sRNA and mRNA turnover in Gram-positive bacteria

FEMS Microbiology Reviews, fuv007, 39, 2015, 316–330 doi: 10.1093/femsre/fuv007 Review Article REVIEW ARTICLE sRNA and mRNA turnover in Gram-positive bacteria 1 CNRS FRE 3630 (affiliated with Univ. Paris Diderot, Sorbonne Paris Cité), Institut de Biologie Physico-Chimique, 13 rue Pierre et Marie Curie, 75005 Paris, France and 2 Architecture et Réactivité de l’ARN, Université de Strasbourg, CNRS, IBMC, 15 rue René Descartes, F-67084 Strasbourg, France ∗ Corresponding author: Institut de Biologie Physico-chimique, 13 rue Pierre et Marie Curie F-75005 Paris, France. Tél: 158415123; E-mail: [email protected] One sentence summary: This review provides an update of the ribonucleases, their mechanisms of action and their roles in regulation in Gram-positive bacteria. Editor: Wolfgang Hess ABSTRACT It is widely recognized that RNA degradation plays a critical role in gene regulation when fast adaptation of cell growth is required to respond to stress and changing environmental conditions. Bacterial ribonucleases acting alone or in concert with various trans-acting regulatory factors are important mediators of RNA degradation. Here, we will give an overview of what is known about ribonucleases in several Gram-positive bacteria, their specificities and mechanisms of action. In addition, we will illustrate how sRNAs act in a coordinated manner with ribonucleases to regulate the turnover of particular mRNA targets, and the complex interplay existing between the ribosome, the ribonucleases and RNAs. Keywords: Gram-positive bacteria; ribonucleases; sRNA; mRNA stability INTRODUCTION Regulation of mRNA degradation is a key method of controlling gene expression to allow bacteria to adapt to their environmental conditions. The genome sequencing projects of the late 90s revealed significant differences in the degradation machineries of Gram-positive and Gram-negative bacteria (Condon and Putzer 2002). For example, RNase E which is the major ribonuclease involved in mRNA degradation in Escherichia coli is absent in Bacillus subtilis, the paradigm of Grampositive organisms (Fig. 1). Moreover, an RNase can be essential in one organism and not in another. It is therefore clear that many of the blueprints that have been established for both regulated and constitutive RNA turnover in the model enterobacterium E. coli have to be re-established in the Firmicutes. Recent studies have permitted a detailed characterization of the RNases present in B. subtilis that have been extended to other Gram-positive bacteria (reviewed in Condon and Bech- hofer 2011; Morrison and Dunman 2011; Jester, Romby and Lioliou 2012). The degradation of any specific mRNA is influenced by a number of factors including its secondary structure and its rate of translation. These parameters can be affected by RNA-binding proteins and, as has been shown more recently, by regulatory RNAs. Regulatory RNAs encoded in cis (antisense or asRNA) or in trans (small or sRNA) with respect to their targets are now recognized as important actors in the modulation of gene expression. Although these regulatory RNAs act through a wide variety of mechanisms, most of them described in the literature block translation and/or affect the stability of their mRNA targets (Lalaouna et al. 2013). The majority of these pioneering studies have been performed in Gram-negative bacteria such as E. coli or Salmonella typhimurium. In these bacteria, the Sm-like protein Hfq is required to protect the sRNA from degradation and to stimulate basepairing interactions between the sRNA and mRNA (e.g. De Lay and Gottesman 2011; Vogel and Luisi 2011; Régnier Received: 23 December 2014; Accepted: 1 March 2015  C FEMS 2015. All rights reserved. For permissions, please e-mail: [email protected] 316 Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 Sylvain Durand1 , Arnaud Tomasini2 , Frédérique Braun1 , Ciarán Condon1,∗ and Pascale Romby2 Durand et al. 317 and Hajnsdorf 2013; Sauer 2013; Wagner 2013). Hfq has also been proposed to recruit RNase E to the binding site of the sRNA and thus facilitate RNA degradation (Morita and Aiba 2011; Prevost et al. 2011). In Gram-positive bacteria, recent transcriptome data have shown that regulatory RNAs are also widely present, but Hfq does not seem to play a major role (Bohn, Rigoulay and Bouloc 2007; Dambach, Irnov and Winkler 2013; Hammerle et al. 2014), again forcing a re-thinking of an established paradigm. The differences in the mRNA degradation machinery (Fig. 1) and in the mode of action of regulatory RNAs have incentivized their characterization in Gram-positive bacteria to the level currently enjoyed in Gram-negative organisms. In this review, we will describe progress made in the determination of the mRNA degradation pathways in Gram-positive bacteria and how the RNases involved are influenced by the action of regulatory sRNAs. mRNA DEGRADATION IN GRAM-POSITIVE BACTERIA Escherichia coli and B. subtilis have different sets of endo- and exoribonucleases to degrade mRNAs (Fig. 1). A few of them including 3′ -5′ exoribonucleases (PNPase, RNase R, RNase PH) and several endoribonucleases (RNase III, RNase P, RNase Z) are conserved in Gram-negative as well as in low GC and high GC Grampositive bacteria (Fig. 2). In contrast, one of the main RNases (the endoribonuclease Y) involved in mRNA degradation in the low GC Gram-positive bacteria is absent from high GC Gram-positive bacteria such as Mycobacterium tuberculosis, which instead encode the endoribonuclease E/G (Figs 1 and 2). Below we will focus on the major enzymes required for the degradation of mRNAs in Gram-positive bacteria. Key enzymes involved in mRNA degradation in Firmicutes Ribonuclease Y The key endoribonuclease of E. coli mRNA turnover, RNase E, has been replaced by RNase Y in B. subtilis (Fig. 1). RNase Y was first shown to be responsible for the endonucleolytic cleavage between the cggR and gapA open reading frames of the gapA operon (Commichau et al. 2009) and in S-adenosylmethionine riboswitch turnover (Shahbabian et al. 2009) in B. subtilis. CggR encodes a transcriptional regulator of the gapA gene encoding glyceraldehyde 3-phosphate dehydrogenase (GAPDH). RNase Y cleavage of this mRNA allows differential expression of these two genes (100-fold more GAPDH than CggR). Depletion of RNase Y also increased the half-life of bulk RNA more than 2-fold in B. subtilis (Shahbabian et al. 2009), the first suggestion that RNase Y could have an important role in global mRNA degradation in this organism, equivalent to RNase E in E. coli. Interestingly, both enzymes have similar cleavage specificities (AUrich single-stranded regions) and are localized to the cytoplasmic membrane. In the case of RNase E, membrane targeting is through an amphipathic helix and the enzyme rapidly diffuses around the inner membrane to form short-lived foci, which have been attributed to transient RNA degradation bodies (Strahl et al. 2015). Such a dynamic clustering of RNase Y has not yet been observed in Gram-positive bacteria, which is targeted to membranes through an N-terminal transmembrane domain. Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 Figure 1. Ribonucleases and their functions. (A) Ribonucleases found in E. coli, B. subtilis and M. tuberculosis. Essential RNases are indicated in red. RNase III and Orn were only found essential for cell growth in some bacteria (see the text). In this review, only RNases involved in mRNA degradation have been discussed in detail, such as RNase J1/J2, RNase Y, PNPase and RNase III (see the text). (B) Activity (endo- vs exoribonucleolytic) and substrates of RNases in panel A. The toxins listed are all type II. 318 FEMS Microbiology Reviews, 2015, Vol. 39, No. 3 Like RNase E, RNase Y endoribonuclease activity is thought to be stimulated by a 5′ monophosphate, although this has only been documented so far for one substrate, the yitJ riboswitch (Shahbabian et al. 2009). While RNase Y was originally thought to be essential in B. subtilis (as RNase E is in E. coli), this was recently shown not to be the case (Figaro et al. 2013). Bacillus subtilis strains completely lacking RNase Y are viable, but they grow slowly and have pleiotropic phenotypes (failure to sporulate or become competent for DNA uptake, for example). The rny deletion mutant can be combined with rnc and pnp, but not rnjA (Figaro et al. 2013), suggesting that it is not possible to inactivate both major endo and 5′ -exonucleolytic pathways (see below). Three different global transcriptome analyses of RNase Y depleted B. subtilis strains showed that between 13 and 23% of individual protein-encoding genes and many non-coding RNA genes were upregulated, confirming a global role in mRNA turnover for this enzyme. In Staphylococcus aureus and Streptococcus pyogenes, the geneencoding RNase Y was first identified in a screen for mutants with attenuated virulence and called cvfA (Kaito et al. 2005). Although deletion mutants of cvfA have only slightly reduced growth rates in both cases compared to wild-type, transcriptome analyses have led to divergent conclusions as to the global importance of RNase Y in mRNA turnover in these two organisms. In S. pyogenes, deletion of the RNase Y/CvfA gene led to an up- regulation of a significant number (14%) of genes in stationary phase (Kang, Caparon and Cho 2010) and a 2-fold increase global mRNA half-life in late-log phase (Chen et al. 2013), suggesting a global role for RNase Y similar to that observed in B. subtilis. In S. aureus, however, only a small subset (4%) of mRNAs and sRNAs were upregulated in the absence of RNase Y (Marincola et al. 2012), suggesting that there may be some functional redundancy with another endoribonuclease in this organism. Although some Firmicutes, such as Listeria and Clostridia, do have an ortholog of the RNase E catalytic domain (RNase E/G) to potentially provide such functional redundancy, S. aureus is not one of these organisms (Fig. 2). In S. aureus, RNase Y is required for the processing and stabilization of the transcript encoding the global virulence regulatory system SaeRS, and was shown to activate