Isolation of B. thuringiensis

Samples obtained from soil, larvae of Spodoptera frugiperda J.E.Smith (Lepidoptera: Noctuidae), larvae of Chlosyne lacinia ssp. lacinia Geyer (Lepidoptera: Nymphalidae), adults of Aidemona azteca Saussure (Orthoptera: Acrididae), leaves of Nerium oleander L. and stubble of Zea mays L. These materials were obtained from different regions of the Mexican Southeast the soil samples were collected eight in the state of Tabasco, eight in Yucatan, one in Campeche and one in Quintana Roo. Regarding the other materials were obtained, 23 samples of larvae of C. lacinia, two samples of adults of A. azteca and four samples of leaves of N. oleander in the municipality of Conkal, six samples of larvae of S. frugiperda and five samples of corn stubble in the municipality of Tekax, both in the state of Yucatán (Table 1). For the isolation of bacteria from soil, 5 g of the sample were weighed and placed in a 15 mL Falcon tube, which was calibrated with sterile distilled water until said volume and the mixture was resuspended by vigorous vortexing for 1 min. It was pasteurized in a water bath at 80 °C for 15 min and cooled immediately on ice.

The 5 mL of the suspension was taken and added to a flask containing 50 mL of nutrient broth (BD Bioxon^®) and 0.25 M sodium acetate (Fermont) (to inhibit the germination of B. thuringiensis spores), incubated for 4 h at 200 rpm and at 29 °C. Then 1 mL of the culture was taken and diluted in 1 mL of sterile distilled water. This sample was again subjected to the pasteurization process, which is recommended to inactivate the growth of other microorganisms in the culture medium (^{Travers et al., 1987}). At the end of the second pasteurization process, an aliquot of 100 μL was taken and evenly distributed in a Petri dish with nutritive agar. The Petri dishes were incubated at 30 °C for 24 h to favor the germination of B. thuringiensis spores and the development of bacterial colonies.

For the bacterial isolation from larvae and adult insects, 3 to 5 individuals were processed per sample depending on the size of these, which were macerated directly in the tube with the help of a glass rod and then hardened with sterile distilled water until 15 mL, vigorously shaken (^{Aldebis et al., 1994}). In the case of isolates from plant leaves, 2 g of the sample was weighed, disinfected in 2% sodium hypochlorite solution, for 10 min and double washing was applied with distilled water, the sample was finely macerated in a mortar and 15 mL of sterile distilled water were added, the solution was transferred to a 15 mL Falcon tube (^{Monnerat et al., 2009}). In both cases, a pasteurization process was carried out and the described process was continued to obtain isolates from soil samples.

Individual colonies were selected according to the typical characteristics reported for Bacillus (^{Sneath, 1986}). Subsequently tests were performed to verify positive Gram stain and positive catalase reaction. Selected isolates were individually cultured on nutrient agar (Bioxon^®) and stored at 4 °C for further studies.

Morphological characterization of the parasporal body

The presence and parasporal morphology of the selected isolates was carried out by direct observation of the crystals under DM500 optical microscope (Leica Microsystems, Switzerland) (1000X), morphology was confirmed by direct smear in phase contrast and staining with Coomassie blue (^{Sharif and Alaeddinoĝlu, 1988}).

Extraction of genomic DNA

Total DNA extraction was performed according to the modified method reported by ^{Rosso and Delécluse (1997)}. A bacterial culture of 2 mL was obtained in Luria-Bertani^® medium (Invitrogen) at 28 °C with constant agitation at 250 rpm overnight, centrifuged at 8 000 rpm for 5 min, the supernatant and the cell pack were removed it was resuspended in 500 μL of buffer J (1.0 M TrisHCl, 0.1 M EDTA and 0.15 M NaCl, pH 8). Again, it was centrifuged under the same conditions as above the cell pack was resuspended in 300 μL of buffer J containing 40 mg mL of lysozyme (Sigma Aldrich Quimica) and incubated at 37 °C for 1 h. Subsequently, 20 μL of 10% SDS was added, it was homogenized gently and incubated at 70 °C for 20 min.

