Regulation of gene expression through processing and turnover of RNA is a key mechanism that allows bacteria to rapidly adapt to changing environmental conditions. Consequently, RNA degrading enzymes (ribonucleases; RNases) such as the endoribonuclease RNase E, frequently play critical roles in pathogenic bacterial virulence and are potential antibacterial targets. RNase E consists of a highly conserved catalytic domain and a variable non-catalytic domain that functions as the structural scaffold for the multienzyme degradosome complex. Despite conservation of the catalytic domain, a recent study identified differences in the response of RNase E homologues from different species to the same inhibitory compound(s). While RNase E from Escherichia coli has been well-characterised, far less is known about RNase E homologues from other bacterial species. In this study, we structurally and biochemically characterise the RNase E catalytic domains from four pathogenic bacteria: Yersinia pestis, Francisella tularensis, Burkholderia pseudomallei and Acinetobacter baumannii, with a view to exploiting RNase E as an antibacterial target. Bioinformatics, small-angle x-ray scattering and biochemical RNA cleavage assays reveal globally similar structural and catalytic properties. Surprisingly, subtle species-specific differences in both structure and substrate specificity were also identified that may be important for the development of effective antibacterial drugs targeting RNase E.
For survival, microorganisms must have the ability to rapidly adapt to environmental changes. One mechanism for achieving this is regulation of gene expression through the differential processing and/or turnover of RNAs by ribonucleases (RNases)1. The importance of RNases in these processes is underlined by the roles that they play in pathogenic bacterial virulence2 and, as a consequence, RNases are emerging as attractive antibacterial drug targets2,3,4. For example, the endoribonuclease RNase E is essential for cell viability in Escherichia coli5,6,7 where roles in mRNA decay and the maturation of tRNA and rRNA have been well documented [reviewed in1 and8]. RNase E is also required for cell viability in Salmonella enterica9 and has been implicated in the pathogenicity of both S. enterica and Yersinia pestis10,11. Furthermore, since homologues of RNase E are predicted to be present in many bacteria12, including pathogenic species, but are not found in animals or humans, RNase E is an ideal potential antibacterial target2,3,4,13.
E. coli RNase E (EcRNase E) is the founding member of Type I RNase Es, found in betaproteobacteria, gammaproteobacteria and cyanobacteria12,14, and has been extensively characterised. It is a large protein containing 1061 amino acids and can be divided into two domains. The N-terminal domain (NTD) is responsible for the endoribonuclease activity15 and the C-terminal domain (CTD) forms the structural scaffold for the RNA-degrading multienzyme complex, the degradosome16,17. The catalytic NTD consists of five subdomains: an RNase H domain, an S1 domain, a 5′ sensor, a deoxyribonuclease (DNase) I domain and a small domain18 (Supplementary Fig. S1). It is a homotetramer formed by interactions between small domains18,19,20,21. The catalytic site is located in the DNase I domain and harbours a hydrated magnesium ion, coordinated by two aspartic acids, Asp303, positioned by asparagine Asn305, and Asp346, that is essential for the hydrolytic cleavage of the RNA substrate18. EcRNase E cleaves single-stranded A/U-rich regions22 and has a strong preference for a 5′ monophosphate23. Specificity for a uracil at the +2 position relative to the cleavage site is defined by the uracil pocket in the S1 domain24. This pocket is comprised of Phe67, positioned by Phe57, and the Lys112-Gly113-Ala114-Ala115 (KGAA) loop24. In addition, recognition of a 5′ monophosphorylated substrate requires the 5′ sensor pocket, formed by amino acids Gly124, Val128, Arg141, Arg142, Arg169, Thr170 and Arg373 with Val128, Arg169 and Thr170 playing the critical roles in 5′ monophosphate detection18,25. Binding of the RNA substrate by the 5′ sensor is predicted to induce a significant structural conformational change that helps to correctly position the RNA substrate for cleavage26.
A number of RNase E homologues from other species have been partially characterised [reviewed in12]. In general, the sequence of the catalytic region of RNase E is highly conserved, whereas the sequence of the degradosome scaffold region is more variable12. Consequently, the focus has mostly been on the degradosome scaffold and its interacting partners12. However, there are indications that the properties of the catalytic domain are species-specific27 and that species-specific differences can extend to the response to potential antibacterial compounds shown to inhibit EcRNase E13. Therefore, characterising the properties of the RNase E catalytic domain from a variety of species may be critical for the development of effective antibacterial drugs targeting RNase E. In this study, we have characterised the structural and biochemical properties of the catalytic domain of the RNase E homologue from four bacterial pathogens, chosen because of their importance in both the health and defence sectors. Specifically, RNase E homologues were chosen from bubonic plague causing Y. pestis (YpRNase E), tularemia causing Francisella tularensis (FtRNase E), melioidosis causing Burkholderia pseudomallei (BpRNase E) and ESKAPE pathogen28 Acinetobacter baumannii (AbRNase E). The sequence and structure of the RNase E catalytic domains were initially investigated in silico, before comparing their low-resolution solution structures to that of EcRNase E by small-angle x-ray scattering (SAXS). The substrate specificity was also compared to that of EcRNase E using biochemical assays. Collectively, these studies have revealed subtle species-specific differences in the properties of previously uncharacterised RNase E catalytic domains from pathogenic species that may prove to be important for the development of effective antibacterial compounds targeting RNase E.