the synthesis of exotoxins independently of the agr and sae pathways by an as yet undefined mechanism (Marincola et al. 2012). Ribonucleases J1 and J2 RNases J1 and J2 have been shown to form a 5′ -3′ exoribonuclease complex in B. subtilis involved in the 5′ -3′ degradation of mRNAs and in the maturation of the 5′ -end of 16S ribosomal RNA (Mathy et al. 2007). Indeed, RNase J1 was the first 5′ -3′ exoribonuclease identified in prokaryotes and its discovery explained how mRNA can be greatly stabilized by a stalled ribosome or a Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 Figure 2. The phylogenetic distribution of Gram-positive ribonucleases. The endonucleases are shown in the top row and the exonucleases in the bottom row of blocks for each species. The phylogenetic relationship between the different organisms was calculated by comparing 16S rRNA sequences using Clustal X. The phylogenetic tree was drawn using Phylodendron (http://iubio.bio.indiana.edu/soft/molbio/java/apps/trees/). The abbreviations for the different RNases are as follows: III = RNase III; P = RNase P; Z = RNase Z; M5 = RNase M5; G = RNase G; EG = RNase E/G; H1 = RNase HI; H2 = RNase HII; H3 = RNase HIII; Y = RNase Y; mIII = MiniRNase III; Maz = MazF/EndoA; Yhc = YhcR; Bsn = RNase Bsn; Bar = Barnase; R = RNase R; II = RNase II; Pnp = polynucleotide phosphorylase; PH = RNase PH; Orn = oligoribonuclease; T = RNase T; D = RNase D; Yha = RNase YhaM; J = RNase J; NrA = NanoRNase A; NrB = NanoRNase B. n Number potential orthologues present. This figure is an updated version of a figure shown in Condon and Putzer (2002). Durand et al. Ribonuclease III Almost all bacteria have the double-strand-specific endoribonuclease RNase III, with Deinococcus radiodurans being a notable exception (Drider and Condon 2004). In the archaea, RNase III activity has been replaced by bulge-helix-bulge nuclease, which has similar specificity (Heinemann, Soll and Randau 2010). The catalytic domain of RNase III is also found conserved in the two enzymes Dicer and Drosha, which are involved in siRNA and miRNA processing in eukaryotes (Jaskiewicz and Filipowicz 2008). Bacterial RNase III is primarily known for its role in the processing of rRNAs. However, the enzyme is non-essential in most organisms studied so far with the notable exception of B. subtilis, where RNase III is required to prevent expression of two prophage-encoded type I toxins via asRNA regulation. Depletion of RNase III has limited impact on mRNA expression levels in both B. subtilis and E. coli, with only 11–12% of mRNAs affected in either organism (Stead et al. 2011; Durand et al. 2012a). In B. subtilis, many of these effects were shown to be indirect and occur at the transcriptional level, so the actual number of mRNAs directly cleaved by RNase III is considerably smaller (Durand et al. 2012a). In S. pyogenes, RNase III is a key partner required for the RNA-mediated immunity CRISPR/Cas system against phages and plasmids (Deltcheva et al. 2011). A small sRNA called tracrRNA directs RNase III-mediated maturation of the short repeatspacer-derived crRNA to silence the foreign DNA in a sequencespecific manner. Co-immunoprecipitation of RNAs bound to RNase III identified a significant number of specific sRNA and mRNA substrates of this enzyme in S. aureus (Lioliou et al. 2012). In addition, S. aureus RNase III was shown to contribute to the correct maturation of rRNAs and tRNAs, to autoregulate its synthesis by cleaving the coding region of its own mRNA, to enhance the stability of cspA mRNA by cleaving in the 5′ UTR and to cut overlapping 5′ UTRs of divergently transcribed genes. In this organism, RNase III has also been proposed to play a role in the removal of background transcriptional noise from both strands of the chromosome (Lasa et al. 2011), a phenomenon not clearly seen in B. subtilis (Durand et al. 2012a) but recently observed in E. coli (Lybecker et al. 2014b). Whether these cryptic transcripts possess regulatory functions or are by-products of transcription events awaits further experimental work (Lasa, Toledo-Arana and Gingeras 2012; Lybecker, Bilusic and Raghavan 2014a). Interestingly, it was recently suggested that pervasive transcription might be considered as a genome surveillance mechanism for DNA damage, enabling a robust action of the nucleotide excision repair machinery (Kamarthapu and Nudler 2015). Studies performed in S. aureus and Listeria monocytogenes also revealed the unexpected but common presence of mRNAs with overlaps of either their entire length or over their 5′ or 3′ -untranslated regions, which were degraded in a RNase III-dependent manner (Lasa et al. 2011; Lioliou et al. 2012; Ruiz de los Mozos et al. 2013). In line with these observations, a new gene organization has been discovered in L. monocytogenes, the so-called excludon, where overlapping UTRs regulate the expression of divergent genes-encoding proteins with opposing functions (Wurtzel et al. 2012). Besides antisense regulation, RNase III plays also an important role in the regulation of some specific mRNAs by sRNAs in the Firmicutes, similar to observations in the Enterobacteria (see examples below). PNPase In 1996, Luttinger, Hahn and Dubnau (1996) performed a mini Tn-10 insertion screen to identify mutants of B. subtilis impaired in competence development. One of the genes identified was comR, which was renamed pnpA due to its strong sequence similarity to the PNPase gene of E. coli. Biochemical studies suggested that PNPase plays an important role in mRNA degradation in B. subtilis (Deutscher and Reuven 1991; Wang and Bechhofer 1996), and the degradation products of several mRNAs, such as ermC, rpsO or the trp RNA leader (Bechhofer and Wang 1998; Oussenko et al. 2005; Deikus and Bechhofer 2009) were shown to be dependent on PNPase. A recent global transcriptome analysis (Liu et al. 2014) confirmed that PNPase is an important player in global mRNA degradation in B. subtilis. This work showed that the steady-state levels of 10% of B. subtilis mRNAs (412 genes) were increased >1.5-fold in the pnpA mutant strain. Among these mRNAs, 178 were highly dependent on PNPase activity and none of the other 3′ exoribonucleases could replace PNPase in the degradation process. The number of direct mRNA targets of PNPase was probably underestimated for different reasons. First, only half of the B. subtilis genes were detected in the condition of the study. Second, PNPase is likely to have some functional redundancy with the three other 3′ -5′ exoribonucleases in the cell. Last, the effect of PNPase on some mRNAs might be masked if the first endoribonucleolytic cleavage, which Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 secondary structure near the 5′ end, even if the rest of the mRNA is completely free of ribosomes (Bechhofer and Zen 1989). Although both RNase J1 and J2 have been shown to additionally have endoribonuclease activity in vitro, they are thought to act primarily as a 5′ -3′ exoribonuclease complex in vivo, with RNase J1 providing most of the activity and RNase J2 helping to stabilize (or regulate) RNase J1 in both S. aureus (Linder, Lemeille and Redder 2014) and B. subtilis (Mathy et al. 2010). RNase J2 appears to play a more important role in S. mutans, where it has been proposed to act as an endonuclease independently of RNase J1 (Liu et al. 2015). Most Firmicutes have at least two RNase J paralogs, while outside of this phylum typically only one ortholog is observed per genome. In B. subtilis, RNase J1 was originally thought to be essential until it was shown that knockout of the gene was possible (Figaro et al. 2013). The lack of RNase J1 also causes pleiotropic effects in cell growth, cell morphology and in the development of competence and sporulation, while B. subtilis RNase J2 mutants grow normally. In S. pyogenes, both RNase J1 and J2 mutants are thought to be non-viable (Bugrysheva and Scott 2010), while in S. aureus and S. mutans, neither enzyme is essential, although in the former growth is restricted to a very narrow window around 37◦ C (Linder, Lemeille and Redder 2014; Chen et al. 2015; Liu et al. 2015). Synthetic lethality of the B. subtilis rnjA deletion has been tested with a number of the other key RNase mutants: rnjA rnc mutants are viable (unpublished results); while we were unable to make rnjA pnp or rnjA rny double mutants (Figaro et al. 2013). The inability to delete both the pnp and rnjA genes in the same strain suggests that cells cannot function in the absence of one or other of the major exonucleolytic (5′ or 3′ ) pathways. Bacillus subtilis cells severely depleted for RNase J1 show increased abundance of about 21% of its transcripts (Durand et al. 2012a), considerably more than an earlier study performed under milder depletion conditions (Mader et al. 2008). A global role for the exoribonuclease activity of RNase J1 in mRNA turnover was also seen in S. aureus, as well as for the correct maturation of the 5′ end of both the 16S rRNA and the RNA subunit of RNase P (Linder, Lemeille and Redder 2014). Degradation of a number of specific mRNAs was shown to be dependent on both RNase J1 and J2 in S. pyogenes (Bugrysheva and Scott 2010), but a global analysis of the role these enzymes in mRNA decay has not yet been performed. 