Sweetpotato (Ipomoea batatas) is a vital food crop, rich in carbohydrates, fiber, and micronutrients, widely grown in Asia, Africa, and Latin America ^{[1, 2]}. Sweetpotato is highly adaptable to poor soils and drought, making it essential for global food security ^[3]. Pests and diseases present significant challenges to the sweetpotato industry, causing major economic losses. Pathogens impair plant growth and reproduction by altering resource use, metabolism, and physiology, extracting nutrients from the host ^[4]. In China, 23 infectious diseases, including 14 fungal ones, affect sweetpotato. A major disease, Sweetpotato Black Rot caused by Ceratocystis fimbriata, targets roots and seedlings ^[5–10]. The disease spreads via infected seed potatoes, soil, and tools, especially in warm, humid environments ^{[9, 10]}. Black spot disease causes black spots on roots, internal rot, and significant yield loss, reducing market value ^{[11, 12]}. Black spot disease can cause up to 50% yield loss without control measures ^{[13, 14]}.

Black spot disease is a major barrier to sweetpotato cultivation, necessitating effective control strategies. Research on controlling Black spot disease focuses on methods like perilla aldehyde and polyamine, which inhibit spore germination ^{[13, 14]}. Studies have cataloged effector proteins to understand the infection mechanism of C. fimbriata, aiding in effector-assisted breeding ^[15]. Research has identified key disease-resistant genes in sweetpotato, such as IbCHI (chitinase) and IbOLP1 (Osmotin-like proteins) ^{[11, 16]}. The regulatory mechanisms and biochemical functions of sweetpotato disease resistance need further investigation.

Plant growth and adaptation are driven by complex gene regulation and metabolic networks, with omics technologies providing powerful analytical tools. Metabolomics analyzes small molecular metabolites to assess plant physiological states under different growth stages or environmental stresses ^[17–21]. Transcriptomics analyzes gene expression patterns and how genes regulate cellular functions and processes ^{[22, 23]}. Metabolomics and transcriptomics are interrelated; combined analysis forms a gene-metabolite network, helping to understand how gene regulation affects metabolite synthesis and accumulation, distinguishing direct and indirect effects, and identifying key regulatory factors and core genes for functional verification ^{[21, 24]}. Joint analysis of metabolomics and transcriptomics improves data reliability, reveals biological networks, and accelerates functional gene identification ^[25–28]. The transcriptome and metabome are crucial for identifying key metabolites and genes involved in plant diseases. Studies on rice blast disease identified upregulated genes in the JA and SA signaling pathways, as well as metabolites like coumaric acid and alkaloids that inhibit disease ^[29]. Transcriptomic and metabolomic studies on drought-stressed sweetpotato show upregulation of drought-tolerant genes (IbMYB, IbNAC) and accumulation of proline and soluble sugars to improve drought tolerance ^[30]. Genome, transcriptome, and metabolite analysis of Darwin’s orchids showed that oxime synthesis is a result of convergent evolution, indicating that different species may achieve similar functions through different genetic mechanisms ^[31]. A study integrated metabolomics and transcriptomics to analyze the metabolic regulatory network of rice across its life cycle, identifying dynamic metabolite changes in tissues and discovering new regulatory factors ^[27]. "Guilty-by-association" in crop metabolism and gene regulation provides crucial resources and tools for precision agriculture and crop improvement ^[26–28]. Currently, no studies have used transcriptomics and metabolomics to investigate secondary metabolic mechanisms in sweetpotato infected with black spot disease.

Terpenes, the largest class of plant secondary metabolites, play roles in signaling, disease defense, environmental adaptation, and growth regulation ^[32–35]. Research on terpenes in sweetpotato has advanced, highlighting their roles in aroma, flavor, resistance, and nutrition, particularly their synthesis pathways and key enzymes. The patterns of disease-resistant genes and terpenes in sweetpotato are vital for understanding their resistance and functions.

This study proposes using comparative transcriptomics and metabolomics to analyze two sweetpotato varieties with different resistance to Black spot disease, aiming to identify key genes and metabolic markers. The study focuses on the MVA and MEP pathways of terpenoids, key enzymes involved in their synthesis, and accumulation patterns of terpenoids after Black spot disease infection. This will provide a theoretical foundation for exploring anti-Black spot disease genes and developing resistant sweetpotato varieties. It also provides a reference for exploring terpenes in sweetpotato.

