Application of atmospheric pressure plasma on polyethylene for increased prosthesis adhesion

    Application of Atmospheric Pressure Plasma on PE for Increased Prosthesis Adhesion S. Van Vrekhem, P. Cools, H. Declercq, A. Van Tongel, C. Vercruysse, M. Cornelissen, N. De Geyter, R. Morent PII: DOI: Reference:

S0040-6090(15)00831-7 doi: 10.1016/j.tsf.2015.08.055 TSF 34606

To appear in:

Thin Solid Films

Received date: Revised date: Accepted date:

1 April 2015 5 August 2015 6 August 2015

Please cite this article as: S. Van Vrekhem, P. Cools, H. Declercq, A. Van Tongel, C. Vercruysse, M. Cornelissen, N. De Geyter, R. Morent, Application of Atmospheric Pressure Plasma on PE for Increased Prosthesis Adhesion, Thin Solid Films (2015), doi: 10.1016/j.tsf.2015.08.055

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

D

MA

Application of Atmospheric Pressure Plasma on PE for Increased Prosthesis Adhesion

AC CE P

TE

By Stijn Van Vrekhem, Pieter Cools, Heidi Declercq, Alexander Van Tongel, Chris Vercruysse, Maria Cornelissen, Nathalie De Geyter and Rino Morent

Corresponding author: Stijn Van Vrekhem Research Unit Plasma Technology (RUPT), Department of Applied Physics, Faculty of Engineering, Ghent University, St-Pietersnieuwstraat 41 B4, Ghent, 9000, Belgium E-mail: [email protected] Tel: +32 9 264 38 38

ACCEPTED MANUSCRIPT Application of Atmospheric Pressure Plasma on PE for Increased Prosthesis Adhesion S. Van Vrekhem a , P. Cools a , H. Declercq a,b , A. Van Tongel c , C. Vercruysse b , M. Cornelissen b , N. De Geyter a , R. Morent a a

RI

Tissue Engineering Group, Department of Basic Medical Sciences, Faculty of Medicine and Health Sciences, Ghent University, De Pintelaan 185 6B3, 9000 Ghent, Belgium Department of Orthopaedic Surgery and Traumatology, Ghent University Hospital, De Pintelaan 185 13K12, 9000 Ghent, Belgium

MA

NU

c

SC

b

PT

Research Unit Plasma Technology (RUPT), Department of Applied Physics, Faculty of Engineering and Architecture, Ghent University, Sint-Pietersnieuwstraat 41 B4, 9000 Ghent, Belgium

Abstract

Biopolymers are often subjected to surface modification in order to improve their surface

TE

D

characteristics. The goal of this study is to show the use of plasma technology to enhance the adhesion of ultra-high molecular weight polyethylene (UHMWPE) shoulder prostheses. Two

AC CE P

different plasma techniques (low pressure plasma activation and atmospheric pressure plasma polymerization) are performed on UHMWPE to increase the adhesion between (1) the polymer and a polymethylmethacrylate (PMMA) bone cement and (2) the polymer and osteoblast cells. Both techniques are performed using a dielectric barrier discharge (DBD). A previous paper showed that low pressure plasma activation of UHMWPE results in the incorporation of oxygen-containing functional groups, which leads to an increased surface wettability. Atmospheric pressure plasma polymerization of methylmethacrylate (MMA) on UHMWPE results in a PMMA-like coating, which could be deposited with a high degree of control of chemical composition and layer thickness. The thin film also proved to be relatively stable upon incubation in a phosphate buffer solution (PBS). This paper discusses the next stage of the study, which includes testing the adhesion of the plasma-activated and plasma-polymerized samples to bone cement through pull-out tests and 2

ACCEPTED MANUSCRIPT testing the cell adhesion and proliferation on the samples. In order to perform the pull-out tests, all samples were cut to standard dimensions and fixed in bone cement in a reproducible

PT

way with a sample holder specially designed for this purpose. The cell adhesion and proliferation were tested by means of an MTS assay and live/dead staining after culturing

RI

MC3T3 osteoblast cells on UHMWPE samples. The results show that both plasma activation

SC

and plasma polymerization significantly improve the adhesion to bone cement and enhance cell adhesion and proliferation. In conclusion, it can be stated that the use of plasma

MA

NU

technology can lead to an implant with improved quality and a subsequent longer lifespan.

Keywords: Plasma activation, plasma polymerization, prosthesis adhesion, dielectric barrier

AC CE P

TE

D

discharge.

3

ACCEPTED MANUSCRIPT 1. Introduction Total shoulder arthroplasty (TSA) is commonly used as treatment for diseases of the glenohumeral joint, such as symptomatic osteoarthritis, rheumatoid arthritis or avascular

PT

necrosis [1, 2]. The shoulder joint is a ball-and-socket joint, which consists of two main parts:

RI

the humerus (upper arm) and the glenoid cavity (located in the shoulder blade). An

SC

anatomical TSA mimics this anatomy and therefore also consists of a humeral and glenoid component. The former is a modular system containing a stem and a rounded head or ball,

NU

usually made of titanium or a cobalt-chromium alloy. The latter component is a concave socket made of ultra-high molecular weight polyethylene (UHMWPE). The major problem

MA

concerning total shoulder replacements today is the loosening of the glenoid component [3]. The standard for fixation of a full-UHMWPE component is using bone cement, usually a

D

polymethyl methacrylate (PMMA) resin. Other techniques such as the use of metal-backed

TE

components and full-UHMWPE components without bone cement can be applied based on

AC CE P

patient’s requirements [4]. However, in spite of these additional techniques and the fact that the glenoid component can come in different geometrical designs, e.g. pegged or keeled, the incidence of glenoid component loosening is still too high and accounts for about 80% of all complications [2-4]. The region of failure is mostly the interface between the glenoid component and the PMMA bone cement [3]. This is mostly due to the nature of the used material. Although UHMWPE has excellent mechanical properties, it is an inert and nonpolar material. As a consequence, it is difficult to adhesively bind UHMWPE [5]. By improving the adhesive properties of medical grade UHMWPE using surface modification, it can be possible to improve the lifespan and quality of a shoulder implant. Literature shows extensive research, using plasma technology, on the improvement of the wettability and adhesion of polymer surfaces [6-12]. Plasma activation has been widely used in the past to introduce functional groups such as hydroxyls and alkanoates onto polymer

4

ACCEPTED MANUSCRIPT surfaces, making them less inert and more polar. Research has shown that an increase in wettability can also lead to an increase in cell adhesion and proliferation [10, 13, 14]. Another

PT

plasma-based surface modification technique is plasma polymerization which is a method to produce thin polymer films on a substrate, allowing to change the surface properties by

RI

depositing a coating. The produced plasma polymerized films are in general pinhole-free,

SC

highly cross-linked, insoluble in nearly all solvents and also strongly adhere to numerous

NU

substrates [15-21].

The goal of this study is to show the potential of plasma technology as a surface modification

MA

technique to enhance the adhesion of UHMWPE shoulder prostheses. Low pressure plasma activation as well as atmospheric pressure plasma polymerization of methyl methacrylate

D

(MMA) will be performed on medical grade UHMWPE to (1) increase the adhesion between

TE

the polymer and a PMMA bone cement and (2) to enhance the cellular interactions with UHMWPE. Plasma polymerization of MMA will result in the deposition of a plasma-PMMA

AC CE P

(PPMMA) thin film on a pre-activated UHMWPE surface. This intermediate layer will show a chemical composition similar to the bone cement used in TSAs and may therefore improve the adhesion between the implant and the bone cement. Both plasma activation and plasma polymerization experiments will be performed using a dielectric barrier discharge (DBD) at medium pressure and atmospheric pressure respectively. DBDs are known for their easy formation of a stable discharge and their scalability. The efficiency of medium pressure DBDs has already been demonstrated in the field of plasma activation as well as plasma polymerization [9, 10, 16, 19, 21, 22]. In a previously published paper, results and insights in UHMWPE plasma surface modifications have already been described by using a varied set of analysis techniques [23]. Low pressure plasma activation of UHMWPE resulted in the incorporation of oxygen-containing hydrophilic functional groups, which led to an increased surface wettability. Atmospheric pressure plasma polymerization of MMA on UHMWPE 5

ACCEPTED MANUSCRIPT resulted in a PMMA-like thin coating, which could be deposited with a high degree of control of chemical composition and layer thickness. The deposited thin films also proved to be

PT

relatively stable upon incubation in a phosphate buffer saline (PBS) solution. This paper however will discuss the next stage of the study, which is testing the effect of

RI

plasma technology on characteristics that are important towards the application of a shoulder

SC

prosthesis. This includes testing the adhesion of plasma-activated and plasma-coated samples

NU

to bone cement through pull-out tests and testing the cell adhesion and proliferation on

2. Materials and methods 2.1. Chemicals and materials

MA

plasma-modified samples.

D

For the plasma polymerization experiments, MMA (99%, 30 ppm MEHQ as inhibitor) was

TE

purchased from Sigma-Aldrich (Belgium) and used as such. Dry air, argon (Ar), helium (He) and nitrogen (N2) (Alphagaz 1) were purchased from Air Liquide (Belgium). Specifically for

AC CE P

the cell tests, an UHMWPE film purchased from Goodfellow Cambridge Ltd. (England) with a thickness of 0.5 mm is used as substrate. All other experiments are performed on medical grade UHMWPE pieces with a thickness of 4 mm, which were mulled from a Chirulen 1020 plate purchased from Quadrant EPP Belgium. The bone cement was obtained from Huge Dental Material Co. (China) and consists of a PMMA resin and a MMA liquid component. Upon adding the liquid component to the PMMA resin, a bone cement paste with a curing time of 20 to 30 minutes is formed. 2.2. Plasma methods Both plasma activation and plasma polymerization experiments are performed in a DBD reactor. Figure 1 shows the schematic of the DBD reactor used for activation. The discharge is generated between 2 circular copper electrodes, which have a diameter of 65 mm and are covered by a glass plate. The gap between the electrodes is 7 mm. The lower electrode is 6

ACCEPTED MANUSCRIPT connected to earth through a 50 Ω resistor or a 10 nF capacitor, while the upper electrode is connected to a 5 kHz AC high voltage power source. The UHMWPE samples are fixed on the

PT

lower glass plate using tape followed by pumping the reactor to 0.1 kPa using a rotary vane pump. Subsequently, the reactor is filled with the gas of choice at a rate of 3 standard liters

RI

per minute (slm) until a pressure of 90 kPa is reached. At this sub-atmospheric pressure, the

SC

plasma reactor is flushed at 3 slm with the working gas for 3 min to obtain a controllable gas composition. Finally, the pressure is lowered to 5.0 kPa, the gas flow is adjusted to 1 slm and

NU

plasma activation is performed by turning on the plasma source for a specific time interval.

