Abstract: Bacillus thuringiensis is a bacterium of huge agronomic and scientific interest. The subspecies of this bacterium colonize and kill a large variety of host insects and nematodes with a high degree of specificity. In the present investigation, 32 native isolates of Bacillus thuringiensis recovered from different regions in India and 8 known Bacillus thuringiensis strains were analyzed at the molecular level using random amplified polymorphic DNA (RAPD) markers. Seven RAPD markers were used for diversity analysis. The RAPD banding pattern data was subjected to dendrogram construction using Unweighted Pair Group Method with Arithmetic Mean (UPGMA) analysis using NTSYSpc2.2 software. Eight main clusters were formed at 25% similarity level and four isolates were standalone of these clusters. Most of the isolates were found to be diverse, even though they were isolated from the same source or location. The RAPD markers were found to be effective to distinguish B. thuringiensis native isolates recovered from different sources and locations. The results of present investigation help to understand diversity of B. thuringiensis in India, which would be exploited to find new types of B. thuringiensis endotoxins.
INTRODUCTION
B. thuringiensis is a gram-positive bacterium distinguishable from other closely related 3 Bacillus spp. viz., B. cerus, B. anthracis and B. mycoides, because of its ability to synthesize crystal (Cry) proteins during sporulation (Hofte and Whiteley, 1989). B. thuringiensis has been used to control insect pests due to its environmental safety and target specificity. When Cry proteins in B. thuringiensis are ingested by insect larvae, the protoxin is dissolved and activated under alkaline conditions present in the midgut of target insects and the active toxin is released. This active toxin then binds to specific receptor in the insect’s midgut epithelial cells resulting in pore formation in the membrane which eventually leads to disruption of osmotic balance, cell lysis and death (Kaur, 2000). The different Cry proteins are toxic to a variety of insects that cause serious damage to crop and lead to decrease in crop yield (Martin and Travers, 1989; Iriarte et al., 1998).
Genotyping of bacteria employing Polymerase Chain Reaction (PCR)-based randomly amplified polymorphic DNA (RAPD) method (Williams et al., 1990) is a commonly used approach for strain typing because of its simplicity and cost effectiveness (Gurtler and Mayall, 2001, Ronimus et al., 1997). It is useful, fast and informative in differentiating B. thuringiensis strains (Hansen et al., 1998; Gaviria and Priest, 2003).
Taxon-specific markers with different specificities can be generated by using RAPD fingerprinting technique (Cherif et al., 2003; La Duc et al., 2004). This technique has successfully been used for grouping of highly related genotypes of B. thuringiensis (Sadder et al., 2006). RAPD has been used to differentiate Bacillus anthracis from B. cereus, B. mycoides and B. thuringiensis by identifying unique bands from these species (Kim et al., 2007). The unique PCR product was cloned to develop species-specific marker for identification of B. anthracis. Therefore, RAPD is a suitable technique to discriminate the B. thuringiensis isolates for their diversity analysis.
In this study, an attempt was made to characterize the genetic diversity of 32 native isolates of B. thuringiensis isolated from different sources collected from various states in India and 8 type strains used as reference by using RAPD markers. We have previously carried out molecular typing of native B. thuringiensis isolates from diverse habitats in India by using repetitive extragenic palindromic sequence (REP)-PCR and enterobacterial repetitive intergenic consensus (ERIC) sequence (ERIC)-PCR analysis (Katara et al., 2012).
MATERIALS AND METHODS
Isolation and culture of B. thuringiensis strains: A total of 32 native isolates of B. thuringiensis collected from diverse habitats of India (Kaur and Singh 2000a, b) and a set of eight type strains obtained from the Bacillus Genetic Stock Centre (BGSC), Department of Biochemistry, Ohio State University and Columbus, Ohio 43210, USA, were used as reference in this study (Table 1). All the strains were grown at 28°C for 14 h on Luria Bertani Agar (LA) containing penicillin (10 μg μL-1) as the selective antibiotic. B. thuringiensis isolates had characteristic resistance against penicillin and this was used for initial screening. Single colony from LA plate was selected to grow in the Luria Bertani broth (LB) to harvest cells for chromosomal DNA isolation in each case.
