Introduction

Faba bean (Vicia Faba L.) is one of the important and oldest crops and is largely cultivated globally. It provides a significant protein source for animals and humans and has medicinal properties and health benefits (Rahate et al. 2021). According to FAO statistics, since 2016, more than 60 countries in the world have been cultivating faba bean. China contributes to 34.0% of the planting area, 36.1% of the total output, and 36.5% of the total consumption of faba bean globally (FAOSTAT 2019). However, with the advances in science and technology, the production under modern agriculture has become increasingly large-scale and intensive. Long-term continuous planting of faba bean leads to various problems including the occurrence of soil-borne diseases such as Fusarium wilt, which is an important factor that limits the production of faba bean (Yang et al. 2022c). Fusarium wilt promotes devastating damages in other legume species worldwide, as an example, pea (Pisum sativum L.) (Khalifa 1965), peanut (Arachis hypogaea L.) (Li et al. 2023), common bean (Phaseolus vulgaris L.) (Batista et al. 2017), chickpea (Cicer arietinum L.) (Jiménez-Díaz et al. 2015), lentil (Lens culinaris Medik.) (Campanella and Miceli 2021), Alfalfa (Medicago sativa L.) (Ramírez-Suero et al. 2010), Cowpea (Vigna unguiculata L.) (Smith et al. 1999), Lupin (Lupinus polyphyllus) (Armstrong and Armstrong 1964) and Soybean (Glycine max L.) (Armstrong and Armstrong 1965). Fusarium wilt is caused by Fusarium sp., which grows in saprophytic manner and survives in soil for several years without any host (Yang et al. 2022b; De Borbat et al. 2017; Bhar et al. 2021).

Fusarium activity can be regulated by plant allelopathy (Wu et al. 2009). Allelopathy is a natural process in which plants produce and release certain chemicals into environment to promote or inhibit the growth of themselves or other plants (Mahé et al. 2022). Autotoxicity is a specific form of allelopathy, in which the autotoxins are mainly secreted by the roots inhibiting own growth and promoting the occurrence of Fusarium wilt (Zhang et al. 2022). Benzoic acid is an autotoxin secreted by plants (Li et al. 2014a; Singh 2015). Compared with no p-hydroxybenzoic acid addition, exogenously added p-hydroxybenzoic acid could reduce the growth and promote the occurrence of Fusarium wilt in cucumber (Jin et al. 2020).

At present, soil-borne diseases are generally chemical, physical control and grafting cultivation in agricultural production. The tomato (Solanum lycopersicum L.) grafted onto rootstocks (Solanum torvum Sw.) reduced incidence rate of Fusarium wilt in tomato compared with nongrafted plants (Awu et al. 2023). However, because of the high cost of grafting technology, this method is not widely used for most plants (Awu et al. 2023; Davis et al. 2008). Exposure of soil to sun is a simple and effective method to prevent and control wilt in olive; however, it could easily lead to secondary colonization of pathogenic microorganisms and negatively impact the subsequent growth of crops (López-Escudero and Blanco-López 2001; Tjamos et al. 1991; Goicoechea 2009). Cucumber nurseries used 98% dazomet to disinfect, which greatly reduced the abundance of Fusarium (Mao et al. 2012). Although it eliminated soil pathogens, it also killed beneficial microorganisms and destroyed soil microecological balance to aggravate environmental pollution.

All of the aforementioned control methods for Fusarium wilt have some limitations. Therefore, the development of environmentally friendly management strategies for soil-borne diseases has been under focus (Wang et al. 2023). Intercropping involves cultivation of two or more crops together, which is an environmentally friendly and efficient cultivation method that suppress soil-borne diseases in actual production (Li et al. 2014b; Li et al. 2021a; He et al. 2016). Effective mitigation of soil-borne diseases via intercropping has been observed in onion and tomato (Zhou et al. 2023), watermelon and wheat (Xu et al. 2015a), rice and watermelon (Ren et al. 2008), and corn and soybean (Gao et al. 2014). Soil microorganisms play a crucial role in soil biogeochemical processes, and their diversity and community composition determine soil functions (Xu et al. 2021). Soil microbial community plays a role in the inhibition or promotion of soil-borne diseases (Cheng et al. 2021; Wei et al. 2015). Intercropping can promote the growth of beneficial microorganisms to inhibit the soil-borne diseases. Compared with soybean monocropping, corn and soybean intercropping could increase the Shannon index of bacterial community; change bacterial community structure and composition; and increase the abundance of beneficial bacteria such as Streptomyces sp., Bacillus sp., Pseudomonas sp., and Microbacterium sp. in the rhizosphere soil of soybean and could reduce the occurrence of root rot in soybean (Chang et al. 2022). Compared with V. villosa monocropping, intercropping with V. villosa and banana could reduce the Shannon index of fungal community, change the fungal community structure and composition, reduce the relative abundance of Fusarium in the rhizosphere soil of banana, thus reducing the occurrence of Fusarium wilt in banana (Yang et al. 2022a). Plant roots can produce exudates and secondary metabolites and strongly impact pathogens causing soil-borne diseases and beneficial microorganisms in the rhizosphere (Wu et al. 2016; Rolfe et al. 2019).

However, studies are limited on how intercropping can promote the growth of beneficial microorganisms to suppress the occurrence of soil-borne diseases under the stress of autotoxic substances. Moreover, only few studies are available assessing the synergistic effect of F. commune and autotoxins and alleviating effect of faba bean–wheat intercropping on the occurrence of Fusarium wilt in faba bean.

