Functional impact of subunit composition and compensation on Drosophila melanogaster nicotinic receptors–targets of neonicotinoids

Abstract

Neonicotinoid insecticides target insect nicotinic acetylcholine receptors (nAChRs) and their adverse effects on non-target insects are of serious concern. We recently found that cofactor TMX3 enables robust functional expression of insect nAChRs in Xenopus laevis oocytes and showed that neonicotinoids (imidacloprid, thiacloprid, and clothianidin) exhibited agonist actions on some nAChRs of the fruit fly (Drosophila melanogaster), honeybee (Apis mellifera) and bumblebee (Bombus terrestris) with more potent actions on the pollinator nAChRs. However, other subunits from the nAChR family remain to be explored. We show that the Dα3 subunit co-exists with Dα1, Dα2, Dβ1, and Dβ2 subunits in the same neurons of adult D. melanogaster, thereby expanding the possible nAChR subtypes in these cells alone from 4 to 12. The presence of Dα1 and Dα2 subunits reduced the affinity of imidacloprid, thiacloprid, and clothianidin for nAChRs expressed in Xenopus laevis oocytes, whereas the Dα3 subunit enhanced it. RNAi targeting Dα1, Dα2 or Dα3 in adults reduced expression of targeted subunits but commonly enhanced Dβ3 expression. Also, Dα1 RNAi enhanced Dα7 expression, Dα2 RNAi reduced Dα1, Dα6, and Dα7 expression and Dα3 RNAi reduced Dα1 expression while enhancing Dα2 expression, respectively. In most cases, RNAi treatment of either Dα1 or Dα2 reduced neonicotinoid toxicity in larvae, but Dα2 RNAi enhanced neonicotinoid sensitivity in adults reflecting the affinity-reducing effect of Dα2. Substituting each of Dα1, Dα2, and Dα3 subunits by Dα4 or Dβ3 subunit mostly increased neonicotinoid affinity and reduced efficacy. These results are important because they indicate that neonicotinoid actions involve the integrated activity of multiple nAChR subunit combinations and counsel caution in interpreting neonicotinoid actions simply in terms of toxicity.

Author summary

In this paper, we show that the Drosophila melanogaster nicotinic acetylcholine receptor (nAChR) Dα3 subunit is co-expressed in ejaculatory duct neurons with Dα1, Dα2, Dβ1, and Dβ2 subunits. All 5 subunits combine to form 12 functional nAChRs in Xenopus laevis oocytes. The functional expression of 18 nAChRs generated from combinations of subunits Dα1−4 and Dβ1−3 are also reported. Dα1 and Dα2 reduced the affinity of D. melanogaster heteromeric nAChRs for imidacloprid, thiacloprid, and clothianidin, whereas Dα3 enhanced it. RNAi of Dα1, Dα2 or Dα3 in adult flies reduced expression of the targeted subunits but commonly enhanced Dβ3 expression; other subunits were also affected in some cases. RNAi targeting either Dα1 or Dα2 reduced neonicotinoid toxicity in larvae but targeting Dα2 led to hyper-neonicotinoid sensitivity in adults consistent with the affinity-reducing effect on neonicotinoids of Dα2. Since RNAi induced subunit compensation was detected, each of Dα1, Dα2, and Dα3 subunits was substituted by Dα4 or Dβ3 subunit. Such subunit compensation mostly increased neonicotinoid affinity and reduced efficacy, impairing the climbing ability of the flies. These results are important because they indicate that neonicotinoid action and toxicity involve the integrated actions of multiple nAChR subunit combinations and counsel caution in interpreting neonicotinoid actions in terms of reduced toxicity.

Introduction

The nicotinic acetylcholine receptors (nAChRs) are cys-loop ligand-gated cation channels playing a pivotal role in fast cholinergic neurotransmission [1]. In mammals, nAChRs underlie memory [2], learning [2], circadian rhythm [3] and immune responses [4] as well as locomotion [5] and hearing [6, 7]. In insects, nAChRs are involved primarily in afferent synaptic transmission [8]. Roles for insect nAChRs include escape responses [9], circadian rhythm [10,11] and regulation of the mating-induced germline stem cell growth [12]. Hence, several classes of synthetic and natural-product based insecticides target nAChRs [13–18].

Neonicotinoid insecticides targeting insect nAChRs are effective, broad-spectrum insecticides [16,19–23]. Following the discovery of the nitromethylene heterocyclic compound nithiazine, this initial lead compound was modified extensively resulting in 3 generations of commercial neonicotinoids [15]. They exhibit higher affinity for insect over vertebrate nAChRs, thereby resulting in selective toxicity to insects [24,25]. Their high systemic activity in plants has permitted seed treatment which has accelerated their deployment for crop protection. However, potential adverse effects on pollinators such as honeybees, bumblebees and solitary bees are a concern [16,20,26,27], even though reduced numbers of pollinators and other non-target insects are not simply due to the effects of neonicotinoids [16,26]. Adverse effects on aquatic invertebrates and birds are also reported [28,29]. It is vital to understand in detail the mechanism of insect nAChR-neonicotinoid interactions but until recently that was precluded by the difficulty of heterologously expressing robust insect nAChRs.

The recent finding that the transmembrane thioredoxin-related protein 3 (TMX3) enables robust functional expression of insect nAChRs in X. laevis oocytes [20, 30, 31] permitted a demonstration that honeybee (A. mellifera) and bumblebee (B. terrestris) α1/α8/β1 and α1/α2/α8/β1 nAChRs were more sensitive to neonicotinoids than the fruit fly (D. melanogaster) Dα1/Dβ1, Dα1/Dα2/Dβ1, Dα1/Dβ1/Dβ2, and Dα1/Dα2/Dβ1/Dβ2 nAChRs. Thiacloprid and clothianidin modulated honeybee and bumblebee nAChRs at picomolar concentrations, lower than those commonly encountered in treated fields [30]. However, the extent to which other subunits participate in the formation of native nAChRs and how that affects neonicotinoid actions is not known.

To begin to address this shortfall in understanding, we first showed that the D. melanogaster Dα3 subunit co-exists with the Dα1, Dα2, Dβ1, and Dβ2 subunits in ejaculatory duct neurons of adult D. melanogaster. Co-expression in X. laevis with the Dα3 subunit increased possible nAChR subtypes from 4 to 12. All the 12 recombinant fruit fly nAChRs will be explored and their sensitivity to the transmitter acetylcholine (ACh), neonicotinoids (imidacloprid, thiacloprid, and clothianidin) and α-bungarotoxin compared. The impact of the Dα1, Dα2, and Dα3 subunits on recombinant nAChR receptor affinity is addressed. The impact of RNAi targeting particular subunits on toxicity to larvae and adult toxicity and behaviour is investigated. Finally, we studied the effects of replacing one of the α subunits in the Dα1/Dα2/Dα3/Dβ1/Dβ2 nAChRs by either the Dα4 or the Dβ3 subunit on the neonicotinoid actions to address whether such subunit compensation further expands the diversity of neonicotinoid actions in insects.

