Synthetic bacterium-facilitated colonization of nitrogen-fixing bacteria for remodeling the rhizosphere microbiome and improving plant yield Authors: Shuo Liu, Zirun Zhao, Yuxin Ji, Hangqi Zhu, Yingfeng Sun, Mingchun Li, Qilin Yu Published: 2025 ABSTRACT AbstractBackgroundNitrogen-fixing bacteria (NFBs) play a critical role in biological nitrogen fixation for supplying essential nitrogen nutrients to plants in agriculture and natural ecosystems. Especially, these bacteria and Leguminosae plants form symbiosis to improve plant growth and soil fertility. Theoretically, the inoculation of NFBs into soils increases biological nitrogen fixation, but the efficiency of NFBs is frequently compromised by the low capacity of NFB root colonization. In this study, we introduced the synthetic bacterium EcCMC, which was genetically engineered to express the surface-displayed artificial polysaccharide (PS)-recognizing protein Cmc, to test if it can improve NFBs root colonization in representative Leguminosae plants, including Astragalus sinicus and Medicago sativa. Rhizosphere microbiomes, biochemical indicators, and plant yields were evaluated after 28 days in the three treatments, i.e., the control group without addition of any exogenous bacterium, the NFBs plus EcM (bacteria only expressing mCherry rather than Cmc) group, and the NFBs plus EcCMC group (n = 3).ResultsOwing to its polysaccharide-binding capacity, EcCMC strongly bound to the surface of A. sinicus roots. This binding was followed by the increased recruitment of the exogenous NFBs, Sinorhizobium meliloti and Sphingomonas endophytica, on the roots. As revealed by amplicon sequencing of the 16S rRNA gene, a combined inoculation of EcCMC and the NFBs increased the relative abundance of both Rhizobiales and Sphingomonadales, two important bacterial groups involved in nitrogen fixation. Consistently, metabolomic analysis showed that the metabolites involved in nitrogen fixation remarkably accumulated in the rhizosphere soils inoculated with NFBs plus EcCMC. Moreover, inoculation of NFBs plus EcCMC increased the activity of nitrogenase from 10.8 ~ 11.3 to 16.2 nmol/min/g (significant difference, p < 0.05, t-test), together with the total soil nitrogen levels from 217 ~ 258 to 414 mg/kg (significant difference, p < 0.05), and the soil organic matter levels from 19.5 ~ 20.8 to 23.6 mg/kg (significant difference, p < 0.05). Consequently, the yield of A. sinicus was remarkably improved by the inoculation of NFBs plus EcCMC. Similar results were observed in the experiments using Medicago sativa.ConclusionsThis study sheds a novel light on a synthetic biology-assisted regulation of rhizosphere microbiomes for enhanced nitrogen fixation and soil fertility in Leguminous plants. The designed polysaccharide-binding protein may be used as a universal tool to promote plant growth and enhance crop resilience in the future. Video AbstractGraphical Abstract Supplementary InformationThe online version contains supplementary material available at 10.1186/s40168-025-02189-5.Keywords: Rhizosphere microbiome, Nitrogen-fixing bacteria, Synthetic bacterium, Colonization, Green manure Introduction Rhizosphere microbiomes include microbial communities closely associated with the plant roots [1–3] and influenced by the root metabolites [4–6]. These microbiomes can have positive or negative effects on plant performance, including growth and defence. The beneficial rhizosphere microbiomes, such as the plant growth-promoting rhizobacteria (PGPR), are indispensable for plant growth. These microbes may produce growth-promoting agents and facilitate the release of bioavailable nutrient elements from insoluble minerals [7–9]. Consequently, rational regulation of rhizosphere microbes is of great importance for phytoremediation, soil fertility promotion, and crop production [10–13]. With the advances of synthetic biology, more and more efforts have been devoted to remodel natural microbiomes with rhizosphere inoculants [14–16]. For example, synthetic microbial communities involved in nutrient utilization have been designed and applied to assist the rhizosphere microbiomes in improving plant growth or plant disease resistance [17, 18]. Previously, we also constructed a polysaccharide (PS)-recognizing synthetic bacterium, which enhanced phytoremediation by strengthening the physical contact between natural PS-producing microbes and plant roots [19]. In addition, to avoid the potential ecological risk of artificial DNA, an engineered protein was further designed to improve the interaction between functional microbes and plant roots [20]. Nitrogen-containing compounds are critical nutrients for plant growth in agriculture. Traditionally, the well-known chemical nitrogen fertilizers, e.g., ammonium and nitrate, have been widely used to provide inorganic forms of nitrogen for plant productivity. However, their wide application in agriculture leads to their abundant leaching in the underground water and severe ecological pollution problems [21–23]. Therefore, there is a growing need to develop sustainable strategies for reducing the use of chemical nitrogen fertilizers. As such, biological nitrogen fixation is a green and economic way for the conversion of N2 to bioavailable nitrogen forms [24–26]. The nitrogen-fixing bacteria (NFBs), commonly functioning as PGPR, provide about 200 million tons of nitrogen to the agricultural lands per year [27–30]. Leguminous plants serve as nutrient suppliers in economic crops to improve soil fertility and crop yield. Some of these manures, e.g., Astragalus sinicus and Medicago sativa, are also used as food for animals [31, 32]. Owing to the symbiosis between Leguminous plants and rhizosphere NFBs, these plants import considerable organic nitrogen compounds into the soil during their growth [33, 34]. Theoretically, inoculation of exogenous NFBs could increase the degree of nitrogen fixation. However, the activity of these NFBs is frequently compromised by their low root colonization efficiency, leading to rapid replacement of these NFBs by indigenous rhizosphere microbes (i.e., the microbes inhabiting the soil before the exogenous NFBs are introduced) [35, 36]. Up to now, it remains a great challenge to enhance the root colonization and persistence of these functional bacteria. The development of synthetic biology techniques is strongly promoting rational regulation of the rhizosphere microbiome [19, 20]. To enhance the root colonization of exogenous NFBs in legumes, this study introduced the synthetic bacterium (i.e., EcCMC), which was genetically engineered to express the