- Email: [email protected]
Lysine–poly(2-hydroxyethyl methacrylate) modified polyurethane surface with high lysine density and fibrinolytic activity
Acta Biomaterialia 7 (2011) 954–958
Contents lists available at ScienceDirect
Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat
Lysine–poly(2-hydroxyethyl methacrylate) modified polyurethane surface with high lysine density and fibrinolytic activity Dan Li a,b, Hong Chen a,⇑, Shasha Wang a,b, Zhaoqiang Wu a,c, John L. Brash a,c a
College of Chemistry, Chemical Engineering and Materials Science, Soochow University, 199 Renai Rd., Suzhou 215123, Jiangsu, PR China School of Materials Science and Engineering, Wuhan University of Technology, Wuhan 430070, PR China c School of Biomedical Engineering and Department of Chemical Engineering, McMaster University, Hamilton, Ontario, Canada b
a r t i c l e
i n f o
Article history: Received 21 July 2010 Received in revised form 9 October 2010 Accepted 20 October 2010 Available online 25 October 2010 Keywords: Fibrinolytic surface Polyurethane Poly(2-hydroxyethyl methacrylate) Protein adsorption Graft density
a b s t r a c t We have developed a potentially fibrinolytic surface in which a bioinert polymer is used as a spacer to immobilize lysine such that the e-amino group is free to capture plasminogen when in contact with blood. Adsorbed plasminogen can be activated to plasmin and potentially dissolve nascent clots formed on the surface. In previous work lysine was immobilized through a poly(ethylene glycol) (PEG) spacer; however, the graft density of PEG was limited and the resulting adsorbed quantity of plasminogen was insufficient to dissolve clots efficiently. The aim of the present work was to optimize the surface using graft-polymerized poly(2-hydroxyethyl methacrylate) (poly(HEMA)) as a spacer to increase the grafting density of lysine. Such a poly(HEMA)–lysine modified polyurethane (PU) surface is expected to have increased plasminogen binding capacity and clot lysing efficiency compared with PEG–lysine modified PU. A lysine density of 2.81 nmol cm2 was measured on the PU–poly(HEMA)–Lys surface vs. 0.76 nmol cm2 on a comparable PU–PEG–Lys surface reported previously. The poly(HEMA)–lysine-modified surface was shown to reduce non-specific (fibrinogen) adsorption while binding plasminogen from plasma with high affinity. With increased plasminogen binding capacity these surfaces showed more rapid clot lysis (20 min) in a standard in vitro assay than the corresponding PEG–lysine system (40 min). The data suggest that poly(HEMA) is superior to PEG when used as a spacer in the immobilization of bioactive molecules at high density. This method of modification may also provide a generic approach for preparing bioactive PU surfaces of high activity and low non-specific adsorption of proteins. Ó 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
1. Introduction Surface modification with bioactive agents capable of inhibiting enzymes in the coagulation cascade is a widely used strategy for improving the blood compatibility of a biomaterial [1]. Polyethylene glycol (PEG) has been used as a spacer to couple these bioactive moieties to surfaces because of its excellent protein resistance [2,3]. We have developed the concept of a fibrinolytic surface in which PEG is used as a spacer to immobilize lysine such that the e-amino group is free to capture plasminogen and tissue type plasminogen activator (t-PA) when in contact with blood and dissolve nascent clots formed on the surface [4–6]. However, the surface density of PEG achievable by ‘‘grafting’’ is limited due to steric hindrance [7], and the density of terminally conjugated bioactive molecules is correspondingly limited. An alternative approach is to generate the spacer via surfaceinitiated polymerization (SIP) [7]. SIP is well known to generate much denser polymer layers, and, more importantly, if the surface
⇑ Corresponding author. Tel./fax: +86 512 65880827. E-mail address: [email protected] (H. Chen).
grafted polymers have abundant side-chains with active chain ends this permits the generation of a high concentration of chemically active sites on the surface to bind bioactive molecules. Of the various monomers available to form such grafts, poly(2-hydroxyethyl methacrylate) (poly(HEMA)) has found the most widespread use. Due to its excellent biocompatibility and physical properties similar to those of living tissues, poly(HEMA) has been extensively studied for applications in tissue engineering [8], drug delivery [9], antifouling materials [10] and biosensors [11]. For the purposes of generating non-fouling bioactive surfaces, HEMA has been graft polymerized on various substrates, with attachment of bioactive moieties via reaction with the hydroxyl groups in the side chains. For example, Xu et al. prepared poly(HEMA) modified silicon surfaces by surface-initiated atom transfer radical polymerization (SI-ATRP) and coupled collagen to the pendant hydroxyl groups to control the adhesion of 3T3 fibroblasts [12]. Hu et al. developed a versatile method of constructing glycosylated membrane surfaces in which poly(HEMA) was polymerized on microporous polypropylene membranes (MPPMs) and then conjugated with glucose pentaacetate. The glycosylated membranes showed a high affinity for concanavalin A protein while effectively preventing non-specific protein adsorption [13,14].
1742-7061/$ - see front matter Ó 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.actbio.2010.10.021
D. Li et al. / Acta Biomaterialia 7 (2011) 954–958
Although PEG is generally believed to be the most effective protein repellent polymer and is widely used in surface modification of various biomaterials, it has been shown to be prone to oxidation in the presence of oxygen and transition metal ions. Its long-term stability is thus in question [15]. In contrast, poly(HEMA), with high chemical and hydrolytic stability [16], may be more suitable for longterm applications such as medical implants. The aim of the present work was to optimize the properties of a fibrinolytic polyurethane (PU) surface using graft-polymerized poly(HEMA) as a spacer to prevent non-specific protein adsorption and increase the grafting density of lysine. It is expected that a high density of lysine should result in increased adsorption of plasminogen from plasma and increased fibrinolytic potential. HEMA was polymerized on a vinyl-functionalized PU surface [17] and lysine was then coupled to the hydroxyl groups of the tethered poly(HEMA) such that the e-amino group was free. Such a poly(HEMA)–lysine modified PU surface was expected to be rich in lysine and to have increased plasminogen binding capacity and clot lysing efficiency compared with PEG–lysine modified PU. This method of modification may provide a generic approach for preparing bioactive PU surfaces of high activity and low non-specific adsorption of proteins. 2. Materials and methods 2.1. Materials N,N0 -Disuccinimidyl carbonate (DSC) (anhydrous, P95% pure), trifluoroacetic acid (TFA), H-Lys(t-BOC)-OH and 4-nitrobenzaldehyde were from Sigma–Aldrich Chemical Co. and used without further purification. 2-Hydroxyethyl methacrylate (HEMA) (Acros, 98% pure) was distilled under reduced pressure to remove inhibitors. 2,20 -Azoisobutyronitrile (AIBN) was recrystallized from methanol solution and dried under vacuum prior to use. Fibrinogen (plasminogen-free) was from Calbiochem (La Jolla, CA). Recombinant tissue plasminogen activator (t-PA) was from Genentech (San Francisco, CA). N,N-Dimethylformamide (DMF), acetonitrile, triethylamine (TEA) and methanol were from the Shanghai Chemical Reagent Co. and purified before use.
