Genetic algorithm-based medium optimization for enhanced production of fluorescent pseudomonad R81 and siderophore

Fluorescent pseudomonad R81, a root-colonizing bacterium, is a potential bio-inoculant due to its plant growth promoting characteristics. It produces hydroxamate-type siderophore which is involved in disease suppression in plants. Genetic algorithm (GA) methodology was applied for the optimization of siderophore and cell mass production simultaneously in shake flask experiments. A total of 10 medium components were optimized within 80 experiments. A high siderophore concentration of 1.9 g/L and cell mass concentration of 2.8 g/L was achieved in the optimized medium. The application of GA was well suited for determination of optimum concentration levels of the medium constituents for a bi-objective function. GA was able to increase the siderophore concentration by 2.8-fold when compared to RSM-based optimization. Further, the batch fermentation of the GA-optimized medium in 14 L bioreactor without pH control produced 2.2 g/L siderophore in 36 h, the highest reported so far. GA was also successfully used to estimate the kinetic parameters of the mathematical models of the batch fermentation.     

Optimization of culture medium composition for cellulolytic bacteria by mathematical methods

The culture medium composition for cellulolytic bacteria growing on sugar cane wastes was optimized. A modified method of Rosenbrock was employed for shaker culture medium and a factorial plan design for fermentor culture medium optimization. A much more economical and productive medium was obtained for the production of single cell protein (SCP). A biomass concentration of 4.3 g/L was obtained in the optimized medium in batch fermentation, in comparison with 2.8 g/L previously obtained in the traditional medium under similar conditions.     

3-Hydroxypropionaldehyde guided glycerol feeding strategy in aerobic 1,3-propanediol production by <Emphasis Type="Italic">Klebsiella pneumoniae</Emphasis>

3-Hydroxypropionaldehyde (3-HPA) is a toxic intermediary metabolite in the biological route of 1,3-propanediol biosynthesis from glycerol. 3-HPA accumulated in culture medium would arouse an irreversible cessation of the fermentation process. The role of substrate (glycerol) on 3-HPA accumulation in aerobic fermentation was investigated in this paper. 1,3-Propanediol oxidoreductase and glycerol dehydratase, two key enzyme catalyzing reactions of 3-HPA formation and consumption, were sensitive to high concentration of 3-HPA. When the concentration of 3-HPA increased to a higher level in medium (ac 10 mmol/L), the activity of 1,3-propanediol oxidoreductase in cell decreased correspondingly, which led to decrease of the 3-HPA conversion rate, then the 3-HPA concentration increasing was accelerated furthermore. 3-HPA accumulation in culture medium was triggered by this positive feedback mechanism. In the cell exponential growth phase, the reaction catalyzed by 1,3-propanediol oxidoreductase was the rate limiting step in 1,3-propanediol production. The level of 3-HPA in culture medium could be controlled by the substrate (glycerol) concentration, and lower level of glycerol could avoid 3-HPA accumulating to a high, lethal concentration. In fed batch fermentation, under the condition of initial glycerol concentration 30 g/L, and keeping glycerol concentration lower than 7–8 g/L in cell exponential growth phase, 3-HPA accumulation could not be incurred. Based on this result, a glycerol feeding strategy was set up in fed batch fermentation. Under the optimized condition, 50.1 g/L of 1,3-propanediol was produced in 24 h, and 73.1 g/L of final 1,3-propanediol concentration was obtained in 54 h.     

基于人工神经网络-遗传算法的樟芝发酵培养基优化

采用优化模型对药用丝状真菌樟芝的复杂发酵过程进行建模,并获得最优发酵培养基组成.对樟芝发酵过程中的形态变化过程进行了观察,并分别采用人工神经网络(ANN)和响应面法(RSM)对樟芝发酵过程进行建模,同时采用遗传算法(GA)优化了发酵培养基组成.结果表明,ANN模型比RSM模型具有更好的实验数据拟合能力和预测能力,GA计算得到樟芝生物量理论最大值为6.2 g/L,并获得发酵最佳接种量及培养基组成:孢子浓度1.76× 105个/mL,葡萄糖29.1 g/L,蛋白胨9.4 g/L,黄豆粉2.8 g/L.在最佳培养条件下,樟芝生物量为(6.1±0.2)g/L.基于ANN-GA的优化方法可用于优化其他丝状真菌的复杂发酵过程,从而获得生物量或活性代谢产物.     

