Long-distance transport of sucrose in source leaves promotes sink root growth by the EIN3-SUC2 module

In most plants, sucrose, a major storage sugar, is transported into sink organs to support their growth. This key physiological process is dependent on the function of sucrose transporters. Sucrose export from source tissues is predominantly controlled through the activity of SUCROSE TRANSPORTER 2 (SUC2), required for the loading of sucrose into the phloem of Arabidopsis plants. However, how SUC2 activity is controlled to support root growth remains unclear. Glucose is perceived via the function of HEXOKINASE 1 (HXK1), the only known nuclear glucose sensor. HXK1 negatively regulates the stability of ETHYLENE-INSENSITIVE3 (EIN3), a key ethylene/glucose interaction component. Here we show that HXK1 functions upstream of EIN3 in the regulation of root sink growth mediated by glucose signaling. Furthermore, the transcription factor EIN3 directly inhibits SUC2 activity by binding to the SUC2 promoter, regulating glucose signaling linked to root sink growth. We demonstrate that these molecular components form a HXK1-EIN3-SUC2 module integral to the control of root sink growth. Also, we demonstrate that with increasing age, the HXK1-EIN3-SUC2 module promotes sucrose phloem loading in source tissues thereby elevating sucrose levels in sink roots. As a result, glucose signaling mediated-sink root growth is facilitated. Our findings thus establish a direct molecular link between the HXK1-EIN3-SUC2 module, the source-to sink transport of sucrose and root growth.


Introduction
Sugars produced from photosynthesis affect plant growth and development throughout the life cycle, from germination to senescence. Sucrose is a major photosynthetic product and the predominant transported sugar [1]. The loading of sucrose into the phloem is essential for Arabidopsis growth, development and reproduction. The sucrose transporter SUC2 in Arabidopsis is a phloem-specific SUC (SUCROSE TRANSPORTER), which is specifically expressed in companion cells [2]. SUC2 has a central role in the loading of sucrose into the phloem and is essential for high-performance sucrose transport from source to sink organs in Arabidopsis [2][3][4]. The suc2 mutant exhibits stunted growth, delayed development and sterility. Further, these mutant plants accumulate massive amounts of starch and sucrose in the leaf blades and transport of sugar to sink organs, including roots, the shoot apex and inflorescences is blocked [3]. Therefore, SUC2, is a key sucrose transporter that mediates sucrose phloem transport and its loss-of-function leads to seriously delayed growth and thereby accumulated sucrose in source tissues. A previous report has shown that growth of sink roots may be promoted by exogenously provided glucose acting through either signaling or metabolism [5]. However, whether and how plant root growth is mediated by the loading of sucrose into the phloem is largely unknown.
In plants, sucrose cannot be directly utilized, rather it is irreversibly hydrolyzed by invertase to the hexoses, glucose and fructose [3]. Glucose is perceived and transduced through two principal mechanisms: indirect sensing via a variety of energy and metabolite sensors including [SNF1-RELATED PROTEIN KINASE1 (SnRK1) and target of rapamycin (TOR)] or direct sensing through glucose sensors [6]. Glucose is indirectly sensed by the cellular energy sensor, SnRK1 (SNF1-RELATED KINASE1), which in turn directly interacts with and subsequently phosphorylates and promotes the degradation of a key ethylene signaling component, ETHY-LENE-INSENSITIVE3 (EIN3), a plant-specific transcription factor [6]. On the other hand, the direct glucose sensor, Hexokinase1 (HXK1), has both signaling and enzymatic functions [7][8] and is the only known nuclear glucose sensor [7]. Glucose indirectly and negatively controls the stability of the EIN3 transcriptional regulator via HXK1, whereas the mechanistic link between glucose sensing and EIN3 degradation has not to date been demonstrated [9]. EIN3 stability rapidly increases with ethylene treatment, while in the absence of ethylene signaling, EIN3 is targeted via the SCF EBF1/EBF2 and subsequently degraded via the 26S proteasome [10][11]. Through binding to specific promoter element termed an EIN3 binding site (EBS), EIN3 modulates the expression of numerous downstream target genes.
In this study, we show that glucose decreases the stability of EIN3 by the activity of a glucose sensor HEXOKINASE 1, which in turn elevates SUC2 function within source tissues, enhancing the loading of sucrose into the phloem in these tissues and thereby promoting the growth of sink organs, such as roots.

