Mechanism of zju‐miR156c‐mediated network in regulating witches' broom symptom of Chinese jujube Authors: Yunjie Wang, Zhi Luo, Xuan Zhao, Hongqiang Sun, Jiaxin Liu, Dongfeng Zhang, Haonan Cao, Changfeng Ai, Lihu Wang, Li Dai, Mengjun Liu, Lixin Wang, Zhiguo Liu Published: 2024 ABSTRACT AbstractJujube witches' broom, caused by phytoplasma, is a destructive disease of Chinese jujube. Studies have shown that zju‐miR156s play an important role in phytoplasma infection in jujube, but the regulatory mechanism between zju‐miR156c and witches' broom remains unexplored. In the current study, miRNA‐seq and gene expression analysis showed that zju‐miR156c was more highly induced in infected jujube plants than the other miRNAs and its target gene was ZjSPL3. In addition, the expression levels of thymidylate kinase gene (TMK JWB ) and secreted jujube protein (SJP1JWB) in diseased materials were higher than those in healthy controls. The expression level of zju‐miR156c was significantly upregulated, while ZjSPL3 was sharply downregulated and the content of cytokinin (CTK) significantly increased. Overexpression of zju‐miR156c in Arabidopsis significantly reduced the expression of AtSPL10 (homologous gene of ZjSPL3) but increased the content of CTK, and the transgenic plants exhibited witches' broom symptoms. In addition, yeast two‐hybrid and co‐immunoprecipitation assays confirmed that SJP1JWB interacted with ZjERF18. Yeast one‐hybrid analysis showed that ZjERF18 could interact with the promoter of zju‐MIR156c. In conclusion, our results demonstrated a novel pathogenic module of ZjERF18‐zju‐miR156c‐ZjSPL3‐CTK has an important function in the formation of witches' broom caused by SJP1JWB. INTRODUCTION Chinese jujube (Ziziphus jujuba), native to China, is one of the most economically valuable species compared to other genotypes of Ziziphus. Jujube has been introduced to more than 50 countries around the world (Liu et al., 2020). Jujube witches' broom (JWB) is a devastating disease for Chinese jujube caused by phytoplasma. It has the characteristics of a strong infection, difficult to cure and a high fatality rate. When jujubes were infected by phytoplasma, they would die about 3 years later, with a fatality rate of almost 100% (Liu et al., 2020). Previous studies showed that phytoplasma obligately parasitized the phloem of plants and secreted effectors into the host, such as effector SAP05, which causes arbuscular symptoms in Arabidopsis thaliana plants by mediating the simultaneous degradation of both SPL and GATA (Huang et al., 2021). In addition, more effectors have been identified in different plant species, including SAP11AYWB (secreted AY‐WB protein 11), TENGU, Phyllogen (phyllody inducing gene) PHYL1, SAP54, Pawb‐SAP54, Pawb20, SWP1, SWP12, SJP3 and Zaofeng3 (Chen et al., 2024). The effectors could not only destroy the defence system but also regulate the activity of related transcription factors in the branches and leaves of host plants and disrupt the balance of hormone metabolism, which finally caused the host to display symptoms of witches' broom, dwarfism and phyllody and affected the growth and development of host plants (Jiang et al., 2019; Li et al., 2019; Sugio, MacLean, et al., 2011). Witches' broom is a typical symptom with continuous axillary bud germination due to the infection of phytoplasma, accompanied by smaller leaves, thinner stems and shortened internodes. Up to now, several phytoplasma protein effectors that cause jujube witches' broom symptom have been identified, including SJP1, SJP2 and Zaofeng 6 (Chen et al., 2022; Deng et al., 2021; Zhou et al., 2021). However, the molecular mechanism of how the effectors regulate the corresponding miRNAs was ambiguous. The symptom of witches' broom is similar to the appearance of increased and clustered branches or tillers. miR156 is one of the major and highly conserved members of micro RNAs (miRNAs), which is involved in the formation of plant branches or tillers (Wang et al., 2023). In terms of branching, it was reported that A. thaliana ath‐miR156 positively mediates the formation of branching by regulating the expression of AtSPL9/15 and AtSPL2/10/11 (Tian et al., 2014). Overexpression of the sco‐miR156 in tomato (Solanum lycopersicum) causes a 2‐ to 3‐fold increase in the number of branches compared with the wild type (Cui et al., 2020). In terms of tiller regulation, pvi‐miR156 overexpression in switchgrass (Panicum virgatum) causes increased tiller number (Chuck et al., 2011). In wheat (Triticum aestivum), tae‐miR156 promotes the increase of tiller number by inhibiting the expression of TaSPL3 and TaSPL17 (Liu et al., 2017). Overexpression of osa‐miR156b and osa‐miR156h inhibits the expression of OsSPLs, releases the inhibition of OsSPLs on OsTB1 (TEOSINTE BRANCHED 1), and leads to an increase of rice tillers (Lu et al., 2013; Wang et al., 2015; Xie et al., 2006). These results confirmed that miR156s are involved in the generation and development of branches/tillers in plants. In