(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .
G3’MTMD3 in the insect GABA receptor subunit, RDL, confers resistance to broflanilide and fluralaner [1]
['Yichi Zhang', 'Key Laboratory Of Integrated Pest Management On Crops In East China', 'Ministry Of Agriculture', 'Nanjing Agricultural University', 'Nanjing', 'People S Republic Of China', 'Qiutang Huang', 'Chengwang Sheng', 'Genyan Liu', 'Key Laboratory For Green Chemical Process Of Ministry Of Education']
Date: 2023-08
Meta-diamides (e.g. broflanilide) and isoxazolines (e.g. fluralaner) are novel insecticides that target the resistant to dieldrin (RDL) subunit of insect γ-aminobutyric acid receptors (GABARs). In this study, we used in silico analysis to identify residues that are critical for the interaction between RDL and these insecticides. Substitution of glycine at the third position (G3’) in the third transmembrane domain (TMD3) with methionine (G3’M TMD3 ), which is present in vertebrate GABARs, had the strongest effect on fluralaner binding. This was confirmed by expression of RDL from the rice stem borer, Chilo suppressalis (CsRDL) in oocytes of the African clawed frog, Xenopus laevis, where the G3’M TMD3 mutation almost abolished the antagonistic action of fluralaner. Subsequently, G3’M TMD3 was introduced into the Rdl gene of the fruit fly, Drosophila melanogaster, using the CRISPR/Cas9 system. Larvae of heterozygous lines bearing G3’M TMD3 did not show significant resistance to avermectin, fipronil, broflanilide, and fluralaner. However, larvae homozygous for G3’M TMD3 were highly resistant to broflanilide and fluralaner whilst still being sensitive to fipronil and avermectin. Also, homozygous lines showed severely impaired locomotivity and did not survive to the pupal stage, indicating a significant fitness cost associated with G3’M TMD3 . Moreover, the M3’G TMD3 mutation in the mouse Mus musculus α1β2 GABAR increased sensitivity to fluralaner. Taken together, these results provide convincing in vitro and in vivo evidence for both broflanilide and fluralaner acting on the same amino acid site, as well as insights into potential mechanisms leading to target-site resistance to these insecticides. In addition, our findings could guide further modification of isoxazolines to achieve higher selectivity for the control of insect pests with minimal effects on mammals.
Meta-diamides (e.g. broflanilide) and isoxazolines (e.g. fluralaner) are members of group 30 of compounds according to the Insecticide Resistance Action Committee, and are defined as γ-aminobutyric acid (GABA)-gated chloride channel allosteric modulators. However, their mode of action on the GABAR remains to be fully elucidated. In the current study, we provide in silico, in vitro and in vivo evidence that G3’ TMD3 , the glycine residue at the third position (G3’) in the third transmembrane domain (TMD3) of the insect GABAR subunit, resistant to dieldrin (RDL), is critical for the action of these insecticides. Furthermore, the homozygous but not heterozygous lines of the fruit fly Drosophila melanogaster bearing G3’M TMD3 showed high resistance to broflanilide and fluralaner but were unable to survive to the pupal stage. These findings enhance our understanding of resistance-related and homozygous-lethal gene mutations, and may prove useful for the future synthesis of related insecticides to gain long-term effectiveness and higher selectivity.
In the present study, differences between arthropod RDL and mammalian GABAR subunits (e.g. α, β and γ) were explored to highlight and reinforce residues in the TMDs as being important for the interaction between GABARs and meta-diamides (e.g. broflanilide) or isoxazolines (e.g. fluralaner). Also, D. melanogaster, the classical insect model organism, was edited using the CRISPR/Cas9 system to determine whether mutations at G3’ TMD3 in RDL in vivo lead to resistance to broflanilide and fluralaner. Knowledge gained from this study can guide further modification of novel insecticides to achieve highly selective toxicity to insect pests with minimal effect on mammals, as well as provide insights into a potential mechanism underlying resistance to meta-diamides and isoxazolines in the field.
