‘Candidatus Phytoplasma ziziphi’ Changes the Metabolite Composition of Jujube Tree Leaves and Affects the Feeding Behavior of Its Insect Vector Hishimonus hamatus Kuoh

Simple Summary Phytoplasma are wall-less phytopathogens that invade hundreds of plant species, causing symptoms, including witches’ broom, phyllody, and leaf yellowing. The difficulty in phytoplasma artificial cultivation makes research on the interactions of phytoplasma–plant–vector insects lag behind the research on interactions of other phytopathogen–plant–vector insects. The spread of phytoplasma heavily relies on piercing–sucking insects. In this study, the feeding behavior of the leafhopper Hishimonus hamatus Kuoh fed on healthy Chinese jujube leaves and on jujube witches’ broom (JWB) leaves was investigated to find whether JWB infection changed the feeding behavior or preference of H. hamatus. Then, we performed a metabolomic analysis to inspect the metabolite composition of healthy and JWB-infected jujube leaves and tried to explain why the leafhopper tended to feed on JWB-infected leaves. We found that more small-molecular carbohydrates, free amino acids, and free fatty acids and less lignans, coumarins and triterpenoids were accumulated in JWB-infected leaves, which might be related to more vector leafhopper feeding with a higher frequency. Abstract Hishimonus hamatus Kuoh is a leafhopper species native to China that feeds on Chinese jujube leaves. This leafhopper species has been verified to transmit jujube witches’ broom (JWB) disease, caused by phytoplasma, a fatal plant pathogen, which belongs to the phytoplasma subgroup 16SrV-B. The transmission of JWB phytoplasma largely relies on the feeding behavior of piercing–sucking leafhoppers. However, the specific mechanisms behind how and why the infection of JWB influences the feeding behavior of these leafhoppers are not fully understood. To address this, a study was conducted to compare the feeding patterns of H. hamatus when feeding JWB-infested jujube leaves to healthy leaves using the electrical penetration graph (EPG) technique. Then, a widely targeted metabolome analysis was performed to identify differences in the metabolite composition of JWB-infected jujube leaves and that of healthy jujube leaves. The results of EPG analyses revealed that when feeding on JWB-infected jujube leaves, H. hamatus exhibited an increased frequency of phloem ingestion and spent longer in the phloem feeding phase compared to when feeding on healthy leaves. In addition, the results of metabolomic analyses showed that JWB-infected leaves accumulated higher levels of small-molecular carbohydrates, free amino acids, and free fatty acids, as well as lower levels of lignans, coumarins and triterpenoids compared to healthy leaves. The above results indicated that the H. hamatus preferentially fed on the phloem of infected leaves, which seems to be linked to the transmission of the JWB phytoplasma. The results of metabolomic analyses partially imply that the chemical compounds might play a role in making the infected leaves more attractive to H. hamatus for feeding.


Introduction
Phytoplasma to Mollicutes are prokaryotic plant pathogens, sharing similar morphology and ultrastructure with mycoplasma [1]. They obligately inhabit the phloem

H. hamatus Collection and Rearing
To establish the laboratory population of H. hamatus, the adults and nymphs were collected through net capture from JWB-infected jujube orchards in Yanchuan, located on the west side of Shanxi-Shaanxi Gorge, from July to August of 2022. Then, the collected H. hamatus were transferred to healthy jujube plants (height ≈ 50 cm) that were placed in nylon cages (60 × 45 × 100 cm) in the laboratory, with each cage accommodating around forty adult leafhoppers. The insects were reared at 23 ± 2 • C with a photoperiod Insects 2023, 14, 750 3 of 16 of L:D = 14 h:10 h and with no humidity control. This rearing process occurred from the beginning of September to the middle of October 2022. For the EPG experiments, adults from 7 to 17 days old after emergence were selected, since, in this period, they were sexually mature and actively feeding. Prior to the initiation of EPG experiments, a preliminary assessment was conducted to determine whether the tested leafhopper carried JWB phytoplasma. This assessment involved amplifying a segment of 16S ribosomal DNA (rDNA) of JWB phytoplasma using the polymerase chain reaction (PCR) technique [16].

