1. Neuroscience

Low doses of the organic insecticide spinosad trigger lysosomal defects, elevated ROS, lipid dysregulation, and neurodegeneration in flies

  1. Felipe Martelli
  2. Natalia H Hernandes
  3. Zhongyuan Zuo
  4. Julia Wang
  5. Ching-On Wong
  6. Nicholas E Karagas
  7. Ute Roessner
  8. Thusita Rupasinghe
  9. Charles Robin
  10. Kartik Venkatachalam
  11. Trent Perry
  12. Philip Batterham
  13. Hugo J Bellen  Is a corresponding author
  1. School of BioSciences, The University of Melbourne, Australia
  2. Department of Molecular and Human Genetics, Baylor College of Medicine, United States
  3. Department of Integrative Biology and Pharmacology, McGovern Medical School at the University of Texas Health Sciences Center, United States
  4. Neurological Research Institute, Texas Children Hospital, United States
  5. Howard Hughes Medical Institute, Baylor College of Medicine, United States
https://doi.org/10.7554/eLife.73812
Share this article
Cite this article
  1. Felipe Martelli
  2. Natalia H Hernandes
  3. Zhongyuan Zuo
  4. Julia Wang
  5. Ching-On Wong
  6. Nicholas E Karagas
  7. Ute Roessner
  8. Thusita Rupasinghe
  9. Charles Robin
  10. Kartik Venkatachalam
  11. Trent Perry
  12. Philip Batterham
  13. Hugo J Bellen
(2022)
Low doses of the organic insecticide spinosad trigger lysosomal defects, elevated ROS, lipid dysregulation, and neurodegeneration in flies
eLife 11:e73812.
https://doi.org/10.7554/eLife.73812
  2. Download BibTeX
  3. Download .RIS

Abstract

Large-scale insecticide application is a primary weapon in the control of insect pests in agriculture. However, a growing body of evidence indicates that it is contributing to the global decline in population sizes of many beneficial insect species. Spinosad emerged as an organic alternative to synthetic insecticides and is considered less harmful to beneficial insects, yet its mode of action remains unclear. Using Drosophila, we show that low doses of spinosad antagonize its neuronal target, the nicotinic acetylcholine receptor subunit alpha 6 (nAChRα6), reducing the cholinergic response. We show that the nAChRα6 receptors are transported to lysosomes that become enlarged and increase in number upon low doses of spinosad treatment. Lysosomal dysfunction is associated with mitochondrial stress and elevated levels of reactive oxygen species (ROS) in the central nervous system where nAChRα6 is broadly expressed. ROS disturb lipid storage in metabolic tissues in an nAChRα6-dependent manner. Spinosad toxicity is ameliorated with the antioxidant N-acetylcysteine amide. Chronic exposure of adult virgin females to low doses of spinosad leads to mitochondrial defects, severe neurodegeneration, and blindness. These deleterious effects of low-dose exposures warrant rigorous investigation of its impacts on beneficial insects.

Editor's evaluation

Insecticides have been implicated in the decline of beneficial insect species. The organic insecticide Spinosad has emerged as a alternative to synthetic insecticides and is thought to be less harmful to beneficial insects than synthetic insecticides. This article used the insect model Drosophila to analyze the impact of Spinosad. It reveals that low doses of Spinosad antagonize its neuronal target, the nicotinic acetylcholine receptor subunit alpha 6, affecting the physiology of Drosophila. This study reveals that although being organic and assumed to be less harmful, spinosad can have profound impact on non-target insects.

https://doi.org/10.7554/eLife.73812.sa0

Introduction

Insecticide applications maximize crop yield, but negatively impact populations of insects that provide beneficial services in agriculture and horticulture (Sánchez-Bayo and Wyckhuys, 2019). The global decline in population sizes of these beneficial insects creates challenges for ecosystems and farming. Although estimates differ depending on the regions and the methodologies used (Wagner et al., 2021), one recent study suggests an approximately 9% decline in terrestrial insect abundance per decade since 1925 (van Klink et al., 2020). While the precise extent to which insecticides are involved remains undetermined, they have consistently been associated as a key factor, along with climate change, habitat loss, and increased levels of pathogens and parasites (Cardoso et al., 2020; Sánchez-Bayo and Wyckhuys, 2019; Wagner et al., 2021). Much attention has been given to neonicotinoid insecticides, both in the scientific literature and in public discourse, because of the evidence that these chemicals contribute to the bee colony collapse phenomenon (Lu et al., 2014; Lundin et al., 2015).

In assessing the risk posed by insecticides, it is important to study the molecular and cellular events that unfold following the interaction between the insecticide and its target. Many insecticides target ion channels in the nervous system. At the high doses used to kill pests, these insecticides produce massive perturbations to the flux of ions in neurons, resulting in lethality (Breer and Sattelle, 1987; Perry and Batterham, 2018; Scott and Buchon, 2019). But non-pest insects are likely to be exposed to much lower doses, and the downstream physiological processes that are triggered are poorly understood. In a recent study, low doses of the neonicotinoid imidacloprid were shown to stimulate an enduring flux of calcium into neurons via the targeted ligand-gated ion channels (nicotinic acetylcholine receptors [nAChRs]) (Martelli et al., 2020). This causes an elevated level of reactive oxygen species (ROS) and oxidative damage that radiates from the brain to other tissues. Mitochondrial stress leads to a significant drop in energy levels, neurodegeneration, and blindness (Martelli et al., 2020). Evidence of compromised immune function was also presented, supporting other studies (Chmiel et al., 2019). Many other synthetic insecticides are known to elevate the levels of ROS (Karami-Mohajeri and Abdollahi, 2011; Lukaszewicz-Hussain, 2010; Wang et al., 2016) and may precipitate similar downstream impacts. Given current concerns about synthetic insecticides, a detailed analysis of the molecular and cellular impacts of organic alternatives is warranted. Here, we report such an analysis for an insecticide of the spinosyn class, spinosad.

Spinosad is an 85%:15% mixture of spinosyns A and D, natural fermentation products of the soil bacterium Saccharopolyspora spinosa. It occupies a small (3%) but growing share of the global insecticide market (Sparks et al., 2017). It is registered for use in more than 80 countries and applied to over 200 crops to control numerous pest insects (Biondi et al., 2012). Recommended dose rates vary greatly depending on the pest and crop, ranging from 96 parts per million (ppm) for Brassica crops to 480 ppm in apple fields (Biondi et al., 2012). Like other insecticides, the level of spinosad residues found in the field varies greatly depending on the formulation, the application mode and dose used, environmental conditions, and proximity to the site of application. If protected from light, spinosad shows a half-life of up to 200 days (Cleveland et al., 2002).

Spinosad is a hydrophobic compound belonging to a lipid class known as polyketide macrolactones. Studies using mutants, field-derived-resistant strains, and heterologous expression have shown that spinosad targets the highly conserved nAChRα6 subunit of nAChRs in Drosophila melanogaster (hereafter Dα6) and a range of other insect species (Perry et al., 2015; Perry et al., 2007). Spinosad is an allosteric modulator, binding to a site in the C terminal region of the protein (Puinean et al., 2013; Somers et al., 2015). Salgado and Saar, 2004 found that spinosad allosterically activates non-desensitized nAChRs, but that low doses were also capable of antagonizing the desensitized nAChRs. It is currently accepted that spinosad causes an increased sensitivity to ACh in certain nAChRs and an enhanced response at some GABAergic synapses, causing involuntary muscle contractions, paralysis, and death (Biondi et al., 2012; Salgado, 1998). However, a recent study (Nguyen et al., 2021) showed that both acute and chronic exposures to spinosad cause Dα6 protein levels in the larval brain to decrease. A rapid loss of Dα6 protein during acute exposure was blocked by inhibiting the proteasome system (Nguyen et al., 2021). As Dα6 loss-of-function mutants are both highly resistant to spinosad and viable (Perry et al., 2021; Perry et al., 2007), it was suggested that the toxicity of spinosad may be due to the overloading of protein degradation pathways and/or the internalization of spinosad where it may cause cellular damage. Higher doses of spinosad than the ones used here have been shown to cause cellular damage via mitochondrial dysfunction, oxidative stress, and programmed cell death in cultured insect cells (Spodoptera frugiperda Sf9) (Xu et al., 2017; Yang et al., 2017).

Here, we show that spinosad by itself does not increase Ca2+ flux in Drosophila neurons. Indeed, the response elicited by a cholinergic agonist is stunted upon spinosad treatment. Following exposure to spinosad, Dα6 cholinergic receptors are endocytosed and trafficked to the lysosomes, leading to lysosomal dysfunction. This dysfunction is associated with high levels of oxidative stress. Antioxidant treatment prevents the accumulation of ROS, but not lysosomal expansion. ROS is a key factor in the mode of action of spinosad at low doses, triggering a cascade of damage that results in mitochondrial stress and reduced energy levels. Low chronic exposures lead to extensive neurodegeneration in the central brain and blindness. Flies carrying a Dα6 loss-of-function mutation show a mild increase in ROS, but no evidence of lysosomal dysfunction. This indicates that the lysosomal defect observed in wild-type flies is not due to the absence of Dα6 from neuronal membranes but rather trafficking of Dα6 to lysosomes under conditions of spinosad exposure. Given the high degree of conservation of the spinosad target between insect species (Perry et al., 2015), our data indicate that this insecticide has the potential to cause harm in non-pest insects at low doses.

Results

Low doses of spinosad affect survival and prevent Ca2+ flux into neurons expressing Dα6

As a starting point to study the systemic effects of low-dose spinosad exposure, a dose that would reduce the movement of third-instar larvae by 50% during a 2 hr exposure was determined. This was achieved with a dose of 2.5 ppm (Figure 1A). Under this exposure condition, only 4% of wild-type larvae survived to adulthood (Figure 1B), whereas 88% nAChRα6 knockout (Dα6 KO) mutants survived (Figure 1C). The effect of this dose was measured on cultured primary neurons of third-instar larva brain, where the Dα6 gene promoter was used to drive the GCaMP5G:tdTomato cytosolic [Ca2+] sensor. As no alterations in basal Ca2+ levels were detected in neurons expressing Dα6 response to 2.5 ppm (Figure 1D and E), a dose of 25 ppm was tested, again with no measurable impact (Figure 1D and E). After 5 min of spinosad exposure, neurons were stimulated by carbachol, a cholinergic agonist that activates nAChR. Spinosad-exposed neurons exhibited a significant decrease in cholinergic response when compared to unexposed neurons (Figure 1D and E). Total Ca2+ content mobilized from ER remained unaltered as measured by thapsigargin-induced Ca2+ release (Figure 1D and E). While it was not determined whether the Ca2+ transients reflect reduced influx from internal or external sources (Campusano et al., 2007), spinosad exposure led to a diminished Ca2+ transient and reduced cholinergic response. Hence, in contrast to imidacloprid, which leads to an enduring Ca2+ influx in neurons (Martelli et al., 2020), spinosad reduces the Ca2+ response mediated by Dα6.

Figure 1
Download asset Open asset
Low doses of spinosad are lethal and fail to increase Ca2+ levels in neurons.

(A) Dose–response to spinosad in Line14 wild-type larvae by an assay of larval movement over time, expressed in terms of relative movement ratio (n = 100 larvae/treatment). (B) Adult eclosion rate after Line14 larvae were subjected to a 2 hr exposure to 2.5 parts per million (ppm) spinosad, rinsed and placed back onto insecticide-free medium (n = 100 larvae/treatment). (C) Adult eclosion rate of Canton-S and Canton-S Dα6 KO larvae subjected to a 2 hr exposure at 2.5 ppm spinosad, rinsed and placed back onto insecticide-free medium (n = 100 larvae/treatment). (D) Cytosolic [Ca2+] measured by GCaMP in neurons expressing Dα6. Measurement is expressed as a ratio of the signals of GCaMP5G signal and tdTomato. Spinosad (2.5 ppm or 25 ppm) was added to the bath solution at 1 min after recording started. At 6 min and 8 min, the spinosad-exposed and unexposed groups were stimulated with 100 µM carbachol and 5 µM thapsigargin, respectively. Each point represents the average of at least 50 cells. (E) Peak [Ca2+] responses to spinosad and carbachol. Error bars in (A) and (E) represent mean ± SEM and in (B) and (C) mean ± SD. (A, E) One-way ANOVA followed by Tukey’s honestly significant difference test; (B, C) Student’s unpaired t-test. *p<0.05, ***p<0.001.

Spinosad exposure causes lysosomal dysfunction in a Dα6-dependent manner

Spinosad exposures cause a gradual reduction in the Dα6 signal in brains (Figure 2A and B; Nguyen et al., 2021). To test whether spinosad affects lysosomes, we stained larval brains with LysoTracker. No phenotype was observed after 1 hr exposure, but after a 2 hr exposure at 2.5 ppm, spinosad caused an eightfold increase in the area occupied by lysosomes (Figure 2C and E). 6 hr after the 2 hr initial exposure ended, the area occupied by lysosomes in brains was 24-fold greater than in unexposed larvae (Figure 2C and E). In contrast, no increased number of enlarged lysosomes were observed in Dα6 KO mutants in the presence or absence of spinosad exposure (Figure 2D and F). These data indicate that the lysosomal expansion is dependent on both the presence of the Dα6 receptors and spinosad. Significantly, the Dα6 receptors were found to colocalize with the enlarged lysosomes (Figure 2G), indicating that enlarged lysosomes are trapping Dα6 receptors in response to spinosad exposure.

Figure 2
Download asset Open asset
Spinosad exposure causes lysosomal expansion, and Dα6 colocalizes with enlarged lysosomes.

(A) Dα6 signal in larval brains exposed to 2.5 parts per million (ppm) spinosad for 30 min, 1 hr, or 2 hr. Larvae obtained by crossing UAS-Dα6 (CFP tagged) in the Line14 Dα6nx loss-of-function mutant background to a native Da6 driver (the Gal4-L driver) in the same background. OL, optic lobe; VNC, ventral nerve cord. (B) Quantification of (A) (n = 3 larvae/condition; three brain sections/larva). (C) LysoTracker staining shows lysosome expansion in the brain of Line14 larvae exposed to 2.5 ppm spinosad for 1 hr, 2 hr, or 6 hr post the 2 hr exposure. (D) LysoTracker staining shows lysosomes in the brain of Canton-S and Canton-S Dα6 KO larvae exposed to 2.5 ppm spinosad for 2 hr. (E) Quantification of (C), LysoTracker area (%) (n = 7 larvae/treatment; three optic lobe sections/larva). (F) Quantification of (D), LysoTracker area (%) (n = 7 larvae/treatment; three optic lobe sections/larva). (G) Larvae expressing Dα6 tagged with CFP exposed to 2.5 ppm spinosad for 2 hr show colocalization of the Dα6 and lysosomal signals. Pink arrowheads indicate Dα6 CFP signal colocalizing with lysosomes identified with LysoTracker staining. Microscopy images were obtained with a Leica SP5 Laser Scanning Confocal Microscope. Error bars in (E) and (F) represent mean ± SD. (B, E, F) One-way ANOVA followed by Tukey’s honestly significant difference test. ***p<0.001.

Spinosad exposure affects mitochondria turnover and reduces energy metabolism that is counteracted by antioxidant treatment

Defects in lysosomal function have been shown to impact other organelles, especially mitochondria (Deus et al., 2020). To investigate whether mitochondrial function was also affected by spinosad exposure, we assessed mitochondrial turnover using the MitoTimer reporter line (Laker et al., 2014). A 2 hr spinosad exposure induced an increase of 31% and 36% for the green (healthy mitochondria) and red (stressed mitochondria) signals in the optic lobes of the larval brain, respectively (Figure 3A and B). For the digestive tract, a 19% and 32% increase was observed in the proventriculus for green and red signals, respectively (Figure 3A and C). The mito-roGFP2-Orp1 construct, an in vivo mitochondrial H2O2 reporter (Albrecht et al., 2011), was used to identify the origin of ROS induced by spinosad exposure at 2.5 ppm for 2 hr. A subtle, but significant, increase in the 405 (oxidized mitochondria signal)/488 (reduced mitochondria signal) ratio was observed in the brain (20% on average) and anterior midgut (10% on average) (Figure 3—figure supplement 1), pointing to a rise in H2O2 generation upon a few hours of exposure. Similarly to the MitoTimer reporter, an increase in the oxidized mitochondrial signal was accompanied by the increase in the reduced mitochondrial signal, accounting for the subtle increase in 405/488 ratio. To further examine whether the results obtained with the mitochondrial reporters were connected to increased ROS production in mitochondria, we measured the enzyme activity of mitochondrial aconitase (m-aconitase), a highly ROS-sensitive enzyme (Yan et al., 1997). We observed a mean 34% reduction in m-aconitase activity (Figure 3D), indicating an increased presence of ROS in mitochondria after the 2 hr exposure. Immediately after the 2 hr exposure, a mean 36% increase in systemic ATP levels was observed, followed by a 16.5% reduction 12 hr after the 2 hr exposure (Figure 3E). The initial increase in energy levels is consistent with the increase in the signals of healthy and stressed mitochondria identified by the MitoTimer and mito-roGFP reporters at this time point. However, the reduction in ATP levels 12 hr after the exposure shows that the mitochondrial energy output is eventually impaired.

Figure 3 with 1 supplement see all
Download asset Open asset
Spinosad exposure impacts mitochondria and energy metabolism, and antioxidant treatment reduces spinosad toxicity.

