Selectivity of mycoinsecticides and a pyrethroid to the egg parasitoid Cleruchoides noackae (Hymenoptera: Mymaridae)

Plants of the genus Eucalyptus, cultivated in many countries, have great importance for the world economy. In Brazil, this culture occupies a total of 5.7 million hectares, but native and exotic insect pests can reduce its productivity. Thaumastocoris peregrinus Carpintero & Dellapé (Hemiptera: Thaumastocoridae), an exotic Australian pest, damages Eucalyptus plants. Biological control using the egg parasitoid Cleruchoides noackae Lin & Huber (Hymenoptera: Mymaridae), Heteroptera predators and entomopathogenic fungi, such as Beauveria bassiana and Metarhizium anisopliae, have potential for managing T. peregrinus. Chemical insecticides, including bifenthrin and acetamiprid + bifenthrin, also control this insect. The compatibility of chemical and biological control methods favors integrated pest management. The objective of this study was to evaluate the selectivity of commercial products based on B. bassiana, M. anisopliae and the chemical bifenthrin on the parasitoid C. noackae and its parasitism on T. peregrinus eggs. The selectivity test followed the standards recommended by the International Organization for Biological Control (IOBC). Beauveria bassiana has selectivity to parasitism as well as viability, but was slightly harmful to C. noackae adults; M. anisopliae was innocuous to adults and to the viability of the offspring of this parasitoid, but it reduced the parasitism rate; and bifenthrin did not show selectivity to this parasitoid.

The area of commercially planted forests in the world increased from 167.5 to 277.9 million hectares from 1990 to 2015 1 . Brazil presently has 5.7 million hectares of Eucalyptus plantations with 24%, 17% and 15% of them in the states of Minas Gerais, São Paulo and Mato Grosso do Sul, respectively. The wood from these plantations is mainly destined for the pulp industry, with a production of 21 million tons in 2018 2, 3 .
Insect pests of Australian origin detected in planted forests during the last three decades on a global scale may reduce Eucalyptus productivity 4 . The bronze bug, Thaumastocoris peregrinus Carpintero & Dellapé (Hemiptera: Thaumastocoridae), was first detected in Brazil in 2008 in the states of São Paulo and Rio Grande do Sul, and has since dispersed to other Eucalyptus-producing states 5 . This insect develops and produces fertile offspring on most Eucalyptus plantations in Brazil 6 . Thaumastocoris peregrinus perforates and causes silvering, tanning, drying and defoliation from Eucalyptus plants 7 .
Biological control is the most widely-used method for managing T. peregrinus 8 . This method includes the egg parasitoid Cleruchoides noackae Lin & Huber (Hymenoptera: Mymaridae), imported from Australia 8,9 , the predators Atopozelus opsimus Elkins (Hemiptera: Reduviidae) 10 and Supputius cincticeps Stäl (Heteroptera: Pentatomidae) 11,12 and entomopathogenic fungi 13,14 . Beauveria bassiana and Metarhizium anisopliae, registered Scientific RepoRtS | (2020) 10:14617 | https://doi.org/10.1038/s41598-020-71151-2 www.nature.com/scientificreports/ commercially 15 and considered to offer reduced risks, are the most studied entomopathogenic fungi 16,17 . The chemical insecticides bifenthrin and acetamiprid + bifenthrin are also used to control T. peregrinus in Eucalyptus plantations 15 . Natural enemies are important in pest control in planted forests, justifying the search for compatible microbial and chemical products 18 . The mycoinsecticides and chemical insecticides must have selectivity to the pest natural enemies 19,20 in order to maintain the effectiveness of the combined use of these methods.
The objective of this study was to evaluate the effect of commercial products based on B. bassiana and M. anisopliae and of the chemical insecticide bifenthrin on the egg parasitoid C. noackae and on its parasitism on T. peregrinus eggs.

