Citrus CitNAC62 cooperates with CitWRKY1 to participate in citric acid degradation via up-regulation of CitAco3

Two novel transcription factors, CitNAC62 and CitWRKY1, were involved in citric acid degradation in citrus fruit, potentially via enhanced expression of CitAco3.


Introduction
Organic acids, including quinic, citric, malic, and oxalic acids, are present in most plants and vary among species, organ, and tissue types, developmental stages, and environmental conditions (Badia et al., 2015). In Arabidopsis, organic acids influence carbohydrate perception in germinating seedlings (Hooks et al., 2004), fumarate accumulation plays an essential role in low temperature sensing (Dyson et al., 2016), malate is involved in cellular pH regulation and stomatal movement (Hurth et al., 2005;Lee et al., 2008), and citrate contributes to metal resistance in plant roots .
Organic acid metabolism and degradation have been widely studied. For instance, MxCS2, a gene encoding a putative citrate synthase in Malus xiaojinensis, was introduced into Arabidopsis, resulting in increased citrate content (Han et al., 2015). In contrast, inhibition of aconitase activity resulted in the accumulation of citrate (Gupta et al., 2012;Hooks et al., 2014). In addition to biosynthesis and degradation, some transporters, including a tonoplast dicarboxylate transporter (AttDT) (Hurth et al., 2005), aluminum-activated malate transporter (ALMT) (Kovermann et al., 2007), and some V-ATPase/V-PPase genes Hu et al., 2016), also influence organic acid content in plants. In citrus, a vacuolar citrate/H + symporter was isolated that could mediate citrate efflux and play a role in citric acid homeostasis (Shimada et al., 2006). In recent years, some transcription factors have been demonstrated to have important roles in the regulation of organic acids. In Arabidopsis, WRKY46 functions as a transcriptional repressor of ALMT1, regulating aluminuminduced malate secretion (Ding et al., 2013). In tomato fruits, overexpression of SlAREB1 resulted in increased citric and malic acid contents, and the expression of the mitochondrial citrate synthase gene (mCS) was up-regulated (Bastías et al., 2011), while CgDREB-overexpressing tomato fruits showed higher levels of organic acids (Nishawy et al., 2015). However, transcriptional regulatory information is still very limited.
In citrus fruit, especially acidic varieties, citric acid is the predominant organic acid, accounting for more than 90% of total organic acids (Albertini et al., 2006;Baldwin et al., 2014). The difference in the acidity of various citrus fruits at the commercial mature stage is due to expansion of the fruit, citrate synthesis and vacuole storage, and is also largely determined by the degradation pathway, including the gamma-aminobutyric acid (GABA) shunt and the glutamine and acetyl-CoA pathways Walker et al., 2011;Lin et al., 2015). Among these, the GABA shunt was considered to be the dominant pathway; the first step of this pathway is the conversion of citrate to isocitrate by aconitase (Terol et al., 2010). In citrus fruits, inhibition of mitochondrial aconitase activity contributes to acid accumulation, and increasing cytosolic aconitase activity reduces the citrate level toward fruit maturation (Degu et al., 2011;Sadka et al., 2000). Transcript analysis from multiple sources indicated that CitAco3 is negatively correlated with citric acid content in citrus fruit and CitAco3 may contribute to citrate degradation (Chen et al., 2012(Chen et al., , 2013. However, understanding of the molecular basis of fruit citrate degradation has been limited to transcript analysis, including CitAco3 and the other structural genes. Because of the difficulty of producing transgenic citrus material, the in planta roles of these genes in citrate degradation, and the in vivo mechanisms regulating their transcripts, remain unknown.
In the present research, gene expression and partial functional verification of CitAco3 in relation to citrate degradation were studied. In order to understand the regulation of CitAco3 expression, a set of 16 transcription factors was isolated on the basis of their co-expression with CitAco3. The potential regulatory roles of the transcription factors were investigated and two of them showed transactivation activity of the CitAco3 promoter. In addition, the interaction and synergistic effects of two transcription factors, protein-protein interaction, and the possible movement of transcription factors within the plant cell were evaluated with regard to citrate degradation.

