Biomaterials 35 (2014) 5261e5277

Contents lists available at ScienceDirect

Biomaterials journal homepage:

Poly (hexamethylene biguanide) adsorption on hydrogen peroxide treated TieAleV alloys and effects on wettability, antimicrobial efficacy, and cytotoxicity Gerald Müller a, Hicham Benkhai a,1, Rutger Matthes a,1, Birgit Finke b, Wenke Friedrichs c, Norman Geist c, Walter Langel c, Axel Kramer a, * a b c

Institute of Hygiene and Environmental Medicine, Walther-Rathenau-Str. 49a, University Medicine, D-17475 Greifswald, Germany Leibniz-Institute for Plasma Science and Technology (INP e.V.), Felix-Hausdorff-Str. 2, D-17489 Greifswald, Germany Institute of Biochemistry, Department of Biophysical Chemistry, Felix-Hausdorff-Str. 4, D-17487 Greifswald, Germany

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 January 2014 Accepted 14 March 2014 Available online 3 April 2014

An effective amount of the antiseptic agent PHMB cannot simply be placed on the surface of titanium alloys where hydrocarbons were removed by different purification procedures. Pre-treatment of Ti6Al4V specimen with 5% H2O2 in 24 h results in extra introduced eOH and eCOOH groups as well as an adsorbed water film on the surface, which provide the base for the subsequent formation of a relatively stable and multi-layered PHMB film. The superficially adhering PHMB film produces no adverse effects on MG63 cells within a 48 h-cell culture, but promotes the initial attachment and spreading of the osteoblasts on the modified Ti6Al4V surface within 15 min. After direct bacterial inoculation of the active sample, the PHMB film reacts antimicrobially against Gram-positive (Staphylococcus aureus, Staphylococcus epidermidis) and Gram-negative strains (Pseudomonas aeruginosa, Escherichia coli) after surface contact. The bactericidal efficacy is only slightly reduced after using of the same specimen for re-testing the antibacterial activity. MG63 cells adhere and proliferate within 48 h on a PHMB film-containing Ti6Al4V surface, which has been pre-contaminated with S. aureus. Bacterial biofilms were only revealed in controls without PHMB. Ó 2014 Elsevier Ltd. All rights reserved.

Keywords: Titanium-aluminum-vanadium alloy Hydrogen peroxide Antimicrobial PHMB Biocompatibility In vitro tests

1. Introduction While titanium implants are becoming more popular in clinical practice the prevalence of infections associated with these implants and other medical devices rises [1]. To prevent implant-associated infections, one approach is to improve the antimicrobial properties using antibacterial coatings [2]. For this, one way is to create a permanently effective, non-leaching and immobilised antimicrobial compound on the surface that prevents the attachment of bacteria or kills them on contact [3]. Another attempt is the controlled release of an antibiotic from a coated surface [4] suffering primarily from providing perfect conditions for resistance development for bacteria. Coating of titanium implant

* Corresponding author. Tel.: þ49 3834 51552; fax: þ49 3834 515541. E-mail address: [email protected] (A. Kramer). 1 Contributed equally. 0142-9612/Ó 2014 Elsevier Ltd. All rights reserved.

material with antiseptics [5,6], essential oils [7], antimicrobial peptides [8], and effective silver or copper ion implantation [9,10] may be good alternatives reducing the risk of drug resistance. The basis for a persistent microbicidal activity is the adsorption of effective amounts of the antimicrobial agent to the surface of the implant material. The antiseptic agent chlorhexidine digluconate, a biguanide, can be adsorbed on the surface of titanium demonstrating antimicrobial effects gradually over a period of several days [11]; but the risk of allergic and anaphylactic reactions limited the usage of chlorhexidine. Another biguanide, poly(hexamethylene biguanide) hydrochloride (PHMB), has been widely used as antimicrobial agent in cosmetics, as sanitizer, or in contact-lens multipurpose solutions lacking any systemic toxicity [12]. Both biguanides are directly adsorbed on cellulose by electrostatic forces and hydrogen bridge bonding [13,14] providing final products, which may be used as antimicrobial textiles. However, up to date PHMB has not been tested for direct adsorption on the titanium surface in order to create antimicrobial coated implants. The agent