addition, it has been reported that miR156s of jujube and Paulownia fortunei are significantly upregulated in response to phytoplasma infection (Cao et al., 2018; Fan et al., 2015; Shao et al., 2016). However, whether miR156s are responsible for the regulation of witches' broom symptom of jujube and what the regulating mechanism behind it is has not been clarified. To answer these questions, firstly, miRNA‐seq analysis of Dongzao healthy tissue‐culture plants (DH) and phytoplasma‐infected Dongzao tissue‐culture plants (DHF) was conducted to identify the phytoplasma‐responding miRNAs. Then, the expression patterns of phytoplasma genes thymidylate kinase gene (TMK JWB ), phytoplasma protein effector secreted jujube protein (SJP1JWB) and zju‐miR156c‐mediated genes were tested in phytoplasma‐infected jujube materials, and the hormone content was measured. The function of zju‐miR156c in regulating plant branching was verified by genetic transformation in A. thaliana and sour jujube seedlings. Finally, the transcription factors that can interact with the promoter of zju‐MIR156c were identified, and the relationship between SJP1JWB and the transcription factors was revealed. These results can enrich the understanding of the molecular mechanism of how SJP1JWB mediates the zju‐miR156c regulating networks in the development of witches' broom symptoms and provide new ideas for comprehensive control of JWB and molecular design breeding for phytoplasma resistance. RESULTS miRNA‐seq analysis in DH and DHF plants To explore the function of miRNAs in response to phytoplasma infection, DH and DHF plants were selected for miRNA‐seq analysis. A total of 108.13 Mb of raw reads were obtained by miRNA‐seq analysis, and in each sample, the raw reads reached above 21.74 Mb. Clean reads data reached 103.52 Mb after removing reductants and adapter sequences filtering, and the error rates were 0.0226%–0.0221% with 98.97% and 96.72% for bases scoring Q20 and Q30 in all samples. Finally, for each sample, the valid clean reads came to 21.19–27.60 Mb (Data S1). In addition, the number of 18–32 nucleotide (nt) sequences was 16.55–27.60 Mb when the small RNA sequences were aligned to the jujube reference genome. At least 7.05 Mb of valid data were assigned to the corresponding chromosomes; the most distributed pseudochromosome was NW_0154544482.1, followed by NC_029680.1 (Data S2a). The further chromosome density analysis confirmed that the largest number of positive strands were in NW_015454482.1 and the highest number was in NC_029680.1 (Data S2b). To further confirm the reliability of miRNA sequencing data, the Pearson correlation analysis was performed. The value of R 2 reached 0.873 among all the samples, indicating that the correlation of the four samples was high. In addition, the value of R 2 reached 0.991 between DHF_1 and DHF_2, while it reached 0.927 between DH_1 and DH_2 (Data S2c). The high Pearson correlation analysis indicated the high reliability of the miRNA sequencing data. Moreover, miRNA sequencing of the four samples yielded 180, 173, 161 and 175 miRNAs, respectively (Data S2d), which provide enough data to explore the main phytoplasma‐responding miRNAs. Identification of differentially expressed miRNAs analysis As different numbers of miRNAs existed in healthy and phytoplasma‐infected plants, the Venn diagram analysis showed that 153 miRNAs existed in DH and DHF plants (Figure 1a). Among them, the differentially expressed miRNAs were identified, including 27 upregulated miRNAs and 34 downregulated miRNAs in the compared group (DH_vs_DHF) (Figure 1b). The upregulated genes mainly included miR156, miR157, miR390, miR398 and miR399 families. Among them, the expression level of miR156 and miR157 was higher compared to the other miRNAs in DHF plants, indicating miR156 and miR157 might be the main differentially expressed miRNAs in response to phytoplasma infection (Figure 1c). All the identified miR156 family members had the same mature sequence, which was the same as miR156c reported by Shao et al. (2016); thus, we designated the identified miR156 as miR156c. To further confirm whether miR156c or miR157 functioned more importantly in phytoplasma infection, the expression levels of them in Pozao healthy plants (PZ‐H) and phytoplasma‐infected Pozao plants (PZ‐F) were examined. As shown in Figure 1d, the expression trend of them was consistent with both of them being significantly induced under phytoplasma infection, but the expression level of zju‐miR156c was significantly higher than that of zju‐miR157, which further indicated zju‐miR156c might be the more significant one in response to phytoplasma infection (Figure 1d). Identification of differentially expressed miRNAs analysis. (a) Venn diagram analysis of miRNAs in Dongzao healthy tissue‐culture plants (DH) and phytoplasma‐infected Dongzao tissue‐culture plants (DHF) plants. (b) Up‐ and downregulated differentially expressed miRNAs in the DH_vs_DHF group. (c) Heatmap of the expression of five miRNAs in DH and