Post-transcriptional regulation is a rapid and efficient way to modulate gene expression used by all organisms. In bacteria, this regulation can involve cis-acting regulatory elements such as riboswitches, which bind small molecules and modify transcription elongation, translation or mRNA stability. It can equally involve trans-acting factors such as proteins (RNases or RNA binding proteins) and small regulatory RNAs (sRNAs). sRNAs can be encoded in cis (antisense RNA) or in trans relative to their target mRNAs and these regulatory mechanisms can be superimposed on each other to create complex control networks.

The two best-studied model bacteria Escherichia coli and Bacillus subtilis have evolved with different arsenals of ribonucleases and alternative approaches to the fundamental cellular processes of RNA maturation and degradation (Bechhofer, 2009; Durand et al., 2015b; Condon and Putzer, 2002). Only eight of the >30 ribonucleases currently identified in these two bacteria are present in both organisms. E. coli depends primarily on endonucleolytic cleavage by RNase E to initiate RNA turnover, followed by degradation of the resulting fragments in a 3’-to-5’ orientation by exoribonucleases. B. subtilis, on the other hand, has replaced RNase E by an enzyme with similar properties called RNase Y and has the additional option of exonucleolytically degrading RNAs from the 5’ end using an enzyme called RNase J1 (Mathy et al., 2007). In this regard, B. subtilis shares features with the eukaryotic model of RNA decay and this observation has obscured some of the traditional divisions between « prokaryotic » and « eukaryotic » pathways of RNA degradation.

Small regulatory RNAs have been mostly studied in Gram-negative bacteria, especially in E. coli and its pathogenic relatives. In these organisms, sRNAs generally act by an imperfect base pairing with their target mRNAs. The interaction between sRNA and mRNA often requires the chaperone Hfq, an Sm-like protein that both stabilizes the sRNA and facilitates the pairing with mRNA targets (for review, see Updegrove et al., 2016). The sRNA-mRNA interaction can lead to a positive or a negative effect on gene expression. In bacteria, most sRNAs examined to date directly affect mRNA translation with an indirect effect on mRNA half-life (reviewed in Wagner and Romby, 2015). However, some more recent studies have shown that sRNAs are also able to directly affect mRNA stability without affecting translation, by recruiting the degradation machinery or interfering with its action (reviewed in Lalaouna et al., 2013). In one particular example, the ternary complex formed between the sRNA (MicC), the mRNA (ompD) and Hfq can recruit RNase E to a site within the open reading frame and stimulate its cleavage by presenting the 5’ monophosphate of the sRNA to RNase E (Bandyra et al., 2012). In another, the sRNA RydC increases the stability of the cfa mRNA by pairing to the 5’-UTR of this mRNA, far upstream of the ribosome binding site, and hiding a cleavage site from RNase E (Fröhlich et al., 2013).

Small RNAs have also been studied in some Gram-positive pathogenic bacteria such as Staphylococcus aureus and Streptococcus pyogenes, but in B. subtilis, the paradigm of Gram-positive bacteria, only a handful of sRNAs have been characterized to date and the global mechanisms of sRNA regulation have not been fully resolved (Durand et al., 2015b). In these bacteria, Hfq is generally not required for sRNA-mRNA pairing (Hämmerle et al., 2014; Rochat et al., 2015) although there are some exceptions in specific organisms (Nielsen et al., 2010; Boudry et al., 2014). One key question is to what extent RNases J1 and Y are influenced by sRNA binding to mRNA targets in B. subtilis, i.e. can E. coli models involving RNase E be transposed to these RNases in its Gram-positive counterpart?

RoxS (related to oxidative stress) is the unique trans-acting sRNA, apart from the ubiquitous 6S RNA, that is conserved between B. subtilis and the Gram-positive pathogen S. aureus. First identified in S. aureus where it is named RsaE, it was shown to regulate genes involved in amino acid and peptide transport, cofactor synthesis, lipid metabolism, carbohydrate metabolism and the TCA cycle (Geissmann et al., 2009). In a previous study, we showed that RoxS transcription is induced during nitric oxide (NO) stress by the two-component system ResDE and regulates the expression of numerous genes linked to oxido-reduction processes in B. subtilis (Durand et al., 2015a). In searching for additional direct targets of RoxS that may not have been expressed under the experimental conditions of the first study, we found that RoxS potentially interacts with the extreme 5’-end of the yflS mRNA, encoding a malate transporter in B. subtilis. We show here that RoxS expression is greatly increased in the presence of malate in the medium and acts as a positive regulator of transporter expression by directly blocking yflS mRNA degradation by the 5’−3’ exoribonuclease J1. This is the first known case of an sRNA controlling 5'-exoribonuclease activity in bacteria. RoxS can further stimulate translation of the stabilized yflS mRNA, possibly by opening a specific fold of the mRNA that partially occludes the ribosome binding site, providing an additional layer of complexity to the control mechanism. Lastly, we show that RoxS expression is controlled by Rex, a transcriptional repressor of genes linked to fermentation processes that senses the NAD⁺/NADH balance in the cell. The increase in RoxS expression in the presence of malate is explained by a release from Rex-mediated repression, allowing a far more efficient regulation of RoxS targets than seen previously.

In a previous study we showed that RoxS, either directly or indirectly, negatively regulates the expression of up to a hundred genes, including two mRNAs encoding proteins involved in central carbon metabolism: ppnkB (encoding an inorganic polyphosphate/ATP-NAD kinase) and sucCD (encoding succinate dehydrogenase) (Durand et al., 2015a). In both cases, RoxS binds to the Shine-Dalgarno (SD) sequences and inhibits translation initiation. To expand the number of direct RoxS targets, we searched for new potential binding sites for this sRNA (TargetRNA2; Kery et al., 2014) that we might have missed in the first study because of the specific growth conditions used (rich medium at 37°C). One such predicted target, the yflS mRNA (encoding one of at least four malate transporters in B. subtilis) drew our attention because RoxS was predicted to bind to the extreme 5’-end of the mRNA well upstream of the ribosome binding site (Figure 1).

Figure 1 with 2 supplements see all

Download asset Open asset

RoxS regulates yflS mRNA stability

The yflS gene encodes a malate transporter whose expression is induced by the two-component system (MalKR), which senses the presence of malate in the medium (Tanaka et al., 2003). To determine whether RoxS had an effect on yflS expression, we compared the half-life of this mRNA in WT and ∆roxS strains grown in 2xTY containing 0.5% malate (Figure 2A, Figure 2—figure supplement 1). We detected a single band of approximately 1.5 kb corresponding to the predicted size of the full-length yflS mRNA on Northern blots of total RNA isolated at different times after addition of rifampicin to stop new transcription. The half-life of the yflS mRNA was 2.5 min in a WT strain but decreased to 0.7 min in the absence of RoxS. Thus, RoxS stabilizes the yflS mRNA about 3.6-fold, presumably by protecting it from degradation by cellular RNases.

