Antagonistic Plant Defense System Regulated by Phytohormones Assists Interactions Among Vector Insect, Thrips and a Tospovirus

Abstract

The western flower thrips (Frankliniella occidentalis) is a polyphagous herbivore that causes serious damage to many agricultural plants. In addition to causing feeding damage, it is also a vector insect that transmits tospoviruses such as Tomato spotted wilt virus (TSWV). We previously reported that thrips feeding on plants induces a jasmonate (JA)-regulated plant defense, which negatively affects both the performance and preference (i.e. host plant attractiveness) of the thrips. The antagonistic interaction between a JA-regulated plant defense and a salicylic acid (SA)-regulated plant defense is well known. Here we report that TSWV infection allows thrips to feed heavily and multiply on Arabidopsis plants. TSWV infection elevated SA contents and induced SA-regulated gene expression in the plants. On the other hand, TSWV infection decreased the level of JA-regulated gene expression induced by thrips feeding. Importantly, we also demonstrated that thrips significantly preferred TSWV-infected plants to uninfected plants. In JA-insensitive coi1-1 mutants, however, thrips did not show a preference for TSWV-infected plants. In addition, SA application to wild-type plants increased their attractiveness to thrips. Our results suggest the following mechanism: TSWV infection suppresses the anti-herbivore response in plants and attracts its vector, thrips, to virus-infected plants by exploiting the antagonistic SA–JA plant defense systems.

Introduction

Tospovirus is a type species in the family Bunyaviridae that consists of three RNA segments (L, M and S) (German et al. 1992), and it is the only group in the family Bunyaviridae that infects plants. Tomato spotted wilt virus (TSWV), a member of the genus Tospovirus (Murphy 1995), infects many agricultural and horticultural crops and is a typical insect-transmitted plant virus (Sherwood et al. 2001). Interestingly, TSWV is transmitted exclusively by thrips (Jones 2005, Whitfield et al. 2005). Among thrips, the western flower thrips [Frankliniella occidentalis (Pergande)] is the main species that transmits TSWV (Inoue et al. 2004, Whitfield et al. 2005). This thrips is one of the most important insect herbivores, with a wide host range that includes many crops, vegetables, fruits and flowers (Parrella 1995, Reitz 2009). The worldwide emergence of insecticide-resistant thrips makes it difficult to control them (Jensen 2000). Both the feeding damage caused by this thrips and the tospovirus disease it transmits are serious problems.

Most plant viruses are transmitted by insect vectors that include aphids, whiteflies, leafhoppers, planthoppers and thrips (Hogenhout et al. 2008). Because the plant virus cannot move from infected to healthy host plants by itself, an effective means of attracting vector insects is a ‘life or death’ issue for viruses. Several studies have demonstrated the existence of such attraction mechanisms. Mauck et al. (2010) reported that infection with Cucumber mosaic virus (CMV) increases the host plant attractiveness to vector aphids. Maris et al. (2004) also reported that TSWV-infected pepper plants attract thrips more effectively than uninfected plants. Similar findings have been reported for plant pathogens other than viruses, such as phytoplasmas (Mayer et al. 2008a, Mayer et al. 2008b). However, it is not well understood how pathogen infection triggers changes in host traits that attract the insect vector. TSWV infection is also reported to affect thrips’ performance. Maris et al. (2004) showed that thrips grow faster and better on TSWV-infected pepper plants than on healthy plants. Therefore, TSWV infection is likely to affect many physiological properties of host plants, some of which may attract or benefit the vector insect.

Our previous studies have shown that jasmonate (JA) plays an important role in a plant's response and resistance to thrips, and that the JA-regulated plant defense negatively affects the performance of thrips and their preference (Abe et al. 2008, Abe et al. 2009). Numerous studies have examined the functions of JA in a plant's defense against pathogen attack, insect feeding and mechanical wounding (Howe and Jander 2008, Thomma et al. 1998). On the other hand, salicylic acid (SA) regulates a different set of basal defense responses in plants and functions in a plant's response to attack by pathogens such as fungi, bacteria and plant viruses (Pieterse et al. 2006). Importantly, negative interaction between the JA and SA pathways has been clearly shown in previous studies (Niki et al. 1998).

In this study, we investigated the mechanism of a tritrophic (three-level) interaction among TSWV, western flower thrips and Arabidopsis. We suggest that the antagonistic JA- and SA-regulated defense systems in host plants increase both the attractiveness and the food quality of TSWV-infected plants for the vector thrips.