955
DSC (0.05 mmol ml1) and TEA (0.05 mmol ml1) and stirred at room temperature for 6 h. The resulting surfaces (PU–pHEMA– NHS) were incubated overnight in phosphate-buffered saline (PBS), pH 7.4 containing 5 mg ml1 H-Lys(t-BOC)-OH. The surfaces with e-NH-t-Boc groups were deprotected by treatment with 25% TFA for 90 min and subsequently washed with PBS. The resulting surface, with the e-NH2 groups of lysine exposed (PU–pHEMA– Lys), is of the greatest interest in terms of its fibrinolytic potential. 2.3. Lysine density The lysine graft densities were determined by reacting the surface amino groups with 4-nitrobenzaldehyde to form imines and subsequent hydrolysis to liberate 4-nitrobenzaldehyde [19]. Typically, six discs of PU–pHEMA–Lys or precursor PU–pHEMA surface were immersed in anhydrous ethanol (10 ml) containing 4-nitrobenzaldehyde (40 mg) and acetic acid (0.008 ml) under nitrogen at 50 °C for 3 h. The surfaces were then washed and sonicated in absolute ethanol for 2 min. The discs were immersed in water (1 ml) containing acetic acid (0.002 ml) and the solution was kept at 40 °C for 3 h. The hydrolysis was run in triplicate (three groups of three discs). The 4-nitrobenzaldehyde liberated, equivalent to the surface amino content, was determined by measuring absorbance at 268.5 nm. 2.4. Protein adsorption Fibrinogen was labeled with 125I (ICN Pharmaceuticals, Irvine, CA) using the iodine monochloride (ICl) method. The product was passed through an AG 1-X4 column (Bio-Rad Laboratories, Hercules, CA) to remove free iodide. Labeled fibrinogen was mixed with unlabeled fibrinogen (1:19, labeled:unlabeled) in PBS at a total concentration of 1 mg ml1. Surfaces were incubated with the protein solution for 3 h at room temperature, rinsed three times (10 min each time) with PBS, wicked onto filter paper and transferred to clean tubes for radioactivity determination using a Wizard 300 1480 Automatic Gamma Counter (PerkinElmer Life Sciences, Shelton, CT). Protein adsorption was expressed as moles per unit surface area. 2.5. Western blotting
2.2. Preparation of surfaces PU (TecothaneÒ TT-1095A, Thermedics, Wilmington, MA) was extracted (Soxhlet) for 48 h with methanol to remove impurities. Films of this material were cast from a 5% (w/v) solution in DMF, dried in air at 75 °C for 48 h and vacuum dried at 60 °C for 48 h to remove solvent. Discs 5 mm in diameter and 0.5 mm thick were punched from the PU elastomer films. Methacryloyl isothiocyanate (MI) monomer was synthesized as reported previously [18]. The PU discs were immersed in 25 ml of an acetonitrile solution containing 8 mmol MI. After stirring at 65 °C for 12 h the vinyl group functionalized PU surface (VPU) obtained was washed with acetonitrile and dried in a vacuum oven at 40 °C for 24 h. Radical graft polymerization on the VPU surface was performed using AIBN as the initiator. Briefly, HEMA (0.65 g, 5.0 mmol) and AIBN (0.0082 g, 0.05 mmol) were dissolved in 4 ml of anhydrous methanol. VPU discs were immersed in the methanol solution. The solution was degassed by bubbling with nitrogen for 20 min. The grafting reaction was carried out at 65 °C for 12 h. The poly(HEMA) grafted PU surface was washed successively with methanol, distilled water and methanol and dried in a vacuum oven at 40 °C for 24 h. For the covalent conjugation of e-lysine the poly(HEMA) grafted PU surfaces were first added to an acetonitrile solution containing
Surfaces were incubated in 100% pooled normal human plasma (PNP) (from blood bank) for 3 h at room temperature. Adsorbed proteins were eluted using 2% aqueous sodium dodecyl sulfate (SDS). Polyacrylamide gel electrophoresis and immunoblotting were performed as described previously [20]. Briefly, the eluates were run on 12% reduced SDS–PAGE gels to separate the proteins according to molecular weight. The proteins were then transferred from the gel onto an Immobilon PVDF membrane (Millipore, Bedford, MA). The membranes were blocked with 5% non-fat dry milk and incubated with primary antibodies (dilution of 1:1000) to plasminogen or fibrinogen and then with an alkaline phosphatase-conjugated secondary antibody (dilution of 1:1000). The substrate system used to develop a color reaction for alkaline phosphatase was 5-bromo-4-chloro-3-indolyl phosphate (BCIP) and nitroblue tetrazolium (NBT) (both from Bio-Rad), prepared as described by the supplier. 2.6. Plasma clot lysis Surfaces were incubated in PNP for 3 h at room temperature in microtitre plate wells. The films were removed from the wells, rinsed three times in Tris-buffered saline (TBS), pH 7.4 and then placed in clean wells. t-PA was added to the wells at a concentration of 0.1 mg ml1 in TBS and incubated for 30 min at room
956
D. Li et al. / Acta Biomaterialia 7 (2011) 954–958
temperature. The films were rinsed extensively with buffer to remove any unbound proteins. To the extent that plasminogen is adsorbed from plasma, this procedure provides surfaces bearing a layer of bound plasmin. The clot lysing potential of the surfaces with adsorbed plasmin was assessed using a modified plasma recalcification assay. 100 ll of PNP was added to the wells containing the surfaces. Following 5 min equilibration at 37 °C an equal volume of 0.025 M CaCl2 was injected into the wells. Absorbance at 405 nm was measured at 30 s intervals over a 1 h period. 3. Results and discussion 3.1. Preparation of poly(HEMA)–lysine-modified PU surfaces The poly(HEMA)–lysine modified PU surfaces were prepared as outlined in Scheme 1. MI was used as a functional monomer that could be bound covalently to the PU surface through reaction of its NCS groups with NH groups in the PU. Poly(HEMA) was then grown on the surface-bound vinyl groups by free radical polymerization. These reactions were confirmed (X-ray photoelectron spectroscopy) in a previous study [17]. The poly(HEMA) grafted surface contains a high concentration of hydroxyl groups, which provide reactive sites for the attachment of lysine to give surfaces of high lysine density. DSC was chosen as a bioconjugate reagent to link lysine to these tethered hydroxyl groups. Lysine was conjugated through its a-amino groups using e-amino protected lysine. The protective groups were removed by hydrolysis after grafting. This procedure is required since lysine exhibits an affinity for plasminogen and t-PA only when its e-NH2 and COOH groups are free [21]. Such a poly(HEMA)–lysine modified PU surface was expected to be superior in plasminogen binding capacity to PEG–lysinemodified surfaces reported previously. 3.2. Lysine density The lysine density is an important parameter that determines the extent of plasminogen binding. Surface immobilized lysine was determined by reacting the e-NH2 groups of lysine on the surface with 4-nitrobenzaldehyde to form a Schiff base. The 4-nitrobenzaldehyde liberated by hydrolysis of the Schiff base is measured [19] by absorbance at 268.5 nm. The lysine density
was found to be 2.81 nmol cm2 on the PU–pHEMA–Lys surface, much higher than on a comparable PU–PEG–Lys surface (0.76 nmol cm2) determined using the same method [6]. This is presumably due to the higher density of hydroxyl groups available for reaction with lysine. The lysine density is comparable with that reported in previous studies in which the lysine-containing surfaces were prepared by a photochemical method; the lysine density ranged from 0.2 to 3.2 nmol cm2 [21]. The photochemical method, however, is difficult to apply to complex shapes, such as the inner surface of a small diameter vascular graft. The high density of amino groups present on the PU–pHEMA–Lys surface suggests a correspondingly high density of lysine. It should be noted that the values of lysine density reported for these surfaces are based on the nominal surface area, which is likely to be less than the effective area. Densities of 3 nmol cm2 imply areas per lysine of the order of 6 Å2, i.e. unrealistically small. However, it is expected that the pHEMA layer to which the lysine is attached has a significant thickness dimension. Thus the effective area will be greater than the nominal and the true lysine densities will be smaller than the reported ones.
