Divergent MYB paralogs determine spatial distribution of linalool mediated by JA and DNA demethylation participating in aroma formation and cold tolerance of tea plants

Summary

Linalool not only is one of characteristic flavour volatiles of tea, contributing to floral aroma, but also a kind of defensive compounds, playing essential roles in resistance against biotic/abiotic stresses. Although the linalool synthases have been identified, much is unknown about the regulation mechanism in tea plants. We identified two pairs of MYB paralogs as linalool biosynthesis activators, in which one pair (CsMYB148/CsMYB193) specifically expressed in flowers, and another (CsMYB68/CsMYB147) highly expressed in flowers, leaves, fruits and roots. These activators interacted with CsMYC2 to form MYC2‐MYB complexes to regulate linalool synthase. While Jasmonate ZIM‐domain (JAZ) proteins served as the linalool biosynthesis repressors by interfering MYC2‐MYB complex. Further, we found that the transcripts of CsMYB68/CsMYB147 were significantly upregulated by jasmonic acid (JA) to improve linalool products during tea processing and that linalool pathway may as one of the downstream pathways of JA signalling and DNA methylation processes to participate in cold resistance. Under cold stress, JA signalling was activated to elevate the abundance of MYC‐MYB complexes; meanwhile, DNA demethylation was also activated, leading to declining methylation levels and increasing transcripts of CsMYB68/CsMYB147. Our study provides a new insight into synergistically improving tea quality and tea plant resistance.

Introduction

Tea is popular in the world for its pleasure flavour and enriched beneficial compounds, such as catechins, theanine, caffeine and volatiles (Zhao et al., 2020a). Particularly, the volatiles, contributing to tea aroma, have been the principal index when consuming tea. In tea, more than 700 volatile compounds have been detected and many volatiles not only play the central roles of tea aroma formation, but also participate in resistant to multiple stresses of tea plants (Ho et al., 2015; Jin et al., 2023b; Zhu et al., 2008). Therefore, it is benefit for breeding tea cultivars with high aroma and high resistant to understand the mechanism of tea aroma biosynthesis during tea processing and under stress condition.

In the volatiles, linalool, the most abundant volatile terpenoid compound, is the characteristic compound in several types of tea, such as green tea, white tea and oolong tea (Feng et al., 2019). Linalool belongs to monoterpene with floral aroma, and derives from geraniol diphosphate (GPP) catalysed by terpene synthase (TPS) via methylerythritol phosphate (MEP) pathway. In tea plants, the linalool synthases have been identified and these enzymes mainly expressed in flowers and leaves (Liu et al., 2018, 2024a; Mei et al., 2017; Zhou et al., 2020b). Tea linalool synthase can be stimulated by biotic and abiotic stresses, such as wounding, insect attack and MeJA treatment (Liu et al., 2018, 2024a; Zhou et al., 2020b). The quick response of linalool synthase to stresses plays key roles in tea aroma formation. In fresh tea leaves, the concentration of linalool was low, but a large amount of linalool was quickly synthesized during tea processing (Zuo et al., 2023). The reason maybe is that the expression pattern of linalool synthase was sharply activated by multiple stresses from tea processing, such as wounding and dehydration (Kong et al., 2024; Zeng et al., 2019). Linalool also serves as a kind of defensive volatiles (Taniguchi et al., 2014; Yactayo‐Chang et al., 2024). In tea plants, the biosynthesis of linalool was sharply induced by cold stress, and the cold‐induced linalool subsequently primed neighbouring plants for cold stress (Jin et al., 2023a; Zhao et al., 2020b). However, the mechanism of linalool biosynthesis under multiple stresses, such as tea processing and cold condition, remains largely unknown in tea leaves.

Linalool biosynthesis is strictly regulated by complex transcriptional module, in which the transcription factors could directly bind to and activate the linalool synthase, playing the central roles in linalool biosynthesis regulation. In dicotyledonous plant Arabidopsis, AtMYB21 and AtMYC2 were shown to activate the expression of linalool synthase by forming MYC2‐MYB complex (Yang et al., 2020). Additionally, their orthologous genes in monocotyledonous plant Freesia hybrida, FhMYB21L2, together with FhMYC2, was also involved in directly modulating linalool synthesis gene in flowers (Yang et al., 2020). Recent study also found that in dicotyledonous plant Osmanthus fragrans, OfMYB21, the orthologs of AtMYB21, directly targeted OfTPS2 to positively regulate linalool biosynthesis in flowers (Lan et al., 2023). Meanwhile, OfMYB21 interacted with OfMYC2 forming an activated MYC2‐MYB complex, but the activation of MYC2‐MYB complex was regulated by JAZ protein (Lan et al., 2023). Those data together not only implied the conserved functions of MYC2‐MYB complex on linalool biosynthesis in angiosperm, but also highlighted the positive roles of JA in linalool biosynthesis (Taniguchi et al., 2014; Wang et al., 2024a). Besides that, other type transcription factors were also reported in regulating linalool biosynthesis, such as PpERF61 and PpbHLH1 in peach fruit (Wei et al., 2021, 2022), CpNAC56 in papaya fruit (Yao et al., 2023), LiNAC100 in Lilium ‘Siberia’ flower (Liu et al., 2024b), LaMYC7 in Lamiaceae angustifolia flower (Dong et al., 2024) and DobHLH4 in Dendrobium officinale flower (Yu et al., 2021). All these transcription factors could directedly target the corresponding linalool synthases. In addition, DNA methylation was also involved in linalool biosynthesis regulation via acting on linalool synthase and linalool biosynthesis‐related transcription factors. During peach fruit ripening, demethylation of PpERF61 and PpTPS3 was positively associated with the linalool concentration (Wei et al., 2021, 2022). Although several transcription factors related to linalool biosynthesis have been characterized in flowers and fruits in other plants, the regulatory mechanism of linalool biosynthesis in tea plants, particularly in tea leaves, remains largely unknown.

Occasional cold stress seriously declines the quality and yield of tea during tea season. Plants have evolved many adaptive mechanisms to survive cold stress, in which the ICE (Inducer of CBF Expression 1)‐CBF (C‐repeat Binding Factor) regulatory cascade was conserved and played a central role in cold response in plants (Hu et al., 2013; Shi et al., 2018). In Arabidopsis, the ICE‐CBF pathway was modulated by JA signalling. The cold‐induced JA could activate the degradation of JAZ proteins to free ICE to activate CBF pathway (Hu et al., 2013). On the contrary, JA catabolism was upregulated by warm temperature, leading to stability of JAZ proteins to promote plant growth (Zhu et al., 2021). Recent study has showed that JA mediated by LUX ARRHYTHMO (CsLUX) also could act on the conserved ICE‐CBF pathway by CsJAZ1 protein to improve cold tolerance of tea plants (Wang et al., 2024c). On the other hand, JA also participated in regulating the biosynthesis of secondary metabolites with cold tolerance, such as anthocyanins, under cold stress (Schulz et al., 2015). But the downstream metabolic pathways mediated by JA under cold stress are largely unknown in tea plants. Linalool was induced by cold tress and emitted from cold‐stressed tea plants to prim cold tolerance of their neighbours via CBF pathway (Zhao et al., 2020b). However, the mechanism of linalool biosynthesis under cold stress and whether cold‐induced JA participated in linalool biosynthesis were still unknown in tea plants.

In this study, we identified two pairs of MYB transcription factors with divergent expression patterns and their positive roles in regulating linalool biosynthesis in tea flowers or tea leaves. These two pairs of MYB transcription factors could interact with CsMYC2 to form activated MYC2‐MYB complexes to directedly regulate linalool synthase, but the activated complexes could be destroyed by JAZ proteins. Further, the MYC2‐MYB complexes mediated by JA signalling participated in regulating linalool biosynthesis during tea processing and improving tolerance to cold stress of tea plants by upregulating linalool biosynthesis. We also found the DNA methylation levels of CsMYB68 and CsMYB147 were associated with the linalool contents under cold stress. Overall, the study unveiled the regulatory complexes of linalool biosynthesis, and their roles in tea quality formation and tea plants cold tolerance, providing a new insight into synergistically improving tea quality and tea plant resistance.

Results

Screening the transcription factors of linalool biosynthesis in tea plants

In tea plants, linalool biosynthesis pathway has been well studied, in which the linalool synthases are the determining components of linalool concentration in flowers and leaves (Liu et al., 2018, 2024a; Zhou et al., 2020b). Recently, numerous studies have showed that MYB21‐type transcription factors were the crucial regulators of linalool biosynthesis in plants, such as in Arabidopsis and Freesia hybrida (Yang et al., 2020). To dig linalool biosynthesis regulators in tea plants, the MYB21 homologues, including CsMYB148 and CsMYB193, were identified from tea genome (Li et al., 2022b) (Figure 1a; Figure S1). Expression analysis showed that the two MYB21‐type homologues were specifically expressed in tea flowers, but not in leaves (Figure 1b; Table S1). However, metabolic analysis showed that linalool mainly accumulated in tea flowers and young leaves (Figure 1c). Additionally, the linalool synthase gene Terpene Synthase76 (CsTPS76) (Zhou et al., 2020a) was highly expressed in both flowers and leaves (Figure 1b; Figure S2; Table S1).

