Silencing an E3 Ubiquitin Ligase Gene OsJMJ715 Enhances the Resistance of Rice to a Piercing-Sucking Herbivore by Activating ABA and JA Signaling Pathways

The RING-type E3 ubiquitin ligases play an important role in plant growth, development, and defense responses to abiotic stresses and pathogens. However, their roles in the resistance of plants to herbivorous insects remain largely unknown. In this study, we isolated the rice gene OsJMJ715, which encodes a RING-domain containing protein, and investigated its role in rice resistance to brown planthopper (BPH, Nilaparvata lugens). OsJMJ715 is a nucleus-localized E3 ligase whose mRNA levels were upregulated by the infestation of gravid BPH females, mechanical wounding, and treatment with JA or ABA. Silencing OsJMJ715 enhanced BPH-elicited levels of ABA, JA, and JA-Ile as well as the amount of callose deposition in plants, which in turn increased the resistance of rice to BPH by reducing the feeding of BPH and the hatching rate of BPH eggs. These findings suggest that OsJMJ715 negative regulates the BPH-induced biosynthesis of ABA, JA, and JA-Ile and that BPH benefits by enhancing the expression of OsJMJ715.


Introduction
To defend themselves from herbivorous insects, plants have developed sophisticated defensive mechanisms known as constitutive and inducible defenses [1,2]. The inducible defense starts with the perception of herbivore-associated molecular patterns and/or damage-associated molecular patterns, followed by the activation of an array of early signaling events, such as the increase in levels of cytosolic calcium ion, burst of reactive oxygen species (ROS), and activation of mitogen-activated protein kinase (MAPK) cascades. These early signaling events elicit defense-related signaling pathways mediated by phytohormones, which in turn lead to a global reconfiguration of transcriptome, proteome, and metabolome, thereby enhancing direct and/or indirect defenses of plants to herbivores [2]. These phytohormones mainly include jasmonic acid (JA), salicylic acid (SA), and ethylene (ET). Recently, other phytohormones, such as abscisic acid (ABA), gibberellic acids, cytokinins, and auxins, have also been reported to modulate herbivore resistance through synergistically or antagonistically interacting with JA signaling [3]. For instance, Vos et al. [4] found that in Arabidopsis thaliana, the activation of primed JA-regulated defenses in response to secondary herbivore attack is dependent on ABA signaling. In contrast, ABA has been shown to negatively regulate resistance to Hirschmanniella oryzae by inhibiting JA biosynthesis in rice (Oryza sativa) roots [5].
Ubiquitin (Ub) is a polypeptide consisting of 76 amino acids, and its mediated proteolysis extensively exists in eukaryotes. The Ub-targeted proteins are first modified by a cascade of reactions catalyzed mainly by three enzymes involving Ub-activating enzymes (E1), Ub-conjugating enzymes (E2), and Ub ligases (E3); then, the mono-or Figure 1. OsJMJ715 encodes a nucleus-localized RING-type E3 ubiquitin ligase. (A) In demethylation assay of OsJMJ715. Bulk histone was incubated with (+) or without (-(left) or GST::706 dem (right) fusion proteins in the reaction buffer for 2 h at 30 °C and immunoblotting by using anti-H3K9me2, anti-H3K9me3, anti-H3, and anti-GST antib vitro ubiquitination assay of OsJMJ715. Fusion proteins of MBP::715 RING and its m MBP::715 RING (C208S; C211S) were assayed for E3 activity in the presence of E1, E2, and ub Anti-Ub and anti-MBP antibodies were used to detect ubiquitinated proteins and fus respectively. Molecular weight markers (kDa) are shown on the right. (C) Subcellular of OsJMJ715. The EGFP::715 fusion protein and the EGFP protein were expressed in to tiana benthamiana) leaves. Florescence assay was performed using a confocal microsco tographs were taken in UV light, visible light, RFP, or in combination (overlay) (bar = In addition to the JmjC domain, OsJMJ715 harbors a RING-finger doma terminal ( Figure S2), which is known to function as ubiquitin E3 ligases [9 whether OsJMJ715 possesses E3 ligase activity, we expressed the RING OsJMJ715 