Results and discussion

We first examined whether the Dα3 subunit is co-expressed with Dα1, Dα2, Dβ1, and Dβ2 subunits, previously shown to be present in ejaculatory neurons of D. melanogaster [30], and if so, how such co-expression influences actions of neonicotinoids in vitro and in vivo. To analyse the expression of Dα3, we stained male ejaculatory neurons with a tyrosine decarboxylase 2 targeting antibody (anti-Tdc2) in the animals expressing GFP under control of Dα3-T2A-Gal4 and found that Dα3 is indeed expressed in the male ejaculatory neurons (Fig 1A). Another recent study has shown similar findings for the oviduct neurons in female fruit flies [12], suggesting that Dα1, Dα2, Dα3, Dβ1, and Dβ2 subunits can potentially generate diverse heteromeric nAChRs in male and female adult neurons involved in reproductive functions.

Fig 1. Expression of Dα3 in male ejaculatory duct neurons of D. melanogaster and characteristics of nAChRs expressed in X. laevis oocytes.

(A) Dα3 is expressed in the ejaculatory ducts of male flies. Male ejaculatory ducts from Dα3-2A-Gal4>UAS-mCD8::GFP were immunostained for Tdc2 (magenta) and GFP (green). If both signals are overlapped, merged signals are shown in white. Scale bar: 25 μm. The Dα3 subunit is expressed in the neurons where Dα1, Dα2, Dβ1, and Dβ2 subunits are also expressed [30]. (B) Current responses to 100 μM ACh of nAChRs reconstructed with Dα1, Dα2, Dα3, Dβ1, and Dβ2 subunits in X. laevis oocytes. Each box plots represents 75 and 25% percentiles of data and horizonal line in each box indicates the median of data (n = 10). Whiskers indicate the range of data. The current amplitude of the ACh-induced response was compared by Kruskal-Wallis tests (*, P < 0.05; **, P < 0.01). ns: not significant.

Given their co-localisation in certain neurons, we investigated in terms of responses to bath-applied 100 μM ACh how many kinds of robust, functional nAChRs the five subunits can reconstitute in X. laevis oocytes when co-expressed with co-factors DmRIC-3, DmUNC-50, and DmTMX3 [30]. We found that by co-expressing the Dα3 subunit, 8 more robust nAChRs (Dα3/Dβ1, Dα1/Dα3/Dβ1, Dα2/Dα3/Dβ1, Dα1/Dα2/Dα3/Dβ1, Dα3/Dβ1/Dβ2, Dα1/Dα3/Dβ1/Dβ2, Dα2/Dα3/Dβ1/Dβ2, and Dα1/Dα2/Dα3/Dβ1/Dβ2 nAChRs) were generated in addition to previously described Dα1/Dβ1, Dα1/Dα2/Dβ1, Dα1/Dβ1/Dβ2, and Dα1/Dα2/Dβ1/Dβ2 nAChRs (Fig 1B). We then determined concentration-response relationships for ACh on the 12 nAChRs (Fig A in S1 Text and Table 1). Replacing either the Dα1 or Dα2 subunit by the Dα3 subunit led to increased current amplitude of the response to 100 μM ACh. For example, the ACh response amplitudes of the Dα3/Dβ1 and Dα1/Dα3/Dβ1 nAChRs were 58.2 and 7.1-fold larger than those of Dα1/Dβ1 and Dα1/Dα2/Dβ1 nAChRs, respectively (ANOVA, P < 0.05, n = 10, Fig 1B). The affinity in terms of pEC₅₀ for ACh varied from 4.14 to 6.14 depending on subunit combinations (Fig 2A, Table 1, and Table A in S1 Text), indicative of 12 distinct nAChRs. Notably, the Dα1/Dα2/Dα3/Dβ1/Dβ2 nAChR is the first functional recombinant insect nAChR consisting of five different subunits. It is also the first time that all possible subunit combinations for a single insect neuron have been reported.

Table 1. Agonist actions of ACh and neonicotinoids on D. melanogaster nAChRs expressed in X. laevis oocytes^*.

