Fed-batch Cell Culture Research Articles

Computational Modeling of Fed-Batch Cell Culture Bioreactor: Hybrid Agent-Based Approach

A hybrid simulation framework was proposed to predict the dynamics in cell culture bioreactors. The model is based on a multi-agent approach where CHO cells are considered as individuals (agents) following a rule base governing their behavior, while a flux balance model is embedded in agents to predict quantitative changes in nutrient and metabolite concentrations. The model takes the measured dissolved oxygen, and sodium data as input along with initial cell culture conditions and predicts the dynamics of viable cell density, viability, concentrations of glucose and lactate. The model showed good agreement with the experimental findings from our laboratory for two sets of cell culture experiments.

Rapid high-throughput characterisation, classification and selection of recombinant mammalian cell line phenotypes using intact cell MALDI-ToF mass spectrometry fingerprinting and PLS-DA modelling

Despite many advances in the generation of high producing recombinant mammalian cell lines over the last few decades, cell line selection and development is often slowed by the inability to predict a cell line's phenotypic characteristics (e.g. growth or recombinant protein productivity) at larger scale (large volume bioreactors) using data from early cell line construction at small culture scale. Here we describe the development of an intact cell MALDI-ToF mass spectrometry fingerprinting method for mammalian cells early in the cell line construction process whereby the resulting mass spectrometry data are used to predict the phenotype of mammalian cell lines at larger culture scale using a Partial Least Squares Discriminant Analysis (PLS-DA) model. Using MALDI-ToF mass spectrometry, a library of mass spectrometry fingerprints was generated for individual cell lines at the 96 deep well plate stage of cell line development. The growth and productivity of these cell lines were evaluated in a 10L bioreactor model of Lonza's large-scale (up to 20,000L) fed-batch cell culture processes. Using the mass spectrometry information at the 96 deep well plate stage and phenotype information at the 10L bioreactor scale a PLS-DA model was developed to predict the productivity of unknown cell lines at the 10L scale based upon their MALDI-ToF fingerprint at the 96 deep well plate scale. This approach provides the basis for the very early prediction of cell lines’ performance in cGMP manufacturing-scale bioreactors and the foundation for methods and models for predicting other mammalian cell phenotypes from rapid, intact-cell mass spectrometry based measurements.

Optimization and robustness analysis of hybridoma cell fed-batch cultures using the overflow metabolism model

The maximization of biomass productivity in fed-batch cultures of hybridoma cells is analyzed based on the overflow metabolism model. Due to overflow metabolism, often attributed to limited oxygen capacity, lactate and ammonia are formed when the substrate concentrations (glucose and glutamine) are above a critical value, which results in a decrease in biomass productivity. Optimal feeding rate, on the one hand, for a single feed stream containing both glucose and glutamine and, on the other hand, for two separate feed streams of glucose and glutamine are determined using a Nelder-Mead simplex optimization algorithm. The optimal multi exponential feed rate trajectory improves the biomass productivity by 10 % as compared to the optimal single exponential feed rate. Moreover, this result is validated by the one obtained with the analytical approach in which glucose and glutamine are fed to the culture so as to control the hybridoma cells at the critical metabolic state, which allows maximizing the biomass productivity. The robustness analysis of optimal feeding profiles obtained with different optimization strategies is considered, first, with respect to parameter uncertainties and, finally, to model structure errors.

Addition of Valproic Acid to CHO Cell Fed-Batch Cultures Improves Monoclonal Antibody Titers

