Comparison of a production process in a membrane-aerated stirred tank and up to 1000-L airlift bioreactors using BHK-21 cells and chemically defined protein-free medium

The applicability of a protein-free medium for the production of recombinant human interleukin-2 with baby hamster kidney cells in airlift bioreactors was investigated. For this purpose, a BHK-21 cell line, adapted to grow and produce in protein-free SMIF7 medium without forming spheroids in membrane-aerated bubble-free bioreactors, was used as the producer cell line. First, cultivation of the cells was established at a 20-L scale using an internal loop airlift bioreactor system. During the culturing process the medium formulation was optimized according to the specific requirements associated with cultivation of mammalian cells under protein-free conditions in a bubble-aerated system. The effects of the addition of an antifoam agent on growth, viability, productivity, metabolic rates, and release of lactate dehydrogenase were investigated. Although it was possible to establish cultivation and production at a 20-L scale without the use of antifoaming substances, the addition of 0.002% silicon-oil-based antifoaming reagent improved the cultivation system by completely preventing foam formation. This reduced the release of lactate dehydrogenase activity to the level found in bubble-free aerated stirred tank membrane bioreactors and led to a reduction in generation doubling times by about 5 h (17%). Using the optimized medium formulation, cells were cultivated at a 1000-L scale, resulting in a culture performance comparable to the 20-L airlift bioreactor. For comparison, cultivations with protein-containing SMIF7 medium were carried out at 20- and 1000-L scales. The application of protein supplements did not lead to a significant improvement in the cultivation conditions. The results were also compared with experiments performed in a bubble-free aerated stirred tank membrane bioreactor to evaluate the influence of bubbles on the investigated culture parameters. The data implied a higher metabolic activity of the cells in airlift bioreactors with a 150% higher glucose consumption rate. The results of this study clearly demonstrate the applicability of a protein-free chemically defined medium for the production of recombinant proteins with BHK cells in airlift bioreactors.     

Bacterial cellulose production by Acetobacter xylinum in a 50-L internal-loop airlift reactor

Bacterial cellulose (BC) production was realized in a batch cultivation of Acetobacter xylinum subsp. sucrofermentans BPR2001 in a 50-L internal-loop airlift reactor. When the bacterium was cultivated with air supply, 3.8 g/L of BC was produced after 67 hours. When oxygen-enriched gas was supplied, the concentration of BC was doubled and the production rate of BC was 0.116 g/L. h, which was two times higher than that of air-supplied culture and comparable to that in a mechanically agitated stirred-tank fermentor. Bacterial cellulose produced by the airlift reactor formed a unique ellipse pellet (BC pellet), different from the fibrous form which was produced in an agitated stirred-tank fermentor. The BC-pellet suspension was demonstrated to have a higher volumetric oxygen transfer coefficient than the fibrous BC suspension in a 50-L internal-loop airlift reactor. The mixing time of BC-pellet suspension in the airlift reactor was also shorter than that in water.     

Low-pressure airlift fermenter for single cell protein production: II. Continuous culture of pichia yeast

Experiments using Pichia yeast grown on n-paraffins have been conducted in laboratory 10-L airlift fermenters and in a 640-L module of commercial scale. Results confirmed the design concept of combining oxygen transfer and fermenter cooling with low-pressure air. However, in the absence of mass transport constraints, the build up of toxic factors in the fermenter appeared to be a major variable limiting cell productivity. Foaming in the large fermenter also presented a serious problem, which must be solved before low-pressure airlift fermenters become practical.     

响应面法优化自养小球藻产生物柴油油脂

利用响应面法对小球藻(Chlorella vulgaris)在2L气升式生物反应器中对自养产生物柴油油脂的培养条件进行了优化。首先用Plackett-Burman方法对10个相关影响因素的效应进行评价并筛选出对产油有显著影响的3个因素:KNO3浓度、温度和CO2浓度;再用最陡爬坡实验逼近最大产油区域;最后由中心组合实验及响应面分析确定了影响产油主要因素的最佳条件为:KNO3浓度0.31g/L,温度26.5℃,CO2浓度6.80%,最高产油量达到0.42g/L,比优化前提高了近2倍。优化后,在10L气升式生物反应器中进行了扩大培养。     

