Chemical Composition and Insecticidal Activity of Eschweilera jefensis Organic Extracts Against Aphis gossypii

Abstract

Aphis gossypii is a major pest that harms crops like industrial tomatoes in Panama. Recent resistance to synthetic insecticides has prompted interest in using plant secondary metabolites as eco-friendly alternatives. While some plants with insecticidal properties are well-known, others remain unexplored but could offer effective solutions. This study aimed to evaluate the insecticidal activity of ethanolic extracts from the stems and leaves of Eschweilera jefensis against nymphs and adults of Aphis gossypii. Extracts were tested at three concentrations (25, 50, and 100 µg/L), with mortality assessed at 24, 48, and 72 h post-application. The LC₅₀ values for the stem extract were 66.5, 36.8, and 31.0 μg/L, and for the leaf extract, they were 37.3, 28.4, and <25 μg/L at 24, 48, and 72 h, respectively. An advanced metabolomic analysis was conducted to identify the active compounds in each extract. This analysis uncovered several pentacyclic triterpenes, which, known for their insecticidal properties, are likely the key bioactive components responsible for the observed effects. Advanced metabolic analyses also revealed that the leaf extract, displaying the strongest insecticidal activity, is primarily composed of friedelin, while the stem extract contains betulin as their key active compounds. Furthermore, 29 known compounds were identified across both extracts, representing the first comprehensive report on the metabolic composition of E. jefensis, which underscores the significance of these findings. Together, these results suggest that E. jefensis extracts could serve as a promising natural alternative to synthetic insecticides for the management and control of A. gossypii.

1. Introduction

Agriculture is a crucial activity for humanity, serving as the foundation of food production and playing a fundamental role in the economy and social development of many countries, particularly those that are agriculturally focused. In recent years, the importance of sustainable agriculture has been emphasized as essential for protecting the environment and ensuring food security for future generations. Every year, agricultural crops face constant threats from various pests, including insects, making pesticides essential for mitigating economic losses associated with reduced crop yields [1,2]. In natural habitats, insects seek food sources, and crop fields provide an accessible option, leading to significant decreases in yields and substantial economic losses. In this context, synthetic insecticides are important and effective tools in modern crop management. However, these products also pose serious threats to the environment and human health [3].

Many plant species are known to produce compounds that protect them from insect pests by either killing or repelling them. This presents the possibility of developing natural insecticides from these compounds to protect crops which may have lost their natural defenses due to the cultivation of plants that have been domesticated and grow thanks to human planting and care. Compared to synthetic pesticides, natural alternatives offer several benefits, including lower environmental costs. They are biodegradable, leave fewer residues in the soil, and are generally safer for humans and animals. Additionally, they tend to be more affordable and easier for farmers in developing regions to access [3,4].

Plants are known for their traditional use as insecticides, yet some plants with potential insect-repelling properties remain overlooked or unidentified, offering opportunities for discovering new natural insecticides. One such lesser-known plant is Eschweilera jefensis, a woody species in the Lecythidaceae family, native to the cloud forests of Panama at elevations between 800 and 1000 m [5]. Despite being documented as recently as 2017, there is still limited information available on its secondary metabolites and possible biological activities.

To address this problem, our research group launched a bioprospecting project focused on exploring the rich biodiversity of Panama to identify potential natural products that could be used to control A. gossypii. Given the significant economic damage caused by this pest in agriculture, developing more effective pest control methods is essential for farmers [4]. Currently, synthetic insecticides are the most widely used solution for managing A. gossypii. However, the overuse of these chemicals has led to the rapid development of resistance in A. gossypii to multiple insecticide classes, including organophosphates, pyrethroids, and neonicotinoids [6]. Moreover, the environmental and ecological impacts of these insecticides, along with their harmful effects on non-target species, are well-documented [7]. As a result, the heavy reliance on synthetic insecticides highlights the urgent need for more sustainable alternatives. Natural insecticides, derived from plants or other natural sources, have emerged as a promising alternative, offering an eco-friendlier pest management approach with fewer negative consequences for both the environment and farming communities [4].

