Understanding the Stabilizing Effect of Histidine on mAb Aggregation: A Molecular Dynamics Study

Abstract

Histidine, a widely used buffer in monoclonal antibody (mAb) formulations, is known to reduce antibody aggregation. While experimental studies suggest a nonelectrostatic, nonstructural (relating to secondary structure preservation) origin of the phenomenon, the underlying microscopic mechanism behind the histidine action is still unknown. Understanding this mechanism will help evaluate and predict the stabilizing effect of this buffer under different experimental conditions and for different mAbs. We have used all-atom molecular dynamics simulations and contact-based free energy calculations to investigate molecular-level interactions between the histidine buffer and mAbs, which lead to the observed stability of therapeutic formulations in the presence of histidine. We reformulate the Spatial Aggregation Propensity index by including the buffer–protein interactions. The buffer adsorption on the protein surface leads to lower exposure of the hydrophobic regions to water. Our analysis indicates that the mechanism behind the stabilizing action of histidine is connected to the shielding of the solvent-exposed hydrophobic regions on the protein surface by the buffer molecules.

Molecular Dynamics Simulation of an Iron(III) Binding Site on the Fc Domain of IgG1 Relevant for Visible Light-Induced Protein Fragmentation

“…The radius of gyration ( R g ), an indicator of protein structure compactness, was used to evaluate whether Fe(III) inside the pocket would affect the structure of the entire Fc fragment. In the absence of Fe(III), the R g value of the Fc fragment is 2.56 ± 0.04 nm, which is similar to that of another Fc fragment (PDB code: 1HZH), 2.6 ± 0.09 nm . In the presence of Fe(III) in the pocket, the R g values are 2.60 ± 0.07 nm for the THR259 (0) state and 2.56 ± 0.04 nm for the THR259 (1−) state.…”

Solid-State NMR Spectroscopy to Probe State and Phase Transitions in Frozen Solutions

“…30 The titration curve of histidine (from low to high pH) reveals three pK a values (1.8, 6.0, and 9.2) resulting from sequential deprotonation of its carboxylic group, imidazole ring, and α-amine group, respectively (Figure 5). 31 Moreover, the neutral imidazole side chain exists in two tautomeric forms, the N ε2 -protonated τ-tautomer and the N δ1 -protonated π-tautomer. 32 Thus, the NMR chemical shifts of the relevant atoms are pH-dependent.…”

“…1 H− 31 P CP experiments were carried out on the HXY triplechannel probe in the H/P/C mode with the 31 P frequency at 162.0 MHz. 31 P CP experiments were performed with a CP mixing time of 2 ms and a linear ramp from 50% to 100%. For the crystalline DPDH powder and frozen histidine−NaP buffer solution (pH 7 with 50 mM histidine and 100 mM NaP), the number of scans were 32 and 512, respectively.…”

Optimizing Excipient Properties to Prevent Aggregation in Biopharmaceutical Formulations

“…38 Molecular dynamics (MD) simulations have been applied in the study of surfactant behavior in solution; 39,40 protein−surfactant interaction, 29 including stability modulation 41 and binding; 42 protein aggregation 43 and folding; 44 and the modulation of protein stability by excipients such as histidine. 45 Atomistic or pseudoatomistic MD models often have a prohibitively high computational expense to be applied in large numbers of simulations that examine microsecond-time scale events, such as many aspects of protein dynamics. 46 There are few investigations of the nonspecific interaction between excipients and APRs as a mechanism of aggregation prevention that considers all areas of the protein.…”

“…Computational techniques can provide mechanistic insights into length and time scales that are inaccessible to conventional wet lab methods . Molecular dynamics (MD) simulations have been applied in the study of surfactant behavior in solution; , protein–surfactant interaction, including stability modulation and binding; protein aggregation and folding; and the modulation of protein stability by excipients such as histidine . Atomistic or pseudoatomistic MD models often have a prohibitively high computational expense to be applied in large numbers of simulations that examine microsecond-time scale events, such as many aspects of protein dynamics …”