The insect γ-aminobutyric acid receptor (GABAR) subunit, RDL (resistant to dieldrin), is the molecular target for various types of insecticides, such as cyclodienes, phenylpyrazoles, macrocyclic lactones, meta-diamides and isoxazolines [ 3 ] ( S1 Fig ). Meta-diamides and isoxazolines are novel classes of compounds, which have been classified into group 30 by the Insecticide Resistance Action Committee, and are used to protect crops and animals from insect pests [ 4 ]. Despite acting on the same molecular target, meta-diamides and isoxazolines do not show cross-resistance with cyclodienes or phenylpyrazoles [ 3 ]. Previous studies demonstrated that mutations at A302 (the fruit fly, Drosophila melanogaster, numbering, otherwise referred to as A2’ TMD2 ) in the second transmembrane domain (TMD2) of RDL underlie resistance to fipronil and dieldrin [ 3 , 5 , 6 ]. In contrast, the potency of meta-diamides and isoxazolines is unaffected by mutations at A2’ TMD2 , suggesting that they act on a site different to that of cyclodienes or phenylpyrazoles [ 7 – 10 ]. In line with this, G3’M TMD3 , a mutation in TMD3 of heterologously expressed RDL, substantially reduced the potency of meta-diamides and isoxazolines ( S2 Fig ) [ 10 – 13 ]. Attempts to generate diamondback moth Plutella xylostella that were resistant to broflanilide were unsuccessful after ten generations of selection [ 14 ]. Thus, potential mechanisms underlying resistance to meta-diamides, including mutations in RDL, remain to be identified in vivo.
High and efficient agricultural activity is required to meet the demand of an ever-growing human population. However, agricultural productivity is hampered by insect pests, which can lead to 20%-30% loss of crops [ 1 ]. To date, insecticides are still the most widely used tool for controlling insect pests. However, crop protection efforts are undermined by the development of resistance to insecticides. Exploring the molecular targets of insecticides forms an important basis for understanding mechanisms leading to resistance [ 2 ].
A crawling assay was used to investigate the potential fitness cost caused by homozygous mutations at the RDL G3’ TMD3 residue. As shown in Fig 4 and S1 Video , D. melanogaster larvae homozygous for G3’M TMD3 , G3’Q TMD3 or G3’S TMD3 showed an average speed of 0.52, 0.85 and 1.13 mm/min, respectively ( S10 Table ). In contrast, the wild-type w 1118 larvae were faster at 2.92 mm/min ( Fig 4B and 4C ), indicating a significant mutation-associated decrease in locomotion.
Because the survival rate of the homozygous G3’M TMD3 larvae was higher than 90% at 24 h after hatching ( Fig 4A ), the sensitivity of newly-hatched larvae towards fluralaner and broflanilide were measured. As shown in Table 1 , LC 50 values of fluralaner and broflanilide for the w 1118 larvae were 1.08 and 0.47 mg/kg, respectively. In contrast, the mortality of the homozygous G3’M TMD3 larvae was 35.71% and 39.54% to fluralaner and broflanilide at 1000 mg/kg, respectively, which indicated that they have developed > 900-fold resistance to these two insecticides compared to the w 1118 larvae. Such high levels of cross-resistance suggested that the G3’ TMD3 residue plays a critical function in the interaction between RDL and fluralaner or broflanilide in vivo. In addition, it is worth noting that the homozygous G3’M TMD3 larvae also showed resistance (6.80-fold) to avermectin but not to fipronil.
(A) Temporal characteristics of lethality in homozygous mutants. Data were analyzed using the Log-rank test for trend and the Mantel-Cox test (ns, not significant; *, P < 0.05; ***, P < 0.001). (B) Crawling speed of homozygous mutants. Error bars indicate the standard error of the mean (SE). Significant difference was determined by Student’s t-test (***, P < 0.001). (C) Representative motion path of homozygous mutants in 2 minutes ( S1 Video ).
After hatching of the three homozygous G3’M TMD3 strains of D. melanogaster, a significant difference was observed in their survival rate ( Fig 4A and S9 Table ). Approximate 90% of their larvae survived for one day and > 70% for three days. However, only 30% and 20% of homozygous G3’Q TMD3 and G3’S TMD3 larvae, respectively, survived for three days. All homozygous mutant flies stayed at the larval stage until death within 7 days, whilst the control strain, w 1118 , progressed to pupation.
