The tomato crop has a great economic and social importance to Brazil. In 2011, the Brazilian production was 3,653,017 t in 57,355 ha with a productivity of 63,729 kg/ha (IBGE, 2015). This culture, however, provides a favorable habitat for several species of insects, which can achieve high rates of reproduction and survival, compromising the production (Picanço and Guedes, 1999; Picanço and Marquini, 1999). Among the insects, the tomato fruit borer (TFB) Neoleucinodes elegantalis (Guenée) (Lepidoptera: Crambidae) is a major pest in most tomato-growing regions. It severely infests the fruit, making them unsuitable for consumption and industrial processing (Gravena and Benvenga, 2003), which can lead to 22% production loss (Picanço et al., 2007), and even damage other solanaceous fruits, such as eggplant, scarlet eggplant, jua, jurubeba and pepper (Toledo, 1948; Zucchi et al., 1993).

 

Furthermore, control this pest is difficult due to its behavior. The female lays the eggs, preferably under the sepals of small green fruit (23 mm) and after five days, the larvae hatch, penetrating the fruit be-tween the first and second hour of scotophase, where they remain for about 16 days (Fernández and Salas, 1985; Blackmer et al., 2001). Therefore, N. elegantalis control has been done almost exclusively with the use of synthetic insecticides, applied according to pre-established timetables (Lima et al., 2001).

 

Management practices aiming at the proper use of insecticides are necessary, thus reducing their applica-tion and maximizing effects. Among these practices, the use of repellent insecticide can be considered an important tool for deterring oviposition. Host-plant location is not only related to feeding, but also to oviposition site selection. In Lepidoptera, finding the oviposition site is mediated by the presence of one or more substances that form the specific volatiles or blends of volatiles of the host (França et al., 2009a). The alteration of the plant natural aroma by applying non-specific volatiles can promote rejection. In this sense, some synthetic chemical and botanical insecticides have shown an oviposition deterrent effect on N. elegantalis (Cook et al., 2007), becoming strong allies in the management of this pest.

 

Also, the development of tactics for pest management based on behavior manipulation has been shown as a promising alternative (Witzgall et al., 2008). Chemicals involved in behavioral management, such as feeding stimulants and semiochemicals, can become excellent tools in pest control (Foster and Harris, 1997). Likewise, attractive toxic baits containing feeding stimulants make pest management by behavioral manipulation an effective tool, helping to reduce quantity and increase efficiency of insecticides used in agricultural production systems (Potts, 1999). According to Arruda-Gatti and Ventura (2003), these baits are widely used for the integrated management of insects of agricultural importance, assisting in the decision making of whether or not to control a particular pest. However, the mortality factor is added to the toxic bait, which usually uses a synthetic insecticide. One of the advantages of toxic baits is the benefit to the environment, since they have less influence on some natural enemies, because they are not applied in the total area cultivated (Galli et al., 2004).

 

The objective of this study was to improve strategies that assist in the integrated management of N. elegantalis, directing insecticide applications at different development stages, in order to increase efficiency. Therefore, the direct effects of insecticides on eggs, pupae and pre-pupae were evaluated, as well as insecticide effects on adults (with toxic baits)   and repellency/deterrence    of    N.    elegantalis oviposition. 

 

The experiments were conducted in the Agronomy Department vegetable garden and at the Laboratory of Agricultural Entomology, Federal Rural University of Pernambuco (UFRPE).

 

 

Insect

 

The insects used in the experiments were obtained from a population kept under 25 ± 2°C, 65 ± 10% relative humidity and 12 h photophase, at the Laboratory of Agricultural Entomology. Breeding was adapted from the methodology developed by Prof. Marcelo Coutinho Picanço at the Laboratory of Integrated Pest Management, Federal University of Viçosa. Branches with green fruit (about 3 cm in diameter) and leaves of the tomato cultivar Yoshimatsu L-1 were put in plastic bottles with water and placed inside cages, as a site for N. elegantalis oviposition. The fruit were daily changed, and the eggs transferred to green fruit of organic scarlet eggplant. Each fruit was infested with about four to six eggs, according to size. The scarlet eggplant fruit were put in plastic trays lined with paper towel and kept for about 15 days, until the larvae reached the last instar, when they abandoned the fruit, spending the pupal stage on the paper towel. Pupae were transferred to wooden cages (60 × 60 × 60 cm) covered with organza until adult emergence. Adults were fed with a 10% sucrose solution.

