Abstract

   Excessive soil cadmium (Cd) and the accumulation of pathogens pose
   serious threats to legume growth. However, it remains unclear whether
   intercropping (IFcd) and its combined treatment with silicon
   nanoparticles (Si-NPs) (IFcd + Si) can alleviate these challenges under
   Cd stress, as well as the underlying mechanisms involved. This study
   systematically elucidated the mechanism of faba bean-wheat
   intercropping and Si-NPs regulating faba bean growth under Cd stress
   using rhizosphere metabolomics and 16 S rRNA microbiome analysis. The
   results showed that IFcd and IFcd + Si treatments significantly reduced
   Cd accumulation by 17.3% and 56.2%, and Fusarium wilt incidence by
   11.1% and 33.3%, respectively, compared with monoculture faba bean
   (MFcd) while promoting root and plant growth. These treatments reduced
   oxidative stress markers, including H[2]O[2], MDA, and O[2]^−, and
   increased the activity of defense enzymes, such as SOD, APX, and POD in
   plants. Furthermore, they increased NH[4]^+-N and available potassium
   levels in rhizosphere soils. Interestingly, the NH[4]^+-N content
   increased and was significantly positively correlated with urease (URE)
   activity and negatively correlated with Cd. Beneficial bacteria and
   functional metabolites were enriched in the rhizosphere of faba bean.
   Joint analysis revealed increased relative abundances of Sphingomonas,
   Intrasporangium, and Streptomyces, which were positively correlated
   with antibacterial metabolites, such as sordarin, lactucin, and
   15-methylpalmate. This explains the reduced Cd accumulation and
   Fusarium wilt in plants. These findings provide mechanistic insights
   into how intercropping with Si-NPs mitigates Cd stress and controls
   soil-borne diseases by regulating rhizosphere metabolites, bacterial
   communities, and plant resistance.

Graphical Abstract

   [32]graphic file with name 12951_2025_3330_Figa_HTML.jpg

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12951-025-03330-0.

   Keywords: Abiotic and biotic stress, Nanoparticles, Intercropping, Faba
   bean fusarium wilt, Microbial- metabolomics analysis

Introduction

   Soil is a critical foundation for supporting crop health and for
   ensuring the production of reliable agricultural products. However,
   cadmium (Cd) pollution in farmland and soil-borne pathogens are common
   abiotic and biotic stresses that affect crops in southern China [[33]1,
   [34]2]. Authoritative data indicate that 7% of China’s land area is
   affected by varying degrees of Cd pollution [[35]3]. The primary
   sources of Cd in farmland soil include atmospheric deposition, sewage
   irrigation, excessive use of phosphate fertilizers, and improper
   application of animal manure [[36]4]. Soil is considered
   Cd-contaminated when its Cd concentration exceeds 1 ppm [[37]5]. Even
   at lower concentrations of Cd, plants can readily absorb and accumulate
   Cd through their roots, causing adverse effects on crop growth and
   posing risks to agricultural food safety [[38]6]. Additionally,
   Fusarium can easily accumulate in continuous cropping soil, leading to
   frequent outbreaks of soil-borne diseases in various crops such as
   cucumber, tomato, and faba bean [[39]7]. As China’s population grows
   and urbanization advances, the availability of cultivated land has
   become increasingly limited, leading to large-scale intensive planting
   [[40]1]. Therefore, to address these challenges, sustainable
   agricultural practices, such as intercropping, have gained significant
   attention owing to their potential to enhance soil health and crop
   resilience.

   Faba bean (Vicia faba L.) is a crucial cold-season legume crop that
   provides dietary protein as a vegetable, feed, and grain in many
   developing countries [[41]8]. Faba bean cultivation in Yunnan, China
   has made significant contributions to both national and global legume
   production. However, studies have shown that local faba bean
   cultivation in Yunnan is increasingly challenged by Cd pollution risks
   and frequent outbreaks of faba bean Fusarium wilt disease [[42]9]. The
   wheat-faba bean intercropping system, which can maximize spatial and
   temporal differentiation, soil moisture, nutrients, and light resources
   while managing soil-borne diseases and other stresses, is an
   indispensable planting practice in the entire southwest faba bean
   cultivation [[43]10]. Research has demonstrated that intercropping faba
   bean with the hyperaccumulating plant Sonchus asper (L.) Hill can
   effectively reduce the available cadmium (Cd) content of faba bean
   plants [[44]9]. Intercropping of Lolium perenne and Medicago sativa can
   significantly reduce Cd levels in alfalfa plants and inhibit effects of
   Cd on plant growth [[45]11]. Although faba bean is not a typical
   hyperaccumulator of Cd, heavy metal stress (such as Cd) can severely
   hinder its growth, product quality, and safety [[46]12]. Our previous
   research has suggested that wheat-faba bean intercropping can
   effectively control Fusarium wilt in faba bean and increase crop yield
   [[47]7]. Additionally, strategies including beneficial elements such as
   silicon may play a role in alleviating this pressure. However, there
   are no reports on whether intercropping wheat-faba bean under Cd stress
   can reduce Cd toxicity and control faba bean wilt disease to enhance
   growth.

   Silicon (Si) is the second most abundant chemical element in the
   Earth’s crust after oxygen. Although Si is not considered an essential
   nutrient for plants, it is recognized as a beneficial element in most
   plant species [[48]13, [49]14]. Silicon stress resistance in plants is
   evident not only in their ability to mitigate abiotic stresses such as
   Cd toxicity [[50]15, [51]16] but also in their unique silicification
   protective layer and strong adhesion properties. These features help
   crops resist pests and diseases, enhance disease resistance, and reduce
   pesticide residues [[52]17]. In soil, Si mainly exists in the form of
   silicates and oxides, which plants cannot absorb directly. The
   absorbable form of Si in plants is primarily monosilicic acid
   (Si(OH)[4]) [[53]16]. Compared with conventional silicon products,
   silicon nanoparticles (Si-NPs) have been introduced in agriculture
   owing to their small size, large surface area, stability, and
   biocompatibility [[54]18]. Studies have demonstrated that Si-NPs not
   only promote plant growth and flowering [[55]17, [56]13] but also
   enhance seed vitality, increase enzyme activity, improve root growth,
   boost water and fertilizer utilization efficiency, accelerate
   metabolism, and strengthen resistance to diseases and pests, ultimately
   improving crop yield and quality [[57]19–[58]21]. These findings
   suggest that Si-NPs are viable alternatives to alleviate Cd toxicity
   and promote plant growth. Given that plants under biotic and abiotic
   stresses often exhibit stunted growth and reduced systemic resistance,
   it is crucial to explore how Si-NPs can mitigate these effects,
   particularly in the context of Cd toxicity and Fusarium wilt.
   Nonetheless, the potential of Si-NPs to concurrently modulate the
   rhizosphere microorganisms and metabolic pathways in faba bean, thereby
   mitigating Cd toxicity and enhancing resistance to Fusarium wilt, has
   yet to be fully elucidated.

