• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of genesdevCSHL PressJournal HomeSubscriptionseTOC AlertsBioSupplyNetGenes & Development
Genes Dev. Jun 1, 2007; 21(11): 1353–1366.
PMCID: PMC1877748

Staphylococcus aureus RNAIII coordinately represses the synthesis of virulence factors and the transcription regulator Rot by an antisense mechanism

Abstract

RNAIII is the intracellular effector of the quorum-sensing system in Staphylococcus aureus. It is one of the largest regulatory RNAs (514 nucleotides long) that are known to control the expression of a large number of virulence genes. Here, we show that the 3′ domain of RNAIII coordinately represses at the post-transcriptional level, the expression of mRNAs that encode a class of virulence factors that act early in the infection process. We demonstrate that the 3′ domain acts primarily as an antisense RNA and rapidly anneals to these mRNAs, forming long RNA duplexes. The interaction between RNAIII and the mRNAs results in repression of translation initiation and triggers endoribonuclease III hydrolysis. These processes are followed by rapid depletion of the mRNA pool. In addition, we show that RNAIII and its 3′ domain mediate translational repression of rot mRNA through a limited number of base pairings involving two loop–loop interactions. Since Rot is a transcriptional regulatory protein, we proposed that RNAIII indirectly acts on many downstream genes, resulting in the activation of the synthesis of several exoproteins. These data emphasize the multitude of regulatory steps affected by RNAIII and its 3′ domain in establishing a network of S. aureus virulence factors.

Keywords: Regulatory RNA, translational repression, antisense regulation, RNase III, virulence, Staphylococcus aureus

The ability of organisms to use RNA to modulate gene expression is a relatively new concept. In bacteria, antisense RNA regulates essential functions in extrachromosomal elements (Wagner et al. 2002). Novel small RNAs (sRNAs) were discovered mainly in Escherichia coli that are a part of regulatory networks required for fast adaptive response to stress and environmental changes (Gottesman 2005; Storz et al. 2005). Functional studies revealed that many of the sRNAs act as antisense RNAs on mRNAs that encode for outer membrane proteins in E. coli and Salmonella typhymurium. This suggests that sRNAs modulate the surface composition of Gram-negative bacteria (for review, see Guillier et al. 2006; Vogel and Papenfort 2006). Novel sRNAs have also been identified in Gram-positive pathogens such as Staphylococcus aureus (Pichon and Felden 2005; Roberts et al. 2006) and Listeria monocytogenes (Mandin et al. 2007). Several S. aureus sRNAs are encoded by pathogenicity islands (Pichon and Felden 2005); some of the Listeria sRNAs are unique to the pathogenic species (Mandin et al. 2007).

These observations suggest that RNA–RNA interactions would be decisive for virulence regulation or for adaptive stress responses needed for the survival in a host. This has been shown for the largest regulatory RNA, S. aureus RNAIII (514 nucleotides [nt] long), which controls the switch between the expression of surface proteins and excreted toxins (Novick et al. 1993). The expression of RNAIII is induced by the agr system, which functions as a sensor of the population density. While RNAIII is not essential for S. aureus growth in vitro, several studies have shown its importance for virulence in animal models (Cheung et al. 1994; Gillaspy et al. 1995). Nearly all S. aureus clinical isolates from acute infections have been shown to produce RNAIII. This RNA has a small ORF near the 5′ end that encodes for the [partial differential]-hemolysin, and the noncoding regions are suspected to have regulatory functions (Novick 2003). The structure of RNAIII is characterized by 14 stem–loop motifs and two long-distance interactions that define independent structural domains (Benito et al. 2000). Among the conserved domains, three of the hairpin motifs contain C-rich sequences in their apical loops (Fig. 1A) that could potentially base-pair with the Shine and Dalgarno (SD) sequences of target mRNA (Benito et al. 2000; Novick 2003). Two direct mRNA targets of RNAIII have been identified so far, both of them encoding virulence factors. These studies revealed that RNAIII uses structural domains to act as an antisense RNA for hla (α-hemolysin) mRNA translational activation (Morfeldt et al. 1995) and for repression of the synthesis of the main surface adhesin protein A (Huntzinger et al. 2005). In the latter case, the 3′-end domain, in particular stem–loop 13, binds to the ribosome-binding site (RBS) of the spa mRNA, preventing ribosome binding and inducing rapid mRNA degradation initiated by the double-strand-specific endoribonuclease III (RNase III). RNAIII is thus an example of a multiple target regulator, which modulates the expression of genes involved in pathogenesis. It is likely that RNAIII may also indirectly control virulence gene expression by targeting regulatory proteins that account for the regulation of many downstream genes (Dunman et al. 2001; Novick 2003). Similarly, S. aureus has evolved a plethora of sensory systems that turn on/off the expression of virulence genes in response to both environmental and host signals. A complex regulatory network involves interplay between several two-component systems and transcriptional regulatory proteins from the Sar family (SarS, SarT, Rot, SarA) (Novick 2003). Among these pleiotropic transcription factors, Rot (for repressor of toxins) has been identified as a general antagonist of agr regulation. Since transcription of rot is constitutive, it was proposed that RNAIII regulates the activity of Rot by sequestrating the protein (Said-Salim et al. 2003; Oscarsson et al. 2005). More recently, it was shown that the RNAIII-dependent regulation of Rot occurs at the translational level (Geisinger et al. 2006).

Figure 1.
Secondary structure of RNAIII and potential base pairings with target mRNAs. (A) The secondary structure of RNAIII is from Benito et al. (2000). Deletion of the central domain (RNAIII-Δ7–9), of hairpin 13 (RNAIII-Δ13) or of hairpin ...

We show here that the 3′-end domain of RNAIII represses the synthesis of several surface and secreted proteins specific to S. aureus, using a shared antisense regulatory mechanism. This domain uses two redundant hairpin loop motifs to target rot mRNA. In all these systems, the formation of RNAIII–mRNA duplexes leads to the inhibition of ribosome binding and favors specific recognition by RNase III. A proteomic analysis also reveals that the 3′ domain by itself is sufficient to promote the synthesis of several exoproteases and exotoxins. This multiregulatory domain of RNAIII acts directly on the expression of a class of virulence factors and indirectly via the translational regulation of a pleiotropic regulatory protein.

Results

Identification of potential mRNA targets of RNAIII

Many sRNAs act as antisense RNAs by interacting with mRNAs (for review, see Storz et al. 2005). While only a small number of targets have been experimentally identified for a given sRNA, it is expected that many of them regulate more than one mRNA transcript. We hypothesized that the highly conserved 3′ domain of RNAIII targets the RBS of other mRNAs in a manner similar to spa mRNA. Thus, the strategy used here to predict mRNA targets is based on base pairing between the RBS of mRNAs and the stem–loop 13 of RNAIII. The sequence of each RBS of all mRNAs was concatenated to the 5′ side of the hairpin 13 of RNAIII by a linker of 8 nt. Each construct was then submitted to RNAfold for predicting the base pairing between each concatenated sequence. This approach allows the estimation of the hybridization of two RNA sequences and calculates the minimum free energy of hybridization. The mRNA– RNAIII hybrids that were considered had a minimum free energy of −19 kcal/mol or less (Fig. 1B; Supplementary Fig. S1). Among the best candidates, we found several mRNAs that encode virulence factors (SA1000, a fibrinogen-binding protein; the staphylocoagulase precursor, protein A; and SA2353 and SA2093, two proteins homologous to the secretory antigen precursor SsaA), and the peptidoglycane hydrolase (Fig. 1B; Supplementary Fig. S1). We show here that the best target SA1000 mRNA produces a virulence factor conserved in all S. aureus strains, which is anchored to the cell surface and adheres to fibrinogen and fibronectin (Supplementary Fig. S2). This fibrinogen-binding protein was recently shown to contribute to bacterial adhesion and invasion of epithelial cells (Liang et al. 2006). Noticeably, we also found rot mRNA, which encodes Rot protein, a transcriptional regulator of the sar family that affects virulence gene expression (Novick 2003). In an independent search, we looked for base pairing between RNAIII and the 5′ leader regions of the sar family mRNAs. In addition to rot mRNA, sarT mRNA may form a duplex with RNAIII, albeit characterized by a higher minimum free energy (Supplementary Fig. S1). Note that the expression of Rot and SarT was shown to be repressed by the agr system (Schmidt et al. 2003; Geisinger et al. 2006). Other mRNA candidates encoding surface transporter proteins, and unknown proteins were also predicted (Supplementary Fig. S1). Thus, this analysis suggested that RNAIII regulates multiple targets directly and indirectly via the repression of transcriptional regulators. We analyzed the regulatory mechanisms for several potential target mRNAs that encode for virulence factors (SA1000, SA2353) and for the transcriptional regulator Rot.

