Monoclonal antibodies (mAbs) used as biological drugs are a class of biopharmaceuticals that has rapidly expanded over the past three decades. The majority of therapeutic products on the market that are based on mAbs are liquid formulations, in which the mAbs can have complex degradation pathways that are difficult to predict. Chemical denaturation is an alternative approach to thermal stress for determination of conformational stability (denaturation) of proteins (such as mAbs) in solution. The purpose of this project was to investigate the conformational changes of a model IgG1 mAb in the presence of chemical denaturants using both experimental and theoretical approaches. Moreover, we wanted to improve upon the stability-testing approaches in the literature for mAbs through the determination of faster and easier approach(es) to indicate differences in mAb stabilities. We studied the stability of the model mAb under the influence of two widely used denaturants (guanidine hydrochloride [GuHCl], urea) and three alternative denaturants (guanidine thiocyanate, N-methyl-urea, N-ethyl-urea), both experimentally and mathematically. We measured the changes in the intrinsic fluorescence signals due to the unfolding of the mAb
under different denaturant concentrations at pH 4.5 to pH 8.5 over 24 h and 15 days. From the denaturation curves (intrinsic fluorescence versus denaturant concentration and time), we calculated the denaturant-concentration-independent changes in the Gibb’s free energy (ΔG) and the half times (t1/2) of unfolding, again, experimentally and through mathematical modelling. To evaluate the predictions of these stability-indicating methods, we also performed accelerated stability studies at 40 °C over 12 weeks. GuHCl was the only investigated denaturants that provided complete data and can be used as a denaturant in buffer solutions of this mAb. By tracking the kinetics of the unfolding over the
first 24 h, we show that t1/2 correlates with ΔG from chemical denaturation curves, with a Pearson’s correlation coefficient of 0.78. The formulation stability ranking based on t1/2 required a third of the protein used in the approach where ΔG was calculated, and provided data with smaller standard deviations. This approach using the calculation of t1/2 can therefore substitute for the standard ΔG calculations. When the chemical denaturation was tracked over 15 days, shifts were seen for the intrinsic fluorescence signals in the transitional regions of the denaturation curves. Dynamic light scattering measurements showed that these are most likely due to the formation of aggregates. By fitting the intrinsic fluorescence signal changes to a pseudo first-order kinetic model of aggregation, the aggregational times (tagg) were obtained. Small shifts in the intrinsic fluorescence signals were also seen in the pre-transitional region at ~1 M GuHCl following incubations of up to 15 days. These shifts correlated well (Pearson’s correlation coefficient, 0.89) with the mAb monomer degradation rates in accelerated stability studies at 40 °C over 12 weeks. Measurements of the intrinsic fluorescence (F350/330) shifts therefore provide a novel stability-indicating tool to predict the outcomes of accelerated stability studies.