Biotechnol. Prog. 1995, 11, 231-234
231
Penicillin-G Enhanced Production of Thuringiensin by Bacillus thuringiensis sp. darmstadiensis Yew-Min Tzeng* and Yu-Hsiang Young Food and Biological Engineering Center, Da-Yeh Institute of Technology, 112 Shanjiao Road, Dahtsuen 515, Taiwan, Republic of China
The effect of penicillin-G on the production of the potential microbial insecticide thuringiensin by Bacillus thuringiensis sp. darmstadiensis was studied. Shake flask and 3-L jar fermentor studies showed t h a t the addition of 360 units/mL penicillin-G at 9 h, when the fermentable sugar in the medium was about to be mostly consumed, improved thuringiensin production by more t h a n l-fold relative to the control. The dosage of 360 units/mL penicillin-G had only a modest effect on the growth of the microorganism. However, cell growth was inhibited at higher dosages of the antibiotic. Since penicillin-G could interfere with cell wall synthesis, which facilitated the release of thuringiensin, a high thuringiensin productivity of 2600 mg/L was attained in this study, which is about 2-10-fold higher t h a n those values reported in the literature.
Introduction Increasing environmental concerns and the ability of insects to develop resistance to chemical insecticides have drawn new interest in microbial insecticides. Bacillus thuringiensis is one of the most commonly used and most extensively investigated producers of microbial insecticides. Thuringiensin is a secondary metabolite of the microorganism that is known to be effective as a means of controlling the larval development of flies. However, one drawback to the production of the exotoxin by fermentation is low productivity. Generally speaking, efforts toward increasing the productivity of thuringiensin fall into four major categories. The first is strain improvement. This method of increasing productivity relies heavily on screening for the best producer (Ohba et al., 1981). The second is to improve the fermentation medium and fermentation conditions (Le., temperature, pH, etc.) by finding optimal composition and fermentation conditions (Mohd-Salleh et al., 1980; Mummigatti and Raghunath, 1988). The third is the study of fermentation processes, in which comparison of the different possibilities of fermentation processes is the major task (Holmberg et al., 1980; Rowe and Margaritis, 1987; Paige and Cooper, 1990; Wu et al., 1993). The fourth one involves the use of antibiotics either to ensure an advantageous environment or to manipulate the physiological conditions of the microorganism to the advantage of producing the target compound. Penicillin-G was found to interfere with cell wall synthesis and facilitate the excretion of metabolites during fermentation and is widely used in glutamic acid fermentation (Phillips and Somerson, 1963). Details of the antibiotic's mode of action is described in the literature (Betina, 1983). Like glutamic acid, thuringiensin is a secondary metabolite. Therefore, it is reasonable to expect that appropriate addition of penicillin-G during the course of fermentation will also have a positive effect on the production of thuringiensin. In this research, effort was devoted to the quantitative study of the effect of penicillin-G on the production of thuringiensin. The present work is concerned with the optimal dosage of penicillin-G and the appropriate time of application of the antibiotic.
Materials and Methods Microorganism and Chemicals. An isolate of Bacillus thuringiensis sp. darmstadiensis (HD-199,provided by Dr. de Barjac, Institute Pasteur, Paris, France) was used in this study. The microorganism was transferred every month on slants of nutrient agar medium (0.8% nutrient broth and 2% agar, both from Difco) and stored a t 4 "C. All chemicals used were of reagent-grade quality. Media. A fermentation medium of the following composition was used (per liter): 15 g of molasses (Taiwan Sugar Co.), 30 g of soy protein, 5 g of KH2PO4, 5 g of KzHPOI, 50 mg of MgSOc7Hz0, 30 mg MnS04. 4H@, 10 mg of FeSO4*7Hz0,50 mg of CaCly7Hz0,1.5 g of NaNH4HP04.4HzO. This medium was modified from selected media reported by Arcas et al. (1987) and Wu et al. (1993). All ingredients, except molasses, were autoclaved together. Preculture. The medium used for preculture contained yeast extract (5 g/L, Difco) and nutrient broth (8 g/L, Difco). A loopful of cells from a nutrient agar plate was inoculated into a l-L Erlenmeyer flask containing 250 mL of medium and cultivated at 30 "C on a rotary shaker (Hotech, Taiwan) a t 200 rpm for 14 h. Fermentation. An inoculum (10% of the working volume) was transferred from the flask of the preculture to the fermentor, which contained 2 L of the initial liquid volume. The fermentation temperature was controlled a t 30 "C. The pH was controlled a t 7.0 using 6 N HzSO4 and 6 N NaOH. The air flow rate was controlled a t 1.5 wm. Equipment. The experimental setup of a 3-L jar fermentor (Mituwa) system was assembled in this laboratory. The sparger located a t the bottom of the reactor was a porous plate with a pore size of 40-50 pm. The temperature of the fermentor was controlled by circulating water. The inlet air was passed through the sparger into the bottom of the reactor after filtration and controlled by a mass flow controller (58513, Brooks). The dissolved oxygen and pH of the fermentation broth were measured by DO (Ingold) and pH sensors (F-600, BradleyJames), respectively. pH and foam control was accomplished by a pH controller (P-171, Mitake) and a foam controller (21 F-GC, National), respectively. All of the signals from the sensors were transferred to an HP-3478
