Research Article

Received: 31 August 2009 Revised: 2 December 2009 Accepted: 11 January 2010 Published online in Wiley Interscience: 29 March 2010

(www.interscience.wiley.com) DOI 10.1002/sia.3247

Effect of oxygen plasma on surface properties and biocompatibility of PLGA lms

Nesrin Hasirci,a,b,c,d,e Tugba Endogan,a Elif Vardar,b Aysel Kiziltayc and Vasif Hasircia,b,c,d,f
In this study, poly(D, L-lactide-co-glycolide) (PLGA) lms were prepared by solvent casting method and the surfaces of the lms were modied by application of oxygen plasma. A radio frequency (RF) generator working at 13.56 MHz was used to create plasma, and powers at different levels changing between 20 and 300 W were applied. The variations in chemistry, topography and surface free energy (SFE) of the lms were investigated by electron spectroscopy for chemical analysis (ESCA), atomic force microscopy (AFM) and goniometer, respectively. The cellmaterial interactions of the modied samples were evaluated by cell culture tests using 3T3 broblast cell line. As the applied power of the RF generator was increased from 20 to 300 W, the surface oxygen content (examined by ESCA) rst increased up to 100 W, and then decreased mostly because of crosslink formation by elimination of oxygen. Surface roughness (examined by AFM) and hydrophilicity (examined by water contact angle measurements) increased parallel to the applied power. SFE and the basic component of SFE also increased while the acidic component did not show a signicant change with power according to the geometric mean approach. In vitro materialcell interaction studies showed that oxygen plasma modication enhanced the cell attachment and cell proliferation on PLGA samples. Copyright c 2010 John Wiley & Sons, Ltd. Keywords: PLGA; surface free energy; oxygen plasma; surface modication; cell adhesion; in vitro compatibility

Introduction
The materials which are used to replace a part of a living system (e.g. intraocular lenses) or to help the biological system in intimate contact with living tissue (e.g. dialysis membranes) are called as biomaterials.[1,2] The surface of these materials is the most important part and determines their compatibility in in vivo applications. When the material is implanted into the body, some signal molecules become effective after the rst biological interactions, and control the adherence and proliferation of the cells to the material surface. In some cases, it is required that the cells should not adhere on the material, such as in the case of articial blood vessels, while there are other cases where the surfaces are expected to attract cells, as in tissue engineering scaffolds. Polymers are very important materials in biomedical devices because of their versatility, ease of processing and low cost. Poly(glycolic acid) (PGA), poly(lactic acid) (PLA) and their copolymer poly(D,L-lactideco-glycolide) (PLGA) are the most commonly used polymers in the production of bone plates, drug delivery systems, sutures and scaffolds.[3,4] One of the indicative processes used to assess the biocompatibility of a synthetic surface is cell adhesion onto the material and their spreading. Cells have membranes consisting of proteins, phospholipids and polysaccharides all of which possess various functional groups that are involved in adhesion. In an aqueous media, the material and the cell exhibit short-range electrostatic repulsions and relatively longer-range attractive forces which need to be in balance for achieving adhesion.[2] Therefore, surface free energy (SFE), chemistry and topography are the parameters affecting the cell adhesion. In some biomedical applications, the material may have the desired bulk properties but not the surface and needs to be modied. Surface modication can be achieved by different techniques such as ame, laser, ion beam, electron beam, ozone

treatment, or UV irradiation and chemisorption of molecules.[5] Glow discharge plasma technique is an efcient method which does not affect the bulk but introduces the desired functional groups onto the surface. The particles present in the gas plasma collide with the surface of the samples placed in the plasma reactor and create free radicals opening the possibility for further reactions. Plasma treatment can be used to increase the surface energy of most polymers by making their surfaces more polar and wettable. For biomaterial surfaces this includes improvement in biocompatibility, promotion in adhesion, increase in surface wettability, functionalization of surface, reduction of friction and increase in tackiness.[6,7] Oxygen plasma generally increases the hydrophilicity of the surface due to the formation of oxygen