Plant materials
Ponkan (Citrus reticulata Blanco cv. Ponkan) fruits received from a commercial orchard in Quzhou, Zhejiang, China, were used in this study. Fruits of uniform size and appearance were collected at each sampling point, from six different trees. Sampling points were at 60, 90, 120, 150, and 180 days after full blossom (DAFB). The flesh was frozen in liquid nitrogen and stored at −80 °C for further experiments.

Citric acid measurement
The citric acid content of Ponkan fruits and leaves was measured according to Lin et al. (2015). Fruits (0.1 g) and leaves (0.05 g) were ground in liquid nitrogen and extracted with 1.4 ml methanol at 70 °C for 15 min, and then centrifuged at 10 000 g. The upper phase was removed and stored at −80 °C until analysis. Aliquots of 100 μl upper phase were dried in a vacuum. The residue was dissolved in 40 μl 20 mg ml −1 pyridine methoxyamine hydrochloride, and incubated at 37 °C for 1.5 h. The sample was then treated with 60 μl Bis(trimethylsilyl)trifluoroacetamide (1% trimethylchlorosilane) at 37 °C for 30 min. Ribitol (20 μl, 0.2 mg ml −1 ) was added to each sample as an internal standard. A 1 μl aliquot of each sample was absorbed with a split ratio of 1:1 and injected into a GC-MS fitted with a fused-silica capillary column (30 m×0.25 mm internal diameter, 0.25 μm DB-5 MS stationary phase). The injector temperature was 250 °C and the helium carrier gas had a flow rate of 1.0 ml min −1 . The column temperature was held at 100 °C for 1 min, increased to 184 °C at a rate of 3 °C min −1 , then increased to 230 °C at a rate of 15 °C min −1 and held for 1 min. The MS operating parameters were as follows: ionization voltage 70 eV, ion source temperature as 230 °C, and interface temperature 280 °C.

RNA extraction and cDNA synthesis
Total RNA was extracted from frozen tissues according to the protocol described by Chen et al. (2012). The genomic DNA in total RNA was degraded with RNase-free DNase I (Ambion). First-strand cDNA synthesis was initiated with 1.0 μg DNA-free RNA and GoScript™ Reverse Transcriptase (Promega) following the manufacturer's protocol. Ten-fold diluted cDNA was used as the template for quantitative real-time PCR analysis. RNA extraction and cDNA synthesis were performed with three biological replicates for each sampling point.

Real-time PCR
The PCR mixture (20 μl total volume) comprised 10 μl Lightcycler480 SYBR Green I Master (Roche), 1 μl of each primer (10 mM), 2 μl diluted cDNA and 6 μl PCR-grade H 2 O. PCR was performed on a Lightcycler 480 instrument (Roche), initiated by 5 min at 95 °C, followed by 50 cycles of 95 °C for 10 s, 60 °C for 10 s, and 72 °C for 15 s, and completed with a melting curve analysis program. No-template controls and melting curve analyses were included in every reaction. Citrus actin (XM_006464503) was used as a control to quantify cDNA abundance (Chen et al., 2012). The sequences of the primers used are described in Supplementary Table S1 at JXB online.

Dual luciferase assays
Dual luciferase assays were performed as described in our previous reports (Xu et al., 2014). The promoter of the CitAco3 gene was amplified with the primers described in Supplementary Table S2. Full-length transcripts of the transcription factors (Table 1) were inserted into the pGreen II 0029 62-SK vector (SK) with the primers described in Supplementary Table S2, while the promoter was inserted into the pGreen II 0800-LUC vector. Details of the vectors are given in Hellens et al (2005).
All the constructs were electroporated into Agrobacterium tumefaciens GV3101. The dual luciferase assays were performed with tobacco (Nicotiana benthamiana) leaves. Agrobacterium cultures were prepared with infiltration buffer (10 mM MES, 10 mM MgCl 2 , 150 mM acetosyringone, pH 5.6) to an OD 600 of 0.7-1.0. Mixtures of cultures of Agrobacterium expressing transcription factors (1 ml) and the promoter (100 µl) were infiltrated into tobacco leaves using needleless syringes. The tobacco plants were grown in a glasshouse with daylight extension to 16 h. Three days after infiltration, firefly luciferase and renilla luciferase were assayed using dual luciferase assay reagents (Promega). For each transcription factor-promoter interaction, three independent experiments were performed (five biological replicates in each experiment).

