Observation on growth and development whole period of Z. latifolia
The development and growth period of Z. latifolia had not been determined yet, and it is a pretty important parameter to find a reasonable period for us to accurately grasp the experiment and conduct experimental treatment. Therefore, we observed and photographed the Z. latifolia at 10 days intervals for 20 days after transplanting. The photos were divided into front view, cross section and vertical section (Fig. 1). In the 150 days after setting, we took photos every 3 days, and a total of 27 group of photos were taken in the whole process. The period was divided into three leaves (including three leaves) and 28 days after transplanting. At the fifth leaf stage, 79 days after the transplanting of Z. latifolia. At the 7th leaf, 110 days after the determination of Z. latifolia; 153 days after transplanting Z. latifolia, before gall formation; After gall formation first period, 156 days after transplanting; After gall formation second period, 159 days after transplanting; After gall formation third period, 162 days after transplanting; After gall formation fourth period, 165 days after transplanting; After gall formation fifth period, 168 days after transplanting.
Microscopic observation of U. esculenta in different section reveals the invasion process of Z. latifolia
In the growth of Z. latifolia, we found that normally grown can formation swelling gall (Fig. 2A). However, TDF treat Z. latifolia without formation of swelling gall (Fig. 2B). We observed the samples of before and after gall formation in CK and TDF. It was found that no swelling gall was formed in the before, but in the late period of untreated, the Z. latifolia had continues to expand (Fig. 2C). Optical microscope observation find that aniline blue staining showed that hyphae could only be observed at after gall formation in CK using ordinary section and paraffin section (Fig. 2D, E). The infection processes of U. esculenta in the gall of CK and TDF treatment were observed using scanning electron and ultrastructure (Fig. 2F, G). By shown in Fig. 2F, G no difference between JB_A and JB_C was observed at swelling gall formation before. In JB_B, we find many spores showing round. The sporophytes are clustered together and attached to the cell wall around. The fungal cell wall (FCW) can be clearly find in the ultrastructure, and it is closely attached to the plant cell wall and may communicate with the plant body.
Analysis of different hormone response to U. esculenta infection
To detection the levels of hormone response to U. esculenta infection, UHPLC was used to measure IAA, ABA, Z, ZR and GA3 contents (Fig. 8, S3) in Z. latifolia gall formation before and after CK and TDF group. As shown in Fig. 8A, Z content was high in JB-B, decreased sharply by 8.21% from JB-B to JB-D and then no significant change from JB-A to JB-C. To our surprise, we also found that GA3 and IAA contents were most than others, indicating the contents of IAA, ABA and GA3 may be due to the symbiosis of U. esculenta infection plant, which stimulates the plant to produce a large amount of growth hormone and causes the swelling gall formation. ABA and ZR content were relatively low during the JB-B. Both Z and ZR belong to cytokinin (CTK) and are usually considered as CTK together. Meanwhile, we compared the ratio of each hormone to ABA content and divided them into 7 groups including IAA/ ABA (Fig. 8B), Z+ZR/ABA (Fig. 8C), GA3/ABA (Fig. 8D), IAA+ GA3/ABA (Fig. 8E), IAA+Z+ZR/ABA (Fig. 8F), Z+ZR+ GA3/ABA (Fig. 8G) and IAA+ GA3+Z+ZR/ABA (Fig. 8H). We found that the content of JB-B in these ratios was always higher than that of JB-D, while there was no significant change in JB-A and JB-C. The results indicated that the U. esculenta infection Z. latifolia could stimulate the host to produce a large amount of growth-promoting hormone and cause the gall formation and expansion.