surface-displayed artificial polysaccharide (PS)-recognizing protein Cmc. Our previous studies have revealed that this bacterium may bind to the roots of bioremediation plants via Cmc, leading to the remodeling of the rhizosphere microbiome [19]. Herein, we further investigated the effect of this synthetic bacterium on NFB colonization, rhizosphere microbiome remodeling, and the growth of two typical Leguminous plants (i.e., Astragalus sinicus and Medicago sativa). This study aimed to develop a facile synthetic biology-based strategy for promoting the growth of legumes during soil fertility improvement. Materials and methods Strains The synthetic E. coli strains EcM and EcCMC were constructed and described in our previous study [19]. Briefly, the E. coli strain BL21 at the competence state was transformed using the pET-28a-based plasmids pET-28a-CMC and pET-28a-M, obtaining the strains EcCMC and EcM, respectively. The synthetic E. coli strain EcCMC expresses the Cmc protein containing the glucan-binding domain of Clostridium cellulolyticum CipC (CBMCipC), mCherry, and another CBMCipC domain. To realize the expression of Cmc, the gene CMC was de novo synthesized using the optimized codon of E. coli and cloned into pET-28a, generating pET-28a-CMC for further transformation. The control E. coli strain EcM expresses the fluorescence protein mCherry not containing the glucan-binding domain. The NFB strains, including Sinorhizobium medicae NKN632 and Sphingomonas endophytica NKN643, were isolated from the rhizosphere soil of Vicia dasycarpa, Tianjin, China. Both strains are free-living nitrogen-fixers and can promote plant growth under low levels of nitrogen fertilizers. The solid BNN medium (glucose 5 g, CaCl2 0.15 g, MgSO4 0.1 g, FeSO4·7H2O 0.04 g, NaMoO4 0.005 g, K2HPO4 0.8 g, dH2O 1000 mL, agar 15 g, pH 6.8 ~ 7.0) was used for this isolation. The exogenous control bacteria not producing extracellular polysaccharides, i.e., Pseudomonas soli NK1431 and Achromobacter denitrificans NK531, were isolated from the rhizosphere soils of Amygdalus persica plants, Shandong Province, China. The LB medium (yeast extract 5 g, tryptone 10 g, NaCl 10 g, dH2O 1000 mL, agar 15 g) was used to isolate the two strains. Both NFBs and the control strains were cultured at 30 °C for 24 ~ 48 h for further use. Plant culture To culture the green manure plants, the seeds of Astragalus sinicus and Medicago sativa were surface-sterilized using 75% ethanol for 5 min and 2% NaClO for 10 min. We washed the surface-sterilized seeds with sterile water 10 times for 1 min. The surface-sterilized seeds were incubated in sterile water at room temperature for 2 h and cultured on the sterile agar medium (agar 1 g, distilled water 100 mL) for 48 h. The above experiments were performed in clean flowerpots. The germinated seeds were planted in sterile medium (consisting of equal volumes of perlite and vermiculite) in flowerpots (three seeds per flowerpot) and irrigated with Fahareus nitrogen-free nutrient solution as needed. Fahareus nutrient solution was prepared by mixing equal volumes of solutions A and B. Solution A consisted of MgSO4·7H2O (0.12 g), Na2HPO4·12H2O (0.15 g), KH2PO4 (0.10 g), and distilled H2O (0.5 L). The pH of solution A was adjusted to 6.8 ~ 7.2. Solution B consisted of CaCl2·2H2O (0.01 g), ferric citrate (5.0 mg), and distilled H2O (0.5 L). The pH was adjusted to 6.8 ~ 7.2 with NaOH solution (10 M). Gibson trace element solution consisted of H3BO3 (2.86 g), Na2MoO4·2H2O (0.13 g), CuSO4·5H2O (0.08 g), ZnSO4·7H2O (0.22 g), and distilled H2O (1 L). The prepared Gibson solution was filtered to remove microbes. The mixture of solutions A and B was autoclaved at 121℃ for 20 min. After cooling to room temperature, 1 mL of Gibson trace element solution was added to the mixture for further use. After culturing for 30 days under natural sunlight and at room temperature, the roots of healthy plants were sampled for further binding assays. For evaluating the effect of the bacterial cells on the microbiome composition and plant growth, the young plants were transferred into new flowerpots containing 1 L of sterile soil for further treatment of the bacterial cells. Bacterium binding assays The capacity of the synthetic bacteria (i.e., EcCMC and EcM) to bind the exogenous NFBs, including Sinorhizobium medicae NKN632 alone, Sphingomonas endophytica NKN643 alone, NKN632 plus NKN643 at equal cell numbers (i.e., NFBs), and the exogenous control bacteria not producing extracellular polysaccharides (i.e., Pseudomonas soli NK1431, Achromobacter denitrificans NK531) were evaluated. To obtain the synthetic bacterial cells, EcCMC and EcM were resuscitated on the solid LB medium containing 50 mg/L kanamycin from the refrigerator at − 80 ℃. The bacterial cells from single colonies on the LB plates were inoculated into the liquid LB medium containing 50 mg/L kanamycin (50 μg/mL) liquid medium and cultured overnight at 37 ℃ in a shaking bed at 160 rpm. Then, 100 μM of isopropyl-beta-d-thiogalactopyranoside, which induces expression of Cmc or mCherry in the synthetic cells, was added to the cultures. The cultures were further incubated with shaking at 37 ℃ and 160 rpm for 8 h, followed by centrifugation at 12,000 rpm for 3 min, obtaining the EcCMC or EcM cells. The cells were suspended in PBS (pH = 7.4) to a final concentration of 1 × 108 cells/mL. The exogenous bacterial cells were obtained using similar procedures. Briefly, Sinorhizobium medicae NKN632 and Sphingomonas endophytica NKN643 were cultured in liquid BNN medium at 30 ℃ for 24 h. Pseudomonas soli NK1431 and Achromobacter denitrificans NK531 were cultured in liquid LB medium at 30 ℃ for 24 h. The cells were also harvested by centrifugation at 12,000 rpm for 3 min, stained with 10 mg/L DAPI, and then suspended in PBS (pH = 7.4) to a final concentration of 1 × 108 cells/mL. One milliliter of the EcCMC or EcM suspension was mixed with 1 mL of the exogenous bacterium suspensions and incubated at 30◦C for 30 min. Ten microliters of FITC-labeled concanavalin A (FITC-ConA, 5 mg/mL, prepared in PBS, GeneTex, USA) was added to the suspensions at a final concentration of 10 mg/L, followed by observation with confocal microscopy (Olympus, Japan). To evaluate the ability of EcCMC and EcM cells to recognize NFB polysaccharides, these polysaccharides were extracted from NKN632 and NKN643 cultures using hot water, followed by alcohol precipitation and vacuum drying. Five milligrams of the purified polysaccharides were suspended in 10 mL of 10 mM NaHCO3 buffer (pH = 9.0) containing 0.1 mg of FITC. The mixture was incubated at