Abstract
Background:
The practice of tobacco monoculture usually leads to increased occurrence of soil-borne diseases and reduced yield and quality, posing a significant obstacle to sustainable farming. The soil microbiome is central to soil health, but the mechanisms by which crop rotation alleviates monoculture-related obstacles by reconstructing microbial communities and their interaction networks remain poorly understood.
Methods:
A field study was conducted in two regions to compare tobacco grown on soils after rotation (with wheat or rapeseed) and tobacco grown on soils with continuous tobacco cropping. A comprehensive analysis was performed on soil physicochemical properties, enzyme activities, microbial biomass, 16S rRNA bacterial and ITS fungal gene sequencing, microbial co-occurrence networks and phenotypic data, including tobacco yield, disease index and chemical quality of cured tobacco. A multi-omics coupling framework was constructed to integrate environmental variables, community structure and phenotype.
Results:
Crop rotation notably increased the content of available potassium in the soil and enhanced the activities of key enzymes involved in carbon and phosphorus cycling, such as β-glucosidase, sucrase and acid phosphatase. These changes led to improved nutrient availability. Across the two regions, crop rotation consistently restructured both bacterial and fungal communities, enriching beneficial taxa, including Chitinophagaceae, Rhodanobacteraceae, Xanthobacteraceae, Cercophora and Montagnula, while suppressing the proliferation of potential fungal pathogens. The microbial co-occurrence networks under rotation conditions were more intricate, cooperative and functionally integrated. Path analysis, incorporating composite variables, revealed a comprehensive causal pathway: crop rotation enhances the abundance of beneficial microbes and increases network complexity, which, in turn, improves the soil health indexes. This ultimately results in higher tobacco yield, reduced disease incidence and superior chemical and sensory quality of the cured tobacco.
1 Introduction
Tobacco (Nicotiana tabacum L.), a member of the Solanaceae family, is an economically significant crop with wide distribution and substantial commercial value (Lisuma et al., 2021). However, as urbanization accelerates and population growth persists, China’s arable land faces twin pressures—shrinking acreage and declining soil quality. Meanwhile, the increasing global demand for tobacco has intensified the challenges of maintaining effective crop rotation systems (Ding et al., 2024). The sustainability of modern agriculture is often threatened by long-term monoculture, which often leads to “soil sickness” or crop rotation obstacles (Liu et al., 2019; Yang et al., 2020). In this study, ‘continuous tobacco monoculture (continuous cropping)’ refers to tobacco planted on the same field for five consecutive years prior to sampling. Continuous tobacco monoculture affects soil health by impairing physicochemical properties (e.g., pH, electrical conductivity, organic matter, and available N–P–K), disrupting microbial community structure and promoting the accumulation of soil-borne pathogens, ultimately leading to severe declines in yield and quality (Huang et al., 2013; Chen et al., 2025).
The soil microbiome is widely recognized as the foundation of soil vitality and ecosystem functionality (Jansson et al., 2023; Ali et al., 2024). Increasing evidence indicates that the adverse effects of continuous cropping stem from ecological dysregulation within the soil microbial community (Chen et al., 2022; Li et al., 2024b). Such disruption is typically reflected in reduced microbial diversity, diminished populations of beneficial microorganisms and the proliferation of pathogenic fungi (Wu et al., 2022; Chen et al., 2025). Moreover, continuous cropping often destabilizes microbial interaction networks (i.e., co-occurrence networks inferred from abundance correlations, reflecting potential ecological associations and community organization), leading to simplified, fragmented ecological structures (Wang et al., 2022). In sharp contrast, crop rotation—a time-tested and sustainable agricultural strategy—has been shown to alleviate soil-borne diseases by reversing the degenerative processes that drive them (Zou et al., 2024; Wen and Zang, 2025). By introducing diverse plant species, rotation systems modify the rhizosphere microenvironment, alter root exudate composition and vary organic matter inputs, thereby reshaping the composition and functional dynamics of the soil microbiome (Zhang et al., 2021a; Tian et al., 2025).
Recent studies have begun to clarify the microbial mechanisms underlying the benefits of crop rotation. Evidence shows that rotation systems selectively recruit beneficial bacterial communities that promote plant growth and suppress soil-borne pathogens (Sun et al., 2022; Wu et al., 2025). In parallel, crop rotation tends to enrich saprophytic and symbiotic fungi, which facilitate nutrient cycling and strengthen plant health (Alsunuse et al., 2024). Under rotation regimes, microbial interactions reorganize into more complex and modular co-occurrence networks, enhancing the functional stability and resilience of soil ecosystems (Lurgi et al., 2019). This microbial reconfiguration is closely associated with improved soil multifunctionality—the coordinated performance of ecosystem processes such as nutrient turnover, organic matter decomposition and disease suppression (Zheng et al., 2019; Xiao et al., 2025).
Despite these advances, several key gaps remain. Most existing studies have examined bacterial or fungal communities in isolation, overlooking the integrated dynamics of these two microbial domains that jointly sustain soil health (Wagg et al., 2019; Lan et al., 2022). Furthermore, while correlations between management practices and microbial attributes are well documented, quantitative assessments of how network properties, soil multifunctionality and crop performance are linked remain scarce (Singh et al., 2025). A critical next step is to bridge these gaps by establishing causal relationships, moving beyond descriptive correlations. This advancement is essential for developing a mechanistic understanding of how crop rotation enhances agroecosystem productivity and stability.
Accordingly, this study tests the hypothesis that crop rotation alleviates the constraints of tobacco monoculture through multiple, interrelated mechanisms. Specifically, we propose that rotation enriches beneficial microbial taxa and optimizes microbial network architecture, thereby enhancing soil multifunctionality and ultimately improving crop health, yield and quality. To evaluate this hypothesis, a comparative field experiment was conducted in two distinct regions, contrasting tobacco rotation systems with continuous monoculture. An integrated analytical framework was employed, encompassing assessments of soil physicochemical properties, enzyme activities, microbial biomass and high-throughput sequencing of bacterial and fungal communities. Microbial co-occurrence network analyses and comprehensive evaluations of tobacco agronomic performance and chemical composition further complemented these data. The objectives of this work were threefold: (1) to quantify the effects of crop rotation on soil attributes, microbial community structure and network organization; (2) to identify key microbial taxa and functional modules characteristic of rotation systems; and (3) to validate the hypothesized causal relationships using composite variable modeling. Collectively, this study provides mechanistic insights into how crop rotation restructures soil microbiomes to sustain ecosystem function, thereby offering a robust scientific foundation for advancing sustainable tobacco cultivation.
2 Materials and methods
2.1 Field experiment design
Field experiments were conducted in Linyi City, Shandong Province, China, at two sites located in Taibao Town, Pingyi County (35.57°N, 117.71°E) and Jinling Town, Lanling County (34.86°N, 118.04°E). The soils at both sites were classified as brown soils with relatively uniform fertility. Four cropping-system treatments were evaluated: Pingyi wheat→tobacco rotation (PYR1), Pingyi continuous tobacco cropping (PYC2), Lanling rapeseed→tobacco rotation (LLR1), and Lanling continuous tobacco cropping (LLC2). Continuous-cropping plots had been under tobacco monoculture for five consecutive years prior to sampling. Rotation plots implemented one rotation cycle, in which tobacco was planted after the preceding crop (i.e., wheat/rapeseed → tobacco). Both sites were sampled in the same year. The tobacco cultivar ‘Zhongchuan 208’ and the rotation crop cultivars ‘Linmai 9’ (wheat) and ‘Huangjin 100’ (rapeseed) were selected because they are widely planted and agronomically representative cultivars in the two regions. Except for the cropping sequence, field management practices (tillage, fertilization, irrigation, and pest\disease/weed control) were kept consistent across treatments following local standards for high-quality tobacco production.
