Selectivity and sublethal effect of botanical extracts on adults of Trichogramma pretiosum Riley (Hymenoptera: Trichogrammatidae)

The objective was to assess the selectivity and sublethal effect of botanical extracts on adults of Trichogramma pretiosum Riley (Hymenoptera: Trichogrammatidae). Plant species used: Rosmarinus officinalis L., Mikania glomerata Spreng., Varronia curassavica Jacq., Chenopodium ambrosioides L., Vermonia polyanthes Spreng., Plectranthus amboinicus Spreng., Tetradenia riparia Codd., Artemisia absinthium L., Cymbopogon citratus Stapf. Plant material were reared in laboratory and leaves collected during the morning were used to prepare the extracts as following: 100 g of leaves from each plant were macerated and immersed in 300 mL of triple-extracted methyl alcohol and evaporated on rotavapor. Then, 250 mg of crude extract was diluted in 1000 mL of water. The experimental design for selectivity was done in randomized blocks with five repetitions, following a standardized protocol of the International Organization for Biological Control (IOBC) and accordingly classified. In sublethal effect, the design was completely randomized with twenty repetitions, where the eggs exposed to parasitism were sprayed with the treatments. In both experiments the control was twofold (water and chlorpyrifos). Regarding selectivity, the botanical extracts of M. glomerata, C. ambrosioides, V. polyanthes, P. amboinicus, A. absinthium and C. citratus were classified as slightly harmful (class 2), chlorpyrifos as moderately harmful (class 3). In the evaluation of sublethal effects, the botanical extracts of C. ambrosioides and C. citratus initially presented reduced number of parasitized eggs, reduction in total parasitized eggs and female longevity. It can be concluded that C. citratus botanical extract has small selectivity and causes sublethal effects on T. pretiosum adults.

he evaluation of selective products is done by using natural enemies of agricultural pests, with members of the parasitoid wasp genus Trichogramma being indicated as subjects by the International Organization for Biological Control (IOBC) standardized protocols. Trichogramma are parasite wasps of lepidopteran eggs species which are considered to be agricultural pests. The genus has cosmopolite distribution, with several species having their hosts associated, a trait that favors massive breeding and hence also favors their usage in biological control (Parra et al. 2015).
Previously, selectivity traits used to classify insecticides were based on short-term lethality parameters over natural enemies of pests. However, even though some of these insecticides may not quickly kill the subject, they may lead to sublethal effects such as: lowering of the longevity and ability to prey/parasitize, changing in different levels of biological or behavioral characters, lowering of the proper biological development, morphological deformations and lowering of reproductive traits (fecundity/fertility/fitness), among other problems (Croft 1990;Desneux et al. 2007;Vogt et al. 2007). Botanical extracts were largely used in agricultural and medical pest control as insecticides between 1930 and 1950, but were later substituted by synthetic molecules (MaChaDo et al. 2007;Martinez 2011). From these compounds, several chemical substances synthetized in the secondary metabolism of plants can be obtained and be used as insecticides, allowing the expansion of agroecological products (silVa et al. 2017). However, one of the limitations of botanical insecticides usage is the lack of knowledge about the effect of these compounds on the benefic fauna of agroecosystems (VenDraMiM & Castiglioni 2000), especially in regard to pest's natural enemies. The majority of works dealing with pest management by using botanical products highlight that the usage of these substances must be compatible with other pest control tactics, especially biological control. However, some researches have been demonstrating the existence of variations in the effects caused by botanical products on natural enemies (ribeiro et al. 2009).
Several natural or botanical products are commonly related as causers of negative effects in Trichogramma spp. One of the most famous examples is the Neem (Azadirachta indica A. Juss), which is sprayed in host-eggs, resulting in significant reduction of the parasitism (oliVeira et al. 2003;gonçalVes-gerVásio & VenDraMiM 2004;hohMann et al. 2010;. The Azidrachtin is the main compound among the secondary compounds. However, several other triterpenoids are found, such the meliantriol, limonene, oderatone, as well as more than 150 other compounds already isolated, which may present synergy with insecticides, enhancing its action (siDiqui et al. 2003;Morgan 2009).
However, despite the benefits above-mentioned, these plants have the disadvantage of lacking data related to phytotoxicity, persistency and their effects in benefic organisms (Costa et al. 2004;aguiar-Menezes 2005). Plant effects can be measured by checking if there is lowering in feeding, inhibition of chitin production, growth interruption, changings in ecdysis, reduction in fecundity, morphological deformations, increasing in death rate and repellency and toxicity to men and domestic animals (MeDina et al. 2003;Mossini & KeMMelMeier 2005;nDione et al. 2007;Morgan 2009).
One of the biggest challenges in the usage of plants as bioinsecticides is the elaboration of formulas which are able to keep their ingredients active through longer periods of time in the target body (nerio et al. 2009), aside from getting information about biologic control agents and the effects of plant-based products.
Thus, it is demonstrated the need for development of new studies about the effect of natural products in agents of biological control, as new products are still arriving at the market. In this context, the aim of this work was to investigate the selectivity and sublethal effects of botanical extracts in adults of Trichogramma pretiosum Riley.