This project employs direct toxicity tests using a mixture of adult aphids and four-stage nymphs of A. gossypii within the same treatment. This approach reflects real-world conditions, where the natural composition of the colonies of this aphid that directly affect tomato crops in Panama primarily consists of nymphs and adults in varying proportions. Furthermore, our objective is to assess the overall effect of the extract on the colony, rather than to examine differences in susceptibility between aphid life stages. Additionally, advanced metabolomic techniques were employed to identify potential bioactive components. Through this initiative, we aim to contribute to the rapid discovery of new natural sources with insecticidal potential, while also promoting sustainable agriculture. By reducing dependence on synthetic pesticides, we seek to mitigate their negative impact on the environment.

2. Materials and Methods

2.1. Plant Material

The permission for the plant collection was obtained from the country’s Ministry of the Environment of Panama (MiAmbiente). E. jefensis (Lecythidaceae) was collected in February 2021 in Cerro Jefe (9°13′33.0″ N, 79°22′36.9″ W), province of Panama, Republic of Panama. The plant was identified for comparison with a E. jefensis previously deposited in the herbarium of the University of Panama and using the work reported by Batista et al. 2017 [5]. The plant aerial parts (leaves and stems) were carefully collected, placed in plastic bags, and transported to the laboratory for subsequent extract preparation. During the plant collection process, great care was taken to select organs free of disease and without apparent insect damage [4]. The plant species’ type material was identified and entrusted to the Laboratory of the Vice-Presidency of Research and Postgraduate Studies at the University of Panama, where it was deposited for future reference.

2.2. Preparation of Extracts

The collected plant material was thoroughly washed to remove any sand, dust, or other contaminants, then left to air dry at room temperature for 5–6 days in a clean environment. After drying in the shade, the stems were separated from the leaves and then crushed. The crushed plant material was subjected to a maceration process. In this process, 500 g of material from each plant part was placed in Erlenmeyer flasks and fully covered with 99% ethanol. The same amount of material was used for each organ (leaves and stems), which were processed separately. The mixture (ground leaves or stems plus extraction solvent) was left to macerate for 48 to 72 h in a cool, dark environment. Afterward, the liquid was filtered and concentrated under a vacuum in a rotary evaporator until dryness was reached. The dried extract was placed in a sterile vial and stored at 5 °C until the bioassay was performed. For the ethanol extract bioassay, 10 μL samples containing 100 µg/L, 50 µg/L and 25 µg/L of the crude extract were taken [4].

2.3. Bioassay

In this study, a bioassay was conducted using adult and fourth-instar nymphs of A. gossipii, following the residual toxicity (leaf-dip) method outlined by Ahmed et al. (2020) [8] with modifications. Aphid colonies were collected from agricultural fields in Villa Lucre, Panama City, and were maintained in the laboratory on Catharanthus roseus plants without prior exposure to insecticides. These colonies were acclimatized for 48 h in a controlled chamber set to 28 ± 4 °C, 70 ± 5% humidity, and a 12:12 h light/dark cycle before starting the bioassay. Adult aphids were distinguished from fourth-instar nymphs by body size (>1.2 mm), sclerotization of the thoracic dorsum, and the presence of fully developed siphunculi and cauda, whereas fourth-instar nymphs were identified by smaller size (0.9–1.1 mm) and incomplete sclerotization. 10 mL of each ethanolic extract solutions was prepared at concentrations of 25, 50, or 100 µg/L using the necessary amount of extract for the desired final concentration diluted in 9 mL of distilled H₂O and 1 mL of 99% ethanol. For each treatment, C. roseus leaves, approximately 5 cm in diameter, were dipped for 10 s into ethanolic solutions of extracts (10 μL). This exposure time was chosen based on a validated protocol to ensure consistent residue coverage without damaging the leaf tissue. The treated leaves were placed in 6 cm Petri dishes containing a 2% agar layer to maintain moisture, and allowed to sit at room temperature for 24 h to allow the solvent to evaporate and prevent excess moisture, which could negatively impact aphid survival. Ten aphids were carefully transferred to each treated leaf using a fine brush, taking care not to injure them. Aphids were allowed to feed freely for the entire exposure period. Treatments were randomly assigned to the Petri dishes, and mortality was assessed by an independent observer blinded to the treatment groups. Each treatment and control group were replicated three times with 10 aphids per dish, and the experiment was repeated once under identical conditions, resulting in six total replicates (60 individuals in total per treatment). The negative control consisted of a mixture of 9 mL distilled water and 1 mL 99% ethanol, while the positive control used Imidacloprid^® at a concentration of 0.04 mg/L, reflecting the field-recommended dose for aphid control. Mortality was recorded at 24, 48, and 72 h after treatment, with individuals deemed dead if no movement was observed after being gently prodded with a fine brush [4]. The following formula was used to obtain the mortality rate for each sample analyzed:

