Abstract
Introduction:
Intercropping is a sustainable agricultural practice known to enhance farmland biodiversity and productivity. However, research on rhizosphere microbial communities in walnut-soybean intercropping systems remains limited.
Methods:
This study investigated a 5-year-old orchard on the Loess Plateau. Rhizosphere soil characteristics and microbial communities were compared between walnut monoculture and walnut-soybean intercropping treatments, where soybean had been intercropped for three consecutive years.
Results:
Intercropping significantly increased soil organic matter (SOM), microbial biomass carbon (MBC) and nitrogen (MBN), and the activities of carbon- and phosphorus-cycling enzymes (e.g., β-glucosidase), whereas it markedly reduced total nitrogen (TN) and total phosphorus (TP). Bacterial and fungal community composition and abundance differed significantly between intercropping and monoculture treatments, with Mortierellomycota and Proteobacteria enriched under intercropping. Mantel tests revealed that bacterial community composition was correlated solely with SOM, while fungal community composition was correlated with both TN and SOM. Bacterial co-occurrence networks exhibited higher connectivity than fungal networks. The mean soil multifunctionality (SMF) index was significantly higher under intercropping (0.54) than under monoculture (0.43, P < 0.01). Bacterial PC1 emerged as the dominant positive predictor of SMF (parameter estimate = +0.39, P < 0.05), and bacterial variables collectively explained 70.51% of the variation in SMF, further indicating that bacteria were the primary microbial drivers of soil multifunctionality in this intercropping system.
Conclusion:
This study suggests walnut-soybean intercropping may improve soil properties via regulating soil properties and microbial dynamics, providing preliminary guidance for sustainable walnut orchard management.
1 Introduction
Intercropping, an ancient agricultural practice that involves the simultaneous cultivation of two or more crop species on the same land area, has reemerged as a promising strategy for sustainable agriculture (Zhao et al., 2016). This agroecological system improves agro-ecosystem performance by enhancing the spatiotemporal use efficiency of critical resources such as water, light, and soil nutrients (; ), and by promoting the interspecific transfer of essential substances and signaling molecules through root-mediated interactions (Zhang et al., 2011). Legumes are especially advantageous in intercropping systems due to their ability for biological nitrogen fixation and their function as green manure (Olivier et al., 2017). Incorporating legumes into orchard systems can counteract the decline in soil microbial diversity and low nitrogen use efficiency associated with long-term monoculture, while simultaneously mitigating environmental degradation caused by bare inter-rows (; ).
Soil health and ecosystem multifunctionality (the ability of soil to simultaneously provide multiple ecological services, e.g., nutrient cycling, organic matter decomposition, and disease suppression) are fundamental to the long-term sustainability of agricultural systems (; ). Rhizosphere microbial communities, particularly fungi and bacteria, play a central role in regulating these functions: fungi contribute to organic matter breakdown and establish mycorrhizal symbioses with plant roots, whereas bacteria are instrumental in decomposing recalcitrant organic compounds and producing bioactive antibiotic (; ; ). Evidence from previous studies indicates that legume intercropping can alter rhizosphere microbial assemblages in orchard ecosystems. For example, intercropping with faba bean in French walnut orchards has been shown to increase colonization by arbuscular mycorrhizal (AM) fungi (Trinsoutrot-Gattin, 2021), and clover cover crops enhance ectomycorrhizal (ECM) fungal associations in pecan trees (Rodriguez-Ramos et al., 2022). Additionally, walnut-soybean intercropping increased the relative abundance of beneficial bacteria (e.g., Burkholderia, Rhodopseudomonas, Pseudomonas, and Flavobacterium) associated with soil organic matter accumulation and nitrogen availability (; Wang et al., 2023). However, the regulatory mechanisms by which legume intercropping affects soil health and multifunctionality, particularly through the interaction between rhizosphere soil properties and microbial communities, remain unclear in walnut systems.
Walnut (Juglans regia L.) is a dry-fruit tree species with both economic and ecological significance (). Between 2000 and 2020, global walnut production increased by 163% (FAOSTAT Crop Production Database, https://www.fao.org/faostat/en/#data/QCL). However, due to the increased planting area and poor management practices, the yield and quality of walnuts are declining, leading to reduced economic benefits. For this reason, the introduction of intercropping patterns has become a simple and effective measure to improve the cultivation and management practices in walnut orchards (). Furthermore, walnut is one of the most widely studied tree species for temperate agroforestry intercropping worldwide, highlighting great research and application significance of walnut intercropping systems. Accordingly, to support the development of the walnut industry, research on intercropping in walnut orchards and resolving knowledge gaps regarding rhizosphere microbiota in intercropping systems is essential.
In this study, we surveyed five-year-old walnut orchards to characterize rhizosphere soil properties and soil microbial communities of walnuts under two cultivation patterns: legume intercropping and bare alley management. We cultivated legumes in walnut orchards located on the Loess Plateau in Northwest China and evaluated them during a production season to explore the following questions: (i) What essential rhizosphere soil physicochemical properties have been altered by legume planting, thereby improving the orchard soil’s suitability for walnut growth? (ii) How do changes in rhizosphere soil physicochemical properties influence soil enzyme activities and microbial community composition? Furthermore, what are the interrelationships among these three key components?
2 Materials and methods
2.1 Site description
As shown in Figure 1, the experimental field is located at the Lanzhou Germplasm Resource Nursery (36°10′N, 103°42′E), the capital city of Gansu Province in Northwest China, which has a temperate continental climate (hot and humid summer, cold and dry winter, with mean annual precipitation of 327 mm). The main soil type is irrigation-silted soil. According to the conditions of the resource nursery, the tested fields had been used for early-bearing walnut (Juglans regia cv. Yuanlin) production for over five years. The walnut trees were planted with a spacing of 4 m between each tree along north-south rows and an interrow distance of 5 m, giving a planting density was 495 trees ha-1. Farming management is implemented on the soil surface of the walnut cultivation area. Before winter every year, each interrow is plowed to a depth of 30 cm and fertilized with approximately 4,500 kg ha-1 of organic fertilizer. The applied organic fertilizer contained 45% organic matter, 1.8% total N, 1.2% P2O5, 1.5% K2O, and its moisture content was below 30%, conforming to the conventional quality standard of commercial compost for orchard use. This 30 cm tillage depth is the conventional local overwintering management practice on the Loess Plateau, rather than the sampling depth designed for this experiment. Soil basic fertility was analyzed before the start of the experiment; the results showed that the soil had an organic carbon (SOC) content of 11.99 g·kg-1, total nitrogen (TN) content of 1.26 g·kg-1, total phosphorus (TP) content of 1.24 g·kg-1, total potassium (TK) content of 21.57 g·kg-1, available nitrogen (AN) content of 113.48 mg·kg-1, available phosphorus (AP) content of 32.52 mg·kg-1, available potassium (AK) content of 318.26 mg·kg-1, and a pH value of 8.31 in the topsoil layer at depths ranging from 0 to 60 cm. The deeper 0–60 cm layer was adopted solely to comprehensively characterize the initial vertical nutrient distribution of the experimental site before treatments were implemented, considering the deep rooting feature of walnut trees.
Figure 1
2.2 Experimental design and soil sample collection
Two 0.5 ha plots were designated for two experimental treatments: walnut plant monoculture (Monoculture) and walnut intercropping with soybean (Intercrop). Walnut was transplanted into the Lanzhou Germplasm Nursery in the spring of 2019. Walnut-soybean intercropping was initiated in the spring of 2021, with soybeans sown between walnut rows at a distance of 1 m away from walnut trunks. Root interspecific competition is mainly distributed 1.0-2.5 m from walnut rows. Sampling at 0.5-1.0 m from trunks largely avoids soybean root disturbance. Rhizosphere soil samples were collected in mid-August 2023, corresponding to the third consecutive year of soybean intercropping. Rhizosphere soil was collected using the conventional shaking root method widely applied in rhizosphere ecology research. Briefly, surface litter was cleared away prior to sampling for each selected walnut tree. A soil block sized 40 cm × 40 cm × 40 cm was carefully excavated at a horizontal distance of 0.5-1.0 m from the walnut trunk, an area characterized by the highest fine root density of walnut trees. Fine walnut roots (diameter < 2 mm) were isolated from the soil block; loosely attached bulk soil was removed via gentle shaking, and the soil tightly adhering to root surfaces was collected as rhizosphere soil. Rhizosphere soil obtained from five walnut trees within each subplot was thoroughly combined to form one composite sample, and three independent biological replicates were obtained for each cultivation treatment. Each plot was further divided into three spatially independent subplots, with each subplot serving as one replicate. Within each subzone, rhizosphere soil from five walnut trees (at a soil depth of 0–40 cm) was collected and mixed into one sample, yielding three independent samples per treatment. This sampling depth was selected because previous regional studies confirmed that roughly 55.7% of walnut fine roots on the Loess Plateau are distributed within the 0–40 cm soil layer, which constitutes the core rhizosphere zone for root-soil-microbe interactions and is most responsive to legume intercropping regulation. In addition, soybean fine roots mostly exist in 0–20 cm soil, and our 0–40 cm sampling depth further minimizes soybean-derived interference, guaranteeing representative walnut rhizosphere samples. Soil samples were transported to the lab at 4 °C, homogenized and sieved (2 mm) to remove roots and gravel, then split into three subsamples: -80 °C storage for DNA sequencing (≤ 2 weeks), 4 °C storage for microbial biomass and enzyme tests (≤7 days), and air-dried samples at room temperature for physicochemical analysis. This standard protocol prevents sample degradation.
