Higher-Order Structure of Adeno-Associated Virus Serotype 8 by Hydrogen/Deuterium Exchange Mass Spectrometry.

The higher-order structure (HOS) is a critical quality attribute of recombinant adeno-associated viruses (rAAVs). Evaluating the HOS of the entire rAAV capsid is challenging because of the flexibility and/or less folded nature of the VP1 unique (VP1u) and VP1/VP2 common regions, which are structural features essential for these regions to exert their functions following viral infection. In this study, hydrogen/deuterium exchange mass spectrometry (HDX-MS) was used for the structural analysis of full and empty rAAV8 capsids. We obtained 486 peptides representing 85% sequence coverage. Surprisingly, the VP1u region showed rapid deuterium uptake even though this region contains the phospholipase A2 domain composed primarily of α-helices. The comparison of deuterium uptake between full and empty capsids showed significant protection from hydrogen/deuterium exchange in the full capsid at the channel structure of the 5-fold symmetry axis. This corresponds to cryo-electron microscopy studies in which the extended densities were observed only in the full capsid. In addition, deuterium uptake was reduced in the VP1u region of the full capsid, suggesting the folding and/or interaction of this region with the encapsidated genome. This study demonstrated HDX-MS as a powerful method for probing the structure of the entire rAAV capsid.

Characterization of Adeno-Associated Virus Capsid Proteins with Two Types of VP3-Related Components by Capillary Gel Electrophoresis and Mass Spectrometry.

Recombinant adeno-associated virus is a leading platform in human gene therapy. The adeno-associated virus (AAV) capsid is composed of three viral proteins (VPs): VP1, VP2, and VP3. To ensure the safety of AAV-based gene therapy products, the stoichiometry of VPs of AAV vector should be carefully monitored. In this study, sodium dodecyl sulfate-polyacrylamide gel electrophoresis, capillary gel electrophoresis (CGE), and liquid chromatography-UV-mass spectrometry (LC-UV-MS) were performed to evaluate the VP components of AAV1, AAV2, and AAV6. Two types of VP3-related components, VP3 variant and VP3 fragment, were identified. The VP3 variant was the N-terminal shorter VP3, of which the translation started at M211, not at the conventional initiation codon, M203. The VP3 variant could be generated by leaky scanning of the first initiation codon of VP3. We also showed that the VP3 variant was identified in a minor peak before VP3 in CGE measurement. Meanwhile, the VP3 fragment was the C-terminal cleaved VP3, of which the sequence of VP3 ended at D590 or D626, indicating that cleavage occurred between D590 and P591, or D626 and G627. The cause of the cleavage of the DP or DG sequence was hydrolysis due to low pH of the mobile phase and high temperature of the column oven in the LC system, which was necessary to clearly separate the peak of VPs. VP3 fragments, detected only in LC-UV-MS in small amount account with less than 3% of total peak area, should be included in the quantification of VP3. Finally, the relationship of VP stoichiometry determined by the above three methods was discussed. From this study, we proposed that the VP components of AAV should be complementarily evaluated by CGE and LC-UV-MS.

Relation of Colloidal and Conformational Stabilities to Aggregate Formation in a Monoclonal Antibody.

Aggregation of therapeutic monoclonal antibodies has a potential risk of immunogenicity, requiring minimization of aggregate formation. We have developed a fitting formula for antibody aggregation at 40°C based on physicochemical parameters, including colloidal and conformational stabilities. An IgG1 monoclonal antibody, MAb-T, was formulated in 24 combinations of different buffer types and pH with or without sodium chloride. The fitting formula for monomer loss was successfully established by nonlinear regression analysis of the results from accelerated stability testing. Calculated monomer fraction values by the fitting formula were strongly correlated with experimental values (R2 = 0.92). The model includes secondary virial coefficient, B22, as the representative parameter of colloidal stability, and aggregation temperature, Tagg, representing conformational stability. Then, we examined charge state, conformational flexibility, and thermal unfolding profile of MAb-T to clarify the molecular basis for the different aggregation propensities in sodium acetate buffer and in sodium citrate buffer at the same pH and buffer concentration. We concluded that the accumulation of citrate anions on the surface of MAb-T is the primary source of the less colloidal and conformational stabilities, resulting in the higher aggregation propensity in sodium citrate buffer.

Identification of IgG1 Aggregation Initiation Region by Hydrogen Deuterium Mass Spectrometry.

Antibody aggregates are a potential risk for immunogenicity; therefore, rational approaches to improve associated aggregation properties need to be developed. Here, we report the amino acid region responsible for aggregation initiation. Two types of therapeutic IgG1 antibody monomer samples were prepared: IgG1 mAb40-3M stored at 40°C for 3 months, which existed in monodisperse state, and the monomer mAb65-5m, which was dissociated from small soluble aggregates by heating at 65°C for 5 min. Hydrogen deuterium exchange mass spectrometry of mAb40-3M identified 2 sites in the Fc region (site 1, F239-M256; site 2, S428-G450) with increased exchange rates. Site 1 includes a region reported as being susceptible to structural change induced by stress. Exposure of site 1 was undetected after 2 months of storage at 40°C but was subsequently detectable after 3 months. As site 2 is spatially close to site 1, the structural change of site 1 could propagate site 2. Besides these 2 regions, hydrogen deuterium exchange mass spectrometry of mAb65-5m identified an exposure of I257-W281 in Fc (site 3), within which a peptide sequence with high aggregation tendency was discovered. We thus concluded that exposure of site 3 is a trigger for the association of a partially denatured antibody.