 

 

Survey of insecticides used to control N. elegantalis

 

The questionnaires about insecticide use for N. elegantalis control in tomato crops were given to farmers from the Agreste municipalities of Camocim de São Félix and Bezerros, which are major tomato producers in the state of Pernambuco (IBGE, 2012). The survey was conducted in six properties, three for each municipality, from September 2011 to February 2012.

 

 

Ovicidal effect of synthetic and botanical insecticides

 

Tomato plants of the cultivar Yoshimatsu L-1, with fruit and leaves, were placed in breeding cages for N. elegantalis oviposition. After 48 h, 15 eggs were selected per fruit, the remaining removed and discarded with the help of a brush. Oviposition sites for the selected eggs were marked with a marker pen. The fruit were sprayed with a 5.0 ml solution of each product, with the aid of a Paasche electric micro-atomizer coupled to a compressor, calibrated with nine pounds of pressure (9 mmHg). The insecticides were applied at 10 cm from the spray table, and the control was sprayed with distilled water. The following insecticides were tested: azadirachtin A/B (Azamax^®), emulsifiable neem oil (Natuneem^®) (1 ml L^-1), chlorpyrifos (Lorsban^® 480 BR) (1.5 ml L^-1), methomyl (Lannate^® BR) (1 ml L^-1), beta-cypermethrin (Akito^®) (0.4 ml L^-1), fenpropathrin (Danimen^® 300 EC) (1.5 ml L^-1) and etofenprox (Safety^®) (0.5 ml L^-1). The number of hatched larvae and the average number of entrance holes in the fruit were counted six days after spraying; the effect of insecticides on newly-hatched larvae (Shock Effect) was also observed, because the pest has direct contact with the dry residues. The assessment of average number of holes per fruit was carried out 21 days after application. To calculate the efficiency of the treatments on egg mortality (Ovicidal Efficiency), on preventing fruit damage by newly-hatched larvae (Larvicidal Efficiency: Shock Action) and on the full development of the pest inside the fruit (Larvicidal Efficiency: Physiological Action), the formula proposed by Abbott and adapted by Benvenga was used for comparing the treatments to the control (Abbott, 1925; Benvenga, 2009). A completely randomized design with eight treatments and five replicates, each consisting of 15 eggs was carried out; results  were submitted to analysis of variance and means were compared by the Tukey's test (P ≤ 0.05).

 

In the formula of biological efficiency, indices were established for the control and treatment, with an initial reference on number of eggs in relation to the number of exit holes in fruit at the end of the study (Table 1).

 

 

Effect of synthetic and botanical insecticides on pre-pupae and pupae

 

Pre-pupae and pupae were taken from the N. elegantalis breeding collection and packed in plastic pots containing tomato leaves; they were then sprayed with the same insecticides, following application techniques used in the previous experiment. Development time, mortality and possible morphological changes were observed.

Mortality values were corrected using the Abbot's formula (Abbott, 1925). A completely randomized design with six treatments and five replications, each consisting of 20 pupae or pre-pupae was carried out; results were submitted to analysis of variance and means were compared by the Tukey's test (P ≤ 0.05).