Subcellular localization analysis
35S-CitNAC62-GFP and 35S-CitWRKY1-GFP were transiently expressed in tobacco leaves by Agrobacterium-mediated infiltration (GV3101) according to previous reports with some modification (Xu et al., 2014). The green fluorescent protein (GFP) fluorescence of tobacco leaves was imaged 3 d after infiltration using a Zeiss LSM710NLO confocal laser scanning microscope. Primers used for GFP construction are described in Supplementary Table S3.
Yeast two-hybrid assay Protein-protein interactions were investigated in yeast with the DUAL hunter system (Dual-systems Biotech, Switzerland). Fulllength coding sequences of CitWRKY1 were cloned into the pDHB1 vector as bait, and the full length of CitNAC62 was cloned into pPR3N vector as prey. The primers used for vector construction are described in Supplementary Table S4.
All constructs were transformed into the yeast strain NMY51 according to the manufacturer's instructions. The assays were performed with different media: (i) SD medium lacking Trp and Leu (DDO); (ii) SD medium lacking Trp, Leu, His, and Ade (QDO); and (iii) SD medium lacking Trp, Leu, His, and Ade, and supplemented with 60 mM 3-amino-1,2,4-triazole (QDO+3AT). Auto-activations were tested with empty pPR3-N vectors and target genes with pDHB1, which were co-transformed in NMY51 and plated on QDO. Autoactivations were indicated by the presence of colonies. Protein-protein interaction assays were performed with co-transformation of CitNAC62 in pPR3N and CitWRKY1 in pDHB1. The presence of colonies in QDO and QDO+3AT indicated a protein-protein interaction.

Bimolecular fluorescence complementation assay
Full-length CitNAC62 and full-length CitWRKY1 were cloned into either C-terminal or N-terminal fragments of yellow fluorescent protein (YFP) vectors (Sainsbury et al., 2009). Primers used are listed in Supplementary Table S4. All constructs were transiently expressed in tobacco leaves by Agrobacterium-mediated infiltration (GV3101) based on previous reports with some modifications . The YFP fluorescence of tobacco leaves was imaged 3 d after infiltration using a Zeiss LSM710NLO confocal laser scanning microscope.

Transient overexpression in citrus leaves and fruits
Full-length coding sequences of target genes (CitAco3, CitNAC62, and CitWRKY1) were amplified with primers (listed in Supplementary  Table S5) and inserted into the SK vector. Information regarding the SK vector is given in Hellens et al. (2005). The constructs were electroporated into Agrobacterium GV3101. For transient overexpression in leaves, Agrobacterium cultures carrying empty vector (SK) or target genes were infiltrated into different sides of the same leaf. In fruit, two uniform sections were chosen from one Ponkan fruit, and were infiltrated with Agrobacterium cultures carrying empty vector (SK) or target genes, respectively. Five days after infiltration, the infiltrated leaves and sections were sampled and used for citric acid analysis.

Statistical analysis
Least significant difference (LSD) was calculated by using DPS 7.05 (Zhejiang University, Hangzhou, China). The statistical significance of differences was calculated using Student's t-test.

Association between CitAco3 and citrate degradation
The correlation of CitAco3 expression and citric acid degradation has been widely supported (Chen et al., 2012; al., 2015). In the present study, we found that CitAco3 is more abundant in late developmental stages (150 and 180 DAFB), when the fruit citric acid decreased from a peak of 32.07 mg g −1 at 120 DAFB to 6.51 mg g −1 at 180 DAFB (Fig. 1A, B). To directly investigate CitAco3 function, we introduced a cDNA, under the control of the constitutive CaMV 35S promoter, into citrus leaves and fruits using Agrobacteriummediated transient transformation (Hellens et al., 2005). Compared with the control (empty vector), transient overexpression of CitAco3 significantly reduced the citric acid content in citrus leaves and fruits. In leaves transformed with CitAco3 or the empty vector, citric acid contents were 1.16 and 1.74 mg g −1 , respectively ( Fig. 2A). Similar results were observed in citrus fruits, where transient overexpression of CitAco3 significantly reduced citric acid content to 12.11 mg g −1 , compared with the empty vector, at 15.52 mg g −1 (Fig. 2B).