	ACh	Imidacloprid		Thiacloprid		Clothianidin
Subunits	pEC50	pEC50	Imax	pEC50	Imax	pEC50	Imax
Dα1/Dβ1	5.24 ± 0.04	7.16 ± 0.15	0.122 ± 0.007	7.79 ± 0.21	0.091 ± 0.009	7.13 ± 0.10	0.508 ± 0.015
Dα3/Dβ1	6.14 ± 0.06	8.32 ± 0.21	0.032 ± 0.003	7.80 ± 0.08	0.016 ± 0.001	7.54 ± 0.04	0.691 ± 0.013
Dα1/Dα3/Dβ1	5.85 ± 0.03	7.58 ± 0.23	0.170 ± 0.020	8.07 ± 0.20	0.047 ± 0.006	7.20 ± 0.09	0.627 ± 0.025
Dα1/Dα2/Dβ1	4.14 ± 0.05	6.71 ± 0.14	0.031 ± 0.002	6.92 ± 0.08	0.025 ± 0.001	5.07 ± 0.11	0.699 ± 0.034
Dα2/Dα3/Dβ1	5.45 ± 0.07	6.88 ± 0.29	0.101 ± 0.016	6.92 ± 0.19	0.045 ± 0.006	6.81 ± 0.08	0.882 ± 0.076
Dα1/Dα2/Dα3/Dβ1	5.43 ± 0.05	7.08 ± 0.15	0.094 ± 0.007	7.73 ± 0.19	0.047 ± 0.004	6.55 ± 0.07	0.817 ± 0.024
Dα1/Dβ1/Dβ2	4.75 ± 0.04	6.99 ± 0.18	0.076 ± 0.006	7.86 ± 0.10	0.043 ± 0.002	6.64 ± 0.10	0.360 ± 0.016
Dα3/Dβ1/Dβ2	6.19 ± 0.07	8.28 ± 0.09	0.039 ± 0.002	8.32 ± 0.21	0.022 ± 0.003	7.46 ± 0.08	0.492 ± 0.016
Dα1/Dα3/Dβ1/Dβ2	6.43 ± 0.02	7.52 ± 0.12	0.330 ± 0.018	8.00 ± 0.25	0.107 ± 0.015	7.40 ± 0.07	0.684 ± 0.022
Dα1/Dα2/Dβ1/Dβ2	4.92 ± 0.05	6.45 ± 0.21	0.165 ± 0.016	7.10 ± 0.30	0.074 ± 0.014	6.31 ± 0.10	0.800 ± 0.033
Dα2/Dα3/Dβ1/Dβ2	5.99 ± 0.07	7.00 ± 0.29	0.125 ± 0.018	8.08 ± 0.33	0.017 ± 0.002	7.29 ± 0.07	0.634 ± 0.020
Dα1/Dα2/Dα3/Dβ1/Dβ2	4.49 ± 0.04	6.47 ± 0.14	0.142 ± 0.011	7.18 ± 0.17	0.059 ± 0.006	6.59 ± 0.07	0.812 ± 0.019
Dα2/Dα3/Dα4/Dβ1/Dβ2	5.49 ± 0.06	7.40 ± 0.20	0.027 ± 0.002	7.99 ± 0.20	0.018 ± 0.002	7.11 ± 0.12	0.912 ± 0.041
Dα2/Dα3/Dβ1/Dβ2/Dβ3	6.01 ± 0.04	7.75 ± 0.31	0.021 ± 0.003	8.31 ± 0.17	0.010 ± 0.001	7.46 ± 0.06	0.695 ± 0.014
Dα1/Dα3 Dα4/Dβ1/Dβ2	4.79 ± 0.04	7.30 ± 0.17	0.101 ± 0.008	7.96 ± 0.10	0.026 ± 0.001	6.94 ± 0.06	0.425 ± 0.008
Dα1/Dα3/Dβ1/Dβ2/Dβ3	5.83 ± 0.02	7.72 ± 0.11	0.140 ± 0.008	8.63 ± 0.23	0.089 ± 0.009	7.60 ± 0.12	0.531 ± 0.026
Dα1/Dα2/Dα4/Dβ1/Dβ2	4.64 ± 0.05	6.48 ± 0.28	0.172 ± 0.020	7.01 ± 0.19	0.066 ± 0.007	6.72 ± 0.09	1.001 ± 0.031
Dα1/Dα2/Dβ1/Dβ2/Dβ3	4.42 ± 0.02	6.27 ± 0.21	0.080 ± 0.007	7.71 ± 0.43	0.023 ± 0.004	5.67 ± 0.11	0.974 ± 0.041

*All data are shown as mean ± standard error (n = 5). For this study, we determined pEC₅₀ and I_max values of the ligands for the Dα1/Dβ1, Dα1/Dα2/Dβ1, Dα1/Dα2/Dβ1/Dβ2 and Dα1/Dα2/Dβ1/Dβ2 nAChRs for which data were reported previously [30], because it was necessary to compare all the data determined using the same protocol (see Methods).

Fig 2. Concentrations-response relationships of ACh for D. melanogaster nAChRs expressed in X. laevis oocytes and actions of α-BTX on the ACh-induced responses of nAChRs.

(A) Concentration-response curves of ACh and (B) heatmap representing α-BTX for the 12 nAChRs expressed in X. laevis oocytes. In (A), each data plot indicates the mean ± standard error (n = 5). In (B), high to low α-BTX sensitivity is shown in blue and white, respectively. The Dα1 subunit underpins the α-BTX sensitivity of the nAChRs tested. See Fig B in S1 Text for the ACh-induced currents measured in the absence and presence of α-BTX (Fig Ba in S1 Text) and bar graph representations of the effects of α-BTX (Fig Bb in S1 Text).

α-Bungarotoxin (α-BTX), a peptide toxin known to block certain insect nAChRs [32], was tested on D. melanogaster nAChRs expressed in X. laevis oocytes. We found that 100 nM α-BTX effectively blocked the ACh-induced responses of nAChRs containing the Dα1 subunit (Fig 2B, and Fig B in S1 Text). In contrast, α-BTX was ineffective on nAChRs lacking the Dα1 subunit. Notably, α-BTX showed a minimal blocking effect on the Dα3/Dβ1, Dα2/Dα3/Dβ1, and Dα3/Dβ1/Dβ2 nAChRs (Fig 2B and Fig B in S1 Text). These findings accord with an earlier observation that Dα1/chicken β2 nAChR was sensitive whereas Dα2/chicken β2 nAChR was resistant to this neurotoxin [33]. Diverse pharmacology in terms of the sensitivity to α-BTX confirms that 12 distinct, robust, and functional nAChRs results from combinatorial assembly from the five subunits.

We measured agonist activity of imidacloprid, thiacloprid, and clothianidin, in terms of their pEC₅₀ and I_max values for 8 Dα3-containing nAChRs and analysed factors determining them (Fig 3A, Fig C in S1 Text and Table 1). The pEC₅₀ value for each neonicotinoid relies primarily on the subunit properties (Fig 3B and Table 1). For imidacloprid, the EC₅₀ value for the Dα3/Dβ1 nAChR was 14.4-fold lower than that for the Dα1Dβ1 nAChR. Similarly, substituting either Dα1 or Dα2 subunit by Dα3 subunit enhanced affinity of imidacloprid and clothianidin (Dα1/Dβ1/Dβ2 nAChR (for imidacloprid, pEC₅₀ = 6.99, EC₅₀ = 102 nM; for clothianidin, pEC₅₀ = 6.64, EC₅₀ = 229 nM) vs Dα3/Dβ1/Dβ2 nAChR (for imidacloprid, pEC₅₀ = 8.28, EC₅₀ = 5.25 nM; for clothianidin, pEC₅₀ = 7.46, EC₅₀ = 34.7 nM), Fig 3B, Table 1, and Tables B and F in S1 Text for ANOVA analysis). For thiacloprid, the affinity for the Dα1/Dα3/Dβ1 nAChR (pEC₅₀ = 8.07, EC₅₀ = 8.51 nM) was higher than that for the Dα1/Dα2/Dβ1 nAChR (pEC₅₀ = 6.92, EC₅₀ = 120 nM, Fig 3B, Table 1, and Table D in S1 Text for ANOVA analysis). Inversely, the Dα2 subunit reduced the affinity of neonicotinoids (imidacloprid and clothianidin, Dα2/Dα3/Dβ1 nAChR < Dα3/Dβ1nAChR; thiacloprid, Dα2/Dα3/Dβ1 nAChR < Dα1/Dα3/Dβ1 nAChR, Tables B, D, and F in S1 Text for ANOVA analysis). Compound properties also contribute to determining the affinity as indicated by the highest pEC₅₀ values of thiacloprid for most of the nAChRs (Fig 3A and 3B, Table 1, Tables B, D, and F in S1 Text for ANOVA analysis). Such compound factors were more evident in the I_max values (Fig 3C, Table 1, and Tables C, E, and G in S1 Text for ANOVA analysis). For all the nAChRs tested, the order of I_max was clothianidin > imidacloprid > thiacloprid, similar to the efficacy order observed in the fruit fly neurons [34], which supports the utility of using the X. laevis oocytes to express nAChRs for the evaluation of neonicotinoid actions in insects.

