Revealing the Roles of the JAZ Family in Defense Signaling and the Agarwood Formation Process in Aquilaria sinensis

Jasmonate ZIM-domain family proteins (JAZs) are repressors in the signaling cascades triggered by jasmonates (JAs). It has been proposed that JAs play essential roles in the sesquiterpene induction and agarwood formation processes in Aquilaria sinensis. However, the specific roles of JAZs in A. sinensis remain elusive. This study employed various methods, including phylogenetic analysis, real-time quantitative PCR, transcriptomic sequencing, yeast two-hybrid assay, and pull-down assay, to characterize A. sinensis JAZ family members and explore their correlations with WRKY transcription factors. The bioinformatic analysis revealed twelve putative AsJAZ proteins in five groups and sixty-four putative AsWRKY transcription factors in three groups. The AsJAZ and AsWRKY genes exhibited various tissue-specific or hormone-induced expression patterns. Some AsJAZ and AsWRKY genes were highly expressed in agarwood or significantly induced by methyl jasmonate in suspension cells. Potential relationships were proposed between AsJAZ4 and several AsWRKY transcription factors. The interaction between AsJAZ4 and AsWRKY75n was confirmed by yeast two-hybrid and pull-down assays. This study characterized the JAZ family members in A. sinensis and proposed a model of the function of the AsJAZ4/WRKY75n complex. This will advance our understanding of the roles of the AsJAZ proteins and their regulatory pathways.

Various signaling models involving JAZ have been proposed in Arabidopsis. Specifically, it is hypothesized that, in the absence of the active forms of JAs, JAZs can bind to transcription factors and recruit co-repressors to repress gene transcription. In response to pathogen attack or wounds, newly synthesized JAs will facilitate the interaction between JAZ and the SCF complex, leading to the degradation of JAZ and alleviating the repression interaction between AsJAZ1 and AsMYC2 was confirmed by a pull-down assay [48,49]. Together, these studies indicate that the AsJAZ1 and AsMYC2 transcription factors can participate in JA-mediated signaling pathways to regulate the production of sesquiterpenes in A. sinensis. However, to date no study has systematically analyzed the AsJAZ family members and the regulatory roles of AsJAZ proteins in relation to AsWRKY transcription factors. Moreover, while a genome-wide analysis of the A. sinensis WRKY transcription factor family has been published, it lacks publicly available genome annotation [50,51]. In the current study, based on the recently released high-quality A. sinensis genome, 12 putative AsJAZ genes in the AsTIFY family and 64 putative AsWRKY genes were systematically revealed and characterized. Through transcriptomic and quantitative real-time PCR (qPCR) analysis, the tissue-specific or hormone-induced expression patterns of the AsJAZ genes and AsWRKY genes were determined. Furthermore, the interaction between AsJAZ4 and AsWRKY75n was validated by in vivo and in vitro experiments. The findings suggest that, in addition to the AsJAZ1/AsMYC2 module, the AsJAZ4/AsWRKY75n module is another module that exists in the JA pathways in A. sinensis. The findings of this study contribute to our understanding of the defense signaling pathways of A. sinensis and the agarwood formation process.

Phylogenetic Relationships and Major Domains of A. sinensis TIFY Family Members
The plant-specific TIFY family members, including the JAZ family proteins, play essential roles in plant growth and anti-stress processes. The TIFY family proteins are characterized by the TIFY domain (previously known as the ZIM domain), which contains a highly conserved amino acid pattern (TIF[F/Y]XG; pfam06200) [2]. Here, the hidden Markov model (HMM) profile of the TIFY domain (PF06200) was extracted and the HM-MER search program was used to screen the A. sinensis TIFY family proteins from all the predicated A. sinensis proteins in the GigaDB database. The resulting candidate sequences were submitted for further domain analysis. Finally, 19 A. sinensis TIFY family members were retrieved. Of these, the mRNA sequence of AsJAZ4 was validated by cDNA sequencing. The mRNA and genomic sequences of AsJAZ4 were submitted to Genbank, with accession numbers OK507019 and OK507018, respectively. The coding sequences of the 12 AsJAZ genes are presented in Table S1. Sequence alignment revealed a high similarity between the coding sequences of AsJAZ4 (375 bp) and AsJAZ5 (408 bp). The dissimilarity between the coding sequences of AsJAZ4 and AsJAZ5 lay in the A/G variant at site 338, and a single-base insertion at site 341 in AsJAZ4. This insertion changed the reading frame of AsJAZ4, resulting in a shorter coding sequence of AsJAZ4 compared with that of AsJAZ5. According to the phylogenetic relationships with Arabidopsis TIFY members, the 19 A. sinensis TIFY members were divided into four subfamilies (TIFY, JAZ, PPD, and ZML) ( Figure 1). In these subfamilies, the ZML subfamily included two ZIM proteins (AsZIM1 and AsZIM2) and two ZIM-like proteins (AsZML1 and AsZML2). The 12 JAZ subfamily members from A. sinensis were further divided into five groups: JAZ6, -7, -8, and -9 were in group JAZ-I; AsJAZ12 was in group JAZ-II; AsJAZ10 was in group JAZ-III; AsJAZ4, -5, and -11 were in group JAZ-IV; and AsJAZ1, -2, and -3 were in group JAZ-V. The major domain or motif architecture of the A. sinensis TIFY family members was retrieved by searching against the SMART database; manual modification was performed based on related references [3,52,53] (Figure 2). Among the 19 A. sinensis TIFY family members, the JAZ subfamily had 12 members, each containing a TIFY domain and a Jas motif; the PPD subfamily had a single member, which contained a PPD domain, a TIFY domain, and a Jas motif; the ZML subfamily had four members, each containing a TIFY domain, a Jas motif, and a GATA zinc-finger domain; and the two TIFY subfamily members contained only a TIFY domain. A further search was conducted for specific motifs of the AsJAZ proteins. In Arabidopsis, AtJAZ8 lacked a canonical degron that included a short conserved 'LPIAR' motif; however, it had an LxLxL-type EAR motif that mediated the transcriptional repression of JAs [54,55]. In the present study, like AtJAZ8, each of the AsJAZ4 and AsJAZ5 proteins had a canonical LxLxL-type EAR motif (LELRL) and lacked the canonical degron. In Arabidopsis, there exists a cryptic MYC2-interacting domain (CMID) in AtJAZ1 or AtJAZ10, and this is related to their interactions with AtMYC transcription factors [52,56,57]. Through sequence comparison, AsJAZ10 was found to contain an N-terminal "SQKFLDRRR" motif that was highly similar to the "CMID" core sequence "FQKFLDRRR" in AtJAZ10, which suggests the possible presence of interactions between the AsJAZ10 and MYC proteins. The sequence characteristics of all of the TIFY family genes are summarized in Table 1. It can be seen that, with the exclusion of AsJAZ12, all the remaining putative AsJAZ proteins were small alkaline proteins consisting of no more than 400 amino acids. AsJAZ4 consisted of only 124 amino acids, making it the shortest protein among the putative AsJAZ proteins. The AsJAZ genes were located in different chromosomes; their chromosome locations are presented in Figure S1. Based on the sequence locations and positions, AsJAZ7 and AsJAZ8 differed only in one base pair and both genes were on chromosome 4; this implies that the two genes are probably paralogous genes that originated from duplication. members contained only a TIFY domain. A further search was conducted for specific motifs of the AsJAZ proteins. In Arabidopsis, AtJAZ8 lacked a canonical degron that included a short conserved 'LPIAR' motif; however, it had an LxLxL-type EAR motif that mediated the transcriptional repression of JAs [54,55]. In the present study, like AtJAZ8, each of the AsJAZ4 and AsJAZ5 proteins had a canonical LxLxL-type EAR motif (LELRL) and lacked the canonical degron. In Arabidopsis, there exists a cryptic MYC2-interacting domain (CMID) in AtJAZ1 or AtJAZ10, and this is related to their interactions with AtMYC transcription factors [52,56,57]. Through sequence comparison, AsJAZ10 was found to contain an N-terminal "SQKFLDRRR" motif that was highly similar to the "CMID" core sequence "FQKFLDRRR" in AtJAZ10, which suggests the possible presence of interactions between the AsJAZ10 and MYC proteins. The sequence characteristics of all of the TIFY family genes are summarized in Table 1. It can be seen that, with the exclusion of AsJAZ12, all the remaining putative AsJAZ proteins were small alkaline proteins consisting of no more than 400 amino acids. AsJAZ4 consisted of only 124 amino acids, making it the shortest protein among the putative AsJAZ proteins. The AsJAZ genes were located in different chromosomes; their chromosome locations are presented in Figure S1. Based on the sequence locations and positions, AsJAZ7 and AsJAZ8 differed only in one base pair and both genes were on chromosome 4; this implies that the two genes are probably paralogous genes that originated from duplication.