MATERIAL AND METHODS
The experiment was conducted at the Plant Protection Department of the Faculty of Agronomic Sciences, Universidade Estadual Paulista "Júlio de Mesquita Filho" (FCA/UNESP), Botucatu Campus, São Paulo, Brazil, under controlled conditions (12 h of photophase, 25 °C e 70% of relative humidity).

Plant collect and botanical extract preparation.
The extracts were obtained at the Laboratory of Medicinal Plants of the Horticulture Department of the Faculty of Agronomic Sciences, Universidade Estadual Paulista Júlio de Mesquita Filho" (FCA/UNESP). Leaves were collected during morning, reared under organic management at Medicinal Garden and then taken to the laboratory. A hundred grams of fresh material of each plant were macerated in methanol (immersed in 300 mL of methyl alcohol for 24, 48 and 72 h, in triple extraction, resulting in 900 mL of methanolic extract). The solvent was evaporated in a rotatory evaporator (Quimis) and the crude extract diluted in the proportion of 1000 mL to 250 mg of distilled water, then conserved in a glass recipient of amber coloration.
Rearing maintenance of T. pretiosum. Adults of T. pretiosum were kept in a four liters recipient, under controlled conditions of temperature (25 ± 2 °C), relative humidity (70 ± 10%) and photophase (14 h), fed with crude honey chunks; fresh eggs of Ephestia kuehniella Zeller (Lepidoptera: Pyralidae), glued to a double-sided tape which was glued over a coated paper, were available during 24 h to stimulate parasitism. Past this time period, the cards were individualized in new recipients until the larvae emerged from the eggs, starting a new rearing cycle.
Treatment selection. The botanical extracts of Rosmarinus officinalis L., Mikania glomerata Spreng., Varronia curassavica Jacq., Chenopodium ambrosioides L., Vermonia polyanthes Spreng., Plectranthus amboinicus Spreng., Tetradenia riparia Codd., Artemisia absinthium L., Cymbopogon citratus Stapf used in the bioassays were chosen based on previous results obtained by testing in second instar nymphs of Spodoptera frugiperda (Smith) (Lepidoptera: Noctuidae). The tests consisted of pulverizing the extracts on the nymphs by using a Potter spray tower® (by contact) and by incorporating the extracts in the artificial diet (by ingestion). Thus, the criteria of the plant species selection were the insecticide effect of the plants, by contact and ingestion, in the above-mentioned pest insect, in the concentration of 1000 mL of crude extract to 250 mL of distilled water, as previously described. The active ingredient chlorpyrifos, an insecticide of the organophosphate chemical group, was used as positive control (as recommended by IOBC) and distilled water as negative control.

