Genome-wide analysis of JAZ family genes expression patterns during fig (Ficus carica L.) fruit development and in response to hormone treatment

Background Jasmonate-ZIM domain (JAZ) repressors negatively regulate signal transduction of jasmonates, which regulate plant development and immunity. However, no comprehensive analysis of the JAZ gene family members has been done in the common fig (Ficus carica L.) during fruit development and hormonal treatment. Results In this study, 10 non-redundant fig JAZ family genes (FcJAZs) distributed on 7 chromosomes were identified in the fig genome. Phylogenetic and structural analysis showed that FcJAZ genes can be grouped into 5 classes. All the classes contained relatively complete TIFY and Jas domains. Yeast two hybrid (Y2H) results showed that all FcJAZs proteins may interact with the identified transcription factor, FcMYC2. Tissue-specific expression analysis showed that FcJAZs were highly expressed in the female flowers and roots. Expression patterns of FcJAZs during the fruit development were analyzed by RNA-Seq and qRT-PCR. The findings showed that, most FcJAZs were significantly downregulated from stage 3 to 5 in the female flower, whereas downregulation of these genes was observed in the fruit peel from stage 4 to 5. Weighted-gene co-expression network analysis (WGCNA) showed the expression pattern of FcJAZs was correlated with hormone signal transduction and plant-pathogen interaction. Putative cis-elements analysis of FcJAZs and expression patterns of FcJAZs which respond to hormone treatments revealed that FcJAZs may regulate fig fruit development by modulating the effect of ethylene or gibberellin. Conclusions This study provides a comprehensive analysis of the FcJAZ family members and provides information on FcJAZs contributions and their role in regulating the common fig fruit development. Supplementary Information The online version contains supplementary material available at 10.1186/s12864-022-08420-z.


Introduction
Jasmonic acid and its oxylipin derivatives (Jasmonates, JAs) are a class of lipid-phytohormones. These components participate in plant growth, development, abiotic and biotic stress responses [1,2]. A previous study reported that JA and methyl jasmonate (MeJA) induce accumulation of protease inhibitors and adaptation mechanisms to withstand insect attack [3][4][5]. MeJA play a role in the initiation and modulation of fruit developmental processes by regulating JA levels [6].
Jasmonate-ZIM domain (JAZ) proteins are key repressors that negatively regulate the plant response to Jas in the signal transduction pathway. Studies report that when JAs are absent, JAZs bind to the adaptor protein novel interactor of jaz (NINJA), use the ERF-associated amphiphilic repression (EAR) domain to recruit TOP-LESS (TPL) to form a co-repressor complex. They further Open Access *Correspondence: chuanlinzheng@cau.edu.cn 1 College of Horticulture, China Agricultural University, Beijing 100193, China Full list of author information is available at the end of the article bind to downstream transcription factors thus interfering with transcription coactivator MED25 binding [7]. On the contrary, the biologically active JA-Ile conjugate mediate JAZ protein binding to corona insensitive 1 (COI1), the F-box component of the E3 ubiquitin ligase complexes (Skip/Cullin/F-box-type, SCFCOI1), which is subsequently ubiquitinated and degraded by the 26S proteasome, thus releasing the inhibitory effect of JAZ protein on transcription factors or signal transduction components [7]. Expression patterns of JAZs are spatiotemporal-specific. The proteins can interact with different transcription factors, implying the functional specificity of different JAZs. In addition, they can interact with other hormone signaling pathway elements to modulate biological processes [8,9].
JAZ family proteins are plant-specific TIFY family members. There are 13 members have been reported in Arabidopsis (Arabidopsis thaliana), 15 in rice (Oryza sativa), 23 in corn (Zea mays), 12 and 16 in two Petunia progenitors (Petunia axillaris and Petunia inflata) and 13 in tea plant (Camellia sinensis) [10][11][12][13][14]. Recently, Heidari done a genome-wide analysis of TIFY family genes in tomato and maize [15]. JAZ family members share a conserved TIFY sequence (TI[F/Y]XG) within the ZIM motif, which mediates interaction of JAZ and NINJA or is implicated in dimerization of JAZ-JAZ proteins. MYC2 is a basic helix-loop-helix-like transcription factor that binds the G-box-containing promoters of the downstream genes, and interacts with MED25 subunit [16,17]. The C-terminal Jas domain is implicated in interaction of JAZs and MYC2, thus suppressing JA responses and distinguishes JAZs members from the other members of the TIFY family (such as PEAPOD subfamily) [16,[18][19][20]. In the ubiquitin JAZ-mediated degradation pathway, the Jas domain facilitates interaction between JAZ and COI1, with the conserved motif of SLX2FX2KRX2RX5PY [21]. ZIM and Jas domains are highly conserved among the JAZ protein members, indicating there are functional importance and synergy among the JAZ family members [8]. Single gene loss-of-function mutant plants of JAZ5 or JAZ13 do not show significant difference compared with the wild type plant. On the contrary, JAZ5/10 and JAZ10/13 double mutants are significantly sensitive to JA application. Furthermore, the ten JAZ mutant plant jazU (JAZ1/2/3/4/5/6/7/9/10/13) showed more significant phenotype compared with the five mutant jazQ (JAZ1/3/4/9/10), including severe growth inhibition and high sensitivity to JA application [22].
Members of JAZ family interact with different transcription factors (MYC/JAM/ICE/MYB/TOE) to regulate specific downstream genes expression, involved in regulation of root elongation and growth [23], leaf senescence [24], insect resistance [25], freezing stress [26,27], flowering time of plants [28]. Moreover, JAZs participate in regulation of various biological processes, by combining with the components of the other hormone signal transduction pathways [8]. For example, Arabidopsis JAZ protein can interact with ethylene insensitive3 (EIN3) and EIN3-LIKE1 (EIL1), the core transcription factors in the ethylene signaling pathway, and interfere with their activity in root growth regulation [29]. DELLA protein, the inhibitor of gibberellin (GA) signal transduction, can also interact with JAZ protein thus regulating the balance between plant growth and development, and defense responses [30].
Common fig (Ficus carica L.) is one of the oldest horticultural crops found in many parts of the world. Fig fruit has antioxidant properties with pleasant taste and high nutritional value [31]. Fig fruit is an aggregate fruit consisting small single drupes. The edible part develops from a closed ovary, called syconium, which contains many unisexual female flowers [32]. During the process of fig fruit development, a typical double sigmoid growth curve is observed, known as fast-slow-fast growth period [33]. After long-term evolution, common fig has established a complex resistance mechanism to resist stress by abiotic factors, pathogens and insects. JAs are reported to play an important role in this regulation [34]. However, genome-wide characterization of the JA signaling-related components and their expression patterns in fig fruit has not been performed. In the current study, wholegenome data of "Horaishi" and "Dottato" [35,36] and transcriptome data of "zibao" fig from different developmental stages (NCBI accession No. PRJNA723733, data uploaded by our team) were used for prediction of the JAZ family members. Characteristics of FcJAZs, genomewide gene co-expression network and expression patterns were analyzed. The findings provide understanding of the roles of FcJAZs in regulating fig fruit development.  (Table 1). Analysis using InterProScan and SMART webservers indicated that all of the 10 putative JAZ proteins had highly conserved TIFY and Jas domains. The CDS lengths of the 10 FcJAZ genes ranged from 381 bp (FcJAZ7) to 1155 bp (FcJAZ3). FcJAZs genes were distributed in 7 chromosomes. The theoretical molecular weights ranged from 14,163.97 Da (FcJAZ8) to 42,241.05 Da (FcJAZ7) and isoelectric points (pI) of these proteins ranged from 4.92 (FcJAZ5) to 9.81 (FcJAZ7). All predicted proteins showed basic pH (pI > 7) apart from FcJAZ5. Prediction of protein subcellular localization indicated that all FcJAZs proteins were localized in the nucleus.