At the tobacco maturity stage, soil samples were collected from the 0–20 cm plough layer. Each treatment comprised five independent replicate plots. Within each plot, 10 soil cores were randomly collected and composited to obtain one plot-level composite sample. All five composite samples per treatment were used for soil physicochemical properties, microbial biomass, and enzyme assays (n = 5). For amplicon sequencing, three plot-level composite samples were randomly selected from the five replicate plots in each treatment (n = 3). Samples were transported to the laboratory on ice and divided into two subsamples: one stored at −80°C for molecular analyses and the other air-dried for soil physicochemical and enzyme measurements.
2.2 Determination of tobacco disease index, quality and yield
At tobacco maturity, a standardized grading system was used to assess the extent of damage to tobacco roots caused by soil-borne diseases (Sparks et al., 1996). The specific grading criteria are as follows: 0: no lesions on the root system; 1: lesion area less than one-third of the total root area; 2: lesion area between one-third and two-thirds of the total root area; 3: lesion area greater than two-thirds of the total root area. Upon reaching maturity, tobacco leaves were promptly harvested and cured. Cured leaves from the middle stalk position (1.5 kg per replicate) were collected and sent to the Tobacco Industry Quality Supervision and Testing Center of the Ministry of Agriculture and Rural Affairs for standardized analysis. The analyzed chemical components included reducing sugars, total sugars, total plant alkaloids (nicotine), total nitrogen, potassium and chloride content. Concurrently, certified experts performed sensory evaluations of the cured leaves, assessing aroma quality, aroma quantity, aftertaste, off-flavors and irritancy. Finally, the total weight of tobacco leaves from each plot was recorded and converted to yield per unit area (kg·hm-2).
2.3 Determination of soil physicochemical properties
Soil physicochemical properties, microbial biomass and enzyme activities were analyzed by Qinhuangdao Ruike Biotechnology Co., Ltd., following standard methodologies. Soil pH was measured potentiometrically using a glass electrode at a soil-to-water ratio of 1:2.5 (w/v). Soil organic matter was determined by the potassium dichromate external heating oxidation-titration method (Walkley and Black, 1934). Alkaline hydrolyzable nitrogen was measured using the alkaline hydrolysis diffusion method. Available phosphorus was extracted with 0.5 mol/L sodium bicarbonate (NaHCO3, pH 8.5) and measured using the molybdenum-antimony colorimetric method (Olsen et al., 1954; Murphy and Riley, 1962). Available potassium was extracted using 1 mol/L ammonium acetate (NH4OAc, pH 7.0) and measured by flame photometry. Microbial biomass carbon (MBC), nitrogen (MBN), and phosphorus (MBP) were determined using the chloroform fumigation–extraction method. Briefly, paired fumigated and non-fumigated subsamples were extracted with 0.5 M K2SO4 for MBC and MBN and with 0.5 M NaHCO3 for MBP. Microbial biomass was calculated as the difference between fumigated and non-fumigated extracts divided by the corresponding k factors (kEC = 0.45, kEN = 0.54, kEP = 0.4) (Brookes et al., 1982; Vance et al., 1987). Urease activity was measured by incubating soil with urea at 37°C, and the released NH4+-N was determined colorimetrically (Kandeler and Gerber, 1988). Sucrase activity was measured by incubating the soil with sucrose as the substrate in a phosphate buffer system at pH 5.5 and 37°C (Kotroczó et al., 2014). The produced reducing sugars were quantified using the 3,5-dinitrosalicylic acid colorimetric method.
2.4 High-throughput sequencing
In this study, soil genomic DNA was extracted from 0.5 g of frozen soil samples using the FastDNA® SPIN Kit for Soil (MP Biomedicals, USA) according to the manufacturer’s instructions. The concentration and purity of the extracted DNA were measured using a NanoDrop spectrophotometer (Thermo Fisher Scientific, USA). For the bacterial 16S rRNA gene’s V3-V4 hypervariable region, PCR amplification was performed using the universal primers 338F (5’-ACTCCTACGGGAGGCAGCAG-3’) and 806R (5’-GGACTACHVGGGTWTCTAAT-3’). For the fungal ITS gene’s ITS1 region, amplification was carried out using the primers ITS1F (5’-CTTGGTCATTTAGAGGAAGTAA-3’) and ITS2 (5’-GCTGCGTTCTTCATCGATGC-3’). All PCR reactions were conducted in triplicate. The amplicon libraries were constructed by Shanghai Meiji Biotechnology Co., Ltd. and sequenced on an Illumina MiSeq platform (16S: 2 × 250 bp; ITS: 2 × 300 bp). The obtained raw data were processed on the Shanghai Meiji Biotechnology Cloud Platform (www.majorbio.com). The process is briefly as follows: Raw paired-end sequences were merged using FLASH (v1.2.11), and quality control was performed using fastp (v0.19.6) to obtain high-quality clean tags. Chimeras were detected and removed using the UCHIME algorithm, yielding the final valid tags. In the QIIME2 (v2020.11) analysis pipeline (Bolyen et al., 2019), these tags were clustered into amplicon sequence variants (ASVs) using the DADA2 denoising method, which provides single-nucleotide resolution (Callahan et al., 2016).
Subsequently, bacterial and fungal ASVs were annotated using the SILVA (v138) and UNITE (v8.0) databases, with a confidence threshold of 70%. Subsequent analyses encompassed measures of α-diversity (ACE, Chao1 and Shannon indices) and β-diversity, evaluated through principal coordinates analysis (PCoA) based on Bray–Curtis distance. All computations were carried out using QIIME2 and the vegan package in R (v4.0.3). To test for significant differences in microbial community composition among treatments, an analysis of similarity (ANOSIM) was conducted with 999 permutations.
2.5 Bioinformatics and statistical analysis
All statistical analyses were conducted using R software (version 4.0.3). One-way analysis of variance (ANOVA) was employed to assess the differences in soil physicochemical properties, enzyme activities, microbial biomass, alpha diversity indices, and tobacco agronomic and quality traits across the four treatments. Where the ANOVA indicated significant overall effects (p < 0.05), post-hoc comparisons were performed using Tukey’s Honestly Significant Difference (HSD) test. Data are reported as the mean ± standard deviation (SD) (Okoye and Hosseini, 2024). In the tables, significant differences among treatments at the p < 0.05 level are indicated by different lowercase letters. Asterisks denote the significance levels of pairwise comparisons in the figures: *p < 0.05, **p < 0.01, ***p < 0.001; ‘ns’ indicates no significant difference.
For beta diversity analysis, a permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis distance was performed, with 999 permutations to test for differences in community structure between treatments, using the adonis2 function. Linear discriminant analysis effect size (LEfSe) was used to identify microbial taxa that significantly differed between the rotation group (PYR1-LLR1) and the monoculture group (PYC2-LLC2). This analysis was performed on the Meiji BioCloud platform, with a log-transformed LDA score threshold of 2.5 to identify differential features (Lu et al., 2016).
A co-occurrence network of key microbial genera was constructed based on Spearman correlation coefficients. Only strong correlations with |ρ| > 0.6 and p < 0.05 were retained. Network analysis was also performed on the Meiji BioCloud platform, and its topological properties (node count, connectivity and average degree) were calculated.
To test the causal pathway hypothesis of “crop rotation→enrichment of beneficial microbes/network structure optimization→enhancement of soil function→improvement of crop yield and quality,” a Path Analysis of Composite Variables approach based on composite variables was used (Mcdonald, 1996). Multiple observational indicators were aggregated into four core composite variables that represent the key constructs of the theoretical model. Soil health composite index: calculated as the standardized mean of sucrase activity, acid phosphatase activity, β-glucosidase, microbial biomass carbon and microbial biomass phosphorus (Lehmann et al., 2020); microbial network complexity: integrated modularity index of bacterial and fungal co-occurrence networks; beneficial microbial abundance: the sum of the relative abundances of beneficial bacterial genera selected by LEfSe analysis; crop performance: a standardized composite of yield and disease index (with the disease index as a negative value). All composite variables were standardized using Z-scores to ensure dimensional consistency. Model parameters were fitted using maximum likelihood estimation, and the standard errors and significance levels of path coefficients were calculated through the bootstrap method (1000 replications). Model fit was assessed using the coefficient of determination (R²) and the significance of path coefficients.
3 Results and discussion
3.1 A systematic restoration of soil health multidimensional attributes by crop rotation
Comprehensive analysis of soil physicochemical properties and microbial biomass revealed that the core effect of crop rotation lies in enhancing the operational efficiency of soil health, rather than merely increasing nutrient pools (Figures 1A, B). Although soil pH and total organic matter did not differ significantly between rotation and continuous cropping, the available potassium content was significantly increased by 20% and 14.18%, respectively. Available potassium is a key factor influencing tobacco leaf quality. Conversely, electrical conductivity (EC) was significantly higher in the continuous cropping treatments (Figure 1A), with increases of 1.95% for PYC2 versus PYR1 and 11.85% for LLC2 versus LLR1, a typical indicator of soil degradation. This phenomenon reflects the accumulation of root exudates and ion imbalance in monoculture systems (Huang et al., 2013). More profound shifts were observed in microbial physiological status (Six et al., 2004). In the Lanling region (Figure 1B), microbial biomass phosphorus (MBP) was significantly greater in the rotation treatment (LLR1) than in continuous cropping (LLC2) (2.7-fold). In contrast, MBP did not differ significantly between rotation and continuous cropping in Pingyi, indicating a site-dependent response that may be influenced by baseline soil P availability and local environmental conditions. Overall, P-related indicators showed site-dependent patterns. In Lanling, rotation increased MBP and was accompanied by higher P-cycling enzyme activities, suggesting a more efficient microbially mediated P-cycling process. In Pingyi, AP was higher in the continuous-cropping treatment (PYC2), which may reflect compensatory P-acquisition processes under monoculture stress rather than improved soil functional efficiency. In contrast, the continuous cropping treatment PYC2 exhibited higher microbial biomass carbon (MBC) concurrently with lower enzyme activities, suggesting a potentially “idle boom” within the microbial community. This pattern suggests that under stress conditions, microbial communities allocate more energy to maintenance than to functional processes, indicating a state of functional impairment.
Figure 1
Soil enzyme activity profiles provided direct functional evidence supporting these inferences (Figure 1C). In rotation treatments, activities of enzymes involved in carbon and phosphorus cycling (βG, SA, ACP) showed synergistic enhancement, indicating a more active and efficient microbial functional engine (Tiemann et al., 2015). Meanwhile, the significantly higher urease (UA) activity in the LLC2 continuous cropping system indicated a potential functional “decoupling” among elemental biogeochemical cycling pathways, which may reduce nutrient use efficiency.
Analysis of soil aggregates further extended the benefits of crop rotation to the physical structure dimension (Supplementary Figure 1). The continuous cropping system LLC2 exhibited the highest mass fraction of large-sized aggregates (>5 mm and 3–5 mm), whereas the rotation system LLR1 was dominated by medium-sized aggregates (0.5–1 mm). This finding challenges the conventional wisdom that “a higher proportion of large aggregates is always better.” The observed enrichment of large aggregates under continuous cropping may not be an optimal indicator of soil health, but rather a result of over-proliferation of specific fungi whose hyphae can bind particles into large aggregates. However, these structures often exhibit poor stability and may correlate negatively with plant health. Medium-sized aggregates in rotation systems are typically formed through synergistic bacterial-fungal interactions and microbial secretions, representing a more stable and healthy soil structure (Six et al., 2004; Rillig and Mummey, 2006).
Principal Component Analysis (PCA) provided an integrated perspective that validated these findings (Figure 1D), with the first two principal components collectively explaining 59.65% of the total variance. The analysis clearly revealed that rotation systems, particularly PYR1, were closely associated with the “high-functional-efficiency” core module defined by the positive direction of PC1, which comprised key functional indicators such as MBP, SA, ACP and βG. In contrast, continuous cropping systems exhibited two distinct dysfunctional pathways: PYC2 was linked with EC and MBC, representing a “high-load, low-efficiency” state. At the same time, LLC2 was associated with UA, indicating a “disrupted nitrogen cycling” state. This overall pattern powerfully demonstrates that the fundamental role of crop rotation is to shift the soil ecosystem from a dysfunctional state toward a healthy state characterized by coordinated functionality and efficient nutrient cycling (Wang et al., 2023).
3.2 Reconstruction of soil microbial community diversity and structure by crop rotation
Given that microbial processes fundamentally govern soil biochemical functionality, it is essential to explore how crop rotation influences the diversity and structure of soil microbial communities. Accordingly, we examined bacterial and fungal assemblages to reveal the compositional and functional rearrangements underpinning soil health restoration under different cropping regimes. Bacterial communities demonstrated pronounced responsiveness to crop rotation. Alpha diversity analysis revealed that rotation significantly enhanced bacterial richness, with both PYR1 and LLR1 showing substantially higher ACE and Chao1 indices compared to their corresponding continuous cropping treatments (Figure 2A). Because the Shannon index reflects both species richness and community evenness, our results indicate that crop rotation mainly increased bacterial richness while having limited effects on evenness, resulting in no significant differences in Shannon values among treatments. Principal Coordinates Analysis (PCoA) showed that PC1 and PC2 explained 26.97% and 15.87% of the total variance in bacterial community composition, respectively. A clear separation between the rotation (PYR1 and LLR1) and continuous cropping treatments (PYC2 and LLC2) was observed along the PC1 axis (Figure 2B), indicating that the cropping system was the dominant factor shaping bacterial community assembly. For fungi, alpha diversity indices showed no significant differences among treatments (Figure 2C), suggesting that crop rotation did not markedly alter overall fungal diversity levels. However, PCoA results revealed a distinct segregation of fungal community structures between rotation and continuous cropping systems (Figure 2D). The separation along the PC2 axis demonstrated that fungal community composition was not only regulated by the cropping pattern but also strongly influenced by site-specific environmental factors. The more pronounced response of bacteria to rotation suggests their role as primary initiators of the improvements in soil nutrient cycling and enzyme activities documented here.