DNA extraction: Bacteria were shaken in 10 mL of LB at 28°C for 14 h until cultures reached late exponential phase. DNA was isolated from the cell suspensions (4.0 mL) by using OmniPrep™ kit for Gram Positive Bacteria, in accordance with the manufacturer’s instructions (G-Biosciences, St Louis, MO, USA). The concentration of DNA was determined spectrophotometrically at 260 nm and finally DNA samples were diluted to 10 ng μL-1 concentration.
RAPD-PCR amplification: In this study, RAPD markers were used for DNA fingerprinting analysis of B. thuringiensis isolates. In preliminary experiments, 12 RAPD primers were used for amplification of two native B. thuringiensis isolates. Out of these, 7 primers were selected on the basis of reproducibility and used for genotyping of all the B. thuringiensis isolates. Thermo-cycling conditions for the RAPD markers system were: DNA denaturation at 94°C for 5 min followed by 35 cycles of denaturation at 94°C for 1 min, primer annealing at 36°C for 1 min and extension at 72°C for 2 min. It was followed by a final extension step at 72°C for 10 min and then samples were stored at 4°C till used for electrophoresis.
Electrophoretic analysis: RAPD-PCR patterns were visualized by agarose gel electrophoresis. Aliquots of 12.5 μL each of the amplification products were loaded on to 0.8% agarose gel and run in 0.5X TBE buffer at 2 V cm-1 for 5 h. Thereafter, gel slabs were stained with 0.4 μg mL-1 of ethidium bromide and documented with Alpha Innotech Gel Doc System. Molecular weight analysis of bands was performed using FluorChem 5500 software with 1Kb DNA ladder (MBI Fermentas) as the molecular weight marker. All reactions were repeated at least twice and only bands that were bright and reproducible, were scored for analysis.
Statistical analysis of marker characteristics: All seven primers were used twice for PCR amplification to detect reproducible bands. Only reproducible bands were considered for analysis. Each band was considered as a marker and scored ‘1’ for presence of band and ‘0’ for absence. A binary data matrix of RAPD markers for all the isolates were recorded on Microsoft office Excel spread sheet for further analysis.
Diversity analysis: Cluster analysis was used to examine genotypic relationships among the tested strains. Amplification products were scored as either present (1) or absent (0). A data matrix was prepared to determine similarities between each pair of genotypes. This genotypic matrix was used for diversity analysis by NTsys-2.1 software tool. In NTsys-pc, the matrix was subjected to unweighted pairwise group method for arithmetic average analysis (UPGMA) to generate a dendrogram using the SAHN subroutine and Tree plot of NTSYS-pc. Cophenetic correlation was calculated as a measure of the faithfulness of cluster analysis and data subjected to Principle Coordinates Analysis (PCA).
Discriminatory power of RAPD markers: The discriminatory power of RAPD markers was evaluated by PIC (Polymorphic information content), RP (resolving power) and MI (marker index). PIC for each marker was calculated as proposed by Powell et al. (1996) as PIC = 1-ΣPi2 where Pi is the frequency of ith allele. As each band line has been considered as a marker, PIC was separately calculated for each band line and averaged to estimate PIC for the particular RAPD primer. Resolution power of the primer was calculated by Rp = ΣIb where Ib (band informativeness) was calculated by =1-[2x |0.5-p|], p is the proportion of genotype having a band. MI was calculated according to Powell et al. (1996). MI is the product between diversity index (equivalent to PIC) and Effective Multiplex Ratio (EMR) where EMR is the product of fraction of polymorphic loci and number of polymorphic loci.
