IntroductionAphids are small sap-sucking insects belonging to the order Hemiptera and the superfamily Aphidoidea. They feed on plants by extracting phloem sap and often transmit viruses present in their saliva to the host plant. Approximately 275 aphid species are known to transmit plant viruses, many of which cause economically significant diseases1. However, the overall economic impact of aphid infestations remains difficult to quantify due to their complex interactions with crops and ecosystems2,3. In addition to direct damage, aphids promote the growth of sooty mould through honeydew excretion and disrupt plant physiology, indirectly reducing crop yield and quality4.Aphids reproduce parthenogenetically and viviparously, with short life cycles and the capacity to produce more than 40 generations per year. This rapid population growth, combined with their adaptability and reproductive plasticity, makes them particularly challenging to manage in open-field conditions5. In field crops, aphid control typically relies on synthetic insecticides such as pyrethroids, neonicotinoids, and carbamates. However, these compounds often lose effectiveness because aphids can develop resistance to multiple insecticide classes6. Moreover, synthetic insecticides pose risks to human health, non-target organisms, and the environment. Their widespread use has been associated with pollinator decline, disruption of natural enemy populations, and contamination of soil and water resources7. These limitations have prompted growing interest in alternative pest control strategies, including the use of essential oils (EOs) as biopesticides8,9,10. EOs are complex mixtures of volatile secondary metabolites, primarily terpenes and their oxygenated derivatives (e.g., aldehydes, esters, alcohols), which confer diverse biological properties. Their insecticidal activity is influenced by volatility, stability, bioavailability, and interactions with the target organism’s biological systems11. In insects, the rapid onset of action and behavioural changes observed following EO exposure suggest a neurotoxic mode of action. The most extensively studied mechanisms include acetylcholinesterase inhibition, modulation of gamma-aminobutyric acid (GABA) receptors, and activation of octopaminergic signaling pathways. Additionally, synergistic and antagonistic interactions among EO constituents, as well as differences in activity between whole oils and isolated compounds, support the hypothesis of multifactorial insecticidal mechanisms12,13,14.Beyond their insecticidal potential, essential oils offer ecological advantages, including biodegradability, low environmental persistence, and compatibility with integrated pest management (IPM) programs. Their selectivity toward target pests, coupled with reduced toxicity to beneficial arthropods, makes them promising candidates for sustainable agriculture11,12,13. EOs are produced by specialized secretory structures such as glandular trichomes, whose density and morphology influence both the quantity and quality of oil production15.EOs are particularly abundant in the Lamiaceae family, where terpenoids—especially monoterpenes, sesquiterpenes, and phenylpropanoids—are predominant constituents3. Within this family, species of the genus Lavandula are well recognized for their insecticidal properties16,17. Lavandula dentata, native to the Mediterranean region, is widely used as a medicinal and aromatic plant. Its essential oils exhibit a broad spectrum of biological activities, including antibacterial, antifungal, anti-inflammatory, antioxidant, and sedative effects17,18. The presence of key terpenoids such as linalool, camphor, eucalyptol, and linalyl acetate contributes to its insecticidal potential18. Despite its pharmacological relevance, the insecticidal properties of wild L. dentata chemotypes remain underexplored, particularly in Mediterranean ecosystems16,17.This study aimed to evaluate the insecticidal activity of essential oil extracted from a wild chemotype of Lavandula dentata collected in northern Algeria against the citrus aphid (Aphis spiraecola), a destructive pest and virus vector in Mediterranean orchards. Histological analyses were performed to visualize essential oil–secreting trichomes, while the chemical composition of the oil was characterized using gas chromatography–mass spectrometry (GC–MS). In addition, an in silico molecular docking study was conducted to investigate interactions between key insecticidal compounds and the insect neurological receptor acetylcholinesterase. The selectivity of the essential oil was assessed against non-target beneficial insects, including aphid predators (ladybird beetles, cecidomyiid, and chrysopid larvae) and a pollinator species (adult honeybee). Collectively, this work provides the first evidence of the aphicidal activity of L. dentata essential oil against A. spiraecola, demonstrates its selective activity with minimal non-target effects, highlights the role of a chemotype specific to North Algerian populations, and suggests a possible neurotoxic mechanism of action through acetylcholinesterase inhibition. These findings advance botanical pesticide strategies and broaden current knowledge of the bioactivity of L. dentata essential oil.Materials and methodsPlant materialsLavandula dentata L. (Lamiaceae) is a widely distributed aromatic and medicinal plant in Algeria and is not protected. Leaf samples were collected during the flowering period in April 2024 from a thermo-Mediterranean zone in the Blidean Atlas, specifically in Hammam Melouane (36°29′12″ N, 3°02′42″ E)19. The collection of the plant material was authorized by the Directorate of Forest Conservation of the Blida Province, and complied with relevant institutional, national, and international guidelines and legislation. Botanical identification was confirmed through morphological comparison with authenticated reference specimens housed at the herbarium of Chrea National Park, and further validated by comparison with specimens from the herbarium of the National Higher School of Agronomy (ENSA), Algiers (voucher specimen no. 109–29, Lavandula dentata L.), and validated by Algerian plant specialists (see Supplementary Material). The leaves were washed, dried and ground into a fine powder, and stored in glass jars. This plant was selected based on its natural abundance and its documented essential oil content, which has previously demonstrated pesticidal and insecticidal properties19.Hydrodistillation of essential oilEssential oil was extracted by hydrodistillation using a Clevenger-type apparatus in the Laboratory of Natural and Local Bio-resources, Hassiba Ben Bouali University, Algeria. A total of 100 g of dried Lavandula dentata leaf powder was placed in a distillation flask containing 1200 mL of distilled water and subjected to hydro-distillation for 3 h20.During hydrodistillation, volatile aromatic compounds were released and carried by rising steam, which condensed into two distinct phases: an aqueous phase (hydrosol) and an organic phase (essential oil). The essential oil was carefully separated from the hydrosol and dried over anhydrous sodium sulfate (Na₂SO₄) to remove residual moisture. The purified oil was stored in opaque glass vials at 4°C until further analysis.Insect materialAll insect specimens used in the bioassays were collected from citrus orchards located in the Mitidja plain, North-Central Algeria, during April 2025. The study encompassed both target pests and non-target beneficial insects naturally associated with citrus agroecosystems.Aphids (Aphis spiraecola)Due to the absence of a laboratory-reared population, field-collected specimens of the green citrus aphid (Aphis spiraecola Patch, 1914) were used. Infested citrus leaves were sampled from an unsprayed orchard (36° 45.082′ N, 2° 55.296′ E) and transported to the laboratory in ventilated plastic containers lined with moist absorbent paper to preserve leaf turgidity and aphid viability. Upon arrival, aphids were sorted and identified using a standard taxonomic key21. No signs of parasitism, predation, or disease were observed, and the population was considered healthy and morphologically homogeneous.Non-target beneficial insectsThe selection of non-target beneficial insects was based on their natural occurrence and availability within the orchard ecosystem. Specimens were sampled directly from citrus shoots infested with aphids, including ladybird beetle larvae (Coccinellidae), cecidomyiid larvae (Diptera: Cecidomyiidae), and chrysopid larvae (Neuroptera: Chrysopidae). Honeybees (Apis mellifera) were obtained separately from a nearby apiculture farm. All specimens were subsequently sorted and identified under a stereomicroscope prior to bioassays.Yield and chemical composition of the essential oilThe yield of Lavandula dentata essential oil was calculated according to Eq. (1).$${\text{Yield}} \left( \% \right) = \left( {{\text{Mass of oil }}\left( {\text{g}} \right) / {\text{Mass of dry plant material }}\left( {\text{g}} \right)} \right) \times 100$$(1)Chemical characterization of the essential oil was performed at the Department of Physics, Chemistry, and Pharmacy, University of Southern Denmark. A volume of 0.2 µL of each essential oil sample was injected in split mode into a Hewlett-Packard Agilent 6890 Plus gas chromatograph coupled to a Hewlett-Packard Agilent 5973 mass spectrometer. The system was equipped with a fused silica capillary column (HP-5MS; 30 m × 0.25 mm × 0.25 µm film thickness). The oven temperature was initially set to 60°C for 8 min, then increased at 2°C min⁻1. until reaching 250°C. The total run time was 113 min. Helium (purity 99.9999%) was used as the carrier gas at a constant flow rate of 0.5 mL min⁻1. Mass spectra were recorded at an ionization energy of 70 eV. Data acquisition and processing were carried out using Agilent MSD Productivity ChemStation software. Compound identification was based on retention time (Rt) and comparison of mass spectra with those in the NIST 2011 spectral library. In the absence of retention index (RI) data, identification was further supported by recording the following parameters for each compound: IUPAC chemical family, molecular weight (MW, g mol⁻1), spectral similarity scores (Match/Reverse from NIST/Wiley), and key diagnostic ions (m/z), including base peak and molecular ion. The relative abundance of each compound was expressed as a percentage of the total peak area (area %).Microscopic examination of cross-sections of the leaf blade of L. dentataHistological examination was performed to investigate the internal leaf anatomy of Lavandula dentata. Fresh leaf samples were manually sectioned using freehand techniques to obtain thin transverse sections. The sections were treated with sodium hypochlorite (12°C, 20 min) to remove cytoplasmic contents and clarify cellular structures. Subsequently, a brief immersion in 1% acetic acid (1–2 min) was applied to enhance dye fixation on the cell walls.Double staining was carried out using methyl green (20 min) to highlight nuclei and lignified tissues, followed by Congo red (10 min) to stain cellulose-rich