Fusion dsRNA targeting ecdysteroid signaling provides efficient and selective RNAi‐based control of spider mites

Abstract

BACKGROUND

Tetranychus urticae is a major agricultural pest mite whose rapid evolution of pesticide resistance necessitates innovative control strategies. RNA interference (RNAi) offers a promising alternative to chemical pesticides for its high specificity, efficiency, and environmental safety. Ecdysteroid signaling genes, which regulate molting and development in arthropods, represent potential RNAi targets for inducing lethal molting defects. This study investigated the roles of ecdysteroid signaling pathway genes in T. urticae molting and evaluated the potential of fusion double‐stranded RNA (dsRNA) constructs targeting multiple genes to enhance RNAi efficacy.

RESULTS

The expression profiling revealed that TuE74, TuECR, TuHR4, TuE93, and TuHR3 were highly expressed during the molting stage of T. urticae. RNAi‐mediated silencing of these genes, particularly TuECR, TuHR3, TuHR4, and TuE74, caused distinct molting defects and high mortality. Fusion dsRNAs targeting combinations of these genes (ECR‐HR3‐HR4 or ECR‐HR3‐E74) achieved stronger and more coordinated gene silencing, lower median lethal concentration (LC₅₀) and LC₉₀ values than single‐gene dsRNA (dsTuHR3), and exhibited cross‐species efficacy against Tetranychus cinnabarinus and Tetranychus evansi. Importantly, the fusion dsRNAs showed no adverse effects on the predatory mite Neoseiulus californicus, supporting their biosafety. Spray‐based delivery of fusion dsRNA significantly delayed molting, confirming its biological activity, although optimization of delivery efficiency is still required to achieve complete developmental arrest.

CONCLUSION

The present study demonstrates that multi‐target fusion dsRNA can substantially enhance RNAi efficacy, broaden control spectra across pest mite species, and maintain environmental safety. These results provide a molecular basis and applied framework for the development of next‐generation RNAi‐based acaricides for sustainable spider mite management. © 2026 Society of Chemical Industry.

Our study reveals that silencing key ecdysteroid signaling genes (TuECR, TuHR3, TuHR4, TuE74) disrupts molting in Tetranychus urticae. Fusion double‐stranded RNA (dsRNA) targeting these genes enhances RNA interference (RNAi) efficacy, demonstrates cross‐species activity, and is safe for a predatory mite, offering a promising strategy to develop next‐generation RNAi‐based acaricides for sustainable spider mite management.

1. INTRODUCTION

The two‐spotted spider mite (Tetranychus urticae Koch, Acariformes: Tetranychidae) is one of the most polyphagous and economically important pests among the Acari. It infests over 1100 plant species from 140 families, including vegetables, fruits, and ornamentals, and occurs worldwide across diverse agroecosystems. ¹ Feeding through a piercing‐sucking mechanism, T. urticae extracts cell contents from mesophyll tissues, leading to chlorosis, desiccation, and premature leaf abscission, ultimately reducing crop vigor and yield. ² Despite decades of research, chemical acaricides remain the predominant tool for T. urticae management. However, extensive and repeated acaricide applications have resulted in the rapid development of resistance. Owing to its short life cycle, high reproductive rate, arrhenotokous parthenogenesis, and strong detoxification and metabolic capacities, T. urticae readily adapts to selective pressures. ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ Resistance to multiple acaricides, including abamectin and bifenthrin, has been reported in field populations from several regions of China. ⁹ These characteristics make T. urticae an important model for studying acaricide resistance mechanisms and the evolution of detoxification and adaptation in mites.

Given the limitations of chemical control, alternative strategies consistent with integrated pest management (IPM) principles are urgently needed. RNA interference (RNAi) technology represents a promising new molecular tool for both basic functional genomics and applied pest control. RNAi is a post‐transcriptional gene‐silencing mechanism conserved among eukaryotes that degrades target messenger RNA (mRNA) through double‐stranded RNA (dsRNA) or small interfering RNA (siRNA), thereby suppressing gene expression. Owing to its high specificity, efficiency, and environmental safety, RNAi has attracted attention as a potential ‘green’ alternative to chemical pesticides. ¹⁰ Successful cases of RNAi‐based pest control have been demonstrated in insects: transgenic maize expressing dsRNA against Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae) (SmartStax® Pro) protects roots from larval damage, ¹¹ and the sprayable RNAi biopesticide Ledprona®, targeting Leptinotarsa decemlineata (Coleoptera: Chrysomelidae), has achieved field registration. ¹² Such successes highlight the potential of RNAi for suppressing key physiological processes, which are development, reproduction, and feeding, via dsRNA delivery through sprays, transgenic expression, or microbial vectors. ¹³ In contrast, progress in RNAi‐based control of spider mites remains limited. ¹⁴ Although T. urticae shares many traits with insect pests, including rapid resistance evolution and high fecundity, the physiological and cellular factors underlying RNAi efficiency in mites are poorly understood. The development of effective RNAi delivery systems and identification of essential gene targets in mites are therefore key biological challenges for advancing acarological molecular research.