RESULTS AND DISCUSSION
In present investigation, very wide diversity was observed among the B. thuringiensis isolates which were isolated from different regions in India (Table 1). All the primers used for RAPD analysis showed amplification and generated RAPD fingerprint for B. thuringiensis isolates (Fig. 1a, b). A total of 108 bands were amplified from all these seven primers.
| Fig. 1: | Agarose gel showing RAPD profiles generated using primers, (a) RAPD11 and (b) RAPD14 for 32 native B. thuringiensis isolates from diverse Indian habitats and 8 known strains. All the native isolates numbers have SK prefix, M: 1kb molecular weight ladder |
| Table 1: | Details of native B. thuringiensis isolates from different sources from diverse locations of India and type strains used in this study |
The discriminatory power of all 7 RAPD markers was evaluated by PIC (Polymorphic information content), RP (resolving power) and MI (marker index). All 7 primers were found polymorphic for all the amplified bands (Table 2). Out of 7 markers, RAPD12 was giving maximum number (22) of polymorphic bands having 0.18, 5.21 and 4.01 of PIC, RP and MI, respectively. An average of ±0.13 Polymorphic Information Content (PIC), 4.08 resolution power and 3.16 marker index was obtained for individual band in 40 isolates.
Eight main clusters were observed in dendrogram produced by cluster analysis (UPGMA) based on Jaccard coefficient by using molecular fingerprint of 32 native isolates recovered from diverse locations of India and 8 type strains used as reference at 25% similarity level (Fig. 2). One type strain B. thuringiensis serovar finitimus (4B1) and three native isolates SK-304, SK-935 and SK-722 were standalone of these clusters Cluster-I has three native isolates and one reference strain. In this cluster, all 3 native isolates are from the sources related to chickpea, namely 2 from the chickpea field soil in Rohtak, Haryana and 1 from chickpea leaves from IARI field in New Delhi (Table 3).
| Table 2: | List of RAPD primers used for fingerprinting of native Bt isolates |
| *Average PIC of individual band in 40 isolates | |
All these isolates were collected from different regions. It is interesting to get such similarity among isolates from very wide locations. In cluster-II, two reference strains B. thuringiensis serovar aizawai/pacificus (4J3), serovar finitimus (4B1) and a native isolate SK-1060 (obtained from barren land in Dungarpur, Rajasthan) were grouped together. Cluster III has 2 native B. thuringiensis isolates SK-4 and SK-758 which were collected from Chickpea field in Rohtak, Haryana and Jowar grain dust in Guntur, Andhra Pradesh respectively.
| Table 3: | Grouping of B. thuringiensis isolates and type strains as per dendrogram of RAPD profiles at 25% similarity |
| Fig. 2: | Dendrogram produced by cluster analysis (UPGMA) based on Jaccard coefficient by using molecular fingerprint of 40 isolates including 32 native isolated from diverse locations of India and 8 type strains used as reference obtained from BGSC |
Cluster IV has highest number of isolates containing 4 reference strains and 15 native isolates, representing different regions namely, New Delhi, Uttar Pradesh, West Bengal and Rajasthan. This cluster was further divided into 3 subgroups IVa, IVb and IVc. While IVa and IVb had 5 isolates each from different regions, the subgroup IVc had 4 isolates from different regions and 5 type strains. Cluster V had two isolates SK-229 and SK-729, collected from phyllosphere in a pigeonpea field in IARI, New Delhi and Cotton seeds, Guntur, Andhra Pradesh respectively. Cluster VI has two native isolates SK-5 (recovered from Cotton field in Hissar, Haryana) and SK-1055 (recovered from Maize field in Udaipur, Rajasthan). Cluster VII has three native isolates SK-51, SK-63 (collected from Orchard soil in Shimla, Himachal Pradesh) and SK-700 (cotton field soil in Guntur, Andhra Pradesh). Cluster VIII having two native isolates SK-94 and SK-680 obtained from Grain dust in Shimla, Himachal Pradesh and a soil sample from Ganges Sangam river bank, Uttar Pradesh, respectively.
It was observed that B. thuringiensis isolates are not clustered according to their geographical locations and sources. Not a single cluster occurred according to geographic location or source except cluster I where all 3 native isolate were from chickpea soil and leaf samples. However, native isolate SK-4, also collected from chickpea field, Rohtak, Haryana, fell in different cluster (cluster III). Some isolates collected from same source, such as SK-51 and SK-63, collected from orchard soil, Shimla, Himachal Pradesh showed only 26% similarity.