Optimization of process parameters for ethanol production from sugar cane molasses by Zymomonas mobilis using response surface methodology and genetic algorithm

Ethanol is a potential energy source and its production from renewable biomass has gained lot of popularity. There has been worldwide research to produce ethanol from regional inexpensive substrates. The present study deals with the optimization of process parameters (viz. temperature, pH, initial total reducing sugar (TRS) concentration in sugar cane molasses and fermentation time) for ethanol production from sugar cane molasses by Zymomonas mobilis using Box–Behnken experimental design and genetic algorithm (GA). An empirical model was developed through response surface methodology to analyze the effects of the process parameters on ethanol production. The data obtained after performing the experiments based on statistical design was utilized for regression analysis and analysis of variance studies. The regression equation obtained after regression analysis was used as a fitness function for the genetic algorithm. The GA optimization technique predicted a maximum ethanol yield of 59.59 g/L at temperature 31 °C, pH 5.13, initial TRS concentration 216 g/L and fermentation time 44 h. The maximum experimental ethanol yield obtained after applying GA was 58.4 g/L, which was in close agreement with the predicted value.     

蝉拟青霉液体发酵条件优化

采用液体发酵蝉拟青霉,对蝉拟青霉的发酵条件进行优化,以提高蝉拟青霉胞外多糖产量及生物量。摇瓶发酵条件下,在单因素基础上设计正交实验确定各因素的最佳组合。优化后得最佳发酵培养基:蔗糖8%,牛肉膏0.75%,酵母膏0.125%,MgSO_4·7H_2O 0.3%,KH_2PO_4 0.2%,麸皮0.5%。该条件下胞外多糖产量为5.96 g/L,生物量为42 g/L,较优化前提高了1倍。采用发酵罐进行扩大培养,对分批发酵时的初糖浓度进行了优化,并分析了补料分批发酵对发酵过程的影响。发酵罐培养时最适初糖浓度为5%,此时生物量最高为38 g/L,多糖含量最高为5.5 g/L;采用补料分批发酵时,多糖产量最高为5.89 g/L,生物量最高为40 g/L,效果优于分批发酵。     

树状多节孢发酵生产紫杉醇工艺条件的初步研究*

研究了树状多节孢HQD33内生真菌融合子TPF-1摇瓶发酵工艺条件,进行了2.8L和10L通用式机械搅拌罐的发酵试验。结果表明,HQD33适宜发酵工艺条件是:发酵时间16～18d,培养基中蔗糖、苯丙氨酸、醋酸钠、酪氨酸、2,4-二氯苯氧乙酸和亚油酸的加量分别为10g/L、1mg/L、1.5g/L、15mg/L、5mg/L和15mg/L,摇瓶装量为150ml/500ml,在此条件下摇瓶发酵液中紫杉醇平均含量为448.52 g/L; 2.8L和10L罐发酵液中紫杉醇含量达406.95和395.12g/L(平均值)。     

Optimization and scale-up of 2,3-butanediol production by <Emphasis Type="Italic">Bacillus amyloliquefaciens</Emphasis> B10-127

The effects of culture conditions on 2,3-butanediol (2,3-BD) production and its possible scale-up have been studied. A newly isolated Bacillus amyloliquefaciens B10-127, belonged to GRAS microorganisms and showed a remarkable 2,3-BD producing potency, was used for this experiment. Corn steep liquor, soybean meal and ammonium citrate were found to be the key factors in the fermentation according to the results obtained from the Plackett–Burman experimental design. The optimal concentration range of the three factors was examined by the steepest ascent path, and their optimal concentration were further optimized via response surface methodological approach and determined to be 31.9, 22.0 and 5.58 g/l, respectively. The concentration of the obtained 2,3-BD increased significantly with optimized medium (62.7 g/l) when compared with unoptimized medium (45.7 g/l) and the 2,3-BD productivity was about 2.4-fold (The fermentation time was shorten from 72 to 42 h). To observe scale-up effects, batch fermentation was carried out at various working volumes. At a working volume of 20.0 l, the final 2,3-BD concentration and yield were 61.4 and 0.38 g/g at 36 h with a 2,3-BD productivity of 1.71 g/l h. This result shows similar amount of 2,3-BD obtained in lab-scale fermentation, and it is possible to scale up to larger fermentors without major problems.     