Molting represents one of the most critical physiological processes in arthropod development. It is orchestrated by ecdysteroid hormones, primarily ecdysone and 20‐hydroxyecdysone, which regulate cuticle synthesis and ecdysis through a complex transcriptional cascade. ¹⁵ Ecdysteroids act via a heterodimeric receptor composed of the ecdysone receptor (EcR) and ultraspiracle protein (USP), initiating a hierarchy of gene activation. Upon elevation of ecdysteroid titers, early‐response transcription factors such as E74, E75, and E93 are induced, which in turn regulate the expression of late‐response genes including HR3, HR4, HR38, HR39, and E78. ¹⁶ , ¹⁷ , ¹⁸ This signaling network controls epidermal cell reprogramming and cuticle remodeling during molting. Experimental interference with key ecdysteroid signaling genes often results in lethal molting defects. ¹⁹ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ , ²⁵ For example, feeding Helicoverpa armigera (Lepidoptera: Noctuidae) larvae on transgenic tobacco expressing dsHaEcR significantly reduced EcR mRNA levels, causing larval‐pupal intermediates and mortality due to incomplete molting. ²⁶ These studies illustrate how hormonal signaling genes serve as fundamental regulators of arthropod growth and development. Comparable ecdysteroid signaling mechanisms appear to operate in spider mites. Functional studies have demonstrated that suppression of E75 and E78 in T. urticae, HR3 and EcR in Panonychus citri (Acariformes: Tetranychidae), and EcR in Tetranychus cinnabarinus (Acariformes: Tetranychidae) all impair molting and development. ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ Transcriptome analyses by Li et al. ³² revealed that T. urticae ecdysteroid‐related genes, including TuECR, TuHR3, TuHR4, TuHR38, and TuHR38b, show marked expression changes during molting, yet their precise regulatory interactions and functional significance remain poorly defined. Understanding how these genes coordinate molting physiology is crucial for elucidating the developmental endocrinology of mites and identifying molecular vulnerabilities for control.

Most RNAi‐based pest control studies to date have targeted individual genes, but single‐gene silencing may not provide sufficient suppression in species with variable RNAi sensitivity, such as T. urticae. Multi‐target or fusion dsRNA constructs, which combine sequences from multiple essential genes, can enhance silencing efficiency, prolong biological effects, and reduce the likelihood of RNAi resistance. ³³ , ³⁴ Exploring the physiological responses of mites to such constructs could provide both mechanistic insights into gene regulatory networks and applied value for sustainable mite management. In the present study, we therefore investigated the expression profiles and physiological functions of key ecdysteroid signaling genes (TuECR, TuE74, TuE93, TuHR3, TuHR4, TuHR38, TuHR38b, and TuHR39) across the developmental stages of T. urticae. We first characterized temporal expression patterns during molting to identify stage‐specific regulatory genes. Next, we evaluated the roles of these genes in molting and development through RNAi‐mediated gene silencing. Finally, we designed multi‐target fusion dsRNA constructs and assessed their biological activity using both microinjection and foliar spray methods (spray‐induced gene‐silencing (SIGS)). Together, these findings provide new insights into the molecular regulation of molting in T. urticae, broaden understanding of ecdysteroid signaling in the Acari, and establish a foundation for developing RNA‐based approaches for spider mite control.

2. MATERIALS AND METHODS

2.1. Mites

Spider mites used in this study, including Tetranychus evansi (Acariformes: Tetranychidae), T. urticae, and T. cinnabarinus, were obtained from the Institute of Entomology, Guizhou University, Guiyang, China. Mites were maintained on common bean (Phaseolus vulgaris) (Fabaceae: Fabales) and tomato (Solanum lycopersicum) (Solanaceae: Solanales) plants under controlled laboratory conditions: 27 ± 1 °C, 65 ± 5% relative humidity, and a photoperiod of 14 h light followed by 10 h darkness. ³⁵ The predatory mite Neoseiulus californicus (Parasitiformes: Phytoseiidae) was obtained from Jiyuan Baiyun Industrial (Henan Province, China) and reared on T. urticae as prey, as reported in a previous study. ³⁶

2.2. Developmental expression profiling of ecdysteroid signaling genes in T. urticae

To characterize temporal expression patterns of eight ecdysteroid signaling genes (TuECR, TuE74, TuHR3, TuHR4, TuHR39, TuE93, TuHR38, and TuHR38b), newly molted (0–8 h post‐molting) and pre‐molting (within 8 h of molting initiation) individuals were collected at different developmental stages: 1000 eggs, 800 larvae, 500 protonymphs, 100 deutonymphs, and 50 adults per replicate, with four biological replicates per stage. Additionally, deutonymphs aged 0–44 h were collected at 4‐h intervals (0–4, 4–8, …, 40–44 h) with 100 mites per replicate and four biological replicates.