Figure 2
To visualize the distribution patterns of core microbial taxa across different treatments, Circos plots were constructed for both bacterial and fungal communities (Supplementary Figure 2). The observed increase in bacterial richness is consistent with ecological theory, which posits that diversified plant inputs in rotation systems provide broader ecological niches and carbon sources, supporting the coexistence of more diverse bacterial taxa (Tiemann et al., 2015; Venter et al., 2016). The pronounced restructuring of bacterial communities thus establishes a direct link between cropping systems and microbiome assembly processes. These reconfigured bacterial assemblages are likely a primary driver of the enhanced soil multifunctionality and nutrient cycling observed in rotation systems (Delgado-Baquerizo et al., 2016).
By comparison, the comparatively stable fungal diversity and site-dependent structural shifts reflect the distinct ecological strategies of fungi. Fungal communities form extensive mycelial networks and exhibit substrate specialization, rendering them more resilient to short-term management effects but more susceptible to persistent local conditions and crop legacy effects (Tedersoo et al., 2020). Although fungal richness remained stable, the compositional turnovers across treatments indicate taxon replacement within functional guilds, which may alter the balance among saprotrophic, pathogenic and symbiotic groups (Sun et al., 2024).
3.3 Soil environmental factors differentially drive bacterial and fungal community assembly
To identify the environmental drivers underlying microbial community variation, redundancy analysis (RDA) was performed at the genus level. For bacterial communities (Figure 3A), the first two RDA axes collectively explained 59.59% of the fitted variance (RDA1 = 40.51%; RDA2 = 19.08%), demonstrating a strong coupling between soil variables and bacterial community structure. Clear separation of the different treatments was observed along the RDA1 axis. The LLR1 rotation community was positioned at the positive end of RDA1, mainly associated with AN and microbial biomass nutrients (MBN, MBP, MBC), suggesting a nutrient enrichment pathway. The PYR1 rotation community was located at the negative end of RDA1, strongly correlated with key enzyme activities (UA, βG, ACP, SA) as well as AK and EC, pointing to an enhanced nutrient mineralization pathway. Continuous cropping samples (PYC2, LLC2) showed greater dispersion, indicating lower consistency in community assembly under monoculture. The strong and consistent clustering of bacterial communities along the major environmental gradient underscores that crop rotation fosters distinct, function-oriented bacterial consortia through modification of the soil habitat (Fierer, 2017).
Figure 3
Distinct environmental drivers were identified for fungal communities (Figure 3B). The lower model fit (RDA1 = 32.27%; RDA2 = 22.12%) and more scattered sample distribution indicate that fungal community assembly is more complex and exhibits stronger site specificity, aligning with their recognized sensitivity to local conditions and host plants. UA activity was identified as a key driver for the LLC2 continuous cropping community, potentially indicating that nitrogen cycling stress under monoculture favors specific fungal assemblages. Overall, fungal communities were influenced by a different suite of factors, with EC, MBC and AP contributing most strongly to the negative end of RDA1. These fundamental differences highlight distinct spatiotemporal patterns in niche occupation and contrasting responses to agricultural management between bacterial and fungal communities (Bahram et al., 2018).
3.4 Crop rotation enriches specific beneficial microorganisms and inhibits potential pathogens
Through LEfSe analysis, specific microbial taxa responsive to the cropping system were identified. In bacterial communities (Figure 4A), the rotation treatment (PYR1_LLR1) was characterized by significant enrichment of families including Chitinophagaceae, Rhodanobacteraceae and Xanthobacteraceae, as well as the genus Sphingomonas and the order SC-I-84. Members of Chitinophagaceae are known to degrade complex polymers (e.g., chitin and cellulose) and contribute to organic matter turnover, whereas Sphingomonas and Xanthobacteraceae are frequently linked to the transformation of recalcitrant organic compounds and plant-beneficial functions (e.g., plant growth promotion and bioremediation) in soil ecosystems (Fierer, 2017; Gatheru Waigi et al., 2017). Within fungal communities (Figure 4B), the rotation system enriched saprotrophic taxa such as Cercophora and Montagnula, which are involved in lignin and cellulose decomposition. The concomitant enrichment of taxa with biocontrol potential, including Curvularia (a genus containing known endophytes and biocontrol agents) and Tubeufiaceae (a group of parasitic fungi), suggests the establishment of a soil environment with a more substantial potential to suppress fungal pathogens (Ge et al., 2008). Together, these microbial taxa contribute to a healthier, more resilient and functionally optimized soil microbiome, forming the foundation for enhanced crop productivity and quality (Wang et al., 2025).
Figure 4
Conversely, continuous cropping (PYC2 and LLC2) enriched a different set of taxa, including the bacteria Gemmatimonas, Streptomyces, Niallia and Paenisporosarcina, as well as fungal families such as Mortierellaceae_gen_Incertae_sedis, Chaetomiaceae and Microascaceae. While some members of Streptomyces are considered beneficial, the genus also includes recognized plant pathogens. Furthermore, the co-enrichment of potentially stress-adapted groups such as TM7a (Saccharimonadaceae) may indicate a community shift toward higher soil stress conditions (Zong et al., 2024). Notably, the enrichment of plant-pathogenic genera such as Botryotrichum and Gibellulopsis indicates soil environment degradation and a disrupted microbiome balance, increasing the system’s susceptibility to crop disease (Yuan et al., 2020).
The identification of these biomarker taxa directly links agricultural management practices to the assembly of a functionally specific microbiome. Rotation enriched specific microbes with defined roles in decomposing complex organics and promoting plant health. Their enrichment aligned with increased soil enzyme activities and nutrient availability (Figure 1), providing a direct microbiological mechanism for the “soil function enhancement” pathway.
3.5 Crop rotation synergistically improves tobacco chemical quality, sensory quality, yield and disease resistance
The enhanced performance of the tobacco crop ultimately confirmed the ameliorative effects of crop rotation on soil and microbiome properties. Analysis of leaf chemical composition, sensory quality, yield and disease incidence revealed significant differences among management regimes. Under rotation, the tobacco chemical composition was significantly optimized (Table 1). The PYR1 and LLR1 rotation treatments exhibited significantly higher reducing sugar and total sugar levels than their continuous cropping counterparts; these sugars are critical for producing a mild, smooth smoke. In contrast, continuous cropping, especially PYC2, resulted in significant accumulation of total plant alkaloids and total nitrogen, compounds associated with increased smoke irritancy and strength (Li et al., 2024a). A plausible explanation is that rotation improved nutrient balance (notably K availability) and reduced soil-borne disease pressure, which can favor carbon assimilation and sugar accumulation, whereas monoculture-associated stress and/or relatively higher N availability may stimulate nicotine (alkaloid) synthesis as a defense-related metabolic shift. Consequently, the key quality parameter—the sugar-to-alkaloid ratio—was considerably higher in LLR1 (13.18) and PYR1 (12.96) than in PYC2 (11.02), placing rotation-treated tobacco within the optimal range for premium leaf quality (10–15) for high-quality tobacco (Pinfield, 2001). Additionally, the significantly higher chlorine content observed in PYC2 is known to negatively combustion efficiency.