 

 

Effects of synthetic and botanical insecticides on oviposition

 

Tomato branches approximately 30 cm long, containing two leaves and two to three fruits (2 to 3 cm diameter), placed in plastic bottles with water, were sprayed with a 5.0 mL solution of each insecticide by a Paasche electric spray gun coupled to a compressor, calibrated with 12 mmHg. After drying, they were placed inside wooden cages covered with organza (1 × 1 × 1 m), in a hexagonal arrangement with six plants treated or untreated. The following insecticides were tested: methomyl (Lannate^®) (1 ml L^-1), chlorpyrifos (Lorsban^®) (1.5 ml L^-1), lambda-cyhalothrin (Karate Zeon^®) (0.4 ml L^-1), beta-cypermethrin (Akito^®) (0.4 mL L^-1), deltamethrin (Decis^®) (0.4 ml L-1) etofenprox (Safety ^®) (0.5 ml L^-1) chlorantraniliprole   (Premio^®)  (0.02  ml  L^-1),  all  registered  for  the control of N. elegantalis. The following neem-based botanical insecticides were used at a concentration of 1%: azadirachtin A/B (Azamax^® CE - 12000 mg kg^-1 of azadirachtin A and B) and emulsifiable neem oil (Natuneem^® 1500 mg kg^-1). Botanical and synthetic insecticides, as well as the control were tested separately in a randomized complete block design with eight replications. Inside each cage, 30 mated females (three to four days old) were released. A 10% honey solution was used as food source in the cage. The evaluations were made 48 h after infestation by counting the number of eggs per fruit. The mean percentage of egg reduc-tion for each insecticide was calculated, according to a formula:

 

PR = [(NC – NT) / (NC + NT) x 100]

 

where PR = mean percentage of egg reduction, NC = mean number of eggs in the control and NT = mean number of eggs in the treatment (Obeng-Ofori, 1995).

The results were analyzed by the Student's t-test at 5% probability, after homogeneity and normality tests; for lambda-cyhalothrin, chlorpyrifos and beta-cypermethrin, the data were transformed into arcsine √ (x / 100) to meet the assumptions of ANOVA, using the SAS version 8.02 program (SAS Institute, 1999-2001).

 

 

Use of toxic baits and oviposition-repellent insecticides

 

Tomato plants were grown in the Agronomy Department vegetable garden. When fruit reached 2 to 3 cm in diameter, four cages were assembled using PVC pipes (25 mm), with the following dimensions: 1.20 m length × 1.00 m width × 1.30 height m, covered by a voile-type cloth and having side vents with a Velcro closure system. The cages were fixed to the beds with the aid of iron clamps, to provide greater stability. Four plants with about 10 to 15 fruits were placed in each cage. Plants were sprayed with the insecticides lufenuron (Match^® EC), deltamethrin (Decis^®) and indoxacarb (Rumo^®) at commercial concentrations, associated  with 2.5% sucrose (França et al., 2009b). The application was carried out with a manual backpack sprayer with a light-jet cone nozzle. After spraying, 45 female N. elegantalis (aged 48 to 72 h) were released per cage. At 48 h of spraying, the insects were collected and the dead females were counted. Then, bunches were removed and taken to the laboratory of Agricultural Entomology for egg quantification. Each insecticide and the control were tested separately in a completely randomized design with eight replications.

 

Square root data transformation (x + 0.5) was carried out when necessary to meet ANOVA prerequisites. The results were submitted to analysis of variance, and the means were compared by the t-test using the SAS version 8.02 program (SAS Institute, 1999-2001).

 

Survey of insecticides used to control N. elegantalis

 

After analyzing the questionnaires, it has been found that the main insecticides used by tomato growers in the Agreste region of Pernambuco are pyrethroids and organophosphates (Table 2); they act as sodium channel modulators and acetylcholinesterase inhibitors in nerve synapses, respectively, causing the continuous passage of nerve impulses, leading the insect to fatigue and, consequently, death (IRAC-BR, 2012).

 

 

Ovicidal effect of synthetic and botanical insecticides

 

The etofenprox, methomyl and fenpropathrin were primarily responsible for inviability of N. elegantalis eggs, with percentages of 98.6, 91.9 and 45.3%, respectively, differing from the control; etofenprox and methomyl markedly reduced larvae survival and fruit damage. Although fenpropathrin has not been as effective in redu-cing larval survival, it greatly decreased their entrance into the fruit, thus avoiding possible damage. The other insecticides were not considered effective. Regarding larvicidal efficiency, etofenprox, methomyl, fenpropathrin and azadirachtin produced better results, showing a reduction of larvae holes of 98, 90.63, 85.94 and 64.06%, respectively, when compared to the control. The insecticides tested  showed  similar  performance  to  the control, in relation to larvae holes in tomato fruit (Table 3).

