The Correlation of the Presence and Expression Levels of cry Genes with the Insecticidal Activities against Plutella xylostella for Bacillus thuringiensis Strains

The use of Bacillus thuringiensis (Bt) strains with high insecticidal activity is essential for the preparation of bioinsecticide. In this study, for 60 Bt strains isolated in Taiwan, their genotypes and the correlation of some cry genes as well as the expression levels of cry1 genes, with their insecticidal activities against Plutella xylostella, were investigated. Pulsed field gel electrophoresis (PFGE) and random amplified polymorphic DNA (RAPD) results revealed that the genotypes of these Bt strains are highly diversified. Also, a considerable number of the Bt strains isolated in Taiwan were found to have high insecticidal activities. Since strains that showed individual combined patterns of PFGE and RAPD exhibited distinct insecticidal activities against P. xylostella, thus, these genotypes may be useful for the identification of the new Bt strains and those which have been used in bioinsecticides. In addition, although the presence of cry2Aa1 may have a greater effect on the insecticidal activity of Bt strains in bioassay than other cry genes, only high expression level of cry1 genes plays a key role to determine the insecticidal activity of Bt strains. In conclusion, both RAPD and PFGE are effective in the differentiation of Bt strains. The presence of cry2Aa1 and, especially, the expression level of cry1 genes are useful for the prediction of the insecticidal activities of Bt strains against P. xylostella.


Bacterial Strains and Cell Cultivation
Bt strains (TT1-TT62) used in this study were obtained from Taiwan Agriculture Chemical and Toxic Substances Research Institute (TACTRI), Taichung, Taiwan. These strains were mainly isolated from soil and granaries in Taiwan. Two strains, i.e., TT12 and TT13, which were isolated from imported biopesticide products, were used as reference strains. Bacterial cells were cultivated in Brain Heart Infusion (BHI) broth (Difco™, Becton, Dickinson and Company, Sparks, MD, USA) overnight at 37 °C with rotary shaking (150 rev/min).

Bioassay of the Insecticidal Activity
The insecticidal activities of the Bt strains (TT1-TT62 except TT14) were assayed according to the methods described by Pang et al. [15]. Bt cells were cultivated, and the spores as well as crystal toxins were collected, mixed with Tris buffer (10 mM Tris-HCl, 10 mM EDTA, pH 7.4, Sigma-Aldrich, St. Louis, MO, USA), and assayed for protein concentration and insecticidal activity against P. xylostella. Two Bt strains, i.e., TT12 and TT13, isolated from two commercial insecticide products, DiPel™ and XenTari™ (Abbott, Chicago, IL, USA), were used as positive control. Tris buffer was used as negative control. Protein concentration was assayed using bicinchoninic acid (BCA) protein assay kit (Pierce, Rockford, IL, USA). For each Bt strain, two protein doses were repeated 3 times using 10 larvae per assay. The spores and crystal mixtures at protein concentrations of 25 mg/L and 250 mg/L, were sprayed onto a 15 cm 2 leaf piece of cabbage by a potter spray tower (Burkard Manufacturing Co. Ltd., Hertfordshire, UK) and the leaf were then fed to the third instar larvae at 25 °C under 60% humidity and incubated for 72 h. The insecticidal activities of the Bt strains were defined as mortality rates after incubation for 24, 48 and 72 h, which were calculated according to Abbott's formula, i.e., corrected mortality (%) = [(test mortality − blank control mortality)/ (1 − blank control mortality)] × 100% [16]. All data were expressed as mean ± standard deviation (n = 3).

Genotyping by PFGE and RAPD
The isolation of chromosomal DNA of Bt strains (TT1-TT62 except TT23 and TT41) and NotI digestion were according to the methods described by Kolstø et al. [17]. Generally, bacterial cells harvested by centrifugation were washed, resuspended, and then mixed with low melting point agarose (NuSieve™ GTG™ agarose, FMC BioProducts, Rockland, ME, USA) to obtain the agarose plugs. Following the cell lysis and proteolysis, the restriction digestion was performed by placing a 2 mm slice of each plug into 100 μL of restriction buffer containing 20 U of NotI (New England Biolabs, Beverly, MA, USA). After incubation at 37 °C for 12-16 h, the plugs were placed into the slots of a 1.2% agarose gel in 0.5× TBE buffer (89 mmol/L Tris-borate, pH 8.3, 2 mmol/L EDTA, Sigma-Aldrich, St. Louis, MO, USA). Electrophoresis was performed by using CHEF-DR ® II System (Bio-Rad, Hercules, CA, USA). The conditions used were 180 V for 28 h at 15 °C (pulse time: 7 to 90 s). Bacteriophage λ DNA concatemers (Bio-Rad, Hercules, CA, USA) were used as molecular weight markers.
PFGE and RAPD patterns were analyzed by the NTSYSpc software (Numerical taxonomy and multivariate analysis system, version 2.10e, State University of New York, Stony Brook, NY, USA). Strains were clustered by using the Dice coefficient of similarity, and cluster analysis by unweighted pair group method with arithmetic averages (UPGMA). The final judgment of whether the patterns were identical was done by visual comparison.

