Selectivity of Insecticides Against Adult Trichogramma pretiosum Riley ( Hymenoptera : Trichogrammatidae ) on Cassava

In this study, the effects of various chemical and biological insecticides on adults of the parasitoid species Trichogramma pretiosum (Riley) were examined in cassava. The chemical insecticides zeta-cypermethrin, lufenuron + profenofós, imidacloprid, thiamethoxam, and teflubenzuron and the biological agents Bacillus thuringiensis and Baculovirus erinnyis were evaluated. For each treatment group, the effects were evaluated using T. pretiosum mortality rates, longevity of females, the number of eggs parasitized by the F0 generation, and number of emergent F1 parasitoids. The insecticides were classified using the International Organization for Biological Control categories. The tested insecticides had little effect on longevity and survival of adults, but all affected parasitism (F0) and were moderately harmful to the parasitoid. In relation to the emergent (F1) generation, the pesticides were either harmless or slightly harmful, except for zeta-cypermethrin which was classified as harmful with a 100% reduction in parasitoid emergence. The results indicate that the tested insecticides were moderately harmful to T. pretiosum in the laboratory; future studies will examine the effects of the insecticides under semi-field and field conditions to confirm their toxicity.


Introduction
Cassava is an important food plant that is widely cultivated in Africa, Asia, Oceania and Latin Americat (CONAB, 2017).In Brazil, the largest producer in Latin America (CONAB, 2017), with crop showed an increase of 3.7% between 2016 and 2017.An estimated 21 million tons of cassava roots have been planted, corresponding to more than 1.4 million hectares of cultivated land (IBGE, 2017).Unfortunately, crop productivity can be reduced by the action of pest insects, such as the moth Erinnyis ello, which is a major pest that can cause defoliation of cassava plants (Pratissoli, Zanuncio, Barros, & Oliveira, 2002).In general, attempts to control pests generally make use of chemical or biological insecticides.
Biological control of caterpillars is carried out by application of Baculovirus erinnyis or Bacillus thuringiensis, which are selective for some beneficial species of entomofauna on cassava plants.Chemical control mainly involves application of pyrethroid and organophosphate insecticides (MAPA, 2016).However, the use of broad-spectrum chemical insecticides can also directly or indirectly affect beneficial insects such as parasitoids and predators.These organisms are of great importance in agroecosystems for maintenance of populations of insect pests below the level of economic damage (Carvalho, Parra, & Baptista, 2001).
Among the natural enemies of pest species present on cassava plants, more than 30 natural control agents have been described for E. ello, such as parasitoids, predators, and pathogens (bacteria, fungi, and viruses) that act on eggs, caterpillars, and pupae.Parasitoids of the genus Trichogramma attack eggs and therefore have the advantage of controlling the pest before the occurrence of damage to the crop (Botelho, 1997;Parra, Botelho, Côrrea, & Bento, 2002).Oliveira, Gomez, Rohden, Arce, and Duarte (2010) recorded natural parasitism of E. ello eggs by Trichogramma marandobai, Trichogramma manicobai and Trichogramma pretiosum in cassava cultivars in Mato Grosso do Sul.Integrated pest management using chemical and biological methods is feasible in cassava as in other crop species; one of the important aspects of an integrated management program is that the different control methods employed do not adversely interact to reduce their individual effectiveness (Oliveira, Antigo, Carvalho, & Glaeser, 2013).
With regard to cassava, few studies have investigated the selectivity of chemical insecticides on T. pretiosum parasitoids.Therefore, the objective of this study was to evaluate the effects of various insecticides on T. pretiosum in cassava crops.
2.2 Side-effects on the Maternal Generation (= F 0 ) Parasitoids 20 T. pretiosum females, up to 24 hours old, obtained from the creation of laboratory were placed into individual glass tubes (8.5 cm high × 2.5 cm diameter) and fed with a honey droplet deposited on the inner wall of the tube.Tubes were sealed with PVC plastic film perforated by an entomological pin for aeration.20 blue cards (1.0 × 1.0 cm), each containing 30 eggs of the alternate host Anagasta kuehniella (Zeller) (Lepidoptera: Pyralidae) adhered using 10% diluted arabic gum, were immersed for five seconds in the various insecticides described in section 2.1.The cards were then placed on a paper towel at room temperature for one hour to remove excess solution and placed into each tube containing a single T. pretiosum female for 24 hours; this protocol is based on those proposed by Brugger et al. (2010), andVianna et al. (2009).
The culture tubes were maintained in an air-conditioned room at 25±2 °C, 60±10% RH, and 12-hour photophase.Each treatment consisted of 20 replicates, each replicate consisting of a card carrying 30 A. kuehniella eggs that had potentially been parasitized.A randomized experimental design was used.Mortality of the adults from potentially parasitized eggs was evaluated; the longevity of females after parasitism, and the parasitism rate of the F 0 generation (number of eggs parasitized by females over 24 hours) were also evaluated.In order to verify possible effects on individuals of the F 1 generation, the rate of emergence [(Number of eggs with parasitoid exit orifice/Total number of parasitized eggs) × 100] of parasitoids from the treated eggs was evaluated.

