Evaluating the selectivity of registered fungicides for soybean against Trichogramma pretiosum Riley, 1879(Hymenoptera: Trichogrammatidae)

The aim of this study was to evaluate the effects of fungicides registered for soybean on the parasitoid Trichogramma pretiosum Riley, 1879 (Hymenoptera: Trichogrammatidae). Bioassays wereconducted in laboratory exposing adult insects to dried residues of fungicides, using the methodology proposed by the International Organization for Biological and Integrated Control of Noxious Animals and Plants (IOBC). The experimental design was completely randomized with four replications per treatment. The parameter used for classification of fungicides was based on the reduced parasitism (RP) evidenced by the number of parasitized eggs per female in the control treatment. Based on these results, we found that fungicides are classified in different classes of selectivity to adults of T. pretiosum regarding the tested fungicides. The fungicides cyproconazole, epoxyconazole + kresoxim-methyl, metconazole, thiophanatemethyl (CS), pyraclostrobin, difenoconazole, were classified as harmless (Class 1); carbendazim, carbendazim + thiram, tetraconazole, tetraconazole+ azoxystrobin, epoxyconazol + pyraclostrobin, tebuconazole, prothioconazole + trifloxystrobin, tebuconazole + trifloxystrobin, azoxystrobin, azoxystrobin + cyproconazole, and thiophanate methyl (WG) were slightly harmful (Class 2); flutriafol, thiophanate methyl (WP), epoxyconazole + pyraclostrobin, and cyproconazole+ trifloxystrobin were moderately harmful (Class 3); and Kumulus DF was harmful (Class 4) to the adult parasitoids.


INTRODUCTION
Around the world, Brazil is the second largest producer of soybeans, producing around 86 million tonnes of soybean grains, in the past harvest, with expectations to export about 46 million tonnes (CONAB, 2015).As *Corresponding author.E-mail: maganodeivid@gmail.comAuthor(s)agreethatthisarticleremainpermanentlyopenaccessunderthetermsoftheCreativeCommons Attribution License4.0InternationalLicensesoybean is grown commercially for human consumption and animal feed, it is considered a major source of protein and vegetable oil in the world.In addition, soybean is emerging as an alternative for biodiesel production (USDA, 2013).
The occurrence of pest attacks is a biotic factor that presents a greater interference on soybean yield.Among the many diseases that attack soybean, the Asian soybean rust (Phakopsora pachyrhizi Syd & P. Syd) identified in Brazil in 2001, stands out as a pathogen capable of causing losses ranging from 10 to 90%, mainly damaging early foliage, which prevents full grain development leading to reduced productivity (Roese, 2011).Brazil has an estimated accumulated loss of 15 million tonnes of soybean grains due to the presence of this pathogen in crops (Goulart, 2010), with chemicals as the main controlling strategy to suppress the pathogen (Sosa-Gómez et al., 2003).
In this context, the use of fungicides for soybean crops has intensified, leading in some cases with up to three applications as needed (Alessio, 2008), due to the spread of Asian soybean rust.According to Carneiro et al. (2012), for the periods between 2002 and 2011, Brazilian soybean crops showed an increase use of 1.6 L ha -1 of fungicides.
As of today, there is great difficulty in predicting and monitoring the presence of this pathogen in soybean crops.In an attempt to avoid losses and minimize the spread of this disease, a preventive application of fungicides has become an alternative (Augusti et al., 2014).However, this management, suffering deleterious effects (Sosa-Gómez, 2005), may affect beneficial organisms in the agroecosystem.
Therefore, in order to evaluate the selectivity of fungicides on parasitoid eggs Bueno et al., 2008;Carmo et al., 2009;Carvalho et al., 2012); we used the standardized methodologies of International organization for Biological and Integrated control of noxious Animals and Plants (IOBC) on soybean crops.Thus, it is apparent that poor support of IPM in soybean crops, regarding indications of selective products to natural enemies, is impairing the suppression of pests by biological control or its association with chemical control.
In order to assess the effects of fungicides on beneficial insects (non-targets), the IOBC has developed a standard protocol, using egg parasitoid specie of the genus Trichogramma based on the cosmopolitan distribution, ease of reproduction, high parasitism capacity, and sensitivity to fungicide exposure (Hassan et al., 2000).Based on the facts presented, the objective of this research was to evaluate the selectivity of registered fungicides for soybean on adult parasitoid Trichogramma pretiosum.