   Plants subjected to Cd and Fusarium stress often exhibit stunted
   growth, poor root development, reduced systemic resistance, and
   diminished yield and quality [[59]7]. When Cd infiltrates plants, it
   binds to proteins and DNA, causing denaturation and damage, leading to
   metabolic disorders, inhibited photosynthesis and respiration,
   increased production of reactive oxygen species (ROS), exacerbated
   membrane lipid peroxidation, and damage to plant cell structure,
   ultimately inhibiting plant growth [[60]16, [61]22]. Intercropping
   target crops with hyperaccumulating plants can utilize interactions
   between plants and associated microorganisms. This practice promotes
   the secretion of organic acids and enrichment of degrading bacteria,
   which affects the bioavailability of HMs in the crop rhizosphere
   [[62]23, [63]24]. Si can reduce Cd absorption and toxicity by
   regulating antioxidant enzyme activities and the expression of Cd
   transport genes in plants [[64]25, [65]26] while enhancing systemic
   resistance by influencing the microbial communities in the rhizosphere
   and altering the secretion of citric acid and other substances [[66]24,
   [67]27]. It is currently uncertain whether exogenous Si-NPs have the
   ability to enhance interspecific root interactions in intercropped faba
   bean and thereby promote its growth. Furthermore, the mechanisms by
   which rhizosphere microbial-metabolite interactions might be involved
   in these processes are also not yet clear.

   This study aimed to elucidate the effects of intercropping and the
   combination Si-NPs on Cd toxicity and resistance to Fusarium wilt in
   faba bean under Cd and Fusarium stress, as well as the interaction
   mechanism between rhizosphere microorganisms and metabolites. A pot
   experiment with the exogenous addition of Si-NPs was conducted to
   verify that (1) intercropping and Si-NPs superposition alleviated Cd
   toxicity and Fusarium wilt occurrence in faba bean; (2) Si-NPs and
   intercropping enhanced the antioxidant activity and systemic defense
   ability of faba bean plants; (3) Si-NPs and intercropping increased the
   NH[4]^+-N content and related enzyme activity in the faba bean
   rhizosphere; and (4) the rhizosphere interaction effect of
   microbiota-metabolomics of faba bean explains the prominent effects of
   intercropping and Si-NPs superposition. These findings provide a
   theoretical basis for using intercropping and Si-NPs in agriculture to
   improve soil health, control disease, and increase crop yield.

Materials and methods

Test location and test materials

   A soil pot experiment was conducted in a greenhouse at Yunnan
   Agricultural University (25°7’N, 102°44’E, altitude: 1950 m) using
   natural day-night lighting (daylight, 11 h), with temperatures ranging
   from 20 °C to 30 °C and relative humidity between 50% and 60%. The
   experiment was conducted from early October 2023 to mid January 2024,
   using artificial paddy soil in Efeng Village, Yuxi City, Yunnan
   Province as the test soil. The region has a tradition of continuous
   faba bean cultivation, which has led to a significant accumulation of
   the soil-borne Fusarium pathogen in the soil, causing faba bean
   Fusarium wilt. The basic physical and chemical properties of the sample
   soil were: organic matter 22.4 g kg^− 1, total nitrogen 1.5 g kg^− 1,
   total phosphorus 0.8 g kg^− 1, total potassium 12.6 g kg^− 1, alkali
   nitrogen 102.0 mg kg^− 1, Olsen phosphorus 16.9 mg kg^− 1, available
   potassium 116.5 mg kg^− 1, Cd^2+ 0.075 mg kg^− 1, and pH 7.1. The
   utilized fertilizers were urea (N 46.0%), ordinary superphosphate
   (P[2]O[5] 16.0%), and potassium sulfate (K[2]O 50.0%).

   Cd was added using CdCl[2]⋅2H[2]O (CAS: 10108-64-2, ≥ 99.0%), purchased
   from Beijing Huawei Ruike Chemical Technology Co., Ltd. (Beijing,
   China). Peptide-coupled fluorescent silicon nanoparticles (Si-NPs)
   (CAS: 7440-21-3, ≥ 99.0%) prepared by microwave irradiation were
   selected as nano-silicon materials. This material, which has high
   stability and dispersibility and is highly absorbable by plants, was
   obtained from Xi’an Qiyue Biotechnology Co., Ltd. (Shaanxi, China).

   The tested varieties of faba bean and wheat were Vicia faba L. cv.
   Yuxidalidou and Triticum aestivum L. cv. Yunmai-112, purchased from the
   Grain and Crop Research Institute of Yunnan Academy of Agricultural
   Sciences.

Experimental design and crop management

   The experiment utilized a single-factor design with three soil pot
   treatments: (a) a monoculture of faba bean under Cd stress (MFcd), with
   three rows per pot, three faba bean per row, plant spacing of 10 cm,
   and row spacing of 8 cm; (b) faba bean-wheat intercropping under Cd
   stress (IFcd); and (c) faba bean-wheat intercropping under Cd stress
   with the addition of Si-NPs (IFcd + Si). In the intercropping
   treatments, the planting arrangement consisted of one row of faba bean
   (three plants/row), followed by three rows of wheat (six plants/row),
   and another row of faba bean (three plants/row). Each treatment
   contained three biological replicates, resulting in a total of nine
   pots. The plastic pots were 22 cm in height, 34 cm in upper diameter,
   and 19 cm in lower diameter.

   The topsoil (0–20 cm) collected from the field was air-dried in a
   greenhouse for one week. Afterward, debris were removed, and the soil
   was sieved through a 2 mm sieve. Each pot was filled with 9.0 kg of
   soil. To prepare Cd-contaminated soil (2 mg kg^− 1 Cd^2+), 36.61 mg
   CdCl[2]⋅2H[2]O was mixed with 9.0 kg of experimental soil [[68]14]. The
   soil was then left to equilibrate for four weeks, with soil moisture
   maintained at 60% of the field capacity to ensure consistent Cd
   concentrations across different soil layers, simulating typical
   Cd-polluted natural soil. Si-NPs were thoroughly mixed with distilled
   water and then uniformly mixed with the experimental soil at a dosage
   ratio of 250 mg kg^− 1 [[69]14] before being transferred into the pots.

   To eliminate uncertainties related to nutrients, seed emergence rate,
   and light exposure, fertilizers were uniformly applied at the time of
   potting, with nitrogen, phosphorus, and potassium added at equal rates
   (N: 100 mg kg^− 1, P[2]O[5]: 100 mg kg^− 1, K[2]O: 100 mg kg^− 1) for
   all treatments. Before starting the pot experiment, wheat and faba bean
   seeds were carefully selected to ensure that each seed had a uniform
   color, size, and plumpness. The seeds were surface sterilized with 10%
   (v/v) H[2]O[2] for 30 min, thoroughly rinsed, and germinated in the
   dark for three days [[70]7]. The greenhouse was maintained at
   temperatures between 20 °C and 30 °C under natural light and a relative
   humidity of 50%. The pots were randomly placed and repositioned every
   three days. Watering was performed every two days with 200–250 mL of
   distilled water per pot, maintaining the soil moisture at 60–70% of the
   maximum field capacity. Other daily management practices were
   consistent with typical field management.

Plant and soil sample collection

   After three months of co-cultivation, two plants were selected for
   destructive sampling. The entire faba bean plant was carefully removed
   from the pot, and agronomic parameters of the aboveground parts were
   quickly recorded, along with the investigation of faba bean Fusarium
   wilt and biomass measurement (dried at 105℃ for 15 min and then at
   65 °C to constant weight). Rhizosphere soil was collected using the
   “shaking soil method” [[71]28]. When collecting plant roots, they were
   first rinsed with distilled water, and the plants were divided into
   different organs (roots and aboveground parts) using scissors. Special
   care was taken to maintain the integrity of root structures. The roots
   were then soaked in 20 mM Na[2]-EDTA for 15 min to remove Cd and Si
   ions attached to the surface, followed by thorough rinsing with
   distilled water [[72]29]. The roots were then prepared for subsequent
   root scanning analysis and determination of defense substances and
   enzyme activity parameters.

   The collected rhizosphere soil samples were sieved through a 2 mm
   aperture and divided into two parts. One part was used to determine the
   conventional physicochemical properties and Cd content of the
   rhizosphere soil, while the other part was stored in a sterile tube at
   -80℃ for bacterial and metabolomic analyses. The collection process
   strictly followed the sterile operating procedures to prevent
   contamination.

