The RNA degradation enzyme RNase E is essential for early flagellar assembly in Escherichia coli

Abstract

Escherichia coli endoribonuclease E (RNase E), encoded by the essential rne gene and conserved across γ-Proteobacteria, plays a central role in RNA processing and decay. We show here that rne-null strain, rne-null strain complemented with catalytic-null RNase E mutant, and C-terminal-truncated strain (Rned500) all lack flagellar biogenesis and motility under both aerobic and anaerobic conditions, which are restored by wild-type RNase E complementation. The Rned500 displays dysregulated expression of the three-tiered flagellar transcriptional cascade, increased stability of flagellar mRNAs, and reduced flagellar protein levels through sRNA-dependent translational inhibition. However, ectopic expression of flagellar master regulators or flagellar proteins fails to restore flagellar biogenesis and motility. To investigate the underlying defect, we examined the cellular localization of the early flagellar structural protein FliF and found it mislocalized in Rned500, indicating a disruption of early flagellar assembly. This defect is further supported by the impaired secretion of the flagellar anti-sigma factor FlgM in Rned500, a process that requires a functional flagellar basal body. Complementation with wild-type RNase E in Rned500 fully restores expression of the flagellar cascade, proper membrane localization of FliF, flagella formation, and motility. Wild-type RNase E-expressing strains, but not Rned500, activate Toll-like receptor 5 (TLR5)-dependent nuclear factor-kappa B signaling in THP-1 human monocytic cells through flagellin. This response, confirmed by a TLR5 dual-luciferase reporter assay in transfected HEK293T human embryonic kidney cells, highlights RNase E's role in enabling flagellar expression required for cellular immune activation. Collectively, these results identify RNase E as a key flagellar biogenesis regulator, revealing novel posttranscriptional control mechanisms.

Keywords: RNase E; TLR5-NF-κB signaling; flagella; posttranscriptional regulation; sRNAs.

Fig. 1.

RNase E and its catalytic activity, but not RNase G, are essential for FliC expression, flagella formation, and bacterial motility. A) Western blot of RNase E, FliC, and TolA (loading control) in KSL2010*/pBAD, KSL2010*/pRNE, KSL2010*/pDM, and MG1655_mvΔfliC. The MG1655_mvΔfliC served as a FliC negative control and an RNase E–positive control. Antibodies against RNase E, FliC, and TolA were used. Asterisk, the nonspecific binding signal of FliC antibody. Protein markers are indicated in kD. Relative RNase E abundances are shown. Values are mean ± SD (n = 3 replicates). ND, not detected. B) Northern blot of RNA I abundance over time following rifampicin (rif.) treatment in KSL2010*/pBAD, KSL2010*/pRNE, and KSL2010*/pDM (n = 5 biological replicates). Methylene blue staining of 5S rRNA served as loading control. Half-lives (t_1/2) are shown in minutes (m). Values are mean ± SD. C, D) Representative TEM images (C) and quantification of flagellated/nonflagellated cell percentage (D) from TEM images of KSL2010*/pBAD, KSL2010*/pRNE, KSL2010*/pDM, and MG1655_mvΔfliC cells. Red arrows, flagellar filaments. Scale bar, 1 μm (n = 200 cells/strain). E) Motility zone diameters of KSL2010*/pBAD, KSL2010*/pRNE, KSL2010*/pDM, and MG1655_mvΔfliC grown on LB soft-agar plates supplemented with or without 0.1% arabinose (ara.) for 60 hours (n = 20 biological replicates). Values are mean ± SD. F) Motility zone diameters of MG1655_mv, MG1655_mvΔrng, and MG1655_mvΔfliC grown on LB soft-agar plates for 24 h (n = 10 biological replicates). Values are mean ± SD.

Fig. 2.

Disruption of RNase E C-terminal integrity or silencing its catalytic activity impairs flagella formation and motility in E. coli under aerobic and anaerobic conditions, both restored by wild-type RNase E complementation. A) Schematic representation of rne variants encoded by the chromosomes of E. coli strains and plasmids. Full-length RNase E with its microdomains is depicted above the schematics. Green rectangle, RhlB binding domain; blue rectangle, enolase binding domain; purple rectangle, PNPase binding domain; blue triangle, FLAG tag. The encoded amino acids and deleted binding sites are indicated by numbers. B, E, I) Western blot of FliC and GAPDH (loading control) in rne mutant strains (B, aerobic, LB) and RNase E complemented strains (E, aerobic, LB; I, anaerobic, LB with 0.2% pyruvate). MG1655_mvΔfliC served as negative control. The pCM127, an empty vector plasmid, was used as a negative control for the pFL, p500, and pCter500 plasmids. Antibody against FliC, and GAPDH were used. Asterisk, the nonspecific signal of FliC antibody. Protein markers are indicated in kD. Relative FliC abundances are shown. Values are mean ± SD (n = 3 replicates). NA, not available; ND, not detected. Aerobic (+) and anaerobic (−) conditions are indicated. C, F, J, L) Quantification of flagellated/nonflagellated cells from TEM images of rne mutant strains (C, aerobic, LB), RNase E complemented strains (F, aerobic, LB; J, anaerobic, LB with 0.2% pyruvate), and KSL2010* variant strains (L, anaerobic, LB + 0.2% pyruvate and 0.1% arabinose). n = 150 or 200 cells/strain. D, G, H, K, M) Motility zone diameters of rne mutant and MG1655_mvΔfliC strains (D, aerobic, LB), RNase E–complemented strains (G, aerobic, LB; K, anaerobic, LB with 0.2% pyruvate), and MG1655_mv with different carbon sources (H, aerobic, LB, LB with 0.2% pyruvate, LB with 0.2% glucose) after 24 h of incubation; and KSL2010* variant strains (M, anaerobic, LB + 0.2% pyruvate and with or without 0.1% arabinose (ara.)) for 60 hours of incubation (n = 5, 10 or 20 biological replicates as indicated; one-way ANOVA (P < 0.001), followed by Dunnett's post hoc test against the control group). Values are mean ± SD.

Fig. 3.

RNase E–dependent flagellar expression correlates with TLR5-mediated NF-κB activation in E. coli–challenged THP-1 and HEK293T cells. Aa) High content fluorescence microscopy of NF-κB (green) and Hoechst-stained nuclei (blue) in THP-1 cells challenged by heat-killed MG1655_mv, Rned500, Rned500/p500, Rned500/pFL, and MG1655_mvΔfliC (negative control) E. coli strains. Scale bars, 20 μm. Ab) Quantification of the NF-κB nuclear:cytoplasmic ratio from NF-κB immunostaining microscopy images (n = 35 individual THP-1 cells; one-way ANOVA, P < 0.001, followed by Tukey's honestly significant difference (HSD) post hoc test). Ba, Bb) NF-κB activity of HEK293T/Control (white bar) and HEK293T/TLR5 (gray bar) cells stimulated with purified flagellin (0 to 800 ng) (Ba) or E. coli strains (MOI = 100) (Bb) for 6 h (n = 6–7 biological repeats; each repeat with three technical replicates; two-way ANOVA followed by Tukey's HSD post hoc test). Values are mean ± SD.

Fig. 4.

Dysregulation of flagellar gene expression and increased flagellar mRNA half-lives in the Rned500 mutant. A, B) Differential mRNA levels in Rned500 mutant compared with parental MG1655_mv, as determined by RNA-Seq analysis (A), and qRT-PCR validation (B). Genes are arranged according to the function of their encoded proteins, as described below the figure. Values are mean ± SD of log₂ relative expression fold-change (n = 3 or 4 biological replicates as indicated; multiple unpaired t-test). ***P < 0.001; **P < 0.01; *P < 0.05; ns, P ≥ 0.05. White bar, class I genes; blue bar, class II genes; pink bar, class II&III genes; yellow bar, class III genes; red inverted triangles, genes selected for RNA half-life analysis and GFP translational fusion assays. Ca–Dd) Remaining RNA abundances of flhD (Ca), flhC (Cb), flhB (Da), fliF (Db), fliM (Dc), and fliI (Dd) transcripts in MG1655_mv and Rned500 plotted against time after rifampicin treatment. Dashed line, 50% remaining RNA. Half-lives (t_1/2) are shown in minutes (m). Values are mean ± SD (n = 4 biological replicates; nonlinear regression fit; unpaired t-test).

Fig. 5.

Deletion of negative regulatory sRNAs restores flagellar protein synthesis but fails to restore flagella formation and motility in Rned500. Aa-Ae) Relative protein abundances from GFP translational reporter assay on FlhD (Aa), FlhB (Ab), FliF (Ac), FliM (Ad), and FliI (Ae) in MG1655_mv, Rned500, and Rned500Δ4sRNAs. Values are mean ± SD (n = 4 biological repeats; unpaired t-test). Schematics above each graph show flagellar GFP fusion constructs, including the native promoter, RBS, and the coding region of first 26 or 30 amino acids of each protein. Numbers indicate the nucleotide regions included in the construct relative to the transcription start site (+1). sRNA regulatory sites on the flhD transcript are indicated. B) Percentage of remaining FLAG-FlhD and FlhC-HA protein levels in MG1655_mv and Rned500 were plotted against time after chloramphenicol treatment. Dashed line, 50% protein remaining. Half-lives (t_1/2) are shown in hours (h). Values are mean ± SD (n = 4 replicates; nonlinear regression fit). C) Northern blot of OxyS, ArcZ, OmrA, and OmrB sRNAs in MG1655_mv and Rned500 during exponential growth. Methylene blue–stained 5S rRNA used as a loading control. Relative expression levels are shown. Values are mean ± SD (n = 3 biological replicates). n.d., nondetectable. D) Quantification of flagellated/nonflagellated cells from TEM images of sRNA-deletion strains (n = 150 cells/strain). E) Motility zone diameters of sRNA-deletion strains on LB soft-agar over 24 h incubation. Values are mean ± SD (n = 10 biological replicates). ND, not detected.

Fig. 6.

Defects in hook-basal body assembly and FlgM secretion in Rned500. Aa) Representative image of the motility zone of MG1655_mvΔfliF and MG1655_mvΔfliF/fliF-Bs1 on an LB soft-agar plate supplemented with 0.2% arabinose for FliF-Bs1 induction after 36 h of incubation. Scale bar, 1 cm. Ab) Representative fluorescence microscopy images and respective brightfield images of FliF-Bs1 in MG1655_mv, Rned500, Rned500/pFL, and MG1655_mvΔflhD (negative control) under 0.2% arabinose induction (n = 3 biological replicates). Scale bars, 5 and 2 μm, as indicated. Areas within white boxes are enlarged at right. Ac) Dot plot of bacterium length (μm) for strains used in (Ab) (n = 100 cells/strain; one-way ANOVA). Red line, median. B) Western blot of intracellular and extracellular (secreted) FLAG-FlgM of MG1655_mv, Rned500, and Rned500/pFL. MG1655_mvΔfliF used as negative control for FLAG-FlgM secretion (n = 3 replicates). GAPDH used as loading control. Antibody against FLAG tag and GAPDH were used. A schematic diagram depicting the FLAG-FlgM constructs is shown.