Figure 2.
Distribution of the major domains and motifs for the A. sinensis TIFY family proteins. T AsTIFY proteins are arranged in order of their phylogenetic groups. The rectangles at the botto represent the complete AsTIFY protein sequences. The domains or motifs were predicated usi bio-software SMART (http://smart.embl-heidelberg.de/, accessed on 21 August 2021) or by manu sequence checking; they are presented as small boxes on the rectangles. The proteins, domains, a motifs are differentiated by size and color. In these domains or motifs, AsJAZ4 and AsJAZ5 conta an N-terminal "LELRL" motif (EAR motif). AsJAZ10 contains an N-terminal "XQKFLDRRR" mo (CMID motif) similar to a core motif of the CMID domain of AtJAZ10.  AsTIFY proteins are arranged in order of their phylogenetic groups. The rectangles at the bottom represent the complete AsTIFY protein sequences. The domains or motifs were predicated using bio-software SMART (http://smart.embl-heidelberg.de/, accessed on 21 August 2021) or by manual sequence checking; they are presented as small boxes on the rectangles. The proteins, domains, and motifs are differentiated by size and color. In these domains or motifs, AsJAZ4 and AsJAZ5 contain an N-terminal "LELRL" motif (EAR motif). AsJAZ10 contains an N-terminal "XQKFLDRRR" motif (CMID motif) similar to a core motif of the CMID domain of AtJAZ10.

Locations and Structures of the AsTIFY Genes and Promoter Elements of the AsJAZ Genes
The structural characteristics of the TIFY family genes were analyzed, particularly in relation to the features of the AsJAZ genes ( Figure 3a). In terms of the number of exons and their arrangement, AsJAZ6, -7, -8, and -9 were classified as one group (each having five exons), AsJAZ11 and AsJAZ12 were classified as one group (each having five exons and a conserved intron in the TIFY domain), AsJAZ4 and AsJAZ5 were classified as one group (each having three exons), AsJAZ10 was classified as one group (with four exons), and AsJAZ1, -2, and -3 were classified as one group (each having seven exons). The intron phase is defined as the position of the intron within or between codons (0, 1, and 2). The AsJAZ genes had two intron phase patterns: long (102012) for three genes (AsJAZ1, -2, and -3), and short for the others. The intron position was highly conserved in the sequences encoding the core TIFY domains and the Jas motifs. As shown by the sequence alignment results, a conserved phase 0 intron was inserted in eight of the twelve AsJAZ genes after the position encoding the 20th amino acid of the TIFY motifs ( Figure 3b). In 10 of the 12 JAZ family members, the 27-amino-acid Jas motif was split into 20 N-terminal and 7 C-terminal (X5PY) amino acid segments by an intron (the Jas intron) located in phase 2 of the codon specifying the Arg residue (R20) at position 20 ( Figure 3c). These characteristics are analogous to those reported for Arabidopsis JAZ genes [58]. Furthermore, the TIFY motifs were conserved for all 12 AsJAZ proteins, but the 12 Jas motifs were separated into two groups. Nine canonical AsJAZs in several plant species contained the consensus Jas motif "SLX2FX2KRX2RX5PY" [59]. However, AsJAZ4, AsJAZ5, and AsJAZ11 contained divergent Jas motifs like those of AtJAZ7 and AtJAZ8 in the TIFY5 clade [54]. The cis-acting element in the promoter region potentially suggests a relationship with growth and defensive responses. The 2 kb upstream sequences of the AsJAZ translation initiation sites of the AsJAZ genes were searched against the PlantCare database, and the cis-acting elements located in the promoters were determined ( Figure S2). For the hormoneresponsive elements, the promoter regions of the AsJAZ genes, except for AsJAZ3, -4, -5, and -11, contained the core TGACG (CGTCA) motif, which was homologous to the motifs involved in MeJA responsiveness in Hordeum vulgare. The conserved ABA-responsive elements (ABRE) were distributed in all gene promoters. The conserved gibberellin-responsive elements (P-box, TATC box, and GARE motif), auxin-responsive element (TGA element), and salicylic acid-responsive element (TCA element) were also distributed in the upstream promoters of the AsJAZ genes. An investigation of the transcription factor binding sites showed that the conserved MYC and MYB transcription binding sites were densely distributed in the promoter regions, while the conserved WRKY transcription factor binding sites (W box) were only distributed in the promoter regions of AsJAZ2, -3, -4, -5, -6, and -10. The findings also revealed several conserved tissue-specific elements, such as motifs or elements, probably related to endosperm expression or meristem expression.  The cis-acting element in the promoter region potentially suggests a relationship with growth and defensive responses. The 2 kb upstream sequences of the AsJAZ translation initiation sites of the AsJAZ genes were searched against the PlantCare database, and the cis-acting elements located in the promoters were determined ( Figure S2). For the hormone-responsive elements, the promoter regions of the AsJAZ genes, except for AsJAZ3, -4, -5, and -11, contained the core TGACG (CGTCA) motif, which was homologous to the motifs involved in MeJA responsiveness in Hordeum vulgare. The conserved ABA-responsive elements (ABRE) were distributed in all gene promoters. The conserved gibberellin-responsive elements (P-box, TATC box, and GARE motif), auxinresponsive element (TGA element), and salicylic acid-responsive element (TCA element) were also distributed in the upstream promoters of the AsJAZ genes. An investigation of The arrow indicates the location of the intron at the Jas motif. The intron was positioned invariably in phase 2 of the codon, specifying the Arg (R) at position 20 of the motif in the sequences of the remaining 10 AsJAZs, except for AsJAZ4 and AsJAZ5. The amino acid alignment of the Jas motifs of the A. sinensis JAZ members revealed that AsJAZ4, AsJAZ5, and AsJAZ11 had divergent Jas motifs.