Plant Rearing
The jujube plants chosen for EPG tests, which were free from JWB phytoplasma, were two-year old seedlings that had been cultivated with a height of about 80 cm through tissuecultured techniques. These healthy seedlings were cultivated under controlled conditions where the temperature was maintained at 23 ± 2 • C, with a photoperiod of L:D = 14 h:10 h and with no humidity control. Subsequently, approximately one hundred H. hamatus, captured from JWB-infected jujube orchards, were introduced to ten healthy seedlings and maintained for two weeks. Both the uninfected healthy plants and those infested with leafhoppers were grown in pots with a capacity of 12.0 L and were watered once a week. The infestation status of jujube plants for electropenetrography or wide metabolomic analysis was determined by amplifying 16S-rDNA from PWB phytoplasma using a nested PCR technique as well as witches' broom symptoms [16].

Electropenetrography
A Giga-8 DC EPG system was used (W.F. Tjallingii, Dept. of Entomology, Wageningen Agricultural University, Wageningen, The Netherlands) to evaluate the probing activities of H. hamatus that feed on JWB-non-infected and JWB-infected jujube leaves. The recordings took place in the Entomology Laboratory at Yan'an University. To ensure consistent conditions, the adults were starved for 4 h before the tests. Five minutes before the tests, the leafhoppers were anesthetized using CO 2 for two seconds. A gold wire (3 cm length; 20 µm diameter) was attached to the prothorax of each leafhopper using silver glue (EPG Systems, Wageningen, The Netherlands). The other end of this wire was connected to a copper cylinder (2 cm length, 2 mm diameter) that was inserted into the input of the EPG head stage amplifier. Each leafhopper was connected to an amplifier input before being placed on a jujube leaf. Another copper wire electrode (20 cm length, 2 mm diameter) was inserted into the soil of the jujube container. The plants, leafhoppers, and amplifiers were all set up inside a Faraday cage (80 × 60 × 100 cm) to shield the static electricity and other noise sources. The EPG recordings were made in the laboratory at 23 ± 2 • C with a relative humidity of 60 ± 10%. Output was set at 50 × gain, and plant voltage was adjusted so that the EPG signal fit between +5 V and −5 V. All recordings were made between 12:00 and 12:30 every day. A total of 51 recordings were made, and each day, six recordings were run. Each single recording was represented by a different plant-leafhopper combination, one male or one female on JWB-non-infected or JWB-infected jujube. In the case of falling from leaves, the leafhopper was repositioned. At the end of the recording, dead individuals were noted and excluded from further analyses.
EPG signals were acquired and analyzed using Stylet+d software. The EPG waveforms were identified and measured based on the capture signals (as shown in Figure S1): (1) waveform Np, representing non-probing behavior; (2) waveform A, corresponding to stylet pathway phase; (3) waveform C, representing active ingestion of xylem sap; (4) waveform E, representing passive ingestion of phloem sap.

EPG Statistic Analysis
All of the EPG recordings were analyzed using R-based software (v. 4.2.2, R core team). Then, this table was read in the R package dplyr, and descriptive statistics were run. In the table of original recording data, each row corresponded to a single recording (a unique combination of one leafhopper and one sex), while each column represented an EPG variable. Univariate analyses were conducted starting from the Generalized Linear Model (GLM): quasi-Poisson for counts and Gamma or inverse-Gaussian for positive continuous variables [33,34]. Pairwise comparison (post hoc) between effects due to JWB infection status and sex was calculated with estimated marginal means and 95% confidence intervals with the function emmeans from the "emmeans" package and p-value adjustment using the method of Tukey [35].

Widely Targeted Metabolome Analysis
The biological samples from Chinese jujube leaves were freeze-dried in a freeze-dryer (Scientz-100F). The dried specimens were homogenized using a mixer mill (MM400, Retch) for 90 s at 30 Hz. Fifty milligrams of lyophilized powder from each sample was dissolved in a 1.2 mL 70% methanol solution. Then, we vortexed 30 s every 30 min six times. After centrifuging at 12,000 rpm for 3 min, the supernatants were filtrated before UPLC-MS/MS analysis. The widely targeted metabolome analysis was carried out on the UPLC-MS/MS system (UPLC: Shim-pack UFLC SHIMADZU CBM30A, Kyoto, Japan; MS/MS: SCIEX QTRAP 6500, Applied Biosystems, Framingham, MA, USA). The analytical conditions were as follows: UPLC: column, Agilent SB-C18 (1.8 µm, 2.1 mm × 100 mm). The mobile phase consisted of solvent A (0.1 formic acid in pure water) and solvent B (0.1% formic acid in acetonitrile). The sample measurements were performed with a gradient program that employed starting conditions of 95% A and 5% B. Within 9 min, a linear gradient of 5% A and 95% B was programmed. The ratio of 95% A and 5% B was kept for 1 min. Subsequently, a composition of 95% A, 5.0% B was adjusted within 1.1 min and kept for 2.9 min. We set the flow velocity at 0.35 mL/min, the column oven at 40 • C, and the injection volume at 4 µL. The primary and secondary spectral properties of metabolites were determined based on public databases of metabolites and the self-constructed MWDB V2.0 database (Metware Biotechnology Co., Ltd. Wuhan, Hubei, China). The metabolites were quantified with triple-quadrupole mass spectrometry.

