ISOLATION, TOXICITY AND MOLECULAR CHARACTERIZA- TION OF NATIVE Bacillus thuringiensis ISOLATES FROM EGYPTIAN SOIL

Thirty three local soil samples were collected from Westward till Sewa Oasis in Egypt to search novel isolates of Bacillus thuringiensis (Bt) and evaluate their toxic potentiality to overcome the serious problem of evolved resistance by insects to the pesticidal activity. The first instars larvae of cotton leaf worm (Spodoptera littoralis) were used to test their toxic potentiality in the presence of the two standard strains kurstaki (K) and neoleonensis H24a (N). The results showed that three isolates from Alexandria (AL3, AL7 and AL11) and two isolates from Kafr - el-Dawar (KD2 and KD3) were presumptively confirmed as Bt. by morphological and microscopic characters. The treated larvae with Insecticidal crystal protein (ICP) exhibited mortality percentage around 50% for all isolates except AL3 isolate. The vegetative insecticidal proteins (VIPs) of KD3, KD2 and AL11 isolates revealed mortality percentages more than 75%. While, AL3 and AL7 isolates showed differential toxicity. Crystal proteins analysis by SDS-PAGE showed KD2, KD3 and AL11 isolates gave similar profiles as those of "K" strain which was characterized with 135 kDa and 70 kDa bands. AL3 isolate failed to show any noticeable bands. According to their plasmids patterns, 4 kb was shown in all isolates and the reference strains except AL3 isolate. Cry1, Vip1 and Vip2 genes of the isolates was detected by polymerase chain reaction (PCR). The results indicated that one band with 550bp in size was present in all isolates except the AL3 isolate. Vip1 primer succeeds to amplify a band 400 bp in size in all isolates and “K" strain. Vip2 primer failed to react with any genome of the studied isolates or reference strains. This study suggested that KD2, KD3 and AL11 isolates may lead to be identified as potential strains of Bt. for their use in the development of bioinsecticide to control insect pests in Egypt.


ISOLATION, TOXICITY AND MOLECULAR CHARACTERIZA-TION OF NATIVE Bacillus thuringiensis ISOLATES FROM EGYPTIAN SOIL AZZA MAKHLOUF, AMIRA A. EL KEMANY, A. S. RIAD, Y. A. ABOU-YOUSSEF
Genetics Department, Faculty of Agriculture, Alexandria University, Egypt acillus thuringiensis (BT) is known to have pathogenic effects against several insects, worms and their effects on humans and beneficial insects are negligible (Sanchis and Bourguet, 2008). The toxicity of Bt is due to its ability to produce crystalline inclusions (Cry toxins) during sporulation, These inclusions are composed of proteins (Cry proteins) or δendotoxins which are highly toxic to a wide variety of insect pests and some invertebrates (Chattopadhyay et al., 2004;Vilas-Bôas et al., 2007). Recently, Cry toxins were classified based on their primary amino acid sequences, and more than 500different cry gene sequences have been classified into 67 groups (Bravo et al., 2011). During vegetative growth of BT, a number of pesticidal proteins unrelated to Cry proteins are produced by some strains (Mesrati et al., 2005). These toxins are named VIPs that do not form parasporal crystal proteins and are apparently secreted from the cell into culture medium. These proteins have shown toxicity against certain important lepidopteran pests like black cut worm (Agrotis ipsilon) and cotton leaf worm (Spodoptera littolaris), which could not be curbed earlier by the crystal toxins. Moreover, some of the lepidopteran insects that are susceptible to the Cry toxin at microgram level could be killed completely by the VIPs at nano gram level (Doss et al., 2002). Due to their high specificity and safety of BT to most non-target organisms and to the environment. The crystal proteins are preferred and widely used as an alternative to chemical pesticides in pest management strategies against insect pests of agricultural crops. At the same time, the wide spread use of Bt (Crystal Protein) in pest control has led to the potential for development of resistance by target organisms to Bt toxins (Liu et al., 2001;Sayyed et al., 2004). So, this has necessitated new isolation and novel strain that could to overcome the serious problem of evolved resistance by insects to the pesticidal activity of BT protein contributing to insect pests control and limit the use of toxic chemicals hazardous to the environment (Kamel et al., 2010). Therefore, the present study was aimed to search in the Egyptian soil for novel active isolates of BT that may be more efficient in using their toxin against insect pests of agricultural crops as cotton leaf worm and investigate their Cry and Vip genes using PCR techniques to determine the distribution of Cry and Vip genes within a collection of the studied isolates. In addition, B SDS-PAGE crystal protein and plasmid profiles were also studied in these isolates.