Sequence analysis of RNase E NTDs
In order to compare the protein sequences of the RNase E homologues from the gammaproteobacteria E. coli, Y. pestis, F. tularensis, and A. baumannii and betaproteobacteria B. pseudomallei a multiple sequence alignment of the five full-length proteins was generated and trimmed to the boundaries of the EcRNase E NTD (Fig. 1). As expected, based on their phylogeny, YpRNase, FtRNase E, AbRNase and BpRNase E all belong to the Type I class of RNase Es with catalytic domains of a similar length to EcRNase E NTD located at the N-terminus of the protein. Consequently, from this point forwards, we will also refer to these catalytic domains as NTDs. The alignment reveals that the amino acid sequence of the NTD is highly conserved, with similarities of 69.4% for FtRNase E, 70.5% for AbRNase E, 75.4% for BpRNase E and 96% for YpRNase E, relative to EcRNase. Furthermore, for these five sequences, there is complete conservation of the key residues known to be critical for the specific recognition of substrates containing a 5′ monophosphate by the 5′ sensor pocket (Val128, Arg169 and Thr170, using numbering for EcRNase E;18,25) and for substrate cleavage at the active site (Asp303, Asn305 and Asp346, using numbering for EcRNase E;18) (Fig. 1).
In silico structural analysis of the RNase E NTDs
In order to begin to investigate whether the amino acid sequence similarity also correlates with structural similarity, we decided to generate homology models for YpRNase E, FtRNase E, BpRNase E and AbRNase E NTDs using available EcRNase E NTD crystal structures as templates. The crystal structure of EcRNase E NTD has been solved in the presence18,25 and absence26 of RNA substrates and collectively these structures suggest a large conformational change between an open and closed conformation upon substrate binding26. Both open and closed conformation homology models were generated using the EcRNase E NTD crystal structures 2VMK26 and 2BX218 as the template structure, respectively (Fig. 2). Globally, the homology models are very similar to the EcRNase E NTD crystal structures and this is reflected in the low root-mean-square deviation (RMSD) obtained for each model compared to the template crystal structure (Fig. 2). Even regions that are poorly conserved at the sequence level (e.g. amino acids 175–203 of the 5′ sensor, 233–263 of the RNase H domain and 457–508 of the small domain (Fig. 1)) are predicted to adopt a similar conformation (Fig. 2; Supplementary Fig. 2). The RMSDs for the homology models generated using the closed EcRNase E NTD crystal structure 2BX218 as a template are slightly lower than those generated using the open EcRNase E NTD structure 2VMK26 as the template. However, this most likely reflects the higher resolution of the closed structure (2.85 Å)18 compared to the open structure (3.3 Å)26. Overall, these data suggest that YpRNase E, FtRNase E, BpRNase E and AbRNase E NTDs are likely to fold into a similar structure and adopt similar conformations to EcRNase E NTD. This also implies that the conformational change from open to closed proposed for EcRNase E26 is, at least theoretically, possible.
Purification of recombinant RNase E NTDs
In order to structurally and functionally characterise YpRNase E, FtRNase E, BpRNase E and AbRNase E NTDs, and compare their properties to those of EcRNase E NTD, pure recombinant RNase E NTDs were produced. An expression plasmid that was previously used for the expression, purification and characterisation of an N-terminally His-tagged EcRNase E NTD19,29 was obtained from Prof. Ben Luisi, University of Cambridge, UK. We adopted a similar cloning strategy for YpRNase E, FtRNase E, BpRNase E and AbRNase E NTDs. All five RNase E NTDs were expressed as N-terminally His-tagged proteins and purified by immobilised Ni2+-affinity chromatography, followed by size-exclusion chromatography, as described previously for EcRNase E NTD19. All five RNase E NTDs were successfully purified using this approach (Supplementary Fig. S3).
Low-resolution solution structure of RNase E NTDs
The catalytically active form of EcRNase E NTD has been shown to be homotetrameric in solution19. During purification, YpRNase E, FtRNase E, BpRNase E and AbRNase E NTDs all eluted from the size-exclusion column at a similar volume to EcRNase E NTD (Supplementary Table S1), which would be consistent with them also forming homotetramers. In order to further investigate, and compare, the structures of the RNase E NTDs, SAXS was employed. SAXS is a technique that is able to provide low resolution structural information about the size and shape of proteins in solution30. Parameters including molecular weight, radius of gyration (Rg), which indicates the overall size of the protein, and maximum particle dimension (Dmax) can be derived from the scattering data, along with information about protein flexibility30. The scattering data for the RNase E NTDs are presented in Fig. 3a. Molecular masses were estimated from the SAXS profiles (Supplementary Table S1) and are consistent with each protein forming a tetramer in solution. For each RNase E NTD, the Rg was determined from Guinier analysis (Fig. 3aii). Rgs ranged from 48 to 51 Å, indicating that all five RNase E NTDs are a similar overall size. The Dmax was determined from the pair distance distribution (p(R)) plot (Fig. 3b), a plot of the distances between all possible pairs of atoms within the protien30, and ranged from 149 to 183 Å, suggesting some variability in the degree of extension of the proteins. Finally, dimensionless Kratky plots were used to assess the folding state and flexibility of the proteins30. The Kratky plots (Fig. 3c) were similar for each RNase E NTD and had the characteristic multiple bell-shaped profile of a multi-domain globular protein.
