Abstract Background Charcoal rot of soybean is caused by the hemibiotrophic fungus Macrophomina phaseolina, a global crop destroyer and an important pathogen in the midwestern USA. The quantitative nature of host resistance and the complexity of the soybean-M. phaseolina interaction at the molecular level have hampered resistance breeding. A previous study showed that L-ascorbic acid (LAA) pre-treatment before M. phaseolina inoculation reduced charcoal rot lesion length in excised soybean stems. This study aimed to elucidate the genetic underpinnings of M. phaseolina-induced senescence and the mitigating effects of ascorbic acid on this physiological process within the same pathosystem. Results RNA was sequenced from M. phaseolina-resistant and -susceptible soybean genotypes following M. phaseolina inoculation, LAA, and hydrogen peroxide (H[2]O[2])—an oxidative stress inducer—application followed by inoculation. More genes were down-regulated in the resistant and susceptible genotypes than up-regulated when the M. phaseolina-inoculated treatments were compared to mock-inoculated control treatments. Gene ontology (GO) term and KEGG pathways analysis detected M. phaseolina-induced up-regulation of receptor-like kinase genes. In contrast, many genes related to antioxidants, defense, and hormonal pathways were down-regulated in both genotypes. LAA pre-treatment induced genes related to photosynthesis and reactive oxygen species responses in both genotypes. H[2]O[2] pre-treatment following inoculation up-regulated many stress-response genes, while hormone signal transduction and photosynthesis-related genes were down-regulated in both genotypes. Conclusions Results revealed transcriptional variation and genes associated with M. phaseolina-induced senescence in soybean. Ascorbic acid induced many photosynthetic genes, suggesting a complex regulation of defense and immunity in the plant against the hemibiotroph. Soybean plants also exhibited enhanced stress responsiveness when treated with H[2]O[2] followed by inoculation with M. phaseolina. This study will broaden more research avenues related to transcriptional regulation during the M. phaseolina-soybean interaction and the potential role of receptor-like kinases, oxidative stress-responsive genes, ethylene-mediated signaling and enhanced photosynthetic gene expression when mounting host resistance to this important soybean pathogen. Supplementary Information The online version contains supplementary material available at 10.1186/s12864-024-11023-5. Keywords: Charcoal rot, Macrophomina phaseolina, Ascorbic acid, ROS, Hydrogen peroxide, RNA-seq, Comparative transcriptomics, Host senescence Background Soybean (Glycine max (L.) Merr.) is an economically important field crop produced in many countries and contributes to global food security. Charcoal rot of soybean is caused by the aggressive polyphagous soilborne hemibiotrophic fungus Macrophomina phaseolina (Tassi.) Goid. Charcoal rot is a common root rot disease of soybean that negatively impacts crop quality and causes considerable economic loss, especially in dry and non-irrigated cropping regions [[26]1]. External environmental factors induce charcoal rot, including drought, or limited soil moisture, and high temperatures. Progression of the fungus from the soybean root to the stem eventually blocks the vascular bundles of the plant and induces wilting and death. Charcoal rot causes considerable economic losses in soybean yields, with reported losses in the United States and Ontario, Canada, totaling 219,605 thousand bushels from 2010 to 2014, and an additional 85,259 thousand bushels from 2015 to 2019 [[27]2, [28]3]. Charcoal rot disease management is critical as the fungus remains in an endophytic stage after infecting the roots without any external symptoms. Disease symptoms are visible after the fungus switches from an endophytic biotrophic phase to a necrotrophic phase [[29]4], where visible symptoms include wilting, leaf chlorosis, and plant death. Considering this phenomenon, genetic resistance could be an effective and eco-friendly control measure. However, genetic mechanisms that govern charcoal rot resistance in soybeans remain unknown due to the complex molecular mechanisms underlying the host-pathogen interaction. Also, resistance genes to charcoal rot have not been reported. After the establishment of the plant innate immunity theory, pattern-triggered immunity (PTI) and effector-triggered immunity (ETI) are recognized as the key regulators of plant and pathogen co-evolution [[30]5]. PTI and ETI operate simultaneously on the physiological level in defense signaling and downstream plant defense. The reactive oxygen species (ROS) burst is the first sign of defense signaling that activates the mitogen-activated protein kinase (MAPK) pathway and callose deposition [[31]6]. However, the accumulation of ROS intermediates, including superoxide anions or H[2]O[2], causes oxidative damage to lipids, proteins, nucleic acids, and photosynthetic machinery [[32]7]. Pathogens with necrotrophic stages, like M. phaseolina, could exploit this response. Studies from the Rhizoctonia solani-wheat [[33]8] and Phytophthora infestans-potato [[34]9] pathosystems have demonstrated this phenomenon. Plants also generate antioxidant enzymes such as superoxide dismutase (SOD), catalase (CAT), and peroxidase (PX) to reduce or eliminate ROS. Zhang et al. (2006) found that major ROS, including H[2]O[2], could be scavenged by PX, and the toxic byproduct of the reaction could eliminate Verticillium dahliae proliferation [[35]10]. However, necrotrophic invasion might take place before antioxidant accumulation. Several previous studies have established the basis of this speculation. A microarray experiment using R. solani-inoculated wheat lines showed significantly expressed wheat genes associated with reactive oxygen species (ROS) and redox regulation [[36]8]. Similarly, two ROS/redox regulation-related genes were expressed in R. solani that were homologous to an Alternaria brassicicola ROS regulation gene, which suggested that the simultaneous interaction of ROS-related genes from the host and pathogen results in pathogenesis in the host [[37]8]. A study by Rossi et al. (2017) showed the role of chloroplast-generated ROS in the pathogenesis of the necrotrophic fungus Botrytis cinerea. These authors used two transgenic tobacco (Nicotiana tabacum) lines that generated lower ROS in the chloroplast due to the expression of a plastid-targeted cyanobacterial flavodoxin protein (the transgenic lines were named pfld lines), demonstrated reduced fungal invasion, host tissue damage, and fungal growth compared to the wild-type tobacco line [[38]11]. L-ascorbic acid (LAA) is a plant-derived antioxidant known to perform several physiological activities associated with stress tolerance in plants [[39]12]. Ascorbic acid, an ROS scavenger, participates in plant enzymatic reactions by donating electrons. The mechanism by which L-ascorbic acid scavenges ROS, such as H[2]O[2,] involves its conversion to monodehydroascorbate. This occurs either through reaction with free radicals and Fe^3+ or via ascorbate peroxidase, which catalyzes the conversion of H[2]O[2] to water (H[2]O) [[40]13]. L-ascorbic acid-altered mutants (vtc mutants) have been developed in Arabidopsis. These show a fivefold decrease in ascorbic acid levels, and although the mutations do not interfere with plant growth and development, their capacity for abiotic stress tolerance is greatly compromised [[41]14]. The biotic stress tolerance conferred by L-ascorbic acid is complex and depends on the pathogen type. For biotrophic pathogens, reduced levels of L-ascorbic acid protected against oxidative damage induced by the pathogen [[42]15, [43]16], while in the necrotrophs, higher levels of ascorbic acid enhanced resistance against pathogens [[44]17, [45]18]. It has been reported that ascorbic acid reduces disease symptoms in several pathosystems. Chung et al. (2019) investigated the role of ascorbic acid in eliminating necrotic stress induced by the necrotrophic pathogen Phytophthora infestans from transgenic potato plants overexpressing a D-galacturonic acid reductase gene with enhanced levels of ascorbic acid. The study revealed reduced necrotic lesions and disease symptoms in the transgenic potato plant compared to the control [[46]9]. Also, biomarkers of oxidative stress such as hydrogen peroxide (H[2]O[2]) and malondialdehyde (MDA) levels were significantly lower in the transgenic plant, suggesting the role of ascorbic acid in scavenging ROS and alleviating oxidative stress [[47]9]. Botanga et al. (2012) showed that inoculation of Arabidopsis with Alternaria brassicicola reduced the ascorbate level of the host, and the host’s redox status became imbalanced, which promoted disease severity. This study also suggested a potential role of ascorbate in eliminating pathogen-induced ROS to contribute to host redox homeostasis, as seen from the reduced level of ascorbate in the inoculated plant [[48]18]. We previously hypothesized that M. phaseolina induces oxidative stress-mediated senescence in the host. To address this hypothesis, a study was conducted with an exogenous application of ascorbic acid prior to pathogen inoculation to observe whether ascorbic acid functions as a reactive oxygen scavenger and can rescue soybean plant health from oxidative stress-induced senescence by M. phaseolina [[49]17]. Also, whether ROS such as H[2]O[2] induces oxidative stress and if the oxidative stress is utilized by the subsequent inoculation with M. phaseolina was investigated using exogenous pre-treatment with H[2]O[2] followed by inoculation with M. phaseolina [[50]17]. This study resulted in reduced charcoal rot lesions in