Biomacromolecules 2004, 5, 869-876
869
Thermodynamic Studies on the Adsorption of Fibronectin Adhesion-Promoting Peptide on Nanothin Films of Poly(2-vinylpyridine) by SPR Xiao Li, Xiaolin Wei, and Scott M. Husson* Department of Chemical Engineering, Clemson University, Clemson, South Carolina 29634-0909 Received July 30, 2003; Revised Manuscript Received December 12, 2003
This paper describes a methodology for preparing uniform, nanothin polymer films for the study of biomolecule adsorption by surface plasmon resonance spectroscopy (SPR). The methodology combines molecular self-assembly of alkanethiols on gold with surface-confined atom transfer radical polymerization (ATRP). Poly(2-vinylpyridine) was chosen to demonstrate the methodology, and growth kinetics were studied by ex situ ellipsometry. Atomic force microscopy (AFM) indicated that the polymer films were uniform with RMS roughness of ∼0.5 nm. Subsequent SPR measurements were done to determine thermodynamic adsorption properties (∆G, ∆H, and ∆S) between fibronectin adhesion-promoting peptide and the surfaceconfined poly(2-vinylpyridine) at 15, 20, and 25 °C. The flexibility in synthesis conditions and the opportunities for manipulating film thicknesses and graft densities that ATRP provides to grow polymer films from gold surfaces holds advantages over conventional spin-coating and grafting to approaches in the design of model polymer films for biomolecule adsorption studies. These advantages are described. Introduction The design of surfaces that interface with biological systems requires knowledge of, and better still, control of how biomolecules (i.e., proteins, peptides, etc.) adsorb to such surfaces. Oftentimes, the design aim is to make a surface inert, i.e., to prevent biomolecule adsorption. Examples include the surfaces of biomedical devices and heat exchangers in the food and dairy industries. In other cases, the design aim is to make a surface that interacts specifically with biomolecules in solution. Examples here include immunosensors and affinity chromatography stationary phase materials, where differences in adsorption behavior are exploited to effect a separation. For these and other applications, understanding biomolecule-surface interactions would allow better a priori design of materials surfaces to achieve the desired response. Since the early 1990s, surface plasmon resonance (SPR) spectroscopy has been used routinely to measure the binding properties of proteins and peptides on functionalized surfaces. In 1991, Prime and Whitesides1 began using self-assembled monolayers (SAMs) as model substrates for studying protein adsorption. Since then, a number of other groups have adopted SAMs with SPR to examine protein-surface interactions, with the aim of designing inert and functional surfaces and understanding the mechanisms that govern the adsorption process. Recently, the Whitesides group has begun to explore whether their results from SAM surfaces can be extended to polymeric films formed by grafted polymers onto the SAMs.2 A motivation for using polymer films is that they represent a closer model to real biomedical devices than * To whom correspondence should be addressed. Phone: (864) 6564502. Fax: (864) 656-0784. E-mail:
[email protected].
do functional SAMs. Another advantage put forward2 is that thin grafted films might mask any heterogeneities in the gold surface that a SAM layer would not. Green et al.3 established that SPR was useful for studying protein adsorption onto polymer films by examining the adsorption of albumin, fibrinogen, and IgG onto spin-coated polystyrene. That same group studied albumin adsorption to Pluronic PEO/PPO/PEO tri-block copolymers of varying PEO and PPO content.4,5 Here, the tri-block copolymers were physisorbed onto a polystyrene substrate that had been spincoated on silver. Pavey and Olliff6 also used SPR to study protein adsorption to gold surfaces coated directly with PEO/ PPO block copolymers. Frazier et al.7 investigated protein resistant dextran layers by SPR. The dextran layers were formed by thiol derivatization of dextran, followed by chemisorption to silver or gold SPR slides, an approach that lends itself to the preparation of other chemisorbed polymers on gold.8-11 Ratner and co-workers have also used SPR to study protein adsorption on PEO-like coatings prepared by radio frequency glow discharge polymerization.12 In all of these examples, the polymer films were either (plasma) deposited, spin-coated, physisorbed, or covalently grafted to the underlying surface. This paper describes a methodology for preparing uniform, nanothin polymer films for the study of biomolecule adsorption. The methodology combines molecular self-assembly of alkanethiols on gold with surface-confined atom transfer radical polymerization. The objectives of this research were to grow and characterize molecularly uniform, nanothin, surface-confined polymers from flat, gold surfaces that would be suitable for SPR measurements and to conduct fundamental measurements of thermodynamic adsorption properties (∆G, ∆H, and ∆S) between fibronectin adhesion-
10.1021/bm034266k CCC: $27.50 © 2004 American Chemical Society Published on Web 04/08/2004
870
Biomacromolecules, Vol. 5, No. 3, 2004
promoting peptide and these surface-confined polymers. In this study, poly(2-vinylpyridine) was chosen to demonstrate the methodology. Structural characterization was done to ensure that a uniform surface was presented to the peptide adsorbing onto it during the adsorption studies. The primary difference between this work and previously reported SPR studies of biomolecule-polymer interactions is that atom transfer radical polymerization (ATRP) was used here to grow polymer films from the surface, whereas physisorption or “grafting to” approaches were used previously. ATRP was used to prepare the polymer films, because it has been shown to produce uniform surface films,13 relative to conventional radical polymerization methods. Relative to the grafting to approach, surface-confined polymer growth also leads to higher graft densities.14,15 This fact is an important consideration for biomolecule adsorption studies. Green et al.3 note that careful film preparation is the most important factor for SPR studies on spin-coated polymer films. They explain that the polymer surface chemistry must be reproducible and free of defects for meaningful SPR analysis. Osterberg et al.16 and Frazier et al.7 showed that, for grafted to layers, relative surface coverage, rather than relative layer thickness, was a more important factor for determining protein adsorption behavior. Furthermore, protein adsorption onto immobilized dextran layers has been shown to depend on the configuration of dextran at the surface.17 Thus, high graft densities should prevent accessibility to the underlying substrate, thereby giving a true reflection of the adsorption properties of biomolecules to polymer layers. We and others have used ATRP previously to grow polymer films from SAMs on gold,18-21 making these substrates available for biomolecule adsorption studies by SPR. The flexibility offered by ATRP allows the use of numerous functional monomers, and post polymerization modifications are possible.22 By controlling the areal density of initiator on the surface,23 opportunities exist for manipulating chain densities. By controlling monomer and catalyst concentrations, temperature, and time, polymer molecular weight is controllable also. Thus, we believe that using ATRP to grow polymer films from gold surfaces provides advantages over conventional spin-coating and grafting to approaches in the design of model polymer films for biomolecule adsorption studies. Materials and Methods Materials. The fibronectin adhesion-promoting peptide (Trp-Glu-Pro-Pro-Arg-Ala-Arg-Ile) (95%) was used as received from Sigma Chemical. Gold substrates were used as received from BIAcore, Inc. (SIA Au kit, BR-1004-05). All chemicals were purchased from Aldrich and used as received, unless noted otherwise; they were 11-mercapto-1-undecanol (97%), (4-chloromethyl)benzoyl chloride (97%), 2-vinylpyridine (97%), copper (I) bromide (99.999%), copper (II) chloride (99.999%), and tris-(2-aminoethyl)amine (TREN) (96%). Solvents were purchased from Aldrich as ACS reagent grade; they were ethyl alcohol (99.5%), anhydrous toluene (99.8%), and acetonitrile (99.9+%). All percentages
Li et al.
are in wt %. 2-Vinylpyridine was purified by vacuum distillation at 25 mmHg before use to remove the inhibitor (p-tert-butyl catechol). Cleaning of Gold Substrates. Prior to use, the gold-coated glass substrates (1 cm × 1 cm) were cleaned in a UV cleaner (Boekel, Inc., model 135500) and then rinsed with deionized water. Following this cleaning procedure, the plain gold substrates were characterized. Background spectra were collected in external reflectance Fourier transform infrared spectroscopy (ER-FTIR); refractive index (N) and extinction coefficient (K) values were measured in spectroscopic ellipsometry; and water contact angle values were measured by static contact angle goniometry. Preparation of Initiator-Functionalized Surfaces. The gold substrates were modified by a two-step process to allow subsequent growth of surface-confined polymer. In a first step, 11-mercapto-1-undecanol was dissolved in ethanol at a concentration of 1 mM, and gold substrates were incubated in this thiol solution for 14 to 20 h at room temperature to form a thiol self-assembled monolayer (SAM). The surfaces were then washed in ethanol using an Aquasonic ultrasonic cleaner for 10 s, rinsed with ethanol and deionized water, and dried in a stream of nitrogen. SAM layers were characterized by static water contact angle, ellipsometry, and ER-FTIR methods. In the second step, (4-chloromethyl)benzoyl chloride was dissolved in toluene at a concentration of 2 mM in a waterfree (