Indicator determination and methods

Investigation of faba bean Fusarium wilt

   Wilt disease investigation was conducted on all faba bean plants upon
   detachment of their aboveground parts. The classification of faba bean
   wilt disease followed a 5-level classification standard [[73]7]. After
   investigation, the wilt disease index was calculated using the
   following formula:undefined
   graphic file with name d33e525.gif

   {\rm{index}} = {\matrix{\Sigma
   \>({\rm{Number}}\>{\rm{of}}\>{\rm{diseased}}\>{\rm{plants}}\>{\rm{at}}
   \hfill \cr \>{\rm{each}}\>{\rm{level}} \times
   \>{\rm{level}}\>{\rm{value}}) \hfill \cr} \over
   \matrix{{\rm{The}}\>{\rm{highest}}\>{\rm{level}}\> \times
   \>{\rm{total}}\> \hfill \cr
   {\rm{number}}\>{\rm{of}}\>{\rm{plants}}\>{\rm{studied}} \hfill \cr} }
   \times \>100]]>

Determination of rhizosphere soil physicochemical properties

   The collected soil samples were sieved and analyzed using the following
   methods: the soil pH was measured with a pH meter in a soil/water
   suspension (1:2.5); the organic matter was determined using the
   K[2]Cr[2]O[7]-H[2]SO[4] oxidation method; and ammonium nitrogen and
   nitrate nitrogen were measured via the indigo phenol blue colorimetric
   method and ultraviolet spectrophotometry, respectively. Available
   phosphorus was extracted with NaHCO[3] and measured using the
   molybdenum-antimony colorimetric method, and total phosphorus was
   determined using the HClO[4]-H[2]SO[4] digestion method. The flame
   photometer method was used to determine the available and total
   potassium contents. All of the above physicochemical parameters were
   determined according to the method described by Bao [[74]30]. The
   determination of rhizosphere soil enzyme activity was carried out using
   air-dried soil (5.000 g; 1 mm sieve screening) and placed in a 50mL
   centrifuge tube. The extraction solution and buffer solution were then
   added according to the manufacturer’s instructions for
   constant-temperature cultivation. The activities of catalase (CAT),
   neutral phosphatase (NPase), dehydrogenase, urease (URE), cellulase
   (cell), and sucrase (SUC) in the rhizosphere soil were measured using
   soil enzyme activity assay kits purchased from Beijing Solaboo
   Technology Co., Ltd. (Beijing, China) according to the detailed steps
   and procedures provided in the manufacturer’s instructions [[75]31].

   The bioavailable Cd content in the rhizosphere soil was extracted using
   diethylenetriaminepentaacetic acid (DTPA) [[76]4], and the total Cd in
   the soil and plant leaves was analyzed using a mixed solution of
   HNO[3]-HClO[4]-HF (5:1:1,v/v/v) and HNO[3] digestion [[77]32], followed
   by inductively coupled plasma mass spectrometry (ICP-MS; Agilent 7500a,
   USA). The standard test results were within the allowable error range
   with a recovery rate of 90–105%.

Scanning of root morphological characteristics

   After washing the collected and separated roots with deionized water,
   they were placed on the root plate of a root scanner (EPSON Perfection
   4990 Photo, China), injected with an appropriate amount of deionized
   water, and scanned using preset parameters. The scanned images were
   processed using image processing software (Adobe Photoshop CC 2019),
   and root morphology data, such as total root length, average diameter,
   perimeter, surface area, and root length-to-diameter ratio, were
   extracted using specialized root analysis software (WinRHIZO Pro
   2007d13).

Determination of antioxidant capacity and defense enzyme activity of faba
bean

   Fresh third true leaf (0.1 g) was weighed, and 1 ml of extraction
   solution (containing buffer solution) was added to homogenize the
   mixture in an ice bath. After homogenization, the supernatant was
   centrifuged at 4℃ (10000 ×g, 10 min), collected, and placed on ice to
   detect the contents of hydrogen peroxide (H[2]O[2]), malondialdehyde
   (MDA), and superoxide anion (O[2]^−), and the activities of superoxide
   dismutase (SOD), ascorbate peroxidase (APX), and peroxidase (POD). MDA,
   H[2]O[2], and O[2]^− contents were determined using trichloroacetic
   acid-thiobarbituric acid (TCA-TBA), titanium sulfate, and hydroxylamine
   methods, respectively, following the procedures outlined by Yan et al.
   [[78]14]. SOD activity was measured according to Giannopolis and Ries,
   APX activity was measured according to Nakano and Asada [[79]33], and
   POD activity was measured according to the guaiacol method described by
   Li et al. [[80]34].

Determination of rhizosphere soil bacteria

   Total DNA of the soil microbial genomes was extracted using the
   EZ.N.A^® Soil DNA Kit (Omega Biotek, Norcross, GA, USA) from 0.5 g of
   freeze-dried soil sample. The AxyPrep DNA Gel Extraction Kit (Axygen
   Biosciences, USA) was used to purify the recovered product. A 2%
   agarose gel electrophoresis was employed to assess the quality of the
   extracted DNA, and NanoDrop2000 (Thermo Scientific, USA) was used to
   measure DNA concentration and purity. Genomic DNA was then sent to
   GENEWIZ for relative quantitative sequencing of the highly variable
   V3–V4 region of the 16S rRNA gene, which was selected for soil
   bacterial sequencing. The primers used were 338F
   (5’-ACTCCTACGGGAGGCAGCAG-3’) and 806R (5’-GACTACHVGGGTWTCTAAT-3’)
   [[81]35]. Paired-end sequencing (2 × 250 bp) was performed using an
   Illumina NovaSeq 6000 sequencing platform. The PCR system and procedure
   are described in the Supporting Information. Processing, sequencing,
   operational taxonomic unit (OTU) clustering, and bacterial species
   annotation of the PCR products were conducted on the cloud platform of
   Shanghai Meiji Biotechnology Co., Ltd. ([82]https://cloud.majorbio.com)
   following the method described by Zhang et al. [[83]36]. Raw sequence
   data were first stored in the Sequence Read Archive (SRA) of the
   National Center for Biotechnology Information (NCBI) and then processed
   and analyzed using the Quantitative Insight Microbial Ecology (QIIME)
   software package version 1.8.0 [[84]37]. The raw data provided in this
   study were submitted to NCBI under the accession number PRJNA1190895.

Metabolomics analysis of rhizosphere soil

Sample preparation

   First, 1000.0 ± 5.0 mg of the freeze-dried rhizosphere soil sample was
   accurately weighed into a 2 mL centrifuge tube and a 6 mm grinding bead
   was added. Subsequently, 1000 µL of extraction solution (methanol:
   water = 4:1, v: v) containing 0.02 mg/mL of the internal standard
   (L-2-chloroalanine) was added for metabolite extraction. The sample
   solution was ground using a frozen tissue grinder for 6 min (-10℃,
   50 Hz), followed by low-temperature ultrasonic extraction for 30 min
   (5 °C, 40 kHz). The sample was then placed at -20 °C for 30 min,
   centrifuged for 15 min (4 °C, 13,000 × g), and the supernatant was
   transferred to an injection vial with an inner tube for analysis.
   Additionally, 20 µL of supernatant from each sample was pooled to
   create a quality control sample. During the analysis, a QC sample was
   inserted every three samples to monitor the repeatability of the entire
   process.

