Atomic layer deposition enhanced grafting of phosphorylcholine on stainless steel for intravascular stents

Colloids and Surfaces B: Biointerfaces 121 (2014) 238–247

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Atomic layer deposition enhanced grafting of phosphorylcholine on stainless steel for intravascular stents Qi Zhong a , Jin Yan a , Xu Qian a , Tao Zhang a,b,∗ , Zhuo Zhang c , Aidong Li a a b c

College of Engineering and Applied Sciences, Nanjing University, Nanjing 210093, China Nanjing Excellence Technology Center for Interventional Medical Devices, Nanjing 210093, China Department of Chemistry, Stony Brook University, Stony Brook, NY 11790, USA

a r t i c l e

i n f o

Article history: Received 22 February 2014 Received in revised form 5 June 2014 Accepted 9 June 2014 Available online 16 June 2014 Keywords: Atomic layer deposition Phosphorylcholine Surface modification Stainless steel Intravascular stents

a b s t r a c t In-stent restenosis (ISR) and re-endothelialization delay are two major issues of intravascular stent in terms of clinical safety and effects. Construction of mimetic cell membrane surface on stents using phosphorylcholine have been regarded as one of the most powerful strategies to resolve these two issues and improve the performance of stents. In this study, atomic layer deposition (ALD) technology, which is widely used in semiconductor industry, was utilized to fabricate ultra-thin layer (10 nm) of alumina (Al2 O3 ) on 316L stainless steel (SS), then the alumina covered surface was modified with 3aminopropyltriethoxysilane (APS) and 2-methacryloyloxyethyl phosphorylcholine (MPC) sequentially in order to produce phosphorylcholine mimetic cell membrane surface. The pristine and modified surfaces were characterized using X-ray photoelectron spectroscopy, atomic force microscope and water contact angle measurement. Furthermore, the abilities of protein adsorption, platelet adhesion and cell proliferation on the surfaces were investigated. It was found that alumina layer can significantly enhance the surface grafting of APS and MPC on SS; and in turn efficiently inhibit protein adsorption and platelet adhesion, and promote the attachment and proliferation of human umbilical vein endothelial cells (HUVEC) on the surfaces. In association with the fact that the deposition of alumina layer is also beneficial to the improvement of adhesion and integrity of drug-carrying polymer coating on drug eluting stents, we expect that ALD technology can largely assist in the modifications on inert metallic surfaces and benefit implantable medical devices, especially intravascular stents. © 2014 Elsevier B.V. All rights reserved.

1. Introduction As a pioneer of implanted medical devices, intravascular stents have been subjected to clinical application such as coronary [1], renal [2], carotid [3], peripheral [4], and even intracranial arteries implantations [5]. Currently, most of the intravascular stents are made from metals such as stainless steel (SS), cobalt base alloy (e.g. Co–Cr alloy), titanium alloy (e.g. Nitinol) and platinum alloy (e.g. Pt–Cr alloy). Upon implantation, the small metal mesh tubes help support the inner wall of artery for months to years restoring or preserving the patency of the vessel. Unfortunately, in-stent restenosis (ISR) always impede the application of intravascular stents [6,7]. Approximately, one third of the patients who accept percutaneous transluminal angioplasty (PTA) with bare metal stents (BMS) have

∗ Corresponding author at: College of Engineering and Applied Sciences, Nanjing University, Nanjing 210093, Jiangsu Province, China. Tel.: +86 25 83680009; fax: +86 25 83594668. E-mail address: [email protected] (T. Zhang). http://dx.doi.org/10.1016/j.colsurfb.2014.06.022 0927-7765/© 2014 Elsevier B.V. All rights reserved.

suffered from ISR [8], which resulted in recurrent angina and minor/major heart attack. Patients who developed severe restenosis even have to be subjected to repeat revascularizations. Although drug eluting stents (DES) have made a great progress in terms of reducing restenosis (ISR reduced to less than 10%) [9–13] and achieved huge commercial success in the past decade. However, a new problem, the late stent thrombosis (LAST) which is resulted from the proinflammatory properties of drug-carrying polymers is in urgent need of solutions [14]. Additionally, for intravascular stents (comparing to coronary artery), ISR is still an issue for the reason that restenosis rates are still high. A 2003 study of selective and systematic stenting for limb-threatening ischemia reported a restenosis rates of 32.3% for selective stenting patients and 34.7% for systematic stenting patient at one-year follow-up [15]. In a 2009 study, which compared bare nitinol stents with PTA in superficial femoral artery disease, restenosis was 34.4% for stented patients at one-year follow-up [16]. In 2013, a study of 206 cases in Iran reported a restenosis rate of 15.9% at 10.8 months (average time of 6–24 months) after stenting of symptomatic vertebral artery stenosis [17].

