Protein stability at high concentrations (200-400 mg/ml) is of broad interest in science and human health. In cellular environments, a variety of macromolecules crowd proteins into in a compact state, stabilizing folded proteins, as described theoretically and observed experimentally. However, when trying to achieve such high concentrations in vitro, short-ranged attractive and other interactions between closely spaced proteins often produce irreversible aggregation and precipitation. In vitro, these undesirable events dominate, thus novel concepts are needed to mimic nature successfully to stabilize proteins at high concentration for protein drug delivery applications. We have created concentrated antibody protein dispersions in the form of dense, 50-400 nm, reversible clusters of protein molecules, by addition of macromolecular crowding agents. The antibody molecules are conformationally stabilized by crowding, retaining therapeutic activity in vitro and exhibiting expected pharmacokinetics in vivo. Whereas the locally concentrated environment within nanoclusters stabilizes the native state of protein molecules, weak interactions between nanoclusters result in colloidally-stable, translucent dispersions with low viscosities. Not only i the protein native structure maintained within the nanocluster, but upon dilution in vitro, the clusters reversibly dissociate into native monomeric protein molecules with high biological activity. The viscosities of antibody nanocluster dispersions are sufficiently low to allow subcutaneous injection to mice at typical therapeutic antibody dosages. Building on our promising initial work, we aim here to characterize nanocluster fate in vivo, after subcutaneous injection. We will perform three studies crucial for further development of this technology, using large (~300 nm diameter) and small (<100 nm) nanoclusters and solution control antibody. First, we will characterize the mechanism and rate of in vivo cluster dissociation, using fluorescently-labeled antibodies and whole animal imaging. Second, we will compare induction of immune responses in terms of anti-drug antibodies bi-weekly doses of antibody nanoclusters or monomers. Third, we will perform an in vivo efficacy study to compare protection against whooping cough when mice are dosed with nanoclusters or monomers of 1B7, a known protective antibody. The reversible antibody nanoclusters described here may represent a transformative advance to enable patient self-administration of biological pharmaceuticals and provide insight into protein stability in vivo.
Protein-based drugs represent some of the most promising therapies for a wide range of diseases, including cancer. Subcutaneous injection is the preferred method of delivery, but its usefulness is currently limited by unwanted outcomes such as protein aggregation and gelation that occur at the high concentrations necessary to provide a full dose in the 1.5 ml volume compatible with sub-cutaneous injection. Previous attempts to address these problems by modifying the amino acid sequence of potential therapeutics have been expensive and often unsuccessful. The investigators have recently reported a new method for creating highly concentrated, low-viscosity dispersions of stable protein nanoclusters that could provide a basis to solve this major challenge. However, at present, the answers to basic questions about behavior of protein nanoclusters in vivo and their potential to induce unwanted immune responses are unknown. This project will quantitatively measure in vivo dissociation kinetics, compare antibody and T cell responses upon repeated dosages and demonstrate efficacy of antibodies delivered as nanoclusters to treat disease.
|Laber, Joshua R; Dear, Barton J; Martins, Matheus L et al. (2017) Charge Shielding Prevents Aggregation of Supercharged GFP Variants at High Protein Concentration. Mol Pharm 14:3269-3280|