Effects of novel meta-diamide insecticides on GABA type A receptors α1β2γ2 and α1β3γ2 and on glycine receptor α1β

Abstract

Novel meta-diamide insecticides [3-benzamido-N-(4-(perfluoropropan-2-yl)phenyl)benzamides] are a distinct class of insect RDL GABA receptor antagonists that have been suggested to act at or near G336 in the M3 region of the Drosophila RDL GABA receptor. Here, we demonstrate the low-level antagonist activities of novel meta-diamides against the human GABA type A receptor (GABA_AR) α1β2γ2S, mammalian GABA_AR α1β3γ2S, and the human glycine receptor (GlyR) α1β. The Gly residue at position 336 in the M3 region of the Drosophila RDL GABA receptor is essential for its high sensitivity to meta-diamides. Therefore, the effects of an equivalent mutation (A288G) in human GlyR α1 on the antagonist activities of meta-diamides were studied. The A288G mutation dramatically increased the antagonist activities of meta-diamides against GlyR α1. Thus, A288 in GlyR α1 is important for this receptor’s sensitivity to meta-diamide insecticides. Our study is useful for understanding the mechanisms underlying meta-diamide selectivity.

Introduction

The insect ionotropic γ-aminobutyric acid (GABA) receptor is a major target of insecticides.^1,2) The GABA receptor is comprised of five subunits, each with a large extracellular agonist-binding N-terminal domain and four membrane-spanning regions designated M1–M4.

Cyclodienes and lindane are first-generation insecticides that act on insect RDL GABA receptors as noncompetitive antagonists. They are thought to bind to the pores formed by the M2 regions. In particular, A2′ and T6′ (index numbers for M2)³⁾ are important for their binding.^4–7) Two amino acid substitutions (A2′S and A2′G) have been reported to be associated with cyclodiene resistance in various insect species.^8–15)

Phenylpyrazoles, such as fipronil, are second-generation insecticides that act on insect RDL GABA receptors as noncompetitive antagonists. Fipronil has greater target site specificity for GABA receptors in insects, but not in mammals, than do first-generation noncompetitive antagonists such as dieldrin, α-endosulfan, and lindane.^16,17) The binding of fipronil was also suggested to be related to A2′ and T6′ of RDL GABA receptors,^4–7) and A2′S and A2′G mutations produce cross-resistance to fipronil. However, the level of resistance is low.¹⁴⁾ We found A2′N mutation in the RDL GABA receptor in serious rice pests, the white backed planthopper Sogatella furcifera,^18–20) and the small brown planthopper, Laodelphax striatellus,²¹⁾ and suggested that this mutation confers fipronil resistance. Thus, insects have developed resistance to first- and second-generation noncompetitive antagonists that target insect RDL GABA receptors.

In a previous study, we demonstrated that novel meta-diamide insecticides [3-benzamido-N-(4-(perfluoropropan-2-yl)phenyl)benzamides] are insect RDL GABA receptor antagonists and suggested that meta-diamides act at a different site than do the first- and second-generation noncompetitive antagonists.^22,23) Meta-diamides have been suggested to act at or near G336 in the M3 region of the Drosophila RDL GABA receptor. Furthermore, studies using a series of G336 mutant Drosophila RDL GABA receptors have demonstrated that Gly in position 336 is essential for high sensitivity to meta-diamides.²²⁾ However, the target site specificities of meta-diamides have never been studied in the GABA receptors of insects and mammals.

We studied the effects of meta-diamides on the human GABA type A receptor (GABA_AR) α1β2γ2S, mammalian GABA_AR rat α1/human β3/rat γ2S in WSS-1 cells,²⁴⁾ and the human glycine receptor (GlyR) α1β. GABA_AR α1β2γ2 is most abundant in the brain²⁵⁾ and is considered a representative GABA_AR. There has been a lot of research on mammalian GABA_AR α1β3γ2 because the general anesthetic etomidate binds M286 of mammalian GABA_AR β3 and M236 of GABA_AR α1.²⁶⁾ M286 of mammalian GABA_AR β3 is equivalent to G336 in the M3 region of the Drosophila RDL GABA receptor. GlyR and GABA_AR belong to the Cys-loop ion channel family and A288 in human GlyR α1, which corresponds to G336 in the M3 region of the Drosophila RDL GABA receptor, is important for insensitivity to ivermectin. A288G mutation confers high ivermectin sensitivity to human GlyR α1.²⁷⁾

In this study, we demonstrate that the antagonist activities of meta-amides are considerably lower in human GABA_AR α1β2γ2S, mammalian GABA_AR α1β3γ2S, and human GlyR α1β than in insect RDL GABA receptors. Furthermore, we suggest that A288 in human GlyR α1 is important for insensitivity to meta-diamides.

Materials and Methods

1. Chemicals

GABA, picrotoxin, and ZnCl₂ were obtained from Wako Pure Chemical Industries, Ltd. (Chuo-ku, Osaka, Japan). Fipronil (92.8%) and meta-diamides 1, 7, and 9 (>99%; Fig. 1) were synthesized at the Agrochemical Research Center, Mitsui Chemicals Agro (Mobara, Chiba, Japan).

