Defense Responses of Cucumber and Cowpea to Frankliniella occidentalis Infestation Influence the Growth, Development, and Host Preferences of the Pest

TL;DR

Cucumbers and cowpeas respond to thrips infestation by strengthening their cell walls, but thrips still prefer cowpeas even though they reproduce better on cucumbers.

Contribution

The study reveals how plant defense mechanisms and thrips adaptability interact through molecular pathways and host selection.

Findings

Cucumbers and cowpeas both attract F. occidentalis, but thrips reproduce better on cucumbers.

Thrips infestation activates plant defense pathways like jasmonic acid and lignin biosynthesis.

F. occidentalis adjusts its host preference and reproduction to overcome plant defenses.

Abstract

Frankliniella occidentalis (Pergande) is a globally invasive pest that inflicts significant damage on economically important vegetable crops such as cucumbers (Cucumis sativus L.) and cowpeas (Vigna unguiculata L. Walp). To elucidate the interactions between host plants and F. occidentalis and to support the development of sustainable management strategies, this study evaluated the host selectivity and life history parameters of F. occidentalis living on these plant species to assess its adaptability. Transcriptome–metabolome profiles and associated metabolites were analyzed in healthy plants and in those infested by F. occidentalis for 48 h to characterize the defense responses of both host species. The results showed that both plant species are attractive to F. occidentalis, with a stronger preference observed for cowpeas. However, the reproductive output of F. occidentalis was significantly higher on cucumbers (16.99 ± 0.43 eggs/female) than on cowpeas (12.00 ± 0.38 eggs/female) plants, indicating a mismatch between host preference and performance. Feeding by F. occidentalis strongly induced the brassinolide and jasmonic acid signaling pathways, activated the phenylpropanoid metabolic pathway, increased the accumulation of the lignin precursor sinapyl alcohol, and promoted lignin biosynthesis, thereby enhancing cell wall rigidity as a physical defense barrier. These findings demonstrate that cucumbers and cowpeas coordinately regulate lignin synthesis through hormone–metabolism crosstalk as a defensive strategy against thrips attack. In response, F. occidentalis adjusts its host selection and reproductive investment to overcome plant defenses, reflecting an adaptive counter-strategy in host–herbivore interactions. This study provides new insights into the molecular mechanisms underlying plant–thrips interactions and supports the development of environmentally friendly pest control approaches.

Figures

Full text

1. Introduction

Frankliniella occidentalis (Pergande) is an invasive alien species and a major global agricultural pest. First described in California, USA, in 1895, it began spreading worldwide in the late 1970s [1] and has since been reported in at least 69 countries [2]. In China, it was first detected in Kunming in 2000 and has rapidly spread across multiple provinces. It is now one of the most destructive pests in both greenhouse and open-field crops, causing annual economic losses in the billions of yuan [3]. F. occidentalis has a broad host range, infesting over 500 horticultural and ornamental plant species, with severe damage to key crops such as cucumbers, snap beans, eggplants, zucchinis, peppers, cabbage, and lettuce [4]. It has a short generation cycle, high reproductive capacity, and concealed feeding behavior, leading to extensive crop injury [5]. F. occidentalis causes damage by feeding on the sap of host plant tissues—including flowers, leaves, buds, and fruits—using its rasping–sucking mouthparts. This feeding activity results in the formation of white stippling on the leaves, followed by leaf wrinkling, curling, chlorosis, and desiccation, as well as fruit deformation. These symptoms compromise fruit quality and ultimately lead to reduced crop yields [6]. Moreover, as a vector of multiple plant viruses, including tospoviruses such as tomato spotted wilt virus (TSWV), it significantly intensifies the overall damage to host plants. These symptoms reduce fruit quality and can result in major yield losses or total crop failure [7]. Cucumbers and cowpeas are important vegetable crops widely cultivated worldwide. In recent years, F. occidentalis has emerged as a major pest of both crops, significantly compromising their yield and quality. However, research on the molecular response mechanisms of cucumbers and cowpeas to F. occidentalis infestation remains limited, particularly in comparative studies across species.

