Arboviruses manipulate rice’s volatile emissions, protecting insect vectors from natural enemies in the field

Abstract

Vector-borne plant viruses depend on insect vectors for transmission and often suppress host defenses that limit vector survival and spread. However, their impact on volatile-mediated indirect defenses remains unclear. Here, we show that rice viruses inhibit methyl salicylate (MeSA) emission, impairing parasitoid recruitment and promoting vector persistence. Field experiments demonstrate that MeSA, a key herbivore-induced volatile, suppresses vector populations by attracting egg parasitoids. Viruses counter this by targeting basic-helix-loop-helix transcription factor OsMYC2, a jasmonic acid signaling hub, thereby down-regulating OsBSMT1 and MeSA biosynthesis, responses conserved across diverse rice viruses and vector species. MeSA applications in the field restore parasitoid-mediated vector suppression, highlighting its potential for sustainable disease control. MeSA is a central ecological signal in a previously unidentified viral strategy that enhances transmission.

Rice viruses modify plant volatiles to protect insect vectors from natural enemies.

INTRODUCTION

Vector-borne plant pathogens pose a major threat to global agriculture, causing devastating epidemics that reduce crop yield, threaten food security, and disrupt ecosystems (1). Over 75% of plant viruses rely on insect vectors—such as aphids, whiteflies, and planthoppers—for transmission, making them particularly difficult to control (2). In rice agroecosystems, planthoppers and leafhoppers not only inflict direct feeding damage but also transmit viruses that decimate harvests. While chemical insecticides have long been used to manage these threats, they carry serious environmental and health concerns (3, 4). Although alternative control strategies have been developed, their effectiveness remains limited by an incomplete understanding of the complex ecological interactions among hosts, vectors, pathogens, and the surrounding environment.

Plants have evolved sophisticated defenses to counter insect herbivores and pathogens, including direct defenses that impair feeding (5–9) and indirect defenses that recruit natural enemies such as parasitoids (5, 8, 10–14). Salicylic acid (SA) plays a central role in antiviral immunity by restricting viral replication and movement and by inducing RNA silencing pathways (15–18). Its methylated volatile derivative, methyl salicylate (MeSA), acts as an “alarm call” that attracts natural enemies to herbivore-infested plants (19–22). Notably, many parasitoids and predators have specialized olfactory receptors for MeSA, enabling them to respond with high sensitivity to this cue by modifying their foraging and oviposition behaviors (23–25). MeSA applications have successfully enhanced biological control in various crops such as vineyards, hop yards (19), red maple plantations (26), strawberry (27), soybean (28), and wheat fields (29). However, unlike complex blends of herbivore-induced volatiles, single-compound releases such as MeSA offer a tractable strategy for ecological deployment. In rice, a system where viral transmission contributes more substantially to yield loss than herbivory alone, the ecological function of MeSA and its utility as a control tool remain largely unexplored.

Here, we uncover an ecological strategy used by rice-infecting arboviruses to suppress MeSA-mediated indirect defenses. Combining field experiments, behavioral assays, and molecular analyses, we show that MeSA enhances parasitoid populations while suppressing insect vectors, whereas viruses disrupt this defense by inhibiting the OsMYC2-Oryza sativa benzoic acid/salicylic acid carboxyl methyltransferase 1 (OsBSMT1) pathway. Field experiments further demonstrate that MeSA application restores parasitoid-mediated control in virus-infected rice, suggesting a sustainable pest management strategy. This work uncovers a previously unidentified MeSA-mediated ecological interaction, providing critical insights into host-pathogen-vector-parasitoid dynamics, and suggests that MeSA-based strategies for integrated pest and pathogen management are feasible for rice agroecosystems.

RESULTS

MeSA supplementation to rice fields increases parasitoids and decreases insect viral vectors

To evaluate whether MeSA supplementation could suppress vector populations and reduce viral transmission, we deployed sustained-release spheres emitting 1.45 mg/m² per day across an 8-acre (3.2375-ha) rice field in Jiangsu Province during a 10-week growing season (Fig. 1A and fig. S1A). This release rate was estimated to be within an order of magnitude of the brown planthopper (BPH)–elicited MeSA emissions from a rice crop. Considering the effects of MeSA volatilization and diffusion, a block design was adopted to minimize potential interference between nearby test points. Two fields of similar area were designated as the control and MeSA-release treatment fields, respectively. Each field contained five focal areas. Family-level surveys of insect populations conducted at five focal areas within both MeSA-release and control portions of the fields revealed marked proportional and absolute numerical responses to the MeSA release. No significant differences were observed in either the proportion or absolute numbers of natural enemy and herbivorous insects between the two fields before the MeSA release. However, over the subsequent 10 weeks, significant differences emerged in multiple surveys, with both the percentage and number of insects varying between the MeSA-treated and control fields. Natural enemies, particularly parasitic wasps (Apocrita), increased (Fig. 1, B and C; fig. S1B; and table S1), while herbivorous insects declined (Fig. 1, D and E; fig. S1, C and D; and table S1). Notably, virus-vectoring planthoppers (Delphacidae) and leafhoppers (Cicadellidae) exhibited substantial reductions. These responses were observed in the first survey after the start of the MeSA releases and were largely sustained throughout the 10-week experiment.

Fig. 1. MeSA applications enhance parasitoid abundance and reduce vector insect populations in rice plantations.

(A) Aerial schematic of the MeSA release field experiment. (B to E) Proportions of insect natural enemies (B), parasitic wasps (C), herbivorous insects (D), and insect viral vectors (E) among total insects at each time point. Wilcoxon rank sum test: not significant (ns), P ≥ 0.05; *P < 0.05; **P < 0.01. Error bars indicate SD of five biological replicates. (F) Preferences of A. nilaparvatae for MeSA (10 μl, 100 ng/μl) versus n-hexane control in Y-tube olfactometer assays. Fifty-four A. nilaparvatae made choices. Data are shown as percentages; χ² test: *P < 0.05. (G) Schematic of L. striatellus choice assay using different volatiles. (H) Preferences of SBPH for MeSA (10 μl, 100 ng/μl) versus n-hexane in H-tube olfactometer assays. Student’s t test: ****P < 0.0001. Error bars indicate SD of five biological replicates.

The parasitoid wasp Anagrus nilaparvatae (Pang et Wang) is a key natural enemy of planthoppers, parasitizing their eggs and contributing to the regulation of vector populations (30). To assess its behavioral response to MeSA releases, we conducted Y-tube olfactometer assays. A. nilaparvatae showed a significant preference for MeSA over a control (n-hexane) treatment (Fig. 1F). On the other hand, H-tube olfactometer assays revealed that MeSA significantly repels the Laodelphax striatellus (Fallén) [small BPH (SBPH)], one of the most extensively studied planthoppers that transmits multiple arboviruses during feeding (Fig. 1, G and H). These findings suggest that MeSA supplementation of rice plots can mitigate pest damage by enhancing natural enemy populations and repelling insect vectors of viruses, which may, in turn, reduce virus transmission and crop losses.

Rice viruses weaken indirect defenses against vector insects by reducing MeSA emissions

Rice stripe virus (RSV), transmitted by the SBPH, is one of the most destructive rice pathogens in East Asia, causing severe yield losses across major rice-growing regions (31). As with many arboviruses, RSV is presumed to enhance its transmission by suppressing host defenses that constrain vector survival. Given that MeSA plays a key role in both repelling insects and attracting parasitoid wasps, the effect of SBPH infestation and RSV infection on MeSA emission was examined. Mock-inoculated Nipponbare (NPB) rice plants and RSV-infected NPB rice plants (rice plants exhibiting RSV symptoms) were first obtained by inoculating rice plants with nonviruliferous and viruliferous SBPH, respectively. SBPH-infested rice plants and “RSV + SBPH”–treated rice plants were obtained by infesting these two types of plants with nonviruliferous adult SBPH for 48 hours (fig. S2A). Volatile compounds were subsequently measured and identified in mock-inoculated, SBPH-infested, and RSV + SBPH–treated rice plants using solid-phase microextraction (SPME) coupled with gas chromatography–mass spectrometry (GC-MS) (fig. S2B). Among the identified volatiles, nine were considered reliable candidates through National Institute of Standards and Technology (NIST) mass spectral library matching (table S2). Of these, three—namely, 1,6-octadien-3-ol, 3,7-dimethyl, MeSA, and butanoic acid, butyl ester—were emitted in significantly higher amounts from SBPH-infested plants compared to mock-inoculated plants (Fig. 2A and table S2). These results indicate that SBPH infestation can induce the release of many volatile compounds, including MeSA, from the rice plants. When comparing the volatile profiles of RSV + SBPH–treated plants compared to SBPH-infested plants, significantly lower MeSA emissions were detected in RSV + SBPH–treated plants, while the emissions of the other two volatiles remained unchanged (Fig. 2A and table S2). This suggests that RSV suppresses the SBPH-induced increase in MeSA emission from rice plants.

Fig. 2. Rice viruses weaken indirect defenses against vector insects by reducing MeSA emission.

