Chemicals
The fipronil used in this study was obtained from the commercial product “Frontline® Spot On Gatti”, purchased from Boehringer Ingelheim Animal Health Italia S.p.A. (Milan), with the excipients as non-reactive compounds. The pyrethrins were obtained from a commercial product, “Piretro Compo®Bio”, an insecticide for organic farming, purchased from COMPO GmbH (Münster), with the excipients as non-reactive compounds.
To simulate a real-world scenario where wastewater reuse and sewage sludge are common agricultural practices, plants were irrigated with the environmentally relevant concentrations of 100 ng/L (coded as F100) of fipronil, and with a higher concentration of 1,000 ng/L (F1000) of fipronil. The F100 and F1000 concentrations were obtained by serial dilution of fipronil stock solutions in distilled water.
For the analysis of fipronil in the zucchini plants, analytical standards of fipronil and the internal standard fipronil-13C215N2 were purchased from Cymit Quimica S.L. (Spain). Stock solutions were prepared in acetonitrile for fipronil and in methanol for fipronil-13C215N2, with the fipronil stock solution used to prepare calibration curves through serial dilution.
Zucchini plant cultivation
Zucchini (Cucurbita pepo L.) seedlings of the cultivar ‘San Pasquale’ were used in this study. The seeds were kept for germination on wet cotton disks in sterile Petri dishes at 20 ± 2°C. After one week, the germinated seeds were sown in pots (14 cm diameter) filled with synthetic soil, consisting of a mixture of peat, perlite, sand and clay (see Trotta et al. 2024b for details). The pots were grown under controlled conditions in a climate chamber at 22 ± 1°C, 60 ± 5% relative humidity (RH) and an 18:6 Light:Dark (L:D) photoperiod until the end of the experiments. The lights used for plant growth consisted of 20-W, 130-lm/W white LED tubes (6500 K) coupled with 36-W, 100-lm/W full-spectrum LED tubes. The seedlings were subsequently watered with water plus 10 mL of 0.3% NPK (7.5-3-6 + Fe and microelements) nutrient solution PIANTE VERDI (Compo®, Ravenna, Italia). These conditions were maintained throughout the entire duration of the experiments, which lasted 35 days.
Insect rearing
The Aphis gossypii strain used in this experiment was collected from a zucchini field near Potenza, Italy (40°34′N, 15°45E) in August 2022 (Forlano et al. 2022) and reared in the laboratory on zucchini plants. No chemicals were used on this strain from the time of collection until the start of the experiment. Aphids were reared at room temperature under a 16:8 L:D photoperiod for two years. Aphis gossypii can show a wide range of colour variation, from yellow to dark green, as a response to developmental temperature, crowding conditions and host plant; it can also produce winged morphs in response to nutritional factors and crowding. Dark green apterous parthenogenetic females of A. gossypii were used for the following experiments.
To eliminate maternal and grandmaternal effects from aphids, 100 adult virginoparae females were isolated from the aphid culture, placed on a fresh host plant and allowed to reproduce for 24 h in a Binder KBF climatic chamber at 22 ± 1°C, 75 ± 5% RH and an 18:6 L:D photoperiod. The adults were then removed, and the cohort of new-born nymphs was reared on the plant for 7 days, corresponding to the beginning of the adult stage. This procedure was carried out for 2 generations and repeated independently for 3 groups of aphids, resulting in individuals from 3 independent replicates.
Experimental design of zucchini irrigation with fipronil
In the control treatment (Co), zucchini plants were watered with tap water. The two experimental treatments involved adding 100 ng/L of fipronil (F100) and 1,000 ng/L of fipronil (F1000) to the irrigation water. These concentrations were selected to reflect environmentally relevant levels (Sadaria et al. 2017, 2019; Zhang et al. 2023; Kim and Kim 2024), with the higher concentration used to explore potential degradation pathways. Experimental watering began one week after transplanting the seedlings into pots. At each watering, a single plant was watered with 150 mL of the respective water solution. To ensure consistent soil moisture between 30% and 90% of water-holding capacity, nine watering sessions were conducted over the 35-day period, with each plant receiving a total of 1,350 mL of experimental water by the end of the experiment. This resulted in a cumulative exposure of 135 ng of fipronil for plants in the F100 treatment and 1,350 ng for those in the F1000 treatment, while control plants received no fipronil.
The experiments were conducted in three separate periods, spaced 7 days apart, producing three independent replicates of 7 plants each, for a total of 21 plants per treatment group.
Aphid survival and fecundity
For each treatment, two adult female aphids of the same age (8–9 days old) were independently placed on 21 zucchini plants (7 plants per replicate) in the climatic chamber described above. After a few minutes, the two aphids were enclosed inside a clip cage (2 cm of diameter). The clip cages were placed on the adaxial side of the youngest fully developed leaf of each plant. After 24 h, adults and excess newborn nymphs were removed with a wet paintbrush, leaving only three nymphs per clip cage.
For the survival measures, these three nymphs were monitored to adulthood, with aphid mortality assessed on day 7. The offspring of the experimental adults was then assessed by daily counting and removal of new-born nymphs for 7 days (the same number of days as the pre-reproductive period). Aphid fecundity was estimated as the mean number of nymphs per aphid per day.
Acute toxicity bioassays
To assess the susceptibility of A. gossypii to the insecticides fipronil and pyrethrins at sub-lethal doses, adult aphids that had completed two generations on control (Co) and on plants treated with the F100 and F1000 doses of fipronil were used. The nymphs produced on day 6 and day 7 by the experimental adults used for the fecundity experiment were removed from the clip cages but left to become adults on another leaf of the same plant and then used for the experiments. The tests were performed according to the leaf dip method described in the Insecticide Resistance Action Committee Susceptibility Test Method No. 019 (IRAC 2024), with some modifications. The solutions used were 1 mg/L of fipronil (F-acute), 1 mg/L of pyrethrins (P-acute), and a control (tap water, TW). The doses of fipronil and pyrethrins were chosen based on previously determined acute lethal doses for A. gossypii (data not shown). The two insecticides and the tap water were applied to untreated zucchini disc leaves (3 cm of diameter) by dipping them in the test suspensions for 5 seconds with gentle agitation. Treated disc leaves were placed with their dorsal side on four layers of wet (saturated with distilled water) filter paper in a Petri dish, and 15 adult female aphids were placed on the surface of each treated leaf. Test and control discs leaves were maintained at 22°C in the thermostatic chamber described above.
For the insecticide susceptibility, nine experimental treatments were generated by combining high-dose insecticide treatments (TW, F-acute and P-acute) and the water treatment applied to the plants where the aphids developed (Co, F100 and F1000). Mortality was estimated after 48 hours as the number of live individuals relative to the initial number of aphids; an aphid was considered alive if it could walk or at least move when gently touched with a soft brush. Three replicates per treatment were used for this experiment, with each replicate consisting of at least 3 disc leaves containing 10–15 individuals, giving a total of at least 100 individuals per treatment.
Plant sample collection, fipronil extraction
Plant sampling included collection of leaves and flowers. At the end of the experiment (35 days after transplanting), leaf and flower samples were collected from three separate plants of each experimental treatment, weighed and lyophilised. The extraction of fipronil from the zucchini’s leaves and flowers was performed in triplicates, via QuEChERS extraction and dSPE PSA-C18 clean-up protocol (adapted protocol, Montemurro et al., 2020). In short, 1 g of freeze-dried and milled leaves or flowers was placed in a 50 mL falcon tube and hydrated with 9 mL HPLC water. The tubes were vortexed for 10 sec, shaken for 2 min at 2500 rpm and left to hydrate for 1 h. 10 mL of acetonitrile and 50 µL of formic acid were added in the tubes, vortexing was repeated and the extraction salts (1 g NaCl and 4 g MgSO4) were added in the tubes. The mixture was instantly shaken to prevent the formation of crystalline agglomerates. Tubes were vortexed for 10 sec, shaken for 2 min at 2500 rpm and centrifuged at 4°C for 10 min at 4,000 rpm. The supernatant, containing the organic phase, was transferred into glass tubes, and left overnight at -20°C for the precipitation of fatty acids and waxes. The following day, the clean-up step involved the transfer of 6 mL of the supernatant into the PSA (primary secondary amine) tubes (150 mg PSA, 150 mg C18, 900 mg MgSO4) and the mixture was vortexed and shaken, then centrifuged at 4°C for 5 min at 4,000 rpm. On the day of the analysis, 1 mL of the supernatant was spiked with the internal standard mixture at a concentration of 20 µg/L, the sample was evaporated until dryness under nitrogen at room temperature, reconstituted with 1 mL of methanol:water (2:8) solution and injected for UHPLC-MS/MS analysis.
The analysis of fipronil in the zucchini extracts was performed with a Waters Acquity Ultra-PerformanceTM liquid chromatography system coupled to a 5500 QTRAP hybrid triple quadrupole-linear ion trap mass spectrometer (Applied Biosystems, Foster City, CA, USA) with a turbo Ion Spray source. Fipronil was analysed in negative ionization mode (method from Castaño-Trias et al., 2023) using an Acquity BEH C18 column (50 × 2.1 mm i.d., 1.7 µm particle size) for chromatographic separation. The mobile phase consisted of (A) ACN and (B) 5mM ammonium acetate/ammonia (pH = 8). The total run was 3.7 min with a flow rate was 0.6 mL/min, and the elution gradient was as follows: initial condition 95% B; 0-1.5 min 5–60% A, 1.5-2 min 60–100% A, 2–3 min 100% A, 3-3.2 min return to initial conditions, 3.2–3.7 min equilibration of the column. The column temperature was set at 30°C and an injection volume was set at 5µL. The quantification of fipronil analytes in zucchini flowers and leaves, as well as the detection (without possibility for quantification) of its metabolites, were performed by selective reaction monitoring by monitoring two mass transitions between the precursor ion and the most abundant fragment ions for each compound. The one at higher intensity was used for quantification purposes, while the second one was used for confirmation of the compound identification. Data acquisition and processing was performed with Analyst 1.5.1 software.
Statistical analysis
Data on nymph-to-adult survival and on the acute toxicity bioassays were analysed using generalized linear mixed-effects models with binomial error (logit link function), because a binomial distribution is appropriate for modelling binary data (dead / alive). P-values for differences among experimental treatments were obtained by analyses of deviance (type II Wald chi-squared tests). For the nymph-to-adult survival model, “water treatment” (three levels: Co, F100, and F1000) was the main factor and “replicate nested in treatment” was the random effect. For the acute toxicity bioassay model, “water treatment” and “insecticide treatment” (three levels: TW, F-acute, P-acute) were the fixed factors together with their interactions, and “replicate” nested within the fixed factors was the random effect.
The Shapiro-Wilk test was used to test the normality of the fecundity data and then analysed using a linear mixed-effects model fitted with restricted maximum likelihood, with “water treatment” (three leaves) and “reproductive interval” (seven levels) as a fixed factors and “replicate” nested within the fixed factors as random effect. The two main effects and the interaction term were fitted for this model.
A backward procedure was used to sequentially remove non-significant effects, allowing the identification of the most parsimonious model. All insect data were analysed using the statistical software R version “4.4.1-Race for Your Life” (R Core Team 2024) with the packages lme4 (Bates et al. 2015) and lmerTest (Kuznetsova et al. 2017).