G. Müller et al. / Biomaterials 35 (2014) 5261e5277

is known for being used for other antiseptics due to its significant promotion of wound healing [15]. Therefore, PHMB is the agent of choice for treatment of chronic wounds [16]. To increase the adsorption of desirable substances unknown impurities of the titanium surface should be removed quantitatively in combination with an oxidation process to create more binding sites. Plasma-cleaning [17], acid passivation procedures [18], exposure to UV light [19], plasma immersion ion implantation (PIII) [20], and other chemical or physical oxidation procedures [21] are useful to eliminate surface contaminants and to produce a consistent and reproducible titanium oxide surface layer. Hydrogen peroxide treatment generates oxidation and hydroxylation of the titanium surface removing the native layer by stripping away organic contaminants and producing a purified and disinfected final product [22]. Only few investigations [23,24] have examined the optimal H2O2 treatment concentration/exposure time without additional supplements (HCl, H2SO4, phosphate, electrolyte) improving protein adsorption and enhanced cell attachment to titanium. There are barely reports related to the surface modification of the T6Al4V alloy using H2O2. In the present study, the surface of Ti6Al4V alloy was modified using H2O2 to enhance the adsorption of the antiseptic agent PHMB. A simulation model for the increased adsorption of PHMB is presented based on both a theoretical model using standard calculation methods in classical all-atom force field theory [25,26] and the surface changes resulting from the H2O2 treatment procedure. The resulting Ti6Al4V specimens before and after PHMB coating were analyzed by XPS and characterized for wettability, effects on human cells, and antimicrobial activity.

positively charged ions of the plasma were accelerated from the plasma to the sample and implanted to the subsurface of the sample. The pulses had a rise time of 200 ns and a repetition rate of 50 Hz. For the implantation, a pulse length of 18 ms and a mean implantation current of 48 mA were used. The samples were heated to temperatures of about 600  C, correspondingly. For assessing the optimal hydrogen peroxide concentration each specimen was transferred into a cavity of a 24-well cell culture plate (Biochrom, TPP, Germany) with 1.0 ml of aqueous 0.5e30% (v/v) hydrogen peroxide (not stabilized, Sigma 95313, Germany). The hydrogen peroxide treatment was carried out at room temperature in the dark for 16 h. The optimal operation time was derived applying 5% H2O2 for 2e72 h. The standard operation procedure for specimen treatment was 5% H2O2 for 24 h. 2.3. Measurement of PHMB The microplate reader PowerWave XS (Biotek Instruments, Germany) was used to quantify aqueous PHMB concentrations based on the maximum absorption at 236e238 nm [27]. Concentrations between 0 and 20 mg/ml PHMB served to establish the calibration curve of 0.2 ml-samples in a 96-well quartz microplate (Hellma, 730.009 QG, Germany). PHMB adsorption on each Ti6Al4V specimen was conducted in 1.00 ml of 20 mg/ ml PHMB in 24-well cell culture plates (Biochrom, TPP, Germany). Cavities were precoated with aqueous PHMB solution for 24 h to avoid adsorption of PHMB on polystyrene. The non-adherent PHMB was quantified after 2 h of incubation at room temperature in the dark. The amount of adsorbed PHMB was calculated by using the difference of applied and residual PHMB concentration. To evaluate the stability of the PHMB linkage, Ti6Al4V discs after PHMB adsorption were placed in sterile water for 24, 48, and 72 h at room temperature. Release of PHMB was monitored by scanning the eluent at 230e240 nm. 2.4. Surface analysis

Commercial Ti6Al4V discs of 11 mm diameter and 2 mm thickness were purchased from DOT GmbH, Rostock, Germany. Specimens were manually polished by the manufacturer to mirror finish. The Ti6Al4V discs were delivered protected from scratching separately sealed in a plastic foil. Each specimen was transferred into a cavity of a 24-well cell culture plate (Biochrom, TPP, Germany) and washed for 24 h with 1.0 ml 80% (v/v) ethanol (RotisolvÒ HPLC gradient grade, Carl-Roth, Germany) at room temperature in the dark on an orbital shaker Titramax 1000 (Heidolph, Germany) followed by rinsing with 2  1.0 ml of sterile water (B. Braun, Germany), each for 1 h. Specimens were placed on a sterile cellulose tissue and dried under laminar airflow in a clean room cabin. Prior to surface treatment procedures and cell culture experiments the Ti6Al4V discs sealed in TyvecÒ-SteriClin foil (vp group, Germany) were ɣ-sterilized with 25 kGy at the Synergy Health Radeberg GmbH (Germany). Surface roughness of the ɣ-sterilized specimen was measured with a Veeco Dektak 3ST surface profiler (USA) using a velocity of 0.08 mm/s and a stylus force of 30 mg. The resulting surface parameters were: Ra ¼ 0.43  0.20 mm, Rq ¼ 0.52  0.22 mm, Rmax ¼ 0.11  0.06 mm, Rz ¼ 0.42  0.34 mm, and Rt ¼ 2.41  0.34.

