Identification and phylogenesis of the JAZ subfamily in Brassica napus
The JAZ subfamily contains 12 members (JAZ1 to JAZ12) in Arabidopsis, which are divided into five groups [49]. Based on this, a total of 56, 28 and 31 orthologues were identified from in the reference genomes of B. napus as well as its two progenitors B. rapa and B. oleracea, respectively (Table 1). The characteristics of these genes, including the length of gene, coding sequence, and amino acid, the molecular weight of protein, isoelectric point, and the subcellular localization were analyzed. The length of 56 identified BnJAZ genes in B. napus showed a wide range from 556 to 9131 bp, indicating its large variation. The amino acid length of 56 BnJAZs ranged from 116 (BnJAZ40) to 564 (BnJAZ53), with molecular weights of 13.13 kDa to 62.22 kDa. The predicted isoelectric points of 56 BnJAZ proteins ranged from 5 (BnJAZ4) to 10 (BnJAZ56). Except for five in Chlo, four in cyto, one in E.R. golgi and mito, the other 44 BnJAZ proteins were predicted to be located in the nucleus, indicating that most of them should be transcription factors.
To elucidate the evolution and phylogenetic relationships among the JAZ gene family, all of the 127 JAZ proteins (including 56 in B. napus, 28 in B. rapa, 31 in B. oleracea and 12 in A. thaliana) were used to construct a phylogenetic tree (Fig. 1 and Table S2). All the JAZs were distinctly divided into five groups (I, II, III, IV, and V) the same with Arabidopsis, which contained12, 23, 49, 21 and 22 members, respectively. It should be noted that nearly half were classified into Group III, which contained 22 BnJAZs (half were from the A and C subgenome), 11 BraJAZs, 12 BolJAZs and 4 AtJAZs. Except for Group I, the phylogenic relationships among JAZ orthologues were consistent with the evolutionary relationship of their species origin.
Gene structure and motif composition of BnJAZ subfamily
To explore the possible structural evolution of JAZs in B. napus, all of the 56 BnJAZs were analyzed for their gene structure, protein motif composition, and cis-elements (Fig. 2 and Table S3A).
A total of 69 to 193 cis-elements were recognized in the 2 kb upstream regulatory sequence of the 56 BnJAZs genes, which were divided into 1 to 33 types (Fig. 2B and Table S3B). Summary statistics of cis-elements number for the different types showed that “core promoter element-TATA box” has the largest number, followed by “Light responsive element”, “common cis-acting element-CAAT box” and “Short function”; “abscisic acid responsiveness”, “MeJA responsive element” and “the anaerobic induction” accounted for a considerable number; whereas the other 26 types were few. It should be noted that the main cis-elements types were involved in phytohormone, biotic or abiotic stress. For the identified 25 types of light responsiveness element, G box and Box4 had the largest number, followed by GT1, TCT, AE, GATA, MRE, I-box, ATCT, and TCCC, whereas the other 15 types were few. Additionally, several cis-elements were responsive to phytohormones, including abscisic acid (e.g., ABRE), auxin (e.g., TGA-element), gibberellin (e.g., GARE-motif, P-box, TATC-box), MeJA (e.g., CGTCA-motif, TGACG-motif), and salicylic acid (e.g., TCA-element) (Table S3A). In addition, some cis-elements were responsive to abiotic stress, including low temperature (LTR), drought (MBS), defense and stress (TC-rich repeats), and the anaerobic (ARE) response. Unexpectedly, there was a large difference in the type and number of cis-elements among the different groups and even within the same group.
All BnJAZ genes possessed two to nine exons and the corresponding one to eight introns, with the means of 5.27 and 4.27, respectively (Table 1). Only one Gene (BnJAZ40) had only two exons and one intron.