Ab initio models for EcRNase E NTD were generated based on the SAXS data and were compared to three available EcRNase E NTD crystal structures, 2BX2 (closed conformation)18, 2VMK (open conformation)26 and 5F6C (a transitioning conformation)25 (Supplementary Fig. S4a). None of the crystal structures could be fitted entirely within the average molecular envelope. The experimental scattering data for EcRNase E NTD was also compared to theoretical scattering data for the three crystal structures (Supplementary Fig. S4b). We compared the resulting χ2 values to evaluate which of the crystallised conformations the solution scattering data most closely resembles. For EcRNase E NTD, the data are most consistent with the transitioning conformation (Supplementary Table 2; Supplementary Fig. S4b). A similar comparison of the experimental scattering data for YpRNase E, FtRNase E, BpRNase E and AbRNase E NTDs to the theoretical scattering data for the three EcRNase E crystal structures was also performed. This revealed that, in solution, YpRNase E, FtRNase E and AbRNase E also most closely resemble the transitioning EcRNase E NTD conformation (Supplementary Table S2). In contrast, BpRNase E most closely resembles the closed EcRNase E NTD conformation (Supplementary Table S2).
Endoribonuclease Activity of RNase E NTDs
Having established that EcRNase E, YpRNase E, FtRNase E, BpRNase E and AbRNase E NTDs are all homotetramers that adopt similar global conformations in solution, we next wanted to investigate their catalytic properties. We decided to assess the endoribonuclease activity of each of the RNase E NTDs using a model 5′ p-RNA13-FAM-3′ substrate in a discontinuous end-point assay (Fig. 4). RNA13 has the same nucleotide sequence as the 13-mer oligoribonucleotide that was crystallised in complex with EcRNase E NTD18 and is based on a known EcRNase E cleavage site within RNAI from the pBR322 plasmid31. Endoribonucleolytic cleavage is expected to occur between the 8th and 9th nucleotide from the 5′ end of the substrate to generate an unlabelled octamer and a 3′ FAM-labelled pentamer (Fig. 4). Both the disappearance of the 3′ FAM-labelled 13-mer substrate and the appearance of the 3′ FAM-labelled pentamer product can be monitored by denaturing polyacrylamide gel electrophoresis (PAGE). As expected, EcRNase E NTD efficiently cleaved this substrate such that 95% of the substrate had been degraded at the assay end-point (Fig. 4). YpRNase E, FtRNase E, BpRNase E and AbRNase E NTDs also endoribonucleolytically cleaved 5′ p-RNA13-FAM-3′ (Fig. 4). The cleavage efficiency for YpRNase and FtRNase E NTDs was similar to EcRNase E NTD (96% and 93% of the substrate had been degraded at the assay end-point, respectively). However, BpRNase E and AbRNase E NTDs both cleaved 5′ p-RNA13-FAM-3′ less efficiently than EcRNase E NTD (58% and 88% of the substrate had been degraded at the assay end-point, respectively).
Substrate specificity of RNase E NTDs
EcRNase E is known to have a strong preference for substrates with a 5′ monophosphate23 (e.g. 5′ p-RNA13-FAM-3′). In order to investigate whether YpRNase E, FtRNase E, BpRNase E and AbRNase E NTDs also share this preference we decided to compare the degradation of a 5′-OH-RNA13-FAM-3′, which has a 5′ hydroxyl, and 5′-p-RNA13-FAM-3′ in the discontinuous end-point assay (Fig. 4). As discussed above, EcRNase E NTD efficiently cleaved the 5′-p-RNA13-FAM-3′ substrate with 95% substrate degraded at the assay end-point. However, as expected, cleavage of the 5′-OH-RNA13-FAM-3′ substrate was relatively inefficient and only 10% of this substrate was degraded by EcRNase E NTD by the assay end-point. Similarly, compared to 5′ p-RNA13-FAM-3′, 5′-OH-RNA13-FAM-3′ was a poor substrate for YpRNase E, FtRNase E, BpRNase E and AbRNase E NTDs. However, while EcRNase E, YpRNase E, BpRNase E and AbRNase E NTDs degraded 10% or less of the 5′-OH-RNA13-FAM-3′ by the assay end-point, FtRNase E NTD cleaved 25% of this substrate, 2.5-fold more than any of the other RNase E NTDs. Therefore, while all five RNase E NTDs display a preference for 5′ monophosphorylated substrates, this preference is less pronounced for FtRNase E.
