Unconventional mRNA processing and degradation pathways for the polycistronic yrzI (spyTA) mRNA in Bacillus subtilis

Laetitia Gilet; Magali Leroy; Alexandre Maes; Ciarán Condon; Frédérique Braun

doi:10.1002/1873-3468.70027

. 2025 Mar 8;599(9):1222–1235. doi: 10.1002/1873-3468.70027

Unconventional mRNA processing and degradation pathways for the polycistronic yrzI (spyTA) mRNA in Bacillus subtilis

Laetitia Gilet ¹, Magali Leroy ¹, Alexandre Maes ², Ciarán Condon ^1,^✉, Frédérique Braun ^1,^✉

PMCID: PMC12067863 PMID: 40055937

Abstract

The ribosome‐associated endoribonuclease Rae1 cleaves the Bacillus subtilis yrzI operon mRNA in a translation‐dependent manner. This operon encodes up to four small peptides, S1027, YrzI, S1025, and S1024, whose functions are unknown. Here, we identified the function of YrzI and S1025 and deciphered the degradation pathways of the yrzI polycistronic mRNA. We show that YrzI is toxic at high concentrations, but co‐expression with S1025 abolishes its toxicity, and that, in the absence of Rae1, S1025 is the major antidote to the YzI toxin. We show that a highly stable mRNA species containing the YrzI and S1025 open reading frames results from endoribonucleolytic cleavage upstream of yrzI followed by the arrest of 5′‐exoribonucleolytic processing by ribosomes bound to its exceptionally strong Shine–Dalgarno sequence. Degradation of this mRNA requires either translation‐dependent cleavage within S1025 by Rae1 or direct attack from the structured 3′‐end by 3′‐exoribonucleases. Neither pathway is common for a B. subtilis mRNA.

Keywords: B. subtilis, Rae1, RNA maturation and decay, RNases, translation

The S1025 peptide is the major antidote to the YrzI toxin, which we renamed here as SpyT (Small Peptide YrzI Toxin) and SpyA (Small Peptide YrzI Antitoxin) (1). Degradation of the toxin–antitoxin spyTA mRNA, either by a translation‐dependent cleavage by the endoribonuclease Rae1 (2) or by direct attack by 3′‐exoribonucleases (3), also contributes to counteract SpyT/YrzI toxicity.

graphic file with name FEB2-599-1222-g007.jpg

Abbreviations

aa, amino acid

nt, nucleotide

ORF, open reading frame

SD, Shine‐Dalgarno

spyA, small peptide yrzI antitoxin

spyT, small peptide yrzI toxin

TA, toxin‐antitoxin

Two major mRNA degradation pathways have been characterized in Bacillus subtilis thus far [1, 2]. The 5′‐exoribonucleolytic (5′‐exo) pathway resembles the mRNA decapping and Xrn1‐dependent pathway found in eukaryotes [1, 2] After deprotection of the mRNA by an RNA pyrophosphohydrolase, e.g., BsRppH, the 5′‐exoribonuclease complex RNase J1/J2 degrades the mRNA all the way to the 3′ end [3]. In the endoribonucleolytic (endo) pathway, the mRNA is cleaved internally and the resulting fragments degraded by either 3′‐ or 5′‐exoribonucleases. The key endoribonucleases known to initiate mRNA decay in B. subtilis are the single strand‐specific enzyme RNase Y, a functional analog of E. coli RNase E, and the double strand‐specific enzyme RNase III [4]. In general, the activity of these enzymes is inhibited by translation, with the flow of ribosomes occluding access to cleavage sites within open reading frames (ORFs) [5, 6, 7, 8, 9]. In 2017, we identified a novel component of the B. subtilis mRNA decay machinery, the ribosome‐associated endoribonuclease Rae1 [10] that, in contrast to RNase III and RNase Y, requires translation for cleavage [11, 12].

We have thus far identified two Rae1 targets: the bmrBCD operon mRNA [12], encoding a multidrug transporter, and the yrzI operon mRNA [10]. Ribosome profiling experiments have suggested that the latter operon, in addition to encoding the annotated 49‐amino acid (aa) YrzI peptide, expresses three supplementary peptides S1027, S1025 and S1024, that are 38, 17, and 52 aa's in length, respectively [13]. We previously showed that Rae1 cleaves within the S1025 ORF in a translation‐dependent manner, confirming that the 17‐aa S1025 peptide is indeed translated [10]. However, the functions of the different peptides expressed from this operon remain mysterious.

Here, we show that the overexpression of the YrzI peptide impairs B. subtilis growth, an effect that is counteracted by S1025, suggesting that these two peptides constitute a toxin‐antitoxin (TA) system. We also show that expression of this operon is governed at the transcriptional level by the transition state regulator, AbrB, and at the post‐transcriptional level via non‐canonical mRNA processing and degradation pathways. Two primary transcripts undergo processing to a short highly stable ~ 500 nt mRNA fragment that is protected from 5′‐degradation by ribosomes bound to an exceptionally strong Shine‐Dalgarno (SD) sequence located upstream of yrzI. With the 5′‐degradation pathway blocked by initiating ribosomes, elimination of this mRNA fragment requires degradation by one of two alternative pathways (i) translation‐dependent endoribonucleolytic cleavage by Rae1, followed by degradation of the unprotected upstream and downstream fragments by 3′‐ and 5′‐exoribonucleases or (ii) direct attack by 3′‐exoribonucleases from the 3′ end of the mRNA despite the presence of a protective transcription terminator.

Materials and methods

Strain construction

Oligonucleotides and strains used in the paper are listed in Tables S1 and S2, respectively.

The rph::spc, rnr::tc, yhaM::pm, pnp::kan and pnp::cm cassettes to make strains CCB329, CCB395, CCB396, CCB407, CCB409 and CCB1210 in our lab background (SSB1002 = W168 trpC+) were a kind gift from David Bechhofer, and have been described previously [14]. Strain CC376 was constructed by transferring the rae1::pMUTIN construct from strain CCB375 into CCB396. Strain CC761 was constructed by transferring the rny::spec construct from strain CCB441 into CCB375.

The plasmid overexpressing the YrzI peptide was constructed by amplifying the yrzI gene by PCR using oligos CC1731/1732 and cloned in pDG148 cleaved with SalI and HindIII. The resulting plasmid pDGYrzI (pl. 687) contains the yrzI gene under Pspac control, followed by an E. coli rRNA transcription terminator to limit transcription of downstream plasmid DNA. This plasmid was transformed into CCB375 (Δrae1) to create strain CCB815. We initially failed to transform this plasmid into wild‐type cells. We thus re‐isolated it from strain CCB815, transformed it with high efficiency into SSB1002 (WT) to create strain CCB839. The plasmid was re‐isolated from CCB839 to confirm its sequence. pDGYrzI‐S1025 (pl. 726) and pDGS1025 (pl. 727) were constructed by amplifying yrzI‐S1025 and S1025 fragments by PCR using oligo pairs CC1731/1809 and CC1808/1809. PCR fragments were cleaved with SalI and HindIII and cloned in pDG148. The resulting plasmids were transferred to B. subtilis WT and ∆rae1 cells to create strains CCB913, CCB914 (pl. 726) and CCB915, CCB916 (pl. 727), respectively.

Northern blots and primer extension assays

Northern blots were performed on total RNAs isolated either by the glass beads/phenol method described in [15] or by the RNAsnap method described in [16]. Northern blots were performed as described previously [4]. Primer extension assays were performed using the oligo CC1589 on glass bead/phenol extracted RNAs as described previously [17].

Spot dilution assays

Cells were grown in 2xYT medium to OD₆₀₀ = 0.6. When cells contained either the empty plasmid pDGYrzI (pl687), pDGYrzI‐S1025 (pl726) and pDGS1025 (pl727), serial dilutions were spot on LB plates containing kanamycin (5 μg·mL⁻¹) to maintain the plasmid. When indicated, IPTG at 100 μm to induce the expression of the inserted gene was added.

