Background

Chrysanthemum, a perennial plant belonging to the family Asteracae, has been cultivated for thousands of years in China and has a high cultural value [1]. Chrysanthemum follows rose in the cut flower trade and holds a prominent position in the global flower market [2]. Chrysanthemum Fusarium wilt is a common soil-borne disease of chrysanthemum that occurs throughout its growth period. Fusarium oxysporum is the primary pathogen causing Chrysanthemum Fusarium wilt. It is a facultative parasitic fungus that can survive in plants and soils. F. oxysporum invades the roots of plants, resulting in blockage of vascular bundles, and causes water shortage in the above-ground parts of plants, resulting in wilting, yellowing, and eventually death of plant leaves [3,4,5]. Additionally, F. oxysporum can secrete a pathogenic toxin during the process of infection known as fusaric acid, which has allelopathic effects on plants and affects their healthy growth [6]. F. oxysporum spreads widely and causes more than 100 plant diseases in a wide host range, such as cucumber, cotton, banana, and flowers [7,8,9]. In recent years, with the continuous expansion of chrysanthemum planting area, F. oxysporum has spread rapidly in the soil and among chrysanthemums, causing frequent obstacles to the continuous cropping of chrysanthemum and a severe reduction in its production.

Plants have evolved to rely on their defense systems to resist biotic stress. After the recognition of pathogen associated molecular patterns (PAMP) by pattern recognition receptors on plant cell membranes, PAMP-triggered immunity is activated. When pathogenic bacteria secrete effectors into the plant, the plant can rely on its resistance proteins to activate effector-triggered immunity and effectively inhibit the spread of pathogens [10,11,12]. Furthermore, during biological stress, the osmotic adjustment substances accumulated in plants can regulate, protect and scavenge active oxygen and perform other functions. Phenylalanine ammonia-lyase (PAL) and peroxidase (POD) are critical enzymes involved in the biosynthesis of secondary metabolites in plants, and improving the activity of these defense enzymes can increase the stress resistance of plants [13]. More importantly, the defense mechanisms of plants in response to F. oxysporum infection have gradually been discovered. When F. oxysporum infects tomatoes, vanillin, vanillic acid, and other phenolic metabolites in tomato roots can enhance the deposition of root cell walls, thus delaying the invasion of pathogens [14]. A study on responses of soybean to F. oxysporum infection revealed that the changes in secondary metabolites in wild soybean were more significant than those in cultivated soybean, and the disease symptoms were weaker [15]. In addition to plant defense enzymes and secondary metabolites, plant transcription factors (TFs) play an important role in the resistance pathway. Overexpression of TaNACL-D1 enhances the resistance of wheat plants to Fusarium wilt [16]. In recent studies on cotton Fusarium wilt, IIc WRKY TFs were shown to upregulate the expression of GhMYC2-mediated flavonoid synthesis genes, accumulate flavonoid content, and enhance resistance to F. oxysporum [17]. However, only few studies have investigated the mechanism of chrysanthemum response to F. oxysporum infection. For example, CmWRKY6-1 and CmWRKY8-1 in chrysanthemum could respond to F. oxysporum infection via the SA pathway [18, 19]. At present, more about the mechanism of chrysanthemum response to F. oxysporum infection is still unclear.

The defense mechanisms of plants are complex, and RNA-seq technology can be used to effectively analyze changes in gene expression during the interaction between plants and pathogens. RNA-seq is used to sequence specific species in a specific period to analyze the expression of some genes during this period, providing an important reference for screening key genes [20]. Comprehensive metabolomic and transcriptomic analysis showed that genes related to the flavonoid biosynthesis pathway, IAA, SA, and JA signal transduction pathways were significantly enriched in wheat leaves inoculated with Fusarium graminearum [21]. RNA-seq technology to study the effect of CmWRKY53 on chrysanthemum implied that CmWRKY53-mediated aphid sensitivity might be related to secondary metabolites [22]. In the present study, we used RNA-seq technology to analyze the differential expressed genes (DEGs) in chrysanthemum during the early stage of inoculation with F. oxysporum, investigate the physiological response and gene expression changes in chrysanthemum in response to F. oxysporum infection. We analyzed the changes of related pathways in chrysanthemum after inoculation with F. oxysporum, and provided candidate genes for developing new cultivars resistant to Fusarium wilt in chrysanthemum.