3.3. Fibrinogen adsorption from buffer Fibrinogen is a key protein of the coagulation cascade and plays a leading role in mediating platelet adhesion to biomaterials. In this research fibrinogen adsorption was measured to provide an indication of the protein resistance of the modified surfaces. As shown in Fig. 1, the unmodified PU surface showed the highest level of adsorption, 0.91 lg cm2, while only 0.1 lg cm2 fibrinogen was adsorbed on the poly(HEMA) modified surface, a decrease of 90%. An optimized PU–PEG surface was shown previously to decrease fibrinogen adsorption by 95%, i.e. only slightly more than the PU–pHEMA surface in the present work. The protein resistant properties of poly(HEMA) may be attributed, at least in part, to its hydrophilicity, which is generally recognized as essential for protein resistance. Hu et al. prepared polypropylene microporous membranes grafted with poly(HEMA) and found that with increasing grafting density the water contact angle decreased and the protein resistance increased [10]. Furthermore, Yoshikawa et al. reported that both high (0.7 chains nm2) and medium density (0.06 chains cm2) poly(HEMA) brushes could resist larger proteins
Scheme 1. Schematic illustrating the processes of surface modification.
D. Li et al. / Acta Biomaterialia 7 (2011) 954–958
957
Fig. 1. Fibrinogen adsorption from buffer (3 h exposure) on modified and unmodified surfaces (mean ± SD, n = 3).
while low density brushes (0.007 chains nm2) showed poor resistance to protein adsorption [22]. They concluded that the size exclusion effect associated with dense poly(HEMA) brushes plays an important role in suppressing protein adsorption. In the present study poly(HEMA) was grafted by free radical polymerization giving rise to a distribution of chain lengths with some chains of high molecular weight. Thus so-called ‘‘primary’’ adsorption [23], by which a protein diffuses through the grafted layer and adsorbs on the substrate, was apparently suppressed. The interaction of proteins with the NHS functionalized surface may involve covalent binding, as suggested previously based on resistance to elution by SDS [4]. As seen in Fig. 1, fibrinogen adsorption was much higher on the PU–pHEMA–NHS surface than on PU–pHEMA, presumably due to the NHS groups on the surface. Furthermore, the amount adsorbed on the PU–pHEMA–NHS surface was almost twice as great as on the PU–PEG–NHS surface reported previously [4], again suggesting that poly(HEMA) may be superior to PEG in terms of achievable graft density of bioactive molecules when used as a spacer. The poly(HEMA)–Lys surface was also fibrinogen resistant, although less so than the poly(HEMA). The slight loss of protein resistance is presumably due to the decreased concentration of hydroxyl groups on the surface and to electrostatic interactions between the partially protonated amino groups and the negatively charged fibrinogen. 3.4. Western blotting Plasminogen is of the greatest importance in this study because of its central role in fibrinolysis. Plasminogen contains five ternary loop structures known as ‘‘kringles’’, two of which (K1 and K4) contain a lysine binding site (LBS). It is through the LBSs that plasminogen binds with high affinity to exposed C-terminal lysine residues in partially degraded fibrin [24]. In this study the immobilized lysine was in the same form as C-terminal lysine residues in fibrin and is thus expected to have high affinity for plasminogen. Immunoblots of proteins eluted from the surfaces following a 3 h exposure to human plasma were probed with antibodies directed against plasminogen and fibrinogen, and are shown in Fig. 2. It should be emphasized that the intensity of a band should be approximately proportional to the adsorbed quantity of the corresponding protein, so for a given protein adsorption may be compared semi-quantitatively from one surface to another. As expected, only the PU–pHEMA–Lys blot showed clear evidence of plasminogen adsorption, with a strong band at 94 kDa, corresponding to the intact form of plasminogen, while for the other
Fig. 2. Immunoblots of eluates from modified and unmodified surfaces after exposure to 100% plasma for 3 h, probed with antibodies against plasminogen and fibrinogen.
two surfaces there was no detectable signal for plasminogen. Thus only the lysine-containing surface has affinity for plasminogen. The intensity of the bands on the fibrinogen blots showed trends similar to those seen in the single protein adsorption data (Fig. 1). Strong bands were detected at about 68, 56 and 48 kDa on the PU blot, corresponding to the a-, b- and c-chains, respectively. Only weak responses were observed in the blots for the PU–pHEMA and PU–pHEMA–Lys surfaces. The quantity of fibrinogen adsorbed from plasma on these two surfaces appears to be very low, presumably due their inherent protein resistance and to competition from other proteins. The trace amounts of fibrinogen adsorbed on the latter two surfaces may not be sufficient to support significant platelet adhesion. 3.5. Plasma clot lysis Plasminogen is the principal zymogen of the fibrinolytic pathway. Proteolytic cleavage of the Arg560–Val561 bond in plasminogen by t-PA transforms the single chain molecule into its two chain enzymatically active form plasmin, which degrades fibrin clots [24]. To assess the clot dissolving properties of the lysine-modified surface a plasma clotting assay based on turbidity measurements was used, and typical data are shown in Fig. 3. The onset of coagulation is indicated by a steep rise in the absorbance vs. time curve following recalcification of the plasma. The PU and PU–pHEMA surfaces showed typical clot formation curves: a plateau in absorbance was reached and maintained, indicating a fully formed, stable clot. In contrast, for the PU–pHEMA–Lys surface the absorbance returned to baseline, indicating that the clot formed and then was lysed by the action of surface localized plasmin. It is important to note that clot lysis was quite rapid and was complete within 20 min, while for a comparable PU–PEG–Lys surface reported previously 40 min was required for the same effect [4]. This difference may be attributed to the increased lysine density on the PU–pHEMA–Lys surface and thus the enhancement of plasminogen binding capacity. It should be noted that clot dissolution by this lysine derivatized surface requires the availability of t-PA or other plasminogen activator. The normal content of t-PA in plasma is very low (6 lg l1),
958
D. Li et al. / Acta Biomaterialia 7 (2011) 954–958
Fig. 3. Clot formation in plasma expressed as absorbance at 405 nm vs. time for modified and unmodified PU surfaces.
