EFFECTIVENESS OF TRANSGENIC BT MAIZE TO CONTROL Helicoverpa armigera (LEPIDOPTERA: NOCTUIDAE)

The caterpillar Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae), identified in Brazil in 2013, greatly worries farmers because of its destructive power. The technology of transgenic plants that induces resistance to insects is a great ally for controlling caterpillars. In this study we compared the performance of transgenic Bt maize in the controlling of different larval instars of H. armigera in semi-field seedlings. There were tested the following genotypes: (1) Non-Bt maize Iso-hybrid (Control); (2) Cry1F; (3) Cry1F+ Cry1A.105 + Cry2Ab2; (4) Cry1A.105 + Cry2Ab2; (5) Cry1Ab + Cry1F; (6) Cry1Ab + Vip3Aa20. The experimental design was organized in randomized blocks with eight replications, each consisting of a pot with 5 plants artificially infested with larvae of 2, 3 or 4 instar. At 1, 3, 7, 10 and 14 days after infestation, plants were evaluated for defoliation. In the 3 instar larvae bioassay, there was high predation by birds, but until seven days after infestation all transgenic maize were efficient to the pest. All transgenic Bt maize were effective in the control of tested H. armigera larvae instars.


INTRODUCTION
In recent years, the cotton bollworm Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae), was introduced and spread throughout South America, especially in Brazil, where in recent years caterpillars of the subfamily Heliothinae have produced great damages in several crops, regardless if transgenic or not (Jones et al.,2019).
Although this pest has been considered a quarantine pest A1 in Brazil, it was recently detected in the states of Goiás, Bahia and Mato Grosso, mainly associated with cotton and soybean crops .
It is an extremely polyphagous pest whose larvae have been recorded in more than 60 species of wild and cultivated plants and in more than 60 host families, including Asteraceae, Fabaceae, Malvaceae, Poaceae and Solanaceae. This pest can cause serious damage to different economically important crops, such as cotton, legumes, sorghum, corn, tomatoes, ornamental plants and fruit trees (Reed, 1965;Pawar et al., 1986;Fitt, 1989;Pogue, 2004;Moral Garcia, 2006).
Besides being very polyphagous and voracious, this pest develops very fast, completing its life cycle in four to six weeks, reaching several generations in a year. Insect adults can fly over 1,000 km, giving the species a large dispersibility (Fitt, 1989;Pedgley, 1985;. Moreover, it is a pest with several records of insecticide resistance (Mabbett et al., 1980;Maelzer & Zalucki, 2000).
In the year 2013, maize crop was marked by high H. armigera outbreaks in the main producing zones, perhaps the largest ever recorded since the start of commercialization of genetically modified Bt maize seeds in Brazil. In some regions of the states of Mato Grosso and Mato Grosso do Sul, several maize producers also reported the occurrence of Helicoverpa sp. in Bt maize crops, notably in the transgenic cultivar expressing the Cry1F toxin (Ávila et al., 2013).
The Bt transgenic maize was developed with refined laboratory techniques using a gene from the entomopathogenic bacterium Bacillus thuringiensis (Bt), which was introduced into maize plants, conferring high plant resistance to some species of lepidopteran pests (Armstrong et al. al., 1995). The introduced gene encodes Bt protein expression with insecticidal action, effective in controlling lepidoptera such as Spodoptera frugiperda (J.E. Smith) (Lepidoptera: Noctuidae) (Huang et al., 2002).
However, little is known about the effect of bt toxins to H. armigera, in this study, the objective was to compare the performance of Bt transgenic maize seedlings in the control of H. armigera caterpillars at different instars.

