Next Article in Journal
Rapid Seed Viability Detection Using Laser Speckle Weighted Generalized Difference with Improved Residual Networks
Previous Article in Journal
Cellulolytic Microbial Inoculation Enhances Sheep Manure Composting by Improving Nutrient Retention and Reshaping Microbial Community Structure
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Agronomy
Volume 16
Issue 1
10.3390/agronomy16010080
agronomy-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

New Insights into the Formation Mechanism of Continuous Cropping Obstacles in Dioscorea opposita Thunb. cv. Tiegun Yam from Rhizosphere Metabolites and Microflora

by
Pengfei Zhang
1,2,
Wanghui Guan
2,
Lili Han
2,
Xiaoli Hu
2,
Ailing Xu
2,
Hui Wang
2,
Xiaomin Wang
2,* and
Xiaoyan Jiao
1,*
1
College of Resources and Environment, Shanxi Agricultural University, Taiyuan 030031, China
2
Cotton Research Institute, Shanxi Agricultural University, Yuncheng 044000, China
*
Authors to whom correspondence should be addressed.
Agronomy 2026, 16(1), 80; https://doi.org/10.3390/agronomy16010080
Submission received: 5 November 2025 / Revised: 22 December 2025 / Accepted: 24 December 2025 / Published: 26 December 2025
(This article belongs to the Section Farming Sustainability)
Browse Figures
Versions Notes

Abstract

Continuous cropping (CC) poses serious challenges to the sustainable production of Dioscorea opposita Thunb. cv. Tiegun yam. The aim of this study is to illustrate the formation mechanisms of CC obstacles by analyzing rhizosphere soil from yam fields with 0 to 2 years of replanting. Metabolomic and microbiome sequences were used to assess variations in yam yield, underground tuber traits, soil properties, metabolite profiles, and microbial communities. The results show that CC significantly reduced tuber yield, shortened stalk length, and altered tuber morphology, leading to the accumulation of soil available phosphorus and potassium and a notable decrease in pH. A total of 38 differentially expressed metabolites, including organoheterocyclic compounds, lipids, and benzenoids, were identified and linked to pathways such as starch and sucrose metabolism, linoleic acid metabolism, and ABC transporters. Microbial alpha diversity increased with CC duration, and both bacterial and fungal community structures were notably reshaped. Metabolite profiles correlated more strongly with fungal than bacterial communities. Partial least squares path modeling revealed that CC years had a negative indirect impact on tuber yield and morphology (the path coefficient was −0.956), primarily through direct effects on soil properties (p < 0.01) and metabolites (p < 0.001), which, in turn, influenced microbial diversity. These findings emphasize the vital role of soil properties in reshaping the rhizosphere environment under CC and provide a theoretical basis for mitigating CC obstacles through rhizosphere regulation.

1. Introduction

Dioscorea opposita Thunb. cv. Tiegun yam is the rhizome of the Dioscorea plant, and its tubers are nutritionally rich in starch, mucus protein, crude fiber, and minerals. Yam has been used for the treatment of inflammation, diarrhea, and diabetes because of its functional bioactive components, such as antioxidants, polysaccharides, diosgenin, polyphenols, allantoin, and flavonoids [1,2]. Due to its high nutritional value and strong antimicrobial properties, yam is in great demand in the consumer market. Thus, large-scale cultivation is carried out to meet market demands. However, because of the geoherbalism of the Tiegun yam and the limitation of cultivated land, continuous cropping (CC) of yam in production has become inevitable, which results in poor plant growth and decreased yam yield and quality [3,4]. Generally, fields used for Tiegun yam cultivation require an 8–10-year interval before replantation. Continuous cropping obstacles (CCOs) have severely restricted the sustainable development of the yam industry.
The formation of CCOs is often closely related to imbalances in the rhizosphere microecosystem, involving the deterioration of soil physicochemical properties [5], changes in enzyme activities, the disruption of the microbial community structure [6], and the accumulation of plant-derived metabolites [7,8,9]. In terms of soil nutrients, because of the preferential uptake of specific nutrients and a relatively consistent root depth, CC can easily lead to the depletion or unbalanced accumulation of specific nutrient elements (e.g., potassium, phosphorus, and trace elements) in the rhizosphere [10]. Meanwhile, soil acidification is also a notable feature under CC conditions [5,11], which reduces key enzyme activities and affects nutrient turnover [10]. Regarding the characteristics of the microbial community in yam, a previous study showed that with an increase in CC duration in Yongfeng yam, the relative abundance of beneficial bacteria (Proteobacteria, Actinobacteria, and Chloroflexi) decreased, while that of harmful bacteria (Gemmatimonadetes and Acidobacteria) increased [12]. As one of the most important pathogenic nematodes in yam, it was found that Meloidogyne incognita infection could enrich Chitinophageceae, Xanthobacteraceae, Kickxellomycota, Ramicandelaber, Fusarium, Cladosporium, and Alternaria in monocropping, which drastically altered the rhizosphere microbial community, and the average tuber length and fresh weight decreased by 60 cm and 161 g, respectively [3,4]. Medicinal plants continuously release a large number of secondary metabolites into the rhizosphere soil during their growth and development. In rotation or first cropping systems, these metabolites may participate in beneficial allelopathic interactions or signal communication. For example, some important bioactive components, such as trehalose 6-phosphate, D-sucrose, D-glucose 1,6-bisphosphate, and D-trehalose, are secreted by Shandong and Henan Tiegun yam. An et al. [13] found that several secondary metabolites, including four flavonoids, batatasin III and IV, and gingerenone A, were upregulated during the late harvest stages. However, as CC continues, certain accumulated metabolites among other medicinal plants have strong allelopathic autotoxicity to themselves and affect the plant’s health and physiological functions [6,7,8]. It is a pity that little is known about the impacts of Tiegun yam CC on rhizosphere metabolites.
Among the many medicinal plants affected by CCOs, the case of Tiegun yam (Dioscorea opposita Thunb. cv. Tiegun) is particularly severe and unique. The underground tubers of Tiegun yam grow vertically downward, reaching depths of over 1.5 m. This deep-rooted characteristic causes significant and lasting disturbance to the physical structure, chemical properties, and microbial flora of the deep soil layers, potentially creating a rhizosphere microenvironment that is vastly different to that of shallow-rooted crops. Therefore, it is essential to elucidate the formation mechanism of serious CCOs in Tiegun yam, which is of crucial importance for the development of relevant regulation technologies to subsequently relieve the CCOs. Although previous studies have begun to focus on the changes in the microbial communities of yam under CC or have identified metabolic components in yam under normal growth conditions, significant gaps remain in integrated research on the rhizosphere microecosystem of Tiegun yam under CC stress: it remains unclear what specific metabolites are secreted by Tiegun yam into the rhizosphere under CC conditions and which of these may act as key autotoxic substances or microbial community regulatory signals. How does CC link the causal chain of “changes in soil physicochemical properties and enzyme activities → root metabolic secretion → evolution of microbial community structure and function”? How is the interaction network among these elements—particularly how metabolites drive microbial responses—manifested in the deep-root system of Tiegun yam? Regarding this unique deep-rooted variety, does the microecological mechanism of its CCOs differ to that of other yam varieties? How does its distinct growth habit exacerbate or alter the aforementioned interaction processes? Targeted research is needed to answer these questions. Therefore, this study proposes a clear hypothesis: increasing the Tiegun yam CC years significantly alters the rhizosphere exudate metabolic profile; these changes, synergizing with the deterioration of soil physicochemical properties and enzyme activities, directionally select for microorganisms (e.g., promoting pathogens and inhibiting beneficial bacteria) through chemical signaling, thereby leading to a rhizosphere microecological imbalance. To test this hypothesis, non-targeted metabolome and high-throughput sequencing was performed on the rhizosphere soil of Tiegun yam obtained from three CC years (0, 1, and 2, namely, CC0, CC1, and CC2 treatments). The objectives were to (1) examine the influence of the CC years of yam on rhizosphere metabolites, physicochemical properties, and key enzyme activity; (2) evaluate the impact of the CC years of yam on the diversity and community composition of bacteria and fungi; and (3) construct an interaction network between metabolites and microorganisms through a correlation analysis, clarifying the potential response mechanisms of microbes to specific metabolites, thereby systematically uncovering the microecological causes of CCOs in Tiegun yam. This study can provide a theoretical basis for developing green mitigation strategies that target the regulation of the rhizosphere microenvironment.

2. Materials and Methods

2.1. Overview of Study Area and Experimental Design

The study area was located in the Nanhua Experimental Base of the Cotton Research Institute at Yuncheng, Shanxi province, China (N 35°4′3.44″, E 110°59′48.28″). This region has a warm temperate continental monsoon climate with four distinct seasons, with concentrated rainfall and heat in the same season. The annual mean temperature, mean precipitation, and sunshine duration are 13.4 °C, 525 mm, and 2040 h, respectively. The frost-free period is 212 days. The average altitude is about 350 m. The soil type is carbonate cinnamon soil, which formed on loess and loess-like parent materials. Prior to yam cultivation, the fields were predominantly used for wheat and maize rotation. In 2021, the basic physicochemical properties of the soil were as follows: organic carbon 11.0 g·kg−1, total nitrogen 0.067%, available nitrogen 37.1 mg·kg−1, available phosphorus 5.4 mg·kg−1, rapidly available potassium 130 mg·kg−1, and pH 8.45. Up to 2024, fields that were cultivated with Tiegun yam for 1, 2, and 3 years were chosen and labeled as CC0, CC1, and CC2. The different treatment plots were adjacent to each other, and they were managed in the same way before the experiment. Five replicates were set up for each field experimental plot with different cultivation years. Seed yams, which were approximately 30 cm long underground tubers growing from bulbils, were planted in early April and harvested in the first half of November in the same year. The sowing density of the yams was approximately 66,000 plants·ha−1, with a row spacing of 100 cm and a plant spacing of 15 cm. Apart from the different cultivation years, all agronomic measures were uniform. The annual fertilization amount was 1200 kg·ha−1 of NPK equilibrium fertilizer (N 17%, P2O5 17%, and K2O 17%) and 1200 kg·ha−1 of microbial fertilizer (effective viable bacteria count ≥ 200 million·g−1), with the addition of sheep manure at approximately 30,000 kg·ha−1.