RNAIII rapidly forms a duplex with SA1000 mRNA covering the RBS

The in silico data shows that SA1000 mRNA is the best candidate to form a long duplex with RNAIII. Before analyzing the formation of the RNAIII–mRNA duplex in vitro, we mapped the transcriptional start site of SA1000 mRNA by primer extension in WA400 strain (ΔrnaIII). A major start site was detected 22 nt upstream of the SD sequence of SA1000 mRNA (Supplementary Fig. S3). Band shift experiments were performed with a 200-nt mRNA fragment encompassing the whole 5′ untranslated region and the RBS. We showed that the 3′-end domain and RNAIII bound equally well to the 5′-end-labeled SA1000 mRNA (Kd 20 nM), whereas the deletion of hairpin 13 in RNAIII (RNAIII-Δ13) failed to bind the mRNA (Supplementary Fig. S3). A time-course analysis estimates the initial rate of RNAIII binding to SA1000 mRNA to be 1 (±0.3) × 105 M−1sec−1, a value close to that obtained for several efficient natural antisense RNAs (Wagner et al. 2002).

The conformation of SA1000 mRNA was probed using Pb(II)-induced cleavages, RNase T1 (specific for unpaired guanines), RNase T2 (specific for unpaired nucleotides), and RNase V1 (specific for helical regions). For SA1000 mRNA, the data support the existence of the two long hairpin structures I and II (Fig. 2A). Binding of RNAIII induced structural changes in a restricted region (from A10 to A60) covering the RBS of the mRNA. Strong protections were mainly observed in hairpin I, accompanied by new RNase V1 cleavages at positions 17 and 50–53, and a new Pb(II)-induced cut at position 23 (Fig. 2A). Binding of SA1000 mRNA to the 3′ domain or to RNAIII induced similar changes in hairpin 13. Strong protections against RNase T2 and Pb(II) were located in region 420–439 (Fig. 2B). Concomitantly, increased cleavages were induced by RNase T2 and Pb(II) at positions 440–452 and by RNase V1 at positions 429, 433–434, and 438. These data are consistent with the rapid formation of the RNAIII–mRNA duplex of nearly 40 base pairs (bp) that overlaps the RBS of SA1000 mRNA (Fig. 2C).

Figure 2.
RNAIII and its 3′ domain bind to the RBS of SA1000 mRNA. (A,B) Enzymatic hydrolysis (RNases T2 and V1) and lead-induced cleavages (Pb2+) of 5′-end-labeled SA1000 mRNA free (−) or with a twofold excess of RNAIII (+) (A) or 5′-end-labeled ...

The 3′ domain represses the synthesis of the fibrinogen-binding protein SA1000 at the post-transcriptional level

We investigated whether the repression of SA1000 synthesis occurs through specific binding of RNAIII to SA1000 mRNA in vivo. The entire regulatory region of SA1000 mRNA was fused in frame with lacZ under the control of an agr-independent promoter (PrpoB) (Table 1). The β-galactosidase activity was determined in the S. aureus strains RN6390 (rnaIII+, agr+), which express RNAIII, and in WA400 (ΔrnaIII), which carries a deletion of rnaIII gene (Table 1). Complementation assays were also done in WA400 transformed with plasmids expressing either RNAIII, RNAIII-Δ13 (deletion of hairpin 13), or the 3′-end domain (Fig. 3A). The β-galactosidase activity was reduced by half in RN6390 as compared with WA400 and was strongly decreased in the WA400 strains expressing either RNAIII or the 3′ domain. Conversely, the expression of RNAIII-Δ13 in WA400 did not affect β-galactosidase synthesis, in agreement with the fact that RNAIII-Δ13 cannot bind to SA1000 mRNA. In the Δrnc strain, a RN6390 derivative in which the RNase III gene (rnc) has been deleted, β-galactosidase activity was enhanced as compared with RN6390 (Fig. 3A).

Table 1.
Strains and plasmids
Figure 3.
RNAIII and its 3′ domain regulate the expression of SA1000 mRNA at the post-transcriptional level. (A) β-Galactosidase activity measured from PrpoB (+1/+200)::lacZ fusions in S. aureus RN6390 (rnaIII+, agr+), RN6390-Δrnc ...

We also analyzed the steady-state level of SA1000 mRNA in different S. aureus strains in late-exponential phase (Fig. 3B). The mRNA was not detected in RN6390 (rnaIII+), and conversely, its level was high in WA400 (ΔrnaIII). In the RN6390-Δrnc strain, the level of SA1000 mRNA was reproducibly found to be slightly higher than in the parental RN6390 strain (Fig. 3B). This result suggests that the RNase III-dependent degradation of the mRNA contributes in part to the disappearance of the mRNA pool. Complementation assays were also done with WA400 (ΔrnaIII) transformed with plasmids expressing RNAIII or several variants. The expression of RNAIII, the 3′ domain, or the 3′ domain carrying mutations in the hairpin loop 14 (ΔL14) strongly reduced the level of SA1000. Conversely, significant levels of the mRNA were still detected in WA400 expressing RNAIII-Δ13 or H14 (Fig. 3B, cf. lanes 7 and 9), and they were almost identical to those found in WA400 transformed with the empty plasmid. Taken together, these data show that hairpin 13 is the main regulatory element of RNAIII required for SA1000 repression, and that RNase III contributes to the efficiency of repression.

Hfq protein is not required for the regulation of SA1000 expression

The bacterial Sm-like Hfq protein is a key player in translational control mediated by sRNAs (Valentin-Hansen et al. 2004). Therefore, we analyzed the role of S. aureus Hfq in RNAIII-dependent regulations. Allelic replacement of the entire coding sequence of hfq gene by the cat (chloramphenicol resistance) gene was done in RN6390 (LUG911) (Table 1). The mutant Δhfq strain formed normal and hemolytic colonies on blood-agar medium plates, and cell growth was not affected in rich medium. The stability of RNAIII was not altered in the Δhfq strain (Supplementary Fig. S3). Furthermore, Hfq protein did not alter the formation of SA1000 mRNA–RNAIII complex in vitro (data not shown), and in vivo SA1000 mRNA was not detectable at the late exponential phase in Δhfq strain as well as in RN6390 (Fig. 3A, lane 3). These data strongly suggest that S. aureus Hfq is not a key player in the regulation of SA1000 expression.

The RNAIII–SA1000 mRNA duplex prevents the binding of the ribosome and promotes RNase III cleavages

RNAIII covers the RBS of SA1000 mRNA; therefore, we tested whether the duplex was able to prevent ribosome binding using toeprinting assays. The ternary initiation complex constituted by S. aureus 30S subunits, the initiator tRNAMet, and SA1000 mRNA blocked the elongation of a cDNA primer by reverse transcriptase (RT), and produced a toeprint 16 nt downstream from the initiation codon (Fig. 3C). Binding of RNAIII significantly decreased the yield of the toeprint since 80% of inhibition was observed at equimolar concentrations of RNAIII and SA1000 mRNA (Fig. 3C). Since the ternary complex forms irreversibly, inhibition was only observed when initiator tRNA was added after RNAIII. The addition of RNAIII-Δ13, which does not form a complex with SA1000 mRNA, did not alter ribosome binding; <20% of the inhibition was observed at a 20-fold molar excess of RNAIII-Δ13 (Fig. 3C). Thus, inhibition of ribosome binding to SA1000 mRNA resulted primarily from a direct interaction of the hairpin 13 of RNAIII to the mRNA.

