• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Structure. Oct 10, 2012; 20(10): 1641–1648.
PMCID: PMC3507626

The Crystal Structure of the Intact E. coli RelBE Toxin-Antitoxin Complex Provides the Structural Basis for Conditional Cooperativity


The bacterial relBE locus encodes a toxin-antitoxin complex in which the toxin, RelE, is capable of cleaving mRNA in the ribosomal A site cotranslationally. The antitoxin, RelB, both binds and inhibits RelE, and regulates transcription through operator binding and conditional cooperativity controlled by RelE. Here, we present the crystal structure of the intact Escherichia coli RelB2E2 complex at 2.8 Å resolution, comprising both the RelB-inhibited RelE and the RelB dimerization domain that binds DNA. RelE and RelB associate into a V-shaped heterotetrameric complex with the ribbon-helix-helix (RHH) dimerization domain at the apex. Our structure supports a model in which relO is optimally bound by two adjacent RelB2E heterotrimeric units, and is not compatible with concomitant binding of two RelB2E2 heterotetramers. The results thus provide a firm basis for understanding the model of conditional cooperativity at the molecular level.


► The E. coli RelB2E2 complex has an open V-shaped structure ► Isolated RelE is conformationally flexible ► The structure is not compatible with two copies binding adjacently on DNA ► The structure suggests a model for conditional cooperativity


The Escherichia coli relBE locus encodes a bacterial type II toxin-antitoxin (TA) complex consisting of a toxin, RelE, and its associated antitoxin, RelB (Gerdes et al., 2005). During nutritional stress, the labile RelB is degraded by Lon protease, rendering RelE capable of cleaving messenger RNA (mRNA) during translation on the ribosome, and thus globally downregulating translation (Pedersen et al., 2003). Structural studies of the RelE-ribosome interaction have shown that the toxin employs a reaction mechanism similar to that of bacterial RNase T1, but requires components of the ribosomal RNA in order to properly orient the substrate for endonucleolytic cleavage between nucleotides 2 and 3 of the A-site codon on mRNA (Neubauer et al., 2009). Binding of RelB inhibits RelE by displacement of the C-terminal, flexible α-helix, which contains a tyrosine residue (Tyr87) that is critical for catalysis (Li et al., 2009).

RelB is additionally known to dimerize through formation of a ribbon-helix-helix (RHH) DNA binding motif that allows transcriptional autoregulation through binding to the relO operator region (Li et al., 2008; Overgaard et al., 2009). Of interest, it was shown both in vivo and in vitro that transcriptional repression is subtly regulated and depends on the overall RelB:RelE ratio (Overgaard et al., 2008). RelB on its own binds DNA with a relatively low affinity, but binding is dramatically stimulated by addition of RelE up to a RelB:RelE ratio of 2:1. At higher RelE concentrations, the affinity for DNA is lost by a mechanism that has been termed conditional cooperativity (Cataudella et al., 2012; Overgaard et al., 2008, 2009). Mathematical modeling suggests that this system serves two main purposes: (1) to lower the level of free toxin in rapidly growing cells, and (2) to allow cells to return quickly to a low-toxin situation at the end of a starvation period (Cataudella et al., 2012).

The mechanism by which increased amounts of RelE are able to release RelB from DNA is unknown, but the structural basis for conditional cooperativity has been described for another TA locus, phd/doc (Garcia-Pino et al., 2010). In this case, the toxin (Doc) contains two antitoxin-binding sites, one with high affinity and one with low affinity. Binding of the antitoxin (Phd) to both sites is required for cooperative DNA binding and transcriptional repression. However, when the relative levels of toxin in the cell increase, the antitoxin molecules eventually will bind only to high-affinity sites, leading to the formation of a Doc-Phd2-Doc structure that is not compatible with DNA binding. The presence of high- and low-affinity binding sites for RelB on RelE has been proposed but never experimentally demonstrated (Overgaard et al., 2008). Furthermore, Phd/Doc and RelBE are unrelated in terms of sequence and structure, so a direct functional relationship between them cannot be inferred. Thus, despite numerous structural and functional studies of RelBE, the structural basis for the observed conditional cooperativity, and how transcriptional regulation occurs, remain unclear.

