Introduction

Modern agriculture has significantly improved production efficiency in greenhouse cultivation; however, this progress is often accompanied by trade-offs in agricultural sustainability1. This issue stems from long-standing poor farming practices and the inherent limitations of greenhouse cultivation, which lead to soil eutrophication, imbalances in soil microbial communities, and the exacerbation of soil-borne diseases2,3. However, soil biodiversity, as the cornerstone of agricultural ecosystem health and stability, plays a crucial role in greenhouse production. Studies have shown that cultivating plant diversity (e.g., intercropping, companion planting, and relay cropping) enhances ecosystem health and promotes agricultural sustainability by increasing spatial and temporal diversity4,5. Furthermore, intercropping offer significant advantages in resource efficiency and crop yield compared to monoculture6. Therefore, investigating how intercropping systems may affect plant growth and soil ecosystem health is crucial for developing strategies to promote long-term agricultural sustainability.

Soil microbial community, as a key component of agricultural ecosystems, are influenced by various factors such as soil physicochemical properties, plant diversity, and agricultural management practices (e.g., cultivation patterns)7,8. In-depth studies reveal that establishing plant diversity significantly enhances ecosystem resilience and agricultural sustainability9,10 by driving functional remodeling of soil microbiomes, exemplified by the enrichment of beneficial microbes such as Bacillus, Pseudomonas, and Rhizobium. Such functional microbes synthesize phytohormones, secrete organic acids, and fix atmospheric nitrogen, thereby directly or indirectly promoting crop development through improvements in soil microenvironments and nutrient availability11,12,13. However, existing studies have primarily focused on monoculture or conventional crop rotation in field systems, with limited research on the impact of diversified cropping systems, particularly intercropping, on microbial community structure and function in controlled environments. Therefore, a comprehensive investigation into the changes in microbial communities under different cropping systems in such facilities, and their implications for agricultural sustainability and soil health, is crucial for optimizing crop management practices and enhancing agricultural productivity.

Tomato (Solanum lycopersicum L.) as a globally significant economic crop, widely cultivated in controlled environments. However, due to long-standing poor farming practices and the drive for economic gain, monoculture is widely practiced in these systems, leading to secondary soil salinization, acidification, nutrient imbalances, disruptions in microbial communities, and exacerbation of soil-borne diseases. These factors significantly reduce both the quality and yield of tomatoes14. Potato-onion (Allium cepa L. var. aggregatum G.Don) is varieties of onions. Studies have shown that certain potato-onion varieties with stronger allelopathic properties have positive effects in diversified planting systems15,16. In our preliminary research, we found that tomato/potato-onion intercropping can alleviate tomato diseases17, recruit beneficial microbial communities such as phosphate solubilizing bacteria and arbuscular mycorrhizal fungi, promote tomato root absorption of potassium and phosphorus15,18, making them an ideal set of intercropping partners. However, other ecological functions of this cultivation model remain unknown.

In this study, we first evaluated the effects of monocropping and intercropping systems on tomato biomass, quality, yield, and soil physicochemical properties. Subsequently, the absolute abundance (quantified via qPCR targeting 16S/ITS genes) and taxonomic composition (resolved by Illumina MiSeq sequencing) of soil microbial communities were comparatively analyzed between systems. The aim was to assess the potential impact of the intercropping on tomato growth, soil nutrient environment, and soil microbial community. We hypothesized that, compared to tomato monoculture: (1) the intercropping system would improve the rhizosphere nutrient environment and promote tomato growth; (2) the intercropping system would increase the abundance and diversity of rhizosphere microbial community, enrich potentially beneficial microorganisms, and enhance microbial community stability, and have the potential to promote tomato growth and improve tomato yield and quality.

Results

Soil physicochemical properties

Compared to tomato monoculture, intercropping treatment significantly increased pH, AP, AK, and NO3−-N in the tomato rhizosphere soil (P < 0.05), while significantly reduced EC (P < 0.05) (Table 1).