2.2. Collection of Rhizosphere Soil Samples

A 5 m long row with approximately 33 yams was randomly selected from each replicate. Large chunks of soil loosely attached to the roots were wiped off via vigorous shaking, and the soil within 1–2 mm of the fibrous roots was carefully collected via brushing. The rhizosphere soil samples collected from the upper, middle, and bottom sections of these yams were mixed to form a single composite sample, and a total of 15 soil samples were obtained [4]. All samples were homogenized and passed through a 2 mm mesh to remove plant residues. Each soil sample was then divided into two portions: one was stored in a −80 °C freezer and transported to the testing company on dry ice for metabolomic and microbiological analyses, and the other was stored at 4 °C for the analysis of soil physicochemical properties and enzyme activities.

2.3. Determination of Soil Physicochemical Properties, Enzyme Activities, and Yam Yield

Soil pH was measured in a 1:2.5 (w/v) soil-to-water suspension using the potentiometric method, soil organic carbon (SOC) was measured via potassium dichromate (K2Cr2O7) oxidation with external heating, and total nitrogen (TN) was estimated using the acid digestion–Kjeldahl nitrogen determination method. Nitrate nitrogen (NN) was extracted using 0.01 M CaCl2 and determined using a continuous-flow analyzer. Available phosphorus (AP) was extracted using 0.5 M NaHCO3 (soil–solution = 1:20, pH 8.5) and determined using the molybdenum blue colorimetric method. Available potassium (AK) was extracted using 1.0 M NH4Ac (soil–solution = 1:10) and measured using the atomic absorption spectrometry method after melting with sodium hydroxide [14]. The exchangeable Ca2+ and Mg2+ were extracted with 1.0 M ammonium acetate and quantified via atomic absorption spectrometry.
Soil cellulase activity (S-CL), soil urease activity (S-UE), soil polyphenol oxidase activity (S-PPO), soil alkaline phosphatase activity (S-ALP), and soil peroxidase activity (S-POD) were determined using the potassium permanganate titration method, the indophenol blue colorimetry method, the purple gallic acid colorimetry method, the organic group content method, and ultraviolet–visible spectrophotometry, respectively [15].
Yam underground tubers from each replicate were weighed. To measure the stalk diameter and length, 10 tubers were randomly selected in each replicate and measured with a Vernier caliper and straightedge.

2.4. Soil Metabolite Extraction, UHPLC-MS/MS, and Metabolite Data Analysis

Untargeted metabolites in the rhizosphere soil were determined using a liquid chromatography–tandem mass spectrometry (LC–MS/MS) platform (Thermo Fisher Scientific, Waltham, MA, USA). In brief, 100 mg of the rhizosphere soil was placed in a 2 mL centrifuge tube, and 300 µL of pre-cooled methanol water internal standard extraction solution containing 5 ppm of L-2-chloropropylglycine (Aladdin, Shanghai, China) was added. After adding 2 steel balls, the mixed liquids were vortexed for 3 min. The tube was inserted into a high-throughput tissue grinder (TL-48R; Jinsin Industrial Development Co., Ltd., Shanghai, China) and ground for 60 s at 55 Hz, and this step was repeated once. Subsequently, the mixture was placed in an ultrasonic cleaning machine (SB-800D; Jinsin Industrial Development Co., Ltd., Shanghai, China) and ultrasonicated for 10 min, and then it was frozen for 30 min in a −20 °C refrigerator. The mixture was then freeze–centrifuged for 10 min at 4 °C and 12,000 rpm (5430R; Eppendorf International Trade Co., Ltd., Shanghai, China). Afterward, the supernatants were pipetted, filtered with a 0.22 μm PTFE membrane, and stored in a glass liner tube in a brown injection bottle for LC-MS/MS detection [16].
Chromatographic separation was conducted on a Waters ACQUITY UPLC HSS T3 column (1.8 μm, 2.1 mm × 100 mm, 40 °C) with a gradient elution using 0.1% formic acid in water (mobile phase A) and 0.1% formic acid in acetonitrile (mobile phase B) at a flow rate of 0.4 mL/min. The initial condition was set to 95% A and 5% B. Within 5 min, a linear gradient was programmed to reach 35% A and 65% B. Within 1 min, a linear gradient was programmed to 1% A and 99% B, which was then maintained for 1.5 min. Subsequently, a composition of 95% A and 5.0% B was adjusted within 0.1 min and maintained for 2.4 min. Mass spectrometric detection was carried out on a Thermo Orbitrap Exploris 120 instrument operating in both positive and negative ion modes with an HESI source, using data-dependent acquisition (DDA) at resolutions of 60,000 (MS1) and 15,000 (MS2). Raw data were processed using MS-DIAL (v4.9.221218) for peak picking, alignment, gap filling, and normalization. The specific process was as follows: The format data were imported into MS-DIAL software (version 4.9.221218). Peak detection, alignment, filtering, metabolite identification, and other operations were performed using this software. Peaks not detected in more than 50% of the QC samples were filtered. Missing values for undetected peaks were imputed using the gap-filling algorithm within the software, followed by normalization.
Metabolite identification was conducted by searching against the Persenon PSNGM (PerSonalbio Next-Generation Metabolomics) Database, which includes in-house standard libraries, the mzCloud library (https://beta.mzcloud.org/, accessed on 20 April 2025), LIPIDMAPS (https://www.lipidmaps.org/, accessed on 20 April 2025), HMDB (https://hmdb.ca/, accessed on 3 May 2025), MoNA (https://mona.fiehnlab.ucdavis.edu/, accessed on 3 May 2025), NIST_2020_MSMS, and an AI-predicted MSMS spectral library. The main parameters for database searching were as follows: the MS1 tolerance for identification was 0.01, MS2 tolerance for identification was 0.05, smoothing level was 3, minimum peak height was 10,000, minimum peak width was 5, mass slice width was 0.05, and identification score cut-off was 70.
The OPLS-DA (R, 3.3.2, ropls packages) was used to visualize overall differences in metabolic profile across different groups. On the basis of the statistical test used to calculate the p value, the OPLS-DA dimensionality reduction method was used to calculate the variable projection importance (VIP), and the fold change (FC) was used to calculate the multiplicity of difference between groups. Differential metabolites were screened with variable importance of projection (VIP) values > 1.0 and fold change (FC) ≥ 2 or ≤0.5. The enriched pathways were used to browse the differentially abundant metabolites and pathway maps via the KEGG Mapper visualization tool. Venn diagrams, heatmaps, chord diagrams, and volcano plots were generated through https://www.genescloud.cn (accessed on 18 July 2025).

2.5. Soil DNA Extraction, Library Construction, and Microbial Sequencing Data Analysis

The total microbial DNA was extracted from the soil samples using an OMEGA Soil DNA Kit. The purity and quantity of the extracted DNA were assessed through agarose gel electrophoresis and a NanoDrop NC2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), respectively. The primers 338F (5′-ACTCCTACGGGAGGCAGCA-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′) were selected to amplify the variable V3-V4 region of the bacterial 16S rRNA genes, and the primers ITS5 (5′-GGAAGATCGTAACAAGG-3′) and ITS2 (5′-GCTGTGTTCATGATGC-3′) were selected to amplify the fungal ITS1 region. Each primer pair was equipped with specific 7 bp barcodes to facilitate sample multiplexing. Amplification was carried out in 25 µL reactions comprising 5 µL of 5× buffer, 0.25 µL of Fast pfu DNA Polymerase, 2 µL of dNTPs (2.5 mM), 1 µL of each primer (10 µM), 1 µL of DNA template, and 14.75 µL of dd H2O. The thermal protocol was as follows: 98 °C for 5 min; 25 cycles of 98 °C for 30 s, 53 °C for 30 s, and 72 °C for 45 s, with a final 5 min extension at 72 °C. The resulting amplicons were purified using Vazyme VAHTSTM DNA Clean Beads (Vazyme, Nanjing, China), quantified using a Quant-iT PicoGreen dsDNA Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA) and subsequently paired-end sequenced (2 × 250 bp) on an Illumina NovaSeq platform at Shanghai Personal Biotechnology Co., Ltd. (Shanghai, China).
Microbiome bioinformatics were performed with QIIME2 version 2022.11, with slight modifications according to official tutorials. Briefly, raw sequence data were demultiplexed using the demux plugin, followed by primer cutting with the cutadapt plugin [17]. The sequences were then quality filtered, denoised, merged, and chimera removed using the DADA2 plugin [18]. The obtained sequences were merged, and characteristic sequence amplicon sequence variants (ASVs) and abundance data tables were generated. ASVs with a total frequency of less than 10 across all samples were removed. For alpha- and beta-diversity analyses, all samples were rarefied to a depth of 100,000 sequences per sample. This depth retained the majority of samples and was sufficient to capture microbial diversity. Before performing PERMANOVA tests, we assessed the homogeneity of group dispersions using the PERMDISP procedure (based on Bray–Curtis distances). The results confirmed no significant difference in beta-dispersion among groups (p > 0.05).
Alpha diversity indices, including Chao1, Shannon, Simpson, and Pielous’s evenness indices, were used to evaluate bacterial and fungal diversity and richness within individual samples. The Bray–Curtis distance combined with a principal coordinate analysis (PCoA) was performed to investigate the structural variation in the microbial communities among treatments. Bacterial and fungal biomarkers were identified using the linear discriminant analysis effect size (LEfSe), with an α value of 0.05 (the significance level of the Kruskal–Wallis test was used for preliminary screening) and an LDA threshold greater than 3.0 (log10, for reporting markers with significant biological effects).

2.6. Statistical Analysis

All statistical comparisons were performed in IBM SPSS Statistics version 24.0. For multi-group experiments, a one-way analysis of variance (ANOVA) supplemented with the least significant difference (LSD) was applied, considering p < 0.05 as statistically significant.
To investigate the relationships between the metabolomic and microbial community datasets, we conducted a Procrustes analysis within QIIME2 version 2022.11, with significance determined via 999 permutation tests. An integrated analysis of metabolites and microbial taxa was performed using two-way orthogonal partial least squares (O2PLS). Pairwise associations between metabolite levels and microbial relative abundance were quantified using Spearman’s rank correlation in Mothur. Furthermore, the complex interplay among key variables was modeled through partial least squares path modeling (PLS-PM), implemented with the “plspm” package in R (v4.3.3), using a significance level of p < 0.05.