and spontaneously adsorbed quantities are likely to be insufficient to cause significant activation of adsorbed plasminogen. It does seem possible that in the case of implants within the vascular system t-PA from damaged endothelium could be available to activate adsorbed plasminogen. However, the ultimate design for clot dissolving properties will most likely include provision of a source of activator, perhaps by controlled release. 4. Summary and conclusions PU surfaces having high lysine density and fibrinolytic activity when exposed to plasma followed by t-PA were prepared by graft polymerization of HEMA and subsequent reaction with e-aminofree lysine. Due to the higher density of hydroxyl groups available for reaction with lysine, a lysine density of 2.81 nmol cm2 was achieved, much higher than for a comparable PU–PEG–Lys surface (0.76 nmol cm2) reported previously. Radiolabelled protein adsorption and immunoblotting experiments showed that the modified surfaces were fibrinogen-resistant and only the lysinecontaining surface adsorbed plasminogen from plasma. When activated by t-PA the poly(HEMA)–lysine-modified surface showed more rapid clot lysis (20 min) than the corresponding PEG–lysine system (40 min), attributed to the increase in lysine density and thus the enhancement of plasminogen binding capacity. These results indicate that poly(HEMA) is superior to PEG when used as a spacer in the immobilization of bioactive molecules at high density. This method of modification also provides a generic approach for preparing bioactive PU surfaces of high activity. Acknowledgements This work was supported by the Canadian Institutes of Health Research, the Natural Sciences and Engineering Research Council of Canada, and the National Natural Science Foundation of China (20634030, 20920102035). Appendix A. Figures with essential colour discrimination Certain figures in this article, particularly Figure 2 is difficult to interpret in black and white. The full colour images can be found in the on-line version, at doi:10.1016/j.actbio.2010.10.021 References [1] Chen H, Yuan L, Song W, Wu Z, Li D. Biocompatible polymer materials: role of protein–surface interactions. Prog Polym Sci 2008;33:1059–87.
[2] Chen H, Chen Y, Sheardown H, Brook MA. Immobilization of heparin on a silicone surface through a heterobifunctional PEG spacer. Biomaterials 2005;26:7418–24. [3] Chen H, Brook MA, Sheardown HD, Chen Y, Klenkler B. Generic bioaffinity silicone surfaces. Bioconjug Chem 2006;17:21–8. [4] Chen H, Zhang Y, Li D, Hu X, Wang L, McClung WG, et al. Surfaces having dual fibrinolytic and protein resistant properties by immobilization of lysine on polyurethane through a PEG spacer. J Biomed Mater Res 2009;90A:940–6. [5] Chen H, Wang L, Zhang Y, Li D, McClung WG, Brook MA, et al. Fibrinolytic poly(dimethyl siloxane) surfaces. Macromol Biosci 2008;8:863–70. [6] Li D, Chen H, McClung WG, Brash JL. Lysine–PEG-modified polyurethane as a fibrinolytic surface. effect of PEG chain length on protein interactions, platelet interactions and clot lysis. Acta Biomater. 2009;5:1864–71. [7] Xu FJ, Neoh KG, Kang ET. Bioactive surfaces and biomaterials via atom transfer radical polymerization. Prog Polym Sci 2009;34:719–61. [8] Lahooti S, Sefton MV. Microencapsulation of normal and transfected L929 fibroblasts in a HEMA–MMA copolymer. Tissue Eng 2000;6:139–49. [9] He HY, Cao X, Lee LJ. Design of a novel hydrogel-based intelligent system for controlled drug release. J Control Release 2004;95:391–402. [10] Hu MX, Yang Q, Xu ZK. Enhancing the hydrophilicity of polypropylene microporous membranes by the grafting of 2-hydroxyethyl methacrylate via a synergistic effect of photoinitiators. J Membr Sci 2006;285:196–205. [11] Yu BZ, Wang CY, Ju YM, West L, Harmon J, Moussy Y, et al. Use of hydrogel coating to improve the performance of implanted glucose sensors. Biosens Bioelectron 2008;23:1278–84. [12] Xu FJ, Zhong SP, Yung LYL, Tong YW, Kang ET, Neoh KG. Collagen-coupled poly(2-hydroxyethyl methacrylate)–Si(111) hybrid surfaces for cell immobilization. Tissue Eng 2005;11:1736–48. [13] Hu MX, Wan LS, Liu ZM, Dai ZW, Xu ZK. Fabrication of glycosylated surfaces on microporous polypropylene membranes for protein recognition and adsorption. J Mater Chem 2008;18:4663–9. [14] Hu MX, Wan LS, Xu ZK. Multilayer adsorption of lectins on glycosylated microporous polypropylene membranes. J Membr Sci 2009;335:111–7. [15] Ostuni E, Chapman RG, Holmlin RE, Takayama S, Whitesides GM. A survey of structure–property relationships of surfaces that resist the adsorption of protein. Langmuir 2001;17:5605–20. [16] Kalal J. The use of methacrylic polymers in medicine. Makromol Chem Suppl 1984;7:31–9. [17] Wu Z, Chen H, Huang H, Zhao T, Liu X, Li D, et al. A facile approach to modify polyurethane surfaces for biomaterial applications. Macromol Biosci 2009;9:1165–8. [18] Wu ZQ, Meng LZ. Novel amphiphilic fluorescent graft copolymers: synthesis, characterization, and potential as gene carriers. Polym Adv Technol 2007;18:853–60. [19] Moon LH, Kim JH, Kim K-j, Kang T-H, Kim B, Kim C-HK, et al. Absolute surface density of the amine group of the aminosilylated thin layers: ultraviolet– visible spectroscopy, second harmonic generation, and synchrotron-radiation photoelectron spectroscopy study. Langmuir 1997;13:4305–10. [20] Mulzer SR, Brash JL. Identification of plasma-proteins adsorbed to hemodialyzers during clinical use. J Biomed Mater Res 1989;23:1483–504. [21] McClung WG, Clapper DL, Hu SP, Brash JL. Adsorption of plasminogen from human plasma to lysine-containing surfaces. J Biomed Mater Res 2000;49:409–14. [22] Yoshikawa C, Goto A, Tsujii Y, Fukuda T, Kimura T, Yamamoto K, et al. Protein repellency of well-defined, concentrated poly(2-hydroxyethyl methacrylate) brushes by the size-exclusion effect. Macromolecules 2006;39:2284–90. [23] Currie EPK, Norde W, Stuart MAC. Tethered polymer chains: surface chemistry and their impact on colloidal and surface properties. Adv Colloid Interface Sci 2003;100:205–65. [24] Anglescano E. Overview on fibrinolysis–plasminogen activation pathways on fibrin and cell-surfaces. Chem Phys Lipids 1994;67–68:353–62.