$$\mathrm{M}\mathrm{o}\mathrm{r}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}\left(\mathrm{\%}\right)=\left({\displaystyle \frac{\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r} \mathrm{o}\mathrm{f} \mathrm{d}\mathrm{e}\mathrm{a}\mathrm{d} \mathrm{a}\mathrm{p}\mathrm{h}\mathrm{i}\mathrm{d}\mathrm{s}}{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{n}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r} \mathrm{o}\mathrm{f} \mathrm{a}\mathrm{p}\mathrm{h}\mathrm{i}\mathrm{d}\mathrm{s}}}\right)\times 100$$

Statistical analyses were performed using R software v4.5.0 (R Foundation for Statistical Computing, Vienna, Austria), and LC₅₀ values were estimated using the drc package by fitting a log-logistic model to the mortality data [9,10].

2.4. GC-MS-Based Metabolomic Analysis of the E. jefensis Extracts

An untargeted metabolomic analysis of the organic extracts prepared from the leaves and stems of from E. jefensis was carried out using an Agilent 8890 Gas Chromatograph coupled to an Agilent 5977C mass spectrometer. Chromatographic analysis was performed using an HP-5MS capillary column (30 m length, 25 mm ID, 0.25 μm film thickness, Agilent), and high-purity helium was used as the carrier gas at a constant flow rate of 1.1 mL/min. The method used for gradual increase in the gas chromatograph’s column oven temperature during analysis to enhance the separation and resolution of compounds contained in the organic extracts analyzed started at 150 °C, followed by a ramp of 3 °C/min to 250 °C and a subsequent ramp of 2 °C/min to 305 °C, with a 5 min hold at 305 °C. For mass analysis, the electron impact (EI) ion source was maintained at 250 °C with a filament bias of −70 V. Data were collected in full scan mode (m/z 30–600) at a rate of 20 spectra per second in the MS mode. Data acquisition and processing were conducted using Agilent Mass Hunter Workstation software (version 10.1.49). The mass spectra for each peak were matched against the NIST20 spectral library, using a retention similarity threshold of 80% or higher [4,11].

2.5. Molecular Networking

Molecular networks based on mass spectrometry data were created using the GNPS platform (http://gnps.ucsd.edu, accessed on 6 August 2025), using GC-EI/MS data that was first processed with Agilent’s Mass Hunter software (version 10.1.49). Since the data from the Electron impact ionization experiments lacked pre-selected precursor ions (using the DIA acquisition format), spectral deconvolution was required. To address this, the GC-MS data were analyzed with the MSHub algorithm contained within the same GNPS platform [12].

The raw data was sent to the GNPS platform for processing using the spectral network algorithm with the following settings: a fragment ion mass tolerance of 0.5 Da, at least 5 matched peaks, and a score threshold of 0.7. On the other hand, regarding the search options, the following parameters were considered: a gold library class, selecting the top 10 history per spectrum, and utilizing both NIST and GNPS spectral libraries. The minimum pair cosine similarity was set to 0.85, and the network topK to 15 in the advanced network options. Specific details of the network generated with our data in GNPS can be found by visiting the webpage: https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=07c08e8793a947c4a97a65650a4076c5, accessed on 22 August 2025. For better visualization of the network, the data obtained from GNPS were analyzed using Cytoscape v.3.4.3. In the reported network, the colors and sizes of the nodes were established according to the metadata files, and the thickness of the edges expresses the cosine similarity scores, with thicker lines representing a higher degree of similarity [4,11].