(A) MeSA emissions from mock-inoculated and RSV-infected NPB rice plants treated with or without SBPH infestation, quantified by GC-MS. Student’s t test: **P < 0.01. Error bars indicate SD of three biological replicates. h, hour. (B) Schematic of parasitism rate assays in glasshouse experiment. Mock-inoculated and RSV-infected NPB rice plants were placed at opposite ends of the box, separated by a barrier in the middle. The same number of gravid female SBPH adults was added to each side and allowed to oviposit for 2 days. Parasitoid wasps were placed on the barrier in the middle of the box and allowed to select between plants at either end. Rice plants were collected 3 days later to determine the parasitism rate. White and red eggs represent unparasitized and parasitized eggs, respectively. (C) Parasitism rates of planthopper eggs by A. nilaparvatae: SBPH on RSV-infected versus mock plants. Wilcoxon rank sum test: ****P < 0.0001. Error bars indicate SD of 15 plants. (D to G) Y-tube olfactometer bioassays of A. nilaparvatae preferences: (D) control versus SBPH-infested plants: 56 A. nilaparvatae made choices; (E) RSV-infected versus RSV + SBPH plants: 57 A. nilaparvatae made choices; (F) control versus SBPH-infested plants ± MeSA: 34 (in the top graph) and 39 (in the bottom graph) A. nilaparvatae made choices; (G) RSV-infected versus RSV + SBPH plants ± MeSA: 39 (in the top graph) and 32 (in the bottom graph) A. nilaparvatae made choices. Data are shown as percentages. χ² test was performed to determine statistical significance. ns, P ≥ 0.05; *P < 0.05; **P < 0.01.

Similar suppression of MeSA was observed in plants infected with southern rice black–streaked dwarf virus (SRBSDV) and rice ragged stunt virus (RRSV), transmitted by the Sogatella furcifera (Horváth) [white-backed planthopper (WBPH)] and the Nilaparvata lugens (Stål) (BPH), respectively (fig. S3, A and B). Both viruses are agriculturally devastating, causing extensive yield losses in major rice-producing regions of Asia (32–35). These findings suggest that diverse rice viruses actively attenuate vector-induced MeSA production, potentially undermining indirect defenses and promoting vector survival.

Given that virus infection suppresses MeSA emission and field supplementation enhances parasitoid abundance, the impact of viruses on parasitoid recruitment and plant indirect defenses was evaluated. The effect of RSV infection on the parasitism of SBPH eggs by A. nilaparvatae was examined in a controlled glasshouse setting (Fig. 2B and fig. S4). Fifteen RSV-infected and 15 mock-inoculated NPB rice plants were placed in separate clear containers with a barrier between them. Forty-five female SBPHs, unable to cross the barrier, were allowed to lay eggs for 3 days. Afterward, 75 female A. nilaparvatae, capable of crossing the barrier, were introduced to assess parasitism rates. Parasitism rates were significantly lower on RSV-infected plants (38.08% of 625 eggs) compared to mock-inoculated plants (67.17% of 920 eggs) (Fig. 2C and table S3). Further glasshouse experiments showed that parasitism rates of WBPH eggs on SRBSDV-infected NPB plants (19.48% of 1304 eggs) and BPH eggs on RRSV-infected NPB plants (18.98% of 1064 eggs) were significantly lower than those on mock-inoculated plants (for WBPH, 57.69% of 1066 eggs; for BPH, 66.30% of 641 eggs) (fig. S5, A and B, and tables S4 and S5). These results indicate that arboviruses broadly suppress rice’s indirect defenses against insect vectors.

To assess whether this reflects changes in parasitoid behavior, we performed behavioral choice assays using volatiles from insect-infested and virus-infected NPB rice plants. To obtain SBPH-infested and control (uninfested) plants, we either infested or uninfested 2-week-old NPB rice plants by nonviruliferous SBPH for 48 hours, respectively (fig. S6). When given a choice between SBPH-infested and control plants, A. nilaparvatae showed a significant preference for SBPH-infested plants (Fig. 2D). However, when offered a choice between volatiles from SBPH-infested NPB rice plants and RSV + SBPH–treated NPB rice plants, A. nilaparvatae showed a significant preference for SBPH-infested plants (Fig. 2E). These results indicate that RSV attenuates host indirect defenses by modulating rice volatile organic compounds (VOC) emissions. Moreover, spraying MeSA on control and SBPH-infested rice plants eliminated parasitoid wasps’ preference for SBPH-infested plants (Fig. 2F). MeSA application also restored parasitoid preference, eliminating the difference between SBPH-infested and RSV + SBPH–treated rice plants (Fig. 2G). These observations are consistent with the inference that MeSA is a crucial volatile responsible for the reduced attractiveness of RSV-infected rice plants to A. nilaparvatae and suggest that RSV infection disrupts parasitoid attraction by suppressing MeSA emissions, ultimately weakening rice’s indirect defenses.

Given that MeSA significantly repels SBPH (Fig. 1H), we further investigated the impact of virus-infected rice on this insect. Behavioral assays revealed that nonviruliferous SBPHs significantly preferred RSV-infected rice plants over mock-inoculated rice plants (fig. S7, A and B). This result indicates that RSV-infected rice plants attract SBPH and is consistent with the finding that arboviruses suppress MeSA emission.

Viruses suppress MeSA emission by down-regulating OsBSMT1 transcripts

To investigate the molecular basis of virus-induced suppression of MeSA, we examined the transcriptional responses of both catabolic and biosynthetic pathways. Catabolism was ruled out, as none of the 12 MES genes, which potentially demethylate MeSA into SA (36), exhibited significantly decreased transcripts upon SBPH infestation and increased transcripts upon RSV infection (fig. S8, A to C). In contrast, SA levels increased upon SBPH infestation and further under RSV infection (Fig. 3A). Among 21 SABATH genes that could be detected in rice, only SABATH3 (also known as OsBSMT1) exhibited the expected pattern of increased transcript accumulation upon SBPH infestation and decreased transcript accumulation upon RSV infection (fig. S8, C and D). OsBSMT1, which encodes an enzyme responsible for methylating SA in both Arabidopsis and Escherichia coli (37, 38), was thus considered a likely candidate responsible for the observed alteration in MeSA emission. Monitoring OsBSMT1 transcript levels confirmed a continuous increase along SBPH infestation and a continuous decrease following RSV infection (fig. S8, E to G). Notably, RSV infection resulted in a reduced level of OsBSMT1 transcripts in SBPH-infested plants (Fig. 3B and fig. S8H).

Fig. 3. Viruses suppress MeSA emission by down-regulating OsBSMT1 transcription.

(A) SA levels in mock-inoculated, RSV-infected, and RSV + SBPH–treated NPB rice plants quantified by high-performance liquid chromatography–MS. Error bars represent SD of six biological replicates. (B) OsBSMT1 transcript levels in corresponding samples analyzed by quantitative real-time polymerase chain reaction (qRT-PCR). OsEF-1α was used as an internal control. Error bars represent SD of three biological replicates. (C to F) Preferences of A. nilaparvatae in Y-tube olfactometer assays: (C) Kitaake wild-type (WT) versus OsBSMT1ox: 30 (in the top graph) and 41 (in the bottom graph) A. nilaparvatae made choices; (D) SBPH-infested wild-type versus osbsmt1: 31 (in the top graph) and 48 (in the bottom graph) A. nilaparvatae made choices; (E) control versus SBPH-infested plants in wild-type or osbsmt1 backgrounds: 31 (in the top graph), 32 (in the middle graph), and 38 (in the bottom graph) A. nilaparvatae made choices; (F) SBPH-infested versus RSV + SBPH plants in wild-type or osbsmt1 backgrounds: 34 (in the top graph), 30 (in the middle graph), and 32 (in the bottom graph) A. nilaparvatae made choices. Student’s t test was used in (A) and (B), and χ² test was used in (C) to (F). ns, P ≥ 0.05; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

To further explore the role of OsBSMT1 in MeSA biosynthesis, we generated transgenic rice plants expressing either the pOsBSMT1::GUS reporter construct or the pUbi::OsBSMT1-3xHA overexpression construct in the japonica rice variety, Kitaake, which is widely used as a model for rice transformation. β-glucuronidase (GUS) staining with pOsBSMT1::GUS plants revealed strong promoter activity in seedling roots, stems, and leaves (fig. S9A). Overexpression plants [OsBSMT1 overexpression (OsBSMT1ox)] accumulated higher OsBSMT1 transcript and protein levels without phenotypic changes, exhibited increased MeSA emissions, and were significantly more attractive to A. nilaparvatae than wild-type plants in olfactometer assays (Fig. 3C and fig. S9, B to E). These findings established OsBSMT1 as the key enzyme catalyzing MeSA biosynthesis, directly influencing MeSA emission and the attraction of A. nilaparvatae. Two OsBSMT1 knockout (osbsmt1) rice lines, osbsmt1#7 and osbsmt1#9, were generated in the Kitaake background using the CRISPR-Cas9 genomic editing technology (fig. S9F). The osbsmt1 plants exhibited no significant phenotypic differences compared to wild-type Kitaake plants (fig. S9G). GC-MS analysis demonstrated a complete absence of MeSA emission in SBPH-infested osbsmt1 plants (fig. S9H), revealing that OsBSMT1 is the major rice SABATH responsible for methylating SA to produce MeSA under SBPH infestation. Furthermore, A. nilaparvatae parasitoids significantly preferred SBPH-infested wild-type Kitaake rice plants over SBPH-infested osbsmt1 plants (Fig. 3D). These results underscore the crucial role of MeSA in the attractiveness of rice plants to A. nilaparvatae and highlight the critical function of OsBSMT1 in mediating these effects.