Correspondence to: Nesrin Hasirci, Department of Chemistry, Middle East Technical University, Ankara 06531, Turkey. E-mail: nhasirci@metu.edu.tr

Paper published as part of the ECASIA 2009 special issue. a Department of Polymer Science and Technology, Middle East Technical University, Ankara 06531, Turkey b Department of Biomedical Engineering, Middle East Technical University, Ankara 06531, Turkey c Department of Biotechnology, Middle East Technical University, Ankara 06531, Turkey d Department of Micro and Nanotechnology, Middle East Technical University, Ankara 06531, Turkey e Department of Chemistry, Middle East Technical University, Ankara 06531, Turkey f Department of Biological Sciences, Middle East Technical University, Ankara 06531, Turkey

486

Surf. Interface Anal. 2010, 42, 486491

Copyright c 2010 John Wiley & Sons, Ltd.

Oxygen plasma effects on PLGA containing groups, such as ester, ether or carboxylate on the surface.[8] Plasma treatment also alters the SFE of the material. SFE (s ) is dened as the work needed to increase the area of a substance by a unit amount and it has polar ( p ) and dispersive ( d ) components. The polar component itself also has acidic ( + ) and basic ( ) parts. For solids, these values can be calculated from the contact angles formed with various liquids and by using some mathematical approximations. Attachment of cells onto a material depends on many parameters. In this study, PLGA lms were prepared, their surfaces were treated with oxygen plasma and the effect of plasma modication on chemistry, topography, SFE, hydrophilicity, as well as attachment and proliferation of 3T3 broblast cells were studied. spectroscopy using a Perkin Elmer Spectrum BX-FTIR spectrometer equipped with a ZnSe crystal at 45 . The samples were analyzed over the 6004000 cm1 range with a resolution of 4 cm1 . All spectra were averaged over 32 scans. X-ray photoelectron spectroscopy The elemental compositions of the surfaces of the pristine and modied samples were analyzed using XPS spectra obtained by a SPECS SAGE spectrometer (Germany) with an Al radiation source (1486.6 eV). The survey scan spectra were obtained between the range 0 and 1000 eV, using a pass energy of 48 eV. Atomic force microscopy Surface topographies of the pristine and modied PLGA lms were examined by AFM. The samples were analyzed with Quesant Universal SPM with a noncontact mode at room temperature using silicon cantilevers (Ambios Technology Inc., USA). Contact angle measurements and SFE calculations Contact angles of different liquids were detected with KSV Cam200 Goniometer (Finland) by sessile drop method at 20 C. SFE, polar and dispersive, as well as acidic and basic components of SFE were calculated using geometric mean, harmonic mean and acidbase approaches. In vitro studies 3T3 broblast cells (passage 16) were cultivated in high glucose DMEM supplemented with 10% FBS, 100 units/ml penicillin and 100 units/ml streptomycin at 37 C in a carbon dioxide incubator (5% CO2 , MCO-17AIC, Sanyo Electric Co. Ltd, Japan). Twenty thousand cells in 30 l were seeded on each lm (1 1 cm2 ) which were placed in 24-well plates and sterilized by UV for 15 min at room temperature (RT) prior to the experiments. After 2 h, 1000 l high glucose DMEM supplemented with 10% FBS, 100 units/ml penicillin and 100 units/ml streptomycin was added. Untreated PLGA lms were used as control. Cell numbers were evaluated by MTS assay and also by Nucleo Counter (ChemoMetec A/S, Denmark) Fluorescence microscopy (Leica DM2500 Confocal Microscope) was used to image the cells on untreated and oxygen plasma treated lms after xing the cells with paraformaldehyde (PFA) and staining with phalloidin.