JAZ family members were identified in fig
We identified 21 duplicate gene pairs in the JAZ family of figs (Supplementary Table 4). The Ka/Ks ratios of FcJAZ1/FcJAZ10, FcJAZ1/FcJAZ7, FcJAZ1/FcJAZ3, FcJAZ6/FcJAZ2, FcJAZ9/FcJAZ10, FcJAZ10/FcJAZ8, FcJAZ10/FcJAZ2, FcJAZ4/FcJAZ2 and FcJAZ5/FcJAZ2 were > 1, suggesting positive selection. The Ka/Ks ratio of the remaining 12 pairs were < 1, indicating that they are negatively selected. Negative selection is redundant with positive selection and (Ka/Ks < 1) indicates that harmful mutations are eliminated in the gene family and the protein remains unchanged, which is a purification selection.

Phylogenetic analysis and sequence alignment of FcJAZs showed high conservation of JAZs proteins
To explore the evolutionary relationships among the TIFY (acc. PF06200) and jas domain (acc. PF09425) conserved motifs were explored in all FcJAZ proteins ( Fig. 1 B). Motif analysis was performed using MEME webserver to identify the conserved domain in these proteins. The findings showed that ZIM domain of TIFY contained a conserved TIFYXG motif, and the core sequence of the jas domain was SLXRFLEKRX(2) R ( Fig. 1 B; Fig S1). Notably, the N-terminus of FcJAZ8 and the C-terminus of FcJAZ6 contained an EAR domain (LxLxLx type), which is implicated in repressing transcription through interaction with the TOPLESS (TPL) co-repressor ( Fig. 1 B; Fig S1) [21]. Multiple sequence alignment of the protein sequences indicated that the Jas domain contains the degron sequence (LPIARR) which plays an important role in COI1-JA-Ile-JAZ complex formation and is physically associated with JA-Ile ( Fig  S1).

