Mutations M918T, M918L, and L1014F impair pyrethroid binding and alter gating of the Megalurothrips usitatus sodium channel

Abstract

BACKGROUND

Mutations M918T/L and L1014F in the voltage‐gated sodium channel (VGSC) are major causes of pyrethroid resistance in insects, but their functional effects in thrips remain unclear. We used Megalurothrips usitatus sodium channel (MuNa_v1‐1) expression system to determine how these mutations affect channel gating and pyrethroid sensitivity.

RESULTS

Electrophysiological analysis revealed that mutations shifted the voltage dependence of activation and inactivation. Notably, the L1014F mutation shifted the voltage dependence of activation by 8.32 mV in the depolarizing direction compared with wild‐type MuNa_v1‐1, whereas M918L resulted in a hyperpolarizing shift of 3.15 mV. Similarly, compared with wild‐type MuNa_v1‐1, L1014F and the double mutant M918T/L1014F shifted the voltage dependence of fast inactivation in the depolarizing direction by 5.52 and 6.23 mV, respectively. All mutants drastically reduced channel sensitivity to bifenthrin, permethrin, deltamethrin, and λ‐cyhalothrin. The M918L mutant showed the greatest reduction in sensitivity (up to 40‐fold), whereas the double mutant exhibited a strong synergistic resistance effect. Computational modeling indicated that the mutation directly impairs the binding affinity through the loss of hydrophobic contacts, with M918T severely disrupting hydrophobic interactions and L1014F inducing steric hindrance.

CONCLUSION

This study provides direct evidence that M918T/L and L1014F mediate pyrethroid resistance in M. usitatus and other thrips species by altering VGSC gating properties and reducing pyrethroid sensitivity. These findings advance the functional understanding of kdr and super‐kdr mutations in thrips and provide a critical basis for targeted resistance monitoring and the rational design of insecticide management strategies. © 2026 Society of Chemical Industry.

M918T/L and L1014F mutations alter MuNa_v gating, reduce pyrethroid sensitivity, and disrupt binding, supporting genotype‐based resistance monitoring and targeted management.

1. INTRODUCTION

Agricultural practices have long relied on pesticides to control crop diseases and pests, ensuring stable grain production. However, escalating pesticide use has driven the evolution of resistance in target pests, posing a major challenge to global integrated pest management. ¹ , ² Pyrethroids, the second most widely used class of commercial insecticides globally, are extensively applied against arthropod pests in agricultural and urban settings. ³ , ⁴ , ⁵ They are synthetic analogs of natural pyrethrins derived from Chrysanthemum cinerariifolium. Based on chemical structure and symptomology, pyrethroids are categorized into two types: type I (lacking an α‐cyano group) and type II (containing an α‐cyano group). ⁴ , ⁵ , ⁶ Despite their cost‐effectiveness and broad‐spectrum efficacy, overreliance and improper application have led to their overuse. This practice not only fosters high‐level resistance in pest populations, but also results in a persistent decline in field control efficacy. ⁷

The primary mechanism of action of pyrethroids involves interaction with voltage‐gated sodium channels (VGSCs) in the insect nervous system. By binding to VGSCs, pyrethroids inhibit channel inactivation and deactivation, increase neuronal excitability, and ultimately produce neurotoxic effects. ⁸ , ⁹ VGSCs, which serve as the critical molecular target for pyrethroids, are evolutionarily conserved proteins composed of a pore‐forming α subunit and auxiliary β subunits. The α subunit alone can constitute a functional channel and contains four homologous domains (I–IV), each comprising six transmembrane segments (S1–S6). The β subunits, by contrast, primarily modulate channel expression, voltage dependence, and gating kinetics. ¹⁰ , ¹¹ , ¹² Structural homology modeling has identified two putative pyrethroid‐binding sites on the VGSC: pyrethroid receptor site 1 (PyR1), located at the interface of domains II and III; and pyrethroid receptor site 2 (PyR2), located at the interface of domains I and II. ¹³ , ¹⁴ Between these two regions, there are numerous reported mutation sites, including the kdr and super‐kdr mutations. These mutations often confer resistance to pyrethroids and Dichlorodiphenyltrichloroethane (DDT) by altering the spatial structure of sodium channel and reducing the binding affinity of these insecticides. ¹⁵