The short 5′-p-RNA13-FAM-3′ model substrate is cleaved by all of the RNase E NTDs at a specific, single site. However, longer, more complex substrates have the potential to be cleaved at multiple sites. For example, a partially double-stranded target-guide substrate consisting of a 5′ DABCYL-labelled, 3′ FAM-labelled 27-mer and a 5′ phosphorylated 13-mer with partial complementarity to the 5′ end of the 27-mer (Fig. 5ai) is known to be cleaved by EcRNase E NTD at multiple sites, each with a different susceptibility to EcRNase E32. Therefore, we decided to use the target-guide substrate in a discontinuous assay to compare cleave site specificity of the RNase E NTDs. In an initial experiment, 5 nM RNase E NTD was incubated with 1 μM target-guide RNA at 28 °C and the cleavage products were analysed by denaturing PAGE after 10 minutes (Fig. 5aii). EcRNase E NTD cleaved the 27-mer component of the target-guide substrate at five positions: between nucleotides 15 and 16, 16 and 17, 19 and 20, 22 and 23 and 24 and 25, from the 5′ end, to generate 3′ FAM-labelled cleavage products 12, 11, eight, five and three nucleotides in length, respectively. Similar cleavage patterns were observed for YpRNase E, BpRNase E and AbRNase E NTDs. Each of these RNase E NTDs also cleaved the 27-mer RNA at the positions cleaved by EcRNase E NTD. Although additional cleavage sites were detected between nucleotides 17 and 18 for AbRNase E NTD, generating a 3′ FAM-labelled product ten nucleotides in length, and between nucleotides 23 and 24 for YpRNase E NTD, generating a 3′ FAM-labelled product four nucleotides in length. The relative predominance of the bands representing the common cleavage sites were also similar for EcRNase E, YpRNase E, BpRNase E and AbRNase E NTDs. In contrast, FtRNase E NTD displayed a markedly different cleavage pattern to the other four RNase E NTDs. In addition to cleaving the 27-mer RNA at the five positions that are cleaved by the other RNase E NTDs, FtRNase E NTD also cleaved the 27-mer RNA between nucleotides 18 and 19 and between nucleotides 21 and 22, generating 3′ FAM-labelled products nine and six nucleotides in length, respectively. Furthermore, the relative predominance of the bands representing the common cleavage sites was different for FtRNase E NTD relative to the other four RNase E NTDs.
Given the observed similarities and differences in the cleavage patterns of the RNase E NTDs, time-course experiments were then used to identify initial cleavage sites and/or preferred cleavage sites (Fig. 5b). For EcRNase E NTD, all five cleavage products began to appear from the start of the time-course, consistent with a distribution of single cleavage events. The octamer accumulated at the fastest rate, closely followed by the 12- and 11-mers suggesting that the preferred cleavage site is between nucleotides 19 and 20, from the 5′ end of the 27-mer. A similar accumulation of products was observed for YpRNase E, BpRNase E and AbRNase E NTDs except that the 11-mer and the 12-mer accumulated at the fastest rate for BpRNase E and AbRNase E NTDs, respectively. For FtRNase E NTD, the 12-mer accumulated significantly faster than the other cleavage products indicating that the preferred cleavage site is between nucleotides 15 and 16. Interestingly, after three minutes, the abundance of this cleavage product began to decline while the abundance of the trimer, pentamer and hexamer dramatically increased. This suggests that the initial 12-mer product can undergo a secondary cleavage to produce the shorter trimers, pentamers and hexamers.
We have investigated the structural and biochemical properties of the NTD of four, previously uncharacterised, RNase E homologues from bacterial pathogens and compared them to those of the well-characterised NTD from E. coli RNase E. Fundamentally, these NTDs are highly conserved at the sequence level and, as might have been expected, they have similar properties in that they are all homotetrameric, endoribonucleases. However, there are also subtle differences between them. The NTD from RNase E from B. pseudomallei is predicted to adopt a more closed conformation than the other RNase E NTDs (Fig. 3 and Supplementary Table S2) while the substrate specificity of F. tularensis RNase E NTD differs from the other RNase E NTDs (Figs 4 and 5).
Three principle regions of sequence variability were identified in the RNase E NTDs, amino acids 175–203, 233–263 and 457–508 (Fig. 1). Despite this sequence variability, the structure of these regions is predicted to be conserved (Figs 2 and S2). Two of these regions, one in the 5′ sensor domain (amino acids 175–203) and one in the RNase H domain (amino acids 233–263), are located at the peripheral edges of the tetrameric protein structure (Supplementary Fig. S2). These regions have no known direct catalytic function; however, given their location, it is possible that they may play a structural role in the movement of the 5′ sensor and S1 domains during the transition from an open to a closed conformation upon substrate binding. Interestingly, in solution, the RNase E NTD from B. pseudomallei adopts a more closed conformation than the other four RNase E NTDs (Fig. 3 and Supplementary Table S2) which could be related to the divergent sequence in this region. The third region of sequence variability, in the small domain (amino acids 457–508), is located at the protomer-promoter interface (Supplementary Fig. S2), where amino acid substitutions could affect tetramerisation. However, all five RNase E NTDs appear to be homotetrameric in solution (Supplementary Table S1), suggesting that the function of the small domain is conserved.