Q. Zhong et al. / Colloids and Surfaces B: Biointerfaces 121 (2014) 238–247

Previous studies [6,7,18] have demonstrated that the delay of the re-endothelialization as well as the injury-induced migration and subsequent proliferation through smooth muscle cells are the major pathophysiological events that lead to neointima formation. The vessel endothelium plays an important role in the maintenance of the integrity of the vessel by preventing thrombosis and hyperplasia from happening. A rapid re-endothelialization has been proven to be a potential technique to prevent restenosis and late stent thrombotic [19–21]. Surface modification is believed to be one of the most effective ways to improve re-endothelialization and stent-vascular compatibility [22–24]. The modification methods have been seen to evolve from early simple physical polishing to current chemical modification, and even to the immobilization of biological molecules [23], such as coating or surface grafting of phosphorylcholine moieties. Phosphorylcholine is a zwitter ionic moiety, that can benefits from its reduced protein [25], platelet [26] and cell adhesion [27] in vitro [28]. Therefore, modification of intravascular stents with phosphorylcholine can lead to antithrombotic effects. DES with phosphorylcholine polymer-coated surfaces have been commercially available for over 5 years [29]. Significant differences in cardiac death/myocardial infarction and composite endpoints favored treatment with phosphorylcholine polymercoated DES over comparator BMS have been observed in the 5-year clinical applications. Moreover, rates of clinical restenosis and safety events, including stent thrombosis beyond the first year of revascularization, have been proved to remain stable, which is a significant improvement comparing to first-generation DES [30]. The successes of phosphorylcholine-coated coronary stents inspired researches to explore other implantable materials. An increasing number of functionalized and complicated phosphorylcholine containing polymers [27,31,32], along with more other coated/grafted substrates [33–35] are under investigation right now. Meanwhile simple routes to fabricate biocompatible surfaces are still being desired to avoid complicated evaluations and to prevent uncertain risks, especially by potential implantable medical devices manufacturer. Therefore, direct grafting of organic biofunctional molecules onto metal surface are still an attractive and cost-efficient method for intravascular stents development [36–38]. Unfortunately, organic reactions on an inert metallic surface are usually difficult, especially for those biomolecules which can only tolerate mild reaction conditions. Thus, how to introduce a reactive surface for intravascular implantable metals become a stepping stone to success. In the previous reports, surface-initiated atom transfer radical polymerization (ATRP) [39,40], RF-glow discharge plasma [41], and both air plasma and ATRP [42] processes were applied in order to enhance the stainless steel surface grafting of polymeric moieties. In this study, we creatively used atomic layer deposition (ALD) technology to achieve a biofunctional surface by enhancing the graft of phosphorylcholine onto stainless steel. ALD was originally developed in the 1970’s [43] to enable manufacturing of thin film electroluminescent displays, and then used in integrated circuit fabrications in the 1990’s, normally worked at pressure of 0.01–1 kPa and temperatures of 100–400 ◦ C. ALD is a self-limiting (the amount of film material deposited in each reaction cycle is constant), sequential surface chemistry that deposits conformal thin-films of materials onto substrates of varying compositions. Due to ALD’s self-limiting as well as surface-reactive nature, ALD film growth makes atomic scale deposition control possible. In terms of the chemistry, ALD is similar but different to chemical vapor deposition (CVD). By pulsing a purge gas (typically N2 or Ar) after each precursor pulse to remove excess precursor from the process chamber and prevent ‘parasitic’ CVD deposition on the substrate, ALD keeps the precursor materials separate during the reaction. This separation is maintained throughout the coating process, thus atomic layer control of film growth can be obtained as fine as about 0.1 A˚ (10 pm) per cycle. Currently, ALD can be used

239

to deposit wide variety of thin films such as metals (e.g. Ru, Ir, Pt), metal nitrides (e.g. TiN, TaN, WN, NbN), metal sulfides (e.g. ZnS) and more important to us, various metal oxides (e.g. Al2 O3 , TiO2 , SnO2 , ZnO, HfO2 ) [44,45]. It is the metal oxide that offers opportunity to fabricate intermediate layer between metallic substrate and organic biomolecules. Narayan et al. [46] reported a ZnO and Pt coated alumina nanoporous membrane which was achieved by ALD. The surface was PEGylated to promote medical and environmental health applications. In 2011, the stainless steel with ALD mediated SiO2 coating was found to be superior in terms(NY) of its performance in electrochemical testing. The novel coating that was achieved with ALD technology was also proved to be functionalized with biologically significant carbohydrates, implying that ALD has the potential of being adapted to the functionalization of metallic biomedical implants with other biologically relevant carbohydrates [47]. In this paper, we report a method to adopt ALD for the deposit of Al2 O3 thin film onto model stainless steel sheets, which is followed by the direct connection of phosphorylcholine moiety onto the Al2 O3 surface. The physical–chemical properties, hema-compatibilities and interactions with endothelial cells of the surfaces were investigated. Furthermore, the actual improvements of surface performance the ALD film have on intact coronary stents were studied. 2. Experimental 2.1. Materials High-purity (99.9999%) trimethylaluminum (TMA, Al(CH3 )3 ) was supplied by Jiangsu Nata Opto-electronic Material Co., Ltd. (Nanjing, China). Silane coupling agent, 3-aminopropyltriethoxysilane (APS) was purchased from Nanjing Shuguang Chemical Group Co. (Nanjing, China), and inhibitor-free 2-methacryloyloxyethyl phosphorylcholine (MPC, purity at 96%) was purchased from Joy-Nature Institute of Technology (Nanjing, China). Other chemicals were all purchased from Sinopharm Chemical Reagent Co. (Shanghai, China) and used as received. Platelet-rich plasma (PRP), prepared on a Baxter CS-3000 Plus blood cell separator with 106 platelets/␮L, was supplied by Jiangsu Province Blood Center (Nanjing, China) and used within 24 h after collection. Human umbilical vein endothelial cells (HUVEC) were supplied by Lifeline Cell Technology (Distributed by Beijing Qingyuanhao Biologics, Beijing, China). VEGF LS-1020 culture medium were purchased from Thermo Fisher Scientific China Branch (Shanghai, China). Fetal bovine serum (FBS) was purchased from Zhejiang Tianhang Biological Technology (Hangzhou, China), sterilized at 56 ◦ C for 30 min and then stored at −20 ◦ C. Carboxyfluorescein diacetate succinimidyl ester (CFDA-SE) was purchased from Dojindo Molecular Technology (Shanghai, China). 2.2. Preparation of samples Electropolished (kindly customized by Promed Medical Tech. Co., Ltd., Suzhou, China) or mechanical polished (only used in platelet adhesion experiments) 316L stainless steel (SS) sheets (20 × 20 mm) were ultrasonic cleaned in acetone, ethanol and distilled water for 3 times, respectively. Then 10 nm amorphous Al2 O3 layer (100 cycles of ALD deposition) was deposited on SS surface by a Picosun SUNALETM R-150B ALD reactor (Picosun, Finland), in which pulsed chemical precursors of Al(CH3 )3 (Al precursor) and H2 O (oxygen precursor) were supplied alternately in a N2 carrier gas at 150 ◦ C. Then SS sheets and the SS sheets with Al2 O3 film (SS–Al) were immersed in 1% APS ethanol solution for 30 min at room temperature and then washed with ethanol for 3 times. The

