Characterization of the 5-enolpyruvylshikimate-3-phosphate Synthase Gene from Walnut (Juglans regia L.)

Abstract

Walnut (Juglans regia L.) is a major nut crop of the Juglandaceae family and is well-known for its high nutritional value, which is achieved by a rich array of polyphenolic compounds. Phenolics are considered beneficial to human health because of their antioxidant, antimutagenic, and free radical scavenging properties. However, the phenolic biosynthetic pathway in walnut remains poorly studied. In this study, we cloned a 5-enolpyruvylshikimate 3-phosphate synthase (JrEPSPS) gene from walnut, a key gene involved in the shikimate pathway that catalyzes the penultimate step of the shikimate pathway toward the biosynthesis of aromatic amino acids. Subsequent sequence analysis revealed that the JrEPSPS protein harbors an N-terminal helix-turn-helix-like motif, which is known to mediate EPSPS function by acting as a transcription factor and regulating the expression of genes in the phenylpropanoid pathway in poplar. Subcellar localization analysis suggested JrEPSPS was localized in chloroplasts. The transient overexpression of JrEPSPS in persimmon (Diospyros kaki Thunb.) leaves and fruit discs showed significantly increased phenolic accumulation by elevating the expression of phenolic biosynthetic pathway genes. These results provide novel insights into the roles of EPSPS involved in phenolic biosynthesis in plants.

Introduction

Phenolic compounds are the most abundant secondary metabolites in plants (Rühmann et al., 2002; Treutter, 2001). These compounds are associated with the taste (e.g., bitterness and astringency) and quality (e.g., pigmentation, flavor, and oxidative stability) of fruits and vegetables (Pandey and Rizvi, 2009). Phenolics also play important roles in plant defense mechanisms triggered by biotic and abiotic stresses (Dixon and Paiva, 1995; Pandey and Rizvi, 2009). As the most ubiquitous antioxidant group in plants, phenolics are excellent for human health and can reduce the risk of cardiovascular diseases (Miller and Ruiz-Larrea, 2002; Pulido et al., 2000; Silva et al., 2004).

Phenolic compounds are biosynthesized through the shikimate pathway and are abundant in plants. The shikimate biosynthetic pathway is a common metabolic pathway through which the precursors of aromatic molecules are produced in microorganisms and plants, but are not present in animal cells (Mittelstädt et al., 2013). In this pathway, phosphoenolpyruvate and erythrose 4-phosphate are catalyzed to produce chorismate via seven consecutive steps (Bentley and Haslam, 1990). As one of the key enzymes involved in this pathway, 5-enolpyruvylshikimate-3-phosphate (EPSP) synthase (EPSPS, EC 2.5.1.19) catalyzes the penultimate step of the shikimate pathway toward the biosynthesis of aromatic acids. However, previous studies on EPSPS have mostly been related to its inhibition by the herbicide glyphosate (Baerson et al., 2002; Pline-Srnic, 2006; Schönbrunn et al., 2001). Akagi et al. (2009) found that EPSPS is highly expressed in persimmon fruit and that its expression is associated with tannin accumulation. Xie et al. (2018) isolated an EPSP isoform from poplar (Populus trichocarpa), named PtrEPSP-TF, which harbors an N-terminal helix–turn–helix (HTH) motif. PtrEPSP-TF regulates the transcription of genes involved in the phenylpropanoid pathway, in addition to its canonical catalytic function in the shikimate pathway. Interestingly, increased phenolic compound levels were detected in PtrEPSP-TF overexpression lines (Xie et al., 2018), indicating that this gene may also promote phenolic compound accumulation. Notably, the HTH motif is conserved in many dicots, indicating that the transcriptional regulatory function of EPSPS may also be conserved in other dicots, although this remains to be clarified.

Walnut (Juglans regia L.) is a major nut crop of the Juglandaceae family. As a highly nutritive and healthy nut, it is extensively cultivated in many regions of the world. The health benefits of walnut are endowed by its rich array of fatty acids, tocopherols, and polyphenols. Phenolics are considered potent components for reducing the risk of coronary heart diseases because of their remarkable antioxidant potential (Li et al., 2006). Extensive pharmacological studies have shown that in addition to the walnut seed (kernel), which is typically preferred for consumption, other parts of the plant such as the leaves, stems, and bark contain polyphenolic compounds (Pereira et al., 2008; Wang et al., 2015). In particular, the husk (exocarp and mesocarp of the walnut), which is considered to be agricultural waste, accumulates high levels of polyphenols with strong antioxidant potential (Fernández-Agulló et al., 2013; Yang et al., 2014). Trandafir et al. (2016) found that the mean total phenolic content (TPC) of the pellicle was 12.7-fold higher than that of the kernel, and pellicle removal reduced 90% of the total antioxidant activity potential of the kernel (Arcan and Yemenicioglu, 2009). As a dichotomous woody plant rich in unsaturated phenolic compounds, it may be interesting to elucidate the potential roles of EPSPS in the phenolic biosynthetic pathway.

An Agrobacterium-mediated genetic transformation system has been developed for walnut (McGranahan et al., 1988), but it usually takes a long time (5–6 years) to transform and observe fruit phenotypes. Persimmon (Diospyros kaki Thunb.) is a species that accumulates abundant polyphenols (the major type is proanthocyanidin, known as condensed tannin) in both its fruit and other tissues. The major polyphenolic compounds in walnut are hydrolysable tannins, but it also accumulates condensed tannins (Regueiro et al., 2014). The biosynthesis of both hydrolysable tannins and condensed tannins is derived from the shikimate pathway (Muir et al., 2011; Ossipov et al., 2003). Condensed tannins are produced through the shikimate pathway leading to the production of anthocyanins, a pathway that has been extensively investigated at the biochemical and genetic levels (Deavours and Dixon, 2005; Falcone Ferreyra et al., 2012). Hydrolysable tannins biosynthesis starts from gallic acid (Muir et al., 2011; Ossipov et al., 2003) which is produced from the shikimate pathway (Muir et al., 2011).