2.3 Soil properties determination
In the experiment, the rhizosphere soil was sifted through a 0.5-mm sieve, and pH was determined using the potentiometric method in a soil:water solution (1:2.5 w/v). SOC was determined by the K2Cr2O7 oxidation method (Salam et al., 1999). The concentrations of TN, TP, and TK were measured using the Kjeldahl method (Qiu et al., 2015), the ammonium molybdate colorimetric method (), and the flame photometric (NaOH) method (Yue et al., 2022), respectively. AN was assayed by the alkaline hydrolysis diffusion method (Lu, 1999). AP was analyzed by the molybdenum blue method, and AK was extracted with ammonium acetate and measured by flame photometry (Mohammad, 2000). Soil microbial biomass carbon (MBC) and nitrogen (MBN) were quantified using the chloroform fumigation-extraction technique (; Vance et al., 1987). The soil β-1,4-glucosidase (BG) activity was assayed using the p-nitrophenyl-β-D-glucoside as substrate, as described by Li et al (Li et al., 2019). The N-acety-β-D-glucosaminidase (NAG) activity was estimated using the method as described by Paz-Ferreiro et al (Paz-Ferreiro et al., 2012). The cellobiohydrolase (CBH) activity was assayed fluorimetrically. The L-leucine aminopeptidase (LAP) activity was assayed with a fluorogenic 7-amino-4-methycoumarin as a substrate (Saiya-Cork et al., 2002). The acid phosphatase activity was measured according to Du et al ().
2.4 Genomic DNA extraction and Illumina sequencing
The CTAB/SDS technique was used to extract whole genome DNA samples. The concentration and purity of DNA were examined on 1% agarose gels, and subsequently, the DNA was diluted to 1 ng/μL using sterilized water based on concentration measurements.
Bacterial and fungal genomic DNA were amplified using the 520F/802R and ITS1F/ITS2 and primer pairs, respectively, in a 25 μL reaction mixture. All PCR reactions were carried out in 30 μL reactions with 15 μL of Phusion® High-Fidelity PCR Master Mix (New England Biolabs), 0.2 μM of forward and reverse primers, and approximately 10 ng template DNA. Thermal cycling consisted of initial denaturation at 98 °C for 1 min, followed by 30 cycles of denaturation at 98 °C for 10 s, annealing at 50 °C for 30 s, and elongation at 72 °C for 30 s, followed by a final extension at 72 °C for 5 min.
Equal volumes of 1X loading buffer (containing SYBR Green) were mixed with PCR products and subjected to electrophoresis on a 2% agarose gel for detection. Samples with bright main bands between 400–450 bp were selected for further experiments. PCR products were pooled in equimolar amounts. Then, the mixture of PCR products was purified using the GeneJET Gel Extraction Kit (Thermo Scientific). Sequencing libraries were generated using the Illumina TruSeq DNA PCR-Free Library Preparation Kit (Illumina, USA) following the manufacturer’s recommendations, with index codes incorporated during library construction. The library quality was assessed on the Qubit@ 2.0 Fluorometer (Thermo Scientific) and Agilent Bioanalyzer 2100 system. Finally, the library was sequenced on an Illumina NovaSeq platform, and 250 bp paired-end reads were generated.
2.5 Bioinformatics analysis
Paired-end reads from the original DNA fragments were merged using FLASH (Mago and Salzberg, 2011), a very fast and accurate analysis tool designed to merge paired-end reads when there are overlaps between read1 and read2. Paired-end reads were assigned to each sample according to the unique barcodes. Sequences were analyzed using the QIIME () software package (Quantitative Insights Into Microbial Ecology), and in-house Perl scripts were used to analyze alpha- (within samples) and beta- (among samples) diversity. First, reads were filtered by QIIME quality filters. Then, the script pick_de_novo_otus.py was used to pick operational taxonomic units (OTUs) by creating an OTU table. Sequences with ≥ 97% similarity were assigned to the same OTUs. A representative sequence was selected for each OTU and the RDP classifier (Wang et al., 2007) to annotate taxonomic information for each representative sequence. To compute alpha diversity, we rarefied the OTU table and calculated three metrics: Chao1 (estimating species richness), observed species (estimating the number of unique OTUs in each sample), and Shannon index (estimating species diversity). QIIME calculates both weighted and unweighted UniFrac, which are phylogenetic metrics used to measure beta diversity (reflecting differences in community composition among samples).
2.6 Calculation of soil ecosystem multifunctionality
Soil ecosystem multifunctionality (SMF) was quantified using two complementary approaches (Maestre et al., 2012; ; Manning et al., 2018). In the averaging approach, each of the 11 variables (Supplementary Table 1) was Z-score standardized across all samples using the pooled mean and standard deviation, and SMF was calculated as the mean of the standardized values. In the multi-threshold approach, SMF was calculated as the number of functions simultaneously exceeding a given percentage (T = 25%, 50%, and 75%) of their observed maximum; the area under the multi-threshold curve was used as an integrative SMF index.
The 11 variables covered carbon storage (SOM), nutrient availability (TN, TP, AN, AP), microbial biomass (MBC, MBN), and extracellular enzyme activities (BG, ACP, CBH, NAG, LAP) (Saiya-Cork et al., 2002; Paz-Ferreiro et al., 2012). All were measured as described in Section 2.3, with no missing data. Microbial diversity or ordination metrics were not included in the index to avoid circular inference when later relating microbial community attributes to SMF (Maestre et al., 2012; ; Manning et al., 2018).
2.7 Bacterial functional prediction using PICRUSt2 and KEGG
Bacterial functional profiles were inferred from 16S rRNA OTU representative sequences using PICRUSt2 v2.5.2 () in a conda environment with Python 3.9. Closed-reference OTUs were placed against the IMG database (release 2024_01) using the default hidden-state prediction pipeline (HMMER v3.3.2), with the sequential steps place_seqs.py, hsp.py, and pathway_pipeline.py. EC profiles were mapped onto KEGG Orthology pathways via the MetaCyc-to-KEGG conversion table embedded in PICRUSt2. OTUs with NSTI > 2.0 were excluded, and pathways with mean relative abundance < 0.01% across all six samples were filtered out. These profiles are predictions based on 16S taxonomy rather than direct measurements (; ).
2.8 Fungal functional-guild assignment using FUNGuild
Fungal guilds were assigned from ITS OTU representative sequences using FUNGuild v1.1 (Nguyen et al., 2016) with the default UNITE taxonomic database (release 8.3, 2023-09-15). Only assignments with ‘Highly Probable’ or ‘Probable’ confidence were retained; ‘Possible’ assignments were discarded. Unassigned OTUs were classified as ‘Undefined’. Guild relative abundances were calculated per sample, and guilds with mean abundance < 0.01% across all six samples were removed. Guild assignments are database-derived predictions rather than directly measured functional traits (Nguyen et al., 2016).
2.9 Statistical analysis of data
The normality of the data was analyzed using the Kolmogorov–Smirnov (K–S) test, and the homogeneity of variances was analyzed using Levene’s test. Statistical product and service solutions (SPSS, version 22.0, USA) was used to perform independent-samples t-test for comparing differences in soil physicochemical parameters between the two treatments. To investigate variations in the community structures of fungi and bacteria across different treatment groups, we used unweighted Unifrac for principal coordinate analysis (PCoA) and unweighted pair-group method with arithmetic mean (UPGMA) hierarchical clustering (Ramette, 2007). PCoA helps to obtain principal coordinates and visualize them from complex, multidimensional data. To gain deeper insights into microbial community composition differences between treatments, a permutational multivariate analysis of variance (PERMANOVA) was performed using the vegan package in R version 3.5.2, based on Bray-Curtis distance. This rigorous statistical approach involved 999 permutations to ensure the robustness and reliability of our results. Correlations between rhizosphere microbial communities and each soil factor were tested using the Mantel test from the vegan package. Furthermore, linear regression analyses were conducted between the first principal component (PC1) of fungal and bacterial community composition and the concentrations of TN and soil organic matter (SOM). Relationships between specific soil nutrients were investigated using Pearson correlations.
