Over the past three decades, significant progress has been made in the production of therapeutic proteins in the biopharmaceutical industry, driving sustained growth in the market. In particular, the rise of monoclonal antibodies (mAbs) has become the mainstay of therapeutic protein products. Monoclonal antibodies belong to the immunoglobulin family, among which IgG antibodies have become the most widely used type of antibody currently due to their excellent physicochemical properties and broad indications.
At present, approximately 70% of therapeutic recombinant proteins are produced using mammalian cells, among which Chinese Hamster Ovary (CHO) cells are the most mainstream expression system. Due to their advantages such as the ability to perform complex protein glycosylation, high adaptability, and safety, CHO cells occupy an irreplaceable position in industrial production. Their advantages are reflected not only in high productivity and strong expression stability but also in their ability to control drug quality (such as glycosylation patterns), which can effectively avoid immunogenicity issues. Therefore, CHO cells dominate both the research stage and commercial scale production.
Medium Composition and Key Components
Mammalian cell culture media usually have complex formulations, covering a variety of nutrients. Glucose, as the main carbon source and energy source, is the core of cellular metabolic activities. Glutamine, as the main nitrogen source, not only provides precursors for amino acid synthesis but also directly affects the energy metabolic pathways of cells. Different amino acids, as building blocks of proteins, also play important roles in regulating cell growth and product synthesis. In addition, lipids contribute to the formation of cell membrane structures and energy storage, vitamins are indispensable as cofactors for enzyme reactions, and inorganic ions and trace elements maintain the osmotic pressure and pH stability of the medium.
Amino acids play a pivotal role in cellular metabolism and are divided into two categories: “essential amino acids” and “non-essential amino acids”. The former cannot be synthesized by the cells themselves and must be supplemented exogenously, while the latter, although synthesizable by the cells, may still become limiting factors under high-density culture conditions and are therefore often added extra to avoid growth bottlenecks.
Byproduct Metabolism and Metabolic Burden Issues
Although CHO cells have excellent protein expression capabilities, their energy metabolism efficiency is relatively low. Under high-density culture, cells usually take up glucose and glutamine at a high rate, leading to the production of a large number of metabolic byproducts, especially the accumulation of lactate and ammonium ions, which become key factors affecting cell status and antibody yield.
Lactate, as the main product of anaerobic glucose metabolism, is often produced at a high rate in the early stage of culture. High lactate concentration not only lowers the pH of the culture medium but also has cytotoxicity, inhibiting cell growth and affecting the glycosylation quality of proteins. It has been confirmed that lactate concentration exceeding 20 mM has obvious toxic effects on cells. Therefore, researchers have focused on how to induce the metabolic shift of cells from “lactate production” to “lactate reconsumption”. This metabolic switch can not only reduce the accumulation of toxicity but also improve culture stability and product quality. Although the mechanism is not yet fully clear, it may be comprehensively affected by factors such as glucose concentration, glutamine depletion, pH changes, and cellular oxidative status.
Currently commonly used strategies include controlling glucose concentration at a low level, supplementing different carbon sources (such as glycerol and lactate) to induce oxidative metabolic pathways, and applying fed-batch or perfusion culture processes to finely regulate metabolic burden. Especially in the perfusion culture process, stable product quality and higher overall yield can be achieved through cell viability maintenance and byproduct dilution.
Similar to lactate, ammonium ions are products of amino acid deamination reactions, especially glutamine decomposition. The accumulation of ammonium also affects cellular energy metabolism, interferes with ion balance, and inhibits cell growth and antibody expression by competitively inhibiting the function of ion pumps such as Na⁺/K⁺-ATPase. Controlling the addition rate of glutamine or using its substitutes (such as propionamide and glutamic acid) has become one of the effective strategies to reduce ammonium accumulation.