In order to generate information about the secondary metabolites produced by E. jefensis and gain a comprehensive understanding of the compound’s types produced by this plant, we analyzed the data obtained by GC-MS using NPClassifier (Version 1.5), a deep neural network designed to automatically classify natural products based on their structural fingerprints [13]. NPClassifier converts the structures of natural products into molecular fingerprints, which are then processed by a supervised deep neural network with a feedforward architecture. During training, the model learns to predict the hierarchical classification of natural products. Once trained, it classifies new natural products by processing their fingerprints through the network and generating probabilities for each category. The model parameters are optimized using techniques such as batch normalization, data loss minimization, and the hyperband algorithm for hyperparameter tuning.

2.6. Heatmap of the Compounds Identified in the Leaf and Stem Extracts of Eschweilera jefensis

To visualize the relative abundance of the chemical compounds identified in the leaf and stem extracts, a heatmap was generated using the Python programming language within the Google Colaboratory environment (also known as Colab) offered by Google [14], utilizing the Pandas, Matplotlib, Seaborn, NumPy, and Matplotlib.colors libraries. First, the abundance data, expressed in peak area, were normalized and processed to ensure numerical consistency and address any missing values. A log1p transformation was then applied to the abundance data to compress the range of values, allowing for effective visualization of both high and low abundances. The resulting table of transformed abundances was transposed to position the sample types on the X-axis and the chemical compounds on the Y-axis. Finally, the heat map was generated using a viridis color palette to visually represent the transformed abundance levels, accompanied by a color bar indicating the applied logarithmic scale.

3. Results

3.1. Insecticidal Activity

The mortality data for A. gossypii obtained after exposure to ethanolic extracts of E. jefensis are in

Table 1. Toxicity of E. jefensis extracts against A. gossypii after an exposure of 24, 48 and 72 h by the residual/leaf dip method. Note: LC₅₀ (Lethal Concentrations); S.E. (Standard Error); χ² (chi-square); F.L. (Fiducial Limit). * Indeterminate data within the concentration ranges tested.

Plant Extract	Time (h)	LC50 (μg/L)	95% F.L.		Slope ± SE	χ2
			Lower	Upper
Stem	24	66.5	27.63	332.7	1.63 ± 0.23	0.16
	48	36.8	23.26	109.16	1.24 ± 0.27	1.10
	72	31.0	18.61	87.35	0.99 ± 0.22	0.88
Leaves	24	37.3	16.27	195.94	0.43 ± 0.14	0.99
	48	28.4	19.21	90.14	1.03 ± 0.22	0.91
	72	>25 *	*	*	*	*

. These data are presented in terms of the lethal concentration 50 (LC₅₀). The results clearly demonstrate that the mortality of A. gossypii was directly correlated with both the concentration of the plant extracts and the duration of exposure. The maximum mortality occurred after 72 h of exposure to E. jefensis extract from both leaves and stems (LC₅₀ < 25 μg/L and 31.0 μg/L, respectively), at a concentration of 100 μg/L. In contrast, after 48 h of exposure, the LC₅₀ increased to 28.4 and 36.8 μg/L, respectively. The highest LC₅₀ values were observed at 24 h, being 37.3 and 66.5 μg/L, respectively. Additionally, the commercial synthetic insecticide Imidacloprid^® was used as a positive control at a single concentration, corresponding to the recommended field application concentration on the product label. This positive control resulted in 80 ± 3.33% mortality at 24 h, with mortality reaching 100% at 48 h. As a negative control, ethanol solution was applied under the same conditions as the samples, showing minimal mortality of 5.00 ± 1.66%, observed only at 72 h. Probit analysis was performed to determine the LC₅₀ values, slope, chi-square, and fiducial limits at the 95% confidence interval. In general, the leaf extract was found to be more toxic than the stem extract.

