Bioactive potential of stinging nettle, round-leaved mint, and crowned chrysanthemum extracts against Tuta absoluta: Chemical characterization and molecular docking insights

2.1 Plant material

The collection sites, botanical identification, and herbarium deposition of the studied species have been previously documented by Krache et al., (2025).

2.2 Methanolic extract preparations

After being harvested, the leaves were dried in the dark and ground into a homogeneous powder. They were then stored in airtight containers at room temperature to preserve the integrity of the bioactive compounds.

To prepare the methanol extracts, 20 g of leaf powder from each plant species was first subjected to a delipidation step. The plant powder was then mixed with 10 ml of petroleum ether and shaken manually for 10 min to remove surface lipids and wax. This process was performed under a hot vacuum to allow complete evaporation of the solvent. The samples were subsequently placed in a cartridge and subjected to Soxhlet extraction. A volume of 600 ml of methanol was then added to the flask. Following five cycles of exhaustion, the cartridge was removed, and the methanol loaded with the plant extract was recovered. Next, the solvent was removed via rotary evaporation at 45°C (Bichra et al., 2012). The yield was subsequently calculated, after which the dry residue was recovered from the flask via dilution with acetone at a ratio of 10 ml/g.

The crude extract obtained was stored at 4°C in the dark to maintain its chemical stability and avoid any degradation by light or heat. This is essential to ensure the reliability of subsequent chemical and biological analyses.

2.3 Secondary metabolite determination

2.4 GC–MS analysis

The separation and identification of the compounds present in the unmodified methanol extracts were performed via a Clarus 690 SQ8T GC‒MS gas chromatograph equipped with a MultiPrep™ autosampler. The oven temperature was initially set at 50°C for 10 min and then increased to 280°C at a rate of 10°C/min.

It was then held constant at 200°C for 10 min before being raised to 240°C at a rate of 1°C/min. The injector temperature was set to 280°C. A volume of 0.5 µl was injected into a capillary column via microsyringes. The carrier gas was helium (1 ml/min), an inert gas that pushes the volatile compounds in the sample toward the detector. For this analysis, the split ratio was set to 20:1. This corresponds to the ratio of the volume injected into the column to the volume directed outside the device to avoid column saturation.

The results are analyzed via an automatic qualitative method that introduces integration and rejection parameters to achieve the best possible integration and, consequently, the most accurate identification. This method involves automatic identification by national institute of standards and technology (NIST), as the number of peaks is far too large to be performed manually. This method produces reliable results and enables the presence of molecules or families of molecules to be accurately identified.

2.5 Insect collection

Tomato leaves infested with T. absoluta were collected on multiple occasions from the experimental farm of the University of Mostaganem. The samples were subsequently transported to the laboratory to obtain the various life stages of T. absoluta for use in in vitro bioassays.

2.6 Contact toxicity assays

Biological tests were conducted in a laboratory under controlled conditions at 25.5 ± 3 °C, 53 ± 8% relative humidity, and a 12:12 h light:dark photoperiod for the four larval stages of T. absoluta. For each test, five healthy larvae of the same developmental stage were meticulously placed on a tomato leaflet in a Petri dish. The methanolic extract was tested at six different concentrations (5%, 10%, 15%, 20%, 25% and 30%). Additionally, control larvae were maintained under identical conditions in the absence of any extract. The positive control group was treated with 10% acetone, while the negative control group was administered distilled water. For each concentration as well as for the control, five repetitions were performed. The number of larvae that had perished in each Petri dish was recorded at 24-h intervals throughout the seven-day exposure period. The mortality percentage was calculated following the formula established by Abbott (1925). To determine the Lethal Concentration 50 (LC₅₀) and Lethal Concentration 90 (LC₉₀) values, probit analysis, as outlined by Finney (1971), was employed.

2.7 In silico molecular docking

Molecular docking was carried out to examine how major plant-derived compounds interact with key protein targets of Tuta absoluta. The 3D structures of the ecdysone receptor (2R40) and AChE (4EY7) were retrieved from the Protein Data Bank and prepared in Chemira 1.18 by removing water and ligands, adding hydrogens, and assigning Assisted Model Building with Energy Refinement force field 14SB (AMBER ff14SB) charges. The processed protein models were then saved for subsequent docking analyses (Kouider Amar et al., 2025).