Results

Effect of TSWV infection on the performance of thrips in Arabidopsis

To analyze the effect of TSWV infection on thrips feeding, we analyzed the difference in feeding damage between TSWV-viruliferous (virus-containing) thrips and non-viruliferous thrips. We put seven female adults of either viruliferous or non-viruliferous thrips on non-infected Arabidopsis plants and compared the feeding damage between the two types of insects. We did not find a notable difference within the first 6 d (Fig. 1A, D). However, we found a remarkable difference after 10 d (Fig. 1B, E), and the magnitude of this difference increased subsequently (Fig. 1C, F). We also found that plants that had been mechanically inoculated with TSWV suffered more thrips feeding than did mock-infected plants (data not shown). Next, we analyzed the numbers of adult and larval thrips to examine the effect of TSWV infection on the thrips population. After 14 d, few adult females were found due to the adult life span of this species (Fig. 2). However, whereas an average of 18 larvae were found on each of the plants treated with viruliferous thrips, only about nine were found on the plants treated with non-viruliferous thrips (Fig. 2).

Fig. 1

Comparison of feeding damage caused by viruliferous and non-viruliferous thrips. The symptoms caused by attacks by viruliferous thrips and non-viruliferous thrips were compared. WT plants were grown for 16 d in each pot. Seven adult females with or without TSWV were then allowed to feed on each plant. The photo shows plants 6 (A, D), 10 (B, E) and 14 d (C, F) after the beginning of feeding.

Fig. 2

Effect of TSWV infection on the number of thrips. WT plants were grown for 16 d in each pot. Seven adult females with or without TSWV were allowed to feed on each plant. The number of adults and larvae was measured after 14 d, and is shown as the mean ± SD of five independent determinations. Asterisks indicate significant differences (Student's t-test), **P < 0.01.

Effect of TSWV infection on plant defense mechanisms

To analyze the mechanism by which TSWV infection enhanced the thrips' performance, we analyzed the defense response of the host plants. Plant defenses are regulated by several complex signaling pathways (Van der Ent et al. 2008). SA and JA are global signals in plant defense systems, and, through cross-talk mechanisms, they are antagonistic to one another (Reymond and Farmer 1998, Pieterse et al. 2006). Therefore, we examined the expression of several SA- and JA-regulated ‘marker genes’ in Arabidopsis, which are used to indicate regulation by these two signaling pathways. Plants were mechanically inoculated with TSWV. We monitored TSWV accumulation by using the expression of the NSs gene, which encodes the 52.4 kDa non-structural protein of TSWV. In mechanically infected plants, TSWV could be detected after 7 d and had increased dramatically by 14 d (Fig. 3B). We also visually confirmed the accumulation of TSWV after mechanical infection by performing tissue blot analyses of whole rosette plants (data not shown); the results were consistent with those of the expression analyses of the viral NSs gene. The expression of the SA-regulated marker genes PR1 (pathogenesis-related gene 1) and BGL2 (β-1,3-glucanase 2) was up-regulated in association with virus proliferation (Fig. 3C, D). The expression of the JA-regulated marker genes VSP2 (vegetative storage protein 2) and LOX2 (lipoxygenase 2) was also up-regulated by TSWV infection (Fig. 3E, F); however, the magnitude of their expression was much lower compared with that of the SA-regulated marker genes. To analyze the effect of TSWV infection on the gene expression induced by thrips feeding, we put 10 non-viruliferous female adults on plants 14 d after mechanical inoculation or mock treatment and analyzed gene expression after 2 and 3 d. Thrips feeding did not change the expression of the SA-regulated marker genes induced by TSWV infection (Fig. 3C, D). Importantly, thrips feeding on mock-treated plants dramatically increased the expression of JA-regulated marker genes VSP2, LOX2, ChiB (β chitinase) and PR4 (pathogenesis-related gene 4), but the increase in the expression of these genes caused by thrips feeding on TSWV-infected plants was much smaller (Fig. 3E–H). Thrips feeding did not affect the level of TSWV accumulation (Fig. 3B).

