Microbial Differences in Rhizospheric and Ednophytic Microbiota in Healthy Chinese Yam Roots and Those Affected by Yam End Black Disease

TL;DR

This study explores how harmful microbes in Chinese yam roots contribute to a damaging disease, offering insights for eco-friendly solutions.

Contribution

The study reveals internal microbial imbalances and nematode increases as key drivers of yam end black disease.

Findings

Diseased yam roots show increased Meloidogyne spp. and disrupted bacterial communities.

Endophytic microbiota imbalances are more severe than rhizosphere changes in diseased roots.

Transcriptomic analysis links polyamine metabolism and hormone signaling to disease response.

Abstract

Chinese yam is threatened by yam end black disease, a soil-borne disease that severely reduces yields. In this study, we investigated its causes by analyzing changes in the plant’s roots and their associated microbial communities. We found that diseased roots contain a large increase in harmful Meloidogyne spp. and show major imbalances in their internal bacterial communities. These internal disruptions appear to be key drivers of the disease. Our findings provide important insights for developing eco-friendly strategies to control yam end black disease, helping protect crop yields and support sustainable agriculture.

Yam end black disease (YEBD) is a devastating soil-borne disease that severely compromises the yield of Chinese yam (Dioscorea opposita Thunb.). Despite its agricultural importance, the etiological agents and molecular mechanisms underlying YEBD remain poorly understood. In this study, we employed an integrated multi-omics approach, combining transcriptomics and microbiome analysis, to dissect the host responses and microbial shifts associated with YEBD. De novo transcriptome assembly revealed significant enrichment of differentially expressed genes involved in polyamine metabolism and hormone signaling pathways. Microbiome profiling identified a substantial increase in nematodes (Meloidogyne spp.) in diseased samples, which correlated negatively with the beneficial fungus Cladosporium. Bacterial community analysis showed an increase in Proteobacteria and Bacteroidetes and a decrease in Actinobacteria and Firmicutes in YEBD-affected roots. Notably, the rhizosphere microbiome was less affected than the endophytic community, suggesting that internal microbial dysbiosis plays a critical role in disease progression. These findings provide new insights into the interactions among yam, nematodes, and microbes, offering potential strategies for biocontrol and disease management.

Figures

Full text

1. Introduction

Chinese yam (Dioscorea opposita Thunb.) is an economically important crop valued for both its nutritional and medicinal properties [1,2]. Extracts of Chinese yam have demonstrated significant medicinal potential [3,4,5]. However, its production faces increasing threats from various devastating diseases [6]. Among them, yam anthracnose, primarily caused by Colletotrichum gloeosporioides, is one of the most destructive foliar diseases globally, capable of causing yield losses of up to 90% [7,8]. It is characterized by black or brown lesions on leaves, stems, and vines, leading to defoliation and plant dieback under favorable warm and humid conditions [6]. Another major constraint is infection by root-knot nematodes (RKNs, Meloidogyne spp.), which cause crazy root syndrome. RKNs induce characteristic galling (knots) on roots and tubers, leading to stunted growth, tuber deformation, and significant yield losses ranging from 24% to 80% in China [9]. While anthracnose predominantly affects aerial parts and RKN causes distinct root galls, a newly emerging and distinct threat is yam end black disease (YEBD). YEBD specifically targets the tubers, causing corking, necrosis, and tumor-like outgrowths on the epidermis, while the above-ground parts remain asymptomatic. This symptom profile clearly differs from the foliar spots of anthracnose and the root galls induced by RKNs. Disease incidence of YEBD can reach up to 70% in continuous cropping systems, yet its etiology remains unclear. The specific causal agent of YEBD has not been determined. The observed tuber symptoms, though involving outgrowths, lack the typical root-knot structures caused by Meloidogyne species, suggesting a potential involvement of other pathogens such as fungi or bacteria or a distinct pathogenic mechanism.

