Development of cost-effective media for the culture of Chilo partellus larvicide in Kenya

Stem borers (Chilo partellus) are important field insect pests of maize (Zea mays L.) and sorghum (Sorghum bicolor L.) in Africa. They account for more than 30% yield losses depending on the composition of the pest community. C. partellus larvicide like Bacillus thuringiensis have been widely and effectively used in C. partellus control programs, but the industrial production of theses bacilli is expensive. Here we have attempted to develop three cost-effective media, based on legumes, potato, and whey. Growth and production of the insecticidal proteins from these bacteria were satisfactory; protein concentration yields of 27.60 mg/ml, spore counts of 5.60 × 10 8 CFU/ ml and first-instar Chilo partellus larvicidal activity (LC50) of 78 μg/l were obtained with a 72 h culture of this bacterium. Therefore, this investigation suggests that legume, potato and whey-based culture media are more economical and effective for the industrial production of B. thuringiensis insecticidal crystal proteins.


INTRODUCTION
In Kenya, the spotted stem borers (Chilo partellus) destroy an estimated 400,000 metric tons or 13.5% of farmers annual harvest of maize costing about >US$72 million (De Groote, 2002).The management of C. partellus has largely been based on chemicals, which are rarely effective particularly due to misuse and resistance development by the pest (Camilla, 2000).In addition, small scale farmers, who form the bulk of the maize producers in Kenya, cannot afford those (Bonhof et al., 2001).Technologies that can reduce yield losses from C. partellus damage are necessary to increase maize *Corresponding authors.E-mail: dankwalimwa@gmail.com.Tel: +254-714-002-880.production to cope with increasing demand for maize in Kenya.Berliner (Bt) was thus considered as a possible component in such a pest management system and a series of investigations was undertaken to elucidate its potential for inclusion in such a program (Mugo et al., 2007).In the recent past, it has been the most successful commercial biopesticide with its worldwide application (Pena et al., 2007).When compared with the chemical pesticides, Bt has the advantages of being biologically degradable, selectively active on pests and less likely to cause resistance (Lambert and Peferoen, 1992).Bt synthesizes an insecticidal cytoplasmic protein inclusion during the stationary phase of its growth cycle (Pena et al., 2007).These crystalline inclusions comprise relatively high quantities of one or more glycoproteins known as delta-endotoxins or cryotoxins (Schnepf et al., 1998).The insecticidal activity of Bt was attributed largely or completely (depending on the insect) to these parasporal crystals (Bravo et al., 2007).The crystal proteins accumulate in the mother cell and are released upon completion of sporulation (Aronson et al., 1995).Usually Bt strains isolated locally are more effective than imported strains due to higher specificity on target host, greater field persistence due to higher adaptation to the natural environment and toxicity at a higher temperature range (Brownbridge, 1991).To derive full benefit from the Bt based biopesticides, there is a need for studying influence of carbon-source from cost-effective raw materials on growth, sporulation and crystal protein production by local Bt strains that would be used as biopesticides for the indigenous crops.

Bacteria
Cultures of B, thuringiensis subspecies kenyae (Bt) were provided by Mr. Richard Rotich, from the Kenya Agricultural Research Insitute (KARI), Nairobi Kenya.Also used in this study were Bt strains isolated from soils and termite mounds collected from Kalunya Glade and Lirhanda Hill in Kakamega Forest, Kenya and also from soil samples from Juja, Kenya.

Bacterial culture media
The conventional laboratory culture broth Nutrient Yeast Extract Medium (NYSM), used as reference medium in the present study was prepared by mixing glucose (5 g), peptone (5 g), NaCl (5 g), beef extract (3 g), yeast extract (0.5 g), mineral solutions (10.0 g/l cow blood; 0.02 g/l MnCl2.4H20(s.d. fine-chem ltd); 0.05 g/l MgSO4.7H20(Lab Tech Chemicals); and 1.0 g/l CaCO3 (Sigma-Aldrich, Germany) (10 ml) in an appropriate volume (1 L) of double distilled water (pH 7.5).Legumes, potato and whey-based culture media was collected from farms and brought to the laboratory, washed in tap water, air-dried and stored at room temperature.A known quantity of these dried substrates (10 g/l) was boiled in ordinary tap water for 15 min.After cooling, the extracts were filtered and the pH of the filtrate was adjusted (pH 7.5). 1 L volume of the extract medium was dispensed in each of the three conical flasks (2 L capacities each) for culturing Bt.Similarly, flasks were kept for conventional medium (NYSM) also.All the culture media were autoclaved (at 120 °C/20 lb/in 2 /20 min).

