We maintained Canton-S WT of Drosophila melanogaster on standard cornmeal-dextrose agar (prepared according to manufacturer’s protocol: Genesee Scientific, San Diego, California) on a 12-hour light cycle at 25 °C and 60% relative humidity, unless otherwise specified. Upon eclosure, we transferred adult flies to experimental media (see below). All experimental individuals were three-day-old adult female flies.
Chemical reagents
To create a 1:1 stock solution, we dissolved IMD (Sigma-Aldrich, USA) in dimethyl sulfoxide (DMSO), which was then pipetted into the food during preparation at concentrations ranging from 0, 10, 50, and 100 µM following Daisley et al. 201741. We supplemented food media with indigo-carmine dye (Sigma-Aldrich, USA) to visually confirm food consumption, and additionally, before beginning our experiments, we performed a separate pilot survival assay to confirm the exposure protocol (Supplemental Material, Fig. S1).
Experiment 1: Single neuron function
We performed single-unit olfactory sensory neuron (OSN) recordings on flies (N = 14) following49. Flies were exposed to diets prepared as described above to contain 0, 10, and 100 µM of IMD. All recordings were conducted as previously described50,51. Briefly, we recorded neuron activity from the AB3 sensilla immediately following stimulation with methyl hexanoate. Methyl hexanoate is a compound found in ripe fruit (such as pineapples, passion fruit, and strawberries), and is known to elicit a strong response in olfactory receptor neurons52. We presented odor stimuli using a Pasteur pipette containing 50 µL of methyl hexanoate diluted in paraffin oil (10-4 dilution) placed on a Whatman 13-mm filter paper disk (Millipore). We prepared these odor cartridges shortly before odor presentation, and each cartridge was never used more than three times. Three recordings were taken from three separate AB3 sensilla for each individual where possible (n = 20, 30, and 22 recordings per treatment, respectively). Then, we assessed the spike frequency of the AB3a neuron in the first 500 milliseconds post-stimulus. The mean response frequency to DMSO alone (which includes spontaneous response) was subtracted from each odor response frequency for each recording.
Experiment 2: Antennal response
We performed whole antenna electroantennograms (EAG) following53,54 on flies exposed to diets prepared as described above to contain either 0 or 100 µM of IMD (N = 5 and 4 individuals, respectively). We placed the recording electrode in the antennal region rich in AB3 sensilla and then conducted three recordings per individual (n = 14 and 10 recordings, respectively) immediately following stimulation with methyl hexanoate diluted in paraffin oil (10-4 dilution, preparing odor cartridges as above). We quantified antennal response as the minimum, the integral of the response curve, and the time to return to baseline (0 mV) for each recording.
Experiment 3: Performance on an olfactory test
To assess the potential impacts of IMD on olfactory-guided behavior, we used a modified trap assay29 where flies were simultaneously offered two odorants representing ripe vs. unripe fruit. Assays took place inside experimental chambers (dimensions 24.75 x 14.5 x 11.5 cm) housed in an incubator on a 24-hour dark cycle to reduce the influence of visual stimuli. Odor traps were constructed from clear plastic 12-dram vials. Each odor trap was fitted with the cut end of a 1000 µL pipette tip, allowing flies to enter the trap, but not exit.
One odor trap contained 1000 µL unadulterated canned liquid pineapple juice (Dole Food Company and Total Produce, USA, which was kept frozen until use), and the other contained pineapple juice adulterated with 10% ethanol by volume. Ethanol, a byproduct of the fermentation process, should be attractive to mated female fruit flies, which lay eggs in fermenting fruit44. In addition, ethanol has been shown to elicit a strong EAG response in D. melanogaster, and when used in a blend with other compounds commonly found in fermenting fruit, the blend lured and trapped more flies than compounds in isolation55. Flies were exposed to IMD as in Exp. 1 and 2, but in this experiment, 10 female and 10 male flies were placed in each vial to allow mating. After three days of exposure, adult females were transferred to an experimental chamber. Within a chamber, all flies were assigned to the same treatment (0, 10, or 100 µM). After 48 hours, we counted flies in each odor trap and the number of un-trapped flies remaining in the experimental chamber.
Analyses
All statistical analyses were carried out in R version 4.0.256. To compare OSN response by treatment, we used a linear mixed-model approach treating individual as a random effect using the lme4 package57 and type II ANOVA using the ANOVA function from the car package58 and Tukey comparison was performed using the glht function from the multcomp package59. Electroantennogram response analysis again used a linear mixed-model approach treating individual as a random effect, but here we used three separate models for each of our three response variables: minimum, integral, and return to baseline for each treatment. To assess performance on an olfactory test, here we used a binomial linear regression to compare the total number of flies in either odor trap (“trapped flies”) to flies in the larger experimental chamber (“un-trapped flies”) for each treatment using the stats package60 and type II ANOVA using the ANOVA function from the car package58 and Tukey comparison was performed using the glht function from the multcomp package59 as before. This allowed us to estimate the effects of the insecticide on fly movement and/or feeding motivation, two aspects of behavior for which IMD has well-documented effects18. To assess if IMD also alters odor discrimination, we used a separate binomial regression to compare the number of flies in the unadulterated juice odor trap and the adulterated (juice + 10% ethanol) juice odor trap.