- Email: [email protected]
Cultivation of Bacillus thuringiensis in an airlift reactor with wire mesh draft tubes
Biochemical Engineering Journal 7 (2001) 35–39
Cultivation of Bacillus thuringiensis in an airlift reactor with wire mesh draft tubes Ting-Kuo Huang, Pei-Ming Wang, Wen-Teng Wu∗ Department of Chemical Engineering, National Tsing Hua University, Hsinchu 30043, Taiwan, ROC Received 22 May 2000; accepted 2 September 2000
Abstract An aeration strategy was proposed for foam control in an airlift reactor with double wire mesh draft tubes. The airlift reactor was employed in the cultivation of Bacillus thuringiensis for thuringiensin production. The aeration strategy involved two situations. If the foam rose and touched the foam probe, the air flow rate was dropped to a low value for a certain period. However, if the DO value was already below 10% of the saturation when the air flow rate was dropped, the conventional foam control was employed. The production of thuringiensin based on the proposed strategy was up to 70% higher than that of using the conventional cultivation method with addition of antifoam agents for foam control. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Foaming; Antifoam agent; Aeration; Airlift reactor; Bacillus thuringiensis
1. Introduction Thuringiensin is a bioinsecticide which has shown efficacious control against flies [1], Colorado potato beetles [2], lygus bugs [3] and several species of mites [4]. Production of thuringiensin is commonly carried out using fermentation in cultivation of Bacillus thuringiensis. Thuringiensin which is an exotoxin is a thermostable and water-soluble metabolite of B. thuringiensis. During cultivation of B. thuringiensis for thuringiensin production, oxygen uptake rate of cells is quite high. Hence, the cultivation vessels should have a high oxygen transfer rate. Although stirred tank fermenters have good oxygen transfer rates, the fermenters have high shear stress which is not appropriate in cultivation of B. thuringiensis for thuringiensin production [5]. In this work an airlift reactor with double wire mesh draft tubes was employed. The reactor has a high oxygen transfer rate and low shear stress [6]. It is suitable for cultivation of B. thuringiensis. Foaming in airlift reactors usually is a serious problem. Foam suppression via antifoam agent is commonly employed. Since antifoam agents have negative effects on both cell growth and oxygen transfer, selection of a proper antifoam agent is required. In this study, an antifoam agent which had minimum negative effects on cell growth was selected among six different antifoam agents. For further min∗ Corresponding author. Tel.: +886-3-5742509; fax: +886-3-5728212. E-mail address: [email protected] (W.-T. Wu).
imization of the effects of antifoam agent, an aeration strategy was introduced to foam control in the present study. The aeration strategy involved two situations, if the foam rose and touched the foam probe, the air flow rate dropped to a low value for a period of time and then increased back to the normal operation. However, if the DO value in the fermenter was already below 10% of the saturation, the conventional foam control by adding the antifoam agent was employed. Based on the proposed aeration strategy, antifoam agent was almost not added during the cell growth stage. Cell growth and thuringiensin production had significantly improved.
2. Materials and methods 2.1. Equipment A schematic diagram of the experimental equipment is shown in Fig. 1. The fermentation was carried out in a modified airlift reactor which was 13 cm in diameter and 200 cm high. The reactor contained two separable concentric wire mesh draft tubes which were 1 m high and, respectively 6 and 8.5 cm in diameter. The mesh number of the draft tubes was 3. The reactor which was surrounded by a jacket for temperature control was made of stainless steel (AISI 304). The sparger was a perforated pipe (30 with 1 mm in diameter) located at the bottom of the reactor and between the draft tubes. Without the draft tubes, the airlift reactor
1369-703X/01/$ – see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 1 3 6 9 - 7 0 3 X ( 0 0 ) 0 0 0 9 9 - 1
36
T.-K. Huang et al. / Biochemical Engineering Journal 7 (2001) 35–39
Fig. 1. Schematic diagram of the fermentation system.
became the bubble column. The DO sensor (Ingold) and pH sensor (Bradley–James, Model F-600) were positioned at 45 and 70 cm from the bottom of the reactor, respectively. The foam probe was located at 20 cm from the top plate. All the sensors and probe were interfaced with a control unit, an IBM PC/AT with a PC-LabCard AD/DA Card (Advantech, PCL-812PG, Taiwan).