Sample collection
Thirty three Soil samples were collected from nine locations in Egypt around Alexandria city and extending from Westward till Sewa Oasis where the samples were collected (Fig. 1) during 2006 season, these locations differed in their vegetation and cropping patterns as shown in Table (1). They were taken by scraping off soil surface with sterile spatula and 10 cm below the surface. 1kg/sample was placed in sterile plastic bags, covered with sterile cheesecloth and stored at 4C. Two strains of Bt kurstaki (K) and neoleonensis H24a (N) were obtained from the Department of Entomology, Faculty of Agriculture, Alexandria University and Department of Biological control, Agriculture research center, respectively and used as standard (negative control) since their toxicity against this insect is well known.

Cotton leaf worm (Spodoptera littoralis)
Cotton leaf worm (S littoralis) was provided by the Department of Entomology, Faculty of Agriculture, Alexandria University. The artificial diet was prepared according to Shorey and Hale (1965). It was used to evaluate toxic potential of the studied isolates.

Isolation of bacterial strains
One gram soil sample was suspended in 20 mL of nutrient broth in a test tube (50 mL). Then heated to 80C for 10 min to remove non-spore forming bacteria found in the soil. The samples were diluted 100-fold, and 200 μL of diluted sample was spread on a nutrient agar plate. After that, left to dry and incubated at least 3 days at 30C.

Viability of bacterial spores and microscopic examination
The viability of bacterial spores was determined as describe by Zweig (1963). 1mLof bacterial culture was mixed with about 9 mL of sterile water. Two serial dilutions (10 -3 and 10 -5 ) were prepared from this suspension as describe by Baker (1962) and Collin (1964). Dishes contain numbers of colonies ranged from 30-300 were studied. Ten single colonies with morphology similar to Bt were randomly selected from each plate after 72 h, stained with crystal violet and examined for the presence of endospores and parasporal inclusion bodies under phase contrast and light microscopes.

Insect toxicity assays
The toxicity of the studied isolates was screened on 1 st instars neonate larvae of S littoralis. Three dilutions (1, 0.75 and 0.5 mL.) and fifteen 1 st instars larvae of S littoralis were used for each dilution for studying the toxicity of protein crystals and vegetative insecticidal proteins from the isolates. The LC 50 values were determined for Cry and Vip extracts using probit analysis without correction, since the non treated controls gave mortality percentages less than 20% (Abbott, 1925).

Cry bioassay
Cry bioassay was preformed according to Kalfon and De Barjac (1985) with some modifications. Bacterial cell suspension were incubated more than 48 h for each bacterial isolate, 100 mL of culture was centrifuged under cooling (4C) at 3000 rpm for 10 min. Bacterial pellets were resuspended in 5 mL of extraction buffer (Phosphate buffer 0.01 M, EDTA 5 mM, NaCl 0.1% and SDS 2%) for 2h and centrifuged under cooling (4C) at 6000 rpm for 10 min. Pellets were taken and stirred in 50 mM NaOH for 1h. The insoluble particles were pelleted at 6000 rpm for 10 min and supernatant was stored at -20C until usage. Mortality was recorded after 48; 72 and 120 h.