Results
Mortality of Cleruchoides noackae adults. The mortality of C. noackae adults was higher with bifenthrin after the first and tenth hours of exposure to this chemical, with 67% and 90.6%, respectively. Biological products based on B. bassiana and M. anisopliae caused mortality of 40.8% and 22.6%, respectively, of C. noackae adults after 10 h of exposure, higher than the control, distilled and autoclaved water, which was 18% (Fig. 1). Bifenthrin was moderately harmful, B. bassiana was mildly harmful, and M. anisopliae was innocuous to C. noackae adults, presenting a reduction in the beneficial ability of the parasitoid [%E = 100 − (average for each insecticide/average for the percentage in the control treatment) × 100] of 90.60, 40.80 and 22.60, respectively, according to IOBC classification (Table 1)   www.nature.com/scientificreports/ Viability of C. noackae in eggs parasitized and treated after one and 10 days of exposure to the insecticides. C. noackae viability in T. peregrinus eggs treated with the insecticides after one day of parasitism differed between treatments (ANOVA; F = 4.301, P = 0.0126), with a lower value for the bifenthrin than in the control and with the product B. bassiana having a value similar to that of M. anisopliae (Table 3). Bifenthrin reduced the viability of this natural enemy on parasitized eggs after 10 days (ANOVA; F = 6.460, P = 0.0018). This chemical was slightly harmful (%E = 30.11 and 34.08), and M. anisopliae (%E = 13.02 and 1.68) and B. bassiana (%E = 4.75 and 4.60) innocuous, after one and 10 days, respectively, for the parasitoid C. noackae (Table 3).

Discussion
The entomopathogenic fungi tested were chosen according to the knowledge and use of these microorganisms in biological control, as well as their reduced environmental impact 21 . Selectivity tests show the low impact of products on non-target organisms, and allow the recommendation of combinations of mycoinsecticides and chemical insecticides to manage harmful organisms. The higher mortality of C. noackae adults produced by bifenthrin shows that this chemical is moderately harmful, like most pyrethroids that keep the sodium channels of the neuron membranes open and reach the insect peripheral and central nervous systems 22 . At the cellular level, these compounds stimulate the neurons to produce repetitive discharges, leading to membrane depolarization and synaptic disorders 23 . Cyanide pyrethroids cause hypersensitivity, choreoathetosis, tremors, paralysis and insect mortality 24,25 . The classification of B. bassiana as slightly harmful to C. noackae adults may be related to the production of secondary metabolites, such as organic acids involved in the infection process and linear and cyclic peptidic toxins such as beauvericin from the mycelium of this fungus 26,27 . The lack of Metarhizium anisopliae toxicity to C. noackae adults is similar to that reported with this fungus on Cotesia flavipes Cameron (Hymenoptera: Braconidae) 28 and Trichogramma galloi Zucchi (Hymenoptera: Trichogrammatidae) 29 . Metarhizium anisopliae is important for biological control because some isolates may be highly specific and others infect a wide host range 30 . The mortality of C. noackae adults at shorter intervals is due to the reduced longevity of this parasitoid: 0.8 to 1.6 days when they were not fed 31 and 3.5 days when they were fed with undiluted honey 32 , evidencing the importance of the evaluations during the first day of this natural enemy life.
Beauveria bassiana and M. anisopliae, with high selectivity and low impact through contact with adults and in the parasitism and offspring of C. noackae, respectively, have potential for joint use with this parasitoid in pest management programs. However, these mycoinsecticides should be applied 3 days after releasing this parasitoid, avoiding contact with their adults at the time of parasitism. Bifenthrin, the first chemical insecticide registered to control T. peregrinus 15 , cannot be used with the C. noackae egg parasitoid to manage this pest.
Beauveria bassiana-based mycoinsecticides have selectivity to parasitism and viability and are slightly harmful to C. noackae adults; Metarhizium anisopliae was innocuous to adults and to the viability of this natural enemy offspring, but it reduced C. noackae parasitism on T. peregrinus; bifenthrin did not show selectivity in all bioassays.