K2 HPO4 5 g/l, MnSO4 ·4H2 O 0.03 g/l, MgSO4 ·7H2 O 0.05 g/l, CaCl2 ·7H2 O 0.05 g/l, FeSO4 ·7H2 O 0.01 g/l and NaNH4 HPO4 ·4H2 O 1.5 g/l. All substrates except sucrose were sterilized together. The cultivation was carried out with 13 l medium at 30◦ C under an aeration of 32.5 SLPM (2.5 VVM). A 10% (v/v) inoculum ratio was used, pH was controlled at 7.0 by the addition of 1N NaOH and 1N H2 SO4 .
2.2. Microorganism and medium
2.3. Aeration strategy
The microorganism used in this study was Bacillus thuringiensis ssp. darmstadiensis, HD-199 ( provided by Dr. de Barjac of Institute Pasteur, Paris, France). The composition of the fermentation medium was as follows: sucrose 10 g/l, soy flour 60 g/l, KH2 PO4 5 g/l,
The aeration strategy which was employed for foam control was based on foaming and the DO value in the fermenter. The aeration strategy was applied, when the foam rose and touched the foam probe. Based on the DO value, the strategy consisted of two approaches. Firstly, if the DO
T.-K. Huang et al. / Biochemical Engineering Journal 7 (2001) 35–39
concentration was higher than 10% of the saturation, the air flow rate was reduced from 32.5 SLPM (2.5 VVM) to 10 SLPM (0.77 VVM) for 5 s and then increased back to the normal aeration. Secondly, if DO value was lower than 10% of the saturated DO value, the conventional foam control via addition of antifoam agent was employed. 2.4. Analytical methods Sucrose concentration was measured by using the DNS method. The number of spores and vegetative cells were determined using a microscope with a counting chamber. The analysis method of thuringiensin was modified from the method of Levinson et al. [7] with a high performance liquid chromotograph (HPLC) using a Inertsil 7 ODS-3 column, 4.5 mm diameter and 15 cm long. The mobile phase was 50 mM KH2 PO4 /H3 PO4 (pH = 2.8) at 1.5 ml/min.
37
of oxygen transfer between gas and liquid. Six different antifoam agents were tested for their effect on cultivation of B. thuringiensis. They were KM 72 (Shin-Etsu, Japan), Degressal SP 21 (BASF, USA), Degressal SP 22 (BASF, USA), Degressal SP 23 (BASF, USA), Pluronic Pe 6000 (BASF, USA) and Pluronic PE 3100 (BASF, USA). The cells were cultivated in a 1 l flask with 300 ml medium. The flask was in a shaker with 200 rpm at 30◦ C. The pH value of the system was initially at 7, 3 ml of antifoam agent were added after 6 h of cultivation. The time course of cell growth is shown in Fig. 2. Obviously, without adding antifoam agent, cells grew faster and cell number was highest. Among six different antifoam agents, KM 72 showed less inhibition of cell growth. The main component of the antifoam agent was the silicone resin (Shin-Etsu, Japan). Hence, KM 72 was selected as the antifoam agent for cultivation of B. thuringiensis in the airlift reactor. 3.2. Culture in the airlift reactor
3. Results and discussion 3.1. Effect of antifoam agent Antifoam agent has some disadvantages in fermentation performance such as inhibition of cell growth and reduction
Fig. 2. Time courses of cell growth with different antifoam agents, (䊏) without antifoam agent, (䊊) KM 72, (䉱) SP 21, (5) SP 22, (䉬) SP 23, (×) PE 6100, (䊉) PE 3100.
A batch culture of B. thuringiensis was carried out in the airlift reactor with double wire mesh draft tubes. Foaming was controlled by adding the previously selected antifoam agent which was KM 72. The time course of the culture is shown in Fig. 3. The thuringiensin production reached 3.4 g/l after 48 h of cultivation. A batch culture in the bubble column
Fig. 3. The time course of the culture in the airlift reactor with conventional foam control by adding antifoam agent.
38
T.-K. Huang et al. / Biochemical Engineering Journal 7 (2001) 35–39
Fig. 4. The time course of the culture in the bubble column with conventional foam control by adding antifoam agent.
