Real-Time Assessment of the Size Changes of Individual Sub-Visible Protein Particles under Buffer Variations: A Microfluidic Study

Abstract

Protein particles in biological drugs can significantly impact drug efficacy and carry the risk of adverse effects. Despite advancements, the understanding and control of particle formation in biopharmaceutical manufacturing remain incomplete. Therefore, further investigation into protein particles is warranted, especially considering that novel formats of biological drugs may be more susceptible to aggregation and particle formation than conventional monoclonal antibodies. In this study, we introduce a microfluidic approach for the real-time analysis of individual sub-visible protein particles during buffer exchange. We find that the modulation of intermolecular forces, achieved by changing the buffer pH or urea concentration, leads to the reversible swelling and shrinkage of particles by up to 50%, which is a consequence of altered intermolecular distances. Additionally, we identify a discrepancy in the biophysical behavior of protein particles compared to monomeric protein. This finding highlights the limited predictive power of commonly applied biophysical characterization methods for particle formation in early formulation development. Moreover, the observed particle swelling may be associated with manufacturing deviations, such as filter clogging. These results highlight the importance of studying individual particles to gain a comprehensive insight into particle behavior and the impact of formulation variations in the biopharmaceutical industry.

Keywords: aggregation; biologics; intermolecular forces; microfluidics; monoclonal antibody; optical tweezers.

Figure 1

Experimental set-up. (a) A typical sub-visible protein particle prepared by stirring a solution of mAbs. The particles can be moved and rotated with optical tweezers as quasi-rigid bodies (Video S1). The scale bar indicates 50 μm. (b) Schematic representation of the experiments in the diffusion chamber. The chamber is a microcavity extending from the main channel of the microfluidic system. The length of the cavity is 250 μm, its width is 100 μm, and its depth is 40 μm. Protein particles are transferred into the chamber by optical tweezers and the buffer surrounding the particles in the chamber is then exchanged by diffusion from the main channel. The buffer in the main channel can be exchanged by a standard pipette. As the chamber is effectively flow-free, the particles remain undisturbed and can be monitored in real time during repeated buffer exchange.

Figure 2

Response of protein particles to altered electrostatic interactions. (a) Three representative protein particles in the microfluidic diffusion chamber at the beginning of the experiment at pH 7 (left) and after the pH was changed to 4 (right). Particle size was quantified as the surface area of the particle outline (A) relative to the initial state (A₀). Scale bars correspond to 50 μm. (b) Dependence of the relative size of protein particles on the value of pH in the absence of NaCl. All the changes were reversible, except at pH 3, at which the particles irreversibly decayed. (c) Dependence of the relative size of protein particles on the concentration of NaCl at pH 4.

Figure 3

Effect of pH on the protein-protein interaction coefficient (k_D) as measured by dynamic light scattering; k_D is a measure of protein–protein interaction: k_D > 0 repulsive interaction, k_D < 0 attractive interaction.

Figure 4

Response of protein particles to urea. (a) Dependence of the relative particle size on urea concentration at pH 7 (blue) and pH 4 (yellow). (b) Reversible changes in particle sizes as the buffer was cyclically changed from 0 M urea at pH 7 to 0 M urea at pH 4, 3 M urea at pH 4, 6 M urea at pH 4, and, finally, to 9 M urea at pH 4. (c) Rapid changes between 0 M and 9 M urea are reversible, but if the particles are left in 9 M urea for a longer time, they disintegrate irreversibly. The scale bars represent 50 μm.