Total RNA was isolated using TRIzol® (Sangon Biotech, Shanghai, China), assessed on a 1% agarose gel, and quantified by Nanodrop (Thermo Fisher Scientific, Waltham, MA, USA). First‐strand complementary DNA (cDNA) was synthesized from 1 μg of DNase‐treated RNA using StarScript II RT Mix (GenStar, Beijing, China). The cDNA was diluted 25‐fold and subjected to real‐time quantitative polymerase chain reaction (RT‐qPCR) on a CFX384 system (Bio‐Rad, Hercules, CA, USA) using RealStar Green Fast Mixture (GenStar) under the following cycling conditions: 95 °C for 120 s; 40 cycles of 95 °C for 15 s and 60 °C for 30 s. Each 10‐μL reaction contained 5 μL of 2× Master Mix, 1 μL of gene‐specific primers, and 4 μL of cDNA. Relative gene expression was calculated using the 2^−ΔΔCt method against the TuATP reference gene, with three technical replicates per sample. ³⁷ Gene sequences were retrieved from the T. urticae genome database (https://bioinformatics.psb.ugent.be/orcae/overview/Tetur), and primers were designed using the National Center for Biotechnology Information (NCBI) Primer‐BLAST (Supporting Information, Table S1).

2.3. Design and synthesis of dsRNA

Primers for dsRNA targeting TuECR, TuE74, TuHR3, TuHR4, TuHR39, TuE93, TuHR38, TuHR38b, and the enhanced green fluorescent protein gene (EGFP, negative control) were designed via NCBI Primer‐BLAST with lengths of 350–550 bp. To synthesize dsRNA, template DNA fragments flanked by T7 promoter sequences were first generated by PCR using first‐strand cDNA as the template. The 25‐μL PCR reaction system contained 12.5 μL of 2× Taq PCR StarMix with Loading Dye, 0.5 μL each of forward and reverse primers, 2 μL of cDNA template, and 8.5 μL of diethyl pyrocarbonate (DEPC)‐treated water. The thermal cycling conditions were as follows: initial denaturation at 95 °C for 2 min; 40 cycles of denaturation at 95 °C for 30 s, annealing at 60 °C for 30 s, and extension at 72 °C for 40 s (extension time was adjusted according to the length of the amplicon); followed by a final extension at 72 °C for 5 min. The PCR products were separated on a 1% agarose gel and purified using the SanPrep Column DNA Gel Extraction Kit (Sangon Biotech) according to the manufacturer's instructions. Finally, dsRNAs were synthesized in vitro using the Transcript Aid T7 High Yield Transcription Kit (Thermo Fisher Scientific, Shanghai, China).

Fusion dsRNAs were designed using the dsRNAEngineer platform (https://dsrna-engineer.cn/) following established procedures. ³⁸ The mRNA sequences of TuECR, TuE74, TuHR3, and TuHR4 were computationally digested into 16‐, 21‐, and 26‐bp fragments. These fragments were then evaluated using defined sequence matching criteria: a 21‐bp perfect match was set as the major condition to ensure effective on‐target silencing, while a 16‐bp perfect match and a 26‐bp almost perfectly matched served as additional conditions for comprehensive risk screening. Specifically, each fragment was assessed for on‐target conservation among four pest mite species: T. urticae, T. evansi, T. cinnabarinus, Tetranychus truncatus (Acariformes: Tetranychidae); and for potential off‐target risks in six beneficial species: Apis cerana (Hymenoptera: Apidae), Coccinella septempunctata (Coleoptera: Coccinellidae), Geocoris occidentalis (Parasitiformes: Phytoseiidae), Harmonia axyridis (Coleoptera: Coccinellidae), Neoseiulus barkeri (Parasitiformes: Phytoseiidae), and Neoseiulus californicus). DNA templates were chemically synthesized (Sangon Biotech, Shanghai, China) and transcribed into dsRNA. Purification was performed via chloroform/phenol extraction, and integrity and concentration were confirmed by 1% agarose gel electrophoresis and NanoDrop/Agilent 2100 Bioanalyzer. The dsRNAs were stored at −80 °C. Primer sequences are listed in Table S1; fusion dsRNA sequences are provided in Table S2.

2.4. Microinjection

Deutonymphs (8 h post‐molt) were dorsally injected with approximately 2.25 nL of dsRNA (10 μg μL⁻¹) using an SYS‐PV830 system (World Precision Instruments, Sarasota, FL, USA). Controls received an equal volume of dsEGFP. Four biological replicates were established with at least 20 mites per replicate. Post‐injection, mites were maintained on bean leaves under standard conditions, with molting, mortality, and death phenotypes recorded every 12 h for 96 h. Gene silencing was quantified via RT‐qPCR 24 h post‐injection, with four biological replicates (≥ 50 mites each) and three technical replicates. Relative expression was normalized to TuATP, TeATP, or TcATP as appropriate using the 2^−ΔΔCt method. Dose‐dependent toxicity of target and fusion dsRNAs was evaluated at 5000, 1000, 500, 100, 50, 40, and 25 ng μL⁻¹, following the same protocol with four replicates of ≥ 20 mites each.