Fig 3. Concentration-response relationships for agonist activity of imidacloprid, thiacloprid, and clothianidin for D. melanogaster nAChRs expressed in X. laevis oocytes and heatmap representations of the affinity and efficacy of neonicotinoids.

(A) Concentration-response relationships of agonist activity of neonicotinoids for D. melanogaster nAChRs. Each data plot represents mean ± standard error (n = 5). (B) Heatmap representation of the affinity in terms of pEC₅₀ values for the neonicotinoids tested. (C) Heatmap representation of the efficacy in terms of I_max values for the neonicotinoids tested. See Table 2 for the results of multivariate analyses of subunit and ligand factors governing variations in pEC₅₀ and I_max values.

To clarify the relationship of the nAChR subunits and the neonicotinoids tested with the affinity and efficacy of neonicotinoids for the 12 fruit fly nAChRs (Table 1), we quantitatively analysed the factors governing the variations in the agonist activity indices (Table 2, and Table H in S1 Text for parameter and data sets). The adjusted coefficients of Dα1 and Dα2 subunits for affinity were -0.306 and -0.754, respectively (Table 2), suggesting that both subunits reduced the neonicotinoid affinity, the Dα2 contribution being higher than the Dα1 contribution, while the coefficient of Dα3 was 0.524, indicating that the subunit enhanced affinity (Table 2). Also, the compound properties underpin the affinity (Table 2). It was impossible to elucidate the contribution of the Dβ1 subunit since it is common to all the nAChRs being an essential subunit. However, we showed previously that the R81T mutation in the Dβ1 subunit strikingly reduced the affinity and efficacy of the neonicotinoids [30], indicating its critical role in determining neonicotinoid action [16, 35, 36]. On the other hand, the I_max relied mainly on the compound properties even though the values also varied with subunit composition (Fig 2C and Table 2). The highest efficacy of clothianidin probably results from hydrogen bond formation of NH of its guanidine moiety with the backbone carbonyl of the tryptophan in loop B conserved in the insect α subunits [37].

Table 2. Multivariate analysis of factors determining the affinity (pEC₅₀) and efficacy (I_max) of neonicotinoids.

	pEC50	Imax
Factors	Coefficients (95% CI)a	Coefficients (95% CI)a
Subunits
Dα1	-0.306 (-0.567 − -0.045)*	0.053 (-0.008 − 0.113)
Dα2	-0.754 (-0.967 − -0.541)*	0.046 (-0.003 − 0.096)
Dα3	0.524 (0.264 − 0.785)*	0.059 (-0.002 − 0.119)
Dβ2	0.092 (-0.121 − 0.305)	-0.003 (-0.052 − 0.047)
Compounds
Acetylcholine	Reference	Reference
Imidacloprid	1.414 (1.113 − 1.715)*	-0.333 (-0.403 − -0.263)*
Thiacloprid	1.785 (1.484 − 2.086)*	-0.881 (-0.951 − -0.811)*
Clothianidin	2.229 (1.928 − 2.530)*	-0.951 (-1.020 − -0.881)*

Data number and degree of freedom were 48 and 40, respectively. R² = 0.878 and 0.957 for pEC₅₀ and I_max, respectively. Akaike’s information criterion was 46.7 and -93.5 for pEC₅₀ and I_max, respectively. CI, confidence interval. Coefficients for Dβ1 subunit could not be calculated because it was essential for the functional expression of the nAChRs compared.

^aCoefficients were obtained by adjusted models using all covariates.

*Significant by t-test (P < 0.05).

The subunit factors governing variations in neonicotinoid affinity for the various fruit fly nAChRs were derived solely from the multivariate analyses. Therefore, to confirm the results, we performed the chaid (Fig 4A) and lattice (Fig 4B) analyses of the affinity of the neonicotinoids. In the chaid analysis, the Dα1 and Dα2 subunits were negative determinants, whereas the Dα3 subunit was a positive determinant of the affinity, Dα2 being a higher contributor than Dα1 and Dα3 (Fig 4A). Mean pEC₅₀ of all the neonicotinoids tested for nAChRs without Dα1 and Dα2, but with Dα3 was highest (7.506), indicating that Dα3 is the most critical determinant of high sensitivity for all the neonicotinoids tested of the D. melanogaster nAChRs. In the lattice analysis (Fig 4B), Dα2 subunit was a significant negative factor for the affinity (Table I in S1 Text).

Fig 4. Chaid and lattice analyses of subunit factors determining the affinity of neonicoticotinoid for the D. melanogaster nAChRs.

(A) Chaid analysis of subunit factors determining the affinity of neonicotinoids for the nAChRs. pEC₅₀ and data numbers are shown in each bracket. Abbreviations: with, w; without, wo. (B) Lattice analysis of subunit factors determining the neonicotinoid affinity for the nAChRs.

Based on these results, we knocked down Dα1, Dα2, and Dα3 subunit genes by using a pan-neuronal Gal4 driver (elav-Gal4) and quantified the expression of genes encoding all nAChR subunits in Control (elav-Gal4>w¹¹¹⁸) and RNAi animals in both developmental (white prepupae) and adult stages of D. melanogaster (Fig 5A). At the same time, these RNAi animals were used to examine toxicity of imidacloprid, thiacloprid, and clothianidin (Fig 5B). We here focused on Dα1, Dα2, and Dα3 subunits, because these three subunits play critical roles in determining the affinity of the neonicotinoids (Tables 1 and 2). Knockdown of Dα1, Dα2, and Dα3 differentially affected the other subunit gene expression level, depending on stage and sex (Fig 5A). During development, knockdown of each of Dα1, Dα2, and Dα3 hardly affected other subunit gene expression except for Dα2 RNAi, which significantly reduced Dα1 expression. By contrast, knockdown of Dα1 enhanced Dβ3 expression in both males and females and Dα7 expression in adult females. Knockdown of Dα2 reduced Dα1, Dα6, and Dα7 expression in both males and females, and Dβ1 expression in adult males. Furthermore, knockdown of Dα2 enhanced Dβ3 expression in adult males. On the other hand, knockdown of Dα3 reduced Dα1 expression and enhanced Dα2 expression in both males and females while enhancing Dβ3 expression in adult females (Fig 5A). These findings indicate that the subunit compensation occurs more frequently in adults than during development. Such subunit compensation can also enhance the inhibitory effect on the climbing behaviour, thereby inducing hypersensitisation to neonicotinoids.