Contents lists available at ScienceDirect
Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat
Lysine–poly(2-hydroxyethyl methacrylate) modified polyurethane surface with high lysine density and fibrinolytic activity Dan Li a,b, Hong Chen a,⇑, Shasha Wang a,b, Zhaoqiang Wu a,c, John L. Brash a,c a
College of Chemistry, Chemical Engineering and Materials Science, Soochow University, 199 Renai Rd., Suzhou 215123, Jiangsu, PR China School of Materials Science and Engineering, Wuhan University of Technology, Wuhan 430070, PR China c School of Biomedical Engineering and Department of Chemical Engineering, McMaster University, Hamilton, Ontario, Canada b
a r t i c l e
i n f o
Article history: Received 21 July 2010 Received in revised form 9 October 2010 Accepted 20 October 2010 Available online 25 October 2010 Keywords: Fibrinolytic surface Polyurethane Poly(2-hydroxyethyl methacrylate) Protein adsorption Graft density
a b s t r a c t We have developed a potentially fibrinolytic surface in which a bioinert polymer is used as a spacer to immobilize lysine such that the e-amino group is free to capture plasminogen when in contact with blood. Adsorbed plasminogen can be activated to plasmin and potentially dissolve nascent clots formed on the surface. In previous work lysine was immobilized through a poly(ethylene glycol) (PEG) spacer; however, the graft density of PEG was limited and the resulting adsorbed quantity of plasminogen was insufficient to dissolve clots efficiently. The aim of the present work was to optimize the surface using graft-polymerized poly(2-hydroxyethyl methacrylate) (poly(HEMA)) as a spacer to increase the grafting density of lysine. Such a poly(HEMA)–lysine modified polyurethane (PU) surface is expected to have increased plasminogen binding capacity and clot lysing efficiency compared with PEG–lysine modified PU. A lysine density of 2.81 nmol cm2 was measured on the PU–poly(HEMA)–Lys surface vs. 0.76 nmol cm2 on a comparable PU–PEG–Lys surface reported previously. The poly(HEMA)–lysine-modified surface was shown to reduce non-specific (fibrinogen) adsorption while binding plasminogen from plasma with high affinity. With increased plasminogen binding capacity these surfaces showed more rapid clot lysis (20 min) in a standard in vitro assay than the corresponding PEG–lysine system (40 min). The data suggest that poly(HEMA) is superior to PEG when used as a spacer in the immobilization of bioactive molecules at high density. This method of modification may also provide a generic approach for preparing bioactive PU surfaces of high activity and low non-specific adsorption of proteins. Ó 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
1. Introduction Surface modification with bioactive agents capable of inhibiting enzymes in the coagulation cascade is a widely used strategy for improving the blood compatibility of a biomaterial [1]. Polyethylene glycol (PEG) has been used as a spacer to couple these bioactive moieties to surfaces because of its excellent protein resistance [2,3]. We have developed the concept of a fibrinolytic surface in which PEG is used as a spacer to immobilize lysine such that the e-amino group is free to capture plasminogen and tissue type plasminogen activator (t-PA) when in contact with blood and dissolve nascent clots formed on the surface [4–6]. However, the surface density of PEG achievable by ‘‘grafting’’ is limited due to steric hindrance [7], and the density of terminally conjugated bioactive molecules is correspondingly limited. An alternative approach is to generate the spacer via surfaceinitiated polymerization (SIP) [7]. SIP is well known to generate much denser polymer layers, and, more importantly, if the surface
⇑ Corresponding author. Tel./fax: +86 512 65880827. E-mail address: [email protected] (H. Chen).
grafted polymers have abundant side-chains with active chain ends this permits the generation of a high concentration of chemically active sites on the surface to bind bioactive molecules. Of the various monomers available to form such grafts, poly(2-hydroxyethyl methacrylate) (poly(HEMA)) has found the most widespread use. Due to its excellent biocompatibility and physical properties similar to those of living tissues, poly(HEMA) has been extensively studied for applications in tissue engineering [8], drug delivery [9], antifouling materials [10] and biosensors [11]. For the purposes of generating non-fouling bioactive surfaces, HEMA has been graft polymerized on various substrates, with attachment of bioactive moieties via reaction with the hydroxyl groups in the side chains. For example, Xu et al. prepared poly(HEMA) modified silicon surfaces by surface-initiated atom transfer radical polymerization (SI-ATRP) and coupled collagen to the pendant hydroxyl groups to control the adhesion of 3T3 fibroblasts [12]. Hu et al. developed a versatile method of constructing glycosylated membrane surfaces in which poly(HEMA) was polymerized on microporous polypropylene membranes (MPPMs) and then conjugated with glucose pentaacetate. The glycosylated membranes showed a high affinity for concanavalin A protein while effectively preventing non-specific protein adsorption [13,14].
1742-7061/$ - see front matter Ó 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.actbio.2010.10.021
D. Li et al. / Acta Biomaterialia 7 (2011) 954–958
Although PEG is generally believed to be the most effective protein repellent polymer and is widely used in surface modification of various biomaterials, it has been shown to be prone to oxidation in the presence of oxygen and transition metal ions. Its long-term stability is thus in question [15]. In contrast, poly(HEMA), with high chemical and hydrolytic stability [16], may be more suitable for longterm applications such as medical implants. The aim of the present work was to optimize the properties of a fibrinolytic polyurethane (PU) surface using graft-polymerized poly(HEMA) as a spacer to prevent non-specific protein adsorption and increase the grafting density of lysine. It is expected that a high density of lysine should result in increased adsorption of plasminogen from plasma and increased fibrinolytic potential. HEMA was polymerized on a vinyl-functionalized PU surface [17] and lysine was then coupled to the hydroxyl groups of the tethered poly(HEMA) such that the e-amino group was free. Such a poly(HEMA)–lysine modified PU surface was expected to be rich in lysine and to have increased plasminogen binding capacity and clot lysing efficiency compared with PEG–lysine modified PU. This method of modification may provide a generic approach for preparing bioactive PU surfaces of high activity and low non-specific adsorption of proteins. 2. Materials and methods 2.1. Materials N,N0 -Disuccinimidyl carbonate (DSC) (anhydrous, P95% pure), trifluoroacetic acid (TFA), H-Lys(t-BOC)-OH and 4-nitrobenzaldehyde were from Sigma–Aldrich Chemical Co. and used without further purification. 2-Hydroxyethyl methacrylate (HEMA) (Acros, 98% pure) was distilled under reduced pressure to remove inhibitors. 2,20 -Azoisobutyronitrile (AIBN) was recrystallized from methanol solution and dried under vacuum prior to use. Fibrinogen (plasminogen-free) was from Calbiochem (La Jolla, CA). Recombinant tissue plasminogen activator (t-PA) was from Genentech (San Francisco, CA). N,N-Dimethylformamide (DMF), acetonitrile, triethylamine (TEA) and methanol were from the Shanghai Chemical Reagent Co. and purified before use.
