Regulation of major cultural components for designing a cost effective medium to increase δ-endotoxin synthesis by Bacillus thuringiensis

The present study was aimed at designing a cost effective medium for increasing the δ-endotoxin (Cry protein) synthesis by Bacillus thuringiensis (Bt) kurstaki HD-73 and indigenous Bt JSc1 harboring potential cry genes active against Lepidoptera insect orders. In this regard, locally available cheap molasses as carbon source, soybean extract as nitrogen source, sea water as trace elements source, cystine as sporulation/growth factor were considered to design a cost effective medium. Molasses and soybean extract in place of glucose and peptone in glucose-peptone (GP) medium supported 78.85% increase in δ-endotoxin synthesis in shake flask culture. The effect of cystine on endotoxin synthesis was highly pronounced in two media with a range of 80.32 to 110% higher δ-endotoxin under comparable fermentation conditions. While, substituting basal salts with sea water, Btk HD-73 yielded satisfactory and comparable endotoxin (74.3% of yield with basal salts). It was detected that the rapid decrease of endotoxin synthesis in the culture after 24 h was due to the degradation by the endogenous protease, synthesized with the progress of fermentation. This degradation of the endotoxin was much better protected (1.23 mg/ml endotoxin versus 0.312 mg/ml) by adding 4% ammonium sulfate in the optimized medium. The medium thus formulated with molasses, soybean extract, ammonium sulfate, cystine and sea water was then used in 3.0 L bioreactor cultivation for endotoxin synthesis by both Btk HD-73 and Bt JSc1 under 30% saturation of dO2 through cascade control of agitation and aeration producing a higher yield of δ-endotoxin (2.1 and 2.63 mg/ml, respectively). The present results may successfully be used for large scale production of Bt biopesticide in Bangladesh.