The 5 μL of RNAse (Promega) (10 mg mL), 10 μL of proteinase K (10 mg mL) (Thermo Fisher Scientific) were added and incubated at 60 °C for 90 min. 50 μL of 5 M NaCl (Baker, J. T.) was added and incubated on ice while stirring constantly. It was centrifuged at 13 000 rpm for 20 min, the supernatant was recovered, precipitated with the same volume of isopropanol (Sigma) and incubated at -70 °C for 30 min. Subsequently, it was centrifuged under the above conditions and the supernatant was removed, the pellet was washed with 200 μL of 70% ethanol (Fermont) and allowed to dry at room temperature, the DNA was resuspended in 20 μL of sterile distilled water and preserved at -20 °C until use.

Amplification of cry and hag genes

To determine the presence of cry genes in the isolates, PCR amplification was performed with the universal primers of block 1 (5’ TATGCWCAAGCWGCCAATYTWCATYT3’) and block 5 (5’ GGRATAAATTCAATTYKRTCWA 3’) (Sigma Aldrich Química) according to ^{Noguera and Ibarra (2010)}.

The presence of the hag gene of flagellin in the native isolates was determined by PCR with the specific primers Bthag-F1 (5’-AGTACATGCGCCAAAACCAAG) and Bthag-R1 (5’-GTTTGCTTGAGAAAGCATGCT) (Sigma Aldrich Chemistry) according to ^{Xu and Cote (2006)}. The PCR products were verified by 0.8% agarose gel electrophoresis (Invitrogen) and visualized in a Biorad Molecular Imager Gel Doc XR photodocument (Bio-Rad^®).

Protein profile

The protein profile was obtained by SDS-PAGE electrophoresis of the spore-crystal complex of the isolates according to the reported methodology (^{Laemmli 1970}). For this, each isolate was inoculated in 5 mL of LB medium and incubated in an orbital shaker at 30 °C with 200 rpm for 72 h. The spore-crystal complex was centrifuged at 10 000 rpm for 10 min and washed with sterile distilled water, this operation was repeated three times. The tablet was recovered and stored at 4 °C. A mixture was prepared with 5 μL of loading buffer with 5% β-mercaptoethanol (Sigma Aldrich Química), 3 μL of the spore-crystal complex sample and 2 μL of water, incubated at 95 °C for 10 min.

For the electrophoresis a 12% separating gel and the 4% polyacrylamide compactor (Sigma Aldrich Química) were prepared, the components used were 30% polyacrylamide, Tris-HCL buffer pH 8.8 (Sigma Aldrich Química), Tris-HCL buffer pH 6.8, 10% SDS, sterile distilled water, TEMED (Invitrogen^®) and 10% ammonium persulphate (Sigma Aldrich Química), a volume of 5 mL was prepared for each gel and the concentrations were calculated according to the instructions in the manufacturer’s manual. Electrophoresis was performed in a modular vertical electrophoresis chamber Enduro™ (Bio-Rad), using the Page Ruler™ Plus Prestained Protein Ladder molecular weight marker (Bio-Rad^®).

The electrophoresis was carried out in two phases, that of compaction at 40 V and 100 A for 1 h and the separation phase was carried out at 90 V and 100 A for 3 h. Subsequently the gel was stained with a solution of Coomassie blue dye G-250 0.1% for 1 h and stained in methanol/acetic acid. The molecular weight of the protein bands present in each sample was estimated according to the molecular weight marker used. The protein profiles observed were compared with the profile of the reference strain B. thuringiensis subsp. kurstaki HD1.

