SEARCH

SEARCH BY CITATION

Keywords:

  • chromosomal toxin–antitoxin systems;
  • E. coli;
  • dinJ-yafQ operon

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

Bacterial toxin–antitoxin (TA) systems are operons that code for a stable toxic protein and a labile antitoxin. TA modules are widespread on the chromosomes of free-living Bacteria and Archaea, where they presumably act as stress response elements. The chromosome of Escherichia coli K-12 encodes four known TA pairs, as well as the dinJ-yafQ operon, which is hypothesized to be a TA module based on operon organization similar to known TA genes. Induction of YafQ inhibited cell growth, but its toxicity was counteracted by coexpression of dinJ cloned on a separate plasmid. YafQ(His)6 and DinJ proteins coeluted in Ni2+-affinity and gel filtration chromatography, implying the formation of a specific and stable YafQ-DinJ protein complex with an estimated molecular mass of c. 37.3 kDa. Induction of YafQ reduced protein synthesis up to 40% as judged by incorporation of [35S]-methionine, but did not influence the rates of DNA and RNA synthesis. Structure modelling of E. coli YafQ revealed its structural relationship with bacterial toxins of known structure suggesting that it might act as a sequence-specific mRNA endoribonuclease.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

Microorganisms respond to environmental stress by numerous highly sophisticated means, sometimes using intriguing molecular mechanisms. As it has recently emerged from several studies, bacterial toxin–antitoxin (TA) systems could be involved in adaptation to stress conditions by modulating the global level of biological processes such as translation or DNA synthesis (Buts et al., 2005; Gerdes et al., 2005; Condon, 2006).

Bacterial TA systems typically consist of two genes organized in an operon that codes for a stable toxin and proteolytically labile antitoxin. Toxins inhibit essential cellular processes, causing cell death or bacteriostasis, while antitoxins bind to toxins and block their activity by forming a nontoxic complex (Gerdes et al., 2005). TA systems were firstly observed on low-copy plasmids of Escherichia coli and other bacteria where they are responsible for the postsegregational killing effect, resulting in the death of cells that have lost the plasmid (Gerdes et al., 1986). In the absence of plasmid-directed de novo protein synthesis in a daughter cell, the antitoxin is degraded faster by cellular proteases than the more stable toxin, allowing the toxin to kill the cell. The plasmid-borne TA systems have been named ‘addiction modules’ as they cause cells to be ‘addicted’ to the short-lived antitoxin product because its synthesis is essential for the survival of the cells (Gerdes et al., 1986).

Pairs of genes homologous to some of the extrachromosomally borne addiction modules have been found on the chromosome of E. coli and appear to be widely spread in the genomes of other Bacteria and Archaea (Pandey & Gerdes, 2005).

Chromosomal TA systems have a typical organization, whereby the antitoxin gene in most cases precedes the toxin gene, and both genes are coexpressed (Gerdes et al., 2005). Toxins are usually basic proteins of ∼11–12 kDa, whereas antitoxins (∼9 kDa) are acidic and highly unstable. Antitoxins are quickly degraded by specific cellular proteases, and when the expression of a TA operon decreases for some reason, the more stable toxins are released from the protein complex and inhibit cell growth (Buts et al., 2005).

There are at least five E. coli chromosome-encoded TA loci: relBE, mazEF, chpBIK, yefM-yoeB and dinJ-yafQ (Gerdes et al., 2005). It was demonstrated recently that toxins RelE, MazF, ChpBK and YoeB inhibit translation through cleavage, either direct or indirect, of translated mRNAs at specific codons (Christensen et al., 2003, 2004; Pedersen et al., 2003; Zhang et al., 2003, 2005). mRNA degradation, triggered by RelE and MazF, is induced under nutritional stress such as amino acid and glucose starvation (Christensen et al., 2001, 2003). An inhibitory effect on translation, caused by MazF or RelE toxins, was reversed by the subsequent expression of their cognate antitoxins (Pedersen et al., 2002). These observations propose a model in which chromosome-borne TA systems such as MazEF and RelBE might function as modulators of metabolic processes in response to nutritional or environmental stress (Gerdes et al., 2005; Condon, 2006).