Figure 1.

Character of candidate MYB transcription factors involved in regulating linalool biosynthesis in tea plants. (a) Phylogenetic tree of candidate tea MYBs with others related to linalool biosynthesis. The numbers at the nodes indicate the bootstrap value with 1000 replicates. The tea MYB proteins were highlighted with black circles. (b) Expression patterns of candidate MYB genes in different tissues of tea plants. The tea linalool synthase (CsTPS76) was used as the reference gene. (c) Linalool profiling in different tissues were measured by GC–MS. (d) Expression divergence of MYB genes in a natural tea population (n = 51). The expression data were all generated from young leaves by RNA‐seq. (e) Expression patterns of MYB genes in tea leaves under biotic stress. CK, control; EA, tea geometrid biting. The foldchange of expression values (EA/CK) was listed on the bars. (f) Synonymous substitution rates (Ks) analysis of duplicated‐MYB gene pairs in tea and Arabidopsis genome. The dotted line indicated the Ks value of the recent whole‐genome duplication (WGD) event of tea plants (Ks = 0.36). (g) Ectopically overexpressing CsMYB68, CsMYB147, CsMYB148 and CsMYB193 in tobacco plants leading to linalool accumulation compared with the wildtype respectively. (h) Representative pictures of leaf‐derived transgenic hairy roots overexpressing CsMYB68, CsMYB148 or GFP as the control. The hairy roots were remarked with white arrows. Ectopic overexpressing CsMYB148 (i) and CsMYB68 (j) significantly activated the expression of CsTPS76 in tea hairy roots. The control was the overexpression of Empty Vector (GFP). (k) The expression of CsMYB68 and CsMYB147 was simultaneously reduced significantly in the tea leaves treated with antisense oligonucleotides (asODN)‐CsMYBs (CsMYBs‐KD) as compared with Sense ODN. The effect of CsMYB68 and CsMYB147 downregulation on the expression of tea linalool synthase CsTPS76 (l) and the concentration of linalool (m) in tea leaves. (n) Subcellular localization of GFP‐CsMYBs in tobacco epidermal cells. The data are from at least three biological replicates and are expressed as means ± SD (*P < 0.05; **P < 0.01; Student's t‐test). CsACTIN was introduced to normalize the expression data.

In order to continue to search for linalool regulators in tea leaves, we reexamined the paralogs of MYB21 in tea plants, and another pair of MYB paralogs caught our attention. The pair of paralogs, including CsMYB68 and CsMYB147, belonged to MYB21‐like type based on the phylogenic tree (Figure 1a; Figure S1). Expression analysis showed that the expression patterns of CsMYB68 and CsMYB147 were different with CsMYB148 and CsMYB193. CsMYB68 was highly expressed in tea flowers and apical buds, while CsMYB147 was highly expressed in tea flowers, fruits, roots and young leaves (Figure 1b; Table S1). For further confirming the expression divergence between two pairs of MYB genes, their expression patterns were investigated in young leaves from a natural population (n = 51). RNA‐seq data showed that the CsMYB148 and CsMYB193 did not express in almost all samples, meanwhile CsMYB68 and CsMYB147 showed higher expression levels significantly than CsMYB148 and CsMYB193 (Figure 1d; Table S2). Further, similar expression profiling was observed in tea leaves under biotic stress (tea geometrid treatment) (Figure 1e; Table S3). Interestingly, these two pairs of MYB genes have been generated before the divergence of Thea section plants based on the synonymous substitution rate (Ks) value calculation (Figure 1f), and similar evolutionary process was observed for their homologues in Arabidopsis (Figure 1f). Ks value also indicated that CsMYB68 and CsMYB147 were notedly younger than CsMYB148 and CsMYB193 (Figure 1f). Together, these data suggest that these MYB transcription factors with divergent expression patterns maybe participate in regulating linalool biosynthesis in tea different tissues.

CsMYB68, CsMYB147, CsMYB148 and CsMYB193 proteins redundantly regulate linalool biosynthesis in tea plants

In order to validate our hypothesis, we firstly created the stable transgenic tobaccos overexpressing CsMYB68, CsMYB147, CsMYB148 and CsMYB193 under the control of 35S promoter, respectively (Figure S3A). Gas chromatography/mass spectrometry (GC/MS) analysis showed that ectopically overexpressing these MYB transcription factors significantly led to linalool biosynthesis in tobacco plants compared with the untransformation plants (WT) (Figure 1g; Figure S3B). Gene expression analysis showed that compared with the WT, the expression levels of tobacco linalool synthases were significantly induced in overexpression lines (Figure S4).

To characterize these MYB transcription factors in tea plants, two of them, CsMYB68 and CsMYB148, were typically selected for overexpression in Agrobacterium‐mediated tea transgenic hairy roots (Figure 1h). The empty vector was the control. Both gene expression by quantitative real‐time PCR (qRT‐PCR) and Green fluorescent protein (GFP) signalling by microscope were used for positive transformation screening (Figure 1h–j). Although no linalool was accumulated in tea roots, qRT‐PCR analysis showed that the expression levels of CsTPS76 were significantly upregulated in the hairy roots overexpressing CsMYB68 or CsMYB148, compared with the GFP control (Figure 1i,j). Heterogenous and homogeneous overexpression assays implied that four MYB proteins played conserved roles in positively regulating linalool biosynthesis.

An antisense oligodeoxynucleotide (asODN) interfering experiment was also adopted to interfere the expression of target MYB genes in tea leaves. Given the gene redundancy and divergence of gene expression pattern, the CsMYB68 and CsMYB147 were simultaneously suppressed in young tea leaves. The CsMYB68 and CsMYB147 transcript levels were significantly reduced following 72 h of asODN treatment (CsMYBs‐KD) compared with sense ODN treatment (Sense ODN), as verified by qRT‐PCR (Figure 1k). Metabolic analysis showed that the linalool concentration was significantly decreased in CsMYBs‐KD compared with the Sense ODN, which was consistent with the lower expression level of CsTPS76 in CsMYBs‐KD tea leaves (Figure 1l,m), indicating that CsMYB68 and CsMYB147 were the positive regulators of linalool synthesis in tea leaves.

Then, we analysed the subcellular localization of those transcription factors by transient expression in tobacco leaves. Subcellular localization assay showed that, except that GFP‐CsMYB193 was observed in both nucleus and cytoplasm, whereas GFP‐CsMYB68, GFP‐CsMYB147 and GFP‐CsMYB148 were all observed in the nucleus of tobacco epidermal cells (Figure 1n). Summarily, two pairs of homologous MYB transcription factors with different expression patterns have been identified as the regulators of linalool biosynthesis in tea plants.

CsMYB68, CsMYB147, CsMYB148 and CsMYB193 interact with CsMYC2 to directly regulate linalool synthase

For exploring how these MYB transcription factors regulate linalool biosynthesis, yeast one hybrid (Y1H) assay was performed to check the interaction between MYB proteins and CsTPS76 promoter (approximately 2 kb). The data indicated that all of MYB proteins could bind to the CsTPS76 promoter region (Figure 2a). Further, four MYB binding sites were detected in the CsTPS76 promoter region, namely P1, P2, P3 and P4 (Figure 2b). In order to confirm the binding sites of MYB proteins, we performed the Electrophoretic Mobility Shift Assay (EMSA). EMSA data showed that CsMYB68 could bind to the four sites, and CsMYB193 could bind to two of them, including P3 and P4. While CsMYB147 and CsMYB148 only could bind to the P3 site (Figure 2c–f).

Figure 2.

CsMYB68, CsMYB147, CsMYB148, CsMYB193 and CsMYC2 directly regulate CsTPS76 in tea plants. (a) Yeast one‐hybrid assay to identify the interaction between CsTPS76 promoter and CsMYB68, CsMYB147, CsMYB148 and CsMYB193 proteins, respectively. The four MYB genes were respectively subcloned into yeast one‐hybrid vectors pGADT7 to fused with the GAL4 AD domain. The CsTPS76 promoter region was cloned into the pHIS2.1 vector to fused with HIS3. The interaction was verified with growth in SD‐His. Images are representatives from at least three biological repeat experiments. (b) Predicted MYB binding sites in the promoter region of CsTPS76. Four MYB binding sites were predicted, including P1, P2, P3 and P4, and the sequence information was also listed. ATG, initiation codon. Identification of the MYB binding sites recognized by CsMYB68 (c), CsMYB147 (d), CsMYB148 (e) and CsMYB193 (f) in the promoter region of CsTPS76 by electrophoretic mobility shift assay (EMSA). The competitor represents the putative motif without the biotin label. The concentration of competitors with different rations with a biotin‐labelled motif is 50×. The biotin signal was indicated by a black arrow and highlighted with red colour of binding site names. (g) The expression of CsMYC2 was reduced significantly in the tea leaves treated with antisense oligonucleotides (asODN)‐CsMYC2 (CsMYC2‐KD) as compared with Sense ODN. The effect of CsMYC2 downregulation on the expression of CsTPS76 (h) and the concentration of linalool (i) in tea leaves. (j) Representative picture showing the process of transient expression assay in tea leaves. EV, empty vector. Transiently overexpressing CsMYC2 significantly activated the expression of CsTPS76 (k) and linalool content (l) in tea leaves. (m, n) Ectopically overexpressing CsMYC2 in Arabidopsis plants increased the expression levels of linalool synthases compared with the wildtype (WT). (o) Yeast one‐hybrid assay to identify the interaction between CsTPS76 promoter and CsMYC2 protein. The CsMYC2 was subcloned into yeast one‐hybrid vectors pGADT7 to fused with the GAL4 AD domain. The CsTPS76 promoter region was cloned into the pHIS2.1 vector to fused with HIS3. The interaction was verified with growth in SD‐Trp‐Leu‐His (SD‐L‐T‐H). Images are representatives from at least three biological repeat experiments. (p, q) Identification of the MYC binding sites recognized by CsMYC2 in the promoter region of CsTPS76 by EMSA. The competitor represents the putative motif without the biotin label. The concentration of competitors with different rations with a biotin‐labelled motif is 50×. The biotin signal was indicated by a black arrow and highlighted with red colour of binding site names. The data are expressed as means ± SD (*P < 0.05; **P < 0.01; Student's t‐test). CsACTIN or AtACTIN were introduced to normalize the expression data.