955
DSC (0.05 mmol ml1) and TEA (0.05 mmol ml1) and stirred at room temperature for 6 h. The resulting surfaces (PU–pHEMA– NHS) were incubated overnight in phosphate-buffered saline (PBS), pH 7.4 containing 5 mg ml1 H-Lys(t-BOC)-OH. The surfaces with e-NH-t-Boc groups were deprotected by treatment with 25% TFA for 90 min and subsequently washed with PBS. The resulting surface, with the e-NH2 groups of lysine exposed (PU–pHEMA– Lys), is of the greatest interest in terms of its fibrinolytic potential. 2.3. Lysine density The lysine graft densities were determined by reacting the surface amino groups with 4-nitrobenzaldehyde to form imines and subsequent hydrolysis to liberate 4-nitrobenzaldehyde [19]. Typically, six discs of PU–pHEMA–Lys or precursor PU–pHEMA surface were immersed in anhydrous ethanol (10 ml) containing 4-nitrobenzaldehyde (40 mg) and acetic acid (0.008 ml) under nitrogen at 50 °C for 3 h. The surfaces were then washed and sonicated in absolute ethanol for 2 min. The discs were immersed in water (1 ml) containing acetic acid (0.002 ml) and the solution was kept at 40 °C for 3 h. The hydrolysis was run in triplicate (three groups of three discs). The 4-nitrobenzaldehyde liberated, equivalent to the surface amino content, was determined by measuring absorbance at 268.5 nm. 2.4. Protein adsorption Fibrinogen was labeled with 125I (ICN Pharmaceuticals, Irvine, CA) using the iodine monochloride (ICl) method. The product was passed through an AG 1-X4 column (Bio-Rad Laboratories, Hercules, CA) to remove free iodide. Labeled fibrinogen was mixed with unlabeled fibrinogen (1:19, labeled:unlabeled) in PBS at a total concentration of 1 mg ml1. Surfaces were incubated with the protein solution for 3 h at room temperature, rinsed three times (10 min each time) with PBS, wicked onto filter paper and transferred to clean tubes for radioactivity determination using a Wizard 300 1480 Automatic Gamma Counter (PerkinElmer Life Sciences, Shelton, CT). Protein adsorption was expressed as moles per unit surface area. 2.5. Western blotting
2.2. Preparation of surfaces PU (TecothaneÒ TT-1095A, Thermedics, Wilmington, MA) was extracted (Soxhlet) for 48 h with methanol to remove impurities. Films of this material were cast from a 5% (w/v) solution in DMF, dried in air at 75 °C for 48 h and vacuum dried at 60 °C for 48 h to remove solvent. Discs 5 mm in diameter and 0.5 mm thick were punched from the PU elastomer films. Methacryloyl isothiocyanate (MI) monomer was synthesized as reported previously [18]. The PU discs were immersed in 25 ml of an acetonitrile solution containing 8 mmol MI. After stirring at 65 °C for 12 h the vinyl group functionalized PU surface (VPU) obtained was washed with acetonitrile and dried in a vacuum oven at 40 °C for 24 h. Radical graft polymerization on the VPU surface was performed using AIBN as the initiator. Briefly, HEMA (0.65 g, 5.0 mmol) and AIBN (0.0082 g, 0.05 mmol) were dissolved in 4 ml of anhydrous methanol. VPU discs were immersed in the methanol solution. The solution was degassed by bubbling with nitrogen for 20 min. The grafting reaction was carried out at 65 °C for 12 h. The poly(HEMA) grafted PU surface was washed successively with methanol, distilled water and methanol and dried in a vacuum oven at 40 °C for 24 h. For the covalent conjugation of e-lysine the poly(HEMA) grafted PU surfaces were first added to an acetonitrile solution containing
Surfaces were incubated in 100% pooled normal human plasma (PNP) (from blood bank) for 3 h at room temperature. Adsorbed proteins were eluted using 2% aqueous sodium dodecyl sulfate (SDS). Polyacrylamide gel electrophoresis and immunoblotting were performed as described previously [20]. Briefly, the eluates were run on 12% reduced SDS–PAGE gels to separate the proteins according to molecular weight. The proteins were then transferred from the gel onto an Immobilon PVDF membrane (Millipore, Bedford, MA). The membranes were blocked with 5% non-fat dry milk and incubated with primary antibodies (dilution of 1:1000) to plasminogen or fibrinogen and then with an alkaline phosphatase-conjugated secondary antibody (dilution of 1:1000). The substrate system used to develop a color reaction for alkaline phosphatase was 5-bromo-4-chloro-3-indolyl phosphate (BCIP) and nitroblue tetrazolium (NBT) (both from Bio-Rad), prepared as described by the supplier. 2.6. Plasma clot lysis Surfaces were incubated in PNP for 3 h at room temperature in microtitre plate wells. The films were removed from the wells, rinsed three times in Tris-buffered saline (TBS), pH 7.4 and then placed in clean wells. t-PA was added to the wells at a concentration of 0.1 mg ml1 in TBS and incubated for 30 min at room
956
D. Li et al. / Acta Biomaterialia 7 (2011) 954–958
temperature. The films were rinsed extensively with buffer to remove any unbound proteins. To the extent that plasminogen is adsorbed from plasma, this procedure provides surfaces bearing a layer of bound plasmin. The clot lysing potential of the surfaces with adsorbed plasmin was assessed using a modified plasma recalcification assay. 100 ll of PNP was added to the wells containing the surfaces. Following 5 min equilibration at 37 °C an equal volume of 0.025 M CaCl2 was injected into the wells. Absorbance at 405 nm was measured at 30 s intervals over a 1 h period. 3. Results and discussion 3.1. Preparation of poly(HEMA)–lysine-modified PU surfaces The poly(HEMA)–lysine modified PU surfaces were prepared as outlined in Scheme 1. MI was used as a functional monomer that could be bound covalently to the PU surface through reaction of its NCS groups with NH groups in the PU. Poly(HEMA) was then grown on the surface-bound vinyl groups by free radical polymerization. These reactions were confirmed (X-ray photoelectron spectroscopy) in a previous study [17]. The poly(HEMA) grafted surface contains a high concentration of hydroxyl groups, which provide reactive sites for the attachment of lysine to give surfaces of high lysine density. DSC was chosen as a bioconjugate reagent to link lysine to these tethered hydroxyl groups. Lysine was conjugated through its a-amino groups using e-amino protected lysine. The protective groups were removed by hydrolysis after grafting. This procedure is required since lysine exhibits an affinity for plasminogen and t-PA only when its e-NH2 and COOH groups are free [21]. Such a poly(HEMA)–lysine modified PU surface was expected to be superior in plasminogen binding capacity to PEG–lysinemodified surfaces reported previously. 3.2. Lysine density The lysine density is an important parameter that determines the extent of plasminogen binding. Surface immobilized lysine was determined by reacting the e-NH2 groups of lysine on the surface with 4-nitrobenzaldehyde to form a Schiff base. The 4-nitrobenzaldehyde liberated by hydrolysis of the Schiff base is measured [19] by absorbance at 268.5 nm. The lysine density
was found to be 2.81 nmol cm2 on the PU–pHEMA–Lys surface, much higher than on a comparable PU–PEG–Lys surface (0.76 nmol cm2) determined using the same method [6]. This is presumably due to the higher density of hydroxyl groups available for reaction with lysine. The lysine density is comparable with that reported in previous studies in which the lysine-containing surfaces were prepared by a photochemical method; the lysine density ranged from 0.2 to 3.2 nmol cm2 [21]. The photochemical method, however, is difficult to apply to complex shapes, such as the inner surface of a small diameter vascular graft. The high density of amino groups present on the PU–pHEMA–Lys surface suggests a correspondingly high density of lysine. It should be noted that the values of lysine density reported for these surfaces are based on the nominal surface area, which is likely to be less than the effective area. Densities of 3 nmol cm2 imply areas per lysine of the order of 6 Å2, i.e. unrealistically small. However, it is expected that the pHEMA layer to which the lysine is attached has a significant thickness dimension. Thus the effective area will be greater than the nominal and the true lysine densities will be smaller than the reported ones.
