Introduction: Many disease states are characterized by the presence, absence, or change of molecular biological markers, or biomarkers. Common affinity-type recognition elements used in the detection of biomarkers are antibodies, either Immunoglobulins (IgGs) or recombinant fragments. While highly specific, the dissociation kinetics of antibodies highly favors the antibody-antigen complex, resulting in single, endpoint measurements [1]. Endpoint measurements are useful for ex vivo assays, but further antibody engineering is required for in vivo continuous monitoring. A key burden to overcome for using antibodies in continuous monitoring applications is the lack of binding site regeneration [2].Thus, we aim to create pH-modulable single chain fragment variables (scFvs), in which the KD of the scFv can be controlled to favor dissociation. We posit that through mutagenesis of residues within the complementarity-determining regions (CDRs) to histidine, we are able to modulate antibody binding kinetics through a change in pH [3,4]. There is a basis in the literature that supports this antibody design approach [5]. Those working with therapeutic antibodies currently employ similar techniques to effectively “recycle,” or regenerate the binding site of the antibody in vivo [6]. We seek to expand these techniques into the biosensing space to develop continuous, antibody-based biosensors. Methods: ClusPro was utilized to predict the antigen-binding residues in the CDRs of an anti-insulin scFv. From the predicted residues, we selected three amino acid residues to mutate to histidine with the intent of generating pH-modulable mutants. Three single mutants and a triple mutant containing all mutations were designed. The wild-type (WT) scFv, as well as the four mutants were all expressed as a soluble protein using Escherichia coli (E. coli) as a host microorganism. The binding ability of the WT and four mutants was assessed via two assays in differing pH with Bio-Layer Interferometry (BLI). The BLI assay workflow is as follows: biotinylated insulin was immobilized to Octet Streptavidin tips in buffers of either pH 7.4 or 6.0. Following immobilization, either the WT or mutants were introduced to bind to insulin. The resulting association and dissociation sensorgrams were recorded, and nonlinear fitting of the data was performed in GraphPad Prism 10.0. Results and Discussion: Compared to the WT, the triple histidine mutations in the CDR did not eliminate the scFv’s ability to bind to insulin. For the WT, the apparent KD decreased slightly between pH 7.4 and pH 6.0. This could be attributed to the change in pH affecting the streptavidin-biotin solution, or could be due to an artifact of the assay. Further studies would be needed to discern between the two hypotheses. However, there was a pronounced increase in the apparent KD of the triple mutant in a solution of pH 6.0 compared to pH 7.4. These results indicate the combination of mutations led to an observed pH-dependent binding. Further experimentation is needed with the single mutants to elucidate which residue is responsible for pH dependence, or if it is a combination of the three. Conclusions: We employed a rational mutation strategy to generate pH-dependent binding of an anti-insulin scFv. Upon a change in solution pH from 7.4 to 6.0, the triple mutant’s binding equilibrium is shifted, leading to an increase in KD. Further studies are needed to understand the reversibility of this interaction as well as to elucidate what specific residue interactions are contributing to this change.
Read full abstract