Plant Volatiles and Essential Oils Induce Sex-Specific Behavioral Responses and Concentration-Dependent Toxicity in the Invasive Pest Bagrada hilaris

Abstract

Bagrada hilaris (Burmeister) (Hemiptera: Pentatomidae) is an invasive pest that causes significant damage to Brassica crops worldwide. This study evaluated behavioral and toxicological responses of adults B. hilaris to plant volatiles and essential oils (EOs). Y-tube olfactometer assays revealed sex-specific responses to plant-emitted volatiles: females were repelled by Coriandrum sativum and Petroselinum crispum, while males responded to Pelargonium hortorum. Essential oils exhibited non-linear concentration-dependent effects, with C. sativum EO inducing repellency at 40–80 µg/µL and P. hortorum at 160–320 µg/µL. In contrast, repellency index was not influenced by sex, but strongly driven by concentration, with C. sativum and P. hortorum most effective, and P. crispum showing weaker yet consistent responses. Toxicity assays demonstrated greater male susceptibility, with lower LC₅₀ and LC₉₀ values for C. sativum and P. hortorum. Gas chromatography-mass spectrometry (GC-MS) analysis of EOs matrix identified linalool, β-citronellol, trans-geraniol, and myristicin as the predominant constituents. Importantly, repellency occurred at lower concentrations than mortality thresholds, indicating distinct behavioral and physiological mechanisms. These findings support integrating C. sativum and P. hortorum essential oils into sustainable pest management strategies for B. hilaris.

1. Introduction

Bagrada hilaris (Burmeister, 1835) (Hemiptera: Pentatomidae), commonly known as the painted bug, is an invasive pest species that has rapidly expanded its geographical range, causing significant damage to crops, particularly in the Brassicaceae family [1,2]. Native to Asia and Africa, B. hilaris has been recorded in North America, southern Europe, and the Middle East [3,4]. Its recent introduction to South America, including Chile [5,6], and its subsequent spread in Argentina [7], has raised significant concern regarding its invasive potential and capacity to cause severe yield losses in Brassica crops [6,8].

Bagrada hilaris primarily feeds on the cotyledons, young leaves, and the shoot apical meristem of host plants, causing chlorotic lesions that can progress to severe plant damage or plant death [8]. This has raised serious concerns regarding its agricultural impact, with crop losses reported to reach up to 30% of total production [6,8]. Current pest control relies on synthetic insecticides, such as pyrethroids, neonicotinoids, and carbamates, risking non-target organisms and the environment by disrupting natural pest control regulation and reducing biodiversity [9,10,11]. Thus, there is a growing need to develop alternative and sustainable pest management strategies that reduce dependence on chemical insecticides [6,12,13,14].

One promising alternative for sustainable pest management involves the use of plant-derived semiochemicals, particularly volatile organic compounds (VOCs) [15,16] emitted by aromatic and ornamental plants. These compounds play a crucial role in mediating insect-plant interactions by influencing behaviors such as host location, feeding, and oviposition [17,18]. Plant species from the families Apiaceae, Lamiaceae, Myrtaceae, and Geraniaceae are especially rich in terpenoids, alkaloids, and phenylpropanoids, many of which possess strong repellent properties and may interfere with host plant recognition by herbivorous insects [19,20,21,22]. Field and laboratory experiments have shown that the strategic use of repellent companion plants, such as Coriandrum sativum L. (coriander), Apium graveolens Linn. (celery), and Mentha haplocalyx Briq. (mint) may significantly reduce pest populations like Myzus persicae (Sulzer) (Hemiptera, Aphididae). This effect is achieved by the release of volatile compounds that mask or interfere with plant/host chemical signaling [23,24]. However, the effectiveness of these strategies may depend on environmental conditions, plant developmental stage, and cropping practices. In addition, logistical challenges such as tillage, irrigation, and spatial requirements often hinder the integration of companion plants in field settings [25,26].

Beyond whole plants, essential oils (EOs) (concentrated mixtures of plant-derived VOCs) have generated considerable interest as eco-friendly pest control agents [16,27,28]. These oils contain diverse bioactive compounds such as terpenes, alcohols, aldehydes, and esters, many of which have been demonstrated to exert both repellent and toxic effects on a wide range of insect species [29,30,31,32]. For example, essential oils extracted from plants of the Pelargonium genus [33,34] and C. sativum are known to affect pest physiology and behavior [35,36,37]. The Pelargonium graveolens L’Her. oil, rich in citronellol, β-pinene and citronellyl formate, has shown high repellency and toxicity against Tribolium castaneum (Herbst) (Coleoptera, Tenebrionidae) [33,34], while C. sativum oil, characterized by its high contents in linalool, has induced mortality in T. castaneum and Sitophilus oryzae (L.) (Coleoptera: Curculionidae) [38]. Similarly, oils from Mentha piperita L. and Thymus vulgaris L. have demonstrated insecticidal activity against pentatomids, such as Nezara viridula (L.) (Hemiptera, Pentatomidae), mediated by compounds like menthol and thymol [39].

