Efficacy and toxicity of insecticides to green peach aphid

Outbreaks of Myzus persicae (Sulzer) (Hemiptera: Aphididae) populations are not rare in Brassica crops in Pernambuco State (Brazil). The aim of the present study was to perform laboratory assessments in seven insecticide formulations against M. persicae. The direct spray and leaf dip methods were used in the experiment. M. persicae mortality rates were subjected to variance and Probit analyses. Based on the results, thiamethoxam + lambdacyhalothrin, thiamethoxam, lambda-cyhalothrin and thiamethoxam + chlorantraniliprole were the most efficient insecticides after 24-hour exposure, since they caused over 95% mortality through the direct spray method. LC50s ranged from 0.10 to 9.1 mg a.i./L in chlorfenapyr and chlorantraniliprole, respectively, in the leaf dip method. All insecticides were effective against M. persicae, except for chlorantraniliprole. Chlorfenapyr, lambda-cyhalothrin, pymetrozine and thiamethoxam + lambda-cyhalothrin were the most toxic to this aphid. RESUMO: Surtos populacionais de Myzus persicae (Sulzer) (Hemiptera: Aphididae) não são raros em cultivos de Brassica no Estado de Pernambuco (Brasil). O objetivo desse estudo foi realizar avaliação laboratorial de sete formulações de inseticidas contra M. persicae. Para tanto, foram utilizados o método de pulverização direta e o método de imersão de folha. A mortalidade de M. persicae foi submetida à análise de variância e análise de Probit. Os resultados mostraram que após 24 horas de exposição, tiametoxam + lambda-cialotrina, tiametoxam, lambda-cialotrina e tiametoxam + clorantraniliprole foram mais eficientes, causando mortalidade acima de 95% pelo método de pulverização direta. Utilizando o método de imersão de folha, as CL50s foram de 0,10 até 9,1 mg i.a/L para clorfenapir e clorantraniliprole, respectivamente. Com exceção ao clorantraniliprole, todos os inseticidas foram eficazes para M. persicae sendo que clorfenapir, lambda-cyhalothrin, pimetrozina e tiametoxam + lambda-cialotrina foram mais tóxicos para o afídio.


Introduction
Green peach or peach-potato aphid, Myzus persicae (Sulzer) (Hemiptera: Aphididae), is among groups of pest-insects that have successfully exploited the agricultural environment and acknowledged worldwide as one of the most harming agricultural pests (Margaritopoulos et al., 2009). M. persicae is a sucking insect able to transmit more than 100 viruses and to use over 400 plants from 50 different families (Schoonhoven et al., 2005). Outbreaks of M. persicae populations in brassica crops often account for losses faced by producers from Pernambuco State (Northeastern Brazil). M. persicae, Brevicoryne brassicae (L.) and Lipaphis erysimi (Kalt.) are the most harming pest aphids in Brassica crops, since they cause up to 70%-80% yield reduction under favorable conditions (Khan et al., 2015) High density M. persicae populations attacked Brassica rapa var. pekinensis (known in Brazil as "couve-chinesa") at the experimental field from the Universidade Federal de Pernambuco (UFRPE). The M. persicae population in this site does not have history of exposure to any insecticide type; therefore, it is possible assuming that this population was susceptible to new insecticides such as chlorantraniliprole, chlorfenapyr and pymetrozine in Brassica rapa var. Pekinensis crops at UFRPE. There is no register of any insecticide to control M. persicae in B. rapa var. Pekinensis (Brasil, 2003) Chemical control is the main method to mitigate damages caused by M. persicae; but the insecticides has put strong selection pressure on M. persicae populations leading to evolution of insecticide-resistance (Bass et al., 2014). Nowadays, the Arthropod Pesticide Resistance Database (APRD) makes available 469 cases (80 compounds) of insecticide resistance in M. persicae; this species was preceded by Tetranychus urticae Koch (Acari: Tetranychidae) (517 cases) and Plutella xylostella L. (Lepidoptera: Plutellidae) (866 cases) and in the rank of resistance to insecticides.
Studies about the efficacy and toxicity of novel insecticides are important to help managing pest aphids, mainly in crops with few registers pesticides to chemical-control. Chlorantraniliprole, pymetrozine and chlorfenapyr are part of novel groups of insecticides, namely: anthranilic diamides, pyridine azomethine and pyrrole analog, respectively. Chlorantraniliprole activates insect ryanodine receptors and lead to fast muscle dysfunction and paralysis (Cordova et al., 2006). This process takes place due to depletion in calcium reserves, which are essential for muscle contraction (Cordova et al., 2006).
The exposure to pymetrozine causes immediate and irreversible feeding cessation, but not because of the deterrent action. Actually, its mode of action results from stylet penetration blockage (Kayser et al., 1994), but this compounds' target mechanism has not yet been identified. Chlorfenapyr is a proinsecticide and its oxidative removal from the N-ethoxymethyl group by mixed function oxidases leads it to its toxic form, which uncouples oxidative phosphorylation in mitochondria. Consequently, ATP production is disrupted and followed by energy loss, which leads to cell disjunction and, eventually, to the death of exposed organisms (Raghavendra et al., 2011). Two more compounds were included in the present study, namely: thiamethoxam (neonicotinoid) and lambda-cyhalothrin (pyrethroid). Thiamethoxam (neonicotinoid) is a nicotinic receptor agonists of acetylcholine and lambda-cyhalothrin (pyrethroid) is a sodium channel modulator. Thiamethoxam bonds account for the persistent activation of nicotinic receptors and both compounds thiamethoxam (neonicotinoid) and lambda-cyhalothrin (pyrethroid) cause super excitation of the nervous system.