The Escherichia coli dinJ-yafQ operon codes for a protein pair that has been grouped into the bacterial relBE TA family (Gerdes et al., 2005). Except for its genomic organization, typical of TA modules, and a weak sequence similarity to some chromosomal TA gene pairs, no structural or functional features have been reported for dinJ-yafQ so far.

In the present work, we studied whether E. coli dinJ-yafQ system is an active bacterial TA module. Towards this goal, we examined cell growth and viability upon expression of the operon genes, cloned on separate plasmids. Experiments addressing the interactions between YafQ and DinJ proteins were carried out. The effect of induction of YafQ on DNA, RNA and protein syntheses was also tested in order to determine its possible cellular target.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

All the materials listed below were purchased from Sigma-Aldrich, Merck and Roth. Restriction enzymes, other DNA- and RNA-modifying enzymes and kits for molecular biology were from AB Fermentas. All enzymes were used as recommended by the supplier.

Bacterial strains, plasmids and growth conditions

The bacterial strains and plasmids used in this study are listed in Table 1. Escherichia coli strains were grown in liquid or solid Luria–Bertani (LB) and in M9 minimal medium with 0.5% glycerol. Antibiotics were added at the following concentrations: ampicillin, 100 μg mL−1, kanamycin, 60 μg mL−1, and chloramphenicol, 33 μg mL−1.

Table 1.   Bacterial strains and plasmids
Strain/plasmidGenotype/descriptionReference/source
Strain
 BL21 (DE3)FompT rBmBStudier & Moffat (1986)
 BW25113Δ(araD-araB)567 ΔlacZ4787(::rrnB-4) lacIp-400(lacIQrpoS396(Am) rph-1 Δ(rhaD-rhaB)568 rrnB-4 hsdR514Datsenko & Wanner (2000)
 BW25113F′BW25113/F′proA+B+lacIqΔlacZ)M15 zzf::mini-Tn10 (KanR)This work
Plasmid
 pBAD30Expression plasmidGuzman et al. (1995)
 pBAD-dinJ620 bp DNA fragment with dinJ gene, cloned into pBAD30Motiejūnaitėet al. (2005)
 pBAD-dinJyafQ931 bp DNA fragment with dinJ-yafQ operon cloned into pBAD30Motiejūnaitėet al. (2005)
 pBAD-yafQ376 bp fragment with yafQ gene cloned into pBAD30Motiejūnaitėet al. (2005)
 pET28bExpression plasmidNovagen
 pET28b-dinJyafQ(His)6559 bp DNA fragment, with dinJ and yafQ genes, cloned into pET28bThis work
 pET28b-dinJ831 bp DNA fragment containing dinJ gene, cloned into pET28bLaboratory collection
 pUHE 25-2Expression plasmidBujard et al. (1987)
 pUHE 25-2(cat)pUHE 25-2 with with bla gene replaced by cat geneThis work
 pUHE dinJ667 bp fragment with dinJ gene, cloned into pUHE 25-2(cat)This work

DNA manipulation and plasmid construction

Plasmids pBAD-yafQ, pBAD-dinJ and pBAD-dinJyafQ were constructed as described previously (Motiejūnaitėet al., 2005). Plasmid pUHE-dinJ was constructed by firstly replacing the bla gene with a cat gene in pUHE 25-2 plasmid (Bujard et al., 1987). Then, a 667 bp DNA fragment harbouring the dinJ gene was cut with EcoRI and HindIII from the pBAD-dinJ plasmid and inserted between the EcoRI and HindIII sites of pUHE 25-2(cat). To construct plasmid pET28b-dinJyafQ(His)6, a 559 bp DNA fragment with the dinJ-yafQ operon was PCR-amplified from plasmid pBAD-dinJyafQ with primers HisJ1 (5′-TACCCCATGGCTGCTAAC-3′) and YafQ6 (5′-GAAGCGGCTCGAGCCCAAAGA-3′), which contain NcoI and XhoI restriction sites (underlined), respectively. The resulting fragment was cut with NcoI and XhoI and inserted between the NcoI and XhoI sites of vector pET28b (Novagen). In this construct, an in-frame translational fusion with a six-His tag was created at the C-terminus of YafQ.