8756-7938/95/3011-0231$09.00/0 0 1995 American Chemical Society and American Institute of Chemical Engineers
Biotechnol. Prog., 1995, Vol. 11, No. 2
232 digital voltmeter and a n HP-3497A data acquisition system (Hewlett-Packard) and were monitored by an IBM PC/AT computer. Assay. Total Sugar. The total sugar concentration was analyzed by using the 2,4-dinitrosalicylic acid reagent and measuring the absorption a t 540 nm (Miller, 1959). Cells and Spores. The numbers of vegetative cells and spores of the strain were counted directly under a microscope (BH2-RFC, Olympus). Thuringiensin. Samples were centrifuged a t 8820g (J2-21M/E, Beckman) for 15 min. The supernatant was then filtered by a membrane (0.45 pm, Gelman). The analysis of thuringiensin was done according to Campbell et al. (1987))Levinson et al. (1990), and Wu et al. (1993) with some modifications. The thuringiensin concentration was measured by HPLC (Shimadzu). The HPLC includes the following components: a n LC 9A pump, a n SPD-6A detector monitoring a t 254 nm, and a 6A integrator. A Merck RP-18 column (15 cm, 3.9 mm i.d.1 was used. The mobile phase was 50 mM KH2PO4, and the pH was adjusted to 2.8 with phosphoric acid. A calibration curve was established using a thuringiensin powder standard (provided by Dr. de Barjac); the thuringiensin concentration was calculated on the basis of the HPLC results and the calibration curve. Penicillin-G Concentration. According to Sigma Co., there are 1690 units of penicillin-G per gram of reagent-grade penicillin-G. In this study, the penicillin-G concentration is calculated according to this figure and expressed as unitdmilliliter.
Results and Discussions Shake Flask Study. Since thuringiensin is a secondary metabolite, the intention is to facilitate the release of thuringiensin by interfering with cell wall synthesis using penicillin-G. Therefore, the time of application and the amount of penicillin-G are both important factors in maximizing the production of thuringiensin. In order to observe the effect of penicillin-G on the production of thuringiensin, different amounts of the antibiotic were added, at 9 or 12 h, to flask cultures. The fermentation temperature was controlled a t 30 "C and the pH at 7.0. Figure 1 shows the effect of different dosages of penicillin-G in terms of thuringiensin production relative to the control (no penicillin-G added). Although the antibiotic was applied a t different times during the course of the culture, the results showed that the dosage of 360 units/ mL gave the highest productivity among the dosages adopted in this study. To determine the appropriate time of application of penicillin-G, a dosage of 360 units/mL was added a t different times to compare relative thuringiensin production. The results are shown in Figure 2. Among the various times of application (5,9, and 12 h) of penicillin-G tested, all showed improved product concentration relative to the control. However, Figure 2 shows that 9 h is the best time of application of the antibiotic for enhancing thuringiensin production. Jar Fermentor Study. The formation of thuringiensin during the course of the fermentation process was believed to be closely related to the time of sporulation (Sebesta et al., 1973). The excretion of thuringiensin from the cell happens when the cell population increases to a certain density, especially during sporulation. On the other hand, Holmberg et al. (1980) suggested that thuringiensin production occurs during the exponential growth phase of the microorganism. Furthermore, recent work completed in this laboratory showed that the
1.2 I
'
0.0 0
-control 240 unWmL (9hr) +360 uniUmL (9 hr) v 720 uniUmL (9 hr) 0 120 unlVmL (1 2 hr) 0 240 unR/mL (1 2 hr) -0- 360 untt/mL (1 2 hr)
10
20
40
30
I
0
Time (hr) Figure 1. Effect of different dosages of penicillin-(=., added at 9 12 h, on relative thuringiensin productivity. The final production of thuringiensin was 1.06 mg/mL (the control was taken as 1 (dimensionless)). 1.2
1.o h
8
P
-
0.8
8
.->
-m e 0.6 v
.-C ul c ?!. 0.4
-
F
't 3
0.2
I
0.0 0
-
P 10
Control 5 hr 9hr - - 0 - - 12hr 0
20
30
40
Time (hr) Figure 2. Effect of different times of application of 360 units/ mL penicillin-(; on relative thuringiensin productivity. The final production of thuringiensin was 1.08 mg/mL (the control was taken as 1 (dimensionless)). maximum rate of thuringiensin release happened a t the maximum rate of exponential growth phase (Wu et al., 1993); the exotoxin accumulation rate was slower after sporulation was almost complete. All of these results lead to the conclusion that thuringiensin is excreted mostly from the viable cells and also released by disruption of the cell through sporulation; thus, it may be possible to improve or enhance thuringiensin productivity by controlling the physiological conditions of the microorganism, for example, acceleration of exotoxin excretion from the viable cells. High-biotin-containing molasses especially is a common carbon source used for fermentation in industry. The presence of biotin in molasses has the effect of increasing the thickness of the cell wall. This, in turn, decreases
Biofechnol. Prog., 1995, Vol. 11, No. 2
233
-
-
Control
0
1o9
360 unWmL 450 uniVmL 675uniVmL
-1
E
1
$ n
5
-C
s
108
J
0 240unWmL +360 uniVmL -450 unWmL 0 675untVmL
PG add
107
I
0
,
1,
t, 10
20
30
Time (hr)
40
50
Figure 3. Viable cell numbers in jar fermentation under different dosages of penicillin-G added at 9 h.