However, the biological effects of EOs are highly context-dependent, varying with concentration, insect sex, developmental stage, and physiological condition, often exhibiting non-linear or hormetic responses [40]. For instance, Podisus nigrispinus Dallas (Hemiptera, Pentatomidae) nymphs were repelled by lemongrass oil, while adults showed tolerance to the same compounds [21]. Similarly, Halyomorpha halys (Hemiptera, Pentatomidae) exhibits seasonal shifts in response to vetiver oil Chrysopogon zizanioides (L.) (Poaceae), with repellent effects to summer active individuals and attraction to individuals entering overwintering [41]. These findings indicate that low and high concentrations can produce contrasting behavioral and toxicological responses, complicating prediction but offering opportunities for targeted pest management strategies [20]. Additionally, essential oils may act synergistically with live plants, enhancing repellency and disrupting host location [30,42]. Despite this potential, studies on their effects on B. hilaris remain scarce. Preliminary evidence suggests that compounds such as geraniol may act as repellents for this species [43], highlighting the need for further mechanistic and species-specific research.

In this context, the present study evaluates the repellent and toxic effects of volatile compounds emitted by aromatic and ornamental plants, as well as their essential oils, against adults of B. hilaris. We hypothesize that: (1) Behavioral responses of B. hilaris to plant volatiles and EOs are sex-specific, and (2) EOs elicit non-linear, concentration-specific behavioral and toxicological responses rather than simple monotonic concentration–response relationships. To test these hypotheses, we assessed insect behavioral responses using Y-tube olfactometer assays and toxicity through exposure bioassays under laboratory conditions. This research aims to identify potential plant species and essential oils for sustainable pest management and to establish a scientific basis for their integration into future integrated pest management (IPM) programs targeting B. hilaris.

2. Materials and Methods

2.1. Insects

The B. hilaris colony was established using individuals collected from two organic cabbage fields in Quillota, Valparaíso Region, Chile, and was periodically restocked with field-collected specimens. Field 1 was located in La Palma (32°53′14″ S, 71°12′9″ W) and Field 2 in San Pedro (32°54′42″ S, 71°15′31″ W). Insects were collected using fine paintbrushes and transferred to sealed glass jars for transport to the laboratory, where species identity and sex were confirmed [44]. Following identification, insects were reared in mesh-covered acrylic cages (25 cm × 25 cm × 40 cm) under controlled environmental conditions at 25 ± 1 °C, 60 ± 10% relative humidity (RH), and a 16:8 h light: dark photoperiod (L:D), similar to those described by [45]. Adults were fed fresh cabbage leaves every two days. Before bioassays, fourth- and fifth-instar nymphs were isolated in separate cages until adult emergence. Newly emerged males and females were then sexed during copulation and maintained individually in labeled Petri dishes (60 mm × 15 mm) [46] until use in experiments.

2.2. Plant Material

To evaluate the effect of volatiles emitted by aromatic and ornamental plants, the species Allium sativa (L.) (garlic), C. sativum (coriander), Mentha spicata L. (mint), Ocimum basilicum L. (basil), P. graveolens (rose geranium), Pelargonium hortorum LH Bailey (geranium), Pelargonium peltatum (L.) (ivy geranium), Petroselinum crispum (parsley), and Thymus vulgaris L. (thyme) were tested. Plants were propagated from seeds, except P. peltatum and P. graveolens, which were grown from cuttings. Pots of 0.29 L were used and filled with a 3:1 peat: perlite potting mix. Irrigation was carried out every 3 days based on substrate moisture. Plants were kept in a tunnel covered with anti-aphid mesh (mesh size: 0.27 mm × 0.79 mm). Experimental plants were used when they had developed 3 to 4 true leaves.

2.3. Olfactometer Bioassays: General Conditions

A Y-tube glass olfactometer was used, adapted from [43]. The Y-tube consisted of a 17-cm-long main arm and two side arms that were 13 cm in length and 2.4 cm in diameter, with an angle of 130° between the arms. Each arm was connected by polytetrafluoroethylene (PTFE) hoses to a glass chamber, approximately 270 mm in height, 14 mm in width, and with a volume of approximately 4 L. The olfactometer system was connected to a constant charcoal-purified air source, as indicated by [45], which flushed the volatile compounds into the glass chamber at a flow rate of 0.2 L/min and then directed them through Teflon hoses to the shorter arms of the Y-tube, where the odor streams converged toward the central arm. This setup exposed insects to opposing odor plumes, enabling directional behavioral responses. A control stimulus and a repellent stimulus (plant or essential oil, depending on the evaluation to be performed) were placed in each glass chamber (cloches). Each replicate consisted of a single adult of B. hilaris, carefully placed at the base of the central arm of the olfactometer using a paintbrush. The number of insects evaluated ranged from 25 to 35 individuals of each sex, as not all individuals gave a valid response. A valid response was considered when an individual of B. hilaris walked beyond the middle of one of the short arms of the Y-tube. If the insect did not move toward either arm during the five-min test, it was scored as a nonresponse, and the individual insect was discarded and was not included in the statistical analysis. Before testing, insects were starved for 5 h before each run, keeping individuals in 50 mm × 15 mm Petri dishes, based on the methods of [46] and preliminary assays (unpublished data). Pseudoreplication was avoided by testing and using the individual insects only once [47]; discarded insects were eliminated in 90% alcohol. The Y-tube was rotated between each replicate [48] and replaced every 5. After 10 replicates, the olfactometer system was cleaned with neutral soap and distilled water, wiped with acetone, and dried at 40 °C for 1 h [45]. All bioassays were conducted in controlled environmental conditions: 25 ± 1 °C, 60 ± 10% RH, and 1200 ± 100 lux intensity.

