Dataset Information

Quantitative evaluation of yeast's requirement for glycerol formation in very high ethanol performance fed-batch process.

ABSTRACT: BACKGROUND: Glycerol is the major by-product accounting for up to 5% of the carbon in Saccharomyces cerevisiae ethanolic fermentation. Decreasing glycerol formation may redirect part of the carbon toward ethanol production. However, abolishment of glycerol formation strongly affects yeast's robustness towards different types of stress occurring in an industrial process. In order to assess whether glycerol production can be reduced to a certain extent without jeopardizing growth and stress tolerance, the yeast's capacity to synthesize glycerol was adjusted by fine-tuning the activity of the rate-controlling enzyme glycerol 3-phosphate dehydrogenase (GPDH). Two engineered strains whose specific GPDH activity was significantly reduced by two different degrees were comprehensively characterized in a previously developed Very High Ethanol Performance (VHEP) fed-batch process. RESULTS: The prototrophic strain CEN.PK113-7D was chosen for decreasing glycerol formation capacity. The fine-tuned reduction of specific GPDH activity was achieved by replacing the native GPD1 promoter in the yeast genome by previously generated well-characterized TEF promoter mutant versions in a gpd2Delta background. Two TEF promoter mutant versions were selected for this study, resulting in a residual GPDH activity of 55 and 6%, respectively. The corresponding strains were referred to here as TEFmut7 and TEFmut2. The genetic modifications were accompanied to a strong reduction in glycerol yield on glucose; the level of reduction compared to the wild-type was 61% in TEFmut7 and 88% in TEFmut2. The overall ethanol production yield on glucose was improved from 0.43 g g(-1) in the wild type to 0.44 g g(-1) measured in TEFmut7 and 0.45 g g(-1) in TEFmut2. Although maximal growth rate in the engineered strains was reduced by 20 and 30%, for TEFmut7 and TEFmut2 respectively, strains' ethanol stress robustness was hardly affected; i.e. values for final ethanol concentration (117 +/- 4 g L(-1)), growth-inhibiting ethanol concentration (87 +/- 3 g L(-1)) and volumetric ethanol productivity (2.1 +/- 0.15 g l(-1) h(-1)) measured in wild-type remained virtually unchanged in the engineered strains. CONCLUSIONS: This work demonstrates the power of fine-tuned pathway engineering, particularly when a compromise has to be found between high product yield on one hand and acceptable growth, productivity and stress resistance on the other hand. Under the conditions used in this study (VHEP fed-batch), the two strains with "fine-tuned" GPD1 expression in a gpd2Delta background showed slightly better ethanol yield improvement than previously achieved with the single deletion strains gpd1Delta or gpd2Delta. Although glycerol reduction is known to be even higher in a gpd1Delta gpd2Delta double deletion strain, our strains could much better cope with process stress as reflected by better growth and viability.

SUBMITTER: Pagliardini J

PROVIDER: S-EPMC2887396 | biostudies-literature | 2010

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Quantitative evaluation of yeast's requirement for glycerol formation in very high ethanol performance fed-batch process.

Pagliardini Julien J Hubmann Georg G Bideaux Carine C Alfenore Sandrine S Nevoigt Elke E Guillouet Stéphane E SE

Microbial cell factories 20100521

<h4>Background</h4>Glycerol is the major by-product accounting for up to 5% of the carbon in Saccharomyces cerevisiae ethanolic fermentation. Decreasing glycerol formation may redirect part of the carbon toward ethanol production. However, abolishment of glycerol formation strongly affects yeast's robustness towards different types of stress occurring in an industrial process. In order to assess whether glycerol production can be reduced to a certain extent without jeopardizing growth and stress ...[more]

PMID: 20492645

Similar Datasets

Project description:Although process intensification by continuous operation has been successfully applied in the chemical industry, the biopharmaceutical industry primarily uses fed-batch, rather than continuous or perfusion methods, to produce stable monoclonal antibodies (mAbs) from Chinese hamster ovary (CHO) cells. Conventional fed-batch bioreactors may start with an inoculation viable cell density (VCD) of ~0.5 × 106 cells/mL. Increasing the inoculation VCD in the fed-batch production bioreactor (referred to as N stage bioreactor) to 2-10 × 106 cells/mL by introducing perfusion operation or process intensification at the seed step (N-1 step) prior to the production bioreactor has recently been used because it increases manufacturing output by shortening cell culture production duration. In this study, we report that increasing the inoculation VCD significantly improved the final titer in fed-batch production within the same 14-day duration for 3 mAbs produced by 3 CHO GS cell lines. We also report that other non-perfusion methods at the N-1 step using either fed batch or batch mode with enriched culture medium can similarly achieve high N-1 final VCD of 22-34 × 106 cells/mL. These non-perfusion N-1 seeds supported inoculation of subsequent production fed-batch production bioreactors at increased inoculation VCD of 3-6 × 106 cells/mL, where these achieved titer and product quality attributes comparable to those inoculated using the perfusion N-1 seeds demonstrated in both 5-L bioreactors, as well as scaled up to 500-L and 1000-L N-stage bioreactors. To operate the N-1 step using batch mode, enrichment of the basal medium was critical at both the N-1 and subsequent intensified fed-batch production steps. The non-perfusion N-1 methodologies reported here are much simpler alternatives in operation for process development, process characterization, and large-scale commercial manufacturing compared to perfusion N-1 seeds that require perfusion equipment, as well as preparation and storage vessels to accommodate large volumes of perfusion media. Although only 3 stable mAbs produced by CHO cell cultures are used in this study, the basic principles of the non-perfusion N-1 seed strategies for shortening seed train and production culture duration or improving titer should be applicable to other protein production by different mammalian cells and other hosts at any scale biologics facilities.

Project description:Biomass and cell-specific metabolic rates usually change dynamically over time, making the "feed according to need" strategy difficult to realize in a commercial fed-batch process. We here demonstrate a novel feeding strategy which is designed to hold a particular metabolic state in a fed-batch process by adaptive feeding in real time. The feed rate is calculated with a transferable biomass model based on capacitance, which changes the nutrient flow stoichiometrically in real time. A limited glucose environment was used to confine the cell in a particular metabolic state. In order to cope with uncertainty, two strategies were tested to change the adaptive feed rate and prevent starvation while in limitation: (i) inline pH and online glucose concentration measurement or (ii) inline pH alone, which was shown to be sufficient for the problem statement. In this contribution, we achieved metabolic control within a defined target range. The direct benefit was two-fold: the lactic acid profile was improved and pH could be kept stable. Multivariate Data Analysis (MVDA) has shown that pH influenced lactic acid production or consumption in historical data sets. We demonstrate that a low pH (around 6.8) is not required for our strategy, as glucose availability is already limiting the flux. On the contrary, we boosted glycolytic flux in glucose limitation by setting the pH to 7.4. This new approach led to a yield of lactic acid/glucose (Y L/G) around zero for the whole process time and high titers in our labs. We hypothesize that a higher carbon flux, resulting from a higher pH, may lead to more cells which produce more product. The relevance of this work aims at feeding mammalian cell cultures safely in limitation with a desired metabolic flux range. This resulted in extremely stable, low glucose levels, very robust pH profiles without acid/base interventions and a metabolic state in which lactic acid was consumed instead of being produced from day 1. With this contribution, we wish to extend the basic repertoire of available process control strategies, which will open up new avenues in automation technology and radically improve process robustness in both process development and manufacturing.

Project description:Introduction: Hybrid modeling combining First-Principles with machine learning is becoming a pivotal methodology for Biopharma 4.0 enactment. Chinese Hamster Ovary (CHO) cells, being the workhorse for industrial glycoproteins production, have been the object of several hybrid modeling studies. Most previous studies pursued a shallow hybrid modeling approach based on three-layered Feedforward Neural Networks (FFNNs) combined with macroscopic material balance equations. Only recently, the hybrid modeling field is incorporating deep learning into its framework with significant gains in descriptive and predictive power. Methods: This study compares, for the first time, deep and shallow hybrid modeling in a CHO process development context. Data of 24 fed-batch cultivations of a CHO-K1 cell line expressing a target glycoprotein, comprising 30 measured state variables over time, were used to compare both methodologies. Hybrid models with varying FFNN depths (3-5 layers) were systematically compared using two training methodologies. The classical training is based on the Levenberg-Marquardt algorithm, indirect sensitivity equations and cross-validation. The deep learning is based on the Adaptive Moment Estimation Method (ADAM), stochastic regularization and semidirect sensitivity equations. Results and conclusion: The results point to a systematic generalization improvement of deep hybrid models over shallow hybrid models. Overall, the training and testing errors decreased by 14.0% and 23.6% respectively when applying the deep methodology. The Central Processing Unit (CPU) time for training the deep hybrid model increased by 31.6% mainly due to the higher FFNN complexity. The final deep hybrid model is shown to predict the dynamics of the 30 state variables within the error bounds in every test experiment. Notably, the deep hybrid model could predict the metabolic shifts in key metabolites (e.g., lactate, ammonium, glutamine and glutamate) in the test experiments. We expect deep hybrid modeling to accelerate the deployment of high-fidelity digital twins in the biopharma sector in the near future.

Dataset Information

Quantitative evaluation of yeast's requirement for glycerol formation in very high ethanol performance fed-batch process.

Publications

Quantitative evaluation of yeast's requirement for glycerol formation in very high ethanol performance fed-batch process.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets