Molecular Characterization of Bacillus thuringiensis cyt Genes and Their Effect Against Fall Armyworm , Spodoptera frugiperda

Bacillus thuringiensis (Bt) endotoxins (Cyt and Cry) have been extensively explored for biological control of fall armyworm, Spodoptera frugiperda, an important pest in major corn-producing countries of America. Five hundred Bt isolates with different toxicities against S. frugiperda were characterized by PCR for the presence of cyt genes (cyt1, cyt2, cyt1Aa, cyt1Ab, cyt2B and cyt2Ba), and the effect of insecticidal proteins Cry1Ba, Cry1Ca, Cry1Da and Cyt on S. frugiperda larvae were evaluated. Six isolates showed the presence of cyt genes, three isolates harbored two gene families (cyt1 and cyt2), and three isolates harbored only one of the cyt gene families (cyt1 and cyt2). It was not possible to correlate the presence/absence of cyt genes with toxicity against S. frugiperda. In this study, cyt genes were present in toxic and nontoxic isolates to this insect pest. Bioassays against S. frugiperda larvae showed that only Cry1Ca protein had toxicity, with 77.08% of mortality. Synergism among Cry and Cry proteins used in this study against S. frugiperda was not observed.


Introduction
Crop damage due to insect pests and diseases could account for up to 35% of total losses (Pardo-Lopéz et al., 2013).The fall armyworm, Spodoptera frugiperda (J.E. Smith), is an important yield-and quality-limiting pest in major corn-producing countries of America (Storer et al., 2012).In tropical regions, the insecticide application is frequently needed to maintain the S. frugiperda population below economic thresholds as it is the main method of control.In recent years, microbial insecticides have become a viable alternative to control fall armyworms, and are considered a safe tool in Integrated Pest Management system (Valicente et al., 2010).
Cry toxins bind to specific protein receptors in the microvilli of the mosquito midgut cells (Bravo et al., 2007).In contrast, Cyt toxins do not bind to protein receptors but directly interact with membrane lipids inserting into the membrane and forming pores.However, synergism between Cyt and Cry toxins with some insecticidal proteins present in the mosquitocidal Bacillus thuringiensis subsp.israelensis (Bti) strain has been observed (Chang et al., 1993;Wu et al., 1994).Together, the Cyt1A and Cry11Aa proteins bind to the apical membrane in the same midgut region of the mosquito larvae, increasing the activity of Bti crystals (Pérez et al., 2005).To present, the molecular mechanism of synergism is unknown; however, the use of Cyt proteins combined with other mosquitocidal toxins could be used to delay the development of resistance and increase the activity of microbial insecticides (Wu et al., 2008).This work aimed the molecular characterization of some Bt isolates with different toxicities against S. frugiperda for the presence of cyt genes and evaluated the effect of insecticidal proteins Cry1Ba, Cry1Ca, Cry1Da and Cyt on S. frugiperda larvae.

Bacterial Strains
Five hundred Bt isolates obtained from different geographical regions from Brazil from Embrapa Maize and Sorghum Bt Collection were used.These isolates expressed different toxicities against S. frugiperda larvae (Valicente & Barreto, 2003).In addition, Bt var.israelensis (Bti) and var.kurstaki (HD1) were kindly provided by São Paulo State University (UNESP/Jaboticabal, São Paulo, Brazil) and United States Department of Agriculture (Columbus, Ohio, USA), and were used as positive and negative controls, respectively.Bt isolates were grown accordingly as described in previous work (Valicente et al., 2010).

Sample Preparation and PCR Analysis
DNA isolation was performed by heat shock protocol (Bravo et al., 1998) Proteins were isolated as previously described (Valicente et al., 2010), with some modifications.Bacteria (E.coli) were grown at 37 ºC containing 100 µg/mL of ampicillin, and 28 ºC (Bti) in 250 ml LB medium, with continuous shaking at 150 rpm.One mM of IPTG was added in half of the treatments, and each protein had two replicates.After 4 days, the cultures were centrifuged at 10,000 rpm for 10 min and the pellet resuspended in 15 ml 0.01% Triton solution.This step was repeated three times.Pellets composed of a mixture of spore-crystal were solubilized in 10 mL of solubilization buffer (0.01% Triton, 10 mM NaCl and 50 mM Tris-HCl, pH 8.0).The mixture was centrifuged at 10,000 rpm for 10 min, and the pellet was resuspended in 10 mL of sodium bicarbonate buffer (50 mM sodium bicarbonate and 10 mM β-mercaptoethanol, pH 10.5), and maintained at 37 ºC for 3 h under continuous shaking.Samples were centrifuged at 10,000 rpm for 15 min, and the pellet resuspended in 5 mL of 0.1 M Tris, pH 8.0, whereas the supernatant was transferred to a new tube and the pH adjusted to 8.0-8.5.The supernatant of the half treatments was treated with trypsin (0.5 µg µl -1) and incubated at 37 ºC for 2 h under continuous shaking, whereas the other half of the treatments were centrifuged without this step.The reaction was inactivated with 0.15 mM PMSF (Phenyl Methyl Sulfonyl Fluoride).Samples were centrifuged at 10,000 rpm for 15 min and washed with distilled water.This step was repeated three times.SDS-PAGE analysis and observation of the crystals under a phase contrast microscope were performed to confirm the presence of proteins.The proteins were lyophilized, weighed and eluted in distilled sterile water and maintained at concentrations of 50 mg/mL and 100 mg/mL.

Insects Colony and Bioassays
Two-day-old healthy S. frugiperda larvae were provided by the Biological Control Laboratory at Embrapa Maize and Sorghum Research Center (Sete Lagoas, Minas Gerais, Brazil).Each bioassay was composed of four replicates of 12 larvae each, and a total of 48 larvae per treatment.
In the first bioassay that aimed to evaluate the toxicity of Cry and Cyt proteins to S. frugiperda larvae isolated, trypsin-activated and non-activated E. coli clones were grown in LB medium with 1 mM of IPTG, and the proteins were activated and non-activated with trypsin (0.5µg µl -1).In the second bioassay that aimed to evaluate the synergism between the Cry (Cry1Ba, Cry1Ca and Cry1Da) and Cyt proteins (Bti strain) against S. frugiperda larvae, all proteins were treated with trypsin.The proteins were obtained as previously described (Valicente et al., 2010).The protein concentration used in first bioassay was 50 mg/mL of bacterial powder.In protein combinations (second bioassay), ratios were used, 1:1 and 1:3, corresponding to 50 mg/mL:50 mg/mL and 50 mg/mL:150 mg/mL, respectively, of each protein powder, with conditions adapted from a previous study (Pérez et al., 2005).
The larvae were transferred individually to 50 ml disposable cups, each one containing 5 g of artificial diet (123.6 g/L cooked bean, 59.3 g/L wheat germ, 38.0 g/L brewer's yeast, 3.82 g/L ascorbic acid, 2.36 g/L Nipagin® M, 1.23 g/L sorbic acid, 15.35 g/L agar, 3.1 g/L formaldehyde, 0.131 mL/L fosforic acid, 1.3 mL/L propionic acid) with 50 µl of protein preparation.Larvae were kept at 25 ºC, 70% humidity and 14 h/10h photophase.Mortality (%) was evaluated determined 5 days after larvae toxin exposition.A negative control was composed of 5 g of artificial diet containing distilled water and Tween 20.

Statistical Analysis
The assumptions of normality of errors and homogeneity of variance were analyzed with the programs UNIVARIATE and GLM.The data were submitted to Kruskal-Wallis test at 5% significance by the NPAIRWAY procedure of SAS version 9.2.

Screening of cyt Genes in Bt Collection
Molecular analysis of PCR products showed that fragments with expected size were generated, and six strains out of 500 showed cyt genes.Three strains (1168C, 1646JAB and 1656JAB) showed two gene families (cyt1 and cyt2), and three strains only showed one of the cyt families (257A with cyt1, LT09 and P283 with cyt2).Figures 1  and 2 shows the electrophoretic profiles of the strains using specific primers for cyt1 and cyt2, respectively.The amplified fragments were confirmed by sequencing (Table 2).Synthetic oligonucleotide primers were designed corresponding to the regions of no specific homology for the cyt1 and cyt2 families, and were used in these strains that harbored cyt genes.All strains containing cyt1 genes showed the gene cyt1Aa; only the strain 257A also showed the cyt1Ab gene.Out of 500 strains tested for cyt2 genes, only one strain (P283) showed the cyt2Ba gene.

Bioassays
The first bioassay aimed to evaluate the toxicity of Cry and Cyt proteins to S. frugiperda larvae using each protein isolated, both trypsin-activated and non-activated.The wild-type Bti strain was used to provide the Cyt proteins and recombinant E. coli strains expressing Bt genes to recover the Cry proteins (Cry1Ba, Cry1Ca and Cry1Da).In Bti, the endotoxin parasporal body makes up approximately 30% of the dry weight of sporulated cells (Ibarra & Federici, 1986).The concentration of toxins within the Bti parasporal body and percentage of the recombinant E. coli strains powder were estimated based on SDS-PAGE analyses.The protein concentration used in this experiment was 50 mg/mL of bacterial powder.Bioassays data showed that only Cry1Ca powder was effective against S. frugiperda, with 77.08% of average mortality (Table 3), and the activation by trypsin increased the mortality of S. frugiperda larvae.The Cry1Ca non-activated powder showed an average mortality of 18.75%.The others two Cry proteins used (Cry1Ba and Cry1Da) showed low toxicity, with values lower than 8.3% mortality.The Cry1Ba protein was not toxic to S. frugiperda larvae with 0% of mortality, showing no statistical difference between the control treatment in both parameters evaluated.The Cry1Da protein caused a reduction in S. frugiperda larvae average weight, however without mortality.The second bioassay aimed to evaluate the synergism between the Cry (Cry1Ba, Cry1Ca and Cry1Da) and Cyt proteins (Bti strain) against S. frugiperda larvae.Two protein powder concentrations were used (50 mg/mL and 100 mg/mL) for each protein evaluated separately.In protein combinations, ratios were used, 1:1 and 1:3, corresponding to 50 mg/mL:50 mg/mL and 50 mg/mL:150 mg/mL, respectively, of each protein powder.
The bioassay results showed that only treatments containing Cry1Ca protein was significantly different from control in both variables (Table 4).Higher average mortality was obtained when larvae were subjected to Cry1Ca protein isolated.The use of Cry1Ca+Cry1Da (1:1), Cyt+Cry1Ca (1:1) and Cyt+Cry1Ca (1:3) protein combinations showed significant reduction in average mortality of S. frugiperda larvae, compared to Cry1Ca protein isolated used in two concentrations (50 mg/mL and 100 mg/mL) with percentage of 56.25%, 22.92% and 22.92%, respectively.No significant difference was observed in the toxicity of Cyt proteins isolated or in combination with Cry1Ba or Cry1Da proteins to S. frugiperda larvae in both parameters.

Discussion
The literature reports the high toxicity of Cyt proteins to Diptera (Guerchicoff et al., 2001;Wu et al., 2008;Mahalakshmi et al., 2012).The focus of this study was to investigate the distribution of cyt genes in the native Bt isolates and its toxicity to S. frugiperda larvae.The synergism between the Cry and Cyt proteins was also assessed.
The investigation of cyt genes in native Bt isolates shows that the cyt genes are not a common in the Bt strains, compared to Lepidoptera-specific toxins frequency.In this study, among the 500 Bt isolates, six (1.2%) harbored cyt genes.In a previous study performed by Costa et al. (2010) 1073 isolates were evaluated as the presence of dipteran specific genes.The frequency of cyt1Aa, cyt1Ab an cyt2Aa genes was 1.58%, 0,65% and 1,11%, respectively.Wu et al. (2008) observed only three isolates containing cyt genes investigating the presence of cyt genes in 143 isolates from soil samples of China using two pairs of primers (cyt1 and cyt2 gene).The identification of B. thuringiensis delta-endotoxin genes by PCR has proven to be a very useful method for strain characterization isolated worldwide and its use as a preliminary selection step that offers many advantages in terms of rapidity and reproducibility (Porcar & Juárez-Pérez, 2003).Thus, from the rapid identification of potential genes, its toxic activity has to be tested directly by bioassay against a series of insect species.
In present study, there was no correlation between the presence/absence of cyt genes and toxicity of Bt isolate to S. frugiperda, since these genes were present in effective and not effective isolates.All strains harboring cyt1 genes also contained cyt1Aa gene.Identification that the subclass level of the genes presents on Bt strains proves useful due to the high specificity of the δ-endotoxin to insect groups.The Cyt1Aa protein is nontoxic or less toxic for Aedes aegypti, Anopheles stephensi, and Culex pipiens, but synergizes the toxic effect of Cry proteins, as Cry11Aa (Oestergaard et al., 2007).
It is known that the Cyt proteins are found particularly, but not exclusively, among strains that display antidipteran activity (Wu et al., 2008).However, cyt genes were identified in Bt strains toxic to insects of the orders Lepidoptera and Coleoptera (Guerchicoff et al., 1997(Guerchicoff et al., , 2001)).Our results show the need for additional bioassays involving other groups of insects such as S. frugiperda, in order to elucidate the actual role of these genes against this group of insects and its toxicity.
The strains used in this study were previously characterized for cry1 genes (Valicente et al., 2010), and the presence of both cyt and cry genes in the strains used in this study suggest a source for new investigations involving synergism, as its possible effect against S. frugiperda and other lepidopteran insetcs.Previous studies have shown that the Cyt proteins have no toxicity alone, acting on the cell membrane of the insect gut synergistically with Cry proteins, and many authors suggest that the Cyt proteins can interact synergistically with one or more of the Cry proteins thereby increasing toxicity from 5-to 20-fold (Wu & Chang, 1985;Wu et al., 1994).Pérez et al. (2005) observed that synergism between Cyt1Aa and the Cry proteins of Bti, and the activity of the Bti crystals is much higher than that of the isolated proteins.
In our study, we tested the ability of Cyt proteins to increase the toxic activity of Cry proteins toxic and nontoxic to S. frugiperda larvae as a tool for suppression of insect resistance.The bioassay analysis demonstrated that only Cry1Ca protein was effective against S. frugiperda larvae.The effectiveness of Cry1 proteins against lepidopteran larvae have been shown in several studies (Aranda et al., 1996;Bravo et al., 2007;Hernández-Martínez et al., 2008).Aguiar et al. (2006) constructed a truncated version of cry1Ca gene from B. thuringiensis strain 1644 inserted into the baculovirus Ac-MNPV that showed toxicity to second instar S. frugiperda larvae.
Positive synergism was not observed among the proteins evaluated against S. frugiperda larvae.There was no increase in toxicity when Cyt and Cry proteins were used together.These results suggest the use of new protein combinations in future studies.Synergism can lead to a reduction in the amount of Bt required for the use in insect pests control, increase the spectrum of action and reduce the evolution of resistance of these insects (Liu & Tabashnik, 1997).
The response of different species of insects to Bt proteins due to intrinsic differences of each specie, where different toxicity patterns can be observed, even within the same Order.Santos et al. (2009) showed that Cry1A and Cry1Ab proteins caused the highest mortality against S. frugiperda.López-Pazos et al. (2010) reported a low toxicity of the Cry1Ba toxin against Tecia solanivora (Lepidoptera), and it was also observed in this study.However, Hernández-Martínez et al. (2008) obtained a high mortality when S. exigua was tested with Cry1Da protein, however, this toxicity was not observed in our studies.This difference in larval mortality may be related to the use of S. frugiperda.
This study was able to detect cyt genes in Bt isolates tested, contributing to the Bt Collection cataloging.The discovery of cyt genes enables for their use in future studies of insect pests.The development of insect-resistance and cross-resistance to Bt toxins in target insects threatens the future use of these proteins in insect control.Thus, the constant search for effective Bt strains and genes against S. frugiperda and other insect pests is useful for managing resistance to insects and in the control of vectors of human diseases.

Table 3 .
Average mean (± standard error) weights and mortality of S. frugiperda larvae subjected to Bt toxins.(T) -trypsin activated, (*) -trypsin non-activated Means followed by the same letter in columns are not significantly different by Kruskal-Wallis test (p≤ 0.05).

Table 4 .
Means (± standard error) of larval weight and mortality of S. frugiperda subjected to toxins from Bt