The elemental chemical surface composition and chemical binding properties of the surface were determined by X-ray photoelectron spectroscopy (XPS). The Axis Ultra (Kratos, UK) ran with the monochromatic Al-Ka line at 1486 eV (150 W) under high vacuum conditions. The spot size for high-resolution measurements was limited to 250 mm. Charge neutralization was implemented. Spectra were recorded at pass energy of 80 eV for the estimation of the chemical elemental composition and of 10 eV for highly resolved C1s peaks. C1s, O1s, and Ti2p spectra were recorded at a take-off angle of 90 for quantification of surface composition. The peak fitting procedure was carried out with CasaXPS software version 2.2 (Casa Software Ltd., UK) using the GausseLorentz (30% Lorentz) distribution, Shirley baseline and a fixed FWHM (full width at half maximum). All values are presented in at.-% and corresponding element ratios. The CeC/CeH component of the aliphatic C1s peak was adjusted to 285.0 eV [28]. The other components of the C1s peak were fixed to known values: amines (eCeNH) and secondary carboxyls (OeC]O) at 285.7  0.1 eV; hydroxyls, ethers and imines, nitriles (eCeO, eC]N, eC^N) at 286.6  0.2 eV; aldehydes, ketones and amides (eC]O, eNeC]O) at 288.0  0.3 eV; esters and acids (OeC]O) at 289.2  0.2 eV. The FWHM was between 0.9 and 1.2 eV for the C1s highly resolved peak. The O1s spectra were fitted with three peaks: the main peak at 530 eV assigned to O1s in TiO2, TieOH, O]C at 531.2  0.2 eV and adsorbed H2O, OeC at 532.1  0.2 eV [28]. The FWHM for the O1s peak was between 1.1 and 1.5 eV. The Ti2p spectra were represented by two dominating peaks Ti2p3/2 at 458.9  0.2 eV and Ti2p1/2 at 464.5  0.2 eV for Ti4þ. Further oxidation states of titanium Ti(0)2p3/2 at 453.5  0.2 eV, Ti(II) and Ti(III) were included [28]. The FWHM was between 1 and 2.5 eV for the Ti2p3/2 and Ti2p1/2 peaks.

2.2. Surface treatment procedures

2.5. Wettability

For the Hellmanex treatment procedure specimen were ultrasonically cleaned for 30 min each in acetone (RotisolvÒ HPLC gradient grade, Carl-Roth, Germany), chloroform (SupraSolvÒ for gas chromatography, Merck, Germany), a 2% (v/v) Hellmanex III solution (Hellma, Germany) and sterile water (B. Braun, Germany). Acid passivation was performed to ASTM F86-04 [18]. The specimens were solvent-cleaned each for 5 min in 1.0 ml methyl ethyl ketone (ACS reagent, CarlRoth, Germany), rinsed for 15 min in 3  2.0 ml sterile water, acid-passivated for 30 min in 1.0 ml of 30% (v/v) nitric acid (Rotipuran, Carl-Roth, Germany), and again rinsed in 4  2.0 ml sterile water each for 5 min. Low temperature plasma treatments were performed for 1e30 min within the reactor V55G (PLASMA-finish, Germany) using a gas mixture of 60 ml O2/40 ml Ar at a power of 1200 W, a pressure of 100 Pa, a pulse rate of 10 ms on/90 ms off, and a distance of 110 mm. Plasma ion immersion implantation (PIII) was carried out for 15 min in a capacitively coupled radio-frequency discharge (13.56 MHz) at low working pressure (2 Pa) and a power of 20 W with coplanar electrodes in an ultrahigh vacuum reactor [10] for 15 min. The samples were completely bathed by immersion plasma in the ultrahigh vacuum reactor. Oxygen was used as working gas. A high voltagepulse-generator delivered also a high voltage pulse (5 kV) to the sample, which was negatively charged with reference to the plasma potential. Therefore, the

Wettability of the surface of Ti6Al4V specimens was examined by an optical water contact angle (WCA) measuring device (Dataphysics OCA40 Micro, Germany) using 1 ml of deionized water at 25  C and 45% humidity. WCAs were measured with the profiles of droplets deposited on the control and modified surfaces immediately after stabilization by the SCA software (Dataphysics, Germany).

2. Materials and methods 2.1. Ti6Al4V test specimen

2.6. Poly(hexamethylene biguanide) hydrochloride (PHMB) Cosmocil PG (Lot 0000695544, Arch Chemicals, USA) containing 20% (w/v) PHMB in water was used as the antibacterial agent. 2.7. PHMB Binding simulation Six different PHMB product types have been identified possessing combinations of amine, cyanoamine, guanidine, or cyanoguanidine end-groups [29]. A theoretical PHMB oligomer (n ¼ 9) terminated with amino and cyanoguanidine end-groupings [30] was used for all calculations (Fig. 1). PHMB was parameterized using the standard procedure of Köppen [31]. The PHMB oligomer was simulated without TiO2 and on different surfaces. A stable flexible model TiO2 surface of (100) rutile was obtained by combining default charge parameters [32] with appropriate van der Waals parameter. All simulations of

G. Müller et al. / Biomaterials 35 (2014) 5261e5277

+ H 3N (CH2)3 (CH2)3 HN Cl





NH NH (CH2)3 (CH2)3 HN HCl 9


Fig. 1. Structural formula of PHMB with nine hexamethylene biguanide units terminated by one amino and cyanoguanidine group.

structural and energetic properties of molecules were done using the AMBER (Assisted Model Building and Energy Refinement) force field tool kit [26] and the molecular dynamics package NAMD (Not (just) Another Molecular Dynamics Program) for high-performance simulation of large bio-molecular systems [33]. The simulation protocol contained energy optimization, temperature/pressure equilibration and final production runs. Hydroxylated surfaces were negatively charged, which was compensated by adding sodium ions to the solution. Five systems were simulated, (1) one PHMB oligomer in solution, (2),(3),(4) one PHMB molecule on TiO2 slabs in rutile modification with increasing hydroxylation and negative charge density, and (5) 64 PHMB molecules on a hydroxylated surface. All cells were filled up by explicit TIP3P (transferable intermolecular potential 3 points) water [34], which provides point charges on each of the three water atoms (þ0.417e for H and 0.834e for O) and is one of the most common and computationally efficient description of liquid water in force field calculations. The folding state of PHMB oligomers was characterized by the radius of gyration (rg) and the solvent accessible surface area (sasa). Adsorption was analyzed by normalized distance histograms between TiO2 and PHMB. Data were analyzed by the visual molecular dynamics (VMD) program [35].

3.82 mg phenazine methosulfate was diluted 1:3 with DMEM/10% FBS without phenol red). Specimens were incubated for 3 h in a humidified atmosphere of 5% CO2/95% air at 37  C. Thereafter, the absorbance of 100 ml culture medium containing colored formazan product was read in a 96-well plate by a spectrophotometric microplate reader PowerWave XS (BioTek Instruments, USA) at 450 nm. 2.12. Crystal violet staining Specimens with cultured cells were fixed in 1.0 ml 70% (v/v) ethanol for 30 min, dried at room temperature and covered with 1.0 ml 0.1% (w/v) aqueous crystal violet (Carl-Roth, Germany) for 5 min in a 24-well plate. The excess of dye was removed washing specimens twice with 2.0 ml sterile water each for 1 min using an orbital shaker. Specimens were dried at room temperature. The absorbed dye was eluted with 0.4 ml 2% (w/v) sodium desoxycholate (AppliChem, Germany). Plates were agitated for 30 min on an orbital shaker. Extracts were diluted with 0.6 ml sterile water. The absorbance of 100 ml was read in a 96-well plate by a spectrophotometric microplate reader PowerWave XS (BioTek Instruments, USA) at 590 nm. 2.13. Cell actin filament staining

2.8. Determination of titanium release The photometrical analysis of Ti4þ is based on an intense orange color following the addition of H2O2 to an acidic solution [36]. Thoroughly mixed supernatants of 0.2 ml were taken after 2, 4, 8, 16, 24, 32, and 48 h (triplicates for each time) during the standard operation procedure for specimen treatment in 5% H2O2 and transferred into 0.05 ml 0.1 M H2SO4. The molar absorption coefficient of the resulting titanium-peroxy complex of 724 M1 cm1 [37] was used for its quantitation at 405 nm.

The cells were fixed for 10 min with 1.5% (v/w) PBS-buffered paraformaldehyde (Carl-Roth, Germany), pH 7.2. After washing with PBS, cells were permeabilized for 10 min with 0.5% (v/w) TritonÔ X-100 (SigmaeAldrich, Germany). MG63 actin filaments were stained for 1 h with 5 mg/ml Alexa FluorÒ 488 conjugated phalloidin (Molecular Probes/Life Technology, Germany) in PBS. All procedures were performed in the dark at room temperature, washed again, embedded and examined using confocal laser microscopy (LSM510 Exciter, Carl-Zeiss, Germany). 2.14. Microorganisms, inactivation, and testing of microbicidal action

2.9. Cell culture Human osteoblast-like MG63 cells were cultured in Dulbecco’s modified Eagle medium (DMEM) (Gibco, Life Technologies, Germany) supplemented with 10% fetal bovine serum (FBS) (Gibco, Life Technologies, Germany) in a humidified atmosphere of 5% CO2/95% air at 37  C. For experiments, MG63 cells were harvested at 80e90% confluence by trypsin/EDTA (PAA Laboratories, Germany). After centrifugation the cell pellet was resuspended either in DMEM (for cell adherence experiments) or DMEM/10% FBS for incubations longer than 1 h. Cells were adjusted to 1  105 cells/ ml; 0.5 ml cell suspension was applied to the surface of the specimen, which was jacketed in a sterile silicon tube sleeve, so that exposure with MG63 cells would be limited to the flat, prepared surface of the Ti6Al4V disc. Non-treated specimen served as control. Tests were performed in 24-well plates for 48 h in a humidified atmosphere of 5% CO2/95% air at 37  C. After 24 h the silicon tube sleeve was removed and cell culture was continued for 24 h in 1.0 ml fresh medium. In a further setup, each specimen was placed on a socket of a cellulose disc of 12.7 mm in diameter and 0.75 mm thickness (Carl-Roth, Germany), which was subsequently saturated by sterile water. MG63 cells were applied in a density of 5  104 cells in 50 ml DMEM/10% FBS and evenly spread on the surface of the specimen using an inoculation loop. After 1 h of incubation in a humidified atmosphere of 5% CO2/95% air at 37  C, each specimen with cultured cells was rinsed with culture medium and transferred into a 24-well cell culture plate containing 1.0 ml of fresh DMEM/10% FBS. The medium was changed after 24 h and culturing continued for further 24 h. 2.10. Cell adhesion experiments Ti6Al4V discs (one per well) were centrally arranged into the cavities of a 24well cell culture plate. MG63 cells were applied in a density of 1  105 cells in 50 ml DMEM and evenly spread on the surface of specimen using an inoculation loop. After 2, 5, and 15 min specimens were fully covered with 1.0 ml DMEM. Nonadhering cells were counted twofold for each specimen in a MOXI Z mini automated cell counter (ORFLO Technologies, USA). For control experiments 24-well cell culture plates were used. Experiments were conducted in triplicates. 2.11. Cell proliferation assay Cell proliferation was evaluated by the activity of mitochondrial dehydrogenase enzymes of metabolically active cells with XTT measuring the conversion of 2,3-Bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide salt (XTT) into a colored aqueous soluble formazan product. After incubation, specimens with cultured cells were washed with DMEM/10% FBS and transferred into a 24-well cell culture plate containing 1.0 ml of XTT reagent (a stock of 1 mg/ml XTT containing

The Gram-positive bacteria Staphylococcus aureus (ATCC 6538) and Staphylococcus epidermidis (ATCC 14990) and the Gram-negative bacteria Escherichia coli (ATCC 11229) and Pseudomonas aeruginosa (ATCC 15442) were used as test microorganisms. For all tested bacteria, aqueous PHMB (

Poly (hexamethylene biguanide) adsorption on hydrogen peroxide treated Ti-Al-V alloys and effects on wettability, antimicrobial efficacy, and cytotoxicity.

An effective amount of the antiseptic agent PHMB cannot simply be placed on the surface of titanium alloys where hydrocarbons were removed by differen...
5MB Sizes 0 Downloads 3 Views