LC-MS detection

   Detection was performed using a Thermo Fisher Scientific
   ultra-high-performance liquid chromatography-tandem Fourier transform
   mass spectrometry (UHPLC-Q Exactive HF-X) system. The chromatographic
   column was ACQUITY UPLC HSS T3 (100 mm × 2.1 mm i.d., 1.8 μm; Waters,
   Milford, USA). Mobile phase A consisted of 95% water and 5%
   acetonitrile with 0.1% formic acid, while mobile phase B contained
   47.5% acetonitrile, 47.5% isopropanol, and 5% water with 0.1% formic
   acid. The injection volume was 3 µL, column temperature was set to
   40 °C, and flow rate was 0.4 mL/min. The gradient elution program was
   as follows: 0–3.5 min, 0–24.5% B; 3.5–5 min, 24.5–65% B; 5–5.5 min,
   65–100% B; 5.5–7.4 min, 100% B, 0.4–0.6 mL/min; 7.4–7.6 min, 100–51.5%
   B; 7.6–7.8 min, 51.5–0% B, 0.6–0.5 mL/min; 7.8–9 min, 0% B, 0.5–0.4
   mL/min; 9–10 min, 0% B, 0.4 mL/min. The mass spectrometry conditions
   included a spray ion source (ESI) temperature of 425 °C, mass spectrum
   voltages of 3500 V (positive mode) and − 3500 V (negative mode), and a
   mass scanning range of 70–1050 m/z. The sheath gas flow rate was 50
   psi, auxiliary gas flow rate was 13 psi, ion transport tube temperature
   was 325 °C, and normalized collision energy was 20–40–60 V in the
   cyclic collision energy. The resolution of the primary mass
   spectrometry was set to 60,000, and that of the secondary mass
   spectrometry was 7,500, with the data collected in DDA mode.

Computer data processing and differential metabolite analysis

   The raw data were imported into the metabolomics processing software
   Progenesis QI v2.3 (Waters Corporation, Milford, USA) for analysis. The
   software was used for feature peak searching and library
   identification, matching MS and MS/MS spectrometry information with
   metabolic databases, with the MS error set to less than 10 ppm.
   Metabolites were identified based on secondary mass
   spectrometry-matching scores. Principal component analysis (PCA) and
   partial least squares discriminant analysis (PLS-DA) were performed on
   the metabolites using the ropls package in R, and model stability was
   evaluated using a 7-cycle interaction validation. Additionally,
   Student’s t-test and fold-change analysis were performed. Differential
   metabolites were screened using variable weight values (VIP values > 1
   and p < 0.05), and the top 30 significantly different metabolites in
   each group were intersected. The metabolic pathway enrichment analysis
   of the differential metabolites was conducted using the KEGG database
   ([85]https://www.kegg.jp/kegg), with the P-values < 0.05, indicating
   the significant pathway enrichment.

Statistics and analysis

   Microsoft Excel 2010 and GraphPad Prism 8.0 were utilized to organize
   and plot the test data. Statistical analysis was performed using
   one-way analysis of variance (ANOVA) and Duncan’s multiple range test
   (SPSS 19.0, IBM, USA) to evaluate the differences in indicators, such
   as Fusarium wilt occurrence, root morphology characteristic parameters,
   growth parameters, soil enzyme activity, Cd content, and soil
   physicochemical factors among the MFcd, IFcd, and IFcd + Si treatments,
   with statistical significance set at P < 0.05. Alpha diversity of
   bacterial communities (Chao, Shannon, and Simpson indices) was
   evaluated using the Rogers’s [[86]38] method, and inter-group
   differences were assessed using the Wilcoxon rank sum test. Beta
   diversity analysis based on Bray-Curtis distance was used to compare
   the microbial community structure between treatments, which was
   verified and visualized through principal coordinate analysis (PCoA).
   Data in the tables and figures are presented as mean ± standard
   deviation (SD) of three replicates. Correlation analysis between
   parameters was conducted using Spearman’s method. Metabolomics of the
   rhizosphere soil, microbiota, and their association maps with
   differential metabolites were analyzed and generated using the free
   online bioinformatics cloud platform of Shanghai Meiji Biomedical
   Technology Co., Ltd. ([87]https://cloud.majorbio.com).

Results

Exogenous Si-NPs and intercropping superposition alleviate cd toxicity and
Fusarium wilt in faba bean

   Compared with the monoculture of faba bean (MFcd) under Cd stress, both
   intercropping (IFcd) and intercropping with Si-NPs (IFcd + Si) improved
   faba bean growth (Fig. [88]1a and b). The IFcd and IFcd + Si treatments
   significantly increased the dry weight (both aboveground and
   underground) by 80.7% and 126.7%, respectively, and plant height by
   28.3% and 57.8%, respectively, but also significantly reduced the
   incidence of Fusarium wilt by 11.1% and 33.3% and Cd accumulation in
   plant leaves by 17.3% and 56.3%, respectively (P < 0.05)
   (Fig. [89]1c–f). However, IFcd alone did not significantly reduce the
   incidence of Fusarium wilt compared to MFcd (P > 0.05). These results
   indicated that intercropping with Si-NPs could notably reduce Cd
   accumulation in plants and decrease the occurrence of soil-borne
   diseases.

Fig. 1.

   [90]Fig. 1
   [91]Open in a new tab

   Effects of intercropping (IFcd) and Si-NPs superposition (IFcd + Si) on
   faba bean growth and Fusarium wilt occurrence. Panels (a)–(f) show
   plant growth status, faba bean phenotype per plant, dry weight of
   aboveground and belowground tissues, plant height, Fusarium wilt
   incidence, and Cd content in plant leaves, respectively. Values are
   presented as the mean ± SD (n = 3). Different lowercase letters
   indicate significant differences between treatments (P < 0.05)

   Further analysis of the root morphology of faba bean under different
   treatments (Table [92]S1) indicated that both the IFcd and IFcd + Si
   treatments increased the total root length, perimeter, root length
   diameter (rang1), and surface area diameter (rang 1 and rang 2) to
   varying degrees, while reducing the average root diameter and root
   surface area. The increases in total root length by 7.6% and 8.1% and
   surface area diameter (rang 1 and rang 2) by 5.4%, 26.4%, 23.6%, and
   18.2% were significant (P < 0.05). Compared with IFcd alone, IFcd + Si
   did not significantly increase the total root length, average diameter,
   circumference, or surface area, but significantly enhanced the root
   length diameter (rang 1) and surface area diameter (rang 1) of the main
   root parts (P < 0.05). This revealed that intercropping with Si-NPs
   could effectively increase the diameter and total root length of the
   main root parts, improving resistance to Cd toxicity and Fusarium
   pathogen stress. Additionally, Si-NPs can also promote faba bean growth
   to a certain extent.

Exogenous Si-NPs and intercropping superposition enhance antioxidant activity
and systemic defense in faba bean plants

   The contents of H[2]O[2], MDA, and O[2]^− in plant leaves reflect
   oxidative stress and membrane lipid peroxidation levels, which serve as
   key indicators of plant stress resistance. Compared with the MFcd
   treatment, the IFcd and IFcd + Si treatments showed a trend of reducing
   H[2]O[2], MDA, and O[2]^− levels, and the IFcd + Si treatment
   significantly lowered these levels by 37.5%, 74.1%, and 90.2%,
   respectively (P < 0.05) (Fig. [93]2a–c). Another analysis of the
   defense enzyme activity in faba bean leaves revealed that both the IFcd
   and IFcd + Si treatments significantly increased the activities of SOD
   (16.5% and 32.9%), APX (15.4% and 40.1%), and POD (80.0% and 209.9%)
   enzymes (P < 0.05). Additionally, IFcd + Si treatment resulted in a
   significant increase in the activities of these enzymes compared to
   IFcd alone (Fig. [94]2d–f). These results indicated that intercropping
   with Si-NPs, both individually and in combination (IFcd + Si), could
   effectively enhance the antioxidant and systemic defense mechanisms of
   faba bean plants under Cd and soil-borne pathogen stress.