All of the RNase E NTDs investigated endoribonucleolytically cleaved model RNA substrates (Figs 4 and 5). This was not surprising given that the key residues known to be involved in RNA cleavage at the active site are absolutely conserved (Fig. 1). This study utilised model RNA substrates that we would expect to be cleaved via a 5′ end-dependent mechanism in which a 5′ monophosphate is specifically recognised by the 5′ sensor pocket18,23,31. The key residues known to be required for discriminating substrates with a 5′ monophosphate within this pocket are absolutely conserved in the RNase E NTDs investigated and, not surprisingly, all of the RNase E NTDs showed a preference for a 5′ monophosphorylated substrate (Fig. 4). However, more complex natural substrates such as 9S RNA, the precursor of 5S rRNA, can also be cleaved through an internal entry mechanism which relies on the recognition of a stem loop structure upstream of the cleavage site33,34,35. A recent study identified eight amino acids (Arg3, Gln22, His268, Tyr269, Gln270, Lys433, Arg488 and Arg490) in the E. coli RNase E NTD domain that are important for recognition of the stem loop25. These residues are absolutely conserved in Y. pestis RNase E but there are substitutions at two of the positions in B. pseudomallei and at six of the positions in F. tularensis and A. baumannii RNase Es. It remains to be seen whether these substitutions affect RNA cleavage through the internal entry mechanism for these RNase Es. Finally, substrate cleavage patterns varied between the RNase E NTDs, with the most striking differences observed for F. tularensis RNase E (Fig. 5), hinting at possible differences in the reaction mechanism.
In conclusion, we have characterised the RNase E NTD homologues from Y. pestis, F. tularensis, A. baumannii and B. pseudomallei. This is the first demonstration that these RNase E NTDs have endoriboucleolytic activity. Furthermore, through a comparison with EcRNase E NTD, we have shown that, despite the expected global structural and catalytic similarities, there are also subtle species-specific differences between them. These observed subtle differences would not have been predicted de novo from amino acid sequence. The most significant differences were identified in two of the RNase E NTDs with the lowest sequence similarity, B. pseudomallei and F. tularensis. However, B. pseudomallei RNase E NTD varied in its structural properties while F. tularensis RNase E NTD varied in its biochemical properties. It remains to be determined, how significant these differences prove to be, but it does suggest caution when applying generalities based on studies with model organisms. In terms of the development of antibacterial drugs targeting RNase E, it has already been reported that RNase Es from different species respond differently to compounds selected based on inhibition of RNase E from E. coli13. Therefore, these subtle differences may prove to be critical when designing antibacterial approaches targeting RNase E in pathogens.
Protein sequence alignment
RNase E amino acid sequences were downloaded from UniProtKB using the following accession codes: EcRNase E (P21513), YpRNase E (Q74TC3), FtRNase E (Q5NFK7), BpRNase E (A0A0H3HN63) and AbRNase E (A0A0B9WR03). The sequences were aligned in MOE (Molecular Operating Environment, 2013.08; Chemical Computing Group Inc., 1010 Sherbrooke St. West, Suite #910, Montreal, QC, Canada, H3A 2R7) using the default parameters, including a BLOSUM62 substitution matrix. This alignment was used to define the catalytic NTD for each of the RNase Es as the amino acids that aligned with amino acids 1–529 of EcRNase E. Sequences were then trimmed to the NTD boundaries and the similarity of each homologue to EcRNase E NTD was calculated in MOE.
RNase E NTD homology models
RNase E NTD homology models were generated for YpRNase E, FtRNase E, BpRNase E and AbRNase E in MOE using the sequences that had been trimmed to the NTD boundaries (Fig. 1), EcRNase E NTD crystal structures as the template structure (2BX218 for the closed conformation and 2VMK26 for the open conformation) and the default settings. Ten initial modes were produced and averaged to give the most energetically favourable model. The RMSD between each model and the template EcRNase E NTD structure was calculated in MOE.
Cloning of RNase E NTD expression strains
RNase E NTDs were cloned into the pET16b expression plasmid (Novagen), in order to express them as N-terminally His10-tagged proteins. EcRNase E NTD (aa 1–529) cloned into pET16b, was kindly provided by Prof. Ben Luisi (University of Cambridge, UK). The coding sequences for the RNase E NTDs were obtained from the European Nucleotide Archive: YpRNase E (aa 1–529; CAL20235.1), FtRNase E (aa 1–543; KFJ71366.1), BpRNase E (aa 1–532; EET07304.1), and AbRNase E (aa 1–544; A0A0B9WR03). The genes were codon-optimised in silico using GeneOtimizer® software (GeneArt, Life Technologies), synthesised by GeneArt and ligated between the NdeI and BamHI restriction sites of pET16b. All constructs were confirmed by DNA sequencing and E. coli BL21(DE3)pLysS was transformed with the plasmids for protein expression.
RNase E NTD expression strains were grown in 500 ml LB supplemented with 100 μg/ml ampicillin, at 37 °C, with shaking (250 rpm), until the OD600 reached 0.6. Recombinant protein expression was induced by the addition of isopropyl β-D-1-thiogalactopyranoside (IPTG) to a final concentration of 1 mM and cells were incubated for a further 3 hours, at 37 °C, with shaking. Cells were harvested by centrifugation at 7,000 rcf and 4 °C for 20 minutes and stored at −20 °C.
