Isolation and Characterization of an LBD Transcription Factor CsLBD39 from Tea Plant (Camellia sinensis) and Its Roles in Modulating Nitrate Content by Regulating Nitrate-Metabolism-Related Genes

Nitrate nitrogen is an important nitrogen source for tea plants’ growth and development. LBD transcription factors play important roles in response to the presence of nitrate in plants. The functional study of LBD transcription factors in tea plants remains limited. In this study, the LBD family gene CsLBD39 was isolated and characterized from tea plants. Sequence analysis indicated that CsLBD39 contained a highly conserved CX2CX6CX3CX domain. The phylogenetic tree assay showed that CsLBD39 belonged to class II subfamily of the LBD family. CsLBD39 was highly expressed in flowers and root; we determined that its expression could be induced by nitrate treatment. The CsLBD39 protein was located in the nucleus and has transcriptional activation activity in yeast. Compared with the wild type, overexpression of CsLBD39 gene in Arabidopsis resulted in smaller rosettes, shorter main roots, reduced lateral roots and lower plant weights. The nitrate content and the expression levels of genes related to nitrate transport and regulation were decreased in transgenic Arabidopsis hosting CsLBD39 gene. Compared with the wild type, CsLBD39 overexpression in transgenic Arabidopsis had smaller cell structure of leaves, shorter diameter of stem cross section, and slender and compact cell of stem longitudinal section. Under KNO3 treatment, the contents of nitrate, anthocyanins, and chlorophyll in leaves, and the content of nitrate in roots of Arabidopsis overexpressing CsLBD39 were reduced, the expression levels of nitrate transport and regulation related genes were decreased. The results revealed that CsLBD39 may be involved in nitrate signal transduction in tea plants as a negative regulator and laid the groundwork for future studies into the mechanism of nitrate response.


Introduction
Nitrogen (N) is an important macronutrient necessary for the normal growth and development of higher plants [1,2]. For most plants, nitrate (NO 3 − ) is the primary source of nitrogen and can be assimilated to nitrite, ammonium and amino acids [3]. Nitrate serves as an essential nutrient and an important signaling molecule [4]. Nitrate is the most established and probably the dominant nitrogen signal that modulates root architecture, leaf development, and anthocyanin accumulation [5][6][7]. The nitrate transport (NRT) gene families, nitrate reductase (NIA) and nitrite reductase (NIR) genes are involved in nitrate transport and assimilation [8,9]. Some studies have demonstrated that NLP, LBD, NRG, and other transcription factors play vital roles in regulating nitrate metabolism [1,10,11].
LBD (Lateral organ boundaries domain) gene family is one of the plant-specific transcription factor (TF) families. LBD TFs have a highly conserved LOB (Lateral organ bound-

Sequence and Phylogenetic Tree Analysis of the CsLBD39
CsLBD39 was isolated from tea plant 'Longjing 43'. Sequence analysis showed that CsLBD39 gene was 687 bp in length and encoded 228 amino acids. Multiple sequence alignments showed that CsLBD39 and other LBDs had a typical zinc finger domain (CX 2 CX 6 CX 3 CX) ( Figure 1A). In order to understand the classification of CsLBD39, the sequence of CsLBD39 and the LBDs of Arabidopsis were used to construct a phylogenetic tree. The results showed that CsLBD39 belongs to the class II subfamily ( Figure 1B).

Relative Expression Level of CsLBD39 in Tea Plant
The transcript levels of CsLBD39 in different developmental stages and tissues were determined. The results showed that CsLBD39 gene was expressed in all tested tissues, and the expression levels were higher in flowers and roots (Figure 2A). CsLBD39 gene was affected by different concentrations of nitrate, and its expression reached the maximum at 1 mM KNO3 treatment ( Figure 2B).

Relative Expression Level of CsLBD39 in Tea Plant
The transcript levels of CsLBD39 in different developmental stages and tissues were determined. The results showed that CsLBD39 gene was expressed in all tested tissues, and the expression levels were higher in flowers and roots (Figure 2A). CsLBD39 gene was affected by different concentrations of nitrate, and its expression reached the maximum at 1 mM KNO 3 treatment ( Figure 2B). The relative expression levels CsLBD39 in the root after adding 0.1, 1, 10 mM KNO3 to N-limited tea plant seedlings. The data expressed as mean ± standard deviation of three replicates (n = 3). Different lowercase letters in cate significant differences at p < 0.05.