Expression Patterns of the AsJAZ Genes
The raw transcriptome data of a five-year-old A. sinensis tree were downloaded from the NCBI SRA database with the accession number SRP068230 (released in 2016). Since the genome data of the A. sinensis tree had not been published at the time, Ye et al. used the de novo transcriptome assembly method to analyze the data [50]. Here, the A. sinensis tree genome released by Ding et al. was used as a reference [39], and the transcriptome data of the five-year-old A. sinensis tree were re-analyzed. The FPRM values of all of the genes were re-calculated. The tissue-specific expression levels of the AsJAZ genes were estimated through the comparison of their FPKM values, and are shown in a heatmap ( Figure 4a). The results revealed that, following the formic acid injection of the A. sinensis tree for 12 months, AsJAZ1, AsJAZ2, and AsJAZ12 exhibited relatively higher expression levels in the white wood (B1), the brown agarwood (J3), and the transition part of the trunk (W2). The expression of AsJAZ12 was highest in the brown agarwood, and its expression level was significantly higher than the others. In contrast, AsJAZ3, AsJAZ7, AsJAZ8, AsJAZ9, and AsJAZ10 had relatively lower expression levels in the B1, J3, and W2 tissues. However, the five genes displayed relatively higher expression levels in the brown agarwood compared to the other tissues. Among the genes, AsJAZ6 exhibited a very low expression level in the white wood, and the expression levels of AsJAZ4, AsJAZ5, and AsJAZ11 were too low to be detected in the various tissues. The expression patterns of the AsJAZ genes in the suspension cells upon various treatments for long (24 h) and short (1 or 2 h) durations were also investigated through RNA sequencing (RNA-seq) or quantitative real-time PCR (qPCR) (Figures 4b and 5). The results showed that the expression levels of AsJAZ3, AsJAZ6, and AsJAZ11 remained very low in suspension cells. This suggests that the three genes may be redundant or pseudogenes. Most of the other genes exhibited different expression patterns in suspension cells compared to that in the A. sinensis tree. For example, although AsJAZ12 exhibited the highest expression in brown agarwood, it did not always exhibit a higher expression level than the others in suspension cells (Figures 4b and 5). Therefore, it appears that AsJAZ12 may play a more significant role in the A. sinensis tree than in suspension cells. RNA-seq analysis revealed that, among all AsJAZ genes, the combined expression of AsJAZ4 and AsJAZ5 (AsJAZ4/5) upon MeJA treatment for 24 h exhibited the highest expression level in suspension cells (Figure 4b). AsJAZ4/5 also showed a very high expression level upon ABA treatment for 24 h. AsJAZ10 is another member that was actively expressed in the suspension cells, particularly after the MeJA and ABA treatments for 24 h (Figure 4b). A previous study reported that many AtJAZ genes were rapidly induced by JA application for 2 h [11]. Thus, we decided to investigate the expression changes in AsJAZ genes after short-duration (1 or 2 h) treatments. The qPCR analysis revealed that the expression levels of AsJAZ4/5, AsJAZ7/8, AsJAZ9, and AsJAZ10 were in a higher range than those of the other AsJAZ genes after the various hormone treatments ( Figure 5). Under the MeJA treatment for a short duration (1 or 2 h), the expression of AsJAZ4/5 increased to the highest among the 12 AsJAZ genes. After 2 h of MeSA or ABA treatment, AsJAZ10 had the highest expression level. After 1 or 2 h of ETH treatment, AsJAZ7/8 had the highest expression level. To analyze the fold changes in the expression levels of each individual gene under the various hormone treatments, the expression data presented in Figure 5 were re-used to construct Figure S3 and the number of genes with significant changes in expression was analyzed (Table S3). As shown in Figure S3, in suspension cells upon several specific short-duration hormone treatments, some of the AsJAZ genes were down-regulated. For example, after MeJA or ETH treatment, the expression of AsJAZ12 was slightly up-regulated, but after MeSA or ABA treatment the expression was down-regulated. Further, AsJAZ1 and AsJAZ2 were down-regulated under short-duration MeJA, MeSA, and ABA treatments. ETH treatment, the expression of AsJAZ12 was slightly up-regulated, but after Me ABA treatment the expression was down-regulated. Further, AsJAZ1 and AsJAZ2 down-regulated under short-duration MeJA, MeSA, and ABA treatments.   [50]. The B1, J3, W2, and F5 samples represent different tissues obtained from the trunk of a five-year-old A. sinensis. Agarwood was induced after formic acid injection for 12 months. The bark of the A. sinensis tree was removed. The white wood in the trunk was the B1 sample. The brown agarwood was defined as the J3 sample. The transition part between the B1 and J3 samples, with a light-brown color, was defined as the W2 sample. The rotten wood in the inner area of the trunk was defined as the F5 sample.  Similarly, because the sequences of AsJAZ4 and AsJAZ5 were highly similar, the total expression levels of AsJAZ4 and AsJAZ5 are represented by AsJAZ4/5. Normal conditions are represented by "0 h", that is, before treatment, and "1 h" and "2 h" represent 1 h or 2 h after the specific treatment. (a-d) Indicate the fold changes in the expression levels of the 12 AsJAZ genes under the MeJA, MeSA, ABA, or ETH treatment, respectively. The data represent the mean + the standard error of three independent biological replicates. Under each treatment, the relative expression level of AsJAZ7/8 at 0 h was taken as 1. The AsJAZ genes were separated into two groups based on their relative expression level ranges. The significance of the expression difference between the control (0 h) and treatments (1 h or 2 h) for each gene was evaluated using two-tailed Student's t-test, with "*" representing a significant difference at the p = 0.05 level and "**" representing a highly significant difference at the 0.01 level. Similarly, because the sequences of AsJAZ4 and AsJAZ5 were highly similar, the total expression levels of AsJAZ4 and AsJAZ5 are represented by AsJAZ4/5. Normal conditions are represented by "0 h", that is, before treatment, and "1 h" and "2 h" represent 1 h or 2 h after the specific treatment. (a-d) Indicate the fold changes in the expression levels of the 12 AsJAZ genes under the MeJA, MeSA, ABA, or ETH treatment, respectively. The data represent the mean + the standard error of three independent biological replicates. Under each treatment, the relative expression level of AsJAZ7/8 at 0 h was taken as 1. The AsJAZ genes were separated into two groups based on their relative expression level ranges. The significance of the expression difference between the control (0 h) and treatments (1 h or 2 h) for each gene was evaluated using two-tailed Student's t-test, with "*" representing a significant difference at the p = 0.05 level and "**" representing a highly significant difference at the 0.01 level.