Figure 2 with 1 supplement see all

Download asset Open asset

RoxS protects yflS mRNA from in vitro degradation by RNase J1

The results described above suggested that RoxS protects the yflS full-length mRNA from attack by RNase J1 in vivo. To confirm this observation in vitro, we transcribed a fragment corresponding to the first 94 nucleotides of the yflS mRNA using T7 RNA polymerase and submitted this transcript to degradation by RNase J1 in the presence or absence of the RoxS sRNA. The RNA contained a ³²P-labelled 5'-monophosphate group to mimic the in vivo situation following RNA pyrophosphohydrolase activity. The yflS RNA fragment was attacked by RNase J1 in 5'-exoribonuclease mode as indicated by the accumulation of free radio-labelled AMP at the bottom of the gel, beginning after only 2 min of incubation with the enzyme (Figure 3, lanes 2–5). In contrast, almost no AMP was visible when the reaction was performed in the presence of RoxS, showing clearly that the 5' end of yflS is protected from RNase J1 by the sRNA (Figure 3, lanes 8–11).

Figure 3

Download asset Open asset

Figure 3—source data 1

Source data for the quantification of yflS degradation by RNase J1 in vitro.

https://doi.org/10.7554/eLife.23602.009

Download elife-23602-fig3-data1-v2.xlsx

RoxS uses C-rich region 3 (CRR3) to interact with the extreme 5’-end of yflS mRNA

RoxS has four C-rich regions (CRRs) that can potentially base-pair with G-rich complementary sequences in its targets (Figure 1). In our previous work, we showed that RoxS mainly uses CRR3 to regulate the expression of the ppnkB mRNA (Durand et al., 2015a). To determine which CRR was most important for the interaction with yflS, we repeated the RNase J1 protection experiment using three different mutants of RoxS: one in CRR1, one in CRR3 and one combining both mutations (CRR1+3) (Figure 1). The RoxS CRR1 mutant still protected yflS against RNase J1 (Figure 3, lanes 14–17), suggesting that this region of RoxS is not important for the interaction with yflS. In contrast, the yflS RNA was no longer protected by the CRR3 RoxS mutant (Figure 3, lanes 20–23) or by the double CRR1+3 mutant (Figure 3, lanes 26–29). These results suggest that the CRR3 region of RoxS is the key C-rich motif for the initiation of base-pairing and protection of the 5' end of the yflS mRNA.

To confirm this result, we made compensatory mutations in the CRR3 of RoxS (CRR3-2) and at the 5' end of yflS (yflS^mut5') (Figure 4A) and assayed the capability of these variants to protect yflS from RNase J1 degradation in vitro. The RoxS CRR3-2 mutant showed significantly weaker protection of the yflS^WT transcript than wild-type RoxS (Figure 4B,C). Even though the yflS^mut5' RNA was intrinsically a better substrate for RNase J1 than yflS^WT, wild-type RoxS similarly showed weaker protection of the yflS^mut5' RNA than the CRR3-2 compensatory mutant (Figure 4B,C). These experiments strongly support the predicted base-pairing between RoxS CRR3 and the 5' end of the yflS mRNA.

Figure 4 with 1 supplement see all

Download asset Open asset

Figure 4—source data 1

Source data for the quantification of yflS and yflS^mut degradation by RNase J1 in vitro.

https://doi.org/10.7554/eLife.23602.011

Download elife-23602-fig4-data1-v2.xlsx

Lastly, we performed structure probing of RoxS in the presence and absence of yflS using the RNase V1 (specific for double stranded RNA) and RNase T2 (preference for unpaired adenines). The cleavage sites were in good agreement with the secondary structure of RoxS, as most of the strong RNase T2 cuts were located in the apical loops, while the RNase V1 cuts were primarily located in the helices (Figure 4—figure supplement 1). The addition of yflS RNA caused limited changes in reactivity restricted to the region encompassing CRR2 and CRR3. Several RNase V1 cleavages were induced close to CRR3 (at nts 63–65). Concomitantly, several RNase T2 cleavages located between CRR2 and CRR3 decreased under the same conditions. Together, these observations show convincingly that, as for the ppnKB mRNA, RoxS primarily uses the CRR3 region to hybridize to yflS mRNA.

RoxS also activates translation of the yflS mRNA

The data presented thus far suggests that RoxS binds to the 5’ end of the yflS mRNA principally via the CRR3 motif to protect it from degradation by RNase J1. To determine the impact of RoxS binding on the conformation of the yflS mRNA and potentially on translation, we also probed the secondary structure of the leader portion of yflS alone or bound to RoxS. The probes used in this case were RNase V1, RNase T1 (specific for unpaired guanines) and RNase T2. A summary of the results is presented in Figure 5. In the absence of RoxS, two equilibrium secondary structure models best explain the enzymatic cleavage patterns detected. In the two proposed models, the SD and the AUG start codon are engaged in weak basepairing interactions, while the four guanines at the extreme 5’ end of the yflS mRNA are accessible to RNase T1 and are therefore in single-stranded conformation. However, we noticed that adenines 11 to 14 were cleaved poorly by RNase T2 compared to other A-stretches (A50-54, A63-66, Figure 5—figure supplement 1). Their three-dimensional conformation may therefore be more complex than shown here. Upon addition of RoxS, we observed a significant decrease in the RNase T1 cleavages in the first 10 nucleotides of the yflS leader, while adenines 11 to 14 become more accessible to RNase T2. A similar effect was seen with the RoxS CRR1 mutant. However, no major effect was observed with the CRR3 variants, consistent with the idea that CRR3 is the key mediator of the interaction with the 5’ end of the yflS mRNA. In the absence of RoxS, we also noticed a strong reverse transcriptase (RT) stop between C34 and U35 of the yflS leader. This RT stop is located between the SD and AUG start codon and was suppressed by the binding of the wild-type RoxS or the CRR1 mutant, but not the CRR3 variants. These results strongly support the prediction that RoxS base-pairs with the extreme 5’ end of the yflS mRNA through CRR3, and suggest that RoxS additionally induces a long-range perturbation of the mRNA structure close to the SD sequence.

Figure 5 with 1 supplement see all

Download asset Open asset

Since the SD and the AUG codon are predicted to be poorly accessible and some structural changes close to the SD sequence were detected, we asked whether, in addition to protecting the yflS mRNA from degradation by RNase J1, RoxS might also have an impact on the translation of this mRNA. We therefore performed toeprinting experiments on the same 94-nt yflS leader fragment used for the RNase J1 in vitro degradation and structure probing assays (Figure 5). The yflS transcript was incubated with the 30S ribosomal subunit and initiator tRNA either with or without pre-incubation with RoxS (Figure 6). In the absence of RoxS, a characteristic reverse transcriptase stop was observed at position +16 relative to the start codon, showing that the translation initiation complex can be formed without RoxS. Addition of wild-type RoxS or the CRR1 mutant increased the signal at +16 approximately 3-fold, suggesting that RoxS stimulates the formation of the translation initiation complex. Significantly less stimulation was observed with the two CRR3 variants; it is possible that the apparent residual effect of these mutants comes from the extended base-pairing around CRR2 in these in vitro experiments. We also detected the strong RT stop between the SD and AUG start codon and, as previously, this species diminished with increasing concentration of wild type RoxS or CRR1 mutant, but not with the CRR3 mutants. The decrease of this signal between C34 and U35 is thus a signature of RoxS binding.