Materials and Methods

Materials PLGA [(C6 H8 O4 )x (C4 H4 O4 )y ]n , (a 50 : 50 molar ratio copolymer of D,Llactide and glycolide units with MW 40 00075 000) was obtained from Boehringer-Ingelheim, Germany. Chloroform (CHCl3 ) was obtained from Lab-Scan, Ireland, and was used as a solvent for PLGA. Liquids used for contact angle measurements were formamide (HCONH2 ) from Merck, Germany, diodomethane (CH2 I2 ) and dimethyl sulfoxide (C2 H6 OS) from Acros, USA. Double distilled water was used in all the experiments. All the liquids used in contact angle measurements were of reagent grade. Materials used in in vitro tests were fetal bovine serum (FBS, from Biochrome KG, Germany), Dulbeccos Modied Eagle Medium (DMEM, from Gibco Invitrogen Corporation, New Zealand), trypsin-EDTA (0.05%, from Sigma Chemical Corporation, USA) and phalloidin (from BDH Chemicals Ltd, UK). 3T3 cell line was purchased from Foot-andMouth Disease Institute of Ministry of Agriculture and Rural Affairs, Ankara, Turkey. PLGA lm preparation and modication PLGA lms were prepared by casting PLGA solutions (20% w/w in chloroform) on Teon sheets. The samples were dried for 35 days at the room temperature and then in a vacuum oven. Before plasma treatment, all the samples were cut in rectangular shapes of approximately 1 cm 1 cm. For contact angle measurements, the lms were casted on microscope slides. All experiments were carried out 24 h after the plasma treatment. Plasma modication of PLGA lms Plasma modications were performed by Advanced Plasma Systems Inc. plasma system (USA). The lms on microscope slides were placed into the reactor and the system was evacuated to a pressure of 20 mTorr. Then oxygen gas was ushed through the system and the ow rate was adjusted to keep the pressure at 20 mTorr. Discharge at various powers such as 20, 50, 100, 200 and 300 W was applied for 10 min. After turning off the power, the reactor was left under oxygen atmosphere for extra 30 min to enable the reaction of surface active radicals with oxygen. FTIR-ATR spectroscopy Chemical composition of the surfaces was examined with Fourier transform infrared-attenuated total reectance (FTIR-ATR)
Surf. Interface Anal. 2010, 42, 486491

Results and Discussion

Surface characterization Radio frequency (RF) plasma is a commonly used technique in surface modications of materials. Application of oxygen plasma to PLGA lms altered the chemistry, topography and SFE of the samples. These, in return, affected the attachment and proliferation of the cells on the material surfaces. Untreated and plasma-treated PLGA lms were studied with FTIR-ATR for the changes in chemical composition but no difference could be detected (data not shown) most probably due to the excessive penetration of IR rays into the bulk. It is known that FTIR-ATR examines the surfaces up to a depth of 100 nm, but the plasma effect is in molecular level and alters only a few nanometers of the surface.[9,10]

487

Copyright c 2010 John Wiley & Sons, Ltd.

www.interscience.wiley.com/journal/sia

N. Hasirci et al.

Figure 1. XPS spectra of (A) C 1s region of PLGA lms; (B) O 1s region of PLGA lms; (C) C 1s region of control lm; (D) O 1s region of control lm; (E) C 1s region of 100 W plasma-treated lm; (F) O 1s region of 100 W plasma-treated lm.

XPS provides semiquantitative data about surface atomic composition and types of chemical bonds. It has a sampling depth of approximately 0.58 nm. XPS results obtained for the PLGA lms for C 1s and O 1s regions are given in Fig. 1 and the detailed spectra of the same regions are given for control (Fig. 1(C) and (D)) and for 100 W plasma-treated samples (Fig. 1(E) and (F)). The C 1s core level spectra of all lms had a peak maximum at about 285 eV at different intensities assigned to the CChydrocarbon backbone. As reported in the literature, the peak at about 287 and 289 eV resulted from the ether bonds (CO) and the carboxylic groups (COOH) in PLGA, respectively.[11 13] It was observed that the intensity of the peak at about 285 eV decreased as the plasma power increased to 100 W, indicating conversion of CCbonds to oxygenated forms. At 200 and 300 W plasma applications, the increase in the peak intensity at 285 eV can be explained as the decrease in oxygen content because of breaks in the oxygencarbon bonds and the formation of CCcrosslinks. On the other hand, O 1s region of PLGA lms can be deconvoluted into two peaks at around 532 and 533 eV which corresponds to carbonyl (C O) and ester (COC O) groups, respectively. After plasma application some slight shifts occurred but the signal intensities of both peaks were almost equal for the control and for the plasma-treated samples. The surface atomic compositions obtained from electron spectroscopy for chemical analysis (ESCA)