(A) Optic lobes of the brain and proventriculus of MitoTimer reporter expressing larvae. 2.5 parts per million (ppm) spinosad exposure for 2 hr increased the signal of healthy (green) and unhealthy (red) mitochondria. (B) Normalized mean fluorescence intensity of MitoTimer signals in optic lobes (n = 20 larvae/treatment; three image sections/larva). (C) Normalized mean fluorescence intensity of MitoTimer signals in proventriculus (n = 20 larvae/treatment; three image sections/larva). (D) Relative m-aconitase activity in whole Line14 larvae exposed to 2.5 ppm spinosad for 2 hr (n = 25 larvae/replicate; six replicates/treatment). (E) Relative systemic ATP levels in Line14 larvae immediately after the 2 hr exposure to 2.5 ppm spinosad and 12 hr post 2 hr exposure (n = 20 larvae/ replicate; six replicates/ treatment). (F) Pretreatment with N-acetylcysteine amide (NACA) improves the movement of spinosad-exposed Line14 larvae. Dose–response to insecticide by an assay of larval movement over time, expressed in terms of relative movement ratio (n = 25 larvae/replicate; four replicates/treatment). (G) Pretreatment with NACA improves survival of Line14 larvae exposed to spinosad. Adult eclosion rate (%) (n = 100 larvae/treatment). OL, optic lobe; VNC, ventral nerve cord; Pv, proventriculus; GC, gastric caeca; AM, anterior midgut. Error bars in (B) and (C) represent mean ± SD; in (F), mean ± SEM; and in (G), corrected percentage survival (Abbot’s correction). Microscopy images were obtained with a Leica SP5 Laser Scanning Confocal Microscope. (B, C, E) One-way ANOVA followed by Tukey’s honestly significant difference test; (D, F, G) Student’s unpaired t-test. **p<0.01, ***p<0.001.

To quantify the extent to which oxidative stress generated by 2.5 ppm spinosad exposure for 2 hr could affect larval motility and survival, larvae were treated with the antioxidant N-acetylcysteine amide (NACA) (Schimel et al., 2011) for 5 hr prior to insecticide exposure. NACA treatment improved larval motility by ~50% at the 2 hr exposure time point when compared to larvae not treated with the antioxidant (Figure 3F). Adult eclosion rates increased from an average 4% to 15% when larvae exposed to spinosad were treated with NACA (Figure 3F). These results show a causal role for oxidative stress in the mode of action of spinosad at low doses. Nonetheless, the results also suggest that oxidative stress is not the only responsible mediator for the detrimental effects of spinosad exposure. Cross-talk between mitochondrial stress and lysosome dysfunction may be the major culprit for the highly toxic effects of low levels of spinosad exposure. This relationship is further investigated below.

Antioxidant treatment prevents ROS accumulation but not lysosomal expansion

Given the evidence for increased ROS production, we next examined the levels of the anion O2¯ (superoxide), a primary ROS produced by mitochondria (Valko et al., 2007), as well as other ROS sources (Zielonka and Kalyanaraman, 2010), using dihydroethidium (DHE) staining. After a 1 hr exposure to 2.5 ppm spinosad, there was a mean 89% increase of ROS levels in the brain. After 2 hr, the levels were lower than that at the 1 hr time point, but still 44% higher than that in the unexposed controls (Figure 4A and B). A different pattern was observed in the anterior midgut. A significant increase of ROS levels compared with the controls (28%) was observed only at the 2 hr time point (Figure 4A and C). Unexposed Dα6 KO mutants showed a mild (17%) increase in ROS levels in the brain when compared to unexposed wild-type larvae (Figure 4—figure supplement 1). Exposure to spinosad caused no alteration of ROS levels in Dα6 KO mutants (Figure 4—figure supplement 1). Hence, the absence of Dα6 subunit by itself is able to modestly increase the oxidative stress (Weber et al., 2012), but higher levels of ROS are only observed in the presence of Dα6 and spinosad. To assess the mitochondrial origin of the ROS measured with DHE, flies expressing the superoxide dismutase gene Sod2 in the nervous system with the elav-GAL4 driver were exposed to 2.5 ppm spinosad for 2 hr. Sod2 is the main ROS scavenger in Drosophila and is localized to mitochondria (Missirlis et al., 2003). Sod1 is present in the cytosol (Missirlis et al., 2003), and expression of this gene was used as a control. While exposure to spinosad caused an average 63% increase in ROS levels in control larvae overexpressing Sod1, an average increase of only 28% was found in larvae overexpressing Sod2 (Figure 4D and E).

Figure 4 with 1 supplement see all
Download asset Open asset
Spinosad exposure increases oxidative stress, and antioxidants prevent reactive oxygen species (ROS) accumulation, but not lysosome expansion.

(A) Dihydroethidium (DHE) staining of ROS levels in the brain and anterior midgut of Line14 larvae exposed to 2.5 parts per million (ppm) spinosad for either 1 hr or 2 hr. (B) DHE normalized fluorescence intensity in brains (n = 15 larvae/treatment; three sections/larva). (C) DHE normalized fluorescence intensity in anterior midgut (n = 15 larvae/treatment; three sections/larva). (D) DHE staining of ROS levels in the brains of larvae expressing Sod2 (elav-Gal4>UAS-Sod2) or Sod1 (control cross; elav-Gal4>UAS-Sod1) in the central nervous system and exposed to 2.5 ppm spinosad for 2 hr. (E) DHE normalized fluorescence intensity in brains (n = 7 larvae/genotype/treatment; three sections/larva). (F) DHE staining of ROS levels in the brain of Line14 larvae treated with N-acetylcysteine amide (NACA) and exposed to 2.5 ppm spinosad for 2 hr. (G) DHE normalized fluorescence intensity in brains (n = 8 larvae/treatment; three sections/larva). (H) LysoTracker staining showing lysosomes in brains of Line14 larvae treated with NACA and exposed to 2.5 ppm spinosad for 2 hr. (I) LysoTracker area (%) (n = 8 larvae/treatment; three optic lobe sections/larva). OL, optic lobe; VNC, ventral nerve cord; Pv, proventriculus; GC, gastric caeca; AM, anterior midgut. Error bars represent mean ± SD. Microscopy images were obtained with a Leica SP5 Laser Scanning Confocal Microscope. (B, C, E) One-way ANOVA followed by Tukey’s honestly significant difference test; (G, I) Student’s unpaired t-test. *p<0.05, ***p<0.001.

To further dissect the relationship between lysosome dysfunction and mitochondrial stress, we exposed larvae treated with NACA to spinosad and once again quantified the levels of ROS and the area covered by lysosomes in brains. Whereas NACA treatment was able to prevent ROS accumulation in exposed animals (Figure 4F and G), it did not prevent lysosome expansion (Figure 4H and I). The presence of enlarged lysosomes in the absence of ROS suggests that the onset of the lysosomal phenotype is not caused by the rise in oxidative stress levels. NACA, however, reduced the severity of the lysosomal phenotype (mean 1.63% of lysotracker area [Figure 4I] versus mean 2.39% of lysotracker area [Figure 2E]). This suggests that, once initiated, the increase in ROS levels may worsen the phenotype associated with lysosomal dysfunction.

Brain signals trigger the impact of spinosad on metabolic tissues

Oxidative stress has the ability to affect the lipid environment of metabolic tissues, causing bulk redistribution of lipids into lipid droplets (LDs) (Bailey et al., 2015). The RNAi knockdown of mitochondrial genes, Marf and ND42, in Drosophila neurons increases the levels of ROS in the brain and precipitates the accumulation of LD in glial cells (Liu et al., 2015). Martelli et al., 2020 showed that the knockdown of the same mitochondrial genes in Drosophila neurons can also precipitate the accumulation of LD in fat bodies and a reduction of LD in Malpighian tubules. Such changes in the lipid environment of metabolic tissues were recapitulated by low imidacloprid exposures, which, like spinosad, also causes an increase of ROS levels in the brain that further spreads to the anterior midgut (Martelli et al., 2020). To test if spinosad could also affect the lipid environment of metabolic tissues, LD numbers were assessed using Nile Red staining. Larvae exposed to 2.5 ppm spinosad for 2 hr showed a 52% increase in the area covered by LD in the fat body (Figure 5A and B), with a significant reduction in the number of large LD and an increase in small LD (Figure 5—figure supplement 1).

Figure 5 with 2 supplements see all
Download asset Open asset
Spinosad triggers reactive oxygen species (ROS)-driven lipid changes in metabolic tissues of wild-type larvae but not Dα6 loss-of-function larvae.

(A) Nile Red staining showing lipid droplets in larval fat bodies of Line14 and Canton-S strains and their respective Dα6 loss-of-function mutant strains. Larvae exposed to 2.5 parts per million (ppm) spinosad for 2 hr. (B) Area covered by lipid droplets in fat body (%) (n = 3 larvae/treatment/genotype; five image sections/larva). (C) Nile Red staining showing lipid droplets in fat bodies of Line14 larvae treated with N-acetylcysteine amide (NACA) and exposed to 2.5 ppm spinosad for 2 hr. (D) Area covered by lipid droplets in fat body (%) (n = 3 larvae/treatment; five image sections/larva). (E) Dihydroethidium (DHE) staining of ROS levels in the fat body of Line14 larvae exposed to 2.5 ppm spinosad for 2 hr. (F) DHE normalized fluorescence intensity in fat body (n = 3 larvae/treatment; five sections/larva). (G) Amount of lipids in hemolymph (µg/µL) by colorimetric vanillin assay of Line14 and Canton-S larvae and their respective Dα6 loss-of-function mutants. Larvae exposed to 2.5 ppm spinosad for 2 hr (n = 10 replicates/treatment/time point; 30 larvae/replicate). Microscopy images were obtained with a Leica SP5 Laser Scanning Confocal Microscope. Error bars in (F) represent mean ± SD. (B, D, G) One-way ANOVA followed by Tukey’s honestly significant difference test; (F) Student’s unpaired t-test. *p<0.05, **p<0.01, ***p<0.001.

Once inside the insect body, spinosad could theoretically access any tissue via the open circulatory system. Given that Dα6 is present in the nervous system (Perry et al., 2015; Somers et al., 2015), and that elevated levels of ROS were observed earlier in the brain than in metabolic tissues, we addressed the following question: Do the interactions between spinosad and Dα6 in the brain provide the signal that ultimately leads to the observed disturbance of the lipid environment in the metabolic tissues? No expression of Dα6 has been reported in gut and fat body, but it is abundantly and widely expressed in most CNS neurons with little to no expression in glia (Li et al., 2021). Two different Dα6 loss-of-function mutants (Line14 Dα6nx loss-of-function mutant and Canton-S nAChRα6 knockout) and their respective genetic background control lines were tested for LD. Larvae were submitted to the same exposure conditions, 2.5 ppm of spinosad for 2 hr. Neither of the mutants tested showed an increase in the area occupied by LD, compared to their respective genetic background, under conditions of spinosad exposure (Figure 5A and B). Interestingly, Line14 Dα6nx loss-of-function mutant and Canton-S Dα6 KO mutant showed on average 16 and 20% larger area covered by LD in fat body than their respective background control lines (Figure 5A and B). These data show that the Dα6 loss of function by itself affects the lipid environment of metabolic tissues.

Treatment with NACA prior to insecticide exposure significantly ameliorated spinosad effects on fat body LD accumulation (Figure 5C and D). This indicates that ROS induced by spinosad exposure is indeed involved in the LD phenotype in fat bodies. However, measurements of ROS levels in fat bodies showed no differences between exposed and unexposed larvae (Figure 5E and F). This result indicates that the presence of a ROS signal other than the one measured here causes the bulk redistribution of lipids into LD. That no accumulation of LD was observed in the absence of Dα6 and presence of spinosad (Figure 5A and B) suggests that in wild-type flies spinosad exposure generates a signal in the brain that triggers fat body to respond. It is possible that oxidizing agents, such as peroxidized lipids, are transported through the hemolymph to the fat body (Martelli et al., 2020; Padmanabha and Baker, 2014). Alternatively, other secreted signals from the brain affected fat body metabolism.

To test for alterations of lipid levels in hemolymph, we used the vanillin assay (Cheng et al., 2011). Whereas the wild-type Line14 and Canton-S strains showed an average increase of 14 and 11% in lipid levels in hemolymph in response to spinosad, neither of the Dα6 loss-of-function mutants showed significant changes after exposure (Figure 5G). This result supports the notion that changes in the lipid environment upon spinosad exposure depend on the insecticide interacting with Dα6 receptors. However, Dα6 mutants showed higher lipid levels than their respective wild-type controls (Figure 5G).

Spinosad doses that do not impact larval survival were also examined for perturbations in the lipid environment. Doses of 0.5 ppm for 2 hr or 0.1 ppm for 4 hr were used as they had no impact on survival rate (Figure 5—figure supplement 2). Both doses caused on average a 29% increase in the area occupied by LD in fat bodies (Figure 5—figure supplement 2). This impact is smaller than that observed with 2.5 ppm, indicating that this phenotype is dose dependent. Once again, an increase in the number of small LD and reduction in the number of large LD was observed (Figure 5—figure supplement 2). In contrast, both doses caused on average a 72% reduction in the total number of LD in the Malpighian tubules (Figure 5—figure supplement 2).

Spinosad triggers major alterations in the lipidome pointing to impaired cell membrane function and a severe decrease in mitochondrial cardiolipins

To further investigate the impacts on lipid environment, we performed a lipidomic analysis on whole larvae exposed to 2.5 ppm spinosad for 2 hr. Significant changes were observed in the levels of 88 lipids out of the 378 detected by mass spectrometry (Figure 6A, Figure 6—source data 1). A significant portion of the changes in lipids corresponds to a reduction in phosphatidylcholine (PC), phosphatidylethanolamine (PE), and some triacylglycerol (TAG) species. Multivariate analysis (Figure 6B) indicates that the overall lipidomic profiles of exposed larvae form a tight cluster that is distinct from the undosed control. The use of whole larvae for lipidomic analysis reduces the capacity to detect significant shifts in lipid levels that predominantly occur in individual tissues but allows the identification of broader impacts on larval biology. In this context, the observed 65% reduction in the levels of identified cardiolipins (CLs) is particularly noteworthy (Figure 6C). CLs are highly enriched in mitochondria and are required for the proper function of the electron transport chain, especially Complex 1, the major ROS generator when dysfunctional (Quintana et al., 2010; Ren et al., 2014), consistent with the increase in ROS in mitochondria described earlier.

Figure 6
Download asset Open asset
Spinosad disturbs the lipid profile of exposed larvae.

Lipidomic profile of larvae exposed to 2.5 parts per million (ppm) spinosad for 2 hr (n = 10 larvae/replicate; three replicates/treatment). (A) 88 lipid species out of the 378 identified were significantly affected by insecticide treatment (one-way ANOVA, Turkey’s honestly significant difference test, p<0.05). The cell colors represent the z-scores, that is, the standardized scores on the same scale, calculated by dividing a score’s deviation by the standard deviation in the row. The features are color-coded by row with red indicating low intensity and green indicating high intensity. (B) Principal component analysis of 378 lipid species. Each dot represents the lipidome data sum of each sample. First component explains 41.4% of variance and second component explains 24.7% of variance. (C) Relative proportion of cardiolipins in exposed animals versus control. Error bars represent mean ± SEM, Student’s unpaired t-test, **p<0.01.

Figure 6—source data 1

Impact of spinosad on the lipidomic profile.

Lipidomic profile of larvae exposed to 2.5 parts per million (ppm) spinosad or control (equivalent dose of dimethyl sulfoxide) for 2 hr as detected by liquid chromatography-mass spectrometry. Values are expressed as peak intensity area normalized to sample weight.

https://cdn.elifesciences.org/articles/73812/elife-73812-fig6-data1-v1.docx

Chronic low exposure to spinosad causes neurodegeneration and progressive loss of vision

Next, we investigated the effects of chronic exposure of low levels of spinosad in adult female virgin flies. A dose of 0.2 ppm spinosad, which kills 50% of adults within 25 days (Figure 7A), was used in all chronic exposure experiments. Two different behavioral outputs were initially assessed: bang sensitivity (BS) and climbing. BS is associated with seizures in flies. Several fly mutants that exhibit BS have been previously described (Kanca et al., 2019; Saras and Tanouye, 2016). The assay measures the time it takes for flies to recover to a standing position following mechanical shock induced by vortexing. Wild-type flies recover in a few seconds, whereas BS mutants require typically between 5 and 60 s. Exposures for 10 and 20 days to 0.2 ppm increased the BS phenotype that has been associated with perturbations in synaptic transmission (Saras and Tanouye, 2016). These can arise from various defects including defective channel localization, neuronal wiring, and mitochondrial metabolism (Fergestad et al., 2006; Figure 7B). Exposed flies also performed poorly in climbing assays, a phenotype that is often linked to neurodegeneration (McGurk et al., 2015). Indeed, 16, 73, and 84% of flies failed to climb after 1, 10, and 20 days of exposure, respectively (Figure 7C). These data suggest that low doses of spinosad induce neurodegenerative phenotypes.

Figure 7
Download asset Open asset
Chronic effects of spinosad exposure are more severe than loss of Dα6 expression in adult virgin females.

(A) A chronic exposure to 0.2 parts per million (ppm) spinosad kills 50% of flies within 25 days (n = 25 flies/replicate; four replicates/treatment). (B) Chronic exposure to 0.2 ppm spinosad increases bang sensitivity after 10 and 20 days of exposure. Time to regain normal standing posture (seconds) after flies were vortexed in a clear vial for 10 seconds (n = 100 flies/time point/treatment). (C) Chronic exposure to 0.2 ppm spinosad reduces climbing ability. Percentage of flies that failed to climb after 1, 10, and 20 days of exposure (n = 100 flies/time point/treatment). (D) Longevity of unexposed Canton-S and Canton-S Dα6 KO mutants (n = 100 flies/genotype). (E) Chronic exposure to 0.2 ppm spinosad for 25 days has no impact on survival of Canton-S Dα6 KO mutants (n = 25 flies/replicate; four replicates/genotype). (F) Chronic exposure to 0.2 ppm spinosad does not increase bang sensitivity of Canton-S Dα6 KO mutants. Time to regain normal standing posture (seconds) after flies were vortexed in a clear vial for 10 s (n = 100 flies/time point/genotype/treatment). (F) Chronic exposure to 0.2 ppm spinosad does not reduce climbing ability of Canton-S Dα6 KO mutants. Percentage of flies that failed to climb (n = 100 flies/time point/genotype/treatment). Error bars in (A), (D), and (E) represent mean ± SEM. (A, D, E) Kaplan–Meier method and the log-rank Mantel–Cox test. (B, C, F, G) Kruskal–Wallis followed by Dunn’s multiple comparisons test. *p<0.05, ***p<0.001.

