Integrated transcriptomic and metabolomic analyses reveal flavonoid and lipid metabolic reprogramming in Dendrobiumofficinale during Colletotrichum fructicola-induced anthracnose

Fungal isolation and morphological characterization

Fungal isolation and morphological characterization were performed on D. officinale leaves exhibiting anthracnose symptoms collected from Wenzhou, Taizhou, Jinhua, and Hangzhou in Zhejiang Province. Samples were surface-sterilized by immersion in 75% ethanol for 30 s, followed by rinsing with sterile water three times, treatment with 0.1% mercuric chloride for 1 min with continuous shaking, and final rinses with sterile water three times. Sterilized tissues (5 mm × 5 mm) were excised from the junction of healthy and diseased areas and placed on potato dextrose agar (PDA) plates, which were incubated at 25 °C in the dark for 5 days. Pure cultures were obtained via monospore isolation. Conidial morphology was observed under a microscope, revealing colorless to brown conidiophores producing single-celled, oblong conidia.

Genomic DNA was extracted using a Plant Genomic DNA kit. For molecular identification, fragments of multiple phylogenetic markers were amplified by polymerase chain reaction (PCR). These included the internal transcribed spacer (ITS) region, β-tubulin (TUB), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), actin (ACT), calmodulin (CAL), and chitin synthase 1 (CHS-1) genes (Zhang, Li & Zhu, 2022). The primers used for amplification are listed in Table S3.

The PCR reaction mixture (25 µL) consisted of 2.5 µL of 10× PCR buffer with MgCl₂, 0.5 µL of dNTPs, 0.5 µL of Taq DNA polymerase, 0.5 µL of each forward and reverse primer, 0.5 µL of genomic DNA template, and 20 µL of ddH₂O. The same thermal cycling protocol was used for the amplification of all gene fragments: initial denaturation at 94 °C for 5 min; followed by 29 cycles of denaturation at 94 °C for 1 min, annealing at 50 °C for 40 s, and extension at 72 °C for 1 min; with a final extension at 72 °C for 10 min.

The amplified products were sequenced, and similarity analysis was conducted using NCBI BLAST. Multiple sequence alignments and a Maximum Likelihood phylogenetic tree were constructed in MEGA 11.0 with 1000 bootstrap replicates based on the concatenated dataset (ITS, TUB, GAPDH, ACT, CAL, CHS-1). The sequences of C. fructicola used for phylogenetic analysis are listed in Table S4.

Pathogenicity test

To fulfill Koch’s postulates, pathogenicity tests were conducted on healthy, one-year-old D. officinale plants. A conidial suspension (1 × 10⁶ spores/mL) of the representative C. fructicola isolate was prepared from a 7-day-old culture grown on PDA. Leaves were gently wounded with a sterile needle and inoculated by placing a 10 µL droplet of the suspension on each wound. Control plants were treated with sterile distilled water. All plants were covered with plastic bags to maintain high humidity and incubated in a greenhouse at 25 ± 2 °C. Symptoms typical of anthracnose, such as dark brown, sunken lesions, developed on inoculated leaves within 7n each wound. Control plants were treated with sterile distilled water. All plants were covered with plastic bag C. fructicola based on morphological and molecular characteristics, thereby fulfilling Koch’ s postulates.

Soluble sugar, reactive oxygen species, and phospholipase activity analysis

For the detection of soluble sugar content, frozen leaf samples (100 mg) were ground in liquid nitrogen and extracted with one mL of 50% ethanol at 90 °C for 30 min. After centrifugation at 12,000× g for 15 min, the supernatant was collected, and the ethanol was evaporated under vacuum. The residue was dissolved in distilled water, and the soluble sugar content was determined using the anthrone-sulfuric acid method. Briefly, 0.5 mL of the sample solution was mixed with two mL of anthrone reagent (0.2% anthrone in concentrated sulfuric acid), heated at 100 °C for 5 min, and the absorbance was measured at 620 nm using a spectrophotometer. Glucose was used as a standard to construct a calibration curve, and results were expressed as mg glucose equivalents per gram fresh weight.