INTRODUCTION
The control of pest populations by using biological pesticides has been an attractive alternative to the application of chemical pesticides (Glazer and Nikaido, 1995;Ciche and Ensign, 2003).As hazardous and recalcitrant chemical agents damage the environment by causing soil, water and air pollutions and also triggers development of insect resistance, biological pest management is more preferable for the specific toxicity of *Corresponding author.E-mail: mmzhoq@gmail.com.Tel: +8801717083673.
Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution License 4.0 International License its component (Massaoud et al., 2010).An entomopathogenic bacterium, Bacillus thuringiensis (Bt), has been used for more than half a century as important biopesticide of which during sporulation produces crystal (Cry) proteins, toxic against various pests (Özkan et al., 2003).Although, Bt based microbial insecticides are available in world market, the high cost makes its application impracticable in large-scale in developing countries.The use of Bt as commercial insecticides would remain prohibitively and relatively expensive if it is not produced with high titer of insecticidal proteins by large-scale fermentation.More attention has been given to the regulation mechanisms that ensure the efficient production of the insecticidal proteins which could be achieved by application of an adequate fermentation technology (Zouari et al., 2002), essentially with use of appropriate media (Zouari and Jaoua, 1999), overcoming metabolic limitations (Zouari et al., 2002) etc.It was reported that the commercial application depends on the cost of raw materials, strain efficiency, fermentation cycle, maintenance of process parameters, bioprocessing of fermentation fluid, and formulation of the final product.The cost of raw materials is one of the principal costs involved in overall Bt production.In the conventional Bt production process, the cost of raw materials varied between 30 and 40% of the total cost depending on the plant production capacity (Ejiofore, 1991).Therefore, local production of this insecticide in developing countries should depend on the use of production media made of cheap, locally available sources including agro-industrial by-products (Ampofo, 1995).
For large scale production of Bt biopesticide, different approaches were investigated to develop suitable media that could support good production of spores and toxins at reasonable costs.Various agricultural and industrial by-products used as raw material in Bt production were citrus peels, wheat bran, corn meal, seeds of dates, beef blood, silkworm pupal skin, ground nut cake, cane molasses, fish meal, cotton seed meal, soybean meal, residues from chicken slaughter house, fodder yeast, cheese whey and corn steep liquor etc (El-Bendary, 2006).Agro-industrial residues and by-products available in southeastern Brazil were used as ingredients for lowcost culture media for liquid fermentation of B. thuringiensis var.kurstaki.Highest spore yield was obtained with a medium containing cheese whey, soya bean milk and molasses (Alves et al., 1997).Other wastes, such as, sludge and broiler poultry litters were also utilized for biopesticides production (Adams et al., 2002;Vidyarthi et al., 2002).In general, two methods of fermentations are used for production of microbial products, submerged fermentation and solid state fermentation.Bt biopesticides are usually composed of spores and crystals protein mixtures, harvested from the production media, readily produced by aerated liquid fermentation.They are easily harvested and have a long shelf life when formulated properly (Ghribi et al., 2007).
Optimizing different culture conditions and regulating some critical factors, it is possible to obtain higher yield in terms of cell mass, Cry protein concentration and toxicity to develop efficient Bt formulations (Dulmage et al., 1990).Critical factors e.g.sugars have significant impact on cell growth but when used at high concentrations, they can cause adverse effects on sporulation due to the acids produced by Bt δ-endotoxin from carbohydrates (Dulmage et al., 1990) and moreover, the balance of the ratio between carbon and nitrogen, itself is directly important for the crystal protein production (Farrera et al., 1998).
Other important components for the production of crystal proteins are the trace minerals (Rose, 1979).Again, amino acids are important in the formation of spores and crystal proteins (Sachidanandham et al., 1997).Moreover, it was reported that decreasing the proteolytic activity in the fermentation medium increased the accumulation of δ-endotoxin in the insecticidal crystal proteins (Ennouri et al., 2013).Indiscriminate use of chemical pesticide is a common practice in Bangladesh, an agriculture dependant developing country, where agro-industrial wastes like defatted soybean meal, defatted mustard seed meal, molasses, rice husk, rice bran, citrus peels etc are generated in huge amount every year.The present study was, therefore, carried out with a view to design a cost effective medium comparing the effects of carbon and nitrogen sources, amino acid such as cystine and basal salts on growth, sporulation and δendotoxin synthesis for large scale production of Bt biopesticide with locally available cheap raw materials such as defatted soybean extract and molasses as nitrogen and carbon source respectively and sea water as the substituent of basal salts.The present study also reports the regulation of certain critical factors that affect the growth, sporulation and δ-endotoxin synthesis by reference Btk HD-73 and indigenous Bt strain JSc1.

B. thuringiensis strains and inoculum
Reference strain B. thuringiensis kurstaki (Btk) HD-73 and the indigenous Bt strain JSc1 were used.Reference strain B. thuringiensis kurstaki (Btk) HD-73 was kindly provided by Okayama University, Japan from their Bt stock collection and the strain Bt JSc1 was isolated from Bangladesh (Shishir et al., 2014).Inoculum was prepared by inoculating a single Bt colony into a 250 ml Erlenmeyer flask containing 50 ml of Luria Bertani (LB) broth (per litre: tryptone 10.0 g, yeast extract 5.0 g, NaCl 10.0 g) and incubated 12 h at 30°C and 180 rpm.Each time, inoculum was added into the medium aseptically with sterile micropipette in a manner so that the process starts with an OD600nm = 0.1.

Production in shake flasks
The fermentation method performed in this study was submerged and Batch type.Each flask containing 100 ml of different media were inoculated aseptically with the bacterial inocula of Btk HD-73 and Bt JSc1 as described in section 2.1 and incubated in an orbital shaker at 30°C with 180 rpm for 72 h.The experiments were performed in triplicate and at least three samples were collected in each case at 24 h interval both from shake flasks and bioreactor for spore count, estimation of δ-endotoxin concentration.