in Escherichia coli as a fusion with maltose-binding protein (MBP). ence of ubiquitin, E1 and E2, purified MBP::715 RING was able to self-ubiquitina MBP fused with an RING mutant, 715 RING(C208S, C211S) , which was constructed wi tions of Cys-208 and Cys-211 with Ser-208 and Ser-211 in the RING domain, to produce ubiquitinated bands ( Figure 1B). These findings suggest that OsJM ligase activity, and the intact RING domain is required. OsJMJ715 encodes a nucleus-localized RING-type E3 ubiquitin ligase. (A) In vitro histone demethylation assay of OsJMJ715. Bulk histone was incubated with (+) or without (-) GST::715 dem (left) or GST::706 dem (right) fusion proteins in the reaction buffer for 2 h at 30 • C and analyzed by immunoblotting by using anti-H3K9me2, anti-H3K9me3, anti-H3, and anti-GST antibodies. (B) In vitro ubiquitination assay of OsJMJ715. Fusion proteins of MBP::715 RING and its mutant form MBP::715 RING (C208S; C211S) were assayed for E3 activity in the presence of E1, E2, and ubiquitin (Ub). Anti-Ub and anti-MBP antibodies were used to detect ubiquitinated proteins and fusion proteins, respectively. Molecular weight markers (kDa) are shown on the right. (C) Subcellular localization of OsJMJ715. The EGFP::715 fusion protein and the EGFP protein were expressed in tobacco (Nicotiana benthamiana) leaves. Florescence assay was performed using a confocal microscopy. The photographs were taken in UV light, visible light, RFP, or in combination (overlay) (bar = 20 µm).
In addition to the JmjC domain, OsJMJ715 harbors a RING-finger domain at the Nterminal ( Figure S2), which is known to function as ubiquitin E3 ligases [9]. To assess whether OsJMJ715 possesses E3 ligase activity, we expressed the RING domain of OsJMJ715 in Escherichia coli as a fusion with maltose-binding protein (MBP). In the presence of ubiquitin, E1 and E2, purified MBP::715 RING was able to self-ubiquitinate, whereas MBP fused with an RING mutant, 715 RING(C208S, C211S) , which was constructed with substitutions of Cys-208 and Cys-211 with Ser-208 and Ser-211 in the RING domain, was unable to produce ubiquitinated bands ( Figure 1B). These findings suggest that OsJMJ715 has E3 ligase activity, and the intact RING domain is required.
To elucidate the subcellular localization of OsJMJ715, we constructed an OsJMJ715::EGFP fusion gene, driven by a CaMV 35S promoter ( Figure S3), and transiently expressed the construct in Nicotiana benthamiana leaves. The result showed that OsJMJ715 specifically localized to the nucleus ( Figure 1C). We further performed quantitative real-time PCR to analyze the expression profiles of OsJMJ715. As shown in Figure 2, the transcript level of OsJMJ715 increased in response to JA and ABA treatment (24 h and 8-24 h after treatment, respectively), mechanical wounding (8 h after treatment), and BPH infestation (4-24 h), especially at a late stage, after these treatments. The data suggest that OsJMJ715 may help regulate rice-herbivore interactions. specifically localized to the nucleus ( Figure 1C). We further performed quantitative realtime PCR to analyze the expression profiles of OsJMJ715. As shown in Figure 2, the transcript level of OsJMJ715 increased in response to JA and ABA treatment (24 h and 8-24 h after treatment, respectively), mechanical wounding (8 h after treatment), and BPH infestation (4-24 h), especially at a late stage, after these treatments. The data suggest that OsJMJ715 may help regulate rice-herbivore interactions.

Silencing OsJMJ715
To study the function of OsJMJ715 in the herbivore-induced rice defenses, we constructed OsJMJ715-RNAi transgenic plants, and obtained two independent T2 homozygous irJMJ lines (R9 and R10) containing a single insertion ( Figure S4). Transcriptional analysis showed that wounding-induced mRNA levels of OsJMJ715 in R9 and R10 lines were only 47.3% and 42.3% of those in WT plants ( Figure 3A). No obvious differences in growth phenotypes were observed between WT and transgenic lines ( Figure 3B). Moreover, the shoot height, root length, seed size, and mass of R9 and R10 plants were all similar to those of WT plants ( Figure 3C-F).

