Sensor combination and chemometric variable selection for online monitoring of Streptomyces coelicolor fed-batch cultivations

Fed-batch cultivations of Streptomyces coelicolor, producing the antibiotic actinorhodin, were monitored online by multiwavelength fluorescence spectroscopy and off-gas analysis. Partial least squares (PLS), locally weighted regression, and multilinear PLS (N-PLS) models were built for prediction of biomass and substrate (casamino acids) concentrations, respectively. The effect of combination of fluorescence and gas analyzer data as well as of different variable selection methods was investigated. Improved prediction models were obtained by combination of data from the two sensors and by variable selection using a genetic algorithm, interval PLS, and the principal variables method, respectively. A stepwise variable elimination method was applied to the three-way fluorescence data, resulting in simpler and more accurate N-PLS models. The prediction models were validated using leave-one-batch-out cross-validation, and the best models had root mean square error of cross-validation values of 1.02 g l⁻¹ biomass and 0.8 g l⁻¹ total amino acids, respectively. The fluorescence data were also explored by parallel factor analysis. The analysis revealed four spectral profiles present in the fluorescence data, three of which were identified as pyridoxine, NAD(P)H, and flavin nucleotides, respectively.     

On-line process monitoring and chemometric modeling with 2D fluorescence spectra obtained in recombinant E. coli fermentations

2D spectrofluorometry produces a large volume of spectral data during fermentation processes with recombinant E. coli, which can be analyzed using chemometric methods such as principal component analysis (PCA), principal component regression (PCR) and partial least square regression (PLS). An analysis of the spectral data by PCA results in scores and loadings that are not only visualized in the score-loading plots but are also used to monitor the fermentation processes on-line. The score plots provided useful qualitative information on four fermentation processes for the production of extracellular 5-aminolevulinic acid (ALA). Two chemometric models (PCR and PLS) were used to examine the correlation between the 2D fluorescence spectra and a few parameters of the fermentation processes. The results showed that PLS had slightly better calibration and prediction performance than PCR.     

Simple control of fed-batch processes for recombinant protein production with E. coli

A very simple but effective process control technique is proposed that leads to a high batch-to-batch reproducibility with respect to biomass concentration as well as the specific biomass growth rate profiles in E. coli fermentations performed during recombinant protein production. It makes use of the well-established temperature controllers in currently used fermenters, but takes its information from the difference between the controlled culture temperature T (cult) and the temperature T (coolin) of the coolant fed to the fermenter's cooling jacket as adjusted by the fermenter temperature controller. For process control purposes this measured difference is corrected regarding stirrer influences and cumulated before it is used as a new process control variable. As a spin-off of this control, it becomes possible to estimate online the oxygen mass transfer rates and the corresponding k(L)a values during the real cultivation process.     

Intelligent real-time performance monitoring and quality prediction for batch/fed-batch cultivations

Supervision of batch bioprocess operations in real-time during the progress of a batch run offers many advantages over end-of-batch quality control. Multivariate statistical techniques such as multiway partial least squares (MPLS) provide an efficient modeling and supervision framework. A new type of MPLS modeling technique that is especially suitable for online real-time process monitoring and the multivariate monitoring charts are presented. This online process monitoring technique is also extended to include predictions of end-of-batch quality measurements during the progress of a batch run. Process monitoring, quality estimation and fault diagnosis activities are automated and supervised by embedding them into a real-time knowledge-based system (RTKBS). Interpretation of multivariate charts is also automated through a generic rule-base for efficient alarm handling. The integrated RTKBS and the implementation of MPLS-based process monitoring and quality control are illustrated using a fed-batch penicillin production benchmark process simulator.     

Open-loop control of specific growth rate in fed-batch cultures of recombinant E.coli

Summary A computer-based algorithm was used for the open-loop control of specific growth rate in fed-batch cultures of recombinant E.coli.. The control of nutrient feed rate to an exponential trajectory resulted in growth of the culture at a constant specific growth rate. Stable specific growth rates between 0.08 and 0.4 h^–1 were achieved.     

Robust artificial intelligence tool for automatic start-up of the supplementary medium feeding in recombinant E. coli cultivations

One of the most important events in fed-batch fermentations is the definition of the moment to start the feeding. This paper presents a methodology for a rational selection of the architecture of an artificial intelligence (AI) system, based on a neural network committee (NNC), which identifies the end of the batch phase. The AI system was successfully used during high cell density cultivations of recombinant Escherichia coli. The AI algorithm was validated for different systems, expressing three antigens to be used in human and animal vaccines: fragments of surface proteins of Streptococcus pneumoniae (PspA), clades 1 and 3, and of Erysipelothrix rhusiopathiae (SpaA). Standard feed-forward neural networks (NNs), with a single hidden layer, were the basis for the NNC. The NN architecture with best performance had the following inputs: stirrer speed, inlet air, and oxygen flow rates, carbon dioxide evolution rate, and CO₂ molar fraction in the exhaust gas.     

