Synergistic application of canna-derived biochar and arbuscular mycorrhizal fungi enhances cadmium immobilization and soil ecological function in contaminated soils Authors: Zequn Tang, Ziwei Jiang, Zhen Liu, Jiaxin Li, Hailan Li, Tao Liao, Liang Qiu Published: 2025 ABSTRACT Cadmium (Cd) is a highly toxic and persistent heavy metal pollutant that poses a serious threat to soil ecosystems and food safety. In this study, biochar derived from three wetland plants (reed, typha, and canna) at various pyrolysis temperatures were evaluated for their capacity to immobilize Cd in contaminated soil. Canna-derived biochar prepared at 500 °C (Ca500) exhibited the most favorable physicochemical properties, including a high surface area and a developed porous structure. Subsequently, ICP-MS analysis, antioxidant activity assays, and microbial community analysis were conducted to further investigate the effects of combined Ca500 and arbuscular mycorrhizal fungi (AMF) application on Cd immobilization in soil and on soil microbial communities. The results revealed that co-application of AMF and biochar (at 6% of soil dry weight) reduced the proportion of transportable Cd in soil from 62.9 to 35.42%, and significantly inhibited the translocation of Cd to plant stems and leaves by 31.6% and 22.6%, respectively, thereby effectively enhancing the edible safety of vegetables. Biochemical analysis showed that the combined application of AMF and biochar synergistically activated the antioxidant defense system, with the activities of SOD, POD, and CAT increasing by 8.9%, 79.1%, and 21.1%, respectively. Meanwhile, the malondialdehyde (MDA) level decreased by 47.2%, indicating alleviation of Cd-induced oxidative damage. Combined analysis of the microbial community using Kraken2 and Bracken revealed that the combined treatment enriched beneficial bacterial genera such as Streptomyces and Pseudomonas, and upregulated genes associated with stress response, sulfur metabolism, and ethanol metabolism, thereby improving soil fertility. In conclusion, the work highlights the ecological advantages of integrating AMF and biochar for sustainable remediation of Cd-contaminated soils, offering both mechanistic insights and practical implications for safe vegetable production.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-025-29752-2. Introduction Heavy metal pollutants in soil represent a persistent and widespread global environmental threat. Among them, cadmium (Cd) is considered one of the most toxic heavy metal due to its high toxicity, persistence, and non-degradability in the environment1,2. Although Cd is a non-essential element for plant growth and reproduction, it has a higher bioaccumulation index in plants than many essential nutrients, posing significant health risks to humans through the food chain3. According to the Chinese Soil Pollution Survey Bulletin, the proportion of sites exceeding Cd pollution standards is as high as 7.0%, making it one of the major contaminants in current soils and arable lands. Cd in soil can be classified into exchangeable, carbonate-bound (CB-bound), Fe–Mn oxide-bound (Fe–Mn-bound), organic matter-bound (OM-bound), and residual fractions. At present, nearly 20% of cultivated land in China is contaminated with Cd. Therefore, controlling Cd pollution and reducing its impact on agricultural products are essential to minimize food safety risks and improve the utilization of farmland4,5. In recent years, immobilization to reduce Cd risk in soil has attracted increasing attention due to its low cost, high efficiency and environmental friendships6,7. The methods of immobilization include chemical and biological methods8,9. Compared with chemical methods, biological immobilization is more environmentally friendly and beneficial to crop growth, as it attenuates the toxic effects of Cd on living organisms without causing secondary pollution or degrading soil properties10. Thus, using microorganisms as an “immobilizing agent” is a promising approach for the remediation of vast areas contaminated with Cd, e.g., agricultural soil11. A variety of biological immobilizing agents have been identified, including members of the genera Pseudomonas, Delftia and Arthrobacter12–14. These microorganisms can immobilize heavy metals in soil through mechanisms such as adsorption, precipitation, complexation, and biotransformation, thereby effectively reducing their bioavailability and mobility. Arbuscular mycorrhizae fungi (AMF), a common soil microorganism, are a parasitic fungus and has a close relationship with plant roots, can promote plant growth, improve plant resistance to heavy metals, and reduce heavy metal transfer to the upper part of plants15. Hence, there is extensive research on using AMF to immobilize Cd16–18. For example, Reda et al.19 found that under Cd stress (2.25 mM), AMF inoculation increased plant growth parameters, chlorophyll, and protein content, and decreased Cd concentration in plants. The main limitation of biological immobilizing agents (such as AMF) is that contaminated soil cannot provide sufficient nutrients for inoculated microorganisms. Additionally, indigenous microorganisms will compete with inoculated microorganisms20. Therefore, improving the survival rate and the available space for microorganisms proliferation is very important. Many studies have attempted to increase the utilization of microorganisms by providing carriers, such as zircon and peat. However, these carriers may alter the soil’s pH balance and activate toxic elements, which can have adverse effects on crop growth21–23. Biochar can provide an ideal growth environment for microorganisms, greatly improving its remediation efficiency due to its porous structure, large surface area, high pH, and nutrient availability (Carbon, potassium, and phosphorus, etc.)24,25. Ma et al. combined biochar with a bacterium (biochemical composite material, BCM) to bioremediate cadmium-contaminated soil. They found that the percentage of acetic acid-extractable Cd in BCM treatments was significantly decreased by 11.34% compared to the control. Meanwhile, BCM significantly increased ryegrass dry weight by 77.78%26. However, current studies on the combined application of biochar and AMF for the remediation of Cd-contaminated soils remain limited. Most previous research has focused on their individual effects—biochar for physicochemical immobilization of Cd and AMF for enhancing plant tolerance27,28—while the interactive mechanisms between these two agents, particularly their influence on soil microbial abundance, community structure, and functional metabolism, are still poorly understood. Therefore, in this work, biochar was pyrolyzed from wetland plants waste (Phragmites australis (reed, Re)), (Typha orientalis (typha, Ty)), (Canna indica (canna, Ca)) at temperatures of 400, 500, and 600 °C. These products were used as representatives to evaluate the immobilization capacity of Cd-contaminated soil through comprehensive studies. Based on these experiments, we further combined the most effective biochar with arbuscular mycorrhiza to study the immobilization properties of Cd in soil. Pot experiments were also carried out to explore the effects of the combined system on the Cd accumulation in plants, antioxidant indicators, and soil microorganisms. Materials and methods Soil samples and reagents Soil samples for incubation experiments were obtained from Fuzhou, Jiangxi Province, which has soil with minimal heavy metal pollution. Samples were taken to the laboratory, air-dried, homogenized, and sieved through a 2 mm sieve. After that, the soil samples were mixed with Cd2+ solution, and the Cd concentration in the soil was controlled at 4 mg/kg. Then, deionized water was used to maintain soil moisture at 70% for 30 days at room temperature. Cd concentration was analyzed using ICP optical emission spectroscopy (ICP-MS; PerkinElmer NexION 300X, USA) following soil digestion with HF + HNO3 + H2SO4 (5: 3: 1)29. The resulting Cd content in soil samples was 3.6 mg/kg. To determine the chemical speciation of Cd in contaminated soil, the Tessier sequential extraction method was employed30. This method utilized the following reagents: acetic acid (HOAc), sodium hydroxide (NaOH), hydrogen peroxide (H2O2), ammonium acetate (NH4-OAc), sodium acetate (NaOAc), nitric acid (HNO3), calcium chloride (CaCl2), and hydroxylammonium chloride (OH-NH2·HCl). The Cd-contaminated soil used in this experiment was artificially prepared using cadmium nitrate Cd(NO3)2. All reagents were purchased from CNW (Shanghai, China) and GL Biochem Ltd. (Shanghai, China). Preparation of biochar Biochar was pyrolyzed from wetland plants waste (reed (Re), typha (Ty), canna (Ca)) at the experiment base (Wuhan, Hubei Province; Institute of Hydrobiology, Chinese Academy of Sciences). Each sample was air-dried, crushed, and sieved through 40 mesh sieves. The temperatures of the pyrolysis process were 400, 500, and 600 °C (each biochar from here referred to as Re400, Re500, Re600, Ty400, Ty500, Ty600, Ca400, Ca500, and Ca600)31. The biochar was subjected to continuous stream of N2 for 4 h to keep the temperature constant. Finally, biochar was crushed and sieved through 100 mesh sieves32. Characterizations The samples were mounted on conductive adhesive and sputter-coating with gold. Surface morphology was analyzed by scanning electron microscopy (SEM) at appropriate accelerating voltages. The specific surface area and pore size distribution were characterized by nitrogen adsorption–desorption measurements using a fully automated surface area and pore size analyzer at 77 K. Prior to testing, the samples were degassed under vacuum at 200 °C for 4 