Arsenic-induced enhancement of diazotrophic recruitment and nitrogen fixation in Pteris vittata rhizosphere

Abstract

Heavy metal contamination poses an escalating global challenge to soil ecosystems, with hyperaccumulators playing a crucial role in environmental remediation and resource recovery. The enrichment of diazotrophs and resulting nitrogen accumulation promoted hyperaccumulator growth and facilitated phytoremediation. Nonetheless, the regulatory mechanism of hyperaccumulator biological nitrogen fixation has remained elusive. Here, we report the mechanism by which arsenic regulates biological nitrogen fixation in the arsenic-hyperaccumulator Pteris vittata. Field investigations and greenhouse experiments, based on multi-omics approaches, reveal that elevated arsenic stress induces an enrichment of key diazotrophs, enhances plant nitrogen acquisition, and thus improves plant growth. Metabolomic analysis and microfluidic experiments further demonstrate that the upregulation of specific root metabolites plays a crucial role in recruiting key diazotrophic bacteria. These findings highlight the pivotal role of nitrogen-acquisition mechanisms in the arsenic hyperaccumulation of Pteris vittata, and provide valuable insights into the plant stress resistance.

Fig. 1. Arsenic stress promotes biological nitrogen fixation in Pteris vittata, alleviating nitrogen limitation for plant growth.

a Differences in N indices including total N, dissolved total N (DTN), NH₄⁺ and NO₃⁻, across the bulk (Bulk) and rhizosphere (Rhiz.) soils under low- (n = 10 for each habitat) and high-As stresses (n = 15 for each habitat). b Correlation between beta-diversity patterns of rhizospheric bacterial communities and As(V) concentration. c Variations of nifH, nifD, and nifK homologs in the rhizosphere between the low- (n = 10) and high-As stresses (n = 15). d Photograph portraying Pteris vittata cultivated for 50 days under the control and As(V) addition, along with the aboveground dry weight of plants under control, As(V) and N amendment. e Comparison of the total N content of root and shoot between the control and As treatments. f Variations in the abundance of nifH across the bulk and rhizosphere soils under the control and As treatments based on qPCR quantification. In box plots, the center line represents the median, box edges delimit lower and upper quartiles and whiskers show the highest and lowest values. For (d–f), data are shown as mean ± SEM (n = 5). Each data point represents a biologically independent replicate. P values were determined through two-sided Wilcoxon test.

Fig. 2. Identification of diazotrophs regulated by arsenic stress.

a The nifH palmprint spans three well-conserved sequence motifs (A, B and C) and intervening variable regions (V1 and V2), exemplified within the Azotobacter vinelandii nifH structure (Protein Data Bank code: 6Q93). Conservation was determined by analyzing nifH alignments from RefSeq, GeneBank, and NR databases, and depicted using motif sequence logos. aa, amino acids. b Maximum-likelihood phylogenetic tree of nifH representatives identified from metagenomic assemblies and a custom nifH database. The nifH sequences, categorized into canonical groups I to VIII, are depicted by branches of varying colors. The innermost circle represents log₂ fold changes (LFC) for up-regulated (red) and down-regulated (blue) nifH genes under high As stress versus low As stress. Green labels highlight branches with significant changes (two-sided Wilcoxon tests, adjusted using the Benjamini-Hochberg method). The remaining circles depict the taxonomic annotation of nifH at domain, phylum, class, and order levels.

Fig. 3. Isolation, identification, and functional validation of diazotrophic bacteria.

a Sankey plot showing the taxonomy of aioA carriers at the domain, phylum, class and order levels. b Changes of optical density (OD) in N-free medium amend with As(III) over time after inoculation with seven isolates from seven different genera of Rhizobiales or Burkholderiales (n = 3). c Effects of isolates inoculation on plant biomass. d Effects of isolates inoculation on plant rhizosphere nifH abundance. e Effects of isolates inoculation on N concentration of root and shoot. f Effects of treatments on the relative abundances of Bradyrhizobium, Cupriavidus, Mesorhizobium and Rhizobium in the rhizosphere using amplicon. For (b–f) data are shown as mean ± SEM (n = 4). Each data point represents a biologically independent replicate. P values were determined through two-sided Wilcoxon tests for each treatment compared to the control group.

Fig. 4. Identification of key root exudates regulated by arsenic and their chemotactic effects on diazotrophic isolates.

a Bar plots representing the log fold changes in the abundances of carbohydrates, alpha amino acids and flavonoids in the rhizosphere between the low- and high-As stresses. The purple-colored bars represent significant differences while the orange and green triangles, respectively, mark the significantly enriched and depleted metabolites in the high-As rhizosphere compared to the corresponding bulk soil. Blue solid circles indicate up-regulated metabolites in the endosphere under the high-As stress compared with the low-As counterparts. b Normalized signal intensity of catechin in the soil-root habitats of wild Pteris vittata, including bulk soil, rhizosphere (Rhiz.), and endosphere (Endo.), under low- (n = 10 for each habitat) and high-As stresses (n = 15 for each habitat), assessed using untargeted metabolomics, and the concentration of catechin in the rhizosphere soil of Pteris vittata with and without As exposure in greenhouse experiments, determined using targeted metabolomics (n = 6). N/A indicate that the metabolite concentration in the soil was below 5 × 10⁻³ nmol g⁻¹. P values were determined through two-sided Wilcoxon tests. c Chemotaxis index of Rhizobium sp. G5 and Bradyrhizobium sp. J3 exposed to four key metabolites based on In-Situ Chemotaxis Assay (ISCA) microfluidic platform. Chemotaxis index denotes the concentration of cells in ISCA wells containing each metabolite, normalized by the concentration of cells in corresponding wells containing phosphate-buffered saline (PBS). Each metabolite treatment was biologically replicated across four independent ISCA chips (n = 4). P values between the metabolite treatments and PBS control are determined through one-sided Wilcoxon tests. For (b, c) data are shown as mean ± SEM.

Fig. 5. Contribution of biological nitrogen fixation under arsenic stress and synthetic communities (SynComs) to nitrogen nutrition of roots and shoot.

a Overview of ¹⁵N isotope tracer experiment. b δ¹⁵N values and contribution of biological N fixation (BNF) to the total N accumulation in the roots and shoots of Pteris vittata (n = 6). Each data point represents a biologically independent replicate. P values were determined by the two-sided Wilcoxon test. In box plots, the center line represents the median, box edges delimit lower and upper quartiles and whiskers show the highest and lowest values.

Fig. 6. Proposed model for arsenic-regulated biological nitrogen fixation in Pteris vittata.

In high-As soils, high content of chemoattractants in root exudates result in an enrichment of specific diazotrophic bacteria (e.g. Bradyrhizobium sp. J3 and Rhizobium sp. G5), which boosts plant growth and N uptake by enhancing biological N fixation.