 

 

Effect of synthetic and botanical insecticides on pre-pupae and pupae

 

The insecticides did not affect the pre-pupal period in relation to the control, except that methomyl lengthened this period, but with no significant morphological changes (Table 4). The insecticides tested did not affect pupal development, when compared to the control, although, among the insecticides, Fenpropathrin presented a higher pupal development period (Table 5). Etofenprox had an upper percentage of morphological changes in pre-pupae, but none of the insecticides caused significant changes (Table 4). Insecticides caused varied morphological changes in pupae, such as deformed abdomens and two-headed individuals, incomplete biological cycle and adults with atrophied wings (Figure 3). However, none of the insecticides provided a number of significant changes in pupae morphology, compared to the control, with methomyl, azadirachtin and neem emulsifiable oil causing the fewest alterations (Table 5). Etofenprox provided greater corrected mortality in pupae and pre-pupae, 35.89 and 25.30% respectively (Tables 4 and 5).

 

 

 

Effect of synthetic insecticides on oviposition

 

The following insecticides reduced significantly the number of eggs, when compared to the control: chlorpyrifos (t = -10.39; P < 0.0001; GL = 17.8), lambda-cyhalothrin (t = -10.63; P < 0.0001; GL = 20.4), beta- cypermethrin (t = -7.95; P < 0.0001; GL = 23.8), deltamethrin (t = -10.54; P < 0.0001; GL = 34), azadirachtin (t = -5.54; P < 0.0001; GL = 34), methomyl (t= -3.72; P = 0.0007; GL = 34) and etofenprox (t = -3.91; P = 0.0004; GL = 34); reduction percentages were 89.51, 74.34, 45.79, 45.16, 36.86, 25.27 and 23.37%, respectively. Chlorantraniliprole (t = -1.60; P = 0.11; GL = 34) reduced the number of eggs in just 10.30%, and neem emulsifiable oil (t = -1.94; P = 0.06; GL = 18.9) achieved a 15% reduction, thus not differing from the control (Figure 1).

 

 

Use of toxic baits and oviposition-repellent insecticides

 

Deltamethrin (t = -6.41, P < 0.0001; GL = 14) and lufenuron (t = -3.49; P = 0.007; GL = 8.19) reduced significantly the number of eggs when compared to the control, but female mortality was not significant (P> 0.05). Indoxacarb (t = 2.64; P = 0.02; GL = 9.44) caused a significant increase in the number of eggs compared to the control (Figure 2).

 

 

 

Pyrethroid insecticides are the most used by tomato growers, due to their low mammalian toxicity, environ-mental impact and quantity/ha. Organophosphates have a high biological activity, relative instability and a half-life in plants from two to 10 days, but in general, they are very toxic to vertebrates (Santos et al., 2007a; Santos et al., 2007b). However, despite the growth in insecticide use over the years, according to reports from producers, the damage caused by pests has also increased, probably due to selection of resistant insect populations. Thus, insecticide resistance management is an important tool in integrated pest management programs and requires the controlled use of chemicals by varying the concentration and frequency of applications, using insecticides in a rotation system or in a mixture, or applying sequence of chemicals that have different modes of action, for example (Gullan and Cranston, 2007).

 

The effects of chemical insecticides have been investi-gated more frequently to optimize their application, directing them to the most susceptible stage of the target-pest. In relation to ovicidal effect, etofenprox, fenpropathrin and methomyl proved to be a promising alternative for controlling N. elegantalis larvae before they could pene-trate the fruit. Azadirachtin (NeemPro) had the highest slope of a curve  dosage-mortality of N. elegantalis eggs compared to other insecticides, meaning that a small increase in the concentration causes significant increases in mortality; however, deltamethrin (Decis^® 250 CE) was 144.58 times more toxic than the neem-based insecticides tested (França et al., 2009a).

