Introduction
Plutella xylostella L. (Lepidoptera: Plutellidae), Diamondback Moth, is a pest that causes serious damages in plants of Brassicaceae family, particularly in cabbage culture (Brassica oleraceae var. capitata) (Castelo Branco et al., 1996; França and Medeiros, 1998) in Brazil and other countries. The damage in the production is around 60% (Imenes et al., 2002).

In the central region of Brazil, the attack of P. xylostella in the field occurs along the year, although its biggest occurrence is from July till September. The critical period of attack occurs from the fourth till seventh week after transplant, and either in the head formation (Castelo Branco et al., 2003).

The main method of its control is the use of chemical insecticides. There are a lot of reports showing four insecticide applications per week increasing the control cost and creating the potential for insect resistance, principally where the cultivation is uninterrupted (Villas Boas et al., 2004). This indicates the necessity of reducing the consuming of insecticides and the importance of using safe alternatives of control.

The use of entomopathogenic bacteria Bacillus thuringiensis (Bt) is a very important alternative for insect control. During the sporulation process, B. thuringiensis produces one or more proteins that are toxic to diamondback moth that are specific, innocuous to mammals, vertebrates and plants, and are not toxic to environment (Monnerat et al., 1999; Broderick et al., 2006).

The Laboratory of Entomopathogenic Bacteria at Embrapa Genetic Resources and Biotechnology has been studying, since 2003, a new way of controlling P. xylostella using B. thuringiensis in its endophytic form, in order to protect cabbage crops and other brassicas from this important endophagous pest (Monnerat et al., 2003; 2009).

The aim of this present study was to select and characterize new Bt strains toxic to P. xylostella to be used as a systemic bioinsecticide.

1 Results and Analysis
1.1 Bioassay of the B. thuringiensis strains against P. xylostella
The three strains S1905, S2122 and S2124 were toxic to P. xylostella. S1905, S2122 and HD-1 caused 100% of mortality after 48 hs, although S2124 caused 58.33% of mortality after 48 hours (Kruskal-Wallis: H3 =10.8, P =0.013) and 98.33% of mortality after five days (Kruskal-Wallis: H3 = 3.0, P = 0.392) (Table 1).
 

Table 1 Perceptual of mortality of second instar larvae of P. xylostella (media ± standard deviation) caused by Brazilian B. thuringiensis strains

After selective bioassays, the insects were submitted to lethal concentration bioassays. The CL₅₀ of the tree B. thuringiensis strains varied between 2.336 to 4.842 µg/mL (Table 2) and all of them were similar to HD-1 (ANOVA: F = 0,673, P = 0,595).
 

Table 2 Estimation of Lethal Concentration (LC50) of B. thuringiensis strains toxic to second instar larvae of P. xylostella after five days

1.2 Characterization of B. thuringiensis strains toxic to P. xylostella
The analysis of the spore-crystals mixtures by SDS-PAGE showed the presence of two proteins of 130 and 65 kDa in S1905 and S2124 strains and in S2122 showed only the 65 kDa protein (Figure 1).
 

Figure 1 Protein profile of B. thuringiensis strains HD1, S1905, S2193 e S2124

S1905 produced amplicons for genes cry1Aa, cry1Ab, cry1Ac, cry1B, cry2Aa and cry2Ab (Table 3). S2122 produced amplicons for seven genes cry1Aa, cry1Ab, cry1Ad, cry1C, cry1D, cry1F and cry2Ab. S2124 strain produced amplicons for cry1Ab, cry1E, cry2Aa and cry2Ab (Table 3).
 

Table 3 Protein profile, cry genes and types of crystal present in B. thuringiensis strains

1.3 Scanning and transmission electron microscopy of B. thuringiensis strains
The morphological characterization through scanning and transmission microscopy showed three different types of crystal protein inclusions. The strains produced bipyramidal, cuboidal and spherical crystals (Figure 2; Figure 3).
 

Figure 2 Scanning Electronic Microscopy of spore-crystals of Bacillus thuringiensis strains

Figure 3 Transmission of electronic microscopy of spores-crystals mixtures of B. thuringiensis strains

2 Discussion
The three strains S1905, S2122 and S2124 were tested against P. xylostella in selective bioassay showing toxicity. In the first 48 hours, S1905 and S2122 strains killed the insects faster than S2124 when compared with Btk HD-1 standard. Although with 96 hours, these strains didn’t present significative differences in the mortality. These results were similar to others researches with B. thuringiensis for the control of P. xylostella that showed high toxicity (Medeiros et al., 2005; Monnerat et al., 2007; Viana et al., 2009; Xie et al., 2010).