Feeding Behavior of H. hamatus
The EPG results of H. hamatus stylet behavior are listed in Table 1 leaves and on healthy leaves (t = 0.332, p = 0.988), while there were significant differences in the durations of waveforms A per male between the two treatments (t = 3.149, p = 0.015). Neither JWB infection (t = 0.270, p = 0.789) nor the interaction between infection and sex (t = −0.725, p = 0.472) affected the number of non-probing frequencies. However, for male adults, the average duration of non-probing waveforms was significantly shorter on JWB-infected jujube leaves than on healthy jujube leaves (t = −3.083, p = 0.017), while for females, the average duration of non-probing waveforms did not show a significant difference between JWB-infected and healthy leaves (t = 0.628, p = 0.923).
Each column reports a single combination of Treat and Sex. Each row represents a specific variable: quasi-Poisson or negative-binomial for counts, Gamma or inverse-Gaussian for continuous time variables, and beta-regression for proportions. In the case of no effect for Sex and Treat × Sex, the GLM was run with only Treat as an explanatory variable (indicated in the tables with the * sign after the specific variable name). In the case of effect for Sex or Treat × Sex, GLM was run with all three explanatory variables (indicated in the tables with the ** sign after the specific variable name). In the case of only an effect for Sex, the GLM was run with only Sex as an explanatory variable (indicated in the tables with the # symbol after the specific variable name). Post-hoc comparisons were conducted with the Tukey method for p-value adjustment at significant levels of 0.05 and 95% confidence intervals. Different letters in each column represented that there is significant difference among different groups after post-hoc comparisons.
To examine the comprehensive impact of the explanatory variables treatment, sex, and their interactions on H. hamatus behavior, a constrained Canonical Correspondence Analysis (CCA) was performed ( Figure 1). The non-multi-collinear variables that are more closely associated with the various groups are represented by CCA. In particular, ellipses were drawn containing 99% confidence intervals for standard errors associated with the treatment variable, considering the absence of effects for sex and treat × sex ( Table 2). This representation emphasized the distinction between H. hamatus feeding on JWB-infected and healthy jujube leaves. The results of the perMANOVA supported the findings of the CCA (Table 2), which highlighted significant differences between JWBinfected and healthy leaves, while neither sex nor the combination of treat and sex showed any significant differences. case of only an effect for Sex, the GLM was run with only Sex as an explanatory var in the tables with the # symbol after the specific variable name). Post-hoc compar ducted with the Tukey method for p-value adjustment at significant levels of 0.05 and intervals. Different letters in each column represented that there is significant differe ferent groups after post-hoc comparisons.

Metabolomic Analysis of Ziziphus jujuba Leaves Infected by Phytoplasma
To reveal potential metabolic pathways responding to phytoplasma used UPLC-MS/MS termed as widely targeted metabolomics to qualitative tatively detect metabolites in jujube leaves. Firstly, the result of princip

Metabolomic Analysis of Ziziphus jujuba Leaves Infected by Phytoplasma
To reveal potential metabolic pathways responding to phytoplasma infection, we used UPLC-MS/MS termed as widely targeted metabolomics to qualitatively and quantitatively detect metabolites in jujube leaves. Firstly, the result of principal component analysis (PCA) suggested that PC1 could annotate 77.3% variety, indicating a clear separation of metabolomics from two groups and the high reliability of this experiment (Figure 2a). In addition, the result of orthogonal projections to latent structure-discriminant analysis (OPLS-DA) indicated a distinct division of metabolites between the phytoplasma-infected group and the healthy group (R 2 X = 0.939, R 2 Y = 1, Q 2 = 0.999) (Figure 2b). As the Q 2 value of the OPLS-DA model was larger than 0.9, the evaluation model was stable.