Previous studies have showed that MYC2 transcription factor is involved in regulating linalool biosynthesis in other plants (Lan et al., 2023; Yang et al., 2020). Here, we further investigated the function of CsMYC2 in linalool biosynthesis of tea plants. When the expression of CsMYC2 was suppressed in young tea leaves, we found that the expression level of CsTPS76 and the linalool concentration were significantly inhibited compared with the Sense ODN (Figure 2g–i). Further, CsMYC2 was transiently overexpressed in tea leaves. Transient overexpression of CsMYC2 led to significant elevation of CsTPS76 expression and linalool concentration compared with the control (Figure 2j–l). We also created the stable transgenic Arabidopsis overexpressing the CsMYC2 (Zhu et al., 2022) (Figure 2m). Gene expression analysis showed that compared with wildtype (WT), the expression levels of several monoterpene synthases, such as Terpene Synthase11 (AtTPS11), AtTPS14 and AtTPS21, were significantly induced in overexpression lines (Figure 2n). Then, we explored how CsMYC2 regulated linalool biosynthesis in tea plants. Y1H assay indicated that CsMYC2 could directedly bind to the promoter region of CsTPS76 (Figure 2o). EMSA was performed to validate the interaction between CsMYC2 and the G‐box sites in the CsTPS76 promoter region (Figure 2p,q).

Given both four MYB transcription factors and CsMYC2 participating in linalool biosynthesis in tea plants, the relationship between CsMYBs and CsMYC2 was investigated. Yeast two hybrid (Y2H) assay indicated the interactions between CsMYC2 with CsMYB68, CsMYB147 and CsMYB193, but not with CsMYB148 (Figure 3a). To validate the interactions, Bimolecular fluorescence complementation (BiFC) and glutathione S‐transferase (GST) pull‐down assays were performed. BiFC and pull‐down assays further indicated that CsMYC2 could interact with CsMYB68, CsMYB147, CsMYB148 and CsMYB193 to form divergent MYC‐MYB complexes respectively (Figure 3b–f; Figure S5). To explore the functions of MYC2‐MYB complexes in regulating linalool biosynthesis, we created a transient transactivation system with CsTPS76 promoter (Figure 3g). In Arabidopsis protoplasts, we found that the CsTPS76 promoter was slightly activated when MYB or MYC2 were separately expressed, while significantly activated by MYC2‐MYB complexes (Figure 3h).

Figure 3.

MYC2‐MYB complexes activate CsTPS76 promoter. (a) Yeast two‐hybrid assay to screen CsMYB proteins interacting with CsMYC2. CsMYBs and CsMYC2 in Matchmaker Gold yeast two‐hybrid vectors pGBKT7 and pGADT7 with reading frame in fusion with the GAL4 BD and AD were tested in different combinations for their interaction. The interaction was verified with growth in SD‐Trp‐Leu‐His‐Ade (SD‐T‐L‐A‐H). Images are representatives from at least three biological repeat experiments. (b) Bimolecular fluorescence complementation (BiFC) verification of the interaction between CsMYBs and CsMYC2 in tobacco epidermal cells. Pull‐down assays confirming the interaction between CsMYC2 and CsMYB68 (c), CsMYB147 (d), CsMYB148 (e) and CsMYB193 (f), respectively. Purified GST, GST‐CsMYB68, GST‐CsMYB147, GST‐CsMYB148, or GST‐CsMYB193 recombinant proteins were incubated with anti‐MBP‐conjugated magnetic beads containing MBP‐CsMYC2. Immunoblotting analyses with anti‐GST and anti‐MBP antibodies were performed to detect MBP‐CsMYC2 or GST‐CsMYBs. (g) Activator and repressor constructs and CsTPS76 promoter‐driven reporter gene constructs used in transient expression assay. (h) The activation of the CsTPS76 promoter in transient expression assay using Arabidopsis protoplasts with Renilla luciferase activity as reference. The data are from at least three biological replicates and are expressed as means ± SD. The difference is significant when **P < 0.01 in Student's t‐test.

CsJAZ12 interferes MYC2‐MYB complex formation to repress linalool biosynthesis in tea plants

CsMYC2 is not only a positive regulator of linalool biosynthesis, but also the key JA signal regulator in tea plants, implying the crucial role of JA in linalool biosynthesis. To comprehensively explore the function of JA on linalool biosynthesis, we focused on the JAZ proteins, which serve as the repressors of JA signal transduction. In tea genome, at least 13 JAZ members were detected, with dramatic tissue‐specific expression patterns (Gao et al., 2024; Shen et al., 2020). In which, only four JAZ genes, including CsJAZ1, CsJAZ8, CsJAZ12 and CsJAZ13, were highly expressed in young leaves and in developing flowers (Figure 4a,b; Table S4). In order to investigate whether JAZ proteins regulated linalool biosynthesis together with other linalool biosynthesis regulators, CsJAZ1, CsJAZ8, CsJAZ12 and CsJAZ13 were introduced into Y2H system to be screened by the MYB transcription factors. The results indicated that only CsJAZ12 could interact with all four MYB transcription factors (Figure 4c). Further, both BiFC and GST pull‐down assays were performed to verify the interactions between CsJAZ12 and CsMYBs in vivo and in vitro respectively (Figure 4d–h; Figure S5). In addition, we also detected the strict conserved JAZ‐MYC2 complex mediated by CsMYC2 and CsJAZ12 with Y2H and BiFC assays (Figure S6). These data imply that CsJAZ12 maybe is the major JAZ protein in regulating linalool biosynthesis via interfering the formation of MYC2‐MYB complexes.

Figure 4.

Identification of JAZ proteins in regulating linalool biosynthesis. Expression patterns of JAZ family genes in different tissues (a) and developing flowers (b) of tea plants. Blue rectangles were used to highlight the high‐expressed JAZ genes in tea plants. (c) Yeast two‐hybrid assay to screen the CsJAZ proteins in regulating linalool biosynthesis in tea plants. CsJAZs and CsMYBs in Matchmaker Gold yeast two‐hybrid vectors pGBKT7 and pGADT7 with reading frame in fusion with the GAL4 BD and AD were tested in different combinations for their interaction. The interaction was verified with growth in SD‐Trp‐Leu‐His‐Ade (SD‐T‐L‐A‐H). Images are representatives from at least three biological repeat experiments. (d) BiFC verification of the interaction between CsMYBs and CsJAZ12 in tobacco epidermal cells. Pull‐down assays confirming the interaction between CsJAZ12 and CsMYB68 (e), CsMYB147 (f), CsMYB148 (g) and CsMYB193 (h), respectively. Purified GST, GST‐CsMYB68, GST‐CsMYB147, GST‐CsMYB148, or GST‐CsMYB193 recombinant proteins were incubated with anti‐MBP‐conjugated magnetic beads containing MBP‐CsJAZ12. Immunoblotting analyses with anti‐GST and anti‐MBP antibodies were performed to detect MBP‐CsJAZ12 or GST‐CsMYBs.

To verify the function of CsJAZ12 on linalool biosynthesis, we generated the stable transgenic Arabidopsis overexpressing CsJAZ12 (Wang et al., 2024c). Expression analysis showed that compared with the WT, the expression levels of several monoterpene synthases, such as AtTPS11, AtTPS14 and AtTPS21, were significantly suppressed in CsJAZ12‐overexpression lines (Figure 5a). Then asODN assay was performed to interfere the CsJAZ12 expression. qRT‐PCR assay validated the suppression of CsJAZ12 expression in CsJAZ12‐KD compared with Sense ODN (Figure 5b). Metabolic analysis showed that linalool concentration was significantly increased in CsJAZ12‐KD compared with Sense ODN, which was consistent with the elevation of CsTPS76 expression level (Figure 5b,c).

Figure 5.