DHF plants. (d) The relative expression level of zju‐miR156 and zju‐miR157 in Pozao healthy plants (PZ‐H) and diseased plants (PZ‐F). (e) zju‐miR156c target gene prediction analysis. (f) The expression analysis of target gene ZjSPLs in the transcriptome data of Pozao (PZ) healthy (H) and diseased (D) plants at three growth stages (S1, S2 and S3). ***Significant at p < 0.001. To further reveal the molecular mechanism of how zju‐miR156c functions in response to phytoplasma infection, the target genes for zju‐miR156c were predicted (Figure 1e). A total of 18 SPL members (ZjSPLs1–ZjSPL18) were identified by Shao et al. (2017), and they functioned importantly in the process of phytoplasma infection. Thus, the expression levels of ZjSPL genes were analysed in the transcriptome data of healthy plants (PZ‐H) and diseased plants (PZ‐D) at three developmental stages (S1 to S3) (Wang et al., 2022), the results showed that most of the ZjSPL genes, such as ZjSPL3/4/15/17, were highly expressed in healthy plants but not in diseased plants. Compared to ZjSPL4/15/17, the expression level of ZjSPL3 was always higher in the healthy plants at stages S1 and S2. Moreover, the expression level of ZjSPL10/13 was induced in the healthy plants at stages S1 and S2, but they were also induced in diseased plants at stage S2. Thus, all the related results indicated that ZjSPL3 could be the main target gene of zju‐miR156c in the regulation of phytoplasma infection (Figure 1f). The prediction analysis of transcription factors that might regulate the expression of zju‐miR156c The key factor driving the expression of zju‐miR156s is transcription factors; thus, analysis of the transcription factors binding to the promoter of the zju‐miR156c precursor sequence could help to understand the biological functions of zju‐miR156c. Fifteen types of transcription factors (Figure 2), including ZjMYB, ZjWRKY, ZjNAC, ZjWOX (WUSCHEL‐related homeobox), ZjbHLH (basic helix‐loop‐helix) and ZjbZIP (basic leucine) zipper, ZjAP2 (APETALA2) and ZjARF (auxin response factor), were identified. Among them, the transcription factors that respond to hormone regulation are brassinolide signalling transcription factor ZjBES1 (BRI1‐ethyl methyl sulfon‐SUPPRESSOR), ethylene signalling transcription factor ZjEIN3 (Ethylene insensitive 3) and auxin response factor ZjARF. The zju‐miR156c might be regulated by transcription factors such as ZjMYB, ZjAP2, ZjbHLH, ZjbZIP and ZjGATA. In addition, ZjSPL (Squamosa promoter‐binding‐like proteins) transcription factors participated in the transcriptional regulation of zju‐miR156c. Moreover, ZjC2H2, ZjYABBY, ZjWOX and other transcription factors involved in morphogenesis development were found to be involved in the transcriptional regulation of zju‐miR156c. The transcription factor prediction analysis of zju‐MIR156c. Expression analysis of zju‐miR156c‐regulated genes and detection of hormones in jujube plants with witches' broom miR156s participate in the development of tillering and branching of plants by regulating the expression of target genes (Wang et al., 2023). The witches' broom symptom of jujube is caused by the continuous germination of axillary buds. Thus, the expression profiles of zju‐miR156c‐regulated genes and the hormone content were analysed in witches' brooms of jujube. As shown in Figure 3a–c, the axillary buds of the plants infected by phytoplasma continued germinating, eventually leading to the formation of shoot proliferation, while the axillary buds of the healthy plants did not show this symptom. By analysing the expression levels of phytoplasma gene TMK JWB and witches' broom protein effector SJP1JWB, it was found that the expression levels of TMK JWB and SJP1JWB in infected plants were significantly higher than that in healthy control (Figure 3d). Furthermore, the target gene of zju‐miR156c and the expression level of zju‐miR156c‐regulated genes were analysed. As shown in Figure 3e, the target gene of the zju‐miR156c was found to be ZjSPL3 by sequencing comparison analysis, and luciferase assay further confirmed that ZjSPL3 was the target gene of zju‐miR156c. The expression of zju‐miR156c in Dongzao plants with phytoplasma infection (DZ‐D) was about 90‐fold higher than that in healthy controls (DZ‐H). However, the expression of ZjSPL3 in DZ‐D plants was significantly down‐regulated (more than 2‐fold lower) compared to healthy controls (Figure 3f). The phenotype and expression profiles of zju‐miR156c‐regulated genes and hormone contents in jujube materials with witches' broom. (a) Phenotype of tissue culture Dongzao healthy plants (DZ‐H) and the corresponding plants with jujube witches' broom (DZ‐D). (b) Zoomed area of axillary buds from green box. (c) Zoomed area of axillary buds from the red dashed box. Bar = 1 cm. (d) Expression profiles of TMK JWB and SJP1 JWB in above healthy and diseased plants. (e) Sequence complementation and dual‐luciferase assay analysis of zju‐miR156c and its target gene ZjSPL3. (f) Expression profiles of zju‐miR156c