For reactive oxygen species (ROS) detection using the nitroblue tetrazolium (NBT), infected and control leaf samples (50 mg) were excised and incubated in 0.1 mM NBT solution for 10 min in the dark at 25 °C. Following incubation, leaf samples were dipped in 95% ethanol and heated at 80 °C for 5 min to remove chlorophyll. After cooling to room temperature, the decolorized discs were photographed to visualize blue formazan deposits indicative of superoxide anion (O₂^•−) accumulation. For quantitative analysis, the formazan product was extracted with two mL of dimethyl sulfoxide by shaking at 37 °C for 30 min. The extract was centrifuged at 10,000× g for 10 min, and the absorbance of the supernatant was measured at 700 nm using a spectrophotometer.

Phospholipase (PLA) activity was assayed using egg yolk lecithin as a substrate. Leaf samples (200 mg) were ground in liquid nitrogen and extracted with two mL of ice-cold 50 mM Tris–HCl buffer (pH 7.5) containing one mM EDTA and 1% polyvinylpyrrolidone. The extract was centrifuged at 10,000× g for 10 min at 4 °C, and the supernatant was used as the enzyme source. The reaction mixture (one mL) contained 50 mM Tris–HCl (pH 7.5), 5 mM CaCl₂, 1% (w/v) egg lecithin, and 100 µL of enzyme extract. After incubation at 37 °C for 30 min, the reaction was terminated by adding two mL of chloroform/methanol (2:1, v/v). The mixture was vortexed and centrifuged at 3,000× g for 10 min, and the organic phase was collected. The released fatty acids were quantified by measuring the absorbance at 234 nm after reacting with copper sulfate and 2,2’-bipyridyl. One unit of PLA activity was defined as the amount of enzyme releasing 1 µmol of fatty acid per minute at 37 °C.

Transcriptome analysis

Total RNA was extracted from 100 mg of frozen infected leaves using the TRIzol reagent (Invitrogen) according to the manufacturer’s protocol. Methods for RNA quantity, cDNA library construction, and transcriptomic analysis were performed according to our previous work (Zhan et al., 2022a). Briefly, RNA quality and quantity were assessed using a NanoDrop 2000 spectrophotometer. Libraries were sequenced on an Illumina HiSeq platform using 150 bp paired-end reads. Raw reads were filtered to remove adaptor sequences, low-quality reads (Q-score <20), and reads with >10% unknown nucleotides using fastp. Subsequently, the clean reads were mapped to the reference genome of D. officinale (Zhang et al., 2016). Differential gene expression analysis was then carried out using DESeq2 (version 1.30.1), with the criteria set as —log₂(fold change)— ≥ 1 and adjusted p - value <0.05. The raw of RNA-seq data of C. fructicola infected plants has been submitted to the BIG Data Center of the Chinese Academy of Sciences (https://bigd.big.ac.cn) with accession number CRA002691 (Zhan, Qian & Mao, 2022).

Metabolome analysis

The infection leaves were ground into a fine powder in liquid nitrogen from two treatment groups (CK and IN). The methods of metabolites extraction, LC-MS/MS detection and metabolomics analysis were performed according to our previous study (Zhan et al., 2020). Briefly, Metabolite extraction was carried out by using a methanol:water mixture (50:50, volume/volume). The resultant extracts were then analyzed via an Orbitrap Exploris 240 Mass Spectrometer (manufactured by Thermo Fisher Scientific), which was coupled with a Vanquish Horizon UHPLC system (also from Thermo Fisher Scientific). The mobile phase consisted of 0.1% formic acid in water (A) and 0.1% formic acid in acetonitrile (B), with a gradient program: 5% B (0–1 min), 5%–95% B (1–14 min), 95% B (14–16 min), 95%–5% B (16–16.1 min), and 5% B (16.1–20 min) at a flow rate of 0.3 mL/min. The mass spectrometer operated in both positive and negative ion modes with a spray voltage of 3.2 kV (positive) and 2.8 kV (negative), capillary temperature of 320 °C, and sheath gas flow rate of 35 arb. Full-scan MS spectra were acquired from m/z 100–1,500 at a resolution of 60,000, and data-dependent MS/MS scans were performed at 15,000 resolution.

