Effect Toxic and Behavioral of Annona mucosa (Annonaceae) on the Tomato Leaf Miner

Tuta absoluta (Meyrick) is considered a pest with high destructive potencial and its control depends mainly on successive applications of insecticides. Therefore, new alternatives for the control of the tomato leaf miner using plants with insecticidal potential have been examined. This study was aimed at evaluating the toxic effect of Annona mucosa extract on the developmental stages of T. absoluta. Larval survival bioassay was performed in which newly-hatched caterpillars were inoculated in tomato leaflets sprayed with A. mucosa extract in the LC50 and LC90 treatments and the insecticidal controls chlorfenapyr, methanol, and water. To identify the mode of action of the extract in caterpillars, histological analyzes of the integument and gut were carried out. To evaluate ovicidal activity and oviposition repellency, only the LC50 treatment and controls (water and methanol) were carried out. In the ovicidal bioassay 75 eggs/treatment were used, and for the oviposition repellency, 10 couples/treatment, with 10 replicates. In the larval survival bioassay, a significant difference among survival curves, and the crude extract of A. mucosa significantly reduced the survival of T. absoluta caterpillars. The mode of action of the extract occurred by contact and ingestion, as indicated by changes in the integument and gut. The extract of A. mucosa also interfered in the embryonic development of T. absoluta, with a viability of more than 90% of the eggs. Regarding the behavioral effect, the extract reduced oviposition rates of T. absoluta females. Thus, A. mucosa extract had toxic effects on the different stages of pest development.


Introduction
The tomato leaf miner, Tuta absoluta (Meyrick, 1917) (Lepidoptera: Gelechiidae) is an economically significant pest of tomato (Desneux et al., 2010), as it causes damage at all stages of plant development.
The control of this pest insects is carried out through successive applications of synthetic insecticides. However, this method has been inefficient, due to the high reproductive potential, short generation time and the leaf miner behavior and formation of galleries in fruits (Ortega, 2013;Lietti, Botto, & Alzogaray, 2005;Silva et al., 2011;Roditakis et al., 2015), which can generate losses of up to 100% of the production in tomato crops when there is no effective control (Lopez, 1991). The excessive use of insecticides also has adverse effects on the natural control of pests insect, environmental pollution and damage to human health (Picanço, Bacci, Crespo, Miranda, & Martins, 2007;Aktar, Sengupta, & Chowdhury, 2009).
In contrast to the problems presented, there is a need to search for safer forms to be used in the control of this pest, through the use of bioactive substances in plants which chemical characteristics potential for the synthesis of new products.
In this context, insecticidal plants can be an alternative for playing an important role in the sustainable control of insect pests, since they have combinations of complex molecules that allow biological effects to be the result of a synergism between all of them, unlike synthetic insecticides in which one or a few molecules cause the observed effects (Pino, Sánchez, & Rojas, 2013;Zoubiri & Baaliouamer, 2011).
In view of the results that confirm the insecticidal action of A. mucosa, the aim of this study was to test the toxic effect of this species and the possible changes cells in caterpillars in laboratory bioassays.

Rearing of Tomato Leaf Miner
The rearing was started with caterpillars and pupae collected in commercial tomato plantations in the municipality of Tangara da Serra, Mato Grosso State. After collection, the insects were kept in an air-conditioned room in the laboratory, under controlled conditions (Temperature of 25±1 °C, Relative Humidity of 60±10%, and fotofase of 12 hours). The method of rearing was an adaptation of the methodology proposed by Krechemer (2010), in which pet bottles were used with the bottoms cut, and soon inserted tomato leaves, to Santa Cruz cultivars, to feed the caterpillars. This pet bottles which were covered with voil to avoid their escape. The leaf base was dipped in a plastic tube, which contained water to maintain the turgescence, and every two days healthy tomato leaves were offered for feeding caterpillars.
The pupae were removed and destined for the wooden cage coated with voil. After the emergence of the adults tomato leaflets were inserted into the oviposition, with the petiole dipped in a vial with water to maintain the turgescence of the leaf. Every two days the leaves were replaced, and the eggs transferred to transparent plastic pots until hatching.

Obtaining the Extract of A. mucosa
The fruits of A. mucosa were collected in the periurban area of Tangara da Serra-MT (latitude of 14°37′55″S and longitude 57°28′05″W) and deposited in the Herbarium of the State University of Mato Grosso in Tangara da Serra (Voucher 964). In the entomology laboratory, the fruits were pulped and the seeds dried in an oven with regulated air circulation (40 °C) for 72 hours. Afterwards the seeds were ground in a knife mill to obtain a fine powder. This powder was mixed with methanol 1:3 (extract:solvent) and percolated for 72 hours. After this period the mixture was filtered and the solvent evaporated in a rotary evaporator at a temperature of 40 °C under reduced pressure (Piton, Turchen, Butnariu, & Pereira, 2014). With the remaining residue the process was repeated three times, and at the end the extract was collected in a single container totaling 201,45 g. The extract obtained showed dark coloration and aspects of density and oiliness characterized by its apolarity (lipophilicity). From the crude extract the concentrations used in the conduction of the bioassays were obtained, in which methanol was used as the solvent.
The application of the treatments was done by Arprex spray gun, model 5AD, gravity type, coupled to a direct air compressor. The sprays were made to the point of draining and the leaflets and/or plants remained at room temperature for 15 minutes. After application of the treatments, the petiole of the leaflets were wrapped with humidity cotton to maintain the turgescence.