Rhizosphere and endophytic microbiomes are critical for plant health, influencing nutrient uptake, stress responses, and pathogen resistance [10,11]. Rhizospheric microorganisms form a complex assembly encompassing both harmful pathogens and beneficial microbes that directly support plant growth and development through mechanisms such as phosphate solubilization and phytohormone production [12,13]. The composition of the rhizosphere microbiome acts as a primary filter, significantly shaping the structure and assembly of endophytic communities [14]. Endophytes, which colonize plant tissues through natural openings or wounds, coexist with their hosts and contribute to balanced plant micro-ecosystem stability, often enhancing host resilience to biotic and abiotic stresses [15,16,17,18,19]. Given the strong association between YEBD and long-term monoculture—a condition known to deplete beneficial microbes and favor pathogen buildup [20,21]—it is likely that dysbiosis in the root microbiome, driven particularly by interactions between pathogenic nematodes and fungi, is a key driver of disease development [22].

Previous studies have shown that continuous cropping alters both rhizospheric and endophytic microbial communities [23,24]. Since the occurrence of YEBD is closely linked to continuous cropping, it can be hypothesized that the structure and abundance of plant endophytic and rhizospheric microorganisms in YEBD-affected Chinese yam differ from those in healthy plants. Certain microorganisms may act as causal agents of this “scorch” disease. In this study, we employed high-throughput sequencing of transcriptomes and microbial communities to compare healthy and YEBD-affected yam roots from two geographic locations. Our aim is to identify key microbial taxa and host genetic mechanisms associated with YEBD, thereby providing a foundation for developing microbiome-based disease management strategies.

2. Materials and Methods

2.1. Sample Collection

The Chinese yam cultivated varieties Xiao Bai Zui and Bang Yao were used in this study. On 25 October 2018, three parallel healthy samples and three parallel YEBD samples of Bang Yao were collected from a field in Qingyuan (115.49267 E 38.76709 N), Hebei, China. Similarly, three parallel healthy samples and three parallel YEBD samples of Xiao Bai Zui were collected from a field in Anguo (115.32321 E 38.41391 N), Hebei, China.

After 150 days of growth, the roots were carefully excavated from the soil. Loosely attached soil was removed by means of vigorous shaking, and the soil adhering tightly to the root surface was defined as rhizosphere soil. Rhizosphere soil was placed into 50 mL of sterile 0.9% NaCl solution and subjected to moderate agitation for 5 min, followed by centrifugation at 8000× g for 10 min. For endosphere sampling, the root system was submerged in tubes containing 10 mL of potassium phosphate buffer (0.1 M, pH 8) and shaken vigorously twice. The washed roots were transferred into 20 mL of fresh buffer solution. Ultrasonic cleaning was performed three times for 10 min each (160 W, 30 s on/30 s off). The resulting washing solutions were combined and passed through a 0.22 μm filter membrane. The membranes and root tissues were frozen with liquid nitrogen and stored at −80 °C until DNA and RNA extraction.

2.2. DNA Extraction, PCR Amplification and Illumina MiSeq Sequencing

Microbial community genomic DNA was extracted from roots using the FastDNATM SPIN Kit for Soil (MP Biomedicals, Solon, OH, USA) according to the manufacturer’s instructions. For each sample, at least 10 roots were pooled for DNA extraction. DNA quality was assessed via 1.0% agarose gel electrophoresis and a NanoDrop 2000 spectrophotometer (Thermo Scientific, Waltham, MA, USA). The hypervariable V3-V4 region of the bacterial 16S rRNA gene was amplified using primers 799F (5′-AACMGGATTAGATACCCKG-3′) and 1193R (5′-ACGTCATCCCCACCTTCC-3′). The ITS2 region of the fungal community was amplified using primers ITS3F (5′-GCATCGATGAAGAACGCAGC-3′) and ITS4R (5′-TCCTCCGCTTATTGATATGC-3′). PCR amplification was performed using an ABI Gene Amp^®^ 9700 instrument. PCR conditions for 16S rRNA amplification were as follows: initial denaturation at 95 °C for 3 min, followed by 30 cycles of denaturation at 95 °C for 30 s, annealing at 55 °C for 30 s, and extension at 72 °C for 45 s, lastly followed by a final extension at 72 °C for 10 min and storage at 4 °C. The ITS2 region was amplified using 35 cycles under similar conditions. Purified amplicons were pooled in equimolar amounts and subjected to paired-end sequencing (Miseq PE300) on an Illumina MiSeq platform by Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China) using standard protocols.