The L1014F mutation was first identified as a kdr mutation in the housefly (Musca domestica), and is located on the domain II S6 helix of the sodium channel. This leucine‐to‐phenylalanine substitution reduces the volume of the original binding pocket, thereby hindering the binding of pyrethroid insecticides. ¹⁶ , ¹⁷ , ¹⁸ A super‐kdr mutation (M918T + L1014F) has also been identified in the housefly, with M918T located on the domain II S4–5 linker of the sodium channel. ¹⁷ The two mutations reside within the PyR1‐ and PyR2‐binding regions, respectively, and frequently occur together, as M918T has been demonstrated in Xenopus oocytes to significantly alter the gating properties of the sodium channel. ¹⁹ , ²⁰ Under ongoing resistance evolution, these two mutation types have diversified into L1014H/S/C/W and M918L/I/V variants in different insect species resistant to pyrethroids. ²¹ , ²² , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ Among these, L1014H/S and M918I have been validated in Xenopus oocyte systems to mediate sodium channel resistance to deltamethrin (DMT) and permethrin (PMT). ²⁵ , ²⁶ , ²⁷ In addition to the L1014F + M918T combination, the mutation E435K + C785R + L1014F has been identified in the German cockroach, Blattella germanica. The double mutations (E435K + L1014F or C785R + L1014F) confer a 100‐fold reduction in sensitivity to DMT, while the triple‐mutation (E435K + C785R + L1014F) results in a 500‐fold reduction. ²¹ , ²⁶ In Apolygus lucorum, the combination M918I + L1014F has been found. This double mutation exhibits a significantly greater reduction in sensitivity to four pyrethroids [PMT, DMT, bifenthrin (BFT), and λ‐cyhalothrin (CHT)] compared with single mutations alone. ²⁷ Therefore, mutations at the L1014 and M918 positions are frequently associated with pyrethroid resistance mediated by the sodium channel.

Mutations at the M918T/L and L1014F sites were also reported in thrips species in a previous study. In Thrips palmi, the double mutation I265T + L1014F was initially identified and reported in field populations between 2020 and 2022, with allele frequencies ranging from 53.33% to 96.67%. This double mutation conferred a 23‐fold resistance to DMT and β‐cypermethrin in T. palmi. ²⁹ In western flower thrips (Frankliniella occidentalis), the T929I + L1014F mutation combination has also been detected, which has been confirmed as the key mutation conferring resistance to pyrethroids. Among these, the combined mutant allelic frequencies (MAF) of T929I and T929V reached 100%, and MAF of L1014F reached 37.29%. ³⁰ , ³¹ In the thelytokous population of Thrips tabaci, mutations of M918L (triploid heterozygous) and M918T + L1014F (diploid heterozygous) have been identified, confirming that super‐kdr mutations conferring pyrethroid resistance have emerged in the evolutionary resistance of T. tabaci populations. ³² , ³³ In Hainan Province, the widespread Megalurothrips usitatus has developed high‐level resistance due to complex field insecticide application, and has even displaced F. occidentalis to become the dominant thrips species in Hainan fields. The primary drivers of this high‐level resistance are target‐site resistance mutations and the upregulation of detoxification metabolic enzymes. ³⁴ Between 2021 and 2023, the resistance ratio of M. usitatus to CHT insecticide continued to rise and remained consistently higher than that of T. palmi. ³⁵ Furthermore, no mutations of the M918L or M918T + L1014F types have been detected in the sodium channel of M. usitatus to date; only the T929I and K1774N mutations have been identified. ³⁶ This raises the question of whether the presence of M918L and M918T + L1014F in the sodium channel of M. usitatus would result in resistance ratios higher than those conferred by T929I + K1774N. Moreover, although other thrips species carry the kdr and super‐kdr mutations L1014F and M918T, their resistance ratios remain lower than those observed in M. usitatus—suggesting that the adaptive cost of resistance evolution may be greater in M. usitatus than in other thrips species. To address this issue, we used the Megalurothrips usitatus sodium channel (MuNa_v1‐1), which shares high homology with thrips populations and has an established heterologous expression system, as a template to introduce the point mutations M918L, M918T, L1014F, and the double mutation M918T + L1014F into MuNa_v1‐1, and heterologously expressed the mutant channels in Xenopus oocytes. Using this system, we examined the regulation effects of these mutations on channel gating properties as well as their pharmacological sensitivity to four pyrethroid insecticides (PMT, DMT, BFT, and CHT) across a range of concentrations. Complementary molecular docking simulations were performed to visualize and analyze the binding interactions between these insecticides and the mutant VGSCs. This study elucidates the resistance mechanisms mediated by the M918T/L, L1014F, and M918T + L1014F mutations in the M. usitatus sodium channel against pyrethroid insecticides. It also serves as a warning regarding the potential risk of pyrethroid resistance in M. usitatus and other thrips populations. Our findings provide strategic direction for future resistance monitoring of M. usitatus and contribute a theoretical basis for the integrated control of M. usitatus under field conditions.

2. MATERIALS AND METHODS

2.1. Cloning of VGSC from M. usitatus

Total RNA was extracted from 30 adult M. usitatus of the pyrethroid‐sensitive HK‐S strain (provided by Prof. Fan Yongmei, Hainan University, China) using the SteadyPure RNA Extraction Kit (Hunan Aikrui Bioengineering Co., Ltd, Hunan, China). Subsequently, complementary DNA (cDNA) was synthesized by reverse transcription with PrimeScript RT Master Mix (Takara Biomedical Technology, Beijing, China). A fragment of VGSC gene was amplified by Reverse Transcription Quantitative Polymerase Chain Reaction (RT‐qPCR) with the following primers: Mu‐F: 5′‐ATGTCCGAGTGCTCGGAGTCG‐3′ and Mu‐R: 5′‐TCAGACATCCGCAAGGCGGGA‐3′. The PCR product was purified using an Omega Purification Kit (OMEGA Biotech, Inc., Norcross, USA) and cloned into the pGH19 vector. The ligation product was transformed into Stbl2 competent cells (Shanghai Weidi Biotechnology Co., Ltd, Shanghai, China). Positive clones were identified and sent for sequencing to Beijing Qingke Biotechnology Co., Ltd (Beijing, China). The obtained sequences were assembled and analyzed using DNASTAR Lasergene v.7.1 software. Finally, plasmid DNA was extracted from cultured positive clones using a Plasmid Mini Kit (OMEGA Biotech, Inc., Norcross, USA).