In this work, we describe the crystal structure of the intact E. coli RelB2E2 TA complex determined to 2.8 Å resolution, corresponding to the fully RelE-saturated complex. The structure reveals the spatial arrangement of inhibited RelE relative to the DNA-binding module of RelB and shows that RelB inhibits RelE through a conserved sequence motif that is also found within RelE itself. Two additional crystal structures of isolated RelE further show that the C-terminal helix uses this motif to adopt multiple, defined conformations. Finally, structural superpositioning shows that the distance between binding sites on DNA is incompatible with concomitant binding of two copies of the RelB2E2 heterotetramer, and therefore provides a structural and mechanistic framework for understanding the phenomenon of conditional cooperativity for the RelBE-type TA loci.


RelBE Complex Formation and Structure Determination

Due to translational coupling between the genes of the relBE operon, RelB is present in large excess over RelE when the proteins are coexpressed in E. coli from a construct representing the natural, genomic context. In order to isolate the fully RelE-saturated RelBE complex, we therefore employed a reciprocal denaturation-renaturation procedure for reconstitution (Overgaard et al., 2009). First, untagged RelE was obtained by on-column denaturation of complex of the R81A active-site mutant and His-tagged RelB, and likewise, untagged RelB was obtained by denaturation of a complex between RelB and His6-RelE. As noted previously, isolated RelB behaves as a tetramer in solution, whereas RelE is in a monomeric form (data not shown) (Cherny et al., 2007; Li et al., 2008). Finally, the RelB2E2 complex was reconstituted by mixing the proteins in the presence of excess RelE to produce the fully RelE-saturated complex.

Large, hexagonal crystals of the complex appeared in 1.9 M (NH4)2SO4 at pH 4.6 and diffracted to 2.8 Å. The crystals belong to space group P6122 with relatively large unit cell dimensions, and structure solution by molecular replacement using the known structures of RelE was not successful. Consequently, the structure was determined by the isomorphous replacement via an anomalous scattering method using multiple heavy atoms (MIRAS), built by iterative model-building, and refined to R (Rfree) = 25.3% (28.5%) (see Supplemental Experimental Procedures available online; Table 1). The final structure comprises three RelBE heterodimers in the crystallographic asymmetric unit (ASU). Two of these heterodimers associate into a heterotetramer by noncrystallographic symmetry, and one engages in a similar interaction through a crystallographic 2-fold axis. The structure covers most of RelE (residues 2–80 of 95) and one molecule of RelB (residues 2–79 of 79), whereas the other two RelB molecules lack either the C terminus (residues 2–69) or N terminus (residues 33–79) due to poor electron density (Figure 1A). The flexible, C-terminal helix of RelE (helix α3, residues 85–95) is disordered in all molecules in this crystal form. We also obtained two different crystal forms of isolated, untagged RelE in 30% w/v PEG 5.000, 0.2 M (NH4)2SO4 at pH 6.5, one of which was previously described (Neubauer et al., 2009). These crystals belong to space group P1211 and contain three molecules per ASU. The other crystal form, belonging to P212121, contains two molecules per ASU. Both structures were readily determined by molecular replacement using known structures of RelE (PDB 3KHA or 2K29) (Li et al., 2009; Neubauer et al., 2009), and were refined by iterative model building to R (Rfree) = 18.4% (21.9%) for the P1211 form and R (Rfree) = 23.6% (28.2%) for the P212121 form (Table 1).