Fig. 1. Structures of meta-diamides 1, 7, and 9

2. Construction of vectors to express human GABA_ARs

The sequence human GABA_AR α1 was amplified from the human brain Quick-Clone cDNA library (Clontech Laboratories Inc., Mountain View, CA, USA) using the primers ALF1-FF (CCG GAG ATC TTA TAA ATA TGA GGA AAA GTC CAG GTC TGT CTG) and ALF-RR-R (CCG GGA ATT CAT TGA TGT GGT GTG GGG GCT). The PCR products were cloned into a pCR2.1-TOPO® Vector plasmid (Invitrogen, Carlsbad, CA, USA). The resultant plasmid was digested with BglII and NotI, and DNA encoding human GABA_AR α1 was inserted between the BamHI and NotI sites of pENTR1A (Invitrogen).

The N-terminus of human GABA_AR α1 was tagged with a 14-amino acid V5 epitope (GKPIPNPLLGLDST) by overlap extension PCR. Two PCR products were amplified using the primer sets ALF1-FF/3′haNV5 (ACC GAG GAG AGG GTT AGG GAT AGG CTT ACC TTG TAA TGA CGG CTG TCC) and 5′haNV5 (CCT AAC CCT CTC CTC GGT CTC GAT TCT ACG GAT GAA CTT AAA GAC AAT)/ALF-RR-R. Next, the resultant PCR product was amplified using the primer set ALF1-FF/ALF-RR-R, with the two previous PCR products as templates. This PCR product was digested with BglII and NotI and inserted between the BamHI and NotI sites of pcDNA3.1-hygro (Invitrogen). The resultant plasmid was named pcDNA-hGABARα1NV5-hygro.

The sequence encoding human GABA_AR β2 was amplified from the human brain Quick-Clone cDNA library using the primers BETA2F (CCG GAG ATC TTA TAA TAT GTG GAG AGT CCG GAA AAG GGG C) and BET-2RR-R (CCG GGA ATT CTT TTA GTT CAC ATA GTA AAGCCAATA). The PCR products were cloned into the pCR2.1-TOPO® Vector plasmid. The resultant plasmid was digested with BglII and EcoRI, and DNA encoding human GABA_AR β2 was inserted between the BamHI and EcoRI sites of pENTR1A. The resultant plasmid was designated pENTR1A-hGABARβ2.

The N-terminus of human GABA_AR β2 was tagged with an 8-amino acid FLAG epitope (DYKDDDDK) by overlap extension PCR. Two PCR products were amplified using the primer sets BET2F (CTC GGA TCC TAT AAT ATG TGG AGA GTC CGG AAA AGG GGC)/3′hb2NFLAG (GTC ATC GTC GTC CTT GTA GTC ATT GAC ACT CTG CGC ACA) and 5′hb2NFLAG (TAC AAG GAC GAC GAT GAC AAG GAC CCT AGT AAT ATG TCG)/BET-2RR-R. Next, the resultant PCR product was amplified using the primer set BET2F/BET-2RR-R, with the two previous PCR products as templates. The resultant PCR product was digested with BamHI and EcoRI and inserted between the BamHI and EcoRI sites of pENTR1A. The resultant plasmid was named pENTR1A-hGABARβ2NFLAG. All of the amplified DNA sequences were verified by sequencing. To construct a vector to express the GABA_AR ^(FLAG)β2 gene, the ccdB gene of pcDNA™3.2/V5-DEST (Invitrogen) was replaced with the GABA_AR ^(FLAG)β2 gene using the Gateway® cloning system (Invitrogen). The expression vector for the GABA_AR ^(FLAG)β2 gene was designated pcDNA-GABARβ2NFLAG-neo.

The sequence encoding human GABA_AR γ2S was amplified from the human brain Quick-Clone cDNA library using the primers GAN2FF (CCG GAG ATC TCT TAT AAA TAT GAG TTC GCC AAA TAT ATG G) and GAN-2RR-R (CCG GGA ATT CTC CTC ACA GGT AGA GGT AGG AGA CCC A). The PCR products were cloned into the pCR2.1-TOPO® Vector plasmid. The resultant plasmid was digested with BglII and EcoRI, and DNA encoding human GABA_AR γ2S was inserted between the BamHI and EcoRI sites of pENTR1A. The resultant plasmid was designated pENTR1A-hGAB_AR γ2S.

The N-terminus of human GABA_AR γ2S was tagged with a 10-amino acid myc epitope (EQKLISEEDL) by overlap extension PCR. Two PCR products were amplified using the primer sets GAN2F (CTC GGA TCC TAT AAA TAT GAG T TCG CCA AAT ATA T)/3′hr2γmyc (TTC TTC TGA TAT TAG CTT TTG TTC ATC AGA TTT CTG GCT AGT) and 5′hr2Nmyc (AAG CTA ATA TCA GAA GAA GAC CTA GAT GAC TAT GAA GAT TAT)/GAN-2RR-R. Next, the resultant PCR product was amplified using the primer set GAN2F/GAN-2RR-R, with the two previous PCR products as templates. This PCR product was digested with BamHI and EcoRI and inserted between the BamHI and EcoRI sites of pENTR1A. The resultant plasmid was named pENTR1A-hGABARγ2SNmyc. All of the amplified DNA sequences were verified by sequencing. To construct a vector to express the GABA_AR ^(myc)γ2S gene, pENTR1A-hGABARγ2SNmyc was digested with BamHI and EcoRI, and the BamHI–EcoRI DNA fragment containing the human GABA_AR γ2S gene was inserted between the BamHI and EcoRI sites of pcDNA3.1/Zeo. The expression vector for the GABA_AR ^(myc)γ2S gene was designated pcDNA-GABARγ2SNmyc-zeo.