(A) Fluorescence detection of embryos. Heterozygous mutant or TM3 Sb GFP/ TM3 Sb GFP embryos showing a GFP signal. Homozygous mutant embryos labeled with white triangles showed no GFP signal. BF, bright field. (B) The proportion and hatching rate of embryos bearing homozygous mutations. For each strain, genotypes of 100 embryos were identified. 70.83% (17 of 24), 70.37% (19 of 27) and 76.92% (20 of 26) embryos hatched that were homozygous for G3’M TMD3 , G3’S TMD3 and G3’Q TMD3 , respectively. (C) Genotypes of the homozygous mutant larvae were confirmed by sequencing of genomic DNA. The codon of the G3’ TMD3 residue is boxed and the synonymous mutations are labeled with black equilateral triangles.
In order to further investigate the substitution-associated lethality of homozygous mutations at G3’ TMD3 in RDL, we crossed the heterozygous mutants (G3’M TMD3 /TM2 Ubx 130 , G3’S TMD3 /TM2 Ubx 130 or G3’Q TMD3 /TM2 Ubx 130 ) with the w 1118 ; dIRE1 Δ /TM3 Sb GFP strain to obtain mutants carrying a GFP-labeled balancer. Analysis of embryos showed that the GFP and homozygous mutations at G3’ TMD3 in RDL did not hinder embryogenesis of D. melanogaster ( Fig 3A ). The proportion of embryos homozygous for G3’M TMD3 , G3’S TMD3 or G3’Q TMD3 was 24%, 27% and 26%, respectively ( Fig 3B ). Furthermore, their corresponding hatching rate was 70.83%, 70.37% and 76.92%, respectively, and that of w 1118 was 73.00%. These results demonstrated that the homozygous embryos of G3’M TMD3 , G3’S TMD3 or G3’Q TMD3 could develop and hatch normally.
Heterozygous D. melanogaster adults bearing a mutation at G3’ TMD3 (G3’M TMD3 , G3’Q TMD3 or G3’S TMD3 ) were sensitive to fipronil, fluralaner, broflanilide and avermectin without significant alteration in resistance ratio ( S8 Table ). The findings in the current and previous studies indicated that mutations at G3’ TMD3 reduced potency of these insecticides when using in vitro and in silico approaches [ 10 , 12 , 15 ].
In the complementation experiment, all progeny produced by these crosses carried the parental balancer chromosomes, thus no complementation was apparently viable. This result confirmed that the observed lethality was linked to the corresponding genomic region containing Rdl, presumably to G3’M/S/Q TMD3 . Therefore, we concluded that the substitutions at G3’ TMD3 of the RDL subunit might have a severe impact on channel function affecting viability.
Because G3’Q TMD3 homozygous lines could not be generated with the w 1118 strain, the nos.Cas9 strain with a different genetic background in the X chromosome was used for establishing the G3’Q TMD3 homozygous lines injected with specific gRNAs/donor plasmid mixture ( S6 Fig ), which was defined as G 0nos . The G 0nos adults were individually crossed with the nos.Cas9 flies, and the G 1nos progeny were identified by genomic DNA sequencing. The G3’Q TMD3 mutation was detected in 1 out of 18 G 1nos lines, indicating that the homology-directed repaired (HDR) allele (G3’Q TMD3 ) was present in the G 0nos strain. Unfortunately, similar to G3’M TMD3 and G3’S TMD3 , G3’Q TMD3 was homozygous lethal. Therefore, a heterozygous G3’Q TMD3 strain was used for further study ( Fig 2 ).
(A) Diagram of the genome editing strategy. Black triangles indicated the gRNA-targeted sites. A 2114 bp homologous region with a modified codon corresponding to the G3’ TMD3 residue was cloned into the donor plasmid, and nine synonymous mutations indicated by vertical lines were designed around the G3’ TMD3 residue to prevent repeated editing. (B) Genotypes of the heterozygous mutant strains were confirmed by sequencing of genomic DNA. The corresponding codon of the G3’ TMD3 residue is boxed, and the synonymous mutations are indicated by black equilateral triangles. (C) Balancer-associated phenotype of heterozygous mutant strains. Additional bristles on the haltere are indicated by a black triangle.