2.4. Olfactometer Bioassays: Response to Plant Volatiles

To evaluate the repellent activity of the selected plant species, each plant was tested against a control with only the potting mix [49], ensuring exposure to a single odor source. Pots were randomly assigned to the left or right of the olfactometer system, and their position was alternated after every 10 replicates to minimize directional bias. After system assembly, individual insects were placed with a paintbrush at the entrance of the Y-tube. A maximum response time of five minutes was allowed per insect. Bioassays were conducted from 11:00 to 17:00 h. The selected response time of the insect and the testing schedule were based on preliminary experiments (unpublished data) to reduce the number of nonresponsive insects in the trials.

2.5. Olfactometer Bioassays: Response to EOs

Commercial EOs of C. sativum, P. hortorum, and P. crispum (Pranarôm, Ghislenghien, Belgium), previously shown to exhibit repellent activity at the plant level, were used in this experiment. Essential oils were first weighed (0.1 g) and dissolved in 1 mL (1000 µL) of SupraSolv^® (98%) hexane (Merck) to obtain a concentrated stock solution (1 µg/µL). Serial dilutions were then performed by transferring 100 µL of the stock solution into 900 µL of hexane (1:10 dilution), generating intermediate solutions. From these, working concentrations of 40, 80, 160, and 320 µg/µL were prepared and used in the olfactometer bioassays. For each trial, a 6 cm² piece of Whatman No. 1 filter paper was used to apply the EO solutions, which were placed at the center of the filter paper. The EO was allowed to evaporate for 10 min on an uncovered glass Petri dish (5 cm in diameter). After evaporation, a potted cabbage plant with 3 to 4 true leaves was placed inside the olfactometer chamber. Each plant pot was randomly assigned either a control (hexane only) or the respective EO treatment (EO solution diluted in hexane), and treatment positions were systematically alternated between replicates, as described in the previous experiment. In each bioassay, valid responses were recorded from 25 males and 25 females of B. hilaris, following the methodology described by [43]. Insects that did not move toward the lateral arms within the trial period were counted but excluded from the analysis.

2.6. Repellency Assay of EO

To assess the repellent effect on B. hilaris adults, the protocol described by [33] was employed. A 9 cm diameter Whatman No. 1 filter paper was divided into two halves. One-half was treated with 0.5 mL of hexane (control), and the other half was treated with 0.5 mL of essential oil (EO) solution at the previously tested concentrations. After allowing the solvent to evaporate for 10 min, the two halves of the filter paper were reassembled to form a complete disc and placed at the bottom of a Petri dish. Five adult B. hilaris of each sex were then released in the center of the dish, and their distribution between the treated and control sides was recorded 24 h later. Each assay was independently replicated five times. Repellency was quantified using the proportion of repellency (PR) index, calculated with the formula:

$$PR={\displaystyle \frac{Nc-Nt}{Nt+Nc}}$$

(1)

where N_c is the number of insects in the control section and N_t is the number of insects in the treatment section. The IR value was used to classify the intensity of repellency or attractancy [41], with significant positive values indicating repellency and significant negative values indicating attractancy, depending on its value:

PR ≥ 0.8	:	very high repellency
0.6 ≤ PR < 0.8	:	high repellency
0.4 ≤ PR < 0.6	:	medium repellency
0.2 ≤ PR < 0.4	:	low repellency
0.2 ≤ PR < −0.2	:	neutral
−0.2 ≤ PR < −0.4	:	low attraction
−0.4 ≤ PR < −0.6	:	medium attraction
−0.6 ≤ PR < −0.8	:	high attraction
PR ≥ −0.8	:	very high attraction

2.7. Toxicity

To assess the contact toxicity of EOs against B. hilaris, 15 mL glass vials were coated with 1 mL of EO solution diluted in hexane or with hexane alone (control). The vials were gently rotated until the solvent completely evaporated, leaving a uniform residue on the inner surface, constituting a residual contact bioassay. Minor vapor exposure was minimized using mesh-covered vials, and no food or water was provided during the 24-h exposure period. Five male and female adults were exposed separately in each treated vial, using the same concentrations as in the previous experiment. Ten independent replicates were performed for each treatment. Insect mortality was assessed 24 h after exposure; individuals were considered dead if no movement of legs or antennae was observed.