Following certification by Xuzhou Academy of Agricultural Sciences, the SS23 sweetpotato cultivar (SS23) demonstrated strong resistance to dark spot disease, while the GS08 sweetpotato cultivar (GS08) exhibited disease susceptibility. Analytical data from the Agricultural Product Quality Testing Center (Ministry of Agriculture) revealed distinct dry matter contents: 25% in SS23 versus 40% in GS08. This diametric contrast in both disease resistance and biochemical composition formed the primary rationale for selecting these cultivars as experimental subjects.

At 6 days post-infection (dpi), both SS23 and GS08 displayed pronounced melanotic symptoms relative to uninfected controls (Fig. 1A). External melanotic lesions (indicated by yellow circles) demonstrated greater severity in GS08-6dpi compared to SS23-6dpi. Transverse section analysis revealed that internal melanotic lesions (marked by red circles) exhibited broader distribution in GS08-6dpi compared to SS23-6dpi, indicative of enhanced pathogen susceptibility in GS08. Quantitative assessment of epidermal melanotic macules (Fig. 1B) showed significant post-infection increases in lesion diameter for both cultivars. GS08-6dpi exhibited the maximal lesion width among infected specimens, with statistically significant differences observed compared to SS23-6dpi. Similarly, transverse section measurements (Fig. 1C) showed significantly greater melanotic lesion widths in infected specimens compared to controls across both cultivars. GS08-6dpi exhibited the maximal lesion width among infected specimens, with statistically significant differences observed compared to SS23-6dpi. The experimental data demonstrate greater susceptibility of GS08 to melanotic infection compared to SS23, as evidenced by more pronounced external and internal pathological manifestations. The experimental results are consistent with the identification results of Xuzhou Academy of Agricultural Sciences.

Principal Component Analysis (PCA; Fig. 2A) demonstrated clear segregation of experimental cohorts, with distinct clustering patterns reflecting differential metabolic profiles across genotypes. LC-MS-based metabolomic analysis revealed 1,892 metabolites organized within a three-tiered taxonomic hierarchy comprising 8 superclasses, 58 classes, and 244 subclasses (Fig. 2B). This taxonomic architecture underscores the metabolic sophistication underlying Ipomoea batatas pathogen response mechanisms. Venn diagram analysis (Fig. 2C) delineated differentially accumulated metabolites (DAMs) distribution patterns across four comparative groups, revealing sixteen conserved DAMs common to all comparisons that represent core regulatory elements in the infection response network. Hierarchical clustering analysis (Fig. 2D) demonstrated infection-induced metabolic pattern divergence, with colorimetric gradients (red: upregulation; blue: downregulation) quantitatively mapping metabolic flux directions. K-means clustering identified nine characteristic metabolite modules (Fig. 2E) with differential accumulation kinetics among experimental cohorts, elucidating pathway-level regulatory mechanisms governing disease resistance phenotypes. The integrated metabolomics approach uncovered genotype-specific metabolic remodeling during pathogen challenge in sweetpotato, establishing a framework for decoding the molecular basis of differential disease resistance.

We performed RNA-seq analysis to investigate gene expression in response to black spot stress. Sequencing data quality metrics (Q20:97.85%, Q30:93.56%, GC: 45.44%; Table S1) validated analytical reliability. Post-filtering (≈ 40M reads/sample; Table S2), reference genome alignment (Ipomoea batatas cv. Taizhong No.6) achieved 83% mapping rate, ensuring analytical fidelity. Volcano plot analysis (Fig. 3A) delineated constitutive transcriptional divergence: GS08-CK vs SS23-CK exhibited 1,532 DEGs (|log2FC|>1, FDR < 0.05), revealing baseline genotypic variation. Post-infection comparison (GS08-6dpi vs SS23-6dpi) identified 2,879 genotype-specific DEGs, demonstrating differential stress adaptation mechanisms. GS08 comparison (CK vs 6dpi) revealed 3,412 infection-responsive DEGs, indicating pathogen-induced transcriptional remodeling. SS23 displayed enhanced transcriptional plasticity (4,619 CK vs 6dpi DEGs), surpassing GS08's response magnitude by 35.4%. Venn analysis (Fig. 3B) delineated conserved (977) and comparison-specific DEG clusters across experimental paradigms. The core conserved stress-responsive regulon (977 genes) spanned all experimental conditions. SS23-specific infection-responsive signature (3,619 unique DEGs) confirmed its superior transcriptional adaptability during pathogenesis. The distribution of unique and overlapping DEGs suggests that both genotype and infection significantly influence gene expression. GO functional annotation (Fig. 3C) stratified DEGs into three ontological domains: biological processes (BP), molecular functions (MF), and cellular components (CC). Pathogen-responsive BP terms dominated infected specimens, including defense signaling (GO: 0006952), immune activation (GO: 0002253), and stress acclimation (GO: 0009628). Stress-induced MF enrichment featured metabolic reprogramming (GO: 0008152), ATPase activity (GO: 0016887), and structural maintenance (GO: 0005198). CC subcellular localization patterns indicated infection-driven membrane reorganization (GO: 0016020), cytoskeletal dynamics (GO: 0005856), and organelle trafficking (GO: 0046907).