The same phenotypes were also assessed in adult female virgin Dα6 KO mutants. Unexposed mutant flies show a significant reduction in longevity compared to unexposed wild-type individuals, but that difference is only noticeable later in life; Canton-S Dα6 KO mutants have a median survival of 68 days compared to 82 for Canton-S (Figure 7D). A 25-day exposure to 0.2 ppm spinosad leads to a 91% survival of Canton-S Dα6 KO mutants, whereas only 40% of Canton-S wild-type flies survive this exposure (Figure 7E). No changes in BS or climbing ability were observed between exposed and unexposed Dα6 KO mutants over the course of a 20-day exposure (Figure 7F and G). However, at the 20-day time point, twice as many of the unexposed Dα6 KO mutants failed to climb (36%) compared to unexposed Canton-S wild-type flies (18%). (Figure 7G). These data support that the deleterious effects of spinosad are mediated by its binding to Dα6 receptors. They also indicate that loss of Dα6 by itself causes mild but significant behavioral and life span phenotypes not previously reported.

We then examined the retinas of adult flies for evidence of neurodegenerative markers, such as the accumulation of LD based on Nile Red staining (Liu et al., 2015). Adult virgin females (3–4-day-old) exposed to 5 ppm of spinosad for 6 hr showed a significant accumulation of LD in the glial cells of the retina (Figure 8A and B), indicating the ability of spinosad to induce another ROS-related phenotype (Liu et al., 2017) within a few hours of exposure. Chronic exposures to 0.2 ppm, however, did not lead to a clear LD phenotype in glial cells of retinas. However, Nile Red-positive accumulations were observed decorating the plasma membrane of photoreceptor neurons (PR) after 10 and 20 days of exposure (Figure 8C and D). Even though Dα6 is not expressed in the retina, it is widely expressed in the adult brain, notably in the lamina neurons that synapse with the PR (Figure 9A). The accumulation of lipids in neurons suggests that the postsynaptic cells that express Dα6 somehow affect lipid production or transfer to PR.

Figure 8
Download asset Open asset
Chronic exposure to spinosad causes lipid deposits in retinas.

(A) Nile Red staining of lipid droplets in the retinas of virgin females exposed to 5 parts per million (ppm) spinosad for 6 hr. Cluster of rhabdomeres delimited with yellow dotted line, purple arrowheads point to lipid droplets. (B) Number of lipid droplets per ommatidium (n = 5 flies/treatment; three image sections/retina). (C) Nile Red staining of lipid deposits in retinas of virgin females exposed to 0.2 ppm spinosad for 1, 10, and 20 days. Green arrowheads point to lipid deposits. (D) Percentage of flies with lipid deposits in retinas (n = 8 flies/time point/treatment). (E) Nile Red staining of defective rhabdomeres in retinas of virgin females Canton-S and Canton-S Dα6 KO mutants 1, 10, and 20 days old. Yellow arrowheads point to defective rhabdomeres. (F) Percentage of flies that show defective rhabdomeres (n = 8 flies/time point/genotype). Microscopy images were obtained with a Leica TCS SP8 Laser Scanning Confocal Microscope. (B) Student’s unpaired t-test; (D, F) one-way ANOVA followed by Tukey’s honestly significant difference test. *p<0.05, **p<0.01, ***p<0.001.

Figure 9
Download asset Open asset
Chronic exposure to spinosad impairs the visual system.

(A) Expression pattern of Dα6 in the Drosophila adult female brain (Dα6 T2A Gal4>UASGFP.nls). Detail of the expression in lamina and medulla (optic lobe). OL,-optic lobe; CB, central brain. (B) Electroretinograms (ERGs) of virgin females exposed to 0.2 parts per million (ppm) spinosad for 1, 10, and 20 days. Red dotted circles indicate the on-transient signal, and green arrow indicates the amplitude. (C) Measurements of on-transient signal and amplitude after 1, 5, 10, 15, and 20 days of exposure to 0.2 ppm spinosad (n = 8–10 flies/time point/treatment). (D) ERGs of virgin females Canton-S and Canton-S Dα6 KO mutants 5, 10, and 20 days old. (E) Measurements of on-transient signal and amplitude in Canton-S and Canton-S Dα6 KO mutants (n = 8–10 flies/time point/genotype). Microscopy images were obtained with a Leica TCS SP8 Laser Scanning Confocal Microscope. (C, E) One-way ANOVA followed by Tukey’s honestly significant difference test. *p<0.05, **p<0.01, ***p<0.001.

The retinas of unexposed Dα6 KO mutants were also examined. Adult virgin females, 1, 10, and 20 days old, showed no Nile Red-positive accumulations in retinas. However, 37% of 10-day-old mutants and 50% of 20-day-old mutants showed abnormal rhabdomeres (Figure 8E and F). These visual system defects have not been described previously in a Dα6 KO mutant and are obviously different from the ones arising from the interaction between Dα6 and spinosad.

Given the Nile Red-positive accumulation in retinas of chronically exposed flies, we investigated the impact on visual function. Electroretinograms (ERGs) were performed at regular intervals over the 20 days of exposure (Figure 9B and C). ERG recordings measure the impact of light pulses on PR. The on-transient is indicative of synaptic transmission between PR and postsynaptic cells, whilst the amplitude measures the ability of photons to impact the phototransduction cascade (Wang and Montell, 2007). A large reduction in the on-transient was observed from day 1 of exposure, whereas the amplitude was only significantly impacted after 20 days of exposure (Figure 9B and C). The reduction in the on-transient is evidence of a rapid loss of synaptic transmission in laminar neurons (Wang and Montell, 2007) and hence impaired vision after just 1 day of exposure. In examining the visual system of Dα6 KO mutants reared without spinosad, mild impacts were identified in ERG amplitude but a very significant reduction in on-transient was also observed, consistent with a requirement for Dα6 in lamina postsynaptic cells of the PR (Figure 9D and E).

To investigate the ultrastructure of the PR synapses, we used transmission electron microscopy. Compared to unexposed flies (Figure 10A), severe morphological alterations were detected in transverse sections of the lamina of flies exposed for 20 days (Figure 10B–F). Vacuoles in photoreceptor terminals or postsynaptic terminals of lamina neurons were observed in the lamina cartridges (Figure 10B). On average, 75% of the lamina cartridges contained vacuoles (Figure 10E). Large intracellular compartments were also observed in dendrites of the postsynaptic neurons in the lamina (Figure 10C and D). These do not correspond to normal structures found in healthy lamina (Figure 10A) and suggest the presence of aggregated material. The lamina of exposed flies also showed a mean 34% increase in the number of mitochondria (Figure 10F), many of which appear defective (Figure 10B). No morphological alterations were detected by TEM in lamina of 20-day-old Dα6 KO mutants reared in the absence of spinosad (Figure 10G and H).

Figure 10
Download asset Open asset
Chronic exposure to spinosad leads to neurodegeneration.

(A–D) Transmission electron microscopy (TEM) of the lamina of virgin females exposed to 0.2 parts per million (ppm) spinosad for 20 days. (A) A regular cartridge of a control fly; blue arrowheads indicate normal mitochondria. (B) Cartridge of an exposed fly; pink arrowhead indicates a vacuole and green arrowhead indicates a defective mitochondrion. (C, D) Cartridges of exposed flies indicating an enlarged digestive vacuole (yellow arrowhead) and the presence of large intracellular compartments (red arrowheads). (E) Percentage of images showing vacuoles in lamina cartridges of exposed flies (10 images/fly; three flies/treatment). (F) Number of mitochondria per cartridge of exposed flies (n = 3 flies/treatment; 16 cartridges/fly). (G) TEM of the lamina of virgin 20-day-old females Canton-S and Canton-S Dα6 KO mutant. (H) Number of mitochondria per cartridge (n = 3 flies/genotype; 16 cartridges/fly). (I) Hematoxylin and eosin (H&E) staining of adult brain of virgin females exposed to 0.2 ppm spinosad for 20 days. (J) Quantification of neurodegeneration in terms of percentage of brain area vacuolated (n = 3 flies/treatment). (E, F, H, J) Student’s unpaired t-test. **p<0.01, ***p<0.001.

Lastly, hematoxylin and eosin stain (H&E) of adult flies painted a broader picture of the neurodegeneration caused outside the visual system by chronic low-dose exposure to spinosad. 20 days of exposure caused numerous vacuoles in the central brain (Figure 10I). On average, 17% of the total central brain area was consumed by vacuoles in exposed flies (Figure 10J).

Discussion

Spinosad antagonizes neuronal activity

In this study, we explore the mechanisms and consequences of exposure to low doses of spinosad upon binding to its target Dα6, comparing these phenotypes side by side with the ones caused by Dα6 loss of function. Low levels of spinosad lead to a lysosomal dysfunction associated with mitochondrial stress, elevated levels of ROS, lipid mobilization defects, and severe neurodegeneration. Spinosad has been characterized as an allosteric modulator of the activity of its primary target, the Dα6 subunit, causing fast neuron over-excitation (Salgado, 1998). Here, the capacity of spinosad to interact with its target to stimulate the flux of Ca2+ into neurons was quantified. The results obtained with the GCaMP assay showed that spinosad caused no detectable increase or decrease in Ca2+ flux into Dα6-expressing neurons, but reduces the cholinergic response (Figure 1D and E). Given that spinosad binds to the C terminal region of the protein (Crouse et al., 2018; Puinean et al., 2013; Somers et al., 2015), these findings are consistent with a noncompetitive antagonist mode of action for spinosad on nAChRs.

Spinosad’s toxicity involves more than causing an absence of Dα6 from neuronal membranes

Dα6 loss-of-function mutants are viable and highly resistant to spinosad (Figure 1C; Perry et al., 2007), yet the loss of Dα6 from the membrane precipitated by exposure to spinosad in wild-type flies leads to death. This creates a conundrum that can be resolved if a significant component of spinosad’s toxicity is due to molecular events that play out elsewhere in the cell. Blocked neuronal receptors can be recycled from the plasma membrane through endocytosis (Saheki and De Camilli, 2012). Here, we demonstrate that Dα6 is removed from neuronal membranes in response to spinosad exposure (Figure 2A) and localizes to lysosomes (Figure 2G), resulting in a significant lysosomal expansion (Figure 2C) and increase in ROS levels (Figure 4A and B). In exposed Dα6 mutants, spinosad does not lead to lysosome expansion (Figure 2D) or an increase in ROS levels in the brain (Figure 4—figure supplement 1). These two phenotypes precede all other phenotypes observed in wild-type exposed larvae. In unexposed mutants, the mild ROS levels found in brains (Figure 4—figure supplement 1) seem to be a direct consequence of Dα6 absence. Indeed, Dα6 has been associated with the response to oxidative stress and Dα6 mutants are more susceptible to oxidative damage (Weber et al., 2012). The mildly elevated ROS levels in unexposed Dα6 mutants cannot be ignored, nor can the altered lipid environment (Figure 5A, B, and G), a minor reduction in longevity and increased climbing defects with age (Figure 7D and G), rhabdomere degeneration (Figure 8E and F), as well as loss of synaptic transmission in photoreceptors (Figure 9D and E). These are all previously unreported metabolic and nervous system defects associated with Dα6 loss of function.

Spinosad causes lysosomal dysfunction and mitochondrial stress

Lysosomal dysfunction and mitochondrial stress are the key players in the cascade of impacts following spinosad exposure. Whether spinosad molecules are ferried to lysosomes along with Dα6 subunits and accumulate into these organelles remains to be clarified. However, that the increased severity in the lysosomal phenotype after exposure ceases (Figure 2C and E) is consistent with the poisoning of these organelles. Lysosomes become enlarged as they accumulate undigested material, which typically lead to recycling problems (Darios and Stevanin, 2020). If spinosad remains bound to the receptor and is ferried into the lysosomes, it may contribute to a lysosomal dysfunction akin to lysosomal storage disease (LSD) (Darios and Stevanin, 2020). To date, there is little published evidence of spinosad metabolites in insects. Spinosyns are polyketide macrolactones, and we speculate that their complex molecular structure may prevent them from being degraded efficiently by metabolic enzymes in lysosomes, triggering a severe lysosomal dysfunction and expansion.

Extensive evidence connects LSDs with mitochondrial dysfunction (Plotegher and Duchen, 2017; Stepien et al., 2020; Yambire et al., 2018). Mitochondrial dysfunction is widespread in LSD and is involved in its pathophysiology, although the exact mechanisms remain unclear. Lysosomal disorders may lead to cytoplasmic accumulation of toxic macromolecules like ceramides (Lin et al., 2018), impaired mitophagy and dysregulation of intracellular Ca2+ homeostasis, resulting in an increase in ROS generation and reduced ATP levels (Plotegher and Duchen, 2017). The severe lysosomal dysfunction observed here is a likely cause for mitochondrial dysfunction and increased ROS generation triggered by spinosad exposure. Treatment with antioxidant NACA was able to prevent the increase in ROS levels at the 2 hr time point exposure (Figure 4F and G) but did not prevent the lysosome expansion (Figure 4H and I). That lysosomes still expand in the absence of the ROS generated by mitochondria supports the notion that it precedes and triggers the mitochondrial stress (Figure 3A–E).

While we cannot rule out the generation of ROS by other mechanisms, we provide compelling evidence that a significant part of ROS that is generated by spinosad exposure is of mitochondrial origin, arguing that mitochondrial impairment is a key element of spinosad mode of action at low-dose exposure. The evidence is based on increased mitochondrial turnover and mito-roGFP 405/488 ratio, reduced activity of the ROS sensitive enzyme m-aconitase, and reduced ATP levels (Figure 3). In addition, we observed a highly significant reduction of CL levels (Figure 6C) typically associated with defects in the electron transport chain and increased ROS production as they are required for the anchoring of Complex 1 in mitochondria (Dudek, 2017; Quintana et al., 2010) Increased levels of ROS in the larval brain have been shown to disturb mitochondrial function triggering changes in lipid stores in metabolic tissues (Martelli et al., 2020). Oxidative stress promotes the redistribution of membrane lipids into LD, reducing their exposure to peroxidized lipids (Bailey et al., 2015; Liu et al., 2015). Here, we observed increases in lipid stores in the fat body (Figure 5A and B), reduction in LD numbers in the Malpighian tubules (Figure 5—figure supplement 2), and changes in lipid levels in the hemolymph (Figure 5G). Our lipidome analysis also revealed reduction of PE and PC levels (Figure 6A, Figure 6—source data 1), consistent with impaired membrane fluidity and altered LD dynamics (Dawaliby et al., 2016; Guan et al., 2013; Krahmer et al., 2011).

The use of the antioxidant NACA reduces the accumulation of LD in the fat body linking this phenotype to oxidative stress (Figure 5C and D). NACA also diminished spinosad toxicity by reducing the impact on larval movement and survival (Figure 3F and G). Exposure to spinosad (7.7 parts per billion for 24 hr) was previously shown to cause the vacuolation of epithelial cells of the midgut and Malpighian tubules of honeybees (Apis mellifera) (Lopes et al., 2018). It is not clear whether this is due to the spinosad:Dα6 interaction precipitating elevated levels of ROS. While the dysfunction of lysosomes and mitochondria and elevated levels of ROS can account for the defects observed here under conditions of spinosad exposure, we cannot rule out this insecticide having other impacts that would contribute to its toxicity. Yet, the loss of Dα6 strongly suppresses the phenotypes caused by exposure. Given that Dα6 is not expressed in the gut and fat body, this suggests that the observed brain defects are at the root of most observed defects.

The LSD-like dysfunction is also likely the underlying cause for the severe vacuolation of adult central brain under spinosad chronic exposure (Figure 10I and J). Recycling defects in neuronal cells caused by LSD impair cell function, ultimately triggering neurodegeneration (Darios and Stevanin, 2020). A model for the low-dose mode of action of spinosad that is consistent with the data presented here is illustrated in Figure 11.

Figure 11
Download asset Open asset
Proposed mechanism for internalization of spinosad after binding to Dα6 targets.

(A) Spinosad binds to Dα6 subunit of nicotinic acetylcholine receptors (nAChRs) in the neuronal cell membranes. (B) The binding of spinosad leads to Dα6-containing nAChR blockage, endocytosis, and trafficking to lysosome. (C) Spinosad accumulates in lysosomes, while receptors and other membrane components are digested. (D) Expanded lysosomes due to accumulation of undigested material do not function properly, leading to cellular defects that may include mitochondrial dysfunction, increased mitochondrial reactive oxygen species (ROS) production, and eventually cell vacuolation and neurodegeneration.

Spinosad causes neurodegeneration and affects behavior in adults

Both LSD (Darios and Stevanin, 2020) and oxidative stress (Liu et al., 2017; Martelli et al., 2020) can cause neurodegeneration. The evidence for spinosad-induced neurodegeneration comes from the reduced climbing ability and increased BS caused by chronic low-dose exposures (Figure 7B and C), vacuolation of the lamina cartridges, and severe vacuolation of the adult CNS (Figure 10). The neurodegeneration observed in the central brain (Figure 10I and J) seems to be largely contained to the functional regions of the optic tubercle, mushroom body, and superior lateral and medial protocerebrum. Dα6 is widely expressed in the brain, including these regions (Davie et al., 2018; Li et al., 2021). These regions are important centers for vision and memory, and learning and cognition in flies (Schürmann, 2016). Neurodegeneration in these regions indicates that a wide range of behaviors will be compromised in exposed flies.

Dα6 is not known to be expressed in PRs or glial cells, but its expression in lamina neurons (Figure 9A) supports its presence in postsynaptic cells of PR. The Nile Red-positive accumulations in PRs of wild-type flies after chronic spinosad exposure (Figure 8C) suggest the existence of cell nonautonomous mechanisms initiated by this insecticide in postsynaptic cells. Liu et al., 2017 showed that ROS induce the formation of lipids in neurons that are transported to glia, where they form LD. Here, a ROS signal generated by spinosad exposure in postsynaptic cells might be carried to PRs, affecting lipid metabolism and triggering LD accumulation. This hypothesis needs further investigation.

