Comparison of descriptors for predicting selectivity of protein-imprinted polymers

Molecular imprinting is a technique that is used to create artificial receptors by the formation of a polymer network around a template molecule, creating a molecularly imprinted polymer. These artificial receptors may be used in applications that require molecular recognition, such as enantioseparations, biosensors, artificial catalysis, drug delivery and others. Small molecules, such as drugs, have been imprinted with high efficiency and, combined with the low cost of preparation, molecularly imprinted polymers have acquired commercial usage. While attempts at imprinting proteins have been significantly less successful, the great potential of protein-imprinted polymers (PIPs) in medicine and industry attracted much research. Multifunctionality, conformational flexibility, large size of the proteins, and aqueous polymerization environment are some of the obstacles faced by protein imprinting. We explore the relation between PIP selectivity and the properties of the template and competitor proteins. A comprehensive statistical analysis of published studies reveals a statistically significant correlation between four protein descriptors and the corresponding selectivity of PIPs. Namely, a PIP will generally be more selective against large competitor proteins with a smooth surface, whose isoelectric point and aspect ratio are significantly different than those of the template protein. The size of the protein, as measured by its molecular weight, appears to be independent of the template protein characteristics. Copyright © 2016 John Wiley & Sons, Ltd.