The three strains S1905, S2122, S2124 exhibited LC₅₀ values similar to the standard strain Btk HD-1. All of them presented a high toxicity against P. xylostella showed results similar to the others previously described (Monnerat et al., 2007).

The result of protein profile showed in SDS-PAGE is typical to lepidopteran active crystal protein from the Cry1 and Cry2 classes (Schenpf et al., 1998; Praça et al., 2004) and it is consistent with the results of the bioassays presented in this work.

In the cause of cry genes, the three strains produced PCR amplicons for cry1 and cry2 genes that are effective for lepidopteran pests. The strain S1905 presented six cry genes like Btk HD-1 and S997 strains that were toxic to Spodoptera frugiperda and Anticarsia gemmatalis (Praça et al., 2004). The strains S2122 presented seven genes and three of them (cry1Ab, cry1C and cry1D) presented in this strain showed a similarity to B. thuringiensis aizawai, a standard strains toxic to P. xylostella. The strain S2122 and S2124 present cry1E and cryF genes respectively and it is important to comment that these genes were less abundant. Some researchers say that they are found in the rains tropic (Bravo et al., 1998).

The genes cry1C and cry1D present in S2122 encoded toxins active against S. frugiperda and S. exigua (Bravo et al., 1998; Monnerat et al., 2006). All the strains presented cry2A genes that encoded proteins toxic to lepidopteran and dipteran species (Cárdenas et al., 2001). The genes that were found most commonly in the strains reported here were cry1Ab and cry2Ab. The presence of cry1 and cry2 genes in S1905 and S2124 strains was consistent with their production of the above proteins of 130 and 65 kDa (Figure 1), respectively.

It’s important to comment that Cry1 and Cry2 toxins are the most studied toxins and that the cry1 gene is the most abundant in the nature (Bravo et al., 1998; Rosa-Garcia et al., 2008).

In the cause of S2122 strain, it didn’t present the 65 kDa protein because it probably could have silent genes and this characteristic will be studied with more details in the future. The high toxicity of the strains to P. xylostella suggests that the Cry1Ab, Cry1Ac and Cry1C toxins present in the strains can be responsible to toxicity against this insect (Monnerat et al., 1999).

This crystals morphology of the strains was similar to B. thuringiensis kurstaki HD-1 as described previously (Praça et al., 2004; Monnerat et al., 2007; Santos et al., 2009; Sezen et al., 2010). The bypiramidal crystals may be related to the presence of Cry1 proteins and show the activity against lepidopteran and coleopteran insects, the cuboidal are associated to Cry2 proteins, which show the activity to lepidopterans and dipterans (Dankocsik et al., 1990; Wu et al., 1991; Bravo et al., 1998; Praça et al., 2004; Monnerat et al., 2007; Xie et al., 2010). Cuboidal crystals appear in the studied strains, although this crystal is known as toxicity to coleopteran insects B. thuringiensis strains characterized in this study can be used in the formulation of new bioinsecticide to manage the insect’s resistance and it’s also possible to be used endophytically for controlling P. xylostella.

3 Materials and Methods
3.1 Strains and culture conditions
Three strains of B. thuringiensis from the Collection of Invertebrate Bacteria of Embrapa Genetic Resources and Biotechnology were used in this work. These strains were obtained from soil and water samples from different parts of Brazil and were previously identified as pathogenic to Lepidopteran species of insects (http://plataformarg.cenargen. embrapa.br/rede-microbiana/colecoes-de-ulturas/colecao-e-bacterias-de-invertebrados/colecao-de-bacterias-de-invertebrados). B. thuringiensis subspecie kurstaki (Btk) HD-1, obtained from the Collection of Bacillus thuringiensis and Bacillus sphaericus of the Institute Pasteur, Paris, was used as standard.