Subcellular Localization and Transcriptional Activation Activity Analysis of CsLBD39
Studying where a protein is expressed is essential to determine its function [2 CLBD39 was fused with GFP to construct recombinant vector CsLBD39-GFP. The reco binant plasmid, CsLBD39-GFP, was bombarded into onion epidermal cells using the ge gun to observe the subcellular localization. The result found that pA7-GFP fluoresce signal permeated the onion cell, and CsLBD39-GFP fusion protein is expressed in the n cleus ( Figure 3A). To detect the transcriptional activation activity of CsLBD39, CsLBD39 was c structed into pGBKT7 vector containing GAL4-binding domain to obtain the yeast expr sion vector pGBKT7-CsLBD39. Positive control (pCL1), negative plasmid (pGBKT7), a pGBKT7-CsLBD39 were transferred into Y2H yeast receptor cells, respectively. The ye The relative expression levels of CsLBD39 in the root after adding 0.1, 1, 10 mM KNO 3 to N-limited tea plant seedlings. The data are expressed as mean ± standard deviation of three replicates (n = 3). Different lowercase letters indicate significant differences at p < 0.05.

Subcellular Localization and Transcriptional Activation Activity Analysis of CsLBD39
Studying where a protein is expressed is essential to determine its function [27]. CLBD39 was fused with GFP to construct recombinant vector CsLBD39-GFP. The recombinant plasmid, CsLBD39-GFP, was bombarded into onion epidermal cells using the gene gun to observe the subcellular localization. The result found that pA7-GFP fluorescence signal permeated the onion cell, and CsLBD39-GFP fusion protein is expressed in the nucleus ( Figure 3A). CsLBD39 in the root after adding 0.1, 1, 10 mM KNO3 to N-limited tea plant seedlings. The expressed as mean ± standard deviation of three replicates (n = 3). Different lowercase let cate significant differences at p < 0.05.

Subcellular Localization and Transcriptional Activation Activity Analysis of CsLBD
Studying where a protein is expressed is essential to determine its funct CLBD39 was fused with GFP to construct recombinant vector CsLBD39-GFP. The binant plasmid, CsLBD39-GFP, was bombarded into onion epidermal cells using gun to observe the subcellular localization. The result found that pA7-GFP fluo signal permeated the onion cell, and CsLBD39-GFP fusion protein is expressed in cleus ( Figure 3A). To detect the transcriptional activation activity of CsLBD39, CsLBD39 w structed into pGBKT7 vector containing GAL4-binding domain to obtain the yeas sion vector pGBKT7-CsLBD39. Positive control (pCL1), negative plasmid (pGBK To detect the transcriptional activation activity of CsLBD39, CsLBD39 was constructed into pGBKT7 vector containing GAL4-binding domain to obtain the yeast expression vector pGBKT7-CsLBD39. Positive control (pCL1), negative plasmid (pGBKT7), and pGBKT7-CsLBD39 were transferred into Y2H yeast receptor cells, respectively. The yeast strains transformed with pCL1 were cultured on SD/Leu − solid medium, the other two were cultured on SD/Trp − solid medium, respectively. The positive yeast screened by SD/Leu − and SD/Trp − were inoculated on SD/His − Ade − deficient medium with or without X-α-Gal, respectively. The results showed that pCL1 and pGBKT7-CsLBD39 could grow on SD/His − Ade − +X-α-Gal solid medium and showed blue color, while pGBKT7 could not grow on SD/His − Ade − +X-α-Gal solid medium, indicating that CsLBD39 had transcriptional activation activity in yeast ( Figure 3B).

Overexpression of CsLBD39 in Arabidopsis
The CsLBD39 gene was inserted into pCAMBIA1301 vector carrying β-glucuronidase (GUS) reporter gene, allowing that CsLBD39 and GUS separately driven by CaMV 35S promoter, so that CsLBD39 and GUS could co-express in transgenic plants (Supplementary Figure S1A). GUS staining were performed for the identification of transgenic Arabidopsis, finding that the cotyledons and roots of 7-day-old Arabidopsis showed blue color (Supplementary Figure S1B). GUS gene was expressed in filaments, anthers, stigmas, sepals and siliques of Arabidopsis (Supplementary Figure S1D). The cDNAs of WT and transgenic Arabidopsis were amplified by PCR to further identify the expression of CsLBD39 in transgenic plants, showing that the corresponding bands could be detected in the transgenic lines (Supplementary Figure S1C). Then, the RT-qPCR assay also indicated that CsLBD39 was overexpressed in transgenic Arabidopsis plants (Supplementary Figure S1E).

Changes in Fresh Weight and Roots of Transgenic Arabidopsis Overexpressing CsLBD39
Transgenic Arabidopsis was cultured in MS medium with different concentrations (0.2, 1, and 5 mM) of KNO 3 for 15 days ( Figure 4A). The fresh weight of transgenic lines was significantly lower than that of wild type (WT) at 1 mM and 5 mM KNO 3 treatments ( Figure 4B). The morphology of taproots and lateral roots of Arabidopsis were observed after 15 d ( Figure 5A). The results showed that under 0.2 mM and 1 mM KNO 3 treatments, the transgenic taproots were shorter than the WT ( Figure 5B). Under the treatment of KNO 3 at three concentrations, the number of lateral roots of transgenic Arabidopsis was less than that of the WT, especially the number of lateral roots of OE-1 was significantly lower than that of the WT ( Figure 5C). strains transformed with pCL1 were cultured on SD/Leu − solid medium, the other two were cultured on SD/Trp − solid medium, respectively. The positive yeast screened by SD/Leu − and SD/Trp − were inoculated on SD/His − Ade − deficient medium with or without X-α-Gal, respectively. The results showed that pCL1 and pGBKT7-CsLBD39 could grow on SD/His − Ade − +X-α-Gal solid medium and showed blue color, while pGBKT7 could not grow on SD/His − Ade − +X-α-Gal solid medium, indicating that CsLBD39 had transcriptional activation activity in yeast ( Figure 3B).

Overexpression of CsLBD39 in Arabidopsis
The CsLBD39 gene was inserted into pCAMBIA1301 vector carrying β-glucuronidase (GUS) reporter gene, allowing that CsLBD39 and GUS separately driven by CaMV 35S promoter, so that CsLBD39 and GUS could co-express in transgenic plants (Supplementary Figure S1A). GUS staining were performed for the identification of transgenic Arabidopsis, finding that the cotyledons and roots of 7-day-old Arabidopsis showed blue color (Supplementary Figure S1B). GUS gene was expressed in filaments, anthers, stigmas, sepals and siliques of Arabidopsis (Supplementary Figure S1D). The cDNAs of WT and transgenic Arabidopsis were amplified by PCR to further identify the expression of CsLBD39 in transgenic plants, showing that the corresponding bands could be detected in the transgenic lines (Supplementary Figure S1C). Then, the RT-qPCR assay also indicated that CsLBD39 was overexpressed in transgenic Arabidopsis plants (Supplementary Figure S1E).

Changes in Fresh Weight and Roots of Transgenic Arabidopsis Overexpressing CsLBD39
Transgenic Arabidopsis was cultured in MS medium with different concentrations (0.2, 1, and 5 mM) of KNO3 for 15 days ( Figure 4A). The fresh weight of transgenic lines was significantly lower than that of wild type (WT) at 1 mM and 5 mM KNO3 treatments ( Figure 4B). The morphology of taproots and lateral roots of Arabidopsis were observed after 15 d ( Figure 5A). The results showed that under 0.2 mM and 1 mM KNO3 treatments, the transgenic taproots were shorter than the WT ( Figure 5B). Under the treatment of KNO3 at three concentrations, the number of lateral roots of transgenic Arabidopsis was less than that of the WT, especially the number of lateral roots of OE-1 was significantly lower than that of the WT ( Figure 5C).

Analysis of Nitrate, Anthocyanin and Chlorophyll Contents in Transgenic Arabidopsis
Overexpressing CsLBD39 Gene 35-day-old Arabidopsis plant was used to test the nitrate content ( Figure 6A). The nitrate content in the leaves and roots of the transgenic Arabidopsis was lower than that of the WT, especially in leaves ( Figure 6B). The content of anthocyanins was affected by nitrogen stress in plants. Here, determined total anthocyanins content in transgenic Arabidopsis leaves was significantly reduced ( Figure 6C). We observed that the leaves of transgenic Arabidopsis were light green and those of the WT was dark green ( Figure 6A). The content of chlorophyll a and chlorophyll b in transgenic Arabidopsis were reduced compared to the WT Arabidopsis ( Figure 6D).

Analysis of Nitrate, Anthocyanin and Chlorophyll Contents in Transgenic Arabidopsis
Overexpressing CsLBD39 Gene 35-day-old Arabidopsis plant was used to test the nitrate content ( Figure 6A). The nitrate content in the leaves and roots of the transgenic Arabidopsis was lower than that of the WT, especially in leaves ( Figure 6B). The content of anthocyanins was affected by nitrogen stress in plants. Here, determined total anthocyanins content in transgenic Arabidopsis leaves was significantly reduced ( Figure 6C). We observed that the leaves of transgenic Arabidopsis were light green and those of the WT was dark green ( Figure 6A). The content of chlorophyll a and chlorophyll b in transgenic Arabidopsis were reduced compared to the WT Arabidopsis ( Figure 6D).

Expression Analysis of Nitrate Uptake and Transport-Related Genes in Transgenic Arabidopsis Plants Overexpressing CsLBD39 Gene
The effect of overexpression of CsLBD39 gene on the expression of nitrate transportrelated genes was analyzed. As showed in Figure 7, the expression levels of several nitrate transport genes, such as AtNRT1.1, AtNRT1.4, AtNRT1.6, AtNRT1.11, AtNRT1.13, AtNRT2.7, AtNIA1, and AtNIA2 were significantly lower in transgenic Arabidopsis leaves than in the WT. Similarly, several TFs, such as AtNLP5, AtNLP6, AtNLP9, and AtLBD37, also showed a downward trend.

