Functional analysis of down‐regulated CYP6AE gene clusters involved in the insecticidal mechanism of lycorine against Spodoptera litura

Abstract

BACKGROUND

Plants have evolved abundant defensive secondary metabolites to resist insect herbivores. Lycorine is an alkaloid with insecticidal activity from Amaryllidaceae plants, which the destructive pest Spodoptera litura naturally avoids. Cytochrome P450 enzymes are central to xenobiotic detoxification in insects, but the mechanism by which lycorine acts against S. litura remains unknown. This study aimed to reveal the toxic mechanism of lycorine focusing on P450‐mediated detoxification.

RESULTS

Lycorine exhibited substantial toxicity to first‐instar S. litura larvae (LD₅₀ = 0.55 μg larva⁻¹). Subsequently, when fifth‐instar larvae were exposed to a sublethal dose (LD₃₀) of lycorine, Lyc disrupted metabolic pathways, damaged Malpighian tubules, and induced oxidative stress. Furthermore, lycorine strongly repressed a CYP6AE gene cluster (CYP6AE47, CYP6AE50, CYP6AE70, CYP6AE138 and CYP6AE139) and decreased total P450 activity to 45% in the Malpighian tubules. RNAi co‐silencing of these cluster genes increased larval mortality (+30%) under lycorine treatment. Finally, molecular docking and microscale thermophoresis analyses further confirmed direct binding between Lyc and this CYP6AE gene cluster, with the strongest affinity observed for CYP6AE47 (K_d = 518.5 nM). A key residue, ARG170, may be vital for the interaction between Lyc and CYP6AE47.

CONCLUSIONS

These results demonstrate that the insecticidal mechanism of Lyc involves suppressing the expression and function of a CYP6AE gene cluster, thereby impairing detoxification capacity, which leads to Lyc accumulation and larval mortality. Elucidation of the detoxification system‐targeted mechanism for this plant‐derived compound provides a foundation for developing novel, sustainable pest management strategies against S. litura and potentially other noctuid pests. © 2026 Society of Chemical Industry.

Lycorine, a promising botanical insecticidal compound, exerts its insecticidal action against Spodoptera litura by inhibiting P450 enzyme activity. This induces oxidative damage in the Malpighian tubule of larvae, which ultimately leads to larval growth retardation or mortality.

1. INTRODUCTION

The tobacco cutworm, Spodoptera litura, is a highly polyphagous global agricultural pest which feeds on over 120 crops. ¹ , ² , ³ Currently, the control of this insect relies primarily on conventional synthetic insecticides (e.g., chlorantraniliprole, indoxacarb, isoprocarb) and certain botanical inhibitors (e.g., berberine, azadirachtin). ⁴ , ⁵ , ⁶ , ⁷ , ⁸ Unfortunately, this pest, known for its remarkable adaptability, has developed significant resistance to many of these chemicals, often rendering them ineffective. ¹ , ⁹ , ¹⁰ Thus, eco‐friendly insecticides with novel modes of action are urgently needed as alternatives for the effective management of resistant populations. Among various natural candidates, plant secondary metabolites, particularly alkaloids, represent ‘treasuries’ of structurally diverse compounds with great potential for novel insecticide development.

Alkaloids represent a chemically diverse class of plant secondary metabolites with broad pharmacological and insecticidal properties. ¹¹ , ¹² , ¹³ , ¹⁴ Exemplifying these traits, the Amaryllidaceous plant Crinum asiaticum is toxic throughout its tissues and organs. Diverse pharmacological properties associated with anti‐inflammatory, antiviral and antitumor activities are all attributed to its alkaloid constituents, including lycorine. ¹⁵ , ¹⁶ Importantly, lycorine is considered to be the primary cause of the toxicity of C. asiaticum, which also exhibits antifeedant activity against most insects. ¹⁷ In this, lycorine (Fig. 1(A)) exhibits multiple deleterious biological and biochemical activities, including inhibition of protein synthesis, disruption of cellular redox balance, and cytotoxicity. ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ Previous studies demonstrate that lycorine suppresses feeding in the desert locust (Schistocerca gregaria), ¹⁶ synergizes with fenvalerate in the common aphid (Aphis gossypii), ²² and induces mortality in both larval and adult yellow fever mosquitoes (Aedes aegypti). ²³ It also exhibits excellent insecticidal activity against the diamondback moth (Plutella xylostella) with an LC₅₀ of 10.6 mg L⁻¹. ¹⁹ Additionally, S. litura naturally avoids Amaryllidaceae containing high levels of lycorine, ²⁴ and Lin et al. reported that lycorine impairs S. litura fitness by inhibiting secretory phospholipase A2 (sPLA2) expression and disrupting lipid homeostasis. ²⁵ PLA2 hydrolyzes membrane phospholipids to yield arachidonic acid (AA), which is metabolized via cyclooxygenase (COX), lipoxygenase (LOX) and cytochrome P450 monooxygenase pathways. ²⁶ Among these, the P450 pathway not only participates in AA metabolism but also plays a central role in the detoxification of xenobiotics. Intriguingly, the P450 enzyme system itself has been suggested as a target of lycorine analogues in mammals, where its activity is inhibited. ¹⁸ However, whether lycorine exerts any effect on P450 genes in S. litura remains unclear.

Figure 1.

Physiological and metabolic responses of Spodoptera litura larvae to lycorine exposure. (A) Chemical structure of lycorine. (B) Schematic diagram illustrating the workflow for detecting physiological and metabolic indicators in S. litura larvae exposed to lycorine. (C) Dose‐dependent effects of lycorine on growth and development of third‐instar larvae over 72 h. Higher concentrations caused greater developmental disruption and increased mortality. LD₃₀ and LD₅₀ values were calculated from mortality statistics. (D) Schematic diagram of S. litura larvae exposed to different medicinal reagents. (E) Morphological phenotypes of S. litura larvae after 72 h of treatment with different chemical reagents. Larvae were treated with berberine, isoprocarb, or lycorine, using an equal volume of DMSO treatment as the solvent control. Sky blue triangles denote larvae with inhibited development; magenta triangles denote dead larvae. (F) Larval mortality after exposure to tobacco leaves treated with different chemical reagents. (G) Trehalose and uric acid levels measured in hemolymph 24 h after feeding on an artificial diet containing 1.5 mg g⁻¹ lycorine (close to the LD₃₀ dose). (H) Uric acid content in frass under the same conditions as in (I). (I) Larval body‐weight gain, rates of frass production, frass water content, and water‐loss. AD, artificial diet without lycorine; AD+Lyc, artificial diet containing 1.5 mg g⁻¹ lycorine. Values are presented as mean ± SEM (n = 40). Asterisks indicate statistically significant differences (*** P < 0.001, Student's t‐test).

Cytochrome P450 monooxygenases play a central role in the detoxification capacity of insects, targeting both plant allelochemicals and synthetic insecticides. ⁴ , ²⁷ , ²⁸ , ²⁹ The diversity, tissue specificity, and substrate selectivity of these agents underpin insect adaptability to chemical stress. Examples include CYP321A8‐mediated insecticide resistance in the beet armyworm, Spodoptera exigua, ³⁰ CYP6AB14‐ and CYP6AB60‐mediated metabolism of plant allelochemicals in S. litura, ³¹ , ³² and reduced larval detoxification capacity after CRISPR/Cas9 knockout of the CYP9A gene cluster. ³³ Furthermore, CYP6AE subfamily P450s have been verified as core functional genes responsible for xenobiotic metabolism and pyrethroid insecticide detoxification in S. litura. ³⁴ Additionally, lycorine analogues inhibited mammalian CYP3A4 activity. ¹⁸ These observations raise the hypothesis that lycorine similarly targets P450 enzymes in S. litura.

To address this hypothesis, in the present study we systematically evaluated the insecticidal effects of lycorine on S. litura using an integrated approach which included metabolic profiling, transcriptomic analysis, microscale thermophoresis (MST), and enzyme‐linked immunosorbent assay (ELISA). Our results reveal that lycorine induces accumulation of reactive oxygen species (ROS), alters P450 gene expression in larval Malpighian tubules, and suppresses P450 enzymatic activity. These changes compromise nutrient utilization, impair larval growth, and cause structural damage to critical detoxification tissues. Collectively, our findings provide mechanistic insights into the insecticidal action of lycorine, offer a molecular basis for identifying key toxicity‐related target genes for the control of noctuid pests, and establish a foundation for developing this alkaloid as a plant‐derived pesticide with potential applications in integrated pest management (IPM).