2.5. Safety assessment on Neoseiulus californicus

The safety of fusion dsRNA to Neoseiulus californicus was assessed using a droplet‐feeding assay with dsEGFP as the control. ³⁶ Newly molted deutonymphs were starved for 24 h on leaf discs and then fed 10 μL of artificial liquid diet containing 1000 ng μL⁻¹ dsRNA, 2 μL of saturated sucrose, and 0.6 μL of blue food dye. Following 24 h of feeding, ≥ 20 mites per replicate were transferred to fresh discs with T. urticae prey and monitored every 12 h for molting and survival (three replicates). Based on the sequence‐matching analysis described in Section 2.3, the Neoseiulus californicus transcript with the highest number of predicted non‐target matching sites was selected as a representative potential non‐target indicator for evaluating possible non‐target effects. For expression analysis, 50 blue‐pigmented mites per replicate were collected to assess RNAi efficiency by RT‐qPCR, normalized to NcRPL13a, with four biological replicates and three technical repeats.

2.6. RNAi delivery via spray

Fusion dsRNAs (dsfusion1, dsfusion2) and dsEGFP were diluted to 1000 ng μL⁻¹ and combined with star polycation (SPc) nanoparticles at a 1:1 mass ratio, incubated 15 min at room temperature, and 0.05% detergent was added as surfactant. ³⁹ Then 1 μL of the mixture was verified for binding by 1% agarose gel electrophoresis. Deutonymphs within 8 h post‐molt were treated using a spray tower. Post‐treatment, the mites were maintained on the sprayed leaf discs under standard conditions, with molting, mortality, and death phenotypes recorded at 12 h intervals for 72 h (T. urticae) or 96 h (T. evansi). Four biological replicates were performed with ≥ 20 mites per replicate. Gene silencing was quantified via RT‐qPCR 24 h post‐spray treatment, with four biological replicates (≥ 50 mites each) and three technical replicates. Relative expression was normalized to TuATP or TeATP as appropriate using the 2^−ΔΔCt method.

2.7. Statistical analysis

Data were tested for normality and homogeneity of variance using Kolmogorov–Smirnov and Levene's tests in SPSS 19 (IBM, Armonk, NY, USA). One‐way analysis of variance (ANOVA) followed by Duncan's multiple range test was used to compare relative gene expression across stages and time points, and to analyze dsRNA‐induced mortality. RNAi efficiency was assessed using Student's t‐test. Differences were considered significant at P < 0.05. Data visualization was performed in GraphPad Prism 8.0. Toxicity regression equations for dsRNAs were calculated using SPSS 19.

3. RESULTS

3.1. Developmental expression analysis of ecdysteroid signaling pathway genes in T. urticae

Across developmental stages, TuE74, TuHR3, TuHR4, TuHR39, and TuECR (except in deutonymphs), as well as TuE93 (except in protonymphs), exhibited higher expression levels during the 8 h quiescent period than at 8 h post‐molting, showing a characteristic negative sawtooth pattern (Fig. 1(A)). This coordinated pattern suggests these genes may share functional roles in regulating molting. In contrast, TuHR38 (except in larvae) and TuHR38b displayed higher expression at 8 h post‐molting, producing a positive sawtooth pattern.

Figure 1.

Temporal expression profiles of eight ecdysteroid signaling genes in Tetranychus urticae. (A) Expression levels across developmental stages (eggs, larvae, protonymphs, deutonymphs, adults) comparing the 8 h quiescent period before molting and 8 h post‐molt. (B) Time‐course expression during the deutonymph stage (4–44 h post‐molting). Data represent mean ± standard deviation of four biological replicates. Detailed expression values for each gene at different stages and time points are provided in Supporting Information, Table S3.

Time‐course analysis of deutonymphs revealed distinct peak expression patterns for individual genes: TuHR38b peaked at 4 h and remained low thereafter; TuHR38 peaked at 8 h with minor fluctuations; TuHR39 peaked at 12 h; TuE74 began increasing at 20 h and peaked at 24 h; TuHR4 and TuHR3 rose at 24 h, peaking at 36 h and 40 h, respectively; TuECR peaked at 32 h; and TuE93 peaked at 36 h (Fig. 1(B)). Notably, 24–44 h corresponds to the molting phase of deutonymphs.

These data suggest that TuE74, TuHR3, TuHR4, TuHR39, TuECR, and TuE93 may regulate molting via a negative sawtooth pattern during quiescence, while the cascading peaks of TuHR4, TuHR3, TuECR, and TuE93 during the critical molting window indicate their central regulatory roles in molting physiology.

3.2. Knockdown of ecdysteroid signaling genes impairs T. urticae development

RNAi via microinjection of gene‐specific dsRNAs effectively silenced all eight target genes in deutonymphs within 8 h post‐molting, with efficiencies ranging from 50% to 88% (Fig. 2(A)). In dsEGFP controls, survival and molting rates were 92% and 87%, respectively. Survival and molting rates were 79% and 78% for dsTuHR38; 91% and 88% for dsTuHR38b; 89% and 88% for dsTuHR39; and 94% and 91% for dsTuE93 (Fig. 2(B),(C)). Thus, silencing these genes did not cause acute molting defects.

Figure 2.

Effects of dsRNA injection on gene expression and development in Tetranychus urticae deutonymphs. (A) Relative expression levels (mean ± standard error) of target genes 24 h after dsRNA injection. Asterisks indicate statistically significant differences compared to dsEGFP controls (Student's t‐test, P < 0.01). (B) Survival (%) of mites after dsRNA treatments. (C) Molting rate (%) of mites after dsRNA treatments. (D) Representative phenotypes of deutonymph mites following dsRNA injection. Data represent four biological replicates per treatment.

The normal molting cycle of T. urticae comprises a feeding/wandering stage, a quiescent stage (characterized by immobility and the convergence of all legs, particularly the posterior two pairs), and finally ecdysis. Our RNAi experiments revealed distinct blocks at specific points in this process. In contrast, dsTuECR caused 93% of mites to fail molting, with 86% showing reduced mobility and inability to enter quiescence (legs not converged), ultimately dying. dsTuHR3 treatment led to 95% molting failure; 89% entered quiescence but failed to shed the old cuticle, dying trapped within exuviae. Injection of dsTuHR4 and dsTuE74 resulted in molting failure in 75% and 93% of mites injected, respectively, producing two distinguishable lethal phenotypes: incomplete apolysis with body shrinkage or post‐apolysis death with body shrinkage (Fig. 2(D)).

These findings demonstrate that TuECR, TuHR3, TuHR4, and TuE74 are essential for molting in T. urticae and represent promising RNAi targets for spider mite management.

3.3. Design and efficacy evaluation of fusion dsRNA for controlling T. urticae

Silencing of TuECR, TuHR3, TuE74, and TuHR4 induced distinct lethal phenotypes, suggesting differential roles in molting. Fusion dsRNAs targeting multiple genes were designed based on these observations (Fig. 3(A)). Bioinformatic analysis confirmed on‐target and off‐target sites across four pest mite species (T. urticae, T. evansi, T. cinnabarinus, T. truncatus) and six beneficial species (A. cerana, C. septempunctata, G. occidentalis, Harmonia axyridis, Neoseiulus barkeri, Neoseiulus californicus), yielding the following: HR3 (4844 on‐targets, 790 off‐targets), ECR (8593 on‐targets, 507 off‐targets), HR4 (10 649 on‐targets, 2652 off‐targets), E74 (7640 on‐targets, 1275 off‐targets). Conserved fragments with minimal off‐target species risk were assembled into two fusion dsRNAs: dsfusion1 (ECR‐HR4‐HR3) and dsfusion2 (HR3‐E74‐ECR) (Fig. 3(B) and Table S4).

Figure 3.

Design and efficacy evaluation of fusion dsRNA for controlling Tetranychus urticae. (A) Workflow for fusion dsRNA design: Step 1, input mRNA sequences of TuECR, TuHR3, TuHR4, and TuE74; Step 2, compare sequences with homologs from four pest mite species and six beneficial organisms using 21 bp as the primary matching criterion, and 16 bp and 26 bp as secondary criteria; Step 3, select highly conserved, low off‐target species dsRNA fragments (151–200 bp); Step 4, assemble the fusion dsRNA. (B) Distribution of on‐ and off‐target sites for the four target genes. The x‐axis shows gene position. Red line (left y‐axis) indicates cumulative off‐target hits (16 and 26 bp) in beneficial organisms; blue line (right y‐axis) indicates on‐target hits (16, 21, and 26 bp). (C) Relative expression levels (mean ± standard error) of target genes in deutonymphs 24 h post‐injection. Asterisks indicate significant differences versus dsEGFP control (Student's t‐test, P < 0.01). (D) Survival (%) of mites following fusion dsRNA injections. (E) Molting (%) of mites following fusion dsRNA injections. (F) Representative phenotypes of deutonymph mites after fusion dsRNA injection. Data represent four biological replicates per treatment.

Injection of dsfusion1 reduced expression of TuHR4 (71%), TuHR3 (85%), and TuECR (87%), while dsfusion2 suppressed TuE74 (54%), TuHR3 (32%), and TuECR (82%) (Fig. 3(C)). Molting rates dropped from 91% in dsEGFP controls to 5.2% (dsfusion1) and 4.2% (dsfusion2) (Fig. 3(D),(E)). Phenotypically, 73% (dsfusion1) and 62% (dsfusion2) exhibited locomotor incapacity and aberrant quiescence, similar to dsTuECR‐treated mites, indicating effective disruption of upstream ecdysteroid signaling (Fig. 3(F)). Mortality reached 95% and 92%, confirming strong inhibition of development.

