The BAD project: data mining, database and prediction of protein adsorption on surfaces

Protein adsorption at solid-liquid interfaces is critical to many applications, including biomaterials, protein microarrays and lab-on-a-chip devices. Despite this general interest, and a large amount of research in the last half a century, protein adsorption cannot be predicted with an engineering level, design-orientated accuracy. Here we describe a Biomolecular Adsorption Database (BAD), freely available online, which archives the published protein adsorption data. Piecewise linear regression with breakpoint applied to the data in the BAD suggests that the input variables to protein adsorption, i.e., protein concentration in solution; protein descriptors derived from primary structure (number of residues, global protein hydrophobicity and range of amino acid hydrophobicity, isoelectric point); surface descriptors (contact angle); and fluid environment descriptors (pH, ionic strength), correlate well with the output variable-the protein concentration on the surface. Furthermore, neural network analysis revealed that the size of the BAD makes it sufficiently representative, with a neural network-based predictive error of 5% or less. Interestingly, a consistently better fit is obtained if the BAD is divided in two separate sub-sets representing protein adsorption on hydrophilic and hydrophobic surfaces, respectively. Based on these findings, selected entries from the BAD have been used to construct neural network-based estimation routines, which predict the amount of adsorbed protein, the thickness of the adsorbed layer and the surface tension of the protein-covered surface. While the BAD is of general interest, the prediction of the thickness and the surface tension of the protein-covered layers are of particular relevance to the design of microfluidics devices.

Injection-molded high-density polyethylene-hydroxyapatite-aluminum oxide hybrid composites for hard-tissue replacement: Mechanical, biological, and protein adsorption behavior

In this article, we report the mechanical and biocompatibility properties of injection-molded high-density polyethylene (HDPE) composites reinforced with 40 wt % ceramic filler [hydroxyapatite (HA) and/or Al2O3] and 2 wt % titanate as a coupling agent. The mechanical property measurements revealed that a combination of a maximum tensile strength of 18.7 MPa and a maximum tensile modulus of about 855 MPa could be achieved with the injection-molded HDPE20 wt % HA20 wt % Al2O3 composites. For the same composite composition, the maximum compression strength was determined to be 71.6 MPa and the compression modulus was about 660 MPa. The fractrography study revealed the uniform distribution of ceramic fillers in the semicrystalline HDPE matrix. The cytocompatibility study with osteoblast-like SaOS2 cells confirmed extensive cell adhesion and proliferation on the injection-molded HDPE20 wt % HA20 wt % Al2O3 composites. The cell viability analysis with the 3(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay revealed a statistically significant difference between the injection-molded HDPE20 wt % HA20 wt % Al2O3 composites and sintered HA for various culture durations of upto 7 days. The difference in cytocompatibility properties among the biocomposites is explained in terms of the difference in the protein absorption behavior. (C) 2011 Wiley Periodicals, Inc. J Appl Polym Sci, 2012

Dependence of Protein Adsorption on Wetting Behavior of UHMWPE-HA-Al2O3-CNT Hybrid Biocomposites

Ultrahigh-molecular-weight polyethylene (UHMWPE) is used as an articulating surface in total hip and knee joint replacement. In order to enhance long-term durability/wear resistance properties, UHMWPE-based polymer-ceramic hybrid composites are being developed. Surface properties such as wettability and protein adsorption alter with reinforcement or with change in surface chemistry. From this perspective, the wettability and protein adsorption behavior of compression-molded UHMWPE-hydroxyapatite (HA)-aluminum oxide (Al2O3)-carbon nanotube (CNT) composites were analyzed in conjunction with surface roughness. The combined effect of Al2O3 and CNT shows enhancement of the contact angle by similar to 37A degrees compared with the surface of the UHMWPE matrix reinforced with HA. In reference to unreinforced UHMWPE, protein adsorption density also increased by similar to 230% for 2 wt.%HA-5 wt.%Al2O3-2 wt.%CNT addition to UHMWPE. An important conclusion is that the polar and dispersion components of the surface free energy play a significant role in wetting and protein adsorption than do the total free energy or chemistry of the surface. The results of this study have major implications for the biocompatibility of these newly developed biocomposites.