All BnJAZ proteins were subjected to the MEME motif analysis, and a total of fifteen conserved motifs were identified (Fig. 2C and Table S4). It should be noted that all of the 56 BnJAZ proteins processed only two common motifs, which were motif 1 and motif 2 (CCT_2 and TIFY domain), suggesting that they were the core domains for the JAZ subfamily. As expected, the BnJAZ members belonging to the same groups displayed a similar motif composition. For example, motif 3 was unique to group III, whereas motif 14 was particular to group II. The clustered BnJAZ pairs, i.e. BnJAZ18/49, BnJAZ9/35, showed the same motif distribution. Overall, the BnJAZ genes within the same group generally have the similar/same gene structures motif compositions, strongly supporting the reliability of the group classifications.
Syntenic analysis of BnJAZ genes
To explore the evolution of JAZ subfamily from the comment ancestor, their syntenic relationship was analyzed between Arabidopsis and Brassica genus (Table S5). AtJAZ4 and AtJAZ11 had no orthologues in Brassica and nine BnJAZs (BnJAZ1, BnJAZ2, BnJAZ19, BnJAZ20, BnJAZ28, BnJAZ29, BnJAZ3, BnJAZ50, BnJAZ53) have no orthologues in Arabidopsis, indicating the regeneration or loss of these genes during evolution. Interestingly, AtJAZ2, AtJAZ5, AtJAZ7, AtJAZ9, AtJAZ10 have three orthologues in both B. rapa and B. oleracea, and expected six orthologues in B. napus, in line with the previously proved triplication from the common ancestor [54, 55]. Whereas, AtJAZ3 and AtJAZ6 have two orthologues in both B. rapa and B. oleracea and expected four copies in B. napus, supporting the proposed hypothesis that triplication was usually followed by diploidization [55].
To elucidate the evolutionary constrictions on the BnJAZ family, the Ka, Ks, and their ratio were calculated for the syntenic gene-pairs between B. napus and Arabidopsis/B. rapa/B. oleracea (Table S6). The result showed that most (91.0%) of the synthetic BnJAZ gene pairs had a Ka/Ks ratio of < 1, indicating purifying selection pressure during the evolution. Whereas, the other (9.0%) had a Ka/Ks ratio of > 1, suggesting accelerated evolution under positive selection (Table S6). For example, the three syntenic gene pairs (BraJAZ2/BnJAZ2; BraJAZ19/BnJAZ19; BolJAZ3/BnJAZ29) showed the smallest Ka/Ks value, all of which lost their syntenic relationship with AtJAZ11 and AtJAZ12 in the Group Ⅴ.
Briefly, the 56 BnJAZ genes were unevenly distributed on the 17 chromosomes (except for A04 and C07), of the 27 and 29 ones were on the A and C subgenomes, respectively. Chromosome A02 contained the largest number of BnJAZs (6), followed by A07, A08, C02, C06 and C08 (5), whereas some chromosomes (e.g. A03, A05, C04) have only one gene. The length of chromosomes was not correlated with the number of JAZ genes. Besides, 15 pairs of paralogous genes were recognized between A and C sub-genomes, except for A03, A04, A09, C04 and C07 chromosomes (Fig. 3). It should be noted that all of the 15 pairs of paralogous genes were located in the synthetic regions, which should originate from the same genomic segments of the common ancestor. Highly accordant with this, no tandem repeat paralogues were found for them, therefore they might be originated from whole-genome duplication (WGD) rather than gene proliferation. The chromosomal locations of almost all JAZ genes in B. napus were similar to those of their orthologues in B. rapa and B. oleracea, with a few exceptions. For example, the BolC07g051380.2J in B. oleracea had no orthologues in B. napus. Interestingly, there was no JZA gene distributing on chromosome A04 in B. rapa and B. napus.