When parasitoid wasps were given the option between odors from SBPH-infested and uninfested control osbsmt1 rice plants, their preference for SBPH-infested wild-type plants disappeared (Fig. 3E). While previous studies suggested that multiple volatile compounds may contribute to the attraction of A. nilaparvatae (19–21), these results emphasize the central role of MeSA emissions in vector-induced attractiveness of A. nilaparvatae. In addition, when A. nilaparvatae parasitoids were given a choice between odors from RSV + SBPH–treated and SBPH-infested osbsmt1 rice plants, the preference for SBPH-infested wild-type plants was eliminated (Fig. 3F). These findings, along with the observed suppression of OsBSMT1 transcripts following RSV infection (Fig 3B), underscore the critical role of OsBSMT1 in mediating RSV-induced reductions in MeSA emission and the subsequent decrease in parasitoid attraction. While RSV infection enhanced the attractiveness of rice plants to SBPH (fig. S7B), SBPH significantly preferred Kitaake plants over OsBSMT1ox plants and preferred osbsmt1 plants over Kitaake plants (fig. S10, A and B). This suggests that RSV also attracts insect vectors through mechanisms dependent on the OsBSMT1 pathway.

Similarly, other planthoppers (WBPH and BPH) up-regulated OsBSMT1 transcript levels (fig. 11, A and B), while their associated viruses (SRBSDV and RRSV) down-regulated OsBSMT1 (fig. S11, C and D) transcript levels. In addition to transmitting RSV, SBPH can also transmit RBSDV, while BPH, in addition to spreading RRSV, also transmits rice grassy stunt virus. The transcript levels of OsBSMT1 are similarly significantly down-regulated in RBSDV/rice grassy stunt virus–infected rice plants compared to mock-inoculated plants (fig. S11, C and D). In contrast, Nephotettix cincticeps (Uhler), the green rice leafhopper (GRL), infestation increased OsBSMT1 transcript levels (fig. S11E), but rice dwarf virus, transmitted by GRL (33, 34), did not suppress OsBSMT1 transcript accumulation (fig. S11, C and D). These observations suggest that many rice viruses, but not all, suppress OsBSMT1 accumulation, which weakens host defenses against their vector insects.

Viruses inhibit OsBSMT1 transcription by disrupting OsMYC2 activity

To investigate the molecular basis of virus-induced suppression of OsBSMT1, we examined the transcriptional regulation. In pUbi::OsBSMT1 transgenic plants, RSV infection did not alter OsBSMT1 transcript or protein levels, ruling out posttranscriptional effects (fig. S12, A to D). However, in pOsBSMT1::GUS reporter lines, GUS transcripts decreased significantly upon RSV infection, indicating that RSV negatively regulates OsBSMT1 transcription through its promoter at the transcriptional level (fig. S12, E and F). Chromatin immunoprecipitation (ChIP) using a polymerase II (Pol II)–specific antibody showed reduced polymerase occupancy at the OsBSMT1 promoter in RSV-infected plants, consistent with transcriptional repression (Fig. 4A and fig. S12G). To identify upstream regulators, we found G box motifs, typically bound by MYC2 (39, 40), in the OsBSMT1 promoter and gene body. ChIP-quantitative polymerase chain reaction (qPCR) confirmed binding of OsMYC2 to the OsBSMT1 promoter in both protoplasts and OsMYC2-overexpressing (OsMYC2ox, Kitaake background) plants (Fig. 4B and fig. S13) (41). Electrophoretic mobility shift assays (EMSAs) further verified direct binding of purified OsMYC2 protein to the OsBSMT1 promoter (Fig. 4C). Quantitative real-time PCR (qRT-PCR) analysis showed that OsBSMT1 transcript levels were elevated in OsMYC2ox plants and reduced in osmyc2 mutants (Fig. 4, D and E). Coexpression of OsMYC2 with pOsBSMT1::GUS in protoplasts significantly up-regulated GUS expression (fig. S14A), suggesting that OsMYC2 activates OsBSMT1 transcription. These results indicate that RSV may reduce MeSA release by inhibiting OsMYC2 function, thereby down-regulating OsBSMT1 expression.

Fig. 4. RSV decreases OsBSMT1 transcription via OsMYC2.

(A) ChIP-qPCR analysis of Pol II binding to the OsBSMT1 promoter in mock-inoculated and RSV-infected NPB rice plants. ChIP assays were performed with Pol II antibody. The OsEF-1α promoter served as a negative control. (B) ChIP-qPCR analysis of OsMYC2 binding to the OsBSMT1 promoter in Kitaake and OsMYC2ox plants. Schematic indicates positions of amplicons near the OsBSMT1 locus. (C) EMSA showing OsMYC2 binding to the OsBSMT1 promoter. MBP was a negative control. Top band indicates DNA-protein complex; bottom band indicates free probe. (D and E) qRT-PCR analysis of OsBSMT1 transcript levels in Kitaake wild-type and osmyc2 (D) and OsMYC2ox (E) rice plants. (F) qRT-PCR of OsBSMT1 in mock-inoculated and RSV-infected Kitaake and osmyc2 plants. OsEF-1α was used as internal control. (G) Y-tube olfactometer assay of A. nilaparvatae preference for Kitaake versus OsMYC2ox or osmyc2 plants: 51 (in the top graph) and 84 (in the bottom graph) A. nilaparvatae made choices. Data are shown as percentages. (H and I) Western blot detection of OsMYC2 in total, nuclear, and cytoplasmic fractions of mock-inoculated and RSV-infected plants. Tubulin and histone H3 were used as cytoplasmic and nuclear markers, respectively. Representative blots (H) and quantification (I) are shown. (J) Immunofluorescence of OsMYC2-FLAG in p35S::OsMYC2-FLAG transgenic plants. Green, anti-FLAG signal; red, 4′,6-diamidino-2-phenylindole (DAPI)–stained nuclei. Scale bars, 50 μm. (K and L) Quantification of OsMYC2-FLAG fluorescence signal in nuclei (K) and cytoplasm (L) using ImageJ (n = 500 for mock and 502 for RSV). Student’s t test [(A), (B), (D) to (F), (I), (K), and (L)] and χ² test (G) were used for significance testing. ns, P ≥ 0.05; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Error bars represent SD of three biological replicates, unless otherwise indicated.

To test whether OsMYC2 mediates RSV-dependent suppression of OsBSMT1, we inoculated both osmyc2 and wild-type Kitaake rice with RSV. While OsBSMT1 transcripts decreased in infected Kitaake plants, this reduction was attenuated in osmyc2 mutants, indicating that RSV suppresses OsBSMT1 transcription via an OsMYC2-dependent mechanism (Fig. 4F and fig. S14B). In Y-tube dual-choice olfactometer assays, A. nilaparvatae significantly preferred OsMYC2ox plants over wild-type Kitaake rice plants (Fig. 4G). Conversely, A. nilaparvatae significantly preferred wild-type Kitaake rice plants over osmyc2 mutant plants (Fig. 4G), underscoring the crucial role of OsMYC2 in A. nilaparvatae attraction. Moreover, the transcript levels of OsMYC2 rose following SBPH infestation, and SBPH significantly preferred osmyc2 over wild-type Kitaake and wild-type Kitaake over OsMYC2ox plants in behavioral choice assays (fig. S15, A to C). These findings highlight the central role of OsMYC2 in modulating OsBSMT1-mediated insect defenses, which are disrupted by RSV infection.

To investigate how RSV manipulates OsMYC2 function to suppress OsBSMT1 transcription, we examined OsMYC2 transcript and protein levels following RSV infection. While OsMYC2 transcript levels remained unchanged, total protein abundance increased following RSV infection (Fig. 4H and fig. S16). Despite MYC2 being a key transcription factors involved in transcriptional reprogramming triggered by increased jasmonic acid (JA) levels (39, 42, 43) and its well-established localization in the nucleus (39), fractionation and immunohistochemistry revealed that OsMYC2 is a nuclear-cytoplasmic protein, localizing to both the nucleus and cytoplasm, and RSV reduced OsMYC2 levels in the nucleus while increasing cytoplasmic accumulation, indicating subcellular redistribution (Fig. 4, H to L). This relocalization likely impairs OsMYC2-dependent transcription of OsBSMT1, compromising MeSA-mediated defenses. Similarly, the infestations of other insect vectors (WBPH, BPH, and GRL) increased the accumulation of OsMYC2 transcripts (fig. S17A). In contrast, vector-borne rice viruses either reduced OsMYC2 transcript accumulation or decreased its nuclear protein levels (fig. S17, B to D). These findings suggest that insect vectors generally activate plant defenses by stimulating OsMYC2 function, while vector-borne rice viruses attenuate host integral defenses by suppressing OsMYC2’s functions.

To identify the mechanism underlying OsMYC2 relocalization, we examined the interaction between RSV proteins and OsMYC2. Yeast two-hybrid screening identified a specific interaction between OsMYC2 and the RSV nonstructural 2 (NS2) protein (fig. S18A). Further coimmunoprecipitation in Nicotiana benthamiana and in vitro pull-down assays confirmed a specific interaction between RSV NS2 and OsMYC2 (fig. S18, B and C). Bimolecular fluorescence complementation showed cytoplasmic localization of the NS2-OsMYC2 complex (fig. S18D). Fractionation assays in NS2-overexpressing (NS2ox, NPB background) plants (44) revealed that NS2 primarily localized in the cytoplasm (fig. S19, A and B). A reduction in the nuclear accumulation of the OsMYC2 protein was consistently observed in NS2ox plants (fig. S19, A and B). These plants also exhibited decreased OsBSMT1 transcript levels and reduced attraction to A. nilaparvatae, while SBPH showed a significant preference for NS2ox plants (fig. S20, A to C). These results demonstrate that RSV NS2 targets OsMYC2, sequestering it in the cytoplasm to disable its regulatory role in insect defenses.

