(M1230-09-52) Ranking mAb-excipient Interactions in Biologics Formulations by NMR and Molecular Dynamics

Purpose: Excipients are added to biopharmaceutical formulations for enhancing protein stability and developing robust formulations which have acceptable physicochemical properties, but the mechanism by which they confer stability remains not fully understood¹. In this study, we aimed to elucidate the mechanism, by providing direct experimental evidence to binding affinity of different excipients (sucrose, trehalose, mannitol, sorbitol, succinic acid and glycine) to a monoclonal antibody (mAb) and ranking a series of excipients with respect to their dissociation constant (K_D) and non-specific binding constants (N_s). Our finding will aid with excipient selection in biologic formulations, by providing insights into mAb-excipient affinities before conducting a conventional excipient screening study which is time consuming.

Methods: BMS mAb stock solution and excipient (sucrose, trehalose, mannitol, sorbitol, succinic acid and glycine) were prepared. BMS mAb-excipient series with constant concentration of BMS mAb and different concentration of excipients were prepared for saturation transfer difference (STD) nuclear magnetic resonance (NMR) spectroscopic method to rank the excipients. Molecular dynamic (MD) simulations were done in parallel, to rank the excipient proximities to the proteins and thereby corroborating the ranking by STD NMR. Finally, the excipient ranking by NMR was correlated to mAbs conformational stability (thermal stability, Tm and colloidal stability, B22).

Results: K_D and N_s were determined by monitoring the excipient STD NMR signals as a function of the excipient concentration. The STD amplification factor (STD-AF)² was calculated first. The experimental buildup curves of STD-AF with respect to ligand concentration were first fitted with the Equation containing both single-site binding and non-specific binding. If failed, they were fitted with the Equation containing only non-specific binding. Among all the excipients, only the experimental data for sucrose is well fitted to single-site binding at both 283 K and 293 K. As for other excipients, the buildup curves are fitted to non-specific binding (Figure 1a). The N_s values for all six excipients were calculated and ranked. The excipient ranking by NMR was validated by correlating the ranking to both the Tm and B22, as well as MD simulation. The ranking of excipients all the methods (Figure 1b) are in good agreement that the sucrose is the highest, followed by trehalose and other excipients. STD-NMR experiment has also been used to characterize the binding site of ligand to protein complexes since the strongest contact of the ligand to the protein shows the most intense NMR signals, enabling the binding epitope mapping^2,3. Sucrose, mannitol and trehalose were studied because their protons can be distinguished in 1D ¹H NMR experiments. For sucrose, H7 is the most involved proton in binding and the closest to mAb, followed by H13, H12 and H10. Other protons are more distant from mAb and more solvent exposed. For mannitol, either H3 or H11 is the most involved proton to the mAb binding site. As for trehalose, multiple protons were overlapped in peak 3,4 and 5, but the general trend is still in good agreement. The results from MD simulation greatly support our statements above (Figure 2).

Conclusion: Our STD-NMR results show that sucrose is the best excipient to BMS mAb in study, followed by trehalose, mannitol and sorbitol. The strong correlation between degree of binding by STD NMR, Tm, and B22 for the different excipients suggests that excipient binding is likely causing improved protein stability. The closer proximity of the excipient to the mAb also provides better conformational stability to the mAb. This approach of ranking the excipients using STD-NMR experiments and MD simulations has aided and accelerated the excipient selection in biologic formulations by providing direct evidence into mAb-excipient affinities before conducting a conventional excipient screening study which is long and time consuming. And by correlating to Tm and B22 values and therefore to the protein stability of future mAbs, this approach can perhaps direct the study to merely using the ranking by MD simulations for excipient selection in protein formulations.

References: 1. T. J. Kamerzell, et al. Advanced Drug Delivery Reviews 2011; 63:1118-1159.
2. M. Mayer, et al. Journal of the American Chemical Society 2001; 123:6108-6117.
3. A. Viegas, et al. Journal of Chemical Education 2011; 88:990-994.

Acknowledgments: Authors would like to thank Rob Garmise, Janet Cortes, Tracy Gabele and Arvid Mathur for this collaboration opportunity.

Figure 1: (a) Fitting plots for all six excipients. (b) Ranking of excipients by Ns, Tm and MD.



Figure 2: Binding site of sucrose(left), mannitol(middle) and trehalose(right) with BMS mAb characterized by STD effect analysis and MD simulation.