and ZjSPL3 in above healthy and diseased plantlets. (g) Shoot phenotypes of field Dongzao‐H (healthy trees) and Dongzao‐D (diseased trees) plants. (h) The axillary buds of Dongzao‐H from green box. (i) The axillary buds of Dongzao‐D from the red dotted box. (j) Expression profiles of TMK JWB , SJP1 JWB , zju‐miR156c and ZjSPL3 in Dongzao‐D and Dongzao‐H plants. (k) The indole‐3‐acetic acid (IAA), indolebutyric acid (IBA), trans‐zeatin (tZ) and 6‐benzylaminopurine (6‐BA) contents in Dongzao‐D and Dongzao‐H plants. Significance level: *p < 0.05, **p < 0.01, ***p < 0.001. In addition, compared with Dongzao‐H (healthy controls that grew in the field), Dongzao‐D (plants in the field infected by phytoplasma) showed axillary bud germination and shoot proliferation (Figure 3g–i). The expression levels of TMK JWB and SJP1JWB were significantly upregulated in Dongzao‐D, higher than that in Dongzao‐H (Figure 3j). The expression level of zju‐miR156c in Dongzao‐D was approximately 3.1‐fold higher than that in Dongzao‐H, but the expression level of ZjSPL3 was 0.59‐fold lower in Dongzao‐D (Figure 3j). Furthermore, the contents of indole‐3‐acetic acid (IAA) and indolebutyric acid (IBA) in Dongzao‐D were significantly lower than that in Dongzao‐H, but the contents of trans‐zeatin (tZ) and 6‐benzylaminopurine (6‐BA) in Dongzao‐D were significantly higher than that in Dongzao‐H (Figure 3k). Functional analysis of zju‐miR156c in formation of witches' broom The above results indicated that the zju‐miR156c‐mediated pathway may play an important role in the formation of witches' brooms of jujube. Therefore, we transferred zju‐miR156c into A. thaliana to examine whether it is involved in regulating the germination of axillary buds. Interestingly, the results showed that zju‐miR156c‐overexpression lines (OE#1 and OE#2) had more branches compared with the wild type (WT) (Figure 4a). Meanwhile, its target gene AtSPL2/10/11 was significantly downregulated in zju‐miR156c‐overexpression lines. Among these SPL genes, AtSPL10 showed the most downregulated expression level (Figure 4b). In addition, phenotype analysis showed a significant difference in the morphology of rosette leaves, stem leaves and siliques (Figure 4c–g), such as shorter leaf length of rosette and stem leaves, thinner diameter of inflorescence stems and shorter siliques and horn stalks in zju‐miR156c‐overexpression line OE#2 (Figure 4h). This phenotype is consistent with the phenotype of jujube witches' broom. Phenotype observation and hormone contents of zju‐miR156c‐overexpressing lines of Arabidopsis thaliana and sour jujube seedlings. (a) Phenotypes of zju‐miR156c‐overexpressing lines of A. thaliana. (b) Expression profiles analysis of zju‐miR156c and its target genes AtSPL2/10/11 in zju‐miR156c‐overexpressing lines. (c) Phenotypes of zju‐miR156c‐overexpressing line OE#2. (d) Zoomed area of branches from the green box and the red dashed box from Figure 5c. (e–g) Phenotype of rosette leaf, stem leaf and siliques between wild‐type (WT) and zju‐miR156c‐overexpressing line OE#2. (h) Statistical analysis of morphological differences between WT and zju‐miR156c‐overexpressing line OE#2. (i) Phenotypes of zju‐miR156c‐overexpressing sour jujube seedlings. (j) Zoomed areas of branches from pink box and red box from Figure 5i. (k) Expression profiles of zju‐miR156c in zju‐miR156c‐overexpressing sour jujube lines. (l) Phenotypes of zju‐miR156c‐overexpressing lines of 84 k poplar. (m) Zoomed area of branches from pink box and red box from Figure 5l. (n) Expression profiles of zju‐miR156c in zju‐miR156c‐overexpressing 84 k poplar. (o) The indole‐3‐acetic acid (IAA), indolebutyric acid (IBA), trans‐zeatin (tZ) and 6‐benzylaminopurine (6‐BA) contents in zju‐miR156c‐overexpressing sour jujube lines. Bar = 1 cm. Significance level: *p < 0.05, **p < 0.01, ***p < 0.001. Furthermore, compared to the WT sour jujube seedlings, the axillary buds of sour jujube seedlings germinated rapidly after transient transformation with zju‐miR156c (Figure 4i, j) and the expression level of zju‐miR156c increased more than 2‐fold (Figure 4k). In addition, overexpression of zju‐miR156c in poplar caused axillary bud germination, but this was not seen in 84 K and 35S#84K (vector infection) at the same period (Figure 4l–n). Hormone analysis found that the contents of cytokinins (tZ and especially 6‐BA) were increased in zju‐miR156c‐overexpressing jujube (Figure 4o). Regulatory relationship between SJP1JWB and zju‐miR156c Transcription factors can bind the corresponding promoter sites to regulate gene expression. Using PlantPAN 3.0 to predict the transcription factors that might interact with the zju‐MIR156c promoter region, 15 transcription factors were identified, such as MYB, WRKY, NAC, TCP and GATA (Figure 2). The transcription factors were screened by yeast one‐hybrid (Y1H) analysis for their ability to interact with the zju‐MIR156c promoter. The results showed that the optimal concentration of inhibitor (3‐AT) was 50 mM (Figure 5a), and