All strains and HD-1 were grown in Embrapa medium (Monnerat et al., 2007), for 72 h., 200 rpm, 30℃ Then, they were centrifuged at 12,800 xg for 30 min, at 4℃ (BR4i centrifuge Jouan). The pellets, containing spores and crystals were frozen for 16 h and lyophilized for 18 h in Labconco model Lyphlock18 freeze-dryer.

3.2 Bioassays against P. xylostella
Bioassays were carried out using second instar larvae of P. xylostella reared in the laboratory at (26 ± 2)℃, (60 ± 10)% humidity and photophase of 12 h (Medeiros et al., 2003). Two kinds of bioassays were carried out, the selective one to confirm the toxicity of the strains and the quantitative one to determine the lethal concentration required to kill 50% of the tested insects (LC₅₀).

The selective bioassays were performed by the procedure described by Tabashnik et al (1990) with some modifications. Pieces of cabbage (Brassica oleracea L.) leaves (6 cm × 3 cm) were dipped vertically for 10 minutes in spore/crystal powder diluted 10% in distilled water. Leaves were then air dried vertically for 1 h at 25℃ then placed in a Petri dish (80 mm× 15 mm) with 10 larvae, in a total of tree replicates. The procedure of the selective bioassay was the same for each strain. After 48 h the surviving larvae were transferred to a new leaf and the mortality was assessed. Larval mortality was assessed again at day five (Monnerat et al., 1999) and the rate of mortality was determined and compared by variance analyses using Sigmastat analyses statistic (Kuo et al, 1992).

In order to determine the LC₅₀, 6 concentrations (between 100 µg/mL and 0.01 µg/mL) of the lyophilized mixture of spores and crystals were diluted in sterile water. Then, the cabbage leaves were dipped in each concentration as described for the previous bioassay. The bioassays were repeated three times. The mortality data were evaluated on the fifth day analyzed by Probit analyses and LC₅₀ (Robertson et al., 2002) was determined and analyzed by variance analyses (Kruskal-Wallis) using Sigma Stat analyses statistic (Kuo et al., 1992).

3.3 Characterization of the B. thuringiensis strains toxic to P. xylostella
3.3.1 SDS-PAGE (Polyacrylamide gel electrophoresis)
The proteins were extracted according to Lecadet et al (1991). The molecular mass and integrity of the proteins were determined by SDS-PAGE 10% (Laemmli, 1970). The electrophoresis was realized in Hoefer miniVE vertical electrophoresis system – Amersham Pharmacia in voltage of 120 V three hours and 30 minutes. The gel was stained and fixed in 40% methanol, 10% acetic acid and Coomassie blue (0.1%) for about 16 h, under slight shaking and it was distained in 40% methanol and 10% acetic acid for 2 h, with agitation. The HD-1 strain of B. thuringiensis subsp. kurstaki was used as control.

3.3.2 DNA sample preparation and PCR
The strains were grown on Embrapa agar medium (Monnerat et al., 2007) for 14~16 h, at 30℃ and DNA was extracted as described by Bravo et al. (1998). Molecular characterization through PCR was performed to identify the toxin-coding genes, by using a variety of oligonucleotide pairs specific for the following genes/gene families: cry1, cry2, cry3, cry4, cry5, cry7, cry8, cry9, cry10, cry11, cry12, cry13, cry14, cry17, cry19, cry21, cry24, cry25, cry27, cry29, cry30, cry32, cry39, cry40, cyt1 and cyt2 (Ceron et al., 1994; Ceron et al., 1995; Ben-Dov et al., 1997; Bravo et al., 1998; Ibarra et al., 2003). 15 µL of the supernatant was used as DNA sample in the PCR mixture containing 2.5 U Taq DNA polymerase enzymes (5.0 U), 0.5 µM of each primer used in the reactions and 0.2 mM of deoxynucleoside triphosphates mix in a total volume of 50 µL. The amplification reactions were carried out in PTC-100 (Programmable thermal Controller – MJ Research, inc). The PCR products were visualized in 1% agarose gel in tris-borate buffer at 150 V for 1 h and stained with ethidium bromide (10 µg/mL).

3.3.3 Scanning and Transmission electron microscopy
In order to observe the spores and crystals produced by the different strains in scanning electronic microscopy, the strains were grown in sporulation medium (Lereclus et al., 1995) at 28℃ for 72 h. Crystals were isolated by centrifugation in sucrose gradients (Chang et al., 1993). These preparations were washed and lyophilized before being deposited on a metallic support. The samples were covered with gold for 180 seconds, using a sputter coater (EMITECH model K550) and observed in a ZEISS model DSM 962 scanning electron microscope.

