iTRAQ-Based Quantitative Proteomics Analysis Reveals the Mechanism Underlying the Weakening of Carbon Metabolism in Chlorotic Tea Leaves

To uncover mechanism of highly weakened carbon metabolism in chlorotic tea (Camellia sinensis) plants, iTRAQ (isobaric tags for relative and absolute quantification)-based proteomic analyses were employed to study the differences in protein expression profiles in chlorophyll-deficient and normal green leaves in the tea plant cultivar “Huangjinya”. A total of 2110 proteins were identified in “Huangjinya”, and 173 proteins showed differential accumulations between the chlorotic and normal green leaves. Of these, 19 proteins were correlated with RNA expression levels, based on integrated analyses of the transcriptome and proteome. Moreover, the results of our analysis of differentially expressed proteins suggested that primary carbon metabolism (i.e., carbohydrate synthesis and transport) was inhibited in chlorotic tea leaves. The differentially expressed genes and proteins combined with photosynthetic phenotypic data indicated that 4-coumarate-CoA ligase (4CL) showed a major effect on repressing flavonoid metabolism, and abnormal developmental chloroplast inhibited the accumulation of chlorophyll and flavonoids because few carbon skeletons were provided as a result of a weakened primary carbon metabolism. Additionally, a positive feedback mechanism was verified at the protein level (Mg chelatase and chlorophyll b reductase) in the chlorophyll biosynthetic pathway, which might effectively promote the accumulation of chlorophyll b in response to the demand for this pigment in the cells of chlorotic tea leaves in weakened carbon metabolism.


Introduction
Tea (Camellia sinensis) is a perennial evergreen leafy woody plant native to southwest China. Recently, chlorophyll-deficient chlorina tea plant cultivars have become valuable materials in processing high quality green tea because of their high amino acid content and low catechin content [1,2]. The natural mutant of tea, "Huangjinya", exhibits chlorotic leaves and lower carbon metabolism than non-chlorotic varieties under sunlight [1,2]. In our previous studies [2,3], metabolomics and transcriptomics analyses were performed on green and chlorotic shoots of "Huangjinya" to gain an overview of the amino acid, flavonoid, and carbohydrate metabolism. These analyses revealed that the weakening of carbon metabolism is accompanied by nitrogen accumulation, suggesting that the metabolism of carbon and nitrogen are unbalanced [3]. Satou et al. [4] have shown similar results in the pale green mutants of Arabidopsis thaliana. However, the correlation of protein expression with weakened carbon metabolism in chlorotic tea leaves remains to be elucidated.
Chlorophyll consists of chlorophyll a and chlorophyll b, and plays indispensable roles in harvesting and transferring light energy during photosynthesis and carbon assimilation [5]. Chlorosis in tea leaves has always been attributed to chlorophyll deficiency. Studies on chlorophyll biosynthesis have been widely reported, and at least in angiosperm plants represented by Arabidopsis thaliana, genes for all 15 steps in the chlorophyll biosynthesis pathway, starting from the biosynthesis of glutamyl-tRNA to that of chlorophylls a and b, have been identified [6]. However, the effect of chlorophyll metabolism on photosynthesis is largely unclear. Mutants defective in chlorophyll biosynthesis have been identified in higher plants [7,8]. For example, the leaf phenotype was yellow-green in the chlorophyll mutant (Oryza sativa) and the level of chlorophyll decreased, meanwhile, chloroplast development was delayed [8]. Thylakoid proteome analysis of a novel rice (Oryza sativa) mutant, Zhenhui 249Y, and the wild type has shown that the reduction of chlorophyll b affects the assembly of light harvesting complex I (LHC-I) more severely than that of LHC-II [9].
Proteomics of leaf color mutants of tea plant has been performed using both isobaric tags for relative and absolute quantification (iTRAQ) [10] and two-dimensional gel electrophoresis (2-DE)-mass spectrometry [11,12]. In these studies, 437 differentially accumulated proteins have been identified between the tea plant cultivars "Longjing43" and "Zhonghuang1" [10] and 46 differentially abundant proteins between tender purple and mature green leaves of tea plant [12]. However, it is difficult to clarify the mechanism of weakened carbon metabolism because of the complex genetic background or inconsistent developmental stages of experimental material in different tea varieties [10]. In this study, we used chlorotic and normal green leaves ("Huangjinya", the albino tea plant cultivar) with the same genetic background and developmental stage as the experimental material to compare the protein expression profiles of shaded and non-shaded leaves by iTRAQ technique.
It is hypothesized that the inhibition of carbon assimilation results in a down-regulation of protein expression in carbohydrate synthesis and transport pathways, further weakening primary carbon metabolism. Meanwhile, flavonoid metabolism, as the major secondary metabolism, may also be suppressed by the down-regulation of the expression of related proteins. In this study, iTRAQ-based quantitative proteomics with phenotypic, biochemical, and transcriptome data confirmed our findings on the differences in protein expression profiles underlying the weakening of carbon metabolism in chlorotic "Huangjinya" tea leaves. The results of this study also provide new insights into the expression level of proteins to understand the mechanisms responsible for chlorophyll deficiency in etiolated tea plant leaves.