Cryogenically cured hydroxyapatite-gelatin nanobiocomposite for bovine serum albumin protein adsorption and release

This research paper presents the first results on the protein adsorption and release kinetics and in vitro biodegradability of cryogenically cured hydroxyapatite-gelatin based micro/macroporous scaffolds (CHAMPS). While the adsorption and release of bovine serum albumin (BSA) protein exhibits steady state behavior over an incubation period of up to 10 days, Fourier transform infrared (FT-IR) analysis importantly confirms the absence of any change in the secondary structure of BSA proteins due to interaction with the CHAMPS scaffold. The compression properties of the CHAMPS scaffold with interconnected porosity (pore size similar to 50-200 mm) is characterized by a non-linear stress-strain response with a strength close to 5 MPa and a maximum strain of up to 24%. The slow but systematic increase in weight loss over a period of 7 days as well as apatite layer formation indicates its good bioactivity. The extensive micro-computed tomography (micro-CT) analysis establishes cancellous bone-like highly interconnected and complex porous architecture of the CHAMPS scaffold. Importantly, the excellent adsorption (up to 50%) and release (up to 60% of adsorbed protein) of BSA has been uniquely attributed to the inherent porous microstructure of the CHAMPS scaffold. Overall, the present study provides an assessment of the interaction of protein with the gelatin-hydroxyapatite macroporous scaffold in vitro, as well as reporting for the first time the efficacy of such scaffolds to release 60% of BSA loaded onto the scaffold in vitro, which is significantly higher than earlier literature reports.

Competitive protein adsorption studied with atomic force microscopy and imaging ellipsometry

The adsorption and competitive adsorption of collagen and bovine serum albumin (BSA) were directly visualized and quantified using atomic force microscopy (AFM) and imaging ellipsometry. Chemically modified silicon surfaces were used as hydrophilic and hydrophobic substrates. The results showed that collagen and BSA in single component solution adsorbed onto a hydrophobic surface two times more than that onto a hydrophilic surface. The competitive adsorption between collagen and BSA showed that serum albumin preferentially adsorbed onto a hydrophobic surface, while collagen on a hydrophilic surface. In the binary solution of BSA (1 mg/ml BSA) and collagen (0.1 mg/ml), nearly 100% of the protein adsorbed onto the hydrophobic surface was BSA, but on the hydrophilic surface only about 6% was BSA. Surface affinity was the main factor controlling the competitive adsorption.

Analysis of steric mass-action model for protein adsorption equilibrium onto porous anion-exchange adsorbent

The ion-exchange equilibrium of bovine serum albumin (BSA) to an anion exchanger, DEAE Spherodex M, has been studied by batch adsorption experiments at pH values ranging from 5.26 to 7.6 and ionic strengths from 10 to 117.1 mmol/l. Using the unadjustable adsorption equilibrium parameters obtained from batch experiments, the applicability of the steric mass-action (SMA) model was analyzed for describing protein ion-exchange equilibrium in different buffer systems. The parametric sensitivity analysis was performed by perturbing each of the model parameters, while holding the rest constant. The simulation results showed that, at high salt concentrations or low pHs close to the isoelectric point of the protein, the precision of the model prediction decreased. Parametric sensitivity analysis showed that the characteristic charge and protein steric factor had the largest effects on ion-exchange equilibrium, while the effect of equilibrium constant was about 70%-95% smaller than those of characteristic charge and steric factor under all conditions investigated. The SMA model with the relationship between the adjusted characteristic charge and the salt concentration can well predict the protein adsorption isotherms in a wide pH range from 5.84 to 7.6. It is considered that the SMA model could be further improved by taking into account the effect of salt concentration on the intermolecular interactions of proteins. (c) 2006 Elsevier Ltd. All rights reserved.

Optimization of evanescent wave imaging for the visualization of protein adsorption layers

We have optimized the settings of evanescent wave imaging for the visualization of a protein adsorption layer. The enhancement of the evanescent wave at the interface brought by the incident angle, the polarized state of light beam as well as a gold layer is considered. In order to improve the image contrast of a protein monolayer in experiments, we have optimized three factors-the incident angle, the polarization of light beam, and the thickness of an introduced thin gold layer with a theoretical simulation.

Modeling of protein adsorption in membrane affinity chromatography

For the design of affinity membranes, protein adsorption in membrane affinity chromatography (MAC) was studied by frontal analysis. According to fast mass transfer, small thickness of affinity membranes and high affinity between the protein and the ligand, an ideal adsorption (IA) model was proposed for MAC and was used together with equilibrium-dispersive (E-D) model to describe the adsorption of bovine serum albumin (BSA) onto cellulose diacetate/polyethyleneimine (CA/PEI) blend membranes with and without Cu2+ chelating. E-D model was found to better describe the initial region of experimental breakthrough curves. The influence of axial dispersion was revealed and it showed the importance of design of the module to homogenously distribute feed solution. IA model was found to be better for the whole experimental breakthrough curve. According to it, the capacity of affinity membranes and the specificity of the interaction are of equal importance for the design of affinity membranes. An optimum feed concentration was also found in the operation of MAC. The discrepancy between experimental optimum feed concentrations and predicted ones from IA model may be due to the ignorance of some experimental effects such as axial dispersion.

A comparison of gold nanoparticle surface co-functionalization approaches using Polyethylene Glycol (PEG) and the effect on stability, non-specific protein adsorption and internalization

To create clinically useful gold nanoparticle (AuNP) based cancer therapeutics it is necessary to co-functionalize the AuNP surface with a range of moieties; e.g. Polyethylene Glycol (PEG), peptides and drugs. AuNPs can be functionalized by creating either a mixed monolayer by attaching all the moieties directly to the surface using thiol chemistry, or by binding groups to the surface by means of a bifunctional polyethylene glycol (PEG) linker. The linker methodology has the potential to enhance bioavailability and the amount of functional agent that can be attached. While there is a large body of published work using both surface arrangements independently, the impact of attachment methodology on stability, non-specific protein adsorption and cellular uptake is not well understood, with no published studies directly comparing the two most frequently employed approaches. This paper compares the two methodologies by synthesizing and characterizing PEG and Receptor Mediated Endocytosis (RME) peptide co-functionalized AuNPs prepared using both the mixed monolayer and linker approaches. Successful attachment of both PEG and RME peptide using the two methods was confirmed using Dynamic Light Scattering, Fourier Transform Infrared Spectroscopy and gel electrophoresis. It was observed that while the 'as synthesized' citrate capped AuNPs agglomerated under physiological salt conditions, all the mixed monolayer and PEG linker capped samples remained stable at 1M NaCl, and were stable in PBS over extended periods. While it was noted that both functionalization methods inhibited non-specific protein attachment, the mixed monolayer samples did show some changes in gel electrophoresis migration profile after incubation with fetal calf serum. PEG renders the AuNP stable in-vivo however, studies with MDA-MB-231 and MCF 10A cell lines indicated that functionalization with PEG, blocks cellular uptake. It was observed that co-functionalization with RME peptide using both the mixed monolayer and PEG linker methods greatly enhanced cellular internalization compared to PEG capped AuNPs.

Probing the functionalization of gold surfaces and protein adsorption by PM-IRRAS

The control of morphology and coating of metal surfaces is essential for a number of organic electronic devices including photovoltaic cells and sensors. In this study, we monitor the functionalization of gold surfaces with 11-mercaptoundecanoic acid (MUA, HS(CH(2))(10)CO(2)H) and cysteamine, aiming at passivating the surfaces for application in surface plasmon resonance (SPR) biosensors. Using polarization-modulated infrared reflection-absorption spectroscopy (PM-IRRAS), cyclic voltammetry, atomic force microscopy and quartz crystal microbalance, we observed a time-dependent organization process of the adsorbed MUA monolayer with alkyl chains perpendicular to the gold surface. Such optimized condition for surface passivation was obtained with a systematic search for experimental parameters leading to the lowest electrochemical signal of the functionalized gold electrode. The ability to build supramolecular architectures was also confirmed by detecting with PM-IRRAS the adsorption of streptavidin on the MUA-functionalized gold. As the approaches used for surface functionalization and its verification with PM-IRRAS are generic, one may now envisage monitoring the fabrication of tailored electrodes for a variety of applications.

Protein Adsorption onto Polyelectrolyte Layers: Effects of Protein Hydrophobicity and Charge Anisotropy

Ellipsometry was used to investigate the influence of ionic strength (I) and pH on the adsorption of bovine serum albumin (BSA) or beta-lactoglobulin (BLG) onto preabsorbed layers of two polycations: poly(diallyldimethylammonium chloride) (PDADMAC) or poly(4-vinylpyridine bromide) quaternized with linear aliphatic chains of two (QPVP-C2) or five (QPVP-C5) carbons. Comparisons among results for the three polycations reveal hydrophobic interactions, while comparisons between BSA and BLG-proteins of very similar isoelectric points (pI)-indicate the importance of protein charge anisotropy. At pH close to pI, the ionic strength dependence of the adsorbed amount of protein (Gamma) displayed maxima in the range 10 < I < 25 mM corresponding to Debye lengths close to the protein radii. Visualization of protein charge by Delphi suggested that these ionic strength conditions corresponded to suppression of long-range repulsion between polycations and protein positive domains, without diminution of short-range attraction between polycation segments and locally negative protein domains, in a manner similar to the behavior of PE-protein complexes in solution.(1-4) This description was consistent with the disappearance of the maxima at pH either above or below pI. In the former case, Gamma values decrease exponentially with I(1/2), due to screening of attractions, while in the latter case adsorption of both proteins decreased at low I due to strong repulsion. Close to or below pI both proteins adsorbed more strongly onto QPVP-C5 than onto QPVP-C2 or PDADMAC due to hydrophobic interactions with the longer alkyl group. Above pI, the adsorption was more pronounced with PDADMAC because these chains may assume more loosely bound layers due to lower linear charge density.