Recently, we developed an Agrobacterium-mediated transient transformation system in persimmon leaves and fruit discs to initially examine the role of candidate genes in polyphenol biosynthesis (Chen et al., 2021; Mo et al., 2015). Given the time-efficiency of this transient system and the similar polyphenol biosynthesis pathway in walnut, this system could be an alternative way to examine the function of walnut candidate genes in polyphenol biosynthesis.

To this end, in this study, JrEPSPS, a homologous gene of PtrEPSP-TF, was cloned from walnut. Consistent with previous reports, an HTH-like motif was detected at the N-terminus of JrEPSPS. In addition, JrEPSPS overexpression significantly increased TPC through direct or indirect upregulation of genes involved in phenolic biosynthesis. This study will further improve our understanding of the overall effects of EPSPS overexpression on plant phenotypes.

Materials and Methods

Plant materials

The husks and pellicles of the walnuts ‘Wen 185’ (Juglans regia L.) and ‘Xinxin 2’ cultivated in Wensu, Xinjiang, China, were obtained. The husk and pellicle samples were collected five times at different growth stages of the fruit from May 15 to September 15, 2018: fruit-bearing stage (S1), fruit expansion stage (S2), kernel enrichment stage (S3), hard-store stage (S4), and maturate stage (S5). The walnut fruit was transferred to the Huazhong Agricultural University (Wuhan, China) at a temperature below freezing for a short time. The walnut husks were peeled, kernels were removed, and pellicles were peeled. Both husks and pellicles were stored at −80°C after freezing in liquid nitrogen.

The leaves of persimmon (Diospyros kaki Thunb.) ‘Eshi 1’ and the fruits of ‘Gongcheng Shuishi’ cultivated in the orchard of Huazhong Agricultural University, were used for transient overexpression analysis.

RNA isolation and qRT-PCR

Total RNA was isolated from frozen tissue samples using the RNAplant Plus Reagent (Tiangen, Beijing, China) with three biological replicates. cDNA was synthesized using the PrimeScript RT Kit with a gDNA Eraser (TaKaRa Bio Inc., Dalian, China) according to the manufacturer’s protocol. qRT-PCR was performed with the QuantStudio 7 Flex Real-Time PCR system (Applied Biosystems, Foster City, CA, USA) using SYBR^® Premix Ex Taq^TM II (TaKaRa Bio). The primers used are listed in Table S1.

JrEPSPS isolation and characterization

The JrEPSPS sequence was derived from the walnut genome database (Martínez-García et al., 2016) using the JrEPSPS-F and JrEPSPS-R primers (Table S1). The full-length coding region of JrEPSPS was amplified by PCR using PrimerSTAR Max DNA Polymerase (TaKaRa Bio), cloned into the pEASY-Blunt Simple Cloning Vector (TransGen Biotech Co., Ltd., Beijing, China), and transformed into Trans1-T1 cells (TransGen Biotech) for sequencing.

RNA-seq data analysis

To profile the JrEPSPS expression across tissues, the RNA-seq data generated from 15 samples of walnut tissue (Table S2) by Chakraborty et al. (2016) were re-analyzed. Briefly, the download raw reads data were trimmed using fastp (Chen et al., 2018) with a default parameter. Then, the clean reads were mapped to the walnut genome (Marrano et al., 2020) with HISAT2 (Pertea et al., 2016), and the transcript abundances were estimated using StringTie (Pertea et al., 2016) in different samples.

Structural modeling analysis

JrEPSP models were built using the iterative threading assembly refinement (I-TASSER, v5.1) on-line server <https://zhanglab.ccmb.med.umich.edu/I-TASSER> (Yang and Zhang, 2015), with the Agrobacterium CP4 crystallographic structure as the template with default parameters.

Phylogenetic analysis

Phylogenetic analysis of the EPSPS sequences was performed using sequences retrieved from Phytozome <http://phytozome.jgi.doe.gov/pz/portal.html>. DNAMAN 8.0 (Lynnon Biosoft) was used for the multiple protein sequence alignment. A phylogenetic tree was constructed in MEGA 7 (Kumar et al., 2016) using the neighbor-joining method with 1,000 bootstrap replicates.

Binary vector construction and plant transformation

To generate binary vectors for JrEPSPS overexpression, cDNA was reverse transcribed from the RNA of ‘Wen 185’ using PCR with the JrEPSPS_attB-F and JrEPSPS_attB-R primers (Table S1). The resulting PCR products were cloned into pDONR221 (Invitrogen, Carlsbad, CA, USA) at the attB sites using Gateway^TM BP Clonase II Enzyme Mix (Invitrogen) and then into the pK2GW7 binary vector using Gateway LR Clonase^TM II Enzyme Mix (Invitrogen) to generate the JrEPSPS-pK2GW7 binary vector via recombination. The construct was verified using PCR and sequencing and then transformed into Agrobacterium tumefaciens GV3101 cells via heat shock. The GV3101 cells were allowed to infect the persimmon leaf and fruit discs for transient expression analysis.