3.3. Fibrinogen adsorption from buffer Fibrinogen is a key protein of the coagulation cascade and plays a leading role in mediating platelet adhesion to biomaterials. In this research fibrinogen adsorption was measured to provide an indication of the protein resistance of the modified surfaces. As shown in Fig. 1, the unmodified PU surface showed the highest level of adsorption, 0.91 lg cm2, while only 0.1 lg cm2 fibrinogen was adsorbed on the poly(HEMA) modified surface, a decrease of 90%. An optimized PU–PEG surface was shown previously to decrease fibrinogen adsorption by 95%, i.e. only slightly more than the PU–pHEMA surface in the present work. The protein resistant properties of poly(HEMA) may be attributed, at least in part, to its hydrophilicity, which is generally recognized as essential for protein resistance. Hu et al. prepared polypropylene microporous membranes grafted with poly(HEMA) and found that with increasing grafting density the water contact angle decreased and the protein resistance increased [10]. Furthermore, Yoshikawa et al. reported that both high (0.7 chains nm2) and medium density (0.06 chains cm2) poly(HEMA) brushes could resist larger proteins
Scheme 1. Schematic illustrating the processes of surface modification.
D. Li et al. / Acta Biomaterialia 7 (2011) 954–958
957
Fig. 1. Fibrinogen adsorption from buffer (3 h exposure) on modified and unmodified surfaces (mean ± SD, n = 3).
while low density brushes (0.007 chains nm2) showed poor resistance to protein adsorption [22]. They concluded that the size exclusion effect associated with dense poly(HEMA) brushes plays an important role in suppressing protein adsorption. In the present study poly(HEMA) was grafted by free radical polymerization giving rise to a distribution of chain lengths with some chains of high molecular weight. Thus so-called ‘‘primary’’ adsorption [23], by which a protein diffuses through the grafted layer and adsorbs on the substrate, was apparently suppressed. The interaction of proteins with the NHS functionalized surface may involve covalent binding, as suggested previously based on resistance to elution by SDS [4]. As seen in Fig. 1, fibrinogen adsorption was much higher on the PU–pHEMA–NHS surface than on PU–pHEMA, presumably due to the NHS groups on the surface. Furthermore, the amount adsorbed on the PU–pHEMA–NHS surface was almost twice as great as on the PU–PEG–NHS surface reported previously [4], again suggesting that poly(HEMA) may be superior to PEG in terms of achievable graft density of bioactive molecules when used as a spacer. The poly(HEMA)–Lys surface was also fibrinogen resistant, although less so than the poly(HEMA). The slight loss of protein resistance is presumably due to the decreased concentration of hydroxyl groups on the surface and to electrostatic interactions between the partially protonated amino groups and the negatively charged fibrinogen. 3.4. Western blotting Plasminogen is of the greatest importance in this study because of its central role in fibrinolysis. Plasminogen contains five ternary loop structures known as ‘‘kringles’’, two of which (K1 and K4) contain a lysine binding site (LBS). It is through the LBSs that plasminogen binds with high affinity to exposed C-terminal lysine residues in partially degraded fibrin [24]. In this study the immobilized lysine was in the same form as C-terminal lysine residues in fibrin and is thus expected to have high affinity for plasminogen. Immunoblots of proteins eluted from the surfaces following a 3 h exposure to human plasma were probed with antibodies directed against plasminogen and fibrinogen, and are shown in Fig. 2. It should be emphasized that the intensity of a band should be approximately proportional to the adsorbed quantity of the corresponding protein, so for a given protein adsorption may be compared semi-quantitatively from one surface to another. As expected, only the PU–pHEMA–Lys blot showed clear evidence of plasminogen adsorption, with a strong band at 94 kDa, corresponding to the intact form of plasminogen, while for the other
Fig. 2. Immunoblots of eluates from modified and unmodified surfaces after exposure to 100% plasma for 3 h, probed with antibodies against plasminogen and fibrinogen.
two surfaces there was no detectable signal for plasminogen. Thus only the lysine-containing surface has affinity for plasminogen. The intensity of the bands on the fibrinogen blots showed trends similar to those seen in the single protein adsorption data (Fig. 1). Strong bands were detected at about 68, 56 and 48 kDa on the PU blot, corresponding to the a-, b- and c-chains, respectively. Only weak responses were observed in the blots for the PU–pHEMA and PU–pHEMA–Lys surfaces. The quantity of fibrinogen adsorbed from plasma on these two surfaces appears to be very low, presumably due their inherent protein resistance and to competition from other proteins. The trace amounts of fibrinogen adsorbed on the latter two surfaces may not be sufficient to support significant platelet adhesion. 3.5. Plasma clot lysis Plasminogen is the principal zymogen of the fibrinolytic pathway. Proteolytic cleavage of the Arg560–Val561 bond in plasminogen by t-PA transforms the single chain molecule into its two chain enzymatically active form plasmin, which degrades fibrin clots [24]. To assess the clot dissolving properties of the lysine-modified surface a plasma clotting assay based on turbidity measurements was used, and typical data are shown in Fig. 3. The onset of coagulation is indicated by a steep rise in the absorbance vs. time curve following recalcification of the plasma. The PU and PU–pHEMA surfaces showed typical clot formation curves: a plateau in absorbance was reached and maintained, indicating a fully formed, stable clot. In contrast, for the PU–pHEMA–Lys surface the absorbance returned to baseline, indicating that the clot formed and then was lysed by the action of surface localized plasmin. It is important to note that clot lysis was quite rapid and was complete within 20 min, while for a comparable PU–PEG–Lys surface reported previously 40 min was required for the same effect [4]. This difference may be attributed to the increased lysine density on the PU–pHEMA–Lys surface and thus the enhancement of plasminogen binding capacity. It should be noted that clot dissolution by this lysine derivatized surface requires the availability of t-PA or other plasminogen activator. The normal content of t-PA in plasma is very low (6 lg l1),
958
D. Li et al. / Acta Biomaterialia 7 (2011) 954–958
Fig. 3. Clot formation in plasma expressed as absorbance at 405 nm vs. time for modified and unmodified PU surfaces.