Growth rate analysis

Escherichia coli BW25113F′ cells, harbouring derivatives of pBAD30 and pUHE 25-2(cat) plasmids (Table 1), were grown at 37°C in liquid LB medium with 0.2% glucose to an A590 nm=0.3–0.5. Then, the culture was divided into two equal parts. The growth medium of one part was changed into fresh LB medium with 0.2% glucose, and that of the other into fresh LB medium with 0.2% arabinose and 1 mM isopropyl-β-D-thiogalactopyranoside (IPTG). Incubation was continued at 37°C. Bacterial growth was assessed by measuring A590 nm every 30 min.

For growth analysis on solid medium, BW25113F′ cells harbouring derivatives of pBAD30 and pUHE 25-2(cat) plasmids were grown in liquid LB medium with 0.2% glucose as described above, diluted 106 times and plated on solid LB and minimal M9 medium plates with 1 mM IPTG and 0.2% arabinose. The plates were incubated at 37°C overnight. When needed, 0.1% of casamino acids were added to M9 medium.

Protein purification and analysis

For purification of DinJ and YafQ(His)6 proteins, plasmid pET28b-dinJyafQ(His)6 was introduced into E. coli strain BL21(DE3). Both proteins were induced by incubation with 0.4 mM IPTG for 2 h. The DinJ/YafQ(His)6 protein complex was purified by affinity chromatography on a nickel-NTA agarose column (Qiagen). Proteins were eluted with 20 mM Tris-HCl buffer (pH 7.9) containing 500 mM NaCl and 250 mM imidazole. Eluted fractions were dialysed against 0.05 M NaH2PO4 buffer (pH 7.0) containing 0.15 M NaCl and further analysed by FPLC gel filtration chromatography using Superose 12 HR column (Pharmacia). The protein molecular mass standard curve included bovine serum albumin (67 kDa); ovalbumin (43 kDa); β-lactoglobulin (35 kDa); carbonic anhydrase (29 kDa); mioglobin (17.8 kDa); and RNase A (13.7 kDa). The eluted fractions were analysed by 13.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The relative quantities of the proteins present in the eluted fractions were determined by an Image Master densitometer (Amersham Biosciences).

YafQ(His)6 was purified by dissociating it out of the DinJ/YafQ(His)6 complex in 6 M guanidine-HCl. YafQ(His)6 was then retrapped on a Ni-NTA resin and refolded by stepwise dialysis in 25 mM HEPES buffer (pH 7.5), containing 0.1 mM DTT and 200 mM NaCl. The DinJ protein with a C-terminal hexa-histidine tag was expressed from plasmid pET28b-dinJ(His)6 (Table 1), purified by affinity chromatography using Ni-NTA agarose and used for reference.

Assays for DNA and RNA synthesis rates in vivo

Escherichia coli BW25113 cells containing pBAD30-yafQ plasmid were grown in M9 minimal medium with 0.2% glucose. At an A590 nm=0.1, cells were centrifuged and resuspended in fresh M9 medium without glucose. The culture was divided into three equal portions. To one portion, arabinose was added to a final concentration of 0.2%, to the second portion, water was added and to the third portion, nalidixic acid or rifampicin at final concentrations of 30 μg mL−1 and 100 μg mL−1 were added, respectively. For assaying DNA synthesis, [2-14C]-thymidine was added to a final concentration of 1 μCi mL−1 to all culture parts. To examine the rate of RNA synthesis, [2-14C]-uridine was added to a final concentration of 1 μCi mL−1. Cells were incubated at 37°C for 2 h, and at different time points as indicated in Fig. 3c and d, 250 μL of culture was taken and mixed with 2.5 mL of chilled 10% TCA solution. Reaction mixtures were kept on ice for 20 min and then applied to nitrocellulose filters (Millipore). Filters were washed twice with chilled 10% TCA and once with 95% ethanol. Radioactivity was determined with a liquid scintillation counter.