Statistical Analysis and Classification of Insecticides
Datasets were first tested for a normal distribution (Kolmogorov test) and homoscedasticity (Bartlett test) and transformed if required; the data were then subjected to ANOVA, and significant differences between means were identified using the Tukey test (α < 0.05).Statistical analyses were performed using Statgraphics ® Centurion version XVI (Statistical Graphics Corp. 1994-2000).

Effect of Insecticides on Mortality of Parasitoids in the Maternal (F 0 ) Generation
The mortality rates of T. pretiosum (F 0 ) females that came into contact with A. kuehniella eggs treated with all products were less than 20% and thus were considered innocuous according to IOBC toxicity classes (Table 1).Sterk et al. (1999).
The effect of an insecticide on beneficial organisms varies according to the mode of action, the dose, the crop treated, and the natural enemy species studied.Neurotoxic insecticides generally have poor selectivity with regard to parasitoids and other natural enemies (Parra et al., 2002).However, the insecticides used in the present study were largely innocuous in terms of adult mortality.
The neonicotinoid thiamethoxam was reported to cause 100% mortality in adult Trichogramma galloi (Zucchi 1988) and was therefore assessed as harmful (Oliveira et al., 2013).The difference in rates of mortality found here and by Oliveira et al. (2013) may be related to the dose used, as the latter study used the maximum dose in a sugar cane crop (1000 g/ha), while the present study used the maximum dose suggested by the manufacturer (150 g/ha).
Zeta-cypermethrin is a broad spectrum pyrethroid and, in the present study, induced 20% mortality in adult T. pretiosum under laboratory conditions (Table 1).Therefore, this insecticide was classified as innocuous.This finding is in agreement with Souza (2011), who did not observe negative effects using a compound of the same chemical group on T. pretiosum.
The effect of B. erinnyis on the survival of adult T. pretiosum has not previously been studied.This insecticide was classified as innocuous to adults of this parasitoid (Table 1).Baculovirus anticarsia has been previously been used on adult T. pretiosum and found to be innocuous (Amaro, Bueno, Pomari-Fernandes, & Neves, 2015).
T. pretiosum females that came in contact with insecticides through 24 hour exposure to treated eggs showed an average longevity of approximately nine days (Table 1); no significant differences were present among the treatments (p > 0.05).

Effect of Insecticide Treatment on the Number of Parasitized Eggs (F 0 )
The number of eggs parasitized by T. pretiosum (F 0 ) females was evaluated for each treatment.The results indicated that all of the tested insecticides reduced the rate of parasitism.Therefore, the treatments were classified as moderately harmful to the parasitoid (Table 2).Carvalho, Moura, Bueno, (2006) classified teflubenzuron as innocuous with respect to parasitism by T. pretiosum.This difference in classification may be related to the use of a higher insecticide dose in the present study.
Baculovirus erinnyis was moderately harmful to parasitism by T. pretiosum (F 0 ).One possible mechanism for these could be related to a repellent activity of the virus.In a test of free choice in the laboratory, it was found that females avoided contact with virus-treated eggs (unpublished data).
In contrast to other published results that classified B. thuringiensis as innocuous (Amaro et al., 2015;Vianna et al., 2009;Filho, Botton, Grutzmacher, Giolo, & Manzoni, 2006), our findings show a moderately harmful effect of B. thuringiensis (Table 2).These differences may be related to the source of the biological agent, which may vary with regard to the number of viable spores.

