AVS2001 Session BI-TuA: Non Fouling Surfaces and Theoretical Concepts

Tuesday, October 30, 2001 2:00 PM in Room 102

Tuesday Afternoon

Start	Invited?	Item
2:00 PM	Invited	BI-TuA-1 Surface Forces and Coating Properties Involved in Protein Repellency P. Kingshott, H. Thissen, L. Meagher, P. Hamilton-Brown, H.J. Griesser (CSIRO, Molecular Science, Australia) PEO coatings have attracted much interest as non-fouling coatings. However, literature data show varying extents of reduction in protein adsorption with PEO coatings prepared in various ways, and use of techniques that may not always have been sufficiently sensitive to support claims of non-fouling (as opposed to low-fouling). We have immobilized PEG chains of different lengths onto surfaces with different densities of pinning groups and at room temperature as well as under cloud point conditions to study how these parameters affects macromolecular conformations and protein resistance. We have also investigated the limits of detection of adsorbed proteins on the 'best' coatings by the sensitive surface analytical methos XPS, ToF-SSIMS and MALDI. Cell colonization was found to be totally inhibited and this could be attributed to the inability of fibronectin and vitronectin to adsorb to the coating. Using a laser ablation technique, patterns were then created of cell-adhesive islands within a PEO-coated surface area. It was shown that the cells recognized edges with high precision. Finally, surface force curves were acquired using a colloid-modified AFM tip in order to probe for the interfacial forces that contribute to incomplete or complete protein repellency. Our PEO coatings differ markedly in structure and some properties from oligo-EO coatings prepared by SAM methodology, yet give analogous results in terms of resistance to fouling. Based on this and data with polysaccharides (Hartley et al, this meeting) we speculate that protein resistance does not require a 'magic' chemistry or a fully extended 'brush' structure; a highly hydrated coating that possesses a repulsive surface force due to steric-entropic-osmotic effects on compression, of sufficient magnitude and range to screen attractive van der Waals and electrostatic forces emanating from the substrate, is sufficient. The chemical composition may not matter as long as the coating is well hydrated and of a minimal thickness, and protein repellency may solely be a result of appropriate physico-chemical properties. Moreover, charge neutrality is required, as negatively charged coatings such as hyaluronan are effective only against some fouling situations. For insance our PEO coatings are neutral and screen substrate charges, thus repelling proteins of both charge signs.
2:40 PM		BI-TuA-3 Water-Uptake of Poly(ethylene glycol)-terminated Self-Assembled Monolayers during Film Formation J. Fick, S. Tokumitsu, M. Himmelhaus, M. Grunze (Universität Heidelberg, Germany) Self-assembled monolayers (SAMs) terminated by poly(ethylene glycol) (PEG; MW = 2000 Dalton) formed on polycrystalline gold have proven to provide an interesting model system for the study of grafted PEG chains with different morphologies, such as mushroom, polymer brush, and crystalline-like phase. The desired structure can be obtained simply by varying the immersion time of the substrate in solution, because the adsorbed molecules adopt the respective conformations as a function of coverage. As there is still a controversial discussion about the origin of the unique properties of PEG in terms of protein resistance and the roles that both, morphology of the PEG and bound water molecules might play, we have studied the water-uptake of the PEG-SAMs as a function of surface coverage by optical spectroscopies. The SAMs were adsorbed from solvent mixtures with distinct amounts of water added. We present the dependency of the adsorption kinetics as a function of various parameters, such as polarity, water-content of the solvent, and temperature during adsorption.
3:20 PM		BI-TuA-5 Novel PEO-containing Copolymers as Protein Repellent Additives In Polyurethanes: Evaluation of Protein Interactions by Radiolabelling, XPS and MALDI J.H. Tan (McMaster University, Canada); K.M. McLean, T.R. Gengenbach, H.J. Griesser (CSIRO Molecular Science, Australia); J.L. Brash (McMaster University, Canada) Polyurethanes (PUs) have long been used for medical applications, mainly because of their excellent mechanical properties. However, there is a need to improve the biocompatibility of these materials. Polyethylene oxide (PEO) has gained recognition as a biocompatible material and appears to interact minimally with proteins and cells. In this work, materials have been developed based on PEO-containing additives that can be applied to conventional PUs. The additives are amphiphilic triblock copolymers, PEO-PU-PEO, the middle segment of which has the same structure as the PU substrate. We hypothesize that such additives should interact strongly and be compatible with any polyurethane of structure similar to the middle segment, and that they should migrate to the PU-aqueous interface. Copolymers were synthesized using PEO blocks of varying MW (550, 2000, 5000) and a central PU block of MW 5000. Materials were prepared by blending the block copolymers with a base PU. The surfaces were characterized by water contact angle and XPS. Adsorption of proteins was investigated by radiolabelling and by XPS. The water contact angle data showed that the blends became more hydrophilic with increasing copolymer content. Radiolabelled fibrinogen expts showed that adsorption was much lower on the blends than on the unmodified PU, in some cases showing reductions of greater than 99%. For the 10% blends, surprisingly, adsorption decreased in the order PEO5000>2000>550. This "inverse dependence" is attributed to slower diffusion of the higher MW copolymers to the interface. The protein adsorption characteristics were also investigated using XPS and surface-MALDI using a range of individual plasma proteins (fibronectin, vitronectin, albumin, insulin, fibrinogen and IgG) and whole plasma. XPS results confirmed that protein adsorption on the blends was negligible compared to the unmodified PU. Ongoing surface MALDI experiments also indicate low adsorption on the copolymer-PU blends.