Fig. 2.

   [95]Fig. 2
   [96]Open in a new tab

   Effects of intercropping (IFcd) and Si-NPs superposition (IFcd + Si) on
   antioxidant and defense ability of faba bean plant leaves. Panels
   (a)–(c) show the contents of H[2]O[2], MDA, and O[2]^− in plant leaves.
   Panels (d)–(f) show the enzyme activities of SOD, APX, and POD in plant
   leaves. Values are presented as the mean ± SD (n = 3). Different
   lowercase letters indicate significant differences between treatments
   (P < 0.05)

Effects of exogenous Si-NPs and intercropping superposition on rhizosphere
soil physicochemical properties of faba bean

   The available Cd content in the rhizosphere of faba bean could be a
   vital indicator for assessing the effectiveness of planting patterns
   and exogenous measures in treating Cd-contaminated soil. Compared to
   the MFcd treatment, the IFcd and IFcd + Si treatments increased the
   total Cd content to varying degrees while reducing the available Cd
   content. The IFcd + Si treatment significantly reduced both the
   available Cd content (32.2%) and its proportion in total Cd compared to
   MFcd alone (Fig. [97]3a and b). As an important indicator of soil
   ecosystem dynamics and health, soil enzyme activity was significantly
   increased by the IFcd and IFcd + Si treatments compared with the MFcd
   treatment. Notable increases included 24.1% and 27.0% for CAT, 47.5%
   and 101.1% for dehydrogenase, 104.8% and 78.0% for URE, 13.5% and 42.1%
   for Cell, and 23.2% and 42.4% for SUC enzymes (P < 0.05)
   (Fig. [98]3c–h). IFcd + Si treatment further increased the activities
   of NPase, dehydrogenase, Cell, and SUC enzymes compared with IFcd,
   whereas it did not significantly increase CAT and URE enzyme activities
   (P > 0.05).

Fig. 3.

   [99]Fig. 3
   [100]Open in a new tab

   Effects of intercropping (IFcd) and Si-NPs superposition (IFcd + Si) on
   rhizosphere soil physicochemical properties of faba bean. (a) and (b)
   show the total Cd and available Cd contents in the rhizosphere soil.
   (c)–(h) represent the activities of CAT, neutral phosphatase,
   dehydrogenase, urease, cellulase, and sucrase in the rhizosphere soil
   of faba bean, respectively. Values are presented as the mean ± SD
   (n = 3). Different lowercase letters indicate significant differences
   between treatments (P < 0.05)

   Another analysis of the physicochemical properties of faba bean
   rhizosphere soil revealed that compared with the MFcd treatment, both
   the IFcd and IFcd + Si treatments exhibited an increasing trend in pH,
   OM, NH[4]^+-N, and AK content, whereas the NO[3]^−-N and Olsen-P
   contents decreased. The NH[4]^+-N content significantly increased by
   23.1% in the IFcd + Si treatment, whereas the NO[3]^−-N content
   significantly decreased by 72.3% (P < 0.05). The IFcd + Si treatment
   reduced the NO[3]^−-N content by 64.6% compared with IFcd (P < 0.05),
   whereas the increase in other physicochemical factors was not
   significant (Table [101]1). Spearman correlation analysis (|r|>0.6 and
   P < 0.05) showed a significant positive correlation between NO[3]^−-N
   and nutrient cycling enzymes such as URE, whereas Cd was significantly
   negatively correlated with OM, A-K, H[2]O[2], and URE (Figure [102]S1).
   This suggests that intercropping and its reinforcement treatment
   (IFcd + Si) primarily affected the transformation of ammonium and
   nitrate nitrogen, leading to improvements in soil physicochemical
   properties.

Table 1.

   Changes in physicochemical factors of rhizosphere soil of faba bean
   under different treatments
   Treatments pH Organic matter (OM)
   (g kg^− 1) NH[4]^+ -N (mg kg^− 1) NO[3]^−-N (mg kg^− 1) Total
   phosphorus
   (TP)
   (g kg^− 1) Olsen-P
   (AP)
   (mg kg^− 1) Total potassium (TK)
   (g kg^− 1) Available potassium
   (AK)
   (mg kg^− 1)
   MFcd 7.04 ± 0.07^a 27.43 ± 1.32^a 27.48 ± 3.02^b 2.67 ± 0.28^a
   0.53 ± 0.07^a 63.72 ± 3.71^a 0.17 ± 0.00^a 176.00 ± 1.22^b
   IFcd 7.24 ± 0.15^a 30.25 ± 0.68^a 29.85 ± 1.67^ab 2.09 ± 0.26^a
   0.69 ± 0.02^a 47.06 ± 3.56^b 0.18 ± 0.01^a 209.50 ± 4.16^a
   IFcd + Si 7.21 ± 0.25^a 30.22 ± 1.10^a 33.83 ± 1.75^a 0.74 ± 0.04^b
   0.45 ± 0.06^a 53.52 ± 2.59^b 0.17 ± 0.00^a 203.00 ± 4.49a
   [103]Open in a new tab

   Note: Different lowercase letters in the same column indicate
   significant differences between treatments (P < 0.05). Data are
   presented as mean ± SD (n = 3)

Effects of exogenous Si-NPs and intercropping superposition on the
rhizosphere bacterial community of faba bean

   The 16 S rRNA sequencing results of the rhizosphere soil demonstrated
   that compared with the MFcd treatment, both IFcd and IFcd + Si
   treatments showed an increasing trend in the richness (Chao index) and
   diversity (Shannon index) indices. Specifically, IFcd treatment
   significantly reduced the Simpson index, whereas IFcd + Si treatment
   significantly increased the Chao and Shannon indices (P < 0.05)
   (Fig. [104]4a), indicating an increase in bacterial community diversity
   and richness. Both the IFcd and IFcd + Si treatments reduced the number
   and proportion of unique OTUs in the bacterial community compared to
   the MFcd treatment. Analysis of community β diversity revealed a clear
   differentiation between the MFcd and the IFcd and IFcd + Si treatments
   but no significant separation between the IFcd and IFcd + Si treatments
   (Fig. [105]4c), suggesting that the addition of Si-NPs did not
   influence the rhizosphere bacterial community.

Fig. 4.

   [106]Fig. 4
   [107]Open in a new tab

   Effects of intercropping (IFcd) and Si-NPs superposition (IFcd + Si) on
   the rhizosphere bacterial community of faba bean. (a) Alpha diversity
   indices (Chao, Shannon, and Simpson) of bacterial community under
   different treatments (MFcd, IFcd, and IFcd + Si); (b) Venn diagram
   showing bacterial OTU overlap across different treatments; (c) Beta
   diversity analysis of the bacterial community (PCoA); (d) composition
   analysis of bacterial communities at the phylum and genus levels; (e)
   intergroup comparison of dominant bacterial genera (Top 15); (f)
   Spearman correlation analysis between dominant bacterial genera (Top
   15) and environmental factors (Cd, Npase: neutral phosphatase, OM:
   organic matter, TP, TK, AK, URE: urease, NO[3]^−-N, Cell: cellulase,
   pH, H[2]O[2], SUC: sucrase, AP, Si, NH[4]^+-N, Dehydrogenase).
   *0.01 < p ≤ 0.05, **0.001 < p ≤ 0.01, ***p ≤ 0.001

   Analysis of the rhizosphere bacterial composition revealed that
   compared to the MFcd treatment, the IFcd and IFcd + Si treatments
   increased the relative abundance of Actinobacteria, Proteobacteria, and
   Chloroflexi at the phylum level, while reducing the abundance of
   Acidobacteriota. At the genus level, the relative abundances of
   Sphingomonas, Intrasporangium, Marmoricola, and Streptomyces increased,
   whereas that of Gemmatimonas decreased. Further intergroup difference
   analysis revealed significant differences among Sphingomonas,
   Intrasporangium, and Nocardioids. Nocardioids were significantly
   positively correlated with URE, H[2]O[2] activity, and NO[3]^−-N
   content, whereas Sphingomonas was positively correlated with NO[3]^−-N,
   cellulase, pH, H[2]O[2], SUC, AP, and Si content. Intrasporangium was
   significantly positively correlated with URE (P < 0.05). These findings
   suggested that intercropping and Si-NPs treatment primarily recruited
   beneficial bacteria, thereby improving the rhizosphere bacterial
   community by influencing specific enzymes and soil physicochemical
   properties.