For transient overexpression analysis in persimmon leaves, we followed the methods outlined in our previous paper (Mo et al., 2015). For persimmon fruit disc transient overexpression analysis, the fruit flesh of ‘Gongcheng Shuishi’ at the equator was made into discs with a diameter of 1 cm and a thickness of 0.5 cm. Samples transformed with an empty vector were used as a control (CK) (Chen et al., 2021). Briefly, the discs were incubated for 30 min with A. tumefaciens carrying constructs in the same buffer used for the persimmon leaf infection, then transferred to filter paper (with MS medium) in tissue-culture dishes and placed in an incubator at 25°C for two days. Then, the discs were dried on filter paper, frozen in liquid nitrogen, and stored at −80°C until further use. Given the relatively low efficiency of transforming persimmons, the status of leaves and developmental stage of fruit may affect the transformation efficiency. In our experience, the thicker leaves that directly grow on the branches and younger fruits have relatively higher transformation efficiency.

JrEPSPS subcellular localization

The coding sequence (CDS) of JrEPSPS without the stop codon was inserted into the pCAMBIA 1302 vector and fused with GFP under the control of the 35S promoter to generate the JrEPSPS-GFP construct. The fusion construct was then transformed into A. tumefaciens GV3101. The GV3101 cells containing the 35S: JrEPSPS-GFP construct were resuspended in infiltration buffer (10 mM MES, 10 mM MgCl₂, and 150 μM acetosyringone; pH 5.6) and infiltrated into tobacco (Nicotiana benthamiana) leaves using a needleless syringe. Following infiltration, the plants were kept at 24°C for 48 h. The GFP fluorescence was observed under a confocal laser scanning microscope (TCS SP8; Leica).

TPC quantification

TPC was measured colorimetrically with the Folin-Ciocalteu reagent following the method described by Singleton and Rossi (1965) with slight modification. Briefly, the samples (100 mg) were ground to a powder in liquid nitrogen and extracted twice with 500 μL MeOH for 1 h at room temperature (23–25°C). Following centrifugation at 12,000 × g for 5 min, the supernatant was diluted with distilled water up to 10 mL and used as the working solution. For TPC measurement, 1 mL working solution, 7.5 mL distilled water, and 0.5 mL phenol reagent were mixed, and 1 mL saturated Na₂CO₃ was added to the mixture after 5 min. After incubation for 1 h at room temperature, absorbance was measured at 765 nm (UV-2450; Shimadzu Corp., Kyoto, Japan); 8.5 mL water and reagents alone were used as the blank solution. The results were calculated according to the standard curve of gallic acid equivalents.

Statistical analysis

Differences between the genotypes were determined using the Student’s t-test with GraphPad Prism (GraphPad Software Inc., San Diego, CA, USA).

Results

JrEPSPS harbors an N-terminal HTH-like motif

Only one EPSPS gene (Jure_02874.t1) was identified in the walnut genome (Martínez-García et al., 2016). Based on the sequence, the CDS of Jure_02874.t1 from ‘Wen 185’ was isolated and named JrEPSPS. The length of the open reading frame of JrEPSPS was 1,569 bp, encoding 522 amino acids. To investigate the sequence divergence and detect potential domains, the amino acid sequence of JrEPSPS was aligned and compared with the sequences of EPSPS from other species. Consistent with a previous finding, JrEPSPS harbors an N-terminal HTH-like domain and is conserved in most dicots (Figs. 1A and S1) (Xie et al., 2018). Moreover, a start codon and a conserved MAQV(L/I)S(T) amino acid residue in the N-terminal were shared by dicots. However, this domain is divergent among species, and much more amino acid was missing in this region in the monocots, while it was completely lost in one of the homolog genes (XP_039787007.1) in Panicum virgatum. To confirm this finding, we further employed atomistic modeling and molecular dynamics simulations to characterize JrEPSPS with I-TASSER <https://zhanglab.ccmb.med.umich.edu/I-TASSER/> using the Agrobacterium CP4 crystallographic structure as the template. Characterization of the secondary structure ascending from the N terminus revealed a putative HTH motif spanning amino acid residues 40 to 100 of JrEPSPS (Fig. 1B). Moreover, a clear HTH structure was observed in the predicted tertiary structure (Fig. 1C). These results imply a potential regulatory function of JrEPSPS. To further investigate the evolutionary relationships, a phylogenetic tree of the EPSPS protein family was constructed by comparing JrEPSPS with EPSPS from sixteen other plant species. As previously observed by Xie et al. (2018), the phylogenetic relatedness of sequences reflects the broader classification delineating dicot and monocot clades of the kingdom Plantae (Fig. 1D). In addition, sequence alignments suggested that the divergence between dicots and monocots is mostly caused by the HTH domain in the N-terminal of ESPS, which implies that the clade of the phylogenetic tree could also reflect the diversity in the HTH motif.

Fig. 1

Structural and phylogenetic analysis of JrEPSPS. (A) ClustalW alignment of the amino acid sequences of JrEPSPS and other EPSPS homologous proteins. The HTH domain is indicated by a black bar. Dicots and monocots are indicated by red and green boxes, respectively. Different amino acid residues are indicated by different color shadings. Prediction of the secondary (B) and tertiary (C) structure of EPSPS, respectively. The HTH domain is indicated with H. (D) Phylogenetic tree of JrEPSPS and EPSPSs from seven other dicots species. The protein sequences used for multiple sequence alignment and phylogenetic tree construction were retrieved from Phytozome <http://phytozome.jgi.doe.gov/pz/portal.html> and National Center for Biotechnology Information <https://www.ncbi.nlm.nih.gov>.