the ascorbic acid pre-treated and inoculated soybean stems compared to the non-pre-treated and M. phaseolina-inoculated stems [[51]17]. To reveal potential mechanisms of how ascorbic acid reduces charcoal rot symptoms in the inoculated plant and the role of H[2]O[2] in oxidative stress-mediated senescence, transcriptome analysis was necessary to screen the associated genes and pathways. Therefore, an RNA-seq experiment was conducted to analyze the global transcriptome profiles from ascorbic acid pre-treated and inoculated plants, non-pretreated and inoculated plants, H[2]O[2] pre-treated and inoculated plants, ascorbic acid pre-treated control plants, non-pre-treated control plants, and H[2]O[2]-treated control plants. The objective of the study was to determine: (i) if oxidative stress-related genes were expressed due to inoculation with M. phaseolina; (ii) if plant antioxidant/ROS or oxidative stress response-related genes were expressed by pre-treatment with ascorbic acid following M. phaseolina inoculation; (iii) if H[2]O[2] pre-treatment following inoculation with M. phaseolina induced enhanced expression of oxidative stress and pathogenesis-related genes. Results Correlation analysis between biological replicates of normalized read count data/RPKM for all samples Correlation analysis was conducted to assess the consistency of gene counts within biological replicates of the same treatment and the resistant and susceptible genotypes. This analysis focused on the reads kilobase^−1 million^−1 mapped reads (RPKM) data. A higher correlation coefficient value, approaching 1.0, signifies a robust correlation among biological replicates within the same sample. This strong correlation within biological replicates validates the RNA-seq data, ensuring its consistency and reliability in capturing gene expression patterns. Furthermore, correlation analysis offers a comprehensive overview of the samples while identifying potential outliers [[52]19]. The correlation plot depicting biological replicates of treatments and genotypes illustrates a robust correlation among all three biological replicates within each sample, whether resistant or susceptible (Fig. [53]1). Fig. 1. [54]Fig. 1 [55]Open in a new tab Pearson correlation coefficients of RPKM values between biological replicates of (A) susceptible genotypes (18 samples) and (B) resistant genotypes (18 samples). Abbreviations: Agar = Mock-inoculated control; ASC = Ascorbic acid; H[2]O[2] = Hydrogen peroxide; Inoc = M. phaseolina-inoculated; NPT = Non-pretreated; Pre = Pre-treatment; Trt = Inoculation or other treatment following pre-treatment Alignment of the RNA-seq reads with the reference genome The total input raw sequence read across all 36 samples was around 1,311 million (1,311,492,419 reads), and the total uniquely mapped reads with the reference genome (Glycine_max_v2.1) [[56]20] were approximately 1,000 million (1,000,008,672 reads). Table [57]1 summarizes the read count data across all three biological replicates and 36 samples. Table 1. The average sequence read mapping summary of the RNA-seq data of the resistant (DT97-4290) and susceptible (Pharoah) genotypes and treatments across the biological replicates Genotype Pre-treatment Treatment Input reads Uniquely mapped reads Uniquely mapped (%) DT97-4290 NPT^a Inoc 109255596 90748314 75.39 DT97-4290 ASC Inoc 113881743 81263061 71.26 DT97-4290 NPT Agar 104640629 80999777 77.42 DT97-4290 ASC Agar 110712502 86115587 78.00 DT97-4290 NPT H[2]O[2] 108805730 83424985 77.00 DT97-4290 H[2]O[2] Inoc 111523929 84235777 75.52 Pharaoh NPT Inoc 115167092 86568012 75.13 Pharaoh ASC Inoc 121185775 89920289 75.00 Pharaoh NPT Agar 102266177 78699715 77.00 Pharaoh ASC Agar 102022305 78737938 77.00 Pharaoh NPT H[2]O[2] 106631099 82227534 77.00 Pharaoh H[2]O[2] Inoc 105399842 77067683 73.00 [58]Open in a new tab ^a Abbreviations: ASC Ascorbic acid, Agar Mock-inoculated (agar plug only), Inoc M. phaseolina-inoculated, NPT Non-pre-treated Differential gene expression analysis Differential gene expression analysis compared six treatments for both resistant and susceptible genotypes. Differential gene expression analysis between pairwise treatment comparisons within genotypes showed an anti-conservative type p-value distribution in the histogram for the normalized read counts of the informative genes. The null p-value distribution in the x-axis of the histogram was uniform, with distribution ranges between 0 and 1. This uniform histogram distribution allows proper control of the false discovery rate (FDR) conferred by type I error (reject the null hypothesis when it is true/false positive). The p-value histograms for the informative genes are summarized in Supplementary Fig. [59]S1. The DESeq2 analysis summarized the total number of informative genes, the total number of significant genes, significantly and differentially up- and down-regulated genes, and the significantly and differentially up- and down-regulated two-fold and four-fold genes between comparisons of two treatment groups within the same genotype (Table [60]2). Volcano plots (Fig. [61]S2) illustrate the differentially expressed genes between two treatment groups of the same genotype. The plots compare the log2 fold changes (magnitude of changes) against the negative log10 P-value. Genes that are significantly differentially expressed are highlighted, using an adjusted P-value cut-off of 0.05. Up-regulated genes are highlighted on the right side of the plot and down-regulated genes are highlighted on the left side. Table 2. Differential gene expression summary of the pairwise comparisons of treatments within genotypes Genotypes versus Informative genes Significant genes Up-regulated Down-regulated Pre Trt Pre Trt 1-2fc 2-4fc  > 4fc Total 1-2fc 2-4fc  > 4fc Total DT97-4290 (R) NPT Inoc NPT Agar 40020 1570 117 179 83 379 185 508 498 1191 ASC Inoc ASC Inoc 39034 7428 1808 1134 171 3113 1452 1452 1411 4315 NPT H[2]O[2] H[2]O[2] Inoc 39187 1077 162 268 187 617 199 202 59 460 Pharoah (S) NPT Inoc NPT Agar 40082 1617 143 138 53 334 355 536 392 1283 ASC Inoc ASC Agar 39341 4532 1245 199 31 1475 1414 894 749 3057 NPT H[2]O[2] H[2]O[2] Inoc 39097 3820 587 1177 950 2714 455 514 137 1106 [62]Open in a new tab Abbreviations: ASC Ascorbic acid, Agar Mock-inoculated (agar plug only), H[2]O[2] Hydrogen peroxide, Inoc Macrophomina phaseolina-inoculated, NPT Non-pre-treated, R Charcoal rot-resistant genotype, S Charcoal rot-susceptible genotype Gene ontology (GO) term analysis Gene ontology (GO) term analysis for the differential gene expression of pairwise treatment comparisons within genotypes revealed significantly enriched up- and down-regulated genes and designated GO terms. These GO terms were classified into biological process (BP), cellular component (CC), and molecular function (MF). GO term enrichment analysis was based on a P-value cutoff of 0.05 and a q-value of < 0.05 (to control positive FDR). In all pairwise comparisons within genotypes, most of the GO terms were designated as BP, followed by MF and CC. Differentially expressed genes induced by M. phaseolina inoculation and ascorbic acid pre-treatment following inoculation in both genotypes The Venn diagram comparing mock versus inoculation and ascorbic acid pre-treated mock versus ascorbic acid pre-treated and M. phaseolina-inoculated controls between resistant and susceptible soybean genotypes illustrated a numerical representation of unique and overlapping genes between treatments and genotypes (Fig. [63]2) [[64]21]. In the resistant genotype, the comparison between mock-inoculated control and M. phaseolina-inoculated treatment revealed 317 unique up-regulated genes, 1,021 unique down-regulated genes, and only one gene overlapping between the up- and down-regulated categories. Similarly, the comparison between ascorbic acid pre-treated mock-inoculated control and ascorbic acid pre-treated M. phaseolina-inoculated treatment in the resistant genotype identified 2,742 unique up-regulated genes and 3,691 unique down-regulated genes, with one overlapping between these categories (Fig. [65]2). Fig. 2. [66]Fig. 2 [67]Open in a new tab Venn diagram showing significantly and differentially expressed genes between treatments within the (A) resistant and (B) susceptible soybean genotypes In the resistant genotype, the comparison of up-regulated genes between mock-inoculated control versus the M. phaseolina-inoculated treatment, and the ascorbic acid pre-treated mock-inoculated versus the M. phaseolina-inoculated treatment revealed 118 overlapping up-regulated genes (Fig. [68]2). Additionally, the comparison within the resistant genotype for down-regulated genes between these treatments showed an overlap of 930 genes (Fig. [69]2). In the susceptible genotype, the mock-inoculated control versus the M. phaseolina-inoculated treatment comparison showed 290 up-regulated genes and 1,104 down-regulated genes, while one gene overlapped between the up- and down-regulated categories. Furthermore, the comparison between ascorbic acid pre-treated M. phaseolina-inoculated treatment versus ascorbic acid pre-treated mock-inoculated control and mock-inoculated control versus the M. phaseolina-inoculated treatment identified 35 overlapping up-regulated genes and 883 overlapping down-regulated genes (Fig. [70]2). DE genes and associated GO terms in the resistant and susceptible genotypes induced by M. phaseolina inoculation and exogenous application of ascorbic acid and H[2]O[2] In the comparison of the non-pre-treated and M. phaseolina-inoculated treatment with the non-pre-treated and mock-inoculated control in the resistant genotype (DT97-4290), 379 genes were up-regulated, of which 117 and 83 genes were two- and > four-fold significantly up-regulated, respectively (Table [71]2). Up-regulated BP GO terms included cellular response to nitrate (GO:0071249), salicylic acid biosynthesis process (GO:0080142), regulation of salicylic acid metabolic process (GO:0009696), phenol-containing compound biosynthetic process (GO:0046189), cellular response to nitrate (GO:0071249) and nitrate transport (GO:0015706); cellular morphogenesis (GO:0000902), defense responses (GO:0006952), response to biotic stimulus (GO:0009607), and immune system process (GO:0002376) (Fig. [72]3). Among the BP, a higher number of genes were associated with defense responses, response to biotic stimulus (GO:0009607), and immune system process (GO:0002376) (Fig. [73]3) [[74]22]. MF GO terms included terpene synthase activity (GO:0010333), ubiquitin protein ligase binding (GO:0031625), enzyme binding (GO:0019899), magnesium ion binding (GO:0000287), kinase activity (GO:0016301), transmembrane signaling receptor activity (GO:0004675), ligand-gated ion channel activity (GO:0099507), ionotropic glutamate receptor activity, glutamate receptor activity (GO:0004970), and protein serine/threonine kinase (GO:0004672) and protein serine kinase activity (GO:0106310) (Fig. [75]3) [[76]22]. Protein serine/threonine kinase activity-related genes represented the major group of up-regulated genes within the MF category. In the CC GO category, extracellular region, plasmodesma, symplast, cell–cell junction, cell junction, and anchoring junction were up-regulated with > 10 genes each (Fig. [77]3). Fig. 3. [78]Fig. 3 [79]Open in a new tab Dot plot showing gene ontology (GO) term analysis for the non-pretreated and M. phaseolina-inoculated treatment versus the non-pre-treated mock-inoculated control for up- and down-regulated GO terms in the resistant and susceptible soybean genotypes. Abbreviations: Agar = mock-inoculated control, BP = biological processes, CC = cellular components, Inoc = M. phaseolina-inoculated, MF = molecular functions, NPT = non-pre-treated For the same comparison, 1191 genes were significantly and differentially down-regulated, of which 185 and 498 genes exhibited two- and four-fold down-regulation, respectively (Table [80]2). The GO terms for differentially expressed down-regulated genes were higher than in the up-regulated category. Most down-regulated BP GO terms included metabolic functions such as glucosinolate metabolic process (GO:0019760), flavonoid metabolic (GO:0009812) and biosynthetic process (GO:0009813), hydrogen peroxide and reactive oxygen species metabolic process (GO:0042743), oxoacid metabolic process (GO:0043436), and organic acid metabolic process (GO:0006082). Other significant BP GO terms were naringenin-chalcone synthase activity, cellular oxidant detoxification (GO:0098869), cellular response to toxic substances (GO:0097237), and detoxification-related (GO:0098754) genes (Fig. [81]3) [[82]22]. As for the up-regulated category GO terms, the majority fell in the MF category, but most of the down-regulated GO terms were BP. Down-regulated MF-associated GO terms were negative regulation of proteolysis, peptidase, and hydrolase activity, peroxidase activity, glucosidase and beta-glucosidase activity, and antioxidant activity (Fig. [83]3). The CC category represented the extracellular region with a larger number of genes compared to other CC categories, including external encapsulating structure and plant-type cell wall (Fig. [84]3). In the susceptible genotype Pharaoh, the pairwise comparison between the non-pre-treated M. phaseolina-inoculated treatment with the non-pre-treated mock-inoculated control showed 334 differentially and significantly up-regulated genes with 143 two-fold and 53 > four-fold significantly up-regulated genes (Table [85]2). Significant BP GO terms associated with a higher number of genes were response to reactive oxygen species (GO:0000302), cell surface receptor signaling pathway (GO:0007166), recognition of pollen (GO:0048544), cell recognition (GO:0008037), and response to inorganic substance (GO:0010035) (Fig. [86]3). The MF category included the largest number of significant genes per GO category. Defense-related MF GOs were protein serine/threonine kinase activity, signaling and transmembrane signaling receptor activity, and ligand-gated channel activity. CC included nucleoid, plastid nucleoid, and chloroplast nucleoid (Fig. [87]3). There were 1283 significantly down-regulated genes including 355 two-fold and 392 > four-fold significantly down-regulated genes. The down-regulated BP GO terms associated with most genes were cellular oxidant detoxification, defense response, hydrogen peroxide metabolic process (GO:0042743) and secondary metabolic process (GO:0019748), glucosinolate metabolic process (GO:0019760), and negative regulation of molecular function (GO:0044092) and catalytic activity (GO:0043086). The major MF GO categories were heme and tetrapyrrole (GO:0046906) and iron ion binding (GO:0005506), transferase activity (GO:0016704), antioxidant (GO:0016209), monooxygenase (GO:0004497) and glucosidase activity (GO:0015926), peroxidase activity (GO:0004601), oxidoreductase activity (GO:0016491), endopeptidase (GO:0004175), and peptidase regulator activity (GO:0061134). The CC GO categories included cell wall (GO:0005618), external encapsulating structure (GO:0030312) with a higher number of genes, extracellular region (GO:0005576), and vacuolar lumen (GO:0005775) (Fig. [88]3) [[89]22]. In the down-regulated defense-related GO (GO:0006979) category, 17 oxidative stress-responsive genes were commonly expressed in DT97-4290 and Pharaoh (Fig. [90]4). M. phaseolina-induced over-expression of genes related to response to oxidative stress were GLYMA_19G011700, 14G201700, 10G222400, 20G169200, 15G129200, 04G220600, and 15G128700 in both genotypes (Fig. [91]4). These genes are primarily associated with peroxidases. The defense-related GO terms kinase activity and serine-threonine protein kinases differentially up-regulated and commonly expressed 34 genes in both genotypes. Genes that showed higher differential expression due to M. phaseolina inoculation compared to mock-inoculated control are related to receptor-like serine/threonine-protein kinase (RLS/TPK), Leucine-rich receptor (LRR) like serine/threonine-protein kinase (LRR S/TPK), receptor-like protein kinase (RLPK), receptor-like cytoplasmic protein kinase (RLCPK), and leucine-rich repeat protein kinase (LRR PK) family protein and protein kinase (PK) domain-containing protein (Fig. [92]5). The highly differentially expressed kinase activity-related genes are GLYMA_07G188800, 14G206200, 17G103400, 08G127400, 10G069500, 10G023300, 10G023400, 06G197600, 08G187400, 20G139500, 15G074600, 06G226700, 15G064900, 05G168800, 08G094400, 05G024000, 20G139400, 15G253500, and 04G230500 (Fig. [93]5). Fig. 4. [94]Fig. 4 [95]Open in a new tab Heatmap for the non-pre-treated and M. phaseolina-inoculated treatment versus the non-pre-treated mock-inoculated control in the resistant genotype (R; DT97-4290) and susceptible genotype (S; Pharaoh) showing differentially expressed oxidative stress responsive genes. The gene expression heatmap was based on the average normalized read counts of three biological replicates Fig. 5. [96]Fig. 5 [97]Open in a new tab Heatmap for the non-pre-treated M. phaseolina-inoculated treatment versus the non-pre-treated mock-inoculated control in the resistant genotype (R; DT97-4290) and the susceptible genotype (S; Pharaoh) showing differentially expressed kinase activity-related genes. The gene expression heatmap was based on the average normalized read counts of three biological replicates DE genes and associated GO terms in the resistant and susceptible genotypes induced by exogenous application of ascorbic acid Comparisons of ascorbic acid pre-treated and M. phaseolina-inoculated versus the ascorbic acid pre-treated mock-inoculated control treatments in the resistant genotype revealed 3113 up-regulated genes with 1808 two-fold and 171 greater than four-fold significantly up-regulated (Table [98]2). The GO terms associated with the up-regulated BP category were cellular response to light stimulus (GO:0071482), cell wall biogenesis (GO:0042546), photosynthesis (GO:0015979), cellular and cell wall polysaccharide metabolic process (GO:0010383), glucan metabolic process (GO:0044042), cellular glucan metabolic process, xyloglucan metabolic process (GO:0010411), and cellular carbohydrate metabolic process (GO:0005975). GO terms, iron ion binding (GO:0005506), chlorophyll binding (GO:0016168), tetrapyrrole binding (GO:0046906), glycosyltransferase (GO:0016757), hexosyltransferase (GO:0016758), and galactosidase activity (GO:0015925) included some of the higher gene numbers associated with molecular functions (MF). In the CC category, many components involved the photosynthetic machinery including chloroplast, plastid (GO:0009536), chloroplast and plastid thylakoids (GO:0031976), plastid envelope (GO:0009526), photosynthetic membrane (GO:0034357), and photosystem II (Fig. [99]6) [[100]22]. This comparison showed 4,315 down-regulated, differentially expressed genes, with 1452 two-fold and 1411 > four-fold significantly differentially expressed genes (Table [101]2). BP GO terms were hydrogen peroxide metabolic process (GO:0042743) and glucosinolate metabolic process (GO:0019760), as well as reactive oxygen species (GO:0072593) and flavonoid metabolic process (GO:0009812); many genes were associated with cellular oxidant detoxification (GO:0098869) and cellular response to toxic substance (GO:0097237). The MF category included glutathione transferase (GO:0004364), antioxidant (GO:0016209), and peroxidase (GO:0004601) activities. Plant-type cell wall (GO:0009505) and external encapsulating structure (GO:0030312) were expressed in the