2.8. Essential Oil Analysis by Gas Chromatography/Mass Spectrometry (GC/MS)

EOs were diluted with dichloromethane (CH₂Cl₂) (99.9% pure), and a 1 µL sample was analyzed using a GC–MS/MS (Thermo Scientifics, GC model Trace 1300 and MS model TSQ8000Evo) operating in electron ionization (EI) mode at 70 eV and equipped with a splitless injector (250 °C). The transfer line temperature was set at 250 °C. Helium was used as the carrier gas at a constant flow rate of 1.2 mL min⁻¹, and the capillary column was a Rtx-5ms column (60 m × 0.25 mm i.d., film thickness 0.25 µm). The temperature program was 40 °C (5 min) to 300 °C (5 min) at a rate of 5 °C min⁻¹. The chemical composition of the oil was identified by comparing its spectra with the MS Spectral Library 2014 of the National Institute of Standards and Technology (NIST) and confirmed by contrasting the retention rates with data published in other studies and NIST, in addition to co-injection of compounds. β -pinene, β-myrcene, γ-terpinene, 4-terpineol and α-terpineol standards of 99.8% purity were purchased from Sigma-Aldrich (St. Louis, MO, USA) [50]. The retention index (RI) was calculated considering the injection of n-alkane standards of C8-C20 and C21-C40, obtained from Sigma-Aldrich.

2.9. Statistical Analysis

All statistical analyses were conducted using R software (Version 4.3.2; R Core Team, 2023), and figures were generated using SigmaPlot software (Version 14.0; Systat Software Inc, San Jose, CA, USA). Olfactometer choice data were analyzed using χ² goodness-of-fit tests, as the response variable consists of categorical frequency data (binary choice: treatment vs. control). This test is appropriate for evaluating deviations from an expected 1:1 distribution under the null hypothesis of no preference [51]. The repellency index (PR) was analyzed using generalized linear models (GLMs), with concentration and sex included as explanatory variables. A Gaussian error distribution with an appropriate link function was applied to accommodate the continuous nature of PR values, allowing robust assessment of main effects. Model performance was evaluated through analysis of deviance using Wald χ² statistics, implemented via the lme4 package [52]. Significant main effects and interactions between sex and EO concentration were further analyzed using the Tukey test with single-step multiple comparison correction, performed using the multcomp package [53]. Finally, concentration-mortality relationships from bioassays were analyzed using probit regression to estimate lethal concentration values (LC₅₀) with associated 95% confidence intervals. All statistical tests employed α = 0.05 as the significance threshold.

3. Results

3.1. Response to Plant Volatiles

Adult females of B. hilaris (n = 25) were significantly repelled by volatiles from C. sativum (χ² = 9; df = 1; p < 0.01) and P. crispum (χ² = 6.76; df = 1; p = 0.01). In contrast, no repellency was detected for A. sativa (χ² = 1; df = 1; p = 0.32), M. spicata (χ² = 0.36; df = 1; p = 0.55), O. basilicum (χ² = 1; df = 1; p = 0.32), P. graveolens (χ² = 0.36; df = 1; p = 0.55), P. hortorum (χ² = 1; df = 1; p = 0.32), P. peltatum (χ² = 1; df = 1; p = 0.32), and T. vulgaris (χ² = 0.36; df = 1; p = 0.55). For adult males (N = 25), repellency was observed only with P. hortorum volatiles (χ² = 6.67; df = 1; p = 0.01). By contrast, A. sativa (χ² = 0.04; df = 1; p = 0.84), C. sativum (χ² = 0.04; df = 1; p = 0.84), M. spicata (χ² = 1; df = 1; p = 0.32), P. graveolens (χ² = 0.04; df = 1; p = 0.84), P. peltatum (χ² = 1.96; df = 1; p = 0.16), P. crispum (χ² = 1; df = 1; p = 0.32), and T. vulgaris (χ² = 0.36; df = 1; p = 0.55) did not show repellency. Interestingly, O. basilicum was attractive to males (χ² = 4.48; df = 1; p = 0.03) (Figure 1).

3.2. Response to EOs

The olfactometer bioassays revealed concentration-dependent behavior in the responses of B. hilaris adults to EOs. Males and females were significantly repelled by volatiles from C. sativum EO at low concentrations (40 and 80 µg/µL), whereas no repellency was detected at higher concentrations (160 and 320 µg/µL) for either sex. In contrast, P. hortorum elicited no response at 40 and 80 µg/µL, but at 160 and 320 µg/µL it exhibited repellency in both sexes. P. crispum showed consistent repellency toward both sexes at 80 and 160 µg/µL; however, at 40 µg/µL only males were repelled, and at 320 µg/µL no repellency was observed (

Table 1. Results of the olfactometer choice tests: behavioral responses (χ², df, and p-value) of female (♀) and male (♂) B. hilaris exposed to different concentrations of essential oils from C. sativum (coriander), P. hortorum (geranium), and P. crispum (parsley).