240

Q. Zhong et al. / Colloids and Surfaces B: Biointerfaces 121 (2014) 238–247

Fig. 1. Atomic layer deposition illustration and phosphorylcholine grafting route. The repeat ALD cycle N = 100.

samples were dried in air for 1 h at 110 ◦ C, and then were soaked in 1% ethanol solution of MPC for 24 h at room temperature. After washed with ethanol and dried for 24 h at 40 ◦ C, SS and SS–Al sheets with MPC layer were obtained and labeled as SS–PC and SS–Al–PC, respectively. The sample preparation and ALD process were illustrated as Fig. 1. 2.3. Instrumentations X-ray photoelectron spectroscopy (XPS) of samples were recorded on a Thermo Scientific K-Alpha Photoelectron Spectrometer equipped with a monochromatic Al-K␣ X-ray source to determine the surface elemental composition of each sample; the XPS spectra were acquired in the constant analyzer energy (CAE) mode at a 90◦ take-off angle. XPS Peak software (v4.1) was utilized to analyze and deconvolute the XPS peaks. Peak deconvolutions were performed using optimized Lorentzian–Gaussian components after a Shirley background subtraction. A tapping mode NanoScope IIIa-Phase atomic force microscope (AFM) (Digital Instruments, USA) was utilized to observe the micromorphology of the samples. The water contact angles of samples were recorded with a KSV CAM200 optical contact angle and surface tension meter system (KSV Instruments, Finland) at 20 ◦ C with relative humidity of 85%.

unlabeled) in PBS at a total concentration of 1 mg/mL. The steel sheets were cut into 5 × 5 mm and were prepared as before. All samples were immersed in PBS for 12 h to achieve complete hydration and then transferred to 96-well microtiter plates, with each well containing 250 ␮L of protein solution. Adsorption was allowed to proceed for 3 h under static conditions at room temperature. Following adsorption, the surfaces were immediately rinsed in fresh PBS (3 times, 10 min for each time) to remove loosely adsorbed proteins. The samples were then wicked onto filter paper, and transferred to clean counting vials for radioactivity determination using a Wizard 3 1480 Automatic Gamma Counter (PerkinElmer Life Sciences, USA). 2.5. Platelet adhesion Twenty microliter PRP was dropped onto the sample surface and kept for 30 min at room temperature. The samples were gently washed with PBS for 3 times and then dipped into 2.5% glutaraldehyde solution for 30 min to fix the platelets. After that, the samples with fixed platelets were dehydrated for 15 min in each of a series of ethanol/water solutions at the concentration of 50, 60, 70, 80, 90, 95 and 100% (v/v) and then dried in air at room temperature. All samples with gold coating (approx. 40 nm thickness) were observed by FEI Quanta 200 scanning electron microscope (SEM, FEI, USA) at an acceleration voltage of 20 kV.

2.4. Protein adsorption 2.6. Proliferation of HUVECs Fibrinogen (Fbg) was labeled with 125 I (Isotope Company of China, Beijing, China) using the iodine monochloride (ICI) method. The labeled protein was dialyzed overnight against phosphate buffer solution (PBS, 0.1 mol/L, pH7.4) to remove the free iodide. Labeled Fbg was mixed with unlabeled Fbg (1:19, labeled:

HUVEC cell density was counted and adjusted to 5 × 105 cells/mL. Four samples (SS, SS–Al, SS–PC and SS–Al–PC) were placed in a glass culture dish as one group. The cells were finally seeded in complete cell culture medium and incubated for 4 h, 8 h, 16 h, 1 day,

Q. Zhong et al. / Colloids and Surfaces B: Biointerfaces 121 (2014) 238–247

241

Fig. 2. XPS survey scan of SS and SS–Al; high resolution scan for C of SS–PC and SS–AL–PC, and high resolution scans for O, N, P and Si of SS, SS–PC, SS–Al and SS–Al–PC, respectively.

3 days, and 5 days at 37 ◦ C in a humidified 5% CO2 atmosphere. After incubation, the cell culture medium was removed and the samples were washed 3 times in PBS and fixed with 4% paraformaldehyde solution for 15 min, then washed 3 times in PBS. All the cells fixed were stained with 20 ␮L 20 ␮M CFDA-SE solution for 15 min in the incubator. Finally, the CFDA-SE solution was removed and samples were washed with distilled water. The adhesion and proliferation of HUVECs were observed with a Zeiss Axio Scope A1 fluorescent microscopy (Zeiss, Germany) with 100 times magnification. 2.7. Coronary stents coating and investigating The 316L stainless steel coronary stents (expanded nominal diameter (DN) of 3.0 mm and length of 18 mm) used in this study were kindly supplied by Promed Medical Tech. Co., Ltd. (Suzhou, China). For those stents with Al2 O3 surface layer, Al2 O3 with thickness of 10 nm was deposited using ALD reactor as described before. Poly(lactide-co-glycolide) (PLGA, PURASORB PDLG 5004, Purac Biomaterials, Netherlands) were also customized coated in the cleaning shop of Promed Co. using ultrasonic spray, the thickness of coating was 6 ± 1 ␮m. The coating on stents were then observed with an S-3400N II SEM (Hitachi, Japan) at an acceleration voltage of 20 kV. 3. Results and discussion The surface of metallic materials will be naturally oxidized by sparse hydroxyl groups distributed on the surface, which is due to the lattice imperfection as well as water on the surface that is absorbed by van der Waals force [48]. However, the concentrations of surface hydroxyl groups are too low to participate in reactions effectively; therefore metals are usually stable and inert

in organic chemical reactions. In order to graft phosphorylcholine moiety onto the surface of intravascular stents made from metallic materials, such as stainless steel, ALD was employed to deposit Al2 O3 thin film onto stainless steel surface. According to the principle of ALD [45], the aluminum atoms at the top of the surface are all hydroxylated, which can provide enough hydroxyl group to react with coupling agent APS which will be subsequently chemically link to phosphorylcholine through Michael addition reaction [49]. The phosphorylcholine moiety is expected to improve the biocompatibility of stainless steels, which is one of the key properties of intravascular stents.