In our previous studies, we demonstrated that benzoic acid is the main autotoxic substance secreted by faba bean roots, and it stably exists in soil (Lv et al. 2020; Yang et al. 2022c). In this study, we aimed to explore the changes in microbial community in rhizosphere soil of faba bean from the perspective of the key interaction between the host and pathogenic fungus. The main aims of the present study were to assess (1) the synergistic effect of F. commune and benzoic acid and alleviating effect of faba bean–wheat intercropping on the occurrence of Fusarium wilt in faba bean, (2) the effects of F. commune and benzoic acid stress on the diversity and composition of microbial communities in the rhizosphere and the regulatory effect of faba bean–wheat intercropping, and (3) to determine the relationship between the amino acid content and active microbial flora in the rhizosphere soil of faba bean and to identify potential microbial taxa related to the changes in amino acid content.

Materials and methods

Test materials

Faba bean and wheat varieties used in the experiment were “Yunmai 53” and “89–147,” respectively, which were purchased from the Agricultural science of Yunnan Province. The fungal pathogen F. commune was isolated from the infected faba bean plant as described previously (Zhang et al. 2023b) and cultivated on glucose potato agar (PDA; distilled water 1 L, agar 20 g·L−1, potato 200 g·L−1, and glucose 20 g·L−1) for 7 days at 28 °C. The spores were collected from the culture plate via filtration through four layers of gauze, and their suspension was prepared at a concentration of 1 × 106 CFU·mL−1 to inoculate the plants (Sharma et al. 2005).

Experimental design

From October 2020 to May 2021, field experiments were conducted in Efeng Village, Eshan County, Yuxi City, Yunnan Province (24° 11′ N, 102° 24′ E; 1540 m a.s.l.). According to Yang et al., the content of autotoxins benzoic acid and its derivatives (benzoic acid and p-hydroxybenzoic acid) was 32.52 μg∙g−1 in the rhizosphere soil of faba bean when faba bean was continuous cultivated for 7 years (Yang et al. 2023). We selected the soil (noncontinuous cropping soil) that had not been planted with faba bean for 7 years at a distance of 20 m to avoid biases due to geographical factors. Soil was randomly sampled from 5 locations at a depth of 10–30 cm. From September to December 2021, the collected noncontinuous cropping soil was used in the greenhouse of Yunnan Agricultural University for the pot experiments. The basic chemical properties of soil before planting were as follows: organic matter 20.2 g∙kg−1, available phosphorus 31.64 g∙kg−1, available potassium 107.74 g∙kg−1, available nitrogen 94.26 g∙kg−1, and pH 6.63.

The trial has a completely randomized multi-factor block design and involved following six treatments:

  1. (1)

    CK-M: faba bean monocropping (M) without F. commune inoculation and benzoic acid addition

  2. (2)

    MF: faba bean monocropping (M) with F. commune inoculation without benzoic acid addition

  3. (3)

    MF100B: faba bean monocropping (M) with F. commune inoculation and addition of 100 mg/L benzoic acid

  4. (4)

    CK-I: faba bean–wheat intercropping (I) without F. commune inoculation and benzoic acid addition

  5. (5)

    IF: faba bean–wheat intercropping (I) with F. commune inoculation and without benzoic acid addition

  6. (6)

    IF100B: faba bean–wheat intercropping (I) with F. commune inoculation with the addition of 100 mg/L benzoic acid

Each treatment was performed using five black plastic basins (top diameter 21 cm, bottom diameter 16 cm, and height 15 cm) filled with 3.5 kg of noncontinuous cropping soil, and each basin had six plants. The trial was repeated 3 times.

In total, evenly sized 900 faba bean seeds and 400 wheat seeds were placed in 10% (v/v) H2O2 for 30 min for germination in dark for 12 h. Further, the seeds were placed on a dark porcelain dish containing saturated CaSO4 for 48 h. The sprouted faba bean and wheat seeds were placed in a pre-sterilized quartz sand to keep them fully moist until the wheat seedlings reached the trileaf stage (20 days after sowing) and the faba bean seedlings exhibited 4–6 true leaves (15 days after sowing). Therefore, wheat seedlings were sown 5 days before faba bean seedlings so that they could be transplanted on the same day. After, the faba beans and wheats together grow 45 days to harvest at the seedling stage and early stage of onset. Under monocropping, 6 faba bean seedlings were planted. Under intercropping, 3 faba bean and 3 wheat seedlings were planted. The planting was performed according to the method by Yang et al. (2022c). According to the treatment, 1 mL of F. commune spore suspension (1 × 106 CFU·mL−1) and 1 L of benzoic acid (100 mg·L−1) were added to the non-rhizosphere soil. Benzoic acid solution was added every 2 days until the faba bean plants were harvested (45 days after transplanting, early stage of onset) (Tian et al. 2019a). The greenhouse was maintained under natural light with relative humidity of 70%–85% and temperature of 26/19 °C day/night.

Measurement of faba bean growth parameters and the disease index of Fusarium wilt

After 45 days of transplanting, 3 faba bean plants were randomly selected to measure the root length and plant height of faba bean plants, and 15 faba bean plants were selected from each treatment for the investigation of Fusarium wilt. The faba bean plants with Fusarium wilt were classified according to 5 classification criteria (Yang et al. 2022b): Grade 0: asymptomatic; Grade 1: local lesions or slight discoloration of the root or stem base (except for the main root); Grade 2: diseased spots on the main lateral root or stem base but not contiguous; Grade 3: 1/3–1/2 of the root or stem base with disease spots, discoloration or rot, and lateral roots significantly reduced; Grade 4: most of the roots discolored and rotten or the stem base surrounded by disease spots; and Grade 5: plants wither and die.