MeSA supplementation in rice fields restores parasitoid-mediated control of vector insect eggs on virus-infected plants

Last, a critical inference was tested in a natural setting by conducting field experiments to assess whether arboviruses affect plant-mediated indirect defenses against insect vectors. Specifically, the effect of infections with different rice viruses on the egg parasitism of their respective vectoring planthoppers by A. nilaparvatae was examined under field conditions.

To assess the effect of RSV infection on parasitism of SBPH eggs, we infested RSV-infected and mock-inoculated NPB rice plants with SBPH females (10 per plant) for 3 days to allow oviposition (Fig. 5A). A total of 64 RSV-infected and 64 mock-inoculated plants were then divided into two groups and transplanted into separate rice fields (field A and field B) in Jiangsu Province, China. Four days after transplantation, plants were collected, and egg numbers and parasitism rates were assessed. The parasitism rate for SBPH eggs on mock-inoculated plants were 62.73% (field A; 856 eggs) and 62.03% (field B; 838 eggs), whereas it was significantly reduced to 41.95% (field A; 1504 eggs) and 40.22% (field B; 1611 eggs) on RSV-infected plants (Fig. 5B and table S6), revealing that RSV infection reduces the effectiveness of parasitoid recruitment under field conditions.

Fig. 5. MeSA application in rice fields restores parasitoid-mediated control of vector insect eggs on virus-infected plants.

(A) Schematic of the field experiment. Mock-inoculated (blue) and RSV-infected (yellow) NPB rice plants were infested with gravid SBPH females for 3 days to allow oviposition and then transplanted into rice field. The eggs of SBPH on rice plants were subsequently parasitized by parasitic wasps under natural field conditions. After more than 4 days, the parasitized eggs turned red. At this point, the rice plants were collected for parasitism rate quantification. White and red eggs represent unparasitized and parasitized eggs, respectively. (B) Parasitism rates of SBPH eggs by A. nilaparvatae on mock-inoculated and RSV-infected plants in two separate field sites (A and B). Data are shown as percentages. Wilcoxon rank sum test was used to assess significance. **P < 0.01; ***P < 0.001. Error bars represent SD of 26 biological replicates. (C) Schematic of field experiment in which mock-inoculated and RSV-infected NPB rice plants were individually infested with SBPH for 3 days to lay eggs. The plants were then transplanted to rice fields in the Jiangsu Province with or without MeSA applications [1.0 g, 50 parts per million (ppm)]. White and red eggs represent unparasitized and parasitized eggs, respectively. (D) Parasitism rates of SBPH eggs by A. nilaparvatae on RSV-infected and mock-inoculated plants with or without MeSA treatment. (E) Parasitism rates of WBPH eggs on SRBSDV-infected and mock-inoculated plants with or without MeSA treatment. (F) Parasitism rates of BPH eggs on RRSV-infected and mock-inoculated plants with or without MeSA treatment. Different letters indicate statistically significant differences (Wilcoxon rank sum test, α = 0.05, n = 26 plants per treatment).

To test whether virus-mediated reductions in parasitism were caused by suppressed MeSA emission, we baited both RSV-infected and mock-inoculated plants harboring planthopper eggs with MeSA released at approximately 3.5 times the natural emission rate of infested plants (Fig. 5C). MeSA supplementation restored parasitoid attraction, with no significant difference in parasitism rates between MeSA-treated RSV-infected and mock-inoculated plants (62.84% of 1168 versus 69.45% of 766 eggs). Notably, MeSA treatment significantly elevated the parasitism rate on RSV-infected plants from 41.33% of 1459 eggs (without MeSA) to 62.84% of 1168 eggs (with MeSA) (Fig. 5D and table S7). Consistent patterns were observed for SRBSDV- and RRSV-infected plants. Virus infection significantly reduced parasitism rates: WBPH eggs on SRBSDV-infected plants showed 27.91% parasitism (5392 eggs), and GRL eggs on RRSV-infected plants showed 7.40% parasitism (3945 eggs) compared to 57.82% (1066 eggs) and 71.48% (2752 eggs) on mock-inoculated plants, respectively. MeSA supplementation restored parasitism, with no significant differences between MeSA-treated SRBSDV-infected and mock plants (61.86% of 4266 eggs versus 60.82% of 2257 eggs) or between MeSA-treated RRSV-infected and mock plants (54.03% of 5815 eggs versus 56.14% of 4710 eggs) (Fig. 5, E and F, and tables S8 and S9). These findings confirm that MeSA mediates parasitoid recruitment and highlight its potential as a field-applicable biocontrol strategy.

DISCUSSION

Vector-borne viral pathogens rely on insect vectors for their transmission, and successful viruses must overcome plant defenses that limit not only their own success but also the survival and dispersal of their vectors. This study shows that MeSA, a methylated SA phytohormone important to viral defense responses, plays a key role in rice’s indirect defense mechanisms by attracting parasitoid wasps that suppress vector insect populations. Moreover, this study shows that rice arborviruses actively suppress MeSA emissions, thereby enhancing vector insect survival and their own spread. This study identifies OsMYC2-OsBSMT1 as a key regulator of this pathway and demonstrates that viral infection disrupts OsMYC2-OsBSMT1 function by abrogating its nuclear localization, leading to reduced MeSA biosynthesis (Fig. 6). These findings suggest that arboviruses suppress indirect host plant defenses to promote between-plant transmission. Hence, OsMYC2-OsBSMT1 emerges as a central regulatory node integrating JA and SA signaling and as a major target of viral interference. Moreover, field experiments show that MeSA supplementation restores parasitoid suppression of vector insects, positioning MeSA as a key ecological signal in plant-virus-vector-parasitoid interactions and offering potential for pest and disease management.

Fig. 6. Proposed model illustrating how rice-infecting viruses attenuate indirect defenses against insect vectors.

(A) Ecological layer: Healthy rice plants indirectly defend against the insect vector, the SBPH, through parasitism by the parasitoid wasp (A. nilaparvatae), which targets planthopper eggs. RSV infection reduces this parasitism and weakens the indirect defense. (B) Chemoecological layer: SBPH infestation enhances rice plant attractiveness to parasitoid wasps through elevated emission of MeSA. RSV infection suppresses this effect, reducing wasp attraction to infested plants. (C) Molecular layer: SBPH feeding up-regulates OsMYC2 expression, which activates OsBSMT1 transcription via binding directly to its promoter. OsBSMT1 catalyzes the methylation of SA to produce MeSA, mediating parasitoid recruitment and enhancing indirect insect defenses. RSV-encoded NS2 protein binds to OsMYC2, retaining it in the cytoplasm and reducing its nuclear activity. This leads to down-regulation of OsBSMT1 and compromised MeSA-mediated parasitoid attraction, facilitating vector survival and viral spread.

Viruses have evolved multiple strategies to manipulate plant defenses in ways that promote vector performance and transmission. Previous studies on virus-induced suppression of plant defenses have primarily focused on direct defenses, where successful vector-borne viruses disrupt plant resistance to promote vector infestations (45, 46). In response to insect infestation, plants release herbivore-induced plant volatiles (HIPVs) to deter these herbivores (12). Increasing evidence shows that arboviruses manipulate HIPVs to attract their vector insects. For example, cucumber mosaic virus enhances aphid attraction by repressing the JA signaling pathway (47). Similarly, tomato yellow leaf curl China virus suppresses terpene biosynthesis and emission, increasing the attraction of tomato plants to whitefly vectors (46). These studies primarily focused on dicots. Our findings indicate that rice viruses attenuate MeSA emission, weakening the plant’s ability to deter vector insects. In addition, cucumber mosaic virus decreases MeSA emission, increasing aphid attraction to infected plants (45). Insect natural enemies also use plant volatiles as foraging cues (10, 12, 14, 48). Plants can achieve indirect defense against herbivorous pests by attracting insect natural enemies. Infection by cucurbit aphid-borne yellows virus alters the plant’s volatile profile, reducing its attractiveness or even causing repellence to the parasitoid wraps (49). In addition to using plant volatiles, infection of host plants by tomato spotted wilt virus can indirectly reduce the predation risk for its natural enemies by accelerating the vector’s growth and development (50). This study shows that virus-induced suppression of MeSA emission disrupts plant-mediated parasitoid attraction, compromising the plant’s indirect defenses in the fields. These observations reveal that modulation of plant volatile emissions is a key viral strategy that enhances vector performance by weakening both direct and indirect defenses of host plants.

On the other hand, parasitoid wasps can counteract virus-vector cooperation, potentially restoring plant defenses against insect vectors (51, 52). For example, Candidatus Liberibacter asiaticus (Las), the causal agent of citrus Huanglongbing disease, increases MeSA emissions, attracting psyllid vectors and facilitating disease spread (53). However, the parasitoid Tamarixia radiata, which preys on psyllids, also responds to MeSA and preferentially attacks psyllids on Las-infected citrus (54). In aphid-virus interactions, cucumber mosaic virus infection not only increases emissions of aphid-attracting volatiles but also enhances aphid parasitism rates by Aphidius colemani (55, 56). Infection by cucumber mosaic virus or potato virus Y alters the VOC profile of pepper plants, making them more attractive to the parasitoid wrap A. colemani (57). These examples illustrate how parasitoids may use virus-induced changes in plant volatiles to locate and attack vector insects, potentially mitigating the spread of viral diseases. Furthermore, cucumber mosaic virus infection can indirectly influence predator behavior by modifying host plant phenotypes; however, these effects appear secondary to prey availability (58). In addition to plant hosts, infection of vector insects by virus can also affect their natural enemies. Aphids carrying cereal yellow dwarf virus show significantly higher parasitism rates by the wasp A. colemani than virus-free aphids (59), revealing the complex interrelationships among pathogens, host plants, vector insects, and natural enemies.