2.2. Site‐directed mutagenesis of VGSC from M. usitatus

Four mutant versions of the VGSC from M. usitatus (MuNa_v1‐1) were constructed: M918T, M918L, L1014F, and the double mutant M918T/L1014F. The mutation numbers throughout this paper are based on the numbering system of the VGSC in Musca domestica (Q94615_MUSDO). All mutants were generated by site‐directed mutagenesis using the Phusion™ High‐Fidelity PCR Master Mix (Thermo Fisher Scientific, USA), following the manufacturer's protocol. The primer sequences used for mutagenesis are listed in Supporting Information, Table S1. Successful introduction of each mutation was confirmed by Sanger sequencing (Beijing Qingke Biotechnology). The resulting sequences were aligned and verified against the wild‐type MuNa_v1‐1 sequence using DNASTAR Lasergene v.7.1 software.

2.3. Expression of MuNa_v1‐1 in Xenopus oocytes

The wild‐type and mutant pGH19‐MuNa_v1‐1 plasmids were linearized downstream of the insert using NotI. cRNA was synthesized in vitro from the linearized templates using the mMESSAGE mMACHINE T7 Transcription Kit (Thermo Fisher Scientific) according to the manufacturer's instructions. The synthesized cRNA was purified using the MEGAclear Kit (Thermo Fisher Scientific), quantified by spectrophotometry (NanoDrop, Thermo Fisher Scientific), and its integrity was verified by 1% agarose gel electrophoresis. Oocytes were harvested from female Xenopus laevis (Xenopus 1, MI, USA). Oocyte preparation, capped complementary RNA (cRNA) synthesis, and injection procedures followed established methods. ³⁷ To enhance functional expression, cRNA encoding the wild‐type or mutant MuNa_v1‐1 α‐subunit was co‐injected at a 1:1 m ratio with cRNA encoding the auxiliary β‐subunit BgTip E from Blattella germanica, which is known to facilitate robust expression of insect sodium channels in the oocyte system. ³⁸

2.4. Electrophysiological recording and analysis

Sodium currents were recorded from oocytes using a standard two‐electrode voltage clamp configuration at room temperature (20–22 °C) in ND96 bath solution (96 mm NaCl, 2 mm KCl, 1.8 mm CaCl₂, 1 mm MgCl₂, 5 mm HEPES, pH adjusted to 7.5 with NaOH). ²⁷ , ³¹ , ³² Microelectrodes were pulled with a P‐1000 puller (Sutter Instrument) and filled with 3 m KCl and 0.5% agarose (resistance <1 MΩ). Signals were acquired using a Digidata 1550B interface and pCLAMP 10.7 (Axon Instruments). The peak sodium current was maintained between 2.0 and 4.0 μA by adjusting cRNA injection, with leakage currents kept below 0.3 μA. The voltage dependence of activation and fast inactivation was assessed using step‐protocols as previously described. ²⁷ , ³² Briefly, activation started from a holding potential of −120 mV and was measured by stepping from −80 mV to +65 mV (5 mV increments), and steady‐state inactivation was determined using a 100‐ms prepulse series (−120 mV to −10 mV) followed by a test pulse at −10 mV. Resulting conductance and inactivation curves were fitted with standard Boltzmann equations, from which the half‐maximal voltage (V _1/2) and slope factor (k) were derived. ³² All fittings were performed in OriginPro 2024 (OriginLab).

2.5. Determination of tail currents induced by BFT, DMT, PMT, and CHT

Pyrethroids were applied to the recording system as previously described. ³⁹ Effects were measured 10 min after drug application. Pyrethroid‐induced tail currents were recorded during a 5‐ms step depolarization from a holding potential of −120 mV to 0 mV, using a series of 100 pulses (5‐ms interpulse interval). The percentage change in channel conductance under pyrethroid action was calculated using the formula: M = {[I _tail/(E _h − E _Na)]/[I _Na/(E _t − E _Na)]} × 100, where I _tail is the maximum tail current amplitude, and E _h, E _Na and E _t are the holding potential, reversal potential, and test potential, respectively. I _Na is the amplitude of the peak current during depolarization prior to pyrethroid exposure. The dose–response curve was fitted to equation: M = M _max{1+ [EC₅₀/(pyrethroids)] ⁿ }, where pyrethroid concentration and EC₅₀ are the pyrethroid concentration and the concentration producing a 50% maximum effect, respectively, and n is the Hill coefficient. M _max represents the maximum percentage change in VGSCs.