Phenotype, Ratio of Pigment Content, Photosynthesis of Chlorotic and Green Leaves
Compared with shaded leaves of tea plants ( Figure 1A), leaves of tea plants grown under full sunlight were chlorotic and exhibited a yellow phenotype ( Figure 1B). Transmission electron microscopy showed clear differences in leaf ultrastructure between chlorotic and shaded green leaves ( Figure 1C,D). Compared with green leaves ( Figure 1C), chlorotic leaves showed chloroplasts with abnormal structural development-thylakoids were observed, but the stacks of grana were not found ( Figure 1D). The contents of chlorophyll a, chlorophyll b, total chlorophyll, and carotenoids in chlorotic mutants and green leaves has been reported previously [3]. In this study, the contents of these four pigments were significantly lower in chlorotic leaves than those in green leaves. Furthermore, the ratio of chlorophyll a to chlorophyll b and that of total chlorophyll to carotenoids were significantly lower in chlorotic leaves than in green leaves ( Figure 2). Leaf gas exchange analysis showed that net photosynthesis and intercellular CO2 concentration were reduced by approximately 21.7% and 36.13%, respectively, in chlorotic leaves compared with green leaves. By contrast, the stomatal conductance and transpiration rate of chlorotic leaves were increased by approximately 15.2% and 21.4%, respectively, compared with green leaves (Table 1). The contents of chlorophyll a, chlorophyll b, total chlorophyll, and carotenoids in chlorotic mutants and green leaves has been reported previously [3]. In this study, the contents of these four pigments were significantly lower in chlorotic leaves than those in green leaves. Furthermore, the ratio of chlorophyll a to chlorophyll b and that of total chlorophyll to carotenoids were significantly lower in chlorotic leaves than in green leaves ( Figure 2). The contents of chlorophyll a, chlorophyll b, total chlorophyll, and carotenoids in chlorotic mutants and green leaves has been reported previously [3]. In this study, the contents of these four pigments were significantly lower in chlorotic leaves than those in green leaves. Furthermore, the ratio of chlorophyll a to chlorophyll b and that of total chlorophyll to carotenoids were significantly lower in chlorotic leaves than in green leaves ( Figure 2). Leaf gas exchange analysis showed that net photosynthesis and intercellular CO2 concentration were reduced by approximately 21.7% and 36.13%, respectively, in chlorotic leaves compared with green leaves. By contrast, the stomatal conductance and transpiration rate of chlorotic leaves were increased by approximately 15.2% and 21.4%, respectively, compared with green leaves (Table 1).

Figure 2.
Ratio of chlorophyll a to chlorophyll b and that of total chlorophyll to carotenoids. Values represent mean ± SD of three biological replicates (** p < 0.01).
Leaf gas exchange analysis showed that net photosynthesis and intercellular CO 2 concentration were reduced by approximately 21.7% and 36.13%, respectively, in chlorotic leaves compared with green leaves. By contrast, the stomatal conductance and transpiration rate of chlorotic leaves were increased by approximately 15.2% and 21.4%, respectively, compared with green leaves (Table 1).

Quantitative Identification of Tea Leaf Proteins Using iTRAQ
Differentially accumulated proteins in chlorotic and green leaves were identified using iTRAQ technique, and 302,042 spectra were obtained. Analysis using the Mascot software revealed that the number of matched spectra and unique spectra were 15,804 and 14,943, respectively. A total of 6157 unique peptides were identified. Distributions of protein mass, peptide number, and peptide length are shown in Supplementary Figures S1-S3.
We identified 2110 proteins. According to GO analysis, 1354, 1284, and 1349 proteins were annotated as cellular components, functional molecules, and those involved in biological processes, respectively ( Figure 3). The main biological function categories included nucleoside phosphate metabolic process, photosynthesis, and carbohydrate derivative catabolic process. The proteins classified as having functional molecular properties were mainly classified based on their activity: hydrolase activity, acting on glycosyl bonds, alpha-glucosidase activity, translation elongation factor activity, glucosidase activity, ATP-dependent peptidase activity, and oxidoreductase activity. A total of 1540 proteins were assigned to 22 categories using the Clusters of Orthologous Groups of proteins (COG) database; the main functional categories were transport and metabolism (21.5%); protein turnover, chaperones (10.5%); energy production and conversion (7.2%); and translation, ribosomal structure, and biogenesis (7.7%) ( Figure 4). Additionally, 1268 proteins were annotated in 119 pathways using the Kyoto Encyclopedia of Gene and Genomes (KEGG) database. The main pathways were metabolic pathways (30.52%); biosynthesis of secondary metabolites (17.03%); plant-pathogen interaction (4.18%); protein processing in endoplasmic reticulum (3.79%); starch and sucrose metabolism (3.39%); and pyruvate metabolism (3.15%) (Supplementary Table S1).

Quantitative Identification of Tea Leaf Proteins Using iTRAQ
Differentially accumulated proteins in chlorotic and green leaves were identified using iTRAQ technique, and 302,042 spectra were obtained. Analysis using the Mascot software revealed that the number of matched spectra and unique spectra were 15,804 and 14,943, respectively. A total of 6157 unique peptides were identified. Distributions of protein mass, peptide number, and peptide length are shown in Supplementary Figures S1-S3.
We identified 2110 proteins. According to GO analysis, 1354, 1284, and 1349 proteins were annotated as cellular components, functional molecules, and those involved in biological processes, respectively ( Figure 3). The main biological function categories included nucleoside phosphate metabolic process, photosynthesis, and carbohydrate derivative catabolic process. The proteins classified as having functional molecular properties were mainly classified based on their activity: hydrolase activity, acting on glycosyl bonds, alpha-glucosidase activity, translation elongation factor activity, glucosidase activity, ATP-dependent peptidase activity, and oxidoreductase activity. A total of 1540 proteins were assigned to 22 categories using the Clusters of Orthologous Groups of proteins (COG) database; the main functional categories were transport and metabolism (21.5%); protein turnover, chaperones (10.5%); energy production and conversion (7.2%); and translation, ribosomal structure, and biogenesis (7.7%) ( Figure 4). Additionally, 1268 proteins were annotated in 119 pathways using the Kyoto Encyclopedia of Gene and Genomes (KEGG) database. The main pathways were metabolic pathways (30.52%); biosynthesis of secondary metabolites (17.03%); plant-pathogen interaction (4.18%); protein processing in endoplasmic reticulum (3.79%); starch and sucrose metabolism (3.39%); and pyruvate metabolism (3.15%) (Supplementary Table S1).

Regulation of Proteins in Response to Chlorosis
In this study, 173 proteins showed significant difference (ratio of protein abundance > 1.2/0.8 fold; p < 0.05) between chlorotic and green leaves, including 80 up-regulated and 93 down-regulated proteins ( Table 2). A total of 23, 49, and 87 proteins reproducibly decreased by 0.50-, 0.67-, and 0.83-fold, respectively, in chlorotic leaves compared with green leaves (Supplementary Table S2). On the other hand, levels of 3, 23, and 75 proteins increased by more than 2.0-, 1.5-, and 1.2-fold, respectively, in chlorotic leaves compared to in green leaves (Supplementary Table S3).

