Soil sterilization accumulates higher content of antibacterial compound and recruits beneficial fungus for konjac soft rot disease management by multi-omics analyses Authors: Chengfen Nong, Sauban Musa Jibril, Xinyun Yang, Lu Huang, Hao Qu, Xingrun Yang, Ruoping Wang, Xiaojuan Wu, Chengyun Li, Yi Wang Published: 2025 ABSTRACT AbstractKonjac soft rot is a typical soil-borne disease, and while soil sterilization is known to reduce pathogen inoculum and disease incidence, the underlying mechanisms remain unclear. This study investigates how sterilization alters soil metabolite profiles and microbial communities to suppress pathogens and promote konjac health. We identified the antibacterial compound 1,4-naphthoquinone, which accumulates more abundantly in sterilized soil. Transcriptome analysis revealed that 1,4-naphthoquinone disrupts gene expressions involved in protein synthesis, thereby inhibiting the occurrence of soft rot disease. Furthermore, microbiome analysis of sterilized replanting soil showed significant changes in microbial community structures. Beneficial fungi for disease management are favored to recolonize in sterilized replanting soil. An antagonistic Aspergillus sp. NCF-16 was identified, demonstrating strong inhibitory effects on soft rot disease management. Consequently, our findings suggest that soil sterilization can foster a novel soil environment by enhancing antibacterial compounds and promoting the colonization of beneficial microorganisms for effective soil-borne disease management.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12870-025-07985-7.Keywords: Soil sterilization; Soft rot disease; 1,4-naphthoquinone; Beneficial microorganisms Introduction Konjac, as a fundamentally economic crop, is extensively cultivated throughout Asia [1]. Konjac food products have gained global popularity among consumers due to their high concentration of konjac glucomannan (KGM), earning them the distinction of being named one of the “top 10 health foods” by the World Health Organization [2]. KGM is regarded as an indigestible polysaccharide that may possess significant health benefits for humans [3]. Soft rot disease poses a significant constraint on the production and cultivation of konjac, leading to substantial economic losses. Outbreaks of this disease exhibit a range of symptoms, including the yellowing and wilting of plants, tissue decay, and the release of a foul odor. Infected konjac plants, particularly their leaves, petioles, and bulbs, often display yellowing and wilting of leaves, soft rot in the internal petiole tissues and subterranean tubers, culminating in the death and rotting of the entire plants [4, 5]. The average incidence rate of bacterial soft rot in fields exceeds 35%, potentially reducing konjac yield by 30–70%, or even up to 100%. While most commercial konjac varieties are derived from the Amorphophallus konjac species, which is susceptible to soft rot pathogens, a recent discovery has identified another species, A. muelleri, that exhibits resistance to soft rot disease. Furthermore, the resistant mechanisms have been extensively studied at multi-omic levels [5–7]. Several pathogens have been reported to cause konjac soft rot, with Pectobacterium spp., Dickeya spp., and Erwinia spp [8, 9]. Pectobacterium spp. belongs to the soft-rot Enterobacteriaceae family. The pathogenic mechanisms have been extensively characterized. Most strains possess a vast array of plant cell wall-degrading enzymes (PCWDEs) as their key virulence factors for infection [4]. Several soft-rot phytopathogens encode predicted effectors that share homology with pectinases and other potential PCWDEs. These enzymes are prominently featured in the genomes of Dickeya dadantii and D. solani and are anticipated to be delivered via Type VI Secretion System (T6SS) [10]. Furthermore, an analysis of seven Pectobacterium spp. strains identified cellulases and a range of pectin-degrading enzymes as potential T6SS effectors, which were prevalent across the majority of the strains [11]. The functions of T3SS were also identified. The T3SS can induce host cell death to facilitate leaf maceration in P. carotovorum. Recent findings suggest that maceration occurs due to the suppression of callose deposition in the plant, which is related to resistance by pathogenesis-related genes. While P. carotovorum PccS1 produces pectinolytic enzymes [12]. Moreover, the bacterial flagellum represents a sophisticated molecular apparatus with diverse functions in pathogenesis, encompassing the ability to reach optimal host sites, facilitate colonization or invasion, maintain residency at the infection site, and enable dispersal post-infection [13]. Quorum sensing (QS) system has been well identified in P. carotovorum subsp. carotovorum which plays a pivotal role in regulating the expression of PCWDE genes, virulence factors, and biofilm formation [14]. N-acylhomoserine lactones (AHLs), which serve as QS activation signals, accumulate to critical threshold concentrations as the population density increases, thereby triggering the activation of pathogenicity-related genes [15, 16]. Consequently, the utilization of reduced AHL-like QS compounds emerges as a promising approach for effective disease management [17, 18]. Soil is teeming with biological activity and diversity, particularly among microbes, endowing it with the capacity to perform numerous valuable functions. Roughly 108 to 109 bacteria, 105 to 108 actinomycetes, 10 to 102 nematodes, and 103 to 105 invertebrates can be found in a mere gram of soil [19]. This vast microbial population can be broadly categorized as beneficial, neutral, or harmful to plants. Soil-borne disease management is a tough challenge for researchers and farmers. Soil ecosystems contain abiotic, biotic factors and their complex interactions that aggravate the difficulties for soil-borne disease management [20]. Soil sterilization is effectively used to decrease the pathogen inoculum and alleviate disease incidence in soil [21]. Nevertheless, it is essential to elucidate the mechanism that soil sterilization influences the relationship between disease incidence and soil factors. Konjac soft rot disease is a typical soil-borne disease, and soil sterilization has been proven to reduce its incidence. While the reduction of pathogen inoculum is a known benefit, the effects of soil sterilization on the changes of metabolome and microbiome in soft rot disease soil remains relatively unexplored. This study reveals that soil sterilization not only reshapes the soil metabolic profile but also elevates the concentration of the antagonistic compound 1,4-naphthoquinone, which hinders protein synthesis by disrupting ribosomes. Furthermore, the fungal communities in konjac replanting soil that has undergone sterilization exhibit significant alterations when compared to those in diseased soil. Notably, beneficial fungi such as Aspergillus, Penicillium, and Trichoderma are more abundant in sterilized replanting soil, leading to the formation of a more intricate network. We have successfully isolated an Aspergillus strain, designated as NCF-16, which exhibits a high inhibitory capability against Pcc. Consequently, our findings indicate that soil sterilization fosters the rebuilding of a reinvigorated soil metabolome and microbiome, conducive to the healthy growth of konjac and effective management of soft rot pathogens. Materials and methods Soil collection and treatment for metabolome Soil samples for metabolome analysis were meticulously collected from three distinct konjac cultivation fields in Fuyuan County, Yunnan Province. Sample collection was performed in accordance with local guidelines and did not require specific permits, as the study involved minimal disturbance to agricultural land with the consent of the cooperating farmers. These samples, taken from a depth of 10 cm, came from areas primarily affected by konjac soft rot disease, which exhibited visible symptoms including leaf yellowing, petiole collapse, and bulb decay. For sterilized soil preparation, the diseased soil samples from the three distinct fields were mixed together and underwent autoclaving at 121℃ for 20 min. Metabolite extraction For metabolite extraction, 50 mg of soil (wet weight) was processed with methanol: water (4:1, v: v) containing 0.02 mg/mL L-2-chlorophenylalanine as an internal standard. To ensure thorough extraction, the samples underwent grinding using a frozen tissue grinder for 6 min at -10 °C and 50 Hz. This was followed by a low-temperature ultrasonic extraction process lasting 30 min at 5 °C and 40 kHz. Subsequently, the samples were cooled at -20 °C for 30 min and then centrifuged for 15 min at 4 °C and 13,000 g to separate the supernatant. This supernatant was subsequently utilized for analysis via UHPLC-MS/MS. For quality control purposes, Quality control (QC) samples were prepared by mixing equal volumes of the test samples. UHPLC-MS/MS analysis The sample was subjected to UHPLC-MS/MS analysis on a Thermo UHPLC-Q Exactive HF-X system, which was coupled with an ACQUITY HSS T3 column (100 mm × 2.1 mm i.d., 1.8 μm) from Waters, USA. This analysis was carried out at Majorbio Bio-Pharm Technology Co. Ltd., located in Shanghai, China. For mass spectrometric data acquisition, the system was equipped with a Thermo UHPLC-Q Exactive HF-X Mass Spectrometer featuring an electrospray ionization (ESI) source that operated in both positive and negative modes. The raw data processing was handled by Progenesis QI software (Waters, Milford, USA). Furthermore, metabolite identification was accomplished by rigorously searching through databases such as HMDB (http://www.hmdb.ca/), Metlin (https://metlin.scripps.edu/), and Majorbio database (https://www.majorbio.com/flow). Significantly different metabolite identification Principal component analysis (PCA) and orthogonal least partial squares discriminant analysis (OPLS-DA) were performed using the R package ropls (v1.6.2). The significantly different metabolites were determined on the variable importance in the projection (VIP) scores obtained by the OPLS-DA model and the p-value calculated using Student’s t test (VIP > 1, p < 0.05). Inhibitory experiments of metabolites in vitro The stored bacterial isolates were recovered on LB agar medium by streak plating. The single colony of bacterium was selected and transferred into liquid LB medium, which was then incubated at 28℃ with shaking at 180 rpm for a duration of 2 days. Subsequently, 7 metabolites including L-theanine, phthalamic acid, bisphenol A, 1,4-naphthoquinone, 1-naphthylamine, ferulic acid and 3-indolybutyric acid were individually introduced into fresh liquid LB medium at final concentration of 1 µg/mL, 10 µg/mL and 100 µg/mL to assess their inhibitory effects on bacterial growth. For the inhibition assay, 100 µL of the P. carotovorum (~ 1.4 × 109 CFU/mL) suspension was inoculated into 100 mL of fresh liquid LB medium containing the test metabolite. This mixture was then incubated at 28℃ while being shaken at 180 rpm. After a 2-day incubation period, the optical density (OD600) of the bacterial suspension was measured to evaluate the impact of the metabolites on bacterial proliferation. Inhibitory effect of 