Effect of protein fusion on the transition temperature of an environmentally responsive elastin-like polypeptide: a role for surface hydrophobicity?

The limited throughput, scalability and high cost of protein purification by chromatography provide motivation for the development of non-chromatographic protein purification technologies that are cheaper and easier to implement in a high-throughput format for proteomics applications and to scale up for industrial bioprocessing. We have shown that genetic fusion of a recombinant protein to an elastin-like polypeptide (ELP) imparts the environmentally sensitive solubility property of the ELP to the fusion protein, and thereby allows selective separation of the fusion protein from Escherichia coli lysate by aggregation above a critical temperature (T(t)). Further development of ELP fusion proteins as widely applicable purification tools necessitates a quantitative understanding of how fused proteins perturb the ELP T(t) such that purification conditions (T(t)) may be predicted a priori for new recombinant proteins. We report here the effect that fusing six different proteins has on the T(t) of an ELP. A negative correlation between T(t) and the fraction hydrophobic surface area on the fused proteins was observed, which was determined from computer modeling of the available three-dimensional structure. The thermally triggered aggregation behavior of ELP-coated, functionalized gold colloids as well as ligand binding to the tendamistat-ELP fusion protein support the hypothesis that hydrophobic surfaces in molecular proximity to ELPs depress the ELP T(t) by a mechanism analogous to hydrophobic residue substitution in the ELP repeat, Val-Pro-Gly-Xaa-Gly.

Genetic engineering of a single-chain antibody fragment for surface immobilization in an optical biosensor.

The development of a biosensor based on a genetically engineered biomolecule offers many potential advantages to sensors that rely on natural proteins only. Here we present how protein engineering techniques can be used to introduce a functional unit for surface immobilization into a single-chain antibody fragment (scFv). A peptide known to mimic the binding properties of biotin was fused to the carboxyterminus of the phosphorylcholine-binding scFv fragment of IgA McPC603. This fusion protein could be immobilized on a streptavidin monolayer. The resulting scFv monolayer was capable of binding a fluorescently labeled phosphorylcholine analog, as detected by total internal reflection fluorescence. In contrast, an scFv monolayer formed by introducing biotin through chemical modification was not capable of binding phosphorylcholine. These results demonstrate the utility of site-specific, oriented attachment strategies in the formation of protein monolayers in optical sensors, made possible by the use of protein engineering techniques.

Assessment of heterogeneity in antibody-antigen displacement reactions.

The intrinsic binding characteristics of monoclonal antibodies are modified upon immobilization onto a solid-phase matrix. Factors such as the distribution in affinity must therefore be taken into consideration in order to predict the kinetics of antibody binding at solid-liquid interfaces. A mathematical analysis is presented herein that allows the assessment of heterogeneity in the affinity of monoclonal antibodies immobilized onto a solid support. This model is based on a modified version of the Sips distribution function adapted to the conditions of a solid-phase displacement assay in flow. An assay for trinitrotoluene (TNT) provides the data to evaluate the extent of heterogeneity introduced by immobilization of antibodies in a flow immunoassay. We determined the index of antibody heterogeneity on two solid supports, controlled-pore glass beads and agarose beads, coated with a monoclonal anti-TNT antibody at varying densities. The data confirm that the threshold for crossover from homogeneous to heterogeneous forms of the reaction isotherm is different in displacement reactions than in association-dissociation reactions. Our analysis shows that the measured displacement isotherm is consistent with a homogeneous or only moderately heterogenous distribution of relative affinities.