All about Clustering in Protein Solutions

                                                                         

Prasad Sarangapani1,2, Jai Pathak1, Steven D. Hudson2, Ronald Jones2, and Kalman Migler2

 

                        1MedImmune, One MedImmune Way, Gaithersburg, MD 20878

2Materials Science and Engineering Division, National Institute of Standards and Technology, Gaithersburg, MD 20899

 

 

Equilibrium and non-equilibrium clustering are ubiquitous phenomena in soft matter physics and are typically observed in systems ranging from colloidal suspensions to solutions of monoclonal antibodies (mAbs). Such phenomena are central to understanding and preventing irreversible aggregation in addition to controlling viscosity challenges related to formulation and drug delivery of protein therapeutics. Curiously, little is known regarding the cluster size dependence of low-shear viscosity and intrinsic viscosity in protein solutions in a controlled manner. In this work, we carefully tune cluster size of reversible and irreversible clusters formed by globular proteins or monoclonal antibodies over a concentration range of 2 mg/mL-500 mg/mL and pH from 3-7.4.  The reversible clusters are formed via depletion interactions by the addition of appropriate amounts of crowding agents such as Trehalose and Poly(ethylene glycol). Irreversible clustering is induced upon addition of urea over a range of 2 M - 8 M. We find a marked dependence of low-shear viscosity on cluster size using custom-designed silicon-based microfluidic viscometers fabricated at the NIST NanoFab facility. Measurements of cluster sizes using SANS reveal a correlation of low shear viscosity as well as intrinsic viscosity with the average cluster size. Classical models of the concentration (volume fraction) dependence of viscosity from the colloid rheology literature are found to fail for clustered globular protein and antibody solutions. We attempt to model the composition dependence of viscosity for the case of equilibrium and non-equilibrium clusters using an adaptation of a model recently presented by Minton for protein mixtures.