Throughout their evolutionary history, plants have developed complex defense systems to counteract herbivorous insect attacks. These defense mechanisms are conventionally classified into constitutive and inducible types [8]. Constitutive defenses are continuously present and primarily involve physical barriers (e.g., trichomes and cuticular waxes) and chemical compounds (e.g., alkaloids and terpenoids). In contrast, inducible defenses are activated upon herbivory or oviposition, where specific signaling pathways trigger the synthesis of defensive proteins and secondary metabolites that deter further pest feeding [9]. Evidence indicates that the jasmonic acid (JA) and salicylic acid (SA) signaling pathways play central roles in mediating plant resistance to insect pests. JA predominantly regulates responses to chewing and piercing–sucking insects, while SA is more involved in the defense against pathogens and certain phloem-feeding insects [10]. In tomato, F. occidentalis infestation induces a pronounced upregulation of key JA pathway genes, accompanied by rapid accumulation of defensive terpenoids and phenolic amides [11]. The phenylpropanoid metabolic pathway serves as a key component of plant defense systems and plays a critical role in mediating resistance to herbivorous pest invasion. This pathway is rapidly activated by pest feeding, regulating defense-related genes and promoting the synthesis of secondary metabolites [12], including lignin, flavonoids, phenolic compounds, and coumarins, which contribute to both physical and chemical plant defenses [13]. Furthermore, the pathway branches into major routes such as those for lignin and flavonoid biosynthesis [14]. Lignin deposition enhances cell wall rigidity through lignification, forming a physical barrier; for example, rice plants resist brown planthoppers by thickening their cell walls, which hinders insect feeding [15]. Flavonoids and related compounds such as phenols and isoflavones chemically deter pests by inhibiting their feeding, development, and oviposition [16]. When vitexin, a flavonoid extracted from spinach (Spinacia oleracea L.) leaves, was administered to larvae of the diamondback moth (Spodoptera litura Fabricius), larval growth was significantly suppressed [17]. The phenylpropanoid pathway plays a key role in direct plant defense and forms an intricate network with hormone signaling pathways such as JA and SA; this network functions as a regulatory hub, enabling plants to mount specific responses against different pest feeding behaviors and optimize defense resource allocation [18].

In response to diverse plant defense strategies, insects have evolved a range of defense mechanisms to minimize harm. These adaptive capabilities, which enable insects to overcome plant resistance, are primarily manifested in three domains: behavioral responses, growth and development, and physiological and biochemical adaptations [19]. Behavioral adaptation to plant defenses is largely driven by the insect’s ability to perceive chemical and physical cues, including secondary metabolites and herbivore-induced plant volatiles. This recognition allows insects to make strategic decisions, such as selecting suitable host plants for feeding and oviposition [20]. For instance, F. occidentalis exhibits a stronger preference for untreated Arabidopsis and pepper plants compared to those treated with JA. When the key defense-related genes LOX in pepper, DEF-1 in tomato, and COI1-1 in Arabidopsis are silenced, the thrips show increased feeding preference for these plants [21]. Changes in insect growth and development contribute to their counter-defense adaptations, including alterations in life cycle duration to enhance adaptation to host plants [22]. When cowpeas are treated with exogenous elicitors, F. occidentalis can adapt by extending its developmental duration [23]. In the coevolutionary arms race between phytophagous insects and host plants, insects have evolved specific detoxification mechanisms to neutralize toxic compounds produced through plant-induced defenses [24]. For example, insects can upregulate detoxification gene (e.g., CYP450) expression or enhance digestive enzyme (e.g., trypsin, amylase, and proelastase) activity to facilitate adaptation to plant defense compounds [25,26]. Elucidating these reaction mechanisms at three levels systematically reveals the multi-level anti-defense strategies evolved during insect–plant coevolution. This provides a theoretical foundation for understanding insect-plant interactions and herbivorous pest ecological adaptation.

In this study, an integrated transcriptomic and metabolomic approach was employed to systematically investigate the dynamic changes in differentially expressed genes in cucumber and cowpea leaves following herbivory, while concurrently profiling defense-related metabolites through metabolomic analysis. Furthermore, by integrating measurements of key plant defense enzyme activities with the quantification of secondary metabolites, we identified critical compounds strongly associated with insect resistance phenotypes. The findings not only facilitate the screening of potential resistance-associated genes and metabolic markers but also provide valuable molecular targets and theoretical support for breeding resistant crop varieties and developing sustainable, green control strategies.

2. Results

2.1. Effects of Host Plants on the Fitness of Frankliniella occidentalis

Selective experiments showed that healthy cucumber and cowpea plants were significantly more attractive to F. occidentalis than was the blank control (p < 0.001). Thrips preferred healthy cowpea plants (VUC) over healthy cucumber plants (CSC) (p < 0.001), but showed no significant difference in selectivity between infested cucumber and infested cowpea plants (p > 0.05) (Figure 1).