Results

Morphological identification of F. oxysporum and Root Rot of Chrysanthemum

The genomic DNA of the pathogenic fungus was extracted, amplified using PCR, and sequenced. The sequencing results were compared with National Center for Biotechnology Information (NCBI) nucleic acid sequences to establish a phylogenetic tree, which showed that it had the highest homology with Fusarium oxysporum strain WZ 321 (Fig. 1a). Furthermore, macroscopic observations revealed a pink fluff filamentous shape on the PDA plate (Fig. 1b). In addition, microscopic observations showed that the spores were curved at both ends (Fig. 1c). Therefore, we named this strain Fusarium oxysporum A2.

Fig. 1
figure 1

Identification and pathogenic phenotype of F. oxysporum A2. a Phylogenetic tree of F. oxysporum A2, the scale bar is 0.01. b Macroscopic observation of the phenotype of F. oxysporum A2, the scale bar is 2 cm. c Microscopic observation of the phenotype of F. oxysporum A2, the scale bar is 100 μm. d Phenotypic observation of ‘Jinba’ before and after infection by F. oxysporum A2, the scale bar is 4 cm. e Changes of cell membrane permeability (Left) and MDA (Right) in ‘Jinba’ after infection by F. oxysporum A2

Full size image

The ‘Jinba’ cultivar was inoculated with F. oxysporum. At 72 h after inoculation, the roots were browned and rotted, and the leaves above the roots of ‘Jinba’ began to wilt slightly (Fig. 1d). The results showed that F. oxysporum began to invade the shoot from the root, blocking the vascular bundle and causing a water shortage. Furthermore, we determined the cell membrane permeability and MDA content of ‘Jinba’ before and after inoculation. The results showed that the cell membrane permeability and MDA of ‘Jinba’ were significantly higher than those of the control at 3 and 72 h after inoculation and gradually increased (Fig. 1e).

Library Construction and sequencing

The roots of ‘Jinba’ inoculated with F. oxysporum at 0, 3, and 72 h were analyzed using RNA-seq (the control group was named SM-CK-0 h, SM-CK-3 h, and SM-CK-72 h; the experimental group was named SM-A-3 h and SM-A-72 h). 720,233,772 clean data points were obtained, and the percentage of Q30 bases in all samples was ≥ 90% (Additional file 1: Table S1). Correlation analysis showed that the samples within the group were highly correlated (Fig. 2a). All the sequencing results showed that the quality of the sequencing data was reliable and could be used for further analysis. The RNA-seq data has been uploaded to the NCBI SRA database. The current sequencing data were deposited in the NCBI SRA under the accession number PRJNA926886.

Differentially expressed genes analysis

To identify the genes of ‘Jinba’ that respond to F. oxysporum infection in the early stage, we compared the DEGs of the experimental group at 3 h after inoculation and the control group at 3 h without inoculation, as well as the DEGs of the experimental group at 72 h after inoculation with the control group at 72 h without inoculation. Using edgeR, we set the thresholds for log2 fold-change ≥ 1 and FDR (padj) < 0.05. As a result, a total of 8677 genes were downregulated, 15,211 genes were upregulated between 3 h after inoculation and 3 h without inoculation, 9415 genes were downregulated, and 13,001 genes were upregulated between 72 h after inoculation and 72 h without inoculation (Fig. 2b). Among these DEGs, 7985 were co-expressed at both time points (Fig. 2c). Among the 7985 co-expressed genes, we analyzed their expression trends at 0, 3, and 72 h after inoculation. The results showed that 1959 genes were downregulated first and then upregulated, 960 genes were upregulated first and then downregulated, 205 genes were upregulated continuously, and 554 genes were downregulated continuously (Additional file 3: Fig. S1).

As our study aimed to identify the genes and pathways involved in the early response to F. oxysporum infection, 7985 co-expressed genes were the focus of our next analysis.