A novel recycle batch immobilized cell bioreactor for propionate production from whey lactose

Recycle batch fermentations using immobilized cells of Propionibacterium acidipropionici were studied for propionate production from whey permeate, de-lactose whey permeate, and acid whey. Cells were immobilized in a spirally wound fibrous sheet packed in a 0.5-L column reactor, which was connected to a 5-L stirred tank batch fermentor with recirculation. The immobilized cells bioreactor served as a breeder for these recycle batch fermentations. High fermentation rates and conversions were obtained with these whey media without nutrient supplementation. It took approximately 55 h to ferment whey permeate containing approximately 45 g/L lactose to approximately 20 g/L propionic acid. Higher propionate concentrations can be produced with various concentrated whey media containing more lactose. The highest propionic acid concentration obtained with the recycle batch reactor was 65 g/L, which is much higher than the normal maximum concentration of 35 to 45 g/L reported in the literature. The volumetric productivity ranged from 0.22 g/L . h to 0.47 g/L . h, depending on the propionate concentration and whey medium used. The corresponding specific cell productivity was 0.033 to 0.07 g/L . g cell. The productivity increased to 0.68 g/L . h when whey permeate was supplemented with 1% (w/v) yeast extract. Compared with conventional batch fermentation, the recycle batch fermentation with the immobilized cell bioreactor allows faster fermentation, produces a higher concentration of product, and can be run continually without significant downtime. The process also produced similar fermentation results with nonsterile whey media. (c) 1995 John Wiley & Sons, Inc.     

Production of xylitol in cell recycle fermentations of Candida tropicalis

Xylitol was produced a in two-substrate, batch fermentation with cell recycling of Candida tropicalis ATCC 13803. A series of cell-recycle experiments showed that the feeding of xylose, glucose and yeast extract in the xylitol production phase was most effective in enhancing xylitol productivity. The optimized cell recycle fermentation resulted in 0.82 g xylitol/g xylose yield, 4.94 g xylitol l^–1 h^–1 productivity, and final xylitol concentration of 189 g l^–1. These results were 1.3 times higher in volumetric xylitol productivity and 2.2 times higher in final product concentration compared with the corresponding values of the optimized two-substrate batch culture.     

Substrate and product inhibition kinetics in succinic acid production by Actinobacillus succinogenes

The inhibition of substrate and products on the growth of Actinobacillus succinogenes in fermentation using glucose as the major carbon source was studied. A. succinogenes tolerated up to 143 g/L glucose and cell growth was completely inhibited with glucose concentration over 158 g/L. Significant decrease in succinic acid yield and prolonged lag phase were observed with glucose concentration above 100 g/L. Among the end-products investigated, formate was found to have the most inhibitory effect on succinic acid fermentation. The critical concentrations of acetate, ethanol, formate, pyruvate and succinate were 46, 42, 16, 74, 104 g/L, respectively. A growth kinetic model considering both substrate and product inhibition is proposed, which adequately simulates batch fermentation kinetics using both semi-defined and wheat-derived media. The model accurately describes the inhibitory kinetics caused by both externally added chemicals and the same chemicals produced during fermentation. This paper provides key insights into the improvement of succinic acid production and the modelling of inhibition kinetics.     

Multi-stage high cell continuous fermentation for high productivity and titer

We carried out the first simulation on multi-stage continuous high cell density culture (MSC-HCDC) to show that the MSC-HCDC can achieve batch/fed-batch product titer with much higher productivity to the fed-batch productivity using published fermentation kinetics of lactic acid, penicillin and ethanol. The system under consideration consists of n-serially connected continuous stirred-tank reactors (CSTRs) with either hollow fiber cell recycling or cell immobilization for high cell-density culture. In each CSTR substrate supply and product removal are possible. Penicillin production is severely limited by glucose metabolite repression that requires multi-CSTR glucose feeding. An 8-stage C-HCDC lactic acid fermentation resulted in 212.9 g/L of titer and 10.6 g/L/h of productivity, corresponding to 101 and 429% of the comparable lactic acid fed-batch, respectively. The penicillin production model predicted 149% (0.085 g/L/h) of productivity in 8-stage C-HCDC with 40 g/L of cell density and 289% of productivity (0.165 g/L/h) in 7-stage C-HCDC with 60 g/L of cell density compared with referring batch cultivations. A 2-stage C-HCDC ethanol experimental run showed 107% titer and 257% productivity of the batch system having 88.8 g/L of titer and 3.7 g/L/h of productivity. MSC-HCDC can give much higher productivity than batch/fed-batch system, and yield a several percentage higher titer as well. The productivity ratio of MSC-HCDC over batch/fed-batch system is given as a multiplication of system dilution rate of MSC-HCDC and cycle time of batch/fed-batch system. We suggest MSC-HCDC as a new production platform for various fermentation products including monoclonal antibody.     