2.3. RNA Extraction and Illumina Sequencing

Three biological replicates were used for the RNA-Seq analysis. Total RNA was extracted from the root samples using an RNeasy^®^ Plant Mini Kit (Qiagen GbmH, Hilden, Germany). After digestion with DNase I, approximately 10 μg of total RNA was used to construct cDNA libraries following standard Illumina procedures. RNA sequencing (150PE, paired-end) was performed on an Illumina NovoSeq 6000 platform by Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China) using standard protocols.

2.4. Processing of Sequencing Data

Raw 16S rRNA gene sequencing reads were first filtered using FASTP version 0.20.0. Filtered reads were then merged using FLASH version 1.2.7 following these criteria: (i) reads were truncated from the start of the 50 bp sliding window if the average quality score fell below 20, and truncated reads shorter than 50 bp were discarded; (ii) paired reads with a minimum overlap of 10 bp were merged into a single sequence, allowing a maximum mismatch ratio of 0.2 in the overlapping region; and (iii) samples were distinguished using exact barcode matching and primer matching, allowing up to two nucleotide mismatches.

UPARSE (v 7.1) was used to cluster the OTUs with a 97% similarity cutoff, and the chimeric sequences were removed. Taxonomic assignment of each OTU representative sequence was performed against the Silva v132 database using RDP classifier v 2.2 with a confidence threshold > 0.7. The taxonomic nomenclature follows the SILVA reference database (release v132).

For reads annotated as “unclassified_k_Fungi”, representative OTU sequences were selected for further taxonomic classification. These sequences were subjected to BLASTn (v 2.10.0) analysis against the NCBI non-redundant nucleotide (nt) database. Sequences yielding alignments with >90% identity were assigned to the corresponding species.

For RNA-Seq data, adapters were removed, and low-quality reads were filtered using Fastp (v 0.19.4). Clean reads were assembled de novo using Trinity with default parameters. Initial unigenes from each sample were merged and further assembled using TGICL v 2.1 with a minimum overlap length of 100 bp. The resulting unigenes were used for subsequent analyses. Bowtie2 was used to map clean reads to the unigenes, and mapping statistics were calculated accordingly.

2.5. Statistical Analysis

Mothur v 1.30.2 was used for the alpha-diversity analyses, including community richness parameters and sequencing depth indices. Principal coordinate analysis (PCoA) based on OUT level data was performed, and the R package vegan was used for difference significance analysis. LEfSe software (v 1.0) was used to identify significantly different genera between groups. The OTUs with an abundance greater than 50 sequences were used to construct co-occurrence networks, which were analyzed with NetworkX software (v 2.7.1) using default parameters. All values presented in this study represent the average of duplicate or triplicate measurements. Differences between groups were considered statistically significant at p < 0.05.

3. Results

3.1. De Novo Transcriptome Assembly and Functional Annotation

YEBD severely affects the quality and yield of Chinese yam (Figure S1), yet its pathogenic mechanisms remain unclear. Transcriptome analysis provides insights into host biological processes and cellular activities in diseased tissues. RNA from healthy and YEBD-affected yams collected in Qingyuan was sequenced and assembled de novo, generating 94,593 unigenes with an N50 of 1135 bp (Table 1). Approximately 50% of the unigenes (46,300) were successfully annotated using four databases (NR, Swiss-port, KEGG, GO) (Figure 1A). Among them, 23,876 unigenes were assigned to five classes containing 122 KEGG pathways (Figure 1B), including pathways associated with plant growth, development, and stress responses. After filtering out low-expression unigenes (average TPM > 1 in at least one sample), 54,719 unigenes were retained for further analysis. GO annotation indicated that expressed genes were primarily associated with “cellular process”, “metabolic process”, “membrane”, “catalytic activity”, and “binding” (Figure 1C). Additionally, 1442 unigenes (6.8%) were linked to “response to stimulus”, including “response to biotic stimulus” (129), “response to external stimulus” (175), and “response to endogenous stimulus” (231). This high-quality transcriptome dataset forms a foundation for investigating the relationship between gene expression patterns and microbial community changes in YEBD.

3.2. Hormonal and Defense-Related Pathways Are Activated in YEBD

To identify host genes affected by YEBD, differential expression analysis was performed. Using a threshold of |log_2_ (fold change)| >1.0 and an adjusted p-value < 0.05, (DESeq2) 2388 unigenes were identified as differentially expressed genes (DEGs), of which 62.14% were up-regulated, and 37.86% were down-regulated. GO and KEGG enrichment revealed significant enrichment of signal-transduction-related categories, such as “cell communication”, “signal transduction”, and “response to endogenous stimulus” (Figure 2). Multiple genes related to plant hormone signaling pathways were differentially expressed. For instance, the ARF gene (TRINITY_DN60141_c0_g1), a downstream transcription factor in the auxin signaling pathway, was down-regulated. In the cytokinin signaling pathway, two type-A ARR genes were up-regulated, while one type-B ARR gene was down-regulated. BZR1, a key transcription factor in the brassinosteroid signaling pathway, was up-regulated. The “plant-pathogen interaction” pathway was also significantly enriched, including genes encoding disease resistance proteins and WRKY transcription factors (Table S1). These results indicate that hormonal regulation and defense signaling pathways are actively involved in the plant response to YEBD.

3.3. YEBD-Affected Endophytic Samples Characterized by Diverse Fungal Communities and Root-Knot Nematodes

The fungal ITS2 rRNA gene was sequenced from healthy and YEBD-affected endophytic samples from Qingyuan (QHEF, QDEF) and Anguo (AHEF, ADEF). After quality filtering, 363 fungal OTUs were retained for analysis. Thirty-two abundant OTUs (>1000 reads) represented 88.81% of total reads, including members of unclassified_k_Fungi, Ascomycota, Basidiomycota, and Rozellomycota (Figure 3A). Notably, compared with healthy samples (QHEF and AHEF), both YEBD-affected samples (QDEF and ADEF) were dominated by unclassified_k_Fungi (Figure 3A), and the reads that mapped to the unclassified_k_Fungi accounted for 49.90% of all sequences. To clarify the taxonomic identity of the sequences annotated as unclassified_k_Fungi, we conducted further alignment against the NCBI nt database. The results indicated that most of these sequences exhibited high similarity to species within the nematode genus Meloidogyne spp., suggesting a potential misannotation in the initial taxonomic classification. Following this finding, we performed a more detailed analysis of nematode composition. This analysis revealed distinct enrichment patterns of nematode species in YEBD samples from different regions: M. incognita was predominantly found in the QDEF sample, while M. luci was the dominant species in the ADEF sample (Figure 3B).

Principal coordinate analysis (PCoA) based on OTUs showed significant differences in community composition between YEBD and healthy samples (p = 0.001, Adonis), but no significant differences were detected among healthy samples or among YEBD samples from different locations (Figure 4A). Among 128 genera, eight differed significantly between groups (Kruskal–Wallis H test). Dominant genera (relative abundance > 5% in at least one sample) included unclassified_k_Fungi, unclassified_p_Ascomycota, Acrocalymma, Saitozyma, and Cladosporium (Figure 4B). Unclassified_k_Fungi was significantly higher in YEBD samples (92.96% in QDEF and 81.49% in ADEF) (Figure 4C). Correlation analysis showed that unclassified_k_Fungi was negatively correlated with Cladosporium (Figure 4D).

3.4. Bacterial Community Shifts in YEBD

To further examine the influence of YEBD on the composition of the root microbiota, the bacterial community was analyzed. A total of 659 bacterial OTUs were obtained from 302,640 reads. Among them, 151 OTUs were shared across all samples (Qingyuan/Anguo healthy/YEBD sample), and the number of shared OTUs was higher in YEBD samples (311) than in healthy samples (205) (Figure 5A). The dominant phyla (relative abundance > 1%) were Proteobacteria, Actinobacteria, Bacteroidetes, and Firmicutes (Figure 5B). YEBD samples from both regions (QDEB and ADEB) showed similar trends; the proportions of Proteobacteria and Bacteroidetes increased, while those of Actinobacteria and Firmicutes decreased. In particular, Actinobacteria declined substantially—from 40% in the Anguo healthy endophytic bacterial sample (AHEB) to 20% in the corresponding diseased sample (ADEB) (Figure 5B). These shifts indicate that YEBD altered the bacterial community structure by increasing or decreasing key dominant taxa. Furthermore, the PCoA analysis based on the detected OTUs showed that the YEBD samples clustered together, whereas healthy samples showed greater dispersion. In addition, there was a significant difference in the community composition (R^2^ = 0.58, p = 0.001 in PERMANOVA) between the YEBD (QDEB and ADEB) and healthy (QHEB and AHEB) samples, whereas no significant difference was observed between QDEB and ADEB (R^2^ = 0.34, p = 0.1 in PERMANOVA) (Figure 5C). These findings suggest that YEBD infection strongly restructures the bacterial community, likely due to microbial recruitment or selective enrichment within diseased roots, resulting in distinct microbial profiles compared with healthy roots.

Next, differences in microbial profiles between YEBD and healthy root samples were examined independently to determine whether pathogenesis-related shifts occurred in the microbial communities of the two root types. Linear discriminant analysis (LDA) effect size (LEfSe) was used to identify the key genera responsible for differences between groups (Figure 6). Based on the LDA score (LDA > 3 and p < 0.05), LEfSe analysis revealed that 35 and 37 bacterial genera differed between QDEB and QHEB (Figure 6A) and between ADEB and AHEB (Figure 6B), respectively. Only seven of these genera were different in both the QDEB vs. QHEB and ADEB vs. AHEB comparisons. Among them, Polaromonas, Pseudomonas, Starkeya, Alphaproteobacteria, SM2D12 and Ilumatobacteraceae were enriched in both YEBD samples, whereas Pseudorhodoplanes displayed contrasting enrichment patterns -it was enriched in QDEB but in AHEB in the Anguo samples. In addition, although the effects of YEBD on bacterial species was between regions (Qingyuan and Anguo), most taxa exhibiting significant changes belonged to the phylum Proteobacteria. These findings suggest that YEBD drives a substantial restructuring of the rhizosphere microbiota and that alterations within the Proteobacteria likely contribute to the YEBD phenotype.

3.5. Rhizosphere Microbiome Is Less Affected Than the Endosphere

We further analyzed the microbial community composition in the rhizosphere of healthy and YEBD-affected plants. For this purpose, three rhizosphere samples were collected from both healthy (QHRF) and YEBD (QDRF) plants from Qingyuan, and data were processed using the same analytical pipelines described above. A total of 331 fungal OTUs were obtained from 185,760 reads. Compared with the endophytic fungi, the fungal composition of QDRF was almost identical to that of QDEF, with Ascomycota, unclassified_k_Fungi and Basidiomycota as the dominant phyla; however, Zygomycota was an additional dominant phylum in the QDRF (Figure S2A). To clarify the origin of sequences initially classified as unclassified_k_Fungi in QDRF samples, we re-aligned them against the NCBI nt database. This re-classification revealed that most were actually derived from nematodes of the genus Meloidogyne, with M. incognita alone accounting for 94.74% of the re-assigned reads. These results were the same as those of the QDEF, which had more Meloidogyne, especially M. incognita. In addition, QDRF and QDEF differed significantly in fungal community composition (PERMANOVA) (R^2^ = 0.68, p = 0.001), whereas QDRF and Qingyuan healthy rhizosphere fungi (QHRF) could not be significantly separated (R^2^ = 0.31, p = 0.1 in PERMANOVA) (Figure 7A). These results indicate that the rhizosphere fungi had a limited effect on the YEBD. It is worth noting that the genus of Cladosporium was significantly reduced in the QDRF (p = 0.038, Wilcoxon rank-sum single-tailed test) (Figure 7B). This was consistent with the results from the QDEF, suggesting that Cladosporium was negatively correlated with the YEBD in both the rhizosphere and endophytic fungal communities.