3.1. Surface characterizations Surface chemical composition of the samples can be conveniently investigated by XPS with high sensitivity. Fig. 2 shows the survey scan of SS and SS–Al. Typical elemental scan of pristine SS and modified samples SS–PC, SS–Al, and SS–Al–PC were also performed and identified according to the binding energy (BE). The survey scan of SS shows the presence of iron (BE 711, 725 and 784 eV), chromium (BE 577 and 587 eV), nickel (BE 855 eV), molybdenum (BE 232 eV), and manganese (BE 641 eV). The existence of oxygen (BE 531 eV) and carbon (BE 285 eV) revealed oxidation of the surface. The results are in agree with the chemical composition of 316L stainless steel and identical to the previous reports [50,51]. After Al2 O3 was deposited on SS surface, the most obvious change on XPS spectrum is the appearing of alumina correlated signals at binding energy of 75 eV and 120 eV, attributed to Al2p and Al2s, respectively. Meanwhile, the intensity increasing of O1s signal, decreasing of C1s signal, and the completely disappearance of other metallic elemental signals revealed the complete cover of the SS surface by Al2 O3 . In O1s high resolution scan, the O* H

242

Q. Zhong et al. / Colloids and Surfaces B: Biointerfaces 121 (2014) 238–247

Fig. 3. AFM 3D images (A) and roughness analysis (B) of pristine SS and stepwise modified surfaces.

signal strengthened dramatically while the O2− signal almost disappeared entirely after Al2 O3 deposition. These results indicated that the metallic surface of SS has been successfully converted into metal–oxide surface, and hydroxyl group distribution density on the surface has increased significantly. In the high resolution scans of elements correlated to APS and phosphorylcholine, the signals of nitrogen (BE 398.4 eV for N1s), phosphorus (BE 134.1 eV for P* O) and silicon (BE 102.1 eV for Si* O) in sample SS–PC and SS–Al–PC are all enhanced remarkably comparing to SS and SS–Al. These changes are all due to the surface coating of APS and MPC. By comparing O1s scan of SS–PC to O1s scan of SS, a small change was found. This small change indicates that not much APS was grafted following MPC. In contrary, the grafting of APS and MPC on SS–Al is more efficient. In the comparison of C1s scan of SS–PC and SS–Al–PC, the signals at 285.1 eV are attributed to C C* bonds, the signals at 288.7 eV are attributed to C* O bonds, while the signals at 286.8 eV are attributed to N C* bonds. More importantly, the proportion of N C* bonds in SS–Al–PC is much higher than that of SS–PC, which proved that the grafting efficiency on SS–Al is higher than that on SS. The existence of the mentioned bonds indicated successful grafting of APS and MPC, which further confirmed the chemical grafting of phosphorylcholine moiety as what we designed. The changes in topography of the SS surface after each functionalization step were investigated using AFM under dry conditions, using a tapping mode at a scan rate of 0.5 Hz over an area of 2 × 2 ␮m. Fig. 3A shows 3D images of the pristine and stepwise functionalized surfaces in an area of 4 ␮m2 . Fig. 3B shows the roughness statistical analysis on these regions. We can see from Fig. 3A that the pristine electropolished SS surface is fairly homogeneous and smooth with a root-mean-square surface roughness (Ra) of about 0.23 nm in the analyzed 4 ␮m2 region. The Ra value of the SS–Al surface slightly increased to about 0.40 nm, indicating that ALD can deposit Al2 O3 homogenously and maintain the smoothness of SS surface. Additionally, the direct APS and MPC modification on SS surface has no significant effects on the roughness because the Ra value of SS–PC remains at about 0.65 nm. After the surface grafting of APS and MPC, the Ra value for SS–Al–PC surface increased to about 1.82 nm, indicating the enhancement of grafting efficiency by Al2 O3 surface on SS. The Ra values in 4 ␮m2 area of analysis are surprisingly small; so the analyzed areas are enlarged to 100 ␮m2 . The obtained Ra values from the enlarged area for SS, SS–PC, SS–Al and SS–Al–PC are 2.24, 2.56, 2.38 and 4.72 nm, respectively, which is identical to the previous reported data [52]. It is obvious that the variation tends of roughness are similar either in 4 or in 100 ␮m2 region. Both of them demonstrate that ALD can maintain the smooth surface of SS. Meanwhile the trends also showed that more MPC can be introduced on SS–Al by grafting than on SS; therefore, ultimately increase the roughness of the surface.

Fig. 3B shows the peak-height distribution of every point of areas in Fig. 3A and the inset shows the average heights with error ranges. The images in Fig. 3A were sampled as many as 262,144 points, the height of every point was recorded and all data were analyzed statistically. It is found that the average height of SS, SS–PC, SS–Al and SS–Al–PC are 3.14, 3.69, 5.22, and 10.08 nm, respectively. As showed in histogram, the height distributions of SS and SS–Al surface are concentrated, while the distribution broadened obviously after modified with APS and MPC on SS–PC and SS–Al–PC. Moreover, it is apparent that the broadening on SS–Al–PC is much higher than that on SS–PC, indicating the grafting reaction on SS–Al is more efficient than that of on SS due to the increase in the quantity of hydroxyl groups. 3.2. Wettability, protein adsorption and platelets adhesion of surfaces Water contact angle measurement is a convenient way to assess the wettability of a surface, the method can provide macroscopically information of the interaction between surface and water [53]. Fig. 4A shows the water contact angle of pristine SS and the modified stainless steel surfaces. As shown in this figure, after being grafted with zwitter ionic phosphorylcholine, the water contact angle significantly decreased from 85◦ of SS to 59◦ of SS–PC, and 51◦ of SS–Al to 45◦ of SS–Al–PC in 85% relative humidity. This is mainly caused by the formation of hydration layer via electrostatic interaction and hydrogen bonding between phosphorylcholine groups and water in air [34]. The deposition of Al2 O3 on SS also caused the significant decrease of water contact angle because of the introduction of large amount of hydroxyl groups which can adsorb water in air through hydrogen bonding interactions. Comparing to SS, the water contact angles of both SS–AI and SS–AI–PC decreased. The decrease of water contact angle of SS–Al–PC is remarkably more than that of SS–PC, indicating higher grafting efficiency on SS–Al due to the enrichment of hydroxyl groups on Al2 O3 layer. It is expected that the improvement of wettability of SS sheets will be beneficial to the compatibility and the interactions of SS with biomolecules. The amount of protein absorbed on surface is reported to be one of the most important factors in evaluating the biocompatibility of materials, because protein adsorption is the first event that happens during the blood–material interactions, especially in terms of the proteins such as clotting enzymes and fibrinogen [54,55]. Therefore, in order to improve the biocompatibility of a material surface, the extent of protein adsorption is always expected to be reduced to as low as possible. In this study, Fbg adsorptions are studied in view of its significance in platelet adhesion: Fbg can bind to the platelet GP IIb/IIIa receptor and active platelets [56]. Fig. 4B shows the results of Fbg adsorption on four different samples. In comparison with pristine SS, the amounts of Fbg on SS–Al, SS–Al–PC was significantly smaller while statistically indistinctive on SS–PC. In the four