Production of extracellular protease from crude substrates with dregs in an external-loop airlift bioreactor with lower ratio of height to diameter

Bacillus subtilis AS1.398 was cultivated in a 11.5-L total volume external-loop airlift bioreactor with a low height-to-diameter ratio of 2.9 and a riser-to-downcomer diameter ratio of 6.6 for the production of protease from crude substrates with dregs. The influence of aeration rate, liquid volume, and sparger hole diameter on protease production was investigated. An average of 8197 u/mL protease activity was obtained after a total fermentation time of 32 h in the external-loop airlift bioreactor with a liquid volume of 8.5 L, air flow rate of 1.5 vvm, and sparger hole diameter of 1.5 mm. The addition of one stainless steel sieve plate in the riser of the airlift bioreactor increased productivity of protease. After 32 h of fermentation, an average of 8718 u/mL protease activity was achieved in the external-loop airlift bioreactor with one sieve plate and an air flow rate of 1.2 vvm, liquid volume of 8.5 L, and gas sparger hole diameter of 1.5 mm. This was 9.0% higher than the typical averages of about 8000 u/mL protease activity in the mechanically stirred tank bioreactors of the enzyme factory using the same microorganism. It is possible to make a scale-up of the external-loop airlift bioreactor and feasible to operate it for production of protease from crude substrate with dregs.     

Design and scale up of a process for manganese peroxidase production using the hypersecretory strain Phanerochaete chrysosporium I-1512.

Manganese peroxidase (MnP) production was performed in a airlift bioreactor in which Phanerochaete chrysosporium I-1512, an MnP hypersecretory strain, was immobilized on a stainless steel mesh. Production was scaled up from a 2.5-L bench scale to a 100-L bioreactor. The yield of MnP was increased 2-fold and reached 6600 U L(-1). These results indicate the feasibility of MnP production on a medium scale, which promises sufficient MnP availability for its use in pulp bleaching at industrial scale.     

Processing characteristics of submerged fermentation of Antrodia cinnamomea in airlift bioreactor

Three 5-L airlift bioreactors including airlift reactor with solid draft tube (ALs), airlift reactor with net draft tube (ALn) and bubble column reactor (BC) were investigated for their suitability for cultivating Antrodia cinnamomea, and a stirred tank reactor (ST) was used for comparison. Results indicated that after 7 days fermentation, ALs yielded the highest mycelium content (313 mg/100 mL) and had the lowest dissolved oxygen in the broth. Among different aeration rates (0.025, 0.05, 0.1, 0.5, 1 vvm) used during cultivation of A. cinnamomea in ALs, the aeration rate 0.1 vvm resulted in a volumetric oxygen transfer coefficient of 10.8 h⁻¹ and produced the highest mycelium content. When the optimal conditions were used for the fermentation of A. cinnamomea in an industrial 500-L ALs, the mycelium content in the broth reached 542 mg/100 mL in 28 days. The IC₅₀ values of the ethanol extracts of A. cinnamomea mycelium cultivated in 5-L and 500-L ALs for 28 days were 23 and 17 μg/mL, respectively, for hepatocellular carcinoma cells HepG2. And after 42 days cultivation in 500-L ALs, the IC₅₀ value of the mycelium ethanol extract was reduced to 10 μg/mL.     

A comparison of some methods of estimating volumetric mass-transfer coefficients in an external-loop airlift fermenter

Volumetric mass-transfer coefficients were measured in an 11-L external-loop airlift fermenter with deionized water, a fermentation medium, and during a fermentation. Both a Mackareth oxygen electrode and a novel rapid-response probe were used. When the conventional step-change dynamic method was used for water, the long, nonlinear response time of the Mackareth electrode made correction of its readings difficult; this problem did not occur when the rapid-response probe was used. A comparison was made with a method of mass-transfer coefficient determination which does not involve any assumptions about the gas residence time distribution. However, this method requires that the liquid phase is well-mixed and this requirement was not met in the airlift fermenter. Comparison of the present results with other K(L) a determinations for airlift fermenters showed that K(L) a in water depends on the active gas holdup, the value of K(L) a/epsilon at 20 degrees C being ca. 0.37 s(-1). Although higher gas holdups were obtained with the fermentation medium than for water, the values of K(L) a/epsilon were lower, ca. 0.22 s(-1) at 20 degrees C.     