2.6. Homology modeling and system preparation

A homology model of the wild‐type MuNa_v1‐1 was built using the Prime module within the Schrödinger Suite. ⁴⁰ The cryo‐electron microscopy structure of the Periplaneta americana VGSC (PDB ID: 5X0M) was used as the template because of its high sequence identity and resolved insect Na_v channel architecture. The resulting wild‐type model was subjected to standard protein preparation and restrained energy minimization. The three‐dimensional structures of four commercially relevant pyrethroid insecticides, BFT, DMT, PMT, and CHT, were retrieved from the PubChem database and prepared using the LigPrep module. ⁴⁰

2.7. Induced‐fit docking

Based on previously defined pyrethroid receptor sites in insect sodium channels, the putative PyR1 and PyR2 regions were mapped onto the wild‐type MuNav1‐1 homology model. Molecular docking was performed using the Induced‐Fit Docking protocol within Schrödinger to account for potential ligand‐induced side chain rearrangements. Two docking grids (20 Å side length) were generated to constrain ligand placement within the predefined PyR1 and PyR2 regions. The PyR1 grid was centered at the geometric centroid of residues 951, 959, and 1050, whereas the PyR2 grid was centered at the geometric centroid of residues 267, 421, and 1042. To further examine the possible structural effects of resistance‐associated substitutions, in situ residue substitution was performed in selected wild‐type ligand‐bound complexes. Specifically, residues M944 and L1040 in MuNav1‐1, corresponding to canonical M918 and L1014, respectively, were individually replaced with threonine, leucine, or phenylalanine. The resulting local structural changes around the docked ligand were then examined after restrained minimization.

2.8. Statistical analysis

All data analyses were performed using IBM SPSS Statistics 23. All mutants were individually compared pairwise with the MuNa_v1‐1 channel, and significant differences were determined by one‐way analysis of variance with Scheffe's post hoc analysis. P value <0.05 indicated significant differences.

3. RESULTS

3.1. Cloning and mutagenesis of the MuNa_v1‐1 from M. usitatus

A full‐length cDNA encoding the MuNa_v1‐1 was successfully cloned from the pyrethroid‐sensitive HK‐S strain of M. usitatus. Sequence analysis confirmed the identity of the cloned gene as a homolog of the insect VGSC. Subsequently, four mutant versions of MuNa_v1‐1 were constructed via site‐directed mutagenesis: three single mutants (M918T, M918L, and L1014F) and one double mutant (M918T/L1014F). As illustrated in Fig. 1(A), all targeted mutations are located within domain II of the channel protein. Sanger sequencing verified the precise introduction of each intended mutation without any unintended nucleotide changes. Alignment of MuNa_v1‐1 with the domain II regions of three other thrips species sodium channels revealed that the sequences at positions M918 and L1014, where mutations have been reported in other thrips populations, are highly conserved. This ensures that subsequent functional validation can be established in the M. usitatus system (Supporting Information, Fig. S1). Figure 1(B) further illustrates the relative positions of these mutations (M918 and L1014) in relation to the predicted pyrethroid‐binding sites PyR1 and PyR2 within the channel's three‐dimensional topology.

Figure 1.

Mutation sites of the Megalurothrips usitatus sodium channel (MuNa_v1‐1). (A) Schematic topology of the MuNa_v1‐1, highlighting the locations of the three resistance mutations (M918T, M918L, L1014F). (B) Predicted pyrethroid‐binding sites pyrethroid receptor site 1 (PyR1; green, residues in domains II/III) and pyrethroid receptor site 2 (PyR2; yellow, residues in domain I/II). Residues are labeled according to domain (I–IV), segment type (i, S6 inner helix; k, S4–S5 linker; o, S5 outer helix; p, p‐loop). The mutations studied herein are underlined and highlighted in red.

3.2. Mutations of M918 and L1014 alter the gating properties of MuNa_v1‐1

The functional expression of wild‐type and mutant MuNa_v1‐1 channels was successfully achieved in Xenopus oocytes. Electrophysiological characterization revealed that M918L and L1014F significantly altered the voltage dependence of channel activation, whereas L1014F and M918T/L1014F significantly altered the voltage dependence of fast inactivation (Fig. 2 and Table 1; P < 0.05). The half‐maximal activation voltage (V _{1/2,activation}) of wild‐type MuNa_v1‐1 was −17.54 mV. The M918L, M918T, and M918T/L1014F mutations all shifted the half‐activation voltage to hyperpolarized values of −20.69, −18.41, and −20.20 mV, respectively, among which only the M918L mutation caused a significant shift. By contrast, L1014F shifted the half‐activation voltage in the depolarizing direction to −9.22 mV. For steady‐state inactivation, the wild‐type half‐inactivation voltage (V _{1/2,inactivation}) was −51.46 mV. M918T, L1014F, and M918T/L1014F all shifted the half‐inactivation voltage in the depolarizing direction to −50.39, −45.94, and −45.23 mV, respectively, with significant shifts observed for L1014F and M918T/L1014F. Similar to the changes in the voltage dependence of activation, M918L shifted the half‐inactivation voltage in the hyperpolarizing direction to −53.73 mV (Table 1).

Figure 2.

Gating properties of wild‐type and mutant Megalurothrips usitatus sodium channels (MuNa_v1‐1). (A) Representative current trace from a −10 mV test pulse. (B) Current–voltage relationship. (C) Voltage dependence of activation. (D) Voltage dependence of fast inactivation. Data were fitted with the Boltzmann equation (mean ± SEM).