Effects of exogenous Si-NPs and intercropping superposition on rhizosphere
metabolites of faba bean

   Metabolomic analysis was conducted on faba bean rhizosphere soil
   samples influenced by exogenous addition of Si-NPs and intercropping.
   PLS-DA results demonstrated a clear separation among the three
   treatments and strong clustering among the repeated samples
   (Fig. [108]5a). The model permutation test parameters R^2Y and Q^2 were
   0.9883 and − 0.5532, respectively, indicating that the PLS-DA model
   performed well and supported further data analysis. Compared to MFcd
   treatment, IFcd treatment resulted in significant changes in 64
   differential metabolites (DMs) (32 upregulated and 32 downregulated),
   whereas IFcd + Si treatment showed significant changes in 26 DMs (10
   upregulated and 16 downregulated) compared to IFcd. Venn analysis of
   all DMs revealed 17 unique DMs for IFcd and six for IFcd + Si compared
   to MFcd. KEGG compound classification of these metabolites grouped them
   into seven major categories (Fig. [109]5f), with flavonoids being the
   most representative, followed by terpenoids and phenylpropanoids.

Fig. 5.

   [110]Fig. 5
   [111]Open in a new tab

   Classification and analysis of differential metabolites in the
   rhizosphere of faba bean using intercropping (IFcd) and Si-NPs
   superposition (IFcd + Si). (a) Partial least squares discriminant
   analysis (PLS-DA) of metabolites between MFcd, IFcd, and IFcd + Si
   treatments; (b) PLS-DA model permutation test; (c) Venn diagram of
   differential metabolites across treatments; volcano plot of
   differential metabolites in (d) IFcd_vs_MFcd and (e) IFcd + Si_vs_IFcd
   comparisons; (f) KEGG compound classification and statistical chart

   Further analysis of the enrichment of dominant DMs revealed that
   compared to MFcd, the IFcd treatment increased the secretion of
   compounds such as scymnol, 5-deoxyribose-1-phosphate, and
   4-hydroxy-5-phenyltetrahydro-1,3-oxazin-2-one, while reducing the
   secretion of compounds such as dehydrotropingosine and cis-p-coumaric
   acid. Compared to IFcd, IFcd + Si treatment increased the levels of
   metabolites, such as ectoine, isopentanol, and gemfibrozil, while
   reducing the levels of compounds, such as isorhamnetin and phthalic
   acid (Fig. [112]6a and b). The differential metabolites influenced by
   IFcd treatment were primarily enriched in certain pathways such as Fc
   gamma R-mediated phagocytosis, GnRH signaling, and carotenoid
   biosynthesis. In contrast, the differential metabolites affected by
   IFcd + Si treatment were mainly enriched in ABC transporters and cell
   cycle-related metabolic pathways (Fig. [113]6c and d).

Fig. 6.

   [114]Fig. 6
   [115]Open in a new tab

   Cluster analysis of rhizosphere metabolites and KEGG enrichment
   analysis of differential metabolites in faba bean using intercropping
   (IFcd) and Si-NPs superposition (IFcd + Si). (a) Cluster heatmap
   analysis of the top 30 differential metabolites in IFcd_vs_MFcd and (b)
   IFcd + Si_vs_IFcd metabolome (based on Bray-Curtis distance); (c) KEGG
   pathway enrichment analysis of differential metabolites in IFcd_vs_MFcd
   and (d) IFcd + Si_vs_IFcd

Joint analysis of exogenous Si-NPs and intercropping superposition on
rhizosphere metabolome and microbiome interactions of faba bean

   Understanding the relationship between metabolites and rhizosphere
   microorganisms is crucial for optimizing crop growth mechanisms through
   intercropping and Si-NPs treatment. Hence, we first analyzed the
   overall correlation between the rhizosphere bacterial community and
   metabolite expression. The results with a Monte Carlo p-value < 0.01
   and M^2 > 0.5 indicated a significant consistency in trends between
   microbial community structure and metabolite expression levels across
   different groups, demonstrating a strong correlation between the two
   datasets (Fig. [116]7a). Further analysis of the relationship between
   major differential metabolites and dominant bacteria (Top 15) revealed
   that Nocardioids, Sphingomonas, and Intrasporangium exhibited varying
   degrees of negative correlation with metabolites such as aflatoxin B2
   and Cinobufagin, while presenting positive correlations with
   antibacterial compounds such as sordarin, lactucin, and
   15-methylpalmate (Fig. [117]7b). These correlations highlight the
   fundamental reasons for the changes in soil physicochemical properties
   and enhancement of crop growth due to the application of exogenous
   Si-NPs and intercropping.

Fig. 7.

   [118]Fig. 7
   [119]Open in a new tab

   Joint analysis of metabolome and microbiome interactions in the
   rhizosphere of faba bean. (a) Correlation analysis between rhizosphere
   metabolites and microbial communities using the Procrustes test; (b)
   Spearman correlation heatmap illustrating the relationship between key
   differential metabolites and dominant bacterial genera (Top 15).
   *0.01 < p ≤ 0.05, **0.001 < p ≤ 0.01, ***p ≤ 0.001

Discussion

   Cd and Fusarium are two major abiotic and biotic stresses affecting
   current faba bean production. The synergistic effect of exogenous
   Si-NPs and intercropping leverages the beneficial properties of Si-NPs
   and interspecific root interactions. This combination enhances the
   activity of rhizosphere soil enzymes and the transformation of soil
   physicochemical factors, diminishes the adverse effects of soil-borne
   pathogens and Cd on host plants, and improves the plant’s stress
   resistance. The findings of this study support the use of exogenous
   Si-NPs to strengthen the ability of intercropped faba bean to mitigate
   Cd toxicity and Fusarium wilt and to elucidate the interaction
   mechanisms between root microbiota and metabolites.