Expression profiling of the BnJAZ genes in main organs/tissues
To predict possible functions through overlapping expression patterns for the BnJAZ genes, the expression levels of 11 representative BnJAZ genes from five groups were detected in eight different tissues and organs (including root, stem, leaf, bud, petal, stamen, stigma and silique) by real-time quantitative RT-PCR (Fig. 4). The correlation between the expression levels of 11 BnJAZ genes in the eight organs/tissues was analyzed. The results showed that high correlations were observed among bud, petal and stamen, and between silique and stigma. In addition, similar expression pattern was observed between several pairs of genes, such as BnJAZ7/17, BnJAZ44/49, BnJAZ3/52, BnJAZ44/54, BnJAZ7/27. Generally, the expression levels of these genes showed great variations in different tissues, which were high in stem, followed by leaf, stamen and petal, and relatively low in other tissues. In addition, BnaJAZ17, BnaJAZ7, BnaJAZ37, and BnaJAZ27 showed relatively high expression, whereas the expression levels of other five genes were low. Interestingly, all of these genes showed a very low expression in roots, except for BnJAZ24, indicating its functional differentiation. Overall, the diverse expression patterns of JAZ genes in distinct tissues and organs suggested that these members might play rich functions.
Expression pattern of BnJAZ genes in response to different abiotic stresses and hormonal treatments
Based on the cis-elements analysis, these BnJAZ genes were predicted to be involved in the response to abiotic stresses and phytohormones. To further confirm this, the expression of the above-mentioned 11 representative BnJAZ genes from five groups were detected after different treatments (PEG, NaCl, cold, waterlogging, ABA, GA, MeJA and IAA) (Fig. 5). Generally, all of the 11 BnJAZ genes were significantly induced/repressed by multiple treatments, in line with their functional prediction by cis-elements analysis. Specially, among the eight treatments, MeJA showed the largest effects on the expression of these BnJAZ genes, in line with their most important core function domain of jas. Interestingly, MeJA and waterlogging treatment tended to induce the expression of these BnJAZ genes except for BnJAZ24, while NaCl likely repressed their expression. ABA and GA showed similar effects on the expression of these genes, which repressed the expression of eight BnJAZ genes in Group Ⅰ to Ⅲ except for BnJAZ 27, BnJAZ 32 and BnJAZ 54. Under the treatment of IAA, cold and PEG, the expression of 11 BnJAZ genes could be divided into four patterns, i.e., a gradual rise or decline, first rise then descend, or first fall after rise.
Obviously, none of the 11 genes showed consistent performance in expression patterns under all of the eight treatments. Especially, several genes showed contrast/opposite expression patterns between the different treatments. For example, the expression of BnJAZ24 was remarkably up-regulated after 4h of the IAA treatment, while down-regulated under the treatments of the other three phytohormones ABA, GA and MeJA. For instance, BnJAZ37 was significantly repressed by ABA and GA, whereas displayed substantially higher expression at 1–8 h under MeJA treatment. Under the abiotic stimuli, BnJAZ37 was rapidly repressed by NaCl, cold and PEG stress, however, was up-regulated under waterlogging stress. BnJAZ52 was significantly repressed by ABA, GA, IAA, PEG and NaCl treatments, while induced under MeJA, cold and waterlogging.
Overexpression of BnJAZ52 ( BnC08.JAZ1-1 ) increased seed weight in Arabidopsis
Under the hormone treatment, BnJAZ52 was up-regulated by MeJA, whereas repressed by the GA, ABA and IAA, which suggested that BnJAZ52 particularly takes part in the JA signal pathway, the same with their orthologue AtJAZ1 in Arabidopsis. In addition, the previous study showed that JA signaling pathway was involved in seed development and size [12, 37, 56, 57]. More importantly, the expression of BnJAZ52 in large-seed lines was about twice than small-seed lines in a previously reported RNA-seq study using the bulked seeds at 25 days after flowering (Li et al., 2019). These results highly suggested that BnJAZ52 might also have a role in regulating seed size. Therefore, BnJAZ52 was selected for further experimental exploration. As BnJAZ52 is homologous to JAZ1 in Arabidopsis that has two orthologues on the C08 chromosome of B. napus, so it’s named as BnC08.JAZ1-1 hereafter.