Chromosome Location of AsWRKY Genes and Phylogenetic Analysis of AsWRKY Proteins
In a previous study, Xu et al. performed a genome-wide analysis of the A. sinensis WRKY family [51]. However, because the genome sequences were not completely assembled at that time, the analysis results indicated that eight of the predicted 70 WRKY genes were located on seven contigs rather than on the chromosome. Additionally, the 70 WRKY sequences were not publicly available. In the present study, based on the high-quality A. sinensis genome presented by Ding et al. [39], the sequence characteristics of A. sinensis WRKY genes and the phylogenetic relationships between A. sinensis WRKY proteins were re-examined. Using the HMMER search program, 64 AsWRKY protein sequences were acquired. The 64 AsWRKY genes were unevenly located on eight chromosomes; chromosome seven contained the greatest number of WRKY genes (15), but only four AsWRKYs were located on chromosome five ( Figure S4). There were seven, seven, nine, five, six, and ten AsWRKY genes on chromosomes one, two, three, four, six, and eight, respectively.
To determine the phylogenetic relationships between the WRKY proteins from A. sinensis and Arabidopsis, an unrooted phylogenetic tree was constructed. In contrast to previous reports [60][61][62][63], the WRKY proteins from A. sinensis were classified into three groups based on their WRKY domain sequences, using the Arabidopsis WRKY proteins as references. The group I members were further divided into subgroup IN (mainly including the N-terminal WRKY domain) and subgroup 1C (mainly including the C-terminal WRKY domain). As shown in Figure 6 and Table S4, group I WRKYs contained ten AsWRKYs, nine of which contained two WRKY domains, with Scaffold70.27 containing only one N-terminal WRKY domain. The group II members were divided into five subgroups-subgroup IIa, subgroup IIb, subgroup IIc, subgroup IId, and subgroup IIe-which contained four, eight, eighteen, six, and ten AsWRKYs, respectively. Group III contained eight AsWRKYs. Based on the phylogenetic relationships, the orthologous pairs of AsWRKYs and AtWRKYs were determined using a relatively strict criterion; that is, the nodes of the phylogenetic tree with bootstrap support values (1000 re-sampling) exceeding 50% were considered as orthologous pairs, like in the analysis method of Ling et al. [64]. Furthermore, through homology analysis it was discovered that the nucleic acid sequence for Scafford369.3 was identical to that of AsWRKY44, which represses the expression of the wound-induced sesquiterpene biosynthetic gene ASS1 in Aquilaria sinensis [49]. The resulting 25 AsWRKYs orthologous to AtWRKYs are shown in Table S4. To facilitate the analysis, new names were given to the A. sinensis proteins orthologous to Arabidopsis WRKY transcription factors: AsWRKY33n, AsWRKY44n, AsWRKY1n, etc. This provided a distinction from the nomenclature of Xu et al. [51]. Two gene models with the same Arabidopsis orthologous protein AsWRKY40 were named AsWRKY40.1 and AsWRKY40.2. The same approach was taken to naming AsWRKY69.1, AsWRKY69.2, AsWRKY22.1, and AsWRKY22.2. New names were not given to Scaffold21.273 and Scaffold3.179, as each had two Arabidopsis orthologous proteins. Although the AsWRKY proteins (Scafford70.29 and Scafford9.167) are not the orthologous proteins for AtWRKY51 and AtWRKY75, according to the strict criterion, the two proteins were closely related to AtWRKY51 and AtWRKY75, respectively, so they were named AsWRKY51n and AsWRKY75n. The IDs and names of the AsWRKYs are shown in Table S5. Because the sequences of the 70 WRKY protein sequences described by Xu et al. [51] are not available now, only the possible Scaffold IDs were given for some of the AsWRKYs of interest by comparison with their phylogenetic groups (Table S5). Some members of the group II AsWRKYs, especially group IIc, are highly expressed in agarwood [51].

Expression Patterns of AsWRKY Genes
The expression patterns of the AsWRKY genes in the suspension cells in response to hormone and H2O2 treatments (at a final concentration of 100 µM) for 24 h were investigated based on our primary transcriptomic data. The results are presented in Figure  7. The significance of gene expression changes (treatment vs. control) is shown in Table  S6.
The results revealed that, of the AsWRKY genes specifically expressed in agarwood, as described by Xu et al. [51], the expression levels of AsWRKY75n and AsWRKY69 were significantly up-regulated after all of the treatments. In contrast, the expression levels of AsWRKY65n were decreased to different degrees after the treatments. However, the

Expression Patterns of AsWRKY Genes
The expression patterns of the AsWRKY genes in the suspension cells in response to hormone and H 2 O 2 treatments (at a final concentration of 100 µM) for 24 h were investigated based on our primary transcriptomic data. The results are presented in Figure 7. The significance of gene expression changes (treatment vs. control) is shown in Table S6.
The results revealed that, of the AsWRKY genes specifically expressed in agarwood, as described by Xu et al. [51], the expression levels of AsWRKY75n and AsWRKY69 were significantly up-regulated after all of the treatments. In contrast, the expression levels of AsWRKY65n were decreased to different degrees after the treatments. However, the expression level of AsWRKY49 was very low and was not induced by the treatments. The expression level of AsWRKY58 increased after the hormone and H 2 O 2 treatments.
Previous studies have proposed that some AtWRKYs can be regulated by AtJ-AZs, such as AtWRKY51, AtWRKY57, AtWRKY75, AtWRKY40, AtWRKY18, and AtWRKY60 [33][34][35]37,38]. In this study, the expression patterns of several AsWRKY genes that have encoding proteins orthologous to the AtWRKY proteins, such as AsWRKY51n, -57n, -75n, -40.1, -40.2, -18n, and -60n, were investigated in suspension cells through primary transcriptomic analysis. In these AsWRKY genes, AsWRKY75n has been included in the AsWRKY genes specifically expressed in agarwood, and its expression has been analyzed to be increased under all the treatments. For others, the expression of AsWRKY51n was a significant up-regulation after the ABA, ETH, JA, and SA treatments, but there were no obvious changes in response to H 2 O 2 treatment. There were no significant changes in the expression of AsWRKY57n after the various treatments. The expression levels of AsWRKY40.1 and AsWRKY40.2 were low during normal conditions, but the expression level of AsWRKY40.1 increased significantly following all treatments. The expression level of AsWRKY18n increased after all of the treatments, but the expression level of AsWRKY60n decreased after all of the treatments. AsWRKY44 has been identified as a repressor of ASS1 expression and can be induced by MeJA [49]. In the current study, the expression of AsWRKY44 was found to be significantly increased by all of the treatments. expression level of AsWRKY49 was very low and was not induced by the treatments. The expression level of AsWRKY58 increased after the hormone and H2O2 treatments. Previous studies have proposed that some AtWRKYs can be regulated by AtJAZs, such as AtWRKY51, AtWRKY57, AtWRKY75, AtWRKY40, AtWRKY18, and AtWRKY60 [33,34,35,37,38]. In this study, the expression patterns of several AsWRKY genes that have encoding proteins orthologous to the AtWRKY proteins, such as AsWRKY51n, -57n, -75n, -40.1, -40.2, -18n, and -60n, were investigated in suspension cells through primary transcriptomic analysis. In these AsWRKY genes, AsWRKY75n has been included in the AsWRKY genes specifically expressed in agarwood, and its expression has been analyzed to be increased under all the treatments. For others, the expression of AsWRKY51n was a significant up-regulation after the ABA, ETH, JA, and SA treatments, but there were no obvious changes in response to H2O2 treatment. There were no significant changes in the expression of AsWRKY57n after the various treatments. The expression levels of AsWRKY40.1 and AsWRKY40.2 were low during normal conditions, but the expression level of AsWRKY40.1 increased significantly following all treatments. The expression level of AsWRKY18n increased after all of the treatments, but the expression level of AsWRKY60n decreased after all of the treatments. AsWRKY44 has been identified as a repressor of ASS1 expression and can be induced by MeJA [49]. In the current study, the expression of AsWRKY44 was found to be significantly increased by all of the treatments.