Figure 6

Download asset Open asset

To confirm the effect of RoxS binding on yflS mRNA expression in vivo, we made a yflS-lacZ translational fusion in which the promoter, regulatory sequences and the first 436 amino acids (consisting of 11 out of 12 trans-membrane domains) of YflS were fused in frame to LacZ and inserted at the amyE locus of the chromosome. We measured the ß-galactosidase activity of this fusion in the presence and absence of RoxS, at different times after induction with malate (Figure 7). In both strains, ß-galactosidase activity increased progressively after the addition of malate to the growth medium. After 60 min induction, the specific ß-galactosidase activity was 2-fold higher in a WT background compared to the RoxS mutant strain, showing that RoxS stimulates YflS synthesis in vivo.

Figure 7

Download asset Open asset

Figure 7—source data 1

Source data for the quantification of ß-galactosidase activity of the yflS-lacZ fusion.

https://doi.org/10.7554/eLife.23602.017

Download elife-23602-fig7-data1-v2.xlsx

RoxS blocks yflS degradation independently of translation

Since RoxS had an impact on both translation and degradation of the yflS mRNA (see above), it was conceivable that the primary effect of RoxS is on yflS mRNA translation. An increase of ribosome trafficking on the yflS mRNA could be sufficient to explain the increase in yflS stability as has been demonstrated for numerous trans-acting sRNAs in bacteria (Lalaouna et al., 2013). To uncouple the effect of degradation and translation, we mutated both the SD sequence (AGGAGA replaced by CAAAAA) and the fifth codon (AAA to TAA) to introduce a premature stop codon in the chromosomal copy of the yflS gene. This non-translated variant is called yflS^mutTR. The half-life of yflS^mutTR RNA was slightly decreased compared to the wild-type strain (1.6 min vs 2.5 min respectively; Figure 8, Figure 8—figure supplement 1), likely due to the decreased ribosome occupancy of the coding sequence and increased access to RNase Y. However, the yflS^mutTR RNA was still destabilized about 2.3-fold in the absence of RoxS compared to the parental strain (1.6 min vs 0.7 min; Figure 8, Figure 8—figure supplement 1). These results suggest that the stimulation of translation by RoxS is not a prerequisite for the protection of yflS mRNA from RNase J1. In other words, RoxS-mediated stimulation of yflS mRNA translation and its protection from RNase J1 are independent phenomena.

Figure 8 with 1 supplement see all

Download asset Open asset

RoxS is transcriptionally repressed by Rex

As mentioned previously, the yflS gene encodes a malate transporter whose expression is induced by the two-component system MalKR when malate is present in the environment (Tanaka et al., 2003). Since RoxS post-transcriptionally regulates yflS gene expression, we asked whether carbon sources like glucose and malate had an impact on RoxS expression. To test this, we added 0.2% glucose and/or malate to the medium at mid-exponential phase in 2xTY medium and followed RoxS expression for 1 hr by Northern blot (Figure 9A). RoxS levels increased for about 30 min after addition of glucose before falling again at the 60 min time point. A similar profile, but of much greater amplitude was seen upon addition of malate to the medium and addition of both carbon sources together (malate + glucose) had a cumulative effect on RoxS levels.

Figure 9 with 2 supplements see all

Download asset Open asset

Figure 9—source data 1

Source data for the quantification of RoxS derepression by malate in Bacillus subtilis mutant strains (∆resDE, ∆malK, ∆rex).

https://doi.org/10.7554/eLife.23602.022

Download elife-23602-fig9-data1-v2.xlsx

We performed a number of experiments to determine the regulator of RoxS expression in the presence of malate from among three potential candidates ResDE, MalKR and Rex. We had previously shown that RoxS is induced by nitric-oxide (NO) in the growth medium via the two-component system ResDE (Durand et al., 2015a). We therefore first tested the involvement of ResDE in the induction of RoxS expression by malate (Figure 9B). In the ∆resDE mutant strain, RoxS levels were lower than in the WT strain, as observed previously (Durand et al., 2015a). However, RoxS levels were still inducible by malate in the absence of ResDE, with the same kinetics as in the WT strain, showing that ResDE is not involved in this regulation.

The two-component system MalKR, known to respond to malate levels in B. subtilis and responsible for induction of yflS expression (see above), was also a good candidate for RoxS regulation by malate. In a ∆malKR strain mutant, RoxS levels no longer increased upon addition of malate (Figure 9C), suggesting a role for this two-component system in regulation. However, a comparison of the RoxS promoter sequence to those of the maeN, maeA and yflS genes, all known to bind MalR, did not reveal any obvious similarities. We purified the MalR protein and performed an electrophoretic mobility shift assay (EMSA) with a DNA fragment corresponding to the RoxS promoter, with the maeA promoter as a positive control and a portion of the yflS coding region as a negative control (Figure 9—figure supplement 1). We confirmed binding of MalR to the maeA promoter but found no specific interaction with the RoxS promoter region. These data suggest that the effect of the MalKR two-component system on RoxS levels is indirect.

Further examination of the sequence around the RoxS transcription start site revealed the presence of a potential binding site for the Rex transcriptional repressor (Härtig and Jahn, 2012) immediately downstream of the transcription start site (Figure 9—figure supplement 2). We therefore examined the induction of RoxS expression upon addition of malate to a ∆rex strain. In the absence of Rex, RoxS levels were already high before addition of malate induction and accumulated to even higher levels over time (Figure 9D). Furthermore, we no longer observed the reduction in RoxS levels one hour after malate addition. This result suggests that Rex represses RoxS transcription and that addition of malate to the growth medium causes a transient release from this repression for at least 30 mins (see discussion). To demonstrate that Rex directly binds the RoxS promoter region, we performed EMSA experiments with the roxS and the lactate dehydrogenase (ldh) gene promoter sequences, and with a portion of the yflS coding region as a negative control (Figure 10). As expected, we did not observe complexes with the yflS coding region but we detected gel-shifted species with the ldh promoter, known to bind Rex (Wang et al., 2008). We also obtained a gel-shifted species for the roxS promoter fragment, confirming that Rex can directly bind this promoter region and explaining how it can inhibit transcription. Interestingly, the Rex protein and its predicted binding site just downstream of the roxS promoter are conserved in S. aureus and in Staphylococcaceae in general (Pagels et al., 2010), suggesting similar regulation may occur in these pathogenic bacteria (Figure 9—figure supplement 2).