Bacterial growth
A small amount of Bt was inoculated separately in 2 ml each NYSM medium and allowed to grow for 12 h at 37°C as pre-cultures (50 µl each) were inoculated into sterile culture media.The cultures were allowed to grow under constant agitation (120 rev/min) at 37°C in an orbital shaker.Culture samples (2 ml) were drawn from each culture medium at 6 h intervals from 0 to 72 h.The pH and culture turbidity were measured using a digital pH meter and UV-VIS spectrophotometer.These were also examined microscopically for the presence of spore-crystal mixtures (Maniatis et al., 1982).

Total viable cell count and spore count
Total viable cell and spore counts were determined in the final whole culture by the pour plate method.Serial decimal dilutions of the final whole culture were made in sterile 1% peptone water (Oxoid), and 0.5 ml of each dilution in triplicate was added to a Petri dish, followed by the addition of 10 ml of plate count agar (Oxoid) at 45°C.The culture and agar were mixed thoroughly and allowed to set.Plates were incubated at 32°C for 24 to 48 h.Plates with 30 to 300 colonies were counted with a colony counter (Gallenkamp Ltd.).For spore counts, cultures were pasteurized at 65°C for 20 min before serial dilutions were made (Hoben and Somasegaran, 1982).

Spore-crystal toxin recovery from culture media
As soon as the cultures were fully sporulated, the spore-crystal toxin complex was recovered by centrifugation (10,000 g/30 min/4°C) using super speed centrifuge and the spore-crystal free supernatants were discarded.The spore-crystal mixtures were thoroughly washed three times each with 0.1 M NaCl and sterile double distilled water (10,000 g/30 min/4°C).Finally, these were washed with protease inhibitor, phenyl methyl sulphonyl fluoride (PMSF, 1mM, Sigma), re-suspended in the required volume of double distilled water and stored at -20°C, until further use, for biochemical studies and toxicity bioassays (Wessel and Flugge, 1984).

Protein estimation
A small volume of the stored spore/crystal sample was centrifuged (10,000 g/15 min/4°C) and the pellets were solubilized in solubilization buffer (50 mM NaHCO3, 10 mM dithiothreitol, pH 10) and incubated for 2 to 3 h, at 25 to 30°C.After centrifugation and extraction (same rpm as above), the pure solubilized protein (from Bt) was quantified for protein estimation (Bradford, 1976) with bovine serum albumin (BSA, Sigma) as standard.

SDS-PAGE
A total of 5 µg protein equivalent samples from Bt spores-crystals (NYSM and the test media) was mixed with an equal volume of sample loading buffer and boiled for 5 min and separated on sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) unit, according to Laemmli (1970).The protein bands were stained with Coommasie Brilliant Blue R-250 and visualized.

Toxicity studies
Powders of Bt produced from the three cost-effective media and control were assayed against laboratory-reared first instar larvae of C. partellus.A standard primary powder of Bt subspecies kurstaki (Btk) was included in the assay for comparison.100 mg of NYSM, potato, legumes, and whey powder was suspended respectively in 1,000 ml of distilled water containing 1% (vol/vol) Tween 80.Serial dilutions of this suspension were made in distilled water.15 larvae were added to 150 ml of each dilution in 250 ml white plastic cups.Three cups were used per dilution.Controls consisted of three cups each containing 150 ml of distilled water and 15 larvae for each powder assayed.Larval food was provided by adding a small portion of finely ground oaf flakes (Quaker) mixed with dried yeast powder.Each experiment was incubated at 20 ± 5°C for 48 h, and each assay repeated three times.Observations were made at 6 h for paralysis and knockdown effects.Mortality counts were made at 24 h and 48 h.A larva was presumed dead if it did not move when touched with a blunt needle.

Statistical studies
A one-way ANOVA test was used to compare mean maximum spore count among media and pairwise comparison of the media was done using the Duncan's multiple comparison test based on least significance difference.Probit analysis for calculation of LC50 values was carried out using the statistical software SPSS 18.0 for windows.