For the bacterial communities, 950 OUTs were detected from 82,236 reads. Compared with QDEB, Gemmatimonadetes and Acidobacteria emerged as additional predominant phyla in the rhizosphere bacteria (QDRB) (Figure S2B). The detected OTUs showed a significant difference in community composition between QDEB and QDRB (R^2^ = 0.87, p = 0.001 in PERMANOVA), while no significant difference was observed between QDRB and QHRB (R^2^ = 0.54, p = 0.1 in PERMANOVA) (Figure 7C). Furthermore, the genera Acidobacteria, Nitrospirae, Chloroflexi and Planctomycetes were significantly enriched in QDRB, whereas Spirochaetae were significantly enriched in QHRB (p = 0.040, Wilcoxon rank-sum single-tailed test) (Figure 7D). These findings suggest that the effect of YEBD on the rhizosphere microbiome is limited compared to the root endophytic community.

4. Discussion

YEBD is a disease that significantly affects the yam quality, yet its exact cause remains unclear. In this study, next-generation sequencing technology, including RNA-Seq and MiSeq, were used to comprehensively compare the dynamic changes in mRNA expression and microbial communities between YEBD-affected and healthy roots. Additionally, YEBD samples from two different regions were analyzed to identify factors that may induce the disease. It should be noted that the geographic and cultivar scope of this study, while informative, necessitates validation across broader agroecological contexts to confirm the generalizability of the identified signatures. Our results indicate that nematodes may be a primary trigger of the disease. We also found statistically significant differences in microbiota composition and relative abundance between healthy and unhealthy samples. However, it is important to recognize that the multi-omics data generated here are primarily correlative. While they reveal strong associations, future pathogenicity assays are required to establish causal relationships between the observed microbial shifts and disease development.

Using the fungal-specific ITS3-ITS4 primer pair, we observed significant amplification of nematode-derived sequences in diseased yam samples alongside target fungal communities, and BLASTn confirmation identified these non-target amplicons as Meloidogyne spp. This aligns with prior reports that fungal ITS primers can co-amplify phylogenetically distant organisms when (i) sequence homology exists in conserved primer-binding regions and (ii) non-target DNA dominates the template pool, as evidenced by documented cross-amplification of plant ITS sequences [25,26]. However, the precise reason why these fungal primers also amplify nematode DNA remains unclear and warrants further investigation.

Notably, the symptoms of YEBD differed from those of typical root-knot nematode disease, which is classically characterized by the formation of root galls (knots) and tuber malformations caused by Meloidogyne species [6]. We therefore hypothesized that nematodes may influence the composition of microbial communities, leading to the occurrence of YEBD. This is consistent with earlier findings showing that nematode-induced wounds, whether caused by micropuncture or by rupture, increase the likelihood of infection by other soil-borne pathogens by creating more infection sites [27,28]. However, the correlative nature of our multi-omics data cannot establish causation; the specific mechanistic roles of nematodes in initiating dysbiosis or disease require future validation through controlled inoculation experiments. Previous studies have also demonstrated interactions between nematodes and other microbes. A previous study demonstrated that infestation by M. incognita significantly altered the endophytic microbial composition of yam roots, enriching opportunistic fungal pathogens such as Fusarium and contributing to a synergistic disease complex [29]. Additionally, similar interactions have been documented in other crops, such as those between M. incognita and F. oxysporum in bananas [30] and between M. incognita and Thielaviopsis basicola, the causal agent of black rot in cotton [31]. These studies highlight the acute impact of nematodes on plant growth and development. Although nematodes can increase seedling mortality, this effect depends on factors such as nematode density, sampling date, and soil texture [32]. In this study, we found that the nematode density in YEBD samples was significantly higher than in healthy samples, suggesting that nematode abundance may play an important role in the development of YEBD. However, the reason for the unusually high nematode numbers in YEBD roots requires further investigation.