Ten major compounds identified in the methanolic extracts of M. rotundifolia, C. coronarium and U. membranacea were selected as ligands for docking. Their 3D structures were optimized and assigned AM1-BCC partial charges to ensure accurate electronic representation. The selected compounds included neophytadiene, photochemical acid, palmitic and linolenic acid methyl esters, phytol, a cyclobutaneacetonitrile derivative, farnesene, dehydrohumulinic acid, and a pyrazolinone derivative.

Docking calculations were performed via the AutoDock Vina algorithm within Chemira 1.18. For the ecdysone receptor (PDB ID: 2R40), the grid box was centered at coordinates (25.70, −12.19, −20.48) Å with dimensions of 47.38 × 55.91 × 64.38 Å. For AChE (PDB ID: 4EY7), the grid was centered at −3.23, −50.39, 1.39) Å with dimensions of 53.84 × 73.92 × 123.66 Å. The simulation was configured to generate 10 binding modes per ligand with an exhaustiveness of 8. The best-ranked pose for each complex, which is based on the Vina score (kcal/mol), was visualized via Discovery Studio 2024 to analyze intermolecular interactions (Amara et al., 2025).

2.8 Data analysis

Two-way Analysis of variance (ANOVA) was performed with plant species and concentration as fixed factors. Temporal measurements of daily mortality over seven days were analyzed via repeated-measures ANOVA. The assumptions of normality and homogeneity of variance were verified via the Shapiro–Wilk test and Bartlett test, respectively. Significant differences between means were determined via Tukey’s multiple comparison test at P ≤ 0.05. All the statistical analyses were conducted with R software (version 4.2.1).

3.1 Plant extract yield

The results of the crude extracted yields of the plants studied showed that, under the same extraction conditions, the amount of extract obtained varied greatly from one plant species to another. The highest yield obtained was 55% for the crude extract of membrane nettle, followed by that of crown chrysanthemum (35%). However, the round-leaved mint extract had the lowest yield, at 26% (Fig. 1). The extraction yield is a measure of the efficiency of the extraction method and the solvent used to extract the specific constituents from the plant matrix (Gurnani et al., 2016; Adam et al., 2019). The efficiency of an extraction process depends on the chemical nature of the phytochemical compounds, temperature, solvent, polarity and time of extraction (Do et al., 2014).

Close

Fig. 1.

Yields (%) of the plant extracts obtained via Soxhlet extraction.

Export to PPT

3.2 Quantitative analysis

3.3 Chemical composition analysis

3.4 Insecticidal effects of the methanolic extracts on T. absoluta

3.4.1 Biocidal efficacy of the methanol extract of U. membrancea

An in vitro test revealed that the nettle extract had a larvicidal effect on tomato leaf miner (TLM) larvae, which varied according to dose and time. Mortality rates ranging from 3.44 to 22.41% were observed after only 24 h of application. At the end of the test, significant percentages of 75, 76.92, and 83.33% were obtained at the respective doses of 20, 25, and 30%, respectively. The lowest percentage (40%) was obtained with the 5% formulation (Figs. 2 and 3).

Close

Fig. 2.

Corrected mortality rate of T. absoluta as a function of exposure time and concentration of the U. membranacea methanolic extract.

Export to PPT

Close

Fig. 3.

Dead T. absoluta larvae treated with extracts of U. membranacea (a) Second larval instar, (b) Fourth larval instar.

Export to PPT

The variation in the percentage mortality as a function of the methanolic extract of U. membranacea and the controls on the T. absoluta larval population was highly significantly different (P<0.001), revealing quite different rates between the negative controls and the treated boxes and a slight difference in the latter compared with the positive controls (Fig. 4).

Close

Fig. 4.

Effects of the U. membranacea methanolic extract on the mortality of T. absoluta larvae seven days after treatment. The values represent the mean mortality percentages ± standard errors (n = 5). Different letters above the bars indicate significant differences among the treatments (one-way ANOVA followed by multiple comparison test, p < 0.05).