Q. Zhong et al. / Colloids and Surfaces B: Biointerfaces 121 (2014) 238–247

243

Fig. 4. Water contact angle (A) and Fibrinogen adsorption (B) on sample surface. The inset in (A) shows representative 5 ␮L water drop images on the sample surfaces.

kinds of surfaces, SS–Al–PC shows the minimum Fbg absorption. The results are consistent with the water contact angle analysis, indicating that the increased wettability of the modified SS surface might be the key factor for the resistance to protein adsorption. As described in AFM observation and water contact angle sections, the zwitter ionic and hydrophilic phosphorylcholine on SS surface constructed a hydration layer through the electrostatic interactions and hydrogen bonds with water molecules in air, which prevented adsorption from happening on Fbg. As described by Ishihara et al. [25], water molecules can also bind to the hydrophobic surface through van der Waals interactions, which is the so-called “hydrophobic hydration”. These bound water molecules cause protein adsorption by hydrophobic interaction. When a protein molecule is adsorbed on a surface, water molecules between the protein and surface needed to be replaced, thus protein adsorbed on the surface lost bound water at the surface-contacting portion. This phenomenon induced conformational change in the proteins, wherein the hydrophobic part of the protein is exposed to the surface. The longer time a protein is bound to a surface, the greater the conformational changes would be. Ultimately, irreversible adsorptions would occur. However, on a hydrophilic surface, the hydrophobic interaction does not occur between protein and surface. Moreover, the conformational change resulted from protein adsorption or contact with the hydrophilic surface is also suppressed. When the free water on the surface is kept at a higher level, proteins can contact the surface reversibly, without significant conformational change and irreversible adsorption. As a result, the protein adsorption on the surface decreases. Acute thrombus formation is one of the most dangerous events during interventional therapy. Platelet spreading and aggregation are marks of platelet activation and are considered to be a major mechanism of thrombosis [24]. Concerning the safety of intravascular stents, the platelet adhesion on surface of stent materials must be dealt with. We evaluated the platelet adhesion on our samples in vitro using fresh PRP. The results of the evaluation were also part of the comparison of the difference of different surfaces. It should be noted that the sample sheets used in this experiment are not polished by electro polishing, but are mechanically polished with 3000 mesh abrasive paper and modified in the same way as electro polished ones. The reason for not using electro polishing is that the plantlets adhered on the electro polished surface is too little to distinguish for different grafting modifications. It’s well known that there would be much more platelets adhered on a rough than on a smooth surface [23]. Therefore, the mature electro polishing technology, which would yield a very smooth surface, used in intravascular stents has decreased the extent of platelet adhesion to a very low level. In our preliminary study, we also found that the difference of platelet adhesion on electro polished samples is too

small to exhibit the effects of modification itself. Because of all the reasons stated above, the mechanical polished samples were used. Fig. 5A shows the representative SEM photographs of platelets adhered on SS and modified SS surfaces. Fig. 5B is the mean value of the statistic results of platelets adhesion. It can be seen that numerous platelets aggregated on surface of pristine SS, but after modification, especially with phosphorylcholine, the amounts of platelet adhesion on SS–PC and SS–Al–PC decreased dramatically. The phenomena clearly demonstrated the benefits of phosphorylcholine [33,53,57]. Comparing SS–AI–PC to SS–PC, the platelet adhesion reduction on SS–Al–PC is more remarkable, implying the advantages of grafting phosphorylcholine on Al2 O3 bottom layer and the potentials of applying ALD in medical devices. Similar conclusions were also reported by Narayan et al. previously [46]. 3.3. Proliferation of HUVECs on various sample surfaces The endothelium is an active organ that maintains vessel integrity with dynamic mechanisms which prevent thrombosis and intimal hyperplasia from happening [58]. Studies that involve in vitro endothelialization of intravascular stents with cultured endothelial cell prior to implantation showed that a confluent endothelium prevents thrombogenic complications and improves long-term patency [22,31,32,58]. Hence, materials that promote in situ endothelialization of cardiovascular implants (without intimal hyperplasia or thrombus formation during endothelium development) would be highly desirable [59]. Utilizing ALD, the grafting of phosphorylcholine on stent material surface has been successfully enhanced. The modified surfaces have exhibited excellent inhibitory effects on Fbg adsorption and platelet adhesion, thus, the endothelialization on the surfaces are also expected to improve. The initial in vitro response of HUVECs to different surfaces were evaluated in terms of cell attachment, while the long-term HUVEC growth was investigated as proliferation [58]. Fig. 6 shows the HUVEC surface images after several incubation periods. In the initial stage (4–16 h), HUVECs attached to the surfaces and their pseudopodia extended but the cells distributed separately. The interaction between surface and cells is mainly adsorptive action. 1 day after of cell planting, a number of cells begin to divide and many binucleate cells were found, which means the cells entered mitotic phase. 3 days after cell planting, the cell number increased significantly and the interactions of cell to cell started. Cells adhering to the surfaces continues to grow. In the 5th day, the cells almost covered the entire surface. The number of cells attached on the surfaces were calculated based on the images shown in Fig. 6. Fig. 7 shows the statistical results of HUVEC density calculated from the images. Increasing