the conversion efficiency was greater than 106 cfu (Figure 5b). Three transcription factors (ZjMYB44 [LOC107424887], ZjWRKY24 [LOC107411616] and ZjERF18 [LOC107404986]) were identified by the Y1H assay as able to bind to the zju‐MIR156c promoter region (Figure 5c). Interestingly, ZjMYB44 and ZjWRKY24 were significantly overexpressed in PZ‐D plants, but ZjERF18 was significantly down‐regulated (Figure 5d). Regulatory relationship analysis between SJP1JWB and zju‐miR156c. (a) Screening of optimal inhibitor (3‐AT) concentration. (b) Analysis of yeast co‐conversion efficiency. (c) Verification of the interaction between zju‐MIR156c promoter and transcription factors by yeast‐one hybrid assay. (d) Expression profiles of transcription factors interacting with the zju‐MIR156c promoter in phytoplasma‐infected Pozao jujube plants (PZ‐D) compared to healthy Pozao plants (PZ‐H). (e) Detection of the pGBKT7‐SJP1JWB self‐activation profile. (f) The interaction analysis of SJP1JWB with ZjMYB44, ZjWRKY24 and ZjERF18 by the yeast‐two hybrid assay. (g) The interaction analysis between SJP1JWB and ZjERF1 by co‐immunoprecipitation assay. Significance level: **p < 0.01, ***p < 0.001. The results described above proved that ZjMYB44, ZjWRKY24 and ZjERF18 have the ability to interact with the promoter of zju‐miR156c. To explore the regulatory relationship between the pathogenic effector SJP1JWB and zju‐miR156c, the interaction of SJP1JWB with ZjMYB44, ZjWRKY24 and ZjERF18 was analysed by a yeast two‐hybrid (Y2H) assay. The results confirmed that SJP1JWB‐BD and ZjERF18‐AD did not autoactivate (Figure 5e), and SJP1JWB did not interact with ZjMYB44 and ZjWRKY24 but could bind to ZjERF18 (Figure 5f). In conclusion, SJP1JWB can regulate the expression of zju‐miR156c through ZjERF18. To further confirm the interaction of SJP1JWB and ZjERF18, a co‐immunoprecipitation (Co‐IP) assay was performed. Compared to the control groups, we could see the anticipated protein line in the groups of ZjERF18‐FLAG and SJP1JWB‐GFP within GFP‐IP, which further confirmed that SJP1JWB could bind to ZjERF18 (Figure 5g). DISCUSSION The relationship between miR156c and witches' broom formation of jujube Jujube witches' broom is a devastating disease caused by phytoplasma. miRNAs are a class of endogenous, noncoding small RNA molecules encoded by the genomes of higher eukaryotes that play important roles in the regulation of gene expression in plants. Previous studies indicated that miR156s play an important role in jujube in response to phytoplasma. For example, transcriptome, metabolome and miRNAs analysis of Paulownia fortunei seedlings identified various miRNAs and their target genes involved in the metabolism of arbuscule disease, plant defence response and pathogen interaction (Cao et al., 2018; Fan et al., 2015). Moreover, through miRNA sequencing analysis, it was found that multiple miRNAs are differentially expressed between healthy jujube trees and phytoplasma‐infected trees. Among them, miR156s were the most significantly upregulated miRNAs (Shao et al., 2016). We also found that zju‐miR156s were significantly upregulated in plants with jujube witches' broom by miRNA‐seq analysis. We found the expression patterns of TMK JWB and SJP1JWB in diseased materials (tissue culture plants and field‐grown jujube trees with witches' broom) were significantly upregulated. When overexpressing zju‐miR156c in A. thaliana, sour jujube seedlings and 84k poplar, transgenic plants showed more branches and smaller leaves, which were similar to the witches' broom symptom. It has previously been found that atspl9/15 mutants can cause increased branching in plants (Xie et al., 2020). In this study, overexpression of zju‐miR156c inhibited the transcription level of SPLs. Thus, the miR156c‐SPL pathway might play a key role in witches' broom formation caused by phytoplasma. In addition, the plant collaterals originate from the axillary bud meristem, and auxin and cytokinin play an important role in axillary meristem formation. In this study, in the transgenic plants of Arabidopsis, poplar and jujube overexpressing zju‐miR156c, we found cytokinin played an important role in the witches' broom symptoms. Mechanism of zju‐miR156c‐mediated network in regulating witches' broom of jujube Pathogenic effectors are the main factors that lead to plant disease. Unravelling the interaction mechanism between pathogenic effectors and target proteins in the host can help to understand the mechanism of pathogen attack. Fifty‐six secreted proteins (effectors) have been predicted in plants, including 30 in onion, 37 in wheat and 27 in jujube (Bai et al., 2009; Chen et al., 2014; Hoshi et al., 2009; Wang et al., 2018). Three types of effectors have been identified from phytoplasmas, namely, SAP11, Phyllogen and TENGU. Among them, SAP11 has been reported to cause shoot proliferation in host plants (Hoshi et al., 2009; Sugio, Kingdom, et al., 2011). The jujube pathogen effectors have been reported to be SJP1, SJP2, SJP3, Zaofeng3 and