Obtainment of pest and essential oils
The experiments were conducted at the Natural Insecticide Chemical Investigation Lab of the Agronomy Department of Universidade Federal Rural de Pernambuco (UFRPE) using a population of Plutella xylostella created years earlier without selection with insecticides. Rearing of P. xylostella was performed following the method described by Torres et al. (2006), with the insects maintained at a temperature of 25 ± 1ºC, with relative humidity of 74 ± 5% and a 12-hour light/darkness photoperiod.
Essential oils from Mentha arvensis, Mentha piperita and Mentha spicata were purchased from the Ferquima company. The commercial insecticides Azamax® and Decis 25 EC were purchased from agricultural product stores in the city of Recife, PE, Brazil.
Gas chromatography-mass spectrometry
Qualitative analysis involving gas chromatography-mass spectrometry (GC-MS) (220-MS IT GC, Varian, Walnut Creek, CA, USA) was performed using a system with a mass selective detector, mass spectrometer in EI 70 eV with a scanning interval of 0.5 s and fragments from 40 to 550 Da. fitted with the same column and temperature program as that for the GC-FID experiments, with the following parameters: carrier gas = helium; flow rate = 1 mL min− 1; split mode (1:30); injected volume = 1 µL of diluted solution (1/100) of oil in n-hexane.
Identification of components
Identification of the components was based on GC-MS retention indices with reference to a homologous series of C8-C40 n-alkanes calculated using the Van der Dool and Kratz equation (Van den Dool and Kratz 1963), computer matching against the mass spectral library of the GC-MS data system (NIST 11 and WILEY 11th), co-injection with authentic standards as well as matching against other published mass spectra (Adams, 2007). Area percentages were obtained from the GC-FID response without the use of an internal standard or correction factors.
Topical bioassay
The effect of topical contact was investigated using the method described by Wei et al. (2015). A micropipette was employed (Kumar et al., 2014) for the application of 0.5 µL of each dose of EO on the prothoracic region of 3rd instar larvae of P. xylostella. The ranges of the doses used were established in preliminary tests. Solutions were obtained by the dilution of the EOs in water containing DMSO (0.5%) and the commercial insecticides in water and Triton-X100® (0.01%).
Nine doses were used for the EOs from M. arvensis and M. piperita, ranging from 2.58 µg larva− 1 to 19.78 µg larva− 1 and from 2.67 µg larva− 1 to 9.79 µg larva− 1, respectively. Eight doses were used for the EOs from M. spicata, ranging from 2.3 µg larva− 1 to 15.18 µg larva− 1. A control treatment was also established with only water and dimethyl sulfoxide (DMSO) (0.5%).
The following doses were tested for the conventional insecticide Decis (Deltametrina) 25 EC: 0.039 µg larva− 1 to 5 µg larva− 1 of deltamethrin diluted in water and Triton-X100® (0.01%). A control treatment was also established with only water and Triton-X100® (0.01%).
Larvae were placed in Petri dishes with disks of collard greens measuring 5 cm in diameter separated into three groups of 10 individuals (30 larvae for each dose tested). The same quantity of individuals was used a second time for the test on another day (repetition in time). In the analysis with the maximum likelihood ratio, the slopes were compared for the determination of the hypotheses of parallelism and equalness (Robertson et al. 2007). When no difference was found, the tests performed on different days were combined, giving a final total of 60 larvae per dose tested for each treatment.
Mortality achieved with the EOs was determined by counts of the number of dead larvae 24 hours after application of the doses. As mortality is slower for the commercial insecticides, the count of the number of dead larvae was performed 48 hours after the application of Decis 25 EC (Jan et al., 2015) and 72 hours after the application of Prêmio® 200 CS (Gong et al. 2014). Larvae were considered dead when not exhibiting any response when prodded with a brush. Dose-mortality data were analyzed using the Probit model (Finney 1971) with the aid of the POLO plus software program (LeOra-Software, 2005) for the determination of LD25, LD50 and LD95 and respective 95% confidence intervals.