image

Figure 3.  Effect of overexpression of YafQ on protein, DNA and RNA syntheses. Escherichia coli BW25113 cells containing the pBAD-yafQ plasmid were grown as described in Materials and methods. Incorporation of [35S]-methionine (a), [14C]-thymidine (c) and [14C]-uridine (d) was measured as a function of time in the absence or presence of 0.2% arabinose, or in the presence of inhibitors of protein (chloramphenicol), DNA (nalidixic acid) and RNA (rifampicin) synthesis. Data are means of at least three independent experiments. Bars indicate SD. (b) SDS 15% PAGE analysis of protein synthesis in vivo under YafQ-inducing and noninducing conditions. Proteins of the same cultures as shown in (a) were analysed.

Download figure to PowerPoint

Assays for protein synthesis in vivo

Escherichia coli BW25113 cells containing pBAD30-yafQ were grown in M9 medium containing 0.2% glucose without methionine and cysteine. At an A590 nm=0.1, cells were centrifuged and resuspended in fresh M9 medium without glucose. The culture was divided into three parts as described above, except that chloramphenicol was added to the third part to a concentration of 33 μg mL−1. Then, [S35]-L-methionine was added to a concentration of 3 μCi mL−1 to all culture parts and cells were incubated at 37°C for 2 h. Samples were taken at different time points as indicated in Fig. 3a. Two hundred and fifty microliters of the samples were used for the radioactivity assay as described above. Another 1.5 mL was chilled on ice, cell pellets were collected by centrifugation and dissolved in 50 μL of SDS-PAGE loading buffer. Fifteen microliters of the samples were analysed by electrophoresis on an SDS 15% polyacrylamide gel.

YafQ structure modelling

Modelling of the E. coli YafQ structure was performed using an automated protein homology modelling server, SWISS-MODEL (Schwede et al., 2003).

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

Inhibitory effect of YafQ on E. coli growth is relieved by coexpression of dinJ present on separate plasmid

In the previous work, we demonstrated that overexpression of the yafQ gene of E. coli dinJ-yafQ operon from an arabinose-inducible expression plasmid inhibits E. coli growth both in liquid and solid medium (Motiejūnaitėet al., 2005). Overexpression of the whole dinJ-yafQ operon from the same plasmid relieved the inhibitory effect of YafQ, indicating that the operon might code for an active TA system. To obtain more evidence that yafQ and dinJ genes code for a toxin protein and its antitoxin, respectively, we individually cloned yafQ and dinJ genes into pBAD30 and pUHE 25-2(cat) expression vectors. The resulting plasmids pBAD-yafQ and pUHE-dinJ were introduced into E. coli strain BW25113F′ (Table 1). Cells were grown in LB medium with 0.2% glucose. At an A590 nm=0.25, bacteria were resuspended in fresh medium with 0.2% arabinose and 0.1 mM IPTG, and incubation was continued. As can be seen in Fig. 1a, the growth of the cells harbouring plasmids pBAD-yafQ/pUHE 25-2(cat) significantly decreased in the presence of the inducers, whereas simultaneous induction of pBAD-yafQ/pUHE-dinJ, pBAD30/pUHE-dinJ and pBAD30/pUHE 25-2(cat) plasmid pairs did not affect E. coli growth. A similar growth pattern was observed when E. coli cells containing pBAD30/pUHE 25-2 plasmids and their derivatives were grown in liquid minimal medium in the presence of 0.2% arabinose (data not shown).

image

Figure 1.  Toxin and antitoxin activity of Escherichia coli YafQ and DinJ. Growth of BW25113F′ containing the combinations of pUHE 25-2(cat)/pBAD30 vectors and their derivatives expressing dinJ and yafQ. (a) Cultures were grown in liquid LB medium in the presence of 0.2% arabinose and 1 mM IPTG. Growth was monitored as described in Materials and methods. Data are means of five independent experiments. Bars indicate SD. (b) Strains were grown on solid LB and glycerol M9 minimal medium (MM) with 0.2% arabinose and 1 mM IPTG.