Table 1.

Gating properties of wild‐type and mutant Megalurothrips usitatus sodium channel (MuNa_v1‐1)

Na+ channel type	Activation		Fast inactivation		N
	V 1/2 (mV)	K (mV)	V 1/2 (mV)	K (mV)
MuNav1‐1	−17.54 ± 0.75	5.34 ± 0.16	−51.46 ± 0.39	5.65 ± 0.21	11
M918T	−18.41 ± 0.89	6.57 ± 0.39*	−50.39 ± 0.69	5.06 ± 0.08	12
M918L	−20.69 ± 0.56*	5.02 ± 0.19	−53.73 ± 0.55	5.12 ± 0.18	11
L1014F	−9.22 ± 0.59*	5.60 ± 0.20	−45.94 ± 0.57*	4.78 ± 0.21*	11
M918T + L1014F	−20.20 ± 0.68	4.47 ± 0.19	−45.23 ± 0.68*	5.04 ± 0.07	11

Note: The voltage dependence of conductance and fast inactivation data was fitted using a two‐state Boltzmann equation to determine the maximum half of the conductance and fast inactivation voltages (V _1/2) and slope (K). Each value represents the mean ± SEM. N, number of oocytes tested.

^*Significant difference compared with the MuNa_v1‐1 channel using one‐way analysis of variance with Scheffe's post hoc analysis (P < 0.05).

3.3. Mutations of M918 and L1014 confer resistance to pyrethroids

The effects of type I (BFT, PMT) and type II (DMT, CHT) pyrethroids were assessed by recording the slowly decaying tail currents they induce in Xenopus oocytes expressing wild‐type or mutant MuNa_v1‐1 channels. ³⁹ The modification of VGSCs was evaluated by measuring the amplitude of pyrethroid‐induced tail currents. As shown in Fig. 3(A)–(D), all four insecticides at 1 μm elicited robust tail currents in wild‐type channels. Specifically, the VGSC tail current induced by type I pyrethroids rapidly decays after the depolarizing pulse ceases, typically returning to baseline levels within 1–2 s. By contrast, the tail current induced by type II pyrethroids demonstrates a significantly slower decay, requiring several seconds to over 10 s for complete inactivation.

Figure 3.

Effects of mutations M918T, M918L, L1014F and M918T/L1014F on the sensitivity of Megalurothrips usitatus sodium channels (MuNav1‐1) to permethrin (PMT), deltamethrin (DMT), bifenthrin (BFT) and λ‐cyhalothrin (CHT). (A)–(D) Tail currents induced by PMT (A), DMT (B), BFT (C) and CHT (D) from MuNa_v1‐1 and mutant channels. (E) and (F) Percentage of channel modification by 1.0 μm PMT, BFT (E) and DMT and CHT (F) as determined using the method of Tatebayashi and Narahashi. ⁴¹ The number of oocytes for wild‐type MuNa_v1‐1 and each mutant was n = 3. Error bars indicate mean ± SEM. Asterisks indicate a significant decrease from the MuNav1‐1 channel as determined by using one‐way analysis of variance with Scheffe's post hoc analysis. Significant values were set at P < 0.05.

The M918T, M918L, L1014F, and M918T/L1014F mutations markedly decrease the sensitivity of MuNa_v1‐1 to pyrethroids at 1 μm (Fig. 3(E),(F)). In wild‐type MuNa_v1‐1, the percentage of modified channels was 75.14% (DMT), 76.34% (CHT), 73.76% (BFT), and 65.80% (PMT). In stark contrast, these values were severely diminished in all mutants. For the M918L mutant, the modifications were 7.17% (DMT), 1.91% (CHT), 6.60% (BFT), and 3.68% (PMT). The M918T mutant showed slightly higher, but still greatly reduced modifications: 9.11% (DMT), 2.49% (CHT), 11.51% (BFT), and 7.79% (PMT). Both the L1014F single mutant and the M918T/L1014F double mutant also established significant levels of resistance. Subsequently, we increased the drug concentration to 10 μm, and the tail currents of the four mutants were markedly larger than those observed under modification with 1 μm pyrethroid. Under modification with 10 μm type I pyrethroid, the channel modification percentages of the four mutants remained significantly lower than that of the wild‐typeMuNav1‐1 at 1 μm. Under modification with 10 μm type II pyrethroid, the channel modification percentage of the L1014F mutant showed no difference from that of the wild type, whereas those of the M918L and M918T/L1014F double mutants were still significantly lower than that of the wild‐type (Supporting Information, Fig. S2).

Dose–response curves for BFT, DMT, PMT, and CHT were generated for the wild‐type MuNa_v1‐1 and the four mutant channels (M918T, M918L, L1014F, M918T/L1014F) (Fig. 4). All mutants exhibited virtually no response at concentrations of 0.01, 0.03, and 0.1 μm, hence subsequent analysis focused on the concentrations of 0.3, 1, 3, 10, and 30 μm. The EC₅₀ values of all mutants were significantly lower than that of the wild‐type MuNa_v1‐1. Among the single‐point mutants, M918L exhibited the lowest sensitivity to DMT and CHT compared with the other mutation types. For the double mutant M918T/L1014F, because of its strong resistance, the dose–response curves failed to achieve convergence of Hill function fitting under DMT and BFT treatment; however, under CHT treatment, the double mutant exhibited a significant synergistic effect (Supporting Information, Table S2). This enhanced resistance was evident across the entire tested concentration gradient (0.3–30 μm).