Regulation of Proteins in Response to Chlorosis
In this study, 173 proteins showed significant difference (ratio of protein abundance > 1.2/0.8 fold; p < 0.05) between chlorotic and green leaves, including 80 up-regulated and 93 down-regulated proteins ( Table 2). A total of 23, 49, and 87 proteins reproducibly decreased by 0.50-, 0.67-, and 0.83-fold, respectively, in chlorotic leaves compared with green leaves (Supplementary Table S2). On the other hand, levels of 3, 23, and 75 proteins increased by more than 2.0-, 1.5-, and 1.2-fold, respectively, in chlorotic leaves compared to in green leaves (Supplementary Table S3).  KEGG pathway enrichment analysis was employed to explore the metabolic and biosynthetic pathways, which changed in response to the chlorotic mutation with those differentially accumulated proteins. A number of such pathways were identified, including chlorophyll biosynthesis, carbohydrate transport and metabolism, energy production and conversion, flavonoid metabolism, nitrogen metabolism, chloroplast function, and oxidative stress (Table 3). Table 3. Differentially accumulated proteins identified in pathways potentially associated with chlorophyll deficiency in green and chlorotic leaves of the tea plant mutant cultivar "Huangjinya".  The differentially accumulated proteins involved in chlorophyll biosynthesis included nine proteases. The up-regulation of six of these proteins, including glutamyl-tRNA (Gln) amino-transferase, geranylgeranyl, glutaminyl-tRNA synthetase, magnesium chelatase, magnesium protoporphyrin, and porphobilinogen deaminase, was increased in the chlorotic leaves than in green leaves by 1.02to 1.68-fold, whereas that of the remaining three proteins, including chlorophyll(ide) b reductase, protochlorophyllide reductase, and violaxanthin de-epoxidase, was reduced in the chlorotic leaves by 0.8-to 0.93-fold.

Identity Proteins
The differentially accumulated proteins related to carbohydrate transport and metabolism mainly comprised six rate-limiting enzymes of the glycolytic pathway (6-phosphofructokinase, fructokinase, hexokinase, pyruvate kinase, phosphoglycerate mutase, and phosphopyruvate hydratase), five glycosidases (beta-fructofuranosidase, xylosidase, galactose oxidase, UDP-L-arabinosidase, and beta-glucosidase), and three proteases related to photosynthesis (ribulose-bisphosphate carboxylase, fructose-1,6-bisphosphatase, and granule-bound starch synthase). The level of eight proteins was up-regulated by 1.01-to 1.77-fold in the chlorotic leaves. Additionally, the level of six proteins was down-regulated by 0.38-to 0.99-fold in the chlorotic leaves.
The most relevant pathway for energy generation and conversion is the tricarboxylic acid (TCA) cycle or the Krebs cycle, which generates the highest amount of energy in the most efficient way through the oxidation of sugars and other substances. The differentially accumulated proteins involved in the Krebs cycle mainly include five dehydrogenases (isopropyl-malate dehydrogenase, dihydrolipoyl dehydrogenase, malate dehydrogenase, pyruvate dehydrogenase, and succinate dehydrogenase), two dihydrolipoyllysine-residue transferases (dihydrolipoyllysine-residue acetyltransferase and dihydrolipoyllysine-residue succinyltransferase), the other are ATP citrate synthases and aconitate hydratase. Among these proteins, 16 were up-regulated by 1.05-to 1.48-fold and 4 were down-regulated by 0.95-to 0.76-fold in the chlorotic mutation compared to green leaves.
Of the eleven differentially accumulated proteins involved in flavonoid metabolism, the levels of eight proteins (cinnamate acid 4-hydroxylase (C4H), chalcone isomerase (CHI), chalcone synthase (CHS), anthocyanidin synthase (ANS), anthocyanidin reductase (ANR), anthocyanidin 3-O-glucosyltransferase (A3Glc), 3-dehydroshikimate dehydratase (3DSD), and 3-dehydroquinate synthase (3DHQ)) were increased by 1.09-to 1.84-fold in chlorotic leaves than in green leaves, whereas those of three proteins (phenylalanine ammonia-lyase (PAL), 4-coumarate-CoA ligase (4CL), and flavonol synthase (FLS)) were reduced by 0.62-to 0.75-fold in the chlorotic leaves compared to in green leaves. Phenylalanine is a precursor of the flavonoid biosynthesis pathway, and PAL and 4CL play key roles in the conversion of phenylalanine to coumaroyl CoA. Our results showed that levels of PAL and 4CL proteins were reduced in the chlorotic leaves, indicating that flavonoid metabolism was inhibited. From the branching of flavonoid biosynthesis (i.e., the synthetic pathway of anthocyanins and flavonols), the accumulation of anthocyanins was promoted and the synthesis of flavonols was inhibited.
Among the proteins involved in nitrogen metabolism, the levels of nine proteins' expression (3-deoxy-7-phosphoheptulonate synthase, alanine transaminase, anthranilate synthase, aspartate kinase, cysteine synthase, glutamate synthase, homoserine kinase, methionine synthase, and glutathione reductase, GR [NADPH; nicotinamide adenine dinucleotide phosphate]) were increased and the expression levels of three proteins (ferredoxin-nitrite reductase, glycine hydroxymethyl transferase, and S-adenosyl methionine synthase) were reduced. These results indicated that the up-regulation of the expression of most amino acid synthase proteins might promote nitrogen assimilation and recycling.
Photosystem Q(B) plays an important role in chloroplast function, and the expression level of its protein was down-regulated by 0.4-fold in chlorotic leaves compared with green leaves, indicating that chloroplast function was inhibited under strong light stress. Dehydrins, also known as LEA D-11 or LEA II, are proteins whose expression is induced by various environmental stress factors [13]. B5TV66_CAMSI Putative dehydrin was annotated as an oxidative stress protein, and its expression level was significantly up-regulated by 2.29-fold. Therefore, it is speculated that the antioxidant capacity of chlorotic tea leaves might be enhanced compared to that of green leaves.