Exogenous Si-NPs and intercropping effectively reduce Cd toxicity in faba
bean and control the Fusarium wilt

   The toxicity of HM Cd can affect crop growth and product safety and
   threaten human health through the food chain [[120]39]. This study
   revealed that compared to the monoculture faba bean under Cd stress
   (MFcd), the wheat-faba bean intercropping (IFcd) significantly reduced
   Cd accumulation in faba bean, promoted plant growth, and reduced
   soil-borne diseases, consistent with the findings of similar studies on
   Sonchus asper (L.) Hill intercropped with faba bean [[121]9]. However,
   some studies have indicated that intercropping faba bean with rapeseed
   (Brassica napus L.), Sedum alfredii Hance, or Arabidopsis alpine can
   promote Cd enrichment in the aboveground parts of faba bean [[122]12,
   [123]40, [124]41]. These differences may be attributed to intercropping
   plant types, soil physicochemical properties, and experimental
   conditions (e.g., hydroponics, soil culture, or field experiments)
   [[125]1]. Additionally, the wheat-faba bean intercropping stimulates
   nitrogen fixation in faba bean through interspecific competition
   [[126]10, [127]42], with nitrogen acting as a key detoxifying factor
   for Cd stress [[128]1]. Nitrogen enhances the biosynthesis of cysteine
   and other compounds, alleviates Cd stress, increases nutrient
   absorption, and reduces forage alfalfa disease [[129]11, [130]43].
   Notably, this study also found that intercropping increased the
   effective Cd content in the faba bean rhizosphere, possibly because of
   the enhanced root interactions between wheat and faba bean, stimulating
   the secretion of organic acids, such as citric and malic acid, which
   may increase the effective Cd content in the rhizosphere [[131]9]. Most
   current intercropping systems focus on using hyperaccumulating plants
   and crop intercropping to study the absorption and toxicity of Cd
   [[132]1, [133]44]. However, few studies have addressed the toxic
   effects of crop-crop intercropping on target crops. This study confirms
   that intercropping faba bean with wheat can mitigate the toxic effects
   of Cd on faba bean.

   This study proved that compared to the IFcd treatment, intercropping
   with Si-NPs (IFcd + Si) significantly reduced Cd accumulation in faba
   bean, decreased rhizosphere Cd availability, reduced the incidence of
   Fusarium wilt, and improved plant growth conditions (Fig. [134]1). Si
   plays a key role in mitigating Cd toxicity in plants by
   compartmentalizing heavy metals [[135]45] and inhibiting their
   transport to aboveground parts [[136]46]. It also strengthens the plant
   cuticle and forms a silica layer in the cell wall, making pathogen
   penetration more difficult and reducing disease occurrence [[137]47].
   As a new type of nanomaterial, Si-NPs offer enhanced stress resistance
   in plants under both biotic and abiotic stresses [[138]14]. Si-NPs have
   demonstrated greater potential for promoting plant growth and stress
   resistance than traditional silicon materials because they can be
   absorbed by plants without the need for silicon-related transport
   proteins [[139]48, [140]49]. The excellent adsorption capacity of
   Si-NPs and their ability to reduce Cd flux in plants are key mechanisms
   for alleviating Cd toxicity [[141]14]. The Si-NPs stacking with
   intercropping treatment exhibited stronger effects in reducing Cd
   toxicity and controlling faba bean Fusarium wilt than the MFcd
   treatment. This could be attributed to the precipitation of Si-NPs in
   plant epidermal tissues, forming silicified cells that could delay and
   hinder the invasion of Fusarium pathogens, thereby enhancing their
   resistance to Fusarium wilt [[142]17, [143]18]. Additionally, we first
   found that the IFcd + Si treatment significantly increased the total
   root length, root length diameter (range 1), and surface area diameter
   (range 1), which could provide a crucial foundation for nitrogen
   fixation, nutrient acquisition, and promotion of plant growth in faba
   bean root nodules.

Enhancement of defense resistance and soil enzyme activity in faba bean
plants through intercropping with exogenous Si-NPs

   The toxic effects of Cd on plants are seen not only in inhibited growth
   but also in disruptions to physiological and biochemical processes,
   such as increased oxidative stress and altered resistance enzyme
   activity [[144]50]. This study suggested that compared to the MFcd
   treatment, both the IFcd and IFcd + Si treatments reduced the levels of
   MDA and O[2]^− and significantly increased the activity of antioxidant
   enzymes such as SOD, APX, and POD. The decrease in oxidation levels and
   Cd content may be related to interspecific root interactions that
   promote Cd-tolerant endophytic bacteria, which in turn enhances the
   activity of oxidoreductases in plants [[145]51]. Plants respond to
   stress by overproducing reactive oxygen species (ROS), sulfur species
   (RSS), and nitrogen species (RNS). If not cleared in time, they can
   cause oxidative damage to cells and even lead to plant death [[146]16].
   Our experimental results indicate that the addition of Si-NPs and
   intercropping can enhance the antioxidant capacity and Fusarium wilt
   resistance of host plants, which is similar to the findings of Sun et
   al. [[147]52] in cucumber. Notably, IFcd + Si treatment was more
   effective than IFcd alone in reducing oxidative stress and increasing
   resistance enzyme activity (Fig. [148]2), indicating that wheat-faba
   bean intercropping combined with Si-NPs synergistically enhanced plant
   antioxidant capacity and defense performance, making it an effective
   strategy for managing Cd and soil-borne pathogen stress. However, it is
   important to note that the Si-NPs effect observed in this study was
   based on Cd stress conditions, and caution should be exercised in
   determining whether Si-NPs can improve the antioxidant capacity and
   promote growth of faba bean in conventional soil.

   Rhizosphere soil is most closely associated with plants and serves as a
   critical reference for assessing the bioavailability of HMs and overall
   plant health through changes in enzyme activity and physicochemical
   properties [[149]53]. This study indicated that IFcd and IFcd + Si
   treatments significantly increased the activities of CAT, NPase,
   dehydrogenase, URE, Cell, and SUC enzymes in the soil, which are vital
   for nutrient cycling, plant growth, and reducing soil-borne diseases
   [[150]25]. Moreover, the IFcd + Si treatment significantly increased
   the NH[4]^+-N and AK contents while reducing the NO[3]^−-N and Olsen-P
   levels (Table [151]1), unlike the IFcd treatment alone. This may be due
   to the enhanced activity of NH[4]^+-N microorganisms facilitated by
   Si-NPs, which have a large specific surface area and strong adsorption
   capacity, thereby promoting phosphorus adsorption. As a result, this
   leads to a decrease in the contents of NO[3]^−-N and Olsen-P in the
   rhizosphere soil. The significant negative correlation between Si and
   Cd, along with their positive correlation with most enzymes, further
   highlighted the beneficial effects of Si-NPs in regulating soil
   properties (Figure [152]S1). These findings indicate that intercropping
   and Si-NPs synergistically enhanced soil enzyme activity and promoted
   nutrient cycling. Furthermore, URE can serve as an enzyme sensitive to
   HM pollution and is an important indicator for evaluating Cd toxicity
   [[153]54]. When applying intercropping and Si-NPs in the field,
   attention should be paid to the potential adverse effects of other
   nutrients, such as managing soil TK content, in mitigating Cd toxicity.

Exogenous Si-NPs and intercropping superposition regulate rhizosphere
microbiota

   Rhizosphere soil microorganisms, often referred to as the “second
   genome”, can serve as vital indicators of crop health [[154]36]. Under
   HMs stress, they typically experience certain changes, such as
   homogenization of species composition and reduced functional diversity
   [[155]13]. This study discovered that intercropping, especially when
   combined with Si-NPs, increased bacterial community diversity and
   richness compared with MFcd treatment. Although the IFcd treatment
   exhibited an improvement trend, the IFcd + Si treatment significantly
   enhanced these microbial parameters (Fig. [156]4a–c), which is
   consistent with previous intercropping studies [[157]29, [158]44].
   Recent studies have suggested that Si can positively influence plant
   diversity and nutrient availability, including soil Si, while reducing
   Cd toxicity [[159]4, [160]25, [161]55]. Both the IFcd and IFcd + Si
   treatments increased the available Cd content in the rhizosphere, but
   the IFcd + Si treatment reduced this content compared to the IFcd
   alone, indicating that the exogenous Si-NPs lowered the Cd
   bioavailability in the soil [[162]56, [163]57]. These changes are
   closely linked to increased rhizosphere microbial diversity [[164]25,
   [165]58]. Interestingly, this study first revealed that, under Cd
   stress, intercropping had a limited effect on enhancing rhizosphere
   bacterial diversity and richness. In contrast, intercropping combined
   with Si-NPs was found to be more effective in improving microbial
   diversity in the rhizosphere of faba bean.

