Integrated Hfq-interacting RNAome and transcriptomic analysis reveals complex regulatory networks of nitrogen fixation in root-associated Pseudomonas stutzeri A1501

Abstract

The RNA chaperone Hfq acts as a global regulator of numerous biological processes, such as carbon/nitrogen metabolism and environmental adaptation in plant-associated diazotrophs; however, its target RNAs and the mechanisms underlying nitrogen fixation remain largely unknown. Here, we used enhanced UV cross-linking immunoprecipitation coupled with high-throughput sequencing to identify hundreds of Hfq-binding RNAs probably involved in nitrogen fixation, carbon substrate utilization, biofilm formation, and other functions. Collectively, these processes endow strain A1501 with the requisite capabilities to thrive in the highly competitive rhizosphere. Our findings revealed a previously uncharted landscape of Hfq target genes. Notable among these is nifM, encoding an isomerase necessary for nitrogenase reductase solubility; amtB, encoding an ammonium transporter; oprB, encoding a carbohydrate porin; and cheZ, encoding a chemotaxis protein. Furthermore, we identified more than 100 genes of unknown function, which expands the potential direct regulatory targets of Hfq in diazotrophs. Our data showed that Hfq directly interacts with the mRNA of regulatory proteins (RsmA, AlgU, and NifA), regulatory ncRNA RsmY, and other potential targets, thus revealing the mechanistic links in nitrogen fixation and other metabolic pathways.

Importance: Numerous experimental approaches often face challenges in distinguishing between direct and indirect effects of Hfq-mediated regulation. New technologies based on high-throughput sequencing are increasingly providing insight into the global regulation of Hfq in gene expression. Here, enhanced UV cross-linking immunoprecipitation coupled with high-throughput sequencing was employed to identify the Hfq-binding sites and potential targets in the root-associated Pseudomonas stutzeri A1501 and identify hundreds of novel Hfq-binding RNAs that are predicted to be involved in metabolism, environmental adaptation, and nitrogen fixation. In particular, we have shown Hfq interactions with various regulatory proteins' mRNA and their potential targets at the posttranscriptional level. This study not only enhances our understanding of Hfq regulation but, importantly, also provides a framework for addressing integrated regulatory network underlying root-associated nitrogen fixation.

Keywords: Hfq; Pseudomonas stutzeri; eCLIP-seq; nitrogen fixation; posttranscriptional regulation.

Fig 1

Global patterns of Hfq-binding targets. (A) IGV (Integrative Genomic Viewer) snapshot of genomic regions with eCLIP-seq data. Reads from eCLIP-seq with A1501 Hfq-3×Flag IP cells compared with eCLIP-seq Input cells and 3×Flag cells. (B) Pie chart depicting the region distribution of Hfq eCLIP-seq peaks relative to the predicted coding sequence (CDS), ncRNA, 5′UTR, and 3′UTR. (C) Venn diagram depicting genes with significant expression changes according to the different transcriptomic approaches. eCLIP-seq peaks without associated annotations are not shown. (D) GO Gene ontology (GO) enrichment analysis of differentially expressed genes between the WT and Δhfq strains according to both RNA-seq and Hfq eCLIP-seq peak data. “Gene ratio” shows the ratio of genes related to the GO term to the total number of differentially expressed genes annotated with the given GO term of biological processes and molecular functions identified using DAVID to be enriched. The adjusted P value (Padj) scale is determined through a correction process applied to the raw P value, utilizing the Benjamini and Hochberg methods to control the false discovery rate. A threshold for significantly differential expression was established with Padj < 0.05 and |log2(foldchange)| > 0.

Fig 2

Effects of Hfq on the expression of genes involved in nitrogen fixation and central carbon metabolism. (A) Effect of Hfq on nitrogen fixation island genes. (B, C) Relative expression levels of glnA, nifH, and nifA (B) and β-galactosidase activity (C) in the WT versus the hfq deletion mutant under nitrogen fixation conditions for 4 h. (D) Effects of Hfq on the expression of genes involved in central carbon metabolism in P. stutzeri A1501. (E, F, G, and H) The bars depict the β-galactosidase levels conferred by the chromosomally integrated translational zwf::lacZ (E) gcd::lacZ (F), oprB::lacZ (G), and gtsA::lacZ (H) fusion constructs expressed in WT and Δhfq, respectively, in the K medium plus glucose. The CA motif indicates the sequence containing the AANAANAA motif in the 5′UTR of the above genes. The asterisks indicate that a Hfq eCLIP-seq peak is associated with the gene. Red indicates repression by Hfq, blue indicates activation, and black indicates no change (NC) in gene expression. The number indicates the fold change (log₂FC). Asterisks indicate statistical significance determined by one-way ANOVA with Tukey’s post hoc test: ****P ≤ 0.0001, ***P ≤ 0.001, **P ≤ 0.01, and *P ≤ 0.05; ns: not significant.

Fig 3

Effect of Hfq on the expression of genes involved in chemotaxis and biofilm formation. (A) The chemotaxis ability of WT and Δhfq. (B, C) Relative expression levels of cheZ, cheR, and fliG (B) and β-galactosidase activity (C) in the WT and Δhfq. (D) Summary of the effects of Hfq on biofilm formation and Psl production-related genes that were differentially expressed in the hfq mutant compared with the WT. (E) Effect of Hfq on biofilm formation after 48 h inoculation and comparison of the biofilm biomass obtained with the WT, hfq deletion mutant, and complemented strains. (F) Relative expression levels of algU, rsmA, rsmY, and sadC in the WT versus the hfq deletion mutant under nitrogen fixation condition. (G) β-Galactosidase activity of rsmA in the WT and Δhfq. (H, I) Analysis of the rsmA (H) and algU (I) mRNA half-life in the WT and hfq mutant strains under nitrogen fixation condition. Rifampicin (400 µg/mg) was added at time 0 min. The error bars show the standard deviations of the means of three independent experiments. Asterisks indicate that a Hfq eCLIP-seq peak is associated with the gene. Red indicates repression by Hfq, blue indicates activation, and black indicates NC in gene expression. The number indicates the log₂FC. Asterisks indicate statistical significance determined by one-way (C) and two-way ANOVA with Tukey’s post hoc test: ****P ≤ 0.0001, ***P ≤ 0.001, **P ≤ 0.01, and * P ≤ 0.05; ns, not significant.

Fig 4

Integration of Hfq into P. stutzeri regulatory networks. (A) Top three enriched sequence motifs corresponding to Hfq eCLIP-seq peaks identified with Homer. (B) qRT-PCR analysis of rpos, algU, rpoN, fleQ, fur, ntrC, sigX, gltR, hexR, and rpoH RNA levels normalized to the 16S rRNA levels. (C) β-Galactosidase activity of the above genes in the WT and Δhfq strains. (D, E) qRT-PCR analysis of the fur (D) and rpoS (E) mRNA half-life in the WT and hfq mutant strains. Rifampicin (400 µg/mL) was added at time 0. The error bars show the standard deviations of the means of three independent experiments. (F) β-Galactosidase levels conferred by the chromosomally integrated transcription PcrcZ::lacZ fusion in the WT and Δhfq strains. (G, H, and I) Half-life of CrcZ (G), RsmY (H), and PrrF (I) in the WT and Δhfq strains. All the strains were cultured under nitrogen fixation conditions for 4 h. The asterisks indicate that a Hfq eCLIP-seq peak is associated with the gene. Asterisks indicate statistical significance determined by a t-test (F) and two-way ANOVA (B, C) with Tukey’s post hoc test: ****P ≤ 0.0001, ***P ≤ 0.001, **P ≤ 0.01, and *P ≤ 0.05; ns, not significant.

Fig 5

The proposed regulatory networks controlling nitrogen fixation and other related pathways in P. stutzeri A1501. The data are derived from both this study and previous research. In this model, Hfq acts as a central player, working with other global regulators, ncRNAs, and their target genes. Targets directly influenced by Hfq are depicted in red, while indirectly affected targets are represented in black. Arrows and T-shaped bars indicate positive and negative regulation, respectively. For details, please refer to Table 1 and Table S1.