Integrated Analysis of Transcriptomic and Proteomic Datasets
A total of 5051 differentially expressed genes (DEGs), with differences between chlorotic and shaded green leaves, were selected for bioinformatics analysis. The combination of transcriptomic and proteomic datasets revealed correlations between 126, 52, and 19 genes and proteins in identification, quantitation, and differential expression levels, respectively (Table 4). Nineteen genes and proteins with significant differences at the quantitative level are shown in Table 5. These were classified in the following categories: chloroplast structure and function; carbohydrate and amino acid metabolism; and flavonoid biosynthesis and oxidative stress.
Genes and proteins related to chloroplast structure and function are related to the chloroplast stroma thylakoids (i.e., V-type proton ATPase subunit C, tRNA (cytosine38-C5)-methyltransferase, chloroplast small heat shock protein). The C metabolism pathway mainly involves proteins for starch synthesis (glycogen synthase), alpha-maltase (alpha-glucosidases), and lignin metabolism (L-ascorbate oxidase). Metabolic processes associated with nitrogen metabolism are mainly arginine metabolism (arginase). Oxidative stress caused by the etiolating mutation of tea leaves was also related to gene expression and protein accumulation levels. Oxidative stress-related proteins and enzymes changed significantly (fructose-bisphosphate aldolase and dehydrin). The change of the polyphenol metabolic pathway is mainly related to its upstream pathway, such as 4-coumarate-CoA ligase (CL48129Contig1), phenylalanine ammonia lyase (comp64735_c0_seq1_2), two key genes and enzyme levels were significantly down-regulated in etiolated leaves. Table 4. The number of proteins and genes identified, quantified, and differentially expressed in green and chlorotic leaves of the tea plant mutant cultivar "Huangjinya".