Residual contact bioassay
The immersion method was used. Disks of collard greens (5 cm in diameter) were immersed in 20 ml of the different concentrations of the test solution (essential oils and positive controls) and negative control for 10 seconds. After drying for 10 minutes at room temperature, ten 3rd instar larvae of P. xylostella were placed on each leaf disk. Mortality was recorded after 48 hours of exposure. Six repetitions were performed per treatment and repeated in time, corresponding to 120 larvae. To determine the toxicity of the EO solutions, the results were compared to the positive controls (chemical insecticide containing deltamethrin and botanical insecticide containing azadirachtin) as well as the negative control (distilled water + DMSO and dodecylbenzene sulfonic acid).
Preliminary tests had been conducted to define the upper and lower concentrations of 0 and 100%. Concentrations were created with serial dilutions and ranged from 0.5 to 60 µL mL− 1 for the EO from M. arvensis, 5 to 80 µL mL− 1 for the EO from M. spicata, 7 to 100 µL mL− 1 for the EO from M. piperita, 2.5 to 170 µL mL− 1 for Azamax® and 12 to 800 µL mL− 1 for Decis®. Concentration-mortality data were analyzed using the Probit model (Finney 1971) with the aid of the POLO plus software program (LeOra-Software, 2005) for the determination of the LC50 and LC90 with respective 95% confidence intervals.
Phytotoxicity test
After the residual contact test, phytotoxicity tests were conducted with the concentration that caused the 95% mortality using the method adapted from Torres et al. (2006). Collard leaf disks measuring 5 cm in diameter were immersed in the solutions prepared with the EOs diluted in aqueous solutions and the negative control (1.0% DMSO, 0.1% de dodecylbenzene sulfonic acid and distilled water) for 10 seconds and then set to dry at room temperature. After 48 h, the phytotoxicity index (PI) was determined for each leaf disk with the aid of the AFSoft program using the following formula: PI = TA% – SA%, in which TA is the total area of the leaf and SA is the sound (unaffected) area of the leaf. The images were analyzed using the criteria of the phytotoxicity scale proposed by Monchiero et al. (2015): 0.00 to 9.99% = no effect or evident symptoms; 10.00 to 19.99% = negligible effects and symptoms; 20.00 to 29.99% = mild discoloration with no leaf scorch; 30.00 to 39.99% = moderate discoloration with leaf scorch; 40.00 to 99.99% = severe leaf scorch; 100% = severe leaf scorch and dead plant.
Biological parameters
The assessment of the sublethal effects of the EOs on biological variables was based on the method described by Han et al. (2012). The following parameters were determined: survival from larva to adult; larval viability; weight of pupa; emergence rate; fecundity; longevity; egg viability of first generation (F1); larval duration of F1; and larval viability of F1.
The experiment was begun with 100 larvae of P. xylostella, which were weighed and distributed among 10 Petri dishes to ensure equal final weight of the treatments. Each larva was submitted to topical application of the sublethal dose (LD25) of the EOs from M. arvensis (5.1 µg larva− 1), M. piperita (4 µg larva− 1) and M. spicata (5.1 µg larva− 1) previously calculated in the toxicity test. A collard green leaf disk measuring 5 cm in diameter was placed in each Petry dish. The leaf disks were exchanged daily until all larvae reached the pupal phase. The same procedure was performed with the control containing only water and DMSO (0.5%).
Dead larvae in the period of 24 hours after exposure were removed and not counted in the determination of larval survival and viability (as this mortality was expected), thus avoiding the overestimation of the effect. Therefore, counts of dead larvae began 48 hours after exposure, with the assessment of mortality and development.
Larval development was recorded daily until the formation of the pupae, which were transferred to duly identified ELISA plates until the emergence of the adults. For each treatment, 15 pupae 24 hours of age were randomly selected and weighed on an analytical scale. Thus, larval viability, weight of the pupae and the periods of the pulpal phase and emergence were determined.
Data on larval viability and weight of the pupae were submitted to analysis of variance (ANOVA) and differences between means were determined using Tukey’s test, with a P-value < 0.05 considered indicative of statistical significance. Data on the periods of the pupal phase and emergence were submitted to the nonparametric Kruskal–Wallis test. Larval survival data were analyzed using the log-rank survival test. All analyses were performed with the aid of the SAS statistical package (SAS Institute 2008).