Detection of insecticide activity

A preliminary test was carried out at a high concentration with the isolates obtained to determine the insecticidal activity against Manduca sexta, since several studies report low lethal doses for several lepidopteran insects ranging from 1 to 164 ng cm^-2 (^{Uribe et al., 2003}; ^{Sharma et al., 2010}). The lyophilisate of the spore-crystal complex was obtained in a Labconco Lyph-Lock 4.5 lyophilizer (Labconco^®), a final concentration of 10 μg cm^-2 was used. The sample was weighed and diluted in 200 μL of Tween 80 (Sigma Aldrich Chemistry) (0.02%), added to the Petri dish with the artificial diet (^{Yamamoto, 1969}) for the insect and distributed evenly over the entire surface of the diet, the surface was allowed to dry for 45 min and 20 neonatal larvae were placed for each isolate with two repetitions and the mortality was recorded at 96 h, as a negative control, 0.02% Tween 80 was used.

Isolation of B. thuringiensis

Sixty-two samples of various collected materials were processed, including 18 soil samples, 10 of Spodoptera frigiperda, 23 of Chlosyne lacinia ssp. lacinia, two of Aidemona azteca, four of Nerium oleander leaves and five samples of stubble of Zea mays. A total of 210 colonies were selected, of which only 27 were observed in the presence of structures in the sporangia, among which were observed spherical, amorphous, bipyramidal, cubic, ovoid and some bodies adhered to the spores. The Coomassie blue staining test (Figure 1) confirmed the presence of parasporal crystals in only three isolates, which were designated with the ITCBT34, ITCBT61 and ITCBT62 keys (Table 2). These selected isolates were used in subsequent tests.

Figure 1 Morphology of the parasporal body by staining with Coomassie blue. A= ITCBT34. B= ITCBT61. C= ITCBT62; b) vegetative cell; c) parasporal crystal; and e) spore. 

Previous studies have reported obtaining native B. thuringiensis isolates from habitats such as soil (^{Adbullah et al., 2014}), plants (^{Alper et al., 2014}) and insects (^{Alquisira-Ramírez et al., 2014}). In this diversity of environments, it is possible to find isolates that have dipyramidal, spherical, cubic crystals (^{Cicero et al., 2009}, ^{Azizoglu et al., 2011}) and irregular ones (^{Assaeedi et al., 2011}). In relation to southeastern Mexico, there are reports of B. thuringiensis isolates only from soils in the states of Tabasco (^{Bravo et al., 1998}) and Yucatán (^{Ornelas-Pérez et al., 2016}).

In the present work the selected isolates that were confirmed as B. thuringiensis presented oval crystals, sample obtained from soil, and dipyramidal samples obtained from larvae of Chlosyne lacinia (Table 2). The isolation index of B. thuringiensis obtained was 0.11. This result is within the values previously reported, since it is possible to find reports with indexes of 0.04 to 1 (^{Rosas-García et al., 2008}; ^{Alper et al., 2014}).

Cry gene amplification

The three isolates showed the presence of cry genes, presented amplification bands with approximate sizes of 1400 bp (ITCBT61), 1500 bp (ITCBT62), 1650 bp (ITCBT34) and 1300 bp in the reference strain (HD1) (Figure 2a). The use of universal primers is common to determine the presence of cry genes (^{Patel et al., 2012}; ^{Reinoso-Pozo et al., 2016}). According to the dipyramidal morphology of the parasporal crystals of the ITCBT61 and ITCBT62 isolates, it is likely that the amplification bands are correlated with cry1 and cry2 genes, which encode toxic proteins mainly towards Lepidoptera larvae (^{Sun et al., 2007}; ^{Sauka and Benintende, 2008}). On the other hand, the fragment amplified in the ITCBT34 isolate shows a different band size (Figure 2a), so it could be another type of cry gene.

Figure 2 Amplification of cry (a) genes and hag (b) genes in the isolates of B. thuringiensis. PM: molecular marker 1 Kb Plus DNA Ladder (Invitrogen, USA), Lane 1: HD1, Lane 2: ITCBT61, Lane 3: ITCBT62, Lane 4: ITCBT34. 