Bioconversion of inulin to 2,3-butanediol by a newly isolated Klebsiella pneumoniae producing inulinase

Inulin could be converted to bio-based chemicals by an inulinase producer without external inulinase, and the production of 2,3-butanediol was less than 50 g/L. In this work, a novel inulinase producer of Klebsiella pneumoniae H3 was isolated, and inulinase catalytic properties as well as 2,3-butanediol fermentation were investigated. The enzyme was an intracellular inulinase with an optimal pH of 6 ∼ 7 and a temperature of 30 °C. The use of inulin by H3 was dependent on the degree of polymerization (DP), and the average DP of inulin in fermentation broth increased from 2.82 to 8.08 in 24-h culture of batch fermentation. Acidic pretreatment was developed to increase inulin utilization by adjusting medium pH to 3.0 prior to sterilization. In batch fermentation with optimized medium and fermentation conditions, the concentration of target product (2,3-butanediol and acetoin) was 80.4 g/L with a productivity of 2.23 g/(L⋅h), and a yield of 0.426 g/g inulin.     

Optimization of bio-hydrogen production from biodiesel wastes by Klebsiella pneumoniae

Biodiesel wastes containing glycerol were utilized by Klebsiella pneumoniae DSM 2026 to produce hydrogen. The optimization of medium components was performed using both Plackett-Burman and uniform design methods. Using the Plackett-Burman design, glycerol, yeast extract, NH(4)Cl, KCl and CaCl2 were found to be the most important components, which were further investigated by uniform design and second-order polynomial stepwise regression analysis. The optimized medium containing 20.4 g.L(-1) glycerol, 5.7 g.L(-1) KCl, 13.8 g.L(-1) NH(4)Cl, 1.5 g.L(-1) CaCl(2) and 3.0 g.L(-1) yeast extract resulted in 5.0-fold increased level of hydrogen (57.6 mL/50 mL medium) production compared to initial level (11.6 mL/50 mL medium) after 24 h of fermentation The optimization of fermentation condition (pH, temperature and inoculum) was also conducted. When the strain grew in the optimized medium under optimal fermentation condition in a 5-L stirred tank bioreactor for batch production, hydrogen yield and production reached 0.53 mol/mol and 117.8 mmol/L, respectively. The maximum hydrogen evolution rate was 17.8 mmol/(L.h). Furthermore, 1,3-propanediol (6.7 g.L(-1)) was also obtained from the liquid medium as a by-product.     

An unstructured kinetic model to study NaCl effect on volatile ester fermentation by <Emphasis Type="Italic">Candida etchellsii</Emphasis> for soy sauce production

Salt-tolerant aromatic yeast is an important microorganism arising from the solid state fermentation of soy sauce. The fermentation kinetics of volatile esters by Candida etchellsii was studied in a batch system. The data obtained from the fermentation were used for determining the kinetic parameters of the model. Batch experimental results at four NaCl levels (180, 200, 220, and 240 g/L) were used to formulate the parameter estimation model. The kinetic parameters of the model were optimized by specifically designed Runge-Kutta Genetic Algorithms (GA). The resulting mathematical model for volatile ester production, cell growth and glucose consumption simulates the experimental data well. The resulting new model was capable of explaining the behavior of volatile ester fermentation. The optimized parameters (μ_o, X _max, K _i, α, β, Y _X/S, m, and Y _P/S) were characterized by a correlation of functions assuming salinity dependence. The kinetic models optimized by GA describe the batch fermentation process adequately, as demonstrated by our experimental results.     

Rhamnolipid production in batch and fed-batch fermentation using<Emphasis Type="Italic">Pseudomonas aeruginosa</Emphasis> BYK-2 KCTC 18012P

The optimization of culture conditions for the bacteriumPseudomonas aeruginosa BYK-2 KCTC 18012P, was performed to increase its rhamnolipid production. The optimum level for carbon, nitrogen sources, temperature and pH, for rhamnolipid production in a flask, were identified as 25 g/L fish oil, 0.01% (w/v) urea, 25 and pH 7.0, respectively. Optimum conditions for batch culture, using a 7-L jar fermentor, were 200 rpm of agitation speed and a 2.0 L/min aeration rate. Under the optimum conditions, on fish oil for 216 h, the final cell and rhamnolipid concentrations were 5.3 g/L and 17.0 g/L respectively. Fed-batch fermentation, with different feeding conditions, was carried out in order to increase, cell growth and rhamnolipid production by thePseudomonas aeruginosa, BYK-2 KCTC 18012P. When 2.5 g of fish oil and 100 mL basal salts medium, containing 0.01% (w/v) urea, were fed intermittently during the fermentation, the final cell and rhamnolipid concentrations at 264 h, were 6.1 and 22.7 g/L respectively. The fed-batch culture resulted in a 1.2-fold increase in the dry cell mass and a 1.3-fold increase in rhamnolipid production, compared to the production of the batch culture. The rhamnolipid production-substrate conversion factor (0.75 g/g) was higher than that of the batch culture (0.68 g/g).     