and spontaneously adsorbed quantities are likely to be insufficient to cause significant activation of adsorbed plasminogen. It does seem possible that in the case of implants within the vascular system t-PA from damaged endothelium could be available to activate adsorbed plasminogen. However, the ultimate design for clot dissolving properties will most likely include provision of a source of activator, perhaps by controlled release. 4. Summary and conclusions PU surfaces having high lysine density and fibrinolytic activity when exposed to plasma followed by t-PA were prepared by graft polymerization of HEMA and subsequent reaction with e-aminofree lysine. Due to the higher density of hydroxyl groups available for reaction with lysine, a lysine density of 2.81 nmol cm2 was achieved, much higher than for a comparable PU–PEG–Lys surface (0.76 nmol cm2) reported previously. Radiolabelled protein adsorption and immunoblotting experiments showed that the modified surfaces were fibrinogen-resistant and only the lysinecontaining surface adsorbed plasminogen from plasma. When activated by t-PA the poly(HEMA)–lysine-modified surface showed more rapid clot lysis (20 min) than the corresponding PEG–lysine system (40 min), attributed to the increase in lysine density and thus the enhancement of plasminogen binding capacity. These results indicate that poly(HEMA) is superior to PEG when used as a spacer in the immobilization of bioactive molecules at high density. This method of modification also provides a generic approach for preparing bioactive PU surfaces of high activity. Acknowledgements This work was supported by the Canadian Institutes of Health Research, the Natural Sciences and Engineering Research Council of Canada, and the National Natural Science Foundation of China (20634030, 20920102035). Appendix A. Figures with essential colour discrimination Certain figures in this article, particularly Figure 2 is difficult to interpret in black and white. The full colour images can be found in the on-line version, at doi:10.1016/j.actbio.2010.10.021 References [1] Chen H, Yuan L, Song W, Wu Z, Li D. Biocompatible polymer materials: role of protein–surface interactions. Prog Polym Sci 2008;33:1059–87.
[2] Chen H, Chen Y, Sheardown H, Brook MA. Immobilization of heparin on a silicone surface through a heterobifunctional PEG spacer. Biomaterials 2005;26:7418–24. [3] Chen H, Brook MA, Sheardown HD, Chen Y, Klenkler B. Generic bioaffinity silicone surfaces. Bioconjug Chem 2006;17:21–8. [4] Chen H, Zhang Y, Li D, Hu X, Wang L, McClung WG, et al. Surfaces having dual fibrinolytic and protein resistant properties by immobilization of lysine on polyurethane through a PEG spacer. J Biomed Mater Res 2009;90A:940–6. [5] Chen H, Wang L, Zhang Y, Li D, McClung WG, Brook MA, et al. Fibrinolytic poly(dimethyl siloxane) surfaces. Macromol Biosci 2008;8:863–70. [6] Li D, Chen H, McClung WG, Brash JL. Lysine–PEG-modified polyurethane as a fibrinolytic surface. effect of PEG chain length on protein interactions, platelet interactions and clot lysis. Acta Biomater. 2009;5:1864–71. [7] Xu FJ, Neoh KG, Kang ET. Bioactive surfaces and biomaterials via atom transfer radical polymerization. Prog Polym Sci 2009;34:719–61. [8] Lahooti S, Sefton MV. Microencapsulation of normal and transfected L929 fibroblasts in a HEMA–MMA copolymer. Tissue Eng 2000;6:139–49. [9] He HY, Cao X, Lee LJ. Design of a novel hydrogel-based intelligent system for controlled drug release. J Control Release 2004;95:391–402. [10] Hu MX, Yang Q, Xu ZK. Enhancing the hydrophilicity of polypropylene microporous membranes by the grafting of 2-hydroxyethyl methacrylate via a synergistic effect of photoinitiators. J Membr Sci 2006;285:196–205. [11] Yu BZ, Wang CY, Ju YM, West L, Harmon J, Moussy Y, et al. Use of hydrogel coating to improve the performance of implanted glucose sensors. Biosens Bioelectron 2008;23:1278–84. [12] Xu FJ, Zhong SP, Yung LYL, Tong YW, Kang ET, Neoh KG. Collagen-coupled poly(2-hydroxyethyl methacrylate)–Si(111) hybrid surfaces for cell immobilization. Tissue Eng 2005;11:1736–48. [13] Hu MX, Wan LS, Liu ZM, Dai ZW, Xu ZK. Fabrication of glycosylated surfaces on microporous polypropylene membranes for protein recognition and adsorption. J Mater Chem 2008;18:4663–9. [14] Hu MX, Wan LS, Xu ZK. Multilayer adsorption of lectins on glycosylated microporous polypropylene membranes. J Membr Sci 2009;335:111–7. [15] Ostuni E, Chapman RG, Holmlin RE, Takayama S, Whitesides GM. A survey of structure–property relationships of surfaces that resist the adsorption of protein. Langmuir 2001;17:5605–20. [16] Kalal J. The use of methacrylic polymers in medicine. Makromol Chem Suppl 1984;7:31–9. [17] Wu Z, Chen H, Huang H, Zhao T, Liu X, Li D, et al. A facile approach to modify polyurethane surfaces for biomaterial applications. Macromol Biosci 2009;9:1165–8. [18] Wu ZQ, Meng LZ. Novel amphiphilic fluorescent graft copolymers: synthesis, characterization, and potential as gene carriers. Polym Adv Technol 2007;18:853–60. [19] Moon LH, Kim JH, Kim K-j, Kang T-H, Kim B, Kim C-HK, et al. Absolute surface density of the amine group of the aminosilylated thin layers: ultraviolet– visible spectroscopy, second harmonic generation, and synchrotron-radiation photoelectron spectroscopy study. Langmuir 1997;13:4305–10. [20] Mulzer SR, Brash JL. Identification of plasma-proteins adsorbed to hemodialyzers during clinical use. J Biomed Mater Res 1989;23:1483–504. [21] McClung WG, Clapper DL, Hu SP, Brash JL. Adsorption of plasminogen from human plasma to lysine-containing surfaces. J Biomed Mater Res 2000;49:409–14. [22] Yoshikawa C, Goto A, Tsujii Y, Fukuda T, Kimura T, Yamamoto K, et al. Protein repellency of well-defined, concentrated poly(2-hydroxyethyl methacrylate) brushes by the size-exclusion effect. Macromolecules 2006;39:2284–90. [23] Currie EPK, Norde W, Stuart MAC. Tethered polymer chains: surface chemistry and their impact on colloidal and surface properties. Adv Colloid Interface Sci 2003;100:205–65. [24] Anglescano E. Overview on fibrinolysis–plasminogen activation pathways on fibrin and cell-surfaces. Chem Phys Lipids 1994;67–68:353–62.