Experimental Validation of the Interaction between AsJAZ4 and AsWRKY75n
It has been proposed that the expression levels of AtWRKY70, AtWRKY18, AtWRKY40, and AtWRKY60 could be regulated by JAZ repressors [65,66]. Interactions between specific JAZ and WRKY proteins-such as AtJAZ4/8 and AtWRKY57 or AtJAZ8 and WRKY51/75-have been reported [33][34][35]. As homologous proteins often exhibit analogous functions, hypotheses were made regarding the potential for AsJAZ4 or AsJAZ5 (AtJAZ8 protein homolog) to inhibit the activation of specific AsWRKY transcription factors, including, but not limited to, AsWRKY51n, AsWRKY57n, AsWRKY75n, AsWRKY40.1, and AsWRKY40.2. On the basis of the former analysis, the AsWRKY75n was specifically expressed in agarwood and its expression was increased after some hormone treatments, so we chose to analyze the interaction between AsJAZ4 and AsWRKY75n as a first step to reveal the interactions of AsJAZs and AsWRKYs. Using the RNA extracted from the A. sinensis calli as the raw material, the coding sequences of AsJAZ4 and AsWRKY75n were acquired through reverse transcription PCR (RT-PCR). Then, the interaction between AsWRKY75n and AsJAZ4 was investigated using the yeast two-hybrid system. The coding sequences of AsWRKY75n and AsJAZ4 were fused to the Gal4 DNA binding domain of the bait vector (BD-AsWRKY75n) and the Gal4 DNA activation domain of the prey vector (AD-AsJAZ4), respectively. Then, these two vectors were co-transformed into yeast. The Gal4 DNA binding domain expressed by pGBKT7 (BD-Empty1) or the Gal4 DNA activation domain expressed by pGADT7 (AD-Empty2) was used as a negative control. As shown in Figure 8a, after the co-transformation of BD-AsWRKY75n and AD-AsJAZ4, the transformant was able to grow on the synthetic dropout medium lacking Leu, Trp, His, and Ade, while the co-transformation of BD-AsWRKY75n and AD-Empty2, or the co-transformation of BD-Empty2 and AD-AsJAZ4, could not produce yeast with the ability to grow on the synthetic dropout medium lacking Leu, Trp, His, and Ade. These results provide in vivo experimental evidence for the interaction between AsWRKY75n and AsJAZ4, and exclude the possibility of the self-activation of these two proteins. To provide additional confirmation of the interaction between AsJAZ4 and AsWRKY75, a Mag-Beads GST Fusion Protein Purification kit was used to perform a pull-down assay. The results revealed that the immobilized AsJAZ4-GST fusion protein was capable of binding to the AsWRKY75n-His fusion protein, whereas the GST protein was not (Figure 8b). Together, the yeast two-hybrid and pull-down assay results suggest that AsJAZ4 can bind to AsWRKY75n in a similar manner to their Arabidopsis homologs.

Experimental Validation of the Interaction between AsJAZ4 and AsWRKY75n
It has been proposed that the expression levels of AtWRKY70, AtWRKY18, AtWRKY40, and AtWRKY60 could be regulated by JAZ repressors [65,66]. Interactions between specific JAZ and WRKY proteins-such as AtJAZ4/8 and AtWRKY57 or AtJAZ8 and WRKY51/75-have been reported [33,34,35]. As homologous proteins often exhibit analogous functions, hypotheses were made regarding the potential for AsJAZ4 or AsJAZ5 (AtJAZ8 protein homolog) to inhibit the activation of specific AsWRKY transcription factors, including, but not limited to, AsWRKY51n, AsWRKY57n, AsWRKY75n, AsWRKY40.1, and AsWRKY40.2. On the basis of the former analysis, the AsWRKY75n was specifically expressed in agarwood and its expression was increased after some hormone treatments, so we chose to analyze the interaction between AsJAZ4 and AsWRKY75n as a first step to reveal the interactions of AsJAZs and AsWRKYs. Using the RNA extracted from the A. sinensis calli as the raw material, the coding sequences of AsJAZ4 and AsWRKY75n were acquired through reverse transcription PCR (RT-PCR). Then, the interaction between AsWRKY75n and AsJAZ4 was investigated using the yeast two-hybrid system. The coding sequences of AsWRKY75n and AsJAZ4 were fused to the Gal4 DNA binding domain of the bait vector (BD-AsWRKY75n) and the Gal4 DNA activation domain of the prey vector (AD-AsJAZ4), respectively. Then, these two vectors were co-transformed into yeast. The Gal4 DNA binding domain expressed by pGBKT7 (BD-Empty1) or the Gal4 DNA activation domain expressed by pGADT7 (AD-Empty2) was used as a negative control. As shown in Figure 8a, after the co-transformation of BD-AsWRKY75n and AD-AsJAZ4, the transformant was able to grow on the synthetic dropout medium lacking Leu, Trp, His, and Ade, while the co-transformation of BD-AsWRKY75n and AD-Empty2, or the co-transformation of BD-Empty2 and AD-AsJAZ4, could not produce yeast with the ability to grow on the synthetic dropout medium lacking Leu, Trp, His, and Ade. These results provide in vivo experimental evidence for the interaction between AsWRKY75n and AsJAZ4, and exclude the possibility of the selfactivation of these two proteins. To provide additional confirmation of the interaction between AsJAZ4 and AsWRKY75, a Mag-Beads GST Fusion Protein Purification kit was used to perform a pull-down assay. The results revealed that the immobilized AsJAZ4-GST fusion protein was capable of binding to the AsWRKY75n-His fusion protein, whereas the GST protein was not (Figure 8b). Together, the yeast two-hybrid and pulldown assay results suggest that AsJAZ4 can bind to AsWRKY75n in a similar manner to their Arabidopsis homologs.