High-level production of heterologous proteins using untreated cane molasses and corn steep liquor in Escherichia coli medium

To develop an economical industrial medium, untreated cane molasses (UCM) was tested as a carbon source for fermentation culturing of Escherichia coli. To test the industrial application of this medium, we chose a strain co-expressing a carbonyl reductase (PsCR) and a glucose dehydrogenase (BmGDH). Although corn steep liquor (CSL) could be used as an inexpensive nitrogen source to replace peptone, yeast extract could not be replaced in E. coli media. In a volume of 40 ml per 1-l flask, a cell concentration of optical density (OD₆₀₀) 15.1 and enzyme activities of 6.51 U/ml PsCR and 3.32 U/ml BmGDH were obtained in an optimized medium containing 25.66 g/l yeast extract, 3.88 g/l UCM, and 7.1% (v/v) CSL. When 3.88 g/l UCM was added to the medium at 6 h in a fed-batch process, the E. coli concentration increased to OD₆₀₀ of 24, and expression of both PsCR and BmGDH were twofold higher than that of a batch process. Recombinant cells from batch or fed-batch cultures were assayed for recombinant enzyme activity by testing the reduction of ethyl 4-chloro-3-oxobutanoate to ethyl (S)-4-chloro-3-hydroxybutanoate (CHBE). Compared to cells from batch cultures, fed-batch cultured cells showed higher recombinant enzyme expression, producing 560 mM CHBE in the organic phase with a molar yield of 92% and an optical purity of the (S)-isomer of >99% enantiomeric excess.     

Efficient conversion of lactic acid to butanol with pH-stat continuous lactic acid and glucose feeding method by Clostridium saccharoperbutylacetonicum

In order to achieve high butanol production by Clostridium saccharoperbutylacetonicum N1-4, the effect of lactic acid on acetone–butanol–ethanol fermentation and several fed-batch cultures in which lactic acid is fed have been investigated. When a medium containing 20 g/l glucose was supplemented with 5 g/l of closely racemic lactic acid, both the concentration and yield of butanol increased; however, supplementation with more than 10 g/l lactic acid did not increase the butanol concentration. It was found that when fed a mixture of lactic acid and glucose, the final concentration of butanol produced by a fed-batch culture was greater than that produced by a batch culture. In addition, a pH-controlled fed-batch culture resulted in not only acceleration of lactic acid consumption but also a further increase in butanol production. Finally, we obtained 15.5 g/l butanol at a production rate of 1.76 g/l/h using a fed-batch culture with a pH-stat continuous lactic acid and glucose feeding method. To confirm whether lactic acid was converted to butanol by the N1-4 strain, we performed gas chromatography–mass spectroscopy (GC-MS) analysis of butanol produced by a batch culture during fermentation in a medium containing [1,2,3-¹³C₃] lactic acid as the initial substrate. The results of the GC-MS analysis confirmed the bioconversion of lactic acid to butanol.     

Fast conversion of glycerol to 1,3-propanediol by a new strain of Klebsiella pneumoniae

Five bacterial strains screened from a batch of 39 samples could convert glycerol anaerobically to 1,3-propanediol (1,3-PD). One of the strains, XJ-Li, which could synthesize 1,3-PD with a higher concentration, was identified and characterized. Phylogenetic analysis of the strain XJ-Li included the study of morphology, physiological and biochemical characteristics. In addition, 16SrDNA sequences were created. The results indicated that this strain is a member of Klebsiella pneumoniae. The optimal cultivation parameters for pH and temperature were determined as 8.0 and 40 °C, respectively. The optimized nitrogen source and carbon source were 6.0 g/L of (NH₄)₂SO₄ and 20 g/L of glycerol, respectively. After 8 h in batch fermentation, both the 1,3-PD concentration and glycerol consumption reached the maximum, with 12.2 g/L of 1,3-PD and 1.53 g/L h of productivity, and a molar yield of 1,3-PD to glycerol of 0.75. Fed-batch fermentation also indicated a higher molar yield of 0.70, and the concentration of 1,3-PD reached 38.1 g/L after 66.4 g/L of glycerol consumption. The results of batch and fed-batch fermentations demonstrated that K. pneumoniae XJ-Li would be an excellent 1,3-PD producer.