Ribosomal subunit (30S) protection assay

Five nanomoles of a 3′‐labeled (p³²Cp) yrzI RNA was mixed with 500 nm B. subtilis 30S ribosomal subunits isolated from an RNase J1‐depleted strain as described in [17] in 4 μL RNase J1 reaction buffer. The yrzI RNA was synthesized in vitro by T7 RNA polymerase using a MegaShortScript kit (Thermo Fischer Scientific, Waltham, MA, USA) from PCR fragments amplified from chromosomal DNA using oligo pair CC3176/CC3042 (CC3176 had an integrated T7 RNA polymerase promoter sequence). Subunits were allowed to bind for 10 min at 37 °C, followed by 10 min on ice. A total of 1.8 μg (1 μL) RNase J1 was added to start the reaction. Reactions were incubated at 25 °C for 10 min, stopped as above, and loaded directly onto 5% polyacrylamide denaturing gels.

Results

YrzI encodes a toxic peptide that is counteracted by S1025

The yrzI operon encodes two proteins (YrhF and YrhG) and four small peptides S1027, YrzI, S1025, and S1024, whose functions are all unknown (Fig. 1A). We previously showed that Rae1 cleaves within the S1025 ORF in a translational‐dependent manner between codons 13 and 14 to initiate the degradation of the yrzI polycistronic mRNA and that deletion of the rae1 gene led to an accumulation of three major RNAs of ~ 2.4 kb (P1‐T4), ~ 0.8 kb (P3‐T4) and ~ 0.5 kb in size (R‐T4), all containing the yrzI open reading frame [10] (Fig. 1A). Since small peptides can have toxic effects on bacterial cell growth [18], we wondered whether the Rae1 cleavage site within S1025 might serve to eliminate this mRNA to protect cells from potential toxicity of one of its encoded peptides. Because YrzI was the originally annotated peptide of this operon and we had direct evidence that S1025 was translated, we began by analyzing the effect of these two peptides on cell growth by constructing two replicative plasmids, pDG‐YrzI and pDG‐S1025, permitting overexpression of YrzI and S1025 under control of an IPTG‐dependent promoter. Induction of yrzI expression by addition of IPTG led to a 3‐ to 4‐log inhibitory effect on B. subtilis growth in WT and Δrae1 cells (Fig. 1B), suggesting that the YrzI peptide is indeed toxic at high doses. In contrast, no toxic effect of S1025 overexpression was observed in either WT or ∆rae1 cells (Fig. 1B). Since many toxins are encoded in operons with their antitoxins just downstream [18], we assayed the effect of producing both peptides together by cloning the YrzI and S1025 region under control of an IPTG‐dependent promoter (pDG‐YrzI‐S1025). Remarkably, co‐expression of S1025 abolished the toxicity of YrzI (Fig. 1B). The protective effect of adding S1025 is not simply due to destabilization of the yrzI‐S1025 mRNA through addition of the Rae1 cleavage site, since expression of S1025 neutralized the toxicity of YrzI even in absence of Rae1 (Fig. 1B).

Thus, two different levels of controlling YrzI expression/toxicity exist: cleavage of its mRNA by Rae1 within the S1025 ORF and a Rae1‐independent antitoxin effect of the S1025 peptide.

The polycistronic yrzI mRNA is regulated by AbrB at the transcriptional level

It has previously been shown that the expression of the bmrBCD operon is repressed by the transition state regulator AbrB at the transcriptional level [19], in addition to the post‐transcriptional regulation mediated by Rae1 [12]. Interestingly, transcriptional repression by AbrB is also predicted to be shared by the yrzI operon [20] and an AbrB binding site was mapped upstream of the P3 promoter [21]. To confirm this, we measured the levels of yrzI mRNA in strains lacking rae1, abrB or both, at mid‐exponential growth phase, by Northern blot (Fig. 2). Deletion of abrB alone caused a small accumulation of the P3‐T4 yrzI transcript compared to WT (Fig. 2, compare lane 3 to 1), and deletion of both rae1 and abrB caused greater increase in accumulation of both P3‐T4 and R‐T4 RNAs than deletion of rae1 alone (Fig. 2, compare lane 4 to 2). These results confirm that AbrB is a repressor of yrzI operon expression in mid‐exponential phase. Expression of the yrzI operon is thus modulated at two levels, transcriptionally by AbrB and post‐transcriptionally by Rae1.

Fig. 2 — The polycistronic *yrzI* mRNA is regulated by AbrB at the transcriptional level. Northern blot showing steady state level of the *yrzI* and *abrB* transcripts in WT, *∆rae1*, *∆abrB* and *∆rae1 ∆abrB* strains at mid‐log phase (OD600 = 0.5). The blot was probed with oligo CC1589 (*yrzI*) and with a probe complementary to 16S rRNA (CC058) as a loading control. This experiment was repeated twice.

Mapping of the major RNAs produced from the yrzI locus

A transcriptome study by [22], identified several potential promoter (P1, P2, P3, P4, and P5) and terminator sequences (T2, T3, and T4) around the yrzI locus (Fig. 1A). However, Northern blots probed for yrzI only accounted for transcripts originating at P1 or P3 and ending at T4 (Fig. 3A, left panel). To get a more complete picture of the RNAs mapping to this locus, we hybridized Northern blots with oligonucleotide probes complementary to the S1027 or S1024 ORFs, upstream and downstream of yrzI, respectively. To determine their sensitivity to Rae1, these blots contained total RNA isolated from WT, ∆rae1 and plasmid complemented strains grown in the absence or presence of IPTG.

As seen previously [10], only the 2.4 kb (P1‐T4), 0.8 kb (P3‐T4) and 0.5 kb (R‐T4) RNAs, which contain S1025, had Rae1‐dependent expression profiles (Fig. 3A and B). The S1027 probe hybridized to the Rae1‐sensitive P1‐T4 and P3‐T4 transcripts, as expected, and to three additional Rae1‐insensitive species of ~ 1.9 kb, ~ 220 nts and ~ 260 nts (Fig. 3A, middle panel and Fig. 3B). These additional bands were not detected with the yrzI and S1024 probes and their sizes are consistent with transcripts originating at P1 and P3 and terminating at T2 or T3, i.e. P1‐T2, P1‐T3 (indistinguishable in size on the Northern blot) and P3‐T2 and P3‐T3, respectively (Fig. 3A). The S1024 probe hybridized to same three major transcripts as the yrzI probe (Fig. 3A, right panel). We also detected a very weak additional minor Rae1‐independent species (around ~ 350 nts), whose origin remains unclear. We did not find any evidence for the use of the putative P2, P4 or P5 promoters under the rich medium conditions tested. More importantly, the most abundant (0.5 kb) species in the rae1 deleted strain, R‐T4, lacks an obvious promoter. We therefore concluded that it most likely results from processing of a primary transcript originating from P1 or P3.

The R‐T4 species results from 5′ degradation by RNase J1 and is degraded via two alternative pathways

To determine the maturation event(s) that led to the production of the R‐T4 species and which RNases eliminate the cleavage products, we performed high‐resolution Northern blot analysis in different B. subtilis RNase mutants, containing or lacking Rae1.