JrEPSPS is localized in the chloroplasts

We investigated the subcellular localization of JrEPSPS in tobacco leaves using the 35S: JrEPSPS-GFP construct, the empty vector 35S, while GFP was used as a control (Fig. 2A). Chloroplasts could be visualized by red autofluorescence, and when the red image was merged with the green GFP fluorescence, the chloroplasts became bright yellow. The green GFP fluorescence also overlapped with the red autofluorescence of chloroplasts and returned a bright yellow signal (Fig. 2B), indicating that JrEPSPS localized in chloroplasts, which is consistent with the subcellular localization of most EPSPS proteins (Fiedler and Schultz, 1985).

Fig. 2

Subcellular localization of EPSPS in tobacco leaves. (A) Schematic diagrams of the constructs used for the subcellular localization assay. (B) Subcellular localization of JrEPSPS. Yellow arrows indicate the merged signals from chloroplasts and JrEPSPS: GFP. Since we did not check the signals of chloroplasts in the empty vector (35S: GFP), there is no chloroplast fluorescence in the empty vector panel. GFP, fluorescence of fusion construct; Chlorophyll, fluorescence of chloroplast; Merge, merge of GFP, Chlorophyll, and bright-field images.

TPC accumulation and JrEPSPS expression in walnut

To investigate the potential role of JrEPSPS, the association between phenolic accumulation and the JrEPSPS expression pattern was examined. The seasonal TPC and the corresponding JrEPSPS expression level of the husks and pellicles of the two cultivars tested were analyzed (Fig. 3A). In the husks, the TPC gradually decreased in both cultivars as fruit development progressed and was much higher in ‘Xinxin 2’ than in ‘Wen 185’ in all stages (Fig. 3B). In contrast to green husks, the TPC in the pellicles gradually increased and was much higher in ‘Wen 185’ than in ‘Xinxin 2’ in all developmental stages (Fig. 3C).

Fig. 3

Total phenolic content (TPC) and expression patterns of JrEPSPS in the husks and pellicles of ‘Xinxin 2’ and ‘Wen 185’. (A) Photographs of cultivars and tissues used for study. Changes in total phenolic content of the walnut husks (B) and pellicles (C), respectively. JrEPSPS expression patterns in the walnut husks (D) and pellicles (E), respectively. S1, S2, S3, S4, and S5 indicate the bearing fruit, fruit-expansion, kernel enrichment, hard store, and maturate stage, respectively. (F) RNA-seq analysis of the expression level of JrEPSPS in 15 samples of walnut tissue (Table S2). TPM, transcripts per kilobase million; n.d., not detected.

Next, we checked the expression of JrEPSPS in each tissue, and found the expression pattern of JrEPSPS was similar in the green husks of two cultivars and higher in ‘Xinin 2’ than ‘Wen 185’ in the all fruit development stages (Fig. 3D). However, JrEPSPS displayed a larger and different expression pattern in the pellicle (Fig. 3E) that was highly expressed in ‘Xinxin 2’ but less expressed in ‘Wen 185’ at S4. Then, we performed a correlation analysis between TPC accumulation and JrEPSPS expression in ‘Wen 185’ and ‘Xinxin 2’, respectively, using Pearson’s analysis. The correlation coefficients were 0.09 (P-value = 0.88) and 0.27 (P-value = 0.66) in green husks and were 0.10 (P-value = 0.34) and 0.49 (P-value = 0.39) in pellicles, respectively. These results indicate that there was no correlation between TPC accumulation and JrEPSPS expression. In addition, we further analyzed the expression level of EPSPS in 15 samples of walnut tissue (Table S2) using previously published RNA-seq data (Chakraborty et al., 2016), and found that JrEPSPS was highly expressed in the calluses and young leaves, but not expressed in pistillate flowers (Fig. 3F).

JrEPSPS overexpression increased the TPC of persimmon leaves

Walnut is a perennial fruit tree that takes years to generate transgenic fruit with stable transformation technologies. Meanwhile, the induction of persimmon leaves by A. tumefaciens is a relatively rapid and efficient method of gene function analysis (Chen et al., 2021; Guan et al., 2017; Mo et al., 2015; Zhu et al., 2018), and persimmon leaves accumulate high levels of phenolic compounds and express both positive and negative phenolic regulators. Thus, we first tested the JrEPSPS function in persimmon leaves. JrEPSPS overexpression in persimmon leaves significantly increased TPC compared with that of the empty vector control (CK) (Fig. 4A, B). To verify the prediction that JrEPSPS overexpression increased TPC by upregulating the phenolic biosynthetic pathway, the expression of structural genes involved in this pathway was examined using qRT-PCR. The transcript levels of 3-dehydroquinate dehydratase (DHD), phenylalanine ammonia-lyase (PAL), chalcone synthase (CHS), chalcone isomerase (CHI), flavonoid 3′-hydroxylase (F3′H), dihydroflavonol 4-reductase (DFR), and anthocyanidin synthase (ANS) increased (Fig. 4C). These results indicate that both early and late flavonoid biosynthesis-related genes were affected by JrEPSPS overexpression.

Fig. 4

Transient overexpression of JrEPSPS in persimmon leaves in vivo. (A) Analysis of JrEPSPS transcript level. (B) Determination of total phenolic content (TPC) of the leaves overexpressing JrEPSPS. (C) Expression of phenolic biosynthesis-related genes in persimmon leaves following JrEPSPS overexpression. ‘Eshi 1’ leaves were sampled seven days after agroinfiltration. OE and CK represent JrEPEPS-overexpressing and controls transformed with an empty vector, respectively. Error bars represent the SD (n > 10). The data correspond to the means ± SD relative to an ACTIN housekeeping control and normalized against the control value. Asterisks indicate significant differences compared with the control based on the Student’s t-test. * P < 0.05; ** P < 0.01.