F.occidentalis fed on cucumber leaves exhibited significantly higher fecundity than those fed on cowpea leaves (p < 0.05). No significant differences were observed in pre-adult duration, pre-oviposition period, or adult longevity between the cucumber and cowpea groups (Figure 2).

2.2. Transcriptome Analysis of Cucumbers and Cowpeas Following Infestation by Frankliniella occidentalis

A total of 2281 differentially expressed genes (DEGs) were identified in the CSC (clean surface control) vs. CST (cucumber systemic treatment) comparison, with 1163 upregulated and 1118 downregulated (Figure 3a). In the VUC (clean surface control) vs. VUT (cowpea systemic treatment) comparison, 3392 DEGs were detected, including 1815 upregulated and 1577 downregulated (Figure 3b). KEGG pathway enrichment analysis showed that DEGs in the CST group were significantly enriched in key pathways: DNA replication (ko03030), phenylalanine metabolism (ko00360), porphyrin and chlorophyll metabolism (ko00860), phenylpropanoid biosynthesis (ko00940), and mismatch repair (ko03430) (Figure 3c). In the VUT group, DEGs were mainly enriched in regards to plant circadian rhythm (ko04712); carbon metabolism (ko01200); carotenoid biosynthesis (ko00906); valine, leucine, and isoleucine degradation (ko00280); and phenylpropanoid biosynthesis (ko00940) (Figure 3d).

2.3. Metabolomic Analysis of Cucumbers and Cowpeas Following Infestation by Frankliniella occidentalis

CST showed 1327 upregulated and 855 downregulated differentially accumulated metabolites (DAMs) compared to the results for CSC (Figure 4a). VUT displayed 1347 upregulated and 837 downregulated DAMs compared to the results for VUC (Figure 4b). KEGG pathway enrichment analysis revealed that DAMs in the CST group were significantly enriched in key pathways: isoflavone biosynthesis (ko00943), linoleic acid metabolism (ko00591), arginine biosynthesis (ko00220), porphyrin metabolism (ko00860), and monoterpene biosynthesis (ko00902) (Figure 4c). In the VUC vs. VUT comparison, DAMs were enriched in glyoxylate and dicarboxylate metabolism (ko00630), the tricarboxylic acid (TCA) cycle (ko00020), oxidative phosphorylation (ko00190), arginine biosynthesis (ko00220), and phenylpropanoid biosynthesis (ko00940) (Figure 4d).

2.4. Integrated Analysis of Metabolomics and Transcriptomics

Pearson correlation analysis was used to examine KEGG pathway enrichment of DEGs and DAMs in the CSC vs. CST and VUC vs. VUT comparisons. In CSC vs. CST, enriched pathways included phenylalanine metabolism, porphyrin and chlorophyll metabolism, phenylpropanoid biosynthesis, carotenoid biosynthesis, histidine metabolism, benzoxazinoid biosynthesis, isoflavonoid biosynthesis, zeatin biosynthesis, pyrimidine metabolism, and plant hormone signal transduction (Table 1). In VUC vs. VUT, enriched pathways included carbon metabolism; carotenoid biosynthesis; valine, leucine, and isoleucine degradation; glycerolipid metabolism; phenylpropanoid biosynthesis; stilbenoids, diarylheptanoids, and curcuminoids biosynthesis; carbon fixation in photosynthetic organisms; phosphatidylinositol metabolism; ABC transporters; and porphyrin and plant hormone signal transduction (Table 2). Phenylpropanoid biosynthesis and plant hormone signal transduction were significantly enriched in both thrips-infested cucumber and cowpea plants.

2.5. Changes in Phenylpropanoid Metabolism and Hormone Signaling Pathways

Compared to CSC, CST showed a higher expression of genes in the phenylpropanoid biosynthesis pathway, including 4CL, CCR, F5H, CAD, and COMT. In the plant hormone signal transduction pathway, upregulated genes included BAK1, BRI1, TCH4, and MYC2. Metabolomic analysis revealed increased levels of phenylalanine, coniferaldehyde, sinapyl alcohol, jasmonic acid, and brassinosteroids in thrips-infested cucumber plants, with phenylalanine increasing by 1.1-fold, sinapyl alcohol by 3.4-fold, jasmonic acid by 1.2-fold, and brassinosteroids by 1.2-fold. Following F. occidentalis infestation, the expression of key genes in the JA and BR signaling pathways was significantly upregulated in both cucumbers and cowpeas within 48 h. In cucumbers, the JA pathway gene JAR1increased 1.13-fold, while the BR pathway gene BRI1 increased 3-fold (Figure 5).