3.4. Toxicity bioassays of fusion dsRNA against spider mites

Graded injection assays revealed dose‐dependent effects. At 0.05625 ng, mortality of dsfusion1, dsfusion2, and dsTuHR3 was 31%, 27%, and 37%, respectively. At 0.09 ng, mortality increased to 47%, 57%, and 33%. At 0.1125 ng, mortality reached 75%, 66%, and 73%, and at 0.2250 ng, 73%, 77%, and 64%. Above 1.125 ng, mortality plateaued at approximately 90%. Logistic regression calculated median lethal concentration (LC₅₀) values of 0.069 ng (dsfusion1) and 0.066 ng (dsfusion2), lower than dsTuHR3 (0.105 ng); LC₉₀ values were 1.928 ng, 2.09 ng, and 2.145 ng, respectively (Fig. 4(A),(B)). These results demonstrate higher efficacy of fusion dsRNAs at lower doses.

Figure 4.

Mortality of Tetranychus urticae at different concentrations of dsfusion1, dsfusion2, and dsTuHR3, and corresponding toxicity regression. (A) Dose‐dependent mortality of T. urticae after treatment with dsfusion1, dsfusion2, or dsTuHR3 at graded concentrations. (B) Toxicity regression curves for dsfusion1, dsfusion2, and dsTuHR3, generated using SPSS software. Data represent mean ± standard deviation of four independent biological replicates per treatment.

Cross‐species assays showed that at LC₅₀ (0.0675 ng), T. evansi mortality was limited (17% and 20% for dsfusion1 and dsfusion2). At LC₉₀ (2.250 ng), dsfusion1 suppressed TeECR (19%), TeHR4 (42%), TeHR3 (54%), TcHR4 (50%), and TcECR (29%), while dsfusion2 reduced TeECR (72%), TeE74 (31%), TeHR3 (15%), TcE74 (48%), and TcECR (69%) (Supporting Information, Fig. S1). Phenotypically, dsfusion1 induced 80% mortality in T. evansi and 84% in T. cinnabarinus, whereas dsfusion2 caused 83% and 88% mortality, respectively (Fig. 5(A),(B)).

Figure 5.

Effects of fusion dsRNA injection on the survival of Tetranychus evansi and Tetranychus cinnabarinus. (A) Survival (%) of T. evansi and T. cinnabarinus following injection with dsfusion1, dsfusion2, or dsEGFP control. (B) Representative phenotypes of mites after dsRNA injection, showing lethal and molting‐defective outcomes. Data represent four biological replicates per treatment.

3.5. Efficacy of SIGS in controlling spider mites

Spray delivery of SPc‐encapsulated dsRNAs delayed molting in T. urticae and T. evansi (Fig. 6(A)). In T. urticae, molting at 48 h decreased to 35% (dsfusion1) and 41% (dsfusion2), with SPc further reducing rates to 25% and 27% and delaying molting by approximately 12 h (Fig. 6(B)). In T. evansi, molting at 60 h was 24% and 35%, further reduced to 18% and 22% with SPc, delaying molting by approximately 24 h.

Figure 6.

Effects of sprayed fusion dsRNA on the molting of Tetranychus urticae and Tetranychus evansi. (A) Schematic representation of the SIGS bioassay in spider mites. (B) Molting rate (%) of T. urticae and T. evansi after treatment with fusion dsRNA compared to dsEGFP control. (C) Relative expression levels (mean ± standard error) of target genes in deutonymphs of T. urticae and T. evansi at 24 h after spray treatment with fusion dsRNA or SPc‐encapsulated fusion dsRNA. The asterisks indicate a significant difference compared with the dsEGFP control (Student's t‐test, P < 0.01). Each value represents the mean of four biological replicates, each with three technical repetitions.

RT‐qPCR analysis confirmed that spray‐delivered fusion dsRNAs induced target‐gene knockdown in both mite species (Fig. 6(C)). In T. urticae, dsfusion1 reduced TuECR, TuHR3, and TuHR4 expression by 2%, 5%, and 22%, respectively, and SPc‐dsfusion1 increased the corresponding knockdown efficiencies to 28%, 19%, and 36%. Dsfusion2 reduced TuECR, TuHR3, and TuE74 by 2%, 21%, and 6%, respectively, while SPc‐dsfusion2 increased them to 36%, 16%, and 31%. In T. evansi, dsfusion1 reduced TeECR, TeHR3, and TeHR4 by 4%, 23%, and 5%, respectively, and SPc‐dsfusion1 increased them to 41%, 50%, and 8%. Dsfusion2 reduced TeECR, TeHR3, and TeE74 by 22%, 3%, and 16%, respectively, while SPc‐dsfusion2 increased them to 29%, 45%, and 25%. These data indicate that SIGS delayed molting and induced measurable target‐gene silencing, while SPc encapsulation generally enhanced RNAi efficacy, although the knockdown efficiency varied among target genes.

3.6. Non‐target species assessment in predatory mites of Neoseiulus californicus

In Neoseiulus californicus, sequence‐matching analysis identified eight and nine potential non‐target sites with 16‐bp contiguous matches for dsfusion1 and dsfusion2, respectively (Fig. 7(A)). NcMPP3 was selected as a representative potential non‐target indicator gene because its corresponding transcript, CL1641.Contig2, was matched by both fusion dsRNAs and contained two adjacent 16‐bp contiguous matching fragments in each construct.

Figure 7.