In the absence of the 5′‐exoribonuclease RNase J1 (ΔrnjA), the P3‐T4 primary transcript was still detected but the R‐T4 band was replaced by a band migrating about 100 nts slower (referred as E1‐T4), detected with both the yrzI and S1024 probes (Fig. 4A, lanes 3 and 11). This suggests that the R‐T4 species is generated by RNase J1‐mediated degradation from an upstream entry site, E1, until RNase J1 reaches the site labeled R (Fig. 4B). The E1‐T4 species was more clearly visible in the Δrae1 rnjA double mutant strain, consistent with the ability of Rae1 to trigger its degradation (Fig. 4A, lanes 4 and 12). A weak band (indicated with an asterisk) that migrates just above the R‐T4 species was also detected in the absence of the RNase J1, which may result from inhibition of 5′ exoribonucleolytic degradation by RNase J1 at the T2 or T3 terminator hairpins and thus, may correspond to T2‐T4 or T3‐T4. This species as well as the R‐T4, P3‐T4, and E1‐T4 RNA were each detected using a probe that hybridized to the T4 terminator confirming they all retain this structure at their 3′ ends (Fig. S1).

Fig. 4 — The R‐T4 species results from 5′ degradation by RNase J1 is degraded via two alternative pathways (A) Expression of *yrzI* in different RNase mutants. High‐resolution Northern blot, probed with oligo CC1589 (*yrzI*; green dot) or CC1600 (*S1024*; gray dot), showing expression of *yrzI* in strains lacking specific *B. subtilis* RNases. Gene abbreviations are as follows: Rae1 (*rae1*), the four 3′‐exoribonucleases PNPase, RNase R, RNase PH and YhaM (*4exos*), RNase J1 (*rnjA*), RNase Y (*rny*). Blots were rehybridized with a probe complementary to 5S rRNA (oligo HP246) as a loading control. The size of the RNAs detected was determined using compilation of 5′ end mapping experiments from this study and previous papers [10]. This experiment was repeated twice. (B) Summary of degradation pathways of short *yrzI* operon mRNAs. The structure of the region between S1027 and S1024 is shown. ORFs are represented by gray arrows, putative promoters by rightward‐pointing arrows and transcription terminators by the hairpin structures. The sizes of the ORFs are given in amino acids (aa) and the positions of probes used are indicated by green and gray bars. The primary transcript (P3‐T4) and the degradation intermediates are depicted. Endoribonucleases are symbolized by scissors and exoribonucleases by a Pacman symbol. Ribosomes protecting the 5′ end of the R‐T4 transcript and involved in Rae1 cleavage are indicated.

We addressed the possibility that the major endoribonuclease RNase Y was responsible for the cleavage at E1, by asking whether the E1‐T4 and of the R‐T4 species were still detectable in the ∆rny rae1 double mutant strain (Fig. 4A, lane 8 and data not shown). The level of R‐T4 species was weaker in the ∆rny rae1 strain compared to the ∆rae1 mutant alone and the E1‐T4 RNA was no longer detected (Fig. 4A, compare lanes 2 and 8). This suggests that RNase Y contributes to the cleavage at E1 and that either a second endoribonuclease is involved or, in the absence of cleavage by E1 RNase Y, RNase J1 can gain access to the mRNA from a dephosphorylated 5′‐end of the primary transcript to generate R‐T4.

Strikingly, the R‐T4 RNA also strongly accumulated in a ∆4exos strain, lacking the four known B. subtilis 3′‐exoribonucleases (PNPase, RNase PH, RNase R and YhaM) but still expressing Rae1 (Fig. 4A, compare lane 5 to lane 1), suggesting that R‐T4 can also be directly degraded from its 3′‐end, independently of Rae1. This is unusual as 3′‐terminator structures are not typically thought to be vulnerable to direct attack by 3′‐exoribonucleases in bacteria [5, 23, 24, 25, 26]. Three shorter species were additionally detected in the ∆4exos strain with the yzrI probe, all of which disappeared upon further deleting the rae1 gene, indicating that they are all dependent on Rae1 cleavage and contain the S1025 ORF (Fig. 4A, lane 6). The shortest one likely corresponds to the upstream fragment resulting from Rae1 cleavage of R‐T4 (labeled R‐Rae1 in Fig. 4A, lane 5) consistent with the predicted size (241 nts) (Fig. 4B), while the two others could correspond to E1‐Rae1 (347 nts) and T3‐Rae1 (289 nts) respectively. As mentioned above, we suspect that the T3‐Rae1 species results from an inhibition of 5′ exoribonucleolytic degradation by RNase J1 by the hairpin of the T3 terminator structure. In the Δ4exos rae1 strain, an additional degradation intermediate of ~ 270 nts was detected whose origin also remains unclear (Fig. 4A, lane 6).

The use of a probe complementary to S1024 permitted the detection of the ~ 280‐nt downstream cleavage product of the R‐T4 species by Rae1 (Rae1‐T4) in the ΔrnjA mutant strain but not in the Δrae1 rnjA strain, as expected (Fig. 4A, compare lanes 11 and 12; Fig. 4B). The upstream and downstream fragments resulting from Rae1 cleavage were also detected in the ∆4exos and rnjA strains, respectively, in a global study to map sites of endoribonucleolytic cleavage to single nt resolution [27]. A longer species, which we call E2‐T4 (~330 nts) (Fig. 4B), was also detected in the Δrae1 rnjA strain (Fig. 4A, lane 12). As for the R‐T4 species, we confirmed that the E2‐T4 species possessed the T4 terminator using the T4 probe (Fig. S1, lane 12). The 5′‐end of the E2‐T4 species corresponds to an additional entry site (E2) for RNase J1, previously mapped by primer extension assay to within the yrzI ORF [10]. The endoribonuclease responsible for the cleavage at E2 remains unknown.

Globally, these data indicate that the primary transcripts originating from P1 or P3 are cleaved at the E1 site, followed by 5′‐trimming by RNase J1 until it reaches the site labeled R to generate the stable intermediate R‐T4. This intermediate can be degraded by either of two pathways: the first is initiated by Rae1 cleavage, followed by degradation of the products by 5′‐ and 3′‐exoribonucleases, and the second is mediated by 3′‐exoribonucleases from the 3′‐extremity, independently of Rae1. A summary of the RNA processing/degradation pathways for the different RNAs of the yrzI operon, combining data from this and our previous paper [10] is shown in Fig. 4B.

The R‐T4 species is protected from RNase J1 degradation by ribosomes initiating translation of yrzI

The yrzI ORF has an exceptionally strong Shine‐Dalgarno (SD) sequence, with 11 out of 12 possible Watson‐Crick (WC) base pairs with the 3′ end of 16S rRNA (Fig. 5A). We therefore postulated that the 5′‐end of the R‐T4 RNA might result from the blocking of RNase J1 by ribosomes tightly bound to the yrzI SD sequence, similar to the phenomenon reported for the hbs mRNA in B. subtilis and the cryIIIA mRNA from B. thuringiensis expressed in B. subtilis, which possess exceptionally strong SD, or SD‐like sequences, respectively [28, 29, 30, 31]. The hall‐mark of these mRNAs that are protected from RNase J1 degradation is a ribosome ‘heel‐print’ that occurs 8 nts upstream of the first G‐residue of the consensus GGAGG SD sequence (Daou‐Chabo et al., 2009) (Fig. 5A).