The eight clusters were subdivided into subgroups at 50% similarity level and analyzed to find any basis behind grouping. However no correlation was observed among the isolates in any of the subgroup. A 3D plot of principal coordinate analysis showing the diversity of 32 native isolates and 8 type strains is depicted in Fig. 3. In the present study, native strains of B. thuringiensis were isolated from different places and sources. Thirty two B. thuringiensis native isolates and 8 type strains were characterized by RAPD markers which differentiate the isolates recovered from different places.
| Fig. 3: | 3D plot of principal coordinate analysis showing the diversity of 40 isolates including 32 native isolated from diverse locations of India and 8 type strains obtained from BGSC |
The RAPD analysis is considered to be a fast and simple method. Once the primers revealing the polymorphism were identified and PCR conditions optimized, slight differences in primer sequences caused significantly different RAPD patterns and enabled the easy discrimination among the strains (Sikora et al., 1997).
The technique of RAPD has been used for characterization of B. thuringiensis isolates by some other workers (Sadder et al., 2006; Zara et al., 2006; Konecka et al., 2007; Manzano et al. 2009). Sadder et al. (2006) carried out RAPD analysis on 7 B. thuringiensis isolates recovered from Jordan and 5 reference B. thuringiensis strains and found high degree of polymorphism. A clearly defined habitats locational pattern was also not observed by Malkawi et al. (1999) in their RAPD-PCR analysis of 16 B. thuringiensis isolates recovered from different Jordanian habitats. Pattanayak et al. (2001) also performed RAPD based fingerprinting of 21 serovars of B. thuringiensis using 19 random decamer primers and found very low similarity values range from 3-68% indicating high genetic divergence. Gaviria and Priest (2003) carried out RAPD analysis of 126 B. thuringiensis representing 57 serovars, which were allowed to 58 genomic type based RAPD patterns. It was concluded that while the species is genomically diverse, the presence of one or more homologous serovar represents clonal lineages of successful pathogens. Vilas-Baas and Lemos (2004) also observed remarkable genetic diversity among 218 B. thuirngiensis isolates from Brazil suggesting that the diversity resulted from influence of different ecological factors and spatial separation between strains generated by conquest of different habitats. A high degree of genetic divergence with only 16 per cent similarity among B. thuringiensis serovars was observed in RAPD analysis and was recommended as a complement to flagellar serology for genetic characterization of B. thuringiensis (Chaves et al., 2010). Kumar et al. (2010) have also carried out RAPD-PCR of 70 B. thuringiensis isolates collected from soil samples of water field and found that RAPD provided a high degree of discrimination.
RAPD analysis is considered an important molecular biology technique for the identification of indigenous B. thuringiensis isolates. Our study showed that the regional B. thuringiensis isolates had a high degree of genetic diversity. Most of the isolates were found to be diverse even though they were isolated from the same source or location. The results indicate that the population of native Bt isolates from different geographical locations and sources is very diverse.
Thus, RAPD analysis could distinguish the native isolates of B. thuringiensis recovered from different regions and sources. It has contributed to our efforts to understand the diversity of the B. thuringiensis isolates in various climatic zones of India and will help in selection of isolates for screening of new sources of cry gene alleles.
CONCLUSION
Wide diversity was observed among 32 native B. thuringiensis isolates recovered from different regions in India and 8 known B. thuringiensis strains used as reference by RAPD analysis. Out of 7 RAPD markers employed in this study, marker RAPD12 was found to yield maximum number of polymorphic bands. The isolates and reference strains fell into 8 main clusters at 25% similarity level in UPGMA clustering. B. thuringiensis isolates were not clustered according to their geographical locations and sources. B. thuringiensis isolates had a high degree of genetic diversity as most of the isolates were found to be diverse even though recovered from same habitat. Thus, RAPD analysis has been useful to distinguish native isolates of B. thuringiensis isolated from different regions and sources.
ACKNOWLEDGMENT
JK was recipient of fellowship from the Council of Scientific and Industrial Research (CSIR), India. Authors are grateful to Rakesh K. Narula for excellent technical assistance.