The CRISPR/Cas9 genome editing system was used to introduce substitutions at G3’ TMD3 (G335) in DmRDL ( Figs 2A and S6 ). For the G3’M TMD3 or G3’S TMD3 , a mixture including Cas9 mRNA, donor plasmid and gRNAs was injected into the embryos of the w 1118 strain (defined as G 0 generation). In the G 0 generation, 2 out of 12 or 4 out of 25 individuals carrying the G3’M TMD3 or G3’S TMD3 allele, respectively, were identified. Subsequently, the G 1 individuals from these positive lines were crossed with the balancer strain (w 1118 ; TM2 Ubx 130 /TM6B Tb 1 ) to retain the mutant allele in the G 2 generation ( S7 Fig ). After self-crossing of G 2 generation, G3’M TMD3 or G3’S TMD3 homozygous lines were not observed in the progeny of the G 3 generation, indicating that either mutation at G3’ TMD3 may cause homozygous-lethality in D. melanogaster adults. Therefore, strains carrying the TM2 Ubx 130 balancer chromosome, which are also heterozygous for G3’M TMD3 or G3’S TMD3 , were generated and verified by genomic DNA sequencing ( Fig 2 ).
To examine the contribution of G3’ TMD3 of mammalian GABARs towards fluralaner insensitivity, the M3’G TMD3 was introduced into the mouse Mus musculus GABAR β2 subunit (Mmβ2). There was no detectable response to GABA in X. laevis oocytes injected with Mmα1 or Mmβ2 alone. However, co-expression of Mmα1 and Mmβ2 subunits generated a functional heteromeric GABA-gated channel (Mmα1β2) ( Fig 1 and S6 Table ). GABA potency on the Mmα1β2-M3’G TMD3 channel was enhanced as shown by a significantly lower EC 50 value compared to that of the wild-type ( Fig 1G and S6 Table ). Fluralaner showed decreased potency on Mmβ2-M3’G TMD3 at concentrations of 10 −8 and 10 −9 M as shown by significantly less inhibition of GABA-induced currents ( Fig 1H and S7 Table ).
To verify whether G3’M TMD3 can potentially give rise to resistance to fluralaner in different arthropod species, the mutation was introduced into RDL from Hymenoptera (Apis mellifera), Hemiptera (Laodelphax striatellus), Arachnoidea (Tetranychus urticae), Lepidoptera (C. suppressalis), and Diptera (D. melanogaster). The mutant RDL subunits were individually expressed in X. laevis oocytes. In each case, G3’M TMD3 reduced the potency of GABA by 8–34-fold and showed significantly different EC 50 values compared with that of the corresponding wild-type RDL ( Fig 1E and S5 Table ). In addition, the G3’M TMD3 in RDL of each species reduced the potency of fluralaner as shown by significantly increased IC 50 values ( Fig 1F and S5 Table ).
Both fipronil and avermectin strongly inhibited the GABA-induced current in the homomeric wild-type CsRDL channel with an IC 50 of 10.02 and 69.90 nM, respectively ( Fig 1 and S4 Table ). However, the G3’M TMD3 CsRDL channel was less sensitive than the wild-type to fipronil or avermectin as indicated by significantly greater IC 50 values. In particular, avermectin at 10 −5 M only inhibited 38.06% of the GABA-induced response in CsRDL with the G3’M TMD3 as opposed to almost abolishing the GABA response in wild-type channels ( Fig 1D ).
( A ) and ( B ) Concentration-response curves of GABA (A) and inhibition of GABA-induced currents by fluralaner (B) from wild-type or mutant RDL receptors. ( C ) and ( D ) Inhibition of GABA-induced currents by fipronil (C) and avermectin (D) in wild-type or G3’M TMD3 RDL. ( E ) and ( F ) Effect of the G3’M TMD3 mutation in RDL of different arthropod species on response to GABA (E) or fluralaner (F). ( G ) and ( H ) Concentration-response curves of GABA (G) and inhibition of GABA-induced currents by fluralaner (H) from heteromeric Mmα1β2 or mutant Mmα1β2-M3’G TMD3 channels. Error bars indicated the standard error of the mean (SE). Significant difference was determined by Student’s t-test (ns, not significant; **, P < 0.01).