Improving the productivity of a biopharmaceutical Chinese hamster ovary (CHO) fed-batch cell culture can enable cost savings and more efficient manufacturing capacity utilization. One method for increasing CHO cell productivity is the addition of histone deacetylase (HDAC) inhibitors to the cell culture process. In this study, we examined the effect of valproic acid (VPA, 2-propylpentanoic acid), a branched-chain carboxylic acid HDAC inhibitor, on the productivity of three of our CHO cell lines that stably express monoclonal antibodies. Fed-batch shake flask VPA titrations on the three different CHO cell lines yielded cell line-specific results. Cell line A responded highly positively, cell line B responded mildly positively, and cell line C did not respond. We then performed factorial experiments to identify the optimal VPA concentration and day of addition for cell line A. After identifying the optimal conditions for cell line A, we performed verification experiments in fed-batch bioreactors for cell lines A and B. These experiments confirmed that a high dose of VPA late in the culture can increase harvest titer >20 % without greatly changing antibody aggregation, charge heterogeneity, and N-linked glycosylation profiles. Our results suggest that VPA is an attractive and viable small molecule enhancer of protein production for biopharmaceutical CHO cell culture processes.

Parameter Identification for State Estimation: Design of an Extended Kalman Filter for Hybridoma Cell Fed-Batch Cultures

The monitoring and optimization of hybridoma cell fed-batch cultures depend on the availability of appropriate on-line sensors for the main culture components. A simple and efficient approach to maintain hybridoma cultures in the optimal operating conditions is to regulate the substrate concentrations at the critical values (G=Gcrit and/or Gn=Gncrit) such as to control the hybridoma cells at the critical metabolism state. However, reliable glucose and glutamine probes are currently rare and/or very expensive on the market and it is necessary to design software sensors which are at same time cheap and reliable and that can be used for online measurement. In this study, the overflow metabolism model is used to develop an extended Kalman filter for online estimation of glucose and glutamine in hybridoma cell fed-batch cultures based on the considered available measurements (biomasses (on-line), lactate and ammonia (on-line or off-line)). The observability conditions are examined, and the performances are analysed with simulations of hybridoma cell fed-batch cultures. Glutamine estimation sensitivity is enforced by minimizing a cost function combining a usual least-squares criterion with a state estimation sensitivity criterion.

Analysis of Chinese hamster ovary cell metabolism through a combined computational and experimental approach

Optimization of cell culture processes can benefit from the systematic analysis of experimental data and their organization in mathematical models, which can be used to decipher the effect of individual process variables on multiple outputs of interest. Towards this goal, a kinetic model of cytosolic glucose metabolism coupled with a population-level model of Chinese hamster ovary cells was used to analyse metabolic behavior under batch and fed-batch cell culture conditions. The model was parameterized using experimental data for cell growth dynamics, extracellular and intracellular metabolite profiles. The results highlight significant differences between the two culture conditions in terms of metabolic efficiency and motivate the exploration of lactate as a secondary carbon source. Finally, the application of global sensitivity analysis to the model parameters highlights the need for additional experimental information on cell cycle distribution to complement metabolomic analyses with a view to parameterize kinetic models.

Chemometrics and in-line near infrared spectroscopic monitoring of a biopharmaceutical Chinese hamster ovary cell culture: Prediction of multiple cultivation variables

In the present study near infrared (NIR) spectroscopy was used to monitor the cultivation of mammalian Chinese hamster ovary (CHO) cells producing a monoclonal antibody in a fed-batch cell culture process. A temperature shift was applied during the cultivation. The cells were incubated at 37°C and 33°C. The Fourier transform near infrared (FT-NIR) multiplex process analyzer spectroscopy was investigated to monitor cultivation variables of the CHO cell culture from 10 independent batches using two channels of the FT-NIR. The measurements were performed on production scale bioreactors of 12,500L. The cell cultures were analyzed with the spectrometer coupled to a transflection sterilizable fiber optic probe inserted into the bioreactors. Multivariate data analysis (MVDA) employing unsupervised principal component analysis (PCA) and partial least squares regression methods (PLS) were applied. PCA demonstrated that 96% of the observed variability was explained by the process trajectory and the inter-batch variability. PCA was found to be a significant tool in identifying batch homogeneity between lots and in detecting abnormal fermentation runs. Seven different cell culture parameters such as osmolality, glucose concentration, product titer, packed cell volume (PCV), integrated viable packed cell volume (ivPCV), viable cell density (VCD), and integrated viable cell count (iVCC) were monitored inline and predicted by NIR. NIR spectra and reference analytics data were computed using control charts to evaluate the monitoring abilities. Control charts of each media component were under control by NIR spectroscopy. The PLS calibration plots offered accurate predictive capabilities for each media. This paper underlines the capability for inline prediction of multiple cultivation variables during bioprocess monitoring.