h. The crystal structure of the biochar was qualitatively analyzed using XRD. The test conditions were as follows: Cu Kα X-rays were used as the excitation source, with a tube current of 40 mA and a voltage of 40 kV. The scanning range was 2–60°, and the scanning speed was 2°/min. Immobilization of Cd by different biochar To analyze the immobilization effectiveness of different biochar, each biochar was applied to 500 g of Cd-contaminated soil. Cd-contaminated soil without biochar was used as the control (CK), and three replicates were set for each treatment. Each biochar treatment was done in triplicate. The contact time was 16 weeks, and the Cd concentration was determined at 4, 8, 14, and 16 weeks. Biochar was applied at 10, 20, 30, 40, 50 g (2, 4, 6, 8, 10%). The amount of Cd was determined by ICP-MS. After establishing a calibration curve using a series of standard solutions, the digested, diluted, and filtered sample solutions were directly introduced into the ICP-MS for measurement. The Cd concentration in the soil (mg/kg) was then automatically calculated and reported by the instrument software. Experiment of arbuscular mycorrhizal infection Arbuscular mycorrhiza (G. etunicatum, G. intraradices, and G. mosseae) were purchased from the Institute of Nutrition and Resources, Beijing Academy of Agricultural and Forestry Sciences. The infection rate of arbuscular mycorrhiza fungi on the roots of Chinese cabbage and amaranth was determined using the acetic acid ink method. To start, a 20% KOH solution (v/v) was added to the cleaned 1 ~ 1.5 cm root segment. The segment was heated in a water bath at 60 °C for 60 min. After heating, it was then removed and washed under running water for 5 min, then soaked in acetic acid for five min. The segments were then dyed using an acid ink solution (5% solution, acetic acid: ink, 95 mL:5 mL) in a 60 °C water bath for 30 min. After the water bath, the segments were soaked in water for 12 h to decolorize. Sudan IV staining solution (3 g Sudan IV solid plus 1000 mL 70% ethanol) was used for restaining in a 60 °C water bath for 60 min. The Segments were then rinsed with running water for 3 min, and subsequently soaked in 70% ethanol solution for 5 min. Finally, the dyed root segments were observed under an optical microscope, and the proportion of dyed fibrous roots was calculated. Pot experiment on the combination of biochar and arbuscular mycorrhiza Before the experiment, the AMF spore was mixed with the soil (20 g AMF spore per 1 kg soil). One week later, the soil moisture content was measured, and when the field water-holding capacity stabilized at approximately 60%, seeding was initiated to ensure the successful germination of AMF spores. The experiment comprised four groups: CK (without adding biochar and arbuscular mycorrhiza), BC (biochar only), AMF (adding arbuscular mycorrhiza fungi alone), and AMF + BC (adding arbuscular mycorrhiza fungi and Ca500), and each group was conducted in triplicate. In the pre-treatment phase before planting, biochar was added to the soil at a rate of 6% and mixed thoroughly. Compound chemical fertilizer was added at a ratio of 0.3 g fertilizer per 1 kg of soil to ensure normal plant growth. One day after the pretreatment, several vegetable seeds (Chinese cabbage and amaranth) were added to each container. The soil moisture content was maintained at approximately 70%. All pots were weighed daily at 9:00 a.m., and water was replenished according to the recorded weight loss to maintain consistent soil humidity. The day/night temperature was set to 28/22 °C, with a 14/10 h light–dark photoperiod. No pesticides were applied during the experiment. Seven days later, seedlings were thinned to 10 per incubator (using 50 cm-long, 30 cm-wide, and 20 cm-high plastic planting pots for this). The culture period was 42 days. After harvest, Cd content in soil, plant biomass, Cd content in different parts of the plant, and antioxidant indices (superoxide dismutase (SOD), peroxidase (POD), catalase (CAT), and malondialdehyde (MDA)) were determined. The bioconcentration factor (BF) and transfer factor (TF) were then calculated. DNA analysis of soil microbial samples Soil samples were collected from the experimental soil after harvesting the vegetables in the pot experiment, which involved combining biochar and arbuscular mycorrhiza. A soil sample without treatment was collected as the blank group. The samples were stored at − 20 °C until use. The experimental process was implemented according to the standard protocol provided by Illumina, including sample quality assessment, library construction, library quality assessment, library sequencing, and other steps. The specific operations were as follows: the total microbial DNA in the soil was extracted using the cetyltrimethyl ammonium bromide (CTAB) method33. After passing the quality