1,4-naphthoquinone on rot disease by pot experiments To assess the inhibitory effects of 1,4-naphthoquinone on P. carotovorum during konjac growth, a series of pot experiments were meticulously designed. Konjac bulbs were collected from Fuyuan County growing field, Yunnan Province. The bulbs were washed under running tap water for 2 hours, after which the bulb surfaces were surface-sterilized by soaking in 75% alcohol for 30 s, followed by eight rinses with sterile distilled water. Each bulb was carefully planted into sterilized soil within a pot and allowed to grow for 15 days. Subsequently, 20 mL of 1,4-naphthoquinone solution was introduced into the soil at a concentration of 10 µg/g of soil. As a control, an equivalent volume of bacterial suspension (~ 8 × 10⁷ CFU/mL) without the 1,4-naphthoquinone additive was added to separate pots. After 15 days, the disease rates were investigated. . Disease rates were recorded from 8 konjac plants for each treatment, with the entire experiment being replicated 3 times. RNA sequencing The transcriptome samples treated with 1,4-naphthoquinone were prepared according to the following procedure. A 100 µL suspension of Pcc was cultivated in 100 mL of LB liquid medium for 24 h at 28℃ and a shaking speed of 180 rpm. Subsequently, 1,4-naphthoquinone was added to the culture to achieve a final concentration of 10 µg/mL and incubated for an additional 4 h. As a control, an equivalent volume of DMSO was used in place of 1,4-naphthoquinone. The harvested bacteria were flash-frozen in liquid nitrogen and dispatched to Shanghai Majorbio Bio-pharm Technology Co., Ltd. for sequencing. Each experimental trial was conducted in 3 biological repeats. RNA sequencing was carried out using the Illumina NovaSeq 6000 Sequencing Systems, and the resultant data underwent a low-quality filter process. Bowtie2 software was employed for mapping, with the reference genome being P. carotovorum subsp. carotovorum (GCF_009931215.1). To identify differentially expressed genes (DEGs), the DESeq2 tool was utilized for analysis. Soil sample collection and treatment for microbiome The konjac bulbs were planted in both diseased and sterilized soil and allowed to grow as seedlings for 15 days, with the soil undergoing two distinct treatment methods. One method involved the addition of a bacterial suspension, whereas the other added with 1,4-naphthoquinone into soil at 10 µg/g final concentration. Following the 7-day inoculation period, bulk soil samples were gathered from the following conditions: (1) untreated diseased or sterilized soil (CKD or CKS), (2) diseased soil or sterilized soil treated with bacteria (PAD or PAS), (3) and diseased soil or sterilized soil treated with 1,4-naphthoquinone (NQD or NQS). These soil samples were subsequently stored at -80 °C. DNA extraction Total microbial genomic DNA was extracted from soil samples using the E.Z.N.A.® soil DNA Kit (Omega Bio-tek, Norcross, GA, U.S.). The quality and concentration of DNA were assessed by 1.0% agarose gel electrophoresis and a NanoDrop2000 spectrophotometer. The high-quality genomic DNA with high integrity was sent to Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China) PCR amplification The hypervariable region V3-V4 of the bacterial 16 S rRNA gene were amplified with primer pairs 338 F (5’-ACTCCTACGGGAGGCAGCAG-3’) and 806R(5’-GGACTACHVGGGTWTCTAAT-3’).For fungi, the internal transcribed spacer was amplified using primer pairs ITS1F (5’-GCATCGATGAAGAACGCAGC-3’) and ITS2R (5’-TCCTCCGCTTATTGATATGC-3’). The PCR product was extracted from 1% agarose gel and purified using the PCR Clean-Up Kit (YuHua, Shanghai, China) and quantified using Qubit 4.0 (Thermo Fisher Scientific, USA). Amplicon sequence processing and analysis Purified amplicons were pooled in equimolar amounts and paired-end sequenced on an Illumina nextseq platform (Illumina, San Diego, USA) according to the standard protocols by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). After demultiplexing, the resulting sequences were quality filtered with fastp (0.19.6) [22] and merged with FLASH (v1.2.11) [23] and then were de-noised using DADA2 [24] plugin in the Qiime2 [25] pipeline with recommended parameters, which obtains single nucleotide resolution based on error profiles within samples. Taxonomic assignment of ASVs (Amplicon Sequence Variants) was performed using the Naive bayes consensus taxonomy classifier implemented in Qiime2 and the SILVA 16 S rRNA database (v138). The metagenomic function was predicted by PICRUSt2 [26] based on ASV representative sequences. Bioinformatic analysis of the soil microbiota was carried out using the Majorbio Cloud platform. Based on the ASVs information, rarefaction curves and alpha diversity indices Chao1 richness was calculated with Mothur (v1.30.1) [27].The similarity among the microbial communities in different samples was determined by principal coordinate analysis (PCoA) based on Bray-curtis dissimilarity using Vegan (v2.5-3). The linear discriminant analysis (LDA effect size (LEfSe) [28] was performed to identify the significantly abundant taxa of bacteria among the different groups (LDA score > 2, P < 0.05). The co-occurrence networks of fungi and bacteria were constructed to explore the internal community relationships across the samples, respectively [29]. A correlation between two nodes was considered to be statistically robust if the Spearman’s correlation coefficient over 0.6 or less than − 0.6, and the p < 0.01. Results and discussion Identification of metabolites In order to gain a deeper insight into the soil metabolomic features, we conducted untargeted metabolomic