Biotechnology Letters Vol i0 No 7 Received as revised.June 16
451-456
(1988)
EFFECT OF CULTURE CONDITIONS ON GROWTH AND SPORULATION OF BACILLUS THURINGIENSIS SUBSP. ISRAELENSIS AND DEVELOPMENT OF MEDIA FOR PRODUCTION OF THE PROTEIN CRYSTAL ENDOTOXIN
D. Pearson x and O.P. Ward 2. 1Delta Biotechnology, Castle Court, Castle Blvd, Nottingham, NG7 1FD, England. 2Department of Biology, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1. SUMMARY The influence of medium composition on the inoculum and production stages of the Bacillus thu.rin$iensis subsp, israelensis bioinsecticide fermentation was investigated. Media which inhibited sporulation were selected for inoculum development stages. Bioinsecticide production media were designed to produce high cell counts and > 90% sporulation in a 48h fermentation. Maximum insecticidal activity occurred at the point of maximum bacterial cell lysis/spore release. A process involving two inoculum stages and a 48h production stage in a 40 1 fermenter yielded a viable cell count of 6.5 x 109/ml with greater than 95% sporulation. Good correlation existed between spore counts and bioinsecticide activity. INTRODUCTION Bacillus thuringiensis is an insect pathogen which produces a prote{naceous crystalline toxin that causes paralysis of the larval gut on ingestion (Stockdale 1985). A variant strain, B.thuringiensis subsp, israelensis is the most potent strain of the,, species capable of producing bioinsecticide against larvae of dipteran species such as mosquito and black fly. Production of this crystal follows the end of log phase growth and involves the assembly of proteins synthesised de novo at the beginning of sporulation (Goldberg & Margalit, 1977). Spore and crystal are released into the culture on bacterial lysis. Media for production of bioinsecticide by B.thuringiensis must be designed to optimise biomass, sporulation and cell lysis (Couch & Ross, 1980). It has been suggested that growth and Sporulation of the organism are maximised by high levels of aeration at temperatures of 28-32~ (Dulmage, 1981; Luthy et al., 1982). It is essential that inoculum cultures are produced in non-sporulation media so as to minimise growth tags following inoculation. Our objective was to evaluate media for production of B.thuringiensis subsp, israelensis bioinsecticide and to develop a fermentation process involving two inoculum stages and a production stage. MATERIALS AND METHODS Sources of Raw Materials: Soya flour, defatted (Sheffield Proteins, U.K); Casein (Chamco, Ireland); yeast extract YEP16, corn steep solids and (NH4)2SO4 (Biocon, Ireland); Molasses (United Molasses, U.K.); Starch (Wheat Industries, Ireland); Sucrose (Irish Sugar); CaCO3 (Eglington Stone, U.K.); Na,HPO 4 (Albright and Wilson, Ireland); Siicolapse 5000 (Imperial Chemical Industries, U.Ri.); Tryptone water, plate count agar, nutrient agar (Merck, Germany). All other chemicals were of analytical or reagent grade. Bacterial Culture Conditions: Shake flask cultures were carried out in 11 Erlenmeyer flasks containing 100 and 200ml media incubated at 30~ on an LH Engineering MK-IIIB
451
orbital shaker set at 150 R.P.M. Laboratory fermenter cultivation was carried out in a New Brunswick Microgen Fermenter containing 10 1 medium. Foaming was controlled by addition of 20-30ml o f 1:10 aqueous emulsion of polypropyleneglycol before sterilization. Pilot fermenter cultivation was carried out in a 75 1 fermenter (New Brunswick Inc.) containing 401 medium. Foaming was controlled by use of Silcolapse 5000 antifoam during the fermentation. In both fermenters temperature, impeller speed and aeration settings were 30~ 400 R.P.M. and 1.0 l/l/rain respectively. First Inoculum Stage Medium (1S): 15.0g/1 tryptone water, 7.1g/1 Na,HPO4, 0.2g/1 MgSO4.7HzO , 0.001g/1 Fe.SO4.7H20, 0.005g/1 ZnSO4.7H20, 0.005g/1 C1~SO4.5H20 in distilled water, pH 7.0-7.2. Media were loop inoculated using agar cultures. Second Inoculum Stage Medium (2S): Unless otherwise stated, the medium contained 20g/1 casein, 5.0g/1 dorn steep solids, 2.7g/1 yeast extract, 5.0g/1 cane molasses, 5.0g/1 NazHPO4, pH7.0. Culture Analysis: Proportions of vegetative cells (VC), sporulated cells (SC) and free spores (FS) were determined by diluting the culture 1:5 in sterile diluent and determining the ratios of VC:SC:FS microscopically using a Nikon Optiphot phase contrast microscope at a magnification of X400. The results, presented as ratios of percentages, are the average of three determinations. Biomass optical density measurements were determined by absorhance measurements at 600nm using a 1.0cm light path in a Pye Unicam SP6-500 spectrophotometer. Other methods have been previously described (Pearson & Ward, 1987). RESULTS
In a preliminary investigation using the medium of Drake and Smythe (1963), designated P1, sporulation was never observed. However, good growth and sporulation was obtained when casein was omitted from this medium and the cells were cultured under conditions of high aeration. With other media, high aeration also accelerated sporulation.