JrEPSPS-overexpression significantly increased the TPC of persimmon fruit discs

To further validate the function of JrEPEPS in promoting phenolic biosynthesis, we transiently overexpressed JrEPSPS in persimmon fruit discs in vivo. The expression level of JrEPSPS was checked with qRT-PCR (Fig. 5A). We further measured the TPC of CK and JrEPSPS overexpression lines. The TPC of JrEPSPS overexpression lines was significantly increased compared with that of CK (Fig. 5B).

Fig. 5

Transient overexpression of JrEPSPS in ‘Gongcheng Shuishi’ persimmon fruit discs in vivo. (A) Analysis of JrEPSPS transcript level. (B) Determination of the total phenolic content (TPC) in JrEPSPS-overexpressor fruit discs; (C) Expression of polyphenolic biosynthesis-related structural genes in persimmon leaves after overexpression of JrEPSPS. Fruit discs infiltrated with JrEPSPS were sampled at two days after agroinfiltration. OE and CK represent JrEPEPS-overexpressing and controls transformed with an empty vector, respectively. Error bars represent the SD (n > 10). The data correspond to the means ± SD relative to an ACTIN housekeeping control and normalized against the control value. Asterisks indicate significant differences compared with the control based on Student’s t-test. * P < 0.05; ** P < 0.01.

To confirm that JrEPSPS promotes phenolic accumulation by upregulating the phenolic biosynthetic pathway, fruit discs overexpressing JrEPSPS were subjected to qRT-PCR. The expression levels of phenolic biosynthesis-related genes, specifically PAL, F3′H, and DFR, were increased in JrEPSPS-overexpression lines (Fig. 5C). Overall, the results of transient overexpression were comparable between persimmon leaves and fruit discs, suggesting that JrEPSPS promotes phenolic accumulation.

Discussion

Until the characterization of a poplar EPSPS (PtrEPSP-TF) paralog as a transcriptional regulator of the expression of genes involved in the phenylpropanoid pathway (Xie et al., 2018), EPSPSs were only considered to catalyze the production of EPSPS from shikimate-3-phosphate in the shikimate pathway in prokaryotes and eukaryotes (Maeda and Dudareva, 2012; Mir et al., 2015). The shikimate pathway is upstream of the phenylpropanoid pathway and produces precursors of aromatic molecules in microorganisms and plants (Mittelstädt et al., 2013; Tohge et al., 2013). Here, we showed that JrEPSPS, an EPSPS homolog in walnut, may have a similar function to PtrEPSP-TF (Fig. 1C), which was similar to that observed in PtrEPSP-TF (Xie et al., 2018) and may be conserved in most dicots (Fig. 1B, C). This motif is a common denominator of the basal and specific transcription factors from the three super kingdoms of life (Aravind et al., 2005). At its core, the domain comprises an open trihelical bundle, which typically binds DNA at the third helix (Aravind et al., 2005). In addition, this domain may also function as a DNA-binding motif, such as the cAMP-dependent catabolite activator protein (Ohlendorf et al., 1982; Sauer et al., 1982), regardless of the considerable diversity in terms of its sequence and binding specificity among organisms (Aravind and Iyer, 2012). Similar to specific transcription factors, the HTH domain also mediates the sigma factors to recognize DNA sequences (Gribskov and Burgess, 1986).

The most striking observation from the comparison of EPSPS between dicots and monocots was that the dicots shared a start codon and a conserved MAQV(L/I)S(T) amino acid residue in this additional exon (Fig. S1), and the HTH motif was almost missing in monocot clades. It is noteworthy that the results from phylogenic analysis found that two EPSPS homologs exist in some species (Fig. 1D), which may be the result of gene duplication in diploid species (Panchy et al., 2016), whereas polyploidization could be the reason for multiple EPSPS genes in polyploid species such as potatoes (Solanum tuberosum) and apples (Malus × domestica) (Glover et al., 2016). The evolutionary origin of the HTH motif has led to novel repressor activity in PtrEPSP-TF and a putative transcriptional regulatory function in JrEPSPS. It is intriguing that this shikimate pathway derived-EPSP synthase isoform appears to have acquired a regulatory function that modulates some pathway gene expressions in dicots relative to other plants; however, the presence of secondary cell walls is a key feature that separates dicots from algae and mosses, and the composition of monocot secondary cell walls also differs from that of dicots (Carpita and Gibeaut, 1993; Yokoyama and Nishitani, 2004). In this respect, it is possible that the domain co-option may have occurred when dicotyledonous plants attained complex cell wall structure rather than after species differentiation (Tohge et al., 2013; Weng et al., 2008; Xie et al., 2018). Therefore, the possible regulatory function of the EPSPS isoform in other dicots should be elucidated in the future.

Transient overexpression analysis in persimmon leaves and fruit discs elucidated the role of JrEPSPS in promoting phenolic accumulation (Figs. 4B and 5B). The subcellular location indicated that JrEPSPS localized in the chloroplasts (Fig. 2B), suggesting its conserved role in canonical catalytic function in the shikimate pathway. Due to the lack of a marker to indicate the nuclear localization signal, the subcellar localization results could not elucidate the potential nucleus localization of JrEPSPS. However, the homolog nuclear localization of PtrEPSPS-TF suggests that JrEPSPS also localizes in the nucleus because the homolog protein always has a similar function and subcellular localization. We propose that JrEPSPS may have TF activity, although the direct targets of JrEPSPS remain to be determined and the success of a transient assay was not fully validated in the present study, qRT-PCR suggested that JrEPSPS may directly or indirectly up-regulate the expression of gene expressions involved in the various phenolic pathways (Figs. 4C and 5C). Physiologically, high levels of monolignols are produced via phenolic pathways during fruit development. In the current transient overexpression analyses, we did not check whether or not the overexpression of JrEPSPS affected lignin biosynthesis. The precursors produced by EPSPS are shared the lignin and phenolic biosynthesis pathways, and Xie et al. (2018) also found that polyphenols including quercetins, dihydromyricetin, and catechins exhibited up to 2.8-fold increases in PtrEPSPS-TF overexpression lines. Thus, the function of JrEPSPS is partially consistent with PtrEPSPS-TF, indicating that the transcriptional regulatory function of EPSPS may also be conserved in other dicots, but this needs to be verified in other species and the molecular mechanism underlying JrEPSPS in phenolic biosynthesis requires much more direct evidence in the future.