3.2. GC-MS-Based Metabolomic Analysis

Figure 1 (view between 12 and 29 min) and Figure 2 (detailed view between 44 and 58 min) display the chromatographic results obtained under optimal conditions. These chromatograms highlight clear differences in the major components between the leaf and stem extracts. The main compounds in the leaf extract appear at 51.51 and 53.29 min, while in the stem extract, they are observed at 51.32 and 56.73 min. The expanded chromatogram also reveals notable differences in minor components between the extracts from both plant parts. Furthermore, differences in peak retention times suggest that certain metabolites may contribute to the observed differences in biological activity.

Table 2. Chemical composition of Eschweilera jefensis stem extract.

No.	Compound	RT (min)	A%	MW (g/mol)	MF
1	Neophytadiene	12.47	0.13	278.5	C20H38
2	Methyl decyl ketone	12.63	0.12	184.32	C15H26O
3	5-(2,5-Dimethoxyphenyl)-5-oxopentanoic acid	14.43	0.49	252.26	C13H16O5
4	Hexadecanoic acid, methyl ester	14.79	0.91	270.45	C17H34O2
5	Hexadecanoic acid	15.67	1.63	256.42	C16H32O2
6	1,3,5-Trimethoxy-2-propenylbenzene	16.32	0.14	208.25	C16H16O3
7	Hexadecanoic acid, ethyl ester	16.66	8.50	284.47	C18H36O2
8	Heptadecanoic acid, ethyl ester	19.43	0.49	298.50	C19H38O2
9	7,10,13-Hexadecatrienoic acid, methyl ester	19.61	0.29	264.40	C17H28O2
10	Phytol	19.92	0.36	296.50	C20H40O
11	Methyl stearate	20.37	0.31	298.50	C19H38O2
12	9,12,15-Octadecatrienoic acid	20.56	0.34	278.42	C18H30O2
13	Linoleic acid ethyl ester	21.28	2.93	308.50	C20H36O2
14	9,12,15-Octadecatrienoic acid, ethyl ester	21.47	2.28	306.48	C20H34O2
15	Ethyl Oleate	21.62	0.88	310.50	C20H38O2
16	Octadecanoic acid, ethyl ester	22.20	2.34	312.53	C20H40O2
17	Perhydro-3-(3,4,a-trimethoxy-2-nitrobenzylidene)-2-oxofuran	23.71	0.28	309.27	C14H15NO7
18	Eicosanoic acid, ethyl ester	27.55	0.15	340.58	C22H44O2
19	Lichexanthone	33.92	0.09	286.28	C16H14O5
20	Stigmasterol	49.74	0.69	412.70	C29H48O2
21	Sitosterol	51.33	10.01	302.50	C20H30O2
22	Beta-Amyrin	51.99	2.51	426.70	C30H50O
23	Lupeol	53.28	47.56	426.72	C30H50O
24	13,15-Octacosadiyne	56.04	4.51	386.70	C28H50
25	Friedelin	56.73	27.73	426.700	C30H50O

RT: Retention Time, A%: Area Percentage, MW: Molecular Weight, MF: Molecular Formula.

and

Table 3. Chemical composition of Eschweilera jefensis leaf extract.