Fig 5. Effects of α subunit gene RNAi on neonicotinoid toxicity in the fruit flies.

(A) Relative expression of genes encoding all nAChR subunits by pan-neuronal knockdown of Dα1, Dα2, and Dα3 in white prepupae and adults of D. melanogaster. elav-Gal4>UAS-dicer2 was used to induce RNAi in all neurons. Each bar indicates the mean ± standard deviation (n = 3). Asterisk indicates that the gene expression level changes as compared with the control (elav-Gal4>w¹¹¹⁸) (t-test, P < 0.05). (B) Toxicity of neonicotinoids by pan-neuronal knockdown of Dα1, Dα2, and Dα3 in larvae and adults of D. melanogaster. Each bar represents the mean ± standard deviation (pupariation rate assay, n = 3; adult climbing assay, control n = 20, RNAi n = 10). Asterisk indicates that the toxicity level changes compared with the control (one-way ANOVA, Bonferroni test, P < 0.05).

Several studies investigated the effects of knocking out nAChR subunit genes in fruit flies on the toxicity of neonicotinoids to larvae or adults and showed that in almost all cases, such mutations resulted in reduced sensitivity to neonicotinoids [38–40].

Interestingly, Perry et al. showed that Dα2 knockout enhanced nitenpyram sensitivity in larvae [38]. Also, Chen et al. showed that Dα1/Dβ2 double knockouts reduced imidacloprid resistance level compared to that observed in a Dα1 or a Dα2 single gene knockout in D. melanogaster [41]. Still, the mechanism of these findings is not known. Liu et al. showed a higher Dα3 subunit contribution to the interactions with clothianidin than for other subunits tested in terms of lethal activity [40]. However, no such Dα3 preference for clothianidin was evident in our study. These findings may be attributable, at least in part, to the differences in the nAChR subunits involved in toxicity. Here we show that RNAi of Dα1 and Dα2 reduced toxicity of imidacloprid, thiacloprid, and clothianidin in larvae (Fig 5B), which is attributable to increased non-target/target nAChR ratio. However, as predicted by the multivariate analyses, RNAi of Dα2 led to hypersensitivity to imidacloprid and thiacloprid in adult males and females and to clothianidin in adult males (Fig 5B), which counsels caution in believing that reduction of drug sensitivity generally happens in response to suppressing the primary target proteins. A direct interpretation of such an observation is that Dα2 subunit is the negative factor reducing the affinity of neonicotinoids (Fig 4 and Table 2), hence reduced Dα2 gene expression results in enhanced neonicotinoid sensitivity. The qRT-PCR data (Fig 5A) revealed that in response to RNAi of Dα2, expression of genes encoding Dα5, Dα6, and Dα7 subunits, of which the Dα5 and Dα6 subunits form low imidacloprid-sensitive nAChRs [42], was reduced, offering another explanation for the enhanced toxicity of neonicotinoids. Reduced toxicity by knockdown of Dα2 and concomitant reduced Dβ1 expression was also observed in adult males (Fig 5A), which can reduce numbers of nAChRs with neonicotinoid sensitivity since the Dβ1 subunit is essential for functional expression (Fig 1 and Table 1). As such, subunit compensation in response to the knockdown of Dα1, Dα2, and Dα3 varies with developmental stages and sexes as well as the primary target of RNAi, resulting in diverse neonicotinoid actions.

These data indicate that Dα1, Dα2, and Dα3 subunits all underpin the interactions with neonicotinoids in the fruit fly. Nevertheless, a contribution of the other subunits to the neonicotinoid actions should not be underestimated because subunit compensation, which can cause replacement of subunits in nAChRs, occurs in response to RNAi of each subunit gene (Fig 5A). The t-SNE representations of the single cell gene expression data indicate co-expression of the Dα1, Dα2, Dα3, Dβ1, and Dβ2 subunits with the Dα4 subunit (Fig D in S1 Text). The Dα1, Dα2, Dα3, Dβ1, and Dβ2 subunits also co-exist with Dβ3 subunit, although such cases are limited [43]. Hence, for the first time we evaluated the effects of replacing one of the Dα1, Dα2, and Dα3 subunits in the Dα1/Dα2/Dα3/Dβ1/Dβ2 nAChRs by the Dα4 or Dβ3 subunit on the agonist activity of imidacloprid, thiacloprid, and clothianidin as well as ACh (Fig 6, Fig E in S1 Text and Table 1). Except for the substitution of the Dα3 subunit, such switching increased affinity of neonicotinoids (Fig 6, Table 1, and Table J in S1 Text). For example, pEC₅₀ values of imidacloprid, thiacloprid, and clothianidin for the Dα1/Dα2/Dα3/β1/Dβ2 nAChR increased from 6.47, 7.18, and 6.59 to 7.40, 7.99, and 7.11, respectively by switching the Dα1 subunit to the Dα4 subunit. On the other hand, switching the Dα1 subunit of the Dα1/Dα2/Dα3/Dβ1/Dβ2 nAChR to the Dα4 or Dβ3 subunit reduced the efficacy of imidacloprid (Dα1/Dα2/Dα3/Dβ1/Dβ2 nAChR, 0.142, Dα2/Dα3/Dα4/Dβ1/Dβ2 nAChR, 0.027; Dα2/Dα3/Dβ1/Dβ2/Dβ3 nAChR, 0.021) and thiacloprid (Dα1/Dα2/Dα3/Dβ1/Dβ2 nAChR, 0.059, Dα2/Dα3/Dα4/Dβ1/Dβ2 nAChR, 0.018; Dα2/Dα3/Dβ1/Dβ2/Dβ3 nAChR, 0.010). Similarly, replacing the Dα2 subunit of the Dα1/Dα2/Dα3/Dβ1/Dβ2 nAChR with the Dα4 subunit reduced the efficacy of imidacloprid (0.142 to 0.101) and clothianidin (0.812 to 0.425), impairing fruit fly motility in accordance with the enhanced neonicotinoid toxicity by RNAi of Dα2 (Fig 5B). By contrast, switching the Dα3 subunit to the Dα4 subunit had a minimal impact on the efficacy of imidacloprid and thiacloprid, while increasing that of clothianidin (0.812 to 1.001). Furthermore, switching the Dα3 subunit to the Dβ3 subunit reduced the efficacy of imidacloprid (0.142 to 0.080) and thiacloprid (0.059 to 0.023) while increasing that of clothianidin (0.812 to 0.974, Fig 6, Table 1 and Table J in S1 Text), explaining why targeting Dα3 had less impact on neonicotinoid toxicity than targeting Dα1 and Dα2. These results suggest that both Dα4 and Dβ3 subunits can form heteromeric nAChRs with either of Dα2/Dα3/Dβ1/Dβ2, Dα1/Dα3/Dβ1/Dβ2, and Dα1/Dα2/Dβ1/Dβ2 combinations and show unique pharmacological features in neonicotinoid actions. Also, it is conceivable that the nAChRs containing the Dα4 or Dβ3 subunit also contribute to the change of neonicotinoid toxicity in response to RNAi of Dα1, Dα2, and Dα3.