3. Construction of vectors to express human GlyRs

The sequence encoding human GlyR α1 was amplified from the human brain Quick-Clone cDNA library. DNA encoding the N-terminus of human GlyR α1 was amplified using the primers Gly5′ (TTT TGG ATC CAA AAA ACC TAT AAA ATG TAC AGC TTC AAT ACT CT) and GlyHin3′ (TCC AAA GCT TTC CAG TTG CAT). DNA encoding the central region of human GlyR α1 was amplified using the primers GlyHin5′ (ATG CAA CTG GAA AGC TTT GG) and GlyCla3′ (GAT CCA AAT ATC GAT GGC TTT). DNA encoding the C-terminus of human GlyR α1 was amplified using nested PCR. The pairs of primers used for the first and second PCR amplifications were GlyHin5′/Gly3′ (TTT TGC GGC CGC TCA CTG GTT GTG GAC GTCCTC) and GlyCla5′(AAA GCC ATC GAT ATT TGG ATG)/Gly3′, respectively. The PCR products were cloned into the pCR2.1-TOPO® Vector plasmid. During these PCRs, a ClaI site was introduced without altering the amino acid sequence. DNAs encoding the N-terminus, central region, and C-terminus of human GlyR α1 were digested with BamHI and HindIII, HindIII and ClaI, and ClaI and NotI, respectively, and inserted between the BamHI and NotI sites of pENTR1A. The resultant plasmid was named pENTR1A-hGlyRα1. An A288G mutation was introduced by PCR using the primer set GlyG5′ (GCC ATC GAT ATT TGG ATG GGA GTT TGC CTGCTC)/Gly3′. The PCR products were digested with ClaI and NotI and inserted between the ClaI and NotI sites of pENTR1A-hGlyRα1. The resultant plasmid was designated pENTR1A-hGlyRα1-A288G. All amplified DNA sequences were verified by sequencing. To construct a vector to express GlyR α1 and GlyR α1-A288G genes, the ccdB gene of pcDNA™3.2/V5-DEST was replaced with GlyR α1 gene or GlyR α1-A288G gene using the Gateway® cloning system. The resultant plasmids were named pcDNA-GlyRα1 and pcDNA-GlyRα1-A288G, respectively.

A plasmid containing DNA encoding human GlyR β was obtained from Thermo Fisher Scientific KK (Yokohama, Kanagawa, Japan). The sequence encoding human GlyR β was amplified from the plasmid using primers Glb50 (TTT TGG TAC CTA TAA ATA TGA AGT TTT TAT TGA CAA CT) and Glb30 (TTT TGC GGC CGC TCA TAA ATA TAT AGA CCAATATAT) and was cloned into the pCR2.1-TOPO® Vector plasmid. After digestion with KpnI and NotI, DNA encoding human GlyR β was inserted between the KpnI and NotI sites of pcDNA3.1-hygro (Invitrogen). The resultant plasmid was named pcDNA-GlyRβ.

4. Cell lines and cell culture

HEK-293-H cells were obtained from Invitrogen and cultured in Dulbecco’s Modified Eagle’s Medium+GlutaMax™-1 (Invitrogen) supplemented with 10% fetal calf serum, penicillin (50 U/mL), and streptomycin (50 µg/mL) at 37°C in a 5% CO₂ incubator. WSS-1 cells²⁴⁾ that expressed mammalian GABA_AR rat α1/ human β3/ rat γ2S were obtained from the American Type Culture Collection (Manassas, VA, USA). The culture medium of the neomycin-resistant WSS-1 cells and cells stably expressing GlyR α1 or GlyR α1-A288G was supplemented with 500 µg/mL of G418 (Calbiochem-Novabiochem, San Diego, CA, USA). The culture medium of the neomycin- and hygromycin-resistant cells that expressed GABA_AR ^(V5)α1^(FLAG)β2, GlyR α1β, or GlyR α1-A288Gβ was supplemented with 500 µg/mL of G418 and 200 µg/mL of hygromycin B (Wako Pure Chemical Industries, Ltd.). The culture medium of the neomycin-, hygromycin-, and zeocin-resistant cells that expressed GABA_AR ^(V5)α1^(FLAG)β2^(myc)γ2S was supplemented with 500 µg/mL of G418, 200 µg/mL of hygromycin B, and 50 µg/mL of zeocin (Invitrogen).

5. Establishment of stable cell lines expressing GABA_ARs and GlyRs

Stable transfection was performed using Lipofectamine 2000 Reagent (Life Technologies, Rockville, MD, USA), according to the manufacturer’s recommendations. Cells were seeded into 6-well plates at a density of 1.6×10⁶ cells/well. After incubation for 16–24 hr, but prior to transfection, the cells were typically 80–90% confluent.