Table 1 The effects of tomato monoculture (T) and tomato/potato-onion intercropping (TO) systems on the soil physicochemical properties of tomato.Values are mean ± standard deviation, and the valves followed by different letters for a given factor are significantly different (P < 0.05;  Student’s t-test).   
Full size table

Tomato and potato-onion root distribution

The root distribution results showed that in tomato monoculture, the root length density value was evenly distributed on both sides of the plant (Fig. 1a). In contrast, in the intercropping treatment, the larger root length density value of tomato was distributed on the left side, and the root showed growth away from potato-onion root (Fig. 1b), while the root length density value of potato-onion was distributed in the tomato root zone (Fig. 1c). In addition, in the presence of potato-onion, the contour line of the tomato root length density at 30 cm/400 cm3 reached a soil depth of 40 cm, whereas in monoculture tomato farming cultivation, it only reached about 32 cm. Furthermore, the tomato root length density value in intercropping was higher than in tomato monoculture.

Fig. 1
Fig. 1
Full size image

Root distribution of tomato (a) in tomato monoculture system and root distribution of tomato (b) and potato-onion (c) in tomato/potato-onion intercropping system.

Tomato plant growth, quality, and yield

Intercropping significantly increased tomato root, shoot, and total plant biomass compared to tomato monoculture (P < 0.05) (Fig. 2a). Yield measurements showed that intercropping significantly increased the first and second fruit clusters, as well as total yield (P < 0.05) (Fig. 2b). Additionally, intercropping significantly increased the levels of soluble protein, soluble sugar, VC, and lycopene in tomato (P < 0.05) (Fig. 2d, e, f, g). Furthermore, intercropping significantly reduced the number of tomatoes affected by blossom-end rot (P < 0.05) (Fig. 2c).

Fig. 2
Fig. 2
Full size image

Plant dry weight (a), fruit yield (b), blossom-end rot incidence (c), soluble protein (d), soluble sugar (e), VC (f), and lycopene (g) contents in tomato under monoculture (T) and tomato/potato-onion intercropping (TO) systems. Asterisks indicate significant differences between monoculture and intercropping treatments, as determined by a Student’s t-test (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001). 

Soil microbial community abundance

The qPCR results showed that intercropping significantly increased the abundance of bacteria, Bacillus spp., and Pseudomonas spp. compared to tomato monoculture (P < 0.05) (Fig. 3a, c, d). However, there was no significant difference in the fungal abundance between the two treatments (Fig. 3b). In addition, the proportion of bacterial copies / fungal copies in intercropping treatment was significantly higher than that in tomato monoculture treatment (P < 0.05) (Fig. 3e).

Fig. 3
Fig. 3
Full size image

Microbial community abundance of bacteria (a), fungi (b), Bacillus spp. (c), Pseudomonas spp. (d), and bacterial copies/fungal copies (e) in tomato monoculture (T) and tomato/potato-onion intercropping (TO) systems. Values are mean ± standard deviation. Asterisks indicate significant differences between monoculture and intercropping treatments, as determined by a Student’s t-test (****P < 0.0001). ns: no significant difference. 

Amplicon sequencing data

In our experiment, high throughput sequencing yielded 212,147 high-quality bacterial sequences and 417,957 high-quality fungal sequences. The average read lengths were 421 bp for the 16S rRNA gene and 260 bp for the ITS1 region. Good’s coverage exceeded 97.46% for all samples (data not shown), demonstrating sufficient representation of microbial diversity. Rarefaction curves for OTUs at 97% sequence similarity approached a saturation plateau, confirming comprehensive coverage of microbial diversity.

Soil microbial community diversities and structures

Intercropping significantly increased the soil bacterial Shannon (P = 0.047) and Pielou_e indices (P = 0.009) (Fig. 4a), as well as the fungal Chao1 (P = 0.016), Shannon (P = 0.009), and Pielou_e indices (P = 0.009) (Fig. 4b), compared to tomato monoculture.

Fig. 4
Fig. 4
Full size image

Alpha diversity of bacteria (a) and fungi (b). (T) tomato monoculture, (TO) tomato/potato-onion intercropping. Values are mean ± standard deviation. Asterisks indicate significant differences between monoculture and intercropping treatments, as determined by a Student’s t-test (*P < 0.05; **P < 0.01). 

PCoA analysis indicated that the five replicate samples for each treatment clustered closely, demonstrating good reproducibility of bacterial and fungal communities in this experiment. Anosim analysis showed significant differences in soil bacterial (P < 0.05) (Fig. 5a) and fungal (P < 0.05) (Fig. 5b) community structures between the two systems.

Fig. 5
Fig. 5
Full size image

Beta diversity of bacteria (a) and fungi (b). (T) tomato monoculture, (TO) tomato/potato-onion intercropping.

Soil microbial community composition

In this experiment, we identified 26 bacterial phyla and 12 fungal phyla. Among the top 10 bacterial phyla, intercropping treatment significantly increased the relative abundance of Firmicutes, Cyanobacteria, and Actinobacteria, but decreased that of Proteobacteria, Acidobacteria, Chloroflexi, and Gemmatimonadetes (P < 0.05) (Fig. 6a). Among the top 10 fungal phyla, intercropping treatment significantly increased the relative abundance of Ascomycota, Chytridiomycota, and Glomeromycota, but decreased that of Basidiomycota (P < 0.05) (Fig. 6b).

Fig. 6
Fig. 6
Full size image

Relative abundance of the top 10 bacteria (a) and fungi (b) at the phylum level, and histogram of the linear discriminant analysis (LDA) scores computed for differentially abundant bacterial (c) and fungal (d) genera between the tomato monoculture and intercropping system. The threshold for discriminative features was LDA score ≥ 2.0. (T) tomato monoculture, (TO) tomato/potato-onion intercropping. Values are mean ± standard deviation. Asterisks indicate significant differences between monoculture and intercropping treatments, as determined by a Student’s t-test (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001). 

Among the top 50 bacterial genera, histogram of the LAD scores showed that 16 genera of Rubrobacter, f_67_14, MB_A2_108, Subgroup_6, Solirubrobacter, MM2, Conexibacter, and Bacillus etc. were significantly enriched in the intercropping treatment (P < 0.05), while 12 genera of Sphingomonas, Gemmatimonas, KD4_96, and Ramlibacter etc. were significantly enriched in the tomato monoculture treatment (P < 0.05) (Fig. 6c). Among the top 50 fungal genera, 20 genera of unclassified_Fungi, unclassified_Pyronemataceae, Mortierella, unidentified, Fusarium, and Solicoccozyma etc. were significantly enriched in the intercropping treatment (P < 0.05), while 8 genera of Tausonia, Oidiodendron, Acremonium, and Didymella etc. were significantly enriched in the tomato monoculture treatment (P < 0.05) (Fig. 6d).

Relationship between rhizosphere microbial community and various environmental variables

Redundancy analysis (RDA) analyzed the relationship between changes in bacterial and fungal community structure and environmental factors among different treatments. The results showed that AP (r2 = 0.969, P = 0.004), pH (r2 = 0.919, P = 0.003), NO3−-N (r2 = 0.936, P = 0.006), EC (r2 = 0.648, P = 0.02), and yield (r2 = 0.918, P = 0.003) were the main influencing factor of soil bacterial community structure (Fig. 7a). AP (r2 = 0.961, P = 0.008), pH (r2 = 0.923, P = 0.006), NO3−-N (r2 = 0.951, P = 0.006), EC (r2 = 0.893, P = 0.002), OM (r2 = 0.662, P = 0.025), and yield (r2 = 0.917, P = 0.003) were the main influencing factor of soil fungal community structure (Fig. 7b).

Fig. 7
Fig. 7
Full size image

Redundancy analysis (RDA) of bacterial (a) and fungal (b) communities in tomato monoculture (T) and tomato/potato-onion intercropping (TO) systems with environmental variables.*** indicates statistically significant correlation at the P < 0.001.

Soil microbial community co-occurrence network analysis

The results of microbial co-occurrence network analysis indicated that the modularity index for each treatment was greater than 0.5, suggesting that the symbiotic networks for all treatments had a typical modular structure. Compared to tomato monoculture, the intercropping treatment increased the total number of correlated edges, average degree, network density, and average clustering coefficient in the tomato rhizosphere bacterial community, while decreasing the modularity index, network diameter, number of modules, and average path length. In contrast, for fungi, these topological showed exhibited the opposite trends. In addition, the co-occurrence network analysis of bacterial and fungal communities showed that intercropping increased the total number of correlated edges, average degree, and modularity index, while reducing network density, average clustering coefficient, and average path length (Fig. 8 and Table 2).

Fig. 8
Fig. 8
Full size image

Co-occurrence networks for bacterial (a, b), fungal (c, d), and integrated bacterial-fungal (e, f) communities under tomato monoculture (T) and tomato/potato-onion intercropping (TO) systems.