Export to PPT

3.4.2 Biocidal efficacy of the methanol extract of M. rotundifolia

The results demonstrated that T. absoluta caterpillars were sensitive to mint extract. The treatments were dependent on time, resulting in variable mortality rates with increasing concentration and time.

The temporal evolution of mortality was observed to result in deaths of 5.35, 16.07, 19.64, 26.78, 30.36 and 32.14% of the caterpillars after 24 h following the application of 5, 10, 15, 20, 25 and 30% of the extract, respectively. On the final day of observation, the highest mortality rate was recorded at the 30% concentration, at 76.74%, whereas the lowest mortality rate was recorded at the 5% concentration, at 37.49% (Figs. 5 and 6).

Close

Fig. 5.

Corrected mortality rate of T. absoluta versus time and concentration of the methanolic extract of M. rotundifolia.

Export to PPT

Close

Fig. 6.

Dead T. absoluta larvae treated with the extract of M. rotundifolia (a) Third larval instar, (b) Fourth larval instar.

Export to PPT

Analysis of variance was applied to the different doses of M. rotundifolia extract against the four larval stages of T. absolua. The results revealed a highly significant difference (P <0.001) between the doses of the same plant and the controls (Fig. 7).

Close

Fig. 7.

Effects of the M. rotundifolia methanolic extract on the mortality of T. absoluta larvae seven days after treatment. The values represent the mean mortality percentages ± standard errors (n = 5). Different letters above the bars indicate significant differences among the treatments (one-way ANOVA followed by multiple comparison test, p < 0.05).

Export to PPT

3.4.3 Biocidal efficacy of the methanol extract of C. coronarium

Twenty-four h after treatment, the mortality rate reached 50% at concentrations of 25 and 30%. The values obtained for concentrations of 5, 10, 15 and 20% were lower than those obtained for the other concentrations at 3.57, 17.85, 31.81 and 39.28%, respectively. An increase in the mortality rate was observed from the second day onward, except for the 5% dose, where mortality was estimated at 3.63%. Conversely, the highest mortality rate of 76.74% was observed on the final day of the experiment at the 30% concentration (Figs. 8 and 9).

Close

Fig. 8.

Corrected mortality rate of T. absoluta versus time and concentration of the C. coronarium methanolic extract.

Export to PPT

Close

Fig. 9.

Dead T. absoluta larvae treated with the extract of C. coronarium (a) second larval instar, (b) fourth larval instar.

Export to PPT

The percentages of temporal mortality as a function of the different dosages of chrysanthemum extract and the control over the four larval stages of the Tomato leaf miner (TLM) were highly significantly different (p <0.001). The in vitro test demonstrated that, in comparison with the other stages, the first and second larval stages presented the greatest sensitivity to all concentrations (Fig. 10).

Close

Fig. 10.

Effects of the C. coronarium methanolic extract on the mortality of T. absoluta larvae seven days after treatment. The values represent the mean mortality percentages ± standard errors (n = 5). Different letters above the bars indicate significant differences among the treatments (one-way ANOVA followed by multiple comparison test, p < 0.05).

Export to PPT

3.5 Analysis of variance of the insecticidal effects of hydroalcoholic extracts on T. absoluta larvae

The variance analysis revealed that the methanolic extracts had significantly different effects on T. absoluta (p < 0.05). The Chrysanthemum extract had the greatest contact toxicity, whereas the U. membranacea and M. rotundifolia extracts were less effective (Fig. 11).

Close

Fig. 11.

Effects of three methanolic extracts on the mortality of T. absoluta larvae seven days after treatment. The values represent the mean mortality percentages ± standard errors (n = 5). Different letters above the bars indicate significant differences among the treatments (one-way ANOVA followed by multiple comparison test, p < 0.05).

Export to PPT

3.6 Estimation of _{the LC50s} and _LC90s of the methanolic extracts tested on T. absoluta larvae

Seven days after exposure, probit analysis was conducted on the corrected mortalities, resulting in LC₅₀ values of 6.30, 6.91 and 9.12% for the U. membranacea extract, M. rotundifolia extract and C. coronarium extract, respectively. In contrast, the LC₉₀ was 56.23% for the nettle extract, 46.77% for the round-leaved mint extract and 67.60% for the crown chrysanthemum extract.