Integrated transcriptomics uncovered genotype-specific transcriptional plasticity, with SS23 exhibiting 2.1-fold greater differential expression magnitude versus GS08. The conserved stress regulon and activated pathway modules elucidate the molecular basis of differential resistance in sweetpotato.

Integrated multi-omics analysis was performed across four experimental contrasts (Fig. 4) to delineate genotype-specific transcriptome-metabolome coordination patterns. Baseline comparison (GS08-CK vs SS23-CK; Fig. 4A) identified 55 shared multi-omics features, with 46 transcriptome-specific and 28 metabolome-specific signatures. This demonstrated moderate transcriptome-metabolome coupling (r = 0.62, p < 0.01) under baseline physiological states.GS08's infection response (CK vs 6dpi; Fig. 4B) revealed 56 shared biomarkers, with transcriptomic perturbations (n = 53) significantly exceeding metabolic alterations (n = 18) (χ²=21.4, p < 0.001). This indicates pathogen-induced transcriptional cascade amplification in GS08, with only 33.9% (19/56) of DEGs showing metabolite-level manifestations. Genotypic divergence under infection (GS08-6dpi vs SS23-6dpi; Fig. 4C) showed substantial intersection (n = 62) with 45 transcriptomic and 22 metabolomic discriminators. The 62-node interaction network (FDR < 0.05) manifested genotype-dependent infection responses through coordinated molecular phenotypes. SS23's infection response (CK vs 6dpi; Fig. 4D) exhibited the largest intersection size (n = 66), with 46 transcriptional regulators and 19 metabolic effectors. This tight coupling (ρ = 0.79 vs GS08's ρ = 0.54) demonstrates SS23's enhanced metabolic implementation of transcriptional commands during pathogenesis. Differential co-regulation dynamics (F_(condition) = 38.2, F_(genotype) = 41.6) revealed genetic background × environment interactions shaping molecular phenotype integration. SS23 displayed superior systems-level integration (covariation index CI = 0.71 vs GS08's 0.58), particularly during infection, indicating enhanced regulatory network fidelity. Multi-omics integration uncovered fundamental divergence in molecular strategy: SS23 employs direct genotype-metabolotype coupling whereas GS08 utilizes post-translational buffering mechanisms. The SS23-specific linear transcriptional control cascade (r²=0.82) contrasts with GS08's non-linear metabolic homeostasis maintenance (r²=0.31), defining distinct evolutionary adaptation strategies.

KEGG enrichment analysis (Fig. 5) demonstrated significant associations between transcriptomic and metabolomic variations. Multiple pathways exhibited coordinated transcriptional and metabolic regulation, suggesting their involvement in genotype-specific differentiation and stress adaptation. Baseline comparison (GS08-CK vs SS23-CK; Fig. 5A) revealed pathway enrichment in flavonoid biosynthesis, phenylpropanoid metabolism, and arginine/proline metabolism, accompanied by substantial differentially expressed genes (DEGs) and differentially accumulated metabolites (DAMs). These results indicate that secondary metabolite pathways underlie metabolic divergence between GS08 and SS23 genotypes. In GS08's infection response (CK vs 6dpi; Fig. 5B), amino acid-, lipid-, and carbohydrate-related metabolic pathways showed marked dysregulation, indicating significant metabolic remodeling upon pathogen challenge. The abundance of DEGs in these pathways implies transcriptional reprogramming in GS08 under stress conditions, likely modulating pathogen defense-related metabolites. SS23's infection response (CK vs 6dpi; Fig. 5D) showed pronounced enrichment in plant hormone signaling, starch/sucrose metabolism, and cyanoamino acid metabolism, indicating SS23's strategic mobilization of hormonal regulators and carbohydrate reserves to combat infection. Defense-associated secondary metabolism pathways showed marked enrichment, featuring elevated DEG/DAM quantities, which reflects tightly synchronized transcriptional-metabolic coordination. Genotypic divergence under infection (GS08-6dpi vs SS23-6dpi; Fig. 5C) identified flavonoid biosynthesis, phenylalanine metabolism, and terpenoid biosynthesis as the most prominently enriched pathways. These findings demonstrate distinct infection-response strategies: SS23 preferentially activates secondary metabolite-driven defenses, whereas GS08 predominantly employs primary metabolic restructuring.