structures. The stained sections were preserved in distilled water supplemented with a few drops of 1% glycerol to maintain hydration and clarity.Microscopic observations were conducted under a light microscope at magnifications × 10 and × 40. Representative micrographs were captured using a digital camera, and scale bars were calibrated using a stage micrometer.In silico prediction of ligand–receptor interactionsThe molecular docking study was conducted to predict the binding affinity of L. dentata essential oil constituents to acetylcholinesterase (AChE), a key enzyme in insect neurophysiology. The three-dimensional structure of AChE was retrieved from the Protein Data Bank (PDB ID: 6XYU) for Drosophila melanogaster. This structure was determined experimentally by X-ray diffraction at a resolution of 2.51 Å, ensuring high reliability.Prior to docking, all crystallographic water molecules and co-crystallized ligands were removed. Hydrogen atoms were added, and partial charges were assigned using standard protocols. Ligand structures were obtained from the PubChem database, converted to 3D format, and geometry-optimized using Open Babel to generate energetically favorable conformations. Docking input files were prepared in PDBQT format.Initial docking simulations were performed in the Molecular Operating Environment (MOE), followed by refinement using AutoGrid and AutoDock. The active site of AChE was identified based on known catalytic residues from the complex with the tacrine derivative 9-(3-iodobenzylamino)-1,2,3,4-tetrahydroacridine, and the docking grid was centered accordingly.Docking calculations were executed, and the resulting ligand–protein complexes were analyzed. Binding poses were visualized using Discovery Studio Visualizer 2021, and interaction profiles were evaluated based on hydrogen bonding, hydrophobic contacts, and binding energy scores. Computational results were further interpreted in light of experimental bioactivity data. This workflow follows standard docking protocols widely applied in insect AChE studies22.In vitro bioassays on aphidsLaboratory bioassays were conducted under controlled conditions (26 ± 1 °C; 60–70% relative humidity) using a completely randomized design. Each experimental unit consisted of a Petri dish (90 mm diameter) containing ten adult Aphis spiraecola individuals, transferred using a fine damp brush. Only adults were selected for testing, as nymphs are generally more susceptible to chemical treatments23. Each Petri dish contained a citrus leaf disc placed on a thin layer of nutrient agar (1% agar) to maintain leaf turgidity. The dish lid was perforated at the center (1.2 cm diameter) and covered with transparent tulle to allow ventilation while preventing insect escape. The experimental protocol of the insecticidal effect is illustrated in Supplementary Fig. 1. Four concentrations of essential oil were tested: D₁ = 10 µL mL⁻1, D₂ = 7 µL mL⁻1, D₃ = 4 µL mL⁻1, and D₄ = 1 µL mL⁻1. Solutions were prepared by diluting the oil in sterile distilled water containing Tween 20 (0.1%) as an emulsifier. Treatments were applied using a calibrated mini sprayer, delivering 0.25 mL per application24,25. Mortality was recorded at 6, 12, 24, 36, 48, 60, and 72 h post-treatment. Each treatment was replicated four times. Two controls were included: (i) Positive control (W⁺): synthetic insecticide (Astrad®, Acetamiprid 20%, Dekachim, Algeria), and (ii) Negative control (W⁻): sterile distilled water with Tween 20 (0.1%).Corrected mortality rates (MT %) were calculated using Abbott’s formula26. to account for natural aphid mortality observed in the control group:$$MT\%=\frac{\text{Mortality in the treated group}-\text{mortality in the control group}}{100-\text{mortality in the control group}}\times 100$$(2)Selectivity assessment of essential oil on non-target insectsTo evaluate the selectivity of the essential oil, the dose corresponding to the LD₉₀ for Aphis spiraecola at 48 h (20.02 µL mL⁻1) was applied to non-target beneficial insect species. Bioassays were conducted in Petri dishes (90 mm diameter), each containing 5–8 individuals depending on species availability. Predatory insects were placed on citrus leaf discs infested with aphid colonies to simulate natural feeding conditions. Selectivity on honeybees (Apis mellifera) was tested separately in ventilated glass vials, with individuals fed powdered sugar to maintain viability. Each treatment was replicated four times per non-target insect species. A positive control was included, consisting of the synthetic insecticide Astrad® (Acetamiprid 20%, Dekachim, Algeria), while untreated controls were not used in this phase. Mortality was recorded at 6, 12, 24, 36, 48, 60, and 72 h post-treatment.Data processing and statistical analysisLethal doses (LD₅₀ and LD₉₀), corresponding to the concentrations causing 50% and 90% mortality, respectively, were estimated using Probit regression. Dose values were log-transformed, and mortality percentages were converted to probit units. A linear regression model (y = ax + b) was fitted, where y is the probit-transformed mortality, x is the logarithm of the dose, a is the slope, and b is the intercept. All dose–response analyses and statistical comparisons were performed in RStudio27. (version 2024.09.0.25). Data were analyzed using two-way ANOVA, and results were expressed as mean ± standard deviation (SD). Dunnett’s post hoc test was applied to compare treated groups with the control, while Tukey’s test was used to assess differences among treatments. Statistical significance was defined as p