Regulation of Heterogenous LexA Expression in Staphylococcus aureus by an Antisense RNA Originating from Transcriptional Read-Through upon Natural Mispairings in the sbrB Intrinsic Terminator

Abstract

Bacterial genomes are pervasively transcribed, generating a wide variety of antisense RNAs (asRNAs). Many of them originate from transcriptional read-through events (TREs) during the transcription termination process. Previous transcriptome analyses revealed that the lexA gene from Staphylococcus aureus, which encodes the main SOS response regulator, is affected by the presence of an asRNA. Here, we show that the lexA antisense RNA (lexA-asRNA) is generated by a TRE on the intrinsic terminator (TT_sbrB) of the sbrB gene, which is located downstream of lexA, in the opposite strand. Transcriptional read-through occurs by a natural mutation that destabilizes the TT_sbrB structure and modifies the efficiency of the intrinsic terminator. Restoring the mispairing mutation in the hairpin of TT_sbrB prevented lexA-asRNA transcription. The level of lexA-asRNA directly correlated with cellular stress since the expressions of sbrB and lexA-asRNA depend on the stress transcription factor SigB. Comparative analyses revealed strain-specific nucleotide polymorphisms within TT_sbrB, suggesting that this TT could be prone to accumulating natural mutations. A genome-wide analysis of TREs suggested that mispairings in TT hairpins might provide wider transcriptional connections with downstream genes and, ultimately, transcriptomic variability among S. aureus strains.

Keywords: Staphylococcus aureus; antisense RNA; lexA; post-transcriptional regulation; transcriptional read-through; transcriptional termination.

Figure 1

lexA-asRNA is produced from the SigB-dependent sbrB promoter upon transcriptional read-through of TT_sbrB. (A) Schematic representation of the sbrB-lexA-sosA locus according previous transcriptomic maps [4]. ORFs and promoters (P) from sbrB, lexA and sosA genes are represented as blue, orange and gray arrows, respectively. Transcriptional terminators (TT) are illustrated as colored hairpins. Transcripts (RNAs) generated from both DNA strands are represented as dashed arrows. (B) Validation of sbrB mRNA boundaries. The transcriptional start and termination sites of the sbrB were determined by visualizing the S. aureus TSS sequencing data [30] and performing a simultaneous mapping of the 5′ and 3′ mRNA ends by circularization (mRACE) [31]. A Jbrowser image showing the RNA-Seq reads mapping on the sbrB promoter region is also included. The complete transcriptomic map is available at http://rnamaps.unavarra.es, accessed on 8 December 2021 [7]. Red bars represent the frequency of each nucleotide position at the 5′ and 3′ ends identified by mRACE. (C) Northern blots showing the sbrB, lexA-asRNA and sosA mRNA levels expressed from the S. aureus WT and ΔTTsbrB strains. Transcripts were developed using ³²P-radiolabelled riboprobes designed to specifically target the sbrB (RP_sbrB), lexA-asRNA (RP_AS) and sosA (RP_sosA) mRNAs. The single-stranded transcript sizes from the RNA Millennium marker are indicated. Midori green-stained ribosomal RNAs are included as loading controls.

Figure 2

Alkaline stress activates the sbrB promoter and lexA-asRNA expression. Schematic representations of plasmid constructions including (A) the P_sbrB transcriptional reporter and (C) TT_sbrB transcriptional read-through reporter. (B,D) Western blots showing the GFP protein levels from the WT and ΔsigB strain transformed by plasmids illustrated in A and C, respectively. Bacteria were grown until exponential phase and, when necessary, challenged with KOH for 60 min to induce alkaline stress. Proteins were transferred to nitrocellulose membranes, incubated with anti-GFP monoclonal antibodies and developed using peroxidase-conjugated goat anti-mouse antibodies and a bioluminescence kit. Coomassie stain gel portions are included as loading controls.

Figure 3

The TT_sbrB read-through is not modulated by SbpB translation. (A) Schematic representation of the sbrB mRNA indicating the ribosome binding site (RBS), the putative start codons (AUG), the small protein SbpB sequence (lysines are colored in red) and the transcriptional terminator structure. The nucleotide modifications performed in the sbrB mRNA are highlighted in yellow. (B) Schematic representation of the different plasmids harboring translation and the transcriptional read-through reporters, respectively. WT and mutant mRNAs were expressed under the control of the P_blaZ+1 promoter. (C) Western blot analyses showing the GFP levels produced from the different plasmids. Membranes were incubated with monoclonal anti-GFP and developed with peroxidase-conjugated goat anti-mouse antibodies and a bioluminescence kit. Coomassie gel portions are included as loading controls (LC).

Figure 4

A single nucleotide change is responsible for the TT_sbrB read-through. (A) Putative TT_sbrB structures of the S. aureus 15981 and MW2 strains. The mutation found in the S. aureus 15981 strain that generated the nucleotide mispairing in the TT_sbrB is highlighted in red. (B) Northern blot analyses showing the lexA-asRNA and sbrB expression levels from the S. aureus 15981, MW2, A112G and ΔTT strains. Bacteria were grown at 37 °C in MHg until exponential phase (OD₆₀₀ 0.3). Midori green-stained ribosomal RNAs are included as loading controls.

Figure 5

Natural mutations found in the TT_sbrB sequence influence the read-through levels. (A) Localization of nucleotide variations found in the TT_sbrB. The changes in the TT structure produced by the mutations are indicated. (B) Putative RNA structures of selected TT_sbrB variants. Mispairing nucleotides are highlighted in red. Frequencies of the mutations among staphylococcal strains are indicated in Table S4. (C) Western blot analysis showing the levels of transcriptional read-through (RT) produced by the different nucleotide variations were monitored by measuring the GFP expression. The different sbrB mRNA variants were cloned upstream of the gfp gene reporter and expressed under the control of the P_blaZ+1 promoter. Total proteins transferred to nitrocellulose membranes were incubated with monoclonal anti-GFP and developed with peroxidase-conjugated goat anti-mouse antibodies and a bioluminescence kit. Coomassie gel portions are included as loading controls (LC).

Figure 6

lexA-asRNA impacts the mCherry LexA reporter expression depending on the P_sbrB and P_lexA ratio. (A,C,E) Schematic representations of dual fluorescent reporter plasmids carrying the sbrB-lexA locus and its corresponding variations. These plasmids monitored the lexA-asRNA expression and P_lexA promoter activity by producing GFP and mCherry, respectively. In order to avoid autoregulation, the lexA CDS was substituted by the mCherry CDS while preserving the 3′- and 5′-UTRs of lexA mRNA. When indicated, the P_sbrB and P_lexA promoters were replaced by the P_blaZ+1 and P_fmtC promoters, respectively. For each promoter combination, a dual fluorescent reporter plasmid carrying the TT_sbrB sequences from the S. aureus 15981 (TT₁₅₉₈₁, pink hairpin) and MW2 (TT_con, blue hairpin) strains and a ΔTT_sbrB mutant was constructed. (B,D,F) Western blot analyses showing the GFP and mCherry levels expressed from the strains carrying the dual reporters shown in (A,C,E). Total proteins were transferred to nitrocellulose membranes and incubated with monoclonal anti-GFP or anti-mCherry antibodies. These were then developed with peroxidase-conjugated goat anti-mouse or goat anti-rabbit antibodies, respectively, and a bioluminescence kit. Coomassie gel portions are included as loading controls (LC).

Figure 7

Regulation of heterogenous LexA expression by lexA-asRNA. Time-lapse fluorescence microscopy was performed to monitor GFP and mCherry expression at the single-cell level in S. aureus 15981 strains carrying one of the four dual fluorescent plasmid reporter variants: (A) TT₁₅₉₈₁, (B) ΔP_sbrB, (C) ΔTT_sbrB and (D) TT_con. The main genetic elements of these constructions are illustrated on top of each corresponding panel. Bacteria were grown at 37 °C in CellAsic microfluidic plates with a continuous flow of MHg and challenged with 30 mM of KOH for 4 h. Images were taken in 15 min intervals. The mCherry and GFP fluorescence, the combination of both signals (merge panels) and the differential interference contrast (DIC) images are shown. Quantification of bacterial growth and single-cell GFP and mCherry levels are shown in Supplementary Figure S4.

Figure 8

Evolutionarily selected mispairing nucleotides in TTs are relevant for read-through events. (A) Structures of predicted TTs (1828, 1022 and 619) including mispairings and restored TT variants (res) that compensated the mispairings. Nucleotide changes are colored in red. (B) The TT read-through levels from each variation were monitored by Western blot analyses using the dual fluorescent reporter plasmid shown in Supplementary Figure S6. The GFP expression levels indicate the amount of transcriptional read-through while mCherry monitors the transcript level expressed from the P_sbrB promoter. Proteins transferred to nitrocellulose membranes were developed as indicated in Figure 6. Coomassie-stained gel portions are included as loading controls. (C) Identification of natural nucleotide variations in the corresponding TTs. The sequences of the selected TTs were compared by BLASTN against all S. aureus genomes available in the NCBI database. n represents the number of genomes compared for each TT. The nucleotide variations and the number of genomes carrying such variations are indicated in the corresponding TT position.