cis-elements and gene structure analysis of FcJAZs
To explore the gene structure and evolutionary trajectory of FcJAZs, cis-elements located in the promotor region and organizations of exon-intron were analyzed. Cis-elements analysis was performed using the 2000 bp region upstream of the FcJAZ genes using PlantCARE and PlantPan tools. The findings showed that most of the promoters in FcJAZs contained various plant hormone response-related elements (Fig. 2 A). In addition to MeJA, abscisic acid (ABRE), Auxin (TGA and HD-Zip 3), gibberellic acid (P-box, TATA box and CARE), ethylene (W-box) and salicylic acid (TCA) related elements were identified in the promotor regions. Notably, lightresponsive elements were the most abundant cis-elements. FcJAZ5 and FcJAZ9 contained a MYB binding site [aC(G/C)GTTA] which is implicated in flavonoid biosynthesis (Table S2) [37]. High intron number correlates with complex regulation, implying a more recent origin of the gene [38]. Analysis showed presence of 1-8 introns in FcJAZ genes (Fig. 2B). FcJAZ8 had lowest number of introns, whereas FcJAZ9 and FcJAZ10 had the highest number of introns. Notably, a high sequence similarity observed in the exon regions implies a complex regulatory during evolutionary process.
For the fig fruit development, the RNA-seq data for FcJAZ genes expression patterns were obtained from the NCBI database (Accession No. PRJNA723733), and further verification of the RNA-seq results were performed by the qRT-PCR. The red line indicated the qRT-PCR results, and the blue line indicated the RNAseq results. For the different tissues, the expression levels FcJAZ genes were determined by qRT-PCR, different shapes and colors represent different tissues as indicated in the figure. a FcJAZ family members with FPKM value greater than 100 in RNA-seq results. b Members of the FcJAZ family with FPKM between 0-100 in the RNAseq results. Different letters indicated significant differences according to Duncan's new multiple range test for the qRT-PCR (red) and RNA-seq (blue) data respectively, P < 0.05.

Expression analysis of FcJAZs in development stage and different tissues
To explore the expression pattern of FcJAZ genes during the developmental process of fig fruits, female flower and fruit peel were collected at different stages of growth. RNA isolation and gene expression analysis was then performed using the female flower and fruit peel samples. Analysis of female flower development showed that most FcJAZ genes were expressed at a high level in the early stages then expression levels decreased in the later stages. Analysis of expression profiles of FcJAZ2 and FcJAZ8 revealed lower expression level in the early stages than in later stages (Fig. 3). Analysis of fruit peel development showed that most of FcJAZs had low expression levels in the later stages whereas FcJAZ2 was slightly upregulated in the later stages. Most FcJAZs were significantly downregulated from stage 3 to 5 in the female flower whereas most genes were upregulated from stage 4 to 5 in the fruit peel. These findings indicate that these regulated FcJAZ genes may play significant roles at special windows during the process of fig fruit development.
To examine the expression level of FcJAZ genes in specific tissues, root, stem, young leaf, old leaf, female flower and fruit peel of fig tree was sampled and used for RNA analysis and gene expression analysis. The findings showed that the expression patterns of FcJAZ1, FcJAZ4 and FcJAZ8 were similar in the group of FPKM > 100. FcJAZ1, FcJAZ4 and FcJAZ8 were highly expressed in the root, followed by female flower and stem. FcJAZ3 was highly expressed in the female flower, stem and root in that order whereas FcJAZ6 was highly expressed in stem and root (Fig. 3A). In the group of 0 < FPKM < 100, FcJAZ2 and FcJAZ7 genes were highly expressed in the female flower, FcJAZ5 was highly expressed in the root, and the other FcJAZs showed   negatively correlated with FcJAZs expression patterns (correlation coefficient < -0.5, P value < 0.05) in the female flower and 2,688 genes were obtained (Fig S2 A). On the contrary, FcJAZ2 and FcJAZ8 showed opposite regulatory correlations with the other 8 FcJAZs. These findings imply that these two genes may play a role as inhibitors during female flower development. Analysis of the peel showed that six modules including tan, salmon, green, purple, blue and turquoise were positively correlated with FcJAZs expression patterns, and 16,194 genes were screened out. Four modules including brown, yellow, grey60 and red were negatively correlated with FcJAZs expression patterns, and 3,783 were screened out. Notably, FcJAZ2 showed significant differential regulation in the peel compared with the other genes (Fig S2 A). There are 12,158 genes showed positive or negative correlations with FcJAZs in the female flower and peel. KEGG analysis was performed for these genes. During fruit development, the expression trend of JAZs in female flower and peel was mainly clustered into plant hormone signal transduction (87 genes) and plant-pathogen interaction (98 genes). (Fig  S2 B). There are 1,547 genes showed diverse trends, and were mainly enriched in flavonoid biosynthesis, phenylpropanoid biosynthesis, phenylalanine metabolism, and flavone and flavonol biosynthesis pathway (Fig S2 B, C).