2. MATERIALS AND METHODS

2.1. Insect rearing and lycorine treatment

S. litura larvae were obtained from Henan Keyun Biological and maintained as previously described. ³⁵ Larvae were reared on an artificial diet (Nosanko, Japan) in a climate‐controlled chamber at 27 ± 1 °C, 60 ± 10% relative humidity (RH), with a photoperiod of 16 h light followed by 8 h darkness, as reported. ¹ , ³ , ³⁵ Adult moths were supplied with 10% sucrose. All rearing conditions were designed to minimize physiological variation prior to experimental treatments.

2.2. Cell culture

Spli221 adherent cells, originally derived from S. litura pupal ovaries ³⁶ and provided by Hunan Agricultural University, were cultured in Sf‐900™ II SFM medium (Thermo Fisher Scientific, USA) supplemented with 10% heat‐inactivated FBS (Gibco, USA), 100 U mL⁻¹ penicillin (Gibco, USA), and 100 μg mL⁻¹ streptomycin (Gibco, USA). Cells were maintained at 27 °C without CO₂ supplementation.

2.3. Bioassay

Lycorine, isoprocarb and berberine toxicities were evaluated using a diet‐incorporation bioassay. Lycorine (≥98%, MCE, USA) was dissolved in 0.1% DMSO and diluted in distilled water containing 0.05% Tween‐80 and 0.1% Triton X‐100. Berberine (>98%, Shanghai yuanye Bio‐Technology, Shanghai, China) and isoprocarb (Yeasen, Shanghai, China) were dissolved in 0.1% DMSO and diluted in distilled water containing 0.1% Triton X‐100. First‐day third‐instar larvae were transferred to treated diets (20 larvae per dish, three replicates) at final lycorine concentrations of 0 (solvent control), 0.5, 1.0, 2.5, 5.0, 7.5, 10.0, and 15.0 mg g⁻¹, isoprocarb concentrations of 0 (solvent control), 150, 225, 300, 375, 450, 600 and 900 μg g⁻¹，and berberine concentrations of 0 (solvent control), 2.5, 5.0, 7.5, 10.0, 12.5, and 15.0 mg g⁻¹. Insecticide concentrations were based on preliminary tests. Mortality was recorded after 72 h; larvae unresponsive to probing were scored as dead. LD₃₀ and LD₅₀ values were calculated following POLO Plus 2.0. ³⁷ , ³⁸

For insecticide testing, a standard leaf‐dip bioassay was performed on third‐instar S. litura larvae. ³⁹ Insecticide (lycorine, berberine and isoprocarb) stock solutions were prepared in DMSO and diluted with sterile water containing 0.1% Triton X‐100. Tobacco leaf discs (3–6 cm²) were dipped in test solutions for 10 s, air‐dried at room temperature for 10 min, and placed in clean Petri dishes. Control discs were treated with DMSO +0.1% Triton X‐100 solution. Leaves were replaced as needed upon observing they were severely dehydrated. Each dish received 15 newly molted third‐instar larvae (5–6 mg). Mortality was assessed after 72 h, with larvae considered dead if unresponsive to gentle prodding with a duck‐hair brush. ³⁹

2.4. Analysis of lycorine effects on larval excretion, water balance, trehalose and uric acid homeostasis

To evaluate effects on metabolic and excretory homeostasis, fifth‐instar larvae were fasted for 2 h, weighed (W₀), and assigned to lycorine‐treated diet (AD+Lyc, 1.5 mg g⁻¹; n = 40) or untreated diet (AD; n = 40). Individual larvae were housed in separate 25‐cm diameter Petri dishes. Frass produced over a 24 h period was collected and its fresh weight (W₁) recorded along with larval body mass at the end of the feeding period (Wt). Frass was then dried at 60 °C to a constant mass, and the dry weight (W₂) recorded. Calculated metrics included: (i) Body‐weight gain (%) = (W_t ‐ W₀) / W_t × 100; (ii) Frass water content (%) = (W₁ ‐ W₂)/W₁ × 100; and (iii) Frass production rate (mg dry/g/24 h) = W₂/W₀/24. The protocol was adapted from Babamir‐Satehi ⁴⁰ and Mousavi. ⁴¹

Initial body water mass (Q₁) was recorded after frass collection. Larvae were maintained on dry filter paper at 25 °C and 50% RH without food or water. After 6 h body mass (Q_t) was measured to calculate water loss rate as (%/h) = (Q₁ ‐ Q_t)/Q₁/6. After 24 h of feeding, hemolymph was collected to measure trehalose levels using the CheKine™ Micro Trehalose Assay Kit (Abbkine, China) and uric acid levels with the Elabscience Colorimetric Uric Acid Assay Kit (Elabscience, China). Frass uric acid levels were quantified from homogenized and centrifuged samples. All data are reported as mean ± SEM and analyzed by the Student's t‐test (P < 0.05).

2.5. Hematoxylin–eosin (HE) staining, fluorescence staining and scanning electron microscopy (SEM) of larval tissues

Larvae (n = 20) were fed AD+Lyc (1.5 mg g⁻¹) or AD for 72 h. Midgut and Malpighian tubules were dissected, rinsed in PBS, and processed for histological analysis. Tissues were fixed in 4% PFA (Beyotime, China) for 24 h at 4 °C. For HE staining, midgut and Malpighian tubule samples underwent ethanol dehydration, xylene clarification, paraffin embedding, and microtome sectioning. Sections (5 μm) sliced from the paraffin blocks were stained with hematoxylin–eosin; sample preparation was then completed by LiLai Biomedicine (Cheng Du, China). For fluorescent staining, Malpighian tubules were treated with DAPI (1 μg mL⁻¹) (Beyotime, China) and Alexa Fluor™ 488‐phalloidin (Invitrogen, USA), and then observed and photographed using light microscopy (BX53, Olympus, Tokyo, Japan). For SEM, Malpighian tubules were dehydrated, dried with tert‐butanol, sputter coated with gold, and imaged using a Hitachi SEM SU3500 at 10 kV. ⁴²

2.6. Determination of H₂O₂, MDA, antioxidant enzyme and P450 enzyme activities

Fifth‐instar larvae were fed AD or AD+Lyc (1.5 mg g⁻¹) for 72 h. Malpighian tubules were dissected, washed in PBS, and stored at −80 °C. Tissue homogenates were prepared in PBS and centrifuged; supernatants were used to quantify H₂O₂ (G0168W), malondialdehyde (MDA, G0110W), total antioxidant capacity (T‐AOC, G0142W), and activities of superoxide dismutase (SOD, G0101W), catalase (CAT, G0106W) and peroxidase (POD, G0107W) using commercial kits (Suzhou Grace Biotechnology, China) according to published protocols. ⁴³ , ⁴⁴ , ⁴⁵ , ⁴⁶ Protein concentrations were determined using a BCA assay (Beyotime, China).

P450 activity was determined using a P450 activity assay kit (GY0934W48; Suzhou Grace Biotechnology, China) following the manufacturer's instructions. This assay kit is based on the classic O‐deethylation reaction catalyzed by cytochrome P450. Using 7‐ethoxycoumarin (7‐EC) as the fluorogenic substrate, cytochrome P450 activity is directly determined by measuring the increase in fluorescence resulting from the formation of the O‐deethylated product, 7‐hydroxycoumarin (7‐HC). Protein was extracted from 0.1 g of Malpighian tubule homogenate, and the assay was performed under a constant protein concentration by monitoring the alteration in absorbance at 390 nm. P450 activity was expressed as nmol min⁻¹g⁻¹protein.

2.7. RNA extraction and transcriptome sequencing

Malpighian tubules from 3rd day fifth‐instar larvae were dissected and washed in PBS (pH 7.2), frozen in liquid nitrogen, and stored in RNAiso Plus (Takara). RNA was extracted using a MAGEN HiPure Stool RNA Kit. Poly(A) + mRNA was enriched with Oligo(dT) beads and sequenced on the BGI MIGSEQ2000 platform. ⁴⁷ Raw reads were filtered with SOAPnuke, ⁴⁸ and clean reads were aligned to the S. litura genome (NCBI SRP095695) using Bowtie2 v2.4.5. ⁴⁹ Gene expression levels were quantified as FPKM using RSEM v1.3.1. ⁵⁰ Differentially expressed genes (DEGs) were identified using |log₂FC| ≥ 2 and FDR < 0.05. ⁵¹ GO and KEGG enrichment analyses were conducted using the Phyper hypergeometric test (q ≤ 0.05) and visualized with OmicShare.