Compared to VUC, VUT showed a higher expression of genes in the phenylpropanoid biosynthesis pathway, including C4H, 4CL, and CAD. Genes upregulated in plant hormone signal transduction included BAK1, BRI1, TCH4, and MYC2. Metabolomic analysis revealed increased levels of phenylalanine, coniferaldehyde, sinapyl alcohol, jasmonic acid, and brassinosteroids in thrips-infested cowpea plants, with phenylalanine increasing by 1.1-fold, coniferaldehyde by 3.3-fold, sinapyl alcohol by 1.2-fold, jasmonic acid by 1.2-fold, and brassinosteroids by 1.1-fold. Following F. occidentalis infestation, the expression of key genes in the JA and BR signaling pathways was significantly upregulated in both cucumbers and cowpeas within 48 h. In cowpeas, the upregulation was more pronounced: JAR1expression rose 3.07-fold, and BRI1expression rose 3.61-fold (Figure 6).

2.6. qRT-PCR Analysis

Nine DEGs were selected from cucumbers and six from cowpeas for validation of relative expression levels. The expression patterns of these DEGs derived from RNA-seq showed high consistency with those obtained by qRT-PCR (quantitative real-time PCR), demonstrating the reliability of the RNA-seq data (Figures S1 and S2).

2.7. Effects of Frankliniella occidentalis Infestation on Defensive Compounds in Plants

In cucumbers, lignin content increased significantly after F. occidentalis infestation (Figure 7b). In cowpeas, both PAL activity and lignin content increased significantly (Figure 7d,e). Flavonoid levels remained unchanged in both cucumbers and cowpeas following infestation (Figure 7c,f).

3. Discussion

Over the course of long-term coevolution, herbivorous insects and their host plants have developed intricate and diverse adaptive interactions [27]. Insect adaptation to plants is influenced by factors such as plant species, morphology, nutrition, and secondary metabolites, which significantly affect insect growth, development, and reproduction [28]. Life tables provide a systematic method for evaluating insect survival, population dynamics, and host adaptability across different host plants [29]. The host selection assay demonstrated that F. occidentalis exhibited a clear preference for both healthy host plants, with higher selectivity toward cowpeas than cucumbers. The host selectivity assays demonstrated that both healthy cucumber and cowpea plants were attractive to F. occidentalis, with a stronger preference for cowpeas. However, this initial preference difference disappeared when the plants were damaged, as thrips showed no significant selection bias between the two damaged hosts. This indicates that herbivore-induced defensive compounds elicited in both plant species reduce the initial preference disparity. Following thrips infestation, phenylalanine ammonia-lyase activity and lignin content increased significantly in cowpeas, whereas only lignin content increased in cucumbers, indicating a stronger defensive response in cowpeas. Although cowpeas initially attracted more thrips, the chemical composition after feeding was less favorable for thrips reproduction. Consequently, thrips reproductive performance was higher on cucumbers over one generation. These results further suggest a potential divergence between thrips adaptability and initial host preference across different vegetable crops. Research indicates that thrips host preference is strongly linked to vegetable species [30]. Furthermore, life history parameter analysis revealed that thrips populations reared on cucumbers laid significantly more eggs than those reared on cowpeas after prolonged adaptation, indicating stronger physiological adaptation to cucumbers. These findings suggest that while F. occidentalis displays a distinct host preference, it also possesses a high degree of phenotypic plasticity, enabling it to adjust its performance in response to host defense responses. Consistent with previous studies, variation in fecundity across hosts serves as a reliable indicator of thrips adaptability to specific plant species [31].

This study reveals that cowpeas contain significantly higher levels of lignin and total flavonoids compared to cucumbers. Elevated lignin content contributes to the reinforcement of plant sclerenchyma tissues, thereby enhancing structural defense against herbivore attack [32]. Elevated total flavonoid content enhances the plant’s defensive response to biotic stresses such as pathogens and herbivores, and can directly deter insect feeding, development, and oviposition [33]. Furthermore, upon thrips infestation, cowpeas exhibited significant increases in both phenylalanine ammonia-lyase (PAL) and lignin levels, whereas in cucumbers, only lignin showed a significant rise. This suggests that cowpeas mount a stronger constitutive and induced defense response compared to that of cucumbers. Although cowpeas are more attractive to F. occidentalis, the higher accumulation of lignin and flavonoids following herbivory creates a less favorable physiological environment for thrips reproduction. Consequently, reproductive performance of F. occidentalis is higher on cucumbers. These findings further indicate a potential decoupling between host preference and adaptation in F. occidentalis across different vegetable crops [34].