Expression Analysis of Nitrate Uptake and Transport-Related Genes in Transgenic Arabidopsis Plants Overexpressing CsLBD39 Gene
The effect of overexpression of CsLBD39 gene on the expression of nitrate transportrelated genes was analyzed. As showed in Figure 7, the expression levels of several nitrate transport genes, such as AtNRT1.1, AtNRT1.4, AtNRT1.6, AtNRT1.11, AtNRT1.13, AtNRT2.7, AtNIA1, and AtNIA2 were significantly lower in transgenic Arabidopsis leaves than in the WT. Similarly, several TFs, such as AtNLP5, AtNLP6, AtNLP9, and AtLBD37, also showed a downward trend. The expression levels of nitrate transport genes AtNRT1.1, AtNRT1.4, AtNIA2, AtNRT1.7 and nitrate response TFs AtNLP5, AtLBD37 in transgenic Arabidopsis roots were lower than WT. The expression levels of nitrate transport related genes AtNRT1.11, AtNRT2.2 and nitrate response TFs AtNLP2, AtNLP4 and AtNLP9 in transgenic Arabidopsis roots were significantly higher than WT. These results suggested that overexpression of Figure 7. The expression levels of nitrate response related genes in WT and transgenic Arabidopsis plants hosting CsLBD39 gene. The data are expressed as mean ± standard deviation of three biological replicates (n = 3). AtSAND was used as reference gene. Asterisks (*) indicate that the value is significant difference compared to the WT (* p < 0.05; ** p < 0.01; *** p < 0.001).
The expression levels of nitrate transport genes AtNRT1.1, AtNRT1.4, AtNIA2, AtNRT1.7 and nitrate response TFs AtNLP5, AtLBD37 in transgenic Arabidopsis roots were lower than WT. The expression levels of nitrate transport related genes AtNRT1.11, AtNRT2.2 and nitrate response TFs AtNLP2, AtNLP4 and AtNLP9 in transgenic Arabidopsis roots were significantly higher than WT. These results suggested that overexpression of CsLBD39 gene leads to changes in the expression of nitrate responsive genes in Arabidopsis plants.

Analysis of Nitrate, Anthocyanins and Chlorophyll Contents in Transgenic Arabidopsis Overexpressing CsLBD39 under Nitrate Treatment
A detailed summary of Arabidopsis growth and treatment conditions is shown in Figure 8A, the WT and transgenic Arabidopsis were grown in the cultivation medium for 25 d and then transferred to KCl and KNO 3 hydroponic nutrient solution for seven days ( Figure 8B). The nitrate content in the leaves of the transgenic lines decreased after treatment, and the nitrate content under the KNO 3 treatment was higher than that under the KCl treatment at seven days ( Figure 9A). The same trend was observed in roots. The nitrate content in the roots of the transgenic plants was significantly reduced after treatment for seven days, the nitrate content in Arabidopsis roots under the KNO 3 treatment was higher than that under the KCl treatment ( Figure 9B). Nitrogen deficiency in plants will cause stress responses, which will affect the synthesis of anthocyanins. The anthocyanins content of Arabidopsis increased at seven days of KCl treatment, and the anthocyanins accumulation of transgenic lines was lower than that of WT. After seven days of KNO 3 treatment, anthocyanins content in Arabidopsis plants increased compared with 0 d of KNO 3 treatment, and transgenic lines also showed lower anthocyanins accumulation in contrast to WT. The anthocyanins content of Arabidopsis plants treated with KNO 3 was still lower than that treated with KCl ( Figure 10A). content of Arabidopsis increased at seven days of KCl treatment, and the anthocyanins accumulation of transgenic lines was lower than that of WT. After seven days of KNO3 treatment, anthocyanins content in Arabidopsis plants increased compared with 0 d of KNO3 treatment, and transgenic lines also showed lower anthocyanins accumulation in contrast to WT. The anthocyanins content of Arabidopsis plants treated with KNO3 was still lower than that treated with KCl ( Figure 10A).     The Arabidopsis plants in KCl treatment group showed more yellow leaves compared to that in KNO3 treatment group. The chlorophyll contents were measured. The results showed that the contents of chlorophyll a and chlorophyll b in the transgenic lines were significantly lower than those in the WT plants at 0 d of the KCl or KNO3 treatment. At seven days, both chlorophyll a and chlorophyll b of WT and transgenic Arabidopsis treated with KNO3 were higher than those treated with KCl, especially the chlorophyll a is signif- The Arabidopsis plants in KCl treatment group showed more yellow leaves compared to that in KNO 3 treatment group. The chlorophyll contents were measured. The results showed that the contents of chlorophyll a and chlorophyll b in the transgenic lines were significantly lower than those in the WT plants at 0 d of the KCl or KNO 3 treatment. At seven days, both chlorophyll a and chlorophyll b of WT and transgenic Arabidopsis treated with KNO 3 were higher than those treated with KCl, especially the chlorophyll a is significantly increased (Figure 10B-D).