Classification of the AsJAZ Family Members
The current study identified 19 AsTIFY genes, including 12 AsJAZ genes. Several conserved cis-elements were found in the promoter regions of specific AsJAZ genes; these are anticipated to be responsive to various phytohormones, such as MeJA, ABA, GA, auxin, and SA. Previous studies have shown that Arabidopsis JAZs can be grouped into four clades [24,25]. Nevertheless, Bai et al. suggested that-based on the topology of the phylogenetic tree, clade support values, and manual inspections-JAZ proteins from angiosperms belong to five groups [58]. Here, the AsJAZ proteins were classified into five groups according to the phylogenetic analysis results. This classification is consistent with that proposed by Bai et al. [58]. Gene structure analysis revealed that conserved intron phases exist in the TIFY and Jas motifs of the same classes of AsJAZ genes. This result is also consistent with the findings of Bai et al. [58] with regard to TIFY genes found in Arabidopsis. In these AsJAZ proteins, AsJAZ10 contained the conserved core motif of the CMID domain and belonged to the JAZ-III group. Since AsJAZ10 is a homolog of AtJAZ10, it was postulated that AsJAZ10 behaves in a similar manner to AtJAZ10 and potentially interacts with MYC transcription factors [52,57]. In phylogenesis, AsJAZ4, -5, and -11 were located in the JAZ-IV clade, with all three containing the non-canonical Jas motif. Specifically, AsJAZ4 and AsJAZ5 possessed the EAR motif in their N-terminal regions, a characteristic that was also shared by AtJAZ8; this mediates the transcription repression of JA responses in Arabidopsis [54]. Therefore, it is hypothesized that AsJAZ4 or AsJAZ5 perform an analogous function to AtJAZ8. However, in the same JAZ-IV group AsJAZ11 did not have the N-terminal EAR motif and exhibited weak relationships with AsJAZ4 and AsJAZ5. On the other hand, in terms of the number of exons and their arrangement, AsJAZ11 was not grouped with AsJAZ4 and AsJAZ5 and instead was grouped with AsJAZ12. Therefore, AsJAZ11 may have different functions to AsJAZ4 or AsJAZ5.

Expression Patterns of the AsJAZ Family Members
The tissue-specific expression patterns of AsJAZ genes in a five-year-old A. sinensis tree injected with formic acid for 12 months were investigated [50]. The findings revealed that AsJAZ1, AsJAZ2, and AsJAZ12 exhibited relatively higher expression levels in the white wood (B1), brown agarwood (J3), and transition part of the trunk (W2), as compared to the other tissues. However, the three genes exhibited lower expression levels in the suspension cells. The expression levels of the three genes were even down-regulated upon short-duration treatment with some hormones. This suggests that AsJAZ1, AsJAZ2, and AsJAZ12 may be specifically expressed in mature tissues. Furthermore, the three genes may be more strongly correlated with the process of sesquiterpene production and agarwood formation in the A. sinensis tree than other AsJAZ genes. Consistent with this speculation, Liao et al. recently reported the cloning and characterization of a gene named AsJAZ1 (GenBank accession number: KP677281) from A. sinensis, which was identical to the AsJAZ2 gene in this article [67]. It has been demonstrated that AsJAZ2 (or AsJAZ1, in the report of Liao et al.) can interact with the AsMYC2 transcription factor and may be involved in repressing the expression of sesquiterpenes in Aquilaria plants. It is speculated that the functions of AsJAZ1, AsJAZ2, or AsJAZ12 would be to repress the over-production of defense molecules, such as sesquiterpenes, so as to avoid the excessive energy consumption of the tree. As for the investigation of the expression patterns of the AsJAZ genes in suspension cells upon the various hormone treatments, AsJAZ4/5, AsJAZ7/8, AsJAZ9, and AsJAZ10 exhibited higher expression ranges. It is suggested that the functions of AsJAZ4/5, AsJAZ7/8, AsJAZ9, and AsJAZ10 would be to repress the excessive cell defense response under various treatments. The expression of AsJAZ4/5 in the suspension cells after short-duration MeJA treatment was significantly higher than the other genes under the various hormone treatments. This suggests that AsJAZ4/5 plays important roles in the JA-mediated signaling pathways of A. sinensis suspension cells. Together, these findings provide important guidance for future in-depth analysis of the functions of the 12 AsJAZ genes, particularly in relation to their functions in the JA-mediated defense response and the agarwood formation process.

Classification and Expression Patterns of the AsWRKY Genes
This study identified 64 AsWRKY transcription factors based on the high-quality A. sinensis genome. Upon phylogenetic analysis, the 64 AsWRKYs were divided into three major groups; the group II WRKRs were further divided into five subgroups (IIa, IIb, IIc, IId, and IIe). The group IIa WRKY genes regulate plant disease resistance by participating in disease-resistance signaling pathways. Some of them, such as GaWRKY1 in cotton and AaWRKY1 in Artemisia annua, regulate the biosynthesis of sesquiterpene phytoalexin [68][69][70]. The present study revealed that the A. sinensis genome contains four genes belonging to the group IIa WRKY family: AsWRKY18n, AsWRKY40.1, AsWRKY40.2, and AsWRKY60n. These likely have similar functions to AtWRKY18, AtWYKY40, and AtWRKY60, which are also group IIa WRKYs. AsWRKY33n belongs to group I and may have a similar function to AtWRKY33, which is an important contributor to necrotrophic pathogen disease resistance and interacts with sigma factor binding proteins [71,72]. Previous transcriptome analyses have revealed five AsWRKY genes specifically expressed in agarwood, i.e., AsWRKY13, -38, -49, -58, and -69 [51]. Sequence comparisons revealed that AsWRKY75n was identical to the AsWRKY13 described by Xu et al. [51]. The transcriptome analysis in the current study indicated that AsWRKY75n was significantly induced by ABA, ETH, MeSA, MeJA, and H 2 O 2 in suspension cells, implying a possible relationship between this gene and the stress response in agarwood. Further, among the AsWRKY genes, AsWRKY51n and AsWRKY75n were significantly induced by short-duration MeJA treatment in suspension cells. This indicates that AsWRKY75n plays a possible role in JA-mediated plant defense against insect and necrotrophic pathogens.

Possible Roles of the AsJAZ4/WRKY75n Complex
In a previous study, the AsJAZ1/AsMYC2 complex was found to be involved in the regulation of ASS1 expression [48,67]. Previous studies have also shown that in Arabidopsis, the AtJAZ4/AtWRKY57, AtJAZ8/AtWRKY57, AtJAZ8/AtWRKY51, and AtJAZ8/AtWRKY75 complexes play significant roles in JA-mediated defense pathways [33][34][35]. As homologous proteins often display analogous functions, it was speculated that AsJAZ4/5 (AtJAZ8 protein ortholog) may be a repressor of several specific AsWRKY transcription factors. In keeping with our speculation, a yeast two-hybrid experiment showed an interaction between AsJAZ4 (AtJAZ8 ortholog) and AsWRKY75n (AtWRKY75 ortholog) in vivo and a pull-down assay verified this interaction in vitro ( Figure 8). Consistent with this, the qPCR and transcriptomic analysis results indicated that the expression levels of AsJAZ4/5 and AsWRKY75n increased after MeJA treatment, which implies that they are potentially involved in the JA-mediated signaling pathways (Figures 5 and 7). Since unique JAZ/transcription factor complexes are key factors in determining the specificity of the JA signaling pathways, it would be important to investigate the specific roles of the As-JAZ4/WRKY75n complex in the JA-mediated defense pathways. In Arabidopsis, AtWRKY75 was found to be involved in various biological processes, such as seed germination, leaf senescence, flowering, salt tolerance, and responding to necrotrophic pathogens [35,[73][74][75][76]. In recent years, it has been discovered that DcWRKY75 (a homologue of AtWRKY75n) promotes ethylene-induced petal senescence in carnations [77,78]. As a homologue of AtWRKY75, AsWRKY75n may be a key player actively involved in developmental and defense pathways in A. sinensis, and AsJAZ4 may be its repressor. Based on this speculation, a simple model depicting the regulatory role of the AsJAZ4/WRKY75n complex in A. sinensis suspension cells was developed. As shown in Figure 9, in normal conditions, AsJAZ4 represses the activities of AsWRKY75n and inhibits the gene's transcription. Upon MeJA treatment, AsJAZ4 will be degraded and AsWRKY75n will be activated to promote the transcription of the defensive genes. In the meantime, the expression level of the AsJAZ4 gene was increased to produce new AsJAZ proteins. The newly produced AsJAZ4 proteins will inhibit the activities of AsWRKY75n and avoid the injury originated from the overexpression of the defense genes through a feedback mechanism. This model will provide new insights into the defensive signaling of A. sinensis. of the AsJAZ4/WRKY75n complex in A. sinensis suspension cells was developed. As shown in Figure 9, in normal conditions, AsJAZ4 represses the activities of AsWRKY75n and inhibits the gene's transcription. Upon MeJA treatment, AsJAZ4 will be degraded and AsWRKY75n will be activated to promote the transcription of the defensive genes. In the meantime, the expression level of the AsJAZ4 gene was increased to produce new AsJAZ proteins. The newly produced AsJAZ4 proteins will inhibit the activities of AsWRKY75n and avoid the injury originated from the overexpression of the defense genes through a feedback mechanism. This model will provide new insights into the defensive signaling of A. sinensis.