Fig. 5. The time course of the culture in the airlift reactor with the proposed aeration strategy.
was also carried out for comparison. The time course of the cultivation is shown in Fig. 4. Under the same operation conditions, the thuringiensin production was only 1.46 g/l after 48 h of cultivation. Since the dissolved oxygen in the bubble column was always at a very low level, the production of thuringiensin was limited. For minimizing the addition of antifoam agent during the cultivation, the aeration strategy was applied to foam control. The cultivation was carried out in the airlift reactor. The time course of the culture with the proposed aeration strategy is shown in Fig. 5. By applying the aeration strategy, almost no antifoam agent was added during the period of cell growth. When the air flow rate was reduced from 32.5 SLPM (2.5 VVM) to 10 SLPM (0.77 VVM), the level of foaming dropped immediately. Hence, the strategy for foam control could be effectively carried out. The total added antifoam agent was only 20 ml while the conventional method of foam control used 200 ml of antifoam agent for 48 h of cultivation. By using the proposed strategy, the thuringiensin production reached 5.79 g/l after 48 h of cultivation which was about 70% higher than that with the conventional foam control by adding antifoam agent. Jong et al. [8] used glucose as carbon source with single wire mesh draft tube under the conventional foam control and obtained 1.67 g/l of thuringiensin.
4. Conclusion A simple aeration strategy was developed for foam control in an airlift reactor with double wire mesh draft tubes for cultivation of B. thuringiensis. By using the aeration strategy, antifoam agent was seldom added during the cell growth stage. Foam control was mainly carried out by manipulating the air flow rate. The oxygen demand in cultivation of B. thuringiensis for thuringiensin production was quite high. By using the airlift reactor, the dissolved oxygen was always higher than 10% of the saturation. Hence, the proposed aeration strategy could be employed in the airlift reactor. However, the bubble column could not provide sufficient dissolved oxygen. It was not suitable to apply the aeration strategy for foam suppression in the bubble column. The production of the thuringiensin in the airlift reactor with KM 72 as the antifoam agent for foam control was 2.3 times higher than that in the bubble column under the same operating conditions. By using the proposed aeration strategy, the production of thuringiensin in the airlift reactor was further 1.7 times higher than that of using the conventional foam control by adding antifoam agent. The aeration strategy significantly improved the cultivation of B. thuringiensis for thuringiensin production.
T.-K. Huang et al. / Biochemical Engineering Journal 7 (2001) 35–39
Application of the proposed aeration control to other fermentation systems is possible. Nevertheless, the duration and the degree of decrease of the air flow rate should be further investigated. Acknowledgements This work is supported by the National Science Council of Taiwan under the grant NSC 87-2622-E-259-002. References [1] G. Carlberg, R. Lindstrom, Testing fly resistance to thuringiensin produced by Bacillus thuringiensis serotype H-1, J. Inverte. Pathol. 49 (1987) 194–197. [2] G.E. Cantwell, E. Dougherty, W.W. Cantelo, Activity of -exotoxin of Bacillus thuringiensis var. thuringiensis against the Colorado potato beetle and the Ames test, Environ. Entomol. 12 (1983) 1424–1427.
39
[3] L.K. Tangigoshi, D.F. Mayer, H.M. Babcock, Identification of -exotoxin of Bacillus thuringiensis to lygus hesperus, J. Econ. Entomol. 83 (1990) 2200–2206. [4] M.R. Paige, R.D. Cooper, Scale-up of beta exotoxin production in fed-batch Bacillus thuringiensis fermentation, in: Proceedings of the 5th Meeting of European Congress on Biotechnology, 1990, pp. 146–159. [5] J.-Z. Jong, W.-T. Wu, Y.-H. Young, Y.-M. Tzeng, T.-H. Hsu, Morphological variation on cultivation of Bacillus thuringiensis for thuringiensin production, in: Proceeding of Asia-Pacific Biochemical Engineering Conference, Singapore, 1994, pp. 390–392. [6] H.-L. Tung, S.-Y. Chiou, C.-C. Tu, W.-T. Wu, An airlift reactor with double net draft tubes and its application in fermentation, Bioproc. Eng. 17 (1997) 1–5. [7] B. Levinson, K.J. Kasyan, S.S. Chiu, Identification of -exotoxin production, plasmido encoding -exotoxin and a new exotoxin in Bacillus thuringiensis by HPLC, J. Bacteriol. 172 (1990) 3172– 3179. [8] J.-Z. Jong, D.-Y. Hsiun, W.-T. Wu, Fed-batch culture of Bacillus thuringiensis for thuringiensin production in a tower type bioreactor, Biotechnol. Bioeng. 48 (1995) 207–213.