Development of an inoculum production process: In order to compare the effects of using vegetative cells or spores to seed inoculum development medium (1S), flasks (1 1 containing 200ml of medium) were inoculated with spores and vegetative cells. Biomass and pH patterns are presented in Figure 1. Apart from a growth lag observed from the spore inoculum, similar fermentation patterns were observed. A spore suspension was therefore used routinely to inoculate the 1S medium. In a related experiment the effects of inoculating the production culture (100ml P3(a) medium in 1 1 flasks) with spores or vegetative cells was investigated. The vegetative cell inoculum was a 5%, 24h culture grown in the 1S medium. The spore suspension was prepared from nutrient agar stock cultures. Biomass, pH patterns, spore formation and cell lysis paterns were followed. The results are presented in Figure 2. The initial vegetative inoculum produced greater sporulation efficiency and spore release on lysis at the production stage. For a scaled up process a second inoculum stage was required. The 1S medium used at the first inoculum stage gave extremely reproducible results and prevented sporulation while maintaining good viability for an extended period of time. However, this medium was too costly for use in a second larger scale inoculum stage. Shake flasks containing various media (100ml) were inoculated with a 24h 5% inoculum of the 1S culture incubated under standard conditions. Viable counts observed and ratios of vegetative 452
cell to sporulated cells to free spores were characterised in the culture after a 20, 24 and 30h incubation (Table 1). Media 2S1 to 2S4 contained no sporulated cells after a 20h incubation. Media 2S2-2S4 contained the highest protein concentration and medium 2S1 had a high protein to carbohydrate ratio. Medium 2S1 was selected for the second inoculum stage as it exhibited superior cell densities by microscopic examination. Viable ceil counts, recorded particularly after a 20h and 24h incubation, were considered to be substantially lower than actual as severe clumping was observed in all cultures at these times. With Medium 2S1, sporulation could be prevented completely up to at least 30h by
using 200ml medium rather than 100ml per 1 1 flask, thereby reducing aeration. Table
1.
No.
Effect o f composition oF stae
2 tnoculget laedium on cell 9rovth and sporulat~on
Hedi,m Composition, g/Z*
Casein
Viable Counts (x 107/ml)
Corn Yeast Steep Holasses Sucrose extract solids
2S1
2S2 20.0 2S3 20.0 2S4 2S5 ,2S6 ZS7 2S8 -
2,5
5.0
5,4 2.7 20.0 10.0 15.0 5,4 5.4
10.0 5.0 ]0.0 10.0 10.0 20.0 20.0
5.0
VC:SC:FS**
Incubation Time (hi 20 24 30
20
Incubation Time (h) 24 30
-
31
28
25
100:0:0
50:50:0
10:90:0
30.0 15.0 30.0 30.0 30.0 30.D 15.0
24 23 )4 30 7 33 19
18 35 31 39 21 29 35
24 23 50 81 60 33 46
100:0:0 100:0:0 |00:0:0 30:70:0 30:70:0 30:70:0 10:90:0
95:5:0 85:15:0 25:75:0 10:80:)0 10:60:30 ]0:90:0 10:85:5
60:40:0 10:90:0 10:90:0 10:25:65 5:50:45 5:85:15 10:80:10
* I n addition all media contained 5.0 g / l Na2HPO4. **Vegetative Cells:Slmrdlated Cells:Free Spores
Characterisation and development of production media-" For the fol]owing production studies a two stage development process was used. A 5% v/v 24h inoculum, developed in the 1S medium, was used to seed the 2S1 medium and a 5% 24h inoculum from this medium was used to seed the production culture.