Notably, no correlation was found between JrEPSPS expression and TPC accumulation in walnut green husks and pellicles, which may be partially caused by the initial position of EPSPS in secondary metabolism pathways that functions to produce the precursors of aromatic molecules for many other metabolites like lignin (Mittelstädt et al., 2013; Tohge et al., 2013).

In conclusion, the present characterization of the molecular mechanism linking JrEPSPS function to phenolic accumulation in walnuts provides novel insights into the function of EPSPSs in plants. As a plant species rich in unsaturated phenolic compounds, walnut may serve as a model system for woody dicot species to investigate the potential transcriptional regulatory function of EPSPSs and its involvement in phenolic biosynthesis in the future.

Literature Cited

•  Akagi,  T.,  A.  Ikegami,  Y.  Suzuki,  J.  Yoshida,  M.  Yamada,  A.  Sato and  K.  Yonemori. 2009. Expression balances of structural genes in shikimate and flavonoid biosynthesis cause a difference in proanthocyanidin accumulation in persimmon (Diospyros kaki Thunb.) fruit. Planta 230: 899–915.
•  Aravind,  L. and  L. M. J.  Iyer. 2012. The HARE-HTH and associated domains: novel modules in the coordination of epigenetic DNA and protein modifications. Cell Cycle 11: 119–131.
•  Aravind,  L.,  V.  Anantharaman,  S.  Balaji,  M. M.  Babu and  L. M.  Iyer. 2005. The many faces of the helix‐turn‐helix domain: Transcription regulation and beyond. FEMS Microbiol. Rev. 29: 231–262.
•  Arcan,  I. and  A.  Yemenicioglu. 2009. Antioxidant activity and phenolic content of fresh and dry nuts with or without the seed coat. J. Food Compos. Anal. 22: 184–188.
•  Baerson,  S. R.,  D. J.  Rodriguez,  M.  Tran,  Y. M.  Feng,  N. A.  Biest and  G. M.  Dill. 2002. Glyphosate-resistant goosegrass. Identification of a mutation in the target enzyme 5-enolpyruvylshikimate-3-phosphate synthase. Plant Physiol. 129: 1265–1275.
•  Bentley,  R. and  E.  Haslam. 1990. The shikimate pathway—a metabolic tree with many branche. Crit. Rev. Biochem. Mol. Biol. 25: 307–384.
•  Carpita,  N. C. and  D. M.  Gibeaut. 1993. Structural models of primary cell walls in flowering plants: consistency of molecular structure with the physical properties of the walls during growth. Plant J. 3: 1–30.
•  Chakraborty,  S.,  M.  Britton,  P. J.  Martínez-García and  A. M.  Dandekar. 2016. Deep RNA-Seq profile reveals biodiversity, plant-microbe interactions and a large family of NBS-LRR resistance genes in walnut (Juglans regia) tissues. AMB Express 6: 1–13.
•  Chen,  S. F.,  Y. Q.  Zhou,  Y. R.  Chen and  J.  Gu. 2018. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics 34: i884–i890.
•  Chen,  W. X.,  Q. Y.  Zheng,  J. W.  Li,  Y.  Liu,  L. Q.  Xu,  Q. L.  Zhang and  Z. R.  Luo. 2021. DkMYB14 is a bifunctional transcription factor that regulates proanthocyanidin accumulation in persimmon fruit. Plant J. 106: 1708–1727.
•  Deavours,  B. E. and  R. A.  Dixon. 2005. Metabolic engineering of isoflavonoid biosynthesis in alfalfa. Plant Physiol. 138: 2245–2259.
•  Dixon,  R. A. and  N. L.  Paiva. 1995. Stress-induced phenylpropanoid metabolism. Plant Cell 7: 1085–1097.
•  Falcone Ferreyra,  M. L.,  S. P.  Rius and  C.  Paula. 2012. Flavonoids: biosynthesis, biological functions, and biotechnological applications. Front. Plant Sci. 3: 222. DOI: 10.3389/fpls.2012.00222.
•  Fernández-Agulló,  A.,  E.  Pereira,  M. S.  Freire,  P.  Valentão,  P. B.  Andrade,  J.  González-Álvarez and  J. A.  Pereira. 2013. Influence of solvent on the antioxidant and antimicrobial properties of walnut (Juglans regia L.) green husk extracts. Ind. Crop. Prod. 42: 126–132.
•  Fiedler,  E. and  G.  Schultz. 1985. Localization, purification, and characterization of shikimate oxidoreductase-dehydroquinate hydrolyase from stroma of spinach chloroplasts. Plant Physiol. 79: 212–218.
•  Glover,  N. M.,  H.  Redestig and  C.  Dessimoz. 2016. Homoeologs: what are they and how do we infer them? Trends Plant Sci. 21: 609–621.
•  Gribskov,  M. and  R. R.  Burgess. 1986. Sigma factors from E. coli, B. subtilis, phage SP01, and phage T4 are homologous proteins. Nucleic Acids Res. 14: 6745–6763.
