(M1130-09-55) Unraveling the Impact of Ionic Strength on Antibody Self-Association Using a Novel QCM-D Approach

Purpose: Therapeutic monoclonal antibodies (mAbs) form the cornerstone of modern biologics and are critical for the treatment and management of numerous diseases. However, the colloidal behavior and long-term stability of these drug products can be compromised by protein self-association, elevated viscosity, and aggregation. These physicochemical liabilities can affect multiple stages in the lifecycle of a mAb product, including development, manufacturing, and in-vivo performance. As such, understanding and mitigating the interactions and risks between protein molecules is critical for developability assessments and formulation optimization. Among the existing tools for quantifying protein–protein interactions (PPI), the diffusion interaction parameter (k_D) from dynamic light scattering (DLS), is a widely used biophysical gold standard in academic and industrial laboratories (Kingsbury et al., 2020). However, considering the sensitivity of k_D to sample conditions and experimental parameters, there remains a need for orthogonal approaches that provide complementary insights into stability under formulation-relevant conditions. In this regard, a recent study has explored the use of quartz crystal microbalance with dissipation monitoring (QCM-D) to measure the frequency shift (Δf) associated with the loosely bound layer (LBL) of mAbs adsorbed on a sensor, as a surrogate for self-association (Rahman et al., 2024). While this work offers promise for optimal candidate selection, its potential for formulation screening, specifically how excipient-induced changes in LBL correlate with solution behavior, remains unexplored. Thus, the present study explores the application of QCM-D to quantify the self-association behavior of a model mAb across varying ionic strengths in a histidine buffer system with NaCl.

Methods: Omalizumab was extracted and purified from the commercialized product and buffer-exchanged into 10.0 mM histidine buffer (pH 6.0) to prepare a stock solution of 10.0 mg/mL. The mAb solution was then filtered and diluted in the histidine buffer system containing 0, 80, or 160 mM NaCl. For QCM-D experiments, the gold-coated sensors were cleaned by a standard protocol and each run comprised three key phases: baseline stabilization in the buffer system; 5.0 mg/mL protein injection until steady-state adsorption and stabilization; and rinsing with the respective buffer to remove LBL. The third overtone of the QCM-D measurements was used for primary analysis, and the LBL was quantified by subtracting the Δf of the irreversibly adsorbed layer after rinse from that of the total adsorbed layer at steady state. All experiments were performed in triplicate with freshly prepared samples and filtered buffers. To compare these surface-sensitive biophysical readouts with solution-based PPI metrics, DLS was utilized to determine k_D, and static light scattering (SLS) was employed to calculate the second virial coefficient (A₂) for assessing bulk thermodynamics. Light scattering experiments were performed simultaneously using 96-well flat-bottom microplates and a DynaPro Plate Reader III. A concentration series of Omalizumab was prepared by diluting the stock solution to obtain samples ranging from 1.0 to 10.0 mg/mL in 1.0 mg/mL increments for each ionic strength condition. A volume of 80 μL of each concentration was loaded into individual wells, and 10 acquisitions were recorded over 5 seconds for each measurement. The data were analyzed using Dynamics software, where the autocorrelation function of scattered light intensity was fitted using the method of cumulants to determine k_D, and the reduced Rayleigh ratio was plotted against mAb concentration to calculate A₂.

Results: Among the various factors governing mAb self-association, ionic strength plays a critical role by modulating electrostatic interactions between individual molecules, and accordingly, all three techniques showed consistent ionic strength–dependent trends. A₂ became progressively more negative with increasing salt concentration (–5.8×10^-5 ± 6.9×10^-6, –8.0×10^-5 ± 4.4×10^-6, and –9.5×10^-5 ±  2.1×10^-5 mol·mL/g² for 0, 80, and 160 mM NaCl, respectively), signifying enhanced thermodynamic attraction between solute molecules. Similarly, k_D became more negative from 0 to 80 mM (–16.6 ± 0.3 to –29.8 ± 0.3 mL/g), but showed a partial reversal at 160 mM (–23.7 ± 0.1 mL/g), which may signify that higher salt content dampens some shorter-range attractive forces. QCM-D mirrored these trends with a sharp increase in LBL from –24.7 ± 11.3 Hz at 0 mM to –59.3 ± 13.1 Hz at 80 mM suggesting that ionic strength promotes more loosely associated mAb at the interface, followed by a drop to –41.6 ± 8.7 Hz at 160 mM. These results indicate that the addition of NaCl initially screens repulsive electrostatic interactions, elevating self-association, while further increases in salt concentration suppress short-range attractions as well, leading to decreased Omalizumab association.

Conclusion: The present study applies a novel QCM-D–based strategy to quantify ionic strength–dependent self-association of mAbs, complementing established light scattering techniques. The interpretation of LBL provides mechanistic insight into the impact of ionic strength and electrostatic screening on interfacial-mediated PPI behavior that parallels trends observed in k_D and to some extent A₂. Together, these findings establish QCM-D as a formulation screening tool, offering a new lens for multi-dimensional evaluation of antibody behavior and stability.

References: (1) Kingsbury JS et al. A single molecular descriptor to predict solution behavior of therapeutic antibodies. Science Advances. 2020.
(2) Rahman Y et al. Prediction of self-association and solution behavior of monoclonal antibodies using the QCM-D metric of loosely interacting layer. Molecular pharmaceutics. 2024.

Acknowledgements: This project is supported by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under Award Number R21AI178218. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.