Fig. 3

Expression of JA- and SA-regulated marker genes. (A) Time course of treatments used for expression analyses. At time 0, plants were mechanically inoculated with TSWV (TSWV). Mock treatment was performed on other plants as a control (Mock). The plants were sampled after 7 and 14 d. At 14 d, adult female non-viruliferous thrips were allowed to feed on inoculated and mock-treated plants (thrips). A control treatment without thrips was also performed (control). After 2 and 3 d, the plants were sampled. Each value represents the average of three replications of three plants each ± SD. (B) The TSWV NSs gene; (C, D) PR1 and BGL2, marker genes of the SA-regulated plant defense; (E–H) VSP2, LOX2, ChiB and PR4, marker genes of the JA-regulated plant defense. The expression level of each gene was normalized to the expression of CBP20 and the relative value is shown.

We then analyzed the SA and JA contents of the plants 10 d after mechanical TSWV infection. TSWV infection clearly increased the endogenous levels of both SA and JA (Fig. 4A, B), but the increase in the level of SA was much greater than that in the level of JA. We previously reported that thrips feeding increased the endogenous level of JA but not of SA (Abe et al. 2008), and we observed the same result here (Fig. 4A, B). Thrips feeding plus TSWV infection seemed to increase the JA level over that seen with TSWV infection alone (Fig. 4A). However, the SA levels in TSWV-infected plants were similar, with or without thrips feeding (Fig. 4B).

Fig. 4

Effect of TSWV infection and thrips feeding on the biosynthesis of JA and SA. Endogenous levels of JA (JA + methyl JA; A) and SA (B) were measured. Mechanical inoculation of TSWV was done on 16-day-old WT plants (TSWV). Mock treatment was also performed as a control (Mock). Plants were sampled after 10 d. Ten adult female non-viruliferous thrips were allowed to feed on TSWV-inoculated or mock-treated plants for 3 d, 7 d after inoculation/treatment (TSWV + Thr, Mock + Thr). Control plants were maintained for the same amount of time without TSWV inoculation or thrips feeding (Control). The results shown are means ± SD of at least four independent measurements. Different letters indicate statistically significant differences between treatments (Tukey–Kramer HSD test; P < 0.05).

Effect of TSWV infection on host plant attractiveness

Finally, we analyzed the effect of TSWV infection on the thrips' host plant preference (i.e. host plant attractiveness to thrips). Maris et al. (2004) reported that thrips prefer TSWV-infected pepper plants over mock-treated pepper plants. To examine the thrips' preference in the Arabidopsis plant system, we performed a choice test by using TSWV-infected and mock-treated wild-type (WT) Arabidopsis plants (Fig. 5A); the thrips clearly preferred the TSWV-infected plants (replicated G-test, G_p = 33.72, df = 1, P < 0.001; Fig. 5A). To analyze the importance of JA- and SA-regulated plant defense systems to this preference, we performed the same assay using JA-insensitive coi1-1 (coronatine insensitive 1-1) mutant plants; interestingly, thrips were not attracted to TSWV-infected coi1-1 mutants (G_p = 3.67, df = 1, P = 0.055; Fig. 5B). We also compared the TSWV accumulation between WT plants and coi1-1 mutants and found no significant difference in TSWV accumulation between them (Fig. 5C). To analyze further the effect of elevated SA levels after TSWV infection on the attraction of TSWV-infected plants to thrips, we applied SA directly to WT plants with an ink brush and compared the thrips' preference for SA-treated plants over water-treated plants. The SA-treated plants were significantly more attractive to thrips than the water-treated plants (G_p = 59.67, df = 1, P < 0.001; Fig. 5D). We also found many more thrips feeding scars on the SA-treated plants than on the water-treated plants (G_p = 21.35, df = 1, P < 0.001; Fig. 5E).

Fig. 5

Effect of TSWV infection on host plant preference of thrips. (A, B) Attractiveness of non-infected plants and infected plants to thrips was compared using WT (A) and coi1-1 mutant plants (B). Non-infected and infected plants were grown at each end of a rectangular pot. Fifty adult females were collected in a 1 ml tube, which was laid between the plants. The number of adult thrips on each plant was counted after 2 d. ‘Other’ represents the number of adult thrips found somewhere other than on the plants. Bars indicate the mean ± SD based on more than three independent determinations. Asterisks indicate a significant difference between the two plants, replicated G-test, ***P < 0.001. NS designates not significant at P < 0.05. (C) Virus accumulation as indicated by the level of NSs expression on WT plants and coi1-1 mutant plants. (D, E) Effect of SA treatment on the attraction of plants to thrips. SA-treated (1 mM) and water-treated plants were grown at each end of a rectangular pot. The number of adult thrips on each plant was counted after 2 d (D). The relative area of feeding scars on the SA-treated and water-treated plants is shown in (E). Bars indicate the mean ± SD based on more than three independent determinations. Asterisks indicate a significant difference between the two plants, replicated G-test, ***P < 0.001.

Discussion

Here, we report that TSWV infection indirectly promotes attacks by thrips through the plant defense system. We found that TSWV infection enhanced the feeding of its thrips vector on Arabidopsis (Fig. 1) and also increased the population density of thrips of the next generation (Fig. 2). Maris et al. (2004) reported that TSWV infection of pepper plants also increases thrips' performance. The enhancement of thrips' performance by TSWV infection in both pepper (Solanaceae) and Arabidopsis (Brassicaceae) indicates that this phenomenon is not restricted to a particular plant family.

Many previous reports have described the importance of the immunity-related plant phytohormones JA and SA in both the response and the resistance of plants to biotic stresses (Pieterse et al. 2006, Bari et al. 2009). Numerous studies have examined the functions of JA in plant defenses against pathogen attack and insect feeding (Thomma et al. 1998, Howe and Schaller 2008). We previously reported that JA functions to increase a plant's resistance to thrips and that a reduction in the JA-regulated plant defense enhanced the thrips' performance characteristics such as feeding and oviposition, thus leading to an increase in the thrips population under virus-free conditions (Abe et al. 2009). On the other hand, SA regulates a different set of basal immune system responses in plants and functions in the plant's response to attack by pathogens such as fungi, bacteria and plant viruses (Pieterse et al. 2006). The JA and SA pathways have different effects on plant resistance to pathogens (Uknes et al. 1992, Thomma et al. 1998) and insect herbivores (Cipollini et al. 2004, Howe and Jander 2008). A negative interaction between the JA and SA pathways has been clearly shown in previous studies (Niki et al. 1998). Leon-Reyes et al. (2009) also noted the antagonistic relationship between JA and SA in the context of plant resistance to thrips, reporting that JA-induced resistance to a thrips attack is decreased by SA.

Importantly, our results here clearly indicate that TSWV infection activated SA biosynthesis and the SA-regulated plant defense (Figs. 3, 4). The inducibility of SA-regulated gene expression depended on the accumulation of TSWV (Fig. 3B). We also analyzed the antagonistic effect of TSWV infection on feeding-induced JA-regulated gene expression, and found that JA-regulated gene expression in TSWV-infected plants was lower than that in mock-treated plants (Fig. 3E–H). It is possible that the increased performance of thrips on TSWV-infected plants was caused by a reduction of the JA-regulated plant defense, which was antagonistically regulated by an increase in SA-regulated plant defense responses. In this model, TSWV infection indirectly enhances the performance of its vector, thrips, by means of the antagonistic JA- and SA-regulated host plant defense systems.

Viruses are immobile and thus need to be transported by vectors such as insects. We further examined the attractiveness of TSWV-infected Arabidopsis plants to thrips. TSWV-infected plants attracted more thrips than did mock-treated plants (Fig. 5A). These results are consistent with the findings of Maris et al. (2004) in pepper. Interestingly, we found here that the attraction to thrips observed in TSWV-infected WT Arabidopsis plants disappeared in TSWV-infected, JA-insensitive coi1-1 mutants (Fig. 5B). This result indicates the importance of JA signaling to the attractiveness to thrips of TSWV-infected plants. We previously reported that non-infected JA-insensitive coi1-1 mutants showed a dramatically increased attraction (or decreased avoidance) to thrips compared with WT plants (>70 thrips per coi1-1 mutant vs. 5 thrips per WT plant) (Abe et al. 2009). Taken together, these data indicate that TSWV infection affects thrips' preference by interrupting the JA-regulated plant defense. To analyze the effect of SA on the attractiveness of a plant to thrips, we applied SA directly to WT plants. SA-treated plants were clearly preferred to untreated plants and suffered much more feeding by thrips than did untreated control plants (Fig. 5D, E). These results suggest that the increased level of SA in a plant after TSWV infection increased its attraction to thrips. When considering in combination the importance of JA signaling and the increased SA levels, SA–JA antagonism in the plant defense system can explain the enhanced performance of thrips on TSWS-infected plants and their preference for such plants.

In this tritrophic interaction, thrips benefit from being viruliferous because of the response of the host plant defense system to TSWV. Feeding by adult TSWV-viruliferous thrips results in TSWV infection of the host plant, which reduces the JA-regulated host plant defense. Both the adult thrips and their progeny can be expected to benefit from this lowering of the host plant's defense. Note that TSWV is not transmitted transovarially (Wijkamp et al. 1996): thrips acquire TSWV only at the larval stage from infected host plants (Whitfield et al. 2005). We found more larvae on plants inoculated with viruliferous adult females than on those inoculated with non-viruliferous females (Fig. 2B). These results indicate that TSWV infection facilitates the production of candidate thrips of the next generation, which will spread the TSWV. In addition, the reduction in the JA-regulated plant defense is also beneficial to TSWV because it increases the attractiveness of the plants to thrips and thus assists in the spread of TSWV.

Our next topic is to clarify the possible effects of plant chemicals induced by TSWV infection on the behavior of thrips. TSWV infection probably changes the emission levels of volatile plant chemicals that can act as attractants or repellents of thrips, as reported for other plant–herbivore interactions (de Vos et al. 2010). For example, overexpression of the dual linalool/nerolidol synthase (FaNES1) gene causes Arabidopsis to release these two terpenes, which act as repellents of green peach aphids (Myzus persicae) (Aharoni et al. 2003). Further, it is possible that a change of the composition of plant volatiles after TSWV infection decreases the attraction of natural enemies to thrips-infested plants, the efficiency of their foraging success, or both.

Note that TSWV infection increased the JA level as well as the SA level, and also induced JA-regulated marker gene expression. Several types of plant virus infection have been reported to elevate endogenous JA levels as well as SA levels (Jameson and Clarke 2002); Conconi et al. (1996) proposed that the JA elevation is caused by disruption of the plant cell membrane. Two opposing arguments have been made about the function of elevated JA after plant virus infection, namely that JA either inhibits plant virus replication or assists in the multiplication of the plant virus (Jameson and Clarke 2002). The biological function of elevated JA after TSWV infection is still unknown. Further analyses will be needed to reveal the details of the relationships among TSWV, thrips and host plants.

In this study, we demonstrated the importance of the antagonistic effects of JA- and SA-regulated plant defense systems in enhanced thrips' performance and in the attractiveness of TSWV-infected plants to thrips. TSWV induces the host plant SA-regulated defense system, which leads to a decrease in the JA-regulated defense. Many remaining questions need to be answered before we will understand the entire picture of this tritrophic interaction at the molecular level. Most recently, Stafford et al. (2011) reported that TSWV infection directly influences the feeding behavior of male thrips, but not that of female thrips, and enhances the transmission efficiency of the virus. This observation raises the question as to how plants respond to attacks by thrips of different sex. Moreover, TSWV infection seems to have more wide-reaching effects on multitrophic interactions. For example, TSWV infection via thrips may affect the survival and development of other, non-vector herbivores (e.g. spider mites; Belliure et al., 2010). Further studies, including transcriptome and metabolome analyses, are required to improve the molecular basis of our understanding of additional multitrophic interactions, based on the tritrophic plant–herbivore–virus interactions.

Materials and Methods

Plants

Plants (Arabidopsis thaliana ecotype Col-0) were grown in soil as described previously (Weigel and Glazebrook 2002). JA-insensitive coi1-1 mutants (Col-0 genetic background) were also used in this study. Arabidopsis seeds were sown on sterile soil in pots, moistened and placed in a cold room at 4°C in the dark to synchronize germination. The pots were then transferred to 22°C with a long-day photoperiod (16 h light/8 h dark). Plants at the four-leaf stage were transferred individually to pots and grown to the rosette stage.

Virus isolation

The TSWV O isolate, originating from green pepper (Capsicum annuum) in Japan, was used in this study (Tsuda et al. 1993). The isolate was first maintained by mechanical transmission on Nicotiana rustica and later by thrips transmission on Arabidopsis plants. Mechanical inoculation of the virus was performed as described previously (Tsuda et al. 1993).

Thrips maintenance and virus acquisition

A colony of western flower thrips, Frankliniella occidentalis, was maintained as described previously (Murai and Loomans 2001). Datura stramonium infected with TSWV was used as the virus acquisition source. First-instar larvae were confined with D. stramonium leaves infected with TSWV in plastic dishes for an acquisition access period of 24 h. After this period, the larvae were reared to adulthood. Viruliferous and non-viruliferous thrips were distinguished by using the petunia leaf disk assay (Wijkamp and Peters 1993). Thrips determined to be viruliferous or non-viruliferous were used for the following analyses.