Fig. 5 — The 5′ end of the R‐T4 RNA is protected by a ribosome bound to the *yrzI* SD sequence (A) Potential base pairing between the *yrzI* SD sequence and the 3′ end of 16S rRNA. (B) Primer extension assay (oligo CC1589) on total RNA isolated from wild‐type (WT) and strains lacking Rae1 (*∆rae1*), RNase J1 (*∆rnjA*) and both (*∆rae1 rnjA*). Sequence lanes are labeled as their reverse complement to facilitate direct read‐out. Mapped 5′ ends are shown to the right of the autoradiogram. Reverse transcriptase (RT) stops corresponding to the T2 and T3 terminator structures and the predicted P3 promoter are also indicated. The origin of the band indicated by an asterisk remains unknown. This experiment was repeated twice. (C) Mapping of RNase J1 and Rae1‐dependent 5′ ends to the predicted secondary structure (RNAfold; http://rna.tbi.univie.ac.at/cgi‐bin/RNAWebSuite/RNAfold.cgi) of the S1027‐*yrzI* intergenic region. Co‐ordinates are given relative to the AUG start codon of YrzI (boxed). The predicted T2 and T3 terminator stem loops are shown. Nucleotides complementary to the anti‐Shine Dalgarno sequence of *B. subtilis* 16S rRNA are underlined (SD). (D) Protection by 30S ribosomes against RNase J1 degradation of the *yrzI* transcript. The autoradiogram shows a heel‐print assay on a 3′‐labeled (p³²Cp) RNA fragment of the *yrzI* transcript in the presence or absence of 30S ribosomes, incubated with RNase J1 for various times. A sequencing reaction on the template DNA was loaded on the left of the gel and labeled as reverse complement to facilitate direct read‐out. Note: there is a slight difference in migration (equivalent to about 5 nts for the full‐length (FL) transcript and about 4 nts for the heel‐print) between the DNA sequence and the RNA reactions due to the 16 Da difference in MW between DNA and RNA nts. This experiment was repeated twice.

To identify the precise 5′‐ends of the main yrzI RNAs in rnjA+ versus rnjA‐ cells, we performed primer extension analysis on total RNA isolated from WT, Δrae1, ΔrnjA, and Δrae1 rnjA double mutant strains, using a primer hybridizing to yrzI. In all four strains, we detected a 5′‐end corresponding to the predicted P3 promoter upstream of S1027 (Fig. 5B). The calculated size of the fragment extending from P3 to the terminator following S1024 is 798 nts, corresponding well to the 0.8 kb fragment seen on Northern blots. In Δrae1 cells, a 5′‐end mapping to nt −21 relative to the yrzI start codon was strongly visible compared to WT cells (Fig. 5B). This 5′ end was located precisely 8 nts upstream of the G‐rich motif in the yrzI SD sequence (Fig. 5A,C), identical what was observed for the hbs and cryIIIA ribosome heel‐prints [30, 31]. The predicted size of a RNA extending from this position to the transcription terminator downstream of S1024 (T4) is 521 nts and correlates well with the size of the R‐T4 band seen on Northern blots. In the Δrae1 rnjA double mutant, the 5′ end at nt −21 was no longer visible and was replaced by a species mapping to position −125/126, corresponding to the E1 site (Fig. 4A).

To show that the R‐T4 species results from direct ribosome protection of the 521 nt yrzI RNA from RNase J1 activity, we transcribed a portion of the yrzI gene (−55 to +80 relative to the start codon) using T7 RNA polymerase. The RNA was 3′‐labeled with ³²P‐pCp, pre‐incubated with 30S ribosomal subunits and then subjected to RNase J1 degradation in vitro for different times. A protected species of the expected size accumulated over time in samples containing 30S subunits (Fig. 5D), confirming that the 5′ end of the 521 nt yrzI species is protected from RNase J1 degradation by ribosomes initiating translation of the yrzI ORF.

Altogether, these data indicate that RNase J1 gains access to the yrzI RNAs at the upstream E1 site and degrades until it is blocked by a ribosome bound to the yrzI SD at nt −21, giving rise to the R‐T4 species.

Degradation of the yrzI polycistronic mRNA by 3′‐ 5′ exoribonucleases

The data presented above showed that the highly stable R‐T4 species can be degraded from its 3′ end independently of Rae1. We therefore analyzed the ability of individual 3′‐exoribonucleases to promote R‐T4 degradation by analyzing the profile of yrzI RNAs in strains deleted for one or more 3′‐exoribonucleases (Fig. 6A). The R‐T4 species was present at a higher level in the ∆rae1 ∆4exos strain compared to the ∆rae1 strain alone, suggesting that the two degradation pathways involving either Rae1 or the 3′ exoribonucleases, destabilize R‐T4 independently (Fig. 6A compare lane 2 and 4). The R‐T4 species was not detected in triple mutant ∆yhaM rph rnr that only retains PNPase (Fig. 6A, compare lane 7 with lane 5), but was observed in all strains lacking PNPase, albeit at a lower level than in the ∆4exos strain (Fig. 6A, compare lanes 6–10 with lane 5). Thus, PNPase appears to be the major, but not the only 3‐exoribonuclease involved in the degradation of R‐T4. A band below P1‐T4 (indicated by an asterisk) was also detected in all strains lacking PNPase (Fig. 6A, lanes 5–6 and 8–10). This band may result from exoribonucleolytic degradation mediated by the other 3′‐exoribonucleases that are blocked by secondary structure or translating ribosomes. Furthermore, a species migrating slightly faster than the upstream product of Rae1 cleavage (R‐Rae1) was additionally observed in all strains lacking PNPase, including the triple mutant ∆pnp rph rnr that only retains YhaM (Fig. 6A, lane 6, white dot). This suggests that YhaM catalyzes the initial shortening of R‐Rae1 and that PNPase is the principal enzyme that finishes the job of degrading it. In contrast, the R‐Rae1 species was only detected in the ∆4exos mutants (Fig. 6A, lane 5).

Fig. 6 — Degradation of the *yrzI* polycistronic mRNA by 3′‐ 5′ exoribonucleases. (A) Northern blot, probed with oligo CC1589 (*yrzI*), showing expression of *yrzI* in strains lacking specific *B. subtilis* RNases. Gene abbreviations are as follows: Rae1 (*rae1*), the four 3′‐exoribonucleases PNPase, RNase R, RNase PH and YhaM (*4exos*), PNPase (*pnp*), RNase PH (*rph*), RNase R (*rnr*) and YhaM (*yhaM*). The origin of the band indicated by a black dot blot remains unknown. This experiment was repeated twice. (B) Half‐lives were determined by Northern blot analysis in WT, *Δrae1, Δ4exos* and *Δrae1Δ4exos* strains at mid‐log phase in rich medium at times after rifampicin addition, probed with oligo CC1589 (*yrzI*). Blots in panels A and B were rehybridized with a probe complementary to 16S rRNA (oligo CC058) as a loading control. Note that the measured half‐life of the R‐Rae1 is impacted by the fact that it is being continually generated by processing of P3‐T4 as it is being decayed. This experiment was repeated twice. (C) Degradation pathways of R‐T4 RNA. The primary transcript (P3‐T4) and the degradation intermediates are depicted. Endoribonucleases are symbolized by scissors and exoribonucleases by a Pacman symbol.

Lastly, we examined the relative contribution of the Rae1 cleavage pathway and the 3′‐exoribonuclease pathway to R‐T4 degradation by measuring its stability after addition of rifampicin to the culture medium in the ∆rae1 and ∆4exos mutants (Fig. 6B). The R‐T4 RNA had a half‐life of around 13.5 min in the ∆4exos strain, but >18 min in either the ∆rae1 strain or the ∆rae1 ∆4exos quintuple mutant, suggesting that the greater contribution comes from the Rae1 pathway (Fig. 6B). The fact that the half‐life of the R‐T4 RNA increases in the ∆4exos strain compared to WT, despite the presence of Rae1, confirms that the effect of the deletion of the four exoribonucleases is at the post‐transcriptional level and this 3′‐degradation pathway can indeed attack the 3′ end of the mRNA independently of Rae1 (Fig. 6C).

The degradation of R‐T4 is thus promoted by two independent rate‐limiting degradation pathways one promoted by a translation‐dependent cleavage mediated by Rae1 within S1025 and the other by 3′‐exoribonucleases from the 3′ end of the mRNA. We postulate that translated R‐T4 mRNAs are efficiently degraded by Rae1 but protected by ribosomes from the 3′‐degradation pathway, while untranslated R‐T4 mRNAs, which cannot be cleaved by Rae1, can be degraded by 3′‐exoribonucleases.