Concentration-Mortality Bioassay
To determine the lethal concentrations (LC 50 and 90) of the A. mucosa extract on T. absoluta caterpillars, the concentration-mortality test was performed. Newly hatched larvae were submitted to different concentrations of the extract (i.e. 100, 50, 25, 15, 10 and 5 ppm) and to the control (methanol). In this procedure, the leaflets of the middle third of the tomato leaves were sprayed with the concentrations and, subsequently, a newly hatched caterpillar was inoculated by leaflet (n = 20 larvae/treatment). Next were then transferred to 150 ml plastic pots where they remained for 24 hours. After this period the mortality was recorded and the data were used to estimate the lethal concentrations, being used as reference for the other bioassays.

Acute Toxicity of A. mucosa on T. absoluta Caterpillars
In this bioassay was used the lethal concentration (LC50 and LC90) of the A. mucosa extract, the insecticide chlorfenapyr as a positive control and methanol and water were used as negative controls. An adaptation of the methodology of Ferreira, Vendrami and Forin (2012) was used, spraying 1.5 ml of each treatment per leaflet, for a total of 20 replicates. After drying the products, a caterpillars was inoculated with leaflets and packed in transparent plastic bottles of 150 ml. The leaflets were replaced every three days with a new untreated. Evaluations were performed daily until mortality or pupation.

Histological Analysis of the Action of A. mucosa on T. absoluta Caterpillars
To verify the mode of action of the extract, tomato leaflets were sprayed with the LC50, then first instar caterpillars were added, which remained in contact with the leaf for 24 hours after application. The same procedure was performed with the water control.
The histological analyzes followed the protocols of the Laboratory of Insect Tissue Biology of the State University of Londrina (UEL), with whole caterpillars fixed in Karnovsky solution (2.5% glutaraldehyde + 4.0% paraformaldehyde in 0.1 M phosphate buffer and pH 7.2) for 6 hours. After this time the material was washed in 0.1 M sodium phosphate buffer and pH 7.2 (5 minutes), followed by dehydration in 70% ethyl alcohol (5 minutes), ethyl alcohol 90% (20 minutes) and ethyl alcohol 100% (20 minutes). After this procedure, the material was pre-infiltrated in resin + ethyl alcohol solution (1:1) for 4 hours at room temperature, basic resin infiltration + activator at room temperature for 24 hours and inclusion on appropriate polyethylene molds. After the polymerization was complete, the blocks were cut in microtome and the sections (5 μm) were fixed on slides, stained with Hematoxylin and Eosin (HE) and photographed.

Toxicity of A. mucosa on T. absoluta Eggs
To evaluate the ovicidal effect of the A. mucosa extract, tomato leaves were inserted into the adult cage for 24 hours, according to the methodology of Trindade, Marques, Xavier, and Oliveira (2000). After this period, 20 leaflets containing 15 eggs each (repetitions) were separated, totaling 75 eggs/treatment.
On the leaflets were sprayed 1.5 ml of solution corresponding to 21.13 ppm of the A. mucosa extract, and chlorfenapyr, methanol and water controls, totaling four treatments with five replicates each. Evaluations were performed daily, recording of embryonic development and/or egg inviability.

Behavioral Bioassay: Oviposition Preference
The preference of T. absoluta adults for oviposition was performed in bioassays with a chance of choice, adapting to the methodology of Ribeiro et al. (2015).
In this bioassay were used tomato plants with approximately 30 days of age grown in disposable cups of 500 ml. Initially, these plants were sprayed with 6 ml of each solution tested, according to previous bioassay. After drying the solutions applied, the plants were distributed randomly into cages (70 × 35 × 35 cm), four plants per cage in a total of 10 replicates (cages). Then, in each cage 10 adult couples were released and after 48 hours the eggs were counted in each treatment.

Statistical Analysis
The normality and homoscedasticity of the residues were checked with the Shapiro-Wilk and Bartlett tests, respectively. The lethal concentrations LC50 and LC90 were calculated by Probit analysis (Finney, 1971).
Survival of the caterpillars over time was subjected to survival analysis using the Weibull model. Survival curves were then estimated and compared using contrast analysis (P < 0.05).
The number of eggs (i.e., hatchability and/or oviposition preference) were submitted to deviance analysis and generalized linear models (GLM) with adjustment for Poisson error distribution, for which the log link was used. When necessary the treatments were compared by contrast analysis (P < 0.05), always using software R (version 3.1.1) integrated with the stats, survival and car packages (R-CoreTeam, 2017).

Toxic Effects of A. mucosa on T. absoluta Caterpillars
The lethal concentrations estimated for the crude extract of A. mucosa in first instar caterpillars of the tomato leaf miner are presented in Table 1.