Interactions among FcJAZs and between FcJAZs with FcMYC2
JAZ family members can form homodimers or heterodimers through interaction at the ZIM domain to regulate downstream gene expression and exhibit their biological functions [39]. JAZs-JAZs gene expression correlation networks were analyzed based on expression of the genes during the development stage of fig fruit (RNA-seq data was used in this analysis as described in Materials and methods section). The correlation was presented in heat maps. Protein-protein interaction networks were constructed using STRING tool. The top three coexpression pairs were FcJAZ4-FcJAZ6, FcJAZ4-FcJAZ1 and FcJAZ6-FcJAZ1; whereas the top four pairs with high values in the protein-protein interaction network were FcJAZ2-FcJAZ4, FcJAZ4-FcJAZ6, FcJAZ1-FcJAZ4 and FcJAZ1-FcJAZ6. The top three co-expression genes were part of the top four protein-protein interaction network (Fig. 5A). Further analysis showed that, besides FcJAZ2 and FcJAZ8, the other eight members showed high correlation. FcJAZ5 showed the highest correlation with the other five FcJAZs with a co-expression score above 0.5. Notably, FcJAZ5 was located at the center of the coexpression network. Analysis of the protein interaction network showed that although FcJAZ2 had no significant correlation with other members in the network, it had a higher protein interaction coefficient with four FcJAZs (FcJAZ4, FcJAZ5, FcJAZ6 and FcJAZ10). FcJAZ8 did not show interaction with the other FcJAZs, and did not form dimers with the other FcJAZs (Fig. 5B). FcJAZ6 protein contained EAR motif, thus FcJAZ1 and FcJAZ4 may perform their functions by recruiting FcJAZ6 to form heterodimers.
The C-terminal Jas domain of JAZ plays an important role in interaction of JAZ-JAZ or JAZ-MYC2 [16]. NCBI CD-search was used to predict a MYC2 like gene-FcMYC2 from the fig genome. This gene had the conserved domain of bHLH-MYC_N, which mediates interaction with the JAZ repressor, and HLH DNA-binding domain which mediates heterodimerization of JAZs ( Figure S3A) [40]. Sequence alignment showed that the basic domain was highly conserved between FcMYC2 proteins and other MYC2s proteins ( Figure S3B). Phylogenetic analysis showed that FcMYC2 was highly homologous to AtMYC2 (Arabidopsis thaliana), MnMYC2 (Morus notabilis), CsMYC2 (Citrus sinensis) and VvMYC2 (Vitis vinifera) (Fig. 5C). Y2H was used to perform interaction analysis of FcJAZs with FcMYC2. The finding showed that all of the ten combinations used for transformation of yeast grew normally on the plate of DDO (SD/-Leu/-Trp) + X-a-Gal/ + AbA. The clones were further incubated on And the FcJAZs-FcJAZs proteins interaction network were analyzed in the STRING database, the data in the cells with blue color were indicated the required confidence (combined score). b The co-regulatory network of FcJAZ was illustrated by Cytoscape. The augmented icon displayed higher reliability. c Phylogenetic analysis was performed tree that shows the homology of each MYC2 sequences by MEGA 6.0 using the neighbor-joining method with 1000 bootstrap replicates. d Y2H assay were used to analyze the interaction between FcJAZ proteins and transcription factor FcMYC2. A combination of recombinant pGBKT7-MYC2 (bait) and corresponding to pGADT7-FcJAZ constructs (prey) were co-transformed into yeast strain Y2H, and further plated on DDO (SD/-Leu/-Trp) medium as control, or on SD/-Ade/-His/-Leu/-Trp/ + X-a-Gal/ + AbA* medium to test the protein interactions SD/-Ade/-His/-Leu/-Trp/ + X-a-Gal/ + AbA medium. All FcJAZs-FcMYC2 combinations showed growth, whereas the negative control did not grow, implying that all the ten FcJAZs directly interact with FcMYC2 (Fig. 5D).
The results showed that, besides the combinations of FcJAZ7/FcMYC2 and FcJAZ8/FcMYC2, all the rest 8 combinations of FcJAZs-nYFP and cYFP-FcMYC2 could detect a strong fluorescence signal, suggesting they may have directly interaction between them. (Figure S5).
The * indicated significant difference from control according to Student's t test, P < 0.05.

Expression pattern of FcJAZs and FcMYC2 in response to MeJA
To explore the response of FcJAZs and FcMYC2 to JA application, the female flower and peel of fig were treated with MeJA. Gene expression analysis showed that most FcJAZs were upregulated in female flower tissues at 1 h after treatment compared with control treatment (Fig. 6). FcJAZ6 and FcJAZ7 were the most upregulated FcJAZs genes at 1 h after MeJA application, showing 6.96 and 6.35 fold upregulation, respectively. Induced expression of FcJAZ2 and FcJAZ3 were not significant in the female flower. Analysis of the peel showed that the effect of MeJA application was not significant compared with that in the female flower.   [41,42]. Analysis showed that opposite responses of FcJAZs expression profiles between fig female flowers and peels (Fig. 7A,  B). In the early stage at day 2 or 4 after ethyphon treatment, the findings showed that FcJAZ 1 and FcJAZ 3-8 were significantly downregulated (≥ twofold, q = 0.05) either at one of these two time points or at the two time points in the peel. However, analysis did not significant change for FcJAZ2 and FcJAZ 9-10 at these two time points. Analysis of the female follower showed that the main significant modulations were on FcJAZ1, FcJAZ3, FcJAZ4 and FcJAZ8, which were significantly upregulated at day 6 after ethyphon treatment. Notably, FcJAZ6 was the only significantly downregulated gene after ethyphon treatment, whereas other genes were not significantly modulated by ethyphon. Analysis showed less regulation of FcJAZs by GA 3 treatment compared with ethyphon treatment. FcJAZ6 and FcJAZ10 were downregulated after GA 3 application in peels whereas FcJAZ4 and FcJAZ7 were upregulated. Analysis showed that FcJAZ9 was downregulated at day 2 and upregulated at day 4 after GA 3 treatment of female flowers. Notably, FcJAZ10 was downregulated at day 4 after GA 3 application, FcJAZ4 and FcJAZ9 were up regulated after GA 3 treatment, whereas GA 3 treatment did not significantly affect expression of the other genes (Fig. 7C, D).