3. 2 Vip bioassay
Vip bioassay was performed as describe by Estruch et al. (1996) with some modifications. Fresh bacterial culture was grown in PY broth medium for 12 h at 30C for harvest the vegetative insecticidal proteins. After centrifugation at12,000 xg for 10 min at 4C, the cell pellet was discarded and the supernatant containing the proteins of interest was retained. Proteins present in the supernatant were precipitated with ammonium sulphate (70% saturation) and collected by centrifugation at 12,000 x g for 10 min at 4C. The pellet was re-suspended in minimum volume of 20 mM Tris HCl buffer (pH 7.4) and dialyzed overnight at 4C.
The crude extract obtained was subjected to ultra centrifugation at 56,000 x g for 30 min at 4C to remove traces of suspended matter. The effect of vegetative insecticidal proteins (bacterial extracts not more than 18 h of culturing) was performed by recording the mortality.

Cry protein analysis using SDS-PAGE
The protein profiles of the studied isolates were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis

Plasmid isolation
DNA plasmid was isolated according to the method described by Kado and Liu (1981) and separated on 1% agarose. Electrophoresis was carried out at 75 volt for 2 h. The gel was visualized by UV light and photographed.

Statistical analysis
Probability test (Fisher, 1935) was used to estimate the significant differences between the isolates and the reference strains. The probabilities were calculated according to the 2x2 contingency tables using a simple basic computer program suggested by Forbes (1984).

Characterizations of B. thuringiensis
Figure (2) showed phase contrast and light microscope photomicrographs of the fixed preparations for five isolates which stained with crystal violet. The results revealed that five isolates as AL3, AL7, AL11, KD2 and KD3 were only identified as Bt based on the ability of bacteria to form endospores and parasporal crystals, irrespective of their crystal shapes. These isolates were only distributed in the soil of Alexandria and Kafr -el-Dawar where, AL3 and AL7 isolates showed typical Bacilli characteristics with endospore inclusions under light microscope and gave a scarce number of parasporal crystals under phase contrast microscopy as shown in Figs (2A and 2B). The other three isolates AL11, KD2 and KD3 produced noticeable number of parasporal crystals as the other typical Bt characteristics (Figs 2C,2D and 2E). No significant differences were detected in the average numbers of endospores and protein crystals between the isolates and the reference strains except AL3 isolate which illustrated significant differences in the average numbers of protein crystals as shown in Table (2). These results indicated that AL7, AL11, KD2 and KD3 isolates were presumptively confirmed as Bt and can be studied as a local strain of this bacterium. Previous studies suggested that Bt can be identified by conventional methods like morphological and microscopic characters (Keshavarzi, 2008). In addition, these isolates obtained from the soil characterized with dense cropping as Alexandria and Kafr -el-Dawar. In contrast, no possible B. thuringiensis was identified from the other locations with specific limited vegetation and mostly characterized as calcareous soils as Urabi; El-Agami; Janaklies; Burg -el -Arab; Mareena; Mersa Matrouh; and Sewa. The present results are in accordance with Thaphan et al. (2008) and Raymond et al. (2010) who reported that the properties of the soil could affect the abundance and distribution of Bt. The presence of Bt population in the soil samples was affected negatively by soil sand percentage and this may because it contains a smaller amount of nutrient and water. Vilas-Bôas and Lemos (2004) stated that the recovery of the bacterium is more frequently from agricultural soils and to some degree from urban soils, however in beach soil and desert sands the incidence of this bacterium may be very low or absent and this is probably explained by the variation of the biotic and abiotic factors in these different soil habitats and their possible effects on the distribution of B. thuringiensis.

