Model-driven Media and Process Optimization in Mammalian Cell Lines

描述（由申请人提供）：基于蛋白质的治疗产品对医疗保健做出了巨大贡献，并在总药物中占很大的比例，而且还在不断增长。这些FDA批准的产品大多数是使用哺乳动物细胞培养系统生产的。在过去的10-20年里，在克服大规模哺乳动物细胞培养的一些关键障碍方面取得了实质性进展。尽管有了这些改进，但开发新的生物制药产品仍然是一个昂贵而漫长的过程，其中总成本的20-30%与工艺开发和临床生产有关。目前，大多数工艺优化策略都是使用试错法来执行的，在这种方法中，单元被视为“黑匣子”，通过几个月的艰苦实验来改进工艺输出。这些经验优化技术被广泛使用，因为在大多数情况下，对影响宿主细胞系生长和蛋白质产生的潜在生理相互作用知之甚少。通过计算建模和模拟技术，对细胞系生理和代谢的基本理解可以极大地改善和加速哺乳动物细胞系系统的培养基和过程发展。在SBIR研究项目的第一阶段，我们利用我们的计算建模平台和在代谢建模和哺乳动物细胞培养方面的专业知识，与SAFC （Sigma-Aldrich Fine Chemicals）生物科学公司合作，减少了GS-NS0小鼠骨髓瘤细胞系的副产物形成。为了实现模型驱动的介质优化方法，我们重建了包含456种代谢物和470种代谢反应的NS0细胞代谢模型。利用该模型对基础培养基进行营养改良，以减少副产物的形成，提高生长和生产力。在NS0细胞培养中，我们基于模型的培养基配方的实验评估显示，与传统的培养基分析方法相比，有了显著的改进，乳酸浓度降低了约12%，最终产品滴度提高了67%。随着I期概念验证研究的成功完成，我们现在提出了II期提案，旨在完成媒体和工艺优化的商业平台的开发，从而显着改善与哺乳动物细胞系治疗性蛋白质生产相关的现有时间表。我们计划通过以下方式来实现这一目标：(1)改进和扩展NS0细胞系的代谢模型，(2)整合瞬时通量平衡方法来定量实施培养基设计，以及(3)通过三个案例研究来验证最终框架，用于NS0细胞系的抗体生产。我们将通过缩短开发优化培养基的时间来衡量第二阶段的总体成功，从而降低副产物的形成，提高细胞培养的生产率。成功完成拟议计划中概述的具体目标将通过合理选择营养补充和工艺优化策略，对模型驱动方法在重组蛋白生产中的商业价值进行基准测试。公共卫生相关性：基于蛋白质的治疗产品对医疗保健做出了巨大贡献，在整个制药市场中所占的比例很大，而且还在不断增长。这些FDA批准的产品大多数是使用哺乳动物细胞培养系统生产的。提出的工作旨在开发计算策略，显着改善与哺乳动物细胞中治疗性蛋白质生产相关的时间表和成本。降低治疗性蛋白质的开发和制造成本，将确保下一代药物能够大量生产，以满足患者的需求，并以患者能够负担得起的价格生产。

DESCRIPTION (provided by applicant): Protein-based therapeutic products have contributed immensely to healthcare and constitute a large and growing percentage of the total pharmaceutical drugs. The majority of these FDA approved products are manufactured using mammalian cell culture systems. Over the past 10-20 years substantial progress has been made to overcome some of the key barriers to large-scale mammalian cell culture. Despite these improvements, the development of new biopharmaceutical products remains an expensive and lengthy process, where 20-30% of the total cost is associated with process development and clinical manufacturing. Currently most process optimization strategies are performed using a trial and error approach where cells are treated as a `black box' and process outputs are improved over several months by laborious experimentation. These empirical optimization techniques are widely used because in most cases little is known about the underlying physiological interactions that impact growth and protein production in the host cell lines. A fundamental understanding of cell line physiology and metabolism, enabled by computational modeling and simulation technologies, can greatly improve and accelerate media and process development in mammalian cell line systems. In the Phase I of this SBIR research project, we utilized our computational modeling platform and expertise in metabolic modeling and mammalian cell culture to reduce byproduct formation in a GS-NS0 murine myeloma cell line in collaboration with SAFC (Sigma-Aldrich Fine Chemicals) Biosciences. To implement our model-driven media optimization approach, we reconstructed a metabolic model for NS0 cell containing 456 metabolites and 470 metabolic reactions. The model was used to develop nutritional modifications to the basal media to reduce byproduct formation and improve growth and productivity. Experimental evaluation of our model-based media formulations in NS0 cell culture showed significant improvements over traditional methods for media analysis and resulted in approximately 12% lower lactate and up to 67% higher final product titers. With the successful completion of our Phase I proof-of-concept study, we now put forth a Phase II proposal that aims at completing the development of a commercial platform for media and process optimization that significantly improves the existing timelines associated with therapeutic protein production in mammalian cell lines. We plan to achieve this goal through: (1) refinement and expansion of the metabolic model of NS0 cell line, (2) integration of a transient flux balance approach for quantitative implementation of media designs, and (3) validation of the final framework using three case studies for antibody production in NS0 cell line. We will measure the overall success in Phase II by our ability to reduce the timelines to develop an optimized media that results in lower byproduct formation and higher productivity in cell culture. Successful completion of the specific aims outlined in the proposed plan will benchmark the commercial value of a model-driven approach in recombinant protein production through rational selection of nutrient supplementation and process optimization strategies. PUBLIC HEALTH RELEVANCE: Protein-based therapeutic products have contributed immensely to healthcare and constitute a large and growing percentage of the total pharmaceutical market. The majority of these FDA approved products are manufactured using mammalian cell culture systems. The proposed work aims at developing computational strategies that significantly improves the timelines and cost associated with therapeutic protein production in mammalian cells. Reducing the cost of therapeutic protein development and manufacturing would ensure that the next generation of medicines can be created in amounts large enough to meet patients' needs and at a price low enough that patients can afford them.