Production in 3.0 L bioreactor
Production experiments were carried out at 30°C in a 3.0 L fully controlled bioreactor (New Brunswick Scientific, USA) containing 2.0 L of finally optimized medium (10% soybean extract, 0.5% molasses, 20 ml sea water, 300 mg/l cystine and 4% ammonium sulfate) as in shake flask.Dissolved oxygen (dO2) level in the medium was automatically controlled and maintained at 30% by providing aeration, set at 1.0 standard liter per minute (SLPM) and agitation at 250 rpm.Dissolved oxygen was continuously monitored by an oxygen sensor but the pH was not controlled.

Estimation of spore count
The spore counting was performed each time in triplicate with 1.0 ml of sample collected during the culture both from shake flask and bioreactor.It was then heat treated at 80°C for 15 min, serially diluted in sterile distilled water and inoculated 0.1 ml sample using glass rod spreader on the LB agar medium by spread plate method.The plates were then incubated at 30°C for 24 h.The colonies were then counted and multiplied by the dilution factor to estimate their number.

Partial purification and determination of δ-endotoxin concentration
The purification of crystal protein was done by the modified method of Liu et al. (1994) and Öztürk et al. (2009) where 1 ml culture was washed twice with sterile distilled water by centrifugation at 10000 rpm for 10 min.The pellet was treated with 1.0 M NaCl and 5.0 mM EDTA and later with 5.0 mM EDTA alone.Finally, the pellet was re-suspended in 1.0 ml 0.1 N NaOH solution for 1 h at room temperature.Thus, the partially purified crystal protein concentration in the supernatant was estimated by Bradford method (Bradford, 1976).

Proteolytic activity assay
Protease activity was determined by a modified method described by Kreger and Lockwood (1981) using azo-casein (Sigma, USA) as the substrate.In this enzyme assay, 400 µl of the culture supernatant and 400 µl of 1% azo-casein solution (suspended in 0.05 M Tris-HCI; pH 8.5) was taken in microfuge tube.The mixture was then incubated in a water bath at 37°C for 60 minutes.The reaction was stopped by adding 135 µl of 35% TCA and kept on ice for 15 min.The solution was then centrifuged at 13000 rpm for 10 min and 750 µl of the supernatant was collected in which equal volume of freshly prepared 1.0 N NaOH was added by gentle mixing.Absorbance (OD440nm) of the solution was then measured using the solution from a parallel reaction as blank where TCA was added before the enzyme.Enzyme activity was then estimated from the absorbance [OD440nm = 1.0 is equivalent to 100 U Enzyme activity].

SDS-PAGE analysis of delta-endotoxin
SDS-PAGE analysis was performed with the partially purified δendotoxin in a 10% separating gel recovered from (NH4)2SO4 supplemented and non-supplemented medium to see its influence on inhibiting protease activity (Sambrook et al., 1989).

Efficiency of molasses and soybean extract as C and N source
The δ-endotoxin production was found to be higher in soybean extract-molasses (SeM) medium in comparison to the glucose-peptone (GP) medium by Btk HD-73 (Table 1) which can be correlated to the replacement of glucose and peptone with 0.5% molasses and 10% soybean extract as these two formulations differed in carbon and nitrogen sources only.At 24 h, the δendotoxin concentration was found to be 0.2 mg/ml in SeM medium which was 78.85% higher than that (0.042 mg/ml) in GP medium.In paired sample t-test analysis, significant difference in endotoxin yield between GP (M = 0.0445, SD = 0.0139) and SeM (M = 0.20933, SD = 0.0139) conditions; t (4) = 14.48 and p = 0.05; was found.