All RNase E NTDs were purified essentially as described previously for EcRNase E NTD19. Briefly, frozen cell pellets were thawed on ice and resuspended in 50 ml of lysis buffer (20 mM Tris-HCl pH 8, 500 mM NaCl and 20 mM imidazole) supplemented with a cOmpleteTM ethylenediaminetetraacetic acid (EDTA)-free protease inhibitor cocktail tablet (Roche). Cells were lysed by sonication for 10 minutes (3.3 seconds on, 9.9 seconds off) using a Sonics Vibra Cell VCX 500 sonicator. Each lysate was clarified by centrifugation at 40,000 g, at 4 °C, for 20 minutes and was subsequently loaded onto a 5 ml HisTrap FF column (GE Healthcare) equilibrated in lysis buffer using an ÄKTA Purifier (GE Healthcare). Bound proteins were eluted with a linear gradient from 20 to 500 mM imidazole (0 to 100% elution buffer (20 mM Tris pH 8, 500 mM NaCl and 500 mM imidazole)) applied over six column-volumes. Fractions containing RNase E NTD were pooled and buffer-exchanged into storage buffer (20 mM Tris pH 8, 500 mM NaCl, 10 mM MgCl2, 10 mM EDTA, 10 mM dithiothreitol (DTT) and 10% (v/v) glycerol) using a HiPrep 26/10 Desalting column (GE Healthcare) and an ÄKTA Purifier. RNase E NTD was concentrated using a Vivaspin 20 centrifugal concentrator with a molecular weight cut-off (MWCO) of 10 kDa (Sartorius) and loaded onto a HiLoad 16/600 Superdex 200 prep grade size-exclusion column (GE Healthcare) equilibrated in storage buffer using an ÄKTA Purifier. Fractions containing RNase E were pooled, concentrated using a Vivaspin 20 (MWCO 10 kDa) centrifugal concentrator and stored at −80 °C. The size and homogeneity of the purified RNase E NTDs was assessed by 12% SDS-PAGE. The identity of the purified protein was confirmed by liquid chromatography tandem mass spectroscopy (LC-MS-MS; Astbury Centre, University of Leeds, UK).
Size-exclusion chromatography coupled small-angle x-ray scattering (SEC-SAXS)
SEC-SAXS experiments were performed at the B21 beamline at the Diamond Light Source (DLS, Didcot, UK). RNase E NTDs were concentrated using a Vivaspin 20 (MWCO 10 kDa) centrifugal concentrator to a concentration of 5–10 mg/ml in sample buffer (10 mM DTT, 10 mM MgCl2, 0.5 M NaCl, 20 mM Tris (pH 8)). 45 μl were loaded onto a 4.6 ml KW403–4F size-exclusion column (Showdex) equilibrated in sample buffer at a flow rate of 0.1 ml/min, using an Agilent 1260 HPLC system. SAXS data were collected at 3-second intervals, at a wavelength of 12.4 KeV and a fixed camera length of 4.014 m using a Pilatus 2 M photon counting detector (Dectris), for 32 minutes. The data were normalised to the intensity of the incident beam and the scattering from the buffer was subtracted using an in-house program. Further data processing was then performed with the ATSAS suite36. PRIMUS37 was used to calculate the Rg and forward scattering intensity (I(0)) for the RNase E NTDs. I(0) was used to calculate the molecular weight (MW). The distance distribution function p(R) was generated with GNOM38 and was used to determine the Dmax. For EcRNase E NTD 20 ab initio models were generated in DAMMIF39 and averaged in DAMAVER40. Crystal structures for EcRNase E NTD in a closed conformation (2BX2;18), open conformation (2VMK;26) and a transitioning conformation (5F6C;25) were fitted to the averaged model using CHIMERA41. The theoretical scattering, based on the crystal structures, for EcRNase E NTD in each of these three conformations was compared to the experimental scattering for each RNase E NTD in CRYSOL42.
RNase E discontinuous assay
Discontinuous RNase E assays were carried out using one of three model RNA substrates: 5′ phosphorylated, 3′ FAM-labelled UUU ACA GUA UUU G (5′-p-RNA13-FAM-3′) 13-mer18,31; 5′ hydroxylated, 3′ FAM-labelled UUU ACA GUA UUU G (5′-OH-RNA13-FAM-3′) 13-mer or a partially double-stranded substrate (target-guide;32) generated by annealing a 5′ 4-(4-dimethylaminophenyl) diazenylbenzoic acid (DABCYL)-labelled, 3′ FAM-labelled GGA UCG GAG UUU UAA AUU AAU AAU AUA 27-mer and a 5′ phosphorylated UUU UCU CCG AUC C 13-mer at 28 °C for 15 minutes. The labelled 27-mer and unlabeled 13-mer were annealed at a ratio of 1:1.14 to ensure that all of the labelled 27-mer was partially double-stranded. All RNAs were obtained from Dharmacon. 30 μl reaction mixtures containing 1.25 to 5 nM RNase E NTD (see Figure legends for details), 1 μM substrate (see Figure legends for details), 25 mM Tris pH 8, 100 mM NaCl, 15 mM MgCl2, 1 mM DTT, 37.5 mg/ml Ficoll 70 and 5% (v/v) dimethyl sulfoxide (DMSO) were incubated at 28 °C for the indicated time(s). Reactions were terminated by the addition of 0.5 volumes of quench buffer (95% (v/v) formamide, 20 mM EDTA) and reaction products were resolved by denaturing 7.5 M urea 20% PAGE. Gels were visualised using a G:Box UV transilluminator (Syngene) and, where indicated, quantified by densitometry using ImageJ (Rasband WS, ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, http://imagej.nih.gov/ij/, 1997–2016).
SAXS data have been deposited at SASBDB under Accession Codes SASDE52, SASDE62, SASDE72, SASDE82 and SASDE92.
Arraiano, C. M. et al. The critical role of RNA processing and degradation in the control of gene expression. FEMS Microbiol. Rev. 34, 883–923 (2010).
Lawal, A., Jejelowo, O., Chopra, A. K. & Rosenzweig, J. A. Ribonucleases and bacterial virulence. Microb. Biotechnol. 4, 558–571 (2011).
Eidem, T. M., Roux, C. M. & Dunman, P. M. RNA decay: a novel therapeutic target in bacteria. Wiley Interdiscip. Rev. RNA 3, 443–54 (2012).
Canestrari, E. & Paroo, Z. Ribonucleases as drug targets. Trends Pharmacol. Sci. 39, 855–866 (2018).
Apirion, D. & Lassar, A. B. A conditional lethal mutant of Escherichia coli which affects the processing of ribosomal RNA. J. Biol. Chem. 253, 1738–1742 (1978).
Ono, M. & Kuwano, M. A conditional lethal mutation in an Escherichia coli strain with a longer chemical lifetime of messenger RNA. J. Mol. Biol. 129, 343–357 (1979).
McDowall, K. J., Hernandez, R. G., Lin-Chao, S. & Cohen, S. N. The ams-1 and rne-3071 temperature-sensitive mutations in the ams gene are in close proximity to each other and cause substitutions within a domain that resembles a product of the Escherichia coli mre locus. J. Bacteriol. 175, 4245–4249 (1993).
Carpousis, A. J., Luisi, B. F. & McDowall, K. J. Endonucleolytic initiation of mRNA decay in Escherichia coli. Prog. Mol. Biol. Transl. Sci. 85, 91–135 (2009).
Hammarlöf, D. L., Lijas, L. & Hughes, D. Temperature-sensitive mutants of RNase E in Salmonella enterica. J. Bacteriol. 193, 6639–6650 (2011).
Lee, E. J. & Groisman, E. A. An antisense RNA that governs the expression kinetics of a multifunctional virulence gene. Mol. Microbiol. 76, 1020–1033 (2010).
Yang, J., Jain, C. & Schesser, K. RNase E regulates the Yersinia type 3 secretion system. J. Bacteriol. 190, 3774–3778 (2008).
Aït-Bara, S. & Carpousis, A. J. RNA degradosomes in bacteria and chloroplasts: classification, distribution and evolution of RNase E homologs. Mol. Microbiol. 97, 1021–1035 (2015).
Kime, L. et al. The first small-molecule inhibitors of members of the ribonuclease E family. Sci. Rep. 5, 8028 (2015).
Lee, K. & Cohen, S. N. A Streptomyces coelicolor functional orthologue of Escherichia coli RNase E shows shuffling of catalytic and PNPase-binding domains. Mol. Microbiol. 48, 349–360 (2003).
McDowall, K. J. & Cohen, S. N. The N-terminal domain of the rne gene product has RNase E activity and is non-overlapping with the arginine-rich RNA-binding site. J. Mol. Biol. 255, 349–355 (1996).
Vanzo, N. F. et al. Ribonuclease E organizes the protein interactions in the Escherichia coli RNA degradosome. Genes Dev. 12, 2770–2781 (1998).
Callaghan, A. J. et al. Studies of the RNA degradosome-organizing domain of the Escherichia coli ribonuclease RNase E. J. Mol. Biol. 340, 965–979 (2004).
Callaghan, A. J. et al. Structure of Escherichia coli RNase E catalytic domain and implications for RNA turnover. Nature 437, 1187–1191 (2005).
Callaghan, A. J. et al. Quaternary structure and catalytic activity of the Escherichia coli ribonuclease E amino-terminal catalytic domain. Biochemistry 42, 13848–13855 (2003).
Callaghan, A. J. et al. “Zn-link”: a metal-sharing interface that organizes the quaternary structure and catalytic site of the endoribonuclease. RNase E. Biochemistry 44, 4667–4675 (2005).
Caruthers, J. M., Feng, Y., McKay, D. B. & Cohen, S. N. Retention of core catalytic functions by a conserved minimal ribonuclease E peptide that lacks the domain required for tetramer formation. J. Biol. Chem. 281, 27046–27051 (2006).