Bioinformatic Analysis of TIFY Family Members in A. sinensis
The protein sequences of A. sinensis were downloaded from the GigaDB database [39]. The hidden Markov model (HMM) profiles of the TIFY domain (PF06200) were extracted from the Pfam database (http://Pfam.sanger.ac.uk, accessed on 21 August 2021) and searches with the HMM model were carried out using the hmmsearch program from the HMMER package against the predicted A. sinensis proteins [79]. The significant hits (E value < 10 10 ) were chosen as possible TIFY proteins. All candidate sequences were

Bioinformatic Analysis of TIFY Family Members in A. sinensis
The protein sequences of A. sinensis were downloaded from the GigaDB database [39]. The hidden Markov model (HMM) profiles of the TIFY domain (PF06200) were extracted from the Pfam database (http://Pfam.sanger.ac.uk, accessed on 21 August 2021) and searches with the HMM model were carried out using the hmmsearch program from the HMMER package against the predicted A. sinensis proteins [79]. The significant hits (E value < 10 10 ) were chosen as possible TIFY proteins. All candidate sequences were submitted for domain analysis using the SMART tool (http://smart.embl-heidelberg.de/, accessed on 21 August 2021) for further validation. For the putative TIFY protein family members, the theoretical pI and predicted molecular weight (MW) were acquired using the Expasy portal tool (http://web.expasy.org/compute_pi/, accessed on 21 August 2021). The conserved motif was analyzed using the MEME online server (http://meme-suite.org/ tools/meme, accessed on 21 August 2021). The domain and motif analysis results were visualized using the IBS online illustrator (http://ibs.biocuckoo.org/online.php, accessed on 25 August 2021) [80]. Exon-intron gene structures were visualized using the GSDS 2.0 server (http://gsds.gao-lab.org/, accessed on 25 September 2021) [81]. To analyze the cis-acting regulatory elements, 2 kb upstream of the AsJAZ genes was extracted and searched against the PlantCARE database (http://bioinformatics.psb.ugent.be/webtools/ plantcare/html/, accessed on 25 September 2022) [82,83]. The promoter element analysis results were visualized using TBtools (version 1.09876) [84]. The chromosome location maps were constructed using Mapchart (version 2.3) [85].

Bioinformatic Analysis of WRKY Family Members
The HMM profiles of the WRKY domain (PF03106) were extracted from the Pfam database, and HMMER 3.0 was used to perform local HMM searches against the A. sinensis proteins in the GigaDB database. The significant hits (E value < 10 10 ) were chosen, and the proteins were further screened and validated using the SMART tool. The chromosome location map was constructed using Mapchart (version 2.3) [85].

Sequence Alignment and Phylogenetic Analysis
The representative TIFY protein and WRKY protein sequences of Arabidopsis were downloaded from the TAIR database (www.arabidopsis.org, accessed on 25 September 2022), with the accession numbers obtained from the descriptions of Vanholme et al. [2] and Rushton et al. [86]. The TIFY phylogenetic tree construction used the full-length TIFY protein sequences, but the WRKY tree used the WRKY conserved domains. The WRKY conserved domains were extracted from the sequences of the Arabidopsis and A. sinensis WRKY proteins using a local perl script. The N-terminal WRKY domain and C-terminal WRKY domain were separately extracted for the two WRKY domains containing proteins. Both the TIFY tree and WRKY tree were constructed using MEGA-X (version 10.2.6) [87]. First, the amino acid sequences were aligned using MUSCLE, and then the alignment files were imported into MEGA-X for the construction of the Maximum Likelihood tree using 1000 bootstrapped replications. The JTT + G + I + F model was used to construct the TIFY tree, and the JTT + G model was used to construct the WRKY tree. Finally, the phylogenetic trees were modified using the online EvolView tool, with only the nodes with a cutoff greater than 50% shown [88].

Plant Sample Preparation
The suspension cells were derived from Aquilaria sinensis (Lour.) Spreng calli and prepared based on a previously reported method [89]. Briefly, the fresh young leaves of A. sinensis were sterilized using 10% sodium hypochlorite solution, cut into small pieces and cultured on Murashige-Skoog (MS) solid culture medium supplemented with Phytagel (Sigma), α-naphthalene acetic acid (NAA) (2.0 mg·L −1 ), and 6-benzylaminopurine (6-BA; 1.0 mg·L −1 ). After the leaves were cultured and transferred about 5 times, energetic and loose calli were generated and used to prepare suspension cells by shaking in liquid MS medium containing NAA (2.0 mg·L −1 ), 6-BA (1.0 mg·L −1 ), and casein hydrolysate (0.5 g·L −1 ). After the suspension cells were sub-cultured 5-6 times, the fresh and energetic suspension cells acquired after growth for one more week were subjected to one of the following treatments at a final concentration of 100 µM: methyl salicylate (MeSA), methyl jasmonate (MeJA), ethephon (ETH), or (±) abscisic acid (ABA). The suspension cells were treated with various hormones for 0 h (control sample without treatment), 1 h or 2 h, then collected by centrifugation and frozen in liquid nitrogen for real-time quantitative PCR (qPCR). The suspension cells were also treated with MeSA, MeJA, ETH, H 2 O 2 , or ABA (100 µM final concentration) for 24 h and the samples were stored in liquid nitrogen for Illumina-based high-throughput sequencing.

Real-Time Quantitative PCR
Total RNA was extracted using the EASYspin Plus Plant RNA Kit (Aidlab, Bejing, China), according to the manufacturer's instructions. First-strand cDNA synthesis was performed using M-MLV Reverse Transcriptase according to the instructions (Promega, Madison, WI, USA). The real-time qPCR was performed using TransStart Top Green qPCR Supermix (Transgene, Beijing, China) on a Roche LightCycler96 Real-Time System. The 2 −∆∆Ct method was used to evaluate the expression levels. Three independent replications were performed for each experiment. The housekeeping gene AsTUB was used as the internal control, as described by Gao et al. [90]. All primers were designed using the online tool Primer3Plus (https://www.primer3plus.com/, accessed on 25 September 2022) and are listed in Table S1. Because AsJAZ7 and AsJAZ8 had highly similar sequences, they used the same pair of qPCR primers; AsJAZ4 and AsJAZ5 also used the same pair of primers for their highly similar sequences.