In vivo regulatory effects of transcription factors the on CitAco3 promoter
In order to study the transcriptional regulation of CitAco3, we searched the RNA-Seq data from our previous report (Lin et al., 2015) to identify 16 transcription factors whose abundance was highly correlated with CitAco3 (Table 1). Dual luciferase assays indicated that in the presence of CitNAC62 or CitWRKY1, CitAco3 promoter activity was significantly enhanced, with approximately 2.4-and 2.0-fold induction, respectively (Fig. 3).
Analysis of CitNAC62 and CitWRKY1 expression indicated that both transcription factors had expression patterns similar to that of CitAco3, being more abundant at the late stages of fruit development (Fig. 4).

Subcellular localization and interaction of CitNAC62 and CitWRKY1
To visualize the subcellular locations of the two transcription factors, we performed a subcellular localization assay in tobacco leaves by using GFP tagging. CitWRKY1 gave strong signals in the nucleus (Fig. 5); CitNAC62 was not located in the nucleus and the signals indicated that its subcellular location was within plastids (Fig. 5). Despite the different locations of the two transcription factors, protein-protein interactions were observed between CitNAC62 and CitWRKY1 in yeast two-hybrid assays (Fig. 6A). This interaction was also verified by bimolecular fluorescence complementation assays (BiFC) using tobacco leaves. The results showed that negative combinations, such as YFP N /CitNAC62-YFP C , CitWRKY1-YFP N /YFP C , and YFP N /YFP C did not produce any detectable fluorescence signal, while co-expression of CitNAC62-YFP C and CitWRKY1-YFP N gave strong signals in the nucleus (Fig. 6B).

Citric acid content is negatively regulated by CitNAC62 and CitWRKY1
CitNAC62 and CitWRKY1, under the control of the CaMV 35S promoter, were introduced into citrus fruits using  Agrobacterium-mediated transient transformation (Hellens et al., 2005). Compared with an empty vector control, transient overexpression of CitNAC62 and CitWRKY1 significantly decreased the citric acid content in citrus fruits, with values of 13.61 and 13.98 mg g −1 , respectively, compared with 18.37 mg g −1 for the empty vector control. Transient overexpression of the   combination of CitNAC62 and CitWRKY1 resulted in lower citric acid content in citrus fruits, at 10.59 mg g −1 (Fig. 7A). Transient overexpression of CitNAC62 or CitWRKY1 significantly increased CitAco3 abundance (Fig. 7B). Furthermore, co-introduction of both CitNAC62 and CitWRKY1 resulted in even lower citric acid content and higher CitAco3 expression (Fig. 7), indicating that the two transcription factors can act in combination to increase the level of CitAco3 and decrease the citric acid content.