room temperature with shaking at 60 rpm for 12 h, and the unbound dye was removed by dialysis, obtaining the FITC-labeled NFB polysaccharides (FITC-NFB PS). The freshly cultured EcCMC and EcM cells were resuspended in PBS containing the prepared FITC-NFB PS (20 mg/L), stained for 30 min, and washed using PBS for confocal microscopy. The Pearson correlation coefficient (PCC) was calculated by the ImageJ software (version 1.53e) to reflect the co-localization degree of the fluorescence. Root binding assays Several fresh and young roots were cut by scissors from healthy Astragalus sinicus plants and washed using PBS. The roots were divided into three groups, including the control group with no treatment, the NFBs + EcM group for NFBs and EcM binding, and the NFBs + EcCMC group for NFBs and EcCMC binding, each group was replicated six times. The roots were immersed in PBS, and then, the freshly cultured functional bacteria (NFBs plus EcM or NFBs plus EcCMC) were added into the suspensions with a concentration of 1 × 107 cells/L and incubated at 30◦C for 1 h. Notably, NFBs were stained using 4’,6-diamidino-2-phenylindole (DAPI, 5 mg/mL) before incubation with the roots. The roots were gently washed using PBS, stained with the FITC-labeled ConA (FITC-ConA, green fluorescence, indicating polysaccharides), and observed using confocal microscopy (A1 + , NIKON, Japan). The root-removing suspensions were further sampled, and the number of viable NFBs was evaluated using the colony-forming unit (CFU) assays on the BNN medium plates. The CFUs were counted after 72 h of incubation at 30◦C. The percentage of root-binding bacteria was calculated using the following formula: (the CFUs of the control supernatant—the CFUs of the treated suspensions after root removal)/the CFUs of the control suspension × 100%. Root colonization assays The fresh Astragalus sinicus roots were immersed in PBS (pH = 7.4) containing NFBs plus EcM or NFBs plus EcCMC as described above for 48 h. The roots were fixed by 4% formaldehyde, dehydrated by ethanol (30%, 50%, 70%, 95%, 100%), and freeze-dried using a vacuum desiccator. The obtained samples were observed using a scanning electron microscope (SEM, TESCAN MIRA LMS). For histopathological observation, fresh roots in each group were fixed using 4% formaldehyde, sliced, embedded in paraffin, and then stained with hematoxylin and eosin for further microscopy. Rhizosphere microbiome analysis The plants were cultured in flowerpots as described in Sect. 2.2. On day 14 after the seeds were planted into the soil, the suspensions of NFBs and the synthetic bacteria were added into the flowerpots as described below: the control group with addition of 20 mL PBS, the NFBs + EcM group with addition of 20 mL bacterial suspension containing 5 × 108 NKN632 cells, 5 × 108 NKN643 cells and 10 × 108 EcM cells, the NFBs + EcCMC group with addition of 20 mL bacterial suspension containing 5 × 108 NKN632 cells, 5 × 108 NKN643 cells, and 10 × 108 EcCMC cells. Twenty-eight days after inoculation, the roots from six plants were randomly selected, excised, and mixed as a sample. Each treatment was replicated three times. The total rhizosphere microbial DNA of each sample was extracted using TGuide S96 magnetic bead extraction kits. The libraries of 16S rDNA (bacteria) and ITS (fungi) were constructed using the bacterial 16S V3 + V4 primers (338F, 5′-ACTCCTACGGGAGGCAGCA-3′; 806R, 5′-GGACTACHVGGGTWTCTAAT-3′) and the ITS primers (ITSF, 5’-CTTGGTCATTTAGAGGAAGTAA-3’; ITSR, 5’-GCTGCGTTCTTCATCGATGC-3’) for analysis of rhizosphere bacteria and fungi, respectively. The constructed libraries were sequenced at Biomarker Technologies Corporation, China, and analyzed with the online platform (http://www.biocloud.net). The sequences of each sample were clustered into operational taxonomic units (OTUs) based on a 97% sequence similarity using USEARCH (Edgar, 2013, version 10.0). The OUT abundances across samples were normalized by the abundance sums per sample divided by the OTU abundance for each sample. The relative abundance and α-diversity values between the groups were compared by ANOVA analysis and the post hoc test with a P value < 0.05. Metabolomic analysis Six roots were evenly selected across the three-dimensional space of each plant and mixed as a sample with a weight of 50 mg, and each group had six samples (n = 6). The samples were suspended in a mixture of 2 mL methanol, 2 mL acetonitrile, and 1 mL water containing 20 mg/L of the internal reference substance, and then broken by a grinder for 10 min. The mixture was then kept in a refrigerator at − 20 °C for 1 h and centrifuged at 4 °C and 12,000 rpm for 15 min. The supernatants were sampled for further analysis. The metabolites in the supernatants were separated by Waters UPLC Acquity I-Class PLUS system, using 0.1% formic acid in water as the mobile phase A, and 0.1% formic acid in acetonitrile as the mobile phase B. The separation time was 15 min, and the flow velocity was 400 μL/min. The metabolites were then detected by the Waters UPLC Xevo G2-XS QTOF system. The detection parameters were demonstrated as follows: the capillary voltage at 2500 V or − 2000 V, the core voltage at 30 V, the ion source temperature at 100 °C, the desolvation gas temperature at 500 °C, the backflow velocity at 50 L/h, the desolvation gas velocity at 800 L/h, and the m/z range at 50 ~ 1200. The identified compounds were classified, and the pathways were annotated using Kyoto Encyclopedia of Genes and Genomes (KEGG, version 2). The t-test was used to determine a significant difference at P < 0.05 between the two groups. The P value of the orthogonal partial least-squares discriminant analysis (OPLS-DA) model was adopted to screen the differential metabolites. The differential metabolites between the two groups were screened with a fold change (FC) > 2 or < 0.5. Determination of soil microbial and biochemical parameters To evaluate the soil living NFB numbers in the rhizosphere soil, 1 g of the fresh rhizosphere soil samples in each group (n = 3) was added to 100 mL of sterile distilled water containing glass beads, and shaken for 5 min. The soil suspensions were then diluted using sterile water with 10 gradients. One hundred microliters of the diluted suspensions was added to the BNN medium plates. The CFUs were counted after 72 h of incubation at 30◦C. The soil urease (S-UE), soil acid phosphatase (S-SAP), and soil catalase (S-CAT) were determined using S-UE kit (Solarbio, BC0120), S-ACP kit (Solarbio, BC0140), and S-CAT kit (Solarbio, BC0100), according to the manufacturer’s instructions. The determination