Download figure to PowerPoint

Thus, taken together, results on the expression of the dinJ gene either from the dinJ-yafQ operon or from a separate plasmid show that DinJ counteracts the toxicity of YafQ in vivo, indicating a TA relationship between YafQ and DinJ.

It must be noted that YafQ appeared less detrimental to cell growth than the other E. coli chromosomal toxins RelE, MazF, ChpBK and YoeB, which, as previously reported, in the presence of 0.2% arabinose rapidly and severely reduced E. coli growth in liquid medium, caused loss in CFU and completely inhibited colony formation on solid medium (Zhang et al., 2003, 2005; Cherny & Gazit, 2004). In the present study, induction of YafQ in liquid LB medium containing 0.2% arabinose did not cause a loss of CFU (data not shown). A strain, bearing the pBAD-yafQ/pUHE 25-2(cat) plasmids, grew on solid LB medium supplemented with 0.2% arabinose and even at higher concentrations of the inducer, although the colony growth rate was significantly reduced as compared with that of pBAD-30/pUHE-dinJ, pBAD-yafQ/pUHE-dinJ and pBAD-30/pUHE 25-2(cat) strains (Fig. 1b). However, the growth of the pBAD-yafQ/pUHE 25-2(cat) plasmids containing strain was disabled on solid glycerol M9 minimal medium with 0.2% arabinose (Fig. 1b). This effect was not a result of an amino acid requirement as pBAD30/pUHE25-2 derivatives bearing cells did not grow in the presence of casamino acids under otherwise identical conditions (data not shown).

YafQ(His)6 and DinJ proteins form a complex

The inhibitory effect on bacterial growth and viability, observed for TA toxins, was shown to be neutralized by their cognate antitoxins through physical interactions with the toxins (Buts et al., 2005). It is thought that under normal growth conditions, cells produce a constant amount of both TA proteins in order to keep the toxin sequestered in a complex with its antitoxin.

In order to investigate whether E. coli DinJ and YafQ proteins interact, we have cloned dinJ-yafQ operon into an IPTG-inducible pET28b expression plasmid. In this construct, a six-His tag is attached to the C-terminus of YafQ. The resulting plasmid, pET-dinJyafQ(His)6, was introduced into BL21(DE3) cells. Coexpression of DinJ and YafQ(His)6 proteins was induced by adding 0.4 mM IPTG and incubating for 2 h. The two proteins were copurified by affinity chromatography using nickel-NTA agarose, as determined by 15% SDS-PAGE (data not shown). The sizes of the purified DinJ and YafQ(His)6 corresponded well with the theoretically calculated molecular masses of 9.4 kDa for DinJ and 10.8 kDa for YafQ (without the histidine tag). YafQ(His)6 was further purified from the complex by dissociating it from DinJ in 6 M guanidine-HCl. YafQ(His)6 was then retrapped on nickel-NTA agarose, eluted as described above and refolded by stepwise dialysis (Fig. 2b, lane 2).

image

Figure 2.  Purification of the DinJ-YafQ(His6) protein complex by gel filtration. The molecular mass of DinJ-YafQ(His6) complex was determined by FPLC gel filtration chromatography using Superose 12HR. (a) The protein molecular mass standard curve includes bovine serum albumin (67 kDa), ovalbumin (43 kDa), β-lactoglobulin (35 kDa), carbonic anhydrase (29 kDa), mioglobin (17.8 kDa) and RNase A (13.7 kDa). The vertical arrow indicates the position of the DinJ-YafQ(His6) complex. (b) Protein analysis by 13.5% SDS PAGE. Gel was stained with Coomasie Brilliant Blue. Lane 1, DinJ(His6) expressed from plasmid pET-dinJ and purified by affinity chromatography on nickel-NTA agarose; lane 2, YafQ(His6), purified from the DinJ-YafQ(His6) complex; lane 3, protein fraction eluted from gel filtration column after loading it with the DinJ-YafQ(His6) complex expressed from plasmid pET28b-dinJyafQ(His)6 and purified by affinity chromatography on nickel-NTA agarose; lane 4, protein molecular mass markers (band sizes in kilodaltons are shown on the right).