Raw data were processed using Thermo Fisher Scientific Compound Discoverer 3.4, including peak detection, alignment, and annotation. Differential metabolite analysis was conducted via partial least squares discriminant analysis (PLS-DA) using the R, with variables having variable importance in projection (VIP) >1 and p-value <0.05 considered significant. Metabolites were identified by matching accurate masses and MS/MS spectra against the self-established database and the Kyoto Encyclopedia of Genes and Genomes (KEGG). Hierarchical clustering and principal component analysis (PCA) were performed using the R packages pheatmap and FactoMineR to visualize metabolic profiles.

Statistical analysis

The data are displayed as means ± standard deviations (SD). Data set were treated by hierarchical clustering using the R package pheatmap (v1.0.12) and by PCA using the R package FactoMineR (v2.6). Gene enrichment analysis was used in the R package clusterProfiler (v4.2.2) (Yu et al., 2012).

Pathogen isolation and identification

The symptoms of anthracnose on D. officinale were initially observed in Wenzhou, Zhejiang Province. The disease manifests as black spots on leaves, which progress to dark brown lesions with irregular margins and a tan halo (Fig. 1A). To identify the causal pathogen, infected leaf tissues were surface-sterilized, and then incubated on potato dextrose agar (PDA) at 25 °C for 5 days under dark conditions. Five morphotypic groups (Den1–Den5) were obtained and purified through single-spore isolation, based on colony characteristics (size, color, mycelial morphology), edge traits, and spore morphology (Fig. 1B). To confirm the pathogenicity of these five groups and identify the causal agent, pathogenicity tests strictly following Koch’s postulates were conducted. For each group (Den1–Den5), conidial suspensions (1 × 10⁵ conidia/ml) were prepared and inoculated onto the leaves of one-year-old D. officinale seedlings without causing mechanical injury; sterile water was used as the control. Each treatment included six biological replicates. All inoculated plants were covered with plastic bags to maintain high humidity and placed in a greenhouse at 25 °C with a 12 h photoperiod. At 25 days post-inoculation (dpi), only seedlings inoculated with the Den3 group exhibited lesions consistent with the field symptoms, including leaf discoloration and wilting (Fig. 1C). In contrast, seedlings inoculated with Den1, Den2, Den4, Den5, and the control group remained asymptomatic. To confirm that all strains within Den3 were pathogenic and conspecific, six isolates were randomly selected from the 87 Den3 strains for supplementary pathogenicity testing (following the same protocol as above, six biological replicates per isolate). All six tested Den3 isolates induced identical typical anthracnose symptoms at 25 dpi, confirming the pathogenicity of Den3 strains as a whole (Fig. S1).

For Koch’s postulate validation, pathogen re-isolation was performed from the lesions of Den3-inoculated seedlings. The re-isolated strains showed identical phenotypic characteristics to the original Den3 group. Microscopic examination of the Den3 group showed conidiophores ranging from colorless to brown, producing single-celled, oblong conidia (Fig. 1E). Molecular identification of both the original and re-isolated Den3 strain was conducted using internal transcribed spacer (ITS), β-tubulin, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), actin (ACT), calmodulin (CAL), and chitin synthase 1 (CHS1) gene sequences; the results of the amplification, sequencing, and homology analysis of these markers revealed that the isolate is C. fructicola (Fig. 1F; Fig. S1). No fungi were recovered from the leaves of control seedlings. Additionally, quantitative analysis revealed that the Den3-inoculated sites had significantly elevated ROS levels compared to the control tissues (Fig. 1D). Together, these findings confirm C. fructicola as the causative agent of anthracnose in D. officinale and establish the foundation for subsequent transcriptomic and metabolomic analyses of host-pathogen interactions.

Transcriptome and metabolome analysis

To comprehensively elucidate the molecular and metabolic responses of D. officinale to C. fructicola infection, transcriptome and metabolome analyses were performed on infected and control leaves at 15 dpi. At this time point, significant increases in ROS and soluble sugar content were observed in infected tissues (Fig. 2), indicating a robust physiological response to pathogen invasion. RNA sequencing yielded high-quality clean reads ranging from 20.13 Gb to 21.89 Gb across all samples (Table S1). PCA revealed a clear separation between control (CK) and infected (IN) groups, with the first two principal components (PC1 and PC2) accounting for 47.18% and 22.49% of the total variance, respectively (Fig. 3A). Differential gene expression analysis identified 6,934 differentially expressed genes (DEGs) in IN versus CK, including 3,105 significantly upregulated and 3,829 significantly downregulated genes (Fig. 3B).