methods are detailed in the manual. The total nitrogen (TN) of the soil was measured using the Kjeldahl method (assisted by the Shiyanjia Lab). In this method, 1 g of soil samples was digested with 20 mL concentrated H2SO4 containing 6 g K2SO4 and 0.2 g CuSO4, followed by ammonia release using 10 mL 40% NaOH solution (W/V). The released ammonia was collected using 10 mL 2% H3BO3 solution (W/V) and titrated using the standard HCl solution. Soil organic matter (SOM) was measured using a chemical oxygen demand detector (Hangzhou Qiwei Instrument Co., LTD). Briefly, 100 mg of an air-dried soil sample (sieved through a 50-mesh sieve) was poured into 5 ml of deionized water to obtain a suspension. One milliliter of the suspension was added to the working solution, incubated at 165℃ for 30 min, cooled, and tested. The solution concentration was determined according to the standard curve, and then, the SOM content in the soil was calculated. Nitrogenase activity was tested using the acetylene reduction assay [37]. The rhizosphere soil was placed in a vial and cultured at 30℃ for 48 h. The vial cap was replaced with a rubber plug under sterile conditions, and 1 mL of the gas in the vial (10% volume of the vial) was extracted with a sterile syringe, then 1 mL of acetylene was injected and cultured at 30℃ for 48 h. A sterile syringe was used to extract 0.2 mL of the vial gas. The ethylene contents in the gas were determined using GC–MS (7890A-5975C, Agilent, USA), which reflects the activity of nitrogenase. All treatments were replicated three times. Biochemical analysis and SEM observation of roots and leaves For biochemical analysis, six roots (and leaves) were obtained from each plant and three plants were sampled in each treatment. The contents of chlorophyll a and b in the plant leaves were detected according to our previously reported method [19]. The mature leaves of each group were selected at the same position of the plants and fixed using 4% formaldehyde. The fixed leaves were dehydrated using 30%, 50%, 70%, 95%, and 100% ethanol and dried in a vacuum desiccator. The obtained samples were observed by SEM. Statistical analysis Most of the experiments were performed using three replicates (n = 3). Metabolomic analysis was repeated six times (n = 6). The data were shown with the means ± the standard errors. Statistical analysis of plant biomass and biochemical factors was performed by t-test or the type I ANOVA method and the post hoc test. Multiple comparisons were done between the NFB + EcCMC group and the control group, and between the NFB + EcCMC group and the NFB + EcM group. The significance of the difference between the groups was determined at P < 0.05, using the Statistical Product and Service Solutions (SPSS) software (version 22.0, IBM, USA). Results Exogenous NFB fails to efficiently colonize rhizosphere soil Initially, two nitrogen-fixing bacteria (i.e., Sinorhizobium medicae NKN632 and Sphingomonas endophytica NKN643), which were isolated from the rhizosphere soil of Vicia dasycarpa, were inoculated in the rhizosphere soil of Astragalus sinicus to explore the growth-promoting effects of these exogenous nitrogen-fixing bacteria. As shown in Fig. S1a, the CFU numbers of the control group without inoculation of exogenous bacteria increased with the increase in culturing time, which may be attributed to the slow proliferation of local bacteria inhabiting the Astragalus sinicus roots. Upon inoculation of NKN632, NKN643, or both strains together (NFBs), the number of nitrogen-fixing bacteria in rhizosphere soil increased significantly. However, after 14 or 28 days, the number of nitrogen-fixing bacteria between the inoculation treatment and the control did not show a significant difference. Although inoculation of functional bacteria into the rhizosphere soil is an ideal strategy in sustainable agriculture [38–41], we did not obtain ideal results by using the two exogenous bacteria. Nitrogenase activity and total nitrogen (TN) contents were further measured. Interestingly, the inoculation of NKN632 or NKN643 alone had no impact on the nitrogenase activity and total nitrogen (TN) contents in the rhizosphere soil, and the inoculation of both strains (NFBs) only slightly increased both indicators (Fig. S1b, S1c). For example, the NFBs only led to a slight increase in the nitrogenase activity from 11.0 to 11.5 μmol/(h·g soil) (Fig. S1b) and to an increase in TN contents from 214.8 to 242.5 mg/kg after 28 days (Fig. S1c). Similarly, after 28 days of inoculation, the mixed NFBs, or NKN632 and NKN643 alone, had only a mild impact on plant fresh weight (FW, Fig. S1d). This indicated that the exogenous NFBs, when they were solely inoculated, failed to efficiently colonize the rhizosphere. These results suggest that other strategies are needed to improve the colonization and duration of NFBs in the rhizosphere soil for a significant improvement of nitrogen fixation and plant yield. Therefore, in the following experiments, we used synthetic biology techniques to improve root binding and colonization of NFBs. The synthetic bacterium EcCMC strongly binds polysaccharide-exposed NFBs Previously, we found that the introduction of polysaccharide-binding synthetic bacteria (EcCMC) enhanced the interaction between plant roots and functional bacteria, which further improved the accumulation of heavy metal cadmium, leading to enhanced phytoremediation [19]. Moreover, the introduced EcCMC was not detectable after 14 days of culturing, suggesting that the potential ecological risk caused by the synthetic bacteria is very small, and its application can be further expanded. To investigate whether these proteins could also function during binding and colonization of NFBs on roots, the distribution of polysaccharides on NFBs and its binding to the synthetic bacterium exposing polysaccharide-binding proteins (i.e., EcCMC) were observed by using confocal microscopy. After staining of the bacterial cells by FITC-ConA, which specifically binds polysaccharides, the nitrogen-fixing bacteria NKN632 and NKN643 alone or in combination (NFBs) were strongly stained by FITC-ConA, while the control strains NK1431 and NK531 did not bind the dye (Fig. S2a, Fig. S3). Co-localization analysis by the ImageJ software further revealed that the NFBs had a higher PCC than the control strain NK1431 (Fig. S2b). Therefore, the used NFBs exposed polysaccharides on their cell surfaces, while the control strains did not expose these macromolecules. Based on these results, we further introduced the engineering bacterium EcCMC. It has a strong capacity to recognize