Figure 10

Download asset Open asset

RoxS regulates ppnkB and sucC mRNA stability far more efficiently in the presence of malate

In our previous studies, we showed that RoxS blocks translation of the ppnkB and sucC mRNAs and indirectly destabilizes these transcripts (Durand et al., 2015a). Since RoxS expression is derepressed in the presence of malate, we retested the impact of RoxS on these two mRNAs in these growth conditions. We measured the half-lives of the ppnKB and sucCD mRNAs in the presence or in absence of malate by Northern blotting of RNAs isolated at different times after rifampicin addition. We previously showed that RoxS destabilized the ppnkB about 2-fold and had little impact on sucCD mRNA half-life in 2xTY with no added carbon source (Figure 11A and [Durand et al., 2015a]). In 2xTY plus malate, however, RoxS destabilized the ppnkB mRNA about 20-fold and the sucD mRNA about 8-fold (Figure 11B). This result confirms our previous data and shows that RoxS has a far greater impact on gene expression in the presence of malate.

Figure 11

Download asset Open asset

In this study, we report for the first time that a small regulatory RNA (RoxS) can directly prevent 5’-to-3’ mRNA degradation of its mRNA target. RoxS binds to the 5' end of the yflS mRNA through base-pairing interactions primarily with the C-rich region CRR3 (Figure 1). The inhibitory effect on RNase J1 5'−3' exoribonuclease activity can easily be understood in structural terms; the RNA binding channel that leads to the RNase J1 catalytic site is only long enough to accommodate 4–5 nts of single stranded RNA. The unbound yflS mRNA is predicted to have at least six unpaired nts before the first hairpin structure (Figure 5) and would thus be expected to be a good substrate for RNase J1 once the 5'-triphosphate group has been removed. When bound to RoxS, the duplexed 5' end of the yflS RNA can no longer reach the catalytic site, explaining how it is protected from degradation.

To our knowledge, there are only three other examples of direct mRNA stabilization by a trans-acting regulatory RNA in Gram-positive bacteria. In Streptococcus mutans, the 5’-UTR of the intact irvA mRNA interacts with the coding sequence of the gbpC mRNA, encoding a surface lectin implicated in dextran dependent aggregation stress (DDAG). This interaction blocks endoribonucleolyic cleavage of the gbpC mRNA by RNase J2 (Liu et al., 2015). Obana et al. (2010, 2013) showed that the Clostridium perfringens colA mRNA (encoding collagenase) is stabilized by the VR-RNA. Binding of the sRNA to the colA mRNA leads to a cleavage its 5’-UTR, which releases the colA SD sequence and stimulates translation initiation. The stabilization of colA depends on the binding of a ribosome to the SD sequence, but translation has no impact on mRNA stability. In the pathogen Streptococcus pyogenes, the FasX sRNA stabilizes the streptokinase encoding mRNA (ska) (Ramirez-Peña et al., 2010) by binding to its extreme 5' end via a sequence resembling that of the RoxS CRR3 (four consecutive C's matched with four consecutive G's). Our data would suggest that RNase J1 is a prime candidate for FasX regulation of the ska mRNA. In fact, the sRNA-dependent 5' end protection mechanism discovered here could in theory be used to activate expression of any mRNA in any microorganism by shielding it from enzymes (e.g. RppH, RNase J1, RNase E) requiring a single stranded 5’ extremity to initiate degradation.

Our data also indicate that binding of RoxS to the 5' end of the yflS mRNA both protects it from degradation by RNase J1 (Figures 2, 3 and 4) and stimulates 30S binding to the SD sequence in vitro (Figure 6), suggesting an activation of translation. Moreover, the increased expression of the yflS(436)-lacZ fusion in the presence of RoxS is consistent with an increase in translation efficiency (Figure 7). We were able to uncouple the effect of RoxS on yflS mRNA stability and translation by making a non-translated variant (yflS^mutTR) and showed that RoxS is still able to stabilize this mRNA independently of translation (Figure 8). However, in all likelihood, the two events are indissociable in vivo with the wild-type yflS mRNA, that is, stabilization of the mRNA and increased translation go hand in hand to produce more of the YflS transporter in the presence of malate.

The enzymatic probing of yflS mRNA indicated that the binding of RoxS, in addition to protecting the 5' end from RNase T1 cleavage, causes subtle modifications of the secondary structure of the yflS leader, that could account for the stimulation of initiation complex formation. Most noticeably, a strong RT stop between C34 and U35 (five nucleotides upstream of the AUG) is lost upon binding of WT RoxS or the CRR1 mutant, but is retained when yflS mRNA is incubated with the inactive CRR3 variant. This suggests a structural modification to the apex of the stem-loop in the vicinity of the SD sequence upon binding to RoxS. Surprisingly, however, no other nucleotides in this neighborhood became more reactive to single stranded probes (e.g. RNase T1) in the presence of RoxS, in particular the two G-C pairs (25G/34C and 26G/33C) that are predicted to close the apical loop (Figure 5). This suggests that the 3D structure of yflS leader and the impact of RoxS may be more complex than the model presented in Figure 5. The yflS mRNA sequence contains several stretches of consecutive G-residues in the first 50 nts that could potentially form G-quadruplex RNA structures (Kikin et al., 2006). Alternatively, part of the 5’ leader including the AAACAA sequence (nts 12–17) of yflS mRNA could potentially form a triple helix with the stem of the adjacent hairpin rich in AU pairings. In both cases, RoxS binding would disrupt a tertiary fold favoring the recognition of the ribosome. Our structure probing data are consistent with a model in which the duplex formed by binding of RoxS to the 5’ end of the yflS mRNA not only blocks RNase J1 access but also disrupts a particular fold of the 5’UTR that is unfavorable to ribosome recruitment.

Our study showed that the expression of the RoxS sRNA is not only induced by nitric oxide (NO) as we described previously (Durand et al., 2015a), but also by the presence of malate in the growth medium (Figure 9). The malate-induced increase in RoxS expression is indirectly dependent on the two-component system MalKR and directly dependent on release from Rex repression (Figures 9 and 10). The Rex repressor is a sensor of the NAD⁺/NADH ratio in the cell. It has a 20,000-fold higher affinity for NADH than for NAD⁺ and the binding of NADH to Rex provokes a structural rearrangement which prevents its binding to target promoters and allows the expression of genes involved in fermentation (Larsson et al., 2005; Wang et al., 2008; McLaughlin et al., 2010). The temporary release from Rex-mediated repression of RoxS expression may be linked to the generation of NADH in the early steps of malate metabolism, either in the conversion of malate to oxaloacetate by malate dehydrogenase (Mdh) in the TCA cycle, or its conversion to pyruvate by the malic enzymes MaeA, MalS and MleA (Doan et al., 2003; Lerondel et al., 2006; Meyer and Stülke, 2013) (Figure 12). The effect of the MalKR mutation on RoxS expression (Figure 9C) is most likely explained by the decrease in expression of the MalKR regulon, consisting of the malate transporter genes yflS and maeN, and that of the malic enzyme maeA, and the consequent impact on NADH levels and Rex activity.

Figure 12

Download asset Open asset

B. subtilis has at least four malate transporters: MaeN (Wei et al., 2000) and YflS (Tanaka et al., 2003), both regulated by the two-component system MalKR (Tanaka et al., 2003), and MleN (Wei et al., 2000) and CimH, the latter of which is primarily a transporter of citrate (Sobczak and Lolkema, 2005). Malate is a common source of carbon for B. subtilis in its natural habitat since it is released by the roots of many plants (Rudrappa et al., 2008). Most of the malate imported by B. subtilis cells serves in gluconeogenesis and the excess is sent towards overflow metabolism/fermentation and the production of lactate, acetate, acetoin and 2,3-butanediol (Kleijn et al., 2010) (Figure 12). This overflow metabolism/fermentation allows NAD⁺ regeneration, primarily catalyzed by lactate dehydrogenase (Ldh/LctE) that converts pyruvate to lactate (Cruz Ramos et al., 2000) (Figure 12). Transcription of the maeN mRNA, encoding the main malate transporter, is essentially turned off during fermentation and ethanol stress (Nicolas et al., 2012). It is possible that RoxS specifically stabilizes the yflS mRNA to compensate for this decrease in maeN expression, thus facilitating the continued uptake of malate during conditions where the fermentation/overflow pathway is turned on.

The identification of Rex as a regulator of RoxS expression significantly expands the Rex regulon and contributes to our understanding of the pivotal role of Rex in Bacilli (Laouami et al., 2014); up to now only a handful of fermentation pathway genes or operons were known to be directly repressed by this transcriptional regulator. Our previous study showed that 30% of genes with increased expression in a ∆roxS strain are linked to oxido-reduction processes (Durand et al., 2015a). Its down-regulated targets include ppnKB, encoding NAD-kinase, and the TCA cycle components sucCD and citZ. PpnKB and the TCA cycle are significant contributors to the production of NADPH and NADH, respectively (Figure 12). These observations, together with the fact that RoxS is regulated by the NADH-sensitive transcriptional repressor Rex, suggest that a key role of RoxS is to temporarily turn down part of the TCA cycle to help restore the NAD⁺/NADH balance. In the presence of malate, this re-equilibration is further aided by the increased activity of Ldh regenerating NAD⁺ (see above). Once the NAD⁺/NADH ratio is restored, Rex would be predicted to become active again as a repressor, explaining the decrease in RoxS levels seen in Northern blots about one hour after addition of malate and the loss of this down-regulation in the Rex mutant (Figure 9D). In further agreement with a role for RoxS in managing NADH levels, we showed in our previous report that expression of the lactate dehydrogenase gene (ldh) and the cydA gene, encoding a subunit of cytochrome bd, is up-regulated in ∆roxS cells (Durand et al., 2015a). Neither of these mRNAs is predicted to be a direct target of RoxS, but both are members of the Rex regulon. Their increased expression is thus most likely indicative of higher intracellular NADH levels and Rex derepression in the ∆roxS strain. Interestingly, a link between NO levels and regulation of virulence via NADH and Rex has been proposed in S. aureus (Pagels et al., 2010). The conservation of the Rex binding site downstream of the RsaE/RoxS promoter in S. aureus suggests that this pathway is evolutionarily conserved to best serve the adaptation of B. subtilis and S. aureus to their respective niches.

1. 1. Arnaud M
   2. Chastanet A
   3. Débarbouillé M

   (2004) New vector for efficient allelic replacement in naturally Gram-positive bacteria

   Applied and Environmental Microbiology 70:6887–6891.

   https://doi.org/10.1128/AEM.70.11.6887-6891.2004

   • PubMed
   • Google Scholar
2. 1. Bandyra KJ
   2. Said N
   3. Pfeiffer V
   4. Górna MW
   5. Vogel J
   6. Luisi BF

   (2012) The seed region of a small RNA drives the controlled destruction of the target mRNA by the endoribonuclease RNase E

   Molecular Cell 47:943–953.

   https://doi.org/10.1016/j.molcel.2012.07.015

   • PubMed
   • Google Scholar
3. 1. Bechhofer DH
   2. Oussenko IA
   3. Deikus G
   4. Yao S
   5. Mathy N
   6. Condon C

   (2008) Analysis of mRNA decay in Bacillus subtilis

   Methods in Enzymology 447:259–276.

   https://doi.org/10.1016/S0076-6879(08)02214-3

   • PubMed
   • Google Scholar
4. 1. Bechhofer DH

   (2009) Messenger RNA decay and maturation in Bacillus subtilis

   Progress in Molecular Biology and Translational Science 85:231–273.

   https://doi.org/10.1016/S0079-6603(08)00806-4

   • PubMed
   • Google Scholar
5. 1. Bohn C
   2. Rigoulay C
   3. Chabelskaya S
   4. Sharma CM
   5. Marchais A
   6. Skorski P
   7. Borezée-Durant E
   8. Barbet R
   9. Jacquet E
   10. Jacq A
   11. Gautheret D
   12. Felden B
   13. Vogel J
   14. Bouloc P

   (2010) Experimental discovery of small RNAs in Staphylococcus aureus reveals a riboregulator of central metabolism

   Nucleic Acids Research 38:6620–6636.

   https://doi.org/10.1093/nar/gkq462

   • PubMed
   • Google Scholar
6. 1. Boudry P
   2. Gracia C
   3. Monot M
   4. Caillet J
   5. Saujet L
   6. Hajnsdorf E
   7. Dupuy B
   8. Martin-Verstraete I
   9. Soutourina O

   (2014) Pleiotropic role of the RNA chaperone protein hfq in the human pathogen Clostridium difficile

   Journal of Bacteriology 196:3234–3248.

   https://doi.org/10.1128/JB.01923-14

   • PubMed
   • Google Scholar
7. 1. Britton RA
   2. Wen T
   3. Schaefer L
   4. Pellegrini O
   5. Uicker WC
   6. Mathy N
   7. Tobin C
   8. Daou R
   9. Szyk J
   10. Condon C

   (2007) Maturation of the 5' end of Bacillus subtilis 16S rRNA by the essential ribonuclease YkqC/RNase J1

   Molecular Microbiology 63:127–138.

   https://doi.org/10.1111/j.1365-2958.2006.05499.x

   • PubMed
   • Google Scholar
8. 1. Condon C
   2. Putzer H

   (2002) The phylogenetic distribution of bacterial ribonucleases

   Nucleic Acids Research 30:5339–5346.

   https://doi.org/10.1093/nar/gkf691

   • PubMed
   • Google Scholar
9. 1. Cruz Ramos H
   2. Hoffmann T
   3. Marino M
   4. Nedjari H
   5. Presecan-Siedel E
   6. Dreesen O
   7. Glaser P
   8. Jahn D

   (2000) Fermentative metabolism of Bacillus subtilis: physiology and regulation of gene expression

   Journal of Bacteriology 182:3072–3080.

   https://doi.org/10.1128/JB.182.11.3072-3080.2000

   • PubMed
   • Google Scholar
10. 1. Daou-Chabo R
    2. Mathy N
    3. Bénard L
    4. Condon C

    (2009) Ribosomes initiating translation of the hbs mRNA protect it from 5'-to-3' exoribonucleolytic degradation by RNase J1

    Molecular Microbiology 71:1538–1550.

    https://doi.org/10.1111/j.1365-2958.2009.06620.x

    • PubMed
    • Google Scholar
11. 1. Doan T
    2. Servant P
    3. Tojo S
    4. Yamaguchi H
    5. Lerondel G
    6. Yoshida K
    7. Fujita Y
    8. Aymerich S

    (2003) The Bacillus subtilis ywkA gene encodes a malic enzyme and its transcription is activated by the YufL/YufM two-component system in response to malate

    Microbiology 149:2331–2343.

    https://doi.org/10.1099/mic.0.26256-0

    • PubMed
    • Google Scholar
12. 1. Durand S
    2. Braun F
    3. Lioliou E
    4. Romilly C
    5. Helfer AC
    6. Kuhn L
    7. Quittot N
    8. Nicolas P
    9. Romby P
    10. Condon C

    (2015a) A nitric oxide regulated small RNA controls expression of genes involved in redox homeostasis in Bacillus subtilis

    PLOS Genetics 11:e1004957.

    https://doi.org/10.1371/journal.pgen.1004957

    • PubMed
    • Google Scholar
13. 1. Durand S
    2. Gilet L
    3. Bessières P
    4. Nicolas P
    5. Condon C

    (2012) Three essential ribonucleases-RNase Y, J1, and III-control the abundance of a majority of Bacillus subtilis mRNAs

    PLoS Genetics 8:e1002520.

    https://doi.org/10.1371/journal.pgen.1002520

    • PubMed
    • Google Scholar
14. 1. Durand S
    2. Tomasini A
    3. Braun F
    4. Condon C
    5. Romby P

    (2015b) sRNA and mRNA turnover in Gram-positive bacteria

    FEMS Microbiology Reviews 39:316–330.

    https://doi.org/10.1093/femsre/fuv007

    • PubMed
    • Google Scholar
15. 1. Fechter P
    2. Chevalier C
    3. Yusupova G
    4. Yusupov M
    5. Romby P
    6. Marzi S

    (2009) Ribosomal initiation complexes probed by toeprinting and effect of trans-acting translational regulators in bacteria

    Methods in Molecular Biology 540:247–263.

    https://doi.org/10.1007/978-1-59745-558-9_18

    • PubMed
    • Google Scholar
16. 1. Figaro S
    2. Durand S
    3. Gilet L
    4. Cayet N
    5. Sachse M
    6. Condon C

    (2013) Bacillus subtilis mutants with knockouts of the genes encoding ribonucleases RNase Y and RNase J1 are viable, with Major defects in cell morphology, sporulation, and competence

    Journal of Bacteriology 195:2340–2348.

    https://doi.org/10.1128/JB.00164-13

    • PubMed
    • Google Scholar
17. 1. Fröhlich KS
    2. Papenfort K
    3. Fekete A
    4. Vogel J

    (2013) A small RNA activates CFA synthase by isoform-specific mRNA stabilization

    The EMBO Journal 32:2963–2979.

    https://doi.org/10.1038/emboj.2013.222

    • PubMed
    • Google Scholar
18. 1. Geissmann T
    2. Chevalier C
    3. Cros MJ
    4. Boisset S
    5. Fechter P
    6. Noirot C
    7. Schrenzel J
    8. François P
    9. Vandenesch F
    10. Gaspin C
    11. Romby P
    12. Franc P
    13. Francois P

    (2009) A search for small noncoding RNAs in Staphylococcus aureus reveals a conserved sequence motif for regulation

    Nucleic Acids Research 37:7239–7257.

    https://doi.org/10.1093/nar/gkp668

    • PubMed
    • Google Scholar
19. 1. Gendron N
    2. Putzer H
    3. Grunberg-Manago M

    (1994) Expression of both Bacillus subtilis threonyl-tRNA synthetase genes is autogenously regulated

    Journal of Bacteriology 176:486–494.

    https://doi.org/10.1128/jb.176.2.486-494.1994

    • PubMed
    • Google Scholar
20. 1. Guérout-Fleury AM
    2. Shazand K
    3. Frandsen N
    4. Stragier P

    (1995) Antibiotic-resistance cassettes for Bacillus subtilis

    Gene 167:335–336.

    https://doi.org/10.1016/0378-1119(95)00652-4

    • PubMed
    • Google Scholar
21. 1. Helfer A-C
    2. Romilly C
    3. Chevalier C
    4. Lioliou E
    5. Marzi S
    6. Romby P

    (2014)

    Handbook of RNA Biochemistry

    205–230, Probing RNA structure in vitro with enzymes and chemicals, Handbook of RNA Biochemistry, Wiley-VCH Verlag GmbH & Co.

    • Google Scholar
22. 1. Hämmerle H
    2. Amman F
    3. Večerek B
    4. Stülke J
    5. Hofacker I
    6. Bläsi U

    (2014) Impact of hfq on the Bacillus subtilis transcriptome

    PLoS One 9:e98661.

    https://doi.org/10.1371/journal.pone.0098661

    • PubMed
    • Google Scholar
23. 1. Härtig E
    2. Jahn D

    (2012)

    Advances in Microbial Physiology 

    195–216, Chapter five – regulation of the anaerobic metabolism in Bacillus subtilis, Advances in Microbial Physiology .

    • Google Scholar
24. 1. Kery MB
    2. Feldman M
    3. Livny J
    4. Tjaden B

    (2014) TargetRNA2: identifying targets of small regulatory RNAs in bacteria

    Nucleic Acids Research 42:W124–W129.

    https://doi.org/10.1093/nar/gku317

    • PubMed
    • Google Scholar
25. 1. Kikin O
    2. D'Antonio L
    3. Bagga PS

    (2006) QGRS Mapper: a web-based server for predicting G-quadruplexes in nucleotide sequences

    Nucleic Acids Research 34:W676–W682.

    https://doi.org/10.1093/nar/gkl253

    • PubMed
    • Google Scholar
26. 1. Kleijn RJ
    2. Buescher JM
    3. Le Chat L
    4. Jules M
    5. Aymerich S
    6. Sauer U

    (2010) Metabolic fluxes during strong carbon catabolite repression by malate in Bacillus subtilis

    Journal of Biological Chemistry 285:1587–1596.

    https://doi.org/10.1074/jbc.M109.061747

    • PubMed
    • Google Scholar
27. 1. Lalaouna D
    2. Simoneau-Roy M
    3. Lafontaine D
    4. Massé E

    (2013) Regulatory RNAs and target mRNA decay in prokaryotes

    Biochimica Et Biophysica Acta (BBA) - Gene Regulatory Mechanisms 1829:742–747.

    https://doi.org/10.1016/j.bbagrm.2013.02.013

    • PubMed
    • Google Scholar
28. 1. Laouami S
    2. Clair G
    3. Armengaud J
    4. Duport C

    (2014) Proteomic evidences for Rex regulation of metabolism in toxin-producing Bacillus cereus ATCC 14579

    PLoS One 9:e107354.

    https://doi.org/10.1371/journal.pone.0107354

    • PubMed
    • Google Scholar
29. 1. Larsson JT
    2. Rogstam A
    3. von Wachenfeldt C

    (2005) Coordinated patterns of cytochrome bd and lactate dehydrogenase expression in Bacillus subtilis

    Microbiology 151:3323–3335.

    https://doi.org/10.1099/mic.0.28124-0

    • PubMed
    • Google Scholar
30. 1. Lerondel G
    2. Doan T
    3. Zamboni N
    4. Sauer U
    5. Aymerich S

    (2006) YtsJ has the Major physiological role of the four paralogous malic enzyme isoforms in Bacillus subtilis

    Journal of Bacteriology 188:4727–4736.

    https://doi.org/10.1128/JB.00167-06

    • PubMed
    • Google Scholar
31. 1. Liu N
    2. Niu G
    3. Xie Z
    4. Chen Z
    5. Itzek A
    6. Kreth J
    7. Gillaspy A
    8. Zeng L
    9. Burne R
    10. Qi F
    11. Merritt J

    (2015) The Streptococcus mutans irvA gene encodes a trans-acting riboregulatory mRNA

    Molecular Cell 57:179–190.

    https://doi.org/10.1016/j.molcel.2014.11.003

    • PubMed
    • Google Scholar
32. 1. Mathy N
    2. Bénard L
    3. Pellegrini O
    4. Daou R
    5. Wen T
    6. 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.

    https://doi.org/10.1016/j.cell.2007.02.051

    • PubMed
    • Google Scholar
33. 1. McLaughlin KJ
    2. Strain-Damerell CM
    3. Xie K
    4. Brekasis D
    5. Soares AS
    6. Paget MS
    7. Kielkopf CL

    (2010) Structural basis for NADH/NAD+ redox sensing by a Rex family repressor

    Molecular Cell 38:563–575.

    https://doi.org/10.1016/j.molcel.2010.05.006

    • PubMed
    • Google Scholar
34. 1. Meyer FM
    2. Stülke J

    (2013) Malate metabolism in Bacillus subtilis: distinct roles for three classes of malate-oxidizing enzymes

    FEMS Microbiology Letters 339:17–22.

    https://doi.org/10.1111/1574-6968.12041

    • PubMed
    • Google Scholar
35. 1. Nicolas P
    2. Mäder U
    3. Dervyn E
    4. Rochat T
    5. Leduc A
    6. Pigeonneau N
    7. Bidnenko E
    8. Marchadier E
    9. Hoebeke M
    10. Aymerich S
    11. Becher D
    12. Bisicchia P
    13. Botella E
    14. Delumeau O
    15. Doherty G
    16. Denham EL
    17. Fogg MJ
    18. Fromion V
    19. Goelzer A
    20. Hansen A
    21. Härtig E
    22. Harwood CR
    23. Homuth G
    24. Jarmer H
    25. Jules M
    26. Klipp E
    27. Le Chat L
    28. Lecointe F
    29. Lewis P
    30. Liebermeister W
    31. March A
    32. Mars RA
    33. Nannapaneni P
    34. Noone D
    35. Pohl S
    36. Rinn B
    37. Rügheimer F
    38. Sappa PK
    39. Samson F
    40. Schaffer M
    41. Schwikowski B
    42. Steil L
    43. Stülke J
    44. Wiegert T
    45. Devine KM
    46. Wilkinson AJ
    47. van Dijl JM
    48. Hecker M
    49. Völker U
    50. Bessières P
    51. Noirot P

    (2012) Condition-dependent transcriptome reveals high-level regulatory architecture in Bacillus subtilis

    Science 335:1103–1106.

    https://doi.org/10.1126/science.1206848

    • PubMed
    • Google Scholar
36. 1. Nielsen JS
    2. Lei LK
    3. Ebersbach T
    4. Olsen AS
    5. Klitgaard JK
    6. Valentin-Hansen P
    7. Kallipolitis BH

    (2010) Defining a role for hfq in Gram-positive bacteria: evidence for Hfq-dependent antisense regulation in Listeria monocytogenes

    Nucleic Acids Research 38:907–919.

    https://doi.org/10.1093/nar/gkp1081

    • PubMed
    • Google Scholar
37. 1. Obana N
    2. Nomura N
    3. Nakamura K

    (2013) Structural requirement in Clostridium perfringens collagenase mRNA 5' leader sequence for translational induction through small RNA-mRNA base pairing

    Journal of Bacteriology 195:2937–2946.

    https://doi.org/10.1128/JB.00148-13

    • PubMed
    • Google Scholar
38. 1. Obana N
    2. Shirahama Y
    3. Abe K
    4. Nakamura K

    (2010) Stabilization of Clostridium perfringens collagenase mRNA by VR-RNA-dependent cleavage in 5' leader sequence

    Molecular Microbiology 77:1416–1428.

    https://doi.org/10.1111/j.1365-2958.2010.07258.x

    • PubMed
    • Google Scholar
39. 1. Pagels M
    2. Fuchs S
    3. Pané-Farré J
    4. Kohler C
    5. Menschner L
    6. Hecker M
    7. McNamarra PJ
    8. Bauer MC
    9. von Wachenfeldt C
    10. Liebeke M
    11. Lalk M
    12. Sander G
    13. von Eiff C
    14. Proctor RA
    15. Engelmann S

    (2010) Redox sensing by a Rex-family repressor is involved in the regulation of anaerobic gene expression in Staphylococcus aureus

    Molecular Microbiology 76:1142–1161.

    https://doi.org/10.1111/j.1365-2958.2010.07105.x

    • PubMed
    • Google Scholar
40. 1. Piton J
    2. Larue V
    3. Thillier Y
    4. Dorléans A
    5. Pellegrini O
    6. Li de la Sierra-Gallay I
    7. Vasseur JJ
    8. Debart F
    9. Tisné C
    10. Condon C