Cytological Observation on Leaves and Stems of Transgenic Arabidopsis
The transgenic Arabidopsis plants overexpressing the CsLBD39 gene showed dwarfing and small rosette leaves in this study. The cytological morphological changes were further observed and analyzed. Leaves and stems of WT and transgenic Arabidopsis treated with KCl were selected for observation. The results showed that the phloem and xylem tissues of transgenic Arabidopsis leaves were smaller than that of WT ( Figure 11). This phenomenon was also observed in the stem cell section of transgenic Arabidopsis. The diameter of stem cells was shortened and the cells became significantly smaller. The longitudinal observations of the stem showed that the cells in the transgenic Arabidopsis stem were small and compact ( Figure 11). Under KNO 3 treatment, the results of cell sections were similar to those of under KCl treatment ( Figure 12). Regardless of the leaves or stems, the cells of transgenic Arabidopsis are reduced and compact, and the diameter of the stem cross section was also smaller. The cells of stem longitudinal section become slender and denser.

The Expression Analysis of Nitrate Uptake and Transport-Related Genes in Transgenic Arabidopsis Plants Overexpressing CsLBD39 under Nitrate Treatment
Arabidopsis plants were cultured in KCl and KNO 3 hydroponic nutrient solution for seven days and sampled for RT-qPCR experiments. As is shown in Figure 13, the expression levels of AtNRT1.1, AtNRT1.6, AtNRT2.1, AtNRT2.7, AtNLP5, AtNLP7, AtNIA2, AtLBD37, and AtLBD39 in transgenic lines were significantly lower than those in WT plants both under KCl treatment and KNO 3 treatment. The expression levels of AtNRT1.2, AtNRT1.5, AtNRT1.9, AtNRT1.11, AtNRT1.13, AtNRT2.4, AtNLA, AtNLP2, and AtNLP8 genes were significantly higher in transgenic lines than those in WT plants both under KCl treatment and KNO 3 treatment.
The expression levels of genes related to nitrate response in roots were different under KCl and KNO 3 treatments. As is shown in Figure 14, the expression levels of AtNRT1.1, AtNRT2.1, AtNRT2.2, AtNLP5, AtNLP7, AtNLP8, and AtNLP9 in Arabidopsis roots of KCl treatment group were different from that of KNO 3 treatment group, that is, the expression levels of these genes in transgenic lines were lower than those in WT plants under KCl treatment, whereas the results were opposite under KNO 3 treatment. The expression levels of AtNRT1.4, AtNRT1.5, AtNRT1.7, and AtNRT2.7 were decreased in transgenic lines than in WT under KNO 3 treatment, and the opposite results were found under KCl treatment.
Arabidopsis plants were cultured in KCl and KNO3 hydroponic nutrient solution for seven days and sampled for RT-qPCR experiments. As is shown in Figure 13, the expression levels of AtNRT1.1, AtNRT1.6, AtNRT2.1, AtNRT2.7, AtNLP5, AtNLP7, AtNIA2, At-LBD37, and AtLBD39 in transgenic lines were significantly lower than those in WT plants both under KCl treatment and KNO3 treatment. The expression levels of AtNRT1.2, AtNRT1.5, AtNRT1.9, AtNRT1.11, AtNRT1.13, AtNRT2.4, AtNLA, AtNLP2, and AtNLP8 genes were significantly higher in transgenic lines than those in WT plants both under KCl treatment and KNO3 treatment. Figure 13. The expression levels of nitrate response related genes in leaves of WT and transgenic Arabidopsis plants hosting CsLBD39 gene under KCl and KNO3 conditions. The data are expressed as mean ± standard deviation of three biological replicates (n = 3). Asterisks (*) indicate that the value is significant difference compared to the WT (* p < 0.05; ** p < 0.01; *** p < 0.001).
The expression levels of genes related to nitrate response in roots were different under KCl and KNO3 treatments. As is shown in Figure 14, the expression levels of AtNRT1.1, AtNRT2.1, AtNRT2.2, AtNLP5, AtNLP7, AtNLP8, and AtNLP9 in Arabidopsis roots of KCl treatment group were different from that of KNO3 treatment group, that is, the expression levels of these genes in transgenic lines were lower than those in WT plants under KCl treatment, whereas the results were opposite under KNO3 treatment. The expression levels of AtNRT1.4, AtNRT1.5, AtNRT1.7, and AtNRT2.7 were decreased in  conditions. The data are expressed as mean ± standard deviation of three biological replicates (n = 3). Asterisks (*) indicate that the value is significant difference compared to the WT (* p < 0.05; ** p < 0.01; *** p < 0.001).

Discussion
As one of the main nutrients required by plants, nitrogen regulates many aspects of plant growth, development and metabolism. In some higher plants, inorganic nitrogen is mainly composed of two forms, NO3 − and NH4 + , and nitrate is the preferential nitrogen source for most higher plants [28,29]. In tea plants, the absorption rate of ammonium ni- Figure 14. The expression levels of nitrate response related genes in roots of WT and transgenic Arabidopsis plants hosting CsLBD39 gene under KCl and KNO 3 conditions. The data are expressed as mean ± standard deviation of three biological replicates (n = 3). Asterisks (*) indicate that the value is significant difference compared to the WT (* p < 0.05; ** p < 0.01; *** p < 0.001).