Time (hr) Figure 4. Effect of different dosages of penicillin-G in jar
fermentation on sporulation. the permeability of the cell wall to the secondary metabolite and diminishes the productivity of thuringiensin. The use of penicillin-G to interfere with cell wall synthesis to a proper degree will not only increase the permeability of the cell wall but also hasten the excretion of the secondary metabolite. Another advantage of using penicillin-G is the increase in tolerance to contamination due to the inhibitory effect of penicillin-G on most other undesirable bacteria in the process. To further confirm the optimal dosage of penicillin-G obtained from the shake flask study, and to observe the growth and sporulation behavior under different dosages of the antibiotic, fermentation was carried out in a 3-L jar fermentor system assembled in this laboratory. Operation conditions were rigorously controlled, as described in the Material and Methods section. Initial cell concentrations were controlled a t about 1.7 x los cells/ mL. As can be seen from Figure 3, cell growth was only modestly affected by the presence of the antibiotic a t dosages up to 360 units/mL compared to the control. The maximum cell population was about 5.1 x lo9 cells/mL, which is similar to that reported by Wu et al. (1993) where no antibiotic was applied. However, cell growth was significantly inhibited a t higher dosages (450 and 675 units/mL) of penicillin-G. The effect of penicillin-G concentration on sporulation follows a similar pattern, as shown in Figure 4. The maximum spore concentration was about 5.0 x lo9 spores/mL. Figure 5 shows the effect of penicillin-G dosage on the production of thuringiensin. In comparing the final thuringiensin concentrations resulting from different dosages of penicillin-G, the 360 units/mL dosage improved thuringiensin production by more than 1-fold relative to the control (2.6 mg/mL for a dosage of 360 units/mL and 1.1 mg/mL for the control). The effect of the penicillin-G dosage of 360 units/mL in the jar fermentor was more profound than that in the shake flask, as can be seen by comparing Figure 5 with Figures 1 and 2. This is probably due to the better controlled environment for microbial growth in the jar fermentor. Figure 6 shows a record of the controlled parameters (dissolved oxygen and pH) and the time course of the concentration of total sugar, cell numbers, spore numbers, and the productivity of thuringiensin. The penicil-
3.0
t
2.5
.-
b
v)
C
a,
1 -
0)
C ‘C
1.0
-
3
e
0.5
-Control 0 240uniVmL
0.0
450 unWmL 675unWmL
360 unWmL
0
0
10
20
30
40
50
Time (hr) Figure 5. Effect of different dosages of penicillin-G,added at 9 h, on thuringiensin production.
lin-G was applied a t 9 h, when the fermentable sugar in the medium was about to be mostly consumed. The dissolved oxygen dropped to a minimal level a t about 12 h. According to the time course of the cell population in Figure 6, this corresponds to the end of the exponential growth phase. Sporulation took place about 5 h later. The rate of increase in the thuringiensin concentration is at maximum during the transition between the exponential growth phase and the stationary phase. The rate of increase in the exotoxin concentration was much slower near the end of the stationary phase. Thuringiensin productivities reported in the literature vary from 50 mg (Bond et al., 19691, to 120 mg (Sebesta et al., 1973), to 250-300 mg per liter of cultivation medium (de Barjac and Lecadet, 1976) and 1340 mg/L for the 3-L fermentation without adding penicillin-G (Wu et al., 1993). The production of thuringiensin witnessed in this study was as high as 2600 mg/L with the addition of 360 units/mL penicillin-G 9 h into a 3-L fermentation.