•  Guan,  C. F.,  X. Y.  Du,  Q. L.  Zhang,  F. W.  Ma,  Z. R.  Luo and  Y.  Yang. 2017. DkPK genes promote natural deastringency in C-PCNA persimmon by up-regulating DkPDC and DkADH expression. Front. Plant Sci. 8: 149. DOI: 10.3389/fpls.2017.00149.
•  Kumar,  S.,  G.  Stecher and  K.  Tamura. 2016. MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol. Biol. Evol. 33: 1870–1874.
•  Li,  L.,  R.  Tsao,  R. M.  Yang,  C. M.  Liu,  H. H.  Zhu and  J. C.  Young. 2006. Polyphenolic profiles and antioxidant activities of heartnut (Juglans ailanthifolia Var. cordiformis) and Persian walnut (Juglans regia L.). J. Agric. Food Chem. 54: 8033–8040.
•  Maeda,  H. and  N.  Dudareva. 2012. The shikimate pathway and aromatic amino acid biosynthesis in plants. Annu. Rev. Plant Biol. 63: 73–105.
•  Marrano,  A.,  M.  Britton,  P. A.  Zaini,  A. V.  Zimin,  R. E.  Workman,  D.  Puiu,  L.  Bianco,  E. A. D.  Pierro,  B. J.  Allen,  S.  Chakraborty,  M.  Troggio,  C. A.  Leslie,  W.  Timp,  A.  Dandekar,  S. L.  Salzberg and  D. B.  Neale. 2020. High-quality chromosome-scale assembly of the walnut (Juglans regia L.) reference genome. GigaScience 9: giaa050. DOI: 10.1093/gigascience/giaa050.
•  Martínez-García,  P. J.,  M. W.  Crepeau,  D.  Puiu,  D.  Gonzalez-lbeas,  J.  Whalen,  K. A.  Stevens,  R.  Paul,  T. S.  Butterfield,  M.  Britton,  R. L.  Reagan,  S.  Chakraborty,  S. L.  Walawage,  H. A.  Vasquez-Gross,  C.  Cardeno,  R. A.  Famula,  K.  Pratt,  S.  Kuruganti,  M. K.  Aradhya,  C. A.  Leslie,  A. M.  Dandekar,  S. L.  Salzberg,  J. L.  Wegrzyn,  C. H.  Langley and  D. B.  Neale. 2016. The walnut (Juglans regia) genome sequence reveals diversity in genes coding for the biosynthesis of non-structural polyphenols. Plant J. 87: 507–532.
•  McGranahan,  G. H.,  C. A.  Leslie,  S. L.  Uratsu,  L. A.  Martin and  A. M.  Dandeker. 1988. Agrobacterium-mediated transformation of walnut somatic embryos and regeneration of transgenic plants. Bio. Technol. 6: 800–804.
•  Miller,  N. J. and  M. B.  Ruiz-Larrea. 2002. Flavonoids and other plant phenols in the diet: their significance as antioxidants. J. Nutr. Environ. Med. 12: 39–51.
•  Mir,  R.,  S.  Jallu and  T. P.  Singh. 2015. The shikimate pathway: Review of amino acid sequence, function and three-dimensional structures of the enzymes. Crit. Rev. Microbiol. 41: 172–189.
•  Mittelstädt,  G.,  L.  Negron,  L. R.  Schofield,  K.  Marsh and  E. J.  Parker. 2013. Biochemical and structural characterisation of dehydroquinate synthase from the New Zealand kiwifruit Actinidia chinensis. Arch. Biochem. Biophys. 537: 185–191.
•  Mo,  R. L.,  Y. M.  Huang,  S. C.  Yang,  Q. L.  Zhang and  Z. R.  Luo. 2015. Development of Agrobacterium-mediated transient transformation in persimmon (Diospyros kaki Thunb.). Sci. Hortic. 192: 29–37.
•  Muir,  R. M.,  A. M.  Ibánez,  S. L.  Uratsu,  E. S.  Ingham,  C. A.  Leslie,  G. H.  McGranahan,  N.  Batra,  S.  Goyal,  J.  Joseph,  E. D.  Jemmis and  A. M.  Dandekar. 2011. Mechanism of gallic acid biosynthesis in bacteria (Escherichia coli) and walnut (Juglans regia). Plant Mol. Biol. 75: 555–565.
•  Ohlendorf,  D. H.,  W. F.  Anderson,  R. G.  Fisher,  Y.  Takeda and  B. W.  Matthews. 1982. The molecular basis of DNA-protein recognition inferred from the structure of cro repressor. Nature 298: 718–723.
•  Ossipov,  V.,  J. P.  Salminen,  S.  Ossipova,  E.  Haukioja and  K.  Pihlaja. 2003. Gallic acid and hydrolysable tannins are formed in birch leaves from an intermediate compound of the shikimate pathway. Biochem. Syst. Ecol. 31: 3–16.
•  Panchy,  N.,  M.  Lehti-Shiu and  S.  Shiu. 2016. Evolution of gene duplication in plants. Plant Physiol. 171: 2294–2316.
•  Pandey,  K. B. and  S. I.  Rizvi. 2009. Plant polyphenols as dietary antioxidants in human health and disease. Oxid. Med. Cell. Longev. 2: 270–278.
•  Pereira,  J. A.,  I.  Oliveira,  A.  Sousa,  I.  Ferreira,  A.  Bento and  L.  Estevinho. 2008. Bioactive properties and chemical composition of six walnut (Juglans regia L.) cultivars. Food Chem. Toxicol. 46: 2103–2111.