Optimal Operation of Hybridoma Cell Fed-Batch Cultures Using the Overflow Metabolism Model: Numerical and Analytical Approach

The maximization of biomass productivity in fed-batch cultures of hybridoma cells is analyzed based on the overflow metabolism model. Due to overflow metabolism, often attributed to limited oxygen capacity, lactate and ammonia are formed when the substrate concentrations (glucose and glutamine) are above a critical value, which results in a decrease in biomass productivity. Optimal feeding rate, on the one hand, for a single feed stream containing both glucose and glutamine and, on the other hand, for two separate feed streams of glucose and glutamine are determined using a Nelder-Mead simplex optimization algorithm. The optimal multi exponential feed rate trajectory improves the biomass productivity by 10% as compared to the optimal single exponential feed rate. Moreover, this result is validated by the one obtained with the analytical approach in which glucose and glutamine are fed to the culture such as to control the hybridoma cells at the critical metabolism state, which allows maximizing the biomass productivity.

Macroscopic modelling of overflow metabolism and model based optimization of hybridoma cell fed-batch cultures

A macroscopic model that takes into account phenomena of overflow metabolism within glycolysis and glutaminolysis is proposed to simulate hybridoma HB-58 cell cultures. The model of central carbon metabolism is reduced to a set of macroscopic reactions. The macroscopic model describes three metabolism states: respiratory metabolism, overflow metabolism and critical metabolism. The model parameters and confidence intervals are obtained via a non linear least squares identification. It is validated with experimental data of fed-batch hybridoma cultures and successfully predicts the dynamics of cell growth and death, substrate consumption (glutamine and glucose) and metabolites production (lactate and ammonia). Based on a sensitivity analysis of the model outputs with respect to the parameters, a model reduction is proposed. Finally, the maximization of biomass productivity of hybridoma cell fed-batch cultures is analyzed. This model allows, on the one hand, quantitatively describing overflow metabolism in mammalian cell cultures and, on the other hand, will be valuable for monitoring and control of fed-batch cultures in order to optimize the process. This is illustrated in this contribution with the determination of optimal feeding profiles aiming at maximizing biomass productivity.

Analyzing the dynamics of cell growth and protein production in mammalian cell fed-batch systems using logistic equations

The logistic modeling approach was used to describe experimental viable cell density (X) and product concentration (P) data from two industrial fed-batch mammalian cell culture processes with maximum product concentrations in the 3.0-9.4 g/l range. In both cases, experimental data were well described by the logistic equations and the resulting specific growth rate and protein productivity profiles provided useful insights into the process kinetics. Subsequently, sensitivity equations for both the X and P models were analyzed which helped characterize the influence of model parameters on X and P time courses. This was augmented by conventional sensitivity analyses where five values of each model parameter, 25% apart, were used to generate X and P time courses. Finally, results from sensitivity analysis were used to simulate X and P time courses that were reflective of typical early- and late-stage fed-batch cell culture processes. Different combinations of the logistic model parameters were used to arrive at the same final product concentration demonstrating the ability of the logistic approach to describe the multitude of process paths that result in the same final product concentration. Overall, the capability of the logistic equations to well describe X and P data from fed-batch cultures, coupled with their ability to simulate the multitude of paths leading up to the desired cell density and product concentration profiles, make them a useful tool during mammalian cell fed-batch process development.

Analysis of dynamic changes in the proteome of a Bcl‐XL overexpressing Chinese hamster ovary cell culture during exponential and stationary phases

Mammalian cell cultures used for biopharmaceutical production undergo various dynamic biological changes over time, including the transition of cells from an exponential growth phase to a stationary phase during cell culture. To better understand the dynamic aspects of cell culture, a quantitative proteomics approach was used to identify dynamic trends in protein expression over the course of a Chinese hamster ovary (CHO) cell culture for the production of a recombinant monoclonal antibody and overexpressing the antiapoptotic gene Bcl-xl. Samples were analyzed using a method incorporating iTRAQ labeling, two-dimensional LC/MS, and linear regression calculations to identify significant dynamic trends in protein abundance. Using this approach, 59 proteins were identified with significant temporal changes in expression. Pathway analysis tools were used to identify a putative network of proteins associated with cell growth and apoptosis. Among the differentially expressed proteins were molecular chaperones and isomerases, such as GRP78 and PDI, and reported cell growth markers MCM2 and MCM5. In addition, two proteins with growth-regulating properties, transglutaminase-2 and clusterin, were identified. These proteins are associated with tumor proliferation and apoptosis and were observed to be expressed at relatively high levels during stationary phase, which was confirmed by western blotting. The proteomic methodology described here provides a dynamic view of protein expression throughout a CHO fed-batch cell culture, which may be useful for further elucidating the biological processes driving mammalian cell culture performance.

Batch, fed‐batch, and microcarrier cultures with CHO cell lines in a pressure‐cycle driven miniaturized bioreactor

Miniaturized bioreactors for suspension cultures of animal cells, such as Chinese Hamster Ovary (CHO) cells, could improve bioprocess development through the ability to cheaply explore a wide range of bioprocess operating conditions. A miniaturized pressure-cycled bioreactor for animal cell cultures, described previously (Diao et al., 2008), was tested with a suspension CHO cell line producing commercially relevant quantities of human IgG. Results from the suspended CHO cell line showed that the cell growth was comparable to conventional flask controls and the target protein production was enhanced in the minibioreactor, which may be due to the relatively high oxygen transfer rate and the moderate shear stress, measured and simulated previously. Microcarrier culture using an anchorage-dependent CHO cell line and Cytodex 3 also showed a similar result: comparable growth and enhanced production of a model protein (secreted alkaline phosphatase or SEAP). Various fed-batch schemes were applied to the CHO cells producing human IgG, yielding cell numbers (1.1 × 10(7) /mL) at day 8 and titers of human IgG (2.3 g/L) at day 14 that are typical industrial values for CHO cell fed-batch cultures. The alteration of the volumetric oxygen transfer coefficient is a key parameter for viability of the CHO cell line producing human IgG. We conclude that the minibioreactor can provide favorable cell culture environments; oxygen transfer coefficient and mixing time can be altered to mimic values in a larger scale system allowing for potential prediction of response during scale-up.

Optimizing amino acid composition of CHO cell culture media for a fusion protein production

The amino acid composition of CHO cell culture media for manufacturing an antibody fusion protein (B1) was optimized using data obtained through metabolic flux analysis. The metabolic flux analysis was performed using the metabolic rates derived from two semi-steady states of continuous culture. The amino acid composition was the same for media used in both the fed-batch and continuous cultures, which made the modification of amino acid composition suggested by the metabolic flux analysis applicable to the fed-batch culture. The flux analysis indicated that methionine, tryptophan, asparagine and serine were limiting, while alanine, arginine, glutamine and glycine were in excess at the semi-steady states. Based on this result, the limiting amino acids were increased while those in excess were reduced to obtain balanced amino acid composition in both the batch and feed media. The fed-batch cell cultures using the modified media generated less lactate and ammonia, and more significantly, increased peak cell densities by 55% and B1 protein titer by 27% without any impact on final sialic acid content, a quality attribute of the B1 protein.

Decreasing lactate level and increasing antibody production in Chinese Hamster Ovary cells (CHO) by reducing the expression of lactate dehydrogenase and pyruvate dehydrogenase kinases

Large-scale fed-batch cell culture processes of CHO cells are the standard platform for the clinical and commercial production of monoclonal antibodies. Lactate is one of the major by-products of CHO fed-batch culture. In pH-controlled bioreactors, accumulation of high levels of lactate is accompanied by high osmolality due to the addition of base to control pH of the cell culture medium, potentially leading to lower cell growth and lower therapeutic protein production during manufacturing. Lactate dehydrogenase (LDH) is an enzyme that catalyzes the conversion of the substrate, pyruvate, into lactate and many factors including pyruvate concentration modulate LDH activity. Alternately, pyruvate can be converted to acetyl-CoA by pyruvate dehydrogenases (PDHs), to be metabolized in the TCA cycle. PDH activity is inhibited when phosphorylated by pyruvate dehydrogenase kinases (PDHKs). In this study, we knocked down the gene expression of lactate dehydrogenase A (LDHa) and PDHKs to investigate the effect on lactate metabolism and protein production. We found that LDHa and PDHKs can be successfully downregulated simultaneously using a single targeting vector carrying small inhibitory RNAs (siRNA) for LDHa and PDHKs. Moreover, our fed-batch shake flask evaluation data using siRNA-mediated LDHa/PDHKs knockdown clones showed that downregulating LDHa and PDHKs in CHO cells expressing a therapeutic monoclonal antibody reduced lactate production, increased specific productivity and volumetric antibody production by approximately 90%, 75% and 68%, respectively, without appreciable impact on cell growth. Similar trends of lower lactate level and higher antibody productivity on average in siRNA clones were also observed from evaluations performed in bioreactors.

Dynamic model of CHO cell metabolism

Fed-batch cultures are extensively used for the production of therapeutic proteins. However, process optimization is hampered by lack of quantitative models of mammalian cellular metabolism in these cultures. This paper presents a new kinetic model of CHO cell metabolism and a novel framework for simulating the dynamics of metabolic and biosynthetic pathways of these cells grown in fed-batch culture. The model defines a subset of the intracellular reactions with kinetic rate expressions based on extracellular metabolite concentrations and temperature- and redox-dependent regulatory variables. The simulation uses the rate expressions to calculate pseudo-steady state flux distributions and extracellular metabolite concentrations at discrete time points. Experimental data collected in this study for several different CHO cell fed-batch cultures are used to derive the rate expressions, fit the parameters, and validate the model. The simulations accurately predicted the effects of process variables, including temperature shift, seed density, specific productivity, and nutrient concentrations.

Effect of temperature, pH, dissolved oxygen, and hydrolysate on the formation of triple light chain antibodies in cell culture

THIOMABs are recombinant antibodies with reactive cysteine residues used for forming THIOMAB-drug conjugates (TDCs). We recently reported a new impurity associated with THIOMABs: one of the engineered cysteines forms a disulfide bond with an extra light chain (LC) to generate a triple light chain antibody (3LC). In our previous investigations, increased LC expression increased 3LC levels, whereas increased glutathione (GSH) production decreased 3LC levels. In this work, on three stably transfected CHO cell lines, we investigated the effects of temperature, pH, dissolved oxygen (DO), and hydrolysate on 3LC formation during THIOMAB fed-batch cell culture production. Although pH between 6.8 and 7.0 had no significant impact on 3LC formation, temperature at 35°C instead of 33 or 31°C generated the lowest 3LC values for two cell lines. The decreased 3LC level correlated with increased GSH production. We implemented a 35°C temperature process for large-scale (2,000 L) production of a THIOMAB. This process reduced 3LC levels by ∼50% compared with a 33°C temperature process. By contrast, DO and hydrolysate had modest effect on 3LC levels for the model cell line studied. Overall, we did not find significant changes in LC expression under the conditions tested, whereas changes in GSH production were more evident. By investigating the impact of bioreactor process and medium conditions on 3LC levels, we identified strategies that reduced 3LC levels.

Control of misincorporation of serine for asparagine during antibody production using CHO cells

A recombinant monoclonal antibody produced by Chinese hamster ovary (CHO) cell fed-batch culture was found to have amino acid sequence misincorporation upon analysis by intact mass and peptide mapping mass spectrometry. A detailed analysis revealed multiple sites for asparagine were being randomly substituted by serine, pointing to mistranslation as the likely source. Results from time-course analysis of cell culture suggest that misincorporation was occurring midway through the fed-batch process and was correlated to asparagine reduction to below detectable levels in the culture. Separate shake flask experiments were carried out that confirmed starvation of asparagine and not excess of serine in the medium as the root cause of the phenomenon. Reduction in serine concentration under asparagine starvation conditions helped reduce extent of misincorporation. Supplementation with glutamine also helped reduce extent of misincorporation. Maintenance of asparagine at low levels in 2 L bench-scale culture via controlled supplementation of asparagine-containing feed eliminated the occurrence of misincorporation. This strategy was implemented in a clinical manufacturing process and scaled up successfully to the 200 and 2,000 L bioreactor scales.

Proteomic Profiling of a High-Producing Chinese Hamster Ovary Cell Culture

The productivity of mammalian cell culture expression systems is critically important to the production of biopharmaceuticals. In this study, a high-producing Chinese hamster ovary cell culture which was transfected with the apoptosis inhibitor Bcl-X(L) gene was compared to a low-producing control that was not transfected. Shotgun proteomics was used to compare the high and low-producing fed-batch cell cultures at different growth time points. The goals of this study were twofold; it would be of value to find a biomarker that could predict cell lines with higher growth efficiency and to gain mechanistic insights into the effects of the introduction of a foreign gene that is known to have growth regulating properties in human cells. A total of 392 proteins were identified in this study, and 32 of these proteins were determined to be differentially expressed. In the high-producing cell culture, several proteins related to protein metabolism were upregulated, such as eukaryotic translation initiation factor 3 and ribosome 40S. In addition, several intermediate filament proteins such as vimentin and annexin, as well as histone H1.2 and H2A, were downregulated in the high producer. The expression of these proteins may be indicative of cellular productivity. A growth inhibitor, galectin-1, was downregulated in the high producer, which may be linked to the expression of Bcl-X(L). The molecular chaperone BiP was upregulated significantly in the high producer and may indicate an unfolded protein response due to endoplasmic reticulum (ER) stress. Several proteins involved in regulation of the cell cycle such as RACK1 and GTPase Ran were found to be differentially expressed, which may be due to a differentially controlled cell cycle between low- and high-producing cell cultures.

Process monitoring and quality variable prediction utilizing PLS in industrial fed-batch cell culture

Partial least squares is a data-driven modeling technique that has been utilized for process monitoring in a variety of industrial processes. This paper develops a novel online partial least squares approach (evolving PLS) and compares it with an existing online PLS technique (global PLS). Both methods are applied to an industrial fed-batch mammalian cell culture process, where process variables are used to predict a key quality variable, product titer. Fault detection and diagnosis are performed using PLS models and statistical metrics. This new detection approach was able to recognize a variety of faults during online monitoring.

Elucidation of metabolism in hybridoma cells grown in fed‐batch culture by genome‐scale modeling

Genome-scale modeling of mouse hybridoma cells producing monoclonal antibodies (mAb) was performed to elucidate their physiological and metabolic states during fed-batch cell culture. Initially, feed media nutrients were monitored to identify key components among carbon sources and amino acids with significant impact on the desired outcome, for example, cell growth and antibody production. The monitored profiles indicated rapid assimilation of glucose and glutamine during the exponential growth phase. Significant increase in mAb concentration was also observed when glutamine concentration was controlled at 0.5 mM as a feeding strategy. Based on the reconstructed genome-scale metabolic network of mouse hybridoma cells and fed-batch profiles, flux analysis was then implemented to investigate the cellular behavior and changes in internal fluxes during the cell culture. The simulated profile of the cell growth was consistent with experimentally measured specific growth rate. The in silico simulation results indicated (i) predominant utilization of glycolytic pathway for ATP production, (ii) importance of pyruvate node in metabolic shifting, and (iii) characteristic pattern in lactate to glucose ratio during the exponential phase. In future, experimental and in silico analyses can serve as a promising approach to identifying optimal feeding strategies and potential cell engineering targets as well as facilitate media optimization for the enhanced production of mAb or recombinant proteins in mammalian cells.