Plant hormones serve as key signaling molecules that directly or indirectly regulate plant defense responses, playing a central role in mediating the defense signaling network during herbivorous insect attack [35]. The plant–insect interaction is coordinately regulated by multiple hormonal pathways, whose synergistic interactions can amplify defensive outputs and enhance resistance to herbivores [36]. In this study, feeding by the F. occidentalis significantly upregulated genes encoding MYC2, BRI1, and BAK1 in both cucumbers and cowpeas. MYC2 functions as a key transcription factor in the endogenous jasmonic acid (JA) signaling pathway and directly activates the expression of defense-related genes [37]. BAK1 functions as a co-receptor of BRI1, and together they regulate the brassinosteroid (BR) signaling pathway, promoting cell wall reinforcement and thereby playing a critical role in activating structural defenses in plants [38]. The upregulation of these genes indicates that F. occidentalis infestation may activate the JA and BR signaling pathways in both cucumbers and cowpeas. The JA and BR signaling pathways are key components of plant hormone-mediated defense networks and exhibit high efficiency and broad conservation in plant responses to herbivore stress. For example, oviposition by the Bactrocera minax (Enderlein) activated both JA and SA signaling pathways in Citrus reticulata [39]. In tea plants, infestation by the tea green leafhopper Empoasca onukii (Matsuda) suppressed the expression of BR biosynthesis-related genes, leading to reduced BR levels [40].

Plants frequently activate the phenylpropanoid metabolic pathway in response to herbivore attack, leading to the synthesis of defensive compounds such as lignin to enhance resistance. In this study, feeding by the F. occidentalis significantly upregulated key genes in the phenylpropanoid pathway in cucumbers, including phenylalanine ammonia-lyase (PAL), caffeoyl-CoA O-methyltransferase (COMT), cinnamoyl-CoA reductase (CCR), and cinnamyl alcohol dehydrogenase (CAD). PAL serves as a critical regulatory node in the phenylpropanoid pathway, catalyzing the first committed step and directing carbon flux from primary metabolism into phenylpropanoid biosynthesis [40]. COMT is a key enzyme involved in lignin monomer methylation [41]. CCR and CAD coordinately participate in the biosynthesis of lignin monomer precursors, such as p-coumaryl alcohol, coniferyl alcohol, and sinapyl alcohol, thereby contributing to lignin accumulation [42]. The observed upregulation of these genes likely reflects activation of the phenylpropanoid pathway in response to thrips feeding. Metabolomic analysis showed a significant increase in lignin content in both cucumbers and cowpeas following infestation. The upregulation pattern of lignin biosynthesis genes observed in this study is highly consistent with findings from other plant–herbivore interaction systems. For example, multiple phenylpropanoid pathway genes (including PAL, CCR, and COMT) were co-activated, collectively promoting lignin synthesis in tomato plants infested by the Tetranychus urticae (Koch) [43]. Similarly, the coordinated upregulation of PAL, CAD, CCR, HCT, and COMT genes drives increased lignin accumulation in tea plants attacked by tea aphids [37]. These results collectively demonstrate that the phenylpropanoid metabolic pathway, particularly its lignin biosynthesis branch, plays a conserved and widespread role in plant defense against herbivores. Notably, although both cucumbers and cowpeas activate lignin biosynthesis in response to F. occidentalis feeding, they exhibit distinct regulatory patterns. The cucumber plant relies on the coordinated upregulation of COMT, CCR, and CAD to enhance lignin production, whereas cowpeas achieve this primarily through specific activation of the CAD gene. This divergence may reflect species-specific defense strategies shaped by differences in metabolic architecture or evolutionary adaptation to herbivore pressure.

F. occidentalis infestation triggers distinct gene expression and regulation patterns associated with brassinosteroid (BR) and jasmonic acid (JA) signaling and phenylpropanoid biosynthesis in cucumbers and cowpeas. This results in divergent accumulations of secondary defensive metabolites such as sinapyl alcohol, which modulates the adaptability and host preference of thrips. Host preference further affects thrips growth and development, with high lignin content in plants exerting negative effects on these processes (Figure 8).