3. Results

3.1. Yam Yield, Tuber Morphology, and Rhizosphere Soil Properties in Different CC Groups

CC significantly decreased the yield and altered the traits of the yam underground tubers. According to Table 1, compared with CC0, the yam yield in CC1 and CC2 decreased by 28.08% (p < 0.05) and 52.86% (p < 0.05), respectively. In fact, the yam yield in CC2 also decreased by 34.48% (p < 0.05) compared with that in CC1. The SL, FW, and DW of the yam underground tubers showed a significant downward trend from CC0 to CC2 (p < 0.05), while the SD observably increased (p < 0.05).
Meanwhile, the nutrient contents and enzyme activities in the yam rhizosphere soil in different CC years were measured. Compared with CC0, the SOC, TN, AP, and AK contents in CC2 sharply increased by 66.37%, 49.15%, 213.13%, and 43.42%, respectively (p < 0.05), while the NN content and pH significantly decreased by 46.54% and 5.32% (p < 0.05). Otherwise, the ExMg and ExCa contents showed no obvious changes among the CC groups (Table 2).
Furthermore, CC significantly altered major enzyme activities (Table 2). Compared to CC0, the PPO and ALP levels in CC2 significantly increased by 56.25% (p < 0.05) and 66.97% (p < 0.05), respectively, while the POD and UE levels declined by 10% and 38.81% (p < 0.05), respectively. Additionally, the CL in CC1 was the highest among the groups. Overall, CC significantly reduced the yam yield, made the underground tubers shorter and thicker, and altered the rhizosphere soil’s physicochemical properties and enzyme activities.

3.2. Metabolomic Analysis of Rhizosphere Soil of Yam in Different CC Groups

A total of 950 metabolites were annotated in 15 samples via metabolome characterization and qualitative analysis, and these metabolites were mainly categorized as organoheterocyclic compounds (16.53%), lipids and lipid-like molecules (13.37%), phenylpropanoids and polyketides (12.63%), benzenoids (10.42%), organic acids and derivatives (8.74%), organic oxygen compounds (4.74%), and others (Figure 1A; Table S1). PLS-DA score plots (Q2 was 0.991) show that metabolite data from CC0, CC1, and CC2 formed three clusters that were completely separated (Figure 1B). Venn diagrams clearly show that the number of unique metabolites was 34, 28, and 19 in CC0_vs_CC1, CC0_vs_CC2, and CC1_vs_CC2, respectively. A total of 388 common metabolites were found among the three rhizosphere soil groups (Figure 1C). Compared to CC0, the number of upregulated differentially expressed metabolites (DEMs) in CC1 and CC2 was 288 and 260, and the number of downregulated DEMs was 40 and 42. It is worth noting that the number of upregulated DEMs in CC1_vs_CC2 decreased to 54, indicating that there were no significant differences in more than 86% of metabolites under short-term CC (Figure 1D; Tables S2–S4).
A total of 38 DEMs related to different CC years were identified, comprising organoheterocyclic compounds (8/38); organic acids and derivatives (8/38); lipids (7/38); benzenoids (5/38); organic nitrogen compounds (3/38); phenylpropanoids and polyketides (3/38); and organic oxygen compounds, organosulfur compounds, hydrocarbons, and nucleosides (1 each). Obviously, the first four categories accounted for 73.68% of the total (Table S5). A heatmap revealed significant differences between CC years. The CC2 treatment was driven mainly by organoheterocyclic compounds (trachelanthine, pughiinin A, and norharman), organic nitrogen compounds (phytosphingosine and trimethylamine N-oxide), lipids (octyl phenylacetate, pyrropirone F, sczukidine, and 1-palmitoyl-2-oleoyl-sn-glycero-3-(phospho-rac-(1-glycerol)), and benzenoids (4-(butoxymethyl) phenol and 1,1-dibutyl-3-(3-chloro-2-methylphenyl) urea). In addition, phenylpropanoids and polyketides (spinosan B and 5,7,4′-trimethoxy-4-phenylcoumarin), organoheterocyclic compounds (morpholine), lipids (nopalyl acetate), organosulfur compounds (isopropyl Isothiocyanate), and organic acids (N-tetradecylcyclohexanecarboxamide and L-isoleucine) increased significantly in the CC1 treatment (Figure 1E; Table S5). Furthermore, a chord plot revealed that phytosphingosine, which was significantly enriched in CC2, was positively correlated with trimethylamine N-oxide and adenosine pentaphosphate. Tsukubadiene (enriched in CC0) was positively correlated with monobutyl phthalate and fludioxonil (Figure 1E and Figure S1).
The DEMs among groups were annotated using the KEGG database and enriched in metabolic pathways. The top 20 pathways with the most significant enrichments were selected. Starch and sucrose metabolism showed the highest enrichment degree, containing 4 DEMs out of 37 total metabolites in the pathway and yielding a rich factor of 0.108. This was followed by linoleic acid metabolism, which contained three DEMs and a rich factor of 0.107. The degree of enrichment was also relatively high among ABC transporters, phosphotransferase system (PTS), purine metabolism, and galactose metabolism. The most significant enrichment was detected in metabolic pathways, where 39 DEMs were enriched, resulting in an FDR of 0.059 (Figure 2; Table S6).

3.3. Analysis of Rhizosphere Microbial Communities of Yam in Different CC Groups

High-throughput sequencing was adopted for the rhizosphere soil of yam in different CC years. A total of 561,200 and 851,244 high-quality sequences were obtained for bacteria and fungi, respectively (Tables S7 and S8). An analysis of the bacterial composition at the phylum level showed that Proteobacteria and Actinobacteriota accounted for more than 60% of the relative abundance, but there were no significant differences in their proportions among the CC groups. Compared to CC0, the relative abundance of Bacteroidota and Myxococcota in CC2 increased by 12.09% (p < 0.05) and 26.69% (p < 0.05), respectively, while that of Chloroflexi, Methylomirabilota, and GAL15 decreased by 9.58% (p < 0.05), 30.90% (p < 0.05), and 62.24% (p < 0.05), respectively (Figure 3A; Table S9). For the fungal composition at the phylum level, Ascomycota was dominant, accounting for more than 95% of the relative abundance among the CC years. Compared to CC0, the relative abundance of Basidiomycota decreased by 62.57% (p < 0.05) in CC1 and by 66.83% (p < 0.05) in CC2 (Figure 3A; Table S10). The microbial composition at the genus level showed greater diversity than that at the phylum level among the CC years, and the top 20 genera of bacteria and fungi accounted for 40% and 45~65% of the total, respectively (Figure 3B). For bacteria, compared to CC0, the relative abundance of Streptomyces increased by 100.06% in CC1 and by 94.19% in CC2, and that of Kribbella decreased by 33.49% in CC1 and by 36.96% in CC2. Additionally, in comparison with CC0, the relative abundances of Lysobacter, Dongia, Steroidobacter, Nonomuraea, and MB-A2-108 declined to varying degrees. Meanwhile, the relative abundances of Pseudomonas, Amycolatopsis, MM2, and Chitinophaga showed an upward trend to varying degrees in CC1 and CC2 (Table S11). For fungi, CC generally enhanced the relative abundances of Preussia, Setophoma, Botryotrichum, Cephalotrichum, Cercophora, and Acrocalymma and decreased the relative abundance of Sarocladium (Table S12).
An alpha diversity analysis was conducted to evaluate the richness and diversity of the microflora. Regarding the rarefaction curve, when the numbers of bacterial and fungal sequences reached 10,000, the curve gradually leveled off (Figures S2 and S3). Combined with the fact that the coverage of all sample libraries at the phylum classification level was above 98.8%, this indicates that, at the sequencing depth in this experiment, all bacteria and fungi in the soil were evenly distributed and analyzed. We found that the Chao1, Simpson, and Shannon indices of the rhizosphere bacteria community significantly increased with the increase in CC years. Especially when compared to CC0, the Chao1 index in CC1 and CC2 increased by 15.66% and 30.44%, respectively. For fungal communities, the top four indices were ranked in the order of CC2 > CC1 > CC0. In addition, Good’s coverage among treatments exceeded 99.99% (Table S13). A principal coordinate analysis (PCoA) was performed to explore the differences in the bacterial and fungal community structures among the CC groups (Figure 3C). We observed that the rhizosphere microflora of the yam in the different CC groups were markedly separated. Venn diagrams were constructed to show the numbers of shared and unique ASVs of bacteria and fungi among the CC groups (Figure 3D). In terms of bacteria, a total of 1426 ASVs were shared among groups, with CC2 having the most unique ASVs (5620) and CC0 having the least (4348). It is worth noting that the number of ASVs shared between CC1 and CC2 (1117) was significantly greater than that shared between CC0 and CC1 (419) and that shared between CC0 and CC2 (423). For fungi, 71 ASVs were shared among groups, and the number of unique ASVs was 235 in CC0, 255 in CC1, and 285 in CC2. Similar to bacteria, the number of ASVs shared between CC1 and CC2 (64) was significantly greater than that shared between CC0 and CC1 (14) and that shared between CC0 and CC2 (9).
The LEfSe was used to identify the biological markers and contribution size of bacteria and fungi with an LDA score > 3.0 as biomarkers. The results revealed that different bacterial taxa were enriched in each CC group (Figure 4A). The number of enriched species was ranked in the order of CC0 (22) > CC2 (15) > CC1 (13). At the genus level, high abundances of Dongia, Nonomuraea, Lechevalieria, MND1, Steroidobacter, Inquilinus, Gaiella, and Phenylobacterium were detected in CC0. Streptomyces, Pseudomonas, and Agromyces were found to be significantly enriched in CC1. Amycolatopsis, Aeromicrobium, Chitinophaga, Sphingomonas, Saccharimonadales, Ohtaekwangia, Chloroplast, and Sphingobium showed a higher abundance in CC2. For fungal biomarkers, the number of enriched species was ranked in the order of CC2 (24) > CC1 (15) > CC0 (11). At the genus level, Preussia, Cephalotrichum, Botryotrichum, Plectosphaerella, Acrocalymma, Trichocladium, Murispora, Purpureocillium, Pseudogymnoascus, Typhula, and Arthrobotry were enriched in CC2. Setophoma, Cercophora, Chaetomium, Botryoderma, Cephaliophora, Niesslia, and Coprinellus showed a higher abundance in CC1. Sarocladium, Coprinopsis, Ilyonectria, Dichotomopilus, Hypocreales_gen_Incertae_sedis, and Podospora were significantly enriched in CC0 (Figure 4B).