Detection of cry Genes
Chromosomal DNA of each of the Bt strains was prepared by phenol-chloroform extraction [18]. All isolates were screened by PCR analysis for the presence or absence of 13 selected cry genes, including those coding for Cry proteins with insecticidal activity against P. xylostella [14]. The sequences of the primers were shown in Table 1. PCR amplifications were performed as previously described [19][20][21][22][23][24][25].

Expression Levels of cry1 Genes
The expression levels of cry1 genes in Bt strains were determined by a two-step reverse transcription real-time PCR. For each Bt strain, total RNA was prepared from 10 8 CFU of Bt using PureLink™ RNA Mini Kit (Invitrogen Life Technologies, Carlsbad, CA, USA). Then, cDNA was synthesized in reactions containing 2 μL of total RNA from Bt isolates using a reverse transcription kit (SuperScript First-Strand Synthesis System, Invitrogen Life Technologies, Carlsbad, CA, USA). The cDNA was used as a template for real-time PCR to determine the expression level of cry1 genes in Bt strains. Real-time PCR was performed in 20 µL reaction mixtures containing 1× KAPA FAST SYBR green I master mix (KAPA Biosystems, Woburn, MA, USA), 2 μL of the cDNA, and 0.25 mM of each primer to amplify all cry1 genes. Real-time PCR was performed using the ABI 7500 system (Applied Biosystems, Foster, CA, USA). The amplification conditions in a thermocycler were according to the methods of Gaviria Rivera and Priest [26]. The expression levels of cry1 genes were shown by threshold cycle (Ct) values.

Statistical Analysis
Statistical tests were performed using SPSS 17.0 (SPSS Inc., Chicago, IL, USA), p < 0.05 was taken as statistically significant. A paired samples t-test was applied to compare the mortality rates at different time points. Since multiple cry genes may be involved in the insecticidal activity of a Bt strain, a multiple regression analysis was applied to determine the relationship between the presence or absence of cry genes and the insecticidal activities of Bt strains. A Mann-Whitney U test was used to determine if the Bt strains with high expression levels of cry1 genes showed higher insecticidal activities against P. xylostella than those with low expression levels of cry1 genes.

Insecticidal Activity of Bt Strains
To evaluate the insecticidal activity of Bt strains, and the effect of the dose of Cry proteins and the incubation time of larvae with these proteins, two concentrations of global Cry protein, i.e., 25 mg/L and 250 mg/L, and three incubation time periods, i.e., 24, 48, and 72 h, respectively, were used. Results were shown in Table 2. At the protein concentration of 250 mg/L, it was found that 35, 46, and 46 Bt strains showed mortality rates ≥90% after incubation for 24, 48, and 72 h, respectively. On the other hand, at the protein concentration of 25 mg/L, 11, 28, and 29 strains showed mortality rates ≥90% after incubation for 24, 48, and 72 h, respectively. At least, 19 strains showed higher insecticidal activities than those of the reference strains, i.e., strains TT12 and TT13, at the toxin concentration of 25 mg/L, after incubation for 24 h (Table 1). A paired samples t-test was applied to compare the mortality rates at different time points. At the protein concentration of 25 mg/L, mortality rates after 72, and 48 h incubation were significantly higher than those after 24 as well as 48 h, and 24 h incubation, respectively (p < 0.05, degree of freedom (df) = 182). At the protein concentration of 250 mg/L, mortality rates after 48 h incubation were significantly higher than those after 24 h incubation (p < 0.05, df = 182); mortality rates after 72 h incubation were higher than those after 48 h incubation but not statistically significant. The above results indicate that the mortality of larvae gradually increases with the length of incubation time, and an assay at a protein concentration of 250 mg/L may not effectively differentiate the insecticidal activities of these Bt strains. Part of the results assayed at the global Cry protein concentration of 25 mg/L and 250 mg/L, respectively, with incubation time of 24 h, was shown in Figure 1. Thus, the insecticidal activity against P. xylostella at the global Cry protein concentration of 25 mg/L was used for further analysis.