Discussion
CitAco3 is a contributor to citric acid degradation Multiple reports have correlated gene expression with citric acid degradation in citrus fruit, including an aconitase gene, CitAco3 (Chen et al., 2013;Lin et al., 2015). In the present study, the association of CitAco3 and citric acid degradation was confirmed during Ponkan fruit development. However, owing to the difficulty of transformation in perennial fruit such as citrus, validation of the function of CitAco3 has not been performed. With the development of a citrus transtransformation system Yin et al., 2016), we have now shown that transient overexpression of CitAco3 led to lower citric acid content in citrus fruit and leaves, supporting a role for CitAco3 in citric acid degradation. A similar function for Aco3 has been reported in other plants, including Arabidopsis (Hooks et al., 2014) and tomato (Morgan et al., 2013). The present results support the potential role of CitAco3 in citric acid degradation in citrus fruit.
Transcription factors CitNAC62 and CitWRKY1 up-regulate CitAco3 transcript abundance and decrease citric acid content In recent years, molecular and genetic studies have identified numerous transcription factors participating in the regulation of fruit quality . For instance, AP2/ERF transcription factors are involved in citrus fruit degreening (CitERF13; Yin et al., 2016) and volatile metabolism (CitAP2.10; Shen et al., 2016); and PavMYB10.1 is involved in anthocyanin biosynthesis in sweet cherry fruit (Jin et al., 2016). For organic acid metabolism, an EIN3-like transcription factor was characterized as the regulator of the ALMT1-like protein in apples (Bai et al., 2015). In addition, Fig. 6. (A) Interaction between CitWRKY1 and CitNAC62 in yeast two-hybrid assays. Liquid cultures of double transformants were plated at OD 600 =0.1 dilutions on synthetic dropout selective media: (i) SD medium lacking Trp and Leu (DDO); (ii) SD medium lacking Trp, Leu, His and Ade (QDO); and (iii) SD medium lacking Trp, Leu, His, and Ade, and supplemented with 60 mM 3-amino-1,2,4-triazole (QDO+3AT). Protein-protein interactions were determined on QDO and QDO+3AT. pOst1-NubI, positive control; pPR3-N, negative control. (B) In vivo interaction between CitNAC62 and CitWRKY1, determined using BiFC. N-and C-terminal fragments of YFP (indicated on the figure as YN and YC) were fused to the C-terminus of CitNAC62 and CitWRKY1, respectively. Combinations of YC or YN with the corresponding CitNAC62 and CitWRKY1 constructs were used as negative controls. Fluorescence of YFP represents protein-protein interaction. Bars=50 µm. MdMYB1 in apple fruits could activate the expression of two vacuolar H + -ATPase genes (MdVHA-B1 and MdVHA-B2), affecting malate accumulation (Hu et al., 2016). However, transcriptional regulation of citrate-related genes is largely unexplored. Here, we showed that CitNAC62 and CitWRKY1 regulate CitAco3 transcript abundance in vivo. Furthermore, transient overexpression of CitNAC62 and CitWRKY1 resulted in lower citric acid content in citrus fruit. Thus, we propose that CitNAC62 and CitWRKY1 are negative regulators of citric acid content, acting via up-regulation of the CitAco3 promoter.

Protein-protein interaction between CitNAC62 and CitWRKY1 also involves translocation
An interesting finding was the protein-protein interaction between CitNAC62 and CitWRKY1, which suggests that the complex of transcription factors may contribute to citric acid degradation. Protein-protein interaction between transcription factors has been widely demonstrated in many plants, including fruit species. For example, MYBs, bHLHs, and WD40s have been shown to act together in a ternary regulatory MYB-BHLH-WD40 complex in order to regulate target genes, especially in anthocyanin biosynthesis (Schaart et al., 2013), and EjAP2-1 regulates lignin biosynthesis via interaction with EjMYB1 and EjMYB2 in loquat fruits . However, such an interaction has not been reported for the regulation of organic acid metabolism. Thus, the impact of the interaction of CitNAC62 and CitWRKY1 on citric acid degradation may be only moderate (according to the transient overexpression data), but the interaction provides a novel clue about citric acid regulation.
BiFC analysis indicated that interaction between CitNAC62 and CitWRKY1 happens in the nucleus, but subcellular localization analysis indicated that only CitWRKY1, and not CitNAC62, is located within the nucleus. These results suggested that CitWRKY1 may translocate CitNAC62 to the nucleus. Translocation of genes by protein-protein interactions plays important roles in plants. In Arabidopsis, AtEBP may move from the nucleus to the cytoplasm via protein-protein interaction with ACBP4 (Li et al., 2008); in rice, OsSPX4 could prevent OsPHR2 from being targeted to the nucleus through its interaction with OsPHR2 when phosphate is sufficient (Lv et al., 2014). The present findings suggest that translocation of CitNAC62 may also contribute to citric acid degradation; however, the specific role of this translocation requires further investigation. In particular, the role and mechanism of CitWRKY1 for translocation, as well as the triggers of translocation, are unclear, and it is important to evaluate the function of such translocation in citric acid degradation.

Supplementary data
Supplementary data are available at JXB online. Table S1. Primers for real-time quantitative PCR analysis. Table S2. Primers used for amplification of the promoter of CitAco3 and SK construction. Table S3. Primers used in subcellular localization analysis. Table S4. Primers for yeast two-hybrid and BiFC assays. Table S5. Primers used in transient overexpression analysis.