The bioassay results demonstrated that all the tested methanolic extracts had concentration- and time-dependent toxic effects on T. absoluta larvae. Mortality rates increased significantly with both exposure duration and extract concentration, revealing clear differences in efficacy among the three plant species. As demonstrated by Bokobana et al., (2014), mortality rates in insect bioassays are strongly influenced by both exposure time and product concentration. In the present study, the application of methanolic plant extracts under laboratory conditions clearly induced behavioral and physiological alterations in T. absoluta larvae. After treatment, larvae exhibited feeding inhibition, reduced mobility and progressive paralysis, eventually leading to color changes and death (Figs. 3,6, and 9). These symptoms are consistent with the neurotoxic effects common with botanical insecticides.

According to Keddar et al., (2022), such effects may result from the interference of plant-derived compounds with the normal functioning of insect organs and nervous tissues, even at low concentrations. This disruption could involve alterations in nerve cell membranes or the inhibition of key enzymes essential for neurotransmission. The larvicidal activity observed in the present study is therefore attributed to the combined action of multiple bioactive constituents detected in the phytochemical analysis, including phytol, neophytadiene, α-farnesene, and dehydro-cohumulinic acid, which are known for their insecticidal and neurophysiological activities (Gurjar et al., 2012).

In the context of the literature, no previous research has reported the insecticidal potential of hydroalcoholic extracts from U. membranacea, M. rotundifolia or C. coronarium against T. absoluta. However, comparative studies with other aromatic and medicinal species support our findings. For example, Ait Taadaouit et al., (2012) reported mortality rates of up to 97% for Thymus vulgaris, followed by 80%, 65% and 55% for Ricinus communis, Peganum harmala, and U. dioica, respectively. Similarly, Ndereyimana et al., (2019). reported mortality rates ranging from 10.6% to 35.1% after five days of exposure to extracts from Tithonia diversifolia, Tephrosia vogelii, Phytolacca dodecandra and Vernonia amygdalina. Comparable results were also obtained by de Brito et al., (2015), who demonstrated that hydroalcoholic extracts of four Piper species induced mortality rates between 43.81% and 67.62% under laboratory conditions.

These findings confirm that the efficacy of botanical extracts depends on both their phytochemical profiles and extraction methods. In our study, the strong insecticidal potential, particularly that of C. coronarium, can be attributed to the synergistic action of its major constituents, such as dehydro-cohumulinic acid and α-farnesene, which, as revealed by molecular docking analysis, exhibit high affinity toward key insect targets, including AChE and the ecdysone receptor. This dual mechanism likely disrupts both neuromuscular activity and molting processes, leading to increased larval mortality.

3.7 Molecular docking

To understand why the plant extracts were effective at the molecular level, we performed in silico docking simulations. We selected ten major compounds identified from the extracts of U. membranacea, M. rotundifolia and C. coronarium. These phytochemicals were tested against two key insect proteins essential for survival: the ecdysone receptor (PDB ID: 2R40), which regulates growth and development, and AChE (PDB ID: 4EY7), a critical enzyme in the nervous system. The binding affinities, which indicate the strength of the interaction between a compound and its target, are summarized in Fig. 12.

Close

Fig. 12.

Heatmap showing the binding affinities (kcal/mol) of the ten major phytochemicals from the three extracts against the ecdysone receptor (2R40) and AChE (4EY7). Darker colors indicate stronger binding.

Export to PPT

The results clearly identified a top-performing compound. Dehydro-cohumulinic acid (9) exhibited the strongest binding affinity for both the ecdysone receptor (–7.5 kcal/mol) and AChE (–8.9 kcal/mol). Farnesene (7) also demonstrated very strong binding to both targets. Importantly, both of these highly effective compounds are major constituents of the C. coronarium extract, which was the most potent insecticide in our laboratory bioassays. This direct correlation between the docking results and larval mortality confirmed that the chemical composition of the extract strongly determines its biological activity.

These docking results therefore provide a molecular explanation for the bioassay results. By binding tightly to both the ecdysone receptor and AChE compounds such as dehydro-cohumulinic acid can simultaneously disrupt two critical physiological processes: hormonal regulation of molting and nerve signal transmission. This dual-target mechanism likely contributes to the high mortality observed in the bioassays. Furthermore, the presence of multiple active compounds in the crude extracts suggests that they may act synergistically, increasing overall toxicity to T. absoluta larvae and supporting their development as effective biopesticides.

Fig. 13 provides a comprehensive illustration of the 2D interaction maps and 3D binding poses of dehydro-cohumulinic acid within the active sites of the ecdysone receptor and AChE . As a top-performing ligand, it has binding affinities of –7.5 kcal/mol for 2R40 and –8.9 kcal/mol for 4EY7, making its detailed interactions particularly significant. In Fig. 13(a), which depicts the ecdysone receptor, the 2D representation reveals a strong conventional hydrogen bond with the phenolic hydroxyl group of TYR D:408, serving as a key anchoring point. Additionally, a notable π–π stacking interaction with the aromatic ring of PHE D:397 was observed, indicative of stable parallel stacking. Further alkyl contacts with residues such as LEU D:349, PRO D:311, and ALA D:398 contribute to the favorable positioning of the ligand within the hydrophobic pocket of the protein, as shown in the 3D surface representation.

Close

Fig. 13.

Representative 2D and 3D binding interactions of dehydro-cohumulinic acid with target proteins: (a) ecdysone receptor (PDB ID: 2R40) and (b) AChE (PDB ID: 4EY7). Key interactions such as hydrogen bonds, hydrophobic contacts, and π–π interactions are highlighted.

Export to PPT

For AChE (4EY7), as shown in Fig. 13(b), dehydro-cohumulinic acid forms critical interactions within the active site. The 2D diagram highlights prominent π–σ interactions with the aromatic residues TRP A:286 and TRP A:86, indicating favorable edge-to-face or parallel-displaced aromatic contacts. A significant π–π stacked interaction is also observed with TYR A:341, suggesting a strong parallel arrangement between the ligand’s ring system and this tyrosine residue. Additionally, an alkyl interaction with VAL A:294 contributes to overall hydrophobic binding. The 3D surface model clearly illustrates how the ligand’s structure is optimally accommodated within the enzyme’s binding pocket, maximizing these combined aromatic and hydrophobic interactions. This multimodal binding profile aligns with the high larval mortality observed in vitro, linking these molecular interactions directly to the biological activity of the extract.

A broader analysis of all ten compounds (Figs. 14(a,b) revealed that although hydrophobic interactions are the primary driving force for binding, other forces, such as hydrogen bonds, play crucial roles in stabilizing the compounds within the target sites. The key takeaway is that efficacy does not depend on a single type of interaction but rather on a combination of forces. This diversity in binding mechanisms likely explains the variation in insecticidal potency observed among the plant extracts in the bioassays.

Close

Fig. 14.

Modes and distances of ligand interactions with both receptors. Panel (a) illustrates the interaction modes for the ecdysone receptor, and panel (b) illustrates the interaction modes for AChE . Panels (c–e) depict details of the binding of compounds from the three extracts to 2R40, whereas panels (f–h) show binding details to 4EY7. The distances are in Å, and the interaction types are indicated (hydrogen bonds, π–π stacking, hydrophobic contacts).

Export to PPT

Figs. 14(c-h) provide a detailed visual representation of the specific interaction types and their corresponding distances between the selected plant ligands and the active sites of the ecdysone receptor and AChE.

The group of ligands, including neophytadiene (1), photochemical A (2), palmitic acid methyl ester (3), linolenic acid methyl ester (4), and phytol (5), corresponds to the major compounds found in U. membranacea, as shown in Fig. 14(c). The binding of these compounds to the ecdysone receptor is predominantly driven by π−alkyl and alkyl hydrophobic interactions. These interactions are widely distributed across numerous amino acid residues with distances generally ranging from 3 to 5 Å. Additionally, some π–σ, carbon hydrogen bonding, and conventional hydrogen bonding interactions are observed, although less frequently. Notably, dehydro-cohumulinic acid (9) exhibited the strongest binding affinity of –7.5 kcal/mol to 2R40, likely because its extended hydrophobic chain facilitates extensive alkyl and π−alkyl contacts.