2.8. Cloning and bioinformatic analysis of CYP6AE cluster genes

Total RNA was extracted using RNAiso Plus and reverse‐transcribed using the Hifair® III cDNA Synthesis Kit (Yeasen, Shanghai, China). Subsequently, 1 μg of the qualified total RNA was reverse‐transcribed into first‐strand cDNA using the Hifair® III cDNA Synthesis Kit. ³⁵ The gene‐specific F/R primers (Supporting Information, Table S4) for cloning the CYP6AE cluster genes were generated using Primer 5.0 (https://primer5.ut.ee/). The 2× Hieff Canace® PCR Master Mix (Shanghai, China) was employed to perform the reverse transcription PCR (RT − PCR) following the manufacturer's protocol. Subsequently, a Universal DNA Purification Kit (Tiangen; Beijing) was employed to purify the CYP6AE cluster genes fragments, which were ligated into the gene cloning vector (pEASY®‐Blunt Zero Cloning) and further transfected into Trans1‐T1 phage resistant chemically competent cells (TransGen Biotech) for sequencing by Sangon Biotech (Sangon, China). Finally, P450 protein sequences with complete ORFs of S. litura (NCBI: SRP095695) and other species were obtained from the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/) and were aligned with ClustalW. ¹ , ⁴²

RT‐qPCR ⁵² , ⁵³ reactions contained 2 μL of cDNA, 0.8 μL of gene‐specific primers (Supporting Information, Table S4), 10 μL of Hieff® SYBR Green Master Mix (Yeasen), and ddH₂O to 20 μL. All RT‐qPCR procedures were performed on a 7500 Fast Real‐Time PCR System (ABI, CA, USA) with the following PCR program: 95 °C for 3 min; 40 cycles of 95 °C for 10 s and 60 °C for 30 s, and a dissociation stage of 95 °C for 15 s and 60 °C for 1 min. ⁵⁴ RPL7A and RPS3 (KC866374) served as reference genes, ³⁵ and mRNA relative expression was calculated via the 2^−ΔΔCt method. Primer amplification efficiency ⁵² , ⁵³ in Supporting Information, Fig. S3. Each treatment included three biological replicates.

2.9. RNAi bioassay

dsRNA targeting candidates for CYP6AE genes were synthesized from RT‐PCR templates containing T7 promoters (Supporting Information, Table S4) using the T7 High Yield RNA Synthesis Kit (Yeasen), as previously reported. ³⁵ Third‐instar larvae were starved for 2 h prior to injection. Approximately 5 μL of dsRNA (1500 ng μL⁻¹) was injected into the hemolymph; DEPC‐water and dsEGFP served as controls. A second injection was performed 24 h later. Larvae recovered for 2 h before exposure to AD+Lyc. Each treatment included three replicates of 20 larvae. Malpighian tubules were sampled after 48 h for confirmation of knockdown efficiency using RT‐qPCR.

2.10. Construction of P450‐EGFP fusion plasmids

The pSL1180 vector containing the OpIE2 promoter was digested with BamHI/XmaI. Target P450 coding sequences and EGFP were amplified with primers containing homologous arms and linker sequences (Supporting Information, Table S5). Fusion constructs were generated by overlap RT‐PCR and inserted via homologous recombination as previously described. ⁴⁰ , ⁴¹ Positive clones were verified by Sanger sequencing and used for preparation of endotoxin‐free plasmids.

2.11. Transfection and expression analysis

Spli221 cells were transfected with pSL1180‐P450‐EGFP plasmids using X‐tremeGENE HP Transfection Reagent (Roche) as previously reported. ⁵⁵ , ⁵⁶ Cells were seeded in 25‐cm² culture flasks (5 × 10⁶ cells) and incubated overnight. Subsequently, the plasmids were transfected into Spli221 cells, with a total dosage of 10 μg per culture flask. Each plasmid was tested in triplicate. Fluorescence was monitored 72 h post‐transfection, after which cell lysates were collected for RT‐qPCR, western blotting, and MST analyses. Western blotting was performed to verify P450 protein overexpression using a GFP antibody (Thermo Fisher Scientific) following established procedures. ⁵⁵ , ⁵⁶

2.12. MST and P450 content assays

MST assays were conducted according to established protocols. ⁵⁷ , ⁵⁸ , ⁵⁹ Spli221 cells expressing EGFP or P450‐EGFP fusions (CYP6AE138, CYP6AE139, CYP6AE70, CYP6AE47, and CYP6AE50) were harvested 72 h post‐transfection, lysed in RIPA buffer with protease inhibitors, and centrifuged. Protein concentration was determined using a BCA assay (Beyotime, China).

For MST, lycorine (3 mM stock in DMSO) was serially diluted 1:2 to produce 16 concentrations. Each reaction contained 19 μL of lysate and 1 μL of lycorine, incubated for 15 min at 25 °C. Each untreated protein sample had an initial fluorescence value from 500 to 600 units. Samples were loaded into Monolith NT.115 capillaries (470 nm excitation; LED 40%) for monitoring fluorescence signals. Binding curves (F_norm = F_hot / F_cold) were generated using NT Analysis software (v1.2.231, NanoTemper Technologies, Munich, Germany). All assays were performed in triplicate.

For assays of P450 content, 2 μL of lycorine (6 mM stock) was mixed with 198 μL of cell lysate and incubated overnight at 4 °C; equivalent volumes of DMSO served as controls. Protein concentrations (0.5 μg mL⁻¹) were equalized across treatments. Cytochrome P450 levels were quantified using insect‐specific ELISA kits (Boshen Biotechnology; BS‐E19103O1) according to the manufacturer's instructions. Data were normalized to the DMSO control (set as 1.0) to minimize baseline variations and avoid negative values derived from assay background noise. Each condition included three replicates.

2.13. Molecular modeling and docking

P450 protein sequences (SRP095695) were used to predict 3D structures using AlphaFold3, ⁶⁰ , ⁶¹ with heme (an essential prosthetic group of P450s) incorporated into the active site of the protein during modeling. Lycorine docking was performed with AutoDock Tools 1.5.6. ⁵⁸ Protein‐ligand interactions were visualized and analyzed using PyMOL 2.4. ⁵⁸

2.14. Statistical analysis

Statistical analyses were performed with GraphPad Prism 8.0. Differences in RT‐qPCR expression, oxidative stress markers (H₂O₂, MDA), antioxidant enzyme activities, and RNAi knockdown efficiency were evaluated using Student's t‐tests or one‐way ANOVA, with significance thresholds of P * < 0.05, ** P < 0.01, or *** P < 0.001. LD₃₀ and LD₅₀ values were calculated using POLO Plus 2.0.

3. RESULTS

3.1. Lycorine exposure affects the growth and development of S. litura

To investigate the effects of lycorine on the growth and development of S. litura, we assessed its toxicity across different larval instars. The median lethal dose (LD₅₀) for highly susceptible first‐instar larvae was 0.55 μg larva⁻¹ (Table 1). In contrast, third‐instar larvae, which possess a more developed detoxification system, exhibited markedly higher tolerance with LD₃₀ and LD₅₀ values of 65 μg larva⁻¹ and 86 μg larva⁻¹ (Fig. 1(C), Table 2). Berberine is a multi‐target botanical pesticide with direct antimicrobial/insecticidal activity and synergistic effects when combined with other pesticides. ⁶² , ⁶³ , ⁶⁴ Berberine exhibited 3‐fold and 3‐fold higher LD₃₀ and LD₅₀ values of 194 μg larva⁻¹ and 261 μg larva⁻¹ against S. litura (Table 2), indicating that Lycorine exhibited stronger toxicity against S. litura larvae than berberine. The conventional pesticide isoprocarb exhibited higher toxicity against S. litura than lycorine, with a 2‐fold lower lethal dose of 31 μg larva⁻¹ and a 2‐fold lower median lethal dose (LD₅₀) of 43 μg larva⁻¹ (Table 2). In a feeding experiment with third‐instar S. litura larvae involving treatment with lycorine and berberine at the equivalent LD₅₀ dose of isoprocarb, larvae in the DMSO control group grew normally, the isoprocarb group showed over 50% mortality, the lycorine group exhibited more than 30% mortality with half of the larvae displaying developmental delay, and the berberine group only showed growth inhibition without mortality (Fig. 1(D)–(F)).