Cultivation of Bacillus thuringiensis in an airlift reactor with wire mesh draft tubes Ting-Kuo Huang, Pei-Ming Wang, Wen-Teng Wu∗ Department of Chemical Engineering, National Tsing Hua University, Hsinchu 30043, Taiwan, ROC Received 22 May 2000; accepted 2 September 2000
Abstract An aeration strategy was proposed for foam control in an airlift reactor with double wire mesh draft tubes. The airlift reactor was employed in the cultivation of Bacillus thuringiensis for thuringiensin production. The aeration strategy involved two situations. If the foam rose and touched the foam probe, the air flow rate was dropped to a low value for a certain period. However, if the DO value was already below 10% of the saturation when the air flow rate was dropped, the conventional foam control was employed. The production of thuringiensin based on the proposed strategy was up to 70% higher than that of using the conventional cultivation method with addition of antifoam agents for foam control. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Foaming; Antifoam agent; Aeration; Airlift reactor; Bacillus thuringiensis
1. Introduction Thuringiensin is a bioinsecticide which has shown efficacious control against flies [1], Colorado potato beetles [2], lygus bugs [3] and several species of mites [4]. Production of thuringiensin is commonly carried out using fermentation in cultivation of Bacillus thuringiensis. Thuringiensin which is an exotoxin is a thermostable and water-soluble metabolite of B. thuringiensis. During cultivation of B. thuringiensis for thuringiensin production, oxygen uptake rate of cells is quite high. Hence, the cultivation vessels should have a high oxygen transfer rate. Although stirred tank fermenters have good oxygen transfer rates, the fermenters have high shear stress which is not appropriate in cultivation of B. thuringiensis for thuringiensin production [5]. In this work an airlift reactor with double wire mesh draft tubes was employed. The reactor has a high oxygen transfer rate and low shear stress [6]. It is suitable for cultivation of B. thuringiensis. Foaming in airlift reactors usually is a serious problem. Foam suppression via antifoam agent is commonly employed. Since antifoam agents have negative effects on both cell growth and oxygen transfer, selection of a proper antifoam agent is required. In this study, an antifoam agent which had minimum negative effects on cell growth was selected among six different antifoam agents. For further min∗ Corresponding author. Tel.: +886-3-5742509; fax: +886-3-5728212. E-mail address: [email protected] (W.-T. Wu).
imization of the effects of antifoam agent, an aeration strategy was introduced to foam control in the present study. The aeration strategy involved two situations, if the foam rose and touched the foam probe, the air flow rate dropped to a low value for a period of time and then increased back to the normal operation. However, if the DO value in the fermenter was already below 10% of the saturation, the conventional foam control by adding the antifoam agent was employed. Based on the proposed aeration strategy, antifoam agent was almost not added during the cell growth stage. Cell growth and thuringiensin production had significantly improved.
2. Materials and methods 2.1. Equipment A schematic diagram of the experimental equipment is shown in Fig. 1. The fermentation was carried out in a modified airlift reactor which was 13 cm in diameter and 200 cm high. The reactor contained two separable concentric wire mesh draft tubes which were 1 m high and, respectively 6 and 8.5 cm in diameter. The mesh number of the draft tubes was 3. The reactor which was surrounded by a jacket for temperature control was made of stainless steel (AISI 304). The sparger was a perforated pipe (30 with 1 mm in diameter) located at the bottom of the reactor and between the draft tubes. Without the draft tubes, the airlift reactor
1369-703X/01/$ – see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 1 3 6 9 - 7 0 3 X ( 0 0 ) 0 0 0 9 9 - 1
36
T.-K. Huang et al. / Biochemical Engineering Journal 7 (2001) 35–39
Fig. 1. Schematic diagram of the fermentation system.
became the bubble column. The DO sensor (Ingold) and pH sensor (Bradley–James, Model F-600) were positioned at 45 and 70 cm from the bottom of the reactor, respectively. The foam probe was located at 20 cm from the top plate. All the sensors and probe were interfaced with a control unit, an IBM PC/AT with a PC-LabCard AD/DA Card (Advantech, PCL-812PG, Taiwan).