This integrated KEGG analysis elucidates critical metabolic and transcriptional networks underlying genotype-specific and infection-triggered responses in sweetpotato. The observed genotypic differences between GS08 and SS23 imply divergent stress tolerance mechanisms, where SS23 displays superior metabolic coordination in defense strategies compared to GS08.

Sweetpotato employs two distinct terpenoid biosynthesis pathways: the cytosolic mevalonate (MVA) pathway and plastid-localized methylerythritol phosphate (MEP) pathway (Fig. 6A). Key MVA pathway enzymes - including acetyl-CoA acetyltransferase (AACT), hydroxymethylglutaryl-CoA synthase (HMGS), and hydroxymethylglutaryl-CoA reductase (HMGR) - demonstrated upregulated expression under specific conditions, indicating their catalytic dominance in terpenoid precursor biosynthesis. The MEP pathway displayed condition-dependent transcriptional regulation, with genes encoding 1-deoxy-D-xylulose 5-phosphate synthase (DXS), DXR reductase-isomerase (DXR), and HDS synthase exhibiting differential expression patterns that correlate with distinct physiological demands. Post-IPP/DMAPP synthesis stages involve three key synthases: geranyl diphosphate synthase (GPPS), farnesyl diphosphate synthase (FPPS), and geranylgeranyl diphosphate synthase (GGPPS) (Fig. 6B), which channel precursors into specific terpenoid classes. GPPS exhibited pronounced transcriptional activation under monoterpene-inducing conditions, confirming its rate-limiting function in C₁₀ terpenoid production. FPPS expression strongly correlated with sesquiterpene (C₁₅) accumulation, whereas GGPPS transcriptional patterns aligned with diterpene (C₂₀) biosynthesis, establishing precursor-product relationships. Terpene synthase (TPS) genes demonstrated stimulus-responsive expression divergence: monoterpene (mTPS), sesquiterpene (sTPS), and diterpene synthases (diTPS) displayed unique activation patterns corresponding to specific environmental or developmental signals. Coordinated transcriptional regulation across MVA/MEP pathways and TPS subfamilies reveals a multi-tiered control system governing terpenoid diversification in sweetpotato. Gene expression plasticity reflects genotype-environment interactions that fine-tune metabolic outputs for either ecological adaptation or specialized metabolite biosynthesis. Targeted metabolomic profiling (Fig. 7) uncovered substantial quantitative differences in terpenoid accumulation patterns among experimental groups. Six terpenoid subclasses were quantified: sesquiterpenoids (C₁₅), monoterpenoids (C₁₀), diterpenoids (C₂₀), triterpenoids (C₃₀), carotenoids (C₄₀), and derivatives, each demonstrating unique spatiotemporal accumulation dynamics. Sesquiterpenoid metabolites (plumericin, ajugol, catalpol) showed peak accumulation in genotype-specific contexts, with plumericin attaining maximum levels in SS23-6dpi accessions. The preferential accumulation of ajugol and catalpol in GS08-6dpi and SS23-6dpi genotypes implies their dual function in stress mitigation and genetic background-dependent metabolism. While monoterpenoids and diterpenoids displayed moderate fluctuation, elevated Brucine H and GGPP levels in particular accessions suggest their functional importance in terpenoid metabolic flux redirection. The differential accumulation of bioactive diterpenoid tanshinone I across genotypes highlights genetic determinants influencing specialized metabolite production. Triterpenoids 3-deacetylkhivorin and musaroside emerged as dominant metabolites showing genotype-dependent accumulation patterns, potentially contributing to accession-specific adaptive traits. The consistent detection of carotenoid derivative 4-ketolutein F across treatments underscores the metabolic versatility of terpenoid biosynthesis pathways. The observed metabolite accumulation gradients demonstrate genotype-treatment interactions that shape terpenoid biosynthesis trajectories. Sesquiterpenoids displayed the most pronounced metabolic plasticity, consistent with their dual roles in stress adaptation and developmental phase transitions. Our metabolomics approach delineates dynamic terpenoid remodeling in sweetpotato, particularly highlighting sesquiterpenoids as key metabolic responders to environmental challenges. The genotype-dependent metabolite signatures point to specialized transcriptional regulators that orchestrate terpenoid biosynthesis, providing mechanistic insights into stress resilience and secondary metabolic engineering.