244

Q. Zhong et al. / Colloids and Surfaces B: Biointerfaces 121 (2014) 238–247

Fig. 5. Platelet adhesive evaluation on sample surfaces. Where (A) SEM images of platelets adhered on four samples with magnification of 600 times, and (B) average number of platelets adhesion.

the incubation time from 4 to 16 h results in about 50% increase in HUVECs attached to surfaces, but in the whole 5-day incubation, the proliferation of HUVEC increased near linearly. In the first day, the difference among cell density on SS–PC, SS–Al, and SS–Al–PC is

not significant but slightly more than that of on SS. In the third day, the cell densities on SS–PC, SS–Al and SS–Al–PC are all significantly larger than that of on SS. Especially on SS–Al–PC, the cell density is significantly higher than that on SS–PC and SS–Al–PC, while the

Fig. 6. HUVEC attachment and proliferation on SS and modified sample surfaces with time.

Q. Zhong et al. / Colloids and Surfaces B: Biointerfaces 121 (2014) 238–247

245

Fig. 7. HUVEC densities (A) and HUVEC proliferation rates (B) on pristine SS and modified surface.

latter two have similar cell density (Fig. 7A). After been proliferated for 5 days, the increase of cell densities keep going on. In general, HUVEC proliferation rate, which is reflected by the slope of cell proliferation curves as shown in Fig. 7B, on SS–Al–PC is significantly higher than that on other surfaces, indicating that the phosphorylcholine grafting on Al2 O3 bottom layer is advantageous than that on pristine SS. By depositing Al2 O3 onto SS surface, abundant hydroxyl groups were introduced and then phosphorylcholine moieties were grafted high-efficiently, which fabricated a mimetic cell membrane surface [27,32,57] and increased the roughness and hydrophility of the surfaces, all of these transformation can contribute to the observed increase of HUVEC proliferation. The similar positive effects of hydrophility on HUVEC proliferation has also been reported in the literatures [31,46,53]. Besides this advantage, better HUVEC proliferation on SS–Al–PC are also found to be related to lower extent of toxic species such as Cr- and Ni-ions releasing at the cell/surface interface [60,61] due to the complete cover of SS surface by Al2 O3 , which is also proved by better HUVEC proliferation on SS–Al compared to SS. It’s well known that phosphorylcholine-grafted surface is a typical non-fouling surface with the ability to resist protein adsorption, platelet adhesion and cell attachment. In our investigation, we also found that the phosphorylcholine grafted SS–Al–PC surface can effectively resist the adsorption of Fbg and the adhesion of platelets, but the enhancement of HUVEC attachments and proliferations were also observed. Actually, the attachment and proliferation of cells are much more complicated than that of protein adsorption and platelet adhesion, which are influenced not only by the chemical composition but also effected by surface morphology, hydrophility and cell microenvironment. Moreover, there are no spacers between the phosphorylcholine structure and the substrate Al2 O3 layer in our system, which may acutely limited the anti-fouling effects of phosphorylcholine moieties. Relative to the phosphorylcholine itself, the changes on surface morphology, hydrophility, the lower extent of toxic species due to the ALD process and grafting of phosphorylcholine moieties probably played greater roles on the cell interactions thus caused the enhancement of HUVEC attachments and proliferations. 3.4. Profits of ALD on coronary stents with biodegradable coating Coronary stents are representative implantable intravascular stents, most of which are manufactured with 316L stainless steel [23]. By depositing a thin layer of Al2 O3 followed by grafting with phosphorylcholine, 316L stainless steel can inhibit protein adsorption and platelet adhesion, meanwhile induce the attachment and proliferation of HUVEC due to the formation of mimetic cell membrane surfaces. The Al2 O3 bottom layer as thin as 10 nm provides

significant changes to SS surface in terms of its chemical property, micromorphology, wettability and biocompatibility. It was also expected to improve the metal–polymer interface properties of DES. Currently, most of DES are manufactured by coating polymeric dug carrier onto metallic stent surfaces, for example, polyethylene co-vinyl acetate and poly-n-butyl methacrylate, poly(styreneb-isobutylene-b-styrene), and polylactide (PLA) on 316L stainless steel; polyvinylidene fluoride co-hexafluoropropylene and poly-n-butyl methacrylate, phosphorylcholine, and poly(lactideco-glycolide) on Co–Cr alloy [62,63]. The coating properties such as uniformities, integrities as well as mechanical properties can dramatically impact the drug release profiles and even the clinical effects of DES, especially for those DES using biodegradable polymers [64,65], while the coating performance itself is closely related to the metal–polymer interactions on their interface. Thus, the modification on metallic surface with ALD deposited Al2 O3 might also influence the coating properties in DES. Utilizing Al2 O3 covered 316L coronary stents, PLGA coated coronary stents were prepared and investigated using SEM to inspect uniformity and integrity of coating under unexpanded and balloon-expanded status. As a control, PLGA-coated stents without Al2 O3 bottom layer are also prepared at the same time. Fig. 8 shows the representative SEM images of the stents. In Fig. 8, images A–C show the coating on stents without Al2 O3 bottom layer. As shown in Fig. 8A, the stents surface with PLGA coating looks uniform, but upon zooming in, wrinkles can be found (Fig. 8B). After being expanded, stent occurs plastic deformations and the coating on struts merely keeps its integrity. But at the curved portion, the coating cracks heavily and the stent surface has been completely exposed to air (Fig. 8C). Such fracture is unacceptable for a coating stent because the crack will cause the coating peel and detach. But on Al2 O3 covered stent, the situation has been improved dramatically as shown in Fig. 8D–F. As presented in Fig. 8D, the stent surface with PLGA coating on Al2 O3 bottom layer looks smooth and integrity, wrinkles have completely disappeared. After being expanded and zoomed in to 1000 times magnifications, even the coating at curved portion still keep perfect (Fig. 8E). Only when it was investigated with magnification of 3000 times, tiny fissures was found at the curve portion, as shown in Fig. 8F. Even so, the coating still completely wraps on the stent. Apparently, while all other conditions remain the same, the application of Al2 O3 bottom layer deposited by ALD is of great worth. It is the Al2 O3 bottom layer that provided abundant hydroxyl groups which changed the polarity and wettability of stent surface; therefore PLGA solution can spread on the surface with higher levelling property. Furthermore, hydroxyl groups can afford strong interactions with PLAG polymer chains through hydrogen bonds and electrostatic forces.

