Article pubs.acs.org/Langmuir
Regulation of Protein Binding Capability of Surfaces via Host−Guest Interactions: Effects of Localized and Average Ligand Density Xiujuan Shi,†,§ Wenjun Zhan,†,§ Gaojian Chen,*,†,‡ Qian Yu,† Qi Liu,† Hui Du,† Limin Cao,† Xiaoli Liu,† Lin Yuan,† and Hong Chen*,† †
College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123, P. R. China Center for Soft Condensed Matter Physics and Interdisciplinary Research, Soochow University, Suzhou 215006, P. R. China
‡
S Supporting Information *
ABSTRACT: The protein binding capability of biomaterial surfaces can significantly affect subsequent biological responses, and appropriate ligand presentation is often required to guarantee the best functions. Herein, a new facile method for regulating this capability by varying the localized and average ligand density is presented. Binding between lysine and plasminogen relevant to a fibrinolysis system was chosen as a model. We integrated different lysine-modified β-cyclodextrin (βCD) derivatives onto bioinert copolymer brushes via host−guest interactions. The localized and average lysine density can be conveniently modulated by changing the lysine valency on β-CD scaffolds and by diluting lysine-persubstituted β-CD with pure βCD, respectively. Both the plasminogen adsorption and the plasminogen binding affinity were enhanced by lysine-persubstituted β-CD compared with those of lysine-monosubstituted β-CD, which is possibly due to the higher localized lysine density and the multivalent binding of plasminogen on lysine-persubstituted β-CD surfaces. With a change in the ratio of lysine-persubstituted βCD to β-CD, the average lysine density can be tuned, leading to the linear regulation of the adsorption of plasminogen on surfaces.
1. INTRODUCTION The ligand presentation at interfaces produces a significant influence on their interaction with biomolecules, thereby affecting subsequent biological responses.1,2 Modulating ligand presentation on biomolecule-functionalized surfaces, such as ligand density,3 valency,4 type,5 spatial distribution,5,6 spacer length,4 etc., can affect biomolecular recognition (RGD− integrin,5 sugar−lectin,3 hapten−antibody,1 amino acid/short peptide−protein,7 etc.), resulting in the subsequent influence on cell adhesion, proliferation and migration,5 cell−toxin interactions,6 fibrinolytic activity,7 and other biological functions. Appropriate ligand presentation is required to yield the best functions of biomaterials. In some cases, high ligand density was beneficial for the enhancement of biomolecular recognition;7 but a closely packed ligand cluster inhibited ligand−protein binding6,8 and decreased the level of cell targeting of nanoparticles.9 In addition, too many ligands in polymer brushes weakened the ability to capture specific proteins or even attenuated the antifouling performance.10 For these reasons, it is important to modulate the ligand density of biomaterials. However, how to achieve it remains a challenge. Ligand-functionalized surfaces with an antifouling layer can enhance their specific interactions with proteins and therefore are widely used.11−13 Ligands are usually introduced onto the well-defined bioinert polymer brushes by a covalent postmodification method,11,12 and ligand density on this kind of © XXXX American Chemical Society
surface can be tuned by changing the upper polymer thickness, co-monomer ratio, ligand concentration, etc. However, changing the upper polymer thickness of block copolymer to regulate protein immobilization would inevitably impair the antifouling property of surfaces,14 and it was difficult to regulate protein adsorption to a large extent by changing the ratio of monomers containing reactive groups in a random copolymer.10,15 Although changing the ligand concentration in solution could modulate ligand density in some cases,16,17 these covalent modification methods require multistep reactions or activation and blocking, unavoidably increasing fabrication time and producing side effects, and the involved organic solvents may corrode vulnerable surfaces. Because of the dynamic self-assembled nature, facile fabrication process, and good biocompatibility, host−guest interaction as a new noncovalent modification method has attracted a great deal of attention.18−22 Among different host− guest pairs, the β-cyclodextrin (β-CD)−adamantane pair has been widely used as a linker to construct biofunctional surfaces.23−26 The primary hydroxyl groups located on the narrower ring of β-CD can be selectively modified by many biomolecules, such as sugars,27 peptides,24 ssDNA,28 etc.; Received: April 15, 2015 Revised: May 14, 2015
A
DOI: 10.1021/acs.langmuir.5b01380 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Scheme 1. Methods for Modulating the Localized and Average Lysine Density of Surfaces via Host−Guest Interactions
meanwhile, the wider ring of β-CD remains open to suitable guests. β-CD used as a scaffold for ligands is capable of regulating ligand valency, type, spatial distribution, etc.29 The introduction of these properties onto surfaces will make it easy to regulate ligand presentation. Lysine−plasminogen (Lys−Plg) is a well-known specific pair in the fibrinolysis system30 and was chosen as a model to investigate the specific protein binding capability of ligandmodified surfaces in this work. The influence of surface lysine presentations on the extent of binding and the binding affinity of Plg for surfaces was investigated in detail. A bioinert surface, i.e., poly[2-hydroxyethyl methacrylate-co-(adamantan-1-yl)methyl methacrylate] (PHAda), statistical copolymer-modified silicon wafer was prepared by the surface-initiated atom transfer radical polymerization (SI-ATRP) method, and then lysinemonosubstituted {mono-6-lysine-β-CD [CD(Lys)1]} and lysine-persubstituted β-CD {per-6-lysine-β-CD [CD(Lys)7]} were integrated onto the surfaces via host−guest interactions between β-CD and adamantane. The localized density and the average lysine density of surfaces are modulated by changing the lysine valency on the β-CD scaffold and by using pure βCD to dilute per-6-lysine-β-CD as shown in Scheme 1, respectively.