CsJAZ12 is a repressor of linalool biosynthesis in tea plants. (a) Ectopically overexpressing CsJAZ12 sharply inhibited the expression of genes related in linalool biosynthesis in Arabidopsis. WT, Wildtype. The effect of CsJAZ12 downregulation on the expression of CsTPS76 (b) and the concentration of linalool (c) in tea leaves. (d, e) CsJAZ12 inhibiting the activation of CsTPS76 promoter in transient expression assay using Arabidopsis protoplasts with Renilla luciferase activity as reference. (f, g) The effect of CsJAZ12 on the activation of CsTPS76 promoter in GUS‐report system in tobacco leaves. EV, empty vector. The data are from at least three biological replicates and are expressed as means ± SD. The difference is significant when *P < 0.05; **P < 0.01 in Student's t‐test.

To confirm the inhibition effect of CsJAZ12 on linalool biosynthesis, we conducted a transactivation/repression of firefly luciferase assay with CsTPS76 promoter (Figure 5d). The promoter activation assay showed that the activation activity of CsTPS76 promoter by the representative MYC2‐MYB complexes was significantly repressed, when CsJAZ12 was co‐expressed in Arabidopsis protoplasts (Figure 5e). Further, similar inhibition effect of CsJAZ12 on the activation activity of CsTPS76 promoter by the representative MYC2‐MYB complexes was observed in tobacco leaves with β‐glucuronidase (GUS) report system (Figure 5f,g). These data indicated the negative role of CsJAZ12 in linalool biosynthesis in tea plants.

MYC2‐MYB complexes regulating linalool biosynthesis during tea processing mediated by JA signalling

Most aroma compounds were generated largely during tea processing, particularly the terpenes. For example, in processing of white tea, green tea, black tea and oolong tea, the withering or spreading technologies were required to improve the floral aroma of tea (Feng et al., 2019). Linalool is the characterized volatile of several types of tea, such as white tea and green tea (Feng et al., 2019). The linalool, with a low level in fresh leaves, mostly was generated during tea processing, but the mechanism has not been unveiled (Zuo et al., 2023). For exploring the mechanism, we collected tea samples at 0, 12, 24 and 48 h during withering (Figure 6a), which was the key process of white tea processing. GC–MS analysis showed that withering significantly increased the linalool content, particularly at 24 h (Figure 6b). Consistently, we found that the expression levels of CsTPS76, CsMYB68 and CsMYB147 also increased during withering stage, but CsMYB148 or CsMYB193 not (Figure 6c,d), suggesting that CsMYB68 and CsMYB147 participated in linalool biosynthesis during tea withering.

Figure 6.

JA regulating linalool biosynthesis in tea processing via MYC2‐MYB complexes. (a) Schematic diagram of withering program of white tea processing. The samples were analysed at 0, 12, 24 and 48 h after treatment. (b) Linalool profiling during withering processing of white tea. (c) The expression pattern of CsTPS76 during withering processing of white tea. (d) The expression patterns of CsMYB68, CsMYB147, CsMYB148 and CsMYB193 during withering processing of white tea. (e) The expression of genes involved in JA biosynthesis and signal transduction were significantly activated during withering processing. (f) Withering processing promoted the JA and JA‐Ile accumulation in tea leaves. Exogenous MeJA treatment promoting the expression of CsMYB68 and CsMYB147 (g) and CsTPS76 (h), as well as the production of linalool (i) in tea leaves. (j) Schematic diagram of Diethylcarbamodithioic acid (DIECA) treatment on the tea leaves during withering processing. (k) Exogenous DIECA treatment significantly inhibited linalool biosynthesis during withering processing. The effect of DIECA treatment on the expression of CsTPS76 (l) and related transcription factors (m) during withering processing. (n, o) Ectopically overexpressing CsLOX2 sharply induced the expression of genes related linalool biosynthesis in Arabidopsis. WT, Wildtype. The data are from at least three biological replicates and are expressed as means ± SD. The difference is significant when *P < 0.05 and **P < 0.01 in Student's t‐test. CsACTIN or AtACTIN were introduced to normalize the expression data.

To answer why tea withering activated the expression of CsMYB68 and CsMYB147, we investigated the expression patterns of genes involved in JA biosynthesis and signalling during withering, for that many studies have illuminated that the burst of JA during tea processing played key roles for tea quality formation (Kong et al., 2024; Shi et al., 2019). Heatmap data suggested that JA biosynthesis and signal transduction pathways were sharply activated during withering processing (Figure 6e), as well as the concentrations of JA and jasmonoyl‐isoleucine (JA‐Ile) (Figure 6f). Subsequently, methyl jasmonate (MeJA) was used for tea plant treatment, and we found that both the expression levels of CsTPS76, CsMYB68 and CsMYB147 and the linalool content were significantly upregulated (Figure 6g–i). Then, sodium diethyldithiocarbamate (DIECA), the JA biosynthesis inhibitor, was applied in withering process (Figure 6j). DIECA treatment significantly reduced the linalool concentration (Figure 6k) and the expression level of CsTPS76 (Figure 6l) compared with control (CK). Additionally, the expression levels of CsMYB68, CsMYB147 and CsMYC2 were also suppressed by DIECA treatment (Figure 6m). For further exploring the effection of JA on linalool biosynthesis, we analysed the transgenic Arabidopsis plants with high endogenous JA concentration by expressing a key JA biosynthetic gene lipoxygenase2 (CsLOX2) (Wang et al., 2024c). Compared with wildtype (WT), elevation of endogenous JA strongly activated the expression of genes involved in linalool biosynthesis and regulation (Figure 6n,o). In conclusion, our data indicated that the increased linalool in tea processing partially relied on the activation of MYC2‐MYB complexes by JA signalling derived from sustained stresses (Figure S7).

Linalool contributes to cold tolerance of tea plants mediated by JA signalling

Previous studies have showed that linalool plays a crucial role in priming cold tolerance of tea plants via CBF‐dependent pathway (Zhao et al., 2020b); however, the regulation mechanism of linalool biosynthesis under cold stress is largely unknown in plants. Here, we also found that the linalool biosynthesis was significantly activated in tea leaves by cold treatment (Figure 7a,b). Consistently, the expression levels of CsTPS76, CsMYB68, CsMYB147 and CsMYC2 were upregulated by cold treatment (Figure 7c). In tea plants, JA plays a positive role in cold tolerance, which acts on the ICE‐CBF pathway (Wang et al., 2024c). It was not surprised that JA signalling participated in linalool biosynthesis regulation under cold stress, for the expression elevation of linalool regulator CsMYC2 under cold stress. Expectedly, the JA biosynthesis and signal transduction pathways were significantly activated in tea leaves under cold stress (Figure 7d), as well as the concentrations of JA and JA‐Ile (Figure 7e). Inversely, according to the chlorophyll fluorescence imaging, Fv/Fm (variable fluorescence/maximal fluorescence) values and nitroblue tetrazolium (NBT) staining, DIECA treatment aggravated the cold damage of tea leaves compared with control (Figure 7f–h). These data indicate that linalool pathway may as one of the downstream pathways of JA signalling to participate in cold resistance of tea plants. To validate our hypothesis, we turned back to investigate the cold tolerance of tea plants with silencing the CsMYBs, CsMYC2 or CsJAZ12, respectively. Both Fv/Fm values and the indicators of hydrogen peroxide indicated that CsMYB68, CsMYB147 and CsMYC2 served as the positive regulators of cold tolerance (Figure 7i,j), while CsJAZ12 as the negative regulator (Figure 7k), which were consistent with their regulatory functions on linalool biosynthesis in tea leaves.

Figure 7.

CsMYBs‐CsMYC2‐CsJAZ12 complex regulating cold tolerance in tea plants. (a) The production of linalool was significantly upregulated in tea leaves under cold treatment. (b) Elevation of the expression of CsTPS76 in tea leaves under cold treatment. (c) Cold treatment promoting the expression of CsMYB68, CsMYB147 and CsMYC2 in tea leaves. (d) Cold treatment activated the expression of genes involved in JA biosynthesis and signal transduction in tea leaves. (e) Cold stress significantly promotes JA concentration in tea leaves. (f–h) Exogenous DIECA reduced the cold tolerance of tea plants. (i) CsMYB68 and CsMYB147 positively regulated cold tolerance of tea plants. (j) CsMYC2 positively regulated cold tolerance of tea plants. (k) CsJAZ12 negatively regulated cold tolerance of tea plants. In f–k, chlorophyll fluorescence images were for indicating the cold tolerance of tea leaves. Photosystem‐II activity (Fv/Fm) values were indicated by pseudo‐colour images. The dark colour indicated the low Fv/Fm. The Fv/Fm values were used for quantitative analysis of cold tolerance. NBT (Nitrotetrazolium Blue chloride) staining was used as the indicator of ROS (Reactive oxygen species). The data are from at least three biological replicates and are expressed as means ± SD. The difference is significant when *P < 0.05 and **P < 0.01 in Student's t‐test. CsACTIN was introduced to normalize the expression data.