Figure 1
Structural Overview of the E. coli RelBE Complex
Table 1
Crystallographic Data and Refinement Statistics

RelB2E2 Has an Open V-Shaped Structure

In the RelBE complex structure, each RelE tightly binds the C-terminal region of its associated RelB (residues 50–79) through displacement of helix α3 as described previously (Li et al., 2009). Two neighboring RelB molecules dimerize at their N termini to form a RHH-type DNA binding motif that closely resembles the structure of the isolated dimerization domain determined by nuclear magnetic resonance (NMR) (Li et al., 2008). In the resulting heterotetrameric complex, the extended conformation of helix α3 of RelB results in an overall very open, V-shaped complex with an RelE-RelB2-RelE architecture and approximate dimensions of 105 × 60 Å (Figure 1B, top). The open structure of the heterotetramer suggests a flexible structure; however, closer inspection of the RelB linker region reveals a unique and stable turn structure centered on Pro45* (throughout this work, residues in RelB are marked with *). In the heterotetramer, the two equivalent proline residues are juxtaposed in a symmetrical arrangement that anchors the hydrophobic core of the dimerization domain through interactions between several residues, including Tyr37*, Phe46*, Gln48*, and Arg43*. The functional properties of these residues are quite well conserved among RelB homologs (Figure 1A and Figure S1, purple boxes), so in summary, we believe that the V-shaped structure is relatively rigid and therefore most likely represents the conformation found in vivo.

Inside the crystal, three heterotetramers pack together to form a remarkably compact, nearly spherical superstructure with three RelB dimerization domains at the surface (Figure 1C). This dodecamer has a strikingly large buried surface area of 31.920 Å2 (ΔGint = −164.0 kcal/mol), compared with 6.810 Å2 (−43.6 kcal/mol) for the RelB2E2 tetramer as estimated via the Protein Interactions, Surfaces and Assemblies (PISA) server (Krissinel and Henrick, 2007). To assess the potential biological significance of this higher-order structure, we analyzed the RelBE complex in solution by analytical ultracentrifugation (AUC; Table S1) and single-particle cryoelectron microscopy. However, the AUC results show that 96%–99% of the complex is in a tetrameric form in solution, and we were also not able to observe any high-molecular-weight species using cryoelectron microscopy or chemical cross-linking (data not shown). We therefore conclude that the dodecamer is a result of crystal packing and is not representative of the architecture of RelBE in solution.

Isolated RelE Dimerizes via a Domain-Swap Interaction

To better assess the structural rearrangements that take place in RelE upon RelB binding, we determined the structures of isolated RelE present in the two crystal forms obtained for the isolated toxin (P21 and P212121). The structure of the P21 form was previously resolved at 2.5 Å (Neubauer et al., 2009). However, reinvestigation of this structure using better data collected to 1.8 Å revealed that although one of the three molecules in the ASU is indeed in the monomeric form reported earlier (Figure 2A), the two other molecules dimerize via a symmetrical domain-swap interaction involving helix α3. In the structure of the P212121 crystal form, which has two molecules in the ASU, RelE also forms a domain-swap dimer involving the C-terminal helix; however, this interaction is not symmetrical (Figure 2A). The interaction patterns observed between helix α3 and the core domain are identical in the RelE monomer, symmetrical dimer, and one molecule of the asymmetrical dimer, whereas they differ in the other molecule (Figure 2B). The most common interaction is characterized by a strong salt bridge between Arg93 and Glu14 in addition to multiple contacts between hydrophobic residues such as Val86 and Tyr87 (Figure 2B, left). In the alternative conformation observed in the asymmetrical dimer, the helix is pulled farther back and Arg93 now makes a hydrophobic interaction by stacking its guanidinium group on Phe74 (Figure 2B, middle).