We selected some genes for Quantitative reverse transcription PCR (qRT-PCR) analysis (Fig. 8) to verify the reliability and authenticity of transcriptome data. The significant upregulation of IbHMGR-g489 in pathogen-challenged GS08 (GS08-6dpi) indicates enhanced mevalonate (MVA) pathway activation during infection response. Co-upregulation of IbMDD-g25746 and IbIDI-g60768 in both infected genotypes demonstrates their conserved function in isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP) biosynthesis during pathogen-induced stress (Fig. 8). SS23-6dpi exhibited stronger upregulation of sesquiterpene synthase genes IbTPS-g21052 and IbTPS-g21061 than GS08-6d, revealing genotype-dependent regulation of C₁₅ terpenoid biosynthesis. GS08-6dpi showed preferential activation of IbTPS-g21081 (monoterpene synthase) and IbTPS-g5935 (diterpene synthase), specifying C₁₀ and C₂₀ terpenoid production pathways (Fig. 8). Elevated expression of cytochrome P450 genes IbCYP-71-g21041 and IbCYP-SF-g21049 in SS23-6d suggests their enzymatic role in terpenoid structural diversification through oxygenation and functional group modifications (Fig. 8). The upregulation of IbCYP-71-g21059 and IbCYP-SF-g21073 in GS08-6dpi implies their involvement in generating stress-adaptive terpenoid derivatives through specialized modification reactions (Fig. 8). SS23 displayed more pronounced transcriptional induction of terpene synthase (TPS) and cytochrome P450 (CYP) gene families under infection, establishing its superior capacity for defensive terpenoid metabolism during biotic stress. GS08 prioritized MVA pathway activation through upregulated IbHMGR-g489 and IbMDD-g25746, implementing a precursor-accumulation strategy to fuel terpenoid biosynthesis under pathogen pressure. The qRT-PCR validation of gene expression profiles showed strong correlation with transcriptomic variation patterns, confirming the technical reproducibility and biological validity of RNA-seq datasets (Figs. 7 and 8). SS23's coordinated induction of downstream TPS and CYP enzymes enhances defensive terpenoid chemodiversity through structural diversification mechanisms. Conversely, GS08's preferential upregulation of upstream MVA genes reveals an alternative metabolic prioritization focused on precursor pool expansion rather than downstream diversification. These regulatory divergences underscore genotype-dependent metabolic adaptation strategies in sweetpotato, with distinct evolutionary solutions for biotic stress resistance.

Terpenoids, as characteristic volatile metabolites, were extracted from sweetpotato storage roots using solid-phase microextraction (SPME) followed by gas chromatography-mass spectrometry (GC-MS) analysis. Integrated analysis of Fig. 9 and Supplementary table 7 revealed 28 volatile compounds in GS08 (22 terpenoids), compared to 18 compounds (9 terpenoids) in SS23. Figure 9A demonstrates that healthy GS08 storage roots contained abundant terpenoids. Post-infection with dark spot disease, six additional compounds were identified, including four terpenoids. Figure 9B indicates SS23 storage roots initially contained four fewer terpenoids than GS08. Following infection, 13 new compounds emerged, with four terpenoids identified. Figure 9C compares post-infection metabolite levels between varieties: GS08 showed significantly higher concentrations in 7 of 9 detectable compounds, with 2 compounds exhibiting lower levels compared to SS23. Five terpenoids - including (Z)-/(E)-linalool oxides, linalool, α-terpineol, and (R)-isocarvestrene - showed significant accumulation in GS08, contrasting with isoterpinolene dominance in SS23. The disease-resistant SS23 variety synthesized 13 novel volatile compounds post-infection compared to 6 in susceptible GS08, suggesting enhanced pathogen response through rapid metabolic reprogramming. Persistent accumulation of specific volatiles in GS08 suggests compensatory biosynthesis mechanisms counteracting its inherent susceptibility through prolonged metabolite production. An integrated analysis combining transcriptomic sequencing (Fig. 7), qRT-PCR validation (Fig. 9), and terpenoid metabolomic profiling (Figs. 8, 10) was conducted. These findings establish a framework for investigating terpenoid biosynthesis regulation in plant-pathogen interactions.