Discussion
JA is an important hormone widely involved in plant growth, development, and abiotic and biotic stress [1,2,29]. JAZ are the key components of JA signal transduction pathway. JAZ plays a role as a repressor and negatively regulates plant responses to JA. JAZs have been reported in some important agricultural crops, such as grape [43], maize [18], bread wheat [44], tomato [45], strawberry [46] and tea [14]. However, no comprehensive analysis has been conducted on the JAZ protein members in common fig.

Structural characteristics and evolution of the JAZ gene family in fig
There are 10 JAZ family members were isolated in figs.   [47]. The TIFY motif in the core sequence of most FcJAZs comprised TIFYXG, whereas in FcJAZ9 it comprised TIFFXG and IISFXG in FcJAZ7. This finding indicates diversity of the core TIFY motif structure, which may be correlated with different functions of JAZs. The sequence of the Jas domain is relatively conserved among plants, including figs. Jas domain plays an important role in regulation of JA sensitivity, perception and interaction with transcription factors [10]. In addition, FcJAZ6 and FcJAZ8 displayed EAR type motif that may be implicated in binding of TPL to repress JA responses without the presence of NINJA [19]. This indicates that JAZ proteins with EAR motif may play a role as negative feedback regulators to inhibit the jasmonic acid signaling pathway, thus modulating plant growth, development and defense response [19]. Analysis of the subfamilies based on the characteristics of the motif showed that FcJAZ2 and FcJAZ3 belonged to the V subfamily. In addition, FcJAZ2 and FcJAZ3 were mainly found in the early stage of the evolutionary process, which contained the typical ZIM and Jas conserved domains [48]. These two genes may be involved in plant growth, reproduction, senescence, hormone synthesis and defense response [49]. Although FcJAZ1 and FcJAZ6 grouped in subfamily I contained the typical ZIM and Jas conserved domains, the EAR motif was identified at the C-terminus of FcJAZ6. This finding implies that this family members may have appeared in latest stage of the evolution process, and may be involved in some plantspecific biological processes (Fig. 1).
JAZ genes structures are diverse in plants as mainly reflected in the length and number of introns [12]. Shorter introns are more often retained, whereas longer introns are mostly alternately spliced [50]. Analysis did not show significant differences in lengths of intros in the FcJAZs of fig, which may can be attributed to longterm selection pressure. Members of the JAZ gene family mainly contain 0-7 introns in some typical plants, such as rice, wheat and Arabidopsis. The number of introns in FcJAZs were 1 to 8. FcJAZ9 and FcJAZ10 in subfamily IV showed the highest number of introns (8), whereas FcJAZ8 had only 1 intron and FcJAZ7 had two introns, and all belonged to subfamily IV. Notably, JAZ genes with fewer introns within the same subfamily respond faster to environmental stress compared with JAZ genes higher number of introns [44,51]. Therefore, FcJAZ8 may play an important role in fast response to environmental stress. The findings of the present study showed that most FcJAZs were significantly downregulated from stage 3 to 5 in the female flower and from stage 4 to 5 in the fruit peel. These findings were consistent with the results on strawberry that most FaJAZs were downregulated during strawberry development and ripening [54]. WGCNA analysis of FcJAZs showed that FcJAZs were negatively correlated with expansin and polygalacturonase, and were positively correlated with superoxide dismutase, ascorbic acid and ficin synthesis during fig fruit development (Table S3). Similar results were reported in the rubber tree, although the direct correlation between JAZ and ficin has not been verified yet [55]. However, a significant correlation has been reported for the expression of components of COI1-JA-Ile-JAZ complex, such as high expression of HbCOI1 and abundance of HbMYC1 and HbMYC2 in latex cells [56].