analyses using UHPLC-MS/MS. A comprehensive total of 22,443 peaks were identified, comprising 9,046 positive ion peaks and 13,397 negative ion peaks across all samples (Table S1). Our analysis led to the identification of 2,987 metabolites, which belonged to 16 KEGG pathways and 24 compound classifications based on the annotation database (Figure S1 and Table S2). Notably, 2,206 metabolites were detected in all disease soils from the three collections (D1, D2, and D3) and the pooled sterilized soil (S), while 348 compounds were exclusively identified in sterilized soil (S) (Fig. 1a). Correlation analyses revealed strong correlations among biological replicates (Fig. 1b, Table S3). Furthermore, principal component analysis (PCA) demonstrated a clear separation between metabolites from sterilized soil and those from diseased soil (Table S4). Interestingly, samples from disease soil 1 and disease soil 2 exhibited relatively closer relationships, whereas samples from disease soil 3 showed distinct individual separation. The first principal component accounted for 58.10% of the variance between sterilized soil and diseased soil (Fig. 1c, Table S4). The clustering of disease soil 1 and 2 might be related to the similarities in nutrient profiles due to fertilizer application or potentially similar microbial and metabolic profiles [30]. Consistent findings were observed through PLS-DA, although a slightly greater distance was noted between disease soil 1 and disease soil 2 (Fig. 1d, Table S5). These results suggest that soil sterilization significantly impacts metabolites, and different diseased soils may also possess distinct characteristic components. Soil metabolomic profiles distinguish sterilized and diseased konjac soils. a Venn plot of identified metabolites among different soil samples. b Correlation analyses of identified metabolites among different soil samples. Principal component analysis (c) and Partial least squares-discriminant analysis (d) plots of identified metabolites among different soil samples Differentially abundant metabolite analysis Konjac soft rot disease, being the most perilous soil-borne pathogen, significantly curtails konjac production [31]. Soil sterilization emerges as a potent strategy to mitigate the incidence of this disease. This process not only diminishes the population of pathogens and other microbes but also alters the metabolic profile of the soil, fostering an environment that suppresses pathogen growth. To delve into and discern the differentially abundant metabolites (DAMs) present in disease soil as opposed to sterilized soil, we identified 1568, 1593, and 1548 DAMs across various comparisons (Fig. 2a), with 1355 DAMs being shared among them (Fig. 2b, Table S6). Furthermore, our analysis uncovered 18 distinct categories of DAMs with the majority appearing consistently across all comparisons. Notably, however, beta-lactams, eicosanoids, and DAMs associated with 19-carbon atoms were found to be uniquely distributed in certain comparisons (Fig. 2c). Based on KEGG analyses, DAMs involving amino acid pathways were enriched in all comparisons (Figure S2). The differentially abundant (DA) score serves as a gauge reflecting the fluctuations of all metabolites within specific pathways. Upon examining these scores, it became evident that DA scores of nearly all DAM-enriched pathways were negative (Fig. 2d). This trend was particularly pronounced for amino acid-related pathways which was consistent with our compound classification findings. Comparative metabolomic profiling of different soils. a Analysis of down or up regulated DAMs in 3 disease soil compared with sterilized soil. b Shared and unique DAMs among diseased and sterilized soils. c The categories of DAMs in all comparisons. d The differentially abundant score of all metabolites within specific pathways Classification of differentially abundant metabolites Based on the classifications of DAMs, we found DAMs belonging to amino acid, carboxylic acid, phospholipid and monosaccharide were consistently present across all comparisons. Consequently, we focused on these DAMs in these four classifications. Amino acids serve as the primary constituents of proteins. Our research revealed that the concentrations of citrulline, argininosuccinic acid, phenylalanine, L-dopa, L-tyrosine, pyroglutamic acid, and L-leucine were all lower in disease soils compared to their sterile counterparts. Additionally, L-isoleucine and D-isoleucine exhibited reduced accumulation in disease soils 1 and 3. Interestingly, L-tryptophan levels were elevated in disease soils 2 and 3, whereas changes in the content of L-glutamate and L-glutamic acid were exclusive to disease soil 3 when compared to sterile soil (Fig. 3a). Regarding monosaccharide content variations (Fig. 3b), we observed decreases in glucosamine, rhamnose, 2-Deoxy-D-glucose, D-erythrose, and muramic acid across all diseased samples. Conversely, D-sorbitol levels were found to be higher. In terms of lipid-related metabolites, a total of 13 DAMs were detected (Fig. 3c). Notably, LysoPC(P-18:1(9Z)/0:0) levels were reduced, while PE(18:3(9Z,12Z,15Z)/15:0), PS(14:1(9Z)/20:1(11Z)), PG(20:4(5Z,8Z,11Z,14Z)/20:4(5Z,8Z,11Z,14Z)), and PE(14:0/20:5(5Z,8Z,11Z,14Z,17Z)) levels were increased in all diseased soil samples. The remaining lipid-related DAMs were only identified in select samples. Furthermore, our analysis of carboxylic acid-related metabolites showed that D(-)-beta-hydroxybutyric acid, butyric acid, and 4-hydroxybutyric acid accumulated more in diseased soils. Conversely, 2-oxovaleric acid, acetoacetic acid, oxoadipic acid, trans-aconitic