Materials and Methods
The M. persicae colony was collected at the Brassica rapa var. pekinensis crop in the production site from the Universidade Federal Rural de Pernambuco (UFRPE) in July, 2015 (Pernambuco State -Brazil). The site has no history of exposure to any insecticide. Leaves with natural M. persicae infestation were taken to the Laboratory of Interaction Insect-Toxic (LIIT).

Direct spray method
B. rapa var. Pekinensis leaves with natural M. persicae infestation were cut into 5.0-cm diameter discs for the direct spray method. In total, 20 insects were kept on each disc, which was subsequently sprayed with a certain insecticide with the aid of a mini plastic sprayer. Each treatment had 12 replicates, which were individualized in Petri dishes (5.0cm diameter) containing 1 ml of 2% water + agar solution. The same procedure was applied to the control (Table 1). The experiment followed a completely randomized design and was performed at 25 ± 0.2 ºC and 65% relative humidity under 12-h photoperiod.

Leaf dip method
Preliminary tests were made to establish an "all or none" response for all insecticides in the leaf dip method. Insecticides were diluted in water added with 0.01% Triton X-100 (Lima Neto et al., 2017). At least six concentrations were used in the ultimate bioassays applied to each insecticide -each treatment had three replicates. Discs (5.0-cm diameter) of B. rapa var. pekinensis leaves were immersed into the insecticide or into the control solution [water+ Triton X-100 (0.01%)] for 10s; subsequently, they were left to dry at room temperature for 1-2h (Lima Neto et al., 2017). After drying, the leaf discs were taken to Petri dishes coated with filter papers and filled with 1 ml of 2% water + agar (2%) solution. At least ten insects were placed on each treated leaf disc. The discs were kept at 25 ± 0.2 ºC and 65 ± 5% relative humidity (RH) under 12-h photoperiod. Bioassays were repeated twice. Mortality was assessed after 72-h exposure to chlorantraniliprole, pymetrozine and thiamethoxam + chlorantraniliprole because these insecticides need longer to cause death than the ones that only needed 4-h exposure to be lethal. When control mortality was >10% , the bioassys were discarded.
Tukey's test was applied to compare means recording 95% confidence level to the mortality rates recorded for different insecticides. Data were transformed through x + 0.5 .
Mortality data applied to dose-response curves were subjected to Probit analysis (Finney, 1971) at P > 0.05 in the Polo-Plus 2.0 software (LeOra Software, 2005) after they were corrected for the natural mortality recorded for the control (Abbott, 1925). Toxicity ratios were calculated through the "lethal ratio test"; results were significant when confidence interval (CI) of 95% did not include the value one (Robertson et al., 2007). The insecticide presenting the lowest LC 50 was the reference for comparisons to other insecticides.
The present study provided toxicological evidence of some novel insecticides, such as chlorfenapyr, chlorantraniliprole and pymetrozine, to M. persicae. In addition, a mix of insecticides (thiamethoxam + lambda-cyhalothrin and thiamethoxam + chlorantraniliprole) seen as important tools to' insect pest management were also tested. Based on results of the direct spray method, thiamethoxam-based formulations were efficient in leading to over 95% mortality before 72-h exposure. LC 50s indicated that thiamethoxam and chlorantraniliprole were less toxic when they