Sweetpotato Black spot disease, caused by C. fimbriata, significantly affects global sweetpotato production. The disease causes root rot in sweetpotato, reduces yield, and negatively impacts market value and storage stability ^{[36, 37]}.Therefore, detailed investigation into the effects and molecular mechanisms of sweetpotato Black spot disease is essential for improving sweetpotato production and breeding disease-resistant varieties. In this study, a combined transcriptomic and metabolomic analysis was employed to investigate the immune response mechanisms of sweetpotato roots following infection by C. fimbriata.

Besides differences in terpenoid biosynthesis and metabolism, amino acid metabolism also significantly contributes to plant resistance in GS08 and SS23 following infection by Black spot disease. Amino acids not only serve as alternative energy sources but also function as signaling molecules that regulate plant defensive responses ^{[38, 39]}.Non-targeted metabolomics analysis indicated a significant increase in leucine, valine, and proline levels in SS23 after infection (Table S3). These amino acids are closely associated with plant disease resistance ^[24]. In contrast, GS08 exhibited decreased glutamine and aspartic acid levels after infection (Table S3). This reduction could disrupt nitrogen metabolic balance and thus weaken disease resistance ^[25]. Comparative analysis of metabolic changes in SS23 and GS08 before and after infection revealed that the resistant variety SS23 rapidly activated amino acid metabolism and increased various amino acid levels (Table S3). Conversely, the susceptible variety GS08 exhibited limited capacity to regulate amino acid metabolism (Table S3). These results suggest that SS23 may enhance its disease resistance through metabolic reprogramming that promotes amino acid metabolism. In contrast, insufficient metabolic regulation makes GS08 more susceptible to pathogen infection.

Transcriptomic analysis was conducted to investigate gene expression changes in sweetpotato infected by C. fimbriata. Results revealed significant differences in the expression of multiple defense-related pathways between the resistant variety SS23 and the susceptible variety GS08. Overall, SS23 exhibited stronger transcriptional activation after infection, with significant upregulation of various defense-related genes, including pathogenesis-related proteins, peroxidases, and chitinases ^{[32, 33]}. In contrast, GS08 showed a weaker transcriptional response to these genes, indicating limited defense capacity under pathogen stress. In response to pathogen or pest invasion, plants activate a series of defense-related enzymes, including peroxidase (POD), phenylalanine ammonia-lyase (PAL), superoxide dismutase (SOD), and polyphenol oxidase (PPO), to strengthen their defense mechanisms ^[40–45]. The present study found greater upregulation of defense-related genes in SS23 after infection (Table S4), whereas GS08 exhibited a relatively weaker response (Table S5). These results indicate that SS23 experiences more pronounced changes in the activity of defense-related enzymes compared to GS08 under pathogen stress. By enhancing enzyme activities, SS23 effectively restricts pathogen proliferation and damage during the early stages of infection.

Terpenoids in sweetpotato demonstrate diverse bioactivities, with notably pronounced antioxidant, anti-inflammatory, anticancer, and antimicrobial properties ^{[46, 47]}. In this study, transcriptomic analysis identified key genes involved in the terpenoid biosynthetic pathway in sweetpotato. Changes in 50 terpenoid metabolites were detected using LC-MS-based non-targeted metabolomics and Volatile compound detection method (Figs. 7 and 9;Table S3 and S7). Among them, several terpene metabolites with relatively large differences and variations, such as sweroside ^[55], plumericin ^[56], linalool ^[57] and α-terpineol ^[58], all have significant antifungal activities. Terpenoid metabolites play critical roles in plant antifungal defense through diverse mechanisms. Multi-omics studies in maize ^[59], rice ^[60], pine ^[61], and apple ^[62] reveal terpenoids inhibit fungal growth via membrane disruption (e.g., selinene in maize) ^[59] and ROS-mediated cytotoxicity (α-farnesene in apple) ^[62]. They prime systemic resistance by activating JA signaling and synergizing with flavonoids. Resistant rice cultivars elevate sesquiterpenoid biosynthesis (e.g., phytocassanes) ^[60], while yeast-induced terpenoid accumulation in apples enhances Botrytis suppression ^[62]. Cross-species conservation underscores terpenoids' dual function as antimicrobial agents and immune amplifiers, offering targets for metabolic engineering in disease-resistant crop development. This study reports the inaugural detection of 50 terpenoid metabolites in sweetpotato. These findings significantly expand the documented phytochemical diversity in sweetpotato, establishing a foundational dataset for elucidating their organoleptic properties and defense mechanisms. The present study revealed a high correlation between expression changes of sesquiterpenoid synthase genes and corresponding metabolic alterations in sesquiterpenoids across four experimental groups. This finding provides essential data support for subsequent identification of disease-resistance-related sesquiterpene synthase genes and verification of their enzyme activities both in vivo and in vitro. Additionally, this research provides a robust theoretical basis for identifying further disease-resistance genes and advancing disease-resistant sweetpotato breeding programs.