Biotechnol. Prog., 1995, Vol. 11, No. 2
234 8,
Thuringiensin
'I 5
0
t
lo
Po add
20
40
30
Time (hr)
Figure 6. Effect of the addition of 360 units/mL penicillin-(; at 9 h on thuringiensin production for HD-199 fermentation in a 3-L jar fermentor.
Conclusion The addition of penicillin-G facilitates the release of thuringiensin and results in a significant improvement in thuringiensin production. The penicillin-G dosage of 360 unitdml applied at 9 h, when the fermentable sugar in the medium was about to be mostly consumed, increased thuringiensin production by more than 1-fold relative to the control. A high productivity of thuringiensin of 2600 mg/L was attained in this study, which is about 2-10-fold higher than those values reported in the literature. In addition, the results for the optimal dosage of penicillin-G and the appropriate time of application of the antibiotic would be useful in high-celldensity fed-batch cultures of Bacillus thuringiensis for thuringiensin production. Acknowledgment This work was supported by a grant from the National Science Council of Taiwan (NSC-82-0406-E-212-004). Literature Cited Arcas, J.; Yantorno, 0.;Ertola, R. Effect of High Concentration of Nutrients on Bacillus thuringiensis Cultures. Biotechnol. Lett. 1987, 9, 105-110. Betina, V. In The Chemistry and Biology of Antibiotics; Nauta, W. H., Rekker, R. F., Eds.; Elsiver Scientific Publishing Co.: New York, 1983; pp 17-33. Bond, R. P. M.; Boyce, C. B. C.; French, S. K. A Purification and Some Properties of an Insecticide Exotoxin from Bacillus thuringiensis Berilner. Biochem. J . 1969, 114,477-488. Campbell, D. P.; Dieball, D.-E.; Brackete, J. M. Rapid HPLC Assay for Beta-Exotoxin of Bacillus thuringiensis. J . Agric. Food Chem. 1987,35,156-158. de Barjac, H.; Lecadet, M. M. Dosage Biochemique de l'exotoxin #RNA Polymerase Bacterienes. C. R. Acad. Sci. Paris 1976, 282,2119-2122. Holmberg, A.; Sievanen, R.; Carlberg, G. Fermentation of Bacillus thuringiensis for Exotoxin Production: Process Analysis Study. Biotechnol. Bweng. 1980,22, 1707-1024.
Levinson, B. L.; Kasyan, K. J.;Chin, S. S. Identification of BetaExotoxin Production Plasmids Encoding Beta-Exotoxin, and a New Exotoxin in Bacillus thuringiensis by Using High Performance Liquid Chromatograph. J . Bacterzol. 1990,172, 3172-3179. Miller, G. L. Use of Dinitrosalicylic Acid Reagent for Determination of Reducing Sugar. Anal. Chem. 1959,31,426-428. Mohd-Salleh, M. B.; Beegle, C. C.; Lewis, L. C. Fermentation Media and Production of Exotoxin by Three Varieties of Bacillus thuringiensis. J . Znvertebr. Pathol. 1980, 35, 7583. Mummigatti, S. G.; Raghunath, A. N. Production of Bacillus thuringiensis var. kurstaki by Three Different Methods and Ita Relative Toxicity to Bombyx mori. J . Znvertebr. Pathol. 1988,51,115-118. Ohba, M.; Tantichodok,A.; Aizawa, K. Production of Heat-Stable Exotoxin by Bacillus thuringiensis and Related Bacteria. J. Znvertebr. Pathol. 1981,38, 26-32. Paige, M. R.; Cooper, R. D. Scale-up of Beta-ExotoxinProduction in Fed-Batch Bacillus thuringiensis Fermentation. Eur. Congr. Biotechnol., 5th 1990,146-149. Phillips, T.; Somerson, N. L. Production of Glutamic Acid. US. Patent No. 3,080,297, 1963. Rowe, G. E.; Margaritis A. Bioprocess Developments in the Production of Bioinsecticides by Bacillus thuringiensis. CRC, Crit. Rev. Biotechnol. 1987, 6, 87-127. Sebesta, K.; Horska, K.; Vankova, J. Estimation of Exotoxin Production by Different Strains of Bacillus thuringiensis Using 32P-labeUedExotoxin. Collect. Czech. Chem. Commun. 1973,38, 298-303. Wu, M. M.; Tzeng, Y. M.; Hsu, T. H. A Study on High-Yield Fermentation of Thuringiensin Formation Monitored by HPLC Method. J. Technol. 1993, 8, 223-230. Accepted October 3, 1994.@ BP9400856 ~~~
~
~~
~~~
Abstract published in Advance ACS Abstracts, December 15, 1994. @