Application of botanical extracts in selectivity and sublethal bioassays.
The experimental design was made in randomized blocks with eleven treatments and five repetitions for selectivity and entirely randomized with 20 repetitions for sublethal effect. The Potter spray tower® was pre-calibrated to pulverize 250 L ha -¹, with pressure varying between 5-7 psi and deposition of 1.75 ± 0.25 mg cm 2 , measured in analytical balance, following the standard procedure suggested by the IOBC. For the pulverization, a volume of 1.13 mL of each treatment was separated in Eppendorfs®.
Cages for selectivity assessment. The cages used followed the IOBC suggestion, made with aluminum frames juxtaposed with screws. Three sides of the cage have entirely tightened orifices, assuring the imprisonment of the subject and allowing for gas exchange (hassan et al. 2000). On the fourth side there are two openings: one made to introduce the egg-cards and another used to release the parasitoids in the cages. The top and bottom sides are sealed with glass plates and interconnected by a hose-system connected to a pressure pump used to remove air by suction.
Selectivity. The top and bottom glass-plates of the cages were built to expose the subjects to the pulverizations. Prior to the exposure, cards with parasitized eggs (± 600 eggs) were individualized in Duran® tubes to obtain a mean of 200 females per cages (hassan et al. 2000). The individuals of 24 hours-old were exposed to the treatments. After the colonization of the interior of the cages, cards of 6.25 cm² with fresh eggs of E. kuehniella (± 1.200 eggs) were made available to stimulate parasitism, during the period of 24, 48 and 72 h. After 96 hours, the cages were disassembled and the eggcards collected and individualized by date and treatment in plastic bags to posterior analyses. The parasitism (P) was analyzed with the equation P = number of parasitized eggs / number of offered eggs x 100. The viability and parasitism reduction in relation to the negative control was assessed by the efficiency equation: E (%) = (1 -(mean percentage of the treatment / mean percentage of the treatment control) x 100), where: E (%) is the percentage reduction of the benefic capacity of the parasitoid (sterK et al. 1999;Manzoni et al. 2006). Thus, the results were classified according to mean reduction, as suggested by the IOBC standards: class 1innocuous (E < 30%); class 2 -slightly harmful (30 ≤ E ≤ 79%); class 3 -moderately harmful (80 ≤ E ≤ 99%); class 4 -harmful (E > 99%) (hassan 1992).
Sublethal effects. Females of T. pretiosum were individualized in a Duran® tube based on antenna structure (querino & zuCChi 2011) and fed with a drop of honey. During their life-time, the females were daily offered cards of 0.25 cm² with fresh eggs (± 180 eggs) of E. kuehniella, pulverized with the treatments, to observe the sublethal effects of those females which did not suffer lethal effect. The egg-cards were labeled and stored to posterior analyses. The total number of parasitized eggs and longevity of parasitoids was analyzed, although the sublethal effects of the treatments may affect the insect's physiology and interfere in the analyzed parameters (Desneux et al. 2007).
Statistical analyses. The results were analyzed by an Analyses of variance (ANOVA) and the means compared with a Tukey's test with 5% of significance with the software SISVAR 5.6. The extract of C. citratus caused alteration of the parasitism (16.18), as well as reduction (76.44%), numbers close to those of the positive control (chlorpyrifos), which is considered to be moderately harmful (class 3) (Table 1).
On the second D.A.E., the extracts of R. officinalis, M. glomerata, C. ambrosioides and V. polyanthes induced the reduction of the parasitism and were classified as slightly harmful (class 2), as well as the extract of C. citratus, with a parasitism proportion of 6.74 and reduction of 78.99%, numbers which are also close to those of chlorpyrifos, with 1.83 and 94.3% respectively (Table 2).
On the third D.A.E., the extracts of M. glomerata, C. ambrosioides, V. polyanthes, C. citratus and chlorpyrifos were classified as slightly harmful (class 2) ( Table 3). The remaining treatments did not cause changings or reductions of the parasitism.
Plants of C. citratus have therapeutic properties, such as bactericide and fungicide, and are also popularly used against lices (lorenzi & Matos 2008). These features are possible due to the presence of terpenes, like the citral (a mix of neral and geranial isomers), myrcene, citronellol and nerol (CarriConDe et al. 1996). Tests with three different concentrations (1%, 5% and 10%) of lemon grass essential oil in Thyrinteina arnobia Stoll (Lepidoptera: Geometridae) resulted in insecticide effect, causing 100% of mortality in subjects submitted to concentrations of 5% and 10% (soares et al. 2011). This lethal effect in T. arnobia may explain the low amount of selectivity found in T. pretiosum. The chlorpyrifos is commonly included

e-ISSN 1983-0572
Selectivity and sublethal effect of botanical extracts on adults of… Gladenucci et al. (2020) in tests of selectivity as positive control of toxicity to allow for comparisons with other insecticides, due to its negative effects in Trichogramma species (stefanello Júnior et al. 2008;aMaro et al. 2015).

