The Modification of Single Domain Antibody Framework and Variable Regions to Impart Novel Pharmacodynamic and Pharmacokinetic Properties

Abstract

In recent years, nanobodies have become important theranostic molecules, and research has focused on improving their pharmacokinetics and binding for novel targets. Nanobodies are characterized by their high kidney uptake, leading to potentially unsafe exposure of the kidneys when delivering cytotoxic payloads. Researchers hypothesize that positively charged amino acids contribute to this kidney uptake. To address this, we assessed the impact of substituting lysine residues within the 2Rs15d HER2 nanobody scaffold to alanine on its in vivo pharmacokinetics using positron emission tomography (PET) imaging. We found that nanobodies bearing lysine-to-alanine substitutions maintained their affinity for HER2. Characterization of the K88A variant showed a 60% decrease in the kidney uptake of the tracer compared to the parent ⁶⁴Cu-NOTA-2Rs15d. Another variant, K62-66A, was similarly assessed in mice bearing HER2⁺ tumors. We found that ⁶⁴Cu-NOTA-K62-66A reduced kidney uptake by 35% while having no effect on the accumulation of the tracer within the tumor, indicating that scaffold charges within the nanobody scaffold exert significant influence on the pharmacokinetics of nanobodies and should be explored to determine the extent of this effect. Novel nanobodies are typically developed by the immunization of camelids to create immune libraries. However, this technique has drawbacks, such as the inability to generate nanobodies for nonimmunogenic antigens and the lack of control over framework sequences encoded by the germline. To circumvent these issues, we developed a modular phage display nanobody vector based on the 2Rs15d nanobody. We generated a library of nanobodies with a randomized CDR2 with 1.1x10⁶ unique amino acid sequences/µL. We panned this library for nanobodies binding calreticulin, a proof-of-concept target known for its role in the immunogenic cell death of cancer cells. A clone, Cal3, has a 37nM affinity for calreticulin and the ability to bind cellular calreticulin. We labeled Cal3 with ⁶⁴Cu for the PET imaging of calreticulin and were able to detect changes in calreticulin in the tumors of treated mice, indicating that synthetic nanobody libraries are versatile and can generate nanobodies with in vivo efficacy. This body of work demonstrates that engineering the framework and binding regions of nanobodies can yield molecules with theranostic impact.