Kinetic models of ribonucleic acid fermentation and continuous culture by <Emphasis Type="Italic">Candida tropicalis</Emphasis> no.121

During ribonucleic acid fermentation, the fermentative processes were researched at pH controlled at 4.0 and under natural conditions. Unstructured models in a 50-L airlift fermentor were established for batch RNA production at pH 4.0 using the Verhulst equation for microbial growth, the Luedeking–Piret equation for product formation and a Luedeking–Piret-like equation for substrate uptake. Parameters of the kinetic models were determined using origin 7.5. Based on the models estimated above, another batch fermentation experiment was conducted in a 300-L airlift fermentor, which demonstrated that the models could be useful for RNA production on an industrial scale. Additionally, continuous fermentation based on kinetic models was proposed to make full use of substrates and reduce the cost of waste water treatment. As a result, although the DCW and RNA concentration were 11.5 and 1.68 g L⁻¹, which were lower than that of batch fermentation, the sugar utilization increased by 14.3%, while the waste water decreased by more than 90%.     

Microbial biomass production from rice straw hydrolysate in airlift bioreactors

Rice straw is a by-product of rice production, and a great bioresource as raw biomass material for manufacturing value-adding protein for animal feedstock, which has been paid more and more attention. In the present work, utilizing rice straw hydrolysate as a substrate for microbial biomass production in 11.5L external-loop airlift bioreactors was investigated. Rice straw hydrolysate obtained through acid-hydrolyzing rice straw was used for the culture of yeast Candida arborea AS1.257. The influences of gas flow rate, initial liquid volume, hole diameter of gas sparger and numbers of sieve plates on microbial biomass production were examined. The best results in the external-loop airlift bioreactor were obtained under 9.0 L initial liquid volume, 1.1 (v/v)/min gas flow rate during culture time of 0-24 h and 1.4 (v/v)/min gas flow rate of 24-48 h at 29+/-1 degrees C. The addition of the sieve plates in the riser of the external-loop airlift bioreactor increased productivity. After 48 h, under optimized operation conditions, crude protein productivity with one sieve and two sieves were 13.6 mg/mL and 13.7 mg/mL, respectively, comparing 12.7 mg/mL without sieves in the airlift bioreactor and 11.7 mg/mL in the in the 10-L mechanically stirred tank bioreactor. It is feasible to operate the external-loop airlift bioreactors and possible to reduce the production cost for microbial biomass production from the rice straw hydrolysate.     

Improvement of Panax notoginseng cell culture for production of ginseng saponin and polysaccharide by high density cultivation in pneumatically agitated bioreactors

A Panax notoginseng cell culture was successfully scaled up from shake flask to 1.0-L bubble column reactor and concentric-tube airlift reactor. High-density bioreactor batch cultivation was carried out using a modified MS medium. The maximum cell density in batch cultures reached 20.1, 21.0 and 24.1 g/L in the shake flask, bubble column and airlift reactors, respectively, and their corresponding biomass productivity was 950, 1140 and 1350 mg/(L x d) for each. The productivity of ginseng saponin was 70, 96 and 99 mg/(L x d) in the flask, bubble column and airlift reactors, respectively; and the polysaccharide productivity reached 104, 119 and 151 mg/(L x d) for each. Furthermore, a fed-batch cultivation strategy was developed on the basis of specific oxygen uptake rate (SOUR), i.e., sucrose feeding before a sharp decrease of SOUR, and the highest cell density of 29.7 g/L was successfully achieved in the airlift bioreactor on day 17 with a very high biomass productivity of 1520 mg/(L x d). The concentrations of ginseng saponin and polysaccharide reached about 2.1 and 3.0 g/L, respectively, and their productivity was 106 (saponin) and 158 mg/(L x d) (polysaccharide). This work successfully demonstrated the high-density bioreactor cultivation of P. notoginseng cells in pneumatically agitated bioreactors and the reproduction of the shake flask culture results in bioreactors. The cell density, biomass productivity, production titer and productivity of both ginseng saponin and polysaccharide obtained here were the highest that have been reported on a reactor scale for all the ginseng species.     