Twelve homomeric mutant C. suppressalis (CsRDL) subunits were expressed in X. laevis oocytes ( S2 Table ). Inward currents upon GABA application were not detected for one mutant, I477D. The maximum GABA-induced current (I max ) was notably lower in the mutant channels than that of the wild-type CsRDL channel except for G3’M TMD3 . All mutations except for I258T and M473V decreased the potency of GABA as indicated by a significant increase in EC 50 ( Figs 1A and S5 and S2 Table ). The EC 50 values of G3’M TMD3 and G3’S TMD3 increased 34- and 64-fold, respectively, compared with that of the wild-type CsRDL channel ( Fig 1A and S2 Table ).
Amino acid residues of arthropod RDL and vertebrate GABAR subunits were aligned and the four TMDs were identified ( S3 and S4 Figs and S1 Table and S1 Note ). The effect of mutant residues in the TMDs on binding of fluralaner to a three-dimensional homology model of RDL from the rice stem borer, Chilo suppressalis, was assessed. Altering several residues in RDL to the equivalent amino acid present in vertebrate GABARs increased the binding energy of fluralaner ( S1 Table ). Twelve of these mutations were selected for functional expression in oocytes of the African clawed frog, Xenopus laevis, in order to determine if they play a role in the interaction between RDL and fluralaner ( S1 Table ).
Discussion
To date, insects have developed varying levels of resistance to traditional GABAR-targeting insecticides such as cyclodienes, phenylpyrazoles and macrocyclic lactones [5,16–18]. Previous studies investigating mechanisms of resistance have provided insights into their binding site(s). For example, A2’ TMD2 of RDL is recognized as the common binding site of cyclodienes and phenylpyrazoles. A mutation at A2’ TMD2 is the critical basis for resistance to dieldrin (A2’S TMD2 ) or fipronil (A2’N TMD2 ) [19–23]. However, RDL with resistance-related mutations of A2’S/G/N TMD2 is still inhibited by fluralaner [24–28].
Meta-diamides (e.g. broflanilide) and isoxazolines (e.g. fluralaner) are defined as non-competitive antagonist (NCA)-II compounds that share a coupled action site in proximity to the interface of TMD1/TMD3 in RDL [3,29]. In the current study, the amino acid sequences of arthropod and mammalian GABARs were aligned in order to explore the potential binding site of fluralaner. A similar research approach was performed on the bumblebee Bombus impatiens BiNa v 1 sodium channel, which identified additional amino acid residues that underlie the sensitivity of B. impatiens to pyrethroids as well as selective resistance to tau-fluvalinate [30]. With mammals being relatively insensitive to fluralaner [24,31], we identified 35 amino acid residues in the TMDs that differ between arthropod and mammalian GABAR subunits (S1 Table). Heterologous expression studies showed that the G3’M TMD3 mutation led to the greatest decrease in fluralaner potency (S3 Table) reinforcing this residue as being crucial for isoxazoline activity, which is consistent with previous studies [10,12]. Meanwhile, G3’M TMD3 also considerably decreased the antagonist actions of avermectin (S4 Table). Avermectin is a macrocyclic lactone and is thought to act primarily on the glutamate-gated chloride channel (GluCl) with GABARs being a secondary target [32]. In line with this, both G323D in TuGluCl and G315E in PxGluCl (equivalent position to RDL G3’ TMD3 ) are associated with resistance to avermectin [33,34]. Although the G3’M mutation might allosterically affect GABA binding to the orthosteric site leading to the significantly different EC 50 values compared to that of wild-type (S2 Table), the concentration-response curve indicated that the channel was still able to function in response to GABA (Fig 1A and 1B). Other studies found that fluralaner [11,25] and avermectin inhibited [3H]fluralaner binding on M. domestica membranes [28]. Fluralaner binding was also inhibited when the G3’M TMD3 was present in the homomeric RDL channels of the common cutworm, Spodoptera litura, and D. melanogaster [10,11].
Desmethyl-broflanilide, which has a common genesis with broflanilide, has a site of action near G3’ TMD3 in the DmRDL subunit [10,35]. Nakao et al. (2013) reported that the volume of amino acid at G3’ TMD3 is a factor that determines the inhibitory activity of desmethyl-broflanilide [10]. Therefore, three mutations (G3’M/S/Q TMD3 ), which eliminate the insecticidal sensitivity with only a minor change in activity of GABAR, were selected for further assay in vivo in this study.