3.4. Relationships Among CC Years, Rhizosphere Metabolites, Microbes, Soil Properties, and Tuber Yield and Traits

A partial least squares path model (PLS-PM) analysis elucidated the mutual relationships among the CC years of yam, rhizosphere metabolites, soil properties, microbial diversity, and tuber yield and traits. The model goodness of fit (GOF) was 0.707 (cut-off = 0.36). Internal consistency measures (AVE > 0.5, CR > 0.7) and cross—validated redundancy > 0 confirmed the reliability of the latent constructs (Figure 5). Generally, CC years had a significant direct impact on rhizosphere metabolites (p < 0.001) and soil properties (p < 0.01) and no direct impact on the microbial diversity or tuber yield and traits. Importantly, soil properties had an extremely noteworthy direct influence on metabolites (p < 0.001) and microbial diversity (p < 0.001) and a negative direct influence on tuber yield and traits (p > 0.05). Rhizosphere metabolites had a significant direct effect on bacterial diversity (p < 0.001) and fungal diversity (p < 0.05), whereas it had no significant influence on soil properties (p > 0.05) or tuber yield and traits (p > 0.05). Microbial diversity had a negative direct impact on tuber yield and traits (p > 0.05). Additionally, the number of CC years had a negative indirect effect on tuber yield and traits, and the standardized path coefficient was −0.956 (Table S16).
To reveal the correlations among various indicators within each module, we conducted Spearman’s correlation analysis between 15 tuber traits and soil variables and the relative abundances of 16 different metabolites and 30 microbes (Figure 6A,B). Synthetically, the yield and SL of tubers and the pH, UE, and POD of soil were significantly positively correlated with the metabolites benzamidine and aspergildiene C (p < 0.01); the bacteria Steroidobacter, Kribbella, Nonomuraea, and MND1 (p < 0.05); and the fungi Coprinopsis and Sarocladium (p < 0.05). Additionally, they were significantly negatively correlated with the metabolite norharman (p < 0.01); the bacteria Niastella, Chitinophaga, and Amycolatopsis (p < 0.01); and the fungi Cephalotrichum, Acrocalymma, Trichocladium, Preussia, Plectosphaerella, and Botryotrichum (p < 0.01). Meanwhile, the TN, AP, AK, and ALP of soil were positively correlated with the metabolites norharman, spiculisporic acid, arthriniumnin B, and pyrropirone F (p < 0.05); the bacteria Niastella, Chitinophaga, and Amycolatopsis (p < 0.01); and the fungi Cephalotrichum, Acrocalymma, Trichocladium, Preussia, Plectosphaerella, and Botryotrichum (p < 0.01). Additionally, they were significantly negatively correlated with the metabolites benzamidine and aspergildiene C (p < 0.001); the bacteria Steroidobacter, Kribbella, Nonomuraea, and MND1 (p < 0.01); and the fungi Coprinopsis and Sarocladium (p < 0.01). Additionally, the soil CL and tuber SD were significantly positively correlated with the metabolites spinosan B (p < 0.001) and L-isoleucine (p < 0.05); the bacteria RB41, Streptomyces, Pseudomonas, and Allorhizobium (p < 0.01); and the fungi Setophoma, Cercophora, and Chaetomium (p < 0.01). Additionally, they were significantly negatively correlated with the metabolites fludioxonil, monobutyl phthalate, and tsukubadiene (p < 0.05); the bacteria Dongia and MB-A2-108 (p < 0.05); and the fungi Poaceascoma (p < 0.01). Compared to the relationships between the metabolites and microbes and the soil CL and tuber SD, the soil NN showed the contrary tendency.
A Procrustes analysis based on the Bray–Curtis distance showed that metabolome datasets had a stronger correlation with fungal datasets (p = 0.004) than bacterial datasets (p = 0.133). In addition, sample separation among treatments and aggregation within each treatment displayed a high reproducibility and confidence level (Figure S4). Meanwhile, an O2PLS analysis of metabolites and microbial taxa also confirmed the relational degree. Concretely, spinosan B, tsukubadiene, monobutyl phthalate, fludioxonil, aspergildiene C, norharman, and benzamidine levels were significantly associated with the microbiome. Similarly, the relative abundance of Fusarium, Botryoderma, Lysobacter, Sphingomonas, Chaetomium, Cephalotrichum, Botryotrichum and Plectosphaerella was highly correlated with the metabolome (Figure 6C and Figure S5; Tables S14 and S15). Additionally, to conduct an in-depth analysis of the correlation between specific metabolites and genus-level microflora in rhizosphere soil, we generated Spearman’s correlation heatmaps based on 38 different metabolites and the top 20 dominant bacterial and fungal genera in terms of relative abundance in the microbial communities. Organic acids and derivatives (spiculisporic acid and tributyl phosphate), lipids and lipid-like molecules (aspergildiene C, pyrropirone F, and alpha-tocopherol), organoheterocyclic compounds (fludioxonil, norharman, and N, N-Didemethylalfileramine), benzenoids (monobutyl phthalate, benzamidine, and alverine), phenylpropanoids and polyketides (spinosan B and arthriniumnin B), and hydrocarbons (tsukubadiene) were the main types of metabolites that significantly affected the relative abundance of the microbial community (Table S5). Specifically, the N, N-didemethylalfileramine, benzamidine, aspergildiene C, and alpha-tocopherol metabolite levels were significantly positively correlated with the relative abundance of the bacteria Steroidobacter, Vicinamibacteraceae, Nonomuraea, and MND1 and the fungi Sarocladium and Coprinopsis while being significantly negatively correlated with the relative abundance of the bacteria Niastella and Chitinophaga and the fungi Preussi, Botryotrichum, Cephalotrichum, Acrocalymma, Plectosphaerella and Trichocladium (p < 0.05). Similarly, arthrinium B, pyrropirone F, trimethylamine N-oxide, spiculisporic acid, norharman, and tributyl phosphate metabolite levels were negatively correlated with the relative abundance of the bacteria Steroidobacter, Vicinamibacteraceae, Nonomuraea, and MND1 and the fungi Sarocladium and Coprinopsis while being positively correlated with the relative abundance of the bacteria Amycolatopsis and Niastella and the fungi Preussia, Botryotrichum, Cephalotrichum, Alternaria, Acrocalymma, and Plectosphaerella (p < 0.05). Additionally, the monobutyl phthalate, tsukubadiene, and fludioxonil metabolite levels were also significantly associated with the relative abundance of Streptomyces, Dongia, Pseudomonas, RB41, Setophoma, Cercophora, Chaetomium, and Poaceascoma (p < 0.05) (Figure 6D).

4. Discussion

This study provides a comprehensive analysis of the shifts in rhizosphere soil properties, metabolite profiles, and microbial communities associated with the decline in the productivity of Dioscorea opposita Thunb. cv. Tiegun yam under continuous cropping (CC). The significant reduction in tuber yield and altered plant morphology (Table 1) corresponded with profound changes in the rhizosphere environment across CC years. Crucially, our integrated analysis revealed a series of interconnected changes, where alterations in soil chemistry and the metabolome were closely linked to the restructuring of the microbial community, which, in turn, correlated with negative yam growth and tuber development.

4.1. Changes in the Rhizosphere Chemical Environment Associated with CC

Increasing CC years was strongly associated with significant alterations in the rhizosphere chemical environment, encompassing both soil physicochemical properties and the small-molecule metabolite profile. Compared to CC0, CC2 soils were characterized by sharp increases in SOC, TN, AP, and AK, alongside decreases in NN and soil pH (Table 2), which is consistent with the results of CC for peanut [19], patchouli [20], and Yongfeng yam [12]. The significant decrease in rhizosphere pH (likely driven by H+ release from plant preferential cation uptake and/or the potential application of physiological acidic fertilizers) could profoundly influence nutrient solubility and availability [20]. Although the final pH in this study (7.23) remained mildly alkaline, this shift toward neutrality from the initial extreme alkalinity is generally favorable for nutrient availability. The surge in SOC, AP, and AK with increasing CC years likely stems from increased fertilizer inputs, reduced decomposition rates, and insufficient plant utilization. This accumulation of excessive nutrients leads to nutrient imbalances [11]. The disruption of nitrogen cycling was central to the decline in NN and accumulation of TN, which can be explained by the changes in enzyme activity. Urease facilitates nitrogen cycling by hydrolyzing urea into CO2 and NH3. Conversely, cellulase drives carbon cycling by decomposing plant-derived carbohydrates, thereby acting as a significant component of soil organic matter [21]. The suppression of urease and cellulase activities impaired urea hydrolysis, while reduced peroxidase (POD) activity limited lignin degradation. These effects collectively slowed nitrogen mineralization, particularly in soil containing organic matter with a high C:N ratio. Moreover, the prominent increase in polyphenol oxidase (PPO) activity suggested an accumulation of phenolic root exudates or lignin derivatives from the plant root system. These compounds likely further inhibited nitrification and N availability through allelochemical effects. These interacting processes created a “pseudo-fertile” environment: the total nutrient accumulation masked severely compromised physiological availability to the yam.
Rhizosphere metabolites exuded from plant roots, residues, and microbial secondary products have been identified as a bridge of material and signal exchange between medicinal plants and soil, and they are the key factors in the regulation of rhizosphere microecology. Due to their long cultivation period, medicinal plants release a mass of secondary metabolites into the rhizosphere, and these metabolites include not only the pharmacological components of the medicinal plants but also allelochemicals that cause allelopathic autotoxicity and CCOs. Consequently, a metabolomic analysis provided a more elaborate view of the rhizosphere chemical variation. The distinct separation of CC0, CC1, and CC2 groups in the PLS-DA model and the identification of 38 DEMs directly linked to CC years underscore a profound restructuring of the rhizosphere metabolome (Figure 1B; Table S5). In the present study, we found that the accumulation of self-toxic metabolites in yam rhizosphere soil was a dynamic process, and the types and concentrations of these substances changed regularly with the increase in CC years. The differential enrichment of metabolite classes—such as phenylpropanoids and lipids in CC1, and organoheterocyclic compounds and benzenoids in CC2 (Figure 1E, Table S5)—indicates a progressive change in the chemical signature of the rhizosphere. Specific metabolite changes, such as the decline in the prenol lipid aspergildiene C [22] and alpha-tocopherol (a compound associated with antioxidant activity) [23] and the increase in various carboxylic acids (e.g., spiculisporic acid) and carboximidic acids (e.g., octadecanamide) in CC2 highlight potential biochemical markers associated with CC stress. As integral components of plant metabolism, carboxylic acids and their derivatives act as key metabolic intermediates in the catabolism of amino acids, lipids, and carbohydrates [24]. Concretely, spiculisporic acid, a fermentation product of several fungi, including Aspergillus cejpii [25], Aspergillus candidus [26], and Talaromyces trachyspermus [27], is a known mycotoxin with significant broad-spectrum antimicrobial activity [26]. Octadecanamide, as a type of carboximidic acid, is a bioactive lipid signaling molecule with a variety of physiological activities that play key roles in biological functions [28]. Previous research has also shown that octadecanamide secreted by Bermuda grass has strong allelopathic effects on the seed germination and seedling growth of Tall fescue [29]. The enrichment of pathways such as starch and sucrose metabolism, linoleic acid metabolism, and ABC transporters in the KEGG analysis (Figure 1C) suggests associated disturbances in core metabolic and transport functions within the rhizosphere ecosystem. These changes likely stem from multiple sources: altered root exudation patterns driven by plant stress or developmental changes under CC, the accumulation of microbial metabolites, incomplete decomposition of previous crop residues, and autotoxic compounds from decaying yam roots/tubers. The metabolome, therefore, represents a dynamic chemical fingerprint of the CC-stressed rhizosphere, integrating inputs from the plant, microbes, and altered soil matrix.

4.2. Impact of CC on Rhizosphere Microbial Diversity and Community Structure

The altered chemical environment under CC was closely linked to significant changes in the structure and diversity of the rhizosphere microbial community. CC duration was associated with increased microbial diversity. This increase in taxonomic richness and evenness may be linked to the creation of new ecological niches driven by accumulated organic matter, specific metabolite profiles, and altered pH, potentially favoring a broader range of saprophytic or specialized heterotrophic taxa [30]. However, increased diversity may reflect ecosystem instability or the proliferation of opportunistic taxa rather than a functionally beneficial state for the plant. Furthermore, a PCoA analysis confirmed that the overall composition of both the bacterial and fungal communities was significantly different across CC years (Figure 3C). Venn diagrams further illustrated this divergence, showing a large pool of unique ASVs in each treatment, with CC1 and CC2 communities showing a greater similarity to each other than to CC0 (Figure 3D). This pattern indicates that CC drives the rhizosphere microbiome toward a distinct, alternative state that becomes more established over time. These results align with those of studies on the CC of Rehmannia glutinosa [31], sugarcane [32], and sweet potato [33].
A taxonomic analysis revealed specific associations between CC duration and the relative abundance of key microbial groups. At the phylum level, the stability of dominant phyla (Proteobacteria and Actinobacteriota) suggested a resilient core community [34]. However, the increase in Bacteroidota and Myxococcota in CC2 may be associated with their known roles in decomposing complex organic matter, correlating with the higher SOC content [35,36,37,38]. Conversely, the decrease in oligotrophic phyla such as Chloroflexi and the fungal phylum Basidiomycota suggests a shift away from communities adapted to lower nutrient availability and potentially a reduction in certain beneficial fungal guilds [39,40]. Shifts at the genus level were more pronounced and have direct implications for hypothesized plant–microbe interactions. We observed a significant reduction in the relative abundance of genera such as Kribbella, Lysobacter, and Sarocladium. In the literature, these genera are commonly associated with plant growth promotion, phosphate solubilization, or antagonism against pathogens [20,41,42]. Conversely, genera including Streptomyces (some species of which are pathogenic) [43] and Fusarium (a well-known phytopathogen) [43,44,45,46] showed increased or variable trends in abundance under CC. As plant growth-promoting rhizobacteria, Pseudomonas are involved in plant pathogen inhibition in many soils. Different to the decreased abundance of Pseudomonas in the third year of cut chrysanthemum CC [37], our study showed a significant increase with yam CC, probably because not all species of Pseudomonas are beneficial, and some are even plant pathogenic bacteria.

4.3. Synthesis: An Integrated Model of Associations Linking CC to Yam Performance Decline

Integrating our findings through partial least squares path modeling (PLS-PM) enabled us to propose a coherent, though correlative, model of system-level associations (Figure 5). The model supports a cascade where the duration of continuous cropping (CC) is not a direct driver of yield loss or microbial change. Instead, it is strongly and directly associated with primary alterations in the soil physicochemical properties and its metabolic profile. These chemical changes—the “pseudo-fertile” nutrient status, acidification, enzyme activity shifts, and the accumulation of specific metabolites—exhibit strong statistical associations with the restructuring of the microbial community. This includes links to both the increased alpha diversity and, more importantly, the pronounced shift in beta diversity (community composition) [47,48]. For example, the pathway analysis suggests associations where soil acidification and reduced urease activity are linked to the decreased abundance of metabolites such as aspergildiene C and benzamidine. These chemical changes are further associated with an increased relative abundance of fungal genera such as Plectosphaerella, Botryotrichum, and Cephalotrichum. This entire chain of correlations culminates in a strong negative association with the final tuber yield and plant growth metrics. Among these soil properties, soil acidification under CC was the most critical influencing factor, which altered the availability of soluble nutrients and impacted their utilization by the soil microbial community, thus significantly affecting microbial diversity [49,50]. Similarly, the accumulation of total nitrogen and available phosphorus is linked to the enrichment of metabolites such as norharman and spiculisporic acid, which, in turn, are associated with shifts in the microbial community structure. Therefore, the observed decline in yam productivity under CC is best understood as the outcome of a series of tightly linked associations within the rhizosphere ecosystem. Continuous cropping initiates changes in the soil chemical environment, which are correlated with a reprogramming of the rhizosphere metabolome. Together, these altered chemical parameters appear to be key factors associated with the selection for a distinct and potentially less plant-beneficial microbial community. This restructured microbiome is then correlated with negative plant performance outcomes, completing a feedback loop that characterizes the CCOs.

4.4. Study Limitations

Several important limitations must be considered when interpreting these results. First, this is an observational field study. While it reveals strong correlations between CC duration and multiple soil–plant parameters, it cannot definitively establish causality. The observed effects could be influenced by unmeasured confounding variables inherent to agricultural fields. Second, the functional roles assigned to specific metabolites and microbial taxa are inferred from the existing literature and metabolic pathway analyses. We lack direct experimental evidence, such as bioassays or microbial isolation/functional tests, to confirm their specific effects on yam growth or microbial interactions in this system. Third, while multivariate models (PLS-DA and PLS-PM) are valuable for identifying patterns and potential linkages in complex datasets, they carry a risk of overfitting, especially with a limited number of biological replicates. Their outputs suggest relationships but do not prove mechanistic drivers.

5. Conclusions and Future Directions

In conclusion, this integrated analysis demonstrates that the decline in Tiegun yam under continuous cropping is associated with a comprehensive and sequential restructuring of the rhizosphere ecosystem. We observed correlated shifts from soil chemistry and metabolism to microbial ecology, culminating in poor plant performance. The value of this work lies in its identification of key correlated factors—specific soil properties, metabolite families, and microbial taxa—that form the CCO signature. These factors serve as priority targets for future manipulative experiments. Future research should employ pot experiments with soil sterilization and re-inoculation, functional bioassays with key metabolites, and isolation of target microbes to move beyond correlation and empirically test the causative roles of the components identified in the complex phenomenon of CCOs.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/agronomy16010080/s1, Figure S1: Chord diagrams of differentially expressed metabolites among different CC groups; Figure S2: Rarefaction curves of 16S rRNA genes; Figure S3: Rarefaction curves of ITS genes; Figure S4: Correlation analysis of rhizosphere metabolites with bacteria (a) and fungi (b) in different CC groups. Table S1: List of quantitative metabolite identification; Tables S2–S4: differentially expressed metabolites (DEMs) in CC0_vs_CC1, CC0_vs_CC2, CC1_vs_CC2, respectively; Table S5: DEMs related to different CC years; Table S6: DEMs were annotated using KEGG and enriched into metabolic pathways; Tables S7 and S8: Sample sequencing volume status of bacteria and fungi; Tables S9 and S10: The proportions of top 20 dominant bacterial communities and top 5 dominant fungal communities at phylum level; Tables S11 and S12: The proportions of top 20 dominant bacteria and fungi communities at genus level; Table S13: Diversity index of bacterial and fungal community in rhizosphere soil in different CC years; Table S14: The important variables that affected the microbiome; Table S15: The important variables that affected the metabolome; Table S16: The standardized path coefficients related to PLS-PM.

Author Contributions

Conceptualization, X.W. and X.J.; methodology, investigation, data curation and writing—original draft preparation, P.Z.; data analysis and figures, W.G.; field sampling and data collection, L.H.; statistical analysis and writing assistance, X.H.; investigation and resources, A.X.; equipment, drugs and manuscript revise, H.W.; writing—review and editing, P.Z. and X.J.; funding acquisition, P.Z., X.J. and X.W. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Science and Technology Innovation Project of Higher Education Institutions in Shanxi Province (No. 2023L054), the Basic Research Program of Shanxi (No. 202403021212076), the Youth Fund of Cotton Research Institute, Shanxi Agricultural University (No. SJJQN2023-04), the Earmarked Fund for China Agriculture Research System (CARS-21).

