大规模哺乳动物细胞培养工程的现状与展望

Present situation and prospect for large-scale mammalian cell culture engineering

近年来,随着对疫苗和治疗性蛋白类药物等多种生物制品需求量增加以及产品质量要求的提高,细胞大规模培养技术也不断发展。为了增加产量、降低成本,生产更安全有效的药物,大规模细胞培养过程的开发至关重要,而动物细胞的工艺优化和规模放大具有挑战性。提高细胞培养工艺表达量、扩大细胞培养生产规模、保证表达抗体质量稳定成为目前大规模细胞培养过程中亟待解决的问题,迫切需要进一步研究和开发细胞培养工艺。本文围绕以上问题,系统综述了通过优良细胞株的构建、培养基设计与无血清培养基的开发、基于过程分析技术(PAT)培养工艺的优化与放大,建立合适的大规模培养体系,实现细胞的高密度培养和产物的高效表达。与此同时,细胞培养过程中产生的多源异质数据基本依靠低效的人工处理与判断,缺乏深层次的全局因素考虑。为此,未来希望通过人工智能深度挖掘数据之间的关系并指导细胞培养过程工艺优化与放大,实现真正的智能生物制造。

With requirements for the improvement of the quantity and quality of vaccines,therapeutic protein drugs and many other biological products,large-scale cell culture technologies continue to be developed.In order to increase production and reduce cost to produce safer and more effective drugs,the development of large-scale cell culture processes is essential,while the process optimization and scale up of animal cell culture are challenging.Ongoing research and development of animal cell culture technology are urgently required to increase the expression level of target poducts,to expand the scale of cell culture and to ensure stable product quality.This article focuses on above issues and systematically summarizes the establishment of suitable large-scale culture to achieve high-density and high efficiency production through the construction of excellent cell lines,the design of medium,particularly serum-free medium,as well as the optimization and scale-up of the culture process based on process analysis technology(PAT).Genomics and gene editing tools are frequently applied to construct an excellent cell line that supports high-density growth and secretes a large amount of therapeutic proteins.With the in-depth study of cell physiological and metabolic characteristics,serum-free culture can be designed based on multi-omics study,and through a variety of online sensing technologies to monitor and analyze the biological process,precise feedback control can thus be performed.On the other hand,the generation of a large amount of inconsistent data during the cell culture process is basically due to inefficient manual processing and judgment,and lacking of an in-depth consideration of global factors.Therefore,for the high efficiency industrial biomanufacturing process,it is necessary to visualize a large number of process parameters,establish a database of process parameters for subsequent big data analysis,and conduct deep learning and data mining to perform real-time bioprocess intelligent analysis,diagnosis and precise control,and then realize intelligent manufacturing.