acid, valeric acid, and succinic acid semialdehyde were detected at lower concentrations in disease soil samples (Fig. 3d). Changes in differentially abundant metabolites belonging to amino acid a, monosaccharide b, phospholipid (c) and carboxylic acid (d) across diseased and sterilized soils comparisons Comparative analysis has revealed higher concentrations of amino acids, sugars, and carboxylic acid metabolites in sterilized soil, whereas phospholipid-like metabolites were less prevalent (Fig. 3). These findings echo observations in soils suppressing and not suppressing Rhizoctonia solani AG8, a pathogen affecting cereal crops. Suppressive soils exhibited a richer sugar content, whereas non-suppressive soils had higher lipid and terpene levels [32]. In the metabolome analysis of the tomato rhizosphere, it was observed that sugars were more prevalent in the healthy tomato rhizosphere compared to that of diseased plants [33]. Furthermore, similar research has demonstrated a negative correlation between sugar compounds and the high abundance of Ralstonia solanacearum pathogen in the tomato rhizosphere [34]. Additionally, our findings reveal the differences in main compound classification between sterilized and diseased soils, suggesting that metabolomics could be employed to improve soil conditions conducive for disease-free konjac growth. 1,4-naphthoquinone has inhibitory effects on bacterial growth and disease occurrence Metabolome analysis revealed that a plethora of accumulated metabolites were detected in sterilized soil, which could potentially inhibit bacterial growth. Among them, seven metabolites were chosen for further testing. Notably, the samples treated with 1,4-naphthoquinone at concentrations of 10 and 100 µg/mL exhibited the most significant reduction in P. aroidearum (Pa) and P. carotovorum (Pcc) growth (Fig. 4a and b) but the inhibitory effects of other six metabolites were not effective (Figure S3). Given that 1,4-naphthoquinone is known to be a raw material for pesticide production, 1,4-naphthoquinone and its derivatives are prevalent in numerous naturally occurring alkaloids and have been identified as promising antibacterial agents [35, 36]. These findings suggest that soil sterilization may lead to the accumulation of metabolites that inhibit the growth of pathogens. Validation of 1,4-naphthoquinone compound antimicrobial activity against soft rot pathogens. a and b The inhibitory effects on the growth of Pa and Pcc. Pathogen suppression on konjac tuber surfaces infected with Pcc c and Pa d. The effects of on Pcc inoculated on the konjac in the pots e and f. Each experiment was conducted with 3 biological repeats. Significant differences were calculated by Student’s t-test, *p < 0.05. Error bars represent the means ± SD To evaluate the inhibitory effects of 1,4-naphthoquinone on Pa and Pcc during infection, bacterial suspensions were mixed with varying concentrations of 1,4-naphthoquinone and inoculated onto the konjac surface (Fig. 4c and d) as well as into the soil (Fig. 4e and f). The disease lesions were decreased significantly on the konjac surface with 1,4-naphthoquinone (Fig. 4c and d). Previous studies have also demonstrated that soil concentrations of 1,4-naphthoquinone below 20 mg/kg appear to be environmentally safe [37]. The effective inhibitory concentrations employed in our study also fall within this environmentally secure range. Moreover, addition of 1,4-naphthoquinone into pathogen inoculated soil could decrease the disease occurrence. The addition of 1,4-naphthoquinone promoted healthy seedling growth, whereas Pcc inoculation without treatment resulted in root suppression and clear disease symptoms (Fig. 4e and f). The findings further demonstrated that the incorporation of 1,4-naphthoquinone led to a reduction in the incidence of soft rot disease. Transcriptome analysis Previous results showed that 1,4-Naphthoquinone and its derivatives exert their effects by disrupting cell wall synthesis, inhibiting protein production, suppressing nucleic acid synthesis, inducing apoptosis, enhancing reactive oxygen species (ROS) generation, and compromising cell membrane integrity [35, 38 ,39]. We conducted a transcriptome analysis to investigate the inhibitory mechanisms of 1,4-naphthoquinone on Pcc. Our findings revealed that upon treatment with 1,4-naphthoquinone, there were 895 upregulated genes and 891 downregulated genes in Pcc (Figure S4, Table S7). Through GO-enrichment analysis, we observed that the downregulated genes were primarily involved in peptide biosynthesis and metabolic processes, RNA binding, and ribosome and translation-related GO terms, all of which are closely linked to protein synthesis. Hfq was identified as an RNA binding protein in Pcc, regulating PCWDE activities, T6SS gene expressions, and bacterial competition [40]. Moreover, the ribosome protein RplY is essential for Pcc virulence [41]. These results indicated the importance of RNA binding and ribosome genes for bacterial virulence. In contrast, the upregulated genes were predominantly associated with transport and S-adenosylmethionine metabolism. Furthermore, KEGG pathway analysis corroborated these results, showing that downregulated genes were enriched in the ribosome pathway, while upregulated genes were mainly classified under ABC transporters. Soft rot pathogens typically encode over 80 ATP-binding cassette transporters per genome, representing a 30% increase compared to closely related animal pathogens. ABC transporters are required for resistance to antimicrobial compounds, chemotaxis, gene regulation, plant cell wall