Download figure to PowerPoint

Next, fractions obtained by affinity chromatography and containing DinJ and YafQ(His)6 proteins were further analysed by gel filtration chromatography on a Superose 12HR column (Pharmacia). The FPLC elution profile showed a single peak corresponding to an entity with a molecular mass of c. 37.3 kDa as determined by protein a molecular mass calibration curve (Fig. 2a). Analysis of the eluted fraction by 13.5% SDS-PAGE showed two bands after staining with Coomassie Brilliant Blue, which corresponded to the DinJ and YafQ(His)6 proteins (Fig. 2b, lane 3). The average ratio of DinJ to YafQ(His)6 protein was found to be 1 : 1 as determined by densitometry.

The observations that DinJ and YafQ proteins copurify in affinity and gel filtration chromatography, and that YafQ may be dissociated from DinJ in the presence of 6 M guanidine-HCl, indicate that DinJ and YafQ interact, forming a specific and stable protein heterocomplex. The experimentally determined molecular mass of DinJ/YafQ(His)6 is most close to the predicted molecular mass of DinJ/YafQ(His)6 complex consisting of two DinJ and two YafQ(His)6 monomers (∼42.2 kDa). Thus, the results of gel filtration chromatography of DinJ/YafQ(His)6 complex described above suggest that DinJ/YafQ(His)6 may represent a tetramer.

Of chromosomal TA protein complexes studied to date, MazE and MazF proteins have been shown to build a linear heterohexamer composed of an antitoxin MazE homodimer, sandwiched between two toxin homodimers (Kamada et al., 2003). In contrast, E. coli YoeB toxin and YefM antitoxin form a 1 : 2 heterotrimer, whereas the archaeal RelB–RelE protein complex from Pyrococcus horikoshii consists of two molecules of each antitoxin RelB and toxin RelE (Kamada & Hanaoka, 2005; Takagi et al., 2005). Therefore, chromosomal TA protein pairs exhibit structural diversity in their macromolecular complexes, which may be important in regulating the stability of the TA components and/or have other regulatory roles. It has been shown that the susceptibility of some free antitoxins to proteases results in part from the presence of unfolded protein domains in their structures, which become ordered when complexed with their toxins (Kamada & Hanaoka, 2005).

Escherichia coli YafQ inhibits [35S]-methionine incorporation

To determine which essential cellular processes are inhibited by overexpression of YafQ, we next studied the incorporation of radiolabelled precursors of replication, transcription and translation upon induction of YafQ synthesis in E. coli cells harbouring the pBAD-yafQ plasmid. Overexpression of YafQ was induced by adding 0.2% arabinose to the growth medium and incubating the cells for 2 h. Incorporation of [35S]-methionine was reduced to c. 60% as compared with that observed in the same cells grown under noninducing conditions (in the presence of 0.2% glucose) (Fig. 3a). Ten percent SDS-PAGE analysis of total E. coli proteins in arabinose-induced and control cells showed that overexpression of YafQ reduces protein synthesis (Fig. 3b).

In contrast, the incorporation of precursors of DNA and RNA biosynthesis, [2-14C]-thymidine and [2-14C]-uridine, was not affected by overexpression of YafQ under inducing conditions (Fig. 3c and d). These results demonstrate that E. coli YafQ affects protein synthesis, but not DNA or RNA synthesis.