This study utilized integrated transcriptomic and metabolomic analyses to investigate distinct molecular responses in two sweetpotato varieties, SS23 and GS08, infected by Ceratocystis fimbriata, the causal agent of black spot disease(Fig. 10). Phenotypically, GS08 exhibited greater susceptibility, developing more severe external and internal melanotic lesions. Metabolomic profiling revealed genotype-specific metabolic responses, emphasizing substantial differences in amino acid and terpenoid metabolism between the two varieties. SS23 exhibited more pronounced metabolic reprogramming, particularly increasing amino acids such as leucine, valine, and proline, which are essential for disease resistance. Transcriptomic analysis revealed that SS23 strongly activated defense-related pathways, including peroxidases, chitinases, and pathogenesis-related proteins, whereas GS08 exhibited a weaker transcriptional defense response.

Integrative analyses identified strong correlations between differentially expressed genes and differentially accumulated metabolites, particularly within terpenoid pathways. SS23 significantly activated downstream terpene synthase and cytochrome P450 genes, enhancing terpene diversity for defense, whereas GS08 predominantly upregulated upstream precursor synthesis genes. Comparative metabolomic profiling revealed significantly enhanced terpenoid biosynthetic capacity in SS23 cultivars post-infection, exhibiting higher activity than GS08 controls, with transcriptional dynamics of TPS and CYP450 genes demonstrating molecular concordance. These findings indicate that effective disease resistance in sweetpotato requires tightly coordinated transcriptional and metabolic reprogramming, with terpenoid metabolism playing a pivotal role. This study offers valuable insights for breeding sweetpotato varieties with improved resistance to black spot disease.

This study utilized two sweetpotato varieties with differing resistance to Black spot disease: SS23 (resistant) and GS08 (susceptible). Sweetpotato roots were supplied by the Institute of SweetPotato Diseases and Pests, Xuhuai Regional Academy of Agricultural Sciences, Jiangsu Province. The test roots were mature sweetpotato roots harvested four months after seedling transplantation. For inoculation preparation, the sweetpotato pathogen Ceratocystis fimbriata was pre-cultured on fresh potato dextrose agar (PDA) medium at 28℃ for approximately one month until mycelium and spores fully covered the plate. The spores were then rinsed with sterile water under aseptic conditions to obtain a spore suspension. A hemocytometer was used to determine the spore concentration, which was adjusted to 1×10⁶. A 1-mL sterile syringe was used to inject 10 µL of spore suspension into sweetpotato roots at a depth of approximately 2.5 cm. Each root was evenly inoculated at 16 sites. The control group received sterile water instead of the spore suspension. Following inoculation, the roots were incubated at 28℃ and 100% relative humidity for six days before sampling for metabolomic and transcriptomic analyses. After six days of incubation, samples were labeled as SS23_6dpi (SS23_6dpi_1, SS23_6dpi_2, SS23_6dpi_3) and GS08_6dpi (GS08_6dpi_1, GS08_6dpi_2, GS08_6dpi_3). Control samples were designated as SS23_ck (SS23_ck_1, SS23_ck_2, SS23_ck_3) and GS08_ck (GS08_ck_1, GS08_ck_2, GS08_ck_3). All 12 samples were immediately flash-frozen in liquid nitrogen and sent to Qingke Biological Technology Co., Ltd. (Beijing, China) for transcriptomic and metabolomic analyses.

LC-MS/MS analysis of non-polar metabolites was conducted using a Vanquish UHPLC system (Thermo Fisher Scientific) equipped with a Phenomenex Kinetex C18 column (2.1 mm × 100 mm, 2.6 µm) and an Orbitrap Exploris 120 mass spectrometer (Thermo). The mobile phase consisted of 0.01% acetic acid in water (phase A) and a 1:1 (v/v) mixture of isopropanol (IPA) and acetonitrile (ACN) (phase B). The column temperature was maintained at 25℃. The auto-sampler was set at 4℃, with an injection volume of 2 µL. The Orbitrap Exploris 120 mass spectrometer was operated in information-dependent acquisition (IDA) mode under the control of Xcalibur software (Thermo), enabling MS/MS spectrum acquisition. In this mode, the software continuously monitors the full-scan MS spectrum. The ESI source conditions were as follows: sheath gas flow rate of 50 Arb, auxiliary gas flow rate of 15 Arb, capillary temperature of 320℃, sweep gas of 1 Arb, vaporizer temperature of 350℃, full MS resolution of 60,000, MS/MS resolution of 15,000, collision energy set to stepped normalized collision energy (SNCE) at 20/30/40, and spray voltage of 3.8 kV (positive mode) or -3.4 kV (negative mode).