degradation, plant and bacterial hormone secretion and nutrient uptake [42]. The induction of ABC transporter genes could potentially reduce the intracellular concentration of 1,4-naphthoquinone as a compensatory mechanism. We have also delved into the expression patterns of various pathogenicity-related gene sets, such as T2SS, T3SS, T6SS, pectate lyase, elongation, and flagellar genes (Figure S5). Notably, the differentially expressed genes (DEGs) associated with T3SS and elongation-related gene sets were uniformly downregulated. Previous research suggested that elicitation of programmed cell death by a type III secretion system (T3SS) in plant leaves can promote virulence in Pcc [43]. P. carotovorum uses the T3SS to induce plant cell death in order to promote leaf maceration rather than to suppress plant immunity [12]. However, the DEGs in the remaining four gene sets did not exhibit uniform regulatory trends. Consequently, these findings suggest that 1,4-naphthoquinone specifically interferes with the gene expressions of T3SS and elongation to inhibit bacterial growth. P. carotovorum and Escherichia coli are both members of the Enterobacterales, signifying their close genetic affiliation. In an inhibitory test of protein synthesis using 1,4-naphthoquinone, E. coli cells were tagged with GFP. Observations of GFP fluorescence intensity revealed a significant decrease in E. coli following treatment with 1,4-naphthoquinone (Fig. 5e). Consequently, transcriptome and fluorescence microscopy analysis indicate that 1,4-naphthoquinone treatment may suppress the expression of genes involved in protein synthesis, thereby disrupting bacterial growth. Transcriptomic response of Pcc to 1,4-naphthoquinone treatment. Gene Ontology (GO) enrichment analysis of (a) upregulated and (b) downregulated differentially expressed genes (DEGs). Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis of (c) upregulated and (d) downregulated DEGs. e The effects of 1,4-naphthoquinone on green fluorescent protein expressions in E. coli Amplicon sequencing of Konjac planting soil To combat soil-borne diseases, sterilization is frequently employed to sanitize the soil environment and reduce the pathogen inoculum. Moreover, soil sterilization can create a “biological desert” conducive to microbial re-colonization, ultimately benefiting plant growth and diminishing pathogen colonization [44, 45]. Through our soil treatment and sample collection and sequencing, we collected 928,931 rRNA and 1,254,375 reads from 18 soil samples, encompassing 1480 bacterial and 664 fungal ASVs (Fig. 6a, Table S8). Previous results showed that soil sterilization could change the composition and structure of soil microbiome [46]. Our analysis delved into the α diversity across these samples. Notably, the Chao index of bacterial communities revealed a significant difference solely between NQD and NQS (Fig. 6b). The significant difference in bacterial diversity between NQD and NQS likely reflects the interactive effect of soil sterilization and 1,4-naphthoquinone treatment on microbial community structure with sterile soil allowing for more open ecological niches and greater recolonization potential, even in the presence of 1,4-naphthoquinone than NQD potentially due to the selective inhibitory effects of 1,4-naphthoquinone on specific bacterial taxa within an already complex community. In contrast, the Chao index of fungal communities exhibited significant disparities between disease soil and sterilized soil under varying treatments (Fig. 6c). These findings suggest that soil sterilization exerts a more pronounced effect on the composition of fungal communities, whereas bacterial inoculation and the addition of 1,4-naphthoquinone appear to have no significant impact. Principal Coordinates Analysis (PCoA) was employed to conduct β diversity analysis among the various samples. The findings revealed that both fungal and bacterial communities, identified from diseased soil and sterilized soil, were distinctly segregated along the first principal coordinate (Fig. 6d and e). Previous studies on the effects of sterilization and maturity of compost on soil bacterial and fungal communities and wheat growth implied that elimination of native soil microorganisms by sterilization enhanced the prevalence and abundance of compost-associated microorganisms. Adding fresh compost increased wheat biomass production, but the positive effects of compost on plant growth were strongest when sterile composts were used [47]. Thus, our observation suggests that soil sterilization exerts a significant selective influence on the composition of the soil microbiome. Microbiome analyses in konjac planting soil under different treatments. a The diagram plot showing the konjac planting soil with different treatments. The Chao α-diversity index among various treatments in bacterial and (b) fungal (c) communities. PCoA analyses of various treatments in bacterial and (d) fungal (e) communities. The asterisk indicates a significant difference among groups according to Kruskal-Walls H test. *, p<0.05. **, p<0.01. ***, p<0.001 Bacterial and fungal community compositions and network structures Previous results showed that there were common plant-beneficial microbial functions that were promoted when peach, corn and tomato were planted in the sterilized soil [48]. The variations in the composition of bacterial and fungal communities are closely linked to soil sterilization. Regarding bacterial community shifts, soil sterilization resulted in a reduced presence of genera Bacillus, Ammoniphilus, Geobacillus, and Hyphomicrobium across all treatments. Conversely, the