Figure 2
RelE Dimerization and RelB Binding

These structures are consistent with NMR studies of RelE and its complex with the interacting helix of RelB, which showed that the helix from RelB lies at an angle of 36° with respect to the orientation of helix α3 in isolated RelE (Li et al., 2009). However, closer analysis reveals that the backbones overlap nearly perfectly at the C-terminal end of the RelB helix, where the interactions are strongest. At the atomic level, the interactions observed here are also surprisingly similar: Arg93 in RelE, which interacts strongly with Glu14 in isolated RelE, superimposes perfectly with Arg65* from RelB in the complex (Figure 2B, right). Likewise, both Val86 and Ala90 have structurally equivalent residues in RelB (Leu58* and Val62*; Figure 1A and Figure 2B, right). At the position of Tyr87, which is required for the endonuclease activity of RelE, RelB has a valine (Val59*), thus providing the hydrophobicity while removing the functional group. In summary, there appears to be a consensus motif by which helical interactions with the RelE core occur, which can be expressed as ZXnnZnnRZ (where n is any amino acid, X is a hydrophobic amino acid, and Z is a small hydrophobic amino acid [Val, Ile, Leu, or Ala]). Looking across a wider range of RelE and RelB sequences from various bacteria, this pattern appears to be well conserved (Figure S1).

A Dimer of RelB2E—But Not RelB2E2—Can Bind to DNA

The RelB dimerization domain belongs to the RHH family of DNA binding proteins, which function by inserting two adjacent β-strands into the major groove (Li et al., 2008; Schreiter and Drennan, 2007). A well-described member of this family is the bacteriophage P22 Arc repressor, for which a DNA-bound structure has been determined (PDB ID 1BDT) (Raumann et al., 1994). The arc operator consists of two binding sites, each of which has an AT-rich center that allows DNA bending, and has many similarities to the relO operator even though it is one basepair shorter (Figure 3A). The crystal structure of the DNA-bound Arc repressor showed that the two sites support binding of a dimer of Arc in each of two adjacent major grooves, and this has been used to create a model for RelB binding to DNA based on the observation that either RelB2 or RelB2E can bind each leg of the operator (Overgaard et al., 2009; Figure 3B).

Figure 3
RelB Binds DNA via an Arc-Like Motif

Functional studies of operator binding have shown that gradual addition of RelE greatly stimulates DNA binding by RelB at subequimolar quantities but diminishes binding at equimolar ratios, a phenomenon that has been termed conditional cooperativity (Overgaard et al., 2008, 2009). To explain this phenomenon, Overgaard et al. (2008) proposed that RelE harbors two RelB binding sites: a high-affinity binding site used for the catalytic inhibition of RelE, and a low-affinity site required for binding of the complex to DNA in a 1:2 RelE:RelB ratio. This model suggests that when the RelE:RelB ratio increases, RelB will only bind RelE via the high-affinity site and thus release the complex from DNA.

To understand the proposed model in structural terms, we used the structure of DNA-bound Arc to analyze the consequences of binding of the complete RelB2E2 tetramer to DNA. Structural alignment of two complete RelB2E2 tetramers with their RelB dimerization domains in adjacent major grooves immediately suggests why the complex cannot bind at high RelE concentrations: Due to the close proximity of the binding sites on DNA, binding of two adjacent tetramers generates an overlapping, W-shaped complex in which the two RelE molecules at each end of the complex (denoted as RelE and RelE′ in Figure S2A) clash. In contrast, if the two most central RelE molecules are removed (corresponding to lowering of the RelE:RelB ratio to 1:2), there are no significant clashes and the four remaining RelB molecules pack accurately together along the axis of the DNA duplex (Figure 3C). Thus, importantly, this architecture permits extensive interactions between all four RelB molecules in the complex, but does not require a secondary low-affinity binding site on RelE. Our structure thus predicts that the RelB2E species, which has been shown experimentally to have the highest affinity for DNA, is in fact a RelE-RelB2-RelB2-RelE W-shaped heterohexameric complex.