Statistical Analysis
Statistically significant differences (* p < 0.05 or ** p < 0.01) were determined using the IBM SPSS statistics 26 package (International Business Machines Corp., New York, NY, USA). Data were expressed as the mean + standard error of the mean. Twotailed Student's t-tests were performed to compare two groups from three independent biological experiments.

Transcriptomic Analysis and Heatmap Illustrating
A. sinensis suspension cells were treated with 100 µM MeJA, MeSA, ETH, H2O2, or ABA for 24 h. Three replicates were collected by centrifugation and pooled to form a sample for RNA sequencing on the Illumina Novaseq platform. Sequencing libraries were generated using the NEBNext ® UltraTM RNA Library Prep Kit for Illumina ® (NEB, Ipswich, MA, USA), following the manufacturer's recommendations. The generated 150 bp paired-end reads were filtered to obtain clean reads and the HISAT2 v2.0.5 program was used for mapping the clean reads against the A. sinensis genome [91]. Gene expression levels were estimated by calculating their FPKM (Fragments Per Kilobase of transcript per Million fragments mapped) value using Cufflinks (version 2.2.1) [92]. Genes with FRKMs values of >1 were regarded as expressed. For the significance analysis of gene expression difference, edgeR software (version 3.38.1) was used [93]. Briefly, the read count was normalized and the p-value was calculated via the negative binomial distribution model. Then, the false discovery rate or the error detection rate (FDR, or padj) was calculated. Finally, the genes with significantly changed expression levels between each group (treatment vs. CK) were screened out based on |log 2 FoldChange| > 1 and a padj value of <0.05. The tissue-specific AsJAZ gene expression data of a five-year-old A. sinensis tree were acquired by re-analyzing the transcriptome data previously published by Ye et al. [50]. In brief, the raw data of the transcriptome of the five-year-old A. sinensis tree injected with formic acid for 12 months (SRP068230) were downloaded and the reads were re-mapped against the genome released by Ding et al. [50] using the HISAT2 v2.0.5 program, and then the FPKM values for the genes were re-calculated using Cufflinks (version 2.2.1) [92]. Finally, the gene expression levels were compared according to the FPKM values, and the heatmap reflecting the gene's expression patterns was produced using TB-tools [86].

Protein Expression and Pull-Down Assay
Coding sequences of AsWRKY75n and AsJAZ4 were PCR-amplified from A. sinensis calli cDNAs, with the primers shown in Table S2. The coding sequence of AsWRKY75n was inserted into plasmid pET28a (+) behind the His tag sequence and AsJAZ4 was inserted into plasmid pGEX-4T-1 behind the GST tag sequence by a homologous recombination method. His-fused AsWRKY75n and GST-fused AsJAZ4 were expressed in E. coli BL21-CodonPlus (DE3). Suitable inducing conditions for the soluble protein were determined. The His-fused AsWRKY75n protein was expressed in the supernatant under the induction of 0.2 mM IPTG for 4 h at 18 • C, while the GST-fused AsJAZ4 protein was only partially expressed in the supernatant under the induction of 0.2 mM IPTG for 0.5 h at 18 • C. The pull-down assay was performed using the Mag-Beads GST Fusion Protein Purification kit (Sangon, Shanghai, China), according to the manufacturer's instructions. Briefly, the AsJAZ4-GST protein or the GST protein produced by the empty vector was adsorbed onto magnetic agarose microspheres pretreated with binding buffer (140 mM NaCl, 2.7 mM KCl, 10 mM Na 2 HPO 4 , and 1.8 mM KH 2 PO 4 , pH 7.4). Then, the AsWRKY75n-His protein was incubated with the AsJAZ4-GST protein and magnetic agarose microspheres complex for 30 min. The unbound proteins were removed by washing twice with washing buffer (same composition as the binding buffer). Finally, the proteins were eluted from the magnetic agarose microspheres using the elution buffer (50 mM Tris-HCl, 10 mM reduced glutathione, pH 8.0), separated by SDS-PAGE and checked with Western blot.

Yeast Two-Hybrid
The yeast two-hybrid experiment was performed via the mating protocol described in the Matchmaker Gold Yeast Two-Hybrid user manual (Clontech, Mountain View, CA, USA). Firstly, the AsWRKY75 coding sequence was fused to the DNA binding domain of the bait vector pGBKT7 to construct the vector pGBKT7-BD-AsWRKY75n via a seamless cloning method using the EasyGeno Single Assembly Cloning Kit (TIANGEN, Beijing, China). Then, the AsJAZ4 coding sequence was fused to the activation domain of the prey vector pGADT7 to construct the plasmid pGADT7-AD-AsJAZ4 using a similar method. Then, the two constructs were co-transformed into the yeast strain Y2H Gold. The co-transformation of the empty vector pGBKT7 with pGADT7-AD-AsJAZ4, or pGADT7 with pGBKT7-BD-AsWRKY75n, was used as a negative control. Finally, the success of the co-transformation was detected by the ability of the cells to grow on the synthetic dropout medium lacking Leu and Trp, and the interaction between the co-transformed proteins was detected by the ability of the cells to grow on synthetic dropout medium lacking Leu, Trp, His, and Ade. The primers used for the vector construction are listed in Table S2.

Conclusions
This study aimed to examine the characteristics of JAZ and WRKY gene family members in A. sinensis and their possible regulation roles in defense signaling and agarwood formation. The findings revealed 12 AsJAZ genes. The structures, chromosome locations, and expression patterns of these genes were determined. The phylogenetic relationships between the putative AsJAZ proteins and their Arabidopsis homologs were elucidated. The expression analysis showed that AsJAZ1, -2, and -12 exhibited higher expression levels than the others in mature trees, while AsJAZ4/5, AsJAZ7/8, AsJAZ9, and AsJAZ10 were expressed at higher levels in suspension cells. This study identified 64 AsWRKY genes. Some AsWRKY genes appear to be involved in JA-mediated plant defense pathways, based on the analysis of their expression patterns. This is the first systematic study of the AsJAZ genes, and provides experimental evidence for an interaction between AsJAZ4 and AsWRKY75n. The present study has revealed important aspects of the specificity and redundancy of AsJAZ proteins and has uncovered vital insights into their regulation modes via interactions with specific transcription factors. These findings may provide an important basis for future studies of the JA-mediated signaling pathways in A. sinensis and the agarwood induction mechanism. Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: The transcriptomic data for A. sinensis callus-derived suspension cell samples were submitted to the NCBI SRA database with the accession number SRP323444 (the BioProject number was PRJNA736326). The published transcriptomic data for the five-year-old A. sinensis tree injected with formic acid for 12 months was downloaded from the NCBI SRA database with the accession number SRP068230. The gene and peptide sequences are included in the tables in the Supplementary Materials. The genomic and mRNA sequences of AsJAZ4 have been submitted to Genebank with the accession numbers of OK507018 and OK507019, respectively.