survey scan spectra demonstrated an increase in the surface oxygen content from 35 to 48% as the power increased to 100 W (Table 1), but further increase in power caused a decrease in the amount of oxygen similar to the results given in literature.[14] AFM showed that the surface morphology of the PLGA lms had changed signicantly depending on the power applied (Fig. 2). Untreated PLGA lms have smooth surfaces (Fig. 2(A)) while surface roughness signicantly increased with increasing power. An undulated surface was observed after oxygen plasma treatment. Breaking of some CC bonds or functional groups and formation of crosslinks may cause this kind of roughness on the surface of the material. Similar kind of roughness formation is also reported in literature.[13] Hydrophilicity of the lms also changed upon exposure to oxygen plasma. Increasing the power of plasma mainly decreased the water contact angle (from 67 to 38 with 300 W oxygen plasma treated samples having the lowest) and therefore increased the hydrophilicity (Table 2). This was expected since oxygen plasma generally causes formation of acid, alcohol or ether groups and therefore increases the oxygen content and the polarity of the surfaces. SFE and components of SFE values obtained using different approaches (geometric mean, harmonic mean and acidbase) are given in Table 3. Geometric or harmonic mean results
Surf. Interface Anal. 2010, 42, 486491

488

www.interscience.wiley.com/journal/sia

Copyright c 2010 John Wiley & Sons, Ltd.

Oxygen plasma effects on PLGA

Figure 2. The AFM images of PLGA samples in noncontact mode (5 m 5 m). (A) Untreated control, (B) 20 W, (C) 50 W, (D) 100 W, (E) 200 W, (F) 300 W.

Table 1. Oxygen and carbon contents of the prepared PLGA surfaces Sample Control PLGA-20 W PLGA-50 W PLGA-100 W PLGA-200 W PLGA-300 W Oxygen (%) 35.31 48.13 48.18 48.34 42.89 40.99 Carbon (%) 64.69 51.87 51.82 51.66 57.11 59.01

Table 3. SFE values of the plasma applied surfaces obtained from different approaches Method p (mJ/m2 ) d (mJ/m2 ) + (mJ/m2 ) (mJ/m2 ) (mJ/m2 )

Harmonic Geometric Acidbase Harmonic Geometric Acidbase Harmonic Geometric Acidbase Harmonic Geometric Acidbase Harmonic Geometric Acidbase Harmonic Geometric Acidbase

5.41 6.22 0.53 8.59 6.27 0.63 22.89 20.27 0.54 24.3 22.01 0.69 22.91 20.15 0.23 24.29 21.91 0.52

Table 2. Contact angle values of different liquids on PLGA lms Plasma application Liquids

Power (W) Time (min) Water Formamide Diiodomethane DMSO 0 (Control) 20 50 100 200 300 0 10 10 10 10 10 67 69 40 38 39 38 51 52 13 13 7 10 32 34 27 31 23 29 25 51 22 24 25 25

demonstrated that the total SFE and the polar character of the surface generally increased with plasma power. Total SFE increased from 44.92 to 56.66 mJ/m2 , the polar component increased from 6.22 to 21.91 mJ/m2 and the dispersive component decreased from 38.69 to 34.75 mJ/m2 according to geometric mean approach. The acidbase approach showed that the highest values of the acidic ( + ) and basic ( ) components of the SFE were obtained for 100 W plasma applied samples, and an increase from 0.05 to 0.08 mJ/m2 and from 2.38 to 4.47 mJ/m2 , respectively, was obtained. Cell culture studies Among the whole lms prepared, 100 W plasma applied samples were chosen for the cell attachment studies since they demonstrated the highest variations in acidic and basic components of SFE, and were compared with the untreated ones.
Surf. Interface Anal. 2010, 42, 486491

