Mechanistic insights into the role of protein interactions on the aggregation behavior of anti-streptavidin immunoglobulin gamma-1
Date
2015
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
Monoclonal antibodies (mAbs) are one of the leading protein-based drug candidates in the biopharmaceutical industry. The formation of irreversible, non-native protein clusters (hereby called aggregates) is a common degradation route during manufacturing of mAbs and other therapeutic proteins. Aggregation may potentially decrease drug efficacy and jeopardize patient safety. A patient's immune system may mount a response against the therapeutic protein making future treatments ineffective and potentially dangerous. As such, controlling unwanted aggregation is an ongoing and crucial challenge in the development of protein-based therapeutics. The process(es) by which aggregates form or the aggregation mechanism(s) influence the aggregate size distribution, concentration, and structure, all of which may potentially impact drug potency and safety. A deeper understanding of the aggregation process may foster strategies to predict and control aggregation.
Thermally induced aggregation of anti-streptavidin immunoglobulin gamma-1 (AS-IgG1), a monoclonal antibody, is first mapped over a range of pH and NaCl concentration typical in formulated protein products. Aggregation mechanisms are influenced by low-concentration anions, such as acetate and citrate that are common buffer species. The relationships among monomeric protein-protein interactions, aggregate morphology, and aggregation mechanisms are explored. Colloidal interactions (i.e. potential of mean force) arguments are discussed in terms of AS-IgG1 aggregation mechanisms. Static light scattering and observed aggregation mechanisms suggest a citrate/acetate specific-ion-effect that cannot be explained with simplified colloidal interactions. The Kirkwood-Buff integral for protein-protein interactions semi-quantitatively predicts how changes in solution conditions change aggregation mechanisms, and may offer a practical tool to capture specific ion effects.
IgG1 aggregation rates are strongly affected by solution pH, ionic strength, and counter-ion species. Additionally, the rates are highly temperature dependent. For example, an increase of 5 degrees C often increases aggregation rates multiple orders of magnitude. As such, it is challenging to compare aggregation rates across a broad range of solution conditions and temperatures. A Parallel Temperature Initial Rates (PTIR) device and method are introduced to accurately and efficiently determine liquid-state degradation rates. IgG1 aggregation rates determined using PTIR compare well to the traditional isothermal approach. The PTIR method has the advantage of being more sample sparing and providing temperature dependent rates with more statistical certainty. Thermal unfolding and aggregation rates of AS-IgG1 were determined across multiple pH values and NaCl concentrations using differential scanning calorimetry (DSC) or PTIR. AS-IgG1 aggregation rate coefficients partially collapse to a common profile upon rescaling the incubation temperature by the midpoint unfolding temperature (Tm) determined via DSC. However, the effective activation energies (Ea) depend on solution pH, NaCl concentration, and citrate/acetate buffer species. Acetate vs. citrate specific-ion-effects manifest themselves in Tm and Ea values. The roles of protein-protein interactions and protein unfolding are discussed in the context of AS-IgG1 aggregation.
Osmolytes, such as sucrose, trehalose, and sorbitol, are commonly added to protein solutions to prevent unfolding and aggregation. These additive molecules are thought to compete with water for interactions with proteins resulting in so-called preferential interactions with proteins that can increase or decrease the chemical potential of proteins. However, direct measurements of protein preferential interactions remain challenging. AS-IgG1 protein-water and protein-osmolyte interactions are determined using precise density measurements of AS-IgG1 with aqueous solutions of sucrose, trehalose, sorbitol, and PEG (Mn= 6,000 g/mol). AS-IgG1 with sucrose and PEG show preferential interactions that depend on osmolyte concentration, which contradicts conventional wisdom for how these molecules interact with proteins in solution. Preferential interactions are compared to protein thermal unfolding using DSC. AS-IgG1 Tm values increase linearly as a function of sucrose, trehalose, and sorbitol concentrations, and decrease linearly as a function of PEG concentration. Results are compared to available models based on protein solvent exposed surface area (ASA) and discussed in terms of the classical preferential interaction theory. mAb formulations require intermediate to high protein concentrations due to dosing requirements of intravenous administration or subcutaneous injections. Elevated concentrations may promote non-ideal protein-protein and protein-osmolyte interactions. The framework for AS-IgG1 aggregation developed in earlier sections under dilute protein concentrations is extended to AS-IgG1 at 30 mg/mL in deuterated water (D2O) and in the presence and absence of 0.15 M (5 w/w %) sucrose. Protein structural techniques are monitored with circular dichroism, second derivative UV absorbance, and Raman spectroscopy. Results are compared to the aggregate morphology, which is monitored with size exclusion chromatography with in-line laser light scattering, dynamic light scattering, and small angle neutron scattering (SANS).
Structural changes suggest the nucleation-dominated mechanism (at low pD
and ionic strength) has significantly larger structural perturbations, as observed in the
disulfide bond conformation, secondary structure, and tyrosine and tryptophan environments.
The addition of 0.15 M sucrose decreases aggregation rates, particularly
for the nucleation dominated mechanism, but does not alter aggregation mechanisms.
AS-IgG1 monomer SANS structure factor (S(Q)) remains unchanged with the addition
of sucrose, suggesting sucrose does not alter protein-protein interactions. Aggregate
morphology is monitored with SANS, and the Kratky plot analysis shows a unique scattering
profile for each mechanism regardless of the presence of sucrose. These results
are consistent with the current framework that suggests protein-protein interactions
mediate aggregation mechanisms. Overall, results in this dissertation illustrate the effect
of protein-protein, protein-osmolyte, and protein-water interactions on the protein
stability and aggregation behavior of AS-IgG1. Many tools and approaches utilized in
this dissertation can be applied to degradation processes of other soft matter systems.