Expression profiles of
The cis-element analysis on FcJAZs' promotor regions showed that the predicted elements were correlated with the plant hormones (such as JA, ABA, ethylene and GA) and stress response (including low temperature and drought). Ethylene not only regulates plant growth and response to adversity, but researchers also focus on its regulatory mechanism on the ripening and senescence of climacteric fruits [57][58][59] . Fig fruit has been proved to belong to respiratory climacteric type [60]. During the ripening process, it has obvious respiratory peak and ethylene release peak, showing the late characteristics of respiratory climacteric fruit ripening [61][62][63]. Similar to tomato and banana, their ripening is regulated by ethylene [64]. For GA, it is found that the inhibition of GA synthesis and transcription level of response pathway is the reason for fruit shedding of San Pedro fig [41]. For non-parthenocarpy Smirna fig, exogenous treatment of GA hormone can induce parthenocarpy [65]. In addition, GA 3 plays an important role in regulating fruit size [66,67], and GA3 seems to inhibit the ripening of both climacteric and non-climacteric fruits, such as peach and sweet cherries [66,68].
It was reported that the expression levels of AtJAZ1 in Group I, AtJAZ7 and AtJAZ10 in Group IV, and AtMYC2 were induced under MeJA treatment [69,70]; and in our results, the expression level FcJAZ1 and FcJAZ6 from group I, FcJAZ7-10 from group IV were up-regulated 2.43-6.96 fold after MeJA treatment, and the expression of FcMYC2 was up-regulated 3.24 fold. FcJAZs of the same subfamily showed significant different regulation in response to MeJA treatment between the two analyzed tissues. This finding indicates that FcJAZs may play an important role in JA signaling transduction in female flowers. Most FcJAZs in female flower tissues responded faster to MeJA treatment compared with genes in the peels. FcJAZ6 in subfamily I and FcJAZ7 in subfamily IV were the most significantly upregulated. Upregulation of FcJAZ2 in subfamily V was more significant after treatment of peels with MeJA, whereas the other FcJAZs were only slightly upregulated. Similar results have been reported in Brassica oleracea and Camellia sinensis, that several JAZ genes were induced after MeJA treatment [14,71]. Findings on Ethephon or GA 3 treatment showed that some FcJAZs may participate in synergistic response to these two hormone treatments, whereas some FcJAZs implicated in antagonism response to these two hormone treatments. In female flower tissue, FcJAZ4 responded at the early stage to the GA 3 treatment but responded at the later stage of Ethephon treatment. FcJAZ6 was significantly downregulated by ethephon treatment, whereas it showed upregulation in early stages after GA 3 treatment of female flowers.

Potential function of FcJAZs in fig
JAZ proteins can directly interact with several transcription factors and inhibit their expression. Therefore, they effectively regulate various physiological processes by modulating expression of related transcription factors [72,73]. In JA signal transduction pathway, transcription activator MYC2 is inhibited by binding to transcription inhibitor JAZ, which prevents the transcription of downstream genes [74]. While when JAZ protein is ubiquitinated and degraded, MYC2 is released and the transcription of downstream genes were activated. It is known that MYC2 is involved in regulating the development of leaves, roots, flowers, fruits, seeds and other organs of plants. For example in fruit development, the expression of transcription factor MYC2 was up-regulated in apple fruit treated with MeJA, and MYC2 directly combined with promoters of ACS1 and ACO1 genes and enhanced their expression, thus increasing ethylene synthesis and promoting fruit ripening [57]. MYC2 transcription factor protein plays a central role in the response to jasmonic hormones, and directly regulates downstream response genes. It has been proved that MYC2 can activate many physiological processes during fruit development, and the combination of JAZ and MYC2 would inhibit these downstream activation reactions. After JAZ degradation, MYC2 is released and the JA response is activated. In our study, we found one bHLH transcription factor FcMYC2, and Y2H analysis showed that the FcMYC2 transcription factor directly interacted with all FcJAZs members, and 8 combinations of FcJAZs-nYFP and cYFP-FcMYC2 were further proved they may have directly interaction.
Previous studies report that R2R3 MYB transcription factors can form a MBW complex with bHLH transcription factors and WD-repeat proteins [75]. Moreover, JAZ proteins can bind to bHLH or R2R3 MYB transcription factors of the MBW complex, thus inhibiting formation of the transcription complex. These implies that JAZs play an inhibiting role to the related biological processes, such as anthocyanin synthesis and epidermal hair production in plants [76]. Furthermore, the IIIf family members GL3, EGL3 and TRANSPARENT TESTA8 (TT8) of Arabidopsis interact with 8 AtJAZ proteins implicated in regulation of biosynthesis of flavonoids (anthocyanins and proanthocyanins) [76,77]. In addition, MYB transcription factors play important roles in regulation of anthocyanins synthesis. Further, most MYB transcription factors that respond to JAs belong to the R2R3-MYB family. For example, AtMYB75 promotes accumulation of anthocyanins [76]. In the present study, WGCNA analysis of the FcJAZs during the fig fruit development showed that the modules containing differentially expressed genes closely related to FcJAZs were enriched in flavonoid biosynthesis, phenylpropanoid biosynthesis, flavone and flavonol biosynthesis. This implies that FcJAZs may play an important role in regulation of fig fruit coloring. There are 4 bHLH transcription factors in Arabidopsis directly interact with the JAZ repressor to inhibit JA response. Notably, bHLH3/bHLH13/bHLH14/bHLH17 tetraploid mutants showed enhanced activities in plant defenses [78]. Similar results have been reported in rice, that the OsbHLH148 interacts with OsJAZ protein leading to increased drought tolerance [79]. The findings of the current study showed one bHLH transcription factor FcMYC2. Y2H analysis showed that the FcMYC2 transcription factor directly interacted with all FcJAZs members. Notably, both FcJAZ5 and FcJAZ9 promotor regions contain a MYB binding site which is a regulation site (aC(G/C)GTTA) involved in regulation of flavonoid biosynthetic genes [37].
Findings on protein interaction prediction indicated that FcJAZs may play some other important roles in fig. For example, FcJAZ8 can interact with one R2R3-MYB and three WD40 proteins, which is consistent with findings that MYB24 regulates JA-mediated anther development and filament elongation by interaction with JAZs in Arabidopsis [80]. FcJAZ9 and FcJAZ10 were predicted to interact with WD40 protein of SAP, which controls organ size by targeting PPD protein degradation in Arabidopsis. WRKY40, WRKY57 and TPL interact with FcJAZs and negatively control JAinduced leaf senescence and root meristem development [72,81]. FcJAZ1 interacted with S-phase kinase associated protein 1 (SKP1) to connect with cullin 1 (CUL1), and is implicated in mediating the response to auxin and jasmonic acid [82].