MATERIALS AND METHODS
Six bioassays were conducted to evaluate the selectivity in T. pretiosum adults, following the methodology adapted to the species (Giolo et al., 2005;Manzoni et al., 2006a), and in accordance with the guidelines proposed by IOBC (Hassan et al., 2000;Hassan and Abdelgader, 2001).The biological material was used in bioassays comprising the egg parasitoid, which were multiplied in the factious host, Ephestia kuehniella Zeller,1879 (Lepidotera: Pieridae) eggs according to the technique described by Parra (1997).The offspring were maintained in the laboratory, in climate-control chambers of 25±2°C, relative humidity of 70±10%, and photophase of 14 h.Twenty-three fungicides of different chemical groups commonly applied in soybean crops (Table 1) were evaluated.In addition, to these fungicides, a positive control, Lannate BR (Methomyl), which is recognized as a standard toxic pesticide (verify in preliminary tests), is capable of providing 100% of the parasitoid mortality (Magano et al., 2013) and a negative control of distilled water was used.
The parasitoids were then exposed to dry residues of fungicide syrup, which is sprayed on 13×13 cm glass plates, at the maximum recommended rate for use in the field (Agrofit, 2013).Applications were applied using a hand sprayer that provided a syrup deposit of 1.75±0.25 mg cm -2 on each glass plate, measured by a precision balance.Toxicity test were conducted in the laboratory, under the same conditions used for the parasitoid offspring.
To start the experiment we first used emergence tubes (transparentglassbulb120mmlongby20mmindiameter at the baseand7mmattheopening),inwhicheach contained a stationary cardboard circle (STC) (1 cm in diameter) with 250±50 eggs of E. kuehniellapreviously parasitized by T. pretiosum.Approximately 24 h after emergence, the tubes containing the adults of T. pretiosum were in connected cages for 12 h, allowing the entry of insects to be exposed to the fungicides.The feed, which consisted of a honey gelatin mix, of the parasitoid was applied on each STC with eggs viable at 24 (three STCs), 48 (two STCs), and 96 h (one STC).Afterwards, each STC was counted to determine the number of parasitized eggs and the number of females inside the cage, thus, giving the average number of parasitized eggs per female T. pretiosum for each treatment.
The reductions, in the average number of parasitized eggs, depending on the tested products were corrected by Equation 1 (Hassan et al., 2000): Where: RP corresponds to the % reduction in parasitism; Rt is the average value of parasitism for each product; and Rc is the average parasitism observed for the control treatment (negative).Due to the reduction in parasitism, selectivity classes were defined: Class 1 (RP>30%), Class 2 (30<RP<79%), Class 3 (79<RP<99%), and Class 4 (RP>99%).
For data analysis, the completely randomized design was used with four replications; each cage exposed was considered an experimental unit.The data obtained were tested for normality, after subjected to variance analysis, and average significant were compared by the Tukey test (p<0.05)with the software SASM agri (Canteri et al., 2001).
For cyproconazole (Table 2, bioassay III), classified as harmless (Class 1), it showed a reduction of 15.28% in parasitism, even with the studies by Hassan (1998) and Sterk et al. (1999) that utilized fungicide Alto 100 in concentrations of 0.08 and 0.0025%, respectively.The acting mechanism on fungi is due to the demethylation of lanosterol at position 14 or at position 24 of dihidrosterol methylene, which are sterol precursors, one of the basic components in the formation of plasma membrane of organism (Reis et al., 2007).
The commercial product pyraclostrobin was classified harmless to the parasitoid (Class 1) with 17% reduction in parasitism (Table 2, Bioassay I).These results arecorroborating with Manzoni et al. (2006a), that confirmed Class 1 to the same commercial productfor those applied in apple orchards, but is important verify that the dose is different and technology The difenoconazolewas classified as harmless to the parasitoids with a decrease in parasitism by 29.79%.In studies on the selectivity of the fungicide in apple orchards on the species T. pretiosum and Trichogramma atopovirila (Oatman and Platner, 1993) (Hymenoptera: Trichogrammatidae), such results were classified the same, harmless (Class 1) (Manzoni et al., 2007).