Inc. (Shanghai, China). Nε-Boc-L-lysine-tert-butyl ester hydrochloride [H-Lys(Boc)-OtBu·HCl] was from GL Biochemicals Ltd. (Shanghai, China). Acid Orange 7 (Orange II) was purchased from TCI Development Co., Ltd. (Shanghai, China). Toluene, 2-propanol, N,Ndimethylformamide (DMF), acetonitrile, dichloromethane (DCM), trimethylamine (TEA), and dimethyl sulfoxide (DMSO) were from the Shanghai Chemical Reagent Co. and purified according to standard methods before use. Fibrinogen (plasminogen-free) was from Calbiochem (La Jolla, CA). Human serum albumin (HSA) was from Sigma-Aldrich (St. Louis, MO). Plasminogen (94 kDa) was obtained from ICN Pharmaceuticals (Irvine, CA). Na125I was from Chengdu Gaotong Isotope Co., Ltd. Silicon wafers [p-doped, (100)-oriented, 0.45 mm thick, 100 mm diameter] were purchased from the laboratory of Guangzhou Semiconductor Materials (Guangzhou, China). The silicon wafers were cut into square chips of 0.5 cm × 0.5 cm. The deionized water used in all experiments was purified with a Millipore water purification system to give a resistivity of 18.2 MΩ cm. Nitrogen gas was of high-purity grade. 2.2. Modification of Surfaces by PHAda Polymer Brushes and Lysine-Decorated β-CD Derivatives. The synthesis of monomer (adamantan-1-yl)methyl methacrylate (AdMA) is shown in the Supporting Information. The procedures reported in our previous work were followed in the pretreatment of silicon wafers for the immobilization of the initiator.31 Surface-initiated ATRP of HEMA and AdMA statistical copolymer was conducted in a glovebox purged with nitrogen. First, a mixture of 6 mL of 2-propanol, HEMA (1.478 g, 11.35 mmol), and PMDETA (25 μL, 0.12 mmol) was agitated with nitrogen for 30 min. Second, a small round-bottom flask containing a mixture of AdMA (141 mg, 0.6 mmol) and CuBr (18 mg, 0.12 mmol) was placed in a glovebox purged with nitrogen, to which the deoxygenated solution was added. The reaction solution was stirred vigorously, allowing complete dissolution, and then added to the glass vessels containing initiator-functionalized silicon wafers. Polymerization was conducted at ambient temperature under a nitrogen atmosphere for 6 h, and then the copolymer-grafted silicon wafers were taken out of the solution, cleaned ultrasonically in methanol five times (5 min each time), and dried under a flow of nitrogen. CD(Lys)1 and CD(Lys)7 were synthesized via copper-catalyzed azide−alkyne cycloaddition chemistry. The detailed synthesis and characterization are described in the Supporting Information (Figures S1−S5). The PHAda copolymer surface was immersed in a 1 mM
2. MATERIALS AND METHODS 2.1. Materials. β-Cyclodextrin (β-CD, 97%, Sigma-Aldrich) was recrystallized twice from water and dried in a vacuum oven at 100 °C for 2 days prior to being used. Copper(I) bromide (CuBr, 98%, SigmaAldrich) was washed sequentially with acetic acid and methanol and dried under vacuum. 2-Hydroxyethyl methacrylate (HEMA) (98%, Acros) was passed through a short inhibitor remover (Sigma-Aldrich) column to remove inhibitors and stored at 4 °C. 1-Adamantane methanol (99%, Sigma-Aldrich), α-bromoisobutyryl bromide (98%, Sigma-Aldrich), methyl 2-chloropropionate (97%, Sigma-Aldrich), N,N,N′,N″,N″-pentamethyldiethylenetriamine (PMDETA, 99%, Sigma-Aldrich), and 3-aminopropyltriethoxysilane (APTES, 99%, Sigma-Aldrich) were used as received. Methacryloyl chloride and ptoluenesulfonyl chloride (99%) were from Aladdin Reagent Database B
DOI: 10.1021/acs.langmuir.5b01380 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Table 1. Elemental Composition of Surfaces Obtained from the XPS Survey Scan (90° take-off angle) and WCAs of Surfaces elemental composition (%)
a
surface
C
N
O
Br
N/O
N/C
Si−Br PHEMA PHAda PHAda/CD PHAda/CD(Lys)1 PHAda/CD(Lys)7
40.90 65.87 65.27 67.20 64.10 65.90
6.77 0.45 0.43 0.57 1.79 1.83
28.59 30.65 29.93 29.78 31.91 28.71
1.76 0.06 0.12 0.11 0.06 0.18
0.237 0.015 0.014 0.019 0.056 0.064
0.166 0.007 0.007 0.008 0.028 0.028
WCA (deg)a 71.1 52.5 61.9 58.1 57.6 57.7
± ± ± ± ± ±
0.4 0.7 0.9 0.7 0.4 0.6
WCAs are means ± the standard deviation (n = 3).
aqueous solution of β-CD derivatives overnight at ambient temperature to form β-CD derivative-complexed surfaces, rinsed with deionized water to remove the unconjugated β-CD derivatives, and blown dry with N2. The mixture with a concentration of 1 mM was prepared by blending pure β-CD and CD(Lys)7 at different ratios. 2.3. Characterization. 2.3.1. Surface Characterization. The dry thickness of a polymer layer on the silicon substrate was measured with an M-2000 V spectroscopic ellipsometer within a spectral range of 371−1000 nm (J. A. Woollam Co., Inc.). The static water contact angles of silicon surfaces functionalized with PHAda and copolymer surfaces complexed by lysine-modified β-CD derivatives were measured using the sessile drop method on a model SL-200C optical contact angle meter (USA Kino Industry Co., Ltd.) under an atmospheric humidity of