Role of cystine on growth, sporulation and δendotoxin synthesis
The maximum sporulation (11.31Log 10 CFU/ml) and δendotoxin yield (0.215 mg/ml) were obtained at 300 mg/l of cystine at 24 h (Figure 1).The role of cystine (300 mg/l) was also tested in SeM medium and an increase of 1 log in spore concentration and 2 fold in δendotoxin yield were observed (Figure 2).In paired sample t-test analysis, there was significant difference in the endotoxin

Efficacy of sea water as basal salts substituent
The efficacy of sea water (20%) in sporulation and δendotoxin synthesis was comparable with that of basal salts.With both ingredients, maximum sporulation and δendotoxin synthesis were obtained at 72 and 24 h, respectively (Table 2).No significant difference was found between endotoxin yield with basal salt (M = 0.4166, SD = 0.384) and sea water (M = 0.313, SD = 0.384) condition; t (4) = 0.33, p = 0.05; and sporulation with basal salt (M = 6.344,SD = 0.058) and sea water (M = 6.439,SD = 0.058) condition; t (4) = 2.00, p = 0.05 at 24 h by paired sample t-test analysis.Maximum yields of sea water in spore count and δ-endotoxin concentration were respectively 90.88 and 74.29% of Basal salt.

Protection of δ-endotoxin degradation by endogenous protease
Sharp decrease in δendotoxin concentration was observed to be simultaneous with the rise of endogenous protease activity after 24 h determined by protease assay (Figure 3).When the protease activity reached its maximum, that is, 105.9 U/ml gradually, the δ-endotoxin concentration was also reduced to its minimum (0.075 mg/ml) from the peak (0.312 mg/ml).Protease inhibitor PMSF was added into the culture medium at 0.1 and 0.3 mM concentration to control the proteolytic degradation of δ-endotoxin which inhibited the protease activity partially (Data not shown).As an alternative of PMSF, ammonium sulfate was used and the protease activity was monitored up to 72 h at 24 h interval.Ammonium sulfate (4%) resulted in maximum endotoxin yield (1.2 mg/ml) as well as restricted the protease activity within 10.1 U/ml (Table 3).This result corresponded to a 295% increase in δendotoxin productivity (51.43×10 -3 g/L/h) in presence of 4% ammonium sulfate.Significant differences in δendotoxin yield and enzyme activity in the presence and absence of ammonium sulfate were observed while performing the statistical paired sample t-test analysis.The t (4) =62.08, p=0.05 was obtained under condition with endotoxin yield without ammonium sulfate (M=0.32,SD= 0.018) and with ammonium sulfate (M=1.2276,SD= 0.018) at 24 hours.Similarly, the enzyme activity [At 24 hours, enzyme activity without ammonium sulfate (M=44.24,SD= 1.863) and with ammonium sulfate (M= 4.30, SD=1.863) condition; t(4) = 26.2661,p=0.05)]was found to decrease significantly in the presence of ammonium sulfate in paired sample t-test analysis.

Production of reference and indigenous strains in 3.0 L bioreactor
Finally, optimized medium containing soybean extract, molasses, sea water, cystine and 4% ammonium sulfate was used for the production of Btk HD-73 and the indigenous Bt JSc1 under controlled conditions in 3.0 L bioreactor with an working volume of 2.0 L. Maximum δendotoxin yields were 2100 and 2630 mg/l by Btk HD-73 and Bt JSc1, respectively, at 24 h (Figure 4).Protease activity was negligible for both strains.The indigenous Bt JSc1 resulted in 25% higher δ-endotoxin yield than the reference strain.

Qualitative analysis of effect of ammonium sulfate by SDS-PAGE
Prevention of δendotoxin degradation due to    endogenous protease by ammonium sulfate was visualized by SDS-PAGE analysis (Figure 5).Comparatively thicker band of partially purified 133 kD Cry1Ac protein was observed in the presence of ammonium sulfate than in its absence.