Discussion

In this paper, we attribute functions to two of the four peptides encoded by the polycistronic yrzI mRNA from B. subtilis: the short 49‐aa inhibitory peptide, called YrzI, and its antidote, a 17‐aa peptide called S1025. We propose to rename YrzI and S1025 as SpyT and SpyA for Small peptide YrzI Toxin and Small peptide YrzI Antitoxin, respectively. We also describe a second naturally occurring case in B. subtilis of a ribosome bound to an exceptionally strong SD sequence protecting an mRNA fragment from 5′‐degradation by RNase J1 (the other example being hbs) [29, 30, 31]. We further describe the non‐canonical degradation pathways required to eliminate this species involving the translation‐dependent endoribonuclease Rae1, on the one hand, and direct attack from the 3′‐end by exoribonucleases on the other.

The SpyT toxin peptide is highly conserved among the genus Bacillus (Fig. S2A,B). A sequence encoding a peptide highly homologous to the SpyA antitoxin was identified downstream of the spyT ORF in all analyzed Bacillus genomes, except for B. cereus (Fig. S2A,B). In B. cereus, SpyT peptides that exhibit lower homology to SpyT from B. subtilis are present in seven contiguous copies, but the ORF located immediately downstream does not share significant homology with the B. subtilis SpyA peptide (Fig. S2C). Although the spyTA operon is expressed most of the time in B. subtilis, pointing to a general role in the cell's physiology, it shows peaks of expression in the cold, in stationary phase and late in sporulation [22], suggesting it may have specific functions under these conditions.

There are some intriguing parallels between the SpyT/SpyA/Rae1 triad and classical toxin‐antitoxin (TA) systems. As currently observed in TA modules, the SpyT toxin and SpyA antitoxin are encoded within the same operon. TA modules have been divided into seven classes according to the mode of action of the antitoxin [32, 33, 34]. We had previously identified a type II TA system in B. subtilis: the toxin EndoA (an RNase member of the MazF/PemK family of bacterial toxins) and the YdcD antitoxin protein encoded by the gene immediately upstream [35]. Our data suggest that SpyT/SpyA/Rae1 triad could potentially constitute a toxin/double‐antitoxin system, with components resembling both the type II (protein antitoxin) and type V (RNase antitoxin) TA systems. On one hand, protection by SpyA against the toxic effect of SpyT was observed even in the absence of Rae1 indicating that the SpyA peptide is the major antidote to the SpyT toxin. On the other hand, Rae1 could easily counteract SpyT toxicity by promoting the degradation of the toxin‐antitoxin mRNA, if SpyA peptide levels were insufficient to fully neutralize SpyT. Clearly, Rae1 does not serve as an antitoxin in cells growing in rich medium as a ∆rae1 strain has no major phenotype under these conditions. However, we cannot exclude that there may be certain growth conditions where expression of SpyT and SpyA become uncoupled and SpyA levels are insufficient to do the job alone. In this case, the double protection afforded by Rae1 could be beneficial. A parallel can be drawn with the RatA/TxpA and YonT/as YonT type I TA systems of B. subtilis where the toxin‐encoding mRNAs are degraded by RNase III [36]. However, in those cases inactivation of RNase III in rich medium is lethal.

Our data indicate that the spyTA and bmrBCD operons are both governed at the transcriptional level by AbrB and at the post‐transcriptional level by Rae1 [12], suggesting there may be some as yet uncovered link between the two. Post‐transcriptional modulation of spyT expression first involves stabilization of the R‐T4 fragment through cleavage by an endoribonuclease at the E1 site followed by 5′‐trimming by RNase J1 until it is blocked by a ribosome bound at the SD sequence of spyT. The stabilized mRNA is then principally degraded through translation‐dependent cleavage by Rae1. About 200 genes are predicted to have exceptionally strong SD sequences less than 20 nts upstream of their ORFs in B. subtilis, with maximum 2 mismatches out of 12 with the 3′‐end of 16S rRNA equivalent to the experimentally confirmed hbs gene. Thus, at a conservative estimate at least 5% of B. subtilis genes are predicted to benefit from 5′‐protection by ribosomes initiating translation at their SD sequences (Table S3). Note, however, that the original stabilizing motif in the B. thuringiensis cryIIIA mRNA (Stab SD) expressed in B. subtilis had only 8/12 consecutive base pairs [37] and so the actual number of mRNAs falling into this category may be much higher.

The spyTA polycistronic mRNA is targeted by several RNases: two endoribonucleolytic sites (E1 and E2) were mapped in addition to the Rae1 cleavage site. Our data suggest that RNase Y may cleave at E1, but further investigation is needed to determine whether other RNases also participate and to identify the RNase responsible for E2 cleavage. Remarkably, 3′‐exoribonucleases can also participate in the degradation of the mature R‐T4 RNA, despite the presence of the hairpin structure of the transcriptional terminator T4 at its 3′ end. Although Rho‐independent transcription terminators are often described as providing resistance against 3′ degradation by exoribonucleases [5, 23, 24, 25, 26], the phenomenon has been understudied in bacteria and is likely to be dependent on the intrinsic stability of the RNA hairpins and/or additional factors. Indeed, the predicted stability of the transcription terminator at the end of the spyTA operon is 11.1 kcal·mol⁻¹, well below the average stability of experimentally determined Rho‐independent terminators in B. subtilis of 16 kcal·mol⁻¹ [38]. This may explain why the 3′‐degradation pathway comes into play for this particular mRNA that is protected by ribosomes at the 5′ end from the RNase J1. Although the R‐T4 RNA was strongly stabilized in the ∆4exos mutant, it showed no signs of decay over the time‐course of the experiment in strains lacking Rae1. This unusual scenario, in which two rate‐limiting degradation pathways target the same mRNA, can be explained by hypothesizing that translated spyTA operon mRNAs are efficiently degraded by Rae1, while untranslated spyTA mRNAs, which cannot be cleaved by Rae1, are eliminated by 3′‐exoribonucleases.

This study of the spyTA polycistronic mRNA turnover highlights the complex interplay between ribosomes and the mRNA degradation machinery: degradation of the most abundant mRNA R‐T4 species, generated by ribosomes obstructing 5′‐exoribonucleolytic degradation, is influenced by the translational status of the mRNA for Rae1‐dependent cleavage and likely also for the 3′‐exoribonuclease pathway.

Author contributions

LG and ML contributed equally to the execution of the experiments. AM performed the genomic analyses to predict exceptionally strong SD sequences. CC helped supervise the project and wrote the manuscript. FB performed the experiments, supervised the project and wrote the manuscript.

Peer review

The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer‐review/10.1002/1873‐3468.70027.

Supporting information

Fig. S1. Validation of presence of T4 terminator at 3′ end of yrzI transcripts.

Fig. S2. Conservation of yrzI operon in the genus Bacillus.

FEB2-599-1222-s003.pdf^{(568.1KB, pdf)}

Table S1. Oligonucleotides used in this study.

Table S2. Bacillus subtilis strains used in this study.

FEB2-599-1222-s001.docx^{(21.2KB, docx)}

Table S3. Predicted exceptionally strong SD sequences in B. subtilis.

FEB2-599-1222-s002.xlsx^{(30.2KB, xlsx)}

Acknowledgements

We thank lab members for helpful discussion. This work was supported by funds from the CNRS and Université Paris Cité (UMR8261), the Agence Nationale de la Recherche (ARNr‐BasRae1), Labex Dynamo and Equipex Cacsice).

Laetitia Gilet and Magali Leroy contributed equally and are listed alphabetically

Edited by Christian Griesinger

This is an Editor's Choice from the 12 May 2025 issue

Contributor Information

Ciarán Condon, Email: condon@ibpc.fr.

Frédérique Braun, Email: braun@ibpc.fr.

Data availability

The data that support the findings of this study are available from the corresponding author on request.

References

1. Bechhofer DH and Deutscher MP (2019) Bacterial ribonucleases and their roles in RNA metabolism. Crit Rev Biochem Mol Biol 54, 242–300. [DOI] [PMC free article] [PubMed] [Google Scholar]
2. Trinquier A, Durand S, Braun F and Condon C (2020) Regulation of RNA processing and degradation in bacteria. Biochim Biophys Acta Gene Regul Mech 1863, 194505. [DOI] [PubMed] [Google Scholar]
3. Richards J, Liu Q, Pellegrini O, Celesnik H, Yao S, Bechhofer DH, Condon C and Belasco JG (2011) An RNA pyrophosphohydrolase triggers 5′‐exonucleolytic degradation of mRNA in Bacillus subtilis . Mol Cell 43, 940–949. [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Durand S, Gilet L, Bessieres P, Nicolas P and Condon C (2012) Three essential ribonucleases‐RNase Y, J1, and III‐control the abundance of a majority of Bacillus subtilis mRNAs. PLoS Genet 8, e1002520. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Braun F, Le Derout J and Regnier P (1998) Ribosomes inhibit an RNase E cleavage which induces the decay of the rpsO mRNA of Escherichia coli . EMBO J 17, 4790–4797. [DOI] [PMC free article] [PubMed] [Google Scholar]
6. Deana A and Belasco JG (2005) Lost in translation: the influence of ribosomes on bacterial mRNA decay. Genes Dev 19, 2526–2533. [DOI] [PubMed] [Google Scholar]
7. Dreyfus M (2009) Killer and protective ribosomes. Prog Mol Biol Transl Sci 85, 423–466. doi: 10.1016/S0079-6603(08)00811-8 [DOI] [PubMed] [Google Scholar]
8. Duviau MP, Chen F, Emile A, Cocaign‐Bousquet M, Girbal L and Nouaille S (2023) When translation elongation is impaired, the mRNA is uniformly destabilized by the RNA degradosome, while the concentration of mRNA is altered along the molecule. Nucleic Acids Res 51, 2877–2890. [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Iost I and Dreyfus M (1995) The stability of Escherichia coli lacZ mRNA depends upon the simultaneity of its synthesis and translation. EMBO J 14, 3252–3261. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Leroy M, Piton J, Gilet L, Pellegrini O, Proux C, Coppee JY, Figaro S and Condon C (2017) Rae1/YacP, a new endoribonuclease involved in ribosome‐dependent mRNA decay in Bacillus subtilis . EMBO J 36, 1167–1181. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Condon C, Piton J and Braun F (2018) Distribution of the ribosome associated endonuclease Rae1 and the potential role of conserved amino acids in codon recognition. RNA Biol 15, 1–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
12. Deves V, Trinquier A, Gilet L, Alharake J, Condon C and Braun F (2023) Shutdown of multidrug transporter bmrCD mRNA expression mediated by the ribosome‐associated endoribonuclease (Rae1) cleavage in a new cryptic ORF. RNA 29, 1108–1116. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Li GW, Oh E and Weissman JS (2012) The anti‐Shine‐Dalgarno sequence drives translational pausing and codon choice in bacteria. Nature 484, 538–541. [DOI] [PMC free article] [PubMed] [Google Scholar]
14. Oussenko IA, Abe T, Ujiie H, Muto A and Bechhofer DH (2005) Participation of 3′‐to‐5′ exoribonucleases in the turnover of Bacillus subtilis mRNA. J Bacteriol 187, 2758–2767. [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Bechhofer DH, Oussenko IA, Deikus G, Yao S, Mathy N and Condon C (2008) Analysis of mRNA decay in Bacillus subtilis . Methods Enzymol 447, 259–276. [DOI] [PubMed] [Google Scholar]
16. Stead MB, Agrawal A, Bowden KE, Nasir R, Mohanty BK, Meagher RB and Kushner SR (2012) RNAsnap: a rapid, quantitative and inexpensive, method for isolating total RNA from bacteria. Nucleic Acids Res 40, e156. [DOI] [PMC free article] [PubMed] [Google Scholar]
17. Britton RA, Wen T, Schaefer L, Pellegrini O, Uicker WC, Mathy N, Tobin C, Daou R, Szyk J and Condon C (2007) Maturation of the 5′ end of Bacillus subtilis 16S rRNA by the essential ribonuclease YkqC/RNase J1. Mol Microbiol 63, 127–138. [DOI] [PubMed] [Google Scholar]
18. Jurenas D, Fraikin N, Goormaghtigh F and Van Melderen L (2022) Biology and evolution of bacterial toxin‐antitoxin systems. Nat Rev Microbiol 20, 335–350. [DOI] [PubMed] [Google Scholar]
19. Reilman E, Mars RA, van Dijl JM and Denham EL (2014) The multidrug ABC transporter BmrC/BmrD of Bacillus subtilis is regulated via a ribosome‐mediated transcriptional attenuation mechanism. Nucleic Acids Res 42, 11393–11407. [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Chumsakul O, Takahashi H, Oshima T, Hishimoto T, Kanaya S, Ogasawara N and Ishikawa S (2011) Genome‐wide binding profiles of the Bacillus subtilis transition state regulator AbrB and its homolog Abh reveals their interactive role in transcriptional regulation. Nucleic Acids Res 39, 414–428. [DOI] [PMC free article] [PubMed] [Google Scholar]
21. Chumsakul O, Nakamura K, Kurata T, Sakamoto T, Hobman JL, Ogasawara N, Oshima T and Ishikawa S (2013) High‐resolution mapping of in vivo genomic transcription factor binding sites using in situ DNase I footprinting and ChIP‐seq. DNA Res 20, 325–338. [DOI] [PMC free article] [PubMed] [Google Scholar]
22. Nicolas P, Mader U, Dervyn E, Rochat T, Leduc A, Pigeonneau N, Bidnenko E, Marchadier E, Hoebeke M, Aymerich S et al. (2012) Condition‐dependent transcriptome reveals high‐level regulatory architecture in Bacillus subtilis . Science 335, 1103–1106. [DOI] [PubMed] [Google Scholar]
23. Deikus G and Bechhofer DH (2007) Initiation of decay of Bacillus subtilis trp leader RNA. J Biol Chem 282, 20238–20244. [DOI] [PubMed] [Google Scholar]
24. Farr GA, Oussenko IA and Bechhofer DH (1999) Protection against 3′‐to‐5′ RNA decay in Bacillus subtilis . J Bacteriol 181, 7323–7330. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. Lecrivain AL, Le Rhun A, Renault TT, Ahmed‐Begrich R, Hahnke K and Charpentier E (2018) In vivo 3′‐to‐5′ exoribonuclease targetomes of streptococcus pyogenes. Proc Natl Acad Sci USA 115, 11814–11819. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Spickler C and Mackie GA (2000) Action of RNase II and polynucleotide phosphorylase against RNAs containing stem‐loops of defined structure. J Bacteriol 182, 2422–2427. [DOI] [PMC free article] [PubMed] [Google Scholar]
27. Taggart JC, Dierksheide J, Leblanc H, Lalanne JB, Durand S, Braun F, Condon C and Li GW (2024) A high‐resolution view of RNA endonuclease cleavage in Bacillus subtilis . bioRxiv 10.1101/2023.03.12.532304 [PREPRINT] [DOI] [PMC free article] [PubMed]
28. Agaisse H and Lereclus D (1996) STAB‐SD: a Shine‐Dalgarno sequence in the 5′ untranslated region is a determinant of mRNA stability. Mol Microbiol 20, 633–643. [DOI] [PubMed] [Google Scholar]
29. Braun F, Durand S and Condon C (2017) Initiating ribosomes and a 5′/3'‐UTR interaction control ribonuclease action to tightly couple B. Subtilis hbs mRNA stability with translation. Nucleic Acids Res 45, 11386–11400. [DOI] [PMC free article] [PubMed] [Google Scholar]
30. Daou‐Chabo R, Mathy N, Benard L and Condon C (2009) Ribosomes initiating translation of the hbs mRNA protect it from 5′‐to‐3′ exoribonucleolytic degradation by RNase J1. Mol Microbiol 71, 1538–1550. [DOI] [PubMed] [Google Scholar]
31. Mathy N, Benard L, Pellegrini O, Daou R, Wen T and Condon C (2007) 5′‐to‐3′ exoribonuclease activity in bacteria: role of RNase J1 in rRNA maturation and 5′ stability of mRNA. Cell 129, 681–692. [DOI] [PubMed] [Google Scholar]
32. Aakre CD, Phung TN, Huang D and Laub MT (2013) A bacterial toxin inhibits DNA replication elongation through a direct interaction with the beta sliding clamp. Mol Cell 52, 617–628. [DOI] [PMC free article] [PubMed] [Google Scholar]
33. Goeders N and Van Melderen L (2014) Toxin‐antitoxin systems as multilevel interaction systems. Toxins 6, 304–324. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Wang X, Yao J, Sun YC and Wood TK (2021) Type VII toxin/antitoxin classification system for antitoxins that enzymatically neutralize toxins. Trends Microbiol 29, 388–393. [DOI] [PubMed] [Google Scholar]
35. Pellegrini O, Mathy N, Gogos A, Shapiro L and Condon C (2005) The Bacillus subtilis ydcDE operon encodes an endoribonuclease of the MazF/PemK family and its inhibitor. Mol Microbiol 56, 1139–1148. [DOI] [PubMed] [Google Scholar]
36. Durand S, Gilet L and Condon C (2012) The essential function of B. subtilis RNase III is to silence foreign toxin genes. PLoS Genet 8, e1003181. [DOI] [PMC free article] [PubMed] [Google Scholar]
37. Agaisse H and Lereclus D (1994) Structural and functional analysis of the promoter region involved in full expression of the cryIIIA toxin gene of Bacillus thuringiensis . Mol Microbiol 13, 97–107. [DOI] [PubMed] [Google Scholar]
38. de Hoon MJ, Makita Y, Nakai K and Miyano S (2005) Prediction of transcriptional terminators in Bacillus subtilis and related species. PLoS Comput Biol 1, e25. [DOI] [PMC free article] [PubMed] [Google Scholar]
39. Karp PD, Billington R, Caspi R, Fulcher CA, Latendresse M, Kothari A, Keseler IM, Krummenacker M, Midford PE, Ong Q et al. (2019) The BioCyc collection of microbial genomes and metabolic pathways. Brief Bioinform 20, 1085–1093. [DOI] [PMC free article] [PubMed] [Google Scholar]
40. Sievers F and Higgins DG (2018) Clustal omega for making accurate alignments of many protein sequences. Protein Sci 27, 135–145. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Fig. S1. Validation of presence of T4 terminator at 3′ end of yrzI transcripts.