Approximately 3 µg of DNA was mixed with 10 µL of Lipofectamine 2000 and 500 µL of Opti-MEM® (Invitrogen) and incubated at room temperature for 30 min to allow the formation of DNA–liposome complexes. Next, the solution containing the DNA–liposome complexes was added to the cells. After 14–16 hr, the transfectants were trypsinized and replated onto 100-mm dishes. Cells stably expressing GlyR α1 or GlyR α1-A288G were selected in culture medium supplemented with 500 µg/mL of G418. Cells stably expressing GABA_AR ^(V5)α1^(FLAG)β2, GlyR α1β, or GlyRα1-A288Gβ were selected in culture medium supplemented with 500 µg/mL of G418 and 200 µg/mL of hygromycin B. Cells stably expressing GABA_AR ^(V5)α1^(FLAG)β2^(myc)γ2S were selected in culture medium supplemented with 500 µg/mL of G418, 200 µg/mL of hygromycin B, and 50–100 µg/mL of zeocin. Resistant colonies were obtained using cloning cylinders, isolated by trypsinization, and cultured in 24-well tissue culture plates. At confluence, one-seventh of the cells were transferred to 24-well plates as a master plate, and four-sevenths of the cells were split into duplicate poly-D-lysine-coated black 96-well plates with clear bottoms (BD Biosciences, Bedford, MA, USA) for screening. After incubation for 14–16 hr, cells expressing GABA_AR ^(V5)α1^(FLAG)β2 were selected by measuring their response to 1 mM GABA. Cells expressing GlyRs α1 and α1-A288Gβ were selected by measuring the cellular response to 1 mM glycine. Cells expressing GABA_AR ^(V5)α1^(FLAG)β2^(myc)γ2S were selected by measuring the cellular response to 100 µM GABA with or without 10 µM ZnCl₂. Cells expressing GlyRs α1β and α1-A288Gβ were selected by measuring the cellular response to 40 and 20 µM glycine with or without 100 µM picrotoxin, respectively. The selected cells in the master plate were used for further characterization.

6. Immunostaining of cells expressing GABA_ARs ^(V5)α1^(FLAG)β2 and ^(V5)α1^(FLAG)β2^(myc)γ2S

Cells were seeded into each well of 24-well plates at a density of 4×10⁵ cells/well for immunostaining using anti-V5 IgG and anti-FLAG IgG antibodies. For immunostaining using anti-myc antibody, the cells were seeded into each well of 24-well plates at a density of 5×10⁵ cells/well. After culturing for 16 hr at 37°C, the cells were washed with phosphate-buffered saline (PBS) and fixed with 3.7% paraformaldehyde in PBS. The cells were washed three times with PBS and treated with blocking solution (PBS containing 0.5% bovine serum albumin and 10% fetal bovine serum) for 30 min at room temperature. The cells were then incubated with mouse anti-V5 IgG (Invitrogen), mouse anti-FLAG IgG (Sigma-Aldrich, St. Louis, MO, USA), or mouse anti-myc IgG (Invitrogen) antibodies at a dilution of 1 : 1000 for 60 min at room temperature in blocking solution. The cells were then washed two times with PBS and one time with blocking solution. Next, the cells were incubated for 60 min at room temperature in blocking solution containing alkaline phosphatase-conjugated rabbit anti-mouse IgG antibody (Sigma-Aldrich) at a 1 : 1000 dilution. The cells were washed three times with PBS, and a 1.6-mM solution of p-nitrophenyl phosphate disodium salt (Sigma-Aldrich) in 9.8% diethanolamine containing 0.5 mM MgCl₂ (pH 9.6) was added to the wells. After 15 min, the absorbance was measured at 405 nm. The data were analyzed using one-way ANOVA followed by Dunnett’s test using Prism 6.00 software (GraphPad Software, Berkeley, CA, USA). Differences yielding a p value of <0.05 were considered significant.

7. Membrane potential assay

GABA_ARs and GlyRs were characterized using the FLIPR® Membrane Potential Assay Kit (Molecular Devices Corp., Sunnyvale, CA, USA). Cells that expressed GABA_ARs or GlyRs were split into poly-D-lysine-coated black 96-well plates with clear bottoms at a density of 5×10⁴ cells/well. After incubation for 14–24 hr, the medium was aspirated, and the wells were washed with 200 µL of Hanks’ buffer. One-hundred microliters of Hanks’ buffer was added to each well, followed by 100 µL of loading buffer [blue dye (Invitrogen) dissolved in Hanks’ buffer], and the plate was incubated at room temperature for 30 min. In the antagonism assay, the cells were incubated with antagonists for 1 hr, except for the antagonism assay using GABA_ARs ^(V5)α1^(FLAG)β2 and ^(V5)α1^(FLAG)β2^(myc)γ2S. For the antagonism assay of ZnCl₂ using cells expressing GABA_ARs ^(V5)α1^(FLAG)β2 and ^(V5)α1^(FLAG)β2^(myc)γ2S, the cells were incubated with antagonists for 74 sec. For the antagonism assays of fipronil and meta-diamides using cells expressing GABA_AR ^(V5)α1^(FLAG)β2^(myc)γ2S, the cells were incubated with antagonists for 30 min because the response to GABA in cells expressing GABA_AR ^(V5)α1^(FLAG)β2^(myc)γ2S was reduced after 1 hr in the presence of 0.1% DMSO. GABA and glycine concentrations corresponding to their EC₈₀ values were added to the cells to assay GABA_ARs and GlyRs, respectively. The plates were assayed using a FlexStation® II plate reader (Molecular Devices Corp.) where the emissions were measured at 560 nm and excited at 530 nm. The percentage of inhibition was determined using the formula: % Inhibition=(1−R/R_EC80)×100, where R_EC80 is the fluorescence at EC₈₀ GABA, and R is the fluorescence in response to each antagonist concentration in the presence of an EC₈₀ GABA. The pEC₅₀ [=log (1/EC₅₀)] values, pIC₅₀ [=log (1/IC₅₀)] values, and slope factors (Hill coefficients) were determined using a four-parameter logistic curve-fitting program with Prism 6.00 software (GraphPad Software). The EC₈₀ values were determined using a four-parameter logistic curve-fitting program (SoftMax Pro v5 software, Molecular Devices). These experiments were performed three times in duplicate for each compound.

Results

1. Establishment of HEK cells that stably express human GABA_ARs

Stable cell lines were screened for GABA_AR ^(V5)α1^(FLAG)β2 expression on the basis of their response to 1 mM GABA. Because GABA_AR α1 or GABA_AR β2 homomers cannot respond to GABA,²⁸⁾ cells responding to GABA should express GABA_AR α1β2. Cells expressing GABA_AR ^(V5)α1^(FLAG)β2 were transfected with the GABA_AR^(myc)γ2S gene. Stable cell lines were screened for GABA_AR ^(V5)α1^(FLAG)β2^(myc)γ2S expression on the basis of their response to 100 µM GABA with or without 10 µM ZnCl₂. Cells expressing GABA_AR α1β2γ2S should experience reduced inhibition of the GABA-induced potential change by Zn²⁺ as compared to those expressing GABA_AR α1β2.²⁸⁾

To confirm the expression of GABA_ARs α1β2 and α1β2γ2S, V5, FLAG, and myc epitopes were added to the α1, β2, and γ2S subunits between amino acids 4 and 5 of the mature subunits to generate ^(V5)α1, ^(FLAG)β2, and ^(myc)γ2S, respectively. The expression of GABA_ARs ^(V5)α1^(FLAG)β2 and ^(V5)α1^(FLAG)β2^(myc)γ2S were confirmed by immunostaining (Fig. 2). The addition of these small epitopes to the extra N-terminal domain of GABA_AR subunits appeared to have no effect on their function.²⁹⁾

Fig. 2. Immunostaining of GABA_ARs ^(V5)α1^(FLAG)β2 and ^(V5)α1^(FLAG)β2^(myc)γ2S. Immunostaining of HEK cells (n=4) and HEK cells that stably express GABA_ARs ^(V5)α1^(FLAG)β2 (n=4) and ^(V5)α1^(FLAG)β2^(myc)γ2S (n=4). The cells were immunostained with mouse anti-V5, mouse anti-FLAG, or mouse anti-myc antibody, followed by alkaline phosphatase-conjugated rabbit anti-mouse IgG antibody, and then the absorbance was measured at 405 nm. Data analysis was performed using one-way ANOVA followed by Dunnett’s test using Prism 6.00 software. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. *p<0.05 was considered significant.

2. Effects of GABA and Zn²⁺ on human GABA_ARs

The GABA EC₅₀ values for human GABA_ARs ^(V5)α1^(FLAG)β2 and ^(V5)α1^(FLAG)β2 ^(myc)γ2S were 0.644 µM (pEC₅₀=6.19±0.10) and 1.31 µM (pEC₅₀=5.88±0.03), respectively (Fig. 3A). This indicates that GABA_AR ^(myc)γ2S expression results in a slight increase in the EC₅₀ value. The Hill coefficients for human GABA_ARs ^(V5)α1^(FLAG)β2 and ^(V5)α1^(FLAG)β2 ^(myc)γ2S were 0.85±0.16 and 1.3±0.1, respectively.

Fig. 3. Responses of human GABA_ARs ^(V5)α1^(FLAG)β2 and ^(V5)α1^(FLAG)β2^(myc)γ2S to GABA and ZnCl₂. (A) GABA concentration–response curve for human GABA_ARs ^(V5)α1^(FLAG) β2 (open circle) and ^(V5)α1^(FLAG)β2^(myc)γ2S (solid circle). Results are expressed as the percentages of maximal responses of GABA_ARs ^(V5)α1^(FLAG)β2 and ^(V5)α1^(FLAG)β2^(myc)γ2S to GABA. Vertical bars represent SEM for three independent experiments, which were conducted in duplicate. (B) Concentration–response curves for the inhibition of GABA-induced currents by ZnCl₂. Experiments were conducted with GABA_ARs ^(V5)α1^(FLAG)β2 (open circle) and ^(V5)α1^(FLAG)β2^(myc)γ2S (solid circle) in the presence of GABA. The data are expressed as the percent of inhibition of the GABA_AR response relative to the EC₈₀ values of GABA in the absence of each test compound. The vertical bars represent SEM for three independent experiments, which were conducted in duplicate.

The IC₅₀ values of ZnCl₂ for human GABA_ARs ^(V5)α1^(FLAG)β2 and ^(V5)α1^(FLAG)β2 ^(myc)γ2S were 1.50 µM (pIC₅₀=5.83±0.41) and >100 µM (pIC₅₀<4.00), respectively (Fig. 3B). This demonstrates that GABA_AR ^(myc)γ2S expression dramatically reduces the inhibitory activity of Zn²⁺.

3. Effects of fipronil and meta-diamides on human GABA_AR ^(V5)α1^(FLAG)β2^(myc)γ2S

The IC₅₀ value of fipronil for human GABA_AR ^(V5)α1^(FLAG)β2 ^(myc)γ2S was 0.289 µM (pIC₅₀=6.36±0.04) (Fig. 4). In contrast, the IC₅₀ values of meta-diamides 1, 7, and 9 (Fig. 1) for human GABA_AR ^(V5)α1^(FLAG)β2 ^(myc)γ2S were >3 µM (pIC₅₀<5.52) (Fig. 4). This demonstrates the lower potency of meta-diamides for human GABA_AR ^(V5)α1^(FLAG)β2 ^(myc)γ2S than that of fipronil.

Fig. 4. Responses of human ^(V5)α1^(FLAG)β2^(myc)γ2S to fipronil and meta-diamides. Concentration–response curves for the inhibition of GABA-induced currents by fipronil (solid circle) and meta-diamides 1 (solid square), 7 (solid triangle), and 9 (solid diamond). Experiments were conducted with GABA_AR α1β2γ2S in the presence of GABA. The data are expressed as the percent of inhibition of the GABA_AR α1β2γ2S response relative to the EC₈₀ values of GABA in the absence of each test compound. The vertical bars represent SEM for three independent experiments, which were conducted in duplicate.

4. Effects of GABA, fipronil, and meta-diamides on mammalian GABA_AR α1β3γ2S stably expressed in WSS-1 cells

WSS-1 cells, which have been shown to express rat GABA_AR α1, rat GABA_AR γ2S, and human GABA_AR β3,²⁷⁾ were used to study the effects of meta-diamides on mammalian GABA_AR α1β3γ2S. The GABA EC₅₀ value for mammalian GABA_AR α1β3γ2S was 14.7 µM (pEC₅₀=4.83±0.08). The Hill coefficient was 1.3±0.3. Fipronil inhibited the EC₈₀ GABA-induced membrane change with an IC₅₀ value of 0.417 µM (pIC₅₀=6.38±0.010) (Fig. 5), but the IC₅₀ values of meta-diamides 1, 7, and 9 were >3 µM (pIC₅₀<5.52) (Fig. 5). This demonstrates the lower inhibitory activity of meta-diamides against mammalian GABA_AR α1β3γ2S than that of fipronil.

Fig. 5. Responses of mammalian GABA_AR α1β3γ2S to fipronil and meta-diamides. Concentration–response curves for the inhibition of GABA-induced current by fipronil (solid circle) and meta-diamides 1 (solid square), 7 (solid triangle), and 9 (solid diamond). Experiments were performed with GABA_AR α1β3γ2S. Data are expressed as the percent of inhibition of the GABA_AR α1β3γ2S response to EC₈₀ values of GABA in the absence of each test compound. Vertical bars represent SEM for three independent experiments, which were conducted in duplicate.

5. Establishment of HEK cells that stably express human GlyRs

Cells stably expressing GlyRs α1 and α1-A288G were screened by their response to 1 mM glycine. Cells that demonstrated high glycine activity were selected as cells expressing GlyRs α1 and α1-A288G. To screen for HEK cells expressing GlyRs α1β or α1-A288Gβ, the responses of cells to 40 and 20 µM glycine with or without 100 µM picrotoxin, respectively, were measured. Cells that demonstrated low sensitivity for picrotoxin were selected as cells expressing GlyR α1β or α1-A288Gβ.

6. Effects of glycine and picrotoxin on human GlyRs

The glycine EC₅₀ values for GlyRs α1, α1-A288G, α1β, and α1-A288Gβ were 35.8 µM (pEC₅₀=4.45±0.04), 11.0 µM (pEC₅₀=4.96±0.10), 17.5 µM (pEC₅₀=4.76±0.06), and 4.16 µM (pEC₅₀=5.38±0.11), respectively (Fig. 6A). This indicates that co-expression of GlyRs α1β and α1-A288Gβ decreases the EC₅₀ values. The Hill coefficients for GlyRs α1, α1-A288G, α1β, and α1-A288Gβ were 2.3±0.5, 1.3±0.4, 1.8±0.4, and 1.6±0.5, respectively.

Fig. 6. Responses of human GlyRs to glycine and picrotoxin. (A) Glycine concentration–response curve for human GlyRs α1 (open circle), α1-A288G (open square), α1β (solid circle), and α1-A288Gβ (solid square). Results are expressed as the percentages of the maximal responses of human GlyRs to glycine. Vertical bars represent SEM for three independent experiments, which were conducted in duplicate. (B) Concentration–response curves for the inhibition of glycine-induced currents by ZnCl₂. Experiments were conducted with human GlyRs α1 (open circle), α1-A288G (open square), α1β (solid circle), and α1-A288Gβ (solid square) in the presence of glycine. The data are expressed as the percent of inhibition of the GlyRs response relative to the EC₈₀ values of glycine in the absence of each test compound. The vertical bars represent SEM for three independent experiments, which were conducted in duplicate.

As picrotoxin is a significantly weaker antagonist for GlyR α1β than for GlyR α1,³⁰⁾ GlyR α1β or α1-A288Gβ expression was confirmed by the decreased antagonist activity of picrotoxin against these receptors as compared with that against GlyRs α1 and α1-A288G. The picrotoxin IC₅₀ values for GlyRs α1 and α1-A288G were 8.08 µM (pIC₅₀=5.09±0.09) and 4.20 µM (pIC₅₀=5.38±0.08), respectively (Fig. 6B). In contrast, the picrotoxin IC₅₀ values for GlyRs α1β and α1-A288Gβ were >100 µM (pIC₅₀<4.00) (Fig. 6B).

7. Effects of fipronil and meta-diamides on human GlyRs

As shown in Fig. 7A, the fipronil IC₅₀ values for GlyRs α1 and α1β were 0.967 µM (pIC₅₀=6.02±0.09) and 0.815 µM (pIC₅₀=6.09±0.32), respectively. However, the IC₅₀ values of meta-diamides 1, 7, and 9 for GlyRs α1 and α1β were >3 µM (pIC₅₀<5.52) (Fig. 7B, 7C, and 7D, respectively). This indicates that meta-diamides have lower-level antagonist activity than does fipronil.

Fig. 7. Responses of human GlyRs α1, α1-A288G, α1β, and α1-A288Gβ to fipronil and meta-diamides. Concentration–response curves of the inhibition of glycine-induced currents by fipronil (A) and meta-diamides 1 (B), 7 (C), and 9 (D). Experiments were conducted with human GlyRs α1 (open circle), α1-A288G (open square), α1β (solid circle), and α1-A288Gβ (solid square) in the presence of GABA. The data are expressed as the percent of inhibition of the GlyR response relative to the EC₈₀ values of glycine in the absence of each test compound. The vertical bars represent SEM for three independent experiments, which were conducted in duplicate.

Then, the effects of A288G mutation in GlyR α1 on the antagonist activities of fipronil and meta-diamides were studied. The fipronil IC₅₀ values for GlyRs α1-A288G and α1-A288Gβ were 0.353 µM (pIC₅₀=6.45±0.06) and 2.67 µM (pIC₅₀=5.57±0.06), respectively (Fig. 7A). This demonstrates that A288G mutation is not essential for the sensitivity of GlyRs α1 and α1β to fipronil.

In contrast, the IC₅₀ values of meta-diamides, 1, 7, and 9 for α1-A288G were 0.178 µM (pIC₅₀=6.75±0.08; Fig. 7B), 0.149 µM (pIC₅₀=6.80±0.09; Fig. 7C), and 0.178 µM (pIC₅₀=6.75±0.08; Fig. 7D), respectively. This indicates that A288G mutation is important for the sensitivity of GlyR α1 to meta-diamides. The A288G mutation-induced increase in the inhibitory activity of meta-diamides against GlyR α1-A288Gβ was not as high as that against GlyR α1-A288G. The inhibition of GlyR α1-A288Gβ was approximately 40% at 3 µM (Fig. 7B), and the IC₅₀ values of meta-diamides 7 and 9 for α1-A288Gβ were 0.706 µM (pIC₅₀=6.15±0.10; Fig. 7C) and 3.03 µM (pIC₅₀=5.52±0.07; Fig. 7D), respectively.

Discussion

Novel meta-diamide insecticides, which are insect RDL GABA receptor antagonists, have been suggested to act at different sites as compared with first- and second-generation noncompetitive antagonists.^22,23) The second-generation noncompetitive antagonist fipronil has different target site specificities for GABA receptors in insects and mammals as compared with the first-generation noncompetitive antagonists dieldrin, α-endosulfan, and lindane.^16,17) However, the target site specificities of meta-diamide insecticides for GABA receptors in insects and mammals have never been studied.

The IC₅₀ values of meta-diamides 1, 7, and 9 for human GABA_AR α1β2γ2S and mammalian GABA_AR α1β3γ2S were >3 µM (Figs. 4 and 5, respectively), whereas those for the Spodoptera litura RDL GABA receptor were <10 nM.²²⁾ This indicates that the antagonistic activities of meta-diamides against insect RDL GABA receptors are considerably higher than those against human GABA_AR α1β2γ2S and mammalian GABA_AR α1β3γ2S. The fipronil IC₅₀ values for human GABA_AR α1β2γ2S, mammalian GABA_AR α1β3γ2S, and the S. litura RDL GABA receptor were 0.289 µM (Fig. 4), 0.417 µM (Fig. 5), and 0.105 µM,²²⁾ respectively, so it is likely that the target site specificities of meta-amides are higher than those of fipronil. Furthermore, the IC₅₀ values of meta-diamides 1, 7, and 9 for human GlyR α1β were >3 µM (Fig. 7B, 7C, and 7D, respectively), although the fipronil IC₅₀ value for human GlyR α1β was 0.815 µM (Fig. 7A).

A study that used a series of G336 mutant Drosophila RDL GABA receptors demonstrated that Gly at position 336 was essential for high sensitivity to meta-diamides.²²⁾ Gly at equivalent positions in GABA_AR and GlyR subunits appeared to be important for an increased sensitivity of GABA_ARs and GlyRs to meta-diamides. G336 in the M3 region of the Drosophila RDL GABA receptor is equivalent to A288 in human GlyR α1, which is important for insensitivity to ivermectin.²⁷⁾ Thus, we studied the effects of A288G mutation in GlyR α1.

The IC₅₀ values of meta-diamides 1, 7, and 9 for human GlyR α1 were >3 µM (Fig. 7B, 7C, and 7D). In contrast, the IC₅₀ values of meta-diamides, 1, 7, and 9 for human GlyR α1-A288G were 0.178 µM (Fig. 7B), 0.149 µM (Fig. 7C), and 0.178 µM (Fig. 7D), respectively, which suggests that A288G mutation in GlyR α1 is important for sensitivity to meta-diamides 1, 7, and 9.

If Gly at the equivalent position in GABA_AR subunits is essential for high sensitivity to meta-diamides, it is expected that the target site specificity of meta-diamides should differ in insect and mammalian GABA receptors, which is the case with ivermectin.²⁷⁾ Human GABA_ARs are comprised of 19 genes, including 16 subunits combined as GABA_A (α1–6, β1–3, γ1–3, δ, ρ, ε, and π) and 3 ρ subunits,³¹⁾ but GABA_AR π is the only subunit that contains Gly at an equivalent position.²⁷⁾ The co-expression of α1, β2, and π subunits produced receptors with low sensitivity to ivermectin, probably because a single π was incorporated per receptor.²⁷⁾ Although further studies are required to understand the specificity of meta-diamides, our study provides new insights into the mechanisms that underlie the selectivity of meta-diamides in insect RDL GABA receptors and human GABA and glycine receptors.

References

• 1) S. D. Buckingham and D. B. Sattelle: “GABA Receptors of Insects,” ed. by L. I. Gilbert and S. S. Gill, Elsevier Academic Press, London, pp. 29–44, 2010.
• 2) Y. Ozoe: Adv. Insect. Physiol. 44, 211–286 (2013).
• 3) P. Charnet, C. Labarca, R. J. Leonard, N. J. Vogelaar, L. Czyzyk, A. Gouin, N. Davidson and H. A. Lester: Neuron 4, 87–95 (1990).
• 4) L. Chen, K. A. Durkin and J. E. Casida: Proc. Natl. Acad. Sci. U.S.A. 103, 5185–5190 (2006).
• 5) S. Ci, T. Ren and Z. Su: J. Mol. Model. 13, 457–464 (2007).
• 6) K. Hisano, F. Ozoe, J. Huang, X. Kong and Y. Ozoe: Invert. Neurosci. 7, 39–46 (2007).
• 7) J. E. Casida and M. Tomizawa: J. Pestic. Sci. 33, 4–8 (2008).
• 8) R. H. ffrench-Constant, T. A. Rocheleau, J. C. Steichen and A. E. Chalmers: Nature 363, 449–451 (1993).
• 9) M. Miyazaki, F. Matsumura and R. W. Beeman: Comp. Biochem. Physiol. 111B, 399–406 (1995).
• 10) N. Anthony, T. Unruh, D. Ganser and R. ffrench-Constant: Mol. Gen. Genet. 260, 165–175 (1998).
• 11) S. A. Thompson, M. Z. Smith, P. B. Wingrove, P. J. Whiting and K. A. Wafford: Br. J. Pharmacol. 127, 1349–1358 (1999).
• 12) C. Bass, I. Schroeder, A. Turberg, L. M. Field and M. S. Williamson: Pest Manag. Sci. 60, 1157–1162 (2004).
• 13) W. Du, T. S. Awolola, P. Howell, L. L. Koekemoer, B. D. Brooke, M. Q. Benedict, M. Coetzee and L. Zheng: Insect Mol. Biol. 14, 179–183 (2005).
• 14) J. R. Gao, T. Kozaki, C. A. Leichter, F. D. Rinkevich, T. Shono and J. G. Scott: Pestic. Biochem. Physiol. 88, 66–70 (2007).
• 15) T. Nakao, A. Naoi, M. Hama, N. Kawahara and K. Hirase: J. Econ. Entomol. 105, 1781–1788 (2012).
• 16) D. Hainzl, L. M. Cole and J. E. Casida: Chem. Res. Toxicol. 11, 1529–1535 (1998).
• 17) X. Zhao, V. L. Salgado, J. Z. Yeh and T. Narahashi: J. Pharmacol. Exp. Ther. 306, 914–924 (2003).
• 18) T. Nakao, A. Naoi, N. Kawahara and K. Hirase: Pestic. Biochem. Physiol. 97, 262–266 (2010).
• 19) T. Nakao, M. Hama, N. Kawahara and K. Hirase: J. Pestic. Sci. 37, 37–44 (2012).
• 20) T. Nakao and K. Hirase: J. Pestic. Sci. 38, 157–160 (2013).
• 21) T. Nakao, A. Kawase, A. Kinoshita, R. Abe, M. Hama, N. Kawahara and K. Hirase: J. Econ. Entomol. 104, 646–652 (2011).
• 22) T. Nakao, S. Banba, M. Nomura and K. Hirase: Insect Biochem. Mol. Biol. 43, 366–375 (2013).
• 23) T. Nakao and K. Hirase: J. Pestic. Sci. 38, 123–128 (2013).
• 24) P. A. Davies, E. B. Hoffmann, H. J. Carlisle, R. F. Tyndale and T. G. Hales: Neuropharmacology 39, 611–620 (2009).
• 25) R. W. Olsen and W. Sieghart: Neuropharmacology 56, 141–148 (2009).
• 26) G. D. Li, D. C. Chiara, G. W. Sawyer, S. S. Husain, R. W. Olsen and J. B. Cohen: J. Neurosci. 26, 11599–11605 (2006).
• 27) T. Lynagh and J. W. Lynch: Int. J. Parasitol. 40, 1477–1481 (2010).
• 28) K. J. Gingrich and P. M. Burkat: J. Physiol. 506, 609–625 (1998).
• 29) C. N. Connolly, B. J. Krishek, B. J. McDonald, T. G. Smart and S. J. Moss: J. Biol. Chem. 271, 89–96 (1996).
• 30) A. A. Jensen: Eur. J. Pharmacol. 521, 39–42 (2005).
• 31) R. W. Olsen and W. Sieghart: Pharmacol. Rev. 60, 243–260 (2008).