In nematode–fungus nematode–fungus–plant interactions, the role of nematodes is complex, and even non-susceptible plants may become susceptible in their presence [27,33]. Thus, characterizing the microbiota associated with nematode infection can help elucidate the potential role of microbes in disease. In our study, 363 fungal OTUs were detected in the root endophyte, with most species present at low relative abundance. Only eight fungal genera were identified as differentially abundant. We propose two possible explanations: (i) nematodes may have a limited effect on endophytic fungi in yam or (ii) the high nematode presence interfered with the detection of fungal species. This highlights a technical consideration: the use of ITS primers, while revealing an important nematode signal, may also introduce bias in characterizing the true fungal community due to primer cross-amplification and competitive PCR effects. Nevertheless, we unexpectedly found a negative correlation between Cladosporium and nematodes in both the endosphere and rhizosphere, whereas Elhady et al. reported no apparent relationship between Cladosporium and nematodes in soil [34]. Interestingly, Cladosporium species have been used successfully in the biocontrol of nematode infections in sheep [35]. The differences in these results suggest that different living environmental conditions strongly shape the relationship between Cladosporium and the nematodes. Previous studies have shown that many fungi contribute to soil suppression of cyst nematodes [36,37]. Thus, Cladosporium may be a promising candidate for nematode suppression in yam roots. Screening for Cladosporium strains with antagonistic activity against Meloidogyne spp. may provide an effective approach for nematode biocontrol in yam.

Previous studies have shown that nematodes significantly influence the bacterial communities inhabiting the root endosphere [38,39]. In our study, a total of 659 bacterial OTUs were detected using the 799–1193 primer, and the bacterial phyla Proteobacteria, Actinobacteria, Firmicutes, and Bacteroidetes were highly abundant in the samples. This result is consistent with previous findings in which these phyla were dominant in the root endosphere [38,40,41]. Although most bacterial taxa were shared between YEBD and healthy tissue at the phylum level, there was a significant difference between the microbes in YEBD and healthy tissue. In addition, we found that the relative abundances of these bacterial taxa were primarily determined by the sample. Proteobacteria and Bacteroidetes were more abundant in YEBD samples, whereas Actinobacteria and Firmicutes were less abundant. This suggests that Proteobacteria and Bacteroidetes may be enriched in the YEBD through recruitment from the surrounding soil. In soybean cyst nematodes, when Heterodera glycines was inoculated into suppressive soil, Bacteroidetes became more enriched in the cyst endosphere, whereas Proteobacteria decreased, and Actinobacteria and Firmicutes increased relative to the root endosphere [38]. These contrasting patterns highlight the complexity of nematode effects on the bacterial community. Furthermore, 21 and 14 taxa of Proteobacteria were identified at the family level as significantly different microbial taxa between QDEB and QHEB and ADEB and AHEB, respectively, and these taxa may serve as potential biomarkers. Proteobacteria, an important phylum of Gram-negative bacteria, are among the most abundant groups in the rhizosphere and play critical roles in root development [42]. Many members of this phylum have been identified as pathogens of humans, animals, and plants [43,44,45]. However, the functional role of Proteobacteria in YEBD-affected microbiota remains unclear and requires further investigation. In addition, Nitrospirae and FBP were enriched in the YEBD samples, but their functions in the root endosphere have not been elucidated. We speculate that these taxa may be associated with nematode infection. In addition, nematode-infested roots often act as metabolic sinks, with nutrients transported through the symplast being diverted to the nematodes, resulting in increased rhizodeposition compared to healthy tissue. This in turn can affect the activity of both plant-pathogenic and beneficial microorganisms in the rhizosphere [33,46]. Thus, alterations in rhizosphere bacteria may also contribute to the symptoms of YEBD. From a practical perspective, the distinct microbial signatures associated with YEBD, including the specific bacterial taxa identified here, offer potential as biomarkers for developing early molecular diagnostic tools.