Rational control of insecticide usage

In the public domain, organic insecticides are often assumed to be safer than synthetic ones for the environment and nontarget insect species. The synthetic insecticide, imidacloprid, has faced intense scrutiny and bans because of its impact on the behavior of bees and the potential for this to contribute to the colony collapse phenomenon (Wu-Smart and Spivak, 2016). No other insecticide has been so comprehensively investigated, so it is not yet clear whether other chemicals pose similar risks. This study has revealed disturbing consequences of low doses of an organic insecticide, spinosad. Based on similar assays deployed here, imidacloprid had a lower negative impact on Drosophila than spinosad (Martelli et al., 2020). At the low acute dose used here (2.5 ppm for 2 hr), imidacloprid has no effect on larval survival, while spinosad exposure is lethal. Given differences in the molecular weight, spinosad has a greater biological impact at lower concentration. 2.5 ppm corresponds to 3.4 µM spinosad and 9.8 µM imidacloprid. 4 ppm imidacloprid causes blindness and neurodegeneration, but no brain vacuolation under conditions of chronic exposure (Martelli et al., 2020), whereas 0.2 ppm spinosad causes vision loss and widespread brain vacuolation. Loss of function of Dα6 caused by mutation or by chronic exposure to spinosad leads to vision loss. This suggests that a wider analysis of Dα6 mutant phenotypes may point to other consequences of spinosad exposure not detected in this study. Given that the Dα6 subunit has been shown to be a highly conserved spinosad target across a wide range of insects (Perry et al., 2015), it is likely that low doses of this insecticide will have similar impacts on other species. The susceptibility of different species to insecticides varies, so the doses required may differ. The protocols used here will be useful in assessing the risk that spinosad poses to other beneficial insects. Given the extent to which spinosad affects lysosomes, mitochondrial function, lipid metabolism, and vision, this insecticide very likely compromises the capacity of insects to survive in natural environments when exposed to a variety of stresses, including some of those that are being linked to insect population declines (Cardoso et al., 2020; Sánchez-Bayo and Wyckhuys, 2019).

Two clocks are ticking. The global human population is increasing, and the amount of arable land available for food production is decreasing. Thus, the amount of food produced per hectare needs to increase. Our capacity to produce enough food has been underpinned by the use of insecticides. Approximately 600,000 tonnes of insecticides are used annually around the world (Aizen et al., 2009; Klein et al., 2007), but sublethal concentrations found in contaminated environments can affect behavior, fitness, and development of target and nontarget insects (Müller, 2018). Despite their distinct modes of action, spinosad and imidacloprid produce a similar spectrum of damage (Martelli et al., 2020). This similarity arises because both insecticides trigger oxidative stress in the brain, albeit via different mechanisms. Several other insecticide classes such as organochlorines, organophosphates, carbamates, and pyrethroids have all been shown to promote oxidative stress (Balieira et al., 2018; Karami-Mohajeri and Abdollahi, 2011; Lukaszewicz-Hussain, 2010; Terhzaz et al., 2015; Wang et al., 2016). Many insect populations are exposed to a continuously changing cocktail of insecticides (Kerr, 2017; Tosi et al., 2018), most of which are capable of producing ROS. The cumulative effect of these different insecticides could be significant. Our research clarifies the mode of action of spinosad, highlighting the perturbations and damage that occur downstream of the insecticide:receptor interaction. Other chemicals should not be assumed to be environmentally safe until their low-dose biological impacts have been examined in similar detail.

Materials and methods

Key resources table
Reagent type (species) or resourceDesignationSource or referenceIdentifiersAdditional information
Genetic reagent(Drosophila melanogaster)Armenia60Drosophila
Genomics
Resource
Center		DGRC #103394Line14 is an isofemale line derived from Armenia60
Genetic reagent(D. melanogaster)nAChRα6 T2A Gal4Bloomington
Drosophila
StockCenter		BDSC #76137RRID:BDSC_76137
Genetic reagent(D. melanogaster)UAS-GFP.nlsBloomington
Drosophila
StockCenter		BDSC #4775RRID:BDSC_4775
Genetic reagent(D. melanogaster)mito-roGFP2-Orp1Bloomington
Drosophila
StockCenter		BDSC #67672RRID:BDSC_67672
Genetic reagent(D. melanogaster)UAS-Sod2Bloomington
Drosophila
StockCenter		BDSC #24494RRID:BDSC_24494
Genetic reagent(D. melanogaster)UAS-Sod1Bloomington
Drosophila
StockCenter		BDSC #24750RRID:BDSC_24750
Genetic reagent(D. melanogaster)Elav-Gal4Bloomington
Drosophila
StockCenter		BDSC #458RRID:BDSC_458
Genetic reagent(D. melanogaster)Canton-SBloomington
Drosophila
StockCenter		BDSC #64349RRID:BDSC_64349
Genetic reagent(D. melanogaster)Canton-S Dα6 KO;Canton-S nAChRα6 knockoutThis paper Mutant strain generated by CRISPR and maintained in T. Perry Lab
Genetic reagent(D. melanogaster)Line14 Dα6 loss-of-function mutant;Line14 Dα6nxPerry et al., 2015 (doi:10.1016/j.ibmb.2015.01.017) Mutant strain generated by EMS and maintained in T. Perry Lab
Genetic reagent(D. melanogaster)GCaMP5G:tdTomato cytosolic [Ca2+] sensorBloomington
Drosophila
StockCenter		BDSC #80079RRID:BDSC_80079
Chemical compound,
drug		SpinosadSigma-AldrichProduct #33706 
Chemical compound,
drug		Antioxidant N-acetylcysteine amide;NACALiu et al., 2015 (doi:10.1016/j.cell.2014.12.019) Provided by
Hugo J. Bellen Lab
Chemical compound, drugDHESigma-AldrichProduct #D7008 
Chemical compound,
drug		Nile RedSigma-AldrichProduct #N3013 
Chemical compound,
drug		LysoTracker Red DND-99 (1:10,000)InvitrogenCat #L7528 
Commercial
assay, kit		Mitochondrial aconitase
activity kit		Sigma-AldrichProduct #MAK051 
Commercial
assay, kit		ATP assay kitAbcamProduct #ab83355 

Fly strains and rearing

Request a detailed protocol

Line14 (Perry et al., 2008), an isofemale line derived from Armenia60 (currently named Aashtrak, Drosophila Genomics Resource Center #103394), was used as the susceptible wild-type line for all assays except the following. GCaMP experiment: UAS-tdTomato-P2A-GCaMP5G (III) (Daniels et al., 2014; Wong et al., 2014) was crossed with nAChRα6 T2A Gal4 (BDSC #76137). Expression of nAChRα6 gene in adult brains: nAChRα6 T2A Gal4 (BDSC #76137) was crossed with UAS-GFP.nls (BDSC #4775). Insecticide impact on mitochondrial turnover: the MitoTimer line (Laker et al., 2014) was used. Insecticide impact on mitochondrial ROS generation: the mito-roGFP2-Orp1 line (BDSC #67672) (Albrecht et al., 2011) was used. Overexpressing Sod2 and Sod1 in the central nervous system: UAS-Sod2 strain (BDSC #24494) and UAS-Sod1 strain (BDSC #24750) were crossed with a Elav-Gal4 driver (BDSC #458). GCaMP experiment: UAS-tdTomato-P2A-GCaMP5G (III) (Daniels et al., 2014; Wong et al., 2014) was crossed with nAChRα6 T2A Gal4 (BDSC #76137). Two mutants for the nAChRα6 gene, which confers resistance to spinosad (Perry et al., 2015) and their background control lines, were used to dissect the differences between phenotypes caused by spinosad mode of action and phenotypes caused exclusively by nAChRα6 loss of function. The first of these is Line14 Dα6nx strain, a loss-of-function mutant recovered following EMS mutagenesis in the Line14 genetic background, with no detectable nAChRα6 expression (Perry et al., 2015). The second mutant is a CRISPR knockout of nAChRα6 generated in the Canton-S genetic background (Luong, 2018). For experiments aiming to investigate the trafficking of nAChRα6 in brains, UAS Dα6 CFP tagged strain built in Line14 Dα6nx background was crossed to a native Dα6 driver (Gal4-L driver) in the same background (Perry et al., 2015). For experiments involving larvae, flies were reared on standard food media sprinkled with dried yeast and maintained at 25°C. Early third-instar larvae were used in all experiments involving larval stage. For experiments involving adults, flies were reared in molasses food and maintained at 25°C. In all experiments involving adult flies, only virgin females were used to maintain consistency.

Generation of CFP tagged Dα6 subunit

Request a detailed protocol

To create the CFP tagged Dα6 subunit for expression, a sequential PCR strategy was used to introduce the tag within the TM3-TM4 cytoplasmic loop region. Amplification of the pCyPet-His plasmid (Addgene #14030) with primers adding Gly-Ala-Gly and flanking homology arms to the Dα6 insertion site (A385F_CFP_YFP and A385R_CFP) was performed. This fragment was purified and combined in a PCR reaction with a wild-type cDNA clone of Dα6 (Perry et al., 2015) and reverse primer (da6_cloneR) to produce a fusion product. This fusion fragment was purified and combined in a PCR reaction with a wild-type cDNA clone of Dα6 (Perry et al., 2015) and forward primer (da6_cloneF) to amplify a fragment encoding the full-length Dα6 protein with an incorporated CFP tag (Dα6CFP – sequence provided below), and this was cloned into pUAST (Bischof et al., 2007). Following this, as per Nguyen et al., 2021, transgenic flies in the correct genetic background were made using microinjection into y w M{eGFP.vas-int.Dm}ZH-2A; Dα6nx; M{RFP.attP}ZH-86Fb (Bischof et al., 2007; Nguyen et al., 2021), cre-recombinase was used to excise the 3XP3 RFP and miniwhite regions of the genomic insertion flanked by lox-P sites, and to control expression in this study, Dα6nx; Dα6CFP flies were crossed to a native Dα6 GAL4 driver w; Dα6>GAL4; Dα6nx (Perry et al., 2015).

Primer sequences

Request a detailed protocol

  • >A385F_CFP_YFP

  • CCTCCAAATCCCTGCTGGCCGGAGCAGGAATGTCTAAAGGTGAAGAATT

  • >A385R_CFP

  • TCGTCGATGTCGAGGACATTTCCTGCTCCTTTGTACAATTCATCCATAC

  • >da6_cloneF

  • GTAGCCATTCAACCCGAGAG

  • >da6_cloneR

  • GCTTCCGACGTATCCGTAGC

  • Dα6CFP – nucleotide sequence

  • GTAGCCATTCAACCCGAGAGCCACGCGATACAAACAAGCCAAGGACATGGACTCCCCGCTGCCAGCGTCGCTGTCGCTGTTTGTCCTGTTGATCTTTCTGGCGATAATTAAAGAAAGCTGTCAAGGACCTCATGAAAAGCGCCTGCTGAACCATCTGCTGTCCACCTACAATACGCTGGAGCGACCCGTGGCCAATGAATCGGAGCCCCTGGAGATTAAGTTCGGACTGACGCTGCAGCAGATCATCGACGTGGACGAGAAGAATCAGCTTCTCATAACGAATCTTTGGCTTTCGTTGGAGTGGAACGACTACAATCTGCGCTGGAATGAAACGGAATACGGCGGGGTCAAGGATCTACGAATCACGCCCAACAAGCTGTGGAAGCCCGACGTGCTCATGTACAACAGCGCGGATGAGGGATTCGATGGCACGTATCACACCAACATTGTGGTCAAACATGGCGGCAGTTGTCTGTACGTGCCCCCTGGTATCTTCAAGAGCACATGCAAGATGGACATCACGTGGTTCCCATTTGATGACCAACATTGCGAAATGAAATTCGGTAGTTGGACTTACGATGGAAATCAGTTGGATTTGGTTTTGAGTTCCGAAGATGGAGGGGATCTTTCCGATTTCATAACAAATGGCGAGTGGTACTTGCTTGCCATGCCGGGAAAGAAGAATACGATAGTCTACGCCTGCTGCCCAGAACCATATGTCGATATCACCTTTACTATACAAATTCGTCGCCGTACATTATATTATTTTTTCAATTTAATCGTGCCATGTGTGCTAATCTCATCGATGGCCCTACTGGGCTTCACATTGCCGCCGGATTCGGGCGAGAAACTGACGCTGGGAGTTACAATTCTTCTATCGCTCACAGTGTTTCTCAACCTTGTAGCTGAGACATTGCCCCAAGTATCTGATGCAATCCCCTTGTTAGGCACCTACTTCAATTGCATCATGTTCATGGTCGCATCGTCGGTGGTGCTGACAGTAGTGGTGCTCAACTACCACCATCGCACAGCGGACATTCACGAGATGCCACCGTGGATCAAGTCCGTTTTCCTACAATGGCTGCCCTGGATCTTGCGAATGGGTCGACCCGGTCGCAAGATTACACGCAAAACAATACTATTAAGCAATCGCATGAAGGAGCTGGAGCTAAAGGAGCGCCCCTCCAAATCCCTGCTGGCCGGAGCAGGAATGTCTAAAGGTGAAGAATTATTCGGCGGTATCGTCCCAATTTTAGTTGAATTAGAGGGTGATGTTAATGGTCACAAATTTTCTGTCTCCGGTGAAGGTGAAGGTGATGCTACGTACGGTAAATTGACCTTAAAATTTATTTGTACTACTGGTAAATTGCCAGTTCCATGGCCAACCTTAGTCACTACTCTGACTTGGGGTGTTCAATGTTTTTCTAGATACCCAGATCATATGAAACAACATGACTTTTTCAAGTCTGTCATGCCAGAAGGTTATGTTCAAGAAAGAACTATTTTTTTCAAAGATGACGGTAACTACAAGACCAGAGCTGAAGTCAAGTTTGAAGGTGATACCTTAGTTAATAGAATCGAATTAAAAGGTATTGATTTTAGAGAAGATGGTAACATTTTAGGTCACAAATTGGAATACAACTATATCTCTCACAATGTTTACATCACCGCTGACAAACAAAAGAATGGTATCAAAGCTAACTTCAAAGCCAGACACAACATTACCGATGGTTCTGTTCAATTAGCTGACCATTATCAACAAAATACTCCAATTGGTGATGGTCCAGTCATCTTGCCAGACAACCATTACTTATCCACTCAATCTGCCTTATCTAAAGATCCAAACGAAAAGAGAGACCACATGGTCTTGCTCGAATTTGTTACTGCTGCTGGTATTACCCATGGTATGGATGAATTGTACAAAGGAGCAGGAAATGTCCTCGACATCGACGACGACTTTCGGCACACAATATCTGGCTCCCAAACCGCCATTGGCTCGTCGGCCAGCTTCGGTCGGCCCACAACGGTGGAGGAGCATCACACGGCCATCGGCTGCAATCACAAAGATCTTCATCTAATTCTTAAAGAATTGCAATTTATTACGGCGCGGATGCGCAAAGCTGACGACGAAGCGGAATTGATCGGCGATTGGAAGTTCGCGGCAATGGTTGTGGATAGATTTTGTTTAATTGTTTTCACGCTCTTCACGATTATTGCAACGGTTACGGTGCTGCTCTCCGCTCCGCACATAATCGTGCAATAAGGACGCTCGAATTAGGCCATTAAGCTACGGATACGTCGGAAGC

Insecticide dilution and exposure

Request a detailed protocol

Pure spinosad (Sigma-Aldrich) was used in all assays. The chemical was diluted to create 1000 ppm stocks solution, using dimethyl sulfoxide (DMSO), and was kept on freezer (–20°C). Before exposures, 5× stocks were generated for the dose being used by diluting the 1000 ppm stock in 5% Analytical Reagent Sucrose (Chem Supply) solution (or equivalent dose of DMSO for controls).

Antioxidant treatment

Request a detailed protocol

The antioxidant, NACA, was used as previously described (Martelli et al., 2020). Briefly, larvae were treated with 300 µg/mL of NACA in 5% Analytical Reagent Sucrose (Chem Supply) solution for 5 hr prior to exposure to spinosad exposures.

Fly media used

Request a detailed protocol
Standard food (1 L)Apple juice plates (1 L)Molasses food (1 L)
H2O987 mLH2O720 mLH2O800 mL
Potassium tartrate8.0 gAgar20 gMolasses160 mL
Calcium chloride0.5 gApple juice200 mLMaize meal60 g
Agar5.0 gBrewer’s yeast7.0 gDried active yeast15 g
Yeast12 gGlucose52 gAgar6.0 g
Glucose53 gSucrose26 gAcid mix7.5 mL
Sucrose27 gTegosept6.0 mLTegosept5.0 mL
Semolina67 g  
Acid mix12 mL  
Tegosept15 mL  

Larval movement assay

Request a detailed protocol

Larval movement in response to insecticide exposure was quantified by Wiggle Index Assay, as described by Denecke et al., 2015. 25 third-instar larvae were used for a single biological replicate and 4 replicates were tested for each exposure condition. Undosed larvae in NUNC cell plates (Thermo Scientific) in 5% Analytical Reagent Sucrose (Chem Supply) solution were filmed for 30 s, and then 30 min, 1 hr, 1 hr, and 30 min and 2 hr after spinosad exposure. The motility at each time point is expressed in terms of relative movement ratio (RMR), normalized to motility prior to spinosad addition.

Larval viability and adult survival tests

Request a detailed protocol

For all tests, five replicates of 20 individuals (100 individuals) per condition were used. In assessing third-instar larval viability and metamorphosis following insecticide exposure, individuals were rinsed three times with 5% w/v sucrose (Chem Supply) and placed in vials on insecticide-free food medium. Differences between adult eclosion rates were analyzed using Student’s unpaired t-test. Correct percentage survival of larvae exposed to 0.5 ppm spinosad for 2 hr, or 0.1 ppm spinosad for 4 hr, was analyzed using Abbot’s correction. To examine the survival of adult flies chronically exposed to 0.2 ppm spinosad, five replicates of 20 females (3–5 days old) were exposed for 25 days. The same number of flies was used for the control group. Statistical analysis was based on the Kaplan–Meier method, and data were compared by the log-rank Mantel–Cox test.