Zaofeng6, among which SJP1, SJP2 and Zaoeng6 can cause shoot proliferation. Interestingly, in the bacterial pathogen effectors, SJP2 had the same sequence as Zaofeng6, and SJP3 had the same sequence as Zaofeng3. Through the whole genome sequencing of jujube witches' broom phytoplasma (Xue et al., 2023), we identified a pathogenic effector named SJP1JWB, which is the same as the reported SJP1 and it can cause shoot proliferation in host plants (Chen et al., 2022). In this study, we found that SJP1JWB was abundant in jujube plants infected by phytoplasma, and zju‐miR156c was significantly upregulated, but the expression of ZjSPL3 was significantly inhibited, followed by a significant increase of tZ and 6‐BA. SJP1JWB could bind to and degrade ZjERF18, and ZjERF18 can interact with the zju‐MIR156c promoter. According to the above results, the possible mechanism of witches' broom induced by phytoplasma is hypothesized and shown in Figure 6. Leafhoppers carry phytoplasmas to infect jujube trees. Phytoplasmas proliferate in jujube trees and secrete the pathogenic effector SJP1JWB in host cells. Then, SJP1JWB regulates the activity or expression of transcription factor ZjERF18 by interacting with it. With the downregulation of the transcription factor ZjERF18, zju‐miR156c is induced, which further inhibits the expression of target gene ZjSPL3, resulting in the accumulation of 6‐BA and tZ, which finally promotes the formation of witches' broom symptoms with continuous axillary bud germination, smaller leaves and thinner stems. Model of zju‐miR156c‐mediated network in witches' broom formation of jujube. Prospectives of zju‐miR156c study on phytoplasma disease Many studies have revealed that miRNA plays an important role in plant defence against the invasion of pathogens such as viruses, bacteria and fungi. For example, mir156 mutant and SPL9‐overexpressing A. thaliana plants display elevated sensitivity to the necrotrophic fungus Botrytis cinerea. Constitutive expression of miR156a or reduction of SPL9 expression in the sqn‐1 (squint) mutant restores the sensitivity of sqn‐1 to B. cinerea and jasmonic acid responses. Therefore, it is speculated that the miR156‐SPL9 module mediates plant resistance to B. cinerea by influencing the interaction between SQN and jasmonic acid (Sun et al., 2022). The above studies have shown that miRNA is involved in the regulation of plant disease resistance. However, there are few reports of miRNA participating in plant resistance to phytoplasma infection. Our study found that zju‐miR156c plays an important role in regulating witches' broom formation in response to phytoplasma infection. Therefore, further studies on the resistant mechanism of jujube mediated by zju‐miR156c need to be carried out by using witches' broom‐resistant materials (Wang et al., 2022; Xue et al., 2020), which could provide a reference for the study of how plants resist phytoplasma infection. In conclusion, miRNA sequencing showed that zju‐miR156c was highly expressed in jujube witches' broom and possibly involved in arbuscule formation. By analysing the expression patterns of phytoplasma genes TMK JWB and SJP1 JWB , it was found that the expression levels of TMK JWB and SJP1 JWB in phytoplasma‐infected jujube were significantly increased compared to healthy controls. The expression of zju‐miR156c was significantly upregulated, but its target gene ZjSPL3 was significantly downregulated. The levels of cytokinin hormones (tZ and 6‐BA) increased sharply. When zju‐miR156c was overexpressed in A. thaliana, AtSPL2/10/11 was significantly downregulated and cytokinin hormones (tZ and 6‐BA) were significantly increased. The zju‐miR156c‐overexpressing plants had phenotypes like witches' broom of jujube, such as shoot proliferation, smaller rosette and stem leaves, shorter siliques and thinner diameter of inflorescence stems. In addition, Y2H and Co‐IP assays confirmed that SJP1JWB interacted with ZjERF18. Y1H analysis showed that ZjERF18 could interact with the promoter of zju‐MIR156c. In conclusion, SJP1JWB‐ZjERF18‐zju‐miR156c‐ZjSPL3‐CTK is a new path for the formation of jujube witches' broom. EXPERIMENTAL PROCEDURES Materials The tissue‐culture jujube plants were cultivated at the Chinese jujube research centre in Hebei Agricultural University. The healthy Z. jujuba ‘Dongzao’ (DZ) was named DZ‐H or DH, where H stands for healthy. DZ with jujube witches' broom was named as DZ‐D or DHF, where D or F stands for disease. Shoot and leaf samples on DZ‐D/DH and DZ‐H/DHF were collected with three biological replications. Witches' broom leaves of Dongzao and Pozao that were infected by phytoplasma in the field at the experimental station, Fuping County, China, were selected as other test groups (Dongzao‐D and PZ‐D), and the normal leaves from healthy trees were used as controls (Dongzao‐H and PZ‐H). Three biological replicates were conducted. All samples were taken to the laboratory in a cool box for photography. Then they were frozen rapidly in liquid nitrogen and