EIN3 and EIL1, acting downstream of HXK1, negatively regulate glucose signaling mediated-root growth
Previous data has shown that root growth may be promoted through glucose [5]. Glucose is directly perceived by the function of HXK1, the only known nuclear glucose sensor. Therefore, we explored if HXK1 was involved in the regulation of root growth. Whereas the HXK1/GIN2 loss-of-function mutant, gin2-1, is in the Ler background, another loss-of-function mutant, hxk1-3, is in the Col-0 background. When Arabidopsis seedlings were grown on solid MS medium with 1% glucose, both the hxk1-3 and gin2-1 mutants exhibited a shorter primary root length, compared to wild-type seedlings (Fig 1A and 1B and S1 Fig). By contrast, corresponding seedlings expressing a cauliflower mosaic virus 35S (35S):HXK1 transgene showed elongated primary root length compared to wild-type (Fig 1A and 1B). These findings indicate that HXK1 promotes root growth.
To determine whether EIN3 acts downstream of HXK1 in the regulation of root growth, we crossed hxk1-3 with ein3-1 plants. The ein3-1 line generated elongated primary roots, compared with wild-type plants (Fig 1E and 1F). Further, hxk1/ein3 mutant plants also generated elongated primary roots, similar to that of the ein3-1 mutant (Fig 1E and 1F). These findings indicate that EIN3 acts downstream of HXK1 in the regulation of root growth.
To investigate the link between glucose signaling and EIN3 dependent signaling, we generated a EIN3pro:EIN3-GFP/hxk1-3 line. A decrease of EIN3pro:EIN3-GFP signaling in wildtype, but not in the hxk1-3 line was apparent post glucose application (Fig 2). Thus, increased GFP fluorescence was generated in the hxk1-3 relative to that than in the wild-type background. These findings indicate that glucose decreases EIN3 stability via the glucose signaling sensor, HXK1.
Taken together, our results show that HXK1 positively regulates and EIN3 negatively regulate primary root growth. Further, EIN3 act downstream of HXK1 in the regulation of glucose signaling that mediates root growth.

EIN3 negatively regulates glucose signaling mediated-expression of SUC2
Chromatin immunoprecipitation sequence (ChIP-Seq) analysis of EIN3 [13] indicates that EIN3 may directly regulate expression of the Sucrose Transporter 4 (SUC4) [14]. Further, SUC2, but not SUC4, has been proposed to function as a major transporter of sucrose loading into the phloem in Arabidopsis [1]. Therefore, we determined if SUC2 is a downstream target gene of EIN3.
We first performed quantitative (q)PCR experiments to monitor the transcript abundance of SUC2 in wild-type, ein3-1 and ein3/eil1 seedlings. When seedlings were grown on MS medium with 1% glucose for 5 days, the expression of SUC2 was significantly elevated in ein3-1 and ein3/eil1 plants relative to wild-type ( Fig 3A). These qPCR results are consistent with previously reported data [15]. These findings indicate that EIN3 might repress SUC2 expression. Further, we monitored SUC2 expression over time in response to exogenously supplied glucose. The expression of SUC2 in wild-type, but not ein3/eil1 seedlings, was increased post glucose application (Fig 3B), indicating that glucose signaling promoted SUC2 expression is partially dependent on EIN3 and EIL1.
To monitor EIN3-dependent transcriptional activity during Arabidopsis root growth, we employed a transgenic line containing the b-glucoronidase (GUS reporter gene) driven by five tandem repeats of the EBS driven by a minimal 35S promoter (5 × EIN3 (EIN3 binding site)-GUS) [16]. GUS staining intensity in roots decreased with increasing exogenous glucose ( Fig  3C-3E), suggesting that glucose decreases the transcriptional activity of EIN3.
Together, these findings suggest that with increasing age, glucose accumulation decreases the transcriptional activity of EIN3, which may in turn elevate glucose signaling mediatedexpression of SUC2.