Figure 4.

Dose–response curves for pyrethroid‐induced modification of wild‐type and mutant channels following treatment with (A) permethrin, (B) deltamethrin, (C) bifenthrin, and (D) λ‐cyhalothrin. Curves were fitted using the Hill equation. Data points represent the mean ± SEM (n = 3 oocytes per concentration).

3.4. Structural mapping and docking analysis of pyrethroids in MuNav1‐1

To complement the electrophysiological findings, computational modeling was used to investigate the structural basis of resistance. A homology model of wild‐type MuNav1‐1 was constructed, and the previously defined dual pyrethroid receptor regions, PyR1 and PyR2, were mapped onto the channel structure. (Fig. 5). In the MuNav1‐1 model, residue M944, corresponding to canonical M918, is located in the helix between IIS4 and IIS5, with its side chain extending into the PyR1 pocket. Residue L1040, corresponding to canonical L1014, is located on the IIS6 helix, with its side chain facing the PyR2 region and projecting into the PyR2 pocket.

Figure 5.

Homology model of the Megalurothrips usitatus sodium channel (MuNav1‐1). (A) Overall structure of MuNav1‐1. Domains I–IV are colored in orange, azure, cyan and purple, respectively. Residues M944 and L1040 (corresponding to canonical M918 and L1014, respectively) are shown as red sticks. Pyrethroid receptor site 1 (PyR1) is indicated by a dashed box. (B) A 90° left‐rotated view of panel A showing the pyrethroid receptor site 2 (PyR2). PyR2 is indicated by a dashed box.

Docking of BFT, PMT, DMT, and CHT within the predefined PyR1 and PyR2 regions yielded plausible ligand poses in the wild‐type MuNav1‐1 model (Figs 6 and 7). In both receptor regions, the four pyrethroids could be accommodated within the local binding environment, supporting the structural relevance of PyR1 and PyR2 in MuNav1‐1. These docking results provide a structural framework for pyrethroid interaction with the wild‐type channel.

Figure 6.

Representative docking poses of four pyrethroids in the pyrethroid receptor site 1 (PyR1) region of the Megalurothrips usitatus sodium channel (MuNav1‐1): bifenthrin (A), permethrin (B), λ‐cyhalothrin (C) and deltamethrin (D). The four pyrethroids were docked into the PyR1 region of MuNav1‐1. The protein is shown in cartoon and surface representation, ligands are shown as stick models, and selected residues surrounding the PyR1 region are labeled. Residue M944 of MuNav1‐1, corresponding to canonical M918, is highlighted in red.

Figure 7.

Representative docking poses of four pyrethroid insecticides in the pyrethroid receptor site 2 (PyR2) region of MuNa_v1‐1: bifenthrin (A), permethrin (B), λ‐cyhalothrin (C) and deltamethrin (D). The four pyrethroids were docked into the pyrethroid receptor site 2 (PyR2) region of the Megalurothrips usitatus sodium channel (MuNav1‐1). The protein is shown in cartoon and surface representation, ligands are shown as stick models, and selected residues surrounding the PyR2 region are labeled. Residue L1040 of MuNav1‐1, corresponding to canonical L1014, is highlighted in red.

To further examine the possible effects of resistance‐related substitutions, local residue replacement was performed in selected wild‐type ligand‐bound complexes. In the PyR1 region of the MuNav1‐1–DMT complex, wild‐type M944 showed multiple favorable contacts with the docked ligand, whereas substitution with threonine or leucine reduced these contacts (Fig. 8(A)). In the PyR2 region, wild‐type L1040 showed a spatially compatible arrangement with DMT, whereas substitution with phenylalanine introduced a bulkier aromatic side chain that caused a steric clash with the ligand (Fig. 8(B)). Together, these structural observations suggest that substitutions at positions corresponding to M918 and L1014 may alter the local ligand‐binding environment and reduce favorable ligand–pocket interactions, which is consistent with the reduced pyrethroid sensitivity observed in the electrophysiological assays.

Figure 8.

Structural effects of substitutions at M944 and L1040 on pyrethroid binding in the Megalurothrips usitatus sodium channel (MuNa_v1‐1). MuNa_v1‐1 residues M944 and L1040 correspond to canonical M918 and L1014, respectively. (A) Local view of deltamethrin docked in the pyrethroid receptor site 1 (PyR1) region. Wild‐type M944 is shown as a green stick model, whereas substituted T944 and L944 are shown as a purple and a red stick model, respectively. Deltamethrin is shown as a pale yellow–green stick model. Green dashed lines indicate favorable contacts between the residue and the docked ligand. Compared with wild‐type M944, both substitutions show fewer favorable contacts with deltamethrin. (B) Local view of deltamethrin docked in the pyrethroid receptor site 2 (PyR2) region of MuNa_v1‐1. Wild‐type L1040 is shown as a green stick model, and substituted F1040 is shown as a red stick model. The L1040F substitution introduces a bulkier aromatic side chain that causes a steric clash with the docked ligand.