Table 1.

Toxicity of lycorine to first‐instar Spodoptera litura larvae

Compound	LD30 (μg larva−1) (95% CI)	LD50 (μg larva−1) (95% CI)	Slope ± SE	χ2	P
Lycorine	0.38 (0.17–1.07)	0.55 (0.19–1.52)	2.615 ± 0.478	1.56	0.803

Note: LD₃₀, 30% lethal dose; LD₅₀, 50% (median) lethal dose; CI, confidence interval; χ2, Chi‐square value and degrees of freedom.

Table 2.

Toxicity of lycorine, isoprocarb and berberine to third‐instar S. litura larvae

Compound	LD30 (μg larva−1) (95% CI)	LD50 (μg larva−1) (95% CI)	Slope ± SE	χ 2	P
Lycorine	65 (55–78)	86 (75–99)	2.052 ± 0.098	1.56	0.814
Isoprocarb	31 (24–39)	43 (33–53)	4.086 ± 0.168	5.12	0.275
Berberine	195 (139–272)	261 (187–364)	4.126 ± 0.436	4.12	0.390

Note: LD₃₀, 30% lethal dose; LD₅₀, 50% (median) lethal dose; CI, confidence interval; χ ², Chi‐square value and degrees of freedom.

Accordingly, fifth‐instar larvae were fed an artificial diet containing 1.5 mg g⁻¹ lycorine (approximating the LD₃₀ for third‐instar larvae) for 24 h. This treatment significantly increased levels of hemolymph trehalose (+46%) and uric acid (+58%) (P < 0.05; Fig. 1(G)), whereas uric acid content in frass decreased modestly (−13%) (Fig. 1(H)). Concurrently, frass moisture (−28%) and production rate (−30%) were significantly reduced (P < 0.05; Fig. 1(I)). Subsequently, following lycorine exposure the body water‐loss rate decreased by 37% with a concurrent 37% reduction of body mass (Fig. 1(I)).

3.2. Lycorine induced structural damage in Malpighian tubules

Given that exposing insects to toxic substances such as plant secondary metabolites typically induces primary damage in the midgut, ⁶⁵ , ⁶⁶ and considering the potential involvement of Malpighian tubules in trehalose and uric acid metabolism, we conducted histological examination of both tissues using HE staining. HE staining results revealed that the midgut cells of lycorine‐treated larvae were essentially normal, being elliptical in shape, tightly arranged, and featuring numerous microvilli. Moreover, goblet and columnar cells were distinctly visible and surrounded by a well‐defined peritrophic membrane similar to the control (AD) group (Fig. 2(A)). By contrast, after lycorine exposure, the Malpighian tubules exhibited marked epithelial cell vacuolization compared to the tightly and structurally intact cells in the AD group (Fig. 2(A)).

Figure 2.

Lycorine‐induced structural damage in Malpighian tubules. (A) Hematoxylin–eosin staining of Malpighian tubules and midguts in S. litura after 72 h of feeding on AD with/without lycorine (Lyc). (B) DAPI/phalloidin staining of Malpighian tubules after 72 h exposure to lycorine. Nuclei are stained with DAPI (blue), and F‐actin filaments are visualized with phalloidin (red). Scale bar: 200 μm. (C) SEM micrographs of Malpighian tubule surfaces after 72 h of lycorine treatment. Arrowheads indicate structural alterations. Magenta highlights damaged areas of the Malpighian tubules; blue indicates associated bacteria. (D) Oxidative stress markers (H₂O₂, MDA) and antioxidant enzyme activities (SOD, POD, CAT, T‐AOC) in Malpighian tubules following lycorine exposure. AD, artificial diet without lycorine; AD+Lyc, artificial diet containing 1.5 mg g⁻¹ lycorine; CC, columnar cell; GC, goblet cell; Lu, lumen; MV, microvilli; Bl, basal lamina. Values are mean ± SEM. Asterisks indicate statistically significant differences (*** P < 0.001, Student's t‐test).

DAPI staining revealed that lycorine‐treated tubules contained an increased number of smaller, condensed nuclei compared with controls, and phalloidin staining revealed disorganized and ruptured cytoskeletal filaments (Fig. 2(B)). SEM images corroborated these findings: control tubules exhibited smooth surfaces with clear longitudinal ridges, whereas lycorine‐treated tubules appeared rough, irregular, and perforated, with occasional bacterial colonization (Fig. 2(C)).

3.3. Lycorine induces alterations in oxidative stress in the Malpighian tubules

To assess whether lycorine disrupts redox homeostasis, oxidative stress markers were measured in Malpighian tubules of fifth‐instar larvae 72 h post‐treatment. Compared with the control group, lycorine exposure caused a marked redox imbalance (Fig. 2(D)). Activities of SOD and CAT increased significantly (+35% and + 28%, respectively), whereas POD activity declined sharply (−40%). Concurrently, oxidative damage markers were elevated, with H₂O₂ increasing by 16% and MDA by 27%, whereas total antioxidant capacity (T‐AOC) decreased by 24%.

3.4. Lycorine downregulates the expression of a CYP6AE gene cluster containing CYP6AE138 , CYPAE47 , CYP6AE50, CYP6AE70 , and CYP6AE139 genes

To investigate the molecular mechanisms underlying the structural abnormalities of Malpighian tubules caused by lycorine in S. litura, we performed transcriptomic analysis of Malpighian tubules dissected from fifth‐instar larvae which had been fed AD+Lyc (1.5 mg g⁻¹) for 72 h (Fig. 3). DESeq2 analysis identified 2525 DEGs between lycorine‐treated and control tissues (adjusted P < 0.05, |log₂FC| ≥ 2), including 977 upregulated and 1267 downregulated genes (Fig. 3(B)), indicating that lycorine exposure predominantly inhibited transcriptional activity.

Figure 3.

RNA‐seq analysis revealed candidate targets of lycorine in the Malpighian tubules of S. litura. (A) Schematic overview of RNA‐Seq analysis from third‐instar larvae exposed to LD₃₀ of lycorine. Larvae fed artificial diet (AD) without lycorine served as controls. (B) Volcano plot of differentially expressed genes (DEGs) in Malpighian tubules. Pink dots indicate upregulated genes, light blue dots indicate downregulated genes, and grey dots represent non‐significant genes. (C) Heatmap showing detoxification‐related gene expression in Malpighian tubules following lycorine treatment. (D) RT‐qPCR validation of 11 DEGs identified by RNA‐Seq, to confirm transcriptional induction. (E) Gene Ontology (GO) classification of DEGs in Malpighian tubules; the top 20 categories are shown. (F) KEGG pathway enrichment analysis of DEGs in Malpighian tubules; the top 20 categories are shown. Circle size indicates the number of genes, red color of the circle indicates significant enrichment, blue indicates insignificant enrichment. Numerical values within circles denote DEG counts mapped to respective pathways. (G) Specific activity of P450 enzymes in Malpighian tubules after lycorine exposure. (H) Agarose gel electrophoresis of CYP6AE138, CYP6AE70, CYP6AE47, CYP6AE50, and CYP6AE139. Band sizes (from left to right): 1404 bp, 1431 bp, 1401 bp, 1563 bp, and 1368 bp. M: DNA marker (from top to bottom: 2000, 1000, 750, 500, 250). (I) Genomic organization of the five tandemly aligned CYP6AE gene cluster (CYP6AE138, CYP6AE70, CYP6AE47, CYP6AE50, and CYP6AE139) on chromosome 20, highlighting their close physical proximity. AD, artificial diet (control); Lyc, artificial diet containing 1.5 mg g⁻¹ lycorine; mt, Malpighian tubules. Values are presented as mean ± SEM. Statistical significance was determined using a Student's t‐test (*** P < 0.001).

RT‐qPCR validation of 11 representative DEGs confirmed expression trends consistent with RNA‐Seq results, supporting the reliability of the transcriptomic data (Fig. 3(C), (D)). Systematic analysis was performed on the top 20 enriched terms of GO function and KEGG pathway to identify the biological functions and pathways affected by lycorine. GO terms revealed significant enrichment in molecular functions such as catalytic activity, monooxygenase activity, and heme binding, together with biological processes related to small molecule metabolism (Fig. 3(E), Supporting Information, Fig. S1B). Similarly, KEGG pathway enrichment analysis highlighted key metabolic pathways which included xenobiotic metabolism by cytochrome P450, glutathione metabolism, drug metabolism‐cytochrome P450, other drug metabolism enzymes, and oxidative phosphorylation (Fig. 3(F), Supporting Information, Fig. S1C).

Among the DEGs, cytochrome P450 genes CYP6AE138, CYP6AE47, CYP6AE50, CYP6AE70, and CYP6AE139 were significantly downregulated in response to lycorine (Fig. 3(C), (D)). Furthermore, enzymatic activity assays confirmed that lycorine treatment suppressed total P450 activity in the Malpighian tubules by approximately 45% (Fig. 3(G)). Genomic search analysis indicated that the five genes are tightly clustered in a narrow region on chromosome 20 in S. litura, arranged in tandem repeats with intergenic distances of only 0.3–0.8 kb (Fig. 3(I)). Sequence alignment further confirmed the high sequence homology among these five tandemly clustered genes, showing an overall sequence identity of 73.36% (Supporting Information, Fig. S2). Based on the published S. litura genome assembly, these genes cluster in a distinct clade within the CYP6AE subfamily of the noctuids and show notable expansion relative to the CYP6AE9 gene of B. mori, ¹ which is oligophagous.

3.5. RNAi silencing of P450 genes enhances lycorine toxicity

To evaluate the functional role of the five P450 genes, newly molted third‐instar larvae were injected with gene‐specific dsRNA, followed by a second injection 24 h later. Silencing individual P450 genes increased larval susceptibility to lycorine, with each individual gene knockdown group exhibiting higher sensitivity than the dsEGFP control (Fig. 4(B), (C)). Specifically, individual knockdown of the five P450 genes in the cluster (CYP6AE138, CYP6AE47, CYP6AE50, CYP6AE70, and CYP6AE139) resulted in consistent reductions in body weight (36–50%), decreased food consumption (12–25%), and prolonged larval duration (30–46%) (Supporting Information, Fig. S4A–C). No significant differences in mortality were observed between individual knockdowns and the EGFP control (Supporting Information, Fig. S4D). Knockdown efficiency was confirmed by RT‐qPCR (Fig. 4(B), (C)). Moreover, RNAi‐mediated silencing of a single gene had no significant effect on the expression levels of other related genes in the gene cluster (Supporting Information, Fig. S5).

Figure 4.

Silencing P450 genes increased the susceptibility of S. litura to lycorine. (A) Schematic diagram of the RNAi assay. Third‐instar larvae were injected with dsRNA targeting individual or multiple P450 genes. (B) RNAi‐mediated knockdown of individual genes (CYP6AE138, CYP6AE70, CYP6AE47, CYP6AE50, and CYP6AE139) significantly increased the susceptibility of larvae to lycorine and inhibited their growth and development. (C) Silencing efficiencies of individual P450 genes as determined by RT‐qPCR. (D) Representative molting defects observed in larvae at 48 h after RNAi targeting of the P450 gene cluster. (E) Effects on larval development of co‐silencing the five P450 genes and mortality rate. (F) Knockdown efficiency of the five P450 genes after co‐silencing, measured by RT‐qPCR. dsP450 mix, combined dsRNAs targeting CYP6AE138, CYP6AE70, CYP6AE47, CYP6AE50, and CYP6AE139. Values are presented as mean ± SEM. Statistical significance was determined by Student's t‐test (*** P < 0.001).

In contrast, compared with the EGFP group, simultaneous co‐silencing of all five P450 genes markedly increased lycorine sensitivity and reduced food intake (−35%) and body mass (−60%), even though the instar duration was prolonged (+25%). This resulted in elevated mortality (+30%) due to molting failure (Fig. 4(D), (E)). At 48 h post‐second injection, RT‐qPCR confirmed knockdown efficiencies of CYP6AE138 (−69%), CYP6AE47 (−54%), CYP6AE50 (−72%), CYP6AE70 (−78%), and CYP6AE139 (−80%) (Fig. 4(F)). RNAi also substantially lowered by slightly more than half the LD₃₀ from 65 to 31 μg larva⁻¹ and the LD₅₀ from 86 to 44 μg larva⁻¹ (Table 2, Supporting Information, Table S1).

3.6. Lycorine directly bound to CYP6AE47, CYP6AE50, CYP6AE70, and CYP6AE139 proteins in vitro

To explore further functional relationships within the gene cluster, CYP6AE70, CYP6AE47, CYP6AE50, and CYP6AE139 were overexpressed individually in Spli221 cells. This resulted in substantial transcript increases relative to EGFP controls: 6.92 × 10⁴‐fold (CYP6AE70), 5.05 × 10³‐fold (CYP6AE47), 2.87 × 10⁴‐fold (CYP6AE50), and 1.31 × 10⁵‐fold (CYP6AE139) (Fig. 5(B)). RT‐qPCR analysis indicated that overexpression of any single gene induced expression of the others, suggesting strong cross‐regulation and the formation of a coordinated detoxification unit (Supporting Information, Fig. S6).

Figure 5.

Lycorine directly binds to the CYP6AE gene cluster proteins in vitro. (A) Schematic overview of RT‐qPCR and MST assays in Spli221 cells. (B) Overexpression of CYP6AE70, CYP6AE47, CYP6AE50, and CYP6AE139 in Spli221 cells, confirmed by RT‐qPCR. (C) Western blot analysis of P450 protein overexpression in Spli221 cells. Lane 1: EGFP (negative control); Lane 2: CYP6AE70; Lane 3: CYP6AE47; Lane 4: CYP6AE50; Lane 5: CYP6AE139. Black arrows indicate the positions of EGFP (bottom) and P450‐EGFP (upper) fusion proteins. Immunoblotting with specific antibodies confirmed successful overexpression. (D) Quantification of the binding affinity of lycorine to P450‐EGFP fusion proteins using MST. EGFP alone served as a negative control (Supporting Information in Fig. S7). Values are presented as mean ± SEM. Statistical significance was determined by a Student's t‐test (*** P < 0.001).

To quantify the binding affinity between lycorine and P450 enzymes, MST was performed using P450‐EGFP fusion proteins expressed in Spli221 cells. Western blot analysis confirmed successful overexpression of all four P450 proteins compared with the EGFP control (Fig. 5(C)). Binding affinity was assessed by incubating lysates containing P450‐EGFP fusion proteins with a series of lycorine dilutions while maintaining a constant concentration of the fluorescent fusion protein. Lycorine bound to all four P450 isoforms with varying affinities: CYP6AE70 (Kd = 63.8 μM), CYP6AE47 (K_d = 518.5 nM), CYP6AE50 (K_d = 41.2 μM), and CYP6AE139 (K_d = 85.6 μM) (Fig. 5(D)). EGFP alone exhibited no detectable binding, confirming specificity of the assay ⁶⁷ (Supporting Information, Fig. S7).

3.7. Molecular docking

To explore the structural basis of lycorine binding, molecular docking was performed using high‐confidence 3D models of CYP6AE70, CYP6AE47, CYP6AE50, and CYP6AE139 proteins generated with AlphaFold3 (ipTM scores >0.8; Supporting Information, Fig. S8; Table S2). Docking simulations in AutoDock Tools predicted the binding of lycorine within the active sites of each P450 isoform, yielding binding energies of −7.7 (CYP6AE70), −7.9 (CYP6AE47), −7.9 (CYP6AE50), and − 8.6 kcal mol⁻¹ (CYP6AE139) (Fig. 6(A)–(D)). These results suggested that lycorine acts as a specific ligand for these proteins.

Figure 6.