Thrips attack assay

The thrips attack assay was performed with female adults 14–21 d after their emergence from the pupal stage. The insects were starved for 2–3 h before being placed on test plants. Adult females were grouped and allowed to feed on each whole plant in an acryl cylinder chamber with air ventilation windows covered with a fine mesh.

Expression analyses

Total RNA (2 μg), isolated with Trizol reagent (Invitrogen) and an RNeasy MinElute Cleanup Kit (Qiagen), was treated with RNase-free DNase (TAKARA) to eliminate genomic DNA. First-strand cDNA was synthesized with random oligo-hexamers and Superscript III reverse transcriptase according to the manufacturer's instructions (Invitrogen). Quantitative real-time PCR analysis was performed as described previously (Abe et al. 2008). Expression of CBP20 (cap-binding protein 20) was used for normalization as a standard control gene. Nucleotide sequences of the gene-specific primers were as follows: VSP2 (At5g24770; forward primer, 5′-GTTAGGGACCGGAGCATCAA-3′; reverse primer, 5′-AACGGTCACTGAGTATGATGGGT-3′); LOX2 (At3g45140; forward primer, 5′-TTGCTCGCCAGACACTTGC-3′; reverse primer, 5′-GGGATCACCATAAACGGCC-3′); ChiB (At3g12500; forward primer, 5′-ACGGAAGAGGACCAATGCAA-3′; reverse primer, 5′-GTTGGCAACAAGGTCAGGGT-3′); PR4 (At3g04720; forward primer, 5′-GTGGGATGCTGATAAGCCGT-3′; reverse primer, 5′-AACACTTGCCGCAAGAAGCT-3′); BGL2 (At3g57260; forward primer, 5′-GCCGACAAGTGGGTTCAAGA-3′; reverse primer, 5′-AACCCCCCAACTGAGGGTT-3′); PR1 (At2g14610; forward primer, 5′-GTTGCAGCCTATGCTCGGAG-3′; reverse primer, 5′-CCGCTACCCCAGGCTAAGTT-3′); CBP20 (At5g44200; forward primer, 5′-CCTTGTGGCTTTTGTTTCGTC-3′; reverse primer, 5′-ACACGAATAGGCCGGTCATC-3′); and TSWV NSs (forward primer, 5′-AGTGAAATCTCTGCTCATGTCAGC-3′; reverse primer, 5′-GCCAGAGGCTCAGCTTGAAAT-3′).

Quantification of JA and SA

Quantification of JA and its methyl ester was performed as described previously (Seo et al. 1995) except that an HP6890 gas chromatograph fitted to a quadrupole mass spectrometer (Hewlett-Packard) was used for the analysis. Quantification of free SA was performed as described previously (Seo et al. 1995).

Counting of the thrips population

Arabidopsis plants were grown in soil covered with fine zirconia beads (0.4 mm diameter; Nikkato Co.) to make it easy to find the thrips. At the time of the assay, 16-day-old plants were placed in a cylindrical acryl chamber. Seven adult females with or without TSWV were put on non-infected WT plants. After 2 weeks, the adults and larvae were counted.

Choice assay

The choice assay was described previously (Abe et al. 2009). Fifty adult females were deposited halfway between the plants and allowed to move freely. After 2 d, the thrips on each plant were counted. The results from 3–6 replications of each experiment were subjected to a replicated G-test; the null hypothesis was that the thrips would be distributed equally between the two pots (Sokal and Rohlf 1995). We confirmed that there was no significant heterogeneity among replications in each experiment (P > 0.05 for each G_h), suggesting good reproducibility of the tests.

Funding

This work was supported by the Ministry of Education, Culture, Sports, Science and Technology of Japan [Grant-in-Aid for Scientific Research (C) to H.A.].

Acknowledgments

We thank F. Mori, S. Kawamura and I. Sasaki of RIKEN BRC, and Y. Matsumura of the National Agricultural Research Center for their excellent technical assistance.

Abbreviations

Abbreviations
 
• BGL2

  β-1,3-glucanase 2

 
• CBP20

  cap-binding protein 20

 
• ChiB

  β chitinase

 
• coi1-1

  coronatine insensitive 1-1

 
• JA

  jasmonate

 
• LOX2

  lipoxygenase 2

 
• PR1

  pathogenesis-related gene 1

 
• PR4

  pathogenesis-related gene 4

 
• SA

  salicylic acid

 
• TSWV

  Tomato spotted wilt virus

 
• VSP2

  vegetative storage protein 2

 
• WT

  wild type.