Regulatory Role of Amino Acid Metabolism
In recent years, studies have found that certain amino acids play important regulatory roles in CHO cell metabolism and mAb yield. The limitation of asparagine and serine leads to cell metabolic arrest, indicating their direct impact on the cell cycle and protein synthesis pathways. The supplementation of amino acids such as arginine, leucine, and valine has also been shown to improve cell viability or slow down the accumulation of toxic byproducts. For example, adding valine in fed-batch culture can reduce ammonium concentration, thereby increasing cell growth rate and extending the high-efficiency expression phase.
In addition, studies have shown that there is a significant correlation between the consumption rate of specific amino acids and their relative abundance in the target antibody sequence, which provides a theoretical basis for guiding medium optimization through protein structure information. Analyzing the amino acid composition of the target protein can help formulate a “customized nutrient supply strategy” to achieve optimal resource allocation and control of culture costs.
Intracellular and Extracellular Metabolic Research and Model Construction
In cell culture research, traditional kinetic analysis mostly focuses on changes in the concentration of extracellular substrates and products, such as glucose, glutamine, lactate, and ammonium. However, simple external monitoring cannot fully reveal the cellular metabolic mechanism, especially under high-density culture or complex feeding strategies.
In recent years, the analysis of intracellular metabolites has gradually become a research focus. Studies have shown that the intracellular metabolic state is closely related to the peak of mAb expression. For example, when cells are in a highly glycolytic state, their growth rate reaches a peak; while the peak of mAb expression usually corresponds to the shift of cellular metabolism to the oxidative phosphorylation pathway. Such information cannot be directly inferred from external detection, so the introduction of intracellular metabolic analysis helps to establish more accurate kinetic models.
Furthermore, some studies have begun to attempt to introduce intracellular metabolite concentrations into macrokinetic models. For example, by tracking the dynamic changes of intracellular NADPH, ATP, lactate, or amino acids, feeding strategies can be optimized, or yield differences under different culture conditions can be predicted. Such methods also show great potential in fluxomics and metabolic flux analysis.
Integration of Macroscopic and Microscopic Models
Kinetic modeling of cell culture can be divided into macroscopic and microscopic approaches. Macroscopic models usually only consider changes in extracellular substances, treat cells as a “black box”, and predict cell behavior through the relationship between input and output variables. Such models have fewer parameters, which are easy to use and adjust, but their accuracy is low when dealing with changes in complex conditions.
In contrast, microscopic models simulate intracellular metabolic pathways, including a large number of metabolites, enzyme reactions, and regulatory factors, and have higher biological explanatory power. However, such models have a large number of parameters, high experimental measurement costs, and high requirements for modeling technology and data quality.
To balance complexity and practicality, new methods such as Flux Balance Analysis (FBA) have been introduced. On the premise that cells are in a metabolic steady state, metabolic flux distribution is solved through optimization algorithms. Although this method simplifies parameter requirements, its steady-state assumption may not hold under fed-batch and dynamic environments.
Therefore, combining changes in extracellular concentrations with intracellular metabolic information is an important direction for future kinetic modeling. This hybrid model can not only improve prediction accuracy but also be used for dynamic optimization control, personalized medium design, and evaluation of cell engineering strategies.
Research Trends and Challenges
Despite the substantial achievements in current research on CHO cell metabolism, several challenges remain. First, the lack of a unified high-throughput analysis platform makes it difficult to standardize and integrate intracellular and extracellular metabolite data. Second, the cell lines, medium formulations, and culture processes used in different research contexts vary, resulting in limited comparability and universality between models.
In addition, current kinetic modeling research still focuses on aspects such as changes in lactate metabolism and optimization of feeding strategies, and there is a lack of systematic research on the correlation between intracellular IgG concentration changes and internal and external metabolism. In-depth exploration of this field will not only help break through the bottlenecks of traditional processes but also may provide theoretical support for the next generation of intelligent and digital cell culture systems.
Future research should pay more attention to the development of quantitative detection technologies for key intracellular metabolites, and promote the transformation of cell culture modeling from experience-driven to data-driven, realizing a leap from “black box” operation to a “transparent system”. In addition, AI-assisted modeling, integration of online PAT sensing technology, and embedded application of digital twin systems in bioprocesses will become key technical paths driving kinetic models towards real-time control and predictive optimization.