CC category (Fig. [102]6) [[103]22]. Fig. 6. [104]Fig. 6 [105]Open in a new tab Dot plot showing gene ontology (GO) term analysis for the ascorbic acid pre-treated and M. phaseolina-inoculated treatment versus the ascorbic acid-pre-treated and mock-inoculated control as up- and down-regulated GO terms for the resistant and susceptible genotypes. Abbreviations: Agar = mock-inoculated control, ASC = ascorbic acid, BP = biological processes, CC = cellular components, Inoc = M. phaseolina-inoculated, MF = molecular functions The comparison between the ascorbic acid pre-treated and M. phaseolina-inoculated treatment versus the ascorbic acid pre-treated and mock-inoculated control treatment for the susceptible genotype (Pharoah) revealed 1475 significantly up-regulated genes, which was lower than the resistant genotype, DT97-4290. Of those, 1245 and 31 genes were significantly two-fold and > four-fold significantly up-regulated, respectively (Table [106]2). No BP GO terms were expressed in this comparison; CC and MF mainly represented the up-regulated category (Fig. [107]6). Most GO terms in this comparison were in the MF category and included secondary active transmembrane transporter activity (GO:0022857), monooxygenase activity (GO:0004497), terpene synthase activity (GO:0010333), heme binding (GO:0020037), tetrapyrrole binding (GO:0046906), and symporter activity (GO:0015293). In the CC category were apoplast (GO:0048046), cell wall (GO:0005618), and extracellular region (GO:0005576). There were 3057 down-regulated differentially expressed genes, more than twice that of the up-regulated genes, with 1414 two-fold and 749 > four-fold significantly differentially expressed genes (Table [108]2). The BP GO terms were associated with defense responses, including flavonoid biosynthetic process, hydrogen peroxidase and reactive oxygen species metabolic processes, cellular oxidant detoxification, and glucosinolate metabolic (GO:0019760) and catabolic processes (GO:0019762). The MF category included protein serine/threonine kinase activity (GO:0004674), carbohydrate (GO:0030246), heme (GO:0020037), chitin binding (GO:0008061), and peroxidase (GO:0004601), beta-glucosidase (GO:0008422), and antioxidant (GO:0016209) activities [[109]22]. The down-regulated CC category GO terms were external encapsulating structure (GO:0030312), apoplast (GO:0048046), and extracellular matrix (GO:0031012) (Fig. [110]6) [[111]22]. In the resistant genotype, the highly differentially expressed defense-related genes induced by the ascorbic acid pre-treatment following M. phaseolina-inoculation were associated with flavonoid biosynthetic process (GLYMA_08G109300, 02G130400, 14G098100, 10G292200, and 09G075200), response to oxidative stress (GLYMA_02G008900, 08G179700, and 14G177200) GO terms (Fig. [112]7), and ethylene-responsive transcription factors (GLYMA_08G257300, 09G041500, 11G036500, 10G223200, 10G007000, and 06G148400). In the resistant genotype, along with overexpression of defense-related genes, most significantly differentially expressed genes were related to photosynthesis and light harvesting (GO:0009765), photosystem I (GO:0009522), and photosystem II assembly (GO:0010207) (Fig. [113]8). Along with defense-related genes, ascorbic acid pre-treatment induced many photosynthesis-related genes that were expressed in the resistant genotype. The photosynthesis-related genes were more highly expressed in the ascorbic acid pre-treated mock control treatment than in the ascorbic acid pre-treated and M. phaseolina-inoculated treatment. The highly differentially expressed photosynthesis-related genes included chlorophyll a-b binding proteins (GLYMA_03G060300, 09G154700, 01G115900, 06G194900, and 16G145800), photosystem I (GLYMA_06G150300, 04G112800, 05G022900, 16G165200, and 04G215800) and photosystem II (GLYMA_11G245400, 16G165800, and 09G073900) -related genes (Fig. [114]8). Fig. 7. [115]Fig. 7 [116]Open in a new tab Heatmap for the ascorbic acid pre-treated and M. phaseolina-inoculated treatment versus ascorbic acid pre-treated and mock-inoculated control in the resistant genotype (R; DT97-4290) showing differentially expressed defense-related genes. The gene expression heatmap was based on the average normalized read counts of three biological replicates Fig. 8. [117]Fig. 8 [118]Open in a new tab Heatmap for the ascorbic acid pre-treated and M. phaseolina-inoculated treatment versus the mock-inoculated treatment in the resistant genotype (R; DT97-4290) showing differentially expressed photosynthesis-related genes. The heatmap was based on the average normalized read counts of three biological replicates DE genes and associated GO terms in the resistant and susceptible genotype induced by exogenous application of H[2]O[2] In the resistant genotype, the non-pre-treated and H[2]O[2] treatment versus the H[2]O[2] pre-treated and M. phaseolina-inoculated treatments showed 617 differentially and significantly up-regulated genes with 167 two-fold and 187 genes greater than four-fold significantly differentially expressed (Table [119]2). Based on the p-value and number of genes associated with the GO terms in the up-regulated category, the BP GO terms were response to oxidative stress, including cellular detoxification (GO:1990748), detoxification, and response to toxic substances (GO:0009636); glucosinolate, hydrogen peroxide, reactive oxygen species, oxoacid metabolic process (GO:0043436), and organic acid metabolic processes (GO:0006082); monosaccharide and hexose biosynthetic processes; and gluconeogenesis. MF-classified GOs included catalytic activity such as oxidoreductase activity, hydrolase activity (of glycosyl bonds) (GO:0016798), and intramolecular lyase activity (GO:0016872); transferase activity, NAD(P)H dehydrogenase (quinone) activity (GO:0003955), peroxidase, and monooxygenase activity; and binding of iron. CC-associated GO terms were apoplast and extracellular region (Fig. [120]9). In this treatment comparison, 460 genes were significantly down-regulated and lower than the number of up-regulated genes, with 199 two-fold and 59 greater than four-fold significantly differentially expressed genes (Table [121]2). Down-regulated BP GO terms associated with a higher number of genes are response to oxidative stress and response to oxygen-containing compound; other significant BP GOs are nitric oxide metabolic process (GO:0046209), reactive nitrogen species, nitrogen cycle, nitrate metabolic process, isoprenoid, and cellulose catabolic process. The down-regulated MF category included hydrolase activity, hexotransferase activity, cellulase activity, sequence-specific and unfolded protein and heme binding, polysaccharide (GO:0000272), and cellulose catabolic process (GO:0030245) [[122]22]. The CC category included only the extracellular region (Fig. [123]9) [[124]22]. Fig. 9. [125]Fig. 9 [126]Open in a new tab Dot plot showing gene ontology (GO) term analysis for the non-pre-treated and H[2]O[2] treated treatment versus the H[2]O[2] pre-treated and M. phaseolina-inoculated treatment as up- and down-regulated GO terms for the resistant and susceptible genotypes. Abbreviations: BP = biological processes, CC = cellular components, Inoc = M. phaseolina-inoculated, MF = molecular functions, and NPT = non-pre-treated In the susceptible genotype, Pharaoh, the comparison between the non-pre-treated H[2]O[2] application with the H[2]O[2]-pre-treated M. phaseolina-inoculated treatment showed 2714 significantly up-regulated differentially expressed genes, of which 587 and 950 were respectively two-fold and > four-fold significantly and differentially expressed (Table [127]2). Up-regulated BP GO terms were cellular response to chemical stimulus (GO:0070887), carboxylic acid metabolic process (GO:0019752), monocarboxylic acid metabolic process (GO:0032787), organic acid metabolic process, and oxoacid metabolic process, response to toxic substance, cellular detoxification, and reactive oxygen species metabolic process (GO:0072593). Up-regulated MF GO terms were oxidoreductase activity (GO:0016491), protein serine/threonine kinase activity (GO:0004712), peroxidase activity (GO:0004601), antioxidant activity (GO:0016209), abscisic acid binding (GO:0010427) and naringenin-chalcone synthase activity (GO:0016210). CC GO terms were plasmodesmata, extracellular region, symplast (GO:0055044), cell-cell junction (GO:0030054), and external encapsulating structure (GO:9943591) (GO:0030312) (Fig. [128]9) [[129]22]. There were 1106 differentially expressed down-regulated genes, with 455 two-fold and 137 > four-fold significantly differentially expressed genes (Table [130]2). BP GO terms were response to reactive oxygen species (GO:0000302) and oxygen-containing compound (GO:1901700), negative regulation of molecular function (GO:0044092) and catalytic activity (GO:0043086), and isoprenoid catabolic process (GO:0008300). In the MF category, the GO terms were transmembrane transport (GO:0055085), oxidoreductase activity (GO:0016491), hydrolase activity (GO:0016787), sequence-specific DNA-binding (GO:0043565), and enzyme regulator activity (GO:0030234). CC GO terms with significant numbers of genes were cell wall (GO:0005618) and extracellular region (GO:0005576) (Fig. [131]9) [[132]22]. In the resistant genotype, a majority of the differentially expressed defense-related genes were related to response to oxidative stress and those were peroxidases (GLYMA_15G128700, 01G228700, 20G241500, 01G239600, 11G011500, 16G164200, 20G169200, 02G234200, 14G201700, and 26G164200) and kinases such as leucine-rich repeat receptor-like serine/threonine protein kinases (LRR RLSTPK) (GLYMA_08G235900 and 07G127100) (Fig. [133]10). Fig. 10. Fig. 10 [134]Open in a new tab Heatmap for the non-pre-treated H[2]O[2] applied