To observe the spores and crystals of B. thuringiensis by transmission electron microscopy, the culture was centrifuged at 8 000 rpm and fixed in 2.5% glutaraldehyde and 0.1 M sodium cacodylate buffer at pH 7.0 for 4 h and maintained under agitation overnight. Fixed bacteria were washed three times for 10 minutes with cacodylate buffer and twice with water. The bacteria were post-fixed in 2% osmium tetroxide and 0.2 M cacodylate buffer for 1 h in dark conditions. After washed with cacodylate buffer for three times and twice with water for 10 minutes, the pellets were dehydrated in a 10 to 100% ethanol series for 20 minutes for each step and twice in pure ethanol for completely dehydration. The bacteria were embedded in Epon resin at 4℃ under agitation and incubated at 70℃. The ultrathin sections (60 µm) were stained in 2% uranyl acetate at dark conditions for 1 hour. The ultrathin sections were washed in water and observed in a JEOL1011C Transmission Electron Microscope. Between changes of solution, the samples were centrifuged at 8000 rpm for few seconds to form the pellets.

4 Conclusion
The strains S1905, S2122 and S2124 were toxic to P. xylostella and present different genes. These strains are important to the development of new bioinsecticides that can be used in the control of P. xylostella and the management of insect resistance in the field.

Authors’ contribution
LP conducted all the research for this paper and prepared the manuscript. CC participated in bioassays and in biochemistry and molecular characterization. AG conducted the scanning and transmission electron microscopy. RM reviewed the manuscript and coordinated the project.

Acknowledgement
This work was financially supported by FAP-DF and Embrapa.

References
Ben-Dov E., Zaritsky A., Dahan E., Barak Z., Sinal R., Manasherob R., Khamraev A., Troitskaya E., Dubitsky A., Berezina N., and Margalith Y., 1997, Extended screening by PCR for seven cry-group genes from field collected strains of Bacillus thuringiensis, Applied and Environmental Microbiology, 63(12): 4883-4890
PMid:9406409 PMCid:168816

Bravo A., Sarabia S., Lopez L., Ontiveros H., Abarca C., Ortiz A., Ortiz M., Lina L., Villa-lobos F.J., Guadalupe P., Nunez-Valdez M.E., Soberón M., and Quintero R., 1998, Characterization of cry genes in Mexican Bacillus thuringiensis strain collection, Applied and Environmental Microbiology, 64: 4965-4972
PMid:9835590 PMCid:90950

Broderick N.A., Raffa K.F., and Handelsman J., 2006, Midgut bacteria required for Bacillus thuringiensis insecticidal activity, Proceedings of the National Academy of Science, 103(41): 15196-15199
http://dx.doi.org/10.1073/pnas.0604865103 PMid:17005725 PMCid:1622799

Cárdenas M.I., Galán-Wong L., Ferré-Manzanero J., and Pereyra-Alférez B., 2001, Selección de toxinas cry contra Trichoplusia ni, Ciencia Uanl, 4:51-62

Castelo Branco M., França F.H., Pontes L.A., and Amaral P.S., 2003, Avaliação da suscetibilidade a inseticidas de populações da traça-das-crucíferas de algumas áreas do Brasil, Horticultura Brasileira, 21(3): 553-556
http://dx.doi.org/10.1590/S0102-05362003000300027  

Castelo Branco M., Villas Boas G.L., and França F.H., 1996, Nível de dano da traça das crucíferas em repolho, Horticultura Brasileira, Brasília, DF, 14(2): 154-157

Ceron J., Covarrubias L., Quintero R., Ortiz A., Ortiz M., Aranda E., Lina L., and Bravo A., 1994, PCR analysis of the cryI insecticidal crystal family genes from Bacillus thuringiensis, Applied and Environmental Microbiology, 60: 353-356
PMid:8117089 PMCid:201313

Ceron J., Ortiz A., Quintero R., Guereca L., Bravo A., 1995, Specific PCR primers directed to identify cryI and cryIII genes within a Bacillus thuringiensis strain collection, Applied Environmental Microbiology, 61: 3826-3831
PMid:8526493 PMCid:167686