Table 1
| Treatment | Reduced sugar (%) | Total sugar (%) | Alkaloids (%) | Total nitrogen (%) | Potassium (%) | Cl (%) | Sugar/total sugar | Sugar/ Alkaloids |
|---|---|---|---|---|---|---|---|---|
| PYR1 | 21.58 ± 0.57 a | 27.15 ± 0.69 a | 2.20 ± 0.08 b | 1.62 ± 0.05 b | 1.34 ± 0.04 a | 1.08 ± 0.05 b | 0.78 ± 0.03 a | 12.96 ± 0.61 a |
| PYC2 | 20.02 ± 0.64 b | 25.48 ± 0.73 b | 2.33 ± 0.09 a | 1.70 ± 0.05 a | 1.37 ± 0.05 a | 1.18 ± 0.05 a | 0.80 ± 0.04 a | 11.02 ± 0.59 c |
| LLR1 | 21.27 ± 0.60 a | 26.97 ± 0.71 a | 2.17 ± 0.07 c | 1.58 ± 0.05 c | 1.32 ± 0.04 a | 1.05 ± 0.04 b | 0.77 ± 0.03 a | 13.18 ± 0.63 a |
| LLC2 | 20.74 ± 0.56 b | 25.41 ± 0.76 b | 2.23 ± 0.08 b | 1.64 ± 0.05 b | 1.35 ± 0.05 a | 1.10 ± 0.05 b | 0.79 ± 0.04 a | 12.44 ± 0.65 b |
Chemical composition of flue-cured tobacco under different cultivation systems.
Data depict means ± SD of three biological replicates. Significant differences between treatments (p < 0.05) are illustrated by lowercase letters a-d.
Rotation systems demonstrated significant commercial advantages in sensory quality, disease resistance and yield (Table 2). Tobacco leaves from rotation plots received significantly higher scores for aroma quality, aroma quantity and aftertaste, while exhibiting significantly lower scores for off-flavors and irritancy. This profile resulted in a cleaner smoking experience with enhanced aromatic richness and overall comfort. Notably, rotation nearly completely suppressed soil-borne diseases, with PYR1 showing a disease index of 0 and LLR1 maintaining a very low value (0.67). This contrasted sharply with the severe disease incidence (14.22) observed in the LLC2 continuous cropping system. Enhanced plant health directly translated into significant yield gains, as evidenced by higher yields in LLR1 and PYR1 compared with continuous cropping treatments.
Table 2
| Treatment | Aroma quality (15) | Aroma quantity (20) | Aftertaste (25) | Off-Flavor (18) | Irritation (12) | Combustibility (5) | Ash color (5) | Total score (100) | Disease index (DI) | Yield (kg·hm-2) |
|---|---|---|---|---|---|---|---|---|---|---|
| PYR1 | 11.58 ± 0.23 a | 15.18 ± 0.20 a | 18.84 ± 0.28 a | 13.12 ± 0.19 b | 8.48 ± 0.17 b | 3.00 ± 0.00 a | 3.00 ± 0.00 a | 74.20 ± 0.68 a | 0 ± 0.00 d | 2209.6 ± 64.7 a |
| PYC2 | 10.95 ± 0.25 b | 14.98 ± 0.18 b | 18.35 ± 0.23 b | 13.46 ± 0.21 a | 8.84 ± 0.19 a | 3.00 ± 0.00 a | 3.00 ± 0.00 a | 72.58 ± 0.74 b | 7.56 ± 0.09 b | 1932.7 ± 70.1 b |
| LLR1 | 11.50 ± 0.22 a | 15.24 ± 0.21 a | 18.90 ± 0.27 a | 13.05 ± 0.18 b | 8.40 ± 0.16 b | 3.00 ± 0.00 a | 3.00 ± 0.00 a | 74.29 ± 0.66 a | 0.67 ± 0.03 c | 2162.3 ± 58.9 a |
| LLC2 | 11.10 ± 0.24 b | 15.06 ± 0.09 b | 18.70 ± 0.10 a | 13.20 ± 0.10 b | 8.52 ± 0.18 b | 3.00 ± 0.00 a | 3.00 ± 0.00 a | 73.76 ± 0.70 a | 14.22 ± 0.17 a | 1894.8 ± 66.4 b |
Effects of different planting patterns on the sensory quality and yield of flue-cured tobacco.
This effective disease suppression reflects the ecological function of the restructured microbiome, potentially mediated by a “cry for help” mechanism and direct pathogen inhibition (Yuan et al., 2018). The enrichment of beneficial rhizosphere microbes, such as Sphingomonas and Curvularia, may prime plant systemic resistance and directly suppress soil-borne pathogens, thereby lowering the disease index (Xie et al., 2022). The synergy between improved plant nutrition and enhanced disease resistance ultimately led to yield increases. These results demonstrate that shifting from continuous cropping to rotation enables simultaneous improvements in crop quality, yield and system sustainability.
3.6 Network analysis reveals the microbial mechanisms underlying crop rotation effects
Co-occurrence network analysis of key microbial taxa revealed fundamentally distinct ecological interaction patterns between rotation and continuous cropping systems (Barberán et al., 2012). For bacterial networks, rotation supported a highly complex and cohesive network topology (Supplementary Table 1). Within this network, the disease index functioned as a central hub, displaying the highest degree and betweenness centrality (Figure 5A). Key soil enzymes (SA, βG) also occupied central positions, providing a direct link between microbial activity and soil function. The network was characterized by keystone taxa with multifunctional roles: Flaviaestuariibacter abundance was positively correlated with βG, SA and aroma quality, whereas Rhodanobacter correlated negatively with the disease index and positively with urease activity, demonstrating its combined potential for pathogen suppression and nutrient cycling. Furthermore, a distinct “disease suppression module” was identified, consisting of taxa such as Tumebacillus and Gaiella (Schlatter et al., 2017). The bacterial network under continuous cropping was significantly smaller in scale, less complex and indicative of functional impairment (Figure 5B). The network core was characterized by functionally inhibitory taxa such as Dongia, which demonstrated negative correlations with MBC and SA activity and composite scores, but positive correlations with disease indices. Notably, crop quality indicators (particularly aroma and yield) were either marginalized or absent from the network topology, reflecting a disrupted ecological linkage between bacterial community structure and crop performance.
Figure 5
A parallel pattern was observed in fungal co-occurrence networks. The fungal network was organized around the disease index as its primary hub (Figure 5C). The central taxon Furcasterigmium exhibited positive associations with βG, SA and aroma quality. A clearly defined disease-suppression module, comprising genera such as Curvularia, Poaceascoma and Darksidea, further strengthened this functional role. Conversely, the continuous cropping fungal network (Figure 5D), although comprising more nodes, exhibited clear functional disorder. It lacked a coherent beneficial core and incorporated taxa with conflicting ecological roles, such as Kernia (positively correlated with the disease index but negatively with soil health) and Scedosporium (a known pathogen unlinked to soil fertility functions). Connections between soil functional attributes and crop quality indicators were either fragmented or hostile.
The rotation system exemplifies a “functionally integrated” model. Microbial communities are assembled around the dual objectives of plant health and efficient nutrient cycling. This assembly forms multifunctional keystone taxa and cooperative modules, which underpin a resilient and productive soil ecosystem (Wagg et al., 2019). This integrated architecture directly facilitates the enhancement of soil functions and the optimization of crop yield and quality (Yang et al., 2023). Conversely, continuous cropping exemplifies a “functional collapse” model, where a simplified bacterial network and a disordered, antagonistic fungal network collectively reflect community-level dysbiosis (Sun et al., 2023). The dominance of detrimental taxa and the marginalization of crop quality indicators visually demonstrate the breakdown of critical agroecosystem linkages. The dysfunction at the level of interaction networks provides a novel mechanistic framework for understanding continuous cropping obstacles.
3.7 Causal paths proposed based on path analysis validation using composite variables
Path analysis with key composite variables was employed to statistically evaluate the proposed conceptual framework—”crop rotation → restructured microbiome → soil health → crop performance”—despite the limited sample size (Figure 6). This approach enables effective testing of causal relationships by integrating multiple observed indicators into core constructs (Zhang et al., 2021b). The path model strongly supported our hypothesis. The cropping system (rotation) demonstrated significant direct positive effects on both beneficial microbial abundance (β = 0.82, p < 0.001) and microbial network complexity (β = 0.75, p < 0.01). These microbial properties subsequently served as key drivers, collectively explaining a large proportion of the variance in the soil health index (R² = 0.68). Ultimately, the enhanced soil health index, together with a direct positive path from beneficial microbial abundance, significantly improved crop performance (R² = 0.74). All specified paths were statistically significant (p < 0.05), confirming the robustness of the proposed causal cascade.
Figure 6
This analysis provides solid statistical evidence for the overall narrative. The strong direct effects of rotation on both beneficial taxon abundance and network complexity confirm its role as a primary driver steering the soil microbiome toward a functional state (Banerjee et al., 2019). The model clarifies that these microbial changes are crucial mechanisms through which rotation improves soil conditions. The high explained variance (68%) of the soil health index emphasizes that a complex network enriched with beneficial microbes is fundamental to soil multifunctionality and health (Delgado-Baquerizo et al., 2016).
Ultimately, the analysis delineates a transparent causal chain. Crop rotation cultivates a protective and efficient soil microbiome, thereby enhancing the soil’s functional capacity and directly translating to superior crop performance. This quantified pathway delineates a clear causal chain: rotation → (beneficial microbes & network complexity) → soil health → crop performance. It thus provides a mechanistic and predictive understanding of how rotation alleviates continuous cropping obstacles, establishing a causal framework that explains the agronomic benefits and underpins sustainable management practices (Michael and Li, 2019).
4 Conclusions
This study compared tobacco rotation systems (wheat/rapeseed–tobacco, region-specific) with continuous tobacco cropping across two regions and showed that rotation generally improves soil health, although some responses are site dependent. Rotation alleviated continuous-cropping stress signals (e.g., lower EC) and enhanced nutrient-supply capacity (e.g., higher AK), while promoting microbial functional potential via shifts in microbial biomass and C- and P-cycling enzyme activities. However, P-related indicators (MBP and AP) responded inconsistently between regions, highlighting strong legacy and site effects.
Multi-omics evidence indicates that rotation reshapes the soil microbiome by enriching beneficial taxa (e.g., Chitinophagaceae and Xanthobacteraceae) and saprophytic fungi (e.g., Cercophora and Montagnula), while suppressing potential pathogens (e.g., Botryotrichum and Gibellulopsis). Network and path analyses support a mechanistic pathway in which rotation increases beneficial microbes and microbial network complexity, thereby improving soil health and crop performance.
Overall, these findings support crop rotation as a practical strategy to restore soil functioning in tobacco systems. Management should prioritize rotation-driven microbiome assembly and apply site-specific P management based on local AP/MBP responses rather than relying solely on chemical inputs (Bender et al., 2016).
Statements
Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.ncbi.nlm.nih.gov/, PRJNA1358713.
Author contributions
YB: Software, Visualization, Writing – original draft. XC: Writing – review & editing, Project administration. LT: Writing – review & editing, Supervision. MX: Writing – review & editing, Data curation. PG: Investigation, Writing – review & editing. LW: Investigation, Writing – review & editing. QG: Writing – review & editing, Project administration. XT: Project administration, Writing – review & editing. RX: Writing – review & editing, Validation. ZD: Formal Analysis, Writing – review & editing. CF: Writing – review & editing. YL: Funding acquisition, Project administration, Writing – review & editing. JY: Resources, Writing – review & editing, Supervision.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This research was supported by the Key Project of Shandong Provincial Branch of China National Tobacco Corporation (202302) and the Major Science and Technology Project of China National Tobacco Corporation (110202201029).
Conflict of interest
XC, LT, PG, LW, QG, XlT, RX, and ZD were employed by the Shandong Linyi Tobacco Company.
CF was employed by the Shandong Junan Tobacco Monopoly Bureau.
The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fagro.2026.1753158/full#supplementary-material
References
1
AliQ.AliM.JingH.HussainA.ManghwarH.AliM.et al. (2024). Power of plant microbiome: A sustainable approach for agricultural resilience. Plant Stress14, 100681. doi: 10.1016/j.stress.2024.100681
2
AlsunuseB. T. B.Al-AwwalN.BudhathokiS.ThapaR. B.BaidooM.ZaidF.et al. (2024). Mycorrhizal fungal inoculum potential in crop rotation soil with different levels of irrigation. J. Agric. Sci.16, 34. doi: 10.5539/jas.v16n11p34
3
BahramM.HildebrandF.ForslundS. K.AndersonJ. L.SoudzilovskaiaN. A.BodegomP. M.et al. (2018). Structure and function of the global topsoil microbiome. Nature560, 233–237. doi: 10.1038/s41586-018-0386-6
4
BanerjeeS.WalderF.BüchiL.MeyerM.HeldA. Y.GattingerA.et al. (2019). Agricultural intensification reduces microbial network complexity and the abundance of keystone taxa in roots. ISME. J.13, 1722–1736. doi: 10.1038/s41396-019-0383-2
5
BarberánA.BatesS. T.CasamayorE. O.FiererN. (2012). Using network analysis to explore co-occurrence patterns in soil microbial communities. ISME. J.6, 343–351. doi: 10.1038/ismej.2011.119
6
BenderS. F.WaggC.van der HeijdenM. G. A. (2016). An underground revolution: biodiversity and soil ecological engineering for agricultural sustainability. Trends Ecol. Evol.31, 440–452. doi: 10.1016/j.tree.2016.02.016
7
BolyenE.RideoutJ. R.DillonM. R.BokulichN. A.AbnetC. C.Al-GhalithG. A.et al. (2019). Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat. Biotechnol.37, 852–857. doi: 10.1038/s41587-019-0209-9
8
BrookesP. C.PowlsonD. S.JenkinsonD. S. (1982). Measurement of microbial biomass phosphorus in soil. Soil Biol. Biochem.14, 319–329. doi: 10.1016/0038-0717(82)90001-3
9
CallahanB. J.McMurdieP. J.RosenM. J.HanA. W.JohnsonA. J. A.HolmesS. P. (2016). DADA2: High-resolution sample inference from Illumina amplicon data. Nat. Methods13, 581–583. doi: 10.1038/nmeth.3869
10
ChenY.DuJ.LiY.TangH.YinZ.YangL.et al. (2022). Evolutions and managements of soil microbial community structure drove by continuous cropping. Front. Microbiol.13. doi: 10.3389/fmicb.2022.839494
11
ChenJ.YangX.ZhongD.HuoZ.SunR.DongH. (2025). Continuous cropping alters soil microbial community assembly and co-occurrence network complexity in arid cotton fields. Agriculture15, 1274. doi: 10.3390/agriculture15121274
12
Delgado-BaquerizoM.MaestreF. T.ReichP. B.JeffriesT. C.GaitanJ. J.EncinarD.et al. (2016). Microbial diversity drives multifunctionality in terrestrial ecosystems. Nat. Commun.7, 10541. doi: 10.1038/ncomms10541
13
DingM.DaiH.HeY.LiangT.ZhaiZ.ZhangS.et al. (2024). Continuous cropping system altered soil microbial communities and nutrient cycles. Front. Microbiol.15. doi: 10.3389/fmicb.2024.1374550
14
FiererN. (2017). Embracing the unknown: disentangling the complexities of the soil microbiome. Nat. Rev. Microbiol.15, 579–590. doi: 10.1038/nrmicro.2017.87
15
Gatheru WaigiM.SunK.GaoY. (2017). Sphingomonads in microbe-assisted phytoremediation: tackling soil pollution. Trends Biotechnol.35, 883–899. doi: 10.1016/j.tibtech.2017.06.014
16
GeY.ZhangJ.-b.ZhangL.-m.YangM.HeJ.-z. (2008). Long-term fertilization regimes affect bacterial community structure and diversity of an agricultural soil in northern China. J. Soils. Sediments.8, 43–50. doi: 10.1065/jss2008.01.270
17
HuangL.-F.SongL.-X.XiaX.-J.MaoW.-H.ShiK.ZhouY.-H.et al. (2013). Plant-soil feedbacks and soil sickness: from mechanisms to application in agriculture. J. Chem. Ecol.39, 232–242. doi: 10.1007/s10886-013-0244-9
18
JanssonJ. K.McClureR.EgbertR. G. (2023). Soil microbiome engineering for sustainability in a changing environment. Nat. Biotechnol.41, 1716–1728. doi: 10.1038/s41587-023-01932-3
19
KandelerE.GerberH. (1988). Short-term assay of soil urease activity using colorimetric determination of ammonium. Biol. Fertil. Soils.6, 68–72. doi: 10.1007/BF00257924
20
KotroczóZ.VeresZ.FeketeI.KrakompergerZ.TóthJ. A.LajthaK.et al. (2014). Soil enzyme activity in response to long-term organic matter manipulation. Soil Biol. Biochem.70, 237–243. doi: 10.1016/j.soilbio.2013.12.028
21
LanG.QuanF.YangC.SunR.ChenB.ZhangX.et al. (2022). Driving factors for soil fungal and bacterial community assembly in topical forest of China. Appl. Soil Ecol.177, 104520. doi: 10.1016/j.apsoil.2022.104520
22
LehmannJ.BossioD. A.Kögel-KnabnerI.RilligM. C. (2020). The concept and future prospects of soil health. Nat. Rev. Earth Environ.1, 544–553. doi: 10.1038/s43017-020-0080-8v
23
LiJ.MaZ.DaiH.LiH.QiuJ.PangX. (2024a). Application of PLSR in correlating sensory and chemical properties of middle flue-cured tobacco leaves with honey-sweet and burnt flavour. Heliyon10, e29547. doi: 10.1016/j.heliyon.2024.e29547
24
LiJ.ZuoY.ZhangJ. (2024b). Rhizosphere shifts: reduced fungal diversity and microbial community functionality enhance plant adaptation in continuous cropping systems. Microorganisms12, 2420. doi: 10.3390/microorganisms12122420
25
LisumaJ. B.MbegaE. R.NdakidemiP. A. (2021). The effects of cultivating tobacco and supplying nitrogenous fertilizers on micronutrients extractability in loamy sand and sandy soils. Plants10, 1597. doi: 10.3390/plants10081597
26
LiuJ.YaoQ.LiY.ZhangW.WangG. (2019). Continuous cropping of soybean alters the bulk and rhizospheric soil fungal communities in a Mollisol of Northeast PR China. Land. Degrad. Dev.30, 1725–1738. doi: 10.1002/ldr.3378
27
LuH.RenZ.LiA.ZhangH.JiangJ.XuS.et al. (2016). Deep sequencing reveals microbiota dysbiosis of tongue coat in patients with liver carcinoma. Sci. Rep.6, 33142. doi: 10.1038/srep33142
28
LurgiM.ThomasT.WemheuerB.WebsterN. S.MontoyaJ. M. (2019). Modularity and predicted functions of the global sponge-microbiome network. Nat. Commun.10, 992. doi: 10.1038/s41467-019-08925-4
29
McdonaldR. P. (1996). Path analysis with composite variables. Multivariate. Behav. Res.31 2, 239–270. doi: 10.1207/s15327906mbr31025
30
MichaelB. S.LiH. (2019). Compositional mediation analysis for microbiome studies. Ann. Appl. StatisticsIn.13, 661–681. doi: 10.1101/149419
31
MurphyJ.RileyJ. P. (1962). A modified single solution method for the determination of phosphate in natural waters. Anal. Chim. Acta27, 31–36. doi: 10.1016/S0003-2670(00)88444-5
32
OkoyeK.HosseiniS. (2024). “ Analysis of variance (ANOVA) in R: one-way and two-way ANOVA,” in R programming: Statistical Data Analysis in Research. (Singapore: Springer) 187–209. doi: 10.1007/978-981-97-3385-9_9
33
OlsenS. R.ColeC. V.WatanabeF. S.DeanL. A. (1954). Estimation of available phosphorus in soils by extraction with sodium bicarbonate. USDA. Circ.939.
34
PinfieldN. J. (2001). Tobacco: Production, chemistry and technology Edited by D. Layten Davis and Mark T. Nielsen. 1999. Plant Growth Regul.33, 245–245. doi: 10.1023/A:1017553623701
35
RilligM. C.MummeyD. L. (2006). Mycorrhizas and soil structure. New Phytol.171, 41–53. doi: 10.1111/j.1469-8137.2006.01750.x
36
SchlatterD.KinkelL.ThomashowL.WellerD.PaulitzT. (2017). Disease suppressive soils: new insights from the soil microbiome. Phytopathology107, 1284–1297. doi: 10.1094/PHYTO-03-17-0111-RVW
37
SinghB. K.JiangG.WeiZ.Sáez-SandinoT.GaoM.LiuH.et al. (2025). Plant pathogens, microbiomes, and soil health. Trends Microbiol.33, 887–902. doi: 10.1016/j.tim.2025.03.013
38
SixJ.BossuytH.DegryzeS.DenefK. (2004). A history of research on the link between (micro)aggregates, soil biota, and soil organic matter dynamics. Soil Tillage. Res.79, 7–31. doi: 10.1016/j.still.2004.03.008
39
SparksD. L.PageA. L.HelmkeP. A.LoeppertR. H.SoltanpourP. N.TabatabaiM. A.et al. (1996). Methods of soil analysis. Part 3 - chemical methods (Madison, WI, USA: Soil Science Society of America), 1–1391. doi: 10.2136/sssabookser5.3
40
SunL.WangS.Narsing RaoM. P.ShiY.LianZ. H.JinP. J.et al. (2023). The shift of soil microbial community induced by cropping sequence affect soil properties and crop yield. Front. Microbiol.14. doi: 10.3389/fmicb.2023.1095688
41
SunX.XuZ.XieJ.Hesselberg-ThomsenV.TanT.ZhengD.et al. (2022). Bacillus velezensis stimulates resident rhizosphere Pseudomonas stutzeri for plant health through metabolic interactions. ISME. J.16, 774–787. doi: 10.1038/s41396-021-01125-3
42
SunY.YangX.ElsgaardL.DuT.SiddiqueK. H. M.KangS.et al. (2024). Diversified crop rotations improve soil microbial communities and functions in a six-year field experiment. J. Environ. Manage.370, 122604. doi: 10.1016/j.jenvman.2024.122604
43
TedersooL.BahramM.ZobelM. (2020). How mycorrhizal associations drive plant population and community biology. Science367. doi: 10.1126/science.aba1223
44
TianT.GheysenG.KyndtT.MoC.XiaoX.LvY.et al. (2025). Pepper root exudate alleviates cucumber root-knot nematode infection by recruiting a rhizobacterium. Plant Commun.6, 101139. doi: 10.1016/j.xplc.2024.101139
45
TiemannL. K.GrandyA. S.AtkinsonE. E.Marin-SpiottaE.McDanielM. D.HooperD. (2015). Crop rotational diversity enhances belowground communities and functions in an agroecosystem. Ecol. Lett.18, 761–771. doi: 10.1111/ele.12453
46
VanceE. D.BrookesP. C.JenkinsonD. S. (1987). An extraction method for measuring soil microbial biomass C. Soil Biol. Biochem.19, 703–707. doi: 10.1016/0038-0717(87)90052-6
47
VenterZ. S.JacobsK.HawkinsH.-J. (2016). The impact of crop rotation on soil microbial diversity: A meta-analysis. Pedobiologia59, 215–223. doi: 10.1016/j.pedobi.2016.04.001
48
WaggC.SchlaeppiK.BanerjeeS.KuramaeE. E.van der HeijdenM. G. A. (2019). Fungal-bacterial diversity and microbiome complexity predict ecosystem functioning. Nat. Commun.10, 4841. doi: 10.1038/s41467-019-12798-y
49
WalkleyA.BlackI. A. (1934). An examination of the degtjareff method for determining soil organic matter, and a proposed modification of the chromic acid titration method. Soil Sci.37, 29–38. doi: 10.1097/00010694-193401000-00003
50
WangZ.LiuF.LiE.YuanY.YangY.XuM.et al. (2022). Network analysis reveals microbe-mediated impacts of aeration on deep sediment layer microbial communities. Front. Microbiol.13. doi: 10.3389/fmicb.2022.931585
51
WangB.MengF.ChengT.NiuJ.RaoD.HanZ.et al. (2025). Comprehensive responses of physiology and rhizosphere microbiome to saline–alkaline stress in soybean seedlings with different tolerances. Plants14, 3480. doi: 10.3390/plants14223480
52
WangP.XieW.DingL.ZhuoY.GaoY.LiJ.et al. (2023). Effects of maize–crop rotation on soil physicochemical properties, enzyme activities, microbial biomass and microbial community structure in Southwest China. Microorganisms11, 2621. doi: 10.3390/microorganisms11112621
53
WenY.ZangH. (2025). Crop rotation boosts yields and soil quality. Nat. Food6, 130–131. doi: 10.1038/s43016-024-01098-4
54
WuJ.LiuS.ZhangH.ChenS.SiJ.LiuL.et al. (2025). Flavones enrich rhizosphere Pseudomonas to enhance nitrogen utilization and secondary root growth in Populus. Nat. Commun.16, 1461. doi: 10.1038/s41467-025-56226-w
55
WuH.YanW.WuH.ZhangJ.ZhangZ.ZhangZ.et al. (2022). Consecutive monoculture regimes differently affected the diversity of the rhizosphere soil viral community and accumulated soil-borne plant viruses. Agricult. Ecosyst. Environ.337, 108076. doi: 10.1016/j.agee.2022.108076
56
XiaoY.WangJ.WangB.FanB.ZhouG. (2025). Soil microbial network complexity predicts soil multifunctionality better than soil microbial diversity during grassland-farmland-shrubland conversion on the Qinghai-Tibetan Plateau. Agricult. Ecosyst. Environ.379, 109356. doi: 10.1016/j.agee.2024.109356
57
XieZ.YuZ.LiY.WangG.LiuX.TangC.et al. (2022). Soil microbial metabolism on carbon and nitrogen transformation links the crop-residue contribution to soil organic carbon. NPJ Biofilms. Microbiomes.8, 14. doi: 10.1038/s41522-022-00277-0
58
YangX.HuH.-W.YangG.-W.CuiZ.-L.ChenY.-L. (2023). Crop rotational diversity enhances soil microbiome network complexity and multifunctionality. Geoderma436, 116562. doi: 10.1016/j.geoderma.2023.116562
59
YangT.SiddiqueK. H. M.LiuK. J. G. E. (2020). Cropping systems in agriculture and their impact on soil health-A review. Global Ecol. Conserv.23, e01118. doi: 10.1016/j.gecco.2020.e01118
60
YuanJ.WenT.ZhangH.ZhaoM.PentonC. R.ThomashowL. S.et al. (2020). Predicting disease occurrence with high accuracy based on soil macroecological patterns of Fusarium wilt. ISME. J.14, 2936–2950. doi: 10.1038/s41396-020-0720-5
61
YuanJ.ZhaoJ.WenT.ZhaoM.LiR.GoossensP.et al. (2018). Root exudates drive the soil-borne legacy of aboveground pathogen infection. Microbiome6, 156. doi: 10.1186/s40168-018-0537-x
62
ZhangK.Maltais-LandryG.LiaoH.-L. (2021a). How soil biota regulate C cycling and soil C pools in diversified crop rotations. Soil Biol. Biochem.156, 108219. doi: 10.1016/j.soilbio.2021.108219
63
ZhangL.ZhangJ.WeiY.HuW.LiuG.ZengH.et al. (2021b). Microbiome-wide association studies reveal correlations between the structure and metabolism of the rhizosphere microbiome and disease resistance in cassava. Plant Biotechnol. J.19, 689–701. doi: 10.1111/pbi.13495
64
ZhengQ.HuY.ZhangS.NollL.BöckleT.DietrichM.et al. (2019). Soil multifunctionality is affected by the soil environment and by microbial community composition and diversity. Soil Biol. Biochem.136, 107521. doi: 10.1016/j.soilbio.2019.107521
65
ZongD.ZhouY.ZhouJ.ZhaoY.HuX.WangT. (2024). Soil microbial community composition by crop type under rotation diversification. BMC Microbiol.24, 435. doi: 10.1186/s12866-024-03580-2
66
ZouY.LiuZ.ChenY.WangY.FengS. (2024). Crop rotation and diversification in China: enhancing sustainable agriculture and resilience. Agriculture14, 1465. doi: 10.3390/agriculture14091465
Summary
Keywords
co-occurrence networks, monoculture obstacles, soil microbiome, soil multifunctionality, sustainable agriculture
Citation
Bu Y, Chen X, Tian L, Xu M, Gao P, Wang L, Gao Q, Tan X, Xu R, Deng Z, Fu C, Li Y and Yang J (2026) Crop rotation (wheat/rapeseed–tobacco) alleviates continuous tobacco cropping obstacles and improves yield and quality by restructuring soil microbial networks across two regions. Front. Agron. 8:1753158. doi: 10.3389/fagro.2026.1753158
Received
24 November 2025
Revised
13 February 2026
Accepted
23 February 2026
Published
17 March 2026
Volume
8 - 2026
Edited by
Mohamed T. El-Saadony, Zagazig University, Egypt
Reviewed by
Lalita Rana, Dr. Rajendra Prasad Central Agricultural University, India
Om Prakash Ghimire, Clemson University, United States
Updates
Copyright
© 2026 Bu, Chen, Tian, Xu, Gao, Wang, Gao, Tan, Xu, Deng, Fu, Li and Yang.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Ying Li, liying03@caas.cn; Jinguang Yang, yangjinguang@caas.cn
†These authors have contributed equally to this work
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.