Genome-Wide Identification of the Tify Gene Family and Their Expression Profiles in Response to Biotic and Abiotic Stresses in Tea Plants (Camellia sinensis)

The TIFY family is a plant-specific gene family that is involved in regulating a variety of plant processes, including developmental and defense responses. The chromosome-level genome of the tea plant (Camellia sinensis) has recently been released, but a comprehensive view of the TIFY family in C. sinensis (the CsTIFY genes) is lacking. The current study performed an extensive genome-wide identification of CsTIFY genes. The phylogenetics, chromosome location, exon/intron structure, and conserved domains of these genes were analyzed to characterize the members of the CsTIFY family. The expression profiles of the CsTIFY genes in four organs were analyzed, and they showed different spatial expression patterns. All CsJAZ genes were observed to be induced by jasmonate acid (JA) and exhibited different responses to abiotic and biotic stresses. Six of seven CsJAZ genes (CsJAZ1, CsJAZ2, CsJAZ3, CsJAZ4, CsJAZ7, and CsJAZ8) were upregulated by mechanical wounding and infestation with the tea geometrid (Ectropis obliqua), while infection with tea anthracnose (Colletotrichum camelliae) primarily upregulated the expression levels of CsJAZ1 and CsJAZ10. In addition, CsJAZs were observed to interact with CsMYC2 and AtMYC2. Therefore, the results of this study may contribute to the functional characterization of the CsTIFY genes, especially the members of the JAZ subfamily, as regulators of the JA-mediated defense response in tea plant.


Introduction
The TIFY family is a plant-specific gene family that was first discovered in Arabidopsis, and this family is named after the highly conserved TIFY domain, which was previously known as ZIM (zinc-finger protein expressed in inflorescence meristem) [1][2][3][4][5]. The TIFY domain contains approximately 36 amino acids, and the core motif TIF[F/Y]XG exhibits variant forms, with glycine being completely conserved [1]. According to the diverse domain structure, the TIFY family can be divided into four major subfamilies, including ZML (ZIM/ZIM-like), PPD (PEAPOD), JAZ (jasmonate-ZIM-domain) and TIFY [6]. In addition to the TIFY subfamily, which only contains the TIFY domain, the other three subfamilies also have specific and conserved domains aside from including the ZML, TIFY, and PPD subfamilies, is still lacking. In addition, how CsTIFY genes respond to biotic or abiotic stresses has not been fully elucidated. In the present study, we searched the genome database and identified 16 genes encoding TIFY proteins. Phylogenetics, chromosome location, exon/intron structure, and conserved protein domains were analyzed to characterize the members of the CsTIFY gene family. Moreover, the expression profiles of the CsTIFY genes in different organs and their responses to the exogenous application of JA, mechanical wounding, infestation by E. obliqua, and infection by C. camelliae were analyzed. Furthermore, the interactions between key CsJAZ proteins and CsMYC2 were investigated as well. Our results may facilitate the functional characterization of CsTIFY genes in tea plants.

Identification of the Members of the TIFY Gene Family in the Tea Plant
To identify all the putative TIFY domain-containing proteins in the tea plant, an hidden Markov model (HMM) profile of the TIFY domain (PF06200) was employed to perform a blast search against the chromosome-level genome of the tea plant (downloaded from TPIA). Through this research, 16 candidate loci encoding putative TIFY proteins were identified in the tea plant. Identification within the predicted protein sequences of the tea plant genome was conducted with the keyword "TIFY", and nine proteins were identified. These nine proteins were included in the proteins identified through the HMM file. The 16 proteins were analyzed for conserved domains in the Pfam database to reconfirm that they belong to the TIFY family. All 16 proteins were found to contain the conserved TIFY motif, and the characteristics of the TIFY motif of all CsTIFY proteins are presented in Table 1. In addition to the TIFY domain, three other domains, the CCT domain (PF06203), GATA domain (PF00320), and Jas domain (PF09425), were also identified in the CsTIFY family. All 16 TIFY family members in tea plants were named according to the existing numbering system described in Arabidopsis and the phylogenetic analysis with AtTIFY and AtJAZ proteins described below (Figure 1 and Supplementary Figure S1). Members of the JAZ and PPD subfamilies have a consensus sequence, "TIFYXG", in the TIFY motif, while most ZML proteins possess a conserved "TLSFXG" motif in the same domain. The coding sequences (CDSs) of the CsTIFY genes ranged from 411 bp (CsJAZ7) to 1197 bp (CsJAZ3 and CsJAZ4) with predicted proteins of 136 to 398 amino acids.  CsJAZ1  702  233  TIFYGG  CSS0031905  CsJAZ2  885  294  TIFYAG  CSS0022053  CsJAZ3  1197  398  TIFYAG  CSS0012514  CsJAZ4  1197  398  TIFYAG  CSS0028188  CsJAZ7  411  136  TIFYNG  CSS0034430  CsJAZ8  654  217  TIFYNG  CSS0011019  CsJAZ10  786  361  TIFYGG  CSS0031620  CsPPD  1008  335  TIFYCG  CSS0050449  CsZML7  912  303  TLSFQG  CSS0037100  CsZML8  912  303  TLSFQG  CSS0003047  CsZML1  846  281  TLSFRG  CSS0017505  CsZML2  972  323  TLSFQG  CSS0004261  CsZML3  978  325  TLSFQG  CSS0050427  CsZML4  1092  363  TLAFEG  CSS0045061  CsZML5  1089  362  TSELSI  CSS0016802 CsZML6 1131 376 TLAFEG