Moreover, a total of 1,275 metabolites were detected by nontargeted metabolomics analysis (Table S2). The largest number of metabolites were classified into flavonoids, phenolic acids, and lipids (Fig. S2). PCA of metabolomic data further confirmed distinct metabolic profiles between CK and IN groups, with PC1 and PC2 explaining 45.54% and 23.95% of the variance, respectively (Fig. 3C). PLS-DA identified 210 significantly increased and 176 significantly decreased metabolites in IN versus CK, based on VIP >1 and p-value <0.05 (Fig. 3D).

KEGG pathway enrichment analysis of the transcriptome and metabolome

KEGG enrichment analysis of DEGs and differentially accumulated metabolites (DAMs) showed some of the same pathways, such as ‘flavonoid biosynthesis’ and ‘diterpenoid biosynthesis’ (Fig. 4). However, terpenoid metabolism was non-significant pathway in enrichment analyses, the p-values of most of them were greater than 0.05 (Figs. 4A and 4B). Based on hierarchical clustering, DEGs and DAMs of the ‘flavonoid biosynthesis’ pathway were divided into two clusters (Figs. S2 and S3). Among them, two dihydroflavonol-4-reductase (DFR) genes significantly increased in M-IN vs. M-CK, including LOC110101655 (16.1-fold) and LOC110111528 (14.8-fold). Leucoanthocyanidin dioxygenase (LDOX) genes were upregulated 9.2-fold in M-IN vs. M-CK (Fig. S3). Eleven flavonoids were significantly increased and five glycosylated flavonoid derivatives were identified in the M-IN group (Fig. S4). These results suggest that flavonoid biosynthesis pathway plays important roles in host-pathogen interactions.

Anthracnose affects products of lipid turnover

To investigate the impact of C. fructicola infection on lipid metabolism in D. officinale, we integrated transcriptomic and metabolomic analyses. Although no lipid-related pathways were significantly enriched in KEGG analysis of DEGs, Gene Set Enrichment Analysis (GSEA) revealed that lipid catabolism pathways were highly active in infected tissues (IN vs. CK) (Fig. 5A). This discrepancy highlights the limitations of predefined pathway annotations in capturing subtle but biologically relevant gene set enrichments. Specifically, 53 genes involved in lipid degradation were upregulated, while 25 were downregulated in IN versus CK (Fig. 5B). Metabolite profiling revealed significant reductions in PC and PE levels in infected leaves, alongside decreased free fatty acid content compared to control tissues (Fig. 5C). The conversion of PC to LPC was associated with changes in fatty acid pools and lipid signaling pathways (Fig. 5D). Key genes involved in lipid metabolism, including long-chain acyl-CoA synthetase (LACS) (22.5-fold upregulation), peroxisomal 2,4-dienoyl-CoA reductase (DECR) (22.2-fold upregulation), and 2-hydroxyacyl-CoA lyase (HACL) (1,015.6-fold upregulation), were significantly induced in infected tissues. These genes are central to fatty acid activation, β-oxidation, and peroxisomal lipid metabolism, indicating an active shift toward lipid catabolism during pathogen invasion. Additionally, the fatty acyl-ACP thioesterase B (FATB) gene, which mediates 16:0 fatty acid hydrolysis, was upregulated 17.9-fold in IN versus CK, further supporting the hypothesis that lipid degradation is a key metabolic response to C. fructicola infection.

The impact of pathogen infection on the lipid metabolism of the host is extremely complex. To dissect the mechanism underlying lipid turnover during infection, we measured phospholipase activity and gene expression dynamics at 5, 15, and 25 dpi. Phospholipase activity in infected leaves peaked at 15 dpi, showing a 2.1-fold increase compared to healthy controls (Fig. S5A). qPCR analysis of phospholipase genes (PLA) revealed distinct expression patterns between host and pathogen. Host PLA genes were upregulated at five dpi (2.9- fold) but declined thereafter (Fig. S5B), while pathogen-derived PLA2 transcripts remained elevated throughout the infection cycle (5–25 dpi) (Fig. S5C). This suggests that the initial surge in phospholipase activity at five dpi is host-driven, potentially as part of a defense response, whereas sustained activity at later stages (15–25 dpi) is primarily mediated by pathogen-secreted enzymes.