As RNase III is required for efficient repression in vivo, we analyzed whether this enzyme can cleave the duplex in vitro. A His-tagged RNase III from S. aureus was purified, and the RNase III-dependent cleavages were analyzed on the free RNAs as well as on the native RNAIII–SA1000 mRNA complex (Fig. 3D). In SA1000 mRNA, S. aureus RNase III cleaved the long hairpin structure II at positions A91, A100, A141, and A149. When end-labeled SA1000 mRNA was incubated with RNAIII, two new strong cleavages occurred at positions U37 and A42 in the mRNA (Fig. 3D). Similarly, new cleavages dependent on mRNA binding were detected at A416 and A418 of labeled RNAIII (data not shown). Thus, SA1000 mRNA and the resulting duplex were both cleaved by the enzyme.

The 3′ domain represses the synthesis of another virulence factor, a secretary antigen precursor using a shared mechanism

Among the predicted mRNA targets, SA2353 mRNA, which encodes a secretory antigen precursor, can potentially form a hybrid with RNAIII characterized by a free energy close to that of the spa mRNA–RNAIII hybrid (Fig. 1B). The steady-state level of SA2353 mRNA was analyzed in different S. aureus strains in early and late exponential phase cultures (Fig. 4A). In RN6390 (rnaIII+), the expression of the mRNA is dependent on the growth phase since the level of the mRNA is strongly decreased at the stationary phase as soon as the RNAIII is produced. The same data were obtained with the strain deleted of the hfq gene, while a significant level of SA2353 mRNA was still observed at the late-exponential phase in RN6390-Δrnc. The mRNA was also detected in the WA400 (ΔrnaIII) strain carrying the empty plasmid or expressing hairpin 14 of RNAIII. Conversely, SA2353 mRNA was barely detectable in WA400 transformed with a plasmid expressing RNAIII, its 3′ domain, or the 3′ domain-ΔL14 (Fig. 4A). These experiments argue for an RNAIII-dependent repression of SA2353 mRNA at the late-exponential phase of growth. Furthermore, as for SA1000, depletion of the repressed mRNA requires the RNase III (Fig. 4A).

Figure 4.
RNAIII-dependant regulation of SA2353 mRNA. (A) Northern blot analysis of SA2353 mRNA levels in different strains. RNAs were prepared from early exponential phase (OD 0.5 at 600 nm) (lanes 13) and from stationary phase (OD 5.0 at 600 nm) (lanes ...

For further in vitro studies, we determined the transcriptional start of SA2353 mRNA in WA400 using 5′RACE (rapid amplification of cDNA ends) analysis. This start occurred 126 nt upstream of the AUG codon of the mRNA. Thus, the whole leader region of the mRNA encompassing the RBS and part of the coding sequence (243 nt) was produced in vitro. Preliminary experiments indicate that RNAIII and its 3′ domain rapidly form a stable complex with SA2353 mRNA in vitro (data not shown). Next, we tested the effect of RNAIII on the formation of the ternary ribosomal complex SA2353 mRNA–30S–tRNAfMet in vitro (Fig. 4B). The addition of increasing concentrations of RNAIII or the 3′ domain significantly reduced the yield of the toeprint at position +16, and concomitantly, an RT pause appeared at position G + 3. Conversely, RNAIII-Δ13 failed to compete with the 30S subunit binding, but the RT pause at position +3 disappeared, revealing the signature of the duplex formed between hairpin 13 and the RBS of SA2353 mRNA (Fig. 4C). Thus the 3′ domain of RNAIII forms a stable complex with SA2353 mRNA that is sufficient to prevent ribosome binding in vitro and to induce efficient repression of SA2353 mRNA in vivo. RNase III also appears to be involved in rapid depletion of the repressed mRNA pool.

RNAIII interacts with rot mRNA, encoding a transcriptional regulator, via several loop–loop interactions

It was recently shown that RNAIII represses the synthesis of Rot, a repressor of exoproteins, at the translational level (Geisinger et al. 2006). This inhibition would be dependent on an interaction between distant regions of RNAIII and the 5′ leader region of rot mRNA. Indeed, three hairpin motifs of RNAIII (H7, H13, and H14) carry similar sequences that could potentially base-pair with three complementary UUGGGA sequences, including the SD sequence (GGGAG) of rot mRNA (Fig. 1). While the previous genetic analysis was focused on two of these domains, H7 and H14 (Geisinger et al. 2006), noticeable base complementarities were also observed with the hairpin 13 of RNAIII (Fig. 1B). Using gel retardation assays, we show that the 3′ domain and RNAIII bind equally well to rot mRNA (Kd 10 nM) (Supplementary Fig. S3). Next, we probed the structure of rot mRNA, either free or bound to RNAIII, using several RNases (Fig. 5A; Supplementary Fig. S4). Different variants of RNAIII were used to monitor the reactivity changes in rot mRNA upon binding. These RNAs include RNAIII-Δ14; RNAIII-Δ13; RNAIII-Δ7–9; the 3′-end domain, which carries two out of the three redundant hairpin motifs; and the isolated hairpins 7, 13, and 14 (Fig. 1). Binding of RNAIII induced protection in rot mRNA, mainly at the purine-rich GGGA sequences located in two apical loops (nucleotides G170–G184, and G227–G232) (Fig. 5B). Strikingly, binding of RNAIII-Δ7-9, RNAIII-Δ13, RNAIII-Δ14, or the 3′ domain induced similar effects on rot mRNA structure (Supplementary Fig. S4). Therefore, two out of the three hairpin domains of RNAIII were sufficient to promote identical footprints as did RNAIII in rot mRNA. We also tested whether the isolated hairpin (H7, H13, or H14) bind rot mRNA in a similar way, using RNase T1 hydrolysis. The results show that the hairpins bind with a lower binding affinity to rot mRNA than RNAIII or the 3′ domain. Hairpin 14 protects loop II of rot mRNA at a lower concentration than loop I, whereas hairpins 13 or 7 interact first with loop I and at higher concentrations with loop II (Fig. 5A). Binding of rot mRNA to end-labeled RNAIII or the 3′ domain induced correlated protections restricted to the complementary loop sequences of hairpins 7, 13, and 14 (data not shown).

Figure 5.
RNAIII binds to rot mRNA. (A) RNase T1 hydrolysis of 5′-end-labeled rot mRNA either free (lane 3) or in the presence of increasing concentrations of RNAIII and of the isolated hairpins H7, H13, or H14. (Lanes 48) Shown are 1, 10, 50, ...

Taken together, these data indicate that the three hairpins 7, 13, and 14 are all capable of binding to the three UUGGGA motifs, albeit with a significant preference of hairpin 14 for the SD sequence of rot mRNA. The fact that changes are only restricted to the redundant and complementary sequences favors the existence of several loop–loop interactions rather than the formation of an extended duplex as predicted by the in silico approach (Figs. 1B, ,5C).5C). The propagation of the initial intermolecular helix is probably stopped due to the topological stress.

Two loop–loop interactions are required for efficient repression of rot mRNA translation

In order to test the consequence of the 3′ domain of RNAIII on regulation in vivo, we fused the regulatory region of rot with lacZ downstream from an agr-independent promoter (PrpoB) (Huntzinger et al. 2005). Determination of the transcriptional start of rot mRNA in RN6390 using primer extension and 5′RACE experiments revealed one major start site at 245 nt upstream of the AUG codon and several shorter RNA transcripts, probably originating from internal cleavage events. The 5′ leader sequence of the primary rot mRNA transcript, containing the region that binds RNAIII in vitro, was fused with lacZ. The β-galactosidase activity was assayed in the S. aureus strains RN6390 (rnaIII+) and WA400 (ΔrnaIII). The expression of the 3′ domain in WA400 strongly repressed β-galactosidase activity, as did RNAIII; i.e., 70%–80% inhibition was observed as compared with WA400 (Fig. 6A). However, mutations in the hairpin loop 14 of the 3′ domain (ΔL14) decrease the RNAIII-dependent β-galactosidase synthesis inhibition to only 20%, suggesting that the hairpin 13 by itself is not sufficient to cause efficient repression (Fig. 6A).

Figure 6.
RNAIII regulates the expression of rot mRNA at the post-transcriptional level. (A) β-Galactosidase activity measured from PrpoB (+1/+290)::rot fusions in S. aureus RN6390 (rnaIII+, agr+), RN6390-Δrnc, WA400 (ΔrnaIII), WA400 ...