319 320 FEMS Microbiology Reviews, 2015, Vol. 39, No. 3 Enzymes involved in mRNA degradation in Actinobacteria Although the degradation machinery is remarkably different between low GC Gram-positive and Gram-negative bacteria, Grampositive bacteria with a high GC content such as the Actinobacteria M. tuberculosis and M. smegmatis have a mixture of RNases found in both types of bacteria (Figs 1 and 2). Indeed, the Actinobacteria encode like E. coli, RNase E/G, RNase D, oligo-RNases (Orn) and a plethora of type II toxins with RNase activity. They also have the 5′ -3′ exoribonuclease J and the nano-RNase (NrnA) found in B. subtilis (Taverniti et al. 2011; Grosjean et al. 2014). RNase E/G is essential in M. tuberculosis while RNase J mutants are viable, suggesting a broader role of RNase E in mRNA degradation than RNase J, even if this hypothesis needs to be confirmed (Taverniti et al. 2011). RNase III and RNase J have been given particular attention in Streptomyces species because of their impact on the expression of genes involved in secondary metabolism and antibiotic production. The mechanisms have not yet been clearly established and tend to suggest indirect effects (Gatewood et al. 2012; Lee, Gatewood and Jones 2013; Jones et al. 2014). RNase III is, however, responsible for rRNA maturation in Streptomyces and this was suggested to be critical for the translation of large multicistronic mRNA transcripts (for a review, see Liu et al. 2013). Pathways of mRNA degradation in Firmicutes Primary mRNA transcripts in bacteria are protected at their 5′ end by a triphosphate group, known to inhibit the activity of many RNases (both endo- and exoribonucleases) involved in mRNA degradation (Fig. 3). They are also typically protected at their 3′ -end by a Rho-independent terminator that blocks 3′ -5′ exoribonuclease attack. Initiation of mRNA degradation must therefore override one of these two protective elements. Based on global transcriptome data and detailed studies on individual mRNAs, several different pathways of mRNA degradation have been proposed in Gram-positive bacteria. One degradation initiation pathway is similar to that found in Gram-negative bacteria but with different enzymes. In this case, the limiting step is an endoribonuclease cut to render the mRNA accessible to exoribonucleases or yet other endoribonuclease cleavages. In B. subtilis and other Firmicutes, this endoribonuclease can be RNase Y, specific for single-stranded RNA, or to a lesser extent RNase III, which cleaves double-stranded RNA (Fig. 3). These RNases leave a 5′ monophosphate extremity which can then be attacked by the 5′ -3′ exoribonuclease RNase J1/J2 complex or cleaved further by RNase Y. The 3′ -end of the upstream cleavage product is degraded by 3′ -5′ exoribonucleases, principally PNPase in B. subtilis (Fig. 3). In the alternative degradation pathway, the 5′ triphosphate of the mRNA can be converted to a 5′ -monophosphate by an RNA pyrophosphohydrolase (RppH) (Hsieh et al. 2013), first discovered in E. coli (Deana, Celesnik and Belasco 2008). After removal of the triphosphate, the mRNA can be degraded either directly by the 5′ -3′ exoribonuclease RNase J1/J2 complex or potentially be subjected to stimulated cleavage by RNase Y (Fig. 3). Analysis of the growth phenotypes of rppH rny and rppH rnjA double mutant strains compared to the respective single RNase mutants suggested that RppH may preferentially act in the same degradation pathway as RNase J1. Indeed, the rppH rnjA double has the same doubling time as a single rnjA mutant, whereas the doubling time of the rppH rny double mutant is significantly greater than that of the rny strain alone (Figaro et al. 2013). Importantly, not all mRNAs are substrates for RppH. In B. subtilis, this enzyme has a strong preference for a G-residue in the second position of its mRNA substrate and requires at least three unpaired nucleotides (nts) to act efficiently (Hsieh et al. 2013; Piton et al. 2013). Preliminary data also suggest that other RppH exist in B. subtilis that could affect a larger number of mRNAs (Hsieh et al. 2013; Condon et al. unpublished results). Protein complexes involved in RNA degradation Messenger RNA degradation can be performed by RNases in complex with other proteins, as has been demonstrated with the discovery of the RNase E-based degradosome in E. coli. These interactions are known to optimize mRNA degradation through allosteric activation in some cases and, in others, concentrate different activities in the same subcellular location. Several RNase Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 allows access to PNPase, occurs early in the mRNA. In such cases, the average signal over the whole transcript might not be significantly affected. A role of PNPase in regulating virulence gene expression in S. aureus has been observed recently (Numata et al. 2014). Surprisingly, the phenotype of the mutant strain deficient in RNase Y, i.e. decreased hemolysin production, was suppressed by disrupting the pnpA gene. It has been suggested that RNase Y, in addition to being an endoribonuclease, can convert RNAs bearing 2′ 3′ cyclic phosphate groups (resulting from cleavage by toxin-type RNases) into 3′ -phosphorylated RNAs, which are far more resistant to PNPase activity than RNAs with 2′ -3′ cyclic phosphates. The authors thus propose a model whereby RNase Y and PNPase competitively interfere with the degradation of some mRNAs involved in virulence (Numata et al. 2014). An alternative explanation can be imagined, however. If the initial destabilizing cleavage by RNase Y occurred in the 3′ -UTR, the lack of degradation by PNPase might stabilize the cleaved RNA sufficiently to yield a protein product. As in E. coli, PNPase from B. subtilis is sensitive to RNA secondary structure (Deikus and Bechhofer 2007). Bacillus subtilis has at least three more 3′ -5′ exoribonucleases that help to degrade mRNAs: RNase PH acts by phosphorolysis like PNPase, while RNase R and YhaM are hydrolytic enzymes. RNase R has been shown to help PNPase to degrade structured RNAs (Oussenko et al. 2005). Moreover, E. coli PNPase activity can be helped both by action of the helicase RhlB in the context of the degradosome and by the polyadenylation of the 3′ end of mRNA (Khemici et al. 2004) synthesized either by the poly(A) polymerase or by PNPase. Polyadenylation of mRNA has also been observed in B. subtilis, but the enzyme(s) responsible for this phenomenon and its significance are unknown (Cao and Sarkar 1993; Campos-Guillen et al. 2005). Interestingly, a recent genetic screen in E. coli has shown that PNPase plays an important role in sRNA regulation by protecting these molecules from degradation by RNase E or other ribonucleases, by binding without degradation (De Lay, Schu and Gottesman 2013). PNPase was also shown to be responsible for sRNA degradation when they are not associated with Hfq (Andrade et al. 2012). Nothing is known in Gram-positive bacteria about the role of PNPase in sRNA regulation. However, recent work in L. monocytogenes revealed a unique CRISPR-like sRNA (clustered regularly interspaced short palindromic repeats) whose DNA interference activity and RNA turnover depend on PNPase, although the mechanism is not yet fully understood (Sesto et al. 2014). Durand et al. 321 complexes have been proposed in the Firmicutes, structured around enzymes other than RNase E (Fig. 3, insert). The RNA degradosome In E. coli, and in other Enterobacteria, the RNA degradosome is organized around RNase E, which has an N-terminal catalytic domain (∼500 amino acids) and an equally large intrinsically unfolded C-terminal domain, peppered with structured microdomains that can interact with accessory proteins (Aı̈tBara, Carpousis and Quentin 2014). Escherichia coli RNase E principally binds PNPase, the DEAD-box RhlB helicase and the glycolytic enzyme enolase. However, the composition of the RNase E-based degradosome shows some degree of plasticity from organism to organism and from one growth condition to another. The Caulobacter crescentus degradosome, for example, contains aconitase instead of enolase (Hardwick et al. 2011). In E. coli, RhlB, CsdA and RhlE helicases are interchangeable in vitro (Khemici et al. 2004), and RhlB can be replaced by CsdA in the cold (Prud’homme-Genereux et al. 2004). Furthermore, a number of proteins have been shown to bind non-stoichiometrically to the degradosome in E. coli, to modulate or regulate its activity. In the high GC Gram-positive Actinobacterium, Streptomyces coelicolor, the PNPase interacting domain is found at the N-terminus of the RNase E (Lee and Cohen 2003), while the catalytic domain occupies a central location in the protein. In M. tuberculosis, RNase E was found to interact with an inor- ganic polyphosphate/ATP-NAD kinase (Ppnk), an acetyltransferase and GroEL but the meaning of these interactions is unknown (Kovacs et al. 2005). In B. subtilis, a degradosome-like complex has been proposed, structured primarily around RNase Y (Fig. 3, insert). A number of pairwise interactions were identified by bacterial two-hybrid (B2H) assay, cross-linkage followed by strep-tagged pull-down assay and surface plasmon resonance (SPR) analysis. Like RNase E, RNase Y has an intrinsically unstructured domain, in this case from amino acids 30 to 200 that also have a predicted propensity to form a coiled-coil. RNase Y principally forms dimers in vitro. Although the coiled-coil domain and the transmembrane domains form the strongest self-interaction in B2H assays, it is clear that all domains, the KH domain, the HD domain and the C-terminal domain of RNase Y contribute to its dimerization (Lehnik-Habrink et al. 2011). Only the full-length RNase Y protein gives a positive B2H reaction with its proposed partners below (Lehnik-Habrink et al. 2011). In this review, we will only deal with the more convincing of these interactions. A reciprocal interaction was first seen between B. subtilis RNase Y and enolase by B2H assay. Although a positive interaction between enolase and the negative leucine-zipper control was seen in the same test and many laboratories systematically eliminate enolase as a false positive in two-hybrid assays because of its high abundance in the cell, the interaction between RNase Y and enolase was further supported by Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 Figure 3. A schematic view of the pathways involved in RNA degradation. Insert: a degradosome-like complex has been proposed in B. subtilis structured primarily around RNase Y (Lehnik-Habrink et al. 2011). J1 and J2 are dual endo- and 5′ -3′ exo-ribonucleases, PNP is the 3′ exoribonuclease, Eno is for enolase and CshA is the major RNA helicase associated with the degradosome. The organization of this degradosome-like complex seems to be also conserved in S. aureus (Roux, DeMuth and Dunman 2011). Primary mRNA transcripts in bacteria are protected at their 5′ -end by a tri-phosphate group. Initiation of mRNA degradation can involve an