Phylogenetic Analysis of the TIFY Family
To identify groups of orthologous proteins and explore the evolutionary relationship of CsTIFY, full-length sequences of 16 CsTIFY proteins, along with 18 TIFY proteins from Arabidopsis, were used to build a phylogenetic tree with the maximum-likelihood (ML) method ( Figure 1). All 34 TIFYs were classified into two major groups (Group I and Group II). Among these proteins, proteins containing the GATA domain, including six CsZML proteins and three ZML members from Arabidopsis, were clustered in Group I. AtTIFY8 was clustered in Group I but not in the ZML branch, and no AtTIFY8 homologs were found among CsTIFY in this analysis. Group II contains proteins with the absence of the GATA domain except for CsZML7 and CsZML8, including all the members of the JAZ and PPD subfamilies from Arabidopsis and putative homologues in tea plant.
Gene duplication is an evolutionary event that contributes to the multimember gene family. Thus, to characterize the distribution of CsTIFY family genes, the chromosomal locations of all 16 CsTIFY genes were determined based on the pseudochromosome assembly of the tea plant genome. The 13 CsTIFY genes were mapped on eight chromosomes, including chromosomes 2, 3,6,8,9,12,13, and 15, with 3 CsTIFY genes being mapped to chromosome contigs ( Figure 2). CsTIFY genes were generally dispersed in the tea plant genomes, with no more than three TIFY genes being present on one chromosome. In addition, CsZML7, CsZML8, CsJAZ3, and CsJAZ4, which are homologous genes, are adjacent to each other in the same chromosome; CsZML2 and CsZML3 are also segmental duplication genes, although they were mapped to different contigs or chromosomes. containing the GATA domain, including six CsZML proteins and three ZML members from Arabidopsis, were clustered in Group I. AtTIFY8 was clustered in Group I but not in the ZML branch, and no AtTIFY8 homologs were found among CsTIFY in this analysis. Group II contains proteins with the absence of the GATA domain except for CsZML7 and CsZML8, including all the members of the JAZ and PPD subfamilies from Arabidopsis and putative homologues in tea plant. Gene duplication is an evolutionary event that contributes to the multimember gene family. Thus, to characterize the distribution of CsTIFY family genes, the chromosomal locations of all 16 CsTIFY genes were determined based on the pseudochromosome assembly of the tea plant genome. The 13 CsTIFY genes were mapped on eight chromosomes, including chromosomes 2, 3,6,8,9,12,13, and 15, with 3 CsTIFY genes being mapped to chromosome contigs ( Figure 2). CsTIFY genes were generally dispersed in the tea plant genomes, with no more than three TIFY genes being present on one chromosome. In addition, CsZML7, CsZML8, CsJAZ3, and CsJAZ4, which are homologous genes, are adjacent to each other in the same chromosome; CsZML2 and CsZML3 are also segmental duplication genes, although they were mapped to different contigs or chromosomes.

Sequence Analysis of the CsTIFY Family
To obtain more diversified information within the CsTIFY family, a phylogenetic tree, exonintron organization, and conserved domains were constructed within the CsTIFY gene family containing the GATA domain, including six CsZML proteins and three ZML members from Arabidopsis, were clustered in Group I. AtTIFY8 was clustered in Group I but not in the ZML branch, and no AtTIFY8 homologs were found among CsTIFY in this analysis. Group II contains proteins with the absence of the GATA domain except for CsZML7 and CsZML8, including all the members of the JAZ and PPD subfamilies from Arabidopsis and putative homologues in tea plant. Gene duplication is an evolutionary event that contributes to the multimember gene family. Thus, to characterize the distribution of CsTIFY family genes, the chromosomal locations of all 16 CsTIFY genes were determined based on the pseudochromosome assembly of the tea plant genome. The 13 CsTIFY genes were mapped on eight chromosomes, including chromosomes 2, 3,6,8,9,12,13, and 15, with 3 CsTIFY genes being mapped to chromosome contigs ( Figure 2). CsTIFY genes were generally dispersed in the tea plant genomes, with no more than three TIFY genes being present on one chromosome. In addition, CsZML7, CsZML8, CsJAZ3, and CsJAZ4, which are homologous genes, are adjacent to each other in the same chromosome; CsZML2 and CsZML3 are also segmental duplication genes, although they were mapped to different contigs or chromosomes.

Sequence Analysis of the CsTIFY Family
To obtain more diversified information within the CsTIFY family, a phylogenetic tree, exonintron organization, and conserved domains were constructed within the CsTIFY gene family

Sequence Analysis of the CsTIFY Family
To obtain more diversified information within the CsTIFY family, a phylogenetic tree, exon-intron organization, and conserved domains were constructed within the CsTIFY gene family (Figure 3). The results showed that 16 TIFY proteins were clearly divided into three subgroups: the JAZ subfamily, ZML subfamily, and PPD subfamily, with the exception of CsZML7 and CsZML8 ( Figure 3A). The exon-intron organization within the CsTIFY gene family indicated that all the gene members processed introns, and the number of introns was variable ( Figure 3B). Although the gene structures of the three subfamilies are clearly different from one another, most genes in the same subfamily tend to have the same member of introns and the same splicing positions, such as CsJAZ1, CsJAZ2 and CsZML1, CsZML2. Conserved domains in CsTIFY were analyzed in the Pfam web server ( Figure 3C). In addition to the TIFY domain, seven members of the CsJAZ subfamily also contain a Jas/CCT_2 domain (PF09425), CCT domain (PF06203), and GATA domain (PF00320) and were also found in eight CsZML subfamily proteins. A PPD domain and modified Jas domain were found in one CsPPD. The protein domain architecture was mostly in accord with the clustering mode of the phylogenetic analysis. CsZML7 and CsZML8 were not included in the ZML branch, but they have the same gene and domain structure as other ZML members.
intron organization, and conserved domains were constructed within the CsTIFY gene family ( Figure 3). The results showed that 16 TIFY proteins were clearly divided into three subgroups: the JAZ subfamily, ZML subfamily, and PPD subfamily, with the exception of CsZML7 and CsZML8 ( Figure 3A). The exon-intron organization within the CsTIFY gene family indicated that all the gene members processed introns, and the number of introns was variable ( Figure 3B). Although the gene structures of the three subfamilies are clearly different from one another, most genes in the same subfamily tend to have the same member of introns and the same splicing positions, such as CsJAZ1, CsJAZ2 and CsZML1, CsZML2. Conserved domains in CsTIFY were analyzed in the Pfam web server ( Figure 3C). In addition to the TIFY domain, seven members of the CsJAZ subfamily also contain a Jas/CCT_2 domain (PF09425), CCT domain (PF06203), and GATA domain (PF00320) and were also found in eight CsZML subfamily proteins. A PPD domain and modified Jas domain were found in one CsPPD. The protein domain architecture was mostly in accord with the clustering mode of the phylogenetic analysis. CsZML7 and CsZML8 were not included in the ZML branch, but they have the same gene and domain structure as other ZML members.
Furthermore, the conservation of functional motifs in the TIFY domain among CsTIFY proteins, as well as the Jas motif among CsJAZ proteins, was also performed by drawing their sequence logos. The results showed that the entire TIFY domain showed diverse forms, but most of them are conserved within the TIF[F/Y]XG region and amino acid sites Val-14 and Val-16, as presented in Figure 3D. Among these motifs, the TIFY motifs in the JAZ and PPD subfamilies exhibited conserved TIFYXG, while those of the ZML subfamily deviated from TIF[F/Y]XG, including TLSFXG (five proteins), TLAFEG (two proteins), and TSELSI. Jas motifs are highly conserved, particularly in location sites 6, 7, 9, 10, 14, 17, 23, and 24 ( Figure 3E). In addition, the amino acid PY located at the C-terminus of the Jas motif is entirely conserved, and it is a symbol used to distinguish the Jas motif in the JAZ subfamily from the divergent Jas motif in the PPD subfamily.