Molecular docking, key binding sites, and inhibitory effects of lycorine on CYP6AE70, CYP6AE47, CYP6AE50, and CYP6AE139 proteins. (A) Predicted 3D structures of SlituP450 proteins and molecular docking simulations with lycorine. Enlarged views of lycorine binding sites in SlituP450‐073/074/075/076 are shown in dashed frames. Potential π‐π T‐shaped (magenta), alkyl (yellow), and hydrogen bond (blue) interactions are represented by dashed lines. Light teal represents the protein, light pink the lycorine molecule, and marine the heme molecule. Red triangles indicate π‐π T‐shaped interaction sites. (B) Sequence alignment of key amino acid residues involved in lycorine binding. Conserved residues among noctuid species in CYP6AE47 are highlighted in purple, light blue, cyan, and brown. Numbers indicate the position of the corresponding amino acid residues. (C) Spli221 cells were transfected to express EGFP‐tagged P450 proteins (CYP6AE70, CYP6AE47, CYP6AE50, and CYP6AE139) or EGFP alone. Lysates with equalized protein concentrations were incubated overnight at 4 °C with lycorine (Lyc) or solvent control (DMSO). P450 protein levels were quantified using an insect‐specific ELISA. Data were normalized to the DMSO control (set as 1.0) to minimize baseline variations and avoid negative values derived from assay background noise. (D) Schematic drawing of the interaction between lycorine and CYP6AE47. Bars represent mean ± SEM of three independent replicates. Statistical significance was determined by Student's t‐test (* P < 0.05, ** P < 0.01, *** P < 0.001).

Two‐dimensional docking analyses (Fig. 5(A), Supporting Information, Fig. S8, Table S3) further revealed the key interactions stabilizing lycorine within the P450 active pockets as follows:

CYP6AE70: ARG100 formed a conventional hydrogen bond (‐C…H‐) with lycorine (3.3 Å), while hydrophobic interactions occurred with VAL115, and PHE116 (3.4–4.0 Å).

CYP6AE47: ARG170, SER174, and SER319 formed hydrogen bonds with lycorine (3.5–4.1 Å). Additional hydrophobic interactions involved ARG170 and VAL209 (3.7–3.9 Å), and salt bridges involved GLU210 (3.8 Å), whereas π‐π T‐shape interactions were observed with ARG170 (4.1 Å).

CYP6AE50: Hydrogen bonds were formed with ASN52, SER381, and VAL382 (2.9–3.8 Å); hydrophobic contacts involved PRO384 and PHE493 (3.6–3.7 Å).

CYP6AE139: ARG175, ALA206, ALA209, and ASP210 formed hydrogen bonds (3.0–3.7 Å), while hydrophobic interactions occurred with VAL171, and ALA209 (3.3–3.9 Å).

3.8. Functional analysis of key binding sites

To investigate the functional relevance of critical residues mediating lycorine binding with hydrogen bonding, hydrophobic interactions, salt bridges and π‐π T‐shaped interactions, we analyzed the conservation of SER174, ARG170, VAL209, GLU210, and SER319 across CYP6AE47 homologs from other Spodoptera species. As shown in Fig. 6(B), these residues are highly conserved. Notably, ARG170 participated simultaneously in hydrogen bonds, hydrophobic interactions, and T‐shaped π‐π interactions with lycorine. These residues are preserved in key Spodoptera pests, including Spodoptera littoralis, Spodoptera frugiperda, and Spodoptera exigua, whereas the corresponding site in the human CYP3A4 homolog is ARG substituted with a Lysine (LYS) residue.

3.9. Lycorine reduces P450 protein levels in vitro

To determine whether lycorine directly affects P450 protein levels, an in vitro incubation assay was performed using Spli221 cells expressing P450‐EGFP fusion proteins or empty EGFP vector controls. Cell lysates were incubated with lycorine or DMSO at 4 °C overnight. ELISA quantification revealed that protein levels of all four P450 isoforms were significantly reduced following lycorine treatment compared with controls (DMSO) (Fig. 6(C)), whereas the abundance of P450 in the EGFP group remained unchanged between the treatment and control conditions. In addition, the same treatment with isoprocarb was applied to these P450‐overexpressing cells. The results showed that, compared with the control (DMSO), isoprocarb treatment significantly increased P450 abundance in the cells overexpressing P450 genes, and such elevation was also observed in the EGFP group exposed to isoprocarb (Supporting Information, Fig. S9). This indicates that lycorine can specifically reduce P450 protein levels without affecting control proteins, while isoprocarb is capable of increasing P450 protein abundance.

4. DISCUSSION

The present study reveals the unique mode of action of lycorine against S. litura. The results indicate that lycorine simultaneously regulates P450 function at both transcriptional and protein levels, suppressing gene expression and inhibiting enzymatic activity, which may be among the mechanisms underlying its insecticidal activity. Toxicity assays revealed a distinct stage‐dependent susceptibility: lycorine exhibited high toxicity towards 1st‐instar larvae (LD₅₀ = 0.55 μg larva⁻¹) (Table 1) and moderate toxicity towards 3rd‐instar larvae (LD₃₀ = 65 μg larva⁻¹, LD₅₀ = 86 μg larva⁻¹) (Table 2). In toxicity comparisons against third‐instar larvae, although less potent than the conventional insecticide isoprocarb, lycorine demonstrated notably stronger activity than the plant‐derived alkaloid berberine (Fig. 1(E), (F), Table 2). Additionally, lycorine showed superior efficacy against first‐instar larvae, with an LD₅₀ of 0.55 μg larva⁻¹, lower than the widely used insecticide methomyl (0.84 μg larva⁻¹). ⁶⁸ These results suggest that lycorine will have distinct insecticidal potential against S. litura, particularly in early larval stages, which supports its promise as a candidate for botanical pesticide development. Notably, even at sublethal doses (LD₃₀), lycorine induced significant physiological impairment, highlighting its promise as an agent for IPM.

Physiological assays revealed that lycorine disrupts systemic homeostasis in S. litura larvae. Key indicators included elevated levels of hemolymph trehalose and uric acid, accompanied by reduced uric acid excretion and decreased frass moisture (Fig. 1(G)–(I)). Accumulation of hemolymph metabolites, reduced excretion, and impaired water regulation are characteristic of Malpighian tubule dysfunction. ⁶⁹ , ⁷⁰ , ⁷¹ , ⁷² As the primary excretory and osmotic regulatory organs in insects, Malpighian tubules mediate the synthesis, transport, and excretion of uric acid and contribute to maintaining organismal redox homeostasis. ⁷⁰ , ⁷¹ , ⁷³ Accordingly, alterations in trehalose and uric acid levels serve as sensitive indicators of systemic metabolic stress and Malpighian tubule impairment. ⁷³ , ⁷⁴ , ⁷⁵ These changes indicate impaired osmotic and metabolic balance, consistent with Malpighian tubule dysfunction. ⁷⁶ Concomitant reductions in body mass gain and altered water loss rates align with prior studies showing that xenobiotics targeting Malpighian tubules induce systemic metabolic stress and retard growth. ⁷⁶ , ⁷⁷ Together, these findings suggest that sublethal lycorine exposure impairs energy metabolism and osmotic regulation through disruption of Malpighian tubule functions.

Histological analysis (H&E staining) revealed that lycorine treatment caused no obvious structural damage to the midgut of S. litura, where cells remained tightly packed and intact (Fig. 2(A)). In contrast, Malpighian tubules of treated larvae exhibited obvious vacuolization (Fig. 2(A)). These comparative results between tissues expected to be involved in the response to toxic agents indicate that lycorine primarily targets the Malpighian tubules rather than the midgut. Further histological and ultrastructural analyses confirmed severe damage in the Malpighian tubules characterized by nuclear condensation or fragmentation, ⁷⁸ disintegration of the F‐actin cytoskeleton, ⁷⁸ and perforation of the epithelial wall (Fig. 2(A)–(C)). These findings identify the Malpighian tubules as the primary site of lycorine‐induced damage at the cellular level.

Biochemical assays revealed that lycorine disrupted redox homeostasis, with elevated hydrogen peroxide (H₂O₂) and MDA levels, increased SOD and CAT activities, and decreased POD and T‐AOC activity (Fig. 2(D)). This oxidative imbalance reflects excessive ROS accumulation and lipid peroxidation, ⁷⁷ , ⁷⁹ with increased MDA correlating directly with epithelial structural damage. These physiological changes are consistent with established models in which plant secondary metabolites induce oxidative stress leading to tissue dysfunction. ⁷⁶ , ⁷⁷ , ⁷⁹ Overall, these findings underscore oxidative damage as a central mechanism of lycorine toxicity.