PFGE and RAPD Patterns
When chromosomal DNA of the 60 Bt strains (TT1-TT62 except TT23 and TT41) were cut with NotI and subjected to PFGE analysis, a total of 54 PFGE patterns were found. Only 4 patterns were shared by two or more strains. For RAPD using primer OPF-06, 41 patterns were found. Only 7 patterns were shared by two or more strains ( Table 2). Strains with the same RAPD pattern could be further discriminated by PFGE typing, and vice versa. For example, strains of TT25, TT29, TT42, and TT61, within RAPD pattern of D34, could be further differentiated by PFGE typing; strains in PFGE pattern PT50, i.e., strains TT44 and TT45, could be further discriminated by RAPD typing ( Table 2). For the combined use of PFGE and RAPD methods, only four strains, i.e., strains TT47-TT50, were found in a combined pattern, i.e., D43-PT52. Thus, both PFGE and RAPD results demonstrated that there is considerable genetic diversity among the Bt strains isolated in Taiwan.
Strains with individual combined patterns showed distinct insecticidal activities; strains that showed the same combined patterns of PFGE and RAPD exhibited similar levels of insecticidal activity. For example, the mortality rates of four strains, i.e., TT47-TT50, within the combined pattern of D43-PT52, were all belonging to strains of high insecticidal activities. The genotypes of Bt strains, if determined by highly discriminatory methods, may allow us to identify the newly isolated strains with potential for pesticide use and those strains which have been used in bioinsecticides.

Correlation among Insecticidal Activity, cry Genes, and the Expression Levels of cry1 Genes
A collection of randomly selected 29 Bt strains, which exhibited various insecticidal activities, was screened by PCR for the presence or absence of 13 selected cry genes, including those associated with the insecticidal activity against P. xylostella and those reported by Gaviria Rivera and Priest [26] and Ben-Dov et al. [22]. These cry genes are widely distributed among strains. All the 29 strains were positive for cry1 genes; none of the 29 strains carried cry1E, cry3A, cry4A2, cry8, cry9A, cry9C and cry32Aa genes. Variation of the presence of cry1B, cry2Aa1, cry7, cry9Ea, and cry22 genes was found among strains (Table 3). Thus, these 5 cry genes of Bt strains were used to find the correlation with the insecticidal activity against P. xylostella.