Co-occurrence networks were meticulously constructed, incorporating the dominant soil phylotypes from all organism types. Fungi and bacteria with average relative abundances exceeding 0.05% were selected to establish networks for the two experimental treatments. The network construction was based on robust correlations characterized by Spearman’s correlation coefficients (r) > 0.65 and false discovery rate (FDR)-corrected P-values < 0.001 (). Using the “igraph” package, the topological features of the networks were calculated, and all networks were visualized using the interactive Cytoscape v3.7.1 platform (https://cytoscape.org) (Shannon et al., 2003). The actual networks were juxtaposed with their randomized counterparts, which possessed identical numbers of nodes and edges. These randomized versions were generated using the Network Randomizer plugin incorporated in Cytoscape. Genera characterized by the highest betweenness centrality values (a metric measuring the importance of nodes in a network) were identified as keystone taxa (Vick-Majors et al., 2014). Network complexity was calculated as the averaged Z-score of multiple normalized topological parameters of microbial co-occurrence networks.
3 Results
3.1 Soil physicochemical properties, microbial biomass carbon, nitrogen, and enzyme activities
Under the same fertilization conditions, we analyzed soil physicochemical properties under two planting patterns (monoculture vs. intercropping). Compared with traditional walnut monoculture, intercropping reduced the contents of total nitrogen, total phosphorus, total potassium, available nitrogen, and available phosphorus (Supplementary Figures 1A–E), but significantly increased the content of soil organic matter (Supplementary Figure 1G). In addition, there was no significant difference in the content of available potassium between the two cropping systems (Supplementary Figure 1F). Furthermore, compared with monoculture, the contents of microbial biomass carbon (Supplementary Figure 2A) and microbial biomass nitrogen (Supplementary Figure 2B) were significantly higher in the walnut-soybean intercropping soil (Supplementary Figure 2). Moreover, soil enzyme activity analysis revealed that the activities of β-glucosidase, acid phosphatase (ACP), cellobiohydrolase, NAG, and LAP were significantly higher in the walnut-soybean intercropping soil than in the walnut monoculture soil (Supplementary Figures 3A–E).
3.2 Soil microbial community diversity and structures
After filtering out low-quality reads, a total of 1,387,916 high-quality reads were obtained. The total number of bases was 576,080,099, and the average read length was 415.13 bp. Rarefaction curves tended to approach saturation plateau in all six samples, indicating near-complete community sampling (Supplementary Figure 4). Subsequently, we analyzed diversity indices of soil fungal and bacterial communities under different cropping systems. The results indicated that two planting patterns (monoculture vs. intercropping) had no significant effects on the OTUs, Chao1, Shannon, Simpson, and Phylogenetic diversity indices of soil fungal and bacterial communities (Figure 2).
Figure 2
PCoA results based on microbial species abundance data, proving that walnut monoculture and walnut-soybean intercropping caused significant structural differentiation in rhizosphere bacterial and fungal communities (Figure 3). The first two principal component axes explained 55.28% (PCoA1) and 18.50% (PCoA2) of the variation in the fungal community (Figure 3A). Similar to that observed in soil fungi communities, different cropping systems also changed the bacterial community structure. The first two principal component axes explained 64.29% (PCoA1) and 18.70% (PCoA2) of the variation in the bacterial community (Figure 3B).
Figure 3
3.3 Composition and abundance of soil microbial communities
To characterize the overall composition of soil microbial communities, we analyzed the relative abundance of dominant phyla (defined as ≥1% relative abundance) in fungal and bacterial communities (Figures 3C, D). For fungi, Ascomycota and Basidiomycota abundances showed no obvious differences between monoculture and intercropping, whereas Mortierellomycota was significantly enriched under intercropping (Figure 4C). For the bacteria communities, the dominant taxa at the phylum level were primarily represented by Actinobacteriota, Proteobacteria, and Acidobacteriota. The relative abundance of Proteobacteria in intercropping was significantly higher than that in monoculture (Figure 4D). These results provide a macro-level overview of the differences in dominant microbial phyla between the two cropping systems.
Figure 4
LEfSe analysis (LDA ≥ 3.5, P < 0.05) was performed to screen differential microbial taxa across all taxonomic levels (phylum to species) (Supplementary Figure 5). For fungi, six taxonomic groups showed significant differences: Thermoascaceae, Nectriaceae, Agaricaceae, Agaricales, and Agaricomycetes were significantly enriched in the intercropping system, while Lophotrochus was significantly enriched in monoculture (Supplementary Figure 5A). For bacteria, four distinct taxonomic groups were identified: Proteobacteria, Gammaproteobacteria, and Enterobacteriaceae were significantly more abundant in intercropping, while Acidobacteriota was notably more abundant in monoculture (Supplementary Figure 5B). Consistent with the results in Figure 4, these findings demonstrate that planting regimes greatly reshape rhizosphere bacterial and fungal community structure via altering specific responsive taxa.
3.4 Relationships between microbial communities and soil physicochemical properties
Mantel tests revealed distinct environmental drivers shaping rhizosphere bacterial and fungal communities. Specifically, the bacterial community was only highly significantly correlated with SOM. In contrast, TN and SOM were significantly correlated with the fungal community. Additionally, SOM exhibited a significant negative correlation with TN, TP, AN, and AP, while demonstrating a significant positive correlation with MBN, BG, ACP, CBH, NAG, and LAP (Figure 4A). Similarly, linear regression analysis revealed that the community compositions of fungi and bacteria exhibited a weak negative correlation with SOM, while fungal community composition showed a moderate positive correlation with TN (Figure 4B). Notably, these trends are preliminary exploratory results owing to the limited sample size (n=6).
3.5 Principal component analysis and functional classification analysis of soil microorganisms under two planting patterns
PCA was applied to elucidate the differences in functional composition between the two planting patterns. Two principal components (PC1 and PC2) were selected and analyzed. For fungi, their contributions were 27.84% and 25.27%, respectively; for bacteria, the contributions were 62.29% and 21.45%, respectively. Samples under monoculture and intercropping treatments were largely overlapped in the PCA ordination space, with no obvious separation observed between groups (Figures 5A, C). We conducted further research on predicted functional categories closely related to bacterial and fungal populations. Through KEGG functional annotation analysis (Figures 5B, D), the results indicated that the most prominently enriched fungal guild was related to plant pathogens (Figure 5B). Specifically, the relative abundance of Plant_Pathogen and Plant_Pathogen-Soil & Wood_Saprotroph guilds was consistently high across both monoculture and intercrop systems, accounting for a substantial proportion of the total fungal community. Intercropping slightly lowered the abundance of Plant_Pathogen yet raised Undefined_Saprotroph, implying a predicted functional turnover of fungal communities. For bacterial communities across all treatments, metabolic pathways were inferred to be the primary functional category, followed by genetic information processing and environmental information processing pathways (Figure 5D). Both bacterial and fungal functional profiles are predicted rather than directly measured and should be interpreted with caution.
Figure 5
3.6 Differences in the soil fungi and bacteria networks under different cropping systems
To investigate the effects of different cropping systems on the soil microbial community networks, we analyzed the fungal and bacterial community networks at the OTU level for each cropping system (Figure 6). Both networks were divided into four modules based on topological features. As illustrated in Figures 6A, B, the fungal network had scattered nodes and loose connections, implying weaker interspecies associations, while the bacterial network showed tighter clustering and denser links, reflecting more intensive microbial crosstalk. Intercropping significantly elevated betweenness centralization only in the fungal network; bacterial edge numbers were markedly reduced under intercropping, with no parallel change in fungi. Other topological indices (node number, average degree, path length, clustering coefficient, density, diameter) showed no treatment differences for either kingdom (Supplementary Figures 6, S7). Furthermore, the complexity of the fungal and bacterial networks was assessed, revealing that the network complexity of bacteria was significantly higher than that of fungi (Figure 6C). However, no significant variation in network complexity was observed across different planting systems for either group (Figure 6D). These results provide exploratory clues for understanding the response characteristics of soil microbial networks to different cropping systems.
Figure 6
3.7 Soil ecosystem multifunctionality under different cropping systems
Intercropping significantly enhanced overall SMF. Under the averaging approach, intercropped plots exhibited a mean SMF of 0.54, compared with 0.43 in monoculture (P < 0.01), representing a 25.6% increase (Figure 7A). The multi-threshold approach, which calculates SMF as the number of functions simultaneously exceeding a given percentage of their observed maximum, yielded consistent rankings and significance levels across all tested thresholds (T = 25%, 50%, and 75%), confirming that the intercropping benefit is not attributable to any single function.
Figure 7
Figure 7B summarizes the associations between these SMF values and four microbial attributes. A multiple linear regression model incorporating the four attributes as predictors-bacterial PC1, bacterial Shannon diversity, fungal PC1, and fungal Shannon diversity-fit the SMF data with high explanatory power (adjusted R² = 0.98). Among these, bacterial PC1 emerged as the strongest driver, with a parameter estimate of +0.39 (P < 0.05). Interestingly, bacterial diversity showed a negative association with SMF (parameter estimate = -0.15, P < 0.1), suggesting that the bacterial groups most strongly promoted by intercropping were not necessarily the most diverse ones. For fungal attributes, both PC1 and diversity exerted positive but relatively smaller effects, with estimates of +0.10 and +0.25, respectively (both P < 0.1). Hierarchical partitioning of the explained variance revealed that bacterial attributes jointly accounted for 70.51% of the variation in SMF, whereas fungal attributes contributed 29.49%. This partitioning is consistent with the interpretation that bacteria serve as the primary microbial drivers of soil multifunctionality in the walnut–soybean intercropping system.