Assays of Malpighian tubule transcription products revealed that DEGs were significantly enriched in cytochrome P450‐related pathways determined by using both GO and KEGG analyses (Fig. 3(E), (F), Supporting Information, Fig. S1). Notably, CYP6AE138, CYP6AE70, CYP6AE47, CYP6AE50, CYP6AE139 were among the most downregulated P450‐associated DEGs (Fig. 3(C), (D)). Whereas the transcriptome revealed some upregulated genes, downregulated ones were more prevalent. Considering that lycorine inhibits synthesis of DNA and some proteins, ¹⁸ , ²⁵ we focused mainly on the downregulated genes. These downregulated genes were located on chromosome 20 (Fig. 3(I)) and constituted a CYP6AE gene cluster. ⁸⁰ Consistently, P450 enzyme activity assays indicated that lycorine also suppressed the activity of the corresponding proteins (Fig. 3(G)). Functional validation via RNAi demonstrated that co‐silencing of this gene cluster under LD₃₀ lycorine exposure increased larval mortality by 30% (P < 0.001; Fig. 4(E)). These observations indicate that the insecticidal action of lycorine is mediated, at least in part, by its inhibitory effect on this specific CYP6AE gene cluster. To address the potential involvement of these P450 genes in response to other pesticides, we performed RNAi under isoprocarb exposure (Supporting Information, Fig. S10). Consistent with the lycorine results, co‐silencing of these genes altered larval physiological status, indicating their broad roles in xenobiotic detoxification.

Inhibition of P450 enzymes can lead to substrate accumulation and inefficient oxidation, increasing electron leakage and ROS generation, which exacerbates oxidative damage. ⁸¹ , ⁸² , ⁸³ , ⁸⁴ Comparable effects have been reported for botanical alkaloids such as piperonyl butoxide and piperine, which alters P450 gene expression while enhancing toxicological effects. ⁸⁵ , ⁸⁶ , ⁸⁷ , ⁸⁸ Interestingly, these alkaloids, including lycorine, share the same methylenedioxyphenyl (MDP) group which is known for inhibiting P450 enzymes. Other plant‐derived compounds, including matrine and berberine, also interfere with P450 expression or catalytic function via induction of oxidative stress and disruption of mitochondria. ³⁸ , ⁸⁹ , ⁹⁰ Hence, it is reasonable to infer that the P450 system may perform a dualistic role under lycorine exposure: while attempting detoxification, it inadvertently contributes to oxidative stress which reduces metabolic efficiency and overwhelms larval detoxification capacity. Furthermore, lycorine exhibited higher toxicity than berberine but lower toxicity than isoprocarb (Supporting Information, Table S2). Notably, isoprocarb typically functions as a synergist in insecticide formulations, whereas berberine has been reported to act as both a synergist and an inhibitor of certain enzymes. Such modes of action correspond to established consequences of P450 inhibition in insects and hint at the potential for using lycorine as a botanical P450 inhibitor or a synergist in IPM. This potential is further supported by its notable synergistic interaction with fenvalerate in A. gossypii ²² and its ability to suppress the activity of the mammalian cytochrome P450 isoform CYP3A4. ¹⁸

Mechanistically, RT‐qPCR analysis indicated that overexpression of any single gene induced expression of the others, suggesting strong cross‐regulation and the formation of a coordinated detoxification unit (Supporting Information, Fig. S6). These findings are consistent with phylogenetic and genomic evidence, supporting the notion that these P450s function as an integrated module for metabolizing plant secondary metabolites. ²⁸ MST analysis confirmed that lycorine binds directly to CYP6AE138, CYP6AE70, CYP6AE47, CYP6AE50, and CYP6AE139 with varying affinities. However, CYP6AE47 shows nanomolar affinity which makes it likely to serve as the principal responsive enzyme (Fig. 5(D)). Molecular docking also suggested favorable binding of lycorine to CYP6AE138, CYP6AE70, CYP6AE47, CYP6AE50, and CYP6AE139 (ΔG = −7.7 to −8.6 kcal mol⁻¹) (Fig. 6(A)).

In CYP6AE47, residues ARG170, SER174, VAL209, GLU210, and SER319 mediate specific hydrogen bonding, π‐π T‐shaped stacking, and hydrophobic interactions with lycorine. These key amino acid residues exhibit extremely high conservation among various Spodoptera pests (Fig. 6(B)), implying that the binding pattern of P450 proteins with lycorine may be widely present in this pest taxon, and these residues are crucial for the recognition and binding of lycorine by pest insect P450 proteins. These residues, along with surrounding hydrophobic amino acids, create a complementary binding pocket that stabilizes lycorine and provides a mechanistic basis for its P450 inhibitory activity. Intriguingly, ARG170 is simultaneously involved in three types of molecular interactions, namely hydrogen bonding, hydrophobic interactions and π‐π stacking interactions, highlighting its role as the key binding residue underlying the interaction between CYP6AE47 and lycorine, and its dominant function in stabilizing the protein‐ligand complex.

Moreover, in vitro assays showed that lycorine reduces the abundance of P450 proteins in Spli221 cells (Fig. 6(C)). Concurrently, Malpighian tubules displayed a pronounced inhibition of P450 enzymatic activity in assays in vivo. Together, these findings indicate that lycorine both downregulates detoxification genes and interferes with the function of P450 enzymes they encode. Consistent with the finding by Lin et al. ²⁵ that lycorine suppresses sPLA2 expression and disrupts lipid homeostasis in S. litura, and considering that arachidonic acid produced via PLA2 catalysis serves as an important substrate for P450 enzymes, ²⁶ our results complement their conclusions. Together, these findings elucidate the multi‐level disruptive mechanism of lycorine, which acts sequentially from upstream lipid hydrolysis to downstream detoxification metabolism. These phenomena may lead to growth retardation or mortality for larvae.

Finally, the present study provides several novel insights into the modes of action of lycorine. We identify a mechanism whereby lycorine simultaneously suppresses transcription and inhibits the activity of cytochrome P450 enzymes. Molecular interaction analyses further clarify the binding of lycorine to specific P450 isoforms, particularly CYP6AE47, offering a structural framework for targeted inhibition of P450 activity. Our results also show that oxidative stress and Malpighian tubule dysfunction are central to lycorine‐induced toxicity, thereby linking molecular interactions to physiological effects.

Taken together, our findings highlight lycorine as a promising botanical insecticide or synergist for IPM. Accordingly, its ability to impair detoxification pathways while inducing oxidative stress may enhance the efficacy of existing insecticides, reduce dependence on synthetic chemicals, and support environmentally friendly pest control strategies. The high conservation of key binding residues of these enzymes across noctuid pests supports the broad potential applicability of these targets beyond S. litura. Nevertheless, we note here that several limitations warrant further investigation. First, the biological functions of the CYP6AE gene cluster were initially characterized using RNAi, but no genetic knockout was performed to achieve permanent gene deletion. Future studies should therefore employ CRISPR/Cas9 technology to generate stable knockout strains, which would allow definitive functional validation. Second, the relevance of post‐translational P450 inhibition in vivo under field conditions remains to be validated, and potential off‐target effects, environmental persistence, and risks to non‐target organisms (NTOs) must be carefully assessed prior to practical application. Therefore, we suggest that future studies should also explore site‐directed mutagenesis of key residues (e.g., ARG170) to clarify structure–function relationships, screen lycorine analogues for improved specificity, develop synergistic formulations with conventional insecticides, and evaluate the long‐term impacts of sublethal exposure on population dynamics, evolution of resistance, and interactions with other pest management strategies.

5. CONCLUSIONS

In conclusion, lycorine is a major alkaloid of the Amaryllidaceae which exhibits potent insecticidal activity against S. litura, and high toxicity towards 1st‐instar larvae (LD₅₀ = 0.55 μg larva⁻¹). The present study demonstrates that lycorine disrupts metabolic and osmotic homeostasis in larvae by elevating oxidative stress and impairing Malpighian tubule function. Lycorine simultaneously downregulates multiple detoxification genes, notably the CYP6AE gene cluster (CYP6AE47, CYP6AE50, CYP6AE70, CYP6AE138, and CYP6AE139), and inhibits cytochrome P450 enzymatic activity. Further validating their critical role in detoxification, RNAi‐mediated silencing of these genes substantially enhances larval susceptibility. Additional analyses using molecular docking and microscale thermophoresis show that lycorine directly binds to diverse P450 isoforms through hydrogen bonding, π‐π stacking, and hydrophobic interactions, with CYP6AE47 showing particularly high affinity, and in vitro experiments reveal its capacity to reduce P450 protein abundance. Together, lycorine weakens insect detoxification capacity through simultaneously down‐regulating expression of these genes and inhibiting P450 protein activity, which exacerbates oxidative damage and ultimately impairs larval development. These insights advance our understanding of plant‐derived alkaloids as botanical insecticides, provide a structural and functional framework for P450‐targeted pest control strategies, and support the potential integration of lycorine into environmentally sustainable pest management programs. Future research addressing field efficacy, safety assessment, and development of effective formulations with low environmental impact will be essential for translating these findings into practical agricultural applications.