Identification of JAZ family genes in fig (Ficus carica L.)
The first phase of identification of candidate JAZ family members, Hidden Markov Model (HMM) profiles of TIFY domain (Pfam: PF06200) and jas domain (Pfam: PF09425) were retrieved from Pfam (http:// pfam. xfam. org/). These domains were used for protein screening using HMMER 3.2.1 (e-value < 0.01) based on the whole-genome protein data of "Horaishi" and "Dottato" (https:// www. ncbi. nlm. nih. gov/ genom e/? term= Ficus+ carica) [35,36]. The second phase of identification of candidate JAZ family members involved retrieval of JAZ genes of mulberry and Arabidopsis from MorusDB (http:// morus. swu. edu. cn/ index. html/) and Arabidopsis Hormone Database (AHD) v2.0 (http:// ahd. cbi. pku. edu. cn/) [83]. Identification of non-redundant predicted protein sequences in the fig genomes was conducted using BLASTP program with Arabidopsis and mulberry JAZs protein sequences as query sequences [84]. Genes annotated as JAZ proteins were searched based on RNA-seq database annotations, candidate genes were manually merged and repetitive sequences were deleted to obtain a complete family member library. Candidate JAZ proteins were further analyzed for identification of conserved domain using Pfam and NCBI-CDD (https:// www. ncbi. nlm. nih. gov/ Struc ture/ cdd/ wrpsb. cgi) databases.

Analysis of sequence characteristics and phylogenetic tree construction
JAZs protein sequences were analyzed by Clustal X by [85] and submitted to MEME (http:// meme-suite. org/) for motif prediction. MEME motif sequence length was set to 20-100 aa, maximum number of discoveries was set to 4, and other parameters were set to default values. Alignments were visualized and edited with Jalview tool [86]. To explore intron-exon organization of the JAZs genes, coding sequences with corresponding genomic sequences were aligned. The results retrieved using the Gene Structure Display Server (http:// gsds. cbi. pku. edu. cn/ index. php). The upstream 2000 bp regions of FcJAZs were considered to have the full length promoters. Regulatory cis-elements were predicted using PlantCARE tool (https:// bioin forma tics. psb. ugent. be/ webto ols/ plant care/ html/) and PlantPAN tool (https:// plant pan. itps. ncku. edu. tw/). The theoretical isoelectric point (pI) and molecular weight (Mw) of each JAZ protein were determined using the 'Compute pI/Mw tool' in ExPASy (https:// web. expasy. org/ compu te_ pi/). Subcellular locations were predicted using WoLFPSORT (https:// psort. hgc. jp/ cgi-bin/ runps ort. pl). Chromosomal position of the FcJAZ family genes were obtained based on the genomic sequence information and maps were constructed using Map Gene 2 Chromosome V2 tool (http:// mg2c. iask. in/ mg2c_ v2.0/). Protein sequences were aligned using ClustalW program and phylogenetic trees were constructed using the Neighbor-Joining (NJ) and method in MEGA6 (http:// www. megas oftwa re. net/) and the bootstrap value was set to 1000. The phylogenetic tree of FcJAZ genes and two species (AtJAZ and MnJAZ) was also constructed using the Maximum Likelihood (ML) method in MEGA 6, and the Jones-Taylor-Thornton (JTT) was selected as the model.
The non-synonymous replacement rate (Ka) and synonymous replacement rate (Ks) of the replicated gene pairs were calculated using KaKs_Calculator 2.0, and environmental selection pressure was analyzed by Ka/Ks ratio.

Analysis of interactive signaling network
Protein interaction network (interactome) for JAZ-JAZ proteins was constructed using STRING tool (http:// string-db. org/). After analysis of the relationship between proteins using the protein interaction network, networkX was using in Python environment to visualize the network of genes of interest. Protein-protein interaction networks were built with an interaction score of the highest confidence (0.900); the minimum required interaction score was 0.150 (was considered as low confidence).

