Saldanha Marianne, Shelar Ashutosh, Patil Vaibhav, Warke Vishal G, Dandekar Prajakta, Jain Ratnesh
Department of Biological Sciences and Biotechnology, Institute of Chemical Technology, Matunga, Mumbai, 400019 India.
Shimadzu Analytical (India) Private Limited, Rushabh Chambers, Marol, Andheri East, Mumbai, 400059 India.
Cytotechnology. 2023 Feb;75(1):77-91. doi: 10.1007/s10616-022-00561-z. Epub 2022 Nov 23.
In this case-study, we demonstrate an approach for identifying correlations between nutrients/metabolites in the spent medium of CHO cell cultures and cell growth, mAb titre and critical quality attributes, using multivariate analyses, which can aid in selection of targets for medium and feed optimization. An extensive LC-MS-based method was used to analyse the spent medium composition. Partial least squares (PLS) model was used to identify correlations between nutrient composition and cell growth and mAb titre and orthogonal projections to latent structures (OPLS) model was used to determine the effect of the changing nutrient composition during the culture on critical quality attributes. The PLS model revealed that the initial concentrations of several amino acids as well as pyruvic acid and pyridoxine, governed the early cell growth, while the concentrations of TCA cycle intermediates and several vitamins highly influenced the stationary phase, in which mAb production was maximum. For the first time, with the help of the OPLS model, we were able to draw correlations between nutrients/metabolites during the culture and critical quality attributes, for example, optimizing the supply of certain amino acids and vitamins could reduce impurities while simultaneously increasing desirable glycoforms. The unique correlations obtained from such an exploratory analysis, utilizing conditions that are commonly adopted in early process development, present opportunities for optimizing the compositions of the growth media and the feed media for enhancing cell growth, mAb production and quality, thereby proving to be a useful preliminary step in bioprocess optimization.
The online version contains supplementary material available at 10.1007/s10616-022-00561-z.
在本案例研究中，我们展示了一种方法，通过多变量分析来识别CHO细胞培养物的废培养基中营养物质/代谢物与细胞生长、单克隆抗体滴度和关键质量属性之间的相关性，这有助于选择用于培养基和补料优化的靶点。使用了一种基于液相色谱-质谱联用的广泛方法来分析废培养基的组成。偏最小二乘法（PLS）模型用于识别营养成分与细胞生长和单克隆抗体滴度之间的相关性，而正交投影到潜在结构（OPLS）模型用于确定培养过程中营养成分变化对关键质量属性的影响。PLS模型显示，几种氨基酸以及丙酮酸和吡哆醇的初始浓度控制着早期细胞生长，而三羧酸循环中间体和几种维生素的浓度对稳定期有很大影响，在稳定期单克隆抗体产量最高。借助OPLS模型，我们首次能够得出培养过程中营养物质/代谢物与关键质量属性之间的相关性，例如，优化某些氨基酸和维生素的供应可以减少杂质，同时增加理想的糖型。利用早期工艺开发中常用的条件，从这种探索性分析中获得的独特相关性为优化生长培养基和补料培养基的组成以促进细胞生长、单克隆抗体生产和质量提供了机会，从而证明是生物工艺优化中一个有用的初步步骤。
在线版本包含可在10.1007/s10616-022-00561-z获取的补充材料。