Fig 6. Effects of Dα1, Dα2, and Dα3 subunit substitution by Dα4 or Dβ3 subunit on agonist actions of ligands on Dα1/Dα2/Dα3/Dβ1/Dβ2 D. melanogaster nAChRs.

Dα1 (A), Dα2 (B), and Dα3 (C) subunits were substituted by Dα4 or Dβ3 subunits and we compared the concentration-response curves for ACh, imidacloprid, thiacloprid, and clothianidin on the Dα1/Dα2/Dα3/Dβ1/Dβ2 nAChRs with those on the nAChRs resulting from the subunit switching. Except for the substitution of the Dα3 subunit, the subunit switching resulted in enhanced affinity of neonicotinoids. In the Dα1 and Dα2 subunit switching, Dα4 subunit addition resulted in reduced efficacy and enhanced affinity of imidacloprid and thiacloprid, whereas Dβ3 subunit addition increased the efficacy of thiacloprid. For the actions of clothianidin, Dα2 subunit switching by Dα4 or Dβ3 subunit reduced efficacy, while Dα3 subunit switching enhanced it. Each data point represents the mean ± standard error (n = 5).

In conclusion, by studying 18 subunit combinations of subunits Dα1, Dα2, Dα3, Dα4, Dβ1, Dβ2, and Dβ3, we have found that imidacloprid, thiacloprid and clothianidin can interact with a broad range of D. melanogaster nAChRs formed not only by the Dα1, Dα2, Dα3, Dβ1, and Dβ2 subunits, but also by the Dα4 and Dβ3 subunits, which has not been described to the best of our knowledge. Although co-expression of these subunits does not necessary prove that they co-assemble to form functional nAChRs in neurons, it is clear that the three neonicotinoids exhibited diverse agonist actions on the 18 nAChRs tested, the outcome depending on both the compound as well as subunit composition. Notably, the Dα1, Dα2, Dα3, Dβ1, and Dβ2 subunits co-localise in organs underlying mating and egg laying, predicting that modulation of the nAChRs consisting of these subunits will affect the number of offspring. In future, it will be of considerable interest to test this hypothesis. If such actions are confirmed, not only for the fruit flies, but also for other insect species such as pollinators and disease vectors, this will counsel further caution in identifying target receptor subtypes simply in terms of reduced neonicotinoid sensitivity resulting only from gene disruption or suppression experiments.

Methods

Ethics statement

Oocytes at stage V or VI of development were removed under anesthetic (0.3 g L^-1 benzocaine) from adult female X. laevis according to the U.K. Animals (Scientific Procedures) Act, 1986. Care was always taken to minimise the number of animals used in experiments.

ACh and neonicotinoids

ACh (#A6625) was purchased from MilliporeSigma (USA). The neonicotinoids (imidacloprid, #099–03771; thiacloprid, #205–19081; clothianidin, #034–22581) were purchased from FUJIFILM Wako Pure Chemical (Japan). These reagents were used without further purification.

Flies

All flies were raised at 25°C under 12 h/12 h light/dark cycle. The animals were reared on standard fly food containing 5.5 g agar, 100 g glucose, 40 g dry yeast, 90 g cornflour, 3 mL propionic acid, and 3.5 mL 10% butyl p-hydroxybenzoate (Nacalai Tesque, Japan, #06327–15) in 70% ethanol per liter. The control strain was w¹¹¹⁸, and transgenic flies are as follows: UAS-nAChRα1 RNAi (#28688) was obtained from the Bloomington Drosophila Stock Center (BDSC); UAS-nAChRα2 RNAi (UAS-Dα2 RNAi, #10760), UAS-nAChRα3 RNAi (UAS-Dα3 RNAi, #101806); UAS-dicer2 (#60009) were obtained from the Vienna Drosophila Resource Center (VDRC); and UAS-mCD8::GFP (#108068) [44] was obtained from Kyoto Stock Center. elav-Gal4 (3A3) was obtained from Michael B. O’Connor. nAChRα3-knock-in 2A-Gal4 (Dα3-knock-in T2A-Gal4) was generated by the CRISPR/Cas9 system as described in detail below.

Generation of Dα3-knock-in T2A-Gal4 strain

The Gal4 knock-in D. melanogaster flies were generated by CRISPR/Cas9-mediated homologous recombination. A targeting vector was designed such that the T2A-Gal4 [45] is inserted in frame with the last intracellular region of the protein. The targeting vector and a gRNA expression vector that cuts near the target site were co-injected into fertilised eggs maternally expressing Cas9 protein. The flanking sequences of the insertion are: 5´-GAAAGAGGACTGGAAGTACGTGGCCATG/GTGCTCGATCGCCTGTTCCTGTGGATCTTCACAATAGC-3´ (The site of integration is indicated by a slash, The 20-bp gene-specific sequence of the gRNA is underlined.)

Immunostaining

D. melanogaster male reproductive systems were dissected in Grace’s Insect Medium, supplemented (Thermo Fisher Scientific, USA, #11605094) and fixed in 4% paraformaldehyde in Grace’s medium for 30–60 min at room temperature (RT). The fixed samples were washed three times in phosphate-buffered saline (PBS) supplemented with 0.1% Triton X-100 (Nacalai Tesque, #12967–45). After washing, samples were blocked in the blocking solution (PBS with 0.1% Triton X-100 and 2% bovine serum albumin (MilliporeSigma, #A9647) for 1 h at RT then incubated with a primary antibody in the blocking solution at 4°C overnight. The primary antibodies used in this study were mouse anti-GFP monoclonal antibody (clone GFP-20; MilliporeSigma G6539; 1:1000) and rabbit anti-Tdc2 antibody (Abcam ab128225; 1:1000) [46]. Fluorophore (Alexa Fluor 488, or 546)-conjugated secondary antibodies (Thermo Fisher Scientific, #A11001, #A32732) were used at a 1:200 dilution and incubated for 2 h at RT in the blocking solution. After washing, all samples were mounted in FluorSave reagent (MilliporeSigma, #345789). Samples were visualised on an LSM 700 confocal microscope (Zeiss, Germany). Images were processed using the ImageJ package [47].

cDNAs and cRNAs

cDNAs of the nAChR subunits and co-factors were cloned into pcDNA3.1 (+) vector (Thermo Fisher Scientific). The accession numbers of the nAChR subunits and cofactors are as follows: Dα1 (NP_524481), Dα2 (NP_524482), Dα3 (NP_525079), Dα4 (CAB77445), Dβ1 (NP_523927), Dβ2 (NP_524483), Dβ3 (NP_525098), DmRIC-3 (CAP16647), DmUNC-50 (NP_649813), and DmTMX3 (NP_648847). cRNAs were prepared using the mMESSAGE mMACHINE T7 ULTRA Transcription Kit (Thermo Fisher Scientific, #AM1345) according to the manual with the cDNA template which was cut with appropriate restriction enzymes at the 3’ end of the cDNA.

cRNA expression in X. laevis oocytes

We minimised the use of X. laevis according to the UK Animals (Scientific Procedures) Act, 1986. Female X. laevis were anaesthetised with benzocaine (Nacalai Tesque, #14804–92) prior to oocyte excision. Oocytes were defolliculated after collagenase treatment in Ca²⁺-free standard oocyte saline (Ca²⁺-free SOS). cRNAs of the nAChR subunits and co-factors mixed at a concentration of 0.1 mg/mL was injected into oocytes at a volume of 50 nL. Then the oocytes were incubated in the incubation medium (SOS supplemented with sodium pyruvate (Nacalai Tesque, #13058–12), penicillin-streptomycin (MilliporeSigma, #P4333), gentamycin (Nacalai Tesque, #16672–04), and 4% horse serum (Thermo Fisher Scientific, #26050–070) for 3−4 days prior to electrophysiology [30].

Voltage-clamp electrophysiology

Each defolliculated X. laevis oocyte was secured in a Perspex recording chamber and perfused with the standard oocyte saline (SOS) containing 0.5 μM atropine (SOSA) at a flow rate of 7−10 mL/min [48]. Two glass electrodes filled with 2 M KCl were impaled into each oocyte and the membrane potential was clamped at -100 mV. ACh and α-BTX (Alamone Labs, Israel, #B-100) were dissolved directly in SOSA, while test solutions of the neonicotinoids were diluted to the final concentration from DMSO stock solutions. DMSO at 1% (v/v) or lower had no effect on the responses to neonicotinoids or other ligands tested. ACh and neonicotinoids were applied for 5 s successively at 3 min intervals. α-BTX was tested as previously described in the literature [33]. The peak amplitude of the response was measured and analysed by pCLAMP (Molecular Devices, USA). The agonist response data were normalised to the maximum response to ACh at concentrations at which the response amplitude attained plateau and fitted by non-linear regression using Prism (GraphPad Software, USA), according to the following equation.

$$\mathrm{Y}=\frac{{\mathrm{I}}_{\mathrm{m}\mathrm{a}\mathrm{x}}}{1+{10}^{\left(\log {\mathrm{EC}}_{50}\mathrm{‐}\mathrm{X}\right){\mathrm{n}}_{\mathrm{H}}}}$$

Where Y is the normalised response, X is log[ligand (M)], EC₅₀ is the half maximal concentration (M), I_max is the maximum normalised response and n_H is the Hill coefficient.

Total RNA extraction and quantitative reverse transcription (qRT)-PCR

Animals were collected in 1.5 ml tubes and immediately flash-frozen in liquid nitrogen. Total RNA from white prepupa (0 hour after puparium formation) or adults (3 days after eclosion) was extracted using RNAiso Plus (Takara Bio, Japan, #9109) according to the manufacturer’s instructions. cDNA was generated from purified total RNA using ReverTra Ace qPCR RT Master Mix with gDNA Remover (TOYOBO, Japan, #FSQ-301). qRT-PCR was performed on the Thermal Cycler Dice TP800 system (Takara Bio) using THUNDERBIRD Next SYBR qPCR Mix (Takara Bio, #QPX-201). For absolute quantification of mRNAs, serial dilutions of plasmids containing coding sequences of the target genes or rp49 were used for standards. After the molar amounts were calculated, transcript levels of the target mRNA were normalised to rp49 levels in the same samples. The primers used are listed in Table K in S1 Text. The primers to detect rp49 levels are as previously reported [49].

Pupariation rate assay

Eggs were laid on grape juice plates with yeast paste at 25°C for 6 h. After 24 h, early first instar larvae just after hatching (20 larvae/vial) were transferred into a mini-vial (Sarstedt, Germany, #58.487) with 2.0 g of neonicotinoid feeding assay food: 50 mL eq. of blue food powder (Formula 4–24 Instant Drosophila Medium, Carolina, USA, #173210), 50 mL eq. of yeast powder (Brewer’s yeast, MP Biomedicals, USA, #903312), and 100 mL dH₂O containing 0.1% dimethyl sulfoxide (DMSO; Nacalai Tesque, #13407–45) (for control) or one of the neonicotinoids (imidacloprid, thiacloprid, and clothianidin) in 0.1% DMSO. After a week of incubation at 25°C, pupal numbers were scored in each vial.

Adult climbing assay

Adult D. melanogaster flies were collected within a day following eclosion and placed in normal fly food (less than 30 flies/vial). Flies were transferred daily to new fly food. 2–5 days after eclosion, flies were briefly anesthetised with CO₂, and the sexes were separated and sorted into fly vials containing 1.0% agar food for starvation (10 flies/vial). After 16 h starvation, flies were transferred to vials containing neonicotinoid-containing food without anesthesia and cultured for 6 h (10 flies/vial). Neonicotinoid-containing foods were prepared by mixing 10 μL of diluted neonicotinoids dissolved in DMSO with 990 μL of a solution containing 1% agar and 5% sucrose for each vial. After 6 h cultured in vials containing neonicotinoid-containing food, flies were gently tapped down to the surface of the food, and flies that climbed within 20 s after tapping were recorded by a video camera (GZ-F270-W, JVCKENWOOD, Japan). The maximum climbing heights of the flies within 20 s after tapping were measured using ImageJ1.53v (National Institute of Health, USA). Since the height from the surface of the food to the vial top is 8 cm, the maximum climbing height is 8 cm.

Reproducibility of data

At least two authors participated independently in measuring data to confirm reproducibility of the results. For electrophysiology, five oocytes from at least two frogs were used to determine the agonist activity of each ligand at each concentration.

Statistical analyses

Prism software was employed for the statistical analyses. The peak current amplitude of the agonist actions of ACh at 100 μM was compared between the nAChRs by Kruskal-Wallis test. One-way ANOVA was used to analyse the differences of the ligand agonist activity in terms of pEC₅₀ on the various nAChRs expressed in X. laevis as well as data obtained with D. melanogaster larvae and adults.

Multiple variate analysis

The multiple variate analysis was conducted with python to examine if D. melanogaster nAChR subunits and compounds contribute significantly to the agonist activity in terms of pEC₅₀ and I_max. We used a dataset including 48 samples (Table H in S1 Text). Objective variables are pEC₅₀ and Imax and explanatory variables are subunits (Dα1, Dα2, Dα3, and Dβ2) and compounds (ACh, imidacloprid, thiacloprid, and clothianidin). Data for ACh were used as references when calculating the subunit and compound factors governing the variations in the agonist activity indices.

Chaid analysis

Chaid analysis was conducted with python to examine if the nAChR subunits contribute significantly to the agonist activity in pEC₅₀. Parameter max depth was set as 4. Objective variable is pEC₅₀ and explanatory variables are subunits (Dα1, Dα2, Dα3, and Dβ2) and compounds (ACh, imidacloprid, thiacloprid, and clothianidin).

Lattice visualization

The lattice visualization was used to observe the positive contribution to pEC₅₀ of adding each subunit (Dα1, Dα2, Dα3, Dβ1, and Dβ2). The presence or absence of subunits forms the powerset with lattice structure with respect to the inclusion order. The data are grouped by differences between two sets of subunits and denoted by "+<subunit name>" on each edge. The color bar indicates ΔpEC₅₀, the value obtained by subtracting pEC₅₀ for smaller nAChR subunit set from that for larger nAChR subunit set. The significance of the ΔpEC₅₀ values was analysed by the 95% confidence interval.

Data used in Figs 1−3, 5, and 6 are available from Dryad [50].

Dryad DOI

https://doi.org/10.5061/dryad.qz612jmk5 [50]

Supporting information

S1 Text. Supporting information.

Fig A. Inward current response of oocytes expressing D. melanogaster nAChRs to several concentrations of ACh. Horizontal bar shows application of ACh. Fig B. Effects of α-bungarotoxin (α-BTX) on the response to 100 μM ACh of X. laevis oocytes expressing D. melanogaster nAChRs. (a) Inward currents induced in oocytes expressing D. melanogaster nAChRs in response to 100 μM ACh in the absence and presence of α-BTX. Horizontal lines indicate application of ACh. (b) Bar graph representations of the current amplitude of responses to 100 μM ACh of the nAChR expressing oocytes exposed to 10 nM or 100 nM α-BTX. Error bars are standard error of the mean (n = 5). Fig C. Inward current responses to ACh and neonicotinoids (imidacloprid, thiacloprid, and clothianidin) of oocytes expressing D. melanogaster nAChRs. Horizontal lines indicate application of neonicotinoids. Fig D. t-SNE representations of Dα1, Dα2, Dα3, Dα4, Dβ1, and Dβ2 gene expressions in the adult brain and ventral nerve cord of D. melganogaster. The figure was illustrated by SCope (https://scope.aertslab.org/#/86757313-d473-4f5f-b045-fc035d99451a/*/welcome) using single cell RNA-sequencing data [43]. These six nAChR subunit genes are co-expressed in single cells (See white dots.). Fig E. Inward current responses to ACh and neonicotinoids (imidacloprid, thiacloprid, and clothianidin) of oocytes expressing D. melanogaster Dα2/Dα3/Dα4/Dβ1/Dβ2, Dα2/Dα3/Dβ1/Dβ2/Dβ3, Dα1/Dα3/Dα4/Dβ1/Dβ2, Dα1/Dα3/Dβ1/Dβ2/Dβ3, Dα1/Dα2/Dα4/Dβ1/Dβ2, and Dα1/Dα2/Dβ1/Dβ2/Dβ3 nAChRs. Horizontal lines indicate application of neonicotinoids. Table A. One-way ANOVA of the pEC₅₀ values of ACh for D. melanogaster nAChRs. Table B. One-way ANOVA of the pEC₅₀ values of imidacloprid for D. melanogaster nAChRs. Table C. One-way ANOVA of the I_max values of imidacloprid for D. melanogaster nAChRs. Table D. One-way ANOVA of the pEC₅₀ values of thiacloprid for D. melanogaster nAChRs. Table E. One-way ANOVA of the I_max values of thiacloprid for D. melanogaster nAChRs. Table F. One-way ANOVA of the pEC₅₀ values of clothianidin for D. melanogaster nAChRs. Table G. One-way ANOVA of the I_max values of clothianidin for D. melanogaster nAChRs. Table H. Data set for multivariate analyses. Table I. Mean and 95% confidence intervals of ΔpEC₅₀ values obtained by lattice analysis. Table J. One-way ANOVA of the pEC₅₀ and I_max values of ligands for D. melanogaster nAChRs containing Dα4 or Dβ3 subunit as compared with the values for Dα1/Dα2/Dα3/Dβ1/Dβ2 nAChR. Table K. Primers for qRT-PCR

(DOCX)

Data Availability

The data used for analysis in this study are available from the Dryad (https://doi.org/10.5061/dryad.qz612jmk5).

Funding Statement

This study was supported by KAKENHI (Grant-in-Aid for Scientific Research) from the Japan Society for the Promotion of Science (grant number 21H04718 (KM and RN), 22H02350 (MI),26250001 (HT), and 17H01378 (HT)); and the Cooperative Research Project Program of Life Science Center for Survival Dynamics, Tsukuba Advanced Research Alliance (TARA Center), University of Tsukuba, Japan (grant number 202113 and 202217 (KM)).The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.