control assessment, the DNA was sheared into fragments by ultrasonic disruption. The fragmented DNA was then purified, followed by end repair, addition of adenine (A) to the 3’ ends, and ligation of sequencing adapters. Fragment size selection was performed by agarose gel electrophoresis. Sequencing Library (NEBNext® Ultra™ DNA Library Prep Kit for Illumina®) was amplified by PCR according to the manufacturer’s instructions. First, the established libraries were inspected, and then the qualified libraries were sequenced using Illumina NovaSeq34. After sequencing, MEGAHIT assembly software (version 1.2.9) was used for analysis, and the default parameters of MEGAHIT (generated contigs N50 with high quality) were selected for assembly to obtain the sample contigs. Prodigal was used to predict the gene from the assembly results. Once the prediction was completed, CD-HIT (version 4.6) was used to remove redundant annotated genes. The longest sequence in each cluster was selected as a unique gene. The similarity threshold and coverage threshold were set to 95% and 90%, respectively. Statistical analysis Statistical analyses were performed using Excel 2021 and IBM SPSS 20. One-way analysis of variance (ANOVA) followed by Duncan’s multiple-range test was used to compare means among treatment groups, with a significance level set at p < 0.05. The results were visualized using Origin 2021. Results and discussion Selection of biochar and microbial strains Cadmium immobilization performance of different biochar As shown in Fig. 1, Cd in the untreated soil was predominantly present in the exchangeable form (77.15%), followed by the CB-bound form (14.92%), while the Fe–Mn-bound fraction accounted for only 2.27%. After 14 weeks of incubation, the proportion of exchangeable fractions in all experimental groups declined significantly, settling in the range of 60.62 to 73.72% while other fractions increased. In particular, the fraction of Fe–Mn-bound increased sharply and surpassed the CB-bound fraction, reaching 8.43 to 12.42%. Studies have shown that the exchangeable and CB-bound can be released in cationic solutions and easily utilized by microorganisms or plants35. In contrast, cadmium bound to Fe–Mn oxides and organic matter can be converted into exchangeable forms under low-oxygen or acidic conditions, thus becoming available to plants. The residual state is mostly stable but cannot be directly utilized by plants26. Therefore, a reduction in the proportions of exchangeable and CB-bound Cd indicates that Cd in the soil has been effectively immobilized.Fig. 1Immobilization performance of Cd by different biochar. Immobilization performance of Cd by different biochar. Notably, the biochar derived from Canna, especially that produced at 500 °C (Ca500), exhibited the strongest Cd immobilization capacity, effectively reducing the proportions of exchangeable and CB-bound Cd in the soil. Similarly, Wang et al. reported that biochar derived from Conyza canadensis achieved the highest Pb(II) adsorption capacity at 450 °C36. In contrast, the biochars derived from reed (Re400, Re500, Re600) showed relatively poor performance, with a decrease of less than 5% in the labile Cd fraction compared to the control, indicating that the feedstock source plays a critical role in determining biochar effectiveness. XRD results (Fig. 2) revealed the presence of three minerals in the biochar: potassium, quartz, and calcite. Significant differences in mineral composition were observed among biochars derived from different raw materials. Biochar produced from Canna exhibited a higher proportion of carbonate-type minerals, which are more effective in improving soil properties and immobilizing heavy metals. Consistently, the SBET values presented in Table S1 revealed that reed-derived biochar exhibited considerably lower specific surface areas at the same pyrolysis temperature than those from other sources.Fig. 2XRD patterns of biochar prepared from different raw materials and at different pyrolysis temperatures. (a) Reed, (b) Typha, (c) Canna. XRD patterns of biochar prepared from different raw materials and at different pyrolysis temperatures. (a) Reed, (b) Typha, (c) Canna. SEM images (Fig. 3) clearly reveal the microstructural differences among various biochars pyrolyzed at 500 °C. The surfaces of Re500 and Ty500 appear rough and compact, with limited porosity and predominantly covered by fine particulate solids. These biochars lack well-developed interconnected pores, indicating a relatively low specific surface area. In contrast, Ca500 exhibits a more pronounced porous architecture with uniformly distributed pore sizes, consistent with the SBET results, confirming its higher specific surface area. A larger surface area provides more binding sites for cadmium adsorption, enhancing its capacity to capture metal ions