Biosafety evaluation of fusion dsRNAs in the predatory mite Neoseiulus californicus. (A) Sequence‐matching analysis of potential non‐target sites of fusion dsRNAs in Neoseiulus californicus. (B) Relative expression levels (mean ± standard error) of target genes in deutonymph mites at 24 h after dsRNA feeding. ‘ns’ indicates no significant difference between treatments (Student's t‐test, P > 0.05). (C) Survival (%) of mites after feeding on fusion dsRNA compared with dsEGFP control. (D) Representative phenotypes of Neoseiulus californicus following dsRNA feeding, showing no visible adverse effects. Data represent four biological replicates per treatment.

Feeding assays with fusion dsRNAs showed no significant effect on the potential non‐target species gene NcMPP3 in Neoseiulus californicus (Fig. 7(B)). Survival rates at 96 h were 85% and 88% in dsfusion1‐ and dsfusion2‐fed mites, respectively, compared with 90% in dsEGFP controls (Fig. 7(C),(D)), indicating minimal non‐target species effects and confirming the biosafety of fusion dsRNAs for this predatory mite.

4. DISCUSSION

RNAi technology represents an innovative and environmentally friendly approach for pest management, with its core principle lying in the rational selection of target genes. ⁴⁰ , ⁴¹ , ⁴² This study focused on the ecdysteroid signaling pathway, a highly conserved regulatory network critical for molting and development, to evaluate its potential as a target for RNAi‐mediated control of T. urticae. Our expression analyses revealed that TuE74, TuHR4, TuHR3, TuECR, and TuE93 are highly expressed during the molting phase, with distinct temporal patterns reflecting the progression of developmental quiescence and molting activity. The negative sawtooth expression pattern observed, which is typically a high transcription during quiescence followed by a decline post‐molt, is consistent with the temporally regulated ecdysteroid‐responsive gene expression conserved across arthropods. ¹⁹ , ²² , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ⁴² , ⁴³ , ⁴⁴ , ⁴⁵ , ⁴⁶ Functional RNAi experiments confirmed that silencing TuECR, TuHR3, TuHR4, and TuE74 induced distinct lethal phenotypes: inhibition of TuECR prevented molting initiation, knockdown of TuHR3 blocked old cuticle shedding, and dsRNA‐mediated silencing of TuE74 and TuHR4 resulted in dehydration and body shrinkage. These phenotypes are consistent with those reported for orthologous ecdysteroid signaling genes in other insects and spider mites. ⁹ , ²³ , ²⁷ , ³¹ , ⁴³ , ⁴⁴ , ⁴⁵ , ⁴⁶ By contrast, silencing of TuHR38/TuHR38b caused only molting delay, and knockdown of TuHR39 and TuE93 produced no observable phenotypes. ¹⁹ , ²² , ²³ , ⁴⁷ , ⁴⁸ In summary, these results highlight that TuECR, TuHR3, TuHR4, and TuE74 are core functional nodes in the ecdysteroid signaling pathway of T. urticae, whereas other homologs may play more minor, redundant, or species‐specific roles. It should be noted that the present study focused on functional and developmental phenotypes, and detailed ultrastructural analysis of molting defects was not undertaken.

Recognizing the potential of multi‐target RNAi, we further designed fusion dsRNAs that simultaneously target multiple core genes, a strategy aimed at enhancing interference efficacy and delaying the development of resistance. ³³ , ³⁴ In our experiments, both dsfusion1 and dsfusion2 effectively silenced multiple targets and induced high mortality rates, with LC₅₀ and LC₉₀ values lower than those observed for single‐gene dsRNA (dsTuHR3). By concurrently disrupting multiple critical nodes within the signaling pathway, multi‐target RNAi achieves a stronger developmental blockade, demonstrating superior advantages in enhancing silencing efficacy, reducing the required dose, and increasing lethal efficiency. Similar results have been reported in multi‐target studies on feeding‐related genes in Myzus persicae (Hemiptera: Aphididae) and detoxification genes in Nilaparvata lugens (Hemiptera: Delphacidae). ⁴⁸ , ⁴⁹ These studies collectively support the concept that multi‐target RNAi can more effectively disrupt key physiological processes such as molting, feeding, and detoxification, thereby improving pest mortality.

Among the two fusion dsRNAs tested, dsfusion1 outperformed dsfusion2 in both gene silencing efficiency and lethal toxicity. This difference reflects the importance of coordinated silencing: dsfusion1 achieved balanced and highly efficient suppression of all three target genes (TuHR4, TuHR3, TuECR) exceeding 71% efficiency, whereas dsfusion2 exhibited lower silencing of TuHR3 and TuE74, relying predominantly on TuECR knockdown. The variable silencing efficiency among targets within a fusion dsRNA may reflect differential processing of its segments into active siRNAs, a mechanism that merits future investigation. Thus, this finding suggests that, beyond selecting physiologically relevant targets, successful multi‐target RNAi likely requires careful design to ensure all target genes are effectively processed into siRNAs and silenced in concert, implying a key consideration for future RNAi product development.

