The 3'-untranslated region of mRNAs as a site for ribozyme cleavage-dependent processing and control in bacteria

Abstract

Besides its primary informational role, the sequence of the mRNA (mRNA) including its 5'- and 3'- untranslated regions (UTRs), contains important features that are relevant for post-transcriptional and translational regulation of gene expression. In this work a number of bacterial twister motifs are characterized both in vitro and in vivo. The analysis of their genetic contexts shows that these motifs have the potential of being transcribed as part of polycistronic mRNAs, thus we suggest the involvement of bacterial twister motifs in the processing of mRNA. Our data show that the ribozyme-mediated cleavage of the bacterial 3'-UTR has major effects on gene expression. While the observed effects correlate weakly with the kinetic parameters of the ribozymes, they show dependence on motif-specific structural features and on mRNA stabilization properties of the secondary structures that remain on the 3'-UTR after ribozyme cleavage. Using these principles, novel artificial twister-based riboswitches are developed that exert their activity via ligand-dependent cleavage of the 3'-UTR and the removal of the protective intrinsic terminator. Our results provide insights into possible biological functions of these recently discovered and widespread catalytic RNA motifs and offer new tools for applications in biotechnology, synthetic biology and metabolic engineering.

Keywords: Aptazyme; RNA decay; RNase; bacteria; hammerhead ribozyme; polyadenylation; riboswitch; secondary structure; twister ribozyme.

Figure 1.

Secondary structures and natural genetic contexts of the twister ribozyme motifs. (A) Pseudoknots Pk1 and Pk2 are highlighted in green and blue, respectively. The inactivating mutation is highlighted in yellow. The cleavage site is indicated with a red arrowhead. The constructs Gob-1-1_w/o_P0 and Sva-1-1_w/o_P0 were generated removing the P0 stem at the position indicated by the gray arrow. The flanking sequences of the 3 type P5 motifs are shown in light gray and were employed in the in vivo assay. (B) Natural genetic context of the 6 investigated twister motifs. Genes of phage and bacterial origin are represented in orange and blue, respectively. Transposase, tRNA and hypothetical genes are represented in green, magenta and gray, respectively. Genes with the opposite orientation, whose coding sequence is present on the (-) strand, are indicated with an arrow. The putative promoters (scores between 0.95 and 1.00) are represented with a cornered arrow and putative intrinsic terminators are represented with a hairpin (for details see Table S1 and S2). The only promoter identified in proximity of the Gob motifs has a score of 0.88.

Figure 2.

Kinetics of the investigated twister motifs. The kinetics were determined in 50 mM Tris-HCl pH 7.5, 0.1M KCl and 1 mM MgCl₂. The kinetic traces were fitted with either mono exponential (Cbo-2-1, Cas-1-1 and Cpa-1-1) or double exponential functions (Gob-1-1, Gob-1-1 w/o P0, Sva-1-1 and Sva-1-1 w/o P0). The error bars represent the standard deviations calculated on independent triplicates. The relative kinetic parameters are reported in Table 1 and Table 2. Representative PAGE images of the kinetics and of the EDTA controls are reported in Fig. S1.

Figure 3.

Gene expression assay and correlation analysis. (A) The general design of the reporter type P1 and type P5 constructs. The cleavage site is shown by a red arrowhead. (B) eGFP expression levels of the catalytically active and inactive twister constructs inserted into the eGFP 3′-UTR. Error bars represent standard deviations calculated on independent biological quadruplicates. (C) Scatter plots of the levels of eGFP expression of the active twister constructs against the observed cleavage rate constant k_obs (left panel) and the final fraction of cleaved RNA y₀ (right panel) for each motif. The error bars represent standard deviations derived from the fit with equation (1). Correlation coefficients and p-values were calculated for both data sets.

Figure 4.

Design and performances of the artificial riboswitches in the 3′-UTR. The twister-based artificial riboswitches based on the naturally occurring env-9 motif (A) with shortened P1 and P3 stems (B). Theophylline and TPP aptazymes were generated connecting 2 aptamer domains to the stem P1 of the catalytic motif (C). Previously isolated communication modules were employed (Theo 5, Theo 6, TPP 2 and TPP 4). In addition 2 combinatorial libraries were generated using communication modules of 4 randomized nucleotides. The inactivating mutation is highlighted in yellow. The cleavage site is indicated by an arrowhead. (D,E) left panel: eGFP expression (bulk fluorescence divided by the relative OD₆₀₀) of the selected active and inactive clones in the absence (−) and presence (+) of (D) 2.5 mM theophylline and (E) 1 mM thiamin in the culture medium. Right panel, the performances (fold of activation or inactivation) of the on- and off-switches were calculated as the ratio of fluorescence of active divided by inactive expression states using background subtraction. The levels of reporter gene expression in the different conditions are represented by gray (controls), red (off-switches) and blue bars (on-switches). Error bars represent standard deviation calculated from independent biological triplicates. The identified communication modules are reported. Note that the numbering of the switches was assigned to be consistent with previously identified communication modules for twister-based theophylline and TPP switches in the 5′-UTR.