and retard their mobility37. Consequently, Ca500 was selected for subsequent experiments. The cadmium immobilization performance of Ca500 under different treatment durations and application rates is presented in Tables S2 and S3, showing that a 6% application rate and 14-week incubation yielded the most effective Cd stabilization. As the biochar dosage increased further, the Cd immobilization efficiency declined. This decrease can be attributed to the aggregation and agglomeration of excess biochar particles, which block pores and adsorption sites, thereby reducing the overall adsorption capacity38.Fig. 3Scanning electron microscopy images. (a) Re500, (b) Ty500, (c) Ca500. The red circle marks the main morphological characteristics of biochar. Scanning electron microscopy images. (a) Re500, (b) Ty500, (c) Ca500. The red circle marks the main morphological characteristics of biochar. AMF inoculation experiment As shown in Fig. 4, the colonization rate of AMF in Chinese cabbage roots were relatively low, with a maximum of 6.32%. This suggests that cruciferous plants are not suitable hosts for AMF and are generally ineffective in establishing symbiotic relationships. Numerous studies have demonstrated that AMF host specificity is closely related to factors such as root structure, root exudates composition, and plant taxonomy. For example, plant families such as Fabaceae, Poaceae, and Solanaceae are widely recognized as highly compatible hosts, with typical colonization rates exceeding 30%39–41. Zhang et al. reported that the colonization rate of Glomus intraradices in the rhizosphere of rice (a Poaceae plant) reached up to 38%, significantly enhancing phosphorus uptake and biomass accumulation42. In contrast, studies on AMF colonization in Amaranthaceae species remain limited. In the present experiment, G. intraradices exhibited notable colonization in amaranth roots, with a colonization rate of 25.01%. This phenomenon may be attributed to the secretion of flavonoids and phenolic metabolites by amaranth roots, which act as chemical signaling molecules that attract AMF hyphae in the soil to colonize the host roots43. Therefore, G. intraradices was selected as the representative AMF strain for subsequent pot experiments, with amaranth used as the host plant.Fig. 4Colonization rate of AMF to different vegetables. Colonization rate of AMF to different vegetables. Immobilization effect of combined system on soil Figure 5 illustrates the distribution of Cd species in soil under different treatment conditions. In the CK group, Cd was primarily present in the exchangeable form, accounting for 62.90% of the total, followed by Fe–Mn-bound (24.76%), OM-bound (4.17%), CB-bound (1.03%), and residual (7.14%). In contrast, the BC and AMF treatments significantly reduced the proportion of labile Cd fractions (i.e., exchangeable and CB-bound forms) to approximately 52% and 46%, respectively, while the residual fraction increased to 10.82% and 14.41%. This trend was more pronounced in the AMF + BC group, where the exchangeable Cd decreased further to 35.42% and the residual fraction rose to 16.41%, indicating a stronger immobilization capacity.Fig. 5Immobilization performances of Cd by combined system. Immobilization performances of Cd by combined system. This transformation pattern suggests that both AMF and biochar, whether applied individually or in combination, can effectively reduce Cd bioavailability and enhance its stabilization in soil. Specifically, AMF improve the rhizosphere microenvironment through symbiosis and secrete glycoproteins such as glomalin-related soil protein (GRSP), which are rich in lipids and methylene groups. These compounds exhibit high affinity for heavy metal ions, forming insoluble complexes that reduce metal mobility and toxicity44. Moreover, AMF may also regulate the redox status of the rhizosphere, promoting the conversion of Cd into less soluble or residual forms45. On the other hand, biochar, characterized by its abundant functional groups, porous structure, and large surface area, can immobilize Cd via electrostatic adsorption, ion exchange, surface complexation, and redox reactions, thereby reducing the proportion of exchangeable Cd46. The superior fixation performance observed in the AMF + BC group indicates a synergistic effect between AMF and biochar under the current experimental conditions. AMF appears to play a dominant role in Cd immobilization, while biochar primarily provides a supportive, synergistic effect. These results align with previous studies47–49, further validating the potential of AMF + BC remediation strategies to reduce heavy metal bioavailability and improve soil environmental quality. Effect of combined system on plants Effects on biomass and root length To investigate the effects of the combined application of biochar and AMF on heavy metal