Determination of hag genes

The three isolates presented hag gene amplification products by PCR (Figure 2b). Isolates ITCBT61 and ITCBT62 presented a band of approximately 700 bp in size, similar to the band that showed the reference strain HD1. The isolate ITCBT34 presented a band of different size of approximately 1000 bp. The amplified fragments of the hag genes reported in B. thuringiensis show polymorphism with bands with sizes from 700 bp up to 1.9 kb (^{Xu and Cote, 2006}). The fragment size of the hag gene of the isolate ITCBT34 reinforces that it is an isolate different from the other two (ITCBT61 and ITCBT62) and to the reference strain HD1. The determination of the presence of flagellin genes is a method that is gaining use as an alternative to serotyping in B. thuringiensis (^{Hendriksen and Hansen, 2006}; ^{Reinoso-Pozo et al., 2016}).

Protein profile

By means of the SDS-PAGE analysis, the protein profile of the three isolates under study was determined. The profiles observed were different among the isolates and only the isolate ITCBT61 coincided with the profile of the reference strain HD1. In the case of the isolate ITCBT62, the presence of a band of 45 kDa was observed and the absence of a band of 130 kDa that is present in the isolate ITCBT61 and the strain HD1, the ITCBT34 presented a profile of bands different from 100, 85, 70 and 50 kDa (Figure 3).

Figure 3 SDS-PAGE of the protein profile presented by the native isolates of B. thuringiensis. PM= Molecular weight marker PageRuler ™ Plus Prestained Protein Ladder (Thermo Scientific), lane 1: HD1, lane 2: ITCBT61, lane 3: ITCBT62, lane 4 = ITCBT34. 

We report mainly bands of 130 and 65 kDa for the strain HD1, which are product of proteins encoded by the cry1 and cry2 genes (^{Alper et al., 2014}), so that the bands of 130 and 65 kDa of molecular weight that presented the native isolates, they can belong to the group of proteins Cry1 and Cry2. The protein profile that has been observed in other works varies, but the molecular weight range of the bands is 130 to 44 kDa (^{Silva et al., 2010}; ^{Barathi et al., 2012}). Previous studies have reported the presence of proteins of sizes similar to those obtained in the present work, which showed toxicity mainly against pests of the order Lepidoptera (^{Valicente et al., 2010}; ^{Li and Bouwer, 2012}).

Insecticidal activity

The test of insecticidal activity was carried out with a concentration of 10 μg cm^-2, the three isolates showed insecticidal activity against Manduca sexta larvae. At 96 h, 100% mortality of larvae was observed (Table 3). The larvae exposed to the isolates showed symptoms of general necrotic and internal liquefied, characteristics typical of the toxicity caused by the Cry proteins of B. thuringiensis (^{Silva et al., 2010}) (Figure 4).

Figure 4 Qualitative test of native isolates of B. thuringiensis against newborn larvae of Manduca sexta at a concentration of 10 μg cm^-2. A) negative control (Tween 80), B)= ITCBT34, C)= ITCBT61 and D)= ITCBT62. 

The activity observed in the ITCBT61 and ITCBT62 isolates responds to expectations, since both isolates have a dipyramidal parasporal morphology and protein profile typical of the Cry1 and Cry2 proteins, which have insecticidal activity against Lepidoptera larvae. Isolate ITCBT34 also had activity towards these larvae. This result is not common, since this isolate presented parasporal body morphology and protein profile very different from the type of proteins that act on Lepidoptera larvae. This activity may be related to the presence of cry2 genes, which were reported in B. thuringiensis isolates with only oval morphology crystals (^{Arrieta and Espinoza, 2006}).

The presence of cry1 and cry2 genes and activity against Lepidoptera larvae is characteristic of B. thuringiensis, and is frequently reported in studies on the search and characterization of native isolates. Were used in these works, several species of Lepidoptera such as Cadra cautella and Thaumetopoea wilkinsoni (^{Yilmaz et al., 2013}), Spodoptera littoralis (^{Assaeedi et al., 2011}), Heliothis armígera and Plutella xylostella (^{Xavier et al., 2007}) and even against Coleoptera larvae Carpophilus hemipterus (^{Alper et al., 2014}).