Data Availability Statement

All data supporting the findings of this study are available within the paper and within its Supplementary Materials published online.

Acknowledgments

We gratefully acknowledge the anonymous reviewers and editors for their helpful comments that greatly improved this manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
CCContinuous cropping
CCOsContinuous cropping obstacles
DEMsDifferentially expressed metabolites
PLS-PMPartial least squares path model

References

  1. Haoyu, C.; Xinyang, T.; Huanqi, Y.; Ye, P.; Quancai, S. Chinese yam (Dioscorea opposita) and its bioactive compounds: The beneficial effects on gut microbiota and gut health. Curr. Opin. Food Sci. 2024, 55, 101121. [Google Scholar] [CrossRef]
  2. Ye, L.; Shengyang, J.; Tao, X.; Yongheng, Z.; Minghao, X.; Yuqi, L.; Mingjun, L.; Bei, F.; Fengzhong, W.; Jianbo, X.; et al. Chinese yam (Dioscorea): Nutritional value, beneficial effects, and food and pharmaceutical applications. Trends Food Sci. Technol. 2023, 134, 29–40. [Google Scholar] [CrossRef]
  3. Gao, J.; Chen, L.; Wang, J.; Zhao, W.; Zhang, J.; Qin, Z.; Wang, M.; Chen, X.; Li, M.; Yang, Q. Response of the Symbiotic Microbial Community of Dioscorea opposita Cultivar Tiegun to Root-Knot Nematode Infection. Plant Dis. 2024, 108, 2472–2483. [Google Scholar] [CrossRef]
  4. Qin, Z.; Zhao, W.; Ma, H.; Gao, J.; Wang, R.; Li, C.; Zhang, H.; Li, M.; Yang, Q. Characteristics of microbial community during the different growth stages of yam (Dioscorea opposita Thunb. cv. Tiegun). Appl. Soil Ecol. 2024, 204, 105696. [Google Scholar] [CrossRef]
  5. Chen, S.; Qi, G.; Luo, T.; Zhang, H.; Jiang, Q.; Wang, R.; Zhao, X. Continuous-cropping tobacco caused variance of chemical properties and structure of bacterial network in soils. Land Degrad. Dev. 2018, 29, 4106–4120. [Google Scholar] [CrossRef]
  6. Zhang, B.; Weston, L.A.; Li, M.; Zhu, X.; Weston, P.A.; Feng, F.; Zhang, B.; Zhang, L.; Gu, L.; Zhang, Z. Rehmannia glutinosa Replant Issues: Root Exudate-Rhizobiome Interactions Clearly Influence Replant Success. Front. Microbiol. 2020, 11, 1413. [Google Scholar] [CrossRef] [PubMed]
  7. Wu, Z.-G.; Jiang, W.; Nitin, M.; Bao, X.-Q.; Chen, S.-L.; Tao, Z.-M. Characterizing diversity based on nutritional and bioactive compositions of yam germplasm (Dioscorea spp.) commonly cultivated in China. J. Food Drug Anal. 2016, 24, 367–375. [Google Scholar] [CrossRef]
  8. Xin, A.; Li, X.; Jin, H.; Yang, X.; Zhao, R.; Liu, J.; Qin, B. The accumulation of reactive oxygen species in root tips caused by autotoxic allelochemicals—A significant factor for replant problem of Angelica sinensis (Oliv.) Diels. Ind. Crops Prod. 2019, 138, 111432. [Google Scholar] [CrossRef]
  9. Zhang, B.; Weston, P.A.; Gu, L.; Zhang, B.; Li, M.; Wang, F.; Tu, W.; Wang, J.; Weston, L.A.; Zhang, Z. Identification of phytotoxic metabolites released from Rehmannia glutinosa suggest their importance in the formation of its replant problem. Plant Soil 2019, 441, 439–454. [Google Scholar] [CrossRef]
  10. Wang, X.; Alami, M.M.; Gong, S.; Cheng, Q.; Chen, C.; Li, X.; Zhong, S.; He, Z.; Chen, D.; Feng, S.; et al. Utilizing Microbial Inoculants to Alleviate Continuous Cropping Obstacles: Insights into the Metabolites and Transcriptomic Responses of Pinellia ternata. Metabolites 2025, 15, 189. [Google Scholar] [CrossRef]
  11. Liu, S.; Wang, Z.; Niu, J.; Dang, K.; Zhang, S.; Wang, S.; Wang, Z. Changes in physicochemical properties, enzymatic activities, and the microbial community of soil significantly influence the continuous cropping of Panax quinquefolius L. (American ginseng). Plant Soil 2021, 463, 427–446. [Google Scholar] [CrossRef]
  12. Yao, J.; Wu, C.; Fan, L.; Kang, M.; Liu, Z.; Huang, Y.; Xu, X.; Yao, Y. Effects of the Long-Term Continuous Cropping of Yongfeng Yam on the Bacterial Community and Function in the Rhizospheric Soil. Microorganisms 2023, 11, 274. [Google Scholar] [CrossRef]
  13. An, L.; Yuan, Y.; Chen, H.; Li, M.; Ma, J.; Zhou, J.; Zheng, L.; Ma, H.; Chen, Z.; Hao, C.; et al. Comprehensive widely targeted metabolomics to decipher the molecular mechanisms of Dioscorea opposita thunb. cv. Tiegun quality formation during harvest. Food Chem. X 2024, 21, 101159. [Google Scholar] [CrossRef]
  14. Wang, J.; Yang, X.; Huang, S.; Wu, L.; Cai, Z.; Xu, M. Long-term combined application of organic and inorganic fertilizers increases crop yield sustainability by improving soil fertility in maize-wheat cropping systems. J. Integr. Agric. 2025, 24, 290–305. [Google Scholar] [CrossRef]
  15. Chen, F.; Xie, Y.; Jia, Q.; Li, S.; Li, S.; Shen, N.; Jiang, M.; Wang, Y. Effects of the Continuous Cropping and Soilborne Diseases of Panax Ginseng C. A. Meyer on Rhizosphere Soil Physicochemical Properties, Enzyme Activities, and Microbial Communities. Agronomy 2023, 13, 210. [Google Scholar] [CrossRef]
  16. Song, R.; Lv, B.; He, Z.; Li, H.; Wang, H. Rhizosphere metabolite dynamics in continuous cropping of vineyards: Impact on microflora diversity and co-occurrence networks. Microbiol. Res. 2025, 296, 128134. [Google Scholar] [CrossRef]
  17. Kechin, A.; Boyarskikh, U.; Kel, A.; Filipenko, M. cutPrimers: A New Tool for Accurate Cutting of Primers from Reads of Targeted Next Generation Sequencing. J. Comput. Biol. 2017, 24, 1138–1143. [Google Scholar] [CrossRef] [PubMed]
  18. Callahan, B.J.; McMurdie, P.J.; Rosen, M.J.; Han, A.W.; Johnson, A.J.A.; Holmes, S.P. DADA2: High-resolution sample inference from Illumina amplicon data. Nat. Methods 2016, 13, 581–583. [Google Scholar] [CrossRef]
  19. Shao, S.; Chen, M.; Liu, W.; Hu, X.; Wang, E.-T.; Yu, S.; Li, Y. Long-term monoculture reduces the symbiotic rhizobial biodiversity of peanut. Syst. Appl. Microbiol. 2020, 43, 126101. [Google Scholar] [CrossRef] [PubMed]
  20. Gu, G.; Haq, M.Z.U.; Sun, X.; Zhou, J.; Liu, Y.; Yu, J.; Yang, D.; Yang, H.; Wu, Y. Continuous cropping of Patchouli alters soil physiochemical properties and rhizosphere microecology revealed by metagenomic sequencing. Front. Microbiol. 2025, 15, 1482904. [Google Scholar] [CrossRef]
  21. Liu, P.F.; Zhao, Y.K.; Ma, J.N.; Cao, Y.; Zhang, M.X.; Yu, J.; Guan, H.B.; Xing, Y.S.; Wang, X.Q.; Jia, X. Impact of various intercropping modes on soil quality, microbial communities, yield and quality of Platycodon grandiflorum (Jacq.) A. DC. BMC Plant Biol. 2025, 25, 503. [Google Scholar] [CrossRef] [PubMed]
  22. Guo, J.; Cai, Y.-S.; Cheng, F.; Yang, C.; Zhang, W.; Yu, W.; Yan, J.; Deng, Z.; Hong, K. Genome Mining Reveals a Multiproduct Sesterterpenoid Biosynthetic Gene Cluster in Aspergillus ustus. Org. Lett. 2021, 23, 1525–1529. [Google Scholar] [CrossRef] [PubMed]
  23. Haroon, M.; Khan, W.U.; Munir, B.; Ahmad, S.R.; Rehman, A.; Akram, W.; Munir, A.; Sardar, R.; Yasin, N.A. Seed priming with alpha-tocopherol alleviates microplastic stress in Brassica rapa through modulations in morphological, physiological and biochemical attributes. Chemosphere 2025, 371, 144060. [Google Scholar] [CrossRef]
  24. Qi, S.; Zha, L.; Huang, D.; Peng, Y.; Li, G.; Yin, D. Metabolomics and network pharmacology profiling revealed soilless cultivation reduces the accumulation of harmful compounds in Houttuynia cordata. Sci. Hortic. 2024, 323, 112438. [Google Scholar] [CrossRef]
  25. Osama, S.; El Sherei, M.; Al-Mahdy, D.A.; Bishr, M.; Salama, O.; Raafat, M.M. Antimicrobial activity of spiculisporic acid isolated from endophytic fungus Aspergillus cejpii of Hedera helix against MRSA. Braz. J. Microbiol. 2024, 55, 515–527. [Google Scholar] [CrossRef]
  26. Wang, R.; Guo, Z.K.; Li, X.M.; Chen, F.X.; Zhan, X.F.; Shen, M.H. Spiculisporic acid analogues of the marine-derived fungus, Aspergillus candidus strain HDf2, and their antibacterial activity. Antonie Van Leeuwenhoek 2015, 108, 215–219. [Google Scholar] [CrossRef] [PubMed]
  27. Moriwaki-Takano, M.; Asada, C.; Nakamura, Y. Production of spiculisporic acid by Talaromyces trachyspermus in fed-batch bioreactor culture. Bioresour. Bioprocess. 2021, 8, 1–11. [Google Scholar] [CrossRef]
  28. Li, Z.; Dong, F.; Sun, Y.; Sun, Z.; Song, X.; Dong, Y.; Huang, X.; Zhong, J.; Zhang, R.; Wang, M.; et al. Qualitative and Quantitative Analysis of Six Fatty Acid Amides in 11 Edible Vegetable Oils Using Liquid Chromatography-Mass Spectrometry. Front. Nutr. 2022, 9, 857858. [Google Scholar] [CrossRef]
  29. Liang, Z.Y.; Li, X.Y.; Zhang, H.; Li, J.; Bian, X.J.; Xu, J.C. Allelopathic effects of Bermuda grass (Cynodon dactylon L.) root exudates on seed germination and seedling growth of Tall fescue (Festuca arundinacea Schreb). Allelopath. J. 2018, 44, 25–34. [Google Scholar] [CrossRef]
  30. Zhalnina, K.; Louie, K.B.; Hao, Z.; Mansoori, N.; da Rocha, U.N.; Shi, S.; Cho, H.; Karaoz, U.; Loque, D.; Bowen, B.P.; et al. Dynamic root exudate chemistry and microbial substrate preferences drive patterns in rhizosphere microbial community assembly. Nat. Microbiol. 2018, 3, 470–480. [Google Scholar] [CrossRef]
  31. Wu, L.; Chen, J.; Xiao, Z.; Zhu, X.; Wang, J.; Wu, H.; Wu, Y.; Zhang, Z.; Lin, W. Barcoded Pyrosequencing Reveals a Shift in the Bacterial Community in the Rhizosphere and Rhizoplane of Rehmannia glutinosa under Consecutive Monoculture. Int. J. Mol. Sci. 2018, 19, 850. [Google Scholar] [CrossRef] [PubMed]
  32. Pang, Z.; Tayyab, M.; Kong, C.; Liu, Q.; Liu, Y.; Hu, C.; Huang, J.; Weng, P.; Islam, W.; Lin, W.; et al. Continuous Sugarcane Planting Negatively Impacts Soil Microbial Community Structure, Soil Fertility, and Sugarcane Agronomic Parameters. Microorganisms 2021, 9, 2008. [Google Scholar] [CrossRef] [PubMed]
  33. Li, H.; Wang, J.; Liu, Q.; Zhou, Z.; Chen, F.; Xiang, D. Effects of consecutive monoculture of sweet potato on soil bacterial community as determined by pyrosequencing. J. Basic Microbiol. 2019, 59, 181–191. [Google Scholar] [CrossRef]
  34. Gong, Y.; Yang, S.; Chen, S.; Zhao, S.; Ai, Y.; Huang, D.; Yang, K.; Cheng, H. Soil microbial responses to simultaneous contamination of antimony and arsenic in the surrounding area of an abandoned antimony smelter in Southwest China. Environ. Int. 2023, 174, 107897. [Google Scholar] [CrossRef] [PubMed]
  35. Sui, J.; Yang, J.; Li, C.; Zhang, L.; Hua, X. Effects of a Microbial Restoration Substrate on Plant Growth and Rhizosphere Microbial Community in a Continuous Cropping Poplar. Microorganisms 2023, 11, 486. [Google Scholar] [CrossRef]
  36. Han, S.; Luo, X.; Liao, H.; Nie, H.; Chen, W.; Huang, Q. Nitrospira are more sensitive than Nitrobacter to land management in acid, fertilized soils of a rapeseed-rice rotation field trial. Sci. Total Environ. 2017, 599, 135–144. [Google Scholar] [CrossRef]
  37. Wang, T.; Yang, K.; Ma, Q.; Jiang, X.; Zhou, Y.; Kong, D.; Wang, Z.; Parales, R.E.; Li, L.; Zhao, X.; et al. Rhizosphere Microbial Community Diversity and Function Analysis of Cut Chrysanthemum During Continuous Monocropping. Front. Microbiol. 2022, 13, 801546. [Google Scholar] [CrossRef]
  38. Li, L.; Huang, D.; Hu, Y.; Rudling, N.M.; Canniffe, D.P.; Wang, F.; Wang, Y. Globally distributed Myxococcota with photosynthesis gene clusters illuminate the origin and evolution of a potentially chimeric lifestyle. Nat. Commun. 2023, 14, 1–14. [Google Scholar] [CrossRef]
  39. Liang, M.; Johnson, D.; Burslem, D.F.R.P.; Yu, S.; Fang, M.; Taylor, J.D.; Taylor, A.F.S.; Helgason, T.; Liu, X. Soil fungal networks maintain local dominance of ectomycorrhizal trees. Nat. Commun. 2020, 11, 2636. [Google Scholar] [CrossRef]
  40. Li, H.; Yang, Y.; Lei, J.; Gou, W.; Crabbe, M.J.C.; Qi, P. Effects of Continuous Cropping of Codonopsis pilosula on Rhizosphere Soil Microbial Community Structure and Metabolomics. Agronomy 2024, 14, 2014. [Google Scholar] [CrossRef]
  41. Puopolo, G.; Giovannini, O.; Pertot, I. Lysobacter capsici AZ78 can be combined with copper to effectively control Plasmopara viticola on grapevine. Microbiol. Res. 2014, 169, 633–642. [Google Scholar] [CrossRef]
  42. Cobo-Diaz, J.F.; Baroncelli, R.; Le Floch, G.; Picot, A. Combined Metabarcoding and Co-occurrence Network Analysis to Profile the Bacterial, Fungal and Fusarium Communities and Their Interactions in Maize Stalks. Front. Microbiol. 2019, 10, 261. [Google Scholar] [CrossRef]
  43. Shi, W.; Lv, M.; Wang, R.; Saleem, M.; Wang, L.; Li, M.; Wang, B.; Lin, R.; Xu, B.; Yang, C.; et al. Geocaulosphere soil bacterial community drives potato common scab outcomes beyond pathogen abundance. npj Biofilms Microbiomes 2025, 11, 1–14. [Google Scholar] [CrossRef]
  44. Lu, Y.; Dong, X.; Huang, X.; Zhao, D.-g.; Zhao, Y.; Peng, L. Combined analysis of the transcriptome and proteome of Eucommia ulmoides Oliv. (Duzhong) in response to Fusarium oxysporum. Front. Chem. 2022, 10, 1053227. [Google Scholar] [CrossRef] [PubMed]
  45. Bao, G.; Bi, Y.; Li, Y.; Kou, Z.; Hu, L.; Ge, Y.; Wang, Y.; Wang, D. Overproduction of reactive oxygen species involved in pathogenicity of Fusarium in potato tubers. Physiol. Mol. Plant Pathol. 2014, 86, 35–42. [Google Scholar] [CrossRef]
  46. Shi, M.; Zhou, W.; Zhou, Y.; Yao, Y.; Wang, Y.; Luo, M.; Zhang, S.; Zhang, Y.; Wang, M. Identification and biocontrol of the key pathogenic fungus inhibiting tuber expansion in Rehmannia glutinosa from Henan Province. Biol. Control 2025, 206, 105798. [Google Scholar] [CrossRef]
  47. Wang, Y.; Zhang, H.; Zhang, Y.; Fei, J.; Rong, X.; Peng, J.; Luo, G. Crop rotation-driven changes in rhizosphere metabolite profiles regulate soil microbial diversity and functional capacity. Agric. Ecosyst. Environ. 2023, 358, 108716. [Google Scholar] [CrossRef]
  48. Li, X.-G.; Ding, C.-F.; Hua, K.; Zhang, T.-L.; Zhang, Y.-N.; Zhao, L.; Yang, Y.-R.; Liu, J.-G.; Wang, X.-X. Soil sickness of peanuts is attributable to modifications in soil microbes induced by peanut root exudates rather than to direct allelopathy. Soil Biol. Biochem. 2014, 78, 149–159. [Google Scholar] [CrossRef]
  49. Zhang, X.; Wei, H.; Chen, Q.; Han, X. The counteractive effects of nitrogen addition and watering on soil bacterial communities in a steppe ecosystem. Soil Biol. Biochem. 2014, 72, 26–34. [Google Scholar] [CrossRef]
  50. Wang, B.; Lu, Y.; Li, W.; He, S.; Lin, R.; Qu, P.; Chen, H.; Zhang, F.; Zhao, M.; Shi, X.; et al. Effects of the Continuous Cropping of Amomum villosum on Rhizosphere Soil Physicochemical Properties, Enzyme Activities, and Microbial Communities. Agronomy 2022, 12, 2548. [Google Scholar] [CrossRef]
Agronomy 16 00080 g001
Figure 1. Metabolomic analysis of rhizosphere soil in different CC years. (A) Cycle of primary classification of metabolites. (B) PLS-DA score plots derived from metabolites. (C) Venn diagram of rhizosphere soil metabolites. (D) Volcano map of differential metabolites, where red dots represent upregulated metabolites, green dots represent downregulated metabolites, and gray dots represent no differential metabolites. (E) Heatmap analysis of the relative contents of differentially abundant metabolites. CC0, rhizosphere soil from the first-year yam. CC1, rhizosphere soil after continuous cropping for 1 year; CC2, rhizosphere soil after continuous cropping for 2 years.
Figure 1. Metabolomic analysis of rhizosphere soil in different CC years. (A) Cycle of primary classification of metabolites. (B) PLS-DA score plots derived from metabolites. (C) Venn diagram of rhizosphere soil metabolites. (D) Volcano map of differential metabolites, where red dots represent upregulated metabolites, green dots represent downregulated metabolites, and gray dots represent no differential metabolites. (E) Heatmap analysis of the relative contents of differentially abundant metabolites. CC0, rhizosphere soil from the first-year yam. CC1, rhizosphere soil after continuous cropping for 1 year; CC2, rhizosphere soil after continuous cropping for 2 years.
Agronomy 16 00080 g001
Agronomy 16 00080 g002
Figure 2. An enrichment factor bubble diagram reflecting the degree of enrichment of differential metabolites. Rich factor refers to the ratio of the number of enriched differential metabolites in a pathway to the number of annotated metabolites. A larger rich factor indicates greater enrichment. The FDR ranges from 0 to 1, with lower values indicating more significant enrichment. The top 20 KEGG pathways with the smallest FDR values are shown. Each dot signifies a metabolic pathway, where the abscissa indicates the rich factor value for different metabolic pathways and the ordinate indicates the enrichment pathway; the color depth of the dot represents the p value, where the bluer the color, the more significant the enrichment; and the size of the dot represents the number of enriched differential metabolites.
Figure 2. An enrichment factor bubble diagram reflecting the degree of enrichment of differential metabolites. Rich factor refers to the ratio of the number of enriched differential metabolites in a pathway to the number of annotated metabolites. A larger rich factor indicates greater enrichment. The FDR ranges from 0 to 1, with lower values indicating more significant enrichment. The top 20 KEGG pathways with the smallest FDR values are shown. Each dot signifies a metabolic pathway, where the abscissa indicates the rich factor value for different metabolic pathways and the ordinate indicates the enrichment pathway; the color depth of the dot represents the p value, where the bluer the color, the more significant the enrichment; and the size of the dot represents the number of enriched differential metabolites.
Agronomy 16 00080 g002
Agronomy 16 00080 g003
Figure 3. Microbial analysis of rhizosphere soil in different CC years. (A,B) Composition of the bacterial and fungal communities at the phylum and genus levels, respectively. (C) PCoA analysis of the bacterial and fungal community structures. (D) Venn diagrams of the bacterial and fungal ASVs. CC0, rhizosphere soil from the first-year yam; CC1, rhizosphere soil after continuous cropping for 1 year; CC2, rhizosphere soil after continuous cropping for 2 years.
Figure 3. Microbial analysis of rhizosphere soil in different CC years. (A,B) Composition of the bacterial and fungal communities at the phylum and genus levels, respectively. (C) PCoA analysis of the bacterial and fungal community structures. (D) Venn diagrams of the bacterial and fungal ASVs. CC0, rhizosphere soil from the first-year yam; CC1, rhizosphere soil after continuous cropping for 1 year; CC2, rhizosphere soil after continuous cropping for 2 years.
Agronomy 16 00080 g003
Agronomy 16 00080 g004
Figure 4. LEFSe analysis of rhizosphere soil in different CC years: (A) bacteria and (B) fungi. The cladogram on the left side shows the different classification rank relationships between groups from the phylum to the genus (from the inner ring to the outer ring), where the LDA threshold is >3.0, and the p value is defined as <0.05. The histogram of LDA effect values for marker species on the right side shows the taxonomic units, with significant differences among groups.
Figure 4. LEFSe analysis of rhizosphere soil in different CC years: (A) bacteria and (B) fungi. The cladogram on the left side shows the different classification rank relationships between groups from the phylum to the genus (from the inner ring to the outer ring), where the LDA threshold is >3.0, and the p value is defined as <0.05. The histogram of LDA effect values for marker species on the right side shows the taxonomic units, with significant differences among groups.
Agronomy 16 00080 g004
Agronomy 16 00080 g005
Figure 5. PLS-PM showing the mutual relationships among the continuous cropping years of yam, rhizosphere metabolites, soil properties, microbial diversity, and tuber yield and traits. Coefficients related to measured variables are shown adjacent to the variables. Significant relationships are indicated by red arrows, whereas non-significant relationships are indicated by blue arrows. The numbers above the arrows represent standardized path coefficients. Significance levels are denoted by asterisks: * p ≤ 0.05, ** p ≤ 0.01, and *** p ≤ 0.001. The model is evaluated using goodness-of-fit statistics.
Figure 5. PLS-PM showing the mutual relationships among the continuous cropping years of yam, rhizosphere metabolites, soil properties, microbial diversity, and tuber yield and traits. Coefficients related to measured variables are shown adjacent to the variables. Significant relationships are indicated by red arrows, whereas non-significant relationships are indicated by blue arrows. The numbers above the arrows represent standardized path coefficients. Significance levels are denoted by asterisks: * p ≤ 0.05, ** p ≤ 0.01, and *** p ≤ 0.001. The model is evaluated using goodness-of-fit statistics.
Agronomy 16 00080 g005
Agronomy 16 00080 g006
Figure 6. Spearman’s correlation analysis between tuber traits and soil variables and rhizosphere differential metabolites (A) and microbial communities (B). The red and green squares represent positive and negative correlations, respectively. The color depth indicates the strength of the correlation, and * indicates significance (* p < 0.05, ** p < 0.01, *** p < 0.001). (C) O2PLS load chart of metabolites and microbial taxa. (D) A correlation heatmap between rhizosphere differential metabolites and genus-level microflora.
Figure 6. Spearman’s correlation analysis between tuber traits and soil variables and rhizosphere differential metabolites (A) and microbial communities (B). The red and green squares represent positive and negative correlations, respectively. The color depth indicates the strength of the correlation, and * indicates significance (* p < 0.05, ** p < 0.01, *** p < 0.001). (C) O2PLS load chart of metabolites and microbial taxa. (D) A correlation heatmap between rhizosphere differential metabolites and genus-level microflora.
Agronomy 16 00080 g006
Table 1. Underground tuber yield and traits of yam for different CC years.
Table 1. Underground tuber yield and traits of yam for different CC years.
TreatmentYield
(kg/hm2)		Stalk Diameter
(mm)		Stalk Length
(cm)		Fresh Weight
(g/Tuber)		Dry Weight
(g/Tuber)		Moisture Content
(%)
CC035,504 ± 685 a28.98 ± 1.10 c142.67 ± 5.50 a622.25 ± 10.71 a164.93 ± 4.02 a73.51 ± 0.2 a
CC125,536 ± 324 b41.52 ± 0.61 a114.56 ± 2.75 b511.11 ± 11.84 b142.81 ± 5.18 b72.11 ± 0.49 b
CC216,735 ± 859 c38.51 ± 1.12 b101.44 ± 5.09 b394.44 ± 17.62 c106.24 ± 5.62 c73.14 ± 0.31 ab
SD, stalk diameter; SL, stalk length; FW, fresh weight; DW, dry weight; MC, moisture content. Data are shown as means ± standard deviations (n = 10). Different letters represent significant differences (p < 0.05).
Table 2. Rhizosphere soil physicochemical properties and enzyme activities of yam for different CC years.
Table 2. Rhizosphere soil physicochemical properties and enzyme activities of yam for different CC years.
TreatmentSOC (g/kg)TN (g/kg)NN (mg/kg)AP (mg/kg)AK (mg/kg)pHExCa (cmol/kg)
CC05.65 ± 0.99 b0.59 ± 0.05 b73.07 ± 9.40 a2.97 ± 0.38 c116.67 ± 4.93 b8.27 ± 0.05 a107.77 ± 14.89 a
CC16.66 ± 0.10 b0.67 ± 0.02 b23.50 ± 3.51 c6.30 ± 0.56 b128.67 ± 5.51 b8.15 ± 0.07 b119.67 ± 2.08 a
CC29.40 ± 0.55 a0.88 ± 0.04 a39.06 ± 4.47 b9.30 ± 0.98 a167.33 ± 8.74 a7.83 ± 0.03 c103.00 ± 1.00 a
TreatmentExMg
(cmol/kg)		PPO
(mg/2 h/g)		POD
(mg/2 h/g)		ALP
(mg/2 h/g)		UE
(mg/24 h/g)		CL
(mg/72 h/g)
CC05.20 ± 0.17 a0.016 ± 0.002 b0.050 ± 0.001 a0.221 ± 0.008 b0.621 ± 0.057 a1.432 ± 0.138 c
CC15.25 ± 0.15 a0.022 ± 0.002 a0.048 ± 0.002 ab0.238 ± 0.011 b0.530 ± 0.028 b2.195 ± 0.185 a
CC25.00 ± 0.27 a0.025 ± 0.001 a0.045 ± 0.000 b0.369 ± 0.027 a0.380 ± 0.011 c1.865 ± 0.149 b
SOC, soil organic carbon; TN, total nitrogen; NN, nitrate nitrogen; AP, available phosphorus; AK, available potassium; ExCa, exchangeable calcium; ExMg, exchangeable magnesium; PPO, polyphenol oxidase; POD, peroxidase; ALP, alkaline phosphatase; UE, urease; CL, cellulase. Data are shown as means ± standard deviations (n = 3). Different letters represent significant differences (p < 0.05).
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Zhang, P.; Guan, W.; Han, L.; Hu, X.; Xu, A.; Wang, H.; Wang, X.; Jiao, X. New Insights into the Formation Mechanism of Continuous Cropping Obstacles in Dioscorea opposita Thunb. cv. Tiegun Yam from Rhizosphere Metabolites and Microflora. Agronomy 2026, 16, 80. https://doi.org/10.3390/agronomy16010080