JWB Infection Affected the Feeding Behavior of Hishimonus hamatus
To investigate the potential factors contributing to varying transmission efficiencies of JWB phytoplasma, this study focused on analyzing the feeding behavior of the JWB leafhopper vector H. hamatus on both JWB-infected and healthy jujube leaves. This investigation is crucial since phytoplasmas are phloem-limited plant pathogens. Consequently, the phases of phloem feeding are of particular importance as they are closely linked to the vector's ability to acquire and transmit the phytoplasma pathogen [33].
The EPG technique has been widely employed to study interactions involving plant pathogens, piercing-sucking insect-host plant, and host plants [39][40][41]. In the current study of H. hamatus feeding on jujube leaves, the results showed that male H. hamatus feeding on JWB-infected leaves exhibited a notably shorter non-probing phase and an extended pathway phase compared to males feeding on healthy leaves. Both sexes performed significantly more phloem probing events and longer phloem ingestion durations on JWB-infected leaves than healthy leaves. This observation corresponds with findings from studies involving other vector insects and their interactions with phytoplasma-infected plants. In the study of the leafhopper Scaphoideus titanus Ball that transmits Flavescence dorée, a longer duration of phloem ingestion events was found on the FD-susceptible cultivar Barbera than for the FD-tolerant cultivars Brachetto and Moscato [33]. Similarly, during the EPG analysis of the leafhopper Matsumuratettix hiroglyphicus Matsumura, which transmits sugarcane white leaf (SCWL) disease phytoplasma, longer durations of waveform C (phloem salivation) and waveform D (phloem ingestion) were recorded when feeding on symptomatic and asymptomatic SCWL-infected sugarcane plants compared to healthy plants [42]. However, 'Ca. Phytoplasma mali'-infected apple plants did not affect the phloem ingestion behavior of the summer apple psyllid Cacopsylla picta Flor [43]. Conversely, infection with 'Ca. P. mali' did not appear to affect the phloem ingestion behavior of C. picta. To gain a deeper understanding of these dynamics, more EPG data focused on phytoplasma-insect vectors are necessary.

JWB Infection Changed the Metabolite Composition of Chinese Jujube Leaves
The extended duration spent in the phloem phase by piercing-sucking insects could, indeed, signify an increased likelihood of transmitting phytoplasma to non-infected plants, given that phytoplasmas are obligate inhabitants of the plant phloem. In the context of persistently transmitted phytoplasma, successful transmission requires insect vectors to transfer the pathogens from infected plants to healthy ones [42,43]. However, the relationship between phytoplasma infection and the survival and development of vector insects is complex. While prolonged phloem feeding might enhance transmission potential, phytoplasma infection might not necessarily facilitate the overall survival and development of vector insects. In fact, interactions between phytoplasma and vector insects can be multifaceted. To shed light on the potential reasons making H. hamatus feeding on JWBinfected jujube leaves longer than those on healthy leaves in the phloem ingestion phase, a widely targeted metabolomic analysis was performed to research the differential metabolite composition between JWB-infected jujube leaves and healthy leaves.

Changes of Lipids
In the present study, the metabolomic analyses revealed that there were increased accumulations of 17 free fatty acids, 11 glycerol esters, 28 lysophosphatidylcholines (lysoPCs), and 21 lysophosphatidyl ethanolamines (lysoPEs), and only 1 decreased free fatty acids, 2 glycerol esters, and 2 lysoPEs in JWB-infected leaves compared to healthy leaves ( Figure 3, Table S2). Similar results were observed between phytoplasma-infected and healthy leaves of sweet cherry leaves [45]. The elevated lipid accumulation might be attributed to the destruction of phospholipid bilayers in the host plant caused by phytoplasma infection. Fatty acids serve as fundamental components for constructing membrane lipids, which are essential throughout the life cycle of any cellular organism [46]. Given that phytoplasma lacks the mechanism for de novo fatty acid synthesis, it is plausible that these organisms import the necessary components for membrane lipid assembly from their host plants [14,[45][46][47][48]. A lack of unsaturated fatty acids could influence the survival and development of insects. For instance, the loss of the FATTY ACID DESATURASE 7 (FAD7) Arabidopsis thaliana L mutation variety impacted the growth of Myzus persicae Sulzer [49]. Perhaps the up-accumulation of unsaturated fatty acids in JWB-infected jujube leaves facilitated the survival and development of H. hamatus.
Given that phytoplasma lacks the mechanism for de novo fatty acid that these organisms import the necessary components for membra their host plants [14,[45][46][47][48]. A lack of unsaturated fatty acids could and development of insects. For instance, the loss of the FATTY (FAD7) Arabidopsis thaliana L mutation variety impacted the growt zer [49]. Perhaps the up-accumulation of unsaturated fatty acids leaves facilitated the survival and development of H. hamatus. Figure 3. Heatmap of saccharides. The vertical axis shows the clustering shows the sample names. The depth of the purple color represents the r regulated metabolites, and orange represents down-regulated metabolites

Changes of Amino Acids
Amino acids participate in various processes of plant growth, development, and homeostasis [50]. While phytoplasma lacks genes for de novo amino acid synthesis, amino acids from the sieve tube elements of host plants might facilitate phytoplasma multiplication [45]. Furthermore, the amino acids within host plants can influence the preferences of sap-feeding insects [51]. If increased levels of amino acids in host plants, caused by phytoplasma infection, benefit the survival and development of sap-feeding phytoplasma vectors, this could lead to a feedback loop promoting the spread of phytoplasma. In the current study, elevated levels of arginine, histidine, valine, leucine, asparagine, tyrosine, tryptophan, and phenylalanine were found in JWB-infected jujube leaves (Table 3, Figure 4). Similar observations were noted in sweet cherry virescence leaves, elevated levels tyrosine, tryptophan, and phenylalanine, which were invaded by phytoplasma that belonged to the 16SrV-B phytoplasma subgroup [45]. Likewise, 'Ca. P. mali'-infected apple leaves showed higher amounts of alanine, serine, aspartic acid, asparagine, and threonine compared to healthy apple leaves [52]. The relationship between phytoplasma infection and amino acid composition necessitates further exploration to understand its implications for phytoplasma-vector interactions.
acids from the sieve tube elements of host plants might facilitate phytoplasma tion [45]. Furthermore, the amino acids within host plants can influence the of sap-feeding insects [51]. If increased levels of amino acids in host plants phytoplasma infection, benefit the survival and development of sap-feeding p vectors, this could lead to a feedback loop promoting the spread of phytopla current study, elevated levels of arginine, histidine, valine, leucine, asparagin tryptophan, and phenylalanine were found in JWB-infected jujube leaves (Tab 4). Similar observations were noted in sweet cherry virescence leaves, elevate rosine, tryptophan, and phenylalanine, which were invaded by phytoplasm longed to the 16SrV-B phytoplasma subgroup [45]. Likewise, 'Ca. P. mali'-inf leaves showed higher amounts of alanine, serine, aspartic acid, asparagine, an compared to healthy apple leaves [52]. The relationship between phytoplasm and amino acid composition necessitates further exploration to understand tions for phytoplasma-vector interactions.

Changes of Lignans and Coumarins
Lignans are phenylpropanoid dimers where the phenylpropane units ar the central carbon (C8) of their side chains [53]. These compounds have shown and deterrent activities on various insects, including coleopteran [54], dipter mipteran [56], and lepidopteran pests [57]. In the current research, it was ob sixteen out of twenty-four lignans were down-accumulated in JWB-infected ju while only three lignans were up-accumulated (Table 3, Figure 5). The differ mulation of lignans might be related to the preference of H. hamatus to feed fected jujube leaves.
Coumarins are phenolic substances composed of fused benzene and α-py derived from the shikimate pathway. The coumarin 2H-1-benzopyran-2-on high toxicity to Myzus persicae Sulzer [58]. Natural coumarin shows toxicity to litura Fabricius via the inhibition of detoxification enzymes and glucometabol

Changes of Lignans and Coumarins
Lignans are phenylpropanoid dimers where the phenylpropane units are linked by the central carbon (C8) of their side chains [53]. These compounds have shown antifeedant and deterrent activities on various insects, including coleopteran [54], dipteran [55], hemipteran [56], and lepidopteran pests [57]. In the current research, it was observed that sixteen out of twenty-four lignans were down-accumulated in JWB-infected jujube leaves, while only three lignans were up-accumulated (Table 3, Figure 5). The differential accumulation of lignans might be related to the preference of H. hamatus to feed on JWB-infected jujube leaves.
Coumarins are phenolic substances composed of fused benzene and α-pyrone rings, derived from the shikimate pathway. The coumarin 2H-1-benzopyran-2-one exhibited high toxicity to Myzus persicae Sulzer [58]. Natural coumarin shows toxicity to Spodoptera litura Fabricius via the inhibition of detoxification enzymes and glucometabolism [59]. In the present study, a decrease in coumarin levels could potentially be linked to the feeding preference of H. hamatus for JWB-infected jujube leaves (Table S2).
Insects 2023, 14, x FOR PEER REVIEW the present study, a decrease in coumarin levels could potentially be linked to preference of H. hamatus for JWB-infected jujube leaves (Table S2).