Bioassay of insecticidal potentialities
According to the LC 50 values and mortality percentage of cotton leaf worms after the treatment with bacterial Cry protein extracts, it can be noticed that the two reference strains are highly toxic to cotton leaf-worm since they showed the lowest LC 50 values of 0.192 and 0.242, respectively with their mortality percentages more than 75%. The most effective isolate against cotton leaf worms was AL11, KD3, KD2 and AL7 isolates in order which recorded LC 50 values of 0.313, 0.5710, 0.708 and 0.908, respectively with mortality percentage around 50%. So, these isolates can be considered as moderately toxic. On the other hand, the isolate AL3 illustrated very high LC 50 value (1.342) and turned to be non toxic because it showed mortality percentage less than 25% comparing with the two N and K reference strains as presented in Table (3) and illustrated by Fig. (3).  Fig. (4) showed bacterial cultures mainly in the log phase of growth that were used as the source of vegetative proteins (VIPs) to test their toxicity against cotton leaf worm. Among the five isolates used, only three isolates KD3, KD2 and AL11 showed the lowest LC 50 values of 0.111, 0.222, 0.244, respectively, with mortality percentages more than 75%. These isolates can be considered effective against Cotton leaf worms (Spodoptera litoraltis). The reference strain "K" as well as isolates AL3 and AL7 showed differential toxicity and they can be considered as moderately tox-ic because their LC 50 values ranging from 0.49 to 0.98. However, "N" strain recorded the highest LC 50 (1.1769) value with mortality percent less than 25% accordingly this isolate can be considered not toxic. The present results indicated that the two isolates of Kafr -el-Dawar (KD2 and KD3) and a certain extent Alexandria isolate (AL11) are promising as local strains of this bacterium since, they manifested high toxicity of vegetative insecticidal proteins (VIP) and displayed moderately toxic against cotton leaf-worm due to their insecticidal crystal proteins (Cry proteins). This may point to the potential of these isolates as bioinsecticide against cotton leaf worm (S. litoraltis). The quantitative and qualitative difference in production and composition of protoxin of B. thuringiensis isolates is postulated by various investigators, Chenot and Raffa (1995) and Kashyap and Amla (2007) reported that the larval stage and the LC 50 values reflect that more than one protoxin is present and they are efficiently processed within the gut to provide sufficient active toxin molecules in larvae.

Analysis Insecticidal proteins of bacteria (Cry protein) using SDS-PAGE
Surface protein profile analysis is used in the determination of similarity between bacterial strains (Costa, 1992). Comparison of the protein patterns had been reported to be useful in evaluation of the relationship between Bt isolates (Swiecicka and De Vos, 2003). In the present study, two reference strains were used. The first was Bt var. kurstaki (K) that is known to have at least Cry I (135 kDa) and Cry II (70 kDa) polypeptides. The other reference, i.e. Bt var. neoleoninsis (N) that is known to have the same cry genes in addition to a cyt genes coding for hemolytic and cytolytic toxins but with very limited insecticidal activities (Kashya and Amla, 2007). Figure (5) shows photograph and diagrammatic drawing representing the SDS-PAG patterns of the present five isolates and the two reference strains. The results show strain "K" revealed 8 bands in total. Among these bands, 4 main bands with molecular mass of 135, 70, 28 and 10 kDa were observed. The 135 kDa band was assumed to be Cry 1 protein and 70 kDa can be resembled Cry 2 protein .The reference strain "N" illustrated similar total number of bands but it was characterized with a band with a molecular mass of about 80 kDa that is most probably corresponding to the Cyt polypeptide. Isolates KD2, KD3 and AL11 showed similar profiles as those of the "K" strain. Isolate AL7 showed less numbers of bands with lower molecular mass of polypeptides. In contrast, isolate AL3 did not display any noticeable bands when compared with the references or the other isolates. Figure (6) illustrates differences in band intensities of the four main polypeptides (135, 70, 28 and 10 kDa) relative to the reference strain K. The highest intensities was the two known Lepedopetrian toxic polypeptides (135 and 70 kDa) that were recorded for KD2 and KD3. Concerning the reference strain N as well as the isolates AL11 and AL7 were exhibited lower intensities than the others. While, the isolate AL3 did not show any noticeable intensity in polypeptides because it did not produce any band. The present results suggested that the two KD2 and KD3 isolates are promising novel Bt strains whereas, they are capable to produce of CryI and Cry2 polypeptides, in addition to other toxic and non toxic crystal proteins. Furthermore, these two isolates i.e. KD2 and KD3 are characterized with their toxic vegetative insecticidal proteins (Vip) production when compared with the other isolates used in the present study. The present results are in accordance with the previous results of Kashyap and Amla (2007) and Fakruddin et al. (2012) who obtained 47 isolates from 53 soil samples from diverse agro forest fields in Bangladesh. These isolates generate proteins by 12% SDS-polyacrylamide gel electrophoresis (PAGE) with molecular mass of 10-150 kDa that were almost similar to those of the reference strains of Bt subsp. kurstaki HD-73, Bt subsp. sotto and Bt subsp. japonensis) which confirmed the identity of these isolates as Bt.