To elucidate the temporal dynamics of host plant hormones, sampling at designated time points post-infestation (e.g., 0, 12, 24, 48 h), followed by UPLC–MS/MS-based quantification of core phytohormones, including JA, SA, and bioactive BRs, is recommended. Integrating hormone quantification data with transcriptional levels of defense-related genes will enable rigorous correlation analysis, thereby directly validating or refining the hormone-dependent defense regulatory model. Future research will integrate targeted metabolomics (e.g., HPLC–MS/MS) with histochemical staining and imaging to quantify defense-related phytohormone dynamics (e.g., JA, BR) and visualize spatially resolved lignin deposition upon pest infestation. This approach will clarify plant–pest interactions and provide direct metabolite-level evidence for signaling pathway activation. Moreover, by validating changes in lignin content after exogenous application of JA or BR, whether or not hormone and lignin-related genes may serve as potential targets for RNA interference or marker-assisted breeding may be determined. This would also provide a theoretical basis for screening insect-resistance genes and developing pest management strategies.

4. Materials and Methods

4.1. Plant Materials

Cucumber (‘Gold Medal 618’, Xiangyun Seed Industry Co., Ltd., Xintai, China) and cowpea (‘Jiangshan Xiuli’, Beijing Sibeichi Seed Industry Co., Ltd., Beijing, China) plants were grown individually in pots (12 cm top diameter × 9 cm bottom diameter × 11 cm height) filled with a potting mixture of peat, vermiculite, and perlite (3:1:1, v/v). Plants were maintained in a climate-controlled chamber (BSG-250, Shanghai Bosun, Shanghai, China) under the following conditions: a 14 h light/10 h dark photoperiod, 26 ± 1 °C, and 70 ± 5% relative humidity. All plants were kept free from pests and watered every 2–3 days, as needed. At 28 days after sowing, when the cucumbers had reached the five-leaf stage and cowpeas the six-leaf stage, healthy plants were selected for the experiments.

4.2. Test Insect Source

F. occidentalis individuals were originally collected from a pepper greenhouse in Korla, Xinjiang, and subsequently reared under controlled conditions to establish a stable population at the Xinjiang Academy of Agricultural Sciences. The colony was maintained in plastic containers (21 × 14 × 7.5 cm) using green bean pods Phaseolus vulgaris (L.) as a food source under controlled conditions (14 h light/10 h dark photoperiod, 26 ± 1 °C, and relative humidity of 70 ± 5%).

4.3. Preference of Frankliniella occidentalis for Cucumbers and Cowpeas

The Y-shaped tube was fabricated from transparent glass, with each arm and the base measuring 10 cm in length, an inner diameter of 2 cm, and an inter-arm angle of 45°. The two arms were connected via silicone tubing to a purified airflow system consisting of gas flowmeters (QC1-S, Beijing Municipal Institute of Labor Protection, Beijing, China), transparent sampling bags (33 cm × 50 cm, used to enclose whole plants), gas-washing bottles (for air purification and humidification), and drying towers (250 mL, filled with activated carbon for air filtration; Changzhou Kede Thermal Engineering Instrument Co., Ltd., Changzhou, China), while the straight arm was connected to a suction pump. Healthy cucumber and cowpea plants, as well as those previously infested by F. occidentalis, were carefully uprooted, their root systems were immediately wrapped in moist cotton to prevent desiccation, and they were enclosed in transparent sampling bags to maintain physiological integrity during volatile collection. The Y-shaped olfactometer was used to assess the behavioral preference of F. occidentalis between the following paired odor sources: clean air vs. healthy cucumber plants (CSC), clean air vs. healthy cowpea plants (VUC), healthy cucumber plants (CSC) ves. thrips-infested cucumber plants (CST), healthy cowpea plants (VUC) vs. thrips-infested cowpea plants (VUT), healthy cucumber (CSC) vs. healthy cowpea plants (VUC), and thrips-infested cucumber plants (CST) vs. thrips-infested cowpea plants (VUT). The treated odor source combinations were placed at both arms of the Y-shaped tube. A single female adult of F. occidentalis was introduced into the tube through the straight arm, and the airflow on both sides was maintained at 40 mL/min. Insect choice behavior was observed and recorded in real time. An adult thrip was scored as having chosen a treatment if it moved past the first two-thirds of an arm and remained there for at least 15 s; no choice was recorded if no decision was made within 5 min [44]. Each treatment was tested in 10 replicates, with eight thrips per replicate.