Demethylation of CsMYB68 and CsMYB148 contributes to cold tolerance of tea plants by improving linalool biosynthesis

Our previous study has showed that demethylation process played key roles in cold tolerance of tea plants by reducing methylation levels of a series of cold‐related genes to improve their expression (Tong et al., 2021). To know the function of DNA methylation on linalool biosynthesis under cold stress in tea plans, weighted methylation levels of the genes (including gene body region and 2 kb upstream and downstream of gene body) were analysed in tea leaves under cold stress. Interestingly, we found that the methylation levels of CsMYB68 and CsMYB147 were strongly decreased at 6 and 12 h after cold treatment, however, the methylation levels of their patterners, such as CsMYB148, CsMYB193, CsMYC2 and CsTPS76, showed no significant inhibition under cold stress (Figure 8a). Then we further analysed the expression patterns of demethylase genes in tea leaves under cold stress. Heatmap analysis showed that the demethylase family genes, such as REPRESSOR OF SILENCING 1 (ROS1), were significantly activated by cold treatment (Figure 8b), indicating that the demethylation process induced by cold stress may directly act on CsMYB68 and CsMYB147 genes to upregulate their expression and the linalool biosynthesis for improving cold tolerance of tea plants.

Figure 8.

Cold stress triggers the demethylation process of CsMYB68 and CsMYB147 in tea leaves. (a) Methylation level variation of the genes involved in linalool biosynthesis in tea leaves under cold treatment. The methylation data were extracted from our previous study (Tong et al., 2021), in which the tea leaves were treated by cold stress, and the samples were isolated at 0, 6 and 12 h after treatment. The reduction values were listed above the bars. (b) The REPRESSOR OF SILENCING 1 (ROS1) genes were induced by cold treatment in tea leaves. The expression level (Log10(TPM)) of each gene were showed in the heatmap boxes. (c) 5‐Azacytidine (5‐azaC) treatment improved the expression of CsMYB68 and CsMYB147 in tea leaves compared with the control. 5‐azaC treatment upregulated CsTPS76 expression level (d) and the corresponding linalool production (e). (f) Chlorophyll fluorescence images of tea leaves treated with 5‐azaC and the control. (g) Statistical analysis of Fv/Fm value in (f). (h) Activity of SOD in tea leaves treated with 5‐azaC and the control after cold treatment. (i) A model of MYC2‐MYB complexes coordinately regulating tea aroma formation and cold tolerance of tea plants. Tea flowers accumulate a large quantity of linalool compound, where CsMYB148 and CsMYB193 are the two predominant linalool regulators with highest expression levels compared with CsMYB68, CsMYB147 and CsMYC2. CsMYC2 could interact with the MYB transcription factors, including CsMYB68, CsMYB147, CsMYB148 and CsMYB193, to form MYC2‐MYB complexes synergistically regulating linalool biosynthesis in tea flowers. CsJAZ12 also highly expresses in tea flowers and serves as an inhibitor by interfering the formation of MYC2‐MYB complexes to finely regulate linalool biosynthesis. In tea processing, the sustained damages, such as withering and rolling, trigger endogenous JA biosynthesis and signal transduction pathways, leading to CsJAZ12 degradation by 26S proteinase system to free CsMYC2, CsMYB68 and CsMYB147 proteins, to promote the MYC2‐MYB complex formation to enhance the expression of linalool synthase and linalool biosynthesis. The linalool improves the tea quality by contributing to the floral aroma. Under cold stress condition, demethylase genes are activated to trigger DNA demethylation process, leading to reduced DNA methylation levels of CsMYB68 and CsMYB147 and elevation of their expression levels. Meanwhile, the JA biosynthesis and signal transduction pathways are also activated under cold stress, which further initiates CsJAZ12 degradation to release CsMYC2, CsMYB68 and CsMYB147 proteins. Integration of demethylation and JA signalling pathways promotes the formation of MYC2‐MYB complexes to enhance the expression of linalool synthase and linalool biosynthesis in tea leaves. The linalool enhances the resistance of tea plants against the cold stress by reducing the ROS level. The data are from at least three biological replicates and are expressed as means ± SD. The difference is significant when *P < 0.05 and **P < 0.01 in Student's t‐test. CsACTIN was introduced to normalize the expression data.

To further confirm the function of DNA demethylation on linalool biosynthesis, the methylation inhibitor 5‐azacytidine (5‐azaC) was used for treating the young tea leaves. qRT‐PCR assay showed that the expression levels of CsMYB68 and CsMYB147, but CsMYB148 or CsMYB193 not, were significantly upregulated by 5‐azaC (Figure 8c), which was consistent with the decreased methylation levels of CsMYB68 and CsMYB147 under cold stress. Expectedly, both the expression level of CsTPS76 and the linalool concentration were significantly elevated in tea leaves under 5‐azaC treatment (Figure 8d,e). We then tested the function of 5‐azaC in cold tolerance of tea plants. Both Fv/Fm and superoxidedismutase (SOD) data showed that 5‐azaC could significantly improve the cold tolerance of tea plants (Figure 8f–h), suggesting that linalool pathway may as one of the downstream pathways of demethylation process to participate in cold resistance of tea plants.

Discussion

Divergent regulation mechanism of linalool biosynthesis in tea flowers and leaves

Linalool is one of the most abundance of terpenoids in plants, and its biosynthesis pathway and regulation mechanism have been widely studied. Recently, several linalool biosynthesis regulators have been characterized in Arabidopsis, Freesia hybrida, peach, and Osmanthus fragrans, and all these regulators directly regulated linalool synthases (Lan et al., 2023; Wei et al., 2021, 2022; Yang et al., 2020). Particularly in flowers, the transcriptional regulation mechanism of linalool is seemly conserved. MYB21 and its paralogs, together with MYC2, form transcriptional complexes to regulate linalool biosynthesis in flower. Both previous study and our data showed that the linalool was the predominant terpenoid compound in tea flowers (Liu et al., 2018; Tang et al., 2022). Integrating phylogenic analysis, expression pattern analysis, and heterologous/homologous overexpression assays, a conserved regulatory module of linalool biosynthesis was found in tea flowers. In this module, CsMYB148 and CsMYB193, the MYB21‐type transcription factors, specifically expressed in tea flowers and directly regulated the linalool synthase, meanwhile the regulatory process was modulated by MYC2 and JAZ at post‐transcriptional level, indicating that JA participated in linalool biosynthesis in tea flowers. Further, we found the linalool biosynthesis in tea flowers was also activated by MeJA treatment, and the CsMYB148 with higher expression level might be the major MYB regulator of linalool biosynthesis under JA signalling (Figure S8A).

Abundance of linalool was also detected in tea leaves under both normal condition and stress condition. But, the MYB21‐type transcription factors, CsMYB148 and CsMYB193, did not express in leaves under normal or stress conditions. However, their patterners, CsMYC2 and CsJAZ12 showed high expression levels in both flowers and leaves, indicating the divergent regulation mechanism of linalool biosynthesis in tea flowers and leaves. Based on phylogenic analysis, expression pattern analysis, and heterologous/homologous overexpression assays, two MYB21‐like type transcription factors, CsMYB68 and CsMYB147, with high expression levels in both flowers and leaves, were identified as the linalool regulators in tea leaves. These two MYB21‐like type transcription factors regulated linalool biosynthesis in tea leaves and flowers with the similar way of CsMYB148 and CsMYB193, by being interacted with CsMYC2 and CsJAZ12 to form complexes to directly modulate linalool synthase. In tea leaves, CsMYB68 and CsMYB147 were highly expressed, particularly under multiple stresses, such as cold stress and MeJA treatment. But in tea flowers, CsMYB68 and CsMYB147 were lowly expressed and were slightly regulated by MeJA treatment and cold stress (Figure S8A,B). On the contrary, CsMYB148 and CsMYB193 were specifically expressed in tea flowers, and in which CsMYB148 was significantly regulated by MeJA treatment and cold stress (Figure S8A,B). Interestingly, the expression of CsMYB193 was less affected by MeJA treatment or cold stress in tea flowers (Figure S8A,B). The divergent expression patterns of these MYB transcription factors under normal condition or stress condition imply the important role of CsMYB68 and CsMYB147 in tea leaves, but CsMYB148 and CsMYB193 in tea flowers, in the linalool biosynthesis.

JA and its derivatives play key roles in tea aroma formation

Unsaturated fatty acid degradation is one of the major ways of tea aroma sources (Ho et al., 2015). In the degradation products, JA and its derivates are derived from α‐linolenic acid with intense floral and sweet odours. JA and its derivates are the representative aroma in almost all type tea, such as oolong tea, green tea and black tea (Ho et al., 2015). (Z)‐methyl epijasmonate, derived from jasmonic acid, is the key contributor to orchid‐like aroma properties in tea, which is mainly synthesized and accumulated during tea processing (Feng et al., 2023). Another jasmonic acid derivative is jasmone, a key aroma in oolong tea and green tea (Wang et al., 2024b). These compounds are predominantly generated at early stage of tea processing, with a low concentration in fresh leaves. For example, increased expression of CsLOX and Multifunctional Protein (CsMFP2) mediated by chromatin accessibility led to JA hyperaccumulation during withering process, contributing to aroma formation (Kong et al., 2024).