GCaMP assay

Request a detailed protocol

Cytosolic [Ca2+] in Drosophila primary neurons was measured as previously described (Martelli et al., 2020). Briefly, four brains from third-instar larvae were dissected to generate ideal number of cells for three plates. Cells were allowed to develop in culture plates (35 mm glass-bottom dishes with 10 mm bottom well [Cellvis], coated with concanavalin A [Sigma]) with Schneider’s media for 4 days with the media refreshed daily. Recording was done using a Nikon A1 confocal microscope, ×40 air objective, sequential 488 nm and 561 nm excitation. Measurements were taken at 3 s intervals. Cytosolic Ca2+ levels were reported as GCaMP5G signal intensity divided by tdTomato signal intensity. Signal was recorded for 60 s before the addition of 2.5 ppm or 25 ppm spinosad to the bath solution. 5 min after that, both insecticide and control groups were stimulated by the cholinergic agonist carbachol (100 µM) added to the bath solution, and finally, the SERCA inhibitor thapsigargin (5 µM) was added after a further 1 min. At least 50 neuronal cells were evaluated per treatment. The data were analyzed using one-way ANOVA followed by Tukey’s honestly significant difference test.

Evaluation of mitochondrial turnover

View detailed protocol

Mitochondrial turnover was assessed as previously described (Martelli et al., 2020). Larvae of the MitoTimer line were exposed to 2.5 ppm spinosad for 2 hr. Control larvae were exposed to 2.5 ppm DMSO. Midguts and brains were dissected in PBS and fixed in 4% paraformaldehyde (PFA) (Electron Microscopy Sciences) and mounted in VECTASHIELD (Vector Laboratories). 20 anterior midguts and 20 pairs of optical lobes were analyzed for each condition. Confocal microscopy images were obtained with a Leica SP5 Laser Scanning Confocal Microscope at ×200 magnification for both green (excitation/emission 488/518 nm) and red (excitation/emission 543/572 nm) signals. Three independent measurements along the z stack were analyzed for each sample. Fluorescence intensity was quantified on ImageJ software, and data were analyzed using one-way ANOVA followed by Tukey’s honestly significant difference test.

Evaluation of mitochondrial ROS generation using Mito-roGFP2-Orp1

Request a detailed protocol

The mito-roGFP2-Orp1 (BDSC #67672) was used to measure the production of mitochondrial H2O2 (Albrecht et al., 2011). Larvae were exposed to 2.5 ppm spinosad for 2 hr (controls exposed to 2.5 ppm DMSO). Anterior midguts and brains were dissected in Schneider’s media (Gibco) and immediately mounted in VECTASHIELD (Vector Laboratories) for image acquisition. Confocal microscopy images were obtained with a Leica SP5 Laser Scanning Confocal Microscope under excitation/emission 488/510 nm (reduced) or 405/510 nm (oxidized). Three independent measurements along the z stack were analyzed for each sample. Fluorescence intensity was quantified on ImageJ software, and data were analyzed using Student’s unpaired t-test.

Systemic mitochondrial aconitase activity

Request a detailed protocol

Relative mitochondrial aconitase activity was quantified using the colorimetric Aconitase Activity Assay Kit from Sigma (#MAK051), following the manufacturer’s instructions as previously described (Martelli et al., 2020). A total of six biological replicates (25 whole larvae per replicate) were exposed to 2.5 ppm spinosad for 2 hr, whilst six control replicates (25 whole larvae per replicate) were exposed to DMSO for 2 hr. Absorbance was measured at 450 nm in a FLUOstar OPTIMA (BMG LABTECH) microplate reader using the software OPTIMA and normalized to sample weight. The data were analyzed using one-way ANOVA followed by Tukey’s honestly significant difference test.

Systemic ATP levels

Request a detailed protocol

Relative ATP levels were quantified fluorometrically using an ATP assay kit (Abcam #83355), following the manufacturer’s instructions as previously described (Martelli et al., 2020). A total of six biological replicates (20 larvae per replicate) were exposed to 2.5 ppm spinosad for 2 hr, whilst six control replicates (20 larvae per replicate) were exposed to DMSO for 2 hr. Fluorescence was measured as excitation/emission = 535/587 nm in FLUOstar OPTIMA (BMG LABTECH) microplate reader using the software OPTIMA and normalized to sample weight. The data were analyzed using one-way ANOVA followed by Tukey’s honestly significant difference test.

Measurement of superoxide (O2-) and other ROS levels

Request a detailed protocol

Levels of superoxide and other ROS were assessed by DHE staining (Sigma-Aldrich), as described in Owusu-Ansah et al., 2008. Briefly, larvae were dissected in Schneider’s media (Gibco) and incubated with DHE at room temperature on an orbital shaker for 7 min in dark. Tissues were fixed in 8% PFA (Electron Microscopy Sciences) for 5 min at room temperature on an orbital shaker in dark. Tissues were then rinsed with PBS (Ambion) and mounted in VECTASHIELD (Vector Laboratories). Confocal microscopy images were obtained with a Leica SP5 Laser Scanning Confocal Microscope at ×200 magnification (excitation/emission 518/605 nm). Fluorescence intensity was quantified on ImageJ software, and data were analyzed using one-way ANOVA followed by Tukey’s honestly significant difference test.

Evaluation of lipid environment of metabolic tissues in larvae

Request a detailed protocol

Fat bodies and Malpighian tubules were dissected in PBS (Ambion) and subjected to lipid staining with Nile Red N3013 Technical grade (Sigma-Aldrich) as previously described (Martelli et al., 2020). Three biological replicates were performed for each exposure condition, each replicate consisting of a single tissue from a single larva. Tissues were fixed in 4% PFA (Electron Microscopy Sciences) and stained with 0.5 µg/mL Nile Red/PBS for 20 min in dark. Slides were mounted in VECTASHIELD (Vector Laboratories) and analyzed using a Leica SP5 Laser Scanning Confocal Microscope at ×400 magnification. Red emission was observed with 540 ± 12.5 nm excitation and 590 LP nm emission filters. Images were analyzed using ImageJ software. For fat bodies, the number, size, and percentage of area occupied by LDs were measured in five different random sections of 2500 µm2 per sample (three samples per group). For Malpighian tubules, the number of LDs was measured in five different random sections of 900 µm2 per sample (three samples per group). The data were analyzed using one-way ANOVA followed by Tukey’s honestly significant difference test.

Lipid quantification in larval hemolymph

Request a detailed protocol

Extracted hemolymph lipids were measured using the sulfo-phospho-vanillin method (Cheng et al., 2011) as previously described (Martelli et al., 2020). 30 third-instar larvae were used for a single biological replicate, and 7 replicate samples were prepared for each exposure condition. Absorbance was measured at 540 nm in a CLARIOstar (BMG LABTECH) microplate reader using MARS Data Analysis Software (version 3.10R3). Cholesterol (Sigma-Aldrich) was used for the preparation of standard curves. The data were analyzed using one-way ANOVA followed by Tukey’s honestly significant difference test.

Lipid extraction and analysis using liquid chromatography-mass spectrometry

Request a detailed protocol

Lipidomic analyses of whole larvae exposed for 2 hr to 2.5 ppm spinosad were performed in biological triplicate and analyzed by electrospray ionization-mass spectrometry (ESI-MS) using an Agilent Triple Quad 6410 as previously described (Martelli et al., 2020). Briefly, samples were transferred to CryoMill tubes treated with 0.001% butylated hydroxytoluene (BHT) and frozen in liquid nitrogen. Samples were subsequently homogenized using a CryoMill (Bertin Technologies) at −10°C. Then, 400 μL of chloroform was added to each tube and samples were incubated for 15 min at room temperature in a shaker at 1200 rpm. Samples were then centrifuged for 15 min at 13,000 rpm at room temperature; the supernatants were removed and transferred to new 1.5 mL microtubes. For a second wash, 100 μL of methanol (0.001% BHT and 0.01 g/mL 13C5 valine) and 200 μL of chloroform were added to CryoMill tubes, followed by vortexing and centrifugation as before. Supernatants were transferred to the previous 1.5 mL microtubes. A total of 300 μL of 0.1 M HCl was added to pooled supernatants, and microtubes were then vortexed and centrifuged (15 min, room temperature, 13,000 rpm). Upper phases (lipid phases) were collected and transferred to clean 1.5 mL microtubes, as well as the lower phases (polar phases). All samples were kept at −20°C until analysis. For liquid chromatography-mass spectrometry (LC-MS) analysis, microtubes were shaken for 30 min at 30°C, then centrifuged at 100 rpm for 10 min at room temperature after which the supernatants were transferred to LC vials. Extracts were used for lipid analysis. For statistical analysis, the concentration of lipid compounds was initially normalized to sample weight. Principal components analysis (PCA) was calculated to verify the contribution of each lipid compound in the variance of each treatment. PCA was calculated using the first two principal component axes. To discriminate the impacts of spinosad on the accumulation of specific lipid compounds, we performed a one-way ANOVA followed by Tukey’s honestly significant difference test.

Investigating impacts on lysosomes

Request a detailed protocol

To investigate spinosad impacts on lysosomes, the LysoTracker staining was used on larval brains. Larvae were dissected in PBS and tissue immediately transferred to PBS solution containing LysoTracker Red DND-99 (1:10,000) (Invitrogen) for 7 min. Tissues were then rinsed three times in PBS, and slides were mounted for immediate microscopy at ×400 magnification (DsRed filter) with a Leica SP5 Laser Scanning Confocal Microscope. To investigate the hypothesis of Dα6 nAChRs being endocytosed and digested by lysosomes after exposure to spinosad, brains from larvae obtained by crossing UAS Dα6 CFP tagged (in Line14 Dα6nx background) to Gal4-L driver (in Line14 Dα6nx background) were also subjected to LysoTracker staining. Images were analyzed using the software ImageJ, and data were analyzed using one-way ANOVA followed by Tukey’s honestly significant difference test.

Electrophysiology of the retina

Request a detailed protocol

Amplitudes and on-transients were assessed as previously described (Martelli et al., 2020). Briefly, adult flies were anesthetized and glued to a glass slide. A reference electrode was inserted in the back of the fly head and the recording electrode was placed on the corneal surface of the eye, both electrodes were filled with 100 mM NaCl. Flies were maintained in the darkness for at least 5 min prior to a series of 1 s flashes of white light delivered using a halogen lamp. During screening, 8–10 flies per treatment group were tested. For a given fly, amplitude and on-transient measurements were averaged based on the response to the three light flashes. Responses were recorded and analyzed using AxoScope 8.1. The data were analyzed using one-way ANOVA followed by Tukey’s honestly significant difference test.

Nile Red staining of adult retinas

Request a detailed protocol

For whole-mount staining of fly adult retinas, heads were dissected in cold PBS (Ambion) and fixed in 37% formaldehyde overnight. Subsequently, the retinas were dissected and rinsed several times with 1× PBS and incubated for 15 min at 1:1,000 dilution of PBS with 1 mg/mL Nile Red (Sigma). Tissues were then rinsed with PBS and immediately mounted with VECTASHIELD (Vector Labs) for same-day imaging. For checking the effects of chronic exposures, eight retinas from eight adult female flies were analyzed per treatment/genotype per time point. Images were obtained with a Leica TCS SP8 (DM600 CS), software LAS X, ×600 magnification, and analyzed using ImageJ. The data were analyzed using one-way ANOVA followed by Tukey’s honestly significant difference test.

Expression of Dα6 nAChRs in the brain

Request a detailed protocol

The expression pattern of nAChR-Dα6 gene in adult brains was assessed in the crossing between Dα6 T2A Gal4 (BDSC #76137) and UAS-GFP.nls (BDSC #4775). Adult brains were fixed in 4% PFA (Electron Microscopy Sciences) in PBS for 20 min at room temperature. PFA was removed and tissues were washed three times in PBS. Samples were mounted in VECTASHIELD (Vector Laboratories). Images were obtained with a Leica TCS SP8 (DM600 CS), software LAS X, ×400 magnification, using GFP channel. Images were analyzed using the software ImageJ.

Adult brain histology (H&E staining)

Request a detailed protocol

Adult fly heads were fixed in 8% glutaraldehyde (EM grade) and embedded in paraffin. Sections (10 µm) were prepared with a microtome (Leica) and stained with H&E as described (Chouhan et al., 2016). At least three animals were examined for each group (20 days exposure to 0.2 ppm spinosad plus control group) in terms of percentage of brain area vacuolated. The data were analyzed using Student’s unpaired t-test.

Transmission electron microscopy (TEM)

Request a detailed protocol

Laminas of adult flies chronically exposed to 0.2 ppm spinosad for 20 days (controls exposed to equivalent volume of DMSO) were processed for TEM imaging as described (Luo et al., 2017). TEM of laminas of 20-day-old Canton-S and Canton-S Dα6 KO mutants aged in the absence of spinosad was also investigated. Samples were processed using a Ted Pella Bio Wave microwave oven with vacuum attachment. Adult fly heads were dissected at 25°C in 4% PFA, 2% glutaraldehyde, and 0.1 M sodium cacodylate (pH 7.2). Samples were subsequently fixed at 4°C for 48 hr. 1% osmium tetroxide was used for secondary fixation with subsequent dehydration in ethanol and propylene oxide. Samples were then embedded in Embed-812 resin (Electron Microscopy Sciences, Hatfield, PA). 50-nm ultra-thin sections were obtained with a Leica UC7 microtome and collected on Formvar-coated copper grids (Electron Microscopy Sciences). Specimens were stained with 1% uranyl acetate and 2.5% lead citrate and imaged using a JEOL JEM 1010 transmission electron microscope with an AMT XR-16 mid-mount 16 megapixel CCD camera. For quantification of ultrastructural features, electron micrographs were examined from three different animals per treatment. The data were analyzed using Student’s unpaired t-test.

Bang sensitivity

Request a detailed protocol

The BS phenotype was tested after 1, 10, and 20 days of chronic exposure to 0.2 ppm spinosad. Flies were vortexed on a VWR vortex at maximum strength for 10 s. The time required for flies to flip over and regain normal standing posture was then recorded. The data were analyzed using Kruskal–Wallis followed by Dunn’s multiple comparisons test.

Climbing assay

Request a detailed protocol

Climbing phenotype was tested after 1, 10, and 20 days of exposure to 0.2 ppm spinosad. Five adult female flies were placed into a clean vial and allowed to rest for 30 min. Vials were tapped against a pad, and the time required for the flies to climb up to a predetermined height (7 cm) was recorded. Flies that did not climb the predetermined height within 30 s were deemed to have failed the test. The data were analyzed using Kruskal–Wallis followed by Dunn’s multiple comparisons test.

Graphs and statistical analysis

Request a detailed protocol

Wiggle Index analyses were performed using the software R (v.3.4.3) (Denecke et al., 2015). All other graphs and statistical analyses were performed using GraphPad Prism (v.9.2.0). Image panels were designed using the free image software Inkscape (0.92.4).

Many of the analyses performed here were conducted on spinosad and imidacloprid in parallel with these treatments sharing the same controls, allowing direct comparison of the impact of these insecticides. The imidacloprid data are published in Martelli et al., 2020. The data with shared wild-type control flies (unexposed) are shown in Figures 1A, D, E, 3A–D, Figures 4A–C–5A, B, C, D, G, Figure 5—figure supplements 1 and 2, Figure 6, Figure 6—source data 1, Figures 7A–C–9B and C.

Data availability

All data generated or analysed during this study are included in the manuscript and supporting file; Source Data file has been provided for Figure 6.

References

    1. Lu C
    2. Warchol KM
    3. Callahan RA
    (2014)
    Sub-lethal exposure to neonicotinoids impaired honey bees winterization before proceeding to colony collapse disorder
    Bull Insectology 67:125–130.
  39. Book
    1. Luong HNB.
    (2018)
    In vivo functional characterization of nicotinic acetylcholine receptors in Drosophila melanogaster
    The University of Melbourne.

Decision letter

  1. Bruno Lemaître
    Reviewing Editor; École Polytechnique Fédérale de Lausanne, Switzerland
  2. Utpal Banerjee
    Senior Editor; University of California, Los Angeles, United States

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Decision letter after peer review:

[Editors’ note: the authors submitted for reconsideration following the decision after peer review. What follows is the decision letter after the first round of review.]

Thank you for submitting your work entitled "The organic insecticide spinosad trigger lysosomal defect, ROS driven lipid dysregulation and neurodegeneration in flies" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and a Senior Editor. The reviewers have opted to remain anonymous.

We are sorry to say that, after consultation with the reviewers, we have decided that your work will not be considered further for publication by eLife. But they all agree that they are happy to reconsider a revised version that provides more mechanistic insights on the action of the insecticide.

In this manuscript, it is shown that low doses of spinosad are toxic to Drosophila, suggesting that traces of this organic pesticide in the environment can be damaging to other useful insect species, as with the neonicotinoid imidacloprid. It is not easy to assess the novelty and significance of this claim. Spinosad has been known for a long time to have a large spectrum of action, including many species of Lepidoptera and Diptera, like e.g. the house fly (Liu and Yue J Econ Entomol 2000). In addition, field studies will be asked in any cases to confirm the danger of this pesticide at low doses out of the laboratory.

Mechanistically, the authors propose that spinosad accumulates in lysosomes after binding to nAChRα6 and that this would progressively disrupt their function. Although this hypothesis is suggested by the disappearance of the α6 subunit, reported in another paper this year (Nguyen et al., Pest Manag Sci 2021) this, and lysosomal swelling, is not directly demonstrated in this study. The authors also suggest that lysosome disruption would give rise to mitochondrial impairments, increased ROS production and finally neurodegeneration, but this causal link is hypothetical at this stage, as noted by some reviewers. The authors finally suggest that a toxic signal released from the brain propagates to peripheral tissues in spinosad–intoxicated flies, explaining the non–nervous system defect, but again, this idea is not addressed experimentally. While the topic was found to be of general interest, the reviewers also feel that this manuscript is very descriptive and it does not report enough original breakthrough findings to warrant publication in eLife at this stage. A new submission that includes more mechanistic insights might be considered for eLife but requires substantial improvement.

Reviewer #2:

This article reports that an organic insecticide used at low doses creates nevertheless considerable havoc, including extensive neurodegeneration in adults. The authors show that exposure to spinosad blocks acetylcholine–triggered calcium influx in larval neurons that express the Acetylcholine receptor (AchR) subunit Dalpha6, the target of spinosad. They document that exposure to spinosad leads to a Dalpha6–dependent enlargement of lysosomes in the nervous system where the Dalpha6 subunits accumulates, an increase in reactive oxygen species (ROS) generation in the brain, and a redistribution of lipids in the fat body from large lipid droplets (LDs) to numerous smaller ones, an effect that can be counteracted by pre–exposure to an antioxidant compound. The chronic exposure of adults to low spinosad doses leads to progressively altered behaviors as well as extensive neurodegeneration in the visual system as well as in parts of the central brain.