Fig. S2. Conservation of yrzI operon in the genus Bacillus.

FEB2-599-1222-s003.pdf^{(568.1KB, pdf)}

Table S1. Oligonucleotides used in this study.

Table S2. Bacillus subtilis strains used in this study.

FEB2-599-1222-s001.docx^{(21.2KB, docx)}

Table S3. Predicted exceptionally strong SD sequences in B. subtilis.

FEB2-599-1222-s002.xlsx^{(30.2KB, xlsx)}

Data Availability Statement

The data that support the findings of this study are available from the corresponding author on request.

[feb270027-bib-0001] 1. Bechhofer DH and Deutscher MP (2019) Bacterial ribonucleases and their roles in RNA metabolism. Crit Rev Biochem Mol Biol 54, 242–300. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0002] 2. Trinquier A, Durand S, Braun F and Condon C (2020) Regulation of RNA processing and degradation in bacteria. Biochim Biophys Acta Gene Regul Mech 1863, 194505. [DOI] [PubMed] [Google Scholar]

[feb270027-bib-0003] 3. Richards J, Liu Q, Pellegrini O, Celesnik H, Yao S, Bechhofer DH, Condon C and Belasco JG (2011) An RNA pyrophosphohydrolase triggers 5′‐exonucleolytic degradation of mRNA in Bacillus subtilis . Mol Cell 43, 940–949. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0004] 4. Durand S, Gilet L, Bessieres P, Nicolas P and Condon C (2012) Three essential ribonucleases‐RNase Y, J1, and III‐control the abundance of a majority of Bacillus subtilis mRNAs. PLoS Genet 8, e1002520. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0005] 5. Braun F, Le Derout J and Regnier P (1998) Ribosomes inhibit an RNase E cleavage which induces the decay of the rpsO mRNA of Escherichia coli . EMBO J 17, 4790–4797. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0006] 6. Deana A and Belasco JG (2005) Lost in translation: the influence of ribosomes on bacterial mRNA decay. Genes Dev 19, 2526–2533. [DOI] [PubMed] [Google Scholar]

[feb270027-bib-0007] 7. Dreyfus M (2009) Killer and protective ribosomes. Prog Mol Biol Transl Sci 85, 423–466. doi: 10.1016/S0079-6603(08)00811-8 [DOI] [PubMed] [Google Scholar]

[feb270027-bib-0008] 8. Duviau MP, Chen F, Emile A, Cocaign‐Bousquet M, Girbal L and Nouaille S (2023) When translation elongation is impaired, the mRNA is uniformly destabilized by the RNA degradosome, while the concentration of mRNA is altered along the molecule. Nucleic Acids Res 51, 2877–2890. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0009] 9. Iost I and Dreyfus M (1995) The stability of Escherichia coli lacZ mRNA depends upon the simultaneity of its synthesis and translation. EMBO J 14, 3252–3261. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0010] 10. Leroy M, Piton J, Gilet L, Pellegrini O, Proux C, Coppee JY, Figaro S and Condon C (2017) Rae1/YacP, a new endoribonuclease involved in ribosome‐dependent mRNA decay in Bacillus subtilis . EMBO J 36, 1167–1181. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0011] 11. Condon C, Piton J and Braun F (2018) Distribution of the ribosome associated endonuclease Rae1 and the potential role of conserved amino acids in codon recognition. RNA Biol 15, 1–6. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0012] 12. Deves V, Trinquier A, Gilet L, Alharake J, Condon C and Braun F (2023) Shutdown of multidrug transporter bmrCD mRNA expression mediated by the ribosome‐associated endoribonuclease (Rae1) cleavage in a new cryptic ORF. RNA 29, 1108–1116. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0013] 13. Li GW, Oh E and Weissman JS (2012) The anti‐Shine‐Dalgarno sequence drives translational pausing and codon choice in bacteria. Nature 484, 538–541. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0014] 14. Oussenko IA, Abe T, Ujiie H, Muto A and Bechhofer DH (2005) Participation of 3′‐to‐5′ exoribonucleases in the turnover of Bacillus subtilis mRNA. J Bacteriol 187, 2758–2767. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0015] 15. Bechhofer DH, Oussenko IA, Deikus G, Yao S, Mathy N and Condon C (2008) Analysis of mRNA decay in Bacillus subtilis . Methods Enzymol 447, 259–276. [DOI] [PubMed] [Google Scholar]

[feb270027-bib-0016] 16. Stead MB, Agrawal A, Bowden KE, Nasir R, Mohanty BK, Meagher RB and Kushner SR (2012) RNAsnap: a rapid, quantitative and inexpensive, method for isolating total RNA from bacteria. Nucleic Acids Res 40, e156. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0017] 17. Britton RA, Wen T, Schaefer L, Pellegrini O, Uicker WC, Mathy N, Tobin C, Daou R, Szyk J and Condon C (2007) Maturation of the 5′ end of Bacillus subtilis 16S rRNA by the essential ribonuclease YkqC/RNase J1. Mol Microbiol 63, 127–138. [DOI] [PubMed] [Google Scholar]