CONFLICT OF INTEREST

The authors declare no conflicts of interest with the contents of this article.

Supporting information

Fig S1. Transcriptome analysis of Malpighian tubules in S litura. (A) Heatmap of detoxification‐related gene expression in Malpighian tubules following lycorine treatment. (B) Gene Ontology (GO) classification of DEGs in Malpighian tubules; top 50 categories are shown. Circle size indicates the number of genes, while circle color reflects enrichment significance (red, significant; blue, not significant). Differentially expressed genes were significantly enriched in biological processes such as energy metabolism and glutathione metabolism, which were mainly localized to cell membrane structures, and enriched functions related to oxidoreductases, detoxification enzymes and ion transmembrane transport. (C) KEGG pathway enrichment analysis of DEGs in Malpighian tubules; top 50 pathways are shown. Circle size indicates the number of genes, while circle color reflects enrichment significance (red, significant; blue, not significant). The differentially expressed genes were significantly enriched in pathways including xenobiotic metabolism by cytochrome P450, glutathione metabolism, energy metabolism, ABC transporters and ion homeostasis regulation, which were predominantly involved in xenobiotic metabolism and energy adaptation responses under stress conditions.

Fig. S2. Multiple sequence alignment of five S. litura P450 paralogs. Alignment of CYP6AE138, CYP6AE70, CYP6AE47, CYP6AE50, and CYP6AE139 isoforms. Comparative analysis reveals 73.36% sequence conservation among these paralogs. Amino acid residues are colored according to the ClustalW scheme based on their properties. Fully conserved amino acid residues are marked with an asterisk (*) above the alignment.

Fig. S3. Primer amplification efficiency assay. The abscissa represents the number of PCR cycles, and the ordinate represents the fluorescence intensity. The R² values of the fitted standard curves are all greater than 0.99, and the absolute values of the slopes are between −3.1 and − 3.5, corresponding to the primer efficiency (E) in the range of 90% to 110%, which confirms that the primers used for the assay have stable and reliable amplification efficiency and can meet the needs of subsequent real‐time fluorescent quantitative PCR experiments. Calculation method of primer efficiency (E): Calculated based on the slope of the real‐time fluorescent quantitative PCR standard curve, the core formula is E = 10^−1/slope ‐ 1, and the calculation result multiplied by 100% is the percentage of primer efficiency.

Fig. S4. Effects of individual P450 gene silencing on larval development in S. litura. RNAi‐mediated knockdown of CYP6AE138, CYP6AE70, CYP6AE47, CYP6AE50, and CYP6AE139, with EGFP as the control. (A) Larval duration. (B) Larval weight. (C) Larval food consumption. (D) Mortality rate. Compared to the control, third‐instar larval duration increased by 38%, 47%, 31%, 39%, and 39% (A); body weight decreased by 41%, 40%, 36%, 38%, and 50% (B); food consumption was reduced by 12%, 24%, 21%, 19%, and 25% (C). No significant differences in mortality were observed (D). Bars represent mean ± SEM of three independent replicates. Statistical significance was determined by Student's t‐test (* P < 0.05, ** P < 0.01, *** P < 0.001).

Fig. S5. RNAi‐mediated silencing of a single gene had no significant effect on the expression levels of other related genes in the gene cluster. After interference treatment of the target gene, its expression level of the target gene changed significantly, while the expression levels of other related genes remained unchanged. Bars represent mean ± SEM of three independent replicates. Statistical significance was determined by Student's t‐test (* P < 0.05, ** P < 0.01, *** P < 0.001).

Fig. S6. Overexpression of P450 genes in Spli221 cells and their cross‐regulatory effects on transcript levels. (A) Effect of CYP6AE70 overexpression, which induced the expression of CYP6AE47, CYP6AE50, and CYP6AE139. (B) Effect of CYP6AE47 overexpression, which upregulated CYP6AE70, and CYP6AE139, but had a minimal effect on CYP6AE50. (C) Effect of CYP6AE50 overexpression, which enhanced all three counterparts, with a notable effect on CYP6AE139. (D) Effect of CYP6AE139 overexpression, which strongly induced the other genes, especially CYP6AE50. Bars represent mean ± SEM of three independent replicates. Statistical significance was determined by Student's t‐test (* P < 0.05, ** P < 0.01, *** P < 0.001).

Fig. S7. MST Analysis with EGFP control shows no binding to lycorine. EGFP was used as the fluorescent control protein in MST (related to Fig. 5(D)). A titration series of lycorine was performed while maintaining a constant EGFP concentration. No binding was detected for EGFP. Binding curves were generated using NanoTemper Analysis v1.2.231 software. Normalized fluorescence (F_hot / F_initial) is plotted against lycorine concentration.

Fig. S8. Predicted protein structures of S. litura P450 genes. A‐e: AlphaFold3‐generated models of CYP6AE138, CYP6AE70, CYP6AE47, CYP6AE50, and CYP6AE139, showing pTM (predicted template modeling) and ipTM (interface predicted template modeling) scores, indicating high‐confidence structural predictions.

Fig. S9. Effect of isoprocarb treatment on proteins of the CYP6AE gene cluster. Spli221 cells were transfected to express EGFP‐tagged P450 proteins (CYP6AE70, CYP6AE47, CYP6AE50 and CYP6AE139) or EGFP alone. Lysates with equalized protein concentrations (0.5 μg μL⁻¹) were incubated overnight at 4 °C with lycorine (Lyc) or solvent control (DMSO). P450 protein levels were quantified using an insect‐specific ELISA. Data were normalized to the DMSO control (set as 1.0) to minimize baseline variations and avoid negative values derived from assay background noise. Bars represent mean ± SEM of three independent replicates. Statistical significance was determined by Student's t‐test (* P < 0.05, ** P < 0.01, *** P < 0.001).

Fig. S10. Effects of P450 gene silencing on the development of S. litura larvae under isoprocarb exposure. With enhanced green fluorescent protein (EGFP) serving as the negative control, RNAi was performed to knockdown the expression of SlituP450‐072 (CYP6AE138), SlituP450‐073 (CYP6AE70), SlituP450‐074 (CYP6AE47), SlituP450‐075 (CYP6AE50), and SlituP450‐076 (CYP6AE139), both individually and in combination (SlituP450‐mix). (A) Phenotypic alterations of S. litura larvae upon P450 gene silencing; (B) Validation of the knockdown efficiency of each of the five P450 genes by RT‐qPCR; (C) Changes in larval body weight following individual knockdown of the five P450 genes; (D) Larval food consumption after individual knockdown of the five P450 genes; (E) Larval frass excretion after individual knockdown of the five P450 genes; (F) Validation of the knockdown efficiency of the combined five P450 genes by RT‐qPCR; (G) Determination of physiological indicators in larvae after combined knockdown of the five P450 genes. SlituP450‐mix: it combined dsRNA targeting SlituP450‐072, −073, −074, −075, and − 076. Bars represent mean ± SEM of three independent replicates. Statistical significance was determined by Student's t‐test (* P < 0.05, ** P < 0.01, *** P < 0.001).

Table S1. RNAi‐mediated co‐silencing of enhanced lycorine toxicity in a Spodoptera litura CYP6AE cluster.

Table S2. Predicted pTM and ipTM scores for SlituP450 protein modeling*.

Table S3. Binding energy and nonbonding interaction of lycorine.

Table S4. P450 gene primers for qPCR, cloning and RNAi.

Table S5. P450 Primers for restriction enzyme digestion and vector construction.

DATA AVAILABILITY STATEMENT

Transcriptome analysis data is derived from accession number CNP0007186 in the China National GeneBank (CNGB) Sequence Archive (CNSA) database. All other data that support the findings of this study are available from corresponding author K. M. upon request.