246

Q. Zhong et al. / Colloids and Surfaces B: Biointerfaces 121 (2014) 238–247

Fig. 8. SEM images of PLGA coated 316L coronary stents without (A–C) or with (D–E) Al2 O3 bottom layer. Where (A) unexpanded stents (DN 1 mm) observed at 32 times magnifications; (B) unexpanded stents observed at 200 times magnifications; (C) expanded stents (DN 3 mm) observed at 1000 times magnifications; (D) unexpanded stents observed at 200 times magnification; (E) expanded stents observed at 1000 times magnifications, and (F) expanded stents observed at 3000 times magnifications.

For DES, the coating integrity was once a challenging problem due to polymer coating is difficult to adhere on inert, mirror-like smooth stent surface, especially at the curve portions of stents. Additionally, when stent is undergoing expansion plastic deformation occurs, causing the coating to crack. The problem was resolved by using poly-p-xylylene (Parylene) bottom layer to produce an intermediate between polymer coating and stent metal [66]. For DES with durable polymer coating, the Parylene bottom layer is always covered by top drug-carrying polymers and the configuration is acceptable. But for DES with biodegradable polymer drug-carrying coating, the durable Parylene bottom layer is unfavorable because the layer will always be there and hinder reendothelialization on stent surface. Comparing to Parylene surface, the re-endothelialization on bare metal surface is more rapidly [14]. Therefore, in the development of DES with biodegradable polymer coating, e.g. PLA and PLGA, it is always expected that novel technologies that can avoid the polymeric bottom layers, such as Parylene can be applied [67]. The ameliorative coating integrity obtained from PLGA coating stents using Al2 O3 as bottom layer has pointed to a promising direction: using ALD technology, an ultrathin layer (with the thickness in the range of nanometers) of metal oxide can be produced to resolve the problem bypassing the need of any durable polymers. The ALD technology is also expected to be beneficial to other intravascular stents, or even other interventional polymer coated medical devices. 4. Conclusions Using ALD technology that is adapted from semiconductor industry, an ultra-thin layer (10 nm) of Al2 O3 was easily deposited on 316L stainless steel surface. This ultra-thin layer of Al2 O3 provides sufficient hydroxyl groups to react with APS and MPC for the purpose of surface modification. The resulted surface is more hydrophilic and cell-membrane mimicking than pristine stainless steel surface. The modification prevents protein absorption and platelet adhesion on the stainless steel while promotes HUVEC attachment and proliferation, which will result in the decrease of restenosis and acceleration of re-endothelialization of intravascular stents that’s made from stainless steel. In addition, by introducing Al2 O3 onto DES surface using ALD technology, adhesion of polymer coatings on inert, mirror-like smooth surface of the

intravascular stents can be improved. Furthermore, the integrity of the polymer coating can be maintained. Acknowledgments The authors acknowledge the support of Science and Technology Support Program (Society Development) of Jiangsu Province (No. BE2012735), the Key Project of Chinese Ministry of Education (No. 109070), the National Basic Research Program of China (No. 2011CB933400). The works are also a part of the Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31]

G.G. Stefanini, D.R. Holmes, N. Engl. J. Med. 368 (2013) 254. C.J. White, S.R. Ramee, T.J. Collins, et al., J. Endovasc. Surg. 5 (1998) 71. M.R. Mayberg, Clev. Clin. J. Med. 71 (Suppl 1) (2004) S42. A. Lejay, F. Thaveau, E. Girsowicz, et al., J. Cardiovasc. Surg. 53 (2012) 171. B.A. Gross, K.U. Frerichs, J. Neurol. Neurosurg. Psychiatry 84 (2013) 244. M.R. Bennett, M. O’Sullivan, Pharmacol. Ther. 91 (2001) 149. M.R. Bennett, Heart 89 (2003) 218. P.W. Serruys, F. Unger, J.E. Sousa, et al., N. Engl. J. Med. 344 (2001) 1117. E. Grube, B. Chevalier, G. Guagliumi, et al., Am. Heart J. 163 (2012) 867. V. Dzavik, U. Kaul, G. Guagliumi, et al., Catheter. Cardio. Interv. 82 (2013) E163. M.C. Morice, P.W. Serruys, J.E. Sousa, et al., N. Engl. J. Med. 346 (2002) 1773. G.W. Stone, S.G. Ellis, D.A. Cox, et al., N. Engl. J. Med. 350 (2004) 221. C. Stettler, S. Wandel, S. Allemann, et al., Lancet 370 (2007) 937. R. Wessely, Nat. Rev. Cardiol. 7 (2010) 194. J.P. Becquemin, J.P. Favre, J. Marzelle, et al., J. Vasc. Surg. 37 (2003) 487. P. Dick, H. Wallner, S. Sabeti, et al., Catheter. Cardio. Interv. 74 (2009) 1090. R. Mohammadian, E. Sharifipour, R. Mansourizadeh, et al., Neuroradiol. J. 26 (2013) 454. J.W. Jukema, J.J. Verschuren, T.A. Ahmed, et al., Nat. Rev. Cardiol. 9 (2012) 53. S. Meng, Z. Liu, L. Shen, et al., Biomaterials 30 (2009) 2276. A. de Mel, G. Jell, M.M. Stevens, et al., Biomacromolecules 9 (2008) 2969. Q. Lin, X. Ding, F. Qiu, et al., Biomaterials 31 (2010) 4017. M. Morra, Expert Rev. Med. Devices 4 (2007) 361. F. Nazneen, G. Herzog, D.W. Arrigan, et al., J. Biomed. Mater. Res. B: Appl. Biomater. 100 (2012) 1989. K. Zhang, T. Liu, J.A. Li, et al., J. Biomed. Mater. Res. A 102 (2014) 588. K. Ishihara, H. Nomura, T. Mihara, et al., J. Biomed. Mater. Res. A 39 (1998) 323. Y. Xu, A. Dong, Y. Zhao, et al., J. Biomater. Sci., Polym. Ed. 23 (2011) 2089. K. Jang, K. Sato, K. Mawatari, et al., Biomaterials 30 (2009) 1413. S.W. Jordan, E.L. Chaikof, J. Vasc. Surg. 45 (Suppl A) (2007) A104. J. Bridges, D. Cutlip, Med. Devices 2 (2009) 1. D.E. Kandzari, M.B. Leon, I. Meredith, et al., JACC—Cardiovasc. Interv. 6 (2013) 504. Y. Wei, Y. Ji, L.L. Xiao, et al., Biomaterials 34 (2013) 2588.