In response to pathogen infection or changes in the microbial community within the root endosphere, plants have evolved various self-protection mechanisms, the most fundamental of which is the regulation of gene expression. In this study, RNA-Seq was used to de novo assemble the yam transcriptome and assess gene expression in healthy and YEBD-affected roots. This high-confidence transcriptome assembly provides a valuable resource for understanding interactions between yam and its pathogens. In total, 54,719 unigenes were identified as expressed genes, and 2388 genes were identified as DEGs. According to the GO and KEGG enrichment annotation of the DEGs, the “Plant-pathogen interaction” pathway was significantly enriched, including DEGs encoding plant disease resistance proteins. These DEGs may be induced by nematodes, other pathogens, or a combination of both. Recent studies indicate that polyamines may act as nematode attractants [47]. In our study, we found that two unigenes (TRINITY_DN56716_c0_g2 and TRINITY_DN56625_c2_g1) encoding polyamine oxidase, enzymes involved in polyamine catabolism, were down-regulated in the YEBD samples, suggesting that reduced polyamine catabolism may contribute to nematode accumulation. During plant defense responses, the timing and magnitude of resistance-related gene expression are critical for effective resistance [48]. Some endophytes can induce systemic resistance and thereby enhance plant defenses. For example, the endophytic bacterium Bacillus cereus BCM2 was shown to prime tomato plants to mount a stronger defense response against nematode infection [49]. Therefore, two strategies may be effective for controlling nematodes: (i) screening endogenous microbes for their biological control potential [49,50,51], and (ii) overexpressing resistance-related genes in the host to enhance systemic immunity [52].

In addition, we found that the signal transduction pathways play important roles in the defense response, particularly “Plant hormone signal transduction” and “MAPK signaling pathway—plant”. These pathways are well known for their roles not only in defense but also in plant growth and development. Brassinosteroids (BRs) are critical for plant growth and development and significantly influence root morphology [53]. A previous study showed that BZR1 can inhibit distal meristem differentiation and delay columella cell development [54]. In our study, the homologue gene of BZR1 was up-regulated in YEBD samples. Notably, the key genes of the cytokinin signal transduction pathway were identified as DEGs. Cytokinins play multiple roles in plant development, including regulating cambium activity, secondary growth, and responses to environmental stress [55,56,57]. In plants, ARR-A acts as a negative regulator of cytokinin responses [58], whereas ARR-B acts as a positive regulator of cytokinin signal transduction [59]. Cytokinins are particularly important for yam because the plants use their roots as storage organs. In the YEBD sample, two type-A ARR genes were up-regulated, while the type-B ARR was down-regulated, indicating reduced cytokinin signaling in the YEBD samples. These results indicated that the phytohormone signal transduction pathway associated with cell enlargement and cell division was negatively regulated, which was consistent with the phenotype of the YEBD sample, which had shorter stems than those of the healthy plants. Similarly, cytokinin also responded to nematode infestation in Solanum tuberosum, another species with storage roots [60], underscoring cytokinin’s role in tuber–pathogen interactions. Future research should combine targeted molecular biology techniques with the multi-omics framework used here to move beyond correlation and establish causal links between specific nematode activities, host gene regulatory networks, and microbiome assembly, ultimately informing precise disease management strategies.

5. Conclusions

In this study, we employed an integrated multi-omics approach that combines transcriptomics and microbiome analysis to investigate the host response and microbial dynamics associated with YEBD. The transcriptomic analysis revealed a significant enrichment of differentially expressed genes involved in polyamine metabolism and hormone signaling pathways, suggesting their crucial role in the plant’s defense against YEBD. Concurrently, microbiome profiling indicated a marked increase in the abundance of root-knot nematodes (Meloidogyne spp.) in the diseased samples, which negatively correlated with the levels of the beneficial fungus Cladosporium. Furthermore, bacterial community analysis highlighted shifts in the root microbiota, characterized by an increase in Proteobacteria and Bacteroidetes, along with a decrease in Actinobacteria and Firmicutes. Notably, the rhizosphere microbiome appeared less affected than the endophytic community, implying that internal microbial dysbiosis is a key driver of disease progression. Collectively, these findings enhance our understanding of the complex interactions among yam, nematodes, and associated microorganisms during the development of YEBD. This research lays the groundwork for designing targeted biocontrol strategies and integrated management practices. Future studies should focus on elucidating the functional mechanisms underlying the identified pathways and evaluating potential microbiome-based interventions to mitigate the impact of YEBD on yam production.