An additional innovation of our study is the demonstration of broad‐spectrum potential. Leveraging conserved sequence homology across pest mites, dsfusion1 and dsfusion2 effectively induced molting failure in T. evansi and T. cinnabarinus. ⁷ , ⁵⁰ Simultaneously, ecological safety was confirmed, as neither dsRNA significantly affected the non‐target predatory mite Neoseiulus californicus. This aligns with previous studies showing minimal non‐target species effects in Propylea japonica (Coleoptera: Coccinellidae) and parasitic wasps, supporting the ecological compatibility of RNAi for IPM. ³⁴ , ⁴⁸ The combination of high cross‐species efficacy and target specificity represents a significant advance for RNAi‐based mite control, offering potential for deployment against multiple Tetranychidae species.

SIGS represents a promising field‐applicable delivery strategy, by which dsRNA enters the pest body primarily through cuticular penetration and oral ingestion. ¹⁴ , ⁵¹ , ⁵² In our study, SPc‐encapsulated fusion dsRNAs induced molting delay in both T. urticae and T. evansi, although complete blockage of molting was not achieved. Spray‐delivered fusion dsRNAs induced detectable knockdown of target genes in both mite species, and that SPc nanoparticles generally enhanced the silencing efficiency. However, the knockdown levels were relatively modest and varied among different target genes, which may explain why spray treatment mainly delayed molting rather than inducing the high mortality observed after microinjection. This contrasts with injection‐based RNAi and previous reports of more robust SIGS efficacy, likely reflecting limitations in dsRNA delivery. Key constraints include the cuticular barrier limiting penetration, the absence of SID‐1/2 homologs in T. urticae restricting systemic RNA uptake, and the high tissue‐specific expression of target genes in the central nervous system, which limits effective dsRNA access via feeding. ²⁸ , ⁴⁰ , ⁵³ , ⁵⁴ While SPc nanoparticles improve cuticular penetration, their ability to enhance cellular uptake remains limited. ⁵⁵ Alternative materials, such as chitosan, facilitate endocytosis but are constrained by acidic conditions, limiting practical applicability. ⁵⁶ These findings highlight a critical research gap: development of optimized delivery systems capable of overcoming integumentary barriers, enhancing cellular uptake, and maintaining biosafety is essential for field‐scale SIGS implementation. Emerging approaches include not only the development of novel nanocarriers but also the incorporation of modified nucleotides into dsRNA, which may improve dsRNA stability, nuclease resistance, and RNAi efficacy. ⁵⁷ , ⁵⁸ , ⁵⁹ Additionally, the feasibility of transgenic plants expressing fusion dsRNAs for continuous pest control warrants exploration as a complementary strategy for sustainable mite management.

In summary, this study systematically elucidates the core functions of ecdysteroid signaling pathway genes in the development of T. urticae and preliminarily validates the significant advantages of the multi‐target fusion dsRNA strategy in enhancing RNAi efficacy, broadening the control spectrum, and ensuring environmental safety. However, the most effective dsRNA delivery method in this study (microinjection) is not feasible for field application, while the more promising spray treatment only induced developmental delay, revealing the limitations of current delivery methods for field translation. Therefore, future research should focus on developing novel nanocarriers capable of effectively overcoming cuticular and gut barriers and improving cellular delivery efficiency, as well as exploring transgenic plant‐mediated sustained expression strategies to evaluate their long‐term field efficacy and ecological impact. Addressing these delivery and application bottlenecks is pivotal for transforming efficient RNAi technology from the laboratory into a practical green control tool for field use. Overall, this study provides meaningful advances in understanding mite molting biology and establishes a solid foundation for the applied use of RNAi‐based pest control strategies in sustainable agricultural systems.

CONFLICT OF INTEREST

The authors have no conflicts of interest to declare that are relevant to the content of this article.

AUTHOR CONTRIBUTIONS

Zhongyi L, XG and GL conceived the study and participated in its design. Zhou L and ZC performed the population of spider mites. Zhongyi L performed all of experiments. Zhongyi L, Zhuo L and GL analyzed the data. Zhongyi L, XG and GL wrote the initial draft. GS and Zhuo L edited the article. All authors have read and agreed to the published version of the article.

Supporting information

Figure S1. Relative expression levels of target genes in deutonymphs of Tetranychus evansi and Tetranychus cinnabarinus at 24 h after dsRNA injection. Mean (± SE) expression levels of target genes in deutonymph mites 24 h post‐injection with 2.25 nL of dsRNA (10 μg μL⁻¹). The asterisks indicate a significant difference compared with the dsEGFP control (Student's t‐test, P < 0.01). Each value represents the mean of four biological replicates, each with three technical repetitions.

Table S1. Primer sequences used in this study.

Table S2. All dsRNA sequence used in this study.

Table S3. Temporal expression levels of eight genes in Tetranychus urticae.

Table S4. Bioinformatic analysis of fusion dsRNA for targeting and specificity.

DATA AVAILABILITY STATEMENT

The data that supports the findings of this study are available in the supplementary material of this article.