CRISPR/Cas9 is a powerful tool for generating specific mutations to validate gene function, e.g. insecticide resistance in D. melanogaster [36–41]. So far, several studies adopting this “reverse genetic” approach have successfully demonstrated the mode of action of various types of insecticides by modifying target receptors with compelling association between mutations and phenotypes [42–46]. The mutation of G3’M TMD3 or G3’S TMD3 was successfully introduced into w1118 DmRDL and heterozygote lines were established. Meanwhile, the G3’Q TMD3 was successfully introduced into the nos.Cas9 strain. Finally, nos.Cas9 bearing G3’Q TMD3 in the X chromosome was replaced with the w1118 allele during crossing ensuring that the three mutant (G3’M/S/orQ TMD3 ) strains were generated with a consistent genetic background. Unfortunately, introduction of G3’M/S/orQ TMD3 resulted in lethality making it impossible to obtain the homozygous mutant strains. This may be a reason for the failure of selecting broflanilide-resistant strains of P. xylostella [14]. Similarly, occurrence of homozygous-lethal effects have hampered investigations into the modes of action of particular insecticides using the CRISPR/Cas9 strategy [45,47,48]. Therefore, the G3’M TMD3 , G3’S TMD3 and G3’Q TMD3 were maintained in heterozygous strains, and the direct linkage between homozygous lethality and target-site mutations was proven by a complementation experiment [49].
The GFP-labeled allele can be used as a useful tool to identify the homozygote and heterozygote mutants at the embryo stage of D. melanogaster, especially when the phenotypes of the gene mutations are unknown [50,51]. Thus, heterozygous mutant strains with a GFP-labeled balancer allele enabled accurate selection of homozygous G3’ TMD3 mutants by fluorescence detection. Embryos bearing any of the three homozygous mutations at RDL G3’ TMD3 were able to hatch where 25% of embryos from each heterozygous strain were expected to carry homozygous mutations [52]. In line with this, the proportion of homozygous G3’M TMD3 , G3’S TMD3 and G3’Q TMD3 embryos was 24%-27% (Fig 3B). It is worth noting that the temporal characteristics of lethality varied among the three mutants during the first three days after hatching (Fig 4A).
In this study, heterozygous lines bearing mutations at G3’ TMD3 did not show significant resistance to broflanilide or fluralaner (S8 Table). This is in line with a previous study using electrophysiology, where expression of heterozygous G3’ TMD3 mutant RDL in D. melanogaster Mel-2 cells did not confer resistance to demethyl-broflanilide compared with the wild-type RDL alone [15]. Therefore, these in vivo and in vitro studies indicate that heterozygous mutations at RDL G3’ TMD3 do not confer resistance to meta-diamides and isoxazolines. Drosophila melanogaster homozygous for G3’M TMD3 were selected at the larval stage for bioassays due to their low mortality during the first three days after hatching. These larvae were highly resistant to broflanilide and fluralaner with LC 50 > 1000 mg/kg, providing convincing evidence in vivo that meta-diamides and isoxazolines share the same mode of action and directly interact with the RDL subunit. It is also worth noting that larvae homozygous for G3’M TMD3 showed low-level resistance to avermectin, indicating that avermectin might share an overlapping but weak binding mode with fluralaner and broflanilide on the GABAR. However, the toxicity of fipronil was not affected by the homozygous G3’M TMD3 mutation, which indicated a different binding site, in accord with previous studies in vivo and in vitro [6,11,22,23].
It has been previously reported that the knock-down of RDL can affect the locomotivity of D. melanogaster [53]. In this study, our results showed that D. melanogaster larvae homozygous for G3’M/S/orQ TMD3 displayed a significantly reduced crawling speed compared to wild-type w1118 (Fig 4B and 4C) suggesting that the mutation results in a potential fitness cost. Physical fitness costs caused by point mutations were also reported in other species of insect pests [43,54–58]. For example, D. melanogaster bearing the homozygous R81T in the nicotinic acetylcholine receptor β1 subunit showed an increased tolerance to neonicotinoid insecticides with a dramatic decrease in fertility, locomotivity and longevity [56].
[END]
---
[1] Url:
https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1010814
Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.
via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/