A number of Bacillus thuringiensis
production media, previously described in the literature, were evaluated. Maxium levels of viable counts achieved for each medium and the incubation times required to give 90% cell lysis are recorded in Table 2. SporuIation was not observed in the P1 medium. Table 2.
Comparative Assessment of Bar
Medium Reference
Pl
thurtnqieaSJ~ production
I12
Drake & SIl)-th Plegoa (19631 (1963)
media.
P3
P4
Ir)ulma(je (19711
CRC (19781
CRC (1978)
3.0 1.0
I0~0 Z.O
-
13~o
5
CmpositI~ 9/1 Casein Soya f l o u r (defatted) Yeast extract Corn steep solids
9Molasses Stash Sucrose Glucose
20.0
14.0 5.4 10.0
15~0
17.0
18.6
4S.O 9.0 -
1o-o
l.O
1o
1~o
1.o
0.o
2.0-2.5
2.0-2.5
1.2-2.5
0.3-1.1
0.7-).S
>96
72
48-72
24-30
48-72
caco,
Time of 901[ lysts (h)
-
5.0
(N'4)2S04
gzxtmm Total Vtable Count { x lO~/ml)
14.0
453
A wide range of variations of P1, P2 and P3 media were examined. Media giving maximum viable counts and greater than 90% cell lysis in a 48h fermentation are summarised in TaNe 3. All of these media contained a significant excess of carbohydrate over protein concentration. Notable was the P3(a) medium which yielded up to 3x109 viable counts/ml and greater than 90% cells lysis in a 48h fermentation. Biomass production and spore release/cell lysis was monitored over the course of a fermentation using the P3(a) medium. The data presented in Figure 3 indicated that maximum insecticide activity occurred after 48 h at the point of maximum cell lysis. Repeated fermentations yielded bioinsecticide activities of 11.5-13.0x104 ITU/ml after a 48h incubation. table 3.
Media fonmlations producing maximamviable counts of Bacillus thurincjiensi_ss and greater than 90% cell lysis in 48 h.
Medium
Composition 9/1 Soya bean meal Soya flour (defatted) Yeast extract Corn steep solids Molasses Starch Glucose 9NazHPO4 CaCO3 Raxtmum Viable Counts (x 109/m1)
Pl(a)
Pl(b)
P2(a)
P2(b)
2.7 5.0 20.0
3~0 5.0
14.0 2.0 5.0
14~0 2.0 5.0
20.0
10.0
5.0
1.5-2.0
P3(b)
15.0
15.0
10.0 IO.O 10.0
5~0
1.5-~.0
P3(a)
10~0
1.0
1.0
1.0
1.0
2.5
2.6
Z.O-3.0
Z.O
Scale up of fermentation process: The inoculum and production media developed for shake flasks cultures were then tested out in a variety of pilot scale fermentation runs. Best results were achieved using a 5% 24h inoculum culture grown in 1S medium to seed a laboratory fermenter containing 101 of 2S inoculum medium. A 16h culture produced in this laboratory fermenter was used to inoculate (5% v/v) a pilot scale 401 fermenter -containing P3(a) medium. Patterns of spore formation, cell lysis and pH observed in the pilot fermenter are illustrated .in Figure 4. Viable count at the end of the 48h fermentation was 6.5 by 109/ml with greater than 95% sporulation. DISCUSSION
Media which were relatively rich in protein content inhibited sporulation and were more suitable for inoculum development stages. Suitable media for sporulation and production
454
o
xiO ii
F
'
"
~r
-
| ii
c
-e 5
i
>
~
C
O,
lO0
i0!
8
f,
g
,,=. (,,,
l[
~00
Tirol:(h)
48
o..~
0.4
g
"i"
G,
E .F.O..~ I
!
50
Q.