In this work, we show that the intact, RelE-saturated RelBE complex from E. coli has an unusual V-shaped structure organized by the RelB dimerization domain, and conserved interactions in the loop that connect this domain to the RelE-interaction motif. In this structure, the two RelB-bound RelE molecules are at the distant ends of the V and clash when two RelB dimerization domains bind adjacently on DNA. However, release of one molecule of RelE from each tetramer would allow simultaneous binding of two complexes at the operator site. In contrast to previous models, this structure has the two RelE molecules located on the same side of DNA and is consistent with a tight interaction among all four RelB molecules, thus explaining the observed cooperativity in binding and placing strong restraints on both the distance and angle between the binding sites (Overgaard et al., 2009). Our results thus provide a structural basis for understanding the phenomenon of conditional cooperativity for RelBE-type TA systems at the molecular level.

Furthermore, investigation of other bacterial TA structures reveals that the V shape may be a relatively common architecture that up to now has not been properly appreciated. Methanococcus jannaschii RelBE (MjRelBE) contains a minimal RelB interaction domain that is not of the RHH type, yet the overall V-shaped structure is highly reminiscent of the E. coli RelB2E2 complex (Figure S3; Francuski and Saenger, 2009). The structure of RelBE2 from Mycobacterium tuberculosis also has a similar architecture, although the long helix in RelB is highly bent (Miallau et al., 2012). Finally, it is noteworthy that the tetrameric structure proposed for Doc-Phd2-Doc based on a combination of small-angle X-ray scattering data and a crystal structure of Phd2-Doc-Phd2 shows a similar arrangement even though the proteins are completely unrelated and have different mechanisms of repression (Figure S3; Garcia-Pino et al., 2010). In summary, we suggest that a V-shaped architecture with the DNA-binding domain at the apex may be a general feature of bacterial TA complexes.

The model of heterohexameric RelE-RelB4-RelE allows for a significant number of direct RelB-RelB interactions between the two dimers, both in the RHH domain and along helix α3, thus supporting the observation that pure RelB can form tetramers and bind DNA cooperatively (Cherny et al., 2007; Li et al., 2008). Furthermore, biochemical studies have shown that a RelB fragment covering only residues 1–65 is dimeric. The RelB tetramer model implicates the region around Arg65*, which points toward Asp53* of the adjacent RelB in the model as being important for the weak cooperativity observed for DNA binding by isolated RelB (Figure S2B; Li et al., 2008; Overgaard et al., 2008). On the other hand, the observation that the RelB construct 1–50 does not show cooperativity implies that interactions between the RHH domains are not critical for this phenomenon (Li et al., 2008).

Although our model does not require a second, low-affinity binding site for RelB on RelE to contribute to cooperativity in the toxin-bound form, it is fully consistent with its existence. Superpositioning of the structure of the C-terminal region of RelB (residues 70–79) from the molecule for which this region is visible onto the innermost RelB molecule of the W indicates that a direct interaction with RelE is possible (Figures S2C and S2D). More precisely, the low-affinity binding site on RelE predicted by this model would consist of the loops between α2/β2 and β2/β3 in RelE and possibly involve interactions with Arg38, Asp49, Lys43, and Glu69. Importantly, however, our structure shows that the presence of a high- and a low-affinity binding site for the antitoxin on the toxin is not a prerequisite for conditional cooperativity, because there are direct RelB-RelB dimer interactions that could potentially be fully responsible for the observed cooperativity if they were stabilized additionally by RelE binding to one RelB molecule. In addition, we note that the architecture of RelBE is different from that of Phd/Doc, in which a Doc toxin molecule bridges two Phd dimers on DNA, leading to the possibility of polymerization as observed by multiple DNA gel shift bands in vitro (Garcia-Pino et al., 2010). In contrast, the closed, W-shaped architecture of the RelBE heterohexamer does not allow for polymerization, which is consistent with the observation that a maximum of two DNA band shifts are observed in vitro (Overgaard et al., 2008).