endoribonuclease (RNase Y or RNase III) cut, which is the limiting step. This step generates a 5′ monophosphate extremity, which can be attacked by the 5′ -3′ exoribonuclease, RNase J1/J2 complex or cleaved further by RNase Y. The 3′ -end of the upstream cleavage product is degraded by 3′ -5′ exoribonucleases, principally PNPase in B. subtilis. In the alternative degradation pathway, the 5′ tri-phosphate of the mRNA can be converted to a 5′ -monophosphate by an RppH. After removal of the tri-phosphate, the mRNA can be degraded by the 5′ -3′ exoribonuclease RNase J1/J2 complex or by RNase Y. 322 FEMS Microbiology Reviews, 2015, Vol. 39, No. 3 The RNase J1/J2 complex RNase J1 and RNase J2 are present in similar numbers (2500– 3000 molecules cell−1 ) in B. subtilis and copurify in stoichiometric amounts from B. subtilis or when co-expressed in E. coli, suggesting that the RNase J1/2 complex is the primary form of these enzymes in vivo (Mathy et al. 2010). The complex has been confirmed by strong positive interactions in numerous B2H and yeast two-hybrid (Y2H) assays in both B. subtilis and S. aureus (Commichau et al. 2009; Mathy et al. 2010; Roux, DeMuth and Dunman 2011). The complex consists primarily of heterotetramers at high concentrations in vitro, but easily dissociate to heterodimers, which maybe the more relevant form at physiological concentrations. As mentioned above, the primary role of B. subtilis and S. aureus RNase J2 appears to stabilize or modu- late the activity of RNase J1 (Mathy et al. 2010; Linder, Lemeille and Redder 2014; Gilet et al. 2015). There has been also some controversy about a direct interaction of RNase J1 with RNase Y in the proposed degradosome complex. An interaction has been proposed based on B2H and cross-linked pull-down assay (Commichau et al. 2009; Lehnik-Habrink et al. 2011). However, extensive Y2H (RNase J1 as bait) or Y3H screens (RNase J1/J2 complex as bait) failed to identify any interacting partners other than RNase J1 and J2 (Mathy et al. 2010). Furthermore, no evidence for additional interacting partners was seen in either a FLAGtagged pull-down assay of RNase J1 (Mathy et al. 2010) or in direct measurements of a potential interaction between RNase Y and RNase J1 by SPR (Newman et al. 2012). Finally, no interaction was seen in B2H assays of S. aureus RNase Y and either RNase J1 or J2. All these negative results lead us to believe that RNase J1/J2 and the putative RNase Y-based degradosome act as independent complexes in the Firmicutes (Fig. 3, insert). MODULATION OF mRNA DEGRADATION BY NON-CODING RNAS (sRNA AND asRNA) To initiate degradation, RNases require access to the mRNA and have to deal with secondary structure, RNA-binding proteins and, particularly, translating ribosomes that potentially obscure RNase cleavage sites. An obvious strategy to avoid competition with ribosomes is to block translation. Regulatory RNA seems to be the perfect player for this role (Figs 4 and 5). Indeed, small regulatory RNAs, and in particular those acting in trans (sRNAs) often block translation by basepairing interactions with the Shine-Dalgarno (SD) sequence of mRNA targets. The mRNA free of ribosomes can then be degraded by ribonucleases. Numerous sRNAs seem to act in this way in Gram-positive bacteria and modulate mRNA degradation indirectly (Brantl and Bruckner 2014). Here, we present specific cases where links between the degradation machinery and sRNAs have been revealed (Figs 4 and 5). For a complete list of sRNAs studied in the Firmicutes and their modes of action, see the recent review (Brantl and Bruckner 2014). We will refer to small non-coding RNAs that form imperfect duplexes with their mRNA targets, as sRNAs, and to antisense RNAs, which form fully complementary interactions with the mRNAs since they are encoded on the same gene locus, as asRNAs. Modulation of mRNA degradation via modification of mRNA translation: destabilization Listeria monocytogenes LhrA sRNA LhrA was identified in L. monocytogenes due to its binding to Hfq. The half-life of LhrA is >30 min in wild-type strains and is decreased to <3 min in a hfq mutant strain (Christiansen et al. 2004). This sRNA was shown to post-transcriptionally regulate the lmo0850 mRNA, encoding a small peptide of unknown function, by basepairing interactions close to the ribosome binding sites (RBS). Toeprint and ß-galactosidase assays showed that LhrA blocks translation initiation and also reduces the level of lmo0850 mRNA (Nielsen et al. 2010) (Fig. 4A). The RNases involved in either the turnover of the LhrA sRNA or the target lmo0850 mRNA have not yet been characterized. Interestingly, L. monocytogenes has both a short form RNase E homolog and RNase Y, as well as the 5′ -3′ exoribonucleases J1/J2. The RNase E/G protein of Listeria lacks the C-terminal domain where Hfq is thought to bind in E. coli. This observation raises the question of whether Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 cross-linked pull-down assay (Lehnik-Habrink et al. 2011) and by SPR analysis (Newman et al. 2012). The RNase Y-enolase interaction seems to be conserved in S. pyogenes as demonstrated by cross-linked pull-down assay (Kang, Caparon and Cho 2010) and in S. aureus by B2H (Roux, DeMuth and Dunman 2011). A reciprocal B2H interaction was also observed between B. subtilis RNase Y and both PNPase (Commichau et al. 2009) and the CshA RNA helicase (Lehnik-Habrink et al. 2010). The RNase Y-PNPase interaction was confirmed by SPR, and the equilibrium dissociation constant is in the nanomolar range (Newman et al. 2012). The specificity of the RNase Y-CshA interaction was also confirmed in a domain swapping experiment with CshB helicase, that showed that the C-terminal domain of CshA is responsible for the protein–protein interaction (Lehnik-Habrink et al. 2010). Although it has not been shown that RNase Y can interact with more than one of these partners at a time, it is nonetheless an intriguing case of convergent evolution that B. subtilis and S. aureus can potentially form an RNase Y-based degradosome containing enolase, PNPase and the CshA RNA helicase. The stoichiometry of such a complex (unlike the E. coli degradosome, the partners of the putative B. subtilis complex have only been seen in Western blots), and whether any of these interactions are functionally relevant, remained to be experimentally determined. Although weak interaction between CshA and RNase Y was detected by B2H assay in S. aureus, no evidence for a direct interaction between PNPase and RNase Y was observed, suggesting that complex formation varies within the Firmicutes (Roux, DeMuth and Dunman 2011). CshA has additionally been proposed to interact with a number of other partners in B2H assays and/or cross-link pull-down assays. These include phosphofructokinase, enolase, PNPase, RNase J1, and the DEAD box helicases CshB and DeaD/YxiN in B. subtilis (Lehnik-Habrink et al. 2010) and the protein subunit of RNase P (RnpA) in S. aureus (Olson et al. 2011; Roux, DeMuth and Dunman 2011). This is quite a large number of potential partners for the 55 kDa CshA protein and one wonders whether some of the weaker positive interactions, seen in the B2H assays or in cross-linked pull-down assays, are not simply tethered through RNA. A number of additional minor pairwise interactions identified by these techniques have led authors to present models of large degradosome complexes in B. subtilis with up to eight interacting protein partners that seem very premature (Lehnik-Habrink et al. 2010; Lehnik-Habrink et al. 2011; Roux, DeMuth and Dunman 2011). Studies performed on S. aureus CshA have suggested that the RNA helicase contributes to mRNA degradation (Oun et al. 2013). Indeed, deletion of the cshA gene resulted in dysregulation of biofilm formation and hemolysis due to the dysregulation of agr mRNA stability, encoding the quorum-sensing system (Oun et al. 2013). Durand et al. 323 and how Hfq interacts with RNase E/G or other RNases in this organism. Bacillus subtilis FsrA sRNA The FsrA sRNA was identified in B. subtilis and is involved in the iron-sparing response (Gaballa et al. 2008). The level of several mRNAs encoding genes linked to central metabolism, such as citB and sdhCAB, are upregulated in an FsrA mutant strain (Smaldone et al. 2012). The ribonucleases involved in these regulatory phenomena are still unknown. Interestingly, FsrA requires a number of small basic proteins (FbpA, B and C) to act efficiently on its targets. For example, the regulation of the lutABC operon by FsrA requires the small protein FbpB for full repression. The authors proposed that FbpA, B and C could fulfill the role of Hfq to stimulate basepairing between FsrA and lutABC and possibly even attract the degradation machinery. In particular, the FbpB protein plays a greater role than FsrA in regulating the levels of the lutABC mRNA (Smaldone et al. 2012). The mode of action of FsrA is still speculative but its potential interaction with the RBS of the lutA, citB and sdhCAB mRNAs suggested that FsrA would block their translation to provoke mRNA degradation (Fig. 4A). RsaE/RoxS sRNA RsaE was first identified in S. aureus and this is the sole transacting sRNA to be conserved in B. subtilis, apart from the ubiquitous 6S RNA (Geissmann, Marzi and Romby 2009; Bohn et al. 2010). This RNA regulates several targets linked to folate and central metabolism in S. aureus (Geissmann, Marzi and Romby 2009; Bohn et al. 2010). In B. subtilis, the RsaE homolog (which is called RoxS) also regulates some targets linked to central metabolism, but plays an even greater role in controlling genes involved in oxidative stress or redox homeostasis in response to nitric oxide (NO) (Durand et al. 2015). In most cases studied, RsaE/RoxS is predicted to bind the SD sequences of target mRNAs in S. aureus and B. subtilis leading to translational inhibition (Fig. 4A). However, this may reflect a bias in target prediction programs that often search for basepairing interactions around translation initiation sites. In B. subtilis, RNase Y and RNase III are important enzymes involved in the regulation of target mRNAs stability in response to RoxS binding (Durand et al. 2015). More unexpectedly, RNase Y intervenes at an additional level by processing the 5′ end of RoxS removing about 20 nts. Processing of RoxS was shown to expand the repertoire of targets recognized by this sRNA in B. subtilis (Durand et al. 2015). This study reveals a complex interplay between RNases and RoxS to regulate its mRNA targets. Modulation of mRNA degradation via modification of mRNA translation: stabilization Under specific conditions, sRNAs can also stabilize mRNAs using a variety of mechanisms. The most common mechanism involves a conformational change of the mRNA structure upon binding of the sRNA that liberates the RBS to recruit the ribosome and to enhance translation (Fig. 4B). The activation of translation then protects the mRNA from degradation. In a second mechanism, binding of the sRNA creates or reveals a specific processing site of the mRNA, which renders the RBS accessible for ribosome binding and stabilizes the mRNA (Fig. 4B). The VRRNA in Clostridium perfringens and RNAIII in S. aureus use these strategies and will be described below. Clostridium perfringens VR-RNA The VR-RNA is a 386-nucleotide sRNA encoded in the genome of C. perfringens (Obana et al. 2010). This sRNA binds the 5′ UTR of the colA mRNA, encoding a collagenase. Binding of the sRNA triggers cleavage of the colA mRNA at the 3′ edge of the VRRNA binding site, which has two consequences: (1) it renders the SD sequence accessible to the ribosome and (2) it creates a stable stem-loop structure at the 5′ end of the mRNA with no 5′ Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 Figure 4. Various ways that small RNAs (sRNAs) regulate mRNA stability. (A and B) Degradation can be a consequence of the effect of sRNA on translation. (A) The repression of translation is often subsequently followed by rapid mRNA degradation while (B) the activation of mRNA translation protects the mRNA from RNases. (C) Binding of sRNA to mRNA recruits a specific RNase to destabilize the mRNA. (D) Binding of sRNA can induce a specific mRNA processing site in the 5′ UTR that leads to stabilizing effect. Alternatively, binding of the sRNA (or regulatory mRNA) can prevent the access of an RNase to prevent the degradation of target mRNA. 324 FEMS Microbiology Reviews, 2015, Vol. 39, No. 3 Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 Figure 5. Mechanisms of sRNA-mediated regulation in various Gram-positive bacteria. (A) Clostridium perfringens VR-RNA binds the 5′ UTR of the colA mRNA, encoding a collagenase and triggers cleavage of the colA mRNA, which in turn stabilizes the mRNA (Obana et al. 2010). The sRNA is in blue, the mRNA target is colored in black, SD is for Shine-Dalgarno sequence, 70S is the ribosome, RNases are represented by scissors. (B) Staphylococcus aureus RNAIII represses the translation of rot mRNA by sequestering the SD sequence, and activates the translation of hla mRNA through mRNA conformational changes (Novick 2003). (B) Streptococcus pyogenes FasX sRNA forms basepairing interactions with the first 9 nts of the ska mRNA to create a stable RNA helix at the 5′ end of the mRNA (Ramirez-Pena et al. 2010). (D) The 5′ UTR of the S. mutans irvA mRNA basepairs with the coding sequence of gbpC mRNA to block cleavage by RNase J2 (Liu et al. 2015). (E) Bacillus subtilis type I toxin/antitoxin system TxpA/RatA (Silvaggi, Perkins and Losick 2005). The 3′ ends of RatA forms a large duplex with the txpA mRNA, which is cleaved by RNase III. (F) Bacillus subtilis BsrG/SR4 is a temperature-dependent type I toxin-antitoxin system (Jahn et al. 2012). The basepairing interactions between SR4 and bsrG mRNA (in blue) lead to structural changes of the mRNA around the SD, preventing the ribosome (70S) binding. (G) The short asRNA, SprA1-AS, binds to the RBS of SprA1 through imperfect basepairing interactions to prevent translation of a small toxic peptide (Sayed, Jousselin and Felden 2011). The Rho-independent terminator hairpin of SprA1-AS (in blue), which is fully complementary to the 3′ end of SprA1, does not interact with SprA1. Durand et al. overhang (Figs 4B and 5A). These modifications allow stabilization of the colA mRNA. Indeed, the half-life of the mRNA is four minutes in a wild-type strain but decreases to less than two minutes in a VR-RNA strain (Obana et al. 2010). This study also show that ribosome binding is essential to protect the colA mRNA from degradation. The identity of the RNase responsible for the cleavage after VR-RNA binding is still unknown but RNase III is not involved in this process (Obana et al. 2010). RNase Y could be a good candidate for this cleavage. Modulation of mRNA degradation without affecting translation In the Enterobacteria, the MicC sRNA binds the coding sequence of the ompD mRNA and directly promotes its degradation. It has been proposed that Hfq not only promotes the interaction between MicC and ompD, but also recruits RNase E to this site to activate mRNA degradation, without inhibiting translation (Pfeiffer et al. 2009). It has been further proposed that a 5′ monophosphate group on the MicC sRNA can stimulate cleavage of ompD mRNA by RNase E (Bandyra et al. 2012). Such a di- rect role of sRNA on mRNA stability, without interfering with translation cannot be excluded in Gram-positive bacteria (Fig. 4C and D). Indeed, five regulatory RNAs (three sRNAs, FasX, IrvA, Psm-mec and two antisense RNAs, RatA and SR4) are able to influence mRNA degradation without affecting translation (see below). Different players are presumably involved in these regulatory events, since RNase E is absent from most Firmicutes and Hfq seems to play a less important role in this type of regulation (Geisinger et al. 2006; Heidrich et al. 2006; Bohn, Rigoulay and Bouloc 2007; Boisset et al. 2007; Gaballa et al. 2008; Hammerle et al. 2014). Streptococcus pyogenes FasX sRNA The FasX sRNA was identified in S. pyogenes (Ramirez-Pena et al. 2010). One of the identified targets of FasX is the ska mRNA, encoding the virulence factor streptokinase. The ska mRNA is more stable in a wild-type strain than in a fasX strain. FasX forms basepairing interactions with the first 9 nts of the ska mRNA to create a stable RNA helix directly at the 5′ end of the mRNA (Figs 4D and 5C). One prediction is that this conformation blocks access to both the 5′ -3′ exoribonuclease activity of RNase J1, which requires a single-stranded 5′ -extension of at least 5 nts to gain access to the mRNA. RppH would also be predicted to be inhibited by this double-stranded conformation of the 5′ end, assuming that it has a similar specificity (at least three singlestranded residues with G at position 2) to B. subtilis RppH (Piton et al. 2013). When FasX is absent, the model would predict that the 5′ triphosphate of the ska mRNA, which starts with four consecutive G-residues, would be removed by S. pyogenes RppH and the mRNA would then be attacked by RNase J1. Although this model has not yet been tested, it has been shown that neither RNase Y nor PNPase is involved in this process (Ramirez-Pena et al. 2010) (Fig. 5C). Streptococcus mutans irvA regulatory RNA The irvA gene encodes a putative transcriptional regulator, originally thought to repress the so-called dextran-dependent aggregation (DDAG) stress response in S. mutans. However, a recent study has shown that, in fact, it is the 5′ UTR of the irvA mRNA (and not the repressor encoded by the ORF) that is responsible for the DDAG minus phenotype observed in irvA deletion strains. The 5′ UTR of the fully intact irvA mRNA behaves as a transacting regulatory RNA that stabilizes the gbpC mRNA about 10fold (Fig. 5D). GbpC is the key surface-exposed lectin responsible for the DDAG+ phenotype under stress conditions. The 5′ UTR of irvA interacts with the coding sequence of gbpC (about 110 nt downstream of the GbpC initiation codon) and blocks cleavage by RNase J2 at this site and apparently another about 100 nts further downstream (Liu et al. 2015). This system is a very nice example of a dual function messenger and regulatory RNA (reminiscent of S. aureus RNAIII, where the 3′ UTR provides the regulatory function). Staphylococcus aureus Psm-mec sRNA The Psm-mec sRNA, like irvA and RNAIII, is a dual-function RNA in S. aureus. It encodes a cytolysin of the phenol-soluble modulin (PSM) and also acts as a regulatory RNA. Psm-mec mRNA binding causes a 2-fold decrease in the half-life of the agrA mRNA via RNase III cleavage (Fig. 4C; Kaito et al. 2013). Psm-mec mRNA also inhibits the synthesis of AgrA by binding to its coding sequence (more than 200 nts downstream of the start codon) but it is not clear whether this interaction is required to promote agrA degradation (Kaito et al. 2013). Depending on the strain of Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 Staphylococcus aureus RNAIII RNAIII is long RNA (514 nts) that was identified in S. aureus as one of the main intracellular effectors of the agr quorum-sensing system (Novick et al. 1993). This RNA encodes δ-hemolysin but also acts as a regulatory RNA. RNAIII possesses 14 hairpin motifs and three of them, all containing stretches of consecutive Cresidues, are used to fulfill its repressor activity. The expression of several genes is negatively regulated by RNAIII (spa, SA1000, rot, lytM and coa) using a shared mechanism, i.e. blocking translation upon basepairing interactions with the SD sequence (Boisset et al. 2007) (Figs 4A, B and 5B). Depending on the mRNA signals, the repressor RNAIII-mRNA complexes adopt different topologies. For instance, loop 13 of RNAIII binds to the SD sequence of the spa mRNA to form a long imperfect duplex while the two apical loops H7 and H14 of RNAIII form loop–loop interactions with two G-rich loops including the SD sequence of rot mRNA (Fig. 5B). In both cases, the complexes formed are accompanied by RNase III cleavages, leading to a functional inactivation of the mRNAs. In contrast, RNAIII also increases the level of the hla mRNA encoding α-hemolysin (Morfeldt et al. 1995; Novick 2003). Although RNAIII is most probably consumed together with its repressed mRNA targets (Boisset et al. 2007), the yield of RNAIII is sufficiently abundant at the late-exponential growth phase to activate the translation of hla mRNA. In this case, the 5′ part of RNAIII binds to the 5′ UTR of hla to impair the formation of a secondary structure which normally traps the SD sequence of the hla gene (Fig. 5B). The basepairing interactions with RNAIII stimulate the translation of hla mRNA, which is then presumably responsible for the increase in mRNA levels (Morfeldt et al. 1995). Whether RNase III cleaved hla mRNA bound to RNAIII to generate a shorter 5′ UTR of hla mRNA has not been analyzed. The enhanced hla mRNA levels are also partially explained by RNAIII-dependent repression of the rot mRNA, which encodes a transcriptional repressor of toxin genes, such as hla (Geisinger et al. 2006; Boisset et al. 2007). In this way, the quorum-sensing-dependent RNAIII is involved in a particular regulatory network motif, a double feedforward loop that behaves as a double selector switch to ensure fine-tuned coordination of the inverse expression of two sets of genes (adhesins and exotoxins), tight regulation and filtering of noisy signals (Nitzan et al. 2015). 325 326 FEMS Microbiology Reviews, 2015, Vol. 39, No. 3 Staphylococcus, the targets of the Psm-mec RNA appear to be different (Cheung et al. 2014). Bacillus subtilis bsrG/SR4 BsrG/SR4 is a temperature-dependent type I toxin/antitoxin system, similar to txpA/RatA, also identified in B. subtilis (Jahn et al. 2012). The bsrG mRNA is 294 nts in length and encodes a toxic hydrophobic peptide of 38 amino acids. Its expression is regulated by an asRNA (SR4) complementary to its 3′ -end. Like txpA/RatA, the bsrG/SR4 duplex is cleaved by RNase III (Figs 4C and 5F). The half-life of both RNAs also depends on RNase R and RNase Y, presumably when they are not basepaired with each other. The basepairing interactions between SR4 and bsrG lead to a secondary structure change of the bsrG mRNA around the RBS (Fig. 5F). This rearrangement extends the stem-loop structure encompassing the SD sequence from 4 to 8 nts, inhibiting bsrG translation (Jahn and Brantl 2013). It is not known whether this translation inhibition is necessary to allow RNase III cleavage further downstream. Modulation of translation without affecting mRNA degradation It should be noted that it is also possible for sRNAs to modulate translation without affecting mRNA degradation, as exemplified by the S. aureus SprA1/SprA1as type I toxin/antitoxin system. To avoid the toxicity of the peptide expressed from SprA1 during S. aureus growth, the stable sprA1 mRNA is repressed by high amounts of the unstable antitoxin SprA1as (Sayed et al. 2012). The ribonucleases involved in the degradation pathway of SprA1as are not known. Contrary to most type I antitoxin/toxin systems, SprA1as prevents translation of sprA1 through imperfect basepairing interactions explaining why RNase III does not cleave the SprA1–SprA1as duplex (Fig. 5G). Surprisingly, the functional domain of the asRNA does not involve the Rhoindependent terminator hairpin of SprA1as, which is fully complementary to the 3′ end of sprA1 (Sayed, Jousselin and Felden 2011). Instead, the binding region is located in its 5′ part that is CONCLUDING REMARKS The different studies presented here show that regulatory RNAs in several Gram-positive bacteria have an important role in modulating RNase access to specific mRNAs to target them for degradation as a function of growth conditions, as shown in Enterobacteriaceae. However, modulation of mRNA degradation by sRNA and the RNases involved in these processes are still largely unknown in Gram-positive bacteria and need to be further explored. RNase III seems to play an important role in regulating degradation via antisense RNA in several type I toxin/antitoxin systems and in various sRNA-mediated regulatory events in both S. aureus (RNAIII, RsaE, psm-mec) and B. subtilis (RoxS). Nevertheless, studies of other sRNAs (FasX, VR-RNA, IrvA) suggest that RNases Y, J1 and even J2 may also be important players in mRNA degradation promoted by sRNA. Because these RNases are thought to be able to form diverse complexes, this leads to the question of how these complexes might be involved in sRNAmediated mRNA degradation in vivo. Furthermore, the composition of these RNA degrading complexes might be different depending on growth conditions, as has been observed in E. coli. Hfq plays a central role in RNA-mediated regulation in E. coli by stabilizing sRNAs, by facilitating basepairing interactions with their targets and by stimulating mRNA degradation by RNase E (Bandyra et al. 2012). A similar mechanism could exist in Gram-positive bacteria but most of them have RNase Y in place of RNase E (Fig. 1). LhrA in L. monocytogenes is currently the sole sRNA from a Gram-positive organism requiring Hfq to stimulate basepairing interactions with its target lmo0850 (Christiansen et al. 2004). While other examples will undoubtedly be uncovered, especially in Listeria and Clostridium where Hfq appears to play a more important role than in S. aureus or B. subtilis, these observations make it unlikely that Hfq is the major factor mediating mRNA degradation by sRNA through RNase Y throughout the Firmicutes. It is interesting in this regard that Hfq from E. coli can be replaced by Hfq from either L. monocytogenes or C. difficile (Caillet et al. 2014). Both of these organisms have an RNase E/G homolog. In contrast, S. aureus and Borrelia burgdorferi do not have RNase E and their Hfq proteins are unable to fully complement a hfq strain of E. coli (Vecerek et al. 2008; Rochat et al. 2012). Thus, it could be interesting to determine whether there is any correlation between the interchangeability of Hfq and the degradation machinery present in a particular organism. Bacillus subtilis has numerous annotated RNA-binding proteins of which at least 14 have unknown functions. Moreover, the group of J. Helmann has shown that the small basic proteins FbpA, B and C play a role in regulation by the sRNA FsrA, suggesting that other RNA chaperones can substitute for Hfq in some Firmicutes. These proteins could also help to modulate RNase activity on target mRNAs as has been suggested for FbpB in B. subtilis (Smaldone et al. 2012). In S. aureus, several RNAbinding proteins such as the RNA helicase CshA (Oun et al. 2013) and the transcriptional regulatory factor SarA (Morrison et al. 2012) have been shown to play major regulatory roles in RNA metabolism, although their mechanisms of action are not yet fully understood. A conserved RNA-binding protein YbeY, which contains structural domain similar to the Agonaute protein, has been shown to regulate the accumulation of Hfq-dependent and -independent sRNAs and the target mRNAs in Sinorhizobium Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 Bacillus subtilis txpA/RatA The type I toxin/antitoxin system txpA/RatA was identified in B. subtilis. The antitoxin is an asRNA, which inhibits the expression of the small toxic peptide TxpA expressed from the opposite strand (Silvaggi, Perkins and Losick 2005). The 3′ ends of RatA and the txpA mRNA overlap by about 120 nts. Repression of TxpA toxin expression occurs via RNase III, which cleaves in the complementary region between txpA and RatA (Figs 4C and 5E). This cleavage is absolutely essential to silence TxpA expression. Indeed, an RNase III mutant is lethal due to the expression of this toxin and another called YonT, silenced using a similar mechanism (Durand, Gilet and Condon 2012b). In contrast to txpA, the half-life of RatA is not significantly affected by the RNase III deletion, which was surprising since this RNase cleaves generally both strands of the RNA duplex. This can be explained by the fact that RatA is synthesized in 15-fold excess over the txpA mRNA. In this context, most of RatA in the cell is not paired to txpA and follows a classical mRNA degradation pathway involving RNase Y cleavage around position +90, upstream of the basepairing region with txpA. After this cleavage, the downstream product is degraded by RNase J1 and the upstream product of RatA cleavage is attacked by PNPase (Durand, Gilet and Condon 2012b). This example is one of the best characterized system in terms of the RNases involved in mRNA degradation mediated by small asRNA. partially complementary to the RBS of sprA1 (Fig. 5G). Overproduction of SprA1as has no effect on sprA1 mRNA levels, suggesting that this system functions solely at the translational level. Durand et al. FUNDING This work was supported by funds from the CNRS (UPR 9073, UPR 9002), Université Paris VII-Denis Diderot (CC), Université de Strasbourg (PR), the Agence Nationale de la Recherche. This work has been published under the framework of two LABEX programs: ANR-Dynamo (CC) and ANR-10-LABX-0036 NETRNA (PR) that benefit from a funding from the state managed by the French National Research Agency as part of the Investments for the future program. Conflict of interest. None declared. REFERENCES Aı̈t-Bara S, Carpousis AJ, Quentin Y. RNase E in the gammaProteobacteria: conservation of intrinsically disordered noncatalytic region and molecular evolution of microdomains. Mol Genet Genomics 2014, DOI: 10.1007/s00438-014-0959-5. Andrade JM, Pobre V, Matos AM, et al. The crucial role of PNPase in the degradation of small RNAs that are not associated with Hfq. RNA 2012;18:844–55. Bandyra KJ, Said N, Pfeiffer V, et al. The seed region of a small RNA drives the controlled destruction of the target mRNA by the endoribonuclease RNase E. Mol Cell 2012;47: 943–53. Bechhofer DH, Wang W. Decay of ermC mRNA in a polynu- cleotide phosphorylase mutant of Bacillus subtilis. J Bacteriol 1998;180:5968–77. Bechhofer DH, Zen KH. Mechanism of erythromycin-induced ermC mRNA stability in Bacillus subtilis. J Bacteriol 1989;171:5803–11. Bohn C, Rigoulay C, Bouloc P. No detectable effect of RNAbinding protein Hfq absence in Staphylococcus aureus. BMC Microbiol 2007;7:1–10. Bohn C, Rigoulay C, Chabelskaya S, et al. Experimental discovery of small RNAs in Staphylococcus aureus reveals a riboregulator of central metabolism. Nucleic Acids Res 2010;38:6620–36. Boisset S, Geissmann T, Huntzinger E, et al. Staphylococcus aureus RNAIII coordinately represses the synthesis of virulence factors and the transcription regulator Rot by an antisense mechanism. Gene Dev 2007;21:1353–66. Brantl S, Bruckner R. Small regulatory RNAs from low-GC Grampositive bacteria. RNA Biol 2014;11:443–56. Bugrysheva JV, Scott JR. The ribonucleases J1 and J2 are essential for growth and have independent roles in mRNA decay in Streptococcus pyogenes. Mol. Microbiol 2010;75:731–43. Cahova H, Winz ML, Hofer K, et al. NAD capture Seq indicates NAD as a bacterial cap for a subset of regulatory RNAs. Nature 2014, DOI: 10.1038/nature14020. Caillet J, Gracia C, Fontaine F, et al. Clostridium difficile Hfq can replace Escherichia coli Hfq for most of its function. RNA 2014;20:1567–78. Campos-Guillen J, Bralley P, Jones GH, et al. Addition of poly(A) and heteropolymeric 3′ ends in Bacillus subtilis wild-type and polynucleotide phosphorylase-deficient strains. J Bacteriol 2005;187:4698–706. Cao GJ, Sarkar N. Poly(A) RNA in Bacillus subtilis: identification of the polyadenylation site of flagellin mRNA. FEMS Microbiol Lett 1993;108:281–5. Chen X, Liu N, Khajotia S, et al. RNases J1 and J2 are critical pleiotropic regulators in Streptococcus mutans. Microbiology 2015, in press. DOI: 10.1099/mic.0.000039. Chen YG, Kowtoniuk WE, Agarwal I, et al. LC/MS analysis of cellular RNA reveals NAD-linked RNA. Nat Chem Biol 2009;5:879– 81. Chen Z, Itzek A, Malke H, et al. Multiple roles of RNase Y in Streptococcus pyogenes mRNA processing and degradation. J Bacteriol 2013;195:2585–94. Cheung GY, Villaruz AE, Joo HS, et al. Genome-wide analysis of the regulatory function mediated by the small regulatory psm-mec RNA of methicillin-resistant Staphylococcus aureus. Int J Med Microbiol 2014;304:637–44. Christiansen JK, Larsen MH, Ingmer H, et al. The RNA-binding protein Hfq of Listeria monocytogenes: role in stress tolerance and virulence. J Bacteriol 2004;186:3355–62. Commichau FM, Rothe FM, Herzberg C, et al. Novel activities of glycolytic enzymes in Bacillus subtilis: interactions with essential proteins involved in mRNA processing. Mol Cell Proteomics 2009;8:1350–60. Condon C, Bechhofer DH. Regulated RNA stability in the Gram positives. Curr Opin Microbiol 2011;14:148–54. Condon C, Putzer H. The phylogenetic distribution of bacterial ribonucleases. Nucleic Acids Res 2002;30:5339–46. Dambach M, Irnov I, Winkler WC. Association of RNAs with Bacillus subtilis Hfq. PLoS One 2013;8:e55156. De Lay N, Gottesman S. Role of polynucleotide phosphorylase in sRNA function in Escherichia coli. RNA 2011;17:1172–89. De Lay N, Schu DJ, Gottesman S. Bacterial small RNA-based negative regulation: Hfq and its accomplices. J Biol Chem 2013;288:7996–8003. Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 meliloti (Pandey et al. 2014). Although this protein is conserved in Firmicutes, its proposed function as an RNase (or RNA chaperone) has not been studied in these bacteria. Approaches to fractionate and identify the various classes of sRNP might also provide some clues on the implication of the RNases and RNA chaperones in sRNA-dependent pathways. Many sRNAs, among them FsrA, RsaE and FasX, have C-rich regions (CRRs) predicted to interact with G-rich sequences such as SD elements in their mRNA targets in bacteria. At least 3 sRNAs in B. subtilis and 11 in S. aureus have these CRRs suggesting that these sRNAs belong to a family of regulatory RNAs. The CRR, in addition to representing a potential seed sequence for the interaction with mRNA targets, could represent a binding site for specific proteins and could play a role in sRNA stability, as it was seen for CRR found in the 3′ UTR of some mRNAs in eukaryotes (Makeyev and Liebhaber 2002). Indeed, unpublished results show that mutations in the some of the C-rich domains of B. subtilis RsaE/RoxS render it more unstable than the wild-type sequence, even though the secondary structure is predicted to be unaffected by these mutations (Durand et al. 2015). More recently, a C-rich sequence motif in the 3′ UTR of icaR mRNA was shown to affect the translation of its own mRNA most probably by favoring a 5′ -3′ UTR interaction and by recruiting RNase III to induce rapid degradation of the mRNA (Ruiz de los Mozos et al. 2013). Finally, in addition to non-coding RNA, the modulation of RNase activity or stability can also be achieved by a direct binding of particular proteins to RNases, e.g. RraA to RNase E (Lee et al. 2003; Zhao et al. 2006), by RNase post-translational modification, e.g. acetylation of RNase R (Liang, Malhotra and Deutscher 2011; Liang and Deutscher 2012) or by modification of the 5′ end of E. coli and Streptomyces RNAs, e.g. modification by NAD (Chen et al. 2009; Cahova et al. 2014). The potential for these types of regulation must also be considered in Gram-positive bacteria even if no specific examples have been characterized to date. 327 328 FEMS Microbiology Reviews, 2015, Vol. 39, No. 3 Jahn N, Brantl S. One antitoxin–two functions: SR4 controls toxin mRNA decay and translation. Nucleic Acids Res 2013;41:9870– 80. Jahn N, Preis H, Wiedemann C, et al. BsrG/SR4 from Bacillus subtilis—the first temperature-dependent type I toxinantitoxin system. Mol Microbiol 2012;83:579–98. Jaskiewicz L, Filipowicz W. Role of Dicer in posttranscriptional RNA silencing. Curr Top Microbiol Immunol 2008;320: 77–97. Jester BC, Romby P, Lioliou E. When ribonucleases come into play in pathogens: a survey of gram-positive bacteria. Int J Microbiol 2012;2012:592196. Jones SE, Leong V, Ortega J, et al. Development, antibiotic production, and ribosome assembly in Streptomyces venezuelae are impacted by RNase J and RNase III deletion. J Bacteriol 2014;196:4253–67. Kaito C, Kurokawa K, Matsumoto Y, et al. Silkworm pathogenic bacteria infection model for identification of novel virulence genes. Mol Microbiol 2005;56:934–44. Kaito C, Saito Y, Ikuo M, et al. Mobile genetic element SCCmecencoded psm-mec RNA suppresses translation of agrA and attenuates MRSA virulence. PLoS Pathog 2013;9:e1003269. Kamarthapu V, Nudler E. Rethinking transcription coupled DNA repair. Curr Opin Microbiol 2015;24:15–20. Kang SO, Caparon MG, Cho KH. Virulence gene regulation by CvfA, a putative RNase: the CvfA-enolase complex in Streptococcus pyogenes links nutritional stress, growthphase control, and virulence gene expression. Infect Immun 2010;78:2754–67. Khemici V, Toesca I, Poljak L, et al. The RNase E of Escherichia coli has at least two binding sites for DEAD-box RNA helicases: functional replacement of RhlB by RhlE. Mol Microbiol 2004;54:1422–30. Kovacs L, Csanadi A, Megyeri K, et al. Mycobacterial RNase Eassociated proteins. Microbiol Immunol 2005;49:1003–7. Lalaouna D, Simoneau-Roy M, Lafontaine D, et al. Regulatory RNAs and target mRNA decay in prokaryotes. Biochim Biophys Acta 2013;1829:742–7. Lasa I, Toledo-Arana A, Dobin A, et al. Genome-wide antisense transcription drives mRNA processing in bacteria. P Natl Acad Sci USA 2011;108:20172–7. Lasa I, Toledo-Arana A, Gingeras TR. An effort to make sense of antisense transcription in bacteria. RNA Biol 2012;9: 1039–44. Lee JH, Gatewood ML, Jones GH. RNase III is required for actinomycin production in Streptomyces antibioticus. Appl Environ Microb 2013;79:6447–51. Lee K, Cohen SN. A Streptomyces coelicolor functional orthologue of Escherichia coli RNase E shows shuffling of catalytic and PNPase-binding domains. Mol Microbiol 2003;48:349–60. Lee K, Zhan X, Gao J, et al. RraA. a protein inhibitor of RNase E activity that globally modulates RNA abundance in E. coli. Cell 2003;114:623–34. Lehnik-Habrink M, Newman J, Rothe FM, et al. RNase Y in Bacillus subtilis: a natively disordered protein that is the functional equivalent of RNase E from Escherichia coli. J Bacteriol 2011;193:5431–41. Lehnik-Habrink M, Pfortner H, Rempeters L, et al. The RNA degradosome in Bacillus subtilis: identification of CshA as the major RNA helicase in the multiprotein complex. Mol Microbiol 2010;77:958–71. Liang W, Deutscher MP. Post-translational modification of RNase R is regulated by stress-dependent reduction in the acetylating enzyme Pka (YfiQ). RNA 2012;18:37–41. Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 Deana A, Celesnik H, Belasco J. The bacterial enzyme RppH triggers messenger RNA degradation by 5′ pyrophosphate removal. Nature 2008;451:355–8. Deikus G, Bechhofer DH. Initiation of decay of Bacillus subtilis trp leader RNA. J Biol Chem 2007;282:20238–44. Deikus G, Bechhofer DH. Bacillus subtilis trp Leader RNA: RNase J1 endonuclease cleavage specificity and PNPase processing. J Biol Chem 2009;284:26394–401. Deltcheva E, Chylinski K, Sharma CM, et al.. CRISPR RNA maturation by trans-encoded small RNA and host factor RNase III. Nature 2011;471:602–7. Deutscher MP, Reuven NB. Enzymatic basis for hydrolytic versus phosphorolytic mRNA degradation in Escherichia coli and Bacillus subtilis. P Natl Acad Sci USA 1991;88:3277–80. Drider D, Condon C. The continuing story of endoribonuclease III. J Mol Microb Biotech 2004;8:195–200. Durand S, Braun F, Lioliou E, et al. A nitric oxide regulated small RNA controls expression of genes involved in redox homeostasis in Bacillus subtilis. PloS Genet 2015;11:e1004957. Durand S, Gilet L, Bessieres P, et al. Three essential ribonucleases-RNase Y, J1, and III-control the abundance of a majority of Bacillus subtilis mRNAs. PLoS Genet 2012a;8:e1002520. Durand S, Gilet L, Condon C. The essential function of B. subtilis RNase III is to silence foreign toxin genes. PLoS Genet 2012b;8:e1003181. Figaro S, Durand S, Gilet L, et al. 