DNA Plasmid profile of the isolates
The two reference strains and the five local isolates except AL3 isolate (Lane 7) showed a molecular weight band corresponding to about 4 kb. Also, the intensity of the separated bands decreased in order of K, N, KD2, KD3 and AL11 respectively as shown in Fig. (7). Based on the present results, it can conclude that the intensity of the separated plasmids decreased in order this may could reflect decreasing in plasmid copy numbers; however, this cannot be confirmed due to inaccuracy in loaded samples. Also, the present results displayed that AL3 isolate doesn't have any plasmid and this result confirmed the previous findings that the isolate AL3 wasn't able to reveal any noticeable Cry protein band and its Cry extraction was non toxic against cotton-leaf worm thus, its inefficiency could be attributed to the plasmids missing in it. Gonzazel et al. (1984) reported that isolates of the same Bt subsp are known to contain Cry protein on their plasmids of low copy number and this causes variation in the plasmid profiles. Attallah et al. (2014) found that plasmids with molecular weight ranging between 27344 bp to 3958 bp were observed in sixteen Bacillus local strains isolated from different Egyptian soils and the identified strains; (Bt. subsp. Kurstaki (Bt. 40), B. subtilis subsp. subtilis strain ATCC 168 (Bs) and B. licheniforms strain ATCC (14580) were shown at least one plasmid, which some of them have a maximum of seven plasmids. In general, all plasmid patterns are unique to each strain.

Detection of Cry gene
A band with 550 bp approximately was successfully amplified for Cry I gene with the presence of a gral Cry I general primer in all samples except the isolate AL3 failed to produce any band as shown in Fig. (8). The result was in agreement with Bravo et al. (1998) who used the same primer for obtaining an amplified fragment for Cry I gene that ranged from 543 to 594 bp in length. Performing PCR analysis with specific Cry gene primers could specify this amplicon (Ceron et al., 1995). The absence of Cry I gene in AL 3 genome (Fig. 7, lane 7) confirmed and explained the absence of any Cry protein in this isolate and no insecticidal activity was observed for this isolate against cotton leaf-worm. Similar results were obtained by Zhang et al. (2000) who, identified isolates of Bt. with no Cry gene and no insecticidal activity. In contrast, the reference strain N showed 4 other bands 2200, 1100, 900 and 800 bp in lengths. These bands may reflect the nonspecificity of gral cry1 primer, which can detect 25 of the 27 different cry1 genes (Bravo et al., 1998). In addition, many authors identified more than one cry Itype gene in the genomes of Bt. (Lee et al., 2001;Martinez et al., 2005). Thus, the presence of more than an amplicon in the genome of N reference strain may be due to the amplification of more than one cry I-type gene. Also, this feature may explain the high toxicity of the N strain Cry extract against the cotton leaf-worm. Zhang et al. (2000) stated that there is a correlation between the existences of more than one cryI-type gene and the high insecticidal effects

