HOOKLESS1(HLS1)is a central positive regulator of apical hook development(Lehman et al.,1996;An et al.,2012)(Figures 1A and1B).The mutants eto1-1and ctr1-1,with high level of HLS1expression,exhibited exaggerated hook curvature,while the mutant ein3-1eil1-3,with low levels of HLS1expression, displayed reduced hook curvature(Figures1A and1B).ACC treatment induced HLS1expression and exaggerated apical hook curvature in the wild type,and ACC-induced HLS1expression and hook formation were obviously repressed by JA treatment (Figures1A and1C).These results suggest that JA represses HLS1expression to inhibit ET-enhanced hook formation and imply that JA acts upstream of HLS1to repress hook curvature. To genetically verify whether JA acts upstream of HLS1,we further generated the double mutant coi1-2hls1-1and the triple mutant coi1-2ein3eil1via genetic cross of coi1-2with hls1-1or ein3eil1.The results in Figure2showed that the coi1-2exhibited an exaggerated hook curvature,while no hook was formed in coi1-2hls1-1and hls1-1.Similar data were also observed for coi1-2ein3eil1(Figure2).Suppression of the exaggerated hook curvature in coi1-2by the hls1-1and ein3eil1mutations suggests that COI1acts upstream of the EIN3/EIL1-HLS1cascade to regulate apical hook formation.
To identify the key components responsible for repression of hook curvature in JA signaling pathway,we examined apical hook phenotypes in JA signaling mutants.As expected,the coi1-1mutant exhibited an exaggerated apical hook curvature (Figure3A)(Turner et al.,2002).JAZ1D3A transgenic plants, with high levels of JAZ proteins(Thines et al.,2007),also displayed an exaggerated apical hook curvature(Figure1A). Among the key transcription factors targeted by JAZ proteins, MYB21/MYB24/MYB57(Song et al.,2011)and WD-repeat/ bHLH/MYB complex(Qi et al.,2011)are not involved in the suppression of hook formation,as the myb21myb24myb57 and gl3egl3tt8mutants exhibited wild-type-like hook curva-ture(Figure3A).
Interestingly,the myc2single mutant exhibited a mildly exaggerated apical hook curvature compared with the wild type(Figure3A);the apical hook curvature was clearly en-hanced in the double mutants myc2myc3and myc2myc4 (Figure3A),while the triple mutant myc2myc3myc4dis-played the strongest apical hook curvature(Figure3A),which is similar to that observed in the coi1mutant(Figure3A).The hook curvature of the single or double mutants(myc2,myc2 myc3,and myc2myc4)could be further inhibited by JA
264The Plant Cell
treatment,whereas the triple mutant (myc2myc3myc4)was completely insensitive to JA-inhibited hook curvature (Figure 3A).Furthermore,JA was unable to repress ACC-enhanced hook curvature in myc2myc3myc4(Figure 3A).Consistent with the exaggerated hook curvature,the expression of HLS1was upregulated in the mutants myc2,myc2myc3,myc2myc4,and myc2myc3myc4(Figures 3B).Furthermore,ACC-enhanced HLS1expression in myc2myc3myc4was not repressed by JA treatment (Figures 3C).
Taken together,the results in Figure 3suggest that MYC2,MYC3,and MYC4function redundantly to mediate JA-inhibited hook curvature.
MYC2,MYC3,and MYC4Interact with EIN3and EIL1Having shown that MYC2,MYC3,and MYC4function re-dundantly to repress HLS1expression and mediate JA inhibition of ET-enhanced hook curvature (Figure 3),we further found that MYC2,MYC3,and MYC4were able to interact with EIN3and EIL1(Figure 4),activators of HLS1(An et al.,2012).
The yellow ?uorescent protein (YFP)–based bimolecular ?uo-rescence complementation (BiFC)assays showed that coex-pression of EIN3-nYFP (fusion of EIN3with N-terminal fragment of YFP)or EIL1-nYFP with cYFP-MYC2(fusion of MYC2with C-terminal fragment of YFP),cYFP-MYC3,or cYFP-MYC4
produced
Figure 1.JA Suppresses the ET-induced Apical Hook Formation.
(A)The hook phenotypes of 4-d-old etiolated Arabidopsis seedlings Columbia-0(Col-0;WT),coi1-1,JAZ1D 3A ,eto1-1,ctr1-1,ein2-1,ein3-1ein1-3(ein3eil1),and hls1-1grown in the dark on MS medium supplied without (Mock)or with 5m M MeJA (JA),10m M ACC,or 10m M ACC plus 5m M MeJA (ACC+JA).
(B)Real-time PCR analysis for HLS1in 4-d-old etiolated Col-0(WT),eto1-1,ctr1-1,and ein3eil1.Actin8was used as the internal control.Data are means (6SD )of three biological replicates.Lowercase letters indicate signi ?cant differences by one-way ANOVA analysis with SAS software (P <0.05).(C)Real-time PCR analysis for HLS1in 4-d-old etiolated Col-0(WT)treated with mock,100m M MeJA (JA),100m M ACC,or 100m M ACC plus 100m M MeJA (ACC+JA)for 6h.Actin8was used as the internal control.Data are means (6SD )of three biological replicates.Lowercase letters indicate signi ?cant differences by one-way ANOVA analysis with SAS software (P <0.05).
MYC2-EIN3Mediates JA-ET Antagonism 265
strong YFP signals in the nuclei (Figure 4A),while the negative controls did not (Supplemental Figure 1),demonstrating that MYC2,MYC3,and MYC4interact with EIN3and EIL1.
We also used pull-down assays to representatively examine the interaction of MYC2with EIN3(Figure 4B).Puri ?ed maltose binding protein (MBP)–fused MYC2(MBP-MYC2)resin was in-cubated with total protein from Nicotiana benthamiana leaves with transient expression of ?ag-tagged EIN3(?ag-EIN3)and separated by SDS-PAGE for immunoblotting with anti-?ag an-tibody.As shown in Figure 4B,the MBP-MYC2resin could pull down ?ag-EIN3,suggesting that MYC2interacts with EIN3.Furthermore,we performed coimmunoprecipitation (Co-IP)as-says to examine the interaction between MYC2and EIN3in planta.The ?ag-EIN3was coexpressed with myc-tagged MYC2(myc-MYC2)or myc-COI1,respectively,in leaves of N.benthamiana ,and
the total proteins were then used for coimmunoprecipitation.The results showed that ?ag-EIN3was indeed coimmunoprecipitated with myc-MYC2(Figure 4C),but not with the control protein myc-COI1(Figure 4C).Taken together,the BiFC assay,pull-down assay,and Co-IP assay consistently demonstrate that MYC2,MYC3,and MYC4interact with EIN3and EIL1(Figure 4).
MYC2Inhibits Transcriptional Activity of EIN3and EIL1Having shown that MYC2,MYC3,and MYC4interact with EIN3and EIL1,we then investigated whether such interactions affect the transcriptional activity of EIN3and EIL1using an Arabidopsis mesophyll protoplast transfection-based transcriptional activity assay (Hellens et al.,2005).
A previous study showed that EIN3could bind to the pro-moter of HLS1to activate its expression,leading to hook cur-vature (An et al.,2012).We ?rst examined whether MYC2affects the in ?uence of EIN3on HLS1transcription.As expected (An et al.,2012),expression of EIN3dramatically activated the ex-pression of LUC driven by the HLS1promoter (Figures 5A and 5B).However,coexpression of MYC2with EIN3signi ?cantly re-pressed EIN3-activated P HLS1-LUC activity (Figure 5B).Similarly,expression of EIL1activated P HLS1-LUC activity,whereas coex-pression of MYC2repressed EIL1-activated P HLS1-LUC activity (Figures 5A and 5C).The results in Figures 4and 5A to 5C demonstrate that MYC2interacts with EIN3and EIL1to interfere with their effect on the transcription of HLS1.
Having shown that MYC2suppresses the effect of EIN3and EIL1on HLS1transcription,we further examined whether MYC2could repress the effects of EIN3and EIL1on the transcription of another target gene,ETHYLENE RESPONSE FACTOR1(ERF1)(Solano et al.,1998),a key transcription factor that activates the expression of PDF1.2to induce resistance against necrotrophic pathogens (Préet al.,2008;Zarei et al.,2011).As shown in Figures 5D and 5E,overexpression of EIN3activated the ERF1promoter that controlled expression of the LUC gene (P ERF1-LUC ),whereas such EIN3-activated P ERF1-LUC expression was obviously repressed by coexpression of MYC2(Figures 5D and 5E).Furthermore,we found that expression of EIL1also acti-vated P ERF1-LUC activity,while coexpression of MYC2repressed the EIL1-activated P ERF1-LUC activity (Figures 5D and 5F).Taken together (Figures 4and 5),these results demonstrate that MYC2interacts with EIN3and EIL1to attenuate their effect on the transcription of their target genes HLS1and ERF1.
Disruption of EIN3and EIL1Suppresses Exaggerated Apical Hook Formation and Resistance against a Necrotrophic Pathogen in myc2