Our observations and a recent study (45) highlight the pivotal role of MeSA in plant defense and its frequent targeting by viruses. OsBSMT1 in rice and SA-carboxylmethyltransferase-1 (SAMT1) in N. benthamiana are the primary enzymes responsible for MeSA synthesis. To reduce MeSA emission, cucumber mosaic virus infection in N. benthamiana suppresses SAMT1 expression (45), while we found that several rice viruses inhibit OsBSMT1 expression. However, these viruses appear to use distinct strategies: Cucumber mosaic virus targets NAC2 via its 1a protein to repress SAMT1 expression (45), whereas RSV and other rice viruses target OsMYC2 via their NS2 protein or functionally analogous proteins to inhibit OsBSMT1. This discrepancy likely reflects species-specific differences in transcriptional regulation of MeSA-mediated defense pathways.

Once released, MeSA is perceived by neighboring plants through the SA-binding protein 2 (SABP2) receptor, which triggers immune responses in these plants (45). Functioning as a methyl esterase (MES), SABP2 catalyzes the conversion of MeSA to SA. The rice genome encodes 12 MES genes, and further research is needed to determine which specific member(s) use MeSA as a substrate. MES enzymes capable of recognizing MeSA are likely candidates for its perception in rice. However, the mechanism underlying MeSA perception and the subsequent signal transduction in rice remains to be elucidated.

While previous studies have identified other volatile compounds that contribute to parasitoid attraction (20–22), the diversity of roles that MeSA plays in plant-insect systems suggests that it is a conserved ecological signal. Moreover, herbivore-inducible SA carboxyl methyltransferase and the associated production of MeSA enhance tomato resistance to lepidopteran pests such as Manduca sexta by acting as chemical deterrents (60). Our findings suggest that MeSA acts as a central airborne signal mediating plant-virus-vector-parasitoid interactions and potentially other complex ecological networks. By manipulating MeSA emission, vector-borne viruses appear to suppress host indirect defenses, tipping ecological interactions in favor of their insect vectors while reducing attraction of natural enemies. Given that the different rice viruses used in our study and cucumber mosaic virus, a dicot-infecting virus, all attenuate host MeSA emission (45), additional work is needed to determine whether other vector-borne viruses similarly manipulate MeSA and other HIPV pathways and coordinately suppress host defenses to viruses, as well as the direct and indirect defenses in other crop systems.

Given its key role in plant resistance to insect pests, MeSA has been explored as a biological control strategy for various insect species. However, before MeSA can be widely implemented in large-scale agricultural systems, extensive field experiments are needed to assess its effectiveness under natural conditions, where pathogens, pest insects, and natural enemies coexist. Understanding how MeSA influences these complex ecological interactions across different cropping systems will be crucial for optimizing its application. Our large-scale field release experiment demonstrated that MeSA plays a critical role in regulating the ratio of herbivorous insects to natural enemies. Although the experimental design was not randomized but rather followed a block design with an inherent limitation, repeated long-term observations and subsequent follow-up experiments consistently confirmed the important ecological role of MeSA. Furthermore, this study provides direct evidence that MeSA application counteracts virus-induced suppression of parasitoid attraction in rice fields, highlighting its potential for integrated pest and disease management. Future research should explore the long-term ecological stability of MeSA-based strategies, including their impact on nontarget organisms and potential trade-offs with other defense mechanisms. Combining MeSA application with parasitoid releases or the deployment of resistant rice cultivars may provide a more sustainable and synergistic approach to managing vector-borne diseases. In addition, investigating the effectiveness of MeSA in other cereal crops, such as wheat and maize, could expand its applicability and contribute to the development of broader integrated pest and disease management strategies. Studies have also shown that MeSA functions as an airborne signal, inducing systemic acquired resistance and enhancing viral resistance in neighboring plants under laboratory conditions (36, 45, 61). Further field experiments could provide key insights into the potential of MeSA in airborne defense, further expanding its role in sustainable agricultural practices.

OsMYC2 is a key transcription factor that reprograms gene expression in response to JA, enhancing plant resistance against insect herbivores (6, 39, 43). It has also been shown to positively regulate rice resistance to RSV (41). Our results reveal that OsMYC2 activates OsBSMT1 expression, thereby promoting MeSA biosynthesis and acting as a regulatory node integrating plant responses to insect threats. Moreover, this work shows that rice viruses, including RSV, target the OsMYC2-OsBSMT1 pathway, suppressing indirect defenses against insect vectors, ultimately facilitating virus transmission. Previous studies have demonstrated that RSV NS2 and other viral suppressors inhibit OsMYC2 activity by promoting the degradation of SLENDER RICE 1, a positive regulator of OsMYC2 (44). Beyond rice viruses, several dicot-infecting viruses also manipulate MYC2 to suppress host defenses. For example, proteins from tomato yellow leaf curl China virus and tomato spotted wilt orthotospovirus interfere with MYC2-regulated terpene biosynthesis, thereby weakening direct defenses against vector insects and enhancing viral transmission (46, 62). These findings are consistent with the hypothesis that MYC2-BSMT1 suppression is a conserved viral strategy across plant species. However, while rice dwarf virus infection reduces nuclear OsMYC2 accumulation, it does not down-regulate OsBSMT1 (fig. S11D), suggesting the presence of compensatory mechanisms that sustain MeSA biosynthesis.

We conclude that rice viruses suppress MeSA biosynthesis by targeting OsMYC2, thereby weakening parasitoid-mediated indirect defenses and promoting vector survival. These findings suggest that MeSA-based strategies could help mitigate virus-induced suppression of natural enemies. Future studies should investigate whether similar viral suppression mechanisms operate in other crop-virus-vector systems and assess the feasibility of large-scale MeSA applications for crop protection. Understanding how viruses manipulate plant defense networks will be critical for developing sustainable strategies to control vector-borne plant diseases.

MATERIALS AND METHODS

Plant materials and insects

Two rice cultivars, Oryza sativa L. japonica cultivar NPB and Kitaake, were used in this study. The osbsmt1 mutant, OsBSMT1ox, and pOsBSMT1::GUS transgenic lines were generated in the Kitaake background. The osmyc2 mutant and OsMYC2ox lines in the Kitaake background were previously described (41). NS2ox plants in the NPB background were generated as described in an earlier study (44). All rice plants were cultivated in a glasshouse under controlled conditions (28° to 30°C, 12-hour light/12-hour dark photoperiod).

The insect vectors included L. striatellus (Fallén) (SBPH), S. furcifera (Horváth) (WBPH), and N. lugens (Stål) (BPH), all from the family Delphacidae, as well as N. cincticeps (Uhler) (GRL) from the family Cicadellidae. Insects were maintained on rice seedlings in glass incubators at 25°C following established protocols (63–65). The parasitoid A. nilaparvatae (Pang et Wang) was collected from field populations in Jurong, Jiangsu Province, China, and reared on L. striatellus eggs under laboratory conditions.

Plasmid construction

For rice plant transformation, the coding sequence (CDS) of OsBSMT1 was amplified from rice cDNA, digested with Xma I and Hind III, and ligated into the pHZ104::Ubi:3xHA vector using T4 DNA ligase to generate the pUbi::OsBSMT1-3xHA construct. The OsBSMT1 promoter region (2.1 kb upstream of the start codon) was amplified from rice genomic DNA, digested with Hind III and Nco I, and inserted into the pCambia1301::GUS vector (66) to generate the pOsBSMT1::GUS construct. A CRISPR-Cas9 construct targeting OsBSMT1 was assembled as previously described (67), using the guide RNA (GTTAGCCATGAAGGTAGAGC) to generate the Cas9-OsBSMT1 construct.

For protoplast transformation, the CDS of OsMYC2 was amplified from rice cDNA and cloned into the pENTR vector using the pENTR/SD/D-TOPO cloning kit (Invitrogen, K242020), followed by recombination into the pEarlyGate102 vector using Gateway LR Clonase II enzyme mix (Invitrogen, 11791020) to generate the p35S::OsMYC2-CFP-HA construct. The p35S::GFP-HA construct was described previously (68).

For yeast two-hybrid assays, the CDS of OsMYC2 was inserted into the pGADT7 vector using the ClonExpress Ultra One Step Cloning Kit (Vazyme, C115) to generate AD-OsMYC2. CDSs of RSV proteins (NS2, NS3, SP, NSvc2, CP, and NSvc4) were amplified from RSV-infected rice cDNA and cloned into the pGBKT7 vector to generate BD-NS2, BD-NS3, BD-SP, BD-NSvc2, BD-CP, and BD-NSvc4, respectively.

For coimmunoprecipitation and bimolecular fluorescence complementation (BiFC) assays, the CDS of NS2 was cloned into the pGD-dsRed2 vector using the ClonExpress Ultra One Step Cloning Kit to generate p35S::RFP-NS2 and also into the pENTR vector to produce pENTR-NS2. pENTR-OsMYC2 and pENTR-NS2 constructs were recombined into pSite-nYFP or pSite-cYFP using Gateway LR Clonase II to generate OsMYC2-nYFP and NS2-cYFP, respectively. The GUS-nYFP and GUS-cYFP constructs were described previously (69) and used as negative controls.

For prokaryotic expression, codon-optimized OsMYC2 was cloned into pMAL-c2x to generate pMAL-OsMYC2 for maltose binding protein (MBP)-tagged protein expression. The pET28a-NS2 construct for expressing His-tagged NS2 protein was described previously (70). Primer sequences used for vector construction are listed in table S10.

Transgenic rice generation

The recombinant plasmids pUbi::OsBSMT1-3xHA, pOsBSMT1::GUS, and Cas9-OsBSMT1 were introduced into Kitaake rice calli via Agrobacterium-mediated transformation to generate OsBSMT1ox, pOsBSMT1::GUS, and osbsmt1 knockout lines, respectively. Hygromycin selection and regeneration protocols were used to obtain T₀ transgenic plants. The OsBSMT1ox and pOsBSMT1::GUS lines were verified by qRT-PCR and Western blot analysis. Transgenic plants from the T₂ generation and beyond were used for subsequent experiments. osbsmt1 knockout lines were confirmed by PCR, followed by Sanger sequencing, and homozygous osbsmt1 plants lacking the Cas9 transgene were selected for downstream analyses.

Insect infestation

To obtain SBPH-infested plants, 2-week-old rice seedlings grown in glass bottles covered with nylon mesh were infested with SBPH adults (10 insects per seedling) for 48 hours. For WBPH-, BPH-, or GRL-infested plants, 2-week-old seedlings were infested at the three-leaf stage with 10 nonviruliferous insects per plant for 24 hours. After infestation, insects were removed, and plants were collected for transcript quantification. Each treatment included at least three biological replicates, with three plants per replicate. Uninfested seedlings served as control plants, while insect-infested plants were categorized by vector species: SBPH-infested, WBPH-infested, BPH-infested, or GRL-infested.

Viral inoculation

Viral inoculations were performed as previously described (63–65). RSV can be vertically transmitted to offspring via SBPH (65). To maintain viruliferous SBPH colonies, infection status was routinely verified every 3 months using dot enzyme-linked immunosorbent assay with a monoclonal anti–coat protein (CP) antibody.

For RBSDV, SRBSDV, RRSV, rice grassy stunt virus, and rice dwarf virus, which cannot be vertically transmitted by their respective insect vectors, viruliferous individuals were generated by feeding newly hatched nymphs on virus-infected NPB plants. Briefly, 20 mated female adults of SBPH (for RBSDV), WBPH (for SRBSDV), BPH (for RRSV and rice grassy stunt virus), or GRL (for rice dwarf virus) were allowed to feed on virus-infected plants for 3 to 5 days. These females laid eggs on the same infected plants. After hatching (∼7 days), nymphs were allowed to feed for an additional 3 days on the same infected material and then transferred to healthy NPB plants for more than 9 days to complete the virus circulation period. These insects were then used for subsequent inoculations.

For inoculation, 1-week-old rice seedlings were exposed to viruliferous or nonviruliferous insects for 3 days under confinement. Specifically, SBPH were used for RSV and RBSDV, WBPH for SRBSDV, BPH for RRSV and rice grassy stunt virus, and GRL for rice dwarf virus. After the feeding period, insects were removed, and plants were maintained under glasshouse conditions (28° to 30°C). RSV symptoms appeared ~2 weeks postinoculation, while symptoms of other viruses appeared after 4 weeks. At this stage, plants were harvested for gene expression, volatile profiling, and behavioral assays.

As shown in fig. S2A, plants inoculated with nonviruliferous insects were designated as mock-inoculated, while those exposed to viruliferous insects were classified as virus-infected (e.g., RSV-infected). These plants were subsequently infested with nonviruliferous vectors for 48 hours and were referred to as SBPH-infested or RSV + SBPH–treated plants, respectively (similarly applied to other virus-vector combinations).

Field-based evaluation of MeSA effects on insect populations

Preparation of the slow-release device

The average release rate of MeSA per rice plant following insect herbivory was measured at 1.15 ng/hour. With an average of 60 rice plants (approximately 40 to 80) cultivated per square meter, the total emission was estimated at 69 ng/hour. Over a 10-week period, this corresponds to a cumulative emission of ~1.45 mg of MeSA. To mimic this emission rate, a MeSA slow-release solution was prepared by adding 5.87 g (5 ml) of pure MeSA (Sigma-Aldrich, M6752) into 40 ml of liquid paraffin. Then, 20 g of porous resin particles were placed into 4 cm–by–6 cm nonwoven fabric bags, and 2 ml of the MeSA-paraffin solution was infused into each bag, yielding a total MeSA content of 0.23 g per bag. Control bags were prepared identically using 2 ml of liquid paraffin without MeSA. All bags were inserted into vented plastic spheres (diameter, 4 cm) to create the final slow-release devices. At least 20 MeSA-loaded and 20 control spheres were prepared 1 day before each deployment.

Deployment of slow-release devices and insect survey

The experiment was conducted in an organic ecological rice field in Jurong City, Jiangsu Province, where no pesticides were applied. As shown in Fig. 1A, the field located downwind was designated as the MeSA-release area, while the upwind field served as the control area. Each field contained five treatment plots, with four corner points assigned for slow-release device placement and a central point designated for insect monitoring.

At the end of August 2024, during the rice tillering stage, yellow sticky traps (20 cm by 15 cm) were installed at each survey point. Two weeks later, insect populations were sampled using a sweep net (50 sweeps over 5 m² per point). Sampling was conducted simultaneously by two researchers approaching from opposite directions to ensure uniformity. After sweep netting, the sticky traps were collected and replaced, and the slow-release spheres were deployed at designated positions (blue dots for MeSA and gray dots for control, as in Fig. 1A).

The height of the release devices was adjusted to remain 30 cm above the average height of the rice leaf tips, accounting for plant growth stage. Sweep netting, sticky trap replacement, and release device maintenance were repeated every 2 weeks. All collected insects and traps were brought to the laboratory for taxonomic identification and quantification by family.

Insects were counted by family for each sampling point and time, and family proportions were calculated. Delphacidae, Cicadellidae, Pyralidae, Thripidae, Curculionidae, Pentatomidae, and Aphidoidea were classified as herbivorous insects. Araneae, Coccinellidae, Gerridae, Coenagrionidae, and Apocrita were classified as natural enemies (as listed in table S1). Differences in insect abundance and composition between MeSA-treated and control fields were assessed using the Wilcoxon rank sum test.

Measurement of A. nilaparvatae parasitism rates on virus-infected rice plants in the glasshouse

The parasitism rate of A. nilaparvatae on SBPH eggs was assessed on both RSV-infected and mock-inoculated rice plants. Briefly, 15 rice plants were individually infested with 60 viruliferous or nonviruliferous SBPH for 48 hours. After complete removal of the insects, the plants were returned to glasshouse.

Two weeks later, 15 RSV-infected and 15 mock-inoculated plants were positioned on opposite sides of a glass chamber (1.5 m in length by 0.8 m in width by 1.5 m in height), separated by a central baffle (1.2 m in height). On each side, 150 mated SBPH adults were released and allowed to oviposit for 48 hours. The middle baffle was designed to be 5 cm shorter than the surrounding structure, permitting A. nilaparvatae to move freely between compartments while preventing planthopper migration.

After the 48-hour feeding period, 100 female A. nilaparvatae were released onto the middle baffle and allowed to disperse freely in the chambers. Five days later, both SBPH adults and parasitoids were removed. The rice plants were then dissected under a stereomicroscope to examine the leaf sheath area and quantify egg parasitism.

Eggs displaying a red or orange interior were classified as parasitized, while those with normal appearance were considered unparasitized. Total egg number was calculated as the sum of parasitized and unparasitized eggs, and parasitism rate was defined as the percentage of parasitized eggs among the total.

To compare egg number and parasitism rates for other virus-vector systems, similar assays were conducted using WBPH on SRBSDV-infected plants and BPH on RRSV-infected plants. In each case, virus-infected and mock-inoculated plants were used in parallel. Statistical significance of parasitism differences was evaluated using the Wilcoxon rank sum test.

Measurement of A. nilaparvatae parasitism rates of planthopper eggs on virus-infected rice plants in the field

To evaluate the impact of RSV infection on A. nilaparvatae parasitism of SBPH eggs under field conditions, 32 RSV-infected and 32 mock-inoculated 6-week-old rice plants were each infested with 10 gravid SBPH females per plant for 3 days in a glasshouse. Following insect removal, the plants were transplanted into a rice field in Jurong County, Jiangsu Province, China.

The 32 RSV-infected and 32 mock-inoculated rice plants were arranged in straight lines across the field, with an average of 16 plants per line. RSV-infected and mock-inoculated plants were alternately arranged within each line, spaced more than 3 m apart, and adjacent lines were separated by more than 10 m.

Four days after transplantation, plants were collected and examined for parasitized and unparasitized eggs. Total egg number and parasitism rate were calculated. To reduce potential outlier effects, the three highest and three lowest parasitism values in each group were excluded; the remaining 26 samples per group were used for statistical analysis.

To assess the role of MeSA in modulating parasitism rates, the same experimental procedure was followed, except that 32 RSV-infected and 32 mock-inoculated plants were baited with 50 μg of MeSA [dissolved in lanolin at a final concentration of 50 parts per million (ppm)], while the 64 control plants were treated with lanolin only. A 1.5-ml Eppendorf tube containing MeSA or lanolin was suspended on each plant. The measured volatilization rate of MeSA from the tubes ranged from 3.94 to 8.87 ng/hour over 4 days, with an average rate of 6.41 ng/hour, approximately 3.5 times the natural emission rate from planthopper-infested plants. Four days posttransplantation, total egg counts and parasitism rates were determined. To compare A. nilaparvatae parasitism of WBPH and BPH eggs between SRBSDV- or RRSV-infected and mock-inoculated rice plants, similar field experiments were performed using the corresponding viruses and insect vectors.

Y-tube olfactometer bioassay

Olfactometer assays were conducted as previously described (21) to evaluate the behavioral responses of A. nilaparvatae to different volatiles or rice plant treatments. A Y-tube olfactometer with two 10-cm arms positioned at a 75° angle and a 10-cm stem (1-cm internal diameter) was used. All assays were performed between 10:00 and 18:00 at a controlled room temperature of 25° ± 1°C.

The olfactometer system consisted of an atmospheric sampling pump (Nanjing Shengli, SELI-5215), two odor bottles, a Y-tube, a flowmeter, a humidification bottle, and an activated carbon filter, all connected using Teflon tubing. Filtered and humidified air was passed through the system, with airflow in both Y-tube arms maintained at 200 ml/min.

To assess the preference of A. nilaparvatae for MeSA, a MeSA standard sample in n-hexane at a concentration of 100 ng/μl. As a control, n-hexane alone was used. A 1 cm–by–2 cm piece of filter paper loaded with 1.0 μg of MeSA or n-hexane was placed in the respective odor bottle. Filter papers were replaced after every two wasps.

To test wasp preference between differently treated rice plants, eight plants were placed into each odor bottle as a single odor source. The positions of the odor bottles were switched every five individuals to control for positional bias. A fresh Y-tube was used every 10 wasps. Used Y-tubes were cleaned with anhydrous ethanol and distilled water and then dried at 180°C for 3 hours before reuse.

Female parasitoid wasps that entered in the stem of the Y-tube and moved more than 5 cm within 10 min into either arm were counted as a response. For each combination of odor source, the response of 30 to 60 female wasps was recorded. A chi-square test was used to determine the significance of differences in parasitoid wasp preference between volatiles. Wasps that did not make a choice were excluded from the analysis. The standard error of the volatile preference of parasitoid wasps was calculated as √p(1 − p)/n (71). n represents the number of recorded wasps, and p is the proportion of wasps that chose a particular odor source.

H-tube olfactometer bioassay

The behavioral responses of SBPH to different volatiles or rice plant treatments were assessed using an H-tube olfactometer, following previously described methods (72). Assays were conducted between 10:00 and 18:00 at a room temperature of 25° ± 1°C (Fig. 1H and fig. S7B).

For plant preference tests, an H-tube olfactometer with a 10-cm internal diameter was used. Treated rice plants were placed at each end of the vertical arms of the H-tube, and the ends were sealed with Parafilm. Ten adult SBPH individuals were introduced into the center of the horizontal connecting tube through a 1-cm² opening. After 30 min, the number of insects on each side of the tube was recorded. Individuals that moved less than 5 cm were classified as nonresponsive and excluded from the analysis. The experiment was repeated five times at hourly intervals. To eliminate positional bias, the positions of the plants at both ends were alternated after each replicate.

For volatile preference assays, an H-tube olfactometer with a 3-cm internal diameter was used. Two 2.0-ml vials containing different chemical treatments were placed at the ends of the vertical arms, sealed with Parafilm. Ten SBPH individuals were released into the middle of the horizontal tube through a 1-cm² opening. The experimental procedures were otherwise identical to those of the plant preference assays.

Collection and analysis of rice plant volatiles

Rice volatiles were collected and analyzed as previously described (73). Plant material was wrapped and sealed below the stem with polyethylene gloves and enclosed in a cylindrical glass bottle (2,355 ml). The bottle was sealed, and an SPME fiber (Supelco) was inserted for volatile collection over a 3-hour period (11:00 to 14:00) in a climate chamber maintained at 28°C.

Volatile compounds were analyzed using GC-MS (Agilent, 7890A). Samples were injected at 230°C in splitless mode onto an HP-5MS capillary column (Agilent, 19091S-433) using the following temperature program: initial hold at 40°C for 3 min, followed by a ramp to 230°C at 6°C/min. Electron impact ionization at 70 eV was used to ionize eluted compounds, and mass spectra (mass/charge ratio) were scanned using a quadrupole MS over a range of 35 to 320 u. Retention times were compared with those of authentic standards, and mass spectra were matched to the NIST library for compound identification.

Measurement of volatile MeSA

Volatile MeSA in rice samples was analyzed by GC-MS as described above. Quantification was performed using an external standard curve constructed with MeSA standard solutions ranging from 0 to 1000 ng/μl.

Measurement of rice SA

SA content in rice was measured following sample homogenization in liquid nitrogen. SA quantification was conducted at the Kunming Institute of Botany, Chinese Academy of Sciences (Kunming, China). SA was extracted using solid-phase extraction, and the eluates were analyzed by ultraperformance liquid chromatography–tandem MS according to previously described protocols (74). Statistical significance was assessed using a Student’s t test.

Transient expression in rice protoplasts

Transient expression assays in rice protoplasts were conducted as previously described (75). Protoplasts were isolated from 10-day-old rice seedlings. Seedlings were cut into ~0.5-mm strips with sharp blades and incubated in enzyme solution [0.6 M mannitol, 10 mM MES (pH 5.7), 1.5% cellulase R10, 0.75% macerozyme, 0.1% bovine serum albumin (BSA), and 3.4 mM CaCl₂] at 28°C for 4 to 5 hours in the dark with gentle shaking (60 to 80 rpm). An equal volume of W5 solution [154 mM NaCl, 125 mM CaCl₂, 5 mM KCl, and 2 mM MES (pH 5.7)] was then added to release the protoplasts. The suspension was filtered through 40- and 70-μm nylon mesh, and the flow-through was centrifuged at 100g for 10 min to pellet the protoplasts. The pellet was washed with W5 solution and resuspended in MMg solution [0.6 M mannitol, 15 mM MgCl₂, and 4 mM MES (pH 5.7)] to a final concentration of 2 × 10⁶ cells/ml.

For transfection, 10 μg of plasmid DNA was mixed with 40% PEG solution [0.6 M mannitol, 100 mM CaCl₂, and 40% (v/v) polyethylene glycol, molecular weight 4000] and incubated with the protoplasts. After transfection, cells were washed and incubated overnight in W5 solution.

Phylogenetic analysis of OsMES genes

Twelve rice MES (OsMES) homologs corresponding to five Arabidopsis MES genes known to exhibit MeSA esterase activity (76) were identified through BLAST analysis. These rice genes include OsMES1 (LOC_Os01g25360), OsMES2 (LOC_Os01g37630), OsMES3 (LOC_Os01g37650), OsMES4 (LOC_Os01g57770), OsMES5 (LOC_Os01g70830), OsMES6 (LOC_Os01g70840), OsMES7 (LOC_Os01g70850), OsMES8 (LOC_Os01g70860), OsMES9 (LOC_Os05g30760), OsMES10 (LOC_Os07g41230), OsMES11 (LOC_Os08g01850), and OsMES12 (LOC_Os12g02510).

Protein sequences of rice and Arabidopsis MES genes were retrieved from UniProt. Multiple sequence alignment was performed using MUSCLE v3.8.1551. Phylogenetic analysis was conducted with IQ-TREE v2.1.4-beta using the JTT + I + G4 substitution model. The resulting tree was visualized with FigTree v1.4.4 (77, 78), and bootstrap values were displayed at each node to indicate branch support.

ChIP-qPCR analysis

ChIP assays were conducted as previously described (68). Briefly, 2 g of 4-week-old rice seedlings (mock-inoculated or RSV-infected) were ground in liquid nitrogen and cross-linked with 1% (v/v) formaldehyde. The reaction was quenched with glycine, and nuclei were isolated. Genomic DNA was fragmented by sonication, and chromatin was immunoprecipitated using a Pol II antibody (Abcam, ab193468). Eluted DNA (100 μl Tris-EDTA buffer) was collected, and 2 μl of each eluate was subjected to quantitative PCR. Rabbit immunoglobulin G was used as a negative control, and input DNA (preimmunoprecipitation) served as the input control.

To assess the interaction between OsMYC2 and OsBSMT1 promoter regions, ChIP was performed using 2-week-old Kitaake and p35S::MYC2-FLAG transgenic rice plants. Chromatin was immunoprecipitated with anti-FLAG beads (Sigma-Aldrich, M8823). Primer sequences used for ChIP-qPCR are listed in table S10.

RNA extraction and qRT-PCR analysis

RNA extraction and qRT-PCR were performed as described previously (68). Total RNA was extracted from rice tissue using TRIzol reagent (Ambion, 15596026), following the manufacturer’s protocol. One microgram of RNA was reverse transcribed into cDNA using the PrimeScript RT reagent Kit with gDNA Eraser (Takara, RR047A) to remove genomic DNA.

qRT-PCR was performed using the TB Green Premix Ex Taq (Tli RNase H Plus) kit (Takara, RR420A) on an ABI StepOnePlus RT-PCR system. Gene expression levels were calculated using the 2^−ΔΔCt method. Statistical significance was assessed by Student’s t test. All primers used in the qRT-PCR analysis are listed in table S10.

Protein expression and purification

Plasmid pMAL-c2x, pMAL-OsMYC2, and pET28a-NS2 were individually transformed into E. coli BL21 (DE3) cells (TransGen Biotech, CD601). Protein expression was induced with isopropyl-β-d-thiogalactopyranoside.

For purification of MBP and MBP-OsMYC2 fusion proteins, bacterial cells were harvested and resuspended in lysis buffer [20 mM tris-HCl (pH 7.5), 200 mM NaCl, 1 mM EDTA, and one cOmplete EDTA-free protease inhibitor cocktail tablet per 50 ml (Roche, 4693132001)]. The suspension was sonicated and centrifuged. The supernatant was applied to amylose resin (New England Biolabs), and proteins were eluted with 10 mM maltose.

For purification of His-tagged NS2 protein, transformed E. coli cells were resuspended in lysis buffer [50 mM phosphate buffer (pH 8.0), 300 mM NaCl, and one cOmplete EDTA-free protease inhibitor cocktail tablet per 50 ml], followed by sonication. After centrifugation, the supernatant was loaded onto Ni–nitrilotriacetic acid resin (QIAGEN), and bound proteins were eluted with 200 mM imidazole. The purity of all recombinant proteins was assessed by SDS–polyacrylamide gel electrophoresis (PAGE), followed by Coomassie Brilliant Blue staining.

Pull-down assay

Purified MBP or MBP-OsMYC2 proteins were incubated with amylose resin beads at 4°C for 1 hour with rotation. The beads were washed three times with wash buffer [160 mM NaCl, 20 mM tris-HCl (pH 7.5), and 0.05% NP-40].

Recombinant NS2 protein was then added in 600 μl of wash buffer and incubated with the beads at 4°C for 2 hours with gentle rotation. After six washes, bound proteins were eluted in 5× SDS sample buffer, boiled for 5 min, resolved by 12% SDS-PAGE, and visualized by Coomassie Brilliant Blue staining.

Electrophoretic mobility shift assay

The interaction between OsMYC2 protein and the OsBSMT1 promoter was examined using a chemiluminescent EMSA kit (Thermo Fisher Scientific, 20148) following the manufacturer’s instructions. Purified MBP or MBP-OsMYC2 proteins were incubated with biotinylated or unlabeled double-stranded DNA probes in binding buffer at room temperature for 20 min. Oligonucleotide probes containing G-box motifs were synthesized and biotin-labeled (BGI, Beijing, China). The binding mixtures were resolved on 6% native PAGE in 0.5× tris-borate EDTA buffer, transferred onto nylon membranes, and cross-linked. Biotin-labeled DNA was visualized by chemiluminescence detection. Probe sequences are listed in table S10.

Protein extraction and Western blot analysis

Rice tissues (0.1 g) were ground to a fine powder in liquid nitrogen and suspended in 500 μl of protein extraction buffer [20 mM tris-HCl (pH 7.5), 200 mM NaCl, 5 mM MgCl₂, 5 mM dithiothreitol (DTT), 0.03% Tween 20, and one tablet of cOmplete EDTA-free protease inhibitor cocktail (Roche, 4693132001)]. Samples were incubated on ice for 30 min and centrifuged at 12,000 rpm for 10 min at 4°C. The supernatants were mixed with 5× SDS-PAGE loading buffer [250 mM tris-HCl (pH 6.8), 250 mM DTT, 50% glycerol, 10% SDS, and 0.5% bromophenol blue] and boiled for 5 min at 95°C.

Proteins were separated by SDS-PAGE and transferred to polyvinylidene difluoride membranes. Immunoblotting was performed using the following antibodies: anti–hemagglutinin (HA) (1:2000; EASYBIO, BE7002), anti-H3 (1:2000; EASYBIO, BE3015), anti-TUBULIN (1:5000; EASYBIO, BE0031), anti-RSV-CP (1:3000; GenScript, CKTMPEGKKKERGLT), anti–red fluorescent protein (RFP) (1:2000; EASYBIO, BE3307), anti-FLAG (1:2000; Sigma-Aldrich, F7425), and anti-OsMYC2 (41), depending on experiment requirements. Signal intensities were quantified using ImageJ software.

Coimmunoprecipitation analysis in N. benthamiana

To examine the interaction between OsMYC2 and NS2 proteins, N. benthamiana leaves were coinfiltrated with Agrobacterium tumefaciens strains carrying pEarleyGate102-OsMYC2-CFP-HA and pGD-dsRed2-NS2, or pEarleyGate102-HA-GFP and pGD-dsRed2-NS2 as a control.

Approximately 2 g of infiltrated leaves were ground into fine powder in liquid nitrogen and resuspended in 10 ml of lysis buffer [20 mM tris-HCl (pH 7.5), 200 mM NaCl, 0.25 mM MgCl₂, 0.5% Tween 20, 5 mM DTT, 50 μM MG132 (ElfSea, S2619), and 1× cOmplete EDTA-free protease inhibitor cocktail]. Samples were incubated at 4°C for 30 min and then centrifuged at 5000g for 10 min at 4°C. The supernatant was incubated with 30 μl of HA beads (LABLEAD, HNM-25-1000) for 2 hours at 4°C. Beads were washed four times with wash buffer [20 mM tris-HCl (pH 7.5), 200 mM NaCl, 0.25 mM MgCl₂, 0.3% Triton X-100, 5 mM DTT, 50 μM MG132, and 0.5× protease inhibitor cocktail]. Bound proteins were eluted by boiling the beads in 1× SDS sample buffer and analyzed by SDS-PAGE and Western blotting to detect copurification with HA–green fluorescent protein (GFP) or OsMYC2–cyan fluorescent protein (CFP)–HA.

Yeast two-hybrid assay

To test interactions between OsMYC2 and RSV-encoded proteins, pGAD-OsMYC2 were cotransformed with pGBK-NS2, pGBK-NS3, pGBK-SP, pGBK-NSvc2, pGBK-CP, and pGBK-NSvc4, respectively, into the yeast strain Y2H Gold Chemically Competent Cell (ZOMANBIO, ZC1602). Transformants were plated on SD/-Leu/-Trp agar (ZOMANBIO, ZC1808) and incubated at 29°C for 48 hours to verify successful transformation. Colonies were then diluted 1:10, 1:100, and 1:1000 in double-distilled H₂O and spotted on SD/-Ade/-His/-Leu/-Trp agar (ZOMANBIO, ZC1793). Growth was assessed after incubation at 27°C for 48 hours to evaluate protein-protein interactions.

Bimolecular fluorescence complementation analysis

To assess associations of OsMYC2 with NS2 proteins, A. tumefaciens (GV3101) carrying pSITE-OsMYC2-nYFP and pSITE-NS2-cYFP were coinfiltrated into N. benthamiana leaves. A. tumefaciens containing pSITE-GUS-cYFP or pSITE-GUS-nYFP was used as negative control. Fluorescence signals were visualized 3 days postinfiltration using a Zeiss LSM-710 confocal microscope.

Nuclear-cytoplasmic fractionation assay

Nuclear-cytoplasmic separation was conducted as described previously (79). Briefly, 0.5 g of aerial tissue from virus-infected or control rice plants was ground into fine powder in liquid nitrogen and homogenized in 5 ml of isolation buffer [20 mM tris-HCl (pH 7.4), 5 mM MgCl₂, 25% glycerol, 10 mM DTT, 100 μM MG132, and two tablets per 50 ml of cOmplete EDTA-free protease inhibitor cocktail] using 50-ml centrifuge tubes.

The homogenate was filtered through double-layered Miracloth, and the filtrate was centrifuged at 1500g for 3 min at 4°C. The supernatant (cytoplasmic fraction) was subjected to further centrifugation at 12,000g for 5 min at 4°C to remove debris. The pellet (nuclear fraction) was washed three times with 1 ml of isolation buffer, followed by 1500g centrifugation for 1 min at 4°C after each wash. H3 and tubulin proteins were probed with anti-H3 or anti-tubulin antibodies as nuclear and cytoplasmic loading control, respectively.

Histochemical GUS staining

GUS staining was performed as previously described (66). Tissues from pOsBSMT1::GUS transgenic rice plants were incubated in sodium phosphate buffer containing X-Gluc (Coolaber, SL7160) at 37°C in the dark overnight. After decolorization with 70% ethanol, tissues were photographed by camera.

Immunohistochemistry analysis

Tender leaves from mock-inoculated and RSV-infected rice were fixed in Formalin-Aceto-Alcohol solution and vacuum-infiltrated on ice for 30 min. Tissues were dehydrated through an ethanol gradient (30 min each at 50, 70, 90, 96, and 100%), followed by embedding in Steedman’s wax (polyethylene glycol distearate:1-hexadecanol, 9:1) using a wax-ethanol gradient (50% wax overnight and then 2 hours each in 75, 100, 100, and 100% wax).

Sections (8 μm) were obtained using a Leica RM2255 microtome and rehydrated through an ethanol gradient (10 min each at 100, 96, 90, 70, 50, 25, and 0%). Sections were blocked with 2% BSA in phosphate-buffered saline (PBS) for 2 hours and incubated with anti-FLAG antibody (1:200; Sigma-Aldrich, F3165) for 2 hours.

Following three washes in PBS with 0.1% Tween 20, sections were incubated with Alexa Fluor 488–conjugated goat anti-mouse secondary antibody (1:200; Thermo Fisher Scientific, A11001) for 1 hour. After additional PBS-T washes, 4w,6-diamidino-2-phenylindole (DAPI; 1 μg/ml) was applied for 5 min, followed by three rinses in water. Imaging was conducted using confocal microscope.

Image intensity analysis

Western blot quantification was performed using ImageJ (v1.8.0; National Institutes of Health). Background subtraction and integrated density calculations were conducted using default settings. Fluorescence intensity was quantified using ZEN Blue (v3.7; Carl Zeiss AG) with automated thresholding algorithms.

Quantification and statistical analysis

Statistical comparisons of MeSA field trials and A. nilaparvatae parasitism rates were conducted using the Wilcoxon rank sum test. Y-tube olfactometer preferences were analyzed using the χ² test. t tests were used for Western blot quantification, SA and volatile measurements, qPCR data, and fluorescence intensity analysis. Significance levels were defined as *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, P > 0.05. Full statistical details are provided in figure legends.

Supplementary Materials

This PDF file includes:

Figs. S1 to S20

Tables S1 to S10