Fig. 2
figure 2

RNA-seq analysis. a Correlation among samples. b Number of up-and down-regulated DEGs. c Venn diagram of 3 h vs. 72 h co-expressed genes

Full size image

GO and KEGG Analysis of co-expressed DEGs

To explore the relevant functions of the 7985 co-expressed genes, we performed GO analysis. Enriched to “Biological process”, “Cellular component”, and “Molecular Function” with 5295, 847, and 1981 entries, respectively. We analyzed the top20 GO terms that were significantly enriched. In the “Biological process”, several GO terms relating to redox and plant responses to fungi, water, and organisms are significantly enriched: “Oxidation-reduction process”, “Response to water”, “Response to water deprivation”, “Response to external biotic stimulus”, “Response to biotic stimulus”, “Response to oxidative stress”, “Response to fungus”, “Defense response to fungus”, and “Defense response to bacterium” (Fig. 3a). In the “Cellular component”, we found a significant enrichment of GO terms for membrane: “Membrane part”, “Plasma membrane”, “Cytoplasmic side of membrane”, and “Integral component of membrane” (Fig. 3b). In “Molecular Function”, most of the GO terms were related to activity: “Catalytic activity”, “Oxidoreductase activity”, and “Peroxidase activity” (Fig. 3c).

At the same time, we performed KEGG analysis on these DEGs, and 129 KEGG pathways were enriched. In addition, we analyzed the top 20 KEGG pathways, including “Plant-pathogen interaction”, “MAPK signaling pathway”, “Starch and sucrose metabolism”, and “Biosynthesis of secondary metabolites”, which are associated with plant disease resistance (Fig. 3d).

Fig. 3
figure 3

GO and KEGG analysis of co-expressed DEGs. a The top 20 GO terms with the most significant Q value in “Biological process”. b The top 20 GO terms with the most significant Q value in “Cellular component”. c The top 20 GO terms with the most significant Q value in “Molecular Function”. d The top 20 KEGG pathways

Full size image

Secondary metabolites

As we enriched the 3 and 72 h co-expressed DEGs to the “Biosynthesis of secondary metabolites” pathway through the KEGG analysis, we analyzed the mechanisms concerning the plant response to pathogen infection from the DEGs of the “Biosynthesis of secondary metabolites” pathway (Fig. 4a). The secondary metabolites of plants contain phenolics, terpenoids, and nitrogenous compounds that play important roles in insecticide, antibacterial, and antioxidant resistance [23]. Within 72 h of F. oxysporum infection, five PAL (evm.TU.scaffold_1462.187, evm.TU.scaffold_1650.57, evm.TU.scaffold_6298.50, evm.TU.scaffold_741.126, evm.TU.scaffold_7835.98), three COMT (evm.TU.scaffold_1333.81, evm.TU.scaffold_1583.59, evm.TU.scaffold_1797.93), CCoAOMT (MSTRG.168,044, evm.TU.scaffold_1099.150, evm.TU.scaffold_1099.166), PPO (evm.TU.scaffold_1212.128, evm.TU.scaffold_126.153, evm.TU.scaffold_11661.327, evm.TU.scaffold_461.48, evm.TU.scaffold_687.86), and one POD (evm.TU.scaffold_7660.4) were upregulated. At the same time, to analyze the changes in secondary metabolites more intuitively after inoculation, we determined the enzyme activity of PAL, polyphenol oxidase (PPO), and POD, which are related to phenol synthesis in plants, and the total phenol content. The results showed that the enzyme activities of PAL, PPO, and POD in ‘Jinba’ increased after inoculation with F. oxysporum, and the total phenol content also increased accordingly (Fig. 4b, c, d, e). In conclusion, within 72 h after inoculation of ‘Jinba’, the infection of F. oxysporum affected the genes related to the “Biosynthesis of secondary metabolites” pathway, thus changing the content of enzymes related to phenolic synthesis and total phenolic content in ‘Jinba’.

Fig. 4
figure 4

DEGs involved in the “Biosynthesis of secondary metabolites”. a Expression analysis of DEGs related to signal transduction pathways of secondary metabolites. Legend normalized for gene expression using z-score; the same scheme applies below. b Changes of PAL in ‘Jinba’ after infected by F. oxysporum. c Changes of PPO in ‘Jinba’ after infected by F. oxysporum. d Changes of POD in ‘Jinba’ after infected by F. oxysporum. e Changes of total phenols in ‘Jinba’ after infected by F. oxysporum