Sublethal effects of botanical extracts on adults of
T. pretiosum. Sublethal effects may affect growth and development, by interfering in cellular metabolism (aguiar-Menezes 2005). The reduction of egg-laying rates or inhibition are important effects of botanical extracts over insect's reproduction (Costa et al. 2004). Analyzing the total number of parasitized eggs, in the treatments of chlorpyrifos, C. ambrosioides and C. citratus, the number of parasitized eggs was 0; 2.66 and 2.08 respectively (Table 4). In treatment with chlorpyrifos, it is possible that the high toxicity of the organophosphate compounds may have interfered in the results. souza (2011) reported a mortality of half of the population of T. pretiosum after one hour of exposition to eggs treated with chlorpyrifos.
In the C. ambrosioides treatment, it may have happened due to the presence of ascaridole, a secondary compound. Experiments made with essential oil made from leaves and stem of C. ambrosioides, in the weevil Sitophilus zeamais Motschulsky (Coleoptera: Curculionidae), resulted in a 100% mortality rate, 24 h after exposition to a dose of 500 μL oil/L (JaraMillo et al. 2012). The treatment C. citratus, which has citral among its secondary compounds, may have affected the amount of parasitized eggs. Experiments made with essential oils of Pelargonium graveolens and Lippia alba, which possess geraniol, linalool, carvone and citral among their major compounds, resulted in significant mortality in larvae of S. frugiperda in comparison with the natural bioinsecticide Azamax ® and with the negative control (niCulau et al. 2013).
In agroecosystems, the longevity of egg-parasitoids may be influenced by several factors, such as feeding, temperature, relative humidity, as well as nutritional conditions of the hostegg (PaCheCo & Corrêa-ferreira 1998).
In the present work, the parasitoid was exposed to botanical extracts in controlled laboratory conditions and it was observed alterations in female longevity, with the treatment C. citratus (2.6 days) being the closest one to the positive control chlorpyrifos (1.05 days). In treatment with distilled water, the duration was 10.42 days (Table 4), similar results to those of oliVeira et al (2005), in which the longevity had a variation between 11 and 13 days in laboratory conditions.
The effects of botanical extracts over the parasitoid's longevity may be related to the act of parasitize. When the female parasitoid starts touching the eggs pulverized with treatments, then she starts absorbing the toxic substances (by contact) and thus causing negative effects on the longevity. For a substance to cause toxic effects in an organism biology, it has to penetrate the cells and spread, starting to interfere in physiological or biochemical processes, causing impact in longevity, growth, development, reproduction and behavior (DelPueCh & Meyet 2003).

e-ISSN 1983-0572
Volume 13, 2020 -www.entomobrasilis.org EntomoBrasilis 13: e871 The satisfactory parasitism rate in laboratory rearing is around 80%. In the present experiment, the distilled water reached 80% of parasitism by the 10 th day, as well as the botanical extracts of C. ambrosioides and V. polyanthes. The treatments with A. absinthium. R. oficinais, V. curassavica and C. citratus reached the 80% rate sooner than expected, when comparing with distilled water. There was an extrapolation with the extract of M. glomerate, which reached 80% of parasitism only by the 15 th day, contrary of P. ambrosioides, which reached the 80% rate by the sixth day, both treatments differing from the remaining (Figure 1).
Regarding the daily parasitism rate, the treatment with C. citratus differed from all the others, with approximately 15 parasitized eggs in the beginning, with several peaks of oviposition and days when there was no parasitism. The remaining treatments began with a mean of 25 parasitized eggs, then gradually diminishing over time. It was not possible to calculate the daily and accumulated parasitism rates of the positive control, chlorpyrifos, as all females died in the first 24 h (Figure 1).
According to Moreira et al. (2005), there is a misconception that botanical extracts with insecticide activity have a smaller toxic potential and are safer than the synthetical products, when there are botanical products in the market that are toxic to fishes, pollinators, mammals, benefic insects, natural enemies, etc. In this context, researches to find botanical products that are efficient in controlling the pest, but selective to natural enemies, are especially important.
Given the results found, we conclude that the botanical extracts analyzed are innocuous (class 1) or slightly harmful (class 2) to T. pretiosum. However, it is demonstrated that C. citratus has small selectivity and sublethal effect on adults of T. pretiosum.
The treatments classified as class 1 may be recommended to integrated pest management programs, aiming to unite alternative and biological methods. The remaining treatments need more studies about their selectivity, which should be done in semi-field and field conditions, to confirm their toxic potential and investigate their impacts.