No.	Compound	RT (min)	A%	MW (g/mol)	MF
1	Neophytadiene	12.47	0.13	278.5	C20H38
2	methyl decyl ketone	12.63	0.20	184.32	C12H24O
4	Hexadecanoic acid, methyl ester	14.79	0.48	270.45	C17H34O2
5	n-Hexadecanoic acid	15.67	0.89	256.42	C16H32O2
7	Hexadecanoic acid, ethyl ester	16.66	1.19	284.47	C18H36O2
8	Heptadecanoic acid, ethyl ester	19.43	0.06	298.50	C19H38O2
26	9,12,15-Octadecatrienoic acid, methyl ester	19.63	0.18	292.45	C19H32O2
10	Phytol	19.93	1.90	296.50	C20H40O
11	Methyl stearate	20.35	0.18	298.50	C19H38O2
12	9,12,15-Octadecatrienoic acid	20.56	0.36	278.42	C18H30O2
13	Linoleic acid ethyl ester	21.28	0.29	308.50	C20H36O2
14	9,12,15-Octadecatrienoic acid, ethyl ester	21.47	0.47	306.48	C20H34O2
15	Ethyl Oleate	21.62	0.19	310.50	C20H38O2
16	Octadecanoic acid, ethyl ester	22.19	0.29	312.53	C20H40O2
27	4,8,12,16-Tetramethylheptadecan-4-olide	26.40	0.16	324.54	C21H40O2
28	Alpha-Tocopherol	46.69	0.46	430.71	C29H50O2
29	Fiehn VocBinbase Bin #153	46.98	1.14	-	-
21	Sitosterol	51.33	0.83	302.50	C20H30O2
30	Betulin	51.51	13.81	442.70	C20H50O2
22	Beta-Amyrin	51.99	3.79	426.70	C30H50O
31	Lup-20(29)-en-3-one	52.68	0.82	424.70	C30H48O
23	Lupeol	53.21	1.81	426.72	C30H50O
24	13,15-Octacosadiyne	56.04	1.83	386.70	C28H50
25	Friedelin	56.73	2.19	426.700	C30H50O

RT: Retention Time, A%: Area Percentage, MW: Molecular Weight, MF: Molecular Formula.

list the compounds identified with a match percentage of 70% or higher in the NIST database. A total of 25 compounds were identified in the stem extract, including monoterpenes, sesquiterpenes, triterpenes, diterpenes, steroids, and small aromatic compounds. The leaf extract contained 24 compounds, with 18 of them common to both extracts, indicating shared metabolic pathways. However, there were significant differences in the number and concentration of pentacyclic triterpenes between the two extracts.

3.3. Molecular Networking

Figure 3 presents the molecular network generated with the volatile compounds found in the leaf and stem extracts of E. jefensis using the GNPS platform. In this network, the compounds are grouped according to their structural similarities, with each node representing a structurally related family of secondary metabolites produced by this plant.

Figure 4 categorizes the metabolites by identified compound families, highlighting the presence of triterpenes, sterols, and fatty acids. Numerous unidentified compounds are highlighted in black, either because they do not match the structures in the databases or because they represent new structures. The presence of these nodes underscores the importance of conducting in-depth chemical studies of the organic extracts of E. jefensis to isolate and identify these compounds, which could represent novel bioactive substances identified for the first time.

Figure 5 presents the results obtained using NPClassifier. It reveals a significant number of compounds that the bioinformatics tool could not classify, along with a notable percentage attributed to unidentified compounds, which may include novel substances. Among the identified compounds, the predominant group consists of fatty acids and related compounds, followed by simple phenolic compounds, and finally, various groups of terpene-type compounds.

Figure 6 displays the heatmap generated from the compounds identified with a high identification percentage or cosine value. These parameters indicate the percentage of fragments produced during the fragmentation of the compounds by electron impact, corresponding to the fragments of those compounds recorded in the consulted databases. Since the fragmentation pattern serves as a unique fingerprint for each compound, a higher value increases the certainty that the reported compounds are accurately identified. This heatmap illustrates the quantity of each compound in the analyzed organic extracts, enabling us to predict potential active compounds based on the published literature.

This figure is based on the peak abundances detected by the equipment when both samples were injected at the same concentration. It reveals that lupeol is present in high concentrations in the leaf extract but only in low concentrations in the stem extract. Regarding potentially active compounds, friedelin is found in high amounts in the leaf extract, while it is present in low amounts in the stem extract. Conversely, betulin is abundant in the stem extract but was not detected in the leaf extract.

4. Discussion

Our research group has established a line of inquiry focused on identifying new natural sources that may serve as potential alternatives for controlling A. gossypii [4]. Among these alternatives, we discovered that organic extracts from a recently described endemic plant, E. jefensis, exhibit significant mortality of A. gossypii. As illustrated in Table 1, both extracts demonstrate considerable toxicity. To contextualize the in vitro potential of E. jefensis leaf and stem extracts against A. gossypii, the LC₅₀ values obtained are comparable to those derived from other plants collected in Panama, such as Sphagneticola trilobata [4]. This neotropical plant is considered invasive in several countries, likely due to the development of significant defensive mechanisms that enable it to adapt to various environmental conditions. Notably, S. trilobata produces diterpene compounds similar to the resin components found in conifers, which serve as a defense system against insects and have been extensively studied [15,16]. Direct comparison of these results with those reported in other studies evaluating the activity of plant extracts against A. gossypii is challenging due to variations in evaluation methods, the use of aphids at different developmental stages, and differences in extract concentrations and preparation techniques. However, it is important to note that these results were obtained under laboratory conditions. Therefore, in vivo studies are necessary to confirm the activity in field conditions, as well as to assess potential formulations for the development of products suitable for agricultural use.

The insecticidal activity observed at various concentrations and evaluation times indicates that the types and amounts of active metabolites in each analyzed extract vary significantly. As a result, factors like solubility and bioavailability may play a role in determining the extent of biological activity observed [17]. To further understand the typs of compounds in E. jefensis extracts that contribute to mortality in A. gossypii, a comprehensive metabolomic analysis was performed. Modern metabolomic techniques are well-established for identifying a broad range of compounds, with high accuracy, as documented in existing literature [12]. Figure 1 and Figure 2 show the chromatograms from gas chromatography, highlighting the volatile metabolites present in each extract. These figures reveal that the dominant components differ between extracts (the relative concentrations of compounds in each extract can be observed quantitatively in column A% in Table 2 and Table 3), which could explain the variations in insecticidal activity. These results clearly demonstrate the effect of compartmentalization on the biosynthesis and storage of natural products in plants, as compartmentalization helps coordinate the activity of metabolic pathways with environmental signals and stress conditions, ensuring the efficient use of resources and a dynamic response to the environmental challenges faced by the plant [18].

Subsequently, the data from the GC-MS analysis was processed using the GNPS platform to conduct an advanced metabolomic analysis, which involved molecular networking, library searches, and identification of spectra similar to those included in the database [12]. This approach enables the visualization of active compounds in E. jefensis. Figure 3 shows the network generated from the leaf and stem extracts, revealing the complexity and diversity of secondary metabolites in each extract. It also highlights metabolites that are exclusive to certain extracts, suggesting substantial variation in the production and storage of these compounds. To further understand the types of compounds in the network, data analysis was performed using the NPClassifier bioinformatics tool, which classifies natural products into structural and biosynthetic categories using deep learning techniques [13]. Figure 4 presents these results, with compound families identified by E. jefensis shown as color-coded nodes. Upon examining these families, the triterpene family stands out, as it is well-known for its insecticidal properties [19]. Therefore, it is likely that triterpenes contribute to the insecticidal activity observed in the plant’s organic extracts.

Figure 5 provides a clearer view of the various compound groups identified in the analyzed extracts. This information is valuable as it sheds light on the metabolic pathways that are active in the plant under study [13]. It also highlights a significant proportion of unidentified compounds, which is not unexpected, as this plant has not been extensively researched. Some of these compounds may be novel, rare, or simply not yet documented in the databases consulted. Among the known compounds, terpenes make up the majority. These are produced in plants through two main biosynthetic pathways: the mevalonate (MVA) pathway, which occurs in the cytosol and generates sesquiterpenes, triterpenes, and sterols, and the methylerythritol phosphate (MEP) pathway in plastids, responsible for the production of monoterpenes, diterpenes, and tetraterpenes [20]. Another notable group, based on their abundance, includes simple C₆C₁ aromatic compounds such as phenyl derivatives and phenylpropanoids, which are mainly synthesized via the shikimic acid pathway [20].