4 Discussion
4.1 Effects of intercropping legumes on soil physical and chemical properties
Intercropping, a time-honored agricultural practice in China, has been widely adopted since ancient times. Studies have demonstrated that cultivating intercropped species, such as Chinese chestnuts, fruit trees, and soybeans, in tea plantations enhances soil fertility and elevates tea quality (Li-Feng et al., 2013; Wen et al., 2019). Many studies have consistently shown that plant species diversity is essential for fostering soil fertility and facilitating the uptake of soil nutrients (; Tilman et al., 1996). Previous research has shown that trees integrated into agroforestry systems can significantly enhance soil physicochemical properties (Wang et al., 2022). Intercropping with legumes is considered a productive and sustainable system. In this study, legumes were chosen as intercropping plants. A key advantage is their ability to fix atmospheric nitrogen through biological nitrogen fixation, minimizing competition with other plants for soil N. Intercropping legumes can boost soil microbial activity and improve soil nutrient status via biological nitrogen fixation and root exudate input (; Shen and Lin, 2021). demonstrated that intercropping peanuts with tea plants significantly enhances soil fertility and exerts positive effects on soil health. Our results demonstrate that the walnut-soybean intercropping system led to a significant decrease in total nitrogen content relative to monoculture. It is important to clarify that soil health and fertility are not evaluated solely by inorganic nutrient concentrations (e.g., TN, TP, AN, AP) in farmland ecosystems under institutional fertilization (; Zhang et al., 2025). Instead, SOM content plays a more pivotal role in enhancing soil fertility, particularly in improving soil structure, water-holding capacity, and sustaining long-term productivity, and its practical importance in agricultural production substantially exceeds that of nitrogen and phosphorus. This supports our conclusion that intercropping enhances soil fertility (; ; ). The observed reduction in nitrogen and phosphorus levels in the intercropping system can be reasonably explained by enhanced microbial immobilization and nutrient cycling, rather than a decline in soil fertility (). To decompose the increased plant residues input by intercropping, microorganisms proliferate extensively; during this process, they assimilate additional nitrogen to balance their cellular carbon-nitrogen stoichiometry, thereby improving the efficiency of soil nitrogen retention and fixation (; ; Ray et al., 2025). This short-term reduction in inorganic nutrients (TN, TP, AN, AP) is a temporary phenomenon associated with microbial nutrient immobilization, which ultimately promotes long-term nutrient supply capacity by reducing nutrient leaching loss. Distinguishing between short-term nutrient pools (e.g., inorganic N, P) and long-term soil properties is essential for accurately evaluating the ecological impacts of intercropping. The increase in SOM induced by intercropping contributes more significantly to long-term soil health and serves as a key metric for enhancing carbon sequestration potential in agricultural ecosystems (; ; Morrow et al., 2016). The improvement in SOM content under intercropping aligns with findings from previous studies on legume-cereal intercropping systems ().
Meanwhile, this study revealed significantly higher levels of microbial biomass carbon and microbial biomass nitrogen in intercropped soils compared to monoculture systems. Elevated MBC levels signify vigorous microbial activity, enhanced organic matter decomposition, and efficient nutrient cycling processes (e.g., transformations of carbon, nitrogen, and phosphorus), which correlate with higher soil fertility (Mu-Chun et al., 2012). In addition, higher MBN content demonstrates the microbial capacity to effectively retain inorganic nitrogen, while reducing nitrogen loss and facilitating subsequent release of plant-available nitrogen during organic matter decomposition, further supporting the positive effect of intercropping on soil nutrient cycling (Ting-Ting et al., 2022).
Soil enzyme activity also plays a critical role in walnut yield and quality. Duan et al. demonstrated that intercropping significantly enhances the activities of urease, phosphatase, and invertase in tea garden soils (). The experimental results of this study demonstrated that walnut-soybean intercropping significantly enhanced the activities of five key soil enzymes (β-glucosidase, acid phosphatase, cellobiohydrolase, NAG, and LAP) compared to walnut monoculture systems. Numerous tea plantations have adopted legume-tea intercropping systems, which mitigate interspecific competition through enhanced complementarity/facilitation effects (Pokharel et al., 2023). This practice has been documented to increase tea yield, modify the physicochemical composition of tea leaves, and improve soil nutrient availability and enzymatic activities (; Sedaghathoor and Janatpoor, 2012). Compared with maize monoculture, legume-intercropped maize systems exhibit significantly increased carbon inputs and biological nitrogen fixation. These enhanced biogenic carbon and nitrogen inputs substantially stimulate soil enzyme activities and associated nutrient cycling processes, consequently improving soil fertility (Liu et al., 2024). These findings are consistent with our results demonstrating that walnut-soybean intercropping may positively influence soil fertility through enhanced soil enzyme activities.
4.2 Influence of intercropping legumes in shaping soil microbial community diversity
The observed shifts in aboveground plant species diversity significantly influence soil physicochemical properties, consequently driving divergent patterns in bacterial and fungal community diversity (Yarwood and Hgberg, 2017). Many studies have shown that intercropping crops with peas can enhance the diversity and structure of soil bacterial communities (Petkova et al., 2025). Microorganisms from distinct phylogenetic lineages differ in their response to environmental changes (; Tian et al., 2018; Zhang et al., 2021a, 2019). Thus, intercropping practices may influence soil microbiota compositions (). In the current study, the results indicated that intercropping systems had no significant effects on the richness and abundance of microbial communities, but they had a significant impact on the community structure. This phenomenon can be largely attributed to the relatively short duration of our intercropping experiment, which is insufficient to drive obvious variations in microbial alpha diversity. The results demonstrated that intercropping was not the primary factor affecting microbial richness and abundance. As confirmed by previous studies, other environmental factors such as crop species, nitrogen content, and soil nutrients (e.g., organic matter) may exert a greater influence on soil microbial richness and abundance (Ma et al., 2024). Additionally, bacteria are more responsive to soil chemical properties, whereas fungi rely more heavily on vegetation (; ; Zhang et al., 2023). By regulating soil physicochemical conditions and extracellular enzyme activities, walnut-soybean intercropping induced remarkable shifts in rhizosphere bacterial and fungal community composition. Such soil environmental changes facilitated the enrichment of certain beneficial functional microbial taxa under intercropping relative to walnut monoculture. Plant diversity, soil chemical properties and substrate quality collectively accounted for the observed microbial community variations (Yanqing et al., 2018). Proteobacteria, regarded as copiotrophic organisms, contribute to biocrust formation and soil stability (Yin et al., 2013). Among soil fungi, Ascomycota and Basidiomycota are particularly prominent, playing a crucial role in carbon cycling through the degradation of organic substances. Mortierellomycota and Rozellomycota, as key components of the rhizosphere microbiome, are induced by rhizobia and arbuscular mycorrhizae to enhance plant nutrient uptake (Wang et al., 2021). In this study, intercropping significantly increased the relative abundance of Ascomycota, Mortierellomycota, Proteobacteria, and Basidiomycota. Thus, the rational integration of legumes (e.g., walnut/soybean) in intercropping systems offers dual benefits: enhancing soil fertility (via increased organic matter and enzyme activity) and boosting microbial diversity.
Previous studies have consistently demonstrated that soil physicochemical properties act as critical mediators in shaping microbial community structure. Specifically, numerous bacterial taxa exhibit strong correlations with specific soil environmental factors, a characteristic that has enabled their widespread use as bioindicators for assessing soil quality and ecological status (; Val-Moraes et al., 2016). In this study, we found that the bacterial community was only significantly correlated with SOM, while the fungal community was correlated with both TN and SOM. In addition, studies have found that bacterial activity has also been affected by the “pre-activation effect” induced by intercropped leguminous crops (Olivier et al., 2017). By adding fresh organic matter derived from leguminous crops, an increase in the mineralization of SOM has been observed, which stimulates the activity of soil bacterial communities involved in the mineralization of stable organic matter (; ).
4.3 Intercropping system effects on microbial community network
Previous studies have demonstrated that various agricultural practices can significantly influence the composition of soil microbial communities (; Town et al., 2023). In this study, PCA revealed distinct differences in the structure of soil microbial communities between two planting patterns (monoculture vs. intercropping). These findings further support the notion that different microbial taxa respond differently to environmental perturbations. Moreover, predicted fungal guild analysis (FUNGuild, see Section 2.8) indicated an enrichment of the plant-pathogen guild across both treatments, consistent with previous findings that intercropping systems can alter soil microecological conditions and thereby influence the habitat suitability of pathogenic organisms (Sun et al., 2022). In contrast, predicted bacterial KEGG pathway analysis (PICRUSt2, see Section 2.7) suggested that metabolic pathways were the dominant functional category in bacterial communities, implying a potential role of bacteria in mediating soil nutrient cycling and energy transformation processes. Both bacterial and fungal functional profiles are predicted from taxonomic markers (16S for bacteria, ITS for fungi) and should be interpreted as tentative, hypothesis-generating inferences rather than directly measured functions.
Soil microbial ecological networks exhibit a high degree of complexity. To preliminarily explore the interrelationships between soil microorganisms and environmental factors, this study constructed a microbial co-occurrence network as an exploratory approach (). This approach defines the nodes within the soil microbial network as representative microbial taxa, and establishes linkages between fungal and bacterial communities (). Existing research has demonstrated that intercropping systems can modify the co-occurrence patterns of both bulk soil and rhizosphere microorganisms in arid ecosystems (Zhang et al., 2021b). Our preliminary findings also suggest that the co-occurrence network structures of soil fungi and bacteria may be influenced by different planting regimes, offering exploratory insights for future research. In microbial co-occurrence networks, modules are generally defined as clusters of nodes that exhibit strong internal correlations (or high internal co-variation) while displaying limited connectivity across clusters (or relatively few inter-group connections) (). This modular structure enhances the network’s resilience to external disturbances and contributes to overall system stability. In the present study, the co-occurrence networks of soil fungi and bacteria were each partitioned into four distinct modules. A high betweenness centralization value is generally indicative of a microorganism’s significant role in regulating community interactions or functional pathways (). Our results show that the intercropping system significantly increased the betweenness centralization of the fungal network, suggesting that key nodes in the fungal network may possess enhanced regulatory potential under intercropping conditions. Furthermore, edge density serves as a crucial metric for assessing network complexity (Shi et al., 2016). The results indicated that the edge density of the bacterial network was lower under the intercropping system than under the monoculture system; however, no significant difference was observed in the fungal network. Importantly, certain topological parameters, including the number of nodes and average degree, were not significantly influenced by planting regime in this study. The complexity of microbial co-occurrence networks is a key indicator of the intensity and extent of interactions among microbial taxa (). This study demonstrated that the bacterial network exhibited significantly higher complexity than the fungal network, and this pattern remained consistent across different planting systems. This observational finding suggests that the bacterial network may possess greater potential for stability and adaptability, although this inference requires validation through more in-depth functional experiments.
4.4 The impact of intercropping legumes plants on soil ecosystem multifunctionality
The peanut-cotton intercropping system has been shown to improve soil ecosystem multifunctionality (Xie et al., 2022), but whether this holds for walnut–soybean cropping remained unclear. Our data show that intercropping with soybean did raise SMF (0.54 vs. 0.43, P < 0.01), matching what has been observed in several other legume-based systems (Li et al., 2023). The effect was consistent under both SMF calculation approaches (averaging and multi-threshold), which argues against the idea that a single function is driving the improvement. One plausible explanation is that root-system spatial complementarity, together with more diverse litter inputs in the intercropping setup, facilitates nutrient turnover and carbon storage at the same time ().
Turning to the microbial side, bacterial PC1 came out as the strongest positive predictor of SMF (+0.39, P < 0.05), whereas bacterial diversity showed a negative relationship (-0.15, P < 0.1). The two fungal predictors were both positive but weaker (+0.10 and +0.25, both P < 0.1). According to hierarchical partitioning, bacteria accounted for 70.51% of the explained variation in SMF, versus 29.49% for fungi-reinforcing the view that bacteria are the main microbial drivers here. That makes sense given their known functions in residue decomposition, antimicrobial production (Mitra et al., 2022), and aggregate stabilization via hydrophobic proteins (). It is also worth noting that, since the SMF index was constructed from non-microbial indicators, this comparison between bacterial and fungal contributions is not biased by the microbial attributes used elsewhere in the paper.
4.5 Limitations and future research directions
This study has several limitations that should be acknowledged. First, the spatial design (one orchard with three replicate plots per treatment) limits the generalizability of the findings to other walnut-producing regions on the Loess Plateau, where microbial communities can vary with soil texture and microclimate. Second, the three-year observation window is too short to capture the longer-term dynamics of soil C and N pools or microbial succession under walnut-soybean intercropping. Third, although the revised SMF index now incorporates eleven functional indicators, additional variables such as soil respiration, aggregate stability, or nematode community metrics would further strengthen the multifunctionality evaluation in future work.
Notably, microbial co-occurrence network analysis entails inherent limitations: (1) correlation-based inference cannot distinguish direct biological interactions from indirect environmental co-occurrence, increasing the risk of spurious associations, especially with limited sample size; (2) static, cross-sectional networks fail to capture the spatiotemporal dynamics of microbial interactions; (3) the ecological mechanisms linking network topological features to ecosystem functions are not well defined, precluding direct functional interpretation; and (4) such networks cannot differentiate functionally essential connections from redundant ones, limiting their utility in assessing trade-offs among multifunctional outcomes. Therefore, these network-based findings should be interpreted as exploratory rather than conclusive.
5 Conclusion
This study conducted a systematic investigation of the walnut-soybean intercropping system. The results demonstrated that this planting pattern significantly increased soil organic matter content, microbial biomass, and enzyme activities. Furthermore, it markedly altered microbial β-diversity and community composition, and thereby enhanced the multifunctionality of the soil ecosystem. Notably, bacteria were identified as the key microbial group driving this enhancement in multifunctionality. Considering the limitations of this study, the results provide a preliminary theoretical basis and reference for the sustainable management of walnut orchards in the Loess Plateau region. Future research with more rigorous experimental design and long-term field observation is needed to further verify and supplement these findings.
Statements
Data availability statement
Raw data of 16S rRNA and ITS fungi have been deposited in the National Center for Biotechnology Information (NCBI) under the BioProject accession numbers in review PRJNA1397880 and PRJNA1397886, respectively.
Author contributions
JR: Conceptualization, Data curation, Funding acquisition, Methodology, Project administration, Resources, Writing – original draft, Writing – review & editing. X-FJ: Conceptualization, Data curation, Methodology, Validation, Visualization, Writing – review & editing. Y-NC: Conceptualization, Data curation, Methodology, Software, Visualization, Writing – review & editing. X-YL: Data curation, Formal analysis, Investigation, Writing – review & editing. Y-FW: Methodology, Software, Supervision, Validation, Writing – review & editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the National Natural Science Foundation of China (grant numbers 32260412) and Science and Technology Department of Gansu Province (grant numbers 25YFNA003).
Conflict of interest
The author(s) declared that this work 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.2026.1876922/full#supplementary-material
Supplementary Figure 1The effects of two planting pattern (monoculture vs. intercropping) on soil total nitrogen, total phosphorus, total potassium, available nitrogen, available phosphorus, and available potassium.
Supplementary Figure 2The effects of two planting pattern (monoculture vs. intercropping) on microbial biomass carbon and microbial biomass nitrogen.
Supplementary Figure 3The effects of two planting pattern (monoculture vs. intercropping) on β-glucosidase, acid phosphatase, cellobiohydrolase, N-acety-β-D-glucosaminidase (NAG), and leucine aminopeptidase (LAP).
Supplementary Figure 4The rarefaction curves of fungi (A) and bacteria (B) rRNA sequencing depth and number of species number in rhizosphere soil.
Supplementary Figure 5Evolutionary map of bacterial taxa (LDA ≥ 3.5, p < 0.05) was analyzed using LEfSe. To enhance clarity in visualization, taxonomic groups at the phylum or species level demonstrating mean relative abundance below 1% were aggregated and collectively designated as ‘Others’.
Supplementary Figure 6The effects of sole cropping and intercropping on topological parameters of soil fungal networks: (A) nodes; (B) edges; (C) average degree; (D) average path length; (E) betweenness centralization; (F) clustering coefficient; (G) graph density; (H) graph diameter.
Supplementary Figure 7The effects of sole cropping and intercropping on topological parameters of soil bacteria networks: (A) nodes; (B) edges; (C) average degree; (D) average path length; (E) betweenness centralization; (F) clustering coefficient; (G) graph density; (H) graph diameter.
Supplementary Table 1Variables used to compute the soil ecosystem multifunctionality (SMF) index. All variables were determined in this study using the methods described in Section 2.3. Each variable was Z-score standardized across all six samples prior to SMF calculation. TK, AK and pH were measured in this study but were not included in the SMF index because they are not direct indicators of ecosystem functions related to nutrient cycling, microbial biomass or carbon storage. The averaging approach uses the mean of the 11 standardized values as the SMF index; the multi-threshold approach counts, at each threshold T (25%, 50%, 75%), the number of variables that simultaneously exceed T% of their observed maximum.
References
1
AbdallahI. B.TliliN.Martinez-ForceE.RubioA. G.Perez-CaminoM. C.AlbouchiA.et al. (2015). Content of carotenoids, tocopherols, sterols, triterpenic and aliphatic alcohols, and volatile compounds in six walnuts (Juglans regia L.) varieties. Food Chem.173, 972–978. doi: 10.1016/j.foodchem.2014.10.095
2
AbdElgawadH.AbuelsoudW.MadanyM. M. Y.SelimS.ZintaG.MousaA. S. M.et al. (2020). Actinomycetes enrich soil rhizosphere and improve seed quality as well as productivity of legumes by boosting nitrogen availability and metabolism. Biomolecules10 (12), 1675. doi: 10.3390/biom10121675
3
ArmbrusterW. S.PélabonC.BolstadG. H.HansenT. F. (2014). Integrated phenotypes: understanding trait covariation in plants and animals. Philos. Trans. R. Soc Lond. B. Biol. Sci.369, 20130245. doi: 10.1098/rstb.2013.0245
4
BainardL. D.KochA. M.GordonA. M.KlironomosJ. N. (2013). Growth response of crops to soil microbial communities from conventional monocropping and tree-based intercropping systems. Plant Soil363, 345–356. doi: 10.1007/s11104-012-1321-5
5
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
6
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
7
BarnesA. P. H.H.Manevska-TasevskaG.G.ShresthaS. S.ThomsonS. G. (2015). The influence of diversification on long-term viability of the agricultural sector. Land Use Policy49, 404–412. doi: 10.1016/j.landusepol.2015.08.023
8
BhattiA. A.HaqS.BhatR. A. (2017). Actinomycetes benefaction role in soil and plant health. Microb. Pathog.111, 458–467. doi: 10.1016/j.micpath.2017.09.036
9
BilasR. D.BretmanA.BennettT. (2021). Friends, neighbours and enemies: an overview of the communal and social biology of plants. Plant Cell Environ.44, 997–1013. doi: 10.1111/pce.13965
10
BlagodatskayaЕ.KuzyakovY. (2008). Mechanisms of real and apparent priming effects and their dependence on soil microbial biomass and community structure: critical review. Biol. Fertil. Soils45, 115–131. doi: 10.1007/s00374-008-0334-y
11
BrookerR. W.BennettA. E.CongW. F.DaniellT. J.GeorgeT. S.HallettP. D.et al. (2015). Improving intercropping: a synthesis of research in agronomy, plant physiology and ecology. New Phytol.206, 107–117. doi: 10.1111/nph.13132
12
BrookesP. C.LandmanA.PrudenG.JenkinsonD. S. (1985). Chloroform fumigation and the release of soil nitrogen: A rapid direct extraction method to measure microbial biomass nitrogen in soil - ScienceDirect. Soil Biol. Biochem.17, 837–842. doi: 10.1016/0038-0717(85)90144-0
13
ByrnesJ.LefcheckJ. S.GamfeldtL.GriffinJ. N.IsbellF.HectorA. (2014). Multifunctionality does not imply that all functions are positively correlated. Proc. Natl. Acad. Sci.111, 351–363. doi: 10.1073/pnas.1419515112
14
CaiM.ZhangC.NdunguC. N.LiuG.LiuW.ZhangQ. (2024). Linking ecosystem multifunctionality to microbial community features in rivers along a latitudinal gradient. mSystems9, e0014724. doi: 10.1128/msystems.00147-24
15
CaporasoJ. G.KuczynskiJ.StombaughJ.BittingerK.BushmanF. D.CostelloE. K.et al. (2010). QIIME allows analysis of high-throughput community sequencing data. Nat. Methods7, 335–336. doi: 10.1038/nmeth.f.303
16
Castellano-HinojosaA.StraussS. L. (2020). Impact of cover crops on the soil microbiome of tree crops. Microorganisms8 (3), 328. doi: 10.3390/microorganisms8030328
17
ChaG.MeinhardtK. A.OrellanaL. H.HattJ. K.PannuM. W.StahlD. A.et al. (2021). The influence of alfalfa-switchgrass intercropping on microbial community structure and function. Environ. Microbiol.23, 6828–6843. doi: 10.1111/1462-2920.15785
18
CongW. F.HofflandE.LiL.SixJ.SunJ. H.BaoX. G.et al. (2015). Intercropping enhances soil carbon and nitrogen. Glob. Chang Biol.21, 1715–1726. doi: 10.1111/gcb.12738
19
CuarteroJ.PascualJ. A.VivoJ.-M.ZbolatO.Sánchez-NavarroV.Egea-CortinesM.et al. (2022). A first-year melon/cowpea intercropping system improves soil nutrients and changes the soil microbial community. Agric. Ecosyst. Environ.328, 107856. doi: 10.1016/j.agee.2022.107856
20
CurtrightA. J.TiemannL. K. (2021). Intercropping increases soil extracellular enzyme activity: A meta-analysis. Agric. Ecosyst. Environ.319, 107489. doi: 10.1016/j.agee.2021.107489
21
DalalR. C.AllenD. E.ChanK. Y.SinghB. P. (2011). Soil organic matter, soil health and climate change. 29, 87–106. doi: 10.1007/978-3-642-20256-8_5
22
DouglasG. M.MaffeiV. J.ZaneveldJ. R.YurgelS. N.BrownJ. R.TaylorC. M.et al. (2020). PICRUSt2 for prediction of metagenome functions. Nat. Biotechnol.38, 685–688. doi: 10.1038/s41587-020-0548-6
23
DuZ.ZhuY.ZhuL.ZhangJ.LiB.WangJ.et al. (2018). Effects of the herbicide mesotrione on soil enzyme activity and microbial communities. Ecotoxicol. Environ. Saf.164, 571–578. doi: 10.1016/j.ecoenv.2018.08.075
24
DuanY.WangT.LeiX.CaoY.LiuL.ZouZ.et al. (2024). Leguminous green manure intercropping changes the soil microbial community and increases soil nutrients and key quality components of tea leaves. Hortic. Res.11, uhae018. doi: 10.1093/hr/uhae018
25
Edward JohnstonA.PoultonP. R.ColemanK. (2009). Soil organic matter: its importance in sustainable agriculture and carbon dioxide fluxes. Adv. Agron.101, 1–57. doi: 10.1016/S0065-2113(08)00801-8
26
FarooqT. H.KumarU.MoJ.ShakoorA.WangJ.RashidM. H. U.et al. (2021). Intercropping of Peanut-Tea enhances soil enzymatic activity and soil nutrient status at different soil profiles in subtropical Southern China. Plants (Basel)10 (5), 881. doi: 10.3390/plants10050881
27
FontaineS.AbbadieL.AubertM.BarotS.BloorJ. M. G.DerrienD.et al. (2024). Plant-soil synchrony in nutrient cycles: Learning from ecosystems to design sustainable agrosystems. Glob. Chang Biol.30, e17034. doi: 10.1111/gcb.17034
28
FrancoJ. G.KingS. R.VolderA. (2018). Component crop physiology and water use efficiency in response to intercropping. Eur. J. Agron.93, 27–39. doi: 10.1016/j.eja.2017.11.005
29
FrossardE.BuchmannN.BünemannE. K.KibaD. I.LompoF.ObersonA.et al. (2016). Soil properties and not inputs control carbon, nitrogen, phosphorus ratios in cropped soils in the long-term. Soil2, 83–99. doi: 10.5194/soil-2-83-2016
30
FureyG. N.TilmanD. (2021). Plant biodiversity and the regeneration of soil fertility. Proc. Natl. Acad. Sci. U.S.A.118 (49), e2111321118. doi: 10.1073/pnas.2111321118
31
GaddG. M. (2021). Fungal biomineralization. Curr. Biol.31, R1557–r1563. doi: 10.1016/j.cub.2021.10.041
32
GaoC.BezemerT. M.de VriesF. T.van BodegomP. M. (2024). Trade-offs in soil microbial functions and soil health in agroecosystems. Trends Ecol. Evol.39, 895–903. doi: 10.1016/j.tree.2024.05.013
33
GeX.CaoY.ZhouB.XiaoW.LiM. H. (2020). Combined application of biochar and N increased temperature sensitivity of soil respiration but still decreased the soil CO2 emissions in moso bamboo plantations. Sci. Total Environ.730, 139003. doi: 10.1016/j.scitotenv.2020.139003
34
GuY.JiaoJ.XuH.ChenY.HeX.WuX.et al. (2025). Intercropping improves the yield by increasing nutrient metabolism capacity and crucial microbial abundance in root of Camellia oleifera in purple soil. Plant Physiol. Biochem.219, 109318. doi: 10.1016/j.plaphy.2024.109318
35
HuB.ZhangJ.XiaoJ.YangS.DongK.DongY. (2025). Long-term intercropping with nitrogen management to improve soil quality and control crop diseases. Plant Cell Environ.49. doi: 10.1111/pce.15384
36
HuangZ.CuiC.CaoY.DaiJ.ChengX.HuaS.et al. (2022). Tea plant-legume intercropping simultaneously improves soil fertility and tea quality by changing bacillus species composition. Hortic. Res.9, uhac046. doi: 10.1093/hr/uhac046
37
HussainM.HamidM. I.WangN.BinL.XiangM.LiuX. (2016). The transcription factor SKN7 regulates conidiation, thermotolerance, apoptotic-like cell death and parasitism in the nematode endoparasitic fungus Hirsutella minnesotensis. Rep.6, 30047. doi: 10.1038/srep30047
38
ImranOrtasI. (2025). Mechanism in soil health improvement with soil microbial actions and its role in potential agriculture. Commun. Soil Sci. Plant Anal.56, 2066–2087. doi: 10.1080/00103624.2025.2486597
39
JungersJ. M.DehaanL. H.MullaD. J.SheafferC. C.WyseD. L. (2019). Reduced nitrate leaching in a perennial grain crop compared to maize in the Upper Midwest, USA. Elsevier272: 63–73. doi: 10.1016/j.agee.2018.11.007
40
KarimB. H.AntonellaC.AnnamariaR.PedroG.-C.MarcoS.AntonioH. J.et al. (2021). Halophyte based Mediterranean agriculture in the contexts of food insecurity and global climate change. Environ. Exp. Bot.191, 104601. doi: 10.1016/j.envexpbot.2021.104601
41
KermahM.FrankeA. C.Adjei-NsiahS.AhiaborB. D. K.AbaidooR. C.GillerK. E. (2017). Maize-grain legume intercropping for enhanced resource use efficiency and crop productivity in the Guinea savanna of northern Ghana. Field Crops Res.213, 38–50. doi: 10.1016/j.fcr.2017.07.008
42
KuramaeE. E.YergeauE.WongL. C.PijlA. S.van VeenJ. A.KowalchukG. A. (2012). Soil characteristics more strongly influence soil bacterial communities than land-use type. FEMS Microbiol. Ecol.79, 12–24. doi: 10.1111/j.1574-6941.2011.01192.x
43
KuzyakovY. (2010). Priming effects: interactions between living and dead organic matter. Soil Biol. Biochem.42, 1363–1371. doi: 10.1016/j.soilbio.2010.04.003
44
LangilleM. G. I.ZaneveldJ.CaporasoJ. G. (2013). Predictive functional profiling of microbial communities using 16S rRNA marker gene sequences. Nat. Biotechnol.31, 814–821. doi: 10.1038/nbt.2676
45
LatatiM.PansuM.DrevonJ. J.OunaneS. M. (2013). Effect of Maize-common bean intercropping on nitrogen uptake. 2. IBEMPA Conference “Microorganisms for future agriculture”, Seville, Spain: Universitad de Sevilla, 527, 01268680.