Discussion
Carbohydrates are a direct product of carbon assimilation via photosynthesis. Additionally, carbohydrates represent a source of plant energy and are involved in the formation of the plant cytoskeleton. In this study, the increased stomatal conductance and the decreased CO 2 concentration was accompanied by a reduced net photosynthesis rate in the chlorotic leaves. However, the concentration of CO 2 in the study of Brassica napus increased [14], possibly because of differences in the leaf structure of the two plants. Both studies have shown impaired carbon fixation efficiency in chlorotic leaves. In this study, levels of some proteins involved in carbohydrate metabolism, including fructokinase, hexokinase, phospho-pyruvate hydratase, ribulose-bisphosphate carboxylase, granule-bound starch synthase, and xylosidase, were reduced, which is consistent with this speculation. Rubisco (ribulose-bisphosphate carboxylase/oxygenase) is the rate-limiting enzyme for carbon fixation in photosynthetic reactions, and is essential for improving the photosynthetic efficiency of plants [15]. In this study, the abundance of Rubisco protein was significantly reduced in chlorotic leaves compared with green leaves (Table 3), leading to a reduction in carbohydrate biosynthesis and sugar content. Simultaneously, proteins (6-phosphate fructokinase, pyruvate kinase, phosphoglycerate mutase, and fructose 1,6-bisphosphatase) with higher expression levels were involved in the glycolytic pathway in chlorotic leaves, promoting carbohydrate catabolism. Some scholars have shown that chloroplast endometrial damage activates the expression of glycolysis-related genes [4]. Results of protein accumulation and gene expression indicated that carbohydrate accumulation was decreased in "Huangjinya" leaves under strong light.
Carbon and nitrogen metabolism balance guarantees the normal growth of tea plants, and enhanced nitrogen metabolism is accompanied by the reduced capability of photosynthesis and carbon metabolism [4]. In this study, methionine synthetase, cysteine synthetase, and glutamate synthetase were up-regulated in chlorotic leaves compared with green leaves, indicating that strong light promotes nitrogen metabolism and amino acid accumulation in "Huangjinya", which is consistent with a previous study [16]. We speculate that reduced chlorophyll biosynthesis in turn reduced nitrogen consumption, thus increasing the content of upstream substances (amino acids) in leaves. In addition, environmental stress such as intense light and high temperature affect the normal growth of plants, and reactive oxygen species (ROS) are accumulated as a by-product [17]. Previous studies have shown that increasing the glutathione reductase activity in chloroplasts improves the photochemical ratio in transgenic cotton plants, thereby reducing photoinhibition [18]. This suggests that enhanced glutathione reductase activity protects the leaf cell biofilm and enhances the plant's ability to defend itself against abiotic stress (e.g., UV-B radiation) [19]. In this study, the expression of glutathione reductase protein was increased by 1.3-fold, and as a result the ability to remove ROS produced by UV-B radiation under intense light might be enhanced.
A simplified schematic presentation of the weakening of carbon metabolism in the chlorotic mutation is shown in Figure 5. The ratio of chlorophyll a to chlorophyll b has been considered as an important parameter to measure the light tolerance of plants. An increase in this ratio is beneficial for the absorption of blue-violet light, which is suitable for plant growth in the dark [20]. Generally, the biosynthesis and degradation of chlorophyll a and chlorophyll b in plants occur in a dynamic cycle, and chlorophyll biosynthesis requires relatively low light intensity. However, as a light-sensitive plant, the ratio of chlorophyll a to chlorophyll b in leaves of "Huangjinya" under the shade was significantly higher than that under intense light, suggesting that the conversion of chlorophyll a to chlorophyll b is accelerated under strong light. Increase in chlorophyll b content is beneficial for the plant's adaptation to light stress. Chlorophyll(ide) b reductase is the key enzyme that catalyzes the first step in chlorophyll b degradation, and plays an important role in the process of leaf senescence, with the degradation of LHC-II and chloroplast matrix [21]. Studies have reported that chlorophyll b reductase was involved in the conversion process of chlorophyll b and chlorophyll a, which is considered to be an important clue for chlorophyll b degradation [22]. In this study, the level of chlorophyll(ide) b reductase was reduced by 0.9-fold in chlorotic leaves, and the ratio of chlorophyll a to chlorophyll b was lower in chlorotic leaves than in green leaves, indicating that the activity of chlorophyll b reductase was suppressed. Zhang and Tan [23] also obtained similar results regarding the reduction of chlorophyll content and chlorophyll a/b ratio under salt stress. The studies of Sang et al. [24] and Huang et al. [25] suggested that the decreased ratio of chlorophyll a/b resulted from a higher sensitivity to various environmental stresses for chlorophyll a compared to chlorophyll b. Therefore, the pathway of chlorophyll b biosynthesis is still enhanced although the total chlorophyll content was decreased in the chlorotic leaves. Interestingly, the content of chlorophyll b was not increased with the enhanced pathway of chlorophyll b synthase in "Huangjinya" leaves under strong light. Because the process of chlorophyll biosynthesis is complex and highly conserved, the mutation of a gene can severely affect the chlorophyll content, leading to a different color phenotype of leaves [26]. Magnesium chelatase is another enzyme with a significant effect on chlorophyll biosynthesis. It catalyzes the insertion of Mg 2+ into protoporphyrin IX [27]. In the process of leaf albinism, the abundance of magnesium chelatase subunit proteins in higher plants is significantly increased under light [28,29]. In this study, the expression level of magnesium chelatase protein was higher in chlorotic leaves by 1.34-fold than in green leaves, which is consistent with the study of Walker et al. [29]. Müller et al. [30] showed that oxidative stress improved the quality of monomeric chlorophyll H in Escherichia coli. Moreover, genes involved in the biosynthesis of chlorophyll were induced by ROS in "Huangjinya" leaves, but the sub-structure of magnesium chelatase was not influenced, although its activity was enhanced. Comprehensively, we proposed that a feedback mechanism existed in weakened carbon metabolism, where the pathway of chlorophyll b biosynthesis was positively regulated as a result of increased expression of Mg chelatase proteins and decreased expression of chlorophyll b reductase proteins.  [24] and Huang et al. [25] suggested that the decreased ratio of chlorophyll a/b resulted from a higher sensitivity to various environmental stresses for chlorophyll a compared to chlorophyll b. Therefore, the pathway of chlorophyll b biosynthesis is still enhanced although the total chlorophyll content was decreased in the chlorotic leaves. Interestingly, the content of chlorophyll b was not increased with the enhanced pathway of chlorophyll b synthase in "Huangjinya" leaves under strong light. Because the process of chlorophyll biosynthesis is complex and highly conserved, the mutation of a gene can severely affect the chlorophyll content, leading to a different color phenotype of leaves [26]. Magnesium chelatase is another enzyme with a significant effect on chlorophyll biosynthesis. It catalyzes the insertion of Mg 2+ into protoporphyrin IX [27]. In the process of leaf albinism, the abundance of magnesium chelatase subunit proteins in higher plants is significantly increased under light [28,29]. In this study, the expression level of magnesium chelatase protein was higher in chlorotic leaves by 1.34-fold than in green leaves, which is consistent with the study of Walker et al. [29]. Müller et al. [30] showed that oxidative stress improved the quality of monomeric chlorophyll H in Escherichia coli. Moreover, genes involved in the biosynthesis of chlorophyll were induced by ROS in "Huangjinya" leaves, but the sub-structure of magnesium chelatase was not influenced, although its activity was enhanced. Comprehensively, we proposed that a feedback mechanism existed in weakened carbon metabolism, where the pathway of chlorophyll b biosynthesis was positively regulated as a result of increased expression of Mg chelatase proteins and decreased expression of chlorophyll b reductase proteins. Flavonoids are important secondary metabolites of carbon metabolism in the tea plant, and have several physiological functions. Flavonoids provide plants with vibrant pigmentation, which protects the plants from UV-B radiation and helps attract pollinators as well as seed dispersers [31]. Because photosynthetic carbon assimilation is severely inhibited in chlorotic leaves, flavonoid biosynthesis is accordingly reduced [32]. Similar to the results of Zhang et al. [2], in this study, the Flavonoids are important secondary metabolites of carbon metabolism in the tea plant, and have several physiological functions. Flavonoids provide plants with vibrant pigmentation, which protects the plants from UV-B radiation and helps attract pollinators as well as seed dispersers [31]. Because photosynthetic carbon assimilation is severely inhibited in chlorotic leaves, flavonoid biosynthesis is accordingly reduced [32]. Similar to the results of Zhang et al. [2], in this study, the expression levels of PAL and 4CL proteins (i.e., the rate-limiting enzymes of the flavonoid biosynthesis pathway) were down-regulated by 0.70-and 0.75-fold in chlorotic leaves (Table 3). Additionally, changes in the gene expression of 4CL confirmed the result of 4CL protein expression, thereby inhibiting the accumulation of flavonoids. A previous study showed that most of flavonoids were accumulated on the leaf surface to protect structure and organization in plants from UV-B radiation damage [33], while in this study, our results indicated that abnormally developed chloroplast inhibited the accumulation of chlorophyll and flavonoids because few carbon skeletons were provided as a result of weakening of carbon metabolism, and the reduction of flavonoids contents was unfavorable for the chlorotic mutant to protect against UV-B radiation damage.

Plant Material and Shading Treatment
"Huangjinya" (Camellia sinensis (L.) O. Kuntze cv.) is a light-sensitive albino tea variety, which displays yellow shoots under strong light condition. The low levels of flavonoids in "Huangjinya" reveals that the metabolism of flavonoids has great correlation to strong light stress [1,2]. A mutant of "Huangjinya" was officially released in Zhejiang province in 2008 and specimens were obtained free of charge from the owner of the mutant Deshi Tea Plantation, Yuyao, Zhejiang province [34]. Then, the experimental material was planted in pots at the Tea Research Institute, Chinese Academy of Agricultural Sciences (TRI, CAAS), Hangzhou, China. In March 2014, 60 pots of tea plants with uniform young shoots, with one bud and one leaf, were selected for the experiment. Half of the pots were treated with high-density polyethylene tape with two-pin net (60% sun-shading, 320-800 µmol·m −2 ·s −1 ), and the remaining half were exposed to full sunlight (800-2000 µmol·m −2 ·s −1 ) for ten days. Randomly selected samples of young shoots, with one bud and two leaves, were harvested, immediately frozen in liquid nitrogen, and stored in a −70 • C ultra-refrigerator. Sampling was repeated six times from shaded and unshaded plants.

Electron Microscope Analysis
Transmission electron microscope (TEM, Hitachi Ltd., Tokyo, Japan) was used to observe the ultrastructure of chlorotic leaves. Leaf samples (approximately 1 mm 2 ) were fixed in 2.5% glutaraldehyde solution overnight at 4 • C. Ultrathin sections of fixed leaves were cut, stained, and viewed under a JEM-1230 transmission electron microscope (Nippon Tekno, Tokyo, Japan) at an accelerating voltage of 80 kV as described previously [1].

Leaf Gas Exchange Measurement
The fifth leaf of potted chlorotic mutants and wild-type tea plants was subjected to gas exchange analysis. A Li-Cor 6400 portable photosynthesis system (Li-Cor Inc., Lincoln, NE, USA) with a built-in light source set at 1000 µmol photons·m −2 ·s −1 was used to determine the net photosynthesis and stomatal conductance. All measurements were carried out between 09:00 a.m. and 11:00 a.m., with the leaf temperature adjusted to 25 • C.

Protein Extraction, iTRAQ Labeling, Data Acquisition, and Processing
Samples used for proteomic analysis were the same as those used for transcriptome analysis, and consisted of two biological replicates of shaded and unshaded leaves. All of the steps, including protein extraction, iTRAQ labeling of protein samples, liquid chromatography electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS) analysis based on Q Exactive, mass spectrometer data analysis, and functional annotation of proteins, were performed as described previously [35]. Protein identification was performed using the Mascot search engine (Matrix Science, London, UK; version 2.3.02) against a database containing 133,175 sequences.

Quantitative Real-Time PCR (qRT-PCR) Analysis
Total RNA was isolated using an RNA plant plus kit (Tiangen, China). cDNA was synthesized using a Prime Script TM RT reagent Kit (TaKaRa, Biotechnology Co., Ltd., Dalian, China). qRT-PCR was performed on the Applied Biosystems 7300 machine (Carlsbad, CA, USA). Primer pairs used for qRT-PCR are shown in Supplementary Table S4, and GAPDH was used as the reference gene. For each target gene, triplicate reactions were performed. Relative transcript levels were calculated against that of the internal control (GAPDH) according to the equation 2 −∆∆Ct . All data are shown as mean ± standard deviation (SD) (n = 3).

Bioinformatics Analysis
Functional analysis of the identified proteins was conducted using gene ontology (GO) annotation (Available online: http://www.geneontology.org/), and proteins were categorized according to their biological process, cellular components, and molecular function. The differentially accumulated proteins were further classified into the Clusters of Orthologous Groups of proteins database (Available online: http://www.ncbi.nlm.nih.gov/COG/) and Kyoto Encyclopedia of Gene and Genomes (KEGG) database (Available online: https://www.kegg.jp/kegg/pathway.html). GO and pathway enrichment analyses were performed to determine the functional sub-categories and metabolic pathways in which the differentially accumulated proteins showed significant enrichment. Cluster analysis of differentially accumulated proteins was performed using Cluster 3.0 (Stanford University, California, USA).
Data of integrated transcriptomic analysis reported previously [3] were deposited in the Sequence Read Archive (SRA) database (Available online: https://trace.ncbi.nlm.nih.gov/Traces/sra/) of the National Center for Biotechnology Information (NCBI) under the accession number SRP072792.

Conclusions
In this study, 2110 proteins were identified in "Huangjinya" leaves, the expression levels of 19 of which changed significantly, correlating with RNA expression. Differential protein expression analysis indicated that primary carbon metabolism (i.e., carbohydrate synthesis and transport) was inhibited in chlorotic tea leaves. The differentially expressed genes and proteins combined with photosynthetic phenotypic data suggested that 4-coumarate-CoA ligase (4CL) had a major effect on repressing flavonoid metabolism (secondary carbon metabolism), and abnormal developmental chloroplast inhibited the accumulation of chlorophyll and flavonoids because few carbon skeletons were provided as a result of the weakened primary carbon metabolism. Additionally, a positive feedback mechanism was verified at the protein level (Mg chelatase and chlorophyll b reductase) in the chlorophyll biosynthetic pathway, which might effectively promote the accumulation of chlorophyll b in response to the demand for this pigment in the cells of chlorotic tea leaves in weakened carbon metabolism.

Conflicts of Interest:
The authors declare that they have no conflict of interest.