Expression Profile of CsTIFY Genes in Different Organs
The expression patterns of CsTIFY genes in four main organs (leaf, flower, stem, and root) were investigated using qRT-PCR ( Figure 4). Among the 16 TIFY genes, 15 were expressed in at least one organ, except CsZML6, whose amplification failed because of its low expression level in those Furthermore, the conservation of functional motifs in the TIFY domain among CsTIFY proteins, as well as the Jas motif among CsJAZ proteins, was also performed by drawing their sequence logos. The results showed that the entire TIFY domain showed diverse forms, but most of them are conserved within the TIF[F/Y]XG region and amino acid sites Val-14 and Val-16, as presented in Figure 3D. Among these motifs, the TIFY motifs in the JAZ and PPD subfamilies exhibited conserved TIFYXG, while those of the ZML subfamily deviated from TIF[F/Y]XG, including TLSFXG (five proteins), TLAFEG (two proteins), and TSELSI. Jas motifs are highly conserved, particularly in location sites 6,7,9,10,14,17,23, and 24 ( Figure 3E). In addition, the amino acid PY located at the C-terminus of the Jas motif is entirely conserved, and it is a symbol used to distinguish the Jas motif in the JAZ subfamily from the divergent Jas motif in the PPD subfamily.

Expression Profile of CsTIFY Genes in Different Organs
The expression patterns of CsTIFY genes in four main organs (leaf, flower, stem, and root) were investigated using qRT-PCR ( Figure 4). Among the 16 TIFY genes, 15 were expressed in at least one organ, except CsZML6, whose amplification failed because of its low expression level in those detected organs. CsJAZ4, CsZML4, CsZML5, and CsZML8 showed their greatest relative abundance in leaves; CsJAZ2, CsJAZ7, and CsZML3 were expressed at higher levels in flowers than in the other organs; CsJAZ1, CsJAZ3, CsJAZ8, and CsJAZ10 had a stronger expression level in roots than in the other organs. The expression levels of CsZML1, CsZML2, and CsZML7 in the four organs were similar and exhibited no significant difference. The transcription levels of the only member of the PPD subfamily CsPPD in stems and leaves were higher than those in flowers and roots. Overall, the relative expression levels of the 15 CsTIFY genes in four organs exhibited distinct expression patterns.

Expression Profile of CsTIFY Genes in Different Organs
The expression patterns of CsTIFY genes in four main organs (leaf, flower, stem, and root) were investigated using qRT-PCR ( Figure 4). Among the 16 TIFY genes, 15 were expressed in at least one organ, except CsZML6, whose amplification failed because of its low expression level in those detected organs. CsJAZ4, CsZML4, CsZML5, and CsZML8 showed their greatest relative abundance in leaves; CsJAZ2, CsJAZ7, and CsZML3 were expressed at higher levels in flowers than in the other organs; CsJAZ1, CsJAZ3, CsJAZ8, and CsJAZ10 had a stronger expression level in roots than in the other organs. The expression levels of CsZML1, CsZML2, and CsZML7 in the four organs were similar and exhibited no significant difference. The transcription levels of the only member of the PPD subfamily CsPPD in stems and leaves were higher than those in flowers and roots. Overall, the relative expression levels of the 15 CsTIFY genes in four organs exhibited distinct expression patterns.

Expression Profiles of CsTIFY Genes under JA and Mechanical Wounding Treatment
We first investigated the transcript levels of CsTIFY genes in leaves treated with an exogenous application of JA and mechanical wounding ( Figure 5, Figure 6). The transcription levels of most genes in the JAZ subfamily were significantly increased in response to JA and mechanical wounding. After JA treatment, CsJAZ1, CsJAZ2, CsJAZ7, and CsJAZ8 were induced more than tenfold at most time points from 1.5 to 24 h, and the fold change for CsJAZ3 and CsJAZ4 was approximately fivefold for at least one time point, while CsJAZ10 was slightly induced at 6 h (Figure

