RNase E-dependent processing stabilizes MicX, a Vibrio cholerae sRNA

Abstract

In Vibrio cholerae, bioinformatic approaches have been used to predict the locations of numerous small RNA (sRNA)-encoding genes, but biological roles have been determined for very few. Here, we describe the expression, processing and biological role of an sRNA (previously known as A10) that was identified through such analyses. We have renamed this sRNA MicX as, like the Escherichia coli sRNAs MicA, MicC and MicF, it regulates expression of an outer membrane protein (OMP). MicX appears to be a direct negative regulator of vc0972, which encodes an uncharacterized OMP, and vc0620, which encodes the periplasmic component of a peptide ABC transporter. Hfq is apparently not required for MicX's interactions with and regulation of these targets. The sequence encoding MicX overlaps with vca0943; however, primary transcripts of MicX are processed in an RNase E- and Hfq-dependent fashion to a shorter, still active and much more stable form consisting largely of the vca0943 3' untranslated region. Our data suggest that processing of MicX enhances its effectiveness, and that sRNA cleavage is not simply a means to sRNA inactivation and clearance.

Fig. 1. Expression of micX, an sRNA-encoding gene that overlaps with vca0943

A. Schematic depiction of the micX-encoding region of the V. cholerae genome. The grey arrows represent predicted coding sequences, the black arrow shows the sRNA (MicX, formerly A10) predicted by bioinformatic analysis, and short lines show the positions of complimentary oligonucleotide probes used for Northern hybridizations. The probe A0943R1 is complementary to the final 10 codons of vca0943. The lollipop shows the position of a predicted rho-independent terminator for vca0943. B. Northern blot probed with oligonucleotide A10-2 (upper panel) or a control probe for 5S RNA (vc5Sa; lower panel). Wt (lanes 1 and 3), hfq (lanes 2 and 4) and rpoS (lane 5) V. cholerae were grown in LB to either log (lanes 1 and 2) or stationary (stat) phase (lanes 3–5). C. Northern blot as in (B), probed with A0943R1 (upper panel) and with vc5Sa (lower panel). RNA was electrophoresed on glyoxal gels. D and E. Northern blots of polyacrylamide gels probed with A10-2, A0943R1 and vc5Sa as above. Wt (lanes 1 and 2) and hfq (lanes 3 and 4) V. cholerae were grown in LB (lanes 1 and 3) or M63 + maltose and casamino acids (lanes 2 and 4). a, b and c correspond to the long, intermediate and short forms of MicX subsequently determined to be 346, 238 and 188 nt in length.

Fig. 3

Analysis of MicX endonucleolytic cleavage. Northern blots of polyacrylamide gels probed with A10-2 and vc5Sa (A, B and D) and primer extension analyses (C) were used to monitor MicX processing (a, b and c, as in Fig. 1). A. MicX transcripts in wt V. cholerae (lane 1), V. cholerae micX_Δ196−263 (pMicX_-2-346) (lane 2), and E. coli DH5α (pMicX_-2-346) (lane 3). MicX form ‘a*’ is slightly longer than form ‘a’ due to the presence of additional vector-derived sequences. B. E. coli strains N3433 (wt, lanes 1–3), N3431 [rne 3071 (ts); lanes 4 and 5] and AB301-105 (rnc-105; lane 6) transformed with pMicX_-2-346 were cultured with (lanes 2–6) or without (lane 1) arabinose to induce production of MicX. Cells were maintained at 37°C (lanes 1, 2, 4 and 6) or were transferred to the non-permissive temperature of 44°C (lanes 3 and 5). C. Primer extension analyses of RNA from E. coli N3433(pMicX_-2-346) (lane 1) and wt V. cholerae (lane 2), run adjacent to a sequencing ladder generated with the same primer. D. Northern blot analysis of MicX RNA in wt (lane 1), hfq (lane 2), micX_Δ196−263 (lane 3) and micX_Δ196−263hfq (lane 4) V. cholerae. E. Summary of primer extension and sequence analyses of MicX. The full-length MicX sequence is presented. The stop codon of vca0943 is shown in bold, and cleavage sites (the 5′ ends of MicX forms b and c) are underlined. The region of MicX deleted in micX_Δ196−263 is shown in grey.

Fig. 2

Identification of the MicX promoter. Plasmids pBD1821, pBD1822 and pBD1858 are transcriptional reporter fusions containing DNA fragments (grey boxes) overlapping the 5′ ends of the various forms of MicX (a–c, as in Fig. 1, shown at the top of the figure). Reporter activity was assessed in wt and hfq V. cholerae grown to log phase in M63 + glucose (M63 g), M63 + maltose and casamino acids (M63mc), and LB. The numbers shown are an average derived from at least three independent cultures. The 5 and 3′ ends of a, b and c were determined using 3′ RACE and primer extension. lacZ transcripts derived from pBD1822 have the same 5′ end as the longest form (a) of MicX.

Fig. 4

Northern blot analysis of processed and precursor MicX stability. RNA was isolated from early stationary phase (OD₆₀₀ ∼1.1) cultures of wt and hfq V. cholerae at varying times following addition of rifampicin (50 μg ml⁻¹), and electrophoresed on an acrylamide gel. The blot was hybridized to probe A10-2. The exposure time for the right panel was longer than that for the left panel. Significant overexposure of the left panel (not shown) reveals full-length MicX (form a) in wt cells at t = 0 only. MicX forms a, b and c are as previously described.

Fig. 5

Northern blot analysis of MicX and its potential targets. A–E. RNA from log-phase cultures of wt (lanes 1 and 3) and micX_Δ196−263 (lanes 2 and 4) V. cholerae grown in LB (lanes 1 and 2) or M63 + glucose (M63g; lanes 3 and 4) was hybridized to probes for the genes indicated for each blot. F–H. RNA from log-phase cultures of wt V. cholerae transformed with pAntiMicX_-37-309 (lanes 1 and 2) or pAntiMicX_103-309 (lanes 3 and 4) and grown in M63 + maltose and casamino acids, in either the absence (lanes 1 and 3) or presence (lanes 2 and 4) of arabinose, was hybridized to probes for the genes indicated under each blot. RNA was electrophoresed on glyoxal (A–E, G, H) or polyacrylamide (F) gels.

Fig. 6

Northern blot analysis of the activity of processed and unprocessed MicX. A–D. RNA from micX_Δ196−263V. cholerae transformed with pMicX_-2-346 (lane 1), pMicX_146-346 (lane 2) or pBAD33 (lane 3), or from wt V. cholerae (pBAD33) (lane 4) cultured in M63 + maltose, casamino acids and arabinose was analysed on Northern blots hybridized to probes for the genes noted for each blot. E–G. Northern blots of RNA from hfq (pMicX_-2-346) (lane 1), hfq (pMicX_146-346) (lane 2), hfq (pBAD33) (lane 3), wt (pBAD33) (lane 4) and micX_Δ196−263 (pBAD33) (lane 5), hybridized to probes for the indicated genes. Strains were grown in LB with arabinose. RNA was electrophoresed on polyacrylamide (A, E) or glyoxal (B–D, F and G) gels. MicX forms a, a*, b and c are as previously described; c* contains 12 nt upstream of the downstream MicX processing site plus additional vector-derived sequences.

Fig. 7. Bioinformatic and functional characterization of mRNA sequences likely to interact with MicX

A. Pairing between MicX and vc0620 mRNA according to TargetRNA. MicX is numbered relative to the start of transcription; vc0620 RNA is numbered relative to the start of translation. Vc0620s start codon and probable ribosome binding site are underlined. B. Schematic representation of vc0972::lacZ reporter fusions and their activity (in Miller units) in the presence or absence of MicX. Numbered relative to the start of translation, pBD1846 contains nt −402 to −26 of vc0972, and pBD1847 and pBD1839 contain nt −402 to 55. Primer extension analyses (not shown) suggest that transcription initiates at nt −137. Translation from pBD1839 transcripts utilizes the vc0972 ribosome binding site. Translation from pBD1847 utilizes only the lacZ ribosome binding site, as there are two stop codons downstream of the vc0972 ribosome binding site. Cells were cultured in LB + arabinose. The numbers shown are an average obtained from at least three independent cultures.