translocation and plant physiology in Cd-contaminated soil, the biomass and primary root length of plants in each treatment group were first measured. As shown in Fig. 6, the biomass and root length of amaranth in the CK group were 0.20 g and 3.33 cm, respectively. In the BC group and AMF group, biomass increased significantly, whereas root length did not. In the AMF + BC group, both biomass and root length showed significant increases, with biomass reaching 0.81 g and root length extending to 5.11 cm. This result is consistent with most previous studies. Li et al. found that the combined application of AMF and biochar significantly enhanced the biomass of reed rhizomes, increased root length, root diameter, and root volume, and improved nutrient uptake in reed50. Yang reported that the application of AMF combined with biochar promoted potato growth by enhancing net photosynthetic rate, transpiration rate, stomatal conductance, and chlorophyll content in leaves51. These effects may be attributed to the biochar’s alkaline nature, which can improve soil pH and water retention capacity. In addition, biochar can significantly increase soil organic matter, available nitrogen and phosphorus, and rapidly available potassium, while suppressing the loss of nitrogen and phosphorus in soil, thereby improving fertilizer utilization efficiency and enhancing soil suitability for plant growth52. Furthermore, biochar is rich in carbon sources and can increase soil cation exchange capacity and redox potential, thereby enhancing the abundance of soil microbial communities and enzyme activities, improving soil nutrient availability, and ultimately increasing nutrient use efficiency53.Fig. 6The effects of the combined system on biomass (a) and root length (b) of amaranth. The effects of the combined system on biomass (a) and root length (b) of amaranth. Distribution of Cd content, bioconcentration coefficient, and transfer coefficient in plants Figure 7 illustrates the Cd content in different parts of amaranth. In the CK group, Cd content in roots, stems, and leaves was 1.213 mg kg−1, 0.642 mg kg−1, and 0.623 mg kg−1, respectively, far exceeding the national safety limit for leafy vegetables (< 0.2 mg·kg−1). Both the bioaccumulation factor (BCF) and translocation factor (TF) were greater than 1 (Table S4), indicating that the plant readily absorbed Cd from the environment and transported it to the aerial parts, posing a high ecological risk. Under the BC treatment, Cd content decreased by 20.6%, 29.7%, and 27.6% in roots, stems, and leaves, respectively. Under the AMF treatment, these values decreased by 15.7%, 31.6%, and 22.6%, respectively. However, in both treatments, the Cd content in the edible parts still exceeded the food safety limit. In contrast, the combined application of AMF and biochar significantly reduced Cd levels in all parts of the plant compared to both the CK and AMF groups (P < 0.05), with roots, stems, and leaves Cd contents recorded at 0.953, 0.243, and 0.280 mg kg−1, respectively. These results suggest that the proportion of Cd translocated from roots to aerial tissues was greatly reduced, with Cd preferentially retained in roots, thereby enhancing the safety of the edible parts.Fig. 7The Cd content in different part of amaranth. The Cd content in different part of amaranth. This is consistent with findings by Sidra, who reported that the co-application of biochar and AMF reduced heavy metal uptake in wheat, with increased accumulation in the roots and decreased Cd concentrations and translocation factor in the shoots54. The observed effect may be attributed to enhanced AMF colonization promoted by biochar, where hydroxyl and carboxyl groups on AMF hyphae can bind metal cations and immobilize Cd in the root zone. Moreover, AMF may induce root lignification and branching, thereby strengthening root retention capacity and limiting translocation to the shoots55,56. Antioxidant activity of plants Antioxidant enzymes (SOD, POD, CAT) and the lipid peroxidation product malondialdehyde (MDA) are key biochemical indicators used to evaluate the physiological stress resistance of plants under heavy metal exposure. As shown in Fig. 8, both AMF and AMF + BC treatments activated the antioxidant defense system to varying extents, effectively alleviating oxidative damage caused by Cd stress.Fig. 8The antioxidant enzymes of different groups. The antioxidant enzymes of different groups. Specifically, SOD activity in the CK group was 604 U g−1 fresh weight, which increased to 629.4 U g−1 under the BC treatment and to 682 U g−1 under AMF treatment, representing increases of 4.2% and 12.9%, respectively. However, SOD activity in the AMF + BC group was 622 U g−1, only 2.9% higher than in the CK group and notably lower than in the AMF-alone group. This reduced SOD activity may be attributed to pH