Taken together, our results suggest a model for DNA binding in which during normal, rapid growth, RelB is expressed in excess of RelE, and a mixture of symmetrical RelB2E2 and RelB2 as well as asymmetrical RelB2E complexes will most likely be present, but only the trimer will bind to DNA (Figure 4; Overgaard et al., 2009). Binding of the first trimer strongly promotes binding of an additional complex to the adjacent site on the operator due to favorable interactions between the trimers, eventually causing transcriptional shutdown (Figure 4, top). In contrast, when cells experience nutritional stress and consequently translation slows down, the levels of the labile RelB molecule drop, thus increasing the overall RelE:RelB ratio (Figure 4, bottom). Under these circumstances, an additional RelE molecule will bind to a free RelB C terminus in the heterohexameric complex on the DNA operator, leading to release of RelE-RelB2-RelE. This leaves a single RelB2E trimer bound to DNA, which will also bind an additional RelE molecule. In this context, we note that DNA band-shift experiments conducted at a high RelE:RelB ratio showed a faint protein-DNA complex that might correspond to a single bound tetramer (Li et al., 2008; Overgaard et al., 2008). Furthermore, surface plasmon resonance measurements revealed that titration of RelE into DNA-bound RelB2E led to the formation of a new, stable complex, suggesting that at least in vitro, a single RelB2E2 (or RelB2E) complex may remain bound to the operator even at a very high level of RelE. Finally, it was found that high levels of RelE could not displace RelB2E from a single operator site (Overgaard et al., 2008). However, it is likely that the affinity of a single complex for DNA in vivo is too low to prevent polymerase binding and, hence, transcription.

Figure 4
Model for RelBE DNA Binding and Conditional Cooperativity

Experimental Procedures

Protein Expression and Purification

Untagged RelER81A was purified by denaturation and refolding from E. coli BL21 DE3 (Novagen) harboring a bicistronic construct based on pMG25, expressing both RelER81A and His-tagged RelB as previously described (Christensen-Dalsgaard et al., 2008; Neubauer et al., 2009). Untagged RelB was purified in a reciprocal way using the plasmid pSC2524HE encoding His-tagged RelER81A and untagged RelB (Christensen-Dalsgaard et al., 2008). In both cases, the untagged protein was further purified by ion exchange and gel filtration into a final buffer containing 25 mM HEPES, pH 7.0, 100 mM KCl, and 5 mM β-mercaptoethanol (BME; see Supplemental Experimental Procedures for details). Complex formation was achieved by mixing RelE and RelB in the presence of an excess of RelE before final separation by gel filtration and concentration to 9 mg/ml.

Crystallization and Structure Determination of the RelBE Complex

Hexagonal crystals containing the RelB2E2 complex grew at 4°C in 1+1 μl sitting-drop vapor diffusion drops with a reservoir of 1.6−2.0 M ammonium sulfate and 0.1 M Na acetate, pH 4.6. Cryoprotection was achieved by gradual transfer of the crystals into 20% glycerol and heavy atom soaks prepared by addition of small amounts of heavy atom salts to the cryo solution. Native data and data from HA-soaked crystals were collected at the MAX-Lab (Lund, Sweden) and processed using XDS (Kabsch, 2010) for the derivative data set and Xia2 (Winter, 2010) for the native set. HA positions were initially located using RANTAN (Collaborative Computational Project, Number 4, 1994), and an improved density-modified MIRAS map was subsequently obtained using only the Pt and Hg derivatives in SHARP (Bricogne et al., 2003). Refinement was carried out by iterative model building in Phenix (Adams et al., 2004) and Coot (Emsley et al., 2010) to a final R (Rfree) of 25.3% (28.5%; see Table 1 for details).

Crystallization and Structure Determination of Isolated RelE

Full-length E. coli RelER81A was expressed, purified, and crystallized as described previously (Neubauer et al., 2009). Closer inspection of the crystallization drops revealed that they contained two morphologically different, three-dimensional crystal forms, and native data sets were collected from both types. For the previously described crystal form, belonging to the space group P21 (P1211) with three molecules per ASU, improved data extending to 1.8 Å were obtained at beamline ID29 of the European Synchrotron Radiation Facility (ESRF). The other crystal form turned out to belong to space group P212121 with two RelE molcules per ASU, and for this form, native data were collected at beamline X12 of the European Molecular Biology Laboratory-Deutsches Elektronen Synchrotron (EMBL-DESY) to a maximum resolution of 2.4 Å (Table 1). All data sets were processed using XDS (Kabsch, 2010) or Mosflm via Xia2 (Powell, 1999) and the structures were solved by molecular replacement in Phenix/Phaser (Adams et al., 2004; McCoy et al., 2007) using a search model derived from the published crystal structure of monomeric RelE (PDB ID 3KHA) (Neubauer et al., 2009). From the map generated by Phaser, the models were fitted and rebuilt to include the C-terminal helix by iterative refinement in Phenix and rebuilding in Coot. The final R (Rfree) was 18.4% (21.9%) for the P1211 form and 23.6% (28.2%) for the P212121 form (see Table 1 and Supplemental Experimental Procedures for details).


Purified and reconstituted RelBE complex was analyzed by AUC at a concentration of ~31 μM using an Optima XL-I analytical ultracentrifuge (Beckmann) at 45,000 rpm and 20°C in 25 mM HEPES, pH 7.0, 150 mM KCl, and 5 mM BME. Data analysis was carried out in Sedfit (Schuck, 2000) using a bimodal distribution of f/fo ratios (see Supplemental Experimental Procedures for details).


We thank Kenn Gerdes for kindly providing the RelB-RelE-His6 expression construct and Stephen H. McLaughlin at the MRC-LMB Biophysics Facility for help with AUC. We are also grateful to the staffs of the MAX-Lab, ESRF, and DESY for their assistance with X-ray data collection. This work was supported by the Danish National Research Foundation’s Centre for mRNP Biogenesis and Metabolism, and a grant from the Novo Nordisk Foundation to D.E.B. A.D.E. was supported by a Woolf Fisher Scholarship and grant U105192715 from the Medical Research Council.

Supplemental Information

Document S1. Supplemental Experimental Procedures, Figures S1–S3, and Table S1:


