Minor structural changes in nicotinoid insecticides confer differential subtype selectivity for mammalian nicotinic acetylcholine receptors

Abstract

1. The major nitroimine insecticide imidacloprid (IMI) and the nicotinic analgesics epibatidine and ABT-594 contain the 6-chloro-3-pyridinyl moiety important for high activity and/or selectivity. ABT-594 has considerable nicotinic acetylcholine receptor (AChR) subtype specificity which might carry over to the chloropyridinyl insecticides. This study considers nine IMI analogues for selectivity in binding to immuno-isolated α₁, α₃ and α₇ containing nicotinic AChRs and to purported α₄β₂ nicotinic AChRs.

2. α₁- and α₃-Containing nicotinic AChRs (both immuno-isolated by mAb 35, from Torpedo and human neuroblastoma SH-SY5Y cells, respectively) are between two and four times more sensitive to DN-IMI than to (−)-nicotine.

3. With immuno-isolated α₃ nicotinic AChRs, the tetrahydropyrimidine analogues of IMI with imine or nitromethylene substituents are 3–4 fold less active than (−)-nicotine. The structure-activity profile with α₃ nicotinic AChRs from binding assays is faithfully reproduced in agonist potency as induction of ⁸⁶rubidium ion efflux in intact cells.

4. α₇-Containing nicotinic AChRs of SH-SY5Y cells (immuno-isolated by mAb 306) and rat brain membranes show maximum sensitivity to the tetrahydropyrimidine analogue of IMI with the nitromethylene substituent.

5. The purported α₄β₂ nicotinic AChRs [mouse (Chao & Casida, 1997) and rat brain] are similar in sensitivity to DN-IMI, the tetrahydropyrimidine nitromethylene and nicotine.

6. The commercial insecticides (IMI, acetamiprid and nitenpyram) have low to moderate potency at the α₃ and purported α₄β₂ nicotinic AChRs and are essentially inactive at α₁ and α₇ nicotinic AChRs.

7. In conclusion, the toxicity of the analogues and metabolites of nicotinoid insecticides in mammals may involve action at multiple receptor subtypes with selectivity conferred by minor structural changes.

Introduction

Nicotinic acetylcholine receptors (AChRs) consist of diverse subtypes formed from five homologous subunits in combinations from nine α, four β, γ, δ and ε subunits. Advances in knowledge of nicotinic AChR structure and function provide a means to establish the specific receptor subtypes conferring selectivity for nicotinic drugs. The 6-chloro-3-pyridinyl moiety is present in some of the most potent and/or selective nicotinic agonists as to subtype specificity, e.g. the analgesics epibatidine and ABT-594 (Badio & Daly, 1994; Holladay et al., 1997; Bannon et al., 1998), and it is also important in a new class of synthetic nicotinoid insecticides (Shiokawa et al., 1995) (Figure 1). Imidacloprid (IMI) is the best known example of these highly effective new insecticides and others are acetamiprid (AAP) and nitenpyram (NTP) (Figure 2). These nicotinoid insecticides and their analogues might also be selective in their action on nicotinic AChR subtypes.

Figure 1.

Structural similarity between 6-chloro-3-pyridinyl-containing nicotinic ligands. Imidacloprid (IMI) is an insecticide and desnitro-imidacloprid (DN-IMI) is one of its metabolites. Epibatidine (from the skin of an Ecuadorian frog) and ABT-594 (from structure-activity optimization studies) are analgesics of outstanding potency and/or nicotinic AChR subtype specificity.

Figure 2.

Nicotinoid insecticides (1, 8 and 9) and a metabolite (2) and analogues (3–7). Eight of the compounds contain the 6-chloro-3-pyridinyl substituent with a methylene bridge to an imidazolidine (IMI, DN-IMI, CH-IMI and Cl-TMNI), tetrahydropyrimidine (DN-THP and THPCH-IMI), thiazolidine (SCH-IMI) or acyclic replacement for the heterocyclic ring (AAP and NTP). They include nitroimines (N-NO₂), imines (NH), nitromethylenes (CH-NO₂) and a cyanoimine (N-CN). A chlorothiazole moiety replaces the chloropyridine in Cl-TMNI. All of these chemicals, except DN-IMI and DN-THP, are highly potent insecticides.

α₁-Containing nicotinic AChRs expressed in skeletal muscle and Torpedo electric organ are α₁γ(or ε in adult)α₁δβ₁ heteromers; they are the best understood nicotinic AChRs as to the ligand binding site environment (Karlin & Akabas, 1995; Arias, 1997). Neuronal nicotinic AChR subtypes in brain and ganglia are assembled in combinations of α_2–9 and β_2–4 subunits and are pharmacologically classified into two main groups based on sensitivity to α-bungarotoxin (α-BGT) (Sargent, 1993; Lindstrom, 1997). The α-BGT-insensitive subtypes are formed from combinations of α₂, α₃, α₄ and α₆ with β₂ or β₄ subunits (sometimes with α₅ or β₃). The most prominent subtype of this group is α₄β₂ which represents >90% of high affinity tritiated agonist binding sites in brain (Whiting & Lindstrom, 1986; Flores et al., 1992). α₃-Containing nicotinic AChRs (α₃β₄α₅ and α₃β₂α₅) are expressed in peripheral ganglia and limited regions of the brain (Conroy & Berg, 1995; Lindstrom, 1997). α-BGT-sensitive neuronal nicotinic AChR subtypes have α₇, α₈ and α₉ subunits (Lindstrom, 1997). The abundance of α₇-containing nicotinic AChRs in brain is comparable to that of the α₄β₂ subtype (Clarke et al., 1985; Whiting & Lindstrom, 1988; Lindstrom, 1997). The α₇ nicotinic AChRs are also coexpressed with multiple α₃-containing receptors (α₃β₄α₅ and α₃β₂α₅) in ganglia as well as in human neuroblastoma cells such as SH-SY5Y (Lukas et al., 1993; Peng et al., 1994; Lindstrom, 1997). The α₈ subunit is found only in chickens and the α₉ in limited regions of the rat nervous system (Schoepfer et al., 1990; Keyser et al., 1993; Elgoyhen et al., 1994). The native α₇-containing nicotinic AChRs are considered to be assembled either as a homomer (Couturier et al., 1990; Chen & Patrick, 1997; Lindstrom, 1997) or as heteromers with unknown subunit(s) (Whiting & Lindstrom, 1987; Gotti et al., 1991; Anand et al., 1993b).

The structure and function of insect nicotinic AChRs have been investigated with biochemical, molecular biological and immunohistochemical approaches but are poorly understood relative to those of animals. Although several candidate genes encoding the α and non-α subunits are identified from fruit flies (Drosophila melanogaster) and migratory locusts (Locusta migratoria), their functional coexpression has not been successful in any combination, implying the involvement of unidentified subunit(s) in assembling the native insect receptors (Gundelfinger & Hess, 1992; Tomizawa et al., 1996; 1999; Tomizawa & Casida, 1997; Hermsen et al., 1998). Interestingly, functional ion channel properties are clearly observed when either of two Drosophila α type subunits is coexpressed with chick β₂ subunit and the two reconstituted Drosophila α/chick β receptors display different sensitivities to α-BGT (Bertrand et al., 1994). It is proposed for the cockroach (Periplaneta americana) that the α-BGT-sensitive and -insensitive nicotinic AChRs are expressed in the dorsal unpaired median neurons and that both subtypes are affected by IMI, based on electrophysiology studies (Lapied et al., 1990; Buckingham et al., 1997).

IMI and its desnitro metabolite (DN-IMI) (Figure 1) differ greatly in binding site specificity: IMI is highly potent at insect but not mammalian nicotinic AChRs (Liu & Casida, 1993; Zwart et al., 1994; Yamamoto et al., 1998) whereas DN-IMI is much more active in mammals than insects (Liu et al., 1993; Chao & Casida, 1997; Nauen et al., 1998). On this basis, minor structural variations in nicotinoid insecticides may also alter the subtype specificity for mammalian nicotinic AChRs. The objective of this study is to determine the contribution of α₁-, α₃-, α₇- and α₄β₂-containing nicotinic AChRs to the specificity of nicotinoid insecticide action.

Methods

Chemicals

Structures and abbreviations for the nicotinoids studied are given in Figure 2. They were available from our previous studies (Liu et al., 1995; Tomizawa et al., 1996; Chao & Casida, 1997) except for DN-THP which was synthesized by a procedure analogous to that used for DN-IMI (Latli et al., 1996). The purity of these compounds was >95% and they were stored in amber bottles (under nitrogen atmosphere if needed) at room temperature, and the test sample was freshly prepared for each experiment.

Sources for other chemicals were as follows: (3-[¹²⁵I]iodotyrosyl)α-BGT ([¹²⁵I]α-BGT, >277 Ci mmol⁻¹) and ⁸⁶rubidium chloride (1.7 mCi mg⁻¹ Rb, ⁸⁶Rb⁺) from Amersham Life Science (Arlington Heights, IL, U.S.A.); L-[N-methyl-³H]-nicotine ([³H]-nicotine, 81.5 Ci mmol⁻¹) from NEN Life Science Products (Boston, MA, U.S.A.); α-BGT, (−)-nicotine hydrogen tartrate and poly-L-lysine hydrobromide from Sigma (St. Louis, MO, U.S.A.); Dulbecco's modified Eagle's medium (DMEM), foetal bovine serum (FBS) and penicillin-streptomycin from Gibco Life Technologies (Grand Island, NY, U.S.A.). The nicotinic AChR monoclonal antibodies (mAb) used were mAb 35 against α₁, α₃ and α₅ subunits (Conroy et al., 1992) and mAb 306 against the α₇ subunit (Schoepfer et al., 1990) from Research Biochemicals International (Natick, MA, U.S.A.).

Torpedo receptor preparation

Torpedo electric organ (Biofish Associates, Georgetown, MA, U.S.A.) was homogenized in four volumes of 50 mM sodium phosphate buffer (pH 7.5) containing (in mM): NaCl 1000, EDTA 5, EGTA 5, phenylmethanesulphonyl fluoride (PMSF) 2, benzamidine 5 and iodoacetamide 5 at 4°C using a Polytron for three 30 s periods with 60 s intervals in between. The homogenate was filtered through four layers of cheesecloth and the filtrate was centrifuged at 40,000×g for 30 min at 4°C. The pellet was suspended in lysis buffer (in mM): sodium phosphate (pH 7.5) containing 2% Triton X-100 50, NaCl 50, EDTA 5, EGTA 5, PMSF 2, benzamidine 5 and iodoacetamide 5 (same volume as the homogenate) and the suspension was solubilized by rotation on a rocking platform for 60 min at 4°C. Insoluble material was removed by centrifugation at 40,000×g for 30 min at 4°C.

Human neuroblastoma cell receptor preparation

Cultures of SH-SY5Y cells (Department of Molecular and Cell Biology, University of California, Berkeley) were maintained in DMEM supplemented with 10% FBS, 50 u ml⁻¹ penicillin and 50 μg ml⁻¹ streptomycin at 37°C in 5% CO₂/95% air atmosphere with a medium change every 2–3 days. The cells were harvested with a cell lifter in phosphate-buffered saline (PBS, NaCl 100 mM, sodium phosphate buffer 10 mM, pH 7.5). The harvested cells were disrupted by brief vortexing in five volumes of lysis buffer as above. After 20 min gentle rotation on a rocking platform at 4°C, the sample was centrifuged for 20 min at 4°C in an Eppendorf microcentrifuge and the supernatant was recovered.

Radioligand binding

Supernatants from Torpedo electric organ or SH-SY5Y cell preparations were used for immuno-isolation of receptor subtypes (Anand et al., 1993a; Peng et al., 1997) then radioligand binding assay. mAb 35 or 306 (immunoglobulin at 5 mg ml⁻¹) was coupled to Immulon 4HBX Removawells (Dynex Technologies, Chantilly, VA, U.S.A.) by incubating 4–5 μg mAb (per well) in 0.1 ml of 10 mM sodium bicarbonate buffer (pH 8.8) overnight at 4°C. After three washes with 0.2 ml of the bicarbonate buffer, the wells were quenched with 0.2 ml of 3% bovine serum albumin in PBS-Tween 20 buffer (0.05% Tween 20 in PBS) for 4 h at 4°C. The wells were then washed three times with 0.2 ml of the PBS-Tween 20. The receptor preparation (0.1 ml) was added to each mAb-precoated well and incubated overnight at 4°C. The wells were washed three times with 0.2 ml of the PBS-Tween 20 buffer and then treated with various concentrations of test compound for 20 min. Radioligand binding assay was initiated by addition to this medium of [³H]-nicotine (20 nM for α₃ nicotinic AChRs) or [¹²⁵I]-α-BGT (0.2 nM for α₁ nicotinic AChRs or 2 nM for α₇ nicotinic AChRs) and incubation in 0.1 ml final volume for 60 min (for [³H]-nicotine binding) or overnight (for [¹²⁵I]-α-BGT binding) at 25 or 4°C, respectively. The wells were then rinsed three times with 0.2 ml PBS-Tween 20 buffer and the radioactivity remaining was subjected to liquid scintillation counting. Every experiment included (−)-nicotine as a standard at 50 nM (for α₃ nicotinic AChRs with [³H]-nicotine) or 10 μM (for α₇ nicotinic AChRs with [¹²⁵I]-α-BGT). Background binding was determined using wells lacking mAb. For comparison, the binding affinity for (−)-nicotine in [³H]-nicotine binding to immuno-isolated α₃ receptors from SH-SY5Y cells is 0.02 μM (Peng et al., 1997), and for α-BGT and (−)-nicotine in [¹²⁵I]-α-BGT binding to immuno-isolated α₇ receptors from the same cells are 0.00106 and 2.6 μM, respectively (Peng et al., 1994).

Membranes from male rat whole brain were prepared and assayed for 2 nM [¹²⁵I]-α-BGT binding by the method of Marks et al. (1986) and for 5 nM [³H]-nicotine binding as described by Yamamoto et al. (1995). Data for [³H]-nicotine binding to mouse brain membranes are from Chao & Casida (1997).

⁸⁶Rb⁺ efflux assay

Agonist-induced cation flux in SH-SY5Y cells has been attributed to α₃-containing nicotinic AChRs; α-BGT-sensitive receptors do not contribute detectably to the ion flux measured in this assay (Lukas et al., 1993). SH-SY5Y cells were therefore used to assay α₃-containing receptor function by the procedure of Lukas (1989) with minor modification. The cells were seeded in 24-well (18.5 mm diameter) culture plates at a density of 10⁶ cells well⁻¹ following the coating of each well by treatment with poly-L-lysine (30 μg ml⁻¹) then aspiration off. At confluence, the cells attached to the culture plates were loaded with 0.2 μCi of ⁸⁶Rb⁺ in DMEM supplemented with 10% FBS, 50 u ml⁻¹ penicillin and 50 μg ml⁻¹ streptomycin (0.5 ml) and incubated overnight at 37°C and 5% CO₂/95% air atmosphere. The medium containing ⁸⁶Rb⁺ was removed by aspiration, the cells were rinsed twice with 0.5 ml of fresh medium and then exposed to 0.25 ml of medium with or without a test compound for 5 min. (−)-Nicotine (0.1 mM) was tested as a standard in each experiment. After exposure to the test compound, the assay medium was immediately transferred into a vial for Cerenkov counting. For validation in a preliminary experiment, we confirmed that 0.1 mM d-tubocurarine gave 97–100% blockage of ⁸⁶Rb⁺ efflux induced by 0.1 mM (−)-nicotine as reported by Lukas et al. (1993).

Data calculation

IC₅₀ (molar concentration of test compound for 50% inhibition of specific radioligand binding) and EC₅₀ [molar concentration of test compound to induce 50% specific ⁸⁶Rb⁺ efflux relative to 0.1 mM (−)-nicotine] values were determined by iterative nonlinear least-squares regression using the SigmaPlot program (Jandel Scientific Software, San Rafael, CA, U.S.A.).

Toxicity

Male albino Swiss-Webster mice (20–25 g) were treated i.p. with the test compounds dissolved in water or dimethyl sulphoxide with mortality observations at 24 h as described by Chao & Casida (1997). Toxicity data are from Chao & Casida (1997) for seven compounds and the present determination for three compounds. These studies were carried out in accordance with the Guide for Care and Use of Laboratory Animals as adopted and promulgated by the United States National Institutes of Health.

Results

Interaction with α₁ nicotinic AChRs

The Torpedo electric organ was used as the source of α₁ nicotinic AChRs with immuno-isolation by mAb 35 and binding assay with [¹²⁵I]-α-BGT. DN-IMI is 2 fold more potent than nicotine with IC₅₀ values of 13 and 25 μM, respectively (Table 1). THPCH-IMI has low activity (IC₅₀ 120 μM) and all the other nicotinoids are essentially inactive (Table 1).

Table 1.

Structure-activity relationships of nicotinoid insecticide action on muscle and neuronal nicotinic AChR subtypes

Interaction with α₃ nicotinic AChRs

Human neuroblastoma SH-SY5Y cells express multiple α₃β₂β₄α₅ nicotinic AChRs; the α₃ receptors were immuno-isolated with mAb 35 for [³H]-nicotine binding. DN-IMI is the most potent compound (IC₅₀ 0.014 μM) and nicotine is 3 fold less active (Figure 3, Table 1). The potency order for the other nicotinoids is DN-THP and THPCH-IMI (IC₅₀ 0.14–0.18 μM)>CH-IMI (IC₅₀ 2.3 μM)>IMI, AAP, Cl-TMNI, SCH-IMI and NTP (IC₅₀ 14–48 μM) (Table 1).

Figure 3.

Displacement by DN-IMI and (−)-nicotine of [³H]-nicotine binding to α-BGT-insensitive α3 nicotinic AChRs immuno-isolated from human neuroblastoma SH-SY5Y cells. The extracted cell membranes with lysis buffer were reacted with mAb 35-precoated wells overnight at 4°C, and then the immunoprecipitated α₃ nicotinic AChRs were incubated for 60 min at 25°C with 20 nM of [³H]-nicotine in competition with a test compound. Data points represent means of three experiments with s.d.

Induced ⁸⁶Rb⁺ efflux with intact SH-SY5Y cells provides another means to evaluate nicotinoid agonist effect attributable to α₃-containing receptor function. DN-IMI with an EC₅₀ of 2.4 μM is 4–5 fold more potent than nicotine, DN-THP and THPCH-IMI (EC₅₀ 10–13 μM), and DN-IMI and DN-THP display steep efflux induction curves (Table 1, Figure 4). THPCH-IMI induces higher ⁸⁶Rb⁺ efflux than that induced by 0.1 mM (−)-nicotine. The remaining chemicals do not induce an efflux response that reaches the maximum of the nicotine standard.

Figure 4.

Induction by nicotinoids of specific ⁸⁶Rb⁺ efflux from intact human neuroblastoma SH-SY5Y cells. Data for agonist potency are given on a percentage basis relative to 0.1 mM (−)-nicotine as 100% in the same experiment. The intact SH-SY5Y cells preloaded with 0.2 μCi of ⁸⁶Rb⁺ were exposed to medium with or without a test agonist for 5 min. The value for the (−)-nicotine standard ranged from 6000–7000 c.p.m. versus a background without agonist of 1000–1200 c.p.m. Data points represent means of three experiments with s.d.

The inhibitory potency of eight nicotinoids with immuno-isolated α₃ nicotine AChRs of SH-SY5Y cells is highly correlated (r=0.99) with that for agonist-induced ⁸⁶Rb⁺ efflux from intact cells (Figure 5).

Figure 5.

Correlation for nicotinoids of inhibitory potency for [³H]-nicotine binding to immuno-isolated α3 nicotinic AChRs and of agonist potency to induce specific ⁸⁶Rb⁺ efflux in cultured human neuroblastoma SH-SY5Y cells. Numbers on graph refer to compounds in Figure 2 and Table 1.

Interaction with α₇ nicotinic AChRs

[¹²⁵I]-α-BGT binding was determined to mAb 306 immuno-isolated α₇ nicotinic AChRs of SH-SY5Y cells and in putative α₇ nicotinic AChRs of rat brain membranes. The α₇ nicotinic AChRs of the human neuroblastoma cells are most sensitive to THPCH-IMI (Figure 6) and then CH-IMI (IC₅₀s 1.2 and 6.1 μM, respectively), least sensitive to IMI, AAP and NTP (IC₅₀s 210–>300 μM) and with intermediate sensitivity to nicotine and the other five nicotinoids (IC₅₀s 11–33 μM) (Table 1). The same structure-activity relationships are obtained with the several-fold more sensitive rat brain putative α₇ nicotinic AChR (Figure 6 and Table 1) as clearly apparent by a correlation plot for inhibition of the SH-SY5Y and rat brain receptors (Figure 7).

Figure 6.

Displacement by THPCH-IMI, (−)-nicotine and α-BGT of [¹²⁵I]-α-BGT binding to α₇ nicotinic AChRs immuno-isolated from human neuroblastoma SH-SY5Y cells (top) and to putative α₇ nicotinic AChRs from rat whole brain membranes (bottom). The extracted cell membranes with lysis buffer were reacted with mAb 306-precoated wells overnight at 4°C, and then the immunoprecipitated α₇ nicotinic AChRs were incubated overnight at 4°C with 2 nM of [¹²⁵I]-α-BGT in competition with a test compound. Rat whole brain membranes (200 μg protein) were incubated with 2 nM of [¹²⁵I]-α-BGT for 4 h at 37°C in the absence and the presence of the test compound. Data points represent means of three experiments with s.d.

Figure 7.

Correlation for nicotinoids of inhibitory potency for [¹²⁵I]-α-BGT binding to immuno-isolated α₇ nicotinic AChRs from human neuroblastoma SH-SY5Y cells and to putative α₇ nicotinic AChRs from rat brain membranes. Numbers on graph refer to compounds in Figure 2 and Table 1.

Interaction with putative α₄β₂ nicotinic AChRs

[³H]-Nicotine binding in mouse or rat brain membranes was used to determine putative α₄β₂ nicotinic AChRs. Data for mouse [³H]-nicotine binding were taken from our previous report (Chao & Casida, 1997). There are four potent inhibitors in this assay (IC₅₀ 0.009–0.033 μM for nicotine, THPCH-IMI, DN-IMI and CH-IMI) with moderate activity for SCH-IMI, DN-THP and Cl-TMNI (IC₅₀ 0.093–0.25 μM), lower activity for IMI and AAP and the lowest activity for NTP.

Toxicity

The i.p. toxicity rating in mice of two chloropyridinyl compounds (CH-IMI and THPCH-IMI) and one chlorothiazolyl compound (Cl-TMNI) is similar to that of nicotine while DN-IMI is a little less toxic than nicotine and the others are much less active (Table 1). The poisoning signs at an LD₅₀ dose included tremors and seizures and appeared to be consistent with action on nicotinic AChRs.

Discussion

The chloropyridinyl nicotinoid insecticide IMI has little or no activity in vertebrate systems based on six observations: (1) the failure to recognize [³H]-IMI specific binding site(s) in brain from several mammalian and avian species and the electric eel (Liu & Casida, 1993); (2) low potency as an inhibitor of [³H]-α-BGT binding and low agonistic effect in muscle-type nicotinic AChR from Torpedo electric organ (not only for IMI but also for AAP and NTP) (Tomizawa et al., 1995); (3) little activity as an inhibitor of [³H]-nicotine binding to rat and mouse brain membranes (Yamamoto et al., 1995; Chao & Casida, 1997); (4) very weak agonistic action in mouse N1E-115 neuroblastoma and BC3H1 muscle cells (IMI and an analogue) (Zwart et al., 1992; 1994); (5) low activity in ion channel activation compared to acetylcholine with rat α₄β₂ and α₇ subtypes expressed in Xenopus oocytes (Yamamoto et al., 1998); (6) weak or partial agonistic nature with recombinant chick α₄β₂ receptor (Matsuda et al., 1998). The present study extends these relationships for IMI, AAP and NTP to include very low apparent affinity to α₁, α₃ and α₇ nicotinic AChR subtypes. However, these conclusions are based only on the parent insecticide which might undergo metabolic activation such as the case of IMI to DN-IMI (Klein, 1994; Chao & Casida, 1997). The imine metabolite DN-IMI is much more effective than the parent IMI, not only for putative α₄β₂ receptors (Chao & Casida, 1997) but also for α₁, α₃ and α₇ nicotinic AChRs (this study).

The chloropyridinyl group is an important structural feature for several nicotinic agonists conferring outstanding potency but little selectivity with epibatidine (Holladay et al., 1997) and remarkable α₄β₂ nicotinic AChR specificity with ABT-594 (Bannon et al., 1998). The binding affinities of analogues without the chlorine atom in the insect receptor are several-fold less than those with the chlorine atom (Liu et al., 1993; Tomizawa & Yamamoto, 1993). The chlorothiazolyl replacement for the chloropyridinyl moiety (Cl-TMNI versus CH-IMI) greatly reduces potency in the mammalian receptor assays but not the toxicity to mammals or activity at the insect nicotinic AChR (Liu et al., 1993; Chao & Casida, 1997).

Differential nicotinic AChR subtype selectivity is conferred by minor structural changes in the chloropyridinyl nicotinoid insecticides (Figure 8). The desnitro analogues favour the α₁, α₃ and putative α₄β₂ receptor subtypes and the nitromethylene analogues the α₇ nicotinic AChRs. THPCH-IMI and CH-IMI are much more potent than nicotine on α₇ nicotinic AChRs while SCH-IMI and Cl-TMNI are less active than the first two compounds. All four of these nitromethylenes are much less active than nicotine on α₁ and α₃ nicotinic AChRs. Interestingly, the two desnitro analogues DN-IMI and DN-THP prefer the α₃ over the α₇ nicotinic AChRs. The change from a five-membered imidazolidine to a six-membered tetrahydropyrimidine ring greatly reduces the potency of the imines (DN-IMI versus DN-THP) but increases the activity of the nitromethylenes (CH-IMI versus THPCH-IMI) in all nicotinic AChR subtypes.

Figure 8.

Toxicity of nicotinoid insecticides and selected analogues in mammals involves differential action at multiple receptor subtypes conferred by minor structural changes. The potency rankings in parentheses are generalizations.

The mammalian toxicity of the nicotinoid insecticides and analogues studied on an overall basis is most closely related to their potency at α₇ nicotinic AChRs with decreasing relationships sequentially at the α₄β₂, α₃ and α₁ nicotinic AChRs. More specifically, the nitromethylenes are more potent in the α₇-containing receptors while DN-IMI is particularly potent at α₁, α₃ and putative α₄β₂ receptors (Figure 8). Thus, the toxicity of the nicotinoid insecticides in mammals may involve action at multiple receptor subtypes with selectivity conferred by minor structural changes.

Abbreviations

AAP

acetamiprid

AChR

acetylcholine receptor

α-BGT or [¹²⁵I]α-BGT

α-bungarotoxin or its 125-iodine labelled ligand

CH-IMI

nitromethylene analogue of IMI

Cl-TMNI

chlorothiazolyl analogue of CH-IMI

DMEM

Dulbecco's modified Eagle's medium

DN-IMI

desnitro metabolite of IMI

DN-THP

tetrahydropyrimidine analogue of DN-IMI

EC₅₀

molar concentration of test compound to induce 50% specific ⁸⁶Rb⁺ efflux

FBS

foetal bovine serum

IC₅₀

molar concentration of test compound for 50% inhibition of specific radioligand binding

IMI

imidacloprid

mAb

monoclonal antibody

NTP

nitenpyram

PBS

phosphate-buffered saline

PMSF

phenylmethanesulphonyl fluoride

⁸⁶Rb⁺

⁸⁶rubidium ion

SCH-IMI

thiazolidine analogue of CH-IMI

THPCH-IMI

tetrahydropyrimidine analogue of CH-IMI