DISCUSSION
One of the main reasons of absence of Bt biopesticide in the Integrated Pest Management in Bangladesh agriculture is that it is not yet produced locally.So, industrial production of Bt biopesticide can facilitate its application in eco-friendly pest management for which higher yield in spore and δendotoxin production with low cost medium is one of the prerequisites to keep the product within farmers' buying capacity.In this connection, present study was performed to formulate a low cost medium regulating different cultural components for higher yield.It was reported that high yield of endotoxin could be achieved by using inexpensive defatted soybean, ground nut seed meal extract for large scale production of Bt biopesticide (Vora and Shethna, 1999).In present study, inexpensive substrate such as molasses and soybean was used as carbon and nitrogen sources for biopesticide production instead of relatively expensive glucose and peptone present in commercial media.Soybean extract and Molasses acted as excellent substituent of peptone and glucose respectively as medium containing soybean extract and molasses resulted in 78.85% higher endotoxin yield than that of GP medium.Molasses is cheap, available throughout the year and easy to store.The defatted soybean meal is also a low cost, readily available item and easy to handle.So, a major cost for the production of Bt biopesticide might receive a 10-fold reduction.On the other hand, the increase in δ-endotoxin yield may be due to complex carbon and nitrogen source that permits high sporulation and biomass production.
The optimum concentration of cystine was determined to be 300 mg/l which enhanced the sporulation and δendotoxin synthesis by Bt strains whereas higher concentration inhibited this phenomena.The cystine was also found to keep statistically significant positive impact on yield.This increase could be explained as the fact that cystine might have interfered with some of the macromolecular changes during sporulation and parasporal crystal formation (Rajalakshmi and Shethna, 1980).On the other hand, when basal salts were replaced with sea water containing most of the minerals, comparable effects on δ-endotoxin synthesis and sporulation were observed as ca.75 and 90% yield respectively of basal salts was obtained and the differences were not significant at 95% confidence level.So, another cost incurring ingredient, that is, minerals could successfully be replaced with sea water without much negative influence on yield.It was reported in a study that 20% sea water (source: Mediterranean sea) improved the yield in δ-endotoxin concentration and spore count by 2 and 4%, respectively (Ghribi et al., 2007) whereas this study reports slightly lower yield.It indicates that the yield might be variable based on the source of sea water too.Endogenous protease activity was responsible for the decrease in δ-endotoxin concentration that occurred after 24 h (Figure 3).So, initially PMSF and later ammonium sulfate were used to control this protease synthesis and to protect δ-endotoxin degradation.PMSF inhibited the protease synthesis partially and it might be due to the fact that Btk HD-73 secreted not only serine protease but also some other classes of proteases which could not be inhibited by this serine protease inhibitor (PMSF).But 4% ammonium sulfate inhibited protease synthesis (10.1 U/ml) to great extent and also resulted in higher endotoxin yield (1.23 mg/ml).
Finally, the medium containing 10% soybean extract, 0.5% molasses, 20% sea water, 300 mg/l cystine and 4% ammonium sulfate thus found suitable for optimum production of both spores and δ-endotoxin in shake flask culture, was selected for bioprocess development in a 3.0 L bioreactor.From this study, it was found that the fermentation broth should be harvested at 24 h for the strains both reference, Btk HD-73 and indigenous, Bt JSc1 to recover δ-endotoxin at its maximum yield which also reduces the power consumption, hence cost.So, a low cost medium was designed thus which will facilitate large scale industrial production of Bt biopesticide in Bangladesh in a cost effective manner.

Figure 2 .
Figure 2. Cystine enhanced both endotoxin yield and sporulation by Btk HD-73 similarly in Soybean extract Molasses (SeM) medium.

Figure 3 .
Figure 3.Comparison between increase protease activity and δ-endotoxin degradation with time occurred with Btk HD-73 in cystine supplemented SeM sea water medium.

Figure 4 .
Figure 4. Production kinetics of Bt JSc1 and Btk HD-73 in cystine and ammonium sulfate supplemented SeM sea water medium in a 3.0 L bioreactor.

Table 1 .
Effect of carbon and nitrogen sources on sporulation and δ-endotoxin synthesis by Bacillus thuringiensis kurstaki (Btk) HD-73.

Table 2 .
Efficacy comparison between Sea water and basal salts based on sporulation and δendotoxin synthesis.

Table 3 .
Effect of Ammonium sulfate on protease synthesis and δendotoxin breakdown in cystine supplemented SeM sea water medium.