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. J Bacteriol 2013;195:2340–8. Gaballa A, Antelmann H, Aguilar C, et al. The Bacillus subtilis iron-sparing response is mediated by a Fur-regulated small RNA and three small, basic proteins. P Natl Acad Sci USA 2008;105:11927–32. Gatewood ML, Bralley P, Weil MR, et al. RNA-Seq and RNA immunoprecipitation analyses of the transcriptome of Streptomyces coelicolor identify substrates for RNase III. J Bacteriol 2012;194:2228–37. Geisinger E, Adhikari RP, Jin R, et al. Inhibition of rot translation by RNAIII, a key feature of agr function. Mol Microbiol 2006;61:1038–48. Geissmann T, Marzi S, Romby P. A search for small noncoding RNAs in Staphylococcus aureus reveals a conserved sequence motif for regulation. Nucleic Acids Res 2009;37:7239–59. Gilet L, DiChiara JM, Figaro S, et al. Small stable RNA maturation and turnover in Bacillus subtilis. Mol Microbiol 2015;95: 270–82. Grosjean H, Breton M, Sirand-Pugnet P, et al. Predicting the minimal translation apparatus: lessons from the reductive evolution of mollicutes. PLoS Genet 2014;10:e1004363. Hammerle H, Amman F, Vecerek B, et al. Impact of Hfq on the Bacillus subtilis transcriptome. PLoS One 2014;9:e98661. Hardwick SW, Chan VS, Broadhurst RW, et al. An RNA degradosome assembly in Caulobacter crescentus. Nucleic Acids Res 2011;39:1449–59. Heidrich N, Chinali A, Gerth U, et al. The small untranslated RNA SR1 from the Bacillus subtilis genome is involved in the regulation of arginine catabolism. Mol Microbiol 2006;62:520–36. Heinemann IU, Soll D, Randau L. Transfer RNA processing in archaea: unusual pathways and enzymes. FEBS Lett 2010;584:303–9. Hsieh PK, Richards J, Liu Q, et al. Specificity of RppH-dependent RNA degradation in Bacillus subtilis. P Natl Acad Sci USA 2013;110:8864–9. Durand et al. Nitzan M, Fechter P, Peer A, et al. A defense-offense multi-layered regulatory switch in a pathogenic bacterium. Nucleic Acids Res 2015;43:1357–69. Novick RP, Ross HF, Projan SJ, et al. Synthesis of staphylococcal virulence factors is controlled by a regulatory RNA molecule. EMBO J 1993;12:3967–75. Novick RP. Autoinduction and signal transduction in the regulation of staphylococcal virulence. Mol Microbiol 2003;48: 1429–49. Numata S, Nagata M, Mao H, et al. CvfA protein and polynucleotide phosphorylase act in an opposing manner to regulate Staphylococcus aureus virulence. J Biol Chem 2014;289:8420–31. Obana N, Shirahama Y, Abe K, et al. Stabilization of Clostridium perfringens collagenase mRNA by VR-RNA-dependent cleavage in 5′ leader sequence. Mol Microbiol 2010;77:1416– 28. Olson PD, Kuechenmeister LJ, Anderson KL, et al. Small molecule inhibitors of Staphylococcus aureus RnpA alter cellular mRNA turnover, exhibit antimicrobial activity, and attenuate pathogenesis. PLoS Pathog 2011;7:e1001287. Oun S, Redder P, Didier JP, et al. The CshA DEAD-box RNA helicase is important for quorum sensing control in Staphylococcus aureus. RNA Biol 2013;10:157–65. Oussenko IA, Abe T, Ujiie H, et al. Participation of 3′ -to-5′ exoribonucleases in the turnover of Bacillus subtilis mRNA. J Bacteriol 2005;187:2758–67. Pandey SP, Winkler JA, Li H, et al. Central role for RNase YbeY in Hfq-dependent and Hfq-independent small-RNA regulation in bacteria. BMC Genomics 2014;15:121–37. Pfeiffer V, Papenfort K, Lucchini S, et al. Coding sequence targeting by MicC RNA reveals bacterial mRNA silencing downstream of translational initiation. Nat Struct Mol Biol 2009;16:840–6. Piton J, Larue V, Thillier Y, et al. Bacillus subtilis RNA deprotection enzyme RppH recognizes guanosine in the second position of its substrates. P Natl Acad Sci USA 2013;110:8858–63. Prevost K, Desnoyers G, Jacques JF, et al. Small RNA-induced mRNA degradation achieved through both translation block and activated cleavage. Gene Dev 2011;25:385–96. Prud’homme-Genereux A, Beran RK, Iost I, et al. Physical and functional interactions among RNase E, polynucleotide phosphorylase and the cold-shock protein, CsdA: evidence for a ‘cold shock degradosome’. Mol Microbiol 2004;54: 1409–21. Ramirez-Pena E, Trevino J, Liu Z, et al. The group A Streptococcus small regulatory RNA FasX enhances streptokinase activity by increasing the stability of the ska mRNA transcript. Mol Microbiol 2010;78:1332–47. Régnier P, Hajnsdorf E. The interplay of Hfq, poly(A) polymerase I and exoribonucleases at the 3′ ends of RNAs resulting from Rho-independent termination: a tentative model. RNA Biol 2013;10:602–9. Rochat T, Bouloc P, Yang Q, et al. Lack of interchangeability of Hfq-like proteins. Biochimie 2012;94:1554–9. Roux CM, DeMuth JP, Dunman PM. Characterization of components of the Staphylococcus aureus mRNA degradosome holoenzyme-like complex. J Bacteriol 2011;193:5520–6. Ruiz de los Mozos I, Vergara-Irigaray M, Segura V, et al. Base pairing interaction between 5′ - and 3′ -UTRs controls icaR mRNA translation in Staphylococcus aureus. PLoS Genet 2013;9:e1004001. Sauer E. Structure and RNA-binding properties of the bacterial LSm protein Hfq. RNA Biol 2013;10:610–8. Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 Liang W, Malhotra A, Deutscher MP. Acetylation regulates the stability of a bacterial protein: growth stage-dependent modification of RNase R. Mol Cell 2011;44:160–6. Linder P, Lemeille S, Redder P. Transcriptome-wide analyses of 5′ -ends in RNase J mutants of a gram-positive pathogen reveal a role in RNA maturation, regulation and degradation. PLoS Genet 2014;10:e1004207. Lioliou E, Sharma CM, Caldelari I, et al. Global regulatory functions of the Staphylococcus aureus endoribonuclease III in gene expression. PLoS Genet 2012;8:e1002782. Liu B, Deikus G, Bree A, et al. Global analysis of mRNA decay intermediates in Bacillus subtilis wild-type and polynucleotide phosphorylase-deletion strains. Mol Microbiol 2014;94:41–55. Liu G, Chatr KF, Chandra G, et al. Molecular regulation of antibiotic biosynthesis in streptomyces. Microbiol Mol Biol R 2013;77:112–43. Liu N, Niu G, Xie Z, et al. The Streptococcus mutans irvA gene encodes a trans-acting riboregulatory mRNA. Mol Cell 2015;57:179–90. Luttinger A, Hahn J, Dubnau D. Polynucleotide phosphorylase is necessary for competence development in Bacillus subtilis. Mol Microbiol 1996;19:343–56. Lybecker M, Bilusic I, Raghavan R. Pervasive transcription: detecting functional RNAs in bacteria. Transcription 2014a;5: e944039. Lybecker M, Zimmermann B, Bilusic I, et al. The doublestranded transcriptome of Escherichia coli. P Natl Acad Sci USA 2014b;111:3134–9. Mader U, Zig L, Kretschmer J, et al. mRNA processing by RNases J1 and J2 affects Bacillus subtilis gene expression on a global scale. Mol Microbiol 2008;70:183–96. Makeyev AV, Liebhaber SA. The poly(C)-binding proteins: a multiplicity of functions and a search for mechanisms. RNA 2002;8:265–78. Marincola G, Schäfer T, Behler J, et al. RNase Y of Staphylococcus aureus and its role in the activation of virulence gene. Mol Microbiol 2012;85:817–32. Mathy N, Benard L, Pellegrini O, et al. 5′ -to-3′ exoribonuclease activity in bacteria: role of RNase J1 in rRNA maturation and 5′ stability of mRNA. Cell 2007;129:681–92. Mathy N, Hebert A, Mervelet P, et al. Bacillus subtilis ribonucleases J1 and J2 form a complex with altered enzyme behaviour. Mol Microbiol 2010;75:489–98. Morfeldt E, Taylor D, von Gabain A, et al. Activation of alphatoxin translation in Staphylococcus aureus by the transencoded antisense RNA, RNAIII. EMBO J 1995;14:4569–77. Morita T, Aiba H. RNase E action at a distance: degradation of target mRNAs mediated by an Hfq-binding small RNA in bacteria. Gene Dev 2011;25:294–8. Morrison JM, Anderson KL, Beenken KE, et al. The staphylococcal accessory regulator, SarA, is an RNA-binding protein that modulates the mRNA turnover properties of lateexponential and stationary phase Staphylococcus aureus cells. Front Cell Infect Microbiol 2012;2:26. Morrison JM, Dunman PM. The modulation of Staphylococcus aureus mRNA turnover. Future Microbiol 2011;6:1141–50. Newman JA, Hewitt L, Rodrigues C, et al. Dissection of the network of interactions that links RNA processing with glycolysis in the Bacillus subtilis degradosome. J Mol Biol 2012;416:121–36. Nielsen JS, Lei LK, Ebersbach T, et al. Defining a role for Hfq in Gram-positive bacteria: evidence for Hfq-dependent antisense regulation in Listeria monocytogenes. Nucleic Acids Res 2010;38:907–19. 329 330 FEMS Microbiology Reviews, 2015, Vol. 39, No. 3 Strahl H, Turlan C, Khalid S, et al. Membrane recognition and dynamics of the RNA degradosome. PloS Genet2015;11:e1004961. Taverniti V, Forti F, Ghisotti D, et al. Mycobacterium smegmatis RNase J is a 5′ -3′ exo-/endoribonuclease and both RNase J and RNase E are involved in ribosomal RNA maturation. Mol Microbiol 2011;82:1260–76. Vecerek B, Rajkowitsch L, Sonnleitner E, et al. The C-terminal domain of Escherichia coli Hfq is required for regulation. Nucleic Acids Res 2008;36:133–43. Vogel J, Luisi BF. Hfq and its constellation of RNA. Nat Rev Microbiol 2011;9:578–89. Wagner EG. Cycling of RNAs by Hfq. RNA Biol 2013;10:619–26. Wang W, Bechhofer DH. Properties of a Bacillus subtilis polynucleotide phosphorylase deletion strain. J Bacteriol 1996;178:2375–82. Wurtzel O, Sesto N, Mellin JR, et al. Comparative transcriptomics of pathogenic and non-pathogenic Listeria species. Mol Syst Biol 2012;8:583. Zhao M, Zhou L, Kawarasaki Y, et al. Regulation of RraA, a protein inhibitor of RNase E-mediated RNA decay. J Bacteriol 2006;188:3257–63. Downloaded from https://academic.oup.com/femsre/article/39/3/316/2467786 by guest on 21 March 2023 Sayed N, Jousselin A, Felden B. A cis-antisense RNA acts in trans in Staphylococcus aureus to control translation of a human cytolytic peptide. Nat Struct Mol Biol 2011;19: 105–12. Sayed N, Nonin-Lecomte S, Réty S, et al. Functional and structural insights of a Staphylococcus aureus apoptotic-like membrane peptide from a toxin-antitoxin module. J Biol Chem 2012;287:43454–63. Sesto N, Touchon M, Andrade JM, et al. A PNPase dependent CRISPR System in Listeria. PLoS Genet 2014;10:e1004065. Shahbabian K, Jamalli A, Zig L, et al. RNase Y, a novel endoribonuclease, initiates riboswitch turnover in Bacillus subtilis. EMBO J 2009;28:3523–33. Silvaggi JM, Perkins JB, Losick R. Small untranslated RNA antitoxin in Bacillus subtilis. J Bacteriol 2005;187:6641–50. Smaldone GT, Antelmann H, Gaballa A, et al. The FsrA sRNA and FbpB protein mediate the iron-dependent induction of the Bacillus subtilis lutABC iron-sulfur-containing oxidases. J Bacteriol 2012;194:2586–93. Stead MB, Marshburn S, Mohanty BK, et al. Analysis of Escherichia coli RNase E and RNase III activity in vivo using tiling microarrays. Nucleic Acids Res 2011;39:3188–203.