alterations induced by biochar, changes in rhizosphere microecology, reprogramming of plant oxidative metabolism, and differential enzyme responses under combined treatment57. POD activity showed a more pronounced increase across treatments, rising from 777 U g−1 in the CK group to 866, 863 and 991 U g−1 in the BC, AMF and AMF + BC groups, corresponding to 11.4%, 11.1% and 27.5% increases, respectively. The synergistic effect of biochar significantly enhanced POD activity, likely due to improved soil aeration, elevated microbial activity, and the promotion of beneficial rhizosphere microbial colonization, collectively enhancing the plant’s capacity to scavenge reactive oxygen intermediates58. For the CAT activity, the AMF group exhibited a slight decline compared to CK (67.3 vs. 69.9 U g−1 fresh weight, a 3.5% decrease). In contrast, BC and AMF + BC treatments significantly increased CAT activity to 89 and 125.2 U g−1, corresponding to 27.3% and 79.1% increases, respectively. This indicates that biochar may enhance H₂O₂ decomposition capacity by improving soil physicochemical properties and facilitating AMF symbiosis, thereby reducing the damage caused by cadmium accumulation59. Moreover, MDA content was significantly reduced in both treatment groups. While the CK group recorded 157.6 nmol g−1, BC, AMF and AMF + BC groups decreased to 128.5, 115.4 and 83.1 nmol g−1, representing reductions of 18.5%, 26.7% and 47.3%, respectively. This suggests that both treatments effectively mitigated Cd-induced lipid peroxidation, reducing oxidative damage to plant cell structures60. Analysis of soil microbial community Composition and analysis of species diversity Α-diversity indices were used to species richness and evenness within each sample. As shown in Table 1, the Berger, Parker, Simpson, and Shannon indices were lower in the AMF and AMF + BC treatment groups than in the Control, indicating a reduction in dominant taxa and microbial diversity after AMF inoculation. This may be due to increased competition for rhizosphere resources induced by AMF, which suppressed the growth of some low-abundance taxa61,62. Notably, the Chao1 index was highest in the AMF + BC group, suggesting that this treatment maintained the greatest species richness.Table 1The α-diversity index of community.SampleBerger-parkerchao1SimpsonShannoneBlank0.036445250.002947.23Control0.043445820.003597.3AMF + BC0.040650050.003417.28AMF0.04145610.00337.29 Table 1The α-diversity index of community.SampleBerger-parkerchao1SimpsonShannoneBlank0.036445250.002947.23Control0.043445820.003597.3AMF + BC0.040650050.003417.28AMF0.04145610.00337.29 The α-diversity index of community. Differences in microbial communities between samples were further assessed by the beta diversity. Principal coordinate analysis (PCoA) based on Bray–Curtis distance (Fig. 9) revealed that the first two axes explained 68% and 29% of the β-diversity variation, respectively. The first axis clearly separated the control group from the treatment groups, while the second axis distinguished the biochar-amended groups from the others, indicating that the co-application of biochar and AMF significantly altered the soil microbial community structure.Fig. 9PCoA plot based on bray–curtis distances. (Group1: AMF + BC; Group2: AMF). PCoA plot based on bray–curtis distances. (Group1: AMF + BC; Group2: AMF). Further analysis of microbial composition (Fig. 10) showed that Proteobacteria was the dominant phylum across all samples. At the genus level, the AMF + BC treatment markedly reduced Chlamydia abundance while enriching several functional genera, including Streptomyces, Pseudomonas, Variovorax, Hydrogenophaga, and Sphingopyxis. Among them, Streptomyces can secrete high-affinity siderophores that not only chelate iron but also bind cadmium, forming Cd-Fe carrier complexes. These complexes can further associate with Fe–Mn oxides in the soil, converting Cd into a more stable Fe–Mn-bound form63. In addition, Pseudomonas can effectively adsorb and immobilize Cd, transforming it into an OM-bound form64. These results suggest that the combined application strategy not only reshaped microbial community structure but also promoted colonization by beneficial microorganisms, providing a microbial foundation for the ecological remediation of Cd-contaminated soils65.Fig. 10The diagram of soil microbial abundance. ((a) Phylum; (b) Genus). The diagram of soil microbial abundance. ((a) Phylum; (b) Genus). Gene function annotation and differential analysis KEGG annotation analysis of the four sample groups (Fig. 11a) identified 198,443 functional genes, distributed across four major functional categories: cellular processes, environmental information processing, genetic information processing, and metabolism. Among these, the most abundantly expressed pathways were related to genetic