National Institute of Bioprocessing Research & Training (NIBRT), Dublin 4, Ireland.
Lehrstuhl für Bioprozesstechnik, Otto-von-Guericke-Universität Magdeburg, Magdeburg, Germany.
Adv Biochem Eng Biotechnol. 2021;175:1-36. doi: 10.1007/10_2017_31.
The glycan profile of therapeutic recombinant proteins such as monoclonal antibodies is a critical quality attribute, which affects the efficacy of the final product. The cellular glycosylation process during protein expression is dependent upon a number of factors such as the availability of substrates in the media, the intracellular content of nucleotide sugars, and the enzyme repertoire of the host cells. In order to control the variability of glycosylation it is important to understand the critical process parameters and their acceptable range of values to enable reproducible production of proteins with a predetermined glycan profile providing the desired biological function or therapeutic effect. The depletion of critical nutrients such as glucose or galactose, which may occur toward the end of a culture process, can lead to truncated glycans. Terminal galactosylation and sialyation are particularly variable but may be controlled by the presence of some key media components. Ammonia accumulation, pH, and dissolved oxygen levels are also known to be key bioprocess parameters that affect the glycosylation of recombinant proteins. Specific enzyme inhibitors can be added to the media to drive the formation of selected and predetermined glycan profiles. Various attempts have been made to predict the glycan profiles of cellular expressed proteins and have led to metabolic models based upon knowledge of metabolic flux and the kinetics of individual glycosylation reactions.In contrast to single recombinant proteins, the glycan profiles of viral vaccines are far more complex and difficult to predict. The example of influenza A virus shows that hemagglutinin, the major antigenic determinant, has three to nine N-glycans, which may influence the antigenicity and efficacy of the vaccine. Glycosylation of the influenza A virus has been largely unmonitored in the past as production has been from eggs, where glycan profiles of antigens are difficult if not impossible to control. Over the past decade, however, there have been various commercial influenza vaccines made available from cell technology using animal host cells. Analysis of glycosylation control shows that the type of host cell has the greatest influence on the final analyzed glycan profile. Other factors such as the virus strain, the cultivation system, or various process parameters have been shown to have only a minor effect on the glycosylation pattern. We predict that the analysis of glycan profiles in viral vaccines will become increasingly important in the development and consistent manufacturing of safe and potent vaccines. Graphical Abstract.
治疗性重组蛋白（如单克隆抗体）的聚糖谱是一个关键的质量属性，它会影响最终产品的疗效。蛋白表达过程中的细胞糖基化过程取决于许多因素，例如培养基中底物的可用性、核苷酸糖的细胞内含量以及宿主细胞的酶谱。为了控制糖基化的可变性，了解关键工艺参数及其可接受的数值范围非常重要，这可以使具有预定聚糖谱的蛋白实现可重复生产，从而提供所需的生物学功能或治疗效果。在培养过程接近尾声时，关键营养物质（如葡萄糖或半乳糖）的耗尽可能导致聚糖缩短。末端半乳糖基化和唾液酸化特别不稳定，但可以通过某些关键培养基成分的存在来控制。氨积累、pH 值和溶解氧水平也被认为是影响重组蛋白糖基化的关键生物过程参数。可以向培养基中添加特定的酶抑制剂来驱动特定和预定聚糖谱的形成。人们已经尝试了各种方法来预测细胞表达蛋白的聚糖谱，并基于代谢通量和单个糖基化反应的动力学知识建立了代谢模型。与单一重组蛋白不同，病毒疫苗的聚糖谱要复杂得多，也更难预测。甲型流感病毒的例子表明，主要抗原决定簇血凝素具有三到九个 N-聚糖，这可能会影响疫苗的抗原性和效力。过去，由于生产是从鸡蛋中进行的，因此甲型流感病毒的糖基化在很大程度上没有得到监测，在鸡蛋中，抗原的聚糖谱难以控制，如果不是不可能控制的话。然而，在过去的十年中，已经有各种使用动物宿主细胞的细胞技术生产的商业流感疫苗上市。对糖基化控制的分析表明，宿主细胞的类型对最终分析的聚糖谱影响最大。其他因素，如病毒株、培养系统或各种工艺参数，对糖基化模式的影响较小。我们预测，在病毒疫苗中分析聚糖谱将在安全有效的疫苗的开发和持续生产中变得越来越重要。