Chang C., Yu Y.M., Dai S.M., Law S.K., Gill S.S., 1993, High-level cryIVD and cytA gene expression in Bacillus thuringiensis does not require the 20-kilodalton protein, and coexpressed gene products are synergistic in their toxicity to mosquitoes, Applied Environmental Microbiology, 59: 815–821
PMid:8481007 PMCid:202194

Dankocsik C., Donovan W.P., and Lany C.S., 1990, Activation of a cryptic crystal protein gene of Bacillus thruingiensis subspecies kurstaki by gene fusion and determination of the crystal protein insecticidal specificity, Molecular Microbiology, 4(12): 2087-2094
http://dx.doi.org/10.1111/j.1365-2958.1990.tb00569.x PMid:2089222

França F.H., and Medeiros M.A., 1998, Impacto de combinação de inseticidas sobre a produção de repolho e parasitóides associados com a traça-das-crucíferas, Horticultura Brasileira, 16(2): 132-135

Ibarra J., Rincón C., Ordúz S., Noriega D., Benintende G., Monnerat R.G., Regis L., Oliveira C.M.F., Lanz H., Rodriguez M.H., Sánchez J., Peña G., and Bravo A., 2003, Diversity of Bacillus thuringiensis strains from Latin America with insecticidal activity against different mosquito species, Applied and Environmental Microbiology, 69: 5269-5274
http://dx.doi.org/10.1128/AEM.69.9.5269-5274.2003 PMid:12957913 PMCid:194983

Imenes S.D.L., Campos T.B., Rodrigues Netto S.M., Bergmann E.C., 2002, Avaliação da atratividade de feromônio sexual sintético da traça-das-crucíferas, Plutella xylostella (L.) (Lepidoptera: Plutellidae), em cultivo orgânico de repolho, Arquivos do Instituto Biológico, 69(1): 81-84

Kuo J., Fox E., and Macdonald S., 1992, Sigmastat: statistical software for working scientists, Users manual, Jandel ScientiWc, San Francisco, CA

Laemmli U.K., 1970, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature, 227: 680-685
http://dx.doi.org/10.1038/227680a0 PMid:5432063

Lecadet M.M., Chaufaux J., Ribier J.E., and Lereclus D., 1991, Construction of novel Bacillus thuringiensis strains with different insecticidal activities by transduction and transformation, Applied and Environmental Microbiology, 58: 840-849

Lereclus D., Bourgouin C., Lecadet, M.M., Klier A., and Rapoport G., 1989, Role, structure and molecular organization of the genes coding for the parasporal d-endotoxins of Bacillus thuringiensis. In: Issar Smith R., Slepecky A., Setlow P. (Eds.), Regulation of procaryotic development, American Society for Microbiology, Washington, pp.71-88

Medeiros P.T., Dias J.M.C.S., Monnerat R.G., and Souza N.R., 2003, Instalação e manutenção de criação massal da traça-dascrucíferas (Plutella xylostella), Brasília, DF: Embrapa Recursos Genéticos e Biotecnologia, 4p. (Embrapa Recursos Genéticos e Biotecnologia. Circular técnica, 29)

Medeiros P.T., Ferreira M.N., Martins E.S., Gomes A.C.M.M., Falcão R., Dias J.M.C.S., and Monnerat R.G., 2005, Seleção e caracterização de estirpes de Bacillus thuringiensis efetivas no controle da traça-das-crucíferas Plutella xylostella, Pesquisa Agropecuária Brasília, 40(11): 1145-1148
http://dx.doi.org/10.1590/S0100-204X2005001100014  

Monnerat R.G., Batista A.C., Medeiros P.T., Martins E., Melatti V., Praça L., Dumas V., Demo C., Gomes A.C., Falcão R., and Berry C., 2007, Characterization of brazilian Bacilus thuringiensis strains active against Spodoptera frugiperda, Plutella xylostella and Anticarsia gemmatalis, Biological Control, 41: 291-295
http://dx.doi.org/10.1016/j.biocontrol.2006.11.008  