Discussion
For decades, worldwide screening followed by isolation and characterization of new Bt strains have been undertaken to find out strains with high insecticidal activities [27]. Screening the environment for novel Bt strains with high insecticidal activity has become one of the strategies for insect resistance management [28]. On the other hand, considerable genetic diversity among Bt strains has been reported. The diversity of Bt strains also facilitates the isolation of new types of insecticidal genes [29].
In this regard, a number of techniques have been used to discriminate Bt isolates. PFGE is generally considered as an accurate and reproducible method for typing clinically relevant bacteria, and has been used for typing Bt strains [26,30]. Another typing method, RAPD, which is fast and simple, also has been applied for the discrimination of Bt strains [13]. Although RAPD has been successfully used for the differentiation of Bt isolates based on their location of origin, further investigation is still needed to understand their insecticidal spectrum [13]. For PFGE, it has been reported that the clonal structure of Bt isolates determined by PFGE may not correlate with their cry gene content [10], however, different conclusions have been made by Gaviria Rivera and Priest [26].
In this study, both PFGE and RAPD methods were effective and informative in differentiation of Bt strains collected in Taiwan. This is evidenced by the fact that these two methods are capable to discriminate most of the Bt strains. When these two methods were combined for the analysis of 60 Bt strains isolated in Taiwan, only four strains could not be discriminated. Thus, Bt strains in Taiwan are highly diversified in their genome patterns. Such highly genetic diversity may be due to high level of heterogeneity within the chromosomal gene organization for strains. Nevertheless, since strains with individual genotypes showed distinct insecticidal activity against P. xylostella, the genotyping data may allow us to identify the Bt strains, either newly isolated strains or strains which have been used for bioinsecticide production.
In general, early instar larvae were more susceptible to Cry proteins in many insects than later instar larvae. In this regard, the larval instar used in conducting the bioassays should be considered by not only the most susceptible stage, but also with respect to other factors, such as low mortality in negative controls [31]. Even though older larvae are less susceptible to Cry proteins, to assay the insecticidal activity of Bt strains, we used the third instar larvae since such larvae may allow us to select the Bt strains with high insecticidal activity. In addition, the assay conditions at 25 mg/L of global Cry protein with 24 h incubation time were used to find the strains with high insecticidal activity since under such conditions, the insecticidal activity of different Bt strains could be discriminated effectively ( Figure 1). Furthermore, since general stress of larvae, such as nutritively unbalanced food for larvae due to unsuitable host plants, may affect larval vulnerability to Bt treatment, and leads to high mortality rates in larval populations [32], the factors of general stress were also considered. P. xylostella is considered as one of the most destructive insect pests of cabbage, and effects of insecticides on P. xylostella have been evaluated with the use of cabbage [33,34]. In this study, the insecticidal activity of Bt strains was also evaluated using cabbage, one of the host plants of P. xylostella. Thus, the possibility that general stress of larvae, such as nutritively unbalanced food, affects the mortality, is little. Under such conditions, 19 of the 60 Bt strains isolated in Taiwan were found to have higher insecticidal activities than those of the reference strains.
The cry gene content of Bt strain may be useful for the prediction of its insecticidal potential, and PCR-based identifications of cry genes have been developed to help the screening process [4,8,27]. A previous study revealed that cry1 and cry2 were the most abundant genes in the Bt strains isolated in Taiwan [11]. In this study, all strains were found to be positive for cry1 genes. Also, most of the strains with high insecticidal activity were positive for cry2Aa1. Our results were in agreement with those described by Chen et al. [11]. Moreover, we found that cry2Aa1 gene was positively correlated with the insecticidal activity of Bt strains (Table 4). Regarding cry1 genes, although all Bt strains were positive for cry1 genes, their insecticidal activities varied. Ferrandis et al. [8] reported that the insecticidal toxicity could be related to gene content in most cases, however, one strain without cry1 genes, showed high toxicity against P. xylostella. A cry gene, detected by PCR, can be interrupted, mutated, or under control by a defective promoter; the corresponding Cry protein may not be present or present at reduced levels, therefore, contributing minimally to the toxicity [8]. The variety in the expression levels of individual cry genes also weakens the correlation between cry gene content and the toxicity of Bt strains [4].
Thus, the expression levels of cry1 genes in our Bt isolates collected in Taiwan were assayed. All Bt strains, which showed high expression levels of the cry1 genes, exhibited high levels of insecticidal activity. However, two strains, i.e., TT47 and TT49, belonging to the same genotypes, as strains TT48 and TT50, also exhibited high insecticidal activities despite of their low expression levels of the cry1 genes (Table 5). Thus, there might be some other cry-type genes active to P. xylostella existing in these Bt isolates. Furthermore, other factors, such as β-exotoxins, phospholipases, proteases, chitinases and the secreted VIPs (vegetative insecticidal proteins), are possibly involved in the complete pathogenic effect of a strain [4]. As for the insecticidal activity of Cyt toxins, a review from Frankenhuyzen [14] revealed that results of the activity of Cyt1Aa against P. xylostella are conflicting, while for Cyt2Aa, it was not tested against P. xylostella. The synergistic interactions of Cry toxins also contribute to the toxicity to a specific insect [35]. In this regard, Porcar and Juá rez-Pé rez [4] have suggested that for different Cry toxins, the expected dose needed to kill 50% of the insects need to be calculated if the relative proportions of these toxins and the individual toxicity of toxins were known. Nevertheless, although Bt strains used in this study were not checked for the presence of VIP and Cyt proteins, as well as the protein concentrations of the five Cry toxins shown in Table 4, based on the results of this study, it is possible that Bt strains with high expression levels of cry1 genes should be the strains with high insecticidal activity against P. xylostella.

Conclusions
In conclusion, for the isolation and characterization of novel Bt strains with high insecticidal activity, both the RAPD and PFGE methods are effective and informative in differentiation of Bt strains. Concerning the insecticidal activity against P. xylostella, a considerable number of the Bt strains isolated in Taiwan were found to have high insecticidal activity, as compared to those of the reference strains isolated from imported bioinsecticides. Also, the presence of cry2Aa1 gene determined by PCR may be used as a reference marker to predict the insecticidal activity, and only high expression level of the cry1 genes plays a key role to determine the insecticidal activity of Bt strains against P. xylostella.