 

The insect development stage in which insecticides are applied can influence their effectiveness, hence the importance of applying them at the right time (Neto e Silva et al., 2011). The insecticide lufenuron (Match^® EC) was more effective on Lobesia botrana Den & Schiff eggs at 0 to 24 h from oviposition (Sáenz-De-Cabezón et al., 2006). However, considering insect development stages, eggs are more tolerant to chemical insecticides. The development of Phthorimaea operculella Zeller eggs be-came unaffected when exposed to various concentrations of thiacloprid (Calypso 480 SC), because larvae hatching was not affected; however, larval survival and adult emergence were reduced (Saour, 2008). 12 insecticides at different concentrations on eggs and larvae of Grapholita lobarzewskii Nowicki, had a larvicidal effect greater than the ovicidal (Charmillot et al., 2007).

 

The effects of botanical and synthetic insecticides on oviposition of some Lepidoptera species have been tested. The emulsifiable neem oil (Natuneem), azadirachtin (NeemPro), deltamethrin (Decis^®) and lambda-cyhalothrin (Karate Zeon^®) reduced the number of N. elegantalis eggs, when compared to the control, in tests of oviposition preference.¹⁵ Aqueous extracts of cinnamon leaves and branches (Melia azedarach L.), powdered tobacco (Nicotiana tabacum L.) and the commercial product Dalneem^® (3000 ppm azadirachtin) in the form of emulsifiable oil at 10% concentration caused a reduction in oviposition of Plutella xylostela L. (Dequech et al., 2008). Methanol extracts from neem seeds and cinnamon at 2, 4, 6, 8 and 10%, reduced oviposition of Earias vittella Fabricius (Gajmer et al., 2002). Aqueous solutions of emulsifiable neem oil reduced the number of eggs laid by Leucoptera coffeella on coffee leaves (Martinez and Martinez, 2003). However, further studies should be conducted in order to explore the deterrent or repellent effects of insecticides on N. elegantalis oviposition.

 

In this study, high natural mortality of pre-pupae and pupae was found in the control treatment. However, the soil is the preferred site for N. elegantalis pupation, which may have influenced their development, because during the breeding of the pest in the laboratory, this problem is commonly observed (Salas et al., 1991).

 

The use of toxic baits has been studied for the control of several urban pests and insect vectors of pathogens (Sackmann and Corley, 2007; Müller and Schlein, 2008); however, studies  with  lepidopteran  pests  are  incipient.

 

Research conducted with adult N. elegantalis found that mortality caused by lufenuron and deltamethrin associated with a 10% honey solution  increased directly with the evaluation periods (0, 0.5, 1, 2, 12 and 24 h after exposure) for males, females and adults (male and females), reaching 100% mortality at 2 h of observation (França et al., 2009b).

 

The deterrent effect of insecticides on oviposition can be an important tool to prevent the establishment of pest insect infestations, as well as the use of toxic baits can increase the efficiency of N. elegantalis control.

 

Chemical control is the most used in N. elegantalis management, but its effectiveness is limited, particularly due to the pest behavior. The neonate larvae penetrate the fruit quickly, thus protecting themselves from insecticides (Eiras and Blackmer, 2003). Hence, the results obtained in this study will bring new perspectives to the management of N. elegantalis in tomato crops in the Agreste region of Pernambuco, by using etofenprox, fenpropathrin and methomyl as ovicidal agents, chlorpyrifos and lambda-cyhalothrin as oviposition deterrents, and lufenuron and deltamethrin in toxic baits. A significant reduction in tomato production losses is expected, and consequently, economic and social benefits for the producers, as well as a better quality product for the consumers.

 

The authors have not declared any conflict of interests.

 

To the Conselho Nacional de Desenvolvimento Científico e Tecnológico, for the scholarship granted to Solange M. França.