[feb270027-bib-0018] 18. Jurenas D, Fraikin N, Goormaghtigh F and Van Melderen L (2022) Biology and evolution of bacterial toxin‐antitoxin systems. Nat Rev Microbiol 20, 335–350. [DOI] [PubMed] [Google Scholar]

[feb270027-bib-0019] 19. Reilman E, Mars RA, van Dijl JM and Denham EL (2014) The multidrug ABC transporter BmrC/BmrD of Bacillus subtilis is regulated via a ribosome‐mediated transcriptional attenuation mechanism. Nucleic Acids Res 42, 11393–11407. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0020] 20. Chumsakul O, Takahashi H, Oshima T, Hishimoto T, Kanaya S, Ogasawara N and Ishikawa S (2011) Genome‐wide binding profiles of the Bacillus subtilis transition state regulator AbrB and its homolog Abh reveals their interactive role in transcriptional regulation. Nucleic Acids Res 39, 414–428. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0021] 21. Chumsakul O, Nakamura K, Kurata T, Sakamoto T, Hobman JL, Ogasawara N, Oshima T and Ishikawa S (2013) High‐resolution mapping of in vivo genomic transcription factor binding sites using in situ DNase I footprinting and ChIP‐seq. DNA Res 20, 325–338. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0022] 22. Nicolas P, Mader U, Dervyn E, Rochat T, Leduc A, Pigeonneau N, Bidnenko E, Marchadier E, Hoebeke M, Aymerich S et al. (2012) Condition‐dependent transcriptome reveals high‐level regulatory architecture in Bacillus subtilis . Science 335, 1103–1106. [DOI] [PubMed] [Google Scholar]

[feb270027-bib-0023] 23. Deikus G and Bechhofer DH (2007) Initiation of decay of Bacillus subtilis trp leader RNA. J Biol Chem 282, 20238–20244. [DOI] [PubMed] [Google Scholar]

[feb270027-bib-0024] 24. Farr GA, Oussenko IA and Bechhofer DH (1999) Protection against 3′‐to‐5′ RNA decay in Bacillus subtilis . J Bacteriol 181, 7323–7330. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0025] 25. Lecrivain AL, Le Rhun A, Renault TT, Ahmed‐Begrich R, Hahnke K and Charpentier E (2018) In vivo 3′‐to‐5′ exoribonuclease targetomes of streptococcus pyogenes. Proc Natl Acad Sci USA 115, 11814–11819. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0026] 26. Spickler C and Mackie GA (2000) Action of RNase II and polynucleotide phosphorylase against RNAs containing stem‐loops of defined structure. J Bacteriol 182, 2422–2427. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0027] 27. Taggart JC, Dierksheide J, Leblanc H, Lalanne JB, Durand S, Braun F, Condon C and Li GW (2024) A high‐resolution view of RNA endonuclease cleavage in Bacillus subtilis . bioRxiv 10.1101/2023.03.12.532304 [PREPRINT] [DOI] [PMC free article] [PubMed]

[feb270027-bib-0028] 28. Agaisse H and Lereclus D (1996) STAB‐SD: a Shine‐Dalgarno sequence in the 5′ untranslated region is a determinant of mRNA stability. Mol Microbiol 20, 633–643. [DOI] [PubMed] [Google Scholar]

[feb270027-bib-0029] 29. Braun F, Durand S and Condon C (2017) Initiating ribosomes and a 5′/3'‐UTR interaction control ribonuclease action to tightly couple B. Subtilis hbs mRNA stability with translation. Nucleic Acids Res 45, 11386–11400. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0030] 30. Daou‐Chabo R, Mathy N, Benard L and Condon C (2009) Ribosomes initiating translation of the hbs mRNA protect it from 5′‐to‐3′ exoribonucleolytic degradation by RNase J1. Mol Microbiol 71, 1538–1550. [DOI] [PubMed] [Google Scholar]

[feb270027-bib-0031] 31. Mathy N, Benard L, Pellegrini O, Daou R, Wen T and Condon C (2007) 5′‐to‐3′ exoribonuclease activity in bacteria: role of RNase J1 in rRNA maturation and 5′ stability of mRNA. Cell 129, 681–692. [DOI] [PubMed] [Google Scholar]

[feb270027-bib-0032] 32. Aakre CD, Phung TN, Huang D and Laub MT (2013) A bacterial toxin inhibits DNA replication elongation through a direct interaction with the beta sliding clamp. Mol Cell 52, 617–628. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0033] 33. Goeders N and Van Melderen L (2014) Toxin‐antitoxin systems as multilevel interaction systems. Toxins 6, 304–324. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0034] 34. Wang X, Yao J, Sun YC and Wood TK (2021) Type VII toxin/antitoxin classification system for antitoxins that enzymatically neutralize toxins. Trends Microbiol 29, 388–393. [DOI] [PubMed] [Google Scholar]

[feb270027-bib-0035] 35. Pellegrini O, Mathy N, Gogos A, Shapiro L and Condon C (2005) The Bacillus subtilis ydcDE operon encodes an endoribonuclease of the MazF/PemK family and its inhibitor. Mol Microbiol 56, 1139–1148. [DOI] [PubMed] [Google Scholar]

[feb270027-bib-0036] 36. Durand S, Gilet L and Condon C (2012) The essential function of B. subtilis RNase III is to silence foreign toxin genes. PLoS Genet 8, e1003181. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0037] 37. Agaisse H and Lereclus D (1994) Structural and functional analysis of the promoter region involved in full expression of the cryIIIA toxin gene of Bacillus thuringiensis . Mol Microbiol 13, 97–107. [DOI] [PubMed] [Google Scholar]

[feb270027-bib-0038] 38. de Hoon MJ, Makita Y, Nakai K and Miyano S (2005) Prediction of transcriptional terminators in Bacillus subtilis and related species. PLoS Comput Biol 1, e25. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0039] 39. Karp PD, Billington R, Caspi R, Fulcher CA, Latendresse M, Kothari A, Keseler IM, Krummenacker M, Midford PE, Ong Q et al. (2019) The BioCyc collection of microbial genomes and metabolic pathways. Brief Bioinform 20, 1085–1093. [DOI] [PMC free article] [PubMed] [Google Scholar]

[feb270027-bib-0040] 40. Sievers F and Higgins DG (2018) Clustal omega for making accurate alignments of many protein sequences. Protein Sci 27, 135–145. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Unconventional mRNA processing and degradation pathways for the polycistronic yrzI (spyTA) mRNA in Bacillus subtilis

Laetitia Gilet

Magali Leroy

Alexandre Maes

Ciarán Condon

Frédérique Braun

Abstract

Abbreviations

Materials and methods

Strain construction

Northern blots and primer extension assays

Spot dilution assays

Ribosomal subunit (30S) protection assay

Results

YrzI encodes a toxic peptide that is counteracted by S1025

Fig. 1.

The polycistronic yrzI mRNA is regulated by AbrB at the transcriptional level

Fig. 2.

Mapping of the major RNAs produced from the yrzI locus

Fig. 3.

The R‐T4 species results from 5′ degradation by RNase J1 and is degraded via two alternative pathways

Fig. 4.

The R‐T4 species is protected from RNase J1 degradation by ribosomes initiating translation of yrzI

Fig. 5.

Degradation of the yrzI polycistronic mRNA by 3′‐ 5′ exoribonucleases

Fig. 6.

Discussion

Author contributions

Peer review

Supporting information

Acknowledgements

Contributor Information

Data availability

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases