Rice receptor kinase FLR7 regulates rhizosphere oxygen levels and enriches the dominant Anaeromyxobacter that improves submergence tolerance in rice

Abstract

Oxygen is one of the determinants of root microbiome formation. However, whether plants regulate rhizosphere oxygen levels to affect microbiota composition and the underlying molecular mechanisms remain elusive. The receptor-like kinase (RLK) family member FERONIA modulates the growth-defense tradeoff in Arabidopsis. Here, we established that rice FERONIA-like RLK 7 (FLR7) controls rhizosphere oxygen levels by methylene blue staining, oxygen flux, and potential measurements. The formation of oxygen-transporting aerenchyma in roots is negatively regulated by FLR7. We further characterized the root microbiota of 11 FLR mutants including flr7 and wild-type Nipponbare (Nip) grown in the field by 16S ribosomal RNA gene profiling and demonstrated that the 11 FLRs are involved in regulating rice root microbiome formation. The most abundant anaerobic-dependent genus Anaeromyxobacter in the Nip root microbiota was less abundant in the root microbiota of all these mutants, and this contributed the most to the community differences between most mutants and Nip. Metagenomic sequencing revealed that flr7 increases aerobic respiration and decreases anaerobic respiration in the root microbiome. Finally, we showed that a representative Anaeromyxobacter strain improved submergence tolerance in rice via FLR7. Collectively, our findings indicate that FLR7 mediates changes in rhizosphere oxygen levels and enriches the beneficial dominant genus Anaeromyxobacter and may provide insights for developing plant flood prevention strategies via the use of environment-specific functional soil microorganisms.

Keywords: Anaeromyxobacter; oxygen; rice FLRs; submergence tolerance; the root microbiome.

Figure 1

FLR7 negatively regulates rhizosphere oxygen levels; (A) quantified integral oxygen signal intensity within 1 cm from the root tip; (B) oxygen flux in the rhizosphere measured by NMT; positive values represent efflux from roots, and negative values represent influx into roots; a model diagram of the assay is presented; (C) potential in the rhizosphere measured by NMT; positive values represent oxidation potential, and negative values represent reduction potential; (D) relative expression levels of hypoxia-inducible marker genes in rice roots; (E) 3D-XRM of the aerenchyma near the root 2.2 cm away from the root tip; the distance between the root tip and the center of the examined ~0.7 mm thick root was 2.2 cm; a model diagram and representative 3D images are presented; (F) proportion of aerenchyma volume in E; in A–D and F, data are mean ± s.e.m. (n = 3 for each sample in A, D and F; n = 10 for each sample in B and C), and the significance of differences between the mutants and Nip was analyzed by ANOVA with Tukey’s HSD test; NS, not significant; P values are indicated; three independent experiments were performed with similar results.

Figure 2

FLRs affect the composition of rice root microbiota, especially the dominant anaerobic-dependent genus Anaeromyxobacter;( A) diagram of the experimental design for rice field trials; (B) unconstrained PCoA with Bray–Curtis dissimilarity showing dissimilarity of bacterial diversity between 11 FLR mutants and Nip; (C) unconstrained PCoA with Bray–Curtis dissimilarity showing the separation of the root microbiota of flr7 from that of Nip (P < .001, PERMANOVA by Adonis); ellipses cover 68% of the data for each rice genotype; (D) phylum-level distribution of the root microbiota of Nip, 11 FLR mutants, and bulk soil; the significance of differences in the relative abundance of each phylum between the mutants and Nip was analyzed by ANOVA with Tukey’s HSD test; ^***P < .001; ^**P < .01; ^*P < .05; exact P values are provided in Table S6; (E) relative abundance of the most dominant genus Anaeromyxobacter in Nip, 11 FLR mutants, and bulk soil; box plots show the median with upper and lower quartiles, and whiskers present the 1.5× interquartile range; statistical significance was determined by ANOVA with Tukey’s HSD test and P values are indicated; (F) LEfSe analysis of the root microbiota between Nip and flr7; the LDA score represents the contribution of differential lineages; lineages with LDA > 4 are displayed, and p, c, o, f, g, and s are the bacterial taxon abbreviations for phylum, class, order, family, genus, and species, respectively; (G) LEfSe analysis of the contribution of Anaeromyxobacter to differences in the root microbiota between Nip and each FLR mutant; the numbers of replicated samples in this figure are as follows: Nip (n = 11), each FLR mutant (n = 12), and bulk soil (n = 6).

Figure 3

FLR7 mutation enhances aerobic characteristics and diminishes anaerobic characteristics in the root microbiome; (A) differences in the relative abundance of species between the Nip and flr7 root metagenomes; the top 30 species in relative abundance in the Nip root metagenome are presented; relative abundance of aerobic and anaerobic respiration-related genes (B), aerobic respiratory pathways (C) and antioxidant-related functions (D); (E) bacterial copies of Anaeromyxobacter sp. PSR1 in the rice rhizosphere after 30 days of culture at different water depths (oxygen content); in A–D, the significance of differences in the relative abundance of each species or gene or pathway and function between flr7 and nip was analyzed by Welch’s t-test, and the numbers of replicated samples are as follows: Nip (n = 3) and flr7 (n = 3). In a, ^**P < .01; ^*P < .05; exact P values are provided in Table S6; in B–D, P values are indicated; in E, the significance of differences in the number of bacterial copies between the mutants and Nip was analyzed by ANOVA with Tukey’s HSD test, and the numbers of replicated samples are as follows: Nip (n = 9), flr7–1 (n = 9), and flr7–2 (n = 9); P values are indicated; box plots show the median with upper and lower quartiles, and whiskers present the 1.5× interquartile range.

Figure 4

Anaeromyxobacter improves submergence resistance in rice dependent on FLR7 via a low-oxygen quiescence-based strategy; (A) representative images of rice on the indicated days in normal culture, submergence treatment or recovery treatment after root inoculation with the Anaeromyxobacter strain PSR-1; (B) shoot height and fresh weight of rice for each treatment in A; (C) expression levels of the positive growth regulator OsGPAT1 and the negative growth regulator OsSAUR27 in the transcriptomes of each treatment; P < .05 and fold change >1.5 or < 0.5 (DESeq2) were set as the thresholds for significant differential expression; ^***P < .001; (D) effects of 50 μM GPAT inhibitor FSG67 on rice shoot height under normal culture and submergence stress; DMSO (0.5%; v/v; mock) was used as a negative control; (E) rate of inhibition of shoot height by submergence under treatment with the GPAT inhibitor FSG67; (F) top 20 pathways from KEGG enrichment analysis based on the P values for genes that showed upregulation with recovering-PSR-1 treatment compared with recovering-mock treatment; in B–E, the data are presented as the mean ± s.e.m. of three biological replicates; in B, data were analysed by ANOVA with Tukey’s HSD test and partial P values are indicated; P values for additional comparisons are provided in Table S6; in D and E, the significance of differences between GPAT inhibitor treatment and mock treatment was analyzed by Student’s t-test and P values are indicated; three independent experiments were performed with similar results.