    (2013) Bacillus subtilis RNA deprotection enzyme RppH recognizes guanosine in the second position of its substrates

    PNAS 110:8858–8863.

    https://doi.org/10.1073/pnas.1221510110

    • PubMed
    • Google Scholar
41. 1. Ramirez-Peña E
    2. Treviño J
    3. Liu Z
    4. Perez N
    5. Sumby P
    6. Ramirez-Pena E
    7. Trevino J

    (2010) The group A Streptococcus small regulatory RNA FasX enhances Streptokinase activity by increasing the stability of the ska mRNA transcript

    Molecular Microbiology 78:1332–1347.

    https://doi.org/10.1111/j.1365-2958.2010.07427.x

    • PubMed
    • Google Scholar
42. 1. Richards J
    2. Liu Q
    3. Pellegrini O
    4. Celesnik H
    5. Yao S
    6. Bechhofer DH
    7. Condon C
    8. Belasco JG

    (2011) An RNA pyrophosphohydrolase triggers 5'-exonucleolytic degradation of mRNA in Bacillus subtilis

    Molecular Cell 43:940–949.

    https://doi.org/10.1016/j.molcel.2011.07.023

    • PubMed
    • Google Scholar
43. 1. Rochat T
    2. Delumeau O
    3. Figueroa-Bossi N
    4. Noirot P
    5. Bossi L
    6. Dervyn E
    7. Bouloc P

    (2015) Tracking the Elusive function of Bacillus subtilis hfq

    PLoS One 10:e0124977.

    https://doi.org/10.1371/journal.pone.0124977

    • PubMed
    • Google Scholar
44. 1. Rudrappa T
    2. Czymmek KJ
    3. Paré PW
    4. Bais HP

    (2008) Root-secreted malic acid recruits beneficial soil bacteria

    Plant Physiology 148:1547–1556.

    https://doi.org/10.1104/pp.108.127613

    • PubMed
    • Google Scholar
45. 1. Sinturel F
    2. Navickas A
    3. Wery M
    4. Descrimes M
    5. Morillon A
    6. Torchet C
    7. Benard L

    (2015) Cytoplasmic control of Sense-Antisense mRNA pairs

    Cell Reports 12:1853–1864.

    https://doi.org/10.1016/j.celrep.2015.08.016

    • PubMed
    • Google Scholar
46. 1. Sobczak I
    2. Lolkema JS

    (2005) The 2-hydroxycarboxylate transporter family: physiology, structure, and mechanism

    Microbiology and Molecular Biology Reviews 69:665–695.

    https://doi.org/10.1128/MMBR.69.4.665-695.2005

    • PubMed
    • Google Scholar
47. 1. Stead MB
    2. Agrawal A
    3. Bowden KE
    4. Nasir R
    5. Mohanty BK
    6. Meagher RB
    7. Kushner SR

    (2012) RNAsnap: a rapid, quantitative and inexpensive, method for isolating total RNA from bacteria

    Nucleic Acids Research 40:e156–159.

    https://doi.org/10.1093/nar/gks680

    • PubMed
    • Google Scholar
48. 1. Tanaka K
    2. Kobayashi K
    3. Ogasawara N

    (2003) The Bacillus subtilis YufLM two-component system regulates the expression of the malate transporters MaeN (YufR) and YflS, and is essential for utilization of malate in minimal medium

    Microbiology 149:2317–2329.

    https://doi.org/10.1099/mic.0.26257-0

    • PubMed
    • Google Scholar
49. 1. Updegrove TB
    2. Zhang A
    3. Storz G

    (2016) Hfq: the flexible RNA matchmaker

    Current Opinion in Microbiology 30:133–138.

    https://doi.org/10.1016/j.mib.2016.02.003

    • PubMed
    • Google Scholar
50. 1. Wagner EG
    2. Romby P

    (2015) Small RNAs in bacteria and archaea: who they are, what they do, and how they do it

    Advances in Genetics 90:133–208.

    https://doi.org/10.1016/bs.adgen.2015.05.001

    • PubMed
    • Google Scholar
51. 1. Wang E
    2. Bauer MC
    3. Rogstam A
    4. Linse S
    5. Logan DT
    6. von Wachenfeldt C

    (2008) Structure and functional properties of the Bacillus subtilis transcriptional repressor rex

    Molecular Microbiology 69:466–478.

    https://doi.org/10.1111/j.1365-2958.2008.06295.x

    • PubMed
    • Google Scholar
52. 1. Wei Y
    2. Guffanti AA
    3. Ito M
    4. Krulwich TA

    (2000) Bacillus subtilis YqkI is a novel malic/Na+-lactate antiporter that enhances growth on malate at low protonmotive force

    Journal of Biological Chemistry 275:30287–30292.

    https://doi.org/10.1074/jbc.M001112200

    • PubMed
    • Google Scholar

Author details

1. Sylvain Durand

   UMR8261 CNRS, Université Paris Diderot (Sorbonne Paris Cité), Institut de Biologie Physico-Chimique, Paris, France

   Contribution

   SD, Conceptualization, Formal analysis, Funding acquisition, Validation, Investigation, Methodology, Writing—original draft, Project administration, Writing—review and editing

   For correspondence

   durand@ibpc.fr

   Competing interests

   The authors declare that no competing interests exist.

   "This ORCID iD identifies the author of this article:" 0000-0003-3509-6463

2. Frédérique Braun

   UMR8261 CNRS, Université Paris Diderot (Sorbonne Paris Cité), Institut de Biologie Physico-Chimique, Paris, France

   Contribution

   FB, Conceptualization, Formal analysis, Investigation, Writing—review and editing

   Competing interests

   The authors declare that no competing interests exist.

3. Anne-Catherine Helfer

   Université de Strasbourg, CNRS, Architecture et Réactivité de l’ARN, Strasbourg, France

   Contribution

   A-CH, Formal analysis, Investigation, Methodology

   Competing interests

   The authors declare that no competing interests exist.

4. Pascale Romby

   1. UMR8261 CNRS, Université Paris Diderot (Sorbonne Paris Cité), Institut de Biologie Physico-Chimique, Paris, France
   2. Université de Strasbourg, CNRS, Architecture et Réactivité de l’ARN, Strasbourg, France

   Contribution

   PR, Conceptualization, Formal analysis, Supervision, Funding acquisition, Validation, Visualization, Project administration, Writing—review and editing

   Competing interests

   The authors declare that no competing interests exist.

5. Ciarán Condon

   UMR8261 CNRS, Université Paris Diderot (Sorbonne Paris Cité), Institut de Biologie Physico-Chimique, Paris, France

   Contribution

   CC, Conceptualization, Formal analysis, Supervision, Funding acquisition, Validation, Visualization, Project administration, Writing—review and editing

   For correspondence

   condon@ibpc.fr

   Competing interests

   The authors declare that no competing interests exist.

• 2,448

  views

• 479

  downloads

• 42

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.