Discussion
As one of the main nutrients required by plants, nitrogen regulates many aspects of plant growth, development and metabolism. In some higher plants, inorganic nitrogen is mainly composed of two forms, NO 3 − and NH 4 + , and nitrate is the preferential nitrogen source for most higher plants [28,29]. In tea plants, the absorption rate of ammonium nitrogen is higher than that of nitrate nitrogen [25]. The excessive application of ammonium nitrogen will cause soil acidification. Therefore, the research on the absorption and utilization of nitrate is also particularly important. The remobilization of nitrate between different organs is mainly mediated by nitrate transporters (NRTs) [30][31][32]. Previous studies have reported that overexpression of the AtLBD TF genes suppressed the expression of NRT and NR genes, thus controlling N utilization in Arabidopsis [11]. However, the LBD TFs that regulate nitrate-responsive genes in tea plants have not been studied so far. Searching for LBD TFs that regulate nitrate uptake and assimilation in tea plants is helpful for future molecular breeding in relevant fields.
LBD TFs play significant roles in plant growth, development, and metabolism [11,33,34]. Based on previous studies, LBD is classified as class I and class II [13]. In this work, sequence analysis showed that CsLBD39 belonged to class II subfamily of the LBDs and was homologous to AtLBD39 in Arabidopsis. In Arabidopsis, the expression of class II LBD genes, LBD37/LBD38/LBD39, are induced by nitrogen or glutamine [11]. Overexpression of LBD37/LBD38/LBD39 genes inhibited the expression of NRT and NR genes, and changed the contents of nitrogen, nitrate and amino acids [11]. In this study, the expression of CsLBD39 was induced by nitrate, the nitrate content was reduced, and the expression of NRT genes related to nitrate transport were inhibited in transgenic plants overexpressing CsLBD39. A similar phenomenon was found in apples, overexpression of MdLBD13 altered the nitrate content and the expression of genes related to N metabolism in apple and Arabidopsis [16]. Studies have shown that AtLBD16, AtLBD29, and AtLBD18 regulate the formation of lateral roots [35,36]. The plant weight and root length of Arabidopsis overexpressing CsLBD39 gene were changed under different KNO 3 treatments. Overexpressing CsLBD39 gene in Arabidopsis altered the root morphology under KNO 3 treatment. These results suggested that CsLBD39 may act as a regulator to modulate the growth and development of plants under KNO 3 treatment.
Yordanov and Busov proposed a mechanism model for the regulation of LBD in secondary woody growth, that is, PtaLBD1 and PtaLBD4 are expressed at the cambium/phloem boundary, could regulate secondary phloem development by inhibiting the expression of ARBORKNOX1 and ARBORKNOX2 genes, and could activate APL and other genes transcription to promote phloem development [37,38]. In Eucalyptus grandis, overexpression of EgLBD37 gene resulted in some changes in the phenotype of the transgenic plants, namely, the plant became taller, the leaves became larger, the length of the internodes increased, the diameter of the stem increased, the total width of the cortical area and the xylem components of the secondary xylem increased significantly [39]. In contrast, the most pronounced phenotype of the EgLBD29 transgenic plants was that all transgenic lines exhibited smaller plant height, reduced internode length and declined leaf size [39]. Similar reports have been found in this study, overexpression of the CsLBD39 gene in Arabidopsis resulted in smaller and dwarf plants. Changes in plant phenotypes can cause cytological changes [40]. Further observation and analysis of cytological morphological changes in leaves and stems of transgenic Arabidopsis overexpressing CsLBD39 gene found that the diameter of transgenic Arabidopsis stems was shortened and the cells in leaf and stem sections were smaller. These results suggested that CsLBD39 can affect plant growth and development.

Plant Materials, Growth Conditions
Tea plant cultivar 'Longjing 43' and wild type Arabidopsis 'Columbia' were selected as materials. 'Longjing 43' was planted in artificial climate room of the State Key Laboratory of Crop Genetics and Germplasm Enhancement of Nanjing Agricultural University. The condition of artificial climate room was 25/18 • C and 16/8 h of light/dark, with 70% relative humidity. The growing medium of tea plants is a mixture of peat, vermiculite and perlite (3:2:1; v/v). Arabidopsis plants was grown in the illumination incubator with the environment of 22/18 • C and 14/10 h of light/dark, as well as 70% relative humidity. The growing medium is a mixture of nutrient soil, vermiculite and perlite (18:6:1; v/v).
The young leaves (YL), mature leaves (ML), old leaves (OL), stems, flowers and roots of healthy tea plant with semblable physiological conditions were collected to analyze the expression of CsLBD39 gene. One-year-old tea plant cuttings were transferred into a total nutrient solution as described by Zhang et al. [26]. The tea plants were cultivated for six weeks of normal N supply (2 mM). Subsequently, the tea plants were placed in a culture medium (without N, as CK) for 10 days, and then transferred to different KNO 3 treatments with 0, 0.1, 1, and 10 mM. The tea roots treated with different KNO 3 concentrations as mentioned above were collected after 2 h, frozen in liquid nitrogen, and stored at −80 • C for RT-qPCR tests. All samples were set up for three biological replicates.