Full size image

Plant osmotic substances

When plants are subjected to water or osmotic stress, they protect themselves by regulating osmotic substances, such as soluble sugars and proline [24]. Changes in these osmotic substances are essential mechanism for the survival of plants in adverse environments. To explore the changes in osmotic substances in response to F. oxysporum infection, we analyzed the DEGs of the co-expressed gene-related pathways at 3 and 72 h after inoculation (Fig. 5a, b). In the “Starch and sucrose metabolism” pathway, four BGLU (evm.TU.scaffold_1004.60, evm.TU.scaffold_1042.264, evm.TU.scaffold_1483.114, evm.TU.scaffold_ 25.386), one SPS (evm.TU.scaffold_476.25), and six SS (evm.TU.scaffold_1269.627, evm.TU.scaffold_3092.67, evm.TU.scaffold_485.33, evm.TU.scaffold_952.408, evm.TU.scaffold_10689.19, evm.TU.scaffold_6916.24) were significantly decreased within 72 h of inoculation compared to pre-inoculation. In the “proline and arginine metabolism” pathway, three ALDH (evm.TU.scaffold_11851.88, evm.TU.scaffold_405.102, evm.TU.scaffold_953.54) were significantly decreased at 3 h after inoculation and significantly increased at 72 h after inoculation.

Further, we measured the changes in the content of these osmotic substances in ‘Jinba’ before and after inoculation. We found that soluble sugar content decreased significantly after inoculation (Fig. 5c). Proline levels transiently decrease at 3 h, followed by a significant increase at 72 h (Fig. 5d). The results showed that the genes related to the “Starch and sucrose metabolism” and “proline and arginine metabolism” pathways were affected by the invasion of F. oxysporum in ‘Jinba’ within 72 h after inoculation, thereby changing the content of soluble sugar and proline in ‘Jinba’.

Fig. 5
figure 5

DEGs involved in plant osmotic substances. a Expression analysis of DEGs related to signal transduction pathways of starch and sucrose metabolism. b Expression analysis of DEGs related to signal transduction pathways of proline metabolism. c Changes of soluble sugar in ‘Jinba’ after infected by F. oxysporum. d Changes of proline in ‘Jinba’ after infected by F. oxysporum

Full size image

TF analysis

TFs have attracted much attention in the plant response to pathogen infection. In this study, 38 WRKY, 32 MYB, 42 NAC, 58 ERF, 15 ARF, and 32 bHLH genes were identified in the 3 and 72 h co-expressed DEGs (Fig. 6a). WRKY family members often play an important role in plant disease resistance pathways [25]. To further investigate the mechanism of WRKY response to F. oxysporum in ‘Jinba’, we analyzed the WRKY-target genes containing the “Plant-pathogen interaction” pathway (Additional file 2: Table S2). Among these WRKYs, evm.model.scaffold_62.16 had the highest homology score with CmWRKY6. Song et al. (2014) found that CmWRKY6 is involved in the response to F. oxysporum infection in chrysanthemums [26]. Their conserved structural domains were highly similar comparing the protein sequences of evm.TU.scaffold_62.16 and CmWRKY6 (Additional file 3: Fig. S2). At the same time, we mapped the association network between evm.TU.scaffold_62.16 and target genes containing the “Plant-pathogen interaction” pathway (Fig. 6b). We found that evm.TU.scaffold_62.16 with CNGC (evm.TU.scaffold_1028.190) and CPK (evm.TU.scaffold_1621.172, evm.TU.scaffold_875.90, evm.TU.scaffold_490.149 and, evm.TU.scaffold_2891.82) have a high potential target relationship.

Fig. 6
figure 6

Transcription Factor Analysis. a Transcription factor heat map analysis. b Targeting analysis between evm.TU.scaffold_62.16 and Pathogen-Related Genes. The red circle indicates evm.TU.scaffold_62.16, purple circles indicate target genes, and green triangles represent high-scoring target genes

Full size image

Confirmation of DEGs using qRT-PCR

To confirm the reliability of the RNA-seq data, we verified the expression of DEGs using qRT-PCR. Three DEGs related to secondary metabolism (evm.TU.scaffold_741.126, evm.TU.scaffold_1462.187, evm.TU.scaffold_1099.150), three DEGs related to starch and sucrose metabolism (evm.TU.scaffold_1269.627, evm.TU.scaffold_10689.19, evm.TU.scaffold_25.386), and three DEGs related to proline and arginine metabolism (evm.TU.scaffold_405.102, evm.TU.scaffold_11851.88, evm.TU.scaffold_953.54) were used for validation. The results showed that the qRT-PCR trends were generally consistent with the RNA-seq trends (Fig. 7). These results confirm the reliability and reproducibility of our data.

Fig. 7
figure 7

qRT-PCR validation of DEGs. Red for RNA-seq, purple for qRT-PCR. The different letters mean significant differences according to Duncan’s multiple range test at p < 0.05

Full size image

Discussion

Due to continuous cultivation, the frequent occurrence of Fusarium wilt disease in chrysanthemums has seriously affected their yield and quality. F. oxysporum is the main pathogen causing Fusarium wilt, but there is little research on the molecular mechanism of the chrysanthemum response to F. oxysporum infection. In this study, we inoculated ‘Jinba’ with F. oxysporum and found that root rot and leaf wilt occurred at 72 h of inoculation (Fig. 1d). In addition, there was a gradual increase in cell membrane permeability and MDA in ‘Jinba’ compared with the control (Fig. 1e), indicating that ‘Jinba’ had started to respond to F. oxysporum infection in vivo at the early stage of inoculation. To clarify the molecular mechanism of early response to F. oxysporum infection in chrysanthemum, we analyzed root samples from ‘Jinba’ at 0 h, 3 h, and 72 h after inoculation with F. oxysporum using RNA-seq. A total of 7985 DEGs were co-expressed between 3 h, and 72 h after inoculation. As our study aimed to mine genes for early response to F. oxysporum in chrysanthemum, these 7985 DEGs were candidates for our next analysis using other techniques. We found a significant enrichment of DEGs in response to moisture and biological stimuli, catalytic activity, plant-pathogen interaction, MAPK signaling pathway, starch and sucrose metabolism, and biosynthesis of secondary metabolites using GO and KEGG pathway analysis. We focused our analysis on DEGs in these pathways to provide new insights into the resistance of chrysanthemums to F. oxysporum infection (Fig. 8).

Fig. 8
figure 8

Analysis of the mechanism of ‘Jinba’ response to F. oxysporum infection

Full size image

Secondary metabolites of plants can act as defenses against plant stress [27, 28]. When infested, plants produce large amounts of phenolic compounds that inhibit the invasion of pathogenic bacteria [29]. For example, α-tomatine limits the growth of pathogens in the apoplast and saponins have strong antifungal activity [30, 31]. Coumaric acid, butyric acid, p-coumaric acid, trans-cinnamic acid, p-hydroxybenzamide, ferulic acid, methyl p-hydroxybenzoate, and 3-indolepropionic acid secreted by Arabidopsis thaliana roots inhibit the growth of F. oxysporum [32]. Phenylalanine metabolism is one of the most important pathways in plant secondary metabolism. PAL is a key enzyme in the phenylpropanoid pathway that catalyzes the conversion of phenylalanine to cinnamic acid and promotes the production of flavonoids, coumarins, and other secondary metabolites [33]. We found that when F. oxysporum inoculated ‘Jinba’, PAL gene expression and PAL enzyme activity levels in plants increased. ‘Jinba’ promoted the metabolism of phenylalanine in response to F. oxysporum. Caffeic acid O-methyltransferase (COMT) and caffeoyl-CoA-O-methyltransferase (CCoAOMT) are key enzymes regulating lignin biosynthesis [34]. POD and PPO are involved in the synthesis of phenols [35, 36]. Our study showed that COMT, CCoAOMT, PPO, and POD gene expressions were upregulated. More importantly, the enzymatic activity levels of PPO and POD were elevated. POD and PPO are also related defense enzymes in plants, which can eliminate oxidative damage by scavenging ROS [37]. The increase in their contents showed that the defense system of chrysanthemum played a role in the early stage of infection, further demonstrating that chrysanthemum responded to the infection of F. oxysporum in the early stage of infection. In conclusion, DEGs enriched to the “Biosynthesis of secondary metabolites” pathway involved in the interaction between chrysanthemum and F. oxysporum.

Osmotic stress occurs when F. oxysporum blocks vascular bundles in plants. When plants are subjected to osmotic stress, they regulate small organic molecules, such as proteins, polyamines, amino acids, betaine, and sugars in their bodies [24]. Soluble sugars can be involved in plant-pathogen interactions in many ways [38]. For example, when Botrytis cinerea infects Arabidopsis, it inhibits genes related to photosynthesis and promotes sucrose catabolism [39, 40]. When exposed to adverse environments, plants accumulate proline to increase stress tolerance [41]. In addition, Proline is involved in the redox and hypersensitive responses that recognize pathogens [42]. In the present study, we focused on DEGs related to two pathways: “Starch and sucrose metabolism” and “Proline and arginine metabolism”.

β-glucosidase (BGLU) catalyzes the hydrolysis of sugar-containing compounds to release β-D-glucose and the corresponding monosaccharides, oligosaccharides, or complex sugars [43]. We found that during the early stages of F. oxysporum infection, the expression of BGLU in plants decreased, leading to a reduction in the amount of glucose produced by catabolism in vivo. Sucrose phosphate synthase (SPS) and sucrose synthase (SS) are key rate-limiting enzymes that regulate sucrose synthesis [44, 45]. SS family genes and SPS were downregulated after inoculation with F. oxysporum. Further, the soluble sugar content in ‘Jinba’ gradually decreased after inoculation. We speculate that this may be a mechanism for self-protection in chrysanthemums, as glycogen in the plant can provide nutrients for pathogenic bacteria to infest the plant. When the chrysanthemum experiences invasion by F. oxysporum, it inhibits the rate of fungal invasion by reducing sugar synthesis.

In addition, the ALDH family encodes pyrroline-5-carboxylic acid synthetase (P5CS), which promotes proline synthesis in proline synthesis [46, 47]. We found that the expression of ALDH family genes decreased and then increased after inoculation with F. oxysporum. In addition, the proline content in vivo also decreased and then increased. This suggests that proline is involved in the chrysanthemum defense pathway in the early stages of F. oxysporum invasion.

In this study, the TFs that responded to F. oxysporum infection were mainly in the WRKY, MYB, NAC, ERF, ARF, and bHLH families. TFs, especially the WRKY family, are key members of disease-resistance signaling pathways in plants responding to biotic stress [48]. We found 38 WRKY co-expressed in the present study at 3 and 72 h after F. oxysporum inoculation. In addition, we identified a WRKY (evm.TU.scaffold_62.16) with high homology to CmWRKY6 and highly similar conserved structural domains. Song et al. (2014) found that CmWRKY6 responds to infection by F. oxysporum in chrysanthemums [26]. Meanwhile, we analyzed the targeting relationship between evm.TU.scaffold_62.16 and “Plant-pathogen interaction” pathway DEGs. We identified CPK as a potential target gene. CPK is involved in the plant immune system. In Arabidopsis, AtCPK1 provides broad-spectrum protection against pathogens by regulating the salicylic acid pathway [49]. In wheat, WRKY45-1 and CPK2 may have a common function in multiple disease-resistance pathways [50]. Therefore, evm.TU.scaffold_62.16 or its homologs identified in this study are key candidates for our subsequent analysis of chrysanthemum WRKY in response to F. oxysporum infection.

Conclusions

In conclusion, our study reveals the early response mechanism to F. oxysporum infection in chrysanthemum ‘Jinba’. Under the infection of F. oxysporum, the levels of MDA, POD, PAL, PPO, and total phenols in chrysanthemum increased in the early stages. Furthermore, with F. oxysporum infection, the soluble sugar content in chrysanthemum decreased, and the proline content increased. More importantly, in the early stages, WRKY, MYB, NAC, ERF, ARF, and bHLH were the main TFs responding to F. oxysporum infection. We also analyzed the targeting relationship between WRKY and “Plant-pathogen interaction” pathway DEGs to find the candidate WRKY (evm.TU.scaffold_62.16) for our subsequent study. The transcriptomic data and mechanisms revealed in this study can help alleviate the chrysanthemum’s monocropping obstacle, slow the occurrence of Fusarium wilt, and provide new management ideas for the industrial development of chrysanthemum.

Materials and methods

Experimental materials

The material for this experiment, chrysanthemum ‘Jinba,’ was provided by Chrysanthemum Germplasm Resource Preserving Center, Nanjing Agricultural University (Nanjing, China) and incubated in a 16 h/8 h (light/dark), a temperature of 25°C, and a humidity of 70% light-temperature chamber.

F. oxysporum A2 was isolated from cut flower chrysanthemum ‘Jinba’ plants from the experimental chrysanthemum base at Hushu, Nanjing Agricultural University [18, 19].

Identification of pathogenic bacteria, culture and inoculation

The isolated F. oxysporum A2 was inoculated onto PDA plates and incubated for 5 days at 28 °C. Next, ten cakes of approximately 0.7 cm in diameter were inoculated into 500 mL of PDB and incubated for 4 d at 28 °C, 170 r·min− 1 in a constant shaker. The spore morphology and size were observed under a microscope [18, 19]. The genomic DNA of the fungi was extracted using the Biospin Fungus Genomic DNA Extraction Kit, and the extracted DNA was amplified via PCR using primers ITS1F / ITS4 and sequenced [51]. NCBI-BLAST was used for sequence alignment, and Mega X software was used to construct a neighbor-joining phylogenetic tree.

The roots of 40-day-old cuttings of Chrysanthemum seedlings were cut with scissors, immersed in a 107 CFU ml− 1 spore suspension for 30 min, and planted in a 1:2 (v/v) mixture of soil and vermiculite.

Measurement of physiological indicators

Root samples were collected from plants at 0, 3, and 72 h after inoculation. In addition, enzyme activity in chrysanthemums was measured using POD, PAL, and PPO kits (Comin, Suzhou). In the meantime, the rest of the physiological data were determined using MDA, soluble sugars, proline, and total phenol extraction kits (Jiancheng, Nanjing). The experimental group was inoculated with F. oxysporum A2 and the control group was inoculated with PDB.

RNA extraction and transcriptome sequencing

Root samples were taken from plants at 0, 3, and 72 h after inoculation, and three biological replicates were used for each time point sample. Total RNA was extracted using TRIzol reagent kit (Invitrogen, Carlsbad, CA, USA) and assessed on an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA). After total RNA was extracted, eukaryotic mRNA was enriched by Oligo(dT) beads. mRNA short fragments using NEBNext Ultra RNA Library Prep Kit for Illumina (NEB #7530, New EnglandBiolabs, Ipswich, MA, USA) transcribed into cDNA. Purified double-stranded cDNA fragments were end repaired, A base was added, and ligated to an Illumina sequencing adapter. The resulting cDNA library were sequenced using Illumina Novaseq6000 (Gene Denovo Biotechnology Co., Guangzhou, China). The reads were further filtered using fastp [52]. Finally, paired-end clean reads were mapped to the reference genome using HISAT2 software. 2.4 [53].

The fragment per kilobase of transcript per million mapped reads (FPKM) value was calculated to quantify its expression abundance and variation using RSEM software [54]. The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession numbers (s) can be found in the NCBI BioProject database under the accession number PRJNA926886.

DEGs analysis

Correlation analysis was performed using R. Differential expression analysis was performed using DESeq2 [55]. FDR below 0.05 and log2 fold-change ≥ 1 were considered differentially expressed genes/transcripts.

GO and KEGG Enrichment annotation of DEGs

All DEGs were mapped to the Gene Ontology database (http://www.geneontology.org/) and the Kyoto Encyclopedia of Genes and Genomes (https://www.kegg.jp/) [56,57,58]. The KEGG related figures for this study have been obtained with permission from the KEGG Pathway database. Hypergeometric tests were applied to identify pathways and GO terms significantly enriched in differential genes compared to the background gene set with FDR ≤ 0.05 as the threshold value.

Screening of TFs, and analysis of the relationship between WRKY and the “Plant-pathogen interaction” pathway DEGs targeting

TFs were annotated using the PlantTFDB (http://planttfdb.gao-lab.org/) database, and statistics were based on TF family classification results. The JASPER database was used to obtain TF binding motif information, and the MEME FIMO software was used to predict transcriptional target genes. Network diagram was drawn to present the association between transcription factors and target genes using Omicsmart (http://www.omicsmart.com).

Candidate DEG validation using qRT-PCR

The primers for the candidate DEGs were designed using Primer Premier 5, and EF1α was used as the internal reference gene [59]. qRT-PCR was performed using the SYBR Green PCR Master Mix (TaKaRa). qRT-PCR analysis was performed using a Roche LightCycler 96 fluorescence quantification instrument. Each sample contained three biological replicates and three technical replicates were performed for each biological replicate. PCR reactions were performed according to the following reaction conditions: 95 °C for 2 min, 95 ℃ 15 s, 55 ℃ 15 s, 72 ℃ 20 s, 40 cycles. Finally, the dissolution curve program was developed. The relative expression level of each gene was calculated using the formula 2−ΔΔCT [60].

Statistical analysis

Statistical analyses were performed using SPSS version 25.0. All data were analyzed using analysis of variance (ANOVA) and t-tests to determine significant differences.