K2 HPO4 5 g/l, MnSO4 ·4H2 O 0.03 g/l, MgSO4 ·7H2 O 0.05 g/l, CaCl2 ·7H2 O 0.05 g/l, FeSO4 ·7H2 O 0.01 g/l and NaNH4 HPO4 ·4H2 O 1.5 g/l. All substrates except sucrose were sterilized together. The cultivation was carried out with 13 l medium at 30◦ C under an aeration of 32.5 SLPM (2.5 VVM). A 10% (v/v) inoculum ratio was used, pH was controlled at 7.0 by the addition of 1N NaOH and 1N H2 SO4 .
2.2. Microorganism and medium
2.3. Aeration strategy
The microorganism used in this study was Bacillus thuringiensis ssp. darmstadiensis, HD-199 ( provided by Dr. de Barjac of Institute Pasteur, Paris, France). The composition of the fermentation medium was as follows: sucrose 10 g/l, soy flour 60 g/l, KH2 PO4 5 g/l,
The aeration strategy which was employed for foam control was based on foaming and the DO value in the fermenter. The aeration strategy was applied, when the foam rose and touched the foam probe. Based on the DO value, the strategy consisted of two approaches. Firstly, if the DO
T.-K. Huang et al. / Biochemical Engineering Journal 7 (2001) 35–39
concentration was higher than 10% of the saturation, the air flow rate was reduced from 32.5 SLPM (2.5 VVM) to 10 SLPM (0.77 VVM) for 5 s and then increased back to the normal aeration. Secondly, if DO value was lower than 10% of the saturated DO value, the conventional foam control via addition of antifoam agent was employed. 2.4. Analytical methods Sucrose concentration was measured by using the DNS method. The number of spores and vegetative cells were determined using a microscope with a counting chamber. The analysis method of thuringiensin was modified from the method of Levinson et al. [7] with a high performance liquid chromotograph (HPLC) using a Inertsil 7 ODS-3 column, 4.5 mm diameter and 15 cm long. The mobile phase was 50 mM KH2 PO4 /H3 PO4 (pH = 2.8) at 1.5 ml/min.
37
of oxygen transfer between gas and liquid. Six different antifoam agents were tested for their effect on cultivation of B. thuringiensis. They were KM 72 (Shin-Etsu, Japan), Degressal SP 21 (BASF, USA), Degressal SP 22 (BASF, USA), Degressal SP 23 (BASF, USA), Pluronic Pe 6000 (BASF, USA) and Pluronic PE 3100 (BASF, USA). The cells were cultivated in a 1 l flask with 300 ml medium. The flask was in a shaker with 200 rpm at 30◦ C. The pH value of the system was initially at 7, 3 ml of antifoam agent were added after 6 h of cultivation. The time course of cell growth is shown in Fig. 2. Obviously, without adding antifoam agent, cells grew faster and cell number was highest. Among six different antifoam agents, KM 72 showed less inhibition of cell growth. The main component of the antifoam agent was the silicone resin (Shin-Etsu, Japan). Hence, KM 72 was selected as the antifoam agent for cultivation of B. thuringiensis in the airlift reactor. 3.2. Culture in the airlift reactor
3. Results and discussion 3.1. Effect of antifoam agent Antifoam agent has some disadvantages in fermentation performance such as inhibition of cell growth and reduction
Fig. 2. Time courses of cell growth with different antifoam agents, (䊏) without antifoam agent, (䊊) KM 72, (䉱) SP 21, (5) SP 22, (䉬) SP 23, (×) PE 6100, (䊉) PE 3100.
A batch culture of B. thuringiensis was carried out in the airlift reactor with double wire mesh draft tubes. Foaming was controlled by adding the previously selected antifoam agent which was KM 72. The time course of the culture is shown in Fig. 3. The thuringiensin production reached 3.4 g/l after 48 h of cultivation. A batch culture in the bubble column
Fig. 3. The time course of the culture in the airlift reactor with conventional foam control by adding antifoam agent.
38
T.-K. Huang et al. / Biochemical Engineering Journal 7 (2001) 35–39
Fig. 4. The time course of the culture in the bubble column with conventional foam control by adding antifoam agent.