RESULTS
During activation and multiplication of the isolates, the growth of the local Bt and Btk isolates occurred between 24 to 72 h producing smooth creamish-white colonies which were rough edged and slightly raised from the nutrient agar.Under the culture conditions described, lysis of the Bt cells was complete after 72 h and most of the released protein crystals settled at the bottom of the flask.Almost complete separation of the endotoxin protein crystals from the spores and cell debris was achieved by decantation of the frothy spent culture and high speed centrifugation at 10,000 rpm where the crystals formed a white pellet at the bottom of the tube leaving the spores and cell debris in the supernatant fraction.Serial washing, decanting and centrifugation rid the crystals of spent culture components, spores and cell debris.Microscopy revealed a high concentration of the crystalline inclusions in the pellets obtained (Figure 1).
The highest optical densities were obtained at 37°C from isolates: 62LBG37°C, Bt 20, 63KAG37°C, 1SKAG37°C, Bt 47, Bt 12 and Bt 54 respectively.The results of optical density show that pH 7 was the most suitable pH for the maximum growth of 62LBG37C, Bt 21, Bt 47, and Bt 52 while pH 7.5 was suitable for 14SLA30°C and pH 5.5 was suitable for Bt 20.
The mass of the resulting pellets ranged from 0.460 g for Bt 30 to 0.225 g for Bt 20 (mean = 0.314 g ±0.084 g) (Table 1).The protein mass of the pellets was significantly different among the isolates (t 16.616, 19, 0.000 ).Spore counts in the pellets ranged from 4.89 × 10 8 to 5.60 × 10 8 (mean = 5.23 ±1.53) (Table 1).There was significant difference (t 16.014, 19, 0.000 ) on the percent protein content in the pellets.Isolates Bt 30 and Bt 47 recorded higher contents while 24LBN30°C, 63KAG37°C, and Bt 20 had low protein content.The protein yield in the nutrient broth ranged from 2.22 to 4.60 mg/ml of broth (mean = 3.112 mg ± 0.938 mg) and was significantly different (t 23.328, 19, 0.000 ) across the different Bt isolates.
When compared to high molecular weight standards in SDS-PAGE analysis, the solution and pellet of the dissolved spore-crystal product from each treatment had proteins with molecular weights of approximately 110 to 120 and 60 to 70 kDa (Figures 2 and 3).The major polypeptides present in the spore-crystal complex of Bt produced from Legume medium (1SKAG37°C and 24LBN30°C), NYSM medium (58SLA25°C and Bt 20), potato medium (58SLA25°C and Bt 30) and whey medium   Crystal toxins were correspondingly related to their larvicidal activity.