The strengths of the study lie in the description of several damages induced by exposure to spinosad. The part on the neurodegeneration in adults is especially impressive. The central message of the authors that even organic pesticides have unintended severe effects to a nontarget insect upon exposure to low doses (chronic low doses in the case of adults) is thus supported by experimental evidence, which however might be improved in specific cases.

The major weaknesses of the study are that the authors hardly investigate causal relationships between the different phenotypes they observe and remain too descriptive. For instance, their preferred model is that lysosomal disorders are at the origin of ROS production in mitochondria, with no experimental evidence to back up this claim. This reviewer is also not thoroughly convinced that ROS are produced by mitochondria, as the decreased activity of aconitase and impact on ATP production might be mediated by another mechanism, e.g., increased mitophagy. They do not consider the alternative possibility that lysosomal dysfunction might be caused by ROS production. Thus, one would like to know whether the expansion of lysosomes still occurs when ROS production is inhibited. This would be best achieved using two independent approaches, one based on exposure to reducing agents (do they have any effect on their own (control missing)?), and the other based on genetic strategies that however require identifying the enzyme(s) responsible for ROS production.

They also do not document whether the ROS phenotype depends on the presence of the Dalpha6 subunit of the AchR. It would also be positive to check whether there is any altered production of ROS occurring directly in the fat body or Malpighian tubules. Can the authors exclude that the Dalpha6 nAchR subunit might be expressed at low levels in these tissues, for instance in the enteroendocrine cells of the gut?

One major question raised by this study is whether insects other than those targeted by this pesticide are exposed to low doses, for instance in fields neighboring the treated area and undergo damages similar to those reported here. Some of the techniques used in this study may be employed on other insects collected near fields treated with spinosad, such as examining the distribution of LDs in their fat bodies. In this respect, it might be useful determining whether this phenotype as well as the expanded lysosomes observed in larvae are also found in adults.

As stated in above, a major weakness of the study is a lack of functional connections made between the different phenotypes documented by the authors. While it may not be easy to determine whether ROS production depends on lysosome dysfunction, the reverse can be investigated relatively easily, for instance by using larvae pre–treated with NACA and other reagents as detailed below.

As regards ROS, one definitely would like to know whether they are generated solely in neurons expressing the Dalpha6 nAchR subunit or whether they are found in other neuronal cell types or glial cells. A subcellular resolution would be ideal. To this end, the authors might want to use ratiometric ROS stress reporters such as Mito–roGFP (Albrechts et al., Cell Metabolism, 2011). It would also be desirable to rely on more than one method to block ROS production. In terms of biochemical reducing agents, the authors might want to also check MitoTempo, which is more specifically acting on mitochondrial ROS. On the genetic side, the authors should overexpress Superoxide Dismutase 2 (SOD2), and possibly SOD1 as a control. Another useful tool is a mitochondrially–addressed catalase (UAS–mtcatalase) developed originally by William Orr, e.g., Mockett et al., Free Radic. Biol. Med. , 2003.

Once the mitochondrial origin of ROS is demonstrated, it will then be highly interesting to determine whether lysosome expansion still occurs when the source is inhibited.

The authors rely solely on Nile Red for assessing the distribution of LDs. A control with lipidTox staining, which is more specific for neutral lipids, would be appropriate. Also, while small LDs may increase the surface available for storing oxidized lipids, it would be useful to determine whether this is indeed the case by staining oxidized lipids. It is also not clear whether the overall levels of triglycerides are changed upon spinosad exposure and measuring the levels of TAGs biochemically would provide an interesting counterpoint to lipid stains.

What is the biological function of this redistribution of lipids in the fat body? Would the genetic inhibition of Dalpha6–expressing neurons lead to the LD phenotype (likely unlikely given the loss–of–function phenotype). Conversely, would genetically activating neuronal signaling in these neurons be able to reverse the effects of spinosad exposure, at least as regards the LD phenotype in the fat body?

Reviewer #3:

In this paper, the authors used Drosophila as a model to study the effect of the insecticide spinosad on the non–target insect. They investigated the histological, physiological, and behavioral impacts of spinosad and found that low dose administration of this pesticide causes neuronal lysosomal alterations and mitochondrial impairment and ROS elevation and lead to altered lipids profile and neurodegeneration and subsequent consequences. One of the main results of the study is to show that spinosad induce major systemic metabolic changes through its action in the brain. Since spinosad have been shown to greatly affect the non–pest insect population, the findings here are interesting and appealing to many researchers. I only have two major comments as listed below.

1. The authors suggest that spinosad actions on neurons expressing Dα6 nAChRs inducing subsequent lipid phenotypes, especially through ROS. They also observed an increased ROS level in tissues as gut. Although Dα6 mutant showed a decreased lipid droplets in fatbody, there is no direct evidence linking neuronal ROS with this phenotype. Thus, it remains unclear whether peripheral tissues ROS also contribute to the altered lipid phenotypes in the fly. Furthermore, whether elevated gut ROS level is induced by neuronal ROS burst or independent from neuronal phenotype is not clear. I would suggest using fly genetic tools (Gal4–uas system) to manipulate ROS level in the neurons and peripheral tissues to further elucidate the conclusion.

2. The author's finding pointed to the mechanisms of spinosad on non–target insects. However, since insects are the most diverse group of animals and there is a huge divergence between Drosophila and other insects, whether spinosad causes the same effects in other species needs to be clearly discussed.

3. The authors suggest that the spinosad effect on Drosophila was a consequence of its action on Dα6 nAChRs. It would be better to include survival data of Dα6 mutant upon Spinosad exposure to reinforce the conclusion.

4. It would also be interesting to see if there is also a ROS burst in the fatbody that leads to the metabolic phenotype directly. Overexpression of antioxidant enzymes (sod1, sod2, catalase) using elav–Gal4 (Pan neuron driver), myo–Gal4 (gut enterocytes driver) and c564–Gal4 (fatbody driver) could elucidate the causal relations between the neuronal phenotypes and lipid metabolic phenotypes.

Reviewer #4:

The manuscript by Felipe Martelli and coll. addresses the cellular and metabolic defects induced by poisoning with spinosad in the fruit fly Drosophila. Spinosad is a potent insecticide first identified as being produced by a soil bacterium and so considered as an organic product, which has been widely used in recent years. Published work showed that the main selective target of spinosad in insects is the α6 subunit of the nicotinic acetylcholine receptor (nAchR), which is selectively expressed in the nervous system in flies, and that knockout of nAChRα6 (Dα6) in Drosophila confers resistance to spinosad. This paper complements and extends these previous studies by showing that spinosad exposure triggers lysosomal swelling, mitochondrial impairment and oxidative stress in fly larval brain cells, and also induces a variety of metabolic dysregulation in peripheral tissues, including alterations in the larval lipidomic profile and an accumulation of small lipid droplets in the fat body. Pretreatment with the anti–oxidant N–acetylcysteine amide (NACA) alleviated some of these defects and partially, the larval lethality induced by spinosad. It is also shown that low doses of spinosad promoted neurodegeneration and loss of vision in adult female Drosophila.

Overall, this manuscript represents a substantial amount of experimental work that provides new insight into spinosad–induced toxicity. The authors introduce here several hypotheses to make sense of their various observations. Specifically, they propose that spinosad accumulates in lysosomes after binding to nAChRα6 and because of that would progressively disrupt lysosomal function. This could then give rise to mitochondrial dysfunction, increased ROS production and finally neurodegeneration. The authors also suggest that a toxic signal released from the brain after spinosad intoxication propagates to peripheral tissues, including the fat body, midgut, and Malpighian tubules, to promote disturbance in metabolism and lipid homeostasis. Although these hypotheses are not directly addressed experimentally in this study, they would certainly be interesting to explore in future work.

There are three sets of results for which I have concerns about some of the authors' conclusions:

– In Figure 1, it is not clear whether the lower carbachol–induced response in spinosad–treated cells reflects a decrease in calcium influx through nAChRs, or, alternatively, a reduced calcium release from intracellular stores (see e.g. Campusano et al. Dev Neurobiol 2007). This point has to be carefully considered as the lower response to carbachol could be either a direct or indirect effect of spinosad–induced Dα6 subunit internalization. I think that the reported experiments are not sufficient to decide between these possibilities.

– The second issue is the experiment of Figure 5, in which the authors showed that spinosad treatment of Dα6 KO flies had no effect on lipid level in the hemolymph, in contrast to the increase it induced in wild–type controls. This conclusion is apparently correct, but it does not take in account the fact that the lipid level actually appears at least as high in the Dα6 mutants as in their respective controls treated with spinosad. It seems therefore that the increase in lipid in the hemolymph is more likely caused by Dα6 deficiency than by spinosad treatment. Similarly, the percentage of the area occupied by LD in fat body of both Dα6 mutants is between that of untreated control flies and control flies treated with spinosad. This seems to be consistent with the idea that spinosad only exacerbates an increase in lipids which is for a large part induced by Dα6 deficiency. Incidentally, because Dα6 KO mutants are viable, this observation indicates that altered lipid homeostasis may not be a prominent cause of larval lethality in spinosad–exposed flies.

– My third main comment is that it would be very informative to compare the effects of spinosad in adult wild–type and Dα6 mutant flies, with respect to the survival, bang sensitivity and climbing ability tests (Figure 7), as well as for lipid deposits in the retina (Figure 8). These data would be rather easy to collect and would usefully complete these figures by indicating whether or not spinosad can induce defects in adult flies in the absence of Dα6. A positive answer would argue for the existence of other potential molecular targets of this insecticide.

I have provided more detail on these and other points below.

Main issues

1) Line 117 and 132: from experiments shown in Figure 1D, E, the authors conclude that spinosad exposure "prevent ca2+ flux into neurons expressing nAChRs" and so "blocks the function of Dα6–containing nAChRs". However, Campusano et al. Dev Neurobiol 2007, have shown that thapsigargin treatment reduces nicotine‐evoked calcium increase in Kenyon cells, consistent with a signal amplification by calcium release from intracellular stores. In addition, part of this calcium increase had a voltage‐gated calcium channel–dependent (VGCC) component, indicating that only a fraction of the calcium increase involves calcium influx directly through nAChRs. Therefore, the experiments in Figure 1D, E are not sufficient to decide, in my opinion, whether the reduction in carbachol–induced response in spinosad–treated cells reflects a decrease in calcium influx through nAChRs, or, alternatively, a reduced calcium release from intracellular stores. As mentioned in the public review, this point has to be carefully considered as the decreased response to carbachol could be either a direct or indirect effect of spinosad–induced Dα6 subunit internalization.

2) Line 263: "We also quantified the level of lipids in hemolymph. Whereas Line 14 and Canton S showed an average 10% and 13% increase in response to spinosad, respectively, neither of the Dα6 KO mutants showed significant changes (Figure 5E)." However, the lipid level appears to be at least as high in the Dα6 mutants as in their respective controls treated by spinosad (Figure 5E). It seems therefore that the increase in lipid in the hemolymph is more likely caused by Dα6 deficiency than by spinosad treatment. Similarly, the percentage of area occupied by LD in fat body of both Dα6 mutants is between that of untreated control flies and control flies treated with spinosad (compare Figure 4B and Figure 5B for the Line 14 background, and compare Figure 5C and Figure 5D for the Canton S background). This seems to be consistent with the idea that spinosad only exacerbates an increase in lipids which is largely induced by Dα6 deficiency. If this is correct, please amend the conclusions accordingly. For example, the sentence on page 25, line 479: "Dα6 knockout mutants exposed to spinosad show no accumulation of LD in the fat body or change of lipid levels in hemolymph indicating that these phenotypes are due to the spinosad:Dα6 interaction (Figure 5)" is not fully correct as it does not mention the fact that Dα6 deficiency by itself appears to increase LD accumulation.

3) It would be very informative to repeat the survival and behavior test experiments of Figure 7 in Dα6 mutant background. I was actually surprised that this was not included in the manuscript, as it would not involve a lot of work, and, importantly, it would indicate whether spinosad can induce behavior defects in the absence of Dα6, which would argue for the existence of other molecular targets for this toxin. I would recommend that lipid deposits in the retina (Figure 8A, B) be also quantified in Dα6 background for purpose of comparison.

– Line 125: "primary culture of neurons " could be "primary culture of third–instar larva brain neurons" to avoid misunderstanding.

– In Figure 1A, the statistical test used should be two–way ANOVA and not t–test, I think, as both the genotypes and time differ in each point, and in Figure 1E one–way ANOVA should be used, as the means of more than two groups were compared. Same issue in Figure 2B and in several other figures of the paper where the Student's t–test was systematically used.

– Figure 1B: I understand that spinosad precludes pupation of intoxicated larvae. However, after eight days, a wild–type third–instar larva should have become an adult fly. So, I guess that authors looked at fly survival, and not larval survival, for the wild–type. Please mention it, if it is right. In Figure 1C, did you treat the pupae or the larvae before pupal case formation with the toxin? This could be stated in the legend.

– Line 579: "derived from Armenia60 (Drosophila Genomics Resource Center #103394)" – I could not find this Armenia60 line referenced neither in the DGRC or Flybase sites. Please provide the correct source or link for this strain, and explain why it was used as a wild–type control.

– Line 581: "Expression of nAChR–Dα6 gene" – According to Flybase, this gene should be named: nAChRα6. The correct nomenclature should be mentioned at least once in the manuscript.

– Line 582: "the MitoTimer line (Gottlieb and Stotland, 2015) was used". The exact reference for this line is: Laker et al. J Biol Chem. (2014) 289(17):12005–15. Same correction required on line 187.

– Line 594: "In all experiments involving adult flies only females were used to maintain consistency." Adult Drosophila females have very different physiology compared to males. This is particularly relevant for reproductive organs, of course, but also for the gut and nervous system. Therefore, this particularity of this study is not negligible and should be mentioned in the abstract.

[Editors’ note: further revisions were suggested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "Low doses of the organic insecticide spinosad triggers lysosomal defects, elevated ROS, lipid dysregulation, and neurodegeneration in flies" for further consideration by eLife. Your revised article has been evaluated by Utpal Banerjee (Senior Editor) and Bruno LeMaitre (Reviewing Editor).

The manuscript by Martelli et al. has been clearly improved in this re-submission. In particular, the statistical analysis has been done more accurately and new interesting experiments have been added. But there are some remaining issues that need to be addressed, as outlined below:

Essential revisions:

1) While this revised version is definitely improved, a sticky point remains, namely that of the relationship between the lysosomal disorder and the production of ROS by mitochondria, which is not rigorously established. The authors use NACA to block the production of ROS. Since presumably NACA has a ubiquitous effect, the authors cannot formally exclude that the ROS originate from a source distinct from mitochondria. In this respect, it is noteworthy that the authors did not investigate the impact of NACA treatment on mitochondria. This reviewer does not understand the argument put forth by the authors as regards the lack of subcellular resolution of mito-roGFP reporters. Because the reporters are targeted to mitochondria, it follows that any signal reflects redox conditions in mitochondria.

2) It may be hazardous to rely only on chemicals to study the effects of ROS. Genetic manipulation of ROS by SOD2 (and eventually SOD1 as control), can clarify whether mitochondria are the ROS source and further demonstrate whether ROS is the cause of other phenotypes. As this superoxide dismutase functions in mitochondria, it should quench the generation of ROS induced by spinosad and thus provide an independent, more convincing, demonstration of the role of mitochondria in spinosad-induced toxicity.

3) The authors did not fully address previous issue 3 of reviewer 2, most likely due to a misunderstanding. What I asked for was that the authors check whether spinosad can induce or not behavior defects (bang sensitivity and climbing ability) in adult Dα6 mutants and whether spinosad can cause lipid deposits in the retina of Dα6 mutants. Instead, the authors characterized behavior defects and retina structure in adult Dα6 mutants in the absence of spinosad, which are interesting but incomplete observations in my opinion. In other experiments (e.g. Figure 4D and E, and Figure 5B and G), the authors have indeed compared the effects of spinosad in control and Dα6 mutants. Similarly, I think that it would be worth doing the same test (i.e. check the effects or lack of effects of spinosad on adult Dα6 mutants) for the experiments described in Figure 7A-C and Figure 8A-D.

https://doi.org/10.7554/eLife.73812.sa1

Author response

[Editors’ note: the authors resubmitted a revised version of the paper for consideration. What follows is the authors’ response to the first round of review.]

Reviewer #2:

This article reports that an organic insecticide used at low doses creates nevertheless considerable havoc, including extensive neurodegeneration in adults. The authors show that exposure to spinosad blocks acetylcholine–triggered calcium influx in larval neurons that express the Acetylcholine receptor (AchR) subunit Dalpha6, the target of spinosad. They document that exposure to spinosad leads to a Dalpha6–dependent enlargement of lysosomes in the nervous system where the Dalpha6 subunits accumulates, an increase in reactive oxygen species (ROS) generation in the brain, and a redistribution of lipids in the fat body from large lipid droplets (LDs) to numerous smaller ones, an effect that can be counteracted by pre–exposure to an antioxidant compound. The chronic exposure of adults to low spinosad doses leads to progressively altered behaviors as well as extensive neurodegeneration in the visual system as well as in parts of the central brain.

The strengths of the study lie in the description of several damages induced by exposure to spinosad. The part on the neurodegeneration in adults is especially impressive. The central message of the authors that even organic pesticides have unintended severe effects to a nontarget insect upon exposure to low doses (chronic low doses in the case of adults) is thus supported by experimental evidence, which however might be improved in specific cases.

The major weaknesses of the study are that the authors hardly investigate causal relationships between the different phenotypes they observe and remain too descriptive. For instance, their preferred model is that lysosomal disorders are at the origin of ROS production in mitochondria, with no experimental evidence to back up this claim. This reviewer is also not thoroughly convinced that ROS are produced by mitochondria, as the decreased activity of aconitase and impact on ATP production might be mediated by another mechanism, e.g., increased mitophagy. They do not consider the alternative possibility that lysosomal dysfunction might be caused by ROS production. Thus, one would like to know whether the expansion of lysosomes still occurs when ROS production is inhibited. This would be best achieved using two independent approaches, one based on exposure to reducing agents (do they have any effect on their own (control missing)?), and the other based on genetic strategies that however require identifying the enzyme(s) responsible for ROS production.

