Direct and indirect control of Rho-dependent transcription termination by the Escherichia coli lysC riboswitch

Abstract

Bacterial riboswitches are molecular structures that play a crucial role in controlling gene expression to maintain cellular balance. The Escherichia coli lysC riboswitch has been previously shown to regulate gene expression through translation initiation and mRNA decay. Recent research suggests that lysC gene expression is also influenced by Rho-dependent transcription termination. Through a series of in silico, in vitro, and in vivo experiments, we provide experimental evidence that the lysC riboswitch directly and indirectly modulates Rho transcription termination. Our study demonstrates that Rho-dependent transcription termination plays a significant role in the cotranscriptional regulation of lysC expression. Together with previous studies, our work suggests that lysC expression is governed by a lysine-sensing riboswitch that regulates translation initiation, transcription termination, and mRNA degradation. Notably, both Rho and RNase E target the same region of the RNA molecule, implying that RNase E may degrade Rho-terminated transcripts, providing a means to selectively eliminate these incomplete messenger RNAs. Overall, this study sheds light on the complex regulatory mechanisms used by bacterial riboswitches, emphasizing the role of transcription termination in the control of gene expression and mRNA stability.

Keywords: Rho-dependent transcription termination; lysine; reporter gene assays; riboswitch.

FIGURE 1.

The E. coli lysC riboswitch regulation mechanism. (A and B) The structures of the lysC riboswitch in the ON state (A) and OFF state (B). In the absence of lysine, the riboswitch adopts the ON state conformation in which the antisequestering stem allows efficient translation and prevents RNase E cleavage. Upon lysine binding, the riboswitch folds into the OFF structure containing the P1 stem, which down-regulates lysC expression both by inhibiting the initiation of translation and by promoting RNase E cleavage of the mRNA. The SD sequence and AUG start codon are shown. (C) Secondary structure of the lysC riboswitch in the OFF state. The RNase E cleavage sites Site1 and Site2 are shown. The region corresponding to the riboswitch expression platform is shown in the ON state where access to Site1 and Site2 is prevented through the formation of the anti-P1 stem. The SD and AUG start codon are indicated in blue boxes.

FIGURE 2.

The lysC riboswitch exhibits Rho-dependent transcription termination in vitro. (A) Sequence analysis of the lysC riboswitch showing a C-rich region (position ∼240) within the riboswitch expression platform. The location of the potential rut site is highlighted by a hollow orange rectangle. The red region indicates the location of RNase E cleavage sites. The positions are numbered according to the transcription start site (TSS). (B) In vitro transcription of the lysC riboswitch showing Rho transcription termination. Transcriptions were performed without (NT) or with Rho (50 nM), NusG (50 nM), or both in the absence or presence of 1 mM lysine. The full-length (FL) and terminated transcripts are indicated at the right of the gel. The % termination corresponds to the fraction of all terminated transcripts relative to the FL transcripts. Sequencing lanes are shown to map the termination sites.

FIGURE 3.

Expression of transcriptional lysC–gfp fusions monitored in wild-type (WT), rne-131, and rho-R66S strains performed in the absence and presence of 50 µg/mL lysine. (A–C) Expression of lysC–gfp fusions performed in WT (A), rne-131 (B), and rho-R66S (C) strains. The number of lysC codons is indicated. A strain without the riboswitch (ctrl) is shown as a control. Values were normalized to the expression obtained without lysine for each construct. The average values of three independent experiments with SD are shown. (*) P < 0.05, (**) P < 0.01, (***) P < 0.001, (****) P < 0.0001.

FIGURE 4.

RT-qPCR analysis of endogenous lysC expression and in the context of the plasmid performed in the WT, rne-131, and rho-R66S strains. (A and B) RT-qPCR assays were performed on the endogenous lysC (A) and expressed from a plasmid (B) used for GFP reporter gene assays (417 codons) without and with 50 µg/mL lysine. Values were normalized to the expression obtained without lysine for each construct. The average values of three independent experiments with SD are shown. (*) P < 0.05, (NS) nonsignificant.

FIGURE 5.

ChIP-qPCR analysis of lysC transcriptional activity in the WT and in the rho-R66S strain. The RNAP occupancy was monitored using ChIP. (Top) Schematic representation of ChIP-qPCR targeting the promoter region (pBAD) and different lysC regions in the 5′ UTR (C3) and in the coding region (C5 and C6). (Bottom) ChIP-qPCR data showing RNAP (β subunit) occupancy at indicated locations of lysC. Values represent the relative enrichment of ChIP DNA to the input control in WT cells or cells expressing the rho-R66S mutant. A construct was made in which positions 250–253 were removed (ΔSite1). (***) P < 0.001, (****) P < 0.0001, (NS) nonsignificant.

FIGURE 6.

The lysC riboswitch controls several mechanisms to efficiently modulate lysC expression. (A) In the absence of lysine, transcription of the riboswitch leads to the formation of the antisequestering stem (orange), which prevents Rho binding and RNase E cleavage. The produced transcripts allow efficient translation initiation. The red star depicts the region containing both the RNase E cleavage site and the rut site. (B) In the presence of lysine, the riboswitch adopts the OFF state structure that promotes Rho-dependent transcription termination, thus inhibiting lysC expression. Our data are consistent with rut sites located either in the 5′ UTR (upper panel, red star) or in the ORF domain (lower panel, red triangles). Provided that Rho binds within the 5′ UTR, transcription termination is expected to occur at the G309 pause site. Importantly, while the accessibility of the 5′-UTR rut site is modulated through the riboswitch structure, the access to the ORF rut sites is dependent on the translation efficiency. It is possible that prematurely terminated transcripts are targeted by the RNA degradosome either in the 5′-UTR region and also possibly within the ORF. (C) In the presence of lysine, transcripts that have not been prematurely terminated by Rho may be targeted by the RNA degradosome in the 5′ UTR (red star) and/or in the ORF. For clarity, the cleavage is shown to occur only when RNAP has reached the ORF.

Tithi Ghosh