The disease index was calculated as follows:

$$\textrm{Disease}\ \textrm{index}=\frac{\Sigma \left(\textrm{number}\ \textrm{of}\ \textrm{diseased}\ \textrm{plants}\ \textrm{at}\ \textrm{all}\ \textrm{levels}\times \textrm{corresponding}\ \textrm{grade}\ \textrm{value}\right)}{\textrm{highest}\ \textrm{value}\times \textrm{total}\ \textrm{number}\ \textrm{of}\ \textrm{investigated}\ \textrm{plants}}\times 100\%$$

Collection of rhizosphere soil of faba bean from plastic basins

After 45 days of faba bean transplantion, rhizosphere soil samples were collected as per the method by Yang et al. (2022c). Briefly, the root was gently shaken to remove large pieces of soil, and the remaining soil attached to the root was brushed and collected in a bag. Rhizosphere soil of 15 faba bean plants was mixed and quickly stored at −80 °C for the extraction of soil microbial DNA.

DNA extraction and PCR amplification

The soil microbial DNA was extracted from 0.5 g frozen soil using the E.Z.N.A®® Soil DNA Kit (Omega Bio-Tek, Norcross, GA, USA) according to the manufacturer’s instructions. The quality and integrity of extracted DNA were confirmed using 1% agarose gel electrophoresis, and the concentration was assessed using Nanodrop ND-2000 UV–Vis spectrophotometer (Thermo Scientific, Wilmington, WA, USA). Bacterial 16S rRNA and fungal ITS rRNA genes were amplified in triplicates using a thermocycler PCR system (GeneAmp 9700, ABI, Carlsbad, CA, USA) with primer sets 338F (5′-ACTCCTACGGGAGGCAGCAG-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′) to target the V3-V4 hyper variable regions of the bacterial 16S rRNA genes (Derakhshani et al. 2016) and ITS1F (5′- CTTGGTCATTTAGAGGAAGTAA-3′) and ITS2R (5’-GCTGCGTTCTTCATCGATGC-3′) to target the ITS1 regions of the fungal ITS rRNA genes (Bokulich and Mills 2013). The reaction mixture (20 μL) for the bacterial genes included 10 ng of template DNA, 0.8 μL of each primer (5 μM), 2 μL of 2.5 mM dNTPs, 4 μL of 5× FastPfu Buffer (Takara, Dalian, China), 0.4 μL of FastPfu Polymerase (Takara, Dalian, China), 0.2 μL of BSA, and 1.8 μL of RNase-free ddH2O. The reaction mixture (20 μL) for the fungal genes included 10 ng of template DNA, 0.8 μL of each primer (5 μM), 2 μL of 2.5 mM dNTPs, 2 μL of 10× Buffer (Takara, Dalian, China), 0.2 μL of rTaq Polymerase (Takara, Dalian, China), 0.2 μL of BSA, and 4 μL of RNase-free ddH2O. The PCR conditions were as follows: initial denaturation at 95 °C for 3 min; 30 cycles of 30 s at 95 °C, annealing for 30 s at 55 °C, and elongation for 45 s at 72 °C; and final extension at 72 °C for 10 min. The PCR products were extracted using 2% agarose gel electrophoresis, purified using the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, USA), and quantified using QuantiFluor™ -ST (Promega, USA) according to the manufacturer’s instructions.

Illumina Miseq sequencing and data processing

Purified amplicons were used in equimolar concentrations and paired-end sequenced (2 × 300) on an Illumina MiSeq platform (Illumina, San Diego, CA, USA). Forward and reverse sequences were merged from overlapping paired-end reads using FLASH (V1.2.7, http://ccb.jhu.edu/ software/FLASH/). All sequence reads with the same tag were assigned according to the same sample and the unique barcodes (raw tags). The raw tags were strictly filtered, and quality of the clean tags was assessed using QIIME Pipeline (V1.9.1, http://qiime.org/index.html). Briefly, an average quality score of <20 and ambiguous sequences shorter than 200 bp were identified and removed using UCHIME algorithm (http://www.drive5.com/usearch/manual/uchime algo.html). The effective sequences of bacteria and fungi were clustered into operational taxonomic units (OTUs) at 97% sequence similarity using UPARSE-OTU and UPARSE-OUT REF algorithms of the UPARSE software package (Uparse v7.0.1001, http://drive5.com/uparse/). Finally, the RDP classifier (http://rdp.cme.msu.edu/) was used to assign a representative sequence to the microbial taxa against SILVA 128 (bacteria, https://www.arb-silva.de/) and Unite (fungi, http://unite.ut.ee) using a confidence threshold of 70%. The generated raw sequence data were deposited in NCBI with the accession number PRJNA1000084.

Measurement of amino acid content in rhizosphere soil

Standard solutions (20 μmol/mL) of 22 amino acids were prepared by adding 10 mg of each standard to 0.1 mol/L HCl. The amino acids included alanine (Ala), glutamic acid (Glu), histidine (His), isoleucine (Ile), arginine (Arg), leucine (leu), aspartate (Asn), phenylalanine (Phe), aspartate (Asp), cystine (Cys), glutamine (Gln), cysteine (Cys-Cys), glycine (Gly), hydroxyproline (Hyd), proline (Pro), tryptophan (Trp), lysine (Lys), threonine (Thr), methionine (Met), valine (Val), serine (Ser), and tyrosine (Tyr). The standard solution of 25 μL was mixed with 1 mL of 90% acetonitrile. This was used to prepare the serial dilutions of 0.02, 0.05, 0.1, 0.2, 0.5, 50, 100, 200, 10, 20, 50, 100, and 200 μmol/L.