EOs	Concentration µg/µL	♀			♂
		χ2	df	p-Value	χ2	df	p-Value
C. sativum	40	9.00	1	<0.01	6.76	1	0.01
	80	6.76	1	0.01	4.84	1	0.03
	160	1.96	1	0.16	3.24	1	0.32
	320	1.00	1	0.32	1.00	1	0.32
P. hortorum	40	1.00	1	0.32	1.96	1	0.16
	80	1.96	1	0.16	3.24	1	0.07
	160	6.76	1	0.01	1.00	1	0.32
	320	1.00	1	0.32	6.76	1	0.01
P. crispum	40	1.60	1	0.21	8.10	1	<0.01
	80	6.76	1	0.01	11.56	1	0.001
	160	4.84	1	0.03	6.76	1	0.01
	320	3.24	1	0.07	1.96	1	0.16

, Figure 2).

3.3. Repellency Assay of EO

PR values showed no significant sex-based differences across the three essential oils. C. sativum (χ² = 2.85; df = 1; p = 0.09), P. hortorum (χ² = 1.35; df = 1; p = 0.25), and P. crispum (χ² = 2.01; df = 1; p = 0.16). However, concentration significantly influenced repellent efficacy for all oils (p < 0.001). C. sativum and P. hortorum showed the highest repellency at concentrations of 160 and 320 µg/µL, respectively (χ² = 48.64 and 25.90; df = 3), achieving medium to high repellency levels. P. crispum showed lower overall repellency (χ² = 19.10; df = 3) but maintained concentration-dependent activity (Figure 3).

3.4. Toxicity

A considerable percentage of mortality of B. hilaris was observed as oil concentration was increased. For C. sativum, the LC₅₀ values were 106.13 µg/µL in males and 177.47 µg/µL in females, with LC₉₀ values of 431.39 and 572.96 µg/µL, respectively. This indicates a higher susceptibility of males to this essential oil. In the case of P. hortorum, males again showed greater susceptibility, with an LC₅₀ of 102.41 µg/µL compared to 176.61 µg/µL in females. The LC₉₀ values followed a similar trend (429.07 µg/µL in males and 817.63 µg/µL in females). For P. crispum, both sexes required higher concentrations to reach 50% and 90% mortality, with LC₅₀ values of 220.31 µg/µL (males) and 241.84 µg/µL (females), and LC₉₀ values of 723.22 and 810.44 µg/µL, respectively. Despite higher LC values, the slopes of the regression lines were steeper (2.44–2.48), indicating a more consistent dose-response relationship (

Table 2. Probit analysis parameters and lethal concentrations (LC₅₀ and LC₉₀) of three essential oils against male and female adult insects. FL: Fiducial limits, SE: standard error, df: degrees of freedom, and p values are shown.

Essential Oil	Insect Sex	LC50 (µg/µL) (95% FL)	LC90 (µg/µL) (95% FL)	Slope ± SE	χ2	df	p-Value
C. sativum (coriander)	♀	177.46 (147.49 ± 221.89)	572.96 (406.55 ± 1008.05)	2.51 ± 0.34	20.46	38	0.991
	♂	106.12 (85.30 ± 131.56)	431.39 (301.461 ± 788.690)	2.104 ± 0.30	20.65	38	0.990
P. hortorum (geranium)	♀	176.61 (140.39 ± 238.03)	817.63 (503.63 ± 985.90)	1.925 ± 0.30	18.00	38	0.998
	♂	102.41 (81.74 ± 127.25)	429.07 (298.081 ± 796.64)	2.060 ± 0.30	20.53	38	0.991
P. crispum (parsley)	♀	241.84 (196.22 ± 324.13)	810.44 (535.20 ± 1686.07)	2.440 ± 0.37	26.44	38	0.921
	♂	220.31 (180.77 ± 286.87)	723.22 (490.87 ± 414.73)	2.483 ± 0.36	24.97	38	0.949

).

3.5. EOs Composition

Eighteen main compounds were identified in the essential oil of C. sativum, 30 in P. hortorum, and 17 in P. crispum. The identified molecules were predominantly monoterpenes, oxygenated monoterpenes, sesquiterpenes, and oxygenated sesquiterpenes (

Table 3. The categories and proportions of volatile compounds in essential oil from C. sativum (coriander), P. hortorum (geranium), and P. crispum (parsley).

Classification	Relative Area (%)
	C. sativum	P. hortorum	P. crispum
Monoterpene	11.00	0.00	54.50
Oxygenated monoterpene	32.20	67.10	3.00
Sesquiterpene	0.00	12.10	1.50
Oxygenated sesquiterpene	0.00	11.40	2.90
Other non-terpene molecules	53.90	1.60	33.50
Not identified	2.90	7.80	4.60
Total	100.00	100.00	100.00

).

In C. sativum, the major components were linalool (26.41%; RI: 1114.9), (E)-2-decenal (16.83%; RI: 1281.5), (E)-2-decen-1-ol (14.23%; RI: 1286.3), and decanal (10.6%; RI: 1222.3). The essential oils of P. hortorum were characterized by high proportions of β-citronellol (24.58%; RI: 1245.5), trans-geraniol (12.2%; RI: 1273.0), and citronellil formate (10.0%; RI: 1292.7). In P. crispum, the major constituents were myristicin (21.8%; RI: 1582.0), p-mentha-1,5,8-triene (15.0%; RI: 1132.5), β-phellandrene (13.98%; RI: 1048.3), dimethylstyrene (11.6%; RI: 1106.6), and α-pinene (10.5%; RI: 948.8) (

Table 4. Gas chromatography analysis (GC-MS/MS) of the commercial essential oils tested.

Compound Name	Classification	RI exp	Relative Area %			Identification
			C. sativum	P. hortorum	P. crispum
α-Pinene	Monoterpene	948.8	2.3		10.5	MS, RI a
β-Pinene	Monoterpene	992.1			7.7	MS, RI a, Co-I
β-Myrcene	Monoterpene	1002.3			5.3	MS, RI a, Co-I
α-Terpinolene	Monoterpene	1033.5	2.0			MS, RI a
p-Cymene	Monoterpene	1042.5	2.5		1.1	MS, RI a
β-Phellandrene	Monoterpene	1048.3			13.9	MS, RI a
γ-Terpinene	Monoterpene	1076.4	4.3		0.9	MS, RI a, Co-I
p-Mentha-1,5,8-triene	Monoterpene	1132.5			15.0	MS, RI a
Linalool	Oxygenated monoterpene	1114.9	26.4	5.3		MS, RI a
Rosoxide	Oxygenated monoterpene	1127.8		1.9		MS, RI a,
Alcanfor	Oxygenated monoterpene	1171.0	3.2			MS, RI a
Isomenthone	Oxygenated monoterpene	1187.3		6.7		MS, RI a
4-Terpineol	Oxygenated monoterpene	1199.7			0.2	MS, RI a, Co-I
α-Terpineol	Oxygenated monoterpene	1215.1		0.5		MS, RI a, Co-I
β-Citronellol	Oxygenated monoterpene	1245.5	0.7	24.5		MS, RI a
Cytral	Oxygenated monoterpene	1263.0		0.5		MS, RI a
trans-Geraniol	Oxygenated monoterpene	1273.0	1.6	12.2		MS, RI a
Citronellyl formate	Oxygenated monoterpene	1292.7		10.0		MS, RI a
Thymol	Oxygenated monoterpene	1314.4	0.3		2.7	MS, RI a
Neryl formate	Oxygenated monoterpene	1321.0		3.4		MS, RI a
Carvacrol	Oxygenated monoterpene	1325.2				MS, RI a
Geranyl acetate	Oxygenated monoterpene	1400.8		0.5		MS, RI a
Citronellyl propionate	Oxygenated monoterpene	1463.2		0.6		MS, RI a
Citronellyl butyrate	Oxygenated monoterpene	1550.4		1.0		MS, RI a
α-Cubebene	sesquiterpene	1379.4		0.3		MS, RI a
α-Copaene	sesquiterpene	1409.5		1.0		MS, RI a
β-Bourbonene	sesquiterpene	1421.3		2.3		MS, RI a
β-Elemene	sesquiterpene	1459.7			0.3	MS, RI a
Caryophyllene	sesquiterpene	1460.1		2.3	0.3	MS, RI a
β-Copaene	sesquiterpene	1467.9		0.4		MS, RI a
γ-Elemene	sesquiterpene	1479.6			0.5	MS, RI a
Aromandendrene	sesquiterpene	1501.7		0.4		MS, RI a
Cadinene	sesquiterpene	1510.4		0.5		MS, RI a
γ-Muurolene	sesquiterpene	1513.0		0.3		MS, RI a
α-Muurolene	sesquiterpene	1536.9		1.4		MS, RI a
γ-Cadinene	sesquiterpene	1554.2		0.5		MS, RI a
β-Famesene	sesquiterpene	1558.8			0.4	MS, RI a
Cadinene	sesquiterpene	1556.0		2.7		MS, RI a
α-Agarofurane	Oxygenated sesquiterpene	1594.3		0.6		MS, RI a
Spathulenol	Oxygenated sesquiterpene	1624.5		0.4		MS, RI a
Caryophyllene oxide	Oxygenated sesquiterpene	1633.0		0.3		MS, RI a
Epicubenol	Oxygenated sesquiterpene	1656.5		0.4		MS, RI a
epi-γ-Eudesmol	Oxygenated sesquiterpene	1667.4		7.1		MS, RI a
α-Eudesmol	Oxygenated sesquiterpene	1700.0		0.9		MS, RI a
Carotol	Oxygenated sesquiterpene	1720.6			1.7	MS, RI a
Geranyl tiglate	Oxygenated sesquiterpene	1729.7		1.6		MS, RI a
Dimethylstyrene	Other non-terpene molecules	1106.6			11.6	MS, RI a
Decanal	Other non-terpene molecules	1222.3	10.6			MS, RI a
(E)-2-Decenal	Other non-terpene molecules	1281.5	16.8			MS, RI a
(E)-2-Decen-1-ol	Other non-terpene molecules	1286.3	14.2			MS, RI a
Undecanal	Other non-terpene molecules	1325.6	0.7			MS, RI a
2-Undecenal	Other non-terpene molecules	1384.9	1.6			MS, RI a
Tetradecanal	Other non-terpene molecules	1637.4	0.5			MS, RI a
(E)-Tetradec-2-enal	Other non-terpene molecules	1698.3	2.3			MS, RI a
Myristicin	Other non-terpene molecules	1582.0			21.8	MS, RI a
Geranyl butyrate	Other non-terpene molecules	1582.9		1.6		MS, RI a
Dodecanal	Other non-terpene molecules	1429.8	1.7			MS, RI a
Elemicin	Other non-terpene molecules	1605.9			1.2	MS, RI a
2-Dodecenal	Other non-terpene molecules	1489.3	5.4			MS, RI a
Total identified			97.1	92.2	95.4
Total no identified			2.9	7.8	4.6
			100.0	100.0	100.0