Expression Profiles of CsTIFY Genes under JA and Mechanical Wounding Treatment
We first investigated the transcript levels of CsTIFY genes in leaves treated with an exogenous application of JA and mechanical wounding (Figures 5 and 6). The transcription levels of most genes in the JAZ subfamily were significantly increased in response to JA and mechanical wounding. After JA treatment, CsJAZ1, CsJAZ2, CsJAZ7, and CsJAZ8 were induced more than tenfold at most time points from 1.5 to 24 h, and the fold change for CsJAZ3 and CsJAZ4 was approximately fivefold for at least one time point, while CsJAZ10 was slightly induced at 6 h ( Figure 5). The expression of CsJAZ1, CsJAZ2, CsJAZ7, and CsJAZ8 was also strongly upregulated within 1.5 h of mechanical wounding, with transcription levels gradually decreasing over time ( Figure 6). In contrast, the accumulation of CsJAZ3 and CsJAZ4 induced by mechanical wounding was delayed and peaked at 6 h, while mechanical wounding treatment did not induce the expression of CsJAZ10. Other CsTIFY genes in the ZML and PPD subfamilies showed no significant change in expression level after those two treatments. 5). The expression of CsJAZ1, CsJAZ2, CsJAZ7, and CsJAZ8 was also strongly upregulated within 1.5 h of mechanical wounding, with transcription levels gradually decreasing over time ( Figure 6). In contrast, the accumulation of CsJAZ3 and CsJAZ4 induced by mechanical wounding was delayed and peaked at 6 h, while mechanical wounding treatment did not induce the expression of CsJAZ10. Other CsTIFY genes in the ZML and PPD subfamilies showed no significant change in expression level after those two treatments.   Data are shown by the mean + SE, asterisks indicate a significant difference between treatment and control (Student's t-test, * p < 0.05, ** p < 0.01, *** p < 0.001, n = 6). 5). The expression of CsJAZ1, CsJAZ2, CsJAZ7, and CsJAZ8 was also strongly upregulated within 1.5 h of mechanical wounding, with transcription levels gradually decreasing over time ( Figure 6). In contrast, the accumulation of CsJAZ3 and CsJAZ4 induced by mechanical wounding was delayed and peaked at 6 h, while mechanical wounding treatment did not induce the expression of CsJAZ10. Other CsTIFY genes in the ZML and PPD subfamilies showed no significant change in expression level after those two treatments.   Figure 6. Expression profile of CsTIFY genes elicited by mechanical wounding. Data are shown by the mean + SE, asterisks indicate significant difference between treatment and control (Student's t-test, * p < 0.05, ** p < 0.01, *** p < 0.001, n = 6).

Expression Profiles of CsTIFY Genes in Response to Attack by the Tea Geometrid and Tea Anthracnose
Members of the JAZ family exhibited different expression patterns in response to different biotic stresses in many plant species, leading us to investigate how members of the CsJAZ subfamily respond to attacks by insect herbivores and pathogens. The results showed that six members of the CsJAZ subfamily were upregulated by damage by tea geometrid with different response patterns, as well as CsJAZ10 (Figure 7). Geometrid attack significantly upregulated the expression levels of CsJAZ1, CsJAZ7, and CsJAZ8 at least 15-fold at all tested time points, and CsJAZ2 was also induced at least eightfold at four time points. Transcription levels of CsJAZ3 increased at 3 h, 6 h, and 24 h, and CsJAZ4 only increased at 24 h. On the other hand, CsJAZ1 and CsJAZ10 were the two genes that were upregulated after 72 h of infection with tea anthracnose (Figure 8), suggesting that the induction of these two genes was primarily responsive to tea anthracnose. Other CsTIFY genes in the ZML and PPD subfamilies were not responsive to those two biotic stresses except for CsZML5, which was slightly downregulated, and CsZML7, which was slightly upregulated by geometrid attack. Members of the JAZ family exhibited different expression patterns in response to different biotic stresses in many plant species, leading us to investigate how members of the CsJAZ subfamily respond to attacks by insect herbivores and pathogens. The results showed that six members of the CsJAZ subfamily were upregulated by damage by tea geometrid with different response patterns, as well as CsJAZ10 (Figure 7). Geometrid attack significantly upregulated the expression levels of CsJAZ1, CsJAZ7, and CsJAZ8 at least 15-fold at all tested time points, and CsJAZ2 was also induced at least eightfold at four time points. Transcription levels of CsJAZ3 increased at 3 h, 6 h, and 24 h, and CsJAZ4 only increased at 24 h. On the other hand, CsJAZ1 and CsJAZ10 were the two genes that were upregulated after 72 h of infection with tea anthracnose (Figure 8), suggesting that the induction of these two genes was primarily responsive to tea anthracnose. Other CsTIFY genes in the ZML and PPD subfamilies were not responsive to those two biotic stresses except for CsZML5, which was slightly downregulated, and CsZML7, which was slightly upregulated by geometrid attack.

Interactions of CsJAZ with CsMYC2
Interaction with MYC2 is a general feature of JAZ repressors in other plants. We selected CsJAZ2 and CsJAZ3, which were highly expressed and exhibited different responses to JA, to assess

Interactions of CsJAZ with CsMYC2
Interaction with MYC2 is a general feature of JAZ repressors in other plants. We selected CsJAZ2 and CsJAZ3, which were highly expressed and exhibited different responses to JA, to assess their capacity of interacting with CsMYC2 and AtMYC2 using yeast two-hybrid experiments. As the results of this study showed that CsJAZ2 and CsJAZ3 interacted with both CsMYC2 and AtMYC2, indicating that CsJAZ may also function in the tea plant by interacting with MYC2 ( Figure 9).