Rearing Thaumastocoris peregrinus. Thaumastocoris peregrinus adults were collected on two-year-old
Eucalyptus grandis × E. urophylla plants at the FCA/UNESP and taken to the laboratory for mass rearing 46 .
Branches of the hybrid Eucalyptus urophylla var. platyphylla (clone 433) were collected from two-year-old trees, and arranged in bouquets with their bases in 250-ml Erlenmeyer flasks with water on a rectangular plastic tray (40 cm long × 35 cm wide × 8 cm high) to mass rear T. peregrinus in the laboratory. These bouquets were changed every three or four days depending on the need and leaf conditions. On the day of the exchange, the oldest and driest bouquets were placed next to new ones to facilitate the insect migration to the latter 46 .
noackae, were obtained from the LCBPF. Paper towel strips (1.5 cm wide × 15.0 cm long) were arranged in the upper portion of the leaves of the T. peregrinus breeding bouquets to obtain their eggs. Cleruchoides noackae were multiplied with T. peregrinus eggs with two to three days of age in transparent polystyrene bottles (7.5 cm high × 3.0 cm diameter).
Newly emerged C. noackae adults were transferred with a brush to another transparent polystyrene flask with paper towel strips with two-to three-day-old eggs obtained from the T. peregrinus rearing. Cleruchoides noackae were fed with 50% honey solution in filter paper strips (7.0 cm high × 1.5 cm wide) 47 .
Selectivity test. The selectivity of mycoinsecticides and the bifenthrin-based insecticide to C. noackae adults and to their parasitism was evaluated in four treatments (Table 4), according to the protocol of the IOBC with the standard test cage 48 . www.nature.com/scientificreports/ One ml per replication of the biological and chemical products was applied in a Potter Tower on the surface of the cages designed according to the standard described by the IOBC and on parasitized or non-parasitized T. peregrinus eggs. Three bioassays were performed.
The first test evaluated the indirect action in the mortality of the parasitoid, exposed by contact to the biological and chemical products, using 100 new individuals per treatment, in five replications of 20 individuals each (per cage). The control had only distilled and autoclaved water. The parasitoids were released in the treated cages and their mortality, after contact with the treated surface, was evaluated.
The second bioassay evaluated the direct action on parasitism and the viability of C. noackae on T. peregrinus eggs treated with the insecticides, with five replications per treatment and 10 eggs, each one offered to a pair of the parasitoids per cage. Paper towel strips with one-day-old T. peregrinus eggs were treated with the insecticides, dried at room temperature and offered to each C. noackae couple for 24 h.
The third bioassay evaluated the C. noackae viability with the products. One-day-old T. peregrinus eggs, exposed to each C. noackae couple for 24 h, were treated in a Potter Tower after one and 10 days post-parasitism, respectively. Five replications with 10 eggs each were used per treatment ( Table 4) and age after parasitism (one and 10 days), totaling 400 eggs. Data evaluation. Mortality, parasitism and viability (%) of C. noackae were evaluated. Mortality of this parasitoid was evaluated after the first hour of contact with the insecticides and then every three hours until completing 10 h, due to its reduced longevity. Parasitism and viability of C. noackae were evaluated after 13 days of parasitism (parasitoid cycle), considering emerged and retained parasitoids and non-parasitized and infertile eggs. The percentage of reduction in parasitoid beneficial ability was calculated for each of the analyzed variables (survival, parasitism and viability; %E) with the equation: %E = [100 − (average for each insecticide/average for the percentage in the control treatment) × 100] to classify the products according to IOBC standards: class 1innocuous (E < 30%); class 2-slightly deleterious (30 ≤ E ≤ 79%); class 3-moderately harmful (80 ≤ E ≤ 99%); and class 4-harmful (E > 99%) 48 .
The design was completely randomized, the data submitted to variance analysis and the means compared by the Tukey test at 5% probability using the R Studio software.