treatment versus the H[2]O[2] pre-treated and M. phaseolina-inoculated treatment in the resistant genotype (R; DT97-4290) showing differentially expressed defense-related genes. The gene expression heatmap was based on the average normalized read counts of three biological replicates. Abbreviations: Inoc = M. phaseolina-inoculated and NPT = non-pre-treated The highly differentially expressed genes induced by non-pre-treated H[2]O[2] treatment and H[2]O[2] pre-treatment followed by M. phaseolina inoculation in the susceptible genotype are related to chalcone synthase, peroxidases (GLYMA_15G128700, 13G138300, 12G129500, 10G050800, 09G022400, 20G171900, and 20G169200), receptor-like protein kinases (GLYMA_20G137900, 08G235800, and 08G235900), and glutathione-S-transferases (GLYMA_02G187200, 03G176300, and 14G031000) (Fig. [135]11). Fig. 11. [136]Fig. 11 [137]Open in a new tab Heatmap for the non-pre-treated H[2]O[2] applied treatment versus the H[2]O[2] pre-treated and M. phaseolina-inoculated treatment in the susceptible genotype (Pharaoh) showing differentially expressed defense related genes. The heatmap was based on the average normalized read counts of three biological replicates. Abbreviations: Inoc = M. phaseolina-inoculated and NPT = non-pre-treated KEGG pathway analysis in the genotype by treatment interactions The DE genes that were log2-fold significant for the genotype-by-treatment interaction were pasted in the ShinyGO online platform to obtain the pathway annotation using the KEGG (Kyoto Encyclopedia of Genes and Genomes) database (ShinyGO 0.77) [[138]23, [139]24]. While comparing the non-pre-treated M. phaseolina-inoculated treatment with the non-pre-treated mock-inoculated control in both genotypes, some common pathways were enriched, such as secondary metabolite biosynthesis, isoflavonoid biosynthesis, flavonoid biosynthesis, and phenylpropanoid biosynthesis. However, in both genotypes, the pathways associated with flavonoid (Figs. S3 & S4), isoflavonoid, and phenylpropanoid biosynthesis were down-regulated (Figs. [140]12 & [141]13). Fig. 12. [142]Fig. 12 [143]Open in a new tab Down-regulated KEGG pathways enriched by non-pre-treated M. phaseolina-inoculated treatment versus non-pre-treated mock-inoculated treatment in the resistant genotype (DT97-4290) Fig. 13. [144]Fig. 13 [145]Open in a new tab Up- and down-regulated KEGG pathways enriched by non-pre-treated M. phaseolina-inoculated treatment versus the non-pre-treated and mock-inoculated treatment in the susceptible genotype (Pharaoh) In the susceptible genotype, only the plant-pathogen interaction and MAPK signal pathways were up-regulated (Fig. [146]13). Also in the susceptible genotype (Pharaoh), glucosinolate biosynthesis, amino acid biosynthesis, ubiquinone, and other terpenoid-quinone biosynthesis pathways were down-regulated (Fig. [147]13). However, in the resistant genotype (DT97-4290), stilbenoid, diarylheptanoid, and gingerol biosynthesis pathways were expressed and down-regulated (Fig. [148]12). Host PRR (pattern recognition receptor)-mediated defense-related signaling and mitogen-activated protein kinase (MAPK) signaling pathways were expressed and down-regulated in both genotypes. Several metabolic pathways were expressed and down-regulated in both genotypes including glutathione metabolism, carbon metabolism, nitrogen metabolism, galactose metabolism, starch and sucrose metabolism, and cyanoamino acid metabolism. Only cysteine, methionine, and glutathione metabolism were significantly enriched and down-regulated in the resistant genotype. In the susceptible genotype, alanine, aspartate, and glutamate metabolism were down-regulated (Fig. [149]12). Comparison of the ascorbic acid pre-treatment and M. phaseolina-inoculated treatment versus the ascorbic acid pre-treated and mock-inoculated control treatment in the resistant and susceptible genotypes showed up-regulation of pathways associated with photosynthesis (Fig. [150]S5) in both genotypes (Figs. [151]14 & [152]15). This did not show up in the M. phaseolina-inoculated versus mock-inoculated control comparisons for either genotype. In these comparisons, stress responsive metabolic pathways (flavonoid, isoflavonoid, and phenylpropanoid) were down-regulated in both genotypes (Figs. [153]14 & [154]15). Fig. 14. [155]Fig. 14 [156]Open in a new tab Up- and down-regulated KEGG pathways enriched by ascorbic acid pre-treated, and M. phaseolina-inoculated versus the ascorbic acid pre-treated and mock-inoculated treatment in the resistant genotype (DT97-4290) Fig. 15. [157]Fig. 15 [158]Open in a new tab Up- and down-regulated KEGG pathways enriched by ascorbic acid pre-treated, and M. phaseolina-inoculated treatment compared to the ascorbic acid pre-treated and mock-inoculated treatment in the susceptible genotype (Pharaoh) Treatment comparisons for non-pre-treated and H[2]O[2] treated, and H[2]O[2] pre-treated and M. phaseolina-inoculated were compared in both the susceptible and resistant genotypes for their associated pathway analysis. In both genotypes, metabolic pathways associated with flavonoid, isoflavonoid, and phenylpropanoid biosynthesis were up-regulated with three- to five-fold enrichment (Figs. [159]16 & [160]17). The glutathione metabolism pathway was up-regulated only in the resistant genotype (Fig. [161]16). For the resistant genotype in the down-regulated category, the phenylpropanoid biosynthesis pathway was expressed with nearly five-fold enrichment (Fig. [162]16). Fig. 16. [163]Fig. 16 [164]Open in a new tab Up- and down-regulated KEGG pathways enriched by non-pre-treated and H[2]O[2] treatment versus the H[2]O[2] pre-treated and M. phaseolina-inoculated treatment in the resistant genotype (DT97-4290) Fig. 17. [165]Fig. 17 [166]Open in a new tab Up- and down-regulated KEGG pathways enriched by non-pre-treated H[2]O[2] treatment compared to the H[2]O[2] pre-treated and M. phaseolina-inoculated treatment in the susceptible genotype (Pharaoh) In the susceptible genotype, more pathways appeared in the up-regulated category than in the down-regulated category. For the susceptible genotype, carotenoid biosynthesis was down-regulated and enriched nearly seven-fold (Fig. [167]17). Plant hormone signal transduction was also down-regulated with more than two-fold enrichment. Several biosynthetic and metabolic pathways were up-regulated in the susceptible genotypes including the stilbenoid, diarylheptanoid, and gingerol biosynthesis pathways; phenylalanine, tyrosine, and tryptophan biosynthesis pathways; and the alanine, aspartate, and glutamate metabolism pathways (Fig. [168]17). Note: We recognize the enduring importance of independently validating RNA-seq data. However, recent literature indicates that RNA-seq data is frequently considered sufficiently robust and may not require validation through quantitative real-time PCR (qRT-PCR) if essential conditions such as maintaining an adequate number of biological replicates and conducting precise experimental procedures and data analysis are fulfilled [[169]25, [170]26]. In addition, a comprehensive investigation conducted by Everaert et al. (2017), comparing RNA-seq data with qRT-PCR, demonstrated minimal non-concordance, with as little as 1.8% disparity observed when examining differentially expressed genes (DEG) and fold-change (FC) values > 2. As a result, the need for RNA-seq data validation through qRT-PCR was generally not deemed necessary [[171]25–[172]29] in our study. Our primary objective was to identify sets of differentially expressed genes (DEGs) induced by the inoculation of the pathogen M. phaseolina or through the exogenous application of L-ascorbic acid or H[2]O[2] following inoculation. Instead of focusing on individual DEGs requiring characterization through qRT-PCR, we emphasized identifying broader gene expression patterns and transcriptional changes. Discussion Plants express mechanistically diverse genes and pathways in response to biotic and abiotic stressors [[173]30], including to the fungus M. phaseolina. This study investigated soybean plant responses to biotic stress induced by M. phaseolina, focusing on ROS induced by M. phaseolina in two different soybean genotypes. Additionally, this study investigated the role of the non-enzymatic antioxidant and ROS scavenger, ascorbic acid, in ROS elimination, as well as the involvement of H[2]O[2] in induced oxidative stress-mediated senescence. To comprehensively analyze these responses, we conducted a comparative RNA-seq analysis between a resistant and susceptible soybean genotype. The results outlined above highlight the activation of pathogenicity, stress-responsive, and antioxidant-associated genes in response to M. phaseolina-induced oxidative stress-mediated senescence. Protein kinase-related genes are expressed in response to oxidative stress induced by M. phaseolina In the resistant genotype (DT97-4290), the comparison between non-pre-treated M. phaseolina-inoculated versus the non-pre-treated and mock-inoculated control demonstrated significantly expressed genes that were associated with protein kinase activity (GO:0004672), protein serine/threonine kinase activity (GO:0004674), and kinase activity (GO:0016301). These GO terms are related to receptor-like kinases (RLKs), a large family of plasma membrane proteins that transduce extracellular signals sensed by the plant [[174]31]. In our analysis, when we compared the mock-inoculated versus the M. phaseolina-inoculated treatments in the resistant and susceptible genotypes, six leucine-rich repeat (LRR) protein kinase (PK) and serine/threonine protein kinase (S/TPK) genes (GLYMA_14G206200, 17G103400, 10G069500, 10G023400, 06G197600, and 05G024000) were significantly up-regulated in response to M. phaseolina. Among these