3:40 PM		BI-TuA-6 Surface Modification of Poly(Vinyl Chloride) Intubation Tubes to Control Bacterial Adhesion D.J. Balazs, Y. Chevolot, K. Triandafillu, H. Harms, C. Hollenstein, H.J. Mathieu (Swiss Federal Institute of Technology - Lausanne, Switzerland) Bacterial colonization of intubation tubes is responsible for 30% of all nosocomial pneumonia cases, 40 % of which lead to death, despite aggressive antibiotic therapy.footnote1 Therefore, a strategy to reduce bacterial adhesion is desirable. We are developing an approach based on the surface modification of the polymer used for this application, medical grade poly(vinyl chloride) (PVC). The strategy is to mask the PVC substrate with a chemically inert teflon-like fluoropolymer layer, which serves as an ideal platform for further surface modification due to its low surface energy.² Protein³ and bacterial⁴ repellant molecules, e.g. amphiphilic Pluronics^R, are bound to the fluoropolymer films using hydrophobic-hydrophobic interactions. This paper investigates fluoropolymer films created on PVC substrates through plasma-enhanced chemical vapor deposition. The films are deposited in an RF-plasma reactor, using C₂F₆ as a precursor and H₂ as a carrier gas. XPS data suggest that the films completely mask the substrate, as no remaining signatures of PVC are detectable. Moreover, alpha step measurements show a uniform film, with a thickness of approx. 200 nm. The fluoropolymer films were found to be highly hydrophobic, with a water contact angle > 100 °. Preliminary contact angle measurements of the Pluronic^R surfaces show a significant decrease in contact angle, (approx. 20 °) indicating adhesion to the fluoropolymer layer. Feedback from imaging XPS is then used to optimize Pluronic^R monolayer formation on the fluoropolymer film. Protein adsorption and in vitro bacterial adhesion studies will also be reported. ¹ J. L. Vincent et al (1995) JAMA 274: 639-644 ² I. Noh et al (1997) J Polym Sci Pol Chem 35: 1499-1514 ³ M. Paulsson et al (1993) Biomaterials 14: 845-853 ⁴ M. J. Bridgett et al (1992) Biomaterials 13: 411-416.
4:00 PM		BI-TuA-7 XPS-Mediated Robust Design Used to Optimize Hyaluronic Acid Surface Immobilization T.A. Barber (University of California, Berkeley); R.A. Stile (Northwestern University); D.G. Castner (University of Washington); K.E. Healy (University of California, Berkeley) A major limitation of scaffold-based cartilage tissue engineering approaches is the inability of the delivery scaffolds to adhere to the tissue lining cartilaginous defects. Previously, thermo-responsive P(NIPAAm-co-AAc) hydrogels were engineered to support chondrocyte viability and promote articular cartilage-like tissue formation in vitro. An objective of the current research is to functionalize these hydrogels with bioactive peptides to support specific interactions with components found in cartilaginous extracellular matrix (specifically Hyaluronic Acid (HA)). It is hypothesized that these interactions will significantly enhance scaffold-defect adhesion. A model HA surface was developed to test this hypothesis by quantifying these interactions. A commonly used strategy for HA surface immobilization exploits a carbodiimide reaction between the carboxylic acid groups present in HA, and an aminofunctional surface. However, a wide variety of reaction conditions have been reported, and it is unclear which elements are critical to this HA-grafting approach. Consequently, Robust Design methods were employed to optimize the HA-grafting procedure on aminofunctional glass substrates. First, a 4-factor, 3-level orthogonal array (L₉) was constructed to monitor the effects of HA concentration, coupling buffer (CB), CB pH, and Carbodiimide/N-Hydroxysulfo-succinimide concentration ([EDC/NHS]) on HA-grafting success. XPS C/Si ratios were used to assess factor effects. Deconvolution of the L₉ identified HA concentration, CB, and [EDC/NHS] as the dominant process variables. Subsequently, a 3-factor, 2-level orthogonal array (L₄) was used to further refine the HA-grafting conditions. Evaluation of the L₄ suggested optimal levels for HA concentration (2.5µM), CB (10mM HEPES), and [EDC/NHS] (100mM/25mM). Optimal grafting conditions will be utilized for preparing model surfaces to evaluate the adhesive properties of the functionalized gels using JKR adhesion testing.
5:00 PM		BI-TuA-10 Theoretical Prediction of the Enthalpic and Entropic Contributions of the Change in Gibbs Free Energy for Peptide Residue Adsorption onto Functionalized SAM Surfaces R.A. Latour (Clemson University) The thermodynamic energy contributions of the change in enthalpy (dH) and entropy (dS), and their summation to calculate the change in Gibbs free energy (dG), provide a very useful tool to predict complex biomolecular behavior. This approach has been successfully applied to address a wide range of biomolecular problems such as the prediction of protein and RNA folding and ligand-receptor binding for rational drug design. A similar approach holds great potential to be applied to understand and predict the adsorption behavior of proteins to synthetic surfaces. A protein is composed of specific sequences of peptide residues arranged in a well-defined structural organization. Protein-surface adsorption can be expressed as a set of intermolecular (residue-surface) and intramolecular (residue-residue) interactions with the minimization of these energetic contributions determining the final conformation and orientation of adsorbed protein. In this study, computation chemistry (MOPAC/PM3/COSMO) was combined with wetting data to predict dH, dS, and dG contributions for the adsorption of individual peptide residues (alanine, serine, lysine) on functionalized SAM surfaces (methyl, hydroxyl, carboxyl) as a function of surface separation distance (SSD). The results are in close agreement with other more generalized continuum-based theories of adsorption and predict how dH and dS from residue/surface and solvent restructuring effects contribute uniquely for each residue/surface pair. These results will serve as the foundational building blocks of more advanced treatments to quantitatively predict protein adsorption behavior with subsequent application for biomaterials surface design.