Table 2 Topological properties of the co-occurrence network of rhizosphere microbial communities in tomato monoculture (T) and tomato/potato-onion intercropping (TO) systems.
Full size table

Discussion

Appropriate diversity planting can enable mutual benefits between plants, which is often related to positive plasticity responses of plants to adapt and improve their growth environment19,20. Studies have shown that plant roots usually respond to intercropping strongly, exhibiting either avoidance or attraction behaviors depending on the neighboring species21. In this study, the presence of potato-onion altered the root distribution pattern of tomato, causing tomato roots to avoid the growth of potato-onion roots (Fig. 1b), while potato-onion roots tended to grow toward the tomato roots (Fig. 1c). Furthermore, the presence of potato-onion significantly promoted tomato growth, increasing both yield and quality (Fig. 2). These findings are similar to previous studies showing that maize/legume intercropping altered root length and surface area density in both crops, while enhancing nitrogen absorption through an increased number of legume nodules, thus promoting crop growth22. We also found that intercropping with potato-onion significantly reduced tomato blossom-end rot, a common physiological disorder caused by localized calcium deficiency (often due to uneven distribution and/or insufficient supply). This may be due to organic acids secreted by potato-onion roots, which enhance calcium dissolution, and the microclimate created by intercropping, which promotes tomato transpiration and improves calcium uptake and transport23. In the future, we will further investigate the mechanisms by which intercropping with potato-onion reduces tomato blossom-end rot.

Increased plant species makes the rhizosphere environment more complex, often leading to significant changes in soil conditions24. Previous studies have shown that intercropping can enhance soil pH, NO3−-N, AP, and AK, while reducing soil EC25,26, which aligns with our findings (Table 1). However, no significant changes were observed in NH4+-N and OM, consistent with tomato/potato-onion intercropping system with added biochar6. This may be because increased plant diversity can accelerate the decomposition of organic matter, promoting the conversion of organic nitrogen into mineral nitrogen in the soil. Diverse plant roots and microbial activity together promote the rapid transformation and release of nitrogen in the soil, thereby increasing nitrate nitrogen levels27. Additionally, the roots of potato-onion can recruit phosphorus-solubilizing bacteria and secrete organic acids that activate soil phosphorus and potassium, while also reducing the activity of certain salt ions. This process promotes the dissolution and transport of phosphorus and potassium, thereby lowering soil EC18. Overall, tomato/potato-onion intercropping improves soil conditions by increasing the availability of nutrients such as N, P, and K, while reducing soil salinization. This, in turn, promotes crop growth and development, potentially enhancing tomato yield and quality.

Soil microorganisms are influenced by various environmental factors. In our study, the rhizosphere bacterial and fungal community structure of tomato was significantly positively correlated with AP, NO3−-N and soil pH under intercropping conditions. In addition, there was a significant positive correlation between soil microbial communities and tomato yield (Fig. 7a, b), indicated that soil nitrogen and phosphorus resources, as well as optimal soil pH, may be key environmental factors affecting soil bacterial and fungal communities structure. Meanwhile, soil microbial community may play a positive role in increasing tomato yield. Furthermore, our results showed that bacterial and fungal communities were significantly negatively correlated with EC in intercropping system (Fig. 7a). This suggests that lower salt ion concentrations may contribute to microbial growth and metabolic activity, which may be closely associated with increased microbial diversity in intercropping system28,29.

Intercropping can enhance the diversity, functionality, and stability of soil microorganisms by providing diverse organic sources, improving the soil environment, and promoting plant-microbial interactions11,30. In our study, the intercropping system significantly increased the abundance of total bacteria, Bacillus spp., and Pseudomonas spp. (Fig. 3a, c, d), and the α diversity of bacterial and fungal communities was significantly higher than that of monoculture (Fig. 4a, b). This suggests that increased plant diversity can promote the enrichment of potentially beneficial microorganisms and positively affect bacterial and fungal diversity6,31. The increase in bacterial and fungal diversity in the intercropping system may be related to its resource diversity. Previous studies have shown that intercropping can enhance resource diversity in the rhizosphere, such as organic matter, nutrients, and root exudates, providing a rich material foundation for microbial diversity, thereby improving the diversity of tomato rhizosphere microorganisms13,31. A large number of studies have shown that Bacillus and Pseudomonas have strong antagonistic ability as potential beneficial microorganisms. They play an active role in improving soil health by producing antibiotics, competing for nutrients and space, and inducing plant disease resistance responses that inhibit soil-borne pathogens6,32. Additionally, significant differences were observed in bacterial and fungal community composition and structure between two systems (Fig. 5 and Fig. 6). The intercropping system enriched potentially beneficial phyla (e.g., Firmicutes and Actinobacteria) and genera (e.g., Bacillus) in the tomato rhizosphere. This is similar to the results of Huang et al. (2022), who found that tea tree/bean intercropping improved soil fertility and tea quality by altering the species composition of Bacillus33. Semblat et al. (2024) also found that intercropping wheat and peas increased the abundance of Pseudomonas in the pea rhizosphere, thereby enhancing the absorption of nutrients such as potassium, iron, and zinc by peas34. This suggests that the increase of potentially beneficial microorganisms in intercropping systems may be closely related to the improvement of long-term soil fertility. This change usually has a positive effect on the rhizosphere nutrient environment and plant growth and development. In addition, the intercropping system significantly increased the bacterial/fungal ratio (Fig. 3e), possibly because intercropping system increased the content of available phosphorus and potassium in the soil, and bacteria used easily decomposed carbon sources (such as sugars and amino acids) more efficiently, thus promoting bacterial growth. Intercropping may increase the bacterial/fungal ratio by stimulating bacterial metabolic activity through the input of diverse root exudates and litter.

Complex species interactions are key indicators of community biodiversity, and intercropping can influence interactions between soil bacteria and fungi35. Microbial co-occurrence network analysis showed that the intercropping system increased the total number of correlation edges, average degree, graph density, and average clustering coefficient in the tomato rhizosphere bacterial community, while decreasing the modularity index, network diameter, modularity number, and average path length. This suggests that the intercropping system promotes interactions among microorganisms within the bacterial community (Fig. 8 and Table 2). The increase in the number of edges and the clustering coefficient suggests that more microorganisms form tight connections within the network, potentially improving the stability and complexity of the community6. An increase in average degree indicates that each bacterial species has more interactions with others, demonstrating a more cohesive and interconnected microbial ecosystem36. The reduced modularity implies a shift toward a more cohesive network structure, with increased within-module species clustering (evidenced by higher average clustering coefficient) and enhanced cross-module interactions (supported by fewer negative correlations), potentially facilitating cooperation between functionally distinct microbes37. The reduction in network diameter and average path length means that the shortest path between species becomes shorter, making communication and information transfer among microorganisms more efficient38. Taken together, the intercropping system enhances community synergies and interdependence by facilitating connections among different microorganisms in the bacterial community. This change may improve the functional diversity, ecological stability, and adaptability of rhizosphere bacterial communities to environmental stress. Thus, the intercropping system may enhance soil health and crop growth potential by optimizing the structure and function of microbial communities. In conclusion, potato-onion intercropping may improve the stability and productivity of the system by increasing the diversity of the rhizosphere microbial community, the abundance of potentially beneficial microorganisms, and enhancing the co-occurrence relationships within the microbial community.

Conclusions

In summary, this study indicated that intercropping system altered the distribution of tomato roots, increased plant biomass, improved tomato quality and yield, and reduced the number of diseased fruits caused by tomato blossom-end rot. The intercropping system significantly increased the alpha diversity of rhizosphere bacterial and fungal communities, increased the abundance of Bacillus spp. and Pseudomonas spp., and led to significant changes in the composition and structure of tomato microbial communities. In addition, the changes in the rhizosphere bacterial and fungal community structure of tomato/potato-onion intercropping system were closely related to soil AP, NO3−-N, pH, and tomato yield. Intercropping also significantly improved the stability and functional efficiency of bacterial communities. This study provided strong evidence that intercropping systems improve soil nutrient conditions, increased soil microbial diversity, and altered soil microbial community structure, thereby enhanced system productivity.

Methods

Root box experiment

The experiment was conducted from April to June 2021 in a greenhouse at Northeast Agricultural University, Harbin, China (45°41’N, 126°37’E). The tomato (Solanum lycopersicum L.) cultivar “Dongnong 708” and the potato-onion ( Allium cepa L. var. aggregatum G.Don) cultivar “Wangkui” were used in this study. Tomato seedlings were raised conventionally and transplanted into root boxes made of polyvinyl chloride (PVC) material (length × width × height: 30 cm × 10 cm × 42 cm) at the two-leaf stage. The soil was black soil (Mollisol) with sandy loam texture. The soil pH was 6.59, EC was 116.49 µS cm−1, NO3−-N was 12.28 mg kg−1, NH4+-N was 4.89 mg kg−1, available P (AP) was 20.16 mg kg−1, available K (AK) was 150.80 mg kg−1, Organic matter (OM) was 2.86 g kg−1. The experiment included tomato planting treatment (T) and tomato/potato-onion intercropping treatment (TO) (Fig. 9a). Each treatment was replicated three times, with three root boxes per replicate, for a total of nine boxes. In the monoculture treatment, tomato plants were planted in the center of the boxes, while in the intercropping treatment, tomatoes and potato-onions were planted 10 cm apart in a 1:3 ratio. A randomized block design was used, with protective rows surrounding the plots. Irrigation was applied every three days to maintain soil moisture at ~ 60%. The root length density of tomato seedling was measured 20 days after transplanting.

Fig. 9
Fig. 9
Full size image

Schematic diagram of root box experiment (a) and greenhouse experiment (b). ○ represents tomato and × represents potato-onion.

The plant roots were sampled by stratified subsection excavation method39. Specifically, the transparent PVC plate of the box was removed, and the roots of the plant were sampled in an iron box (length × width × height: 5 cm × 10 cm × 8 cm) in turn. The samples were put into water to wash off the soil on the root surface, and then the plant roots were picked out with tweezers, and the roots of two crops were distinguished by color and smell. The roots of potato-onion were white and transparent with smooth root surface and pungent smell, and the roots of tomato were yellow with whisker roots40. The roots of different plants in each treatment were numbered. After numbering, the roots were scanned with a root analyzer (LA-S2400, Wseen Instruments Inc., Zhejiang), and the root length density values in each treatment were calculated according to the root length and soil volume. Surfer software (Golden Software Surfer 8.0) was used to synthesize the isograms of root length density and root mass density.

Greenhouse experiment

The experiment was conducted from April to September 2021 in a greenhouse (temperature 15℃ − 35℃, humidity 50% − 80%). The soil and its physicochemical properties, and experimental design were the same as in the root box experiment. Each treatment included five replicate plots (5 m × 2 m), with tomato seedlings transplanted at the two-leaf stage. Prior to planting, well-decomposed pig manure (37,500 kg ha−1) was applied, providing the following nutrient composition: organic matter 15%, N 0.5%, P 0.5%, and K 0.4%. Additionally, potassium sulfate-based compound fertilizer (nutrient content ≥ 45%, N:P2O5:K2O ratio of 12:18:15) was applied at a rate of 100 kg ha−1. Each plot contained five rows, with 13 tomato seedlings per row. Three potato-onions were planted approximately 10 cm from each tomato seedling, parallel to the tomato plants in the intercropping treatment. The plant spacing and row spacing of tomato were 30 cm and 50 cm, respectively, and potato-onion bulbs were spaced 10 cm apart (Fig. 9b). Set up protective rows around. Drip irrigation was used throughout the experiment to meet the plants’ water needs, and manual weeding was performed once a week after tomato transplantation.

Rhizosphere soil and root sampling and plant biomass measurements

In the greenhouse experiment, plant samples were collected 45 days after transplantation to determine tomato plant biomass. Tomato rhizosphere soil was collected by shaking the roots after measuring the yield of the second fruit cluster (through a 2 mm mesh). A portion of the soil was subjected to physicochemical properties after natural air drying, and the remaining portion was stored at −80°C for DNA extraction. The tomato seedlings were dried at 105°C for 30 min, then further dried at 80°C until a constant weight was achieved. The tomato plant biomass was measured using a balance (± 0.001 g).

Soil property analysis

A 5.0 g soil was suspended in 12.5 mL of deionized water, and pH and EC were measured using a glass electrode41. Soil inorganic nitrogen (N), available phosphorus (P), and exchangeable potassium (K) were extracted with 2 M KCl, 0.5 M NaHCO3 (pH 8.5), and 1 M NH4OAc (pH 7.0), respectively. Soil filtrates were then analyzed using a Continuous Flow Analyzer (SAN++ , Skalar, Breda, Netherlands). Soil organic matter was quantified using the potassium dichromate volumetric method42.

Soil DNA extraction

Weighed 0.25 g of soil, extracted total DNA using the PowerSoil® DNA Isolation Kit (MO BIO Laboratories, CA, USA) according to the manufacturer’s instructions. Each composite soil sample was extracted three times, and the extracted DNA solution was combined. The DNA samples were stored at −80°C.

Quantitative PCR (qPCR)

qPCR analysis was performed using the IQ5 real-time PCR system (Bio-Rad Laboratories, Hercules, CA, USA). The abundance of bacteria, fungi, Bacillus, and Pseudomonas in tomato soil was quantified using specific primers: 338F/518R43, ITS1F/ITS444, BacF/BacR45, and PsF/PsR46, respectively. The reaction system and conditions followed those described by He et al. (2021)6. After the reaction was completed, the melting curve was analyzed, and the PCR product was verified by agarose gel electrophoresis. A negative control was included in each 96-well plate. A standard curve for the corresponding gene fragments of each bacterial group was prepared according to a method established in this laboratory47, and the abundance of microorganisms in each sample was calculated based on the curve.

Illumina MiSeq sequencing and data processing

The bacterial 16S rRNA gene (V3-V4 regions) and fungal rRNA gene (ITS1 region) were amplified using the primer pairs 338F/806R48 and ITS1F/ITS249, respectively. The 25 µL PCR reaction mixture contained: 5 µL of 5× FastPfu Buffer, 0.5 µL of FastPfu DNA Polymerase (2.5 U/µL), 2 µL of 2.5 mM each dNTP, 0.5 µL of each forward and reverse primer (10 µM), 1 µL of DNA template, and deionized water to 25 µL. PCR amplification was performed on an ABI GeneAmp 9700 system, and 16S rRNA V3-V4 and ITS1 amplification program reference He et al., 20216. The PCR products were visualized by 2% agarose gel electrophoresis, purified using the Agarose Gel DNA Purification Kit (TakaRa), and the purified products that met quality standards were sent to Shanghai Personal Biotechnology Co., Ltd. (Shanghai, China) for sequencing.

The raw sequence reads were quality filtered and processed using FLASH50. MiSeq sequencing data were analyzed using QIIME (Version 1.9.0)51, following these steps: (1) removal of barcode sequences, primer sequences, and low-quality fragments; (2) OTU clustering of high-quality sequences selected through screening with CD-HIT at a 97% similarity threshold52; (3) removal of chimeric sequences; (4) rarefaction to the minimum sequence count for further analysis. Taxonomic classification of bacteria and fungi was performed using the SILVA53 and UNITE54 databases, with a similarity threshold of 0.7. The high-throughput sequencing data for soil bacteria and fungi were deposited in the National Center for Biotechnology Information with the primary accession code PRJNA1189198.

Statistical analysis

Alpha diversity indices, including Shannon, Chao 1, and Pielou_e indices, were calculated using QIIME51. The data on soil physicochemical properties, tomato agronomic traits, total abundance of bacteria, fungi, Bacillus spp. and Pseudomonas spp., α diversity index, bacterial and fungal taxa (phylum and genus) of monoculture and intercropping systems were compared using Student’s t-test in SPSS software (Version 21.0). Anosim and principal coordinate analysis (PCoA) were conducted on soil microbial communities based on Bray-Curtis distance to evaluate differences in microbial community structure. Redundancy analysis (RDA) based on Euclidean distance was used to evaluate the relationship between bacterial and fungal community structure and various factors. PCoA and RDA were analyzed based on the “vegan” package in R (Version 4.1.2). Inter-group differences at the phylum and genus levels were analyzed using Stamp software (Version 2.1.3), and correlation coefficients between microbial genera were calculated using the “Hmisc” and “psych” packages. The Linear Discriminant Analysis (LDA) Effect Size (LefSe) method was employed to identify ecologically representative species. Taxa meeting the criteria of LDA score ≥ 2.0 and P < 0.05 were defined as characteristic biomarkers for downstream analyses. We computed a pairwise correlation matrix using the corr.test function from the Psych package in R 4.3.1, based on Pearson’s correlation coefficient (ρ). A microbial co-occurrence network was constructed by selecting associations with an absolute Spearman’s correlation coefficient (ρ) > 0.6 and P < 0.05, and visualized using Gephi (Version 0.9.2).