EIN3 directly binds to the promoter of SUC2 to repress SUC2 expression
Investigation of the DNA sequence comprising the promoter of SUC2 revealed a few putative EIN3 binding sites (EBSs) [16] (Fig 4A). To explore if EIN3 can interact with any of these EBSs, we employed a transgenic Arabidopsis line possessing a 35S:EIN3-GFP transgene in an ein3/eil1 genetic background (35S:EIN3-GFP/wild-type crossed with ein3/eil1 and subsequently 35S:EIN3-GFP/ein3/eil1 seedlings were identified by qPCR). We subsequently utilized EIN3 negatively regulates SUC2 expression. A, Bar graph showing the differential expression of SUC2 among 5-day-old wild-type, ein3-1 and ein3/eil1 seedlings grown on MS medium with 1% glucose. Quantification of the expression of the SUC2 gene in wild-type seedlings was set as 1.0. B, Bar graph showing the differences in the expression level of SUC2 between wild-type and ein3/eil1 seedlings with different time treatment. Wild-type seeds were sowed and grown in soil for 10 days. And then these seedlings were gathered and treated with 2% glucose for 30 min and 60 min. Total RNA was extracted from these treated seedlings and qPCR was performed. Quantification of 10-day-old wild-type seedlings without treatment (control) is set as 1. chromatin immunoprecipitation (ChIP) analysis to determine whether the transcription factor EIN3 directly binds to the EBSs in the promoter sequence of SUC2. Indeed, in the absence of glucose, EIN3 directly bound to one SUC2 promoter sequence (S2), but not to three other SUC2 promoter sequences (S1, S3 and S4 Figs) ( Fig 4B). Further, glucose treatment decreased the amount of amplified S2 sequence of the SUC2 promotor after an anti-GFP ChIP (Fig 4C), which may result from glucose decreasing EIN3 transcriptional activity (Fig 3C-3E).
To confirm EIN3 could directly bind to the S2 EBS within the promoter of SUC2 in vitro, electrophoretic mobility shift assay (EMSA) experiments were performed. Indeed, EIN3 bound exclusively to the labeled S2 EBS element in vitro ( Fig 4D). Excess unlabeled competitor DNA abolished this binding ability, implying EIN3 binding specificity ( Fig 4D). Moreover, EIN3 binding to the labeled S2 EBS element in vitro was not abolished by the mS2 (mutant S2) DNA sequence in vitro ( Fig 4E). Further, transient expression data in N. benthamiana showed that co-expressions of EIN3 decreased the expression of the reporter gene PGreenSUC2pro: Lucifertase (LUC) (Fig 4F and 4G).
Taken together, these data indicate that EIN3 directly binds to the SUC2 promoter to suppress SUC2 expression. Quantifications were normalized to the expression of UBQ5. C, Bar graph illustrating ChIP-PCR in glucose treatment. Ten-day-old 35S:EIN3-GFP/ein3/eil1 seedlings grown in soil. And then these wild-type seedlings were treated by using 0%, 1% and 3% glucose for 30 mins. The ChIP DNA amount at 0% glucose for 30 mins is set as 1.0. Quantification was normalized to the expression of UBQ5. D, Image illustrating unlabeled promoter of SUC2 were used as competitors to determine the specificity of DNAbinding activity for EIN3. E. Image illustrating a mutant version of the promoter of SUC2 was labeled with biotin and used for EMSA with EIN3 polypeptides. Free probe and EIN3 probe complexes are indicated by an asterisk and arrows, respectively in D and E. F, Image illustrating luminescence units is represented by LUC activity. G, Bar graph illustrating transient expression of the 35S:EIN3 effector construct with the SUC2pro-LUC reporter construct in N. benthamiana leaves. Note PGreen-mSUC2 indicates the conserved sites (TTCAAA) of the S2 region in SUC2 promoter were mutated. The activity of relative LUC represents arbitrary luminescence units, that is, expressing SUC2-LUC is~10, other expressions was quantified by using Adobe Photoshop CS (AdobeSystems) software, as described previously by [28]. Error bars represent SD (n = 3). Student's t test ( �� P < 0.01). https://doi.org/10.1371/journal.pgen.1010424.g004

SUC2 acts downstream of EIN3 to positively regulate root growth
Since EIN3 negatively regulates root growth (Fig 1) and EIN3 directly inhibits the expression of SUC2 (Figs 3 and 4), we therefore determined if SUC2 positively regulates root growth.
In By contrast, SUC2 loss-of-function homozygous suc2-5 (suc2-5 is a null allele) mutant seedlings exhibited severely retarded growth and development [17]. We observed that seedlings of suc2-5 showed a shortened primary root length, compared to wild-type ( Fig 5C and 5D and S5B and S5D Fig). These findings indicate that SUC2 promotes root growth.
To determine whether EIN3 genetically acts upstream of SUC2 in the regulation of root growth, we crossed suc2-5 with ein3-1 seedlings. The ein3-1 line exhibited elongated primary roots, whereas suc2-5 seedlings produced a shortened primary root, compared with wild-type (Fig 5E and 5F). Further, ein3/suc2 mutant seedlings generated a shortened primary root, which was similar to that of the suc2-5 mutant (Fig 5E and 5F).
Together, these findings indicate that SUC2 genetically acts downstream of EIN3 to positively regulate root growth.

Sucrose phloem loading and accumulation in source and sink organs are regulated by an EIN3-SUC2 module
In Arabidopsis, SUC2 encodes a proton/sucrose symporter localized throughout collection and transport phloem within the vasculature and is thus essential for effective transport of sucrose from source to sink organs [4]. Therefore, we determined if the EIN3-SUC2 module mediates the loading of sucrose into the phloem.
To further elucidate EIN3 mediated sucrose phloem loading, we employed a transgenic line containing the GUS reporter gene downstream of five tandem repeats of the EBS (5×EBS: GUS), which has been utilized to check EIN3 transcriptional activity [16]. We observed that EIN3 transcriptional activity was present in the veins of mature leaves of the 5×EBS:GUS line and was also observed in sieve tubes and companion cells in the phloem (Fig 6D and 6E). To confirm sucrose phloem loading was linked to the EIN3-SUC2 module, we assayed the sucrose level of vegetative leaves (source organs) and roots (sink organs) in 9, 11 and 13 dayold wild-type, suc2-5/+ and ein3/eil1 seedlings following exposure to light for 4 and 8 hours, respectively. In all seedlings, sucrose accumulation in both source and sink organs increased with increasing age (Fig 6B and 6C), suggesting total sucrose levels increase with tissue age under these conditions. Further, following exposure to light for 4 hours, sucrose accumulation over time was lower in the ein3/eil1 vegetative leaves and was higher in the suc2-5/+ line relative to wild-type (Fig 6B), because the phloem loading rate of sucrose in vegetative leaves over time was higher in the ein3/eil1 line and lower in suc2-5/+ plants relative to in wild-type ( Fig  6A). As a result, following exposure to light for 8 hours, sucrose accumulation over time was higher in ein3/eil1 roots and was lower in the suc2-5/+ line relative to wild-type (Fig 6C), which lead to elongated roots in ein3/eil1 seedlings and shortened roots in suc2-5/+ mutant seedlings (Fig 5).
Together, these findings indicate that the EIN3-SUC2 module regulates sucrose accumulation in both vegetative leaves (source organs) and roots (sink organs) by mediating the loading of sucrose into the phloem. As a result, sink growth in roots is promoted by sucrose transport.

Discussion
Plants acquire their carbon from atmospheric carbon dioxide. During daytime plant sugar phosphates are largely generated via photosynthetic assimilation and are further metabolized to support plant growth and development [1,2,3]. Excess photoassimilates are stored mostly in a form of starch in Arabidopsis and at relatively low levels of sucrose, both starch and sucrose are in part metabolized via their turnover (synthesis and breakdown) during the day and are stored at night [1,2,3]. Under advantageous environmental conditions, enhanced photosynthesis, will in turn enhance sucrose phloem loading to maintain the balance between source and sink [1]. In whole-plant carbon partitioning, it is noteworthy that sucrose phloem loading in source leaf blades is a step.
SUC2 has a central role in the phloem loading of sucrose and is essential for high-performance sucrose transport from source to sink organs in Arabidopsis and presumably SUC2 homologs in most crop plants undertake similar key roles [2][3][4]. Recently, posttranscriptional modulation of SUC2 in coordination with a change in plant biomass has also been reported [18]. Also, transcriptional modulation of phloem-loading of SUC in sugar beet in response to exogenously fed sucrose has been investigated [19], however, the associated regulatory pathway remained unclear. In this context, signal cues and associated signaling components that directly regulate SUC2 transcription and SUC2 activity have not been identified.
Our findings found that EIN3, acting the downstream of HXK1, functions upstream of SUC2 in the modulation of glucose signaling mediated-root growth (Figs 1 and 5). Further, the transcription factor EIN3 directly binds to the SUC2 promoter to inhibit SUC2 expression (Fig  4) to suppress root growth (Figs 1 and 5). These components thus form the HXK1-|EIN3-| SUC2 module to promote sucrose phloem loading in source tissues and thereby raise sucrose levels in sink roots (Fig 6). As a result, glucose signaling mediated-root growth is facilitated (Figs 2 and 7). Further, EIN3 transcriptionally suppresses SUC2 activities, providing insight for a previous observation that down-regulation of sucrose phloem loading is linked to transcriptional regulation of SUC transporters [14]. Diverse environmental stresses also result in a change in SUC transcriptional levels [14]. However, increases in sucrose phloem loading may also result from post-transcriptional regulation [14], as a recent report has revealed posttranscriptional regulation of SUC2 in response to high-light acclimation [18].
In this study, HXK1 is central factor, and this protein has both signaling and enzymatic functions [7][8]. HXK1 is a key integration point of various plant hormone or external stimuli that modulate plant growth and metabolism [7][8]. Further, HXK1 signaling and enzymatic activities directly or indirectly affect many genes associated with photosynthesis, including key components of the photosynthetic machinery as well as RUBISCO and PAL [7]. Therefore, the HXK1-|EIN3-|SUC2 module may regulate the formation of major photosynthetic product sucrose, and sucrose transport from source leaves to sink roots (Fig 7). The resulting HXK1-| EIN3-|SUC2 module may promote sink root growth by the interaction of glucose and various plant hormone or external stimuli, as glucose and auxin signaling interaction control Arabidopsis sink root growth [5].
In the case of Beta vulgaris (Bv) SUC1 (a sugar beet leaf sucrose symporter), sucrose negatively and specifically regulates both sucrose transport activity and steady-state mRNA levels [19]. By contrast, in this work, with increasing age, sucrose/glucose accumulation was increased (Fig 6B and 6C), which in turn decreased EIN3 transcriptional activity (Fig 3C-3E) by decreasing EIN3 levels [9,21] thereby raising Arabidopsis thaliana SUC2 (AtSUC2) dependent sucrose transport (Figs 6B and 6E and 7). Since Solanum tuberosum SUC4 (StSUC4) may play a role as a suppressor of StSUC1 [20], BvSUC1 and AtSUC2 may have opposite functions in the regulation of sucrose phloem loading.
However, when glucose accumulation in source leaves reaches a given threshold level of excess glucose, ethylene production is stimulated [15] and thus the level of EIN3 increases [9,21], which in turn directly decreases SUC2 transport activity (Figs 4 and 7 and S6 Fig). Consequently, sink root growth is delayed, as ethylene inhibits root growth [22]. Therefore, we speculate that when glucose accumulation reaches a given threshold level, excess glucose stimulates ethylene production, which in turn inhibits root growth by its canonical signal transduction pathway suppressing SUC2 activity (Fig 7). Thus, excess levels of glucose consequently inhibit root growth via increased ethylene levels. Indeed, following a 3.5% glucose application, ein3/eil1 plants had shortened root length (S7A and S7B Fig) and enhanced ethylene production (S7C Fig), relative to wild-type seedlings.
It has long been known that light promotes root growth. Light decreases EIN3 levels, whereas darkness elevates the accumulation of this transcriptional regulator [23]. When EIN3 levels are elevated [23] under darkness, SUC2 transcripts are reduced [24]. Thus, darkness raises EIN3 protein levels and consequently decreases SUC2 expression (Fig 3). As a result, this reduces sucrose phloem loading and thus enhances sucrose accumulation in source leaves. The resulting sucrose in sink roots is decreased, which in turn reduces root growth (S8 Fig).
Ethylene promotes hypocotyl growth [25], however, glucose and ethylene also have antagonistic functions in the regulation of plant growth [9]: hypocotyl length is decreased with an increasing sugar concentration [26]. Under basal conditions, HXK1 positively regulates hypocotyl growth [7]. However, under these conditions, EIN3 and SUC2 do not impact hypocotyl growth (S9 Fig). These findings suggest that the EIN3-SUC2 pathway may not be involved in the regulation of hypocotyl growth.
For simultaneous germination, seeds were treated with vernalization of 4˚C overnight and then sown on solid MS medium supplemented with 1% sucrose, pH 5.8 and 0.8% agar [22]. Seedlings grown on agar were maintained in a growth room under 16/8 h of light/dark cycle with cool white fluorescent light at 21 ± 2˚C for long-day conditions [27]. Unless otherwise noted, plants grown in soil were maintained in a controlled environment growth chamber under short-day conditions (10h light/14 h dark) and normal light conditions (130 μmol quanta PAR m -2 s -1 ) or high light conditions (230 μmol quanta PAR m -2 s -1 ) at 22˚C, as has been described by ref [28].
In all experiments, wild-type is the Col-0 background. Three biological replicates were performed with similar results, and error bars represent SD.

Assay of sugar metabolites
Plant Sucrose and glucose Assay Kit (Beijing Solarbio Science & Technology Co., Ltd, Cat#BC2465; http://www.solarbio.com/goods-9298.html) was used for assay of sugar metabolites. 0.1g mature leaf blades were ground into homogenate at 23˚C and then 0.5 ml extraction buffer added and centrifuged after extensive grinding and finally kept at 80˚C for 10 min and shaken 3-5 times. After cooling these extracts were centrifuged at 4,000g for 10 min, the supernatant was transferred to a fresh tube and 2 mg reagent 5 added to decolorize at 80˚C for 30 min. Then 0.5 ml extraction buffer was added, mixed and centrifuged at 4,000g for 10 min. The supernatant was transferred to a fresh tube for visible light analysis.
Three centrifugal tubes per sample were used with 25 μL of sample and then standard product (reagent 1) and water were added, respectively. Fifteen μL of reagent 2 was added, mixed and boiled at 100˚C for 5 minutes. Subsequently, 175μL of reagent 3 and 50μL of reagent 4 were added, respectively. They were reacted in boiling water for 10 min, followed by measuring the light absorption value at 480 nm after cooling. The sucrose content was calculated accordingly, with respect to the sample standard absorption value and sample's weight.

Phloem loading of sucrose
Nine-day-old seedlings grown on solid MS medium with sugars or without were used to assay phloem loading of sucrose. Following incubation in pretreatment solution (pH5.7; liquid MS medium) for 40-50 mins, the cotyledons were incubated in the phloem loading buffer (pH 5.7; MS medium with 0.1-0.2% Sucrose) with [ 14 C] Sucrose (0.50μCi mL -1 ) and then were cultivated for 2-3 hours. Following three washes in desorption solution (pH 5.7; liquid MS medium and 1% Sucrose), these seedlings were put in scintillation vials and 3.0 mL of scintillation cocktail was added. The amount of [ 14 C] phloem loading of sucrose was analyzed on a scintillation counter and expressed as cpm mg -1 fresh weight.
These cotyledons were observed with a confocal laser scanning microscope (LSM 900; Zeiss) for GFP. All images were gained in the multitrack pattern and analyzed by using LSMimage-examiner software (Zeiss). The data were exported as eight-bit TIFF files and generated by using Adobe Photoshop 5.5 (Adobe Systems).

Ethylene assay
These experiments were completed on 8-day-old wild-type and ein3/eil1 plants by using a gas chromatograph equipped with a flame ionization detector [Echrom Technologies (Shanghai) Ltd Co], as was described [15].

Plasmid constructs
For generating pCB2004EIN3pro::EIN3-GFP plasmids, the relevant primers are F-ggg gac aag ttt gta caa aaa agc agg ct AAT GTG TGG AAA CAT GGA TTT CT, R-ggg gac cac tttg tac aag aaa gct ggg t TTAGAACCATATGGATACAT; F-ggg gac aag ttt gta caa aaa agc agg ct cat att tgc atc tct cta tta gt; R-ggg gac cac tttg tac aag aaa gct ggg t atc gta ttt tta atc act cct aaa c were used.

Quantitative PCR
Total RNA was extracted from tissues indicated in the figures by the TRIZOL reagent (Invitrogen), as has been described by refs [27][28][29]. SYBR green was used to monitor the kinetics of PCR product in real-time RT-PCR, as has been described by ref [27]. Primer pairs of SUC2 (5 0 -ATC CAA TGG AGA AAG CTG CAA A-3 0 and 5 0 -ACC ACA GAG CCA AAT CAG GAA-3 0 ) were utilized. Primer pairs of EIN3 have been described by ref [29].

Transactivation assay
To examine EIN3 activity, we performed the below experiments. To generate proSUC2-LUC, the promoter was PCR amplified with primers proSUC2-F and proSUC2-R (ProSUC2-Fttcctgcagcccgggggatcctaccagatttcggtaaattggtattcc; ProSUC2-R-tgtttttggcgtcttccatggatttgacaaaccaagaaagtaagaa) for the gnomic DNA of Arabidopsis and inserted into the cloning site of the pGreen0800-LUC vector.
The Agrobacterium with the effector or reporter constructs were coincubated for 3 hour and then infiltrated into 20-day-old N. Benthamiana leaf blades. Leaf blades of these seedlings were incubated under middle light conditions (160 μmol quanta PAR m -2 s -1 ) for 2-4 days after infiltration. The firefly LUC activities were photographed after spraying with 1 mM luciferin. To determine the dualluciferase, activities of firefly luciferase and Renilla luciferase were measured, as has been previously described [32]. For LUC images, confocal laser scanning microscopy was performed by using a Epi-fluorescence microscopy (Olympus 80i, Olympus Corporation, Japan). The leaves were infiltrated with luciferin solution, and imaged by using a Tanon 5200S Luminescent Imaging Workstation.