S C S C
The neonate larvae started feeding immediately after introduction into the Petri dish and preferred the underside of the leaf.Their feeding intensity slowed down with time and some larvae moved away from the meal after 24 h.Most larvae in the treatment especially with higher delta-endotoxin concentrations stopped feeding after 48 h, appeared weak and stunted in growth compared to the control upon when death was also observed.For instance, with the 0.015 mg/ml endotoxin treatment, 78% of the larvae were found to be dead from the diet after 48 h.After 48 h, only 10% of larvae were dead with the 0 mg/ml endotoxin treatment, 72% were dead with the 0.15 mg/ml endotoxin treatment, 49% dead with the 1.5 mg/ml endotoxin treatment and 20% of the larvae were dead with the 15 mg/ml endotoxin treatment.Leaf damage was observed to be less on inoculated leaf disks compared to that in the control.Upon death, the larvae appeared dark and shrunk compared to the control.A dead larva was washed, ground and aseptically inoculated onto nutrient agar plate, creamish growth was observed around the larvae which confirmed that the larval mortality was due to ingestion of Bt endotoxins.In the set of starved larvae, mortality was 100% in 144 h (Table 2).A 10% larval mortality on the control treatment was observed after 48 h while 30% recorded after 48 h (Table 2).The cause of this mortality was probably due to drastic changed of weather and laboratory conditions from source of larvae to bioassay laboratory resulting in weak neonates, dehydration or infection; in the subsequent bioassays, eggs were sourced in the yellow state and allowed to acclimatize to the bioassay lab conditions before hatching and maize leaves were also thoroughly washed with distilled water, the filter paper in the Petri dish wetted daily with distilled water and the larvae placed back on the leaf if they had moved far away and got trapped under the filter paper which greatly reduced control larvae mortality.The LC 50 value estimate for reference isolate Btk was 0.011 mg/ml after 48 h (regression coefficient = 0.031005, 95% confidence limit, SE = 0.13747).
Among the different Bt treatments, only isolates 24LBN30°C, 63KAG37°C and Bt 20 recorded mortality of 10% at 48 h of observation (Table 3).However, while Btk increased mortality towards the end of the observation, Bt 30, Bt 47 and Bt 54 stabilized at 60% from 48 h.
Although, isolate 58SLA25°C recorded the first mortality of 40% at 24 h, it is the only isolate that recorded 73% mortality in the observation period, at 48 h.The Btk recorded 83% mortality at 48 h.No mortality was observed with 14SLA30°C, ISKAG37°C, Bt 21, Bt 12 and the control throughout the experiment period.Calculations of LT 50 shows that Btk was the most toxic, causing 50% mortality after only 24 h, followed by 58SLA25°C at 48 h and Bt 30 at 48 h.One way ANOVA (repeated measures) revealed that the difference in percent larval mortalities of the standard isolate Btk and the Bt isolates was statistically significant (p<0.05)except the isolates with less than 20% mortality; Bt 37 (F (1.0,5.0)= 6.4,p>0.05) and Bt 20 (F (1.0,5.0)= 0.122, p>0.05).

DISCUSSION
The growth of the Kenyan isolates on nutrient agar and nutrient broth was similar to that reported by Wamaitha (2006) which shows the viability of local (Kenyan) Bt isolates.The protein purity values recorded by this study were highly variable probably due to the large diversity of isolates that may have differences in optimum growth conditions and a variety of insecticidal crystal protein produced (Aronson et al., 1995) which suggests that the culturing and extraction method may need to be optimized for each isolate in order to obtain equally high delta-endotoxin yields of high purity.The procedure used by this study has been reported to be well suited for large scale production of endotoxin extracts for pesticidal application (Osir and Vundla, 1999).The mean protein yield from the pellets varied from 20 to 30% similar to what is reported by Lereclus et al. (1993) although using different protocol.Higher protein values may have resulted from using lower centrifugation speeds and prior decantation of froth containing cell debris and spores effectively resulting in a lighter pellet.
Two commonly used Bt strains in commercial formulations for control of lepidopterans are Bt kurstaki and Bt aizawai, with the letter strains showing better larval control in situations where Bt kurstaki is becoming less effective due to resistance development of the pests like the diamond black moth (Schnepf et al., 1998;Polanczyk et al., 2000).The findings of this study closely resemble those of Wang'ondu (2001) with the more toxic isolates (Bt 44 and Bt 48) being obtained from the lowest LT 50 values.Significant correlations between endotoxin yield and toxicity among the different isolates illustrates that different Bt isolates produce different deltaendotoxins which may differ in toxicity against different target pests ( Uribe et al., 2003).

RECOMMENDATIONS
These results form a basis for further investigation of the local Bt isolates showing efficacy against C. partellus such as determination of the Cry proteins therein and how temperature and pH would affect their toxicity.It is also recommended that the toxicity of these isolates be investigated against other local lepidopteran pests in order to determine their target range.

Figure 2 .
Figure 2. The protein bands (arrows) of delta-endotoxin and spore mixture of Bt isolates during its fermentation from Legume medium (1SKAG37°C and 24LBN30°C) and NYSM medium (58SLA25°C and Bt 20) as determined by SDS-PAGE.

Figure 3 .
Figure 3.The protein bands (arrows) of delta-endotoxin and spore mixture of Bt isolates during its fermentation

Table 1 .
Masses of pellets of Bt isolates and their protein quantities.

Table 2 .
Percent mortality of neonate C. partellus larvae on treatment with five toxin concentrations of, the standard, Bacillus thuringiensis subspecies kurstaki (Btk).

Table 3 .
Percent cumulative mortality of C. partellus first-instar larvae exposed to 0.015 mg/ ml endotoxins from Bt isolates.