McDowall, K. J., Lin-Chao, S. & Cohen, S. N. A + U content rather than a particular nucleotide order determines the specificity of RNase E cleavage. J. Biol. Chem. 269, 10790–10796 (1994).
Mackie, G. A. Ribonuclease E is a 5′-end-dependent endonuclease. Nature 395, 720–723 (1998).
Chao, Y. et al. In vivo cleavage map illustrates the central role of RNase E in coding and non-coding RNA pathways. Mol. Cell 65, 39–51 (2017).
Bandyra, K. J., Wandzik, J. M. & Luisi, B. F. Substrate recognition and autoinhibition in the central ribonuclease RNase E. Mol. Cell 72, 275–285.e4 (2018).
Koslover, D. J. et al. The crystal structure of the Escherichia coli RNase E apoprotein and a mechanism for RNA degradation. Structure 16, 1238–1244 (2008).
Zeller, M. E. et al. Quaternary structure and biochemical properties of mycobacterial RNase E/G. Biochem. J. 403, 207–215 (2007).
Pendleton, J. N., Gorman, S. P. & Gilmore, B. F. Clinical relevance of the ESKAPE pathogens. Expert Rev. Anti. Infect. Ther. 11, 297–308 (2013).
Redko, Y. et al. Determination of the catalytic parameters of the N-terminal half of Escherichia coli ribonuclease E and the identification of critical functional groups in RNA substrates. J. Biol. Chem. 278, 44001–44008 (2003).
Kikhney, A. G. & Svergun, D. I. A practical guide to small angle X-ray scattering (SAXS) of flexible and intrinsically disordered proteins. FEBS Lett. 589, 2570–2577 (2015).
McDowall, K. J., Kaberdin, V. R., Wu, S. W., Cohen, S. N. & Lin-Chao, S. Site-specific RNase E cleavage of oligonucleotides and inhibition by stem-loops. Nature 374, 287–290 (1995).
Bandyra, K. J. et al. The seed region of a small RNA drives the controlled destruction of the target mRNA by the endoribonuclease RNase E. Mol. Cell 47, 943–953 (2012).
Kime, L., Jourdan, S. S., Stead, J. A., Hidalgo-Sastre, A. & McDowall, K. J. Rapid cleavage of RNA by RNase E in the absence of 5′ monophosphate stimulation. Mol. Microbiol. 76, 590–604 (2010).
Bouvier, M. & Carpousis, A. J. A tale of two mRNA degradation pathways mediated by RNase E. Mol. Microbiol. 82, 1305–1310 (2011).
Clarke, J. E., Kime, L., Romero, A. D. & McDowall, K. J. Direct entry by RNase E is a major pathway for the degradation and processing of RNA in Escherichia coli. Nucleic Acids Res. 42, 11733–11751 (2014).
Franke, D. et al. ATSAS 2.8: a comprehensive data analysis suite for small-angle scattering from macromolecular solutions. J. Appl. Cyrstallogr. 50, 1212–1225 (2017).
Konarev, P. V., Volkov, V. V., Sokolova, A. V., Koch, M. H. J. & Svergun, D. I. PRIMUS – a Windows-PC based system for small-angle scattering data analysis. J. Appl. Crystallogr. 36, 1277–1282 (2003).
Svergun, D. I. Determination of the regularization parameter in indirect-transform methods using perceptual criteria. J. Appl. Cyrstallogr. 25, 495–503 (1992).
Franke, D. & Svergun, D. I. DAMMIF, a program for rapid ab-initio shape determination in small-angle scattering. J. Appl. Crystallogr. 42, 342–346 (2009).
Volkov, V. V. & Svergun, D. I. Uniqueness of ab-initio shape determination in small-angle scattering. J. Appl. Crystallogr. 36, 860–864 (2003).
Pettersen, E. F. et al. UCSF Chimera – a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605–1612 (2004).
Svergun, D. I., Barberato, C. & Koch, M. H. J. CRYSOL – a program to evaluate x-ray solution scattering of biological macromolecules from atomic coordinates. J. Appl. Cyrstallogr. 28, 768–773 (1995).
We thank Prof. Ben Luisi (University of Cambridge, UK) for the EcRNase E NTD expression plasmid and for helpful discussions. We thank Dr Andy Scott (Defence Science and Technology Laboratory, UK) for helpful discussions. We thank Bailei Spelman (University of Portsmouth, UK) for assistance in protein expression and purification. We thank Nathan Cowieson and Nikul Khunti (Diamond Light Source, Didcot, UK) for SAXS technical support and Dr Heather Bruce (University of Portsmouth, UK) for assistance with SAXS data analysis. This work was supported by funding from Defence Science and Technology Laboratory. Funding for the open access charge was also from Defence Science and Technology Laboratory.
H.S.A. is an employee of Defence Science and Technology Laboratory who funded the research. All remaining authors declare no competing interests.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Mardle, C.E., Shakespeare, T.J., Butt, L.E. et al. A structural and biochemical comparison of Ribonuclease E homologues from pathogenic bacteria highlights species-specific properties. Sci Rep 9, 7952 (2019). https://doi.org/10.1038/s41598-019-44385-y