Moreover, azoxystrobin, that belongs to the group strobilurin, which is similar to Domark XL, and thatPriori Commercial dosage (L ou kg.ha -1 ) to soybean crop; 2 Concentration (%) of active ingredient in the syrup used in the bioassays; 3 Means followed by identical letters do not differ significantly (p> 0.05) by the Tukey test, 4 RP = Reduction parasitism compared with the negative control (distilled water) used in the bioassays; 5 IOBC classes, 1-Harmless (PR <30%), 2-Slightly harmful (30 <PR <79 %), 3-Moderately harmful (80 <PR <99%),4-Harmful (RP > 99%).* This fungicide was used in two bioassays, due to the ability of the operating system installed in the lab (6Tx4R).Xtra is associated with triazoles (Table 1).Abdelgadder and Hassan (2002) evaluating the product Amistar 250 SC (azoxystrobin) was identified by the same selectivity class (Class 2) for T. cacoeciea adults, using 6.4 µcm -2 of the commercial product.These results show that sometimes the association between two active products not prevent on selectivity class.With fungicides flutriafol, thiophanate methyl (WP), epoxyconazole + pyraclostrobin, and cyproconazole + trifloxystrobin, they were classified as moderately harmful (Class 3) to the egg parasitoids of T. pretiosum (Table 2), for what was used in the tests accounted for 17.39% of the fungicides tested (Figure 1).
The groups of benzimidazole fungicides all have the active ingredient methyl thiophanate, where reductions observed in parasitism ranged from 14.22 to 88.06% (Table 2).The products Support and Cercobim 500 SC (Table 2, Bioassay II) are considered harmless (Class 1) to the parasitoid and both have suspension concentrate (SC) formulationwhich differentiates them from fungicide Support WG (granules dispersed in water), (Table 2, Bioassay VI), in which classified them as slightly harmful (Class 2), a 49.34% decrease in parasitism.For fungicide Metiltiofan (Tab.2, Bioassay III), it presented a wet table powder formulation with a parasitism reduction of 88.06% (Class 3).Since the active ingredient(s), inert ingredient(s), dosage, and other factors change, they can also influence the behavior and activity of the parasitoid, even to tolerable substances (Carmo et al., 2009).
The commercial product Kumulus DF was considered harmful (Class 4) to T. pretiosum, corresponding to 4.35% of the tested fungicide and providing 100% reduction in parasitism (Table 2, 2007) reported the toxic action of the fungicideKumulus DF was due to its interference in many biochemical processes, since sulfur can form chelates with heavy metals that inhibit breathing through the respiratory tract.
Formerly,Kissmann (1998) reported that each company develops its own formulations in ways more convenient for them, and therefore, commercial products with the same type of formulation from two different companies, may differ in their physical characteristics and inert ingredients.This reinforcedthe importance ofinformation of commercial products used in the bioassay when dealing with selectivity studies as proposed by Hassan et al. (2000).
According to the guidelines of IOBC, fungicides classified as non-safe (Class 2,3, and/or 4) should be tested in their immature stage, where the parasitoid finds lower susceptibility to pesticide exposure due to the protection provided by the egg chorion (Orr et al., 1989), which was carried out in soybeans by Bueno et al. (2008) and Carmo et al. (2009).
Toxicity tests for the insects in the laboratory were subjected to maximum exposure to fungicide residues and constituted the first step of the test sequence recommended by IOBC (Hassan, 1998;Hassan et al., 2000;Hassan and Abdelgader, 2001).The fungicides in Class 1 are no longer to be tested as selective for parasitoids, however, the results obtained in Class 2, 3, and 4 (non-safe) are not to be extrapolated to field conditions.Therefore, the fungicides will be required to be tested in a laboratory greenhouse, with immature forms of the parasitoids eggs, larvas, and pupas, to evaluate the persistence of biological activity necessary to determine the impact against natural enemies in the field.

Figure 1 .
Figure 1.Classification of fungicides tested in bioassays selectivity for the soybean crop to Trichogrammapretiosum in accordance with the parameters of IOBC.

Table 1 .
Fungicides evaluated in tests of selectivity for adults of Trichogramma pretiosum using maximum dosage of the commercial product registered for soybean.
1 Dosage field (L ou Kg.ha -1 of commercial product) considering a spray volume of 200l.ha -1  Concentracion (%) active ingredient present in the syrup used in bioassays.

Table 2 .
Average number of females per cage and effect of fungicides on soybeans on the number (± SE) of eggs parasitized by females, reduction (%) in adults parasitism capacity of T. pretiosum and classification of toxicity depending on conditions IOBC laboratory.