genera Sphingomonas and Streptomyces exhibited an increased presence (Fig. 7a). Turning to fungal community changes, Fungi_gen_Incertae_sedis, Pyxidiophora, and Rozellomycota_gen_Incertae_sedis demonstrated high abundances in diseased soil. In contrast, Aspergillus, Torula, Penicillium, and Trichoderma were more prevalent in sterilized soil (Fig. 7b). Analysis of bacterial and fungal community composition in diseased soil and sterilized soil with different treatments. Changes in the bacterial (a) and fungal (b) microbial communities at the genera level. Multiple comparisons analysis among significantly different bacterial (c) and fungal (d) genus in diseased and sterilized soil with different treatments. The significant differences among all groups were analyzed by the Kruskal-Walls H test. *, p<0.05. **, p<0.01 Multiple comparisons analysis is employed to identify significantly distinct species across various treatments. Notable differences in the composition of bacterial genera, including Bacillus, Ammoniphilus, Geobacillus, Hyphomicrobium, Sphingomonas, Streptomyces, Pseudoxanthomonas, norank_f__67 − 14, Microvirga, and Nocardioides, were observed between diseased and sterilized soils (Fig. 7c). Similarly, the composition of fungal genera such as Fungi_gen_Incertae_sedis, Pyxidiophora, Rozellomycota_gen_Incertae_sedis, Aspergillus, Torula, Cladosporium, Geminibasidiales_gen_Incertae_sedis, Fibulochlamys, Trichoderma, and Saitozyma exhibited significant disparities between diseased and sterilized soils (Fig. 7d). Some isolates of Streptomyces, Trichoderma, and Aspergillus exhibit inhibitory effects on the growth of Pcc [18, 49, 50]. These results indicated that more biocontrol potential fungi could colonize in the sterilized replanting soil. The functions of bacterial and fungal communities are predicted by PICRUSt2. Notably, the abundances of two enriched enzymes in bacterial communities exhibit significant differences: N-acetylmuramoyl-L-alanine amidase (EC 3.5.1.28) and cytochrome-c oxidase (EC 1.9.3.1). These enzymes are found in lower abundances in sterilized soil compared to diseased soil (Figure S6a). Regarding fungal communities, choline dehydrogenase (EC 1.1.99.1) and 3-oxoacyl-[acyl-carrier-protein] reductase (EC 1.1.1.100) display higher abundances in sterilized soil than in diseased soil (Figure S6b). Abiotic and biotic factors can significantly alter the network structure of the soil and plant host microbiome [51–53]. Networks of the top 100 genera in both fungal and bacterial communities were employed for analysis. Notably, bacterial communities exhibited a higher number of edges and average degree compared to fungal communities, irrespective of whether bulk soil was sterilized or not (Figure S7). However, the average degrees of fungal communities in sterilized replanting soil were significantly higher than those in diseased soil (Figure S7 and S8). Previous results showed that the fungal networks were less stable and their communities were more sensitive to Fusarium wilt disease than bacterial communities in pepper [54]. Our results also indicated that the fungal networks were easily changed by soil condition. Comparable findings were observed in the network density between fungal and bacterial communities. The study further examined the degrees of potentially beneficial bacteria and fungi. Among them, Streptomyces, Sphingomonas, and Bacillus had the highest degrees in NQS, PAD, and NQS, respectively (Figure S9a). In terms of fungal genera, there were significant differences in degrees between sterilized soil and diseased soil, indicating an increased complexity of the fungal community in sterilized soil. Furthermore, the degrees of Trichoderma, Aspergillus, and Penicillium reached their respective peaks in CKS, NQS, and PAS (Figure S9b). Isolation of anAspergillus sp.NCF-16 from sterilized replanting soil showing antagonistic activity againstP. carotovorum The LDA Effect Size (LEfSe) stands as an invaluable instrument for evaluating biomarkers in bacterial and fungal communities, especially in relation to diseases and sterilized soil that has undergone various treatments. However, it has been noted that a considerable number of genera are present across all bacterial community comparisons, which complicates the process of identifying treatment-dependent biomarkers (Figure S10). In contrast, fungal communities demonstrate a more distinct and significant correlation between biomarker genera and treatments. Importantly, Aspergillus has emerged as a crucial biomarker within the CKS group (Fig. 8a). We isolated 16 fungi from CKS soil and classified them into seven genera, including Fusarium, Aspergillus, and Trichoderma (Fig. 8b). These isolates were tested for their inhibitory effects on the growth of P. carotovorum, and three Aspergillus isolates, NCF-12, NCF-13, and NCF-16, exhibited antagonistic effects. Among them, NCF-16 demonstrated the strongest inhibitory ability in vitro. When P. carotovorum was introduced into suspensions of NCF-16, its growth was also suppressed (Fig. 8c and d). Furthermore, when P. carotovorum mixed with NCF-16 suspensions was inoculated on the konjac surface, the area affected by rot disease was significantly reduced (p < 0.05) (Fig. 8e and f). Aspergillus terreus ANU-301 was reported to inhibit the growth of potato soft rot pathogen, Dickeya chrysanthemi by secreting an antibacterial compound [50]. NCF-16 inhibit the disease occurrence which might release some antibacterial compounds. However, future study will focus on the isolation, identification, and antibacterial activity of these metabolites