The rhizosphere soil of faba bean (250 mg) was added to 800 μL of 50% acetonitrile solution. The suspension was ground in a freezing grinder (−10 °C, 50 Hz) for 6 min and centrifuged at 13,000 rcf and 4 °C for 5 min. To 500 μL supernatant, 50 μL 50% acetonitrile solution was added and centrifuged at 4 °C and 13,000 rcf for 5 min. The supernatant was used for the further assessments. The target amino acids were qualitatively and quantitatively detected using the LC-ESI-MS/MS (UHPLC-Qtrap). The chromatographic conditions were as follows: the ExionLC AD system liquid chromatography with Waters BEH Amide (100 × 2.1 mm, 1.7 μm) column, temperature of column: 35 °C, velocity of flow: 1.0 mL/min, and injection volume: 2 μL. The mobile phase A (0.4% formic acid and 20 mM ammonium formate in 95% acetonitrile solution) and mobile phase B (0.4% formic acid and 20 mM ammonium formate in 5% acetonitrile solution) were used for gradient elution [A: 100% (0 min) → 90% (1 min) → 85% (2.6 min) → 70% (3.5 min) → 70% (4 min) → 100% (4.1 min) → 100% (6 min) and B: 0% (0 min) → 10% (1 min) → 15% (2.6 min) → 30% (3.5 min) → 30% (4 min) → 0% (4.1 min) → 0% (6 min)]. The mass spectroscopy conditions were as follows: AB SCIEX QTRAP 6500+ adopts positive mode detection, Ion Source Gas1 (GS1) and Ion Source Gas2 (GS2): 70, Collision Gas (CAD): Medium, Curtain Gas (CUR) IS: 35, the temperature (TEM): 350, and the ionSpray Voltage (IS): 5500. The standard linear regression curve was plotted with the analyte (amino acid) concentration as the abscissa and the analyte mass peak area as the ordinate. The amino acid concentration was calculated by adding the mass spectrum peak area of the amino acids in the linear equation.

Statistical analysis

The microbial community diversity and richness indices (Shannon and Ace indices), for various taxonomic levels, were generated using the Mothur program. With the R programming language (version 3.1.0), the relationships between abundant microbial taxa obtained from Spearman’s rank correlation coefficients and soil amino acids were established using heat map analysis and visualized using principal coordinates analysis (PCoA) based on the Bray–Curtis distance dissimilarity. The Canoco software (ver 5.0) was used to perform redundancy analysis (RDA). Each dataset was tested using a normal probability plot from the variance homogeneity in SPSS 18.0 (IBM Inc.). A multi-factor way analysis of variance (ANOVA) was used to analyze the data. All the treatment combinations were not interactive effect. The significance of difference was assessed using a least significant difference (LSD) test, with p ≤ 0.05 considered significant. The data were expressed as mean ± standard deviation.

Results

Effects of intercropping on the Fusarium wilt disease index under F. commune and benzoic acid stress

The MF treatment significantly increased the disease index compared with the CK-M treatment (p < 0.05) (Fig. 1). The MF100B treatment further significantly increased the disease index compared with the MF treatment (p < 0.05) (Fig. 1).

Fig. 1
figure 1

Effects of intercropping on the Fusarium wilt disease index under Fusarium commune and benzoic acid stress. The data were expressed as mean ± standard deviation from three biological replicates. Lowercase letters indicate significance at p < 0.05

Full size image

Compared with the MF and MF100B treatments, respectively, the IF and IF100B treatments significantly reduced the disease index (p < 0.05) (Fig. 1).

Effects of intercropping on the growth of faba bean under F. commune and benzoic acid stress

The MF treatment significantly reduced the root length and plant height of faba bean compared with the CK-M treatment (p < 0.05) (Fig. 2). The MF100B treatment further significantly reduced the root length and plant height of faba bean compared with the MF treatment (p < 0.05) (Fig. 2).

Fig. 2
figure 2

Effects of intercropping on the growth of faba bean under Fusarium commune and benzoic acid stress. A: Plant height. B: Root length. The data were expressed as mean ± standard deviation from three biological replicates. Lowercase letters indicate significance at p < 0.05

Full size image

Compared with the CK-M, MF, and MF100B treatments, the CK-I, IF, and IF100B treatments significantly increased the root length and plant height of faba bean (p < 0.05) (Fig. 2).

Effects of intercropping on the microbial community structure in the rhizosphere soil of faba bean under F. commune and benzoic acid stress

Using Illumina Miseq platform, 1,001,914 bacterial 16S rRNA sequences and 887,219 fungal ITS gene sequences were sequenced with a total of 4603 and 3862 OTUs and an average fragment length of 413 and 240 bp at 97% sequence similarity level in all samples, respectively (Fig. 3A and B). In all bacterial and fungal samples, the number of OTUs tended to saturate in the sparse curve corresponding to 97% sequence similarity and did not increase (Fig. 3). Therefore, the number of sequenced sequences was sufficient to evaluate the fungal and bacterial community diversity in all samples. Further, the microbial community structure was analyzed by flattening according to the minimum sequence number.

Fig. 3
figure 3

Effects of intercropping on the microbial community structure in the rhizosphere soil of faba bean under Fusarium commune and benzoic acid stress. A: Bacteria. B: Fungi. M: monocropping control, MF: monocropping with Fusarium commune inoculation, MF100B: monocropping with Fusarium commune inoculation and addition of 100 mg/L benzoic acid, I: intercropping control, IF: intercropping with Fusarium commune inoculation, IF100B: intercropping with Fusarium commune inoculation and addition of 100 mg/L benzoic acid

Full size image

Effects of intercropping on the alpha diversity of microbial communities in the rhizosphere soil of faba bean under F. commune and benzoic acid stress

The MF100B treatment significantly reduced the Shannon index of bacterial community in the rhizosphere soil compared with the MF treatment (p < 0.05) (Table 1). Compared with the MF100B treatment, the IF100B treatment significantly increased the Shannon index of bacterial community in the rhizosphere soil (p < 0.05) (Table 1).