Silencing OsJMJ715
To study the function of OsJMJ715 in the herbivore-induced rice defenses, we constructed OsJMJ715-RNAi transgenic plants, and obtained two independent T 2 homozygous irJMJ lines (R9 and R10) containing a single insertion ( Figure S4). Transcriptional analysis showed that wounding-induced mRNA levels of OsJMJ715 in R9 and R10 lines were only 47.3% and 42.3% of those in WT plants ( Figure 3A). No obvious differences in growth phenotypes were observed between WT and transgenic lines ( Figure 3B). Moreover, the shoot height, root length, seed size, and mass of R9 and R10 plants were all similar to those of WT plants ( Figure 3C-F).

Silencing OsJMJ715 Enhanced the Resistance of Rice to BPH
To evaluate the effect of silencing OsJMJ715 on the resistance of rice to BPH, we performed a series of bioassays. BPH nymphs fed on irJMJ lines displayed similar survival rates and developmental durations to those fed on WT plants ( Figure 4A,B). However, the amount of honeydew excreted by BPH female adults fed on R9 and R10 lines was significantly decreased, by 29.3% and 38%, respectively, compared to those that fed on WT plants ( Figure 4C). Moreover, the number of eggs laid by female adults that had emerged from irJMJ lines was less than that laid by female adults emerged from WT plants ( Figure 4D). Silencing OsJMJ715 also reduced the hatching rate of BPH eggs: for eggs laid on R9 and R10, the hatching rate was only 55.2% and 66.1%, respectively, of those laid on WT plants ( Figure 4E). These results suggest that silencing OsJMJ715 enhances the resistance of rice to BPH.

Silencing OsJMJ715 Enhanced the Resistance of Rice to BPH
To evaluate the effect of silencing OsJMJ715 on the resistance of rice to BPH, we per formed a series of bioassays. BPH nymphs fed on irJMJ lines displayed similar surviva rates and developmental durations to those fed on WT plants ( Figure 4A,B). However, th amount of honeydew excreted by BPH female adults fed on R9 and R10 lines was signifi cantly decreased, by 29.3% and 38%, respectively, compared to those that fed on WT plants ( Figure 4C). Moreover, the number of eggs laid by female adults that had emerged from irJMJ lines was less than that laid by female adults emerged from WT plants ( Figur  4D). Silencing OsJMJ715 also reduced the hatching rate of BPH eggs: for eggs laid on R and R10, the hatching rate was only 55.2% and 66.1%, respectively, of those laid on WT plants ( Figure 4E). These results suggest that silencing OsJMJ715 enhances the resistanc of rice to BPH.

Silecing OsJMJ715 Enhanced BPH-Induced Levels of JA, JA-Isoleucine, and ABA
To gain further insight into the mechanism underlying OsJMJ715-mediated rice resistance to BPH, we measured levels of JA, JA-isoleucine (JA-Ile), and ABA in WT and irJMJ plants in response to gravid BPH female attack. As previously found [32], levels of JA and JA-Ile were induced quickly and continuously during gravid BPH female infestation ( Figure 5A,B). Silencing OsJMJ715 did not influence the constitutive level of JA but slightly enhanced the constitutive level of JA-Ile ( Figure 5A,B). Infestations by BPH caused the level of both JA and JA-Ile in irJMJ lines to rise, becoming significantly higher than that in WT plant 24 and 12-24 h, respectively, after its onset ( Figure 5A,B). Gravid BPH female infestation also enhanced levels of ABA in both WT and irJMJ plants ( Figure 5C). Silencing OsJMJ715 had no effect on constitutive levels of ABA, but it strengthened and quickened the response of ABA to BPH infestation-compared to that in non-infested plants, the level of ABA in BPH-infested plants began to significantly increase at 4 h after BPH attack in irJMJ lines at 24 h in WT plants; moreover, the BPH-infested level of ABA in irJMJ plants was always significantly higher than that in WT plants at 4-24 h after BPH infestation ( Figure 5C). The data demonstrate that in plants silencing OsJMJ715 enhances the response of JA, JA-Ile, and ABA to BPH infestation.