were used in separate. M. persicae population susceptibility was linked to life history of individuals' exposure to different insecticides in the environment. Accordingly, Nidhi et al. (2013) estimated LC 50 equal to 48.75 mg thiamethoxam/L, whereas Lima Neto et al. (2017) showed LC 50s ranging from 5.00 mg to 9.00 mg thiamethoxam/L against M. persicae. Their study evidenced that the M. persicae population was not exposed to neonicotinoid . Lima Neto et al. (2017) estimated LC 50 ranging from 5.5 mg to 10.00 mg chlorantraniliprole/L; but, based on the present study, chlorantraniliprole is not a promising insecticide against M. persicae. On the other hand, the efficacy of thiamethoxam against M. persicae was confirmed. i./l of water and confidence intervals to 95%; e Toxicity ratio: ratio of estimates of the LC 50s of the most toxic insecticides and the insecticide least toxic (Robertson et al., 2007) and confidence intervals to 95%.
*Toxicity ratio is significant if confidence interval does not comprise the value 1.0 (Robertson et al., 2007).
Unexpectedly, chlorfenapyr was the most toxic insecticide against M. persicae. Based on the results, it was efficient in the direct spray method. Chlorfenapyr was the most toxic insecticide against M. persicae in comparison to results in the studies by Nidhi et al. (2013) (12.98 mg chlorfenapyr/L). Field studies should be carried out in order to better assess chlorfenapyr against M. persicae, because this compound is mainly used as acaricide and lepidoptericide. Pymetrozine was efficient after 72-h exposure, it was the third most toxic insecticide (LC 50 5.1 mg/L); but it was the least toxic to M. persicae in comparison to results in the studies by Sadeghi et al. (2009), who estimated LC 50 equal to 0.01 mg/L for Acyrthosiphon pisum (Harris).
Concentrations tested for all insecticides in the direct spray method overcame LC 90s estimated for the leaf dip method. Assumingly, the direct spray method caused additional intoxication in all aphid individuals due to both insecticide intake and contact. This outcome can explain the efficacy/ toxicity differences between methods. The current results are relevant because LC 90 is a parameter connected to control efficacy. According to the Brazilian Ministry of Agriculture, Livestock and Supply (MAPA), for an insecticide to be considered effective in the field, the dose recommended by the manufacturer must cause at least 80% mortality to the insect pest population under field conditions. All insecticides were efficient to control M. persicae under laboratory conditions, except for chlorantraniliprole. Susceptible M. persicae populations are rare because of the pressure caused by the indiscriminate use of insecticides. Therefore, the present study is important to resistance-management programs based on susceptibility monitoring/baseline and on subsequent evaluations. The herein used insecticides should be sprayed in rotation (between different groups) in order to keep populations' susceptibility and production sustainability. Developing resistance to insecticide is a matter of time, but such process should be delayed as much as possible, in pest populations.

Conclusions
All insecticides were effective against M. persicae, except for chlorantraniliprole. Chlorfenapyr, lambda-cyhalothrin, pymetrozine and thiamethoxam + lambda-cyhalothrin were the most toxic insecticides to this aphid. Thiamethoxam + chlorantraniliprole and thiamethoxam recorded intermediate toxicity to it.