RNA Extraction and cDNA Synthesis
The total RNA of tea plant and Arabidopsis samples were extracted using RNA extraction kit (Huayueyang, China; Pudi, China), and then the total RNA was reverse transcribed into cDNA using the HiScript II Q RT SuperMix for qPCR kit (Vazyme, Nanjing, China).

Isolation and Bioinformatics Analysis of CsLBD39
The sequence of CsLBD39 was downloaded from Tea Plant Information Archive (TPIA) (http://tpia.teaplant.org/index.html) (accessed on 17 January 2020) database [41]. The gene was cloned from 'Longjing 43' by a pair of primers (forward: 5 -ATGAGTTGCAATGGATGTCG-3 and reverse: 5 -TCAGGTGAACAAGTTTAGAAG-3 ) through polymerase chain reaction (PCR). The PCR product was first linked to the pMD19-T vector and then sequenced. Homologous LBD protein sequences and others were obtained using NCBI (https://www.ncbi.nlm.nih.gov/) (accessed on 2 April 2020) and Plant TFDB (http://planttfdb.gao-lab.org/index.php) (accessed on 2 April 2020). The MUSCLE program of MEGA 5 was used to carry out multiple alignments of protein sequences, and then phylogenetic trees were generated by the Neighbor-Joining method [42].

Subcellular Localization of CsLBD39
To confirm subcellular localization of CsLBD39, a pair of specific primers (forward: 5 -CACCATCACCATCACGCCATGATGAGTTGCAATGGATGTCG-3 and reverse: 5 -CACTAGTACGTCGACCATGGCGGTGAACAAGTTTAGAAG-3 ) was used to clone CsLBD39 without stop codon. The PCR product was inserted into pA7 vector via Nco I site. Subsequently, the fusion construct (35S:CsLBD39-GFP) was generated. The 35S:CsLBD39-GFP plasmid and the pA7 plasmid were separately bombarded into the onion epidermal cells (PDS-1000, Bio-Rad, Hercules, CA, USA) and then placed on MS medium in the dark condition [43]. After 14 h, the GFP expression signals was observed using a confocal laser scanning microscope (Zeiss, Germany) and photographed.

Transcriptional Activation Activity Analysis of CsLBD39
To verify the transcriptional activation activity of CsLBD39, a pair of specific primers (forward: 5 -ATGGCCATGGAGGCCGAATTCATGAGTTGCAATGGATGTCG-3 and reverse: 5 -ATGCGGCCGCTGCAGGTCGACTCAGGTGAACAAGTTTAGAAG-3 ) were used to clone CsLBD39. The PCR product was insert into the pGBKT7 vector via EcoR I and Sal I sites to generate a recombinant construct (pGBKT7-CsLBD39). Subsequently, the empty vector (pGBKT7, as the negative control), pCL1 plasmid (as the positive control), and pGBKT7-CsLBD39 were transformed into yeast strain Y2H, respectively. The yeast strains transformed with pCL1 plasmid was cultured on SD/Leu − medium, while the yeast strains hosing pGBKT7-CsLBD39 or pGBKT7 were cultured on SD/Trp − medium, respectively. After 3 d, positive clones were selected and inoculated on SD/His − Ade − medium containing X-α-gal to examine whether they turned blue.

Overexpression Plasmid Construction and Transformation
The full length CsLBD39 ORF was cloned using a pair of specific primers (forward: 5 -TTTACAATTACCATGGGATCCATGAGTTGCAATGGATGTCG-3 and reverse: 5 -ACCGATGATACGAACGAGCTCTCAGGTGAACAAGTTTAGAAG-3 ) and insert into the Sac I and BamH I sites of pCAMBIA1301 vector that containing the β-glucosidase (GUS) gene to construct the recombinant plasmid pCAMBIA1301-CsLBD39. The expression of CsLBD39 and GUS genes was driven by the 35S promoter, respectively. Simply put, the recombinant plasmid pCAMBIA1301-CsLBD39 was introduced into Agrobacterium tumefaciens strain GV3101. The Arabidopsis was transformed by A. tumefaciens-mediated genetic transformation using flower dipping method [44]. Transgenic Arabidopsis was screened on 1/2 MS medium containing hygromycin and carbenicillin. The transgenic lines were verified by GUS staining and PCR amplification tests.