MATERIAL AND METHODS Site locations and insect pest source
Three bioassays were carried out at the Moura Lacerda University Center (Ribeirão Preto -São Paulo state). The H. armigera caterpillars, used in the bioassays for artificial infestations were obtained from laboratory colonies maintained by Bug Agentes Biológicos, (Charqueada -São Paulo state). The caterpillars were kept in artificial diet described by (Hamed & Nadeem, 2008). All insect colonies were reared on artificial diet and maintained in a room with controlled conditions of temperature (25 ± 3°C), relative humidity (60 ± 5%) and photoperiod (14:10 (L: D) h).

Insect infestation and experimental design
The bioassays were conducted from 6 to 23 May and June, 2016. Each biossay was installed in a semi-field, where the corn tested genotypes were sown in 15L plastic pots, fertilized with 25g of NPK 4-14-8 formulation, two days before sowing. Forty-eight vessels were used per assay and covered with voilelike tissue attached to the sides of the vessels by a sewing elastic and supported by a stake about 50 cm behind the vessels. The pots were kept in the open field.
All trials were carried out similarly, with the difference only in the number of caterpillars in plant infestation. Each bioassay compared five different transgenic Bt maize and one Non-Bt maize for the damage caused by 2, 3 and 4 H. armigera larval instars.
The experimental design was organized in randomized blocks, where each of six treatments was repeated eight times, each repetition being a pot with five plants, spaced 10 cm apart. The non-Bt isogenic maize hybrid (iso-hybrid) of the same genetic background was used as control (Table 1). In the first and second bioassays, six days after seedling emergence, a 2 nd and 4 th caterpillar instar, respectively, was placed between the leaves of each plant. In the third bioassay, four days after emergence, 3 rd caterpillar's instar were infested on seedlings. All plants were evaluated for the average defoliation percentage 1, 3, 7, 10 and 14 days after infestation.

Statistical analysis
All data were submitted to analysis of variance (ANOVA). When the F-test of ANOVA indicated a significance of 5% of error probability, the complementary analyzes were carried out by means of the Tukey test at 5% of probability, where the averages were compared. All statistical calculations were performed by the program Statistica for Windows (Statsoft, 1996).

RESULTS AND DISCUSSION
In the bioassay where the 2 nd instar larvae of Helicoverpa armigera were artificially infested in the different maize genotypes, significant differences were found between treatments.
All Bt transgenic maize presented the lower average defoliation percentages compared to non-Bt treatment, every day after infestation. Only at 14 days after infestation did the Cry1F treatment show a greater defoliation than the Cry1F + Cry1A.105 + Cry2Ab2 treatment, both differing statistically from the others Bt genotypes (Figure 1).

Figure 1.
Average defoliation percentage of Bt and non-Bt maize seedlings after artificial infestation with 2 nd instar of Helicoverpa armigera caterpillars. Points values followed by different letters were significantly different by Tukey's test (p≤0.05).
When 3 rd instar caterpillars were infested on maize, all transgenics suffered significantly less defoliation than non-Bt maize, until seven days after infestation. At 10 days, there were no differences between treatments. At 14 days after infestation, only Cry1A.105 + Cry2Ab2 treatment differed from non-Bt treatment, presenting the lowest defoliation average percentage value (Figure 2).

Figure 2.
Average defoliation percentage of Bt and non-Bt maize seedlings after artificial infestation with 3 rd instar of Helicoverpa armigera caterpillars. Points values followed by different letters were significantly different by Tukey's test (p≤0.05).
In this bioassay there was too much predation of caterpillars by birds, which perforated the "voil" protection for capture. This must have interfered with the results, not allowing a safe conclusion about 3 rd instar caterpillar. However, the first evaluations show a tendency of all transgenic maize to be effective to H. armigera.
In the assay of 4 th instar caterpillars, all transgenic maize differed from the non-Bt treatment, which presented the highest defoliation average value on all dates (Figure 3).   Liao et al. (2002), where just Cry1Ab, Cry2Ab, and Vip3A controlled H. armigera larvae. Avilla et al. (2005) also reported that the toxin Cry2Ab2 produced a significant growth inhibition against neonate larvae of H. armígera.