Fig. 5. The time course of the culture in the airlift reactor with the proposed aeration strategy.
was also carried out for comparison. The time course of the cultivation is shown in Fig. 4. Under the same operation conditions, the thuringiensin production was only 1.46 g/l after 48 h of cultivation. Since the dissolved oxygen in the bubble column was always at a very low level, the production of thuringiensin was limited. For minimizing the addition of antifoam agent during the cultivation, the aeration strategy was applied to foam control. The cultivation was carried out in the airlift reactor. The time course of the culture with the proposed aeration strategy is shown in Fig. 5. By applying the aeration strategy, almost no antifoam agent was added during the period of cell growth. When the air flow rate was reduced from 32.5 SLPM (2.5 VVM) to 10 SLPM (0.77 VVM), the level of foaming dropped immediately. Hence, the strategy for foam control could be effectively carried out. The total added antifoam agent was only 20 ml while the conventional method of foam control used 200 ml of antifoam agent for 48 h of cultivation. By using the proposed strategy, the thuringiensin production reached 5.79 g/l after 48 h of cultivation which was about 70% higher than that with the conventional foam control by adding antifoam agent. Jong et al. [8] used glucose as carbon source with single wire mesh draft tube under the conventional foam control and obtained 1.67 g/l of thuringiensin.
4. Conclusion A simple aeration strategy was developed for foam control in an airlift reactor with double wire mesh draft tubes for cultivation of B. thuringiensis. By using the aeration strategy, antifoam agent was seldom added during the cell growth stage. Foam control was mainly carried out by manipulating the air flow rate. The oxygen demand in cultivation of B. thuringiensis for thuringiensin production was quite high. By using the airlift reactor, the dissolved oxygen was always higher than 10% of the saturation. Hence, the proposed aeration strategy could be employed in the airlift reactor. However, the bubble column could not provide sufficient dissolved oxygen. It was not suitable to apply the aeration strategy for foam suppression in the bubble column. The production of the thuringiensin in the airlift reactor with KM 72 as the antifoam agent for foam control was 2.3 times higher than that in the bubble column under the same operating conditions. By using the proposed aeration strategy, the production of thuringiensin in the airlift reactor was further 1.7 times higher than that of using the conventional foam control by adding antifoam agent. The aeration strategy significantly improved the cultivation of B. thuringiensis for thuringiensin production.
T.-K. Huang et al. / Biochemical Engineering Journal 7 (2001) 35–39
Application of the proposed aeration control to other fermentation systems is possible. Nevertheless, the duration and the degree of decrease of the air flow rate should be further investigated. Acknowledgements This work is supported by the National Science Council of Taiwan under the grant NSC 87-2622-E-259-002. References [1] G. Carlberg, R. Lindstrom, Testing fly resistance to thuringiensin produced by Bacillus thuringiensis serotype H-1, J. Inverte. Pathol. 49 (1987) 194–197. [2] G.E. Cantwell, E. Dougherty, W.W. Cantelo, Activity of -exotoxin of Bacillus thuringiensis var. thuringiensis against the Colorado potato beetle and the Ames test, Environ. Entomol. 12 (1983) 1424–1427.
39
[3] L.K. Tangigoshi, D.F. Mayer, H.M. Babcock, Identification of -exotoxin of Bacillus thuringiensis to lygus hesperus, J. Econ. Entomol. 83 (1990) 2200–2206. [4] M.R. Paige, R.D. Cooper, Scale-up of beta exotoxin production in fed-batch Bacillus thuringiensis fermentation, in: Proceedings of the 5th Meeting of European Congress on Biotechnology, 1990, pp. 146–159. [5] J.-Z. Jong, W.-T. Wu, Y.-H. Young, Y.-M. Tzeng, T.-H. Hsu, Morphological variation on cultivation of Bacillus thuringiensis for thuringiensin production, in: Proceeding of Asia-Pacific Biochemical Engineering Conference, Singapore, 1994, pp. 390–392. [6] H.-L. Tung, S.-Y. Chiou, C.-C. Tu, W.-T. Wu, An airlift reactor with double net draft tubes and its application in fermentation, Bioproc. Eng. 17 (1997) 1–5. [7] B. Levinson, K.J. Kasyan, S.S. Chiu, Identification of -exotoxin production, plasmido encoding -exotoxin and a new exotoxin in Bacillus thuringiensis by HPLC, J. Bacteriol. 172 (1990) 3172– 3179. [8] J.-Z. Jong, D.-Y. Hsiun, W.-T. Wu, Fed-batch culture of Bacillus thuringiensis for thuringiensin production in a tower type bioreactor, Biotechnol. Bioeng. 48 (1995) 207–213.