information processing, including DNA transcriptional regulation, signal transduction mediated by protein phosphorylation, and methylation66. Consistent results were obtained from the GO functional annotation analysis (Fig. 11d), which closely aligned with the KEGG findings. These results suggest that remediation treatments may regulate transcriptional activity and signal transduction efficiency in microbial populations, thereby influencing community structure and ecological function.Fig. 11(a) Result classification diagram of KEGG pathway69. (b) Density distribution of transcript expression (A: blank, B: control, C: AMF + BC, D: AMF). (c) PCA result of transcript expression (Group1: AMF + BC, Group2: AMF). (d) Result classification diagram of GO term. (a) Result classification diagram of KEGG pathway69. (b) Density distribution of transcript expression (A: blank, B: control, C: AMF + BC, D: AMF). (c) PCA result of transcript expression (Group1: AMF + BC, Group2: AMF). (d) Result classification diagram of GO term. The results of the normalized gene abundance density distribution (Fig. 11b) and PCA plots (Fig. 11c) shows that the overall gene expression abundance in the AMF and AMF + BC groups was significantly higher than in the control and blank groups, indicating that AMF inoculation markedly enhanced the functional expression levels of soil microorganisms, with biochar addition further reinforcing this effect to some extent. Additionally, the AMF and AMF + BC groups clustered closely in the PCA plot, suggesting high similarity in their gene composition, while both showed distinct separation from the control and blank groups. The differential gene enrichment bubble plot (Fig. 12) further revealed the directional characteristics of functional transitions. Compared with the control group, the AMF group showed significant enrichment in pathways such as macromolecule biosynthesis and cyanophycin synthetase activity (p < 0.05). In contrast, the AMF + BC group exhibited a more complex enrichment profile, including misfolded protein degradation, RNA helicase activity, heat stress response, and sulfate-transporting ATPase activity. These results indicate that biochar enhances the environmental adaptability and stress resistance of microbial communities by modulating microbial stress response systems and sulfur metabolism pathways67. Further comparison between the AMF + BC and AMF groups revealed that the ethanol metabolism pathway was the most significantly enriched (p < 0.05), suggesting that biochar stimulated the activity of certain fermentative genera, thereby enhancing carbon source conversion efficiency68. In addition, the largest number of differentially expressed genes (five in total) between these two groups was associated with bacterial flagellar assembly, suggesting that biochar intervention may affect microbial motility and rhizosphere colonization behavior.Fig. 12Enrichment bubble diagram of different group ((a) blank to control, (b) AMF to control, (c) AMF + BC to control, (d) AMF + BC to AMF). Enrichment bubble diagram of different group ((a) blank to control, (b) AMF to control, (c) AMF + BC to control, (d) AMF + BC to AMF). Conclusion This study evaluated the immobilization efficiency of biochar derived from reed, typha, and canna at different pyrolysis temperatures for the remediation of Cd-contaminated soil. Among them, canna biochar produced at 500 °C (Ca500) showed the best performance. Ca500 was selected for combination with AMF to investigate their synergistic remediation mechanisms and integrated effects on the plant–soil system. Ca500 featured a well-developed porous structure and high surface area. It effectively reduced soil bioavailable Cd. The AMF + BC treatment exhibited a clear synergistic effect, further decreasing Cd mobility and accumulation in amaranth to levels approaching food-safety standards. It also enhanced plant growth, boosted antioxidant enzyme activities (POD, CAT), and mitigated oxidative stress. Microbial community analysis revealed that AMF + BC significantly enriched Streptomyces, Pseudomonas, and other functional genera involved in pollutant degradation and stress tolerance. This treatment also modulated sulfur metabolism and stress-response pathways, indicating that biochar provided a favorable habitat and metabolic support for beneficial microbes. The combined application of Ca500 biochar and AMF offers an effective, eco-friendly strategy for the remediation of Cd-contaminated soils, minimizing the negative impacts on plants and the environment. However, as the current findings are based on pot experiments, further field trials are necessary to validate the results under more complex environmental conditions and to provide a solid theoretical basis for large-scale Cd remediation practices. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1