Nitrate Treatment Conditions in Arabidopsis
WT and transgenic Arabidopsis seeds were plated on MS solid medium. The MS plate was placed in an illumination incubator for cultivation. Arabidopsis seedlings grown in MS medium for seven days were transferred to the cultivation medium. One month later, part of the plants was transferred to nutrient solution containing 1 mM KNO 3 for seven days, and the other part was transferred to nitrogen free nutrient solution for seven days, KCl was used to control the difference in K + concentration. Arabidopsis leaves after treatment were collected for RT-qPCR assay, anthocyanins, chlorophyll and nitrate contents determination. Arabidopsis roots were collected for RT-qPCR assay and nitrate contents determination.
MS nitrogen-free medium was purchased from PhytoTech LABS [45]. KNO 3 was used as the sole nitrogen source. The final concentrations of adding KNO 3 in MS nitrogen-free medium were 0.2 mM, 1 mM, and 5 mM. KCl with final concentrations of 4.8 mM, 4 mM, and 0 mM was added to MS nitrogen-free medium to supplement the corresponding concentration of K + . The seeds of WT and transgenic Arabidopsis were placed on the above-mentioned MS medium to evaluate the effects of KNO 3 treatments at different concentrations on Arabidopsis plant fresh weight and root length.

Measurement of the Nitrate Content
WT and transgenic Arabidopsis were planted in a mixed substrate. 35-day-old Arabidopsis leaves and roots were collected for determination of nitrate content.
Briefly, 0.2 g of freeze-dried sample was added with deionized water and the mixture was boiled and centrifuged. The obtained supernatant was transferred into a new centrifuge tube, and salicylic acid-sulfuric acid solution was first added to mix, and then NaOH solution was added to react, cooling the reaction liquid to room temperature. The absorbance of reaction mixture was measured using microplate reader (Spectramax ID5) at 410 nm [1]. Three replicates were conducted.

Determination of Chlorophyll
WT and transgenic Arabidopsis were planted in a mixed substrate. 35-day-old Arabidopsis leaves were collected for determination of chlorophyll content.
The extraction and determination of chlorophyll (Chl) were carried out with reference to previous studies [46]. Briefly, the leaves are cut into pieces, 0.1 g fresh leaves added with 10 mL of the mixed extract (95% acetone: ethanol: distilled water = 4.5:4.5:1) and soaked in the dark for 24 h until the leaves turn completely white. The mixed extract was used as a blank control, the absorbance was measured by Spectramax ID5 at 645 nm and 663 nm, respectively. Three replicates were conducted.

Determination of Anthocyanins
WT and transgenic Arabidopsis were planted in a mixed substrate and grown to 35 d of age, and leaves were collected for determination of anthocyanins content.
The total content of anthocyanins in Arabidopsis leaves was determined by methanol-HCl method, as described in previous studies [47]. The absorbance was measured using Spectramax ID5 at 530, 620, and 650 nm. The relative anthocyanins concentration was calculated according to the formula. Each sample contains three independent biological replicates.

Histochemical Staining
Cytological observation was conducted according to the method described by Han with slightly modification [48,49]. The samples of leaves and stems are fixed and dehydrated, and then cut into slices with ultramicrotome (Leica, Weztlar, German). Generated slices were treated with multiple steps, including stained with safranin-O, washed with water, discolored with alcohol, and quick-dyed with green dye. Pictures was shot using a charge coupled device (CCD) camera.

Gene Expression Analysis
CsGAPDH and CsTBP were selected as reference genes [41,50], to explore the expression pattern of CsLBD39 gene in different tissues and nitrate response. The expression levels of nitrate-responsive genes in WT and transgenic Arabidopsis were also analyzed. AtSAND and AtActin2 were used as reference genes. RT-qPCR primers were consulted to previous studies and listed in Supplementary Table S1 [1,10,11,16,51,52]. RT-qPCR test was performed with 20 µL reaction mixtures using Hieff qPCR SYBR Green Master Mix (Yeasen, Shanghai, China) on CFX96 system (Bio-Rad, Hercules, CA, USA). The relative expressions of genes were calculated using the 2 −∆∆CT method. Three separate biological replicates were set.

Statistical Analysis
Data were analyzed by SPSS 17.0 software. The difference significance of gene expression levels in tea plant were detected by Duncan's multiple-range test at a 0.05 probability. The statistical differences of data between WT and transgenic Arabidopsis were analyzed by one-way analysis of variance and indicated by asterisks (*) (* p < 0.05; ** p < 0.01; *** p < 0.001).

Conclusions
In conclusion, a novel transcription factor, named as CsLBD39, was identified from 'Longjing 43'. CsLBD39 is an LBD Class II transcription factor. Subcellular localization, transcriptional activation, and overexpression in Arabidopsis were performed to confirm its function. Overexpression of CsLBD39 decreased the nitrate content and the expression of nitrate transport-related genes in transgenic Arabidopsis plants. These results provided evidence that CsLBD39 may play a negative regulatory factor in the nitrate response pathway of tea plants.  Data Availability Statement: In this section, transcriptional data, physiological and anatomic metabolic data were measured by the authors themselves.

Conflicts of Interest:
The authors declare no conflict of interest.