Egg viability bioassay
Male and female pairs of P. xylostella from the previous experiment (24–36 hours of age) were randomly separated in plastic cups, forming seven groups of three pairs (total of 21 pairs per treatment). The cups were perforated with a pin to enable the passage of air and duly identified. A moistened filter paper disk carefully cut to occupy the entire space of the bottom of the cup was placed and a collard green disk measuring 2.5 cm in diameter was placed on the filter paper for oviposition of the females. The top was covered with PVC film. An opening was made in the PVC film with the aid of a precision knife and a cotton ball soaked in a 10% honey solution was placed in the opening. The cotton ball was exchanged daily.
Only the eggs of the first laying found on the leaf disk were used for the observation of hatching. When the egg laying had finished, the test proceeded with the exchange of the cotton balls soaked in the honey solution until the death of all adults, with daily counts of the number of eggs laid by the females.
The leaf disks with the first oviposition were placed in Petri dishes and the number of larvae hatched was counted daily. The ratio between the initial quantity of eggs and the number of larvae hatched was used to calculate egg viability. Female fecundity, longevity and egg viability were assessed in this test.
Fecundity data were submitted to analysis of variance (ANOVA) and differences between means were determined using Tukey’s test, with a P-value < 0.05 considered indicative of statistical significance. For the analysis of longevity and egg viability, the data were submitted to the nonparametric Kruskal–Wallis test. All analyses were performed with the aid of the SAS statistical package (SAS Institute 2008).
Effect on F1
For this experiment, 150 newly hatched larvae from the previous bioassay were distributed among three Petri dishes, totaling 50 larvae per dish for each treatment. The larvae were treated with the LD25 of each oil. A 5-cm leaf disk was placed in each dish, followed by the determination of development and mortality. After growing, the larvae were transferred to larger plastic recipients containing sections of collard green leaves and covered with a piece of voile fabric. The test was concluded when the larvae treated with the oils were unable to form pupae four days after the pupae formed in the control group.
Larval viability and duration of the F1 generation were assessed using the nonparametric Kruskal–Wallis test. The analysis was performed with the aid of the SAS statistical package (SAS Institute 2008).
Feeding deterrence and repellent activity bioassays
The feeding deterrence and repellent activity tests were adapted from Akhtar et al. (2012) and involved the determination of the preference for treated or untreated leaves. Collard leaf disk measuring 2.2 cm in diameter were immersed in 20 ml in concentrations corresponding to the LC30 of the solutions (5 µL mL− 1, 2 µL mL− 1 and 3.75 µL mL− 1 for the EOs from M. avensis, M. spicata and M. piperita, respectively) for ten seconds and then set to dry at room temperature. A treated disk was placed at a distance of 2.0 cm from an untreated disk (immersed in distilled water with 1.0% DMSO and 0.1% dodecylbenzene sulfonic acid for 10 s) in a Petri dish 9.0 cm in diameter. A larva deprived of food for 4 hours was placed equidistant (1.0 cm) between the treated and control disk of each Petri disk and left to feed for 24 h. Ten repetitions were performed for each concentration of each treatment, with one Petri dish used for each repetition. After 24 h of exposure, the larvae were removed and areas of consumed leaf on the control and treated disks were determined with the aid of a Licor-3100 leaf area meter, which has high precision and reproducibility, with the resolution ranging from 0.1 to 1 mm2. After determining the consumption of each leaf disk (treated and control), the feeding deterrence index (FDI) was calculated as follows: FDI = 100{(C – T) / (C + T)}, in which C and T (Isman 1993) refer to the areas consumed on the control and treated disks, respectively. Based on the FDI, the EO solutions were classified as phagodeterrent (positive values) or phagostimulant (negative values) (Ferreira et al., 2022).
The repellent effect was recorded after 1, 2, 4, 6, 12 and 24 h of exposure, with the quantification of the number of larvae on the treated and control leaves. The repellence index (RI) was calculated at follows: RI = 2G / (G + P), in which G is the percentage of larvae found on the disk treated with the oils and positive control and P is the percentage of larvae found on the disk treated only with distilled water. RI values range from zero to 2. RI = 1 indicates neutral action; RI > 1 indicates attraction and RI < 1 indicates repellent action (Mazzonetto & Vendramim, 2003). The repellence intensity scale based on the indices proposed by Mazzonetto & Vendramim (2003) was used to classify the degree of repellence of the EOs and positive control to the larvae of P. xylostella. For this experiment, an entirely randomized design was employed with 10 repetitions per treatment.