They also do not document whether the ROS phenotype depends on the presence of the Dalpha6 subunit of the AchR. It would also be positive to check whether there is any altered production of ROS occurring directly in the fat body or Malpighian tubules. Can the authors exclude that the Dalpha6 nAchR subunit might be expressed at low levels in these tissues, for instance in the enteroendocrine cells of the gut?

One major question raised by this study is whether insects other than those targeted by this pesticide are exposed to low doses, for instance in fields neighboring the treated area and undergo damages similar to those reported here. Some of the techniques used in this study may be employed on other insects collected near fields treated with spinosad, such as examining the distribution of LDs in their fat bodies. In this respect, it might be useful determining whether this phenotype as well as the expanded lysosomes observed in larvae are also found in adults.

We thank the reviewer for these comments and have addressed most issues raised (see below).

As stated in above, a major weakness of the study is a lack of functional connections made between the different phenotypes documented by the authors. While it may not be easy to determine whether ROS production depends on lysosome dysfunction, the reverse can be investigated relatively easily, for instance by using larvae pre–treated with NACA and other reagents as detailed below.

1) The levels of superoxide in brains of Dα6 KO mutant flies were measured prior and after spinosad exposure and results added to the manuscript (Figure 4D, E). Text was amended in the following way (lines 222-229): “To test whether the increase of superoxide levels in brains are a consequence of the spinosad induced Dα6 removal from membranes, the levels of superoxide were measured in exposed and non-exposed Dα6 KO mutants. Non-exposed Dα6 KO mutants showed a mild (17%) increase in the levels of superoxide in brains when compared to non-exposed wild type larvae (Figure 4D, E), and exposure to spinosad caused no alteration of superoxide levels in Dα6 KO mutants (Figure 4D, E). Hence, the absence of Dα6 subunit by itself is able to modestly increase the oxidative stress (Weber et al., 2012), but higher levels of ROS are observed in the presence of Dα6 and spinosad.”

2) Experiments exposing NACA pre-treated larvae to spinosad were added. They showed that NACA prevents ROS accumulation (Figure 4F, G) but does not prevent lysosome expansion (Figure 4H, I). Text was amended in the following way (lines 232-238): “Whereas NACA treatment was able to completely prevent ROS accumulation in exposed animals (Figure 4F, G), it does not prevent lysosome expansion (Figure 4H, I). The presence of enlarged lysosomes in the absence of ROS suggests that the onset of lysosomal phenotype is not caused by the rise in oxidative stress levels. NACA, however, reduced the severity of the lysosomal phenotype (mean 1.63% of lysotracker area – Figure 4I, versus mean 2.39% of lysotracker area – Figure 2E). This suggests that, once initiated, the increase in ROS levels may worsen the lysosomal phenotype.”

As regards ROS, one definitely would like to know whether they are generated solely in neurons expressing the Dalpha6 nAchR subunit or whether they are found in other neuronal cell types or glial cells. A subcellular resolution would be ideal. To this end, the authors might want to use ratiometric ROS stress reporters such as Mito–roGFP (Albrechts et al., Cell Metabolism, 2011).

Upon investigation we concluded that Mito-roGFP could not generate the level of resolution expected to indicate the subcellular origin of ROS. Recent data however

(https://www.biorxiv.org/content/10.1101/2021.07.04.451050v1.article-metrics), points that Dα6 is largely expressed in neurons while not expressed in glia, guts, or fat body. Text was amended in the following way (lines 273-275): “No expression of Dα6 has been reported in gut and fat body but it is abundantly and widely expressed in most CNS neurons with little to no expression in glia (Li et al., 2021).”

It would also be desirable to rely on more than one method to block ROS production. In terms of biochemical reducing agents, the authors might want to also check MitoTempo, which is more specifically acting on mitochondrial ROS. On the genetic side, the authors should overexpress Superoxide Dismutase 2 (SOD2), and possibly SOD1 as a control. Another useful tool is a mitochondrially–addressed catalase (UAS–mtcatalase) developed originally by William Orr, e.g., Mockett et al., Free Radic. Biol. Med. , 2003.

Once the mitochondrial origin of ROS is demonstrated, it will then be highly interesting to determine whether lysosome expansion still occurs when the source is inhibited.

In response to spinosad exposure we show: (a) accumulation of superoxide – main ROS produced by mitochondria – in brain and guts (Figure 4A, B); (b) reduction of mitochondrial aconitase activity (Figure 3D) and (c) ATP (Figure 3E); (d) increased mitochondrial turnover (MitoTimer) (Figure 3A, B); and (e) reduced cardiolipin levels (Figure 6C). These data strongly argue that the observed ROS in wild-type exposed flies is likely not due to a xenobiotic response. However, we cannot rule out that this response does not play a minor role. We further demonstrate that NACA rescues improves motility (Figure 3F) and survival (Figure 3G) of spinosad exposed larvae, while reducing impacts on the lipid environment of metabolic tissues (Figure 5C, D). One of the novelties present in this work is connecting the generation of ROS to spinosad mode of action.

The authors rely solely on Nile Red for assessing the distribution of LDs. A control with lipidTox staining, which is more specific for neutral lipids, would be appropriate. Also, while small LDs may increase the surface available for storing oxidized lipids, it would be useful to determine whether this is indeed the case by staining oxidized lipids.

Microscopy with Nile red securely shows the increase in lipid droplets after spinosad exposure (Figure 5A, B) and the amelioration of this phenotype with NACA pre-treatment (Figure 5C, D). Despite LipiTox higher affinity for neutral lipids, experiments with this staining would not change the lipid droplet count obtained with Nile red.

It is also not clear whether the overall levels of triglycerides are changed upon spinosad exposure and measuring the levels of TAGs biochemically would provide an interesting counterpoint to lipid stains.

We have observed the increase of some TAG species in our lipidomic profiles, as well as the decrease of others (Figure 6A; Figure 6 – table supplement 1). While the use of whole larvae for lipidomic analysis reduced the capacity to detect significant shifts in lipid levels that predominantly occur in individual tissues, it allowed the identification of broader impacts on larval biology. The levels of circulating lipids in hemolymph as well as the lipid droplet counts were assessed in two wild-type strains and their respective Dα6 KO mutants. In all cases, increased lipid levels were found in response to spinosad exposure in wild-type flies (Figure 5A, B and G). And whereas Dα6 KO mutants present higher lipid levels than their wild-type strains, such levels did not change after spinosad exposure (Figure 5A, B and G).

What is the biological function of this redistribution of lipids in the fat body? Would the genetic inhibition of Dalpha6–expressing neurons lead to the LD phenotype (likely unlikely given the loss–of–function phenotype). Conversely, would genetically activating neuronal signaling in these neurons be able to reverse the effects of spinosad exposure, at least as regards the LD phenotype in the fat body?

The knockout of Dα6 leads to mild alterations in the lipid environment. It increases lipid stores in fat body and lipid levels in hemolymph (Figure 5A, B and G). Dα6 loss of function mutation has been linked to an increased susceptibility to oxidative stress

(DOI: 10.1371/journal.pone.0034745). Many sections in the manuscript were restructured to clearly point the difference between phenotypes causes by spinosad exposure and those caused by Dα6 loss of function. Regarding genetically activation of neuron signalling, since nicotinic receptors are involved in a plethora of functions (from muscle activity to hormone levels control and behaviour), the activation of neural signalling would have a diverse range of significant effects that should first be addressed before testing its effect on spinosad exposure.

Reviewer #3:

In this paper, the authors used Drosophila as a model to study the effect of the insecticide spinosad on the non–target insect. They investigated the histological, physiological, and behavioral impacts of spinosad and found that low dose administration of this pesticide causes neuronal lysosomal alterations and mitochondrial impairment and ROS elevation and lead to altered lipids profile and neurodegeneration and subsequent consequences. One of the main results of the study is to show that spinosad induce major systemic metabolic changes through its action in the brain. Since spinosad have been shown to greatly affect the non–pest insect population, the findings here are interesting and appealing to many researchers. I only have two major comments as listed below.

1. The authors suggest that spinosad actions on neurons expressing Dα6 nAChRs inducing subsequent lipid phenotypes, especially through ROS. They also observed an increased ROS level in tissues as gut. Although Dα6 mutant showed a decreased lipid droplets in fatbody, there is no direct evidence linking neuronal ROS with this phenotype. Thus, it remains unclear whether peripheral tissues ROS also contribute to the altered lipid phenotypes in the fly. Furthermore, whether elevated gut ROS level is induced by neuronal ROS burst or independent from neuronal phenotype is not clear. I would suggest using fly genetic tools (Gal4–uas system) to manipulate ROS level in the neurons and peripheral tissues to further elucidate the conclusion.

2. The author's finding pointed to the mechanisms of spinosad on non–target insects. However, since insects are the most diverse group of animals and there is a huge divergence between Drosophila and other insects, whether spinosad causes the same effects in other species needs to be clearly discussed.

We thank the reviewer for these comments and have addressed most issues raised (see below).

3. The authors suggest that the spinosad effect on Drosophila was a consequence of its action on Dα6 nAChRs. It would be better to include survival data of Dα6 mutant upon Spinosad exposure to reinforce the conclusion.

Data showing the survival of Dα6 mutant upon spinosad exposure was added (Figure 1C). Text was amended in the following way (lines 117-119): “Under this exposure condition, only 4% of wild type larvae survived to adulthood (Figure 1B), whereas 88% nAChRα6 knockout (Dα6 KO) mutants survived (Figure 1C)”.

4. It would also be interesting to see if there is also a ROS burst in the fatbody that leads to the metabolic phenotype directly.

Data showing that no burst of superoxide is observed in fat body upon spinosad exposure was added (Figure 5E, F). Text was amended in the following way (lines 286-287): “However, measurements of superoxide levels in fat bodies showed no differences between spinosad exposed and non-exposed larvae (Figure 5E, F).”.

Overexpression of antioxidant enzymes (sod1, sod2, catalase) using elav–Gal4 (Pan neuron driver), myo–Gal4 (gut enterocytes driver) and c564–Gal4 (fatbody driver) could elucidate the causal relations between the neuronal phenotypes and lipid metabolic phenotypes.

We do not exclude the possibility that oxidative stress is directly generated in the midgut but in a previous work published in PNAS (Martelli et al.,2020) we demonstrate that knocking out mitochondrial genes (ND42 and Marf) in the nervous system (elav-Gal4) increases oxidative stress in the brain and, also lead to an accumulation of lipid droplets in fat body as well as a reduction of lipid droplets in Malpighian tubules. The text was amended to emphasised this (lines 257-265): “Oxidative stress has the ability to affect the lipid environment of metabolic tissues, causing bulk redistribution of lipids into lipid droplets (LD) (Bailey et al., 2015). The RNAi knockdown of mitochondrial genes, Marf and ND42, in Drosophila neurons was shown to increase the levels of ROS in the brain and precipitate the accumulation of LD in glial cells (Liu et al., 2015). Martelli et al. (2020) showed that the knockdown of the same mitochondrial genes in Drosophila neurons can also precipitate the accumulation of LD in fat bodies and a reduction of LD in Malpighian tubules. Such changes in the lipid environment of metabolic tissues were recapitulated by low imidacloprid exposures, which like spinosad, also causes an increase of ROS levels in the brain that further spreads to the anterior midgut (Martelli et al., 2020).”

Reviewer #4:

The manuscript by Felipe Martelli and coll. addresses the cellular and metabolic defects induced by poisoning with spinosad in the fruit fly Drosophila. Spinosad is a potent insecticide first identified as being produced by a soil bacterium and so considered as an organic product, which has been widely used in recent years. Published work showed that the main selective target of spinosad in insects is the α6 subunit of the nicotinic acetylcholine receptor (nAchR), which is selectively expressed in the nervous system in flies, and that knockout of nAChRα6 (Dα6) in Drosophila confers resistance to spinosad. This paper complements and extends these previous studies by showing that spinosad exposure triggers lysosomal swelling, mitochondrial impairment and oxidative stress in fly larval brain cells, and also induces a variety of metabolic dysregulation in peripheral tissues, including alterations in the larval lipidomic profile and an accumulation of small lipid droplets in the fat body. Pretreatment with the anti–oxidant N–acetylcysteine amide (NACA) alleviated some of these defects and partially, the larval lethality induced by spinosad. It is also shown that low doses of spinosad promoted neurodegeneration and loss of vision in adult female Drosophila.

Overall, this manuscript represents a substantial amount of experimental work that provides new insight into spinosad–induced toxicity. The authors introduce here several hypotheses to make sense of their various observations. Specifically, they propose that spinosad accumulates in lysosomes after binding to nAChRα6 and because of that would progressively disrupt lysosomal function. This could then give rise to mitochondrial dysfunction, increased ROS production and finally neurodegeneration. The authors also suggest that a toxic signal released from the brain after spinosad intoxication propagates to peripheral tissues, including the fat body, midgut, and Malpighian tubules, to promote disturbance in metabolism and lipid homeostasis. Although these hypotheses are not directly addressed experimentally in this study, they would certainly be interesting to explore in future work.

There are three sets of results for which I have concerns about some of the authors' conclusions:

– In Figure 1, it is not clear whether the lower carbachol–induced response in spinosad–treated cells reflects a decrease in calcium influx through nAChRs, or, alternatively, a reduced calcium release from intracellular stores (see e.g. Campusano et al. Dev Neurobiol 2007). This point has to be carefully considered as the lower response to carbachol could be either a direct or indirect effect of spinosad–induced Dα6 subunit internalization. I think that the reported experiments are not sufficient to decide between these possibilities.

– The second issue is the experiment of Figure 5, in which the authors showed that spinosad treatment of Dα6 KO flies had no effect on lipid level in the hemolymph, in contrast to the increase it induced in wild–type controls. This conclusion is apparently correct, but it does not take in account the fact that the lipid level actually appears at least as high in the Dα6 mutants as in their respective controls treated with spinosad. It seems therefore that the increase in lipid in the hemolymph is more likely caused by Dα6 deficiency than by spinosad treatment. Similarly, the percentage of the area occupied by LD in fat body of both Dα6 mutants is between that of untreated control flies and control flies treated with spinosad. This seems to be consistent with the idea that spinosad only exacerbates an increase in lipids which is for a large part induced by Dα6 deficiency. Incidentally, because Dα6 KO mutants are viable, this observation indicates that altered lipid homeostasis may not be a prominent cause of larval lethality in spinosad–exposed flies.

– My third main comment is that it would be very informative to compare the effects of spinosad in adult wild–type and Dα6 mutant flies, with respect to the survival, bang sensitivity and climbing ability tests (Figure 7), as well as for lipid deposits in the retina (Figure 8). These data would be rather easy to collect and would usefully complete these figures by indicating whether or not spinosad can induce defects in adult flies in the absence of Dα6. A positive answer would argue for the existence of other potential molecular targets of this insecticide.

I have provided more detail on these and other points below.

We thank the reviewer for these comments and have addressed most issues raised (see below).

Main issues

1) Line 117 and 132: from experiments shown in Figure 1D, E, the authors conclude that spinosad exposure "prevent ca2+ flux into neurons expressing nAChRs" and so "blocks the function of Dα6–containing nAChRs". However, Campusano et al. Dev Neurobiol 2007, have shown that thapsigargin treatment reduces nicotine‐evoked calcium increase in Kenyon cells, consistent with a signal amplification by calcium release from intracellular stores. In addition, part of this calcium increase had a voltage‐gated calcium channel–dependent (VGCC) component, indicating that only a fraction of the calcium increase involves calcium influx directly through nAChRs. Therefore, the experiments in Figure 1D, E are not sufficient to decide, in my opinion, whether the reduction in carbachol–induced response in spinosad–treated cells reflects a decrease in calcium influx through nAChRs, or, alternatively, a reduced calcium release from intracellular stores. As mentioned in the public review, this point has to be carefully considered as the decreased response to carbachol could be either a direct or indirect effect of spinosad–induced Dα6 subunit internalization.

The reviewer is correct to state that with the current results it is not possible to determine whether the ca2+ transients reflect reduced influx from internal or external sources. What is clear, however, is that spinosad exposure led to a diminished ca2+ transient. No evidence for stimulus is observed. That is an important observation since past publications would assume that spinosad could provoke a stimulating action on Dα6 receptors. The text was amended in the following way (lines 126-130): “While it was not determined whether the ca2+ transients reflect reduced influx from internal or external sources (Campusano et al., 2007), spinosad exposure led to a diminished ca2+ transient and reduced cholinergic response. Hence, in contrast to imidacloprid, which leads to an enduring ca2+ influx in neurons, spinosad reduces the ca2+ response mediated by Dα6 (Martelli et al., 2020).”

2) Line 263: "We also quantified the level of lipids in hemolymph. Whereas Line 14 and Canton S showed an average 10% and 13% increase in response to spinosad, respectively, neither of the Dα6 KO mutants showed significant changes (Figure 5E)." However, the lipid level appears to be at least as high in the Dα6 mutants as in their respective controls treated by spinosad (Figure 5E). It seems therefore that the increase in lipid in the hemolymph is more likely caused by Dα6 deficiency than by spinosad treatment. Similarly, the percentage of area occupied by LD in fat body of both Dα6 mutants is between that of untreated control flies and control flies treated with spinosad (compare Figure 4B and Figure 5B for the Line 14 background, and compare Figure 5C and Figure 5D for the Canton S background). This seems to be consistent with the idea that spinosad only exacerbates an increase in lipids which is largely induced by Dα6 deficiency. If this is correct, please amend the conclusions accordingly. For example, the sentence on page 25, line 479: "Dα6 knockout mutants exposed to spinosad show no accumulation of LD in the fat body or change of lipid levels in hemolymph indicating that these phenotypes are due to the spinosad:Dα6 interaction (Figure 5)" is not fully correct as it does not mention the fact that Dα6 deficiency by itself appears to increase LD accumulation.