Interactions of CsJAZ with CsMYC2
Interaction with MYC2 is a general feature of JAZ repressors in other plants. We selected CsJAZ2 and CsJAZ3, which were highly expressed and exhibited different responses to JA, to assess their capacity of interacting with CsMYC2 and AtMYC2 using yeast two-hybrid experiments. As the results of this study showed that CsJAZ2 and CsJAZ3 interacted with both CsMYC2 and AtMYC2, indicating that CsJAZ may also function in the tea plant by interacting with MYC2 ( Figure 9).

Discussion
The TIFY family is an important plant-specific gene family and has been identified in multiple species among both monocots and eudicots [1, 8,10,12]. Continuing research has demonstrated that the members of the TIFY gene family, especially members of the JAZ subfamily, play multiple roles in different plant developmental processes and the process of defense against biotic and abiotic stresses [1, 8,10,11,13,31]. As evergreen plants, tea plants are frequently challenged by abiotic and biotic stresses, but information on the TIFY gene family is lacking, which hampers the research investigating resistance mechanisms [27,28,31]. The chromosome-level genomic sequence of the tea plant has been recently released, and the extensive annotated information provided in the tea genome provides fundamental material for comprehensive analysis of the CsTIFY family [32]. All the above-mentioned research enables us to study the evolutionary history and expression diversity of the TIFY family in tea plants.
In the present study, 16 CsTIFY proteins were identified from the chromosome-level tea genome sequence through conserved sequence analysis of the TIFY motif (Table 1). According to the reported plants, members of the TIFY family vary from one in Marchantia polymorpha to 48 in the higher plant Panicum virgatum [6,8]. TIFY in the tea plant ranks next to last with 16 members, further indicating functional diversity and redundancy in this rapidly evolving gene family. The TIFY family in tea plant was also divided into two groups, and eight CsZML proteins in Group I were believed to act as transcription factors [2]. Proteins in Group II contain JAZ repressors and PPD proteins that lack the GATA domain. This result was similar to the findings of the phylogenetic analysis of TIFYs from Arabidopsis and rice, further suggesting functional conservation in different plant taxa, and both groups appeared before divergence in monocots and dicots [1]. However, only three subfamilies of TIFY were found in the tea plant, including seven belonging to the JAZ subfamily, eight belonging to the ZML subfamily, and one belonging to the PPD subfamily without the TIFY subfamily. It was confirmed that not all subfamilies were found in plant species, and the TIFY subfamily is also absent from B. distachyon and Sorghum bicolor [6,13], while the PPD subfamily only exists in the dicots and in lycophyte Selaginella moellendorffii [6]. In addition, orthologues of AtTIFY8, which was the first member of the TIFY subfamily, are also mostly present as unique genes within the species studied [17]. One reason for the absence of the TIFY subfamily in the tea plant is that instead of TIFY subfamily genes, other genes have evolved to function as AtTIFY8 [10]. The other reason for this exception may be the incomplete genomic sequencing of tea plants, such as that found in indica rice [10].
Analysis of chromosome position, intron/exon structures, and conserved domains was also employed to obtain evolutionary clues within the CsTIFY family. Segmental duplications were found in the CsTIFY family, mainly in the CsZML and CsJAZ subfamilies. Our results agree with the duplication phenomenon identified within the TIFY family, supporting the notion that duplication events play an important role in the evolution and expansion of the TIFY family in plants [6][7][8]33]. CsTIFY members in the same subfamily generally possessed a similar gene structure and shared the same distribution of conserved domain patterns, indicating the strong correlation between phylogeny and intron/exon diversification and conserved domain patterns within CsTIFY. The correlation is consistent with findings in Arabidopsis and other plants, which played an important role in the evolution of multiple gene families [7,33]. Among these genes, the three sets of homologous genes (CsZML2 and CsZML3, CsZML7 and CsZML8, CsJAZ3 and CsJAZ4) contained the exact same intron/exon structures and conserved domains but with different genomic lengths, respectively, further indicating that these TIFY genes may be the products of duplication events. Furthermore, sequence logos of the TIFY domain showed that this central typical motif could be detected with slight divergence in all CsTIFY proteins except for CsZML5, sufficiently suggesting broad conservation of the main motifs characteristic of the plant TIFY family during evolution. However, the JAZ subfamily contains the same core motif as "TIFYXG", and CsPPD also contains the "TIFYXG" motif, while it is "TLXFXG" in the ZML subfamily. The CsPPD subfamily, which has the same TIFY motif, was also clustered with CsJAZs together in both polygenetic trees, confirming that PPD proteins may be the evolutionary precursors of JAZ and may have evolved as two independent subfamilies by gene duplication in vascular plants [1,6]. A recent study reported that 12 CsJAZ proteins were found through TIFY and Jas domain blast in tea plants [31]. Among the 12 CsJAZ proteins, six members were identical to those in our study, and the other six CsJAZ proteins in the above study both contain the so-called Jas domain, which was only part of the CCT domain but was not a true Jas domain. In addition, the GATA domain was also found in their protein sequence; therefore, they should be named ZML proteins, rather than JAZ proteins [1, 6,8]. In addition, the Jas motif defines JAZ proteins and mediates complex formation through direct interactions between JAZs and TFs (such as MYC2) [34][35][36]. This motif is also conserved among CsJAZ proteins, and interactions between CsJAZ2, CsJAZ3, and CsMYC2 were also suggested that CsJAZ proteins may act as functional repressors of CsMYC2 in tea plants.