Silecing OsJMJ715 Enhanced BPH-Induced Levels of JA, JA-Isoleucine, and ABA
To gain further insight into the mechanism underlying OsJMJ715-mediated rice resistance to BPH, we measured levels of JA, JA-isoleucine (JA-Ile), and ABA in WT and irJMJ plants in response to gravid BPH female attack. As previously found [32], levels of JA and JA-Ile were induced quickly and continuously during gravid BPH female infestation ( Figure 5A,B). Silencing OsJMJ715 did not influence the constitutive level of JA but slightly enhanced the constitutive level of JA-Ile ( Figure 5A,B). Infestations by BPH caused the level of both JA and JA-Ile in irJMJ lines to rise, becoming significantly higher than that in WT plant 24 and 12-24 h, respectively, after its onset ( Figure 5A,B). Gravid BPH female infestation also enhanced levels of ABA in both WT and irJMJ plants ( Figure 5C). Silencing OsJMJ715 had no effect on constitutive levels of ABA, but it strengthened and quickened the response of ABA to BPH infestation-compared to that in non-infested

OsJMJ715 Negatively Mediated BPH-Elicited Callose Deposition
Callose deposition in sieve plates is an important defense mechanism of rice plants to prevent BPH feeding [33]. To ask whether OsJMJ715 regulates callose deposition and thereby influences BPH feeding as stated above ( Figure 4C), we investigated callose deposition in rice plants of different genotypes before and after BPH infestation. As previously reported [34,35], little callose was observed in non-infested plants ( Figure 6A-C). Upon BPH attack for 48 h, callose deposition had increased in both WT and irJMJ plants ( Figure 6D-F); however, silencing OsJMJ715 enhanced the deposition of callose-the amount of callose deposition in R9 and R10 was 1.9-and 3.2-fold, respectively, of that in WT plants ( Figure 6G). plants, the level of ABA in BPH-infested plants began to significantly increase at 4 h after BPH attack in irJMJ lines at 24 h in WT plants; moreover, the BPH-infested level of ABA in irJMJ plants was always significantly higher than that in WT plants at 4-24 h after BPH infestation ( Figure 5C). The data demonstrate that in plants silencing OsJMJ715 enhances the response of JA, JA-Ile, and ABA to BPH infestation.

OsJMJ715 Negatively Mediated BPH-Elicited Callose Deposition
Callose deposition in sieve plates is an important defense mechanism of rice plants to prevent BPH feeding [33]. To ask whether OsJMJ715 regulates callose deposition and thereby influences BPH feeding as stated above ( Figure 4C), we investigated callose deposition in rice plants of different genotypes before and after BPH infestation. As previously reported [34,35], little callose was observed in non-infested plants ( Figure 6A-C). Upon BPH attack for 48 h, callose deposition had increased in both WT and irJMJ plants ( Figure 6D-F); however, silencing OsJMJ715 enhanced the deposition of callose-the amount of callose deposition in R9 and R10 was 1.9-and 3.2-fold, respectively, of that in WT plants ( Figure 6G).

Discussion
The RING-type E3 ubiquitin ligases play an important role in plant growth and development as well as in defense responses to abiotic stresses and pathogens. In this study, we found that a rice RING-type E3 ubiquitin ligase, OsJMJ715, plays an important role in regulating the resistance of rice to BPH. Several lines of evidence support this statement. First, OsJMJ715 was induced by BPH infestation and mechanical wounding. Second, silencing OsJMJ715 enhances BPH-elicited accumulation of JA, JA-Ile, ABA, and callose, which subsequently reduces the hatching rate of BPH eggs and BPH feeding. Third, OsJMJ715 specifically localizes to the nucleus and has E3 ligase activity, suggesting that OsJMJ715 might be implicated in the degradation of proteins [36]. These data demonstrate