4. DISCUSSION

Target‐site resistance via mutations in the VGSC is a well‐established mechanism conferring resistance to DDT and pyrethroid insecticides across diverse insect species. ¹⁴ , ⁴² Under sustained insecticide selection pressure, a spectrum of kdr mutations has evolved within VGSCs and electrophysiological techniques, ¹⁰ such as patch clamp and two‐electrode voltage clamp, have been instrumental in deciphering the functional link between these mutations and the resistance phenotype. ⁴³

The M918T mutation was first identified in houseflies, and the more recently reported M918L substitution has been found in T. tabaci, Myzus persicae and Rhopalosiphum padi. ³³ , ⁴⁴ , ⁴⁵ , ⁴⁶ Similarly, residue 1014 also exhibits considerable variability, with substitutions including L1014F, L1014H, L1014S, L1014C and L1014W reported in various pests. ²² , ⁴⁷ Among these, L1014H has been documented in housefly populations across multiple continents, and L1014S has been identified in B. germanica. Notably, emergence of the M918T mutation often demonstrates background dependence, frequently being detected only in genetic backgrounds already harboring the L1014F mutation. ⁴⁸ , ⁴⁹ , ⁵⁰ , ⁵¹ , ⁵² This co‐occurrence pattern has also been observed in T. tabaci and the resulting double mutant M918T + L1014F, often heterozygous, is associated with a high fitness cost but confers strong resistance. ³³ Recently, the M918T, M918L, L1014F, and M918T/L1014F mutations were reported in a thelytokous strain of T. tabaci, where they were correlated with pyrethroid resistance, although their precise functional roles remained uncharacterized. ³⁴ Against this backdrop, our study aimed to functionally validate these mutations and quantify their effects in a thrips VGSC model.

The functional consequences of kdr mutations can vary across insect species. In Musca domestica, the M918T mutation induces a hyperpolarizing shift in the steady‐state inactivation curve and confers strong resistance to cismethrin, although it shows no effect against cypermethrin at high concentrations. ¹⁹ In our study using the M. usitatus VGSC, although M918T did not significantly alter channel gating, as a single mutation it conferred high levels of resistance to all four pyrethroids tested. At 1 μm, it reduced channel sensitivity by approximately 8‐fold (PMT), 8‐fold (DMT), 6‐fold (BFT), and 31‐fold (CHT) compared with the wild‐type MuNa_v1‐1. Unlike M918T, the M918L mutation caused a hyperpolarizing shift in the activation curve and exhibited a distinct resistance profile, conferring particularly strong resistance, with sensitivity reductions of approximately 18‐fold (PMT), 10‐fold (DMT), 11‐fold (BFT), and 40‐fold (CHT) at 1 μm. The exceptionally high resistance to CHT (40‐fold) aligns with its character as a type II pyrethroid, which typically elicits a more persistent disruption of channel inactivation compared with type I compounds. For the L1014F mutation, previous work in Drosophila reported a depolarizing shift in both activation and inactivation voltages, an accelerated inactivation process, and a 17‐fold reduction in sensitivity to DMT, attributed to altered channel conformation reducing pyrethroid binding. ²⁰ , ³⁹ Similarly, in houseflies, L1014F alone causes at least a tenfold reduction in DMT sensitivity. ¹⁹ Our results in the homozygous M. usitatus background are consistent with these findings: L1014F caused a significant depolarizing shift in the activation curve and substantially reduced sensitivity to pyrethroids, with reductions of approximately 25‐fold (PMT), 3‐fold (DMT), 24‐fold (BFT), and 3‐fold (CHT) at 1 μm.

Double mutants also exhibit a similar phenomenon. In Musca domestica, the L1014F + M918T combination causes a depolarizing shift in activation and confers extreme resistance, raising the channel activation threshold. In Drosophila melanogaster, this double mutation results in a + 0.9 mV shift in activation potential and a significantly accelerated inactivation process (time constant shortened by 5.6 ms), leading to an approximately 102‐fold reduction in sensitivity to DMT compared with the wild‐type channel. ¹⁹ , ⁵³ In other pests such as Haematobia irritans, Myzus persicae, Liriomyza huidobrensis, and Tuta absoluta (where direct electrophysiological characterization is lacking), the same double mutation is also associated with significantly reduced sensitivity to both type I and type II pyrethroids. ⁵⁴ , ⁵⁵ , ⁵⁶ Previous research on the gating characteristics of M918T and L1014F indicated that, in the para sodium channel, both the V _1/2 activation voltage and the inactivation voltage of the L1014F mutant shifted toward depolarization compared with the wild type. In addition, the V _1/2 activation voltage and deactivation voltage of the L1014F + M918T mutant also shifted in the depolarization direction, although the shift was smaller than that observed in the L1014F mutant type. ³⁹ Furthermore, another study examined the gating characteristics of M918T in the para Na⁺ channel. The results indicated that, compared with the wild type, both the V _1/2 activation voltage and the inactivation voltage of the M918T mutant shifted in the direction of depolarization. However, its offset was larger than that of the L1014F + M918T mutant but smaller than that of the L1014F mutant. ²⁰ In the research results concerning the effect of DDT (100 μm) on the activation, inactivation, and deactivation of the para wild‐type and mutant Na channels, the control group, which received no DDT treatment, exhibited notable differences compared with the wild‐type. The V _1/2 activation by the three mutant types, L1014F, M918T, and L1014F + M918T, all shifted to varying degrees in the depolarization direction. In addition, the V _1/2 inactivation by L1014F also shifted toward depolarization. However, it is important to note that the V _1/2 inactivation by the M918T and L1014F + M918T mutant types shifted toward hyperpolarization, a finding that diverged from the previously mentioned research results. ⁵⁷ The research results on activation differed slightly from those observed in the VGSC of M. usitatus. Our findings indicated that the activation V _1/2 of M918T and L1014F + M918T shifted toward hyperpolarization, facilitating the opening of VGSCs and their binding to pyrethroids. Alterations in the gating control of VGSCs influenced the ease of channel opening or closing, which may have indirectly affected the sensitivity of these channels to pyrethroids. Nevertheless, changes in gating control were not the most significant factor contributing to variations in sensitivity.