All seven CsJAZ genes found in our study were upregulated by JA. The phylogenetic tree showed that the stronger inducible genes CsJAZ1, CsJAZ2, CsJAZ7, and CsJAZ8 were grouped closer than CsJAZ3 and CsJAZ4, which may suggest that JA-induced CsTIFY genes evolved from common ancestors responsive to JA [7,37]. The identical CsJAZ genes found in a previous study showed that CsJAZ2, CsJAZ3, and CsJAZ4 were remarkably upregulated by MeJA [31]. This difference may be attributable to such factors as different tea varieties, discrepancies in the elicitation effects of JA and MeJA, and differing treatment concentrations. However, the expression level of gene members from CsZML and CsPPD was not modulated by JA in the current study. Due to the key role of JA and JAZ in plant responses to biotic and abiotic stresses, our study mainly focused on the influence of biotic and abiotic stresses on the CsJAZ subfamily. Except for CsJAZ10, six other CsJAZs were significantly upregulated after mechanical wounding, and the transcripts of most genes accumulated to higher levels within 1.5 h. Mechanical damage results in rapid accumulation of JA within 30 min and thus reshapes the expression profile of genes related to the JA signaling pathway [37]. The expression pattern of CsJAZ genes in response to E. obliqua infestation is similar to those elicited by mechanical wounding, for the CsJAZ members induced by mechanical wounding were also induced by E. obliqua feeding. However, there was also a notable difference in the induction trend for those genes. The elicitation of all CsJAZ by tea geometrid feeding showed a longer duration, while an induction by mechanical wounding gradually declined at later time points. This finding may be observed because mechanical wounding was not a continuous damage as feeding by tea geometrid, or herbivore-associated molecular patterns from E. obliqua may elongate the elicitation effect, which needs to be clarified in future studies. Previous findings also established the role of JAZ proteins in the regulation of plant anti-insect defense [25,36,37]. Several JAZ genes were proven to be responsive to mechanical wounding and the damage of E. obliqua in a previous study based on transcriptomic analysis [30]. Feeding by generalist Spodoptera exigua increased the transcript levels of most AtJAZ, while transgenic line AtJAZ1∆3A plants that overexpressed a Jas-motif-deleted form (AtJAZ1∆3A) perturbed JA signaling and decreased plant resistance to insect feeding [37]. The herbivory-elicited gene NaJAZh was proven to regulate part of JA-dependent direct and indirect defenses in tobacco, such as the accumulation of nicotine and the emission of volatile organic compounds [25]. The key roles played by JA as signals mediating pathogen defense responses have been widely documented [38]. Changes in the transcription levels of seven CsJAZ genes after fungal pathogen infection exhibited different expression patterns. In our assay, CsJAZ1 and CsJAZ10 were primarily induced by inoculation with the pathogen C. camelliae, providing evidence to support the hypothesis that at least two members of the CsJAZ subfamily are involved in the plant pathogen resistance of tea plants. Eight of the 12 JAZ genes in Arabidopsis were induced after infection with the bacterial pathogen P. syringae strain DC3000, and AtJAZ10 was proven to be involved in the branch of JA-mediated DC3000 symptom development [7,39]. Moreover, CsJAZ2 and CsJAZ3 were also regulated by other hormone treatments, including ETH (ethylene), GA (gibberellin), and ABA (abscisic acid), and abiotic stress drought (PEG) [31]. Thus, although both responses are mediated by JA signaling, there are also unique regulatory events governing the expression of specific JAZ genes in response to different stimuli. Furthermore, the interaction between CsJAZ and CsMYC2 may suggest that CsJAZ participates in signal transduction and gene regulation in the defense response of tea plants by interacting with CsMYC2 to regulate the JA signaling pathway. Taken together, these results indicate that different CsJAZ genes were differentially responsive to various stresses, which is another possible way to provide the mechanism specificity underlying JA responses.
In conclusion, a total of 16 TIFY genes were identified through genome-wide analysis, and most of these genes may share common ancestors with those in other reported species, such as A. thaliana. Expression profile analyses showed that CsJAZ genes were induced by JA, mechanical wounding treatment, and biotic stresses, such as herbivorous attack and pathogen infection. Among these genes, CsJAZ1, CsJAZ2, CsJAZ3, CsJAZ7, and CsJAZ8 were responsive to tea geometrid infestation, while tea anthracnose infection mainly regulated the expression of CsJAZ1 and CsJAZ10. In addition, the interaction complex of CsJAZ-CsMYC2 also indicated their regulatory role in the JA-mediated defense response. Our results would be helpful in laying the foundation for the further functional characterization of CsTIFY genes. Further investigations are also warranted to determine the function of these JAZ genes involved in the tea plant defense response.

