Center for Proteomics and Metabolomics, Leiden University Medical Center, Leiden, the Netherlands.
Center for Proteomics and Metabolomics, Leiden University Medical Center, Leiden, the Netherlands.
Anal Chim Acta. 2021 Nov 1;1184:339015. doi: 10.1016/j.aca.2021.339015. Epub 2021 Sep 2.
Characterization of post-translational modifications (PTMs) of therapeutic antibodies is commonly performed by bottom-up approaches, involving sample preparation and peptide analysis by liquid chromatography-mass spectrometry (LC-MS). Conventional sample preparation requires extensive hands-on time and can increase the risk of inducing artificial modifications as many off-line steps - denaturation, disulfide-reduction, alkylation and tryptic digestion - are performed. In this study, we developed an on-line multidimensional (mD)-LC-MS bottom-up approach for fast sample preparation and analysis of (formulated) monoclonal antibodies and antibody-derived therapeutics. This approach allows on-column reduction, tryptic digestion and subsequent peptide analysis by RP-MS. Optimization of the 1D -and 2D flow and temperature improved the trapping of small polar peptides during on-line peptide mapping analysis. These adaptations increased the sequence coverage (95-98% versus 86-94% for off-line approaches) and allowed identification of various PTMs (i.e. deamidation of asparagine, methionine oxidation and lysine glycation) within a single analysis. This workflow enables a fast (<2 h) characterization of antibody heterogeneities within a single run and a low amount of protein (10 μg). Importantly, the new mD-LC-MS bottom-up method was able to detect the polar, fast-eluting peptides: Fc oxidation at Hc-Met-252 and the Fc N-glycosylation at Hc-Asn-297, which can be challenging using mD-LC-MS. Moreover, the method showed good comparability across the different measurements (RSD of retention time in the range of 0.2-1.8% for polar peptides). The LC system was controlled by only a standard commercial software package which makes implementation for fast characterization of quality attributes relatively easy.
治疗性抗体的翻译后修饰(PTMs)的表征通常通过自下而上的方法进行,涉及通过液相色谱-质谱(LC-MS)进行样品制备和肽分析。传统的样品制备需要大量的手工操作时间,并且由于许多离线步骤(变性、二硫键还原、烷基化和胰蛋白酶消化)的进行,会增加诱导人工修饰的风险。在这项研究中,我们开发了一种在线多维(mD)-LC-MS 自下而上的方法,用于快速制备和分析(配方)单克隆抗体和抗体衍生的治疗药物。该方法允许在柱上进行还原、胰蛋白酶消化和随后的 RP-MS 肽分析。优化一维和二维流动和温度改善了在线肽图谱分析中小极性肽的捕获。这些改进提高了序列覆盖率(95-98%,而离线方法为 86-94%),并允许在单次分析中鉴定各种 PTM(即天冬酰胺的脱酰胺、蛋氨酸氧化和赖氨酸糖化)。该工作流程能够在单个运行中快速(<2 h)表征抗体异质性,并且只需少量蛋白质(10 μg)。重要的是,新的 mD-LC-MS 自下而上方法能够检测到极性、快速洗脱的肽:Fc 在 Hc-Met-252 处的氧化和 Fc N-糖基化在 Hc-Asn-297 处的氧化,这在使用 mD-LC-MS 时可能具有挑战性。此外,该方法在不同测量之间表现出良好的可比性(极性肽的保留时间 RSD 在 0.2-1.8%范围内)。LC 系统仅由标准商业软件包控制,这使得相对容易实现快速表征质量属性。