Discussion
The RING-type E3 ubiquitin ligases play an important role in plant growth and development as well as in defense responses to abiotic stresses and pathogens. In this study, we found that a rice RING-type E3 ubiquitin ligase, OsJMJ715, plays an important role in regulating the resistance of rice to BPH. Several lines of evidence support this statement. First, OsJMJ715 was induced by BPH infestation and mechanical wounding. Second, silencing OsJMJ715 enhances BPH-elicited accumulation of JA, JA-Ile, ABA, and callose, which subsequently reduces the hatching rate of BPH eggs and BPH feeding. Third, OsJMJ715 specifically localizes to the nucleus and has E3 ligase activity, suggesting that OsJMJ715 might be implicated in the degradation of proteins [36]. These data demonstrate that RING-type E3 ubiquitin ligase-mediated proteolysis is involved in regulating the resistance of plants to both pathogens [11,[16][17][18][19][20] and herbivores (this study).
As stated in the introduction, RING-type E3 ubiquitin ligases have been reported to modulate phytohormone biosynthesis and signaling by degrading regulators in their processes. Here, we found that the expression of OsJMJ715 was induced by BPH infestation, and treatment with JA and ABA, whereas silencing OsJMJ715 enhanced BPH-induced levels of JA, JA-Ile, and ABA. This suggests that just as OsEBF1 attenuates OsLOX9 expression by degrading its transcriptional activator OsEIL1, decreasing JA accumulation in rice [37], OsJMJ715 may degrade activators in the biosynthesis of JA, JA-Ile, and ABA when plants were infested by BPH; as such, OsJMJ715 may function as a negative regulator in the biosynthesis of BPH-induced JA, JA-Ile, and ABA by forming a negative feedback loop with the level of JA, JA-Ile, and ABA. Such a negative regulator, like OsNPR1 [38], 9lipoxygenase (Osr9-LOX1) [39], OsWRKY53 [40], and MPK20-5 [41] reported previously in rice, may help ensure appropriate levels of JA, JA-Ile, and ABA in rice plants when infested by BPH, thereby avoiding the autotoxicity that excessive plant defenses may lead to. Further studies should identify the targets of OsJMJ715 under BPH infestation to uncover the mechanism underlying the OsJMJ715-mediated regulation of defensive phytohormone biosynthesis.
We observed that silencing OsJMJ715 significantly reduced the amount of honeydew excreted by BPH and the hatching rate of BPH eggs. Given a significant increase in levels of BPH-induced ABA in irJMJ plants, the above results suggest that ABA-mediated signaling pathway may play an important role in enhancing the resistance of rice to BPH. In addition to its contributions to plant disease resistance by controlling stomatal closure and callose deposition [42,43], the ABA-mediated signaling pathway has also been reported to regulate the resistance of plants to piercing-sucking herbivores, including BPH, by influencing the callose deposition on plant sieve plates, a mechanism by which plants effectively prevent piercing-sucking herbivores from ingesting phloem sap [33,44]. For instance, Zhou et al. [35] found that silencing the ABA hydrolase gene, OsABA8ox3, remarkably increased endogenous levels of ABA and callose deposition in rice, which in turn enhances the resistance of rice to BPH. Moreover, exogenous ABA increases the resistance of rice to BPH by promoting callose deposition, whereas resistant varieties that were treated with fluridone exhibit susceptibility to BPH with reduced callose deposition [45]. In addition to ABA-mediated signaling pathway, JA-mediated signaling pathway also plays an important role in modulating the resistance of rice to BPH by regulating the production of defensive compounds [26,31]. Thus, JA and ABA signaling pathways, both of which were strengthened in irJMJ plants compared to WT plants, might contribute to the enhanced resistance of irJMJ plants to BPH. Some defensive compounds in rice against BPH have been reported. Callose deposition on sieve plates, for example, as stated above, inhibits BPH feeding [34,35,46]. Some flavonoids and phenolamides, such as sakuranetin, p-coumaroyl putrescine, feruloyl putrescine, and cinnamoyl putrescine, also affect the survival and development of BPH [47][48][49]. Although no ovicidal compounds for BPH eggs have been identified, benzyl benzoate in japonica rice has been demonstrated to have a lethal role on the eggs of the white-backed planthopper Sogatella furcifera [50]. Moreover, Zhou et al. [51] recently revealed that feeding on plants treated with benzyl benzoate decreases the fecundity of BPH female adults. We found that silencing OsJMJ715 enhanced the accumulation of callose deposition. Hence, the decrease in BPH feeding on irJMJ plants is at least in part related to the increase in callose deposition. We look forward to investigating whether other compounds are involved in influencing BPH feeding and which compounds are responsible for the decrease in the hatching rate of BPH eggs.
In summary, our results show that OsJMJ715, a RING-type E3 ligase, negatively regulates the biosynthesis of BPH-induced JA, JA-Ile, and ABA. BPH can benefit by inducing the expression of OsJMJ715, which subsequently decreases the level of BPH-induced ABA, JA, and JA-Ile, as well as the resulting defense responses in rice. These findings demonstrate that BPH reduces host plant resistance by regulating a RING-type E3 ligase.