To understand the function of BnC08.JAZ1-1, its CDS sequence was cloned from Zhongshuang11 and over-expressed in A. thaliana, and the phenotypes of transgenic lines were investigated. During the seedling stage, BnC08.JAZ1-1-OE plants flowered about three days earlier than the control (Fig. 6). During the mature stage, relative to CK, the seed weight of nine independent transgenic lines in the T3 generation increased significantly, with the proportions from 17.4–27.2%. Whereas, there was no significant difference between the SNPS (seed number per silique) of the transgenic lines and CK. These results suggested that BnC08.JAZ1-1 was a vital positive regulator that promotes plant growth and development, especially for flowering time and seed weight.
Expression pattern and subcellular location localization of BnC08.JAZ1-1
We performed staining on the selected pBnC08.JAZ1-1: GUS transgenic lines, and found its signal was mainly found in vascular bundle and young flowers including petals, pistil, stamens and developing ovules, but not in the stem, leaf and mature silique and seed (Fig. 7A-C). Therefore, the expression of BnC08.JAZ1-1 was gradually decreased as the development and maturation of flowers. It was speculated that the expression level of BnC08.JAZ1-1 gene might be related to the degree of organ development.
The fusion GFP expression vectors BnC08.JAZ1-1GFP-PD1301S and GFP-PD1301S were transferred into tobacco cells by the transient expression method of tobacco, and the observation was performed by laser confocal microscopy (Fig. 7D). The green fluorescence signal was distributed in both the cell membrane and the nucleus, and the 35S: GFP-BnC08.JAZ1-1 green fluorescence signal was completely coincident with the DAPI nuclear dye, indicating that the BnC08.JAZ1-1 protein was localized in the nucleus, in line with the prediction (Table 1).
Transcriptome analysis of BnC08.JAZ1-1 overexpressing Arabidopsis seeds
The summary statistics of RAN-seq data showed that it meets the standards and the followed experimental requirements (Table S7). Using Hisat2 software, clean reads were efficiently and accurately mapped to genes. Finally, the FPKM method was adopted to calculate the relative expression of each gene, and a total of 25397 expressed genes were detected. At the threshold of P < 0.05 and | log_2 fold change | ≥1, a total of 582 differentially expressed genes (DEGs) were identified, including 123 (21%) up-regulated genes, 459 (79%) down-regulated genes.
The GO enrichment analysis was performed on all the 582 DEGs using Classification Super Viewer (Fig. 8A). From the perspective of biological processes, differentially expressed genes were most abundant in defense response, followed by transcription and signal transduction. This is understandable because the seed filling process involves cell proliferation and the expression of a series of genes, which also depend on related signal transduction processes in the cell. In molecular function, the most enriched category was hydrolytic enzyme activity, followed by transcription factor activity, and other enzyme activities and these were closely related to gene transcription regulation and seed filling. In cellular components, extracellular was most enriched, followed by cell wall nucleus, plasm membrane, chloroplast, and mitochrondria.
In addition, pathway enrichment analysis was conducted on all the 582 DEGs (Fig. 8B and Table S8). They were mainly enriched in ubiquitin-mediated proteolysis, plant-pathogen interaction, circadian rhythm, stilbene compounds, and biosynthesis of cellulosic (Stilbenoid, diarylheptanoid, and gingerol biosynthesis), and glycerophospholipid metabolism. During the development of seeds, the biosynthesis of stilbene compounds and inositol provides a material basis for the synthesis of many metabolites. Many previous studies have shown that the ubiquitin-proteasome pathway regulates seed size [58–62]. In the current study, four DEGs (AT2G03190, AT3G21830, AT4G33270, AT2G25700, AT5G22920) were enriched in ubiquitin-mediated proteolysis, it was speculated that the increased seed weight of BnC08.JAZ1-1 over-expression lines may be through ubiquitin-proteasome pathways. In addition, four DEGs (AT3G03530, AT3G03540, AT2G44810, AT4G01950) are enriched in the phospholipid metabolism pathway, which was understandable as seed development is also a process of continuous accumulation and storage of lipids. Moreover, of all 582 DEGs, three were known genes in regulating seed weight, including CYP78A9 (AT3G61880), LEC2 (AT1G28300), and ASPGB1 (AT3G16150) [63, 64].