We further tested whether RNAIII and its derivatives were able to prevent ribosome binding at rot mRNA using toeprinting (Fig. 6C). While rot mRNA carries two potential RBSs, the experiments revealed one single stop at A258. Thus, the RBS contains the SD sequence GGGAG232. Interestingly, the RNAIII mutants carrying only two redundant hairpin motifs (H7-H14 [RNAIII-Δ13], H7-H13 [RNAIII-Δ14], or H13-H14 [RNAIIIΔ7–9, 3′ domain]) competed with the 30S subunit as efficiently as RNAIII (data not shown). However, even high concentrations of the isolated hairpin (H7, H13, or H14) had no similar effect on ribosome binding (Fig. 6C). These experiments indicate that the 3′ domain is sufficient to repress rot mRNA translation initiation, and that two loop–loop interactions are required for full repression.

A specific signature of RNase III cleavage in rot mRNA when bound to RNAIII

We previously showed that RNase III is associated with the regulatory function of RNAIII (Huntzinger et al. 2005). Here we tested the expression of the rot-lacZ reporter construct in the Δrnc strain. The results showed that the β-galactosidase activity was increased in the Δrnc strain as compared with RN6390 (rnaIII), suggesting that RNase III contributed to repression efficiency (Fig. 6A). Previous Northern blot experiments have shown that the half-life of rot mRNA does not significantly vary in RN6390 and Δagr strains (Geisinger et al. 2006). However, Northern blot experiments performed on total RNAs extracted in the late exponential phase revealed a shorter fragment of rot mRNA that was absent in WA400 (ΔrnaIII) (Fig. 6B). Interestingly, this fragment was also absent in the RN6390-Δrnc strain while the amount of full-length mRNA was significantly increased. These data indicate that a cleavage site in rot mRNA is dependent on both RNAIII binding and RNase III activity.

RNase III cleavage assays were further analyzed on the free RNAs and on native RNAIII–rot mRNA complexes (Fig. 6D). No specific RNase III cleavages were observed in the 5′-end-labeled RNAIII when bound to rot mRNA (data not shown). When labeled rot mRNA was incubated with RNAIII, one strong RNase III-dependent cleavage appeared at A231 and a weaker one at A182 (Fig. 6D) in the two GGGA sequences that base pair with RNAIII. The strong cut, located in the SD of rot mRNA, resulted from binding of hairpin 14. Indeed, this cleavage was absent in complexes formed between rot mRNA and hairpin 13, or RNAIII-Δ14. Furthermore, binding of hairpin 7 to rot mRNA induced a cleavage at A182, which did not occur in rot mRNA bound to the 3′ domain or to RNAIII-Δ7–9. These two specific RNase III cleavages might result from coaxial stacking of several helices forming a long helical structure recognized by the RNase (Fig. 5C). In agreement with the footprinting experiments, these data indicate that hairpin 14 binds preferentially to the SD sequence, and that RNase III cleavage may help to inactivate rot mRNA translation.

The 3′ domain of RNAIII indirectly activates the expression of several exoproteins

Given that the 3′ domain of RNAIII might regulate numerous genes, we analyzed the changes induced by the 3′ domain on the extracellular proteome, which includes several known virulence factors (Sibbald et al. 2006). This was done by two-dimensional gel electrophoresis/mass spectroscopy (MS) on extracts prepared from different strains: RN6390 (rnaIII+), WA400 (ΔrnaIII), and WA400 transformed either by an empty plasmid or by the plasmid expressing the 3′ domain. Proteins isolated from exponential phase cultures of the different strains were solubilized for 2-DGE and arrayed using pH 3–10 IPG two-dimensional gels (Supplementary Fig. S5). Previous work has shown that the synthesis of many of the exoproteins was drastically altered in the Δagr strain (Cordwell et al. 2002). We show here that the expression of the 3′ domain significantly restored the synthesis of several exoproteins, including hemolysins (α- and β-hemolysins) and triacylglycerol lipase, and of several exoproteases (cysteine protease SspB, serine proteases SplA-F).

Discussion

S. aureus is a Gram-positive pathogen responsible for a wide variety of human diseases, and is one of the leading causes of hospital and community-acquired infections. The high diversity of clinical infections caused by S. aureus depends on the expression of numerous virulence factors and stress response pathways, which require the coordinated action of several regulators. Among these intracellular regulators, RNAIII plays a key role in the quorum sensing-dependent central regulatory circuit and coordinately regulates numerous virulence-associated genes (Novick 2003). However, the mechanistic aspects of these regulations have been elucidated only for a few cases (Morfeldt et al. 1995; Huntzinger et al. 2005; Geisinger et al. 2006). The present study not only extends the number of target genes for which a mechanistic model of regulation can be proposed, but also reveals that a specific domain of RNAIII, the 3′ domain, acts primarily as an antisense RNA by targeting a class of mRNAs encoding virulence factors and the transcriptional regulator Rot.

The 3′ domain of RNAIII causes direct repression of SA1000 mRNA (coding for a newly adhesin) translation and induces its rapid degradation in a process that is dependent on the double-strand-specific RNase III (Fig. 7A). Similar results were obtained for SA2353 mRNA (encoding a secretory antigen precursor): In vivo, the depletion of the mRNA pool requires RNAIII or its 3′ domain along with RNase III, and in vitro, the formation of the SA2353 mRNA–RNAIII duplex prevents the binding of the ribosome. The in silico approach predicted several other target mRNAs such as coa (coagulase), lytM (petidoglycane hydrolase), SA2093 (homologous to SsaA), and SA1183 (a choline–glycine transporter), whose expression was shown to be repressed by the agr system, and most likely by RNAIII (Benito et al. 1998; Ramadurai et al. 1999; Dunman et al. 2001). Hence, the mechanism proposed for spa, SA1000, and SA2353 repression may be generalized to many other mRNAs (Fig. 1B; Supplementary Fig. S1). Note that the SA1000 and spa mRNAs have similar structural features; they both carry a 5′ hairpin structure with the SD sequence in the apical loop and a long hairpin structure located downstream from the complementary nucleotides (Fig. 7A). Structure prediction and probing experiments also suggested that coa mRNA has a very similar structural organization (C. Chevalier and P. Romby, unpubl.). We propose that the 5′ hairpin motif facilitates the initial RNAIII–mRNA contacts. Indeed, fast pairings are usually a consequence of a limited number of initial interactions involving either loop–loop or unpaired region–loop interactions (Wagner et al. 2002). Thus the nucleation step probably involves base pairings between loop 13 of RNAIII and the SD sequence located in the first apical loop of SA1000 and spa mRNAs. These interactions, rich in GC base pairs, are then subsequently propagated to form extended duplex of >30 bp (Fig. 1B). The formation of such long duplexes explains why RNase III initiates the rapid decay of these mRNAs. In addition, RNase III induced cleavages in these duplexes and also in the long downstream hairpin of mRNAs, which may provide access to several other endoribonucleases or exoribonucleases for further depletion of the mRNA pool (Fig. 7A). In E. coli, sRNA-dependent translational repression is also followed by rapid decay of the repressed mRNA. However, the enzyme that initiates degradation is the single-strand-specific RNase E since many duplexes involve <20 bp (Morita et al. 2005). In E. coli, degradation of repressed mRNAs is probably a consequence of ribosome exclusion rather than the primary event because the translational repression can occur in the absence of mRNA destabilization (Morita et al. 2006). In S. aureus, we postulate that RNAIII-dependent inhibition of the ribosome is coupled to the RNase III-dependent cleavage to render the arrest of translation irreversible. This event might be facilitated by the fact that RNAIII guides the endoribonuclease to the repressed mRNAs in vivo since the enzyme was purified through association with RNAIII coated beads (Huntzinger et al. 2005).

Figure 7.
Schematic view of the antisense regulatory mechanisms. (A) RNAIII binds to its target mRNAs via one (top panel) or two (bottom panel) loop–loop interactions. In the first case, the initial base pairings are propagated, leading to the formation ...

