Stirred Bioreactors Research Articles

Effects of immobilization by entrapment in alginate and scale‐up on paclitaxel and baccatin III production in cell suspension cultures of Taxus baccata

Paclitaxel and baccatin III-producing cells of Taxus baccata were immobilized within Ca(2+)-alginate beads. Under established optimum conditions for the biosynthesis of both taxanes, the yields of paclitaxel and baccatin III in shake-flask cultures of free cells increased by factors of up to 3 and 2, respectively, in the corresponding cultures of immobilized cells. Although the scale-up from shake-flask to bioreactor culture usually results in reduced productivities when both free and immobilized cells were grown in the same optimum conditions in three different bioreactor types (Stirred, Airlift, and Wave) running for 24 days in a batch mode and with the system optimized in each case, there was a considerable increase in the yields of paclitaxel and baccatin III. Among the reactors, the Stirred bioreactor was the most efficient in promoting immobilized cell production of paclitaxel, giving a content of 43.43 mg.L(-1) at 16 days of culture, equivalent to a rate of 2.71 mg.L(-1).day(-1). To our knowledge, the paclitaxel productivity obtained in this study is one of the highest reported so far by academic laboratories for Taxus species cultures in bioreactors.

A High Performance Fermentation Process for Citric Acid Production by Aspergillus niger NGGCB-101, Using Vermiculite as an Additive in a Stirred Bioreactor

The present study describes the use of vermiculite for enhanced citric acid productivity by a mutant strain of Aspergillus niger NGGCB-101 in a stirred bioreactor of 15.0 l capacity. The maximum amount of citric acid (96.10 g/l) was obtained with the control 144 h after mycelial inoculation. To enhance citric acid production, varying levels of vermiculite were added as an additive into the fermentation medium. The best results were observed when 0.20 μg/l vermiculite was added into the medium 24 h after inoculation resulting in the production of 146.88 g citric acid monohydrate/l. The dry cell mass and residual sugar were 11.75 and 55.90 g/l, respectively. Mixed mycelial pellets (1.08–1.28 mm, dia) were observed in the fermented culture broth. When the culture grown at different vermiculite levels was monitored for Qp, Qs and qp, there was a significant enhancement (P≤ 0.05) in these variables over the control (vermiculite-free). Based on these results, it is concluded that vermiculite might affect mycelial morphology and subsequent TCA cycle performance to improve carbon source utilization by the mould, basic parameters for high performance citric acid fermentation.

Dynamic Behavior of Microbial Populations in Stirred Bioreactors Simulated with Euler−Lagrange Methods: Traveling along the Lifelines of Single Cells

In the present work, an Euler−Lagrange approach has been applied to characterize the behavior of a heterogeneous cell population in a stirred-tank bioreactor with nonideal mixing. It allows one to ...

Optimal Design of Two Interconnected Enzymatic Bioreactors

This paper deals with the optimal design of two interconnected Continuous Stirred Bioreactors in which a single enzymatic reaction occurs. The term “optimal” should be understood here as the minimum of the total volume of the reactors required to perform a given conversion rate, given a quantity of matter to be converted per time unit. The optimal design problem is solved for a wide class of kinetic functions including, in particular, the well known Michaelis-Menten kinetic. The analysis of the optimal configurations is investigated, and it is shown that the concept of “Steady State Equivalent Biological System” (SSEBS) first introduced by Harmand et al., (2003a) for microbial reactions only applies to enzymatic systems which have nonmonotonic kinetics. In addition, a stability analysis of the optimal configurations is performed.

Optimal design of two interconnected enzymatic bioreactors

This paper deals with the optimal design of two interconnected continuous stirred bioreactors in which a single enzymatic reaction occurs. The term “optimal” should be understood here as the minimum of the total volume of the reactors required to perform a given conversion rate, given a quantity of matter to be converted per time unit. In determining the optimal volume, it is considered that the input flow may be distributed among the tanks and also that a recirculation loop may be used. The optimal design problem is solved for a wide class of kinetics functions including, in particular, the well-known Michaelis–Menten kinetics function. The analysis of the optimal configurations is investigated, and it is shown that the concept of “Steady State Equivalent Biological System” (SSEBS) first introduced by Harmand et al. [AIChE J., in press] for microbial reactions only applies to enzymatic systems which have non-monotonic kinetics. In addition, a stability as well as a sensitivity analysis of the optimal configurations are performed.

Optimal design of interconnected bioreactors: New results

AbstractThe optimal design of two interconnected continuous stirred bioreactors in which a single reaction occurs is presented. The term “optimal” should be understood here as the minimum of the total volume of the reactors required to perform a given conversion rate, given a quantity of matter to be converted per time unit. In determining the optimal volume, it is assumed that the input flow may be distributed among the tanks and also that a recirculation loop can be used. The analysis of the optimal configurations is investigated, and the concept of “steady‐state equivalent biological system” is highlighted. The class of growth functions for which the results of this study hold is quite wide and includes, in particular, Monod, Haldane, and Aiba kinetics.

Fed-batch fermentation of Ganoderma lucidum for hyperproduction of polysaccharide and ganoderic acid

A process for simultaneous efficient production of bioactive ganoderic acid (GA) and Ganoderma polysaccharide by fed-batch fermentation of Ganoderma lucidum was developed. Effects of carbon source and initial sugar concentration on the production of GA and polysaccharide were studied at first. Sucrose as a carbon source was suitable for the extracellular polysaccharide (EPS) production although the cells could not grow well. Lactose was beneficial for the cell growth and production of GA and intracellular polysaccharide (IPS). When the initial lactose concentration exceeded 35 g/l, the GA accumulation was decreased. The GA production was remarkably improved by pulse feeding of lactose, when its residual concentration was between 10 and 5 g/l. The fed-batch fermentation process in shake flasks could be successfully reproduced in stirred bioreactors (STR) with dissolved oxygen (DO) controlled between 20 and 35% of air saturation during fermentation. The maximum cell density reached 21.89±0.35 and 20.95±0.01 g DW/l in STR and flasks, respectively. The production of EPS reached 0.87±0.05 (STR) and 0.75±0.05 g/l (flask), and the total production of IPS reached 2.49±0.07 (STR) and 2.40±0.05 g/l (flask). The total production of GA reached 367.1±17.4 and 297.9±13.7 mg/l for STR and flask, respectively, and their corresponding productivity was 29.6±0.5 and 14.3±0.1 mg/l per day. Not only the cell density but also the content, production and productivity of GA and IPS obtained in this work are the highest reported in agitated fermentation of G. lucidum.

Ethanol Production at Flask and Pilot Scale from Concentrated Slurries of Steam-Exploded Aspen

This paper investigates the production of ethanol from steam pretreated aspen. The optimal conditions for both the pretreatment and ethanol production were determined. In the former step the parameters investigated were temperature (180−220 °C) and residence time (2−6 min). The most effective combination was 214 °C and 6 min. The exploded substrates were detoxified in three ways, washing with water at 65 °C proving to be the most effective. The substrate was then converted into ethanol via simultaneous saccharification and fermentation. Influences of the reactor type (shaken flasks and stirred bioreactors) and process parameters (solid-to-liquid ratio, enzyme loading, and stirrer speed) have been investigated. The highest ethanol yield obtained from solid-to-liquid ratios of 0.20 g/g was 85% in shaken flasks and 79% in helical stirred bioreactors. In the former case, the ethanol concentration in the broth was 47 g/L. The fermentation unit returns a solid residue with a calorific value of 5612 kcal/kg. The...

Oxygen Transfer in Stirred Bioreactors Under Animal Cell Culture Conditions

Bioreactors designed for animal cell culture may differ significantly in reactor geometry and operating conditions from those used for microbial fermentations. Small impellers, operating at low impeller speeds and low air flow rates, generate only moderate energy dissipation rates so that conventional chemical engineering correlations for oxygen transfer coeffients are usually out of range and unproven in application. A comprehensive oxygen transfer study was conducted in two animal cell bioreactors of 100 litres and 8m 3 in working volume during the production of recombinant CHO cells using a dynamic method. The results indicated that the correlation of van’ Riet 1 for electrolyte solutions predicted the k L a data for the smaller bioreactor surprisingly well, though it consistently overestimated k L a compared with the actual results obtained in the large scale bioreactor. Additional studies conducted in water indicated that the over-estimation resulted not from the differences in the fld dynamics between the systems compared, but from differences in the composition of the solutions.

On-line biomass monitoring in bench-scale stirred bioreactors using parts of a liquid chromatography system

A cell density monitor, comprising parts of a liquid chromatography system, was adapted for on-line real-time measurement of bacterial cell density up to 75 g l−1. The device dilutes the sample containing cells and measures the absorbancy at 280 nm. This detector gives an excellent correlation with off-line measurements, good reproducibility, a low limit of detection (3.5 mg l−1), and becomes a low cost alternative for fermentation monitoring in small stirred bioreactors.

A 3-D Analysis of Gas-Liquid Mixing, Mass Transfer and Bioreaction in a Stirred Bio-Reactor

Gas-liquid stirred vessel two-phase mixing accompanied by bioreaction has been analysed using a 3-D networks-of-zones, in which non-axisymmetric phenomena can be included. The effect of the feed of liquid nutrient from a single dip-pipe can be incorporated so that previous 2-D limitations of axisymmetry are avoided1. The turbulent swirl flow created by the impeller uses clock-wise and anti-clock-wise swirl coefficients, which can be estimated using image-reconstruction 3-D visual modelling8. The simulations can provide detailed predictions of the local gas hold-up distribution, the local mass transfer area, the partial segregation of both the dissolved oxygen and the nutrient and the extent of oxygen depletion of bubbles. The overall gas hold-up and mass transfer area are obviously summations of the local values and the local and overall reaction rates can be predicted as well as the local and overall oxygen absorption fluxes. Simulations are presented for a 3 X [2X (10 X 10)) X 60] configuration of networks-of-zones for a 3 m3 triple-impeller industrial pilot-plant bioreactor. The theoretical predictions are demonstrated using colour-augmented 3D contour maps and solid-body isosurface images created by AVS graphics. Severe non-uniformity of gas hold-up distribution and consequently spatially uneven oxygen mass transfer create significant partial segregation of both oxygen and nutrient. The simulated bioreactor is predicted to be far from perfectly mixed, so that Tylosin producing microorganisms will experience large variations in dissolved oxygen and nutrient concentrations as they circulate around the stirred fermenter.

A Gain-Scheduling Approach for Control of Dissolved Oxygen in Stirred Bioreactors

This paper discusses control of dissolved oxygen in a bioreactor when the stirrer speed is used as control signal. During batch and fed-batch cultivations the operating conditions change significantly which causes tuning problems, especially when performance requirements are high. A linearized process model reveals that the variations in the oxygen dynamics are mainly due to changes in the volumetric oxygen transfer coefficient KLa. To account for the process variations, a control strategy based on PID control and gain scheduling from the stirrer speed is suggested. Controller parameters in different operating regions are obtained using auto-tuning experiments. The approach is easy to implement and does not require any exhaust-gas analysis. Experimental results from a laboratory reactor are presented.

Enhancing xanthan fermentations by different modes of glucose feeding

This paper is the fourth in a series aimed at improving the understanding and operation of conventional agitated fermenters for the production of the commercially important gum, Xanthan. In the first, reproducible fermentations were established and this protocol was used in studies of different agitator types and of bulk mixing and dissolved oxygen concentration in the next two. Here, building on the previous work, the influence of different glucose feeding strategies on Xanthan production in a 20-L agitated fermenter under equivalent conditions of agitation and dissolved oxygen is reported. The biological performances in three types of fed-batch cultures (a two-step glucose addition, multiple glucose-pulse feeding and continuous feeding of glucose) are compared to two batch fermentations with different initial glucose concentrations. The work confirmed that improved performance cannot be achieved by increasing the initial glucose concentration above 50 g/L nor by a single 10 g/L pulse addition (initial glucose concentration of 40 g/L) while significant nitrogen is still present. On the other hand, the simple pulse and continuous feeding strategies, after nitrogen has been essentially exhausted and under conditions of nonlimiting dissolved oxygen and similar bulk mixing, can result in a greatly enhanced performance compared to batch fermentations. Using the final Xanthan gum concentration, the yield on glucose and the overall productivity as performance indices, values of 62 g/L, 0.82 g of Xanthan/g of glucose, and 0.72 g/(L.h), respectively, were obtained compared to literature values for conventional stirred bioreactors of 15-30 g/L, 0.27-0.86 g of Xanthan/g of glucose, and 0. 12-0.43 g/(L.h).

Hydrodynamics of Stirred Bioreactors

This review of the hydrodynamics of stirred bioreactors begins with an introduction to the agitation problems of particular concern in such systems. This is followed by a brief review of some basic concepts in turbulence and rheology of relevance to bioreactors. Important aspects of single phase mixing in low viscosity, high viscosity and Theologically complex broths are then covered in some detail including flow patterns, power number versus Reynolds number plots (including the modification of the latter to allow for shear thinning broths), flow numbers, energy dissipation rates and flow close to impellers and between multiple impeller systems. From these basic principles, the problem of homogenization is then covered in depth because of its significance for bioreactor performance. Aeration concepts are then introduced and the behavior of traditional Rushton turbine impellers is then treated in detail, covering the flow patterns, aerated power characteristics, mixing time and scale-up considerations. The weaknesses of the Rushton turbine are then discussed which leads into a section describing how more modern impellers are able to improve on many of these, especially emphasising their ability to introduce more energy dissipation into the broth and handle more air before flooding, both of which enhance oxygen transfer. The improvement in bulk blending found with multiple axial flow agitators is brought out too. Finally, the retrofitting of fermenters originally containing Rushton turbines with these more modern impellers is discussed. In conclusion, it is clear that there have been substantial increases in the understanding of stirred bioreactor hydrodynamics. However, whilst further understanding will occur within the framework discussed here, the expectation must be that computational fluid dynamics will increase in importance in spite of the difficulty of handling complex rheology, multiphase systems and biological responses. This review article has 135 references.

γ‐Decalactone Production by Trichoderma harzianumin Stirred Bioreactors

AbstractProduction of γ‐decalactone, a compound having peach‐like aroma characteristics, by a strain of Trichoderma harzianumis reported here for the first time. Cultures were carried out in a 14 L fermentor using castor oil as the sole carbon source. The filamentous‐like growth of the microorganism resulted in highly viscous and non‐Newtonian broths, which caused the formation of stagnant zones in the fermentor. T. harzianum was able to produce 260 mg/L of γ‐decalactone after 7 days of culture in a stirred fermenter operated at 3.3 s−1 and 1 vvm of air flow rate. The production of γ‐decalactone was mainly non‐growth associated. The use of high agitation rates—to improve mixing—resulted in celullar lysis. Cell damage occurred at volumetric power drawn as low as 0.2 W/L. Mycelial morphology (mainly hyphal thickness) changed drastically along the culture and was an important influence on the rheology of the broths. All these characteristics make this system an interesting model of study of hydrodynamically complex fermentations involving up to four phases.

Development of scale-down techniques for investigation of recombinant Escherichia coli fermentations: acid metabolites in shake flasks and stirred bioreactors.

We have developed shake-flask screening conditions that are predictive of specific expression of the chimeric toxin, TGF alpha-PE40, by recombinant Escherichia coli JM109 in stirred bioreactors. When a nutrient-rich stirred bioreactor medium was used in shake flasks, neither the extent of growth nor the specific level of recombinant protein expression duplicated the performance in stirred bioreactor fermentations. Incomplete oxidation of glucose and concomitant accumulation of organic acid metabolites, as well as oxygen limitation and lack of pH control, were examined as contributors to the poorer performance in the flask. The medium buffering capacity, initial glucose level, and flask aeration were evaluated to establish the limits of "scale-down" conditions for expression both in a complex nutrient medium (M101) similar to that used in stirred bioreactors and in a defined (FM) medium. Acid metabolites and ethanol were measured as indicators of carbon flow from glucose as well as indirect indicators of oxygen limitation. For the complex M101 medium, optimal shake-flask performance in 250-mL, nonbaffled flasks at 37 degrees C occurred with 0.3 x medium strength, supplementation with 0.3 m HEPES buffer (pH 7.5), and 10 mL of medium per flask. Cultures grown under these conditions produced a maximum density of 3.6 g of dry cell weight/L (as estimated by absorbance measurements at 600 nm) and maintained a pH near neutrality. Additionally, metabolite markers of anaerobic or microaerobic conditions, such as ethanol, lactate, and pyruvate, were not detected, and specific expression of TGF alpha-PE40 was comparable to stirred bioreactors induced for expression at various biomass levels.(ABSTRACT TRUNCATED AT 250 WORDS)