MA

Plasma polymerization experiments are done in a second DBD reactor, which is schematically presented in Figure 2. The discharge is generated between 2 circular copper electrodes with a

D

diameter of 55 mm, which are both covered by a glass plate. The gas gap between the

TE

electrodes is set to 3.5 mm for the 0.5 mm thick samples used during the cell tests and is set to 8.5 mm for the 4 mm thick samples used for the pull-out tests. The lower electrode is

AC CE P

connected to earth through a 50 Ω resistor or a 10 nF capacitor, while the upper electrode is connected to a 50 kHz AC high voltage power source. The UHMWPE samples are fixed on one of the glass plates using tape, which is followed by pumping the reactor to 0.1 kPa using a rotary vane pump. Subsequently, the reactor is filled with helium (He) at 3 slm to atmospheric pressure (100 kPa). At this pressure, the plasma reactor is flushed at 3 slm for 3 min to obtain a controllable gas composition. After the flushing step, the monomer is introduced into the reactor by sending a small secondary He flow through a gas bubbler containing the MMA monomer. Both the main and secondary He flow are controlled by a mass flow controller (Bronkhorst, The Netherlands). The secondary flow is maintained at 100 ml/min, while the total gas flow is maintained at 3 slm. The discharge power is kept constant at 30 W during the experiments, which are conducted for a fixed treatment time of 4 minutes. These values are optimal for MMA plasma polymerization as can be found in previous work [23]. For both 7

ACCEPTED MANUSCRIPT plasma activation and plasma polymerization, the discharge power is calculated based on Lissajous figures. These Lissajous figures are voltage versus charge plots, which can be

PT

obtained by measuring the applied high voltage using a 1000:1 high voltage probe (Tektronix P6015A) connected to the upper electrode as well as the charge stored on the electrodes. This

RI

charge can be determined by measuring the voltage across the 10 nF capacitor using a 60

SC

MHz oscilloscope probe (Picotech MI007). The resulting Lissajous figures are subsequently used to determine the discharge power, since the electrical energy consumed per voltage cycle

NU

is equal to the area enclosed by the Lissajous figure [24]. The discharge power P is then

MA

calculated by multiplying the electrical energy with the frequency of the feeding voltage (5 kHz).

D

2.3. Surface analysis techniques

TE

Static water contact angle measurements are obtained at room temperature to determine the saturation treatment time for each plasma gas. These measurement are performed using a

AC CE P

commercial Krüss Easy Drop system (Krüss Gmbh, Germany). A water droplet with a controlled size of 2.0 µl is placed on the sample surface. The software provided with the instrument allows automated measurements of the static contact angle. The contact angle values shown in this work are obtained using Laplace-Young curve fitting and are the average of 5 values measured over an extended area of the treated sample. X-ray photoelectron spectroscopy (XPS) is also performed on the UHMWPE samples using a PHI 5000 Versaprobe II spectrometer employing a monochromatic Al Kα X-ray source (hν = 1486.6 eV) operating at 23.3 W. All XPS measurements are conducted in a vacuum of at least 10-6 Pa and are recorded at a take-off angle of 45° relative to the sample surface. Survey scans are recorded with a pass energy of 117.4 eV on 3 different points per sample. These survey spectra are processed using Multipak (9.3) software and from the peak area ratios, the elemental composition of the UHMWPE samples can be determined. 8

ACCEPTED MANUSCRIPT 2.4. Pull-out tests For the pull-out tests, rectangular untreated, plasma-activated and plasma-polymerized UHMWPE samples with an area of 30 mm x 9 mm are fixed in a PMMA bone cement

PT

cylinder with a height of 40 mm and a diameter of 20 mm. This is done in a reproducible way

RI

with a sample holder specially designed for this purpose. The UHMWPE samples are

SC

subsequently pulled out of the bone cement making use of a universal testing machine LRX plus (Lloyd Instruments, Bognor Regis, UK) with a moving speed fixed at 2 mm/min. For

NU

each condition, 10 samples are manufactured and tested.

MA

2.5. Cell tests

The cell experiments consist of two parts: testing of (1) cell adhesion and (2) cell proliferation. Cell adhesion tests are performed on plasma-activated samples using mouse

TE

D

calvaria 3T3 (MC3T3) subclone pre-osteoblast cells, human foreskin fibroblasts (HFF), human adipose tissue-derived mesenchymal stem cells (ATMSC) and human bone

AC CE P

osteocarcoma cells (MG63). During these experiments, these cells are cultured onto untreated and plasma-activated samples. After 24 hours, the cell viability is analyzed with the CellTiter 96® Aqueous Non-Radioactive Cell Proliferation Assay (Promega, USA). This MTS (3-(4,5dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) assay measures the cellular conversion of the MTS into a soluble formazan dye by the mitochondrial NADH/NADPH-dependent dehydrogenase. The absorbance of the formazan dye in the solution is measured with a 490 nm Universal Microplate Reader EL 800 (BioTek Instruments, USA) and the viability is calculated as a percentage of a control culture. Each condition is tested in triplicate. Additionally, a fluorescent live/dead staining with Calcein AM (Anaspec, USA) and propidium iodide (Sigma Aldrich, Belgium) is performed to visualize the cell morphology. The cell proliferation tests are performed by culturing MC3T3 cells onto untreated, plasma-treated and plasma-polymerized samples for 7 days. After 7 days,

9

ACCEPTED MANUSCRIPT the cell viability is again analyzed by means of an MTS assay and the cell morphology is visualized using the same live/dead staining as for the cell adhesion tests.