Temperature gradient-based high-cell density fed-batch fermentation for the production of pyruvate oxidase by recombinant E. coli

Pyruvate oxidase (PyOD) is a very powerful enzyme for clinical diagnostic applications and environmental monitoring. Influences of temperature on cell growth, plasmid stability, and PyOD expression during the PyOD fermentation process by recombinant Escherichia coli were investigated. Based on the influences of temperature on the physiological metabolism, a novel high-cell density fed-batch cultivation with gradient temperature decrease strategy for effective PyOD production was achieved, under which the biomass (OD₆₀₀) of recombinant E. coli could reach to 71 and the highest PyOD activity in broth could reach to 3,307 U/L in 26?hr fermentation.     

Production of soluble human interleukin-6 in cytoplasm by fed-batch culture of recombinant E. coli

The major objective of this study is to identify fed-batch culture conditions optimal for the production of human interleukin-6 (hIL-6) in a soluble form. Five different expression vectors were constructed for the expression of hIL-6 and hIL-6s fused with NusA, maltose binding protein (MBP), thioredoxin (Trx) or ubiquitin (Ubi). A series of flask cultures were conducted in LB medium at 37 degrees C. The intact hIL-6 was expressed mostly in the form of inclusion body. More than 95% of the hIL-6 fused with NusA (NusA/hIL-6) and about 90% of MBP/hIL-6 were expressed in a soluble form, whereas Trx/hIL-6 and Ubi/hIL-6 were expressed mostly in the form of inclusion body. Based on this result, NusA was selected as the fusion partner for the production of hIL-6 in the subsequent experiments. A series of pH-stat fed-batch cultures of an E. coli BL21(DE3) transformed with a NusA/hIL-6 expression vector were conducted in a bioreactor with a working volume of about 3 L. As the amount of nitrogen source was increased in the feeding medium, more soluble NusA/hIL-6 was produced, while the total amount was not significantly changed. Under the best conditions tested, about 90% of NusA/hIL-6 was produced in the soluble form. In this case, the concentration of soluble NusA/hIL-6 was 7.5 g/L with a volumetric productivity of 0.43 g/L-h.     

Influence of process temperature on recombinant enzyme activity in Escherichia coli fed-batch cultures

The influence of proteolysis over recombinant protein quality has been studied using rhamnulose 1-phosphate aldolase (RhuA) production as case example. Progressive induction by means of continuous isopropyl-β-d-thiogalactopyranoside (IPTG) dosage in Escherichia coli fed-batch cultures led to high specific levels of recombinant protein. However, the specific activity profile did not correlate to the specific protein content when the process was run at 37 °C and there was a decrease of the enzyme activity along the induction phase. Specific activity loss depending on the presence of an energy source was observed at short term, but protein degradation due to the action of energy-independent metalloproteases occurred after a longer time period. The effects of lowering the temperature were analysed on both mechanisms, and a reduction of the specific activity loss was observed when the process temperature was decreased to 28 °C. Lower plasmid copy number and specific production rates probably alleviated the metabolic load on host cell during recombinant protein overexpression, and a high increase of the enzyme activity was achieved in high cell density fed-batch cultures under these conditions.     

Identification of critical batch operating parameters in fed-batch recombinant E. coli fermentations using decision tree analysis

To develop a useful fermentation process model, it is first necessary to identify which batch operating parameters are critical in determining the process outcome. To identify critical processing inputs in large databases, we have explored the use of Decision Tree Analysis with the decision metrics of Gain (i.e., Shannon Entropy changes), Gain Ratio, and a multiple hypergeometric distribution. The usefulness of this approach lies in its ability to treat "categorical" variables, which are typical of archived fermentation databases, as well as "continuous" variables. In this work, we demonstrate the use of Decision Tree Analysis for the problem of optimizing recombinant green fluorescent protein production in E. coli. A database of 85 fermentations was generated to examine the effect of 15 process input parameters on final biomass yield, maximum recombinant protein concentration, and productivity. The use of Decision Tree Analysis led to a considerable reduction in the fermentation database through the identification of the significant as well as insignificant inputs. However, different decision metrics selected different inputs and different numbers of inputs to classify the data for each output.     

Industrial control of recombinant E. coli fed-batch culture: new perspectives on traditional controlled variables

Three industrially relevant fed-batch algorithms were applied to the production of recombinant proteins in Escherichia coli BL21 DE3. Starvation dissolved-oxygen (DO)-transient control sustained growth rates greater than those in pH-stat (0.16 h(-1) versus 0.11-0.13 h(-1)) while feed-up DO-transient control better tracked the measured threshold for acetate production (mu approximately 0.2 h(-1)). All controllers supported growth without acetate production, resulting in end concentrations of recombinant protein up to 20 times greater than in batch culture. Both DO-transient control systems were judged superior to pH-stat for their ability to detect and track the acetate threshold. Results also showed that although high cell density at the stationary phase is desirable, this parameter may be dictated by the choice of media and reactor design as opposed to controller type. Controller selection, however, has a great impact on the capacity to track the acetate threshold and therefore to enhance productivity without concomitant acetate production.     

Influence of controlled glucose oscillations on a fed-batch process of recombinant Escherichia coli

The influence of glucose oscillations on cell growth and product formation of a recombinant Escherichia coli culture producing a heterologous alpha-glucosidase was studied in fed-batch cultures in a laboratory bioreactor. Glucose oscillations were created by an on/off-feeding mode in either fast cycles (1 min) or slow cycles (4 min) and compared to a process with constant glucose addition. The study indicates that glucose oscillations influence the product stability and the overgrowth of plasmid-free cells if such cultures are not performed under continuous pressure for selection of plasmid-containing cells. Although the glucose uptake capacity decreased after induction of the recombinant alpha-glucosidase in all cultures performed, the up-growth of plasmid-free cells during the production phase was strongly inhibited by fast oscillations. In contrast, plasmid-free cells grew up when constant feeding or slow cycles were applied. Our data suggest that the various feed protocols effect the specific carbon dioxide formation rate differently, with the highest production of carbon dioxide in the cultivations with fast cycles. In connection to product formation the initial alpha-glucosidase accumulation was the same in all cultures, but the stability of the product was significantly lower in the cultivation with slow cycles. Our results from laboratory experiments are discussed in relation to the mixing situation in large-scale bioreactors.     

Integrated L-phenylalanine separation in an E. coli fed-batch process: from laboratory to pilot scale

Pilot-scale reactive-extraction technology for fully integrated L-phenylalanine (L-Phe) separation in Escherichia coli fed-batch fermentations was investigated in order to prevent an inhibition of microbial L-Phe production by-product accumulation. An optimal reactive-extraction system, consisting of an organic kerosene phase with the cation-selective carrier DEHPA (di-2-ethylhexyl phosphonic acid) and an aqueous stripping phase including sulphuric acid, was found particularly efficient. Using this system with two membrane contactors, mass-transfer coefficients of up to 288 x 10(-7) cm s(-1) for the aqueous/organic and 77 x 10(-7) cm s(-1) for the organic/stripping phase were derived from experimental data using a simple modelling approach. Concentration factors higher than 4 were achieved in the stripping phase as compared to the aqueous donor phase. Reactive extraction enabled a 98% cation portion of L-Phe in the stripping phase, leading to final product purity higher than 99% after L-Phe precipitation. A doubling of L-Phe/glucose yield was observed when kerosene/DEHPA was added to the fermentation solution in the bioreactor to experimentally simulate a fully integrated L-Phe separation process.     

Determination of the maximum specific uptake capacities for glucose and oxygen in glucose-limited fed-batch cultivations of Escherichia coli

A simple pulse-based method for the determination of the maximum uptake capacities for glucose and oxygen in glucose limited cultivations of E. coli is presented. The method does not depend on the time-consuming analysis of glucose or acetate, and therefore can be used to control the feed rate in glucose limited cultivations, such as fed-batch processes. The application of this method in fed-batch processes of E. coli showed that the uptake capacity for neither glucose nor oxygen is a constant parameter, as often is assumed in fed-batch models. The glucose uptake capacity decreased significantly when the specific growth rate decreased below 0.15 h(-1) and fell to about 0.6 mmol g(-1) h(-1) (mmol per g cell dry weight and hour) at the end of fed-batch fermentations, where specific growth rate was approximately 0.02 h(-1). The oxygen uptake capacity started to decrease somewhat earlier when specific growth rate declined below 0.25 h(-1) and was 5 mmol g(-1) h(-1) at the end of the fermentations. The behavior of both uptake systems is integrated in a dynamic model which allows a better fitting of experimental values for glucose in fed-batch processes in comparison to generally used unstructured kinetic models.     

An improved procedure for production of human epidermal growth factor from recombinant E. coli

An improved procedure for the fermentation and purification of human epidermal growth factor (hEGF) was developed. Recombinant Escherichia coli HB-101 [lacUV5omp08hEGF] harboring plasmid lacUV5omp08hEGF encoding hEGF was used in fermentation to increase levels of hEGF. Medium composition, and the levels of inoculum, inducer (isopropyl-beta-D-thiogalactoside) and ampicillin were optimized with respect to volumetric fermentation of hEGF. As a result, the hEGF concentration reached a high value of 242 mg l(-1) and the amount of heterogeneous protein decreased by 62% compared with that before optimization in batch fermentation. High-quality hEGF was purified from the fermentation culture by centrifugation, salting-out, resuspension, recentrifugation and finally gel chromatography on a Grad-iFrac System using Sephadex G-50 superfine. The purity of hEGF and the total yield were more than 94% and higher than 36%, respectively, and SDS-PAGE of the purified hEGF demonstrated a single band corresponding to an hEGF standard. In particular, a very important phenomenon was found, i.e. that the amount of heterogenous protein in fermentation broths cultured in media with high concentrations of lactose is far less than that cultured in media with high concentrations of glucose.     

The use of elements of the E. coli Ntr-system for the design of an optimized recombinant expression system for high cell density cultivations.

The inducible glnA promoter 2 of the E. coli glutamine synthetase gene is suitable as an expression unit for the production of recombinant proteins at low and high cell densities. It is active when the concentration of ammonium as the sole nitrogen source in the culture medium is below 1 mM. This nitrogen regulatory system was optimized by introduction of expression cassettes consisting of additional elements of the ntr-system. These artificial constructions result in enhanced recombinant gene expression in the production phase. Furthermore, the basic recombinant protein level during the growth phase is reduced due to a tighter promoter control. A three- to four-fold higher accumulation of chloramphenicol-acetyltransferase (as reporter protein) and of anti-EGF-receptor miniantibodies was achieved by increasing the amount of the final regulator molecule NtrC approximately P via plasmidal co-expression of the ntrC gene. The introduction of a modified glnA promoter 1 inverse to glnAp2 lowered the basic activity of glnAp2 to about one half. It is assumed that under nitrogen excess conditions sigma 70-RNA polymerase binds at glnAp1 and thereby prevents most of the binding of sigma 54-RNA polymerase at glnAp2. The optimized expression systems were successfully applied in low and high cell density cultivations. In the fed-batch phase of high cell density cultivations recombinant protein formation was induced through external nitrogen limitation under FIA-controlled concentration of glucose as carbon source.     

Electrophysical monitoring of culture process of recombinant Escherichia coli strains

A novel method for monitoring the cell culture process has been developed. The method is based on the measurements of electro-optical characteristics of cell suspension, calculation of cell structure parameters, and the relationship between accumulation of proteins and change of these parameters' employment. Application of the method for the monitoring of a culture process of a recombinant strain is considered. The process of growth of recombinant strains cannot be sufficiently predicted and the direct measurement of cell culture parameters is unlikely to be the most efficient way of solving the problem.Escherichia coli plasmid-free and recombinant strains synthesizing the fusion protein consisting of tumor necrosis factor-alpha (TNF) and thymosin-alpha(1) (T) were studied. It was found that cytoplasmic electroconductivity of the strains investigated increased during the culture process. The accumulation of insoluble recombinant pThy-315-encoded hybrid protein TNF(SINGLEBOND)T in cells resulted in a decrease of the membrane dielectric permeability. To determine variations of membrane dielectric permeability the amount of insoluble recombinant protein TNF(SINGLEBOND)T in the bacterial cells should be calculated.     

Engineering E. coli for improved microaerobic pDNA production

Escherichia coli strains W3110 and BL21 were engineered for the production of plasmid DNA (pDNA) under aerobic and transitions to microaerobic conditions. The gene coding for recombinase A (recA) was deleted in both strains. In addition, the Vitreoscilla hemoglobin (VHb) gene (vgb) was chromosomally inserted and constitutively expressed in each E. coli recA mutant and wild type. The recA inactivation increased the supercoiled pDNA fraction (SCF) in both strains, while VHb expression improved the pDNA production in W3110, but not in BL21. Therefore, a codon-optimized version of vgb was inserted in strain BL21recA⁻, which, together with W3110recA⁻vgb⁺, was tested in cultures with shifts from aerobic to oxygen-limited regimes. VHb expression lowered the accumulation of fermentative by-products in both strains. VHb-expressing cells displayed higher oxidative activity as indicated by the Redox Sensor Green fluorescence, which was more intense in BL21 than in W3110. Furthermore, VHb expression did not change pDNA production in W3110, but decreased it in BL21. These results are useful for understanding the physiological effects of VHb expression in two industrially relevant E. coli strains, and for the selection of a host for pDNA production.