to better understand the antibacterial mechanisms of NCF-16 and potential applications in sustainable plant disease management. Thus, our findings align closely with the LEfSe analysis, suggesting that soil sterilization establishes a selective environment by reducing microbial complexity, thereby facilitating the re-establishment of beneficial microorganisms that contribute to disease suppression. Isolation and antagonistic activity of Aspergillus sp. NCF-16 from sterilized replanting soil. a The LEfSe analysis identifying fungal biomarkers across treatments. b Taxonomic identification of fungal isolates from sterilized soil. c and d Growth inhibition of Pcc by NCF-16 suspension. e and f The effects of NCF-16 suspension on konjac soft disease occurrence. Error bars represent the means ± SD. The different letters above each bar graph indicate significant differences (p < 0.05) calculated by ANOVA and Duncan’s test Conclusion Soil sterilization is a vital strategy for controlling soil-borne diseases such as konjac soft rot, yet its broader ecological impacts have remained unclear. This study demonstrates that sterilization not only reduces pathogens but also fundamentally restructures the soil ecosystem by accumulating antibacterial metabolites like 1,4-naphthoquinone, which inhibits P. carotovorum through disruption of ribosomal function and T3SS-mediated virulence; enriching beneficial fungi such as Aspergillus, Penicillium, and Trichoderma that establish complex disease-suppressive networks; and promoting the dominance of antagonistic strains like Aspergillus sp. NCF-16, which directly suppresses pathogen growth. These synergistic changes foster a suppressive soil environment that supports konjac health. By decoupling the biochemical and microbial mechanisms through a multi-omics approach, this work offers a foundation for designing targeted soil amendments. Supplementary Material Supplementary Material 1. Supplementary Material 2. Figure S1: The KEGG pathways and compound classifications of identified compounds based on the annotation database. Supplementary Material 3. Figure S2: The KEGG enrichment analysis of DAMs in different samples compared to sterilized soil sample. Supplementary Material 4. Figure S3: The inhibition test of six compounds on the growth of Pcc and Pa. Supplementary Material 5. Figure S4: The volcano plot of DEGs in Pcc with 1,4-naphthoquinone treatment. Supplementary Material 6. Figure S5: The expressions of pathogenicity genes involved in T2SS (a), T3SS (b), T6SS (c), pectate lyase (d), elongation (e) and flagellar (f). Supplementary Material 7. Figure S6: The function analyses of bacterial and fungal communities predicted by PICRUSt2. Supplementary Material 8. Figure S7: The co-occurrence network of fungal communities in different soil samples. Supplementary Material 9. Figure S8: The co-occurrence network of bacterial communities in different soil samples. Supplementary Material 10. Figure S9: The degree of beneficial bacteria (a) and fungi (b) in different soil samples. Supplementary Material 11. Figure S10: LEfSe analysis of fungi in different soil samples. Authors’ contributions Y.W., C.L; conceptualization, Supervision. C.N, X.Y, L.H, X.Y, R.W, X.W; methodology. Y.W, H.Q, S.M.J; Formal analysis and visualization. Y.W, C.N; writing the original manuscript. C.L., Y.W, S.M.J; revision and editing. C.L, Y.W. funding acquisition. All authors have read and agreed to the publication of the final manuscript. Funding This work was supported by the National Key R&D Program of China (2023YFD1400800), the National Natural Science Foundation of China (32202254), the Yunnan Provincial Talent Program for Wang Yi (XDYC-QNRC-2023-0421). Data availability 16 S, ITS and transcriptome sequence data are publicly available at the NCBI public database Sequence Read Archive (SRA) under the BioProject number [PRJNA1170501](https:/www.ncbi.nlm.nih.gov/bioproject/PRJNA1170501) . Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Chengfen Nong and Sauban Musa Jibril contributed equally to this work. Contributor Information Chengyun Li, Email: 2004165@ynau.edu.cn. Yi Wang, Email: wyi_0114@163.com. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Figure S1: The KEGG pathways and compound classifications of identified compounds based on the annotation database. Supplementary Material 3. Figure S2: The KEGG enrichment analysis of DAMs in different samples compared to sterilized soil sample. Supplementary Material 4. Figure S3: The inhibition test of six compounds on the growth of Pcc and Pa. Supplementary Material 5. Figure S4: The volcano plot of DEGs in Pcc with 1,4-naphthoquinone treatment. Supplementary Material 6. Figure S5: The expressions of pathogenicity genes involved in T2SS (a), T3SS (b), T6SS (c), pectate lyase (d), elongation (e) and flagellar (f). Supplementary Material 7. Figure S6: The function analyses of bacterial and fungal communities predicted by PICRUSt2. Supplementary Material 8. Figure S7: The co-occurrence network of fungal communities in different soil samples. Supplementary Material 9. Figure S8: The co-occurrence network of bacterial communities in different soil samples. Supplementary Material 10. Figure S9: The degree of beneficial bacteria (a) and fungi (b) in different soil samples. Supplementary Material 11. Figure S10: LEfSe analysis of fungi in different soil samples. Data Availability Statement 16 S, ITS and transcriptome sequence data are publicly available at the NCBI public database Sequence Read Archive (SRA) under the BioProject number [PRJNA1170501](https:/www.ncbi.nlm.nih.gov/bioproject/PRJNA1170501) .