stored at −80°C for DNA, RNA and miRNA extraction. Sour jujube seedlings were used as material to verify the gene function of zju‐miR156c. The seedlings were cultivated under the condition of 25–28°C and 16 h (light)/8 h (dark) photoperiod. The seeds of A. thaliana Col‐0 ecotype were sterilized and placed in a 1/2× Murashige and Skoog (MS) medium with 0.8% agar and 1% sucrose (wt/vol). After stratification at 4°C for 2 days, seeds were transferred and cultivated at 22°C under the 16 h (light)/8 h (dark) condition. After 10 days, seedlings were transferred to growth pots and cultivated under the same environmental conditions. Nicotiana benthamiana was grown in growth pots with 16 h light at an average daily temperature of 22–24°C. The 84k poplar (Populus alba × Populus glandulosa) plants were grown in a greenhouse at about 25°C with an average light duration of 10 h. DNA, RNA, miRNA extraction and cDNA synthesis DNA was extracted by a modified CTAB (Xiao et al., 2015) method, and RNA was extracted by an RNA kit (Tiangen Biotechnology). Five hundred nanograms of total RNA were used for first‐strand cDNA synthesis (cDNA Biotechnology, China) (Bu et al., 2016). miRNA was extracted by using the miRcute polysaccharide polyphenols plant miRNA extraction and isolation kit (Tiangen Biotech), and then miRNA was synthesized into first‐strand miRNA cDNA using the miRcute Enhanced miRNA cDNA First‐strand Synthesis Kit (Tiangen Biotech). Finally, the quality and concentration of DNA, RNA and miRNA were assessed using NanoDrop 2000, and all samples were stored at −20°C for further application. miRNA library construction miRNA library construction was performed by Meji Bio Company (Shanghai) from healthy Dongzao tissue culture plants DH and diseased plants DHF were cultivated at Chinese Jujube Research Centre in Hebei Agricultural University. The concentration and purity of the RNA were tested by NanoDrop 2000, RNA integrity was measured by agarose gel electrophoresis, and the RQN value was determined by Agilent 5300. The miRNA sequencing was performed using the Illumina sequencing platform. miRNA‐seq data analysis Removing adaptors and low‐quality data from the raw data to obtain clean data. By comparing them with the jujube reference genome, the mapped data (reads) for subsequent analysis were obtained, and counting their distribution situation on each chromosome. The miRNA analysis was performed based on the signature hairpin structure of the precursor and predicted the potential new miRNAs. For the new miRNAs, the software miRDeep2 was used to explore the secondary structure, Dicer restriction site information, energy and other features, and finally obtain the sequence information, length, precursor information and score value of the new miRNAs. Expression statistics were performed for known miRNAs and new miRNA in the samples, and expression was normalized by using TPM values. Target gene and transcription factor prediction analysis of zju‐miR156c The target genes of zju‐miR156c were predicted by using the online TAPIR software (http://bioinformatics.psb.ugent.be/webtools/tapir/) and visualized by Gephi software (Bonnet et al., 2010; Wang et al., 2022). The 2000 bp upstream sequence of the precursor sequence of the zju‐miR156c was obtained by TBtools software. The transcription factors regulating the expression of the zju‐miR156c were predicted by the online software PlantPAN3.0 and visualized by Gephi software. Luciferase reporter assay for the zju‐miR156c and target gene ZjSPL3 Based on the predicted results, the miRNA target assessment was performed using the pmirGLO vector. Constructs of pmirGLO‐ZjSPL3 (Data S3), 35S::zju‐miR156c‐GFP as effector, were syringe‐infiltrated into the leaves of 4‐ to 6‐week‐old N. benthamiana plants. At least three leaves were infiltrated. Three days after infiltration, the fluorescence signal in leaves was detected by the intra‐vital imager (Liu et al., 2021). The expression analysis of TMK JWB , SJP1JWB , ZjMYB44 , ZjWRKY24 , ZjERF18 , zju‐miR156c, ZjSPL3 and AtSPL2/10/11 The SJP1JWB used in this study was the same as the reported SJP1 (Zhou et al., 2021), and the current naming method aimed to distinguish different phytoplasmas. Primers for SJP1JWB were designed, and TMK JWB was referred to Wang et al. (2022). cDNA from DZ‐D and DZ‐H, Dongzao‐D and Dongzao‐H were used as templates, and the expression levels of TMK JWB and SJP1JWB were detected by reverse transcription‐quantitative PCR (RT‐qPCR). In addition, the primers for zju‐miR156c, ZjSPL3 and AtSPL2/10/11 were designed, and the expression level of zju‐miR156c and the target gene ZjSPL3 was determined by RT‐qPCR. Transgenic zju‐miR156c Arabidopsis and WT lines were used as materials for expression analysis of zju‐miR156c and its target gene AtSPL10 by RT‐qPCR. Transgenic zju‐miR156c sour jujube seedlings and WT lines were used as materials for expression analysis of zju‐miR156c. The RT‐qPCR was performed on a CFX96 Touch machine with TransStart Top Green qPCR SuperMix (Tiangen Biotechnology). The 20 μL reaction system contained 10 μL of 2 × SuperReal PreMix Plus, 0.4 μL primers, 1 μL diluted cDNA and 8.2 μL double‐distilled water. The reaction process was 15 min at 95°C, followed by 40 cycles of 10 s at 95°C, 20 s at 55–63°C and 20 s at 72°C. ZjActin was used as the housekeeping gene (Bu et al., 2016). For the expression of miRNAs, the 20 μL reaction system was the same as above with 15 min at 95°C, followed by 5 cycles of 20 s at 94°C, 30 s at 55–63°C and 34 s at 72°C, and 40 cycles of 20 s at 94°C, 34 s at 60°C. U6 was used as the housekeeping gene (Liu et al., 2021). The relative expression levels were calculated by the 2−ΔΔCt method (Livak & Schmittgen, 2001), and all the primer sequences of RT‐qPCR were shown in Data S4. The transcriptome data of Pozao infected by phytoplasma (Wang et al., 2022) was used to analyse the expression level of ZjMYB44 (LOC107424887), ZjWRKY24 (LOC107411616) and ZjERF18 (LOC107404986). zju‐MIR156c overexpression analysis in A. thaliana and sour jujube To generate zju‐MIR156c‐overexpression plants of A. thaliana, Agrobacterium tumefaciens GV3101 containing 35S::zju‐MIR156c‐GFP (Data S5) was used to transform A. thaliana Col‐0 and the transgenic lines were identified on plates with hygromycin and RT‐qPCR analysis. Ten zju‐miR156c‐OE and WT plants grown for 25 days were randomly selected and conducted morphological investigation, including leaf length and width of rosette leaves, leaf length and width of stem leaves, the number of inflorescence stems, the thickness of the inflorescence stem (0.3 and 0.8 cm from the substrate surface) and the length of the siliques. Three biological replicates were conducted. 35S::zju‐MIR156c‐GFP was transiently transformed into the sour jujube seedlings by A. tumefaciens GV3101 through cotyledon injection; the uninjected sour jujube seedlings were used as control. The phenotype of axillary bud germination was observed and the expression level of zju‐miR156c was tested 30 days after injection. Transgenic poplar was obtained by transforming 35S::zju‐MIR156c‐GFP into 84k poplar using the leaf disc method. Screening of transcription factors interacting with the promoter of zju‐MIR156c The transcription factors interacting with the zju‐MIR156c promoter were identified by using the cDNA library of jujube leaves. Then, the promoter sequence of zju‐MIR156c was inserted into the pHIS2 vector, and the transcription factors (ZjMYB44, LOC107424887; ZjWRKY24, LOC107411616; ZjERF18 and LOC107404986) obtained by library screening were inserted into the pGADT7 vector (Data S3). Both pHIS2 vector and pGADT7 vector were co‐transformed into yeast strain Y187 for Y1H assays. Y2H assays Individual construction of pGBKT7‐SJP1JWB, pGADT7‐ZjMYB44, pGADT7‐ZjWRKY24 and pGADT7‐ZjERF18 for the Y2H was performed (Data S3). Firstly, self‐activation detection was conducted, then the pGBKT7‐SJP1JWB together with pGADT7‐ZjMYB44, pGADT7‐ZjWRKY24 and pGADT7‐ZjERF18, transformed into yeast strain Y2H and grown in SD/Trp/Leu plates. The yeast strains were then transferred to SD/Trp/Leu/His and SD/Trp/Leu/His/Ade plates, and the growth of yeast was observed and photographed. Co‐IP assay Firstly, the 35S::ZjERF18‐FLAG and 35S:: SJP1JWB‐GFP vectors were constructed using the primers shown in Data S3. Then the vectors were transformed into A. tumefaciens GV3101, and the OD600 of the Agrobacterium infection solution was adjusted to 0.6–1.0. The 1‐month‐old N. benthamiana plants were selected and injected with the above solution. After 2 days, the injected leaves were selected and rapidly frozen for the extraction of total proteins. The process involved preparing beads, binding antibodies, binding bait proteins, boiling off, electrophoresis detection, membrane transfer, antibody hybridization and other steps. Afterwards, the images were exposed and photographed. Detection of hormone contents The indole‐3‐acetic acid (IAA), indolebutyric acid (IBA), trans‐zeatin (tZ) and 6‐benzylaminopurine (6‐BA) contents of Dongzao‐D, Dongzao‐H, zju‐miR156c‐overexpressing sour jujube seedlings and their control were determined by Shandong Guo Cang Jian Biotechnology Co., Ltd. using high‐performance liquid chromatography‐tandem mass spectrometry (HPLC‐MS/MS). Statistical analysis The data were statistically analysed using single factor analysis by SSPS v. 25.0, and all the results were expressed as mean ± SE. Supporting information Data S1. The miRNA sequencing quality information. Data S2. Analysis of the miRNA‐seq data. (a) Statistical analysis of effective miRNAs in chromosomes. (b) Densitometric chromosome density analysis of effective miRNAs. (c) The Pearson correlation analysis of four samples (d) Number of miRNA in samples. Data S3. The primer sequences for yeast two‐hybrid, co‐immunoprecipitation and luciferase complementation imaging assays. Data S4. The primer sequences for reverse transcription‐quantitative PCR analysis. Data S5. The primer sequences for overexpression vector construction.