4.4. Effects of Host Plants on the Growth, Development, and Reproduction of Frankliniella occidentalis

Three green bean pods, each about 8 cm long, were placed in a Petri dish (11 cm diameter) and 100 adult thrips (female: male = 1:1) were collected from each Petri dish (female-to-male ratio, 1:1). The dish was sealed with perforated cling film. After a 12 h oviposition period, adults were removed [45], and the egg stage—from oviposition to first-instar nymph emergence—was recorded. Newly hatched first-instar nymphs were transferred to individual Petri dishes (60 mm diameter) and reared separately on fresh cucumber or cowpea leaves. Nymphal development and survival were monitored every 24 h until pupation [46]. Newly emerged adult F. occidentalis individuals were individually placed in 60 mm Petri dishes (one insect per dish, without pairing) and provided with fresh cucumber or cowpea leaves as a food source. Dishes were sealed with perforated cling film to allow ventilation and prevent escape. Leaves were replaced daily, and egg-laden leaves were kept for three days until all eggs hatched; the number of emerging nymphs was recorded to determine total oviposition per female [45].

4.5. Transcriptome Analysis of Cucumbers and Cowpeas Following Frankliniella occidentalis Feeding

Pupae of F. occidentalis were collected and allowed to develop into adults. Unmated individuals collected within 6 h after eclosion were starved for 3 h before inoculation. Thirty starved adults were transferred onto cucumber plants at the five-leaf stage and cowpea plants at the six-leaf stage, respectively. Plants were immediately covered with insect-proof cages to prevent escape. After 48 h of feeding, infested cucumber leaves (CST) and infested cowpea leaves (VUT) were harvested, flash-frozen in liquid nitrogen, and stored at −80 °C for transcriptome analysis. Non-infested cucumber (CSC) and cowpea leaves (VUC) served as controls. Each treatment was replicated three times.

Transcriptome sequencing was conducted by Beijing Biomarker Technologies Co., Ltd. (Beijing, China). Total RNA was extracted from plant tissues using the RNAprep Pure Plant Kit (Tiangen, Beijing, China). RNA concentration and purity were measured with the NanoDrop 2000 (Thermo Fisher Scientific, Wilmington, DE, USA), and integrity was assessed using the Agilent Bioanalyzer 2100 system (Agilent Technologies, Santa Clara, CA, USA) with the RNA Nano 6000 Assay Kit. For each sample, 1 μg of total RNA was used for library preparation. Libraries were constructed using the Hieff NGS Ultima Dual-mode mRNA Library Prep Kit for Illumina (Yeasen Biotechnology [Shanghai] Co., Ltd., Shanghai, China), with unique index tags added for multiplexing. Sequencing was performed on the Illumina NovaSeq platform to generate 150 bp paired-end reads. Raw sequencing data were stored in FASTQ format. After preprocessing with a custom Perl script, reads containing adapter sequences, poly-N stretches, or low-quality bases were removed to generate high-quality clean reads. Clean reads were aligned to the reference genome using Hisat2. Transcript assembly was performed with StringTie via the Reference Annotation-Based Transcript (RABT) approach to identify known transcripts and predict novel types. Differential gene expression analysis was conducted using DESeq2, with significantly differentially expressed genes (DEGs) defined as |log_2_FC| ≥ 1 and Benjamini–Hochberg adjusted p-value (p) < 0.05. DEGs were functionally annotated using the NR and KEGG databases.

4.6. Metabolomics Analysis of Cucumbers and Cowpeas Following Frankliniella occidentalis Feeding

F. occidentalis were selected onto cucumber and cowpea plants according to the procedure described in Section 4.5. Leaves from both plant species were harvested, flash-frozen in liquid nitrogen, and stored at −80 °C until non-targeted metabolomics analysis. Each treatment included six biological replicates.

Non-targeted metabolomic analysis was performed by Beijing Biomarker Biotechnology Co., Ltd. (Beijing, China). LC–MS was conducted using a Waters Acquity I-Class PLUS UPLC system coupled to a Waters Xevo G2-XS QTof high-resolution mass spectrometer (Waters, Milford, MA, USA). In both positive and negative ion modes, mobile phase A was 0.1% formic acid in water, and mobile phase B was 0.1% formic acid in acetonitrile. The injection volume was 1 μL. Mass spectrometry data were acquired in MSe mode using MassLynx V4.2 software. Data preprocessing—including peak detection, retention time alignment, and noise filtering—was performed using Progenesis QI 4.2 software. Metabolites were identified by integrating the METLIN database in Progenesis QI with Biomarker’s in-house database. Putative annotations were assigned based on accurate mass matches and theoretical fragment ions, with mass error within ±100 ppm. Raw peak areas were normalized to the total peak area, followed by principal component analysis (PCA) and Spearman correlation analysis to evaluate sample reproducibility. Metabolites were annotated using the KEGG, HMDB, and LIPIDMaps databases. OPLS–DA was performed using the ropls package in R to identify differential metabolites. VIP (Variable Importance in Projection) values were derived from cross-validation, and statistical significance was determined by a two-tailed Student’s t-test. Metabolites meeting the criteria of p < 0.05 and VIP > 1 were defined as differentially expressed metabolites (DEMs).

4.7. Integrated Transcriptomic and Metabolomic Analysis of Cucumbers and Cowpeas

Gene expression data for the selected target metabolites were imported into Cytoscape (Version 3.5.1). All datasets were log2-transformed before analysis. Integrated metabolome and transcriptome analysis was conducted using the Pearson Correlation Coefficient (PCC) and its p-value, with a correlation threshold of PCC > 0.8 to identify strongly correlated gene-metabolite pairs; a corresponding heatmap was generated. KEGG pathways were annotated with both DEGs and DEMs. Pathways enriched with DEGs at adjusted p-value < 0.05 were selected for interpretation. DEGs and DEMs from different treatment groups were jointly mapped onto KEGG pathway maps based on differential metabolite and gene expression results to visualize their functional relationships.

4.8. qRT-PCR Analysis

Nine genes (PAL, CAD, COMT, 4CL, UGT, CCR, JAR1, BRI1, BKI1)) were selected for qRT-PCR in cucumbers, with ELF1A as the internal reference gene. Six genes (PAL, COMT, 4CL, JAR1, BRI1, BKI1)) were selected for qRT-PCR in cowpeas, using ELF1A as the reference. Gene-specific primers were designed using Primer Premier 6.0, and sequences are listed in (Table S1). All primers were synthesized by Sangon Biotech Co., Ltd. (Shanghai, China). F. occidentalis was selected onto cucumber and cowpea plants as described in Section 4.5. Healthy and thrips-infested leaves (48 h) were collected from both plant species, flash-frozen in liquid nitrogen, and stored at −80 °C until RNA extraction and qRT-PCR analysis. Each treatment included three biological replicates. Total RNA was extracted using the RNAprep Pure Plant Kit (Tiangen, Beijing, China). cDNA was synthesized with the miRNA First Strand cDNA Synthesis Kit (Stem-Loop Method) (Vazyme, Nanjing, China) following the manufacturer’s protocol. Gene-specific amplification and CT values were measured using ChamQ SYBR qPCR Master Mix (Vazyme, Nanjing, China) on a real-time PCR system. Relative expression levels of target genes were calculated using the 2^–ΔΔCT^ method and normalized to appropriate reference genes.

4.9. Detection of Defensive Compounds in Cucumbers and Cowpeas

Following the infestation procedure described in Section 4.5, cucumber and cowpea plants were infested with F. occidentalis. Leaf samples were collected from both plant species after 48 h of continuous feeding, immediately flash-frozen in liquid nitrogen, and stored at −80 °C until further analysis. Each treatment included three biological replicates. Phenylalanine ammonia-lyase (PAL) activity, lignin, and total flavonoid contents were quantified using specific assay kits: PAL (G0114W), lignin (G0708W), and total flavonoids (G0118W), all from Greiner Bio-One (Suzhou, China), with measurements performed according to the manufacturer’s instructions.

4.10. Statistical Analysis

The preference of F. occidentalis across treatments was assessed using the Chi-square test [47]. Independent samples t-tests were used to compare developmental durations and life table traits among treatments. The bootstrap method (10,000 iterations) was applied to estimate variance and standard errors for key parameters: pre-adult development time, adult longevity, pre-oviposition period, total oviposition duration, and fecundity [48]. One-way ANOVA with Tukey’s HSD post hoc test (p < 0.05) was used to evaluate gene expression differences in F. occidentalis before and after infestation on cucumber and cowpea plants. All statistical analyses were performed in SPSS 19.0.