Effect of Insecticides on Emergence of the First (F 1 ) Generation
Zeta-cypermethrin reduced the rate of emergence of T. pretiosum (F 1 ) from treated A. kuehniella eggs and was classified as harmful.Teflubenzuron and lufenuron + profenophos were classified according table the IOBC as slightly harmful (Table 3).The remaining insecticides were all classified as innocuous (Table 3).Sterk et al. (1999).
The reduced rate of emergence of parasitoids (F 1 ) from eggs treated with teflubenzuron or lufenuron + profenophos may be related to the mode of action of benzoylurea compounds, which act as growth regulators and inhibit synthesis of chitin.Thus, these compounds may affect formation of the parasitoid since the larvae hatch within their host and initiate feeding on a substrate contaminated with insecticide (Cônsoli, Botelho, & Parra, 2001;Oliveira et al., 2013).Oliveira et al. (2013) reported a 98.38% reduction in T. galloi emergence from D. saccharalis eggs treated with triflumuron, another benzoylurea compound, and classified the insecticide as moderately harmful.However, Carvalho et al. (2003) did not find a significant effect of lufenuron (benzoylurea insecticide) on emergence of T. pretiosum in treated A. kuehniella eggs and classified the insecticide as innocuous.
The results obtained for thiamethoxan were consistent with those of Moura, Carvalho, and Rigitano (2005), who classified this insecticide as innocuous for the emergence of T. pretiosum from treated eggs at different immature phases of the parasitoid.However, our results differ from those of Oliveira et al. (2013) for T. galloi, who classified the insecticide as slightly harmful.These differences may be related to the doses used: Oliveira et al.
(2013) used the maximum recommended dose for sugar cane, which was approximately seven-fold higher than the dose used in the present research.The two studies also investigated different Trichogramma and host species.
In the present study, imidacloprid was classified as innocuous in contrast to the studies by Brunner et al. (2001) and Carvalho et al. (2003), who found this insecticide to be toxic to T. platneri and T. pretiosum under laboratory conditions.The different outcomes might be related to the use of different species of parasitoids.
Zeta-cypermethrin was classified as harmful as it caused a 100% reduction in the emergence of T. pretiosum from A. kuehniella eggs.Similar results were obtained by Bastos et al. (2006), who observed a reduction in the emergence of T. pretiosum in the hosts A. kuehniella and S. cerealella.This reduction demonstrates the toxicity the insecticide that would be acts directly on the central nervous system of the insects, resulting in high mortality rates.
No previous investigations have been reported on the effect of B. erinnyis on emergence of egg parasitoids.In the present study, we classified this biological insecticide as innocuous.
The biological insecticide B. thuringiensis was found innocuous for the emergence of parasitoids.Similarly, Amaro et al. (2015) reported that this insecticide was harmless for the emergence of T. pretiosum pupae, while Pratissoli et al. (2006) classified the insecticide as innocuous to T. pratissolii after feeding the parasitoids with honey containing B. thuringiensis.
In general, the different insecticides evaluated were innocuous to the survival of T. pretiosum.However, many showed some effect on parasitism.There is a possibility of parasitoid repellency by the tested products or by transovarian action.
If T. pretiosum is included in integrated pest management programs for the biological control of E. ello cassava crops, our results suggest that the releases be carried out 24 hours before spraying with insecticides so that use of insecticides does not affect the development of T. pretiosum in the field.

Table 1 .
Mortality (±SD), percentage of reduction (PR) and toxicological class for adults of Trichogramma pretiosum after contact in Anagasta kuehniella eggs treated with insecticides used in cassava crops.Means followed by the same lowercase letter in the column do not differ by test Tukey (p > 0.05) Note.* Products in phase of registration by Joint Normative Instruction nº1, of June 16, 2014, dose used supplied by the manufacturer.¹ Average percentage reduction in survival of T. pretiosum.² Toxicity classes recommended by

Table 2 .
Number (±SD) of parasitized eggs and percent reduction (PR) of Trichogramma pretiosum in Anagasta kuehniella eggs treated with insecticides used in cassava crops.Means followed by the same lowercase letter in the column do not differ by test Tukey (p > 0.05) Note.* Products in phase of registration by Joint Normative Instruction nº1, of June 16, 2014, dose used supplied by the manufacturer.¹ Average percentage reduction in survival of T. pretiosum.² Toxicity index recommended by Sterk et al. (1999).

Table 3 .
Emergence (± SD) and percentage reduction (PR) the Trichogramma pretiosum (F 1 ) (Hymenoptera: Trichogrammatidae) in Anagasta kuehniella (Lepidoptera: Pyralidae) eggs treated with insecticides used in cassava crops.Means followed by the same lowercase letter in the column do not differ by test Tukey (p > 0.05) Note.*Product in the stage of registration by Joint Normative Instruction nº1, of June 16, 2014, dose used supplied by the manufacturer.¹ Average percentage reduction in survival of T. pretiosum.² Toxicity index recommended by