^a: Bibliographic retention index for nonpolar column; MS: Mass spectrometry; RI: retention index, Co-I: Coinjection of standard.

).

4. Discussion

This study showed that B. hilaris exhibits sex-specific and concentration-dependent behavioral and toxicological responses to plant volatiles and essential oils, supporting their potential use in pest management. Behavioral assays revealed sexual dimorphism in olfactory sensitivity. Females were repelled by plant volatiles from C. sativum and P. crispum, whereas males responded only to P. hortorum. Essential oil assays confirmed strong concentration dependence. C. sativum was effective at low concentrations but lost efficacy at higher doses, while P. hortorum became repellent only at higher concentrations. P. crispum showed moderate, more consistent repellency at intermediate concentrations. PR index analyses identified concentration as the main driver of behavioral response. Toxicity assays further demonstrated greater male susceptibility, with lower LC₅₀ and LC₉₀ values, especially for C. sativum and P. hortorum.

The repellent effects of the evaluated plant volatiles on B. hilaris, while statistically significant in some cases, were generally weaker than expected, reflecting the complexity of plant–insect chemical interactions. One contributing factor may be the phenological stage of the plants used, as young plants with only 2–4 true leaves likely emit volatile blends that differ qualitatively and quantitatively from those produced at later developmental stages [54]. Because plant volatile profiles change substantially during development, the chemical cues available to B. hilaris may have been less representative of those encountered under natural field conditions, resulting in reduced behavioral responses [55]. In addition, the polyphagous nature of B. hilaris may partly explain the limited repellency observed. The generalist Pentatomidae typically possess flexible olfactory systems, with responses shaped by physiological state and environmental context [56]. Their reliance on ubiquitous cues such as green leaf volatiles, which are common across many plant taxa, may dilute or mask the repellent effects of more specific compounds [57], thereby reducing overall behavioral avoidance. Sex-specific responses further contribute to this variability and are consistent with patterns reported in other heteropteran insects (see below). This difference may stem from sex-specific host use, with males primarily feeding and females engaging in both feeding and oviposition [58].

Olfactometer bioassays revealed pronounced, non-linear concentration-dependent behavioral responses of adult B. hilaris to essential oils (EOs) from three plant species, demonstrating that repellency is governed by species-specific sensory thresholds rather than simple concentration–response relationships. Sub-threshold concentrations failed to elicit avoidance, whereas optimal doses effectively disrupted host-orientation behavior [40], underscoring the critical role of concentration optimization in EO-mediated behavioral interference. C. sativum EO exhibited the most pronounced non-linear response, inducing significant repellency at low concentrations (40–80 µg/µL) but losing efficacy at higher doses (160–320 µg/µL). This inverse dose–response pattern is consistent with biphasic behavioral responses reported in other insect–volatile systems and is commonly attributed to sensory adaptation, odorant receptor desensitization, or masking effects at high stimulus intensities [59]. Insect odorant receptors are known to exhibit concentration-dependent tuning, responding selectively at low concentrations but producing broader or conflicting responses as concentrations increase [60]. The predominance of linalool in C. sativum EO [61,62] likely contributes to this pattern, as linalool elicits biphasic behavioral effects across multiple pest taxa [29,40,61], including heteropterans such as H. halys [63] and Philaenus spumarius (L.) [61]. Additional volatile constituents, including (E)-2-decenal, decanal, and (E)-2-decen-1-ol, may further modulate responses by saturating the headspace or masking key olfactory cues [64,65], highlighting the strong context dependence of repellency. P. hortorum EO showed a threshold-dependent response, with repellency emerging only at higher concentrations (160–320 µg/µL). Its chemical profile, dominated by oxygenated monoterpenes such as β-citronellol, trans-geraniol, and citronellyl formate, is characteristic of Pelargonium species [32,55] and has been associated with repellency against diverse arthropod pests [34,66], including B. hilaris [43]. Such threshold effects are typical of oxygenated monoterpenes [67] and parallel findings in other systems, including citronella oil against Nilaparvata lugens Stål. (Hemiptera: Delphacidae) [68]. Although B. hilaris often exhibits limited responses to EOs, the demonstrated repellency of geraniol alone [43] suggests that synergistic interactions among EO constituents may underlie the observed effects [41,42]. By contrast, P. crispum EO displayed the most complex non-monotonic response, with maximal repellency at intermediate concentrations (80–160 µg/µL) and reduced effects at lower and higher concentrations. Such hormetic or biphasic patterns are increasingly recognized in essential oil research [40,69]. Although constituents such as myristicin exhibit repellent activity in other arthropod systems [70,71], their behavioral roles in Hemiptera remain poorly understood. Similar context-dependent outcomes have been reported for other EOs, including vetiver oil, where attraction or repellency varies with physiological state and environmental conditions [39,41].