Monnerat R.G., Martins E., Queiroz P., Ordúz S., Jaramillo G., Benintende G., Cozzi J., Real M.D., Martinez-Ramirez C.R., Cerón J., Ibarra J.E., Rincon-Castro M.C.D., Espinoza A.M., Meza-Basso L., Cabrera L., Sánchez J., Soberon M., and Bravo A., 2006, Genetic variability of Spodoptera frugiperda Smith (Lepidoptera: Noctuidae) populations from Latin America is associated with variations in susceptibility to Bacillus thuringiensis Cry toxins, Applied and Environmental Microbiology, 72(11): 7029-7035
http://dx.doi.org/10.1128/AEM.01454-06 PMid:16936049 PMCid:1636186

Monnerat R.G., Santos R., Barros P., Batista A., and Berry C., 2003, Isolamento e caracterização de estirpes de Bacillus thuringiensis endofíticas de algodão, Brasília, Embrapa Recursos Genéticos e Biotecnologia, 4p. (Embrapa Recursos Genéticos e Biotecnologia, Comunicado técnico, 98)

Monnerat R., Masson L., Brousseau R., Pusztai-carey M., Bordat D., and Frutos R., 1999, Differential activity and activation of Bacillus thruringiensis insecticidal proteins in Diamond moth, Plutella xylostella, Current Microbiology, 39: 159-162
http://dx.doi.org/10.1007/s002849900438 PMid:10441730

Monnerat R.G., Soares C.M.S., Gomes A.C.M., Jones G., Martins E., Praça L., and Berry C., 2009, Translocation and insecticidal activity of Bacillus thuringiensis bacteria living inside of plants, Microbiol Biotechnology, 2: 1560-1562
http://dx.doi.org/10.1111/j.1751-7915.2009.00116.x PMid:21255282

Praça L.B., Batista A.C., Martins E.S., Siqueira C.B., Dias D.G., De S., Gomes A.C.M.M., Falcão R., and Monnerat R.G., 2004, Estirpes de Bacillus thuringiensis efetivas contra insetos das ordens Lepidoptera, Coleoptera e Diptera, Pesquisa Agropecuária Brasileira 39(1): 11-16
http://dx.doi.org/10.1590/S0100-204X2004000100002  

Robertson, J. R., Preisler, H. K., Russell, R. M., 2002, Polo Plus. Probit and logit analysis user’s guide. LeOra Software, Petaluna, CA

Santos K.B., Neves P., Meneguim A.M., Dos Santos R.B., Dos Santos W.J., Boas G.V., Dumas V., Martins E., Praça L.B., Queiroz P., Berry C., and Monnerat R., 2009, Selection and characterization of the Bacillus thuringiensis strains toxic to Spodoptera eridania (Cramer), Spodoptera cosmioides (Walker) and Spodoptera frugiperda (Smith) (Lepidoptera: Noctuidae), Biological Control, 50: 157-163
http://dx.doi.org/10.1016/j.biocontrol.2009.03.014  

Schnepf E., Crickmore N., Van Rie J., Lereclus D., Baum J., Feitelson J., Zeigler D.R., Dean D.H., 1998, Bacillus thuringiensis and its pesticidal crystal proteins, Microbiology Molecular Biology Review, 62: 775-806
PMid:9729609 PMCid:98934

Sezen K., Kati H., Muratoglu H., and Demirbag Z., 2010, Characterization and toxicity of Bacillus thuringiensis strains form hazelnut pests and fields, Pest Management Science, 66: 543-548
http://dx.doi.org/10.1002/ps.1905 PMid:20024949

Tabashnik B.E., Cushing N.L., Finson N., and Johnson M.W., 1990, Field development of resistance to Bacillus thuringiensis in diamondback moth (Lepidoptera: Plutellidae), Journal of Economic Entomology, 83: 1671-1676

Viana C.L.T.P., De Bortoli S.A., Thuler R. T., Goular T.R.M., Thuler, A. M. G., Lemos, M. V.F., and Ferraudo A.S., 2009, Efeito de novos isolados de Bacillus thuringiensis Berliner em Plutella xylostella (Linnaeus, 1758) (Lepidoptera: Plutellidae), Científica 37(1): 22-31

Villas Bôas G.L., Castelo Branco M., Medeiros M.A., Monnerat R.G., and França F.H., 2004, Inseticidas para o controle da traça-das-crucíferas e impactos sobre a população natural de parasitóides, Horticultura Brasileira, 22: 696-699
http://dx.doi.org/10.1590/S0102-05362004000400006