Tuberous root samples were transported on dry ice and sent to Tsingke Biotechnology Co., Ltd. (Beijing, China) for total RNA extraction and transcriptome sequencing. The extracted total RNA was evaluated for integrity, concentration, and purity to ensure it met the quality standards for library construction. After cDNA library construction and quality assessment, sequencing was performed using the Illumina NovaSeq platform. Raw data were filtered to remove low-quality sequences, including adapter reads and reads with more than 10% unknown bases (N), yielding high-quality clean data. Clean reads from each sample were aligned to the PMT reference genome using version 2.1.0 software ^[51] to determine the positional information of genes and specific sequence characteristics. Reads were assembled into transcripts based on their positional information using StringTie 1.3.4d ^[52]. A comparative transcriptome analysis was conducted on tuberous roots of "SS23" and "GS08" after six days of black spot infection, along with a control group, using high-throughput sequencing technology. Differentially expressed genes (DEGs) were identified using DESeq2 1.22.1 software ^{[53, 54]}, with gene expression quantified as fragments per kilobase of transcript per million mapped reads (FPKM). Log2 fold change Genes satisfying these criteria were classified as DEGs. The identified DEGs were annotated by comparing them against KEGG, GO, KOG, NR, Swiss-Prot, and Pfam databases. DEGs were further analyzed through KOG classification, GO functional enrichment, and KEGG pathway enrichment.

Samples were ground to a powder in liquid nitrogen.500 mg (1 mL) of the powder was transferred immediately to a 20 mL head-space vial (Agilent, Palo Alto, CA,USA) containing a NaCl-saturated solution to inhibit any enzyme reaction. The vials were sealed using crimp-top caps with TFE-silicone headspace septa (Agilent). At the time of SPME analysis, each vial was placed at 60°C for 5 min, and then a 120 µm DVB/CWR/PDMS fiber (Agilent) was exposed to the sample’s headspace for 15 min at 60°C. After sampling, desorption of the VOCs from the fiber coating was carried out in the injection port of the GC apparatus (Model 7890; Agilent) at 250°C for 5 min in the splitless mode.The identification and quantification of VOCs were carried out using an Agilent Model 7890 GC and a 7000D mass spectrometer (Agilent) equipped with a 30 m × 0.25 mm × 0.25 µm DB-5MS (5% phenyl-polydimethylsiloxane) capillary column. Helium was used as the carrier gas at a 1.2 mL/min linear velocity. The injector temperature was kept at 250°C and the detector at 280°C. The oven temperature was programmed from 40°C (3.5 min), increasing at 10°C/min to 100°C, at 7°C/min to 180°C, at 25°C/min to 280°C, and held for 5 min. Mass spectra were recorded in electron impact (EI) ionization mode at 70 eV. The quadrupole mass detector, ion source, and transfer line temperatures were set at 150, 230, and 280°C. The MS was selected in ion monitoring (SIM) mode to identify and quantify analytes.

Pathways enriched with both DEGs and DAMs under the same treatment were identified and visualized using KEGG pathway mapping. A histogram was generated to depict the enrichment levels of the identified pathways.

Total RNA was extracted for deep sequencing to validate gene expression analysis results. The qRT-PCR was conducted as previously described. Gene expression was quantified using TB Green® Premix Ex Taq™ II (TaKaRa, Kusatsu, Japan) on a CFX96 real-time PCR system (Bio-Rad, Hercules, CA, USA). Three independent biological replicates were analyzed, with each replicate measured in triplicate to obtain technical replicates. Relative gene expression was calculated using the 2^−ΔΔCt method and expressed as log₂ fold changes. IbActin was used as the internal reference gene. The primer sequences are listed in Table S6.

All experiments were performed in triplicate. Data analysis was conducted using SPSS 15.0 software. All data were presented as mean ± standard deviation (SD). Comparisons within and between groups were conducted using Duncan’s multiple range test. Student’s t-test was used to calculate P-values, with P < 0.05 considered statistically significant.