genes, GLYMA_14G206200, 17G103400, and 06G197600 shared structural similarities with Arabidopsis’s receptor-like serine/threonine protein kinases (RLS/TPK). This result suggests possible M. phaseolina-induced expression of oxidative stress-responsive genes in both soybean genotypes, as expression of RLK genes is related to direct or indirect perception of ROS and oxidative stress-related responses [[175]32–[176]34]. RLKs are represented by three protein domains: an extracellular domain that senses and perceives a ligand molecule, a transmembrane domain that integrates the protein with the cell membrane, and a cellular kinase that transmits downstream signals by phosphorylation and activation [[177]35, [178]36]. Czernic et al. (1999) showed that the Arabidopsis receptor-like protein kinase gene (At-RLK3) is activated by oxidative stress induced by inoculation with the bacterium Ralstonia solanacearum. The possible role of At-RLK3 in signal transduction revealed that it was activated and induced by H[2]O[2] (ROS) or menadione and salicylic acid, which supports our findings [[179]32]. Zhang et al. (2013) showed that the serine/threonine protein kinase gene GbSTK (a defense-associated gene in cotton) when overexpressed in the model plant Arabidopsis, enhanced the pathogenicity of Verticillium dahliae and increased oxidative stress following the activation of defense-signaling pathways. Their findings revealed that GbSTK responds to infection and oxidative stress induced by the pathogen [[180]37]. M. phaseolina inoculation also induced one G-type lectin S-receptor-like serine/threonine protein kinase (GLYMA_15G064900) and one lectin domain-containing receptor kinase (GLYMA_15G074600). The carbohydrate-binding protein lectin has been found to work as a pathogen and pest resistance gene in soybean. The transgenic expression of soybean lectin in tobacco also defended the plant against pathogen invasion [[181]38, [182]39]. M. phaseolina down-regulates defense and cellular oxidant detoxification genes in both genotypes In the resistant and susceptible genotypes, high numbers of down-regulated genes were associated with the following GO terms: response to oxidative stress (GO:0006979), hydrogen peroxide catabolic process (GO:0042744), cellular oxidant detoxification (GO:0098869), and peroxidase activity (PX) (GO:0004601) when the non-pre-treated M. phaseolina-inoculated versus the mock-inoculated control was compared. Yan et al. (2021) showed that peroxidase activity-related genes contributed to plant defense and work as antioxidants by contributing to pathogen-induced ROS detoxification [[183]7]. Also, if plants are stressed by pathogen invasion, peroxidase activity is enhanced [[184]7]. Our analysis indicated that the peroxidase-related genes GLYMA_15G129200, 04G220600, 15G128700, 19G011700, 14G201700, 10G222400, and 20G169200 were induced by M. phaseolina in both genotypes with higher differential expression compared to the mock-inoculated treatment. These PX-associated genes were down-regulated in both genotypes, which may contribute to host compromise and promote pathogen invasion and proliferation. Cellular oxidant detoxification-related genes were also down-regulated in both genotypes. These cellular oxidant detoxification-related genes were related to different peroxidases such as peroxidase P7, A2, 12, and PC7. Cellular oxidant detoxification is a biological process that eliminates or scavenges cellular superoxides and ROS, such as H[2]O[2], when they reach a toxic level [[185]7]. Thus, our findings suggest that M. phaseolina may induce oxidative stress in both genotypes, as evidenced by the lack of overexpression of genes related to cellular oxidant scavenging. Exogenous ascorbate (L-ascorbic acid) induced contrasting responses to oxidative stress, photosynthesis, and defense-related genes in the resistant and susceptible genotypes Comparison of ascorbic acid pre-treatment following inoculation versus the ascorbic acid pre-treated and mock-inoculated control treatment in both genotypes showed up-regulation of a high number of photosynthesis and photosynthetic machinery-related GO terms, suggesting ascorbic acid-induced over-expression of photosynthesis in both genotypes. Bilgin et al. (2010) studied the effect of different biotic stresses on photosynthesis, and their study revealed that plants restrain the expression of photosynthesis genes during biotic stress while investing more energy related to defense to conserve resources [[186]40]. Chen et al. (2021) showed the role of exogenous ascorbic acid application in rescuing salt stress-induced damage and increasing photosynthesis by modulating endogenous ascorbate levels [[187]41]. Foyer (2015) included ascorbic acid within the key antioxidant molecules required for continued photosynthesis [[188]42]. Chung et al. (2019) studied transgenic potato plants overexpressing a D-galacturonic acid reductase (GalUR) gene with elevated cellular L-ascorbate levels. They showed increased mRNA abundance of cellular antioxidants, pathogenesis-related PR genes, defense-related and hormonal signaling of gibberellic acid (GA) and abscisic acid (ABA)-related biosynthetic pathway genes when inoculated with the hemibiotrophic pathogen Phytophthora infestans. The authors also detected lower levels of hydrogen peroxide (H[2]O[2]) and malondialdehyde (MDA) in the transgenic plants compared to the untransformed plants [[189]9], which supported the speculation of ascorbate-mediated hydrogen peroxide scavenging in the transgenic potato plants. Other than the photosynthesis-related genes, many genes related to carbohydrate metabolic processes were up-regulated in the resistant genotype. Berger et al. (2004) hypothesized that biotrophic and necrotrophic pathogens induced sink-specific genes, including defense genes, and repressed photosynthesis and carbohydrate metabolism-related genes in the plant to weaken plant immunity [[190]43]. They assessed gene expression, photosynthesis, and sugar levels of plant restorative (ComCat, an organic growth agent) treated pathogen-inoculated plants, which suggested that increased photosynthesis and carbohydrate metabolism work as a plant immunity arsenal against pathogen invasion [[191]43]. Our results from the ascorbic acid pre-treated and M. phaseolina-inoculated soybean plants also showed a similar trend of host resistance against the fungus. The significantly reduced level of soybean stem necrosis in the ascorbic acid pre-treated and M. phaseolina-inoculated soybean plants [[192]17] could be due to increased expression of photosynthesis and carbohydrate metabolism genes. We detected the expression of several ethylene-responsive transcription factors while comparing normalized read count data of ascorbic acid pre-treated and M. phaseolina-inoculated versus ascorbic pre-treated mock-inoculated control treatments in the resistant genotype. Ethylene plays a critical role in host defenses through a complex signaling network involving jasmonic acid, salicylic acid, and abscisic acid [[193]44, [194]45]. Transcriptional activation of ETHYLENE RESPONSE FACTOR1 (ERF1) by the convergent signaling pathways of ethylene and jasmonate triggers host defense-related transcription factors [[195]45]. Lanubile et al. (2015) also detected several ethylene responsive transcription factors as highly expressed and induced by a pathogenic isolate of Fusarium oxysporum in soybean [[196]46]. Overexpression of ERF is associated with stress responses and increased host resistance against different pathogens [[197]46–[198]48]. Hydrogen peroxide induced contrasting responses of defense, oxidative stress, and oxidant detoxification genes in the resistant and susceptible genotypes A comparison of H[2]O[2] pre-treatment followed by M. phaseolina inoculation versus the non-pre-treated and H[2]O[2] treatment revealed up-regulation of some oxidative stress-responsive and ROS scavenging-related genes in the resistant genotype. GO terms related to response to oxidative stress and cellular oxidant detoxification, hydrogen peroxide metabolic process, and peroxidase activity were up-regulated. These GO terms are related to ROS homeostasis in the cell. Nevertheless, in the susceptible genotype, a distinct scenario has been observed concerning several genes and their enrichment related to defense and oxidative stress response compared to the resistant genotype. This suggests that H[2]O[2] induces enhanced expression of oxidative stress-responsive genes in the susceptible genotype to a greater extent than in the resistant genotype. Such contrasting responses to H[2]O[2] suggest that resistant and susceptible plants react to oxidative stress differently. Miller et al. (2010) stated that the hypersensitive response in the plant induces excessive H[2]O[2] and NADPH-dependent H[2]O[2] generation-related genes encoded by the ROS burst-associated homolog RBOH (respiratory burst oxidase homolog) [[199]49]. It is an important source of signal-transduction-associated ROS [[200]50, [201]51], supporting H[2]O[2]-modulated pathogen-induced ROS generation. M. phaseolina induced contrasting biosynthetic and defense signal pathways in the resistant and susceptible genotypes KEGG pathway enrichment analysis revealed that M. phaseolina down-regulated the isoflavonoid and flavonoid biosynthetic pathways in resistant and susceptible genotypes when comparing the non-pre-treated M. phaseolina-inoculated treatment with non-pre-treated mock-inoculated (agar plug) control. Studies have shown that pathogen-induced isoflavonoid biosynthetic genes [[202]52, [203]53] are up-regulated in response to pathogen invasion, contributing to the production of antimicrobial compound glyceollin, a phytoalexin [[204]54]. Additionally, flavonoid biosynthesis