Changes of Triterpenoids
Triterpenoids are one of the most numerous and diverse groups of plan metabolites. They exist in either simple, unmodified form or as conjugates wi drates and other macromolecules, especially as triterpene glycosides [60]. Trit cosides contribute to the resistance of pest insects and pathogens, as a gro showed antifeedant, growth inhibition, and poison abilities [60,61]. For exam acid isolated from Shorea robusta Sal inhibited the growth and extended the inst of Oxya fuscovittata Maschall [60]. α-and β-amyrin acetates, extracted from Ma sericea Mart., exhibited potent growth inhibitory effects on two phytophagous insects, Dysdercus peruvianus Guérin-Méneville and Oncopeltus fasciatus Dallas present study, the accumulation level of asiatic acid was 6.82% in JWB-infe leaves compared to healthy jujube leaves. Similarly, the accumulation level of was 25.32% in JWB-infected leaves compared to healthy leaves (Table S2). Th in triterpenoids is hypothesized to influence the feeding preference of H. hama infected jujube leaves. However, further consideration is warranted regarding of specific triterpenoids on the survival and development of this leafhopper sp The presence of higher quantities of low-molecular carbohydrates, free a and free fatty acids in JWB-infected leaves could potentially contribute to the l preference to feed on leaves. These compounds may offer enhanced nutrition or energy sources for the leafhoppers, thereby influencing their feeding beh versely, the reduced amounts of lignans, coumarins and triterpenoids in JW leaves could also play a role in shaping the leafhopper's feeding preference. pounds might deter or affect the leafhopper's perception of the leaves, imp feeding decisions.

Changes of Triterpenoids
Triterpenoids are one of the most numerous and diverse groups of plant secondary metabolites. They exist in either simple, unmodified form or as conjugates with carbohydrates and other macromolecules, especially as triterpene glycosides [60]. Triterpene glycosides contribute to the resistance of pest insects and pathogens, as a group of them showed antifeedant, growth inhibition, and poison abilities [60,61]. For example, asiatic acid isolated from Shorea robusta Sal inhibited the growth and extended the instar duration of Oxya fuscovittata Maschall [60]. αand β-amyrin acetates, extracted from Manilkara subsericea Mart., exhibited potent growth inhibitory effects on two phytophagous hemipteran insects, Dysdercus peruvianus Guérin-Méneville and Oncopeltus fasciatus Dallas [62]. In the present study, the accumulation level of asiatic acid was 6.82% in JWB-infected jujube leaves compared to healthy jujube leaves. Similarly, the accumulation level of β-Amyrone was 25.32% in JWB-infected leaves compared to healthy leaves (Table S2). The reduction in triterpenoids is hypothesized to influence the feeding preference of H. hamatus on JWBinfected jujube leaves. However, further consideration is warranted regarding the impact of specific triterpenoids on the survival and development of this leafhopper species.
The presence of higher quantities of low-molecular carbohydrates, free amino acids, and free fatty acids in JWB-infected leaves could potentially contribute to the leafhopper's preference to feed on leaves. These compounds may offer enhanced nutritional resources or energy sources for the leafhoppers, thereby influencing their feeding behavior. Conversely, the reduced amounts of lignans, coumarins and triterpenoids in JWB-infected leaves could also play a role in shaping the leafhopper's feeding preference. These compounds might deter or affect the leafhopper's perception of the leaves, impacting their feeding decisions.

Conclusions
The results of this work lead us to the conclusion that the leafhopper H. hamatus exhibited distinct feeding behavior when consuming JWB-infected jujube leaves compared to healthy leaves. Specifically, the leafhoppers that fed on JWB-infected leaves displayed increased frequency and a longer duration of phloem ingestion. This altered feeding behavior might be attributed, in part, to the differing metabolite composition between the two types of leaves. This research could contribute to a deeper understanding of plantphytoplasma-insect interactions and their implications for the spread of phytoplasmaassociated diseases.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/insects14090750/s1. Table S1: original data for each leafhopper individual based on EPG analyses. "CK", and "IN" represents the leafhoppers feeding on healthy and JWB-infected leaves, respectively. "F", and "M" represents female and male adults, respectively. A, waveform Np, non-probing; C, waveform A, pathway phase; E, waveform C, active xylem ingestion; G, waveform E, passive phloem ingestion. Table S2: Differential accumulated metabolites between healthy and JWB-infected jujube leaves. Figure

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.