Adams P.D., Gopal K., Grosse-Kunstleve R.W., Hung L.W., Ioerger T.R., McCoy A.J., Moriarty N.W., Pai R.K., Read R.J., Romo T.D. Recent developments in the PHENIX software for automated crystallographic structure determination. J. Synchrotron Radiat. 2004;11:53–55. [PubMed]
Bricogne G., Vonrhein C., Flensburg C., Schiltz M., Paciorek W. Generation, representation and flow of phase information in structure determination: recent developments in and around SHARP 2.0. Acta Crystallogr. D Biol. Crystallogr. 2003;59:2023–2030. [PubMed]
Cataudella I., Trusina A., Sneppen K., Gerdes K., Mitarai N. Conditional cooperativity in toxin-antitoxin regulation prevents random toxin activation and promotes fast translational recovery. Nucleic Acids Res. 2012;40:6424–6434. [PMC free article] [PubMed]
Cherny I., Overgaard M., Borch J., Bram Y., Gerdes K., Gazit E. Structural and thermodynamic characterization of the Escherichia coli RelBE toxin-antitoxin system: indication for a functional role of differential stability. Biochemistry. 2007;46:12152–12163. [PubMed]
Christensen-Dalsgaard M., Overgaard M., Winther K.S., Gerdes K. RNA decay by messenger RNA interferases. Methods Enzymol. 2008;447:521–535. [PubMed]
Collaborative Computational Project, Number 4 The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D Biol. Crystallogr. 1994;50:760–763. [PubMed]
Emsley P., Lohkamp B., Scott W.G., Cowtan K. Features and development of Coot. Acta Crystallogr. D Biol. Crystallogr. 2010;66:486–501. [PMC free article] [PubMed]
Francuski D., Saenger W. Crystal structure of the antitoxin-toxin protein complex RelB-RelE from Methanococcus jannaschii. J. Mol. Biol. 2009;393:898–908. [PubMed]
Garcia-Pino A., Balasubramanian S., Wyns L., Gazit E., De Greve H., Magnuson R.D., Charlier D., van Nuland N.A., Loris R. Allostery and intrinsic disorder mediate transcription regulation by conditional cooperativity. Cell. 2010;142:101–111. [PubMed]
Gerdes K., Christensen S.K., Løbner-Olesen A. Prokaryotic toxin-antitoxin stress response loci. Nat. Rev. Microbiol. 2005;3:371–382. [PubMed]
Kabsch W. Xds. Acta Crystallogr. D Biol. Crystallogr. 2010;66:125–132. [PMC free article] [PubMed]
Krissinel E., Henrick K. Inference of macromolecular assemblies from crystalline state. J. Mol. Biol. 2007;372:774–797. [PubMed]
Li G.Y., Zhang Y., Inouye M., Ikura M. Structural mechanism of transcriptional autorepression of the Escherichia coli RelB/RelE antitoxin/toxin module. J. Mol. Biol. 2008;380:107–119. [PubMed]
Li G.Y., Zhang Y., Inouye M., Ikura M. Inhibitory mechanism of Escherichia coli RelE-RelB toxin-antitoxin module involves a helix displacement near an mRNA interferase active site. J. Biol. Chem. 2009;284:14628–14636. [PMC free article] [PubMed]
McCoy A.J., Grosse-Kunstleve R.W., Adams P.D., Winn M.D., Storoni L.C., Read R.J. Phaser crystallographic software. J. Appl. Cryst. 2007;40:658–674. [PMC free article] [PubMed]
Miallau, L., Chernishof, I., Chiang, J., Arbing, M., Cascio, D., Eisenberg, D. (2012). The crystal structure of the toxin-antitoxin complex RelBE2 (Rv2865-2866) from Mycobacterium tuberculosis. http://www.rcsb.org/pdb/explore/explore.do?structureId=3G5O.
Neubauer C., Gao Y.G., Andersen K.R., Dunham C.M., Kelley A.C., Hentschel J., Gerdes K., Ramakrishnan V., Brodersen D.E. The structural basis for mRNA recognition and cleavage by the ribosome-dependent endonuclease RelE. Cell. 2009;139:1084–1095. [PMC free article] [PubMed]
Overgaard M., Borch J., Jørgensen M.G., Gerdes K. Messenger RNA interferase RelE controls relBE transcription by conditional cooperativity. Mol. Microbiol. 2008;69:841–857. [PubMed]
Overgaard M., Borch J., Gerdes K. RelB and RelE of Escherichia coli form a tight complex that represses transcription via the ribbon-helix-helix motif in RelB. J. Mol. Biol. 2009;394:183–196. [PMC free article] [PubMed]
Pedersen K., Zavialov A.V., Pavlov M.Y., Elf J., Gerdes K., Ehrenberg M. The bacterial toxin RelE displays codon-specific cleavage of mRNAs in the ribosomal A site. Cell. 2003;112:131–140. [PubMed]
Powell H.R. The Rossmann Fourier autoindexing algorithm in MOSFLM. Acta Crystallogr. D Biol. Crystallogr. 1999;55:1690–1695. [PubMed]
Raumann B.E., Rould M.A., Pabo C.O., Sauer R.T. DNA recognition by beta-sheets in the Arc repressor-operator crystal structure. Nature. 1994;367:754–757. [PubMed]
Schreiter E.R., Drennan C.L. Ribbon-helix-helix transcription factors: variations on a theme. Nat. Rev. Microbiol. 2007;5:710–720. [PubMed]
Schuck P. Size-distribution analysis of macromolecules by sedimentation velocity ultracentrifugation and lamm equation modeling. Biophys. J. 2000;78:1606–1619. [PMC free article] [PubMed]
Winter G. xia2: an expert system for macromolecular crystallography data reduction. J. Appl. Cryst. 2010;43:186–190.


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...