Overview of sequencing and transcript identification
To study the differences in RNA-seq between U. esculenta infection Z. latifolia and non-infected Z. latifolia in gall formation before and after, three biological replicates were performed on 12 cDNA samples. After removing the low copy and quality sequences using the generic Perl script (Table 1), a total of 84.42 Gb of clean reads was generated. The quality score of more than 95.45% reads and more is equal or greater than Q30, accounting for 52.16-53.52% of the GC content as shown in Table 1. An average of 88.42% of reads were mapped to the reference genome (http://ibi.zju.edu.cn/ricerelativesgd), of which 86.26% were located at unique locations (Table 1). A total of 17,541 DEGs were optimized by known structures (Fig. 3A). Pearson's significant correlation (Fig. S1) between the FPKM distributions of biological replicates of all samples confirmed the high reproducibility of the sequencing data.
Identification of different expressed genes
By shown in Figure 3, in JB_B vs. JB_A, JB_D vs. JB_B, JB_D vs. JB_C and JB_C vs. JB_A, there were 3122, 2672, 6704 and 5042 DEGs identified by the DESeq R package based on FPKM data respectively (Fig. 3). Principal component analysis (PCA) indicated that replicate samples within each sample group clustered together. Venn and upset plot diagram show that among them, 1430, 2852, 1293 and 2795 DEGs were up-regulated, while 1242, 2191, 1829 and 3909 DEGs were down-regulated in JB_B vs. JB_A, JB_D vs. JB_B, JB_D vs. JB_C and JB_C vs. JB_A, respectively (Fig. 3A). Through the comparison between the four libraries, 3542 unique and 477 common DEGs were identified, among which JB_B pairs JB_A, JB_D pairs JB_B, JB_D pairs JB_C, JB_C pairs JB_A were 713, 537, 1726 and 566 unique DEGs respectively (Fig. S9). Venn and upset plot diagram were made based on the up-regulation and down-regulation data in Fig C and Fig D, from which we could find that the 4 groups of data were not up-regulation or down-regulation common DEGs, suggesting that the gall before and after formation and TDF reached the expected effect in Z. latifolia.
GO and KEGG enrichment analyses for all DEGs
To identify the similarities and differences between JB_B and JB_A and JB_D and JB_B in U. esculenta infection transcriptomes, DEGs were used for GO classification and KEGG functional enrichment analysis. The comparison of the distribution of DEGs and all genes at top 10 in JB_B vs. JB_A and JB_D vs. JB_B were allocated to three categories of the biological process (BP), cellular component (CC) and molecular function (MF) (Fig. S2; Table S3). For the DEGs in JB_B vs. JB_A, the significant GO terms were mostly enriched in adjusting of transcription, DNA-templated, cell wall organization and positive regulation of transcription, in BP category. The DEGs were mainly distributed in extracellular region, membrane anchoring components and plant cell wall in the CC category. The top three GO terms in the MF category were "DNA-binding transcription factor activity", "sequence-specific DNA binding" and "heme binding" (Table S3A). In the DEGs of JB_D vs. JB_B, ‘sinapoylglucose-choline O-sinapoyltransferase activity’ and ‘cellulose synthase activity’ term was remarkable in the BP category, the ‘plasma membrane’, ‘extracellular region’ and ‘plasmodesma’ terms in the CC category. Meanwhile, ‘zinc ion transmembrane transport’ and ‘adventitious root development’ term in the MF category was most significant. The most common GO terms were ‘cell wall’, ‘DNA-binding transcription factor activity’ and ‘extracellular region’ (Table S3B). Therefore, the corresponding genes of these important terms might play a central role in the fight against U. esculenta infection. In the figure 4A, C, we found that the most KEGG classification level was in the environmental information and classification level2 was signal transduction processing in JB_B vs. JB_A and JB_D vs. JB_B. The gene number is 80 and 105 JB_B vs. JB_A and JB_D vs. JB_B (Fig. 4, Table S4).