Fig. 14(d) illustrates the interactions of cyclobutaneacetonitrile 1-methyl-2-(1-methylethenyl)- (6) with 2R40. This ligand primarily engages in alkyl interactions at distances greater than 5 Å. A π–σ interaction with PHE D:397 is observed, along with a significant conventional hydrogen bond formed with ARG D:383 at a very close distance of 2.35 Å. Despite this strong hydrogen bond, the overall binding affinity of −6.4 kcal/mol suggests that the limited number and diversity of interactions, compared with those of other compounds, may reduce its stability within the binding pocket. This finding is consistent with the relatively low insecticidal effect observed in bioassays for M. rotundifolia, where this compound is dominant.

Fig. 14(e) illustrates the binding interactions of farnesene (7), neophytadiene (8), dehydro-cohumulinic acid (9), and 4-dimethylsulfylidene-3-methyl-1-phenyl-2-pyrazolin-5-one (10) with the 2R40 receptor. This group exhibits a broad spectrum of interactions, including predominant π–σ, alkyl, and π–alkyl contacts with various residues. Notably, shorter-distance conventional hydrogen bonds are observed, along with π–π stacking, π–π positive‒positive interactions, and some unfavorable positive‒positive interactions, particularly for 4-dimethylsulfylidene-3-methyl-1-phenyl-2-pyrazolin-5-one (10). The highly favorable binding affinities of dehydro-cohumulinic acid (9) (−7.5 kcal/mol) and 4-dimethylsulfylidene-3-methyl-1-phenyl-2-pyrazolin-5-one (10) (−7.4 kcal/mol) can be attributed to this rich and diverse interaction profile, which facilitates strong anchoring within the receptor’s active site. These strong affinities mirror the high mortality caused by C. coronarium, which contains these top-binding compounds.

The interaction patterns of major compounds from U. membranacea with AChE (PDB ID: 4EY7) are illustrated in Fig. 14(f). Similar to the interactions observed with 2R40, π-alkyl and alkyl interactions predominated. Multiple conventional hydrogen bonds and carbon‒hydrogen bonds are also present at relatively short distances (approximately 2–3 Å), indicating strong polar contacts. The strong binding affinities of neophytadiene (1) and phytol (5) (both −8.0 kcal/mol), as well as linolenic acid methyl ester (4) (−7.9 kcal/mol) to 4EY7, are likely enhanced by this extensive network of hydrophobic and critical hydrogen-bonding interactions. These affinities are consistent with the moderate mortality recorded for U. membranacea via laboratory assays.

As shown in Fig. 14(g), cyclobutaneacetonitrile 1-methyl-2-(1-methylethenyl)- (6), the sole major compound found in M. rotundifolia, interacts with AChE (PDB ID: 4EY7). It forms alkyl, π−alkyl, and some π−σ interactions; however, the most significant interaction is a conventional hydrogen bond with GLY B:121 at a distance of 2.91 Å. Despite this hydrogen bond, its binding affinity of –6.6 kcal/mol is the weakest among all the ligands tested against 4EY7, indicating that the overall number or quality of interactions is insufficient for strong binding. These results align with the weak insecticidal activity observed for this plant.

Finally, Fig. 14(h) illustrates the highly diverse interaction landscape of major compounds from C. coronarium with 4EY7. The plant’s bioactive compounds exhibit a comprehensive array of interactions, including π−alkyl, alkyl, π−σ, π−π stacked, π−π T-shaped, π−sulfur, π−cation, and conventional hydrogen bonds. These interactions involve numerous residues, indicating a highly complementary fit within the enzyme’s active site. The exceptionally strong binding affinity of dehydro-cohumulinic acid (9) (−8.9 kcal/mol), followed by farnesene (7) (−8.6 kcal/mol), directly results from this rich and diverse interaction profile. The presence of specific aromatic (π−π, π−sulfur) and charged (π−cation) interactions, alongside extensive hydrophobic and hydrogen bonding, underscores the multifaceted engagement of these compounds with the enzyme, providing a robust molecular basis for the high mortality observed with C. coronarium extract in bioassays.