Insect and Fungi Pathogen
The E. obliqua larvae and C. camelliae were originally obtained from the tea plantation of Tea Research Institute of the Chinese Academy of Agricultural Sciences (TRICAAS), Hangzhou, China. E. obliqua larvae was reared with fresh tea shoots of Longjing 43. C. camelliae was cultured on potato dextrose agar (PDA). The larvae and pathogen were kept in an incubator at 26 ± 2 • C, 12/12 h photo phase, and after one generation, they were used for the experiments.

Plant Materials and Treatments
Two-year-old Longjing 43 tea plants were planted individually in plastic pots (14 cm diameter × 15 cm high) and grown in a greenhouse (26 ± 2 • C, 12/12 h photo phase, 65 ± 5% room humidity). They were irrigated (once every other day) and fertilized (once a month) with the same standards. Plants without disease and insects were used for the experiments.

Different Organs
Various organ samples, including flowers, leaves, stems, and roots, were harvested and frozen in liquid nitrogen; then, they were stored at −80 • C. Six replications were carried out.

JA Treatment
JA (Sigma Chemical Co., St. Louis, MO, USA) was dissolved in a small amount of ethanol and then made up to 150 µg/mL in 50 mM sodium phosphate buffer (titrated with 1 M citric acid until pH 8). Plants were individually sprayed with 8 mL of JA solution. Control plants were sprayed with buffer with a small amount of ethanol. Samples were harvested at 1.5, 3, 6, and 24 h after the start of treatment, separately. Six replications were carried out.

Mechanical Wounding Treatment
The mechanical wounding damage was made through a fabric pattern wheel for rolling with 4 times on each leaf. Six replications were carried out.

E. obliqua Treatment
The second leaves of tea plants were covered with a fine-mesh sleeve and introduced with 10 2nd-instar caterpillars that had been starved for 8 h. Leaves covered with a fine-mesh sleeve but without insects were set as control. Control and insect-treated leaves were harvested at 1.5, 3, 6, and 24 h after the start of treatment, separately. Six replications were carried out.

C. camelliae Infection
C. camelliae has been cultured on potato dextrose agar for 7 days. The plant leaves were damaged with a sterile needle 5 times, and fungus cake (ID = 6 mm) was applied to the damaged part immediately (n = 12). After inoculation, the potted tea plants were maintained in the greenhouse, in which conditions were 26 ± 2 • C, 90 ± 5% RH (relative humidity), and 12 h photoperiod. Then, 24 and 72 h later, the inoculated leaves were harvested and frozen in liquid nitrogen, after which they were stored at −80 • C. Six replications were carried out.

Identification of TIFY Family in Tea
Genome-wide screen of putative TIFY proteins in tea plant was performed within the Tea Plant Genome Database (http://tpia.teaplant.org/download.html) with a hidden Markov model (HMM) profile of the TIFY domain (PF06200) as a query and set as E-value < e −6 . TIFY was also used as a keyword for searching protein sequences of the chromosome-level genome of tea plant.

Bioinformatics Analysis of TIFY Family in Tea
Sequence information of putative TIFYs in tea plants was also extracted from the Tea Plant Genome Database. The Pfam database (http://pfam.xfam.org/) was used for analyzing the protein domain of the TIFY proteins family. The CsTIFY genes were positioned on the tea chromosomes based on the genomic sequences of these genes in the Pseudochromosome assembly of the tea plant genome available at http://tpia.teaplant.org/download.html. Mapchart software was used to draw the Chromosome position (https://www.wur.nl/en/show/Mapchart.htm). An exon/intron structure schematic diagram was performed with Gene Structure Display Server (http://gsds.cbi.pku.edu.cn/).