Table 1 Effects of intercropping on the alpha diversity of microbial communities in the rhizosphere soil of faba bean under Fusarium commune and benzoic acid stress
Full size table

Effects of intercropping on the beta diversity of microbial communities in the rhizosphere soil of faba bean under F. commune and benzoic acid stress

The bacterial and fungal community in the rhizosphere soil was significantly separated on the PC1 axis in the MF100B treatment compared with the MF treatment (ANOSIM, R = 0.6099 and 0.6218, respectively, p = 0.001) (Fig. 4). The result revealed that F. commune and benzoic acid stress significantly differentially affected fungal and bacterial community structures in the rhizosphere soil of faba bean. Under F. commune and benzoic acid stress, faba bean–wheat intercropping exhibited restorative effects on the fungal and bacterial communities in the rhizosphere soil of faba bean (ANOSIM, R = 0.6218 and 0.6099, respectively, p = 0.001) (Fig. 4).

Fig. 4
figure 4

Effects of intercropping on the beta diversity of microbial communities in the rhizosphere soil of faba bean under Fusarium commune and benzoic acid stress. A: Bacteria. B: Fungi. Beta diversity was visualized using principal coordinate analysis based on Bray–Curtis distance dissimilarity. M: monocropping control, MF: monocropping with Fusarium commune inoculation, MF100B: monocropping with Fusarium commune inoculation and addition of 100 mg/L benzoic acid, I: intercropping control, IF: intercropping with Fusarium commune inoculation, IF100B: intercropping with Fusarium commune inoculation and addition of 100 mg/L benzoic acid

Full size image

Effects of intercropping on the microbial community composition at phylum and genus levels in the rhizosphere soil under F. commune and benzoic acid stress

Classifying the data obtained from Illumina MiSeq sequencing at a similarity level of 97%, 33 bacterial and 17 fungal phyla were detected. At the phylum level, the top-10 bacterial phyla with the highest relative abundance were selected for analysis; they were Actinomycetota, Actinomycetota, Chloroflexota, Patescibacteria, Acidobacteriota, Gemmatimonadota, Bacteroidota, Myxococcota, WPS-2, and GAL15. The top-5 fungal phyla with the highest relative abundance were Ascomycota, Basidiomycota, Mortierellomycota, Glomeromycota, and unclassified_k_Fungi. The MF100B treatment significantly increased the relative abundance of Ascomycota compared with the MF treatment (p < 0.05) (Fig. 5B).

Fig. 5
figure 5

Effects of intercropping on the microbial community composition at phylum and genus levels in the rhizosphere soil under Fusarium commune and benzoic acid stress. A: Phylum level of bacteria (the top-10 phyla were selected with the highest average abundance). B: Phylum level of fungi (the top-5 phyla were selected with the highest average abundance). C: Genus level of bacteria (the top-10 genera were selected with the highest average abundance). D: Genus level of Fungi. M: monocropping control, MF: monocropping with Fusarium commune inoculation, MF100B: monocropping with Fusarium commune inoculation and addition of 100 mg/L benzoic acid, I: intercropping control, IF: intercropping with Fusarium commune inoculation, IF100B: intercropping with Fusarium commune inoculation and addition of 100 mg/L benzoic acid. The data were expressed as mean ± standard deviation from three biological replicates. Lowercase letters indicate significance at p < 0.05

Full size image

Data obtained from Illumina MiSeq sequencing were classified at a similarity level of 97%, and 719 bacterial and 838 fungal genera were detected. The top-10 bacterial genera with the highest relative abundance were norank_f_norank_o_norank_c_AD3, unclassified_f_Micrococcaceae, Sphingomonas, Gemmatimonas, TM7a, HSB_ OF53-F07, norank_f_norank_o_norank_c_TK10, Blastococcus, Burkholderia Caballeronia Paraburkholderia, and Bradyrhizobium. The MF100B treatment significantly increased the relative abundance of unclassified_ f_Micrococcaceae, TM7a, and Burkholderia and reduced that of Sphingomonas, Gemmatimonas, norank_ f_ norank_ o_ norank_ c_ TK10, and Blastococcus compared with the MF treatment (p < 0.05) (Fig. 5C). Compared with the MF100B treatment, the IF100B treatment significantly increased the relative abundance of Sphingomonas, Blastococcus, and Bradyrhizobium and reduced that of norank_f_ Norank_o_Norank_c_AD3, unclassified_ f_ Micrococcaceae, TM7a, and Burkholderia (p < 0.05) (Fig. 5C). At the genus level, the top-10 fungal genera with the highest relative abundance were Mortierella, Fusarium, Cladosporium, unclassified_o_Chaetotyriales, Plectosphaerella, Penicillium, Saitozyma, Entrophospora, Gibberella, and unclassified_o_Helotiales. The MF treatment significantly increased the relative abundance of Fusarium and Penicillium in the rhizosphere soil compared with the CK-M treatment. The MF100B treatment significantly reduced the relative abundance of unclassified_o_Chaetothyriales and Plectosphaerella compared with the MF treatment (p < 0.05) (Fig. 5D). Compared with the MF100B treatment, the IF100B treatment significantly reduced the relative abundance of Fusarium and significantly increased that of Entrophospora (p < 0.05) (Fig. 5D).

Effects of intercropping on the content of amino acids in the rhizosphere soil of faba bean under F. commune and benzoic acid stress

Compared with the CK-M treatment, the MF treatment significantly increased the content of Asp, Glu, Thr, Asn, Gly, Tyr, Ile, Pro, Leu, Val, Ala, Phe, and Gln in the rhizosphere soil of faba bean. Compared with the MF treatment, the MF100B treatment significantly increased the content of Ala, Ile, Leu, Glu, Phe, Thr, Ser, Trp, Asn, Met, Asp, Gly, Gln, Val, Pro, and Tyr in the rhizosphere soil of faba bean (Fig. 6).

Fig. 6
figure 6

Effects of intercropping on the content of amino acids in the rhizosphere soil of faba bean under Fusarium commune and benzoic acid stress. M: monocropping control, MF: monocropping with Fusarium commune inoculation, MF100B: monocropping with Fusarium commune inoculation and addition of 100 mg/L benzoic acid, I: intercropping control, IF: intercropping with Fusarium commune inoculation, IF100B: intercropping with Fusarium commune inoculation and addition of 100 mg/L benzoic acid. The mean values and standard deviations. Lowercase was indicate for significance from statistics (P < 0.05)

Full size image

The IF treatment significantly increased Asp content and significantly reduced Thr, Asn, Glu, Pro, and Gln contents compared with the MF treatment (Fig. 6). The IF100B treatment significantly increased Asp content and significantly reduced Tyr, Thr, Ser, Pro, Asn, Glu, Gly, Leu, Ile, Met, Val, Ala, Phe, Trp, and Gln contents compared with the MF100B treatment (Fig. 6).

Correlation analysis between amino acids and microbial genera in the rhizosphere soil of faba bean

In total, 17 amino acids with significant changes were selected for the correlation analysis between amino acids and key microbial genera in the rhizosphere soil of faba bean. At the bacterial genus level, Val, Ser, Phe, and Met were significantly negatively correlated with Burkholderia (Fig. 7A). Gly, Leu, Ser, and Tyr were significantly negatively correlated with Sphingosine (Fig. 7A). Asp, Trp, Gly, and Glu were significantly negatively correlated with Bradyrhizobium (Fig. 7A). Trp, Asn, Gln, Thr, Arg, Glu, and Pro were significantly positively correlated with Penicillium (Fig. 7B). At the fungal genus level, Gly was significantly positively correlated with Fusarium (Fig. 7B).

Fig. 7
figure 7

The correlation analysis between amino acids and microbial genera in the rhizosphere soil of faba bean. Correlation heat map of the bacterial genera (A), fungal genera (B) and various amino acids. *p < 0.05, ** < 0.01, and *** < 0.001

Full size image

Effect of amino acids in the rhizosphere soil on the faba bean growth and disease index

Arg, Ala, Leu, Phe, Thr, Glu, Ser, Met, Asn, Asp, Trp, Gln, Gly, Ile, Val, Pro, and Tyr were negatively correlated with the faba bean biomass and positively correlated with the disease index (Fig. 8). MF and MF100B treatments could increase the amino acid content in the rhizosphere soil of faba bean, reduce the faba bean growth, and promote the occurrence of Fusarium wilt. However, IF and IF100B treatments could reduce the amino acid content in the rhizosphere soil of faba bean, promote the faba bean growth, and reduce the occurrence of Fusarium wilt. Therefore, the amino acids in the rhizosphere soil may have an important mediating effect in the occurrence of Fusarium wilt.

Fig. 8
figure 8

Effects of amino acids in the rhizosphere soil of faba bean on the faba bean growth and disease index. Biomass: plant height + root length. Rhizosphere soil amino acids and RDA of faba bean growth and disease

Full size image

Conceptual model of amino-acid-mediated action of faba bean–wheat intercropping in reconstructing microbial community structure in the rhizosphere soil of faba bean in response to F. commune and benzoic acid stress (Fig. 9)

Fig. 9
figure 9

Conceptual model of amino-acid-mediated action of faba bean–wheat intercropping in reconstructing the microbial community structure in the rhizosphere soil of faba bean in response to Fusarium commune and benzoic acid stress. The brown line indicated that the faba bean tissue (root system) was damaged by the stress resulted in the amino acids leaked into the soil. The green line indicated that faba bean sought the help of wheat to together defend disease. The red dot indicated the harmful Fusarium taxa. The green dots indicated the beneficial microbiome taxa

Full size image

Discussion

In recent years, allelopathic and autotoxic effects have received increasing attention, and they play a crucial role in promoting the occurrence of soil-borne diseases (Ma et al. 2023; Li et al. 2014a). Exogenously added p-hydroxybenzoic acid, a derivative of benzoic acid, increased the disease index of Fusarium wilt in cucumber compared with no addition of p-hydroxybenzoic acid (Jin et al. 2020). Compared with the MF treatment, the MF100B treatment significantly increased the disease index of Fusarium wilt in faba bean (p < 0.05) (Fig. 1). This indicated that benzoic acid played an important role in promoting the occurrence of Fusarium wilt in faba bean. Several studies report that appropriate intercropping is an effective measure to control soil-borne diseases. Compared with cotton monocropping, onion and cotton intercropping could reduce the disease index of Fusarium wilt in cotton (Zhang et al. 2023a). Compared with the MF and MF100B treatments, the IF and IF100B treatments significantly reduced the disease index of Fusarium wilt in faba bean (p < 0.05) (Fig. 1). This indicated that faba bean–wheat intercropping effectively reduced the occurrence of Fusarium wilt in faba bean. Allelopathic autotoxins not only promote the occurrence of soil-borne diseases but also inhibit the crop growth. Total fresh weight, root length, and plant height of patchouli were significantly reduced after exogenous addition of p-hydroxybenzoic acid compared with no addition of p-hydroxybenzoic acid (Xu et al. 2015b). The MF treatment significantly reduced the root length and plant height of faba bean compared with the CK-M treatment (p < 0.05) (Fig. 2). Compared with the MF treatment, the MF100B treatment further significantly reduced the root length and plant height of faba bean (p < 0.05) (Fig. 2). Some studies reported that intercropping promotes plant growth. Intercropping with onion and tomato significantly increased plant height and dry weight of tomato compared with tomato monocropping (He et al. 2021). The IF and IF100B treatments increased the plant height and root length of faba bean compared with the MF and MF100B treatments, respectively (p < 0.05) (Fig. 2).

Microorganisms, as the main participants in soil material cycling and energy flow, play an important role in the storage and turnover of soil nutrients. Soil microbial community diversity and composition play a crucial role in determining plant health and soil quality (Raaijmakers and Mazzola 2016; Jia et al. 2022;). Community richness can be measured using the Ace index; the larger the index, the higher the community richness. Community diversity can be measured using the Shannon index; the higher the Shannon index, the higher the community diversity (Chiarello et al. 2022). Compared with no p-hydroxybenzoic acid addition, exogenously added p-hydroxybenzoic acid reduced the Shannon index of bacterial community in the rhizosphere soil of cucumber (Jin et al. 2020). The MF100B treatment significantly reduced the Shannon index of bacterial communities in the rhizosphere soil of faba bean compared with the MF treatment (p < 0.05) (Table 1). The decrease in microbial community diversity in rhizosphere soil may increase the incidence rate of soil-borne diseases (Qin et al. 2022). Intercropping with corn and soybean significantly increased the Shannon index of bacterial community in the rhizosphere soil of soybean, compared with soybean monocropping (Chang et al. 2022). In our study, compared with the MF100B treatment, the IF100B treatment significantly increased the Shannon index of bacterial community in the rhizosphere soil of faba bean (p < 0.05) (Table 1). It indicated that faba bean–wheat intercropping could increase the microbial diversity in the rhizosphere soil of faba bean. The fungal and bacterial community structures in the rhizosphere soil of cucumber treated with p-hydroxybenzoic acid were significantly different compared with the control (Jin et al. 2020). According to PCoA, fungal and bacterial community structures in the rhizosphere soil of faba bean were significantly different under the MF100B and MF treatments (ANOSIM, R = 0.6218, p = 0.001; ANOSIM, R = 0.6099, p = 0.001) (Fig. 4). This indicated that benzoic acid played an important role in altering the fungal and bacterial community structures in the rhizosphere soil of faba bean. Intercropping can increase microbial diversity and exhibit a more stable community structure (Zhang et al. 2021; Yu et al. 2019). Compared with the MF100B treatment, the IF100B treatment exhibited a trend of restoring the bacterial and fungal community structures, increasing the microbial community stability in the rhizosphere soil of faba bean (ANOSIM, R = 0.6099, p = 0.001; ANOSIM, R = 0.6218, p = 0.001) (Fig. 4). These results are consistent with previous studies (Tian et al., 2019; Yu et al. 2019; Zhou et al. 2019).

When attacked by pathogens, plants recruit beneficial microorganisms, alter the microbial community composition in the rhizosphere, and enhance their microbial activity to resist pathogens. Some key or functional species are more important for the occurrence and control of soil-borne diseases (Li et al. 2021b; Ofek et al. 2014). To determine the microbial groups as indicators of soil-borne diseases, the microbial community composition is assessed at the phylum and genus levels at various stages the disease. At the phylum level, the MF100B treatment significantly increased the relative abundance of Ascomycota compared with the MF treatment (p < 0.05) (Fig. 5B). This indicated that benzoic acid has more obvious effect on some microorganisms with specific functions; for example, it especially increases the relative abundance of Ascomycota, a pathogenic phylum (Zhou and Wu 2012). The rhizosphere soil of cucumber infected with Fusarium could increase the relative abundance of Fusarium and Bacillus compared with the uninfected rhizosphere soil (Meng et al. 2019). The MF treatment significantly increased the relative abundance of Fusarium and Penicillium compared with the CK-M treatment (p < 0.05) (Fig. 5D). Among them, Penicillium sp. plays an important role in antagonizing pathogenic fungi (Zhao et al. 2021). This may be due to the activation of the stress response of the faba bean itself, recruiting antagonistic microorganisms to respond to the Fusarium stress. The MF100B treatment significantly reduced the relative abundance of Sphingomonas compared with the MF treatment (p < 0.05) (Fig. 5C). Sphingomonas sp. plays an important role in antagonizing pathogenic fungi to promote plant growth and degrading phenolic acids (He et al. 2017; Siddiqui and Shaukat 2002; Liu et al. 2023; Prischl et al. 2012). The possible reason is that the ability of faba bean to recruit antagonistic microorganisms may be limited in response to this biotic stress (Fusarium) and abiotic stress (benzoic acid). Compared with the V. villosa monocropping, intercropping with V. villosa and banana reduced the relative abundance of Fusarium in the rhizosphere soil of banana (Yang et al. 2022a). In this study, the IF100B treatment significantly increased the relative abundance of Sphingomonas and Bradyrhizobium and reduced that of Fusarium in the rhizosphere soil of faba bean, compared with the MF100B treatment (p < 0.05) (Fig. 5C and D). Bradyrhizobium sp. plays an important role in antagonizing pathogenic fungi and promoting plant growth (Tewari and Sharma 2020; Cai et al. 2020). Wheat root exudates could inhibit the growth of Fusarium and recruit specific Pseudomonas species (Lv et al. 2018; Weller et al. 2002). The possible reason is that as the ability of faba bean to recruit antagonistic microorganisms may be limited in response to this combination of biotic stress (Fusarium) and abiotic stress (benzoic acid), faba bean may seek help from neighboring wheat plants, which inhibit Fusarium activity by secreting root exudates or helping faba bean to recruit antagonistic microorganisms to fight against Fusarium.

Changes in soil environment and microbial community composition are regulated by plant root exudates (Rousk et al. 2010; Wen et al. 2022). Among them, free amino acids can be secreted into the rhizosphere soil by plant roots (Xie et al. 2020). Compared with sesame plants not infected with Fusarium, sesame plants infected with Fusarium significantly increased the content of Ala, Ile, Thr, Glu, Leu, Try, Asn, Phe, Ser, Met, Asp, Gly, Val, Arg, Tyr, Lys, His, and Pro in the rhizosphere soil (Radhakrishnan et al. 2013). Under stress conditions, high concentrations of free amino acids were associated with the symptoms of tissue and organ damage in plants (Silveira et al. 2001). The MF treatment significantly increased the content of Ala, Asn, Asp, Ile, Gly, Glu, Val, Thr, Tyr, Pro, Leu, Gln, and Phe in the rhizosphere soil of faba bean, compared with the CK-M treatment (p < 0.05) (Fig. 6). Compared with the MF treatment, the MF100B treatment significantly increased the content of Ala, Ile, Trp, Thr, Ser, Met, Asn, Asp, Gln, Leu, Gly, Glu, Phe, Val, Pro, and Tyr in the rhizosphere soil of faba bean (p < 0.05) (Fig. 6). This may be because the stress response of faba bean is activated in response to the synergistic effect of biotic stress (Fusarium) and abiotic stress (benzoic acid), which increases the amino acid metabolism in the plant secreting the amino acids into the rhizosphere soil through the root system. After infection with Phytophthora pestis, aspartic acid transaminase could promote the bidirectional conversion of Asn to Gln (Radhakrishnan et al. 2013). Under stress conditions, glutamate could be converted into other amino acids such as Pro and Arg (Brauc et al. 2011). The amino acids secreted by cotton roots were positively correlated with the disease index of Fusarium wilt in cotton (Li et al. 2009). Our results from RDA analysis are consistent with a previous study (Fig. 8) (Li et al. 2009). Free amino acids in soil, as an important source of soluble organic nitrogen, play an important effect in the growth of soil microorganisms (More et al. 2020; Hu et al. 2001; Cheng et al. 2022). In this study, a correlation analysis was performed between the amino acids and key microorganisms in the rhizosphere soil of faba bean. Gly, Leu, Ser, and Tyr were significantly negatively correlated with Sphingomonas (Fig. 7A). Asp, Trp, Gly, and Glu were negatively correlated with Bradyrhizobium (Fig. 7A). Gly was significantly positively correlated with Fusarium (Fig. 7B). Trp, Asn, Gln, Thr, Arg, Gln, and Pro were significantly positively correlated with Penicillium (Fig. 7B). The possible reason is that different amino acids may have selectivity toward different antagonists, growth-promoting bacteria, and pathogenic fungi in the rhizosphere (Chapon et al. 2002; Doornbos et al. 2012). Fusarium has the highest Gly utilization efficiency among various organic nitrogen sources (Naim and Sharoubeem 1964). In this study, the IF treatment significantly increased Asp content and reduced Thr, Asn, Glu, Pro, and Gln contents in the rhizosphere soil of faba bean, compared with the MF treatment (p < 0.05) (Fig. 6). Compared with the MF100B treatment, the IF100B treatment significantly increased Asp content and significantly reduced Ala, Ser, Thr, Phe, Glu, Leu, Asn, Gly, Met, Val, Tyr, Ile, Trp, Pro, and Gln contents in the rhizosphere soil of faba bean (p < 0.05) (Fig. 6). The possible reason why only Asp content was increased is that the rhizosphere soil itself is in an oligotrophic state, and Asp replenishes nitrogen deficiency by providing organic nitrogen. On the one hand, the decrease in most amino acids in rhizosphere soil may reduce the availability of resources to the microorganisms in the rhizosphere soil (Wei et al. 2015; Gu et al. 2017; Raaijmakers et al. 2008;). Studies have reported that the wheat root exudates have inhibitory effects on the activity of Fusarium (Lv et al. 2018). The content of most amino acids was reduced under faba bean–wheat intercropping, which may reduce the availability of resources in the rhizosphere environment to the microorganisms, strengthen the competitive ability of antagonistic beneficial microorganisms against Fusarium and reduce the Fusarium stress in faba bean, thus reducing the occurrence of Fusarium wilt in faba bean. Notably, wheat roots also secrete and absorb amino acids in wheat and intercropping systems. In the future, we need to explore more for the changes in the secretion and absorption of amino acids in the roots of wheat, a neighbor crop of faba bean in wheat and intercropping systems, which is crucial for how wheat helps faba bean to defend against Fusarium wilt.

Conclusion

Combined stress of F. commune and benzoic acid could increase the content of certain amino acids, reduce the bacterial community diversity, change the fungal and bacterial community structure and composition, reduce the relative abundance of beneficial microorganisms including Sphingomonas, and increase the relative abundance of Fusarium in the rhizosphere soil of faba bean. Moreover, it reduced the growth of faba bean and promoted the occurrence of Fusarium wilt in faba bean. Faba bean–wheat intercropping could reduce the content of certain amino acids, increase bacterial community diversity, change the fungal and bacterial community structure and composition, increase the relative abundance of beneficial microorganisms including Sphingomonas and Bradyrhizobium, and reduce the relative abundance of Fusarium in the rhizosphere soil of faba bean. Moreover, it played a role in antagonizing pathogens, promoted the faba bean growth, and reduced the occurrence of Fusarium wilt in faba bean. Therefore, faba bean–wheat intercropping is an effective measure for the management of Fusarium wilt in faba bean, and it acts via reducing the abundance of Fusarium, improving rhizosphere habitats, and gradually restoring healthy soil microbial communities.