Q. Zhong et al. / Colloids and Surfaces B: Biointerfaces 121 (2014) 238–247 [32] S.L. Chin-Quee, S.H. Hsu, K.L. Nguyen-Ehrenreich, et al., Biomaterials 31 (2010) 648. [33] S.H. Ye, Y.S. Jang, Y.H. Yun, et al., Langmuir 29 (2013) 8320. [34] C.J. Pan, Y.H. Hou, H.Q. Liu, et al., Colloids Surf., B: Biointerfaces 112 (2013) 508. [35] M. Kyomoto, T. Moro, S. Yamane, et al., Biomaterials 34 (2013) 7829. [36] C.K. Kang, Y.S. Lee, J. Mater. Sci.—Mater. Med. 18 (2007) 1389. [37] C.K. Kang, Y.S. Lee, J. Ind. Eng. Chem. 18 (2012) 1670. [38] C.K. Kang, W.H. Lim, S. Kyeong, et al., Colloids Surf., B: Biointerfaces 102 (2013) 744. [39] S.J. Yuan, S.O. Pehkonen, Y.P. Ting, et al., Langmuir 26 (2010) 6728. [40] S. Yuan, D. Wan, B. Liang, et al., Langmuir 27 (2011) 2761. [41] C. Vreuls, G. Zocchi, B. Thierry, J. Mater. Chem. 20 (2010) 8092. [42] W. Guo, J. Zhu, Z. Cheng, ACS Appl. Mater. Interfaces 3 (2011) 1675. [43] T. Suntola, J. Antson, US Patent 4,058,430, 1977. [44] R.L. Puurunen, J. Appl. Phys. 97 (2005) 121301. [45] S.M. George, Chem. Rev. 110 (2010) 111. [46] R.J. Narayan, S.P. Adiga, M.J. Pellin, et al., Philos. Trans. R. Soc. London, Ser. A 368 (2010) 2033. [47] A.M. Slaney, V.A. Wright, P.J. Meloncelli, et al., ACS Appl. Mater. Interfaces 3 (2011) 1601. [48] E. McCafferty, J.P. Wightman, Surf. Interface Anal. 26 (1998) 549. [49] A. Zhu, B. Shan, Y. Yuan, et al., Polym. Int. 52 (2003) 81.

247

[50] T. Hanawa, S. Hiromoto, A. Yamamoto, et al., Mater. Trans. 43 (2002) 3088. [51] K. Prabakaran, S. Rajeswari, J. Appl. Electrochem. 39 (2009) 887. [52] G. Selvaduray, S. Trigwell, in: W. Baxter, A.P. Lee, et al. (Eds.), Proc. of NanoBio2006 Frontiers in Biomedical Devices Conference, Irvine, CA, USA, 2007, 8–9 June, 2006, ASME, New York. [53] B. Gao, Y. Feng, J. Lu, et al., Mater. Sci. Eng., C 33 (2013) 2871. [54] P. Roach, D. Farrar, C.C. Perry, J. Am. Chem. Soc. 127 (2005) 8168. [55] J.H. Seo, R. Matsuno, T. Konno, et al., Biomaterials 29 (2008) 1367. [56] W.B. Tsai, J.M. Grunkemeier, C.D. McFarland, et al., J. Biomed. Mater. Res. 60 (2002) 348. [57] Q. Ma, H. Zhang, J. Zhao, et al., Appl. Surf. Sci. 258 (2012) 9711. [58] A. Shahryari, F. Azari, H. Vali, et al., Acta Biomater. 6 (2010) 695. [59] J. Aoki, P.W. Serruys, H. van Beusekom, et al., J. Am. Coll. Cardiol. 45 (2005) 1574. [60] A. Shahryari, S. Omanovic, J.A. Szpunar, J. Biomed. Mater. Res. A 89 (2009) 1049. [61] A. Shahryari, W. Kamal, S. Omanovic, Mater. Lett. 62 (2008) 3906. [62] S. Garg, P.W. Serruys, J. Am. Coll. Cardiol. 56 (2010) S43. [63] S. Garg, P.W. Serruys, J. Am. Coll. Cardiol. 56 (2010) S1. [64] T. Xi, R. Gao, B. Xu, et al., Biomaterials 31 (2010) 5151. [65] C.J. Pan, J.J. Tang, Y.J. Weng, et al., J. Controlled Release 116 (2006) 42. [66] S. Venkatraman, F. Boey, J. Controlled Release 120 (2007) 149. [67] M.A. Hiob, S.G. Wise, A. Kondyurin, et al., Biomaterials 34 (2013) 7584.