Plant Material and Growth Conditions
The rice variety Xiushui 110 (XS110, japonica) and OsJMJ715-RNAi transgenic lines (using XS110 as the receptor variety) were used for this work. Water-soaked seeds were germinated in an incubator (28 ± 2 • C, 14 h light, 10 h dark). Seven-day-old seedings were transplanted to a hydroponic box (length × width × height: 45 × 30 × 15 cm) containing nutrient solution [52] and grown in a greenhouse (28 ± 2 • C, 14 h light, 10 h dark, 50-60% relative humidity) for 2 weeks, then were transplanted into individual plastic pots (diameter × height: 8 × 10 cm) that were placed in the greenhouse. Seven days later, the plants were used for experiments.
The transgenic tobacco (N. benthamiana) plants expressing the red fluorescent protein RFP-H2B were used for subcellular localization of OsJMJ715 [53]. Seeds were sown in a seedling pot (diameter × height: 5 × 8 cm) filled with a peat/vermiculite/perlite mixture (2:1:1, v/v/v) and grown in the greenhouse as mentioned above. Plants were watered weekly.

Insects
BPH colonies were originally obtained from rice fields in Hangzhou, Zhejiang province, China, and reared in a growth chamber (26 ± 2 • C, 14 h light) on seedlings of a rice variety TN1, which is susceptible to BPH.

Plant Treatments
For the mechanical wounding treatment, leaf sheaths of individual plant shoots were pierced 200 times with a needle (diameter 0.32 mm). Non-manipulated plants were used as controls. For BPH infestation, plants were individually confined in glass cylinders (diameter 4 cm, height 8 cm, with 48 small holes) into which 10 BPH gravid females were placed ( Figure S5). Plants with empty cylinders were used as controls. For JA or ABA treatment, plants were grown in nutrient solution, and JA or ABA (Sigma-Aldrich, St. Louis, MO, USA) was added (first dissolved in a small volume of 70% (v/v) ethanol) into the nutrient solution to give a final concentration of 100 µM. Control plants were grown in nutrient solution without JA and ABA but with an equal volume of 70% ethanol.

Isolation and Characterization of OsJMJ715
The full-length cDNA of OsJMJ715 was obtained by reverse-transcription PCR from total RNA isolated from BPH-infested leaf sheaths of XS110 plants. The primers 715 ORF -F1 and 715 ORF -R1 (Table S1) were designed based on the sequence of OsJMJ715 (accession number: Os03g31594). PCR-amplified fragments were cloned into the pEASY-Blunt Simple Vector (TransGen, Beijing, China) and sequenced (BioSune, Hangzhou, China).

Quantitative Real-Time PCR
For RNA isolation, rice leaf sheaths were ground into fine powders in liquid nitrogen. Total RNA was isolated by the MiniBEST Plant RNA Extraction Kit (TaKaRa, Dalian, China). A total of 1 µg of RNA was reverse-transcribed using the PrimeScript TM RT Master Mix (TaKaRa, Dalian, China). A quantitative real-time PCR assay was performed on a CFX96 TM Real-Time System (Bio-RAD, Hercules, CA, USA) using the TB Green TM Premix EX Taq TM II (Tli RNaseH Plus) (TaKaRa, Dalian, China). The rice actin gene OsACT (accession number: Os03g50885) was used as an internal standard to normalize cDNA concentrations in tested genes; primers used for qRT-PCR are provided in Table S2. Each treatment at each time point was replicated six times.

Subcellular Localization
The full-length open reading frame (ORF) of OsJMJ715 without the stop codon was amplified by primers 715 ORF -F2 and 715 ORF -R2 (Table S1). The PCR product was inserted into p1301-EGFP vector [27] using a pair of restriction endonuclease, SalI and BamHI, to fuse it with EGFP (enhanced green fluorescent protein), yielding the construct OsJMJ715::EGFP ( Figure S3). The construct was then transferred into Agrobacterium tumefaciens strain EHA105 competent cells, yielding a transformation vector. The vector was introduced into N. benthamiana leaves for transient expression as described in Huang et al. [53]. Tobacco leaves were cut into small pieces for fluorescence analysis at 24 h after agroinfiltration. The fluorescent signals were detected under a laser scanning confocal microscope (Zeiss, Oberkochen, Germany) as described by Huang et al. [53].