pathways play a crucial role in host adaptation [[205]55, [206]56], development [[207]57], reproduction [[208]58], and the alleviation of oxidative damage [[209]59, [210]60]. Our findings indicate that M. phaseolina infection induces a significant down-regulation of the phenylpropanoid biosynthesis pathway, with more than two-fold reduction in expression observed in both soybean genotypes. Phenylpropanoid pathways are known for their great potential in ROS homeostasis, as some polyphenols work as an antioxidant to scavenge or neutralize ROS [[211]61–[212]63], thus mitigating oxidative damage induced by the pathogen invasion. The polyphenols synthesized through flavonoid and phenylpropanoid pathways such as flavonoids and lignins, are integral to soybeans defense mechanisms against pathogens like M. phaseolina [[213]63, [214]64]. Furthermore, these compounds contribute to various aspects of plant physiology, including defense, development and stress responses [[215]63, [216]64]. This interconnected network of metabolic pathways exemplifies the diverse responses of plants to pathogen induced stress and highlights the significance of interconnected pathways in comprehending complex plant-pathogen interaction [[217]63, [218]64]. Mitogen-activated protein kinase signaling (MAPK) was also up-regulated with more than five-fold expression in the susceptible genotype. In contrast, the MAPK pathway was down-regulated in the resistant genotype, exhibiting around three-fold enrichment. The MAPK pathways play a crucial role in perceiving extracellular signal induced by pathogens, and transducing these signals to downstream defense mechanisms, leading to hypersensitive cell death to restrain pathogen proliferation [[219]65–[220]67]. The differential regulation of MAPK signaling in resistant and susceptible soybean genotypes suggests a complex interplay between the fungus M. phaseolina and the soybean responses which might underlie the phenotypic differences. Kovuton et al. (2000), demonstrated the critical involvement of MAPK cascades in plant responses to oxidative stress [[221]66]. The study showed that H[2]O[2]-regulated MAPK cascades in Arabidopsis leaf cells are activated by the MAPKKK, ANP1, through epitope tagging and a protoplast transient expression assay [[222]66]. This finding suggests that reactive oxygen species (ROS), such as H[2]O[2] can activate MAPK signaling pathway leading enhanced host defenses. The up-regulation of MAPK pathways in the susceptible soybean genotype might indicate an overactive defense response, potentially contributing to susceptibility through cellular damage, which could be exploited by M. phaseolina. Conversely, the down-regulation of the MAPK pathway in the resistant genotype could indicate a more controlled and effective signaling mechanism, enhancing the plant’s ability to counter pathogen attacks without incurring significant cellular damage. Glutathione peroxidases are known to work as antioxidants [[223]68] and are synthesized within the peroxisome organelles in response to oxidative stress triggered by biotic agents. The glutathione metabolic pathway was down-regulated by around three- and five-fold in the resistant and susceptible genotypes when comparing mock-inoculated versus M. phaseolina-inoculated treatments, suggesting induction by M. phaseolina. Glutathione peroxidases were essential in detoxifying ROS in sweet cherry when inoculated with the fungus Penicillium expansum [[224]69]. Thus, pathogen-induced ROS-mediated damage was alleviated from the harvested sweet cherry [[225]69]. The up-regulation of glutathione peroxidases and other antioxidants in tomato leaf peroxisomes in response to Botrytis cinerea also supports the notion of antioxidant-mediated defense activation against pathogen invasion [[226]70]. Ascorbic acid pre-treatment induced contrasting pathways in the resistant and susceptible genotypes Ascorbic acid pre-treatment followed by M. phaseolina inoculation versus the ascorbic acid pre-treated and mock-inoculated control expressed genes related to photosynthetic pathways in both genotypes. The major photosynthetic pathways induced by the ascorbic acid pre-treatment were photosynthesis, carotenoid biosynthesis, porphyrin, and chlorophyll metabolism in both genotypes, suggesting exogenous ascorbic acid pre-treatment modulated photosynthesis, which agrees with Foyer's (2015) statement "ascorbic acid as an essential antioxidant for photosynthesis" [[227]42]. Along with maintaining sustained photosynthesis, ascorbic acid is known to detoxify chloroplastic ROS and protect the photosynthetic machinery from ROS damage [[228]71–[229]73]. A genome-wide transcriptional profiling and metabolic pathway analysis from the oxidative stress induced by salinity and heat stress in tomato plants revealed the role of the ascorbate metabolic pathway in controlling redox homeostasis [[230]74], supporting our findings. However, ascorbic acid pre-treatment did not enrich the up-regulation of stress-responsive metabolic pathways, including flavonoid, isoflavonoid, and phenylpropanoid pathways, in either genotype. The circadian rhythm pathway was up-regulated with more than two-fold expression in the resistant genotype. As in the earlier discussion of ascorbate's role in ROS neutralization, circadian clock-associated genes are involved in mounting resistance against pathogens through the assistance of plant innate immunity [[231]75–[232]77], suggesting that enhanced oxidative stress might induce immunity responses in the resistant genotype, DT97-4290. H[2]O[2] pre-treatment following M. phaseolina inoculation induced contrasting pathways in the resistant and susceptible genotypes Hydrogen peroxide pre-treatment followed by M. phaseolina inoculation up-regulated the isoflavonoid, flavonoid, and phenylpropanoid biosynthesis pathways in both genotypes. Interestingly, each of those pathways is stress-responsive and was down-regulated in both genotypes when the non-pre-treated and M. phaseolina-inoculated versus the mock-inoculated control treatment was compared. This suggests that exogenous application of H[2]O[2] might induce the above-mentioned pathways. In the resistant genotype, the glutathione metabolism-related pathway was also up-regulated with more than three-fold expression. As mentioned previously, glutathione metabolism has a significant role in ROS detoxification [[233]78]. This suggests that the enhanced oxidative stress induced by exogenous application of H[2]O[2] followed by M. phaseolina inoculation, may influence the up-regulation of the ROS-neutralizing glutathione metabolic pathway. Furthermore, it implies that the resistant genotype might have a different arsenal for ROS neutralization. In the susceptible genotype, other than the defense-related metabolic pathways, MAPK signaling, and plant-pathogen interaction pathways were enriched around two-fold, both pathways are related to stress responses, host defense, and signal transduction. Conclusions Using RNA-seq assay, we assessed the gene expression in soybean stem tissues affected by M. phaseolina, with pre-treatments of ascorbic acid and hydrogen peroxide. Both M. phaseolina-resistant and susceptible soybean genotypes exhibited increased expression of defense-related genes following M. phaseolina inoculation. These genes were primarily related to receptor-like kinases, receptor-like cytoplasmic kinases, and other kinase-related activity. Additionally, many genes were linked to oxidative stress-responsiveness including peroxidases and antioxidants. Pre-treatment with ascorbic acid followed by M. phaseolina inoculation induced a significant up-regulation of genes related to photosynthesis and oxidative stress responses in both genotypes. This suggests that ascorbic acid induces the augmentation of photosynthesis genes and possibly stabilizes the host’s source-sink activity during biotic stress. Conversely, hydrogen peroxide pre-treatment followed by inoculation resulted in a substantial increase in oxidative stress-responsive genes, indicating an enhanced oxidative stress response due to the presence of reactive oxygen species. Overall, our study highlights the critical role of genes and pathways associated with kinase activity, photosynthesis, carbohydrate metabolism, ethylene-activated signaling, and oxidative stress responses in soybean resistance to M. phaseolina. Methods Soybean genotypes, plant growth, Macrophomina phaseolina isolates, experimental design, inoculation, and plant sample collection Two soybean genotypes DT97-4290 (moderately resistant to M. phaseolina) and Pharaoh (susceptible to M. phaseolina) were used for this experiment [[234]79, [235]80]. This experiment was conducted in the Throckmorton Plant Sciences Center greenhouse facilities at Kansas State University (KSU), Manhattan, Kansas, USA. The experiment was conducted using a randomized complete block design (RCBD) with a factorial arrangement, incorporating three biological replications. It involved two genotypes and six treatments, resulting in a 2 × 6 factorial design (Table [236]3). Table 3. Treatment scheme of soybean cut stem assay for the sample collection and RNA extraction Soybean cut stem assay Treatments Pre-treatment Inoculation (48 h after pre-treatment) 1 NPT Inoc 2 ASC Inoc 3 NPT Agar 4 ASC Agar 5 NPT H[2]O[2] 6 H[2]O[2] Inoc [237]Open in a new tab Abbreviations: Agar Mock-inoculated (agar plug only); H[2]O[2] Hydrogen peroxide; Inoc M. phaseolina-colonized PDA plug; NPT Non-pre-treated For each treatment, six soybean seeds were planted in a 5 × 5 in. (12.7 × 12.7 cm) pot with a vermiculite (Therm-O-Rock East, Inc., USA) soil mix (silt loam soil and vermiculite combined in a 1:1 ratio). Seedlings were thinned to four plants pot^−1 once they reached the