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.

A precise representation of the spatial distribution of hydrophobicity, hydrophilicity and charges on the molecular surface of proteins is critical for the understanding of the interaction with small molecules and larger systems. The representation of hydrophobicity is rarely done at atom-level, as this property is generally assigned to residues. A new methodology for the derivation of atomic hydrophobicity from any amino acid-based hydrophobicity scale was used to derive 8 sets of atomic hydrophobicities, one of which was used to generate the molecular surfaces for 35 proteins with convex structures, 5 of which, i.e., lysozyme, ribonuclease, hemoglobin, albumin and IgG, have been analyzed in more detail. Sets of the molecular surfaces of the model proteins have been constructed using spherical probes with increasingly large radii, from 1.4 to 20 Å, followed by the quantification of (i) the surface hydrophobicity; (ii) their respective molecular surface areas, i.e., total, hydrophilic and hydrophobic area; and (iii) their relative densities, i.e., divided by the total molecular area; or specific densities, i.e., divided by property-specific area. Compared with the amino acid-based formalism, the atom-level description reveals molecular surfaces which (i) present an approximately two times more hydrophilic areas; with (ii) less extended, but between 2 to 5 times more intense hydrophilic patches; and (iii) 3 to 20 times more extended hydrophobic areas. The hydrophobic areas are also approximately 2 times more hydrophobicity-intense. This, more pronounced “leopard skin”-like, design of the protein molecular surface has been confirmed by comparing the results for a restricted set of homologous proteins, i.e., hemoglobins diverging by only one residue (Trp37). These results suggest that the representation of hydrophobicity on the protein molecular surfaces at atom-level resolution, coupled with the probing of the molecular surface at different geometric resolutions, can capture processes that are otherwise obscured to the amino acid-based formalism.

Many areas of biochemistry and molecular biology, both fundamental and applications-orientated, require an accurate construction, representation and understanding of the protein molecular surface and its interaction with other, usually small, molecules. There are however many situations when the protein molecular surface gets in physical contact with larger objects, either biological, such as membranes, or artificial, such as nanoparticles. The contribution presents a methodology for describing and quantifying the molecular properties of proteins, by geometrical and physico-chemical mapping of the molecular surfaces, with several analytical relationships being proposed for molecular surface properties. The relevance of the molecular surface-derived properties has been demonstrated through the calculation of the statistical strength of the prediction of protein adsorption. It is expected that the extension of this methodology to other phenomena involving proteins near solid surfaces, in particular the protein interaction with nanoparticles, will result in important benefits in the understanding and design of protein-specific solid surfaces.

No abstract

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, protein hydrophobicity and spread 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 into two separate subsets representing protein adsorption on hydrophilic and hydrophobic surfaces. 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 absorbed layer and the surface tension of the proteincovered 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.