Generation of Transgenic Plants
A 382bp fragment of OsJMJ715 was amplified by primers 715 RNAi -F1/R1 and 715 RNAi -F2/R2 (Table S1) and inserted into the pCAMBIA-1301 transformation vector to yield an OsJMJ715-RNAi interference construct (p1301-ir715, Figure S4). The construct was transferred into XS110 plants by using A. tumefaciens-mediated transformation. Homozygous lines of T 2 generations were confirmed via GUS staining and PCR screening, as described in Zhou et al. [26]. Finally, two homozygous irJMJ lines (R9 and R10) harboring a single insertion of T-DNA were confirmed by Southern blot analysis were selected for further studies.

JA, JA-Ile, and ABA Measurement
Plants of different genotypes were randomly assigned to BPH and control treatments. For JA, JA-Ile and ABA analysis, outer leaf sheaths of plants were harvested at 0, 4, 8, 12 and 24 h after treatment. Samples (about 150 mg each) were ground in liquid nitrogen, compounds in each sample were extracted with 1 mL of ethyl acetate containing labeled internal standards ( 2 D 2 -JA, 2 D 6 -JA-Ile and 2 D 6 -ABA) and analyzed by a Triple Quad liquid chromatography/mass spectrometry (Agilent Technologies, Santa Clara, USA) following the method described in Lu et al. [27]. Each treatment at each time point was replicated six times.

Callose Measurement
Rice plant shoots of different genotypes were individually confined within glass cylinders into which 10 gravid BPH females were placed. Then, 48 h later, outer leaf sheaths of the plants were collected, cut into 0.3-0.5 cm pieces, and immersed in 10% glycerol. Leaf sheaths were embedded in optimal cutting temperature compound (SAKURA, Torrance, CA, USA) and then cut into 10 µm thick slices via a microtome (Thermo, Waltham, MA, USA) and subsequently placed into 95% ethanol for overnight. Slices were soaked with 1/15 M phosphate buffer (pH 7.0) for 30 min, and then stained with 0.1% (w/v) aniline blue for 60 min. Samples were observed and photographed using a laser scanning confocal microscope (Zeiss) as described by Yang et al. [58]. Callose deposition area was analyzed by Image-Pro Plus (version 6.0). The callose deposition area of each bundle was calculated with 12 replicates for each treatment.

Herbivore Bioassays
To evaluate the effect of OsJMJ715 on the performance of BPH, we individually confined plants of wild-type (WT) and irJMJ lines in cylinders into which 15 newly hatched BPH nymphs were released. The number of surviving nymphs and newly emerged adults on each plant were observed and recorded every day until all the nymphs had become adults. Each treatment was replicated 15 times. The newly emerged adults were then paired (one female with one male) and introduced into a new plant of the same genotype as they emerged. Fourteen days later, the number of BPH eggs in individual plants was counted under a microscope. The experiment was replicated 20 times. We also measured the amount of honeydew excreted by individual newly emerged BPH female adults using a method described in Paguia et al. [59]. Briefly, rice plant shoots were individually confined within transparent plastic cups (diameter 9 cm, height 12 cm) that were placed on plastic Petri dishes lined with filter paper ( Figure S6). Individual newly emerged BPH female adults from different rice genotypes were released into the cups and allowed to feed for 24 h. The filter papers were collected and sprayed with a 0.1% ninhydrin in acetone solution [46] and then dried for 30 min at 60 • C. The area of the ninhydrin-positive deposition was measured with Image-Pro Plus. Each treatment was replicated 10 times. On the basis of these experiments, we calculated the survival rate and developmental duration of BPH nymphs, the fecundity of BPH female adults, and the amount of honeydew excreted by BPH female adults on plants of WT and irJMJ lines.
To investigate the influence of OsJMJ715 on the hatching rate of BPH eggs, we individually infested plants of different genotypes by 10 gravid BPH female adults using the method above. Then, 24 h later, BPHs were removed. The number of newly hatched neonates was recorded every day until no neonates emerged for three days in a row. The number of unhatched eggs was counted under a microscope to calculate the hatching rate of BPH eggs. Each treatment was replicated 15 times.

Data Analysis
Differences between two comparisons were analyzed by Student's t-tests. Multiple comparisons were analyzed by one-way ANOVAs followed by Tukey's post hoc test. The normality of data was tested using the Kolmogorov-Smirnov test (p < 0.05), and the equality of error variances was tested by Levene's test (p < 0.05). In the case of non-normality and/or unequal variances, data were logarithmic-transformed before ANOVAs. All data analysis was carried out with IBM SPSS Statistics, Version 20, International Business Machines Corporation (Armonk, NY, USA).