AMA Style

Zhang P, Guan W, Han L, Hu X, Xu A, Wang H, Wang X, Jiao X. New Insights into the Formation Mechanism of Continuous Cropping Obstacles in Dioscorea opposita Thunb. cv. Tiegun Yam from Rhizosphere Metabolites and Microflora. Agronomy. 2026; 16(1):80. https://doi.org/10.3390/agronomy16010080

Chicago/Turabian Style

Zhang, Pengfei, Wanghui Guan, Lili Han, Xiaoli Hu, Ailing Xu, Hui Wang, Xiaomin Wang, and Xiaoyan Jiao. 2026. "New Insights into the Formation Mechanism of Continuous Cropping Obstacles in Dioscorea opposita Thunb. cv. Tiegun Yam from Rhizosphere Metabolites and Microflora" Agronomy 16, no. 1: 80. https://doi.org/10.3390/agronomy16010080

APA Style

Zhang, P., Guan, W., Han, L., Hu, X., Xu, A., Wang, H., Wang, X., & Jiao, X. (2026). New Insights into the Formation Mechanism of Continuous Cropping Obstacles in Dioscorea opposita Thunb. cv. Tiegun Yam from Rhizosphere Metabolites and Microflora. Agronomy, 16(1), 80. https://doi.org/10.3390/agronomy16010080

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop