Post-Transcriptional Regulation of the MiaA Prenyl Transferase by CsrA and the Small RNA CsrB in Escherichia coli

Abstract

MiaA is responsible for the addition of the isopentyl modification to adenine 37 in the anticodon stem loop of specific tRNAs in Escherichia coli. Mutants in miaA have pleotropic effects on the cell in E. coli and play a role in virulence gene regulation. In addition, MiaA is necessary for stress response gene expression by promoting efficient decoding of UUX-leucine codons, and genes with elevated UUX-leucine codons may be a regulatory target for i⁶A-modified tRNAs. Understanding the temporal nature of the i⁶A modification status of tRNAs would help us determine the regulatory potential of MiaA and its potential interplay with leucine codon frequency. In this work, we set out to uncover additional information about the synthesis of the MiaA. MiaA synthesis is primarily driven at the transcriptional level from multiple promoters in a complex operon. However, very little is known about the post-transcriptional regulation of MiaA, including the role of sRNAs in its synthesis. To determine the role of small RNAs (sRNAs) in the regulation of miaA, we constructed a chromosomal miaA-lacZ translational fusion driven by the arabinose-responsive P_BAD promoter and used it to screen against an Escherichia coli sRNA library (containing sRNAs driven by the IPTG-inducible P_Lac promoter). Our genetic screen and quantitative β-galactosidase assays identified CsrB and its cognate protein CsrA as potential regulators of miaA expression in E. coli. Consistent with our hypothesis that CsrA regulates miaA post-transcriptional gene expression through binding to the miaA mRNA 5' UTR, and CsrB binds and regulates miaA post-transcriptional gene expression through sequestration of CsrA levels, a deletion of csrA significantly reduced expression of the reporter fusion as well as reducing miaA mRNA levels. These results suggest that under conditions where CsrA is inhibited, miaA mRNA translation and thus MiaA-dependent tRNA modification may be limited.

Keywords: RNA binding protein; RNA processing; small RNA; tRNA modification.

Figure 2

The effect of CsrB on miaA expression. (A) Quantitative β-galactosidase assay analysis of P_BAD-miaA27_(P2HS)-lacZ translational fusion activity showing repression by CsrB sRNA. β-galactosidase assays were repeated at least three times and data points represent the mean plus and minus the standard error of the mean (mean ± sem). β-galactosidase activity is quantified by the use of arbitrary machine units as described in Section 4. (B) Northern blot analysis of MiaA steady-state levels following over-expression of CsrB (KMT792) or McaS (KMT799) in comparison to the empty vector. Northern blots were repeated at least three times. (C) Quantitative densitometry of Northern blot analysis. Densitometry signals were acquired using a Fluorochem R Fluorescent/Chemiluminescent imager. Each data point represents an average of at least three experiments and error bars represent the standard error of the mean (mean ± sem) based on the densitometry of Northern blot signals in Section (B). Statistical analysis was executed using One-Way ANOVA with Tukey’s Multiple Comparisons test on GraphPad Prism 9 (* p-value = 0.05, ns = not significant).

Figure 3

Effects of RNase E and PNPase on miaA mRNA stability. (A) Northern blot analysis of miaA mRNA stability. A temperature-sensitive mutant RNase E (rne-3071 zce-726::Tn10) allele was transduced from KMT621 into MG1655 (KMT665) by bacteriophage P1 transduction and selected for tetracycline (Tet^R) resistance KMT801. Wild-type and rne^ts strains were grown in rich media (LB) at 30 °C to an OD₆₀₀ of 0.3. Sample aliquots of 600 μL were collected for total RNA isolation and Northern blot analysis at zero minutes, before transferring cultures to 43 °C. Rifampicin was added, and the samples were collected at 2, 4, 8, 16, and 32 min for total RNA isolation and analysis by agarose Northern blot. Experiments were repeated at least three times and the blots shown are representative blots of the triplicate experiments. (B) Quantitative densitometry of Northern blot in Section (A). Statistical analysis includes mean and standard error of the mean (mean ± s.e.m.). (C) Half-life calculations of Northern blot executed in Section (A). Quantitative Northern blot data from Section (B) were subjected to linear regression analysis using GraphPad Prism 9. (D) Δpnp::kan mutation was transduced from KMT624 into MG1655 (KMT665) by bacteriophage P1 transduction and selected for kanamycin (kan^R) resistance (KMT800). Wild-type and pnpA⁻ strains (KMT665 and KMT800) were grown in rich media (LB) to an OD₆₀₀ of 0.3. Then, 600 μL aliquots of the sample were collected for total RNA isolation and Northern blot analysis at zero minutes. Rifampicin was added, and samples were collected at 2, 4, 8, and 16 min for total RNA isolation and analysis by agarose Northern blot. (E) Quantitative densitometry of Northern blot in Section (A). Statistical analysis includes mean and standard error of the mean (mean ± s.e.m.) (F) Half-life calculations of Northern blot executed in Section (D). Quantitative Northern blot data from Section (B) were subjected to linear regression analysis using GraphPad Prism 9.

Figure 1

Small RNA library screen for regulators of miaA expression. (A) A schematic representing the arabinose-inducible miaA-lacZ translational gene fusion (P_BAD-miaA27_(P2HS)-lacZ) containing the 5′ UTR from the miaA P2 heat shock promoter. (B) A P_BAD-miaA27_(P2HS)-lacZ translational fusion strain was transformed with a library of 30 known sRNAs cloned downstream from an IPTG-inducible promoter in plasmid pBR-pLac and screened for activity on MacConkey-Lactose plates supplemented with ampicillin. Results shown are for sRNA clones that gave a Lac- phenotype, suggesting a role for these sRNAs in the negative regulation of MiaA expression. (C) Quantitative β-galactosidase assay analysis of P_BAD-miaA27_(P2HS)-lacZ translational fusion strain (JIA4000) carrying pBR-pLac (empty vector—JIA4001), pBR-sdsR (JIA4010), pBR-arcZ (JIA4018), pBR-gcvB (JIA4015), pBR-spf (JIA4024), or pBR-csrB (JIA4029). Each time point represents an average of at least three experiments and error bars represent the standard error of the mean (mean ± sem). (D) Quantitative β-galactosidase assay analysis of P_BAD-miaA27_(P2HS)-lacZ translational fusion strain (JIA4000) with deletions–insertions in the genes for the candidate sRNA repressors of miaA picked up in our screen: ΔcsrB::zeo (JIA4042), ΔgcvB::kan (JIA4041), Δspf::cat (JIA4043), ΔarcZ::zeo (JIA4040), and ΔsdsR::kan (JIA4045). Each time point represents an average of at least three experiments and error bars represent the standard error of the mean (mean ± sem). β-galactosidase activity is quantified using arbitrary machine units as described in Section 4.

Figure 4

The expression of miaA in the absence of csrA. (A) Wild-type, ΔcsrA::zeo, and ΔcsrB::zeo versions of P_BAD-miaA27_(P2HS)-lacZ translational fusions (JIA4000, JIA4044, and JIA4042, respectively) were grown in LB supplemented with glucose to an OD₆₀₀ of 0.5, and then shifted to LB supplemented with arabinose; aliquots were obtained for β-galactosidase assay every 10 min for 70 min. (B) CsrA complementation assay using P_BAD-miaA27_(P2HS)-lacZ translational fusion activity. Wild-type (csrA⁺) and ΔcsrA::zeo (csrA⁻) strains of P_BAD-miaA27_(P2HS)-lacZ fusions carrying pBR-pLac or pBR-csrA were grown in rich media (LB) to an OD₆₀₀ of 0.5 and shifted to LB supplemented with arabinose; aliquots were obtained for β-galactosidase assay every 10 min for 70 min. (C) Northern blot analysis of miaA mRNA from total RNA isolated from exponentially growing wild-type, ΔcsrA::zeo, and ΔcsrB::zeo cells. Experiments were repeated at least three times and blots shown are representative blots of the triplicate experiments. (D) Quantitative densitometry of Northern blot analysis in (C). Densitometry signals were acquired using Fluorochem R Fluorescent/Chemiluminescent imager and statistical analysis of densitometry was executed using One-Way ANOVA with Tukey’s Multiple Comparisons test on GraphPad Prism 9 (*** p-value = 0.001, ns = not significant).

Figure 5

Model for CsrA-CsrB regulation of miaA expression. Graphical model of CsrA/CsrB regulation of MiaA was created using Biorender.com. MiaA transcript turnover is mediated through both PNPase and RNase E. CsrA promotes the accumulation of the miaA P2 (heat shock) (HS) mRNA in the absence of CsrB. In the presence of CsrB, CsrA sequestration leads to decreased levels of the miaA P2 (heat shock) mRNA.