•  Pertea,  M.,  D.  Kim,  G. M.  Pertea,  J. T.  Leek and  S. L.  Salzberg. 2016. Transcript-level expression analysis of RNA-seq experiments with HISAT, StringTie and Ballgown. Nat. Protoc. 11: 1650–1667.
•  Pline-Srnic,  W. 2006. Physiological mechanisms of glyphosate resistance. Weed Technol. 20: 290–300.
•  Pulido,  R.,  L.  Bravo and  F.  Saura-Calixto. 2000. Antioxidant activity of dietary polyphenols as determined by a modified ferric reducing/antioxidant power assay. J. Agric. Food Chem. 48: 3396–3402.
•  Regueiro,  J.,  C.  Sánchez-González,  A.  Vallverdú-Queralt,  J.  Simal-Gándara,  R.  Lamuela-Raventós and  M.  Izquierdo-Pulido. 2014. Comprehensive identification of walnut polyphenols by liquid chromatography coupled to linear ion trap–Orbitrap mass spectrometry. Food Chem. 152: 340–348.
•  Rühmann,  S.,  C.  Leser,  M.  Bannert and  D.  Treutter. 2002. Relationship between growth, secondary metabolism, and resistance of apple. Plant Biology 4: 137–143.
•  Sauer,  R. T.,  R. R.  Yocum,  R. F.  Doolittle,  M.  Lewis and  C. O.  Pabo. 1982. Homology among DNA-binding proteins suggests use of a conserved super-secondary structure. Nature 298: 447–451.
•  Schönbrunn,  E.,  S.  Eschenburg,  W. A.  Shuttleworth,  J. V.  Schloss,  N.  Amrhein,  J. N. S.  Evans and  W.  Kabsch. 2001. Interaction of the herbicide glyphosate with its target enzyme 5-enolpyruvylshikimate 3-phosphate synthase in atomic detail. Proc. Natl. Acad. Sci. USA 98: 1376–1380.
•  Silva,  B. M.,  P. B.  Andrade,  P.  Valentão,  F.  Ferreres,  R. M.  Seabra and  M. A.  Ferreira. 2004. Quince (Cydonia oblonga Miller) fruit (pulp, peel, and seed) and jam: antioxidant activity. J. Agric. Food Chem. 52: 4705–4712.
•  Singleton,  V. L. and  J. A.  Rossi. 1965. Colorimetry of total phenolics with phosphomolybdic-phosphotungstic acid reagents. Am. J. Enol. Vitic. 16: 144–158.
•  Tohge,  T.,  M.  Watanabe,  R.  Hoefgen and  A. R.  Fernie. 2013. Shikimate and phenylalanine biosynthesis in the green lineage. Front. Plant Sci. 4: 62. DOI: 10.3389/fpls.2013.00062.
•  Trandafir,  I.,  S.  Cosmulescu,  M.  Botu and  V.  Nour. 2016. Antioxidant activity, and phenolic and mineral contents of the walnut kernel (Juglans regia L.) as a function of the pellicle color. Fruits 71: 177–184.
•  Treutter,  D. 2001. Biosynthesis of phenolic compounds and its regulation in apple. Plant Growth Regul. 34: 71–89.
•  Wang,  X.,  M. M.  Zhao,  G. W.  Su,  M. S.  Cai,  C. M.  Zhou,  J. Y.  Huang and  L. J.  Lin. 2015. The antioxidant activities and the xanthine oxidase inhibition effects of walnut (Juglans regia L.) fruit, stem and leaf. Food Sci. Technol. 50: 233–239.
•  Weng,  J. K.,  X.  Li,  N. D.  Bonawitz and  C.  Chapple. 2008. Emerging strategies of lignin engineering and degradation for cellulosic biofuel production. Curr. Opin. Biotechnol. 19: 166–172.
•  Xie,  M.,  W.  Muchero,  A. C.  Bryan,  K.  Yee,  H. B.  Guo,  J.  Zhang,  T. J.  Tschaplinski,  V. R.  Singan,  E.  Lindquist,  R. S.  Payyavula,  J.  Barros-Rios,  R.  Dixon,  N.  Engle,  R. W.  Sykes,  M.  Davis,  S. S.  Jawdy,  L. E.  Gunter,  O.  Thompson,  S. P.  DiFazio,  L. M.  Evans,  K.  Winkeler,  C.  Collins,  J.  Schmutz,  H.  Guo,  U.  Kalluri,  M.  Rodriguez,  K.  Feng,  J. G.  Chen and  G. A.  Tuskan. 2018. A 5-enolpyruvylshikimate 3-phosphate synthase functions as a transcriptional repressor in Populus. Plant Cell 30: 1645–1660.
•  Yang,  J. and  Y.  Zhang. Protein structure and function prediction using I-TASSER. 2015. Curr. Protoc. Bioinformatics 52: 1–15.
•  Yang,  J.,  C. Y.  Chen,  S. L.  Zhao,  F.  Ge and  D. Q.  Liu. 2014. Effect of solvents on the antioxidant activity of walnut (Juglans regia L.) shell extracts. J. Food Nutr. Res. 2: 621–626.
•  Yokoyama,  R. and  K.  Nishitani. 2004. Genomic basis for cell-wall diversity in plants. A comparative approach to gene families in rice and Arabidopsis. Plant Cell Physiol. 45: 1111–1121.
•  Zhu,  Q. G.,  Z. Y.  Gong,  M. M.  Wang,  X.  Li,  D.  Grierson,  X. R.  Yin and  K. S.  Chen. 2018. A transcription factor network responsive to high CO₂/hypoxia is involved in deastringency in persimmon fruit. J. Exp. Bot. 69: 2061–2070.