Many sections in the manuscript were restructured, and new datasets added to better characterize the phenotypes caused by Dα6 loss of function mutation and distinguish those from the ones caused by spinosad:Dα6 interaction. While phenotypes in mutants are mild compared to the ones induced by exposure, they are not neglectable and are here for the first time characterized.

3) It would be very informative to repeat the survival and behavior test experiments of Figure 7 in Dα6 mutant background. I was actually surprised that this was not included in the manuscript, as it would not involve a lot of work, and, importantly, it would indicate whether spinosad can induce behavior defects in the absence of Dα6, which would argue for the existence of other molecular targets for this toxin. I would recommend that lipid deposits in the retina (Figure 8A, B) be also quantified in Dα6 background for purpose of comparison.

Data characterizing the survivorship, climbing ability, bang sensitivity (Figure 7D-F) and morphology of retinas (Figure 8E, F) of Dα6 mutants were all added to the manuscript.

– Line 579: "derived from Armenia60 (Drosophila Genomics Resource Center #103394)" – I could not find this Armenia60 line referenced neither in the DGRC or Flybase sites. Please provide the correct source or link for this strain, and explain why it was used as a wild–type control.

Armenia60 is a strain received from the Umea Stock Center in 2001, this centre is now closed however the strain was transferred to DGRC and is called Aashtrak (DGRC #103394). https://kyotofly.kit.jp/cgi-bin/stocks/search_res_det.cgi?DB_NUM=1&DG_NUM=103394. It has been used in several studies as both a neonicotinoid susceptible strain (Perry et al. 2008) and a spinosyn susceptible strain (Somers et al. 2015). Line 14 refers to a specific isofemale derived strain of Armenia60 which do not overexpress the P450 gene Cyp6g1 (Perry et al. 2008), a known mechanism of resistance against neonicotinoids.

– Line 125: "primary culture of neurons " could be "primary culture of third–instar larva brain neurons" to avoid misunderstanding.

– In Figure 1A, the statistical test used should be two–way ANOVA and not t–test, I think, as both the genotypes and time differ in each point, and in Figure 1E one–way ANOVA should be used, as the means of more than two groups were compared. Same issue in Figure 2B and in several other figures of the paper where the Student's t–test was systematically used.

– Figure 1B: I understand that spinosad precludes pupation of intoxicated larvae. However, after eight days, a wild–type third–instar larva should have become an adult fly. So, I guess that authors looked at fly survival, and not larval survival, for the wild–type. Please mention it, if it is right. In Figure 1C, did you treat the pupae or the larvae before pupal case formation with the toxin? This could be stated in the legend.

– Line 581: "Expression of nAChR–Dα6 gene" – According to Flybase, this gene should be named: nAChRα6. The correct nomenclature should be mentioned at least once in the manuscript.

– Line 582: "the MitoTimer line (Gottlieb and Stotland, 2015) was used". The exact reference for this line is: Laker et al. J Biol Chem. (2014) 289(17):12005–15. Same correction required on line 187.

– Line 594: "In all experiments involving adult flies only females were used to maintain consistency." Adult Drosophila females have very different physiology compared to males. This is particularly relevant for reproductive organs, of course, but also for the gut and nervous system. Therefore, this particularity of this study is not negligible and should be mentioned in the abstract.

[Editors’ note: what follows is the authors’ response to the second round of review.]

Essential revisions:

1) While this revised version is definitely improved, a sticky point remains, namely that of the relationship between the lysosomal disorder and the production of ROS by mitochondria, which is not rigorously established. The authors use NACA to block the production of ROS. Since presumably NACA has a ubiquitous effect, the authors cannot formally exclude that the ROS originate from a source distinct from mitochondria. In this respect, it is noteworthy that the authors did not investigate the impact of NACA treatment on mitochondria. This reviewer does not understand the argument put forth by the authors as regards the lack of subcellular resolution of mito-roGFP reporters. Because the reporters are targeted to mitochondria, it follows that any signal reflects redox conditions in mitochondria.

Following the reviewer’s suggestion we used the mito-roGFP2-Orp1 strain (BDSC #67672) to investigate the mitochondrial origin of ROS. Figure 3 —figure supplement 1 was included, and the text was amended in the following way (lines 185-192): “The mito-roGFP2-Orp1 construct, an in vivo mitochondrial H2O2 reporter (Albrecht et al., 2011), was used to identify the origin of ROS induced by spinosad exposure at 2.5 ppm for 2 hrs. A subtle, but significant increase in the 405 (oxidized mitochondria signal)/488 (reduced mitochondria signal) ratio was observed in the brain (20% on average) and anterior midgut (10% on average) (Figure 3 —figure supplement 1), pointing to a rise in H2O2 generation upon a few hours of exposure. Similarly, to the MitoTimer reporter, an increase in the oxidized mitochondrial signal was accompanied by the increase in the reduced mitochondrial signal, accounting for the subtle increase in 405/488 ratio.”

2) It may be hazardous to rely only on chemicals to study the effects of ROS. Genetic manipulation of ROS by SOD2 (and eventually SOD1 as control), can clarify whether mitochondria are the ROS source and further demonstrate whether ROS is the cause of other phenotypes. As this superoxide dismutase functions in mitochondria, it should quench the generation of ROS induced by spinosad and thus provide an independent, more convincing, demonstration of the role of mitochondria in spinosad-induced toxicity.

Following the reviewer’s suggestion we used an elav-GAL4 driver (BDSC #458) to express Sod2 (BDSC #24494) and Sod1 (BDSC #24750) in the nervous system and investigate the mitochondrial origin of ROS. Figure 4D, E was included, and the text was amended in the following way (lines 248-254): “To assess the mitochondrial origin of the ROS measured with DHE, flies expressing the superoxide dismutase gene Sod2 in the nervous system with the elav-GAL4 driver were exposed to 2.5 ppm spinosad for 2 hr. Sod2 is the main ROS scavenger in Drosophila and is localized to mitochondria (Missirlis et al., 2003). Sod1 is present in the cytosol (Missirlis et al., 2003), and expression of this gene was used as a control. While exposure to spinosad caused an average 63% increase in ROS levels in control larvae overexpressing Sod1, an average increase of only 28% was found in larvae overexpressing Sod2 (Figure 4D, E).”. We also added (lines 569-576): “While we cannot rule out the generation of ROS by other mechanisms, we provide compelling evidence that a significant part of ROS that is generated by spinosad exposure is of mitochondrial origin, arguing that mitochondrial impairment is a key element of spinosad mode of action at low dose exposure. The evidence is based on increased mitochondrial turnover and mito-roGFP 405/488 ratio, reduced activity of the ROS sensitive enzyme m-aconitase and reduced ATP levels (Figure 3). In addition, we observed a highly significant reduction of cardiolipin levels (Figure 6C) typically associated with defects in the electron transport chain and increased ROS production as they are required for the anchoring of Complex1 in mitochondria (Dudek, 2017; Quintana et al., 2010).”

3) The authors did not fully address previous issue 3 of reviewer 2, most likely due to a misunderstanding. What I asked for was that the authors check whether spinosad can induce or not behavior defects (bang sensitivity and climbing ability) in adult Dα6 mutants and whether spinosad can cause lipid deposits in the retina of Dα6 mutants. Instead, the authors characterized behavior defects and retina structure in adult Dα6 mutants in the absence of spinosad, which are interesting but incomplete observations in my opinion. In other experiments (e.g. Figure 4D and E, and Figure 5B and G), the authors have indeed compared the effects of spinosad in control and Dα6 mutants. Similarly, I think that it would be worth doing the same test (i.e. check the effects or lack of effects of spinosad on adult Dα6 mutants) for the experiments described in Figure 7A-C and Figure 8A-D.

Following the reviewer’s suggestion we have now included experiments to assess the effects of spinosad exposure on viability and behavioral phenotypes of Dα6 mutant flies. Figure 7E-G was included, and the text was amended in the following way (lines 395-408): “The same phenotypes were also assessed in adult female virgin Dα6 KO mutants. Unexposed mutant flies show a significant reduction in longevity compared to unexposed wild-type individuals, but that difference is only noticeable later in life, Canton-S Dα6 KO mutants have a median survival of 68 days compared to 82 for Canton-S (Figure 7D). A 25 day exposure to 0.2 ppm spinosad leads to a 91% survival of Canton-S Dα6 KO mutants whereas only 40% of Canton-S wild-type flies survive this exposure (Figure 7E). No changes in bang sensitivity or climbing ability were observed between exposed and unexposed Dα6 KO mutants over the course of a 20 day exposure (Figure 7F, G). However, at the 20 day time-point, twice as many of unexposed Dα6 KO mutants failed to climb (36%) compared to unexposed Canton-S wild-type flies (18%). (Figure 7G). These data support that the deleterious effects of spinosad are mediated by its binding to Dα6 receptors. They also indicate that loss of Dα6 by itself causes mild but significant behavioral and lifespan phenotypes not previously reported.”

https://doi.org/10.7554/eLife.73812.sa2

Article and author information

Author details

  1. Felipe Martelli

    School of BioSciences, The University of Melbourne, Melbourne, Australia
    Present address
    School of Biological Sciences, Monash University, Melbourne, Australia
    Contribution
    Conceptualization, Data curation, Formal analysis, Investigation, Validation, Visualization, Writing – original draft, Writing – review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-4783-9025

  2. Natalia H Hernandes

    School of BioSciences, The University of Melbourne, Melbourne, Australia
    Contribution
    Formal analysis, Investigation, Writing – review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-5644-6974

  3. Zhongyuan Zuo

    Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, United States
    Contribution
    Formal analysis, Investigation, Visualization, Writing – review and editing
    Competing interests
    No competing interests declared

  4. Julia Wang

    Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, United States
    Present address
    Medical Scientist Training Program, Baylor College of Medicine, Houston, United States
    Contribution
    Investigation, Writing – review and editing
    Competing interests
    No competing interests declared

  5. Ching-On Wong

    Department of Integrative Biology and Pharmacology, McGovern Medical School at the University of Texas Health Sciences Center, Houston, United States
    Present address
    Department of Biological Sciences, Rutgers University, Newark, United States
    Contribution
    Formal analysis, Investigation, Writing – review and editing
    Competing interests
    No competing interests declared

  6. Nicholas E Karagas

    Department of Integrative Biology and Pharmacology, McGovern Medical School at the University of Texas Health Sciences Center, Houston, United States
    Contribution
    Formal analysis, Investigation
    Competing interests
    No competing interests declared

  7. Ute Roessner

    School of BioSciences, The University of Melbourne, Melbourne, Australia
    Contribution
    Investigation, Writing – review and editing
    Competing interests
    No competing interests declared

  8. Thusita Rupasinghe

    School of BioSciences, The University of Melbourne, Melbourne, Australia
    Contribution
    Data curation, Formal analysis, Investigation
    Competing interests
    No competing interests declared

  9. Charles Robin

    School of BioSciences, The University of Melbourne, Melbourne, Australia
    Contribution
    Investigation, Writing – review and editing
    Competing interests
    No competing interests declared

  10. Kartik Venkatachalam

    Department of Integrative Biology and Pharmacology, McGovern Medical School at the University of Texas Health Sciences Center, Houston, United States
    Contribution
    Funding acquisition, Investigation, Writing – review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-3055-9265

  11. Trent Perry

    School of BioSciences, The University of Melbourne, Melbourne, Australia
    Contribution
    Conceptualization, Funding acquisition, Writing – review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-8045-0487

  12. Philip Batterham

    School of BioSciences, The University of Melbourne, Melbourne, Australia
    Contribution
    Conceptualization, Funding acquisition, Investigation, Supervision, Writing – review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-9840-9119

  13. Hugo J Bellen

    1. Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, United States
    2. Neurological Research Institute, Texas Children Hospital, Houston, United States
    3. Howard Hughes Medical Institute, Baylor College of Medicine, Houston, United States
    Contribution
    Conceptualization, Funding acquisition, Investigation, Supervision, Writing – review and editing
    For correspondence
    hbellen@bcm.edu
    Competing interests
    Reviewing editor, eLife
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-5992-5989

Funding

Howard Hughes Medical Institute

  • Hugo J Bellen

Victoria State Government (Latin America Doctoral Scholarship)

  • Felipe Martelli

University of Melbourne (Alfred Nicholas Fellowship)

  • Felipe Martelli

University of Melbourne (Faculty of Science Travelling Scholarship)

  • Felipe Martelli

The Robert Johanson and Anne Swann Fund (Native Animals Trust)

  • Felipe Martelli
  • Trent Perry

University of Melbourne

  • Philip Batterham

National Institute on Aging

  • Kartik Venkatachalam

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

The pCyPet-His plasmid was a gift from Patrick Daugherty (Addgene plasmid # 14030). FM was supported by a Victorian Latin America Doctoral Scholarship, an Alfred Nicholas Fellowship, a UoM Faculty of Science Travelling Scholarship, and The Robert Johanson and Anne Swann Fund – Native Animals Trust (awarded to FM and TP). PB was supported by the University of Melbourne. HJB was supported by the Howard Hughes Medical Institute (HHMI) and is an investigator of HHMI. KV was supported by NIH (NIA) grant. Lipid analyses were performed at Metabolomics Australia at University of Melbourne, which is a National Collaborative Research Infrastructure Strategy initiative under Bioplatforms Australia Pty Ltd (http://www.bioplatforms.com/).

Senior Editor

  1. Utpal Banerjee, University of California, Los Angeles, United States

Reviewing Editor

  1. Bruno Lemaître, École Polytechnique Fédérale de Lausanne, Switzerland

Publication history

  1. Preprint posted: February 4, 2021 (view preprint)
  2. Received: September 11, 2021
  3. Accepted: January 28, 2022
  4. Version of Record published: February 22, 2022 (version 1)

Copyright

© 2022, Martelli et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 7,562
    Page views
  • 341
    Downloads
  • 8
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Felipe Martelli
  2. Natalia H Hernandes
  3. Zhongyuan Zuo
  4. Julia Wang
  5. Ching-On Wong
  6. Nicholas E Karagas
  7. Ute Roessner
  8. Thusita Rupasinghe
  9. Charles Robin
  10. Kartik Venkatachalam
  11. Trent Perry
  12. Philip Batterham
  13. Hugo J Bellen
(2022)
Low doses of the organic insecticide spinosad trigger lysosomal defects, elevated ROS, lipid dysregulation, and neurodegeneration in flies
eLife 11:e73812.
https://doi.org/10.7554/eLife.73812

Categories and tags

Research organism

Further reading

    1. Neuroscience

    Cholinergic and noradrenergic axonal activity contains a behavioral-state signal that is coordinated across the dorsal cortex

    Lindsay Collins, John Francis ... David A McCormick
    Research Article Updated

    Fluctuations in brain and behavioral state are supported by broadly projecting neuromodulatory systems. In this study, we use mesoscale two-photon calcium imaging to examine spontaneous activity of cholinergic and noradrenergic axons in awake mice in order to determine the interaction between arousal/movement state transitions and neuromodulatory activity across the dorsal cortex at distances separated by up to 4 mm. We confirm that GCaMP6s activity within axonal projections of both basal forebrain cholinergic and locus coeruleus noradrenergic neurons track arousal, indexed as pupil diameter, and changes in behavioral engagement, as reflected by bouts of whisker movement and/or locomotion. The broad coordination in activity between even distant axonal segments indicates that both of these systems can communicate, in part, through a global signal, especially in relation to changes in behavioral state. In addition to this broadly coordinated activity, we also find evidence that a subpopulation of both cholinergic and noradrenergic axons may exhibit heterogeneity in activity that appears to be independent of our measures of behavioral state. By monitoring the activity of cholinergic interneurons in the cortex, we found that a subpopulation of these cells also exhibit state-dependent (arousal/movement) activity. These results demonstrate that cholinergic and noradrenergic systems provide a prominent and broadly synchronized signal related to behavioral state, and therefore may contribute to state-dependent cortical activity and excitability.

    1. Neuroscience

    Early language exposure affects neural mechanisms of semantic representations

    Xiaosha Wang, Bijun Wang, Yanchao Bi
    Research Article Updated

    One signature of the human brain is its ability to derive knowledge from language inputs, in addition to nonlinguistic sensory channels such as vision and touch. How does human language experience modulate the mechanism by which semantic knowledge is stored in the human brain? We investigated this question using a unique human model with varying amounts and qualities of early language exposure: early deaf adults who were born to hearing parents and had reduced early exposure and delayed acquisition of any natural human language (speech or sign), with early deaf adults who acquired sign language from birth as the control group that matches on nonlinguistic sensory experiences. Neural responses in a semantic judgment task with 90 written words that were familiar to both groups were measured using fMRI. The deaf group with reduced early language exposure, compared with the deaf control group, showed reduced semantic sensitivity, in both multivariate pattern (semantic structure encoding) and univariate (abstractness effect) analyses, in the left dorsal anterior temporal lobe (dATL). These results provide positive, causal evidence that language experience drives the neural semantic representation in the dATL, highlighting the roles of language in forming human neural semantic structures beyond nonverbal sensory experiences.

    1. Neuroscience

    Two conserved vocal central pattern generators broadly tuned for fast and slow rates generate species-specific vocalizations in Xenopus clawed frogs

    Ayako Yamaguchi, Manon Peltier
    Research Article Updated

    Across phyla, males often produce species-specific vocalizations to attract females. Although understanding the neural mechanisms underlying behavior has been challenging in vertebrates, we previously identified two anatomically distinct central pattern generators (CPGs) that drive the fast and slow clicks of male Xenopus laevis, using an ex vivo preparation that produces fictive vocalizations. Here, we extended this approach to four additional species, X. amieti, X. cliivi, X. petersii, and X. tropicalis, by developing ex vivo brain preparation from which fictive vocalizations are elicited in response to a chemical or electrical stimulus. We found that even though the courtship calls are species-specific, the CPGs used to generate clicks are conserved across species. The fast CPGs, which critically rely on reciprocal connections between the parabrachial nucleus and the nucleus ambiguus, are conserved among fast-click species, and slow CPGs are shared among slow-click species. In addition, our results suggest that testosterone plays a role in organizing fast CPGs in fast-click species, but not in slow-click species. Moreover, fast CPGs are not inherited by all species but monopolized by fast-click species. The results suggest that species-specific calls of the genus Xenopus have evolved by utilizing conserved slow and/or fast CPGs inherited by each species.