Detection of Vip genes
Many Bt. strains secrete non δendotoxins vegetative insecticidal proteins (Vip1, Vip2 and Vip3) during vegetative growth. VIPs represent a new type of insecticidal proteins because they are secret-ed as soluble proteins, rather than forming crystal inclusions inside the B. thuringiensis mother cell (Warren, 1997). In the present study, the genome of the five isolates and the reference strain (K) succeed to amplify only one band with 400 bp approximately in length in presence of specific Vip1 primer that represent vip3LB (Vip1) gene while, the reference strain (N) failed to produce any amplified fragment with the same primer as shown in Fig. (9). Mesrati et al. (2005) obtained a sharp band with 419bp in size coding for vip3LB (Vip1) gene. While, Vip2 primer failed to react with any genome of the isolates or reference strains under the present study. It can suggest that the genome of the reference strain N do not comprise the vip3LB (Vip1) gene as well as, vip3Aa (Vip2) gene do not exist in any genome of the local isolates or reference strains in this study. Osman (2010) stated that the existence reduction of Vip1/Vip2 genes in Bt. is perhaps due to the fact that these two genes have a wider distribution in B. cereus than in Bt., However, the insecticidal effects of Vip proteins for the five local isolates and the reference strain K are probably due to the presence and the expression of a vegetative insecticidal gene with high homology to the vip3LB gene specially, the high toxicity of VIPs extracts for KD2; KD3 and AL11 isolates that could be attributed to the high expression of this gene. In relation to the reference strain N, the absence of both vip genes (vip3LB and vip3Aa) and their protein in it mostly reflected the non-toxicity of its Vip extracts against cotton leafworm. It can conclude from our investiga-tion that, Egyptian agricultural soils rich with dense cropping are important source for providing a large genetic resource of B. thuringiensis strains. KD2, KD3 and AL11 isolates may confer commercial applications in agriculture and may lead the isolates to be identified as potential strains for their use in the development of bioinsecticide to control insect pests in Egypt.

SUMMARY
Thirty three local soil samples were collected from West-ward till Sewa Oasis in Egypt to search novel isolates of Bacillus thuringiensis (Bt) and evaluate their toxic potentiality to overcome the serious problem of evolved resistance by insects to the pesticidal activity. The first instars larvae of cotton leaf worm (Spodoptera littoralis) were used to test their toxic potentiality in the presence of the two standard strains kurstaki (K) and neoleonensis H24a (N). The results showed that three isolates from Alexandria (AL3, AL7 and AL11) and two isolates from Kafr -el-Dawar (KD2 and KD3) were presumptively confirmed as Bt. by morphological and microscopic characters. The treated larvae with Insecticidal crystal protein (ICP) exhibited mortality percentage around 50% for all isolates except AL3 isolate. The vegetative insecticidal proteins (VIPs) of KD3, KD2 and AL11 isolates revealed mortality percentages more than 75%. While, AL3 and AL7 isolates showed differential toxicity. Crystal proteins analysis by SDS-PAGE showed KD2, KD3 and AL11 isolates gave similar profiles as those of "K" strain which was characterized with 135 kDa and 70 kDa bands. AL3 isolate failed to show any noticeable bands. According to their plasmids patterns, 4 kb was shown in all isolates and the reference strains except AL3 isolate. Cry1, Vip1 and Vip2 genes of the isolates was detected by polymerase chain reaction (PCR). The results indicated that one band with 550bp in size was present in all isolates except the AL3 isolate. Vip1 primer succeeds to amplify a band 400 bp in size in all isolates and "K" strain. Vip2 primer failed to react with any genome of the studied isolates or reference strains. This study suggested that KD2, KD3 and AL11 isolates may lead to be identified as potential strains of Bt. for their use in the development of bioinsecticide to control insect pests in Egypt. genic plants. London, UK: Taylor and Francis Ltd. p. 109-121. Zhang, J., F. Song, Y. Zuo, L. Dai and D. Huang (2000). Identification of Cry-type genes of 31 Bacillus thuringiensis isolates and analysis of their expression product. Wei Sheng Wu Xue Bao., 40: 372-378. Zweig, G. (1963). Analytical methods for pesticidies, plant growth regulators and food additive. Academic press, New York and London, Vol. L. 626.