We also show that RNAIII or its 3′ domain represses rot mRNA translation initiation through base-pairing with two short sequences, one of which overlaps with the SD sequence and the other resides further upstream within the 5′ leader region (Figs. 5, ,7A).7A). Our data indicate that hairpin 14 binds preferentially to the SD sequence of rot mRNA to form a perfect match of seven contiguous base pairs. Moreover, the 3′ domain and RNAIII variants, which contain only two of the hairpin loops, bind equally well to rot mRNA and compete with the ribosome as efficiently as RNAIII. Therefore, we suggest that several coexisting complexes with two loop–loop interactions occur (Fig. 5C, H7-H14 and H13-H14), reminiscent of the E. coli OxyS-fhlA mRNA system (Argaman and Altuvia 2000). Our data also stress that one loop–loop interaction is not sufficient to block the access of the ribosome in vitro and for efficient repression in vivo. Since a large proportion of S. aureus mRNAs contain a strong SD sequence (Chang et al. 2006), the three hairpin motifs (7, 13, and 14) of RNAIII could potentially interact with the SD of many mRNAs. To prevent such global effects, which would be detrimental for the bacteria, specificity is obtained either by propagating the initial contacts into more extended duplexes, as seen in the target mRNAs encoding virulence factors (protein A, SA1000, SA2353, coagulase), or by the addition of a second loop–loop interaction, as shown for rot mRNA (Fig. 7A). It is important to note that the inhibition of rot mRNA translation is not followed by total depletion of the mRNA pool although part of rot mRNA is cleaved by RNase III (Fig. 5; Geisinger et al. 2006). Rot is a pleiotropic transcriptional regulator, and RNA regulation is likely to be more dynamic in this case in order to rapidly respond to specific cellular needs.

A main difference with sRNA-dependent regulation in E. coli and S. aureus is the functional requirement of the Sm-like protein Hfq. In E. coli, Hfq stabilizes sRNA or facilitates its interaction with the target mRNAs (Storz et al. 2005). Mutations in hfq also decrease virulence in several pathogens (for review, see Vogel and Papenfort 2006). RNAIII is a potential target of S. aureus Hfq in vivo (Huntzinger et al. 2005). However, the protein exerts no major effect in vivo on SA1000, SA2353, rot mRNA levels, and RNAIII stability and, in vitro, on RNAIII–mRNA complex formation. Furthermore, deletion of hfq had no significant effect on virulence in Caenorhabditis elegans (S. Boisset and F. Vandenesch, unpubl.). Our data are in agreement with recent works that show the deletion of Hfq had no noticeable effect on virulence gene expression and that this protein is not highly expressed in S. aureus (Geisinger et al. 2006; Bohn et al. 2007). The facts that the structures of RNAIII and its target mRNAs have evolved to promote fast pairings and the duplexes formed are substrates of RNase III may obviate the need of Hfq protein.

Several of the proteins (protein A, coagulase) regulated by RNAIII are considered major virulence factors, produced by all S. aureus clinical isolates. We show that SA1000 is a novel conserved fibrinogen-binding protein in agreement with recent observations that SA1000 belongs to the bacterial adhesion machinery and is required for virulence (Liang et al. 2006). Like Efb (extracellular fibrinogen-binding protein) or Eap (extracellular adhesion protein), SA1000 does not contain the C-terminal LPXTG consensus sequence that is required for the processing and anchorage of mature adhesins to the peptidoglycan, and probably interacts with negatively charged surface molecules, such as lipoteichoic acids, by noncovalent electrostatic or hydrophobic interactions (Hussain et al. 2001). The function of proteins SA2353 and SA2093 in virulence is still unknown. However, these two proteins are homologous to the staphylococcal secretory antigen SsaA (Dubrac and Msadek 2004), which is predicted to play a role in the pathogenesis of S. epidermidis infective endocarditis (Lang et al. 2000). Thus, RNAIII seems to directly repress the synthesis of proteins that are required for the earlier phase of the infection, as several of them favor the bacterial adherence (protein A, fibrinogen-binding protein SA1000, coagulase), protect the bacteria against the host defense (protein A, staphylococcal secretary antigens), or are part of the cell wall metabolism (peptidoglycane hydrolase). Interestingly, many of these proteins are exposed at the surface of S. aureus, an observation also made for several sRNAs in enterobacteria and their pathogenic relatives (for reviews, see Guillier et al. 2006; Vogel and Papenfort 2006). Modulation of membrane and surface properties in bacterial pathogens by sRNAs may contribute to virulence in different ways—by ensuring the entry of essential nutrients, by excluding toxic compounds, and by combating the host defense mechanisms.

RNAIII is expected to initiate regulatory cascades to modulate the expression of multiple downstream genes involved in or required for pathogenesis. Previous studies have shown that in vitro repression of cell wall proteins occurs as soon as RNAIII is produced in the exponential phase, whereas a delay is observed for the activation of many of the exoproteins, suggesting that RNAIII regulates an intermediary factor (Vandenesch et al. 1991; Balaban and Novick 1995). Rot protein might be this factor since it represses the transcription of several exoproteases and toxins, conferring a pattern of expression opposite to that seen with RNAIII (Geisinger et al. 2006; Oscarsson et al. 2006). The 3′ domain efficiently represses the expression of Rot and of several virulence factors and concomitantly activates the synthesis of many exoproteases. Since we did not observe obvious complementarities between the 3′ domain and these mRNAs, we suggest that the activation of the synthesis of the exoproteases by the 3′ domain might be the consequence of its direct action on rot mRNA translation (Fig. 7B). Furthermore, Rot protein was also shown to activate the expression of cell surface proteins (Said-Salim et al. 2003; Oscarsson et al. 2006). Hence, several genes appear to be controlled at multiple levels by RNAIII (Fig. 7A). This is particularly well elucidated for spa, whose promoter activity, translation initiation, and mRNA degradation are negatively controlled by RNAIII and its 3′ domain (Huntzinger et al. 2005). Rot binds to the promoter region of spa and activates transcription (Oscarsson et al. 2005). We therefore propose that the expression of spa mRNA is directly repressed by RNAIII via an antisense mechanism and indirectly through the repression of rot mRNA translation. Such a situation is probably more common than initially expected and is likely the case for coagulase, which is also repressed at multiple levels by RNAIII (Lebeau et al. 1994; Benito et al. 1998). Finally, transcriptomic analysis has revealed many overlapping effects between several two-component systems and the quorum-sensing system (Dunman et al. 2001; Dubrac and Msadek 2004; Liang et al. 2005; Luong et al. 2006; Oscarsson et al. 2006). These multiple interconnections indicate that RNAIII can target other intermediary regulatory proteins. For example, a potential base pairing is predicted between RNAIII and sarT mRNA (Supplementary Table S1), a member of the transcriptional regulator of the sar family known to be negatively regulated by SarA and agr (Schmidt et al. 2003).

Such multiple levels of regulation of the genes encoding virulence factors are remarkable and suggest that timing and the amount of production of each of these virulence factors need to be precisely controlled during the course of infection. It also provides the bacteria with various ways to integrate several external signals. The fact that the conserved 3′ domain “concentrates” a direct and an indirect control of expression of many virulence factors suggests that this part of the molecule has been subject to strong evolutionary constraints. In contrast, other parts of the RNAIII molecule are much more divergent, probably owing to species-specific traits. Further studies will be required to identify the complete RNAIII-dependent regulatory networks in staphylococcal strains. Elucidating the functions of the novel noncoding RNAs in S. aureus should also provide a clearer picture of the importance of RNA in biological processes of this pathogen.

Materials and methods

Strains and plasmids

S. aureus RN6390 derives from 8325-4. In WA400 (ΔrnaIII), the P3 operon is deleted and replaced by the chloramphenicol transacetylase gene (cat86) (Janzon and Arvidson 1990). LUG774 and LUG911 strains derive from RN6390 in which the RNase III (rnc) and hfq genes, respectively, have been replaced by cat86 gene (Huntzinger et al. 2005). The deletion/replacement Δhfq/cat mutant of S. aureus RN6390 (LUG 911) was obtained by using pMAD, a thermosensitive plasmid (Arnaud et al. 2004). For experimental details, see the Supplemental Material. Clones were transferred by electroporation into RN4220, a nitrosoguanidine-induced mutant capable of accepting E. coli DNA, before transfer to other strains. Staphylococci were grown either on BM agar plates or in brain–heart infusion (BHI) with erythromycin (5 μg/mL) when appropriate. RNAIII and its variants were expressed in S. aureus WA400 by using plasmid pE194 (see Table 1).

Construction of translation fusions and β-galactosidase measurements

Translation fusions were constructed by using plasmid pLUG220, a derivative of pTCV-lac, a low-copy-number promoter-less lacZ vector (Table 1). Leader regions of SA1000 and rot mRNAs were cloned downstream from the rpoB promoter in frame with lacZ (Huntzinger et al. 2005). β-Galactosidase activity was measured three times on duplicate cultures with the Enzyme Assay System (Promega).

Northern blots and measure of mRNA half-life

Electrophoresis of total RNA, prepared from late exponential growth phase of different S. aureus strains, was done on a 1% agarose gel containing 2.2 M formaldehyde. Hybridizations were carried out with end-labeled-specific probes to detect mRNA, RNAIII, or 5S rRNA (Huntzinger et al. 2005).

RNA preparation

RNAIII, RNAIII derivatives (RNAIII-Δ7–9 deletion of nucleotides G207–U319, RNAIII-Δ13 deletion of nucleotides U409–A451, and RNAIII-Δ14 deletion of nucleotides G483–C511); the isolated hairpins 7, 13, and 14; and mRNA fragments were transcribed in vitro using T7 RNA polymerase as previously described (Benito et al. 2000). Before use, RNAs were renatured by incubation for 2 min at 90°C in RNase-free water, 1 min at 4°C, followed by an incubation step for 15 min at 20°C in TMN buffer (20 mM Tris-acetate at pH 7.5, 10 mM magnesium-acetate, 150 mM Na-acetate).

RNA structure probing and toeprinting assays

Enzymatic hydrolysis, chemical modifications, and RNase III hydrolysis on free or bound RNAs were performed as previously described (Huntzinger et al. 2005). The formation of a simplified translational initiation complex with mRNA and the extension inhibition conditions were strictly identical to those described by (Benito et al. 2000).

In silico approach

All 5′ regions of mRNA of S. aureus N315 were extracted from the genomic sequence as a multifasta file including the following for each mRNA: 28 nt before the start, the start, and 30 nt after the start. Hybridization with the hairpin 13 (H13) domain of RNAIII was simulated by concatenating each 61-nt sequence with a spacer of 8 N and the 5′ region of H13 known to be involved in the interaction with spa mRNA. We used Zuker’s algorithm of the version 1.5 of Vienna RNA package RNAfold (http://www.tbi.univie.ac.at/~ivo/RNA) to compute minimum free energy of each sequence and to predict the secondary structure of the target mRNAs. Only the best folds showing hybridization between H13 and the 5′ region of the mRNA were considered as potential targets.

Acknowledgments

We are thankful to J.C. Cortay for helpful technical advice and for providing the modified pET43 vector, and to C. Guillier and C. Lichte for the analysis of the proteins by mass spectroscopy. We are grateful to G. Wagner, G. Lina, J. Etienne, and E. Westhof for helpful discussions, and to C. Isel and N. Grover for critical reading of the manuscript. This work was supported by the Centre National de la Recherche Scientifique (CNRS), the Institut National pour la Recherche Médicale (INSERM), the Agence Nationale pour la Recherche (ANR), the Ligue Régionale contre le Cancer, the Fondation pour la Recherche Médicale, and the European Community (FOSRAK, FP6-5120; BacRNA, FP6-018618). M.P. was supported by a fellowship from the Fondation pour la Recherche Médicale.

Footnotes

Supplemental material is available at http://www.genesdev.org.

Article is online at http://www.genesdev.org/cgi/doi/10.1101/gad.423507

References

  • Argaman L., Altuvia S., Altuvia S. fhlA repression by OxyS RNA: Kissing complex formation at two sites results in a stable antisense-target RNA complex. J. Mol. Biol. 2000;300:1101–1112. [PubMed]
  • Arnaud M., Chastanet A., Debarbouille M., Chastanet A., Debarbouille M., Debarbouille M. New vector for efficient allelic replacement in naturally nontransformable, low-GC-content, gram-positive bacteria. Appl. Environ. Microbiol. 2004;70:6887–6891. [PMC free article] [PubMed]
  • Balaban N., Novick R.P., Novick R.P. Translation of RNAIII, the Staphylococcus aureus agr regulatory RNA molecule, can be activated by a 3′-end deletion. FEMS Microbiol. Lett. 1995;133:155–161. [PubMed]
  • Benito Y., Lina G., Greenland T., Etienne J., Vandenesch F., Lina G., Greenland T., Etienne J., Vandenesch F., Greenland T., Etienne J., Vandenesch F., Etienne J., Vandenesch F., Vandenesch F. Trans-complementation of a Staphylococcus aureus agr mutant by Staphylococcus lugdunensis agr RNAIII. J. Bacteriol. 1998;180:5780–5783. [PMC free article] [PubMed]
  • Benito Y., Kolb F.A., Romby P., Lina G., Etienne J., Vandenesch F., Kolb F.A., Romby P., Lina G., Etienne J., Vandenesch F., Romby P., Lina G., Etienne J., Vandenesch F., Lina G., Etienne J., Vandenesch F., Etienne J., Vandenesch F., Vandenesch F. Probing the structure of RNAIII, the Staphylococcus aureus agr regulatory RNA, and identification of the RNA domain involved in repression of protein A expression. RNA. 2000;6:668–679. [PMC free article] [PubMed]
  • Bohn C., Rigoulay C., Bouloc P., Rigoulay C., Bouloc P., Bouloc P. No detectable effect of RNA-binding protein Hfq absence in Staphylococcus aureus. BMC Microbiol. 2007;7:10. [PMC free article] [PubMed]
  • Chang B., Halgamuge S., Tang S.-L., Halgamuge S., Tang S.-L., Tang S.-L. Analysis of SD sequences in completed microbial genomes: Non-SD-led genes are as common as SD-led genes. Gene. 2006;373:90–99. [PubMed]
  • Cheung A.L., Eberhardt K.J., Chung E., Yeaman M.R., Sullam P.M., Ramos M., Bayer A.S., Eberhardt K.J., Chung E., Yeaman M.R., Sullam P.M., Ramos M., Bayer A.S., Chung E., Yeaman M.R., Sullam P.M., Ramos M., Bayer A.S., Yeaman M.R., Sullam P.M., Ramos M., Bayer A.S., Sullam P.M., Ramos M., Bayer A.S., Ramos M., Bayer A.S., Bayer A.S. Diminished virulence of a sar-/agr- mutant of Staphylococcus aureus in the rabbit model of endocarditis. J. Clin. Invest. 1994;94:1815–1822. [PMC free article] [PubMed]
  • Cordwell S.J., Larsen M.R., Cole R.T., Walsh B.J., Larsen M.R., Cole R.T., Walsh B.J., Cole R.T., Walsh B.J., Walsh B.J. Comparative proteomics of Staphylococcus aureus and the response of methicillin-resistant and methicillin-sensitive strains to Triton X-100. Microbiol. 2002;148:2765–2781. [PubMed]
  • Dubrac S., Msadek T., Msadek T. Identification of genes controlled by the essential YycG/YycF two-component system of Staphylococcus aureus. J. Bacteriol. 2004;186:1175–1181. [PMC free article] [PubMed]
  • Dunman P.M., Murphy E., Haney S., Palacios D., Tucker-Kellogg G., Wu S., Brown E.L., Zagursky R.J., Shlaes D., Projan S.J., Murphy E., Haney S., Palacios D., Tucker-Kellogg G., Wu S., Brown E.L., Zagursky R.J., Shlaes D., Projan S.J., Haney S., Palacios D., Tucker-Kellogg G., Wu S., Brown E.L., Zagursky R.J., Shlaes D., Projan S.J., Palacios D., Tucker-Kellogg G., Wu S., Brown E.L., Zagursky R.J., Shlaes D., Projan S.J., Tucker-Kellogg G., Wu S., Brown E.L., Zagursky R.J., Shlaes D., Projan S.J., Wu S., Brown E.L., Zagursky R.J., Shlaes D., Projan S.J., Brown E.L., Zagursky R.J., Shlaes D., Projan S.J., Zagursky R.J., Shlaes D., Projan S.J., Shlaes D., Projan S.J., Projan S.J. Transcription profiling-based identification of Staphylococcus aureus genes regulated by the agr and/or sarA loci. J. Bacteriol. 2001;183:7341–7353. [PMC free article] [PubMed]
  • Geisinger E., Adhikari R.P., Jin R., Ross H.F., Novick R.P., Adhikari R.P., Jin R., Ross H.F., Novick R.P., Jin R., Ross H.F., Novick R.P., Ross H.F., Novick R.P., Novick R.P. Inhibition of rot translation by RNAIII, a key feature of agr function. Mol. Microbiol. 2006;61:1038–1048. [PubMed]
  • Gillaspy A.F., Hickmon S.G., Skinner R.A., Thomas J.R., Nelson C.L., Smeltzer M.S., Hickmon S.G., Skinner R.A., Thomas J.R., Nelson C.L., Smeltzer M.S., Skinner R.A., Thomas J.R., Nelson C.L., Smeltzer M.S., Thomas J.R., Nelson C.L., Smeltzer M.S., Nelson C.L., Smeltzer M.S., Smeltzer M.S. Role of the accessory gene regulator (agr) in pathogenesis of staphylococcal osteomyelitis. Infect. Immun. 1995;63:3373–3380. [PMC free article] [PubMed]
  • Gottesman S. Micros for microbes: Non-coding regulatory RNAs in bacteria. Trends Genet. 2005;21:399–404. [PubMed]
  • Guillier M., Gottesman S., Storz G., Gottesman S., Storz G., Storz G. Modulating the outer membrane with small RNAs. Genes & Dev. 2006;20:2338–2348. [PubMed]
  • Horinouchi S., Weisblum B., Weisblum B. Nucleotide sequence and functional map of pE194, a plasmid that specifies inducible resistance to macrolide, lincosamide, and streptogramin type B antibodies. J. Bacteriol. 1982;150:804–814. [PMC free article] [PubMed]
  • Huntzinger E., Boisset S., Saveanu C., Benito Y., Geissmann T., Namane A., Lina G., Etienne J., Ehresmann B., Ehresmann C., Boisset S., Saveanu C., Benito Y., Geissmann T., Namane A., Lina G., Etienne J., Ehresmann B., Ehresmann C., Saveanu C., Benito Y., Geissmann T., Namane A., Lina G., Etienne J., Ehresmann B., Ehresmann C., Benito Y., Geissmann T., Namane A., Lina G., Etienne J., Ehresmann B., Ehresmann C., Geissmann T., Namane A., Lina G., Etienne J., Ehresmann B., Ehresmann C., Namane A., Lina G., Etienne J., Ehresmann B., Ehresmann C., Lina G., Etienne J., Ehresmann B., Ehresmann C., Etienne J., Ehresmann B., Ehresmann C., Ehresmann B., Ehresmann C., Ehresmann C., et al. Staphylococcus aureus RNAIII and the endoribonuclease III coordinately regulate spa gene expression. EMBO J. 2005;24:824–835. [PMC free article] [PubMed]
  • Hussain M., Becker K., von Eiff C., Schrenzel J., Peters G., Herrmann M., Becker K., von Eiff C., Schrenzel J., Peters G., Herrmann M., von Eiff C., Schrenzel J., Peters G., Herrmann M., Schrenzel J., Peters G., Herrmann M., Peters G., Herrmann M., Herrmann M. Identification and characterization of a novel 38.5-kilodalton cell surface protein of Staphylococcus aureus with extended-spectrum binding activity for extracellular matrix and plasma proteins. J. Bacteriol. 2001;183:6778–6786. [PMC free article] [PubMed]
  • Janzon L., Arvidson S., Arvidson S. The role of the δ-lysin gene (hld) in the regulation of virulence genes by the accessory gene regulator (agr) in Staphylococcus aureus. EMBO J. 1990;9:1391–1399. [PMC free article] [PubMed]
  • Kreiswirth B.N., Lofdahl S., Betley M.J., O’Reilly M., Schlievert P.M., Bergdoll M.S., Novick R.P., Lofdahl S., Betley M.J., O’Reilly M., Schlievert P.M., Bergdoll M.S., Novick R.P., Betley M.J., O’Reilly M., Schlievert P.M., Bergdoll M.S., Novick R.P., O’Reilly M., Schlievert P.M., Bergdoll M.S., Novick R.P., Schlievert P.M., Bergdoll M.S., Novick R.P., Bergdoll M.S., Novick R.P., Novick R.P. The toxic shock syndrome exotoxin structural gene is not detectably transmitted by a prophage. Nature. 1983;305:709–712. [PubMed]
  • Lang S., Livesley M.A., Lambert P.A., Littler W.A., Elliott T.S., Livesley M.A., Lambert P.A., Littler W.A., Elliott T.S., Lambert P.A., Littler W.A., Elliott T.S., Littler W.A., Elliott T.S., Elliott T.S. Identification of a novel antigen from Staphylococcus epidermidis. FEMS Immunol. Med. Microbiol. 2000;29:213–220. [PubMed]
  • Lebeau C., Vandenesch F., Greenland T., Novick R.P., Etienne J., Vandenesch F., Greenland T., Novick R.P., Etienne J., Greenland T., Novick R.P., Etienne J., Novick R.P., Etienne J., Etienne J. Coagulase expression in Staphylococcus aureus is positively and negatively modulated by an agr-dependent mechanism. J. Bacteriol. 1994;176:5534–5536. [PMC free article] [PubMed]
  • Liang X., Zheng L., Landwehr C., Lunsford D., Holmes D., Ji Y., Zheng L., Landwehr C., Lunsford D., Holmes D., Ji Y., Landwehr C., Lunsford D., Holmes D., Ji Y., Lunsford D., Holmes D., Ji Y., Holmes D., Ji Y., Ji Y. Global regulation of gene expression by ArlRS, a two-component signal transduction regulatory system of Staphylococcus aureus. J. Bacteriol. 2005;187:5486–5492. [PMC free article] [PubMed]
  • Liang X., Yu C., Sun J., Liu H., Landwehr C., Holmes D., Ji Y., Yu C., Sun J., Liu H., Landwehr C., Holmes D., Ji Y., Sun J., Liu H., Landwehr C., Holmes D., Ji Y., Liu H., Landwehr C., Holmes D., Ji Y., Landwehr C., Holmes D., Ji Y., Holmes D., Ji Y., Ji Y. Inactivation of a two-component signal transduction system, SaeRS, eliminates adherence and attenuates virulence of Staphylococcus aureus. Infect. Immun. 2006;74:4655–4665. [PMC free article] [PubMed]
  • Luong T.T., Dunman P.M., Murphy E., Projan S.J., Lee C.Y., Dunman P.M., Murphy E., Projan S.J., Lee C.Y., Murphy E., Projan S.J., Lee C.Y., Projan S.J., Lee C.Y., Lee C.Y. Transcription profiling of the mgrA regulon in Staphylococcus aureus. J. Bacteriol. 2006;188:1899–1910. [PMC free article] [PubMed]
  • Mandin P., Repoila F., Vergassola M., Geissmann T., Cossart P., Repoila F., Vergassola M., Geissmann T., Cossart P., Vergassola M., Geissmann T., Cossart P., Geissmann T., Cossart P., Cossart P. Identification of new noncoding RNAs in Listeria monocytogenes and prediction of mRNA targets. Nucleic Acids Res. 2007;35:962–974. [PMC free article] [PubMed]
  • Morfeldt E., Taylor D., von Gabain A., Arvidson S., Taylor D., von Gabain A., Arvidson S., von Gabain A., Arvidson S., Arvidson S. Activation of α-toxin translation in Staphylococcus aureus by the trans-encoded antisense RNA, RNAIII. EMBO J. 1995;14:4569–4577. [PMC free article] [PubMed]
  • Morita T., Maki K., Aiba H., Maki K., Aiba H., Aiba H. RNase E-based ribonucleoprotein complexes: Mechanical basis of mRNA destabilization mediated by bacterial noncoding RNAs. Genes & Dev. 2005;19:2176–2186. [PMC free article] [PubMed]
  • Morita T., Mochizuki Y., Aiba H., Mochizuki Y., Aiba H., Aiba H. Translational repression is sufficient for gene silencing by bacterial small noncoding RNAs in the absence of mRNA destruction. Proc. Natl. Acad. Sci. 2006;103:4858–4863. [PMC free article] [PubMed]
  • Novick R.P. Autoinduction and signal transduction in the regulation of staphylococcal virulence. Mol. Microbiol. 2003;48:1429–1449. [PubMed]
  • Novick R.P., Ross H.F., Projan S.J., Kornblum J., Kreiswirth B., Moghazeh S., Ross H.F., Projan S.J., Kornblum J., Kreiswirth B., Moghazeh S., Projan S.J., Kornblum J., Kreiswirth B., Moghazeh S., Kornblum J., Kreiswirth B., Moghazeh S., Kreiswirth B., Moghazeh S., Moghazeh S. Synthesis of staphylococcal virulence factors is controlled by a regulatory RNA molecule. EMBO J. 1993;12:3967–3975. [PMC free article] [PubMed]
  • Oscarsson J., Harlos C., Arvidson S., Harlos C., Arvidson S., Arvidson S. Regulatory role of proteins binding to the spa (protein A) and sarS (staphylococcal accessory regulator) promoter regions in Staphylococcus aureus NTCC 8325-4. Int. J. Med. Microbiol. 2005;295:253–266. [PubMed]
  • Oscarsson J., Tegmark-Wisell K., Arvidson S., Tegmark-Wisell K., Arvidson S., Arvidson S. Coordinated and differential control of aureolysin (aur) and serine protease (sspA) transcription in Staphylococcus aureus by sarA, rot and agr (RNAIII) Int. J. Med. Microbiol. 2006;296:365–380. [PubMed]
  • Peng H.L., Novick R.P., Kreiswirth B., Kornblum J., Schlievert P., Novick R.P., Kreiswirth B., Kornblum J., Schlievert P., Kreiswirth B., Kornblum J., Schlievert P., Kornblum J., Schlievert P., Schlievert P. Cloning, characterization, and sequencing of an accessory gene regulator (agr) in Staphylococcus aureus. J. Bacteriol. 1988;170:4365–4372. [PMC free article] [PubMed]
  • Pichon C., Felden B., Felden B. Small RNA genes expressed from Staphylococcus aureus genomic and pathogenicity islands with specific expression among pathogenic strains. Proc. Natl. Acad. Sci. 2005;102:14249–14254. [PMC free article] [PubMed]
  • Poyart C., Trieu-Cuot P., Trieu-Cuot P. A broad-host-range mobilizable shuttle vector for the construction of transcriptional fusions to β-galactosidase in Gram-positive bacteria. FEMS Microbiol. Lett. 1997;156:193–198. [PubMed]
  • Ramadurai L., Lockwood K.J., Nadakavukaren M.J., Jayaswal R.K., Lockwood K.J., Nadakavukaren M.J., Jayaswal R.K., Nadakavukaren M.J., Jayaswal R.K., Jayaswal R.K. Characterization of a chromosomally encoded glycylglycine endopeptidase of Staphylococcus aureus. Microbiol. 1999;145:801–808. [PubMed]
  • Roberts C., Anderson K.L., Murphy E., Projan S.J., Mounts W., Hurlburt B., Smeltzer M., Overbeek R., Disz T., Dunman P.M., Anderson K.L., Murphy E., Projan S.J., Mounts W., Hurlburt B., Smeltzer M., Overbeek R., Disz T., Dunman P.M., Murphy E., Projan S.J., Mounts W., Hurlburt B., Smeltzer M., Overbeek R., Disz T., Dunman P.M., Projan S.J., Mounts W., Hurlburt B., Smeltzer M., Overbeek R., Disz T., Dunman P.M., Mounts W., Hurlburt B., Smeltzer M., Overbeek R., Disz T., Dunman P.M., Hurlburt B., Smeltzer M., Overbeek R., Disz T., Dunman P.M., Smeltzer M., Overbeek R., Disz T., Dunman P.M., Overbeek R., Disz T., Dunman P.M., Disz T., Dunman P.M., Dunman P.M. Characterizing the effect of the Staphylococcus aureus virulence factor regulator, SarA, on log-phase mRNA half-lives. J. Bacteriol. 2006;188:2593–2603. [PMC free article] [PubMed]
  • Said-Salim B., Dunman P.M., McAleese F.M., Macapagal D., Murphy E., McNamara P.J., Arvidson S., Foster T.J., Projan S.J., Kreiswirth B.N., Dunman P.M., McAleese F.M., Macapagal D., Murphy E., McNamara P.J., Arvidson S., Foster T.J., Projan S.J., Kreiswirth B.N., McAleese F.M., Macapagal D., Murphy E., McNamara P.J., Arvidson S., Foster T.J., Projan S.J., Kreiswirth B.N., Macapagal D., Murphy E., McNamara P.J., Arvidson S., Foster T.J., Projan S.J., Kreiswirth B.N., Murphy E., McNamara P.J., Arvidson S., Foster T.J., Projan S.J., Kreiswirth B.N., McNamara P.J., Arvidson S., Foster T.J., Projan S.J., Kreiswirth B.N., Arvidson S., Foster T.J., Projan S.J., Kreiswirth B.N., Foster T.J., Projan S.J., Kreiswirth B.N., Projan S.J., Kreiswirth B.N., Kreiswirth B.N. Global regulation of Staphylococcus aureus genes by Rot. J. Bacteriol. 2003;185:610–619. [PMC free article] [PubMed]
  • Schmidt K.A., Manna A.C., Cheung A.L., Manna A.C., Cheung A.L., Cheung A.L. SarT influences sarS expression in Staphylococcus aureus. Infect. Immun. 2003;71:5139–5148. [PMC free article] [PubMed]
  • Sibbald M.J., Ziebandt A.K., Engelmann S., Hecker M., de Jong A., Harmsen H.J., Raangs G.C., Stokroos I., Arends J.P., Dubois J.Y., Ziebandt A.K., Engelmann S., Hecker M., de Jong A., Harmsen H.J., Raangs G.C., Stokroos I., Arends J.P., Dubois J.Y., Engelmann S., Hecker M., de Jong A., Harmsen H.J., Raangs G.C., Stokroos I., Arends J.P., Dubois J.Y., Hecker M., de Jong A., Harmsen H.J., Raangs G.C., Stokroos I., Arends J.P., Dubois J.Y., de Jong A., Harmsen H.J., Raangs G.C., Stokroos I., Arends J.P., Dubois J.Y., Harmsen H.J., Raangs G.C., Stokroos I., Arends J.P., Dubois J.Y., Raangs G.C., Stokroos I., Arends J.P., Dubois J.Y., Stokroos I., Arends J.P., Dubois J.Y., Arends J.P., Dubois J.Y., Dubois J.Y., et al. Mapping the pathways to staphylococcal pathogenesis by comparative secretomics. Microbiol. Mol. Biol. Rev. 2006;70:755–788. [PMC free article] [PubMed]
  • Storz G., Altuvia S., Wassarman K.M., Altuvia S., Wassarman K.M., Wassarman K.M. An abundance of RNA regulators. Annu. Rev. Biochem. 2005;74:199–217. [PubMed]
  • Valentin-Hansen P., Eriksen M., Udesen C., Eriksen M., Udesen C., Udesen C. The bacterial Sm-like protein Hfq: A key player in RNA transactions. Mol. Microbiol. 2004;51:1525–1533. [PubMed]
  • Vandenesch F., Kornblum J., Novick R.P., Kornblum J., Novick R.P., Novick R.P. A temporal signal, independent of agr, is required for hla but not spa transcription in Staphylococcus aureus. J. Bacteriol. 1991;173:6313–6320. [PMC free article] [PubMed]
  • Vogel J., Papenfort K., Papenfort K. Small non-coding RNAs and the bacterial outer membrane. Curr. Opin. Microbiol. 2006;9:605–611. [PubMed]
  • Wagner E.G.H., Altuvia S., Romby P., Altuvia S., Romby P., Romby P. Antisense RNAs in bacteria and their genetic elements. Adv. Genet. 2002;46:361–398. [PubMed]

Articles from Genes & Development are provided here courtesy of Cold Spring Harbor Laboratory Press
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...