PT

3. Results and discussion 3.1. Plasma surface modification

RI

Research has shown that the effect of plasma activation on a polymeric substrate, i.e. the

SC

incorporation of functional groups, increases with increasing energy density until the substrate surface is saturated [20, 23, 25]. Once the surface is saturated, an increase in energy density

NU

will not result anymore in an increase in functional group density. As this saturation region

MA

strongly depends on the substrate, it is crucial to first investigate what energy density is required to have a fully saturated UHMWPE surface. To examine this, the treatment time and the corresponding energy density required to reach this saturation need to be first determined

TE

D

for each plasma gas by measuring the contact angle of activated UHMWPE samples at increasing treatment times. The saturation treatment times and energy densities are reached

AC CE P

when the contact angle does not decrease anymore with increasing energy density. The obtained saturation values are presented in Table I for each discharge gas. The UHMWPE samples activated with the resulting energy densities have also been analyzed using XPS and the resulting chemical compositions are also shown in Table I. The saturation energy densities for N2 and air plasma are almost similar and the highest, followed by the energy density for Ar plasma and He plasma. The fact that the saturation energy density for the He plasma treatment is much lower can be explained by the fact that the Penning effect is more efficient in a He atmosphere, creating more reactive species than in an Ar, air or N2 atmosphere [26]. The saturation contact angle ranges between 30° for the N2 plasma and 50° for the He plasma. In comparison, the contact angle of the untreated UHMWPE plate is 85°. The XPS results show that plasma activation results in a higher oxygen concentration on the sample surface compared to the untreated UHMWPE plate, which also contains 5% of oxygen. The amount

10

ACCEPTED MANUSCRIPT of incorporated oxygen ranges between 16% for the He plasma and 24% for the Ar plasma. These results are consistent with other research studies, which have shown that Ar plasma is

PT

more efficient in creating radicals on a polymer surface than He plasma [12]. In the case of N2 and air discharges, also a small amount of nitrogen is incorporated into the surface. The fact

RI

that He, Ar and N2 discharges result in the integration of oxygen is due to a combination of

SC

oxygen impurities present in the discharge gas and post-oxidation of the radicals created by the plasma on the sample surface. The impurities can come from residual air present in the

NU

reactor or from gaseous products that are desorbed from the reactor wall by the plasma, while

MA

the post-oxidation occurs after the samples are taken out of the reactor enabling the created radicals to react with the surrounding air. To examine which chemical groups are present on the surface of the plasma-treated samples, curve fitting of the detailed C1s peaks is also

TE

D

performed. Based on literature, the obtained C1s envelopes of the plasma-treated samples can be decomposed into 4 distinct components: a peak at 285 eV attributed to C-C/C-H bonds, a

AC CE P

peak at 286.7 eV due to C-O bonds (alcohols and ethers), a peak at 287.7 eV corresponding to C=O bonds (aldehydes and ketones) and a peak at 288.9 eV attributed to O-C=O bonds (carboxylic acids and esters) [12]. Within this context, it is also important to mention that the identification of the functional nitrogen-containing groups on the air and nitrogen plasmatreated samples is very difficult, since there are large discrepancies in binding energies published in literature for carbon-nitrogen species and since the energy difference between some carbon-oxygen and carbon-nitrogen species is too small to allow separation of the peaks. Taking into account these issues and the very low incorporation of nitrogen species compared to oxygen species, it does not seem to be opportune to take into account these carbon-nitrogen species during the curve fitting process.

11

ACCEPTED MANUSCRIPT

Based on the fitted C1s envelopes, the concentration of each chemical functionality can be

PT

determined and the results can be found in Table II. As can be seen from this table, oxygen is integrated in the UHMWPE surface as several functional groups, however, it is clear that

RI

most of the incorporated oxygen forms a single bond with carbon. However, there are also

SC

some small amounts of ketones and/or aldehydes (1-2%) as well as carboxylic acids and/or

NU

esters (3-5%) present on the surface of the plasma-treated samples. In comparison, plasma polymerization of MMA results in a PPMMA coating with a carbon

MA

and oxygen concentration of 81% and 19% respectively. The curve-fitted C1s envelop of the deposited coating can be fitted with 4 peaks: a peak at 285.0 eV attributed to C-C/C-H bonds,

D

a peak at 286 eV assigned to the C-C=O bonds, a peak at 286.9 eV due to C-O bonds and a

TE

peak at 288.9 eV attributed to O-C=O bonds [27]. In contrast to the plasma-activated samples,

AC CE P

the peaks at 286.9 and 288.9 eV can now be solely attributed to ketones and esters respectively since previously determined FTIR spectra of the deposited coatings have shown that the deposits do not contain OH groups [23]. Based on the curve-fitted C1s envelop, the concentration of each chemical functionality can be determined and it is found that approximately 7% of the carbon is present as ester groups. This value is low compared to the value obtained for commercial PMMA (14%) due to the partial decomposition of the monomer during the plasma polymerization process. 3.2. Pull-out tests As mentioned before, the major goal of this study is to improve the adhesion between medical grade UHMWPE and bone cement by using plasma activation and plasma polymerization. The adhesive properties of UHMWPE are therefore examined in detail by means of pull-out tests. Plasma-modified UHMWPE samples are pulled out of the bone cement cylinder and the required pull-out force is normalized over the contact area between the sample and the bone 12

ACCEPTED MANUSCRIPT cement. The resulting pull-out stress serves as a measure for the adhesion between UHMWPE and PMMA bone cement and is represented for each plasma modification procedure in Figure

PT

3. Untreated UHMWPE has a mean pull-out stress of 0.47 MPa and a standard deviation of 0.28 MPa, meaning that there is a very high variation in pull-out stresses for untreated

RI

UHMWPE. This might be caused by the milling process. The mean pull-out stress however

SC

clearly increases when the samples are plasma-activated or coated with PPMMA and are situated between 0.92 MPa for He plasma-activated samples and 1.43 MPa for Ar plasma-

NU

activated samples. Applying a Student’s t-test with a significance level of 0.05 to the

MA

measured pull-out stresses shows that all used plasma treatments result in a significantly improved adhesion between the UHMWPE and the PMMA bone cement compared to untreated UHMWPE. Together with the XPS-results, this shows that incorporating oxygen

TE

D

containing functional groups on the polymer surface enhances the adhesion between the two materials. This can be attributed to an increase in van der Waals forces. Applying a Student’s

AC CE P

t-test with a significance level of 0.05 to compare the different plasma treatments with each other shows that the air and argon plasma-activated samples have a statistically better adhesion than nitrogen and helium plasma-activated samples. This can be attributed to the higher amount of oxygen present on the air and argon plasma-activated samples (see XPSresults in Table I). The higher the amount of oxygen, the stronger the intermolecular forces between the two interfaces will be. Next to this, there is no statistical difference in pull-out stress between the plasma-polymerized samples and the air and argon plasma-activated samples. There is less oxygen present on the surface of the plasma-polymerized samples, which should result in a lower pull-out stress for these samples. However, the amount of ester groups on the PPMMA coatings is higher than the amount present on the Ar and air plasmaactivated samples. As the bone cement is based on PMMA, which contains ester functionalities, the more ester functionalities are present on the plasma-treated samples, the

13

ACCEPTED MANUSCRIPT stronger the interaction between both materials will be. The lower amount of oxygen but higher amount of ester functionalities on the plasma-polymerized samples therefore results in

PT

an adhesion that is statistically equal to the adhesion of air and Ar plasma-treated samples. Finally, the high variation in pull-out stress of the untreated UHMWPE is not ideal when

RI

using it in shoulder prostheses and should be avoided. The use of plasma treatment however

SC

decreases this variation to some extent which can be beneficial when using the material as

NU

shoulder prosthesis.

At this point, there seems to be no significant difference between an Ar plasma activation or

MA

plasma polymerization of MMA on the surface. This would give a clear benefit to the use of Ar plasma activation, as this method uses less chemicals and has a simpler set-up than the

D

plasma polymerization process. However, research has shown that the effect of plasma

TE

activation diminishes over time [28-30], while plasma polymer coatings are not prone to ageing effects [31, 32]. To test the ageing effect on the pull-out stress, 3 samples were Ar

AC CE P

plasma-activated and 2 samples were plasma-polymerized. All 5 samples were stored under ambient conditions for 48h and subsequently fixed in bone cement. The mean pull-out stress was 1.01 MPa for the activated samples and 1.53 MPa for the polymerized samples. These results clearly show that after 48h the effect of plasma activation is already significantly diminished, while the pull-out stress of the plasma-polymerized samples has not changed. As the components of a shoulder prosthesis are often stored for a relatively long time, the plasma polymerization treatment is thus preferred. 3.3. Cell tests Cell adhesion and proliferation tests are often used to assess the viability of cells on a surface. In the case of this study, cell tests give important information about the effect of plasma treatment on the adhesion and proliferation of osteoblast and fibroblast cells on UHMWPE. The first step of the cell tests consists in testing the adhesion of different cell lines, in order to 14

ACCEPTED MANUSCRIPT determine a good cell line to proceed to cell proliferation tests. Therefore, four different cell lines (MC3T3, MG63, HFF and ATMSC) are cultured on plasma-activated samples for 24

PT

hours. After these 24 hours, the samples are subjected to an MTS-assay and a live/dead staining. The results from the MTS-assay are represented in Figure 4. From the figure, it is

RI

clear that in general the MC3T3 cell line gives the best adhesion results with the percentage of

SC

viable attached cells ranging from 68% to 78%. In addition, there is little to no difference between the other 3 cell lines. Finally, the figure also shows that the plasma treatments seem

NU

to have no positive effect on cell adhesion after a culture time of 24h.

MA

As MC3T3 cell lines are known to adapt easily to the characteristics of an underlying substrate and show a good adhesion to the UHMWPE, these cells are used to perform

D

proliferation tests for both plasma-activated and plasma-polymerized UHMWPE samples.

TE

Cell proliferation tests asses the ability of, in this case, osteoblast cells to adhere, grow and

AC CE P

multiply on a substrate. Research shows that multiple factors influence the proliferation of cells on a substrate [33-36]. These factors include surface chemistry, topography and wettability. It is clear that each cell type has its own optimal surface with a specific surface roughness, wettability, etc. to adhere to. For the MC3T3 cell line specific, little research on cell proliferation has been performed in the past. However, Nakagawa et al. [37] have found a positive correlation between the oxygen/carbon-ratio of the surface chemistry and the proliferation of MC3T3 cells. For this study, tests are performed by culturing MC3T3 cells on plasma-treated UHMWPE samples. After 7 days, an MTS assay and a life/dead staining are performed. The results of the MTS assay are represented in Figure 5. Ar, He and air plasma activation clearly result in an increased cell proliferation, while N2 plasma activation and plasma polymerization of MMA seem to have no statistically significant effect on cell proliferation. There seems to be no statistically significant difference between the Ar, He and air plasma-activated samples, although the oxygen concentration is lower for He plasma15

ACCEPTED MANUSCRIPT activated samples. A possible explanation for this phenomenon might be the presence of nitrogen-containing species. Research in literature shows that the presence of oxidized

PT

nitrogen (NO2, etc.) can have a negative effect on cell proliferation [38]. However, the fluorescence images in Figures 6-11 show that the number of cells is much higher for all

RI

plasma-treated samples than for untreated samples, clearly showing the beneficial effect of

SC

plasma modification on cell proliferation. The fact that this positive plasma treatment effect is not so obvious in Figure 5 can be attributed to a low sensitivity of the MTS assay for this

NU

specific cell type. Nonetheless, it is clear that plasma treatment has a positive effect on

MA

MC3T3 proliferation. For plasma activation, this can be attributed to the incorporation of oxygen in the surface and the resulting increase in surface wettability. For plasma polymerization, the improvement in cell proliferation can be attributed to the major change in

TE

D

surface chemistry due to the formation of a PPMMA-coating containing a significant amount of oxygen.

AC CE P

Next to the number of cells, also cell morphology is an important feature as it shows if cells are truly attached to a surface. When cells have a round shape, they are not attached to the surface. On the other hand, when cells are attached to the surface they will have a more elongated or triangular shape. Figures 6-11 show fluorescence images after live/dead staining of the different plasma-treated samples after a culture time of 7 days. Not only is it clear that plasma-treated samples have a much higher number of cells on their surfaces, the morphology of the cells has also significantly improved after plasma treatment. The fluorescence images of the plasma-treated samples clearly show that cells have a more triangular shape, proving that these cells are attached to the sample surface. It can be stated that the used plasma treatments not only improve proliferation of MC3T3 cells on the substrate, but also cell morphology and consequent cell adhesion are significantly enhanced. 4. Conclusion 16

ACCEPTED MANUSCRIPT The goal of this study was to use both plasma activation and atmospheric pressure plasma polymerization to enhance the surface characteristics of UHMWPE in order to improve its

PT

performance in shoulder prostheses. This paper therefore focused on two important biomedical aspects: (1) UHMWPE adhesion to bone cement and (2) osteoblast cell adhesion

RI

and proliferation. The results of this paper show that plasma-activating an UHMWPE

SC

substrate or plasma-polymerizing MMA onto an UHMWPE substrate results in the incorporation of oxygen containing functionalities and a subsequent increase in wettability. In

NU

turn, this results in a significant enhancement of the adhesion between UHMWPE and bone

MA

cement. Storage of plasma-treated samples for 48h in ambient conditions before fixation in bone cement showed that the effect of plasma activation diminishes over time. This is an important feature towards the application of a shoulder prosthesis as these components are

TE

D

often stored for a long time before being implanted. On the other hand, plasma-polymerized thin films are not prone to ageing effects. Next to this, adhesion and proliferation of osteoblast

AC CE P

cells on the surface is also increased after plasma activation and plasma polymerization. As both techniques are non-invasive, the excellent mechanical bulk properties can be preserved. Due to the increased adhesion to bone cement and the increased cell proliferation and adhesion, it can thus be concluded that plasma activation and atmospheric pressure plasma polymerization are potentially excellent tools to enhance the quality and lifespan of the UHMWPE component in a shoulder prosthesis. Acknowledgements This research has received the funding from the European Research Council under the European Union’s Seventh Framework Program (FP/2007-2013) / ERC Grant Agreement n. 279022.

17

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

18

ACCEPTED MANUSCRIPT References

[1] L.A. Kaback, A. Green, T.A. Blaine, Glenohumeral arthritis and total shoulder replacement, Medicine and health, Rhode Island, 95 (2012) 120-124.

RI

PT

[2] E.T. Ricchetti, J.A. Abboud, A.F. Kuntz, M.L. Ramsey, D.L. Glaser, G.R. Williams, Jr., Total shoulder arthroplasty in older patients: increased perioperative morbidity?, Clinical orthopaedics and related research, 469 (2011) 1042-1049.

SC

[3] K.I. Bohsali, M.A. Wirth, C.A. Rockwood, Jr., Complications of total shoulder arthroplasty, The Journal of bone and joint surgery. American volume, 88 (2006) 2279-2292.

MA

NU

[4] E.J. Strauss, C. Roche, P.H. Flurin, T. Wright, J.D. Zuckerman, The glenoid in shoulder arthroplasty, Journal of shoulder and elbow surgery / American Shoulder and Elbow Surgeons ... [et al.], 18 (2009) 819-833.

D

[5] R. Oosterom, T.J. Ahmed, J.A. Poulis, H.E. Bersee, Adhesion performance of UHMWPE after different surface modification techniques, Medical engineering & physics, 28 (2006) 323-330.

AC CE P

TE

[6] D. Hegemann, H. Brunner, C. Oehr, Plasma treatment of polymers for surface and adhesion improvement, Nucl Instrum Meth B, 208 (2003) 281-286. [7] J.N. Lai, B. Sunderland, J.M. Xue, S. Yan, W.J. Zhao, M. Folkard, B.D. Michael, Y.G. Wang, Study on hydrophilicity of polymer surfaces improved by plasma treatment, Appl Surf Sci, 252 (2006) 3375-3379. [8] N. De Geyter, R. Morent, C. Leys, Surface characterization of plasma-modified polyethylene by contact angle experiments and ATR-FTIR spectroscopy, Surf Interface Anal, 40 (2008) 608-611. [9] N. De Geyter, R. Morent, T. Desmet, M. Trentesaux, L. Gengembre, P. Dubruel, C. Leys, E. Payen, Plasma modification of polylactic acid in a medium pressure DBD, Surf Coat Tech, 204 (2010) 3272-3279. [10] T. Jacobs, N. De Geyter, R. Morent, T. Desmet, P. Dubruel, C. Leys, Plasma treatment of polycaprolactone at medium pressure, Surf Coat Tech, 205 (2011) S543-S547. [11] R. Morent, N. De Geyter, T. Desmet, P. Dubruel, C. Leys, Plasma Surface Modification of Biodegradable Polymers: A Review, Plasma Process Polym, 8 (2011) 171-190.

19

ACCEPTED MANUSCRIPT [12] N. De Geyter, R. Morent, C. Leys, L. Gengembre, E. Payen, Treatment of polymer films with a dielectric barrier discharge in air, helium and argon at medium pressure, Surf Coat Tech, 201 (2007) 7066-7075.

RI

PT

[13] Y. Arima, H. Iwata, Effect of wettability and surface functional groups on protein adsorption and cell adhesion using well-defined mixed self-assembled monolayers, Biomaterials, 28 (2007) 3074-3082.

SC

[14] C. Lucchesi, B.M. Ferreira, E.A. Duek, A.R. Santos, Jr., P.P. Joazeiro, Increased response of Vero cells to PHBV matrices treated by plasma, Journal of materials science. Materials in medicine, 19 (2008) 635-643.

MA

NU

[15] N. De Geyter, R. Morent, S. Van Vlierberghe, P. Dubruel, C. Leys, L. Gengembre, E. Schacht, E. Payen, Deposition of polymethyl methacrylate on polypropylene substrates using an atmospheric pressure dielectric barrier discharge, Prog Org Coat, 64 (2009) 230-237.

D

[16] F. Benitez, E. Martinez, J. Esteve, Improvement of hardness in plasma polymerized hexamethyldisiloxane coatings by silica-like surface modification, Thin Solid Films, 377 (2000) 109-114.

AC CE P

TE

[17] Y. Sawada, S. Ogawa, M. Kogoma, Synthesis of Plasma-Polymerized Tetraethoxysilane and Hexamethyldisiloxane Films Prepared by Atmospheric-Pressure Glow-Discharge, J Phys D Appl Phys, 28 (1995) 1661-1669. [18] R. Morent, N. De Geyter, S. Van Vlierberghe, E. Vanderleyden, P. Dubruel, C. Leys, E. Schacht, Deposition of Polyacrylic Acid Films by Means of an Atmospheric Pressure Dielectric Barrier Discharge, Plasma Chem Plasma P, 29 (2009) 103-117. [19] F. Arefi, V. Andre, P. Montazerrahmati, J. Amouroux, Plasma Polymerization and Surface-Treatment of Polymers, Pure Appl Chem, 64 (1992) 715-723. [20] T. Jacobs, N. De Geyter, R. Morent, S. Van Vlierberghe, P. Dubruel, C. Leys, Plasma modification of PET foils with different crystallinity, Surf Coat Tech, 205 (2011) S511-S515. [21] C. Vautrin-Ul, C. Boisse-Laporte, N. Benissad, A. Chausse, P. Leprince, R. Messina, Plasma-polymerized coatings using HMDSO precursor for iron protection, Prog Org Coat, 38 (2000) 9-15. [22] U. Lommatzsch, J. Ihde, Plasma Polymerization of HMDSO with an Atmospheric Pressure Plasma Jet for Corrosion Protection of Aluminum and Low-Adhesion Surfaces, Plasma Process Polym, 6 (2009) 642-648.

20

ACCEPTED MANUSCRIPT [23] P. Cools, S. Van Vrekhem, N. De Geyter, R. Morent, The use of DBD plasma treatment and polymerization for the enhancement of biomedical UHMWPE, Thin Solid Films, 572 (2014) 251-259.

RI

PT

[24] H.E. Wagner, R. Brandenburg, K.V. Kozlov, A. Sonnenfeld, P. Michel, J.F. Behnke, The barrier discharge: basic properties and applications to surface treatment, Vacuum, 71 (2003) 417-436.

SC

[25] N. De Geyter, R. Morent, D.N. Ghista, Non-thermal plasma surface modification of biodegradable polymers, in, 2012.

MA

NU

[26] F. Massines, G. Gouda, A comparison of polypropylene-surface treatment by filamentary, homogeneous and glow discharges in helium at atmospheric pressure, J Phys D Appl Phys, 31 (1998) 3411-3420. [27] T.B. Casserly, K.K. Gleason, Effect of substrate temperature on the plasma polymerization of poly(methyl methacrylate), Chem Vapor Depos, 12 (2006) 59-66.

TE

D

[28] Y. Ren, C.X. Wang, Y.P. Qiu, Aging of surface properties of ultra high modulus polyethylene fibers treated with He/O-2 atmospheric pressure plasma jet, Surf Coat Tech, 202 (2008) 2670-2676.

AC CE P

[29] A. Van Deynse, P. Cools, C. Leys, R. Morent, N. De Geyter, Influence of ambient conditions on the aging behavior of plasma-treated polyethylene surfaces, Surf Coat Tech, 258 (2014) 359-367. [30] T. Jacobs, H. Declercq, N. De Geyter, R. Cornelissen, P. Dubruel, C. Leys, A. Beaurain, E. Payen, R. Morent, Plasma surface modification of polylactic acid to promote interaction with fibroblasts, J Mater Sci-Mater M, 24 (2013) 469-478. [31] P.L. Girard-Lauriault, P.M. Dietrich, T. Gross, T. Wirth, W.E.S. Unger, Chemical Characterization of the Long-Term Ageing of Nitrogen-Rich Plasma Polymer Films under Various Ambient Conditions, Plasma Process Polym, 10 (2013) 388-395. [32] M.I. Totolin, I. Neamtu, Positive findings for plasma polymer (meth)acrylate thin films in heritage protective applications, J Cult Herit, 12 (2011) 392-398. [33] J.C. Meredith, J.L. Sormana, B.G. Keselowsky, A.J. Garcia, A. Tona, A. Karim, E.J. Amis, Combinatorial characterization of cell interactions with polymer surfaces, J Biomed Mater Res A, 66A (2003) 483-490.

21

ACCEPTED MANUSCRIPT [34] J.H. Lee, H.W. Jung, I.K. Kang, H.B. Lee, Cell Behavior on Polymer Surfaces with Different Functional-Groups, Biomaterials, 15 (1994) 705-711.

PT

[35] M. Lampin, R. WarocquierClerout, C. Legris, M. Degrange, M.F. SigotLuizard, Correlation between substratum roughness and wettability, cell adhesion, and cell migration, J Biomed Mater Res, 36 (1997) 99-108.

SC

RI

[36] Y. Tamada, Y. Ikada, Cell-Adhesion to Plasma-Treated Polymer Surfaces, Polymer, 34 (1993) 2208-2212.

NU

[37] M. Nakagawa, F. Teraoka, S. Fujimoto, Y. Hamada, H. Kibayashi, J. Takahashi, Improvement of cell adhesion on poly(L-lactide) by atmospheric plasma treatment, J Biomed Mater Res A, 77A (2006) 112-118.

AC CE P

TE

D

MA

[38] J.H. Lee, H.W. Jung, I.K. Kang, H.B. Lee, Cell behaviour on polymer surfaces with different functional groups, Biomaterials, 15 (1994) 705-711.

22

ACCEPTED MANUSCRIPT Table I: Discharge power (W), saturation treatment time (s), saturation energy density (J/cm2), saturation contact angle (°) and resulting atomic concentration (%) for each discharge gas.

Saturation treatment time (s)

Nitrogen

1.89

90

Saturation energy density (J/cm2) 5.12

Air

2.00

90

5.43

48.7

Helium

0.65

90

1.76

50.2

Argon

1.12

130

4.73

36.8

/

/

/

Saturation contact angle (°)

RI

78.4 ± 1.6 75.8 ± 0.4 83.9 ± 0.3 76.4 ± 0.6 95.2 ± 0.4

NU

SC

30.0

Carbon

85.2

Oxygen

Nitrogen

19.3 ± 1.6 22.7 ± 0.4 16.1 ± 0.3 23.6 ± 0.6 4.8 ± 0.4

2.3 ± 1.6 1.5 ± 0.4 / / /

MA

Untreated

Atomic concentration (%)

PT

Discharge power (W)

TE

C-C / C-H (%) 75.8 75.5 78.5 75.9

AC CE P

Nitrogen Air Helium Argon

D

Table II: C1s deconvolution results for each discharge gas. C-O (%) 17.9 17.8 17.7 17.2

C=O (%) 1.5 1.4 0.2 2.1

O-C=O (%) 4.8 5.3 3.6 4.8

23

ACCEPTED MANUSCRIPT Figure captions Figure 1: Schematic representation of the DBD plasma activation reactor. 1: gas bottle; 2: mass flow controller; 3: DBD plate reactor; 4: manometer; 5: pressure valve; 6: oil pump.

RI

PT

Figure 2: Schematic representation of the DBD plasma polymerization reactor. 1: gas bottle; 2: mass flow controller; 3: gas bubbler containing MMA; 4: DBD plate reactor; 5: manometer; 6: pressure valve; 7: oil pump.

SC

Figure 3: Pull-out stress for untreated, plasma-activated and plasma-polymerized UHMWPE samples.

NU

Figure 4: Percentage of viable attached cells for plasma-activated UHMWPE compared to a control culture. Plasma activation was performed with different gases (helium, air, nitrogen and argon).

MA

Figure 5: Results of an MTS assay of MC3T3 cells after a culture time of 7 days for untreated, plasma-activated and plasma-polymerized UHMWPE substrates.

D

Figure 6: Fluorescence images of MC3T3 cells after a culture time of 7 days on nitrogen plasma-activated UHMWPE substrates. Green indicates living cells, red indicates dead cells.

TE

Figure 7: Fluorescence images of MC3T3 cells after a culture time of 7 days on argon plasma-activated UHMWPE substrates. Green indicates living cells, red indicates dead cells.

AC CE P

Figure 8: Fluorescence images of MC3T3 cells after a culture time of 7 days on helium plasma-activated UHMWPE substrates. Green indicates living cells, red indicates dead cells. Figure 9: Fluorescence images of MC3T3 cells after a culture time of 7 days on air plasma-activated UHMWPE substrates. Green indicates living cells, red indicates dead cells. Figure 10: Fluorescence images of MC3T3 cells after a culture time of 7 days on plasma-polymerized UHMWPE substrates. Green indicates living cells, red indicates dead cells. Figure 11: Fluorescence images of MC3T3 cells after a culture time of 7 days on untreated UHMWPE substrates. Green indicates living cells, red indicates dead cells.

24

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

Figure 1

Figure 2

25

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC CE P

TE

D

Figure 3

Figure 4

26

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC CE P

TE

D

Figure 5

27

SC

RI

PT

ACCEPTED MANUSCRIPT

NU

500 µm

AC CE P

TE

D

MA

Figure 6

500 µm Figure 7

500 µm Figure 8

28

SC

RI

PT

ACCEPTED MANUSCRIPT

NU

500 µm

AC CE P

TE

D

MA

Figure 9

500 µm

Figure 10

500 µm

Figure 11

29

ACCEPTED MANUSCRIPT Reseach highlights

TE

D

MA

NU

SC

RI

PT

Plasma activation increases adhesion to bone cement significantly. Plasma activation increases osteoblast cell adhesion and proliferation. Plasma polymerization of methymethacrylate (MMA) increases adhesion to bone cement. Plasma polymerization of MMA increases osteoblast adhesion and proliferation. Effect of plasma activation diminishes after storage for 48 hours.

AC CE P

• • • • •

30