Yeast two-hybrid assay and bimolecular fluorescence complementation assay
The full-length sequences of all FcJAZs were cloned into the pGAD-T7-AD vector, whereas the full-length sequences of FcMYC2 and FcbHLH33 were cloned into the pGBK-BD vector following a protocol by Matchmaker ™ Gold Yeast Two-Hybrid system (Clontech, Mountain View, CA, USA). Primers used for Y2H are listed in Table S1. Yeast AD/BD co-transformations were performed using lithium-acetate method [87]. Interactions between proteins were determined by growth on the DDO medium (SD/-Trp/-Leu/) at 30 °C. After growing the plaque, they were transferred to the QDO medium (SD/-Trp/-Leu/-His/-Ade) containing 5-bromo-4-chloro-3-indoxyl α-D-galactoside (X-α-Gal) and incubated for three days at 30 °C.
To co-transform the paired combination of FcJAZs and FcMYC2, the vectors used are pSPYNE-35S and pSPYCE-35S, and the primers used are listed in Table S1. After infection, the tobacco was cultured in the dark for 24 h, and then grown for 1 day under the conditions of 16 h of light, 8 h of darkness and 25 ℃. Cut the infected leaves and place them on a microscope slide. Use a confocal microscope to observe the fluorescence of YFP (trigger wavelength is 514 nm, radiation wavelength is 527 nm). BiFC experiments were repeated at least two times in independent experiments.

Plant growth conditions and treatments
Common fig cv. Zibao fruits were harvested from Shandong province, P. R. China. Six stages of autumn fruit were sampled for gene expression analysis based on the characteristics of fruit development. The stages were marked as stage 1-6, whereby stage 2, 3, and 4 were the early, middle and late stages of phase II (slow growth period), stage 1 represented phase I (fast growth period), stage 5 and 6 represented phase III (fast growth period). Sixty fruits were randomly selected at each stage and 20 fruits were used as a biological replicate. The peel and female flower tissue were separated at the sampling time. Fresh samples were quick-frozen with liquid nitrogen and stored at -80 °C for subsequent experiments.
The peel and female flower tissue of the stage 3 fruit were obtained for MeJA treatment. The separated tissues were immersed in 200 μM MeJA (sample: MeJA solution = 1:3, w/w), shaking was performed at 25 °C at 30 rpm under dark conditions. The control was treated with the dissolving solution (1% ethanol solution) without MeJA and shaken under the same conditions. Untreated samples were collected and labeled as 0 h. After 1 h, 3 h, 9 h, 24 h of shaking, the solution on the surface of the samples was absorbed using a filter paper, then frozen with liquid nitrogen and stored at -80 °C for subsequent experiments. were analyzed also analyzed. Transcriptome data were obtained from findings of a related research by our team (Accession No. SRP113799 and No. PRJNA606407) [41,42]. For RNA-seq analysis, fragments per kb per million (FPKM) method was used to normalize and determine gene expression levels of JAZ and JAs-related genes in different treatments after hormone application (Wang et al., 2010).

Co-expression network analysis during fig fruit development
Weighted gene co-expression network analysis (WGCNA) [88] was used to analyze gene co-expression network during the fig fruit development. The threshold power was set to nine. The network included phenotypic data, based on filtered data and sample size. Correlations were determined using Pearson's correlation analysis. The common expression module was identified, and the minimum connection value, module merging standard, and minimum module size were set to 0.5, 0.25, and 30, respectively. After identifying the co-expression modules, the modules were associated with each sample type. DESeq2 was used to identify differentially expressed genes (DEGs) (|log 2 (Fold Change)| > 1, p value < 0.05) [89]. Gene ontology analysis was performed using Kyoto Encyclopedia of Genes and Genomes (KEGG) database [90].

Quantitative RT-PCR (qRT-PCR)
Fig fruit RNA was extracted with the CTAB method [91]. RNA concentration and purity were determined by NanoDrop 2,000 (Thermo Scientific, USA). 1% agarose gel electrophoresis was performed to determine RNA integrity, then the RNA concentration was adjusted to a consistent level (1,000 ng). First strand cDNA was synthesized using cDNA Synthesis Kit (Takara, Kyoto, Japan). Ultra SYBR Mix kit (Takara, Kyoto, Japan.) was used to perform qRT-PCR using the ABI Q6 real-time PCR system (Applied Biosystem Inc., MA, USA). The reaction volume of PCR was 30 ml which comprised 3 L diluted cDNA, 3 L of each primer (2 mM), 15 L of SYBR Green Realtime PCR Master Mix Plus and 3 l Plus solution (Toyobo). PCR amplification was initiated by 1 min at 95 °C, followed by 40 cycles of 95 °C for 15 s, 58 °C for 15 s and 72 °C for 45 s. All experiments were performed in triplicate using diluted cDNAs. Gene-specific primers listed in Table S1 were used for qRT-PCR. Relative quantitative analysis of data was performed by the 2 −ΔΔCT method with reference genes Fcactin and 18S-RNA. Three technical replicates were carried out for each sample to ensure reproducibility and reliability. Statistical analysis of variance (ANOVA) followed by Duncan's new multiple range test and Student's t test were performed with SPSS Version 16.0 (Chicago, IL, USA). The significance level was set to P < 0.05.