polysaccharides commonly distributed on the root surface, owing to the high-degree surface exposure of the polysaccharide-binding protein Cmc on EcCMC [19]. As revealed by confocal images, EcCMC exhibited strong green fluorescence after its incubation with FITC-labeled polysaccharides extracted from the NFB cell surface (FITC-NFB PS, Fig. S2c & S2d). In contrast, the control EcM cells, which did not express the Cmc proteins, failed to bind the FITC-NFB PS (Fig. S2c). Moreover, the mCherry fluorescence of Cmc was intactly co-localized with the FITC fluorescence on the EcCMC cells, exhibiting a much higher PCC than that on the control EcM cells (0.87 versus 0.17, Fig. S2d). Hence, the synthetic EcCMC cells, rather than the control EcM cells, could efficiently recognize and bind the polysaccharides produced by NFBs. The high capacity of EcCMC to recognize polysaccharides, which are also distributed on the cell surfaces of NFBs, implies that the root-binding EcCMC may have a high potential to recruit NFBs. As expected, the control EcM cells could not bind the NFBs, with the two kinds of tested cells randomly dispersing without co-localization in the observation fields (Fig. S2e). In contrast, the EcCMC cells strongly bound to the NFBs, leading to drastic aggregation of both cells, with colocalization with the polysaccharides indicated by FITC-ConA (Fig. S2e). Consistently, the EcCMC cells had a higher PCC than the control EcM cells (0.83 versus 0.25), together with a higher PCC of mCherry and FITC (0.80 versus 0.17, Fig. S4). Together, these results confirmed that EcCMC efficiently bound to the NFBs and induced their aggregation. EcCMC improves root recruitment and colonization of exogenous NFB As previously reported, the synthetic bacterium EcCMC can recognize natural polysaccharide-binding microbes and mediate the binding of these microbes to the alive and dead roots of Eichhornia crassipes [19]. Although EcCMC belongs to the intestine-inhabiting species E. coli, it facilitated the root colonization of functional bacteria. It was not detectable on day 14 after its addition [19], indicating that its positive effect on the colonization of functional bacteria just depended on its cell surface-exposed PS-binding protein rather than on its persistent proliferation. Therefore, we assumed that EcCMC could also be used to assist NFB colonization on Leguminosae roots. To verify this, a model terrestrial Leguminous plant, Astragalus sinicus, was used to evaluate the root-binding capacity of EcCMC and its effect on NFB colonization. After incubation of the A. sinicus plants in the soil containing EcCMC and DAPI-stained NFB for observation. As shown in Fig. 1a, NFBs and EcM were hardly observed on the roots of the NFBs + EcM group, indicating that EcM failed to mediate the binding of NFB to the plant roots. In contrast, the roots of the NFBs + EcCMC group exhibited both strong blue (on behalf of NFB) and red (EcCMC) fluorescence. Moreover, the blue and red colors overlapped with the green (root) fluorescence (Fig. 1a, Fig. S5), demonstrating that NFB efficiently bound to the surface of roots in the presence of EcCMC. Meanwhile, the NFBs exhibited a layer-like morphology rather than signal rods (Fig. 1a), indicating that these bacterial cells formed dense biofilms coating the roots. Consistently, the CFU assays of the root-removing suspensions showed that more than 90% of NFBs bound to the roots in the NFBs + EcCMC group, while less than 20% NFBs bound to the roots in the NFBs + EcM group (Fig. 1b), confirming that ScCMC enhanced the NFBs’ root colonization. EcCMC-assisted recruitment and colonization of NFBs to the A. sinicus roots. a Confocal microscopic images of the roots treated by NFBs plus EcM or NFBs plus EcCMC for 1 h. b Percentage of root-binding bacteria revealed by CFU assays of the suspensions after removal of the roots. c SEM images of the treated roots after 48 h of incubation. The regions indicated by the red squares were amplified and demonstrated in the images below. The white arrows indicate the distribution of NFB on the roots. d Percentage of root-colonizing bacteria based on CFU assays. The asterisks (*) indicate significant difference between the groups (ANOVA, P < 0.05) Root colonization of NFBs was further evaluated by 48 h of co-incubation. SEM observation showed that the roots of both the control and NFBs + EcM group only bound a few bacterial cells on the root surfaces, while the roots of the NFBs + EcCMC group bound a high number of bacterial cells (Fig. 1c). Similarly, the percentage of root-binding CFUs representing NFBs and the synthetic bacteria was higher in the NFBs + EcCMC group compared with the control and NFBs + EcM group (Fig. 1d). These results indicated that EcCMC improves the NFBs’ root colonization. EcCMC and exogenous NFB drastically remodel rhizosphere microbiome The microbial diversity of the rhizosphere microbiome was analyzed by bacterial 16S rDNA and fungal ITS sequencing. The rarefaction curves in each sample flattened out with the increase of sequencing amount, suggesting that the sample sequences were sufficient for analysis (Fig. S6). The NFBs + EcCMC treatment had no significant impact on the bacterial Simpson index, but increased the fungal Simpon index (Fig. 2a), indicating the higher rhizosphere fungal diversity of the NFBs + EcCMC group. Abundance analysis of bacterial community further revealed that NFBs + EcCMC led to the increase in the relative abundance of several bacterial groups (Fig. 2b). Especially, the relative abundance of Rhizobiales, Sphingomonadales and Xanthomonadales in the NFBs + EcCMC group increased by 0.494-fold (0.240 versus 0.154), 0.345-fold (0.074 versus 0.055), and 0.676-fold (0.062 versus 0.037), respectively, as compared to that in the Control group. Remodeling of A. sinicus rhizosphere microbiome by NFBs with the assistance of EcM (NFBs + EcM) or EcCMC (NFBs + EcCMC) after 28 days of culturing (n = 3). a Simpson index of the rhizosphere microbiome. b Relative abundance of bacterial community of the rhizosphere microbiome. c Statistical analysis of bacterial abundance involved in biofilm formation in each group. These bacteria include Sphingomonadales and Xanthomonadales, etc. d Relative abundance of the biofilm-formed bacterial orders in each group. e The number of NFBs in rhizosphere soil revealed by CFU assays. f Relative abundance of fungal community of the rhizosphere microbiome. The unidentified and unclassified_Fungi indicate the unknown fungal data and the known fungi without classified into the listed orders, respectively. The asterisks (*) indicate significant difference between the NFBs + EcCMC group and other groups (ANOVA, posthoc test, P < 0.05) It is reported that many soil bacteria adhered to the root surfaces and were organized to form biofilms [42]. The formation of biofilm on the root surface further assists bacterial survival and stability near the rhizosphere [43]. The relative abundance of biofilms in each group was then tested. As shown in Fig. 2c, the abundance of biofilm-formed bacteria in the NFBs + EcCMC group was significantly higher than that in the control and NFBs + EcM groups, indicating that more bacteria could be colonized on the root surface to form biofilms. The microbial diversity of the bacterial orders forming biofilms was further analyzed. As shown in Fig. 2d, the NFBs + EcCMC group had the highest relative abundance of Rhizobiales, Sphingomonadales and Xanthomonadales. The Xanthomonadales order was responsible for early colonization and community establishment [44], while due to the bridging role of EcCMC, much more PGPR strains, such as Rhizobiales and Sphingomonadales tended to colonize on the root surfaces. Consistently, as revealed by CFU assays, the NFBs + EcCMC group had the highest amount of NFBs as compared with the control and NFBs + EcM group (Fig. 2e), confirming the enhanced root colonization of NFBs by EcCMC. ITS sequencing was further used to detect the fungal abundance in rhizosphere soil. As shown in Fig. 2f, the relative abundance of Pleosporales, which was associated with nitrogen fixation [45], increased by 1.87-fold (0.43 versus 0.15) in the NFBs + EcCMC group compared with the control group. In addition, fungal polysaccharide-binding assays revealed that the EcCMC cells could adsorb much higher levels of the polysaccharides (Fig. S7), suggesting that the binding of EcCMC to fungal polysaccharides was involved in the change in fungal abundance. Taken together, the introduction of NFBs plus EcCMC strongly remodeled the rhizosphere microbiome of A. sinicus. Metabolomic analysis of the control, NFBs + EcCMC, and NFBs + EcM groups was then performed. Principal component analysis (PCA) showed the remarkable difference between the groups and sample distribution in each group. As shown in Fig. 3, the control and NFBs + EcCMC groups had significantly different metabolites as the PCA dots were completely separated (Fig. 3a), while the NFBs + EcCMC and the NFBs + EcM groups exhibited less difference (Fig. 3b). A higher level of group differentiation was obtained by using the orthogonal partial least-squares discriminant analysis (OPLS-DA). Moreover, all samples in the same treatment were gathered in the same area (negative or positive), and all the extrapolated intercept values of Q2 were less than 0.05 (Fig. S8), indicating the models in this study have good predictability [46]. Metabolomic analysis of the Astragalus sinicus rhizosphere soils treated by NFB with the assistance of EcM (NFBs + EcM) or EcCMC (NFBs + EcCMC) after 28 days of culturing. a PCA comparison of NFBs + EcCMC and control groups. b PCA comparison of NFBs + EcM and NFBs + EcCMC. c Comparative metabolomic profiling of the root-microbiome system of NFBs + EcCMC versus that of the control group. The differential metabolites with P value < 0.05 and fold change (FC) > 2 (red dots) or < 0.5 (blue dots) were marked with color. d Comparative metabolomic profiling of the root-microbiome system of NFBs + EcCMC versus NFBs + EcM. e Metabolic pathway enrichment map of different metabolites in NFBs + EcCMC and control groups. f Metabolic pathway enrichment map of different metabolites in NFBs + EcCMC and NFBs + EcM groups EcCMC and exogenous NFBs influence rhizosphere metabolism of A. sinicus Differential metabolite analysis was further performed between the control and NFBs + EcCMC groups as well as the NFBs + EcCMC and NFBs + EcM groups. The volcano map in Fig. 3c showed that 438 metabolites of the NFBs + EcCMC group were increased compared with that of the control group, including Tragopogonsaponin B, Avermectin B1a, Adonixanthin, Desoxoglabrolide, Cynarasaponin F, etc. Moreover, 82 metabolites of the NFBs + EcCMC group were increased compared with the NFBs + EcM groups (Fig. 3d). The pathway enrichment analysis of different metabolites in the three groups was conducted with the KEGG (Kyoto Encyclopedia of Genes and Genomes) database. There was a significant difference between the NFBs + EcCMC and control groups in six metabolic pathways, including alpha-linolenic acid metabolism, inflammatory mediator regulation of TRP channels, histidine metabolism, purine metabolism, ascorbate and aldarate metabolism, and serotonergic synapse (Fig. 3e). On the contrary, the NFBs + EcCMC and NFBs + EcM groups had different metabolites in eight metabolic pathways, including cutin, suberine and wax biosynthesis, sulfur metabolism, biosynthesis of alkaloids derived from ornithine, lysine and nicotinic acid, cAMP signaling pathway, glucagon signaling pathway, and serotonergic synapse (Fig. 3f). Together, these results revealed that NFBs plus EcCMC had an impact on rhizosphere metabolism of A. sinicus, which may be related to the remodeled rhizosphere microbiome. EcCMC and exogenous NFBs promote soil fertility and growth of A. sinicus To test the effects of EcCMC cells on soil fertility, the nitrogen-fixing capacity and enzymatic activity of rhizosphere soils were investigated. The activity of nitrogenase, a key catalyst for biological nitrogen fixation [47], was first tested. As shown in Fig. S9a, the nitrogenase activity of the NFBs + EcCMC group was significantly higher than the control and NFBs + EcM groups (16.2 nmol/min/g versus 10.8 ~ 11.3 nmol/min/g), The increased nitrogenase activity may be attributed to the higher abundance of Rhizobiales and the higher numbers of nitrogen-fixing bacteria in rhizosphere soil of the NFBs + EcCMC group induced by EcCMC plus exogenous NFBs (Fig. 2b,e). The total nitrogen (TN) and soil organic matter (SOM) are two typical indicators of soil fertility [48, 49]. As shown in Fig. S9b,c, both TN and SOM values of the rhizosphere soil in the NFBs + EcCMC group were higher than that in the control and NFBs + EcM groups, i.e., 414 mg/kg versus 217 mg/kg (control) and 258 mg/kg (NFBs + EcM) for TN, 23.6 mg/kg versus 19.5 mg/kg (control), and 20.8 mg/kg (NFBs + EcM) for SOM. Soil enzyme activity is an important indicator for soil health [50]. The urease, acid phosphatase (ACP), and catalase (CAT) activity of rhizosphere soil was tested using the corresponding assay kit. As shown in Fig. S9d–f, the NFBs + EcCMC group exhibited > 0.333-fold higher urease activity (1250 versus < 937 U/g, Fig. S10d), > 0.149-fold higher ACP activity (1870 versus < 1628 U/g, Fig. S9e), and > 0.252-fold higher CAT activity (8.26 versus < 6.60 U/g, Fig. S9f) than the control and NFBs + EcM groups. Taken together, the above results revealed that the inoculation of both NFB and the synthetic EcCMC cells significantly improved the soil fertility. Rhizosphere microbiome plays an important role in plant growth. Since the inoculation of EcCMC as well as NFB significantly remodeled the rhizosphere microbiome and improved the soil fertility, it may further affect root growth. After 14 days of culturing, the complete structures of all roots were observed through histological observation. As revealed by Fig. 4a and Fig. S10, while the roots of the control and NFBs + EcCMC groups had abundant damaged cells, the NFBs + EcCMC group only displayed quite few damaged cells. Additionally, the roots of NFBs + EcCMC group were much longer (11.25 versus < 8.25 cm, Fig. 4b) and stronger (0.344 versus < 0.271 mm in diameter, Fig. 4c) than other groups, as well as higher nitrogen, phosphorus, and potassium content in dry root (Fig. 4d), demonstrating the synthetic EcCMC cells had positive effect in root growth. Characteristics of A. sinicus roots treated by NFB with the assistance of EcM (NFBs + EcM) or EcCMC (NFBs + EcCMC) after 28 days of culturing. a Histological observation of the roots. b Root length. c Root diameter. d Content in dry root of Astragalus sinicus with different treatments. The asterisks (*) indicate significant difference, while the letters “ns” indicate no significant difference between the groups (ANOVA, P < 0.05) The effect of EcCMC on plant growth and leaf metabolism was then tested. As shown in Fig. 5, after 14 days of culturing, the Astragalus sinicus of NFBs + EcCMC group exhibited > 0.375-fold higher plant height (17.6 versus 10.2–12.8 cm, Fig. 5a), > 0.367-fold higher fresh weight (2.46 versus 1.44–1.8 g, Fig. 5b) and > 0.417-fold higher leaf area (34 versus 18–24 cm2, Fig. 5c). Chlorophyll quantification showed that the NFBs + EcCMC group had the highest chlorophyll a and chlorophyll b level compared with control and NFBs + EcM group (> 1.5 versus < 1.2 mg/g FW for chlorophyll a and > 0.8 versus < 0.6 mg/g FW for chlorophyll b, Fig. 5d). Moreover, compared with control and NFBs + EcM group, the NFBs + EcCMC group had higher N, P and K content in dry leaf (about 3% versus < 2.2% for N, 0.286% versus < 0.243% for P, and 1.557% versus < 1.286% for K, Fig. 5e), higher content of soluble protein (32.5 versus < 27.5 mg/g FW, Fig. 5f) and more dense protrusion structure on the leaf surface (Fig. 5g). Consequently, the inoculation of NFBs and EcCMC significantly improved the growth of plants and leaves. Plant growth and leaf metabolism of Astragalus sinicus treated by NFBs with the assistance of EcM (NFBs + EcM) or EcCMC (NFBs + EcCMC) after 28 days of culturing. a Plant height; b fresh weight; c leaf area; d chlorophyll a and b content; e nitrogen, phosphorous, and potassium content in dry leaf; f soluble protein; and g SEM images of leaf for Astragalus sinicus with different treatments. The asterisks (*) indicate significant difference, while the letters “ns” indicate no significant difference between the groups (ANOVA, P < 0.05) EcCMC and exogenous NFBs promote growth of Medicago sativa To explore the effect of NFBs and EcCMC on the growth of other legumes, M. sativa, another typical green manure in China, was further used for treatment. After culturing of M. sativa in soil containing EcCMC and NFBs for 7 days, the roots were sampled and processed for confocal microscopy and SEM observation. As shown in Fig. S11a, the blue and red fluorescence of NFBs + EcCMC group overlapped with the green (roots) fluorescence, while NFBs and EcM were hardly observed on the roots of NFBs + EcM group, demonstrating NFBs could colonize on the surface of roots in the presence of EcCMC. Consistently, SEM results showed that the roots of both control and NFBs + EcM group could barely adsorb bacteria on the root surface, while much more bacteria could be observed on the roots of the NFBs + EcCMC group (Fig. S11b). The abundance analysis of bacterial and fungal communities in the M. sativa rhizosphere soils was further investigated. Rarefaction curves revealed that the sequencing data were also enough for further analysis (Fig. S12). Similar to the results in the A. sinicus culturing experiments, the NFBs + EcCMC group had the higher relative abundance of Rhizobiales (Bactereria), Sphingomonadales (Bactereria), and Pleosporales (Fungi), which are important participants in nitrogen fixation (Fig. 6a). Meanwhile, the NFBs + EcCMC group had the highest abundance of the fungal order Pleosporales, as compared with the control and NFBs + EcM group (Fig. 6b). CFU assays further confirmed that the NFBs + EcCMC group displayed the highest amount of NFBs in the rhizosphere soil (Fig. 6c). Moreover, the NFBs + EcCMC group had the highest Shannon index (Fig. S13), indicating its highest bacterial diversity. Metabolomic analysis of the samples also showed a significant difference between the NFBs + EcCMC group and the other two groups (Fig. S14–17). Effect of NFBs + EcM and NFBs + EcCMC on microbiome composition, nitrogen fixation, and plant growth of Medicago sativa after 28 days of culturing. a, b Relative abundance of a bacterial and b fungal community of the rhizosphere microbiome. c The amount of nitrogen-fixing bacteria in rhizosphere soil as revealed by CFU assays. d Soil nitrogenase activity. e Soil total nitrogen. f Soil organic matter. g Representative images of the cultured plants. h–j Plant height (h), fresh weight (i), and leaf SEM images (j) of Medicago sativa with different treatments for 28 days. The asterisks (*) indicate significant difference, while the letters “ns” indicate no significant difference between the groups (ANOVA, P < 0.05) With the higher amount of beneficial microbiome in the rhizosphere soil, the NFBs + EcCMC group exhibited > 0.259-fold higher nitrogenase activity (23.86 versus < 18.95 nmol/min/g, Fig. 6d), > 0.529-fold higher TN (509.2 versus < 333.1 mg/kg, Fig. 6e), and > 0.164-fold high SOM (29.71 versus 25.52 mg/kg, Fig. 6f) than the control and NFBs + EcM groups. After 14 days of culturing, the Medicago sativa plants treated by EcCMC and NFBs was significantly higher and stronger than the other two groups (Fig. 6g), with > 0.818-fold higher plant height (37.21 versus < 20.47 cm, Fig. 6h) and > 0.559-fold higher plant FW (33.72 versus < 21.63 g, Fig. 6i). Additionally, SEM images showed that more dense protrusion structure could be observed on the leaf surface of the NFBs + EcCMC group (Fig. 6j). All the results indicated that the NFBs plus EcCMC had a positive effect in plant growth. Discussion Leguminous plants are effective plants to improve farmland, increase soil organic matter, reduce the use of chemical fertilizers, and increase crop yield. Owing to the symbiosis between these plants and rhizosphere nitrogen-fixing bacteria, Leguminosae plants provided considerable organic nitrogen compounds into the soil during their growth. With the increasing awareness of green environmental protection, the utilization of Leguminous plants and biological nitrogen fixation resources has attracted high attention. In theory, inoculating exogenous nitrogen-fixing bacteria can enhance the growth of leguminous green manure, but how to colonize nitrogen-fixing bacteria quickly and effectively in plant roots is an urgent problem to be solved. Previously, our group constructed a synthetic bacterium (i.e., EcCMC) exposing the artificial polysaccharide-recognizing protein Cmc, which strongly bound to polysaccharides and recruited natural polysaccharide-producing bacterial and fungal cells [19]. However, the specific receptor recognition pathway mediating the binding of EcCMC and the roots remains to be investigated. The synthetic bacteria were demonstrated to increase the microbial biomass on the roots and strongly remodel the rhizosphere microbiome, improving plant growth and oxidative stress resistance in water. To evaluate the regulation ability of rhizosphere colonies in terrestrial plants, in this study, EcCMC was used to enhance the colonization of exogenous NFBs on the plants of legumes for improved yield. This enhancement may be attributed to the function of the synthetic bacterium as a bridge, supporting the physical binding of the functional NFBs to plant roots via exposed polysaccharides. Consequently, the colonization and biofilm formation of the NFBs were strengthened. The colonized NFBs further supplied more nitrogen nutrients to the plant roots for the promotion of plant growth as compared to free-living NFBs. Microbiome analysis revealed that the introduction of EcCMC and NFBs increased the relative abundance of PGPR. Several bacterial phyla (Rhizobiales, Sphingomonadales, and Xanthomonadales) and fungi phyla (Pleosporales) associated with N fixation were significantly upregulated. Rhizobiales are the bacteria that enter the plant cells to form symbiosomes in legumes and perform the biological N2 fixation [51], while Sphingomonadales bacteria are important participants in free-living N2 fixation. The increase of beneficial phyla in the rhizosphere might promote plant growth [52, 53]. Since the two kinds of exogenous NFBs, i.e., Sinorhizobium medicae NKN632 and Sphingomonas endophytica NKN643, were added in both the NFBs + EcM group and the NFBs + EcCMC group, the result that the NFBs + EcCMC group had higher abundance in both Rhizobiales and Sphingomonadales than the NFBs + EcM group (Fig. 2a) indicated the enhanced root colonization of exogenous NFBs by EcCMC. Moreover, the biofilm amount of the NFBs + EcCMC group was higher than other groups, indicating that more bacteria could be colonized on the root surface. Moreover, the amount of nitrogen-fixing microbes in the rhizosphere soil of the NFBs + EcCMC group was obviously upregulated compared with other groups. These results implied that EcCMC facilitated both bacterial and fungal adsorption and growth on the root surfaces, making PGPR strains fully exert their functions to promote plant growth. Consistently, in plant experiments, the parameters for plant growth (including plant height, fresh weight, and levels of leaf chlorophyll a and chlorophyll b) and soil fertility (enzymatic activity of rhizosphere soils, total nitrogen, and soil organic matter) in the NFBs + EcCMC group were significantly higher than other groups. These results indicate that the introduction of EcCMC can recruit functional microorganisms to the rhizosphere, fully exerting their promoting effect on plant growth. Microbiome analysis revealed that the above results and our previous findings indicate that EcCMC can promote the colonization of nitrogen-fixing bacteria, remodel the rhizosphere microbiome, and further improve plant yield. Through screening, more functional microorganisms can be introduced into the plant rhizosphere using this method, achieving effects such as promoting plant growth, soil fertilization, and soil remediation. Conclusion In summary, this study developed a synthetic bacterium-assisted colonization of exogenous nitrogen fixation bacteria techniques for enhanced nitrogen fixation and soil fertility in the application of Leguminous plants. Owing to its polysaccharide-binding capacity, the synthetic EcCMC could act as a “bridge” to improve the colonization of nitrogen-fixing bacteria on the roots. The remodeled microbiome caused by EcCMC further promoted soil fertility and plant growth. With the positive effect of the synthetic bacteria as well as nitrogen fixation bacteria on green manure, this study provides a strategy of engineered bacteria-based regulation of rhizosphere microbiome to improve biological nitrogen fixation as well as soil fertility. Supplementary Information Supplementary Material 1. Acknowledgements This work was supported by National Key R&D Program of China (2024YFD1701100), Joint Funds of the National Natural Science Foundation of China (U23A20158), National Natural Science Foundation of China (32170102), Foundation of Nankai University-Cangzhou Bohai New Area Green Chemical Research Institute (NCC2022-PY-09), and the Fundamental Research Funds for the Central Universities (63253191). Authors' contributions Y.Q. and L.S. conceived and designed the study, L.S., Z.Z. and J.Y. performed the experiments, L.S. and Z.H. drafted the manuscript and analyzed the data, Y.Q., L.M. and S.Y. provided critical reviews of the manuscript, figures and text. All authors read and approved the final manuscript. Funding This work was supported by the National Key R&D Program of China (2024YFD1701101), Joint Funds of the National Natural Science Foundation of China (U23A20158), National Natural Science Foundation of China (32170102), Foundation of Nankai University-Cangzhou Bohai New Area Green Chemical Research Institute (NCC2022-PY-09), and the Fundamental Research Funds for the Central Universities (63253191). Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement No datasets were generated or analysed during the current study.