Fed-batch cultivation of recombinant escherichia coli JM103 and production of the fusion protein SPA::EcoRI in a 60-L working volume airlift tower loop reactor

SPA::EcoRI fusion protein was produced by Escherichia coli JM103 carrying the multicopy expression plasmid pMTC48, the multicopy repressor plasmid pRK248, and the multicopy protection plasmid pEcoR4 in a 60-L working volume airlift tower loop reactor on M9 minimal medium with glucose. Cell mass concentration, total cell count, number of colony-forming units, specific growth rate, yield coefficient, and metabolite (acetate, pyruvate, succinate, lactate, ethanol) concentrations were monitored during the growth phase and gene expression. Gene expression was induced by temperature shift or chemically by isopropyl-thiogalactosidase in the airlift tower loop reactor (ALTR) at constant cultivation time and in a small stirred tank reactor at different cultivation times. During induction, the cultivation medium was supplemented with concentrated Luria-Bertani (LB) medium. The intracellular enzyme activity was evaluated as a function of the time after the start of the induction. It was found that the reduction of the glucose concentration and increase of the dissolved oxygen concentration reduced the acetate produced and increased the intracellular enzyme activity. (c) 1993 John Wiley & Sons, Inc.     

Effect of addition of water-soluble polysaccharides on bacterial cellulose production in a 50-L airlift reactor

Bacterial cellulose (BC) production was carried out in a batch cultivation of Acetobacter xylinum in a 50-L internal loop airlift reactor by addition of water-soluble polysaccharides into the medium. When 0.1% (w/w) agar was added, BC production reached 8.7 g/L compared with 6.3 g/L in the control, and duration of the cultivation period to reach the maximum concentration of BC was almost half of that without addition of polysaccharides. During cultivation, BC was formed into pellets whose size was smaller when the productivity of BC was higher, indicating that increase in the relative viscosity by addition of polysaccharides hindered formation of large clumps of BC and increase in the volumetric oxygen transfer coefficient at high flow rate led to increase in BC productivity.     

Recombinant beta-galactosidase production in serum-free medium by insect cells in a 14-L airlift bioreactor.

Spodoptera frugiperda (Sf9) insect cells were successfully cultured in serum-free medium in a 14-L airlift bioreactor. Cell densities as high as 1 x 10(7) cells/mL were achieved with specific growth rates of approximately 0.0286 h-1 (doubling time of 24 h). This system was also used to demonstrate the expression of a reported gene, beta-galactosidase (beta-gal), when cells were infected with a recombinant baculovirus. Approximately 0.33 mg of beta-gal/mL (i.e., 104,000 units/mL) of medium were obtained at the 14-L scale, while about 0.95 mg of beta-gal/mL (i.e., 285,000 units/mL) of medium were obtained in small-scale shaker flasks. The difference was attributed to a suboptimal infection in the large scale. Specific oxygen consumption rates decreased from 5.58 x 10(-17) mol O2/cell.s in early exponential growth to 3.13 x 10(-17) mol O2/cell.s at 3 days post-infection.     

Characterization study of the sporulation kinetics of Bacillus thuringiensis

A wild-type and an rDNA strain of Bacillus thuringiensis were cultured in a net-draft-tube modified 20-L airlift bioreactor. A comparison of the sporulation patterns suggests that the early sporulation strain has a lower final spore count. Results from off-gas analysis suggests that the CO(2) profile could be an alternative indication to spore counts for the examination of fermentation performance or even the mortality in bioassay of the cultivation product. The difference in mortality tests exhibited by the microorganism was attributed to different patterns of sporulation as well as different levels of gene control inside the cell itself. The sporulation kinetics of B. thuringiensis was simulated by a simple modified Hill equation, where the initial glucose concentration could affect the timing of the onset of sporulation. The equation matches well with the experimental sporulation data for B. thuringiensis in both wild-type and rDNA strains.     