Computational analysis provided structural support for the mutation‐associated resistance patterns observed in the electrophysiological assays. Previous modeling and mutational studies, including those of O'Reilly and Du and colleagues, identified the IIS4–S5 linker and IIS6 as important parts of the pyrethroid receptor environment within the PyR1/PyR2 framework. ¹³ , ⁵⁸ In our MuNav1‐1 model, the positions corresponding to M918 and L1014 were also located close to these receptor regions, which is consistent with the general receptor‐site architecture described previously. The M918T (super‐kdr) mutation substitutes a hydrophobic methionine with a hydrophilic threonine, introducing a polar defect into the core of the otherwise hydrophobic binding pocket. ⁵⁹ The L1014F mutation replaces a branched‐chain leucine with a bulky, rigid phenylalanine. This substitution primarily exerts a steric hindrance effect, forcing rearrangement of surrounding residues and thereby altering the geometry and chemical properties of a key sub‐pocket. Although M918L is a hydrophobic‐to‐hydrophobic substitution, it may still alter the three‐dimensional topology of the local binding pocket via changes in side chain branching. In our structural comparison, M944T and M944L reduced favorable contacts with DMT in the PyR1 region, whereas L1040F introduced steric clash in the PyR2 region. These observations are consistent with earlier functional studies showing that substitutions at M918 and L1014 reduce pyrethroid sensitivity, and they support the view that reduced pyrethroid action in the mutant channels is associated primarily with local changes in the ligand‐binding environment rather than with a single common gating effect.

5. CONCLUSION

This study provides direct functional validation that the M918L, M918T, L1014F, and M918T + L1014F mutations in the M. usitatus VGSC are causative molecular determinants of pyrethroid resistance. By integrating homozygous expression, detailed electrophysiology, and molecular modeling, we have systematically elucidated how these mutations alter channel gating and drastically reduce insecticide sensitivity, with compound‐specific effects. This study clarified the complex functional significance of these mutations and provided the mechanistic explanation for the substitution of M918L. Meanwhile, it revealed the specific gating responses of the two mutant types, M918T and M918T + L1014F, in Xenopus oocytes expression system. These findings advance the theoretical understanding of target‐site resistance from correlation to mechanistic causation. Crucially, the observed mutation and compound‐specific sensitivity profiles inform practical resistance management. The severe resistance conferred by M918T and the double mutation underscores the urgency of early detection. Conversely, the nuanced effects of M918L, which retains sensitivity to BFT, highlight that resistance is not monolithic. This knowledge enables the design of precise insecticide deployment strategies, such as the rational rotation or mixing of compounds based on the prevailing kdr genotype detected in field populations. Ultimately, sustaining pyrethroid efficacy will depend on such science‐driven, mutation‐aware pest management approaches.

AUTHOR CONTRIBUTIONS

LW: Writing—original draft, Methodology, Data curation. RG: Writing—original draft, Methodology. GX: Writing—original draft, Methodology, Data curation. LY: Data curation. PC: Methodology. KZ: Methodology. FL: Writing—review & editing, Supervision, Resources, Methodology, Funding acquisition. SW: Writing—review & editing, Supervision, Resources, Methodology, Funding acquisition.

CONFLICTS OF INTEREST

The authors confirm that there are no financial or personal conflicts of interest that could have influenced this research.

Supporting information

Table S1. Primer sequences for site‐directed mutagenesis of the M. usitatus sodium channel gene (MuNa_v1‐1) for mutations M918T, M918L, L1014F and M918T/L1014F.

Table S2. EC₅₀ values of four pyrethroids for MuNa_v1‐1, M918T, M918L, L1014F, and M918T/L1014F mutants.

Figure S1. Sequence alignment of partial Domain II regions of sodium channels from multiple thrips species.

Figure S2. Effects of 10 μM permethrin (PMT), deltamethrin (DMT), bifenthrin (BFT), and λ‐cyhalothrin (CHT) on the sensitivity of M918T, M918L, L1014F, and M918T/L1014F mutants, and of 1 μM pyrethroid modification on the sensitivity of wild‐type MuNa_v1‐1.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.