3.3 OsMYC2 is a direct target of OsPHR2
OsPHR2 and OsMYC2 are well-known as the central transcription regulators in Pi starvation and JA signaling, respectively (Zhou et al. , 2008; Kazan & Manners, 2013). Interestingly, bioinformatics analysis identified three P1BS cis -elements in the 2-kb promoter region ofOsMYC2 (Figure 3a). Thus, we first performed a yeast one-hybrid (Y1H) assay to examine their interaction. The 2000 bp fragment containing the P1BS elements was used as bait and cloned into a pAbAi reporter vector, while a pGADT7-Rec2-OsPHR2 vector was used as prey. Yeast cells co-transformed with bait and prey were grown on selective media lacking Leu with or without AbA. The assay results demonstrated that OsPHR2 indeed bound to the promoter region of the OsMYC2gene (Figure 3b).
We then performed EMSA using purified His-OsPHR2 fusion protein and biotin-labeled probes (P3 and P4 in Figure 3a). OsPHR2 recombinant protein specifically bound the biotin-labeled probes, and this binding could be suppressed by unlabeled probes (Figure 3c). To test whether OsPHR2 activated the OsMYC2 promoter, we performed transient expression assays with dual effectors. Transient transfection assay revealed OsPHR2-Flag to increase the expression of ProOsMYC2-LUCin N. benthamiana leaves as compared with GFP-Flag control (Figure 3d). Finally, to determine whether OsPHR2 directly binds theOsMYC2 promoter in vivo , we performed ChIP-qPCR with anti-Flag antibody on transgenic plants expressing OsPHR2-Ov1fused with Flag (Zhou et al. , 2008). In this assay, OsPHR2 displayed strong binding to the region containing P1BScis -elements, but not to other regions in the OsMYC2promoter (Figure 3e). Together, these results suggest OsPHR2 is directly targeted to the cis -element in the OsMYC2 promoterin vitro and in vivo, and overexpression of OsPHR2 results in up-regulation of OsMYC2 .
3.4 OsPHR2 confers JA signaling activation and Xoo resistance through regulating expression of OsMYC2
The data above clearly demonstrated that OsPHR2 activates the transcription of OsMYC2 ; accordingly, we further dissected the effects of OsPHR2 on the expression of OsMYC2 -mediated JA-responsive genes and JA production in planta . We first examined OsMYC2 expression in the presence or absence of Pi inOsPHR2-Ov1 , phr2 and the corresponding wild-type NIP plants using RT-qPCR. Under the Pi-sufficient condition, we found thatOsMYC2 was up-regulated in OsPHR2-Ov1 leaves but down-regulated in phr2 plants compared with NIP plants (Figure 4a). Consistently, similar expression patterns of JA-responsive genes were also observed in the leaves of the tested mutants (Figure 4b-h). In addition, the Pi starvation-triggered induction of JA-responsive genes, including OsMYC2 , over three days of growth on Pi-deficient solution was enhanced in OsPHR2-Ov1 plant leaves but suppressed in phr2 mutants relative to the Pi-sufficient condition (Figure 4a-h). These results indicate that OsPHR2 positively modulates the expression of JA-responsive genes.
Furthermore, MeJA content was higher in OsPHR2-Ov1 plants but lower in phr2 mutants compared with NIP control, whereas only small changes was observed for JA-IIe (Figure 5a and S1b). We also tested the resistance of each type to Xoo infection. At 9 to 15 days after inoculation with virulentXoo , bacterial proliferation was lower in OsPHR2-Ov1plants but higher in phr2 plants than in NIP control (Figure 5b). Likewise, blight lesion length was significantly shorter inOsPHR2-Ov1 plants but longer in phr2 mutants than in NIP plants (Figure 5c and d), indicating a positive role of OsPHR2 duringXoo infection in rice. Taking together, our results indicated that the Pi starvation activated OsMYC2 expression, which resulted in the induction of MYC2-regulated JA-responsive genes and the increased MeJA production, and thus promoted resistance to Xooinfection in rice.