Besides that, JA also could upregulate downstream flavour‐related metabolic pathways to improve tea aroma formation. In tea processing, the JA and JA‐Ile induced by sustained stresses could directly upregulate the biosynthesis of multiple flavour‐related compounds by signal transduction pathway, such as (E)‐nerolidol, indole, α‐farnesene (Gu et al., 2022; Wu et al., 2023; Zhou et al., 2020c). In this study, the linalool was synthesized and accumulated during tea processing together with JA hyperaccumulation. Given the JA regulator CsMYC2 also serving as the regulator of linalool biosynthesis, the activation of linalool pathway in tea processing was associated with the increased JA concentration. Subsequently, exogenous JA and JA inhibitor treatment assay confirmed that the linalool biosynthesis in tea processing was partially dependent on the JA biosynthesis and signalling pathways.

As JA signalling repressors, JAZ proteins have been involved in linalool biosynthesis in plants. In rice, OsJAZ8 served as a repressor of JA‐induced linalool biosynthesis, but the regulation mechanism was not fully known (Taniguchi et al., 2014). Here, CsJAZ12 was also identified as a linalool biosynthesis repressor, and we found that CsJAZ12 could interact with not only CsMYC2, but also CsMYB68 and CsMYB147, to interfere activated MYC2‐MYB complex formation to decrease the expression of linalool synthase in tea leaves. CsJAZ12 highly expressed in tea leaves and flowers, meanwhile CsJAZ12 could interact with CsMYB148 and CsMYB193 which highly and specifically expressed in tea flowers, implying that CsJAZ12 maybe also participated in regulating linalool biosynthesis in tea flowers with similar mechanism.

JA‐linalool is a novel cold‐regulation module acting as an upstream signal of CBF pathway

Previous study has illuminated that linalool could improve cold tolerance of tea plants, by upregulating CBF pathway (Zhao et al., 2020b). However, the mechanism of linalool in cold tolerance of tea plants is largely unknown in plants. JA plays a positive role in cold tolerance, by affecting the stability of JAZ proteins to regulate ICE‐CBF signalling pathway. In addition, the MYC2‐JAZ module could further regulate the biosynthesis of cold‐related metabolites, such as polyamine, glutathione, and anthocyanins, under cold stress (Hu et al., 2013). In tea plants, JA could significantly reduce the cold damage, and its dynamic concentration was modulated by LUX ARRHYTHMO (LUX). Meanwhile, LUX could competitively interact with JAZ protein to destroy the inhibition of JAZ protein on ICE‐CBF signalling pathway (Wang et al., 2024c). Here, we further found that JA‐linalool is a novel cold‐regulation module in tea plants. The linalool biosynthesis under cold stress was depend on JA signalling pathway. The degradation of JAZ protein induced by cold stress contributed to the formation of MYC2‐MYB complexes to activate linalool synthase expression and linalool biosynthesis, and subsequently these cold‐induced linalool positively acted on the CBF genes to improve the cold tolerance of tea plants.

The roles of JA‐linalool module in divergence of cold tolerance in CSS and CSA

Divergence of cold tolerance in sinensis type (CSS) and assamica type plants (CSA) has been observed in previous study (Tong et al., 2024); in this study, we further explored the role of JA‐linalool module in the divergence. We collected the RNA‐seq data of CSA from Yunnan province with low latitude and CSS from Zhejiang province with high latitude (Zhang et al., 2020), and found that CSS had a higher level of JA‐Ile because of its higher or lower expression levels of JA biosynthetic or catabolic genes, compared to CSA (Figure S9A,B). Interestingly, we found that CsMYC2, not CsMYB68 and CsMYB147, had higher expression levels in CSS than that of CSA, leading to higher content of linalool in CSS (Figure S9C–E). These data seemly suggested that the divergent linalool concentration was mainly regulated by the JA, not the MYB transcription factors, in the CSA and CSS at population level. To know the mechanism of divergent JA biosynthesis, two factors, including the genetic divergence between CSS and CSA and the growth environment, were analysed. In our previous study, the selected genes in CSS and CSA against Camellia taliensis (CTA, Wild) have been identified (Tong et al., 2024). We further analysed the effect of cold stress on these selected genes. Unexpectedly, in these selected genes, similar numbers of genes activated by cold stress were observed in CSS and CSA (Figure S9F; Table S6), meanwhile, all genes involved in JA biosynthesis or catabolism pathways were not included in the selected genes after annotation analysis. But a JAZ gene, which was a JA signalling component and could be induced by cold stress, was found specifically in CSA/Wild group (Figure S9F; Table S6).

To know the roles of growth environment on JA pathway in CSA, three cultivars of CSA were transferred from Yunnan province to Anhui province (high latitude) to keep similar growth environment. After about 2 years, we found that the growth environment did not significantly improve the cold tolerance of CSA compared to CSS (Figure S9G,H). But we found that these CSA plants contained equivalent linalool and JA compared with CSS under normal condition (Figure S9K–M). Particularly under cold condition, CSA had higher levels of JA and linalool compared to CSS (Figure S9K–M), which was conflicting with their cold resistance. These data indicate that linalool biosynthesis is regulated by the environment‐induced JA in both CSS and CSA, but the linalool, as one of JA downstream pathways, plays a limited role in the divergence of cold tolerance in CSS and CSA.

The positive roles of JA homeostasis in divergence of cold tolerance in CSS and CSA

Given the complex functions of JA on cold tolerance, JA homeostasis is necessary for plants to balance tolerance and growth. CsLUX was a positive regulator of cold stress in tea plants, which could accurately modulate JA concentration in short‐term and long‐term cold stress oppositely to improve the cold tolerance. At early stage of cold stress, CsLUX was suppressed to upregulate CsLOX2 expression and the JA concentration. But the CsLUX was significantly activated at later stage leading to inhibition of JA biosynthesis. Meanwhile, CsLUX not only served as a direct downstream target gene of CsCBF1, but also interacted with JAZ to regulate ICE‐CBF pathway. Those data indicate that CsLUX participated in cold response mainly via regulating JA homeostasis, particularly at later stage of cold stress, in tea plants (Wang et al., 2024c).

Warm temperature promotes JA catabolism by triggering JASMONATE‐INDUCED OXYGENASE (JOX) and sulfotransferase (ST2A) genes (Zhu et al., 2021). In this study, we found that CSA has not only higher expression levels of CsJOX and CsST2A, but also lower expression levels of early genes related to JA biosynthesis, together leading to lower concentration of bioactive JA, compared with CSS. Further, genetic variation analysis showed that no significant selecting signalling of JA metabolic genes was detected in CSS/Wild or CSA/Wild, implying that the difference of JA metabolism between CSS and CSA was associated with the JA signalling pathway or growth condition. Another evidence was that when the CSA plants were transferred to the growth condition of CSS, equal concentration of JA between CSA and CSS was detected. After short‐time cold stress, the JA homeostasis was observed in CSS, but JA burst was detected only in CSA. JA metabolism disorders maybe was the major factor of cold‐hypersensitization in CSA.

JA metabolism disorders in CSA probably was associated with JA signal transduction. In CSA genome, a cold responding JAZ gene was specifically selected. In plants, JAZ proteins not only serve as the repressor of JA signal transduction, but also play inconsistent roles in JA biosynthesis under multiple stresses (Singh et al., 2020; Yan et al., 2018). As the JAZ gene selected, diminished JA signal transduction maybe effected the JA feedback loop in CSA leading to excess JA under cold stress. Subsequently, the excess JA may trigger the leaf senescence of CSA rashly.

Independent roles of DNA methylation with JA signalling on linalool biosynthesis under cold stress

Dynamic DNA methylation plays a crucial role in linalool biosynthesis in plants by finely altering the DNA methylation levels of linalool synthase or related transcription factor genes (Duan et al., 2023; Wei et al., 2021, 2022). In tea plants, several studies have showed that the DNA methylation was also associated with the biosynthesis of flavour‐related compounds, such as indole, in tea processing and in different growth conditions (Gu et al., 2021; Han et al., 2024; Kong et al., 2023; Wang et al., 2019; Yang et al., 2021). Here, we found that DNA methylation was involved in cold‐induced linalool biosynthesis in tea leaves. Unexpectedly, cold stress did not decrease the DNA methylation levels of CsTPS76 or the JA regulator gene CsMYC2, but the DNA methylation levels of the upstream regulator genes, CsMYB68 and CsMYB147, indicating that JA signalling and DNA methylation pathways may be independent in regulating linalool biosynthesis in tea leaves under cold condition. Our previous study has showed that tea plants could globally reshape gene expression patters to improve cold tolerance by changing genome‐wide DNA methylation (Tong et al., 2021). In this study, we also found that the expression levels of DNA demethylase family genes were significantly upregulated under cold stress in tea leaves. Meanwhile, 5‐azaC treatment contributed to cold tolerance of tea leaves, as well as the linalool biosynthesis, but not in tea flowers (Figure S8C). Thus, linalool pathway may as one of the downstream pathways of DNA methylation process to participate in cold resistance of tea plants.

In summary, we uncovered the regulatory mechanism of linalool biosynthesis in flowers and leaves of tea plants, and found the divergent regulation mechanism in flowers and leaves was mediated by the spatiotemporal expression of MYB transcription factors. Further, we identified the positive roles of the regulatory module of linalool biosynthesis on tea aroma formation during tea process and cold tolerance of tea plants (Figure 8i). Detailly, in tea flowers, CsMYB68, CsMYB147, CsMYB148 and CsMYB193 are expressed, and in which CsMYB148 and CsMYB193 play important roles for their highest expression levels. These MYB transcription factors can interact with CsMYC2 to form activated MYC2‐MYB complexes to directly regulate linalool synthase. Meanwhile, the JAZ protein can destroy the MYC2‐MYB complexes by interacting with their components to negatively regulate linalool biosynthesis. In tea leaves, for loss of expression of CsMYB148 and CsMYB193, CsMYB68 and CsMYB147 play dominant roles in regulating linalool biosynthesis by interacting with CsMYC2 to form MYC2‐MYB complexes. In tea processing, JA biosynthesis and signal transduction pathways are triggered, subsequently resulting in JAZ degradation to free the activated MYC2‐MYB complexes to improve aroma formation. Under cold condition, the JA biosynthesis and signal transduction pathways are also activated in tea leaves. The cold‐induced JA reduces the JAZ stability, leading to release of MYC2‐MYB complexes. Meanwhile, cold stress upregulates the expression of CsMYB68 and CsMYB147 by reducing their DNA methylation levels. By integrating JA and DNA demethylation pathways, the abundance of MYC2‐MYB complexes is elevated to enhance the cold tolerance of tea leaves. Our study not only unveil the divergent regulatory mechanism of linalool biosynthesis in tea plants and its functions in tea quality formation and cold tolerance, but also provide a new insight into synergistically improving the tea quality and tea plant resistance.

Experimental procedures

Plant materials and growth conditions

All tea plant varieties (Camellia sinensis L.) were grown at the tea plantation of Anhui agricultural university (Hefei, Anhui Province, China), and were used for gene cloning, phenotype survey and metabolomic analysis. The tea cutting seedlings (about 2‐year‐old Shuchazao and Longjing43 cultivars) were cultivated and treated in the artificial climate chamber (25 °C, 16/8 h photoperiod, 65% humidity). The transgenic Arabidopsis lines, including CsMYC2OE, CsJAZ12OE and CsLOX2OE, were described as previously (Wang et al., 2024c; Zhu et al., 2022). All the tobacco and Arabidopsis plants were grown at 22 °C with 16/8 h photoperiod at approximately 120 μmol m⁻² s⁻¹.

Cloning and construction of binary vectors for plant expressing

The Coding Sequences (CDSs) of CsMYB68, CsMYB147, CsMYB148, CsMYB193, CsMYC2, and CsJAZ12 were amplified from the cDNAs of leaves or flowers prepared from Camellia sinensis L. cv. Shuchazao using gene‐specific primers (Table S7), and all sequences were identified by Sanger sequencing. Then, the genes were subcloned into the binary vectors pBI121 or pLGFP1301, respectively, for tobacco transformation or tea plant hairy root transformation.

Targeted metabolite profiling

Linalool was analysed by the Headspace solid‐phase microextraction GC–MS according to previous method (Li et al., 2020). Linalool was identified by comparing the standard substance. The ethyl caprate (0.2 μg g⁻¹, Sigma, USA) was served as the internal standard to quantification. At least five repeats for each sample were performed. JA concentration was measured on a LC–MS/MS System (SCIEX Triple Quad 5500) equipped with a C18 column (Phenomenex, 4.6 × 250 mm, 5 μm) according to previous method (Wang et al., 2024c). Target hormone components were identified by comparing the retention time and molecular weight to authentic compounds. Five biological replicates were included for each assay. The standard curves were calculated to quantify the compounds.

Transformation of tobacco

For tobacco transformation, the CDSs of CsMYB68, CsMYB147, CsMYB148 and CsMYB193 were recombined into pBI121 and then transformed into Agrobacterium tumefaciens EHA105 respectively. The tobacco (Nicotiana tabacum cv, Xanthi) transformation was performed according to the described methods (Shen et al., 2019). The transformants were selected on half‐strength Murashige and Skoog (MS) medium containing 20 mg L⁻¹ Kanamycin and were further confirmed by genomic Polymerase Chain Reaction (PCR) and Real‐Time Quantitative Reverse Transcription PCR (qRT‐PCR).

RNA extension and qRT‐PCR

Total RNA was isolated from each sample by Trizol reagent (Invitrogen) according to the manufacture's protocol. Then, approximately 1 μg of total purified RNA was used to construct the cDNA library by the SuperScript III kit (Invitrogen). qRT‐PCR was carried out on the iQ5 real‐time PCR machine with gene‐specific primers (Table S7). The real‐time data were acquired and parsed by the manager software (Bio‐Red). The housekeeping genes were used to normalize the relative expression of related genes using a modified 2^−ΔΔCT method. All RT‐qPCR and RT‐PCR analyses were performed with three independent biological replicates.

Yeast two‐hybrid (Y2H) and one‐hybrid (Y1H) assays

For Y2H assay, the Open Reading Frames (ORFs) of CsMYB68, CsMYB147, CsMYB148, CsMYB193, CsJAZ1, CsJAZ8, CsJAZ12 and CsJAZ13 were respectively subcloned in frame with the GAL4 DNA‐binding domain (BD) into the pGBKT7, and CsMYC2 was subcloned in frame with the GAL4 Transcriptional activation domain (AD) into the pGADT7 vector. The constructs with different combinations were co‐transformed into the yeast strain AH109 to identify the protein interactions by the growth test on the SD medium without Tryptophan (Trp), Leucine (Leu), Histidine (His), and Adenine (Ade) (SD‐T‐L‐H‐A) according to the method described previously (Li et al., 2022a). For Y1H assay, approximately 2.0 kb CsTPS76 promoter was inserted into the pHIS2.1 vector. Meanwhile, the ORFs of CsMYB68, CsMYB147, CsMYB148, CsMYB193 and CsMYC2 were inserted into pGADT7 to generate the CsMYB68‐AD, CsMYB147‐AD, CsMYB148‐AD, CsMYB193‐AD and CsMYC2‐AD constructs. The constructs were transformed into Y187 harbouring pHIS2‐proCsTPS76. The interactions between transcription factors and promoter were tested on the SD medium without His. All experiments above were carried out by at least three independent biological replicates.

Bimolecular fluorescence complementation (BiFC) and subcellular localization

For BiFC assay, the ORFs of CsMYB68, CsMYB147, CsMYB148, CsMYB193 and CsJAZ12 (stop codon removed) were subcloned in frame with the YFP^N into the pFGC‐YN173 vector, and the ORFs of CsMYC2 and CsJAZ12 (stop codon removed) were subcloned in frame with the YFP^C into the pFGC‐YC155 vector. The resulting constructs were then transformed into A. tumefaciens strain GV3101 respectively for co‐expression in tobacco plants. The strains harbouring different recombinant plasmids were mixed and then injected into leaves of about 3‐week‐old tobacco plants (Nicotiana benthamiana) as the method described previously (Li et al., 2022a). For subcellular localization analysis, the ORFs of the genes, including CsMYB68, CsMYB147, CsMYB148 and CsMYB193, were subcloned into pLGFP1301, respectively, and then were transformed into A. tumefaciens EHA105 strains for transient expression. Three‐week‐old Nicotiana benthamiana leaves were used for infiltration with strains harbouring the resulting constructs. At 48 h after infiltration, the Green fluorescent protein (GFP) fluorescence signals were detected using confocal microscopy (Leica DMi8 M; Germany) at least three independent biological replicates.

Pull‐down assay

The ORFs of CsMYB68, CsMYB147, CsMYB148 and CsMYB193 were subcloned into the pGEX‐4‐T vector to generate GST‐CsMYB68, GST‐CsMYB147, GST‐CsMYB148 and GST‐CsMYB193, respectively. Similarly, the ORFs of CsMYC2 and CsJAZ12 were inserted into the pMAL‐c2x vector to product MBP‐CsMYC2 and MBP‐CsJAZ12. The GST, GST‐fused CsMYB68, CsMYB147, CsMYB148, CsMYB193, and MBP‐fused CsMYC2, CsJAZ12 were expressed in Escherichia coli BL21 and purified by amylose resin beads as described previously (Gao et al., 2021). For the GST pull‐down assay, approximately 50 mg of purified GST and GST‐fused proteins were respectively incubated with equal MBP‐CsMYC2 or MBP‐CsJAZ12 proteins for 2 h at 4 °C with gentle rocking in pull‐down buffer (20 mm Tris–HCl (pH 7.4), 200 mm NaCl, 1 mm EDTA, 1 mm PMSF, and 1 mm DTT) containing 150 μL amylose resin beads. After centrifugation, the Tris‐Buffered Saline (TBS) buffer (10 mm Tris (pH 7.4), 150 mm NaCl, 1 mm EDTA, and 1% Triton X‐100) was used for washing the residue for five times. Then, the residue was resuspended in SDS loading buffer and immunoblotted using the anti‐MBP antibody (Proteintech Group) (1 : 5000) and anti‐GST antibody (Beyotime) (1 : 5000). The immunoblotting signals were scanned with the BioRad (ChemiDoc MP) imaging system.

Electrophoretic mobility shift assay (EMSA)

The purified GST‐CsMYB68, GST‐CsMYB147, GST‐CsMYB148, GST‐CsMYB193 and MBP‐CsMYC2 proteins were used for EMAS. The EMSA was performed with approximately 2 μg of purified proteins using the EMSA kit (Beyotime, Shanghai, China) in 6.6% nondenatured polyacrylamide gel according to the manufacture's protocol and previous study (Gao et al., 2021).

Promoter trans‐activation assay

Approximately 2.0 kb sequence upstream the translation start codons of CsTPS76 was isolated from tea genomic DNA (Shuchazao cultivar) using the special primers (Table S7), and then subcloned into pGreenII0800‐LUC to generate the luciferase reporter construct. The effecter genes, including CsMYB68, CsMYB147, CsMYB148, CsMYB193, CsMYC2 and CsJAZ12 were cloned, respectively, into pGreenII62‐SK to generate the effector constructs. The reporter and effector constructs with different combination were co‐expressed in Arabidopsis mesophyll protoplasts according to the method described previously (Yoo et al., 2007). Luciferase activities were measured by Dual‐Luciferase Reporter Assay System (Promega) according to the manufacture's protocol. The Renilla luciferase activity was introduced as the internal control to normalize Luciferase activity value. The relative LUC activity was calculated by normalizing to that of REN in three biological triplicates.

Treatment on tea plants

For withering treatment, fresh tea leaves were isolated from Fudingdabai cultivar grown in Anhui tea plantation (Anhui agricultural university). The manufacturing processes were performed according to previous method (Zuo et al., 2023). Tea leaves at various withering degrees (0, 12, 24 and 48 h) were collected for gene expression and metabolism analyses. Three replicates were processed for white tea manufacturing. For hormone treatment, the tender tea shoots of cutting seedlings or blooming tea flowers were sprayed with 100 μm Methyl Jasmonate (MeJA) solution or 100 μm Sodium diethyldithiocarbamate trihydrate (DIECA), while the controls were treated with distilled water. The samples treated by MeJA were harvested at 0 H, 3 H, 6 H, 12 H and 24 H after treatment, and the samples treated by DIECA were collected at 24 H, for metabolism and gene expression analyses. For cold treatment, about 2‐years‐old cutting seedlings or blooming tea flowers were treated under 0 °C, and the samples were collected at 0 H, 6 H, 12 H, 24 H and 48 H for metabolism and gene expression analyses. For cold treatment of tea leaves silenced with AsODN, the materials were treated with 0 °C for 1 h and recovery of 25 °C darkness for 25 min, then the maximum photosynthetic efficiency of photosystem II (Fv/Fm), H₂O₂ and SOD contents were measured with assay kit provided by Nanjing Jiancheng Bioengineering Institute. To evaluate the functions of 5‐azaC and DIECA on the cold tolerance of tea plants, about 2‐years‐old cutting seedlings or blooming tea flowers were treated with 100 μm 5‐azaC or 100 μm DIECA, and the controls were treated with distilled water. At 48 H after treatment, tea leaves were further treated under 0 °C for 1 h.

Stable/transient transformation and antisense oligonucleotides (AsODN) assay in tea plants

For stable transformation of tea hairy roots, the ORFs of CsMYB68 and CsMYB148 were cloned into pLGFP1301 vector, respectively. The pLGFP1301‐CsMYB68, pLGFP1301‐CsMYB148 and pLGFP1301 (empty vector) were transformed into A. rhizogenes ATCC15834, respectively. The leaves of C. sinensis cv. Shuchazao were used for transformation as described method (Ma et al., 2023). For transient transformation, the pBI121‐CsMYC2 and pBI121 (empty vector) plasmids were transformed into A. tumefaciens GV3101 and then were injected into both sides of the leaves with the main vein as the boundary. After 48 h of culture, tea samples were collected respectively to determine gene expression and linalool content. Gene suppression of CsMYBs, CsJAZ12, and CsMYC2 in tea plants using AsODN was carried out as described previously (Xie et al., 2014) with slight modifications. Briefly, the AsODN of target genes was designed and selected respectively using Soligo software (Xie et al., 2014) (Table S7). The tender tea shoots (Shuchazao cultivar) with leaves were cut off and partially inserted into Eppendorf tubes to incubate with 2 mL of 20 μm AsODN solution in a plant incubator, and the solution with sense oligonucleotides of the corresponding gene as the control. After 18–72 h of incubation, the tender shoots were collected for further metabolism and transcriptome analyses. Three independent measurements consisted of 10 tea shoots.

Bioinformatic analysis

All heatmaps of gene expression and Ks analysis were constructed using the TBtools software (Chen et al., 2023). Phylogenetic trees were constructed with MEGA11 software using the neighbour‐joining method with 1000 bootstrap replications and multiple alignments were constructed using the ClustalW2 program (Tamura et al., 2021).

GUS activity assay

The promoter of CsTPS76 amplified from tea cultivar (Shuchazao) was subcloned into pBI121 vector to generate proCsTPS76::GUS reporter system. Then, the GUS driven by CsTPS76 promoter with CsMYB68, CsMYB148, CsMYC2 or GFP control was co‐expressed in a same N. benthamiana leaf (about 3‐week‐old) mediated by A. tumefaciens infiltration. About 2 days after infiltration, the GUS activity was detected by leaf staining method as described previously (Li et al., 2022c). The signals were recorded from three biological replicate.

Statistical analysis

All data were obtained from at least three independent experiments. The confidence limit 95% or 99% were defined as the significant between two‐tailed data in Student's t‐test. For the Y1H assay, Y2H assay, BiFC, and plant phenotypic display, only representative pictures were shown.

Accession numbers

Nucleotide sequences of genes in this study can be download in the TAIR database (https://www.arabidopsis.org/) or Tea Plant Information Archive database (http://tpia.teaplants.cn/) under following accession numbers: CsMYB68, TEA012423; CsMYB147, TEA024384; CsMYB148, TEA024596; CsMYB193, TEA030389; CsMYC2, TEA000833; CsTPS76, TEA007191; CsTPS77, TEA004822; CsJAZ1, TEA001681; CsJAZ2, TEA002032; CsJAZ3, TEA001414; CsJAZ4, TEA033836; CsJAZ5, TEA033832; CsJAZ6, TEA014550; CsJAZ7, TEA004474; CsJAZ8, TEA032228; CsJAZ9, TEA001501; CsJAZ10, TEA013465; CsJAZ11, TEA001821; CsJAZ12, TEA030190; CsJAZ13, TEA027049; CsLOX2, CSS0033612; CsAOS, TEA001041; CsHPL, TEA008699; CsJAR, TEA010095; AtTPS11, AT5G44630; AtTPS14, AT1G61680; AtTPS21, AT5G23960; AtMYB21, AT3G27810; AtMYB24, AT5G40350.

Conflict of interest

The authors declare no conflict of interest in the paper.

Author contributions

P.H.L. conceived and designed the experiments; Y.L.L., R.Y., and Y.J.Q. performed experiments and analysed the data; X.Y.L., Z.Q.Z., Y.R.Z., and Z.L.Y. performed quantitative analysis of the phytohormone; Y.L.L., R.Y., and X.Y.S. performed GC–MS analysis; W.T., and E.H.X. performed RNA‐seq and genetic evolution analyses; P.H.L., and E.H.X. wrote the manuscript.

Supporting information

Table S1 The expression patterns of MYB genes in different tissues of tea plants

Table S2 The expression levels of MYB genes in different cultivars

Table S3 The expression patterns of MYB genes under biotic stress

Table S4 The expression patterns of JAZ family in different tissues of tea plants

Table S5 The expression of genes related to JA pathway in CSS and CSA

Table S6 The expression of genes from CSS/Wild and CSA/Wild in tea leaves under cold treatment

Table S7 List of primers used in this study

Figure S1 Sequence analysis of MYB transcription factors related to linalool biosynthesis in tea plants.

Figure S2 Expression patterns of CsTPS77 in tea plants.

Figure S3 Ectopically expressing CsMYB68, CsMYB147, CsMYB148 and CsMYB193 in tobacco plants.

Figure S4 CsMYB68, CsMYB147, CsMYB148 and CsMYB193 positively regulated linalool synthase in tobacco plants.

Figure S5 Negative controls of BiFC assay.

Figure S6 Conservative JAZ‐MYC2 complex mediated by CsMYC2 and CsJAZ12.

Figure S7 The expression of genes involved in Ca²⁺ signalling and MAPK pathway were significantly activated during withering processing.

Figure S8 The linalool biosynthesis was regulated by MeJA, 5‐azaC and cold treatment in tea flowers.

Figure S9 JA played the key role in divergence of cold tolerance in CSS and CSA.

Data availability statement

All data supporting the findings in this study can be available and found in the Supporting Information. The sequences for new genes, promoters in this study are given in the “Method” section. The research materials related to these findings can be available for distribution upon request, to Penghui Li at lphui2012@126.com; phli@ahau.edu.cn.