46
LiX.ChuY.JiaY.YueH.HanZ.WangY. (2022). Changes to bacterial communities and soil metabolites in an apple orchard as a legacy effect of different intercropping plants and soil management practices. Front. Microbiol.13, 956840. doi: 10.3389/fmicb.2022.956840
47
LiY.NieC.LiuY.DuW.HeP. (2019). Soil microbial community composition closely associates with specific enzyme activities and soil carbon chemistry in a long-term nitrogen fertilized grassland. Sci. Total Environ.654, 264–274. doi: 10.1016/j.scitotenv.2018.11.031
48
LiG.ZhuS.LongJ.MaoH.DongY.HouY. (2023). Differences in microbial community structure and metabolic activity among tea plantation soils under different management strategies. Front. Microbiol.14, 1219491. doi: 10.3389/fmicb.2023.1219491
49
Li-FengM. A.Hong-JingC.Ying-JieS.Ming-BeiJ.Gen-MiaoZ.Lin-TuW. U.et al. (2013). Status and suggestions of tea garden fertilization on main green tea-producing counties in Zhejiang Province. J. Tea Sci.33 (1), 74–84.
50
LiuR.YangL.ZhangJ.ZhouG.ChangD.ChaiQ.et al. (2024). Maize and legume intercropping enhanced crop growth and soil carbon and nutrient cycling through regulating soil enzyme activities. Eur. J. Agron.159, 10. doi: 10.1016/j.eja.2024.127237
51
LuK. (1999). Analytical Methods of Soil and Agricultural ChemistryBeijing: China Agriculture Press.
52
MaH. Y.SurigaogeS.XuY.LiY. C.ChristieP.ZhangW. P.et al. (2024). Responses of soil microbial community diversity and co-occurrence networks to interspecific interactions in soybean/maize and peanut/maize intercropping systems. Appl. Soil Ecol.203, 10. doi: 10.1016/j.apsoil.2024.105613
53
MaestreF. T.QueroJ. L.GotelliN. J. (2012). Plant species richness and ecosystem multifunctionality in global drylands. Science335, 214–218. doi: 10.1126/science.1215442
54
MagoT.SalzbergS. L. (2011). FLASH: Fast length adjustment of short reads to improve genome assemblies. Bioinformatics27, 2957–2963. doi: 10.1093/bioinformatics/btr507
55
ManningP.FonsV. D. P.SoliveresS. (2018). Redefining ecosystem multifunctionality. Nat. Ecol. Evol.2, 427. doi: 10.1038/s41559-017-0461-7
56
MitraD.MondalR.KhoshruB.SenapatiA.RadhaT. K.MahakurB.et al. (2022). Actinobacteria-enhanced plant growth, nutrient acquisition, and crop protection:Advances in soil, plant, and microbial multifactorial interactions32, 22. doi: 10.1016/s1002-0160(21)60042-5
57
MohammadS. (2000). Long-term effects of fertilizers and integrated nutrient supply systems in intensive cropping on soil fertility, nutrient uptake and yield of rice. J. Agric. Sci.133, 365–370. doi: 10.1017/s0021859699007157
58
MorrowJ. G.HugginsD. R.Carpenter-BoggsL. A.ReganoldJ. P. (2016). Evaluating measures to assess soil health in long-term agroecosystem trials. Soil Sci. Soc Am. J.80, 450. doi: 10.2136/sssaj2015.08.0308
59
Mu-ChunY.Ting-TingX.Peng-HuiS.Jian-JunD. (2012). Effects of different cropping patterns of soybean and maize seedlings on soil enzyme activities and MBC and MBN. J. Northeast. Agric. Univ. (English Edition)19, 42–47. doi: 10.1016/s1006-8104(13)60049-5
60
NguyenN. H.SongZ.BatesS. T.BrancoS.TedersooL.MenkeJ.et al. (2016). FUNGuild: An open annotation tool for parsing fungal community datasets by ecological guild. Fungal Ecol.20, 241–248. doi: 10.1016/j.funeco.2015.06.006
61
OlivierD.Jean-FrançoisV.FlorianC. (2017). Intercropping with legume for agroecological cropping systems: Complementarity and facilitation processes and the importance of soil microorganisms. A review. Agric. Ecosyst. Environ.240, 148–161 doi: 10.1016/j.agee.2017.02.019
62
Paz-FerreiroJ.GascóG.GutiérrezB.MéndezA. (2012). Soil biochemical activities and the geometric mean of enzyme activities after application of sewage sludge and sewage sludge biochar to soil. Biol. Fertil. Soils48, 511–517. doi: 10.1007/s00374-011-0644-3
63
PetkovaM.ShilevS.PopovaV.NeykovaI.MinevN. (2025). Intercropping of oats with vetch conducts to improve soil bacteriome diversity and structure. Microorganisms13, 977. doi: 10.3390/microorganisms13050977
64
PokharelS. S.YuH.FangW.ParajuleeM. N.ChenF. (2023). Intercropping cover crops for a vital ecosystem service: A review of the biocontrol of insect pests in tea agroecosystems. Plants12 (12), 2361. doi: 10.3390/plants12122361
65
QiuQ.WuL.OuyangZ.LiB.XuY.WuS.et al. (2015). Effects of plant-derived dissolved organic matter (DOM) on soil CO2 and N2O emissions and soil carbon and nitrogen sequestrations. Appl. Soil Ecol.96, 122–130. doi: 10.1016/j.apsoil.2015.07.016
66
RametteA. (2007). Multivariate analyses in microbial ecology. FEMS Microbiol. Ecol.62, 142–160. doi: 10.1111/j.1574-6941.2007.00375.x
67
RayS.MaitraS.SairamM.SameerS.SagarL.DivyaB. S.et al. (2025). The nexus between intercropping systems, ecosystem services and sustainable agriculture: A review. Res. Crops26, 1–11. doi: 10.31830/2348-7542.2025.roc-1166
68
Rodriguez-RamosJ. C.ScottN.MartyJ.KaiserD.HaleL. (2022). Cover crops enhance resource availability for soil microorganisms in a pecan orchard. Agric. Ecosyst. Environ.337, 11. doi: 10.1016/j.agee.2022.108049
69
Saiya-CorkK. R.SinsabaughR. L.ZakD. R. (2002). The effects of long term nitrogen deposition on extracellular enzyme activity in an Acer saccharum forest soil. Soil Biol. Biochem.34, 1309–1315. doi: 10.1016/s0038-0717(02)00074-3
70
SalamA. K.SutantoE.DesviaY.NiswatiA.DermiyatiKimuraM. (1999). Activities of soil enzymes in fields continuously cultivated with cassava, sugarcane, and pineapple in middle terrace areas of Lampung Province, South Sumatra, Indonesia. Soil Sci. Plant Nutr.45, 803–809. doi: 10.1080/00380768.1999.10414329
71
SedaghathoorS.JanatpoorG. (2012). Study on effect of soybean and tea intercropping on yield and yield components of soybean and tea. J. Agric. Biol. Sci.7, 1–8.
72
ShannonP.MarkielA.OzierO.BaligaN. S.WangJ. T.RamageD.et al. (2003). Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res.13, 2498–2504. doi: 10.1101/gr.1239303
73
ShenF. T.LinS. H. (2021). Priming effects of cover cropping on bacterial community in a tea plantation. Multidisciplinary digital publishing Institute.13 (8) 4345. doi: 10.3390/su13084345
74
ShiS.NuccioE. E.ShiZ. J.HeZ.ZhouJ.FirestoneM. K. (2016). The interconnected rhizosphere: High network complexity dominates rhizosphere assemblages. Ecol. Lett.19, 926–936. doi: 10.1111/ele.12630
75
SunY.ChenL.ZhangS.MiaoY.ZhangY.LiZ.et al. (2022). Plant interaction patterns shape the soil microbial community and nutrient cycling in different intercropping scenarios of aromatic plant species. Front. Microbiol.13, 888789. doi: 10.3389/fmicb.2022.888789
76
TianJ.ZhuD.WangJ.WuB.HussainM.LiuX. (2018). Environmental factors driving fungal distribution in freshwater lake sediments across the Headwater Region of the Yellow River, China. Rep.8, 3768. doi: 10.1038/s41598-018-21995-6
77
TilmanD.WedinD.KnopsJ. M. H. (1996). Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature379, 718–720. doi: 10.1038/379718a0
78
Ting-TingX.An-DongC.Chang-AiL. U.Hong-LingY. E.Hong-LiangW. U.Sheng-ChangH.et al. (2022). Increasing soil microbial biomass nitrogen in crop rotation systems by improving nitrogen resources under nitrogen application21, 1488–1500. doi: 10.1016/s2095-3119(21)63673-0
79
TownJ. R.DumonceauxT.TidemannB.HelgasonB. L. (2023). Crop rotation significantly influences the composition of soil, rhizosphere, and root microbiota in canola (Brassica napus L.). Environ. Microbiome18, 40. doi: 10.1186/s40793-023-00495-9
80
Trinsoutrot-GattinI. (2021). Understanding arbuscular mycorrhizal colonization in walnut plantations: The contribution of cover crops and soil microbial communities. Agriculture12 (1), 1. doi: 10.3390/agriculture12010001
81
Val-MoraesS. P.de MacedoH. S.KishiL. T.PereiraR. M.NavarreteA. A.MendesL. W.et al. (2016). Liming in the sugarcane burnt system and the green harvest practice affect soil bacterial community in northeastern São Paulo, Brazil. Antonie van Leeuwenhoek109, 1643–1654. doi: 10.1007/s10482-016-0764-8
82
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
83
Vick-MajorsT. J.PriscuJ. C.Amaral-ZettlerL. A. (2014). Modular community structure suggests metabolic plasticity during the transition to polar night in ice-covered Antarctic lakes. ISME J.8, 778–789. doi: 10.1038/ismej.2013.190
84
WangX.FangL.BeiyuanJ.CuiY.PengQ.ZhuS.et al. (2021). Improvement of alfalfa resistance against Cd stress through rhizobia and arbuscular mycorrhiza fungi co-inoculation in Cd-contaminated soil. Environ. pollut.277, 116758. doi: 10.1016/j.envpol.2021.116758
85
WangQ.GarrityG. M.TiedjeJ. M.ColeJ. R. (2007). Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl. Environ. Microbiol.73, 5261–5267. doi: 10.1128/aem.00062-07
86
WangC.LiangQ.LiuJ.ZhouR.LangX.XuS.et al. (2023). Impact of intercropping grass on the soil rhizosphere microbial community and soil ecosystem function in a walnut orchard. Front. Microbiol.14, 1. doi: 10.3389/fmicb.2023.1137590
87
WangZ.ZhouM.LiuH.HuangC.MaY.GeH. X.et al. (2022). Pecan agroforestry systems improve soil quality by stimulating enzyme activity. PeerJ10, e12663. doi: 10.7717/peerj.12663
88
WenB.ZhangX.RenS.DuanY.FangW. (2019). Characteristics of soil nutrients, heavy metals and tea quality in different intercropping patterns. Agrofor. Syst.94, 963–974. doi: 10.1007/s10457-019-00463-8
89
XieW.ZhangK.WangX.ZouX.ZhangX.YuX.et al. (2022). Peanut and cotton intercropping increases productivity and economic returns through regulating plant nutrient accumulation and soil microbial communities. BMC Plant Biol.22, 121. doi: 10.1186/s12870-022-03506-y
90
YanqingG.XiaotianC.YuanyuanW.LuZ.ChengJ. (2018). Natural revegetation of a semiarid habitat alters taxonomic and functional diversity of soil microbial communities. Sci. Total Environ.635, 598–606. doi: 10.1016/j.scitotenv.2018.04.171
91
YarwoodS. A.HgbergM. N. (2017). Soil bacteria and archaea change rapidly in the first century of Fennoscandian boreal forest development. Soil Biol. Biochem.114, 160–167. doi: 10.1016/j.soilbio.2017.07.017
92
YinC.HulbertS. H.SchroederK. L.MavrodiO.MavrodiD.DhingraA.et al. (2013). Role of bacterial communities in the natural suppression of Rhizoctonia solani bare patch disease of wheat (Triticum aestivum L.). Appl. Environ. Microbiol.79, 7428–7438. doi: 10.1128/aem.01610-13
93
YueY.DianwuW.MiaomiaoC.XiaoquanM.YangY.BoL.et al. (2022). Distribution patterns of the earthworm community in relation to soil properties and metals in agro-ecosystems in Hebei Province, North China. Community Ecol.23, 389–399. doi: 10.1007/s42974-022-00116-4
94
ZhangZ.ChaiX.TariqA.ZengF.LiX.GracianoC. (2021b). Intercropping systems modify desert plant-associated microbial communities and weaken host effects in a hyper-arid desert. Front. Microbiol.12, 754453. doi: 10.3389/fmicb.2021.754453
95
ZhangW.ChenJ.FanY.HussainM.LiuX.XiangM. (2021a). The E3-ligase AoUBR1 in N-end rule pathway is involved in the vegetative growth, secretome, and trap formation in Arthrobotrys oligospora. Fungal Biol.125, 532–540. doi: 10.1016/j.funbio.2021.02.003
96
ZhangW.HuC.HussainM.ChenJ.XiangM.LiuX. (2019). Role of low-affinity calcium system member Fig1 homologous proteins in conidiation and trap-formation of nematode-trapping fungus Arthrobotrys oligospora. Sci. Rep.9, 4440. doi: 10.1038/s41598-019-40493-x
97
ZhangW. P.LiZ. X.GaoS. N.YangH.XuH. S.YangX.et al. (2023). Resistance vs. surrender: Different responses of functional traits of soybean and peanut to intercropping with maize. Field Crops Res.291, 11. doi: 10.1016/j.fcr.2022.108779
98
ZhangZ.XieD.TengW.GuF.ZhangR.ChengK.et al. (2025). A state of art review on carbon, nitrogen, and phosphorus cycling and efficient utilization in paddy fields. Plant Soil513, 1689–1709. doi: 10.1007/s11104-025-07344-3
99
ZhangG.YangZ.DongS. (2011). Interspecific competitiveness affects the total biomass yield in an alfalfa and corn intercropping system. Field Crops Res.124, 66–73. doi: 10.1016/j.fcr.2011.06.006
100
ZhaoC.ChaiQ.ZhaoY.MuY.ZhangY.YuA.et al. (2016). Interspecific competition and complementation is a function of N management in maize-pea intercropping systems. Crop Sci.56, 3286–3294. doi: 10.2135/cropsci2016.03.0204
Summary
Keywords
bacteria, fungi, intercropping, rhizosphere microbial community, soil multifunctionality, walnut
Citation
Ren J, Jiang X, Chen Y, Liu X and Wang Y (2026) Intercropping with soybean enhances walnut rhizosphere soil physicochemical conditions, microbial community composition, and ecosystem multifunctionality. Front. Agron. 8:1876922. doi: 10.3389/fagro.2026.1876922
Received
09 May 2026
Revised
07 August 2026
Accepted
18 August 2026
Published
11 September 2026
Volume
8 - 2026
Edited by
Antonio Rafael Sánchez-Rodríguez, University of Cordoba, Spain
Reviewed by
Shiqiang Ge, Chinese Academy of Agricultural Sciences, China
Rui Li, Chinese Academy of Sciences (CAS), China
Updates
Copyright
© 2026 Ren, Jiang, Chen, Liu and Wang.
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: Jing Ren, mailrenjing@163.com
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.