time-point, despite the presence of large quantities of bacteria. The initial number of adherent U-2 OS cells adhesion density was greater on both

the 900 s and 1800 s treated surfaces compared with the untreated controls; however, this was not statistically significant (p 5 0.079) [Fig. 2(C)]. The difference in the number of adherent cells was also reflected in the 48 h cells density

FIGURE 2. A: Mean U-2 OS cell density after 1.5 h of static conditions in the PPFC and 48 h of flow conditions for each of the untreated and oxygen plasma-treated surfaces. B: Mean U-2 OS cell confluency after 1.5 h of static conditions and 48 h of flow conditions in the PPFC for each of the untreated and oxygen plasma-treated surfaces. C: Mean number of adherent U-2 OS cells after 1.5 h of static conditions in the PPFC and 48 h of flow conditions in the coculture experiment with S. epidermidis for each of the untreated and oxygen plasma-treated surfaces. Mean U-2 OS cell confluence after 48 h of flow conditions in the PPFC coculture experiment with S. epidermidis for each of the untreated and oxygen plasma-treated surfaces. 6 s.e. n 5 3 *p < 0.05.

4430

ROCHFORD ET AL.

AN IN VITRO INVESTIGATION OF BACTERIA-OSTEOBLAST COMPETITION

ORIGINAL ARTICLE

FIGURE 3. Representative micrographs of U-2 OS cells after the 48 h culture period adhering to: A: untreated PEEK in the absence (left) and presence (right) of S. epidermidis RP12. B: 900 s oxygen plasma-treated PEEK in the absence (left) and presence (right) of S. epidermidis RP12. C: 1800 s oxygen plasma-treated PEEK in the absence (left) and presence (right) of S. epidermidis RP12. The bacteria in the coculture samples were removed by the wash steps before staining. Scale bar 75 mm, phalloidin stained actin cytoskeleton (red) and DAPI stained nuclei (blue). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

as the mean cell density was greater on the treated surfaces compared with the control. After 48 h, confluency was higher on the treated surfaces compared with the untreated control in the coculture experiments, and this difference was significant for the 1800 s treated surfaces (p 5 0.040) [Fig. 2(D)]. Interestingly, the mean confluence of U-2 OS cells on the 1800 s treated surfaces was 32% higher in the presence of bacteria than in the absence, despite the number of adherent cells and initial plating density being similar. However, due to the high variability of the results this difference was not significant (p 5 0.063). When the 48 h U-2 OS cell confluency was plotted against the 48 h cell density a good correlation was

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | DEC 2014 VOL 102A, ISSUE 12

observed (R2 5 0.88). This relationship suggests that confluency was influenced by cell number and not individual cell spreading. U-2 OS cell adhesion and spreading can be seen qualitatively in the micrographs of the fixed and stained cells after 48 h of culture in the PPFC both in the presence and absence of S. epidermidis RP12 (Fig. 3). The decrease in cell confluency on the untreated PEEK film from the coculture experiments is clear in comparison to the treated films.

DISCUSSION

The application of PEEK, like many other polymers, in orthopaedics is restricted due to the relatively low surface

4431

free energy of this biomaterial. This feature limits osseointegration, and therefore can lead to fibrous encapsulation of implanted devices.4 Not only does fibrous encapsulation reduce the efficacy of the body to heal around the device, but it also may provide a niche for infection to develop isolated from vascularization and the immune system.7,8 Recently, oxygen plasma treatment has been used to modify the PEEK surface with the aim of increasing its surface free energy to enhance osseointegration and reducing fibrous encapsulation.10,13 However, as with any surface modification, oxygen plasma treatment may influence bacterial adhesion and therefore increase or decrease infection risk. In our earlier in vitro studies, oxygen plasma treatment of PEEK was shown to increase primary human osteoblast adhesion and differentiation10 without increasing the risk of bacterial adhesion in a preoperative contamination model.11 Using a method of culturing eukaryotic cells on materials contaminated with bacteria in a PPFC,12 it is possible to assess the combined effects of biomaterial choice and bacterial contamination on the host response in vitro. Osteoblasts are important cells in the postoperative bone healing process around an implant. Therefore, the ability of these cells to spread and proliferate on the implant surface, even in the presence of contamination, is critical in the healing response of the host. Oxygen plasma treatment of the PEEK films led to an increase in surface oxygen and a decrease in water contact R angle similar to that previously reported for PEEK OPTIMAV discs.9,10 In the current investigation, bacterial adhesion to the plasma-treated films was investigated in PBS using an established PPFC method. Confirming the results of our previous work, the adhesion density of the bacteria on the treated surfaces did not differ significantly after 2.5 h. In the absence of proteins, bacteria interact with surfaces through non-specific chemical mechanisms. These interactions can be modelled using colloidal theories,14–18 such as the extended Derjaguin Landau Verwey and Overbeek (X-DLVO) theory,19 and include thermodynamic factors such as charge, hydrophobicity, and Van der Waals forces. Previously it has been shown that increasing the surface free energy of materials through plasma treatment can reduce bacterial adhesion.20,21 However, bacterial adhesion has previously been shown to be unaffected by the more subtle plasma treatment used in this study11 and the current result confirms this. By using U-2 OS osteosarcoma cells, the effect of oxygen plasma treatment of PEEK films on osteoblasts was investigated in the PPFC. In the presence of organic molecules, such as those in the modified culture media, eukaryotic cells interact with surfaces indirectly through adsorbed proteins.22 In this investigation, oxygen plasma treatment of PEEK films led to no significant difference in U-2 OS cell adhesion after 48 h compared with the untreated samples in the absence of bacteria. The difference between this and the earlier study could be attributed to a number of different factors. The source of the cell types used, primary or cellline, may alter the way in which the cells interact with the surfaces. The flow method used may also influence cell

4432

ROCHFORD ET AL.

adhesion compared to static conditions. Finally, the differences between the films and the injection moulded discs could influence cell adhesion. To attempt to clarify this discrepancy, primary human osteoblasts were cultured on plasmatreated PEEK films using a previously described method.10 Under these conditions, a clear improvement in adhesion and spreading was observed on the oxygen plasma-treated PEEK film surfaces compared to untreated films after 24 and 48 h. Therefore, the lack of significant difference between the treated and untreated materials in the PPFC experiments could be due to the decreased sensitivity of cell-line osteoblasts to adhere to suboptimal surfaces caused by the inherent up-regulation of adhesion protein production or because of the shear stress in the flow chamber. The coculture method used in this study was designed to model the effect of preoperative bacterial contamination of an implant surface on postoperative bone healing.12 When the U-2 OS cells were cocultured in the presence of S. aureus V8189-94, cell death was observed on all of the test surfaces within 10 h. Though the mechanism of the observed cell death was not further investigated, S. aureus is an aggressive pathogen, which actively releases cytotoxic products into the surrounding environment inducing apoptosis in osteoblasts.23 Previously it has been shown that internalised S. aureus cause osteoblast apoptosis via the TRAIL pathway.24 In contrast, S. epidermidis is less actively cytotoxic and not as readily internalised as S. aureus25; therefore, as in the current investigation, can be cocultured with eukaryotic cells. Lee et al. reported that in a similar microfluidic coculture system using titanium–aluminum– vanadium, S. epidermidis does not compete for the substrate, permitting normal initial MC3T3-E1 osteoblast cell adhesion.26 However, after 24 h, MC3T3-E1 osteoblast cells showed signs of membrane damage. The cell damage was attributed to environmental stress, such as a change in pH, caused by the presence of bacteria in the low shear system.26 Interestingly, in this study, when the U-2 OS cells were cocultured on the treated surfaces in the presence of S. epidermidis RP12, differences in both initial and 48 h adherent cell density and confluence were observed. In the presence of S. epidermidis, the ability of the osteoblast cells to adhere to untreated PEEK was reduced, while on the treated surfaces the presence of bacteria did not affect initial cell adhesion. In an earlier study using the coculture system, the ability of U-2 OS cells to adhere to glass and amine reactive esters (NHS)-reactive polyethylene glycol (PEG) coatings was reduced in the presence of bacteria.27 In the study by Saldarriaga Fernandez et al., the reduction in U-2 OS cell adhesion was attributed to the extracellular bacterial products present in the flow chamber which blocked the NHS functionalities required to enhance cell adhesion. In the current investigation the untreated surfaces may have adsorbed some of the bacterial products in a specific conformation that led to a reduction in U-2 OS cell adhesion compared with the treated surfaces. Alternatively, the bacteria may have been sensed by the U-2 OS cells causing the osteoblast-like cells to become more selective over adhesion sites. Osteoblasts are known to detect the

AN IN VITRO INVESTIGATION OF BACTERIA-OSTEOBLAST COMPETITION

ORIGINAL ARTICLE

presence of bacteria and bacterial products through the use of toll-like receptors.28 Upon detecting bacterial products, osteoblasts phenotypically differentiate and begin to produce signal molecules to initiate an immune response.28 The way in which the detection of bacteria affects osteoblast cell adhesion to biomaterials is yet to be fully explored; however, levels of stimulation may decrease the propensity of these cells to adhere to less suitable surfaces during the initial static adhesion period. This may explain why the osteosarcomal cells adhered nondiscriminately in the absence of bacteria, yet had an improved response to the treated surfaces in the presence of bacteria. An interesting result of these experiments was the increase in mean confluency after 48 h on the treated surfaces in the presence of bacteria compared with the same surface without. This increase in confluency due to the presence of bacteria could again be due to the change in phenotype associated with detecting pathogenic material. It is possible that specific levels of stimulation by bacteria may increase integrin expression. This has previously been reported for fibroblasts and HeLa cells in the presence of microbial stimuli.29,30 However, this feature has yet to be studied in greater depth and whether this is true of osteoblasts has yet to be verified. The results of the current study provide further evidence for an interesting interaction between bacterial cells and osteoblasts which should be further investigated in future work. The results of the bacterial adhesion and coculture experiments suggest that oxygen plasma treatment of PEEK continues to provide a promising method of increasing the surface free energy of PEEK for osseointegration without increasing the risk or impact of preoperative implant contamination. To decrease bacterial adhesion more effectively, other more active methods of modification may be considered.31 However, the effect of this treatment on the immune response to the treated materials and bacterial contamination may also be an important factor in the performance of these materials in vivo. CONCLUSIONS

In this investigation oxygen plasma-treated PEEK films were assessed for bacterial adhesion, U-2 OS osteosarcomal cell adhesion and coculture adhesion between bacteria and U-2 OS cells. Bacterial adhesion to the treated films was not significantly different to the untreated PEEK, thus confirming the findings of our earlier work.11 Additionally U-2 OS adhesion to the PEEK films was not affected by plasma treatment. In the coculture experiments, S. aureus rapidly killed the U-2 OS cells, while S. epidermidis caused a decrease in U-2 OS cell adhesion to the untreated surfaces. In contrast, U-2 OS cell adhesion to the treated surfaces in the presence of S. epidermidis RP12 remained high. These results together with previous studies show that oxygen plasma treatment of PEEK remains a promising treatment for increasing osseointegration without increasing the risk of infection. Additionally oxygen plasma treatment of PEEK may enhance osteoblast adhesion in the presence of contamination, thereby improving wound healing.

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | DEC 2014 VOL 102A, ISSUE 12

REFERENCES 1. Tant MR, McManus HLN, Rogers ME. High-temperature properties and applications of polymeric materials. In: Tant MR, McManus HLN, Rogers ME, editors. ACS Symposium Series. Oxford, UK: ACS; 1995. p 1–20. 2. Kurtz SM, Devine JN. PEEK biomaterials in trauma, orthopedic, and spinal implants. Biomaterials 2007;28:4845–4869. 3. Khonsari RH, Berthier P, Rouillon T, Perrin J, Corre P. Severe infectious complications after PEEK-derived implant placement: Report of three cases. In: 2013. 4. Kieswetter K, Schwartz Z, Dean DD, Boyan BD. The role of implant surface characteristics in the healing of bone. Crit Rev Oral Biol Med 1996;7:329–345. 5. Toth JM, Wang M, Estes BT, Scifert JL, Seim HB, III, Turner AS. Polyetheretherketone as a biomaterial for spinal applications. Biomaterials 2006;27:324–334. 6. Ul A, I, Bartnik A, Fiedorowicz H, Kostecki J, Korczyc B, Ciach T, Brabazon D. Surface modification of polymers for biocompatibility via exposure to extreme ultraviolet radiation. J Biomed Mater Res A 2013. [ePub ahead of print] 7. Arens S, Schlegel U, Printzen G, Ziegler WJ, Perren SM, Hansis M. Influence of materials for fixation implants on local infection. An experimental study of steel versus titanium DCP in rabbits. J Bone Joint Surg Br 1996;78:647–651. 8. Petty W, Spanier S, Shuster JJ, Silverthorne C. The influence of skeletal implants on incidence of infection. Experiments in a canine model. J Bone Joint Surg Am 1985;67:1236–1244. 9. Poulsson AHC, Richards RG. inventors. Polymer surface modification. Switzerland patent WO2009/149827A1; 2009. 10. Poulsson AHC, Richards RG. Surface Modification Techniques of PEEK; Including Plasma Surface Treatment. In: Kurtz SM, editor. PEEK Biomaterials Boston: Elsevier; 2011. p 145–162. 11. Rochford ETJ, Poulsson AHC, Salavarrieta Varela J, Lezuo P, Richards RG, Moriarty TF. Bacterial adhesion to orthopaedic implant materials and a novel oxygen plasma modified PEEK surface. Colloids Surfaces B: Biointerfaces 2014;113:213–222. 12. Subbiahdoss G, Kuijer R, Grijpma DW, van der Mei HC, Busscher HJ. Microbial biofilm growth vs. tissue integration: "the race for the surface" experimentally studied. Acta Biomater 2009;5:1399– 1404. 13. Poulsson AH, Eglin D, Zeiter S, Camenisch K, Sprecher C, Agarwal Y, Nehrbass D, Wilson J, Richards RG. Osseointegration of machined, injection moulded and oxygen plasma modified PEEK implants in a sheep model. Biomaterials 2014;35:3717–3728. 14. Azeredo J, Visser J, Oliveira R. Exopolymers in bacterial adhesion: Interpretation in terms of DLVO and XDLVO theories. Colloids Surfaces B: Biointerfaces 1999;14:141–148. 15. Bayoudh S, Othmane A, Mora L, Ben Ouada H. Assessing bacterial adhesion using DLVO and XDLVO theories and the jet impingement technique. Colloids Surf B Biointerfaces 2009;73:1–9. 16. Katsikogianni M, Missirlis YF. Concise review of mechanisms of bacterial adhesion to biomaterials and of techniques used in estimating bacteria-material interactions. Eur Cell Mater 2004;8:37–57. 17. Katsikogianni MG, Missirlis YF. Bacterial adhesion onto materials with specific surface chemistries under flow conditions. J Mater Sci Mater Med 2010;21:963–968. 18. Sharma PK, Rao KH. Analysis of different approaches for evaluation of surface energy of microbial cells by contact angle goniometry. Adv Colloid Interface Sci 2002;98:341–463. 19. van Oss CJ, Good RJ, Chaudhury MK. The role of van der Waals forces and hydrogen bonds in "hydrophobic interactions" between biopolymers and low energy surfaces. J Colloid Interface Sci 1986;111:378–390. 20. Amanatides E, Mataras D, Katsikogianni M, Missirlis YF. Plasma surface treatment of polyethylene terephthalate films for bacterial repellence. Surface Coatings Technol 2006;200:6331–6335. 21. Balazs DJ, Triandafillu K, Chevolot Y, Aronsson BO, Harms H, Descouts P, Mathieu HJ. Surface modification of PVC endotracheal tubes by oxygen glow discharge to reduce bacterial adhesion. Surface Interface Analy 2003;35:301–309. 22. Kasemo B, Lausmaa J. Material-tissue interfaces: The role of surface properties and processes. Environ Health Perspect 1994;102: 41–45.

4433

23. Alexander EH, Bento JL, Hughes FM Jr, Marriott I, Hudson MC, Bost KL. Staphylococcus aureus and Salmonella enterica serovar Dublin induce tumor necrosis factor-related apoptosis-inducing ligand expression by normal mouse and human osteoblasts. Infect Immun 2001;69:1581–1586. 24. Reott MA Jr, Ritchie-Miller SL, Anguita J, Hudson MC. TRAIL expression is induced in both osteoblasts containing intracellular Staphylococcus aureus and uninfected osteoblasts in infected cultures. FEMS Microbiol Lett 2008;278:185–192. 25. Valour F, Trouillet-Assant S, Rasigade JP, Lustig S, Chanard E, Meugnier H, Tigaud S, Vandenesch F, Etienne J, Ferry T, Laurent F, and Lyon BJI Study Group. Staphylococcus epidermidis in orthopedic device infections: the role of bacterial internalization in human osteoblasts and biofilm formation. PLoS One 2013;8: e67240. 26. Lee JH, Wang H, Kaplan JB, Lee WY. Effects of Staphylococcus epidermidis on osteoblast cell adhesion and viability on a Ti alloy surface in a microfluidic co-culture environment. Acta Biomater 2010;6:4422–4429. 27. Saldarriaga Fernandez IC, Busscher HJ, Metzger SW, Grainger DW, van der Mei HC. Competitive time- and density-dependent adhesion of staphylococci and osteoblasts on crosslinked poly(ethylene glycol)-based polymer coatings in co-culture flow chambers. Biomaterials 2011;32:979–984. 28. Marriott I. Osteoblast responses to bacterial pathogens: A previously unappreciated role for bone-forming cells in host defense and disease progression. Immunol Res 2004;30:291–308. 29. Engels-Deutsch M, Rizk S, Haikel Y. Streptococcus mutans antigen I/II binds to alpha5beta1 integrins via its A-domain and increases beta1 integrins expression on periodontal ligament fibroblast cells. Arch Oral Biol 2011;56:22–28.

4434

ROCHFORD ET AL.

30. Kim M, Ogawa M, Fujita Y, Yoshikawa Y, Nagai T, Koyama T, Nagai S, Lange A, Fassler R, Sasakawa C. Bacteria hijack integrinlinked kinase to stabilize focal adhesions and block cell detachment. Nature 2009;459:578–582. 31. Tateishi T, Kyomoto M, Kakinoki S, Yamaoka T, Ishihara K. Reduced platelets and bacteria adhesion on poly(ether ether ketone) by photoinduced and self-initiated graft polymerization of 2-methacryloyloxyethyl phosphorylcholine. J Biomed Mater Res A 2013. [ePub ahead of print] 32. Bruinsma GM, van der Mei HC, Busscher HJ. Bacterial adhesion to surface hydrophilic and hydrophobic contact lenses. Biomaterials 2001;22:3217–3224. 33. Krsko P, Kaplan JB, Libera M. Spatially controlled bacterial adhesion using surface-patterned poly(ethylene glycol) hydrogels. Acta Biomater 2009;5:589–596. 34. MacKintosh EE, Patel JD, Marchant RE, Anderson JM. Effects of biomaterial surface chemistry on the adhesion and biofilm formation of Staphylococcus epidermidis in vitro. J Biomed Mater Res A 2006;78:836–842. 35. Patel JD, Ebert M, Ward R, Anderson JM. S. epidermidis biofilm formation: Effects of biomaterial surface chemistry and serum proteins. J Biomed Mater Res A 2007;80:742–751. 36. Rowan B, Wheeler MA, Crooks RM. Patterning bacteria within hyperbranched polymer film templates. Langmuir 2002;18:9914– 9917. 37. Rozhok S, Fan Z, Nyamjav D, Liu C, Mirkin CA, Holz RC. Attachment of motile bacterial cells to prealigned holed microarrays. Langmuir 2006;22:11251–11254. 38. Whitehead KA, Verran J. The effect of surfact topography on the retention of microorganisms. Food Bioproducts Proc 2006;84:253–259.

AN IN VITRO INVESTIGATION OF BACTERIA-OSTEOBLAST COMPETITION