An important observation from the detailed analysis is the fact that the major identified triterpenes in each extract differ in their chemical structures, which can be directly correlated with the potency of the insecticidal activity displayed in each extract. This clearly indicates that the main active compounds in each extract have distinct chemical structures from those found in the other extract. In the stem extract, the compound Friedelin is the main contributor to insecticidal activity. Friedelin has shown significant antifeedant effects against insects like Helicoverpa armigera and Spodoptera litura [21]. Further studies have confirmed that Friedelin exhibits concentration-dependent insecticidal properties, particularly against H. armigera, which is more susceptible than S. litura. Friedelin demonstrated both larvicidal and pupicidal effects on these pests, reinforcing earlier research identifying similar larvicidal effects in other compounds. It was found that Friedelin caused growth inhibition, leading to abnormal development in both insect species. A strong correlation was observed between Friedelin concentration and its antifeedant and larvicidal effects, aligning with findings from other studies on natural extracts [21]. Friedelin exhibited low LC₅₀ values against third-instar larvae of both pests, indicating its potent insecticidal activity, with no toxicity observed in freshwater fish, marking the first report of its effects on aquatic organisms [21]. In contrast, the insecticidal activity in the leaf extract is attributed to another triterpene, betulin. Previous research has demonstrated that this pentacyclic triterpene is highly effective against the aphid Myzus persicae [22]. Betulin has thus gained attention as a potential insecticide for aphid control, prompting further studies to understand its mechanism of action. Some studies suggest that betulin targets GABA receptors in insects, which regulate neuronal activity. RNA sequencing and other assays have shown that betulin reduces the expression of MpGABR in aphids, making them more sensitive to the compound. Microscale thermophoresis and voltage-clamp assays have confirmed that betulin binds to MpGABR and inhibits its activity, ultimately leading to aphid death [23].

In addition to the insecticidal compounds friedelin and betulin, which have been well-established in prior studies, the leaf extracts of E. jefensis also contain other pentacyclic triterpenes, such as lupeol and lupeol acetate. Due to their structural similarities to the aforementioned compounds [24], these triterpenes may also contribute to the observed insecticidal activity. Moreover, it is possible that other compounds of varying chemical structures are also playing a role in the insecticidal effects seen in the E. jefensis extracts. The study also notes the presence of non-volatile compounds, which were not the primary focus but could contribute to the insecticidal properties. However, the research concentrated on volatile compounds, as they are commonly recognized as key players in the plant’s defense mechanisms against insect pests [25,26]. As such, these metabolites were the primary focus of the study. Finally, Figure 6 shows that the major compounds identified are those likely responsible for the insecticidal activity. The data reveals that the most abundant compounds in the leaf extract are lupeol and friedelin, while betulin predominates in the stem extract. These findings are consistent with the observed effects in each extract, supporting the idea that these compounds are the active constituents. However, considering that the plant analyzed is wild and that the amount of a natural compound produced by a plant depends on several factors—such as genetic and environmental conditions, growth stage, biotic and abiotic stress, herbivores, pathogens, extreme conditions, and interactions with microorganisms, among other [27]—it is essential to cultivate the plant under controlled conditions to ensure the production of optimal amounts of lupeol, friedelin and betulin. Through cultivation, the levels of active compounds could be increased or kept consistent, potentially through genetic modifications aimed at optimizing compound production.

5. Conclusions

The results of our initial study indicate that organic extracts from the leaves and stems of E. jefensis cause significant mortality in A. gossypii under laboratory conditions, underscoring the plant’s potential as a natural insecticide. This conclusion is based on the use of a mixture of adult aphids and four-stage nymphs, which have been observed to cause the most damage to tomato crops in Panama. This insecticidal activity against aphids is attributed to the high levels of pentacyclic triterpenes, including friedelin in the stem extract and betulin in the leaf extract, both of which have been studied and shown to contribute to their insecticidal properties. In addition, advanced metabolomic techniques were used, allowing us to identify 29 compounds in the databases consulted, which constitute the first report of their presence in the E. jefensis extracts. Notably, this represents a novel contribution, as it is the first report on the metabolic composition of the E. jefensis plant, further enhancing the significance of the present work.