The repellency assays revealed that sex did not significantly influence the behavioral response of B. hilaris adults to any of the tested essential oils, indicating the absence of sexual dimorphism in short-range avoidance behavior. Similar sex-independent responses have been documented in H. halys [41] and N. viridula [39], strengthening the reliability of these patterns and suggesting that formulations are dependent on the application range. In contrast, concentration significantly affected repellent efficacy for all oils, demonstrating clear concentration-dependent responses [41,72]. C. sativum and P. hortorum exhibited the strongest activity at 160 and 320 µg/µL, respectively, reaching medium to high repellency levels. Increasing PR values with concentration indicate that higher doses enhance avoidance behavior through greater volatile emission and olfactory stimulation [21,39]. Although P. crispum showed lower PR values, it maintained significant concentration dependence, suggesting moderate repellency. The documented activity of C. sativum against hemipteran pests further supports its potential [35]. These findings highlight the importance of dosage optimization and identify C. sativum and P. hortorum as promising candidates for behavioral management of B. hilaris. Differences between long- and short-range assays likely reflect distinct sensory contexts, including variation in stimulus concentration, exposure mode, and integration of olfactory and contact cues [40,59].

Toxicity assays revealed clear sexual dimorphism in B. hilaris with males consistently more susceptible than females to essential oils. Similar sex-related differences in insecticide sensitivity have been widely documented across insect taxa [73] and are often associated with physiological and biochemical variation. Differential activity of detoxification enzymes, including cytochrome P450 monooxygenases [74], glutathione S-transferases [75], and esterases [76], likely contributes to this pattern. Morphological traits such as cuticular properties may contribute to differential exposure and insecticide susceptibility by influencing penetration rates [77,78]. Although not evaluated here, these traits merit further investigation.

Differences in toxicity among essential oils were strongly associated with their chemical composition, with C. sativum and P. hortorum exhibiting higher toxicity than P. crispum. In B. hilaris, these results emphasize the importance of oil-specific chemical profiles in determining insecticidal efficacy. C. sativum, characterized by high concentrations of linalool and aliphatic aldehydes, demonstrated strong bioactivity, consistent with concentration- and time-dependent mortality reported in Tribolium confusum Duval (Coleoptera: Tenebrionidae) exposed to coriander oil [37], supporting its potential effectiveness against insect pests. Similarly, P. hortorum, containing β-citronellol, trans-geraniol, and citronellyl formate, showed substantial toxicity, in agreement with its documented effects against T. castaneum [33] and Bemisia tabaci Gennadius (Hemiptera: Aleyrodidae) [66]. These findings suggest that oils rich in oxygenated monoterpenes are particularly effective against B. hilaris, likely due to enhanced bioavailability and interaction with target sites [67]. In contrast, P. crispum, dominated by myristicin, β-phellandrene, and α-pinene, exhibited significantly lower toxicity, consistent with its reduced fumigant activity against Callosobruchus maculatus (F.) (Coleoptera, Chrysomelidae) compared to phosphine [79]. This lower efficacy may reflect the predominance of monoterpene hydrocarbons, which generally show lower activity than oxygenated compounds [39,67]. Overall, these results indicate that essential oil efficacy in B. hilaris depends primarily on compositional traits rather than generalized modes of action.

The combined behavioral and toxicological responses observed in B. hilaris highlight the importance of concentration-dependent mechanisms operating at different biological levels. While sublethal concentrations primarily modulate host-orientation behavior through olfactory interference, higher concentrations induce physiological disruption and mortality. Importantly, the observed non-linear behavioral responses and sex-specific toxicity indicate that EO efficacy is not uniform, but varies with concentration and biological context. Integrating these dual and context-dependent effects provides a framework for optimizing EO-based strategies, where concentration-specific applications can be tailored to either repel or suppress pest populations.

5. Conclusions

This study highlights a divergence between volatile emission ranges and concentrations required for repellency versus toxicity. Repellency occurs at low vapor-phase concentrations via olfactory-mediated behavioral avoidance without systemic damage, whereas toxicity requires higher internal exposure to disrupt physiological processes and overcome detoxification defenses. These mechanistic insights support IPM strategies, where low concentrations deter pests through behavioral disruption, while higher concentrations produce direct toxic effects. Integrating repellent plant volatiles and essential oils may help reduce dependence on synthetic insecticides in brassica systems. Future research should validate field efficacy, optimize formulations, assess compatibility with biological control agents, and evaluate potential impacts on beneficial arthropods and other non-target insect species to ensure sustainable and ecologically sound management of B. hilaris.