Control 30.98 38.69 43.22 0.05 2.38 20 W 10 min plasma 36.35 36.35 43.16 0.05 3.37 50 W 10 min plasma 37.35 36.02 46.9 0.06 4.29 100 W 10 min plasma 36.00 36.00 45.29 0.08 4.47 200 W 10 min plasma 38.3 36.65 48.35 0.03 4.27 300 W 10 min plasma 36.36 34.75 46.05 0.06 4.46

36.98 44.92 43.75 44.94 40.84 43.79 60.24 56.28 47.44 60.31 56.41 45.98 60.31 56.08 48.58 60.64 56.66 46.57

Figure 3 shows uorescence microscopy images of 3T3 cells adhered to untreated and 100 W oxygen plasma treated PLGA lms on day 1 and day 3. During cell culture experiments, it was observed that untreated samples started to degrade in culture media and broke into pieces after 24 h, while the treated ones preserved their lm integrity.[15] Therefore, it can be concluded that oxygen plasma treatment leads to stabilization of the PLGA lms. Since the power of O2 plasma treatment has increased the surface hydrophilicity, the cells uniformly spread over the

489

Copyright c 2010 John Wiley & Sons, Ltd.

www.interscience.wiley.com/journal/sia

N. Hasirci et al.

Figure 3. Fluorescence images of 3T3 broblast cells on PLGA lms on day 1 and day 3.

Figure 4. Cell proliferation determined with an MTS assay.

treated lm surfaces on the rst day, and the borders of cells attaching to these lms can easily be seen (Fig. 3). On the other hand, for the control group, cells were observed as aggregates most probably due to the less spreading of the initially added drop of cell suspension on more hydrophobic surface compared to treated samples. On the third day, dense cell image was observed on 100 W plasma-treated samples due to the proliferation of the cells. On the other hand, clear images could not be obtained for the control lms since these lms were not stable and broke into a few pieces in the culture media. In MTS cell viability assay, a higher cell proliferation was observed for O2 plasma treated samples (Fig. 4). The cell number is a reection of cells adhesion strength on the material. For untreated control samples, the lower cell numbers might be a result of the degradation of material as well as their different chemistry. In fact, there is a signicant contribution of O2 plasma application to the cell adhesion on PLGA lms, since plasma changes the adhesion by changing the chemistry as well as by increasing the surface roughness, which was shown in several studies in literature.[12,13,16] Recent studies concentrated on determining whether a correlation between SFE and the cell attachment might exist. In the study

by Ozcan et al.[17] the correlation between the hydrophilicity, SFE and cell adhesion was examined. The poly(methyl methacrylate) (PMMA) lms were subjected to different oxygen plasma treatments and it was concluded that if there is a relation between SFE and cell attachment, value of SFE should be around 61 mJ/m2 for maximum cell attachment.[17] In the study of Alves et al.[9] the effect of surface modication by oxygen RF glow discharge on protein adsorption and bone cell behavior were discussed. The results showed that the oxygen plasma enhanced surface wettability and hydrophilicity and the proliferation of MG63 osteoblast-like cells on P(D,L-LA) surfaces.[9] The improvement of the binding efciency of basic broblast growth factor to CO2 plasma treated PLGA was also reported in literature.[18] Plasma treatment was suggested as an efcient method for altering the surface topography and the atomic environment on the surface without changing the bulk and thus affecting protein adsorption and cell attachment onto the materials surfaces.[19,20] The SFE results obtained in this study are 60.31 mJ/m2 (harmonic) and 56.41 mJ/m2 (geometric) and are in good agreement with the ones reported in the literature.

Conclusions
Surface properties of biomaterials are crucial since they determine the biocompatibility of the material. In order to improve biomaterialcell interaction, surface modication by plasma application is an effective technique. Oxygen plasma treatment changed the surface chemistry, topography, hydrophilicity and SFE of the prepared PLGA lms, and mainly the hydrophilicity and SFE of the lms increased and the water contact angles decreased. It can be concluded that enrichment in oxygenated functionalities and the roughness of the surface, and the total SFE and polar component of SFE could be effective parameters for the cell attachment, and there is also a certain critical value for each which gives the maximum adherence of the cells. Therefore, in order to adjust the desired biocompatibility, plasma surface modication is a preferable technique to alter the surface properties of the materials.

490

www.interscience.wiley.com/journal/sia

Copyright c 2010 John Wiley & Sons, Ltd.

Surf. Interface Anal. 2010, 42, 486491

Oxygen plasma effects on PLGA Acknowledgment The authors are grateful to Tubitak Nano-Biomat project (TBAG 105T508) and METU-BAP-07-02-2009-00-01 project for funding and supporting the research and H. Ozcelik (METU BIOMAT Group) for AFM analysis.
[8] K. E. Park, K. Y. Lee, S. J. Lee, W. H. Park, Macromol. Symp. 2007, 249, 103. [9] C. M. Alves, Y. Yang, D. Marton, D. L. Carnes, J. L. Ong, V. L. Sylvia, D. D. Dean, R. L. Reis, C. M. Agrawal, J. Biomed. Mater. Res. Part B Appl. Biomater. 2008, 87, 59. [10] A. E. Aksoy, V. Hasirci, N. Hasirci, Macromol. Symp. 2008, 269, 145. [11] S. J. Lee, G. Khang, J. H. Lee, Y. M. Lee, H. B. Lee, Polymer (Korea) 2000, 24, 877. [12] G. Khang, J. Choee, J. M. Rhee, H. B. Lee, J. Appl. Polym. Sci. 2002, 85, 1253. [13] Y. Wan, X. Qu, J. Lu, C. Zhu, L. Wan, J. Yang, J. Bei, S. Wang, Biomaterials 2004, 25, 4777. [14] C. Ozcan, N. Hasirci, J. Biomater. Sci. Polym. Ed. 2007, 18(6), 759. [15] E. Vey, C. Roger, L. Meehan, J. Booth, M. Claybourn, A. F. Miller, A. Saiani, Polym. Degrad. Stab. 2008, 93, 1869. [16] H. Shen, X. Hu, F. Yang, J. Bei, S. Wang, Biomaterials 2007, 28, 4219. [17] C. Ozcan, P. Zorlutuna, V. Hasirci, N. Hasirci, Macromol. Symp. 2008, 269, 128. [18] Y. Ozdemir, K. Serbetci, N. Hasirci, J. Mater. Sci. Mater. Med. 2002, 13, 1147. [19] N. Kayirhan, A. Denizli, N. Hasirci, J. Appl. Polym. Sci. 2001, 81, 1322. [20] H. Shen, X. Hu, J. Shenghuo, Biomaterials 2008, 29(15), 2388.

References
[1] B. J. Park, S. R. Lakes, Biomaterials An Introduction, Plenum Press: New York, 1992. [2] R. Langer, J. P. Vacanti, Tissue Eng. Sci. 1993, 260, 920. [3] V. Hasirci, N. Hasirci, in Control of Polymeric Biomaterial Surfaces, Chapter 2 in Surfaces and Interfaces for Biomaterials, P. Vadgama (ed), CRC Press: England, 2005. [4] M. Ngiam, S. Liao, A. J. Patil, Z. Cheng, C. K. Chan, S. Ramakrishna, J. Bone Miner. Res. 2009, 45, 4. [5] V. Hasirci, A. Tezcaner, N. Hasirci, S. Suzer, J. Appl. Polym. Sci. 2003, 87, 1285. [6] C. Oehr, Nucl. Instrum. Meth. B 2003, 208, 40. [7] H. G. Ryu, W. S. Yang, H. W. Roh, I. S. Lee, J. K. Kim, G. H. Lee, D. H. Lee, B. J. Park, M. S. Lee, J. C. Park, Surf. Coat Tech. 2005, 193, 60.

491
www.interscience.wiley.com/journal/sia

Surf. Interface Anal. 2010, 42, 486491

Copyright c 2010 John Wiley & Sons, Ltd.