cotyledonary stage (VC). The remaining seedlings were grown until the 2nd trifoliate (V2) stage. The M. phaseolina isolate used for this experiment (MP0336) is an aggressive isolate obtained from charcoal rot diseased soybean plants in Oct 2008 from the Ashland Bottoms Research Farm, KSU, Manhattan, Kansas, USA. The identity of the isolate was reconfirmed by PCR and sequencing using the M. phaseolina-specific primers MpKFI and MpKRI [[238]81]. The isolate was grown on a ¼^−strength potato dextrose agar (PDA) media for 5 days. The newly growing portions of the colony edge were cut into a circular cylinder using the 200 (µL) micropipette tip. The mycelium side of the agar plug was directly placed onto the cut stem point for the inoculation. The cut-stem inoculation method of Twizeymania et al. (2012) was followed for this experiment. The apical portion of V2 soybean seedlings was cut 25 mm above the unifoliate node [[239]82]. Two-hundred microliter (µL) micropipette tips, inverted and filled with one of the pre-treatments or treatments listed in Table [240]3, were placed on the cut portion of the seedling. Following stem clipping and pre-treatment, respective post-treatments were applied approximately 48 h after pre-treatment (Table [241]3). Stem tissue was collected 48 h after applying the final treatment (M. phaseolina inoculation, agar plug, or H[2]O[2] application) (Table [242]3). The stem was clipped from the unifoliate node region and immediately put in liquid nitrogen to flash freeze tissue. Stem tissues from three biological replicates across twelve conditions (2 genotypes × 6 treatments = 12) were collected, resulting in a total of 36 samples (3 × 12 = 36), each stored in separate 1.5 ml tubes. Each sample tube contained four stems from individual plants. Stem tissue was stored at −80 °C until RNA extraction. Sample preparation and RNA extraction Soybean stem tissues from each sample tube were pooled and ground using a mortar and pestle, continuously adding liquid nitrogen. Powdered stem tissues were transferred to 1.5 ml tubes and stored in the −80 °C freezer before RNA extraction. Total RNA was extracted from the soybean tissue powder using a Qiagen RNeasy Plant Mini Kit (Qiagen Inc., USA) following the manufacturer’s protocol. RNA quality and quantity were tested using the Nanodrop Spectrophotometer (Nanodrop One) (ThermoFisher Scientific, USA). The RNA integrity number (RIN) was checked using an Agilent 2100 Bioanalyzer (Agilent Technologies Genomics, USA). cDNA library preparation and Illumina sequencing One μg of high-quality total RNA was taken for RNA sequencing (RNA-seq) library construction using the TruSeq Stranded mRNA Library prep kit (Illumina Inc, San Diego, California, USA). RNA-seq libraries were barcoded with TruSeq RNA CD dual 8 bp indexes (Illumina, Inc.). Three pools, each containing twelve libraries, were created and subsequently subjected to sequencing using three high-output 75 cycle (1 × 75 bp) runs on the NextSeq 500 Sequencing System, following the manufacturers protocol (Illumina, Inc.). Library preparation and sequencing were conducted at the KSU Integrated Genomics Facility (IGF) lab (Department of Plant Pathology, Throckmorton Plant Sciences Building, KSU, Manhattan, Kansas). Raw sequence processing for quality check and soybean reference genome alignment The adapter trimmed single-end sequence reads were checked for phred quality score using the FastQC [[243]83] quality test. FastQC resulted in an average phred quality score > 30. Sequence data were not needed to perform any quality trimming. Next, sequence reads were aligned with the soybean (Glycine_max_v2.1) reference genome [[244]20]. The soybean genome FASTA file and gene annotation file (gtf file) were downloaded from the EnsemblPlants database [[245]84]. The genome FASTA file and annotation file were indexed bioinformatically using the clustered computer (Beocat, K-State) before aligning with the sequence read. STAR (Spliced Transcripts Alignment to a Reference) genome mapper software was used for sequence alignment with the reference genome [[246]85]. The STAR genome aligner generated reads per gene count data for all 36 samples. The reads per gene count data were used for downstream analysis. Differential gene expression analysis Differential gene expression analysis was performed using the reads per gene count data (derived after mapping with the reference genome) and the “DESeq2” R package [[247]86]. The DESeq2 tool uses a model based on the negative binomial distribution [[248]86]. The DEseq2 package allows the differential analysis of read count data per gene by estimating dispersion and fold changes [[249]86]. The differential gene expression tool DeSeq2 executes the statistical test from the read per gene count data to find the statistically significant genes, gene expression scenario (up- or down-regulation), and the magnitude of difference between each gene when pairwise comparisons are considered [[250]87]. All possible pairwise comparisons between samples were performed among the 36 samples to find the significantly and differentially expressed genes between two different conditions. To control the false discovery rates (FDR) at the 5% level between multiple comparisons and hypothesis testing, a q-value was determined [[251]88]. Differentially expressed genes with q-value < 0.05 were considered to control FDR at the 5% level. The pairwise comparison for differential expression was performed between two different treatments of the same genotype (resistant or susceptible). Genes with a q-value < 0.05 were considered as significantly and differentially expressed (null hypothesis rejected) with 5% FDR. Functional gene annotation using gene ontology GO term and KEGG pathway analysis using the ShinyGO bioinformatics platform Gene ontology (GO) term analysis was performed using the R software package “goseq” to detect significantly enriched GO terms for differentially expressed genes from the DEseq2 analysis. The gene to GO term (GO functional annotations) was obtained using the genes text file and xref.genes file from EnsemblPlants [[252]84] and running the gene to GO term code in R software. The significantly enriched GO terms were based on a p-value cutoff of 0.05. Gene ontology GO terms differentiated genes into their respective biological processes (BP), molecular functions (MF), and cellular components (CC). The genes associated with the enriched GO terms were pasted on the ShinyGO (ShinyGO 0.77) bioinformatic online platform from South Dakota State University to discover Kyoto Encyclopedia of Genes and Genomes (KEGG)-associated pathways and gene functions [[253]23, [254]24]. This KEGG pathway analysis is based on the number of genes associated to the respective pathway with their log-fold enrichment. Transparency statement Portions of this manuscript are derived from the first author's dissertation: "Understanding the physiological and molecular aspects of charcoal rot resistance mechanisms in sorghum and soybean," archived at [255]https://krex.k-state.edu/items/581d1f57-063f-47cc-b09c-e8b31223c60 8. Supplementary Information [256]12864_2024_11023_MOESM1_ESM.jpg^ (185.5KB, jpg) Additional file 1: Fig. S1. P value histograms of the informative genes of normalized read counts for the treatment comparisons of the resistant genotype [R-G] DT97-4290: (Panels A-C) for the non-pre-treated [NPT] mock-inoculated [Agar] vs. M. phaseolina-inoculated [Inoc] (A), ascorbic acid [ASC] pre-treated inoculated [Inoc] vs. mock-inoculated [Agar] (B), non-pre-treated H[2]O[2 ]vs. H[2]O[2] pre-treated M. phaseolina -inoculated treatment (C). In the susceptible genotype [S-G] Pharaoh: (Panels D-F) for the non-pre-treated mock-inoculated [Agar] vs. inoculated [Inoc] (D), ascorbic acid pre-treated [ASC] inoculated [Inoc] vs. mock-inoculated [Agar] (E), and non-pre-treated [NPT] H[2]O[2] vs. H[2]O[2] pre-treated inoculated [Inoc] treatment (F). [257]12864_2024_11023_MOESM2_ESM.png^ (362.3KB, png) Additional file 2: Fig. S2. Volcano plots displaying differential gene expression of log2 fold changes versus negative log10 p-values for pairwise comparisons within the resistant genotype DT97-4290 for M. phaseolina-inoculated [Inoc] vs. mock-inoculated [Agar]; ascorbic acid pre-treated [ASC] inoculated [Inoc] vs. mock-inoculated [Agar]; and non-pre-treated [NPT] and H[2]O[2]-treated vs. H[2]O[2] pre-treated and inoculated [Inoc]; in the susceptible genotype Pharaoh for inoculated [Inoc] vs. mock inoculated [Agar]; ascorbic acid pre-treated and inoculated [Inoc] vs. mock-inoculated [Agar]; and non-pre-treated [NPT] and H[2]O[2]-treated vs. H[2]O[2] pre-treated and inoculated [Inoc]. Blue dots indicate significantly up-regulated genes, while pink dots represent significantly down-regulated genes. Genes below the yellow p-value threshold line (0.05) are considered non-significant. [258]12864_2024_11023_MOESM3_ESM.png^ (822.8KB, png) Additional file 3: Fig. S3. Flavonoid biosynthesis pathway-associated genes (highlighted) induced in the resistant genotype when comparing the non-pre-treated M. phaseolina-inoculated versus the non-pre-treated mock-inoculated control. [259]12864_2024_11023_MOESM4_ESM.png^ (809.9KB, png) Additional file 4: Fig. S4. Flavonoid biosynthesis pathway associated genes (highlighted) induced in the susceptible genotype when comparing the non-pre-treated M. phaseolina-inoculated versus the non-pre-treated mock-inoculated control. [260]12864_2024_11023_MOESM5_ESM.png^ (2.1MB, png) Additional file 5: Fig. S5. Photosynthesis pathway associated genes (highlighted) induced in the resistant genotype when comparing the ascorbic acid pre-treated and M. phaseolina-inoculated versus the ascorbic acid pre-treated mock-inoculated control. Acknowledgements