   Actinobacteria, Proteobacteria, and Chloroflexi were the dominant
   bacterial phyla in the intercropping systems and Si-NPs treatments.
   They are known for their roles in decomposing organic matter,
   increasing soil enzyme activity, and promoting soil nutrient cycling
   [[166]59]. Actinobacteria have been reported to produce indole-3-acetic
   acid and catechol acid iron carriers and form complexes with Cd to
   reduce Cd content. The presence of Cd can induce Pseudomonas aeruginosa
   to produce pyocyanin, thereby reducing the number of pathogenic fungi,
   such as Fusarium, thereby resisting soil-borne diseases and promoting
   plant growth [[167]25, [168]60]. Further analysis of bacterial genus
   composition across different treatments revealed that the relative
   abundance of microbial genera in the IFcd and IFcd + Si treatments
   increased with improvements in soil chemical properties and plant
   growth parameters compared with the MFcd treatment. Among them,
   Sphingomonas, Intrasporangium, Marmoricola, Streptomyces, and
   Nocardioids genera significantly increased in the IFcd and IFcd + Si
   treatments. Sphingomonas not only promotes plant growth, inhibits
   pathogens, and participates in nitrogen cycling and antioxidation but
   also has the ability to fix soil heavy metals [[169]2, [170]13].
   Beneficial bacteria such as Streptomyces and Nocardioids play crucial
   roles in promoting plant growth and controlling soil-borne diseases in
   the rhizosphere [[171]61]. Additionally, Nocardioides and Sphingomonas
   were identified as metabolically active bacteria in heavy
   metal-contaminated soils, which could enhance the tolerance of Dahlia
   pinata Cav to Cd [25]. These findings further explain the positive
   effects of intercropping under Cd stress, especially when combined with
   Si-NPs, on reshaping rhizosphere microbial communities and improving
   crop health. However, in this study, the addition of Si-NPs under Cd
   stress did not lead to significant enrichment of beneficial bacteria,
   such as Paenibacillus and Pseudomonas aeruginosa [[172]2, [173]63].
   This could be because of the presence of pathogenic bacteria and Cd
   stress in faba bean continuous cropping, where crops could recruit more
   complex functional microorganisms in the rhizosphere than relying on
   single stress-resistant or growth-promoting bacteria, reflecting a
   plant strategy to balance growth and stress resistance.

Regulation of rhizosphere metabolites by exogenous Si-NPs and intercropping
superposition and their interaction mechanisms with microorganisms

   Changes in plant rhizosphere metabolites can serve as key indicators of
   plant phenotypic responses to environmental stress [[174]39, [175]64].
   Rhizospheric metabolites can significantly influence crop growth and
   disease resistance [[176]5, [177]39]. This study compared the root
   metabolites of monoculture, intercropping, and intercropping enhanced
   with Si-NPs and identified a substantial impact on the rhizosphere
   metabolites of faba bean. Flavonoids were the most abundant metabolites
   (Fig. [178]5f), playing a crucial role in regulating plant-environment
   interactions, recruiting beneficial microbial communities, reducing
   free radical damage, alleviating Cd toxicity, and resisting disease
   invasion [[179]4, [180]65]. Terpenoids also contribute to plant
   adaptation to stress, pathogen inhibition, and induced resistance,
   which exert positive effects [[181]66]. Analysis of differential
   metabolites revealed an increase in ribonucleotides related to DNA
   synthesis, along with metabolites including 5-Deoxyribose-1-phosphate
   and 4-Hydroxy-5-phenyltetrahydro-1,3- oxazin-2-one, which promoted
   plant growth and defense. These metabolites were enriched in KEGG
   pathways related to cell phagocytosis and hormone signal transduction.
   Si-NPs enhance the secretion of antioxidant secondary metabolites such
   as ectoine, which helps regulate plant cell permeability [[182]16]. The
   main enriched KEGG pathways, including ABC transporters and the cell
   cycle, support plant resilience to salt and other stressors [[183]67].

   The plant rhizosphere is a crucial zone for soil-plant-microbe
   interactions [[184]23, [185]24], and the microbiome and metabolome
   within this area play key roles in understanding the mechanisms
   underlying plant oxidative stress responses, healthy growth, and
   disease resistance [[186]18, [187]68]. Our association analysis between
   dominant bacteria and differential metabolites revealed a strong
   correlation between microbial and metabolomic datasets. Microorganisms
   linked to plant stress, including Nocardioides, Sphingomonas, and
   Intrasporangium, exhibited strong positive correlations with sordarin,
   lactucin, and 15-methylpalmate (Fig. [188]7), and these antibacterial
   compounds are connected to both plant rhizosphere defense mechanisms
   and overall plant defense capabilities [[189]69, [190]70]. Previous
   studies have indicated that the absorption of Si-NPs by plants can
   enhance the metabolism of rhizosphere sugars, organic acids, and amino
   acids, improve microbial nutrition, and increase the colonization of
   Cd-tolerant bacteria, thereby reducing Cd toxicity in the rhizosphere
   [[191]68, [192]71]. This study suggested a strong correlation between
   the recruited microbiota and metabolites under Cd stress, whether
   through intercropping or Si-NPs treatment. Notably, compounds such as
   sordarin, lactucin, and 15-methylpalmate were significantly enriched.
   In summary, our findings indicate that intercropping and Si-NPs
   treatment significantly enriched the differential metabolites and key
   microorganisms involved in faba bean root zone defense, offering first
   and important insights into enhancing the resilience of faba bean
   against environmental stress. These findings demonstrate the vital role
   of rhizosphere metabolites in shaping microbial communities and
   strengthening plant resistance to soil-borne diseases. The close
   association between microbes and metabolites explains how intercropping
   and Si-NPs could alleviate the dual stress of soil-borne pathogens and
   Cd by enhancing plant antioxidant capacity, boosting rhizosphere soil
   enzyme activity, reshaping the bacterial community structure, and
   regulating key rhizosphere metabolites, ultimately promoting healthy
   crop growth.

Conclusion

   Our results indicated that compared to the monoculture faba bean under
   Cd stress, IFcd and IFcd + Si reduced Cd accumulation in plants,
   decreased the incidence of faba bean Fusarium wilt, and increased the
   diameter and total root length of the main roots. These treatments also
   enhanced the plant antioxidant and systemic defense capabilities,
   increased the activity of root enzymes such as NPase, dehydrogenase,
   and SUC, improved the physicochemical properties of the soil, regulated
   flavonoid secretion, and promoted the enrichment of functional bacteria
   such as Nocardiides, Sphingomonas, and Intraporangium, along with
   secondary metabolites such as sordarin, lactucin, and 15-methylpalmate
   in the rhizosphere. The close interaction between microorganisms and
   their metabolites revealed the rhizosphere microecological mechanisms
   by which IFcd and IFcd + Si alleviated Cd toxicity and resisted
   Fusarium wilt. Although IFcd alone improved these factors to some
   extent, the regulatory effects of intercropping combined with Si-NPs
   were more pronounced. In summary, this study demonstrated the
   significant advantages of using intercropping and Si-NPs to alleviate
   Cd and pathogen stress in faba bean, with promising potential for
   large-scale agricultural practices in the future.

Electronic supplementary material

   Below is the link to the electronic supplementary material.
   [193]Supplementary Material 1^ (627.6KB, docx)

Acknowledgements