Analysis of petroleum biodesulfurization in an airlift bioreactor using response surface methodology

For the first time, growing cells of Gordonia alkanivorans RIPI90A were used for biodesulfurization (BDS) of diesel. This process was carried out in an internal airlift bioreactor. BDS parameters (oil/water phase ratio and initial sulfur concentration) were optimized in flasks using response surface methodology. Predicted results were found to be in good agreement with experimental results. Initial sulfur concentration had a remarkable effect on BDS process. Maximum removal of sulfur (21 mg/l) can be achieved at oil/water phase ratio of 25% (v/v) and initial sulfur concentration of 28 mg/l. Moreover, effect of superficial gas velocity (Ug) and working volume (v) on volumetric gas liquid mass transfer coefficient was studied in an airlift bioreactor for BDS of diesel. The best results were achieved at Ug and v of 2.5l/min and 6.6l, respectively. Subsequently, BDS of diesel was investigated in an airlift bioreactor under optimized conditions. Sulfur reduction after 30 h was 14 mg/l.     

The influence of vessel height and top-section size on the hydrodynamic characteristics of airlift fermentors

Fermentations of the yeast Saccharomyces cerevisiae were carried out in a 90 to 250-L working volume concentric tube airlift fermentor. Measurements of liquid circulation velocity, gas hold-up, and liquid mixing were made under varying conditions of gas flowrate, vessel height, and top-section size. Both liquid circulation velocity and mixing time increased with vessel height. Liquid velocity varied approximately in proportion to the square root of column height, supporting a theoretically based relationship. The effect of vessel height on gas hold-up was negligible. The height of the top-section had a significant effect on liquid mixing. Mixing time decreased with increasing size of the top-section up to a critical height. As the top-section was expanded beyond this height, little improvement in mixing was seen. This indicated the presence of a two-zone flow pattern in the top-section. Liquid velocity and gas hold-up were essentially independent of top-section height. (c) 1994 John Wiley & Sons, Inc.     

Novel airlift reactor fitting for hairy root cultures: developmental and performance studies

In vitro growth of Solanum chrysotrichum hairy roots was carried out in three different types of reactors: shake flasks, a glass-draught internal-loop 2-L basic design airlift reactor (BDR), and a novel modified mesh-draught with wire-helixes 2-L reactor (MR). In each of them, the growth patterns were different, as well as some of the dynamic parameters. The specific growth rates were 0.08, 0.067, and 0.112 d(-1) for shake flasks, BDR, and MR, respectively. In shake flasks and in the MR, growth followed first-order kinetics. In the MR without roots, superficial liquid velocity in the riser and downcomer ranged from 2.1 to 2.7 and 1.4 to 1.7 cm s(-1), respectively (nearly the same as the BDR values). After 42 days in culture, tissue density in the MR was twice that found in the BDR and about the same as that found in the shake flasks. At the tissue densities reached at 42 days, superficial liquid velocities in the MR and BDR downcomers were 4-5 and 7-8 times lower, and mixing times were 11 and 18 times longer than those observed without roots. Tissue densities measured at three points in the MR's downcomer and riser ranged from 10.21 to 12.17 and 4.94 to 5.24 gDW L(-1) respectively. Dynamic gas hold-up dropped faster when roots grew radially in the mesh-draught. In addition, root cultures were scaled-up in a 10-L MR reactor in which some geometric relations were maintained, such as the Q/V radio. Growth in 10-L MR followed first-order kinetics, but despite this, specific growth velocity was 0.09 d(-1) and overall tissue density diminished slightly with respect to that of the 2-L MR. Tissue inoculation, distribution, and harvest were more easily accomplished in the MRs.     

Comparison of stirred tank and airlift bioreactors in the production of polygalacturonases by Aspergillus oryzae

The production of endo and exo-polygalacturonase (PG) by Aspergillus oryzae IPT 301 was studied in a stirred tank bioreactor (STR) and an internal circulation airlift bioreactor. Using a factorial experimental design, a soluble culture medium was defined which allowed the production of exo- and endo-PG comparable to that obtained in a medium containing suspended wheat bran. The soluble medium was used in tests to compare the production of these enzymes in the STR and airlift bioreactor. In these tests, after 96 h, maximum enzymatic activity values achieved for exo- and endo-PG were 65.2 units (U) per mL and 91.3 U mL⁻¹, in the STR, with similar activity values of 60.6 U mL⁻¹ and 86.2 U mL⁻¹, respectively, being achieved in the airlift bioreactor. The airlift bioreactor also showed satisfactory results regarding the oxygen transfer rate in this process, indicating its potential to be used in an eventual larger scale production of exo- and endo-PG, with lower costs for both installation and operation.