The KEGG enrichment top 20 pathway in the two groups were shown in figure 4B, D. In JB_B vs. JB_A, DEGs are dominating enriched expressed in plant hormone signal transduction, phenylpropanoid biosynthesis, starch and sucrose metabolism. Similarly, in JB_D vs. JB_B, plant hormone signal transduction was also enriched most. In plant hormone signal transduction (Ko04075), 48 and 37 DEGs were identified in the JB_B vs. JB_A and JB_D vs. JB_B, respectively. Among 48 DEGs in comparison of JB_D vs. JB_B, 22 genes were overlapped with those in JB_B vs. JB_A, and 26 genes belonging to signal transduction response regulator were unique (Fig. 7A). The different expression patterns of hormone metabolism related DEGs between CK and TDF suggest that plant hormones play a central role in regulating the against response to U. esculenta infection in Z. latifolia.
Metabolism and regulatory pathways analyses of all DEGs
Regulatory pathways in JB_D vs. JB_B were studied by MapMan tool to annotation analyses (Fig. 5, Table S2), and regulation overview enrichment of DEGs were up-regulated and functionally enriched in transcription factors (TFs) including homeobox transcription factor family and receptor kinases (Fig. 5A). Some pathways including protein modification, calcium regulation and protein degradation were also up-regulated or down-regulated in response of JB_D vs. JB_B to U. esculenta infection. Most DEGs linked to plant hormones associated with down-regulation of cytokinin, ethylene and abscisic acid, and only IAA mostly up-regulation expression (Fig. 5A, Table S3A). Through biological stress analysis (Fig. 5B), the relevant genes are divided into signal transduction, PR protein, TFs, hormones ABA, SA, JA and ethylene, further supporting the crucial of these pathways in symbiotic environment regulated sensing and promotion response to U. esculenta infection in Z. latifolia. A more detailed list of all DEGs corresponding to MapMan functional categories was provided (Fig. 5A, Table S3B). TFs were primary regulators of DEGs expression and perform significant functions in the transcriptional symbiosis of plant-fungi genes after U. esculenta. Meanwhile, we described TFs in detail and found that HB, bHLH and AP2-EREBP had more differentially up-regulated expression (Fig. 5A). We showed the expressions of 308 putative TF genes, which can be divided into 42 TF families in JB_D vs. JB_B (Table S3C). Both the B3 and the homeobox transcription factor family have obvious high-expressed TFs, including Zlat_10033978 and Zlat_10028938. B3 transcription factor family was main auxin response factors family that regulate auxin response. HB (homeobox transcription factor family) was a DNA binding motif within TF proteins. These TF might be involved in hormone regulate, cell differentiation and expansion, as well as expression patterns in patterning of different organisms after U. esculenta infection. These TFs seem to be in connection with stimulate gall formation response according to the annotation information.
Changes in plant hormone signal transduction and metabolism related gene expression after U. esculenta infection
The expression of genes related to plant hormone metabolism and signal transduction indicated meaningfully dynamic changes during the U. esculenta infection process. DEGs were related to IAA, CTK, GA, ABA, Eth, BR, JA and SA (Fig. 6; Table 2) were observed from 37 to JB_D vs. JB_B. At JB_B vs. JB_A 48 DEGs, 76 DEGs in JB_D vs. JB_C and 53 DEGs in JB_C vs. JB_A (Fig. 7B).
The IAA-related pathways had the most response to plant phytoplasma infections. In the IAA signal transduction pathway, 9 genes were up-regulated in JB_D vs. JB_B, including 4 auxin-responsive proteins (AUX/IAA), 3 AUX1 family proteins (auxin influx carrier family), 1 auxin responsive GH3 gene family (GH3) and 1 auxin response factor (ARF). On the contrary, 3 AUX/IAA (Zlat_10001127, Zlat_10020059, Zlat_10033997) and an AUX1 (Zlat_10014750) were down-regulated by -1.26, -1.11, -1.52 and -2.29-fold, respectively, at JB_D vs. JB_B. Most of the up-regulation expression which indicated that IAA signaling was obviously affected by U. esculenta infection. We suspect that auxin may be caused by the gall formation caused by the growth substances produced by the infection of U. esculenta. In the CTK signaling transduction pathway, 2 DEGs (Zlat_10010976, Zlat_10018529) were found to be Up-regulated, including two-component response regulator ARR-B family (B-ARR) was up-regulated 1.31-fold and 1.22-fold. In the GA signaling pathway, encoding TF (phytochrome-interacting factor 4, Zlat_10039510), was significantly down-regulated at JB_D vs. JB_B, which indicated that GA signaling was less affected by U. esculenta infection. The 8 DEGs were associated with ABA signal transduction pathway, 6 were up-regulation expression and 2 were down-regulation expression (Fig. 7C). Interestingly, 6 ABA-related DEGs were up-regulated on JB_D vs. JB_B signal transduction pathways (Fig. 7C). The ABA receptor PP2C, SnRK2 and ABF transcription factor family were up-regulated DEGs in the signal transduction pathway and participate in the negative regulation of ABA signaling. For example, PP2C (Zlat_10022265) was up-regulated by 4.96-fold in JB_D vs. JB_B. Up-regulation of these key DEGs in ABA pathway indicated that ABA expression might be increase after TDF treatment causes the cell senescence. On the contrary, ABA lower content after U. esculenta infection and down-regulation of gene expression were characterized by cell proliferation and development in JB_D vs. JB_B. Two genes involved in JA signaling were down-regulated (JAR1, Jasmonic acid amino synthetase) at JB_D vs. JB_B. From 7 SA related genes were identified, with 6 of them related TGA transaction factor and 1 related to regulatory protein NPR1. Most of the DEGs were up-regulated. For example, the largest change in SA-relative gene expression was for TGAL1 (Zlat_10031473), which was up-regulated by 3.57-fold at 3 JB_D vs. JB_B. A few SA-related DEGs were down-regulated, such as one associated with the regulatory protein NPR1 genes, and down-regulated by 2.16-fold. In summary, most of the genes involved in IAA, CTK, and SA are down-regulated, and the hormone content after U. esculenta infection. However, most of the gene expression patterns involving SA, JA, ABA metabolism and signal transduction are up-regulated by U. esculenta infection (Fig. 6, 7; Table 2).
Expression of plant hormone biosynthesis genes and quantitative by qRT-PCR
A set of 25 DEGs (Fig. 9) were selected for quantitative real time PCR analysis to confirm their function of response U. esculenta infection in 17 JB_B vs. JB_A and 15 JB_D vs. JB_B. As shown in Fig. S4, comparison of transcriptome data with qRT-PCR results showed a relatively high correlation (R2 = 0.9723), verifying accountable RNA-seq analysis in the present research. We chose 25 of DEGs including 10 (IAA), 4 (CTK), 3 (ABA), 3 (SA), 3 (JA) and 2 (GA). Among 25 DEGs, 20 DEGs up-regulation and 5 down-regulation were confirmed by qPCR in JB_D vs. JB_B accordance with results of RNA-seq analysis (Fig. 9).
Many phytohormone play a significant role in regulating plant development and modulating diverse biological processes such as IAA, CTK, GA and so on. The auxin response factor (ARF) genes are central components of plant auxin signal transaction. To understand how IAA and CTK participate in and affect swollen gall formation, we identified genes encoding IAA and CTK signal transduction relative genes at the host gall tip and regulated their expression during gall formation. These candidates included 10 IAA genes (ZLIAA23, ZLIAA2, Os11g0169200, ZLSAUR39, ZLPLC2, ZLYUC11, ZLIAA9, ZLIAA10, ZLARF21 and ZLIAA50) and 4 CTK (ZLCKX8, ZLBPA1, ZLRR1 and ZLRR4) hormone related genes. We detected that ZLIAA23, ZLIAA2, Os11g0169200, ZLSAUR39, ZLPLC2, ZLYUC11, ZLIAA10, ZLARF21 and ZLIAA50 were not obviously affected by U. esculenta infection before gall formation, but at U. esculenta infection, its expression was significantly induced (Fig. 9). Similarly, ZLCKX8, ZLBPA1 and ZLRR4 expressed consistently in control before gall formation, but at U. esculenta, its expression level was obviously up-regulated in Z. latifolia.