~0.2
i1171 / IV I
i, U
0.1
o
24
T i m e (h)
48
72
0
24 Time (h)
48
Fiqure I. Comparative effect of using a vegetative and spore inoculum on the pH and c e l l growth pattern in the f i r s t inoculum stage medium (1S). Vegetative inoculum: A, viable counts; 9 pH; B, 0D600 Spore inoculum: A, viable counts; O, pH; O, OD6D0 Fiqure 2. Comparative effect of vegetative and spore inoculum on the B. thurinqiensis israelensls fermentation pattern in the production culture. Vegetative inoculum: A, viable counts; V, pH; 0, spore formation; B, cell lysis. Spore inoculum: A, viable counts; V, pH; O, spore formation; I"1, cell lysis.
subsp.
Fiqure 3: Relationship between insecticidal b i o a c t i v i t y production (in terms of crystal protein concentration) and cell lysis during flask culture of B. thurinqiensis subsp, israelensis. 9 , crystal protein; D, cell lysis. Fjqure 4: Spore formation and cell lysis during a pilot-scale B. thuringiensis subsp, israelensis fermentation. 9 , Spore formation; A , pH; D , cell lysis.
455
of bioinsecticide contained high carbohydrate to protein ratios. In Bacillus species similar mechanisms appear to be necessary for the triggering of sporulation and synthesis of proteases (Debabov, 1982). Spore formation and protease production by various strains of ,B. thuringiensis appear to be regulated by nitrogen catabolite repression (Egorov, Loriya & Yudina, 1983; Shetsov, Krainova, Voloshin & Kosareva, 1982; Rajalkashmi & Shetna, 1977). The effects of protein content on production of vegetative or sporulated cells may therefore be explained in terms of nitrogen metabolite control. The use of spore count data to select appropriate media for bioinsecticide production was justified by the correlations demonstrated between insecticide potencies, spore numbers and celllysis/spore release. In a variety of fermentation experiments, bioinsecticide activity was always in the range 4.6-5.8 x 104 ITU per 109 spores. Production medium P3(a) was substantially more effective than the media of Drake and Smythe (1963), Dulmage (1981), Megna (1963) and CRC (1978) in terms of spore yields produced in a 48h fermentation. Higher yields of spores were obtained in the pilot fermenter than in laboratory cultures. Aeration may have been more effective in the pilot fermenter as this favours growth and sporulation of 13. thuringiensis (Dulmage, 1981). Yields of cells (>95% sporulated) of 6.5 x 109/ml produced in this fermenter compare favourably with other quoted values. Smith (1982) obtained spore yields of 9 x 108/ml after a 72h fermentation and Goldberg (1980) reported a yield of 4 x 109/ml in a chemostat. Because of the relatively large size of B. thuringiensis cells, viable cell or spore count values tend to be less than those obtained with smaller bacteria. REFERENCES
Couch, T.L. and Ross, D.A. (1980). Biotechnol. Bioeng. 22, 1297-1304. CRC, (1978). British Patent 1,501,563. Debabov, V.G. (1982). In: The Molecular Biology of Bacilli, Vol 1, Bacillus subtilis. D.A. 9 Dubnau, ed., Academic Press, New York, p 331. Drake, B.B. and Smythe, C.V. (1963). U S Patent 3,087,865. Dulmage, H.T. (1981). In: Biological Control on OvpProduction. G.C. Papavisas, ed., Allenheld Osmun, Totowa, p 129. Egorov, N.S., Loriya, Z.K. and Yudina, T.G. (1983). Mikrobiologiya 52, 443-445. Goldberg, I., Sneh, B., Battat, E. and Klein, D. (1980). Biotechnol. Lett. 2, 419-426. Goldberg, L.J. and Margalit, J. (1977). Mosquito News 37, 355-358. Goodman, N.S., Gottfried, R.J. and Rogoff, M.H. (1967). J. Bacteriol 94, 485. Luthy, P., Cordier, J-L. and Fischer, A-M. (1982). Microbial and Viral Pesticides. E. Kurstak, ed., Marcel Dekker, New York, p 35. Megna, J.C. (1963). U S Patent 3,073,7491 Pearson, D. & Ward, O.P. Rajalkashmi, S. and Shetna, Y.I. (1977). Z Indian Inst. Sci. 59; 169-176. Shetsov, V.V., Krainova, O.A., Voloshin, A.G. and Kosareva, N.I. (1982). Mikrobiologiya 5/, 777-779 Smith, R.A. (1982). Can. J.. MicrobioL 28, 1089-1092. Stockdale, H. (1985). Comprehensive Biotechnology, M. Moo-Young, ed., Vol. 3, Pergamon, Oxford, p 949.
456