To date, two targets of TA toxins, DNA gyrase and mRNA, have been identified (Buts et al., 2005). Escherichia coli chromosomally encoded toxins, RelE, MazF, YoeB and ChpBK, have been shown to inhibit translation through degradation of mRNA at specific codons (Christensen & Gerdes, 2003; Zhang et al., 2003, 2005). MazF, YoeB and ChpBK toxins showed intrinsic RNase activity, whereas RelE triggered RNase activity of the ribosomes, which have been stalled under some stress conditions (Christensen et al., 2004). As YafQ inhibited translation, as demonstrated in this study, it could be hypothesized to trigger RNA degradation as well. Structural modelling of YafQ has shown that it exhibits a protein fold structurally similar to that of microbial RNases, such as T1 and barnase, as well as E. coli YoeB and P. horikoshii RelE TA toxins (Fig. 4a). A YafQ sequence alignment to bacterial toxins of RelE family revealed that the protein possesses a conservative histidine residue (His87) (Fig. 4b shaded in grey), which has been demonstrated to be crucial for the catalytic activity of E. coli YoeB endonuclease and proposed to serve as a general acid in the catalytic reaction (Kamada & Hanaoka, 2005). However, Glu46 (the proposed general base), Tyr84, shown to be crucial for YoeB toxin catalytic activity and Arg65 (a candidate for phosphate binding), important for YoeB and archaeal RelE (Fig. 4b, shaded in black), are not conserved in YafQ. Several residues at positions different from those of YoeB, which could potentially participate in the active site chemistry, were proposed for YafQ (Kamada & Hanaoka, 2005). For example, Arg83 in E. coli YafQ is highly conservative among YafQ homologues in a number of microorganisms as are Asp67 and Asp61 (Fig. 4b, shaded in grey). A different distribution of amino acids, which could participate in catalysis, in YafQ led to a suggestion that it could possess a type of active centre different from that of YoeB (Kamada & Hanaoka, 2005). This could imply specific substrate requirements for YafQ and could also explain the lower toxicity of YafQ in vivo, observed in this study, compared with that reported for YoeB. In support of this conclusion is our observation that the synthesis of larger proteins was more affected upon YafQ overproduction than that of smaller proteins, thus indirectly indicating that the pool of larger mRNA could be more susceptible to the toxin (data not shown). Noteworthy, among the E. coli chromosomal TA systems tested, YafQ appears to be the most temperate toxin.

image

Figure 4.  Structures and a sequence alignment of RelE family chromosomal toxins. (a) Comparison of crystal structures of Escherichia coli YoeB (PDB code 2A6S) and Pyrococcus horikoshii RelE (aRelE, PDB code 1WMI), and the structure of E. coli YafQ obtained by computer modelling. α-Helixes are shown in grey, and β-strands in black. (b) Sequence alignment of RelE family chromosomal toxins: RelE–RelE of E. coli, aRelE–RelE of P. horikoshii, YoeB–YoeB of E. coli and YafQ–YafQ of E. coli. Gaps were introduced to maximize alignment and are indicated by dashes. Numbers below the aligned sequences denote every tenth amino acid of YafQ. Identical amino acids in three or four proteins are indicated by asterisks. Functionally important amino acids of aRelE (Takagi et al., 2005) and YoeB (Kamada & Hanaoka, 2005) are shaded in black, and catalytically important amino acids, proposed for YafQ, in grey.

Download figure to PowerPoint

The observations that the dinJ-yafQ system, similarly to the other four E. coli chromosomal TA systems, targets translation, and that YafQ shows structural similarity to known ribonucleases, raise the question of whether dinJ-yafQ plays a redundant or a specific physiological role among bacterial TA modules.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

This work was supported by Lithuanian State Science and Studies foundation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References
  • Bujard H, Gentz R, Lancer M, Stüber D, Müller M, Ibrahimi I, Häuptle MT & Dobberstein B (1987) A T5 promoter-based transcription-translation system for analysis of protein in vitro and in vivo. Methods Enzymol 155: 416433.
  • Buts L, Lah J, Dao-Thi M-H, Wyns L & Loris R (2005) Toxin-antitoxin modules as bacterial metabolic stress managers. Trends Biochem Sci 30: 672679.
  • Cherny I & Gazit E (2004) The YefM antitoxin defines a family of natively unfolded proteins: implications as a novel antibacterial target. J Biol Chem 279: 82528261.
  • Christensen SK & Gerdes K (2003) RelE toxins from bacteria and Archaea cleave mRNAs on translating ribosomes, which are rescued by tmRNA. Mol Microbiol 48: 13891400.
  • Christensen SK, Mikkelsen M, Pedersen K & Gerdes K (2001) RelE, a global inhibitor of translation, is activated during nutritional stress. Proc Natl Acad Sci USA 98: 1432814333.
  • Christensen SK, Pedersen K, Hansen G & Gerdes K (2003) Toxin-antitoxin loci as stress-response elements: ChpAK/MazF and ChpBK cleave translated mRNAs and are counteracted by tmRNA. J Mol Biol 332: 809819.
  • Christensen SK, Maenhaut-Michel G, Mine N, Gottesman S, Gerdes K & Van Melderen L (2004) Overproduction of the Lon protease triggers inhibition of translation in Eschericha coli: involvement of the yefM-yoeB toxin-antitoxin system. Mol Microbiol 51: 17051717.
  • Condon C (2006) Shutdown decay of mRNA. Mol Microbiol 61: 573583.
  • Datsenko KA & Wanner BL (2000) One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci USA 97: 66406645.
  • Gerdes K, Rasmussen PB & Molin S (1986) Unique type of plasmid maintenance function: postsegregational killing of plasmid-free cells. Proc Natl Acad Sci USA 83: 31163120.
  • Gerdes K, Christensen SK & Løbner-Olesen A (2005) Prokaryotic toxin-antitoxin stress response loci. Nat Rev Microbiol 3: 371382.
  • Guzman LM, Belin D, Carson MJ & Beckwith J (1995) Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J Bacteriol 177: 41214130.
  • Kamada K & Hanaoka F (2005) Conformational change in the catalytic site of the ribonuclease YoeB toxin by YefM antitoxin. Mol Cell 19: 497509.
  • Kamada K, Hanaoka F & Burley SK (2003) Crystal structureof the MazE/MazF complex: molecular bases of antidote toxin recognition. Mol Cell 11: 875884.
  • Motiejūnaitė R, Armalytė J, Šeputienė V & Sužiedėlienė E (2005) Escherichia coli dinJ-yafQ operon shows characteristic features of bacterial toxin-antitoxin modules. Biologija 4: 915.
  • Pandey D & Gerdes K (2005) Toxin-antitoxin loci are highly abundant in free-living but lost from host-associated prokaryotes. Nucleic Acid Res 33: 966976.
  • Pedersen K, Christensen SK & Gerdes K (2002) Rapid induction and reversal of bacteriostatic conditions by controlled expression of toxins and antitoxins. Mol Microbiol 45: 501510.
  • Pedersen K, Zavialov K, Pavlov MY, Elf J, Gerdes K & Ehrenberg M (2003) The bacterial toxin RelE displays codon-specific cleavage of mRNAs in the ribosomal A site. Cell 112: 131140.
  • Schwede T, Kopp J, Guex N & Peitsch MC (2003) SWISS-MODEL: an automated protein homology-modeling server. Nucleic Acids Res 31: 33813385.
  • Studier FW & Moffat BA (1986) Use of bacteriophage T7 RNA polymerase to direct high-level expression of cloned genes. J Mol Biol 198: 113130.
  • Takagi H, Kakuta Y, Okada T, Yao M, Tanaka I & Kimura M (2005) Crystal structure of archaeal toxin-antitoxin RelE-RelB complex with implications for toxin activity and antitoxin effects. Nat Struct Mol Biol 12: 327331.
  • Zhang Y, Zhang J, Hoeflich KP, Ikura M, Quing G & Inoue M (2003) MazF cleaves cellular mRNAs specifically at ACA to block protein synthesis in Escherichia coli. Mol Cell 12: 913923.
  • Zhang Y, Zhu L, Zhang J & Inoue M (2005) Characterization of ChpBK, an mRNA interferase from Escherichia coli. J Biol Chem 280: 2608026080.